2nd Inter
national Ballast W
Global Ballast Water
Management Programme
G L O B A L L A S T M O N O G R A P H S E R I E S N O . 1 5
ater T

R&D Symposium
2nd International Ballast Water Treatment
R&D Symposium
IMO, LONDON, 21-23 JULY 2003
Eds. Jose Matheickal & Steve Raaymakers
More Information?
el (+44) 020 7928 5888 www
Programme Coordination Unit
Global Ballast Water Management Programme
International Maritime Organization
4 Albert Embankment
London SE1 7SR United Kingdom

Tel: +44 (0)20 7587 3247 or 3251
est & Associates, London. T
Fax: +44 (0)20 7587 3261
Web: http://globallast.imo.org

A cooperative initiative of the Global Environment Facility,
United Nations Development Programme and International Maritime Organization.
Cover designed by Daniel W

GloBallast Monograph Series No. 15
2nd International Ballast Water
Treatment R&D Symposium
IMO London: 21-23 July 2003
Jose Matheickal and Steve Raaymakers (Eds)1

1 Programme Coordination Unit, Global Ballast Water Management Programme, International Maritime Organization,
4 Albert Embankment, London SE1 7SR, UK

International Maritime Organization
ISSN 1680-3078

Published in September 2004 by the
Programme Coordination Unit
Global Ballast Water Management Programme
International Maritime Organization
4 Albert Embankment, London SE1 7SR, UK
Tel +44 (0)20 7587 3251
Fax +44 (0)20 7587 3261
Email sraaymak@imo.org
Web http://globallast.imo.org

The correct citation of this report is:
Matheickal, J.T. and Raaymakers S. (Eds) 2004. 2nd International Ballast Water Treatment R&D Symposium, IMO London,
21-23 July 2003: Proceedings.
GloBallast Monograph Series No. 15. IMO London.

The Global Ballast Water Management Programme (GloBallast) is a cooperative initiative of the Global Environment Facility (GEF),
United Nations Development Programme (UNDP) and International Maritime Organization (IMO) to assist developing countries to reduce
the transfer of harmful organisms in ships' ballast water.

The GloBallast Monograph Series is published to disseminate information about and results from the programme, as part of the
programme's global information clearing-house functions.

The opinions expressed in this document are not necessarily those of GEF, UNDP or IMO.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The following persons and parties are acknowledged for their contributions to making the Symposium
a success:
· The Secretary-General of IMO, Mr William O'Neil for hosting the symposium at IMO
· All session chairpersons for facilitating a smooth running of the technical sessions of the
· All persons who submitted and presented papers, providing the very substance of the
· All other symposium participants, without whom the symposium would not be an event.
· The IMO Conference Section for organizational and facilities support.
· The IMO Catering Section for sustaining symposium delegates.
· Mrs Christine Gregory and Mr Leonard Webster for their `behind the scenes' efforts in
supporting the symposium.
· Mrs Sarah Harden and Ms Fiona Morris of IMarEST.
· Prof Thomas Waite (US National Science Foundation), Mr Mike Hunter (UK Maritime and
Coast Guard Agency) and Dr Ehsan Mesbahi (University of Newcastle Upon Tyne School of
Marine Science and Technology) for organizing sponsorship for the symposium from their
respective organizations.
· Mr Leonard Webster of the GloBallast PCU for layout and formatting of this report.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Some of the delegates at the Symposium ­ nearly 230 attended.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Mr. Steve Raaymakers
Chief Technical Adviser, Global Ballast Water Management Programme
The issue of aquatic invasive species, including the transfer of harmful organisms in ships' ballast
water and sediments, is considered to be one of the greatest threats to global marine bio-diversity and
ecosystems, and also a significant threat to coastal economies and even public health. Global
economic impacts from invasive aquatic species, including through disruption of fisheries, fouling of
coastal industry and infrastructure and interference with human amenity, are estimated to exceed tens
of billions of Euros per year. The US General Accounting Office (2003) has identified biological
invasions as one of the greatest environmental threats of the 21st Century. The United Nations
Environment Programme (UNEP) and World Conservation Union (IUCN), announced at the World
Summit on Sustainable Development (WSSD) in Johannesburg in 2002, that invasive species are the
second greatest threat to global bio-diversity after habitat loss. The impacts are set to increase in
coming years with a three-fold increase in shipping activity predicted in the next decade.
The main management measure to reduce this risk, as recommended under the existing IMO ballast
water guidelines, is ballast exchange at sea. However, it has also been widely recognised that Ballast
Water Exchange at Sea has limitations, including:
· serious safety concerns on ballast water exchange operations at sea; and
· the fact that translocation of species can still occur even when a vessel has undertaken the
ballast exchange in accordance with the current guidelines.
It is therefore extremely important that alternative, more effective ballast water methods are
developed as soon as possible, and the impending Ballast Water Convention provides a powerful,
regulatory-driven incentive to support research and development efforts aimed at alternative methods.
Although significant research and development (R&D) efforts are underway by a number of
establishments around the world there are no formal mechanisms in place to ensure effective lines of
communication between IMO, the R&D community, governments and ship designers, builders and
owners on this issue. These are vital if the R&D effort is to succeed.
To help address this situation, the GloBallast Programme initiated an International Ballast Water
Treatment R&D Symposium series and the first symposium was held in March 2001 in London. The
Symposium was hailed as a major success and participants requested that it become a regular event
held every one or two years. In response, the GloBallast has organised this Second Symposium in
conjunction with the Institute of Marine Engineering, Science and Technology (IMarEST), and with
support from the United Kingdom Maritime and Coast Guard Agency; the University of Newcastle
upon Tyne School of Marine Science and Technology; and the National Science Foundation in the
United States.
The second symposium which had a truly global scope and highly focused objectives, brought
together world's leading experts in the specialised field of ballast water treatment. Over the three
days, thirty-six papers were presented, covering all of the main technologies currently being
researched and updating the latest results from the major R&D projects, thus catalyzing a more co-
ordinated and co-operative global R&D efforts. The symposium attracted nearly 230 participants.
The papers contained in this Symposium Proceedings provide a very useful information resource for
all parties interested in the topic of ballast water treatment, management and control.
Ballast water transfers and invasive marine species are one of the most serious environmental
challenges facing the global shipping industry. I am pleased that the outcomes of the symposium are

2nd International Ballast Water Treatment R&D Symposium: Proceedings
providing important catalysts for progressing the new international ballast water convention and for
moving us closer to a practical solution to the `ballast water problem.'
Symposium Objectives
The objectives of the symposium were to:
· Update the current status of ballast water treatment R&D around the world.
· Enhance communication links between IMO, member countries, the R&D community and
ship designers, builders and owners on ballast water treatment issues.
Major Outcomes
Some of the conference highlights and general conclusions from the conference are given below:
In opening the Symposium, the Director of the IMO Marine Environment Division, Mr Koji
Sekimizu, speaking on behalf of the Secretary-General, Mr William O'Neil, stated that during the
development of the ballast water Convention, it has been widely recognized that the practice of ballast
exchange at sea has many limitations, including serious safety concerns and highly variable biological
effectiveness. As an example, approximately 15 new species have invaded the North American Great
Lakes since 1993, despite mid-ocean exchange becoming mandatory that year for ships entering the
Lakes region. This is the same number of invasions that occurred during the 1970s and 80s, indicating
that current management efforts are not completely effective. Overall, the current rate of invasions in
the Great Lakes is 66% higher than 100 years ago, and similar trends are recorded in other parts of the
world where surveys and monitoring are conducted. Mr Sekimizu stated that it is therefore extremely
important that alternative, more effective ballast water treatment methods are developed as soon as
In delivering the keynote address at the Symposium, Dr Thomas Waite, Programme Director of
Environmental Engineering at the US National Science Foundation stated, inter alia, that the search
for solutions requires far more input from naval architects and marine engineers, that the initial focus
should be on adapting existing water treatment techniques, that the R&D effort should look for
synergies between treatment processes, and that non-chemical, reversible treatments such as heat, de-
oxygenation and pH extremes should be seriously pursued, along with new techniques such as light-
sensitive biocides.
A total of 36 technical papers were presented over the three days covering mechanical and gas-based
treatment systems, heat and electro-based systems, chemical-based approaches, multiple technologies
and combined systems, with a special session on test protocols and verification procedures.
A general picture that emerged from the technical presentations is summarised as follows:
· Overall, there has been a significant increase in R&D and good progress has been made by
several groups in moving closer to viable, practical, effective solutions, although most of the
groups still remain at the basic research stage. The lack of finalised treatment standards in the
IMO Convention (at the time) was identified as still being the major obstacle to the R&D
· It is unlikely that a single treatment technology will suit all vessel types and voyage
characteristics. The R&D community should seek to develop different treatment options for

2nd International Ballast Water Treatment R&D Symposium: Proceedings
different scenarios, as long as they meet the international performance standard. For example,
heat appears to hold significant promise for cruise ships and some tankers that generate
significant waste heat, but is unlikely to be an option for bulk carriers with large volumes of
ballast but little waste heat.
· It appears that treatment systems will need to involve combined technologies, and that
primary filtration or physical separation will almost certainly be necessary, followed by
secondary biocidal treatment(s). If primary filtration alone was implemented now, a
significant reduction in bio-invasions would be achieved.
· The development of internationally standardised test protocols and verification procedures
was identified as the most urgent remaining priority that must be addressed by IMO.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Symposium Programme
Sunday 20 July 2003
1400 ­ 1800:
Monday 21 July 2003: Day 1
0730 ­ 0900:
Opening & Keynote Speakers
0900 ­ 0915:
Opening statement: Mr William O'Neil, Secretary-General, IMO
0915 ­ 0935:
Keynote address: Dr T Waite, National Science Foundation, USA
0935 ­ 0950:
Introduction, background and objectives of the symposium: Mr S Raaymakers, GloBallast
0950 ­ 1010:
Official group photograph
1010 ­ 1040:
Session One: Mechanical and Gas-Based Treatment Systems
1040 ­ 1105:
The Ternary Effect for ballast water treatment
I Kreisel, N Shimron, Y Kolodny, D Sorek, Arkal Filtration Systems, Israel
Y Sasson, The Hebrew University of Jerusalem, Israel
A Cangelosi, Northeast Midwest Institute
C Blatchley, Purdue University, M Blacer, University of Wisconsin,
P Brodie, Balaena Dynamics Ltd,
B Cairns, R Braun, Trojan Technologies Inc
1105 ­ 1130:
Progress report on the `Special Pipe System' as a potential mechanical treatment for ballast
T Kikuchi, K Yoshida, S Kino, The Japan Association of Marine Safety, Japan
Y Fukuyo,University of Tokyo, Japan
1130 ­ 1155:
Progress report on the AquaHabiStat Deoxygenation system
W J Browning, J Parker Davis, W J Browning III, AquaHabiStat, USA
Capt C Thompson, USCG Retired, USA
Dr R Mann, Virginia Institute of Marine Science, USA
1155 ­ 1220:
Evaluations of Venturi Oxygen Stripping as a ballast water treatment to prevent aquatic
invasions and ship corrosions
Dr M N Tamburri, University of Maryland Center for Environmental Science, USA
B J Little, Stennis Space Center, USA
G M Ruiz, Smithsonian Environmental Research Center, USA
P D McNulty, NEI Treatment Systems Inc, USA
1220 ­ 1245:
Ballast water treatment by De-oxygenation with elevated CO2 for a shipboard installation ­ a
potentially affordable solution
M Husain, R Apple, D Altshuller, C Quirmbach, MH Systems Inc, USA
H Felbeck, Scripps Institution of Oceanography, USA
1245 ­ 1315:
Session one panel discussion
1315 ­ 1430:
Session Two: Heat and Electro-Based Treatment Systems
(each presentation 20 min + 5 min questions)

1430 ­ 1455:
Does heat offer a superior ballast water treatment option?
G Rigby, Reninna Pty Limited, Australia
G Hallegraeff, University of Tasmania, Australia
A Taylor, Alan H Taylor & Associates, Australia

2nd International Ballast Water Treatment R&D Symposium: Proceedings
1455 ­ 1520:
Treatment of residual ballast water in the NOBOB ship using heat
D T Stocks, BMT Fleet Technology Ltd, Canada
M O'Reilly, ESG International Inc, Canada
1520 ­ 1550:
1550 ­ 1615:
The use of heat for ballast water disinfection ­ the AquaTherm method
G A Thornton, Hi Tech Marine Pty Ltd, Australia
1615 ­ 1640:
Application study of ballast water treatment by electrolyzing seawater
K Dang, P Yin, P Sun, Y Song, Dalian Maritime University, P R China
1640 ­ 1705:
Electro-sanitization of ballast water
C E Leffler, B Paul, P Trupiano, A Salamone, Marine Environmental Partners Inc, USA
A Rogerson, S Grubbs, C Cox, Nova South Eastern University, USA
1705 ­ 1730:
Superconducting magnetic separator for ballast water treatment
Dr N Saho, H Isogami, T Mizumori, N Nishijima, HITACHI Ltd, Japan
1730 ­ 1800:
Session two panel discussion
1800 ­ 2030:
Reception ­ IMO delegates' lounge
Sponsored by IMarEST, the University of Newcastle upon Tyne and GloBallast
Tuesday 22 July 2003: Day 2
0800 ­ 0900:
Registration (at registration desk, ground floor, IMO)
Session Three: Chemical-Based Treatment Systems
(each presentation 20 min + 5 min questions)

0900 ­ 0925:
Sodium Hypochlorite as a ballast water biocide
D T Stocks, BMT Fleet Technology Ltd, Canada
M O'Reilly, ESG International Inc, Canada
W E McCracken, Consultant, USA
0925 ­ 0950:
Effects of chlorination treatment for ballast water
S Zhang, J Xiao, D Yang, W Gong, Q Wang, Dalian Maritime University, P R China
0950 ­ 1015:
Use of chlorine for ballast water treatment
J da Silva, F da Costa Fernandes, Instituto de Estudos do Mar
Almirante Paulo Moreira ­ IEAPM, Brazil
1015 ­ 1045:
1045 ­ 1110:
SeaKleen®: a potential product for controlling aquatic pests in ships' ballast water
J Cutler, H G Cutler, Garnett Inc, USA
Glinski, Planta Analytica, USA
D Wright, R Dawson, University of Maryland Center for Environmental Science, USA
D Lauren, HortResearch Ruakura, New Zealand
1110 ­ 1135:
Peraclean® Ocean ­ a potential treatment option for ballast water
Dr R Fuchs, Degussa AG, Germany
1135 ­ 1200:
Acrolein as a potential treatment alternative for control of micro-organisms in ballast tanks:
five day sea trial
Dr J E Penkala, M D Law, J K Cowan, Baker Petrolite, USA
1200 ­ 1230:
Session three panel discussion
1230 ­ 1345:
Session Four: Multiple Technologies and Combined Systems
(each presentation 20 min + 5 min questions)

1345 ­ 1410:
Solution to ballast water pollution: ship shape and ports escape?
E. Donkers, Port of Rotterdam, The Netherlands
1410 ­ 1435:
Latest results from testing seven different technologies under the EU MARTOB project ­
where do we stand now?
Dr E Mesbahi, University of Newcastle upon Tyne, UK
1435 ­ 1500:
The TREBAWA ballast water treatment project
Capt K Hesse, Reederei Hesse GmbH and Co, Germany

2nd International Ballast Water Treatment R&D Symposium: Proceedings
1500 ­ 1530:
1530 ­ 1555:
Shipboard trials of ballast water treatment systems in the United States
D A Wright, R Dawson, University of Maryland Center for Environmental Science, USA
P Mackey, Hyde Marine Inc, USA
H G Cutler, S J Cutler, Mercer University, USA
1555 ­ 1620:
Development and design of process modules for ballast water treatment onboard
Dr Ing A Kornmueller, Berkefeld Water Technology/RWO Marine Water Technology,
1620 ­ 1645:
Hydrodynamic cavitation and filtration treatment of ballast water
A Andryushchenko, Engineering Center TRANSZVUK, Ukraine
1645 ­ 1715:
Session four panel discussion
Close day two
Wednesday 23 July 2003: Day 3
0800 ­ 0900:
Registration (at registration desk, ground floor, IMO)
Session Four contd: Multiple Technologies and Combined Systems
(each presentation 20 min + 5 min questions)

0900 ­ 0925:
A new modular concept for the treatment of ships' ballast water - the Hamann project
Dipl Ing H Röpell, Dipl Ing T Mann, Hamann Wassertechnik GmbH, Germany
0925 ­ 0950:
A portable pilot plant to test the treatment of ships' ballast water
S Hillman, Dr P Schneider, Dr F Hoedt, James Cook University, Australia
0950 ­ 1015:
Ballast water treatments R&D in The Netherlands
J L Brouwer, Royal Haskoning, The Netherlands
Dr C C ten Hallers-Tjabbes, Netherlands Institute for Sea Research (NIOZ), The
1015 ­ 1045:
1045 ­ 1110:
Ballast water treatment research and interim approval in Washington State
S S Smith, Washington Department of Fish and Wildlife, USA
1110 ­ 1135:
Corrosion effects of ballast water treatment methods
E Dragsund, A B Andersen, B O Johannessen, J O Nųkleby, Det Norske Veritas, Norway
1135 ­ 1205:
Session four (contd) panel discussion
1205 ­ 1310:
Session Five: Test Protocols and Verification Procedures
(each presentation 20 min + 5 min questions)

1310 ­ 1335:
A proposed frame-work for approving ballast water treatment technologies
Dr D O Mountfort, M D Taylor, T J Dodgshun, Cawthron Institute, New Zealand
1335 ­ 1400:
Ballast water treatment verification protocol ­ DNV proposal
E Dragsund, A B Andersen, B O Johannessen,
Det Norske Veritas, Norway
K Tangen, Oceanor, Norway
1400 ­ 1425:
The Artemia Testing System for ballast water treatment options
Dr M Voigt, dr. voigt-consulting, Germany
1425 ­ 1450:
Development of dinoflagellate "cyst-on-demand" protocol and comparison of particle
monitoring techniques for ballast water treatment evaluation
Dr J T Matheickal, Prof T J Hwa, S Mylvaganam, L Loke, Institute of Environmental
Science and Engineering, Singapore
Dr M Holmes, Tropical Marine Science Institute, Singapore
1450 ­ 1520:
1520 ­ 1545:
Test procedure for evaluation of ballast water treatment systems using copepods as
zooplankton and dinoflagellates as phytoplankton
T Kikuchi, K Yoshida, S Kino, The Japan Association of Marine Safety, Japan
Y Fukuyo, University of Tokyo, Japan

2nd International Ballast Water Treatment R&D Symposium: Proceedings
1545 ­ 1610:
Testing ballast water treatment equipment
Prof A E Holdų, University of Hertfordshire, UK
1610 ­ 1635:
Performance verification testing of ship ballast water treatment technologies by USEPA/NSF
Environmental Technology Verification programme
T G Stevens, NSF International, USA
R M Frederick, US Environmental Protection Agency, USA
R A Everett, J T Hurley, US Coast Guard, USA
C D Hunt, D Tanis, Battelle, USA
1635 ­ 1705:
Session five panel discussion
1705 ­ 1730:
Conclusions and recommendations
Close Symposium

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Acknowledgements ...............................................................................................................................i
Foreword ...............................................................................................................................................iii
Symposium Objectives .......................................................................................................................iv
Major Outcomes...................................................................................................................................iv
Symposium Programme .....................................................................................................................vi
Papers presented
Session 1: Mechanical and Gas Based Systems
The ternary effect for ballast water treatment ..................................................................................5
I. Kriesel, Y.Kolodny, W.L. Cairns, B.S. Galil, Y. Sasson, A.V. Joshi, A. Cangelosi,
E.R. Blatchley III, M.C. TenEyck, M.D. Blacer & P. Brodie
Progress report on the `Special Pipe System' as a potential mechanical treatment for
ballast water ........................................................................................................................................19

T. Kikuchi, K. Yoshida, S. Kino & Y Fukuyo
Progress report on the AquaHabiStat deoxygenation system ....................................................26
W. J. Browning Jr., J. P. Davis & W. J. Browning III
Evaluations of Venturi Oxygen StrippingTM as a ballast water treatment to prevent aquatic
invasions and ship corrosion ...........................................................................................................34

M. N. Tamburri, B. J. Little, G. M. Ruiz, J. S. Lee, & P. D. McNulty
Ballast water treatment by de-oxygenation with elevated CO2 for a shipboard installation ­
a potentially affordable solution ......................................................................................................48

M. Husain, H. Felbeck, R. Apple, D. Altshuller & C. Quirmbach
Session 2: Heat and Electro-based Treatment Systems
Does heat offer a superior ballast water treatment option? ........................................................67
G. Rigby, G. Hallegraeff & A. Taylor
Treatment of residual ballast water in the NOBOB ship using heat ...........................................80
D.T. Stocks, M. O'Reilly, & W. McCracken
The use of heat for ballast water disinfection - the AquaTherm method...................................88
G. A. Thornton & J. E. Chapman
Application study of ballast water treatment by electrolysing seawater .................................103
K. Dang ,P. Yin, P. Sun, J. Xiao & Y. Song
Electro-sanitization of ballast water ..............................................................................................111
C.E. Leffler, A. Rogerson, W. Paul, G. Germaine, M. Elliot, V. Antonelli, S. Grubs, C. Campbell,
G. Beall & A. Salamone
Superconducting magnetic separator for ballast-water treatment ...........................................125
N. Saho, H. Isogami, T. Mizumori & N. Nishijima
Session 3: Chemical-based Treatment Systems
Sodium hypochlorite as a ballast water biocide..........................................................................137
D.T. Stocks, M. O'Reilly & W. McCracken
Effects of the chlorination treatment for ballast water ...............................................................148
S. Zhang, X. Chen, D. Yang, W. Gong, Q. Wang, J. Xiao, H. Zhang & Q. Wang

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Use of chlorine for ballast water treatment.................................................................................. 158
J. S. Vianna da Silva & F. da Costa Fernandes
SeaKleen®, a potential product for controlling aquatic pests in ships' ballast water ........... 164
S. J. Cutler, H. G. Cutler, J. Glinski, D. Wright, R. Dawson & D. Lauren
Peraclean® Ocean ­ A potentially environmentally friendly and effective treatment
option for ballast water ................................................................................................................... 175

R. Fuchs & I. de Wilde
Acrolein as a potential treatment alternative for control of microorganisms in ballast
tanks: five day sea trial ................................................................................................................... 181

J. E. Penkala, M. Law & J. Cowan
Session 4: Multiple Technologies and Combined Systems
Solution to ballast water pollution: ship shape and the ports escape? .................................. 201
E. Donkers
Latest results from testing seven different technologies under the EU MARTOB project -
Where do we stand now? ............................................................................................................... 210

E. Mesbahi
The TREBAWA ballast water project ............................................................................................ 231
K. Hesse, Mike Casey, P. Zhou, F. Aslan, E. Schmid, A. Leigh & A. Santos
Some shipboard trials of ballast water treatment systems in the United States ................... 243
D. A. Wright, R. Dawson, T. P. Mackey, H. G. Cutler & S. J. Cutler;
Development and design of process modules for ballast water treatment on board ........... 258
A. Kornmueller
Hydrodynamic transonic treatment and filtration of ship ballast water .................................. 264
A. Andruschenko, A. Dukhanin, V. Rabotnyov, Y. Skanunov & S. Tishkin
A new modular concept for the treatment of ships ballast water - the Hamann project ...... 271
H. Röpell & T. Mann
A portable pilot plant to test the treatment of ships' ballast water .......................................... 274
S. Hillman, F. Hoedt & P. Schneider
Ballast water treatment R&D in the Netherlands: Ballast water treatment on-board of
ships & evaluation of market potential and R&D requirements ............................................... 282

C.C. ten Hallers-Tjabbes, J. Boon, M.J. Veldhuis, J.L. Brouwer & J.Rvan Niekerk
Ballast water treatment - management and research in Washington State............................ 289
S. S. Smith
Corrosion effects of ballast water treatment methods............................................................... 291
E. Dragsund, B. O. Johannessen, A. B. Andersen & J. O. Nųklebye
Session 5: Test Protocols and Verification Procedures
A proposed frame-work for approving ballast water treatment technologies........................ 303
D. Mountfort, T. Dodgshun & M. Taylor
Ballast Water Treatment Verification Protocol - DNV................................................................. 309
A. B. Andersen, B. O. Johannessen & E. Dragsund
The Artemia Testing System for ballast water treatment options ............................................ 321
M. Voigt
Development of dinoflagellate "cyst-on-demand" protocol, and comparison of particle
monitoring techniques for ballast water treatment evaluation................................................. 326

J. T. Matheickal, Tay Joo Hwa, Chan Mya Tun, S. Mylvaganam, M. Holmes & L. Loke

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Test procedure for evaluation of ballast water treatment system using copepoda as
zooplankton and dinoflagellates as phytoplankton ....................................................................340

T. Kikuchi, K. Yoshida, S. Kino & Y. Fukuyo
Testing ballast water treatment equipment ..................................................................................347
A. E. Holdų
Performance verification of ballast water treatment technologies by USEPA/NSF
Environmental Technology Verification Program .......................................................................353

T. G. Stevens, R. M. Frederick, R. A. Everett, J. T. Hurley, C. D. Hunt, & D. C. Tanis
Papers submitted (not presented at the Symposium)
Design optimisation of a UV system for onboard treatment of ballast water .........................363
M. Casey, A. Leigh & P. Zhou
Ship ballast water treatment: the closed-loop option ................................................................372
P. Brodie, C. Blatchley, M. Balcer, Y. Sasson, I. Kreisel, Y. Kolodni, D. Sorek, N. Shimron, W. Cairns
& R. Braun
Using MCDA methods in an application for outranking the ballast water management
options ...............................................................................................................................................380

C. F. S. Gomes
The eco-friendly ship of the future.................................................................................................391
K. V. Subba Rao
Appendix 1: List of Participants

Papers presented
Papers have been included in these
Proceedings largely as submitted, with
basic editing and formatting only, and
without scientific or technical peer

Neither the GloBallast Programme, the
International Maritime Organization
(IMO), the Institute of Marine
Engineering, Science and Technology
(IMarEST) nor the symposium
sponsors take any responsibility
whatsoever for any statements and
claims made in these papers, for the
quality, accuracy and validity of data
presented, or for any other contents of
these papers.

Individuals and organisations that
make use of any data or other
information contained in these papers
do so entirely at their own risk.

Inclusion of papers in these
proceedings in no way constitutes any
form of endorsement whatsoever by
IMO, GloBallast, IMarEST or the
symposium sponsors.

Session 1:
Mechanical and
Gas Based Systems

The ternary effect for ballast water treatment
I. Kriesel1, Y.Kolodny1, W.L. Cairns2, B.S. Galil3, Y. Sasson4, A.V. Joshi4, A. Cangelosi5,
E.R. Blatchley III6, M.C. TenEyck7, M.D. Blacer7 & P. Brodie8
1 Arkal Filtration Systems, Israel.
2 Trojan Technologies Inc.,
London, Ontario, Canada
3 Israel Oceanographic and Limnological Research,
4 The Hebrew University of Jerusalem, Israel
5 Northeast Midwest Institute, USA
6 Purdue University, USA
7 University of Wisconsin-Superior, USA
8 Balaena Dynamics Ltd., Canada
We have designed assembled and preliminarily tested a hybrid ballast water treatment prototype that
is potentially environmentally safe and compatible.

The concept is based upon an advanced filtration stage combined with controlled generation of
oxygenated free radicals in the water. The technology is designed to cope with the following

· Inactivation of substantially more than 95% of aquatic organisms in a wide range of taxa.
· Targeting the resistant organisms, particularly those smaller than the exclusion limit of the
· Variable water quality (composition and turbidity).
· Avoidance of residual toxic chemicals in discharged water.
· Forestalling corrosion in the ballast tanks.
· Applicability to a range of ship classes.
The technology consists of the following phases:
A. An ex-situ phase (outside the ballast tanks) with the following stages:
Catalytic formation of oxygenated free radicals from precursors that include hydrogen
peroxide at minimal dose and dissolved oxygen. The reaction is enhanced by a metal oxide-
based catalyst that is suspended in the water medium.

Disc filtration using Arkal Spin Klin® patented technology. The channels within the discs
efficiently capture aggregated particles that are created at the first step, and therefore
filtration efficacy is enhanced.

UV irradiation. Direct UV disinfection is enhanced by the higher inlet UV transmittance
induced by the second stage. Disinfection efficacy is enhanced by further photocatalytic
formation of free radicals within the UV reactor in the presence of hydrogen peroxide, a high
oxygen level and the metal oxide catalyst.

The three steps described above are synergistic. The whole process is entitled the Ternary Effect.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
B. An in situ phase (within the ballast tanks):
In this phase, disinfection continues via peroxide /radical mechanisms. Generally, an increased
holding time within the ballast tank improves (to a limit) the organism inactivation.

Phases A & B, integrated together constitute a flexible multiple barrier approach to invasive species

The technology is protected by two pending patents: (IL/US/PCT).
Members of the Consortium
Arkal Filtration Systems: providers of water separation solutions for the agricultural and industrial
sectors. Manufacturers of the unique disc filter system (Spin-Klin®). Typical performance of the latter
is displayed in Figure 1.
Trojan Technologies: providers of systems for UV disinfection and chemical contaminant treatment
of wastewaters, potable waters and industrial waters using UV and advanced oxidation approaches.
NEMW Institute: expertise in the problems and applications of different ballast water treatment
technologies, and closely interfaced with the shipping industry, academic community and
regulatory community that will be making decisions on invasive species control and the selection
criteria that must be met by acceptable technologies.
Bella Galil is a marine biologist with the National Institute of Oceanography, Israel Oceanographic
and Limnological Research. Galil participated in the pioneering studies testing the bioefficacy of
ballast water exchange. She serves as a consultant to MOT.
Yoel Sasson is a Professor of applied chemistry at the Hebrew University of Jerusalem and
consultant to several chemical process industries, and specializes in homogeneous and heterogeneous
catalytic processes R&D.
E.R. Blatchley III is a Professor of Environmental Engineering in the School of Civil Engineering at
Purdue University. He conducts research in the area of physico/chemical processes of environmental
engineering, with particular emphasis on the dynamic behavior of disinfection systems used in the
treatment of water.
Lake Superior Research Institute (Mary Balcer and Matthew TenEyck) are scientists at the
University of Wisconsin-Superior with expertise in aquatic ecology, zooplankton taxonomy and
ecology, and environmental toxicology. They are involved in dose response and residual toxicity tests.
Balaena Dynamics (Paul Brodie): marine biology expertise and originator of closed loop concepts
for ballast water treatment.
Research on process optimization (Ternary Effect) is being conducted by Arkal Filtration System on a
pilot in Israel with consultancy in physics and chemistry from Professor Yoel Sasson and consultancy
in marine biology from Dr. Bella Galil. Research on process optimization (oxidation-disinfection) is
being conducted in parallel with the NEMW Institute and Professor Blatchley using equipment and
guidance from Trojan, Arkal and Yoel Sasson. Balaena Dynamics is assessing optimization of the
closed-loop option both for use with the new integrated oxidation-filtration-disinfection process and
as a means to minimize sedimentation buildup in the ballast tanks.

Kreisal: The ternary effect for ballast water treatment
The transport of potentially invasive species in ballast water is considered a major threat to the
environment, the economy and human health (Ruiz, 2000; Anil, 2002; Elliot, 2003; Topfer, 2002).
Given the safety and efficacy concerns expressed about Ballast Water Exchange (BWE) on the high
seas, there is a strong incentive to explore various methodologies of Ballast Water Treatment (BWT).
A self-contained BWT process allows independence of seas state/ice pack, etc; independence of ship
delays and independence of onshore work disruption. Given a probable 20-25 year phase-out of
existing vessels, retrofitting of validated BWT systems becomes an essential consideration in addition
to designing BWT systems for new ship outfitting (Champ, 2002).
The above described consortium of collaborators has been formed to configure, design, optimize and
validate a new Ballast Water Treatment (BWT) process that is compatible with a set of selection
criteria, design options and strategic considerations that are defined in this article. The novel process
integrates various technologies (enhanced particle aggregation and filtration, UV
disinfection/advanced oxidation and catalytic oxidation). The new platform implements a patent-
pending ex situ (outside the ballast tank) process to promote particle aggregation and improve
filtration, and provides both an ex situ and an in situ (within the ballast tank) disinfection components.
The process has the flexibility of being used in single pass, dual pass (ballasting and deballasting) and
optionally, during the ship voyage. The process is compatible with Ballast Water Exchange (BWE)
where it is deemed practicable and safe, but otherwise is totally self-contained.
The plankton (organisms in the water column that are unable to maintain their position independent of
the movement of water masses) is divided according to their size: microplankton (20-200 µm),
nanoplankton (2-20 µm) and picoplankton (0.2-2.0 µm) including bacterioplankton. It is conceivable
that one disinfection technology will not impact effectively all taxa and that a hybrid solution
incorporating several methodologies is essential. In the course of this study we have demonstrated that
the larger microplankton can be removed by a disc filter of a size between 50µ-100µ, the
microorganisms are eradicated with UV irradiation and the smaller plaktonic organism can be treated
and eliminated using advanced photochemical and catalytic oxidation processes based on hydrogen
The new process is built around two commercially available platforms:
1. the depth filtering capability of the Arkal disk filters and
2. the UV disinfection and advanced oxidation technologies of Trojan Technologies.
The process builds on Trojan's experience in UV photolysis of hydrogen peroxide to produce
hydroxyl radicals within the reactor to disinfect organisms that are resistant to disinfection by UV
alone. The process also incorporates additional proprietary oxidation processes of Arkal to enhance
the filtration efficacy and to continue disinfection within the ballast tank
Primary treatment of ballast water by filtration
Filtration is one of the separation technologies of choice for BWT by virtue of the need to separate
according to particle size and not just particle density as the density of the organisms in many cases is
close to the density of water.
The Spin-Klin® disc filtration system manufactured by Arkal Filtration Systems has several unique
features that are advantageous for ballast water treatment (see Figure 1). It has an efficient and precise
particle separation (down to size of 10 µm), there is no "break through" of the retained material, easy
and effective backwash ­ very low energy and water consumption, very low maintenance and
corrosion free construction materials. The filtration system can be configured to provide a small foot
print suitable for ship installation.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
In a recent study (Parsons, 2002) Spin Klin® system was critically tested and assessed along side other
separation technologies for ballast water filtration in a full scale experiment (340 m3/hr) in the Great
Lakes area (August-September 2001). It was concluded that the 100 µm disc filter performs better in
particle removal as compared to other filters and hydrocyclones. Typically all particles above 200 µm
were removed along with 91.4% of all the particles above 100 µm.
Agglomeration and flocculation induced by hydrogen peroxide
Initial experiments aimed at assessing the potential of hydrogen peroxide (HP) for the sterilization of
ballast water resulted in a surprising discovery. It was established that HP, at concentrations of 10-50
mg/liter performs as a flocculant, inducing the agglomeration of nanoplankton (mainly diatoms,
dinoflagellates and blue-green algae) into filterable flocs. Thus treatment of Mediterranean sea water
with HP with 10-50 seconds of retention time resulted in coagulation of an oxidized biomass which
functions as the basis for further aggregation, since the oxidized organics possess high adhesion
power. When these aggregates were examined under magnification it was observed that they contain
not only oxidized and inert biomass but also occluded inorganic particles as well as live
microorganisms. This oxidation-sorption interaction improves the efficiency of the filtration.
Consequently, the filtration is successful in removing large quantities of particles where the original
size of which, before aggregation, was small enough to escape removal. Using a 50 µm disc filter,
96% of the microplankton (size > 80 µm) was removed. Comparison of the post filtration sample with
the control showed reduction in nanoplankton population (size > 3 µm) in correlation with the
concentration of HP. Application of 30, 40 and 50 mg/liter HP resulted in 53, 56 and 76% reduction
Model I BWT Pilot Unit: Combined activity of disc filtration, uv irradiation and
hydrogen peroxide

With the objective of testing whether organisms that escape disc filtration (mainly nano- and
picoplankton) may be eradicated by exposure to UV irradiation in the presence of hydrogen peroxide,
we designed and built a 10 m3/hr demonstration unit. The pilot unit was placed and tested on a towing
dock at the port of Hadera on the Mediterranean coast of Israel (Figure 2). The pilot was operated
from April to June 2001 and processed more than 3000 tons of sea water. The apparatus contained
two alternating serpentine pipe reactors (2.2 and 20 meter long, residence time 5 and 48 sec.
respectively). HP (30%) was continuously injected into the flowing sea water to generate a 20-50
mg/liter concentration. The treated water was transferred to a disc filter (single or dual) with filtration
degrees ranging between 20-100 micron, followed by a UV irradiation unit (several types of UV
technologies using medium or low pressure lamps were tested) at UV doses ranging between 33-200
The planktonic organisms in the treated and untreated samples were studied. In the untreated samples
eleven microplankton taxa (at different taxonomic levels) were identified. The most common taxa
were the foraminiferans and crustaceans. Comparing the post HP/filtration/UV sample with the
control, a 93% reduction in abundance of the microplankton (size > 80 µm) was observed. Six
nanoplankton taxa (at different taxonomic levels) were identified. The most numerous taxa were
diatoms and Ebriida. Of the nanoplankton specimens (size > 3 µm) 35-42% were removed utilizing a
single disc filter unit, and 61-62% were removed with the dual filter system in the HP/filtration/UV
The microbial diversity in the treated and untreated sea water samples was studied by Professor
Norbert Hulsmann of the Free University of Berlin. He concluded that the treated samples still
contained particles of organismic origin (diatom frustules, dinoflagellate cell walls, loricae of
phytoflagellates up to about 50 µm) with putative plasmatic remnants as well as particles of unknown
origin (mainly fibers with a length of 100 µm and more). However, no living cells could be detected
in the samples, neither after addition of yeast cells as food organisms to sub-samples, nor after a

Kreisal: The ternary effect for ballast water treatment
period of more than three weeks of inspection. In all the treated samples, the development of biofilm
was inhibited or strongly suppressed, indicating quasi-sterile conditions. Addition of living cells from
an untreated control sample to dishes with treated seawater led to moderate biocidal effects mainly for
flagellates and heliozoans, but not for naked amoebae. The controls (untreated and sieved crude
material) showed the normal picture of a moderate microbial diversity typical for marine water of
oligotrophic origin.
We concluded that Model I system was quite efficient in removing the larger microplankton (size >
80 µm), but only moderately effective in removing the nanoplankton (size > 3 µm). Conclusions of
this phase prompted us to explore the potential of catalytic activation of hydrogen peroxide in seeking
higher disinfection activity for the eradication of the smaller nanoplankton and picoplankton.
Advanced Oxidation Processes (AOP) for ballast water treatment
Reactive oxygen species (ROS) such as superoxide radical anion, hydroxyl radical, perhydroxyl
radical, singlet oxygen or ozone are potentially highly appealing disinfection agents for various water
treatment applications. The most potent oxidant in the above list is the hydroxyl radical with oxidation
potential of 2.8V that is second only to elemental fluorine (3.0V). Hydroxyl radical and other ROS are
devastating to various constituents of the living cell, mainly to membrane lipids, proteins and DNA.
Remarkable recent ROS life science applications are in photodynamic therapy (Lane, 2003) and in
crop protection (Heitz, 1995).
An area where hydroxyl radical intermediate is the ordinary mode of action is in industrial effluent
management, particularly of toxic streams that are impervious to standard aerobic or anaerobic
biological treatment. The widespread methodology for generation of hydroxyl radicals is via the
Fenton chemistry which is based on diluted hydrogen peroxide and ferrous ion (Neyens, 2003):
Basic Fenton chemistry cannot be considered for BWT as it requires acidic conditions (pH < 4).
Improved technologies which have emerged in the last decade under the general term AOP
(Advanced Oxidation Processes) integrate and facilitate Fenton reaction with UV-VIS irradiation,
special catalysts, ozone or oxygen. Another novel approach is photocatalysis based on semiconductors
such as titanium dioxide which generates ROS from water and oxygen. The main advantage of ROS
as chemical reagents is their inherent ability to induce a destructive chain process, in the presence of
an organic material, which utilize dissolved oxygen in the propagation step and hence is essentially
self sustained.
In an early paper Waites et al. have demonstrated the synergistic effect of UV and hydrogen peroxide
in destruction of bacterial spores (Waites, 1988). However, no report on disinfection of drinking or
recycled water using catalytic AOP has been published hitherto.
In a series of experiments aimed at assessment of AOP as a core technology for ballast water
treatment (May-October 2002) we have utilized salicylic acid (SA) as a chemical scavenger to
monitor the rate of formation of hydroxyl radicals. In the presence of the latter, salicylic acid is
swiftly oxidized to a mixture of 2,3- and 2,5-dihydroxybenzoic acid which can be assayed by high
pressure liquid chromatography (HPLC) using a UV detector. The degree of conversion of SA is
proportional to the number of hydroxyl radical generated in a given system.
We examined several AOPs for the oxidation of salicylic acid in Mediterranean Sea water under
irradiation using a 150W medium pressure UV lamp, equipped with a quartz sleeve (TQ-150, Heraeus
Nobel Light Ltd.) with calculated UVC average intensities of 50-180 mW/cm2 in the 1-4 litre reaction
vessels used.
Our results clearly show that medium-pressure UV irradiation of sea water results in generation of
hydroxyl radicals. This amount was strongly affected by the presence of oxygen and/or hydrogen
peroxide (HP). Thus, irradiation of sea water containing 250 mg/liter of SA (neutralized by NaOH to

2nd International Ballast Water Treatment R&D Symposium: Proceedings
pH=7.9) and saturated with argon gas resulted in 6.3% conversion of SA after 90 minutes. The same
experiment done under air gave 8.4% conversion. Addition of 10 mg/liter of HP, with otherwise
identical conditions, raised the conversion to 11.6% and saturation with oxygen increased it to 15.1%.
Combined addition of 10 ppm HP and saturation with oxygen boosted the conversion to 20.3%.
We were quite astonished to realize that none of the previously proven metal or metal oxide catalysts
as well as other metal salts or oxides that we have tested generated any detectable superfluous activity
under irradiation. Thus salts and oxides of Fe, Ni, Co. Ag, Ru, Pt, Cu, W and V all failed to show any
synergic catalytic effect with the UV irradiation. TiO2 demonstrated some activity when irradiated
with wavelength above 300 nm but was practically ineffective when exposed to the complete
spectrum of the medium pressure lamp.
The molar extinction coefficient of hydrogen peroxide is relatively low (18.6 dm3. mol-1.cm-1 at
254nm). Consequently, only a small fraction of incident light is actually exploited. The rate of
photolysis of aqueous hydrogen peroxide has been found to be pH dependent and to increase with
higher alkalinity. This may be primarily due to the higher molar absorption coefficient of the peroxide
anion, which at 254nm is 240 dm3.mol-1.cm-1 (Legrini, 1993) These essentials prompt us to examine
the potential role of basic catalysts on the behavior of the system. Indeed, at pH=9.5 and particularly
at pH=10.0 the measured rate of SA oxidation using the UV/H2O2/O2 system was 100 and 140%
respectively, faster than the rate at pH=8.0 (note however that HP is intrinsically unstable at basic
Altering the pH of the ballast water in the course of the treatment is not viable, so we have envisaged
the application of a non-soluble solid base catalyst which would display a surface basicity without
affecting the overall pH of the processed sea water. The obvious material that drew our attention was
magnesium oxide. The latter is a natural refractory mineral (periclase) and an industrial product
(magnesia) with numerous commercial (including pharmaceutical) applications. MgO is a strong solid
base with remarkable surface base strength of +26.5 > H_ > +22.3 (Higuchi et al. 1993) and is only
very slightly soluble in water (0.6 mg/100 ml).
We used a magnesium oxide sample supplied by Aldrich with surface area (BET) of 11.4 m2/gr, bulk
density of 3.58 gr/cm3 and micropore volume of 0.000973 cm3/g.
When a slurry of 20 mg/liter of MgO in sea water (0.8 liter) containing 250 mg/liter of SA was
irradiated with a medium pressure UV lamp (intensity of 180 mW/cm2) conversion of 10.5% was
obtained after 90 minutes (compared with 8.4% in the absence of MgO). The conversion increased to
15.4% when 10 mg/liter of HP was added to the above slurry and to 27.3% when a continuous stream
of oxygen (50 ml/min) was introduced together with MgO and HP in the above capacity. These
experiments are presented in Figure 3.
Although the major rationale to apply MgO stems from its basic properties we attribute part of the
synergic effect observed in the above experiments to the unique characteristic of magnesium oxide to
stabilize ROS on its surface. This attribute was established by Giamello (1993) who showed that
hydroxyl radical and other ROS formed on the surface of the MgO upon contact with HP are stable up
to a temperature of 200°C.
Application of MgO as a component in a hybrid ballast water treatment is particularly appealing due
to several other beneficial attributes of this material which are described below:
· Recent studies disclose the unique bactericidal characteristic of MgO particularly when
fabricated as nano-particles (Sawai, 2000, Stoimenov, 2002). This trait probably stems from
surface ROS (Sawai, 1996).
· MgO was advocated as a scavenger of hydrogen sulfide in wastewater systems. (Higgins,
2003). This is advantageous for ballast water tanks where anaerobic sulfate reducing bacteria
(SRB) are abundant (Parker, 1996).

Kreisal: The ternary effect for ballast water treatment
· Corrosion inhibitor: Due to its neutralizing properties magnesia is used to neutralize acidity in
various boilers and water treatment facilities.
· Coagulant and flocculant: magnesium ions and oxide are used for coagulation-flocculation of
biological material in industrial and municipal waste water treatment (Semerjian 2003,
Hughes 2002).
Effects of HP, MgO and UV radiation on rotifers
Ultraviolet (UV) radiation is known to be effective for inactivation of waterborne microorganisms that
are of concern in conventional (potable) water and municipal wastewater treatment operations,
including many bacteria, viruses, and protozoa. However, among higher life forms, the effects of UV
irradiation on waterborne organisms are not as well defined. For ballast water treatment operations,
available evidence suggests that these higher life forms, which are generally larger than the
microorganisms listed above, may be resistant to UV-based treatment technologies.
To address this issue, experiments were conducted to characterize the responses of relevant
waterborne organisms (February 2003 to date). The focus of experiments described herein is on
freshwater rotifers. Organisms were subjected to UV irradiation in the presence of hydrogen peroxide
(H2O2) and magnesium oxide (MgO). H2O2 and MgO were added to the solution because they have
the potential to generate oxygenated radical species; both compounds have been shown to have
antimicrobial characteristics as well. It was hypothesized that these compounds (and radicals formed
in their presence) would augment the UV-based inactivation of rotifers in that they provided a
different mechanism of stress on the organisms. Furthermore, there is evidence to suggest that these
compounds and UV radiation may act in a synergistic manner, such that the antimicrobial responses
of the combined treatment would be greater than the responses attributable to either UV radiation or
the chemicals alone.
Materials and methods
Two freshwater rotifers (Philodinia sp. and Brachionus calyciflorus) were selected for study. These
organisms are representative of the two major classes of rotifers found in freshwater communities and
the Great Lakes. Brachionus calyciflorus is a loricate member (containing a stiffened cuticular body
wall composed of scleroprotein and glycoprotein) member of the class Monogonota while Philodina
is an illoricate member of the class Bdelloida. A hypothesis of these experiments was that the
presence/absence of a lorica would affect the sensitivity of the organism relative to UV irradiation and
chemical treatments. Brachionus calyciflorus are commonly used as a test specimen in toxicological
studies and has been cited in many studies.
Rotifer cultures were maintained in 2.0 L glass- flasks containing Lake Superior water that had been
passed through a 20 µm nylon mesh screen and allowed to rest for one week prior to use in the rotifer
cultures. Approximately 3.0 mL of Selenastrum capricornutum algae (108 cells/mL) was added three
times per week along with 0.5 mL of Roti-rich©, a yeast based food. Densities were checked weekly
and maintained at approximately 90-100 rotifers per mL of water. Gentle aeration (1-3 bubbles/sec)
was used to facilitate gas exchange and maintain adequate dissolved oxygen concentration.
Test organisms were subjected to UV irradiation in aqueous solutions containing H2O2 and MgO at
concentrations of 20 and 10 mg/L respectively. UV irradiation was accomplished using a large-
diameter collimated-beam. The source of radiation is two low-pressure, high-output mercury amalgam
lamps, which provide essentially monochromatic output at a characteristic wavelength of 254 nm. The
device produced a 15.5-cm diameter beam of radiation at an incident intensity of approximately 1.0
mW/cm2. The beam produced by this device was collimated and spatially uniform.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The background matrix for all rotifer exposures was a laboratory water supply taken from the
municipal drinking water supply for the city of Superior, WI. The municipal drinking water supply in
Superior, WI is chlorinated prior to distribution. The water used in these experiments was further
treated prior to use. First, the water was passed through an activated carbon column to remove
residual chlorine and some residual organics. This was followed by passage though a weak acid cation
exchange column (Amberlite® DP-1) for removal of iron and other soluble metals. In-house treatment
concluded with the addition of sodium sulfite to ensure complete removal of residual chlorine.
Stock solutions H2O2 and MgO were prepared at concentrations of 20 mg/L and 10 mg/L respectively.
This solution was then transferred to a 15-cm diameter Petri dish for UV exposure under the
collimated beam. Approximately 150 rotifers were added per dish. Suspensions were subjected to a
UV dose of 200 mJ/cm2; delivered dose was defined as the product of the depth-averaged radiation
intensity and the period of exposure. All exposures were replicated (n=5) to allow statistical
evaluation of the resulting data.
Following UV irradiation, test solutions were transferred to an Erlenmeyer flask for incubation. Petri
dishes were rinsed several times with appropriate exposed test solutions to promote transfer of
rotifers. Flasks were incubated at a 16:8 hour light:dark cycle at a temperature of approximately
23±2°C. During incubation, organisms were fed 250 µL of algae (Selenastrum, 108 cells/mL) and
250 µL yeast-trout chow-cereal (YTC, total suspended solids = 1720 mg/L). Dissolved oxygen was
maintained at near saturation conditions by continuous, gentle bubbling of air into the suspensions.
Subsamples (25 mL) were collected from each flask for microscopic examination of the condition of
the rotifers after 24, 96, and 168 hours of incubation. Organisms in each sample were classified as
alive or dead based on observed responses of swimming and internal movement of body parts.
Results and discussion
The survival of Philodina following UV irradiation in the aqueous mixtures described above is
illustrated in Figure 4. The number of live organisms may have been slightly underestimated at time
24 and 96 hours due to the tendency of Philodinia sp. to adhere to the sides of the culture flask. At the
end of the recovery period (168 hours), the flask was rinsed to ensure removal of all organisms.
During the period of incubation, the control (unexposed) rotifers reproduced, causing an increase in
the population size. In contrast, the treated samples showed a continuous decline in live numbers over
the period of incubation. While it is possible that reproduction could have taken place, any
reproductive activity in the treated samples was substantially suppressed relative to the controls. Live
numbers of Philodina decreased by more than one order of magnitude over the period of incubation,
as compared with initial numbers. When compared with the number of live organisms in the control,
treatment accomplished roughly 2 log10 units decrease in the concentration of live Philodina from
roughly 10/mL to 0.1/mL.
The survival of Brachionus calyciflorus following UV irradiation in the aqueous mixtures described
above is illustrated in Figure 5. As a loricate rotifer, it was anticipated that Brachionus would be
somewhat more resistant to UV irradiation and the mixed oxidants provided by hydrogen peroxide
and magnesium oxide than the illoricate rotifer Philodinia. The data presented in Figure 5 do not
support this hypothesis. In particular, these data indicate roughly 2 log10 units of inactivation among
the Brachionus in samples that had incubated for 24 hours. Continued incubation of the control
revealed evidence of Brachionus reproduction.
At 96 hours, the average density in the controls peaked at roughly 20 rotifers/mL. Due to
overcrowding, the population then declined to roughly 0.3 rotifers/mL at 68 hours. In the treated
samples, rotifer density remained low through out the recovery period, averaging less than 0.1

Kreisal: The ternary effect for ballast water treatment
BWT Model II prototype device integrating the ternary effect
The three components (filtration, irradiation and catalysis) were incorporated into an integrated Model
II continuous pipe-reactor pilot unit (10m3/hr., Figure 6). The unit was erected on a dock at Hadera
port on the Mediterranean Eastern coast (May 2003 to date).
Sea water is pumped, at a programmed rate, into a first stage reactor into which a measured amount of
hydrogen peroxide and slurry of magnesium oxide (both at a rate resulting in final concentration of 5-
30 ppm) are continuously injected. Air is pumped into the system via an injector at a predetermined
rate to keep the sea water at saturation with oxygen. The mixture is then transferred into a battery of
disc filters followed by a photoreactor (with optional additional injectors of HP at this stage). The
treated water is discharged into the sea. The system is equipped with controllers and monitors of flow
rate for the sea water feed, the hydrogen peroxide concentrated solution (30%), the MgO slurry in
water (30-40%) and air. The system also continuously monitors and records the inlet and outlet pH,
oxygen concentration, redox potential, water turbidity, temperature and irradiation intensity. There is
also a continuous monitoring of the filtration unit delta pressure and backflush regime. Samples of
intake and outflow are withdrawn at regular intervals and analyzed for hydrogen peroxide
concentration and the quality of water (total suspended solids, volatile suspended solids, particle size
distribution and total organic carbon) the material accumulating at the filter is also sampled and
scrutinized chemically and biologically at given intervals.
The diversity of the taxa will also be closely examined.
The issue of corrosion will be closely followed in cooperation with a classification company at
different retention times from hours to several months.
While this manuscript was being written (June 2003) the depicted pilot plant unit was in operation.
Anil, A.C., Venkat, K., Sawant, S.S., Dileepkumar, M., Dhargalkar, V.K., Ramaiah, N., Harkantra,
S.N. & Ansari, Z.A., 2002. Marine bioinvasion: Concern for ecology and shipping. Current Science,
83, pp. 214-218.
Champ, M.A. 2002. Marine testing board for certification of ballast water treating technologies.
Marine Poll. Bull. 44, pp. 1327-1335.
Eliot, M. 2003. Biological pollutants and biological pollution - an increasing cause for concern.
Marine Poll. Bull. 46, pp. 275-280.
Giamello, E., Calosso, l., Fubini, B. & Geobaldo,F. 1993. Evidence for stable hydroxyl radicals and
other oxygen radical species generated by interaction of hydrogen peroxide and magnesium oxide. J.
Phys. Chem.
97, pp. 5735-5740.
Heitz, J.R. & Downum, K.R. (Eds.) 1995. Light activated pest control ACS Symposium Series 616.
American Chemical Society, Washington DC.
Higuchi, K., Onaka, N. & Izumi, Y. 1993. Solid acid and base catalyzed cyanosilylation of carbonyl
compounds with cyanotrimethylsilane. Bull. Chem. Soc. Jpn., 66, pp. 2016-2032.
Hughes, K. 2002. Composition of insoluble magnesium containing minerals for use in fluid filtration.
WO 02/076577.
Lane, N. 2003. New light on medicine, Scientific American, January 26-33.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Legrini, O., Oliveros, E. & Braun, A.M. 1993. Photochemical processes for water treatment. Chem.
,93, pp. 671-698.
Neyens, E. & Baeyens, J. 2003. A review of Fenton's peroxidation as an advanced oxidation
technique. J. Hazard. Mater., 98, pp. 33-50.
Parker, S. 1996. Is there an SRB in your tank? Motor Ship, March, 75, 57.
Parsons, M.G. & Harkins R.W. 2002. Full scale particle removal performance of three types of
mechanical separation devices for the primary treatment of ballast water. Marine Technology, 39, pp.
Ruiz G.M., Rawlings T.K., Dobbs F.C., Drake L.A., Mullady T., Huq A. & Colwell R.R. 2000.
Global spread of microorganisms by ships - Ballast water discharged from vessels harbours a cocktail
of potential pathogens. Nature 408, pp. 49-50.
Sawai, J., Kawada, E., Kanou, F., Igarashi, H., Hashimoto, A., Kokugan, T. & Shimizu, M. 1996.
Detection of active oxygen generated from ceramic powders having antibacterial activity. J. Chem.
Eng. Jpn.
29, pp. 627-633.
Sawai, J., Kojima, H., Igarashi, H., Hashimoto, A., Shoji, S., Sawaki, T., Hakoda, A., Kawada, E.,
Kokugan, T. & Shimizu, M. 2000. Antibacterial characteristics of magnesium oxide powder, World J.
Microbiol. & Biotech.
16, pp. 187-194.
Semerjian, L. & Ayoub, G.M. 2003. High pH coagulation-flocculation in wastewater treatment. Adv.
Environ. Res.
, 7, 389-403.
Stoimenov, P.K., Klinger, R.L., Marchin, G.L. & Klabunde, K.J. 2002. Metal oxide nanoparticles as
bactericidal agents. Langmuir, 18, pp. 6679-6686.
Topfer, K. UNEP, Globallast 8, p.2.
Waites, W.M. Harding, S.E., Fowler, D.R., Jones, S.H., Shaw, D. & Martin, M. 1988. The destruction
of spores of Bacillus subtilis by the combined effects of hydrogen peroxide and ultraviolet light. Lett.
Appl. Microbiol.
7, pp. 139-140.

Kreisal: The ternary effect for ballast water treatment
Figure 1. Performance of Arkal Spin-Klin ® 130µm Disc Filter.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 2. Model I BWT Pilot Unit-General view.
Figure 3. Synergistic Effect of MgO/H2O2/O2 System in Sea Water- Conversion of salicylic acid as function of
time. Reaction Conditions: salicylic acid 250 ppm, H2O2 10 ppm, MgO 20 ppm, O2 50 cm3/min, UV medium
pressure, quartz sleeve, Volume = 800 ml, irradiation intensity = 180 mW/cm2.

Kreisal: The ternary effect for ballast water treatment
Figure 4. Measured responses of Philodina to UV irradiation (dose = 200 mJ/cm2), hydrogen peroxide (20 mg/L)
and magnesium oxide (10 mg/L). Error bars represent the standard deviation among 5 replicate measurements.
Figure 5. Measured responses of Brachionus calyciflorus to UV irradiation (dose = 200 mJ/cm2), hydrogen
peroxide (20 mg/L) and magnesium oxide (10 mg/L). Error bars represent the standard deviation among 5
replicate measurements. Note that at t=24 and 168 hours no error bars are indicated for the treatment because
no live organisms were found in any of the samples; the values indicated on the graph for these conditions
represent the limit of detection for this method (< 1 organism per 25 mL).

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 6. Layout of Model II BWT Pilot Plant prototype-general sketch.

Progress report on the `Special Pipe System' as a
potential mechanical treatment for ballast water
T. Kikuchi1, K. Yoshida1, S. Kino1 & Y. Fukuyo2
1The Japan Association of Marine Safety,
kikuti@oak.ocn.ne.jp, yoshida@lasc.co.jp,
2Asian Natural Environmental Science Center,
University of Tokyo
Name of project
The project "Research and Development of the Special Pipe System for Ballast Water Treatment"
conducted by the Japan Association of Marine Safety under the sponsorship by the Nippon
Foundation has two components: 1) improvement of the special pipe system to achieve better
effectiveness in the termination of zooplankton and phytoplankton, and 2) development of the
procedure and standard for evaluation of the effectiveness. This paper describes the first component,
and the second one is also explained in another article recorded in the same proceedings.
Treatment options being researched
The instrument designed in the special pipe system uses the options that can be categorized in a
mechanical treatment, because it applies shear stress and cavitations generated in the instrument for
termination of organisms in ballast water. During the development stage of the prototype pipe system,
which was reported verbally at the 1st GloBallast Symposium held more than two years before,
injection of ozone into ballast water before a passage of the pipe was tried to increase termination
efficacy. But mixture of such chemicals has not been applied to the new system, mostly because of the
difficulty in the installation of an instrument for provision of chemicals to the system.
Timeframe of the project
The project has three phases, commencing from April 1999.
Phase 1: 1999-2000
Basic research of the special pipe system with and without addition of ozone to the system
Phase 2: 2001-2002
On harbor testing of the improved special pipe systems
Phase 3: 2003 (in planning and to be carried out before March, 2004)
On board testing of the improved special pipe systems
Aims and objectives of the project
The objective of this study is to develop a ballast water treatment system to terminate and eliminate
harmful aquatic organisms contaminated in ballast water with special attention to criteria related to
safety of ship and crew, practicability in terms of operational complexity and installation on board
ships, cost effectiveness, and consequential environment impacts in addition to the effectiveness of

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Research methods test protocols and experimental design
The prototype special pipe system was designed to use shear stress to terminate planktonic organisms.
The potential was high, as reported at MEPC 44 in 2000, and verbally at the 1st International Ballast
Water Treatment Symposium (2001, London) and at the First International Conference on Ballast
Water Management (2001, Singapore). This structure was, however, not suitable for practical use,
because its pressure loss in passing water was high and needed higher pressure in a pipe with a larger
diameter. The higher pressure could not cause higher damage to organisms in the pipe.
Then the special pipe was re-designed with a unit generating shear stress and cavitations. Comparison of
effectiveness between the former and the developed special pipe systems was made to ascertain the higher
level of effect on marine organisms and the smaller pressure loss in the case of developed one.
Evaluation of effectiveness of the prototype special pipe
The analysis of the effectiveness of the prototype special pipe system was conducted in laboratory
with and without adding ozone produced by an ozonizer using natural seawater collected in a harbor
area at Imari Bay in Kyushu Island, western Japan.
The inner diameter of the special pipe used for the experiments was 40 mm. The seawater flow rate
was 20 m3/hr. The concentration of ozone as oxidant in sea water was 1mg/L, when injected.
Evaluation of effectiveness of the improved special pipe
Termination efficacy of the improved special pipe system was analyzed by using the system installed
in the harbor with natural seawater taken in at the harbor area at Imari Bay in Kyushu Island, western
Japan. The experiment flow is shown in the figure 1. Figure 2 shows the appearance of the main part
of the improved special pipe. The inner diameter of the pipe used for the experiments is 100 mm. Two
different flow rates of seawater, 115 m3/hr and 150 m3/hr were applied at the experiments.
In case of 115 m3/hr flow rate, the quantification of live phytoplankton and zooplankton was carried
out 5 times using method described below, and an average individual number of live organisms was
calculated by subdividing all organisms into 4 different size range groups; smaller than 20 um,
between 20 and 50 um, between 50 and 100 um, and larger than 100 um. Total individual number of
phytoplankton and zooplankton was also calculated from the data of these four subgroups.
In case of 150 m3/hr flow rate, only one data set has been available for the moment, as more
experiments are now in planning. Numbers of live organisms were counted separately for those
smaller and larger than 20 um by the method described below.
Measurement of the termination rate by quantification of live organisms
The effectiveness of the special pipes was measured by the termination rate of phytoplankton and
zooplankton, comparing the number of live organisms in initial seawater and treated seawater after
passage of the pipes. Dead or live of the organisms in the water samples was judged based on the
change of appearance, i.e. shape and color, of individual phytoplankton and zooplankton. Examples of
the normal and terminated phytoplankton and zooplankton are shown in Figure 3. Quantification was
made by counting live organisms in one ml portion of water samples taken onto a Sedgewick-Rafter
chamber under a regular compound microscope.
Preparation of seawaters samples for microscopic observation was different between organisms larger
than 20 um and the rest (smaller than that). The former was observed after concentrating the seawater
samples 1,000 times using 20 um plankton net cloth, because individual number larger than 20 um
was not high. On the other hand, the latter was observed without concentration.

Kikuchi: Progress report on the `Special Pipe System'
Relationship between flow rate and termination rate of zooplankton
Termination efficacy of the improved special pipe system in relation to flow rate was analyzed by
using the system installed at Imari Bay in Kyushu Island, western Japan. The inner diameter of the
pipe used for the experiments is 50 mm. Termination rate was calculated using zooplankton larger
than 20 um as test organisms, and quantification of individual zooplankton was made three times at
three flow rate, 10.5, 16 and 21 m3/hr.
Effectiveness of the prototype special pipe
Termination rates of phytoplankton and zooplankton with and without injection of ozone are shown in
Table 1. One-passage treatment gave an effectiveness of about 55% of phytoplankton and about 65%
of zooplankton and they increased to about 99 and 89%, respectively, by injecting ozone.
Effectiveness of the improved special pipe
The improved special pipe system can terminate about 70 and 95% of all phytoplankton and
zooplankton, respectively, in natural seawater in the case of one-passage treatment at the seawater
flow rates 115 m3/hr (Table 1). This effectiveness was obtained using 60% of the energy of the
prototype pipe. This effectiveness increased about 80 and 100%, respectively, by two-times passage
treatment, and furthermore, they reached 85 and 100%, respectively, at flow rates 150 m3/hr.
(Table 1).
Table 1. Termination rate by the prototype and improved special pipe systems
The prototype special pipe system
The improved special pipe system
Flow rate: 20m3/hr
Flow rate: 20m3/hr
Flow rate: 115m3/hr
Flow rate: 115m3/hr
Flow rate: 150m3/hr
concentration: 1mg/L
rate (%) of all
rate (%) of all
Table 2 and 3 show the details of the result with size fractions obtained at 115 and 150 m3/hr,
respectively. These results indicates that larger phytoplankton and smaller zooplankton are more
effectively terminated than the others.
Table 2. Termination rate by one and two-times passage treatments using the improved special pipe at flow rate
115 m3/hr
Cells number/ml
Termination rate (%)
Size range
After one-
After two-
Note: The values in the table are the average of 5 times of experiments

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Table 2 cont. Termination rate by one and two-times passage treatments using the improved special pipe at flow
rate 115 m3/hr
Individuals number/L
Termination rate (%)
Size range
After one-
After two-
Note: The values in the table are the average of 5 times of experiments
Table 3. Termination rate by one time passage treatments using the improved special pipe at flow rate 150 m3/hr
Cells number/ml
Termination rate (%)
Size range
After one-passage
Note: The values in the table are the data of one time experiment
Individuals number/L
Termination rate (%)
Size range
After one-passage
Note: The values in the table are the data of one time experiment
Relationship between flow rate and termination rate of zooplankton
Figure 4 shows the termination rate of zooplankton in response to flow rates in the pipe. It is obvious
that the higher flow rate produced higher treatment effectiveness.
Size of system and installation cost
The main part of the system can be installed as a part of ballast water intake line or discharge line.
The size is 1m long and 0.5m height in case of pipe having the inner diameter 100 mm. Figure 5
shows the model prepared for on board ship test which will be practiced in the latter half of this year
2003. The installation cost of the system could be estimated as 100,000 US$ per a unit, and the
running cost could be 0.01 US$/ton.
Termination efficacy of the improved special pipe system is very high. Only one time passage through
the pipe kills more than 84% of phytoplankton and almost 100% of zooplankton (Table 1 and 3). As
expected from the data using a pipe of inner diameter 50 mm shown in the Figure 4, it is not difficult
to have higher termination rate in the pipe of 100 mm inner diameter, if faster flow speed can be
applied. Multiple passage through the pipe, or application of the pipe system for both intake and
discharging waters can produce higher termination rate.
The mechanical treatment by using the improved special pipe system may be one of the treatment
options by its practicability in terms of easiness in installation on board a ship, safe in operation and
maintenance and cost performance, in addition to effectiveness in termination of organisms

Kikuchi: Progress report on the `Special Pipe System'
contaminated in ballast water. The authors have a plan of on board test in this year, and expect that the
system become available and practical in quite near future.
Japan (1999): Mixer pipe method as an alternative ballast water management technique,
MEPC 44/INF.9, 7pp.
Japan (2002): Outcome of a study on Mechanical Treatment System, MEPC 47/INF.18, 2pp.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 1. The experiment flow.
Figure 2. The improved special pipe.
Figure 3. Examples of the normal and terminated phytoplankton and zooplankton.

Kikuchi: Progress report on the `Special Pipe System'
Figure 4. The termination rate of zooplankton in response to flow rates in the pipe.
Figure 5. The special pipe system prepared for onboard ship.

Progress report on the AquaHabiStat deoxygenation
W. J. Browning Jr., J. P. Davis & W. J. Browning III
Treatment options being researched
The AquaHabiStat, or AHS, system is a patented purely mechanical ballast water treatment system
and removes dissolved oxygen (DO) from the ballast water as it is taken on board. The system utilizes
inbound ballast pumps to transfer water into a specially-constructed steel tank. In the tank, a vacuum
is drawn on the water by a vacuum pump. A centrifugal pump then removes the water from the tank
and moves it to the ships ballast hold. The marine life in the water suffocates in the ballast hold during
the voyage. After two to three days, the marine life is eradicated and the vessel may discharge the
water, which regains the oxygen on discharge and therefore leaves no ancillary environmental side
effects. The entire system of the pumps and tank connects with a controller unit that the operator may
run with a laptop PC. From the computer, the operator can turn the unit on and adjust the rate of flow,
the water level inside the tank and the vacuum force. A shipboard model may also have the
capabilities of satellite monitoring by compliance organizations who would have the ability to
remotely query the system and monitor the vacuum levels, flow rates and the time the system has run.
AHS is investigating the benefits of an additional mechanical hyper-pressurization zone prior to the
above mentioned vacuum process and/or the addition of combustion inert gasses to further lower the
oxygen content and/or to lower the pH of the ballast water being treated. AHS further wishes to
expand the number of samples subjected to phytoplankton and biomass testing, given the encouraging
results from limited ATP testing.
Time frame of the project
The full-scale prototype research in a dockside setting was completed in three 10-day time series tests
in May and June of 2000 and two 10-day time series tests in December of 2000. The results of these
tests were presented at the first IMO GloBallast conference in 2001. The prototype was able to show
an efficacy of 100% reductions of zooplankton over a two day period at an initial 72 ton per hour flow
rate. AHS is in the planning stages of "optimizing" the prototype for full scale onboard testing at
higher flow rates in the summer and fall of 2003. Additional testing will likely include the
determination of the appropriate mix of certain variables such as cost, biological efficacy, corrosion
reduction via deoxygenation and/or pH changes, the introduction of inert gasses into the ballast water
treatment process, and the safety issues or other side effects associated with these variables.
Aims and objectives of the project
The main objective of the original testing of AHS and the current plan for testing is to show that the
AHS system is both effective at reducing nearly all larval aquatic invaders in the ballast water of ships
and capable of doing so at high flow rates. While the AHS prototype has demonstrated functionality
suitable for many commercial vessels, it would like to broaden the spectrum of flow rate capabilities

Davis: Progress report on the AquaHabiStat deoxygenation system
to coordinate with the cargo discharge rates of the normal operational procedures of larger vessels
such as tankers.
The tests performed in 2000 demonstrated this effectiveness at a 72 ton per hour flow rate, using
water piped from the Chesapeake Bay into the prototype through the system and then immediately
into multiple swimming pools, which simulated ballast tanks. Biological testing was performed over a
ten day time period, to simulate a typical transatlantic voyage between Europe and the US East Coast.
Current planning and funding is in process to show that the same prototype system will maintain this
high reduction rate at an anticipated flow rate of about 300 tons per hour, onboard a vessel or barge.
AHS anticipates executing comparison tests that will allow it to gather direct data of the effects of
vacuum deoxygenation as compared directly to ballast exchange procedures. Such a direct
comparison is necessary due to the fact that the tests performed in 2000 showed that the probable
effect of biological oxygen demand rendered the control samples nearly as effective but not as reliable
as the deoxygenated "treated" samples at controlling both zooplankton and total biomass. Therefore, it
may be assumed that over a 10 day voyage, leaving ballast water alone, to become naturally hypoxic,
may be preferable to ballast exchange that actually re-aerates the ballast water.
Research methods, test protocols, experimental designs
Roger Mann, of the Virginia Institute of Marine Science (VIMS), is expected to be the Principle
Biological Investigator in our upcoming demonstration event, and his biological test plan will be
available in the near future.
Completed experiments
The following is a summary of the research methods under our original testing in 2000 under DRs.
Andrew Gordon and Anna Rule. A complete description is in the previous AHS report to IMO in
Microorganisms including zooplankton (>75 and 80 µm) as well as biomass were monitored in treated
and untreated water samples using 18 foot diameter 20,000 liter pools loosely covered with black
plastic for better simulation of a dark ballast tank. The water in the pools was monitored for water
quality (dissolved oxygen, temperature, salinity, conductivity, and pH). Biological samples were
analyzed by two independent laboratories: the Old Dominion University Department of Biological
Sciences (ODU) (at 80 µm), and the Hampton Roads Sanitation District (HRSD) (at 75 µm).
Pool sampling
ODU monitored zooplankton populations from the pool ballast tank simulation using standard 80 µm
plankton net pulls through the swimming pools. This produced samples within the 10 day time series
test that could be compared with prior work done on actual ballast water utilizing similar nets. ATP
levels were monitored, in the > 20 µm fraction and >10 µm fraction. Microscopic evaluation
conducted at ODU utilized one sample collected at the surface and one sample collected at the bottom
for each pool and each day sampled. In addition, a one-liter surface sample was collected from each
pool during every sample date and brought back to ODU for ATP extraction and analysis to determine
biomass. Samples were provided to the Hampton Roads Sanitation Department (HRSD) for
comparative work.
Flow samples were collected for HRSD, which took its samples in 20 liter carboys, all at the time that
the treated water was first put into the swimming pools. They collected 40 carboys in total, 20 treated
and 20 of control untreated water. They were stored in a dark space in ambient air temperature, and
sacrificed 4 per day; 2 treated and 2 of control. Thus no statistical corrections were necessary due to
prior sampling.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Microscopic evaluation (ODU)
All microscopic enumeration focused on the largest zooplankton found in each sample. These are
termed meso-zooplankton and are generally greater than 200 µm in length. The most abundant
zooplankton are the adult stages of cyclopoid, calanoid, and harpactacoid copepods. Equally abundant
are the larval or nauplii stages of these copepods. Copepods are categorized and tallied according to
these two stages. All other zooplankton identified were placed into the following general categories:
Barnacle nauplii, which encompass the early stage of a barnacle, polychaete larvae, ascidian,
cladoceras, crab zoea, which encompass the early stages of a crab, shrimp larvae, and unknown.
All "dead" zooplankton were first enumerated. Zooplankton not moving or slightly twitching were
considered dead or non-viable. The sample was then preserved with Lugol's iodine and all
zooplankton enumerated again. The difference in counts between the initial "dead" counts and the
total preserved counts were the numbers of zooplankton alive and moving within the sample. When a
sub-sample was used, it was preserved immediately after enumeration of dead zooplankton so that the
same water was analyzed.
ATP extraction and analysis (ODU)
Water samples from the surface of each pool were collected in a 1-liter media bottle. The samples
were taken to ODU for analysis. Each one-liter sample was divided for filtration purposes and ATP
extraction. Extracted ATP was stored in the freezer until the last day of sampling. All samples were
then analyzed as a group. Each one-liter surface water sample was divided and 500 ml filtered through
a Whatman #1 filter paper, which retained organisms >20 µm.
HRSD used similar manual microscope counts as a counting technique and the results of both labs
were similar.
Please note graphs of results following the References section. The AHS system has shown the
· The AHS system removed dissolved oxygen (DO) from ballast water to levels below 1 ppm
with a vacuum equivalent of negative 14.2 psi
· The AHS system has shown that in limited ATP testing that it eliminates approximately 80%
of all biomass above 20 microns less than three days, while certain data points (which are
averaged in the attached graph) indicated 100% reduction.
· After three days in the treated water, all larval stages that could become "nuisance species"
and other organisms 75 microns and above were eliminated.
· The system can be "ship friendly" with pumps, tanks and control devices of types normally
found aboard ships. (It can be designed to fit within existing engine room spaces.)
· The AHS system is automated and is run from a laptop computer, which is applicable to any
size of vessel and, for regulatory needs, can be monitored through electronic records that can
be read remotely.
· The system is easily adaptable to match any ship size. (The prototype was built as a one tenth
scale model of a 130,000 dwt bulk carrier with a 72 ton per hour capacity.)
Based upon our findings to date, this technology shows the potential to be one of the lowest cost and
most effective forms of ballast water treatment available within the burgeoning market for such
treatment alternatives. Being entirely mechanical, the AHS system is simple in concept and design

Davis: Progress report on the AquaHabiStat deoxygenation system
and has many benefits to the ship owner. It has been engineered and tested to comply with regulatory
standards as drafted. It is comprised of pumps, tanks, piping, valves and instruments that are suitable
for installation aboard ship and can be installed in the vessel's normal ballast intake piping system.
The system is easy to manage, so that any typical vessel crew can operate the system's controls and
the control device can be easily installed into the typical control board of the vessels cargo and ballast
plan. The AHS system is effective with one pass of the ballast water on intake through the vacuum
tank, therefore eliminating costly and confusing procedures and the need to exchange ballast water at
sea. Most importantly, the system has no harsh environmental side effects because it is not adding
harmful substances during its process.
The system uses an automatic control unit to keep the vacuum tank from overflowing or running
empty. This is run by a simple laptop computer and an off-the-shelf software program that contains a
feature of recording periodic readings of all settings to be stored for later access. If various concerned
authorities accessed these readings, they might know exactly when and where the vessel took on
ballast and to which tank. The readings of any instrumentation could be known at any given time,
such as the oxygen meter valve readings to each tank, for example. Thus, remote query may enable
appropriate authorities to know the vessel's ballast history long before the vessel reached its port of
The system is also flexible. It has the potential to be effective in all types of water, regardless of
turbidity. Other technologies, such as UV, ozone, or biocides, could be added to the process if desired.
Perhaps more important, because it is mechanical, the system can be scaled to broad ballasting flow
rate needs. The current AHS prototype has proven to be effective on land at a flow rate of 72 tons per
hour. The system can support lower or less demanding flow rates such as those for cruise vessels.
AHS is confident in its ability to achieve significantly higher flow rates to meet the needs of larger
vessels such as tankers, given the mechanical nature of the system and the theoretical ability to scale
the design to a larger size without altering the test results. The tanks of the system can be multiplied to
address high flow rates as well.
It is important to note that, under its current configuration, AHS aims to achieve environmental
soundness because absolutely nothing is added to the ballast water. AHS only removes most of the
oxygen, allows natural respiration to remove the rest, and leaves natural suffocation as the actual
control method. Any negative effects of the hypoxic ballast water on discharge areas can be corrected
either by utilizing a simple compressor to re-aerate the ballast water or pumping the ballast out above
the waterline so that it may re-aerate in the open air.
The system is of low cost up front and over the life cycle, with no expensive or hazardous chemical
agents to buy and manage. Relative to the operational cost of the ballast exchange procedure in
particular, the cost of the AHS system reduces overall ballast costs for the ship owner by cutting total
time spent ballasting in half, reducing fuel expenses, and extending the life of the ballast pumps and
the vessel itself. The broader deoxygenation process has demonstrated a reduction in the corrosion of
the ballast tanks and therefore a strong probability of a reduction of inner tank coating costs
(Tamburri, 2001).
Wilson J. Browning, Jr.
Inventor, Coordinating Investigator
Dr. Robert Ash
Eminent Scholar and Professor of Engineering, ODU: Consultant to AHS
Captain Claude Thompson
US Coast Guard (Ret.), Former Chief of the Engineering Faculty of the USCG Academy: Consultant
to AHS, Expected Principle Engineering Investigator for planned AHS experiment

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Dr. Andrew Gordon
Professor and Former Chairman of the Department of Biology, ODU: Principle Biological
Investigator of the 2000 testing procedures
Dr. Anna Rule
Chief of Laboratories, HRSD: Principle Biological Investigator of the 2000 testing procedures
Dr. Roger Mann
Professor of Marine Biology and Deputy Director of the Virginia Institute of Marine Science:
Expected Principle Biological Investigator for planned AHS Experiment
Gordon, Andrew, Old Dominion University and Rule, Anna, Hampton Roads Sanitation District
(HRSD): Biological Evaluation of the AquaHabiStat (AHS) System for Treatment of Ballast Water,
September 2000.
Gordon, Andrew, Old Dominion University: AHS Supplemental Report, January 16, 2001.
Tamburri, et al: Ballast Water Deoxygenation Can Prevent Aquatic Introductions While Reducing
Ship Corrosion
, Elsevier Science Ltd, 2001.
Clesceri, et al. (eds): Standard Methods for the Examination of Water and Wastewater. 17th edn. pp.
10-40 - 10-42. APHA-AWWA-WPCF 1989.

Davis: Progress report on the AquaHabiStat deoxygenation system
Figure 1. AHS System Design.
Figure 2. AHS Prototype.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 3. AHS System Process Diagram.
Figure 4. AHS Vessel Configuration.

Davis: Progress report on the AquaHabiStat deoxygenation system
Figure 5. Dissolved oxygen levels in treated and untreated water.
Figure 6. Test Results, Copepods.
Figure 7. Test Results, Zooplankton.

Evaluations of Venturi Oxygen StrippingTM as a ballast
water treatment to prevent aquatic invasions and ship
M. N. Tamburri, B. J. Little, G. M. Ruiz, J. S. Lee, & P. D. McNulty
Chesapeake Biological Laboratory
University of Maryland Center for Environmental
Treatment option
Gas-Based Deoxygenation
Time frame of current project
January 1, 2003 through June 30, 2004
Aims and objectives of project
Statement of Problem
Invasions by non-native aquatic species are increasingly common worldwide in coastal habitats and it
is widely accepted that ballast water is the most important vector responsible for transporting and
introducing non-native species to new biogeographic regions (Carlton and Geller 1993; Cohen and
Carlton 1998). It has proven challenging, however, to find an environmentally friendly ballast water
treatment that is effective at reducing the potential for introductions and yet also acceptable to the
shipping industry in terms of safety, time, cost, and space constraints. For instance, the offshore
exchange of ballast water is currently recommended to reduce introductions (since coastal organisms
are unlikely to invade open ocean areas, and vice versa), but the process is time-consuming (thus
costly) cannot be performed in rough sea conditions, and has limited effectiveness in some
environments and for certain vessel designs (e.g., Cooper et al. 2002; Ruiz et al. unpublished data).
Analysis of different ballast water treatments by the National Research Council (1996) suggested that
intensive filtration, thermal treatment, and biocides were the most promising options. However,
discharging warm water or water laden with biocides potentially threaten biological communities
around ports, some biocides can be dangerous to crew members, and fine filtration systems can be
expensive to install and maintain (National Research Council 1996). For any ballast water
management strategy to be successful, the shipping industry must be willing and able to comply (e.g.,
non-conflicting with other regulations such as those designed for crew safety). However, the shipping
industry does appear prepared to embrace technologies that are effective, safe, and efficient.
The acceptance and implementation of effective ballast water treatment measures would be hastened
by providing the shipping industry with economic incentives for doing so. Our previous and ongoing
work suggests that deoxygenation may be such a treatment. The economic benefit for ship owners
involves significant corrosion reduction, while simultaneously limiting the number of aquatic
organisms surviving transport in ballast tanks (Tamburri et al. 2002).
Corrosion of ballast tanks from exposure to seawater is typically destructive and costly for individual
vessels and the shipping industry as a whole. Currently painting and sacrificial anodes are used almost
exclusively as the means to prevent ballast tank corrosion, but they are expensive and time-
consuming. Investigators from Sumitomo Heavy Industries, Ltd. of Japan have therefore proposed an

Tamburri: Evaluations of Venturi Oxygen StrippingTM as a ballast water treatment
alternative corrosion prevention technique that purges oxygen from ballast tanks with nitrogen gas
(Matsuda et al. 1999). This new anticorrosion technology was derived from the basic concept that
removing oxygen from the ballast tanks will limit the oxidation of metallic structures and thus greatly
reduce the problems associated with corrosion. Our initial proof-of-principle and laboratory studies on
the effectiveness of deoxygenation to prevent the transport on non-native species and the full-scale,
field study on ballast tank corrosion demonstrated that this approach may both save the shipping
industry money on corrosion prevention while removing a large proportion of the organisms typically
found in ballast waters (Tamburri et al., 2002).
Current objectives
While results from the initial proof-of-principle studies are promising, clearly additional work is
needed to determine if deoxygenation is a feasible and effective treatment for shipboard application to
prevent aquatic invasions and tank corrosion. Our current National Oceanic and Atmospheric
Administration funded investigations are focused on a laboratory scale proof-of-technology.
Specifically, we are: (1) evaluating the Venturi Oxygen StrippingTM system developed by NEI
Treatment Systems, Inc. to optimize the deoxygenation process, (2) examining the impact of this
oxygen stripping technique on the immediate and long-term survival of natural Chesapeake Bay
planktonic organisms, and (3) quantifying corrosion rates and establishing the corrosion mechanism
under deoxygenated conditions (with particular emphasis on microbiologically influenced corrosion).
Although the effects of low oxygen or hypoxia (< 1.0 mg/l oxygen) on aquatic organisms (see reviews
by Grieshaber et al., 1994, Diaz and Rosenberg, 1995; Tamburri et al., 2002) and corrosion (e.g.,
Hardy and Bown, 1984; Lee et al., 1993a) are well documented, our current work is the first large-
scale, direct investigation of both simultaneously. Furthermore, by conducting the experiments across
different scales, we are collecting the critical data required to evaluate the feasibility of deoxygenation
as a shipboard ballast water treatment. These results will ultimately lead to a full-scale evaluation of
deoxygenation as a cost-saving ballast water treatment onboard active vessels.
Background and previous work on ballast water invasions
Sumitomo Heavy Industries found that deoxygenating ballast waters (purging with nitrogen gas to
drop oxygen levels to approximately 0.2 mg/l) decreases the rate of uniform corrosion by 90% and
represents a significant saving for ship owners when compared to other corrosion prevention
approaches currently available (approximately $80,000/year/vessel saved when compared to the
standard painting and maintenance; Matsuda et al. 1999). These results are supported by the anecdotal
observations of the Hellespont Group, who state that corrosion in ballast tanks on their tankers has
been "completely arrested" after the addition of anodes and low-sulphur inert gasses.
To test whether deoxygenation may also limit invasion, we carried out laboratory oxygen tolerance
experiments on the larvae of three widely introduced aquatic nuisance species (Australian tubeworm
Ficopomatus enigmaticus, European zebra mussel Dreissena polymorpha, and European green crab
Carcinus meanas) using oxygen levels comparable to those in the shipboard corrosion study (< 0.8
mg/l). Significant levels of mortality were found in nitrogen treated water after only two or three days
(Tamburri et al., 2002). Two separate literature reviews of oxygen tolerance for various aquatic
species further support the conclusion that few organisms will be able to withstand extended periods
of exposure to deoxygenated ballast water (Table 1). For example, by far the most abundant animals
found in ballast water are copepod crustaceans (Carlton and Geller, 1993; Smith et al., 1999) and
shallow water and estuarine species that are unable to withstand 24 hours of exposure to hypoxia (e.g.,
Roman et al., 1993; Lutz et al., 1994; Stalder and Marcus, 1997).
Small plant and algal parts (fragments, spores, and seeds) as well as single-celled phytoplankton,
protoctists, fungi, and bacteria are also often transported in ballast water. These microscopic
components of ballast water have not been thoroughly characterized. However, it appears from our
reviews that their tolerances for low oxygen environments will vary greatly. There are examples of
species that are very sensitive to hypoxic conditions (e.g., filamentous fungi, Padgett et al., 1989;
zoospores of the seaweed Undaria pinnatifida, Mountfort et al., 1999), as well as counter-examples of

2nd International Ballast Water Treatment R&D Symposium: Proceedings
species that can withstand low oxygen levels (e.g., resistant cysts of dinoflagellates, Hallegraeff,
1998). Marine bacteria, in particular, will have dramatically different responses to the conditions
created in nitrogen treated ballast tanks. While most obligate aerobic strains will be unable to grow
over extended periods of hypoxia, some facultative and obligate anaerobic bacteria may actually
thrive under the conditions found in treated ballast. We therefore conclude that ballast water
deoxygenation (maintaining hypoxia) would likely be highly effective at reducing introductions of
aquatic animals (larvae, juveniles, and adults stages) but may have mixed success at eliminating
introductions by members of other taxa.
Table 1. A representative sample of time until significant mortality (LD50, LT50, or survivorship in treatment
significantly less than control) was found for aquatic organisms held under various low oxygen concentrations.
Adapted from Tamburri et al., 2002.
2 level
Time to significant mortality
Astronotus ocellatus
24 hours
Muusze et al. 1998
fish - adults
Ophiura albida
60 hours
Vistisen and Vismann, 1997
brittle star - adults
Gammarus pseudolimnaeus
24 hours
Hoback and Barnhart, 1996
amphiod - adults
Platichthys flesus
2 hours
Tallqvist et al. 1999
fish - juveniles
Loimia medusa
72 hours
Llanso and Diaz, 1994
polychaete - adults
Meganyctiphanes norvegica
2 hours
van den Thillart et al. 1999
krill - adults
Cancer irroratus
4 hours
Vargo and Sastry, 1977
crab - larvae
Crassostrea virginica
18 hours
Widdows et al. 1989
oyster - larvae
Although other ballast water treatment options might be more comprehensively effective, they may
come at greater environmental and financial cost. For example, some biocides may be hazardous for
the crew as well as for native organisms in the vicinity of the ballast discharge (National Research
Council, 1996). Moreover, these techniques could come at a significant price for ship owners. Our
previous work suggested that widespread voluntary adoption of deoxygenation may result if the
economic benefits for controlling corrosion are demonstrated definitively and become well known.
While ballast water treatments have been controversial, raising conflicts between environmentalists
and ship owners, we felt that deoxygenation represented a working solution that should appeal to both
parties and that deserved further investigation.
Background and previous work on ballast tank corrosion
The vast majority of the world's fleet of ships, including military and commercial vessels, are
constructed of carbon steel. Steel corrodes quickly when exposed to oxygen and water. Ocean-going
vessels are particularly susceptible to corrosion, due to the accelerated corrosion rate in exposure to
salt water. Corroded steel structures on a vessel decrease seaworthiness so extensive measures are
taken to prevent corrosion and, inevitably, in repair. The cost to prevent, maintain, and repair
corrosion on individual vessels can run into the millions of dollars (e.g., $5.5 million to replace 1400
tonnes of ballast tank steel on Wind Conquest, Marine Engineering Review 1991).
One area in a ship where corrosion is of particular concern is in the ballast tanks. Prolonged exposure
of the ballast tank structure to water (often salt water) creates a condition conducive to rapid
corrosion. The cost to paint ballast tanks is typically at least $5.00 to $10.00 per square meter with the

Tamburri: Evaluations of Venturi Oxygen StrippingTM as a ballast water treatment
cost to repair corroded areas at approximately $500 per square meter (Fairplay, 1993). With large
cargo vessels and oil tankers having hundreds of thousands of square feet of ballast tank surface area,
preventing and treating corrosion is extremely costly.
Therefore, any measure for controlling aquatic invasive species in ballast tanks cannot be evaluated
without consideration of the impact on corrosion. For example, both chlorination (McCracken, 2001)
and ozonation (Andersen, 2001) of seawater are believed to exacerbate corrosion of steel. Clearly,
removal or reduction of oxygen will eliminate or reduce direct oxidation reactions related to
corrosion. However, deoxygenation could increase corrosion resulting from the activities of naturally
occurring microaerophilic, facultative or obligate anaerobic bacteria. Acid-producing bacteria (APB)
and sulfate-reducing bacteria (SRB) grow under anoxic conditions and produce corrosive metabolic
by-products (organic acids and sulfides, respectively).
The corrosion rate of carbon steel is not influenced by pH over the range of 4.5 to 9.5 in distilled and
tap waters (Boyer and Gall, 1985). Over this range, corrosion products maintain a pH of 9.5 at the
metal surface. Below pH 4.0, hydrogen evolution begins and corrosion increases dramatically.
Although it is extremely unlikely that APB will change the bulk pH of carbonate buffered seawater,
APB can reduce pH locally under colonies and produce corrosion in carbon steel (Pope, 1995).
All seawater contains 2 gm l-1 sulfate than can be reduced to sulfide by SRB in the absence of oxygen.
Reviews by Miller and Tiller (1970), Iverson (1974) and Postgate (1979) provide examples and
details of microbiologically influenced corrosion of iron and mild steel under anaerobic conditions
caused by SRB. Microbiologically influenced corrosion failures have been reported for mild steel
piping and equipment exposed in the marine environment (Sanders and Hamilton, 1986; Eidsa and
Risberg, 1986; Eashwar et al, 1990) soil (King et al, 1983; Kasahara and Kajiyama, 1986; Alanis et al,
1986; Pope et al., 1988; Dias and Bromel, 1990), oil refining industry (Winters and Badelek, 1987),
fossil fuel and nuclear power plants (Soraco et al., 1988; Licina, 1988; Pope, 1986 and 1987; Bibb,
1986) and process industries (Pacheco, 1987; Honneysett, 1985; Tatnall et al, 1981). Deoxygenation
can also result in putrefaction, anaerobic breakdown of sulfur-rich proteins, and levels of sulfides will
not be limited to the sulfate concentration in the seawater. Sulfide reacts with iron oxide, formed in
the atmosphere or in oxygenated seawater, to produce a non-tenacious iron sulfide layer that can be
removed with stress or converted back to an oxide by the introduction of oxygen. In either case the
sulfide layer is not uniformly removed or oxidized, creating adjacent anodic and cathodic regions and
aggressive corrosion.
The most corrosive operating condition is one in which carbon steel is exposed to alternating
oxygenated/deoxygenated conditions (Hardy and Bown, 1984; Lee et al, 1993a; Lee et al, 1993b).
Under constant oxygenation an oxide will form that provides corrosion resistance. Under anaerobic
conditions a sulfide layer will form and corrosion rate will decrease until oxygen is introduced. The
result of alternating operating conditions is severe pitting. Additionally, concentrations of sulfides can
produce sulfide assisted stress corrosion cracking in carbon steel. Most reported cases of SRB induced
corrosion of carbon steel in marine waters are in environments with some dissolved oxygen in the
bulk medium (Hamilton, 1986). Anaerobic conditions and sulfides form within marine biofilms at
biofilm/metal interfaces, independent of bulk oxygen concentrations. Exposure of iron sulfide
corrosion products to oxygen creates differential aeration cells and localized corrosion.
Research Methods
Optimizing Deoxygenation
A key to the success of deoxygenation as a ballast water treatment is to design and develop the most
efficient method for maintaining levels of oxygen in tanks that both kills the majority of aquatic
organisms while also reducing corrosion rates ­ below 1.0 mg/l. The deoxygenation method proposed
by Sumitomo Heavy Industries for ballast water treatment entails bubbling an inert gas into the ballast
tanks after they have been filled. The shipboard trial by Matsuda and colleagues (1999) included

2nd International Ballast Water Treatment R&D Symposium: Proceedings
vertical pipes installed into a ballast tank from which pure nitrogen gas was pumped into the water for
the "sparging" of oxygen. The tank was also sealed at the deck to permit nitrogen purging of the
headspace. This method may achieve some deoxygenation through the contact of the nitrogen bubbles
with the water, but primarily relies on diffusion of oxygen through the water surface in the tanks.
Although hypoxic conditions were achieved, the sparging and purging of oxygen took days and relied
on both the presence of a large headspace in the ballast tank filled with nitrogen gas (a free surface
condition that is typically avoided since it can destabilize the vessel as water moves within tanks) and
on large volumes of expensive inert gas. Although the basic principles are sound and experimental
results on corrosion significant, the method used by Sumitomo Heavy Industries for deoxygenation
appears to be inefficient and relatively costly to employ (approximately $3.5 million for installation
on a vessel).
Other deoxygenation methods (e.g., vacuums, horizontally placed diffuser plates, biological
processes) use techniques with varying degrees of effectiveness. However, our investigations suggest
that the most efficient way to remove oxygen from ballast water is through introducing microfine
bubbles of an inert gas as water is being pumped into the tanks. The smaller a bubble, the higher the
ratio of surface area to volume and thus the higher gas-to-water contact surface where transfer takes
place. Therefore, we have begun work with NEI Treatment Systems, Inc. to evaluate the
deoxygenation of ballast water through Venturi Oxygen StrippingTM.
Survivorship of natural planktonic organisms subjected to deoxygenated water
Dockside, mesocosm experiments are being conducted at the Chesapeake Biological Laboratory
(CBL), University of Maryland Center for Environmental Science, in Solomons, Maryland (Figure 1).
Natural seawater is pumped from one meter below the surface into 10 identical 25-gallon, airtight
fiberglass cylinders, held inside a laboratory at the end of the CBL pier. All water first passes through
a 1 cm screen (the mesh size commonly used to filter intake into ship ballast tanks) and the cylinders
are kept in the dark during the trails to mimic the light environment onboard vessels. In five control
cylinders, seawater is delivered directly from the pump. In five treated cylinders, the seawater first
passes through the Venturi Oxygen StrippingTM system. Physical conditions such as oxygen,
temperature, pH, and conductivity are monitored throughout the experiments with sensors sealed
within the cylinders. Oxygen levels in the control cylinders are always above 8.0 mg/l whereas water
in the treated cylinders enters and remains hypoxic (< 0.9 mg/l) throughout the experiments.
To examine mortality over time as a result of deoxygenation, one treated and one control cylinder are
drained completely through a bottom valve 1, 24, 48, 72, and 96 hours after filling. The treated and
control cylinder at each sampling period are then compared for abundance or mortality of three
separate planktonic community components. Zooplankton mortality is examined by sieving the entire
volume through a 50 µm screen and determining total abundance and living versus dead individuals.
The percentages of living individuals are quantified by examining reactivity or movement under a
dissecting microscope. Relative abundances of phytoplankton are analyzed by determining
chlorophyll-a concentrations using standard extractive fluorometry techniques. Subsamples are also
examined under a compound microscope to identify major algae groups. Finally, the densities of
bacterial cells in each cylinder are determined by flow cytometry.
Additionally, subsamples of abundant zooplankton (such as copepods and barnacle larvae) that are
scored as dead after the 48-hour deoxygenation treatment are being placed in aerated natural seawater
to determine their ability to recover and resume swimming after removal from hypoxic conditions.
These entire dockside/mesocosm trials are being repeated five times during the seasons when
planktonic organisms are most abundant (April through September 2003) in the Chesapeake Bay.
Rates and mechanism of corrosion under deoxygenated conditions
Laboratory experiments are underway to examine: A) how deoxygenation influences bulk water
chemistry, biofilm formation and biofilm/metal interfacial chemistry, B) if microbiologically
influenced corrosion occurs under deoxygenated conditions and if so by what mechanism, and C) the

Tamburri: Evaluations of Venturi Oxygen StrippingTM as a ballast water treatment
impact of O2 on corrosion mechanisms and rates under deoxygenated conditions. The corrosion
experiments are being conducted at the Naval Research Laboratory (NRL), Corrosion Facility, in Key
West, FL and at the NRL, Stennis Space Center, MS.
Five identical chambers were built to expose 1020 carbon steel (common ballast tank material) and
natural seawater to different conditions (Figure 2). Three chambers are alternating immersion
treatments where for two weeks the chambers are filled with water, then two weeks with gas, and this
cycle is repeated for one year. The first chamber is alternating between raw, oxygenated seawater and
air. The second chamber is alternating between natural seawater that is first deoxygenated by passing
through the Venturi Oxygen StrippingTM system and air. The third chamber is alternating between
deoxygenated water and inert gas containing only trace amounts of oxygen. The two remaining
chambers are held stagnant for one year (no cycling). One was filled with raw, oxygenated seawater
and is being left open to air while the other was filled with deoxygenated seawater and is being stored
in an anaerobic hood. The experimental design is summarized in Table 2.
Table 2. Experimental design for the one year corrosion experiment being conducted at NRL facilities.
Two weeks oxygenated water
Key West, FL
then two weeks air
Two weeks deoxygenated water
Key West, FL
then two weeks air
Two weeks deoxygenated water
Key West, FL
then two weeks inert gas
Oxygenated water open to air
Stennis, MS
no cycling
Deoxygenated water in anaerobic hood
Stennis, MS
no cycling
Samples are collected every month over one year to assess changes in dissolved and particulate water
chemistry (dissolved oxygen, dissolved organic carbon and nitrogen, particulate organic carbon and
nitrogen, bulk pH, sulfide concentration) using standard techniques. Serial dilutions are used to
determine most probable numbers of APB, SRB, general heterotrophic aerobes, and anaerobes
(Bioindustrial Technologies, Inc.).
The carbon steel coupons have been oriented in rows both horizontally and vertically in each chamber
to simulate tank bottoms and sidewalls, respectively. Triplicate samples from both containers are
removed every two months, fixed in glutaraldehyde and examined to assess the extent of biofilm
formation and corrosion morphology. Environmental scanning electron microscopy (ESEM) and
energy dispersive spectroscopy (EDS) are being used to characterize the corrosion morphology,
biofilm structure and corrosion product composition on the metal surface. Swabs made of the coupon
surface and serial dilutions are used to determine the microbial composition of the biofilm and
microelectrodes are used to make O2 profiles through the biofilms. Finally, polarization resistance and
open-circuit potential is being used to monitor electrochemistry and corrosion of the carbon steel
continuously over the one year experiment.
Optimizing deoxygenation
Evaluations of several approaches and a series of pilot studies have led to the conclusion that Venturi
Oxygen StrippingTM represents the most effective and economical method of deoxygenation for use
aboard vessels. Venturi Oxygen StrippingTM is a patent-pending rapid, in-line deoxygenation system
that mixes inert gas directly into ballast water as it is drawn into the vessel. The inert gas is produced

2nd International Ballast Water Treatment R&D Symposium: Proceedings
by combusting low-sulfur marine diesel (generating mostly nitrogen with small amounts of carbon
dioxide and only trace levels of oxygen) in a device similar to the inert gas generators commonly used
on tankers. The gas is mixed with the ballast water using a venturi injector that is installed in-line, just
down-stream of the ballast pump. The venturi injector creates a micro-fine bubble emulsion where
dissolved oxygen quickly diffuses out of the water into the gas. Because adding carbon dioxide in
solution forms both carbonic and carboxylic acid, the pH of treated water is also reduced. This system
is designed so that the same inert gas is also used to blanket all headspaces and the entire ballast tank
when empty to maintain permanent hypoxia. Continuously maintaining a deoxygenated environment
in ballast tanks appears to be a critical factor for corrosion prevention (see below).
Laboratory experiments performed under a variety of environmental conditions show that the time
until low-oxygen equilibrium condition in the water is reached is less than 10 seconds. Treated water
also reoxygenates within seconds after release from test tanks. Further design, development, and
testing by NEI Treatment Systems has found that this ballast water treatment will be simple to install,
operate, and maintain because several component parts are similar to equipment already commonly
used onboard vessels. Finally, cost analysis show that the Venturi Oxygen StrippingTM system will be
relatively inexpensive to install ($100,000 - $700,000 depending on vessel design) and operation
($15,000 - $50,000 / year). These values do not consider the significant decease in ballast tank
maintenance costs through corrosion prevention.
Survivorship of natural planktonic organisms subjected to deoxygenated water
Although the experiments on the ability of Venturi Oxygen StrippingTM to kill planktonic organisms
are still ongoing, initial results are striking. The dissolved oxygen levels and pH in the control
cylinders were between 8.18 ­ 11.01 mg/l and 7.61 ­ 8.20 respectively, whereas the dissolved oxygen
levels and pH in the treated cylinders dropped to 0.26 ­ 0.87 mg/l and 5.46 ­ 5.62 respectively. In
treated tanks, these changes to the physical environment lead to a greater than 99% mortality of
Chesapeake Bay zooplankton (copepods, barnacle larvae, polychaete larvae, cladocerans, crustacean
nauplii, bivalve larvae, and nematodes) in less than 48 hours while the majority of zooplankton
survived in the control cylinders (Figure 3). In addition to hypoxia and lowered pH, many of the
larger zooplankton (mostly copepods) also appeared to be killed instantaneously by being damaged as
they passed through the venturi injector which created large amounts of cavitation and turbulence
(Figure 4). Furthermore, no intact individuals scored as dead after 48 hours recovered after being
placed back in aerated water for 24 hours. Therefore zooplankton are not simply narcotized but are
being effectively killed.
It also appears that the Venturi Oxygen StrippingTM system may reduce the abundance of
phytoplankton (Figure 5). However, because large reductions in chlorophyll-a were also found in the
control cylinders over time, impacts of deoxygenation on phytoplankton are difficult to discern at this
point. Although additional experiments are being run, it is obvious that the abundances of algae are
generally decreasing due to the darkened test conditions (which are meant to mimic ship ballast tank
light levels) regardless of treatment.
Finally, the deoxygenated environment and relatively high organic material available (dead plankton)
after treatment with the Venturi Oxygen StrippingTM system does not appear to enhance bacterial
growth or cause blooms. Initial measurements are showing no obvious difference in bacterial
abundances in control versus treated through time (Figure 6).
Rates and mechanism of corrosion under deoxygenated conditions
Corrosion, biofilm formations, and changes to seawater chemistry as a result of deoxygenation are
relatively slow processes. Therefore, conclusions can only be drawn after the year long study is
completed. However, initial results from the IR compensated Linear Polarization Resistance analyses
(only one of the many parameters being studied) suggested that instantaneous corrosion rates are
significantly lower when the carbon steel in the alternating immersion trials are kept continuously in a

Tamburri: Evaluations of Venturi Oxygen StrippingTM as a ballast water treatment
hypoxic environment. In fact, alternating back and forth from water that is deoxygenated to air may
enhance corrosion rates.
Ballast water treatment technologies should be: 1) effective at killing potentially damaging invaders,
2) safe for shipboard crew, 3) environmentally benign, and 4) affordable for ship owners. As we have
discussed above, deoxygenation through Venturi Oxygen StrippingTM is highly effective at killing
animal invaders but may be less effective for other taxa. However, the number of individuals from
resistant taxa that do survive this treatment may be below the threshold which poses a significant
threat for the establishment of non-native populations (Williamson, 1996; Bailey et al., 2003; Drake et
al., 2003). Furthermore, because most of the components of Venturi Oxygen StrippingTM system are
already found onboard vessels and existing regulations require the measurement of ballast tank
oxygen levels prior to entry, there appear to be no major threats to crew safety. Deoxygenated water
itself is also relatively benign when discharged. Treated water will reoxygenate and mix rapidly with
receiving water in harbors (particularly if released above surface) and therefore create little danger for
native estuarine organisms, which can withstand brief reductions in oxygen levels. However, if
required, water can also be actively reoxygenated prior to release by simply adding an additional
venturi injector connected to air on the outflow piping system. Finally, this ballast water treatment
admirably meets the fourth criterion. Rather than an added expense for ship owners, it actually
represents a net saving, due to the significant decrease in corrosion.
An additional consideration when evaluating any ballast water treatment is how operational efficacy
will be measured and how compliance with regulations will be monitored. Given the results of our
work and the wealth of literature on the oxygen tolerance of aquatic organisms (Grieshaber et al.
1994, Diaz and Rosenberg 1995; Tamburri et al. 2002), determining efficacy and compliance with
future regulations may simply entail continuous measurements of dissolved oxygen levels with
perhaps only periodic biological sampling for validation.
Our fundamental goal is to provide the science necessary for the development of effective ballast
water management strategies and policies. Through rigorous laboratory and dockside/mesocosm
experiments, our work is providing the information required to evaluate the efficacy and feasibility of
deoxygenation through Venturi Oxygen StrippingTM as a ballast water treatment to prevent aquatic
invasions and will be the basis for a definitive shipboard study planned for the near future.
In summary, it appears that rapid and efficient reduction of oxygen levels in ballast water both causes
substantial mortality of a large proportion of transported organisms and minimizes ballast tank
corrosion. As such, it represents a good example of a solution that simultaneously has benefits for
marine conservation and industry.
Alanis, I., Berardo, L., De Cristofaro, N, Moina, C., and Valentini, C. 1986. In Biologically Induced
, 102, Houston, National Association Of Corrosion Engineers.
Andersen, A.B., 2001. Ballast Water Treatment by Ozonation. 1st International Ballast Water
Treatment R&D Symposium. International Maritime Organization, London 26-27 March 2001:
. GloBallast Monograph Series No. 5, IMO London.
Bailey, S.A., Duggan, I.C., van Overdijk, C.D.A., Jenkins P.T., & MacIsaac. H.J. 2003. Viability of
invertebrate diapausing eggs collected from residual ballast sediment. Limnology and Oceanography.
48: pp. 1701-1710.
Bibb, M. 1986 In Biologically Induced Corrosion, 96, Houston, National Association Of Corrosion

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Boyer, H.E. & Gall, T.L., 1985. Metals Handbook, American Society for Metals: Metals Park, OH.
Carlton, J. T. & Geller, J. B. 1993. Ecological roulette: The global transport of nonindigenous marine
organisms. Science, 261, pp. 78-82.
Cohen, A.N., and Carlton, J.T. 1998. Accelerating invasion rate in a highly invaded estuary. Science,
279, pp. 555-558.
Cooper, W.J., Dinnel, P. A., Gensemer, R. W., Herwig, R. P., Kopp, J. A., Ruiz, G. M., Sonnevil, G.,
Stubblefield, W. A., & VanderWende, E. 2002. Ozone, Seawater and Aquatic Nonindigenous
Species: testing a full-scale ozone ballast water treatment system on an American Oil Tanker. 11th
International Conference on Aquatic Invasive Species, Alexandria, VA
Dias, O.C., and Bromel, M.C., 1990. Material Performance, 29: 53.
Diaz, R. J. & Rosenberg, R. 1995. Marine benthic hypoxia: a review of its ecological effects and the
behavioral responses of benthic macrofauna. Oceanography and Marine Biology Annual Review, 33,
pp. 245-303.
Drake, J.M., Lodge, D.M. & Lewis, M. 2003. Combining allometric parameters of biological
processes with recent progress in invasion theory: Implications for ballast water standards. 12th
International Conference on Aquatic Invasive Species, Windson, Ontario, Canada.

Eashwar, M., Subramanian, G., Chandrasekaran, P. & Balakrishnan, K. 1990. Proc. Corrosion '90,
No. 120; Las Vegas, National Association Of Corrosion Engineers.
Eidsa G. & Risberg, E. 1986. Biologically Induced Corrosion, 109, Houston, National Association Of
Corrosion Engineers.
Fairplay, April 1993. Special Report: Paints and Coatings - Covering the Costs.
Grieshaber, M.K., Hardewig, I., Kreutzer, U. & Poertner, H. O. 1994. Physiological and metabolic
responses to hypoxia in invertebrates. In: Blaustein, M.P. (Ed.), Reviews of Physiology Biochemistry
and Pharmacology 125
. Springer-Verlag, New York, pp. 43-147.
Hallegraeff, G.M., 1998. Transport of toxic dinoflagellates via ships' ballast water: Bioeconomic risk
assessment and efficacy of possible ballast water management strategies. Marine Ecology Progress
Series, 168
, 297-309.
Hamilton, W.A. & Maxwell, S. 1986. Biological and Corrosion Activities of Sulphate-Reducing
Bacteria within Natural Biofilms. In Biologically Induced Corrosion, Dexter, S. C. ed., National
Association of Corrosion Engineers: Houston, TX, pp. 131- 136.
Honneysett, D.G., Van Den Bergh, W.D. and O'brien, P.F., 1985. Material Performance, 24: 34.
Hardy, J.A. & Bown, J.L., 1984. The Corrosion of Mild Steel by Biogenic Sulfide Films Exposed to
Air. Corrosion, 40, pp. 650-654.
Hoback, W.W., and Barnhart, M.C. 1996. Lethal limits and sublethal effects of hypoxia on the
amphipod Gammarus pseudolimnaeus. Journal of the North American Benthological Society, 15, pp.
Iverson, W.P. 1974. Microbial Iron Metabolism, New York, Academic Press.
Kasahara, K. & Kajiyama, F. 1986. In Biologically Induced Corrosion, 171, Houston, National
Association Of Corrosion Engineers.

Tamburri: Evaluations of Venturi Oxygen StrippingTM as a ballast water treatment
King, R.A., Skerry, B.S., Moore, D.C.A., Stott, J.F.D. & Dawson, J. L. 1986. In Biologically Induced
, 83, Houston, National Association Of Corrosion Engineers.
Lee, W.C., Lewandowski, Z., Okabe, S., Characklis, W.G., and Avci, R. 1993. Corrosion of Mild
Steel Underneath Aerobic Biofilms containing Sulfate-Reducing Bacteria Part I: At Low Dissolved
Oxygen Concentration. Biofouling 7, pp. 197-216.
Lee, W.C., Lewandowski, Z., Morrison, M., Characklis, W.G., Avci, R. & Nielsen, P. H. 1993.
Corrosion of Mild Steel Underneath Aerobic Biofilms containing Sulfate-Reducing Bacteria Part I: At
High Dissolved Oxygen Concentration, Biofouling 7, pp. 217-239.
Licina, G.C. 1988. Sourcebook For Microbiologically Influenced Corrosion In Nuclear Power Plants,
Palo Alto, Electric Power Research Institute.
Llanso, R.J., and Diaz, R.J. 1994. Tolerance to low dissolved oxygen by the tubicolous polychaete
Loimia medusa. Journal of the Marine Biological Association of the United Kingdom, 74, pp. 143-
Lutz, R.V., Marcus, N.H. & Chanton, J.P. 1994. Hatching and viability of copepod eggs at two stages
of embryological development: anoxic/hypoxic effect. Marine Biology, 119, pp. 199-204.
Marine Engineering Review, January 1991. Life Corrodes Away.
Matsuda, M., Kobayashi, S., Miyuki, H. & Yoshida, S. 1999. An anticorrosion method for ballast
tanks using nitrogen gas
. Ship and Ocean Foundation Technical Report.
McCracken, W. 2001. Ballast Water Treatment with Available Biocides, 1st International Ballast
Water Treatment R&D Symposium. International Maritime Organization, London 26-27 March 2001:
. GloBallast Monograph Series No. 5, IMO London.
Miller, J.D.A. & Tiller, A.K. 1970. In Microbial Aspects Of Metallurgy, 61, New York, Elsevier.
Mountfort, D.O., Hay, C., Dodgshun, T., Buchanan, S. & Gibbs, W. 1999. Oxygen deprivation as a
treatment for ships' ballast water: laboratory studies and evaluation. Journal of Marine Environmental
, 5 (3).
Muusze, B., Marcon, J., Van Den Thillart, G. & Almeida-Val, V. 1998. Hypoxia tolerance of Amazon
fish: respirometry and energy metabolism of the cichlid Astronotus ocellatus. Comparative
Biochemistry and Physiology
A, 120, pp. 151-156.
National Research Council, 1996. Stemming the tide: controlling introductions of nonindigenous
species by ship's ballast water
. National Academy Press, Washington, D.C.
Pacheco, A., Dishinger, T.D. & Tomlin, J. L. 1987. In Proc. Corrosion '87, No. 376, San Francisco,
National Association Of Corrosion Engineers.
Padgett, D.E., Celio, D.A., Hearth, J.H. & Hackney, C. T. 1989. Growth of filamentous fungi in a
surface-sealed woody substratum buried in salt-marsh sediments. Estuaries, 12, pp. 142-144.
Pope, D.H. 1986. A Study Of MIC In Nuclear Power Plants And A Practical Guide For
, Final Report Np-4582, Palo Alto, Ca, Electric Power Research Institute.
Pope, D.H. 1987. Microbial Corrosion In Fossil-Fired Power Plants - A Study Of Microbiologically
Influenced Corrosion And A Practical Guide For Its Treatment And Prevention
, Palo Alto, Electric
Power Research Institute.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Pope, D.H. & Morris, E.A. 1995. Some Experiences with Microbiologically Influences Corrosion of
Pipelines. Materials Performance, 34, pp. 23-28.
Pope, D.H., Zintel, T.P., Kuruvilla, A.K. & Siebert, O. W. 1988. In Proc. Corrosion 88, No. 79, St.
Louis, Mo, National Association Of Corrosion Engineers.
Postgate, J.R. 1979. The Sulphate Reducing Bacteria, Cambridge, England, Cambridge University
Roman, M.R., Gauzens, A.L., Rhinehart, W.K. & White, J.R., 1993. Effects of low oxygen waters on
Chesapeake Bay zooplankton. Limnology and Oceanography, 38, pp. 1603-1614.
Sanders, P.F. & Hamilton, W.A. 1986. In Biologically Induced Corrosion, 47, Houston, National
Association Of Corrosion Engineers.
Smith, L.D., Wonham, M.J., McCann, L.D., Ruiz, G.M., Hines, A.H. & Carlton, J. T. 1999. Invasion
pressure to a ballast-flooded estuary and an assessment of inoculant survival. Biological Invasions, 1,
pp. 67-87.
Stalder, L.C. & Marcus, N. H. 1997. Zooplankton responses to hypoxia: behavioral patterns and
survival of three species of calanoid copepods. Marine Biology, 127, pp. 599-607.
Tallqvist, M., Sandberg-Kilpi, E. & Bonsdorff, E. 1999. Juvenile flounder, Platichthys flesus (L.),
under hypoxia: effects on tolerance, ventilation rate and predation efficiency. Journal of Experimental
Marine Biology and Ecology
, 242, pp. 75-93.
Tamburri, M.N., Wasson, K. & Matsuda, M. 2002. Ballast water deoxygenation can prevent aquatic
introductions while reducing ship corrosion. Biological Conservation, 103, pp. 331-341.
Tatnall, R., Piluso, A., Stoecker, J., Schultz, R. & Kobrin, G. 1981. Material Performance, 19: p. 41.
van den Thillart, G., George, R.Y. & Stromberg, J. O. 1999. Hypoxia sensitivity and respiration of the
euphausiid crustacean Meganyctiphanes norvegica from Gullmarn Fjord, Sweden. Sarsia, 84, pp.
Vargo, S.L. & Sastry, A. N. 1977. Acute temperature and low dissolved oxygen tolerances of
brachyuran crab (Cancer irroratus) larvae. Marine Biology, 40, pp. 165-171.
Vistisen, B. & Vismann, B. 1997. Tolerance to low oxygen and sulfide in Amphiura filiformis and
Ophiura albida (Echinodermata: Ophiuroidea). Marine Biology, 128, pp. 241-246.
Widdows, J., Newell, R.I.E. & Mann, R. 1989. Effects of hypoxia and anoxia on survival, energy
metabolism, and feeding of oyster larvae (Crassostrea virginica, Gmelin). Biological Bulletin, 177,
pp. 154-166.
Williamson, M. 1996. Biological Invasions. Chapman and Hall, London.
Winters, M.A. & Badelek, P.S.C. 1987. In Proc. Corrosion 87, No. 156, San Francisco, National
Association Of Corrosion Engineers.

Tamburri: Evaluations of Venturi Oxygen StrippingTM as a ballast water treatment

Figure 1. Mesocosm experimental setup to examine the effectiveness of Venturi Oxygen StrippingTM at killing
natural planktonic organisms found in Chesapeake Bay.

Figure 2. Experimental setup to examine rates and mechanism of corrosion under the deoxygenated conditions
produced by the Venturi Oxygen StrippingTM system.

2nd International Ballast Water Treatment R&D Symposium: Proceedings

% Survival of Zooplankton
Time (hours)
Figure 3. Percent survival of natural Chesapeake Bay zooplankton (copepods, barnacle larvae, polychaete
larvae, cladocerans, crustacean nauplii, bivalve larvae, and nematodes) in control and treated (deoxygenated)
chambers after 1, 24, 48, 72, and 96 hours for the first four replicate trials of an ongoing experiment.
Figure 4. A damaged copepod (lower middle) after passing through the Venturi Oxygen StrippingTM system. In all
current trials examining the impacts of this treatment on zooplankton, the initial (after 1 hour) percent survival is 5
to 20 percent lower in treated versus control (see Figure 3) because of physical damage to larger individuals.

Tamburri: Evaluations of Venturi Oxygen StrippingTM as a ballast water treatment

Trial 2
Active Chl-a (ug/l)
Trial 1
Time (hours)
Figure 5. Active chlorophyll-a concentrations in control and treated (deoxygenated) chambers after 1, 24, 48, 72,
and 96 hours for the first two trials of an ongoing experiment.

Trial 2
Bacteria (million cells/ml)
Trial 1
Time (hours)
Figure 6. Abundance of bacterial cells in control and treated (deoxygenated) chambers after 1, 24, 48, 72, and
96 hours for the first two trials of an ongoing experiment.

Ballast water treatment by de-oxygenation with
elevated CO2 for a shipboard installation ­
a potentially affordable solution
M. Husain1, H. Felbeck2, R. Apple1, D. Altshuller1 & C. Quirmbach1
1MH Systems, Inc., USA
2Scripps Institution of Oceanography,
University of California
Treatment options being researched
It is estimated that 21 billion gallons of ballast taken on in foreign ports are discharged by commercial
vessels annually in the waters of the United States (Carlton et al. 1993). Specifically, ballast water
transport is a major vector for the introduction of potentially invasive aquatic species.
The concept to combat Aquatic Nuisance Species (ANS) invasion resulting from ballast water
discharge, described in this paper, is a technical extension of MH Systems' American Underpressure
System (AUPS). The AUPS utilises a slight negative pressure in the tank's ullage space, in an inert
environment, to prevent or minimize oil spillage from tankers (Husain et al. 2001).
The ballast water treatment method consists of bubbling the inert gas via a row of pipes (orifices at
the bottom of the pipes) located at the bottom of the tank, while maintaining a negative pressures of
­2 psi at the ullage space. The inert gas from a standard shipboard inert gas generator is composed of
84% Nitrogen, 12-14% CO2 and 2% Oxygen. The ballast water will be equilibrated with gas from an
inert gas generator. As a result, the water will become hypoxic, will contain CO2 levels much higher
than normal, and the pH will drop from the normal pH of seawater (pH 8) to approximately pH 6.
Ballast water treatment standards
Standards for treatment of ballast water are still in a state of flux. Efforts to define standards are
ongoing in the US Congress, International Maritime Organisation (IMO), and other individual
maritime nations. The US Congress (NAISA 2002) proposes an Act that will, among other
considerations, set the interim standards for ballast water treatment (BWT). It states, "The interim
standard for BWT shall be a biological effectiveness of 95% reduction in aquatic vertebrates,
invertebrates, phytoplankton and macroalgae." There are discussions about setting micron standards,
i.e., x microns cut-off for living organisms.
Currently, a fifty (50) micron standard is being discussed in various circles, including IMO and US
Coast Guard. The default standard appears to be the Ballast Water Exchange (BWE), or something
close to it. Cangelosi (2002) states "... the Coast Guard has set forth a "do-it-yourself" approach,
directing interested ship owners to conduct complex shipboard experiments (post-installation) to
undertake direct and real-time comparisons between BWE and treatment. If the comparison is
favourable and defensible, the Coast Guard will approve the treatment. ....."
Current investigative efforts of alternative technologies
Glosten (2002) provides a review of the numerous treatment systems options being investigated.
These include heat, cyclonic separation, filtration, chemical biocides, ultraviolet light radiation,
ultrasound, and magnetic/electric field. The methods not mentioned in this reference are hypoxia,
carbonation, and their combination. In studies of 18 months duration on a coal/ore vessel (Tamburri et

Husain: Ballast water treatment by de-oxygenation with elevated CO2 for a shipboard installation
al. 2002), the ballast water dissolved O2 level was reduced and held to concentrations at or below 0.8
mg/l by bubbling essentially pure nitrogen. The experiments resulted in a treatment "that can
dramatically reduce the survivorship of most organisms found in the ballast water..."
In extensive experiments with gas of varying percent CO2, N2 and O2 (McMahon 1995), the "...results
indicate that CO2 injection may be an easily applied, cost-effective, environmentally acceptable
molluscicide for mitigation and control a raw water system macrofouling by Asian clams and zebra
Corrosion considerations of various treatment systems
Shipboard corrosion mitigation is always a priority consideration. It requires the continual attention of
the crew and, if not carefully controlled, can actually compromise the strength of the ship. Any
installed ballast water treatment system must not under any circumstances increase the potential for
corrosion and, if possible, should decrease the potential. The system discussed in this proposal has
considered the corrosion issue. As reported in literature (Tamburri et al., 2002), corrosion might even
be mitigated by deoxygenation. Perry et al. (1984) states that unless pH level drops below 4, concerns
about corrosion are unfounded.
Timeframe of the project
We present initial proof of concept results, which have been conducted during 2002-2003.
Aims and objectives of the project
Except for ballast water exchange, essentially all treatment concepts involve the chemical change of
the water to cause an environment lethal for ANS. The chemical changes described by Tamburri et al.
(2002) and McMahon (1995) offer promising results, i.e., reduce the dissolved O2 in the one case, and
carbonate and reduce the pH in the other case. In both cases the process involves the exchange of
gases, the extraction of the dissolved O2 and the introduction of CO2. Surface contact area and partial
pressure differentials permit the gas exchanges to occur. The deoxygenation of the ballast water is
based on Henry's Law of gas solubility: The relative proportion of any dissolved gas including
oxygen in the ballast water is a function of the concentration, equivalent to partial pressure of the gas
(e.g. oxygen), within the mixed gases over the ballast water. The depletion of oxygen in the ballast
water is primarily a function of the shared surfaces and concentrations at the interfaces of the inert
gases and water.
The pH of the ballast water is lowered by the chemical reaction:
CO + H O H CO H + + HCO-
All systems described thus far in the literature, including ballast transfer, has left untreated the
sediment buildup in the bottom of the tanks. If the orifices in the lattice work of piping pointed down,
then the sediment could be stirred up facilitating the kill of the embedded ANS.
The purpose of the preliminary experiments described here was to obtain initial data on the effects of
"inert gas" on marine organisms. "Inert gas", hereinafter called trimix, a commercially available gas
mixture of 2% oxygen, 12% CO2 and 84% nitrogen resembles the gas generated by commercially
used marine "inert gas generators". Adult or young adult animals were chosen for two reasons a) to
make the size of specimens amenable for the experimental setup and b) to raise the significance of
possible effects since adults of a species are typically more tolerant of environmental changes than
juveniles or larvae. All animals were collected fresh from the coastal waters off La Jolla, CA and used
immediately. The plankton sample was collected with a plankton net from a small boat.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Research methods, test protocols and experimental design
The schematic of the experimental setup is shown in Figure 1. Three parallel incubations were done
for each experiment. Several organisms were incubated in 1.5L of seawater at 22°C in large
Erlenmeyer flasks. Each incubation was equilibrated with the respective gas using aquarium stones
before any organisms were introduced. The aerobic control was bubbled from an aquarium pump for
approximately 15 min and left open to the atmosphere after addition of specimens. An anaerobic
incubation was bubbled with 99.998% nitrogen for 15 min. After introduction of the organisms, the
bubbling was continued for another 10 min and then the container was closed with a rubber stopper or
the bubbling was continued. The incubation in trimix was treated similarly except that the gas mix
was used instead of nitrogen. The oxygen concentrations were measured after the initial bubbling
period using a Strathkelvin oxygen electrode with a Cameron instruments OM-200 oxygen analyser.
pH values were determined using a combination electrode and a Radiometer pH meter.
Survival of the specimens was determined visually by checking for motile responses to tactile
stimulus (e.g. mussels do not close their shells, barnacles to not withdraw their feet, shrimp do not
move their mouthparts, worms appear limp and motionless). After each testing of the animals, the
incubation flasks were bubbled for 10 min to reestablish original conditions. To verify survival of the
specimens, they were relocated to aerobic conditions and checked again after 30 min. If they still did
not respond, they were considered dead. The survival of the bacterium Vibrio cholerae strain N16961
was monitored by repeated plating on Luria-Bertani Agar with Rifampicin (100 µg/mL).
This setup allowed us to compare responses to nitrogen and "trimix" while making sure that test
specimens were not gravely affected by other experimental parameters. Incubation in pure nitrogen
allow for a comparison with published results by others.
Experimental results and discussion.
The oxygen concentrations were measured at "non-detectable" for the nitrogen incubations and 10%
air saturation (=16 Torr partial pressure) for the "trimix". The pH value of the water bubbled with
trimix reached 5.5 after the initial 10 min of vigorous bubbling. The aerobic and nitrogen bubbled
seawater maintained their pH at 8. The incubations showed clearly that "trimix" kills organisms
considerably faster than incubations in pure nitrogen Table 1. All organisms except of Vibrio cholerae
showed no mortality in aerobic conditions. The shrimp and crabs incubated in "trimix" were dead
after 15 min and 75 min, respectively. Even a transfer into aerated water did not result in any
movement. The brittle stars incubated under nitrogen started to move again after transferred into
aerated water. All the mussels incubated in nitrogen and "trimix" were open after 95 min but only the
ones in nitrogen still responded to tactile stimuli by closing their shells. The barnacles were judged
dead after incubation in "trimix" when they did not withdraw their feet when disturbed, the ones
incubated in nitrogen still behaved normally. The plankton sample mainly contained copepods. They
stopped moving after 15 min and could not be revived in nitrogen and "trimix" incubations. The
results are summarised in Table 1.
Low oxygen concentrations in water are a common natural phenomenon and their effects on live
organisms have been widely discussed in the past. Oxygen may not be available to an organism
because no water for respiratory purposes is present, e.g., during low tide in the intertidal zone.
Oxygen may also be removed in stagnant waters due to bacterial or other "life based" actions, e.g., in
ocean basins, fjords, tide pools, or in waters with high organic content and consequently high bacterial
counts, e.g., in sewage, mangrove swamps, paper mill effluent. In addition, oxygen can also be
removed by chemical reactions, e.g., in hot springs, industrial effluents. The manuscript by Tamburri
et al. (2000) summarises survival of a variety of larvae and adults of organisms including some which
may be significant as "nuisance species" under hypoxic conditions. The publication supports
extensively that most organisms only survive strongly hypoxic conditions for a few hours and only a

Husain: Ballast water treatment by de-oxygenation with elevated CO2 for a shipboard installation
few adults for several days. The authors suggest that 72 h of hypoxia will be sufficient to kill most
eucaryotic organisms, adults or larvae in ballast water.
Table1. Effects of Trimix on Marine Species
Dead after
75 min
Open but
6 dead after
95 min
Dead after
60 min
Dead after
Dead after
84 h
48 h
Dead after
Dead after
19 min
7 min
Brittle star
Most survive up to 3 h,
Most survive up to
Mean of 4
most dead after 26 h
3 h, several dead
after 26 h
Brittle star
Not moving but
Dead after
revivable by air
50 min
Affected but alive after
Dead after
30 min
25 min
2 dead after
5 dead after
30 min
45 min
Dead after
Dead after
15 min
15 min
Dead after
20 min
Dead after
15 min
8 - 10
Dead after 2 h
Many dead after
Mean of 3
2 h
Vibrio cholerae
2.5 x106/ml
>>99% dead after 24 h
>>99% dead after
Aerobic: 30%
24 h
dead after 24 h
*Trimix (2% oxygen, 12% CO2 and 86% nitrogen)
The effects of high CO2 on organisms in natural waters have become a research focus because of
proposals to dispose atmospheric CO2 in the deep ocean (Haugan 1997, Omori et al. 1998, Seibel and
Walsh 2001). Two effects have to be distinguished when looking at "trimix" incubations in seawater:
a) the lowering of the pH from pH 8 to about 5.5 and b) the raised CO2 concentrations in the water.
While the pH change caused by the incubations in "trimix" are in the range of published experiments,
the CO2 concentration in "trimix" (about 14%) is much higher than those investigated in the published
literature (0.1% to 1%). Therefore, the effects of "trimix" incubations should be much stronger than
those published previously.
Several publications have shown the detrimental effect of lower pH values and high CO2 levels on
aquatic life. In a recent publication, Yamada and Ikeda (1999) tested ten oceanic zooplankton species
for their pH tolerance. They found that the LC50 (=pH causing 50% mortality) after incubations of 96
hours was between pH 5.8 and 6.6 and after 48h it was between pH 5.0 and 6.4. Therefore, the pH
value caused by incubations with "trimix" is well within the lethal range for this zooplankton.
Huesemann et al. (2002) demonstrate that marine nitrification is completely inhibited at a pH of 6.
Larger organisms were also investigated, a drop in seawater pH by only 0.5 diminishes the
effectiveness of oxygen uptake in the midwater shrimp Gnathophausia ingens (Mickel and Childress
1978) and Deep sea fish hemoglobin may even be more sensitive to pH changes (Noble et al. 1986). It
appears that a common metabolic response to raised CO2 levels and concomitant lowered pH is a
metabolic suppression (Barnhart and McMahon 1988, Rees and Hand 1990). Most recently, first
papers were published investigating the effects of environmental hypercapnia in detail (Poertner et al.
1998, Langenbuch and Poertner 2002). The effects of pH changes on phytoplankton growth has been
reviewed by Hinga (2002). The review summarises data from 22 studies. Many of the cited studies

2nd International Ballast Water Treatment R&D Symposium: Proceedings
use elevated levels of CO2 to adjust pH. In almost all cases, the growth of unicellular phytoplankton
and diatom species was severely affected by low pH below pH 6.5, only the species Nitzchia
showed significant growth at pH 5.5. Since all of the studies cited were done at high light
levels and in aerobic conditions, it can be safely assumed that the conditions in an hypoxic dark
environment as is found inside of an inert gas treated ballast tank is even more detrimental to
phytoplankton growth.
The trimix combines both of these effects on organisms - hypoxia and hypercapnia. Preliminary
results demonstrate the effectiveness of this combination in quickly killing a variety of sample
organisms. Contrary to methods using additions of biocides or any chemicals in general, nothing is
added to the ballast water and, therefore, nothing will be released into the environment when it is
released again. Methods using radiation, heating, or filtering ballast water before or during a ship's
trip, can be expensive. The equipment needed to establish a rapid gassing of ballast water is available
off the shelf and has been used in the marine environment. The plumbing and gas release equipment
has been optimised and has been used in application such as aquaculture, sewage treatment and
industrial uses. Extensive supporting literature and research about the design and optimisation of
equipment for the aeration of water is available from public resources. Inert gas generators are
available for fire prevention purposes on ships and other structures and are already installed on many
ships, mainly tankers. They can use a variety of fuels including marine diesel to generate the inert gas.
Several topics have to be further investigated before a conclusive recommendation about the treatment
of ballast water with "inert gas" can be made: a) how are larvae, eggs, and plankton effected and b)
what is the affect of trimix type inert gas in fresh water? If ballast water is taken up through a screen,
larger animals will not be included. The initial tests were made with adults because of easy access to
them. However, if adults of a species are effected by "inert gas" it is most likely that their larvae will
also be effected probably even more so.
Future tests will be conducted with specimens from plankton and larval cultures and with incubations
of mixed plankton collected from the ocean. Determinations of viability will be made by microscopic
observations (e.g. movement of mouthparts, swimming behaviour), ATP measurements (the ATP
levels rapidly decreases after death of an organism), and the ability to bioluminesce (many planktonic
organisms emit light, an ability which ceases after death). Fresh water organisms will be of interest
because the pH change is not as much as in seawater. Freshwater in its natural environment can have
pH values around 5.5. It has to be proven that raised CO2 concentrations in combination with hypoxia
will also affect these species. Only then can the method be used for both, fresh and salt water ballast.
Analysis and Design Equations
In this section, we present mathematical descriptions of the deoxygenation process and of the transfer
of carbon dioxide into the ballast water, which, in turn, leads to lowering of the pH to the levels lethal
to most ANS. We obtain closed-form mathematical models, usable in design of a shipboard system
from any set of given specifications. The list of symbols used in the equations is given at the end of
the paper.
The system being analysed places a mixture of nitrogen and carbon dioxide with a relatively small
fraction of oxygen in contact with ballast water. The oxygen level in the ballast water is assumed to
have reached equilibrium with air as a result of prolonged contact, and therefore would contain a
concentration of oxygen sufficient to support a wide spectrum of life forms. The objective is to reduce
the oxygen content to a low level by interchange with the gas mixture. The gas is bubbled through the
ballast water, which assures uniform distribution of dissolved gas throughout the ballast tank. Thus,
diffusion within the tank can be neglected. Bubbles are assumed to be small and variation of
hydrostatic pressure over the vertical dimension of a bubble is neglected.

Husain: Ballast water treatment by de-oxygenation with elevated CO2 for a shipboard installation
We do not discuss here the size of bubbles and the frequency of their generation. These two issues are
addressed in existing reference literature (see, for example, Perry et al. 1984).
We assume that deoxygenation process follows Henry's Law with equilibrium achieved within the
residence time of each bubble. The composition of the mixture in the bubble changes primarily due to
transfer of carbon dioxide, a dynamic chemical process assumed to obey the mass action kinetics.
Deoxygenation Process
As trimix gas is flushed through the system, the total weight of oxygen in the ballast water will be
reduced. For the purpose of analysing the deoxygenation process we neglect the presence of carbon
dioxide in the trimix.
When a small quantity of gas, dQ, is admitted, it contains an oxygen molar fraction y0. By the time
this quantity of gas leaves the system it contains, according to Henry's Law, the molar fraction
Y / k
Therefore, we obtain the following differential equation:
dY =
y0 -
y 0 -Y / k
Integration of this equation yields:
Q = k H ln y0 -Y / k
From this equation it follows that pumping 5,200 m3 of gas into a 32,200 m3 tank reduces oxygen
concentration to 0.83 ppm. This level of hypoxia is lethal to many ANS. With the flow rate of
38.2 m3/min this can be achieved in 135 min. The relationship between the size of the tank and the
time required to deoxygenate it is linear. Therefore, these results can be scaled to any tank size.
Underpressure in Ullage Space of Ballast Water Tank
Deoxygenation is enhanced by the under-pressure, as can be seen from the following simple
argument. Let p be pressure of water at a given depth in the absence of underpressure. Let pu be the
absolute value of the negative pressure at the top. Let Y be the weight fraction of oxygen in the water
without underpressure and Yu ­ the same weight fraction with underpressure. Then by Henry's Law:
Y - Y
yp - k y( p - p )
From this equation we conclude that solubility of oxygen is reduced by underpressure. This factor
becomes even more significant as a bubble rises to the surface, and the pressure inside decreases.
For example, if p=14.7 psi (the usual value at the surface of the tank) and the absolute value of the
underpressure is 2 psi, then the solubility of oxygen is reduced by approximately 14%.
The maintenance of underpressure is not mandatory. The underpressure helps accelerate the de-
oxygenation process because, by reducing the oxygen solubility, it also reduces the amount of inert
gas needed. For example, 2 psig underpressure will speed up the de-oxygenation by 14%; 0.5 psig
underpressure will speed it up by 3.5%. Slight underpressure is also helpful in eliminating the
contaminated gas from the ullage space.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Carbon Dioxide Transfer
Since we assumed that the pressure inside the bubble depends only on the pressure of the liquid
surrounding it, we can write:
dp = - gu
, p = p0 - gut

By definition we have n
. Differentiating this equation we obtain:
2 = xn
CO2 = x
+ n
However, since the reaction of carbon dioxide with water is the dominant cause of change in the
chemical composition, we can write:
Combining this with the Equation (2) yields the following equation:
= (
1 -
In addition, we can solve n = xn + n for n to obtain

1- x
From the Law of Mass Action kinetics we have:
dnCO2 = -kp

For the partial pressure of carbon dioxide we have, according to Dalton's Law pCO2=xp.
Combining the equations (1), (3), (4), and (5) yields:
= -
x 1
( - x)2 ( 0
p - gut
) .
n N
This equation can be integrated to obtain:
I (x) - I ( 0
x ) = -
(2 0
p - gut) ,
I (x) =
+ ln
1- x
1 - x
This equation can be used to calculate parameters of the systems, including the residence time of a
bubble, required to achieve the desired molar fraction of carbon dioxide in the bubble. The latter
quantity is related to the pH and the concentration of carbon dioxide in the water, as we shall see in
the next subsection.

Husain: Ballast water treatment by de-oxygenation with elevated CO2 for a shipboard installation
Concentration of Carbon Dioxide in Water and pH Calculation
Concentration of carbon dioxide in water can be determined as the ratio of the number of moles
transferred from the bubble to the volume of the tank. The number of moles transferred from each
bubble can be determined from the value of x as follows. By definition, we have:
x =
2 +
Solving for n
we find:
2 =
1- x
which gives the following answer for the concentration of carbon dioxide in water:
c =

CO2 -

- x
The concentration of the hydrogen ions in the water can be calculated from c by solving the following
equation for h:
h2 = K
c - h
The pH can be then found by taking the - log h .
We can also solve the Equation (8) for c and substitute the result into the Equation (7). This yields
after some tedious, but straightforward algebra the following relationship between the desired molar
fraction of carbon dioxide in the bubble and the desired concentration of hydrogen ions in the water:
x = 1
KN (n 0
+ n ) + (K - h)hV
The equations (6) and (9) constitute a closed-form mathematical model of carbon dioxide transfer
usable for design of the treatment system.
The MH Systems' Ballast Water Treatment System Description
(Note: The Authors are cognizant that a large tanker of the size as 300,000 DWT may not be an ideal
candidate for ballast water treatment features. However, this hypothetical design study can be easily
modified for smaller tankers.)

The MH Systems Ballast Water Treatment System is a combination of two other effective treatment
systems, i.e. deoxygenation and carbonation. It also is an extension of the MH Systems American
Underpressure System ­ AUPS (Husain et al. 2001). The inert gas, supplied by the standard marine
gas generator, is 84% nitrogen, 12-14% carbon dioxide and about 2% oxygen. This inert gas has all
the ingredients necessary to combine the two very effective treatments of hypoxia and carbonation at
a very reasonable cost. The laboratory tests at Scripps, described previously, show that this gas needs
very little contact time to be effective. The analyses described earlier established the flow rates and
control time for hypoxia carbonated conditions.
Each ballast tank has rows of pipe at the tank floor with downward pointing nozzles. The pressurized
inert gas is jetted downward out of the piping. The jets stir up the sediment for contact with the inert
gas bubbles. The bubbles then rise through the ballast water to the space above the water surface,
which has previously been underpressurized to ­2 psi. For the purposes of this paper, a 300,000 DWT

2nd International Ballast Water Treatment R&D Symposium: Proceedings
single hull tanker was used for design studies of this system to test practicality and affordability.
Applicability to a 300,000 DWT double hull tanker was also examined. Figure 2 shows inboard
profile, deck plan view, piping layout, nozzle detail and section through ballast tank. Figure 3 shows
schematic of the system and Figure 4 shows isometric of one tank. A 300,000 DWT double hull
tanker has somewhat less installation costs since the tank bottom is smooth as shown in Figure 4.
For the 300,000 DWT tanker, there are 8 ballast tanks as follows in Table 2:
Table 2. Ballast Water Tank Capacity
Size M3
Fore Peak
B Engine Room S
B Engine Room P
Aft Peak
From analyses and experience (Tamburri et al. 2002), it is estimated the hypoxia and pH conditions
can be set in at least 8 hours, even in the largest tanks, B3 Port and Starboard. The flow rate is 1350
cfm for each of these tanks. With one 1500 cfm marine gas generator, and treating each tank
sequentially, it is estimated that all 8 tanks can be in a hypoxia, low-pH (5.5 - 6) condition in less than
48 hours. Contact time for essentially total lethality may not require more than another 24 hours
although the remainder of the 2 to 3 week voyage is available.
The space above the liquid in each tank is underpressurized to about ­2 psi and maintained throughout
the voyage. As the gas bubbles rise up to the surface, they are evacuated by a blower to maintain the
underpressure of the inert gas blanket at the surface. The underpressure further facilitates the
solubility of the oxygen (see analysis) and tends to compensate for the oxygen captured in the bubbles
as they rise.
Since the ballast tanks are treated sequentially, only two 700 cfm compressors are required to
compress the gas. The gas is compressed enough to offset the hydrostatic head plus an additional 25%
psi to provide a jet force for stirring the sediment. Two compressors are provided for redundancy. If
there are some concerns with the dumping of hypoxia and carbonated treated water, it is easily
countered with the system discussed in this paper. The compressors will shift over from the gas
generator to atmospheric and the ballast water will be oxygenated within just a few hours. In this
same period of time the CO2 is readily washed out since the air contains no CO2 component.
Sensors are needed to monitor the pH to ensure that it never goes below about 5.5. Sensors will
measure dissolved oxygen content to ensure that adequate deoxygenation is established. Sensors will
also monitor the underpressure. The control system will remotely start and stop the gas generator, the
compressor and the blower. The control system also remotely controls the valves off of the inert gas
manifold to each ballast tank and the valving for the underpressure manifold.
It is expected that system will be controlled by a suitably designed arrangement of programmable
logic controllers (PLCs). These devices are commercially available. They are also easy to program
and maintain.
A control console with displays will integrate the functions of the inert gas generator and the AUPS
ballast water treatment system as well as provide for monitoring, status displays and manual override,
if required.

Husain: Ballast water treatment by de-oxygenation with elevated CO2 for a shipboard installation
Tests were conducted with the AUPS System installed on a naval reserve fleet tanker. They verified
the structural capability of tanks to withstand the pressure of -3 psi and the controls needed to
maintain the required underpressure. These findings are applicable to the equipment and controls that
will be used for the ballast water treatment system.
The following are the design features of the shipboard system:
· Dry docking is not required for the installation of the system. The system can be retrofitted at
pier side.
· The system includes mainly off-the-shelf components.
· The system is fully automated. Data can be transmitted in real time to a shore-side facility, if
· Sensors are installed at different locations inside the tank to determine pH and oxygen levels.
· The system requires low maintenance.
Economic Evaluation of MH Systems' Ballast Water Treatment System for a 300,000 DWT

In making an economic evaluation, the analysis methodology described in Mackey et al. (2000) was
used. This method states, "a logical basis for economic comparisons would be a change in Required
Freight Rate (RFR)." Since there would be no change in cargo capacity, then:
[CRF(i,n) P
+ Y

CRF (i, n) is Capital Recovery Factor for an interest rate i and n for economic payback years,

is change in Capital Cost, and
is net change in annual operating cost and revenue.
Mackey et al. (2000) stated that the economic payback period for conversions is typically 5 years.
The Authors selected a 300,000 DWT tanker for analysis. As stated earlier, a ballast water treatment
system applicable for ships must have the capacity for treating huge quantities of ballast water. If a
system is practical and economical for treating a ship with 8 ballast tanks of 110,823 cubic meters,
then it is practical for all ship types. The economics would have to be assessed for ships of other,
smaller ballast capacity, as the economics might not scale. But obviously, the effectiveness as well as
the practicality of the system would be established.
Table 3 (over) lists the principal parts and materials in the ballast water treatment system together
with estimated prices and labour costs.
The total cost is approximately $3,057,100. All tankers already have some type of inert gas generating
capability. The newer tankers have generators with a gas mixture discharge similar to the mix used in
the experiments at Scripps. Nevertheless, for conservatism, the generator has been included in the
cost. Similarly tankers probably have sufficient excess electrical capacity to supply the load of this
equipment ­ the compressors and blower. This is especially true since this is on the return trip in
ballast and the machinery will only run about 48 hours each trip. Nevertheless, again for extreme
conservation, a 300 KW generator has been included.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Table 3. Preliminary cost estimate for ballast water treatment system.
Note: labor cost is based on US repair shipyard estimates.
Parts and Materials
Capacities or Type
Material Cost
Labor Cost
Material & Labor
Blower (Exhaust)
2000 CFM-100HP
$ 10,000
Reciprocating Compressor
Electric-700 CFM-100HP
$ 40,000
Inert Gas Generator
1500 CFM - 50 HP
$ 175,000
$ 175,000
$ 150,000
Row of pipes at tank bottom PVC
3" SCH 80; Length in Ft.
Header Branch Piping - PVC
8" SCH 80 "
$ 105,000
Header Piping - PVC
10"SCH 80 "
$ 252,000
Header Piping - Steel
10" SCH 40 (Steel)
$ 650,000
Valves-Electric (Ballast)
10" Butterfly
Valves-Electric (Inert Gas)
10" Butterfly
Coarse Bubbles
$ 130,000
Fittiings (Elbows, Tees, Couplings) PVC
$ 140,000
300 KW
$ 60,000
Sub-Total Materials
$ 767,600
Sub-Total Labor
Sub-Total Materials & Labor
Sensors, Controllers & Computer
pH Gauges
Pressure Gauges for Ullage space
Pressure Controllers for Ullage
Controller for Compressor
Oxygen Sensor
Controller for Valves
$ 40,000
Computer Software & Hardware
$ 50,000
Sub-Total - Material
$ 147,000
Labor for Installation
$ 250,000
Sub-Total - Material & Labor
Other costs
Engineering & Maintenance
To make a usefully indicative estimate of operating costs, the following assumptions were made:
· The tanker will operate to 360 days per year.
· Six (6) voyages per year between Persian Gulf and USA.
· Half of the voyages are return trips in ballast, or 6 trips a year.
· Assume the 2 compressors and blower must operate 48 hours to obtain hypoxia and
carbonation in all 8 tanks (note that actually the cfm of both compressors is only required for
tanks B3 port and starboard and B6 port and starboard.
· Operating costs are primarily the fuel costs for the inert gas generator and the 300 KW
· n is 5 years (economic payback period) and i (interest rate) is 8%.
If the gas and electric generators operate 48 hours for each of 6 voyages, then the total operating time
is 288 hours per year for each generator. About 6,000 gallons of diesel fuel would be consumed by the
electric generator and for the gas generator about 16,500 gallons. This is a total of 22,500 gallons. At
a cost $1.25 per gallon, the yearly operating cost will be about $28,125. Considering the few hours

Husain: Ballast water treatment by de-oxygenation with elevated CO2 for a shipboard installation
per year that the machinery operates and the fact that the ship has no cargo and therefore less
requirements of the crew, minimal cost has been allocated for maintenance.
CRF (i,n) = 0 25


0 25
× 057
300 000 × 6
/ ton
In estimating the cost of treatment per ton of ballast water, the estimated annual operating costs of
$28,125 is used. The approximate 4 million cubic feet of ballast is 128,000 tons. Six trips are made in
ballast, which is a total of 768,000 tons treated. Therefore, cost of ballast water treatment is 3.7 cents
per ton.
This ballast water treatment system is focused on treating the huge amounts of ballast water
discharged into US harbours. It has the capacity to readily treat these huge quantities using standard
marine components. For tankers that already have the major components on board, it would be very
affordable. And for tankers with the AUPS spill containment, the added cost would be even less
Also, it appears (although not tested) that this system may be adequately effective in treating
sediments. Ballast Water Exchange leaves sediment and other residue untreated. In fact, only the
filtration concept treats sediment, by eliminating it.
Conclusions and recommendations
Based on the preliminary study, we conclude that a combination of hypoxia and elevated CO2 levels
are expected to kill in excess of 95% of marine phytoplankton, zooplankton, macroalgae, and
invertebrates as required by the interim standard proposed by the US Congress. The treatment system
proposed requires only off-the-shelf components which can be installed at pier side, without dry-
docking. The system can be fully automated. Installing pH and oxygen sensors at multiple locations
inside the tank can assure continuous remote monitoring of the ballast water.
It will be necessary to continue the laboratory tests, especially to include experiments on the effects of
the system on phytoplankton, cysts and spores. In addition, the practical application of the system
should be verified in a large scale effort using land based tanks or ballast water tanks in ships.
Barnhart, M.C. & McMahon, B.R. 1988. Depression of aerobic metabolism and intracellular pH by
hypercapnia in land snails, Otala lactea. J. exp. Biol. 138, pp. 289-299.
Boyd, C.E. 1998. Pond water aeration systems. Aquacultural Engineering 18, pp. 9-40.
Bryant, C. 1991. Metazoan life without oxygen. London: Chapman and Hall.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Cangelosi, Allegra Nov. 14, 2002. Testimony Before the Joint Committee on Resources and Science
of the U.S. House of Representatives.
Fenchel, T. & Finlay, B.J. 1995. Ecology and evolution in anoxic worlds. New York: Oxford
University Press.
Glosten­Herbert-Hyde Marine April, 2002. Full-Scale Design Studies of Ballast Water Treatment
, Prepared for Great Lakes Ballast Technology Demonstration Project.
Grieshaber, M.K. 1994. Physiological and metabolic responses to hypoxia in invertebrates. In Review
of Physiology, Biochemistry, and Pharmacology
, vol. 125 (ed. M. P. e. a. Blaustein), pp. 43-147.
Berlin, New York: Springer verlag.
Haugan, P.M. 1997. Impacts on the marine environment from direct and indirect ocean storage of
CO2. Waste Management 17, pp. 323-327.
Hinga, K. 2002. Effects of pH on coastal marine phytoplankton. Mar. Ecol. Prog. Ser. 238: pp. 281-
Huesemann, M.H., Skilmann, A.D. & Crecelius, E. A. 2002. The inhibition of marine nitrification by
ocean disposal of carbon dioxide. Mar. Poll. Bull. 44, pp. 142-148.
Husain, M., Apple, R., Thompson, G. & Sharpe, R. 2001. Full Scale Test, American Underpressure
System (AUPS) on USNS Shoshone, Presented to Northern California Section, SNAME, September
Johnson, P.D. & McMahon, R.F. 1998. Effects of temperature and chronic hypoxia on survivorship of
the zebra mussel (Dreissena polymorpha) and Asian clam (Corbicula fluminea). Can. J. Aquat. Sci.
55, pp. 1564-1572.
Langenbuch, M. & Poertner, H.O. 2002. Changes in metabolic rate and N excretion in the marine
invertebrate Sipunculus nudus under conditions of environmental hypercapnia: identifying effective
acid-base variables. J. exp. Biol. 205, 1153-1160.
Laudien, J., Schiedek, D., Brey, T., Poertner, H.-O. & Arntz, W.E. 2002. Survivorship of juvenile surf
clams Donax serra (Bivalvia, Donacidae) exposed to severe hypoxia and hydrogen sulphide. J. exp.
mar. biol. ecol.
271, pp. 9-23.
Mackey, T.P., Tagg, R.D., Parsons, M.G., May, 2000. Technologies for Ballast Water Management,
Proc. 8th ICMES/SNAME New York Metropolitan Section Symp.
Matthews, M.A. & McMahon, R.F. 1999. Effects of temperature and temperature acclimation on
survival of Zebra mussels (Dreissena polymorpha) and asian clams (Corbicula fluminea) under
extreme hypoxia. J. Moll. Stud. 65, pp. 317-325.
McMahon, R.F. 2002. Evolutionary and physiological adaptations of aquatic invasive animals:
selection versus resistance. Can. J. Fish. Aquat. Sci. 59.
McMahon, R.F., Matthews, M.A., Shaffer, L.R. & Johnson, P.D. 1995. Effects of elevated carbon
dioxide concentrations on survivorship in zebra mussels (Dreissena polymorpha) and Asian clams
(Corbicula fluminea). In The fifth international zebra mussel and other aquatic nuisance organisms
, pp. 319-336. Toronto, Canada.
Mickel, T.J. & Childress, J.J. 1978. The effect of pH on oxygen consumption and activity in the
bathypelagic mysid Gnathophausia ingrens. Bio. Bull. 154, pp. 138-147.

Husain: Ballast water treatment by de-oxygenation with elevated CO2 for a shipboard installation
NAISA 2002. National Aquatic Invasive Species Act of 2002 ­ Summary, Current as of September 24,
Noble, R.W., Kwiatkowski, L.D., De Young, A., Davis, B.J., Haedrich, R.L., Tam, L.T. & Riggs,
A.F. 1986. Functional properties of hemoglobins from deep-sea fish correlations with depth
distribution and presence of a swim bladder. Biochem. Biophys. Acta 870, pp. 552-563.
Omori, M., Norman, C.P. and Ikeda, T. 1998. Oceanic disposal of CO2: Potential effects on deep-sea
plankton and micronekton- a review. Plankton Biol. Ecol. 45, 87-99.
Perry, R.H., Green, D.W., Maloney, G.O. Perry's Chemical Engineer's Handbook, 5th Ed., McGraw
Hill, 1984.
Poertner, H.-O., Bock, C. & Reipschlaeger, A. 2000. Modulation of the cost of pHi regulation during
metabolic depression: A 31P-NMR study in invertebrate (Sipunculus nudus) isolated muscle. J. exp.
203, pp. 2417-2428.
Poertner, H.-O., Reipschlaeger, A. & Heisler, N. 1998. Acid-base regulation, metabolism and
energetics in Sipunculus nudus as a function of ambient carbon dioxide level. J. exp. Biol. 201, 43-55.
Rees, B.B. & Hand, S.C. 1990. Heat dissipation, gas exchange and acid-base status in the land snail
Oreohelix during short-term estivation. J. exp. Biol. 152, pp. 77-92.
Reipschlaeger, A. & Poertner, H.-O. 1996. Metabolic depression during environmental stress: The
role of extracellular versus intracellular pH in Sipunculus nudus. J. exp. Biol. 199, pp. 1801-1807.
Seibel, B.A. & Walsh, P.J. 2001. Potential impacts of CO2 injection on deep-sea biota. Science 294,
pp. 319-320.
Tamburri, M.N., Peltzer, E.T., Friederich, G.E., Aya, I., Yamane, K. & Brewer, P.G. 2000. A field
study of the effects of CO2 ocean disposal on mobile deep-sea animals. Mar. Chem. 72, pp. 95-101.
Tamburri, M.N., Wasson K., & Matsuda, M. 2002. Ballast water deoxygenation can prevent aquatic
introductions while reducing ship corrosion. Biological Conservation. 103, pp. 331-341.
Watanabe, Y., Ishida, H., Yamaguchi, A. & Ishizaka, J. 2001. III-5. Effects of high concentration of
CO2 on deep-sea plankton. Nippon Suisan Gakkaishi 67, pp. 764-765.
Yamada, Y. & Ikeda, T. 1999. Acute toxicity of lowered pH to some oceanic zooplankton. Plankton
Biol. Ecol.
46, pp. 62-67.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
concentration of carbon dioxide in the water, including ions produced by electrolytic dissociation.
acceleration due to gravity.
concentration of hydrogen ions in the water.
dissociation constant of carbonic acid
( =
4.3 10-
reaction rate constant.
Henry's Law constant for oxygen
= 39.79×10 ).
total number of bubbles generated.
total number of gas moles in the bubble.
number of moles of carbon dioxide in the bubble.
number of moles of nitrogen in the bubble.
total pressure inside the bubble.
partial pressure of carbon dioxide in the bubble.
gas weight flow rate.
bubble speed.
volume of the tank.
molar fraction of carbon dioxide in the bubble.
weight fraction of oxygen in the water.
molar fraction of oxygen in the bubble.

density of the ballast water.
Superscript 0 refers to quantities in the gas bubble when it is first introduced into the tank.
Subscript 0 refers to quantities in the water at the time t=0.
Figure 1. Schematic of the experimental setup.

Husain: Ballast water treatment by de-oxygenation with elevated CO2 for a shipboard installation
Figure 2. Inboard profile, deck plan view, piping layout, nozzle details and ballast tank section view.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 3. Typical ballast water treatment schematic.
Figure 4. Typical ballast treatment piping in a single hull tanker segregated ballast tank. In a double hull vessel,
the piping system is simplified by installing the nozzle grid on the tank bottom without any structural interference.

Session 2:
Heat and Electro-based
Treatment Systems

Does heat offer a superior ballast water treatment
G. Rigby1, G. Hallegraeff2 & A. Taylor3
1Reninna Pty Limited, Australia.
2University of Tasmania
3Alan H Taylor & Associates, Australia
Treatment options being researched
This work involves the use of heat treatment using various engineering designs to kill or inactivate
harmful organisms present in ballast water
Timeframe of the project
The authors' interest in the effect of heat on ballast water organisms and the translation of laboratory
results to practical designs for onboard implementation commenced in 1993. The first full scale
shipboard trials were undertaken on the bulk carrier, Iron Whyalla, in 1998. Further ongoing work has
continued to refine biological temperature thresholds and alternative designs as a means of extending
this technique to a wider range of vessels and voyages.
Aims and objectives of the project
The primary objective of the recent work has been to gain a better understanding of the biological
effects of heat for the range of organisms and conditions likely to be encountered in ballast water and
to extend the initial range of options and designs for future extension and implementation of this
Background and introduction
Mandatory reporting and regulations now exist in many parts of the world for the management and
control of ballast water to minimize the risks of translocating harmful organisms around the world
(Rigby and Taylor 1993). The International Maritime Organisation's (IMO) Maritime Environmental
Protection Committee (MEPC) Ballast Water Management Convention is in its final stagers of being
drafted and is scheduled for submission to a Diplomatic Conference in February or March 2004 for
the signing of the `Final Act' of this Convention.
These regulations require each ship to have on board and implement a Ballast Water Management
Plan (BWMP) that uses an approved management procedure. At the present time this generally
involves the use of an accepted form of Ballast Water Exchange (BWE). In addition to BWE most
Guidelines/Regulations (including the new IMO Draft Convention) have provision for the use of an
alternative treatment option that complies with the approved standard for efficacy (the latter are yet to
be defined and agreed in detail at the Diplomatic Conference).
BWE significantly reduces the number of organisms from the ballasting port being discharged into the
receiving environment and hence is a step in the right direction in reducing the risk of the
establishment of new inoculations establishing. In general, the BWE regulations stipulate that a water

2nd International Ballast Water Treatment R&D Symposium: Proceedings
exchange replacement efficiency of at least 95% be achieved. However for many ships and/or
voyages, although this level of water exchange is achieved (or exceeded) the biological replacement
efficiency for e.g. zooplankton may be considerably less than 95%. Furthermore, for some voyages,
BWE can significantly increase the risk of possible establishments of harmful aquatic organisms as a
result of taking on new organisms during the exchange process that may be more detrimental than
those in the originally ballasted water (Rigby, 2001).
Even though insufficient information is currently available to estimate with certainty what constitutes
a minimal viable inoculum for a biological establishment, it is widely recognized that the ultimate
long term goal for ballast water treatment should be a 100% removal or inactivation of harmful
A variety of alternative technologies have been tested (Rigby & Taylor, 2001) and new options are
continually being proposed as possible candidates. However at the present time, only limited success
has been achieved in achieving superior performance to that available from BWE.
One of the difficulties in comparing the performance of alternative technologies arises from the fact
that no standard for biological efficiency currently exists. A Standards Workshop organized by the
Global Ballast Water Management Programme developed some suggestions in 2001 that were
reviewed and modified at the subsequent meetings. However efforts to refine the standard have been
hampered by the complexities involved in combining a practically achievable as well as a
scientifically acceptable outcome both for the short term and long term and have only been partly
successful to date using currently available best technology. To obtain an achievable starting point an
initial standard might be based on an "equivalent" biological standard to that currently prescribed for
BWE, i.e. 95% removal, kill or inactivation.
The issue of definition of "organisms" is yet another difficulty and the discussion in this paper will be
restricted to more general comments based on currently available data, but will be primarily focused
on zooplankton and phytoplankton observations. Although some concerns about possible risks from
viruses and bacteria in ballast water have been expressed, it has generally been considered that these
are of secondary concern in overall international operations and consequently research on treatment
technologies has not included these organisms.
The most promising treatment option identified by the US National Research Council review for
successful shipboard treatment was constant backwash filtration (NRC, 1996). Extensive research and
demonstration studies have been undertaken internationally using this and other filtration systems to
assess the effectiveness of this option. From work carried out so far, mean particle size count
efficiencies of 91% have been achieved for particles above 50 µm (using a screen filter) and 91.6%
for particles above 100 µm (disk filter) with wide variations in removal efficiencies for organisms
with a mean of 90% for zooplankton (50 µm filter) and 50%-around 95% for phytoplankton (Parsons
& Harkins, 2002; Cangelosi, 2002).
Like BWE, filtration, which is based on a physical separation process, is not directly linked to
biological destruction but rather relies on the efficiency of size separation and the relationship
between size and organism species for removal. Clearly this option has limitations in achieving what
may be regarded as an acceptable level of biological efficiency. Likewise very few other treatment
options have demonstrated an ability to achieve desirable results, especially at the scale of operations
that will be required for many vessels (2000 to 20,000 m3/h ballast water-or an equivalent total
quantity of 25,000 to 200,000 m3).
Heating ballast water to kill or inactivate ballast water organisms, although not yet formally accepted
by IMO or any National Authority as an approved treatment option has been demonstrated in some
full on-board at sea trials to be capable of destroying virtually all of the phytoplankton and
zooplankton present in the ballast water, and as such offers a superior treatment option in cases where
it can be used. This presentation reviews the current status of heat treatment research and
development and recommends its acceptance as one of the superior options for future implementation.

Rigby: Does heat offer a superior ballast water treatment option?
Research methods and review of studies to date
The biological basis of heat to kill or inactivate marine organisms.
High temperatures induce denaturation of key proteins and compromise cell membrane structures
through increased mobility of molecules, thereby inactivating metabolic processes vital to all known
living organisms. As a general rule, the smallest organisms such as bacteria tend to be most heat
resistant, because their minute protoplasm volume allows for less damage from heat-induced mobility
of molecules.
Enterobacteria such as Salmonella, Campylobacter and Escherichia, which are adapted to living
within warm blooded animals, require heat treatments of 60-70°C for complete inactivation. It has
been well established that effective heat treatment is a probability function of both temperature and
treatment time, e.g. milk pasteurisation can equally be achieved by 15 seconds at 72°C ("flash"
pasteurisation) or 30 min at 63-66°C ("holding method"). There is no evidence that heat treatment has
any cumulative effect on cells (Brock & Madigan, 1994). Among the enterobacteria, species that
produce highly resistant endospores (e.g. Clostridium botulinum) are the most heat resistant.
Autoclaving procedures widely used to sterilise laboratory and hospital equipment utilize heat
treatment of 10-15 min at 121°C.
Table 1 lists lethal temperatures for a wide range of marine organisms, from bacteria, microalgae,
seaweed spores, molluscs, starfish, brineshrimp to rotifers. A striking conclusion (Figure 1) is that
most marine organisms, at least in a hydrated stage, can be killed at temperatures of 40-45°C, that is
well below temperatures used in food treatment technology. The only exceptions are marine bacteria
(commonly requiring 45-55°C), the smallest (<5 micron) diatoms and dehydrated brineshrimp cysts
or rotifer eggs. Longer treatment times (hours to days) are generally more effective in achieving heat
transfer into the interior of organisms than using short treatments at higher temperatures. An example
of this is spraying of thick-walled oysters for 40 sec with 70°C water killed associated boring
polychaetes, but did not sufficiently raise the core temperature of the oysters to kill them (Nel et al,
Concerns that heating ballast water to temperatures of 40-45°C would stimulate the growth of harmful
bacteria have not been substantiated by simulated laboratory experiments (Desmarchelier & Wong,
1996). Bacterial growth at those temperatures would only be stimulated when contained in food
products or nutrient broth, but not in nutrient starved seawater.
Table 2 summarises the studies and nature and observations from experimental studies that have been
undertaken to date.
The fact that lower temperatures are generally required for longer treatment times means that
appropriate temperatures can be selected for specific shipboard designs based on the nature and
availability of heat from the ship's main engine or auxiliary sources together with the ballast water
temperatures, pump and tank designs. These facets are explored in more detail in the design case
studies that are included in this paper.
The original heat treatment proposals for use on ships (Rigby, 1994) recommended the use of waste
heat from the ship's main engine cooling water system. The quantity of heat required to heat the total
quantity of water on a large ship (50,000 to 100,000 tonnes) is large and to provide this from a stand
alone independent heat source would be impractical and expensive. As an example, heating the
50,000 tonnes of ballast water on the Iron Whyalla on a once through basis (from 30°C to 45°C)
during ballasting or deballasting) without any heat recovery would require approximately 70 MW
power, which well exceeds the main engine power of 13.7 MW.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Table 1. Summary of Lethal Temperatures for Marine Organisms.
Vibrio cholerae
45°C, 2-3 hrs
McCarthy 1996
in seawater but
Desmarchelier & Wong 1998
survived in
nutrient broth
Diatoms Skeletonema costatum ,
35°C, 30-60 min
Marshall & Hallegraeff (original
Detonula pumila,Pseudo-nitzschia
data); Forbes & Hallegraeff
cuspidata, Thalassiosira rotula
Small diatoms Amphora, Navicula
35°C, 5 hr
Forbes & Hallegraeff 2002
Raphidophyte Heterosigma
35°C, 5 hr
Marshall & Hallegraeff
(original data)
Picoplankton Nannochloropsis
42.5°C, 3 hr
Marshall & Hallegraeff (original
Chlorophyte Dunaliella tertiolecta
42.5°C, 24 hr
Marshall & Hallegraeff (original
Dinoflagellate Amphidinium
35°C, 30 min
Marshall & Hallegraeff (original
Dinoflagellate Alexandrium
45°C, 3 min
Montani et al. 1995
Alexandrium catenella dinocysts
42°C, 30 min
38°C, 4.5hr
Hallegraeff et al.1997
Gymnodinium catenatum
Hallegraeff et al.1997
30-60 sec
1-2 hr
Undaria pinnatifida spores
35-40°C, 0.9-42
Mountfort et al.1999
Dreissena polymorpha
36°C, 10 min
32°C , 3hr
Jenner & Janssen-Mommen
Crassostrea gigas (larvae)
40-48°C, 6-97 min
Mountfort et al.1999
Crassostrea virginica
Sellers & Stanley 1989
Mytilus edulis
40°C, 0,33 hr
Johnson et al. 1983
Corbicula fluminea
Graney et al. 1983
Perna viridis
43°C, 30 min
Coscinasterias calamaria (larvae)
Mountfort et al.1999
1-35 min
Artemia salina
42.5°C, 48 h
Marshall & Hallegraeff (original
(hydrated eggs)
(dry eggs survive)
42.5°C, 1h
Marshall & Hallegraeff (original
(eggs survive)

Rigby: Does heat offer a superior ballast water treatment option?
Table 2. Summary of heat treatment studies undertaken to date.
Research Group
Nature of studies
Summary of results and observations
Bolch and Hallegraeff
Laboratory studies to evaluate the effects
Gymnodinium catenatum cysts killed at 40-45°C for
of heat on dinoflagellate cysts
90-30 s, lower temperatures less effective
Rigby and
Biological observations of organisms in
No survival of phytoplankton and zooplankton
Hallegraeff 1994
heated ocean engine cooling water
Identified sufficient waste heat from main engine to
Rigby 1994,
Design evaluation for heating ballast tank
heat all ballast water to 38°C - most phytoplankton
Hallegraeff et al. 1997
water on Iron Whyalla ­further laboratory
algae tested in vegetative stage killed at 35°C for
tests for effect of extended times on
30m-several h; total mortality of G. catenatum and A.
Rigby et al.1998,
temperature thresholds
catanella cysts at 38°C after 4.5 h.
Full scale trials on Iron Whyalla
All zooplankton and almost all phytoplankton
destroyed-original organisms reduced to amorphous
flocculent detritus. Combined actions of flushing
and heating give dual treatment in single operation
Sobol et al., 1995
Suggested shipboard design using engine
Not tested on board ship, but identified feasibility
hot water and steam to heat ballast water
using design details provided
to 70°C with three additional heat
Thornton, 2000
Shipboard design using additional heat
Small scale system (20m3/h) tested on MV Sandra
exchanger and holding tank to heat ballast
Marie. Plankton mortality of 80-90% achieved-rough
water to 65°C
seas caused problems and tank mixing not
Mountfort et al., 1999,
Laboratory and shipboard studies with
Temperature/time regimes identified for mortality;
2000, 2001
model organisms including larval
long ( 16h at 36°C), medium (10 min to 16h at 36-
mollusks, starfish and seaweed spores
45°C, short ( 10min at 46°C). Trials on Union
at 38°C resulted in all organisms being
killed. Trials on the M/T Iver Stream identified need
for effective mixing in tanks
Zhou, 2002 and
Laboratory and crude oil tanker shipboard
Ship design using waste heat from steam
MARTOB, 2003
design for rapid heating ballast water to
discharged from cargo pumps identified effective
heating can be achieved during ballast water
discharge-design not tested yet ­laboratory studies
using model organisms suggest a treatment
temperature of 50-55°C should be suitable
However 21% of the main engine power (5.71 MW) is discharged in the form of waste heat imparted
to ocean water used to cool the main engine. It was on the basis that this waste energy is available
without the use of additional fuel, that the original concepts of water heating were developed. Energy
balances and engine thermal efficiencies vary widely for different ships based on heat recovery and
utilization, sea water temperatures, ballast water pump capacities and drive arrangements as well as
operational requirements. Consequently the nature and feasibility of using this mode of ballast water
treatment requires a detailed analysis of the specific requirements and ship design features. A number
of these aspects are explored in the case studies below.
Case Study 1
Heating/flushing on the Iron Whyalla
In this study, in addition to achieving a practical design suited to the normal operation of this ship,
two shipboard trials were undertaken in one of the sets of ballast tanks (topside, trunk and double
bottom, containing 6350 tonnes water) on the BHP owned bulk carrier, Iron Whyalla ­ loaded DWT
141,475 tonnes (Rigby et al, 1998). Analysis of the waste energy available from the main engine
cooling system, together with the ship's usual voyage schedule, and ballast water temperature history
and laboratory investigations to identify the desirable temperature/time conditions required to kill the
major organisms likely to be of concern, identified that flushing the heated engine cooling ocean
water through the tanks and allowing the excess to overflow through the breather pipes would be the
best option (Figure 2).
In these trials heated water at approximately 41°C was flushed through the tank at a flowrate of 520
tonnes per hour. Figure 3 shows how the temperatures in various parts of the treatment tank increased
and at the end of the first trial (after 30 hours and 2.5 tank volumes of flushing) the entire tank
contents has exceeded temperatures of 38°C (Rigby et al, 1998).

2nd International Ballast Water Treatment R&D Symposium: Proceedings
On board biological observations and subsequent culturing showed that none of the zooplankton
present (mainly chaetognaths and copepods) and only very limited original phytoplankton (mainly
dinoflagellates) survived the heat treatment. The original organisms were all essentially reduced to
flocculent amorphous detritus. Subsequent culturing on samples only produced growth of some small
(5 µm) diatoms and colourless ciliates which are considered likely to be of little consequence.
Another very significant aspect of this mode of heating is that the flushing (especially with ocean
water that had been heated to some 42°C) in itself is very effective in exchanging the original water.
In the above trial, 90-99% of the original plankton was removed by flushing.
Only minor modifications were necessary on the ship to allow the heating operation to be carried out.
The installation of an additional piece of pipework allowed the overflow water to be pumped via the
existing general services pump. This modification and the operating procedures were approved by the
ship's Classification Society. The estimated total cost of carrying out this treatment (including the
capital cost for the additional pipe installation) has been estimated as 5.56c/m3 (capital 0.9c/m3,
operating 4.66c/m3). The equivalent cost for ballast exchange using continuous flushing with three
tank volumes would be 3.74c/m3 (Rigby & Taylor, 2001).
Based on the successful outcome of this trial together with the quite acceptable cost involved in this
form of ballast treatment, it would be potentially feasible to apply this mode of treatment with a
highly superior biological efficiency (compared to BWE) to most of the international ballast water
(120 million tones annually, Kerr 1994) transported to Australian ports in bulk carriers.
In addition to the added feature of flushing with biologically deficient water, this mode of heating
ensures that all of the water in the tank reaches the final minimum temperature of 38-40°C. Some of
the other options (discussed below and referred to in Table 2) involve recycling water from the ballast
tanks and in these cases mixing becomes an important issue in achieving treatment of all the water
(and organisms) in the tank. Another feature of the flushing mode is that the temperatures in the lower
sections of the tank reach a much higher temperature than the overall final minimum temperature.
This means that the sediment (and any contained biological organisms) are heated to a temperature
approaching that of the inlet water thereby enhancing the effect of destroying organisms in
accumulated sediments.
Other requirements for this form of flushing are sufficient voyage time to allow all tanks to be heated
to the desired temperature (approximately 8 days for the Iron Whyalla) and a temperature differential
between the initial ballast water and desired final temperature compatible with the amount of energy
available in the heated engine cooling water (approximately 14.5°C for the Iron Whyalla). Where
these conditions are not met, such as in colder seas or short voyages, an alternative form of heating (as
detailed below) would be required.
Case Study 2
Heating/recycling on the Iron Whyalla
Although this design has not been tested on the ship, it illustrates an alternative to that used above
which would permit the water to be treated over a shorter voyage time and would be compatible with
lower ocean temperatures. A higher final temperature is also included to demonstrate this possibility.
In this case, it is assumed that the starting water temperature is 20°C and is heated to 45-50°C. Ballast
water is recirculated from the ballast tanks through an additional preheater where it is heated to 35°C
before entering the main jacket water coolers where it reaches a temperature of 45-48°C. This heated
water is then returned to the top of the ballast tank (either via the appropriate length of pipe or via a
tank depending on the biological requirements for time/temperature) after passing through the
preheater where the water is cooled to approximately 30°C (Figure 4).

Rigby: Does heat offer a superior ballast water treatment option?
Although some small scale trials (Thornton, 2000; Mountfort et al, 1999, 2000, 2001) have attempted
to test this basic arrangement, equipment or trial conditions have not allowed the full concept to be
proven. One of the main areas to be explored is the mixing of the treated water within the ballast tank
after recirculation. Further trials are necessary to identify the specific requirements, however it is
expected that 50,000 tonnes could be treated successfully over a period of approximately 4-5 days.
Shorter times could be achieved by utilizing additional heat to allow higher recirculation rates.
This design also has the ability to treat ballast water at significantly lower starting temperatures,
simply by cooling the treated water to a lower end temperature. For example if the initial temperature
was 10°C, the end temperature before recycling back to the ballast tank would be 20°C (compared to
30°C for the earlier example).
The estimated cost of treating the water on the Iron Whyalla using this system has been estimated
(Rigby and Taylor, 2001) as 9.13 c/m3 (capital 6.6 c/m3, operating 2.53 c/m3). This compares with an
estimated total cost of 28 c/m3 for the combined use of filtration and ultraviolet irradiation using
similar cost estimation parameters (capital 27.14 c/m3, operating 0.86 c/m3).
Case Study 3
Heating ballast water during discharge on an oil tanker
Zhou (2002) has examined this case for an "Aframax" 107,000 DWT oil tanker owned by Neptune
Orient Lines. The tanker has a total ballast water capacity of 41,262 m3. The vessel uses steam driven
cargo pumps to discharge the oil product. Waste heat from the condensed steam used to drive these
pumps can be used to heat the ballast water as it is discharged (Figure 5). In this study it has been
assumed that the water needs to be heated to 65°C and this requires additional heat (over that
available from the condenser) which can be obtained from an auxiliary boiler. Using this system the
ballast water can be discharged at its normal capacity of 2,580 t/h over a total pumping tome of
approximately 16 hours.
Using a similar cost basis to that used for the Iron Whyalla analysis (Rigby and Taylor, 2001), the
estimated total cost for heating the water on this tanker would be approximately 22.44 c/m3 (capital
16.9 c/m3, operating (for additional steam cost only 5.54 c/m3).
Based on the information contained in Table 2 related to temperatures required for effective biological
control, the temperature of 65°C chosen for this study is considered to be excessive and it is likely
that the costs would be lower if a lower temperature (45°C) were used.
Case Study 4
Container ship
In the case of container ships, only small amounts of ballast water are involved when compared to
bulk oil or ore carriers. The ballast water is usually carried in a large number of tanks (27 ballast tanks
carrying 12,300 tonnes for a 3,950 TEU container ship, for example, fitted with two 550 m3/h ballast
pumps powered by a 47,520 BHP main engine). Container ships are designed to never be empty in
service and therefore only small amounts of ballast water may be loaded or unloaded at any one time
to ensure trim, stability, propeller immersion and visibility over the bow. They also use ballast water
to maintain the ship in a vertical envelope to allow the container to be slotted into the guides and
clearance under the gantry cranes whilst loading and discharging.
This heating application would also be ideally suited to passenger ships which also carry a small
quantity of ballast water.
These design and operational arrangements for ballast water mean that main engine cooling water
heating is an ideal method for treatment since the process can be carried out at low flow rates utilizing

2nd International Ballast Water Treatment R&D Symposium: Proceedings
only partial quantities of the heated water. This operation can be used over a short period or extended
periods of time to ensure that "biologically acceptable" water is available for discharge when it is
required. A number of options based on the above cases could be adopted, although the flushing
option has many advantages due to the lower capital costs of additional equipment that may be
Case Study 5
Use of heated engine cooling water as a preferred method of ballast exchange
This case offers a biologically superior treatment option for vessels that have sufficient strength to
permit the empty-refill mode of ballast water exchange to be used.
The option would involve firstly pumping the ballast water from a particular tank until it is empty
(loss of pump suction). At this stage hot water from the engine cooling system (at a temperature of
around 45°C and already biologically deficient as a result of being heated to this temperature) is then
used to refill the tank. The process is continued in a sequential pattern until all the original ballast
water has been replaced. In general this process would require longer times than the normal refill
operation but offers a superior option in cases where the safety of the ship can be guaranteed.
This method would be ideally suited to container and passenger ships that carry small quantities of
ballast water in a large number of tanks. The most suitable method would be to undertake this type of
treatment in matched port and starboard tanks, mindful of the bending moment, shear forces and
stability of the vessels.
Conclusions and Recommendations
Ballast water organisms that have the potential to initiate new invasions can be effectively killed or
inactivated by heating them to a temperature sufficient to inactivate the metabolic processes. The
lethal temperature required varies for different organisms, however as a general rule, for most marine
organisms of concern in ballast water, a temperature of 40-45°C is sufficient to achieve mortality.
Longer periods at lower temperatures are generally more effective than using short treatments at
higher temperatures.
Heating of ballast water using waste heat from the ship's main engine cooling system, high and low
temperature centralised cooling water system, auxiliary steam condenser cooling water auxiliary
boiler or other heat sources available can achieve the required conditions for a large proportion of
ships operating on both domestic and international voyages. A variety of designs are possible for a
wide range of ships to optimize the heat availability, voyage duration, sea temperatures and other
operating parameters. A series of case studies have illustrated a number of possible options, however
specific designs need to be considered for each ship based on the criteria outlined above.
Full scale shipboard trials using a combined flushing and heating design have demonstrated high
levels of biological and cost effectiveness with a superior performance to typical ballast water
exchange and other treatment options currently available. Other heating designs suitable for different
voyage conditions and ship energy balances require further exploration and trials to demonstrate and
confirm effectiveness.
Heating offers a potential ballast water treatment option and can make a significant contribution to the
future elimination of biological threats from ballast water discharges.

Rigby: Does heat offer a superior ballast water treatment option?
Bolch, C.J. & Hallegraeff, G.M. 1993. Chemical and physical treatment options to kill toxic
dinoflagellate cysts in ships' ballast water. J. Marine Env. Engg. (1): pp. 23-29.
Brock, T.D. and Madigan, M.T. (1994). Biology of microorganisms. 6th edition.Prentice Hall, USA.
Cangelosi, A. 2002. Filtration as a ballast water treatment measure. In Invasive Aquatic Species of
Europe. Distribution, Impacts and Management
. Leppakoski, E., Gollasch, S., & Olenin, S. (Ed.)
Kluwer Academic Publishers, Dordrecht, pp.511-519.
Desmarchelier P. and Wong, F. 1998. The potential for Vibrio cholerae to translocate and establish in
Australian waters
. AQIS Ballast water Research Series, Report No. 10. 45pp.
Forbes, E. & Hallegraeff, G.M. 2002. Transport of potentially toxic Pseudo-nitzschia diatom species
via ships' ballast water. In: Proceedings from the 15th International Diatom Symposium, Perth 1998
(J.John, ed.)
. Gantner Verlag. pp. 509-520.
Graney, R.L., Cherry, D.S. & Cairns, J. 1983. Heavy metal indicator potential of the Asiatic clam
(Corbula fluminea) in artificial stream systems. Hydrobiologia 102, pp. 81-88.
Hallegraeff, G.M., Vallentine, J.P., Marshall, J.A. & Bolch, C.J. 1997. Temperature tolerances of
toxic dinoflagellate cysts: application to the treatment of ships' ballast water. Aquatic Ecology 31,
pp. 47-52.
Hallegraeff, G.M. 1998. Transport of toxic dinoflagellates via ships' ballast water: bioeconomic risk
assessment and efficacy of possible ballast water management strategies. Mar. Ecol. Prog. Ser. 168:
pp. 297-309.
Jenner, H.A. & Janssen-Mommen, J.P.M. 1993. Monitoring and control of Dreissena polymorpha and
other macrofouling bivalves in the Netherlands. In: Nalepa, T.F. & Schloesser, P.W. (eds), Zebra
Mussels: Biology, Impacts and Control
, pp. 537-554. Lewis Pubs.
Johnson, G., Foerth, J., Keser, M. and Johnson, B. (1983). Thermal backwash as a method
of macrofouling control at Millstone Nuclear Power Station, Waterford, Connecticut, USA. In: Dias-
Tous, I.A., Miller, M.J .& Mussalli, Y.G. (ed), Symposium on condensor macrofouling control
technologies: The state of the art, CS-3343, Electric Power Research Institute, Palo Alto, CA
, 25-1-
Kerr, S. 1994. Ballast water ports and shipping study. Agriculture, Fisheries and Forestry ­ Australia
Ballast Water Research Series, Report No. 5, Feb.
McCarthy, S.A. 1996. Effects of temperature and salinity on survival of toxigenic Vibrio cholerae O1
in seawater. Microbial Ecology 31, pp. 167-175.
MARTOB. 2003. On board treatment of ballast water (technologies development and applications)
and application of low-sulphur marine fuel. Work Package 3 Completion Report. DTR-3.11-VTT-
03.03. www.marinetech.ncl.ac.uk/research/martob.
Montani, S., Meksumpun, S & Ichimi, K. 1995. Chemical and physical treatments for destruction of
phytoflagellate cysts. Journal of Marine Biotechnology. 2: pp. 179-181.
Mountfort, D., Dodgshun, T & Taylor, M. 2001. Ballast water treatment by heat-New Zealand
laboratory and shipboard trials. Proc. 1st International Ballast Water Treatment R&D Symposium.
IMO, London, 26-27 March
, pp. 45-50.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Mountfort, D.O., Dodgshun, T., Gibbs, W. and McCallin, B. 1999. Towards a feasible heat treatment
system for ships' ballast water. Proc. Workshop on Ballast Water Problem-Where to from here?
Ports Corp. of Qld. EcoPorts Monograph Ser. No. 19. July.
Mountfort, D.O., Hay, C., Taylor, M., Buchanan, S. & Gibbs, W. 2000. Heat treatment of ship's
ballast water: development of a model based on laboratory studies. J. Marine Env. Engg. 5, pp. 193-
Nel, R., Coetzee, P.S. & Vanniekerk, G. 1996. The evaluation of two treatments to reduce mud worm
(Polydora hoplura Claparede) infestation in commercially reared oysters (Crassostrea gigas
Thunberg). Aquaculture 141, pp. 31-39.
NRC 1996. Stemming the tide. Controlling introductions of nonindigenous species by ships' ballast
. National Academy Press, Washington, DC USA. 141pp.
Parsons, M.G. & Harkins, R.W. 2002. Full-scale particle removal performance of three types of
mechanical separation devices for the primary treatment of ballast water. Marine Technology, 39, 4,
pp. 211-222.
Rigby, G.R. 2001. Ocean exchange as a means of mitigating the risks of translocating ballast water
organisms ­ a review of progress ten years down the line. J. Marine. Env. Engg. 6, pp. 153-173.
Rigby, G.R. & Taylor, A.H. 2001. Ballast water treatment to minimise the risks of introducing
nonindigenous marine organisms into Australian ports. Review of current technologies and
comparative costs of practical options
. Agriculture, Fisheries and Forestry ­ Australia Ballast Water
Research Series, Report No. 13, Jan.
Rigby, G.R. 1994. Possible solutions to the ballast water problem, AQIS Ballast Water Symposium
Proceedings, Canberra 11-13 May
Rigby, G.R. & Hallegraeff, G.M. 1994. The transfer and control of harmful marine organisms in
shipping ballast water: behaviour of marine plankton and ballast water exchange on the MV "Iron
Whyalla". J. Marine Env. Engg., 1: pp. 91-110.
Rigby, G.R., Hallegraeff, G.M & Sutton, C.A. 1998. Novel ballast water heating and sampling trials
on the BHP ship MV Iron Whyalla.
AQIS Ballast Water Research Series, Report No.11. Oct.
Rigby, G.R., Hallegraeff, G.M. & Sutton, C.A. 1999 Novel ballast water heating technique offers
cost-effective treatment to reduce the risk of global transport of harmful marine organisms. Mar. Ecol.
Prog. Ser
. 191: pp. 289-293.
Sellers, M.A. & Stanley, J.G. 1989. Species profiles: Life histories and environmental requirements of
coastal fishes and invertebrates (North Atlantic). American oyster
. FWS/OBS-82/11.23, US Fish.and
Wildlife Service, Washington, DC. and US Army Engineering Waterways Experiment Station.
Sobol, Z., Wladyslaw, K. & Bohdan, W. 1995. System for destruction of microorganisms occurring in
ballast waters
. Paper presented at IMO MEPC 38.
Zhou, P. 2002. A rapid thermal ballast water treatment system for oil tankers. Proc. RECSO/IMO
Joint Regional Seminar on Tanker Ballast Water Management & Technologies, Burj Al Arab, Dubai,
pp. 107-114.

Rigby: Does heat offer a superior ballast water treatment option?
Figure 1. Relationship between treatment time (plotted on a logarithmic scale; in minutes) and lethal
temperatures (°C) for a wide range of marine organisms. The solid lines for dinoflagellate cysts (D),
seaweeds (W) , starfish (S) and molluscs (M) are based on Mountfort et al. (1999) supplemented by data for
vegetative stages of microalgae (A), crustaceans (C) and rotifers (R) as specified in Table 1. The overwhelming
majority of marine organisms can be killed utilising temperatures of 40-45°C in combination with treatment times

of 100-1000 mins.
Figure 2. Heating circuit used to simultaneously flush and heat ballast water on the Iron Whyalla (Rigby et al.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 3. Tank temperatures during one of the heat treatment trials on the Iron Whyalla.
Figure 4. Heating system involving recirculation of ballast water from the ballast tanks and recovery of heat using
an additional heat exchanger.

Rigby: Does heat offer a superior ballast water treatment option?
Figure 5. System proposed by Zhou (2002) for heating ballast water during deballasting.

Treatment of residual ballast water in the NOBOB ship
using heat
D.T. Stocks1, M. O'Reilly2, & W. McCracken3
1BMT Fleet Technology Ltd, Canada
2ESG Stantec Consulting Inc, Canada
3Consultant, USA
The majority of ships entering and leaving the Great Lakes do so loaded with cargo and therefore
report a "no ballast on board" condition. However, as a result of ballast tank design, residual materials
(ballast water, sediment, and biota) still remain in the bottom of the tanks when emptied. During their
visit to the Great Lakes, NOBOB ships invariably take on ballast water during cargo discharge and
loading operations. This exposes the tank residuals, which may contain invasive biota, to discharge
through re-suspension in the outgoing ballast water.
Numerous treatment technologies for the control of invasive species via the ballast water vector are
presently being evaluated, including heat treatment. Present heat treatment system designs make use
of the waste heat generated by the propulsion engines to increase the temperature of the ballast water
to a point where IAS are reduced or inactivated. To date these studies have been limited to large ships,
with treatments being carried out over relatively long distances and time periods, and in relatively
warm climates. Significant success has been demonstrated in some Australian sponsored projects and
several other waste heat-generating treatment systems have been proposed for study. The systems
may offer significant advantages in that they are relatively simple, do not produce significant
disinfection by-products (DBPs), and are effective against a majority of flora and fauna found in
ballast water.
In general, the water to sediment ratio in the NOBOB condition is considerably lower and more
variable than in full ballast tanks and this presents numerous difficulties when attempting to treat the
tank residuals. Pumping the residuals through any treatment system is made difficult due to the
limitations of the ballast tanks design, the location of the pump intakes in the tank, and proximity of
the pump intakes to the bottom of the tank. More importantly, the high sediment content of the
residual ballast water makes treatment using filtration and Ultra Violet (UV) light methods difficult,
and settled material can act as a refuge to protect certain organisms from treatment by chemical
The treatment limitations resulting from the conditions within a NOBOB ballast tank mean that direct
treatment of the tank residuals is necessary. The demonstrated heating capacity of portable boiler
systems makes this possible.
The objective of the study is to examine the use of heat as a treatment to reduce invasive aquatic
species (IAS) in the ballast tanks of NOBOB ships entering the Great Lakes. The study explores
issues such as the:

Stocks: Treatment of residual ballast water in the NOBOB ship using heat
· thermal tolerance of typical ballast water organisms;
· energy requirements to achieve target treatment temperatures in a NOBOB ship; and
· development of a thermal model to predict heat transfer and dissipation within a typical
ballast tank.
NOBOB ballast tanks are expected to contain residual water and sediment associated with normal
ballasting operations. The amount and types of aquatic species present in a ballast tank will depend on
numerous factors including transit routes, management practices, and ship design. Effective treatment
of ballast tank residuals will require that a majority of the organisms (and lifestages of those
organisms) be inactivated prior to discharge.
Thermo toxicity tests will be used to determine Lethal Temperature to 90% mortality (LTm90) and
Lethal Time to 90% Mortality (LT90) values for organisms and resting stages representative of those
found in ballast water and their associated sediments.
Heat dissipation models have been developed based on a variational finite difference methodology.
The models assume heat loss through all modes, i.e. conduction, convection and radiation, and in
order to calibrate the heat transfer coefficients a full scale trial was conducted on a ship in Toronto
Harbor, Canada. These trials had a second purpose, to demonstrate the practicality of using relatively
low energy input requirements to heat treat the residual contents of the NOBOB ballast tank typical of
ships entering the Great Lakes.
Thermo toxicity tests
The purpose of this portion of the study is to determine the LTm90 and LT90 values for typical ballast
water organisms. This is being achieved in the laboratory through controlled thermo toxicity tests
using the organisms listed in Table X. These organisms were selected for testing because;
· the organism represents an organism/lifestage expected to be found in ballast water;
· the organism is considered to be reasonably tolerant to treatment, and;
· the organisms ease of culture.
The thermo toxicity tests are designed to determine the upper treatment temperature required to
inactivate each species/lifestage, expected in the NOBOB ballast tank.
Table 1 summarizes the organisms that will be used in the thermo-toxicity tests.
Table 1. Summary of organisms selected for thermo-toxicity testing.
Organism Group
E. coli
Bacillus subtilis
Cyclops spp.
% Hatch
Daphnia magna
Daphnia magna
% Hatch
Americamysis bahia
Artemia salina
% Hatch
Dreissena polymorpha
Selenastrum capricornutum
Skeletonema costatum

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Thermotoxicity data will be compared to ballast tank thermal distribution models and energy
requirements to evaluate the efficacy of the heat treatment and determine target treatment
Ship board heat treatments
A thermal gradient model was developed to analyze heat dissipation through the steel structure of a
ballast tank to the water. This model represents the steel structure of the tank, its residual content and
the surrounding environment and uses finite difference techniques to establish a heat balance such that
thermal energy input balance heat dissipation to maintain a given temperature in the portion of ballast
water remaining in the bottom of the ballast tank.
This model can be exercised for the anticipated temperature of the operational environment using
standard coefficients of heat transfer for water and steel and both conduction and convection
components of heat transfer.
There is a significant uncertainty in the model in the dissipation of heat through convection of water
up and around the outer hull as heat is lost to the environment. Ballast tanks, being the part of ship's
structure, are inherently heavily subdivided by girders, floors and other elements. These structural
elements act as baffles and cooling fins and affect both heat losses and temperature distribution within
the tank. Furthermore, heat transfer is affected by the presence of sediment at the tank bottom of a
NOBOB ship and coefficients of thermal conductivity for such case are unknown. It is these factors
that are the primary target of calibration data to be recovered from the experiment.
Upon recovery of temperature and energy data from the experiment the model will be calibrated to
replicate the steady state energy temperature distribution.
Experimental model calibration
The experiments were conducted in Toronto Harbor on-board the ULS ship Canadian Provider in
May and June 2003. This ship is a typical Great Lakes bulk carrier. Ballast is carried in 6 sets of port
and starboard tanks which are integral side and bottom tanks. The tanks do not extend vertically to the
upper deck and for the entire length of the ship there is an access tunnel along the ship side, this
provided an ideal location for the data acquisition system.
The Canadian Provider is currently active carrying grain from Lake Superior ports to St. Lawrence
River terminals and iron ore on the return trips. The typical voyages are thus of short duration, across
lakes were ballast operations are frequent and often in shallow draft high sediment areas,
consequently the amount of sediment build up is significantly higher that in the typical trans ocean
bulk carrier. This afforded the opportunity to perform the experiment in a tank with significant
sediment and after cleaning, a sediment free tank.
Starboard ballast tank No. 6 is available for the experiment.
Equipment and materials
Name: Canadian
Type: Bulk
(as Murray Bay) in 1963 by Collingwood Shipyards, Collingwood, ON
Loa =
222.5 m
B =
22.86 m
T =
8.35 m (7.92 m)
H =
11.94 m
Capacity 27450 t (25600 t) in 6 holds

Stocks: Treatment of residual ballast water in the NOBOB ship using heat
V =
15 kn.
Powered: By a single 9,000 HP John Inglis steam turbine with 2 water tube boilers
Upper Lakes Group, Toronto, ON
Measuring equipment
Equipment used for temperature measurements consisted of Cambbell Scientific's 107B thermistors
connected to Campbell Scientific CR10 datalogger via AM 416 multiplexer and CR10WP Wiring
Panel (Figure 1).
Ambient temperatures were measured by hand-held thermometer. Since the harbour water
temperature is not subject to quick changes, it was measured only before and after each test. Air
temperature was measured and recorded at the same intervals the internal tank temperatures are
The update frequency required was low (between 5 and 15 min reading interval) temperatures were
scanned and recorded manually via CR10KD Keyboard Display unit.
Heat (steam) source
Steam for tank heating was provided by 50 HP oil-fired firetube boiler.
The Boiler was located on shore, just by the ship's side, and positioned between superstructure and aft
hold (No. 6).
Water supply for the boiler was provided by a pump submerged into the lake. Electricity supply and
fuel was provided by the ship.
Steam distribution system
Steam from the boiler was transported via 2 in. flexible pipe vertically to the deck where the pipe
entered the tank vent, routed downwards to the side ballast tank bottom and into the double bottom
tank. Inside the bottom tank the pipe ran transversely towards the ship's centre line and at 3 points in
between girders, longitudinal branches were attached to main pipe. Each of the branches consists of
short piece of 25mm flexible pipe and 1.5 m of heavy perforated steel pipe. These pipes were
submerged into the sediment or water at the bottom of the tank. No fastening were provided since the
weight of the pipes is sufficient to keep them submerged.
Circulating pump
A small pneumatic diaphragm pump was installed close to forward tank bulkhead and centrally
between girders to circulate the tank contents. Pump discharge were routed via appropriate hose to a
place near the middle steam exhaust pipe.
Test procedures
Initial Conditions (water depth in tank, air/water temperatures)
The experiment was designed to resemble the reality of the NOBOB as close as possible and therefore
modifications to initial conditions were kept to a minimum. However, some additional ballast water
was added in order to facilitate steam flow from the perforated pipes. The maximum depth of water in
the ballast tank did not exceed the depth of the bottom longitudinal stiffeners.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Transducer locations
15 107B-type thermistors were located at three longitudinal positions and five across the section.
These thermocouples were placed at the very bottom of the tank, touching the bottom plating.
Three additional probes were located 10 cm above the tank bottom in the NOBOB water, between
longitudinal girders.
One thermistor probe was placed in air in the tank center.
One thermistor was located in the cargo hold, touching the inner bottom plate directly above the
middle steam exhaust pipe.
Cables running from transducers located within the tank were routed via the forward manhole, then
vertically to the tunnel deck to a small platform within a cargo hold.
The thermometer used for outside air temperature measurements was located at the main deck, in the
shaded area near the superstructure.
Temperature sampling
Temperatures were recorded at 5 min sampling intervals
Thermotoxicity tests.
Preliminary results from range-finding experiments are nearly completed for most species. The
definitive thermotoxicity experiments will take place in July and August of 2003. A report detailing
results and methodologies will be available at project completion.
Ship-board heat treatment studies
Experiments were conducted in both mud laden and clean tanks with and without the aid of water
circulating pumps. Results from the experiments demonstrate that the temperature rises from around
15 degrees centigrade to 40 degrees centigrade can be achieved with a 50 HP boiler system over a 4-
hour period. Data from each thermistor are shown in Figure 2. These data also demonstrate the long
cool down period required to get back to ambient temperatures, i.e. in excess of 14 hours.
The temperature distribution throughout the tank with and without the circulating pump operating are
shown in the following figures, These presentations demonstrate that the heat input at one end of a
tank does not generate heat transfer through the tank sufficient to raise the temperature at the forward
end of the tank without the aid of a circulating pump. It is also evident that the large amount of
sediment in these particular tanks enabled higher temperatures to be achieved.
These data will be used to calibrate the heat models and provide a methodology to establish the heat
input necessary to achieve the thermotoxicity levels currently under development.
The heat experiment demonstrates that it is possible to heat the residual content of a NOBOB tank
effectively with a small external power source.

Stocks: Treatment of residual ballast water in the NOBOB ship using heat
Figure 1. Equipment used for temperature measurements.
Figure 2. Data from each thermistor.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 3. Clean tank + circulating pump after 4 hours heat input.
Figure 4. Clean tank no circulation.

Stocks: Treatment of residual ballast water in the NOBOB ship using heat
Figure 5. Mud laden tank no circulation.
Figure 6. Mud laden tank + circulating pump.

The use of heat for ballast water disinfection - the
AquaTherm method
G. A. Thornton & J. E. Chapman
Hi Tech Marine Pty. Ltd.
Treatment method
Physical - heat.
Project timeframe
1995 - 2003
Project aims
When commenced in 1996, the Heat Disinfection (AquaTherm, originally `WaterSafe') Project was
an offshoot of our On-board (SeaSafe-3) system. It was realised that a majority of ballast water could
be treated off the ship, giving a greater efficacy of treatment and diminishing the risks of exotic
organisms invading a marine area.
The project was commenced in 1995 with the aim of killing the Toxic Dinoflagellate Cyst, which is
regarded as a difficult organism to kill. In 1997 the treatment temperature and dwell-times were raised
to include the destruction of Vibrio cholerae which is inactivated at 73°C/30 seconds or 65°c/120
seconds (Australian Quarantine and Inspection Service (AQIS) 1997). In 1999 the treatment
temperature and dwell-time were raised again to include other human pathogens including Hepatitis A
virus which is inactivated at 90°C/60 seconds ­ in order to treat ballast water in the Great Lakes area
(Northeast-Midwest Institute (NEMWI) 1999). We have always aimed at >95% mortality for all of
the organisms on the Australian Ballast Water Management Advisory Council - Marine Target
Species List (Table 2) which may be contained in the ballast water, with on-board treatment, and
100% mortality of all discharged organisms, for shore based treatment.
Our original concept looked at the overall picture and we realised that with the number and types of
ocean-going ships there would be no single answer to the problem and certainly the logistics of fitting
and retrofitting equipment would be an immense undertaking, taxing the capabilities of any one
We determined that there are three basic ways to resolve the problem:
Load clean ballast water ­ for ships that trade on a regular route. The modification of the ship
is straightforward and the ship uses it's own ballast pumps and trimming system. This system
requires the initial removal of all sediment from the ballast tanks, and depending on the biota
to be killed, may require the addition of a residual disinfectant.
Treat the ballast water en-route ­ suitable for vessels on longer voyages. The problem with
this method is that it is not possible to eliminate all organisms; some may remain in the
sediments and in out of the way areas of the ballast tanks.
Disinfect the ballast water at the port/point of discharge. This ensures that 100% of all of the
organisms being discharged are killed, prior to the re-use or disposal of the Ballast Water.

Thornton: The use of heat for ballast water disinfection - the AquaTherm method
Project objectives
The objective of the project is to provide an environmentally acceptable, commercially viable, safe,
ballast water disinfection system. The system would be able to kill/inactivate all pathogens contained
or likely to be contained in ballast water, would be able to operate from suitable existing sources of
waste heat, where available, and would be able to operate either on-board a ship or on shore as either
a fixed installation or transportable system.
Research methods
HTM conducted research on Ballast Water (1991­1993) with the assistance of a Research and
Development laboratory using Ultrasonics, Biocides (in conjunction with Rohm and Haas),
chemicals, ultra violet, and microwaves. In 1995, after further research, it was concluded that the most
cost-effective, user friendly and environmentally acceptable process would be heat treatment.
When the temperature of a body of water containing the target organism is elevated to a temperature
above the thermal-threshold of the target organism, the target organism is killed. The thermal
threshold is the point at which an organism is instantly killed due to either denaturing of cellular
proteins or increasing the organism's metabolism beyond sustainable levels.
This thermal threshold is variable among different species, as is a species' ability to endure periods of
high temperatures that are below their thermal threshold. In general, temperatures close to an
organism's thermal threshold can be tolerated for short periods with little non-reversible damage, and
temperatures sufficiently cooler than the thermal threshold organisms can be survived for longer
The organisms requiring destruction were sourced by way of requests from various organizations
(AQIS and NEMWI). Thermal-threshold temperature/times have been provided privately (Hallegraeff
and Appendix) or are from published sources (AQIS, 1994). By taking the thermal-threshold value of
the most thermo-tolerant organism requiring destruction, and allowing for a margin of safety above
that value, we assume that organisms of lesser thermo-tolerance will succumb at the higher
Test protocols
Testing was carried out with AquaTherm systems of variously, 250, 360, and 1,000 litres per hour
Analysis has been independently carried out by Testing Laboratories accredited by the National
Association of Testing Authorities ­ Australia (NATA) to the relevant International or National
Standard. The ramp-up (25 to 30 seconds), ramp-down (25 to 30 seconds), and residence (2 minutes)
times of the AquaTherm system precludes accurate laboratory replication.
Experimental design
The AquaTherm (and SeaSafe-3) system is based on holding a body of water at a given temperature
for a given period of time, which will be fatal to a given organism and all organisms with a lower
thermo tolerance.
AquaTherm system design
The design of the AquaTherm disinfection system is technologically simple and based on
pasteurisation techniques. AquaTherm uses many off-the-shelf components.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Water to be treated is pumped through a series of heat exchangers into the water heater circuit heat
exchanger, raised to the desired temperature for the required time, and then discharged through the
heat exchanger series into the ballast tank (uptake) or local waters (discharge) after being cooled by
pre-heating the incoming water (Figure 1).
The AquaTherm system is controlled by a Programmable Logic Controller (PLC) and consists of four
Stage 1: Raw product is transferred from the source through a transfer pump into intake of the
preheat heat exchanger (primary circuit), which is recovering the heat from the disinfected product of
Stage 4.
Stage 2: Preheated raw product is then passed through a second heat exchanger (secondary circuit) to
be heated to a temperature of 85°C via a water heater.
Stage 3: Heated product then passes into a holdover tank for a specified time of 2 minutes at 85°C,
allowing disinfection to occur.
Stage 4: Disinfected product is then passed through the preheat heat exchanger (primary circuit) to
recover the heat of the disinfected product, allowing it to be used for preheating the raw product as
part of stage 1 process. The disinfected product on exit from the AquaTherm can be within as little as
2°C of inlet temperature.
The temperature difference between the inlet and discharge water (referred to as Delta T- , Actual
Temperature Difference ­ ATD, or, Approach Temperature) is the heat required (minus losses) to
raise the water to disinfection temperature ­ whatever the disinfection temperature may be (Table 1).
Table 1. Energy calculations.
· Flow in kilograms
· Temperature change in °C
· Heat Exchanger area in metres2
· 1 Kg of water raised 1°C = 1 kilocalorie (Changes marginally with salinity & temperature)
Flow in kg per Hour x Temperature change °C = Heat load in kilocalories
A smaller change in temperature requires a proportional increase in flow
(also requires a greater heat exchanger area)
1,000 kg/hr x 10°C change = 10,000 kilocalories/hr
as does:
2,000kKg/hr x 5°C change = 10,000 kilocalories/hr
There are of course practical limits
kilocalories/hr ÷ 860 = kilowatts/hour
Given the target organism's kill temperature (i.e., water temperature at which target organisms
succumb), AquaTherm will give known exposure time and known disinfection temperature. The
source of energy for the water heater(s) is flexible and can vary with what is available. Industrial
water heaters for our ballast water applications can use several sources of fuel. Other sources of heat
energy are Industrial Processes (Thornton 2000), Central Cooling Systems, Oil Refineries, Power
Stations, and Co-Generation Systems.

Thornton: The use of heat for ballast water disinfection - the AquaTherm method
Energy requirements
Energy requirements are proportional to the incoming water volume, and ATD () of the
AquaTherm (only pumping is required if using a source of waste heat). Hot water flow requirements
are approximately 1/3 of disinfection flow rate.
The organisms listed in the Australian Ballast Water Management Advisory Council - Marine
Target Species List (
ABWMAC) (Table 2) are killed by a disinfection temperature of 50°C for 30
seconds, which includes a margin of safety. The best time-series temperature data for an aquatic
organism known to be carried in ballast water is for Gymnodinium catenatum and Alexndrium
cysts (red-tide Dinoflagellate). Bolch and Hallegraeff (1993) report that 0% of G .
and A. tamarense cysts exposed to 45°C water were able to germinate (100% mortality).
Table 2. ABWMAC Marine Target Species List.*
Species Name
Common Name
Native and Introduced Regions
Organisms that are already in Australia
Sabella spallanzanii
Mediterranean fanworm
Carcinus maenas
European shore crab
Asterias amurensis
Northern Pacific seastar
Japan, Russia, Korea
Corbula gibba
Asian bivalve
Crassostrea gigas (FERAL)*
Pacific oyster
Musculista senhousia
Asian date mussel
China, Taiwan, Philippines
Undaria pinnatifida
Japanese seaweed
Alexandrium catenella
Alexandrium minutum
Alexandrium tamarense
Gymnodinium catenatum
Organisms that have not yet arrived in
Australia but pose a significant threat **
Eriocheir sinensis
Chinese mitten crab
China, Taiwan Japan, Europe,
North America
Hemigrapsus sanguineus
Asian crab
China, Taiwan, Japan, West
Caulerpa taxifolia spp
(Aquarium hybrid)
Mnemiopsis leidyi
Comb jellyfish
West Atlantic, Black and Azov
Seas, Eastern Mediterranean
Potamocorbula amurensis
Asian bivalve
China, Taiwan, North America
Dreissena bugensis
Quagga mussel
Europe, North America
Philline aurioformis
New Zealand sea slug
New Zealand, North America
Sargassum muticum
Japanese seaweed
China, Taiwan, Japan, Eastern
Pacific, Atlantic Europe
*All species listed here have been assessed as having either a severe economic and/or ecological impact. The
list was developed using an Impact Assessments Score Sheet developed by the CSIRO Centre for Research into
Introduced Marine Pests (CRIMP) and the results of the process of listing species were subsequently endorsed
by the Australian Ballast Water Management Advisory Council's Research Advisory Group (RAG) and agreed by

In addition, in their evaluation of exposure period verses effectiveness, Bolch and Hallegraeff found
that whilst 45°C water could reduce germination to 0% with a 30-second exposure time, exposure to
40°C water required 90 seconds to achieve 0% germination. Thus, if a Ballast Water treatment system
were designed for only G. catenatum and A. tamarense cysts (relatively hardy micro-organisms of
concern), 45°C water with a 30 second exposure period in the system would be adequate.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The temperature/dwell times required for G. catenatum and A. tamarense cysts' mortality are not
adequate for the destruction of V. cholerae, which requires 65°C for 2 minutes, or, 73°C for 30
seconds for complete mortality this is exceeded with the AquaTherm system.
The NEMWI required a number of organisms to be killed or inactivated (Table 3) for the Great Lakes
Ballast Water Program
. To kill the most thermo-tolerant organism, Hepatitis A virus and excluding
Clostridium refringens (Appendix), a temperature of 90°C for 60 seconds (or equivalent) is adequate.
Organisms used for testing were known to occur in the water being tested and were not introduced. As
will be discussed in the safety aspects of the AquaTherm, the attainment of the required kill
temperature is incorporated as a function of design, as is the residence time. That is, once the
AquaTherm is at operating temperature the target organisms are killed or pedantically, in the case of
viruses, inactivated, without the need for hands-on process control or adjustment.
Table 3. Great Lakes Ballast Water Program:
Northeast-Midwest Institute organisms required to be killed or inactivated.
Faecal coliforms
Microcystis elebans
Zebra mussel veligers
Faecal streptococci
Spirulina subsalsa
Thalassiosira eccentria
Adult calanoid copepods,
Clostridium perfringens
Chroococcus limneticus
Various crab & shrimp
Salmonella spp
Chroomonas amphoxera
Starfish (Asterias rubins)

E. coli
Euglena proxima
Infectious pancreatic
necrosis (fish related),

Vibrio cholerae
Amphidinium sp ate alga)
Cryptosporidium spp
Gymnodinium catenatum
Giardia spp
Hepatitis A virus
Infectious Hepatitis
Chlorella Vulgaris
Staphylococcus aureas
Nematode eggs ­ Ascaris
Effectiveness of the AquaTherm was determined as mortality/inactivation by independent
Results of Testing
AquaTherm has been tested on 14-day old Pig effluent containing greater than 15% solids,
Secondary-treated human effluent, and estuarine river water.
Queensland (QLD) Department of Primary Industries ­ Centre for Food Technology, analysed
samples from the AquaTherm 360 ­ Estuarine water, Burnett River, QLD (for non-potable industrial
use) (Table 4).
Silliker Microtech Pty Ltd analysed samples from an AquaTherm 250 ­ 14-day old untreated Pig
effluent (Table 5).

Thornton: The use of heat for ballast water disinfection - the AquaTherm method
ECOWISE Environmental Pty Ltd were commissioned to monitor the performance of an AquaTherm
250 ­ connected to an AWTS (secondary treated human effluent), and AquaTherm 1000 ­ Estuarine
water, Hawkesbury River, NSW (non-potable human use). (Tables 6-8, 10)
Environmental Pathogens performed the analysis of samples for viruses using in-house methods.
(Table 9)
Estuarine river water:
1. Burnett River, QLD:
Coliforms from 35 to <1 CFU/100ml (method AS4276.4 ­ 1995)
E. Coli from 3 to <1 CFU/100ml (method AS4276.6 ­ 1995)
Table 4. Test Results for AquaTherm 360: Analysis for pre and post-disinfection samples ­ Burnett River
estuarine water (Queensland Department of Primary Industries ­ Centre for Food Technology).
Aerobic Plate
Aerobic Plate
Count per mL
Count per mL
E. coli
(at 21°C for 72
(at 37°C for 48
Per 100mL
Per 100mL
± 2hrs)
± 2hrs)
Raw water
Raw water
Treated water
Treated water
2. Hawkesbury River, NSW ­ with aggregate and activated Carbon pre-AquaTherm filtration:
Total Coliform from 19,000 to 0 CFU/100ml (method 540.06)
Faecal Coliforms from 11 to 0 CFU/100ml (method 610.04)
E. coli from 11 to 0 CFU/100ml (method 610.04)
Pig effluent
The AquaTherm unit was found to reduce:
Thermo-tolerant Coliforms from CFU/100ml >16,000 to <2 (method M16.2)
E. coli from CFU/100ml >16,000 to <2 (method M16.3)
At 4 Days, the Thermo-tolerant Coliforms were CFU/100 ml <2 (method M16.2)
E. coli CFU/100ml <2 (method M16.3)
Table 5. Test Results for AquaTherm 250: Analysis for pre and post-disinfection samples - 14-day old raw Pig
Effluent (Silliker Microtech Pty Ltd.).
E. coli
Salmonella Stage 2
Further Testing
> 16,000
Not Detected
Further Testing
< 2
Not Detected
Further Testing
Not Detected
at 4 Days

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Secondary-treated human effluent
Daily and re-growth testing on 7-days, over a period of 3 months indicated a 100% kill of::
Total Coliform ­ pre-disinfection <1,000 to 40,000 CFU/100ml (method 540.06)
Faecal Coliforms ­ pre-disinfection 650 to 7,100 CFU/100ml (method 610.04)
E. coli - (pre-disinfection 180 to 5,900 CFU/100ml (method 610.04)
Total Plate Count ­ pre-disinfection 650 to 31,000 CFU/100ml (method 520.06)
P. aeruginosa pre-disinfection 150 to 44,000 CFU/100ml (method APHA 9213 E)
Norwalk virus*
Adenovirus reduced from 8550 to 22 units/L*
(*Environmental Pathogens ­ in house method)
Table 6. Summary of methods used by ECOWISE for all samples.
Detection Limit Method Reference
Method Summary
APHA 4500-NH3-H
Salicylate method ­ Colorimetric FIA
APHA 5210 B
Probe Method ­ 5 day incubation at 20'C
APHA 5220 C
Dichromate reflux ­ Spectrophotometric UV/VIS
E. coli
APHA 9222
Confirmed from Faecal Coliforms
Faecal Coliforms
APHA 9222
Membrane Filtration -
APHA 9213 E
Membrane Filtration -
APHA 2450
Gravimetric analysis
APHA 5310
Determined as non-purgeable organic carbon
measured by non dispersive infra-red
Total Coliforms
APHA 9222
Membrane Filtration -
APHA 4500-N B
Cadmium reduction method ­ Colorimetric FIA
APHA 4500-P I
Ascorbic acid ­ Colorimetric UV/VIS
Total Plate Count
APHA 9215 B
Pour plate method
Table 7. ECOWISE Test Results for AquaTherm 250/AWTS: Physical and Chemical analysis for pre and post-
disinfection samples.

Field Blank
Day 4
Field Blank

Thornton: The use of heat for ballast water disinfection - the AquaTherm method
Table 8. ECOWISE Test Results for AquaTherm 250/AWTS: Biological analysis for pre and post-disinfection
E. coli
P. aeruginosa
Total Plate
Pre-disinfection duplicate
Field Blank
Post-disinfection Day 4
Field Blank
Post-disinfection Day 4
Trip Blank
Trip Blank
Post-disinfection Day 4
Post-disinfection Point 1
Post-disinfection Point 2
Post-disinfection Point 1
Post-disinfection Point 2
Trip Blank
Table 9. Environmental Pathogens Test Results for AquaTherm 250/AWTS: Viral analysis for pre and post-
disinfection samples.
Hepatitis A
Table 10. ECOWISE Test Results for AquaTherm 1000/Hawkesbury River estuarine water: Biological analysis
for pre and post-disinfection samples.
E. coli
Bio ID
Plate Count
Pres Count
Conf Count
Pres Count
Conf Count
35°C 48 Hr

Heat disinfection technology is well established and is an effective application. The principal
limitations for rapid disinfection of incoming or discharging ballast water are engineering/design and
limits on energy-consumption.
The AquaTherm and SeaSafe-3 systems are two potential systems for ballast water treatment
(Thornton 1997), (Thornton 2000),(Rigby and Taylor 2001). Of the two systems, we regard the
AquaTherm as the better of the two in that all of the water is disinfected. The SeaSafe-3 system is
generally limited to the disinfection temperature of 65°C and residence time of 2 minutes, which is
adequate for V. cholerae as previously discussed. Another inherent problem with any on-board system
is to guarantee the treatment of all of the ballast water ­ in that there are dead pockets within ballast
tanks, and there will always be some sediment. As was demonstrated in the AquaTherm 250/AWTS

2nd International Ballast Water Treatment R&D Symposium: Proceedings
tests, disinfected water presents an opportunity for a viable organism to expand exponentially ­ as
there are no other organisms to conflict with or counteract it.
Using the AquaTherm system requires little alteration to the ship. It would be connected to the
outboard side of the ship's ballast pumps, and this would require moving the ship's ballast discharge
pipe to above the waterline if need be. We are led to believe that an electromagnetic coupling has
already been developed which would enable a seamless connection between the ship and the shore
piping. It may also be practical in some instances to fit a pipe from the engine room to the main deck
with a `T' head and discharge the Ballast Water through it.
The SeaSafe-3 system was designed to be easily fitted whilst building or retrofitted to a ship without
the ship being withdrawn from service. Allowance was also made so that the system could be easily
removed from a ship and fitted to another ship of similar engine capacity.
Cost effectiveness
The cost-effectiveness figures are calculated on a 515-m3/h capacity on-board and 6,000-m3/h shore-
based systems. Certain operating parameters have been assumed.
Cape size Bulk Carrier ­ ballast water capacity 55,000 m3.
Twelve voyages per year.
Useful life of ship 15 years.
Life of system 40+ years.
Indicated price of SeaSafe-3 system US$ 200,000 plus say, US$ 91,000 for pipe work.
Amortised over life of ship × voyages = US$ 1616 per voyage.
Therefore cost per m3 of ballast water carried = 2.9381 cents.
N.B. The SeaSafe-3 system uses the engine-cooling pump, as the Ballast Water is substituted for
the engine cooling water during the disinfection process. Should the pressure drop on the Engine-
Cooling Pump be too great to return the disinfected water to the ballast tank, the Auxiliary-Fire or
General Service pump would be used to assist.
This would add the following costs:
Total pumping time required = 128 hours (ballast water x 1.2 volumes) per voyage
Diesel fuel required (80 kW pump power) = 2893 kg
Fuel cost per cubic metre of ballast water on ship = 1.97 cents/m3.
Estimated maintenance cost for pump and generator = 0.56 cents/m3.
Total capital and operating cost = 5.43 cents/m3 7.
AquaTherm systems are a modular design and are manufactured in 316 stainless steel, Titanium, or
other specialised materials as required. Nominally, for marine work, Titanium is the preferred
material. The smallest system available will treat 250 Litres/hour. The largest module will treat 6,000
m3/hour; modules and can be combined to suit any greater flow requirements.

Thornton: The use of heat for ballast water disinfection - the AquaTherm method
AquaTherm system capable of 6,000 m3/h with 1°C Delta T Starting Temperature 15°C
Disinfection Temperature 80°C for 2 minutes.

Life of system 40+ years.
Indicated price of AquaTherm system US$ 40,000,000
Amortised over life of system = US$ 2739.73 per day
Per cu m of ballast water = 0.0190 cents/m3.
Energy required for Delta T of 1°C = 6,978 kW/h
Fuel (light fuel oil) required per hour (calculated at 1kg/kW/h) = 6,978 kg.
At assumed price of 20 cents/kg = US$ 1395.60
= 0.2326 cents/m3.
Total capital and operating cost = 0.2516 cents/m3.
AquaTherm system capable of 6,000 m3/h with 2°C Delta T Starting Temperature 15°C
Disinfection Temperature 80°C for 2 minutes.

Life of system 40+ years.
Indicated price of AquaTherm system US$ 30,000,000
Amortised over life of system = US$ 2053.38 per day
Per cu m of ballast water = 0.0142 cents/m3.
Energy required for Delta T of 2°C = 13,956 kW/h
Fuel (light fuel oil) required per hour (calculated at 1kg/kW/h) = 13,956 kg.
At assumed price of 20 cents/kg = US$ 2791.20
= 0.4652/m3.
Total capital and operating cost = 0.4794 cents/m3.
AquaTherm system capable of 6,000 m3/h with 5°C Delta T Starting Temperature 15°C
Disinfection Temperature 80°C for 2 minutes.

Life of system 40+ years.
Indicated price of AquaTherm system US$ 26,000,000
Amortised over life of system = US$ 1788.05 per day
Per cu m of ballast water = 0.0124 cents/m3.
Energy required for Delta T of 5°C = 34,890 kW/h
Fuel (light fuel oil) required per hour (calculated at 1kg/kW/h) = 34,890 kg.
At assumed price of 20 cents/kg = US$ 6977.60
= 1.1629 cents/m3.
Total capital and operating cost = 1.1753 cents/m3.
The costings for the AquaTherm shore-based systems as set out above illustrate the advantage of a
lower Delta T when considering the balance between equipment costs and operating costs, and also
the quantity of water able to be processed from any available waste heat source. The energy required
in the costings does not include the running of pumps as the piping distances, and thus the pump
capacity cannot be determined.
The comparison between on-board and shore-based systems is weighted towards the AquaTherm
shore-based system over the on-board system. Shore-based disinfection is even more cost effective if

2nd International Ballast Water Treatment R&D Symposium: Proceedings
there is a local source of waste heat available. Environmentally, the balance is weighted towards
shore-based treatment, as the disinfection temperature can be higher and all of the pathogens in the
discharging ballast water are killed.
The use of an AquaTherm or SeaSafe heat disinfection system cannot create safety concerns. The use
of heat to treat Ballast Water will not create a hot-water hazard.
As almost all of the operating heat is recovered in our system, there are very few safety concerns for
the ship, which would otherwise be associated with filling ships' ballast tanks with hot water.
Routing Ballast Water to or from topside or side-of-ship connection points, or to other parts of the
ship, will not alter its stability. The ship takes on or discharges ballast water in a normal manner as
determined by the ship's computer system. The associated shipboard ballast water management piping
may need minor alterations.
AquaTherm - safety
AquaTherm is controlled by two PLCs, one for the system and one for the Hot Water Service (HWS).
System PLC inputs:
HWS temperature ­ heater failure
Heater Plate Heat Exchanger (PHE) return water temperature ­ PHE is at operating temperature
Raw water feed pump flow ­ pump failure, strainer blockage
Raw water Low-level float switch ­ turns raw water feed pump off
Raw water High-level float switch ­ turns raw water feed pump on
Recirculating hot water pump flow ­ pump failure
System PLC outputs:
Raw water feed pump ­ flow control to ensure correct operating temperature
Raw water feed pump - from Low-level and High-level float switches
Chlorine dosing equipment (where applicable)
Pump failure alarm
HWS failure alarm
Digital readout of operational failure
Digital readout of Heater PHE return water temperature ­ this temperature will always equal the
disinfection temperature.
The HWS PLC ensures that the correct operating temperature is maintained to within 1°C. In the
event of any component failing, the AquaTherm system instantaneously shuts down and or reverts
to a bypass mode.
SeaSafe-3 - safety
Lloyds Register and Australian Maritime Safety Authority (AMSA) approved the SeaSafe system in
The system is designed utilising normally closed electro-valves to ensure that in the event of a
malfunction or fire the system shuts down instantaneously.
Ship stability and hull stresses are not affected by the SeaSafe system as water is drawn from the
bottom of the ballast tank and returned to the top of the same ballast tank in a continuous loop until all
of the water (plus 20%) in the tank is disinfected (Figure 3).

Thornton: The use of heat for ballast water disinfection - the AquaTherm method
In 1997, a sea-trial was conducted on the small Australian bulk carrier `Sandra Marie' primarily to
prove that the water in the Ballast tank would remain stratified. Gale force winds and big seas proved
the stratification would remain (Table 11).
The AquaTherm system exceeds the following guidelines for the destruction of pathogens in potable
(drinking) water, and sewage effluent - for water re-use: NSW Department of Health; South Australia
Department of Human Services; Australian National Health and Medical Research Council
(NHMRC); Australian and New Zealand Environment and Conservation Council (ANZECC);
Agriculture and Resource Management Council of Australia and New Zealand (ARMCANZ); U.S.
EPA; World Health Organisation.
The AquaTherm system is able to operate at 121°C with a 4 minute dwell-time, which yields sterile
water and when used for the treatment of discharging Ballast Water, will destroy all pathogens likely
to be found in Ballast Water, excepting Infectious Pancreatic Necrosis Virus (IPNV), and possibly,
Salmonid herpeviruses, Renebacterium salmoinarum, Myxosoma cerebralis, and resting spores of
fungus ichthyophonus (AQIS 1994).
AquaTherm and SeaSafe-3 are fully developed water disinfection systems that have the potential for
ballast water applications. AquaTherm is used commercially and is being considered for a number of
applications in the food, potable water, agricultural, water re-use, and sewage effluent re-use areas,
with some major sites requiring up to 22,000 tonnes per hour disinfection flows. Testing has shown
that AquaTherm is able to kill or inactivate 99.9% of all pathogens likely to be carried in ballast
HTM's AquaTherm and SeaSafe-3 systems already exist, and show the potential to be commercially
and environmentally viable, as may be the case with certain other systems.
The Regulators should set an arbitrary Standard for Ballast Water Treatment and at least stop the
further daily introduction of alien, exotic organisms and biota into the world's fragile native
The Scientific Community have a wealth of knowledge about Ballast Water-transported, non-
indigenous life forms, and should be given the opportunity to apply that knowledge in rectifying these
known, existing, problems.
With such a wide variety of ships and operational variations to be found in the world, it is improbable
that one, single, universally adopted method of treating ship's Ballast Water will emerge. There will
be scope for the use of a variety of systems in various applications, especially in the case of non-
standard vessels engaged in non-mainstream work.
We have demonstrated that for the world's large commercial fleets plying between the world's large
commercial ports, treating Ballast Water by means of heat, whether on-shore or on-board, may be one
of the economical and environment-friendly systems of choice.
A strong case has been put for the preference for on-shore over on-board treatment of Ballast Water.
The logistics of building and installing equipment on approximately 80,000 ships are daunting, as is
the potential time frame of the task.
The sooner we start to take real action the sooner the problem will be solved!

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Hallegraeff G.M. Department of Plant Science, University of Tasmania, Hobart, Australia. (personal
Hoffman, G.L. 1975. An Epidemiological Review of Possible Introductions of Fish Diseases,
Northern Pacific Seastar and Japanese Kelp Through Ship's Ballast Water
AQIS Ballast Water
Research Series, Report No.3, January 1994 - pp. 172-173. (*)
Agriculture Forestry & Fisheries Australia - Ballast Water Research Series Report No. 13, January
2001: Ballast Water Treatment to Minimise the Risks of Introducing Nonindigenous Marine
Organisms into Australian Ports
. 75.
Bolch, C.J. & Hallegraeff, G.M. 1993. Chemical and Physical options to kill toxic dinoflagellate cysts
in ships' ballast water. Journal of Marine Environmental Engineering. 1: pp. 23-29.
American Public Health Association 1998, Standard Methods for the Examination of Water and
, 20th Edition.
Agriculture Forestry & Fisheries Australia - Ballast Water Research Series Report No. 13, January
2001: Ballast Water Treatment to Minimise the Risks of Introducing Nonindigenous Marine
Organisms into Australian Ports
. 33 ­ 34; pp. 67.
Agriculture Forestry & Fisheries Australia - Ballast Water Research Series Report No. 13, January
2001: Ballast Water Treatment to Minimise the Risks of Introducing Nonindigenous Marine
Organisms into Australian Ports
. 67.
Thornton, G.A. 1997. Ballast Water Heat Treatment. Maritime Pollution Control Seminar, Australian
Defence Industries, Operations Group, Marine. Sydney, Australia.
Thornton, G.A. 1997. Ballast Water Heat Treatment. Marine Engineers Review. October 1997,
Thornton, G.A. 2000. Ballast Water Decontamination ­ Using Heat as a Biocide. Sea Australia 2000.
Sydney, Australia.
(*) Also see - Whirling Disease (Myxosoma cerebralis): Control with ultraviolet irradiation and effect
on fish. Journal of Wildlife Diseases 11: pp. 505-507.
Figure 1. Diagram of AquaTherm system.

Thornton: The use of heat for ballast water disinfection - the AquaTherm method
Scientific report re: WaterSafe 250 water treatment program
Prepared for:
Mr Robert Prentice
Hi Tech Marine Pty Ltd
P.O. Box 524, Newport, Sydney, Australia 2106.
Physician & Microbiologist,
Infectious Diseases & Public Health.
Department of Microbiology, Royal North Shore Hospital (Sydney University),
St Leonards, Sydney, Australia 2065.
November 11, 1999
A water treatment program is required to kill a number of microbial species, viruses and plankton.
The treatment process is a heat treatment that can elevate water temperature to (or above) the thermal
threshold of the target organisms. This report tabulates the thermal thresholds for the target
organisms, and makes recommendations on the required temperature and contact times.
Required Temperatures
The thermal thresholds vary for the target organisms. The most resistant of the target organisms are
spores of Clostridium perfringens. The spore stage requires high temperatures of steam at pressure for
inactivation. In reality, however, illness caused by this organism is related to errors in food
preparation that permit proliferation of the vegetative stage of the organism. A thermal process
effective against the vegetative stage of the organism is a more appropriate goal. Apart from
vegetative bacteria, the highest thermal threshold is that for hepatitis A (see table). Any process used,
therefore, should inactivate hepatitis A virus. Additionally, a margin of safety must be provided by the
process. This would be provided by at least 5-10°C above the thermal thresholds outlined in the table.
A temperature of water at 90°C for 1 minute would provide a treatment that offers inactivation of the
target organisms with an acceptable safety margin.
Dr B.J. Hudson

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Appendix table. Thermal inactivation data of known water organisms.
required for
Examples of most heat
required for contact
contact time of > 1
Class of organism
resistant organisms in
time of 1 minute or
minute to
less to inactivate the
inactivate the
Vegetative Bacteria are
Legionella pneumophila
80°C for 0.3 to 0.7
70°C for 0.7 to 2.6
Quoted times are D values
Most rapidly killed at
(see reference 1)
These include: E.coli,
coliforms,Vibrio spp,
Salmonella enteritidis PT4
71.5°C for 15
Tailing occurs at
References 1, 2
faecal streptococci,
60°C for 5mins in
(Enterococcus spp),
food studies
aureus, Fungi
Enteric Viruses
Hepatitis A virus
85°C for 1 minute
60.6°C for 19
Reference 3
achieves complete
minutes achieves
partial inactivation
Cryptosporidium parvum
72.4°C or higher for 1
Oocysts remain
Reference 4
infectious at 67.5°C
for 1 minute
Marine organisms
Gymnodinium catenatum
45°C for 30 seconds
Generally not
Reference 5
Temperature of 65.5°C is
considered above the
thermal threshold for all
aquatic organisms of
Spore forming bacteria
Clostridium perfringens
Not recommended for
121°C for 15
C.perfringens spores are
(non-vegetative forms)
inactivation of
the most susceptible to
bacterial spores
heat among the pathogenic
spore-forming bacteria
Appendix references
Gardner J.F., Peel M.M. 1991. Introduction to sterilization, disinfection and infection control. Churchill
Livingstone. (2nd edition). 49 pp.
Humpheson L., Adams M.R., Anderson W.A. et al. 1998. Biphasic thermal inactivation kinetics in Salmonella
PT4. Appl. Environm Microbiol. 64 : pp. 459-464.
Scheid R., Deinhardt F., Frosner G. et al. 1982. Inactivation of hepatitis A and B viruses and risk of iatrogenic
transmission. In Viral Hepatitis, 1981 International Symposium, Philadelphia. Franklin Institute Press,. 1982.
pp. 627-628.
Harp J.A., Fayer R., Pesch B.A. et al. 1996. Effect of pasteurization on infectivity of Cryptosporidium parvum
oocysts in water and milk. Appl. Environm Microbiol. 62 : pp. 459-464.
Hallegraeff GM, Department of Plant Science, University of Tasmania, Hobart, Australia (personal

Application study of ballast water treatment by
electrolysing seawater
K. Dang1 ,P. Yin2, P. Sun, J. Xiao & Y. Song
1 Dalian Maritime University
P.R. China
2 Dalian Maritime University
P.R. China
Treatment options being researched
The treatment method employed in this paper is the electrolysis of seawater. The raw seawater from
Xinghai bay was used as ballast water and treated by this means.
Timeframe of the project
First phase: experimental study. This phase includes the design and building of an experimental
system as well as test experiments to verify the effectiveness of this method (2002-2003).
Second phase: on board trial. The electrolysing unit is going to be installed on board a cargo ship and
operational trials will be carried out (2003-2004).
Aims and objectives of the project
To develop a model of a ballast water treatment unit that is used to treat ballast water by means of
electrolysing seawater. The capacity can meet the requirements of IMO conventions and the
requirements of ship survey. The unit will be made up of:
· seawater electro-chlorinator;
· control system for the regulating concentration of chlorine;
· piping system; and
· auxiliary equipment.
The system can regulate the chlorine concentration produced according to the content of harmful
organisms in the seawater and the temperature of the seawater. This then is used to kill all harmful
organisms and pathogens with free residual chlorine kept in a minimum level.
To make a blue print for the installation of the system on board.
Research methods
Experimental system
The experimental system has been built as shown in Figure 1. Figure 2 shows a schematic diagram of
the same. The experimental system mainly consists of:
· one storage tank 2.0 m × 1.0 m × 1.0 m;
· one electrolytic unit. 440 V, 50 A, throughput 2.5 m3/h

2nd International Ballast Water Treatment R&D Symposium: Proceedings
· one portable submerged pump, centrifugal type, with a rated capacity of 3.0 m3/h and a rated
water head of 2.5 m;
· one floater type flow meter with a measurement scale of 0-3.0 m3/h;
· large barrels, 6 × 90 L;
· sample receivers; 4 × 10 L;
Experimental procedures and testing methods
Experimental procedures
Natural seawater electrolysing. The raw seawater from Xinghai bay was simulated as ships'
ballast water and treated by electrolysis.
Electrolysing seawater with different concentrations of Artemia salina. Artemia salina
(hatched in seawater for 24-48 hrs) is used as target species to verify the effectiveness of the
process. Then the sample, which is either the hatching seawater or a mixture of the hatching
seawater and natural seawater, is treated by electrolysis. All of the samples flow through the
electrolytic cell and different electrolysing voltages are applied, but the control samples are
not treated in any way.
Testing methods:
Zooplankton: conducted in according with GB 17378.7-98 (NBS 1998) and counting under a
Phytoplankton: conducted in according with GB 17378.7-98 (NBS 1998) and counting under a
Bacteria: see Table 2.
Residual chlorine: measured with the residual chlorine indicator (colour comparison).
Live/dead Artemia salina counting: Three samples for each treated group or control group to be
taken and put into 3 Petri dishes (D = 90 mm). These samples are subjected to visual inspection
under a special light lamp or under an optical microscope.
Tables 1 to 4 are the results of natural seawater electrolysing, where C stands for control group and T
stands for treated group.
Table 1. Electrolysing condition and initial residual chlorine (natural seawater electrolysing).
Initial residual
chlorine ppm
Control group
Table 2. Quantitative analysis of bacteria (natural seawater electrolysing).
Sample volume
8.38×10 3/cm3
1/10 cm3
1/100 cm3

Dang: Application study of ballast water treatment by electrolysing seawater
Table 3. Quantitative analysis of phytoplankton (natural seawater electrolysing).
Nil found
Nil found
Nil found
Nil found
Nil found
Nil found
Nil found
Nil found
Nil found
Melosiea sulcala
Nil found
Nil found
Leptocylindrus danicus
Gomphonema sp
Nil found
Nil found
Synedra sp
Nil found
Nil found
Nil found
Nil found
Nil found
Nil found
Dunaliella sp
Dunaliella salina
Nil found
Chroococcus tenas
Nil found
Nil found
Gloeothece linearis
Oscillatotia sp
Nil found
Dactlococcopsis rhaphidioides
Nil found
Nil found
Synechococcus sp
Chamaesiphon sp
Nil found
Glenodinium sp
Diatom cysts
Nil found
Nil found
From the table it can be drawn out that 4 kinds of alga are destroyed and the total mortality is
(6,891,648-1,929,135)/6,891,648 = 72.00%

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Table 4. Quantitative analysis of protozoa (natural seawater electrolysing).
C Individual/L
T Individual/L
Euciliata sp
Nil found
Nil found
Nil found
Larvae of seashell
Nil found
Nil found
Difflugia sp
Nil found
Nil found
Nil found
Nil found
Nil found
Nil found
Tables 5 to 7 illustrate the results of electrolysing seawater with different contents of Artemia salina.
Sample O1 is natural seawater without Artemia salina. A10, A20 and B0 are control groups and A1,
A2, B1, B2, B3 are treated groups.
Table 5 Electrolysing condition and initial residual chlorine.
Initial residual chlorine
Table 6 Results of group A.
Contact time
Residual chlorine ppm
Total live individual number in 6 ml
Mortality percentage
Note: Test samples of A10, A1, A20, A2 are taken as 3 × 2ml. The average live individual is 6,056 per litre in A10 and
2,704 per litre in A20.
The variation of residual chlorine and mortality as time changes are shown in Figures 3 and 4.

Dang: Application study of ballast water treatment by electrolysing seawater
Table 7 Results of group B.
Contact time
Residual chlorine ppm
Total live individual number in 15 ml
Mortality percentage
Note: Test samples of B0, B1, B2, B3 are taken as 3 × 5 ml. The average live individual is 1,919 per litre in B0.
The sign - stands for Nil found.
The variation of residual chlorine and mortality of group B as time changes are shown in Figures 5
and 6.
AC power consumption and cost
If ships' ballast water is treated by direct electrolysis of seawater, there will be no cost for salt
consumption. The DC power consumption is about 4.5-6.5 kWh/kg active chlorine and the AC power
consumption is only about 6.0-10 kWh/kg active chlorine (NBS 1990). Supposing the secondary
electrolytic cell is employed on shipboard and the cost can be calculated in the following manner:
Assuming the AC power consumption 7.0kWh/kg active chlorine, i.e. 0.007kWh/g (NBS 1990).
Diesel oil consumption rate: 200-230g/kWh. Taking account other cost factors, such as lube oil
consumption use 270g/kWh as an assumed rate (1.25 × [200-230] = [250-287.5]g/kWh).
Diesel oil price: 365 US$/t = 0.365 US$/kg = 0.000365 US$/g.
Cost of 1kWh: 270 × 0.000365=0.09855 US$/kWh 0.10 US$/kWh.
The power cost for 1000 m3 ballast water treatment will be:
Applied chlorine
AC power consumption
Cost US$/1000t
Note: The cost just refers to the electrolytic cell power consumption and it does not include other
costs such as equipment purchase, pump operation etc.
Conclusions and recommendations
· If the raw seawater is treated by electrolysis, it can kill 4 kinds of alga from 18 kinds with an
initial chlorine concentration of 4.0 ppm. The total mortality of phytoplankton can be up
to72% and the mortality of bacteria is 99.99%. Euciliata sp in the seawater can be killed

2nd International Ballast Water Treatment R&D Symposium: Proceedings
· If the seawater with an Artemia salina density increased from 2 individual/ml to 6
individual/ml is treated by electrolysing with an initial chlorine concentration of 4.0 ppm, the
mortality of Artemia salin is more than 95% after 48 hours of contact.
· If the seawater with an Artemia salina density of not more than 2 individual/ml is treated by
electrolysing with an initial chlorine concentration of 8.0 ppm, the mortality of Artemia salina
is more than 95% after 24 hours of contact. With an initial chlorine concentration of 15 ppm,
99.99% of Artemia salina is killed after 12 hours of contact.
· If the residual chlorine in the treated seawater is less than 0.5 ppm, the chlorine will have no
effect on Artemia salina.
It is recommended that the target species used to verify the performance of any new ballast water
treatment unit or system should be selected and standardized as soon as possible.
This project is one (1B4c) of China's activities as part of the GloBallast programme. It is financially
supported by the GloBallast programme and China Ocean Shipping Company (COSCO). We would
like to express our appreciation to the GloBallast Programme Coordination Unit (PCU), GEF, UNDP,
COSCO and China SMA. We are grateful to Mr. Zhao Dianrong for his coordination between PCU
and the implementation team for this project, and further for his contribution and valuable suggestions
to the project.
NBS 1998: China National Bureau of standards, Coastal pollution biological measurement and
ecological survey
, GB 17378.7-98.
NBS 1990: China National Bureau of standards, sodium hypochlorite generator, GB 12176-90.

Dang: Application study of ballast water treatment by electrolysing seawater
Figure 1. The picture of the experimental system.
Figure 2. Schematic diagram of the experimental system.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 3. Changes in residual chlorine (Group A).
Figure 4. Changes in mortality (Group A).
Figure 5. Changes in residual chlorine (Group B).
Figure 6. Changes in mortality (Group B).

Electro-sanitization of ballast water
C.E. Leffler1, A. Rogerson2, W. Paul1, G. Germaine1, M. Elliot1, V. Antonelli3, S. Grubs2,
C. Campbell2, G. Beall4 & A. Salamone1
1Marine Environmental Partners, Inc. (MEP)
2 Oceanographic Center of Nova Southeastern
University (NSU), USA
3ISIR, Genoa, Italy
4Nanospec Company, USA
Treatment options being researched
MEP's continued development of a ballast water treatment system based on electro-ionization is the
subject of this paper. Design criteria include achieving a > 95% kill of marine biota (bacteria and
protists) found in ballast tanks and an ecologically safe discharge effluent while maintaining
engineering standards and cost effectiveness.
Research began three years ago by examining the strengths and limitations of other technologies
being considered: chemical additives, ultraviolet light, heat treatment, and others. Investigation,
through a series of laboratory, pilot, and on-board experiments, produced MEP's current version of an
electro-ionization system for treatment of ballast water. This system in its nascent form was reported
on at the 1st International Ballast Water Treatment R&D Symposium 2001 (Aliotta et al, 2001).
Timeframe of the project
This paper reports on research and development occurring between Spring 2002 and May 2003.
Objectives of the project
Refine laboratory and shipboard electro-ionization systems for ballast water sanitization.
Apply biological and chemical tests to evaluate performance and safety of treatment.
Design a scalable system to treat ballast on diverse ship types.
Biological and chemical test protocols
Electro-ionization treatment systems
On-Board Pilot Treatment System
A pilot system was installed in Spring 2002 aboard the Carnival Cruise ship the M.V. Elation
("Elation") that operated out of Long Beach, California traveling south to Mexico on a 7-day
itinerary. The treatment system was installed as a flow-through system re-circulating ballast water
at approximately 350 gallons per minute (Figure 1). This pilot system consisted of the electro-
ionization technology, comprising an air ionization module and seawater electrolysis generator.
The ionized air and the electrolyzed seawater mixture were delivered into the bulk of the ballast
water flow via a mixing manifold.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
1/20th Scale Model Treatment System
At Nova Southeastern University, a 1/20th scale model of the 2003 Carnival Elation single-pass
system was built for equipment and treatment development. The three components of the system
are undergoing testing and refinement: seawater electrolysis generator, ionized air module, and
the filtering system. Mixing of the ionized air and electrolyzed seawater with the bulk of the
ballast water is accomplished with an inline static mixer.
Current On-Board Treatment System
The electro-ionization system currently being installed on the Carnival Elation is a full-scale
single-pass system with a flow rate of approximately 1000 gallons per minute. It was designed
based on the results to date from the 1/20th scale model system and the 2002 Elation pilot system.
Sample collection and handling
Collection of ballast water samples from the ballast tank during on-board Elation tests (2002) was
effected by suction withdrawal of ballast water from a sampling tube one meter below the ballast
water surface. Sampling at discharge occurred post the discharge pump and the ORP sensor. To
collect water from the pre- and post- treatment ports, samples were collected from sampling taps
(Figure 1, a and b). In all cases, equipment and sampling bottles were sterilized prior to use. Samples
were processed after one hour of collection (except those set aside for re-growth; these were left at
room temperature for 24 h before counting). EPA method 9060A was followed for sample collection.
Baseline samples were taken from seawater alongside shipboard, Los Angeles, Port of San Pedro,
Laboratory procedures
Enumeration of organisms
Since reactive chemicals were generated in the treated ballast water, collected samples were
degassed by shaking them briefly and then allowing them to stand for one hour before processing
for biota.
A. Bacteria
Total culturable bacteria were counted using standard plate counting and/or membrane filtration
methods. For plate counting, samples were processed by serial dilution and aliquots (0.1 ml) were
spread on the surface of a Marine Agar (Difco) plate (EPA method 9215C). The heterotrophic
plate count (HPC), formerly known as the standard plate count, is a procedure for estimating the
number of live heterotrophic bacteria in water. All counting was replicated three times. After
treatment, when bacterial counts were significantly reduced, samples were processed by
membrane filtration counting (EPA method 9215D). Aliquots of treated water (10, 50 and 100 ml)
were filtered through a sterile 0.45 µm filter to collect bacteria. Filters were placed on Marine
Agar plates and incubated. All incubations were at room temperature (around 23°C) for five days.
Thereafter, the number of colony forming units (cfu's) was recorded for both untreated and
treated water.
B. Protists (algae and protozoa)
Protists span a wide diversity of forms including naked amoebae, heterotrophic flagellates,
ciliates, diatoms, autotrophic flagellates, dinoflagellates and non-motile algae. Consequently, the
protists represent an important diverse array of eukaryotic microbes that are useful for verification
testing when dealing with "indigenous" organisms. Some of these protists form resistant resting
stages (cysts), such as those of dinoflagellates. Depending on native populations in the water
column, the protistan count included trophic and resistant stages.
The aliquot method has routinely been used for the enumeration of heterotrophic protozoa, but is
also appropriate for autotrophic protists if inoculated dishes are incubated in light. Small samples
of water withdrawn from the ballast tanks (ca 50 ml) were vortexed to randomly distribute protists

Leffler: Electro-sanitization of ballast water
in the sample (often, protists are located on suspended flocs). To count protists, aliquots (10 to
40 µl) of water were micropipetted into the wells of tissue culture plates. For each of the five
replicates from each sample, 48 tissue culture wells were inoculated. Each well also contained one
ml of sterile seawater and a one cm3 bloc of malt-yeast agar. Nutrients diffusing from the agar
nourished attendant bacteria that in turn were grazed by protozoa present in the inoculums. Plates
were incubated at 20°C in the dark and were examined after seven days for the presence of
protozoa (amoebae, flagellates, and ciliates). An inverted phase contrast microscope was used to
examine the base of wells. It was assumed that a well with positive growth originated from a
single cell added in the inoculums. In this way, the total number of protozoa in the total volume
inoculated (48 × inoculation volume) was calculated and expressed as density per ml.
The number of autotrophic protists in the sample was estimated in a similar manner as above. In
addition, seawater was enriched with a dilute soil extract solution to supply trace nutrients
essential for the growth of algae. Cultures were incubated in the light to promote the growth of
autotrophs in the wells.
Chemical and Physical Evaluation
Chemical and physical parameters were monitored to provide baseline information on the pre-
and post- treated ballast water. This monitoring provided information on changes resulting from
the ballast water treatment. In particular, due to potential chlorine or bromine residuals in the
treated water, trihalomethanes (THM) were analyzed by mass spectroscopy at Spectrum
Laboratories, Ft Lauderdale, Florida, using EPA method 8260. This method is capable of
detecting 58 common organic compounds. Triplicate samples were collected and monitored for
the following (all protocols in Eaton, et al. 1995; the EPA method reference number for each is
indicated in parenthesis):
a) Dissolved oxygen (4500-0 G. Membrane Electrode Method)
b) pH (4500-H+)
c) Temperature (2550)
d) Conductivity/Salinity (2520 B. Electrical Conductivity Method)
e) Turbidity (2130 B, Nephelometric Method)
f) Chlorinated/Brominated organics (Mass Spectroscopy Method 8260)
conducted by Spectrum Laboratories
g) Chlorine/Bromine (4500-ClF. DPD Ferrous Titrimetric Method)
h) Oxidative Reduction Potential (ORP) (2580 Oxidation-Reduction Potential)
i) Reduction Potential Analysis
conducted by Nanospec Company
Acute and Chronic Toxicity
The effluent from the on-board testing was not subjected to acute and chronic toxicity testing,
however, a 1/20th scale model, located at the Oceanographic Center of NSU, was used to generate
treated water for testing. The pilot system is a modified single-pass system similar to the system
currently being installed on Carnival Elation.
Toxicity tests were carried out by Toxikon Corporation, Jupiter, Florida, to determine the acute
toxicity of treated effluents to the mysid shrimp, Mysidopsis bahia. The methodologies used for
the 96-hour acute static definitive studies were based on those described in EPA/600/4-90/027F
"Methods for Measuring the Acute Toxicology of Effluents and Receiving Waters to Freshwater
and Marine Organisms". The tests consisted of exposure of Mysidopsis bahia to nominal

2nd International Ballast Water Treatment R&D Symposium: Proceedings
concentrations of 6.25, 12.5, 25, 50 and 100% treated effluent. The effluents were produced by
treatment per the electro-ionization process of this paper. Samples of the effluent were collected at
discharge in sealed containers and transported for testing in an ice-filled container. Reactive
halogens were measured within four hours (0.04 ppm ­ 0.12 ppm) of sample collection and
toxicity testing was begun within a 24-hour period. A dilution water control of filtered laboratory
saltwater, as well as a raw seawater control (reactive halogen-containing chemicals ­ 0.00 ppm)
from the NSU boat basin were run concurrently.
Seven-day definitive chronic toxicity testing, by Toxikon Corporation, was conducted on electro-
ionization treated effluent containing two times the maximum normally produced residual
chlorine, 1.2 ppm, to simulate extreme conditions. Methods for the seven-day static-renewal
definitive test followed "Short Term Methods for Estimating the Chronic Toxicity of Effluents
and Receiving Water to Marine and Estuarine Organisms" (EPA/600/4-91/003 - U.S. EPA, 1994).
The test consisted of exposure of mysid shrimp to 100% (undiluted), 50%, 25%, 12.5% and
6.25% test effluent, as well as a laboratory saltwater control and de-chlorinated effluent (100%
only). The endpoints observed in this study were survival, growth (via dry weight), and fecundity
(ability of the females to reproduce).
A subsequent seven-day definitive chronic test was also conducted using effluent containing
residual halogens produced at normal operating levels (0.04 ppm).
All statistical analyses for chronic tests were conducted following the decision tree for analysis of
survival or growth (via dry weight) utilizing the statistical program Toxcalc, Version 5.0. The
survival data was transformed using the one-tailed arcsine-square root transformation. Statistical
comparison of each effluent testing concentration was made to the control replicates tested
concurrently with the effluent testing concentrations.
The statistical comparisons for survival data used in this study include Shapiro-Wilk's Test for
normal distribution. Additionally, Dunnett's Test was used for the Hypothesis Test of survival
data and Trimmed Spearman-Karber for the LC50 calculation.
The statistical comparisons for the growth via dry weight data used in this study include Shapiro-
Wilk's Test for normal distribution and Barlett's Test for the equality of variances. Additionally,
Dunnett's Test was used for the Hypothesis Test of growth data.
Marine Environmental Partners Inc. (MEP) (with C. E. Bud Leffler as the lead technical investigator)
and the Oceanographic Center of Nova Southeastern University (NSU), (with Dr. Andrew Rogerson
as the lead independent investigator in biological testing) have evaluated multiple processes for
sanitizing ballast water and found electro-ionization technology to be a promising option. Electro-
ionization is a treatment method which has been used to disinfect freshwater effluents that MEP
modified and applied to treat marine and estuarine waters.
Data on various treatment configurations employing electro-ionization technology was collected over
the last two years. Generally, the results show the technology, generating halogen residues of around
0.5 ppm, to be capable of killing (or inactivating) approximately 95% of indigenous (i.e. native),
culturable bacteria in water from Port Everglades, Florida. On one occasion, up to 99% of bacteria
were killed or inactivated. Trials conducted on indigenous protist (algae and protozoa) indicate a kill
efficiency of around 90%. These promising results guided the evolution of the treatment system to its
present configuration.
Pilot system tests on board the Carnival M/S Elation
In January 2002, MEP installed a pilot system on the Carnival cruise ship Elation. The pilot electro-
ionization system on the Carnival Elation was tested on one 200 m3 ballast tank at a flow rate of

Leffler: Electro-sanitization of ballast water
350 gpm. This pilot system was installed to function as a re-circulating system providing continuous
electro-ionization (sanitization) of the ballast water.
Based on prior laboratory tests, the shipboard pilot system was designed in such a manner that a
slipstream was diverted from the main ballast to feed several electrolysis cells for generation of
primary disinfectants. Each electrolysis cell (three in total) was 1.5" in diameter with an output of
158 oz of halogens/24 hours/cell. This slipstream and the airflow from the gas ion generators were
then introduced via a mixing module, known on the ship as "the octopus", for combining the ionized
air (gas) and halogen species, in tandem, with the ballast water to kill biota. The shipboard prototype
system utilized air compressors so that a precise amount of ionized gases was injected into the system.
Using the continuous recycling system, bacteria and protists were reduced by over 95% during the
first 20 h of treatment (Figures 2 and 3). In these runs, starting bacterial counts were close to 120 ×
103 bacteria per ml and total protists were around 20 per ml. Of course, these counts are significant
underestimates of the true numbers of bacteria and protists in the ballast water, since not all were
amenable to laboratory cultivation (in the enrichment counting methods used). However, they give a
representative `index' of kill rate and it is expected that these underestimates reflect actual die-offs in
the treated water. However, in the next 20 h, numbers of biota in the tanks recovered and reached
levels equivalent to the starting concentration. This was probably due the fact that the system did not
run continuously on-board ship. Although continuous treatment had been planned over the course of
the voyage, ship-operating procedures required the use of the ballast pump resulting in frequent shut-
downs of the system (Figure 4). Hence, the dramatic recoveries in counts probably resulted from the
lengthy down-times around the 35 to 70 hour times. This recovery after treatment is an example of re-
growth. This was investigated further by storing treated samples for 24 hours prior to counting
bacterial levels. Figure 5 shows that immediately after treatment (in the case of the samples at 24 h,
48 h, 72h, and 96 h), the numbers of bacteria were low but that there was a rapid population increase
over the next 24 h as surviving bacteria quickly became re-established because of high nutrient loads
(from lysed biota) and reduced competition. To attain the densities observed, bacteria were replicating
in a few hours. For example, the 72 h data showed that bacteria increased from 94 × 103 bacteria per
ml to 930 × 103 bacteria per ml, which indicated the bacteria were dividing approximately every eight
hours. The scale of recovery in these incubation bottles was comparable to the recoveries observed in
the ballast tank treatments.
Mass spectrometry was used to determine the chemical species generated by this system. The
seawater electrolysis cell module was found to generate reactive bromine ions (Spectrum
Laboratories, Ft. Lauderdale, FL). Concurrently, atmospheric air was ionized into various (undefined)
species of oxygen and nitrogen in the air ionization module. These ionized species probably included
various singlet molecular oxygen species, ionized nitrogen, and peroxyl ions (e.g. O -
2 , O2 , N2 , e-,
H2O2, OH-). The ionized air (gas) stream was fed into the electrolyzed seawater stream where reaction
occurred with the previously electrolyzed bromine, thereby enhancing biota termination.
Utilizing the mass spectrometry data along with reduction potential analysis, the electro-ionization
system was theorized to utilize a combination of hydrogen peroxide, oxygen species, and bromine
species as disinfectants (Nanospec Company, San Marcos, TX). Ozone did not appear to be a factor in
the chemical reactants. This was consistent with the analytical data where bromo species (40­54 ppb
bromoform and 2-11 ppb dibromochloromethane) were the main trace contaminants left in the
seawater (Spectrum Laboratories). Both oxygen and hydrogen peroxide are expected to dissipate
rapidly in oceans to environmental levels. In summary, the net result of the treatment system was the
disinfection of the ballast water using trace amounts of bromoform and even smaller amounts of
dibromochloromethane, with no persistent disinfectant species released to the marine environment. It
must be recognized that even the bromoform concentration was below levels normally applied to
drinking water.
Notably, the use of chlorine in potable water is known to react with organic materials in water and
form a variety of carcinogenic trihalomethanes (THMs) and other molecular species. Therefore, the

2nd International Ballast Water Treatment R&D Symposium: Proceedings
U.S. Environmental Protection Agency (EPA) set an absolute limit of 100 ppb for THMs in any
potable water system. As existing discharge standards do not address the presence of THMs, MEP
tested its shipboard ballast water for THMs utilizing EPA drinking water standards and found it to be
well within the EPA drinking water standards [bromoform (80-100 ppb), dibromochloromethane (0.5-
3.1 ppb), and dibromoethane (1-4 ppb)]. Furthermore, no detectable THMs were present at the point
of discharge. MEP confirmed ballast water processed in its recirculating system remains within EPA's
parameter even when fluid is circulated for several days during the electro-ionization process
(Spectrum Laboratories).
The Carnival Elation tests demonstrated that 0.5 ppm of halogen (0.35 ppm free and 0.15 ppm
combined) produced effective kills in the greater than 95% range.
1/20 scale model single pass system
Because of re-growth issues during idle treatment periods with the recirculating system, constant
treatment of ballast water during passage was deemed undesirable. Therefore, although results from
the recirculating pilot system were promising, it was concluded that sanitization treatment should
occur at de-ballasting to ensure the highest flexibility for on-board operation. Hence, the transition to
the single-pass system is presently being made.
In preparation for installation of the current system on the Elation and to provide a testing platform
for new methodologies and equipment, MEP installed a 1/20-scale model single-pass electro-
sanitization ballast water system at the Oceanographic Center of Nova Southeastern University. To
date, the system has primarily been run to provide treated water for toxicity tests.
Acute exposure results on discharged ballast water indicate no surrogate organism (mysid shrimp)
death at electro-ionization treatment levels required for 95% on-ship ballast water biota kill (0.04 ­
0.12 ppm residual halogens). Chronic static exposure (seven days) testing of discharged ballast water
on mysid shrimp, at treatment levels required for 95% on-ship ballast water biota kill (0.04 ppm
residual halogens), also indicated no-impact to growth or ability to reproduce. All chronic and acute
toxicity testing is being performed at Toxikon Corporation. A critical element of this ballast water
system is its lack of environmental impact upon discharge.
After 96 hours of acute exposure, there was 0% mortality in all test controls and all test effluents
produced by treatment per the electro-ionization process utilizing a full capacity system and even at
1.2 ppm residual halogen if neutralized prior to discharge. The LC50 cannot be calculated due to the
lack of significant mortality during the 96-hour exposure when compared to the controls. Therefore,
the LC50 value is greater than the highest test concentration or > 100% effluent and the NOEC (no
observable effects concentration) can be stated to be 100% effluent.
After seven days of chronic exposure to treated ballast water containing 0.04 ppm of total residual
halogen, mysid mortality was zero percent in testing concentrations of 6.25%, 25%, 50%, and 100%
test effluent in seawater. Testing concentration 12.5% yielded 5% mortality. The LOEC value (lowest
observable effects concentration) for survival was 100% test effluent. The LC50 is calculated to be
100% test effluent ­ the treated ballast water at discharge does not kill mysid shrimp even upon seven
days of full exposure.
The discharged treated ballast water (0.04 ppm total residual halogen) was found to have no affect on
the mysid growth. At test termination, mysid growth (as average dry weight per mysid in each
replicate) ranged from 0.39 mg to 1.1 mg for surviving mysids exposed to treated effluent. The
average dry weight of the laboratory control animals was 0.69 mg; and was 0.55 mg in raw seawater.
There was no statistical difference in dry weight between the treated tests and the controls. Therefore,
again the LOEC was calculated to be > 100% effluent and the NOEC value was 100% effluent. In
summary, the mysid shrimp grew to the same size whether living in discharged treated ballast or raw

Leffler: Electro-sanitization of ballast water
Also the discharged treated ballast water (0.04 ppm total residual halogen) did not affect fecundity,
the ability to produce offspring. Mysid fecundity is expressed as the percentage of gravid and/or
ovigerous females bearing eggs observed at the termination of the study. With treated effluent
containing 0.04 ppm residual halogen, the fecundity of female mysids was 100% in each testing
concentration that contained identifiable females. Laboratory and seawater controls produced 100%
fecundity. Therefore, the LOEC value for fecundity was >100% effluent, while the NOEC value was
100% effluent.
In order to test an extreme situation, seven-day chronic exposure tests with treated effluent containing
1.2 ppm residual halogen (two times the level required for treatment) produced mysid mortality of
zero percent at 6.25% and 12.5% test effluent. Testing concentrations 25% and 50% yielded three
percent mortality. The 100% test treatment yielded one hundred percent mortality. Mortality was zero
percent in 100% de-chlorinated effluent and in the laboratory saltwater control. The LOEC value for
survival was 100% test effluent and the NOEC value for survival was 50% test effluent. The LC50 was
calculated to be 68% test effluent. With this testing series, the mysid shrimp exposed to 50% effluent
or less, demonstrated no statistically significant difference in dry weight when compared to the
laboratory controls or raw seawater controls. Therefore, based upon growth via dry weight, the LOEC
value was calculated at > 50% effluent and the NOEC value was 50% effluent.
Summary of electro-ionization system testing results
Elation pilot recirculating system
1/20 scale single-pass system
Biological testing
Bacteria kill
> 95%
> 95%
Protists kill
> 90%
Effluent toxicology ­ electro-ionization system at full capacity (95% biota kill or higher)
- 96-hr acute
NOEC = 100% effluent
LD50 > 100% effluent
- 7-dy chronic mortality
NOCE = 100% effluent
LOEC > 100% effluent
LD50 > 100% effluent
- 7-dy chronic growth
No effect
- 7-dy chronic fecundity
No effect
Chemical analysis at full capacity (95% biota kill or higher)
- treatment halogen
0.4 - 0.5 ppm
0.4 - 0.5 ppm
- halogen residuals
0.04 ­ 0.12 ppm
- THM residuals on ship
80 ­ 100 ppb bromoform
0.5 ­ 3.1 ppb dibromochloromethane
1 ­ 4 ppb dibromoethane
at discharge
not detectable
System currently being installed on-board the Elation
The most effective ballast water treatment system found to date utilizes three technologies - solids
removal, electrolysis and ionization - in a single pass system. This system will continue to develop
over its five-year testing plan on the Carnival Elation. As the ship changes routing and encompasses
various silt loads and other variables, MEP expects to develop a database of information to refine
engineering of future systems.
Currently, at intake, the ballast water is filtered to remove solids larger than 50 microns. During
transit the ballast water is left without treatment. Just prior to de-ballasting, the ballast water is re-
filtered to remove any large particles or biota formed during transit. This filtrate is handled as required

2nd International Ballast Water Treatment R&D Symposium: Proceedings
by management. The filtered ballast water is then sanitized as it flows through a static mixer with
reactive chemicals formed from electrolysis of the ballast water and ionization of ambient air.
Discharge occurs immediately following sanitization.
MEP has built a 1000 gpm unit, which is being installed on the Carnival Elation. Each of the system
modules will meet class certification and is built to perform over the life of the ship. The system is
designed with integrated power and control systems driven by a PLC (programmable logic controller)
module, which monitors and controls over 300 points. This system, to be tested over a five-year
period, is designed to operate at a cost of $0.005 or 1/2 cent per metric ton based on $0.16/kW hr
energy charge as estimated by Carnival Cruise Lines.
Solids removal
After testing alternative means of solids removal, filtration modules were selected and are deployed
based on the quality of the seawater influent anticipated aboard a given ship as determined by its
proposed routing.
Based on anticipated regulatory action and system efficiency, MEP is using a self-cleaning 50 micron
filtration module (Figure 6) controlled by a PLC (programmable logic controller). When taking on
ballast, the discharge of the filtrate can be returned overboard, as only local species should be present
during ballast intake.
When de-ballasting begins the PLC initiates the filtration process utilizing the same filter on the
ballast water held during transit. Since a ship is normally in port at this stage, the filtrate must be
managed onboard per ship management requirements.
Electrolysis and ionization
The original pilot electro-ionization system on the Carnival Elation was tested on one 200 m3 ballast
tank at a flow rate of 350 gpm. The current electro-ionization system, which is being installed in
May/June 2003, processes 1000 gpm of ballast water with a single pass system. This increase in
treatment rate from 350 gpm to 1000 gpm was accomplished by the addition of parallel electrolysis
and ionization modules as well as module design efficiencies.
The pilot ionization gas-generating cylinders were 30" long by 4" in diameter. Two of these cylinders
generate 1.25 cfm of airflow; four generate 2.5 cfm of airflow, the number used on the pilot Carnival
Elation tests (Figure 7). By doubling the length of the cylinders to 60" and increasing the diameter to
8", MEP effected another quadrupling of output while reducing the equipment footprint and cost. In
other improvements, the original ionization unit was water jacketed for cooling; air-cooling has been
found to be preferable.
The current units are rack-mounted onboard the Elation (Figure 8). They may be remote mounted to a
bulkhead or elsewhere in other applications. The ionization generator rack is monitored for flow from
each cylinder as well as temperature, pressure, etc. Photocopies are included of both the pilot and the
current ionization gas generators.
Following pilot testing on the Carnival Elation and on MEP's 1/20-scale model at Nova Southeastern
University, it was determined to not use a ballast water slipstream for electrolysis of ballast water.
Instead the ballast water, in its entirety, flows through the electrolysis cell as shown in Figure 9. The
current module contains 10 cells for a combined capability of 1,580 oz of halogens/24 hours, which
was designed to treat ballast water pumping at 1000 gpm while being discharged. The halogens which
are sanitizing the ballast water are > 99% bromine species (as detected by mass spectrometry).
Electrolysis cell diameter and parallel electrolysis modules are planned to increase capacity to 6,500
gpm for cargo ship requirements. Further design modifications are planned to reduce sensor costs for
controlling the electrolysis module and to decrease the number of parts.

Leffler: Electro-sanitization of ballast water
Effective mixing of the sanitizing agents within seawater held in the reaction vessel is a key research
area, which has resulted in the engineering of a unique mixing module.
Static mixing
Prior to shipboard installation, MEP designed mock ballast water tanks to simulate the transmissive
qualities of the ionized gases and electrolyzed seawater into the ballast water. In order to expose all of
the ballast water in a tank to the treatment process, the reactants had to be distributed in a manner
where virtually all fluids would come into contact with the sterilizing ionized gases and electrolyzed
bromine species.
In the earlier shipboard pilot testing, the electrolyzed ballast water slipstream and the airflow from the
gas ion generators were introduced to each other via a mixing module known on the ship as "the
octopus" (Figure 10).
With the current electro-ionization system, the "octopus" has been replaced by a mix manifold
module, which provides thorough mixing of the ionized air and electrolyzed seawater with the bulk of
the ballast water for disinfection, while stabilizing the water chemistry in preparation for discharge
(Figure 11).
The controls
The ballast water treatment system is controlled and monitored by an electronic control system. The
controls are installed in four cabinets that start with a stabilized and conditioned power supply to the
PLC, which monitors and drives the entire system (Figure 12). The PLC monitors the system and can
self- repair by turning on additional back-up units if it senses a problem, and is designed to self-report
and generate information for remote troubleshooting.
The testing of the control system was designed to integrate the modules and determine the least
amount of equipment/energy required to fully sanitize the biota. MEP uses periodic biological plate
counts to validate the control system operation.
Conclusions and recommendations
MEP's electro-ionization system shows promise for use in sanitizing ballast water. The system, as
tested on Carnival's Elation and in the 1/20-scale model, disinfected seawater (California coast,
Pacific Ocean, and Florida coast, Atlantic Ocean) to at least a 95% kill of bacteria.
The effluent's safety also appears promising. No detectable trihalomethanes were present at de-ballast
from the Elation pilot trials. The concentrations of reactive halogens present at ballast discharge from
the 1/20th scale model preliminary tests were ecologically non-toxic producing no mysid shrimp
mortality and no effect on mysid shrimp growth or fecundity.
Chemical and biological research methods that were tested provided useful information for system
improvements and for determining efficacy and safety. Further development of an ATP rapid
detection protocol for living biota is planned, as well as further protocol development for effluent
Formal testing, for the California Lands Commission, of the system currently being installed on the
Carnival Elation is expected to begin in Summer 2003. In the meantime, equipment refinements are
now concentrating on defining the least concentration of bromine species required to effect a high
elimination of biota. This concentration, as determined from the 1/20th scale model, will become the
starting point for the upcoming on-ship system testing.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Aliotta, J, Rogerson, A., Campbell, C. & Yonge, M. 2001. Electro-ionization treatment for ballast
water; assessment of effectiveness against marine microbiota typical of those in ballast water.
IMarEST 1st International Ballast Water Treatment R& D Symposium.
Nanospec Company, Gary W. Beall, Director of Southwest Texas State University, Center for
Nanophase Research, Institute for Environmental and Industrial Science, 601 University Drive, San
Marcos, TX 78666-4616, USA.
Spectrum Laboratories, 1460 W. McNab Road, Ft. Lauderdale, FL 33309, USA.
Toxikon Corporation, 106 Coastal Way, Jupiter, FL 33477, USA.

Leffler: Electro-sanitization of ballast water
Figure 1. Diagram of ballast tank with re-circulating pilot treatment system (T) on Carnival Elation. Tank was
baffled (B). Water was taken from an entry hatch (OUT), pumped through the treatment system with the ballast
pump (P), and returned into another entry hatch (IN). Sampling ports (a and b) are indicated. Diagram not to
Figure 2. Percent survival of bacteria in the pilot ballast tank as a function of treatment time. Samples were
collected from the sample port `a' located before the treatment system and at port `b' located after the treatment
system. Means of 5 replicate runs with standard errors.
Figure 3. Percent survival of total protists in the treated tank (solid line) and the control tank (dotted line).
Samples were collected from the `before' sample port.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 4. Time line showing functioning treatment times (white) and treatment down-times (black) of the on-
board recirculating treatment system.
Bacteria per ml
Figure 5. Bacteria per ml in treated samples (collected from sample port `b' at 0 h. 20 h, 72 h, and 96 h) after a
24 h incubation period at room temperature. Data as means of 5 replicates with standard errors. Black bars =
numbers before incubation: open bars = numbers of bacteria after incubation. Note, the time 0 bars are similar
since there was no treatment and therefore no regrowth.
Figure 6. Self-cleaning filter for Carnival Elation current system.

Leffler: Electro-sanitization of ballast water
Figure 7. Racks of ionization generators (30) on Carnival Elation pilot project.
Figure 8. Ionization generator rack module for Carnival Elation current system.
Figure 9. Electrolysis module for Carnival Elation current system.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 10. The "octopus" onboard the Carnival Elation pilot project.
Figure 11. Shear mix manifold module for Carnival Elation current system.
Figure 12. Electrolysis module power cabinet - one of four power and control cabinets.

Superconducting magnetic separator for ballast-water
N. Saho, H. Isogami, T. Mizumori & N. Nishijima
Mechanical Engineering Research
Laboratory of Hitachi Ltd.
Treatment options being research
Mechanical (filtration and magnetic separation).
Timeframe of the project
Phase 1: 2003
Basic research on superconducting magnetic separation system for ballast water treatment.
Phase 2: 2004
Detailed design of superconducting magnetic separation system on board.
Phase 3: 2005
On board testing of the superconducting magnetic separation system
Aims and objectives of the project
The aim is to develop a ballast water treatment system that is suitable for rapidly purifying ballast
water on board.
Research methods, test protocols and experimental design
A prototype water treatment system using a superconducting magnet to clean the ballast water
discharged from ship was developed. The system is capable of treating 100 cubic meters of
contaminated water a day through the following process sequence: mixing contaminated water with
magnetic powder and a flocculant, stirring the mixture to make magnetic flocs, filtering the flocs,
transferring them to a rotary magnetic shell, and dumping them in a sludge tank. The system was
evaluated in experiments on two types of contaminated water samples, one containing kaolin particles
and the other crude oil.
Prototype structure
As the name implies, the new water-treatment system (Saho N., 2000) combines filtration and
magnetic separation. As Figure 1 shows, the treatment process is divided into three steps. First, a pre-
application treatment unit gathers the targeted contaminants in the influent into magnetic flocs.
Second, a filtration unit filters these magnetic flocs through a rotating filter to purify the water. Third,
a magnetic separator unit attracts the flocs deposited on the surface of the filter, washes the surface for
reuse, and recovers the magnetic flocs as highly concentrated sludge.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
In the pre-application-treatment step, ferromagnetic magnetic particles (Fe3O4), a flocculant
(Fe2(SO4)3-nH2O), and a polymer are introduced into the influent. They are then stirred and mixed
into the influent to produce magnetic flocs (containing contaminant particulates and ferromagnetic
particles) to magnetize nonmagnetic contaminants, such as fine organic matter. Three to four minutes
is an adequate time for this stirring and mixing.
Since most of the particulates in the generated magnetic flocs have a diameter of several hundred
microns, we considered a single filter would be sufficient to trap and filter these flocs if it had a
micropore diameter of several tens of microns. The filter unit with a frame and a wire net is shown in
Figure 2. A stainless-steel wire net with a pore diameter of 43 µm is used as the filter, and the width
of the aperture inside the frame is 200 mm. Twelve filter units, forming a rotating micro-pore filter,
are located on the outer circumference of a rotating shell with an outer diameter of 400 mm (see
Figure 1).
To enable continuous treatment, the system has a configuration with a rotating filter fitted to the outer
circumference of a rotating drum, and the pre-treated water is passed from the outside to the inside of
the filter. The magnetic flocs are trapped and accumulated on the surface of the rotating filter, the
influent is purified and flows to the inside of the drum, and the purified water is discharged from the
system. The magnetic flocs accumulated on the filter in the water migrate from the rotating filter
toward the high-temperature superconducting (HTS) bulk magnet positioned near the surface of the
pre-treated water. They are separated from the filter by a shower of water from inside the filter near
the surface, so the filter is continuously cleaned and is thus always ready for reuse.
In the magnetic separator, the HTS bulk magnet, magnetized in advance and chilled by a cooler inside
a vacuum adiabatic chamber, is fixed inside a nonmagnetic rotating cylinder. The separated magnetic
flocs adrift in the magnetic field near the surface migrate swiftly, drawn by the strong attraction of the
HTS bulk magnet. The migrating magnetic flocs adhere to the surface of the cylinder and are then
ejected into the atmosphere above the surface of the water by the rotation of the cylinder. At this
point, the surplus water in the magnetic flocs falls out owing to gravity and its magnetic concentration
increases, resulting in a highly concentrated sludge. This sludge is continuously stripped from the
surface of the cylinder by a claw and dropped under its own weight into a sludge tank. The surface of
the cylinder is continuously scraped for reuse by the claw. This series of operations continuously
purifies the influent, and the by-product is highly concentrated sludge.
Since magnetic flocs may be magnetically drawn to the surface of the cylinder at high speed by the
strong magnetism of the HTS bulk magnet, a large volume of sludge can be separated per unit time. If
the number of revolutions is increased, the new water-treatment system can therefore clean large
volumes of pre-treated water even with a small rotating filter, so even a small treatment system can
process vast amounts of contaminated water.
Configuration of the HTS bulk magnet system
Figure 3 is a photograph of the 33-mm-square, 20-mm-thick YBa2Cu3O7 high-temperature
superconducting bulk superconductor impregnated with epoxy resin (Tomita, M. et al, 2000) used in
the system. Eleven such bulk magnets were used to build a 387-mm-long trial HTS bulk-magnet
system. Figure 4 shows the configuration of the magnetization system. A small, single-stage Gifford-
McMahon helium cryocooler cools the inside of the adiabatic vacuum chamber containing the HTS
bulk superconductorsto a temperature of approximately 35 K. To allow connection to the magnetizing
unit, the HTS bulk superconductors are embedded in the tip of a copper, thermal-conductive bar. The
other end of the thermal-conductive bar is joined to the cold station of the cooler by a flange via an
indium sheet. The low-temperature unit is wrapped with several layers of laminated heat-insulating
material to prevent radiation heat for penetrating it.
Split-solenoid superconducting magnets are used to magnetize the HTS bulks. The magnetic field in
the tunnel between the split magnets is 70 mm in diameter and approximately 100 mm long with a
maximum field of 5.0 T. As the configuration for the magnetizing system in Figure 3 shows, the HTS

Saho: Superconducting magnetic separator for ballast-water treatment
bulk section of the magnet system is inserted into the tunnel between the split superconducting
magnets before excitation. After cooling the bulk superconductors to a temperature of approximately
100 K, i.e., just above its critical temperature Tc, the split superconducting magnets are excited and
emit a specified magnetic field. Since the bulk superconductor do not reach a state of
superconductivity, the magnetic field penetrates the bulk superconductor unassisted. When the bulk
superconductor is cooled further, the temperature falls below its critical temperature Tc, and the
internal magnetic flux gradually begins to be trapped. The split superconducting magnets are
demagnetised at a few degrees above the lowest temperature, and the temperature of the bulk
superconductor then drops to the lowest temperature. Finally, the HTS bulk-magnet system is
extracted from the solenoid magnets (with the cooler still running) and mounted in the sludge-cylinder
of the prototype filtration-magnetic separator.
The magnetized superconductors retain their strong magnetism for good, as long as they are kept
appropriately cooled by the cryocooler. The experimental superconductor-bulk-magnet system is
shown in Figure 5, and a photo of the 1,380-mm-high magnetic separator is shown in Figure 6.
Magnetization of HTS bulk magnets
Figures 7(b) and (c) show the magnetization characteristic of the superconductor bulk magnets under
a 5.0-T magnetic field. As the coordinates show, the upper wall of the vacuum adiabatic chamber has
been assumed to be the x-y plane. Figure 7(a) shows there are 11 bulk superconductors arranged in a
row, and the centre of the sixth from either end of the row meets the zero point of the x-y coordinates.
The symbol Bz stands for the distribution of magnet field intensity along the z-axis (vertical). The
intensity is 3.2 T at the centre on the surface of the chamber, which is the maximum distribution, and
a nearly uniform intensity distribution exists in the range from ­50 mm to +50 mm along the x-axis;
that is, the magnet at the centre and the two magnets on either side have nearly the same magnetic
field intensity. Outside this range, the intensity decreases in proportion to the magnetizer's
distribution of magnetism. [no clear meaning] And a magnetic field of 1.6 T to 3.2 T was produced
within a 200-mm range (­100 to +100 mm). The cooling temperature for the bulk magnets is about 35
K, and the power consumption of the cryocooler is 2.8 kW. (See Table 1). Graph (c) in Figure 7
shows the maximum intensity along the z-axis, which is inversely proportional to the distance from
the chamber-wall surface, in other words, the further away, the lower the intensity. At 6 mm from the
surface at x=0 mm and y=0 mm, the intensity is 1.0 T; this means the bulk magnet generates a
stronger magnet field than that of an ordinary rare-earth metal permanent magnet.
Table 1. Performance of bulk-superconductor cooling system
bulks temperature
34 K
cool-down time
4 hours
electric power consumption
2.8 kW
Treatment of kaolin-contaminated water
The contaminated sample for the treatment experiment was obtained by adding kaolin particles to tap
water. The grain size of the particles was 0.5 µm at a concentration in the water of 92 mg/L. In the
experiment, 100 m3 of sample water were processed per day. The water was mixed with magnetic
powder and a flocculant and stirred to form flocs. The flocs were trapped on the rotary filter and
transferred to the magnet separator, where they accumulated as sludge on the rotary shell. The sludge
was scraped off the shell's surface and dumped into the sludge tank after the water had dripped and
drained off.
Table 2 lists the results of the experiment. The system removed 93% of the suspended particles
("suspended solids"), which means that it can purify plankton-contaminated water (because the

2nd International Ballast Water Treatment R&D Symposium: Proceedings
diameter of plankton is similar to that of kaolin particles or even larger). The concentration of sludge
immediately after magnet separation was 90,000 mg/L, and it was further concentrated as a result of
water dripping and draining into the atmosphere. However, this concentration would be lower in
plankton-contaminated water because the contaminant concentration in the original seawater is a little
lower than that of kaolin in this experiment (i.e., several ten-thousand milligrams per litre). The new
water-treatment system is a continuous one, but the complete purification process takes about five
minutes from when contaminated sample water is input into the flocculation vessel to when it is
dumped in the sludge tank.
Table 2. Treatment test results for kaolin-polluted water at flow rate of 100m3/day.
efficiency (%)
Suspended solids (mg/L)
Concentration of solids in
recovered sludge (mg/L)
Figure 8 is a photo of the water before and after the treatment process. The processed water is more
transparent, which proves the new treatment technology is effective. Figure 9 is a photo of the rotary
filter and magnetic separator in operation. It clearly shows how the suspended solids are captured by
the cage, magnetically transferred to the rotary shell, and scraped off and dumped in the sludge tank.
The scraping position is about 100 mm from the magnetic separator, and magnetic field intensity there
is about 0.05 T, which means magnetic force barely influences the sludge, so scraping is easy.
In the past, we carried out an experiment on a batch-type system using superconductor coil magnets to
remove plankton from red tide (Saho, N. et al., 1999), and the results are listed in Table 3. The
phytoplanktons treated were Chattonella antiqua, and Heterocapsa circularisquama. We made
magnetic flocs of the planktons in the same way as described above for kaolin. The flocs were
magnetically captured on a metal net. The removal rate was 92% or better, which is satisfactory for
water treatment. Although the filter-magnet separation differs from the batch-type system in regards
to the separation of the magnetic flocs, the authors believe that it will have comparable performance.
Table 3. Treatment test results for three kinds of red phytoplanktons (Saho, N. et al, 1999).
Marine Bacteria
Chattonella antiqua
% removal
Heterocapsa circularisquama
% removal
Treatment of oil-contaminated water
A contaminated sample for a treatment experiment was made by adding oil to tap water and agitating
it to emulsify it. As the photomicrograph in Figure 10(a) shows, the diameter of the emulsified
particle suspended in water ranges from 1 to 10 µm. Magnetic powder, a flocculant, and a high-
molecular-weight polymer are mixed in with the sample. When stirred, the oil particles and powder
coagulate and flocs are formed [Figure 10(b)]. Figure 11 plots the results of this experiment. A TOC
evaluation shows that the removal rate is 90% or more. The concentration of TOC in the recovered
sludge was 23,000 mg/L, which is 960 times thicker than the original contaminant concentration of

Saho: Superconducting magnetic separator for ballast-water treatment
24 mg/L. This result suggests that the developed water-treatment system demonstrated in this
experiment can be applied to the treatment of oil-contaminated water from offshore oil rigs, and the
resulting treated water can be discharged directly into the ocean.
A continuous water-treatment system consisting of superconductor bulk magnets, which generate a
high-intensity magnetic field, was developed and experimentally evaluated in tests on purifying
several contaminated-water samples. The experiment showed that more than 90% of the particles in
the contaminated water can be removed in about five minutes. This result indicates that this system is
capable of purifying water continuously and at high speed within a limited space. Moreover, the
recovered sludge is highly concentrated, being tractable for easy disposal. It is concluded that the new
water-treatment system is potentially very effective for the treatment of ballast and oil-contaminated
Saho, N. 2000. German-Japan Workshop on High-Temperature Superconductivity, SRL ISTEC
Tokyo Japan.
Tomita, M. & Murakami, M. 2000. Improvement of the mechanical properties of Bulk
superconductors with resin impregnation
, Supercond. Sci. Technol, No.13 , p. 722.
Saho, N., Isogami, H., Takagi, T., Morita, M., Yamaoka, Y. & Takayama, H. 1999. Development of
Continuous Superconducting-Magnet Filtration System.
The Fourth International Conference on the
Mediterranean Coastal Environment. The Fourth International Conference on the Environmental
Management of Enclosed Coastal Seas, vol. 3, pp. 1398-1410.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 1. Structure of water-treatment system using high-temperature bulk superconducting magnet for filtration-
magnetic separation to clean ballast water.
Figure 2. One filter unit with a frame and a wire net. A stainless steel wire net with a micro-pore diameter of
43 µm is used as the filter, and the width of the aperture inside the stainless steel frame is 200 mm.
Figure 3. YBa2Cu3O7 bulk superconductor impregnated with epoxy resin (20 mm thick). Eleven such bulk
magnets form a 387-mm-long trial HTS bulk magnet.

Saho: Superconducting magnetic separator for ballast-water treatment
Figure 4. Magnetization system for bulk superconductors. The magnetic field in the tunnel between the split
magnets is 70 mm in diameter and approximately 100 mm long.
Figure 5. Superconducting magnet system with bulk superconductors. Under steady-state conditions, the
temperature of the bulk magnets is 34 K and all superconductors are uniformly cooled by the GM cryocooler.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 6. Photograph of the magnetic separator at a treatment flow rate of 100 m3/day.
Figure 7. Measured magnetic field distribution of bulk superconductors on the surface of the vacuum chamber
obtained by zero-field cooling.

Saho: Superconducting magnetic separator for ballast-water treatment
Figure 8. Views of the test influent containing kaolin particles and of the processed effluent
Figure 9. Recovering sludge by magnetic separation.
Figure 10. (a) Photomicrograph of emulsified oil particles in influent and
(b) magnetic flocs in pre-application treatment water.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 11. Efficiency of removing oil from water.

Session 3:
Treatment Systems

Sodium hypochlorite as a ballast water biocide
D.T. Stocks1, M. O'Reilly2 & W. McCracken3
1BMT Fleet Technology Ltd, Canada
2ESG Stantec Consulting Inc, Canada
3Consultant, USA
Political and regulatory context
Legislation was introduced in the Michigan Senate for the regulation of ballast water in January 2000.
This proposed legislation required, among other things, that all ballast water discharged in Michigan
be "sterilized", and that such discharges be authorized by permits issued by the Department of
Environmental Quality. Shortly thereafter, in March 2000, the Department of Environmental Quality
established the Ballast Water Work Group, a group of people with technical knowledge about the
shipping industry. This Work Group was asked to help identify practical methods, currently available,
to minimize the problem of ballast-borne invasive aquatic species in the Great Lakes. "Practical,
currently available methods" were defined as not needing extensive research to establish efficacy, not
needing extensive ship retrofitting, and not needing shore-side facilities. In April 2000, the Ballast
Water Work Group concluded that the only methods which met these criteria were improved
management practices and chemical biocides.
During the summer of 2000, the Ballast Water Work Group set forth lists of improved management
practices for oceangoing vessels ("salties") and Great Lakes vessels ("lakers"). In response to a
request from the Work Group, the Department of Environmental Quality developed plans for
laboratory and shipboard testing of chemical biocides during the fall of 2000. In February 2001, those
plans were finalized, focusing on two biocides, sodium hypochlorite and copper ion. These were the
plans which were presented at the March 2001 IMO Symposium. In June 2001, Michigan selected
BMT Fleet Technology, Ltd., of Kanata, Ontario, as the contractor for the project, and work began.
Meanwhile, much discussion and deliberation took place on the ballast water legislation first
introduced in January 2000. In August 2001, Michigan Act 114 was passed and signed into law. It
states that a goal of the State of Michigan is "to prevent the introduction of and minimize the spread
of aquatic nuisance species within the Great Lakes." However, Act 114 is much different than the
original bill. It does not require sterilization or permits for ballast water. Instead, it requires that the
Department of Environmental Quality make determinations by certain deadlines as to: a) which ships
are complying with the management practices set forth through the Ballast Water Work Group
process, b) what treatment methods can be used to minimize invasive aquatic species, and c) which
ships are complying with these treatment methods. Listings of these determinations are to be made
available to the public. Vessel owners and operators and their customers are not eligible for grants,
loans, or awards from the Department of Environmental Quality unless they are on the listings.
A key provision of Act 114 requires that the Department of Environmental Quality makes a
determination by March 1, 2002, of "Whether one or more ballast water treatment methods, which
protect the safety of the vessel, its crew, and its passengers, could be used by oceangoing vessels to
prevent the introduction of aquatic nuisance species into the Great Lakes." Also, the Act requires the
Department to determine a time period after which the treatment method could be used by all
oceangoing ships on the Great Lakes. If a treatment method is not now available, the Act requires the
Department to determine the actions needed and the time period for finding a viable treatment
method. Although the legislative deadline has passed, the requirement to make the determinations is
still in effect.
During the fall of 2001, the shipboard testing was completed, and the laboratory testing was
completed during the winter of 2002. The draft project report was completed in early March 2002.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Then the draft project report was discussed with the Ballast Water Work Group in late March. The
Work Group expressed concern about several aspects of the draft project report. As a result, the
Department of Environmental Quality requested the Governor to convene a Panel of the Michigan
Environmental Science Board to carry out a critical review of the findings of the project. The
Governor agreed, and the Panel was assembled and met on May 29, 2002. The Panel included four
permanent members of the Michigan Environmental Science Board, and five invited experts in the
areas of ballast water and corrosion. One of the invited Panel members was Stephen Raaymakers of
the IMO staff.
The Michigan Environmental Science Board published the report of its review of the draft project
report in September 2002. The "Major Findings and Conclusions" of the Board are contained in
Attachment 5. Key statements included the following:
For Copper Ion: "In summary, the Study's toxicity data, as a whole, suggest that in sufficiently
high concentrations, copper ion could be an effective biocide. However, at the concentrations
needed to achieve the desired effectiveness, the level would be far too high to be discharged into
the Great Lakes. Given this, and in the absence of any known neutralizing agent that would allow
copper to be safely discharged into the Great Lakes, the MESB Panel concludes that copper ion
cannot be considered to be a viable ballast water biocide alternative at this time."
For Sodium Hypochlorite: "The Study's conclusions regarding the effectiveness of sodium
hypochlorite as a viable ballast water biocide alternative from the shipboard and laboratory
toxicity tests are limited and can only be considered preliminary at best. However, despite the
problems outlined in this critique regarding the testing protocols used, the MESB Panel suggests
that the use of sodium hypochlorite as a ballast water biocide can have a high degree of efficacy
when treating the majority of organisms that were tested in the Study, assuming that sufficient
active hypochlorite concentration can be attained to account for sediment loads from both
suspended and deposited material." The MESB report raised several other questions about the
Study which need to be answered relative to use of sodium hypochlorite, including corrosivity in
ballast tanks.
The project report was finalized by BMT Fleet Technology, Ltd. in November 2002. Based on that
report and on the conclusions of the Michigan Environmental Science Board, the Department of
Environmental Quality could not make a determination under Act 114 as to whether one or more
ballast water treatment methods could be safely used by oceangoing vessels. The Department decided
to fund a second phase of work in order to gather additional information which may allow that
determination to be made for sodium hypochlorite. In November 2002, BMT Fleet Technology, Ltd.
was selected to carry out a contract for the Phase 2 Study.
Michigan Act 114 is being implemented. The Department of Environmental Quality has developed
listings of which ships, both salties and lakers, are complying with the applicable ballast water
management practices. These listings are publicly available on the state of Michigan web page, and
are updated as new information becomes available. A determination has been made that these
management practices are now conditions of passage on the St. Lawrence Seaway. Although the
determination as to whether a treatment method is available has been delayed until further information
is developed under the Phase 2 Study, the Department of Environmental Quality remains under the
mandate of Act 114 to make that determination.
Invasive aquatic species continue to cause great concern in the Great Lakes region. The need for a
solution to the problem, or at least a reduction of the risk of these foreign invaders, is urgent, and
keenly felt in the state of Michigan. Questions remain about the use of sodium hypochlorite to treat
ballast water, including potential corrosion of ballast tanks, and the problems of sediment biocide
demand. However, those working on the Phase 2 Study remain hopeful that it can be shown to be a
viable ballast water biocide. There are several potential advantages of sodium hypochlorite: it is a
widely used, proven biocide; it can be neutralized prior to discharge; and it is readily available.
Although concerns about dangerous byproducts have been raised, the amount of sodium hypochlorite

Stocks: Sodium hypochlorite as a ballast water biocide
needed to treat ballast water would be very small in comparison to other uses of chlorine. Thus, the
increase in relative risk to the environment should be small. This is being further verified as part of
the Phase 2 Study. Criticism that it is not effective on 100% of foreign species should not rule out its
use, at least on an interim basis, because adequate dosages will destroy most of the foreign organisms
in ballast water and greatly reduce the risk of new invasions.
Phase 2 studies
After a review of the MESB findings the MDEQ concluded that follow up efforts were required to
address the issues raised. The issues included;
· The practical implementation, on board ship, of a sodium hypochlorite chlorination and de-
chlorination system to ensure effective and efficient use (minimization) of chemical additives.
· The effect of typical Great Lakes ballast water temperature ranges on the efficacy of sodium
hypochlorite as a biocide.
· The effect of ballast tank sediments on the efficacy of sodium hypochlorite as a biocide.
· The quantification and impact of the formation of chlorinated compounds by typical sodium
hypochlorite treatment of ballast tanks in Great Lakes waters.
The first round of work included an examination of the effect that sodium hypochlorite and copper
might have on the structural integrity of the ship's ballast tank. This work was included in the
evaluation of the biocides in response to ship owners and classification societies concern over
accelerated corrosion of the steel. An investigation of these effects along with an assessment of tank
coating damage effects was undertaken, however, due to the large number of variables necessarily
examined and available time frame for the project only short duration tests were conducted. The
MESB review recommendations suggested longer term test be conducted on hypochlorite be
conducted to quantify the life cycle effects of exposure to biocides in the ballast tank.
The objective of this study is to provide additional information to support the MDEQ's determination
of whether sodium hypochlorite can be recommended for general application as a ballast water
Issue 1: Appropriate dose control
The practical implementation, on board ship, of a sodium hypochlorite chlorination and de-
chlorination system control mechanism to ensure effect and efficient use (minimization) of chemical
The problem can therefore be stated as one to develop an application of existing technology for TRC
monitoring that can be utilized as a monitoring and control system suitable for the ship ballast tank
To address this, an engineering design is being undertaken by BMT/FTL to determine the
appropriate type, placement and operational constraints of a sodium hypochlorite/sodium bisulfite
treatment system.
Issue 2: Temperature
The effect of typical Great Lakes ballast water temperature ranges on the efficacy of sodium
hypochlorite as a biocide.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Great Lakes shipping typically starts during the spring months of March or April and lasts until
December, depending on conditions. During a typical shipping season temperatures in the Great
Lakes varies from lows of 1 to 3°C in cooler months to 15 to 26°C during summer and fall depending
on the lake and water depth. As is the case with many oxidizing chemicals, the temperature of
application can impact their biocidal properties. Given the range of temperatures that can be found in
the Great Lakes during a typical shipping season, more or less hypochlorite may be required to
achieve similar endpoints at different times of the year.
To address this, a literature review was conducted prior to initiating the study. The literature
review discussed studies of a similar nature to determine whether temperature is expected to
impact the biocidal efficacy of hypochlorite.
Issue 3: Sediments
The effects of sediment on the efficacy of sodium hypochlorite as a biocide.
The amount of sediment suspended in new ballast water, and in the re-suspension of residual ballast
water, will impact the amount of biocide needed to account for the chlorine demand. The impact of
the presence sediments on the efficacy of sodium hypochlorite needs to be evaluated.
To address this, toxicity tests involving Hyalella azteca are being conducted at Stantec's
(formerly ESG International) Ecotoxicity Laboratory in Guelph, ON. Tests are 48-hour static
acute toxicity tests conducted at 15°C. Test endpoints include lethal concentration to 90%
mortality (LC90) and lethal time to 90% mortality (LT90). Tests are being conducted in triplicate
using a linear dilution of hypochlorite against a logarithmic concentration of sediment. The
amount of chlorine required to achieve treatment levels for each sediment level is determined.
Issue 4: Disinfection by-products
The quantification and impact of the formation of chlorinated compounds by typical sodium
hypochlorite treatment of ballast tanks in Great lakes waters.
The mixing of hypochlorite in a ballast tank containing water and sediment is expected to produce
some level of disinfection by-products (DBPs). The interaction of hypochlorite with organic matter
commonly found in ballast water and sediments can produce trihalomethanes (THMs), haloacetic
acids (HAAs), and liberate metals from the sediments into the overlying water. Additionally,
dechlorination of treated ballast water using neutralizing agents can reduce the dissolved oxygen
content and pH of the discharged ballast water. This portion of the study is designed to determine the
amount and types of DBPs that may be produced from the interaction of hypochlorite and a
dechlorinating agent with the sediments and water associated with ballast tanks.
To address this, tests to determine the extent and amount of DBPs produced during the
chlorination of ballast water and sediments are being undertaken at Stantec's (formerly ESG
International) Ecotoxicity Laboratory in Guelph, ON. Samples of sediments will were collected
from the ballast tanks of three ocean going ships. Natural water and increasing levels of sediment
were mixed, chlorinated, and dechlorinated (after 48 hours exposure). Samples of overlying
solutions were then collected and analyzed for THM's, HAA,s, metals etc.
Issue 5: Structural deterioration
The quantification and impact of exposure of a ships ballast tank structure, including coating systems,
to periodic doses of Sodium Hypochlorite.

Stocks: Sodium hypochlorite as a ballast water biocide
Aqueous corrosion of steels in natural waters depends entirely on the availability of oxygen. When the
source of oxygen is air in an open natural system with pH between 4 and 10, the rate of attack has
been observed to average approximately 0.1mm/year (0.004 inches/year or 4mpy) at ambient
temperatures and this rate is controlled by the diffusion rate of oxygen from the bulk solution to the
steel surface. In short term exposures, the rate tends to be higher on clean bare surfaces but the rate
tends to decrease with longer exposures as surface scales build up. In the temperature range
encountered in nature, corrosion rates increase with temperature, doubling every 30°C because
diffusion rates increase with temperature. Other factors which accelerate bulk diffusion such as
agitation in the liquid which reduces the thickness of the boundary layer and wetting and drying
cycles which afford atmospheric oxygen better access through the meniscus in the drying stage also
accelerate corrosion. These factors account for the enhanced attack observed at the waterline and
splash zone in marine environments.
Other oxidizing agents added to oxygenated water may have positive or negative effects on corrosion
rates of steels. Some anions, such as chromates or permangenates, are effective inhibitors and result in
corrosion rates approaching zero. Hypochlorite ion has no inhibiting effect and, on the contrary, acts
as additional oxidizing agents to accelerate the corrosion of steel. The hypochlorite ion has been
compared to wet chlorine in its effects on materials. Not many metals show good resistance even at
low temperatures and concentrations.
The fundamental requirements of the barrier system, i.e. tank paint coatings are that the coating
should be (a) impermeable to damaging ionic species and, if possible, to oxygen and (b) that it should
maintain adhesion to the steel under wet corrosion conditions. Sufficient impermeability to water is
not possible except in very thick films (>20 dry mils) and ingress of water leads to de-adhesion.
Impermeability to ionic solutions and oxygen are entirely more practical objectives and, consequently,
these factors are rate determining for corrosion beneath intact barrier films.
To address this, tests to determine the extent and amount of corrosion in the typical ballast tank
exposures including fully submerged and splash zones are being conducted at BMT FTL
laboratories in Kanata Ontario. Tests are also being conducted on the permeability of paint
systems and the effects of surface damage to paint systems. All tests are being conducted in fresh
and salt water with varying exposure levels of hypochlorite.
Issue 6: Life cycle impacts
The practicality and impact of adopting Sodium Hypochlorite as a ballast water biocide for use in the
Great Lakes on the shipping industry and the discharges into the basin.
The economic impact on the ship is significant, systems will have to be engineered to ensure correct
dosing levels are applied; biocide will have to be carried or generated on board and infrastructure
developed to ensure proper training etc. The impacts of discharges need also to be evaluated to assess
the relative risk of damage to the environment.
A life cycle model will be developed based on the results of the toxicity testing which will
establish appropriate dosing levels, the traffic level of ships into the Great Lakes which will
establish the total exposure and the economics of fitting engineered solutions to the ship.
Selected Phase 1 study results
Project composition
The project is comprised of three parts:

2nd International Ballast Water Treatment R&D Symposium: Proceedings
· A field demonstration on-board the MV Federal Yukon,
· Toxicology testing in the biological laboratory, and
· Corrosion testing in the material laboratory.
Field trials on board the M.V. Federal Yukon
This part of the project is characterized as a field trial, rather than a research project. The purpose of
this part was to examine the shipboard application of biocides and assess the efficacy of treatment on
a single typical voyage and to further determine whether the application of biocides adversely affects
the real-life operations of the ship.
The project work was not allowed to interfere with the commercial operations of the ship and certain
on-site modifications to the experimental plan were necessary to accommodate the local biological
conditions and engineering difficulties encountered.
The tests were conducted on the deck of the ship using 55 gallon plastic barrels as test chambers.
Additionally, the deck mounted decant tank (a metal deck tank typically reserved for capturing cargo
wash water prior to discharge) was modified and coated with paint used in the ballast tanks for
additional hypochlorite tests.
A typical voyage profile for a ship on international trade into the Great Lakes consists of loading
cargo overseas and transiting the ocean as a NOBOB. On arrival at a Great Lakes port, the ship will
discharge its cargo and take on ballast to transit to a second Great Lakes port. Here the ship will
discharge that ballast and take on an out-bound cargo. The field trial was conducted during one such
typical international voyage at four ports:
Port #1 ("Coastal Port #1"): (Bilbao Spain) An ocean port in a saltwater environment. Cargo was
off-loaded, and ballast water taken on.
Port #2 ("Coastal Port #2"): (Antwerp Belgium) An ocean port where cargo was taken on, and
ballast water discharged, creating a NOBOB condition.
Port #3 ("Great Lakes Port #1"): (Burns Harbor Indiana) A Great Lakes, fresh water port. Cargo was
off-loaded, and
Port #4 ("Great Lakes Port 2"): (Superior Wisconsin) A Great Lakes port where cargo was taken on
and ballast water discharged.
Laboratory toxicity testing
The ship is an operational platform and its voyage plans may take it anywhere in the world. Given the
variability of ballast water characteristics that this entails, shipboard trials are not as well controlled as
laboratory experiments. For example, given where and when ballast water is taken on, it may not
contain high numbers of specific organisms of concern, and it may not contain high levels of
sediment. Therefore, a series of laboratory toxicity tests were conducted at ESG International's
Ecotoxicity Laboratory (Guelph, Ontario) to complement the shipboard testing of biocide efficacy.
The purpose of this part of the project was to quantify the efficacy of the biocides as it relates to the
treatment of organisms of concern in ballast water. The toxicity testing was conducted on freshwater
and saltwater fish, invertebrates, algae and bacteria. In addition, the toxicity of the biocides to selected
resting stages was evaluated. The organisms were selected to represent the range of pelagic and
benthic organisms and the various lifestages that may be found in ballast water. In general, and where
possible, organisms/lifestages that tend to be more resistant to chemical treatment were selected over
more sensitive organisms. In certain instances, the toxicity of the biocides was tested in both
laboratory water and ballast water collected from the ship. A limited number of tests were conducted

Stocks: Sodium hypochlorite as a ballast water biocide
with and without the presence of a clean, control sediment for characterizing the effect of sediment on
biocide efficacy. Appendix D contains the laboratory for laboratory protocol addenda.
Laboratory corrosion testing
Likewise, the relatively short shipboard trial could not reveal the true corrosion or tank coating
damage potential of the biocides. Thus, complementary laboratory studies of the potential for biocide-
induced damage were undertaken at the Fleet Technology Limited Material Laboratories (Kanata,
The purpose of this part of the project was to examine the possible detrimental effects that the
addition of biocide to ballast water may have on the structural integrity of the vessel. The effects of
biocide treated water on coating systems and base metals typically used in the construction of ships
ballast tanks were investigated in a specially adapted accelerated corrosion tank. The conditions
within a ballast tank (i.e., fully submerged, a splash zone or area of periodic immersion, and the damp
spaces) were simulated along with a "buried" experiment to show the effects on structure covered
with sediment. The experiment used the accelerated corrosion testing concept to compare the effects
of adding biocide to both fresh and saltwater. Corrosion tests were conducted on bare metal coupons,
metal coupons coated with typical marine coating systems, and coated metal coupons that were
scribed through the paint thickness to examine the effects of coating damage.
Efficacy ­ chlorine
Table 1 lists the range of lethal chlorine concentrations for the freshwater and marine species tested in
the laboratory. For both freshwater and marine species, the range between the least and most tolerant
organism was significant (i.e. several orders of magnitude).
Within the freshwater species, algae (including S. capricornutum, S. obliquus, and Nanochloris sp.)
exhibited the lowest tolerance to chlorine (IC99s 0.1 mg/L), followed by D. magna (LC99 = 0.2
based on exposure of neonates (< 24-h old)). The most tolerant species, based on exposure of the
resting egg or "ephippia", was D. magna (IC99 = 76.3 ppm). Lethal effect levels for all other species
were in between this range with LC99s below 10 ppm chlorine.
For the marine species, the bacterium, (V. fischeri) and the alga (S. costatum) exhibited the lowest
tolerance to chlorine with LC99's estimated to be 0.15 and 0.20 mg/L, respectively. The most tolerant
species was the brine shrimp (A. salina), based on exposure of the cyst (LC99 486 ppm). All other
species were in between this range with estimated LC99s below 10 ppm chlorine.
Table 1. Lethal concentrations (estimated LC99, IC99) of chlorine for selected biota.
Lethal Concentration
Range (ppm as TRC)
< 1
Alga S. capricornutum
Bacteria V. fischeri
Alga Nanochloris sp.
Alga S. costatum
Alga S. obliquus
Invertebrate D. magna (neonate)
1 to 10
Bacteria B. subtilis
Amphipod E. estuarius
Mollusc D. polymorpha
Fish C. variegatus
Benthic invertebrate L. variegates
Fish C. carpio
10 to 100
Invertebrate D. magna ­ (ephippia)
100 to 1000
Invertebrate A.salina (cyst)

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Structural integrity.
Figure 1 shows the test tanks set up with test coupons at the start of an experiment. Test coupons are
exposed to varying levels of chlorine (10 ppm and 40 ppm) over a 15 day period in both salt water
and fresh water. A control tank is also provided with no chlorine. Coupons are mounted on a rotating
wheel to simulate the splash zone, suspended in the solution to provide continuous exposure and
suspended in the air space inside the enclosure to simulate the damp conditions of the partially filled
tank. In addition a set of coupons were buried in inert sand in the bottom of the solution tank to
simulate under sediment low oxygen conditions.
It is common in ship building to coat the steel with zinc rich pre-weld primers under an epoxy paint
system. The primer protects the steel during construction but in the presence of damage paint may
provide for an anodic reaction with different metal. A series of tests were also conducted with 4
different paint products in typical use in ship building. These test were done in accordance with the
standard ASTM "scratch" procedure whereby a prescribed damage is introduced in to the paint
surface and the extent of damage increase monitored.
Figure 2 shows the experimental results in terms of annual diminution rates from the accelerated
corrosion tests on bare steel coupons in salt and fresh water. On the basis of the operational scenario
previously developed from the field trial on the Federal Yukon. It is assumed that a ship will be
subject to ballast water treatment on a 30 day cycle and that during that cycle the structure will be
exposed to 2 days of high sodium hypochlorite dose and 4 days at a low dose given in the experiment
(20% of total time) decomposing chlorine reducing the levels to zero over time. Under this
assumption, the effective corrosion rate of bare exposed steel (after removal or damage to the coating
system) in the ballast tank would be increased by 1% and its serviceable life reduced accordingly.
Figure 3 shows a comparative measure of coating loss, in accordance with the ASTM rating score for
three coating systems, and reveals the following:
· Samples exposed to hypochlorite tend to experience slightly more damage than the control
samples; however, this is a small effect and is not quantifiable in terms of life expectancy
from this analysis. The saltwater low hypochlorite exposure showed no difference in damage
to that experienced by the control exposure.
· There is an observable trend in the level of damage experienced relative to the location in the
test tank, i.e., the more aggressive location from a corrosion perspective also provides for
more damage from a coating perspective.
Table 2. Economics of ship installation.
Copper Ion
On Board
Purchase commercial
concentration Sodium
0.2 ppm on
330 kg (725lbs)
Buy and store
Deliver to the
50 tonnes
per day 0.8%
ship as required
Capital cost
$ 437,710
$ 207,025
$ 77,318
Element replacement cost
$ 50,000
$ 10,000
Element replacement (years)
Ballast operations per year
Raw material costs
$ 0.09
$ 504
$ 756
Vessel charter rate (per day)
Return rate
Inflation rate
Amortization period
Increase charter to maintain
$ 48.08
$ 207.73
$ 60.54
%increase to cost of shipping

Stocks: Sodium hypochlorite as a ballast water biocide
Economics of ship installation
Sodium hypochlorite can be purchased as a liquid in concentration of 15% sodium hypochlorite, or it
can be can be generated onboard a ship using a sodium hypochlorite generator. Either option requires
appropriate storage, handling and dosing and metering systems. In addition, any chlorine-based
system will also need a de-chlorination capacity to render discharge environmentally acceptable. This
system will also require control and monitoring of pumps and storage facilities.
A life cycle economic analysis of various systems was conducted and the increase in ship charter rate
(cost to user) necessary to support the ballast water treatment computed.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 1. The test tanks set up with test coupons at the start of an experiment.
Humid In the
damp space
Burried in
constantly wet
Wheel periodic
Figure 2. The experimental results in terms of annual diminution rates from the accelerated corrosion tests on
bare steel coupons in salt and fresh water.

Stocks: Sodium hypochlorite as a ballast water biocide
Figure 3. Comparative measure of coating loss.

Effects of the chlorination treatment for ballast water
S. Zhang, X. Chen, D. Yang, W. Gong, Q. Wang, J. Xiao, H. Zhang & Q. Wang
Dalian Maritime University,
Treatment options being researched
Aims and objectives of the project
This project deals with the effects of the chlorination treatment for ballast water. Chlorination
treatment is selected mainly based on three facts.
Firstly, chemical method is widely adopted to kill organisms and bacteria in large-scale water
treatment. Secondly, among chemical methods chlorination treatment is earliest and most common.
For instances, many countries including China, use chlorination to disinfect and kill bacteria for
potable water, water from hospital and water for aquaculture. Since late 80s, State Entry-exit
Inspection and Quarantine of China has been using chlorine to treat ballast water containing vibrio
bacteria from epidemic area. Thirdly, due to easy operations and low expenses, chlorination is feasible
to be used on board without special apparatus to treat ballast water.
Our experiments selected Sodium Hypochlorite as biocide. The results prove that chlorination
treatment is effective in killing organisms and bacteria in seawater. They also show that available
chlorine with concentration of 20 mg/L is able to kill almost all the bacteria in the seawater. However,
the concentrations of available chlorine for phytoplankton, zooplankton and benthic invertebrate's
treatment vary depending on the species and the density, ranging from 5 mg/L to 100 mg/L. The
exposure duration is not considered in this experiment.
Ame's and luminescent bacteria's tests of treated byproducts in laboratory and onboard field-test have
not been done for various reasons, and we will do them before long.
Test design
Bacteria test
The bacteria test is conducted to determine the efficacy of Sodium Hypochlorite on total anaerobic
bacteria (Membrane filter method, reported as CFU/10cm3), Vibrio (Membrane filter method, reported
as CFU/10cm3) and E.Coli. (MPN fermentation method, reported as CFU/10dm3) in seawater.
Phytoplankton test
We selected four kinds of phytoplankton algae, namely Nitzschia closterum (diatoms), Dicrateria spp.
(chrysophyta), Platymonas spp.(green alga) and Pyramidomonnas sp.(green alga) with the density of
109/L. This is the typical density when "red tide" occurs. The objective was to find out absolute lethal
concentrations (LC99) of sodium hypochlorite for every selected phytoplankton algae.
The culture of adaptability: phytoplankton algae in laboratory were cultured with f/2 general culture
media for phytoplankton algae (in 22°C, 2200 Lux and 12:12 photoperiod) for 3 days prior to being

Zhang: Effects of the chlorination treatment for ballast water
Determination of available chlorine: Available chlorine of sodium hypochlorite (NaOCl) bought from
market was determined using Chinese National Standard methods (GB10666).
Chlorination: Water samples were treated by sodium hypochlorite with the concentrations of available
chlorine (nominal) of 5, 10, 20, 40, 80 and 100 mg/l, respectively.
Regrowth and examination of phytoplankton: After chlorination, water samples were placed in the
light/dark (which simulates the condition of ballast tank) for 7 and 15 days. Then they were regrown
(at 22°C, 2200 Lux and 12:12 photoperiod) for 20 days and examined.
Natural seawater test
The natural seawater used in the experiments was obtained from the sea nearby Lingshuiqiao and was
filtered prior to use. The objective was to find out absolute lethal concentrations (LC99) of sodium
hypochlorite for all organisms in natural seawater.
The culture of adaptability: natural seawater was cultured with f/2 general culture media (in 22°C,
2200 Lux and 12:12 photoperiod) for 3 and 20 days prior to being treated.
Determination of available chlorine: Available chlorine of sodium hypochlorite (NaOCl) bought from
market was determined using Chinese National Standard methods (GB10666).
Chlorination: Water samples were treated by sodium hypochlorite with the concentrations of available
chlorine of 5, 10, 20, 40, 80mg/l, respectively.
Regrowth and examination: After chlorination, water samples were placed in the light/dark (which
simulates the condition of ballast tank) for 7 and 15 days. Then they were regrown (in 22°C,
2200 Lux and12:12 photoperiod) for 20 days and examined.
Amphipod test
We selected a kind of amphipod, Corophium acherusicum Costa, that belongs to the benthic
invertebrate group. The test was conducted using standard toxicity tests. However, the exposure
duration was altered to 48h. The objective was to find out the concentration-effect relationship for
Corophium acherusicum Costa.
Chlorination treatment for Corophium acherusicum Costa was mainly using ASTM methods (E1367-
92 Standard Guide for Conducting 10-day Static Sediment Toxicity Tests with Marine and Estuarine
Amphipods) and Chinese industrial standard (Standard for conducting marine sediment toxicity tests
with amphipods: pending authorisation).
Table1. Summary of test conditions for determining acute lethality to Corophium acherusicum Costa.
Test type:
Test duration:
48- h
Water bath at 22°C
Ambient laboratory illumination
Feeding regime:
No feeding
Beaker volume:
1000 ml
Test solution volume:
500 ml
Thickness of Sediment
Renewal of test solution:
Age of test organisms:
7-10 days old
Number of animals/test beaker:
Number of replicates
Dissolved oxygen
Measured end points:

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Brine shrimp (Artemia salina) test
The test involved cysts, nauplii and adults of Brine Shrimp (Artemia salina), which is a representative
of zooplankton. The test was conducted using standard toxicity tests. The objective was to find out the
concentration-effect relationship for cysts, nauplii and adults of Brine Shrimp.
Chlorination treatment for Artemia salina was carried out as per the Chinese national standard
GB18420.1-2001 (Biological toxicity inspection method for pollutant from petroleum exploration and
exploitation) and Germany ATS benchmark (The ATS-benchmark for chemical treatment options).
Table 2. Summary of test conditions for nauplii (GB18420.1-2001)
Test type
Test duration
Light intensity
1000 lux
12 h light, 12 h dark
Feeding regime
No feeding
Test beaker volume
100 ml
Water volume
50 ml
Age of organisms
Hatch 24-36h
Number of animals/test beaker
Number of replicates
Dissolved oxygen
Saturation prior to treatment
Dilution water
Manmade seawater(35, pH7.9)
Measured end points
Median lethal concentration (LC50)
Test Validity
Invalid if mean 96h died in control >10%
Table 3. Summary of test conditions for cysts, nauplii and adults (Germany benchmark)
Test type:
Test duration:
Light intensity:
1000 lux
12 h light, 12 h dark
Feeding regime:
No feeding (cysts and nauplii)
Feed with phytoplankton (adults)
Test beaker:
100ml (cysts and nauplii)
2000 ml (adults)
Water volume:
50 (cysts and nauplii)
1000 ml (adults)
Age of organisms:
Hatch for 24-36h (nauplii)
14-16 days old (adults)
Number of animals/test beaker
Number of replicates:
Dissolved oxygen
Saturation prior to treatment
Dilution water:
Manmade seawater(35, pH 7.9)
Measured end points:
Hatch rate mean%
Mortality mean%
Breakdown test of available chlorine
In theory the concentration of available chlorine added equals to residual chlorine in water. However,
in practice, because of illumination, volatilization and substances including living things consuming
chlorine in water, some chlorine will be lost, even in distilled water. We designed the experiment in

Zhang: Effects of the chlorination treatment for ballast water
which the available chlorine, in treated solutions with different biomass and initial concentration,
breaks down with time.
We selected Pyramidomonnas sp. and nauplii of brine shrimp with different biomass and added
sodium hypochlorite to make the concentration of available chlorine to be 5, 10, 20 and 40 mg/l
respectively, and then examined total residual chlorine (TRC) in the water samples at 1h, 4h, 24h and
In this test, we examined residual chlorine with colorimetric method (GB5750-85), with a detection
limit of 0.01 mg/l. The sample of phytoplankton chlorination is filtered prior to detection.
Test results
Results of bacteria test
Table 4. Efficacy of chlorination for the bacteria in natural seawater
After chlorination, available chlorine (mg/l)
Bacterial type
(24 hours)
Density of anaerobic
bacteria (Membrane
filter method)
Density of Vibrio
(Membrane filter
Density of
fermentation method)
Table 4 indicates that after 24 hours of chlorination the density of various bacteria decreases
dramatically in water. In the group of 5 mg/l available chlorine, anaerobic bacteria and E.coli are
0.15% and 0.05% of the initial concentrations before chlorination, 14.8% and 6.2% in contrast with
the control group and no vibrio was detected in 10 cm3 of water samples. In the groups above 20 mg/l
available chlorine, there are no bacteria in 10 cm3 of water samples.
These results indicate that chlorination of 5 mg/l available chlorine can kill 99.85% of anaerobic
bacteria, 100% of vibrio and 85.2% of E.coli. respectively, while above 20mg/l available chlorine can
kill almost all bacteria.
Results of phytoplankton test
Table 5. Absolute Lethal Concentrations (LC99) of sodium hypochlorite for selected phytoplankton algae.
Density of
Pyramidomonnas sp.
Nitzschia clostertum
Dicrateria spp.
algae (/ml)
These results indicate that absolute Lethal Concentrations (LC99) varies depending on the species and
the density of algae, ranging from 5 mg/L to 100 mg/L.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The order of the endurance of algae selected in our test from high to low is as follows:
Platymonas spp .>Pyramidomonnas sp and Dicrateria spp.> Nitzschia closterum
We make a comparison between the two conditions, which are in light and in dark. As seen from the
LC99 for phytoplankton algae, there is no difference between the two conditions. However, with the
duration in dark being extended, phytoplankton algae recover more slowly.
Besides killing phytoplankton algae, sodium hypochlorite also has a strong discoloration effect.
Results of natural seawater test
Before chlorination, there were many kinds of phytoplankton algae in natural seawater, such as:
Thalassiosira sp, Navicula spp., Nitzschia sp, Leptocylindrus danicus, Asterionella japonica Cleve,
Cyclotella sp., Dunaliella sp., Gloeothece linearis, Oscillatotia sp., Synechococcus sp, Glenoddinium
sp. etc. There are also some kinds of protozoan, such as: Euciliata sp, Euplotes, Difflugia sp,
Brachionus calyciflorus and planula larva etc.
After 20 days of culture, there were multicellular algae in natural seawater, such as: Cladophora
and Cladophora rudolphiana etc. Total biota density is much greater than that in primary
natural seawater.
Table 6. Absolute Lethal Concentrations (LC99) of chlorination for the organisms in natural seawater.
concentration of available chlorine (mg/l)
Regrowing time
+ = algae can survive and reproduce;
O = protozoan can survive;
- =
neither algae nor protozoan can survive.
* = there are only some benthic diatoms.
These results indicate that 5mg/l available chlorine can kill protozoan while 20 (3 days regrowth)
60mg/l (20 days regrowth) available chlorine can kill both protozoan and algae in natural seawater.
Results of amphipod test
Table 7. Acute lethality of chlorination for Corophium acherusicum Costa
Concentration of
Number of tested
Number of died
available chlorine
Results of brine shrimp test
The LC50 Sodium hypochlorite for nauplii of Artemia sallina:
Table 8. Toxicity test data of Sodium hypochlorite for nauplii of Artemia sallina (GB18420.1-2001).
Available chlorine (mg/L)
Mortality (%)

Zhang: Effects of the chlorination treatment for ballast water
y = 8.7711x + 4.1335
6 . 5
5 . 5
4 . 5
3 . 5
- 0 . 1
0 . 1
0 . 2
0 . 3
Logarithm of concentration
Figure 2. LC50 Calculation for Sodium hypochlorite for nauplii of Artemia sallina.
Based on probit method, we found out that the LC50 of sodium hypochlorite for the nauplius of
Artemia salina in seawater is 1.26mg/L, and the 95% confidence interval is 1.16~1.36mg/L.
Efficacy of chlorination for Artemia sallina (Germany ATS benchmark)
Chlorination effects for hatching of the cysts of Artemia sallina
Firstly, an orthogonal test is designed with three factors (temperature, illumination, salinity) and
four levels. From the summation or average of each factor, we find out the key factor that
influences the hatching rate is salinity, while temperature is the second and the illumination is the
last. Through level selection, the perfect conditions were: temperature 24°C, salinity 35% and
illumination 1000 lux. These were also the conditions of our hatching experiment.
Secondly, we conducted concentration-effect test for hatching rate under different concentrations
of available chlorine.
1 0 0
8 0
1 . 5 5 m g / l
2 . 1 m g / l
6 0
2 . 8 m g / l
4 0
4 m g / l
5 . 6 m g / l
Hatching rate(%)
2 0
8 m g / l
1 3 . 5 m g / l
2 4 h
4 8 h
7 2 h
9 2 h
1 2 0 h
Figure 3. Hatching rate under different concentrations of available chlorine.
Figure 3 indicates that sodium hypochlorite has an obvious constraint efficacy on the hatching.
With the concentration increasing, the constraint efficacy is more obvious. In addition, there is
certain delay effect on the hatching in all treated groups.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Concentration-mortality relationship for the nauplii of Artemia sallina
1 0 0
8 0
6 0
4 0
2 0
Concentration of available chlorine
Figure 4. Mortalities under different concentrations of available chlorine.
With the concentration increasing, mortalities increase. In the treated group of 2.4 mg/l, there is
no survival. In contrast with the adults, nauplii have a higher endurance at low concentrations.
Concentration-mortality relationship for the adults of Artemia sallina
1 2 0
1 0 0
8 0
6 0
4 0
2 0
Concentration of available chlorine
Figure 5. Mortalities under different concentrations of available chlorine
With the concentration increasing, mortalities increase. In the treated group of 2.4 mg/l, there is
no survival.

Zhang: Effects of the chlorination treatment for ballast water
Results of available chlorine's breakdown test
Breakdown of available chlorine with different density of nauplii (The concentration of available
chlorine added is 5 mg/l).

3 . 5
2 . 5
2 8 / m l
2 0 / m l
1 4 / m l
1 . 5
7 / m l
Residual chlorine (mg/l)
0 . 5
1 h
4 h
2 4 h
Figure 6. The breakdown of available chlorine.
Table 9. The breakdown of available chlorine.
Density of
Residual chlorine (pH value)
Breakdown of available chlorine with different density of nauplii. (The concentration of available
chlorine added is 10 mg/l)

5 8 / m l
4 4 / m l
2 9 / m l
1 5 / m l
Residual chlorine(mg/l) 0
Figure 7. The breakdown of available chlorine.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Table 10. The breakdown of available chlorine.
Density of
Residual chlorine (pH value)
Breakdown of available chlorine with different density of nauplii. (The concentration of available
chlorine added is 20 mg/l)

1 5
4 9 / m l
(mg/l) 1 0
3 7 / m l
2 5 / m l
1 3 / m l
Residual chlorine
1 h
4 h
2 8 h
Figure 8. The breakdown of available chlorine.
Table 11. The breakdown of available chlorine.
Density of
Residual chlorine (pH value)
Breakdown of available chlorine with different density of nauplii. (The concentration of available
chlorine added is 40 mg/l)

3 0
2 5
1 7 0 / m l
2 0
1 2 8 / m l
1 5
8 5 / m l
1 0
4 3 / m l
Residual chlorine (mg/l)
1 h
4 h
2 4 h
Figure 9. The breakdown of available chlorine.

Zhang: Effects of the chlorination treatment for ballast water
Table 12. The breakdown of available chlorine.
Density of
Residual chlorine (pH value)
The results indicate that if other parameters are the same, the higher the density of organisms is,
the higher the amount of available chlorine demand. The consumption of available chlorine
becomes faster with the density of brine shrimp increasing. After a period of time, in the treated
group with no brine shrimp, the concentration of available chlorine no longer decrease or decrease
slowly. In the treated groups with the density of nauplii below 30/ml under 40 mg/l available
chlorine, there are no survival after 24h.
Conclusions and recommendations
(1) The chlorination treatment can kill harmful organisms in ballast water, but the concentration of
available chlorine demand varies with different target organisms.
(2) Because chlorinated compounds belong to oxidizing disinfectants, the concentration of available
chlorine demand increases with the biomass of organisms. The maximum concentration of
available chlorine demand for ballast water and natural seawater is 20-60 mg/l.
(3) Because of high pH value, the chlorination treatment using a high concentration of available
chlorine will corrode the ballast tank.
(4) The available chlorine breaks down quickly and it is hard to mix in tank. In addition, chlorination
should be carried out before the cysts and spores with high resistance to biocide are produced. So
the pipe and installation for biocide addition should be installed near to the entrance of ballast
(5) Further tests on harmful effects of the by-products of chlorination treatment should be done.
As part of the Global Ballast Water Management Program in China, this project is funded by
GEF/UNDP/IMO. We would like to thank these international organizations.
We also wish to acknowledge the assistance provided by Maritime Safety Administration of China.

Use of chlorine for ballast water treatment
J. S. Vianna da Silva & F. da Costa Fernandes
Admiral Paulo Moreira Marine Research Institute
Treatment options being researched
This study explored the potencial of using chlorine as a biocide to treat ballast water and the
formation of toxic subproducts like trihalomethane (THM).
Timeframe of the project
The project was carried out from March 1999 to September 2001. The experiments were done during
8 days on board and 3 days in laboratory, between June and July of 2000.
Aims and objectives of the project
The objective of this study was to assess the efficacy of chlorine as a biocide, to determine its
minimum concentration to eliminate organisms in ballast water and to observe the formation of
trihalomethane, on board. This study also is concerned about the evaluation of survival of microalgae
and trihalomethane formation in laboratory in different concentrations of chlorine and cells.
Research methods, test protocols and experimental design
On board of the bulk carrier Frotargetina.
The Experiments were carried out on board during a trip from Port of Forno in Arraial do Cabo, Rio
de Janeiro State, to Areia Branca Terminal in Areia Branca, Rio Grande do Norte State, Brazil. Four
superior lateral tanks, on the port side, were kept as control and other four, on the starboard side, were
treated, with different concentrations of chlorine: 1, 3, 5 and 10 ppm. During 4 days before the
departure of the ship, according to the unload of cargo, 2 tanks (control and test) were ballasted at
Forno Port. Every day, during 6 to 8 days, samples of water from all tanks were collected to analyze
salinity, pH, temperature, dissolved oxygen, ammonia, nitrite, nitrate, phosphate and chlorine. The
analyses of salinity, pH, temperature and chlorine were done on board. The dissolved oxygen was also
analyzed immediately after the sampling, according to the Winkler technique described by Strickland
& Parsons (1972). For qualitative and quantitative analysis of zooplankton, 100 liters of water were
pumped, filtered with a sieve with a mesh of 75µm and analyzed in lab with a Leitz stereomicroscope.
For the qualitative and quantitative analysis of phytoplankton, 500 mL of water were collected. After
sedimentation in 50 mL tubes, the cells were counted in an Olympus inverted microscope by the
Utermöhl method. The samples of zooplankton and phytoplankton were done daily, in all tanks, with
three replicates and fixed in formalin at 4%. In the last day of the experiment, 500 mL of water were
pumped from the test tanks for trihalomethane analysis, using the gas chromatography method.
In the laboratory
For the analysis of the THM formation, 24 liters of seawater were collected, filtered, through
Millipore HA filter of 47 mm with pore of 0.45µm and kept in erlenmeyers of 1 liter. The experiment,
done in duplicate, was kept for 72 hours in the darkness with temperature and salinity similar to those

da Silva: Use of chlorine for ballast water treatment
found in the ballast tanks (25°C and 35 ppp),. Four concentrations of chlorine were tested: 1, 3, 5 and
10 ppm in three different concentrations of organic matter: 10x106, 5x106 and 1x106 cells/liter, using
the microalgae Tretaselmis chui.
Chlorine is the most used biocide in chemical industry and in sewage treatment. For several decades it
has been chosen as a disinfectant in water treatment. However, the efficiency of the chlorine is related
to the neutral pH. In general, the water is usually neutralized before using chlorine. Seawater has an
alkaline pH, around 8, and possibly this is one of the disadvantages of using chlorine in ballast tanks.
The other disadvantage is the combination of chlorine with organic matter that form trihalomethane, a
carcinogenic substance.
Physical and chemical characteristics of the water in the ballast tanks
The physical and chemical variables were relatively stable during all the experiment. The salinity,
around 35, remained constant. The temperature had a slight increase, especially in the tanks on the left
side, due to the solar incidence, varying from 2l.5°C to 25.5°C. The changes of temperature inside of
the ballast tanks are related to the environment temperature to which the ship is submitted (Carlton et
., 1993).
The dissolved oxygen presented a variation from 4.06 to 5.50ml/L. This variation is related to the
initial concentration, density of organisms, size of the tank and to the quantity of air that remains in
the tank after the ballast (Committee on Ships Ballast Operations, 1996).
The water pH was constant, around 8, in all tanks, except in tank 1, treated with a higher
concentration of chlorine (10 ppm), where the pH was around 5, during the whole experiment.
Nutrients also were very constant during all the experiment. The values were kept about the same of
the beginning of the experiment.
Zooplankton: characterization, mortality and survival
The zooplankton abundance on the first day of the experiment varied from 2.010 to 10.281
individuals.m-3. There was a gradual decrease of the individuals during the experiment in all the tanks,
varying, on the last day, from 40 to 276 individuals.m-3 in the control tanks, and from 5 to 130
individuals.m-3 in the chlorinated tanks. The mortality of the organisms, in the last day, in the eight
studied tanks varied from 96.26% to 99.01%.
On the second day, however, 24 hours after the application of the sodium hypochlorite, an important
decrease in the zooplankton mortality was observed in all tanks, varying from 67.32% to 90.88%.
Despite of the difference in the number of individuals in each tank, the organism composition was the
same in all tanks. Among the organisms collected in the control tanks, 33.24% were copepod;
32.23%, nauplii; 32.02%, barnacle larvae; 1.43%, mollusk larvae; 0.74%, polychaete larvae and
0.33%, other groups with low density (foraminiferans, tintininas, cladocerans, appendicularians and
In the chlorinated tanks, the copepods were the most abundant with 52.26%, followed by barnacle
larvae with 26.39%, nauplii with 17.76%, mollusks larvae with 2.06%, polychaete larvae with 0.71%
and others with 0.82%.
Phytoplankton: characterization, mortality and survival
The phytoplankton abundance, on the first day of the experiment, varied from 1.187 × 103 to 2.640 ×
103 cells.m-3. The gradual decrease observed in the zooplankton organisms, during the experiment,
also occurred with the phytoplankton cells, in all tanks, varying, on the last day, from 173 × 103 to

2nd International Ballast Water Treatment R&D Symposium: Proceedings
640 × 103 cells.m-3 in the control tanks, and from 0 to 387 × 103 cells.m-3 in the chlorinated tanks. The
mortality of the organisms on the last day, varied from 75.75% to 100% in all tanks. As occurred with
zooplankton, twenty-four hours after the application of the sodium hypochlorite, the phytoplankton
had an important density reduction in all tanks, varying from 24.50% to 76.47%. The phytoplankton
was gathered into three big groups: diatoms, dinoflagellates and coccoliths. Those groups kept the
same proportions in the control tanks and chlorinated ones, being 93% diatoms, 6% dinoflagellates,
and 1% coccoliths.
Efficiency of the chlorine
To confirm the chlorine concentrations in the tanks, it was measured ten minutes after its application.
On the next day, it was measured again and no chlorine was detected in the tanks with concentrations
of 1, 3 and 5 ppm. In the tank with concentration of 10 ppm, we still observed the presence of chlorine
after 24 hours, what did not happen after 48 hours. So, although the tanks were analyzed for until
eight days, the action of the chlorine in the organism mortality was only effective on the first two days
(Figure 1). From the fourth day on, we could notice that the quantity of individuals was similar in the
control tanks and in the chlorinated ones. We obtained, on the first 24 hours after the application of
the sodium hypochlorite, in all chlorinated tanks, a mortality of the organisms (zooplankton and
phytoplankton) from 24.85% to 76.46% . The statistic analysis comparing the tanks with different
chlorine concentrations using the Test "t de Student", did not show significative differences (p>0.05)
among the treatments.
Trihalomethane formation
The trihalomethane are products formed from the combination of chlorine and organic matter and are
classified as possible carcinogens. The levels of the trihalomethane tend to increase with the pH,
temperature, time and quantity of organic matter. Once it is released, this product persists in the
environment, spreading through the trophic chain, accumulating in the adipose tissue, destroying and
blocking the hormonal system (Jenner et al., 1997).
The Environmental Protection Agency, USA, was the first to recommend, in 1979, that the maximum
limit for the concentrations of THM should be of 100 µg/l in potable water. In Brazil, according to the
Decree n° 36 from 19/1/1990, from the Ministry of Health, the maximum quantity of THM in potable
water was also fixed in 100 µg/l.
In the analysis done in the four chlorinated tanks, it was verified the formation of trihalomethane in all
of them, though only in tank 1, where we used chlorine at 10 ppm, the concentration was above the
one permitted by law, 430 µg/L (Figure 2). The low formation of THM, in the tank with 5 ppm, a
relatively high concentration of chlorine, is probably due to the small quantity of organic matter
existing in the water collected in do Forno Port. The region is little impacted, mainly if compared to
the big Brazilian ports.
In laboratory we tried to simulate places with low cellular density, as do Forno Port, Arraial do Cabo,
RJ (<1 × 106cells.l-1), and eutrophic places like the Guanabara Bay, Rio de Janeiro, RJ
(10 × 106cells.l-1).
The results of the tests carried out in labs can be observed in Table 1.

da Silva: Use of chlorine for ballast water treatment
Table 1. THM concentration in different concentrations of organic matter and chlorine.
Concentration of
Concentration of THM
Values above the ones permitted by law.
As we observe in Table I, only the water treated with 1ppm of chlorine kept the THM levels within
the standards permitted by law, varying from 10 to 30 µg/L, in the three concentrations of organic
matter. In the water treated with 3 ppm of chlorine, in the experiments with a lower cellular density,
the formation of THM was between 20 and 25 µg/L, however it reached 1,105 µg/L, in the highest
cellular concentration. In the other concentrations of chlorine and cellular density, the formation of
THM varied from 480 to 1600 µg/L, making impossible to use 5 and 10 ppm in ballast water
treatments, even with low concentrations of organic matter. All cells of the microalgae Tretaselmis
were dead, in all concentrations of chlorine and organic matter, in 24 hours after the beginning of
the experiment.
· The physical and chemical variables (S°/°°, T°, pH and nutrients) remained relatively stable in
all tanks, control and chlorinated.
· The copepods and nauplii were the predominant organisms of the zooplankton and the
diatoms predominated in the phytoplankton, in all tanks, control and chlorinated. In the
control tanks we could observe higher richness of species, without alteration of the
predominant groups, that were similar in all tanks.
· The samples had a gradual decrease of organisms during the experiment in all tanks and there
were significative differences (p<0,05) among the control tanks and the chlorinated ones in all
treatments. Although there were not 100% efficiency in any treatment, the chlorine increased
the mortality inside the tanks.
· The lowest tested concentration, 1ppm, showed a good performance. It presented low
concentrations of trihalomethanes, far bellow the one permitted by law, ant its efficiency was
the same as the other treatments.
· It seems reasonable the use of low concentrations of chlorine to eliminate the organisms in
ballast tanks. It would be interesting to do complementary studies, with daily applications or
in a continuous flux, with the aim of maximize the chlorine efficiency. It must also be
emphasized that the chlorine is an inexpensive product and easy to handle.
· Chlorine concentrations above 3 ppm should not be used, especially in eutrophic
environment, due to the formation of high concentrations of trihalomethane.
· The chlorine dioxide seems to be more suitable to the treatment of ballast water as an
alternative to chlorine, because it does not form THM, and it is efficient in low concentrations
and in any pH. We suggest experiments with ClO2 to verify its efficiency to eliminate
organisms in ballast tanks, cost and handling on board.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
This project was supported by Brazilian Petroleum S/A ­ Petrobras
Carlton, J. T.; Reid, D. M. & Van Leeuwen, H. 1993. The role of shipping in the introduction of
nonindigenous aquatic organisms to the coastal water of the U.S. and an analysis of control
operations. The National Biological Invasion Shipping Study (NABISS), 390 pp.
Committee on Ships Ballast Operations. 1996. Stemming the Tide. Ed. Nat. Acad. of Sci.
Jenner, H. A.; Taylor, C. J. L.; Van Donk, M. & Khalanski, M. 1997. Chlorination by-products in
chlorinated cooling water of some European coastal power stations. Marine Environmental Research.
43:279 - 293.
Strikland, J. D. & Parsons, T.R. 1972. A practical handbook of seawater analysis. Bull. Fish. Res.

da Silva: Use of chlorine for ballast water treatment
Figure 1. Total density of the organism (zoo and phytoplankton) in the chlorinated and control tanks.
Figure 2. Concentration of trihalomethane in chlorinated tanks.

SeaKleen®, a potential product for controlling aquatic
pests in ships' ballast water
S. J. Cutler1, H. G. Cutler1, J. Glinski2, D. Wright3,
R. Dawson3 & D. Lauren4
1 Garnett, Inc., USA
2Planta Analytica, USA
3University of Maryland
Center for Environmental Science
Chesapeake Biological Laboratory
4HortResearch, Ruakura Research Centre
New Zealand
While investigating the use of various natural products as molluskocidal agents, it was observed that
several agents belonging to the chemical class of naphthoquinones were found to be highly effective.
Further investigation in the structure-activity-relationship led to the biologically active agent
menadione, which is being developed under the trademark SeaKleen®. This product has been shown
to possess significant efficacy against a wide variety of estuarine and fresh water organisms including
Cyprinodon variegatus, Eurytemora affinis, Isochrysis sp., Neochloris sp., and Glenodinium foliacium
cysts. In addition, current studies have shown SeaKleen® is very effective against free swimming
Glenodinium foliacium, Cyclopoidea sp (Cyclops). In order to gain a better understating of its effects,
studies were designed to evaluate SeaKleen® against the edible oyster, Mytilus galloprovincialis .
Based on the broad spectrum activity of SeaKleen® against marine organisms and its high potential
as a commercial product, it was of interest to determine the degradation of the active component,
menadione, when subjected to normal applications. Using an HPLC assay, SeaKleen® was subjected
to sterilized and unsterilized sea and fresh water over a period of 72 hours, and samples taken at 24
hour intervals, to determine longevity and breakdown. Results, to date, indicate that SeaKleen® is an
environmentally friendly and cost effective ballast water treatment to control invasive species.

Treatment options being researched
Chemical (Biocide)
Timeframe of the project
July 2001-June 2003
Aims and objectives of the project
This project includes the evaluation of SeaKleen® against a variety of aquatic nuisance species
residing in ballast tanks of ships. This project fills the gap on tests that have been performed since the
last IMO meeting in 2001. Furthermore, this study includes the degradation of the active principle in
fresh and salt water studies using High Performance Liquid Chromatography (HPLC).

Cutler: SeaKleen®, a potential product for controlling aquatic pests in ships' ballast water
Research methods, test protocols, and experimental design
In mid 1988, Eurasian zebra mussels, Dreissena polymorpha, were found in the Great Lakes, North
America, and their entry was determined to have been in the ballast water carried by ships. Within a
short time, their effect on commercial and recreational water supplies became readily apparent. By
January 2000, twelve years after the initial discovery, zebra mussels had spread from the Great Lakes
to Louisiana and west into Texas. Water intake pipes, for example in power plants, have been clogged
resulting in a 50% reduced water flow rate. And, the mussels apparently secrete metabolites that etch
away ferrous pipes. This occurs whether they are living, or dead. In drinking water, they produce an
off-flavor even after water purification. This effect is associated with the production of polyamines,
especially cadaverine, the latter being a distinctive odor in decaying corpses, animals and certain
plants and fungi. Its main purpose is to attract insects, especially flies, which lay their eggs, then hatch
to produce larvae that feed upon the decaying matter, thus ensuring survival of the next generation. In
the case of plants, the flies act as pollinators, an example being the Deadhorse Aurum, and,
presumably, in fungi they spread propagules to other sites
While attending the 1st World Congress on Allelopathy, 16-20 September, 1996 in Cadiz, Spain, Dr.
Horace Cutler listened to the effects of various natural products, as described by CB Rogers of the
University of Durban ­ Westville, South Africa (Rogers 1996). In his presentation, he informed his
audience that Combretum tree species had no plants growing within its canopy, or dripline and,
therefore, he proposed that certain natural herbicides were produced and exuded by the leaves. As an
afterthought, Rogers added that the sodium salts of bioactive metabolites, which included mollic acid
and imberic acid, were also active against snails, for example Biomphalaria glabrata. Recognizing
the relationship of snails to zebra mussels (Dreissena polymorpha), which are present in the many
areas including the Great Lakes of the United States, it was decided to evaluate these compounds as
potential biocides in controlling this aquatic nuisance species.
Upon evaluation, it was noted that other natural products had similar structural features and, possibly,
similar effects on snails. Furthermore, it was recognized that juglone, a quinone produced by walnuts
shells (Juglans nigra and other species), had been evaluated by the United States Department of
Agriculture as a potential natural herbicide. From this, we began to investigate the usefulness of
various quinones against snails and slugs, particularly those that are recognized as aquatic nuisance
Other ballast pests
The US Fish and Wildlife Service presently calculates that the cost of introducing non-indigenous
pests to North America amounts to more than $100 billion, annually. Other pests, beside zebra
mussel, including the spiny water flea, Bythotrephes cedarstroemi; the Eurasian ruffe,
Gymnocephalus cernuus, a non-game fish; and the round goby Proterorhinus marmoratus, have been
introduced to the US. Other pests, among them the dinoflagellates Prorocentrum, Gymodinium,
and Gonyaulax also hitch rides and become unwelcome visitors. These are well known
for their ability to cause blooms that kill fish and destroy commercial shellfish industries. The best
known of these is the "red tide", which turns water blood red. It has been speculated that the first
plague sent by God to convince Rameses II that he should release the Israelites from captivity, was a
"red tide". As the Bible states this caused the Nile (the only river in Egypt) to turn to blood and, "The
fish in the river will die, and the river will smell so foul that the Egyptians will not want to drink the
water of it
(Exodus 7:18). It should be noted that there followed an interesting development, "But the
magicians of Egypt used their witchcraft to do the same, so that Pharaoh's heart was stubborn......"
(Exodus 7:22). The latter actions lends credence, but not proof, to the dinoflagellate theory since a
period of time obviously followed the first red flush in the Nile, and may have occurred later in its
Another effect caused by dinoflagellates is that they can colonize Caribbean waters, especially where
reef fish, such as grouper, feed. The net result is that upon ingestion certain toxins accumulate in the
flesh and travelers who consume the fish may become stricken with Ciguatera poisoning, which may

2nd International Ballast Water Treatment R&D Symposium: Proceedings
be fatal. This is manifested by characteristic symptoms where the subject may feel tingling, or
numbness at the extremity of the fingers, and will feel cold on hot days to the point of wanting to wear
an overcoat when it is 30°C. Cold liquids, especially those containing ice, will feel scalding hot to the
mouth, and conversely, hot liquids feel ice cold. Recovery may take several months, and death is not
uncommon. The medical costs have not been included in the economic equation for this class of
marine pests.
An algal bloom toxin reported to have killed over fifty sea lions in Monterey Bay, in 1999, may also
be affecting blue and humpback whales, both endangered species. A bloom that appeared in the
summer of 2000 had entered the food chain on which the whales fed (Atlanta Journal, 2001).
Yet another ballast water import is the cholera bacterium, Vibrio cholera. Although some public
concern was voiced over the discovery of this bacterium in ships entering the Chesapeake Bay, it
should be noted that Vibrio bacteria are common in Chesapeake Bay waters while conditions do not
support the development of the disease. In the Southern United States, the conditions do favor the
development of cholera. The vector appears to be planktonic copepods (crustaceans), which emigrate
from South America to the US Gulf coast ports, their initial route being from Europe to South
America. It is, in fact, this tortuous passage via ballast water that points to the seriousness of the
migratory pest problem. So much so, that legislation has been enacted in the USA.
In 1990, the United States Congress passed Public Law 101-646. The legislation, "The
Nonindigenous Aquatic Nuisance Prevention and Control Act", an article of which was the, "National
Ballast Water Control Program." Therein, are mandated studies to control the introduction of aquatic
pests into the United States. Among the propositions posed are ultraviolet irradiation, filtration of all
types including voraxial separators, ultrasonic perturbation, ozonation, thermal and electrical
treatments, reduction in available oxygen, and chemical treatment.
If chemical treatment is to be successfully employed it must be effective, environmentally benign and,
therefore, biodegradable. It must also have high specific activity and be target specific. To reiterate,
such characteristics are typical of organic natural products. Consequently, upon examination of the
three natural products discussed earlier, menadione, 1,2- and 1,4-naphthalenedione, were the most
likely candidates to control aquatic phytoplankton by growth inhibition, or phytocidal activity, was
menadione. Any other effects would be gratuitous, for example, controlling the growth of zebra
mussels, or dinoflagellates, or cholera, and other pests.
Initial experiments (vide infra) showed that many of the quinones evaluated had toxicity towards
aquatic nuisances species. Of these, three naphthalenediones had specific activity against target pests.
It became apparent that if we were to proceed toward the ultimate goal of a practical aquatic pest
control product, price would be an overriding factor, coupled to the availability and generally known
toxicity. Of the three, menadione (vitamin K3) became the selected candidate. On a cost basis,
menadione is 50 cents/gram; 1,2-naphthalenedione is $14.30/gram; 1,4-naphthalenedione is
$1.20/gram (all costs are calculated as price/gram even though, for example, the minimum purchase
for menadione is 5 grams. Aldrich Catalog 2002-2003). However, it is imperative to realize out that
fine laboratory chemicals are extraordinarily expensive relative to the source material.
Vitamins belonging to the K group are polyisoprenoid substituted naphthalenediones with vitamin K3
serving as the parent template. Although initially believed to be simply a synthetic derivative of
vitamin K1 and K2, after therapeutic administration, menadione is readily converted, in the liver, to
vitamin K2 (Marcus and Coulston, 2001). As a group, these are ubiquitous, natural compounds that are
found in microorganisms, plants, and animals. Vitamin K1 and K2 are essential substances that are
required by all mammals, including humans, for the regulation of normal blood clotting factors.
Found in leafy dark vegetables, which may serve as a dietary source, the major requirements of the
body are met by gut microorganisms that feature the ability to synthesize vitamin K1 (Bently and
Meganathan, 1982).

Cutler: SeaKleen®, a potential product for controlling aquatic pests in ships' ballast water
The vitamin is absorbed from the gastrointestinal tract where it circulates in the blood, playing a
critical role in the biosynthesis of prothrombin, a protein responsible for blood clotting.
Posttranscriptional modification of prothrombin, as well as factors I, VII, IX, and X, results in the
formation of blood clotting proteins which may be found in the plasma. This posttranscriptional
process is totally determined by vitamin K in which it converts the glutamate residues of the precursor
proteins to gamma-carboxyglutamate residues of the functional coagulation factors. These specific
products are the sites of the Ca+2 binding, and are essential to their role in the clotting cascade.
Menadione is used in chicken feed to control a hemorrhagic syndrome brought on by feeding
synthetic rations, hence the term vitamin K [for Koagulans vitamin, a term coined by Henrik Dam]
(Dam & Schonheyder, 1935). Dam et al., isolated vitamin K from alfalfa and fishmeal, in the K1 and
K2 forms. In addition, menadione is used in feeds for turkeys, swine, cattle, and catfish, with the latter
serving as an aquatic application of menadione at use rates of 4.4 ppm. (Robinson et al, 2001).
Although primarily produced by synthetically, menadione is found to exist naturally (Mikhlin, 1942;
Mikhlin, 1943). Although there has been skepticism to these publications, the United States
Department of Agriculture Agricultural Research Service (USDA-ARS) has isolated menadione from
various walnuts (Binder et al, 1989; Thomson, 1997) The world's primary source of menadione,
which can be synthesized from -methylnaphthalene by oxidation with chromic oxide under mild
conditions, is used clinically as a prothrombogenic [a blood clotting agent]. One of the most
significant uses today involves hypothrombinemia of newborn infants. In this condition, the neonate is
incapable of producing enough vitamin K, but this can be remedied by administering the deficient
vitamin, as menadione (Committee on Nutrition, 1961). In veterinary medicine it is used in
hypoprothrombinemia and bishydroxycoumarin poisoning, and sweet clover poisoning (Merck Index,
1996). On a molar basis menadione is identically active to vitamin K1 and may be used orally,
intramuscularly, and intravenously. C14 studies indicate that it is converted in vivo to vitamin K2, the
side chain genesis being through mevalonic acid. While it has limited solubility in water, 1 gram
dissolves in ~ 60 mL ethanol. It is stable in air, but is rapidly degraded by sunlight and ultraviolet.
However, its most utilitarian form is as menadione sodium bisulfite [1,2,3,4-tetrahydro-2-methyl-1, 4-
dioxo-2-naphthalenesulfonic acid] and it is this form that is used pharmaceutically: one gram will
dissolve in ~ 2 mL of water.
Initial aquatic experiments with naphthalenediones
As presented at the first International Maritime Organization meeting in London (Wright, 2001),
replicated experiments were conducted with menadione and menadione sodium bisulfite against the
following organisms: the marine alga, Isochrysis galbana; the freshwater green alga, Neochloris sp.;
zebra mussel larvae, Dreissena polymorpha; an estuarine copepod, Eurytemora affinis; a bacterium
congeneric with V. cholerae, Vibrio fischeri; the toxic marine dinoflagellate, Proprocentrum
; dinoflagellate cysts, Glenodinium sp.; the benthic amphipod crustacean, Leptocheirus
; Sheepshead minnow, eggs and larvae, Cyprinodon variegatus; and oyster larvae,
Crassostrea virginica (Table 1).
1,2-naphthalenedione was tested against I. galbana; E. affinis ; and V. fischeri, while 1,4-
naphthalenedione was tested in all these, plus Neochloris, in replicated experiments (Table 2).

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Table 1. A Summary of the Effects of Menadione (Vitamin K3) Against Assorted Ballast Water Pests1.
Toxicity Levels
T. Isochrysis galbana
Toxic at 1.0 ppm and above
Marine Algae
Neochloris sp.
Toxic at 500 ppb and above
Freshwater algae
Dreissena polymorpha
Toxic at 500 ppb and above
Zebra mussel larvae
Eurytemora affinis
Toxic at 1.5 ppm and above in less than
Estuarine copepod
20 hours
Vibrio fisheri
Toxic at 1.0 ppm and above
Congeneric with V. cholerae
Proprocentrum minimum
Toxic at 500 ppb and above
Marine dinoflagellate cysts
Glenodinium sp.
Toxic at 2.0 ppm after 2 hours
Dinoflagellate cysts
Leptocheirus plumulous
Toxic at 2.0 ppm and above
Benthic amphipod crustacean
Cyprinodon variegatus
Eggs: toxic at 1.0 ppm. Kills and/or
Sheepshead minnow
prevents hatch
Larvae: toxic at 1.0 ppm and above
Crassostrea virginica
Toxic at 500 ppb and above
Oyster larvae
1Results obtained from replicated experiments; toxicity represents 100% kill
Table 2. A Summary of the Effects of 1,2- and/or 1,4-Napthalenedione (NLD) Against Assorted Ballast Water
Toxicity Levels
T. Isochrysis galbana
Toxic at 375 ppb after 1 minute
Marine Algae
Toxic and bleaches at 1 ppm
T. Isochrysis galbana
Toxic at 750 ppb after 1 minute
Marine Algae
Toxic at 750 ppb
Vibrio fisheri
Toxic at 500 ppb
Congeneric with V. cholerae
Toxic at 500 ppb
Neochloris sp.
Toxic at 1 ppm after 24 hours
Freshwater algae
1Results obtained from replicated experiments; toxicity represents 100% kill.
Based on these findings, it was of interest to determine the concentration-response that menadione has
against Isochrysis galbana and Glenodinium foliaceum. This report describes the experiments used to
determine these effects. Furthermore, since menadione is being considered for use in ship trials on the
west coast of the United States, it was important to gain understanding of the toxic profile it possess
against the indigenous aquaculture that is present. Studies were performed on Mytilus
in order to determine the effects menadione on this edible mussel. Also, the
degradation of menadione was evaluated in fresh and salt water using high-performance-liquid-
chromatography (HPLC) so that the half-life could be calculated.
Research methods
The media and test organisms Isochrysis galbana and Glenodinium foliaceum were obtained from the
culture facility of Carolina Biological (Burlington, North Carolina USA) and maintained under 16h:8h
light/dark regime at 22°C. Exposure of Glenodinium foliaceum cysts to SeaKleen® were conducted at
the Kalmar Marine Institute, Kalmar, Sweden (Professor E. Graneli) where they were examined by

Cutler: SeaKleen®, a potential product for controlling aquatic pests in ships' ballast water
epifluorescence microscopy. The assays involving Mytilus galloprovincialis were performed by a
contract laboratory, Northwest Aquatic Labs (New Port, Oregon USA).
All cell counts were preformed using an Improved Neubauer, 1/400 square mm Hemacytometer
(Hausser Scientific, Horsham, PA USA). All counts were completed in duplicated and an average of
these counts was used.
Isochrysis and Glenodinium
The alga Isochrysis galbana was cultured in Soil Water Medium as a 1 liter stock culture. The initial
average cell density count was determined to be 816 x 106 cells per liter. From the stock culture 10 ml
was placed into 12 sterile micro Petri dishes (Fisher Scientific, Norcross, Georgia USA). These were
divided into four groups one of which was a control group and the others three concentrations of
testing for SeaKleen®. The concentrations of formulated menadione included 0.0 ppm (control), 0.250
ppm, 0.750 ppm, and 1.5 ppm and were run in triplicate. Formulated menadione was dissolved in
sterile water as a stock solution. Aliquots were taken from the stock menadione solution and added to
each of the 12 sterile Petri dishes. For the control group, only sterile water was added at the same
volume used in the test concentrations. Cell counts were performed at 2, 4, 6, 24, 48, 72, and 96 hours
after the test solution was added. The cultures were maintained under 16h:8h light/dark regime at
22°C during this 96 hour period.
The dinoflagellate Glenodinium foliaceum was cultured in Alga Gro® Medium as a 1 liter stock
culture. The initial average cell density count was determined to be 147 × 106 cells per liter. From the
stock culture 10 ml was placed into 12 sterile micro Petri dishes (Fisher Scientific, Norcross, Georgia
USA). These were divided into four groups one of which served as a control group and the others as
three concentrations of testing for SeaKleen®. The concentrations of formulated menadione included
0.0 ppm (control), 0.250 ppm, 0.750 ppm, and 1.5 ppm and were run in triplicate. Formulated
menadione was dissolved in sterile water as a stock solution. Aliquots were taken from the stock
menadione solution and added to each of the 12 sterile Petri dishes. For the control group, only sterile
water was added at the same volume used in the test concentrations. Cell counts were performed at 2,
4, 6, 24, 48, 72, and 96 hours after the test solution was added. The cultures were maintained under
16h:8h light/dark regime at 22°C during this 96 hour period.
The dinoflagellate cysts were collected from marine sediments cleaned of debris using mild ultrasonic
cleansing and exposed to 2.0 ppm of formulated menadione. Light microscopy and epifluorescence
microscopy were employed to examine the cysts for oxidative damage and chloroplast disruption
following treatment at 2.0 ppm level.
Formulated menadione was tested in order to estimate the chronic toxicity of water discharge from a
source such as a ballast tank. This was performed by assaying bivalve larval development in a 48-hour
static test. This protocol complies with the U.S. EPA West Coast chronic toxicity annual (EPA/600/R-
95/136), ASTM bivalve toxicity method (E 724-89), and the WDOE (Washington State Department
of Ecology) toxicity guidance manual (WQ-R-95-80).
Adult mussels (Mytilus gallaprovincialis) were collected from Yaquina Bay, Oregon and immediately
used in testing. The source of the gametes were from 1 female and 1 male (gametes of male physically
stripped from gonads). Eggs from the female were filtered (200-300 µm) to remove feces and
pseudofeces and adjusted in concentration to about 2500-6000/ml. Eggs were then fertilized by
addition of sperm from the male. Ten minutes after adding the sperm, the egg and sperm mixture was
poured through a 25 µm screen to remove excess sperm; then the eggs were rinsed and resuspended in
water. The embryo density was adjusted to between 1500 and 3000/ml. Embryos were kept suspended
by frequent gentle agitation with a perforated plunger and the temperature was maintained at

2nd International Ballast Water Treatment R&D Symposium: Proceedings
approximately 16°C. The quality of the embryos was verified before testing by microscopic
examination. Embryos were used 2.7 hours post-fertilization.
Three formulations of SeaKleen® were used in this study;100:0, 80:0, and 0:100, menadione sodium
bisulfite:menadione, respectively. Five concentrations of the three formulations were prepared 2 days
prior to the bioassay test and on the day of the test. Vials containing 8 mg of each formulation were
mixed in 4 liters of filtered seawater to make up the highest concentration (2 ppm) and were
subsequently diluted to make up the remaining concentrations of 1.0, 0.5, 0.2, and 0.1 ppm. A control
group (0 ppm) was used for comparison. The 2 day solutions were then either stored under dark or
light conditions at 15°C prior to the bioassay testing.
Larvae were placed in 30 ml glass vials containing 10 ml of test solutions. The average number of
embryo in the 10 ml solution was 247. The temperature was maintained at 20°C with a photo period
of 16:8 hours (light:dark) for 48 hours to permit development into prodissoconch I larvae. Larvae
were subsequently counted to determine the total number of abnormal and normal surviving larvae.
Each sample was performed in quadruplet.
Where data permitted, the EC50s and LC50 were calculated using either the Maximum-Likelihood
Probit or the Trimmed Spearman-Karber methods. NOEC and LOEC values for survival and
normality were computed using either Dunnett's test, T-test with Bonferroni's adjustment, Steel's
Many-one Rank Test, or Wilcoxon Rank Sum Test with Bonferroni Adjustment. The statistical
software employed for these calculations was Toxcalc, v.5.0.23N (Tidepool Scientific Software).
HPLC Studies
The tests were performed using a bulk sample of seawater collected from either Flax Mill Bay or
Raglan, or river water collected from the Waikato River in the Hamilton City area of New Zealand.
For tests involving exposure to aquatic organisms, pond water was collected from catfish ponds in
either Indianola, Mississippi or at the National Warmwater Aquaculture Center, Stoneville,
Mississippi. The water samples were stored at 6°C prior to testing. Samples of SeaKleen® were used
at either a 2.0 ppm or 1.0 ppm concentration of active material. These concentrations were prepared
using the collected water samples and all samples were analyzed at 4, 24, 36, 48, 72, and 96 hours.
For the river and sea water, samples were stored at 6°C in darkness or under full exposure to daylight.
For the pond water assay, SeaKleen® was only evaluated under normal photoperiods of day and night
during the summer of 2002.
Analysis of the samples was accomplished using High-Performance-Liquid-Chromatography (HPLC)
(Hewlett-Packard model 1090) with simultaneous fluorescence and UV detection or with a Diode
Array Detector. The HPLC column used was a YMC C18 ProPack (100 mm x 3 mm, 3 µm particle
size (YMC Co. Kyoto, Japan). The mobile phase consisted of acetonitrile and 0.1% trifluoroacetic
acid. The column oven temperature was set at 450C with a flow rate of 0.6 ml/min. The detection
wavelength as set at 230 nm (to monitor menadione sodium bisulfite) and 263 nm (to monitor
menadione). The injected sample volume was 4 µl. Calibrations of the HPLC response for menadione
sodium bisulfite and menadione was performed using a series of dilution of standards obtained from
the United States Pharmacopeia (USP). The detection limit for menadione was determined to be 0.56
µg/l (0.56 ppb).
Isochrysis and Glenodinium
The effects of SeaKleen® on Isochrysis galbana and Glenodinium foliaceum were found to be within
the range of 0.25 ppm and 1.5 ppm of active ingredient. After the organisms were allowed to
equilibrate, SeaKleen® was added at four concentrations in order to determine the concentration-
response curve. These results are listed in Figure 1 and Figure 2 for Isochrysis galbana and

Cutler: SeaKleen®, a potential product for controlling aquatic pests in ships' ballast water
Glenodinium foliaceum, respectively. SeaKleen® produced a significant inhibition in the growth of
both these organisms at 0.750 and 1.5 ppm. In addition, it produced a significant inhibition of
Glenodinium at the 0.250 ppm.
Using light microscopy and epifluorescence microscopy, Figure 3 shows the effect of SeaKleen® at
2.0 ppm against cysts of Glenodinium foliaceum. It is clear from this figure that the cysts suffered
oxidative damage and chloroplast disruption following treatment.
The edible mussel, Mytilus gallaprovincialis, was used in a bioassay to gain an understanding of the
effects of SeaKleen® released from ballast tanks into United States harbors. Therefore, studies were
conducted to evaluate the effects of SeaKleen® on aquatic organisms present in Puget Sound, USA.
As shown in Tables 3 and 4, SeaKleen® exhibited similar effects whether stored under light or dark
conditions. In the 80:20 (menadione sodium bisulfite:menadione) formulation, there was a sharp drop
off in toxicity from 0.5 ppm to the 0.2 ppm concentrations, while with the 0:100 (menadione sodium
bisulfite:menadione) formulation, this drop off was seen at a lower concentration (0.2 ppm to 0.15
ppm). This is to be expected as the weight of active ingredient is greater for menadione than for
menadione sodium bisulfite since the sodium bisulfite (inactive material) constitutes 37.7% of the
weight for that particular salt. Therefore, the actual active ingredient, menadione, in the salt is 62.3%.
Using this correction, the studies involving the 80:20 formulation yield similar cut-off values to those
obtained in the 0:100 formulation.
Table3. Effects of SeaKleen® towards Mytilus gallaprovincialis (80:20 Formulation).
Dark Storage - 48 Hrs
Dark Storage - 48 Hrs
Light Storage - 48 Hrs
Bioassay - Dark
Bioassay - Dark
Bioassay - Light
Concentration ppm
% Mortality
% Mortality
% Mortality
Table 4. Effects of SeaKleen® towards Mytilus gallaprovincialis (0:100 Formulation).
Dark Storage - 48 Hrs
Dark Storage - 48 Hrs
Light Storage - 48 Hrs
Bioassay - Dark
Bioassay - Dark
Bioassay - Light
Concentration ppm
% Mortality
% Mortality
% Mortality
HPLC Studies
Since SeaKleen® is being developed for use in treating ballast water it is necessary to calculate the
degradation of the active material under normal use conditions. Studies employing HPLC were
executed in order to monitor the degradation of SeaKleen® under dark conditions, which is typical of

2nd International Ballast Water Treatment R&D Symposium: Proceedings
ballast tanks, and light conditions, which is expected after the release of water from a tank into the
environment. In addition, since the mechanism of action for menadione involves interaction with
aquatic organisms, it is expected that interrelationships might facilitate the degradation of SeaKleen®.
As such, it was importance to evaluate the fate of menadione under conditions where aquatic nuisance
species were present in the water.
HPLC studies under dark conditions and in the absence of high levels of aquatic nuisance species
show that menadione is relatively stable in sea water with a drop of 8.5% in menadione concentration
after 72 hours. After 28 days, only 21% of the original starting concentration of active ingredient was
present. Studies under light conditions were differed to the dark studies in that the rate of degradation
was faster. After 72 hours, only 47% of the original starting concentration was present.
In river water, the HPLC studies show that the degradation is slightly different than seen under sea
water conditions. In the absence of high levels of aquatic nuisance species, under dark conditions, the
degradation was almost identical to the dark sea water results. After 72 hours, there was 80.5% of the
original concentration, but after 28 days, the concentration was only 22%. However, under light
conditions, only 8% of the original concentration of SeaKleen® remained after 72 hours. This was
attributed to a higher microbial load present in the river water used in this assay. It is believed that this
phenomenon caused SeaKleen® to be more actively consumed, because for each organism at least one
mole of SeaKleen® is expended. Based on this, it was decided to perform HPLC studies in the
presence of aquatic nuisance species.
When SeaKleen® was used at 0.8 ppm active ingredient, in the presence of the aquatic nuisance
species Oscillatoria perornata, degradation was very rapid. At 48 hours, SeaKleen® was approaching
the lower levels of detection limits (0.56 ppb) and after 72 hours, it was no longer detectable. This
suggests that the degradation of SeaKleen® is dependent on many factors, especially the presence of
aquatic nuisance species. It is believed that there is a direct relationship between the number of moles
of menadione in a given treatment and the number of organisms. Simply, Avogadro's number most
likely plays a definite role in aquatic nuisance species death and in the degradation of SeaKleen®.
During the HPLC studies, non toxic by-products were detected. This was accomplished by following
the degradation of SeaKleen® by HPLC as well as the concomitant use of Liquid-Chromatography-
Mass spectrometry (LC-MS).
Conclusions and recommendations
Initial studies suggested that SeaKleen® has potential as an agent to control aquatic pests in both fresh
and salt water at 1 ppm, or less. Studies, to-date, show that there is a correlation between
concentration and toxicity. For most aquatic organisms, the toxicity is found to be between 0.50 ppm
and 1.5 ppm active ingredient. Further, there is a sharp, acute drop-off in toxicity suggesting that
when released from ballast tanks, the dilution of any residual material should result in below toxic
levels to indigenous organisms.
Additionally, HPLC studies suggest that under typical ballast conditions (i.e., dark environment)
SeaKleen® is present for a sufficient period of time to ensure complete removal of all aquatic nuisance
species. However, it is important to realize that the presence of organisms will facilitate the
degradation of SeaKleen® to non-toxic levels within the normal travel time of most cargo ships.
Furthermore, the HPLC studies, in conjunction demonstrate that the SeaKleen® degrades to harmless
metabolites suggesting that it is an environmentally friendly natural product.
The low use rates and highly acute but transient toxicity implies that the material can be administered
in low amounts, thereby making it cost effective. It is calculated that 1 gram of SeaKleen® will treat 1
metric ton of ballast water. Thus, a ship with a 10,000 metric ton ballast tank should cost
approximately less than US $2,000. Over the ship's lifetime, this cost should represent significant
savings to the owners. In addition, due to the cost of repairs and replacement parts, the savings are

Cutler: SeaKleen®, a potential product for controlling aquatic pests in ships' ballast water
even more pertinent. One final, but high practical point, is the ease with which SeaKleen® can be
Atlanta Constitution, 14 January, pA8, 2001.
Bently, R., Meganathan, R. 1982. Biosynthesis of vitamin K (menaquinone) in bacteria. Microbiol
46: pp. 241-280.
Binder, R.G., Benson, M.E. & Flath, R.A. 1989. Phytochemistry 28(10): pp. 2799-2801.
Committee on nutrition, American Academy of Pediatrics (1961) Vitamin K compounds and the
water-soluble analogues; use in therapy and prophylaxis in pediatrics. Pediatrics 28: pp. 501-506.
Dam, H., Schonheyder, F. 1935. The antihaemorrhaig vitamin of the chick. Nature 135: pp. 652-653.
Marcus, R. & Coulston, A.M. 2001. Fat soluble vitamins. In: Hardman, J.G. & Limbird, L.E. (eds):
Goodman & Gilman's The Pharmaceutical Basis of Therapeutics. 10th ed. pp. 1783-1786.
Merck Index 12th Edition 1996. page 994.
Mikhlin, D.M. 1942. Vitamin K3, an antihaemorrhage: Corn source for Vitamin K. Comptes Rendus
(Doklady) de l'Acadaemie des Sciences de l'URSS
37(5-6): pp. 191-192.
Mikhlin, C.R. 1943. The peculiarities of the antihemorrhagic factor of the maize stigmata (Vitamin
K3) Biokhimiya 8(4): pp. 158-167.
Robinson, E.H., Menghe, H.L. & Manning, B.B. 2001. A practical guide to nutrition, feeds, and
feeding of catfish.
Mississippi Agricultural and Forestry Experiment Station, Office of Agricultural
Communication, Division of Agriculture, Forestry, and Veterinary Medicine, Mississippi State
University. Oxford, Mississippi.
Roger, C.B. 1996. The allelopathic action of Combretum leaf secretions. 20 September 1996.
Symposium 8. 1st World Congress on Allelopathy. Cadiz, Spain.
Thomson, R.H. in Naturally Occurring Quinones IV Recent Advances, Blackie Academic &
Professional, London, 1997, p 112.
Wright, D.A. & Dawson, R. 2001. "SeaKleen® - A potential natural biocide for ballast water
treatment. 1st International Ballast Water Treatment R&D Symposium. International Maritime
Organization, London 26-27 March 2001: Proceedings
. GloBallast Monograph Series No. 5, IMO
London. Pp. 73-75.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 1. The effects of SeaKleen® on Isochrysis galbana.
Figure 2. The effects of SeaKleen® on Glenodinium foliaceum.
Figure 3. Glenodinium foliaceum cysts 2 hours after exposure to SeaKleen®.

Peraclean® Ocean ­ A potentially environmentally
friendly and effective treatment option for ballast water
R. Fuchs & I. de Wilde
Degussa AG, Germany
The transfer of human pathogens and the introduction of non-indigenous species through the ballast
water of ships has been recognized as a significant problem. The introduction can result in
tremendous costs and may impose a threat on local ecosystems. Globally, approximately 3 billion tons
of ballast water are transported per year. Various treatment options for ballast water have been
suggested (Gollasch, 1997).
Chemical and environmentally friendly treatment with Peraclean® Ocean is one method to effectively
remove unwanted organisms and pathogens in ballast water. This paper summarizes the laboratory
results of a partially funded and already finished research project and covers experimental results of a
shipboard test. It provides details on the efficacy and toxicological properties of Peraclean® Ocean.
Name of project
Testing of Peraclean® Ocean as a chemical ballast water treatment option has been part of a research
project in Germany (1998 ­ 2001), that was funded by the industry (Degussa AG) and the German
Federal Ministery of Education and Research (BMBF) with the title `Process for the removal of
organisms from different waters1.
Properties of Peraclean® Ocean
Peraclean® Ocean is a liquid biocide formulation based on peroxygen chemistry. One active
component in the formulation Peraclean® Ocean is peracetic acid (PAA). PAA- containing
formulations are widely used in the food and beverage industry as well as in sewage treatment plants
and other water treatment processes. They are widely used in the treatment of cooling water and as a
pre-treatment of biologically contaminated waters prior to discharge into the environment. PAA is
accepted in the USA as a secondary and indirect food additive at concentrations up to 100 mg/l.
Peraclean® Ocean is a fast-acting oxidizing biocide effective against a broad spectrum of micro-
organisms: bacteria, spores, yeasts and moulds, protozoa, algae and viruses (Block, 1991; Schliesser,
& Wiest, 1979; Baldry, 1983). Peroxyacetic acid products are effective over a wide range of
conditions. Peraclean® Ocean is most active at pH values of 5-7 but also displays good activity even
under mildly alkaline conditions up to pH 9. Peraclean® Ocean remains effective even at temperatures
of 4°C and below. The microbial activity of peroxyacetic acid based products is relatively unaffected
by organic matter, compared to other oxidising biocides (Block, 1991).
The shelf-life of Peraclean® Ocean is more than 1 year, and: more than 90% of the original activity is
still present after one year`s storage at room temperature. Peraclean® Ocean is commercially available

This publication is based on the results of a research project funded and supported by the Ministry for Research and
Technology of Germany under registration number 02/WA9912. The authors are solely responsible for the content of this

2nd International Ballast Water Treatment R&D Symposium: Proceedings
in 220-kg drums, 1 m3-IBC or in 20-m3 bulk containers. Peraclean® Ocean is readily biodegradable
according to OECD Screening Test 301 E guidelines.
Peraclean® Ocean does not persist in the environment and breaks down into innocuous degradation
products, being acetic acid, water and oxygen:
CH3CO3H + H2O CH3CO2H + H2O2
2 H2O2 O2 + 2 H2O
The hydrolysis products of Peraclean® Ocean are also readily biodegradable.
The half-life of Peraclean® Ocean amounts to minutes to hours in seawater, depending on pH value,
salinity and temperature. In fresh water, the half-life of Peraclean® Ocean is 2-24 hours. Enhanced
decomposition of Peraclean® Ocean may occur in contact with sediments.
Efficacy tests ­ laboratory tests
Several studies showed that many organisms from different trophic levels can be found in ballast
water tanks. For that reason the efficacy testing of a chemical treatment should include organisms
from more than one trophic level (Voigt, 1999).
For a first evaluation of the performance of Peraclean® Ocean, the Artemia Testing Standard (ATS)
was applied. This benchmark test uses the brine shrimp, Artemia salina, as indicator organism. The
ATS involves 4 different development stages of the brine shrimp: adults, larvae, nauplius-stages, pre-
incubated eggs and cysts. The results of the benchmark tests are summarized in Table 1.
Table 1. Results of Peraclean® Ocean on different development stages of the brine shrimp, Artemia salina;
Values in brackets represent the highest mortality reached at the end of the experiment.
Concentration of
Max. Hatching
Time (hrs.) needed to
Brine shrimp,
Parameter observed
Peraclean® Ocean (ppm)
Rate after 72 hrs
reach 100% mortality
Artemia salina
Hatching rate
Survival of
hatched Nauplii
1 400
Pre-incubated Eggs2
Hatching rate
Survival of
hatched Nauplii
1 400
(97%; 72 h)
1 400
(38%; 72 h)
1 400
1 = untreated control group: 52 +/- 8,4%
2 = untreated control group: 47,4 +/- 2,2%.
The ATS data showed that the addition of Peraclean® Ocean at levels of above 350 ppm resulted in
100% mortality of all Artemia live stages. The pH of the treated seawater is slightly reduced from pH
8.2 to 6.1, due to the acidic properties of Peraclean® Ocean.
After the initial tests, further experiments were carried out with a number of indicator organisms. The
experimental designs applied included different salinities and temperatures. In each case, the
experimental conditions represented optimum environmental conditions for the test species.

Fuchs: Peraclean® Ocean ­ A potentially environmentally friendly and effective treatment option for ballast water
Experiments with nauplii of the brine shrimp, Artemia salina, indicated, that only 400 ppm Peraclean®
Ocean is required to reach 100% mortality under varying environmental conditions (Tab. 2).
Table 2. Experiments with Peraclean® Ocean in different water qualities. Test organism: nauplii of brine shrimp
(Artemia salina). Values represent average of 3 parallel experiments.
Note: Observations were made after 1, 2, 4, 8, 12, 24, 36, 48 and 72 hours.
Concentration of
Time (hrs.) needed to
Brine shrimp,
Water Quality
Peraclean® Ocean
reach 100% mortality
Artemia salina
Salinity 13.5ppt
Temp. 24°C
1 200
Salinity 13.5ppt
Temp. 32°C
1 200
Salinity 31ppt
Temp. 24°C
1 200
Salinity 31ppt
Temp. 32°C
1 200
ppt= parts per thousand
Experiments with fertilized eggs of Atlantic herring (Clupea harengus) followed. The eggs were pre-
incubated in clean water for one week to assure an undisturbed start of the larval development. In this
case too, 400 ppm were sufficient to reach 100% mortality of the embryos. Concentrations as low as
200 ppm also resulted in high mortalities above 98%, with the lowest killing rate (98.3%) being
observed under marine conditions (salinity = 31 ppt) and temperatures of 12°C (Tab. 3).
Table 3. Experiments with Peraclean® Ocean in different water qualities. Testorganism: pre-incubated eggs of
Atlantic Herring (Clupea harengus). Values represent average of 3 parallel experiments. Note: Observations were
made after 1, 2, 4, 8, 12, 24, 36, 48 and 72 hours. Values in brackets represent the highest mortality reached at
the end of the experiment.
Test organism
Concentration of
Fertilised eggs of
Time (hrs.) needed to
Water Quality
Peraclean® Ocean
Atlantic Herring
reach 100% Mortality
Clupea harengus
Salinity 13.5ppt
Mortality of
Pre-incubated eggs
Temp. 5°C
Salinity 13.5ppt
Mortality of
Pre-incubated eggs
Temp. 12°C
Salinity 31ppt
Mortality of
Pre-incubated eggs
Temp. 5°C
Salinity 31ppt
(98.3%; 72 h)
Mortality of
Pre-incubated eggs
Temp. 12°C

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Organisms of the zooplankton showed even higher sensitivities. The dosing of only 400 ppm
Peraclean® Ocean resulted nearly instantly in 100% mortality of the test organisms. After a maximum
of 2 hours exposure time, all of the organisms were dead (see Tab. 4).
Table 4. Experiments with Peraclean® Ocean with plankton organisms. Testorganisms: crustaceans from
freshwater and brackish water communities. Values represent average of 3 parallel experiments.
Concentration of
Time (hrs.) needed to
Water Quality
Peraclean® Ocean
reach 100% mortality
Freshwater Plankton
Freshwater, room
Cyclops sp. (Copepod)
Bosmina sp. (Cladocera)
Freshwater, room
Daphnia sp. (Cladocera)
Freshwater, room
In situ Plankton Baltic Sea
Brackish water,
(wild catch)
about 13 ppt Sal.
Copepods (30% of taxa)
room temperature
< 1
< 1
Nauplii (66% of taxa)
< 1
< 1
Cladocera (4% of taxa)
< 1
Experiments with phytoplankton cultures (indicator organism: Chlorella sp.) showed similar results:
even 200 ppm Peraclean® Ocean killed the algae within 48 hours (See Table 5). However, higher
concentrations of Peraclean® Ocean (concentration range from 400 ppm to 1600 ppm) did not result in
significantly faster eradication of the algae.
Table 5. Experiments with algae. Testorganism: Chlorella sp.. Parameter: photometric measurement of extinction
at 3 different wave lengths: 750 nm, 663 nm and 645 nm. The following results represent the average of three
parallel experiments each.
Water quality
Concentration of
Time needed to
Peraclean® Ocean
reach 100% mortality
Chlorella sp.
Salinity: 31 ppt
a and b
1 200
1 600
Efficacy tests ­ ship board trial
A ship board trial was organized from Maritime Solutions Inc. at the harbour of Baltimore, USA. On
the vessel "CAPE MAY", a ship with roughly 30,000 dwt and 10,000 tons ballast water capacity. A
field trial was done during summer 2001.

Fuchs: Peraclean® Ocean ­ A potentially environmentally friendly and effective treatment option for ballast water
50 ­ 400 ppm of Peraclean® Ocean without any pre-separation of organisms or solids was dosed into
ballast water (water out of the harbour of Baltimore) that went into the ship`s ballast tanks and into
plastic containers. See Table 6.
Peraclean® Ocean effectively killed:
· Copepod Adults, Copepod Nauplii and Nematodes at 50 ppm Peraclean® Ocean concentration
· Polychaetes, Bivalves, Rotifiers and Nematodes at 100 ppm Peraclean® Ocean concentration
· Ostracods and Protozoans at 200 ppm Peraclean® Ocean concentration.
Table 6. Ship board trials: treatment with Peraclean® Ocean, without any pre-separation of species or solids
Mortality of
Mortality [%] of treated groups
in different tanks
Concentration of
Plastic tank
Ship`s Ballast
Peraclean® Ocean
(Mesocosm tank)
Copepod Adults
Copepod Nauplii
0-NF a)
NF a)
NF a)
(a) Values of different control groups; highest and lowest numbers are given.
(b) NF = not found.
The results of all the experiments indicate that Peraclean® Ocean is potentially an effective biocide for
the treatment of ship`s ballast water. 100% mortality of different test organisms from different trophic
levels were found at Peraclean® Ocean concentrations between 50 ppm and 400 ppm.
The short half-life of Peraclean® Ocean in seawater indicates that even the discharge of great
quantities of ballast water in sheltered areas with limited water exchange (e.g. harbours and bays) may
not have a negative impact on the environment. Furthermore, the physical properties of Peraclean®
Ocean (easy storage and long shelf-life) favour both, on board and land based ballast water treatments
as a stand-alone method, or in combination with filtration and/or gravity separation.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
A lower dosage of Peraclean® Ocean could be sufficient if a separation of solids and bigger organisms
takes place before Peraclean® Ocean is applied.
Baldry, M.G.C. 1983. The bactericidal, fungicidal and sporicidal properties of hydrogen peroxide and
peracetic acid. J. Appl. Bact. 54, pp. 417-423.
Block, S.S. 1991. Disinfection, sterilization, and preservation. 4th ed., Chapter 9, peroxygen
compounds. Lea & Felbinger, Philadelphia PA, pp. 167-181.
Gollasch, S. 1997. Removal of barriers to the effective implementation of ballast water control and
management measures in developing countries.
Desk study carried out for GEF/IMO/UNDP, 187 pp.
See also:
Grenman, D., Mullen, K., Parmar, S. & Friese C. 1997. Ballast water treatment systems: A feasibility
. 61 pp.
From the internet: http:/www.ansc.purdue.edu/signis/publicat/papers/.
Schliesser, T.H. & Wiest, J.M. 1979. Catalase-mediated redox reactions. In: Doyer PB (ed) The
Enzymes vol XIII Oxydation-Reduction Part C
3rd ed. Academic press, New York NY, pp. 389-395.
Voigt, M. 1999. Treatment Methods for Ballast Water. 2nd Annual Conference "Managing
environmental risks in the maritime industry", London, 30 Sept-1 Oct 1999.

Acrolein as a potential treatment alternative for control
of microorganisms in ballast tanks: five day sea trial
J. E. Penkala, M. Law & J. Cowan
Baker Petrolite Corporation
Ballast water discharge by marine vessels at destination ports poses serious health, economical and
ecological repercussions due to the introduction of non-indigenous nuisance organisms into new
environments. Current attempts to mitigate this problem via ballast water exchange programs have
been marginally effective (75% removal of organisms at best) and restricted by ship safety limits. An
alternative chemical strategy being investigated is acrolein, a broad spectrum biocide with proven
efficacy against bacteria, algae, and other microorganisms. Extensive toxicity testing has
demonstrated its effectiveness against macroorganisms as well, including molluscs, crustaceans, fish,
and aquatic plants. Recent laboratory studies demonstrated that 1-3 ppm of acrolein can effectively
control various marine microorganisms.

Based on these findings, a sea trial was conducted on board an 8000 MT DWT container vessel
during a 5 day voyage from Venezuela to Florida. Dedicated ballast tanks were treated with 1, 3, 9,
or 15 ppm of acrolein during ballast intake in Venezuela. Monitoring of viable bacteria and acrolein
residuals was conducted prior to treating, daily during the voyage, and during discharge. When
applied at treatment concentrations of 9 ppm, acrolein maintained 99.99% efficacy for 2 days. At 15
ppm, acrolein was shown to be 99.9999% effective for 3 days as compared to untreated ballast tanks.
En route monitoring confirmed that regrowth of microorganisms was minimized when the acrolein
residual was maintained at >2 ppm. At the time of discharge, the acrolein residuals were zero ppm, a
consequence of its reaction with water, thus allowing its safe discharge overboard. These findings
indicate that the use of acrolein can be an effective treatment strategy which can be managed safely,
can be safely discharged into the marine environment, and can be economical in the control of
organisms in ballast water.

Biopollution via ballast water uptake and discharge
The introduction of invasive marine species into non-native environments via ballast water discharge
by marine vessels poses a serious threat in the form of biopollution (Shipley, et al., 1995; Fuller et al.,
1999). Shipping transfers approximately 3 to 5 billion tons of ballast water internationally each year
and possibly a similar volume domestically within countries and regions (NRC, 1996). Ballast water
provides an essential function in the transport of cargo by ships in maintaining stability and
efficiency. At the same time it serves as a vector for the transfer of non-indigenous organisms,
resulting in a potential threat to the ecology, economics, and even health of a particular region.
It is estimated that globally at least 7,000 different species are being ferried in the ballast tanks NRC,
1996). However, the majority of these species do not survive during transit due to the trauma of
ballasting and deballasting, the incompatibility with the ballast tank environment, and inability to
adapt to the new environment upon discharge. Occasionally, however, a species is able to survive
these conditions and establish a reproductive population in the host environment. In that case, the
consequences can be devastating.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
There are several key examples of the impact of aquatic invasive species on the host environment. In
the USA, the European Zebra Mussel Dreissena polymorpha has infested over 40% of the internal
waterways (Nalepa and Schloesser, 1993). It is estimated that between 1989 and 2000 approximately
$750 million to $1 billion was spent on control measures. In southern Australia, the Asian kelp
Undaria pinnatifida is invading new areas rapidly, displacing the native seabed communities
(Pughuic, 2002). In the Black Sea, the filter-feeding North American jellyfish Mnemiopsis leidyi has
proliferated and in some cases reached densities of 1kg of biomass/m2 (Gollasch, 1998)) It has
decimated native plankton stocks thereby seriously impacting the Black Sea commercial fisheries. In
several countries, the `red tide' algae (toxic dinoflagellates) have been introduced. This type of algae
is responsible for fish kills and when absorbed by filter-feeding shellfish can cause paralysis and even
death in humans who consume the shellfish.
In Texas, it is widely suspected that the brown mussel, Perna perna, which invaded rocky intertidal
surfaces in South Texas, originated from ballast water discharge in the Port Aransas area (McGrath et
al., 1998; Hicks et al., 2000, 2001). Two mussels were found in 1990 at Port Aransas. Within 3 years
they had spread nearly 100 miles south and 50 miles north of Port Aransas. It is estimated that the
Galveston Bay System currently has at least 10 introduced species of fish, most of which originated in
subtropical or tropical environs (Fuller et al., 1999). More recently, an invasive species of encrusting
tunicate, identified as Didemnum perlucidum, has been found covering active petroleum production
platforms as well as decommissioned platforms being used as artificial reefs offshore Galveston
(Harper, 2002, personal communication). Because the epicenter of the tunicate invasion appears to be
very near shipping lanes, it is suspected that ballast water is the source of this tunicate.
Recently, the pathogen responsible for cholera (Vibrio cholerae) has been detected in various ports,
including Chesapeake Bay, off the coast of Mobile, Alabama and the Galveston, Texas (EPA, 2000;
DePaulo et al. 1992; McCarthy et al., 1992). Certain strains have been tracked to Latin American port
waters. It has been determined that one third of the ballast discharge in these Gulf Coast regions
contain V. cholerae.
The examples of nuisance invasive species are numerous and these present serious problems to the
health, economy, and/or ecology of a locale. It is imminent that effective control measures be
implemented to mitigate this threat of biopollution.
Treatment options
The only method currently available to reduce the risk of transfer of harmful aquatic organisms is
ballast water exchange at sea. This method is being recommended by the International Maritime
Organization (IMO) in their pending guidelines on ballast water management (Pughuic, 2002;
Gollasch, 1998). But the protocol is subject to serious ship safety limits and even when fully
implemented, this technique is at most 75% effective. Some reports suggest that reballasting at sea
may itself contribute to the wider dispersal of harmful species, and that island states located `down-
stream' of mid-ocean reballasting areas may be at particular risk from this practice.
It is therefore important that alternative, effective ballast water management and/or treatment methods
are developed as soon as possible, to replace reballasting at sea.
Options being considered include: 1) mechanical treatment methods such as filtration and separation,
2) physical treatment methods such as sterilization by ozone, ultra-violet light, deoxygenation, electric
currents and heat treatment, and 3) chemical treatment methods such as adding biocides to ballast
water to kill organisms.
Although research efforts are being focused on developing such methods, there are certain difficulties
being encountered in 1) accommodating the dynamics of the ballasting schedules, 2) customizing to
the ships' ballast system and operations, 3) effectively treating ballast tanks which have a great deal of
internal structures and contain sediment that harbors and protects the resident organisms, and 4)
expanding the treatment program from lab scale to a magnitude that can deal with the large quantities
of ballast water carried by ships such as tankers. Ultimately, a successful treatment must be

Penkala: Acrolein as a potential treatment alternative for control of microorganisms in ballast tanks: five day sea trial
biologically effective, economically feasible, safe to ship personnel, environmentally acceptable, and
simple to implement.
Acrolein technology
We have been investigating the use of acrolein, an organic biocide, as a treatment option for ballast
water. Marketed by Baker Petrolite Corporation as MAGNACIDE® B Microbiocide, acrolein is used
extensively in the oilfield as a biocide to mitigate bacteria in produced fluids. It is a wide spectrum
biocide that is extremely effective against aerobic microorganisms as well as sulfate reducing
bacteria. In addition this chemical is also marketed as MAGNACIDE® H Herbicide, which is widely
used in irrigation canals throughout the world as an aquatic herbicide to control submerged plants and
algae that can impede water flow.
Acrolein is a small 3-carbon molecule that is highly reactive chemically. The molecule is
a vinyl aldehyde and the reactivity is due to its carbon-carbon double bond conjugated
with the double bond of an aldehyde carbonyl.
The biocidal efficacy of acrolein at low dosages appears to stem from its ability to inhibit several
enzyme systems within the living cell and the denaturing of proteins. This high degree of activity at
low concentrations makes it a very good candidate for the ballast water treatment program.
As a ballast tank treatment alternative, acrolein exhibits several key advantages. It is an extremely
potent biocide against not only bacteria, but algae, mollusks, crustaceans, fish, and aquatic plants.
Laboratory studies have shown an effective concentration of 1-3 ppm when tested against marine
microorganisms (results reported in this paper). At the same time, acrolein has a relatively short half
life, 8-24 hours, in water due to a hydration reaction. The reaction products rapidly break down into
carbon dioxide (CO2) and water. Acrolein can be easily neutralized by reaction with ammonium
bisulfite or soda ash. In addition, acrolein can be safely applied by injecting directly into the ballast
line during ballast tank filling. The application involves no major capital expenditure for equipment or
installation. Furthermore, this product is non-corrosive and compatible with the common epoxy-based
linings used in ballast tanks (Mills, 2002, personal communication). These features render acrolein as
a highly effective and economic treatment option that can be safely applied, is relatively inexpensive,
and due to rapid degradation poses no threat to the marine environment during discharge.
Aim of study
In the current study, the authors undertook to investigate the efficacy of acrolein as a potential ballast
water treatment alternative. To accomplish this, a compilation was made of toxicity data evaluating
the impact of acrolein on a range of freshwater and marine aquatic macroorganisms. The results
indicated the biocidal efficacy of acrolein is achieved at a low (<1 ppm) concentration. The second
phase of the study involved laboratory experiments examining the biocidal efficacy of acrolein against
marine microorganisms under conditions mimicking exposure time in ballast tanks during a voyage. It
was determined that a concentration of 3 ppm would effectively control all species tested. In the third
and final phase of this study, a sea trial was conducted using acrolein for ballast treatment on board an
8000 MT DWT container vessel. Nine ballast tanks were tested with acrolein concentrations ranging
from 0 to 15 ppm. The ballast water was treated inline during ballast uptake by injecting the chemical
at the suction side of the ballast pumps. Comprehensive monitoring of chemical residual and viable

2nd International Ballast Water Treatment R&D Symposium: Proceedings
microorganisms was conducted on the water during uptake and discharge and daily during the voyage.
The results are described in the following report.
Laboratory studies
Efficacy of acrolein against aquatic organisms
As an initial step in evaluating the efficacy of acrolein for control of aquatic organisms in ballast
tanks, a review was made of the existing toxicity and environmental tests conducted to date on this
product over a range of various aquatic species. These aquatic toxicity tests were standard flow-
through tests determining the LC50 or EC50 for organisms that included molluscs, crustaceans, fish,
and aquatic plants. Both marine and freshwater organisms had been tested. The compiled data are
presented in Table 1. The greatest sensitivity to acrolein was observed with freshwater fish and
daphnia: 0.022-0.024 ppm. The aquatic plants, including diatoms, green and blue green algae ranged
from 0.034 ppm for the marine diatom (Skeletonema costatum) to 0.072 ppm for duckweed (Lemna
). The highest tolerance was detected in the marine invertebrates and fish, ranging from
0.180 ppm for the eastern oyster to 0.570 ppm for the sheepshead minnow.
Table 1. Toxicity of acrolein to various aquatic organisms
Acrolein Concentration
Organism Tested
Marine Mollusk: Eastern Oyster (Crassostrea virginica)
96-hr EC50
Marine Crustacean: Mysid Shrimp (Mysidopsis bahia)
96-hr EC50
Marine Fish: Sheepshead Minnow (Cyprinodon variegates)
96-hr EC50
Freshwater Fish: Rainbow Trout (Oncorhynchus mykiss)
96-hr EC50
Freshwater: Bluegill Sunfish (Lepomis macrochirus)
96-hr EC50
Freshwater Crustacean: Water flea (Daphnia magna)
48-hr LC50
Duckweed (Lemna gibba)
14-day EC50
Green Algae (Selenastrum capriconutum)
5-day EC50
Freshwater Diatom (Navicula pellicosa)
5-day EC50
Bluegreen Algae (Anabaena flos-aquae)
5-day EC50
Marine Diatom (Skeletonema costatum)
5-day EC50
These data indicated that acrolein has the potential as an effective treatment for ballast water
organisms in low effective concentrations such that the treatment would be economical and the
chemical could be neutralized by the end of the voyage via its hydration reaction with water,
rendering it safe for discharge.
Efficacy of acrolein against marine microorganisms
Next a series of experiments were conducted to evaluate the efficacy of acrolein against common
marine microorganisms. The first experiment tested the response of four common marine bacterial
strains to acrolein: 1) Pseudomonas fluorescens, a Gram negative, non-sporulating bacterium, 2)
Bacillus cereus a Gram positive, spore-forming bacteria, 3) Bacillus subtilis, a Gram positive, spore-
forming bacteria, and 4) Staphylococcus epidermidis, a Gram positive, non-sporulating bacterium.
Each isolate was streaked and revived on nutrient agar, then transferred to nutrient broth and
maintained until the time of the efficacy test. The culture conditions were 3.5% total dissolved solids
(TDS) and 30°C incubation temperature. Acrolein was tested at 0, 1, 3, and 10 ppm for contact times
of 24 and 72 hours to mimic conditions of a short voyage. Following the contact time, a serial dilution
was made into nutrient both to enumerate the number of surviving bacteria. The results (Figure 1)
indicate that significant reductions in bacterial number occurred at all acrolein concentrations tested.
The control organisms (0 ppm) exhibited at least 106 bacteria per ml for all strains tested. At 10 ppm
acrolein no greater than 101 bacterial per ml were observed for any strain at either 24 or 72 hours

Penkala: Acrolein as a potential treatment alternative for control of microorganisms in ballast tanks: five day sea trial
contact (>99.999% reduction). At 3 ppm acrolein, no greater than 102 bacteria per ml were detected
(>99.99% reduction).
In a second experiment, heterogeneous cultures of general aerobic and facultative anaerobic bacteria
(GAB) and sulfate reducing bacteria (SRB) obtained from seawater in the Gulf of Mexico at
Galveston, Texas were tested to model populations that might be encountered in port water used for
ballast. Cultures were maintained in modified formulations of phenol red dextrose (for GAB) and
Postgate's broth (for SRB) at ambient temperatures (17-23°C) at 3.5% TDS. The test conditions were
identical to those in the first experiment: 0, 1, 3, and 10 ppm of acrolein and 24 and 72 hour contact
times. Enumeration was performed by serial dilution into the appropriate culture media. The results
(Figure 1) show no detectable growth of GAB or SRB at 10 ppm of acrolein, an 11 order of
magnitude reduction. At 3 ppm acrolein limited SRB growth to 102/ml at 24 hour contact and to
below detection with 72 hour contact compared to 1011/ml in the controls. At 3 ppm, GAB growth
was limited to 103/ml after 24 hours contact and 101/ml after 72 hours contact compared to 1011/ml in
the controls. These results show that acrolein is effective at 3 and 10 ppm for control of
microorganisms in Galveston Port water.
A third experiment was conducted using the marine dinoflagellate, Gymnodinium sanguineum, the red
tide former associated with fish and jellyfish mortality. The dinoflagellate was cultured in L1 medium
at ambient temperature (17-23°C) under a 12L/12D photoperiod. For the efficacy test, the cells were
transferred to filter-sterilized Galveston seawater and appropriate concentrations of acrolein (0, 1, 3,
and 10 ppm) were added for 24 and 72 hours contact. Surviving dinoflagellates were enumerated by
serial dilution into L1 media. In addition, light microscopy observations were made on the diluted
organisms to verify motility and viability. These results are presented in Table 2 and Figure 2.
Table 2. Efficacy of acrolein against the marine dinoflagellate Gymnodinium sanguineum
24 hour Contact Time
72 Hour Contact Time
Mean No.
% Reduction
Mean No.
% Reduction
Inact Motile
Inact Motile
from Control
from Control
The results indicate that all concentrations of acrolein were able to reduce the concentration of viable
dinoflagellates to below the detectable limit of the assay. No viable or motile dinoflagellates were
observed in any of the acrolein-treated samples. The integrity of the dinoflagellate cell was
completely destroyed by the treatment with acrolein as seen in Figure 2.
On-board ballast trial
Overview of sea trial
Based on toxicity data and laboratory studies presented in the previous sections, it was concluded that
acrolein would be a good candidate for chemical control of ballast water. The information was
presented to shipping companies and an opportunity for a test voyage was obtained. The sea trial was
undertaken aboard an 8000 MT DWT container vessel sailing from Guanta, Venezuela to Panama
City, Florida. The trial took place from November 4-10, 2002. Ballast uptake, pre-treatment sampling
and chemical treatment were conducted the evening of November 4. The ship set sail the morning of
November 5. Ballast discharge and post-treatment sampling took place on the evening of November 9
at 100 miles offshore Panama City, Florida. This discharge procedure was conducted in accordance
with the US Coast Guard. The ship arrived at the destination port on November 10.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Ballast system, chemical application, and sampling protocol
Ten ballast tanks were selected (5 pairs) so that discharge and some of the tank filling could be done
on parallel tanks. The ballast system and treatment design are illustrated in Figure 3. The ballast water
enters the ship via a single line and then passes through a 5 mm mesh filter. It then separates into two
parallel lines each feeding a charge pump. The normal operating rate of each pump is 250 m3/hour.
The line pressure was approximately 15 psi. The two lines then converge downstream of the pumps to
a common line that transports ballast water to the parallel ballast tanks.
The application and sampling points were set up on each of the parallel lines feeding the charge
pumps. Acrolein was injected into the line on the suction side of one pump and water samples were
obtained on the parallel line on the discharge side of the second pump. In that way, acrolein treatment
and sampling could be carried out simultaneously. Both the acrolein cylinder and the sample drum
were placed on the weather deck, and chemical/sample lines were run down to the ballast room via an
escape hatch. This facilitated operations and optimized both safety (the acrolein could be stored
topside, secluded from all personnel) and convenience (once the sample water was filtered for
collection of the target organisms, it could be dumped over the side of the ship). The acrolein was
delivered from a 58 lb (net weight) cylinder via a standardized BPC manifold using nitrogen pressure.
Chemical volumes were metered using a Sponsler digital flowmeter at a rate to achieve maximum
chemical injection time during ballast tank filling. Injection rates varied between 114 ml/min and
280 ml/min. The minimum metering rate for the flowmeter was reported to be 40 ml/min although
only 80-90 ml/min at best was achievable during the test. Each ballast tank was filled to capacity with
tank volumes ranging from 125 to 340 m3. The tanks were confirmed to be at full capacity by
detecting the level at the sounding tubes. Ballast water injection rates averaged 200 m3/hr. For
accurate monitoring of sample volume, sample collection was conducted using an industrial hose
fitted with a Great Plains Industry (GPI) flowmeter (maximum flow rate of 20 gal/min).
Untreated control tanks were filled first in order to ensure that the ballasting operation, flow rates and
sampling were proceeding properly. The first treated tank (Wing Tank #1 - Port) received 3 ppm (v/v)
of acrolein. The parallel starboard tank (Wing Tank #1 ­ Starboard) received 1 ppm of acrolein. After
measuring residuals in these tanks using differential pulse polarography (DPP; see Appendix for
description of this method) it was determined that the acrolein in these two tanks had been
immediately consumed as no residual was detectable. Therefore adjustments were made and
subsequent tanks were treated with 15 ppm acrolein (Double Bottom #2 Port and Double Bottom #2
Starboard) and 9 ppm of acrolein (Double Bottom #1 Port and Double Bottom #1 Starboard).
On the evening of November 9, ballast discharge and then reballasting were conducted. At the time of
discharge, the chemical residuals in the treated tanks were below the 10 ppb detection limit of the
DPP. Sampling was conducted for acrolein residuals and microbiological specimens during discharge.
However, additional planned tests were aborted due to the presence of a severe tropical storm.
It should be noted that this study was originally designed to also obtain data on a) ballast sediment, b)
ballast tank residual water before filling, and c) macroorganisms, including ichthyoplankton,
zooplankton, and phytoplankton. However, due to logistical complications with the ship's schedule
and inclement weather during discharge, these studies could not be completed during this sea trial.
Therefore, the partial data on macroorganisms will not be included. Preparations are being made for a
more comprehensive analysis on a second voyage which is currently being planned.
Analysis of microorganisms and acrolein residuals
Water samples were collected by delivery topside via a 1/2 inch industrial hose. Three 100 ml samples
were obtained in triplicate during the filling operation of the ballast tanks approximately 15 minutes
apart. These samples were immediately diluted into culture bottles and parallel samples were fixed for
microscopy. This procedure was used during ballast uptake as well as discharge. The discharge
operation was a reverse of the ballast tank filling, using the same sample port and line. One tank was
omitted during this procedure (Double Bottom #3 Starboard). This was one of the four untreated tanks

Penkala: Acrolein as a potential treatment alternative for control of microorganisms in ballast tanks: five day sea trial
and was not sampled in order to save time, since the ship had to remain stationary (engines off) during
this procedure.
During the voyage, each of the test ballast tanks were sampled daily via the sounding tubes, using a
siphon tube with hand-operated pump. The siphon tube was attached to a plumb line to deliver it near
the bottom of the ballast tank. Each tank was sampled in triplicate and tested for acrolein residuals
and bacterial cultures.
A detailed description of the bacterial enumeration methods and measurement of chemical residuals is
included in the Appendix at the end of this report.
Results of sea trial
Physical data on Port Guanta Water used for ballast uptake
At the time of ballast uptake, the water in port was calm, the sky was 100% overcast with a wind at
5 kph and an air temperature of 28.3°C. The port water exhibited a dark green color. Vertical visibility
through the water column was estimated at 1.3 m whereas total water depth was 11.1m
Hydrographic data are shown in Table 3. There was a 2.5°C decrease from surface to bottom and a
1.3 ppm decrease in dissolved oxygen from surface to bottom.
Table 3. Hydrographic characteristics of water column in Port Guanta, Venezuela prior to ballast uptake
Depth in Water Column
Temperature (°C)
Salinity (ppt)
Dissolved Oxygen (ppm)
Results of bacterial monitoring
General aerobic and facultative anaerobic bacteria (GAB) were monitored by serial dilution into the
appropriate culture medium. As seen in Table 4, The GAB concentrations in the ballast uptake
samples ranged from 4-5 log10 GAB/ml for each of the 10 tanks. The average concentration for all
samples collected was 4.3 x 104 GAB/ml. These concentrations are within the range of what is
typically encountered in seawater samples (unpublished observations).
Table 4. Average Log10 Number of GAB/ml in Ballast Water
4-5 Nov
Ballast Tank
Concentration of
Acrolein (ppm)
DB#3 Port
DB#3 Star
DB#4 Port
DB#4 Star
DB#5 Star
1 ppm
DB#5 Port
3 ppm
DB#1 Port
9 ppm
DB#1 Star
9 ppm
DB#2 Port
15 ppm
DB#2 Star
15 ppm

2nd International Ballast Water Treatment R&D Symposium: Proceedings
In the four control tanks, the GAB numbers increased dramatically from 104 to 108 GAB/ml within 24
hours to 36 hours (November 6th) after filling the tanks (Table 4). On November 7th, 48 hours after
filling the tanks, the same concentration was encountered in all control samples. It is important to note
that on November 6th and 7th, only an 8 log10 serial dilution was used, not anticipating growth beyond
this limit. However, the maximum number of bottles had turned positive, so that the concentration
measured was at least 108 GAB/ml, possibly more. For subsequent sampling on November 8th -10th
(72 hours to 120 hours) a 12-bottle serial dilution was used. For each of the subsequent time points,
samples collected from the control tanks showed positive cultures in all 12 bottles, indicating that the
GAB concentrations tanks were greater than or equal to 1012/mL.
In Wing Tanks #5 Port and Starboard, treatments of 3 ppm and 1 ppm of acrolein were used,
respectively. The GAB concentrations were the same as reported for the untreated tanks described
above (Table 4). On November 6th and 7th, the GAB concentrations were greater than or equal to 108
GAB/ml. On November 8th-10th, the GAB concentrations were greater than or equal to 1012 GAB/ml.
Therefore, it can be concluded that the acrolein treatment applied had no impact on the concentration
of GAB in these tanks. This finding is not surprising, since chemical monitoring revealed no residual
acrolein in the tanks immediately after treatment (See Results of Acrolein Residual Monitoring and
Figure 6).
The Double Bottom Tanks #1 Port and Starboard were both treated with 9 ppm acrolein. On
November 6th, a reduction in GAB concentration to approximately 106 GAB/ml was achieved as
compared to the control tanks (Table 4 and Figure 4). However, this still represented a 2 log10 increase
over the intake water. Since approximately 10% of the tank volume remained following discharge and
prior to uptake and chemical treatment, it is presumed that bacteria in this residual ballast water
contributed to the rapid increase in GAB seen the day after filling. On November 7th, the GAB
concentration had increased to approximately 107 GAB/ml. This was still less than the controls but
steadily increasing as the acrolein residuals decreased to less than 1 ppm.
The Double Bottom Tanks #2 Port and Starboard were each treated with 15 ppm acrolein. On
November 6th, the GAB had decreased to well below the concentration in the uptake water: less than
102 GAB/ml (Table 4 and Figure 5). This represents an approximate 99.9% reduction of the bacteria
in the uptake water. Furthermore, this represents greater than 99.9999% decrease in GAB compared to
the untreated tanks. On the November 7th, the GAB concentration was slightly greater than 104
GAB/ml, approximately the same concentration as seen in the uptake water. However, this represents
a greater than 99.99% GAB reduction as compared to the untreated tanks.
Sulfate reducing bacteria (SRB) were enumerated in the water samples by serial dilution into
appropriate culture medium. No SRB were observed in the port water during uptake to fill the ballast
On November 6th (24 ­ 36 hours) only Double Bottom Tank #3 Starboard (control) showed positive
cultures of SRB (Table 5). Until November 8th (72 hours), no SRB were detected in any of the treated
tanks. At this time point, three tanks, one control, one at 1 ppm, and one at 15 ppm were positive for
SRB. The maximum number of SRB detected (102/mL) was observed in Double Bottom Tank #3
Starboard. At the time of ballast discharge (November 10th), seven of the nine tanks discharged for the
test contained SRB. Concentrations ranged up to 103 SRB/ml in Double Bottom Tank #2 Starboard
(15 ppm).
The contamination by SRB may be due to residual populations that have become established in the
tanks over time.

Penkala: Acrolein as a potential treatment alternative for control of microorganisms in ballast tanks: five day sea trial
Table 5. Average Log10 Number of SRB/ml in Ballast Water
4-5 Nov
Ballast Tank
Concentration of
Acrolein (ppm)
DB#3 Port
DB#3 Star
DB#4 Port
DB#4 Star
DB#5 Star
DB#5 Port
DB#1 Port
DB#1 Star
DB#2 Port
DB#2 Star
Results of acrolein residual monitoring
Immediately after filling the tanks, residuals were measured in the sounding tubes for two of the
ballast tanks, those targeted for 1 ppm, 3 ppm acrolein. No residual was detected in either the 1 ppm
or 3 ppm tanks (Figure 6). Accurate sampling of the port and starboard tanks targeted for 9 and 15
ppm acrolein was not possible due to ballasting operations, deck activity, and the schedule of getting
On November 6th (24 hours) the Double Bottom Tanks #1 Port and Starboard (9 ppm) exhibited a
residual of 0.5 ppm and 2.5 ppm, respectively. By November 7th (48 hours), the residuals in both
tanks were less than 1 ppm and analysis on November 8th showed residuals less than 0.3 ppm. No
residuals were detected in either tank on November 9th (96 hours).
Double Bottom Tanks #2 Port and Starboard (15 ppm) exhibited acrolein concentrations of 4 ppm at
the 24 hour time point (November 6th), 2 ppm to 3 ppm on November 7th and less than 0.5 ppm on
November 8th. On November 9th (96 hours) no residual was detected in these tanks.
Persistence of acrolein residuals in ballast tanks during sea trial
The design for treatment of the water in the ballast tanks of the ship was for 1 ppm and 3 ppm
treatments. These treatment concentrations were chosen based on a series of laboratory efficacy
studies conducted on bacteria cultured from Galveston seawater and bacteria and dinoflagellates
obtained from various culture suppliers as well as historical toxicity testing conducted against
mollusks and other aquatic organisms. In addition, the proposed treatment concentrations would limit
the need for neutralization of acrolein prior to discharge due to the hydration of acrolein during a 4-5
day voyage.
During treatment of the tanks, it was discovered that immediately following treatment at 1 ppm and 3
ppm, no residual was detected. This was confirmed with samples taken approximately 30 hours later.
Furthermore, tanks treated with 9 ppm had substantially decreased concentrations (less than or equal
to 2.5 ppm) within 24 hours. Finally, the tanks treated with 15 ppm of acrolein also had decreased
within 24 hours to only 4 ppm. This degradation rate of acrolein is quite rapid. Typically in oilfield
produced waters and canal irrigation waters we have observed half-life ranges from 8-24 hours
depending on the water being treated. In laboratory studies comparing artificial brine to Galveston
Port water we observed half-lives ranging from 20-25 hours for concentrations of 1, 5, and 10 ppm
acrolein. A test of water collected from another Venezuelan port (Puerto Cabello) showed an average
half-life of half-life of 52 hours for acrolein concentrations ranging from 1-10 ppm. Thus in this study

2nd International Ballast Water Treatment R&D Symposium: Proceedings
the rate of degradation was much steeper, indicating components in the ballast water and tanks
contributed to this.
Bacterial control appeared to be immediately diminished in all tanks once the acrolein concentration
decreased below 0.5 ppm. Unfortunately, all tanks were below this concentration threshold by
November 8th, 48 hours before termination of the project and discharge of the ballast tanks. Based on
these findings, a higher concentration of acrolein, possibly 25 ppm or 50 ppm should be considered in
future tests.
In future studies it would be best to pre-test the water for acrolein demand and check various
parameters such as sulfides and pH in the residual water left in the ballast tanks as well as the uptake
water prior to initiating treatment. Although the current study had been designed to examine these
parameters, logistical considerations and shipboard operations interrupted this part of the study.
Several factors have been identified as potentially imposing an unexpected demand on acrolein in the
treated tanks. Although SRB were not detected in the uptake water, they were cultured from the tanks
while en route to Florida. Sulfate reducing bacteria produce soluble sulfide as a result of their
respiration process. Acrolein is highly reactive with sulfides and is consumed at a 2:1 molar ratio (3.7
ppm acrolein: 1 ppm sulfide). It is presumed that the SRB recovered were established in the 10%
residual water left in each tank prior to treatment. If any soluble sulfide was in this residual water, it
would have quickly reacted with acrolein resulting in decreased residuals. Another source of demand
on the acrolein may have been organics, solids and other bacteria resident in the residual water and
sediment in the tank bottoms. Unfortunately, we did not have the option to examine the tanks for
sediments, bacteria or sulfides before treatment. This information might have given a more complete
picture of those factors contributing to acrolein demand.
The ideal treatment concentration cannot be predicted from this study. It is a balance between having
a sufficiently high concentration of acrolein to maintain adequate control of marine organisms and
ensure an adequate residual throughout the voyage while having the concentration low enough so that
the application is economical and neutralization/ discharge issues are minimized. Based on this study,
a suggested concentration range for acrolein treatment during a 5 day voyage might be 15 to 50 ppm.
Efficacy of acrolein treatment in control of bacteria in ballast tanks
Bacterial control is believed to be the most rigorous test of efficacy of biocide treatment in the ballast
tanks. They are the most adaptive organisms, with highly evolved protective mechanisms. Their very
short life cycle, as little as 20 minutes in some cases, allows for rapid proliferation. Their sessile
communities in tank sediments and biofilms represent a difficult challenge for any biocide. In
essence, they are the toughest organisms to control.
Although the GAB levels in the port water during ballast filling were 104 to 105 GAB/ml, within 24-
36 hrs the concentrations in the controls, 1 ppm and 3 ppm-treated tanks were greater than or equal to
108/mL using a 8 bottle dilution series. This may have been due to rapid growth within the tank
environment or it might have been due to a high initial concentration of GAB in the residual water
residing in the tanks prior to filling. Given that the estimated residual water in each tank was 1/10 of
the total volume, had it contained 109 bacteria/ml, then a 10-fold dilution with incoming ballast water
would only reduce this population to 108/mL. The concentrations of bacteria in the residual water of
the tanks are fairly important and should be accounted for prior to treatment. Not only that, but any
sediment or sludge in the tanks which most likely contain sessile bacterial populations could also
significantly impact the initial concentration in the tanks once they are filled. In the future, it is
important to obtain data on the final concentration of bacteria in the tanks immediately after filling.
Acrolein at 15 ppm had a significant impact on the bacterial load in the tanks as compared to the
controls. This comparison is the more critical one when determining efficacy of biocide, not the
comparison with intake water. If one uses this comparison, then 24 hours after treatment with 15 ppm
acrolein the bacterial levels were reduced by at least 6 log10 units or greater than 99.9999%. The

Penkala: Acrolein as a potential treatment alternative for control of microorganisms in ballast tanks: five day sea trial
residual at that time was approximately 4.0 ppm. On the following day (48 hours), when the residual
had decreased to approximately 2 ppm, there was still at least a 6 log10 reduction in the number of
bacteria in the tank.
Acrolein applied at 9 ppm had a lesser impact on the bacterial load in the tanks as compared to the
controls. However, it is important to review what the actual biocide residual is at the time the readings
are being made. After 24 hours, the residual was 0.5 ppm in the Port tank and 3 ppm in the Starboard
tank. The bacterial load at this time point had been reduced by at least 2 log10 units as compared to the
control, or 99% reduction. However, since the dilution limit had been exceeded in the control tanks,
the maximum reduction at this time point could not be obtained. Not having data on the initial load in
the tanks limits our conclusions. Bacterial control in the 9 ppm-treated tanks becomes further reduced
as the residual decreases. In all cases, the tanks no longer exhibited substantial bacterial control after
the third day, assuming the untreated tanks had concentrations no more than 1012 GAB/ml (the
maximum detection limit for this test).
The other class of bacteria examined was sulfate reducing bacteria (SRB). These organisms reduce
sulfate to sulfide during respiration. They are also notorious for their involvement in
microbiologically influenced corrosion (MIC), which is a concern in ballast tanks. Although none of
these organisms were detected in the port water used to fill the tanks, their growth was detected in the
tanks, presumably due to SRB resident in the residual water and sediments in the tanks. By the end of
the voyage, 7 out of 9 tanks had established planktonic populations of SRB, up to 1000 SRB/ml.
These planktonic populations only became established as the acrolein residuals had become negligible
and bacterial control was lost.
One advantage to acrolein technology is that it works well against sessile bacteria and SRB. If SRB
are becoming established in the tanks of a ship, then the tank integrity is under threat from MIC.
Acrolein is non-corrosive and compatible with common ballast tank coatings (Mills, 2002). Therefore,
it is well suited for mitigation of MIC.
Proposed strategy for future sea trial
At this time a second sea trial is being secured. The objectives of the follow-up sea trial will include:
· To test acrolein at treatment concentrations of 15-50 ppm. This will insure a minimal 2 ppm
residual during the voyage and minimize the potential for bacterial regrowth before the
destination port is reached. Measures will be taken to determine the acrolein demand prior to
application and the residual in all the tanks will be measured immediately after filling
· To premonitor tanks before filling. This should include bacterial analysis of the residual water
in the tanks and an analysis of tank sediments. This should also include measurement of
sulfide levels, pH, and dissolved oxygen.
· To conduct a comprehensive analysis of macro-organisms as well as microorganisms. This
will include studies on zooplankton, phytoplankton, and ichthyoplankton. It will also include
analysis of certain indicator microorganisms such as Pseudomonas fluorescens, Vibrio
, etc.
· To conduct a side by side comparison of the acrolein treatment with ballast water exchange
on the same voyage.
· Acrolein at a concentration of 15 ppm was required to have a significant impact on bacteria
present in the ballast tanks after filling the tanks. A concentration of 9 ppm, exhibited a lesser
degree of effectiveness. Whereas, 1 and 3 ppm acrolein was ineffective.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
· Bacteria populations in the control tanks and those tanks treated with <9 ppm increased from
4-5 log10 to greater than or equal to 8 log10 within 24 hours. This extended to greater than or
equal to 12 log10 after 72 hours.
· At 15 ppm, acrolein controlled the bacteria for at least 48 hours, but regrowth occurred by 72
hours as the acrolein residual had diminished below 0.5 ppm. It is estimated that a minimum
residual of 2 ppm would be required to maintain control. It is suggested that future studies
test a concentration range of 15-50 ppm acrolein for a voyage of comparable duration in order
to maintain this minimal residual and prevent the possibility for regrowth.
· The demand for acrolein in the ballast tanks is high, much higher than what was predicted
from laboratory testing. Applied concentrations of 1 and 3 ppm were immediately
undetectable by the time the tanks had been filled.
· SRB were present in 7 of the 9 ballast tanks tested at the end of the voyage (1-3 log10/mL),
although none were present in the seawater that was used to fill the tanks. These organisms
are key players in MIC of ballast tanks and ballast lines and also produce the hazardous and
explosive gas hydrogen sulfide.
· Overall, the results were encouraging. A very high level of control was maintained by 15 ppm
of acrolein, especially given the high concentrations of bacteria observed in the untreated
tanks. This study supports the feasibility of acrolein as an alternative ballast water treatment
method by showing its potential for high efficacy, safe and simple installation and
application, economic viability, and potential for safe discharge.
DePaulo, A, Capers, G.M., Motes, M.L., Olsvik, O., Fields, P.I., Wells, J. et al. 1992. Isolation of
Latin American epidemic strain of Vibrio cholera O1 from US Gulf Coast. Lancet 339: p. 624.
Fuller, P.L., Nico, L.G., & Williams, J.D. 1999. Nonindigenous fishes introduced into inland waters
of the United States
. American Fisheries Society, Special Publication 27, Bethesda, Maryland.
Gollasch, S. 1998. Removal of Barriers to the Effective Implementation of Ballast Water Control and
Management Measures in Developing Countries
. Report prepared for the IMO MEPC Ballast Water
Working Group. 32 pp. IMO MEPC 41, April 1998.
Hicks, D.W., Hawkins, D.L., & McMahon, R.F. 2000. Salinity tolerance of brown mussel Perna
(L.) from the Gulf of Mexico: an extension of life table analysis to estimate median survival
time in the presence of regressor variables. Journal of Shellfish Research 19(1): pp. 203-212.
Hicks, D.W., Tunnell, J.W, Jr., & McMahon, R.F. 2001. Population dynamics of the nonindigenous
brown mussel, Perna perna (L.), in the Gulf of Mexico compared to other world-wide populations.
Marine Ecology-Progress Series 211: pp. 181-192.
Harper, D. 2002. Texas A&M University Galveston, Personal Communication.
Invasive Species Focus Team, Gulf of Mexico Program. 2000. An Initial Survey of Aquatic Invasive
Species Issues in the Gulf of Mexico
McCarthy SA, McPhearson RM, Guarino AM & Gaines JL 1992. Toxigenic Vibrio cholera O1 and
cargo ships entering Gulf of Mexico. Lancet 339: pp. 624-625.
McGrath, M., Hyde, L., & Tunnell, J.W., Jr. 1998. Occurrence and Distribution of the Invasive
Brown Mussel, Perna perna in Texas Coastal Waters
, Center for Coastal Studies. Texas A&M
University-Corpus Christi, March 1998. GBNEP-49.

Penkala: Acrolein as a potential treatment alternative for control of microorganisms in ballast tanks: five day sea trial
Mills, G. 2002. George Mills & Associates International, Inc., Personal Communication.
Nalepa, T.F. & Schloesser (eds). 1993. Zebra Mussels: Biology, Impacts and Control. Boca Raton,
. Lewis Publishers, Inc.
National Research Council. 1996. Stemming the Tide: Controlling Introductions of Nonindigenous
Species by Ships' Ballast Water
. National Academy Press, Washington D.C.
Pughuic, D. 2003. Ballast Water Management and Control: An Overview. 2001. International
Maritime Organization Publication.
Shipley, F.S., Keilsing, R., Prince, K.A., Bentson, M. A., Lane, W.G., Belk, J.O., & Eemisse, J. 1995.
The Galveston Bay Plan. The Galveston Bay National Estuary Program.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 1. Laboratory studies: acrolein efficacy vs. marine microorganisms.
Figure 2. Effect of acrolein treatment on physical integrity of marine dinoflagellate. Bright field microscopy: 400 X

Penkala: Acrolein as a potential treatment alternative for control of microorganisms in ballast tanks: five day sea trial
Sea Chest
WT 5
Port Side
3 ppm
5 mm mesh
DB 4
DB 3
DB 2
DB 1
0 ppm
0 ppm
15 ppm
9 ppm
DB 4
DB 3
DB 2
DB 1
0 ppm
0 ppm
15 ppm
9 ppm
DB = Double Bottom Tank
WT = Wing Tank

WT 5
1 ppm
ppm = acrolein concentration
Starboard Side
Figure 3. Intermarine Industrial Century: ballast line schematic.
Figure 4. Acrolein residual versus GAB concentration in tanks treated at 9 ppm.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 5. Acrolein residual versus GAB concentration in tanks treated at 15 ppm.
Figure 6. Intermarine ­ Industrial Century BW test: acrolein residual profile.

Penkala: Acrolein as a potential treatment alternative for control of microorganisms in ballast tanks: five day sea trial
Serial dilution cultures for bacterial enumeration
Immediately following collection, the water samples were prepared for semi-quantitative enumeration
of viable SRB and viable GAB using the serial dilution technique. Samples were diluted into 3.5%
TDS modified Postgate's SRB media and 3.5% TDS modified aerobic phenol red dextrose media for
GAB growth (C&S Laboratories, Inc. Broken Arrow, OK). Serial dilutions were performed according
to the NACE Standard Test Method 0194-94 Field Monitoring of Bacterial Growth in Oilfield
. The serially diluted culture vials were then incubated at 28°C for 28 days at which time the
log10 number of bacterial growth for each sample was recorded.
Measurement of acrolein residuals
The most sensitive and accurate field method to date for determining acrolein residuals is differential
pulse polarography (DPP). This method employs an EG&G PARC Model 394 electrochemical trace
analyzer connected to an EG&G PARC Model 303A static mercury drop electrode (SMDE).
DPP analysis allows for the determination of a trace chemical, in this case acrolein, which can be
electrochemically oxidized or reduced (in this case, reduced) in a sample. A potential is applied to a
sample via a conductive electrode. The potential, which serves as the driving force in the experiment,
is scanned over a region of interest. When measuring acrolein residuals, all samples are scanned from
an initial potential of -0.9 V to a final potential of -1.5 V. At a potential of approximately -1.2 V the
acrolein in solution is reduced, producing a current at the working mercury electrode. The magnitude
of current produced is proportional to the concentration of acrolein in the solution.
Prior to monitoring acrolein residuals, a standard curve was generated using acrolein standards of 1.0,
5.0, 10.0, and 25.0 ppm. Analysis of each standard was initiated by energizing the solenoid of the
SMDE. The solenoid allows for the flow of mercury through the capillary of the Model 303A
electrode forming a mercury drop that acts as an electrode with a renewable surface. Measurements of
current are performed on these drops. Uninterrupted flow of mercury is established prior to running
the samples to ensure the removal of air bubbles from the capillary. The current is sampled twice over
the life of each drop; the first sample is taken just before the modulation pulse is applied, the second
just before the pulse ends. The modulation pulse is a staircase ramp that is applied to the sample cell
with a fixed height modulation pulse superimposed on the ramp just before each drop is dislodged.
Each step of the drop lasts one second, and each has a magnitude of 2 mV. The two current
measurements are compared, and the change in current becomes the processed signal. The signal is
plotted as the inverse of the current (1/µA) versus the electron potential over voltage (E/V) resulting
in a peak with amplitude proportional to the concentration of acrolein in solution. The peak heights
obtained for these samples were used to generate a standard curve and had a correlation coefficient
(R2) of 0.993 using the least squares method.

Session 4:
Multiple Technologies
and Combined Systems

Solution to ballast water pollution: ship shape and the
ports escape?
E. Donkers
Port of Rotterdam
The Netherlands
With this paper and accompanying presentation, the Port of Rotterdam explains its view on on-board
treatment and management versus on-shore treatment of ballast water. Next to that, this paper aims to
further outline the possible role of the Port of Rotterdam and other ports view to tackle this problem.
Helicopter-view Port of Rotterdam
With 30,000 ships calling every year and a cargo throughput of some 320 million tonnes, the Port of
Rotterdam is one of the biggest ports in the world. Rotterdam is Europe's most important port for oil
& chemicals, containers, iron ore, coal, food and metals. The port and industrial area covers 10,500
hectares (26,000 acres), stretching out over an area of 40 kilometers, from the center of the city to the
North Sea (see figure 1).
Consequently, significant volumes of ballast water are transported in and out of the port area. It has
been calculated that around 6 million and 60 million tonnes are respectively imported to and exported
from this Port (Aquasense, 1998). This vast amount of exported ballast water can be explained by the
high number of oil tankers that arrive with cargo, and leave free of cargo, "in ballast".
Regional context
Within the port water basin, with "De Nieuwe Waterweg" as its main waterway, powerful tidal
currents take place. Regardless of the distance to the sea, in the entire port there is a strong influence
of fresh water as well as sea water movements.
In Dutch coastal and port waters, only around 15 species can be regarded as non-indigenous, and this
number for the Dutch North Sea is estimated to be 44 (Aquasense, 1998). On the entire North Sea
Area, the estimated number of introduced alien species is 108 (North Sea Foundation, 2001). From
these total numbers, there are various "vectors" of introduction, besides ships' ballast water also bio-
fouling on ships' hulls and aquaculture.
From the studies that are currently available, it can be concluded that no significant negative
ecological or economic effects have taken place yet in the Dutch coastal waters and the Dutch part of
the North Sea, because of introduction of alien species by shipping. The question is whether this is
because this water area is relatively less vulnerable for alien introductions, either we are dealing with
an "ecological time bomb". Another question is if "the open character" of the Rotterdam Port Basin
decreases the opportunities for non-native species to be introduced in the port area or upstream. A
more enclosed port area like San Francisco is known for its massive number of introductions (once
every 14 weeks between 1961 and 1995, Pew Oceans Commission 2001), contrary to the estimated
low numbers in Rotterdam.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The forthcoming IMO Convention
Most likely, an IMO convention for the control of harmful introductions of aquatic organisms by
ships' ballast water discharges will enter into force within a few years. The Port of Rotterdam, and
many other ports, are of the opinion that the solution to prevent alien invasions from ballast water
must be found on board of ships. "Ship Shape" should be the starting point, to tackle this problem at
the source. Swift technical development and upgrading must ensure that effective on-board ballast
water treatment techniques (BWTTs) are developed, without causing unacceptable adverse
environmental effects or safety hazards.
Given the fact that the development and broad use of BWTTs will take quite some years, ballast water
exchange (BWE) will be an allowed interim-option in years to come. In the currently circulating text
of the IMO Convention, the possibility for land-based treatment is left open, primarily for "Certain"
or "Special Areas". The Port of Rotterdam has serious concerns about a strong focus on land-based
treatment. First of all, it is important to highlight the developments on the ship side.
Ship side developments
During its involvement in the EU funded assessment-project "SEAM"1, the Port of Rotterdam has
monitored the development on various on board ballast water management techniques. Primarily
techniques that combine hydrocyclone with a secondary treatment such as UV radiation seem to be
promising and get broad stakeholder support. Also, heat and filtering prove to be viable options.
These signals are reflected in the recent placement of an installation based on hydrocyclone with UV
after-treatment on two vessels of Dutch ship-owner "Wagenborg", as well as the cruise vessel "Regal
Princess". For broader application, bottleneck appears to be the flow rate of the incoming ballast
water of the hydrocyclone. Because of this, application of this technique is currently restricted to
relatively small vessels. Pilot-scale testing and upgrading to bigger flow rates is crucial to make such
techniques suitable for larger vessels such as oil tankers. Technical research and pilot-testing aimed to
increase flow rates is taking place continuously, primarily in the United States.
Besides physical and UV treatment, there is also a possibility to use chemicals. Chemical treatment,
often used in combination with hydrocyclone pre-treatment, may cause serious chemical pollution of
port or sea water. Pollution of port water with ballast water cleaning effluents could have significant
negative environmental impacts. If chemicals such as acids are discharged into port waters, it could by
lowering the pH, increase the amount of "free" heavy metals in the environment. Consequently, the
sludge that is dredged from the port area might become more polluted, in Rotterdam possibly leading
to more sludge that has to be discharged in "De Slufter", a big basin for polluted dredged material.
Also, soluble metals are known to accumulate in the ecosystem and food chain. Some chemically
treated ballast water might also contain chlorine and bromine, which also have potential negative
effects on water and dredged material quality.
It is, however, too premature to exclude chemicals from the "tool-box" of on-board ballast water
management techniques. Additional research needs to be done to decide whether the use of chemicals
could be allowed in certain cases. Besides determining hazard profiles, it is crucial to know the
freights (amounts) and frequency of ballast water discharges, to determine the environmental impact.
Besides this, the principle question is whether the risk of release of alien species is more or less
serious than the discharge of certain chemicals.
Apart from treatment, ballast water exchange (BWE) is an often used option to decrease the amount
of invasive species in ballast tanks. Some BWE-techniques, such as the sequential method, have
potential ship safety hazards, some seem to be acceptable. Because of existing ballast water
regulations in the US, Canada and Australia, BWE has been applied by a large number of ships for
many years.

1 Assessing concepts, systems and tools for a Safer, Efficient and Environmentally Aware and Friendly Maritime Transport.

Donkers: Ship shape and the ports escape?
For the future IMO legal instrument, it is crucial that the "Tool Box" of safe ballast water
management options on board is flexible, provided that they comply with performance standards for
effectiveness and do not cause unacceptable adverse environmental effects. Another important
principle is, that the Convention must give a clear signal to the maritime industry about how and
where the problem of alien species introduction by ballast water must be tackled.
With various on-board ballast water management options available and technical progress still
developing, it can be stated that a ship directed approach is feasible.

Port side developments
In Certain Areas, the current IMO Convention text states that additional measures such as the
provision of port reception facilities can be prescribed. At first, it must be said that the possibility of
declaring "more stringent measures" in a certain area is a right that every state has under the United
Nations Law of the Sea (UNCLOS). Therefore, with the IMO Convention serving as "baseline
requirements", countries are free to design stricter regulations. The Port of Rotterdam is of the opinion
that there are more possible measures than port reception facilities. Other possible measures include
mandatory BWE prior to port entry or mandatory use of a treatment technique before entering a
certain port or sea area. Apart from the basic right of very country to design more stringent measures,
the measures itself must be decided on depending on local circumstances.
For the sake of practicality, the Port of Rotterdam has executed a short desk study to visualise and
assess the operational, technical and economical implications of the creation of Port Reception
Facilities. The following main conclusions can be drawn.
Historical context
The provision of adequate and sufficient Port Reception Facilities (PRFs) has been a global debate for
over 20 years. Marpol obligates every state that has ratified the Convention to create and maintain
PRF's for oily, chemical and solid waste (garbage, hazardous and cargo-associated waste). For oily
and solid waste however, no total ban on discharge in the sea has been established, neither mandatory
delivery of waste at a PRF. Also, the inspection regime of the mandatory Oil Record Book has been
While some ports and countries still struggle to provide PRFs, within the Port of Rotterdam there have
been facilities for twenty years now. Because the lack of a "mandatory use policy", an in adequate
inspection regime and high costs, a very low number of ships use the facilities. For ship-generated
waste (bilgewater, sludge and garbage) the percentage of ships that use a PRF have fluctuated
between 3 and 7 % over the last five years. Fortunately, the European Directive 2000/59 on Port
Waste Reception Facilities on Ship-Generated Waste and Cargo Residues is likely to improve this
situation. The Directive prescribes indirect financing, mandatory discharge ashore and the design of
port waste management plans. The latter must ensure an improvement of the provided service level by
However, before the mentioned directive will be implemented, two major Reception Facilities for
sludge and bilge water in the Rotterdam Port Area have suffered a financial loss up to 10 million a
year over the years of 2001 and 2002. During 2002, the Rotterdam Port Management was forced to
buy out one of these companies and entirely redevelop its site, costing many efforts and financial
means. It goes without saying that the Port of Rotterdam does not strive to repeat this scenario.
If the provision of sufficient and adequate ballast water PRFs are made mandatory under the "Special
Area" approach in the Ballast Water Convention, this will be also be without any mandatory
requirement to use those facilities. This is the same approach that has been followed in the above
mentioned Marpol Convention, which obliges ports to provide sufficient and adequate facilities,
without any guarantee that they will be used.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Volumes of ballast water to be treated
In view of the high volume of ballast water influx (6 million tonnes annually, Aquasense, 1998), this
could mean 0.5 million tonnes per month that has to be treated on shore. This will result in a
considerable amount of storage tanks that have to placed on-shore. Apart from storage tanks, also tank
space must be available to clean and process this water. If the processed ballast water must be re-used
on board, even more tank space is needed for the "clean ballast" to be pumped back into vessels.
Adding the space demand resulting from storage, processing and possibly re-use, this might lead to
around 10 tanks of 116,000 m3 (= capacity of one tank at the Maasvlakte Oil Terminal see photo 1). If
the ballast water is stored in a basin, this likely also consumes a significant amount of space. This vast
space-demand will be very hard to allocate on for example container-terminals (see photo 2).
Per ship, the amount of ballast water varies from 10 to 30 % of DWT. This can lead to from around
10,000 or tonnes (container vessels) up to more than 100,000 tonnes (VLCC's). Apart from the large
volumes, the amount ballast tanks can be numerous as well as the location and variation of inlet
points. This is likely to result in the need for numerous on shore-connected pipelines as well as barges
and vessels. The size of these barges will be far bigger than an average bilge oil vessel. A flexible and
expensive infrastructure will be needed to transfer these considerable amounts of collected ballast
water to on-shore installations. Besides this, ships continuously aim to decrease the turn around time
in port. The strong wish for "zero undue delay" for the shipping industry view to the use of oily and
garbage waste facilities will without any doubt re-occur at this debate.
The on shore treatment process
A short desk- and literature study has provided more insight in the needed technical means to treat
large amounts of ballast water on shore. The following assumptions were made:
· The treatment process must be able to meet the performance standards for treatment
techniques, that is a rate of killing every organism (algae etc.) bigger than 10 µm). At the time
of writing this paper, this was stated in the Draft Convention Text;
· Quantities of ballast water that has to be treated varies from 1,000 to 5,000 tonnes (small
container ships) to 10,000 to 15,000 tonnes (large container ships);
· The approximate time at the port berth is 12 hours;
· The oil tankers that regularly visit Rotterdam can have up to 100,000 tonnes of ballast water
on board. The on shore treatment plant has to also to be able to process such amounts;
· If this water is discharged back into port waters, all non-indigenous fishes, shellfish, plants
but also phytoplankton (2-200 µm), zooplankton (50-1000 µm), bacteria's (1-10 µm) and
viruses (0.01-0.1 µm) have to be removed. Because of this, the water has to be disinfected.
The resulting cleaning process is presented in Figure 4.
The necessary cost of the technical means (grid filters, hydrocyclone, UV treatment plant and active
coal filter) are in the direction of 1, 5 million . This is based on a flow rate of 850 tonnes per hour.
This is a rough initial cost estimate, which does not include:
· Maintenance costs (personnel, machinery);
· On shore infrastructure costs (pipelines etc.);
· Vessel infrastructure (reception barges, trucks etc.);
· Costs because of land/space use (rent).
Such costs must be regarded in comparison what ships have to invest for on board treatment
techniques or executing BWE. It is important to realize that if a shore-based treatment approach is
chosen, ships will have to pay in numerous ports for such facilities and infrastructure. A point of
concern is the requested swiftness of service by the shipping industry. If the given tanker with over

Donkers: Ship shape and the ports escape?
50,000 or even 100,000 tonnes of ballast water arrives, it has to berth along the treatment plant. It will
be impossible to discharge such quantities in a mobile vessel.
Above mentioned foreseen problems apply to the reception of ballast water, and not sediment in
ballast tanks. For sediments, provision of reception facilities in ship yards and dry docks has been
common practice for years. The forthcoming Convention is unlikely to cause any problems in this
field. This accounts also for the reception of relatively small volumes of oily ballast water. Because of
the regulations of Segregated Ballast Water tanks, this type of waste is likely to decrease to zero in
years to come.
In view of the magnitude of volumes, operational, technical and financial implications on both port
and ship side, provision of reception facilities for ballast water can be regarded as very complex
and .... most likely very expensive. According to the "Polluter Pays Principle", the financial burden
will be put on the ship. The important question is whether the shipping industry is willing to pay for
the price of these facilities.

The ports escape?
If massive provision of PRFs is regarded as unrealistic, does that exclude ports in general from any
involvement in and responsibility for solving the problem of alien species in ballast water? The Port
of Rotterdam identifies a range of subjects where port involvement can or must take place.
Incoming ballast water
Although seriously harmful introductions seem not to have taken place yet in the nearby region, there
is an urgent need for adequate Risk Analysis. This must lead to an improved knowledge and
understanding of the ecological and economic vulnerability of the port, coastal and adjacent sea area
to alien species from ballast water discharges. Risk Analysis must provide insight to:
· Location and circumstances of ballast water uptake;
· Description of species in ships' ballast tanks;
· Location and circumstances of ballast water discharge;
· Ecological data on existing habitats and ecosystems (presence of solid substrate, tidal
currents, configuration of port basins, water quality);
· Resources that are at risk (biodiversity, fisheries, aquaculture, tourism, port constructions);
· Identification of "high risk" species.
The National institute for Coastal and Marine Management (a branch of the Dutch Ministry of
Transport, Public Works and Water Management) has monitored the contents of ballast water tanks of
ships entering Dutch ports (NICM management, 2001). This is, in fact, a first step to execute the
second above mentioned aspect of Risk Analysis (description of species in ships' ballast water).
Although a significant number of (zoo- and phyto) plankton species were found, it is too soon to draw
general conclusions.
Further monitoring of the ports' and coastal ecosystem must provide insight in the risks associated
with ballast water discharge in the Port of Rotterdam and other Northern-European ports. It is
recommended to integrate such research with general ecosystem monitoring that will be executed in
the light of the EU Habitat and Bird Directive. National governmental authorities are likely to play a
major role in this. On a local level, Port Baseline Surveys might be appropriate to provide specific
local data.
Such a comprehensive Risk Analysis must determine whether more stringent measures than the IMO
Convention are necessary. These measures might be mandatory use of a treatment technique or

2nd International Ballast Water Treatment R&D Symposium: Proceedings
mandatory BWE on mid-ocean before port entry. It is realistic to say that it will take quite some years
before all ships comply with the IMO standards. If a non-complying ship arrives in port, planning to
discharge ballast water from a suspected area were "high risk species" occur, this might lead to an
unacceptable risk. Also, ships might arrive in port that were not able to execute BWE because of sea
and weather conditions. These are issues that have to be dealt with. If for such cases the creation of
PRFs is considered, the former mentioned economical and operational aspects have to be taken into
Export of ballast water
With regard to the outgoing ballast water volumes (as mentioned earlier, approx. 60 million tons from
the Port of Rotterdam annually) risks and effects are largely unknown. It is unrealistic to execute a
comprehensive Risk Analysis for every sea area that the ballast water is transported to. However,
some desk research is recommendable to investigate effects that have already taken place in other port
and sea areas.
The last decades, there have been various studies to examine whether fresh water could be pumped in
the ballast tanks of oil tankers and transferred to for example the Gulf region, where fresh water has a
value comparable to or even higher than the oil transported. With the IMO Convention in sight, the
time could be right to seriously consider the start of a pilot-project in this direction.
Operational port involvement
For both incoming as exported volumes of ballast water, it is preferable that Port Authorities design a
"ballast water-uptake and discharge policy" in or in the vicinity of their port waters. For example,
ships can be advised to avoid ballast water uptake near sewage outlets or water regions that contain
much sediment, which could be the case near dredging operations. The latter minimises uptake of
species as well as the carriage of superfluous weight when sailing not under ballast.
In the World Port Center, the port's main office, a Harbour Co-ordination Centre (HCC) has been in
operation since 2001 (see photos 3 and 4). All ships that enter the territorial waters (or at least at the
time of departure from the last port) are obliged to inform the HCC about:
· Any deficiencies and/or accidents with regard to Marpol and/or Solas prior to arrival and
during stay in the port;
· The carriage and reporting of dangerous and/or noxious goods.
This intense ship-port interface provides good opportunities to integrate ballast water management
procedures. Also, the information provided to the HCC provided by the annual 30,000 incoming ships
can provide useful information that is needed when executing a Risk Analysis. For example, the
origins of ballast water can obtained from this data.
On board survey by Port Inspectors
The Port of Rotterdam has not waited for the Convention to enter into force in getting pro-active in
this field. In the beginning of 2003, an inspection unit of the Rotterdam Port Authority (RPA) has
started an on-board survey on the origins, destination of ballast water and management and treatment
techniques applied on board. The first results are presented in the following table:

Donkers: Ship shape and the ports escape?
Table 1. First results of RPA survey.
No. of ships that execute
Total no. of ships
Percentage of total
BW exchange
Container vessels
Oil tankers
Chemical tankers
Bulk carriers
General Cargo
Source: on board survey by inspectors of the Rotterdam Port Authority
Approximately one third of these vessels had exchanged its ballast water prior to port entry. From the
BWE methods applied, the sequential method was used most (by 29 vessels), followed by flow-trough
(15) and dilution method (3). One vessel, a chemical tanker, declared to have filtering equipment and
UV treatment on board. Approximately one third of the examined vessels had ballast water on board
which originated from outside the so called OSPAR2 region, which could be regarded as a sea region
in which species migrate freely by natural currents. From 10 % of the vessels, the origin of the ballast
water was unknown. Naturally, at a later stage with a higher number of vessels examined, more
structural statistical data and trends can be obtained from this on-board survey.
With about 4,000 Marpol-inspections executed per year by the RPA, there is a potential to survey a
substantial amount of ships. This inspection role could also be important view to the Port Inspections
that will be prescribed in the IMO Convention. The processing of the incoming Ballast Water Record
introductions of harmful aquatic organisms by ballast water. For the Port of Rotterdam and most
likely many other ports in the world, it is preferable that efforts are made to better analyse, assess and
manage the ecological and economical risks of imported as well as exported volumes. Other possible
areas of involvement are advising ships view to ballast water uptake and discharge activities in or
nearby port areas and the establishment of an effective port inspection regime.

2 Oslo and Paris Commission, a regional regulatory body for the sea area between Norway and Portugal.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 1. Map of the Port of Rotterdam Area.
Figure 2. Maasvlakte Olie Terminal, Rotterdam.

Donkers: Ship shape and the ports escape?
Figure 3. ECT terminal, Rotterdam.
ballast water
First filtration step
(20 - 50 µm)
Figure 4. on shore filtration process
(source: desk study ballast water treatment by Witteveen & Bos consultants, July 2003).

Figures 5 and 6. The Harbour Co-ordination Center in the World Port Center, Rotterdam.

Latest results from testing seven different technologies
under the EU MARTOB project -
Where do we stand now?
E. Mesbahi
MARTOB Consortium
School of Marine Science and Technology,
University of Newcastle upon Tyne, UK
MARTOB is a three-year project funded through the Transport and Energy Directorate of the
European Commission (GROWTH Programme). The MARTOB project began in April 2001, and it
has the dual aims of developing methods for treating ballast water on-board ships and for developing
recommendations of best practice for verification and monitoring of compliance of a sulphur cap for
marine fuels. Both of these aims are directed towards making shipping operations more
environmentally friendly.
The main work components to be carried out as part of the MARTOB project are as follows:
· Collection and assessment of data and information on ballast water management methods and
existing relevant legislation, and a review and update of alien species introductions in
European waters.
· Development of selected methods for on-board treatment of ballast water through lab-scale
testing and in-depth analysis.
· Large and full-scale testing of selected ballast water treatment methods.
· Assessment of the financial, technical and operational effects of a sulphur cap on marine
bunker fuel in European waters.
The first phase of the project related to ballast water management was completed in early 2002. This
included collection of information on ballast water management methods that are currently used, that
have been tested on board ships, or that are in an advanced stage of development. In addition to
collecting information on biological effectiveness, information was collected on the safety of
methods, environmental effects, and costs. Information was also collected on existing and proposed
regulations, to give an indication of future directions for ballast water management requirements.
Techniques tested within the MARTOB project
High temperature Thermal Treatment: This method uses heat to incapacitate and kill organisms in
ballast water. Low temperature treatment requires a long time and will not be effective against
bacteria and some of the hardier organisms, but will be cheaper to implement as it uses waste heat.
High temperature treatment is more expensive as in most cases it needs a dedicated heating system,
but is potentially more effective at killing the organisms and requires a much shorter exposure time.
De-oxygenation Treatment: De-oxygenation of ballast water can be achieved mechanically by gas
sparging, chemically by adding reducing chemicals, and biologically by adding nutrients. In
MARTOB only the latter method has been studied in detail. By adding nutrients into the ballast water,
the growth of the naturally occurring bacteria in the water will be stimulated. During the growth they
consume oxygen, and the oxygen in the water will be depleted.

Mesbahi: Latest results from testing seven different technologies under the EU MARTOB project
Ultraviolet Treatment: Ultraviolet (UV) lamps are used to irradiate the organisms in the ballast
water. The UV radiation will induce photochemical changes in the organism; i.e. it will break the
chemical bonds in DNA. This can lead to problems should the organisms survive, as it may carry
mutations. Furthermore, there is a requirement for pre-treatment of the ballast water, as the
performance of the system decreases with the turbidity of the water. Ultraviolet Treatment is well
established and proven as a disinfectant in the wastewater treatment sector.
Ultrasonic Treatment: Ultrasound is generated by a transducer, which converts mechanical or
electrical energy into high frequency vibration. The ultrasound generates cavitation in liquid (in this
case ballast water), which can lead to the cells of organisms rupturing. It has been shown to be
effective with bacteria, plankton and other larger organisms. However, ultrasound may have an
adverse effect on ship/tank coatings and ship structure and would, therefore, need to be tested.
Ultrasonic treatment has been successfully used in water treatment and the food industry to control
Ozonation Treatment: The Ozonation system introduces ozone into the ballast water. As ozone is
unstable at atmospheric pressure it must be generated in situ. Ozone has been used in onshore
applications, such as swimming pools, disinfecting drinking water and controlling microbiological
contamination in various areas. In these applications it has proven to be very effective and a more
powerful biocide than chlorine, which has traditionally been used. Ozone is toxic and therefore it will
have to be used with care. There is also concern that it may cause increased corrosion in the tanks and
Oxicide Treatment: The Oxicide method is an electrochemical method, which generates hydrogen
peroxide from the oxygen present in the ballast water. This decline in the concentration of oxygen and
the presence of hydrogen peroxide is enough to significantly reduce the number of organisms present
in the water. It also decomposes in water and will therefore not cause any problems to the
environment. Hydrogen peroxide is an irritant and it will have to be used with care and it could
possibly lead to increased corrosion.
Advanced Oxidation Technology: Advanced Oxidation Technology (AOT) consisting of a
combination of ozone, UV and catalysts. Thus Ozonolytic / Photolytic / Photocatalytic Redox
processes are operating simultaneously within a reactor. The unique combination is designed to
generate large amounts of radicals, mainly hydroxyl radicals, within the reactor. It is these radicals
that destruct / eliminate microorganisms. This water purifier has successfully been used in land-based
applications such as purification of swimming pool water, drinking water, water used for irrigation in
green houses and water used in fish breeding.
Hurdle Technology: Hurdle technology uses a combination of two or more treatment methods to
reduce the number of microorganisms present. This may increase the effectiveness of the treatment
and if chosen properly, can also eliminate some of the disadvantages of using the treatment methods
Timeframe of the project
MARTOB project, (On Board Treatment of ballast water (Technologies Development and
Applications) and Application of Low-sulphur Marine Fuel, partially funded by European
Commission by contract number GRD1-2000-25383, started in April 2001 for a period of 36 months
under the coordination of University of Newcastle upon Tyne.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Aims and objectives:
MARTOB's main objectives are:
· To investigate methodologies and technologies for preventing the introduction of non-
indigenous species through ships' ballast water.
· To develop design tools and treatment equipment to be used in the further development of
ballast water treatment techniques.
· To assess the effectiveness, safety and environmental and economic aspects of current and
newly developed methods.
· To develop cost-effective (capital and operating), safe, environmentally friendly on board
ballast water treatment methods, which have a minimum impact on ship operations.
· To produce guidelines for crew training and criteria for selecting appropriate ballast water
management method.
· To assess the financial, technical and operational effects of sulphur cap on marine bunker fuel
in European waters, and propose a verification scheme ensuring compliance with a sulphur
cap from all players in the market.
· To help to facilitate the introduction of an important sulphur emission abatement measure
without unintentional distortion of competition in the shipping market.
Research methods, test protocols and experimental design:
Laboratory-Scale Testing of Ballast Water Treatment Methods
The purpose of the laboratory-scale testing phase of the MARTOB project was to test a range of
ballast water treatment methods using a standard mixture of seawater and target organisms.
Specifications for the seawater/organism mixture were developed within the MARTOB project. The
test organisms included three species of zooplankton and two species of phytoplankton. By using a
standard mixture and analysis method it was possible to measure the biological effectiveness of all
methods and to make basic comparisons. In June 2002, laboratory scale testing of selected ballast
water treatment methods was carried out at the School of Marine Science and Technology at the
University of Newcastle upon Tyne.
In addition to assessing biological effectiveness of the treatment methods, information on safety,
corrosion, costs, and potential environmental `side-effects' is being collected for each method. It is
important that the methods are practical, safe for the ship and its crew, environmentally friendly, and
economically viable. These characteristics are in addition to the primary requirement that the methods
have to be effective at controlling the spread of alien species.
Materials and methods
Standard seawater was prepared for all tests 24 hours before use. Deionised water (supplier) was
added to Tropic Marine salt (35g/l) (Aquatics Unlimited, Bridgewater, Wales) in 4 mesocosms of 250
or 450l. Following the addition of water, the mixture was agitated continuously for 24h using
compressed air to ensure that all the salt had dissolved. Salinity was checked using a refractometer.
Cultures were supplied in bulk, zooplankton every 2 days and phytoplankton every 5 days. They were
stored in CT rooms in the aquarium suite at the Ridley Building, University of Newcastle, at 10 and
15°C respectively.
Information on supplied plankton density was available from the suppliers. Samples were measured
out directly from the cultures, each species being stored in a separate bottle. The organisms were
mixed with 70L of seawater that had been pumped into a tank, to create a sample of test organisms,
the `soup' (Table 1). This was the agreed minimum volume to be used in the experiments that would

Mesbahi: Latest results from testing seven different technologies under the EU MARTOB project
be statistically significant regarding the density of the organisms added as well as being cost effective.
However this volume can always be increased in the case larger experiments are wanted to be
conducted. After pouring the samples into the prepared seawater the bottles used to carry them were
rinsed twice in the same water and added to the mixture.
Prior to pumping the soup into test rigs the mixture was gently agitated to ensure a homogeneous
mixture. Following pumping to the test rigs the tank was rinsed with clean seawater to ensure removal
of any residual organisms.
Before initiating the treatments, a 10L initial sample was collected from each test rig for laboratory
analysis (see below). Treatments were carried out and on completion a 60L sample was taken for
A control tank containing one sample was set up and left at room temperature. Sub-samples were
taken at intervals to monitor background mortality (Table 1). Three replicates were made during three
consecutive days (12-14th June)
Table 1. Times after set-up and sample sizes used for control soup sampling.
Time of sampling
Size of sample
0 min
30 min
Biological assessment: sampling and test protocols
Within the MARTOB project it was necessary to assess the performance of various ballast water
treatment techniques. A standard test protocol was therefore required. Because the standards under
discussion at IMO were not finalised, it was necessary to develop a test protocol specifically for this
project. The developed protocol is to some extent based on the draft standards, but also other
suggested protocols were taken into account.
The sampling and test protocol provided standards for:
· water quality,
· species to be used for laboratory tests,
· composition of the test mixture,
· how to assess the biological effectiveness.
The water quality standard specifies the quality and quantity of the artificial seawater (ASW),
including salinity, turbidity, pH and temperature. The chosen salinity was 33-35, achieved by adding
"Tropic Marine seasalt" to distilled water. Seawater may be turbid due to both inorganic and organic
particles. Kaolin was used to simulate the former, while flour was used to simulate the latter. The pH
of the ASW was around 8.3, i.e. close to the normal pH of seawater. The temperature was 10-15°C to
ensure the survival of the introduced marine organisms.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Five different species, three zooplankton species and two phytoplankton species, were selected as test
organisms, and added to the ASW. The zooplanktons were a polychaete (nectochaete larvae of Nereis
), a harpacticoid copepod (Tisbe battagliai), and a calanoid copepod (Acartia tonsa). The
phytoplanktons were a diatom (Thalassiosira pseudonana) and a dinoflaggelate (Alexandrium
). Densities of the species are given in Table 2.
Table 2. Artificial Sea Water or MARTOB Soup
Selected Species
Maximum field densities
Standard mix composition
Standard mix composition
(indivs /m3)
(indivs/ m3)
of a 70 litre test solution
Benthic nectochaete larvae
Nereis virens
Harpacticoid copepod
Tisbe battagliai
Calanoid copepod
Acartia tonsa
Thalassiosira pseudonana
Alexandrium tamarense

Mesbahi: Latest results from testing seven different technologies under the EU MARTOB project
The mix used did not include any fish eggs or larvae. In many countries, including the UK,
experiments involving vertebrates require special licenses. For this reason we excluded them from the
standard test mix and would propose that separate trials of a mix containing fish eggs and larvae
(probably salmon or turbot) be conducted, under licence for the most promising techniques identified
in the trials with the standard mix. The mixture composition describes the density of the species to be
included in the test mixture. The premise here is that densities should reflect the top end of the natural
range for each taxa.
The effectiveness of each individual treatment technique was assessed by determining the number of
live and dead organisms of each species after the treatment. This was done by fixing and staining the
organisms in a manner that allowed living and recently dead material to be easily distinguished. This
will allow the efficiency, expressed as % kill, of each technique for each group of organisms to be
During the first few days of testing, UV, US and Ozone techniques used a high pressure pump for
supplying artificial seawater into the treatment system. Analysis of preliminary results showed that
the pump itself was eliminating almost all of the zooplankton; therefore a gravity system was used to
supply the water for the rest of the tests. Consequently, it was observed that large number of bends,
valves and long pipes could contribute as a source of error for these technologies. Since ASW
flowrate was now much lower than original pump, it was concluded that some of species were
gathered into the slow velocity points, thus altering some of the results. Both living and dead
organisms were found to be hidden in the systems. It was therefore decided to flush these systems
after each test run, when some of zooplankton species were detected from the sample. This could
slightly remedy the source of error but there are still concerns regarding the accuracy of analysis.
Zooplankton fixation and staining
All samples were filtered through a 63 µm sieve. The zooplankton was rinsed from the sieve with
clean seawater into labelled pots.
Zooplankton samples were stained with 0.1% Neutral Red solution in the ratio of 3ml
stain/100 ml sample. After staining for 60 min, 4 ml of 1N Sodium Acetate solution was added
per 100 ml of sample. The specimens were then fixed with 4% Formalin in a volume equal to that
of the sample (50/50). Thereafter all samples were stored overnight at 5°C prior to counting.
Following the overnight storage and before examination of the samples, Glacial Acetic Acid was
added dropwise to each sample, until the colour of the solution changed to magenta. The sample
was filtered through a 48 µm sieve and washed with tap water. During the counting procedure the
sample was kept in water. After counting organisms were preserved in 4% Formalin.
Live copepods stained immediately prior to fixation turned a deep magenta after acidification,
whereas dead specimens were light pink to white. Nereis had to be more carefully observed, as
dark staining did not guarantee viability. Some treatments affected the staining in such way that
`live' organisms varied in colour from magenta to orange. Therefore the assessment of individuals
also included a morphological examination.
For the counting procedure whole organisms as well as bits were taken into account. The quantity
of organisms delivered by the suppliers was a range between two densities therefore we dealt with
volumes and not with exact number of organisms to make the soup samples. The percentage of
mortality was calculated as the number of dead animals divided by the sum of dead and alive
animals found in the after treatment samples. When no material or no whole animals only bits
were found a 100% in mortality was recorded.
Framework of evaluation
Environmental assessment
Environmental assessment includes evaluation of the direct environmental impact resulting from
the discharge of treated ballast water and consideration of the indirect environmental impact.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Direct impacts on receiving waters can result from discharges of the ballast water systems
including that with altered quality, discharge of solids from physical separation methods, and
discharge of living organisms that have survived treatment. The treated ballast water will be
sampled on discharge for those parameters that are expected to change as a result of the treatment.
Data from testing for biological effectiveness will give an indication of the types of organisms
that will survive the treatment and be discharged. Indirect environmental effects of ballast water
management will be assessed by estimating energy use, and calculating amounts of materials used
during both operation and construction of the treatment equipment. Waste generated during
operation or through disposal of worn out components and equipment will also be assessed.
Safety assessment
The assessment of safety aspects of treatment methods within the MARTOB project will be based
on an evaluation of operational aspects. These include use of hazardous chemicals (either
generated or stored on site), hazards related to operation of the equipment, aspects related to the
storage and handling of chemicals and residuals required for, or resulting from, the on-board
treatment of ballast water; and aspects related to unintentional release on board the vessel of
treated ballast water containing residuals. The safety assessment of each method will also
consider possible accident scenarios.
Economic assessment
In order to assess the economic viability of treatment options, two basic cost components are
relevant, i.e. capital costs and operational costs. An interest rate of 8% over a period of 10 years is
recommended to depreciate the investment costs, which fall under capital costs. Material costs,
personnel costs and maintenance costs all fall under operational costs and these must be estimated
in details for each treatment option. Other cost components like those resulting from training and
management issues and those from the economic benefits and disadvantages of treatment options
all need to be estimated in detail for each treatment option. All these cost components mentioned
above must be estimated based on the same basic data e.g. ship type, ballast water capacity,
number of voyages per year, number of ballast pumps, ballast pump capacities etc.
Technical and operational applicability
With respect to on-board ship applicability of treatment options, the options, alongwith their
space requirements, capacity, flow rate and time, should be checked on the vessel for effects on
stability, visibility, longitudinal strength, overpressure in the ballast tanks, liquid motions in the
ballast tanks, thermal stresses, aggressiveness versus materials, corrosion, pressure drop in
pumping system, modification of the piping and pumping system, safety of the crew and
compatibility with trip duration and crew working load.
Objective assessment
Assessment of ballast water treatment technologies have not been limited to their biological
effectiveness only, other criteria such as their compatibility with a particular ship and her route,
overall cost, safety, crew, life cycle assessment, corrosion effects and many other factors have been
considered here as ranking criteria with their individual weighting in the final assessment. MARTOB
has also developed a comprehensive IT based technology to determine attractiveness of a particular
ballast water management system for an individual ship travelling at a particular route. Details of the
developed objective assessment methodology have been illustrated in Figure2.
Results: biological effectiveness
High temperature thermal treatment
Offers one solution if short heating period is required for the effective elimination of unwanted marine
organisms. Treatment during ballasting and treatment at exit (deballasting) are two possible options.

Mesbahi: Latest results from testing seven different technologies under the EU MARTOB project
The treatment at exit does not require the water to be pumped from one tank to the other for treatment,
or additional tanks for storage, both of which can cause problems with stability of a vessel and/or
reduction of the cargo space. There is also no risk of cross-contamination of the treated ballast water,
once treated water is discharged. A possible problem for this system is that the equipment reliability is
critical as the water is not stored and there is therefore no backup.
The effects of temperature on phytoplankton and zooplankton have successfully been tested under
laboratory conditions. This has allowed us to obtain a correlation between kill rate and temperature
for Acartia sp., Nereis sp. and Tisbe sp., three zooplankton species commonly found in ballast water.
For the phytoplankton Alexandrium sp. and Thalassiosira sp., it was stated that all the temperatures
that were used for thermal treatment resulted in a reduction of chlorophyll a. However, experiments
carried out at lower temperatures (40 and 45°C) resulted in a significantly lower reduction of
chlorophyll a. It would therefore appear that temperatures of 55°C and above were more effective at
reducing phytoplankton biomass. However, there was no significant effect between the results for
treatments at 55, 60 and 65°C, which would seem to indicate that increasing the temperature above
55°C does not result in a corresponding reduction of chlorophyll a. Combining the results from the
zooplankton and phytoplankton we have been able to deduce a treatment temperature for the high
temperature thermal treatment system of between 55 to 60°C. Figure 3 shows laboratory scale
equipment for High Temperature Thermal Treatment.
Biological de-oxygenation
The solubility of oxygen in water is low. Biological de-oxygenation is based on the fact that addition
of nutrients to ballast water will stimulate the growth of the indigenous bacteria in the ballast water.
The growth of the bacteria will consume the available oxygen in the water, and when the ballast water
becomes anoxic, organisms that require a steady supply of oxygen will die. The aim of the studies was
to develop a de-oxygenation process that could be applied in large scale, and to test the efficiency
towards selected organisms in the mesoscale trials in Newcastle. See Figure 4.
The time it takes to consume all the oxygen in seawater decreases with increasing temperature. At
4°C it will take 3-4 days, at 10-20°C, 1-2 days and above 20°C less than 1 day to obtain anoxic
Biological de-oxygenation was tested in meso-scale in 50 litre polypropylene vessels covered with
black plastic bags to simulate the darkness in a ballast tank. The efficiency of the treatment was tested
against three species of zooplankton, two copepods (Acartia tonsa and Tisbe battagliai) and one
polychaete (nectochaete larvae of Nereis virens), and two species of phytoplankton, a dinoflagellate
(Alexandrium tamarense) and a diatom (Thalassiosira pseudonana).
Biological de-oxygenation of the seawater killed all the added zooplankton species. The killing rate
increased with increasing time under anoxic conditions. After 4-6 days of anoxia, more than 95% of
all the tested organisms were dead.
The killing effect on phytoplankton of de-oxygenation was limited as measured by the change in the
concentration of chlorophyll a. On average, the chlorophyll concentration decreased by 33%, but there
was no significant difference between the treated tanks and the non-treated controls. The reduction in
chlorophyll may therefore be due to the fact that the micro algae in all cases were incubated in
Corrosion effect estimated with FMECA analysis identified the following issues: a slight decrease of
the pH with possible consequences on metal corrosion, coatings and gaskets, a slight increase of CO2
with possible consequences on metal corrosion and gaskets, the production of H2S with possible
consequences on metal corrosion, coatings and gaskets, the addition of inorganic substances with
possible consequences on metal corrosion, coatings and gaskets, the addition of organic substances
with possible consequences on coatings and a significant increase of the bacteria concentration with
possible consequences on metal corrosion, coatings and gaskets.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Ultraviolet light treatment (UV)
UV irradiation is used for the disinfection of potable, process, aquaculture and waste waters. It
achieves disinfection by inducing photochemical changes of biological components within micro-
organisms, and more specifically by breaking chemical bonds at the DNA and RNA molecules and
proteins in the cell. In the majority of UV disinfection applications, low-pressure mercury arc lamps
have been chosen as the source of UV radiation. Approximately 85% of the output from these lamps
is monochromatic at a wavelength of 253.7 nm. This corresponds to the short wave portion of the UV
spectrum which in all spans from 200-280 nm, and is referred as UV-C. The sensitivity of micro-
organisms to UV radiation depends on the wavelength. Microorganisms are sensible to UV radiation
between 210 and 320 nm, with a peak at 265 nm. See Figure 5.
Maximum reduction rate of 78% with phytoplankton was achieved and regarding zooplankton the UV
method did not inactivate more than 56%. With UV treatment, the greatest percentage change of
chlorophyll a concentration achieved was a 56% reduction.
UV light causes a slight increase of the Redox potential (short term effect) with possible
consequences on metal corrosion, coatings and gaskets.
Ultrasound treatment (US)
Ultrasonic treatment is a relatively new technology in ballast water treatment. Ultrasonic liquid
treatment uses high frequency energy to cause vibration in liquids to produce physical or chemical
effects. Ultrasound, from 20 kHz to 10 MHz, is generated by a transducer that converts mechanical or
electrical energy into high frequency acoustical (sound) energy. The sound energy is then fed to a
horn that transmits the energy as high frequency vibrations to the liquid being processed. The action
of ultrasound is thought to be mediated through various responses that may be fatal to marine
organisms. These are heat generation, pressure wave deflections, cavitation and possibly the
degassing effect of ultra-sound causing removal of much of the oxygen. Cavitation, the formation of
gas cavities within liquids, is affected by the frequency of the ultrasonic, power level, volume of
water, temperature of the water and concentration of dissolved matter and gases. Higher frequencies,
warmer temperatures and lower concentrations of dissolved matter have been found to increase the
effect of ultrasound pulses. Plankton mortality has also been observed in the presence of ultrasound
and is considered in part to be attributable to the cavitation process.
The mortality attained by the US treatment was always below 40% for zooplankton for all the tests.
The highest percentage change of chlorophyll a levels achieved with US was a 71% reduction.
No risk of increased corrosion with respect to coating and gaskets was identified regarding the US
Ozone treatment
O3 is the triatomic form of oxygen which is a gas at room temperature. Marine applications of ozone
include depuration of shellfish, oxidation of colour producing organics and toxins, improvement of
filtration, control of microbiological contamination in aquaria and aquaculture, and control of
biofouling in cooling water systems. Ozone is a fairly powerful but unstable agent which rapidly
destroys viruses and bacteria, including spores, when used as a disinfectant in conventional water
treatment. Salt-water ozone reactors are currently used for salt-water aquariums and fish hatcheries.
The three modules of an ozone treatment system are a generator, ozone contact chamber, and ozone
destructor. In the contact chamber ozone is introduced to the water stream. Biological effectiveness is
a function of concentration and exposure period. The longer the ozone-contact time, the higher the
mortality. Industrial systems use a "bubble contractor" chamber that maximises ozone exposure. A
bubble system was also selected to the ozone device utilized in the Martob project. See Figure 6.

Mesbahi: Latest results from testing seven different technologies under the EU MARTOB project
Mortality rates increased rapidly with increasing contact time. The highest value for the O3 treatment
was 89%, eliminating Nereis. Phytoplankton results showed that O3 reduced chlorophyll a levels with
a 97.2 percentage change against samples taken before treatment.
O3 method caused a significant increase of the Redox potential (short term effect) with possible
consequences on metal corrosion, coatings and gaskets. The production of O3 (short term effect) with
possible effects on metal corrosion, coatings and gaskets was also identified.
Oxicide treatment
Hydrogen peroxide is an oxidising compound and can be produced in-situ by means of an
electrochemical conversion of dissolved oxygen. This new process, the Oxicide process, is carried out
in a specially designed and patented electrochemical reactor. H2O2 destructs plankton and
microorganisms in the ballast water. Hydrogen peroxide is known to be of limited risk to humans,
especially at low concentrations. It decays within a period of days or a few weeks, resulting in
harmless compounds: water and oxygen. Hydrogen peroxide has various applications, among others
treatment of swimming pool water, as an alternative to chlorine based disinfectants. A first design of
the Oxicide cell has been built and tested under laboratory conditions at a scale of 100 dm3 water per
hour. It contained three Oxicide cells in series, each with contactors for supplying oxygen to the sea-
water, the source of which is either pure oxygen or air. The seawater runs along a 3 dimensional
electrode (cathode), where the oxygen is transformed to hydrogen peroxide. The anode compartment
is fully separated from the seawater compartment by means of a conducting membrane. It was found
that the maximum achievable concentration of hydrogen peroxide in seawater is determined by
kinetics and depends on the concentration of dissolved oxygen, temperature, electrical current and cell
voltage. The H2O2 concentration follows a logarithmic trend in batch operation. The highest
concentration of H2O2 achieved at ambient condiments was approx. 400 mg per liter (using pure
oxygen gas) or 150-180 mg per liter (using air). The initial current efficiency (CE) was 70-80%. The
pH of the seawater decreases because of some migration of H+ ions from the anode compartment
through the membrane. The maximum observed pH drop in a batch operated Oxicide cell was from
pH 8.4 to pH 6.5. The 3-dimensional electrode of the Oxicide module showed no plugging or
irreversible retention of particles in tests with kaolin, wheat flour and algae, i.e. particles < 100 µm.
See Figure 7.
H2O2 is efficient against selected organisms: 100% of Nereis and 90% of Acartia were removed in
all experiments at 10-15 mg H2O2/dm3. Tisbe proved more difficult, but was also removed by at least
85% at higher concentrations of H2O2 (> 28 mg/dm3). Furthermore, a reduction in chlorophyll a levels
of 50% was achieved by Oxicide treatment at 10-15 mg/dm3, although some of the other test results
with phytoplankton were inexplicable.
Elevated temperature (up to 35°C) seems to improve the efficiency of H2O2, especially zooplankton.
A literature study and additional tests revealed that some organisms need much higher concentrations
(>100 mg H2O2/dm3) to destruct or inactivate; this especially holds for large organisms.
In summary, various organisms are destructed or inactivated at relatively low concentrations of
hydrogen peroxide (10-30 mg H2O2/dm3). A treatment time of at least 24 hrs is required for H2O2 to
take full effect. However, a combination of Oxicide with other techniques should be considered,
because of the relatively high resistance of some organisms to hydrogen peroxide.
In terms of corrosion assessment, the production of H2O2 and the significant increase of the Redox
potential of the water (several hours to a few days) may have consequences for the metal corrosion,
coatings and gaskets. In addition, it is recommended to consider the electric isolation of the DC
equipment, because of the risk of unexpected current return paths and significant local metal

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Advanced Oxidation Technology (AOT)
AOT consists of a combination of ozone, UV and catalysts. Thus Ozonolytic / Photolytic /
Photocatalytic Redox processes are operating simultaneously within a titanium reactor to generate
large amounts of radicals, mainly hydroxyl radicals, which will destruct and/or eliminate
microorganisms. This technology has successfully been used in land-based applications such as
purification of swimming pool water, drinking water, water used for irrigation in green houses and
water used in fish breeding. The water was circulated through the water purifier. Tests were taken
after 1 ­ 10 cycles. Some tests were carried out with 100 µm filter upstream the water purifier. The
combination of AOT and the 100 µm filter could achieve over 95% killrate of zooplankton. See
Figure 8.
In the samples after treatment with the water purifier and filter the number of dead and alive
zooplankton are low (1.4 - 17% of the number initially included in test water). Organisms are
obviously caught in the filter. Also in the samples after treatment with the water purifier and no filter
the number of zooplankton are low (down to 4% of the number included in test water). This indicates
that organisms are eliminated by the water purifier. It could be that some organisms are left in the
pipes or in the tank. But compared to the number of zooplankton left after a test with only the pump
(35-52% of the number included in test water) some may have been lost.
The combination oxidation technology together with the 100 µm filter achieved a 40-70% reduction
in chlorophyll a compared to samples taken before treatment. This indicates that there has been a
reduction in the phytoplankton biomass. It is possible that the filter caught some of the phytoplankton.
In terms of corrosion assessment a moderate increase of the Redox potential (short term effect) with
possible consequences on metal corrosion, coatings and gaskets and a slight increase of CO2 with
possible consequences with respect to metal corrosion and coatings were recommended for careful
Hurdle Technology
Combining disinfecting technologies offer the option of eliminating the limitations of individual
techniques as well as the advantage of using the synergy of different methods. From the food industry
it is known that combinations of two disinfecting techniques have more effect than the sum of
individual conservation methods. One well known application of hurdle technology in ballast water
treatment is the combination of filter technology (hydrocyclons) and UV disinfection.
During the MARTOB trials various combinations were tested, based on the expected synergistic
effects, i.e. the combination of mechanical filter + US + UV, filter + UV + oxicide (H2O2), H2O2 +
UV, thermal treatment + de-oxygenation and H2O2 + heat treatment.
From the results of the hurdle technologies, the treatment that worked best was the low temperature
thermal treatment (40°C) + de-oxygenation, which had 100% efficiency for Tisbe and Nereis, and
97% for Acartia.
Comparing the efficiency of UV+H2O2 with and without filter (150 µm), the results showed that the
filter did affect the survival of the organisms, as the percentage removal increased for Acartia and
Nereis when the filter was used.
The combination of US and UV achieved a 68% reduction of chlorophyll a levels compared to
samples taken before treatment. The combination of filter, US and UV achieved a 57 % reduction of
chlorophyll a level.
Regarding the phytoplankton results, it is difficult to be certain which of the combinations of
technologies are the most effective. It would appear that combinations of low heat with de-
oxygenation or hydrogen peroxide were not effective at reducing chlorophyll a. The remaining four
treatments were all based on combinations of UV and hydrogen peroxide, sometimes with the added

Mesbahi: Latest results from testing seven different technologies under the EU MARTOB project
combination of a filter. On two occasions this reduced the chlorophyll a by over 70%, on another
occasion the reduction was less than 20% and the fourth run resulted in an increase in chlorophyll a. It
is therefore impossible to say with any certainty whether this combination of technologies is effective.
Results: Environmental impacts, Risk and Safety and Economic aspects
In the laboratory testing phase of the MARTOB project, information from the laboratory scale test
reports and from information provided by system designers for ballast water treatment on a case study
ship formed the basis of the evaluation. Evaluation criteria developed within the MARTOB project
were used to assess each of these effects for each of the methods tested at laboratory scale. To provide
a consistent basis for comparing the individual ballast water treatment techniques, a theoretical case
study approach was used. Data on the case ship and sample voyage were specified and provided to the
technical developers in the project, as well as a list of data needed for assessing cost, environmental
effects, and hazards.
Table 2. Case study SHIP details.
Basic Ship Information
Ship type
Pure Car and Truck Carrier (PCTC)
DWT (Dead Weight Tonnage)
Length Overall
199.1 m
Voyages per year
Southampton ­ New York
Ballast water capacity
8076 m3 (total volume of all ballast water tanks)
Volume of Ballast Water to be treated per trip
2000 m3
Number of ballast pumps
Capacity of ballast pumps
500 m3/h
Additional Data Selected for Economic Assessment
Specified for case study
Power consumption of
Energy use per hour per pump
50 kW
Fuel Type
Fuel type used for BW pumps
Energy content of fuel
Standard factor
42.5 MJ/kg
Fuel notional costs
Have to be standard for all comparative
0.4 /kg
Fuel conversion efficiency
Standard factor
(diesel to electricity)
Fuel conversion efficiency
Standard factor
(diesel to steam)
Depreciation period
Period in years used to annualise capital costs
10 year
Interest rate
Interest rate used to annualise capital costs
Fuel cost
Cost per litre MDO
0.4 /kg
Personnel cost
Average cost per hour
25 /h
Risk and safety effects
For the risk and safety assessment of ballast water treatment methods, hazard identification was
carried out and some recommendations for potential risk control measures were provided. Hazards
can be considered from the perspective of safety/survivability of the vessel and safety of the crew
during ship operations. Categories of hazards related to operation of the ballast water treatment
methods tested in MARTOB include physical hazards such as heat, electrical hazards, ultraviolet or

2nd International Ballast Water Treatment R&D Symposium: Proceedings
ultrasound radiation hazards, and chemical hazards from gases or hazardous liquids used or generated
during treatment. The major hazards associated with most of the treatment methods, including thermal
treatment, UV, US, Oxidation, and Oxicide, were confined to the location of the equipment
installation. None of the on-board treatment methods have the potential to threaten ship structural
integrity in the manner of empty-refill ballast exchange. For biological de-oxygenation and ozone,
ballast water is treated in the ballast tanks, so the hazard would encompass a larger area of the ship.
Most of the ballast water treatment methods, with the exception of biological de-oxygenation and
ozone, require the ballast water to be pumped through treatment systems. This additional piping
means that there is an additional risk for pipe breaks and leaks of treated or untreated ballast water.
However, this is expected to be a minor risk as most additional pipe work would be in a very localized
Other hazards associated with ballast water treatment include the potential for a spill of hazardous
material stored or being used within the treatment system. The UV and AOT treatment systems both
use UV lamps that contain mercury or amalgamated mercury. The oxicide method uses nitric acid as
an anolyte and requires sodium nitrate salt to be stored on board. All of these could result in damages
if accidentally released.
With all methods, there is the potential to reduce risks through appropriate training and safety
procedures. If these systems are installed on new ships additional safety features could be considered
during ship design.
Environmental effects
Environmental impact categories used to assess the effects of each of the ballast water treatment
technologies tested in the MARTOB project included:
· Direct Impact through Discharge to Receiving water:
- Discharge of water with altered quality with respect to the following parameter types:
Physical parameters
Nutrients/Oxygen Demand, Low D.O.
Biocide residuals
- Discharge of surviving organisms
- Discharge of solids (organisms and sediments)
· Other Environmental Impacts
- Energy Consumption (treatment systems, additional pumping, filtration)
- Potential for spill of treatment chemicals
- Materials use (both for consumables and for construction of treatment equipment)
Although some of the treatment methods will result in the discharge of ballast water with altered
quality, none of the discharges will include substances that are identified as `priority hazardous
substances' (under the European Union's Water Framework Directive), or that have the potential to
bio-accumulate. Ballast water quality will undergo the most change with the biological oxygen
removal method, which will produce a discharge that is low in dissolved oxygen and that has
increased concentrations of nutrients and bacteria. The Oxicide and advanced oxidation methods will
both lower the dissolved oxygen concentration of the ballast water. Increased temperature of the
ballast water discharge will occur after thermal treatment and ultrasound treatment. UV treatment has
no effect on ballast water quality.
For all methods, the ballast water discharge will include some form of organic matter in the form of
dead organisms, but this will vary depending on whether filtration is used, treatment type, and the

Mesbahi: Latest results from testing seven different technologies under the EU MARTOB project
concentration of organisms within the intake ballast water. The potential of this would be much less
than if live non-indigenous species are released, but could be of minor concern in eutrophic waters.
All but two of the treatment methods would be operated using a filter as pre-treatment. Biological de-
oxygenation and ultrasound treatment do not require the use of a filter. Methods using the filter as
pre-treatment will need to discharge the filtered material to the receiving environment, which could
cause some turbidity.
All treatment methods require the use of some energy, and this will result in environmental effects
from fuel consumption and associated emissions. Energy use is lowest for biological oxygen removal
and high temperature thermal treatment is the most energy intensive method (although the energy
used is dependent on the selected treatment temperature and the temperature of the ballast water
before treatment).
Stainless steel and titanium are the most commonly used materials for constructing the treatment
systems. Materials used for construction of the treatment equipment will be further refined in the next
phase of the project when the treatment systems are constructed for full scale testing. It should then be
possible to have more detailed information to assess life cycle impacts of these methods.
Economic aspects
Installation of an on-board ballast water treatment system will lead to changes in ships' capital costs,
changes in annual operating costs, and possibly will lead to extra training and management costs and
economic benefits or disadvantages. Generally, the cost calculation results highly depend on some
basic data associated with shipping trade and ballast water treatment. This may include type and
characteristic of the vessel, sailing and trading pattern, including aspects like route, distances, speed,
sailing and harbour time, and number of voyages per year, volume of ballast water, number of ballast
pumps and their capacities, type of fuel used, type of treatment and treatment capacity. Costs can be
easily compared when they are calculated based on the same type of dependants mentioned above.
The theoretical case study approach provided a consistent basis upon which to compare costs.
Table 3. Preliminary calculations for costs
Cost Type
Capital costs

TOTAL capital
one time investment
costs (for 10
costs (including
installation, testing, &
10 year depr. at 8%
* Material
Costs of all materials
needed in the course
of system operation,
including fuel.

Including materials
and labour
Training and
Including training,
Total costs
All costs annualised.
Costs per m3

All costs calculated
BW (/m3 BW)
towards costs per
tonne ballast water

2nd International Ballast Water Treatment R&D Symposium: Proceedings
From the preliminary cost calculations it can be concluded that there are still some data gaps to be
filled in. For some treatment methods the potential cost and cost factors are already quite transparent,
for some other systems there is still a lot of data to be estimated. The differences are partly related to
the status of development of the method. It is expected that during scaling-up of the systems and the
large-scale trials more data will become available. In addition more research into tank cleaning costs,
cost of corrosion control, certification cost, average wages of on-board personnel, total shipping cost
to be able to calculate the impact of ballast water treatment on the total cost of shipping, needs to be
The preliminary cost of treatment of ballast water on "the case study ship" varies considerably,
ranging from 0.10/m3 in the case of biological de-oxygenation up to 2.34/m3 for oxicide.
Nevertheless, it should be kept in mind that not all data were available for the techniques, and some
were preliminary.
Results: evaluation of corrosion risk of the treatment methods
In ships, an important problem is the corrosion of the hull structure, the piping system and the ballast
water handling equipment. Therefore it has been decided to identify if the installation and operation
on board of the considered in the MARTOB project ballast water treatment systems will modify the
water properties in such a way that it could increase the corrosion risk of the ship structure and ballast
water piping network. The target of this study was not to perform a detailed analysis of the corrosion
risk link to each system which will require information about the ship on which they will be installed,
but to provide a warning to the designers and classification societies which will have to approve the
installation on board, on the main possible new risks with respect to corrosion attached to each
system. This approach was carried out utilising FMECA grid support and ranking tables developed by
MARTOB's expert group.
The parameters considered in the analysis with indication of the variation or consequences which
induce a corrosion risk increase were water properties, water content and circuit content. The
resistance list for the chosen coating is important. It appears that the manufacturers of the coatings,
linings, seals, Dresser couplings, pumps, etc. should be asked to provide a resistance list for their
product. The coating maker will have to investigate the resistance of the coating where the ballast
tanks contain treated water.
Therefore, it is possible that the chosen ballast water treatment method needs to be specified first so
that the materials with the best corrosion resistance and coatings compatible with the water content
can be chosen for the detailed specification of coating, piping, pump, valve, seals and alloys etc.,
based on the treatment method.
All risk increases are acceptable considering today's knowledge and can be managed for new ship
design with existing techniques and methods. In existing ships, some treatment systems may be not
acceptable due to the treated water, incompatibility with the existing piping, gaskets or coatings
Full/large scale trials
Strategy for full scale is based on the experience gained from laboratory scale test trials. High
Temperature Thermal Treatment, de-oxygenation and oxidation technologies will be tested onboard a
Care and Truck Carrier. Ultraviolet, ultrasound, ozone and oxicide methods will be tested with large
scale facilities.
In the large scale test phase of US and UV the duration of test runs will be longer in order to minimise
the technical sources of errors, i.e. piping, fittings, valves and small amount of water. The use of sea

Mesbahi: Latest results from testing seven different technologies under the EU MARTOB project
water enables the access to unlimited amount of water and thus the error caused by the small amount
of water can be reduced. Also the link to the actual marine environment is evident. The strategy with
ozone has also been changed. The contact time will be extended with modification of the device in
order to monitor ozone dosage per amount of water versus contact time. Various ozone dosages and
contact times will be studied, and long term test runs might also be carried out.
To assess biological effectiveness of treatment systems, similar procedures as laboratory tests will be
followed. Standard sieves of size ranging from 10 µm (for phytoplankton) to 50 µm (for
zooplanktons) will be used onboard the ship. A large volume of Ballast Water (1000 litres) will be
tested at each sampling period to ascertain true representative of individual Ballast Water tanks. The
effect of time spent in the ballast water tank on species' survival will also be studied.
Conclusions and recommendations
During last two years, MARTOB has gained valuable expertise in the field of Ballast Water treatment
technology, assessment of biological effectiveness (large and small scale), development of test
protocols and procedures and overall objective assessment.
MARTOB believes that all key criteria in the development of Ballast Water technologies should be
weighted and considered accordingly.
MARTOB believes that given time and adequate funding, there are technologies which have the
capability of reaching high standards for Ballast Water treatment. Setting up a high standard of "No
harmful discharge" and deciding on realistic time horizons to achieve such goal, could urge
technology developers to seek more effective solutions. Considering the existing level of expertise, a
primary standard of "No discharge of live species larger than 50 µm" seems justifiable. More
stringent standard (i.e. No discharge of 10-20 µm live species) could be introduced in a 3 or 5 year
time and after re-visiting the level of technological developments.
During last few years, significant progress has been made by various projects all around the world;
MARTOB strongly suggests that additional Research and Development funds through appropriate
channels at national, continental and international levels should be provided to enable technologists
and scientists proceed with further development.
MARTOB Partners
University of Newcastle upon Tyne
T V/den Heuvel Watertechnologie BV (NL)
(Coordinator, UK)
Abo Akademi University (FIN)
The International Association of Independent
Tanker Owners (UK)
VTT Industrial Systems (FIN)
Souter Shipping Ltd. (UK)
Environment, Energy and Process Innovation
SSPA Sweden AB (S)
Institute for Applied Environmental Economics
Three Quays Marine Services (UK)
SINTEF Applied Chemistry (NO)
International Chamber of Shipping (UK)
Fisheries Research Services (UK)
Bureau Veritas (F)

2nd International Ballast Water Treatment R&D Symposium: Proceedings
French Research Institute for the Exploitation of
(MARINTEK) Norwegian Marine Technology
the Sea (F)
Research Institute (NO)
Association of Bulk Carriers (UK)
Shell Marine Products (NO)
Alfa Laval AB (S)
Wallenius Wilhelmsen Lines (SW and NO)
Berson Milieutechniek B.V. (NL)
Environmental Protection Engineering S.A. (EL)
Fueltech AS (NO)
Norwegian Shipowner Association (NO)
MARTOB is partly funded by the European Commission under the 5th Framework Programme for
Research, Technological development and Demonstration activities, Programme GROWTH, and is
managed by the Direction-General for Energy and Transport.
All materials presented in this paper are extracts from publicly available MARTOB reports. Please
visit our website:
where all detailed technical and scientific references may be found.

Mesbahi: Latest results from testing seven different technologies under the EU MARTOB project
Figure.1 Preparation of MARTOB Soup.
Figure 2. Objective assessment flowchart.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 3. High Temperature Thermal Treatment, Laboratory Scale.
Figure 4. Laboratory scale De-oxygenation technique.

Mesbahi: Latest results from testing seven different technologies under the EU MARTOB project
Figure 5. Laboratory scale equipment for UV and US systems.
Figure 6. Laboratory Scale Ozone treatment.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 7. Laboratory scale Oxicide treatment system.
Figure 8. Laboratory Scale Advanced Oxidation Technique.

The TREBAWA ballast water project
K. Hesse1, Mike Casey 2, P. Zhou, F. Aslan 3, E. Schmid, A. Leigh 4 & A. Santos 5
1Reederei Klaus Hesse GmbH
2Department of Naval Architecture and Marine
Universities of Glasgow and Strathclyde
mcasey@strath.ac.uk and
3Technologie-Transfer-Zentrum Bremerhaven
faslan@ttz-bremerhaven.de and
4Willand U.V. Systems Ltd tony.leigh@atgwilland.com.
5ISQ-Instituto de Soldadura e Qualidade
Treatment options being researched
The TREBAWA project is based on a primary mechanical treatment to remove larger organisms and
suspended solids, followed by an ultraviolet (UV) light irradiation to inactivate the remaining
organisms, disinfect the ballast water and make it suitable for discharge.
Timeframe of the project
TREBAWA is a European CRAFT project included in the 5th Framework Programme of the European
Commission started on the 1st of July 2002 and will end on the 31st of June 2004.
Aims and objectives of the project
Ballast water is of great importance for maintaining a ship's stability and limiting share forces and
tensions. During the loading of ballast water, large volumes of sediment are sucked from the water
columns or the harbour floor into the ballast tanks. The movement of some 10 billion tonnes of ballast
water in ships internationally each year has been responsible for the settlement of about 100 million
tons of sediments. Its cleaning and the disposal of the ballast sludge produced involve enormous costs,
as well as job hazards and time. Besides these economic aspects, ballast water has been recognised as
a major vector for the translocation of aquatic species across biogeographical boundaries, which may
prove ecologically harmful when released into a non-native environment. It is estimated that as many
as 3,000 alien species of plants and animals are transported per day in ships around the world.
The Marine Environment Protection Committee is working on developing draft new regulations for
ballast water management to prevent the transfer of harmful aquatic organisms in ballast water. The
working group has confirmed that ballast exchange on the high seas is the only widely used technique
currently available to prevent the spread of unwanted aquatic organisms in ballast water and its use
should continue to be accepted. However, it has been stressed that this technique has a number of
The conclusions are that development of alternative treatment technologies might produce techniques
that were substantially more reliable and that ballast water exchange is an interim solution. There is
considerable demand for efficient ballast water treatment alternatives, and thus, a great market

2nd International Ballast Water Treatment R&D Symposium: Proceedings
potential for the development of such a system, besides the environmental benefits that it would
TREBAWA is a European CRAFT project included in the 5th Framework Programme of the European
Commission. Its objective is the development of a new technically and economically competitive
ballast water treatment system, based on a primary mechanical treatment to remove larger organisms
an/or suspended solids, followed by a ultraviolet (UV) light irradiation to inactivate the remaining
organisms, disinfect the ballast water and make it suitable for discharge.
Seven European SMEs from every sectors of marine field will participate in this project in a
cooperative research, joining their capacities and expertise in the multidisciplinary areas involved in
the development of TREBAWA system. Meanwhile, three RTD performers will participate in order to
carry out the research and development of the system.
The experimental focus of the proposed work, consisting of 5 interdependent work packages, is a
pilot-scale system, which will be constructed to obtain the required data using in-situ analysis
techniques. It will then be installed on board in order to develop the field tests leading to the
validation of the prototype. Critical points are the achievement of:
a high degree of separation of in seawater suspended particles,
(ii) a
performance for the UV system in inactivating and killing all the in water remaining
(iii) an integrated prototype compact in size which fulfil the space requirements of a wide range
of existing ships,
(iv) an economically and operational efficiency for the final system.
Research methods, test protocols and experimental design
The main techniques which have already been assessed or are currently being investigated for
application in the treatment of ballast water are:
· Mechanical separation: is a good ballast water pre-treatment and should in fact be applied on
every ship to reduce residuals.
· Chemical separation: addition of chemicals is no good option in connection with negative
effects to the environment.
· Heat treatment: heating is a big energy consumer, what brings negative effects to the
surroundings; application of UV treatment appears to be an effective disinfection method. Its
application on board needs further research.
Most of these potential technologies haven't yet been demonstrated in a full-scale shipboard
It can be concluded that effective method can be achieved by using a combination of existing
technologies together, using the hydrocyclone as a primary treatment to remove larger organisms
and/or suspended solids, and UV as a secondary treatment, to inactivate the remaining organisms,
disinfect the ballast water and render it suitable to discharge.
The two most feasible treatment technologies for primary solids separation are filtration and cyclonic
separation. Filter has operational problems regarding high-pressure drops and a strong tendency
towards clogging and a low operational dependability.
In contrast to filtration, spin particle separation is a relatively simple and inexpensive way of
removing larger particles and organisms from ballast water.

Aslan: The TREBAWA ballast water project
Cyclonic separators are utilized in various industries such as chemical, coal mining and handling,
metal mining, rock products, plastics and wood products. Their relatively simple construction and
absence of moving parts mean that their capital and maintenance costs are lower than those of other
control devices that are available.
The most viable option for secondary treatment at the present time is considered to be ultraviolet
(UV) light irradiation. It has been the subject of laboratory testing on a range of marine organisms
with positive results and it has already been used in other marine applications for many years. Based
upon currently available information, UV radiation preceded by a primary clarification stage by
cyclonic separation appears to be the method that will provide the best combination of effectivity and
The research method includes computer modelling, laboratory tests, pilot tests and field evaluation.
The work programme has been formulated to address the key technical issues which must be
understood/solved in order to optimise the performance of the ballast water treatment system and
allow for its development. It includes fundamental hydrodynamic studies to measure the dispersion
coefficients and the separation performance as a function of critical parameters not available in
literature for solid ballast water components, such as particle size, aspect ratio and separator loading.
The subsequent data and the CAD/CFD (Computational Fluid Dynamics) model developed will
provide the basis for the design and detailed economic analysis of the system for commercialisation.
The experimental focus of the proposed work is a pilot-scale system which will be constructed to
obtain the required data using in-situ analysis techniques. It will then be installed on board in order to
develop the field tests leading to the validation of the prototype. The biological effectiveness of the
technology under consideration will be measured by comparing zooplankton, phytoplankton and
microbial concentrations with and without ballast water treatment. The deliverable of this project will
be the TREBAWA prototype for the on-board treatment of ballast water which is suitable for
demonstration on a wider basis.
Theoretical studies
Several commercial software tools have been successfully employed to carry out CFD in cyclonic
separator and UV chamber design. These include `PHOENICS'[1] `CFX'[2] and `Fluent'[3]. Fluent
was adopted for the current work and its effectiveness was assessed in meeting the main objectives.
Numerical meshes representing each of the separator and UV chamber geometries were created using
Fluent's GAMBIT preprocessor. Unstructured (tetrahedral) elements were employed to model the
geometry. These elements were found to model the separator and UV chamber components more
accurately than structured elements. The total number of computational cells making up a mesh
typically ranged from 150,000 for straightforward single pipe geometries up to 500,000 for more
complicated geometries. All the simulations were run on a Dell GX260 Pentium 4 PC machine and
the convergence times varied from two to several hours depending on the mesh size. The fluid flow
properties for each of the simulations involved typical values for sea water at 50F (=1027.9 kg/m3
and µ = 0.0014 kg/m-s).
A selection of separator geometries has been investigated. CFD predictions were carried out on
several centrifugal separators. Some designs were based on those previously constructed and
successfully employed onboard ship by OptiMarin [4].
The main objectives of the simulation work associated with the primary system were:
· To simulate the physical flow characteristics within various proposed designs and assess the
effects on the flow field when modifying various geometric components.
· To assess the efficiency of various designs in removing particles of varying size by
introducing them at the inlet and predicting their movement towards the outlet and drain exit.
· To provide the group with design recommendations based on the predictions obtained.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Ultraviolet light has been employed for many years to treat contaminated water due to its ability to
have a detrimental affect on the DNA of harmful micro-organisms that may be present. It is
essentially an environmentally friendly process without damaging side effects to the treated water. It
is for this reason that UV light has been considered as a viable treatment for ballast water for many
years. The main objectives of the secondary UV system investigation were:
· To produce a practical and efficient prototype UV treatment system to be employed in
conjunction with a primary separation system for use onboard ship.
· To carry out design optimisation by taking a wide range of possible UV system geometries
and simulating the physical flow characteristics within them using CFD.
· To focus on the most efficient system (or systems) and undertake construction and physical
testing of the prototypes under laboratory conditions.
· To undertake sea trials of the chosen prototype system.
Experimental tests
First ballast water characterization procedure was prepared in order to identify all the required
parameters for assessing the system performance. Analyses were made to the ballast water and the
ballast tank sediments. Those ballast water analyses were performed by diffraction of laser beam,
using samples from MV Jambo (Reederei Hesse) and from MV Roaz (Vinave).
An experimental platform has been designed and with the delivered cyclonic separators by OptiMarin
several tests have been carried out. The experimental platform was tested with two different feeds:the
harbour water in Bremerhaven and diatomaceous earth ( = 2.6 kg/m3).
The feeding water passes through the centrifugal pump that transforms the pressure charge into
velocity charge, allowing the centrifugal movement of the flow inside the separator. The feeding flow
rate can be controlled by means of the rotameter and the bypass, so that it can be fixed at the wished
value. The inlet flow passes through the cyclonic separator were it is separated in two streams: the
sludge (separated solids) and the outlet of clean water. This separation always involves a pressure
drop. The pressure is measured before the separator and after it, for both outlet streams, in order to
determine the pressure drop.
Water samples have been collected during each test for two different intermediate values of the flow,
after a stabilisation time: sample of clean water and sample of sludge. The samples were analysed in
the laboratory in order to determine the particles separation efficiency .
The results of the ballast water characterisation show (tables 1-2 and figures 1-4) that the average
particle diameter
for the samples analysed is placed in between 19 and 30 µm, while the values for
the sauter average diameter are placed between 8 and 11 µm. In general, those values are smaller than
foreseen and could present serious difficulties for the separation of the particles with the existing
cyclonic separators, which can not separate particles with low density, such as ballast water particles,
smaller than 50µm.
To ensure conditions, showing the characteristics of real ballast water, it was planned to conduct the
experimental tests in the harbour of Bremerhaven, Germany.
Different types of cyclonic separators are under experimental tests in order to improve the
performance of the separators as our aim is to remove the smaller particles.

Aslan: The TREBAWA ballast water project
Table 1. Average diameter size of samples collected in MS Roaz (5/11/02).
D[4.3] µm
D[3.2] µm
Sample 1- Bottle 1
Sample 2- Bottle 1
Sample 1- Bottle 2*
Table 2. Average diameter size of samples collected in MS Jambo (5/11/02)
D[4.3] µm
D[3.2] µm
BW Sample
Sediments Sample
D[4.3] volume average diameter and D[3.2] Sauter average diameter (diameter in the surface)
Regarding the performance of cyclonic separator, the first tendencies are:
· the efficiency of removing particles >150 µm is very good
· the performance of the separator decreases by particles <50 µm. The performance is between
15 ­ 40 %.
The experimental tests results comply with the findings of the simulated cyclonic separators.
Over 20 separator simulations and over 75 UV chamber simulations were carried out (see figures 5
and 6). Simulations were obtained for designs similar to those that are currently commercially
available and for designs that were proposed through collaboration with Willand UV Systems Ltd by
drawing on their expertise in UV chamber design.
The CFD design optimisation has provided a useful insight into the possible behaviour of the
separator and UV system when the geometry is modified. Modification is of little benefit however
unless a proposed design can be practically employed onboard ship. For example, a 15° separator
head angle necessitates a tall unit that is likely to be impractical for fitting onboard ship. Similarly, a
70° separator head angle would require a significant space requirement and material costs would
increase significantly compared to the original 35° design.
The experimental tests are still running as the most efficient separator and UV system has to be found
for the TREBAWA prototype. At this stage additional serious detailed results can not be published.
Developments on the most important two units have been carried out but both units have not been
tested together. After the implementation of the TREBAWA prototype more optimization will be
needed, which will be done by conducting tests and a CFD simulation of the whole prototype.
Conclusions and Recommendations
CFD was proved to be a useful tool in the design optimisation of a ballast water treatment system for
use onboard ship. The `Fluent' package was proved to be a robust and easy to use tool for the current
study. CFD has enabled the TREBABWA group to evaluate a wide range of centrifugal separator and
UV chamber designs under varying conditions and to focus on the most favourable thus avoiding the
need to construct numerous physical prototypes. This approach has helped to speed up the design
optimisation process and reduce the overall project cost. A number of points can be made from the
centrifugal separator optimisation:
· For separators to effectively move a large percentage of particles to the drain exit it is
essential that they are employed at the correct scale and at the correct flow rate when installed
onboard ship.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
· CFD has effectively highlighted possible practical improvements that can be made to the
original designs e.g. the movement of the flow restrictor and the positioning and size of the
drain outlet.
· The simulations have highlighted geometric modifications that are likely to be detrimental to
the separator design.
· The simulations have provided predictions of pressure drops for each design to enable
running costs to be estimated.
Several conclusions were drawn from the UV chamber optimisation:
· The steady state, RTD and UV intensity map calculations have indicated that the `inline'
designs are likely to be the most favourable for use by the TREBAWA group and represent
the most practical and economic solution for a UV chamber design.
· CFD has effectively highlighted possible unfavourable flow conditions such as short-
circuiting and areas of stagnation.
· CFD has successfully shown the effects on the flow caused by changing the chamber
geometry and by changing the lamp configuration.
· The simulations have provided useful comparisons of the pressure drop values between
different designs enabling a comparison to be made of the possible running costs of each
In summary the experimental- and simulation results have shown promising results for the further
development of the TREBAWA system. It is expected that the removal of the smaller particles can be
achieved through the compilation of experimental and simulation results for the construction of the
Holdo, A. 2001. Simulation of Ballast Water Treatment. 1st International Ballast Water Treatment
R&D Symposium, IMO London 26-27 March 2001: Symposium Proceedings
. GloBallast Monograph
Series No 5. IMO London. pp 151-161.
Wright, N.G. & Hargreaves, D.M. 2001. The use of CFD in the evaluation of UV treatment systems,
Journal of Hydrodynamics Vol 3, pp. 59-70.
Lawryshyn, Y. A. & Dong Ming, L. 1999. Reactor Design: Its more than putting a lamp in a pipe,
Water Conditioning and Purification
, February 1999.
Nilsen, B., Nilsen, H. & Mackey, T. 2001. The OptiMar Ballast System, 1st International Ballast
Water Treatment R&D Symposium, IMO London 26-27 March 2001: Symposium Proceedings
GloBallast Monograph Series No 5. IMO London. pp. 126-136.

Aslan: The TREBAWA ballast water project
Figure 1. Report of particle size distribution.
Samples collected in MV ROAZ (Aveiro Harbour, 05.11.02).

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 2. Report of particle size distribution.
Samples collected in MV ROAZ (Aveiro Harbour, 05.11.02).

Aslan: The TREBAWA ballast water project
Figure 3. Report of particle size distribution.
Jambo ship (Hesse) - Tank nŗ5 - 5/11/02 - BW Sample.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 4. Report of particle size distribution.
Jambo ship (Hesse) - Tank nŗ5- 5/11/02 - Sediments Sample.

Aslan: The TREBAWA ballast water project
Figure 5. Selection of Centrifugal Separator geometries.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 6. Selection of UV chamber geometries.

Some shipboard trials of ballast water treatment
systems in the United States
D. A. Wright 1, R. Dawson 1, T. P. Mackey 2,
H. G. Cutler 3 & S. J. Cutler 3;
1University of Maryland Center
for Environmental Science,
2Hyde Marine, Inc., USA
3Mercer University School of Pharmacy, USA
Full-scale ship trials of potential ballast water treatments were conducted aboard the Cape May, a
ship of the U.S. reserve fleet under the auspices of the U.S. Maritime Administration (MARAD). The
Cape May was berthed in Baltimore Harbor, near the Chesapeake Bay, on the east coast of the
United States. Two biocides and an ultraviolet (UV) irradiation system were each tested singly. With
correct dosing, biocide treatments resulted in the total eradication of zooplankton and phytoplankton.
All three technologies were capable of effective removal (>95%) of planktonic organisms without the
need for any primary treatment. This testing led to the development of two commercial ballast water
treatments, one using one of the biocides (SeaKleen®) and the other using the UV technology in
combination with a filtration primary treatment technology (tested separately).

Both systems will be further evaluated, both singly and in combination, in full-scale (3000 gpm) ship
trials aboard the Cape Washington, a MARAD vessel located in Baltimore Harbor. Particular
emphasis is being placed on demonstration of additive or synergistic effects of UV/biocide
combinations and the ramifications for cost effectiveness. In addition, primary treatment of filtration
to 50 microns is being examined to ascertain if economies may be realized in the use of biocide or
UV. Further evaluation of the filtration/UV combination will be tested aboard Princess Cruises'
Coral Princess.

Mechanical: Separation, Filtration
Physical: UV irradiation
Chemical: Biocides
Hybrid systems: Multiple treatments used in combination
Present phase underway from 2001-2005
The purpose of this project is determination of the efficacy of various ballast water treatment
technologies when used alone and in combination. Emphasis is on the identification of a
comprehensive, versatile, effective and economical system that is of immediate practical use to the
shipping industry.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The experimental design of this project is intended to evaluate various ballast water treatment
technologies under `real world' conditions. After preliminary evaluations are performed in a
laboratory setting, treatment technologies are placed aboard operational commercial vessels for
rigorous and thorough analysis. The work described here is designed to test the efficacy of a primary
mechanical treatment consisting of either separation or filtration, and secondary treatments consisting
of ultra-violet (UV) irradiation either alone or in combination with various biocides.
Laboratory analysis is conducted at the University of Maryland's Chesapeake Biological Laboratory,
a fully equipped research facility located in Solomons, Maryland. Shipboard trials are conducted
aboard a variety of vessels including vessels belonging to the U.S. Maritime Administration
(MARAD), and passenger cruise ships. Project vessels are selected for their potential to provide
ballasting conditions representative of typical shipping operations in both scale and difficulty.
The sampling design is based upon the following treatments: A) primary treatment only, B) primary
treatment + UV, C) primary treatment + biocide, D) UV only, E) biocide only, F) UV + biocide, G)
primary treatment + UV + biocide. The test system incorporates multiple sampling ports providing the
capability to simultaneously sample pre-treatment (raw) water, post-treatment water, and between-
treatment water. Water samples are drawn before and after each stage of treatment and directed to
triplicated banks of 200L polyethylene mesocosms, installed in the machinery space of the vessel, to
be held for 24 and 48 hours. All samples from the mesocosms are compared with triplicate untreated
controls. Treated and untreated water is also directed into different ballast tanks. The sampling and
biological endpoint determinations determined below are those used aboard the MARAD ship Cape
in 2001 trials. See Figures 1 and 2. Some modifications to these methods were made for cruise
ships and ongoing MARAD ship trials (e.g. no mesocosms: ballast tank sampling only).
Sampling protocols
Mesocosm and ballast tank samples
Mesocosms are filled from sampling ports located at strategic points in the treatment system and
designed to supply water of different treatment status. They are maintained under dark conditions to
simulate ballast water storage. Water samples are collected at 24 and 48 hour intervals following
initial ballasting/treatment. All mesocosms are thoroughly mixed with a compressed air wand
immediately prior to sampling. Ballast tanks are sampled either by lowering containers into the tanks
or using submersible pumps. A 1L sample is taken for phytoplankton analysis, and 260 ml. taken for
bacterial analysis (10 ml. for AODC and 250 ml. for culturable bacteria). The remainder (40-100 L) is
filtered (20 µm) for zooplankton analysis. Another sample is analyzed for standard water quality
parameters (salinity, temperature, pH, oxygen concentrations) as well as suspended particulate size
distribution analysis using an Accusizer laser obscuration particle analyzer. Control samples are
carried through the complete trial and analyzed for biological endpoints following dark storage.
Biological endpoint determinations
Samples for zooplankton analysis are concentrated to obtain approximately 50 organisms per ml.
subsampled in a circular, `one-time' counting chamber. Selected subsamples are preserved in
0.1% Lugol's solution for possible further taxonomic identification. Representative organisms
from at least 5 taxa are sized using a calibrated reticule eyepiece and records are made of
live/dead counts. The efficiency of a treatment is determined by comparing counts of taxonomic
groups with an untreated control sample. Overall removal efficiencies are calculated on the basis
of total number of organisms counted vs. controls.

Wright: Shipboard trials of ballast water treatment systems in the United States
Extractable chlorophyll a fluorescence determination in treated and untreated ballast water is the
core of the phytoplankton analysis. The fluorimeter used for analysis (Perkin Elmer 650-10 S) is
equipped with monochromators for specific wavelengths (436 nm ex., 680 nm emm.) that read
only emissions from chlorophyll a (Welschmeyer 1994). Whole water samples from ballast tanks
are prefiltered through a 200 µm screen to remove larger zooplankton. Post-treatment and control
water samples (500 ml.) are illuminated for 24 hours to examine capacity for cell doubling as
determined by repeated extraction and measurement of chlorophyll a. The instrument is calibrated
daily with accurate concentrations of chlorophyll a in ethanol.
Assessment of ambient bacterial populations are made using Acridine Orange Direct Counting
(AODC), together with standard plate counts of cultural bacteria. Colony counts are made on
treated and untreated samples following up to 72 hours incubation on Petri Pads impregnated with
marine broth.
ATP analysis
(introduced in 2003 for cruise ship tests and ongoing studies aboard the Cape Washington).
Adenosine triphosphate (ATP) is an obligate constituent of all living organisms and analysis of ATP
has long been performed by biologists to assess live (usually microbial) biomass. ATP is related to
total microbial biomass by determining ATP and applying a well established, laboratory-derived
conversion factor. Following death, ATP is rapidly converted to other phosphorylated compounds.
Therefore ATP would be expected to be at a maximum prior to UV inactivation. In these studies ATP
present in organisms is analyzed by filtration of the biota through a Whatman glass fiber filter and
immediate extraction with boiling Tris buffer. ATP levels in UV-irradiated and control samples are
determined using the firefly luminescence technique with the Deltatox analyzer operated in ATP
mode (Karl 1993).
Particle size analysis
Particulate size distribution is determined by Fraunhofer laser diffraction using an optical particle
sizer (Accusizer 770) fitted with a syringe injection sampler (Accusizer 770/SIS). Triplicate syringe
pulls of 25 ml. volume from a vigorously stirred beaker are counted and averaged to display the
particle size distribution either over the entire size range of the sensor (2 µm - 1000 µm) or selected
size ranges at 5 µm intervals. Particle concentration distributions are expressed as counts per ml. All
samples are analyzed aboard the vessel within two hours of sampling. The optical sensor is factory
calibrated with NIST traceable standards and may be recalibrated on site if necessary.
Laboratory studies
In parallel with 2003/04 shipboard trials, laboratory studies are being carried out to test the hypothesis
that combination of UV treatment and chemical biocide can lead to economies and benefits in the
treatment of ships' ballast water. A benchtop UV system is used to irradiate seawater samples and
organisms. The UV dose delivered at 254 nm is measured with a spectral radiometer. Calibration
points include the following doses: 35 mWatt sec-1 cm-2 ( the standard germicidal dose for many
bacteria), 50 mWatt sec-1 cm-2 ( usually sufficient to inactivate phytoplankton), 100, 150, and
200 sec-1 cm-2 (the range over which zooplankton and fish larvae are killed, based on data from
previous studies, including the Cape May trials). Test organisms include Vibrio fischeri, a
luminescent bacterium congeneric with Vibrio cholera, the basis of the Deltatox test (Azur
International), a toxic strain of the dinoflagellate Prorocentrum minimum, the estuarine copepod,
Eurytemora affinis, and sheepshead minnow (Cyprinodon variegatus) larvae.
Various analytical techniques are employed to determine degree of inactivation and/or mortality
associated with these tests including acridine orange direct counts, chlorophyll a, and ATP analysis.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
In parallel experiments, the chemical biocide SeaKleen® (primary active ingredient menadione) will
be tested with the same species and employing the same end-points to obtain dose response curves
and toxic thresholds. Once minimum lethal doses, of UV and biocide, have been established for the
different test species, a matrix of combinations will probe synergistic effects. The order of treatment
will also be a variable, i.e. experiments will be performed with UV irradiation either preceding or
following chemical biocide treatment.
Shipboard studies
Based upon results of laboratory investigations, shipboard tests are performed aboard the MARAD
ship Cape Washington, utilizing the most effective UV/biocide combination where UV irradiation
precedes biocide dosing and the most effective biocide/UV combination where biocide dosing
precedes UV irradiation. The experimental design follows that used in the 2001 shipboard
experiments, with the addition of the ATP determination described previously.
In the absence of a successful trial of a primary treatment system in 2001, a depth filter (Arkal Inc.),
supplied and installed by Hyde Marine Inc., will be tested as the primary treatment system. The
experimental design allows for evaluation of UV or biocide treatments with and without 50 µm depth
Note on centrifugal separator:
The centrifugal separator failed to operate during the 2001 test period. Reported to have been
corrected, the separator is scheduled for re-testing in the fall of 2003. The focus of the 2001
investigation shifted to a determination of the maximum efficacy that could be provided by the
secondary treatments alone. The question posed was: what was the minimum chemical or UV dose
required to achieve the maximum kill rate without the assistance of any other technology?
A summary of the major findings from the 2001 portion of the project is given below, although a
more detailed description of preliminary data appears in a November 2001 Interim Report to
Maryland Port Administration, and a more extensive report is in preparation.
Peraclean Ocean® experiments
Three experiments with Peraclean Ocean® were conducted during June 2001. The first experiment
applied Peraclean Ocean® at doses of 400 ppm and 200 ppm, which resulted in > 95%
zooplankton mortality. The second experiment applied Peraclean Ocean® at doses of 100 ppm and
50 ppm, which resulted in complete mortality of zooplankton with both doses. Repeat
experiments at doses of 100 ppm and 50 ppm, using different ballast tanks, resulted in overall
mesocosm mortalities of 98% and 100% at 100 ppm and 50 ppm respectively. See Figures 3a-f.
Overall ballast tank mortalities were 96% and 54% for 100 ppm and 50 ppm respectively. Much
of the inconsistency between experiments 2 and 3 was due to incomplete mortalities in protozoans
seen in experiment 3. This taxon was not recorded in experiments 1 and 2.
Exposure of phytoplankton to Peraclean Ocean® doses of 400 ppm and 200 ppm resulted in
complete bleaching of the chlorophyll a pigment with no growth whatsoever.
Success was achieved in controlling bacteria and will be presented in more detail by Dr. Fuchs at
this symposium. See Table 1.

Wright: Shipboard trials of ballast water treatment systems in the United States
Table 1. Peraclean Ocean® and culturable bacteria
Colony Counts of treated and untreated (control) ballast water samples.
(TNTC = too numerous to count)

Colony Count (24h)
Colony Count (48h)
Peraclean 400 ppm
Straight mesocosms
Straight ballast tank
Control mesocosms
Control ballast tank
Peraclean 50 ppm
Straight mesocosms
Straight ballast tank
Control mesocosms
Control ballast tank
SeaKleen® experiments
A synopsis of the results obtained with the biocide SeaKleen® is presented below; a more complete
data set appears in the aforementioned reports.
Three experiments on SeaKleen® were conducted during 2001. SeaKleen® was applied to
mesocosms and ballast tanks, at doses of 1ppm and 5ppm. Dosing at 5ppm SeaKleen® resulted in
complete mortality of all zooplankton examined after 24 hours. See Figure 4a. All developmental
stages of copepod crustaceans showed 100% mortality at 1 ppm SeaKleen® after 24 hours,
however other taxonomic groups showed incomplete mortality at this dose. See Figure 4c. Dosing
with 2 ppm SeaKleen® resulted in overall zooplankton mortalities of 99% and 100% after 24
hours and 48 hours, respectively. See Figure 4b.
SeaKleen® was shown to be effective in controlling phytoplankton irrespective of cell densities at
all concentrations tested. See Table 2.
Table 2. SeaKleen® toxicity to phytoplankton
Effect of SeaKleen® on phytoplankton.
Chlorophyll a expressed as µg Chl a L-1 ± S.D.

1 ppm
5 ppm
5 ppm

12.6 ±
34.2 ± 0.7
27.9 ± 2.5
9.7 ± 0.3
12.3 ± 0.4
5.1 ± 0.3
After 24h
30.6 ± 1.1
26.4 ± 2.6
6.8 ± 0.3
5.1 ± 0.3
6.5 ± 0.6
4.4 ± 0.6
Laboratory studies have shown that low ppm concentrations of menadione are effective in
controlling several microorganisms, including E.coli and Vibrio fischerei, as well as impacting
overall acridine orange direct counts (Cutler pers. com., Wright/Dawson Cape May results).

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Samples analyzed from mesocosms and ballast tanks, based on plate counts, did not present a
clear picture largely due to the latent release of bacteria from decaying phyto-zooplankton.
Ultraviolet irradiation experiments
Two experiments were completed at a nominal UV dose rate (200 mW sec-1 cm-2 at 1450 gpm).
The ambient estuarine water in Baltimore Harbor was measured for UV (254 nm) transmission
periodically, and was always in excess of 90%, measured over a 1 cm path length. True dosing
was seen to be closer to 180 mWatt sec-1 cm-2, when corrected for UV transmission). At this dose
zooplankton mortalities averaged better than 95% in both mesocosms and ballast tanks. See
Figures 5a-d.
UV irradiation at 180 mWatt sec-1 cm-2 resulted in greatly reduced phytoplankton growth relative
to unexposed control water. Results are presented for water samples taken immediately following
treatment and for samples taken after a 24-hour illuminated grow out period. See Table 3.
Table 3. UV treatment of phytoplankton
Effect of UV irradiation on phytoplankton.
Chlorophyll a expressed as µg Chl a L-1 ± S.D.

UV 0h
UV 24h
B e f o r e f l u o r e s c e n t
6.1 ± 0.1
5.0 ± 0.2
5.6 ± 0.2
4.2 ± 0.2
After 24h fluorescent
9.2 ± 0.2
7.5 ± 0.1
4.6 ± 0.1
3.6 ± 0.1
Counts of cultural bacteria were usually much higher in UV-treated samples compared with
untreated (control) samples. We concluded that the increase in bacterial numbers resulted from
the destruction of phytoplankton and zooplankton providing increased nutrient levels for both
free-living bacteria and endogenous bacteria released from decomposing metazoan organisms.
See Table 4.
Table 4. UV and culturable bacteria
Colony Counts of UV-treated and untreated (control) ballast water samples.
(TNTC = too numerous to count)

Colony Count (24h)
Colony Count (48h)
Test No. UV1
UV-treated mesocosms
ca. 5000
ca. 3000
UV-treated ballast tank
ca. 2000
ca. 1330
Control mesecosms
Control ballast tank
ca. 2000
Test No. UV2
UV-treated mesocosms
UV-treated ballast tank
Control mesecosms
Control ballast tank

Wright: Shipboard trials of ballast water treatment systems in the United States
Conclusions and recommendations
There have been two distinct classes of biocides proposed to treat ballast water. These can best be
described as inorganic oxidants (H2O2, Cl2, peracetic acid, Br, O3, K2S2O8) and organic oxidants such
as quinones (e.g. menadione; active ingredient in SeaKleen®). The first class can be classified as
indiscriminate oxidants because they react with carbon and organic matter in general and, to varying
degrees, with metals. Peracetic acid (Peraclean Ocean®), used in trials aboard the Cape May, is a good
example of the first class. While it was evident that in mesocosm tanks (plastic), levels of biocide of
100 ppm and lower were effective in controlling zooplankton, data from ballast tanks (metal)
indicated a slightly lower degree of effectiveness and, analogous to chlorination practices, it is clear
that residual oxidant levels are important in metal ballast tanks. Organic oxidants, such as the
menadione (vitamin K3) used in the SeaKleen® formulation, are selective for cellular structures and,
via futile redox cycling reactions, repeatedly oxidize tissues and membranes (O'Brien 1971). In fact,
many such organic oxidants, particularly the quinones, have been investigated in tumor and cancer
research because of their unique targeting properties. SeaKleen® does not seem to be consumed via
oxidation of metals and corrosion studies (results not included here) have shown SeaKleen® in
seawater to be no more corrosive to bare steel than seawater alone.
Measurement of growth potential through chlorophyll a determination represents a robust and
convenient method for assessing the efficacy of a particular ballast treatment in controlling natural
phytoplankton assemblages. While a healthy phytoplankton population might be expected to double in
a 24-48 hour period, a substantially lower rate of doubling indicates inhibition resulting from toxicity
or a limiting resource. In the experimental design used here, untreated samples serve to control for
any limiting resource unrelated to the ballast water treatment itself. However, dark exposure itself
probably contributes to lack of growth potential depending on the initial densities of phytoplankton.
Interestingly, control water samples taken after 24 hours and subjected to further irradiation showed
no sign of growth. In the SeaKleen® experiment reported here, control phytoplankton standing crop
declines 18% and 42% between 24 and 48 hours, respectively. Control samples collected at 24 and 48
hours showed no capacity for growth under fluorescent light (10.6% and 5.3% loss in chlorophyll a
respectively). It may be concluded from these results that, even without treatment, a dark ballast tank
represents an inhospitable environment for free-living phytoplankton. And certainly, following
biocide treatment, natural phytoplankton populations are much more severely inhibited than controls.
Also, there is little difference between the effects of SeaKleen® on phytoplankton whether sampled in
ballast tanks or dark mesocosms.
UV Irradiation
UV irradiation has been the subject of laboratory and pilot testing on a range of marine organisms
with relatively positive results (Wright and Dawson, 2000). UV works by damaging parts of
organisms' DNA. The biological effect depends on the dose, expressed as mWatt sec-1 c m-2 and is
dependent on power, exposure surface, flow rate and distance from the UV source. With the correct
dose of UV, viruses, bacteria, and most types of zooplankton and phytoplankton can be killed or
rendered nonviable.
The results of UV irradiation testing, without pretreatment, indicate that medium pressure UV
(200 mWatt sec-1 cm-2) reduced viable zooplankton by better than 95%. Phytoplankton growth was
arrested relative to untreated control water. With the addition of effective pretreatment to remove
larger, more UV resistant organisms, consistent mortality of organisms in excess of a 95% standard
should be possible.
Development of commercial Ballast Water Treatment (BWT) systems
As a result of the research and testing on both SeaKleen® and medium pressure UV irradiation
conducted aboard the Cape May , and of separate full-scale testing of solids separation and UV

2nd International Ballast Water Treatment R&D Symposium: Proceedings
systems and components aboard several ships and a barge, two commercial ballast water treatment
systems have been developed. The SeaKleen® biocide is undergoing EPA registration and is awaiting
testing approval from several states in the U.S. It is expected that full-scale testing on operating tank
vessels will begin in 2003. A Hyde Marine BWT system, incorporating disk filtration (Arkal) and
medium pressure UV (Aquionics), was recently installed aboard Princess Cruises' Coral Princess.
Testing of the system is expected to commence in the fall of 2003.
Performance of shipboard BWT systems is based upon biological effectiveness, cost effectiveness,
and adaptability to the ballast pumping and piping systems on ships. SeaKleen®, is intended for ships
with large ballast water volumes such as bulk carriers and tankers. The filtration and UV irradiation
BWT system is intended for ships with ballast water flow rates up to approximately 1000 m3/hr.
Larger flow rates are possible, but the system would consume a very large amount of electrical power
and could be subject to space limitations. The development of both systems continues as part of a
University of Maryland research program designed to improve both performance and economic
effectiveness. Testing of SeaKleen® will include determining the benefits of pretreatment by filtration,
UV irradiation and other technologies to reduce the chemical dosage required. Testing of the filtration
and UV system will attempt to determine the most synergistic combination of filtration level and UV
dose to improve efficiency and adaptability to new and existing ships.
Development of SeaKleen® treatment system
SeaKleen® can be used without pretreatment. Testing to date has indicated it will be effective at
dosage rates of 1.5 ppm to 2 ppm. Application requires mixing with water just prior to use, as
biodegradation of SeaKleen® begins as soon as it is mixed with water. Injection of precise amounts of
SeaKleen® would be accomplished using conventional dosing pumps to feed the chemical at
proportional rates to the ballast water flow into the ballast pump(s) discharge piping. Sediments do
not greatly affect the performance of SeaKleen® (particularly compared with indiscriminant inorganic
oxidants and hydrophobic compounds) and it is not corrosive. No special materials are required for
the dosing system, and many ships already meter chemicals into their ballast tanks. The
environmental degradation of SeaKleen® has been the subject of several investigations (and will be
presented by Dr. Cutler) and was a regulatory requirement of the Cape May study. See Figures 9a-f.
It is expected that the cost of SeaKleen® will be 10-20 cents per tonne of ballast water treated. For
ships with very large ballast volumes and relatively frequent ballasting and deballasting, this could
lead to significant costs over the lifetime of the ship, in addition to a low initial capital expenditure.
Future research on the application of SeaKleen® will focus on pretreatment methods to reduce the
required chemical dosage. Shipboard trials of SeaKleen®, aboard bulk carriers working in Puget
Sound, Washington, are planned for late 2003.
Development of filtration and UV systems
Independent of the Cape May testing, three full-scale Ballast Water Treatment (BWT) Systems were
installed on cruise ships, one on a Panamax container ship and one on a parcel tanker during 2000 and
2001. These systems ranged from 200 to 350 m3 hr-1 ballast water flow rate and included cyclonic
separation of solids as a pretreatment during ballasting and low pressure UV treatment (100 to 125
mWatt sec-1 cm-2) during both ballasting and deballasting. Testing both on board ship and aboard the
Great Lakes Ballast Technology Demonstration Project (GLBTDP) test barge indicated that cyclonic
separation is not an effective means of removing larger organisms from ballast water. Additional
information on such systems may be found in: Parsons and Harkins 2001, Parsons 2003; Cangelosi et
al. 2001; Mackey 2001. The current design philosophy for modern UV/filtration systems is further
described in Mackey and Wright (2002).
There has been a widely held belief that particle separation by filtration or centrifugal separation may
improve UV transmission and several projects have sought to compare the effect of UV treatment
with and without separation. Turbidity is rarely the limiting factor for UV transmission, although it
can be during an extreme red tide event. Molecular absorbance of dissolved organic matter, on the

Wright: Shipboard trials of ballast water treatment systems in the United States
other hand, can have dramatic impact on UV penetration (ex: Duluth Harbor) and no amount of
physical separation will improve transmission. Efforts to remove particulates using back-flush filter
screens, voraxial separators, or hydrocyclones have shown little or no success. Efforts to reduce
turbidity in hopes of aiding secondary treatment of either UV or biocide, have not demonstrated
significant results. Several modern biocides have a low affinity for binding with particulates, therefore
the argument that turbidity consumes these biocides seems misguided. No currently available
separators remove small zooplankton or phytoplankton.
In the case of UV treatment, there does not seem to be a strong correlation between UV treatment
effectiveness and organism size, however, shelled organisms (bivalves) and those with highly
protective carapace pigments (some crustaceans) are noticeably more resistant. The argument may be
made that effective filtration (e.g. down to 50 µm) might remove a substantial fraction of organisms in
this category, however, attempts to demonstrate complimentary benefits of physical separation and
secondary treatment have been confusing, given the current state of available technology. It is also
clear that a quantum leap in the performance of UV systems is necessary to effectively and
economically achieve the doses required to treat ballast water flow rates in excess of 3,000 gpm. In
the future, this may be possible with the use of higher efficiency excimer UV systems (Coogan et al.
1999) but expense remains an issue, particularly if multiple units are employed. The current test
platform, MARAD's Cape Washington, has ballast pumps rated at 3,000 gals min-1. These pumps will
be used in 2004/05 to test the efficiency of two 32 kW UV systems mounted either in series or in
parallel. In series mounting will test the as yet unproven concept that UV irradiation can be additive
where higher ballasting rates and multiple systems are involved.
Combination UV/biocide treatment systems
To our knowledge, there have been no serious attempts to combine UV and chemical biocide
treatments in order to examine synergistic or complimentary effects, although waste-water and pulp
mill effluents have been successfully treated with a combination of UV and peroxide. This
combination increases the oxidation potential via formation of hydroxy radicals. In the case of organic
biocides, we have a little or no information as to whether a UV toxic stress to an organism results in
enhanced susceptibility to chemical toxicity or conversely if sublethal biocide toxicity compromises
an organism's susceptibility to UV. An additional benefit of a combination biocide and UV treatment
may result in an enhanced degradation of the biocide post treatment and, hence, a reduction in the
holding time prior to safe discharge. In view of foregoing considerations, a comprehensive appraisal
of UV/biocide combinations is the focus of our 2003/2004 studies. We hypothesize that exposure to
one treatment may sensitize organisms to the other treatment and that such combination treatments
may offer economies in both UV dose and biocide application. Initial laboratory studies are well
underway at the time of writing. Planned ship-scale test platforms include two MARAD vessels; the
Cape Washington, recently returned from the Middle East and currently berthed in Baltimore and a
fully refurbished liquid containment barge, presently located in Virginia but available for relocation to
Cangelosi, A et al., GloBallast Symposium and Workshop Submission, March 26-30, 2001.
Cooghan, J.C. Byssing, A., Morgan, G.L., Barracato, J., Dawson R. & Wright D.A. 1999. U V
disinfection of ballast water: effects of organism size on system scaling
. National Conference on
Marine Bioinvasions. Massachusetts Institute of Technology, Cambridge, MA. January 24th - 27th,
Karl, D.M. 1993. Total microbial biomass estimation derived from the measurement of particulate
adenosine-5?-triphosphate. In: Kemp, P.A., Sherr B.F. & Cole J.J. (eds.). Handbook of Methods in
Aquatic Microbial Ecology
. Lewis Publ., Boca Raton, Fl.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Mackey, T.P. 2001. Ballast Water treatment Technologies: Including a review of Initial Testing and
Lessons Learned Aboard the Regal Princess, Marine Environmental Technology Symposium 2001:
Engineering for a Sustainable Marine Environment
CD ROM Session A2, May 31 ­ June 1, 2001.
Mackey, T.P. & Wright, D.A. 2002 A Filtration and UV based ballast water treatment technology:
Including a review of initial testing and lessons learned aboard three cruise ships and two floating
test platforms
, ENSUS 2002, Marine Science and Technology for Environmental Sustainability, Univ.
of Newcastle upon Tyne UK and EU MATOB Project, 16, 17 and 18 December 2002.
O'Brien, P.J. 1971. Molecular mechanisms of quinone toxicity. Chem. Biol. Interactions. 80, pp. 1-
Parsons, M.G. & Harkins, R.W. 2002. Full-Scale Particle Removal Performance of Three Types of
Mechanical Separation Devices for the Primary Treatment of Ballast Water, Marine Technology, Vol.
39, No. 4, pp. 211-222.
Parsons, M.G. 2003. Considerations in the Design of the Primary Treatment for Ballast Systems,
Marine Technology, Vol. 40, No. 1, pp. 49-60.
Welschmeyer, N.J. 1994. A fluorescence based approach for determination of chlorophyll a without
interference from phaeopigments. Limnol. Oceanogr 39: pp. 1985-1992.
Wright, D.A. & Dawson, R. 2000. International control of non-indigenous species. Presentation at
3rd International Congress of the Society for Environmental Toxicology and Chemistry. Brighton,
England. May 21st - 25th, 2000.

Wright: Shipboard trials of ballast water treatment systems in the United States

Figure 1. Cape May.
Figure 2. Cape May.
Mesocosms Treated with 200 ppm Peraclean
Ballast Tank Treated with 100 ppm Peraclean
24 hr.
48 hr.
24 hr.
48 hr.
% Mortality
% Mortality
opepod -A
opepod -A
Ballast Tank Treated with 50 ppm Peraclean
Mesocosms Treated with 50 ppm Peraclean
24 hr.
48 hr.
24 hr.
48 hr.
% Mortality
% Mortality
opepod -Adult
Copepod -Adult
Control Mesocosm after 24/48 Hours
Ballast Tank Control after 24/48 Hours
24 hr.
48 hr.
24 hr.
48 hr.
% Mortality
% Mortality
Ostracods Protozoans
opepod -A
Copepod -Adult
Figure 3. Effect of Peraclean Ocean ® on zooplankton.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Mesocosms Treated with 2 ppm Seakleen
Mesocosms Treated with 5 ppm Seakleen
24 hr. 48 hr.
24 hr.
% Mortality
% Mortality
ae L
opepod-A C
Bivalvae Larvae
Control Mesocosms after 24/48 Hours
Mesocosms Treated with 1 ppm Seakleen
24 hr. 48 hr.
24 hr. 48 hr.
% Mortality
% Mortality
ae L
ae Larvae stracods
opepod-A C
Ballast Water Tank Control after 24/48 Hours
Ballast Water Tank Treated with 2 ppm Seakleen
24 hr. 48 hr.
24 hr. 48 hr.
% Mortality
% Mortality
Copepod -A
Bivalvae L
Copepod -A
Bivalvae L
Figure 4. Effect of SeaKleen® on zooplankton.

Wright: Shipboard trials of ballast water treatment systems in the United States
Control Mesocosm after 24/48 Hours
Mesocosms Treated with UV
24 hr. 48 hr.
24 hr. 48 hr.
% Mortality
% Mortality
atodes Rotifers
Ballast Tank Control after 24/48 Hours
Ballast Water Tank Treated with UV
24 hr. 48 hr.
24 hr. 48 hr.
% Mortality
% Mortality
epodite N
atodes Rotifers
Figure 5. Effect of UV on zooplankton.
Figure 6. Coral Princess.
Figure 7. Arkal Galaxy Filter, Coral Princess.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 8. Aquionics UV Chamber, Coral Princess.
Percent Survival, for Paleamonetes pugio, after 96 Hour Exposure
Percent Survival, for Palaemonetes pugio, after 96 Hour Exposure
% Survival
% Survival
% Survival
% Survival
% Seakleen (5 ppm)
% Seakleen (1 ppm)
Percent Survival, for Neomysis americana, after 96 H
Percent Survival, for Neomysis americana, after 96 Hour E
% Survival
% Survival
% Survival 20
% Survival 20
% Seakleen (Mixture of 2, 2.5, 5 ppm)
% Seakleen (2.5 ppm)
Percent Survival for Cyprinodon variegatus, after 72 Hour Exposure
Percent Survival for Cyprinodon variegatus, after 72 Hour Exposure
% Survival
% Survival
% Survival
% Survival
% Seakleen (Mixture of 2, 2.5, 5 ppm)
% Seakleen (2 ppm)
Figure 9. Regulatory discharge survival assay.

Wright: Shipboard trials of ballast water treatment systems in the United States
Figure 10. Cape Washington.
Figure 11. Liquid containment barge.

Development and design of process modules for
ballast water treatment on board
A. Kornmueller
Berkefeld Water Technology, Germany
Treatment options being researched
Various processes have been suggested for ballast water treatment (BWT) in the last years (IWACO
2001). Overall solutions are hard to meet due to the complexity in design specifications, which are
caused predominantly by the variation in the water quality, the technical demands and the specific
requirements by different vessels. Berkefeld Water Technology and its subsidiary RWO Marine
Water Technology are developing new ballast water treatment systems for use onboard ships, which
include particle separation and disinfection steps. Intentionally a concentration on just one treatment
solution is avoided. Different treatment options are investigated, which enables a modular design and
adaptation to each kind of vessel in accordance with the biological, chemical and technical
constraints. Besides the prevention in the introduction of harmful aquatic species, the BWT solves the
problem of sediment accumulation in ballast water tanks by a mechanical separation as the first step.
Timeframe of the project
In the scope of the program "Shipping and marine technology for the 21st century" the R&D-project is
under investigation since Oct. 2002 and will be funded by the Ministry for Research and Technology
(Germany) until the end of 2004.
Aims and objectives of the project
The R&D-project is entitled "Basic examinations of the biological, chemical and physical
characteristics and loadings of ballast water and the design of process modules for its treatment and
disinfection onboard". The aim of the project is the development of efficient and cost-effective
modular process combinations for ballast water treatment onboard. Therefore, a parallel approach is
applied by basic evaluation and practical examinations. The basic evaluation includes the
biological/aquatic and chemical/physical water characterization, and sets a special focus on the basic
conditions by different vessels (such as type, construction and operation). A comprehensive survey of
these influencing parameters results in the definition of requirements on BWT and the development of
system specifications. After the identification and comparison of different treatment options, which
are available in the market and research, suitable processes are studied experimentally for sediment
and organism removal followed by disinfection. From the beginning a modular design is considered,
which is a precondition for the adaptation of BWT systems to each kind of vessel.
Research methods, test protocols and experimental design
The following methodologies are applied in preliminary basic evaluation, which includes the desk
based review like:
· literature and internet inquiry,
· communication with organizations, authorities, research institutes and companies,
· field examinations (like ballast water sampling) to complement existing data

Kornmueller: Development and design of process modules for ballast water treatment on board
· and the assignment of different research institutes and companies to attribute their special
Within the scope of the experimental examinations a test plant for sediment removal is designed,
which will be upgraded stepwise to different BWT options and described later in detail. Two parallel
test lines are installed to examine the reproducibility of the same equipment and the direct comparison
of different processes. A significant statistical component is necessary in the test procedure and water
analysis for the assessment of test data. In particular, time profiles have to be determined, because the
water used as influent to the plant is varied by the tide. Due to the different water qualities and salt
contents found globally, the performance of the test plant has to be verified at different locations
reflecting these variations.
This project takes part in an expert group of participants from German R&D-projects, which works on
land-based type approval tests for BWT systems (Voigt at al. 2003). Certainly its outcome resulted in
our own testing protocols, such as sampling procedure, selection of test organism and test water
Ballast water treatment is limited by specific characteristics and conditions. Its complexity is given by
the global variation in the biological and chemical/physical water characteristic, the high technical
demands and efficiencies, the requirements by vessel construction and operation as well as different
national and international regulations and legislations.
Chemical and physical water characterization
Very small information is available on the chemical and physical characteristic of water used as
ballast. This includes fresh, brackish and sea water depending on the location of ballasting and the
vessel route. The concentrations of some water parameters are known by en-route and end-point
sampling of ballast water tanks or by monitoring of harbour water. A comprehensive and publicly
available database does not exist yet. Monitoring programs often concentrate only on few, mainly
physical water parameters. Therefore it has to be stressed, that various water parameters and
impurities may have an influence on the performance of different treatment processes. For example
the iron content is normally not included in monitoring programs, but has an important influence on
the efficiency in disinfection due to precipitation on UV-lamps. Even extreme water conditions, like
algae blooms or dispersed sediments by vessel propellers nearby intake, have to be accomplished by a
BWT system. Therefore, the influence of different water parameters is defined on the performance of
possible treatment processes within this R&D-project. Concentrations of decisive water parameters
are collected from the literature and authorities to give a comprehensive, global overview in their
variations and to define limitations on treatment performances. This data collection is ongoing and
will be published later.
Biological water characterization
The biological basics of organisms and their invasion have been under investigation for years. The
influences of the diverse organisms properties are important on the performance evaluation of
different treatment processes. While the size distributions of main organisms are known and compiled
(e.g. IWACO 2001), only single data are available on the concentration of organisms in water used as
ballast. Based on these organism densities and sizes, limit values have to be concluded for
performance standards. These standards are urgently needed and discussed internationally at the
moment. This is an international task, which cannot be solved by a single R&D-project. Nevertheless,
from a plant manufacturers perspective the knowledge about organism properties and standards is
decisive for designing BWT systems and for achieving its future required overall efficiency.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Requirements by vessel
In this R&D-project one main focus is set on the requirements by various vessels, which comprise the
vessel routes and therefore the water quality ballasted, the vessel operations and construction of the
vessel itself and its ballast water system. Figure 1 shows a comparison in vessel number of passenger
ships, oil tankers and bulk carriers over the dead weight tonnes (dwt) ­ size class. While passenger
ships feature a high vessel number at low dwt, oil tankers show a more even distribution over the
whole dwt-range. In opposite bulk carriers have a main maximum in vessel number at 30,000 dwt ­
size class. In accordance with the different vessel sizes and transport functions a great variation can be
found in the layout of the ballast water system, vessel routing and operation. Passenger ships have a
more flexible routing, which often occurs near the coast at a regional mode, while oil tankers show a
fixed and recurred routing on a global scale. Ballast water tanks vary in number and sizes having
different total ballast water capacities and flow rates at ballasting/deballasting. The later normally
range from around 200 m3/h for cruise ships up to 6,000 m3/h for Very Large Crude Carriers. A cruise
ship is continuously ballasting at a low rate in order to compensate for the consumption of
consumables (like fuel). Because of its mode of operations at touristy sites, the water quality used as
ballast water is better compared to the intake of harbour water by other kind of vessels, which are
ballasting at the time of unloading.
Conception and design of treatment options
In general three different overall options can be distinguished for the BWT onboard:
· Treatment at intake during ballasting
· Treatment during voyage
· Treatment at discharge during deballasting.
If the treatment concentrates solely on the prevention of non-indigenous species, a treatment at intake
would be the right one and most effective to choose. Additionally the BWT offers the possibility to
avoid simultaneously the problem of sediment accumulation in ballast water tanks. Sediment is
deadweight, which makes the trim and stability of a vessel harder to rate, and causes additional costs
due to loss of cargo, energy consumption and tank cleaning at shipyard. Additionally sediments in
ballast water might hide and protect smaller organisms from disinfectants. Therefore, a BWT
including a mechanical separation of particles during intake provides many advantages for the vessel
operation and maintenance and the prevention of biological invasions. Besides the inorganic (mostly
sediment) particles a part of the organisms is likewise removed by mechanical separation. This
facilitates the following disinfection step resulting in advantages like a lower disinfectant dose and
footprint. Likewise the disinfection step can be carried out as an in-line treatment during ballasting or
can be operated later during the voyage in a bypass to the ballast water tanks. The second options has
the advantage, that lower flow rate can be applied at longer treatment times. Unfavourably
inhomogeneities and incomplete mixing effects result in unstable and insufficient treatment
efficiency. These are caused by the present design of ballast water tanks, which contain a large
number of fixtures for the structural strength of tanks, like longitudinals, intercostals and floors
(Taylor and Rigby 2001).
A possible design of a ballast water treatment system is shown exemplary in Figure 2 for a container
vessel. The treatment modules are set in the main ballast water line, whereas each can be located
independently. For ship safety the BWT system can be bypassed. The treatment takes place during
ballasting using only the two ballast water pumps. For discharge via valves in the hull the stored
ballast water is pumped out of the tanks using the ballast water pumps again. As discussed in details
below, an additional disinfection might be affordable at deballasting, if the disinfectant used during
ballasting does not prevent a regrowth of organism in the ballast water tanks during voyage.
Like mentioned above, some vessels run on fixed routes giving an easier assessment of the water
quality to be treated onboard. Ballast water treatment on vessels with more flexible routes affords a
higher reliance in treatment efficiency and therefore a more redundant and effective treatment

Kornmueller: Development and design of process modules for ballast water treatment on board
combination, which can be obtained by installing modular multi-step processes. The basic treatment
by an one-step mechanical separation followed by a disinfection can be upgraded in the case of higher
loadings in water quality. Another separation module can be integrated in the basic two-step BWT
giving relief to the original one. By applying a different second disinfectants a short- and long-term
deactivation or a supplementary chemical oxidation can be combined. Depending on the
characteristics of the disinfection process used two different process options are possible. A
disinfectant, which provides a depot effect in the ballast water tanks, can guarantee a continuing
disinfection until discharge. In the other case a replicated disinfection is necessary at deballasting due
to a possible regrowth of organisms during voyage.
The potential for various treatment processes is restricted by some constraints, like the high flow rate
and volume to be treated and the low footprint available at ships. Therefore all basic processes in
mechanical, chemical and physical treatment are identified quantitatively by specific parameters, like
· maximal volume flow rate being treatable
· footprint and height
· removal efficiency
· energy consumption
· acquisition costs
· operation expenses.
As an example, some data of different filter types, which are theoretically applicable in the first
treatment step of mechanical separation, are shown in Table 1.
Table 1. Example of some specific parameters for different filters used in the quantitative process assessment
(based on a volume flow rate of 500 m3/h).
Filter type
Specific velocity [m/h]
p [bar]
Overall height [m]
Pressurized gravel filter
15 - 30
0.1 - 0.5
4.5 - 5.5
Cartridge filter
5 - 40
0.2 - 2.5
2 - 3
Edge filter
400 - 600
0.5 - 1.0
1.5 - 2
Disc filter
100 - 200
0.5 - 1.0
After quantitative data collection and the consideration of qualitative parameters, like safety and
environment issues, a comprehensive survey is carried out by combining all relevant biological,
chemical, physical and technical parameters for designing BWT systems. Hereby, a setting of
priorities is necessary. For example, the effectiveness of a sufficiently long sand passage is known for
cyst removal from drinking water, but sand filtration is not a treatment option for shipboard use due to
the enormous space and time requirement.
After the identification and comparison of different treatment options, which are available in the
market and research, suitable processes are studied experimentally for particle and organism removal
and disinfection. Intentionally alternative processes are examined and optimised for mechanical
separation and for disinfection, which can be used as complementary or replacing components and
facilitate a modular design.
In a test plant the performance of single and combined processes will be optimised including the
above mentioned specific parameters. For instance, filtration using a filter fineness of 20 µm is
effective in removing zooplankton and the cysts of toxic dinoflagellate algae (Oemcke and van
Leeuwen, 2003), but certainly smaller organisms like phytoplankton, bacteria and virus are not
removed in this step. Published results of experiments using different filters indicate, that the lowest
achievable filter fineness is around 50 µm for BWT at this time (Parsons and Harkins 2003). An

2nd International Ballast Water Treatment R&D Symposium: Proceedings
optimum has to be determined between the lowest filter fineness giving relief to the post-disinfection
and the increasing footprint of the filtration system with decreasing filter fineness.
At the end of June 2003 a test plant was started to run on river water in order to investigate different
mechanical separation options (results will be available later). Focussing on sediment removal the
performance of different suitable filter are studied first of possible options. Later the pilot plant will
be upgraded stepwise to test and optimise different BWT options followed by designing and testing
Conclusions and Recommendations
· BWT systems exhibit a complexity in design specifications, which has to compromise the
requirements by biological, chemical/physical and technical parameters together with the ones
by different vessels (such as type, construction and operation).
· Therefore different optimised processes should be available in BWT design giving the choice
for the most adequate and efficient treatment system in accordance with the requirements.
· For onboard treatment an adaptation of BWT systems to each kind of vessel has to be
achieved by its modular design.
The research project is funded and supported by the Federal Ministry for Research and Technology of
Germany under registration number 03SX169.
ISL. 2001. Shipping Statistics Yearbook 2001. Institute of Shipping Economics and Logistics (ISL).
Bremen. December 2001.
IWACO. 2001. Standards for ballast water treatment. Report for the Ministry of Transport and Public
Works, North Sea Directorate. Netherlands. February 20, 2001.
Oemcke, D. J., van Leeuwen, J. (H). 2003. Chemical and physical characterization of ballast water ­
part I: Effect on ballast water treatment processes. Journal of Marine Environmental Engineering Vol.
7. 47 - 64.
Parsons, M. G., Harkins, R. W. 2003. Full-scale particle removal performance of three types of
mechanical separation devices for the primary treatment of ballast water. Marine Technology Vol. 39
­ No. 4. pp. 211 ­ 222.
Taylor A. H., Rigby G. 2001. Suggested designs to facilitate improved management and treatment of
ballast water on new and existing ships. Proceedings of the First International Conference on Ballast
Water Management ­ Best Practices and New Directions. November 1-2, 2001. Singapore.
Voigt, M., Gollasch, S., Kornmüller, A., Röpell, H., Kerschek, O., Bahlke, C. 2003. Proposed land-
based type approval tests for ballast water treatment systems. Oral presentation at the International
Conference and Exhibition on Ballast Water and Waste Water Treatment Aboard Ships and in Ports,
June 11-13, 2003, Bremerhaven, Germany.

Kornmueller: Development and design of process modules for ballast water treatment on board
Figure 1. Vessel number of passenger ships, oil tankers and bulk carriers over dwt-size class (based on statistics
by ISL 2001).
Figure 2. Exemplary design of a ballast water treatment system on a container vessel providing full treatment at
ballasting and additional disinfection at deballasting.

Hydrodynamic transonic treatment and filtration of ship
ballast water
A. Andruschenko, A. Dukhanin, V. Rabotnyov,
Y. Skanunov & S. Tishkin
A well-balanced and thoroughly researched choice of method, or complex of methods, for treating
ships' ballast water is an important goal. Surveys of work previously done in this field, as published in
the Ballast Water Treatment R&D Directory August 2002 materials, do not cover all possible methods
for treating ballast waters.
The goal of our research is to develop a technical solution for treating ship ballast waters that would
comply with criteria set forth in the materials of the Marine Environment Protection Committee 48th
(MEPC 48 / WP.15.10 October 2002. Regulations E-1 ... E-4).
This research is being conducted under the auspices of the GEF/UNDP/IMO/GloBallast Program.
The goals of the research are:
developing and putting together a pilot version of a ship-mounted autonomous processing
conducting experiments to determine the effectiveness of the hydraulic transonic method of
decontamination combined with filtering procedures;
examining the effects of hydraulic transonic decontamination of seawater combined with
filtering on any macro and microorganisms, including bacteria and etc.;
reconciling research results with regulations set forth in the materials of the Marine
Environment Protection Committee 48th session (MEPC 48 / WP.15.10 October 2002.
Regulations E-1 ... E-4);
assessment of specific energy consumption of the project.
At the time of writing this paper, the first phase of our work has been completed ­ a complete set of
project and engineering documentation has been created that will allow us to manufacture the pilot
version of an autonomous ship-mounted processing plant. The device is engineered for testing
scenarios both on shore and in actual ship environments. The materials and devices used, the
technologies used to manufacture the processing plant and its various parts, and the operating
parameters are nearly all compliant with corresponding regulations of the Ukrainian Shipping and
Navigation Register for auxiliary equipment installed in seagoing ship machine rooms.
Hydrodynamic transonic decontamination of liquids is based on a complex localized high-intensity
exposure of the liquids processed to a combination of physical energy fields, specifically, a field of
ultrasonic wave influence, appearing during high-speed phase transfers and the nearly instant
(approximate timing is 10-4 to 10-6 seconds) pressure shifts in the system. These field effects are
achieved without using expensive electronic ultrasound generators, electric oscillators, and no heat
energy is used as input in the process. Spontaneous gassing occurs in the pressure shift zone, and,
with certain conditions met, fairly high-intensity ultrasound is also generated due to several
hydrodynamic effects. Ultrasound generated within a narrow segment of the liquid flow destroys the
structure of macro and microorganisms. Additionally, during the second sharp pressure shift from

Andruschenko: Hydrodynamic transonic treatment and filtration of ship ballast waters
vacuum conditions to very high pressure, the so-called pressure jump phenomenon occurs. Gas
bubbles collapse within the area affected by this phenomenon, which means a lot of mechanical
influence is applied to the liquid medium. The bactericidal effect of the ultrasonic vibrations depends
on the specific form of the microorganisms, the durability of the chemical composition of organism
cell walls, the presence of a cell capsule, the age of the culture, the intensity of influence, the
frequency of the ultrasonic waves, and the length of the exposure. It is known that the most lethal
effect is produced by ultrasound with a wavelength roughly the same as the size of the target
The draft regulations set forth at the Marine Environment Protection Committee 48th session (MEPC
48 / WP.15.10 October 2002. Regulations E-1 ... E-4)
specifically stipulate:
Regulation E-2 Ballast Water Management Standard
Option 2: Ships conducting Ballast Water Management in accordance with this Regulation shall
discharge no detectable quantities of viable organisms above [100] µm in size, and discharge no
more than [25 viable individuals of zooplankton per litre, 200 viable cells of phytoplankton per
ml26] smaller than [100] µm in size.
The more recent draft of the regulations produced by the Marine Environment Protection
Committee 49th session
significantly reduce the dimensions of organisms to control, raising the
bar of quality for ship ballast water treatment.
Regulation E-2 Ballast Water Performance Standard
Ships conducting Ballast Water Management in accordance with this Regulation shall discharge
no more than [25] viable individuals per litre of zooplankton greater than [10] µm in size; and no
more than [200] viable cells per ml of phytoplankton greater than [10] µm in size; and discharge
of a specified set of indicator microbes shall not exceed specified concentrations.
Regulation E-3 Additional criteria for Ballast Water Management systems
Ballast Water Management systems used to comply with the Convention must be:
The level of compliance of the trans-sound ballast
water treatment technology to the additional criteria

Safe in terms of the ship and its crew;
Safe for ship and crew members
Environmentally acceptable, i.e., not
Do not provide secondary pollution to the treated
causing more or greater environmental
environment, environmentally safe
impacts than it solves;
Practicable, i.e., compatible with ship
Contain equipment common for ships
design and operations;
Cost effective, i.e., economical; and
Assessment of economical parameters will be
carried out after installation testing
Biologically effective in terms of
Effective on all living forms contained in water
removing, or otherwise rendering
inactive Harmful Aquatic Organisms
and Pathogens in Ballast Water.
In order to validate the proposed technical solution, during the first phase, a working model was
created based on the TCA/3/B-1 processing plant, with a throughput capacity of 0.35 m3/hour and a
maximum water pressure of 10 MPa. A basic installation schematic for the TCA/3/B-1 is shown in
Figure 1.
The preliminary testing studied the effectiveness of the processing plant's influence on zooplankton
organisms living in the waters of the Odessa Bay. Among the principal technical benefits of the
processing plant, which make its use practical on a variety of ship types, are small dimensions,

2nd International Ballast Water Treatment R&D Symposium: Proceedings
productivity, energy economy, safety and easy maintenance. The initial material was a plankton
sampling living in water drawn from "Peschananya Gavani" of the Odessa Port. During the
experiment, seawater was passed through the TCA/3/B-1 processing plant, running the following
operational parameters:
· Water pressure before processing point: 20 bar
· Water pressure in the working area:
0.05 bar
· Water pressure on exit:
5 bar
· Input water temperature:
· Output water temperature:
A control sampling of the original seawater was taken before each series of experiments, and the
populations of the various zooplankton organisms were assessed. The volume of water that passed
through the processing plant during each experiment equaled approximately 10-30 litres of water.
After processing, another sample was taken, and the remaining plankton debris was condensed with
the use of special gauze filters (#60 and #70).
Table 1. Species content and basic dimensions of zooplanton.
Organism Species
Dimensions (microns)
Aztropoda Type, Crustacea Class
Brachiopoda Subclass
Cladocera Order
Podonidae Family, Podon Genus
Pleopis polyphemoides
Daphniidae Family, Daphnia Genus
Daphnia magna
2000-4000; 1500 average
Copepoda Subclass
Calanoida Order
Pseudodiaptomidae Family, Calanipeda Genus
Calanipeda auae dulcis
1000-1200; 1000 average
Acartiidae Family, Acartia Genus
Acartia clausi
1170-1750; 1000-1500 average
Cyclopoida Order
Oithonidae Family, Oithona Genus
Oithona minuta
500-700; 400-600 average
Harpacticoida Order
Canuellidae Family, Canuella Genus
Canuella perplexa
900-1300; 800-1000 average
Harpacticidae Family, Harpacticus Genus
Harpacticus flexus
Harpacticus sp.
Copepoda, nauplii
Cirripedia Subclass
Balanomorpha Superfamily, Balanidae Family, Balanus Genus
Balanus, nauplii
Nematoda Type, Rotatoria Class
Monogononta Order
Brachionidae Family, Brachionus Genus
Brachionus plicatilis
Synchaetidae Family, Synchaeta Genus
Synchaeta sp.
In determining the species of the organisms that survived or were killed, living/dead percentages were
detailed for each type of organism, as well as their sizes; various samples of debris were examined
and their possible origins were accounted for.

Andruschenko: Hydrodynamic transonic treatment and filtration of ship ballast waters
Preliminary processing of the samples was performed without fixing. Then the samples were fixed
with 10-percent formaldehyde solution to conduct further laboratory analysis.
Calculation, content determination and zooplankton measurement was performed with binocular lens
In order to identify the zooplankton forms, Black and Azov Sea Fauna Guide was referenced.
Species content and basic dimensions of zooplankton organisms are given in Table 1.
The first series of tests
Samples were condensed through gauze No.60
Control: The number of zooplankton organisms was 560 per liter. The basic taxonomic groups
present in samples are Rotatoria (wheel animalcules), Cirripedia nauplii, Copepoda nauplii,
Harpacticoida, Cladocera (crustaceans), Nematoda (worms), Polychaeta larvae. Among them,
wheel animacules (36%), Cirripedia nauplii (25%) and Harpacticoida (21%) were prevalent in
the total number. Many Infusoria could be observed. Natural departure of organisms has not been
After processing with the TCA/3/B-1 plant: We detected 9 and 11 (i.e. 0.9 and 1.1 per liter)
Cirripedia nauplii with dimensions of 50-90 microns in 2 analyzed samples, respectively.
Processing efficiency, i.e. death rate, for this kind of species amounted to 99.3%. Representatives
of other organism groups were not detected. The efficiency is 100%. There are neither Infusoria,
nor zooplankton fragments in the water.
The second series of tests
Samples were condensed through gauze No.60
Control: Zooplankton content is similar to that of the first test series. The number of organisms is
400 per liter.
After processing with the TCA/3/B-1 plant: We detected 14 Cirripedia nauplii (i.e. 1.4 per liter)
with dimensions of 40 to 80 microns. Processing efficiency, i.e. death rate, for this kind of species
amounted to 98.6%. The processing efficiency for other species is 100%. The water is clean,
suspended matter and Infusoria are not present.
The third series of tests
Samples were condensed through gauze No.70
Control: The number of organisms is 3200 per liter. The species are Rotatoria, Cirripedia
, Copepoda nauplii, Calanoida, Harpacticoida, Cladocera, Polychaeta larvae. Among
them, Calanoida (34%), Harpacticoida (22%) and Cirripedia nauplii (19%) were prevalent.
Many Infusoria could be observed.
After processing with the TCA/3/B-1 plant: We detected 7 Cirripedia nauplii (i.e. 0.7 per liter)
with dimensions of 40-100 microns. Processing efficiency, i.e. death rate, for this kind of species
amounted to 99.9%. The processing efficiency for other species is 100%. The sample contains
many fragments of crustaceans, some non-destructed chitinous covers are also met. Infusoria are
not observable.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The fourth series of tests
Samples were condensed through gauze No.70
Control: The number of organisms is 6000 per liter. Their species are Rotatoria, Copepoda
, Calanoida, Harpacticoida, Cladocera. Cladocera crustaceans and wheel animalcules
were predominant.
After processing with the TCA/3/B-1 plant: Living zooplankton organisms were not found in
any of the samples taken while the plant was operated. The plant performance efficiency is 100%.
1. Under conditions in which these experiments were performed, passing sea water through the
TCA/3/B-1 processing plant provides for practically complete destruction of zooplankton
organisms. Cirripedia nauplii proved to be the organisms most resistant to effects produced by
the processing plant. Their survival rates observed in three series of tests turned out to be 0.7%,
1.4% and 0.1%, respectively, which can possibly be explained by inadequate cleanliness of the
processing tract before the experiments.
2. Transonic decontamination technologies are one of the potential solutions to the ship ballast water
treatment problem ­ they are characterized by the simplicity of the equipment used, ease of
maintenance, transfer no secondary pollution to the liquid being processed, and are based mostly
on equipment commonly found on ships.
More research on the effectiveness of ballast water transonic decontamination is planned for this year
on the experimental autonomous processing plant created during the first phase of the project. A
schematic of the experimental processing plant is provided in Figure 2. The device complex includes:
Reservoir for input water to be processed.
Tank to mix and store chemical disinfection reagents
Peripheral pump
Mechanical filtration module
Gas saturation module, part of the filtered air / water ejector segment
Holding tank for aerated water under pressure
Multistage centrifugal pump
Transonic hydrodynamic module
Degasification module
10. Reservoir to hold processed water
The processing plant is equipped with a thermometer to measure the temperature of the water being
processed, as well as a manometer and flow meter. The basic installation allows for both successive
processing of the water by all modules, as well as selective processing, with different combinations of
modules. The whole device complex is mounted on a common framework.
The purpose of the chemical reagent tank is to provide means of preliminary decontamination of the
hydrodynamic tract. If necessary, the position and hookup of the tank allows routing controlled doses
of reagent solution directly into the water medium being processed.
The peripheral pump provides a pressure increase in the water being processed, up to 1 MPa. Control
of the pump output pressure is achieved with a bypass valve. Water volume passing through the
mechanical filter can be measured by the flow meter.

Andruschenko: Hydrodynamic transonic treatment and filtration of ship ballast waters
The mechanical filtration module can use a number of different filtering materials with an effective
mesh of no more than 100 micrometers.
The processing plant can function with or without the presence of a filtering material in the
mechanical filtering module.
The purpose of the ejector is to force filtered air into the water being processed, raising the pressure of
the mix and pumping it into the holding tank. The layout of the processing plant also allows it to
function without air input.
The purpose of the holding tank is to saturate the water being processed with air under pressure, and is
equipped with a water level gauge glass, an emergency valve and a device for maintaining the needed
water level. Extra air is released automatically into the atmosphere.
The purpose of the multistage centrifugal pump is to further raise the water pressure to 2.5 MPa.
Output pressure is controlled via a bypass valve. The device complex installation scheme allows for
operation without the centrifugal pump. In that case, a section of pipe is installed in its place.
The hydrodynamic transonic module destroys zooplankton present in the water. This module was
engineered to be easily interchangeable, to allow testing both high-pressure and low-pressure modes
of water treatment. The low-pressure mode uses only the peripheral pump. Maximum water pressure
in this mode is 1 MPa. The high-pressure mode uses both the peripheral and the centrifugal pumps.
Maximum water pressure in this mode is 2.5 MPa.
The degasification module removes gases released in the water by the hydrodynamic cavitational
treatment, and lowers dissolved gas content in the treated water to levels lower than the input water.
Extracted gases are released into the atmosphere. After going through the degasification module,
processed water is collected in a final reservoir.
Preliminary analysis of the results allows us to make the following conclusions:
· Treatment of seawater by a TCA-type processing complex can result in a near absolute
elimination of zooplankton and phytoplankton organisms.
· Transonic decontamination technologies are one of the potential solutions for treating ship
ballast waters. They are characterized by the use of simple and reliable technical components,
ease of maintenance, no secondary processing pollution, and are based on technology that is
commonly found on most ships.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Input water
Processed water
storage tank
collection tank
Figure 1. Basic schematics of model setup of TCA/3/B-1 processing plant.
Input water reservoir
Processed Water
Holding Tank
Figure 2. Composition schematic of hydrodynamic water.

A new modular concept for the treatment of ships
ballast water - the Hamann project
H. Röpell & T. Mann
Hamann Wassertechnik GmbH, Germany

This paper presents the new ballast water treatment option of Hamann Wassertechnik GmbH,
Germany. The system is based on a modular concept which includes a two-step physical separation
(hydrocyclone and 50 µm self-cleaning filter) as well as a secondary treatment with an oxidising
agent (Peraclean®Ocean). Due to the modular approach, the Hamann ballast water treatment option
is flexible in adopting different ships' requirements such as space available, location of ballast water
system on board ship, capacity of ballast water pumps, and others. Furthermore, it can adjust to
different ballast water management scenarios (changing flow rates of ballast water pumps, serving all
ballast water tanks from double bottom to top wing, varying flow rates of ballast water pumps).

Hamann's ballast water treatment option has been tested at full scale flow rates of 135 m3/h to 210
m3/h. The tests were carried out at a number of different land based location with different water
qualities (e.g. Baltic Sea, Lower Elbe River and Port of Hamburg). The Artemia Testing System (ATS)
was applied to produce reliable and reproducible test results.

The key-objectives of the project were the identification of suitable combinations of treatment steps /
methods for various types of ships and ballast water management scenarios as well as the design of a
full scale treatment plant for land-based tests and evaluations. The time frame of the project was 3
years (2000 to 2003). The project was funded jointly be the Federal Ministry Of Research (through
Arbeitsgemeinschaft industrieller Forschungs-vereinigungen, Otto von Guericke e.V., AIF) and the
At this stage, a practicable combination of treatment steps has been identified and a full-scale test unit
has been produced for land-based tests. The tests of this treatment plant were carried out at different
locations and different flow rates.
Technical description of the test treatment plant
The test treatment plant has a modular design, which gives the most possible flexibility in addressing
different ballast water management scenarios and different types of ship. The modular design concept
allows the installation of individual treatment steps at different locations on the ship where room is
available, which makes the system suitable even for refits. Furthermore, new modules can be
integrated or added to an existing system as technology improves or regulations change.
The modular system that was tested consisted of the following treatment steps / modules:
1. A physical separation which included two treatment steps:
A new developed hydrocylone which was specially designed for ballast water applications. It
significantly reduces the sediment load of the ballast water and also removes some of the
organisms. The small size of the individual hydrocylone allows installation on a single deck.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The number of hydrocyclones needed (35 m3 to 45 m3/h each) depends on the flow rate of the
ballast water pump.
A fine filtration with a mesh size of 50 µm, which serves two functions:
1) It removes nearly all organisms with a body length > 100 µm,
2) It increases the stress imposed on the organisms present in the ballast water, resulting in
physical damage of the organisms as well as increased sensitivities towards the secondary
2. A chlorine free oxidising agent (Peraclean®Ocean) was used as a secondary treatment, which was
dosed to the ballast water after the physical treatment at concentrations of only 150 ppm. The
selected oxidising agent is fully bio-degradable and has no corrosive impact on the ballast water
system of the ship.
The tests were carried out at different locations with different types of water:
In the inter-tidal zone of the lower Elbe River near Brunsbüttel. This location is
characterised by changing salinities due to tidal influence and high turbidity with high loads
of suspended solids.
In the Baltic Sea (Kiel), at constantly brackish water conditions (salinity about 13 ppt).
Currently more tests are on the way in the Port of Hamburg at freshwater conditions.
All tests were carried out at flow rates of 135 m3/h to 210 m3/h. The duration of the tests varied
between 4 weeks (Kiel) and 16 weeks (Brunsbüttel). During the tests, the treatment plant was
operated at the above flow rates for an average of 8 to 10 hours of continuous operation during
working days.
Furthermore, the biological efficiency of the treatment plant was evaluated at each of the test sites.
The evaluation was based on the removal/inactivation of the plankton present at the testing sites and
surrogate organisms, respectively. Different life-stages of Artemia salina were used as surrogate
organisms and the ATS (Artemia Testing System) test protocol was applied.
A total of 7 tests have been conducted with the plankton present at the test sites and 6 experiments
were carried out according to the ATS test protocol. The results of the tests are summarised below.
Results of full-scale land-based tests
· The treatment plant performed during the test cycles without mechanical problems, giving
good continuous flow conditions at each of the testing sites.
· The biological efficacy was evaluated for each treatment step separately. The different
qualities of the water at the testing sites had no influence on the biological efficacy.
· Great differences occurred in the separation rates of both, the hydro cyclones and the 50 µm
filter, according to the different sizes and physical properties of the test organisms.
· During all tests, the Hamann Modular Ballast Water Treatment plant was dosed with 150 ppm
of Peraclean® Ocean, which is equivalent to 15 l of Peraclean® per 100 m3 of ballast water.
· After 24 hours of exposure time to Peraclean®Ocean, no living organisms were detected in
any of the samples. The test results are summarized in the following table.

Mann: A new modular concept for the treatment of ships ballast water - the Hamann project
Numbers of frequently found
Range of
average average
kill rate after 24
organisms during the
numbers of
removal removal
hrs exposure to
experiments; single findings
per litre
150 ppm
were not regarded
per litre
cyclone + 50µm
ATS experiments
Artemia salina nauplii
3 ­ 47
Artemia eggs in development
10 - 231
Artemia soaked cysts
17 - 220
Artemia dry cysts
7 - 65
in-situ plankton organisms
Copepod (Cyclops sp.)
2 - 10
Daphnia sp.
0 - 2
Copepod nauplii
1 - 14
1 - 41
0 - 13
*no living organisms were detected in any of the samples after 24 hrs exposure to 150ppm of Peraclean®Ocean.
**the observed differences between the removal rates of the cyclones and the removal rates of the cyclones +
fine filter were not significant, because of the high variance of the input values.
Conclusions and remarks
The Hamann modular BWT addresses all of the following criteria:
· compliance with short term regulations that are currently discussed by IMO and
· options for upgrading to future requirements
· the type of ship and the individual ballast water management plan
· space requirements (footprint of set-up)
· risks involved: safety and handling, environmental risks (aquatic toxicity)
The current treatment modules include a physical separation in two steps, the Hamann Hydrocyclones
and a 50 µm fine filtration unit plus a disinfection with a chlorine free oxidising agent
Test results showed a removal of:
· 97% of all organisms > 100 µm in smallest dimension;
· 80 % of all organisms of < 100 µm in smallest dimension; and
· a killing/inactivation of all organisms, no living organisms were detected after 24 hrs of
Further full-scale tests will be carried out in the port of Hamburg and onboard ship's with updated
testing procedures according to currently developed national and international test standards.

A portable pilot plant to test the treatment of ships'
ballast water
S. Hillman, F. Hoedt & P. Schneider
CRC Reef Research Centre
James Cook University, Australia
Aims and objectives of the project
The objective of project is to build a pilot treatment plant based on existing technologies and off-the-
shelf equipment. The pilot plant uses various technologies, as well as chemicals on a `plug and play'
basis. The medium to longer term aim is to develop a system that will be scaled up and used aboard
Research methods
The overall concept of the pilot plant is to treat ballast water using a variety of techniques individually
or in series as described by Oemcke (1999 (1)), Oemcke and Hillman (2001). They suggested that the
use of filtration followed by ultra-violet radiation would be efficacious in treating a large number of
known ballast water pest species with the notable exception of encysted dinoflagellates. To address
the need to treat ballast water for encysted organisms, the pilot plant incorporates an ultra-sonic shear
device as well as the capability of injecting measured doses of chlorine dioxide.
The pilot plant has been constructed in a 20 foot container and is designed so that it can be used under
laboratory conditions or be moved to different ports for testing under various port environmental
conditions, such as temperature, salinity and sediment load. It can also be put aboard ships for testing
of ballast water treatment in transit.
We consider the best time to treat ballast water is during loading by the ship. This has the benefit of
leaving filter backwash material at the port of origin and ensures all ballast water passes through the
treatment system. Recognising that the quality of discharged ballast water is the essential criteria, the
use of the system to treat water during discharge may be necessary to kill or inactivate organisms that
may have recovered during the voyage. The pilot plant is, therefore, designed to treat water as it
passes through the system. Design flow rate for the pilot plant is between 2 and 3 litres per second (up
to 11 tonnes per hour). This balances our ability to sample the water effectively with the need to
demonstrate pilot plant perfomance. It also investigates treatment at reasonable flow rates, which we
feel can be scaled up to meet required full scale throughputs. A schematic of the pilot plant is found at
Figure 1.
Pilot plant operation
Fresh seawater can be stored in two 27,000 litre tanks. Test water can then be gravity-drained to the
10,000 litre dosing tank, where it is inoculated with the organism of choice. This tank is mixed using
an aeration system to enhance homogeneity. The contents of this dosing tank can be pumped to any,
or all, of the Amiad filter, the sonic disintegrator and the ultra-violet unit. Sampling points are
available before and after the pump and after each treatment method.
This filter can be used with a number of different sized screens and the project has available to it 20,
50 and 80 micron screens. To date only the 80 micron screen has been used. The sonic disintegrator is

Hillman: A portable pilot plant to test the treatment of ships' ballast water
driven by a variable frequency drive that allows the speed of the machine to be varied to optimise
effects. The ultra-violet unit operates at 254 nanometres. All components are designed to be able to be
operated at greater than the design capacity of 3 litres per second.
Test protocols
The pilot plant has been operational for two weeks and protocols are presently being refined. Our
initial objective is to manage variabilities in sampling the 10,000 litre tank and at the sampling points
with no treatment of the water occurs. A sampling regime is being developed that will determine with
95% confidence that the variability in the numbers of organisms counted is less than 5%.
We have chosen to use Artemia for initial testing. The Artemia Testing System protocol (Voigt, 1999)
suggests the use of different life stages of Artemia as a surrogate for pelagic and benthic cysts as well
as numerous planktonic organisms. While it is clear that Artemia cannot be used as surrogates for
smaller organisms ­ due to its size ­ its availability and ease of culture makes it attractive, at least for
the early stages of testing.
It will be necessary to repeat the testing for variability for each class of organism tested. These will
include larvae of bivalves, fish, echinoderms, algae and free-swimming phytoplankton. Dinoflagellate
cysts will also be used.
Post-treatment and control samples will be stored under temperature controlled, dark conditions to
simulate hold-up in ballast tanks. Sub-samples will be taken immediately and after suitable periods of
dark storage (e.g. 24 hourly for 5 days).
Biological testing of the ballast water pilot plant
The pilot plant is comprised of two distinct parts, firstly the shipping container with built-in treatment
options (filters, sonic disintegrator, UV) and sample outlet points and secondly, the external system of
storage and culture tanks and sampling outlet points (see appendix A). The experimental system is
designed to enable the inoculation of a known density of mono-specific culture in the 10,000 litre
culture tank which is then pumped through the pilot plant with selected treatments activated. Samples
are collected to determine changes in population density and the viability of organisms at key
locations in the system (i.e. pre and post-treatment). At a later stage, several smaller culture tanks (2-
500 litre) will be placed in the container to examine longer-term effects on surviving organisms.
The location of sampling points is an important aspect of the process of biological testing and
experimentation of the efficacy of the system. At present there are 7 sampling points in the whole
system (Table 1). The diameter of these sampling points varies, as some were built into the steel
piping system prior to arrival and testing, and the diameter of others is governed by the diameter of
the attachment location. If necessary, sampling points will be changed so that they are identical. An
explanation of the location and role of each sampling point is given in Table 1.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Table 1. Sampling points throughout the external and internal pilot plant system. `Internal' refers to within the pilot
plant container and `external' is outside.
Sampling point
Sample method
Information derived
Point 1
External ­ Samples
25mm suction hose at 5
Accurate estimate of initial
collected in the 10,000 litre
locations and three depth
stocking density of culture
culture tank
Point 2
External ­ PVC tap (50mm
Collect 10 litre sample
Estimate of organism
outlet) with valve at outlet
(allows full-flow flushing)
density exiting tank and
pipe just after 10,000 litre
and take one 250 ml sample
entering pump
tank and prior to inlet pump
from each bucketful
Point 3
External ­ PVC tap (25 mm
Collect 10 litre sample
Estimate of organism
outlet) just after pump and
(allows full-flow flushing)
density after pumping and
before filtration unit
and take one 250 ml sample
prior to treatment.
from each bucketful
Point 4
Internal ­ 6 mm steel tap
Collect 10 litre sample in
Estimate of organism
outlet on pipe just before
bucket and take one 250 ml
density in piping just before
sonic disintegrator
sample from each bucketful
sonic disintegrator and after
Point 5
Internal ­ 20 mm steel tap
Collect 10 litre sample in
Estimate of organism
outlet on pipe just after
bucket and take one 250 ml
density in piping just after
sonic disintegrator
sample from each bucketful
sonic disintegrator
Point 6
Internal ­ 20 mm steel tap
Collect 10 litre sample in
Estimate of organism
outlet on pipe just before
bucket and take one 250 ml
density in piping just after all
exiting pilot plant after UV
sample from each bucketful
Point 7
External ­ 50mm pvc valve
Flush pipe then collect 10
Full-flow sample after all
tap outside of pilot plant
litre sample in bucket and
take one 250 ml sample
from each bucketful
Experiment 1 (18/6/03 and 25/6/03)
Number and volume of subsamples required to estimate Artemia density in the 10,000 litre
culture tank

To examine the effect of treatments, knowledge of pre-treatment and post-treatment density of
organisms is necessary. Further, it is important to understand the variability of subsamples and be able
to state the accuracy of any density estimate based on subsamples. We have chosen the 95%
confidence interval as a statistical measure of accuracy. We undertook a preliminary assessment of
subsampling variability in order to determine the required number of samples to obtain a
predetermined accuracy. Initially it was thought that we would aim for 95% confidence that the
density estimate was within 5% of the actual density.
A series of 15 replicated 1 litre samples (30 total) were collected at 15 different locations within the
10,000 litre culture tank on 18/6/03. The locations were chosen to include all obvious factors that may
cause aggregation. These were:
· 3 depth strata from top to bottom,
· above and between aeration lines, and
· sunny versus shaded parts of the tank (since Artemia are photophyllic).
The 1 litre samples were collected with a 25mm suction hose and then concentrated to 250 ml by
removing water using a small suction hose within a container fitted with 20 micron mesh (to prevent
Artemia removal).
Concentrating samples proved time consuming. Subsequently, 10 × 250 ml samples were collected at
10 locations in the same tank on 25/6/03 to examine the possibility of taking smaller-volume samples.

Hillman: A portable pilot plant to test the treatment of ships' ballast water
The results of the above experiments were used to determine the number and volume of samples
required to obtain a `starting density' estimate with a pre-specified degree of accuracy. Both
experiments used cultures of newly hatched Artemia nauplii (24 hrs post incubation). Samples were
sorted fresh in a Bogorov tray using a stereo microscope.
The results of analysis on the first set of samples are shown in Table 2. The small confidence limits
(mean = 86 +/- 5.8), show that 30 samples enabled a relatively accurate estimation of density. Further
analysis showed that 53 × 1 litre samples would be needed to be 95% confident of obtaining a density
estimate within 5% of the actual mean density in the tank.
Table 2. Statistical analysis of samples taken at 15 locations in the 10,000 litre tank on 18/6/03.
Number of samples
Mean density (# per litre)
Standard error of mean
95% Confidence interval
To examine the possibility of collecting smaller samples; a second set of 250 ml samples were
collected on 25/6/03 and an analysis of the variance was undertaken. Results are shown in Table 3. It
was calculated that 145 × 250ml samples would be needed to get within 5% of the actual mean. These
results suggest that 250 ml samples may not be viable.
Table 3. Analysis of 10 X 250 ml samples collected in the 10,000 litre tank.
Number of samples
Mean density (# per litre)
Standard error of mean
95% Confidence interval
Experiment 2 Analysis of mixing by aeration in 10000 litre culture tank
As yet a statistical analysis of variability in sample density with location (ANOVA on 30 samples ­
18/6/03) has not been undertaken but visual inspection of the data suggests that there were no obvious
points of aggregation and that density estimates in samples were generally similar with random
Experiment 3 Analysis of variation between different sampling points after the 10,000 litre tank
(`control samples') and effect of the intake pump on
Samples were collected at 4 post-culture tank sampling points to make a preliminary comparison of
these and further to examine if passage through the pump was effecting Artemia densities. Table 4
shows that the pump appears to have reduced Artemia density by more than half the tank density.
Further, while there were no dead nauplii in tank samples, there were high proportions of dead nauplii
in most samples taken after the pump. Thus it is evident that the pump may be reducing densities and
killing nauplii. This will require further investigation to quantify. The densities at the three post-pump
locations are comparable to the pre-treatment point. There is a suggestion of some variation between
points which again will require further sampling and analysis, particularly with regard to the
unexpected apparent increase in numbers as the organisms progress through the system.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Table 4. Densities of Artemia in 250 ml samples collected at 4 sampling points with no treatments operational.
(just after pump)
(just before sonic
(external to
Experiment - 3 Treatment effects
Tables 5, 6 and 7 summarise the preliminary experiments on treatment effects. Due to the short time
available and the need for further analysis to determine sampling requirements, these results are of a
preliminary nature having not been done with a systematically established sampling protocol.
Filtration rates with the 80-micron screen are between 92 and 96 % (Table 6). It is likely, given the
size of nauplii (300 by 150 microns), that some Artemia in the post filter samples may have been
caught in the system prior to filter activation since sample point 6 was not flushed prior to sampling.
All post sonic disintegrator samples contained no Artemia (alive or dead) (Table 7). While the sample
numbers were low this is an encouraging result.
Table 5. Preliminary experiments to determine treatment effects.
Post­treatment number of samples
Filter ­ 80 micron
Sonic disintegrator
Filter ­ 80 micron
sonic disintegrator
Table 6. Experiment with 80-micron filter screen ­ mean densities at sample points are given. Number of
samples shown in brackets.
Pre-filter densities
Post-filter at sample Pt. 6 *
% filtration
85.8 (Tank density)
3 (5)
74.4 (10)
25.9 (30)**
2 (10) (all nauplii)
1.2 (10) (alive only)
* On 24/6/03, post filter samples at point 6 were collected while the filter was on; pre-filter samples (pts. 1, 3 & 6) were
taken in a run with no treatment operational just beforehand. Therefore these can only be used to infer the likely system
densities before the filter.
** Mean pre-filtration densities from sample points 1, 3 and 6.

Hillman: A portable pilot plant to test the treatment of ships' ballast water
Table 7. Experiments with sonic disintegrator ­ mean densities at sample points are given. Number of samples
shown in brackets.
Sample pt. 1
Sample Pt. 3
Sample Pt. 4
74.4 (10)
18.4 (10)
26 (10)
0 (4)
52.8 (10)
5.3 (10)**
0 (10)
* Sonic disintegrator samples at point 4 were collected a few hours after the other samples were taken with sonic
disintegrator off. Thus these can only be used to infer the likely system densities before the sonic disintegrator.
** Unusually low pre-treatment densities ­ filter may have been accidentally left on filtering a component (not all due to
plumbing design) of the throughput water.
Utility of the technologies used
There has been much reporting of the difficulties associated with the flow rates required for vessels to
treat ballast water. Often the stress has been on the relatively small number of large vessels (e.g. Cape
Class) with ballast loading rates of 3000 tonnes per hour or more. It is important to weigh this against
the average bulk carrier dead weight tonnage being 35,750 tonnes and the average DWT of tankers
being 38,000 tonnes (Anon, 2002). This indicates that there are a very large number of vessels that
will have loading rates that are more of the order of 500 tonnes per hour.
While filtration and/or ultra-violet treatment at the higher loading rates is possible, albeit difficult,
there is a significant proportion of the world fleet that will be much more easily catered for using this
kind of technology. As has been pointed out by Oemcke (1999 (2)), that small ships will release much
less ballast water than large vessels and, therefore, a given kill rate will only need to be proportionally
smaller to achieve approximately the same level of risk reduction. Furthermore, as suggested by
Hilliard (2001), there is evidence that discharge frequency is at least as important as volume. Thus
there would be great advantages in risk reduction by incorporating treatment measures in the large
number of smaller ships.
The two established technologies being tested, filtration and ultra-violet light, have existing facilities
that can treat upwards of 1000 tonnes of water per hour. In the case of Amiad filtration, this can be
achieved through a single large filtration. However, this throughput would almost certainly require
several ultra-violet units running in parallel. The scaling up of the sonic disintegrator is being carried
out as a related research project within the School of Engineering at James Cook University.
Inspection of ships' pump and engine rooms do not indicate that there will be major issues with space
Conclusions and recommendations
While it is too early to form conclusions regarding the efficacy of the pilot plant, the results to date
indicate that with further development the technologies are capable of effectively removing organisms
from ballast water as it is being loaded aboard ship.
This project is being supported by:
Environment Australia, Ports Corporation of Queensland, Townsville, Mackay and Gladstone Port
Authorities, Amiad Australia, CRC Reef Research Centre, URS Australia, the Great Barrier Reef
Research Foundation and Pasminco.
The researchers gratefully acknowledge this support.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Anon, 2002. Lloyds Register ­ Fairplay, 2002. See http://www.lrfairplay.com/.
Hilliard, R. 2001. Determining Practical Evaluation and Approval Guidelines and Criteria for "First
Generation" Ballast Water Treatment Systems: Some Issues on their Derivation, Application and
Development. 1st International Ballast Water Treatment R&D Symposium, IMO London 26-27 March
2001: Symposium Proceedings
. GloBallast Monograph Series No 5. IMO London.
Oemcke, D. 1999 (1). The Treatment of Ships' Ballast Water. EcoPorts Monograph No 18, Ports
Corporation of Queensland, Brisbane.
Oemcke, D. & Hillman, S. 2001. Ultra-Violet Treatment of Ships' Ballast Water ­ Research to
Reality. Proceedings of the First International Conference on Ballast Water Management, Singapore.
Oemcke, D. 1999 (2). Comparing ballast water treatments on a cost-benefit basis: the balance of
needs. The Ballast Water Problem- Where to from here? Proceedings of a workshop held 5-6 May
1999, Brisbane Australia
. EcoPorts Monograph Series No 19. Ports Corporation of Queensland,
Voigt, M. 1999. Artemia Testing System Protocols, developed by dr.voigt-consulting, Germany. See

Hillman: A portable pilot plant to test the treatment of ships' ballast water
Figure 1. Schematic of pilot plant.

Ballast water treatment R&D in the Netherlands:
Ballast water treatment on-board of ships & evaluation
of market potential and R&D requirements
C.C. ten Hallers-Tjabbes 1, J. Boon 1, M.J. Veldhuis 1, J.L. Brouwer 2 & J. Rvan Niekerk 2
1Royal Netherlands Institute for Sea Research (NIOZ)
The Netherlands
2Royal Haskoning, Shipping HSE services
The Netherlands
In this paper, we give an overview of the future needs for ballast water treatment systems on board of
the worlds' shipping fleet. In our view, such systems should consist of a two-stage design, involving a
primary particle exclusion step followed by a secondary step, that kills the remaining living
organisms. An important prerequisite of the treatment is that the receiving ecosystem should not be
damaged by discharged ballast water. Therefore the use of (toxic) chemicals for this purpose appears
a risky way to go. Assessment of the adequacy of treatment systems requires a reliable evaluation
The minimum size of organisms that should be removed from the seawater during primary treatment
should be in the order of 10 µm. A larger diameter for particles to be removed will result in an
incomplete removal of silt and clay particles and will allow for formation of a substantial sediment
layer that acts as a seafloor offering shelter for living organisms in the ballast tanks. A secondary
treatment step should kill the remaining organisms after primary treatment. These mainly involve part
of the algal species responsible for harmful algal blooms, bacteria and viruses.
The performance of ballast water treatment equipment should in the future be monitored in an
automated way. Flow-cytometry offers good potential to achieve this goal, since it can be fully
automated, and discriminate between living and dead cells.
A future global market potential has been estimated based on a relevant world fleet for ballast water
management requirements of some 33,000 vessels (larger than 1,000 tonnes dwt) and a model general
cargo vessel of 12,000 tonnes dwt with a ballasting capacity in the range of 600 ­ 1,000 m3/h. This
showed an annual market potential ranging from USD 225 million to USD 350 million for the period
between adoption and ratification of the international convention. After ratification of the
international convention this annual market potential is expected to increase to a range from USD 700
million to USD 1,100 million.
Ballast water has been subject to the development of (inter)national legislation and the problem has
been studied for many years, in particular with an aim to minimise the risk of introduction of alien
organisms in marine ecosystems through the transfer of organisms through ballast water.
One of the preliminary results is the "draft international convention for the control and management
of ships' ballast water and sediments", which is expected to be adopted by IMO in 2004.

Brouwer: Ballast water treatment R&D in the Netherlands
Recent studies performed by Royal Haskoning looked into the possibilities and constraints of ballast
water treatment on-board of ships and the global market potential for this equipment. The most
relevant studies are Application of ballast water treatment techniques on Dutch vessels (2001), Global
market analysis of ballast water treatment technology (2001)
and Ballast water treatment; full scale
tests, strategies and techniques (2002, in co-operation with Royal Netherlands Institute for Sea
Research - NIOZ).
This is intended to lead to full-scale tests of treatment equipment on-board of
ships, which is currently in development by NIOZ and Royal Haskoning.
Ballast water: the problem
The use of (sea) water as ballast for the stability and trim of the vessel and to submerge the propeller
is a necessity on one hand, but poses a risk of the movement of non-indigenous marine organisms
between ecosystems on the other hand. This is considered today to be one of the most important
threats to the stability of local ecosystems, and thereby biodiversity.
The size of organisms and sediment particles is a key factor and serves as a classification basis in
ballast water management, as the efficacy of ballast water treatment depends on the potential to
remove particles, including those of a smaller size. This capacity in turn is related to the techniques
that should be applied to analyse (or test) for the presence of such tiny organisms. The natural range
of organisms is very variable, and as an example the size classes of pelagic organisms are given in
figure 1, indicating a wide range of size classes from < 1 µm until > 1000 µm.
Besides marine organisms, ballast water also contains sediment (sand and clay). Sediment in itself is
not a problem in the sense as described above, although it reduces the maximum cargo weight to be
loaded. But a stable sediment layer in a ballast tank provides a `sea-bottom' and thereby a stable
hideout place for organisms and a hothouse for growth and increase of numbers until the ballast tank
are emptied. Many organisms experience an emptying ballast water tank as a low-tide situation, to
which they respond by hiding in the sediment, only to emerge again at the next flood, i.e. when the
ballast tanks are filled again.
The size range of sediment plays an important role; clay particles are generally smaller than 2 µm and
sand particles are in general larger than 50 µm. Sediment can also build up in a ballast water tank by
decaying organisms that were killed or damaged at intake or died during the voyage. The resulting
detritus is often minute in size and by its nature can increase the coherence of the tank sediment.
Depending on the location of intake of ballast water, sediment can be easy or very difficult to remove.
Several NW European ports have sediment that is mainly in the range of 10 ­ 50 µm (i.e. the silt
fraction); such conditions are also common in estuaries and deltas in flat coastal areas elsewhere the
IMO requirements for ballast water management
The current draft international convention for the control and management of ships' ballast water and
sediments (MEPC 49) gives, amongst others, guidelines for ballast water treatment and for efficacy of
treatment by indicating what needs to be absent in treated ballast water that is considered safe to
discharge. This includes a measure for the level of organisms that should not be present in treated
ballast water, standards for ballast water management and the review of standards.
Ballast water that is considered safe to be discharged should minimise the risk of harm to the
environment, human health, property and resources; the standards that are being developed for the
purpose of the convention reflect such quality.
Standards for ballast water quality under development aim to determine a cut-off size class for
different classes of organisms, together with a concentration level that should not be exceeded.
According to the currently proposed levels, ballast water should meet the following performance

2nd International Ballast Water Treatment R&D Symposium: Proceedings
standard: zooplankton greater than 10 µm in size should not exceed 25 viable individuals per litre and
phytoplankton greater than 10 µm in size shall be less than 200 viable cells per litre. These values are
reflecting the current state of understanding and are likely to be further developed and refined
alongside further developments in treatment and testing research. Of even more importance than
minimum numbers is the viability of the remaining cells. In other words can those alien introductions
form new populations in the area of the port of destination. A proper treatment method should in the
end focus on a reduction of viable organisms, and treatment performance should be evaluated
It is proposed to review the standards before the effective date in order to determine the availability of
appropriate technologies; thereby enabling optimum performance and innovation.
Ballast water management systems should be safe in terms of the ship and its crew, environmentally
acceptable (i.e. not causing more or greater environmental impacts than it solves), practicable (i.e.
compatible with ships' design and operations), cost effective (i.e. economical) and biological effective
in terms of removing, or otherwise rendering inactive harmful aquatic organisms and pathogens in
ballast water.
Ballast water treatment options and restrictions
In theory ballast water can be treated on-board of the ship or in a land-based facility. This paper will
focus on on-board treatment only.
The treatment of ballast water can be performed upon intake or discharge of ballast water, and during
the voyage. Each option has its own advantages and disadvantages and the choice in favour of an
option is also dependent on the type/size of the marine organisms and sediment and the treatment
equipment to be used.
Treatment upon intake of ballast water has the advantage to prevent organisms and sediment to enter
the ballast tanks in the first place, but the required equipment will be relatively large. It has also been
proven, since a hundred percent prevention and/or killing is not possible, that some organisms even at
a low initial concentration are able to increase in numbers during the voyage, while others will die and
decay in the sediment. This indicates that treatment upon intake only will not be sufficient.
Treatment upon discharge prevents organisms to enter the threatened marine environment, but this
option also requires a relatively large equipment. A disadvantage of this option is that the removed
and/or killed organisms and sediment will either built up in the ballast tanks or have to be given off as
waste in the respective ports.
Treatment during the voyage requires fairly small equipment, because of the time available for
treatment. On the other hand there is no guarantee that all ballast water (including organisms and
sediment) will be treated during circulation over the ballast tanks, mainly because organisms and
sediment have a tendency to settle during the voyage. Also the removed and/or killed organisms and
sediment will either built up in the ballast tanks or will produce additional waste.
This all indicates that treatment at one moment is not sufficient. As the required equipment for
treatment during intake and discharge are of similar capacity, this seems to be the combination that
has the most potential.
The ship itself also gives a set of restrictions to the treatment equipment because of its design
characteristics and operating circumstances, which might prevent well-proven land-based equipment
to be installed on-board a ship without modifications.
The restrictions due to the ships' design are related mainly to the available space and specific ballast
water piping configuration on-board the ship. The main operating constraints relate to the changing

Brouwer: Ballast water treatment R&D in the Netherlands
atmospheric conditions during the voyages, the highly corrosive atmosphere at sea and the limited
availability of crewmembers to operate the treatment equipment.
Ballast water treatment equipment
The treatment of ballast water aims at reducing the risk of viable organisms entering the marine
environment at destination and can include both removal of marine organisms and sediment and
killing of organisms.
Based on the characteristics/sizes of organisms and sediment and the potential of treatment
equipment, it is not likely that one type of equipment will cure the problem sufficiently. This will
result in the necessity of a combination of techniques to cure the problem to the maximum extent
possible, as will be explained below. The effectiveness of each technique will not be discussed.
Techniques to remove organisms and sediment from seawater include filtration, separation, (hydro)
cyclonation and centrifugation. Such techniques are all based on physical properties, like particle size
and specific gravity. The smaller the particles and the smaller the differences in specific gravity, the
more difficult it becomes to remove the particles from the water. Very small particles (< appr. 10 µm)
will be quite hard to remove; such conditions are likely to be found in many locations where ballast
water is loaded, which is often in ports. Also some organisms consist mainly of water and
consequently have almost the same specific gravity as water, which will decrease the efficiency of
especially hydrocyclonation and centrifugation
Since the application of the above mentioned primary techniques cannot be expected to result in
ballast water of the required quality, secondary techniques that kill the organisms are necessary.
Examples that have been applied in ballast water treatment are UV-irradiation, heat treatment,
chemical treatment, ultrasonic treatment, and biological treatment. For all these techniques, organisms
to be killed should be in actual contact with the active ingredient of the treatment. Without primary -
treatment, this will be hardly possible as high concentrations of suspended sediment will be present
and organisms can then easily "hide", from the mortal secondary treatment.
The above justifies the statement that a combination of primary and secondary treatment techniques
will be required. Sediment and larger organisms should be removed as much as possible, to allow for
a high efficiency of the secondary treatment to kill the remaining organisms. The latter will mainly be
of a size below 10 um, and involve (cysts of) algae that contributes to harmful algal blooms, bacteria,
and viruses.
Promising combinations of techniques include filtration and hydrocyclones as primary treatment,
followed by UV-irradiation as secondary treatment. Other combinations are also explored although
investigations are currently in an earlier phase.
A well-designed ballast water treatment system will contain more than just the equipment to remove
and kill the organisms and sediment. Although the system will be type-approved and as such will not
require proof of effectiveness by analysis of samples on each journey, a testing system will be
required for random checks in harbours or for monitoring the equipment by the crew during the
voyage. As the system is type approved prior to use by a proper evaluation method, this system
requires an independent, automated control and register device that will prove the proper use of the
system. For such evaluation system both the sampling and analysis techniques should be robust,
reliable and reproducible and described in a consistent manner and for personnel that should
(routinely) perform the evaluation. Both sampling of treated ballast water and analysis of the sampled
matter should be developed such that they can be applied on a routine basis, by authorities and
management personnel. Elaborate sampling from ballast water tanks or from discharged ballast water
by a range of sampling equipment and analysis of the sampled material by biological specialists are
important steps on the way to evaluate the problem and the quality of ballast water; to meet the
requirements of a full-proofed type test and routine evaluation system, further development is needed.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
For the measurement techniques required for evaluation of the treatment product, a purpose-oriented
adaptation of flow-cytometry is promising for the realisation of an automated measurement of ranges
of particle sizes and forms present in ballast water before and directly after treatment, and upon ballast
water discharge at the end of a journey. For this purpose, automated equipment should be developed,
which allows monitoring of the performance of the installed ballast water treatment system and which
can also be used by the responsible authorities. The more elaborate, research-oriented forms can also
discriminate between life and dead particles.
Ballast water treatment equipment: market potential
A study (Royal Haskoning, 2001) was performed to estimate the market potential for ballast water
treatment equipment. This study used a three-step approach:
· Step 1: defining the relevant part of the world fleet
· Step 2: determine the "qualified available market"
· Step 3: predict the future market behaviour
This study made use of the data of the world fleet, but based its qualitative analysis on information
from Dutch ship owners.
Step 1: defining the relevant part of the world fleet
In 2001 some 91,000 vessels were registered with Lloyds. Part of the registered vessel types does not
use seawater as ballast, or return always to the same port. Examples of such vessels are tugs,
lighthouse vessels, fishing vessels etc. After excluding these vessels a number of appr. 47,000 vessels
remains. Besides the type of vessel, also the area of operation will determine whether a vessel will
need to comply to ballast water regulations. As a measure to determine whether a vessel makes long
voyages (i.e. international or intercontinental trade) the vessel size was used. Most of the world fleet
is actually quite small (see figure 2). In the study it was concluded that all vessels under 1000 tonnes
dead-weight probably have regional modes of operation. Excluding also these category of vessels
yields an estimate of about 33,000 vessels that will in some way face regulations on ballast water
Step 2: determine the "qualified available market"
It was assumed that after the adoption of the international convention (expected in 2004) the main
driving force for installing treatment equipment, during the first 5 years, would be unilateral
legislation based on this convention. Ship owners with sufficient awareness and financial means were
selected to be the short-term market (the first 5 years); this was based on the 52 high-income
countries. The ship owner can either consider retrofitting or phasing out the vessel .
The age-distribution of the world fleet is important to determine the expected amount of new
buildings in the future, and the number of vessels on which retrofit will be likely. Based on expert
opinions, an age of 10 years (dependent on trades, vessel types, ship owner) was deemed the
maximum age on which a vessel may still be considered for retrofitting ballast water treatment
These analyses resulted in an estimate of appr. 675 vessels to be retrofitted and appr. 450 to be newly
built as replacement for old ships per annum for the short-term market.
Step 3: predict the future market behaviour
After ratification of the convention (expected in 2009?), many more ship owners will be obliged to
either retrofit existing vessels or phase out and replace the vessel. The analysis resulted in an estimate
of appr. 2,400 vessels to be retrofitted and appr. 1,050 to be newly built per annum for the mid-term

Brouwer: Ballast water treatment R&D in the Netherlands
market (after 5 years until all existing vessels have been retrofitted). In the long-term the market will
mainly consist of new-builds only.
Potential market prediction
Based on the analysis of the Lloyds' register, the modal vessel is probably a general cargo vessel of
12,000 tonnes dead-weight. According to a survey for the Royal Netherlands Association of Ship
Owners (KVNR), this coincides with approximately 4000 tonnes ballast capacity, and a ballasting
capacity of 600-1,000 m3/h.
Data from suppliers of treatment equipment, provided cost estimates of USD 200,000 (lower estimate
of 600 m3/h) until USD 310,000 (higher estimate of 1,000 m3/h) per vessel for the modal vessel.
For the short-term period (2004 ­ 2009) the annual turnover is estimated to be in the range of USD
225 million to 350 million. After ratification of the convention the potential annual turnover will
increase and is estimated to be in the range of USD 700 million to 1,100 million. The long-term
annual turnover is estimated to be in the range of USD 200 million to 325 million.
These estimates are, of course, subject to a number of uncertainties and constraints. Firstly, the actual
adoption and ratification of the convention is still uncertain and this will be the main determining
factor. Secondly, the appropriate treatment technologies are under development and so far it is not
clear which technologies can and will be used in the future. A last, but not least, aspect is the market
penetration of the equipment suppliers, which will require a thorough marketing strategy.
Full scale ballast water treatment on-board testing
Before on-board test on commercial ships can be performed, land-based and controlled sea borne
(pilot) tests in a research environment (preferably a research vessel) are required to prevent major
setbacks .
The test program, which is being developed by the NIOZ and Royal Haskoning in co-operation with
ship owners and equipment suppliers, includes three main parts, which are (1) a land-based pilot test
close to NIOZ, (2) a controlled pilot test on the NIOZ research vessel and (3) a full-scale test on-board
of a commercial vessel.
This test program will investigate different treatment options and will simultaneously develop
protocols for sampling and analysis. The sampling technique will also require modifications to the
currently available sampling systems, which will be part of the project.
It is also envisaged as being adequate to cover all seasons of a year, so as to meet the variations in
presence and absence of the relevant organisms over the year.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 1: Size classes of Pelagic organisms.
number of vessels
DWT classes (metric ton)
Figure 2. Distribution of dead-weight in the world fleet.

Ballast water treatment - management and research in
Washington State
S. S. Smith
Washington Department of Fish and Wildlife, USA
Washington State has established a ballast water treatment discharge standard and an interim
approval process for technology evaluation. This process is intended to further the development of
promising treatment systems while international and national standards are being established. The
Washington State interim approval process will be described along with ballast data that identified
high-risk vessels. Planned and completed research on various treatment methods will be reviewed.
Treatment systems currently under consideration include filtration and UV light, SeaKleen,
ozonation, chlorine dioxide, filtration and mixed oxidant process, and the use of treated wastewater
for ballast. A cooperative ballast water treatment research and development project was developed to
promote the installation and testing of ballast treatment systems. Vessel operators, regulators,
technology vendors, ports and governments are sharing the risk and cost associated with on-board
testing. The more we install and test technologies, the faster they will improve, and the sooner we will
solve the problem of ballast related invasive species introductions.

Treatment options
Treatment options currently being considered for interim approval:
· Hyde Marine ­ Filtration and UV Light
· Garnett - Sea Kleen
· EcoChlor - Chlorine Dioxide
· NuTecho3 - Ozone
· Marine Environmental Partners ­ Mixed oxidant process
Other treatment systems will be considered for approval as they become available.
Aims and objectives
Washington Department of Fish and Wildlife is striving to improve our capacity to protect
Washington's waters from aquatic invasive species. The development and implementation of
practical, effective and environmentally sound ballast water treatment systems is key to achieving this
goal. Promising treatment systems must be installed on vessels and tested for operational, biological,
and environmental performance. Each installation yields new information that leads to improvements
in future performance. More installations will increase production, which leads to lower prices.
Improved and cheaper technologies will not just happen; they must be allowed to evolve through use.
Installing experimental technology is risky. What if it's not approved for use? What if it's
impractical? What if it just doesn't work? Programs should be developed that allow for the risk to be
shared, thus minimizing the impact to any one group. Vessel operators, regulators, technology
vendors, ports and governments should share the risk to promote more installations. The more we
install technologies, the faster they will improve, and the sooner we will solve the problem of ballast
related introductions. The Washington state ballast management program is designed to identify high-

2nd International Ballast Water Treatment R&D Symposium: Proceedings
risk vessels that need treatment, and provide incentives for vessel operators to install technologies for
evaluation in our interim approval process.
Research methods
Testing protocols are currently under development. They will be designed to evaluate each treatment
system's capacity to achieve our percent removal standards, which are 95% removal of zooplankton
and 99% removal of phytoplankton and microorganisms.
Washington State has taken the following steps to further the implementation of ballast treatment.
· Implemented a mandatory ballast water reporting program that can identify high risk vessels
that need ballast treatment.
· Established a standard for the discharge of treated ballast.
· Created an interim approval process for ballast treatment technologies.
· Mandated treatment after July of 2004, if exchange cannot be conducted.
· Created a cooperative ballast water treatment research and development project to fund the
installation and testing of on-board treatment systems.
· Established the Ballast Water Work Group made up of representatives from the maritime
industry, environmental organizations, and agencies to further the implementation of ballast
Cooperative ballast water treatment research and development projects should be developed around
the world to promote the installation and testing of ballast treatment systems. Vessel operators,
regulators, technology vendors, ports and governments should share the risk and cost associated with
on-board testing. The more we install and test technologies, the faster they will improve, and the
sooner we will solve the problem of ballast related invasive species introductions.

Corrosion effects of ballast water treatment methods
E. Dragsund, B. O. Johannessen, A. B. Andersen & J. O. Nųklebye
Det Norske Veritas
This paper outlines the main mechanisms and factors affecting corrosion rates in ballast tanks and
associated piping. Further, the paper discusses the main ballast water treatment techniques proposed
with emphasis on their potential effects in relation to corrosion. The paper is based on laboratory
studies and desktop evaluations of ballast water treatment methods carried out by DNV over a period
of time. These treatment methods comprise a range of solutions and operating principles and may in
many cases involve a combination of several technologies, in which case the effect of the individual
components has been assessed. A more detailed assessment for ozone treatment is provided as case
example for the purpose of illustrating corrosion testing methodologies.

During the last years ballast water treatment methods have been subjected to extensive investigations
driven by future regulations and anticipated future requirements for the limitation of transfer of
marine organisms. Some of these methods are now commercially available while others are still at the
development stage. Studies of such technologies have had prioritised focus on the ability to eliminate
various classes of marine organisms. However, the operational applicability of the solutions involves
other aspects, of which the potential for increased ballast water tank corrosion is a main concern. For
the maritime industry, corrosion and corrosion protection is a considerable cost element in the
operation of a vessel. Consequently, any method which significantly accelerates corrosion or can
reduce the efficiency of presently applied protective corrosion measures are likely to be discarded
even if the ballast water treatment performance is good.
Corrosion mechanisms in ballast tanks
Electrolytic corrosion
In relation to ballast tanks, main corrosion mechanism is that of electrolytic corrosion (general
corrosion). For steel submerged in sea water, the accessibility of oxygen to the surfaces is the main
controlling factor for the corrosion rate. The increase in present levels of oxygen (oxygen surface
exposure), will act as a corrosion rate catalyst. Corrosion rates will also increase with increase in
temperatures. An often referred to example illustrating this effect is the double hulled tanker carrying
heated liquids compared to that carrying liquids of a lower (or ambient) temperature. The surfaces
facing the heated tanks will experience the higher rate of corrosion.
However, interrelationship between the various factors and conditions must always be considered, e.g.
oxygen content of the seawater will decrease with increasing temperature and thus have a possible
decelerating effect.
Surface protecting coatings will always have imperfections causing local corrosion with the potential
of spreading if not repaired.
As a first approximation it can be stated that the corrosion rate for different low alloy steel grades in
submerged, static condition is approximately the same, independent of minor alloying elements.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
However, for long term exposures the development of rust deposits and their protective effect and
reduction of direct oxygen flux to the steel surface is a critical factor. Present knowledge on corrosion
for various seawater type exposures is not sufficiently understood to accurately predict the
development of rust deposits as function of either the steel grade or the environmental impact.
The corrosivity of sea water as regards general corrosion on steel, increases with increasing:
· temperature
· oxygen content
· water velocity
· content of corrosive compounds (e.g. H2S, CO2, H2O2)
· velocity of eroding particles
Bacterial corrosion
In stagnant ballast water containing organic material (including oil), microbes may thrive. High
corrosion rates are caused by chemical processes initiated by bacterial and/or fungi activity, and often
proceeds in two or three stages:
During initial microbial proliferation dissolved oxygen in water is used up by microbial and
chemical degradation of organic matter. Already at this stage, mildly acidic organic chemicals
are produced by the microbial oxidation processes which may accelerate ongoing electrolytic
corrosion. The zone near the microbial growth becomes oxygen deficient and anodic.
In some cases, conditions are such that a second stage occurs, where one or a few specialist
species of anaerobic bacteria take over the scene, feeding on the acidic chemicals. The best
known are the Sulphate Reducing Bacteria (SRB) which reduce SO 2-
to S2-. Hydrogen
sulphide gas is also an end product of the SRB activity and promotes corrosion.
Strong acids such as sulphuric acid can be produced from sulphides when oxygen becomes
available again. This will further accelerate the corrosion rate.
Bacterial corrosion is found most frequently underneath sludge or dirt settling out from the water on
bottom plating and other up-facing, horizontal surfaces. Water properties affecting bacterial corrosion
· Low oxygen level (anaerobic)
· Hydrocarbons or other pollutants (carbon sources) nourishing bacteria
· Temperature (20 ­ 40°C)
· Sulphates (in sea water sulphates are always present in excess quantities)
Typical corrosion levels in ballast tanks
DNV (1993) presented ballast tank corrosion rates compiled from available sources and from DNV's
ship surveys. It should be noted that reported corrosion rates in literature may be either for one side or
for both sides of the plate. All rates reported in the following refer to one side.
Corrosion rates vary between different parts of the ballast tank:
· Corrosion rates of bare steel fully submerged in sea water (i.e. the lower part of the ballast
tank) are usually 100 ­ 200 µm/year.
· In the splash zone (mid section of the ballast tank) corrosion rates can be 200 ­ 400 µm/year.
· Highest corrosion rates (350 - 400 µm/year) were found in the upper 2 meter of Side Plating.
This was explained by the combined effect of increased average temperature due to sun

Dragsund: Corrosion effects of ballast water treatment methods
heating, abundant oxygen supply, splashing of sea water, and cyclic temperature changes
leading to cyclic condensation (wet and dry).
However, the variation in the reported rates is large leading to a "normal" range roughly indicated as
200 ± 200 µm/year, and in extreme environments (such as high temperature and bacterial corrosion
underneath sediment) corrosion rates may be several millimetres per year. The variation depends on a
number of factors e.g. (DNV, 1996):
· A layer of built up corrosion products (rust) on a steel surface will have a protective (coating)
effect by limiting the access of oxygen to the steel, thus lowering the corrosion rate.
· A layer of corrosion products may render parts of the surface cathodic in relation to other,
anodic, parts of the surface lacking such layer experiencing increased corrosion rate.
· Surfaces exposed to vibrations and/or high stress levels may have increased corrosion rates
with time, due to the thickness reduction of steel plates causing reinforced vibrations and
stress levels.
· Macro-elements or large aeration cells caused by variations in oxygen concentration, e.g. at
different depth levels in ballast tanks and over or under sediments, may create anodic parts
experiencing accelerated corrosion and other parts cathodic, non-corroding.
· Areas with locally degraded coating may become anodic compared with intact coating,
resulting in pitting corrosion.
Corrosion protection in ballast tanks
The SOLAS Amendment Corrosion prevention of seawater ballast tanks gives direction to include
corrosion protection of ballast tanks in oil tankers and bulk carriers within the scope of classification.
A good coating applied on a well prepared surface at the newbuilding stage is the most effective
means of avoiding corrosion. Coatings will have varying useful lives in ballast tanks, from a few
months to more than 25 years, depending largely on steel surface and edge preparation and
application conditions. An important contributing factor to coating degradation is the increasing
brittleness and loss of flexibility with time, causing cracking and disbonding at structural hotspots,
typically in deckhead structures.
Some of the proposed ballast water treatment systems may accelerate this degradation. This is in
particular cyclic high temperature (heat treatment) causing loss of low molecular weight components
of the coating, and oxidation (treatment by oxidants) of the coating constituents which further
contribute to gradual loss of flexibility and may lead to disbonding.
Ballast water treatment technologies and their expected effects on corrosion
This chapter summarises assessments of potential corrosion effects of various presently known ballast
water treatment methods. The assessment focuses on the treatment categories.
A range of ballast water treatment methods have been developed or are subjected to research and
development. Ballast water treatment systems may involve a single method or a combination of
several principles. The methods may have an instantaneous biological effect or require a prolonged
time for exposure of organisms to achieve the desired effect.
The methods have for convenience been grouped by their general operating principle according to
three main classes:

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Mechanical methods
Methods by which organisms are physically removed, completely or partially, at the ballast water
loading stage resulting in reduced organic material in the ballast tanks. This may provide conditions
that are less favourable for bacterial growth and thereby reducing the oxygen consumption in ballast
water. As a result, this class of methods may lead to higher oxygen levels in ballast water which is
favourable to general corrosion. The effect is not believed to be dramatic.
A higher oxygen level combined with reduced organic material presence, may reduce bacterial
Examples: filters, hydrocyclones, separators. The method does not necessarily imply neutralisation of
Physical methods
Methods by which organisms are rendered harmless (neutralised) by a physical method but remain in
the ballast water to sediment, degrade and/or be offloaded. Typically the methods work by physically
damaging the structure or tissue of the organisms by rapid stress variations on a scale comparable to
the dimensions of the organisms. The methods vary in required exposure time for the desired effect to
be achieved. Examples are ultrasound, pressure fluctuations, aggressive mechanical methods,
temperature treatment (heat), UV, nitrogen/air supersaturation and cavitation.
Ultrasound/Cavitation/Aggressive mechanical methods
Physical disruption of organisms will release easily degradable organic material resulting in a
gradually reduced oxygen level. General corrosion rates may therefore be reduced. Due to the
same effect these methods may provide conditions that promote bacterial corrosion.
Generally increased temperatures lead to higher corrosion rates. However, as a secondary effect,
high temperatures will lead to reduced oxygen level which may lead to lower general corrosion
rates. Organisms killed by treatment may be basis for bacterial decomposition. Reduced oxygen
level and increased carbon sources may lead to bacterial corrosion.
The real effects on corrosion rates related to heat treatment are not clear and should therefore be
Chemical methods
This class includes any method by which aquatic organisms are neutralised by addition of any active
chemical substance. The class may be further subdivided into group of methods. Examples are:
Organisms are neutralised by addition of an oxidant or any other substance that will form oxidants
in reaction with the sea water. Examples: ozone, hypochlorite, chlorine, hydrogenperoxide,
hydroxyl radicals and several combinations called multioxidants.
Increased oxidant level in the sea water will normally lead to an increase in the corrosion intensity
that may range from mild to severe. The corrosion effect may be reduced with time as oxidant is
consumed by the organic material oxidising process. On the other hand immediate corrosion may
be severe at the location of oxidant introduction. Treatment by ozonation leads to higher oxygen
and oxidant level. The build up of rust deposits and their protective effect by reducing the oxygen
flux to the steel surface may substantially change corrosion rates, in particular in short term tests
of corrosion.

Dragsund: Corrosion effects of ballast water treatment methods
Organic macromolecules easily available as nutrient for bacteria will be oxidized thereby
reducing an important source for bacteria growth. These methods are therefore believed to have a
positive effect on bacterial corrosion.
Organisms are neutralised by addition of a biocide or any substance that will form a biocide in
reaction with sea water. Examples: acrolein, formaldehyde, copper sulphate and varying brands of
Unless containing oxidising substances, biocides are not likely to have any effect on general
corrosion levels. However, the organic content in the ballast tanks will not be reduced and effects
on environmental conditions affecting corrosion will therefore be similar to physical methods.
Nutrients or other chemicals are added into the ballast water leading to anaerobic conditions in the
ballast tank. This will neutralise most animals and algae, but not anaerobic bacteria and
Reduced oxygen content will lead to lower corrosion rates. Bacterial corrosion may, however,
increase. If water contains normal oxygen levels for surface water when pumped into ballast
tanks, this may lead to high pitting corrosion by oxidizing sulphides to sulphuric acids.
Table 2 summarises expected corrosion effects for some of the discussed treatment methods.
Table 2. Potential effects of some treatment methods on corrosion in ballast tanks. For details see text. =
Increased level, / = Unchanged level, = Reduced level. Expected corrosion rate: + = higher, - lower, / =
Unchanged. ? = not clear.
Effect on selected environmental conditions
Expected effect on corrosion rate
Total effect



Heat treatment



- ?


* = increased CO2 level.
Conclusion - testing of corrosion effects
Corrosion effects are only occasionally included in tests carried out on ballast water treatment
systems. If included, methods used are most frequently measurements of linear polarisation resistance
or redox-potential. As illustrated in the case example below, the testing of corrosive effects of
ozonation, such measurements can only be indicative and seldom conclusive. Long term tests using
standard coupons should be carried out if significant corrosion effects are feared or likely.
When tests of corrosive effects are carried out in the laboratory, it is important to simulate a real
ballast situation as exact as possible. Corrosion rates in ballast tanks vary between different levels in
the tank. Highest corrosion rates are experienced in the upper segment of the tank. Temperature,
abundance of oxygen supply and splashing of sea water is important factors impacting the corrosion
rates. The tanks should therefore be stirred or moved during the experiment.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The build up of rust deposits and their protective effect by reducing the oxygen flux to the steel
surface may substantially change corrosion rates, in particular in short term (1 ­ 3 months) tests of
corrosion. It is important to keep in mind that a stable "steady state" corrosion rate will not be
achieved before an exposure period of about 6 months.
Furthermore, the water used in experiments often has relatively low organic content and the set-up
seldom includes tank sediments.
The corrosion rates as such are therefore not directly comparable to actual corrosion rates found on
ships in service. This implies that it is the relative differences between the results from the
experiments that are of the greatest interest.
A case example ­ testing of corrosion effects of ozone treatment
A feasibility study of ozone treatment of ballast water was carried out including both biological
efficiency and corrosivity measurements. Similar to several other oxidants, ozone reacts with seawater
and produces a number of corrosive compounds (e.g. several forms of bromine and chlorine). These
corrosive compounds were found to decay after a period from some hours to more than one week
following treatment. The decay rate is a function of ballast water characteristics (presence of organic
compounds, metal ions and organisms). In polluted water the decay rate will be higher compared to
clean water.
The corrosivity study included two phases:
· Short term; determination of corrosion rates based on linear polarisation resistance (Rp)
immediately after treatment. The platinum wire was also used as an electrode for measuring
the redox-potential.
· Long term; this phase was designed to reflect typical ballasting scenarios over a three month
time period and was based on the long term exposure of bare steel and coated coupons.
Weight loss of the bare steel coupons were used as measure for corrosion while visual
observations and coating disbonding were used for the coated coupons. The coupons were
exposed by repeatedly emptying the ballast tanks and refilling with ozonated seawater.
Short term test
1. The corrosion of carbon steel in an ozone-injected system was found to be about 500% higher
than the corrosion rate of steel in normal oxygen rich seawater. The free oxidant level at the
electrode position was 0.85 ppm or higher in all tests.
2. This initial corrosivity did not change due to the presence of the algae or other types of organic
In Figure 1, the electrode to the right was exposed to ozone treated water in one test compared to
untreated electrode.
In real ballast water scenarios, the oxidant concentration will decay after hours to days depending on
water quality. To give an estimate of a maximum level of corrosion based on the Rp measurements,
the following approach was adopted:
· Extensive inspection and surveys done by DNV have established that the average general
corrosion rate of carbon steel in ballast tanks are around 0.1 to 0.2 mm/year (for single side of
a tank).
· This represents the Rp results of no ozone in the test. As an approximation, this measured
corrosion rate is assumed to represents the "base-line corrosion" equivalent to an in-service
corrosion at about 0.1 to 0.2 mm/year. Based on this assumption, the corrosion rates for the

Dragsund: Corrosion effects of ballast water treatment methods
other exposure conditions can be calculated from the "base-line corrosion". This implies that
the corrosion rate of steel will increase to 0.5 to 1 mm/year (for the single side of a tank)
provided the presence of chlorine at a level of 0.85 to 5 ppm.
· The rates of corrosion estimated above assume the presence of a constant level of chlorine
over a period of one year. This will not be the case in a ballast tank in a vessel in normal
service. Oxidant decay will incur immediately following ozonation (ballasting). The
established elevated levels of oxidants can be expected to last from some hours up to 2-3
days. This will dramatically change the corrosion picture. Annual corrosion effects due to
ballast water ozonation will depend upon the ballasting pattern (no. of ballast voyages/ quality
of ballast water).
This approach for estimating maximum corrosion rates, is only a first estimate. Only long term testing
can establish accurate corrosion rates for a given ballast scenario.
Long term test
The tests were designed to simulate three sailing schedules/scenarios, which resulted in the following
Ballasting once per week, i.e. 3.5 days full, 3.5 days empty, totally 12 fillings.
Ballasting every second week, i.e. 7 days full, 7 days empty, totally 6 fillings
Ballasting once per month, i.e. 14 days full, 14 days empty, totally 3 fillings
Corrosion was monitored using steel coupons mounted at three different levels in the test tanks. The
tanks were filled to a level of approximately 80% of the tank height, always leaving the upper steel
coupons out of water. Similarly the tanks were emptied to about 20% always leaving the bottom row
of steel samples permanently in the water phase. For each of the three scenarios, a reference test with
untreated water was carried out.
The steel coupons had three different surface qualities:
Bare steel (grit blasted)
Primed with 15 µm zinc silicate and coated with 2 layers of modified epoxy (totally 300 µm)
- intact
Primed with 15 µm zinc silicate and coated with 2 layers of modified epoxy (totally 300 µm)
- scribed
An average corrosion rate was calculated for the total area of the plates based on the weightloss of
each sample. This provided a "one side corrosion rate". It should be noted that this average corrosion
rate does not provide information on local corrosive attacks.
Corrosion rates did not achieve a steady state during the testing period. The decline in
corrosion rates that was observed in each individual tank over the project period, was caused
by increased quantities (thickness) of corrosion products forming a protective layer covering
the metal surface.
In the lower segment of the tank (permanently submersed in water) testing showed increased
corrosion rate as a result of ozonation. This is assumed to be due to the increased level of
oxygen and corrosive compounds. Normal corrosion in this part of ballast tanks, which is a
minor part of the total tank area, is 100 ­ 200 µm/year indicating an increase if ozonated to
200 ­ 400 µm/year. The absence of sediments may have contributed to this increase.
In the mid-section (air and water) testing demonstrated a lower corrosion rate in the ozonated
tanks than in the untreated tanks. This is assumed to be due to the initial corrosive product
resulting from ozone treatment being denser, less permeable to air and hence more protective.
In a ballast tank this protective layer will probably peel off after some time due to ship

2nd International Ballast Water Treatment R&D Symposium: Proceedings
movement and wave induced loads on the ballast tank side plating. This, however, will also
be the case for uncoated tanks exposed to untreated ballast water. It may therefore be assumed
that ozonation will not represent an addition to normal corrosion rates (200 ­ 400 µm/year) in
the mid- section of ballast tanks.
In the top segment (air only), very limited corrosion was found and the testing did not show a
significant difference in corrosion rates between the treated and untreated tanks. The test
results for this segment is, however, assumed to have limited relevance for ballast tanks as
condensation and splashing normally will occur, wetting also the upper part of the tanks. The
highest corrosion rates (approximately 400 µm/year) in ships ballast tanks are observed in this
segment. This has been explained by the combined effect of increased average temperature
due to sun heating, abundant oxygen supply, splashing of sea water, and cyclic temperature
changes leading to cyclic condensation of water and drying. These factors will not be affected
by ozonation and it is therefore assumed that the corrosion rates will be unchanged after
Ozonation clearly affected the bonds between the topcoat and primer close to coating defects.
In total more than 80% of the coated plates from the mid- and lower level in ozonated tanks
showed disbonding of coating, compared to 0% from the untreated tanks. The oxidising
properties of the ozonated seawater may have affected the properties of the primer at the
topcoat interface leading to a reduced bonding between the primer and the coating.
Figure 2 shows coated coupons with scribed surface showing disbonding of coating after ozone
Askheim, E., Nakken, O & Haugland, B.K. 2000. New class rules for protective coating. Presentation
at the 8th ICMES/SNAME, May 22-23, 2000.
Beech, I., Campbell, S., Mills, G. & Walsh, F. 1996. Engineering problems caused by microbial
corrosion and their prevention. Corrosion Management, June/July 1996.
Brown, B.E. & Duquette, D.J. 1994, A review of the effects of dissolved ozone on the corrosion
behaviour of metals and alloys. Corrosion 94. Paper no. 486.
Campbell, S.A., Scannell, R.A. & Walsh, F.C. 1988. Microbially-assisted pitting corrosion of ships
hull plate. UK Corrosion 88, Brighton 3-5 October 1988, pp. 137-157.
Cleeland, J.H. 1995. Corrosion risks in ships' ballast tanks and the IMO pathogen guidelines.
Engineering Failure Analysis, 2 (1) pp. 79-84.
DNV, 1993. Corrosion wastage rates and corrosion additions in ships ballast tanks. DNV Report 93-
DNV, 1996. Corrosion protection of ships. Guidelines no. 8. July 1996.
DNV, 1999a. PLUS and COAT notations. Fatigue and corrosion prevention. DNV brochure.
DNV, 1999b. Corrosion prevention of tanks and holds. Classification notes, No. 33.1. July 1999.
DNV, 1999c. Bacterial corrosion in ballast tanks. Problem and solution. DNV Brochure 1999.
DNV, 2000. Ballast water treatment by ozonation. DNV Report no. 2000-3229.
Korbin G. (Ed. ) Microbiologically influenced corrosion. Nace International, 1993.

Dragsund: Corrosion effects of ballast water treatment methods
Figure 1. Electrode to the right was exposed to ozone treated water in one test compared to untreated electrode.
Figure 2. Coated coupons with scribed surface showing disbonding of coating after ozone treatment.

Session 5:
Test Protocols and
Verification Procedures

A proposed frame-work for approving ballast water
treatment technologies
D. Mountfort, T. Dodgshun & M. Taylor
Cawthron Institute
New Zealand
Introduction, aims and objectives
Recent progress by the Marine Environmental Protection Committee (MEPC) toward achieving an
interim standard for ballast water treatment (eg the draft report on ballast water management [MEPC
48/WP.1.5]), represents an advance in quality assurance beyond the current protocols for ballast water
management, which are essentially restricted to mid-ocean ballast water exchange
(http://globallast.imo.org). With the pending implementation of a ballast water treatment standard in
2003, it is necessary to develop systems under which the performance of a new treatment technology
can be measured. In this regard we consider it important to also introduce the following:
A framework to evaluate the performance of new treatment technologies;
A certification system leading to the ultimate approval of a technology;
An effective management system for performance review and certification.
In this paper we outline how these objectives might be achieved and describe the operation and
management of the proposed performance evaluation system. Our proposed framework is formulated
on the experience of our team and our collaborators, the Centre for Research on Introduced Marine
Pests (CRIMP), in developing a systems-based approach for risk assessment for marine pests in
Australasia (eg Hayes and Hewitt, 1998) and builds on Cawthron's expertise in the validation and
approval of new methods for marine pest management (e.g. harmful algal blooms and the Asian kelp,
Undaria pinnatifida). It also builds on our experience in developing new methods for ballast water
treatment (Mountfort et al., 1999 a,b; Mountfort et al., submitted) and in working with the shipping
industry in ballast water management since 1995 testing the efficacy of ballast water exchange and
shipboard treatment systems (Taylor and Bruce, 1999: Mountfort et al., 1999c; Mountfort et al.,
2003). More recently we have initiated new international partnerships evaluating the efficacy of
biocides and ozonation.
We provide a rationale for the proposed framework and describe how ongoing developments in the
refinement of international standards for sampling and treatment of ballast water would be
accommodated. The framework aims to provide a pathway for the approval and implementation of
new treatment systems so that the process will have a minimum impact on the shipping industry.
Advancement of the performance evaluation system for new treatment technologies
The "moving target" of global protocols and standards
Any new framework should be flexible enough to accommodate both the present "state-of-the-art"
standards for both ballast water treatment and sampling, and future developments in these areas. For
example, future standards might include the use of representative indicator taxa for the certification of
a particular ballast water treatment system.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The proposed ballast water treatment approval framework
Currently there are a number of agencies involved in the formulation of approval methods for ballast
water treatment. These include Det Norske Veritas (DNV) who, in co-operation with the Norwegian
Institute for Water Research (NIVA), are developing standard methodologies for evaluating the
biological effectiveness of a given treatment (http://projects.dnv.com/). Also, the work of Taylor and
Rigby (2001) emphasises the need to consider design aspects of ballast water treatment systems
during shipboard operations. Such initiatives are timely in the context of developing an internationally
accepted approval system. In our proposal, which stems from our earlier considerations on future
strategies for ballast water treatment (Mountfort, 2000), the following points are considered:
1. The need to minimise the intrusion on ships' operations in the early phase of treatment
2. The need to minimise costly shipboard installation of inefficient or inappropriate units that
have not undergone an appropriate preliminary testing regime;
3. The need to recognise that the criteria for assessing the performance of a shipboard system
may differ from those required for pre-shipboard testing;
4. The need to minimise scientifically indefensible commercial bias;
5. The need to obtain a performance and provisional compatibility assessment before installation
on any ship.
Consequently, we are advancing a three tier framework for the approval of ballast water treatment
systems. The essential components of the framework are:
Tier I
Testing of a promising treatment in an IMO approved testing facility;
Tier II
Pending tier I approval, shipboard assessment of the performance of the treatment or
treatment facility;
Tier III
Ongoing performance review of the technology after it has been certified for shipboard
The criteria for approval (Tiers I and II above) would be based on standards and protocols ratified into
IMO convention by the Marine and Environmental Protection Committee (MEPC) and would include:
Performance of the treatment system measured against the IMO approved ballast water
treatment standard and using an approved international standard for ballast water sampling;
Assessment of the environmental impacts from discharged treated ballast;
An assessment of compatibility with candidate ships including factors such as:
- Ship type
- Ship design including ballast tank configuration
- Ship's operations
- Ship safety
An economic appraisal of the treatment, especially in relation to point 3 above.
Treatment systems proposed by a vendor would be assessed and scored under each criteria. Thus for
example, assessments conducted in regard to point 2 might score filtration more highly than for
biocide treatment. Importantly, the framework recognises that the ballast water standards and
protocols, as set out by the IMO convention, are likely to change over time with new treatment
technologies and member-state priorities. Figure 1 shows how the proposed system would
accommodate such changes over time. As an example of how this might operate, the interim standard
may require that a IMO approved testing facility (Tier I) tests to a standard based on a complete kill or
removal of organisms greater than 100 microns in size. However in the longer term, the testing
facility might be required to test to a different standard, possibly one in which the maximum size of

Mountfort: A proposed frame-work for approving ballast water treatment technologies
the test organisms is reduced, or one in which a universally applied set of indicator taxa is used to
assess performance.
We anticipate that the criteria used for assessing the performance of the technology in a shipboard
situation (Tier II) would be similar to those for the treatment facility (Tier I) except greater emphasis
would be placed on criteria 3 and 4.
In Figure 2 we show diagrammatically how the tiered system of treatment approvals might work in
practice. Implicit in the scheme would be the overseeing of the testing for Tiers I and II, and
performance review (Tier III) by an independent IMO approved expert (or experts). The make-up of
the team might be different for each of the three Tiers. For example, the team for Tier 1 testing might
be limited to an IMO approved scientist. In Tier II testing might include an IMO approved scientist, a
ship's engineer and could include consultation with a wider group including an economist.
We contend that once installed, any newly certified treatment technology should be periodically
subjected to a performance review (Tier III) by IMO approved experts. We suggest that performance
reviews should be carried out on all ships equipped with treatment systems for the first 5 years after
implementation of the interim ballast water treatment standard, and that at least one inspection is
carried out on all units installed during this period. The rationale for this is that (i) the interim standard
will apply for approximately this period, and (ii) the number of ships equipped with ballast water
treatment systems will be relatively few compared to those carrying out ballast water exchange. Over
the longer time period, the system of performance review may need to be revised according to the
number of ships with treatment systems onboard and the development and implementation of new
treatment standards and protocols.
Among the criteria that we believe should be considered for assessments during performance
monitoring are:
comparison of performance against that at the time of certification (see criteria for Tiers I
and II);
inspection for deterioration of plant;
where appropriate, inspection of ships log to determine use of plant.
Should the treatment continue to meet the standards under which it was initially granted, then its
certification should be extended.
How the framework would be implemented
The framework would essentially operate on a report and recommendation basis for each Tier.
Because of the anticipated ongoing improvements to ballast water standards and protocols, it would
be necessary to consider the timing and duration of approvals and certification. Thus, in the case of a
technology that has been approved under Tier I, testing in a shipboard situation should be completed
within a period of time that is consistent with current standards and protocols.
Similarly, approval for shipboard installation should be valid for a minimum period (e.g. 5 years).
Extension of the approval would be subject to performance review on the basis of the criteria for
which the technology was originally approved. On the other hand new treatment technologies entering
the system would have to meet the requirements of any new standards and protocols. We believe our
framework allows for reasonable life-times for treatment systems once installed, but at the same time
is flexible enough to meet the requirements of standards and protocols set by international convention.
The framework also minimises the need to replace newly-installed treatment equipment (eg by retro-
fitting) in order to meet a new standard.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Additional considerations
In addition to the criteria set out for assessing the performance of new treatment systems (Tiers I and
II above), some countries may elect to have stricter standards for the management of ballast water
discharge. These might be applied, for example, after a risk-based assessment of specific ships or
shipping pathways. We believe, therefore, that the framework finally adopted will need to be
sufficiently robust to accommodate such variations.
Conclusions and recommendations
To date there has been very little progress in the development of an effective framework for the
approval of ballast treatment technologies. With the interim standard for ballast treatment
technologies being included in the pending IMO convention for ballast management, a need has arisen
for a framework under which approvals for new treatment systems can be granted. This would ensure
that ballast water treatment facilities are able to meet the standards and protocols set by international
convention. We have outlined how this could be achieved by proposing a three tier framework of
performance testing and review. The framework would:
1. Minimize the impacts on existing ships' operations resulting from the installation and trial of
treatment systems that are either inappropriate or have not met the required standard;
2. Minimize the environmental impact that may be caused by an inappropriate treatment or one
that has not met the required standards;
3. Minimize and structural damage and safety concerns resulting from a treatment that has not
been subjected to performance assessment and review
Remove scientifically indefensible commercial bias in the advancement of new treatment
5. Provide an internationally recognized framework for approval of ballast treatment systems;
Maximize the efficiency with which new treatment technologies can be certified for shipboard
7. Protect the vendor from liabilities that may arise from adverse structural, health or
environmental impacts.
Key future considerations for implementation of the framework include:
Refinement of the criteria for certification;
The composition of the IMO approved teams who would oversee and report on performance
testing and review;
Certification period and frequency of performance review.
We express our appreciation to the shipping industry particularly in providing access to vessels for
ballast management investigations. We express also our thanks to the New Zealand Ministry of
Fisheries (MFish) for their support in preparing this paper, and in particular to Elizabeth Jones and Dr
Chad Hewitt. We also thank members of the Biotoxin Monitoring Team at Cawthron for providing
valuable information in regard to the granting of approvals for new methods in phytoplankton toxin
analysis. We also thank the New Zealand Foundation for Research, Science and Technology for
supporting our work on ballast water management since 1996.

Mountfort: A proposed frame-work for approving ballast water treatment technologies
Hayes, K.R. & Hewitt, C.L. 1998. Risk assessment framework for ballast water introductions.
Technical Report No 14, Center for Research on Introduced Pests, pp 1-75.
MEPC ANNEX. 48/WP.1.5. 2002. Draft international convention for the control and management of
ship's ballast water and sediments. 31 p.
Mountfort, D.O., Hay, C., Dodgshun, T., Buchanan, S. & Gibbs, W. 1999a. Oxygen deprivation as a
treatment for ships' ballast: laboratory studies and evaluation. J. Mar. Environ Engg. 5: 175-192.
Mountfort, D.O., Hay, C., Taylor, M., Buchanan, S. and Gibbs, W. 1999b. Heat treatment of ships'
ballast water: development and application of a model based on laboratory studies. J. Mar. Environ
. 5: pp. 193-206.
Mountfort, D.O., Dodgshun, T., Gibbs, W. & McCallin, B. 1999c. Towards a feasible heat treatment
system for ships' ballast water. Proceedings of AAPMA meeting, Brisbane, May 5-6, pp. 125-126.
Mountfort, D.O. 2000. Ballast water treatment-the future. In "Developing a strategy for Marine
Biosecurity in New Zealand" Proc 1st National Workshop on Marine Biosecurity, Nelson, pp. 19-21.
Mountfort, D.O., Parker, N. & Oemcke, D. Effect of UV irradiation on viability of micro-scale and
resistant forms of marine micro-organisms: implications for the treatment of ships' ballast water
(Submitted J. Mar. Environ Engg).
Mountfort, D.O., Dodgshun, T. & Taylor, M. 2003. Ballast water treatment by heat: New Zealand
laboratory and shipboard trials. In Raaymakers (ed) 1st International Ballast Water Treatment R&D
Symposium, IMO London 26-27 March 2001: Symposium Proceedings
. GloBallast Monograph Series
No 5. IMO London. pp. 45-50.
Taylor, A.H. (1996). Design considerations for ballast water control and treatment. IMAS. Paper 10:
pp. 23-29.
Taylor, M., & Bruce, E. 1999. Mid ocean ballast water exchange: Shipboard trials of methods for
verifying efficiency
. Cawthron Report No. 524.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 1. Response of the tiered system for ballast water treatment approval to perceived changes in the ballast
treatment standard over time.
Figure 2. Scheme for the approval of new treatment technologies.

Ballast Water Treatment Verification Protocol - DNV
A. B. Andersen, B. O. Johannessen & E. Dragsund
Det Norske Veritas
The ballast water treatment challenge has inspired numerous owners of technologies of a wide range
of associated applications resulting in a significant volume of development projects parallel to the
development of the International Convention for the Control and Management of Ships' Ballast
Water and Sediments
(the Convention). Main focus has been that of treatment efficiency and a
number of different assessment approaches have emerged. DNV have also assessed various proposed
treatment concepts.
Evident need for a uniform reference for performance assessments resulted in the development of the
DNV "Model Group Concept" which was presented to the Marine Environmental Protection
(MEPC) at its 46th session. (MEPC 46/3/8). This methodology introduced an approach for
setting standards and allowed performance comparisons between different ballast water treatment
concepts with respect to treatment efficiency to be undertaken.
Experience from the application of this methodology to actual concept assessments including
laboratory work and inputs from the iterative processes related to the ongoing development within
MEPC of the Convention, have formed the basis for the development of a protocol covering a wider
range of issues requiring consideration in light of the issuance of compliance documentation. The
DNV protocol will cover:
· Application and Feasibility.
· Occupational Safety and Health (OSH).
· Treatment Efficiency.
The protocol and its appendices are under development and will be available following the adoption
of the Convention.
Aims and objectives
DNV have established the project; Ballast Water Treatment Technologies ­ Standards for
for the purpose of developing a protocol or standards for certification for the approval of
onboard installations of ballast water treatment systems. These standards will form procedures and
rules that will be developed in coordination with the Guidelines for Type Approval of Ballast Water
Treatment Systems
currently being drafted by IMO. The standards will provide as an Operational
Performance and Testing Programme
that may be adopted by any Party to the convention.
Research methods, test protocols and experimental design proposed
The development of standards for certification is a process representing a methodical approach itself
that requires the definition of:
· Principles
· Acceptance criteria

2nd International Ballast Water Treatment R&D Symposium: Proceedings
· Required information
In order to assess compliance and thus issue an approval document, the considered treatment system
must meet the requirements of the standard which includes the following four areas:
Provision of satisfactory documentation: Compliance to the standard of certification.
Occupational safety and health norms: Safety margins with respect to risks-levels (ship, its
crew, etc.).

Demonstrate sustained performance to meet any applicable IMO conventions: Ballast Water
Performance Standard.

Demonstrate no present unwanted effects: Environmental, other.
The conformity assessment process may be illustrated by step 1, 2 and 3 as illustrated in figure 1. The
starting and ending points represented by Phase I and Phase II respectively, relates to:
Idea and investigation: This represents the motivation of the owner of the treatment
technology or system.
Certification: The issuance of compliance documentation following a successful conformity
assessment process by an Administration (or an appointed party).
Steps 1, 2 and 3 represent the procedures of the actual protocol. The process of identifying content,
criteria and assessment procedures in relation to the areas covered by the standard have rested upon
actual system assessments. These have included the following:
General review (relates to step 1, figure 1):
Technology considerations (feasibility assessments) focussing on occupational safety and health,
potential shipboard impacts, possible constraints in relation to operations (practicality, long term
effects, etc.), treatment sequence and performance characteristics.
The general review has rested upon available literature and in-house knowledge and experience.
Initial small scale laboratory tests have also been performed to improve the understanding of
methodologies and interrelations when applied together in a system.
Laboratory studies (relates to step 2, figure 1):
For the purpose of evaluating treatment performance and thus to provide recommendations on
aspects related to treatment performance reliability, potential scaling effects and particulars of the
treatment concept requiring attention (treatment sequence, occupational safety and health aspects,
energy consumption, residual products, etc.), meso-scale laboratory set-ups have been established
and testing performed.
The driving principle when sizing the laboratory treatment system has been that of simulating the
actual full scale process by dimensioning with emphasis on treatment speed. Thus, the speed of
flow throughout the system including piping to ballast tanks (holding time) has been included.
Conditions in ballast tanks have been simulated (darkness, limited ventilation) over time.
However, our work has not included motions and vibrations as will be an integrated factor in a
full scale application and may impact aspects associated to the operations of some technologies.
In cases where the general review has identified particular aspects of concern, these have been
encountered for in the laboratory testing, e.g. corrosion/ coating impacts.
Treatment performance has been undertaken by applying the principles of the Model Group

Andersen: Ballast Water Treatment Verification Protocol - DNV
Model Group Concept:
The diversified presence of organisms in an ecosystem can be organised and represented by
defining groups of organisms ­ model groups. The model groups are exposed to the treatment
process of the system under consideration and efficiency may then be measured. The concept is
similar to that of eco-toxicological testing (in reverse).
Selection criteria for species for each model group and further, the number of model groups
required will reflect the applicability of the method. When the concept is applied for the purpose
of assessing ballast water treatment system performance, the selection criteria must mirror the
global perspective.
Full scale verification (relates to step 3, figure 1):
Based on output from the above procedures, full scale verification procedures for the treatment
systems tested have been identified. These have included:
· System performance (operational verification)
· Treatment performance (sampling under varying conditions)
· Effects (atmospheres, residuals, etc.)
The above approach has been applied to mechanical, physical as well as chemical treatment systems.
Based on experiences and findings, the frames of a protocol or a standard for certification has been
The draft Ballast Water Treatment Verification Protocol outlines the DNV process for the approval of
ballast water treatment systems to be installed onboard DNV classed vessels.
The protocol, as illustrated in figure 2, consists of three sections:
· Protocol infrastructure: Introduction, Definitions, Conformity Assessment Procedures
· Criteria: Regulations
· Tools: Appendices
The protocol will be reviewed in accordance to any revisions associated to the development of the
Protocol infrastructure
The three introductory chapters (see figure 2) explain the area of application of the protocol and how
it is applied when used for compliance assessments.
For practical purposes, the protocol has established a number of definitions. Some examples are listed
Ballast Water Treatment Systems
Applies to the collection of all components that make up a complete, operating ballast water
treatment arrangement, from ballast water intake to discharge.
Ballast Water Treatment Systems Components
Applies to any of the various components that make up the complete ballast water treatment
system (e.g. filtering unit, primary treatment unit, piping, pumps, control units, etc.).

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Ballast Water Treatment Units
Applies to the component of the ballast water treatment system specifically designed to remove or
render harmless (extinguish) organisms. A ballast water system may be composed of one or
several ballast water treatment units.
Active substance
Applies to any substance added to, or produced by the ballast water treatment unit, with the
purpose to remove marine organisms or render such organisms harmless.
Active physical process
Applies to any physical process utilised by the ballast water treatment unit, with the purpose to
remove marine organisms or render such organisms harmless.
The procedural mechanisms of the protocol are also explained
1) Approval of scaled down system
a. performance testing
b. safety, functionality, quality, documentation
c. unwanted effects

2) Desktop scaling of test results
3) Approval of large scale arrangement

System compliance criteria are formulated as regulations. The areas considered are listed in figure 2
and follows the methodology represented by the conformity assessment processes.
System performance testing
The performance references include all relevant Regulations of the Convention with emphasis on
Regulations E-2 and E-3.
The system performance must be demonstrated through either:
1. Testing of the installed full scale system or
2. Laboratory testing of the full scale system or
3. Testing of a scaled version of the system
Performance testing is subject to the procedure described in Appendix A in the draft DNV
protocol (see figure 2).
For scaled testing, the performance of the full scale system must be established through
documented and scientifically approved methods. A guideline for scaling of test results is also
provided (Appendix B, Guidelines for scaling of performance test data (currently under
development). Furthermore, if the system incorporates two ore more dissimilar ballast water
treatment units, the overall performance estimate must be based on individual testing of all units.
The estimated performance of the scaled-up version of the system must take into effect any
interaction between the two independent systems.
Safety, functionality, quality and documentation
These aspects are all interrelated. Requirements to these are based on relevant applicable existing
standards and norms (e.g. ISO standards, EU Directives (Directive on Marine Equipment), ILO
requirements, DNV rules, etc.).

Andersen: Ballast Water Treatment Verification Protocol - DNV
Unwanted effects
This relates changing characteristics of the ballast water as a consequence of the treatment.
Unwanted effects are those affecting:
- Ballast water tanks and/or associated systems including pumps, piping, valves, etc.
- The environment following the discharge of treated ballast water
The standards for certification refer to recognised standards and norms in relation to both of these
Standard for performance testing of ballast water treatment units
This represents the methodology to be applied for collecting quantitative performance data for ballast
water treatment units under controlled test conditions. A ballast water treatment unit is defined as the
component of a ballast water treatment system that kills, removes or renders marine organisms in
ballast water.
A ballast water treatment system may be composed of a combination of several ballast water
treatment units. The standard may be applied to establish performance data for the combined unit,
as well as for individual units. The boundaries of the tested unit or combination of units, for
which test result will apply, must be clearly defined by the party responsible for testing.
The method is designed to provide performance data for a range of operating conditions for the
treatment unit, including the required performance data to demonstrate compliance/ non-
compliance with relevant IMO Regulations.
The testing methodology may be applied to ballast water treatment units of any size. The test
results however, apply only directly to the unit tested or to identical units. Care should be taken
upon extrapolating test results to differently scaled units. Any such scaling should follow
scientifically acknowledged scaling laws and should be carried out according to directions set
forth in the Appendix B accompanying the protocol.
In order to form basis for estimation of a full scale ballast water treatment system (after
extrapolation), the scaling difference should comply with some limits. At this stage in the process,
the following recommendations have been considered:
- Not to exceed 1: 1000 by volumetric capacity.
- Test arrangement must have a minimum water flow of 1 m3/ h.
Note that these restrictions are preliminary and under further assessment.
It is recommended that basic studies of ballast water cleaning principles are carried out at bench-
scale prior to testing according to the DNV Protocol. Recommendations for such studies are given
in the Appendix C - Guideline for small scale evaluations of ballast water treatment techniques,
also currently under development.
Generalised description of test set-up and methodology
Testing of ballast water treatment units according to the protocol involves three general steps:
1. Preparing simulated ballast water with predefined properties and content of marine
2. Treatment of simulated ballast water by a flow-through ballast water treatment unit,
operated according to its specifications and operating principles.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
3. Sampling, monitoring and analysis of treated ballast water for determination of treatment
performance as a function of test variables and time.
The standard is applicable to any test arrangement and facility that allows control and monitoring
of the test conditions and test parameters. Figure 3 shows two alternative general test
Test facility requirements
To perform testing, certain requirements to the facility undertaking the tests have been identified.
These are:
1. Uninterrupted access to the needed quantity of fresh and salt water with the desired
physical/chemical properties for the entire duration of the testing procedures.
2. Provide means for preparation of simulated ballast water with the desired uniformity in
composition and concentration of marine organisms.
3. Functional provisions for the treatment unit to be tested including pumping capacity and
piping/hosing arrangements appropriately dimensioned.
4. Tanks and containers with sufficient holding capacity for the test water (before pumping)
and for prolonged storage of water for monitoring after pumping
5. Provide a physical environment (temperature, light, air quality) that is not in conflict with
storage of live marine organisms (organisms must not die due to the physical
6. Provide (or have access to) laboratory facilities suitable for storage and analysis of water
samples for registration of presence of marine organisms
7. Provide all necessary means for correct operation of the ballast water treatment units
according to its specifications.
8. Provide all necessary instrumentation for control and monitoring of all operational
parameters during testing, (e.g. fluid flow rate, pressure, turbulence level etc).
If tests are undertaken statically, i.e. conditions do not include vibrations and motions, the
likeliness of this not affecting the performance of the system must be substantiated.
Testing according to this standard will establish quantitative performance data for the unit as a
function of the following parameters:
- Content and concentration of various marine organisms.
- Salt water/ fresh water.
- Content of additional organic material.
- Time for exposure and/or storage.
- Main operating parameters of ballast water treatment unit (such as concentration of active
substance, intensity of active physical process; see definitions).
Composition/ concentration of marine organisms
The Ballast Water Performance Standard, as identified in the current draft convention under
Regulation E-2, identifies absolute quantifiable levels of aquatic organisms in ballast water for
discharge distinguishing between zooplankton, phytoplankton and indicator microbes. Table 1
presents test conditions with reference to required concentrations.

Andersen: Ballast Water Treatment Verification Protocol - DNV
Table 1. Test organisms and concentrations.
Treatment performance
Organism class
Bacteria (I.E)
200 cfu/ ml
Bacteria (E.C)
500 cfu/ ml
200.000 cpl
1,000 cpl
500 cpl
25 cpl
* This will be changed in accordance with future regulation requirements
The selection of test organisms has been based on a criteria weighting model proposed by DNV,
see Table 2.
Table 2. DNV Criteria for selection of model group organisms.
The selected species must be easy to cultivate and handle.
Each organism should represent a typical ballast water organism, i.e. have one or
more pelagic life stages which will represent the test stadium.
The cultivation conditions for each species must be described in detail. The method
for assessing the test results (viability) for each species must be clearly and
unambiguously described. Selective detection methods should be available for each
The species of the model group must be robust compared to the majority of ballast
water organisms implying high tolerance to physical and chemical stress, i.e. salinity,
temperature, oxygen demand.
The species must be non-pathogenic (human and animal). Test should not lead to
risk of spreading pathogen organisms.
The species must be well described and specified with respect to species and strain.
One species per model group should be readily available for testing irrespective of
the geographical localisation of the test laboratory.
The species should preferably have a fairly worldwide distribution. The organisms
must be readily available from culture collection.
Testing shall establish performance data for treatment of water containing the classes of marine
organisms (model groups) defined in Table 3.
Table 3. Alternative model groups.
Model group
Representative species for testing/ certification
Lifestages in test set-up
Bacillus sp.
None at present
Dunaliella salinas
Vegetative stages
Diatoms (Skeletonema costatum,
Vegetative stages
Artemia salina
Red algae (Heterosiphonia japonica)
Simulated ballast water
The test procedure has to encounter for the large variations in ballast water characteristics. The
following variables have been included (se summary in Table 4):
Tests includes both fresh and salt water (typical for coastal surface
The temperature under which the tests shall take place must reflect
the natural temperature range of the organisms used and for coastal
O2 - saturation:
The level of dissolved oxygen shall reflect normal variations.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Must differentiate between fresh- and salt water.
Turbidity varies over a range depending on local conditions and
season and is dependent upon the level of suspended material. Tests
for both clear and dirty water shall be included.
Organic content:
Tests for clean and polluted (eutrophic) harbour water shall be
Table 4. Simulated ballast water test variables.
Salt water
Fresh water
> 32 PSU
< 0.5 PSU
02 saturation
Clear water, 2 FNU
Clear water, 2 FNU
Dirty water, 10 FNU
Dirty water, 10 FNU
Dissolved Organic
1(+/- 0.5) - 5 (+/- 1) mg C/ litre
1(+/- 0.5) - 5 (+/- 1) mg C/ litre
Test concentrations of organisms
For testing, the organisms in Table 3 shall be added to the water in concentrations specified (see
Table 1).
Methods for adding marine organisms
The introduction of marine organisms to the treatment process should apply one of the following
1. By use of a simulated ballast water holding tank (by stirring) from where the water is
pumped to the ballast water treatment unit:
2. By an introduction chamber at the pumping line downstream of the pump (by pressurised
injection to overcome the pump pressure).
3. In the sampling tanks after water has passed through the treatment unit.
Alternative 1 shall be the preferred injection point unless any conditions preclude this. Effects of
pumping and eventual other physical "treatment" that is not a part of the system must be
Alternative 2 may be applied if the mechanical action of the pump or any non-essential
component upstream of the ballast water treatment unit is considered to influence (increase) rate
of mortality of test organisms; and these components are not essential and/or integral parts of the
ballast water treatment principle.
Alternative 3 may be applied if the ballast water treatment method acts by generation or addition
of active substances, and it is evident (or proven) that the immediate effect in the ballast water
treatment unit is of negligible importance.
The uniformity of the simulated ballast water should be demonstrated by sample analyses.
Sampling and analysis
The following water samples should be collected for analysis of:
Water properties before and after treatment.
Content of organisms prior to treatment.
Content of surviving organisms at regular intervals during a minimum period of 2 days (under

Andersen: Ballast Water Treatment Verification Protocol - DNV
Time development of content of active substances in storage over a period one day to one
week depending on characteristics of the substance.
Analysis and methods
Analysis of water samples shall establish content of viable marine organisms as a function of time
following treatment. Analysis methods are currently being assessed, but should include re-growth
of bacteria and microalgae.
Performance parameters
Actual details regarding final performance requirements of ballast water discharge are not
available until the convention has been amended. Thus, parameters to be considered will decided
following the completion of the ongoing work within IMO.
Recommended testing procedures
Some testing procedures have been assessed. The protocol includes recommendations regarding:
· Preparations prior to testing.
· Testing period.
· After testing.
These are detailed into identified sequential tasks, the aim being to ensure that all tests are uniformly
undertaken and thus comparable and at the same time in accordance to protocol procedures.
Quality control
The protocol includes quality control procedures which specify minimum requirements with respect
· Number of samples
· Repetitions
· Test duration
· Test volumes
· General (ambient) conditions
This is vital in order to ensure reliability and to gain acceptance and recognition.
Measurements and reporting
This section of the protocol summarises measurements that must be reported and include:
· Water properties
- Including the parameters listed in table 4.
· Environmental parameters (ambient conditions; air temperature, humidity, barometric
pressure, etc.)
· Operating parameters
· Performance parameters
· Any calculated parameters

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Validity of certificate
A manual for maintenance, calibration and training should follow the treatment system and needs to
be assessed in the evaluation.
The validity of the certificate will be time restricted in compliance with requirements in the
Convention but will also reflect the characteristics of the system in question. Regular tests on board
shall be conducted to document that performance of equipment are in compliance with requirements.
Such control should include overboard discharge control and eventual measurements of
concentrations of active substance in ballast water after treatment.
DNV, 2003 Standardised Methodology for testing of Ballast Water Treatment Systems (Norwegian
Research Council).
DNV 2003. Ballast Water Treatment by Nitrogen Super-saturation, Technical report no.: 2003-0070.
IMO, 2003. Draft International Convention for the Control and Management of Ships` Ballast Water
and Sediments. MEPC 48
. Note: This is not an approved document and the text of the draft new
ballast water convention is still under development, subject to negotiation, and may change
MARTOB, 2003a. On board treatment of ballast water (technologies development and application)
and application of low-sulphur marine fuel.
Work package 2 completion report. Document Id: DTR-
MARTOB, 2003b. On board treatment of ballast water (technologies development and application)
and application of low-sulphur marine fuel
. Work package 3 completion report. Document Id: DTR-
DNV, 2002 Multi-oxidant Treatment of Ballast Water, Technical report no.: 2002-0656.
DNV 2002 National Regulations on Ballast Water Management (Norwegian Maritime Directorate).
DNV 2001. Ballast Water treatment by Ozonation ­ Biology, Technical report no.: 2001-0523.
DNV 2001. Ballast Water treatment by Ozonation ­ Corrosion, Technical report no.: 2001-0524.
DNV 2001. Alternative Ballast Water Treatment Principles and their potential in fulfilling future
IMO requirements
, Technical report no.: 2001-1062.
MEPC 47/INF.11. Harmful Aquatic Organisms in Ballast Water. Description of the proposed model
groups defined under Tier 1. Submitted by Norway.
MEPC 46/3/8. Harmful Aquatic Organisms in Ballast water. Proposal for performance criteria of the
ballast water treatment standard. Submitted by Norway.
MEPC 47/2/13. Harmful Aquatic Organisms in Ballast water. Proposal for performance criteria of the
ballast water treatment standard. Submitted by Norway.
IMO, 1997. IMO, 1997 - Guideline for the control and Management of Ships Ballast Water to
Minimise the transfer of Harmful Aquatic Organisms and Pathogen's
(IMO Assembly Resolution

Andersen: Ballast Water Treatment Verification Protocol - DNV
Figure 1. Steps in the conformity assessment of ballast water treatment systems.
Standard for Certification of Ballast Water Treatment Systems
Conformity Assessment Procedures

I. System performance

II. Operational and Maintenance
Manual and Documentation
A . Standards for Performance

III. Labelling
Testing of Ballast Water

IV. Maintenance and Repairs
Treatment Units
V . Constructional Strength and
B . Guidelines for Scaling of
Material Quality
Performance Test Data

VI. Power Requirements and
C . Guidelines for Small Scale
Studies of Ballast Water

VII. Hydraulic Power Units
treatment Methods

VIII. Pumps and Piping

IX. Safety
Figure 2. Standard for certification of ballast water treatment systems.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Alternative 1 - Loading system (one-pass system)
Alternative 2 - Recirculation system (multiple passes)
Figure 3. Generalised test arrangement.

The Artemia Testing System for ballast water treatment
M. Voigt
dr. voigt-consulting
This paper provides information on a new test system for the evaluation of the efficiency of ballast
water treatment options. The test is based on different life-stages of Artemia salina, an organism often
used in standard tests. The tests can be carried out at any flow rate and the results give a quick and
cost-efficient estimate of the efficiency of the proposed treatment option, avoiding costly and time
consuming multiple experiments and minimise the number of necessary field trials. Different
treatment options have already been tested at flow rates between 130 m3/h and 200 m3/h.

It has been demonstrated in numerous studies, that many organisms from different trophic levels can
be found in ballast water tanks, ranging from viruses to metazoans as well as algae and various cysts.
Any ballast water treatment option has to be able to remove or inactivate all of these different
The biological efficacy of any ballast water treatment option has to be assed with flow rates
representative for ballast water operations The current practice is, to carry out numerous tests, either
land-based or onboard of a ship, with the species present in the water at the testing site, or to prime
the system with individual test species. All of these approaches are very time consuming, expensive
and difficult to standardise. Changes in the species composition at the test site and in the densities of
individual species have a negative impact on the statistical analysis of the experimental data.
When surrogate species are used, the test system can be primed with a given number / density of the
surrogate organisms and the observed changes in numbers / survival rates are mainly attributed to the
A new full-scale test has been developed to account for most of the trophic levels and the different
physical properties of the organisms frequently found in ships' ballast water. The Artemia Testing
System (ATS) involves different larval and development stages of the brine shrimp (Artemia salina)
as surrogates for a variety of organisms commonly found in ballast water (Table 1). The robust
Artemia can be produced in any lab with only little effort. Furthermore, they can be easily added to
the water prior to the treatment system and are easy to recognise / identify even in samples with high
numbers of other taxa and / or high turbidity.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Table 1. Development stages of Artemia salina used in the ATS full-scale tests.
Artemia development stage
Trophic level
Surrogate for
Resting stage
inactive cysts
Floating (pelagic cysts) <100 µm
Soaked cysts
inactive cysts
Demersal (benthic) cysts > 100 µm
Developing eggs
floating / demersal eggs
Larval organisms (plankton) 150 µm
­ 180 µm
larvae (not feeding)
Numerous planktonic organisms >
250 µm
The different physical properties (specific weight, size) and the different behaviour (passive
movement with currents and active swimming) make the above development stages ideal surrogates.
Furthermore, they show rather low sensitivities to physical and chemical stressors, which makes them
a good "worst-case-scenario" for any combination of treatment options as well as for stand-alone
Because of the rapid development of the cysts and larvae (nauplii), the test results can be obtained 24
hours after the experiment.
So far, the ATS test protocol (see Annex) has been downloaded nearly 400 times from users all over
the world (Tab. 2).
Table 2. Country-codes of servers that have downloaded the PDF-file of the ATS full-scale test protocol.
Over seas
New Zealand
Republic of Congo
Furthermore, the ATS test protocol has already been applied in several land-based tests with different
treatment options.
The ATS is a useful tool for the assessment of the biological efficacy of ballast water treatment
options. It can be used as a model for a wide range of organisms with
· different specific gravities
· different sizes and shapes
· different behaviour
· different sensitivities to stress.
The ATS can be used in any location and at any time, independent from seasonal fluctuations of in-
situ plankton organisms. The full ATS test protocol is available free of charge in PDF-format in the
internet (see Annex).
The ATS poses a low environmental risk from the surrogate species. Only little training is required for
the personnel that analysis the samples. High taxonomic skills, as they are essential in most tests

Voigt: The Artemia Testing System for ballast water treatment options
which use in-situ species composition, are not required. As an other advantage, the ATS can be
calibrated against more sensitive species, and the results are highly reproducible.
However, any evaluation of the biological efficacy of a ballast water treatment option should not be
based only on the results of the ATS. It should be applied in combination with at least one more
surrogate organisms that accounts for small (< 50 µm) zooplankton and / or phytoplankton. It has also
to be noted, that the ATS should only be applied in test waters that show physical properties within
the tolerance of Artemia (e.g. salinity > 12 ppt and water temperature above 15°C, and max. 28°C).

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The ATS© full-scale test protocol

© dr.voigt-consulting, Germany, e-mail: m.voigt@drvoigt-consulting.de
Preparation of experiments
The numbers of individuals needed for the tests depend on the capacity of the treatment system. As a
role of thumb, one each of the following cultures is needed for every 30 m3/hour capacity.
Breeding of nauplii
1. Fill a 1-l-bottle with 600 ml of filtered sea water and aerate well.
2. Transfer 1 table spoon of premium grade Artemia eggs to the bottle. Incubate in a water bath
at 24°C for 24 hours.
3. Decant the hatched nauplii through a sieve (mash 10 µm) and transfer to 10 l to 20 l aquarium
filled with sea water (24°C).
Breeding of developing eggs
1. Fill a 1-l-bottle with 600 ml of filtered sea water and aerate well.
2. Transfer 1 table spoon of premium grade Artemia eggs to the bottle. Incubate in a water bath
at 24°C for 12 hours.
3. Take a sample of the eggs and examine under a stereo microscope at 20 x magnification. If
the outer shell of the eggs has opened, the yellowish embryo is clearly visible and the eggs
can be used for the experiments. If the embryo is not clearly visible, incubate the eggs for 4 to
6 more hours. Monitor the development closely
Preparation of soaked cysts
1. Fill a 1-l-bottle with 600 ml of filtered sea water.
2. Transfer 2 to 3 table spoons of premium grade Artemia eggs to the bottle. Allow the cysts to
soak for 2 hours at room temperature
Preparation of resting stages
1. Fill a 1-l-bottle with 600 ml of filtered sea water.
2. Transfer 2 to 3 table spoons of premium grade Artemia eggs to the bottle directly prior the
beginning of the tests
Preparation of the treatment system
1. Install a by-pass to the first pump of the treatment system in order to prime the system with
the cultures of Artemia development stages.
2. Identify the capacity (flow rate) of the by-pass and calculate the passage time of the water
through the system.
3. Adjust the flow rate of the by-pass to allow min. 5 minutes of test run.
4. Start the treatment system and allow to stabilize for at least 1 hour.
IMPORTANT: re-direct the water flow into tanks with sufficient capacity during the test
run to avoid introduction of Artemia to the test site


Voigt: The Artemia Testing System for ballast water treatment options
The ATS test procedure
1. Mix the cultures (resting stages, soaked cysts, developing eggs and nauplii) in a bucket or barrel
(10 litre volume for every 30 m3 of capacity of the treatment system). Top up with sea water and
aerate well.
2. Transfer a sample of 1 litre to a 200 l barrel (control), top up with sea water and aerate.
3. Prime the system with the prepared cultures through the by-pass of the pump.
4. During the passage of the organisms through the treatment system, take samples of 200l each
directly before and after each treatment step (e.g. filtration / separation, disinfection).
5. Mix the water in the 200 l barrels well and take sub-samples (three replicates) of 10 litres each .
6. Put the sub-sample through a sieve (10 µm) and observe under a stereo microscope at
magnification of 10 x.
7. Count the numbers for each of the development stages. Record numbers of damaged or dead
individuals separately.
8. Observation of test organisms directly after the test run:
a. The movements of the antenna and legs of the Artemia nauplii are monitored under a stereo
microscope at 10x magnification. The individual is dead, if no movements of the antenna can
be detected.
b. The resting stages, the soaked cysts and the developing eggs are examined for mechanical
damage under a stereo microscope at 10 x magnification.
9. Cover the barrels and leave without aeration for 24 hours.
10. Repeat steps 6 to 9.
11. Calculate the mortality / removal in percent for the nauplii for each step of the treatment.
12. Calculate the removal /damage rate in percent for the resting stages, soaked casts and developing
If the numbers of developing eggs increases in all three replicates taken after 24 hours in comparison
to the samples taken directly after the test run, the treatment was insufficient for the soaked cysts.
If the numbers of the alive nauplii increases in all three replicates taken after 24 hours in comparison
to the samples taken directly after the test run, the treatment was insufficient for the developing eggs.

Development of dinoflagellate "cyst-on-demand"
protocol, and comparison of particle monitoring
techniques for ballast water treatment evaluation
J. T. Matheickal 1, Tay Joo Hwa 1, Chan Mya Tun 1, S. Mylvaganam 1, M. Holmes 2 & L. Loke 1
1Institute of Environmental Science and Engineering,
2Tropical Marine Science Institute, Singapore
There is an urgent need for development of standardized testing and analytical protocols for
evaluation of various ballast water treatment technologies. This paper presents the initial results from
two separate studies on 1) using particle counts for evaluation of ballast water filtration techniques
and 2) development of a "cyst-on-demand" protocol for mass-culturing dinoflagellate cysts. Particle
counts, size and distribution analysis can provide reasonably sensitive measurements of particulates
in water and as such can be a valuable analytical tool for the evaluation of treatment technologies
that "remove" organisms from ballast water. The use of particle counts based on the particle size
assumes importance from ballast water quality standards point of view, as there is an increasing
consensus on the use of "size" as the basis for ballast water treatment performance standards. In the
past, a number of methods have been used to count and size particles in ballast water ­ each method
uses a different counting technique. There is no standard method for performing particle size analysis
for ballast water, and each method has its own limitations. One key issue is the reported lack of
consistency of such measurements by different instruments in the market as the basic measurement
principles vary. Studies were undertaken in our laboratory to test whether different techniques yield
the same particle count and size distribution information when applied to ballast water quality
monitoring. In this study, two particle counters, based on light obscuration and electrical sensing as
the measurement principles are compared. There are many factors that would affect the reliability of
such measurements and these include, particle concentrations, size ranges, storage etc. The second
part of this paper discusses the initial results of our attempts to mass culture dinoflagellate cysts that
could be used as biological surrogates for the evaluation of various secondary treatment
technologies. Most of the secondary treatment technologies are based on the inactivation or kill of the
organisms that are not removed by a primary treatment process such as filtration. Since using toxic
dinoflagellates cysts as a surrogate is difficult due to practical reasons, our research focused on the
non-toxic, cyst forming dinoflagellate specie,
Scrippsiella sp., as a surrogate. We have observed that
sufficiently large number of dinoflagellate cysts can be produced "on-demand" for lab-scale
evaluation of treatment technologies.

Name of research programme
Singapore Ballast Water Research Programme (SBWRP)
Treatment options being considered
Filtration, Hydrocyclone, UV, Biocides and Photocatalysis

Matheickal: Development of dinoflagellate "cyst-on-demand" protocol
Timeframe of the programme
Overall aims and objectives of the Singapore Programme
· To develop a set of strategies for the control of transfer of non-indigenous species via ballast
water for Singapore shipping and port interests.
· To demonstrate ballast water management schemes at suitable scale in order to generate
treatment effectiveness, and reliability data, as well as life cycle costs.
· To participate in developing appropriate performance verification protocols for the evaluation
and approval of treatment technologies.
· To act as a regional center for coordinating research and development on ballast water
As noted above, one of the major objectives of this programme is to develop appropriate verification
protocols for ballast water treatment systems. This paper, therefore, discusses the initial results from
one such study carried out to evaluate particle counting instruments for ballast water quality
monitoring. The paper also discusses the development of a "cyst-on-demand" protocol for mass
culturing a surrogate organism for ballast water treatment verification.
The main objective of the study reported in this paper, were:
· to evaluate the use of particle counting as a measure of filtration efficiency and to monitor
filter failure
· to compare two different particle counting techniques for monitoring seawater quality
· to develop and optimize a culturing protocol for mass-culturing dinoflagellate cysts
Background information on the study
The ongoing research in Singapore, USA, Europe and elsewhere showed the potential for alternate
treatment technologies for management of ballast water. However, it has been recognized widely that
performance comparison of these technologies has been severely restricted by the lack of
standardization in this area. Although, there are a number of on-going pilot-scale and ship-board trials
of some of these technologies, the protocols used for evaluation of these systems vary and a
reasonable comparison of these test results are, unfortunately, difficult. There is, thus, a need to
develop internationally acceptable methodologies for the harmonization of performance testing of
"removal" based technologies. One of the pre-requisites to development of these protocols is the
identification of suitable water quality parameters or their surrogates and development of reliable
measurement tools and protocols to monitor these parameters/ surrogates.
There is an increasing consensus that primary as well as secondary treatment technologies may be
required to meet the long-term ballast water performance standards. Broadly, these treatment
technologies can be classified as "removal" based technologies and "inactivation/kill" based
technologies. Many primary treatment technologies such as filtration, hydrocyclones etc., belong to
the former category and the secondary treatment will most likely belong to the later one. The
parameters chosen to evaluate technologies therefore should take into account the removal,
inactivation and kill effectiveness of treatment technologies. Moreover, the parameters chosen should
satisfy the requirements of easy and rapid monitoring needs.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
"Size" as a non-biological surrogate measurement
Currently, the scientific basis for establishment of ballast water treatment standards are going through
an intense debate both among the scientists and at the IMO. Although there is a general consensus on
the primary criteria for acceptance of treatment technologies (safety, environmental friendliness,
practicality, cost effectiveness and biological effectiveness), there are two differing views on
measuring biological effectiveness of treatment systems 1) a performance standard based on%
removal of representative species 2) a water quality standard based on organism size. The former one
is based on a perceived risk reduction through % organism reduction and the later one bases its
recommendation on the history of past invasions and attempt to identify and group the past invaders
as per size.
While it is not attempted here to discuss the pros and cons of these options, it should be noted that the
selection of a standard would have significant influence on the way treatment technologies are
evaluated and performance assessed. From the previous discussions and looking at the latest draft text
of IMO regulations (MEPC-49), it can be clearly seen that "organism size" would play a major role as
a surrogate water quality parameter in the long-term as well as short-term definition of ballast water
standards. In fact, recent discussions show signs of some convergence between the two schools of
thoughts. It is argued that a size based standard can still result in a meaningful reduction of risk, while
at the same time "size", as a surrogate measurement, lends itself for easy monitoring. This is
supported by the recent literature study by Waite (2002) who suggested a 100 micron size cut-off as a
preliminary standard, as this would eliminate most, if not all, of the marine organisms with some past
invasion history.
The idea that the material to be removed or inactivated by a shipboard treatment process is larger than
a certain size, is an important concept. Moreover, a size based standard can be used to evaluate
several treatment technologies, including screening, filtration, and cyclonic separation (Waite, 2002)
and if coupled with viability measurement, it can be used for technologies based on
killing/inactivation (biocides, heat etc) as well.
Particle size is also an effective surrogate measurement for measuring biological effectiveness of
mechanical separation systems (e.g., filtration, hydrocyclone) used in ballast water treatment.
Researchers have used particle sizing, counting and size distribution analysis (Parsons and Harkins,
2002) for evaluating performance of screen filters, hydrocyclone and depth filters for ballast water
Particle sizing and counting for filtration evaluation
In the past, a number of methods have been used to count and size particles in seawater ­ each method
uses a different technology, since there is no standard method for performing particle size analysis for
ballast water, and each method has its own limitations. One key issue is the reported lack of
consistency of counts and sizing by different instruments in the market as the basic measurement
principles vary. Also, there are many other factors that would affect the reliability of such
measurements and these include, sample concentration procedures, storage, and even different types
of concentration standards used for calibration of the instrument. Standardization of particle
count/size based measurements and interpretation of data obtained from particle counting are
therefore important components of the development of standard protocols for treatment verification.
Dinoflagellate cysts as biological surrogates
Physical separation technologies, including filtration or hydrocyclone methods, may play an important
role in the primary treatment for the removal of the larger organisms. For example, a screen filter with
mesh size of 50­100 microns could remove the larger planktons. But smaller organisms such as
dinoflagellates and plankton, with sizes of 10­30 microns are problematic to remove by physical
separation. These smaller organisms may be removed using 20­25 micron filters but such filters are
less efficient from an operations point of view (Cangelosi et al., 2001). Although difficult to remove

Matheickal: Development of dinoflagellate "cyst-on-demand" protocol
by physical separation, secondary treatment technologies such as biocides, thermal techniques,
electric pulse and pulse plasma techniques, ultraviolet treatment, acoustics systems, magnetic fields,
deoxygenation, etc are being considered by many research groups, for inactivating or killing the
smaller organisms.
R&D on secondary treatment technologies would benefit considerably if a suitable biological
surrogate can be selected for evaluation of various treatment regimes as well as for technology
verification purposes. Such a biological surrogate should be representative of target organisms that
can be invasive on a global level, should lend itself for mass culturing, should be hardy enough to
ensure treatment efficacy and should be non-toxic so that it can be used in pilot-scale and lab-scale
studies. One possible class of biological surrogates is the cysts of microalgae, dinoflagellates.
Toxic dinoflagellates have been identified as a major invasive problem worldwide, especially since it
can survive long voyages. Blooms of the toxin producing dinoflagellate Gymnodinium catenatum
were first recorded in Tasmanian waters in late 1985 and may have been introduced by ship ballast
water (Hallegraeff et al., 1989). A new toxin producing, benthic dinoflagellate has been isolated from
the fringing coral reefs surrounding the Singapore island of Pulau Hantu (Holmes, 1998).
Dinoflagellate species are globally distributed and many of them are harmful. Toxic blooms have
been reported from many countries (Gollach, 1999).
Dinoflagellate vegetative (motile) cells and their cysts often measured between 20 and 40 microns
(Anderson et al., 1985). Dinoflagellate cysts (hypnozygotes) are often thick walled, highly resistant,
non-motile stages that are formed from sexual re-combination. The cysts can often survive in harsh
environmental conditions and may be resistant to mild disinfection technologies.
In summary, the following rationale is given to select Dinoflagellate Cysts as a surrogate organism to
evaluate the efficiency of the ballast water treatment:
· Dinoflagellate cells and cysts are smaller in size and may escape primary treatment.
· The roles of cysts need more attention as they have thick and special cell walls that are
resistant to mild-disinfection that are normally used in other disinfection techniques
· The spread and damage to environment caused by dinoflagellates is of international concern
· The vegetative stage is easy to culture and calcareous cysts of certain species are easy to
produce (although not necessary representative of the cysts of harmful species).
· The outcome of minimum test conditions of removal/kill/inactivate technologies can be
applied and studied to other organisms including bacteria.
Research methods and protocols
Particle counting for filter evaluation
In order to evaluate the use of particle counting for filtration efficiency, a 30 micron filter was
challenged with Arizona Fine Dust (ISO 12103-1 A4) particles suspended in pre-filtered (0.45
micron) water. Arizona Dust was selected due to its wide range of particle size distribution (1-80
microns), its non-coagulating nature and also due to the non-spherical shapes of the particles. A
Coulter particle counter (Multisizer III) was used to count the particles before and after filtration. In
order to check the usefulness of particle counting for monitoring filter failure, the filter was later on
damaged intentionally, by making a pinhole. Filtration experiments were also carried out using actual
seawater, containing organisms. Scanning Electron Microscope (SEM) analysis of the particle that
passed through the filter was also carried out to study the filtration efficiency. This was performed by
filtering a measured aliquot of water sample through a 0.45 micron membrane filter and subsequently
air-drying the filter in a laminar hood before SEM analysis.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Comparison of particle counters for ballast water quality analysis
Two different particle counters were compared in this study; 1) light-obscuration based particle
counter, that uses both on-line sampling as well as grab-samples. This instrument was size calibrated
with latex spheres and set for a flow rate of 100 mL/min by the manufacturer. Software supplied with
the instruments allowed the operator to choose the size classes 2) electrical sensing zone (ESZ)
instrument using coulter principle which was set to count the particles over a specific period of time at
a set flow rate. Three different water samples were used to generate particle counting data 1) actual
seawater collected from Sembawang Site in Singapore 2) Arizona Dust suspended in filtered seawater
and 3) Dinoflagellate cultures suspended in filtered seawater. Same concentration of particles was
used when the instruments were compared.
Dinoflagellate "cysts-on-demand" protocol development
A temperature-controlled room with lighted growth tables and height-adjustable table platforms were
used to achieve the optimum light intensity required for growth of Scrippsiella. A temperature-light
cutoff set at 30°C was used to switch off all the culture lights if the temperature in the room
significantly exceeded the set temperature of 26°C. A timer controlled all lights on a 12 ­ 12 hour
light: dark cycle photo-period.
Three strains of the known cyst producing and non-toxic dinoflagellates Scrippsiella sp. were
purchased from USA from Provasoli-Guillard National Center for Culture of Marine Phytoplankton
(CCMP 1331 and CCMP 1735) and the University of Texas at Austin culture collection of algae
(LB1017). Only CCMP1735 could be induced to routinely produce hypnozygotes. We subsequently
found that growth of this strain was faster and the hypnozygotes production better at 26°C than the
recommended culture temperature of 20°C. The culture room facilities were adjusted accordingly and
the two other strains were discontinued since they could not survive the higher temperature. The
counting of cells and cysts was made manually using a Sedgewick- Rafter cell counter and
The media, f10, were prepared by adding the necessary nutrients to the sterilized seawater. The
seawater was 0.2 micron filter-sterilized by the pressure vessel system. Initially, 5ml of soil extract
was added per 1 L of media based on the nutrient formulations used by the originating USA culture
collection. However, this was discontinued after few months because of frequent fungal
contamination. Stock cultures were prepared and transferred to new media every two weeks. All the
transfers of cultures were carried out in laminar flow hood to avoid any contamination.
To design and conduct the ballast water treatment experiment efficiently, the growth rate of the
Scrippsiella sp. was determined. During this study, the morphology of the cells was also studied by
using an inverted Olympus IX 70 microscope. The culture was sampled and diluted at known time
interval and manually counted using Olympus IX 70 microscope and Sedgewick-Rafter Cell counter
to determine the growth rate. Hypnozygotes are the sexual (diploid) stage in the life cycle of
dinoflagellates. The strain of Scrippsiella produces hypnozygotes in culture, however, manipulating
nutrients conditions can considerably increase the number of cysts. The following protocol was used
to determine the patterns of formation of cyst (hypnozygotes) in 250 ml flasks. A 10 ml portion of
culture samples were aseptically transferred into a 250 ml flask containing nutrients depleted
sterilized seawater as a medium for the formation of cyst. These samples were regularly monitored
and counted for cysts after dilution with sterilized seawater. All the flasks were kept in the culture
condition of 12:12 hours light: dark photoperiod at 26°C.
Rate of excystment was determined by transferring a 10 ml of culture rich in cyst back to the f10
media. The inoculated samples were kept in the culture condition of 12:12 hours light:dark
photoperiod at 26°C and were monitored and counted daily for the formation of cells.

Matheickal: Development of dinoflagellate "cyst-on-demand" protocol
Results and discussion
Particle counting for evaluation of filtration efficiency
It has been well documented that particle counting provides greater sensitivity and increases lead-time
in surface water filter performance optimization (Hargesheimer and Lewis, 1995, Bellainy et al.,
1993). Particle counters both count and size individual particles as they flow through a sensing zone.
They operate by electrical resistance (Coulter), light blockage or light scatter principles. Other
instruments, called particle size analyzers, provide information on particle size and particle size
distribution, but they do not count. Some will provide an indication of particle numbers by
mathematical estimation.
In order to determine if particle counting could be used as a reliable tool for monitoring filtration
efficiency, a 30-micron filter was challenged with Arizona Dust Particles suspended in water. Filter
integrity monitoring using particle counter was also studied by using a compromised screen. Figure 1
shows the results from these studies. Particle monitoring data are expressed in differential numbers,
differential volume and % differential volume. It can be seen that the particle monitoring gave an
accurate description of the filtering efficiency for various size ranges of the particles. The filtered
water samples showed significantly less number of particles and the same was better expressed when
the data was presented in volume fractions.
When a compromised screen was challenged with the test samples, the particle counter responded
well and showed a significant increase in number of particles in the filtered samples. Again, when
expressed in volume fractions, the particle counter was able to show the sudden peak corresponding to
the larger particles that went through the pinhole in the screen. The instrument was sensitive enough
to count the small number of particles that went through the compromised screen.
Particle counting was also used to evaluate filtration of seawater using a 30-micron screen. The
number of particles in seawater was considerably lower than the Arizona Dust Samples. As shown in
Figure 2, Particle counting again proved to be a useful tool for evaluating filter performance. It can
however be noted from the particle data that the test filter allowed some particles above 30 microns to
pass through. Although this was initially suspected to be the result of a compromised screen, detailed
SEM studies showed that these larger size particles were in fact "needle-shaped" organisms with their
shortest dimensions being considerably smaller than 30 microns. Figure 3 shows the SEM image of
the organisms that had passed through the screen and it can be noted that the diatom skeletonema, one
of the common diatoms in Singapore waters, had passed through the 30-micron screen. As the coulter
particle counter used in this study measures the volume of particles and subsequently calculate the
equivalent spherical diameter, such needle shaped organisms would be shown as larger size particles.
These observations are similar to the one reported by Waite et al. (2003).
Comparison of particle counting techniques
Electrical Sensing Zone Technique: This technique was pioneered by the Coulter (USA) Company
many years ago for blood cell counts in hospitals, where it is still widely used. Particles are suspended
in an electrically conductive fluid (usually saline water with emulsifier) and forced to flow through a
small orifice. Conductors are placed in the fluid on either side of the orifice, and the electrical
resistivity of the orifice is monitored as particles pass. Each particle produces a sharp "spike" in
electrical resistivity as it passes the orifice, and the total area (time x height) under the spike is
approximately proportional to the volume of the particle. Each of the spikes is classified according to
total area, and a particle count is place in a bin that corresponds to the appropriate particle size. After
many thousands of particles have passed the orifice, the bin counts are converted to a particle size
distribution, and the distribution is finally adjusted to account for the statistically finite probability of
"co-incident" counts.
This technique is suitable for a relatively broad range of sizes (0.5 micron to >300 microns, using
different orifice sizes). However, the dynamic size range is limited to about 30 in a single run (from

2nd International Ballast Water Treatment R&D Symposium: Proceedings
about 2% of the orifice size to about 60% of the orifice size). Analysis of broader distributions
requires pre-separation of samples according to size (for example, via sieves), so that the individual
fractions can be run using different orifices.
Light Obscuration Technique: In this method, a laser beam transmits its light through a flow cell to a
photo detector and when there is an absence of particles, the light transmitted is received by the
detector as an equivalent voltage level corresponding to full voltage intensity. As particles interrupt
the laser beam, a shadow equivalent to the particle's size generates a voltage drop that the counting
electronics convert into size and count information (Figure 4).
We compared the ESZ zone technique and light obscuration technique by subjecting different
seawater samples to two different instruments that use these techniques (Coulter Multisizer III and
LaserTech). Our initial experiments included the use of a light scattering type particle counter
(Malvern), however its use was discontinued as it was observed that the particle concentration in
typical seawater was too low for such measurements and the noise levels were high, masking the
actual measurements. Particles and light interact strongly, with lights scattered from refraction,
reflection, and diffraction by the particle. Complex equations apply to quantify the change in
intensity, and one set of parameters does not apply for all size of particles (Van Gelder et al., 1999).
Varying indexes of refraction among the inorganic, organic and biological particles in seawater
further complicates the light-scattering response.
The two instruments (based on ESZ and Light Obscuration) were compared by analyzing the same
samples using the instruments. Throughout the experiments, large differences in the instrument's
counts were seen. Figure 5 shows the results of laboratory counts of three types of samples 1)
seawater samples 2) Arizona Dust samples and 3) seawater containing a mixture of dinoflagellates
and other organisms. The difference in counts between light obscuration instrument and the ESZ
instrument was dramatic and typical of what was observed throughout the research. The light
obscuration instrument consistently undercounted particles compared with the ESZ instrument in the
smaller size classes and consistently over counted the larger particles.
The variations in counts for the samples can mainly be attributed to the coincidence errors that can
happen in light obscuration instruments. This happens when more than one particle passes through the
sensor at a time as shown in Figure 6. In the Arizona Fine Dust samples, there is a significant
proportion of particles in the less than 5-micron size category. It is possible that several particles were
moving through the sensing orifice simultaneously and resulting in several particles detected and
counted as one larger particle. The presence of over-concentration errors can be confirmed by
reviewing the particle size distribution results shown in Figure 5.
From the data shown, it can also be seen that the differences in counts were also dependant on the
samples tested. The difference in counts was maximum (50 times) in the case of 2-10 micron class
category of Arizona Dust, possibly due to the largest concentration of particles in these samples.
Although the total number of particles in the 2-10 micron range was very similar in the case of
seawater samples and samples containing predominantly dinoflagellates, the differences in counts
observed was higher (15 times) in the case of dino samples as compared to seawater samples (8
times). This is possibly due to the large difference in refractive index of the calibration particles (latex
spheres) used in calibration of the light obscuration instrument and the dinoflagellates. It also shows
that appropriate calibration standards need to be selected if light obscuration based instruments are
selected for ballast water monitoring. Research in the past had shown that light obscuration
instruments have difficulty in counting particles between 2 and 3 microns and this also could have
added to the dramatically lesser counts obtained in the light obscuration instrument.
Dynamic size range in light obscuration technique is limited to about 100-200 for a single run.
Analysis of broader distributions requires measurement using two different size sensors. Resolution
appears to suffer with smaller particles. Non-spherical particles reduce resolution, because the cross
section of the particle is evaluated rather than it's volume. The cross-section for a given particle will

Matheickal: Development of dinoflagellate "cyst-on-demand" protocol
depend on both particle shape and orientation as it passes through the detector. A second detector
beam perpendicular to the first would allow a better measurement of volume, but to this author's
knowledge, only single beam instruments are produced.
It can also be seen that ESZ instrument recorded the events when a large size particle (80 ­ 100
microns) passed through the aperture, where as the light obscuration instrument did not capture this.
This again shows the size limitation of light obscuration instruments as many of them are optimized
for the 4-6 micron size ranges. Since the size information is entirely dependent upon the size of the
voltage drop, there is a restriction as to how big a particle can be sized since the biggest voltage drop
that can be sized is down to zero volts (Figure 7). Particles of around 80-100 microns are usually the
highest that can be sized and perhaps represent the extreme range for the instrument used in this
study. Bigger particles can be detected, but the optics and electronics have no way of knowing how
much bigger than 80-100 microns they are.
Cyst-on-demand protocol
Hypnozygotes are the sexual (diploid) stage in the life cycle of dinoflagellates. The CCMP1735 strain
spontaneously produces hypnozygotes in culture, however, manipulating nutrient conditions can
considerably increase the number of cysts. Five nutrient treatments for increasing the production of
hypnozygotes from the CCMP1735 strain of Scrippsiella sp. were trailed. We developed 2 nutrient
protocols (f2 to f10 and f2 to filtered water) to reliably induce hypnozygote-production in 2 litre
culture flasks. These protocols rely upon transferring a large biomass of vegetative cells into nutrient-
deficient media. Details of this protocol will be published elsewhere.
The CCMP1735 hypnozygotes have an oblong to spherical shape with many, but all, producing
calcareous spines (Figure 8). Newly formed cysts are translucent with mature cysts developing a red
accumulation body(s). However, the dimensions of newly formed and mature hypnozygotes (with or
without spines) are not significantly different (Table 1 and 2, P >0.05).
Table 1. Longest length (µm) of newly formed hypnozygotes (NFH) and mature hypnozygotes (MH) with (+) and
without (-) spines.
NFH + spines
MH + spines
NFH - spines
MH - spines
Standard deviation
Sample size
Table 2. Shortest length (µm) of newly formed hypnozygotes (NFH) and mature hypnozygotes (MH) with (+) and
without (-) spines.
NFH + spines
MH + spines
NFH - spines
MH - spines
Standard deviation
Sample size
Development of methods for excystment of hypnozygotes: The excystment of CCMP1735
hypnozygotes were characterized. In 250 ml flask cultures, mature hypnozygotes form after 9-20 days
and begin spontaneously excysting after 2-28 days. It was found that mature hypnozygotes could be

2nd International Ballast Water Treatment R&D Symposium: Proceedings
stored for more than 1 month in a quiescent state in the dark at 5 to 7°C. Hypnozygotes transferred
from cold storage to 24 to 27°C excyst in a time-dependent manner (Fig. 1).
The time to excystment (days) = 0.2 x (days storage at 5 to 7°C) + 0.9 [P < 0.001]
The pattern of encystment and excystment was observed and could be used to store mature
hypnozygotes for more than one month and predict time-to-excystment to about ± 1 day.
Although mature hypnozygotes could be held in cold storage for longer than one month, the
proportion of viable cysts reduces the longer the storage time. There were also indications that the
linear relationship between time of storage and time to excystment may not hold for more than 2
months cold storage.
Cyst formation was minimal for the first three days but increased considerably from the 7th day of the
inoculation (Figure 9). Excystment of cyst was confirmed by the presence of a motile cell and an
empty cyst wall, with or without archeopyle (characteristic excystment pore). Cysts were not excysted
after one month of regular weekly observation nor after two months. This indicates that accumulation
bodies are necessary for excystment and could be a source of energy for the excystment process.
However, when these conditions were met, we observed the excystment pattern as shown in Figure 9.
Conclusions and recommendations
Ballast water standard based on organism size can provide an ideal basis for defining ballast water
quality and treatment technology evaluation. Particle counting and sizing is an extremely useful tool
for ballast water treatment monitoring and verification. Participle counters, if properly used, can
monitor the treatment performance on a continuous basis and offers a sensitive tool for monitoring
events such as filter failure. Nevertheless, particle counters themselves and the ability to check their
sizing and counting accuracy need improvements. During the research, the variation in counts taken
by two different instruments when the same samples of various types was measured was documented.
Dramatic variations in counts were present between electrical sensing zone based particle counters
and the commonly used light obscuration based counters. The later one dramatically undercounted
particles in smaller size classes compared with the research grade ESZ instruments for all types of
samples, and there is ample evidence to suggest that the ESZ instrument was most correct. However,
light obscuration particle counters can give a cheap and practical solution for online monitoring of
ballast water, provided the instrument is calibrated using appropriate calibration standards, right
concentration of particles used and correct flow rate is chosen. It may perhaps be required to improve
the methods the manufacturers use to set the lowest millivolt calibration value. It is strongly
recommended that ballast water monitoring be conducted using an electrical sensing zone based
particle counting instrument for any verification purposes.
The second part of the study developed culture protocols for producing hypnozygotes (cysts) of the
CCMP1735 strain of dinoflagellate Scrippsiella Sp. on demand. It was observed that transferring a
large biomass of motile cells to nutrient deficient media induces cyst formation. Once the
hypnozygotes mature they begin spontaneously excysting after about 2 days. However, hypnozygotes
can be stored in a quiescent state for up to 2 months in the dark at 5 to 7°C, although the proportion of
viable cells drops after about 1 month storage. The time to excystment of cold-stored hypnozygotes
can be predicted from the time of cold storage. Dinoflagellate, being an invasive species of
international concern, can be an ideal surrogate organism for treatment system evaluation. The
protocol developed in this study can be used to produce sufficiently large number of dinoflagellate
cysts. Detailed protocols for the use of these cysts for evaluating various treatment options are
currently under development in our laboratory.

Matheickal: Development of dinoflagellate "cyst-on-demand" protocol
The study was supported by the Agency for Science, Technology and Research (A*STAR) and the
Maritime and Port Authority of Singapore (MPA).
Anderson, D.M., Lively, J.T, Reardon, E.M. & Price C.A. 1985. Sinking Characteristics of
Dinoflagellate Cysts. Lymnol.Ocenogr. 30: pp. 1000-1009.
Bellaķny, W., Cleasby, J.L., Logsdon, G.S. & Allen,M.J. 1993. Assessing Treatment Plant
Performance. Jour. AWWA, 85(12): pp. 34-38.
Gollach, S., Minchin, D., Rosenthal H. & Voigt M. (eds.) 1999. Exotic Across The Ocean, Case
Histories on Introduced Species
Hallegraeff, G.M., Stanley, S.Q., Bolch, C.J. & Blackburn, S.I. 1989. Gymnodinium catenatum
blooms and shellfish toxicity in southern Tasmania, Australia. In: Okaichi T, Anderson,D.M,
Nemoto,T., (eds) Red Tides: biology, environmental science toxicology. Elsvier, Amsterdam, pp. 75-
Hargesheimer, E.E. & Lewis, C.M. 1995. A Practical Guide to On-Line Particle Counting. Denver,
Colo.: American Water Works Association Research Foundation.
Holmes, M.J. 1998. Gambierdiscus Yasumotoi Sp.Nov (Dinophyceae), A Toxic Benthic
Dinoflagellate From South Eastern Asia, Journal of Phycology. pp. 661-667.
Parsons M.G. & Harkins, R.W. 2002. Full-scale Particle Removal Performance of Three Types of
Mechanical Separation Devices for the Primary Treatment of Ballast Water, Marine Technology,
39(4): pp. 211-222.
Van Gelder, A.M, Chowdhury, Z.K & Lawler, D.F. (1999) Conscientious Particle Counting, Jour.
, 91(12): pp. 64-76.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 1. Particle data for raw and filtered Arizona Dust samples. (A. ISO Dust solution without filtration B. ISO
Dust solution filtered with 30um filter C. ISO Dust solution filtered with a failed filter (30um with a pinhole)).
sea wate
Volume % (Differential) 1
85 um 95

Figure 2. Particle counts data for raw and filtered Seawater samples.
Figure 3. SEM image of particles passed through 50-micron screen filters.

Matheickal: Development of dinoflagellate "cyst-on-demand" protocol

Particle shadow
corresponding to

Flow Cell
its size
Figure 4. Light Obscuration Principle.
Mixed Culture - Coulter
Seawater - Coulter
Mixed culture - Light
Seawater - Light obscuration
Number [/ml]
Number [/ml] 3,000
5206 4,123
391 498
5 34
4 7
2 2
2 0
58 208
18 53
8 12
2 4
1 2
1 0
Size (microns)
Size (microns)
Arizona Dust - Coulter
Arizona Dust - Light Obscuration
Number [/ ml]
1 183
1 6
size (microns)
Figure 5. Comparison of particle counts obtained for different water samples.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Count detected by
light obscuration
Counts detected by light
obscuration (particle/ml)
Particle concentration in sample water
Figure 6. Coincidence error in light obscuration instruments.
Voltage level for particle free sample
Voltage level for particle within the sizing range
Voltage level for particle at the limit and beyond
the sizing range
Figure 7. Diagram explaining particle size range limits of light obscuration instruments.
Figure 8. CCMP1735 hypnozygotes formed using the protocols developed in this study.

Matheickal: Development of dinoflagellate "cyst-on-demand" protocol
Rate of Germination
Encystment rate of Scrippsiella
no. of cysts/ml 200
no. of cysts/ml
Time ( days )
Time ( days )
Figure 9. Excystment and encystment patterns for CCMP1735 hypnozygotes.

Test procedure for evaluation of ballast water treatment
system using copepoda as zooplankton and
dinoflagellates as phytoplankton
T. Kikuchi1, K. Yoshida 1, S. Kino 1 & Y. Fukuyo 2
1The Japan Association of Marine Safety,
kikuti@oak.ocn.ne.jp, yoshida@lasc.co.jp,
2Asian Natural Environmental Science Center
University of Tokyo
Name of project
The project "Research and Development of the Special Pipe System for Ballast Water Treatment"
implemented by the Japan Association of Maine Safety with the support of Japan Foundation has two
components: 1) improvement the special pipe system to achieve better effectiveness in the termination
of zooplankton and phytoplankton, and 2) development of the procedure and standard for evaluation
of the effectiveness of the treatment system. This paper describes the second component, and the first
one is also explained in another article recorded in the same proceedings.
Treatment options being researched
The test procedure was first designed to evaluate the special pipe system, one of the mechanical
treatments. But its concept and the procedure itself can be applied to the analysis of the effectiveness
of any other method.
Timeframe of the project
The project commenced in April 1999 and is still on going.
Aims and objectives of the project
The objective of this study is to develop a specific test procedure for evaluation of a ballast water
treatment system to terminate and eliminate harmful aquatic organisms in ballast water based on
biological and ecological nature of the organisms in coastal waters.
Research methods
In order to establish an appropriate test procedure, it is essential to analyze the biological and
ecological features of organisms in port areas where ballast water has been taken. Seasonal change
and regional difference of composition and numbers of plankton in Japanese waters were observed
using several references such as Nomura and Yoshida (1997). Special attention was paid to high
phytoplankton numbers occurring during red tides.
Based on data obtained by the analysis of plankton nature, necessity of selection of test organisms for
evaluation of ballast water treatment system was assessed. For the selection, following criteria were
considered; 1) the test organisms should be available in a certain amount easily anytime and anywhere
to put enough concentration in test water to evaluate the result; 2) the organisms must be found in

Kikuchi: Test procedure using copepoda as zooplankton and dinoflagellates as phytoplankton
both near-shore and off-shore waters easily, as the evaluation experiment includes a test bed test on
land and a onboard test in ship; 3) the organisms should be easily differentiated with respect to its
survival or fatality with high accuracy for evaluation of effectiveness of treatments. A test procedure
and a standard for ballast water treatment were also designed using results of above mentioned
Ballast water has not only planktonic organisms, but also small benthic ones living in bottom
sediment and being re-suspended by water flow, if water is charged at shallow ports. But it is
appropriate to use only planktonic organisms at first for the materials of the present study in order to
simplify the way of discussion. Introduction of benthic organisms such as mussel and seaweeds may
be made not by transport of benthic adult organisms, but by planktonic eggs and larvae, of which
numbers are usually larger by more than several thousand times.
Results and Discussion
Phytoplankton and zooplankton community changes in natural environment
Tokyo Metropolitan Government monitors red tide occurrence in Tokyo Bay regularly and reports
phytoplankton number as one of the parameters observed. In 1999 and 2000, the highest, lowest
and average cell numbers were 188,860, 76 and 16,260 cells/ml, respectively, among 312 samples
(Tokyo Metropolitan Government 2002). Their methods of sampling and observation were not
described in details. But the cell numbers must be based on quantitative analysis of live samples
collected by a bucket and kept without using any fixative reagent under a regular compound
microscope, as commonly applied for red tide research.
Seasonal fluctuation of plankton number is wide in eutrophic temperate areas such as Tokyo Bay.
High nutrient concentration can keep high number of plankton individuals. However, sometimes
other environmental parameters disturb the increase of plankton number, and therefore range of
individual number becomes wider. Nomura and Yoshida (1997) reported the change of
phytoplankton composition and cell numbers of 35 monthly samples collected by a bucket from
surface and preserved by formalin at Tokyo Bay during 1991 and 1993. Summary of their results
is as follows;
1. 55 species (33 diatoms species, 19 dinoflagellates species and 3 other algal species) were
2. plankton community was composed of diatoms (92%), dinoflagellates (7%) and others
3. phytoplankton cell number was 7 ­ 8,607 cells/ml by counting after preservation of
specimen under a regular compound microscope
4. plankton composition and cell number sometimes showed big difference from those
suspected from the chlorophyll a amount analysis and the preliminary observation of live
specimen. This means that phytoplankton was sometimes dominated by unfixable species.
Nomura (1998) reviewed historical phytoplankton records in Tokyo Bay between 1907 and 1997
using more than 45 publications and summarized the number of species reported in certain
duration. In the whole years occurrence of more than 273 species, of which 78, 66, 59, 187 and
119 species were reported in 1900-1940s, 1950-1960s, 1970s, 1980s and 1990s, respectively,
were recorded. The number of species varied depending on environmental condition of the bay
and techniques of sampling and observation used for the study.
Red tide is defined as discolored water caused by high concentration of microscopic unicellular
organisms. This represents one of the highest concentrations of plankton. Japan Fisheries Agency
has been issuing an annual report on red tides that occurred in Seto Inland Sea in western Japan

2nd International Ballast Water Treatment R&D Symposium: Proceedings
since 1973. Following is a summary of data collected between 1992 and 2000 (Japan Fisheries
Agency 1993-2001).
Total case number of red tides: 1020 cases
Number of causative species: 46
Highest and lowest cell number: 476,700 and 10 cells/ml (a half of cases >5,000 cell/ml)
Longest and shortest duration of a red tide: 276 and 1 day (a half of cases <4 days)
Largest and smallest area size covered by a red tide: 1360 and 0.0005 km2 (a half of cases <10
Most of research on zooplankton composition analysis used a plankton net, with a mesh size of
more than 80 µm, as a sampling tool. Quite few data are useful to analyze the change of
individual number of whole zooplankton, i.e. plankton community of all size ranges.
Tokyo Metropolitan Government observes zooplankton number simultaneously at red tide
monitoring research in Tokyo Bay regularly and reports the number as one of the parameters
observed. In 1999 and 2000, the highest, lowest and average individual numbers of zooplankton
were 667,140, 90 and 34,299 ind./l, respectively, among 305 samples (Tokyo Metropolitan
Government 2002). The numbers were based on quantitative analysis of live samples collected by
a bucket and kept without using any fixative reagent. The large part of community was occupied
by unicellular protozooplankton, and large-sized zooplankton such as copepods were usually
minor member in individual number.
Shizuoka Prefecture (1999) reported seasonal change of plankton composition at the central part
of Sagami Bay, which has good water circulation influenced by Kuroshio Current.
Microzooplankton smaller than 22 µm was dominated by unicellular protozoa such as ciliates and
appeared several hundred individuals per liter. Zooplankton larger than the size was organized
into various groups of animals such as variety of copepods (Maxillopoda), arrow worms
(Sagittoidea) and planktonic sea worms (Polychaeta). Among them copepods is common and
dominant organism, and individual number is about 100 ind./l.
Several ecological characters on plankton community in Japanese coastal waters become clearer
from the analysis described above.
1) Phytoplankton species number varies depending on environmental physical, chemical and
biological condition.
2) Phytoplankton cell number also varies greatly from 188,860 to 76 with 16,260 cells/ml as
an average by observation of live specimens, and 8,607 to 7 cells/ml by preserved
specimens in Tokyo Bay. It means that the cell numbers decreased much in preserved
3) In red tides cell number reached as much as 476,700 cells/ml.
4) Zooplankton individual number also varies from 667,140 to 90 ind./l, but average is
34,299 ind./l. Most of the community was occupied by unicellular protozooplankton
smaller than 20 µm.
5) Zooplankton larger than 20 µm contains a large variety of organisms, and copepods
always appears about 100 ind./l.
Quite wide diversity of plankton, both phyto- and zooplankton, is obvious in terms of organism
number and species variety. Phytoplankton cell number differs about 5,000 times, and species
number 3 times by sampling times. Zooplankton individual number also varies about 7,000 times.
Consequently it is fundamentally necessary to define organism(s) to use as test materials for

Kikuchi: Test procedure using copepoda as zooplankton and dinoflagellates as phytoplankton
evaluation of ballast water treatment system. Result using low plankton concentration is not
comparable to those using 7,000 times high concentration.
Biological character influencing to the evaluation
Among phytoplankton, diatoms and dinoflagellates are two major components. Zooplankton has two
groups, i.e., small unicellular protozooplankton has ciliates, and larger microzooplankton has
copepods as major members. For evaluation of effectiveness of treatment the judgment on the
viability of the test organisms is crucial. Change of shape and mobility is indicative character useful
for evaluation.
Diatoms and copepods do not change their shape by preservation using chemicals. But almost all
ciliates burst and disappear by sudden change of temperature or salinity and also by fixative reagents.
Dinoflagellates have both groups. One half has thick cellulose plates on cell surface and they do not
change their shape by fixation, but the other half has no plate and change shape or disappear by
bursting during fixation.
Differentiation of live or dead is easy in organisms that have mobility. All zooplankton actively move
and some phytoplankton such as dinoflagellates also can move by their flagella. But diatoms cannot
move because of lack of any organ for movement and therefore mobility cannot be taken as an
indicating character to judge viability.
Photosynthetic members of phytoplankton, i.e. diatoms and a half of the dinoflagellates, have color
due to photosynthetic pigments. After death of cells, the color disappears gradually, but the color
remains sometimes for more than a day. Damaged cells also have less color, but potential of recovery
cannot be judged by appearance. Therefore color of organisms, either by organism-specific pigment
or stained by chemicals, is not adequate to use for the judgment. Many cells show faint color after
treatment and it makes judgment of treatment effectiveness very difficult.
Test organisms cannot be preserved by fixative chemical to observe if they are alive or dead. It means
that organisms in samples have potential to grow even after treatment. Unicellular organisms can
make cell division often once to several times a day. Diatoms often have short doubling time of
several hours. But dinoflagellates can make cell division once a day at maximum. Therefore samples
after treatment should be analyzed within a few hours to avoid the change of cell number.
Selection of test organisms
In phytoplankton group, diatoms and dinoflagellates are two major members available everywhere
almost always. Judgment of treatment effectiveness using diatoms must be impossible, because they
are immobile and do not change their shape. Dinoflagellates (Dinophyceae) have advantage, as they
stop moving and about a half of them change shape after death. Therefore Dinophyceae have better
indicative feature for a test organisms.
In zooplankton group, protozooplankton smaller than 20 µm are abundant near shore, but rare in off
shore waters. Copepods (Maxillopoda) is major member and appears everywhere any time, but other
zooplankton such as arrow worms occur at certain time in a year.
Based on criteria described in the Methods and data described in the Section 1 of the Results, the
testing organisms could be selected Dinophyceae from phytoplankton and Maxillopoda (Copepoda)
from zooplankton. These individuals with 20 µm or more in size may be used for experiments.
Settlement of test procedure
Plankton number varies in very wide range. High concentration numbers are several thousand times
larger than low concentration numbers. Results of treatment using high concentration of plankton
must be very different from those using low concentration.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The observation of performance and the judgment of effectiveness of the treatment may be conducted
in the following steps. Observation of all samples collected should be conducted within one day after
the sampling of the water, thus avoiding, as far as practical, the change of conditions of targeted
organisms under storage. The environmental parameters of the waters before and after the treatment
should be observed, including temperature, salinity and pH.
Step 1:
Take seawater samples before and after the treatment with 100 litres or more, i.e. at the
points of inlet and outlet of a treatment system. Volume of sample water should be noted for
calculation of plankton concentration in each sample.
Step 2:
Slowly concentrate phytoplankton and zooplankton in the sample water of known volume
by using plankton nets or meshes with pore size of 20 µm. This concentration process
should be conducted to observe many testing organisms by speedy observations under a
microscope. Concentration should be done slowly to avoid any damages to the plankton
through such process.
Step 3:
Transfer the concentrated sample seawater into a clean receptacle such as a beaker, and
adjust to 500ml or one litre with seawater filtrated through GF/F filters.
Step 4:
A certain quantity of the sample water should be taken from the receptacle, and then the cell
number of Dinophyceae with exercising of flagella and normal shape, and the individual
number of Maxillopoda with normal motion and shape must be counted under a compound
microscope and a stereoscope, respectively. The volume of water observed must be noted.
This observation and counting should be repeated, until not less than 100 cells of
Dinophyceae and 100 individuals of Maxillopoda can be obtained, to ensure high reliability.
Step 5:
The phytoplankton and zooplankton counted should be identified at the ranks of genus of
Step 6:
The results of the counting, i.e., the total number of the normal cells of Dinophyceae and
individuals of Maxillopoda, must be recorded together with the total volume of test water
observed. Then the total number of normal cells of Dinophyceae and of normal individuals
of Maxillopoda per liter of test waters before and after the treatment must be calculated and
Step 7:
To ensure the reliability of the data obtained, the test should be conducted not less than 3
times using same seawater under same environmental condition, and the mean and the
deviation values from the results should then be obtainable.
Step 8:
By comparing the number of the indicator organisms (Dinophyceae and Maxillopoda)
before and after the treatment, the rate of diminution and attenuation may be calculated and
the efficiency of the system.
Standard for ballast water treatment approval
According to the analysis of plankton community and its ecological characters, such as wide variation
of cell density, described above, following standards for type approval is suggested.
95% of Dinophyceae and Maxillopoda more than 20 µm in size should be removed, rendering
harmless, inactivated through the process from inlet to outlet of the system.
The percentage looks small, but it should be thought as the starting point of system development.
Higher percentage, i.e. higher efficacy, should be applied after certain period.

Kikuchi: Test procedure using copepoda as zooplankton and dinoflagellates as phytoplankton
Conclusions and Recommendations
As the experiment to evaluate treatment systems will be conducted at various places throughout the
world under various circumstances by both test-bed and on-board tests, the procedure of the
experiment should be clearly defined with special consideration to the reproductivity and reliability of
the result. Use of whole planktonic organisms occurring in the areas of the experiment as test
organisms for the evaluation increases difficulty of experiments themselves and evaluation of results
of the experiments. As the analysis of plankton composition before and after the experiment is by
counting only, live individuals is thought to be essential and inevitable, but it is nearly impossible to
conduct it with scientific accuracy. Diatoms, one of the major components of phytoplankton, are
immobile and the change of diatom cell color may not occur in a short time, even in case the cells
died completely.
The conclusions of the present study are:
The potential test organisms for evaluation of ballast water treatment system can be
Dinophyceae from phytoplankton and Maxillopoda (Copepoda) from zooplankton. These
individuals with 20 µm or more in size can be used for experiments.
Evaluation of efficacy should be based on termination rate of the test organisms before and
after treatment. Live or dead can be distinguished by shape and mobility of the test organisms.
In order to keep reproductivity and accuracy of the evaluation, number of test organisms in
test water should be counted no less than three times.
Standard for treatment approval is termination rate of test organisms more than 95%. The rate
should be set higher along with the development of techniques.
Concerning the cost of experiments, it is difficult to calculate it, because it varies depending on scale
of experiments. Quantitative analysis (triplicate observation) of phytoplankton and zooplankton with
judgment of live or dead costs 200 US$ per sample.
Fisheries Agency 1993. Information on the occurrence of red tide in the Seto Inland Sea, Annual
Report 1992
. Seto Inland Sea Fisheries Coordination Office, 59pp. (in Japanese).
Fisheries Agency 1994. Information on the occurrence of red tide in the Seto Inland Sea, Annual
Report 1993
. Seto Inland Sea Fisheries Coordination Office, 48pp. (in Japanese).
Fisheries Agency 1995. Information on the occurrence of red tide in the Seto Inland Sea, Annual
Report 1994
. Seto Inland Sea Fisheries Coordination Office, 61pp. (in Japanese).
Fisheries Agency 1996. Information on the occurrence of red tide in the Seto Inland Sea, Annual
Report 1995
. Seto Inland Sea Fisheries Coordination Office, 54pp. (in Japanese).
Fisheries Agency 1997. Information on the occurrence of red tide in the Seto Inland Sea, Annual
Report 1996
. Seto Inland Sea Fisheries Coordination Office, 46pp. (in Japanese).
Fisheries Agency 1998. Information on the occurrence of red tide in the Seto Inland Sea, Annual
Report 1997
. Seto Inland Sea Fisheries Coordination Office, 79pp. (in Japanese).
Fisheries Agency 1999. Information on the occurrence of red tide in the Seto Inland Sea, Annual
Report 1998
. Seto Inland Sea Fisheries Coordination Office, 81pp. (in Japanese).
Fisheries Agency 2000. Information on the occurrence of red tide in the Seto Inland Sea, Annual
Report 1999
. Seto Inland Sea Fisheries Coordination Office, 60pp. (in Japanese).

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Fisheries Agency 2001. Information on the occurrence of red tide in the Seto Inland Sea, Annual
Report 2000
. Seto Inland Sea Fisheries Coordination Office, 62pp. (in Japanese).
Nomura H. 1998. Changes in red tide events and phytoplankton community composition in Tokyo
Bay from the historical plankton records in a period between 1907 and 1997. Oceanography in Japan,
7(3), 159-178 (in Japanese with English abstract).
Nomura, H. & Yoshida, M. 1997. Recent occurrence of phytoplankton in the hyper-eutrophicated
inlet, Tokyo Bay, central Japan. La mer, 35, 107-121 (in Japanese with English abstract).
Shizuoka Prefecture 1999. Report of the research on the usability of deep sea water in Suruga Bay.
110pp. (in Japanese).
Tokyo Metropolitan Government 2002. Report of red tide monitoring survey in the innermost part of
Tokyo Bay in 1999 and 2000
. Tokyo environmental office Administrative Division, 235pp. (in

Testing ballast water treatment equipment
A. E. Holdų
Faculty of Engineering and Information Sciences
University of Hertfordshire
Introduction, aims and objectives
Ballast water treatment equipment is reaching a critical stage of development. There are at present
several methods and types of ballast water treatment equipment available for the ship owner or
operator and there are also many considerations in choosing the appropriate type of equipment for a
specific ship. Importantly, when selecting equipment there is a need to ensure that the equipment will
perform to requirements. Computational Fluid Dynamics (CFD) methods can be used with advantage
for the design stage of ballast water treatment equipment together with analytical and scale model
tests. However, due to the complex nature of most proposed and current ballast water treatment
equipment, it is necessary to carry out near full scale tests in order to ensure that the equipment
performs according to specification. Furthermore, there may be a range of tests which needs to be
performed. This range of tests may well include biological sample as well as being of a
hydrodynamic/thermodynamic nature.
In order to ensure that equipment that is installed onboard ships performs to specification and
expectation it is necessary to carry out tests using a purpose made facility. The present paper describes
such a facility at the University of Hertfordshire.
The aim of this presentation is to stimulate discussion about the next stage of ballast water treatment,
namely the application and installation of equipment onboard ships and how to ensure that this
equipment will perform according to plan and expectation.
Research methods, test protocol and experimental design proposed
The focus of most ballast treatment equipment is to separate out or render biological hazards
harmless. The dimensions and methods used for ballast water treatment equipment means that to
ensure physical similarity between a model and a real installation, a number of dimensionless groups
have to be satisfied simultaneously for any given treatment method.
An example is the case of a method based on heat treatment. In this case the equations governing the
convective flow within such equipment due to an applied temperature field are the conservation
equations; namely conservation of mass, momentum and energy. These equations are given as (Holdų
et al, 2000):
Conservation of Mass:
( u
) ( v
) (w)
= 0



Conservation of Momentum:


+ u u = -p + µ u
+ f

i, j
[ i,j]
, j


2nd International Ballast Water Treatment R&D Symposium: Proceedings
Conservation of Energy:


+ u T =
, j
(kT,j) H
, j

The equations give rise to a number of important dimensionless groups or force ratios for convective
flows not including the transport of species. These ratios are:
· The Grashof number; which is the ratio of buoyancy to viscous forces and is given as :
Gr =
· The Rayleigh number; which is the ratio of inertial times buoyant forces to the square of
viscous forces

Ra =
· The Prandtl number; which is the ratio of mass to thermal diffusivity
Pr = k
· The Reynolds number; which is the ratio of inertia to viscous forces and is given as:
Re = µ
In analysing this type of problem it is important to choose the relevant physical and geometrical
quantities in order to obtain the correct values for dimensionless parameters. Demonstrated by the
above equations and dimensionless groups, it is impossible to satisfy them all when using a scale
model of the equipment even for the case where there are no species being transported. This is most
readily seen from the fact that some groups include L whilst other groups include L3.
Similar analysis for physical similarity can be carried out for other type of ballast water treatment
equipment. Clearly, different types of dimensionless groups will be appropriate for scaling, however,
for a majority of the cases it can be shown that is necessary to use near full scale conditions for
ensuring that equipment performs according to specification. For this reason it is essential to have
facilities which can test and consequently certify equipment at near full scale conditions and where a
variety of methods emulating the biological hazards can be introduced.
In order to satisfy ballast water treatment equipment testing a hydrodynamic based test facility for
such purposes has been constructed at the University of Hertfordshire. The facility is based in a
building with 150m2 free surface area. This floor space sits on top of two tanks (Figure 1) which can
be connected to each other using various approaches.
The volume of the tanks are 60m3 and 630 m3 which enables a realistic amount of sample water to be
passed through ballast water treatment equipment on test. It is important to ensure that a sufficient
amount of water is passed through the equipment in order to achieve a realistic statistical basis for
analysis. The necessary volume of water may be debatable, but it is also related to the typical flow
rates that will be used. The flow rates that are used in present equipment is of the order of hundreds of
cubic meters per hour.
The test facility has at present installed pump capacity of 400 m3/hour. This means that the large tank
will give an hour and a half of flushing contaminated water through test equipment, whilst the smaller
tank can be used for 10 minutes tests. The tanks can be connected in several ways so that the smaller
tank can be used as a mixer tank for preparing several types of water conditions to be tested. This may

Holdų: Testing ballast water treatment equipment
well be necessary as in some cases the biological contaminant concentration and conditions may vary
significantly. Furthermore, the presence of sand, mud or other particles in the water may well affect
the performance of the treatment of the biological hazards. If this is the case then such conditions
must also be part of a performance test.
It is clearly also possible to exhaust the water back into any of the tanks and this enables long term
performance tests. Such tests may well be of interest when equipment contains filters which needs to
be cleaned at various stages of operation. Filters may not be the only part of ballast water equipment
that may need the long term tests made possible through exhausting treated water back into either of
the two tanks. However, it is clearly necessary to perform such tests for operational reasons.
Calibrated flow rate meters and pressure transducers are available in the laboratory together with
other standard hydrodynamic instrumentation. Particle counting and sizing equipment available for
quantitative measurements is also available and can be used on a continuous or sampling basis for
assessing the specification of inflow water composition as well as equipment performance in terms of,
for example , separation efficiency of particles.
In many cases, it may not be sufficient to carry out mechanical based particle tests only and it is likely
to be necessary to perform biological tests. Towards this end, the facility is equipped with a variety of
sampling port locations as well as locations for introducing the biological contaminant. On such
sampling point is shown in Figure 3. The biological contaminant may also be introduced and mixed
with water in the smaller of the two tanks.
The laboratory also offers facilities such as heating coils, compressed air (e.g. for back-flush of filters)
and thermal probes and measurements.
Initial tests have been carried out on Optimarin a/s ballast water treatment equipment with some
success. Figure 4 shows the equipment during installation in the test facility.
The tests demonstrated that the flow rates and modes of operation described can be achieved. During
the tests, back flush operations for filters were also performed. Biological sample preparation and
testing was also carried out by Dr Voigt using the Artemis method (Voigt&Gollasch, 2000;
Voigt&Rosenthal, 2000).
Conclusions and Recommendations
· The present paper demonstrates the physical complexity of a ballast water treatment method
and proposes that it is necessary to carry out validation experiments for such equipment.
· A facility for ballast water treatment equipment is demonstrated and presented
· A discussion on the testing protocol for ballast water treatment equipment is recommended
Armstrong, G.M., Rose, A. & Holdų, A.E. 1999. An Analysis of Flow-through Ballast Water, Trans.
Inst. Marine Eng. Vol. 112 pp. 51-65.
Holdų, A.E., Armstrong, G.M. & Rose, A. 2000. An Analysis of Flow-through Ballast Exchange,
2000, Trans. Of 8th Int. Conf on Marine Eng. Systems, Paper A6, New York, May 2000.
Voigt, M., Gollasch, S. 2000. Ballast Water: The Latest Research and Methods of Treatment.
Proceedings of the 3rd Annual Conference "Managing environmental Risk in the Maritime Industry"
25th ­ 26th October 2000, London, UK
, 10 pp.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Voigt, M. & Rosenthal, H. 2000. Management of Ballast Water in Ports. Maritime Conferences
"Ships Waste ­ Management and Treatment in Ports and Shipyards", 28 ­ 30 June 2000,
Bremerhaven, Germany
. 10pp.

Holdų: Testing ballast water treatment equipment
Figure 1. Top of water tanks with access hatches.
Figure 2. Pumping Facility with total capacity of 400 m3/hour.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 3. Injection sampling point for biological samples.
Figure 4. Ballast water equipment of Optimar during installation in the test facility at the University of

Performance verification of ballast water treatment
technologies by USEPA/NSF Environmental
Technology Verification Program
T. G. Stevens 1, R. M. Frederick, R. A. Everett, J. T. Hurley, C. D. Hunt, & D. C. Tanis
1NSF International
Introduction and objectives
The U.S. Coast Guard (USCG) is tasked in the United States with developing and implementing a
program for regulating the discharge of ballast water from ships. The USCG has teamed with the U.S.
Environmental Protection Agency (USEPA), through the Environmental Technology Verification
(ETV) Program's Water Quality Protection Center (WQPC), to develop testing protocols that will
evaluate the effectiveness of technologies to address invasive species present in ballast water. The
cooperative effort was initiated in June 2001, and has progressed to the development of a draft
protocol for evaluation of technologies.
The EPA ETV program provides credible, independent data on the actual performance of
technologies designed to prevent or control degradation of ground and surface waters. Stakeholder
input is an important aspect of the ETV Program, and provides direction for development of testing
protocols and implementation of the Program. Through technically sound protocols and appropriate
QA/QC, testing provides information on the ability of technologies to achieve treatment, and provides
information for potential purchasers and regulators regarding operation and maintenance of the
technology. The information obtained from testing is made public through publishing of a full
verification report and a summary verification statement, both of which are posted for general public
access on the EPA web site. NSF International, a not-for-profit, third-party certification organization,
is the verification partner working with EPA to implement the WQPC. Further information regarding
the ETV Program and the WQPC is available on the EPA web site (www.epa.gov/etv), while
information about NSF and their effort in the WQPC is available on the NSF web site
The primary objective of the USCG's involvement in the ETV effort is to develop a mechanism for
verifying the performance of ballast water treatment technologies. It is likely that many elements of
the protocols will be incorporated into type tests, which will provide the information needed for
USCG certification of the technologies. It is also the objective of the effort to coordinate protocol
development to meet international approval agency needs.
Proposed experimental design
Information and input on the verification approach and experimental design for verification testing of
ballast water treatment technologies was obtained from an initial meeting of a general stakeholder
group. The straw experimental approach developed from the stakeholder input was presented to an
18-member technology panel, representing developers and vendors, the shipping industry, regulators
and researchers. The tech panel has worked with Battelle, the contractor selected to write the protocol,
to develop a draft document that will soon be available for general stakeholder review. In developing
the draft document, the tech panel was reminded that the ETV program is not a research project ­ that
only essential, need-to-know issues be addressed in the protocol, and that the cost to complete testing
needs to be an important consideration in developing the experimental design.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
The following verification factors, or questions to be addressed by the testing protocol, were
identified by the stakeholders and provide the foundation for the protocol:
Biological treatment performance ­ determination of the technology's ability to remove, inactivate
or destroy organisms, as measured by removal efficiency (percent) or a threshold (water quality
standard); this also addresses the potential for organisms to survive treatment to reproduce (regrowth).
Operation and maintenance ­ measure of the operator time, effort, and skill required to achieve the
performance achieved during the testing.
Reliability ­ measure of the ability of the technology to perform consistently over a period of time.
Cost factors ­ determine the amounts of consumables (i.e., chemicals, filter media, power, etc) and
labor hours required to achieve the stated level of performance.
Environmental acceptability ­ evaluation of the compatibility of the treatment technology with the
receiving waters, particularly with regard to residuals of treatment chemicals or by products produced
by the treatment process.
Safety ­ evaluation of potential chemical, electrical, mechanical or biological hazards associated with
the operation and/or maintenance of the technology.
The tech panel agreed that the protocol should only address prefabricated, commercial-ready systems,
and that components of systems not be evaluated separately. While operational data for technologies
under actual use conditions was deemed desirable, the tech panel decided that the protocol should
initially address land-based testing, and that shipboard verification testing be addressed by a later
protocol, as appropriate. This decision was based on:
· Land-based testing is necessary to provide comparable conditions for verifying technology
performance, particularly systems that will not need to be scaled up for shipboard use; and,
· Shipboard testing is critical for evaluating technology-engineering performance and for
systems requiring scale up to accommodate higher flows.
Addressing the verification factors and taking into account the need to provide a repeatable evaluation
protocol, Battelle generated an initial document that was subsequently reviewed by the tech panel. A
special working group was also formed to consider the challenges posed by the biological
measurements necessary to ensure treatment efficacy and to provide input on the experimental design.
The second draft of the protocol, which will be submitted for review by the tech panel, incorporated
comments from the tech panel and the working group.
Proposed protocol approach
The test design in the protocol addresses the physical/chemical and biological challenge conditions,
duration of testing, replication, reliability, biological effectiveness measures, and core measurements.
The protocol further describes test site arrangements to conduct the testing, and includes guidance on
methods for analysis of samples and statistical analysis of the test data. As with all ETV protocols, a
detailed description of QA/QC procedures is provided to assure the credibility of the acquired data.
The draft challenge conditions, shown in Table 1, allow for verification of technology performance
with water conditions that are difficult to treat and representative of the range of conditions found in
the natural environment, excluding rare or extreme conditions that might occur.

Stevens: Performance verification of ballast water treatment technologies by USEPA/NSF
Table 1. Water Quality Challenge Matrix
Water Type
Water Quality Characteristics
Fresh (<1 PSU)
DOC = 8 ­ 12 mg/L
POM = 8 ­ 12 mg/L
MM = 16 ­ 22 mg/L
Sum of POM and MM = 24 ­ 34 mg/L
Temperature: 10 ­ 25ŗC
Marine (~ 33 PSU)
DOC: 8 ­ 12 mg/L
POM: 8 ­ 12 mg/L
MM: 16 ­ 22 mg/L
Sum of POM and MM: 24 ­ 34 mg/L
Temperature: 10 ­ 25ŗC
Where: DOC is dissolved organic content
POM is particulate organic matter
MM is mineral matter
The water available at the testing facility will be amended to achieve the required water quality
conditions by adding organic and mineral matter as necessary. The purpose of the testing is to
evaluate a technology's ability to remove, destroy or inactivate organisms. Three groups of organisms
are included in the test protocol, including bacteria, protists and macroalgae, and zooplankton. Both
ambient populations in the water at the test location and surrogate (added) species will be used during
the testing to determine treatment efficacy. A list of organisms under consideration is included in
Table 2. Additional work, described below, is needed to identify the appropriate surrogates to be
included in the test protocol.
Table 2. List of Potential Surrogate Species
Functional Group
Fresh Water
Marine Water
Bacillus globigii
Bacillus marinus
Bacillus similis
Bacillus licheniformis
Bacillus cereus
Clostridium perfringens
Enterococcus spp.
Oyster larvae
Sea urchin larvae
Protist (resting cyst form)
No surrogate identified ­ natural populations will be used
Macroalgae (fragmentor)
To be determined.
Naturally occurring species, such
a s E n t e r m o r p h a s p p . o r
Caulerpa spp.
For tests to be initiated, threshold concentrations of ambient organisms are required at a test location.
The thresholds are 106 bacteria per liter, 102 ­ 103 zooplankton per liter, and 105 protists per liter.
Surrogate additions will be made in the same concentrations. Surrogates used in the testing will be
obtained from commercial suppliers where available, or prepared at the test facility. In either case, an
assay will be conducted to evaluate the viability of the surrogates upon receipt at the test site and
within 24 hours of a test cycle.
Six biological efficacy tests will be conducted to determine a technology's treatment effectiveness.
Three marine and three fresh water tests, using surrogate and ambient species to determine
effectiveness, will be completed over the course of the technology evaluation.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Two approaches are possible for evaluation of the reliability of the technology ­ based on either
ballasting cycles (as specified in the technology O&M manual) or minimum treated volume. Under
the ballasting cycle approach, a technology will be operated for the number of operational cycles
needed to achieve 150 percent of the vendor specified operation and maintenance cycles. For
example, if the vendor indicates that their technology can operate for 40 operational cycles (or hours
of operation) before requiring O&M, the test would run for 60 cycles (or hours), during which six
biological efficacy tests would be completed. In-tank treatment technologies using chemical biocides
may be operated without active agent addition during non-biological efficacy cycles to evaluate the
electro/mechanical aspects of the technology.
The minimum treated volume approach is based on either the number of ballasting cycles required to
achieve treatment of a minimum of 10,000 m3 of challenge water for in-line treatment technologies
(equivalent to approximately 30 hours of operation at 300 m3/hour), or 1,800 m3 for in-tank
technologies (equivalent to six biological test cycles of 300 m3 of water). Figures 1, 2 and 3 show test
set ups and sampling schemes for evaluation of technologies designed for different ballast operation
The sampling locations for each of the testing arrangements are indicated in the figures. Samples will
be collected simultaneously during efficacy test runs in three one-m3 tanks, providing triplicate time-
integrated samples at each sample location. Each sample will be sub-sampled for the core parameters,
as shown in Table 3. In situ sensors will be used, where possible, to monitor water quality and proxy
(e.g., chlorophyll, turbidity, etc.) parameters during test runs. Standard analytical methods (USEPA
methods, Standard Methods, ASTM, etc.) will be used for analysis, where available, or non-standard
methods will be used and described in the verification Test Plan. The Test Plan will include detailed
information needed to complete the verification evaluation, and will be specifically developed for
each technology and testing site.
Table 3. Core and supplemental parameters
Sample Location and Approach

Challenge Water
Post Treatment
Core Measurements
In situ, Continuous
In situ, Continuous
Test facility
In situ, Continuous
In situ, Continuous
Test facility
Total suspended solids
Discrete grab
Discrete grab
Particulate organic matter
Discrete grab
Discrete grab
Dissolved organic matter
In situ, Continuous,
In situ, Continuous,
Test facility,
Dissolved oxygen
In situ, Continuous
In situ, Continuous
Test facility
Dissolved Nutrients
(N, P, Si)
Indigenous species
Surrogate species
Proxy measures
Turbidity (represents TSS)
In situ, Continuous
In situ, Continuous
Test facility
Chlorophyll a (biomass)
In situ, Continuous
In situ, Continuous
Test facility
ATP (living material)
Discrete grab,
Discrete grab,
Laboratory Test
Continuous as
Continuous as
A key part of the biological efficacy testing will involve determination of the viability of organisms
remaining in the challenge water after treatment by the technology under evaluation. An enrichment

Stevens: Performance verification of ballast water treatment technologies by USEPA/NSF
approach will be used for bacteria and protists. A +/- scoring system will be used, with multiple media
and nutrient levels, and light and dark incubations to separate autotrophs. Zooplankton viability will
be determined by observation of samples for movement.
Part of the evaluation will also include regrowth of the organisms, which will be determined by
holding treated water for up to five days, then measuring abundance and viability. Longer holding
times may be used where the technology dictates, and will be indicated in the Test Plan.
Part of the verification evaluation for technologies employing a biocide involves toxicity testing to
evaluate the potential impact the technology would have on a receiving water , and to be sure that
discharge of the waste from the testing site will have no negative environmental impacts. The toxicity
testing will be completed during the start-up phase of testing and will use standard wastewater toxicity
tests. Favorable toxicity testing is required prior to initiation of the biological efficacy testing.
Subsequently, a technology would have 30 days to take steps to comply with discharge requirements
or testing would be terminated until the toxicity issue is resolved.
The data generated during testing will be evaluated to determine the efficacy of the technology. The
remaining concentration and percent removal for each naturally occurring or surrogate species will be
calculated, and the statistical significance of the data will be evaluated relative to the treatment control
using a t-test of treated removal versus control removal.
Research needed for indicator species
Although a significant amount of work has been completed toward developing a testing protocol,
identification and selection of species that can be used as surrogates during testing still needs
refinement. As mentioned previously, addition of surrogates is important from the standpoint of being
able to have a protocol that will generate meaningful data from different testing sites, where ambient
species would differ in both abundance and resistance to treatment.
The working group formed to address this issue recommended a comparative study to determine the
relative resistance of the proposed surrogates to different treatment methods (not to include filtration)
through various life stages. The goal is to understand the relative responses of potential surrogate
species, as well as ambient organisms present in the source water. The objective of the study is to
significantly reduce the number of surrogates needed to be included in the testing.
The first stage of the comparative study is to screen a number of species and a range of basic
treatment processes that could be used for ballast water treatment to determine the most resistant
species. A second, more detailed evaluation will be conducted on the resistant species identified in the
first round of testing, with the goal of arriving at a minimum number of species that will present the
greatest challenge to treatment technologies submitted for verification.
Planning is also proceeding for completing a pilot round of testing to evaluate the procedures included
in the protocol. Whether this testing will include surrogate species is unclear, but it will be conducted
using ambient species at the selected testing site. Completion of the pilot testing will help to identify
changes that need to be made to the protocol to achieve the objective of having a test procedure that
will generate meaningful and useful information on the performance of ballast water treatment
The work completed on the ETV Ballast Water Treatment Technology Protocol has developed an
approach that will produce data to assist users, purchasers and regulators in making decisions on the
use of technologies. There is still work to be done, as the protocol is still in a draft stage. The
document will be reviewed by the Tech Panel during the summer of 2003, and should be available for

2nd International Ballast Water Treatment R&D Symposium: Proceedings
general stakeholder review in the fall of 2003. While primarily a U.S. effort to this point, international
input into the protocol is welcome, as development of a standardized approach to evaluation of ballast
water treatment technologies on a global basis, to the extent possible, would be of great benefit to all
parties ­ technology vendors, ship owners/operators and regulators.

Stevens: Performance verification of ballast water treatment technologies by USEPA/NSF
Figure 1. In-line treatment on uptake or in combination with in-tank treatment.
Figure 2. In-tank treatment.

2nd International Ballast Water Treatment R&D Symposium: Proceedings
Figure 3. In-line treatment on discharge or in combination with in-tank treatment.

Papers submitted
(not presented at
the Symposium)
Papers have been included in these
Proceedings largely as submitted, with
basic editing and formatting only, and
without scientific or technical peer

Neither the GloBallast Programme, the
International Maritime Organization
(IMO), the Institute of Marine
Engineering, Science and Technology
(IMarEST) nor the symposium
sponsors take any responsibility
whatsoever for any statements and
claims made in these papers, for the
quality, accuracy and validity of data
presented, or for any other contents of
these papers.

Individuals and organisations that
make use of any data or other
information contained in these papers
do so entirely at their own risk.

Inclusion of papers in these
proceedings in no way constitutes any
form of endorsement whatsoever by
IMO, GloBallast, IMarEST or the
symposium sponsors.

Design optimisation of a UV system for onboard
treatment of ballast water
M. Casey1, A. Leigh 2 & P. Zhou 1
1Department of Naval Architecture & Marine
Universities of Glasgow and Strathclyde
2Willand U.V. Systems Ltd, UK
Treatment option being researched
The current work relates to the design optimisation of a physical ballast water treatment system using
ultraviolet (UV) light.
Project timeframe
This work was carried out over a six-month period from October 2002 to April 2003.
Aims and objectives
The harmful environmental effects on the ocean environment as a result of the translocation of foreign
or unwanted aquatic bodies via ballast water is well documented. The TREBAWA group is a
European consortium that addresses this issue by focusing on the development of a new technically
and economically competitive ballast water treatment system to be employed onboard ship. The
proposed system consists of a primary (mechanical) pre-treatment phase together with a secondary
integrated UV system to prevent microorganisms' transport by disinfection of the ballast water. The
current work focuses on the design optimisation of the secondary (UV) system by simulating the flow
regimes present in proposed designs by employing Computation Fluid Dynamics or CFD.
CFD enables engineers to gain a valuable insight into the possible weaknesses and strengths of
proposed designs. This can reduce the need for costly prototypes and consequently reduce the time
from concept to construction. The main objectives of the current work are:
· To produce a practical and efficient prototype UV treatment system to be employed in
conjuction with a primary separation system for use onboard ship.
· To carry out design optimisation by taking a wide range of possible UV system geometries
and simulating the physical flow characteristics within them using CFD.
· To focus on the most efficient system (or systems) and undertake construction and physical
testing of the prototypes under laboratory conditions.
· To undertake sea trials of the chosen prototype system.
This paper provides details of the approaches made to address the first two objectives. At the time of
writing construction and testing of the chosen prototype are about to begin. Details of meeting the
latter two objectives will be discussed in a later publication.
The UV chamber design must have a geometric form that enables it to be installed onboard together
with existing pipe work in straightforward manner with minimal disruption to the operation of the

2nd International Ballast Water Treatment R&D Symposium: Proceedings
ship. Design scale-up must be addressed as the treatment system is likely to be employed in a wide
range of vessels involving units of differing size. The design must promote ease of use for the
onboard personnel and the material costs, fabrication and manufacture must be economically viable.
The proposed system must be able to facilitate regular maintenance and monitoring. Examples include
cleaning of the UV lamp to avoid foul up, lamp replacement and system shutdown in the event of
malfunction. Innovative UV chamber designs are of little use if the pressure loss incurred makes the
running costs too prohibitive, therefore pumping power estimates must be made at early stage of the
design process.
Research methods
Several previous authors have employed CFD to evaluate UV treatment system performance[1]-[5].
Wright et al[1] and Baas[6] have highlighted the usefulness of evaluating different UV system
designs using CFD. Both these works involved the development of numerical algorithms to simulate
the UV dosage given to particles enabling a comparison to be made of the efficiency of each design.
Wright et al[1] found that making relatively minor changes to the positions of the inlet and outlet
branches resulted in marked improvements to the uniformity of the flow and an increase in the
approximated UV dosage. They concluded that the increased pressure drop incurred was acceptable in
view of the improvements to the system. Baas[6] found that an original UV chamber design could be
improved upon by placing several small UV lamps across the main chamber as opposed to having a
single UV lamp running along the chamber.
Employing CFD as a research method enables several areas to be investigated when undertaking
design optimisation:
· Highlight areas within the UV chamber geometry where there is the possibility of
unfavourable flow regimes. Examples include areas of flow stagnation or areas where the UV
lamps are bypassed (so called `short-circuiting').
· Examine the effect on the predictions by having various UV lamp configurations with varying
chamber diameters and varying flow rates.
· Compare predicted pressure drops for each of the proposed designs.
· Obtain predictions for particle tracking and Residual Time Distributions to help estimate the
efficiency of the unit to impart a UV dosage to the ballast water.
· Investigate the effects of placing flow devices upstream or downstream of the UV chamber in
order to eliminate detrimental secondary flow or back flow elements.
Modeling approach
Model set up
Several commercial software tools have been successfully employed to carry out CFD in UV
chamber design including `CFX'[1] and `Fluent'[5]. Fluent was adopted for the current work and
its effectiveness was assessed in meeting the main objectives. Numerical meshes representing
each of the UV chamber geometries were created using Fluent's GAMBIT preprocessor.
Unstructured (tetrahedral) elements were employed to model the geometry. These elements were
found to model the chamber components more accurately than structured elements. The total
number of computational cells making up a mesh typically ranged from 200,000 for
straightforward single pipe geometries up to 400,000 for more complicated geometries. All the
simulations were run on a Dell GX260 Pentium 4 PC machine and the convergence times varied
from 30 minutes to 2 hours depending on the mesh size. The fluid flow properties for each of the
simulations involved typical values for sea water at 50F ( = 1027.9 kg/m3 and µ =
0.0014 kg/m-s).

Casey: Design optimisation of a UV system for onboard treatment of ballast water
The computations were carried out using the standard set of under-relaxation factors used by
Fluent. Similarly, the default values of the first approximation and the standard pressure-velocity
coupling solver were used. The convergence criterion adopted for each of the residuals was set to
0.0001 for each simulation. The high velocity gradients and high shear stresses present at the
chamber walls were modeled using a standard wall function approach. Ideally a very fine mesh
would be employed to model near wall behavior but this was not achievable with the computing
power available. For all cases the boundary condition at the chamber inlet was the specified
velocity based on the inlet branch diameter and the flow rate. At the chamber outlet the relative
pressure was set to zero.
Turbulence modeling
Pipe diameters ranged from 4 inches to 16 inches and flow rates ranged from 30 m3/hr to
1200 m3/hr. For the pipe size and flow rate combinations tested this equated to upstream inlet
velocities of between 0.5ms-1 and 3.0ms-1. The Reynolds number range for the flows investigated
lie in the turbulent region and it was therefore necessary to adopt a turbulence model to obtain
flow solutions. The more comprehensive Reynolds Stress Model (RSM) and Large Eddy
Simulation (LES) models were discounted because of the increased calculation times required
considering the number of chamber designs being assessed and the computing power available.
Fluent has a number of (k-) based turbulence models available that are the most widely used
turbulence models based on work originally developed by Launder and Spalding[7]. For the
current work a (k-) Realizable model was adopted[8]. It was found that this model gave
improved predictions for the swirling flow component due to its anisotropic treatment of
turbulence compared to the standard (k-) model. This model proved to be more stable compared
to the (k-) renormalized (RNG) model for a range of simulations. This has been confirmed by
Schaler et al[9] in modeling similar flows.
Particle tracking and dosage
The standard Discrete Phase Model, or DPM, provided by Fluent was adopted to simulate particle
tracking. The Lagrangian model employed firstly involves obtaining the solution of the continuity
and momentum transfer equations for the continuous phase then setting the injection conditions
(in this case at the UV chamber inlet). The particles are then tracked using visualization or by
collecting a data summary of the particle behavior from the inlet to the chamber outlet. Particle
movement and particle residence time are critical to the levels of UV dosage the chamber imparts
to the ballast water. Particle track predictions assist in determining whether the water is likely to
be directed close to the UV lamps to obtain the required dosage. Residence time distributions, or
RTD, provide an indication as to whether the particles are in the system long enough to acquire
the required dosage.
If the TREBAWA UV system is to achieve a high percentage microorganism kill rate a
measurement of the lowest dose per particle should be made. This is calculated from the lowest
intensity at any point in the chamber multiplied by the contact time. However, this assumes that
the Residual Time Distribution (RTD) has a value of one. That is each particle spends exactly the
same time in the chamber as every other particle. In practice this has been a fair assumption and
systems have performed well without any performance failures as a consequence of this RTD.
Each RTD data summary provided by Fluent contains statistical data relating to the standard
deviation, the mean and the maximum and minimum of the particle residence time. RTD data was
recorded for particles that were predicted to escape out of the exit following their introduction at
the inlet. Particles that are predicted to become `trapped' in the system and thus show
unrealistically long residence times, caused for example by re-circulation, were highlighted in the
RTD data summary.
The "time to first trace" is the ratio of the fastest particle (i.e. the shortest time) to the mean time
for the particles and was calculated for each chamber design. UV intensity mapping was carried
out at Willand U.V. Systems for each of the designs and when combined with the `time to first

2nd International Ballast Water Treatment R&D Symposium: Proceedings
trace' values was used to develop a system whereby the minimum dose was close to the average
dose. A system that falls into this category is most likely to help reduce the problems associated
with the logarithmic effect of UV disinfection.
Case comparison
Over 75 UV chamber simulations were carried out for varying UV chamber geometries.
Simulations were carried out for designs similar to those that are currently commercially available
and for designs that were proposed through collaboration with Willand UV Systems Ltd by
drawing on their expertise in UV chamber design. This study involved varying the UV lamp
configurations, varying the flow rates, varying the chamber orientations and varying the positions
of the inlet and outlet branches.
Design optimisation involving numerical modelling can only carried out by maintaining
consistency in the modelling techniques employed. It is for this reason that when obtaining steady
state solutions for each of the UV chamber designs the same turbulence model, grid density, fluid
properties and convergence criteria were employed. In addition, the same number of particles was
released at the inlet boundary condition for each design when employing the Discrete Phase
Model. The methods described above were implemented with the aim of focusing on a design or
design(s) that consistently gave favorable predictions for a range of flow rates and chamber sizes
bearing in mind the unit's intended use on a wide range of vessels.
A selection of the UV chamber designs that were modelled is shown in Figure 1. These geometries
were generated using the GAMBIT pre-processor and represent the fluid domain over which the
simulations were