BALLAST WATER TREATMENT TECHNOLOGY

Transcription

1 BALLAST WATER TREATMENT TECHNOLOGY Current status June 2007

2 Lloyd s Register, its affiliates and subsidiaries and their respective officers, or agents are, individually and collectively, referred to in this clause as the Lloyd s Register Group. The Lloyd s Register Group assues no responsibility and shall not be liable to any person for any loss, daage or expense caused by reliance on the inforation or advice in this docuent or howsoever provided, unless that person has signed a contract with the relevant Lloyd s Register Group entity for the provision of this inforation or advice and in that case any responsibility or liability is exclusively on the ters and conditions set out in that contract. Lloyd s Register 2007

3 Contents BALLAST WATER TREATMENT TECHNOLOGY 1. Introduction 3 2. Regulation 4 Ballast water quality and standards 4 The processes 5 3. Treatent processes 7 Background 7 Separation processes 8 Disinfection 9 4. Treatent technologies and suppliers 12 Suppliers 12 Technologies 12 Coercial availability 18 Approval status Suary 20 Annex Listing by supplier 21 Glossary of ters and abbreviations 28 Lloyd s Register 1

4

5 1. Introduction The requireent for ballast water treatent arises fro the requireents of regulation D-2 of the International Convention for the Control of Ships Ballast Water and Sedients (BWM Convention). In response to this, a nuber of technologies have been developed and coercialised by different vendors. Many have their basis in land-based applications for unicipal and industrial water and effluent treatent, and have been adapted to eet the requireents of the BWM Convention and shipboard operation. This publication coprises an outline description of water treatent processes and an appraisal of coercially and developing technologies for ballast water treatent. It is the end deliverable of work conducted by the Centre for Water Science at Cranfield University in conjunction with the consultants Vale Water Services and Whitewater Liited on behalf of Lloyd s Register. The publication contains a suary both of the governing regulation that ultiately akes ballast water treatent andatory (Section 2), and of water treatent technology as it relates to ballast water anageent (Section 3). These sections then provide the background knowledge and context for an assessent of the coercial technologies either currently coercially or projected to be arket-ready by 2009 with reference to their efficacy, technical and econoic viability and testing and status (Section 4). Full data, referenced against individual suppliers, are provided in the Annex. The assistance and patience of the technology suppliers who contributed uch of the inforation published herein are gratefully acknowledged. Lloyd s Register 3

6 2. Regulation Ballast water quality and standards Ballast water when taken onboard will contain a variety of organiss including bacteria and viruses and the adult and larval stages of the any arine and coastal plants and anials. While the vast ajority of such organiss will not survive to the point when the ballast is discharged, soe ay survive and thrive in their new environent. These non-native species, if they becoe established, can have a serious ecological, econoic and public health ipact on the receiving environent. The International Maritie Organization (IMO) has developed international legislation (The International Convention for the Control and Manageent of Ships Ballast Water and Sedients) to regulate discharges of ballast water and reduce the risk of introducing non-native species fro ships ballast water. Regulation D-2 of the Convention sets the standard that the ballast water treatent systes ust eet (Table 1). Ballast water treatent systes ust be tested and approved in accordance with the relevant IMO Guidelines. Organis category Plankton, >50 µ in iniu diension Plankton, µ Toxicogenic Vibrio cholera (O1 and O139) Escherichia coli Intestinal Enterococci * colony foring unit Regulation < 10 cells / 3 < 10 cells / l < 1 cfu* / 100 l < 250 cfu* / 100 l < 100 cfu* / 100 l Table 1 IMO D2 standards for discharged ballast water Ships will be required to treat ballast water in accordance with the tietable shown in Table 2. According to this table, a key ilestone arises in 2009, when ships under construction in or after that date having less than ballast water capacity ust have ballast water treatent to eet the D2 Standard in the Convention. This is likely to apply to around 540 ships estiated to be coencing construction in Ballast capacity Year of ship construction* Before < Ballast water exchange or treatent until 2016 Ballast water treatent only fro 2016 Ballast water treatent only Table 2 Tietable for installation of Ballast Water Treatent systes Ballast water exchange or treatent until 2014 Ballast water treatent only fro 2014 Ballast water treatent only > Ballast water exchange or treatent until 2016 Ballast water treatent only fro 2016 Ballast water exchange or treatent until 2016 Ballast water treatent only fro 2016 Ballast water treatent only Ship Construction refers to a stage of construction where: The keel is laid or construction identifiable with the specific ship begins; or Assebly of the ship has coenced coprising at least 50 tonnes or 1% of the estiated ass of all structural aterial, whichever is less; or The ship undergoes a ajor conversion. Major conversion eans a conversion of a ship: which changes its ballast water carrying capacity by 15 percent or greater or which changes the ship type, or which, in the opinion of the Adinistration, is projected to prolong its life by ten years or ore, or which results in odifications to its ballast water syste other than coponent replaceent-in-kind. Conversion of a ship to eet the provisions in the Convention relating to ballast water exchange ( regulation D- 1 ) does not constitute a ajor conversion in relation to the above requireents. Lloyd s Register 4

7 The processes Technologies developed for ballast water treatent are subject to through specific IMO processes and testing guidelines designed to ensure that such technologies eet the relevant IMO standards (Table 1), are sufficiently robust, have inial adverse environental ipact and are suitable for use in the specific shipboard environent. A copany offering a treatent process ust have the process approved by a Flag Adinistration. In general the anufacturer will use the country in which it is based to achieve this, although this is not a specific requireent and soe copanies ay choose to use the Flag State where the testing facility is based or the Flag State of a partner copany. In general the Flag State will probably choose to use a recognised organisation - such as a classification society - to verify and quality assure the testing data. The testing procedure is outlined in the IMO s Guidelines for Approval of Ballast Water Manageent s 1 (frequently referred to as the G8 guidelines ). The consists of both shore based testing of a odel to confir that the D2 discharge standards are et and ship board testing to confir that the syste works in service. These stages of the are likely to take between six weeks and six onths for the shore based testing and six onths for the ship based testing. s using active substances* Approval of environental ipact of discharged ballast water (GESAMP BWWG) Initial Approval of syste (Flag State) Land based testing Shipboard trials Approval of environental ipact of discharged ballast water (GESAMP BWWG) Final Issue of type certificate (Flag State) Type Approval Certificate Fig 1. Suary of pathway for ballast water treatent systes s not using active substances Land based testing Shipboard trials Type Approval Certificate * Includes cheical disinfectants, e.g. chlorine, ClO 2, ozone Includes techniques not eploying cheicals, e.g. deoxygenation, ultrasound Further requireents apply if the process uses an active substance (AS). An AS is defined by the IMO as a substance or organis, including a virus or a fungus that has a general or specific action on or against harful aquatic organiss and pathogens. For processes eploying an AS, basic fro the GESAMP 2 Ballast Water Working Group (BWWG), a working coittee operating under the auspices of IMO, is required before shipboard testing proceeds. This is to safeguard the environent by ensuring that the use of the AS poses no har to the environent. It also prevents copanies investing heavily in developing systes which use an active substance which is subsequently found to be harful to the environent and is not approved. The GESAMP BWWG assessent is based largely on data provided by the vendor in accordance with the IMO approved Procedure for Approval of Ballast Water Manageent s that ake use of Active Substances 3 (frequently referred to as the G9 Guidelines ). The IMO Marine Environent Protection Coitteee has, on the recoendation of the GESAMP BWWG, granted basic to four technologies as of June Guidelines for of ballast water anageent systes (G8) IMO resolution MEPC125(53) 2 Joint Group of Experts on the Scientific Aspects of Marine Environental Protection. An advisory body established in 1969 which advises the UN syste on the scientific aspects of arine environental protection. 3 Procedure for of ballast water anageent systes that ake use of active substances (G9) IMO resolution MEPC126(53). Lloyd s Register 5

8 Basic Approval is the first step in the process when using an active substance. In ost cases Basic Approval has been granted with caveats and the request for further inforation for the purposes of Final Approval. Basic Approval is thus an in principle of the environental ipact of an active substance, which ay then expedite inward strategic investent or arketing within the supplier s organisation and allow testing of a syste at sea. After Basic Approval for active substances, treatent systes can be tested both on land and onboard ship according to the IMO Guidelines for Approval of Ballast Water Manageent s ( G8 guidelines ). Final Approval by the GESAMP BWWG will take place when all testing is copleted. Once final is granted by GESAMP the Flag Adinistration will issue a Type Approval certificate in accordance with the aforeentioned guidelines. If the process uses no active substances the Flag Adinistration will issue a Type Approval certificate without the need for fro the GESAMP BWWG. Whilst there is a considerable aount of published inforation concerning the efficacy of the coercially or developing ballast water treatent technologies, these data have not all been generated under the sae conditions of operation, scale and feedwater quality. This akes appraisal of the technologies difficult. The IMO G8 Guidelines for Approval of Ballast Water Manageent s are therefore designed to create a level playing field for assessent of technological efficacy. The stipulated testing regie and protocols are prescriptive in nature and costly to undertake. The sea-based test alone requires six onths of testing based on a triplicated trial, with biological analysis to be copleted within six hours of sapling. The land-based testing is based on specific organiss which therefore have to be either indigenous in the water or cultured specifically for the test. The land based and shipboard testing is overseen by the Flag Adinistration or a recognised organisation (generally a classification society). Generally it is taking up to two years fro subitting an application for Basic Approval for an active substance to copletion of testing. As at May 2007, no systes have copleted the final stages of the process under the G8 Guidelines for Approval of Ballast Water Manageent s, although this ay soon change. 6 Lloyd s Register

9 3. Treatent Process Background The technologies used for treating ballast water are generally derived fro unicipal and other industrial applications; however their use is constrained by key factors such as space, cost and efficacy (with respect to the IMO discharged ballast water standards). There are two generic types of process technology used in ballast water treatent: solid-liquid separation and disinfection (Fig. 2). Physical solid-liquid separation Treatent: Hydrocyclone Surface filtration Cheical enhanceent: Coagulation/ Flocculation Cheical treatent: Chlorination Electrochlorination ] or electrolysis Ozonation Peracetic acid SeaKleen Chlorine dioxide OR Physical UV irradiation UV + TiO 2 Deoxygenation Gas injection Ultrasonic treatent Cavitation Disinfection Residual control: Cheical reduction (sulphite/bisulphite) Physical enhancent: Ultrasonic treatent Cavitation Fig. 2 Generic ballast water treatent technology process options Solid-liquid separation is siply the separation of suspended solid aterial, including the larger suspended icro-organiss, fro the ballast water, either by sedientation (allowing the solids to settle out by virtue of their own weight), or by surface filtration (reoval by straining; i.e. by virtue of the pores in the filtering aterial being saller than the size of the particle or organis). Disinfection reoves and/or inactivates icro-organiss using one or ore of the following ethods: cheical inactivation of the icroorganis physicocheical inactivation by irradiation with ultraviolet light, which denatures the D of the icro-organis and therefore prevents it fro reproducing. Ultrasound or cavitation (tered icroagitation for the purposes of this publication) are also physico-cheical disinfection ethods deoxygenation either by displaceent of the dissolved oxygen with an inert gas injection or stripping it by eans of a vacuu and thereby asphyxiating the icro-organis. All of the above disinfection ethods have been applied to ballast water treatent, with different products eploying different unit processes. Most coercial systes coprise two stages of treatent with a solid-liquid separation stage being followed by disinfection (Fig. 2), though soe disinfection technologies are used in isolation. One ballast water treatent technology also eploys cheical enhanceent (ie coagulation/ flocculation) upstrea of solid-liquid separation; another eploys titaniu dioxide (TiO 2 ) to intensify ultraviolet irradiation. Lloyd s Register 7

10 Separation processes As previously stated, the cheical or physicocheical unit processes used for disinfection are usually preceded by physical solid-liquid separation, by either filtration or hydrocyclone technology. The filtration processes used in ballast water treatent systes are generally of the autoatic backwashing type using either discs (Fig 3a) or fixed screens. Since the standards relating to treated ballast water are sizebased, technologies capable of reoving aterials above a specific size are ost appropriate. Reoval of larger organiss such as plankton (Table 1) by filtration requires a filter of equivalent esh size between 10 and 50µ. Such filters are the ost widely used solid-liquid separation process eployed in ballast water treatent, and their effective operation relates ainly to the flow capacity attained at a given operating pressure. Maintaining the flow norally requires that the filter is regularly cleaned, and it is the balance between flow, operating pressure and cleaning frequency that deterines the efficacy of the filtration process. In principle, surface filtration can reove sub icron (i.e. less than 1µ in size) icro-organiss. However, such processes are not viable for ballast water treatent due to the relatively low pereability of the ebrane aterial. Hydrocyclone technology is also used as an alternative to filtration, providing enhanced sedientation by injecting the water at high velocity to ipart a rotational otion which creates a centrifugal force (Fig. 3b) which increases the velocity of the particle relative to the water. The effectiveness of the separation depends upon the difference in density of the particle and the surrounding water, the particle size, the speed of rotation and residence tie. Since both hydrocylcones and filters are ore effective for larger particles, pre-treatent with coagulants to aggregate (or flocculate ) the particles ay be used upstrea of these processes to increase their efficacy. However, because flocculation is tie dependent, the required residence tie for the process to be effective deands a relatively large tank. The processes can be advanced, however, by dosing with an ancillary powder of high density (such agnetite or sand) along with the coagulant to generate flocs which settle ore rapidly. This is soeties referred to as ballasted flocculation, and is used in soe unicipal water treatent installations where space is at a preiu and has been used in one of the systes included in this publication. Pressure housing Stacked discs spaced at µ Vortex finder Inlet Clean water overflow Typical trajectory of light particle Fig. 3 (a) Filtration, and (b) Hydrocyclone processes Typical path of a larger heavier particle Dirty water in Clean water out (a) Underflow containing solids (b) 8 Lloyd s Register

11 Disinfection Cheical disinfection A nuber of different cheicals or cheical processes have been eployed in the ballast water treatent systes reviewed including: Chlorination Electrochlorination Ozonation Chlorine dioxide Peracetic acid Hydrogen peroxide Menadione/Vitain K The efficacy of these processes varies according to the conditions of the water such as ph, teperature and, ost significantly, the type of organis. Chlorine, whilst relatively inexpensive is virtually ineffective against cysts unless concentrations of at least 2 g/l are used. Chlorine also leads to undesirable chlorinated byproducts, particularly chlorinated hydrocarbons and trihaloethanes. Ozone yields far fewer harful byproducts, the ost proinent being broate, but requires relatively coplex equipent to both produce and dissolve it into the water. Chlorine dioxide is norally produced in situ, although this presents a hazard since the reagents used are theselves cheically hazardous. Peracetic acid and hydrogen peroxide (provided as a blend of the two cheicals in the for of the proprietary product Peraclean) are infinitely soluble in water, produce few harful byproducts and are relatively stable as Peraclean. However this reagent is relatively expensive, is dosed at quite high levels and requires considerable storage facilities. For all these cheicals pre-treatent of the water with upstrea solid-liquid separation is desirable to reduce the deand on the cheical, because the cheical can also react with organic and other aterials in the ballast water. Post-treatent to reove any residual cheical disinfectant, specifically chlorine, prior to discharge using a cheical reducing agent (sodiu sulphite or bisulphite) ay be appropriate if high concentrations of the disinfectant persist. In potable water treatent this technique is routinely eployed. When used in ballast water treatent, dosing to around 2 g/l of chlorine can take place, leaving a chlorine residual in the ballast water tanks to achieve disinfection. The chlorine level is then reduced to zero ( quenching the chlorine copletely) prior to discharge. This technique is used in at least two of the ballast water treatent systes currently reviewed. Menadione, or Vitain K, is unusual in that it is a natural product (although produced synthetically for bulk coercial use) and is relatively safe to handle. It is awaiting US FDA and EPA which should be granted in It is arketed for use in ballast water treatent under the proprietary nae Seakleen by Hyde Marine. As with other disinfectant cheicals, it is not without a history of application elsewhere and has been used in catfish faring where it is liberally spread into water. Over three tonnes of enadione are used annually for this application alone. Physical disinfection Of the physical disinfection options ultraviolet irradiation (UV) is the ost well established and is used extensively in unicipal and industrial water treatent applications. The process eploys aalga laps surrounded by a quartz sleeve (Fig.3) which can provide UV light at different wavelengths and intensities, depending on the particular application. It is well known to be effective against a wide range of icroroganiss, including viruses and cysts, but relies on good UV transission through the water and hence needs clear water and unfouled clean quartz sleeves to be effective. Lloyd s Register 9

12 The reoval of water turbidity (i.e. cloudiness) is therefore essential for effective operation of the syste. UV can be enhanced by cobining with another reagent, such as ozone, hydrogen peroxide or titaniu dioxide which will provide greater oxidative power than either UV or the suppleentary cheical reagent alone. The reaining physical disinfection processes do not inherently require use of pre-treatent. However, the efficacy of both processes is subject to liitations. Deoxygenation takes a nuber of days to coe into effect due to the length of tie it takes the organiss to be asphyxiated. However, ost voyages will exceed this tie period so this should not be a significant constraint. Cavitation or ultrasonic treatent processes both act at the surface of the icro-organis and disrupt the cell wall through the collapse of icrobubbles. These processes are currently not as well understood as the other ore established disinfection technologies and soe systes use these techniques with cheical disinfection to provide the necessary biocidal efficacy. Outlet Fig. 4 UV tube and syste Wiper Inlet Lap Quartz sleeve Ballast water treatent unit processes Taking account of all of the technologies surveyed, there are currently 16 individual unit processes used in ballast water treatent (Table 3). The coercial systes differ ainly in the choice of disinfection technology and the overall syste configuration (i.e. the coupling of the disinfection part with solid liquid separation, where the latter is used). Alost all have their basis in land-based systes eployed for unicipal and industrial water and wastewater and thus can be expected to be effective for the duty of ballast water, albeit subject to constraints in the precise design arising fro space and cost liitations. 10 Lloyd s Register

13 Table 3 Coercial technologies by generic unit operation type ID Manufacturer Active substance Solid-liquid sepn Cheical disinfection and dechlorination Physical Microagitation disinfection HC Filt None Coag O 3 Cl EL/EC PAA ClO 2 SK Red UV Deox Cav US AOP 1 Alfa Laval Tuba AB 2 Ecochlor Inc 3 Electrichlor Inc 4 RWO Marine 5 Environental Technologies Inc 6 Gauss 7 Greenship 8 Haann AG 9 Hitachi 10 Hyde Marine Inc 11 Mitsui Engineering & Shipbuilding 12 JFE Engineering Corporation 13 Marenco Technology Group Inc 14 Oceansaver AS 15 MH s Inc 16 Mitsubishi Heavy Industries 17 NEI Treatent s LLC 18 Nutech Optiarin AS 20 Qwater 21 Resource Ballast Technology 22 Severn Trent De Nora 23 Techcross 24 ATG Willand 25 Sincerus Water Treatent 1 26 DNV Maritie Solutions TiO Did not return copleted survey fors; 2 Physical pre-treatent options under consideration; 3 Two disinfection options offered 4 Technology ay be considered as using active substancies HC Hydrocyclone Filt Filtration Coag Coagulant (with agnetic particles); UV Ultraviolet irradiation Deox Deoxygenation O 3 Ozonation Cl Chlorination; EL/EC Electrolysis/electrochlorination PAA Peracetic acid (as Peraclean) ClO 2 Chlorine dioxide SK Seakleen Red Reduction Cav Cavitation US Ultrasound AOP Advanced oxidation Lloyd s Register 11

14 4. Treatent technologies and suppliers Suppliers This publication considers only suppliers of coplete systes for ballast water treatent rather than suppliers of unit operations, although individual proprietary unit operations (e.g. filters, electrochlorination devices, disinfectant cheicals and UV sterilisers) ay be included as part of the systes reviewed. Very any copanies ay consider their product(s) to be appropriate for ballast water treatent but it is not necessarily the case that shipboard applications exist. Moreover, a few copanies appear to have withdrawn fro this arket altogether, following an initial assessent of the technology and arket potential. These have been excluded fro the survey. Whilst basic technical inforation is fro 26 copanies, only 24 of these took part in the survey, and it is inforation fro these 24 which fors the basis of this publication. Of these 24, nine are part of a ulti-billion dollar turnover international group of copanies with significant activity in arine and/or engineering areas, whilst 14 appear to be SMEs (sall to ediu enterprises, generally defined as having less than 250 ) all of which have been set up within the past 13 years and seven of which are no ore than seven years old. The reaining one is a governental organisation. Eight different countries are represented by these 24 copanies, with the predoinant nation being the US (Fig. 5). Soe copanies have collaborated with unit process suppliers to produce a syste. For exaple, Alfa Laval use the Wallenius UV-based advanced oxidation process; RWO Marine (part of the ulti-billion dollar turnover Veolia group of copanies) originally partnered Perascand for its Ectosys syste (and have since acquired the Ectosys product), the Mitsui Engineering syste was originally developed by a nuber of partners led by the Japanese Association of Marine Safety; several copanies eploy proprietary filters. A nuber of the saller suppliers, whilst having few in the core copany, have partners, priarily licensees/distributors providing a global network to sell the technology. Korea 2 Norway 3 Other 3 US 10 Fig. 5 Technology suppliers by country: Other coprise the Netherlands, South Africa and the UK Gerany 4 Japan 4 Technologies Categories Treatent technologies offered by the 26 suppliers are suarised in Table 3; since one supplier Hyde Marine offers two technologies, there are 27 technologies in total. All these systes have undergone preliinary pilot trials. The published data fro these trials has shown the systes to be generally effective with reference to the IMO treated water standards applicable to discharged ballast water shown in Table 1. However, these data are based on test protocols which largely do not coply with those stipulated under the IMO G8 Approval of Ballast Water Manageent s and such is still required. All of the 27 products for which inforation is, other than those based on gas injection, are either odular or can be ade so. Ten of the technologies do not eploy cheicals for the ain flow and disinfection and ay be considered to have no active substance. However there is ongoing debate as to whether free hydroxyl radicals produced either by electrolysis or advanced oxidation will be considered to be active substances by the GESAMP BWWG. 12 Lloyd s Register

15 Of the 27 technologies considered 17 eploy upstrea filtration for solid-liquid separation (Fig. 6a), with the filter pore size priarily in µ range. Only one supplier (Hitachi) eploys a sub-10 µ rated filter, which would be expected to provide good disinfection capability against the larger icro-organiss and reduce the loading on the downstrea processes. However, filters of saller pore ratings deand coensurately higher operating pressures, and thus higher energy deands, and also ore rigorous cleaning. One supplier (Marenco) uses cartridge filters which are not backwashable. Four suppliers eploy hydrocyclones as pre-treatent and the reainder use no solid-liquid separation at all. One supplier (Haann) uses both hydrocyclones and filtration. Only one syste (Hitachi) eploys pre-coagulation upstrea of the filter. This particular syste eploys agnetic particles to accelerate the clarification process ( ballasted flocculation ). All solid-liquid separation processes produce a waste strea containing the suspended solids. This waste strea coprises the backwash water fro filtering operations or the underflow fro the hydrocyclone separation. These waste streas require appropriate anageent. During ballasting they can be safely discharged at the point where they were taken up. On deballasting, the solid-liquid separation operation is generally by-passed. Whilst there are a total of eight disinfection processes used for ballast water treatent, the ajority of the systes provided (15 of the 27 reviewed) are based on either electrolytic treatent (electrolysis or electrochlorination) or UV irradiation (Fig. 6b). In one case (the Wallenius process eployed by Alfa Laval), the UV irradiation is suppleented with titaniu dioxide (TiO 2 ) to intensify the oxidative power of the UV light (Section 3.3). Electrolytic treatent is used the ost extensively, with products having different design features but all essentially eploying a direct current to electrolyse the water. Electrolytic technologies provided for ballast water treatent ay be designed to generate either chlorine, as in the classic electrochlorination process, or other oxidative products. Those designed for chlorine generation rely on the salinity of the feedwater for effective chlorine generation; suppleentary brine is necessary when the abstracted ballast water is fresh. This is not an issue for classical chlorine dosing using hypochlorite solution, of which there is only a single exaple (JFE). There are only single exaples of the use of SeaKleen or chlorine dioxide for disinfection. This ay be due to licensing constraints or patenting issues. Ten of the technologies do not eploy cheicals for disinfection and ay be considered to have no active substance. However there is a debate as to whether free hydroxyl radicals produced either by electrolysis (such as in the RWO Marine process) or advanced oxidation will be considered to be active substances by the GESAMP BWWG. Debate on this issue is on-going at the tie of publication of this Publication. Of these ten technologies which do not use cheicals, six eploy UV disinfection, one uses ultrasound with prefiltration and three gas injection for deoxygenation. None 5 Filtration 19 (a) Hydrocyclone 4 Chlorine dioxide, 1 Peracetic acid (Peraclean) 2 Electrolysis/ electrochlorination 8 Chlorination 1 Seakleen 1 Ozonation 3 (b) Coagulant, 1 Ultraviolet irradiation 7 Deoxygenation 3 Ultrasound 1 Fig. 6 Suary of treatent technology options for (a) physical pre-treatent, and (b) disinfection Note: One supplier eploys both hydrocyclones and filtration. Lloyd s Register 13

16 Process configurations With regard to overall process configuration, that is the way in which the unit operations are cobined to produce a treatent technology (Fig 7), over half of the products surveyed (14 out of 27) are either filtration followed by UV (six), filtration followed by electrolytic treatent (five) or hydrocyclone separation followed by either electrolytic treatent or UV (three). The use of filtration upstrea of a UV process is essential for reoving solid particles, and in particular fine particles which inhibit the effective operation of the UV due to their ipact on light transission (Section 3). The UV tubes appear to be ainly ediu pressure type as is appropriate for the syste capacities of these shipboard applications. Other 13 Filtration + electroclysis 5 Filtration + UV 6 Hydrocyclone + electrochlorination or UV 3 Fig. 7 Treatent process configurations: cobinations of filtration or hydrocyclone followed by electro-chlorination/ electrolysis or UV irradiation As with UV disinfection, upstrea solid-liquid separation ahead of cheical treatent is required ainly to reduce the solids loading on the disinfection process. This reduces the risk of fouling and clogging of the disinfection unit, but also reduces the load on the cheical; the cheical is thus used priarily for disinfection rather than cheically oxidising other essentially innocuous dissolved atter in the ballast water. However there are three products (Ecochlor, Nutech O3 and Techcross) where no pre-treatent is used ahead of cheical disinfection, two of which relate to non-chlorine based cheical disinfection (ozonation and chlorine dioxide). The ipact of the disinfectant cheical is, in four cases (ETI, JFE, Mitsui Engineering and Resource Ballast Tecnology), enhanced by the use of icro-agitation, i.e. cavitation or ultrasonic treatent. As already stated (Section 3), the use of icro-agitation for disinfection - either in cobination with a classical cheical disinfectant or, ore unusually, in isolation - is not widespread. Existing exaples for water and wastewater treatent appear to be liited to cooling tower waters. There is one exaple (Qwater) of ultrasonic treatent being used in the absence of a disinfection cheical. Three technologies (Hyde arine, Haan and DNV Maritie Solutions) use a proprietary biocide cheical such as Peraclean or Seakleen. The reaining three technologies (NEI Treatent s, Oceansaver and MH s) are based on deoxygenation using injection of an inert gas (nitrogen), in soe cases cobined with carbon dioxide to displace the dissolved oxygen. This ethod is particularly attractive when a supply of nitrogen gas exists on board, as when transporting flaable or explosive aterials such as liquid fuel. Of the three exaples, two eploy cavitation presuably to enhance the solubility of the gas in water. These systes are not odular but are scaled according to the desired capacity. Eleven of the technologies treat the ballast water both during ballasting and discharge (Table 5 ), effectively treating the ballast water twice. If filtration is used with backwashable filters then the filters are by-passed during discharge to avoid discharging non native organiss and other aterial into the receiving water. Nine of the technologies treat only during ballasting and two only during discharge. A further two copanies treat during ballasting and during the voyage. 14 Lloyd s Register

17 Cost and footprint The key technical features of the syste with respect to ballast water treatent are the flow capacity, footprint, overall size of the syste and costs, the latter coprising capital expenditure (capex) and operating expenditure (opex). Most of the technologies have been developed for a flow rate of about /hr, considered to be the flow rate required for the first phase of ships required to be equipped with ballast water treatent technology. Since the systes are largely odular in design (other than the gas injection type), there is no technical liit to the upper flow rate other than that iposed by size and/or cost. In soe cases there are exaples of systes already for flows above /hr. The ean key data for costs and footprint for all the technologies are suarised in Table 4 and Figures 8 and 9. Full data are provided in Table 5. The ean quoted estiated or projected operating cost of the systes, on the basis of the 13 sets of data provided is $0.1 per 3, within a broad range of values fro $0.01 up to $0.35 per 3 treated water. Nine of the 13 suppliers who provided operational expenditure inforation quoted costs below $0.08 per 3 ; the ean value is skewed by two anoolously high values. There is no correlation between the quoted opex and the type of syste. In ost cases (except for the few technologies that use stored cheicals and the gas injection units that use fossil fuel) the ajority of the opex relates to the power required to operate the process (UV irradiation, electrolysis or ozonation). 11 < $20 4 $ Fig. 8 Estiated plant operating cost per of treated water: 24 products; inforation not or not provided for 11 systes > $ < Fig. 9 Estiated footprint of a /h plant: 24 products; inforation not or not provided for six systes. > Mean Data points Min Max * flow rate Height Capex $ /h* /h* /h* /h* $/ Table 4 Suary of plant footprint, and capital and operating expenditure Lloyd s Register 15

18 Capital cost inforation is not widely, and the ajority of the suppliers regard this inforation as confidential. Fro the 11 sets of data provided, the capital cost of a /h plant ranges fro $135k - $650k, with a ean value of around $340k. For a /h plant, the equivalent values are 250k k with a ean of 800k. As with the opex, fro the liited inforation provided there appears to be no correlation between the quoted capex and the configuration of the process, and variations in prices arise fro differences in assuptions ade by the various suppliers regarding inclusion or exclusion of specific coponents. Prices quoted ust be regarded as tentative since soe of these products are still under developent and the price is to soe extent deterined by the arketplace. The footprint of the systes reviewed varies between 1.6 and 25 2 for a /h unit, with a ean value of 5 2, according to the data provided by suppliers in relation to 17 systes. For a unit of ten ties this flow capacity, there is less inforation, since soe suppliers do not provide units of this size, and the iniu, axiu and ean values are 4, 100 and 25 2 respectively. Thus, whilst the units ay be predoinantly odular, this does not iply that the footprint increases proportionately with flow capacity. The quoted, or calculated fro total volue space occupied, is between 1.0 and 3.0 for the technologies for which inforation was provided. Table 5 key data: capacity, footprint and costs Manufacturer Alfa Laval Tuba AB Ecochlor Inc Electrichlor Inc Environental Technologies Inc Gauss Greenship Haann AG Hitachi Hyde Marine Inc JFE Engineering Corporation Marenco Technology Group Inc MH s Inc Mitsubishi Heavy Industries Mitsui Engineerg. & Shipbuildg. NEI Treatent s LLC Nutech 03 Oceansaver AS Optiarin AS Qwater Resource Ballast Technology RWO Marine Severn Trent De Nora Techcross ATG Willand Treatent protocol A + B+ D A A + B+ D B * A + B + D A A A + B + D A + B + D B A + C A A A A A + B + D A + B + D A + B + D A A + B + D A+B+C+D A A + B + D * 000 s 3 /h 5 >10 >10 >10 >10 2 >10 > >10 >10 >10 >10 >10 >10 >10 >10 >10 >10 Estiated Footprint variable variable A ballasting, B discharging, C during voyage, D bypass filter on deballasting * treatent flow currently (>10 3 /h indicates no stated axiu) Estiated Capex $ 000 ( cost) * Estiated $/ Lloyd s Register

19 Other syste characteristics Other technical features of the products are not necessarily coon to all of the and are specific to generic types of process technology. These process-specific facets can be suarised as follows: Deoxygenation is the only technology specifically developed for ballast water treatent and is effective because the de-aerated water is stored in sealed ballast tanks. However the process takes between one and four days to take effect, and thus represents the only type of technology where voyage length is a factor in process efficacy. This type of technology is also the only one where, technically, a decrease in corrosion propensity would be expected (and, according to one supplier, has been recorded as being suppressed by 50-85%), since oxygen is a key coponent in the corrosion process. The water is reaerated on discharge. s in which cheicals are added norally need to be neutralised prior to discharge to avoid environental daage in the area of discharge. Most ozone and chlorine systes are neutralised but soe are not. Chlorine dioxide has a half life in the region of 6-12 hours, according to the supplier, but at the concentrations at which it is eployed it can be safely discharged after a axiu of 24 hours. Essentially ost UV systes operate using the sae type of ediu pressure aalga laps. A critical aspect of UV effectiveness is the applied UV dose/power of the lap. This inforation has not been given by all suppliers. Another aspect of UV effectiveness is the clarity of the water. In waters with a high turbidity or colloidal content, UV would not be expected to be as effective. Most chlorination systes are applying a dose in the region of 2 g/l residual chlorine which has proven to be effective. Most ozonation suppliers are using an ozone dose of 1-2 g/l which has proven to be effective. UV systes are the least coplex treatent plants to operate. Electrolysis and electrochlorination plants are the ost coplex. Deoxygenation plants are relatively siple devices if an inert gas generator is already on the ship and in the latter case would take up little additional space. Cheical dosing systes such as Peraclean, SeaKleen and chlorine dioxide have low capital costs because only a dosing pup is required but require cheical storage facilities and availability of cheicals in the ports vid. The biggest operating cost for ost systes is power and for large power consuers (electrolytic and advanaced oxidation processes) availability of shipboard power will be a factor. For cheical dosing systes, power is very low and cheical costs are the ajor factor. For these reasons cheical addition ay be better suited to sall ballast capacities. Although the systes operate at generally low pressure and thus do not require additional ballast water puping pressure, those eploying venturi devices (for exerting shear) incur pressure losses of up to 2 bar. For ost systes it is recoended that installation takes place in the engine/achine roo near the existing ballast water pups, although installation on deck ay also be possible if appropriate precautions are taken. If the location is in an explosion zone, then the installation will need explosion proofing. Soe of the technologies can be provided as explosion-proof products, but there is a cost penalty for this. The generation of hydrogen by the electrolytic technologies is not considered an issue, since the gas is vented and diluted with air to safe levels. Lloyd s Register 17

20 Whilst disinfection by-products are an issue, and central to the of ballast water anageent systes that ake use of active substances, suppliers are confident that the levels generated are unlikely to be probleatic. There is a large aount of scientific and technical inforation on disinfection by-products foration that is likely to support this. Coercial availability Ten of the 24 copanies participating in this review have already units on board ship (Table 6). As of May 2007, a total of 28 ballast water treatent units have been. Of these, over half have been by two copanies (Optiarin and Hyde Marine) and are UV based systes. It is very likely that the underlying reasons for UV based technology being adopted lie with the ready availability of coercial UV units in the water disinfection area, relatively low capital cost, very sall footprint and ease of installation. However, the fact that copanies have units and have coercially units does not ean that the systes have the necessary s in accordance with the IMO guidelines. Approval status which is fundaental to coercial acceptance in the future is discussed below. Table 6 status: coercial developent and Manufacturer Alfa Laval Tuba AB Ecochlor Inc Electrichlor Inc Environental Technologies Inc Gauss Greenship Haann AG Hitachi Hyde Marine Inc JFE Engineering Corporation Marenco Technology Group Inc MH s Inc 4 Mitsubishi Heavy Industries Mitsui Engineerg. & Shipbuildg. NEI Treatent s LLC 4 Nutech 03 Oceansaver AS 4 Optiarin AS Qwater Resource Ballast Technology RWO Marine Severn Trent De Nora Techcross ATG Willand (if applicable) Basic final Apr-06 Aug-07 - Jul-07 Dec-05 Mar-08 - Mar-08 - Dec-06 May-06 5 Feb Oct-06 Dec-07 - Dec Oct-07 Mar-09 - Jul-09 Aug Mar Jun-08 - Apr-08 Dec-08 Dec-07 Jun Jul-07 Jun-07 Mar-09 Jul Dec Dec-07 Dec Dec-07 Dec-08 Aug-07 Apr-07 Jun Jul-07 Jun-07 Jul-08 Dec Dec Oct Jul-07 Jan-07 Aug-07 NIVA ND ND ND ND NIOZ NIOZ ND ND ND MLML SIO ND JAMS NOAA KORDI NIVA ND SAMSA/AISA MWB USNRL KORDI ND /y Year coercialised or anticipated for coercialisation for ballast water treatent; 2 refers to existing installations; 3 Figures for 2009, capacity projected to increase in following years; 4 IMO G8 ballast water anageent systes protocol does not lend itself to testing of this type of technology; 5 tests coparable to IMO G8 ballast water anageent systes testing protocol stated to have been copleted prior to introduction of G8 protocol. KORDI Korean Ocean Research and Developent Institute; JAMS Japan Association of Marine Safety; NIVA Norwegian Institute for Water Research; USNRL US Naval Research Laboratory; NOAA US National Oceanic and Atospheric Adinistration; SAMSA South African Departent of Transport; AISA Agricultural Institute of South Africa; SIO Scripps Institution of Oceanography; MLML Moss Landing Marine Laboratories; MWB Motorenwerke Breerhaven. 18 Lloyd s Register

21 Approval status The regulatory fraework requires that a key distinction is to be ade between those systes eploying active substances (priarily disinfectant cheicals) and those which do not. Non-AS systes (of which there are approxiately 9-12 out of the 27 reviewed) ay reach the arketplace earlier because subission to the GESAMP BWWG for is not required. This perceived reduced environental ipact and lack of requireent for GESAMP ay favour non-as systes. However, systes using active substances ay be expected to perfor ore reliably. Notwithstanding this observation, preliinary tests conducted by all the suppliers appear to be positive with respect to disinfection capability. According to inforation provided by the suppliers, all the technologies reviewed are progressing towards, though the scheduling of the testing differs between the different suppliers and thus the projected date for final. To date only four of the active substance systes have received basic fro GESAMP. It is not clear, however, how any other systes are undergoing G8 ballast water anageent systes without having received basic fro GESAMP. Approxiately eight copanies have undertaken or are scheduled to undertake ballast water anageent systes testing at test facilities in Several US copanies who have been operating in this field for over 5 years have carried out extensive testing to satisfy local state legislation. It reains unclear as to whether data fro soe of the above tests, perfored before the IMO regulation was put in place, will be approved - but it appears unlikely. One of the reasons behind the very specific nature of the G8 ballast water anageent systes testing protocol is to overcoe soe of the variability of these early tests and put systes testing on a level playing field. However, the gas injection/deoxygenation technologies do not lend theselves to testing under the strict protocol given in G8, since these rely on sealed tanks and operation over a nuber of days. They are none-the-less effective for journeys in excess of two days. Lloyd s Register 19

22 5. Suary Most of the products included in this publication are either currently coercially or are to be coercialised in Alost all should be coercially by 2009, according to schedules provided and anufacturing capability detailed by the technology suppliers. However, these estiates ay be optiistic and will be reviewed in the future in an update of this publication. There is a great deal of evidence supporting the efficacy of ost of the technologies with respect to the ballast water quality standards, though this is subject to testing under the stipulated IMO conditions. Having said this, and despite the encouraging data fro pilot-scale tests systes provided by the suppliers for these technologies, the efficacy of technologies not eploying active substances is likely to be ore variable than any of those which do, and are also ore likely to be ipacted by ballast water quality (such as turbidity in the case of UV treatent). Coercial technologies eploying generically the sae unit process can be expected to give a siilar perforance and be subject to siilar constraints. Most technologies have their basis in known water and wastewater treatent unit operations used in the unicipal and other industrial sectors, the exception being deoxygenation which is specific to ballast water treatent. Half of the ballast water treatent technology processes reviewed are based on a cobination of either filtration or hydrocyclone separation followed by either electrolytic treatent or ultraviolet irradiation. There is strong coitent towards fro ost copanies. Most copanies have indicated that G8 and if necessary G9 ballast water anageent systes testing will be carried out in 2007 or 2008, though only seven to nine copanies have scheduled or copleted the tests required for IMO. Only one or two of the 24 copanies or organisations surveyed are not in a position to coercialise their products. 20 Lloyd s Register

23 Annex Listing by supplier Lloyd s Register 21

24 Supplier Alfa Laval Tuba AB Partner(s) Wallenius Process Pureballast: Filtration + Ultraviolet/TiO 2 Country Norway used Ballasting + discharging Web fored (if applicable) /y 04/ / / /2007 NIVA * Coents *According to an evaluation of potential growth to 2016, anufacturing not seen as a liiting factor Supplier Ecochlor Inc. Process ClO 2 Country US used Ballasting Web fored > (if applicable) Other regs. /y 08/ / / /2008 ND STEP * Coents *Manufacturing capability in US and China Supplier Electrichlor Hypochlorite Generators Inc. Process Filtration + electrolysis/electrochlorination Country US used Ballasting + discharging Web fored > (if applicable) /y ND Supplier Environental Technologies Inc Process Filtration + ultrasound Country US used Discharging Web fored > (if applicable) /y ND Lloyd s Register