This article was downloaded by: [For Sea Research], [Louis Peperzak] On: 05 September 2014, At: 07:26 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Environmental Technology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tent20 Assessment of didecyldimethylammonium chloride as a ballast water treatment method Cees van Slooten a , Louis Peperzak a & Anita G.J. Buma b a Department of Biological Oceanography, NIOZ, Royal Netherlands Institute for Sea Research, Landsdiep 4, 1797 SZ Den Hoorn (Texel), The Netherlands b Biology, Life Sciences & Technology, Faculty of Mathematics and Natural Sciences, University of Groningen, Linnaeusborg, Building U, Nijenborgh 7, 9747 AG Groningen, The Netherlands Accepted author version posted online: 07 Aug 2014.Published online: 03 Sep 2014. To cite this article: Cees van Slooten, Louis Peperzak & Anita G.J. Buma (2014): Assessment of didecyldimethylammonium chloride as a ballast water treatment method, Environmental Technology, DOI: 10.1080/09593330.2014.951401 To link to this article: http://dx.doi.org/10.1080/09593330.2014.951401 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Transcript
This article was downloaded by: [For Sea Research], [Louis Peperzak]On: 05 September 2014, At: 07:26Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK
Environmental TechnologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tent20
Assessment of didecyldimethylammonium chloride as aballast water treatment methodCees van Slootena, Louis Peperzaka & Anita G.J. Bumab
a Department of Biological Oceanography, NIOZ, Royal Netherlands Institute for SeaResearch, Landsdiep 4, 1797 SZ Den Hoorn (Texel), The Netherlandsb Biology, Life Sciences & Technology, Faculty of Mathematics and Natural Sciences,University of Groningen, Linnaeusborg, Building U, Nijenborgh 7, 9747 AG Groningen, TheNetherlandsAccepted author version posted online: 07 Aug 2014.Published online: 03 Sep 2014.
To cite this article: Cees van Slooten, Louis Peperzak & Anita G.J. Buma (2014): Assessment of didecyldimethylammoniumchloride as a ballast water treatment method, Environmental Technology, DOI: 10.1080/09593330.2014.951401
To link to this article: http://dx.doi.org/10.1080/09593330.2014.951401
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
Assessment of didecyldimethylammonium chloride as a ballast water treatment method
Cees van Slootena∗, Louis Peperzaka and Anita G.J. Bumab
aDepartment of Biological Oceanography, NIOZ, Royal Netherlands Institute for Sea Research, Landsdiep 4, 1797 SZ Den Hoorn(Texel), The Netherlands; bBiology, Life Sciences & Technology, Faculty of Mathematics and Natural Sciences, University of
Groningen, Linnaeusborg, Building U, Nijenborgh 7, 9747 AG Groningen, The Netherlands
(Received 7 March 2014; final version received 30 July 2014 )
Ballast water-mediated transfer of aquatic invasive species is considered a major threat to marine biodiversity, marine indus-try and human health. A ballast water treatment is needed to comply with International Maritime Organization (IMO) ballastwater discharge regulations. Didecyldimethylammonium chloride (DDAC) was tested for its applicability as a ballast watertreatment method. The treatment of the marine phytoplankton species Tetraselmis suecica, Isochrysis galbana and Chaeto-ceros calcitrans showed that at 2.5 μL L−1 DDAC was able to inactivate photosystem II (PSII) efficiency and disintegratethe cells after 5 days of dark incubation. The treatment of natural marine plankton communities with 2.5 μL L−1 DDAC didnot sufficiently decrease zooplankton abundance to comply with the IMO D-2 standard. Bivalve larvae showed the high-est resistance to DDAC. PSII efficiency was inactivated within 5 days but phytoplankton cells remained intact. Regrowthoccurred within 2 days of incubation in the light. However, untreated phytoplankton exposed to residual DDAC showeddelayed cell growth and reduced PSII efficiency, indicating residual DDAC toxicity. Natural marine plankton communitiestreated with 5 μL L−1 DDAC showed sufficient disinfection of zooplankton and inactivation of PSII efficiency. Phytoplank-ton regrowth was not detected after 9 days of light incubation. Bacteria were initially reduced due to the DDAC treatmentbut regrowth was observed within 5 days of dark incubation. Residual DDAC remained too high after 5 days to be safelydischarged. Two neutralization cycles of 50 mg L−1 bentonite were needed to inactivate residual DDAC upon discharge.The inactivation of residual DDAC may seriously hamper the practical use of DDAC as a ballast water disinfectant.
Keywords: DDAC; ballast water treatment; IMO D-2 standard; zooplankton; phytoplankton
1. IntroductionThe ongoing spread of aquatic invasive species throughballast water poses major risks to global biodiversityand may negatively impact marine industries and humanhealth.[1] Ballast water transport through shipping is con-sidered a major vector in the spread of aquatic invasivespecies.[2,3] To halt this spread, the International MaritimeOrganization (IMO) adopted the international conventionfor the control and management of ship’s ballast water andsediments.[4] The convention limits the maximum num-ber of viable organisms allowed to be discharged throughballast water. These requirements are known as the D-2Ballast Water Performance Standard (D-2 standard).[4] Inorder to comply with the D-2 standard ship owners willhave to install ballast water treatment systems (BWTS)aboard their ships to disinfect the ballast water prior todischarge. In recent years, many companies have devel-oped on-board treatment systems capable of disinfectingballast water.[5] In order for a BWTS to be appreciated as aviable and effective system, the IMO has developed elabo-rate testing procedures.[6–8] One of the major phases in theapproval process is the execution of a full-scale land-basedverification test. Prior to full-scale verification testing,
pilot experiments are usually performed. Here, the assess-ment of didecyldimethylammonium chloride (DDAC) as apotential ballast water treatment option is presented.
DDAC is a quaternary ammonium compound which iscommonly used as a general disinfectant for a wide rangeof applications. It was registered by the US environmen-tal protection agency in 1962. DDAC is used to disinfectsurface areas such as household, agricultural and medicalequipment. DDAC is also used as disinfectant in swim-ming pools and as anti-sapstain agent in the wood industry.It is estimated that a total of 396 commercial productscontain DDAC as an active ingredient.[9]
DDAC is a molecule with a positively charged cationichead side (ammonium) and a hydrophobic carbon tail side.Disinfection is achieved by binding of the hydrophobictail into the lipid bilayer of cell membranes while thecationic head sticks out into the water phase. The bindingcauses rearrangement of the lipid bilayer which disruptsthe cell membrane and leads to leakage of cell content andeventually cell death.[10]
To test the efficacy of DDAC in the ballast watertreatment, results of experiments were checked for com-pliance with the D-2 standard for organisms > 50 μm
(limit: < 10 viable organisms m−3), commonly referredto as zooplankton and 10–50 μm phytoplankton (limit:< 10 viable organisms mL−1). In addition to zooplanktonand phytoplankton abundance, also photosystem II (PSII)efficiency, bacterial abundance and DDAC concentrationswere monitored.
Experiments included laboratory-scale trials usingthree phytoplankton monocultures to determine the appro-priate DDAC dose needed for disinfection of marinephytoplankton. During these experiments, phytoplanktonconcentration and fitness was followed during exposureto a range of DDAC concentrations. Three cubic metrevessel trials (cube trials) were performed, using naturalseawater derived from a harbour adjacent to the institute.The cube trials were intended first of all to determine theDDAC dose needed for sufficient zooplankton and phyto-plankton disinfection and then to test for any detrimentaleffects of high total suspended solids (TSS) loads on theefficacy of DDAC. The first and third cube trials were fol-lowed by a laboratory-scale regrowth experiment to test thepotential for phytoplankton regrowth after treatment. Var-ious BWTS do not physically disrupt cells immediately,so the potential for regrowth can be a decisive factor inthe efficacy assessment of a potential BWTS.[11] Finally,a full–scale 100 m3 tank trial was conducted using naturalseawater to test the neutralization system that was neededto render the residual DDAC harmless upon discharge.
2. Materials and methods2.1. Experimental design2.1.1. Laboratory trialThree phytoplankton species, obtained from the NationalCentre for Marine Algae and Microbiota, were selectedfor the laboratory trial. The prasinophyte Tetraselmis sue-cica (CCMP 904), the prymnesiophyte Isochrysis galbana(CCMP 1323) and the diatom Chaetoceros calcitrans(CCMP 1315) were cultured in 500 mL polyethylene bot-tles (Nalgene) in a mix of 1:1 F/2 medium [12] andenriched artificial seawater medium [13] at 15°C and a 16:8light:dark cycle of 50 μmol photons m−2 s−1 light inten-sity. After reaching exponential growth phase, the cultureswere treated with 0, 2.5, 5, 7.5 and 10 μL L−1 DDAC usingan 8% DDAC working stock solution made by dissolving1 mL Bardac
®2280, containing 80% DDAC (Lonza Inc.),
in 9 mL milli-Q. Treated and control cultures were incu-bated at 15°C in the dark. Samples of 50 mL were taken 24h before and 1 h after the DDAC addition and subsequentlyfor 5 days to monitor the following variables: phytoplank-ton abundance, PSII efficiency and DDAC concentration.
2.1.1.1. Cube trial 1. On 21 July 2010, natural seawa-ter was pumped from the saltwater harbour adjacent to theinstitute at high tide. A 300 m3 h−1 centrifugal pump wasused to take up water through a pipeline normally used for
filling 300 m3 subterranean tanks. A bleeding valve wasused to divert a side-stream to three opaque black polyethy-lene cubic metre containers (cube vessels). A 1 L workingstock of 2.5 mL L−1 DDAC was made by mixing 6.25mL Bardac
®2240, containing 40% DDAC (Lonza Inc.),
in 993.75 mL milli-Q. The working stock was added tothe second cube vessel when the vessel was 75% filledwith seawater. After completely filling the cube vessel with1000 L of seawater, a DDAC concentration of 2.5 μL L−1
was reached. The firstly and thirdly seawater filled cubevessels were used as control. The cube vessels were storedinside to shield them from direct sunlight to prevent excessheating of the water in the vessels.
Before samples were taken from the cube vessels, thecontent was stirred with a clean wooden paddle, which wasinserted through an opening on top of the vessel. Samplesfor zooplankton abundance were taken on Days 0 and 5from a tap at the base of the cube vessels. Zooplanktonabundance before treatment from the DDAC-treated vesselwas interpolated from the two adjacent control vessels. Allcube vessels were sampled for phytoplankton abundance,PSII efficiency and DDAC concentration from a tap at thebase of the vessel for 5 days.
2.1.1.2. Regrowth experiment. One-litre dilutions weremade using treated water from Cube trial 1 at Day 5.As dilution water, 1.2 μm filtered and autoclaved seawaterfrom the first cube vessel was used. Untreated water fromCube trial 1 was used for the control incubation. Undi-luted, 10 times and 100 times diluted treated and controlwater was incubated at 15°C with a 16:8 h light:dark cycleat 50 μmol m−2 s−1 light intensity. A similar second dilu-tion series was made to which living organisms were addedby filtering 1 L of freshly collected seawater over a GF/Ffilter (Whatman) and adding one filter to each incubationbottle. Alongside the dilutions, a negative control (sterileseawater) and a positive control (sterile seawater with afreshly added filter containing living organisms) were incu-bated. For 7–10 days, the phytoplankton abundance andPSII efficiency were monitored.
2.1.1.3. Cube trial 2. Two days after the termination ofCube trial 1, the second control vessel from Trial 1 wastreated with 5 μL L−1 DDAC. This test was performed todetermine the specific mortality of bivalve larvae whichsurvived the initial treatment of 2.5 μL L−1 DDAC of Cubetrial 1. A 1 L working stock of 5 mL L−1 DDAC wasmade by mixing 12.5 mL Bardac
®2240 in 987.5 mL milli-
Q. The working stock was added to the second controlvessel and thoroughly mixed with a wooden paddle toreach a concentration of 5 μL L−1 DDAC. Samples forzooplankton were taken at Days 0 and 5. Samples forDDAC concentration were taken at Days 0, 1, 2, 5 and6. As control for the zooplankton concentration, a 20 Lpolycarbonate bottle (Nalgene) was filled with water from
the second control vessel before the DDAC addition andincubated in the dark alongside the cube vessel. At Day 5,the 20 L bottle was completely analysed for zooplanktonabundance.
2.1.1.4. Cube trial 3. Natural sediment was obtainedfrom a saltwater bay adjacent to the institute. The sedimentwas dried at 60°C for 3 days to remove the water frac-tion. Three opaque black polyethylene cube vessels wereused to perform incubation in the dark. The first vesselwas used as control. To the second and third cube vessels,respectively, 45 and 95 mg L−1 dried sediment was added,to ensure high TSS loads. Two 1 L working stocks of 5mL L−1 DDAC were made by mixing 12.5 mL Bardac
®
2240 in 987.5 mL milli-Q. On 27 May 2011, the three cubevessels were filled with 1000 L seawater from the saltwa-ter harbour adjacent to the institute at low tide. The tankswere filled as in Cube trial 1. When the second and thirdcube vessels were 75% filled, DDAC was added from theworking stocks to reach the final concentration of 5 μL L−1
DDAC.In contrast to Cube trial 1, samples for zooplankton
were taken directly from the bleeding valve at the begin-ning, middle and end of the hour it took to fill all threecube vessels. The average zooplankton count of the threesamples was used as the zooplankton abundance on Day 0before treatment for all cube vessels. Samples for DDAC-treated zooplankton on Day 0 were taken from a tap at thebase of the cube vessels. At Day 5, all three cube vesselswere sampled for zooplankton abundance. Before sampleswere taken from the cube vessels, the content was stirredwith a clean wooden paddle, which was inserted throughan opening on top of the vessel. Samples for phytoplank-ton and bacterial abundance, PSII efficiency and DDACconcentration were obtained from a tap at the base of thevessel during 5 days.
2.1.1.5. Regrowth experiment. At Day 5, 500 mL waterfrom each cube vessel was incubated in polycarbonate bot-tles (Nalgene) at 15°C with a 16:8 light dark cycle at a 50μmol m−2 s−1 light intensity. After 9 days the bottles wereanalysed for phytoplankton abundance and PSII efficiency.
2.1.2. Tank trialOn 16 September 2010, 100 m3 of natural seawater waspumped into a 300 m3 subterranean concrete tank situatedonshore. The water was injected with DDAC (Bardac
®
2240) using a dosing pump into the main pipeline after thepump to reach the final concentration of 5 μL L−1 DDAC.A second concrete subterranean tank was filled as a con-trol. After 5 days, the DDAC-treated water was transferredto another tank. During the transfer, 50 mg L−1 of ben-tonite (natural clay mineral) was injected into the water toneutralize the residual DDAC. On Day 6, a second neutral-ization step was carried out. At various moments during the
incubation, samples were taken for phytoplankton abun-dance, PSII efficiency, DDAC concentration and bacterialabundance using a tap at the base of the tank.
2.2. Analytical methods2.2.1. Zooplankton enumerationZooplankton samples containing natural untreated seawa-ter were obtained by filtering 20 L of seawater over a 50μm sieve either directly from the hose used for filling thecube vessels or from a tap at the base of the cube vessels.Cube vessels containing DDAC-treated water were filteredentirely over a 50 μm plankton net.
Organisms retained in a 50 μm net or sieve after sam-pling were suspended in 100 mL 0.2 μm filtered seawaterand immediately stained with neutral red vitality stain.[14]Stained samples were distributed in a Bogorov dish andcounted using a Zeiss microscope with a 20 times magni-fication. Organisms were determined alive on the basis ofmovement or whether they were stained by neutral red.
2.2.2. Enumeration of phytoplanktonSamples for phytoplankton abundance were analysed induplicate. Analyses were performed within 4 h after sam-pling using a Coulter Epics XL-MCL flow cytometer(Beckman Coulter). Phytoplankton cells were discrim-inated from other particles by detecting the red auto-fluorescence produced by chlorophyll when excited at 488nm using the red fluorescence detector (620 ± 15 nm bandpass filter).[15] Phytoplankton cells were enumerated byplotting red fluorescence against forward scatter. Subse-quent data analysis was carried out using FCS Express 4(De Novo Software). A selection gate was made based onthe cluster of untreated cells. DDAC-treated samples wereanalysed using the same gate used for untreated samples.Particles recorded outside of the gate were considered tobe background noise or cell debris.
2.2.3. Enumeration of bacteriaTwo methods for the enumeration of bacteria were used.In the first method, samples for total bacterial countswere fixed with 1.8%/1% formalin/hexamine for 15 min at4°C, snap frozen in liquid nitrogen and stored at − 80°Cuntil analysis. Prior to analysis, bacterial samples werethawed at room temperature and stained with PicoGreen
®
(250 times commercial stock dilution) (Invitrogen) whichmakes the genomic DNA green fluorescent.
The second method involved a live/dead determination.Unfixed samples were double stained with SYBR
®Green
(Invitrogen) and propidium iodide (Invitrogen). Cells withintact membranes were considered alive and cells with per-meable membranes were considered dead. SYBR® Green(10,000 times commercial stock dilution) stains the DNAof all cells green fluorescent. Propidium iodide (500 times
commercial stock dilution) stains the DNA of cells withpermeable membranes red fluorescent. Bacteria were anal-ysed in duplicate using a Coulter Epics XL-MCL flowcytometer (Beckman Coulter) with a 488 nm excitationlaser. Bacterial counts were discriminated from other parti-cles on the basis of green or red fluorescence intensity andinternal complexity using the green fluorescence detector(525 ± 20 nm band pass filter), red fluorescence detector(620 ± 15 nm band pass filter) and side scatter detector,respectively.
2.2.4. PSII efficiencySamples for PSII efficiency analysis were stored at 4°Cin the dark for 30 min to 4 h prior to analysis. ThePSII efficiency was measured in duplicate as Fv/Fm witha pulse-amplitude modulation (PAM) fluorometer (Walz,Germany).[16]
2.2.5. DDAC analysisDDAC samples were analysed colorimetrically within 4h after sampling according to HACH method 8337.[17]Samples of the laboratory trial were analysed once. Sam-ples of the cube trials, regrowth experiments and tank trialwere analysed in triplicate.
2.3. Statistical analysisIn all statistical analyses, the null hypothesis was that thereis no significant difference between treatment and control.When samples were analysed in triplicate, the 95% confi-dence intervals (CI) of the means were calculated using theMS Excel 2010 function CONFIDENCE.T. A Student’st distribution was used instead of a normal distributionbecause the former is more appropriate when dealing withsmall sample sizes. When the 95% CI did not overlap,the difference between means was considered significant(p < .05).
T-tests were carried out using the MS Excel functionTTEST. A two-tailed distribution was assumed in all testsand α = 0.05. Two types of t-tests were used dependingon the equality of variance of the two samples. An F-testwas performed to test for equality of variance using theMS Excel function FTEST using α = 0.05 to decide whichtype of t-test should be used.
Least squares linear regression models were calcu-lated using SYSTAT 13. When data were nonlinearlydistributed, they were elog-transformed. To test whethermodel coefficients were not significantly different from oneanother, the 95% CI was calculated as: 95% CI = SE * t.Whereby SE is the coefficient’s standard error calcu-lated by SYSTAT 13 and t is the two-tailed t-valuecorresponding with α = 0.05 and degrees of freedom(df) = n − 1.
3. Results3.1. Laboratory trial3.1.1. Phytoplankton abundanceIn the control, the cell abundance of C. calcitrans remainedbetween ∼ 626,000 and ∼ 713,000 cells mL−1 throughoutthe 5-day dark incubation. The cell abundance of T. suecica(t0 ≈ 36,000 cells mL−1) and I. galbana (t0 ≈ 464,000cells mL−1) decreased 58% and 52%, respectively, after 24h and fully or partly recovered, respectively, between 1 and5 days of dark incubation (Figure 1(a)). When cell abun-dance on Day 0 was compared with Day 6, only T. suecicahad not significantly changed (t-test: p = .24). Both the I.galbana and C. calcitrans cultures changed significantlybetween Days 0 and 6 (t-test: p = .03; both cultures).
Several factors could have contributed to the apparentrecovery in T. suecica and I. galbana cell abundance in thecontrol incubations. Clumping of T. suecica cells has beenobserved in other studies,[18] so it could be hypothesizedthat T. suecica cells were clumping at the start of the incu-bation and that the successive shaking of the bottles priorto sampling led to an apparent increase of cells over time.Cell clumps were not detected during flow cytometer dataanalysis; however, sometimes clumps or too large to enterthe flow cytometer uptake needle. Also, cell clumps can betoo rare to be picked up in the 92 μL sample volume anal-ysed by the flow cytometer. It was deemed unlikely thatactual growth occurred during the incubations, since theincubation was in the dark.
The patterns observed in the treated incubations weremarkedly different. Both I. galbana and C. calcitransshowed a significant decline (t-test: p < .05, all DDACconcentrations) directly after the DDAC treatment at allconcentrations tested (Figure 1(b)–(e)). On Day 6, the cellabundances of all DDAC-treated cultures combined wereon average (SD): 1091 (557); 6386 (1,845) and 35 (64)cells mL−1 for I. galbana, C. calcitrans and T. suecica cul-tures, respectively. This is equivalent to a 99.8%, 99.0%and 99.9% decrease in cell abundance for treated I. gal-bana, C. calcitrans and T. suecica cultures, respectively.Based on the shape of the clusters observed during flowcytometer data analysis, what appeared as intact cells inDDAC-treated cultures on Day 6 could also consist out ofcell debris. However, no microscopic analysis was carriedout to confirm this hypothesis.
3.1.2. PSII efficiencyIn the control incubations, the PSII efficiency remainedbetween 0.6 and 0.7 over the entire incubation time(Figure 2). In the DDAC-treated incubations, a completeinactivation of the PSII efficiency was observed in all threespecies at all DDAC concentrations tested up until thelast incubation day. The PSII efficiency of T. suecica wasstill detectable within hours after the DDAC addition, butwas below the detection limit after 24 h at all DDAC
Dow
nloa
ded
by [
For
Sea
Res
earc
h], [
Lou
is P
eper
zak]
at 0
7:27
05
Sept
embe
r 20
14
Environmental Technology 5
0
50
100
150
cell
abun
danc
e (%
)
0
50
100
150
0 1 2 3 4
T.suecica
I.galbana
C.calcitrans
5 6
cell
abun
danc
e (%
)
Days
0
50
100
150
0 1 2 3 4 5 6
cell
abun
danc
e (%
)
Days
(a) (b)
(c) (d)
(e)
Control
10 µL L–1 DDAC
7.5 µL L–1 DDAC5 µL L–1 DDAC
2.5 µL L–1 DDAC
Figure 1. Laboratory trial. Normalized cell abundance of T. suecica, I. galbana and C. calcitrans after the addition of control (a), 2.5(b), 5 (c), 7.5 (d) and 10 μL L−1 DDAC (e). DDAC was added 1 h prior to the sampling of t1, as indicated by the vertical dotted line.Results are duplicates with the average represented by a continuous line.
concentrations tested. The PSII efficiency of I. galbanaremained detectable for 2–3 days after the DDAC addition.The PSII efficiency of C. calcitrans decreased to below thedetection limit directly after the addition of DDAC at allconcentrations tested.
3.1.3. DDAC degradationIn most of the incubations, the observed concentrationof DDAC was lower than was actually added (Table 1).Especially at higher DDAC concentrations, much less wasactually observed in the cultures. In all, except the 2.5 μLL−1 DDAC incubation, the highest concentrations wereobserved in the I. galbana samples. The concentration ofDDAC remained fairly constant during the incubation, asindicated by the small standard deviation of the averageDDAC concentration over the 5-day incubation (Table 1).
3.2. Cube trial 13.2.1. Zooplankton abundanceWithin hours after the addition of 2.5 μL L−1 DDAC,the abundance of living zooplankton decreased 51% from
82,050 to 40,250 organisms m−3. After 5 days, the numberof living zooplankton was reduced 98% to 1500 organ-isms m−3. All remaining zooplankton in the DDAC-treatedcube vessel were bivalve larvae. In the control cube vessel,living zooplankton decreased 42% from 82,050 to 47,550organisms m−3 during the 5-day incubation.
3.2.2. Phytoplankton abundanceBoth the DDAC-treated and control cube vessels showeda similar decrease in phytoplankton abundance during the5-day incubation (Figure 3(a)). The decrease in phyto-plankton abundance was logarithmic in both the controland treated vessels. After 4 days the phytoplankton abun-dance decreased to approximately 4000 cells mL−1 in theDDAC-treated vessel which still exceeded the < 10 viablecells mL−1 required by the IMO.
To test whether decrease rates were significantly dif-ferent between treated and control incubations, an elogtransformation on the data was carried out. The linearregression models for elog transformed data of control-and DDAC-treated samples were: y = − 0.46x + 10.1
Dow
nloa
ded
by [
For
Sea
Res
earc
h], [
Lou
is P
eper
zak]
at 0
7:27
05
Sept
embe
r 20
14
6 C. van Slooten et al.
0.0
0.2
0.4
0.6
0.8
0 1 2 3 4 5 6
PSI
I ef
fici
eny
(Fv/
Fm
)
0
0.2
0.4
0.6
0.8
0 1 2 3 4 5 6
PSI
I ef
fici
ency
(F
v/F
m)
Days
0
0.2
0.4
0.6
0.8
0 1 2 3 4 5 6
PSI
I ef
fici
ency
(F
v/F
m)
Days
Control
2.5 µL L–1 DDAC
5 µL L–1 DDAC 7.5 µL–1 DDAC
10 µL L–1 DDAC
(a) (b)
(c) (d)
(e)
T.suecica
I.galbana
C.calcitrans
Figure 2. Laboratory trial. PSII efficiency of T. suecica, I. galbana and C. calcitrans after the addition of control (a), 2.5 (b), 5 (c),7.5 (d) and 10 μL L−1 DDAC (e). DDAC was added 1 h prior to the sampling of t1, as indicated by the vertical dotted line. Results areduplicates with the average represented by a continuous line.
Table 1. Laboratory trial. Average DDAC concentrationduring the 5-day incubation.
(Control) and y = − 0.37x + 10.0 ( + DDAC). The 95%CI for the coefficient was calculated using: SE * t(df=17):0.023 * 2.11 = 0.048. So for the Control, the coeffi-cient ± 95% CI = − 0.46 ± 0.05. For the + DDAC, the95% CI was 0.036 * 2.11 = 0.08. So the coefficient ±95% CI = − 0.37 ± 0.08. The coefficients of both mod-els overlap so it was concluded that the decline rate in
phytoplankton abundance was not significantly differentbetween the control and treated vessels.
3.2.3. PSII efficiencyIn contrast to phytoplankton abundance, a clear differencein PSII efficiency was observed between the control andDDAC-treated vessels. The PSII efficiency measured insamples from the control vessel resulted in yields associ-ated with healthy phytoplankton (Figure 3(b)). Within 2 h,the PSII efficiency was reduced to 0.1 in the DDAC-treatedvessel and remained below 0.1 during the entire incubation.
3.2.4. DDAC degradationThe DDAC concentration remained constant until Day2 (Figure 3(c)). On Day 5, the DDAC concentration inthe treated vessel was not significantly different from thecontrol vessel due to relatively high variability in con-trol measurements. However, the average concentration in
DDAC. The phytoplankton abundance (a), the PSII efficiency (b)and the DDAC concentration (c) in the cube vessels. Error barsrepresent the 95% CI of triplicate measurements.
treated samples was still significantly different from zeroon Day 5 at 0.3 ± 0.1 μL L−1 DDAC (average ± 95%CI).
3.3. Regrowth experiment3.3.1. Phytoplankton abundanceIn all incubations except the negative control, regrowthwas observed (Figure 4). The incubations of DDAC-treatedwater showed a 1–2 days lag time before regrowth. Thecontrol dilutions did not show this lag phase, except for the100 times dilution which showed a 1 day lag phase. Whenuntreated phytoplankton was added to the DDAC-treatedwater, the undiluted incubation showed a 1 day lag phasebefore regrowth started (Figure 4(c)). The diluted incuba-tions showed no lag phase prior to regrowth. Control water
with untreated phytoplankton added showed no lag phasebefore regrowth.
3.3.2. PSII efficiencyIn all incubations except the negative control, a strongrecovery of PSII efficiency was observed (Figure 5).Notably, the PSII efficiency of undiluted DDAC-treatedwater with fresh phytoplankton was much lower on Day0 than the other dilutions (Figure 5(c)).
3.4. Cube trial 23.4.1. Zooplankton abundanceZooplankton abundance declined 62% from 23,850 to9000 organisms m−3 within hours after the addition of 5μL L−1 DDAC. At Day 5, no zooplankton organisms wereobserved in the treated cube vessel. So the zooplanktondisinfection was 100% after treatment at Day 5. The zoo-plankton abundance in the control decreased by 82% from23,850 at Day 0 to 4300 organisms m−3 at Day 5.
3.4.2. DDAC degradationThe observed concentration on Day 0 was 5.8 ± 0.3 μLL−1 DDAC (average ± 95% CI). On Day 5, the DDACconcentration observed in the treated vessel remained sig-nificantly different from the control samples at 1.6 ± 0.4μL L−1 DDAC (average ± 95% CI). On Day 6, the DDACconcentration was not significantly different from Day 5(t-test: p = .30), indicating that a plateau was reached inthe degradation process.
3.5. Cube trial 33.5.1. Zooplankton abundanceAfter 5 days, 85% of the zooplankton present in the controlvessel at Day 1 was still alive. In the DDAC-treated vesselswith extra sediment, virtually no organisms were left onDay 5 except for a phyllodocidae larva and a nereididaelarva in the 45 mg L−1 TSS vessel (Table 2).
The dominant category of zooplankton organisms in thevessels on Day 1 were bivalve larvae. The second mostabundant category was balanidae nauplia. Both these typesof zooplankton were considered as hard to kill by con-ventional BWTS. The most difficult to kill zooplanktoncategory found in the Wadden sea are balanidae cyprid lar-vae (personal communication with Frank Fuhr and Isabelvan der Star). These were however not present at the timeof testing.
3.5.2. Phytoplankton abundanceJust as observed in Cube trial 1, the phytoplankton abun-dance trends in control and treated vessel were remarkablysimilar (Figure 6(a)). The difference from Cube trial 1 was
Dow
nloa
ded
by [
For
Sea
Res
earc
h], [
Lou
is P
eper
zak]
at 0
7:27
05
Sept
embe
r 20
14
8 C. van Slooten et al.
0 1 2 3 4 5 6 7
0 1 2 3 4 5 6 7Days
1
10
100
1000
10,000
100,000
0 1 2 3 4 5 6 7 8 9 10
Phy
topl
ankt
on (
cells
mL
–1)
1
10
100
1000
10,000
100,000
0 1 2 3 4 5 6 7
Phy
topl
ankt
on (
cells
mL
–1)
Days
(a)
(d)(c)
(b)
2.5 µL L–1 DDAC+Living organisms
2.5 µL L–1 DDAC Control
Control+Living organisms
0x
10x
100x
PC
NC
Figure 4. Regrowth experiment. Phytoplankton abundance. DDAC-treated water dilution series (a); control water dilution series (b);DDAC-treated dilution series with addition of living organisms (c) and control water dilution series with addition of living organisms(d). NC = negative control (sterile seawater). PC = positive control (sterile seawater with addition of living organisms). Results areduplicates with the average represented by a continuous line.
aThe average zooplankton concentration at intake for all threecube vessels.
that the decrease in phytoplankton abundance could bedescribed by a linear regression model. The phytoplank-ton trends could be described by the following models:y = − 773x + 4894 (Control); y = − 667x + 4421 (45mg L−1 TSS); y = − 640x + 4591 (95 mg L−1 TSS).
It was tested whether the decrease rate of phytoplank-ton was significantly different for the different treatments.The 95% CI for the coefficients was calculated using: SE* t(df;11). The 95% CI for the three decrease rates were:47.2 * 2.201 = 104 (Control); 62.2 * 2.201 = 137 (45 mgL−1 TSS) and 78.3 * 2.201 = 173 (95 mg L−1 TSS). Thecoefficient ± 95% CI for the control and two treatmentswere: − 773 ± 104 (Control); − 667 ± 137 (45 mg L−1
TSS) and − 640 ± 173 (95 mg L−1 TSS).
All the 95% CI overlapped with each other so thedecrease rates in phytoplankton were not significantlydifferent among the control and two treatments.
3.5.3. PSII efficiencySimilar to Cube trial 1, the PSII efficiency of the treatedvessels was reduced to below 0.1 within hours andremained close to the detection limit for the duration ofthe experiment (Figure 6(b)). The PSII efficiency observedin the control samples were associated with healthy phyto-plankton cells.
3.5.4. DDAC degradationOn Day 1, 70% of the added DDAC was measured in thevessels (Figure 6(c)). Day 5 marked the first day that a sig-nificant reduction in DDAC was observed in both vessels.In none of the vessels, the DDAC concentration decreasedbelow the detection limit.
3.5.5. Bacterial abundanceIn the control vessel, the majority of bacteria were aliveat Day 1. After Day 3, the bacterial abundance decreasedstrongly in the control (Figure 7(a)). The bacterial abun-dance in the DDAC-treated vessels was 30%–40% lower
Dow
nloa
ded
by [
For
Sea
Res
earc
h], [
Lou
is P
eper
zak]
at 0
7:27
05
Sept
embe
r 20
14
Environmental Technology 9
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 1 2 3 4 5 6 7
Qua
ntum
yie
ld (
Fv/
Fm
)
Days
(b)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 1 2 3 4 5 6 7 8 9 10
PSI
I ef
fici
ency
(F
v/F
m)
Days
(a)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 1 2 3 4 5 6 7
Qua
ntum
yie
ld (
Fv/
Fm
)
Days
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 1 2 3 4 5 6 7
PSI
I ef
fici
ency
(F
v/F
m)
Days
(c)
2.5 µL L–1 DDAC
Control
2.5 µL L–1 DDAC+Living organisms
Control+Living organisms
(d)
0x
10x
100x
PC
NC
Figure 5. Regrowth experiment. PSII efficiency. DDAC-treated water dilution series (a); control water dilution series (b); DDAC-treateddilution series with addition of living organisms (c); control water dilution series with addition of living organisms (d). NC = negativecontrol (sterile seawater). PC = positive control (sterile seawater with addition of living organisms). Results are duplicates with theaverage represented by a continuous line.
than that in the control vessels on Day 1 (Figure 7(b)and (c)). Notably, the bacterial assembly in the treatedvessels consisted almost entirely of dead bacteria. Duringthe course of the incubation, bacteria started to regrow inthe DDAC-treated vessels. On Day 5, the bacterial abun-dance in both DDAC-treated vessels exceeded the bacterialabundance in the control vessel.
3.5.6. Regrowth experimentAfter 9 days of light incubation, the control had increasedfrom ∼ 1000 to ∼ 8400 phytoplankton cells mL−1. So,clear regrowth was observed in the control. The PSII effi-ciency in the control had increased from 0.38 to 0.58 soa recovery was also observed there. No regrowth or PSIIefficiency recovery was observed in DDAC-treated sam-ples. On the contrary, the intact phytoplankton cells at Day5 of the dark incubation had degraded after 9 days in thelight. Phytoplankton cell counts ranged from 0 to 7 cellsmL−1 for the 45 and 95 mg L L −1 TSS, respectively.
3.6. Tank trial3.6.1. Phytoplankton abundanceIn contrast to cube trials, phytoplankton abundance didnot decrease in the control tank. The control tank showed
no significant decline in phytoplankton abundance on Day4 (t-test: p = .06) (Figure 8(a)). The 5 μL L−1 DDACtreatment immediately led to a decrease of 75% in phyto-plankton abundance. At Day 5, more than 500 cells mL−1
remained in the treated tank, which is 50 times higher thanthe < 10 viable cells mL−1 allowed by the D-2 standard.[4]
3.6.2. PSII efficiencyWithin hours after the DDAC treatment, the PSII effi-ciency was reduced to 0.04 and remained below 0.1 forthe duration of the experiment (Figure 8(b)).
The strong decrease in phytoplankton numbers andthe inactivation of PSII indicate that 5 μL L−1 DDACwas effective with respect to phytoplankton. However, aregrowth experiment was not performed; therefore the via-bility of the remaining 500 cells mL−1 at Day 5 could notbe determined.
3.6.3. DDAC neutralizationDDAC levels did not decrease significantly during the5-day tank trial (t-test: p = .40) (Figure 8(c)). Afteran initial neutralization with 50 mg L−1 bentonite, theDDAC concentration decreased to 0.75 ± 0.04 μL L−1
DDAC. Phytoplankton abundance (a); PSII efficiency (b) andDDAC concentration (c) in the cube vessels. a,b45 and 95 refer tothe amount of extra TSS in mg L−1 added to the DDAC-treatedvessels. Results of PSII efficiency are duplicates with the averagerepresented by a continuous line. Error bars represent the 95% CIof triplicate DDAC measurements.
(average ± 95% CI) which was still significantly differ-ent from zero. At Day 6, the neutralization procedure wasrepeated which reduced the DDAC concentration to levelsindiscernible from zero.
3.6.4. Bacterial abundanceBacterial abundance was initially reduced due to theDDAC treatment. However, on Days 4 and 5 clearregrowth of bacteria was observed (Figure 8(d)). So justlike in Cube trial 3 (Figure 7(c)), 5 μL L−1 DDAC wasunable to prevent bacterial regrowth for a 5-day period.
DDAC. Bacterial abundance. Results are duplicates with theaverage represented by a continuous line.
3.7. OverviewResults were classified as insufficient, inconclusive andsufficient with respect to compliance with IMO regulations(Table 3). Bacterial disinfection efficacy was also includedin the summary despite the lack of IMO regulation ofmost bacteria. The classification of sufficient or insufficientdisinfection was based on the assessment whether bacte-ria were initially affected by the treatment and whetherregrowth occurred within the 5-day dark incubation.
4. Discussion4.1. Establishing compliance with D-2IMO-type approved BWTS should reduce the concen-trations of certain groups of viable organism to belowstandard (D-2) values as well as being environmentallysafe.[8] For systems using toxic compounds environmen-tally safe means that the residual toxicity of dischargedwater should be below a certain threshold. The residualtoxicity level can be reached by degradation of the toxiccompounds during the 5-day holding time, or if that is not
Dow
nloa
ded
by [
For
Sea
Res
earc
h], [
Lou
is P
eper
zak]
at 0
7:27
05
Sept
embe
r 20
14
Environmental Technology 11
–0.2
0.0
0.2
0.4
0.6
PSI
I ef
fici
ency
(F
v/F
m)
Control
+DDAC
0
3
6
9
12
15
Phy
topl
ankt
on (
×103
cells
mL
–1)
Control
+DDAC
(a)
(c)
0
3
6
9
12
15
01 2 3 4 5 6
Bac
teri
a (×
105 c
ells
mL
–1)
Days
Control
+DDAC
(b)
(d)
–2
0
2
4
6
0 1 2 3 4 5 6
DD
AC
(µL
L–1
)
Days
+DDAC
Neutralized
Neutralized twice
Control
Figure 8. Tank trial. Natural seawater treated with 5 μL L−1 DDAC. Phytoplankton abundance (a), PSII efficiency (b), DDACdegradation (c) and bacterial abundance (d). Error bars represent the 95% CI of triplicate DDAC measurements.
Table 3. Overview of results classified as: insufficient ( − ), inconclusive ( ± ) or sufficient ( + ).
Variable
T. suecica; I. galbana; C. calcitrans Cube trials Tank trial
DDAC concentration (μL L−1)
2.5 5 7.5 10 2,5 5 5
Zooplankton disinfection − +Phytoplankton disintegration ± ± ± a ± ± ± ± ± ± ± ± ± ± ± +PSII inactivation + + + + + + + + + + + + + + +Phyto. regrowth prevention − +Bacterial disinfection + +Bact. regrowth prevention − −DDAC degradation − − − − − − − − − − − − ± − −aThe three consecutive ‘ + ’, ‘ ± ’ or ‘ − ’ signs in the first four columns correspond with the results of T. suecica, I. galbana and C.calcitrans, respectively.
sufficient, by adding a neutralizing agent. The merits andcaveats of DDAC in the ballast water treatment will bediscussed for three organism groups, zooplankton, phyto-plankton and heterotrophic bacteria, their viability as wellas the environmental acceptability of DDAC.
4.1.1. Organism concentrationsThe DDAC treatment affected the physical integrity of phy-toplankton cells in monocultures and the tank trial. In thecube vessels, phytoplankton abundance also decreased inthe control so no significant differences were observed.Since no clear differences in abundance and PSII effi-ciency were observed between 2.5 and 10 μL L−1 DDAC,it appeared that 2.5 μL L−1 DDAC was sufficient to disin-fect all three monocultures. However, in natural seawater,zooplankton and phytoplankton showed resistance to 2.5
μL L−1 DDAC. Eventually, 5 μL L−1 DDAC proved suf-ficient to disinfect natural seawater in compliance withD-2. The rapid disappearance of cells after treatment is incongruence with other active substances like Peraclean
®
Ocean,[19,20] chlorine [21] and chlorine dioxide,[22]which tend to disrupt cells directly after treatment.
In Cube trials 1 and 3, the cause of the similar decreasein phytoplankton abundance between treated and untreatedvessels remained unclear. It could be hypothesized that thepolyethylene cube vessels had adverse effects on phyto-plankton survival. However, flocculation could also causean apparent decline in phytoplankton cells because of sed-imentation of phytoplankton clumps and the inability tocount cell clumps as loose cells using a flow cytometer.[23]Grazing was ruled out as a cause for phytoplankton declinein the treated vessels, since zooplankton was almost com-pletely eradicated by the DDAC treatment. On the other
Dow
nloa
ded
by [
For
Sea
Res
earc
h], [
Lou
is P
eper
zak]
at 0
7:27
05
Sept
embe
r 20
14
12 C. van Slooten et al.
hand, the strong decline in phytoplankton abundance in thetreated tank in the full-scale tank experiment was a goodindicator that the DDAC treatment had been effective.
Although the 2.5 μL L−1 DDAC treatment in Cube trial1 caused a reduction of 98% in living zooplankton, theabundance after 5 days was still well above the limit of< 10 organisms m−3 allowed by the IMO.[4] The appar-ent resistance of bivalve larvae to DDAC could be causedby the ability of bivalves to close their shells and thustemporarily avoid adverse environmental conditions.[24]
In Cube trial 2, the complete eradication of zooplank-ton organisms, including the bivalve larvae, indicated that5 μL L−1 DDAC has sufficient zooplankton disinfectioncapacity to comply with the IMO D-2 standard. The largedecrease in the control bottle indicated, however, that the12-day presence of the seawater in the cube vessel and 20 Lpolycarbonate bottle had negatively impacted the survivalrate of the zooplankton.
The successful eradication of all but two worm larvaein the treated vessels of Cube trial 3 implied that the extraTSS added did not impair the disinfection capacity of 5μL L−1 DDAC with respect to zooplankton. In the absenceof the most challenging balanidae cyprid larvae, 5 μL L−1
DDAC was capable of eradicating virtually all zooplank-ton organisms. So these results were considered promisingwith respect to zooplankton disinfection efficacy of DDACat current concentrations to comply with the IMO D-2standard.
The reduction in bacterial abundance in the control ofCube trial 3 was probably the result of depletion of car-bohydrates and continued grazing by microzooplankton.Carbohydrates are produced by phytoplankton under lightconditions but in the dark no photosynthesis can take place.Carbohydrates produced by phytoplankton are consideredan important organic carbon source for bacteria.[25,26]Apart from three indicator microbes, bacterial abundanceis not regulated by the IMO, so the observed regrowth hasno implications for assessing DDAC as a suitable ballastwater treatment method.
4.1.2. ViabilityThe PSII efficiency results indicated that PSII of the phyto-plankton in all experiments was severely damaged. WhenPSII efficiency levels drop below the detection limit, nophotosynthesis can take place and hence no energy pro-duction. The inactivation of PSII could indicate that thephytoplankton cells were effectively killed by the DDACtreatment. However, recovery of PSII efficiency afterremoval of toxic compounds has been reported before.[27]
In the regrowth experiment of Cube trial 1, the decreasein PSII efficiency observed in undiluted treated water withuntreated phytoplankton added could indicate that watercontaining 0.3 μL L−1 DDAC still contained residual tox-icity to phytoplankton. Nevertheless, despite the inactiva-tion of PSII during the 5-day DDAC treatment, regrowth
and PSII recovery occurred within 3–4 days. The shortlag phase indicated that 2.5 μL L−1 DDAC had insuffi-cient phytoplankton disinfection capacity to comply withthe IMO D-2 standard. Also, it was shown that the loss ofPSII efficiency during treatment was no definitive indica-tor for the loss of phytoplankton viability after treatment.In Cube trial 3, the loss of PSII efficiency at 5 μL L−1
DDAC was comparable to the effect observed at 2.5 μLL−1 DDAC. This indicated that the maximum effect of PSIIdisruption was indeed already reached at the lower DDACconcentration.
4.1.3. Regrowth potentialDuring the cube trials, it became clear that phytoplank-ton cell integrity and PSII disruption were not reliablepredictors of cell viability. In order to establish the effec-tiveness of a treatment, it was deemed essential to carryout regrowth experiments with treated water. The presenceof regrowth in all 2.5 μL L−1 DDAC-treated dilutions ofCube trial 1 showed that residual toxicity to phytoplank-ton probably was negligible after 1 or 2 days incubationin the light and that sufficient viable phytoplankton cellsremained after treatment to enable regrowth. In contrast,the absence of regrowth and the continued degradation ofany remaining cells in Cube trial 3 was a clear indicationthat intact phytoplankton cells at Day 5 of the dark incu-bation were not viable. Although it cannot be ruled outthat, in Cube trial 3, residual DDAC hampered regrowth.Nevertheless, the degradation to < 10 cells mL−1 after9 days in the tank experiment is a strong indication that5 μL L−1 DDAC had sufficient phytoplankton disinfec-tion capability to comply with the IMO D-2 standard.It can be challenging to establish how long it shouldtake for regrowth to occur before it can be reliably con-cluded that the treatment was effective. In official G8land-based tests, it is custom to incubate up to 7 daysafter treatment.[28] Since the regrowth incubation in Cubetrial 3 lasted 9 days, without regrowth, it can be con-cluded with reasonable confidence that the treatment hadeffectively rendered all phytoplankton unviable. Bacterialregrowth in DDAC-treated vessels indicated that certainbacterial species were resistant to DDAC. Possibly, a com-munity shift in bacterial composition took place, sincethe bacteria at Day 1 in Cube trial 3 were almost com-pletely killed by the DDAC treatment. Bacterial resistanceto quaternary ammonium compounds has been reportedbefore [29] and it is therefore not improbable that DDAC-resistant bacteria became dominant. It would be interestingto investigate the causal relationship between the increasein bacterial abundance and the significant decrease inDDAC observed on Day 5 (Figure 6(c)). The decreasein DDAC concentration might be due to either chemi-cal decomposition which could have enabled bacteria toregrow or bacteria may have been responsible for theDDAC degradation.[30]
Dow
nloa
ded
by [
For
Sea
Res
earc
h], [
Lou
is P
eper
zak]
at 0
7:27
05
Sept
embe
r 20
14
Environmental Technology 13
4.1.4. Environmental acceptabilityThe reason that the observed DDAC concentrations werelower than the added DDAC likely was an immediatereaction of DDAC with organic matter in the phytoplank-ton culture after addition and adsorption to phytoplanktoncells. Therefore, in ballast water with a relatively highorganic carbon content, the amount of DDAC added dur-ing ballasting should be higher than in water with a lowcarbon content. Ideally, the DDAC concentration is mon-itored directly after the addition, as is the case in electrochlorination systems where chlorine is measured, so thatthe disinfectant dose can be adjusted during the ships’ballasting operations, thereby reducing the chance of over-dosing.
The persistence of DDAC during most incubationsensured that the disinfection process was continuous dur-ing the whole 5-day period. In contrast to the monocul-ture tests, DDAC did degrade in natural seawater with ahalf-life of approximately 3 days. DDAC half-lives rang-ing from days to years in soil and sewage cultures havebeen reported before. Henderson [30] hypothesized thatmicrobial population compositions could play a majorrole in the degradation rate of DDAC. Indeed, bacteriausing DDAC as sole carbon source have been reportedbefore.[31]
After adding 5 μL L−1 DDAC, degradation was eitherinsufficient or even absent. So, to be responsibly usedas a ballast water treatment option a neutralization stepis required, since it is not allowed by the IMO to dis-charge toxic ballast water.[8] Therefore, monitoring theDDAC concentration upon discharge, in addition to moni-toring at ballasting, appears critical to assess the need for apotential neutralization step in order to be environmentallyacceptable.
Ideally, BWTS exclude or limit the use of activesubstances. Chemical-free treatment techniques are, e.g.,filtration, cavitation and UV-radiation. Whenever activesubstances are applied, preferably these are generated on-site, e.g. in the electro chlorination and chlorine dioxidetreatment to eliminate the transport and storage of dan-gerous chemicals. Usually, residual chlorine is neutralizedusing a sulphur compound like sodium bisulphite, which isinjected as a liquid in the discharge line during deballastingof a ballast tank.[32] Compared to existing technologies,DDAC combined with bentonite neutralization has severaldisadvantages. DDAC and bentonite have to be transportedand stored on-site creating potential logistical and health-related risks. In addition, bentonite is a clay mineral whichis injected as a powder into the ballast water. Althoughprecise turbidity measurements were not performed, thesuspended bentonite was clearly visible with the nakedeye after neutralization. Discharging turbid water couldbe an environmental risk in some areas and it is unclearwhether port authorities would allow the discharge ofbentonite-containing ballast water in their harbours.
5. ConclusionsMost BWTS apply physical separation to remove the largerparticles from the ballast water prior to the chemical treat-ment or UV-radiation.[32] This is because the chemicaltreatment alone is often not sufficient to reliably disinfectzooplankton organisms under all circumstances. Excep-tions exist like the SeaKleen
®trials which resulted in
reliable disinfection of both zooplankton and phytoplank-ton using menadione as sole active substance without theuse of filtration.[33] At 2.5 μL L−1 DDAC, zooplanktonabundance was not reduced to below 10 organisms m−3.In particular, bivalve larvae appeared resistant to the treat-ment. However, at 5 μL L−1 DDAC, zooplankton abun-dance was twice reduced to below 10 organisms m−3. Itcan be concluded that 5 μL L−1 DDAC is probably suffi-cient to disinfect zooplankton organisms from seawater inaccordance with the IMO D-2 standard.[4] Nevertheless,although DDAC results are promising, given the experi-ence of other BWTS manufacturers that filtration seemsnecessary, it is recommended that extensive zooplanktontrials are carried out in full-scale tests in order to val-idate DDAC disinfection efficacy on a wide variety ofzooplankton organisms.
In phytoplankton monocultures DDAC clearly dis-rupted cells at 2.5–10 μL L−1 DDAC. In the cube trials,phytoplankton abundance decreased not only in the treatedsamples, but also in the control samples. So it remainedunclear whether DDAC was solely responsible for thedecline. On the other hand, in the tank trial a clear dif-ference in phytoplankton abundance between control andtreated samples was observed. No regrowth was observedafter 9 days in the light after the 5 μL L−1 DDAC treat-ment. It is therefore concluded that 5 μL L−1 DDAC wassufficient to achieve reliable disinfection in seawater withrespect to phytoplankton in accordance with IMO.[4]
Although initially reduced, bacterial regrowth wasobserved at 5 μL L−1 DDAC both in the cube vessels andtank trials. Therefore, the resistance of the three indicatormicrobes to DDAC or their capability of metabolizing thedisinfectant still needs to be tested.
In none of the experiments DDAC naturally degradedto non-detectable levels after 5 days. Regrowth exper-iments hinted at adverse effects of residual DDAC onphytoplankton regrowth and PSII efficiency. Sufficientneutralization to comply with environmental acceptabilityof 5 μL L−1 DDAC was only achieved after two ben-tonite neutralization cycles. The addition of bentonite clayto ballast water considerably increased the particle loadof the discharge water that may be subject to regulationsin various countries and ports. In addition, the instal-lation of a bentonite injection apparatus in addition tothe transportation to the ship and storage of bentoniteclay and DDAC on board will complicate the practi-cal application of DDAC as a ballast water treatmentmethod.
Dow
nloa
ded
by [
For
Sea
Res
earc
h], [
Lou
is P
eper
zak]
at 0
7:27
05
Sept
embe
r 20
14
14 C. van Slooten et al.
AcknowledgementsThe authors thank all personnel of the plankton laboratory atNIOZ. In particular, the authors thank Frank Fuhr and Isabel vander Star for carrying out all zooplankton sampling and analysisand for supplying valuable information regarding zooplanktonresistance to disinfection. This work has been co-funded by theNorth Sea Region Program under the ERDF of the EuropeanUnion.
References[1] Bax N, Williamson A, Aguero M, Gonzalez E, Geeves W.
Marine invasive alien species: a threat to global biodiversity.Mar Policy. 2003;27(4):313–323.
[2] Gollasch S. Overview on introduced aquatic species inEuropean navigational and adjacent waters. Helgol MarRes. 2006;60(2):84–89.
[3] Drake JM, Lodge DM. Rate of species introductions in theGreat Lakes via ships’ ballast water and sediments. Can JFish Aquat Sci. 2007;64(3):530–538.
[4] Anonymous. International convention for the control andmanagement of ships’ ballast water and sediments. Inter-national Maritime Organisation, 16 February 2004. ReportNo.: 36.
[5] Gregg MD, Rigby G, Hallegraeff M. Review of two decadesof progress in the development of management options forreducing or eradicating the phytoplankton, zooplankton andbacteria in ship’s ballast water. Aq Inv. 2009;4(3):521–565.
[6] Anonymous. Guidelines for approval of ballast water man-agement systems (G8). Annex 3. International MaritimeOrganisation, 2005 MEPC 125/53.
[7] Anonymous. Guidelines for approval of ballast water man-agement systems (G8). Annex 4. International MaritimeOrganisation, 2008 MEPC 174/58.
[8] Anonymous. Procedure for approval of ballast water man-agement systems that make use of active substances (G9).International Maritime Organisation, Committee MMEP;2008 MEPC 57/21.
[9] Anonymous. Reregistration eligibility decision for aliphaticalkyl quaternaries (DDAC). Washington (DC): UnitedStates Environmental Protection Agency; 2006. Availablefrom: http://www.epa.gov/oppsrrd1/REDs/ddac_red.pdf
[10] Ioannou CJ, Hanlon GW, Denyer SP. Action of disinfectantquaternary ammonium compounds against Staphylococcusaureus. Antimicrob Agents Chemother. 2007;51(1):296–306. Epub 23 October 2006. doi: 10.1128/AAC.00375–06.
[11] Liebich V, Stehouwer PP, Veldhuis MJW. Re-growth ofpotential invasive phytoplankton following UV-based bal-last water treatment. Aq Inv. 2012;7:29–36.
[12] Guillard RRL, Ryther JH. Studies of marine planktonicdiatoms I. Cyclotella nana hustedt, and Detonula confer-vacea (cleve) gran. Can J Microbiol. 1962;8:229–239.
[13] Berges JA, Franklin DJ, Harrison PJ. Evolution of an arti-ficial seawater medium: improvements in enriched seawa-ter, artificial water over the last two decades. J Phycol.2001;37:1138–1145.
[14] Crippen RW, Perrier JL. Use of neutral red and Evansblue for live-dead determinations of marine plankton (withcomments on use of rotenone for inhibition of graz-ing). Stain Technol. 1974;49:97–104. PubMed PMID:WOS:A1974S552900009.
[18] Moheimani NR. Inorganic carbon and pH effect on growthand lipid productivity of Tetraselmis suecica and Chlorellasp (Chlorophyta) grown outdoors in bag photobioreac-tors. J Appl Phycol. 2012;25:387–398. Epub 24 July 2012.doi: 10.1007/s10811-012-9873-6.
[19] de Lafontaine Y, Despatie SP, Veilleux E, Wiley C. Onboardship evaluation of the effectiveness and the potential envi-ronmental effects of PERACLEAN (R) ocean for Ballastwater treatment in very cold conditions. Environ Toxicol.2008;24(1):49–65. doi: 10.1002/tox.20394. PubMed PMID:WOS:000263533000007.
[20] Veldhuis MJW, Fuhr F, Boon JP, Ten Hallers-Tjabbers C.Treatment of ballast water: how to test a system with a mod-ular concept? Environ Technol. 2006;27(8):909–921. Epub11 May 2010. doi: 10.1080/09593332708618701. PubMedPMID: WOS:000240910500010.
[21] Peperzak L, Stehouwer P, Boon J. Land-based test accord-ing to IMO Guideline G8 of the Hamworthy Water Sys-tems Ltd Aquarius™-UV Ballast Water Treatment System(2011–2012). Texel: NIOZ; 2012.
[22] Veldhuis MJW, Fuhr F, Stehouwer P. Final report ofthe land-based testing of the Ecochlor® system, for typeapproval according to regulation-D-2 and the relevant IMOguideline (April–July 2008). Texel: NIOZ, 2009 February2009. Report No.
[23] Ozbay G, Jackson M. Flow cytometry application in marinephytoplankton study: a case study investigating effectsof formalin preservation on phytoplankton count and cellintegrity. J Biotech Res. 2010;2:79–100.
[24] Valenti TW, Cherry DS, Currie RJ, Neves RJ, Jones JW,Mair R, Kane CM. Chlorine toxicity to early life stages offreshwater mussels (bivalvia: unionidae). Environ ToxicolChem. 2006;25(9):2512–2518.
[25] Alderkamp AC, van Rijssel M, Bolhuis H. Characteriza-tion of marine bacteria and the activity of their enzymesystems involved in degradation of the algal storage glu-can laminarin. FEMS Microbiol Ecol. 2007;59(1):108–117.Epub 24 October 2006. doi: 10.1111/j.1574-6941.2006.00219.x
[26] Hahnke S, Sperling M, Langer T, Wichels A, Gerdts G,Beardsley C, Brinkhoff T, Simon M. Distinct seasonalgrowth patterns of the bacterium Planktotalea frisia in theNorth Sea and specific interaction with phytoplankton algae.FEMS Microbiol Ecol. 2013;86:185–199. Epub 19 June2013. doi: 10.1111/1574-6941.12151.
[27] Buma GJ, Sjollema SB, Van der Poll WH, Klamer HJC,Bakker JF. Impact of the antifouling agent Irgarol 1051 onmarine phytoplankton species. J Sea Res. 2009;61:133–139.Epub 10 December 2009. doi: 10.1016/j.seares.2008.11.007
[28] Peperzak L. The Cathelco ballast water treatment system:results of the 2012 land-based IMO G8 tests at NIOZ. Texel:NIOZ; 2013.
[29] Mitchell BA, Brown MH, Skurray RA. QacA mul-tidrug Efflux pump from Staphylococcus aureus: compar-ative analysis of resistance to Diamidines, Biguanidines,and Guanylhydrazones. Antimicrob Agents Chemother.1998;42(2):475–477.
[30] Henderson ND. A review of the environmental impactand toxic effects of DDAC. Victoria (BC): EnvironmentalProtection Division, BC Environment, Ministry of Environ-ment, Land and Parks; 1992. 45p.
[31] Van Ginkel CG, Hoenderboom A, Van Haperen AM,Geurts MGJ. Assessment of the biodegradability ofDialkyldimethylammonium salts in flow through systems.J Environ Sci Health. 2003;A38(9):1825–1835. Epub 18August 2006. doi: 10.1081/ESE-120022881.
[32] Tsolaki E, Diamadopoulos E. Technologies for ballastwater treatment: a review. J Chem Technol Biotechnol.2010;85(1):19–32. doi: 10.1002/jctb.2276.
[33] Wright DA, Dawson R, Caceres V, Orano-Dawson CE,Kananen GE, Cutler SJ, Cutler HG. Shipboard testingof the efficacy of SeaKleen (R) as a ballast water treat-ment to eliminate non-indigenous species aboard a workingtanker in Pacific waters. Environ Technol. 2009;30(9):893–910. Epub 06 Aug 2009. doi: 10.1080/09593330902929889, Pii 913696753. PubMed PMID: ISI:000268757800004.
