This article was downloaded by: [Ecole Polytechnique Montreal]On: 18 November 2013, At: 11:30Publisher: 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

Gamma irradiation: a method to produce an abioticcontrol for biological activated carbonC. Stoquartab, G.A. Vázquez-Rodríguezac, P. Servaisb & B. Barbeaua

a NSERC Industrial Chair in Drinking Water, Department of Civil, Mining and GeologicalEngineering, École Polytechnique de Montréal, CP 6079, Succursale Centre-Ville, Montréal,QC, Canada H3C 3A7b Écologie des Systèmes Aquatiques, Université Libre de Bruxelles, Campus de la Plaine, C.P.221, Boulevard du Triomphe, B-1050 Brussels, Belgiumc Centro de Investigaciones Químicas, Universidad Autónoma del Estado de Hidalgo,Carretera Pachuca-Tulancingo km. 4.5, 42184 Mineral de la Reforma, Hgo, MexicoAccepted author version posted online: 08 May 2013.Published online: 05 Jun 2013.

To cite this article: C. Stoquart, G.A. Vázquez-Rodríguez, P. Servais & B. Barbeau (2013) Gamma irradiation: a methodto produce an abiotic control for biological activated carbon, Environmental Technology, 34:23, 3091-3097, DOI:10.1080/09593330.2013.803132

To link to this article: http://dx.doi.org/10.1080/09593330.2013.803132

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

Environmental Technology, 2013Vol. 34, No. 23, 3091–3097, http://dx.doi.org/10.1080/09593330.2013.803132

Gamma irradiation: a method to produce an abiotic control for biological activated carbon

C. Stoquarta,b∗, G.A. Vázquez-Rodrígueza,c, P. Servaisb and B. Barbeaua

aNSERC Industrial Chair in Drinking Water, Department of Civil, Mining and Geological Engineering, École Polytechnique deMontréal, CP 6079, Succursale Centre-Ville, Montréal, QC, Canada H3C 3A7;

bÉcologie des Systèmes Aquatiques, Université Libre de Bruxelles, Campus de la Plaine, C.P. 221, Boulevard du Triomphe, B-1050Brussels, Belgium; cCentro de Investigaciones Químicas, Universidad Autónoma del Estado de Hidalgo, Carretera Pachuca-Tulancingo

km. 4.5, 42184 Mineral de la Reforma, Hgo, Mexico

(Received 16 January 2013; final version received 29 April 2013 )

The aim of this paper was to investigate the feasibility of using gamma irradiation to inhibit the microbial activity of biologicalpowder activated carbon (PAC) without impacting its adsorptive properties. First of all, the range of dose of gamma raysrequired to produce abiotic PAC was selected on the basis of heterotrophic plate counts (HPC) inactivation and methyleneblue (MB) adsorption kinetics. Doses inferior to 10 kGy were not sufficient to inhibit the culture of heterotrophic bacteria.On the other hand, doses superior to 15 kGy were demonstrated to affect the adsorption rate of MB. Consequently, a dosecomprised between 10 and 15 kGy was selected for further investigation. In order to validate the adequacy of the rangeof dose (i.e. 10–15 kGy), adsorption characteristics were tested by monitoring the removal kinetics of refractory dissolvedorganic carbon (RDOC). No significant differences were observed between irradiated and non-irradiated biological PAC forthe adsorption of RDOC. Irradiated, non-irradiated and virgin PAC were also evaluated in terms of abundance of viable (usingthe LIVE/DEAD BacLight™ method) bacteria and in terms of heterotrophic biomass activity. The results of the BacLight™method demonstrated that attachment of the biofilm on the PAC was not impacted by the irradiation and heterotrophic activitymeasurements demonstrated that the latter could be radically reduced in the range of dose selected. In conclusion, when usinga proper dose, the gamma irradiation of colonized activated carbon drastically reduced the heterotrophic activity on activatedcarbon without significantly impacting its adsorptive behaviour.

Keywords: activated carbon; gamma irradiation; abiotic control; adsorption; biodegradation

IntroductionMaintaining stable drinking water quality during water tran-sit in distribution systems is an important challenge forwater utilities. In this context, an increasing number of watertreatment plants have decided to rely on the use of activatedcarbon-based processes. These processes allow to removedissolved biodegradable substrates that promote bacterialregrowth in distribution systems. Others may also elect touse these processes for the reduction of chlorine demand,precursors of chlorinated disinfection by-products, taste andodours and algal toxins.

Initially developed for their adsorption capacity, gran-ular activated carbon (GAC) filters are now often operatedas biological activated carbon (BAC) filters in order toreduce operation and maintenance costs.[1] More recently,alternative processes combining low-pressure membranefiltration with biological powder activated carbon (PAC)have also been proposed as an alternative to BAC/GACfilters.[2] The objective of these biological processes is tofavour the development of a biofilm hosting a consortiumof heterotrophic and nitrifying bacteria acclimated to the

∗Corresponding author. Email: celine.stoquart@polymtl.ca

biodegradation of organic carbon and to the oxidation ofammonia into nitrates. However, both biodegradation andadsorption mechanisms occur simultaneously within acti-vated carbon-based processes used for organic carbon andmicropollutants removal.[3–5] Discriminating the respec-tive contributions of biotic and non-biotic mechanisms inthese processes is crucial to assess the proper fate of pol-lutants, particularly in the case of compounds adsorbablebut also amenable to biological oxidation. For instance,Wang et al. [5] found that adsorption accounted for 70%of the overall removal of microcystine-LR in BAC filters,while microcystine-LA was mainly biodegradated (60%).Such information allows the development of mathemati-cal models and the optimization of operational strategies,and ultimately the enhancement of the process. Yet, therelative importance of biotic and non-biotic phenomenaremains unknown. Consequently, abiotic controls need tobe included in any scientific research. However, this is achallenging task as it is difficult to neutralize the biologi-cal activity of activated carbon samples without modifyingtheir adsorptive characteristics.

© 2013 Taylor & Francis

Dow

nloa

ded

by [

Eco

le P

olyt

echn

ique

Mon

trea

l] a

t 11:

30 1

8 N

ovem

ber

2013

3092 C. Stoquart et al.

Various methods have been proposed in the scien-tific literature to eliminate microbial activity, especiallywhile studying soil microbiology.[6] However, all the tech-niques tested (e.g. autoclaving, treatment with sodiumazide, gamma irradiation, air drying) altered the chem-ical and physical properties of the samples.[7,8] Whilesodium azide (NaN3) is amongst the most common strategyevaluated, bacteria are not equally impacted by this com-pound. Moreover, adding NaN3 is likely to increase theionic strength of the sample, and thus modify the adsorp-tion of dissolved compounds.[9] Finally, interactions withherbicides were observed, affecting the accurate evaluationof their sorption.[10] Autoclaving has also been evalu-ated as a mean to sterilize BAC samples.[5] However,autoclaving was shown to lead to an important decrease(55%) in the specific surface area of soil samples [9]and an increase in the Freundlich adsorption coefficients[11] of soil samples. Moreover, Berns et al. [12] showedthat autoclaving was responsible for structure modifica-tions of exopolysaccharides, which could potentially beresponsible for a loss of integrity of the biofilm and itssubsequent release from the surface where it is attached.This modification to the PAC surface could lead to thereopening of previously hidden adsorption sites. Finally,while studying fluidized bed reactors, Zhao and collabora-tors [13] irradiated virgin GAC samples with gamma raysin order to sterilize their activated carbon. No significantimpact of the gamma irradiation on the toluene adsorp-tion capacity of their samples was observed at the applieddose (20 kGy). To the best of our knowledge, there is cur-rently no published demonstration of an ideal method thatinhibits the microbial activity on activated carbon with-out impacting its adsorption characteristics (i.e. capacityand kinetics). The least destructive method has thus tobe chosen carefully and depends on the objective of thesterilization. Amongst all techniques, gamma-sterilizationseems thus to emerge as the technique with the highestpotential.

Gamma-sterilization was extensively studied for soilssterilization and appeared to have the lowest impact on soilsamples.[8,12] The gamma irradiation is commonly appliedin the food industry and the health sector for sterilizationpurposes. The gamma rays are high-energy photons com-monly produced by the radioactive decay of 60Co. In watermatrices, energy from photons is transferred to electrons,leading to the ionization of the medium and the generationof highly reactive intermediates such as hydroxyl radicals,protons and hydrated electrons.[14,15] These intermedi-ates inhibit bacterial growth by modifying the structure ofproteins, DNA and RNA which subsequently impacts themetabolic processes.[16,17]

In this study, the gamma irradiation is evaluated as apotential technique to generate an abiotic control for colo-nized activated carbon. First, the selection of the adequatedose of gamma irradiation was based on its ability to steril-ize the PAC (in terms of elimination of heterotrophic plate

counts (HPC)) without affecting its adsorptive behaviour(in terms of methylene blue (MB) adsorption kinetics).Once the appropriate dose of gamma rays was established,the biomass viability and activity on biological PAC werethen evaluated before and after gamma irradiation usingviable counts (i.e. the BacLight™ method) and potentialheterotrophic activity methods (i.e. potential 14C-glucoserespiration rate and potential acetate uptake rate). Theeffect of irradiation on refractory dissolved organic carbon(RDOC) adsorption kinetics was also investigated in orderto confirm the MB results.

Material and methodsPowder activated carbonA wood-based PAC (PicaHydro LP39) was colonized in ahybrid membrane process pilot described in details in a pre-vious paper.[18] The pilot was fed with settled water fromthe Ste-Rose drinking water treatment plant (Laval, Qc).Low-pressure membranes were immersed in a 5 and 10 gdry weight (dw) L−1 PAC suspension. The PAC age wasmaintained respectively at 10 and 60 days, which allowedPAC colonization by heterotrophic and autotrophic bacte-ria. The age of the PAC was maintained stable by purgingdaily a fraction of the PAC and replacing it by the sameamount of virgin PAC.

PAC irradiationThe PAC suspension was sampled from the pilot plant inSte-Rose (Laval, QC, Canada). The suspension was filteredon a paper filter (Whatman #41). Dry weights of the PACcake were evaluated in triplicate according to the 2540 Btechnique of the Standard Methods.[19] The PAC cake wasresuspended in UF filtered (i.e. ultrafiltration) water fromthe pilot plant in order to reach a target concentration of 50 gdw PAC L−1. The concentrated suspension (1 L) was thenexposed to gamma irradiation with incremental doses of5 kGy ranging from 0 to 25 kGy (C-188 60Co source, Under-water Calibrator-15A) in Nordion Inc. Facilities (Laval,QC, Canada).

Evaluation of the biomass viability on PACHeterotrophic plate countsSamples were first homogenized in a blender followingthe method developed by Camper et al.[20] A mass of 1 gof PAC wet weight (ww) as well as a mixture of Zwit-tergent (10−6 mol L−1), EGTA (ethylene glycol tetraaceticacid, 10−3 mol L−1), Tris-buffer (0.01 mol L−1, pH 7.0) andpeptone (0.1%) were mixed in a blender at 16,000 rpm for3 min at 4◦C. Homogenized samples were analysed in dupli-cate on R2A agar (incubated 7 days at 20◦C) using themembrane filtration technique (9215 D) of the StandardMethods.[19] HPC measurements are expressed as Log10CFU (g dw PAC)−1.

Dow

nloa

ded

by [

Eco

le P

olyt

echn

ique

Mon

trea

l] a

t 11:

30 1

8 N

ovem

ber

2013

Environmental Technology 3093

BacLight™ measurementsAn epifluorescence staining method using the LIVE/

DEAD� Bacterial Viability Kit (BacLight™) was applied toqualitatively estimate the physical integrity of the biofilm atthe surface of the PAC. Both BacLight™ stains (i.e. SYTO9 and propidium iodide) were mixed together and kept pro-tected from light at 20◦C. A 3 μL BacLight™ stock solutionwas added to 1 mL of diluted PAC suspension withdrawnfrom the pilot plant. Both stains differ in their spectralcharacteristics and in their ability to penetrate the viablecells. The membrane-permeable SYTO 9 dye labels cellswith green fluorescence, while the membrane-impermeablepropidium iodide labels membrane-compromised bacte-ria with red fluorescence. Samples were incubated in thedark at room temperature for 15 min before observation.The stained sample was filtered through a black 0.2 μmNuclepore polycarbonate filter. The filter was mountedin BacLight™ mounting oil. An Olympus microscopeequipped with a mercury lamp, a 470–490 nm excitation fil-ter, and a 520-nm barrier filter were used. Viable cell countsare expressed as the number of green-stained bacteria pergram of dry activated carbon, while total cell counts areexpressed as the sum of both the green and the red-stainedbacteria per gram of dry activated carbon.

Evaluation of the potential activity of heterotrophicbiomass on PACPotential glucose respiration rateThis method was adapted from the potential glucose respi-ration technique developed by Servais et al. [21] to evaluatethe heterotrophic activity of biomass fixed on the surfaceof GAC from biological filters. This method is based onthe measurement of the production of 14CO2 due to themineralization at 20◦C of 14C-glucose added at a saturat-ing concentration. The adaptation of the method lies in thefact that 1 g ww of PAC sample was used to evaluate thepotential activity of the biomass (as opposed to 2 cm3 ofGAC in the original method). The amount of glucose min-eralized was evaluated in triplicate and expressed in nmol14C-glucose respired (g dw PAC)−1 h−1.

Potential acetate uptake rateThis method evaluates the maximum uptake rate of acetateachieved by the active biomass fixed on PAC. A sampleof PAC (10 g ww) was incubated at 20◦C on an orbitalshaker (approximately 150 rpm) in 500 mL of Milli-Q wateramended with an excess acetate solution (15 mg Na-acetateL−1) and nutrients to maintain a C:N:P ratio of 100:10:1.Preliminary assays had confirmed that acetate uptake wasindependent of initial acetate concentration (i.e. zero orderkinetics) when the latter was added in excess. The deple-tion of acetate was monitored for 24 h by sampling 40 mLof the PAC suspension. Samples were immediately filteredon a 0.45 μm PES membrane previously rinsed with 1 L of

Milli-Q water. Acetate concentration was determined by ionchromatography (Dionex Corporation, Sunnyvale, USA).The apparatus was equipped with an IonPac AS18 col-umn and suppressed conductivity detection. The EluGen�

EGC-KOH was used as eluent. Acetate uptake rates werecalculated by linear regression on the first hours of the assayfor colonized PAC in order to reach a linear regression andfor 24 h for virgin PAC. Specific potential acetate uptake(PAU) rates are expressed in μmol of acetate consummed(g dw PAC)−1 h−1.

Evaluation of the adsorption kineticsMB adsorption kineticsMB has often been used as a non-biodegradable surro-gate [22] for organic matter in adsorption studies.[23] A10−5 mol L−1 solution of MB was prepared in Milli-Qwater. Since pH influences MB adsorption,[24] pH wasset to 7.0 using a 1 mol L−1 phosphate buffer. A mass of0.0064 g ww PAC was resuspended in 200 mL of the MBsolution and maintained at 20◦C under constant agitationat approximately 150 rpm. Subsamples were collected atincreasing contact times (i.e. 1, 5, 20, 40, 60 min and thenevery hour) over a period of 7 h and were immediately cen-trifugated at 9633g for 5 min (Heraeus Pico 17, ThermoElectron Corp., Karlsruhe, Germany). MB concentrationwas measured in the supernatant using spectrophotometryat 665 nm.

RDOC adsorption kineticsRDOC adsorption kinetics were monitored in PAC batchreactors including either 1 or 10 g dw PAC L−1. Repli-cate assays were conducted using both settled and rawwaters from the Ste-Rose drinking water treatment plant(Laval, QC). The depletion of RDOC concentration wasmonitored over 60 min at 20◦C by sampling 125 mL of thePAC suspension at increasing contact times (i.e. 1, 5, 10,15, 30 and 60 min). Samples were immediately filtered ona 0.45 μm PES membrane previously rinsed with 1 L ofMilli-Q water. Dissolved organic carbon (DOC) was mea-sured with a TOC-meter (Sievers 5310 C). BiodegradableDOC (BDOC) analysis was completed using the methodadapted from Servais et al. [25] as described by Markarianet al. [26] RDOC concentrations were obtained by subtract-ing BDOC from DOC measurements and are expressed inmg C L−1.

ModellingA pseudo-second-order model (Equation (1)) was used todescribe the adsorption kinetics [27]:

dqt

dt= ∣

∣k(qe − qt)2 , (1)

where qt is the adsorption capacity (in mg (g dw PAC)−1),qe is the equilibrium adsorption capacity and k (in g dw

Dow

nloa

ded

by [

Eco

le P

olyt

echn

ique

Mon

trea

l] a

t 11:

30 1

8 N

ovem

ber

2013

3094 C. Stoquart et al.

PAC mg −1 h−1) is the pseudo-second-order rate constant.After integration and linearization, the parameters k andqe can be obtained from the intercept and slope of theplot of (t/qt) against t.[27] Indeed, the pseudo-second-order kinetic model has allowed to describe accuratelythe adsorption of metal ions, dyes, herbicides, oils, andorganic substances from aqueous solutions.[28] The mainassumption of this model is that chemisorption (i.e. sorp-tion involving sharing or exchange of electrons betweenadsorbent and adsorbate) is the rate-limiting step. Since theadsorption capacity and the rate constant can be determinedeasily from the equation without any prior parameter esti-mation, it is nowadays the expression most commonly usedfor modelling adsorption kinetics.[29]

Results and discussionProper inhibition of the biomass activity on PAC requires todetermine the minimum and maximum doses of irradiationthat can be applied on the samples. The minimum dose isthe dose of gamma rays required to suppress the biomassactivity in the samples. The maximum dose of irradiation isdefined as the maximum dose of gamma rays that could beapplied without modifying the adsorptive properties of thesample. Our research hypothesis is that such range of dosesexists (i.e. biological activity can be minimized withoutsignificantly interfering on adsorption).

HPC and MB adsorption kinetics were the methodsselected to assess the impact of gamma rays (doses rangingfrom 0 to 25 kGy) on the culturability and the adsorptivebehaviour of PAC samples, respectively. These methodswere selected based on their efficiency in terms of cost,time and feasibility. Results gathered enabled to determinerapidly the adequate range of dose recommended to producean abiotic control of colonized PAC. In order to validate therecommended range of gamma ray doses, new samples weresubmitted to irradiation in this range, and activity-basedmethods (i.e. potential glucose respiration (PGR) and PAUrates) were used to confirm the impact of irradiation on thebiomass activity. In parallel, the removal kinetics of RDOCwas monitored in order to confirm the impact of irradiationon the adsorption of the natural organic matter (NOM).

Selection of the appropriate dose of gamma irradiationMinimum dose of gamma raysHPC enumerations were performed on PAC colonized fora 60-day period and irradiated with doses ranging from0 to 25 kGy. Results given in Table 1 demonstrate thata dose of at least 10 kGy is required to inhibit the for-mation of colonies on R2A agar. A dose of 10 kGy ishigher than the 0.35 kGy value previously suggested forachieving 4 log10 units of Escherichia coli inactivationin axenic cultures.[14] However, in that case, the condi-tions of irradiation were completely different. Only onetype of microorganism (E. coli) was irradiated in a suspen-sion saturated in dissolved oxygen. The oxygen saturatingconditions,[14] the absence of shielding effect [8] and thehomogeneity of the bacteria population could explain themuch lower dose required for sterilization. On the oppo-site, when irradiating samples with a complexity closer tothat of colonized activated carbon (e.g. soil samples), thedose found here appears to be consistent with the scien-tific literature. When applying a dose of gamma irradiationcomprised between 10 and 25 kGy on soil samples, mostbacterial groups were found to be no longer culturable.[8]

Maximum dose of gamma raysMB adsorption kinetics was studied using irradiated 60-daycolonized PAC (0–25 kGy). The mass of MB adsorbed permass of adsorbent (i.e. qt) was calculated and expressedin mg MB (g dw PAC)−1. These results were modelledaccording to a pseudo-second-order model (Equation (1);R2 > 0.96) in order to assess the adsorption capacity at equi-librium (i.e. qe) and the kinetic constant (i.e. k). Results areshown in Figure 1. For all gamma doses investigated, thePAC adsorption capacity (i.e. qe) was not influenced bythe irradiation (i.e. no significant linear regression betweenqe and the irradiation dose was observed, p > 0.05). qe-values fluctuated randomly in the range of 290–360 mgMB (g dw PAC)−1. On the other hand, the MB adsorptionkinetics were noted to increase with the dose of irradiation(p = 0.024). It should be noted that the k-value correspond-ing to a 10 kGy dose of irradiation appeared to be an outlier.

Table 1. HPC and total and viable cell counts (using the BacLightTM technique) on colonized60-day PAC exposed to gamma rays doses ranging from 0 to 25 kGy.

Total cell counts Total cell counts Viable cell countsDose applied (HPC) (Log10 CFU (BacLightTM) (Log10 (BacLightTM) (Log10(kGy) (g dw PAC)−1) bact (g dw PAC)−1) bact (g dw PAC)−1)

0 7.5 10.5 10.35 3.8 10.5 10.310 <LD 10.5 10.215 <LD 10.5 10.120 <LD NA NA25 <LD 10.5 10.2

Note: LD = 200 CFU (g dw PAC)−1, NA = Not Available.

Dow

nloa

ded

by [

Eco

le P

olyt

echn

ique

Mon

trea

l] a

t 11:

30 1

8 N

ovem

ber

2013

Environmental Technology 3095

Figure 1. Pseudo-second-order kinetics constants k and qeobtained for MB adsorption on 60-day PAC exposed to gammarays doses ranging from 0 to 25 kGy.

Consequently, this result will not be considered for the restof the discussion. While a 5 kGy dose did not affect thek-value, irradiating the sample with a dose of 15 kGy ledto a 1.5-fold increase in the MB adsorption rate (k). Thisobservation suggests that high doses will modify the sur-face properties in such a way that diffusion of MB withinactivated carbon is increased. Potential release of biofilmswhich were blocking adsorption sites could explain suchobservation. Figure 2 depicts the predicted percentages ofMB removal for a dose of gamma rays ranging from 0 to25 kGy in function of the contact time. Predictions of thepotential removal of MB are presented for 1 h, the high-est realistic contact time considered in the PAC contactorof a hybrid membrane process. These predictions wererealized by considering a pseudo-second-order adsorptionkinetic, the absence of modification of the adsorption capac-ity (i.e. the average value of 319 mg MB (g dw PAC)−1 fromFigure 1 was used for qe) and by using the k-values shown inFigure 1. Curves shown in Figure 2 demonstrate that while a5 kGy dose virtually left unchanged the percentage of MBremoved in 1 h, irradiating with a 15 kGy dose increasedthe amount of MB removed by 8% for the same contacttime. Doses superior to 15 kGy impacted the amount of MBremoved in 1 h by increasing its adsorption by more than10%. The results of a previous study [13] demonstratedthat a slightly higher dose (20 kGy) than that applied inour study do not modify the capacity of toluene adsorp-tion of a virgin GAC sample. The absence of biofilm onthe GAC tested could explain the discrepancy between thedoses proposed in both studies. In conclusion, doses inferiorto 15 kGy should be targeted in order to limit the impact ofthe irradiation on the adsorption kinetics of colonized PAC.

Impact of irradiation on biomass viability and activity oncolonized PACAlthough the HPC measurements demonstrated thatirradiation was an effective method to suppress the ability

Figure 2. Predicted percentages of MB removal for a dose ofgamma rays ranging from 0 to 25 kGy in function of the con-tact time. Predictions were based on pseudo-second-order kineticmodel for 1 h maximum. The 10 kGy curve was excluded.

of bacteria to multiply on a solid culture medium (Table 1),additional information was required to confirm that biomassmetabolic activity had also been significantly inhibited bythis treatment. Therefore, the impact of gamma irradiationon the activity of heterotrophic biomass was evaluated usingmore advanced methods such as the PGR and PAU rates.Figure 3 indicates that gamma irradiation (dose of 13 kGy)reduced by 83% the PGR rate of colonized PAC whilethe PAU rate was reduced by 71%. For this dose, HPCmeasurements were below detection limit (i.e. <200 CFU(g dw PAC)−1). The HPC as well as the PGR and PAUrates were also evaluated on new PAC (i.e. non-colonizedPAC) (Figure 3). These measurements are used as con-trols to demonstrate the impact of irradiation on the activityof the heterotrophic biomass. Submitting colonized PACto gamma irradiation decreased drastically its biologicalactivity by preventing the multiplication of heterotrophicbacteria and by hindering the oxidation of easily assimil-able compounds such as glucose and acetate. However, aresidual potential activity of the biomass was measured withboth PAU and PGR methods.

There was no visible effect of the gamma irradiation ofthe 60-day PAC sample on the viable (p = 0.11) and total(p = 0.30) cell counts expressed in Log10 bact per g dwPAC obtained with the LIVE/DEAD BacLight™ method(Table 1). Indeed, radiolysis products of gamma irradiationare strong oxidants that damage DNA by inducing singleand double-strand breaks, base-free sites and by alteringthe DNA bases.[30] Thus, this result suggests that no sig-nificant cell lysis occurred during the irradiation treatment.In addition, the epifluorescence observations (Figure S1,Supplementary Information) suggest that gamma rays didnot physically alter the biofilm as no major detachment ofbiomass was observed in the conditions tested (dose of15 kGy). This is a desirable characteristic for the abioticcontrol of colonized activated carbon as it is hypothe-sized that biofilm detachment may influence the diffusionof NOM inside the PAC particles by increasing the amountof available adsorption sites.

Dow

nloa

ded

by [

Eco

le P

olyt

echn

ique

Mon

trea

l] a

t 11:

30 1

8 N

ovem

ber

2013

3096 C. Stoquart et al.

Figure 3. Measurements of bacterial abundance and poten-tial heterotrophic biomass activity on virgin (non-colonized),non-irradiated 60-day colonized PAC and irradiated 60-day col-onized PAC (dose of 13 kGy). The methods used were: HPC inLog10 CFU (g dw PAC)−1, PGR in nmol (g dw PAC)−1 h−1, PAUin μmol (g dw PAC)−1 h−1. Standard deviations are representedwith the whiskers bars.

Impact of irradiation on the adsorption of the organicmatter by PACThe impact of the irradiation on the PAC adsorptive charac-teristics was further studied by monitoring RDOC adsorp-tion kinetics. The latter were monitored during laboratoryassays conducted with concentrations of 1 and 10 g dw PACL−1 in settled (coagulated with alum) and raw waters using10- and 60-day colonized PAC irradiated (13 kGy) and notirradiated (Table 2). As RDOC can only be removed byPAC adsorption, these results confirmed the conclusionsobtained with MB at doses between 10 and 15 kGy: nostatistically significant differences were observed for theparameters evaluated by the pseudo-second-order model(p-values for qe and k were 0.74 and 0.40, respectively).It should be noted that qe-values were increased by anorder of magnitude when using 10 times less PAC. Thisphenomenon can be explained by the fact that during the

assays, the adsorption kinetics was probably limited by theamount of adsorbable RDOC rather than by the amount ofadsorbant. When using a smaller concentration of PAC, itappears that its usage is maximized. By lowering the con-centration of the PAC in the reactors, absolute removal wasdecreased but potential of the PAC was better exploited.In addition, k-values were shown to vary on a logarithmicscale depending on water type and PAC ages. Therefore, thisobservation supports the idea that the 1.5-fold variation ink-value observed earlier for MB adsorption is in fact a lowvariation on a logarithmic scale (0.18 log). Finally, mod-elling the impact of the irradiation on the efficiency for theRDOC removal allowed to confirm the previous results. Forall the operating conditions given in Table 2, the impact ofthe irradiation on RDOC removal after 1 h of contact timewas inferior to 0.11 mg C L−1, which corresponds to theusual accuracy of the RDOC measurements. In summary,the adsorption kinetics of MB and RDOC on PAC weremarginally impacted by gamma irradiation (i.e. at dosesof 10–15 kGy). This is in agreement with a previous studyrecommending gamma irradiation as a reliable sterilizationtechnique for soils.[9] These authors concluded that gammairradiation had a minimal effect on the size distribution ofsoil particles and on their BET surface area.

In this study, the optimal dose of irradiation was com-prised between 10 and 15 kGy. This range of doses allowedto reduce the active biomass but did not entirely inhibitits metabolic activity. However, the adsorption kineticsdemonstrated that applying higher doses of gamma rayswould have led to an increased adsorption of NOM and MB.

Our results demonstrated that the irradiation dose shouldremain as low as possible in order to limit its effect onthe physical and chemical properties of the activated car-bon. However, among microbial populations, resistance togamma beams varies and resistant populations are also metin non-extreme environments.[17,31] Furthermore, wateralkalinity can act as a scavenger of radiolysis products,[14]thereby modifying the dose of irradiation necessary for theinactivation of a given sample. Sample size also influencesthe dose required (e.g. a shielding effect occurs when irradi-ating large samples). Consequently, it is advised to validatethe dose of irradiation for each new project involving the

Table 2. Kinetics constants calculated for RDOC adsorption on PAC with non-irradiated andirradiated PAC (13 kGy) of 10- and 60-day in settled (SW) and raw (RW) water.

qe qe,irradiated k kirradiated

Conditions (mg RDOC (g dw PAC)−1) Log10 (g dw PAC (mg RDOC)−1 h−1)

10 g 60-day PAC L−1, SW 0.02 0.06 3.1 3.610 g 60-day PAC L−1, RW 0.36 0.38 2.1 2.310 g 10-day PAC L−1, SW 0.13 0.1 2.6 2.910 g 10-day PAC L−1, RW 0.46 0.41 2.1 2.31 g 60-day PAC L−1, SW 0.44 0.73 2.0 1.31 g 10-day PAC L−1, SW 1.31 0.81 1.2 1.7

p-value 0.73 0.40

Dow

nloa

ded

by [

Eco

le P

olyt

echn

ique

Mon

trea

l] a

t 11:

30 1

8 N

ovem

ber

2013

Environmental Technology 3097

inhibition of activated carbon heterotrophic activity. Theuse of MB adsorption and PAU method is recommended toevaluate the optimal dosage of gamma rays as the formeris a simple method that provided information coherent withNOM adsorption while the latter, although more complexthan HPC, provided a better evaluation of the heterotrophicactivity on the colonized PAC. Future work should eval-uate if the optimal dose range (10–15 kGy) identified forheterotrophic biomass could be extended to the case of theautotrophic biomass fixed onto activated carbon. Finally,irradiating PAC samples remains costly and its accessibil-ity is not guaranteed. Drawbacks of less expansive methods(e.g. autoclaving) have been observed in soil samples. How-ever, comparing their efficiency on PAC samples could beof great interest to potentially reduce spending.

AcknowledgementsThe authors would like to acknowledge J. Mailly for her help inthe implementation of the 14C-glucose method. This work wascompleted at the NSERC Industrial Chair in Drinking Water ofÉcole Polytechnique de Montréal with the financial support of itspartners, namely the City of Montreal, John Meunier Inc., and theCity of Laval.

References[1] Prévost M, Laurent P, Servais P, Joret J-C. Biodegradable

organic matter in drinking water treatment and distribution.1st ed. Denver, CO: American Water Works Association;2005.

[2] Stoquart C, Servais P, Bérubé P, Barbeau B. Hybrid mem-brane processes using activated carbon treatment for drink-ing water production: a review, J Membr Sci. 2011;411–412:1–12.

[3] Persson F, Heinicke G, Hedberg T, Hermansson M, Uhl W.Removal of geosmin and MIB by biofiltration – an investiga-tion discriminating between adsorption and biodegradation.Environ Technol. 2007;28:95–104.

[4] Stoquart C, Servais P, Carrière A, Gadbois A, BarbeauB. Biological PAC: could it replace BAC filters? Amer-ican Water Works Association-Water Quality TechnologyConference, Phoenix, Arizona, USA, 2011.

[5] Wang H, Ho L, Lewis DM, Brookes JD, Newcombe G.Discriminating and assessing adsorption and biodegradationremoval mechanisms during granular activated carbon filtra-tion of microcystin toxins. Water Res. 2007;41:4262–4270.

[6] Trevors J. Sterilization and inhibition of microbial activityin soil. J Microbiol Methods. 1996;26:53–59.

[7] Wolf D, Dao T, Scott H, Lavy T. Influence of steriliza-tion methods on selected soil microbiological, physical, andchemical properties. J Environ Qual. 1989;18:39–44.

[8] McNamara NP, Black HIJ, Beresford NA, Parekh NR.Effects of acute gamma irradiation on chemical, physical andbiological properties of soils. Appl Soil Ecol. 2003;24:117–132.

[9] Lotrario JB, Stuart BJ, Lam T, Arands RR, O’Connor OA,Kosson DS. Effects of sterilization methods on the physi-cal characteristics of soil: implications for sorption isothermanalyses. Bull Environ Contam Toxicol. 1995;54:668–675.

[10] Chefetz B, Stimler K, Shechter M, Drori Y. Interactions ofsodium azide with triazine herbicides: effect on sorption tosoils. Chemosphere. 2006;65:352–357.

[11] Shaw LJ, Beatonb Y, Glover LA, Killham K, Meharg AA.Re-inoculation of autoclaved soil as a non-sterile treatment

for xenobiotic sorption and biodegradation studies. Appl SoilEcol. 1999;11:217–226.

[12] Berns A, Philipp H, Narres HD, Burauel P, Vereecken H,Tappe W. Effect of gamma-sterilization and autoclavingon soil organic matter structure as studied by solid stateNMR, UV and fluorescence spectroscopy. Eur J Soil Sci.2008;59:540–550.

[13] Zhao X, Hickey RF, Voice TC. Long-term evaluation ofadsorption capacity in a biological activated carbon fluidizedbed reactor system. Water Res. 1999;33:2983–2991.

[14] Thompson J, Blatchley E, III. Gamma irradiation for inacti-vation of C. parvum, E. coli, and Coliphage MS-2. J EnvironEng. 2000;126:761–768.

[15] Taghipour F. Ultraviolet and ionizing radiation for microor-ganism inactivation. Water Res. 2004;38:3940–3948.

[16] Kent. Method for sterilizing products with gamma radiation.United States Patent – 5362442. 1994, pp. 1–3.

[17] Thornley MJ. Radiation resistance among bacteria. J ApplMicrobiol. 1963;26:334–345.

[18] Léveillé S, Carrière A, Charest S, Barbeau B. PAC membranebioreactor as an alternative to biological activated carbonfilters for drinking water treatment. J Water Supply Res T.2013;62:23–24.

[19] American Public Health Association (APHA), AmericanWater Works Association (AWWA), and Water EnvironmentFederation (WEF). Standard methods for the examinationof water and wastewater. 21st ed. Washington, DC: Amer-ican Public Health Association, American Water WorksAssociation, and the Water Environment Federation; 2005.

[20] Camper AK, LeChevallier MW, Broadaway SC, McFetersGA. Evaluation of procedures to desorb bacteria from gran-ular activated carbon. J Microbiol Methods. 1985;3:187–198.

[21] Servais P, Billen G, Ventresque C, Bablon GP. Microbialactivity in GAC filters at the Choisy-le-Roi treatment plant.J Am Water Works Assoc. 1991;83:62–68.

[22] Ferreira-Leitão VS, de Carvalho MEA, Bon EPS. Lignin per-oxidase efficiency for methylene blue decolouration: com-parison to reported methods. Dyes Pigments. 2007;74:230–236.

[23] Perry L, Essex D, Giess P, Graham N, Kaur K, Lambert S.Spencer C. Improving the performance of granular activatedcarbon (GAC) via pre-regeneration acid treatment. WaterEnviron J. 2005;19:159–166.

[24] Wang S, Zhu ZH, Coomes A, Haghseresht F, Lu GQ.The physical and surface chemical characteristics of acti-vated carbons and the adsorption of methylene blue fromwastewater. J Colloid Interface Sci. 2005;284:440–446.

[25] Servais P, Anzil A, Ventresque C. Simple method for deter-mination of biodegradable dissolved organic carbon in water.Appl Environ Microbiol. 1989;55:2732–2734.

[26] Markarian A, Carrière A, Dallaire P-O, Servais P, BarbeauB. Hybrid membrane process: performance evaluation ofbiological PAC. J Water Supply Res T. 2010;59:209–220.

[27] Tsai WT, Chang CY, Ing CH, Chang CF. Adsorption of aciddyes from aqueous solution on activated bleaching earth. JColloid Interface Sci. 2004;275:72–78.

[28] Ho Y-S. Review of second-order models for adsorptionsystems. J Hazard Mater. 2006;136:681–689.

[29] Wu F-C, Tseng R-L, Huang S-C, Juang R-S. Characteris-tics of pseudo-second-order kinetic model for liquid-phaseadsorption: a mini-review. Chem Eng J. 2009;151:1–9.

[30] Henner WD, Grunberg SM, Haseltine WA. Sites and struc-ture of gamma radiation-induced DNA strand breaks. J BiolChem. 1982;57:1750–11754.

[31] Daly MJ. Engineering radiation-resistant bacteria forenvironmental biotechnology. Curr Opin Biotechnol.2000;11:280–285.

Dow

nloa

ded

by [

Eco

le P

olyt

echn

ique

Mon

trea

l] a

t 11:

30 1

8 N

ovem

ber

2013