Intimate coupling of photocatalysis and biodegradation in a photocatalytic circulating-bed biofilm...

ARTICLE

Intimate Coupling of Photocatalysis andBiodegradation in a PhotocatalyticCirculating-Bed Biofilm Reactor

Michael D. Marsolek,1 Cesar I. Torres,2 Martina Hausner,2 Bruce E. Rittmann2

1Department of Chemical Engineering, Northwestern University, Evanston,

Illinois 60208; telephone: 206-296-5527; fax: 206-296-2173; e-mail: [email protected] of Civil and Environmental Engineering, Northwestern University,

Evanston, Illinois

Received 6 November 2007; revision received 8 February 2008; accepted 4 March 2008

Published online 19 March 2008 in Wiley InterScience (www.interscience.wiley.com)
. DOI 10.1002/bit.21889
ABSTRACT: Coupling advanced oxidative pretreatmentwith subsequent biodegradation demonstrates potentialfor treating wastewaters containing biorecalcitrant andinhibitory organic constituents. However, advanced oxida-tion is indiscriminate, producing a range of products thatcan be too oxidized, unavailable for biodegradation, or toxicthemselves. This problem could be overcome if advancedoxidation and biodegradation occurred together, anorientation called intimate coupling; then, biodegradableorganics are removed as they are formed, focusing thechemical oxidant on the non-biodegradable fraction.Intimate coupling has seemed impossible because the con-ditions of advanced oxidation, for example, hydroxylradicals and sometimes UV-light, are severely toxic tomicroorganisms. Here, we demonstrate that a novel photo-catalytic circulating-bed biofilm reactor (PCBBR), whichutilizes macro-porous carriers to protect biofilm from toxicreactants and UV light, achieves intimate coupling. Wedemonstrate the viability of the PCBBR system first withUV only and acetate, where the carriers grew biofilm andsustained acetate biodegradation despite continuous UVirradiation. Images obtained by scanning electron micro-scopy and confocal laser scanning microscopy show bacterialiving behind the exposed surface of the cubes. Second, weused slurry-form Degussa P25 TiO2 to initiate photocatalysisof inhibitory 2,4,5-trichlorophenol (TCP) and acetate. Withno bacterial carriers, photocatalysis and physical processesremoved TCP and COD to 32% and 26% of their influentlevels, but addition of biofilm carriers decreased residuals to

Michael D. Marsolek’s present address is Department of Civil and Environmental

Engineering, Seattle University, Seattle, WA 98122.

Cesar I. Torres’s present address is Center for Environmental Biotechnology,

Biodesign Institute, Arizona State University, Tempe, AZ 85287-5701.

Martina Hausner’s present address is Department of Chemistry and Biology, Ryerson

University, Toronto, Ontario, Canada M5B 2K3.

Bruce E. Rittmann’s present address is Center for Environmental Biotechnology,

Biodesign Institute, Arizona State University, Tempe, AZ 85287-5701.

Correspondence to: M.D. Marsolek

Contract grant sponsor: National Science Foundation

Contract grant number: BES-0402510

Contract grant sponsor: Institute for Environmental Catalysis (IEC) at Northwestern

University

� 2008 Wiley Periodicals, Inc.

2% and 4%, respectively. Biodegradation alone could notremove TCP. Photomicrographs clearly show that biomassoriginally on the exterior of the carriers was oxidized(charred), but biofilm a short distance within the carrierswas protected. Finally, we coated TiO2 directly onto thecarrier surface, producing a hybrid photocatalytic-biologicalcarrier. These carriers likewise demonstrated the conceptof photocatalytic degradation of TCP coupled with bio-degradation of acetate, but continued TCP degradationrequired augmentation with slurry-form TiO2.

Biotechnol. Bioeng. 2008;101: 83–92.

� 2008 Wiley Periodicals, Inc.

KEYWORDS: advanced oxidation; biodegradation; biofilm;photobiocatalysis; photocatalysis; trichlorophenol

Introduction

Waters contaminated with toxic and biologically recalcitrantorganics require new treatment technologies (Rodriguezet al., 2005; Scott and Ollis, 1995). One emerging approachis to integrate chemical and biological methods by com-bining an advanced-oxidation process (AOP) with bio-degradation (Bandara et al., 1997; Sarria et al., 2002). UsingAOP to partially transform the recalcitrant compounds intobiodegradable products allows cost-effective and robusttreatment. Prior research in this area has focused onsequentially coupled systems, or those that have the chemicaland biological treatments in separate stages (Hong andZeng, 2002; Parra et al., 2002).

In the ideal case of sequential coupling, advancedoxidation is controlled so that the recalcitrant organicsare only transformed to the point that they are rapidlybiodegradable, at which time they are passed along to thebiodegradation stage. Chemical transformation beyondthis point wastes oxidant and increases costs with no

Biotechnology and Bioengineering, Vol. 101, No. 1, September 1, 2008 83

further benefit. However, since advanced oxidants areindiscriminate and fast acting, transformation only to thispoint is difficult. Instead, a range of products is typicallyproduced, many of which can be toxic themselves, toooxidized, or unavailable for biodegradation. These productcharacteristics can make it difficult to produce an optimizedchemical effluent for subsequent biodegradation. Further-more, the indiscriminate nature of the hydroxyl-radicalreaction can render it inefficient for pre-treating wastewatersthat are largely biodegradable, but contain only a smallfraction of toxic compounds, since much of the oxidantwould be spent reacting with already biodegradableorganics.

In principle, advanced oxidation and biodegradationcould overcome the limits of sequential coupling if thebacteria were present during advanced oxidation and couldbiodegrade the rapidly degradable products as soon as theywere formed. We call this approach intimate coupling. Then,bacteria in close proximity to advanced oxidation couldremove biodegradable products originally present or as theyare produced, focusing the chemical oxidant only on thebiologically recalcitrant compounds. In the past, intimatecoupling has seemed impossible, because the reactantsused in advanced oxidation (e.g., O3, H2O2, Fenton’sreagent, HO. free radical, and UV light) are severely toxic tobacteria, with photocatalysis being used to inactivate or killcells (Belhacova et al., 1999; Blake et al., 1999; Rincon andPulgarin, 2004).

Here, we investigate a novel system that can, for thefirst time, achieve the ideal of intimate coupling inUV-photolysis and UV-photocatalysis settings. Intimatecoupling is achieved by using a photo-catalytic circulating-bed biofilm reactor, or PCBBR, which is an extension of pastwork with a circulating-bed biofilm reactor, or CBBR, usedsuccessfully to degrade volatile organics including toluene

Figure 1. The PCBBR system utilizes macro-porous cellulosic carriers that protect mi

coupling possible. The ESEM image shows the porosity and size of the pores. The orientatio

half of the reactor volume, resulting in a large emission of high-intensity light per unit ar

84 Biotechnology and Bioengineering, Vol. 101, No. 1, September 1, 2008

(Sang et al., 2003; Yu et al., 2002) and benzene, toluene, andp-xylene (BTX) (Yu et al., 2001a,b,c). In the CBBR, air isused to create air-lift circulation of water and biofilmcarriers, while simultaneously providing oxygen for aerobicmetabolism. The CBBR employs a cellulosic macro-porouscarrier that has nearly the same density as water andallows biofilm accumulation in its macro-pores. Thecarriers’ density allows them to circulate with the water,providing a high-turbulence environment with good mixingand external mass transport. Having the bacteria accumulatein the macro-pores protects them from chemical toxicityand loss by abrasion.

The CBBR system is ideal for adapting into the PCBBRsystem, shown in Figure 1, because the microorganismsshould be protected from all kinds of toxic species in bulksolution, as long as the toxic species is reactive (Sang et al.,2003). Accumulation of bacteria within the macro-porouscarrier sets up concentration gradients that relieve inhibi-tion by biodegradable toxins, such as toluene, within thebiofilm (Sang et al., 2003). Hydroxyl free radicals fromphotocatalysis are highly reactive and also should beeliminated inside the macro-porous carriers.

The biofilm in the PCBBR accumulates inside the macro-porous carriers, while the AOP takes place either inbulk solution—with UV light photolysis or slurry-formTiO2-photocatlysis—or on the carrier surface with fixedTiO2-photocatalyst. The bacteria are well protected fromUV light, toxic substrates, free radicals, and physical loss,but are close to the advanced-oxidation reactions so they canimmediately degrade photo-catalytic products that arebiodegradable.

The work we report is the first example of intimatelycoupled UV-biological and photocatalytic-biological sys-tems. A possible precursor example of intimate coupling isthat of Buyuksonmez et al. (1998, 1999), who investigated

croorganisms from toxic reactants (e.g., HO.) and organic compounds, making intimate

n of the circular lamp-head illuminates a disk of area covering approximately the lower

ea.

simultaneous Fenton oxidation with biodegradation fortreatment of perchloroethylene (PCE) and oxalate usingsuspended biomass. They took advantage of the ability ofcertain microorganisms to acclimate to high concentrationsof H2O2 when they are pre-exposed to small non-toxicconcentrations. The approach of Buyuksonmez et al. (1998,1999) allows the Fenton reaction to occur in the presence ofthe microorganisms, but the bacterial suspension would notbe practical in UV-light or UV-light mediated processes andalso cannot be applied to situations in which the targetorganic is present at toxic concentrations. In the PCBBR, themicroorganisms are protected within the interior of thecube from advanced oxidation reactants and from the toxiceffects of the target organic, allowing for application of thistechnology to a wide-range of constituents and removingconcern about toxic byproduct formation.

The model compound for this work, 2,4,5-trichloro-phenol (TCP), is highly recalcitrant and inhibitory(Marsolek and Rittmann, 2007) and is used here inconjunction with sodium acetate as test substrates. TiO2-photocatalysis has been extensively studied for transforma-tion and degradation of a wide variety of toxic organics,including many chlorinated aromatics (Agrios et al., 2003;Choi and Hoffmann, 1997; Stafford et al., 1993, 1997), and ithas been used as a pretreatment in coupled systems forherbicides (Parra et al., 2002) and chlorophenols (Manilalet al., 1992; Shang et al., 2002), but never in an intimatelycoupled system.

Here, we test three scenarios of intimate coupling with aPCBBR: intimately coupled UV-photolysis and biodegrada-tion using acetate, intimately coupled slurry TiO2-photo-catalysis using TCP and acetate, and intimately coupledfixed TiO2-photocatalysis using TCP and acetate. Thesethree scenarios allow us to test the overarching question ofwhether it is possible to intimately couple UV-mediatedAOP with biodegradation, in spite of toxic concentrations oforganics and free radicals in the bulk medium. Furthermore,we explore various methods to employ the titania catalystwith biofilm carriers and interactions between the biofilmmicroorganisms and advanced oxidation.

Materials and Methods

Reactor Set-Up

We used a bench-scale version of the PCBBR illustratedin Figure 1. The liquid volume of the bench-scale reactorwas 122 mL, with roughly equal volumes in the riserand downcomer sides. Circulation was achieved throughbubbling air up the riser, which also provided oxygen foraerobic and photocatalytic oxidation. The air was filtered(gas purifier, Alltech Brand, Grace-Davison, Deerfield, IL)prior to entering the reactor.

The influent was pumped into the reactor, using acalibrated Minipuls 2 pump with Viton pump tubing. In thecase of pumping 0.5 g/L slurry TiO2, the medium was

prepared in 500-mL volumes, exchanged frequently, andmixed vigorously with stirring in a 1.0-L media bottle toensure homogenous dispersion of the catalyst. Hydraulicretention time was 6 h.

The liquid medium consisted of a trace mineral solutionand 10 mM phosphate buffer (Marsolek and Rittmann,2007). TCP and the buffer were filter sterilized and addedaseptically to the autoclaved water, acetate, and traceelement solution, producing the final medium.

Carriers

The biomass carriers, called AQUACEL (Ecomat Inc.,Hayward, CA), were macroporous cellulosic cubes, 4-mmper side. The average porosity of the carriers was 0.93, theirwet density was nearly that of water, 1.04 g/cm3, andtheir average dry weight was 7.4 mg/carrier (Yu et al., 2002).Similar carriers have been used for various applications,including as wet-scrubbers in biotrickling filters (Gabrielet al., 2004), aerobic quinoline degradation (Buchtmannet al., 1997), and for enhanced accumulation of nitrifyingorganisms (Jun et al., 2000).

Inoculum

Biomass was added to the carriers by placing them inactivated sludge inoculum from the Stickney WastewaterTreatment Plant (Calumet City, IL). Biofilm was allowed togrow in and on the carriers for durations of 1 day (TiO2-coated cubes), 2 days (UV experiment), or 10 days (slurry-TiO2 experiment). For the work with carriers studded withTiO2, a 1:5 dilution of the activated sludge was used toensure that any biofilm coating on the surfaces would not betoo dense. For experiments with TCP, the activated sludgewas adapted to the chlorinated aromatic for 7 days at 20 mMprior to adding the carriers. Fifty carriers were added to thePCBBR at the beginning of the UV experiment, while fortywere added for the experiments with TiO2. Five carriers wereperiodically removed to monitor biomass.

Illumination

The reactor was illuminated using a high-intensity UV lamp(UVP-B-100AP, UVP Products, Upland, CA) with a 100-Wlong-wave bulb (21,700 mW/cm2). The emission range is315–400 nm, with a maximum at 365 nm. The bulbdiameter was 6 inches (15 cm), which illuminatedapproximately 30% of the total reactor volume for theorientation used in the UV-photolysis experiments and 50%for the TiO2-slurry and TiO2-coated-cube experiments. Theilluminated area was covered with aluminum foil to reflectlight, maximizing illumination within the reactor, andminimizing extraneous light emitting from the system. Forprotection, the entire reactor was operated in a shieldedenvironment, which was only opened for sampling and toexchange medium.

Marsolek et al.: Photocatalysis and Biodegradation in a PCBBR 85

Biotechnology and Bioengineering

Titanium Dioxide and Its Coating Procedure

Degussa P25 TiO2 was used at 0.5 g/L for slurryphotocatalysis experiments and at 20 g/L for fixing ontothe cellulose carriers. The method used to fix P25 followsthat of Ku et al. (2001). Prior to adding catalyst, 0.1 g ofsurfactant, dioctyl sulfosuccinate (Sigma-Aldrich, St. Louis,MO), was dissolved in 1-L double-deionized water. Then,20 g/L of Degussa P25 was added and mixed overnight(at 130 rpm) on a shaker table, and the cubes were thenadded and shaken with the TiO2 slurry for 24 h. The mixturewas poured into a drying basin, and the slurry was allowed toevaporate onto the cubes. The cubes were baked at 1008C for16 h and rinsed of poorly adhering catalyst by adding themto an Erlenmeyer Flask with double-deionized water andshaking at 130 rpm on a shaker table. We repeatedlyexchanged the rinse water until no more TiO2 was removedfrom the cubes, as monitored by a decrease in cube dry-weight. We determined the mass of TiO2 deposited onto thesurface by comparing to the dry-weight of the cubes prior toadding the catalyst.

Methylene Blue Experiments

Methylene blue oxidation was used to test the photocatalyticactivity of the studded cubes. Methylene blue is ideally suitedfor gauging the effectiveness of TiO2 photocatalysis, as itundergoes a bleaching process in the presence of TiO2/UV,and its concentration is easily monitored by tracking theabsorbance at 660 nm (deTacconi et al., 1997) (Spectronic20 Genesys, Thermo Scientific, Waltham, MA).

We filled a 100-mm� 50-mm crystallization beaker with50 mL of 2.95 mM methylene blue and added 25 studdedcubes. We placed the crystallization dish on a rocking tablelocated directly beneath a high intensity (21,700 mW/cm2)long-wave UV lamp (UVP-B-100AP, UVP Products). Wefiltered the samples and monitored absorbance at 660 nm,calibrated versus a set of standards. Control experiments runwith only methylene blue and UV light showed negligiblemethylene blue degradation; control experiments withmethylene blue and un-studded cubes and UV light showedsome adsorption of methylene blue within the first 2 h, butno continuous degradation.

Sampling

We collected liquid samples from the outlet of the PCBBRreactor, filtered them through a 0.2-mm membrane filter(Whatman #6872-2502, PVDF membrane), and analyzedthem for chemical oxygen demand (COD) and substrateconcentrations. COD was monitored with HACH closed-reflux method COD digester tubes. We used a HitachiD-7000 HPLC system having a Supelco SupelcoSil LC18column (25 cm� 4.6 mm) for HPLC analysis of TCP andhaving a BioRad Aminex 87h column for monitoringacetate. The HPLC elution solvent was 70% pure HPLC-


grade methanol plus 30% acetic acid in water (2%, v/v) forthe C18 column, and dilute phosphoric acid in water(0.04%, v/v) for the Aminex column. We comparedsample peak areas to peak areas of standards to quantifyconcentrations.

We removed cubes periodically from the reactor tomonitor biomass accumulation. Cubes were removed byopening the top section of the PCBBR reactor, removing thescreen, and extracting cubes with a tweezers. The removedcubes were dried and weighed, and the weights werecompared to bare cubes to determine biomass accumula-tion. At the end of the experiment, some cubes were placedin a 10% paraformaldehyde solution for future SEManalysis, and we immediately stained and analyzed othersusing confocal laser scanning microscopy (CLSM). Photo-graphs were taken with a digital camera (Fuji FinepixS3100).

Scanning Electron Microscopy

For cubes with biomass preserved in a paraformaldehydesolution, we dried them using a critical point dryer (Polaroncritical point dryer), coated them with 5 nm of gold(Cressington 208HR Sputter Coater), and visualized themon a Hitachi S-3500N VP SEM. Energy dispersive X-rayspectroscopy (EDS) was performed with a PGT energydispersive X-ray analyzer on the S-3500N. Interiors werevisualized by slicing prior to drying and coating. For cubeswithout biomass (bare cubes, cubes studded with TiO2), weair dried them, coated them with 5 nm of gold, and similarlyvisualized them. We visualized uncoated carriers on an FEIQuanta sFEG environmental SEM (NUANCE facility,Northwestern University).

Confocal Laser Scanning Microscopy

We acquired CLSM images using a Zeiss LSM 510 (CarlZeiss, Jena, Germany). We removed cubes from the reactor,cut them in half if interior images were needed, andimmediately placed in a Live/Dead stain (LIVE/DEADBacLight viability stain; Molecular Probes Inc., Eugene, OR)for 15 min on a rocker table, followed by 15 min of rinsingwith MilliQ water (x2). We then fixed the cubes onto a smallpetri dish using silicone sealant, filled the dish with water,and visualized labeled microbial cells using a water-dippinglens (63�/0.90 NA).

Results and Discussion

UV Photolysis

The over-riding goal of this work is to determine theviability of an intimately coupled process, and to that endthe first step is to test the ability of the PCBBR system toprotect biofilm from the effects of UV irradiation alone. Totest this aspect, we conducted experiments with UV

Figure 2. Acetate degradation in the UV-photolysis PCBBR. The influent acetate

concentration of 5.2 mM and UV illumination were continuous. Acetate was not

removed by photolysis alone. Carriers with biofilm were added at hour 96, after which

acetate was quickly removed and sustained in spite of exposure to UV light. Biomass

initially decreased (expressed as mg dry weight, DW), but regrew over time validating

that bacteria were protected within the carriers.

illumination and acetate. We fed a medium of 5-mM acetateto the PCBBR and, upon establishing a baseline performanceof UV photolysis alone, added 40 cubes with 2 days growthof biofilm. The flow rate was 20 mL/h, giving an HRT ofapproximately 6 h. We monitored acetate in the effluent andbiomass accumulation on the cubes. Finally, we used SEM tovisualize biofilm and CLSM to determine if the biomassinside the carriers was living.

Figure 2 shows the effluent acetate concentration overtime with or without biofilm-containing carriers, as well asthe biomass accumulation on the carriers upon additionto the reactor. UV photolysis alone, up to 96 h, showed noacetate degradation. Once the carriers with biofilm wereadded (hour 96), acetate biodegradation began quickly,increased up to about 200 h, and was sustained to 290 h, orfor over 30 HRTs, with continuous UV illumination.Biomass was initially lost from the carriers, likely dueto shearing and exposure to UV-light, but subsequentlybiomass regrew in correlation with acetate degradation,validating that bacteria were protected within the carriers.

At hour 290, the observed yield was 0.27 g cells DW/gCOD. This low yield for acetate may have been associatedwith the harsh growth environment at the outside of thecubes, where the loss rate from decay or detachment mayhave been high. It may be necessary to have a proportionateincrease in carriers or carrier surface area to compensate forthe losses, if they present a problem during operation.

The biomass and acetate data demonstrate the biomassgrew and sustained itself and metabolic activity in the

Figure 3. (a) SEM images of exterior (top) and interior (bottom) of the cellulose cube

biofilm formation along the walls of the cube, while the image near the exterior indicates that

(b) A projection of 12 single xy CLSM images collected at 2 mm intervals clearly show live c

leaving large pore spaces uninhabited. The image is taken from the center of a carrier after r

of this article, available at www.interscience.wiley.com.]

CBBR under steady UV illumination, but further proof isprovided via SEM and CLSM images, shown in Figure 3.Figure 3a shows that the biofilm was located away fromthe outer surface of the carriers. This confirms the conceptunderlying the PCBBR: that the biofilm grows inside thecarrier, where it is protected from UV irradiation and strongdetachment forces. Figure 3b shows that the biofilm islargely alive and lines the pore walls.

In summary, the UV-photolysis experiment shows thatbacteria inside the macro-porous carriers were alive,

s after 180 h in the UV-photolysis reactor. The interior image (bottom) clearly shows

the exterior face of the cube is free of cells, while a ‘‘ledge’’ behind the face has biofilm.

ells (green) and dead cells (red) following the form of the cellulose walls in the cube,

oughly 180 h in the UV-photolysis reactor. [Color figure can be seen in the online version



growing as a biofilm, and metabolically active, despitecontinuous illumination with UV. The biofilm was pro-tected in the cellulose cubes from toxic irradiation (anddetachment), and it gave substantial substrate removal eventhough the conditions in the bulk liquid were harsh.

Slurry TiO2-Photocatalysis With Acetate and TCP

Photocatalytic conditions challenge the bacteria with toxichydroxyl radicals and UV irradiation, which, together withinhibitory TCP in the influent, confront the biofilm with amore challenging environment than UV-photolysis alone.In this experiment, we continuously applied a 0.5-g/L slurryof TiO2 (Degussa P25) with 2.5 mM acetate and 100 mMTCP, an inhibitory concentration of TCP (Marsolek andRittmann, 2007).

To compare the effectiveness of full intimate coupling ofphotobiocatalysis versus photocatalysis and biodegradationalone, we operated the reactor in four stages: (1) photo-catalysis with UV-light turned on, (2) full PCBBR operationwith simultaneous photocatalysis and biodegradation,(3) light turned off and biodegradation only, and (4) fullPCBBR operation repeated. Figure 4 summarizes the CODand TCP concentrations at the end of each of the variousstages of operation.

Photocatalysis alone (labeled ‘PC only’) showed appreci-able COD and TCP removals due to photocatalysis (�70%of the removal), while air stripping also removed some of thevolatile TCP (20–40% of influent TCP), but COD and TCPresiduals were still 32% and 26%, respectively. Additionof carriers with protected biofilm for full PCBBRoperation, noted as ‘Bio and PC’, decreased COD andTCP to much lower concentrations, giving residuals of 16%and 12%, respectively, during the second phase. Thus, thepresence of biomass resulted in an additional 16% removalof COD and 14% removal of TCP. However, it is unlikelythat the enhanced TCP removal was due to biodegradation,since TCP is highly biorecalcitrant. Thus, the biomassreduced the concentration of easy-to-biodegrade substrates,

Figure 4. COD and TCP concentrations during different operating stages of the

slurry-TiO2 experiment. Full PCBBR operation with simultaneous photocatalysis and

biodegradation showed excellent TCP and COD removal and was reproduced. Error

bars indicate 95% confidence limits. PC, photocatalysis; Bio, biological carriers.


including acetate, as shown through lower COD outputs.This resulted in a higher percentage of the advanced oxidantbeing focused on the TCP and a subsequent lower effluentTCP concentration.

When the light was turned off, eliminating photocatalysis,the TCP concentration quickly and steadily rose to about70 mM in 24 h (4 HRTs), or a residual of 70%. These resultsindicate that photocatalysis was responsible for most of theTCP removal. COD removal initially improved during thisstage, probably due to enhanced biodegradation of acetate inthe absence of toxic UV and free radicals, but the CODbegan to slowly climb in correlation with increasing TCPconcentration. Error bars are not indicated for these data,since the concentrations were still increasing and thereforenot at steady state at the end of this stage, and therefore noconfidence limits could be calculated.

Upon reintroduction of photocatalysis, TCP was quicklyremoved, and residuals of total COD and TCP were onlyabout 4% and 2%, respectively. Here, biodegradationenhanced COD removal by 28% and TCP destruction by22%. Biodegradation is responsible for the enhanced CODremoval, but not the enhanced TCP destruction. Theimproved performance in this stage, versus the first bioand photocatalysis stage, was likely due to greater biomassaccumulation in and on the carriers achieved during the ‘bioonly’ stage, where bacteria were not hindered by UV and freeradicals and had access to a greater fraction of the acetate,since photocatalysis was turned off. This enhanced biomassaccumulation continued to remove a greater fraction ofacetate, thereby focusing more of the chemical oxidationon TCP, resulting in relatively better performance thanthe first fully coupled stage. These results underscore thatUV-driven photocatalysis was necessary for major TCPremoval, biofilm in the carriers was necessary for nearlycomplete COD removal, and photocatalysis and biodegra-dation occurred in an intimately coupled configuration.

The slurry-TiO2 conditions made it impossible tomonitor the biomass concentration by mass difference,since the carriers could capture catalyst particles. However,simple photographs of the cubes show the striking effect ofthe photocatalysis on the carriers and biology. Figure 5shows images of bare cubes versus cubes with biomass; thebiofilm accumulation is noticeable on the left-center image.The two right-most images show the differential effect ofphotocatalysis on the biomass at the end of the slurry TiO2

experiment. The structure of the cubes was not damaged,and they retained their ‘‘spongy’’ make-up, but it is obviousthat the biomass on the outside of the cube was thoroughlyoxidized (i.e., ‘‘charred’’). When the cube was cut (right-most image), the interior still looked identical to the cubesprior to adding them to the reactor, validating that the toxiceffects of hydroxyl radical did not reach deep into the cubeand biomass could be protected within.

In summary, the experiments with a TiO2 slurry showedthat the concepts underlying the PCBBR were validwith photocatalytic production of free radicals and a toxicsubstrate (TCP) present, in addition to UV irradiation. The full

Figure 5. Comparison of a bare carrier (left) and a carrier loaded with biomass (left center), to carriers taken from the PCBBR reactor at the end of the slurry TiO2 experiment

(right center and right). The biomass on the surface (right center) was thoroughly oxidized (‘‘charred’’), but the interior (right) was not charred, illustrating the safe environment for biofilm.

[Color figure can be seen in the online version of this article, available at www.interscience.wiley.com.]

PCBBR system gave nearly complete removals of TCP andCOD, photocatalysis was responsible for quickly removing theTCP while the biomass removed COD, and biofilmaccumulated inside the macro-porous carriers.

Full Intimate Coupling: TiO2-Coated Cubes

In the ideal of intimate coupling, we would like thephotocatalysis to take place as close as possible to the biologicaldegradation. To approach this ideal, titanium dioxide photo-catalyst needs to be coated directly onto the carrier outer surface.Figure 6 shows SEM images of the P25/surfactant coated

Figure 6. SEM images and X-ray analysis of bare carriers (left) and P25/surfactant c

noticeable in the image and corroborated by the strong titanium peak in the EDS. [Color fi

wiley.com.]

carriers, together with EDS analysis that clearly shows thetitanium coating of the cleaned and washed carriers.

The photocatalytic activity of the TiO2-coated carrierswas assayed through methylene blue oxidation. The first-order reaction coefficient was 0.0050/min, and this isequivalent to the first-order reaction rate of slurry TiO2 at aconcentration of 2.1 mg/L for the same experimentalconditions (data not shown). Control experiments with nocubes and uncoated cubes showed no degradation ofmethylene blue. Although the reaction rate was low for thecoated cubes, for example, compared to the 0.5-g/L TiO2

concentration used in the slurry experiments (Fig. 4), theypositively exhibited photocatalytic activity. We therefore

oated carriers (right) after repeated washing. The titanium coating on the surface is

gure can be seen in the online version of this article, available at www.interscience.



progressed to the next stage of the experiment, intimatelycoupled photobiocatalysis using a single photocatalytic-biological carrier, rather than slurry TiO2.

We conducted the PCBBR experiment first with nocarriers, followed by TiO2 studded carriers, and studdedcarriers with biomass. Finally, leaving the studded carrierswith biomass in the reactor, we added 0.2 g/L slurry P25TiO2 at the end to enhance photocatalysis. Influentconditions were 95 mM TCP and 2.5 mM acetate. Airstripping reduced residual TCP to an average of 55 mM, butdid not alter the acetate concentration. The TCP and acetateresults are shown in Figure 7.

As compared to the baseline data (no carriers), the TiO2-coated carriers without biomass gave a small amount ofsustained photocatalytic degradation of TCP. With nocarriers, the residual TCP concentration was 55.4� 2.1 mM,with the decrease due to air stripping. Upon additionof catalyst-coated carriers, the residual TCP decreasedfurther, to 42.3� 1.2 mM. Sorption affects can be ruled out,because TCP-removal did not decrease over time (21.3 h,�3.5 HRTs). This demonstrates that the cubes can bephotocatalytic carriers in the PCBBR. The carriers alonegave no enhanced removal of acetate.

Removing these TiO2 coated carriers and replacingthem with TiO2-coated carriers also containing biofilmshowed sustained TCP removal initially, but TCP removalbegan to diminish; after 20 h, photocatalytic degradation ofTCP was no longer enhanced. Acetate was quickly andconsistently degraded upon addition of coated carriers withbiomass. Thus, biodegradation accounted for completedegradation of 2.5 mM acetate, but no enhanced degrada-tion of TCP. The next stage, addition of slurry-form TiO2,resulted in enhanced TCP degradation with no loss inacetate removal, again showing the unique relationshipbetween the two processes, with photocatalysis degradingTCP, while biomass removes acetate and other easy-to-degrade organics.

Figure 7. Performance of the PCBBR with TiO2 studded carriers. Upon addition

of TiO2 coated carriers, TCP was quickly degraded to lower levels. Addition of

TiO2 coated carriers also inoculated with biomass resulted in quick and complete

degradation of all acetate, but while TCP removal was initially sustained for 6 h (1 HRT)

thereafter TCP destruction diminished until there was no enhanced removal. Addition

of slurry TiO2 once again demonstrated TCP degradation with no loss of acetate

removal capability.


Microbial production of soluble microbial products(SMP) or extracellular polymeric substances (EPS) is aplausible hypothesis explaining the reduction of photo-catalytic activity in the biomass loaded carriers. Bacteria mayproduce SMP and EPS to protect themselves, perhaps byscavenging free radicals. The strong increase in acetateremoval supports that bacteria were active, and activemetabolism generates SMP and EPS (Laspidou andRittmann, 2002a,b). Their ability to scavenge free radicalsis unknown.

CLSM images of TiO2-coated carriers with biomass usedin the full PCBBR, Figure 8, show an edge with a gradientfrom red (dead cells) to green (live cells). Light micrographsof the same region (not shown) indicate a wall along whichthe bacteria are growing. The gradient from red-to-greenalong the edge indicates that some bacteria are protectedfrom toxicity of photocatalytic reactants and TCP, whileother bacteria are not, validating the premise that thecarriers provide an environment capable of protectingthe biofilm even under these harsh conditions.

The results in Figure 7 underscore the importance ofhaving enough TiO2 photocatalytic activity to take fulladvantage of intimate coupling. Clearly, TiO2 photocatalyticactivity was too low with the P25/surfactant carriers alone, asthe residual TCP concentrations remained high. Addition of0.2 g/L of slurry TiO2 showed an increase in TCP removal,while acetate removal was unaffected. While the results withslurry TiO2 reinforce the principle underlying intimatecoupling, adding slurry TiO2 may not be a practical way to

Figure 8. CLSM live/dead image of an edge in a carrier at the end of the full

PCBBR experiment. The graduation from red-to-green (dead to live) shows bacteria

growing along a wall of the carrier, validating the premise that the carriers provide

safe environments for bacteria to grow in spite of AOP and chemical toxicity.

operate an intimately coupled PCBBR. One drawback is thatthe TiO2 would need to be removed from the effluent, a stepincreasing the complexity and cost of the system. Second,slurry-form TiO2 may eventually become trapped in and fillthe carrier pores, decreasing the microbiological efficiencyof the process over time. Therefore, an important goal forfuture research on the PCBBR is obtaining a macro-porouscarrier than can hold more active TiO2 on its outer surface.

This work demonstrates, for the first time, the intimatecoupling of UV-mediated AOP with biodegradation.Biofilm bacteria accumulated and metabolically functionedinside a macro-porous carrier, even though the bulkenvironment was extremely harsh from UV irradiation,hydroxyl free radicals, and toxic TCP. The combinationof TiO2 photocatalysis (from slurry TiO2) and biofilmbiodegradation inside the macro-porous carrier gave by farthe best removals of TCP and COD. Although our attemptto coat the carriers with TiO2 failed to achieve a high densityof active TiO2, the results showed the coated carrierswere photocatalytically active, while the microbial activityinside the carrier remained strong. Putting more TiO2 on thecarrier may achieve results similar to using slurry TiO2, butwithout the problems associated with removing TiO2 fromthe effluent. If its current limitations are overcome, thePCBBR will combine the benefits of the wide-rangingapplicability and robustness of photocatalysis with theefficiency of biodegradation into a single technology,providing a potential solution for treatment of a host ofbiologically recalcitrant and inhibitory wastewaters thatchallenge our conventional treatment methods.

The authors would like to acknowledge the National Science Founda-

tion for supporting this work (BES-0402510), the Institute for Envir-

onmental Catalysis (IEC) at Northwestern University for support, the

NUANCE facility at Northwestern University for electron microscopy

support, and Dr. Kristin Searcy for help in capturing the CLSM image

found in Figure 8.

References

Agrios AG, Gray KA, Weitz E. 2003. Photocatalytic transformation of 2,4,5-

trichlorophenol on TiO2 under sub-band-gap illumination. Langmuir

19(4):1402–1409.

Bandara J, Pulgarin C, Peringer P, Kiwi J. 1997. Chemical(photo-activated)

coupled biological homogeneous degradation of p-nitro-o-toluene-sul-

fonic acid in a flow reactor. J Photochem Photobiol A 111(1–3):253–263.

Belhacova L, Krysa J, Geryk J, Jirkovsky J. 1999. Inactivation of micro-

organisms in a flow-through photoreactor with an immobilized TiO2

layer. J Chem Technol Biotechnol 74(2):149–154.

Blake DM, Maness PC, Huang Z, Wolfrum EJ, Huang J, Jacoby WA. 1999.

Application of the photocatalytic chemistry of titanium dioxide to

disinfection and the killing of cancer cells. Sep Purif Methods 28(1):

1–50.

Buchtmann C, Kies U, Deckwer W-D, Hecht V. 1997. Performance of three

phase fluidized bed reactor for quinoline degradation on various

supports at steady state and dynamic conditions. Biotechnol Bioeng

56(3):295–303.

Buyuksonmez F, Hess TF, Crawford RL, Watts RJ. 1998. Toxic effects of

modified Fenton reactions on Xanthobacter flavus FB7. Appl Environ

Microbiol 64(10):3759–3764.

Buyuksonmez F, Hess TF, Crawford RL, Paszczynski A, Watts RJ. 1999.

Optimization of simultaneous chemical and biological mineralization

of perchloroethylene. Appl Environ Microbiol 65(6):2784–2788.

Choi WY, Hoffmann MR. 1997. Novel photocatalytic mechanisms for

CHCl3, CHBr3, and CCl3CO2-degradation and the fate of photo-

generated trihalomethyl radicals on TiO. Environ Sci Technol 31(1):

89–95.

deTacconi NR, Carmona J, Rajeshwar K. 1997. Reversibility of photoelec-

trochromism at the TiO2/methylene blue interface. J Electrochem Soc

144(7):2486–2490.

Gabriel D, Huub HJC, Deshusses MA. 2004. Conversion of full-scale wet

scrubbers to biotrickling filters for H2S control at publicly owned

treatment works. J Environ Eng 130(10):1110–1117.

Hong PKA, Zeng Y. 2002. Degradation of pentachlorophenol by ozonation

and biodegradability of intermediates. Water Res 36(17):4243–

4254.

Jun BH, Tanji Y, Unno H. 2000. Stimulating accumulation of nitrifying

bacteria in porous carrier by addition of inorganic carbon in a con-

tinuous-flow fluidized bed wastewater treatment reactor. J Biosci

Bioeng 89(4):334–339.

Ku Y, Ma CM, Shen YS. 2001. Decomposition of gaseous trichloroethylene

in a photoreactor with TiO2-coated nonwoven fiber textile. Appl

Catal B 34(3):181–190.

Laspidou CS, Rittmann BE. 2002. A unified theory for extracellular poly-

meric substances, soluble microbial products, and active and inert

biomass. Water Res 36(11):2711–2720.

Laspidou CS, Rittmann BE. 2002. Non-steady state modeling of extracel-

lular polymeric substances, soluble microbial products, and active and

inert biomass. Water Res 36(8):1983–1992.

Manilal VB, Haridas A, Alexander R, Surender GD. 1992. Photocatalytic

treatment of toxic organics in waste-water—Toxicity of photodegra-

dation products. Water Res 26(8):1035–1038.

Marsolek MD, Rittmann BE. 2007. Biodegradation of 2,4,5-trichlorophenol

by mixed microbial communities: Biorecalcitrance, inhibition, and

adaptation. Biodegradation 18(3):351–358.

Parra S, Malato S, Pulgarin C. 2002. New integrated photocatalytic-

biological flow system using supported TiO2 and fixed bacteria

for the mineralization of isoproturon. Appl Catal B 36(2):131–

144.

Rincon AG, Pulgarin C. 2004. Bactericidal action of illuminated TiO2 on

pure Escherichia coli and natural bacterial consortia: Post-irradiation

events in the dark and assessment of the effective disinfection time.

Appl Catal B 49(2):99–112.

Rodriguez SM, Galvez JB, Rubio MIM, Ibanez PF, Gernjak W, Alberola IO.

2005. Treatment of chlorinated solvents by TiO2 photocatalysis and

photo-Fenton: Influence of operating conditions in a solar pilot plant.

Chemosphere 58(4):391–398.

Sang BI, Yoo ES, Kim BJ, Rittmann BE. 2003. The trade-offs and effect of

carrier size and oxygen-loading on gaseous toluene removal perfor-

mance of a three-phase circulating-bed biofilm reactor. Appl Microbiol

Biotechnol 61(3):214–219.

Sarria V, Parra S, Adler N, Peringer P, Benitez N, Pulgarin C. 2002. Recent

developments in the coupling of photoassisted and aerobic biological

processes for the treatment of biorecalcitrant compounds. Catal Today

76(2–4):301–315.

Scott JP, Ollis DF. 1995. Integration of chemical and biological oxidation

processes for water treatment: Review and recommendations. Environ

Prog 14(2):88–103.

Shang NC, Yu YH, Ma HW. 2002. Variation of toxicity during the

ozonation of monochlorophenolic solutions. J Environ Sci Health

Part A 37(2):261–271.

Stafford U, Gray KA, Kamat PV, Varma A. 1993. An in situ diffuse

reflectance FTIR investigation of photocatalytic degradation of

4-chlorophenol on a TiO2 powder surface. Chem Phys Lett 205(1):

55–61.

Stafford U, Gray KA, Kamat PV. 1997. Photocatalytic degradation of 4-

chlorophenol: A mechanistically-based model. Res Chem Intermed

23(4):355–388.



Yu HB, Kim BJ, Rittmann BE. 2001. The roles of intermediates in biode-

gradation of benzene, toluene, and p-xylene by Pseudomonas putida F.

Biodegradation 12(6):455–463.

Yu HB, Kim BJ, Rittmann BE. 2001. A two-step model for the kinetics of

BTX degradation and intermediate formation by Pseudomonas

putida F. Biodegradation 12(6):465–475.


Yu HB, Kim BJ, Rittmann BE. 2001. Contributions of biofilm versus

suspended bacteria in an aerobic circulating-bed biofilm reactor. Water

Sci Technol 43(1):303–310.

Yu HB, Kim BJ, Rittmann BE. 2002. Effects of substrate and oxygen loading

rates on gas-phase toluene removal in a three-phase biofilm reactor.

Water Environ Res 74(3):288–294.

Intimate coupling of photocatalysis and biodegradation in a photocatalytic circulating-bed biofilm...

Documents

Transcript of Intimate coupling of photocatalysis and biodegradation in a photocatalytic circulating-bed biofilm...