Bioresource Technology 101 (2010) 126–130

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Compounds interaction on the biodegradation of acetone and methyl ethylketone mixture in a composite bead biofilter

Wu-Chung Chan *, Tzu-Yu LaiCivil Engineering Department, Chung-Hua University, Hsinchu 30067, Taiwan, ROC

a r t i c l e i n f o a b s t r a c t

Article history:Received 12 May 2009Received in revised form 2 August 2009Accepted 3 August 2009Available online 28 August 2009

Keywords:AcetoneMethyl ethyl ketoneBiodegradationCompounds interactionComposite bead biofilter

0960-8524/$ - see front matter � 2009 Elsevier Ltd. Adoi:10.1016/j.biortech.2009.08.001

* Corresponding author. Tel.: +886 3 5186725; fax:E-mail address: wcchan@chu.edu.tw (W.-C. Chan)

Compounds interaction on the biodegradation of acetone and methyl ethyl ketone (MEK) mixture in acomposite bead biofilter was investigated. The biodegradation rate of two compounds in the exponentialgrowth phase and stationary phase for the single compound and two compounds mixing systems wasdetermined. The microbial growth rate and biochemical reaction rate of biodegraded two compoundswas inhibited at higher compound inlet concentration for the single compound system. The microbialmetabolic activity of biodegraded acetone in the microbial growth process and biochemical reaction pro-cess was inhibited by introducing MEK and was more pronounced at higher MEK inlet concentration andlower acetone inlet concentration for the two compounds mixing system. The maximum eliminationcapacity of acetone and MEK for the single compound system was smaller and greater than those forthe two compounds mixing system, respectively.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

The removal of volatile organic compounds (VOCs) from a pol-luted air stream using biofiltration process is highly efficient andhas low installation and operation/maintenance costs. A sphericalpolyvinyl alcohol (PVA)/peat/KNO3/GAC composite bead was pre-pared and was proven suitable as a filter material in the biofiltra-tion process in our previous works (Chan and Lin, 2006; Chanand Peng, 2008). Acetone, methyl ethyl ketone (MEK), methyl iso-butyl ketone (MIBK) and methyl isopropyl ketone (MIPK) arewidely used industrial chemicals. These ketone compounds weredesigned high-priority toxic chemicals. Large volumes of these ke-tone compounds are released into the atmosphere during manu-facturing processes every year, leading to endanger the airquality and public health.

Acetone biodegraded in trickle-bed biofilters packed with plexi-glass chips was more stable than that packed with coconut fiberdue to the plexiglass chips packed bed has greater specific surfacearea, porosity and durability. The maximum elimination capacitywas 59.46 g-C/h-m3 (Przybylska et al., 2006). Isopropyl alcohol ispreferred substrate in the isopropyl alcohol and acetone mixtures.The elimination capacity of isopropyl alcohol and acetone in-creased but the removal efficiency decreased with increased influ-ent loading (Chang and Lu, 2003). Biofiltration of gas-phase solventmixtures during intermittent loading in a fungal biofilter wasfound that shutdown had no noticeable adverse affect on removal

ll rights reserved.

+886 3 5372188..

of n-butyl acetone while removal of MEK and methyl propyl ketone(MPK) was impacted for a few hours, and toluene removal was ad-versely impacted for a few days (Moe and Qi, 2004). N-butyl ace-tate was the most favorable substrate followed by MEK, tolueneand then o-xylene in a coal based biotrickling filter for the removalof paint solvent mixture (Mathur and Majumder, 2008). Althoughthe lag phase for MIBK removal was shorter than that for MEK re-moval, the biodegradation rate of MEK was significantly higherthan that of MIBK. Both MEK and MIBK biodegradation rates werereduced as MEK was pulsed into the air stream containing MEKand MIBK mixtures, but both rates were not influenced as MIBKwas pulsed. MIBK showed the most inhibitive effect on the degra-dation of the base feed of MEK when hexane, acetone, 1-propanoland MIBK were introduced in a step pulse manner to biofilterdegrading MEK (Deshusses and Hamer, 1993; Deshusses et al.,1996; Deshusses, 1997). Using backwashing and starvation/stag-nant strategies, long-term stable removal efficiency over 99% wasretained for MEK and MIBK in a biotrickling filter packed with pel-letized diatomaceous earth. The maximum elimination capacity ofMEK and MIBK was 5.82 kg COD/day-m3 (60.6 g C/h-m3) and4.82 kg COD/day-m3 (53.2 g C/h-m3), respectively. The pseudo-first-order rate constant decreased with an increase the VOCsvolumetric loading rate (Cai et al., 2004, 2005). The biofilter perfor-mance was evaluated under interchanging the feed VOCs in thesequence MEK, toluene, MIBK and styrene, and then back toMEK. The biofilter easily acclimated to the oxygenated compounds(MEK and MIBK), but re-acclimation was delayed for the aromaticcompounds (toluene and styrene). The destructed aromatic com-pounds were eliminated exclusively by aerobic biodegradation;

W.-C. Chan, T.-Y. Lai / Bioresource Technology 101 (2010) 126–130 127

however, the destructed oxygen compounds were eliminated byaerobic biodegradation and possible denitrification (Cai et al.,2006). MEK and MIBK were completely removed in the upper 3/8media depth. While biofilter depth utilization for the removal ofstyrene and toluene increased with increase of influent concentra-tions. The elimination capacity of styrene, MEK and MIBK was notsignificantly affected by the presence of the other VOCs, however,the removal of toluene was significantly affected. Toluene contentin the mixtures played a major role in the biofilter over perfor-mance (Cai et al., 2007). MEK biodegradation by Pseudomonas sp.KT-3 was suppressed by the addition of MIBK or acetone in liquidculture due to the solubility of acetone is higher than that of MEK,and MIBK and MEK compete with each other for enzyme binding(Lee et al., 2006).

Acetone and MEK are commonly mixed together in the effluentstream of the semi-conductor and optic-electronic manufactures.We had indicated that the exponential growth and stationaryphases were important for controlling the removal efficiency ofbiofilter, and studies the biodegradation kinetic behaviors of theexponential growth and stationary phases for single compoundin a composite bead biofilter (Chan and Lin, 2006; Chan and Peng,2008). However, details of the biodegradation kinetic behaviors oftwo compounds mixture in biofilter are scant in the relevant liter-ature. This article investigates the biochemical kinetic behaviors ofacetone and MEK mixtures in a composite bead biofilter. The com-posite bead is the spherical PVA/peat/KNO3/GAC composite bead.The effect of compound inlet concentration and compounds inter-action on the microbial growth rate and biochemical reaction rateare studied. The inlet concentration of acetone and MEK variesfrom 50 to 300 ppm, respectively.

Fig. 1. The variations of (j) kg, MEK/kg, Acetone with MEK inlet concentration, and (d)kg, Acetone/kg, MEK with acetone inlet concentration: (- - -) single compound system,(—) two compounds mixing system.

2. Methods

Peat (industrial grade from KekkilaOyj, Tuusula, Finland) wasdried at 105 �C before use. It has a dry density of 90 kg/m3, a pHof 5.5, a pore volume of 96%, and an organic substance content of91%. Boric acid, sodium monobasic phosphate, sodium dibasicphosphate, potassium nitrate, methyl ethyl ketone and acetone(extra pure grade from Union Chemical, Hsinchu, Taiwan) wereused as received. Poly(vinyl alcohol) (PVA) powder (industrialgrade from Chung Chun Petrochemical, Hsinchu, Taiwan) andgranular activated carbon (GAC) (industrial grade from TaipeiChemical, Hsinchu, Taiwan) were also used as received.

The procedures for preparing PVA/peat/GAC/KNO3 compositebeads and the apparatus and operation of the biofilter system weredescribed in our previous works (Chan and Lin, 2006; Chan andPeng, 2008). Acetone and methyl ethyl ketone were used as VOCs.Acetone and methyl ethyl ketone were individually poured intoeach Erlenmeyer flask. Before packing, the filter material was im-mersed in 0.384 M KNO3 aqueous solution to adsorb KNO3 and toreach equilibrium (approximately 12 h). The bead moisture con-tent was humidified to more than 1.5 g water/g dry compositebead and the seeding was performed with activated sludge ob-tained from the sludge thickener of an industrial wastewater plant.The inlet gas stream was designed to single compound system andtwo compounds mixing system. The single compound system wasonly one compound at a desired inlet concentration in the gasstream. The two compounds mixing system was one compoundat a fixed concentration mixed with another compound at a variousconcentration in the gas stream. Each desired inlet VOCs concen-tration was obtained by adjusting the amount of evaporated VOCsusing mass flow controllers and it was maintained at this concen-tration during the period of biofilter operation. The gas flow ratewas maintained at 0.102 m3/h for all experiments and conse-quently the empty bed residence time (EBRT) of biofilter column

was 28 s. As the stationary phase had maintained more than3 days, the biofilter operation was stopped according to the varia-tions of the removal efficiency of VOCs. Then, new filter materialwas repacked and following the operation procedures describedas above to carry out another the desired inlet concentrationexperiment. The VOCs concentration in the inlet and exit airstreams of each section was taken by auto-sampling and analyzedusing gas chromatography (GC) (Model GC-8A from Shimadzu, To-kyo, Japan). The VOCs removal efficiency was calculated by the dif-ference of the VOCs concentration between the inlet and exit gasstreams. The relative standard deviation and relative error of theexperimental measurements were less than 2% and 5%,respectively.

3. Results and discussion

The variations of VOCs removal efficiency with operation timefor the single compound system and two compounds mixture sys-tem are appeared in lag phase (phase I), exponential growth phase(phase II) and stationary phase (phase III) three phases. Only thebiochemical kinetic behaviors in the exponential growth phaseand stationary phase was studied in this work.

3.1. Microbial growth process

In the exponential growth phase (phase II), the microbialgrowth rate increased exponentially and was represented by thefollowing equation (Valsaraj, 1995; Chan and Peng, 2008).

ln ðC=C0Þ ¼ �kgt ð1Þ

where, C0 is the concentration of VOCs in the inlet air stream. A plotof ln (C/C0) versus t should correspond to a straight line and kg canbe determined. The microbial growth rate kg of two compounds forthe single compound system and two compounds mixing system atvarious inlet concentrations was calculated from the data in phase IIand Eq. (1).

The variations of microbial growth rate of MEK to acetone(kg, MEK/kg, Acetone) ratio values with MEK inlet concentration forthe acetone inlet concentration maintains at 100 ppm and theMEK inlet concentration varies from 50 to 300 ppm is shown inFig. 1. The kg, MEK/kg, Acetone ratio values of the single compoundsystem decreased from 0.791 to 0.198 as the MEK inlet concen-tration increased from 50 to 300 ppm. An increase in the inletconcentration generally would enhance the transfer rate of theVOCs from the gas phase to the biofilm. This phenomenon leads

Fig. 2. The plot of (C0 � C)/ln (C0/C) versus h/ln (C0/C) for single compound system:(d) acetone, (j) MEK.

128 W.-C. Chan, T.-Y. Lai / Bioresource Te

more microorganisms participating in the biodegradation activity.However, high concentrations of some recalcitrant VOCs may pro-duce inhibitive effects on the metabolic activity of the microbialpopulation (Leson and Winer, 1991). Therefore, the result indicatedthat the microbial metabolic activity of biodegraded MEK wasinhibited and the inhibitive effect resulting from increasing theMEK inlet concentration predominated at higher MEK inlet con-centration for the single MEK system.

The kg, MEK/kg, Acetone ratio values of the two compounds mixingsystem increased from 0.822 to 2.374 as the MEK inlet concentra-tion increased from 50 to 300 ppm. However, this phenomenonwas opposite to that in the single compound system. The resultsindicated that the kg, Acetone values decreased with increasing theMEK inlet concentration and the degree of kg, Acetone values de-creased resulting from increasing the MEK inlet concentrationwas greater than that of kg, MEK values decreased resulting fromincreasing the MEK inlet concentration. Therefore, the microbialmetabolic activity of biodegraded acetone would be inhibited byintroducing MEK and the inhibitive effect resulting from com-pounds interaction was more pronounced at higher MEK inlet con-centration for the two compounds mixing system.

The variations of microbial growth rate of acetone to MEK(kg, Acetone/kg, MEK) ratio values with acetone inlet concentrationfor the acetone inlet concentration varies from 50 to 300 ppmand the MEK inlet concentration maintains at 100 ppm is shownin Fig. 1. The kg, Acetone/kg, MEK ratio values of the single compoundsystem decreased from 1.906 to 0.570 as the acetone inlet concen-tration increased from 50 to 300 ppm. The result indicated that themicrobial metabolic activity of biodegraded acetone was alsoinhibited and the inhibitive effect resulting from increasing theacetone inlet concentration predominated at higher acetone inletconcentration for the single acetone system.

The kg, Acetone/kg, MEK ratio values of the two compounds mixingsystem decreased from 0.921 to 0.262 as the acetone inlet concen-tration increased from 50 to 300 ppm. This phenomenon is thesame as that in the single compound system. The result indicatedthat the kg, Acetone value also decreased with increasing the acetoneinlet concentration for the two compounds mixing system. Theslope of the linear profiles in this concentration range for the singlecompound system and the two compounds mixing system were5.623 � 10�3 ppm�1 and 2.489 � 10�3 ppm�1, respectively. The re-sult indicated that the inhibitive effect for acetone resulting fromincreasing the acetone inlet concentration in the single compoundsystem was more pronounced than that in the two compounds sys-tem. The kg, Acetone/kg, MEK ratio values of the two compounds mix-ing system was smaller than that of the single compound systemand the difference of the kg, Acetone/kg, MEK ratio values betweentwo systems was decreased from 0.985 to 0.308 as the acetone in-let concentration increased from 50 to 300 ppm. The result indi-cated that the microbial metabolic activity of biodegradedacetone would be inhibited by introducing MEK and the inhibitiveeffect resulting from compounds interaction was more pronouncedat lower acetone inlet concentration for the two compounds mix-ing system.

3.2. Biochemical reaction process

In the stationary phase, the population of viable cells was at arelatively constant value. The earliest and commonly used biofil-tration model under steady state condition was proposed byOttengraf. Three basic situations of Ottengraf’s model was first-order kinetics, zero-order kinetics with reaction limitation andzero-order kinetics with diffusion limitation (Ottengraf and vande Oever, 1983; Ottengraf, 1986). The corresponding equationsexpressed the rates of biochemical reaction for each situation asfollows:

1. First-order kinetic

ln ðC=C0Þ ¼ �k1h ð2Þ

chnology 101 (2010) 126–130

2. Zero-order kinetic with reaction limitation

C0 � C ¼ k0h ð3Þ

3. Zero-order kinetic with diffusion limitation

1� ðC=C0Þ1=2 ¼ kdh ð4Þ

where, kd is the rate coefficient of zero-order kinetic with diffusionlimitation (Yang and Allen, 1994).

The substrate utilization rate by microbial was generally ex-pressed by the Michaeilis–Menten relationship. Under the steadystate of microbial population, three possible situations may beencountered in a biochemical reaction system (Yang and Allen,1994): situation (1) if the substrate concentration was very low(Ks� C0), the reaction rate expression could be simplified tofirst-order kinetic; situation (2) if the substrate concentrationwas very high (Ks� C0), the reaction rate expression could be sim-plified to zero-order kinetic; situation (3) if the substrate concen-tration C0 was comparable with Ks, the reaction rate expressioncould not be simplified and it was followed fractional-order ki-netic, and the Ottengraf’s diffusion limiting model was found tobe the most approximate expression.

In order to verify the biochemical reaction kinetic model, as-sume there was a plug air flow in the biofilter column and the fol-lowing equation was derived from the Michaelis–Menten equation(Valsaraj, 1995).

ðC0 � CÞ=ln ðC0=CÞ ¼ Vmðh=lnðC0=CÞÞ � Ks ð5Þ

where Ks is half-saturation constant and Vm is maximum reactionrate. A plot of (C0 � C)/ln (C0/C) versus h/ln (C0/C) should correspondto a straight line, and Ks and Vm can be determined. The plot of(C0 � C)/ln (C0/C) versus h/ln (C0/C) for single compound system isshown in Fig. 2. The calculated Ks values for acetone and MEK were30.04 and 32.43 ppm, respectively. The Vm values for acetone andMEK were 12.34 and 15.74 ppm/s, respectively. The C0/Ks valuesfor acetone and MEK were found to be 1.67–9.99 and 1.54–9.25,respectively. The results indicated that the relationship of C0 andKs was not corresponding to situation 1 or situation 2, and it wascorresponding to situation 3 for two compounds. Therefore, theconcentration C0 was comparable with Ks, and zero-order kineticwith diffusion limitation was regarded as the most adequate bio-chemical reaction kinetic model in this study. The kd value of twocompounds in the single compound system and the two

W.-C. Chan, T.-Y. Lai / Bioresource Technology 101 (2010) 126–130 129

compounds mixing system at various inlet concentrations was cal-culated from the data in phase III and Eq. (4).

The variations of biochemical reaction rate of MEK to acetone(kd, MEK/kd, Acetone) ratio values with MEK inlet concentration forthe acetone inlet concentration maintains at 100 ppm and theMEK inlet concentration varies from 50 to 300 ppm is shown inFig. 3. The kd, MEK/kd, Acetone ratio values of the single compoundsystem decreased from 1.155 to 0.258 as the MEK inlet concentra-tion increased from 50 to 300 ppm. The result indicated that thebiochemical reaction rate of biodegraded MEK was also inhibitedand the inhibitive effect resulting from increasing the MEK inletconcentration predominated at higher MEK inlet concentrationfor the single MEK system.

The kd, MEK/kd, Acetone ratio values of the two compounds mixingsystem increased from 0.894 to 1.545 as the MEK inlet concentra-tion increased from 50 to 300 ppm. However, this phenomenon isopposite to that in the single compound system. The result indi-cated that the kd, Acetone values decreased with increasing theMEK inlet concentration and the degree of kd, Acetone values de-creased resulting from increasing the MEK inlet concentrationwas greater than that of kd, MEK values decreased resulting fromincreasing the MEK inlet concentration. Therefore, the biochemicalreaction rate of biodegraded acetone would be inhibited by intro-ducing MEK and the inhibitive effect resulting from compoundsinteraction was more pronounced at the higher inlet MEK concen-tration for the two compounds mixing system.

The variations of biochemical reaction rate of acetone to MEK(kd, Acetone/kd, MEK) ratio values with acetone inlet concentrationfor the acetone concentration varies from 50 to 300 ppm and theMEK concentration maintains at 100 ppm is shown in Fig. 3. Thekd, Acetone/kd, MEK ratio values of the single compound system de-creased from 1.237 to 0.668 as the acetone concentration increasedfrom of 50 to 300 ppm. The result indicated that the biochemicalreaction rate of biodegraded acetone was also inhibited and theinhibitive effect resulting from increasing the acetone inlet concen-tration predominated at higher acetone inlet concentration for thesingle acetone system.

The kd, Acetone/kd, MEK ratio values of the two compounds mixingsystem decreased from 0.762 to 0.303 as the acetone concentrationincreased from 50 to 300 ppm. This phenomenon is the same asthat in the single compound system. The result indicated that thekd, Acetone values also decreased with increasing the concentrationof acetone for the two compounds mixing system. The slope ofthe linear profiles in this concentration range for the single com-pound system and two compounds mixing system were

Fig. 3. The variations of (j) kd, MEK/kd, Acetone with MEK inlet concentration, and (d)kd, Acetone/kd, MEK with acetone inlet concentration: (- - -) single compound system,(—) two compounds mixing system.

2.221 � 10�3 ppm�1 and 1.744 � 10�3 ppm�1, respectively. The re-sult indicated that the inhibited effect resulting from increasingthe acetone inlet concentration in the single compound systemwas more pronounced than that in the two compounds mixing sys-tem. The kd, Acetone/kd, MEK ratio values of the two compounds mix-ing system was smaller than that of the single compound systemand the difference of the kd, Acetone/kd, MEK ratio values betweentwo systems was decreased from 0.475 to 0.365 as the acetone in-let concentration increased from 50 to 300 ppm. The result indi-cated that the biochemical reaction rate of biodegraded acetonewould be inhibited by introducing MEK and the inhibitive effectresulting from compounds interaction was more pronounced atlower acetone inlet concentration for the two compounds mixingsystem.

3.2.1. Elimination capacityElimination capacity and load were calculated according to

equations presented below (Deviney et al.,1999):

EC ¼ QðC0 � CÞ=V ð6Þ

Load ¼ QC0=V ð7Þ

where Q is the flow rate of inlet air steam and V is the bed volume offilter material as packed. The relationship of elimination capacity(EC) of biofilter versus load for the single compound system isshown in Fig. 4. The maximum elimination capacity of acetoneand MEK were 50.5 and 42.2 g-C/h-m3 bed volume, respectively.The result indicated that the maximum elimination capacity of ace-tone was greater than that of MEK for the single compound system.

The maximum elimination capacity of acetone and MEK were22.6 and 59.9 g-C/h-m3 bed volume, respectively, for the two com-pounds mixing system with the acetone concentration maintainsat 100 ppm and the MEK concentration was in the range of 50 to300 ppm. The result indicated that the maximum EC of MEK in thismixing system (equals 59.9 g-C/h-m3 bed volume) was greaterthan that in the single MEK system (equals 42.2 g-C/h-m3 bed vol-ume) in the same MEK inlet concentration range. The maximumelimination capacity of acetone and MEK were 35.9 and 41.4 g-C/h-m3 bed volume, respectively, for the two compounds mixing sys-tem with the MEK concentration maintains at 100 ppm and theacetone concentration was in the range of 50–300 ppm. The resultindicated that the maximum EC of acetone in this mixing system(equals 35.9 g-C/h-m3 bed volume) was less than that in the singleacetone system (equals 50.5 g-C/h-m3 bed volume) in the sameacetone inlet concentration range. In conclusion, the maximum

0 100 200 300 400 500 600 7000

10

20

30

40

50

60

EC

, g C

/h-m

3 com

posi

te b

ead

Load, g C/h-m3 composite bead

Fig. 4. The variations of elimination capacity (EC) with load for single compoundsystem biofilters: (j) acetone ketone, (d) methyl ethyl ketone, (- - -) 100% removal.

130 W.-C. Chan, T.-Y. Lai / Bioresource Technology 101 (2010) 126–130

EC of acetone and MEK for the two compounds mixing systemwere smaller and greater than that for the single compound sys-tem, respectively.

4. Conclusions

This work was to test the biodegradation of acetone and methylethyl ketone (MEK) individually and as mixtures at various concen-trations and to investigate the interaction between two com-pounds during their biodegradation in a composite bead biofilter.The microbial metabolic activity of biodegraded acetone in expo-nential growth phase and stationary phase was inhibited as MEKwas introduced and this inhibitive effect was more pronouncedat higher MEK inlet concentration and lower acetone inlet concen-tration for the two compounds mixing system.

Acknowledgements

The authors wish to thank the Chung Hua University of theRepublic of China for financial aid through the project, CHU-97-S-001.

References

Cai, Z., Kim, D., Sorial, G.A., 2004. Evaluation of trickle-bed air biofilter performancefor MEK removal. J. Hazard Mater. B114, 153–158.

Cai, Z., Kim, D., Soria, L.G.A., 2005. Removal of methyl isobutyl ketone fromcontaminated air by trickle-bed air biofilter. J. Environ. Eng. 131 (9), 1322–1329.

Cai, Z., Kim, D., Soria, L.G.A., Saikaly, P., Zein, M.M., Oerther, D.B., 2006. Performanceand microbial diversity of a trickle-bed air biofilter under interchangingcontaminants. Eng. Life Sci. 6 (1), 37–42.

Cai, Z., Kim, D., Soria, L.G.A., 2007. A comparative study in treating two VOCmixtures in trickle bed air biofilters. Chemosphere 68, 1091–1097.

Chan, W.C., Lin, Z.Y., 2006. A process to prepare a synthetic filter material containingnutrients for biofiltration. Bioresour. Technol. 26, 223–230.

Chan, W.C., Peng, K.H., 2008. Biodegradation of methyl ethyl ketone and methylisopropyl ketone in a composite bead biofilter. Eng. Life Sci. 8, 167–174.

Chang, K., Lu, C., 2003. Biofiltration of isopropyl alcohol and acetone mixtures by atrickle-bed air biofilter. Process Biochem. 39, 415–423.

Deshusses, M.A., 1997. Transient behaviors of biofilters: start-up, carbon balance,and interactions between pollutants. J. Environ. Eng., 563–568.

Deshusses, M.A., Hamer, G., 1993. The removal of volatile ketone mixtures from airin biofilters. Bioprocess Eng. 9, 141–146.

Deshusses, M.A., Hamer, G., Dunn, I.J., 1996. Transient-state behaviors of a biofilterremoving mixtures of vapors of MEK and MIBK from air. Biotechnol. Bioeng. 49,587–598.

Deviney, J.S., Deshusses, M.A., Webster, T., 1999. Biofiltration for Air PollutionControl. Lewis Publishing Inc., New York.

Lee, T.H., Kim, J., Kim, M.J., Ryu, H.W., 2006. Degradation characteristic of methylethyl ketone by Pseudomonas sp. KT-3 in liquid culture and biofilter.Chemosphere 3, 315–322.

Leson, G., Winer, A.M., 1991. Biofiltration: an innovative air pollution controltechnology for VOC emissions. J. Air Waste Manage. Assoc. 41, 1045–1054.

Mathur, A.K., Majumder, C.B., 2008. Biofiltration and kinetic aspects of a biotricklingfilter for the removal of paint solvent mixture laden air stream. J. Hazard Mater.152, 1027–1036.

Moe, W.M., Qi, B., 2004. Performance of a fungal biofilter treating gas-phase solventmixtures during intermittent loading. Water Res. 38, 2259–2268.

Ottengraf, S.P.P., 1986. Exhaust gas purification. In: Rehm, H.J., Reed, G. (Eds.),Biotechnology, vol. 8. VCH-Verlagsgesellschaft, Weinheim, Germany.

Ottengraf, S.P.P., van de Oever, A.H.C., 1983. Kinetics of organic compound removalfrom waste gases with a biological filter. Biotechnol. Bioeng. 25, 3089–3102.

Przybylska, K.P., Zieminski, K., Szopa, J.St., 2006. Acetone biodegradation in atrickle-bed biofilter. Inter. Biodeterior. Biodegrad. 57, 200–206.

Valsaraj, K.T., 1995. Elements of Environmental Engineering: Thermodynamics andKinetics. Lewis Publishing Inc., New York.

Yang, Y., Allen, E.R., 1994. Biofiltration control of hydrogen sulfide. 2. Kinetics,biofilter performance, and maintenance. J. Air Waste Manage. Assoc. 44, 1315–1321.