Bioresource Technology 146 (2013) 749–753

Contents lists available at ScienceDirect

Bioresource Technology

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

Short Communication

Development of a UBFC biocatalyst fuel cell to generate powerand treat industrial wastewaters

0960-8524/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.biortech.2013.07.065

⇑ Corresponding author at: Department of Food Science and Technology, Facultyof Technology and Community Development, Thaksin University, PhatthalungCampus, Phatthalung 93110, Thailand. Tel.: +66 84 212 1788; fax: +66 74 693 996.

E-mail address: chontisa.s@gmail.com (C. Sukkasem).

Chontisa Sukkasem a,b,⇑, Sunee Laehlah b

a Department of Food Science and Technology, Faculty of Technology and Community Development, Thaksin University, Phatthalung Campus, Phatthalung 93110, Thailandb Research Center in Sustainable Energy and Environment, Thaksin University, Phatthalung Campus, Phatthalung 93110, Thailand

h i g h l i g h t s

� The UBFC system can treat wastewater without chemical or nutrient addition.� COD removal up to 26.7 gCOD/L-d was obtained higher than for existing technologies.� The carbon fiber brush immobilized base increased the performance of the UBFC.

a r t i c l e i n f o

Article history:Received 15 May 2013Received in revised form 12 July 2013Accepted 16 July 2013Available online 20 July 2013

Keywords:Up-flow bio-filter circuitMicrobial fuel cellBiocatalystHigh efficiencyWastewater treatment

a b s t r a c t

Agro-industry wastewaters normally contain high levels of organic matter and require suitable treatmentbefore discharge. The use of Microbial fuel cells, a novel wastewater treatment, can provide advantagesover existing treatment methods. In this study, an up-flow bio-filter circuit (UBFC) for treating wastewa-ters without chemical treatment or nutrient supplement, was developed to solve a clogging problem. Theoptimal conditions included an organic loading rate of 30.0 gCOD/L-d, hydraulic retention time of1.04 day, pH level of 5.6–6.5 and aeration at 2.0 L/min. External resistance of the circuit was tested.COD removal levels of 8.08, 20.1 and 26.67 gCOD/L-d were obtained, while fed with sea food, biodieseland palm oil mill wastewater, respectively. These rates are higher than for conventional technologies.The carbon fiber brush immobilized base increased the performance of the new UBFC by 17.54% over thatobtained in a previous study, while the cost was slightly decreased about 4.48%.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Normally, agro-industry wastewaters contain high levels of or-ganic matter, such as from biodiesel, palm oil mill and sea foodprocessing and hence need to be treated by suitable methods.These include production of biogas, but their use is limited bythe need for special maintenance techniques and the large areafor plant. A novel approach to wastewater treatment is to useMicrobial fuel cells (MFCs) since they can convert organic matterto electricity without combustion, providing a number of advanta-ges over existing methods, e.g. superior efficiency to mineralize or-ganic and inorganic matter, smaller area, less maintenance andhigh stability (Aelterman et al., 2006; Ahn and Logan, 2010; Green-man et al., 2009; Nam et al., 2010; Rabaey et al., 2006; Raghavuluet al., 2009; Rodrigo et al., 2007; Sukkasem et al., 2008; Yi and Har-per, 2009). Therefore, MFC development, when optimized for low

running cost and maximal power generation, could help to reduceenvironmental pollution and be a renewable energy source. Re-cently, biodiesel wastewater and palm oil mill wastewater treat-ment by MFCs has been studied (Feng et al., 2011; Cheng, 2010).

The advantages of MFC applications in wastewater treatmentsystems include higher COD removal, 40–80% compared to 20–50% COD removal by anaerobic treatment for the same retentiontime. Biomass is also 68% higher in this system, 0.07–0.22 g bio-mass COD/gSCOD, but it is 45% lower than for aerobic systems,0.4 g biomass COD/gSCOD (Logan, 2008), which requires lesssludge disposal (Drapcho et al., 2008). The current limits of MFCdevelopment have been described in a previous report (Sukkasemet al., 2011) in which biodiesel wastewater treatment was investi-gated for neutral alkalinity and COD removal by an UBFC. It wascombined with a pre-fermented (PF) and an influent adjusted(IA) procedure. The process could improve the quality of the rawwastewater and obtain a high efficiency of COD removal of15.0 gCOD/L-d. The total investment for UBFC materials wasUS$1775.7/m3 and the total power consumption was 2.28 kW-h/m3. For comparison, the capital cost for anaerobic digesters (AD)is of the order of £100,000 (US$144,000) per ton of COD treated

Fig. 1. Schematic of upflow bio-filter circuit (UBFC) system (Sukkasem et al., 2011).

Ope

Clo

CO

n circ

se circ

D rem

ircuit

emov

uit po

Data

poten

al effi

ten

tentia

fficien

tial (mV)

l at 10

cy

10 kΩ (mV)

GAC

700

270

57

Brush

895

320

67

Efficieniency Enhancem

27.

18.

17.

86%

52%

54%

ent fnt from GAC

Fig. 2. Effect of electrode materials on UBFC performance, operated with biodieselliquor from the PF process at OLR 30.0 gCOD/L-d, HRT 1.04 day, pH 5.6–6.5 andaerated at 2.0 L/min.

750 C. Sukkasem, S. Laehlah / Bioresource Technology 146 (2013) 749–753

per day (Pham et al., 2006) while for the UBFC it is lower at aboutUS$118,380 per ton of treated COD. UBFC systems are suitable forbiodiesel wastewater treatment, however, the problem of bio-par-ticle clogging in the UBFC during operation needs to be addressed.

In this study, a UBFC was developed to solve the clogging prob-lem by changing the immobilized base materials. This was studiedunder different operational conditions to treat various wastewaters.

2. Methods

2.1. A bio-filter circuit system construction

The UBFC system was developed to solve the clogging problemby changing the immobilized base material. The granular activatedcarbon was substituted by carbon fiber brush. Each PF and AF tankwas made from an 8.0 L plastic bin. The UBFC consisted of twoparts: Part (1) two plastic bottles used as UFAF1 and UFAF2 reac-tors and Part (2) two plastic bottles, vertically stacked, used as aBFC reactor (Fig. 1). The total capacity of the UFAF1, UFAF2 andBFC was 4 L. The lower bottle of the BFC was an anaerobic compart-ment called the ‘‘anode’’, and the upper bottle was an aerobic com-partment called the ‘‘cathode’’. As a membrane replacement, a14 cm plastic tube-funnel was inserted between the anaerobicand aerobic compartments to reduce oxygen diffusion into the an-ode. A microorganism immobilized base was prepared by adding555.0 ± 10 g (1.0 L) of inoculated granular activated carbon (parti-cle size 1.0–10.0 mm) in the UFAF1, UFAF2 and BFC anode bottles.The void volume of each bottle was 0.5 L. Also 333.0 ± 7 g (0.6 L) ofinoculated GAC was added to the BFC cathode chamber and purgedwith air via sparkling-holes at the top of the chamber at a rate of2.0 L/min to induce aerobic bacteria growth.

2.2. Pre-fermentation

Typically for wastewaters that contain high levels of fat, oil andgrease (FOG), a high density of suspended solids and strong alkalin-ity are not suitable for microbial growth. These inhibitors are pres-ent in biodiesel wastewater, palm oil mill effluent and sea foodwastewater. To increase the fermentative rate, the wastewaterwas inoculated by 20% (v/v) of activated sludge. After a week, thesupernatant of fermented wastewater was sampled and analyzedfor COD and pH.

2.3. Adjusted influent

The supernatant from the pre-fermentation process, the rawwastewaters, were adjusted to the 30.0 g/L COD concentration (ex-cept for sea food wastewater because the initial concentration was4.5 g/L COD) and their pH neutralized in order to control the influ-ent condition before being fed into the UBFC.

2.4. UBFC inoculation

For the biocatalytic immobilized base electrode preparation,each material was inoculated in activated sludge obtained froman industrial treatment plant (Songkhla, Thailand) at a ratio of 1:1(v/v) for 3 days to immobilize the microorganisms on the electrodesurface as a biocatalyst producer instead of metal catalysts ormediators. Afterwards, the bio-electrode was added into the UBFC.For the start-up process, 6.0 gCOD/L of wastewater influent wascontinuously fed into the UBFC at 1.0 mL/min. The BFC cathodewas aerated at 2.0 L/min to stimulate aerobic microorganisms onthe bio-electrode surface. The UBFC was operated until the outputvoltage stabilized at >0.6 V in an open circuit mode for a week.Then, the circuit was closed by an external resistance of 10 kO. Allexperiments were operated at an ambient temperature of 30 ± 3 �C.

2.5. UBFC operation

The UBFC system was investigated in stages to determine theoptimum conditions. Five external resistances (1.0, 5.0, 10.0, 20.0and 30.0 kO) were applied under controlled conditions: organicloading rate (OLR) 30.0 gCOD/L-d of neutral pH influent, hydraulicretention time (HRT) 1.04 days and aerated at 2.0 L/min. Eachexperiment was carried out for 3 days. The COD concentration ofUBFC and BFC anode influent, as well as the BFC cathode effluentwere analyzed. The UBFC voltage across the external resistor wasmeasured and automatically recorded by a data logger computerprogram (National Instrument, 2008). All the experiments weredone in triplicate.

2.6. Analysis and calculations

Current (I) and power (P) was calculated as Eqs. (1) and (2):

I ¼ V=R ð1Þ

P ¼ V2=R ð2Þ

where V is cell voltage (V) and R is external resistance (O). Volumet-ric current or power is calculated by dividing the current or power

A B

Fig. 3. Effect of external load on % COD removal (A) and volumetric power (B) and at OLR 30 gCOD/L-d, HRT 1.04 day, pH 5.6–6.5, high aerated 2.0 L/min.

ΩΩ

A B

Fig. 4. Effect of external resistance on COD removal efficiency (A) and power generation (B) while the UBFC was fed with palm oil mill wastewater at an OLR of 30 gCOD/L-d,HRT 0.35 day, pH 5.6–6.5, aerated 2.0 L/min.

0102030405060708090

100

1 5 10 20 30 open circuit

COD

rem

oval

(%)

Load (kΩ)

anode cathodeA

0

2

4

6

8

10

12

0 10 20 30 40 50

Pow

er D

ensi

ty (m

W/m

3 )

Current Density (mA/m3)

B

Fig. 5. Effect of external load on % COD removal (A) and volumetric power (B) while the UBFC was fed with sea food wastewater at OLR 8.57 gCOD/L-d, HRT 1.04 day,pH 5.6–6.5, aerated 2.0 L/min.

C. Sukkasem, S. Laehlah / Bioresource Technology 146 (2013) 749–753 751

by the anode volume. HRT and OLR were calculated as described inEqs. (3) and (4):

HRTðdayÞ ¼ ½reactor volumeðLÞ�=½influent flow rateðL=dayÞ� ð3Þ

OLRðgCOD=L-dÞ ¼ ½COD concentrationðgCOD=LÞ�=½HRTðdayÞ� ð4Þ

The COD concentration was analyzed using standard methods(APHA, AWWA, and WPCF 1995). A complete block design (CBD)and Duncan test were used for statistical analysis.

3. Results and discussion

3.1. Start-up period

In this study, the open circuit voltage at the start-up period was0.6 ± 0.1 V. The pH and efficiency of COD removal from each bottle(Fig. 1), UFAF1, UFAF2, BFC anode and BFC cathode, were deter-mined. The results illustrate that the UBFC has a potential to treatCOD with a 10% improvement in a closed circuit mode compared to

Tabl

e1

Com

pari

son

ofM

FCco

mbi

nati

onsy

stem

for

was

tew

ater

trea

tmen

t.

Was

te-w

ater

trea

tmen

tsy

stem

sV

olu

me

(L)

Mic

ro-o

rgan

ism

sM

embr

ane

Cat

alys

tSu

bstr

ates

Cru

dew

aste

-wat

erC

OD

(g/L

)O

LR(g

/L-d

)H

RT

(d)

Pow

erde

nsi

ty(m

W/m

3)

CO

Dre

mov

al(g

/L-d

)R

efer

ence

s

Pre-

ferm

enta

tion

&U

BFC

UB

FC4.

0Sl

udg

eN

oB

ioca

taly

ston

GA

CB

iodi

esel

was

tew

ater

218

±30

30.0

1.04

35.6

220

.1Th

isst

udy

Pre-

ferm

enta

tion

&U

BFC

UB

FC4.

0Sl

udg

eN

oB

ioca

taly

ston

bru

shPa

lmoi

lm

ill

was

tew

ater

50.0

–60.

030

.01.

0411

.59

26.6

7Th

isst

udy

Pre-

ferm

enta

tion

&U

BFC

UB

FC4.

0Sl

udg

eN

oB

ioca

taly

ston

bru

shSe

afo

odw

aste

wat

er3.

0–4.

08.

570.

359.

478.

08Th

isst

udy

MFC

&U

ASB

bM

FC2.

36Sl

udg

eN

oB

ioca

taly

stPa

lmoi

lm

ill

was

tew

ater

504–

5.25

2.0

2.83�

103

5.0

Ch

eng

etal

.(20

10)

Sin

gle

cham

ber

MFC

0.02

8Sl

udg

eN

oPl

atin

um

Dom

esti

cw

aste

wat

er18

254

0.00

912

.8�

103

13.9

3A

hn

and

Loga

n(2

010)

Dou

ble

cham

ber

MFC

1.15

Slu

dge

No

KM

nO

4A

rtifi

cial

was

tew

ater

NA

a6.

00.

516

.8�

103–

70.8�

103

2.42

–3.7

1B

eher

aan

dG

han

grek

ar(2

009)

Sin

gle

cham

ber

MFC

0.25

Act

ive

elec

tro-

chem

ical

bact

eria

No

Plat

inu

mFe

rmen

ted

was

tew

ater

25.9

3.84

1.04

3.0�

103

3.57

Nam

etal

.(20

10)

Sin

gle

cham

ber

MFC

0.25

Slu

dge

Yes

Plat

inu

mD

omes

tic

was

tew

ater

0.2

12.0

–18

.03.

01.

2�

103

0.89

–1.3

4Ji

ang

and

Li(2

009)

Som

eda

taca

lcu

late

dor

mod

ified

from

orig

inal

data

.a

NA

=n

oda

tafr

omor

igin

alda

ta.

bU

ASB

=U

pflow

An

aero

bic

Slu

dge

Bla

nke

t.

752 C. Sukkasem, S. Laehlah / Bioresource Technology 146 (2013) 749–753

in an open circuit mode (normal process). At the end of this pro-cess, the effluent of the system had a neutral pH, which is benefi-cial in downstream processes.

After the circuit was connected, the aqueous phase of biodieselwastewater from the pre-fermentation process was then neutral-ized and fed into the UBFC under controlled conditions of an OLRof 30 gCOD/L-d, HRT of 1.04 day and aeration rate of 2.0 mL/min.The advantage was that this UBFC was able to treat this wastewa-ter without a nitrogen source added, while anaerobic treatment re-quired the addition of urea or yeast extract to reach the optimumC:N ratio.

3.2. Effect of electrode materials on the UBFC performance

In a previous study (Sukkasem et al., 2011), colloidal particlesand biofilm development rapidly caused reactor clogging. At theend of each experiment, the reactor had to be cleaned by backwashing using tapped water.

Substitution of immobilized base GAC (surface area4.39 � 102 m2/g) with iron core carbon fiber brush-£ 2.5 cm ofbrush, length 30 cm with a core of 2 mm twist iron cable to solvethe clogging tendency of the UBFC was investigated. The loosestructure of the iron core fiber brush and high surface area of fiber(1.07 m2/g) is suitable for bacterial immobilization. The sludgeaccumulation was less than that when using GAC over the sameperiod of operation. It promoted better performance at the sameconditions (OLR 30 gCOD/L-d, HRT 1.04 day, pH 5.6–6.5 and aer-ated at 200 ml/min, connecting to 10 kO). After operating for aweek, an open circuit potential was measured at 0.88–0.91 V,which is higher than for the previous material (GAC OCP 0.6–0.8 V). The performance of the new UBFC, using carbon fiber brushimmobilized base, was higher than recorded in the previous study(Sukkasem et al., 2011) by about 27.86% for OCP, 18.52% for theclosed circuit and 17.54% for the COD removal efficiency (Fig. 2).

3.3. Biodiesel wastewater treatment

3.3.1. Effect of external resistance on power generation from biodieselwastewater

The effect of external loads was investigated at 1.0, 5.0, 10.0,20.0 and 30.0 kO, in a stepwise method. The results showed thatthe percentage of COD removal increased from open circuit modeand the highest efficiency was up to 67% (20.1 gCOD/L-d) at anexterior resistance of 10 kO (Fig. 3A). This corresponds to a peakof power on the polarization curve (Fig. 3B). The BFC generatedquite low electricity (35.62 mW/m3 of BFC anode volume). This isbecause of a large internal resistance (10 kO) caused by a low de-gree of biodegradation, by a long distance between electrodes orby material sensitivity: these factors need to be further investi-gated. The cost of new UBFC was slightly decreased by about4.48% from US$1775.7/m3 per 15.0 gCOD removal to US$2273/m3

per 20.1 gCOD removal.

3.4. Palm oil mill effluent treatment

3.4.1. Effect of external resistance on power generation from palm oilmill effluent

The effect of various external resistances (1.0, 5.0, 10.0, 20.0 and30.0 kO) was studied under controlled conditions (OLR 30.0 gCOD/L-d, HRT 1.04 day, pH 5.6–6.5 and aerated at 2.0 L/min). It wasfound that 10 kO revealed the best efficiency of 94.13% COD re-moval (26.67 gCOD/L-d) (Fig. 4A) while fed by 6 gCOD/L of palmoil mill effluent. The result is related to the polarization curvewhich indicated the internal resistance that provided the maxi-mum power of 11.59 mW/m3 (Fig. 4B).

C. Sukkasem, S. Laehlah / Bioresource Technology 146 (2013) 749–753 753

3.5. Sea food wastewater treatment

3.5.1. Effect of the external resistance on the power generation fromsea food process wastewater

The effect of various external resistances (1.0, 5.0, 10.0, 20.0 and30.0 kO) was studied under maximum OLR 8.57 gCOD/L-d, pH 5.6–6.5 and aerated at 2.0 L/min. It was found that 10.0 kO of externalresistance gave the maximum COD removal efficiency of 94.37%(8.08 gCOD/L-d) (Fig. 5A). It is related to the polarization curve thatindicated the internal resistance that gave the maximum power ofthe UBFC at 9.47 mW/m3 (Fig. 5B).

4. Conclusion

The UBFC system can treat wastewaters without chemical treat-ment or nutrient supplementation. COD removal levels of 7.35,20.1 and 26.67 gCOD/L-d were obtained, while fed with sea food,biodiesel and palm oil mill wastewater, respectively; these arehigher than for existing technologies (Table 1). The carbon fiberbrush immobilized base improved the UBFC performance by17.54%. To comply with standards, a complete FOG separation pro-cess, an optimum number of UFAF and a constructed wetland atterminal process, are suggested to meet the discharge allowance.Further development is needed to scale up the UBFC for wastewa-ter treatment.

Acknowledgements

This research was financially supported by the Thai NationalResearch Council of Thailand. Authors would like to thank Dr. JohnThomas Harry Pearce, Dr. Elizabeth Burrows, Assoc. Prof. Dr. Siri-luck Nivitchanyong and Assoc. Prof. Dr. Kasem Asawatreratanakulfor their kind assistance to this work.

References

Aelterman, P., Rabaey, K., Clauwaert, P., Verstraete, W., 2006. Microbial fuel cells forwastewater treatment. Water Science and Technology 54, 9–15.

Ahn, Y., Logan, B.E., 2010. Effectiveness of domestic wastewater treatment usingmicrobial fuel cells at ambient and mesophilic temperatures. BioresourceTechnology 101, 469–475.

APHA, AWWA, and WPCF, 1995. Standard Methods for the Examination ofWater and Wastewater. American Public Health Association, Washington,DC.

Behera, M., Ghangrekar, M.M., 2009. Performance of microbial fuel cell in responseto change in sludge loading rate at different anodic feed pH. BioresourceTechnology 100, 5114–5121.

Cheng, J., Zhu, X., Ni, J., Borthwick, A., 2010. Palm oil mill effluent treatment using atwo-stage microbial fuel cells system integrated with immobilized biologicalaerated filters. Bioresource Technology 101, 2729–2734.

Drapcho, C.M., Nhuan, N.P., Walker, T.H., 2008. Biofuels Engineering ProcessTechnology. The McGraw-Hill Companies, Inc., New York, Chicago, SanFrancisco, Lisbon, London, Madrid, Mexico City, Milan, New Delhi, San Juan,Seoul, Singapore, Sydney, Toronto.

Feng, Y., Yang, Q., Wang, X., Liu, Y., Lee, H., Ren, N., 2011. Treatment ofbiodiesel production wastes with simultaneous electricity generationusing a single-chamber microbial fuel cell. Bioresource Technology 102,411–415.

Greenman, J., Gálvez, A., Giusti, L., Ieropoulos, I., 2009. Electricity from landfillleachate using microbial fuel cells: comparison with a biological aerated filter.Enzyme and Microbial Technology 44, 112–119.

Jiang, J., Zhao, Q., Zhang, J., Zhang, G., Lee, D.J., 2009. Electricity generation from bio-treatment of sewage sludge with microbial fuel cell. Bioresource Technology100, 5808–5812.

Logan, B.E., 2008. Microbial Fuel Cells. John Wiley & Sons, Inc., Hoboken, New Jersey.Nam, J.Y., Kim, H.W., Lim, K.H., Shin, H.S., 2010. Effects of organic loading rates on

the continuous electricity generation from fermented wastewater using asingle-chamber microbial fuel cell. Bioresource Technology 101 (Suppl. 1), S33–S37.

Pham, T.H., Rabaey, K., Aelterman, P., Clauwaert, P., De Schamphelaire, L., Boon, N.,Verstraete, W., 2006. Microbial fuel cells in relation to conventional anaerobicdigestion technology. Engineering Life Science 6, 285–292.

Rabaey, K., Sompel, K.V.D., Maignien, L., Boon, N., Aelterman, P., Clauwaert, P., DeSchamphelaire, L., 2006. Microbial fuel cell for sulfide removal. EnvironmentalScience and Technology 40, 5218–5224.

Raghavulu, S.V., Mohan, S.V., Goud, R.K., Sarma, P.N., 2009. Effect of anodic pHmicroenvironment on microbial fuel cell (MFC) performance in concurrencewith aerated and ferricyanide catholytes. Electrochemistry Communications 11,371–375.

Rodrigo, M.A., Canizares, P., Lobato, J., Paz, R., Saez, C., Linares, J.J., 2007. Productionof electricity from the treatment of urban waste water using a microbial fuelcell. Journal of Power Sources 169, 198–204.

Sukkasem, C., Laehlah, S., Hniman, A., O-thong, S., Boonsawang, P.,Rarngnarong, A., Nisoa, M., Kirdtongmee, P., 2011. Upflow bio-filtercircuit (UBFC): biocatalyst microbial fuel cell (MFC) configuration andapplication to biodiesel wastewater treatment. Bioresource Technology102, 10363–10370.

Sukkasem, C., Xu, S., Park, S., Boonsawang, P., Liu, H., 2008. Effect of nitrate on theperformance of single chamber air cathode microbial fuel cells. Water Resource42, 4743–4750.

Yi, T., Harper, W.F., 2009. The effect of nitrate and sulfate on mediator-less microbialfuel cells with high internal resistance. Water Environment Research 81, 2320–2328.