10Feng C H_ Li F B_ Mai H J_ Et Al. Bio-Electro-Fenton Process Driven by Microbial Fuel Cell for Wastewater Treatment. Environ. Sci. Technol. 2010_ 44 (5)_ 1875-1880

8/13/2019 10Feng C H_ Li F B_ Mai H J_ Et Al. Bio-Electro-Fenton Process Driven by Microbial Fuel Cell for Wastewater Treatm

1/14

Bio-Electro-Fenton Process Drivenby Microbial Fuel Cell forWastewater Treatment

C H U N - H U A F E N G , F A N G - B A I L I , *,

H O N G - J I A N M A I , A N DX I A N G - Z H O N G L I

The Key Lab of Enhanced Heat Transfer and EnergyConservation, Ministry of Education, School of Chemistry andChemical Engineering, South China University of Technology,Guangzhou 510640, PR China, Guangdong Key Laboratory of Agricultural Environment Pollution Integrated Control,Guangdong Institute of Eco-Environmental and Soil Sciences,Guangzhou 510650, PR China, and Department of Civil andStructural Engineering, The Hong Kong PolytechnicUniversity, Hong Kong, PR China

Received October 31, 2009. Revised manuscript receivedJanuary 10, 2010. Accepted January 14, 2010.

In this study, we proposed a new concept of utilizing thebiological electrons produced from a microbial fuel cell (MFC)

to power an E-Fenton process to treat wastewater at neutralpH as a bioelectro-Fenton (Bio-E-Fenton) process. This processcan be achieved in a dual-chamber MFC from which electronswere generated via the catalyzation of Shewanella decol-orationisS12 in its anaerobic anode chamber and transferred

to its aerated cathode chamber equipped with a carbonnanotube (CNT)/-FeOOH composite cathode. In the cathodechamber,the Fentonsreagentsincludinghydrogenperoxide(H2O2)and ferrous irons (Fe2+) were in situ generated. This Bio-E-

Fenton process led to the complete decolorization andmineralization of Orange II at pH 7.0 with the apparent first-order rate constants,kapp ) 0.212 h

-1 andkTOC ) 0.0827 h-1,

respectively, and simultaneously produced a maximum poweroutput of 230 mW m-2 (normalized to the cathode surfacearea). The apparent mineralization current efficiency wascalculated to be as high as 89%. The cathode compositionwasan important factor in governing systemperformance. When

the ratio of CNT to -FeOOH in the composite cathode was1:1, the system demonstrated the fastest rate of Orange IIdegradation, corresponding to the highest amount of H2O2formed.

IntroductionThe electro-Fenton (E-Fenton) process has been widelystudied for the destruction of organic and biorefractorypollutants contained in wastewaters by highly oxidativehydroxyl radicals formed from the reaction of electro-generated H2O2with Fe2+ (1-11). It offers more advantagesthanthechemicalFentonprocessowingtothehighefficiencyof Fentons reagents (e.g., H2O2) utilization and saving costsinduced by the chemical storage and transportation. The

power consumption in the E-Fenton process will mainlycontribute to its operating costs. It should be very attractiveand also challenging to develop an energy-saving E-Fentonsystem.Many reports (12-15) studied themicrobial fuel cell(MFC) and showed that electrons can be continuouslysupplied from the organics existed in wastewaters. Thebioelectro-Fenton (Bio-E-Fenton) concept is thus possible

by using these bioelectrons produced from the microbialmetabolism to drive an E-Fenton process. This can beachieved by properly configuring a MFC (16) reactor whichconsists of two chambers separated by a cation-exchangemembrane: an anaerobic anode chamber filled with bio-degradable organic substrates and an aerated cathodechamber with biorefractory pollutants. The electrons arereleased from thebioreactions at the anode and transportedto the cathode through an external load circuit. The two-electron reduction of oxygen at the cathode results in H2O2formation (17), which then reacts with a Fe2+ source (e.g.,FeSO4 (16)) to produce hydroxyl radicals for pollutantoxidative degradation.

With respect to the E-Fenton reaction, it has been foundthat acidic pH between 2 and 4 is important in facilitating

oxidative degradation of pollutants (1-9). This, however,requires an initial pH adjustment with acids and finalneutralization of the treated water before it is released intotheenvironment; thus results in an increase in thetreatmentcostand also sludgeproduction. Recently,it hasbeen shownthat using these low soluble iron oxides as iron sources inthe electro-Fenton process has the advantages of the abilityto self-regulate the supply of a constant amount of iron ionsall along the reaction time and also the easy recycling of theiron catalyst after treatment (10, 11, 18, 19). Moreover, it canallow the E-Fenton reaction to proceed under a neutralcondition (10, 11). Taking advantages of both the neutralE-Fenton reaction and utilization of bioelectrons as a powersupply, in this study we proposed a MFC-driven E-Fentonprocess as a Bio-E-Fenton reaction system for wastewatertreatment at neutral pH.

To attain high degradation efficiency at neutral pH, wefabricated a carbon nanotube (CNT)/-FeOOH compositecathodefor theBio-E-Fenton system. CNTs were used as thecathode materials for the in situ generation of H2O2owingto their advantages of large surface area, good conductivityand superior electrochemical activity over other carbonmaterials towardthe two-electron oxygen reduction (11,20).The lepidocrocite (-FeOOH), an iron oxide with highersolubility in water than goethite and hematite, functionedmainly as the Fe2+ source of the E-Fenton reactions. Fe2+

was in situ produced at neutral pH by direct electroreductionof-FeOOH to adsorbed ferrous ion, Feads2+, followed by itsdesorption to aqueous solution as Fe2+ (21). The aim of this

study was at demonstrating the feasibility of using such aBio-E-Fenton system to degrade Orange II, a model azo dye(1-3) that is widely used in a variety of industries such astextile, food, and cosmetics and abundant in their waste-waters, in aqueous solution at neutral pH.

Experimental SectionConfigurationand Operation of the Bio-E-FentonProcess.A MFC configuration is shown in Figure 1. It consists of twoequal rectangular chambers (anode chamber and cathodechamber), which were separated by a cation exchangemembrane (Zhejiang Qianqiu Group Co., Ltd. China). Eachchamber has an effective volume of 75.6 mL (6.06.02.1cm). The anode is a piece of carbon felt (4.44.40.5 cm)which was washed in a hot H2O2(10%, 90 C) solution for

* Corresponding author phone: 86-20-87024721; fax: 86-20-87024123; e-mail: [email protected].

South China University of Technology. Guangdong Institute of Eco-Environmental and Soil Sciences. The Hong Kong Polytechnic University.

Environ. Sci. Technol. 2010, 44,18751880

10.1021/es9032925 20 10 A mer ic an C he mi cal So ci et y V OL. 44 , N O. 5 , 20 10 / E NVI RON ME NT AL SC IE NCE & T EC HN OLO GY 9 1875

Published on Web 01/28/2010


2/14

3 h to develop local quinone sites on the carbon surface forimproving the anode biocompatibility (22). The cathode is

a composite electrode of cabon nanotube (CNT) and-FeOOH which was prepared by (1) mixing CNTs (10-15nmwideand3-5mlong,ShenzhenNanoHarboCo.,China)and -FeOOH (homemade according to the proceduresdescribed elsewhere (23)) with polytetrafluoroethylene (PTFE)solution (Dupont) and ethanol in an ultrasonic bath to forma dough-like paste; (2) assembling the paste between twopieces of Timesh(0.1mm thickness)at a pressure of10 MPaand 60 C. The PTFE functions as a promoter (11) for oxygendiffusion in thecathode. Four types of cathodes withdifferentCNT/-FeOOHratiosof1:0,1:0.5,1:1,and1:2werefabricatedwith the same amounts of CNT (5 g) and PTFE (0.5 g), butdifferent -FeOOH contents. A Ti wire (0.5 mm in diameter)was used to connect the anode and cathode by passingthrough anexternal load.Unlessotherwisestated, thecathode

usedwas composed ofCNT and-FeOOHwitha ratio of1:1.The inoculation and operation of the MFC with a pure

culture ofShewanelladecolorationisS12(24) were describedin the Section S1 of the Supporting Information (SI). FourMFCunits including one experimental sample (MFC-A) andthree control samples (MFC-B, MFC-C, and MFC-D) wereused to account forthe decolorization and mineralization ofOrange II. Same anode was used in four MFC units, butexperiments were conducted under different cathode condi-tions as summarized in SI Table S1. Each MFC was initiatedwith an Orange II-free cathode solution (100 mM phosphatebuffer solution, PBS) purged with air. When the cell voltageremained unchanged for over one day, the cathode solutionwas replaced with the fresh solution containing 0.1 mMOrange II dye and 100 mM PBS. The decolorization andmineralization of Orange II occurred in MFC-A once air wascontinuously purged to its cathode chamber. The MFC-Band MFC-C experiments were designed in the absence ofH2O2 andFe2+, respectively. MFC-B consistedof a N2-purgedcathode solution in which Orange II is the sole electronacceptor (25) susceptible for reduction and no H2O2 wasgenerated due to the lack of the dissolved O 2. MFC-C useda CNT only electrode without -FeOOH as the cathode andno Fe2+ was produced in the absence of an iron source.MFC-DwiththesameconfigurationofMFC-Awasconductedunder an open-circuit condition to avoid anyelectrochemicalreactions and study the effects of adsorption on Orange IIremoval.

Analytical Methods.The concentration of Orange II was

determined by a UV-vis spectrophotometry (TU1800-PC,

Beijing China) at 484 nm. The concentration of H 2O2 wasdetermined spectrophotometricallyusing the iodide method

at 351 nm (26). The concentration of Fe2+

was measuredbased on the light absorption of its complex after reactionwith 1, 10-phenantroline at 508 nm. It should be noted thatH2O2and Fe2+ concentrations were detected when OrangeII was absent in the cathode chamber. Total organic carbon(TOC) analysis was carried out with a Shimadzu TOC-VSCNanalyzer. X-raypower diffraction(XRD)measurements wereperformed on a Bruker D8 Advance X-ray diffractometer withCu Ka radiation (1.54178 ).

To evaluate the power performance of the system, thecell polarization curves as well as the anode and cathodepolarization curves were measured by varying an externalresistor in the range of 10-6000 . The anode and cathodepotentials were measured by placing a saturated calomelelectrode (SCE, +0.242 V vs SHE) in the anode and cathode

chambersforreference.Currentdensity(I)andpowerdensity(P) were calculated as follows:

where U is the cell voltage measured; R is the electricalresistance; Iis thecurrent normalized to thecathode surfacearea;Aisthecathodesurfacearea,andPis the power density.

To evaluate the catalytic activity of the cathode towardoxygen reduction, linear sweep voltammetry (LSV) measure-ments were performed in 100 mM PBS at pH 7.0 using anAutolab potentiostat (PGSTAT30, Eco Chemie). The CNT/-FeOOH composite electrode was used as the workingelectrode, while a Pt mesh (2 2 cm) and a SCE were usedas the counter and reference electrodes, respectively. Beforethe measurements, the solution was saturated with oxygen.A scan rate was set at 50 mV s-1 and temperature was 30 C.

Results and DiscussionDegradation of Orange II. Figure 2 showsthe decolorizationand mineralization of Orange II in the cathode chamber atpH 7.0 against time. In Figure 2A, a gradual decolorizationof the solution in MFC-A was observed with eye as timeproceeded. Approximately 100% of the initial Orange II wasdegraded by theBio-E-Fenton process within14 h. However,the Orange II degradations in MFC-B and MFC-C were only

achieved by 10 and 8%,respectively, after 14 h. These results

FIGURE 1. Schematic diagram of the Bio-E-Fenton system having an MFC configuration.

I ) URA

(1)

P )UI (2)

1876 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 5, 2010


3/14

demonstrated much less degradation of Orange II in theabsence of the Fentons regents due to the lack of thedissolved O2in MFC-B to produce H2O2and the lack of Fe2+

in MFC-C. In addition, the concentration of Orange II inMFC-D decreased only by 3%, showing that its adsorptionon the electrode material under the open circuit conditionwas insignificant. Figure 2B indicated that the completemineralization of Orange II within 43 h in MFC-A. Asanticipated,there wasno TOC reductionfor thethree controlsamples. In contrast, a slight TOC increase was observed inMFC-B, MFC-C, and MFC-D possibly due to the release oforganic matters from the CNT surfaces and transport oforganic species from the anode to the cathode through

membrane.In agreement with previous studies (2, 3) concerningOrange II degradation in traditional E-Fenton systems, theexponential decrease of its concentration was observed inthe Bio-E-Fenton process at neutral pH. The experimentaldata in Figure 2A were fitted by the apparent first-orderlogarithmicdecay model (eq3) andan apparentrate constant(kapp) value of 0.212 h-1 was determined.

whereCtandC0are the concentrations of Orange II at timetandtimezero,respectively,and tisthereactiontime.Itwas

found that TOCremoval alsofollowedthe pseudofirst-order

kinetics(eq4)withanapparentmineralizationconstant(kTOC)value of 0.0827 h-1.

where TOCtand TOC0 arethe concentrations of TOC at timetand time 0, respectively, and tis the reaction time. Themuch lower value ofkTOCthan that ofkappindicates that theOrange II dye was first oxidized to colorless intermediatesand then further oxidized to a final product of CO2 (2, 3).

Taking into account the results shown in Figure 2, it canbe seen that this Bio-E-Fenton system enables the completemineralization of Orange II dye at neutral pH and does notneed any power input for in situ generation of Fentonsreagents. The process efficiency was further evaluated interms of theapparentmineralizationcurrent efficiency(MCE)(4, 5, 27, 28) as defined by eq 5.

where(TOC)exp is the experimental TOCremoval at a giventimeand(TOC)theor is thetheoreticalTOCremoval calculatedaccording to the reaction indicated by eq 6, if the electrons

reaching thecathode are fully utilized for themineralizationof Orange II. In the light of eq 6, the destruction of eachmolecule of Orange II consumed 84 electrons.

Accordingly, (TOC)theorcan be calculated based on eq 7.

where Iis the current generated in the MFC, tis the reactiontime,Fis the Faraday constant,Vis the effective volume of

the cathode chamber, andMis the total molecule weight ofcarbon. The value of MCE in this system was determined tobe89%,muchhigherthanthereportedvaluesinotherstudies(4, 5, 27, 28). For example, Ozcan et al. (27) reported amaximum MCE value of 35% when concerning the miner-alization of basic blue 3 dye via the traditional E-Fentonprocess.ThehigherMCEvalueobtainedinthisstudyislikelydue to the fact that the electrical energy from the MFC canbe better utilized for in situ generation of Fentons reagentsthanthatfromanexternalenergysource,andthattheparasitereactions of hydroxyl radicals with H2O2 andFe2+ (27, 28)aresuppressed owing to low amounts of H2O2and Fe2+ (Table1) available in this system.

To investigate the performance stability of the Bio-E-Fenton system, thedegradationexperiments wererepeatedlyconducted for up to 10 runs. The rate constants ofkappandkTOCwere determined as shown in Figure 3. It can be notedthat both the values decreased slightly during the first fourruns and then dropped dramatically in the fifth run. Thisphenomenon is the consequence of the gradual decline inanode performance along withthe increasednumberof runs.As shown in SI Figure S2, the curves of anode polarizationmoved toward less negative potential values with increasedslopes when more experiments were run. The curves ofcathode polarization, however, showed little variation. Thedecrease of pHin theanolytedue to accumulationof protonsand thedepletion of fuel (particularlyin thefifth run) shouldbe responsible for the loss of anode activity. More detailedexplanations can be found in SI Section S2. After replenish-

ment with the fresh substrate in the anode chamber, the

FIGURE 2. Decolorization (A) and mineralization (B) kinetics ofOrange II in four MFC units. The inset shows the color changeover time in MFC-A. The data point shown represents theaverage on triplicate measurements obtained from threeindependent experiments ( standards deviations.

lnCtC0

) -kappt (3)

lnTOCtTOC0

) -kTOCt (4)

MCE )(TOC)exp(TOC)theor

100 (5)

C16H11N2NaO4S +38 H2O f16CO2 +Na+

+SO4-

+

2 NO3-

+87 H+ +84 e- (6)

(TOC)theor ) It

84FV M (7)

VOL. 44, NO. 5, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY9 1877


4/14

values ofkappand kTOCwere well maintained from the sixthrun. Further replacements of the cathode solution onlyresulted in slight decrease in both values, and the tenth testgave substantially decreased values also due to the con-

sumption of anode substrate and the decrease of pH inanolyte solution. These results showed the durability of theBio-E-Fenton system operated at neutral pH over 20 d. TheXRD analysis (SI Section S3 and Figure S3) on the treatedCNT/-FeOOH(1:1,20dayreaction)andthefreshlypreparedelectrodes demonstrated a negligible difference in theirpatterns and indicated that the composite cathode can bereused.

Effect of Cathode Composition on the Process Perfor-mance. In the composite cathode, the CNTs can generateH2O2through a two-electron reduction of oxygen, whereasthe -FeOOH can release free Fe2+ for the Fenton reactionsand also catalyzed oxygen reduction. The ratio of CNT to-FeOOH by weight was an important factor to affectperformance of theBio-E-Fenton process. Table 1 shows theeffects of the cathode composition on the kinetics of OrangeII decolorization and mineralization. By comparing bothvalues ofkappandkTOCamong three samples, it can be seenthat the rate of Organge II degradation with respect todifferentcathode compositions increased in an order of 1:0.5CNT/-FeOOH


5/14


6/14


7/14

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

Supporting Information

Bio-Electro-Fenton Process Driven by Microbial Fuel Cell for

Wastewater Treatment

Chun-Hua Feng,Fang-Bai Li,

*,Hong-Jian Mai, Xiang-Zhong Li

The Key Lab of Enhanced Heat Transfer and Energy Conservation, Ministry of Education,

School of Chemistry and Chemical Engineering, South China University of Technology,

Guangzhou 510640, PR China

Guangdong Key Laboratory of Agricultural Environment Pollution Integrated Control,

Guangdong Institute of Eco-Environmental and Soil Sciences, Guangzhou 510650, PR

China

Department of Civil and Structural Engineering, The Hong Kong Polytechnic University,

Hong Kong, PR China

8 Pages

3 Sections

1 Table

4 Figures

Submitted to Environmental Science and Technology

*Corresponding author phone: 86-20-87024721; fax: 86-20-87024123 ; e-mail: [email protected].
mailto:[email protected]:[email protected]


8/14


9/14

3

1

2

3

4

5

6

7

8

9

10

11

decrease of pH in the anolyte and the depletion of fuel (particularly for the fifth run) should

be responsible for the decrease in kappand kTOC.

Section S3 XRD results. Figure S3 shows XRD patterns of the freshly prepared and treated

CNT/-FeOOH (1:1, 80-h reaction) electrodes. The diffraction peaks appeared at a 2value

of 18.1 and 26.1 were ascribed to CNTs and PTFE, respectively. The remaining peaks in

the composite electrode were characteristics of -FeOOH. These results demonstrate the

successful preparation of the composite electrode. A negligible difference in the patterns

between the treated and freshly prepared electrodes suggests that the resulting cathode can

be reused.


10/14

4

1 Table S1 Cathode operation conditions in four MFC units.

Operation conditions of cathode

SamplesCathode solution

Gas

purged

Cathode

material

Close or

open circuit

MFC-A PBS (100 mM, pH 7.0) + Orange II

(0.1 mM)Air 1:1

CNT:-FeOOHClose

MFC-BPBS (100 mM, pH 7.0) + Orange II

(0.1 mM)N2

1:1

CNT:-FeOOHClose

MFC-CPBS (100 mM, pH 7.0) + Orange II

(0.1 mM)Air Only CNT Close

MFC-DPBS (100 mM, pH 7.0) + Orange II

(0.1 mM)

Air1:1

CNT:-FeOOH

Open

2


11/14

5

1

2

3

4

5

Figure S1 Changes of pH in the catholyte (A) and the anolyte (B) as a function of time. The

data point shown represents the average on triplicate measurements obtained from three

independent experiments standards deviations.

0 10 20 30 405.5

6.0

6.5

7.0

7.5

8.0

8.5

pHi

n

thec

atholyte

Time (h)

(A)

6

0 50 100 150 200 2505.5

6.0

6.5

7.0

7.5

8.0

8.5

pHi

n

theanolyte

Time (h)

(B)

7

8


12/14

6

1

2

3

4

5

6

Figure S2 Dependence of anode and cathode polarization curves on numbers of experiments.

In these experiments, the anode solution remained unchanged, but the cathode solution was

changed 5 times for 5 successive experimental runs. The data point shown represents the

average on triplicate measurements obtained from three independent experiments

standards deviations.

0.0 0.4 0.8 1.2 1.6

-0.4

-0.2

0.0

0.2Cathode

Run 1

Run 2

Run 3

Run 4

Run 5

Anode

Run 1

Run 2

Run 3

Run 4

Run 5

Potential(V)vs.

SCE

Current density (A m-2)

7


13/14

7

1

2

3

4

5

6

7

8

9

10

11

12

Figure S3 XRD patterns of different samples.

10 20 30 40 50 60 70 80

0

300

600

900

10 20 30 40 50 60 70 80

0

200

400

600

10 20 30 40 50 60 70 80

0

150

300

450

10 20 30 40 50 60 70 80

0

1500

3000

4500

Intensity(Counts)

2 ( ))

PTFE

D: Treated CNT+PTFE+-FeOOH

C: Freshly prepared CNT+PTFE+-FeOOH

B: -FeOOH

CNT A: CNT+PTFE


14/14

8

1

2

3

4

5

6

7

8

9

10

11

12

13

Figure S4 Linear sweep voltammograms of oxygen reduction on the CNT/-FeOOH

composite cathodes with different compositions in O2-saturated 100 mM PBS. The scan rate

was 50 mV s-1.

-1.2 -0.9 -0.6 -0.3 0.0 0.3 0.6

-900

-600

-300

0

(4)

(3)

(2)

(1) Only CNT

(2) CNT/-FeOOH = 1:0.5(3) CNT/-FeOOH = 1:1(4) CNT/-FeOOH = 1:2C

urrent(

A)

Potential (V) vs. SCE

(1)

Documents

10Feng C H_ Li F B_ Mai H J_ Et Al. Bio-Electro-Fenton Process Driven by Microbial Fuel Cell for Wastewater Treatment. Environ. Sci. Technol. 2010_ 44 (5)_ 1875-1880