Accepted Manuscript

Bioelectrochemical enhancement of methane production in low temperature

anaerobic digestion at 10 °C

Dandan Liu, Lei Zhang, Si Chen, Cees Buisman, Annemiek ter Heijne

PII: S0043-1354(16)30217-2
DOI: 10.1016/j.watres.2016.04.020
Reference: WR 11977

To appear in: Water Research

Received Date: 28 October 2015

Revised Date: 11 April 2016
Accepted Date: 12 April 2016

Please cite this article as: Liu, D., Zhang, L., Chen, S., Buisman, C., ter Heijne, A., Bioelectrochemical
enhancement of methane production in low temperature anaerobic digestion at 10 °C, Water Research
(2016), doi: 10.1016/j.watres.2016.04.020.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

31

(mg CH4-COD/gVSS)
CH4 yield

PT
10℃

RI
5

U SC
AN
Electricity

M
D
TE
C EP
AC

CH4 CH4

Acetate e- Acetate

BES-AD system AD system

 ACCEPTED MANUSCRIPT

1 Bioelectrochemical enhancement of methane

2 production in low temperature anaerobic digestion

3 at 10 °C

PT
4 Dandan Liu*, Lei Zhang*, Si Chen, Cees Buisman, Annemiek

RI
5 ter Heijne

SC
6 Sub-Department of Environmental Technology, Wageningen University,

7 Bornse Weilanden 9, P.O. Box 8129 6709WG Wageningen, The Netherlands

U
AN
8 Note: Dandan Liu and Lei Zhang contribute same to this publication

9 Corresponding author: Annemiek ter Heijne;

M

10 E-mail: Annemiek.terHeijne@wur.nl
D

11 Abstract
TE

12 Anaerobic digestion at low temperature is an attractive

EP

13 technology especially in moderate climates, however, low

C

14 temperature results in low microbial activity and low rates of

AC

15 methane formation. This study investigated if

16 bioelectrochemical systems (BESs) can enhance methane

17 production from organic matter in low-temperature anaerobic

18 digestion (AD). A bioelectrochemical reactor was operated with

19 granular activated carbon as electrodes at 10 . Our results

1
 ACCEPTED MANUSCRIPT

20 showed that bioelectrochemical systems can enhance CH4 yield,

21 accelerate CH4 production rate and increase acetate removal

22 efficiency at 10 . The highest CH4 yield of 31 mg

23 CH4-COD/g VSS was achieved in the combined BES-AD

PT
24 system at a cathode potential of -0.90 V (Ag/AgCl), which was

RI
25 5.3 to 6.6 times higher than that in the AD reactor at 10 .

26 CH4 production rate achieved in the combined BES-AD system

SC
27 at 10 was only slightly lower than that in the AD reactor at

U
28 30 . The presence of an external circuit between the
AN
29 acetate-oxidizing bioanode and methane-producing cathode

30 provided an alternative pathway from acetate via electrons to

M

31 methane, potentially via hydrogen. This alternative pathway

D

32 seems to result in higher CH4 production rates at low

TE

33 temperature compared with traditional methanogenesis from

34 acetate. Integration of BES with AD could therefore be an

EP

35 attractive alternative strategy to enhance the performance of

C

36 anaerobic digestion in cold areas.

AC

37 Keywords: Anaerobic digestion, bioelectrochemical system,

38 low temperature, methane methanogenic activity

39 1. Introduction

40 Anaerobic wastewater treatment is an attractive

2
 ACCEPTED MANUSCRIPT

41 technology that recovers energy in the form of CH4, has low

42 excess sludge production, and low operational cost (Speece

43 2008). For anaerobic digestion, mesophilic conditions

44 (25-37 ) and thermophilic conditions (55-65 ) are

PT
45 required to ensure optimal microbial activity (Hussain and

RI
46 Dubey 2015). Many types of wastewater are discharged at low

47 temperature, however, such as wastewater from malting,

SC
48 breweries, soft drink industry and domestic sewage, with

U
49 typical temperatures around 10-20 (Lettinga et al. 2001). As
AN
50 low temperature leads to low activity and low growth rate of

51 methanogens, and consequently to low CH4 production rate,

M

52 strategies are needed to improve methanogenic activity of

D

53 anaerobic sludge (Zhang et al. 2013a, Álvarez et al. 2004,

TE

54 Mahmoud et al. 2004). CH4 production at 10 is at this

55 moment unfeasible. In order to keep CH4 production and

EP

56 wastewater treatment at a high rate in a cold area or during cold

C

57 months, many anaerobic digesters are heated up to mesophilic

AC

58 conditions. However, this requires expensive heating and

59 insulation systems, which makes it not practical and

60 uneconomical (Witarsa and Lansing 2015). Use of electrodes

61 instead of heating up could provide an alternative,

62 cost-effective strategy to enhance performance of anaerobic

3
 ACCEPTED MANUSCRIPT

63 digesters.

64 Methane-producing bioelectrochemical systems (BESs)

65 are of special interest as a promising “power to gas” technology

66 for renewable energy storage (Van Eerten-Jansen et al. 2012,

PT
67 Cheng et al. 2009, Villano et al. 2010). In this system, different

RI
68 types of electron donors (e.g. organic matter contained in waste

69 streams) can be oxidized at the anode, and when electrical

SC
70 energy is supplied to the system, the released electrons can be

U
71 transferred to the cathode to produce CH4 from CO2. Reduction
AN
72 of CO2 to CH4 can occur via two pathways: direct, using

73 electrons and protons (H+) and indirect, where protons are first
M

74 reduced to H2, and H2 is used to reduce CO2 in the presence of

D

75 hydrogenotrophic methanogens (Villano et al. 2010).

TE

76 BESs have been integrated with anaerobic digestion to

77 enhance CH4 formation at mesophilic conditions (20 – 40 ).

EP

78 Compared to the digesters without electrodes, integrating these

C

79 two technologies has been shown to lead to enhanced CH4

AC

80 production rates (Feng et al. 2015, Tartakovsky et al. 2011,

81 Zamalloa et al. 2013). Moreover, control experiments with

82 electrodes but without an additional electrical energy supply

83 also resulted in increasing CH4 yield and rate, as a result of

84 enhanced biofilm attachment (De Vrieze et al. 2014). Also, it

4
 ACCEPTED MANUSCRIPT

85 was shown that hydrogenotrophic methanogens and

86 exoelectrogens (e.g. Geobacter sp.) were enriched, especially

87 on the cathode electrode, when integrating BES with AD. This

88 also enhanced CH4 production rate in anaerobic digestion of

PT
89 waste activated sludge at 20-25 (Liu et al. 2016). The

RI
90 integration of BES in anaerobic digestion at low temperature

91 (<20 ), however, has not been studied. While bioanodes

SC
92 have been shown to produce current at low temperature

U
93 (<20 ), albeit at lower rates than at mesophilic conditions,
AN
94 the cathodic production of CH4 at low temperature has not been

95 reported before.
M

96 At low temperature (<20 ), fatty acids are accumulated

D

97 as a result of low methanogenic activity. Hydrogenotrophic

TE

98 methanogenesis is far less temperature-sensitive than

99 acetoclastic methanogenesis (Enright et al. 2009), and the

EP

100 growth of hydrogenotrophic methanogens can be stimulated by

C

101 a moderate release of small amount of hydrogen at the cathode

AC

102 electrode (Liu et al. 2016, Yang et al. 2013). Instead of

103 producing CH4 through the traditional route of acetoclastic

104 methanogenesis and hydrogenotrophic methanogenesis,

105 including elecrodes in the system could result in a new route

106 for methanogenesis, created via the bio-anode (Fig. 1a): acetate

5
 ACCEPTED MANUSCRIPT

107 is oxidized while generating electrons and protons (Lu et al.

108 2011), which, with input of electrical energy can be reduced to

109 H2 or CH4 at the biocathode(van Eerten-Jansen et al. 2015). The

110 different routes via which CH4 can be produced from acetate in

PT
111 this combined BES-AD system are summarized in Fig.1b.

RI
112 The objective of this study was to investigate whether a

113 methane-producing BES could enhance CH4 production rate in

SC
114 low-temperature (10 ) anaerobic digestion. Energy input in

U
115 terms of electricity in this combined BES-AD system was also
AN
116 investigated. Acetate was used as a model substrate. The results

117 were compared to reactors without external voltage and in the

M

118 presence of electrodes at 10 , and to anaerobic digestions

D

119 alone (control reactors without electrodes) at different

TE

120 operational temperatures (10 , 15 and 30 ).

C EP
AC

6
 ACCEPTED MANUSCRIPT

121 2. Materials and Methods

122 2.1BES-AD system set-up

123 2.1.1 BES-AD system

PT
124 The experiment was carried out in reactors constructed by

RI
125 assembling rectangular Perspex frames. These frames had

internal cylindrical volume of 28 mL(dimensions: 3.0 cm

SC
126

127 diameter×4.0 cm length) , and the reactor had a 8.0 mL gas

U
128 collection tube (Fig.3.). The solution(20 mL) in each reactor
AN
129 was continuously stirred with a magnet. For the BES-AD

reactor, an anode and cathode electrode compartment was made

M

130

131 on both sides of the chamber in one reactor. The electrodes

D

132 were made of granular activated carbon(1.5 g, 875 m2/g, Norit

TE

133 PK 1-3) and were packed in a cylindrical cage (3.0 cm diameter

EP

134 ×6.0 mm length), and were separated from the chamber with a

135 spacer (SEFAR, PROPYLTEX). A graphite rod was inserted in

C

136 both anode and cathode compartments as a current collector.

AC

137 Potentials were measured and were reported in this paper

138 versus Ag/AgCl reference electrode (Ag/AgCl 3M KCl,

139 ProSenseQiS, QM 710X, Netherlands) placed in the chamber.

140 For the AD reactor, the spacer and electrodes were replaced

141 with the same volume of a rubber sheet leading to the same

7
 ACCEPTED MANUSCRIPT

142 internal volume as BES-AD system.

143 2.1.2 Inoculum and medium

144 Each reactor was filled with 15mL of anaerobic sludge as

inoculum. The anaerobic sludge was taken from a digester

PT
145

146 (35 ) in the UASB-digester system treating domestic sewage

RI
147 (Zhang et al. 2013b). The volatile suspended solids

SC
148 concentration of the inoculum was 5.9±0.20 g VSS/L. Each

149 reactor was filled with 5mL of concentrated medium (12.8 g/L

150
U
CH3COONa, 0.80 g/L NH4Cl, 0.54 g/L KCL, 18.32 g/L
AN
151 Na2HPO4, 11.08 g/L NaH2PO4·2H2O, 0.80 mL of Wolfe’s
M

152 micronutrients (Clauwaert and Verstraete 2009), including

153 vitamin and mineral solution ). After mixing well with the
D

inoculum, the concentration the solution in each reactor was 3.2

TE

154

155 g/L CH3COONa, 0.05 g/L NH4Cl, 0.032 g/L KCL, 50mM
EP

156 phosphate buffer (4.58 g/L Na2HPO4, 2.77 g/L NaH2PO4·2H2O),

157 0.20 mL of Wolfe’s micronutrients. All of the chemicals used in

C
AC

158 this study were from Merck, Germany. The medium was

159 flushed with N2 (99.99%) for 30 minutes prior to each

160 experiment to provide anaerobic conditions.

161 2.1.3 Operational conditions and start-up

162 Five anaerobic reactors were operated in the conditions as

8
 ACCEPTED MANUSCRIPT

163 shown in Table 1. One reactor (BES-AD-10)was operated at

164 10 with BES in which the cathode potential was controlled

165 at -0.90 V. One reactor (BES-ADC-10) was operated as a

166 control at 10 , with electrodes but without an electrical

PT
167 connection via the potentiostat (open circuit condition).

RI
168 Three other control reactors (ADC-10, ADC-15, ADC-30)

were operated at 10, 15 and 30 to determine CH4

SC
169

170 production rates for AD without electrodes. Low temperature

U
171 digesters were operated in a fridge at 10 and 15 , and the
AN
172 digester at 30 was located in a temperature controlled

cabinet.
M

173

174 To start-up the bioanode for the BES-AD-10 reactor, the

D

anode was set at 0 V vs. Ag/AgCl using a potentiostat (Ivium

TE

175

176 with IviumSoft v2.462, Eindhoven, The Netherlands) to

EP

177 cultivate electroactive bacteria oxidizing acetate. After running

178 for 4 weeks, the anode potential reached around -0.35 V at open
C
AC

179 circuit conditions, and the cultivation of the bacteria oxidizing

180 acetate was considered to be completed (Wang et al. 2009). In

181 the BES-AD-10 reactor, afterwards, the cathode potential was

182 controlled at -0.90 V with the potentiostat.

183 No anaerobic sludge was added in the following batches.

9
 ACCEPTED MANUSCRIPT

184 At the end of each batch, part of the reactor solution (5 mL)

185 was replaced by the same amount of concentrated medium (see

186 section 2.1.2 for the concentrated medium composition). Each

187 batch lasted for 2 weeks. Three sequential batches in total

PT
188 lasted around 41 days.

RI
189 2.2 Analyses

SC
190 2.2.1 Gas production

191 CH4 and H2 were analysed by Gas chromatograph with the

U
AN
192 same conditions as described in the literature (Van

193 Eerten-Jansen et al. 2012). CH4 yield and production rate were
M

194 calculated with the following equation:

D

195 Cumulative CH4 yield (mg CH4-COD/gVSS):

TE

 ×
196
 =  ∙
 ∙  (1)
 ×  ∗
EP

197 Where CCH4 is concentration of CH4 in the headspace (%)

198 which was measured by gas chromatography, same as (Van

C
AC

199 Eerten-Jansen et al. 2012); P is the gas pressure of the reactor

200 (Pa) and it was measured periodically by gas pressure meter

201 (GMH 3151, Germany); V is the volume of headspace (16 mL)

202 of each reactor; R is gas constant value (8.314 m3 Pa/mol K); T

203 is the temperature in Kelvin; 4: per g of CH4 equals to 4 g COD;

204 MCH4 is the molecular weight of CH4 (16 g/mol); Cvss is the
10
 ACCEPTED MANUSCRIPT

205 concentration of volatile suspended solids (5.9±0.20 g/L);

206 Vsludge is the volume of sludge in each reactor (15 mL). 1 mg

207 CH4-COD is equal to 0.35 mL CH4 at standard conditions with

208 a temperature of 273.15 K and an absolute pressure of 1 bar.

PT
209 CH4 production rate (mg CH4-COD/(gVSS·d)) was

RI
210 calculated by dividing the CH4 yields with the time period

between the two sampling points.

SC
211

212 2.1.2 Substrate consumption

U
AN
213 Acetate concentration was measured by Gas

214 chromatograph (HP5890 series GC Germany) same as in

M

215 the literature (Van Eerten-Jansen et al. 2013). Acetate removal

D

216 efficiencies were calculated using equation:

TE

) *)+,
217 ( = × 100 (2)
)+,
EP

218 Where Ct and Ct-1 are the acetate concentration (mg/L) on

219 sample time t and previous sample time t-1.

C
AC

220 The CH4 production efficiency was determined by the

221 produced CH4 over removed acetate.

222 2.2.3 Coulombic and cathodic CH4 recovery efficiency

223 Current of the reactors was recorded each minute by the

224 potentiostat. The average current per hour was used to calculate

11
 ACCEPTED MANUSCRIPT

225 volumetric current density (I/V, where V is the working volume

226 of the BES-AD-10 reactor, 20 mL).

227 Performance of the BES-AD reactor was evaluated by two

228 kinds of efficiencies: namely coulombic efficiency and cathodic

PT
229 CH4 recovery efficiency. Coulombic efficiency(ηCE) represents

RI
230 the ratio of electrons measured as electric current over the

electrons available from the removed substrate (Feng et al.

SC
231

232 2015), as shown in the following equation:

U
)
15 2 34
/0 = × 100% (3)
AN
233 8/,555
6, *5 7∗ ∗=∗>
:;<
M

234 Cathodic CH4 recovery efficiency is the efficiency of

235 capturing electrons from the electric current in CH4, via
236 following equation:
D

> ∗=∗>
237 / = × 100% (4)
TE

)
15 234

238 Where MAc is the molar weight of acetate (60 g/mole),nAB is

EP

239 CH4 production (mole), F is the Faraday constant (96485

C

240 C/mole e-), n is the moles of electrons per mole of CH4 (8moles
AC

241 e-/ mole CH4), I is current (A), and t is time (s).

242 2.3 Polarization curves

243 Bioelectrochemical activity of the cathode in the

244 BES-AD-10 reactor was investigated by polarization curves.

245 These measurements were performed at the end of each batch

12
 ACCEPTED MANUSCRIPT

246 and were also done for a control cathode, in (the same

247 BES-AD-10 reactor with fresh medium but without inoculum).

248 The cathode potential was decreased from -0.80 V to-1.1 V

249 with steps of 0.050 V, each cathode potential lasting for 15min.

PT
250 Current was recorded each minute and the last point at each

RI
251 potential was plotted in the polarization curve (Deeke et al.

252 2012).

SC
253 3. Results and discussion

254
U
3.1 CH4 production under different operational modes
AN
255 The effect of combining BES and AD at low temperature
M

256 was first investigated in three different reactors ( BES-AD-10,

D

257 BES-ADC-10, ADC-10) at the same operational temperature

TE

258 10 As shown in Fig. 3(a), CH4 yields increased with time in

259 all the reactors during all three batches. CH4 yield was higher in
EP

260 the BES-AD-10 reactor compared to the other reactors, and

C

261 varied between 20 and 30 mg CH4-COD/gVSS during one

AC

262 batch period of 14 days. The maximum cumulative CH4 yield

263 was 31 mg CH4-COD/g VSS in BES-AD-10 reactor during the

264 second batch, which was 5.3-6.6 times higher than in the

265 BES-ADC-10 reactor and 5.0-15 times higher than in the

266 ADC-10 reactor. These results showed that BES enhanced CH4

13
 ACCEPTED MANUSCRIPT

267 formation at 10 . This is in line with other studies, that

268 showed that at mesophilic conditions (30 ), BES enhanced

269 the CH4 yield (Feng et al. 2015, Tartakovsky et al. 2011).

270 The lower CH4 yield of the BES-ADC-10, which

PT
271 contained electrodes but was operated at open circuit,

RI
272 compared with the BES-AD-10 reactor, in which electrical

energy was supplied, showed that the enhanced CH4 yield

SC
273

274 was the result of applied electrical energy, rather than of

U
275 biomass retention on the electrodes, as has been found
AN
276 previously at 30 (De Vrieze et al., 2014). CH4 yield of

the BES-ADC-10 reactor was higher than that of ADC-10

M

277

278 (anaerobic digestion alone) in the first batch, and was lower
D

279 in the second and third batch. The decrease of CH4 yield in
TE

280 BES-ADC-10 reactor during three batches might be the result

EP

281 of differences in pH: the pH in the BES-ADC-10 had

282 increased to 8.3 at the end of Batch 1, while the pH in the

C

283 other two reactors was quite stable (<7.8) during three
AC

284 batches (see Fig. S2). This higher pH could have decreased

285 the activity of methanogens (Chen et al. 2008). The pH in

286 BES-AD-10 was 7.5±0.3, higher than ADC-10 (7.1±0.20),

287 possibly the result of consumption of protons at the cathode,

288 a phenomenon also found by previous studies working at

14
 ACCEPTED MANUSCRIPT

289 mesosphilic conditions (Bo et al. 2014).

290 To compare the potential of CH4 yield in the BES-AD

291 system at low temperature to CH4 yields at higher temperatures,

292 we operated two additional AD reactors without electrodes and

PT
293 without electrical energy input, ADC-15 at 15 and ADC-30

RI
294 at 30 . As shown in Fig. 3(b), the cumulative CH4 yields

increased with the increasing temperature during the three

SC
295

296 batches. BES-AD-10 reactor had higher CH4 yield than both

U
297 control reactors at 10 and 15 , and reached a CH4 yield
AN
298 close to the ADC reactor at 30 .
M

299 Table 2 gives an overview of the CH4 production rates of

300 all the reactors at different temperatures during three

D

consecutive batches. In each reactor, the maximum CH4

TE

301

302 production rate obtained from the first time period (0-5 days)
EP

303 was 2-3 times higher than the average CH4 production rate in

304 the whole batch period.

C
AC

305 3.2 Acetate removal in AD with and without BES

306 Acetate removal efficiency for all reactors operated at 10

307 was analyzed to evaluate the overall process (Fig. 4). For all the

308 reactors, acetate was removed throughout each full batch,

309 showing that acetate was not limiting CH4 production. Acetate

15
 ACCEPTED MANUSCRIPT

310 removal efficiency in BES-AD-10 was 45%, while acetate

311 removal efficiency in ADC-10 was 8%. The difference in

312 acetate removal efficiency between BES-AD-10 and ADC-10

313 could be the result of anodic oxidation by electroactive

PT
314 microorganisms (Zhao et al. 2014). However, BES-ADC-10

RI
315 also had high acetate removal efficiency of 40%, which was

316 comparable to that of BES-AD-10, even though there was no

SC
317 electric circuit and CH4 production was much lower. Apparently,

U
318 the presence of electrodes in the form of granular activated
AN
319 carbon, influences acetate removal. It is well known that

320 granular activated carbon possesses excellent adsorption

M

321 capability for organic matter (Gur-Reznik et al. 2008,

D

322 Orshansky and Narkis 1997), although adsorption of acetate to

TE

323 granular activated carbon bioanodes has not been reported or

324 discussed so far. Adsorption tests in batch bottles with acetate

EP

325 and granular activated carbon indeed showed that up to 25% of

C

326 the acetate was adsorbed after around 2 days (data not shown).
AC

327 The acetate removal efficiency of BES-ADC-10 decreased

328 from 40% to 28% during three batches, while CH4 production

329 rates increased, which is another indication of saturation of

330 granular activated carbon with acetate. Not only acetate

331 removal, but also CH4 production efficiency was influenced

16
 ACCEPTED MANUSCRIPT

332 by adsorption, because adsorption of part of the removed

333 acetate would result in lower CH4 production efficiency. Indeed,

334 CH4 production efficiency in the reactors with granular

335 activated carbon as electrode material (both BES-ADC-10 and

PT
336 BES-AD-10), were 8 times lower than that without electrode

RI
337 materials (ADC), see Table S1.

3.3 Bioelectrochemical analysis of anode and cathode

SC
338

339 At low temperature, acetoclastic methanogenesis is more

340
U
rate limiting than hydrogenotrophic methanogenesis (Enright et
AN
341 al. 2009). A potential mechanism for CH4 generation in the
M

342 BES-AD-10 reactor is that introducing a bioanode opens a new

343 route for CH4 formation, in which acetate is converted into

D

electrons (Fig 1, R4), and the applied voltage is used to generate

TE

344

345 either H2 (R5) or CH4 (R6) at the cathode. To further analyze the
EP

346 role of the (bio)electrodes in the BES-AD-10 reactor,

347 volumetric current density, coulombic efficiency and cathodic

C
AC

348 CH4 recovery efficiency were analysed (Fig. 5). Coulombic

349 efficiency can be used to distinguish the contribution of anodic

350 oxidation and acetoclastic methanogenesis in acetate removal

351 (Zhao et al. 2014). Under a constant cathode potential -0.9 V,

352 an average volumetric current density of -10 A/m3 was

353 observed (reported negative because it was measured as a

17
 ACCEPTED MANUSCRIPT

354 cathodic current) (Fig. 5). The current density was quite stable

355 during three batches. Anode potential was regularly measured

356 and was always <-350 mV, indicating that the bioanode was

357 active throughout the experiment. Coulombic efficiency

PT
358 increased from 44% to 60%, while the acetate removal

RI
359 efficiency in BES-AD-10 ranged between 37% and 43%

360 (Fig.6). This increase in coulombic efficiency at stable acetate

SC
361 removal efficiencies showed that the contribution of the

U
362 bioanode to acetate oxidation increased with time, which was
AN
363 likely caused by the growth of electroactive microorganisms in

364 the anode. Because CH4 production rate in BES-AD-10 was

M

365 several times higher than those of BES-ADC-10 and AD-10

D

366 throughout the experiment, we can conclude that BES became

TE

367 an alternative route to degrade acetate and produce CH4.

368 Cathodic CH4 recovery efficiency shows which part of the

EP

369 electrical current is converted into CH4 (either direct of via

C

370 hydrogen) at the cathode. The highest cathodic CH4 recovery

AC

371 efficiency was achieved around 20% in the first batch, and

372 gradually increased to 25% in the third batch, which was

373 relatively low compared to those in previous

374 methane-producing BES study at higher temperature of 30

375 (Siegert et al. 2015). The causes for the low cathodic CH4

18
 ACCEPTED MANUSCRIPT

376 recovery have not been further examined in our study, but

377 could be the result of H2 diffusion to the anode side and

378 re-oxidation by the anode electroactive bacteria (Lee and

379 Rittmann 2009). In our study, hydrogen concentration was

PT
380 analyzed regularly and was below the detection limit during all

RI
381 batches. Even though not detected, hydrogen could be an

382 intermediate product and used in-situ by hydrogenotrophic

SC
383 methanogens (van Eerten-Jansen et al. 2015, Lohner et al.

U
384 2014). Other possible causes for the low CH4 recovery may be
AN
385 biomass growth, or oxygen leakage during medium

386 replacement. The combination of anodic and cathodic

M

387 efficiency shows which part of the acetate is converted into

D

388 CH4 in the BES. This overall efficiency was around 9.5 %,
TE

389 indicating there is still much room for improvement in

390 conversion efficiency.

EP

391 Polarization curves of the biocathode in BES-AD-10 were

C

392 measured at the end of each batch (Fig.6), and also for a control
AC

393 cathode (without inoculum). The biocathode in BES-AD-10 in

394 all of the three batches had similar current densities, which

395 were higher than that of the control cathode. At a cathode

396 potential of -0.90 V, the biocathode in BES-AD-10 had a

397 current density of -23 A/m3, whereas the control cathode had a

19
 ACCEPTED MANUSCRIPT

398 current density around -0.50 A/m3. This indicated the formation

399 of a cathodic biofilm, however, it cannot be concluded if this

400 biofilm was catalytic for the production of H2 and/or CH4.

401 3.4 Evaluation of the enhancement of BES on AD at low

PT
402 temperature

RI
403 The maximum specific methanogenic activity (SMA) of

SC
404 the BES-AD-10 reactor was 3.3 mg CH4-COD/(gVSS·d). SMA

405 of anaerobic sludge treating domestic sewage was about 300

406 mg CH4-COD/ (gVSS·d) at 30

U and 30-40 mg CH4-COD/
AN
407 (gVSS·d) at 15 (Zhang et al. 2013b). Maximum CH4
M

408 production rate of the ADC-10 reactor in this study was lower

409 than related results above, and was probably due to the
D

differences in reactor design, substrate, and lower temperature

TE

410

411 of 10 . Strategies to achieve higher SMA could be to use a

EP

412 longer cultivation time for the sludge in the reactor (De Vrieze

413 et al. 2014), or to inoculate with sludge from low-temperature

C
AC

414 anaerobic digester(Zeeman et al. 1988), aspects that need

415 further research.

416 CH4 production rates of the previous studies on BES-AD

417 systems under mesophilic conditions and the present study are

418 summarized in Table 3. CH4 production rate was different in

20
 ACCEPTED MANUSCRIPT

419 different BES-AD systems, as it was affected by many factors,

420 such as inoculum, reactor configuration, applied voltage and

421 medium. However, the increased CH4 production rate in our

422 study outperformed most BES-AD systems operated at

PT
423 mesophilic conditions, except studies that operated with higher

RI
424 external voltage than this study (Zamalloa et al. 2013, Guo et al.

425 2013).

SC
426 The energy input of the BES-AD system was calculated in

U
427 terms of electrical energy invested per volume (m3) of
AN
428 produced CH4 during one batch. In this experiment, the energy
C×2× ×
input for BES-AD-10 reactor was 39 kWh/m3 CH4 ( ,
M

429
D

430 U is the cell voltage 0.55 V; I is the volumetric current density

D

431 10 A/m3; V is the reactor volume 20 mL ; T is the total three

TE

432 batches time 41 days; P is the total CH4 yield during three
EP

433 batches 78.6 mg CH4-COD/ (gVSS) ). If this same amount of

434 electrical energy would be used to heat up the reactor, which

C

435 would also lead to higher conversion rates, the temperature of

AC

G
436 the reactor would only increase by 1.6 (∆F = H∙ , Q is the

437 electrical energy during one batch; m is the mass of the

438 medium 20 g; C is the heat capacity 4.179 J/ /g). The energy

439 input that would be required for heating up the digester from

440 10 to 30 in our experiment, would be, with the same

21
 ACCEPTED MANUSCRIPT

441 calculation, 407.4 kWh/m3 CH4 , which was almost 10 times

442 higher than electrical energy invested in the BES-AD-10

443 reactor. If the cathodic CH4 recovery efficiency could be

444 increased from 25% to 80% this would lead to a further

PT
445 decrease in electricity input to 12 kWh/m3 CH4. Therefore, in

RI
446 addition to other processes aiming at enhancing CH4 production

447 during low-temperature anaerobic digestion, e.g. co-digestion

SC
448 (Zhang et al. 2013b) and combing a solar energy heating-up

U
449 system (Ren et al. 2012), BESs may serve as a cost-efficient
AN
450 alternative to enhance CH4 production at low temperature.

4. Conclusions
M

451

452 At 10 , CH4 yield in the integrated BES-AD system

D

was 5.3~6.6 times higher than those of the control (no external
TE

453

454 voltage and no electrodes). The comparison between reactors

EP

455 operated at different temperatures suggested that CH4

456 production from a low-temperature anaerobic digestion assisted

C
AC

457 by BES might result in a similar performance as mesophilic

458 anaerobic digestion. Energy input by the form of electricity in

459 BES-AD system was lower than the energy for heating up the

460 digester to mesophilic temperature. This study demonstrated

461 that BES has potential to be an alternative strategy to enhance

462 CH4 production in low-temperature anaerobic digestion.

22
 ACCEPTED MANUSCRIPT

463 Acknowledgements

464 This study was financially supported by Alliander

465 (Netherlands) and DMT Environmental Technology together

466 with the Chinese Scholarship Council.

PT
RI
U SC
AN
M
D
TE
C EP
AC

23
 ACCEPTED MANUSCRIPT
467 Reference
468 Speece, R.E. (2008) Anaerobic Biotechnology and Odor/corrosion Control
469 for Municipalities and Industries, Fields Publishing, Incorporated.
470 Hussain, A. and Dubey, S.K. (2015) Specific methanogenic activity test for
471 anaerobic degradation of influents. Applied Water Science, 1-8.
472 Lettinga, G., Rebac, S. and Zeeman, G. (2001) Challenge of psychrophilic
473 anaerobic wastewater treatment. Trends In Biotechnology 19(9), 363-370.
474 Zhang, L., Hendrickx, T.L.G., Kampman, C., Temmink, H. and Zeeman, G.

PT
475 (2013a) Co-digestion to support low temperature anaerobic pretreatment of
476 municipal sewage in a UASB-digester. Bioresource Technology 148,
477 560-566.

RI
478 Álvarez, J.A., Armstrong, E., Presas, J., Gómez, M. and Soto (2004)
479 Performance of a UASB-Digester System Treating Domestic Wastewater.

SC
480 Environmental Technology 25(10), 1189-1199.
481 Mahmoud, N., Zeeman, G., Gijzen, H. and Lettinga, G. (2004) Anaerobic
482 sewage treatment in a one-stage UASB reactor and a combined
483 UASB-Digester system. Water Research 38(9), 2348-2358.

U
484 Witarsa, F. and Lansing, S. (2015) Quantifying methane production from
AN
485 psychrophilic anaerobic digestion of separated and unseparated dairy
486 manure. Ecological Engineering 78, 95-100.
487 Van Eerten-Jansen, M.C.A.A., Heijne, A.T., Buisman, C.J.N. and Hamelers,
488 H.V.M. (2012) Microbial electrolysis cells for production of methane from
M

489 CO2: long-term performance and perspectives. International Journal Of

490 Energy Research 36(6), 809-819.
D

491 Cheng, S.A., Xing, D.F., Call, D.F. and Logan, B.E. (2009) Direct
492 Biological Conversion of Electrical Current into Methane by
TE

493 Electromethanogenesis. Environmental Science & Technology 43(10),

494 3953-3958.
495 Villano, M., Aulenta, F., Ciucci, C., Ferri, T., Giuliano, A. and Majone, M.
EP

496 (2010) Bioelectrochemical reduction of CO2 to CH4 via direct and indirect
497 extracellular electron transfer by a hydrogenophilic methanogenic culture.
498 Bioresource Technology 101(9), 3085-3090.
C

499 Feng, Y., Zhang, Y., Chen, S. and Quan, X. (2015) Enhanced production of
500 methane from waste activated sludge by the combination of high-solid
AC

501 anaerobic digestion and microbial electrolysis cell with iron-graphite

502 electrode. Chemical Engineering Journal 259, 787-794.
503 Tartakovsky, B., Mehta, P., Bourque, J.S. and Guiot, S.R. (2011)
504 Electrolysis-enhanced anaerobic digestion of wastewater. Bioresource
505 Technology 102(10), 5685-5691.
506 Zamalloa, C., Arends, J.B.A., Boon, N. and Verstraete, W. (2013)
507 Performance of a lab-scale bio-electrochemical assisted septic tank for the
508 anaerobic treatment of black water. New Biotechnology 30(5), 573-580.
509 De Vrieze, J., Gildemyn, S., Arends, J.B.A., Vanwonterghem, I., Verbeken,
510 K., Boon, N., Verstraete, W., Tyson, G.W., Hennebel, T. and Rabaey, K.

24
 ACCEPTED MANUSCRIPT
511 (2014) Biomass retention on electrodes rather than electrical current
512 enhances stability in anaerobic digestion. Water Research 54(0), 211-221.
513 Liu, W., Cai, W., Guo, Z., Wang, L., Yang, C., Varrone, C. and Wang, A.
514 (2016) Microbial electrolysis contribution to anaerobic digestion of waste
515 activated sludge, leading to accelerated methane production. Renewable
516 Energy 91, 334-339.
517 Enright, A.-M., McGrath, V., Gill, D., Collins, G. and O’Flaherty, V. (2009)
518 Effect of seed sludge and operation conditions on performance and archaeal

PT
519 community structure of low-temperature anaerobic solvent-degrading
520 bioreactors. Systematic and Applied Microbiology 32(1), 65-79.
521 Yang, Y., Guo, J. and Hu, Z. (2013) Impact of nano zero valent iron (NZVI)

RI
522 on methanogenic activity and population dynamics in anaerobic digestion.
523 Water Research 47(17), 6790-6800.

SC
524 Lu, L., Ren, N., Zhao, X., Wang, H., Wu, D. and Xing, D. (2011) Hydrogen
525 production, methanogen inhibition and microbial community structures in
526 psychrophilic single-chamber microbial electrolysis cells. Energy &
527 Environmental Science 4(4), 1329-1336.

U
528 van Eerten-Jansen, M.C.A.A., Jansen, N.C., Plugge, C.M., de Wilde, V.,
AN
529 Buisman, C.J.N. and ter Heijne, A. (2015) Analysis of the mechanisms of
530 bioelectrochemical methane production by mixed cultures. Journal of
531 Chemical Technology & Biotechnology 90(5), 963-970.
532 Zhang, L., Hendrickx, T.L.G., Kampman, C., Temmink, H. and Zeeman, G.
M

533 (2013b) Co-digestion to support low temperature anaerobic pretreatment of

534 municipal sewage in a UASB–digester. Bioresource Technology 148(0),
D

535 560-566.
536 Clauwaert, P. and Verstraete, W. (2009) Methanogenesis in membraneless
TE

537 microbial electrolysis cells. Applied Microbiology and Biotechnology 82(5),

538 829-836.
539 Wang, X., Feng, Y., Ren, N., Wang, H., Lee, H., Li, N. and Zhao, Q. (2009)
EP

540 Accelerated start-up of two-chambered microbial fuel cells: Effect of anodic

541 positive poised potential. Electrochimica Acta 54(3), 1109-1114.
542 Van Eerten-Jansen, M.C.A.A., Ter Heijne, A., Grootscholten, T.I.M.,
C

543 Steinbusch, K.J.J., Sleutels, T.H.J.A., Hamelers, H.V.M. and Buisman, C.J.N.
544 (2013) Bioelectrochemical Production of Caproate and Caprylate from
AC

545 Acetate by Mixed Cultures. ACS Sustainable Chemistry & Engineering 1(5),
546 513-518.
547 Deeke, A., Sleutels, T.H.J.A., Hamelers, H.V.M. and Buisman, C.J.N. (2012)
548 Capacitive Bioanodes Enable Renewable Energy Storage in Microbial Fuel
549 Cells. Environmental Science & Technology 46(6), 3554-3560.
550 Chen, Y., Cheng, J.J. and Creamer, K.S. (2008) Inhibition of anaerobic
551 digestion process: A review. Bioresource Technology 99(10), 4044-4064.
552 Bo, T., Zhu, X., Zhang, L., Tao, Y., He, X., Li, D. and Yan, Z. (2014) A new
553 upgraded biogas production process: Coupling microbial electrolysis cell
554 and anaerobic digestion in single-chamber, barrel-shape stainless steel

25
 ACCEPTED MANUSCRIPT
555 reactor. Electrochemistry Communications 45, 67-70.
556 Zhao, Z., Zhang, Y., Chen, S., Quan, X. and Yu, Q. (2014)
557 Bioelectrochemical enhancement of anaerobic methanogenesis for high
558 organic load rate wastewater treatment in a up-flow anaerobic sludge
559 blanket (UASB) reactor. Sci. Rep. 4.
560 Gur-Reznik, S., Katz, I. and Dosoretz, C.G. (2008) Removal of dissolved
561 organic matter by granular-activated carbon adsorption as a pretreatment to
562 reverse osmosis of membrane bioreactor effluents. Water Research 42(6–7),

PT
563 1595-1605.
564 Orshansky, F. and Narkis, N. (1997) Characteristics of organics removal by
565 PACT simultaneous adsorption and biodegradation. Water Research 31(3),

RI
566 391-398.
567 Siegert, M., Li, X.-f., Yates, M.D. and Logan, B.E. (2015) The presence of

SC
568 hydrogenotrophic methanogens in the inoculum improves methane gas
569 production in microbial electrolysis cells. Frontiers in Microbiology 5.
570 Lee, H.-S. and Rittmann, B.E. (2009) Significance of biological hydrogen
571 oxidation in a continuous single-chamber microbial electrolysis cell.

U
572 Environmental Science & Technology 44(3), 948-954.
AN
573 Lohner, S.T., Deutzmann, J.S., Logan, B.E., Leigh, J. and Spormann, A.M.
574 (2014) Hydrogenase-independent uptake and metabolism of electrons by the
575 archaeon Methanococcus maripaludis.
576 Zeeman, G., Sutter, K., Vens, T., Koster, M. and Wellinger, A. (1988)
M

577 Psychrophilic digestion of dairy cattle and pig manure: Start-up procedures
578 of batch, fed-batch and CSTR-type digesters. Biological Wastes 26(1),
D

579 15-31.
580 Guo, X., Liu, J. and Xiao, B. (2013) Bioelectrochemical enhancement of
TE

581 hydrogen and methane production from the anaerobic digestion of sewage
582 sludge in single-chamber membrane-free microbial electrolysis cells.
583 International Journal Of Hydrogen Energy 38(3), 1342-1347.
EP

584 Ren, Z., Chen, Z., Hou, L.-a., Wang, W., Xiong, K., Xiao, X. and Zhang, W.
585 (2012) Design investigation of a solar energy heating system for anaerobic
586 sewage treatment. Energy Procedia 14, 255-259.
C

587
AC

26
 ACCEPTED MANUSCRIPT

Table 1. Overview of experimental conditions at which the reactors were operated

Reactor BES-AD-10 BES-ADC-10 ADC-10 ADC-15 ADC-30

Temperature (℃) 10 10 10 15 30

Electrode materials +a + -b - -

PT
Cathode potential(V) -0.9 0 0 0 0

RI
a. Reactor was equipped with activated carbon granules.
b. Reactor was without any activated carbon granules.

U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 2. Comparison of averagea and maximumb CH4 production rate for the reactors

during each batch

No. Batch 1 Batch 2 Batch 3

mgCOD/(gVSS·d) Average Max Average Max Average Max
BES-AD-10 1.55 2.22 2.37 3.19 2.13 3.32

PT
BES-ADC-10 0.41 0.68 0.16 0.54 0.25 0.48
ADC-10 0.36 0.45 0.40 0.49 0.44 0.51

RI
ADC-15 1.13 2.29 1.35 2.35 1.50 2.41
ADC-30 2.29 5.55 2.62 5.91 2.98 7.62

SC
a. The average CH4 production rate was calculated from the whole batch period (2 weeks).
b. The maximum CH4 production rate was calculated from the first period (0-5 days).

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 3. Comparison between BES-AD and AD in terms of CH4 production rate

Improvement
of BES-AD
Applied Production Rate in AD
compared to
Temperature voltage alone Reference
control
(V) mg CH4-COD/(gVSS·d)
without

PT
electrodes
(Feng et al.
35℃ 0.3~0.6 18 1.22 a times
2015)

RI
(Zhao et al.
25±2℃ 0.7~0.8 557 1.25b times
2014)

SC
(De Vrieze et al.
34℃ 0.5~1 32 1.5a times
2014)

U
37±1℃ 1.4~1.8 0.47 11.4~13.6a times (Guo et al. 2013)
AN
(Zamalloa et al.
30±2℃ 2±0.1 0.17 5a times
2013)
M

(Tartakovsky et al.
35℃ 2.8~3.5 143 1.1~1.25a times
2011)

5~6 a times
D

10℃ 0.55 0.44 This study

TE

a. CH4 yields
b. CH4 production rate
C EP
AC
 ACCEPTED MANUSCRIPT
(a)

PT
RI
SC
(b)

U
AN
M
D
TE

Fig.1 (a) Process scheme of the single chamber BES-AD system. (b) Overview of main
EP

reactions involved in BES-AD system. R1: hydrogenotrophic methanogenesis (H2

CH4); R2: Acetoclastic methanogenesis (acetate  CH4); R3: Acetogenesis and

C
AC

homoacetogenesis, which were specially focused on syntrophic oxidation of acetate to

CO2 and H2, and acetate production from CO2 and H2. R4: Oxidation of organic matter

by Anode-Respiring Bacteria at anode electrode (herein acetateelectrons) ;R5: H+

reduction to H2 at cathode electrode; R6: Direct electron up-take by methanogens

(electronsCH4).
 ACCEPTED MANUSCRIPT

PT
RI
SC
Fig.2 Experimental setup of the BES-AD system

U
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
(a) 

PT
RI
SC
 

U
(b)
AN
M
D
TE
C EP

 
AC

 
Fig.3. CH4 yield of the different reactors and conditions during three consecutive
batches (a) CH4 yield of three reactors at 10℃. (b) CH4 yield of four reactors at
different temperature, 10℃, 15℃ and 30℃. The dashed line indicates medium
replacement.   
 ACCEPTED MANUSCRIPT

PT
RI
SC
Fig.4 Acetate removal efficiency of reactors at 10℃ during three consecutive batches.

U
AN
The dashed line indicates medium replacement.
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
SC
Fig.5 Trend of current density (line), coulombic efficiency (empty circle) and cathodic

U
CH4 recovery efficiency (filled circle) during the three consecutive batches in
AN
BES-AD-10 reactor. The dashed line indicates medium replacement.
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
SC
Fig.6 Polarization curve for BES-AD-10 reactor at the end of each batch. The control

U
AN
cathode was measured for the same BES-AD-10 reactor with fresh medium and without

sludge.
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Highlights

 BES enhances CH4 production in anaerobic digestion (AD) at 10℃

 BES provides an alternative route for CH4 production in AD at 10℃

 Performance of BES-enhanced AD at 10℃ is similar to AD at mesophilic condition

PT
BES uses less energy to assist AD at 10℃than heating up AD to mesophilic condition

RI
U SC
AN
M
D
TE
C EP
AC