Bioresource Technology 249 (2018) 809–817

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Analysis of a bio-electrochemical reactor containing carbon fiber textiles for T

the anaerobic digestion of tomato plant residues
⁎
Shin-ichi Hirano , Norio Matsumoto
Environmental Chemistry Sector, Environmental Science Research Laboratory, Central Research Institute of Electric Power Industry (CRIEPI), Chiba, Japan

G RA P H I C A L AB S T R A C T

A R T I C L E I N F O A B S T R A C T

Keywords: A bio-electrochemical system packed with supporting material can promote anaerobic digestion for several types
Electrochemical fermentation of organic waste. To expand the target organic matters of a BES, tomato plant residues (TPRs), generated year-
Anaerobic digestion round as agricultural and cellulosic waste, were treated using three methanogenic reactors: a continuous stirred
Tomato residues tank reactor (CSTR), a carbon fiber textile (CFT) reactor, and a bio-electrochemical reactor (BER) including CFT
Hydrogenotrophic methanogen
with electrochemical regulation (BER + CFT). CFT had positive effects on methane fermentation and metha-
nogen abundance. The microbial population stimulated by electrochemical regulation, including hydro-
genotrophic methanogens, cellulose-degrading bacteria, and acetate-degrading bacteria, suppressed acetate
accumulation, as evidenced by the low acetate concentration in the suspended fraction in the BER + CFT. These
results indicated that the microbial community in the BER + CFT facilitated the efficient decomposition of TPR
and its intermediates such as acetate to methane.

1. Introduction generate electricity, hydrogen, or other products (Lu and Ren, 2016;
Wang et al., 2015b). Electro-fermentation (EF), a term first proposed by
A bio-electrochemical system (BES) is a type of bioreactor in which Rabaey, is a new BES method that has recently been applied to the
both biological and electrochemical processes can take place to traditional fermentation process (Moscoviz et al., 2016; Roy et al.,

Abbreviations: BER, bio-electrochemical reactor; BES, bioelectrical system; CFT, carbon fiber textile; CSTR, continuous stirred tank reactor; DGGE, denaturing gradient gel electro-
phoresis; EF, electro-fermentation; HRT, hydraulic retention time; OLR, organic loading rate; ORP, oxidation reduction potential; qPCR, quantitative polymerase chain reaction; SS,
suspended solid; TPR, tomato plant residue; VFA, volatile fatty acid
⁎
Corresponding author at: Environmental Chemistry Sector, Environmental Science Research Laboratory, Central Research Institute of Electric Power Industry (CRIEPI), 1646 Abiko,
Abiko-Shi, Chiba-ken 270-1194, Japan.
E-mail address: s-hirano@criepi.denken.or.jp (S.-i. Hirano).

http://dx.doi.org/10.1016/j.biortech.2017.09.206
Received 1 August 2017; Received in revised form 28 September 2017; Accepted 30 September 2017
Available online 12 November 2017
0960-8524/ © 2017 Elsevier Ltd. All rights reserved.
S.-i. Hirano, N. Matsumoto
2016). Electrochemical oxidizing or reducing reactions on electrodes in
a BES can modify the oxidation reduction potential of a medium (ex-
tracellular ORP), which subsequently affects the intracellular ORP via
the reduced/oxidized NAD (NADH/NAD+) balance, impacting the
metabolic process (Berrios-Rivera et al., 2002). In EF, redox environ-
ments are controlled to stabilize fermentation and obtain high yields of
target products via electrochemical reactions on electrodes (Moscoviz
et al., 2016). Unlike other types of BES, EF does not require high current
densities because the electric current is a source of reducing or oxi-
dizing power to optimize extracellular and intracellular redox condi-
tions, and is not the main energy source for microbial reactions. To take
advantage of this property, the method has been applied to the fer-
mentative production of energy products such as ethanol, butanol, and
energy-rich biogas.
Anaerobic digestion is a popular industrial fermentation process for
the treatment of biodegradable solid waste (e.g., food waste, sewage
sludge, and agricultural plant residues) and for renewable energy pro-
duction. Improving the performance of anaerobic digestion is a key
challenge for increasing the energetic and economic performance of
these processes. The EF method was previously applied to anaerobic
digestion with the goal of improving energy recovery from organic
waste, including garbage slurry, sewage sludge, and filter paper as
model cellulose substrates (Sasaki et al., 2011a,b, 2013b). In all of these
studies, an applied reductive potential of −0.6 V or −0.8 V (vs. Ag/
AgCl) in the bio-electrochemical reactor (BER) successfully stabilized
methane gas production at a higher organic loading rate (OLR) and
increased energy yield (methane) in comparison with those obtained in
a control experiment without electrolysis. Although a correlation be-
tween total methane fermentation activity and applied potential on a
working electrode has been suggested, the microbes that are influenced
by the electrochemical potential in the BER and the mechanism by
which methane fermentation is improved remains unclear.
In this study, to evaluate the effect of electrochemical regulation on
microbial communities in anaerobic digestion as well as the mechanism
explaining previous electrochemical methane fermentation results, a
fed-batch BER with an attached carbon fiber textile (CFT) on a working
electrode for anaerobic digestion was developed. Attachment of the CFT
to the working electrode was previously shown to result in high reactor
performance by retaining a high cell density (Sasaki et al., 2011b).
Moreover, since application of the EF method to anaerobic digestion
has not been examined using actual cellulosic waste to date, in this
study, agricultural tomato plant residue (TPR) was used as a cellulosic
feedstock of anaerobic digestion. TPR is an agricultural cellulosic waste
that is targeted for effective utilization because it is discharged year-
round from greenhouses and is usually landfilled in nearby production
sites or combusted as a garbage. Although there has been no in-
vestigation on the use of EF for agricultural waste such as TPR waste, EF
has shown potential for such application with a model cellulose sub-
strate (Sasaki et al., 2011a, 2013a). During the stable anaerobic di-
gestion of TPR in BERs, the gas production rate was measured and
microbial communities were analyzed by both DNA- and RNA-based
denaturing gradient gel electrophoresis (DGGE) and real-time PCR to
detect actively growing microbes.
2. Materials and methods
2.1. Feed material and inoculum
Agricultural TPRs, including stalks, leaves, and residual tomatoes,
were obtained from a plant factory (with a closed growing system for
the year-round production of vegetables) in Abiko, Japan. Collected
TPRs were air-dried and shredded to a particle size of less than 2 mm
using a food processer and kitchen blender. The TPR feedstock for
anaerobic digestion was composed of 10% TPR and medium, including
(p liter) 0.8 g of KH2PO4, 1.6 g of K2HPO4, 1.0 g of NH4Cl, 2.0 g of
NaHCO3, 0.1 g of MgCl2·6H2O, 0.2 g of CaCl2·6H2O, and 0.8 g of NaCl,
810
Bioresource Technology 249 (2018) 809–817
Potentiostat
Gas bag
Reference
electrode
Counter
side
Working Carbon fiber textile
electrode
Fig. 1. Schematic diagram of the bio-electrochemical reactor (BER) containing carbon
fiber textiles (CFT). The same configuration was used for the control reactor without
electrochemical regulation (CFT reactors). The electrodes (working, counter, and re-
ference electrode) were not included in the continuous stirred tank reactor (CSTR).
along with 10 mL of DSMZ medium 131 trace element solution and
DSMZ medium 141 vitamin solution. Anthraquinone 2,6-disulfonate
was added to the medium as an electron mediator to facilitate the
electrochemical redox regulation, as in previous reports (Sasaki et al.,
2011a; Hirano et al., 2013; Chen et al., 2016). The TPR feedstock was
autoclaved at 121 °C for 30 min.
2.2. BERs containing carbon fiber textile (CFT) for electrochemical
methane fermentation
All electrochemical methane fermentation experiments were per-
formed using an apparatus comprising a bottle-type single chamber. A
three-electrode system, which included a working electrode with a CFT,
a reference electrode, and a counter electrode, was used (BER + CFT,
Fig. 1). The anodic bag was formed using a proton exchange membrane
(Nafion 117; DuPont Co., Wilmington, DE, USA) and inserted in the
cathodic working chamber. A carbon plate (25 × 75 × 1 mm) was used
as the working electrode and a smaller carbon plate (10 × 65 × 1 mm)
was used as the counter electrode. On the working electrode, CFT
(25 × 75 × 2.4 mm) was attached to the side of the reference electrode
using Pt wire, as previously described (Sasaki et al., 2011b). An Ag/
AgCl reference electrode was inserted into the cathodic working
chamber. A gas bag (Aluminum Bag; GL Sciences, Tokyo, Japan) was
connected to each chamber via a silicon cap on the top of reactors for
the collection of gas produced during the operation. The cathodic
chamber and the anodic bag were filled with 250 mL of medium and
5 mL of 100 mM NaCl, respectively. In the BER, three electrodes were
connected with a potentiostat (PS-08; Tohogiken, Yokohama, Japan).
Since a positive effect of regulated potential at −0.8 V on stable me-
thane production and decomposition of cellulosic substrate was pre-
viously observed (Sasaki et al., 2011a, 2013a), the potential of the
working electrode was electrochemically regulated to −0.8 V (vs. Ag/
AgCl) in the present study. All reported potentials on the working
electrode pertain to the Ag/AgCl reference electrode (type: saturated
KCl). BER containing CFT was prepared in triplicate. Three reactors
containing CFT in which the potential at the working electrode was not
electrochemically regulated were prepared as control 1 (CFT reactors).
In addition, three CSTRs without CFT and electrochemical regulation
were operated as control 2. The contents of reactors were thoroughly
mixed using a magnetic stirrer.
2.3. Operation of reactors
First, the same volume (250 mL) of sludge from the methane fer-
mentation (55 °C) of artificial solid garbage slurry (Sasaki et al.,
2011a,b) was injected into each working chamber after it was filled
S.-i. Hirano, N. Matsumoto Bioresource Technology 249 (2018) 809–817

Fig. 2. (a) Time-dependent changes in the gas production

(a)
rates and organic loading rate (OLR) in the three reactor
types: continuous stirred tank reactor (CSTR), carbon fiber
400 1200 textile (CFT) reactors, and bio-electrochemical reactors
(BERs) + CFT. Error bars show standard deviations for the
mean of triplicates. Owing to deterioration, CSTRs were
Gas production rate (mL·L-1·day-1)

stopped at day 113. The remaining two reactors types

OLR (mg-COD·L-1·day-1)
containing CFT were stopped at day 122. (b) Electric cur-
300 900
rent changes in BERs + CFT for 3 days from a sampling
point to the next sampling point. Representative data in
each OLR condition are shown.

200 600

100 300

0 0
10 30 50 70 90 110 130
Incubation time (day)

CSTR CFT BES+CFT OLR

(b)

0.0E+00
Electric current (A/reactor)

-5.0E-04

-1.0E-03
day 50 day 85 day 107
-1.5E-03

-2.0E-03

-2.5E-03

-3.0E-03
0 1000 2000 3000 4000 5000

Incubation time (min)

with nitrogen gas (gas phase). Next, all chambers were sealed with a periodically quantified (every 12–15 h) from day 113 to 116. Moreover,
silicon stopper. The cathodic working chambers and the control re- from day 122 to 125, applied potentials in the three BERs + CFT were
actors without electrochemical regulation (control 1) or without CFT turned off and the produced gas and acetate were also periodically
(control 2) were operated in semi-continuous mode at 55 °C as follows. measured. All reactors were stopped on day 125.
Once a day until day 25 and once every 3 days after day 25, each
predetermined volume was discharged and the same amount of fresh 2.4. Analysis of reactor performance
TPR feedstock was added under atmospheric conditions. The pH in the
working chamber was adjusted to 7.2 by adding 1 N NaOH. From days 0 The suspended fraction in all reactors was sampled and analyzed
to 10, high gas production from the residual garbage slurry was de- once every 3 days. CODcr, the chemical oxygen demand, was de-
tected and gradually decreased. After the fluctuation in the gas pro- termined using the HACH method (HACH Co., Loveland, CO, USA).
duction rate declined, the OLR was increased in a stepwise fashion by CODcr removal efficiency was calculated as follows: CODcr removal
reducing the hydraulic retention time (HRT). From days 0 to 56, 56 to efficiency (%) = [(CODcrin − CODcrout)/CODcrin] × 100, where
91, or 91 to 125, the reactors were operated for a period of more than CODcrin is the CODcr of the substrate and CODcrout is the CODcr of the
two times the length of the HRT (15, 11.5, or 8 days) at an OLR of bulk. The suspended solid (SS) content was determined according to the
641.2, 854.9, or 1139 mg-CODcr·L−1·day−1, respectively. Fig. 2(a) Japanese Industrial Standard K-0102. In short, the filtrate was dried at
summarizes the time schedule for the OLR and the HRT in the three 105 °C for 2 h and then weighed to determine the amount of SS. The SS
experimental conditions using the BER + CFT, CFT reactor, and CSTR. removal efficiency was calculated as follows: SS removal efficiency
Three reactors of control 2 without CFT and electrochemical regulation (%) = [(SSin − SSout)/SSin] × 100, where SSin is the SS of the substrate
were stopped owing to deterioration on day 110 (OLR; 1139 mg- and SSout is the SS of the bulk. During the operation, the concentrations
CODcr·L−1·day−1). In the experiment to evaluate the effect of electro- of volatile fatty acids (VFAs) in the suspended fraction were quantified
chemical regulation on the working electrode, the gas produced and by high-performance liquid chromatography with ultraviolet and in-
acetate concentration in the BER + CFT and CFT reactors were frared detectors (LaChrom Elite; Hitachi, Tokyo, Japan). A total of

811
S.-i. Hirano, N. Matsumoto
10 µL of filtered culture was loaded onto a GelPak Column (GL-C610H-
S; Hitachi) at 40 °C and eluted at 0.5 mL/min with 0.1% H2PO4 solu-
tion. The produced gas was collected in an aluminum gas bag and its
volume was measured using the water displacement method. The me-
thane, carbon dioxide, and hydrogen contents of produced gas were
measured using a gas chromatograph equipped with a thermal con-
ductivity detector (GC390B; GL Sciences) and a stainless-steel column
packed with active carbon (30/60 mesh; GL Sciences). The cellulose
components (cellulose, hemicellulose) of TPR were obtained according
to the Technical Association of the Pulp and Paper Industry (Atlanta,
GA, USA) standard.
2.5. Analysis of the microbial community
2.5.1. DNA extraction and quantitative (q)PCR assay
The suspended fraction was sampled for microbial DNA extraction
at days 37, 50, 85, 107, and 125. In addition, at day 125, for one of the
three BERs + CFT and CFT reactors, the retained cell fraction on the
CFT working electrode was collected. DNA was extracted from the
suspended fraction and the retained fraction on the CFT electrode using
the Power Max Soil Isolation Kit (MO Bio Laboratories, Carlsbad, CA,
USA). The abundance of prokaryotic and methanogenic 16S rRNA
genes was evaluated by qPCR, as described previously (Sasaki et al.,
2009). Prokaryotic-specific primer sets Uni340F and Uni806R and the
double-dye probe Uni516F (Takai and Horikoshi, 2000) were used.
Methanogenic-specific primer sets S-P-MArch-0348-S-a-17 and SD-
Arch-0786-A-a-20 and the double-dye probe S-P-MArch-0515-SFig a-25
(Sawayama et al., 2006) were also used. To calculate the total copy
number per reactor, the copy number per DNA-extracted area of CFT or
volume of the bulk was multiplied by the total area of CFT or total
suspended volume.
2.5.2. PCR-DGGE
DGGE was conducted using Bacteria domain-specific primers (341F
with GC clamp and 517R) and the Archaea-specific primers Parch340F
with a GC clamp and Parch519R (Lueders et al., 2004). The PCR con-
ditions were as follows: initial denaturation for 2 min at 98 °C for 1
cycle and 30 cycles of 94 °C for 1 min, 53 °C for 1 min, and 72 °C for
3 min, followed by 1 cycle of 72 °C for 10 min. Amplified PCR products
were confirmed by gel electrophoresis on 1.5% agarose gels. The DGGE
analysis was performed using the D-code Universal Mutation Detection
System (Bio-Rad Laboratories, Hercules, CA, USA). Fixed amounts of
PCR products were loaded onto a 10% (w/v) polyacrylamide gel with a
linear denaturant gradient ranging from 25% to 55%. Electrophoresis
was performed at 60 °C and 200 V for 300 min. After electrophoresis,
the gel was stained with SYBR Green I (Takara Bio, Kyoto, Japan). The
16S rRNA gene fragments in the DGGE gel were cut out and extracted
for sequence determination. Using the primers 341f plus 517r and
Parch340R plus Parch519R, respectively, bacterial and archaeal 16S
rRNA gene fragments were amplified from DNA bands for samples
obtained on day 125 in BER + CFT. The ligated PCR products were
transformed into Escherichia coli JM109 using a pGEM-T Easy Vector
System (Promega, Tokyo, Japan). The plasmids were extracted and
purified using a GenElute Plasmid Miniprep Kit (Sigma, Tokyo, Japan).
Sequencing reactions were carried out with a BigDye Terminator v3.1
Cycler Sequencing Kit (Applied Biosystems, Waltham, MA, USA).
Table 1
Performance of anaerobic digestion for the three reactor types.
HRT (day) OLR (mg-COD·L−1·day−1) Gas production rate (mL·L−1·day−1)
CSTR 11.5 854.9 133.3 ± 17.8
CFT 8.5 1139 180.9 ± 13.4
BER + CFT 8.5 1139 197.2 ± 11.0
Abbreviations: CSTR, continuous stirred tank reactor; CFT, carbon fiber textile; BER + CFT, bio-electrochemical reactor with carbon fiber textile.
812
Bioresource Technology 249 (2018) 809–817
Nucleotide sequencing was performed using an Applied Biosystems
3130xl Genetic Analyzer. All sequences were checked for chimeric ar-
tifacts using CHIMERA_CHECK (Ribosomal Database Project). The nu-
cleotide sequences were compared with those in GenBank/EMBL/DDBJ
using nucleotide BLAST (http://www.ncbi.nlm.nih.gov/blast/Blast.cgi)
and were deposited as accession numbers LC257996 to LC258004.
2.5.3. RNA analysis
RNA was extracted from the suspended fraction in BER + CFT and
CFT reactors at day 114 using the Max Soil RNA Extraction Kit
(MOBIO). Extracted RNA was converted to cDNA using the Prime Script
RT Reagent Kit with gDNA eraser (Takara Bio). qPCR and DGGE ana-
lyses of cDNA were conducted as described above for the DNA analysis.
3. Results and discussion
3.1. Effects of electrochemical regulation on reactor performance
Three reactor types (CSR, CFT reactor, and BER + CFT) were op-
erated with stepwise OLR increases of the TPR feedstock at 641.2 mg-
COD·L−1·day−1 from day 0–56, 854.9 mg-COD·L−1·day−1 from day
57–91, and 1139 mg-COD·L−1·day−1 from 92 to 125 (Fig. 2(a)). In
BERs with the CFT attached to one side of the working electrode, a
potential of −0.8 V was applied on the working electrode. During ex-
periments, the negative current flow of approximately −2 mA to
−0.2 mA per reactor was detected in BERs + CFT after adding the TPR
feedstock every 3 days (Fig. 2(b)). This indicated that the cathodic re-
action occurred on the working electrode. The other two reactor types
without electrochemical regulation showed higher ORPs (−0.46 to
−0.55 V). A comparison of reactor performance indicators, including
gas production rates, methane content, COD removal, and C-recovery in
gas, is summarized in Fig. 2(a) and Table 1. TPR containing 48.5% of
holocellulose (cellulose and hemicellulose) and lignin showed lower
biodegradability and lower biogas yields than those reported for arti-
ficial solid garbage slurry used as seed sludge (Sasaki et al., 2010,
2011a,b). Specifically, changing the substrate from artificial solid gar-
bage slurry to TPR decreased the gas production rate to around
100 mL·L−1·day−1 during the initial phase (OLR 641.2 mg-
COD·L−1·day−1) of the three reactor types. During the second operation
phase at an OLR of 854.9 mg-CODcr·L−1·day−1, the three reactor types
exhibited similar trends in the gas production rate (CSTRs:
133.3 mL·L−1·day−1, CFT reactors: 171.1 mL·L−1·day−1, BERs + CFT:
168.7 mL·L−1·day−1) and methane content (CSTRs: 68.3%, CFT re-
actors: 73.4%, BERs + CFT: 76.3%). Hydrogen was not detected during
the experiments for any of the three reactor types. Acetate accumula-
tion was slightly higher for CSTR (5 mM) than for the other reactor
types (< 1 mM) during this OLR phase (Fig. 3). After the OLR was
raised to 1139 mg-CODcr·L−1·day−1 on day 92, CSTR gas production
decreased substantially to 41 mL·L−1·day−1 and reactor performance
deteriorated. Therefore, the operation of CSTRs was stopped on day
113. After increasing the OLR to 1139 mg-CODcr·L−1·day−1, acetate (as
a major VFA) accumulated in the CSTR to 22 mM. A high VFA con-
centration has been reported to decrease the pH, which leads to low
anaerobic digestion (Sasaki et al., 2010). In contrast, CFT reactors and
BERs + CFT maintained stable gas production, methane contents,
CODcr removal, and C-recovery in gas at an OLR of 1139 mg-
Methane content (%) COD removal ratio (%) C-recovery in gas (%)
68.3 ± 3.2 34.0 ± 10.4 29.3
72.4 ± 5.9 40.7 ± 5.5 29.6
77.3 ± 0.2 43.5 ± 1.7 30.3
S.-i. Hirano, N. Matsumoto Bioresource Technology 249 (2018) 809–817

25 methanogens, respectively (Table 2). The ratios of methanogens to

prokaryotes in the retained fractions in CFT were 16.2% and 15.3% in
CSTR CFT reactors and BERs + CFT, respectively, whereas these ratios were
20 much lower in the suspended fraction in the three reactor types at all
Acetate concentration (mM)

CFT
BER+CFT
15
10
5 Bacterial and archaeal communities in the suspended fractions on
0 showed similar patterns for the suspended fractions: 6 species in the
60 70 80 90 100 110 120
Incubation time (day)
Fig. 3. Time-dependent changes in the accumulation of acetate in the three reactor types
for high organic loading rate (OLR) conditions (854.9, 1139 mg-CODcr·L−1·day−1). Error
bars show standard deviations for the mean of triplicates.
CODcr·L−1·day−1 (Table 1). The rate of cellulose decomposition was
53.1 ± 4.1% in CFT reactors and was 57.3 ± 3.6% in BERs + CFT on
day 122. Although BERs + CFT only showed slightly higher perfor-
mance than that of CFT reactors, there was a clear difference in the
accumulation of acetate observed at 1139 mg-CODcr·L−1·day−1
(Fig. 3). Acetate accumulation increased slowly in CFT reactors, but
remained at a low level in BERs + CFT. Electrochemical regulation on
the working electrode including CFT may explain this difference in
acetate accumulation in the suspended fraction of the reactor.
3.2. Microbial community analysis in the three reactor types
The microbial community structures in the three reactor types were
compared at each HRT condition. Prokaryotic and methanogenic
amounts in suspended fractions and retained fractions on the CFT
working electrode were evaluated by qPCR. No difference in the
abundances of prokaryotes in the suspended fraction was observed for
any HRT among the three reactor types (1.8–3.1 × 1010) (Table 2).
However, methanogens were twofold more abundant in CFT reactors
and BER + CFT at a short HRT (8.5 days) than at a long HRT (15 days),
whereas there was no difference observed with respect to HRT for the
CSTR (Table 2). Retained fractions to CFT in both CFT reactors and
BERs + CFT on day 125 contained large amounts of prokaryotes and
OLRs (1.9–4.1%). Large amounts of microbes and higher ratios of me-
thanogens to prokaryotes in the retained fraction of CFT reactors and
BERs + CFT likely influenced the microbial community in the sus-
pended fraction, and could explain the overall higher reactor perfor-
mances compared with that of the CSTR. These results were consistent
with those of a previous report in which retained cells to CFT in an
anaerobic digester for other substrates contributed to high and stable
reactor performance by avoiding the wash-out of methanogens (Sasaki
et al., 2009, 2010, 2011b).
days 50, 85, and 107 in the three reactor types and initial seed sludge
were compared using PCR-DGGE analysis (Fig. 4a). The DGGE results
130 domain Bacteria were detected (bands 1–6) and 3 species in Archaea
were detected (bands a–c) (Fig. 4c). Bacterial band 1, similar to the
hydrolytic bacterium Defluviitoga tunisiensis isolated from a thermo-
philic biogas plant (KT274710, 100% sequence identity), was present in
the initial seed sludge, and was detected in all experiments, while the
other DNA bands appeared by the addition of TPR feedstocks. Bacterial
bands 1, 3 (Anaerovorax sp., LT631507, 87% sequence identity), and 6
(Thermoanaerobacteraceae bacterium, JN656279, 93% sequence
identity) were constitutively present during the experiments at all OLR
conditions of TPR. In contrast, bacterial bands 2, 4, and 5 were detected
in the CFT reactors and BERs + CFT at higher OLR conditions (854.9
and 1139 mg-CODcr·L−1·day−1). Sequences of band 2 and band 5 were
similar to those of cellulose-degrading bacteria belonging to Herbivorax
saccincola (LN868252, 99% sequence identity) (Koeck et al., 2016) and
Herbinix hemicellulosilytica (NR_136763, 98% sequence identity) (Koeck
et al., 2015), respectively. The sequence of band 4 was similar to that of
the syntrophic acetate-degrading bacterial genus Tepidanaerobacter
(KT274715, 97% sequence identity) (Cibis et al., 2016; Wang et al.,
2015a). The increase in these se specific bacterial species depended on
the TPR loading rate and contributed to the anaerobic digestion of TPR
feedstocks. In the archaeal community, the hydrogenotrophic metha-
nogen Methanobacterium formicicum (KX344121, 99% identity, DNA
band-a) was the dominant species in all reactor types at an HRT of
15 days. At higher OLR conditions, both the aceticlastic methanogen
Methanosarcina thermophila (AP017646, 98%, DNA band-b) and the
hydrogenotrophic methanogen Methanothermobacter marburgensis
(KY684738, 99% identity, DNA band-c) increased in the CFT reactors.
The microbial communities of the retained fractions in CFT sus-
ceptible to electrochemical regulation on the working electrode were
also evaluated by PCR-DGGE after the CFT reactors and BERs + CFT

Table 2
Copy numbers of the 16S rRNA gene of and archaea in the suspended fraction of three reactor types, and in the retained fractions to CFT in two reactor types containing CFT (CFT reactors
and BERs + CFT).

HRT (day) CSTR (copies/mL) CFT reactors BER + CFT

Suspended fraction copies/mL copies/mL

Prokaryotes
15 2.7 × 1010 ± 2.4 × 109 1.8 × 1010 ± 2.1 × 109 1.9 × 1010 ± 7.4 × 109
11.5 2.8 × 1010 ± 7.1 × 109 2.4 × 1010 ± 3.5 × 109 2.4 × 1010 ± 2.2 × 109
8.5 2.2 × 1010 ± 2.3 × 109 3.1 × 1010 ± 1.1 × 1010 2.2 × 1010 ± 1.3 × 1010

Archaea
15 6.9 × 108 ± 1.0 × 108 4.5 × 108 ± 1.0 × 108 3.6 × 108 ± 1.9 × 108
11.5 8.9 × 108 ± 1.4 × 108 7.7 × 108 ± 2.5 × 108 1.0 × 109 ± 1.5 × 108
8.5 6.6 × 108 ± 1.4 × 108 9.1 × 108 ± 5.5 × 108 7.4 × 108 ± 4.7 × 108

Retained fraction to CFT copies/cm2 copies/cm2

Prokaryotes 9.7 × 1010 ± 3.1 × 1010 5.0 × 1010 ± 1.6 × 1010
Archaea 1.5 × 1010 ± 8.0 × 109 7.7 × 109 ± 1.8 × 109

Abbreviations: CSTR, continuous stirred tank reactor; CFT, carbon fiber textile; BER+CFT, bio-electrochemical reactor with carbon fiber textile; HRT, hydraulic retention time

813
S.-i. Hirano, N. Matsumoto
(a)
(b) (c)
FB Most closely related strain
1 Defluviitoga tunisiensis strain ASV2
2 Herbivorax saccincola
3 Anaerovorax sp. Marseille-P3303
4 Tepidanaerobacter sp. AS46
5 Herbinix hemicellulosilytica strain T3/55
6 Thermoanaerobacteraceae bacterium HZ254T
a
b a Methanobacterium formicicum
c b Methanosarcina thermophila
c Methanothermobacter marburgensis
were stopped at day 122. Bacterial DNA bands 1–6 in the suspended
fraction were also detected as the major species in the retained fraction,
although other minor bands were present in the retained fraction. In
contrast, only three methanogenic DNA bands were detected in the
retained fractions of CFT (Fig. 4b). M. marburgensis (DNA band c) was
commonly present in the CFT of both reactor types. M. thermophila
(DNA band b) was present in addition to M. marburgensis in CFT at-
tached to the non-electrochemically regulated carbon electrode of the
CFT reactor. In BERs + CFT, instead of a decrease of the DNA band of
M. thermophila, M. formicum appeared in the CFT on the electro-
chemically regulated electrode of BERs + CFT. These electrochemically
influenced methanogenic communities induced a difference in acetate
accumulation in the suspended fraction between CFT reactors and
BERs + CFT.
3.3. Electrochemical effects on anaerobic digestion performance
To clarify the effects of electrochemical regulation of the working
electrode on acetate accumulation in each reactor, the time courses of
acetate accumulation in the suspended fraction and gas production
were compared in detail between CFT reactors and BERs + CFT for
3 days from day 113 after the addition of TPR feedstock until the next
addition of TPR feedstock. After the addition of TPR feedstock, gas
814
Bioresource Technology 249 (2018) 809–817
Fig. 4. Microbial communities in the domains Bacteria
(upper) and Archaea (lower) of samples from seed sludge,
suspended fractions (a), and the archaeal community re-
tained fractions to CFT (b) for each HRT condition in the
three reactor types determined by denaturing gradient gel
electrophoresis (M: DNA marker, S: seed sludge, C: CSTR, F:
CFT reactors, B: BER + CFT). DNA bands are designated by
serial numbers and letters. (c) Taxon assignments based on
sequence identity for DNA bands corresponding to the serial
numbers in (a, b).
1
2
3
4
5
6
Identity (%)
100
99
87
97
98
93
99
98
99
production exhibited a linear increase in both reactor types, and the gas
production rate in BERs + CFT (2.0 mL/h) was 1.4-times higher than
that in CFT reactors (1.4 mL/h) (Fig. 5a). Acetate accumulation for the
3-day period is shown in Fig. 5b with respect to the initial acetate
concentration at day 113. Acetate accumulated at a similar rate in both
reactor types until 16 h. After 16 h, the acetate concentration in
BERs + CFT decreased immediately, while this decrease was delayed
for 37 h in CFT reactors. This result indicated that the rate of acetate
consumption in BER + CFT after 16 h was higher than that in CFT re-
actors. Accordingly, acetate accumulation in the suspended fraction of
BERs + CFT returned to approximately zero until the next feeding day
(3 days later). The same experiment was then performed for 3 days from
day 122 in BERs + CFT, without electrochemical regulation, and the
results were similar to those obtained with CFT reactors (Fig. 5a, b,
triangle). These results indicated that an applied reductive potential on
CFT influences microbial activity for gas production and acetate con-
sumption.
The effect of the electron supply for electrochemical regulation on
gas production was next evaluated. For 3 days, the negative electric
current corresponding to 75.2 C was detected in BERs + CFT
(Fig. 2(b)). If all electric currents are used for methane production via
CO2 + 8H+ + 8e− = CH4 + 2H2O (the electrochemical reaction on
the cathodic electrode) (Cheng et al., 2009), 2.6 mL of methane could
S.-i. Hirano, N. Matsumoto Bioresource Technology 249 (2018) 809–817

(a) 160
140
120 CFT reactor
Gas volume (mL)

100
BER+CFT
80
60 BER+CFT (off)

40
20
0
0 20 40 60 80
Incubation time (h)

(b)
4.5
4
Acetate concentration (mM)

3.5
3 CFT reactor

2.5
BER+CFT
2
1.5 BER+CFT (off)
1
0.5
0
0 20 40 60 80
Incubation time (h)

Fig. 5. Detailed time-dependent changes in gas production (a) and acetate accumulation
(b) in carbon fiber textile (CFT) reactors and bio-electrochemical reactors (BERs) + CFT
for a 3-day period from day 113. In addition, gas production and acetate accumulation in
BERs + CFT without electrochemical regulation (off) are shown.

be produced by electrolysis during the 3-day period in BERs + CFT at

55 °C. The calculated gas volume (2.6 mL) is substantially lower than
the increased gas volume (53 mL). Therefore, the electric current during
the experiment is not sufficient to explain the increase in gas production
by direct electrochemical conversion. The increased gas production rate
was correlated with a decrease in acetate, an intermediate of the
anaerobic digestion process with TPR feedstock. Cyclic voltammetry
confirmed that electrochemical acetate degradation on the cathodic
electrode did not occur in the voltage range applied in our study (data
not shown). These results indicated the possibility that the cathodic
reaction stimulated gas-producing reactions and acetate-consuming
reactions of microbes around the working electrode.
3.4. Electrochemical effects on the microbial community in BER + CFT
To investigate the microbial species stimulated by electrochemical
regulation, actively grown microbes during electrochemical regulation
were analyzed. Total RNA was extracted from the suspended fractions
obtained from CFT reactors and BERs + CFT at 31 h after the addition
of TPR on day 113 in the experiment, as summarized in Fig. 5. A mi-
crobial community analysis using cDNA reverse-transcribed from 16S
rRNA was performed by PCR-DGGE and qPCR. The quantities of pro-
karyotes and methanogens in BERs + CFT were higher than those in
CFT reactors (Fig. 6a). In addition, the ratio of methanogens to pro-
karyotes was 1.7-times higher in BERs + CFT than in CFT reactors.
Bacterial community analysis based on cDNA for CFT reactors and
BERs + CFT by PCR-DGGE (Fig. 6 (b)) showed similar banding patterns
to those observed in the DNA analysis (Fig. 4). No difference was ob-
served in actively growing bacteria between CFT reactors and
BERs + CFT. In both reactor types, the three most actively growing
bacterial species corresponding to bands 1, 5, and 4 were attributed to
the cellulose-degrading Defluviitoga tunisiensis, Herbinix hemi-
cellulosilytica, and the syntrophic acetate-degrading Tepidanaerobacter.
For the archaeal community, three major cDNA bands identified as M.
Fig. 6. Microbial population structure of actively growing taxa in the two reactor types,
revealed by RNA analysis. (a). Copy number of 16S rRNA in actively growing prokaryotes.
(b) Microbial diversity of actively growing bacteria (left) and archaea (right) in carbon
fiber textile (CFT) reactors and bio-electrochemical reactors (BERs) + CFT. cDNA bands
are designated using the same serial numbers and letters used to describe phylogenetic
affiliations in Fig. 4.
formicum (cDNA band-a), M. thermophila (cDNA band-b), and M. mar-
burgensis (cDNA band-c) were detected in both BERs + CFT and CFT
reactors. Among the three methanogens, the relative intensity of the
cDNA band corresponding to the aceticlastic methanogen M. thermo-
phile in CFT was weaker than that of BER + CFT, while the band cor-
responding to the hydrogenotrophic methanogens M. formicum and M.
marburgensis were similar between CFT reactors and BERs + CFT.
These results indicate that hydrogenotrophic methanogens ex-
hibited relatively active growth during electrochemical regulation in
BERs + CFT, but not in CFT reactors. There are two acetate conversion
pathways to methane in anaerobic digesters (Town et al., 2014). First,
aceticlastic methanogens directly convert acetate to methane. Second,
the syntrophic pathway of hydrogenotrophic methanogens converts
hydrogen and carbon dioxide produced by acetate-degrading bacteria.
Hydrogen consumption increases the degradation of organic substances
such as cellulose, proteins, or VFAs (Sasaki et al., 2013a; Wagner et al.,
2013; Wang et al., 2015a). Syntrophic acetate degradation via hydro-
genotrophic methanogens is typical in thermophilic, high-acetate, and
high-H2 environments in anaerobic digesters, and is correlated with
reactor performance (i.e., low acetate accumulation and high methane
production) (Karakashev et al., 2006; Hattori, 2008; Westerholm et al.,
2012). The present results indicated the possibility that electrochemical
stimulation of hydrogenotrophic methanogens such as M. formicum and
M. marburgensis in BERs + CFT influences the syntrophic relationships
of microbes with acetate-degrading bacteria such as Tepidanaerobacter,
and suppress the accumulation of acetate in BERs + CFT.

815
S.-i. Hirano, N. Matsumoto
There are several potential mechanisms by which electrochemical
regulation can affect the microbial community in BERs + CFT. RNA-
based analysis revealed active growing microbes in BERs + CFT with
electrochemical regulation at −0.8 V. Although the bacterial commu-
nities in CFT reactors and BERs + CFT were similar, the abundance of
total bacteria was higher in BERs + CFT. In contrast, the prevalence of
hydrogenotrophic methanogens over aceticlastic methanogens was
specifically observed in BERs + CFT and not in CFT reactors. The
electrochemically applied potential in BERs + CFT might support the
growth of the detected hydrogenotrophic methanogens. Indeed, pre-
vious studies on the effects of applied cathodic potential on hydro-
genotrophic methanogens, evaluated in a single culture of M. ther-
mautotrophicum (Hirano et al., 2013), co-culture of the protein-
degrading bacterium Coprothermobacter with M. thermautotrophicus
(Sasaki et al., 2012), and co-culture of the cellulose-degrading bac-
terium Clostridium clariflavum with M. thermautotrophicus (Sasaki et al.,
2013a), revealed that a low electric current promoted the growth and
methane-producing activity of the hydrogenotrophic methanogen M.
thermautotrophicum; however, this increase was insufficient for the
production of increased methane via the direct electrochemical con-
version of CO2 to methane. It has been predicted that a low electric
current could be used to control the redox potential for the reducing
environment in the culture medium to achieve a level that is optimal for
the growth and methanogenesis of hydrogenotrophic methanogens.
Although the precise mechanism underlying the electrochemical sti-
mulation of hydrogenotrophic methanogens is still unclear, the lower
redox potential of the culture medium as an exo-cellular environment
has been estimated to increase the activity of enzymes responsible for
hydrogenotrophic methanogenesis, such as hydrogenases (Hirano et al.,
2013; van Dijk and Veeger, 1981). The stimulation of hydrogenotrophic
methanogens by the applied cathodic potential in BERs + CFT could
have contributed to the slight increase in the activity and amount of
cellulose-degrading bacteria and acetate-oxidizing bacteria detected in
community analysis, because hydrogen removal by hydrogenotrophic
methanogens facilitates the degradation of cellulose (Marvin-Sikkema
et al., 1990; Sasaki et al., 2011a, 2013a).
3.5. Effectiveness of the electrochemical method for methane production
from TPR
BERs + CFT and CFT reactors clearly showed higher performance in
anaerobic digestion and methane production from TPR compared to the
CSTR (Fig. 2(a), Table 1). Therefore, the effectiveness of the applied
electrochemical potential in BERs + CFT was considered to highlight
the advantages and applicability of BERs + CFT by comparing the
parameters between the BERs + CFT and CFT reactors in detail.
For 3 days as of day 113 in the experiment (Fig. 5), the detected
electric current and average voltage between the cathodic working
electrode and counter electrode were 75.2 C and 1.8 V, respectively, in
the BERs + CFT. Therefore, an input energy of 135.4 W s/reactor (75.2
C × 1.8 V) or 135.4 J/reactor was consumed in BERs + CFT over
3 days. Over the 3 days, methane gas increased by 38.3 mL due to the
applied electrochemical potential, which was calculated from the gas
production rate (2.0 mL/h, 1.4 mL/h) and CH4 content (77.3%, 72.4%)
in the BERs + CFT and CFT reactors, respectively, under standard
conditions. Using the reaction enthalpy (DrH) for the combustion of
CH4 gas (−890.71 kJ/mol) under standard conditions (Dean, 1999),
this increase of CH4 in BER + CFT corresponded to 1.5 kJ for 3 days. An
additional 1.5 kJ of CH4 as the output energy was generated for 3 days
in the BERs + CFT, compared with that in the CFT reactors. Based on
this result, 11-times larger energy was acquired as CH4 for input electric
energy (135 J) in BERs + CFT at an OLR of 1139 mg-CODcr·L−1·day−1.
To consider the cost-effectiveness of electrochemical regulation in
BERs + CFT, the cost of electrons consumed in BERs + CFT was also
compared with the price of the increased methane relative to that in
CFT reactors (Fig. 5) at an OLR of 1139 mg-CODcr·L−1·day−1.
816
Bioresource Technology 249 (2018) 809–817
BERs + CFT processed for 3 days consumed 75.2 C, corresponding to
7.8 × 10−4 mol of electrons (calculated using the Faraday constant:
F = 96,485 C/mol). When assuming an average voltage of 1.8 V for the
BER, this equates to a price of 0.34¢ US/mol of electrons (Harnisch
et al., 2015). An additional cost of 2.7 × 10−4¢ US as electrons was
necessary to obtain 1.7 mmol greater methane production in
BERs + CFT than that in the CFT reactors for 3 days. When assuming
the price of methane of 23.9¢ US/kg (0.2€/kg in the original paper by
Shafiei et al., 2011), the price of this increased methane gas production
for 3 days in BER + CFT is estimated to be 6.5 × 10−4¢ US. Thus, this
small cost of electric energy consumed in BERs + CFT would produce a
2.4-times greater benefit from increased methane gas production. Al-
though the investment costs including equipment cannot be reliably
assessed because of the early stage of development, overall, these re-
sults show the potential for economic advantages with the application
of an EF method for methane fermentation. Taken together, these data
demonstrate that the use of BERs + CFT for anaerobic digestion is more
beneficial than a CSTR or CFT reactor. However, much further studies
about development for increasing the scale and the effectiveness of the
proposed reactor were necessary before practical application.
4. Conclusions
Anaerobic digestion of TPR was successfully performed without
acetate accumulation using reactors containing CFT attached to a
carbon electrode with electrochemical regulation at −0.8 V. CFT ad-
dition promoted microbial retention for TPR hydrolysis. In addition,
electrochemical cathodic regulation stimulated the growth of hydro-
genotrophic methanogens as well as interspecific interactions between
hydrogenotrophic methanogens and bacteria, contributing to acetate
degradation and methane formation. Although further studies are
needed to determine how electrochemical regulation stimulates hy-
drogenotrophic methanogens, this work highlights that electrochemical
approaches for the anaerobic digestion of agricultural plant waste may
be useful tools to increase reactor performance and stability.
Supporting materials
The characteristics of tomato residues are summarized in Table S1.
Acknowledgements
We would like to thank Editage (www.editage.jp) for English lan-
guage editing.
Funding
This work was supported in part by a Grant-in-Aid for Young
Scientists (B) (grant number 25871209) and Grant-Aid for Challenging
Exploratory Research (grant number 15 K14074).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.biortech.2017.09.206.
References
Berrios-Rivera, S.J., Bennett, G.N., San, K.Y., 2002. The effect of increasing NADH
availability on the redistribution of metabolic fluxes in Escherichia coli chemostat
cultures. Metab. Eng. 4, 230–237.
Chen, M., Tong, H., Liu, C., Chen, D., Li, F., Qiao, J., 2016. A humic substance analogue
AQDS stimulates Geobacter sp. abundance and enhances pentachlorophenol trans-
formation in a paddy soil. Chemosphere 160, 141–148.
Cheng, S., Xing, D., Call, D.F., Logan, B.E., 2009. Direct biological conversion of electrical
current into methane by electromethanogenesis. Environ. Sci. Technol. 43,
3953–3958.
Cibis, K.G., Gneipel, A., König, H., 2016. Isolation of acetic, propionic and butyric acid-
S.-i. Hirano, N. Matsumoto
forming bacteria from biogas plants. J. Biotechnol. 220, 51–63.
Dean, J.A., 1999. Lange’s Handbook of Chemistry, 15th ed. McGraw-Hill Inc., New
York, NY.
Hattori, S., 2008. Syntrophic acetate-oxidizing microbes in methanogenic environments.
Microbes Environ. 23, 118–127.
Harnisch, F., Rosa, L.F., Kracke, F., Virdis, B., Krömer, J.O., 2015. Electrifying white
biotechnology: engineering and economic potential of electricity-driven bio-pro-
duction. ChemSusChem 8, 758–766.
Hirano, S., Matsumoto, N., Morita, M., Sasaki, K., Ohmura, N., 2013. Electrochemical
control of redox potential affects methanogenesis of the hydrogenotrophic metha-
nogen Methanothermobacter thermautotrophicus. Lett. Appl. Microbiol. 56, 315–321.
Karakashev, D., Batstone, D.J., Trably, E., Angelidaki, I., 2006. Acetate oxidation is the
dominant methanogenic pathway from acetate in the absence of Methanosaetaceae.
Appl. Environ. Microbiol. 72, 5138–5141.
Koeck, D.E., Ludwig, W., Wanner, G., Zverlov, V.V., Liebl, W., Schwarz, W.H., 2015.
Herbinix hemicellulosilytica gen. nov., sp. nox., a thermophilic cellulose-degrading
bacterium isolated from a thermophilic biogas reactor. Int. J. Syst. Evol. Microbiol.
65, 23652371.
Koeck, D.E., Mechelke, M., Zverlov, V.V., Liebl, W., Schwarz, W.H., 2016. Herbivorax
saccincola gen. nov., sp. nov., a cellulolytic, anaerobic, thermophilic bacterium iso-
lated via in sacco enrichments from a lab-scale biogas reactor. Int. J. Syst. Evol.
Microbiol. 66, 4458–4463.
Lu, L., Ren, Z.J., 2016. Microbial electrolysis cells for waste biorefinery: a state of the art
review. Bioresour. Technol. 215, 254–264.
Lueders, T., Manefield, M., Friedrich, M.W., 2004. Enhanced sensitivity of DNA- and
rRNA-based stable isotope probing by fractionation and quantitative analysis of
isopycnic centrifugation gradients. Environ. Microbiol. 6, 73–78.
Marvin-Sikkema, F.D., Richardson, A.J., Stewart, C.S., Gottschal, J.C., Prins, R.A., 1990.
Influence of hydrogen-consuming bacteria on cellulose degradation by anaerobic
fungi. Appl. Environ. Microbiol. 56, 3793–3797.
Moscoviz, R., Toledo-Alarcon, J., Trably, E., Bernet, N., 2016. Electro-fermentation: how
to drive fermentation using electrochemical systems. Trends Biotechnol. 34,
856–865.
Roy, S., Schievano, A., Pant, D., 2016. Electro-stimulated microbial factory for value
added product synthesis. Bioresour. Technol. 213, 129–139.
Sasaki, K., Morita, M., Hirano, S., Ohmura, N., Igarashi, Y., 2009. Effect of adding carbon
fiber textiles to methanogenic bioreactors used to treat an artificial garbage slurry. J.
Biosci. Bioeng. 108, 130–135.
Sasaki, K., Morita, M., Hirano, S., Sasaki, D., Ohmura, N., Igarashi, Y., 2010. Efficient
degradation of rice straw in the reactors packed by carbon fiber textiles. Appl.
Microbiol. Biotechnol. 87, 1579–1586.
Sasaki, K., Hirano, S., Morita, M., Sasaki, D., Matsumoto, N., Ohmura, N., Igarashi, Y.,
817
Bioresource Technology 249 (2018) 809–817
2011a. Bioelectrochemical system accelerates microbial growth and degradation of
filter paper. Appl. Microbiol. Biotechnol. 89, 449–455.
Sasaki, K., Morita, M., Sasaki, K., Hirano, S., Matsumoto, N., Watanabe, A., Ohmura, N.,
Igarashi, Y., 2011b. A bioelectrochemical reactor containing carbon fiber textiles
enables efficient methane fermentation from garbage slurry. Bioresour. Technol. 102,
6837–6842.
Sasaki, D., Sasaki, D., Morita, M., Hirano, S., Matsumoto, N., Ohmura, N., 2012.
Bioelectrochemical regulation accelerates facultatively syntrophic proteolysis. J.
Biosci. Bioeng. 114, 59–63.
Sasaki, D., Morita, M., Sasaki, K., Watanabe, A., Ohmura, N., 2013a. Increased growth of
a hydrogenotrophic methanogen in co-culture with a cellulolytic bacterium under
cathodic electrochemical regulation. Biosci. Biotechnol. Biochem. 77, 1096–1099.
Sasaki, D., Sasaki, K., Watanabe, A., Morita, M., Matsumoto, N., Igarashi, Y., Ohmura, N.,
2013b. Operation of a cylindrical bioelectrochemical reactor containing carbon fiber
fabric for efficient methane fermentation from thickened sewage sludge. Bioresour.
Technol. 129, 366–373.
Sawayama, S., Tsukahara, K., Yagashita, T., 2006. Phylogenetic description of im-
mobilized methanogenic community using real-time PCR in a fixed-bed anaerobic
digester. Bioresour. Technol. 97, 69–76.
Shafiei, M., Karimi, K., Taherzadeh, M.J., 2011. Techno-economical study of ethanol and
biogas from spruce wood by NMMO-pretreatment and rapid fermentation and di-
gestion. Bioresour. Technol. 102, 7879–7886.
Takai, K., Horikoshi, K., 2000. Rapid detection and quantification of members of the
archaeal community by quantitative PCR using fluorogenic probes. Appl. Environ.
Microbiol. 66, 5066–5072.
Town, J.R., Links, M.G., Fonstad, T.A., Dumonceaux, T.J., 2014. Molecular character-
ization of anaerobic digester microbial communities identifies microorganisms that
correlate to reactor performance. Bioresour. Technol. 151, 249–257.
van Dijk, C., Veeger, C., 1981. The effects of pH and redox potential on the hydrogen
production activity of the hydrogenasefrom Megasphaera elsdenii. Eur. J. Biochem.
114, 209–219.
Wagner, A.O., Lins, P., Malin, C., Reitschuler, C., Illmer, P., 2013. Impact of protein-,
lipid- and cellulose-containing complex substrates on biogas production and micro-
bial communities in batch experiments. Sci. Total Environ. 458, 256–266.
Wang, H., Fotidis, I.A., Angelidaki, I., 2015a. Ammonia effect on hydrogenotrophic me-
thanogens and syntrophic acetate-oxidizing bacteria. FEMS Microbiol. Ecol. 91, 130.
Wang, H., Luo, H., Fallgren, P.H., Jin, S., Ren, Z.J., 2015b. Bioelectrochemical system
platform for sustainable environmental remediation and energy generation.
Biotechnol. Adv. 33, 317–334.
Westerholm, M., Levén, L., Schnürer, A., 2012. Bioaugmentation of syntrophic acetate-
oxidizing culture in biogas reactors exposed to increasing levels of ammonia. Appl.
Environ. Microbiol. 78, 7619–7625.