ARTICLE
Integrated Biogas Upgrading and Hydrogen
Utilization in an Anaerobic Reactor Containing
Enriched Hydrogenotrophic Methanogenic Culture
Gang Luo, Irini Angelidaki
Department of Environmental Engineering, Technical University of Denmark,
DK-2800, Kgs Lyngby, Denmark; telephone: 45-4525-1429; fax: 45-4593-2850;
e-mail: iria@env.dtu.dk
ABSTRACT: Biogas produced by anaerobic digestion, is
mainly used in a gas motor for heat and electricity produc-
tion. However, after removal of CO2, biogas can be upgraded
to natural gas quality, giving more utilization possibilities,
such as utilization as autogas, or distant utilization by using
the existing natural gas grid. The current study presents a
new biological method for biogas upgrading in a separate
biogas reactor, containing enriched hydrogenotrophic
methanogens and fed with biogas and hydrogen. Both
mesophilic- and thermophilic anaerobic cultures were
enriched to convert CO2 to CH4 by addition of H2. Enrich-
ment at thermophilic temperature (558C) resulted in CO2
and H2 bioconversion rate of 320 mL CH4/(gVSS h), which
was more than 60% higher than that under mesophilic
temperature (378C). Different dominant species were found
at mesophilic- and thermophilic-enriched cultures, as
revealed by PCRDGGE. Nonetheless, they all belonged
to the order Methanobacteriales, which can mediate hydro-
genotrophic methanogenesis. Biogas upgrading was then
tested in a thermophilic anaerobic reactor under various
operation conditions. By continuous addition of hydrogen
in the biogas reactor, high degree of biogas upgrading was
achieved. The produced biogas had a CH4 content, around
95% at steady-state, at gas (mixture of biogas and hydrogen)
injection rate of 6 L/(L day). The increase of gas injection
rate to 12 L/(L day) resulted in the decrease of CH4 content
to around 90%. Further study showed that by decreasing the
gasliquid mass transfer by increasing the stirring speed of
the mixture the CH4 content was increased to around 95%.
Finally, the CH4 content around 90% was achieved in this
study with the gas injection rate as high as 24 L/(L day).
Biotechnol. Bioeng. 2012;xxx: xxxxxx.
2012 Wiley Periodicals, Inc.
Correspondence to: I. Angelidaki
Contract grant sponsor: Bioref-resund
Contract grant sponsor: Hans Christian rsted Postdoc Program
Additional supporting information may be found in the online version of this article.
Received 20 March 2012; Revision received 6 May 2012; Accepted 8 May 2012
Accepted manuscript online 21 May 2012;
Article rst published online in Wiley Online Library
(wileyonlinelibrary.wiley.com).
DOI 10.1002/bit.24557
2012 Wiley Periodicals, Inc.
KEYWORDS: anaerobic digestion; biogas upgrading; H2;
CO2; CH4
Introduction
Production of biogas by anaerobic digestion of organic
wastes is an emerging alternative energy technology (Luo
et al., 2011a; Ueno et al., 2007). Biogas is envisioned as a key
element in emerging renewable energy strategies in Europe,
motivated by the European Union target of achieving 20%
renewable energy by 2020. The Danish government also
proposed a target of using 50% of the manure produced in
Denmark for renewable energy production by 2020, and it
would essentially be met though a strong biogas expansion
(Hamelin et al., 2011). Biogas mainly contains CH4 (40
75%) and CO2 (2560%). Upgrading of biogas to CH4
content to higher than 90% can increase the heating value,
and extend the biogas utilization as a renewable energy
source (Deng and Hagg, 2010). The main advantage of
upgrading is, however, that it would make it possible to use
biogas as alternative to natural gas. Upgrading to natural gas
quality is very much in focus currently as it gives possibilities
for alternative applications such as fuel for road vehicles
(Ryckebosch et al., 2011). An additional advantage of
upgrading is the opportunity to utilize the existing natural
gas grid for transporting the upgraded biogas from rural
areas, where typically biogas plants are located to urban
areas where consumer density is higher.
The common methods of biogas upgrading include water
washing, pressure swing adsorption, polyglycol adsorption,
and chemical treatment (Osorio and Torres, 2009). The
costs of the above methods are relatively high since they need
either high pressure or addition of chemicals. Besides, when
removing CO2 from biogas, small amounts of CH4 are also
removed, which will increase greenhouse gas emissions
(Weiland, 2010). To circumvent these disadvantages and use
milder treatments with minimal chemical and energy,
biological conversion of CO2 to CH4 for biogas upgrading
( Equation 1) can be achieved by utilization of
Biotechnology and Bioengineering, Vol. xxx, No. xxx, 2012 1
hydrogenotrophic archeae, microorganisms which can
bind CO2 with H2 and convert them to methane.
4H2 CO2 CH4 2H2 O;
(1)
DG0
130:7 kJ=mol
Hydrogenotrophic methanogens belong to the orders
Methanobacteriales, Methanococcales, Methanomicrobials,
and Methanosarcinaceae (Karakashev et al., 2005). Anaerobic
sludges from anaerobic reactors contain hydrogenotrophic
methanogens (Demirel and Scherer, 2008).
Nevertheless in order for the whole process to be
considered as renewable, the H2 needed for the biogas
upgrading should also be provided by renewable sources.
One obvious way is to use excess electricity from wind mills,
for water electrolysis to produce hydrogen as we have
mentioned previously (Luo et al., 2012). In addition, H2 can
be also obtained by other sources, including coal gasica-
tion, petroleum renery, petrochemical plants, and soda
manufacture (Ni et al., 2011). Utilization of H2 for biogas
upgrading provides several advantages, such as utilization
of the existing infrastructure of biogas plants and energy
conversion for H2 utilization.
We have proposed an innovative method for in situ
biogas upgrading by the addition of hydrogen to anaerobic
reactor treating cattle manure (Luo et al., 2012). This in situ
upgrading method is very simple, but it has negative effect
on the anaerobic process due to pH increase. Therefore, it
requires special solutions, such as codigestion with acidic
substrates or pH control, to keep the pH within appropriate
limits. Therefore, in the present study we investigated an
innovative process for biogas upgrading, in a separate
reactor enriched with hydrogenotrophic methanogens, and
fed with biogas and hydrogen. The biomethanation
efciencies and microbial community compositions of
both mesophilic- and thermophilic-enriched cultures were
investigated. Furthermore, the biogas upgrading and reactor
performance was studied in a continuously fed thermophilic
anaerobic reactor under varying operation conditions.
Materials and Methods
Inocula
As sources of inoculum, mesophilic anaerobically digested
sewage sludge (Wastewater treatment plant, Lundtofte,
Denmark) and thermophilic anaerobically digested manure
(Biogas Plant, Snertinge, Denmark) were used.
Reactor Setup and Operation
Enrichment Cultures
Mixed cultures from mesophilic and thermophilic full-scale
biogas reactors were used to establish the enrichment
2 Biotechnology and Bioengineering, Vol. xxx, No. xxx, 2012
cultures. Initially, mixtures (380 mL) of mesophilic or
thermophilic sludge and basal anaerobic (BA) medium
(Angelidaki and Sanders, 2004) with VSS concentration 5 g/L
were added in 1,140 mL serum bottles. The BA medium
contained 10 mM PBS buffer. The bottles were then closed
with butyl-rubber stoppers and aluminum crimps and
ushed with pure H2. After that, pure CO2 (190 mL) was
injected into the bottles to achieve a ratio of H2/CO2 4:1. The
bottles were incubated in shakers at 378C (bottles inoculated
with mesophilic sludge) and 558C (bottles inoculated with
thermophilic inoculum) with shaking speed 300 rpm.
Twenty-ve milliliter of the bottle content was daily
replaced with fresh BA medium using 100 mL syringe and
at the same time the gases were also refreshed by ushing
with pure H2 and then adding CO2 as described above.
This procedure continued for approximately 2 months.
The experiment was performed with duplicate bottles. To
determine the kinetics of H2 consumption, enriched
mesophilic or thermophilic cultures (10 mL cultures)
were transferred to serum vials (60 mL serum vials) and
pre-incubated for 1 h under the atmosphere of pure N2, and
then given the appropriate H2 partial pressure (00.2 atm)
by injecting pure H2 and corresponding CO2 using syringe.
All the bottles were pressurized to 1.5 atm by adding
additional N2. The experiment was conducted for 1h to
ensure that the initial hydrogen concentration was not
decreased signicantly (Ahring and Westermann, 1987).
All the bottles were prepared in fume cupboards to avoid
the possible explosion of H2.
Continuously Fed Reactor Experiment
After enrichment for half a month in the previous
experiment, thermophilic condition was shown to be
more efcient than mesophilic condition. So the continu-
ously fed reactor was immediately started up and
operated at thermophilic temperature. The reactor was
1 L bottle with 600 mL working volume. The reactor was
lled with thermophilic inoculum from full-scale biogas
reactor and BA medium containing 10 mM PBS buffer.
The initial VSS concentration was 5 g/L. The feeding gas
was composed of H2, CH4, and CO2 with the ratio 60:25:15.
The gas was injected to the bottom of the reactor through
ceramic gas diffusers. The initial gas ow rate was 3 L/
(L day) and then increased gradually. Forty milliliter
mixture in the reactor was daily replaced with fresh BA
medium. The reactor was magnetically stirred at 500 or
800 rpm. Steady-state was dened as a period of 6
consecutive days with daily variation of biogas production
rate of <10%.
Specic Methanogenic Activity (SMA) Tests
Acetoclastic and hydrogenotrophic methanogenic activities
of the anaerobic cultures from both mesophilic and
thermophilic reactors at the beginning and end of the
enrichment were tested. 20 mL fresh samples were added in
68 mL bottles. Acetate (20 mM) or H2/CO2 (80/20, 1 atm)
were supplemented to each bottle. Bottles with fresh samples
only, but without substrates, were used as controls. The
bottles were incubated in shakers at 37 or 558C with shaking
speed 300 rpm. All the tests were prepared in duplicates.
Microbial Community Composition
Archaeal communities of the enriched cultures (both
mesophilic and themophilic) and the cultures at steady-
states from the continuous reactors were analyzed by
polymerase chain reactiondenaturing gradient gel electro-
phoresis (PCRDGGE). The procedure has been described
in previous publication (Luo et al., 2011b). Dominant bands
from DGGE were sequenced and identied by comparing
the gene sequences with DNA sequences in the National
Centre for Biotechnology Information (NCBI) database
using the BLAST algorithm.
Fluorescent in situ hybridization (FISH) was used to
observe the microbial communities in the reactors and also
validate the results obtained by DGGE analysis. The probes
used and procedure for preparation of the samples were
described previously (Karakashev et al., 2005).
Analytical Methods
The concentrations of acetate, butyrate, and propionate
were determined by gas chromatograph (GC; Hewlett
Packard, HP5890 series II) equipped with a ame ionization
detector and HP FFAP column (30 m  0.53 mm  1.0 mm).
H2 was analyzed by GC-TCD tted with a 4.5 m  3 mm s-m
Figure 1. Proles of H2, CO2, and CH4 at the beginning of the enrichment (a) thermophlic (b) mesophilic and the end of the enrichment (c) thermophlic (d) mesophilic.
Luo and Angelidaki: Integrated Biogas Upgrading and Hydrogen Utilization
stainless column packed with Molsieve SA (10/80). CH4
was analyzed with GC-TCD tted with parallel column of
1.1 m  3/16 Molsieve 137 and 0.7 m  1/4 chromosorb
108. Detailed information about the operation conditions of
above GC or HPLC was previously described (Luo et al.,
2011c). All the values of gas production and consumption
reported in the article were corrected for the standard
temperature and pressure conditions (STP).
Dissolved H2 in the liquid phase was rst extracted by
the method described by Yu et al. (2006) and then was
determined by a reduction gas detector (Trace Analytical
RGD2) with a detection limit of 0.1 ppmv (Heimann
et al., 2007). ATP was determined by using Lumin(ATE)/
Lumin(EX) reagents (Celsis No. 92687) and a luminometer
(Advance coupe, Celsis, Landgraaf, The Netherlands).
ATP content was calculated from an ATP standard curve
prepared from ATP standard salt (Celsis No. 92521)
dissolved in Lumin (PM) buffer (Celsis No. 92588) and
diluted in sterile tap water. Measurements were performed
by the use of internal standard addition (Silhan et al., 2006).
Results and Discussion
Biogas Upgrading Potentials and Characterization of
Mesophilic- and Thermophilic-Enriched Mixed Cultures
Mesophilic- and thermophilic-mixed cultures were enriched
to characterize their biomethanation potentials from H2 and
CO2. Steady-state was achieved after 1-month cultivation,
and the experiment was continued for another month, to
ensure stability of the microbial community. Figure 1 shows
3
Biotechnology and Bioengineering
the H2 and CO2 consumption and CH4 production proles
at the beginning and end of the enrichments. It is obvious
the long-term enrichments signicantly increased the H2
and CO2 consumption rates under both mesophilic and
thermophilic conditions. At the beginning of the enrich-
ments, there was still CO2 left even though H2 was fully
consumed, despite that the H2 and CO2 were added at a
ratio of 4:1 which would stoichiometrically correspond to
methane production. On the one hand, it may be due to
the bicarbonate in the inoculum, contributing to surplus
of CO2 compared to hydrogen. On the other hand, it may
be attributed to the excess CO2 and CH4 production by
the residual organics in the inoculum. The CO2 and H2
were almost fully consumed after long-term cultivation.
At the end of the enrichment, it took around 4 h to achieve
complete conversion of H2 and CO2 to CH4 under
thermophilic condition, while it took around 8 h under
mesophilic condition. It indicated that biogas upgrading at
thermophilic conditions was faster compared to mesophilic.
The linear decrease of H2 under all conditions showed there
was no gasliquid mass-transfer limitation in the experi-
ment (Luo et al., 2012). The higher solubility of CO2 resulted
in the sharp decrease of CO2 initially. The sharp initial
decrease of CO2 was expected, as it was calculated that
around 50 mL CO2 was immediately dissolved in the liquid
phase assuming for simplication, insignicant amount
Figure 2. Specic methanogenic activity (SMA) results (a) mesophilic (b)
bicarbonate in the liquid and gasliquid mass-transfer thermophilic.
limitations.
Acetoclastic and hydrogenotrophic methanogenic activi-
ties at the beginning and end of the enrichments are shown
in Figure 2. Initially, both mesophilic and themophilic The sequencing results of the dominant bands are shown in
cultures had obvious acetoclastic and hydrogenotrophic Table I. The mesophilic and thermophilic inocula contained
methanogenic activities [around 10 mL CH4/(gVSS h)]. archaea belonging to both hydrogenotrophic (bands 5 and
However, the acetoclastic methanogenic activities were 9) and acetoclastic methanogens (bands 6, 7, and 11), which
detected after long-term cultivation, while the hydrogeno- was as expected (Fig. 2). However, the dominant bands
trophic methanogenic activities increased to 198 mL CH4/ changed after long-term enrichment. Under mesophilic
(gVSS h) under mesophilic condition and 320 mL CH4/ condition, bands 3, 4, and 8 were the dominant ones and
(gVSS h) under thermophilic condition. The results indi- they were all related to archaea belonging to the order
cated that, as expected, hydrogenotrophic methanogens Methanobacteriales. Under thermophilic condition, bands 1,
were selectively enriched, and thermophilic conditions provide 2, 10, and 12 were the dominant ones and they were also
high efciency for biogas upgrading. Acetate was found related to Methanobacteriales. Methanobacteriales were also
in both mesophilic (0.5 mM) and thermophilic (6.6 mM)- found in the inocula (bands 5 and 9). The organisms
enriched cultures. If the acetate was also converted to CH4, it belonging to Methanobacteriales mediate hydrogenotrophic
would only account for 0.14% and 1.9% of the produced methanogenesis under both mesophilic and thermophilic
CH4. The acetate was probably produced from homoaceto- conditions (Karakashev et al., 2005). The above results
gens (Luo et al., 2011). demonstrated that hydrogenotrophic methanogens were
After around 2 months of cultivation, DAPI-stained en- selectively enriched, which is consistent with SMA results
riched cultures were observed by microscopy (Supplementary that only hydrogenotrophic methanogenic activities
Fig. S1). Rods and laments were observed in both were detected. Methanococcales, Methanomicrobiales, and
mesophilic- and thermophilic-enriched cultures. However, Methanosarcinales, which can also mediate hydrogeno-
the laments in mesophilic-enriched culture were generally trophic methanogenesis (Garcia et al., 2000; Karakashev
longer than those in thermophilic-enriched culture. In order et al., 2005), were not found in the enriched cultures. The
to better dene the composition of the enriched mixed dominance of Methanobacteriales in the enriched mesophilic
cultures, archaeal communities were analyzed by PCR and thermophilic cultures was further veried by FISH
DGGE. Proles of archaeal communities showed there were observation (Supplementary Fig. S2).
several well-separated bands representing different phylo- The substrate kinetic parameters Km (half-saturation
genetic groups of dominant archaea in the samples (Fig. 3). constant) and Vmax (maximum hydrogen consumption

4 Biotechnology and Bioengineering, Vol. xxx, No. xxx, 2012

 Figure 4. Effect of increasing H2 concentrations on the rate of hydrogen
consumption. T, thermophilic-enriched culture; M, mesophilic-enriched culture.

the H2 consumption rates were almost constant under

Figure 3. DGGE bands of archaeal communities. IM, mesophilic inoculum; IT,
thermophilic inoculum; EM, enriched mesophilic culture; ET, enriched thermophilic
culture.
rate) of mesophilic- and thermophilic-enriched cultures
were determined by measuring the consumption rates of
H2 under different concentrations of H2 in the headspace.
Figure 4 showed that substrate was not the limiting factor
under H2 partial pressures higher than 0.05 atm, because
both mesophilic and thermophilic conditions. It could also
explain the linear decrease of H2 in Figure 1. Ks and Vmax of
the mesophilic-enriched culture were 0.011 atm and
409 mL/(L h), and the values of thermophilic-enriched
culture were 0.0075 atm and 699 mL/(L h). Since there was
intense mixing in the experiment, it can be assumed that
hydrogen inside the reactor was in equilibrium with the
liquid phase. Therefore, the Ks values of mesophilic- and
thermophilic-enriched cultures were calculated based on
Henrys constant [378C, 7.31  10
4 mol/(L atm); 558C,
6.7  10
4 mol/(L atm)] as 8 and 5 mM, respectively.
They were close to the values found for pure cultures
of mesophilic methanogens (16 mM; Kristjansson et al.,
1982). The biomass concentrations of mesophilic- and
thermophilic-enriched cultures were 0.48 and 0.56 g-dry
weight/L. Therefore, the Vmax could also be expressed
as 852 mL/(gVSS h) (mesophilic) and 1,248 mL/(gVSS h)
(thermophilic). The value [1,248 mL/(gVSS h)] was rela-
tively lower than that from pure cultures under thermophilic
condition [2,900 mL/(gVSS h)] (Ahring and Westermann,
1987). It may be due to the unoptimized medium and
pH.

Table I. DGGE band sequencing results.

DGGE band Closest match Identity (%) Order Accession no.

1 Methanobacterium kanagiense 96 Methanobacteriales JN983050
2 Methanothermobacter thermautotrophicus 93 Methanobacteriales JN983051
3 Uncultured Methanobacterium sp. 95 Methanobacteriales JN983052
4 Uncultured Methanobacteriales archaeon clone 98 Methanobacteriales JN983053
5 Uncultured Methanobacteriales archaeon clone 98 Methanobacteriales JN983054
6 Methanosarcina acetivorans 100 Methanosarcinales JN983055
7 Methanosaeta concilii 96 Methanosarcinales JN983056
8 Methanobacterium petrolearium 98 Methanobacteriales JN983057
9 Methanobrevibacter arboriphilus 98 Methanobacteriales JN983058
10 Methanobrevibacter arboriphilus 98 Methanobacteriales JN983059
11 Uncultured Methanosaetaceae archaeon clone 98 Methanosarcinales JN983060
12 Uncultured Methanothermobacter sp. clone 100 Methanobacteriales JN983061

Luo and Angelidaki: Integrated Biogas Upgrading and Hydrogen Utilization 5

Biotechnology and Bioengineering
Table II. Performances of the thermophilic reactor under different operation conditions.

Period (day) 010a 1143 (I) 4473 (II) 7496 (III) 97135 (IV)
Gas injection rate (L/(Lreactor day)) 3 6 12 12 24
Gas retention time (h) 8 4 2 2 1
Mixing speed (rpm) 500 500 500 800 800
Biogas production rate (L/(Lreactor day)) 1.4  0.3 2.5  0.4 5.1  0.6 4.9  0.5 10.1  1.3
Biogas composition
CH4 (%) 93.5  4.4 95.4  2.8 89.9  4.1 94.2  2.8 90.8  2.4
CO2 (%) 4.2  2.5 0.7  0.4 2.6  1.5 1.9  0.5 2.2  1.3
H2 (%) 2.3  2.4 3.9  0.8 7.5  1.2 3.9  0.7 7  1.8
H2 consumption rate (L/(Lliquid day)) 2.9  0.5 5.9  0.4 11.3  0.7 11.6  0.8 22.8  2.1
CH4 production rate (L/(Lliquid day)) 0.9  0.2 1.5  0.3 2.6  0.5 2.7  0.6 5.3  1.4
Yield CH4/H2 0.31 0.26 0.23 0.23 0.23
Acetate concentration (mM) 5.9  1.2 8.8  0.7 9.8  1.3 10.5  1.1 9.4  1.6
a
The data are not in steady-state for period 110.

Biogas Upgrading Performance of Thermophilic
Anaerobic Reactor
The biogas upgrading potentials were then tested in
continuously stirred tank reactor (CSTR) at 558C since
the conversion of H2 and CO2 into CH4 was signicantly
faster under thermophilic condition compared to meso-
philic operation. The operation parameters and results are
shown in Table II. The inoculum had obvious hydrogeno-
trophic methanogenic activity, and therefore H2 in the
feeding gas was efciently consumed even during the start-
up period of the reactor. At day 10, the CH4 content could
be increased from 25% of the feeding gas to around 93% of
the upgraded gas. After that, the gas injection rate increased
to 6 L/(L day) and the reactor was operated under this
condition until steady-state was achieved. The CH4 content
was as high as 95.4%, and there were only small amount
of H2 and CO2 left. The quality of the upgraded biogas
was adequate for utilization as natural gas. At day 44, the
gas injection rate was further increased to 12 L/(L day).
However, the CH4 content in the upgraded gas decreased
to around 90% even after around 1-month operation. The
gasliquid mass transfer has been found as the limiting
factor for bioconversion of gaseous substrate, especially
when the gas (such as H2) has low solubility (Pauss et al.,
1990). Therefore, the mixing speed was increased from 500
to 800 rpm at day 74 to lower the gasliquid mass transfer
limitations. The increase of mixing speed resulted in the
increase of CH4 content to around 95%. It seems that
intensive mixing of the liquid reduced the H2 content in the
biogas, and thereby resulted in higher CH4 content. After
day 97, the gas injection rate increased to 24 L/(L day).
Though a small decrease of CH4 content was observed again,
the methane content still remained around 90%. The above
results demonstrated that biogas upgrading was achieved in
thermophilic anaerobic reactor with high gas injection rate.
Theoretically, the molar ratio of CH4 production rate
to H2 consumption rate is 0.25. Nevertheless, the measured
ratio was higher than 0.31 during the initial 10 days. It may
be explained by the excess CH4 production from residual
organic matter contained in the inoculum. After long-term
operation, the methane contribution originating from
residual organic matter in the inoculum was eliminated.
The CO2:H2 ratios were 0.26, 0.23, 0.23, 0.23 during the
steady-states of phase I, II, III, and IV, respectively. It
showed that the consumed H2 was almost stoichiometrically
converted to CH4.
Acetate was also detected as the main metabolite in
the liquid phase with the concentration between 5.9 and
10.5 mM.
During the experiment, pH was relatively stable (around
7.8), which is within the pH range appropriate for
methanogens (OFlaherty et al., 1998).
The microbial communities at days 50 and 100 were
analyzed by PCRDGGE, and similar distribution of
dominant bands as that found from thermophilic-enriched
culture in the rst experiment were observed (data not
shown).
In the biogas upgrading process, gasliquid mass transfer
is crucial since it determines the available substrate for
methanogens. Therefore, the dissolved H2 during the steady-
state of each operation condition was measured. Figure 5
shows both measured dissolved H2 (H2l, mol/L) and the
calculated dissolved H2 equivalent to gas phase (H2gTh, mol/
L) according to Henrys law. The H2gTh were always three to
eight times higher than H2l in the whole operation periods,
which indicated that there are gasliquid mass transfer
limitations in the process. It is consistent with the fact
that H2 is a kind of gas with low solubility. The H2gTh
changed with the variations of the operation parameters
while H2l were relatively stable. To describe the gasliquid
mass transfer, Equation (2) is used.
rt 22:4kL aH2gTh
H2l (2)
where rt (L/(L day)) is the H2 gasliquid mass transfer rate,
22.4 (L/mol) represents that 1 mol gas is 22.4 L (STP), kLa
(1/day) is the transfer coefcient. It is obvious rt is
determined by the difference between gaseous and liquid
H2 [assuming kLa is constant for the sample operating
conditions (i.e., mixing)]. Higher gasliquid mass transfer
rate could be obtained with larger difference between H2gTh

6 Biotechnology and Bioengineering, Vol. xxx, No. xxx, 2012


Figure 5. Dissolved H2 concentration during the steady-state of each operation
condition.
and H2l. In our experiment, rt is the same as H2
consumption rate during the steady-states of the reactor.
The H2 consumption rate in phase II was almost two-times
higher than that in phase I (Table II), therefore, the
difference between H2l and H2gTh should also be increased
to obtain a higher H2 gasliquid mass transfer rate, which
explained the higher H2 content in the upgraded gas of
phase II. However, the H2 content in the upgraded gas
decreased from 7.5% in phase II to around 3.9% in phase III.
This may be due to the increase of mixing speed resulting
in an increase of the kLa (Kramer and Bailey, 1991), and
thereby decreased the difference between H2l and H2gTh. It is
expected that by increasing kLa in phase IV, the CH4 content
can be further increased. Besides increasing the mixing
speed, other methods to increase kLa include efcient gas
diffusion equipment (such as hollow ber membrane; Kim
et al., 2011), biogas recirculation (Guiot et al., 2011), etc.
Table II shows that H2 consumption rates were increased
with the increase of gas injection rates. The H2 consumption
rates under steady-states can be described by the following
equation:
rc mYX
kd 1
m (4)
SRT
where rc (L/(L day)) is the H2 consumption rate, Y (L/g) is
the yield coefcient (STP), X (g/L) is the cell concentration,
m (1/day) is the specic growth rate of the microorganism,
kd (1/day) is the specic decay rate of the microorganism
(constant), SRT (day) is the sludge retention time. As shown
in Equation (4), m is constant since kd and SRT are constant
in our study. Besides, Y is also constant. Therefore, it
is expected that X was increased with the increase of gas
injection rate. VSS is usually assumed to represent biomass
cell concentration (Hwang and Hansen, 1998). Nevertheless,
in reality this is not always the case, because as was
previously mentioned VSS does not only represent
Figure 6. ATP concentration and H2 consumption rate during the steady-state of
each operation condition.
microbial biomass, but also other residual organic matter.
ATP has been reported as an effective method to assess the
cell concentration in the literature (Hwang and Hansen,
1998). Therefore, ATP concentrations under different
operation parameters were measured (Fig. 6). It is obvious
the increase of ATP was in correspondence with the increase
of H2 consumption rate.
Ex situ biogas upgrading by anaerobic digestion has not
been studied before, and the results from this study showed
that CH4 content in the biogas around 90% or even higher
was achieved under thermophilic condition with gas
injection rate up to 24 L/(L day). In Denmark, there are
more than 20 centralized biogas plants and cattle manure is
the main substrate. The daily average production of biogas is
140 m3 per 100 m3 reactor tank (Raven and Gregersen,
2007). If the biogas upgrading concept proposed in the
present study is applied in these biogas plants, an anaerobic
reactor for biogas upgrading with volume 1/10 of the biogas
reactor is needed. If more efcient gasliquid mass transfer
method is adopted, it is expected that even smaller reactor
for biogas upgrading can be used.
(3)
Conclusions
The present study proposed and demonstrated a method for
biogas upgrading in an anaerobic reactor. Both mesophilic
and thermophilic sludges had obvious biomethanation
potential from H2 and CO2. Enrichment under thermophilic
condition provided higher CH4 production rate from H2
and CO2. Microbial community analysis showed that
different archaeal species were involved in mesophilic-
and thermophilic-enriched cultures, but the dominant
archaeal species were all belonged to the order
Methanobacteriales. Biogas upgrading in a thermophilic
anaerobic reactor showed that CH4 content higher than 90%
was obtained with the gas injection rate as high as 24 L/
(L day). The study also revealed that gasliquid mass transfer
is the rate limiting factor for efcient H2 utilization.
Luo and Angelidaki: Integrated Biogas Upgrading and Hydrogen Utilization 7
Biotechnology and Bioengineering
 We would like to thank Hector Garcia for his technical assistance with
the experiments.
References
Ahring BK, Westermann P. 1987. Kinetics of butyrate, acetate, and hydro-
gen metabolism in a thermophilic, anaerobic, butyrate-degrading
triculture. Appl Environ Microbiol 53:434439.
Angelidaki I, Sanders W. 2004. Assessment of the anaerobic biodegradabil-
ity of macropollutants. Rev Environ Sci Biotechnol 3:117129.
Demirel B, Scherer P. 2008. The roles of acetotrophic and hydrogenotrophic
methanogens during anaerobic conversion of biomass to methane: A
review. Rev Environ Sci Biotechnol 7:173190.
Deng L, Hagg M-B. 2010. Techno-economic evaluation of biogas upgrading
process using CO2 facilitated transport membrane. Int J Greenhouse
Gas Control 4:638646.
Garcia JL, Patel BKC, Ollivier B. 2000. Taxonomic phylogenetic and
ecological diversity of methanogenic archaea. Anaerobe 6:205226.
Guiot SR, Cimpoia R, Carayon G. 2011. Potential of wastewater-treating
anaerobic granules for biomethanation of synthesis gas. Environ Sci
Technol 45:20062012.
Hamelin L, Wesnaes M, Wenzel H, Petersen BM. 2011. Environmental
consequences of future biogas technologies based on separated slurry.
Environ Sci Technol 45:58695877.
Heimann AC, Friis AK, Scheutz C, Jakobsen R. 2007. Dynamics of reductive
TCE dechlorination in two distinct H(2) supply scenarios and at
various temperatures. Biodegradation 18:167179.
Hwang S, Hansen CL. 1998. Evaluating a correlation between volatile
suspended solid and adenosine 50 -triphosphate levels in anaerobic
treatment of high organic suspended solids wastewater. Bioresour
Technol 63:243250.
Karakashev D, Batstone DJ, Angelidaki I. 2005. Inuence of environmental
conditions on methanogenic compositions in anaerobic biogas reac-
tors. Appl Environ Microbiol 71:331338.
Kim HW, Marcus AK, Shin JH, Rittmann BE. 2011. Advanced control
for photoautotrophic growth and CO(2)-utilization efciency using
a membrane carbonation photobioreactor (MCPBR). Environ Sci
Technol 45:50325038.
Kramer HW, Bailey JE. 1991. Mass-transfer characterization of an airlift
probe for oxygenating and mixing cell-suspensions in an NMR spec-
trometer. Biotechnol Bioeng 37:205209.
Kristjansson JK, Schonheit P, Thauer RK. 1982. Different Ks-values for
hydrogen of methanogenic bacteria and sulfate reducing bacteriaAn
explanation for the apparent inhibition of methanogenesis by sulfate.
Arch Microbiol 131:278282.
8 Biotechnology and Bioengineering, Vol. xxx, No. xxx, 2012
Luo G, Talebnia F, Karakashev D, Xie L, Zhou Q, Angelidaki I. 2011a.
Enhanced bioenergy recovery from rapeseed plant in a biorenery
concept. Bioresour Technol 102:14331439.
Luo G, Xie L, Zhou Q, Angelidaki I. 2011b. Enhancement of bioenergy
production from organic wastes by two-stage anaerobic hydrogen
and methane production process. Bioresour Technol 102:8700
8706.
Luo G, Karakashev D, Xie L, Zhou Q, Angelidaki I. 2011c. Long-term effect
of inoculum pretreatment on fermentative hydrogen production by
repeated batch cultivations: Homoacetogenesis and methanogenesis
as competitors to hydrogen production. Biotechnol Bioeng 108:1816
1827.
Luo G, Johansson S, Boe K, Xie L, Zhou Q, Angelidaki I. 2012. Simultaneous
hydrogen utilization and in situ biogas upgrading in an anaerobic
reactor. Biotechnol Bioeng 109:10881094.
Ni BJ, Liu H, Nie YQ, Zeng RJ, Du GC, Chen JA, Yu HQ. 2011. Coupling
glucose fermentation and homoacetogenesis for elevated acetate pro-
duction: Experimental and mathematical approaches. Biotechnol
Bioeng 108:345353.
OFlaherty V, Mahony T, OKennedy R, Colleran E. 1998. Effect of pH on
growth kinetics and sulphide toxicity thresholds of a range of metha-
nogenic, syntrophic and sulphate-reducing bacteria. Process Biochem
33:555569.
Osorio F, Torres JC. 2009. Biogas purication from anaerobic digestion in a
wastewater treatment plant for biofuel production. Renewable Energy
34:21642171.
Pauss A, Andre G, Perrier M, Guiot SR. 1990. Liquid-to-gas mass-transfer in
anaerobic processsesInevitable transfer limitations of methane and
hydrogen in the biomethanation process. Appl Environ Microbiol
56:16361644.
Raven R, Gregersen KH. 2007. Biogas plants in Denmark: Successes and
setbacks. Renewable Sustain Energy Rev 11:116132.
Ryckebosch E, Drouillon M, Vervaeren H. 2011. Techniques for transfor-
mation of biogas to biomethane. Biomass Bioenergy 35:16331645.
Silhan J, Cortzen CB, Albrechtsen HJ. 2006. Effect of temperature and pipe
material on biolm formation and survival of Escherichia coli in used
drinking water pipes: A laboratory-based study. Water Sci Technol
54:4956.
Ueno Y, Fukui H, Goto M. 2007. Operation of a two-stage fermentation
process producing hydrogen and methane from organic waste. Environ
Sci Technol 41:14131419.
Weiland P. 2010. Biogas production: Current state and perspectives. Appl
Microbiol Biotechnol 85:849860.
Yu YH, Ramsay JA, Ramsay BA. 2006. On-line estimation of dissolved
methane concentration during methanotrophic fermentations. Bio-
technol Bioeng 95:788793.