Fuel 149 (2015) 95–99

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Conversion of biogas to synthesis gas over NiO/CeO2–Sm2O3 catalysts

M. Genoveva Zimicz a, Brian A. Reznik b, Susana A. Larrondo c,d,⇑
a
Instituto de Física del Sur (IFISUR) – CONICET, Av. Alem N°1253, 8000 Bahía Blanca, Buenos Aires, Argentina
b
Departamento de Ingeniería Química, Facultad de Ingeniería, Universidad de Buenos Aires, Pabellón de Industrias, Ciudad Universitaria, 1428 Buenos Aires, Argentina
c
División Materiales Funcionales para Celdas de Combustible, CINSO-UNIDEF, Juan B. de La Salle 4397, 1603 Villa Martelli, Buenos Aires, Argentina
d
Instituto de Investigación e Ingeniería Ambiental, Universidad Nacional de San Martín, UNSAM Campus Miguelete, 25 de Mayo y Francia, 1650 San Martín, Provincia de Buenos
Aires, Argentina

h i g h l i g h t s

 NiO/Ce0.82Sm0.18O1.91 cermets were prepared for catalytic and SOFC applications.

 The solids were active without need of pre-treatment.
 The conversion of methane increases with increasing temperature and carbon dioxide to methane molar ratio.
 The catalyst showed a stable behavior during a time on stream of six hours for a carbon dioxide to methane molar ratio of 2.3.

a r t i c l e i n f o a b s t r a c t

Article history: The main objective of this work has been to study the catalytic activity in the conversion of biogas to
Received 15 March 2014 Syngas of a Ni/Ce0.82Sm0.18O1.91 oxide system, which could be used as a component in the anode material
Received in revised form 1 September 2014 of a SOFC directly fed with biogas.
Accepted 6 September 2014
The solids were active without need of pretreatment with a stable behavior during a time on stream of
Available online 19 September 2014
six hours for a carbon dioxide to methane molar ratio of 2.3. This feed composition maximize the produc-
tion of synthesis gas, with stable catalytic behavior without formation of carbonaceous compounds and
Keywords:
involving CO2 capturing, contributing to the reduction of greenhouse gases.
Biogas
SOFCs
It can be concluded that NiO/CeO2Sm2O3 catalysts exhibited promising catalytic performance for their
Ceria-based materials use in the composition of anode material of solid oxide fuel cells fed with biogas.
Syngas Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction processes. In this context, the biomass arises with a renewed role
as a source of renewable fuels, which become an alternative to fos-
Energy is an integral part of society and plays a vital role in sil fuels [2,3]. The main processes of converting biomass into
socio-economic development by raising the standard of living. energy include direct combustion, pyrolysis, gasification, liquefac-
The condition of economic development of a region is directly tion, anaerobic digestion, alcoholic fermentation and trans-esterifi-
related to the patterns of consumption and production of energy. cation, among others. Each technology has its advantages and
In other words, the indexes of energy consumption per capita and disadvantages depending on the type of biomass used [4]. In par-
economic-growth feed each other back and grow together [1]. ticular, the conversion of the biogas (mixtures of H2 and CO) into
In recent years, public and political sensitivity regarding envi- Syngas constitutes a possible way of revalorization of residential,
ronmental care and energy sustainability has led the investigations industrial and agricultural wastes and opens the way for the use
to the developing of clean and renewable energy sources. Two of waste for energy production. Syngas has many uses in the indus-
main concepts have become the guidelines of the development of try. It is primarily converted in methanol, which is mainly used in
new technologies for energy supply. These are the utilization of chemical synthesis. Likewise, it can be used as a fuel in solid oxide
renewable sources of energy and the development of sustainable fuel cells (SOFCs) to provide electrical energy with great efficiency
[5,6].
The biogas consists mainly of methane (CH4) and carbon diox-
⇑ Corresponding author at: División Materiales Funcionales para Celdas de
ide (CO2), with some minor components like H2S and siloxanes that
Combustible, CINSO-UNIDEF, Juan B. de La Salle 4397, 1603 Villa Martelli, Buenos
Aires, Argentina. Tel.: +54 11 47098158. must be eliminated prior any further transformation process [7].
E-mail address: susana@di.fcen.uba.ar (S.A. Larrondo). The direct use of biogas as a fuel in traditional combustion engine

http://dx.doi.org/10.1016/j.fuel.2014.09.024
0016-2361/Ó 2014 Elsevier Ltd. All rights reserved.
96 M. Genoveva Zimicz et al. / Fuel 149 (2015) 95–99

technology to produce heat and electricity is not efficient due to of biogas [13–15]. It was found that the poisoning at low temper-
the low calorific power of biogas. The conversion into Syngas ature is not recoverable just by removal of H2S from the feed
through the Dry Reforming of Methane (DRM) emerges as one pos- stream. However, poisoning at high temperature is easily reversed
sibility to take profit of this methane-rich stream. The DRM reac- just by removal of H2S from the feed stream [13]. Moreover, it is
tion is a deep endothermic reaction that proceeds according to well known that ceria-based materials may develop resistance to
the following global equation: sulfur poisoning, because the adsorption of sulfur on ceria is
reversible for concentration of H2S below 100 ppm [16].
1
CO2 ðgÞ þ CH4 ðgÞ $ 2COðgÞ þ 2H2 ðgÞ DH0 ¼ 247 kJ  mol ð1Þ Based on the information presented in the previous paragraphs,
This reaction has similarities with the Steam Reforming of the main objective of this work has been to study the catalytic
Methane (SRM), activity in the conversion of biogas to Syngas of a Ni/Ce0.82Sm0.18-
O1.91 oxide system, which could be used as a component in the
1
H2 OðgÞ þ CH4 ðgÞ $ COðgÞ þ 3H2 ðgÞ DH0 ¼ 206 kJ  mol ð2Þ anode material of a SOFC directly fed with biogas.

However, in the DRM reaction the H to C atomic ratio in the feed

stream is lower than in the case of the SRM reaction, creating a
higher risk to form carbonaceous deposits. From the thermody- 2. Materials and methods
namic analysis presented in the literature it can be seen that these
carbonaceous compounds can be formed in a wide range of reac- The NiO/Ce0.82Sm0.18O1.91 catalyst was obtained by the incipient
tion conditions [8,9]. It can also be concluded from the thermody- wetness impregnation of ethanolic solutions of Ni(NO3)2 on com-
namic analysis that the DRM reaction is always accompanied by mercial powders of Ce0.82Sm0.18O1.91 (Nextech Materials; specific
the Reverse Water Gas Shift reaction (RWGS), surface area = 28.6 m2 g1) in order to get a final content of
1 9 wt.% of Ni. After the impregnation, the solid is dried at 90 °C
CO2 ðgÞ þ H2 ðgÞ $ COðgÞ þ H2 OðgÞ DH0 ¼ 41 kJ  mol ð3Þ
and afterwards calcined at 350 °C during two hours, with a heating
This is the main drawback of the DRM reaction, recently pointed ramp of 5 °C min1. With this heating treatment all the Ni(NO3)2 is
out by Oyama et al. [10], who called attention to the misuse of the converted to NiO.
DRM reaction for hydrogen production. These researchers noted The catalysts were characterized with X-ray Powder Diffraction
that this reaction is important for the production of Syngas when (XPD), Temperature Programmed Reduction (TPR), Scanning
it can be directly used without further processing, like in the Gas Electron Microscopy (SEM) and N2 physisorption.
to Liquid (GTL) processes to obtain liquid fuels, or, as it was men- XPD experiments were performed with a Phillips PW3710
tioned before, as a fuel for SOFCs. H2 and CO mixture can be easily diffractometer operated with Cu Ka radiation and a graphite
electrochemically oxidized in the anode chamber of that kind of monochromator. Data was collected in the angular region of
fuel cells. In particular, the technology of SOFCs probably consti- 2h = 20–80° with a step size of 0.03° and a time per step of 4 s.
tutes the only technology capable to convert biogas methane Temperature Programmed Reduction (TPR) experiments were
directly to syngas taking advantage of the presence CO2 in the performed in a Micromeritics Chemisorb 2720 equipment using a
anode chamber. However, it is necessary to develop appropriate heating ramp of 10 °C min1 in a flow consisting of 5 mol% H2 in
materials for this technology, in particular anode materials having N2 (50 cm3(STP) min1). The experiments started at room temper-
catalytic activity to produce the internal dry reforming of methane ature and finished at 920 °C.
and avoiding or disfavoring the formation of carbonaceous resi- The morphology of the samples was examined by SEM tech-
dues that would harm the stability of the cell performance [11]. nique. The images were obtained with a Zeiss Electron Beam Field
It is important to note that in addition to the catalytic activity of Emission SEM-Supra 40. In order to avoid charging problems, the
the anode material, this material must have the appropriate ionic samples were placed over carbon conductive ribbon.
and electronic conductivity to allow good performance of the fuel The specific surface area was determined by the BET method
cell. from the N2-physisorption isotherms obtained at (196) °C.
From the point of view of the catalytic activity, many catalytic Catalytic experiments were performed in a fixed bed lab-scale
systems have been studied for the DRM reaction, showing those reactor, operated isothermally and at atmospheric pressure. The
based on Rh better performance [1]. However, the developing of reactor consisted of quartz tube of 12 mm outer diameter and
Ni-based catalytic systems is important in order to make this tech- 11.2 mm internal diameter. The reactor was placed in an oven pro-
nology economically viable in commercial applications. Nickel is vided with a temperature controller. In addition, the catalyst bed
an active dehydrogenating catalysts but it is necessary to support was prepared with catalysts and inert material in a mass ratio 1–
it on a system with the capacity to produce the rapid oxidation 6 in order to favor temperature uniformity in the catalytic bed.
of the methane dehydrogenated fragments in order to prevent The temperature was monitored with a K-type thermocouple
the formation of carbonaceous deposits. Additionally, the support located axially in the center of the catalyst bed, whose signal was
must have basic sites to produce the adsorption of CO2, weakening transmitted to an electronic thermometer to determine the actual
of the CO–O bond and dissociation or the CO2 [3,12]. It is well- reaction temperature. This thermocouple could be moved along
known that ceria-based oxides have the capability to easily the reactor axe in order to control the isothermicity of the fixed
exchange oxygen with the surrounding atmosphere (Oxygen bed. The feed flow was controlled with a mass-flow controller.
Storage Capacity, OSC). The partial replacement of Ce4+ by aliova- The compositions of the feed and exit streams were analyzed by
lent cations, like Sm3+, increases the OSC and the number of oxygen on-line gas chromatography using a Clarus 50 Perkin Elmer gas
vacancies and their mobility, being these vacancies active sites for chromatographer equipped with a Thermal Conductivity Detector
dissociative adsorption of CO2. Besides, Sm–Ce mixed oxides have (TCD) and an automatic injection valve.
emerged as excellent mixed ionic electronic conductors with Two kinds of catalytic tests were made. The first group of tests
appropriate properties to be used as electrolytes in SOFCs operat- studied the evolution of the catalytic performance with tempera-
ing at intermediate temperatures (T < 800 °C). Finally, it is impor- ture: (i) Y0CH4 = 0.12 and molar feed ratio RA ðY0CO2 =Y0CH4 Þ = 1; (ii)
tant to consider the possibility of the existence of sulfur Y0CH4 = 0.12 and molar feed ratio RA in the range 1–2.3; where
compounds in the feed. In the literature we can find reports about Y0CH4 and Y0CO4 represent the methane and carbon dioxide molar
the sulfur poisoning of Ni-based catalysts during steam reforming fraction in the feed.
 M. Genoveva Zimicz et al. / Fuel 149 (2015) 95–99 97

The second type of test involved the stability study performed

at Y0CH4 = 0.12 and molar feed ratio RA = 1.
Previous catalytic tests were performed in order to assure neg-
ligible contribution of the homogeneous reaction, fluid-dynamic
behavior compatible with the plug-flow model and chemical con-
trol during the catalytic tests [17].
The following expressions will be used to present the catalytic
results:
Methane conversion:

F0CH4  FsCH4
XCH4 ¼ ð4Þ
F0CH4

where F0CH4 is the methane molar flow in the feed and FsCH4 the meth-
ane molar flow in the reactor exit.
Carbon dioxide conversion:

F0CO2  FsCO2
XCO2 ¼ ð5Þ Fig. 2. TPR profiles of NiO, support (SDC) and fresh catalyst (SDC 9% Ni).
F0CO2

where F0CO2 is the carbon dioxide molar flow in the feed and FsCO2 the
carbon dioxide molar flow in the reactor exit. profile shows two peaks, the first one at 400 °C and the second
Hydrogen production: one at 570 °C. Quantifying the hydrogen consumption yields to a
percentage of reduction of 100 ± 5%, pointing out that the oxide
FsH2 reduces according to the following equation:
PH2 ¼ ð6Þ
2  F0CH4
0
NiOðsÞ þ H2 ðgÞ ! Ni ðsÞ þ H2 OðgÞ ð8Þ
where FsH2
is the hydrogen molar flow in the reactor exit.
Carbon monoxide production: Analyzing the TPR profile of the Ce–Sm support, it can be seen
that the TPR profile shows a shoulder at the temperature of
FsCO 500 °C and a very broad peak at 800 °C. Quantifying the hydrogen
PCO ¼ ð7Þ
F0CH4 þ F0CO2 consumption, it is obtained a total reduction of 31% of the cerium
sites present in the solid. This is consistent with the fact that the
replacement of Ce4+cations by Sm3+ leads to the formation of oxy-
3. Results and discussion gen vacancies and stabilizes the oxidized state of cerium cations.
The NiO/Ce0.82Sm0.18O1.91 TPR profile is complex and cannot be
3.1. Characterization of fresh catalyst obtained as a superposition of the profiles of the NiO and the Ce–
Sm support, indicating a synergistic effect between the support
In Fig. 1 the XPD patterns of support, nickel oxide and fresh cat- and the supported phase. It can be observed a broad reduction pro-
alyst are plotted together. In the diffraction pattern of the support file with a wide reduction temperature range: 170–900 °C, with
it is possible to see the typical fluorite structure of the ceria-based two unresolved peaks in the 200–300 °C region, with maxima at
materials. The diffraction pattern of the fresh catalyst corresponds 235 °C and 275 °C, a very intense peak at 350 °C and a broad peak
to the overlapping of support and nickel oxide diagrams. No addi- at 730 °C. The reductions of both NiO and Ce–Sm support have
tional peaks were observed. moved to lower temperatures. The NiO is more easily reduced
In Fig. 2 the TPR profiles of Ce–Sm support, nickel oxide and when supported while the main reduction peak of the support
NiO/Ce0.82Sm0.18O1.91 catalyst are depicted together. Pure NiO remains broad but with the maximum slightly shifted to lower
temperatures. It is widely considered that low temperature peaks
are related to relatively free NiO species, while high temperature
peaks are associated to NiO species that strongly interact with
the support. This strong interaction enhances ceria reducibility,
most probably due to enhancement of oxygen mobility in the
Ce–Sm support [18,19].
A SEM micrograph image of the fresh catalyst is shown in Fig. 3.
Two different particle sizes can be observed, one about 30 nm in
size which corresponds to the support and the larger one about
70 nm in size corresponding to NiO. The mapping results obtained
by Electron Diffraction Spectroscopy (EDS) are also presented in
Fig. 3 indicating a good distribution of nickel on the surface. Atomic
percentage values of metal cations determined by EDS in three dif-
ferent points of the sample are consistent with the nominal values.
The BET specific surface area of the catalyst was obtained from
the N2 physisorption isotherm. Five points were considered in the
isothermal region corresponding to relative pressure below 0.3. A
specific surface area of 26 m2 g1 was obtained. Comparing with
the value corresponding to the support, it is clear that a reduction
of a 10.5% in the specific surface area was produced in the impreg-
Fig. 1. XPD diffraction pattern of Ce–Sm support, nickel oxide and fresh catalyst. nation and calcination performed to support the NiO phase.
98 M. Genoveva Zimicz et al. / Fuel 149 (2015) 95–99

3.2. Catalytic tests

In Fig. 4 methane and carbon dioxide conversions are plotted as

a function of reaction temperature for stoichiometric composition
in the feed (RA = 1), space-time of 28.8 g h dm3(CNPT), and
Y0CH4 = 0.12. The conversion of methane and carbon dioxide
increases with increasing temperature. It is important to note that
carbon dioxide conversion is always higher than the methane con-
version at temperatures below 800 °C. This fact is usually attrib-
uted to the occurrence of the RWGS reaction, in which carbon
dioxide reacts with the hydrogen produced to generate water
and carbon monoxide. It is important to note, that in the reactor
exit there is a silica-gel fixed bed to trap any trace of water pro-
duced during the process, in order to avoid its entrance to the
gas chromatographer. At higher temperatures the conversions tend
Fig. 4. CH4 and CO2 conversion vs. temperature: RA = 1; space-
to be equal. In Fig. 5 the production of hydrogen and carbon mon-
time = 28.8 g h dm3(CNPT). Y0CH4 = 0.12.
oxide are depicted together. It is noted that at low temperatures,
yields have a similar value. At temperatures above 700 °C the H2

Fig. 5. Hydrogen and carbon monoxide production vs. temperature; RA = 1; space-

time = 28.8 g h dm3(CNPT). Y0CH4 = 0.12.

production is slightly lower than that of CO production. These

results cannot be explained only with the occurrence of DRM and
RWGS reactions. In order to see whether carbon deposits have
been formed during the experiments, stability studies were per-
formed. In Fig. 6 the evolution of methane and carbon dioxide con-
versions with time on stream are presented. The conversion of
methane is very stable during the test period. However, the carbon
dioxide conversion slightly decreases with time on stream,

Fig. 6. CH4 and CO2 conversions vs. time on stream during stability tests; RA = 1;
Fig. 3. (a) SEM micrograph image of fresh catalyst; (b) EDS mapping of cations. space-time = 28.8 g h dm3(CNPT). Y0CH4 = 0.12.
 M. Genoveva Zimicz et al. / Fuel 149 (2015) 95–99 99

Acknowledgements

The post-doctoral scholarship granted to M. Genoveva Zimicz

by the CONICET is gratefully acknowledged. This research received
financial support from PIDDEF 011/11 MINDEF and Premio FOCA
2013, Banco Galicia.

References

[1] Katuwal H, Bohara AK. Biogas: a promising renewable technology and its
impact on rural households in Nepal. Renew Sust Energ Rev 2009;13:2668–74.
[2] Munster M, Lund H. Optimal use of organic waste in future energy systems –
the danish case. Riso Int Energy Conf 2007.
[3] Dodić SN, Popov SD, Dodić JM, Ranković JA, Zaravago ZZ, Glousin MT. An
overview of biomass energy utilization in Vojvodina. Renew Sust Energ Rev
2010;14:550–3.
[4] Bridgewater AV, Toft AJ, Brammer JG. A techno-economic comparison of power
production by biomass fast pyrolisis with gasification and combustión. Renew
Sust Energ Rev 2002;6:181–246.
[5] Vayenas CG, Bebelis SI. In: Gellings PJ, Bouwmeester HJM, editors. The CRC
Fig. 7. CH4 conversions vs. RA; space-time = 28.8 g h dm3(CNPT). Y0CH4 = 0.12.
handbook of solid state electrochemistry. London: CRC Press; 1997 [chapter
13].
[6] Van Herle J, Membrez Y, Bucheli O. Biogas as a fuel source for SOFC co-
indicating a deactivation process probably occurring by the depo- generators. J Power Sources 2004;127:300–12.
[7] Plou J, Durán P, Herguido J, Peña JA. Purified hydrogen from synthetic biogas by
sition of carbonaceous deposits over the support. XPD pattern of
joint methane dry reforming and steam-iron process: behaviour of metallic
spent catalysts confirmed the presence of the more intense peak oxides and coke formation. Fuel 2014;118:100–6.
of graphite. Besides, the more intense peaks corresponding to Ni° [8] Edwards JH, Maitra AM. The chemistry of methane reforming with carbon
were also detected, indicating that NiO reduces during the dioxide and its current and potential applications. Fuel Process Technol
1995;42:269–89.
experiments. [9] Nikoo MK, Amin NAS. Thermodynamic analysis of carbon dioxide reforming of
Catalytic studies for different feed ratios, in the range 1–2.3, methane in view of solid carbon formation. Fuel Process Technol
were performed. RA values below 1 were no tested since they favor 2011;92:678–91.
[10] Oyama ST, Hacarlioglu P, Gu Y, Lee D. Dry reforming of methane has no future
carbon deposition. In Fig. 7 it is possible to observe that methane for hydrogen production: comparison with steam reforming at high pressure
conversion increases with increments in the feed ratio, reaching in standard and membrane reactors. Int J Hydrogen Energ 2012;37(13):
100% for RA = 2.3 for all the temperatures. Stability tests were per- 10444–50.
[11] Van Herle, Membrez Y, Bucheli O. Biogas as a fuel source for SOFC
formed with this feed ratio observing a stable behavior over 6 h on co-generators. J Power Sources 2004;127:300–12.
stream. Besides, the XRD patterns of the spent catalysts showed no [12] Flores E, Fuentealba P, Mondragon F. Chemical reactivity of oxygen vacancies
peaks corresponding to carbon. on the MgO surface: reactions with CO2, NO2 and metals. Catal Today
2008;133–135:216–22.
[13] Appari S, Janardhanan VM, Bauri R, Jayanti S. Deactivation and regeneration of
4. Conclusions Ni catalyst during steam reforming of model biogas: an experimental
investigation. Int J Hydrogen Energy 2014;30:297–304.
[14] Appari S, Janardhanan VM, Bauri R, Jayanti S, Olaf Deutschmannd. A detailed
Regarding the synthesis of the catalyst it was observed that the kinetic model for biogas steam reforming on Ni and catalyst deactivation due
NiO was in a separate phase with a uniform distribution over to sulfur poisoning. Appl Catal A 2014;471:118–25.
the support surface. The solids were active without need of pre- [15] Ashrafi M, Christoph Pfeifer, Tobias Pröll, Hofbauer H. Experimental study of
model biogas catalytic steam reforming: 2. Impact of sulfur on the
treatment. The conversion of methane increases with increasing deactivation and regeneration of Ni-based catalysts. Energy Fuels
temperature and carbon dioxide to methane molar ratio. The cata- 2008;22:4190–5.
lyst showed a stable behavior during a time on stream of six hours [16] Yun JW, Yoon SP, Park S, Kim HS, Nam SW. Analysis of the regenerative H2S
poisoning mechanism in Ce0.8Sm0.2O2-coated Ni/YSZ anodes for intermediate
for a carbon dioxide to methane molar ratio of 2.3. temperature solid oxide fuel cells. Int J Hydrogen Energ 2011;36:787–96.
It can be concluded that NiO/CeO2–Sm2O3 catalysts exhibited [17] Perego C. Experimental methods in catalytic kinetics. Catal Today
promising catalytic performance for their use in the composition 1999;52:133–45.
[18] Roh HS, Jun KW, Dong WS, Chang JS, Park SE, Joe YI. Highly active and stable
of anode material of solid oxide fuel cells fed with biogas. Further-
Ni/Ce–ZrO2 catalyst for H2 production from methane. J Mol Catal A
more, the conditions that maximize the production of synthesis 2012;181:137–42.
gas, with stable catalytic behavior without formation of carbona- [19] Pengpanich S, Meeyoo V, Rirksomboon T. Methane partial oxidation over Ni/
CeO2–ZrO2 mixed oxide solid solution catalysts. Catal Today 2004;93–
ceous compounds involves capturing CO2, contributing to the
95:95–105.
reduction of greenhouse gases.