You are on page 1of 14

Fuel Processing Technology 92 (2011) 678691

Contents lists available at ScienceDirect

Fuel Processing Technology


j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / f u p r o c

Thermodynamic analysis of carbon dioxide reforming of methane in view of solid


carbon formation
M. Khoshtinat Nikoo a,b, N.A.S. Amin a,
a
b

Chemical Reaction Engineering Group (CREG), Faculty of Chemical Engineering, Universiti Teknologi Malaysia, 81310 UTM, Skudai, Johor, Malaysia
Dept of Catalyst and Process, R&D of Bandar Imam Petrochemical Complex (BIPC), Mahshahr, Iran

a r t i c l e

i n f o

Article history:
Received 23 July 2010
Received in revised form 16 November 2010
Accepted 26 November 2010
Available online 30 December 2010
Keywords:
Thermodynamic equilibrium
Gibbs free energy
Reforming
Carbon formation
Methane
Carbon dioxide

a b s t r a c t
A thermodynamic equilibrium analysis on the multi-reaction system for carbon dioxide reforming of methane
in view of carbon formation was performed with Aspen plus based on direct minimization of Gibbs
free energy method. The effects of CO2/CH4 ratio (0.53), reaction temperature (5731473 K) and pressure
(125 atm) on equilibrium conversions, product compositions and solid carbon were studied. Numerical
analysis revealed that the optimal working conditions for syngas production in FischerTropsch synthesis were at
temperatures higher than 1173 K for CO2/CH4 ratio being 1 at which about 4 mol of syngas (H2/CO = 1) could be
produced from 2 mol of reactants with negligible amount of carbon formation. Although temperatures above
973 K had suppressed the carbon formation, the moles of water formed increased especially at higher CO2/CH4
ratios (being 2 and 3). The increment could be attributed to RWGS reaction attested by the enhanced number of
CO moles, declined H2 moles and gradual increment of CO2 conversion. The simulated reactant conversions and
product distribution were compared with experimental results in the literatures to study the differences between
the real behavior and thermodynamic equilibrium prole of CO2 reforming of methane. The potential of
producing decent yields of ethylene, ethane, methanol and dimethyl ether seemed to depend on active and
selective catalysts. Higher pressures suppressed the effect of temperature on reactant conversion, augmented
carbon deposition and decreased CO and H2 production due to methane decomposition and CO disproportionation reactions. Analysis of oxidative CO2 reforming of methane with equal amount of CH4 and CO2 revealed
reactant conversions and syngas yields above 90% corresponded to the optimal operating temperature and feed
ratio of 1073 K and CO2:CH4:O2 = 1:1:0.1, respectively. The H2/CO ratio was maintained at unity while water
formation was minimized and solid carbon eliminated.
2010 Elsevier B.V. All rights reserved.

1. Introduction
Numerous efforts are being directed at limiting CO2 and CH4
emissions to halt global green-house warming. Utilization of landll
gas (LFG) and a CO2-rich fossil natural gas (NG) to produce highervalue added chemicals seems promising to reduce green house
effects, since CH4 and CO2 are the predominant constituents in both
LFG and fossil natural gas. The CO2/CH4 ratio in LFG varies widely
depending on the type of waste, age of landll and its extraction
systems. Landll gas commonly consists of 4070% CH4 and 3060%
CO2. Similarly the compositions of fossil NG vary from one eld to
another. For example, the CO2/CH4 ratio for NG in Natuna's eld
(located in the South China Sea) [1] and zone D of the Dalan Formation
(located in south of Iran) are 71/28 and 85/2.5 [2], respectively.
However, only several industrial processes involving CO2 as reactant
gas have been developed due to the inert and stable characteristics of
Corresponding author. Tel.: +60 7 553 5579; fax: +60 7 5581463.
E-mail addresses: m_k_nikoo@yahoo.com (M.K. Nikoo), noraishah@fkkksa.utm.my
(N.A.S. Amin).
0378-3820/$ see front matter 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.fuproc.2010.11.027

CO2 and circumvent the thermodynamic constraints [36]. Numerous


efforts have been and are being conducted to activate and utilize CO2
[7].
Traditionally, steam reforming of methane to syngas (H2 and CO)
[68] produces light hydrocarbons or oxygenates including methanol.
But, an inevitable drawback of this process is the H2/CO ratio being 3:1
is higher than the needed ratio for FischerTropsch process [11,12].
Furthermore, a considerable amount of CO2 (the greenhouse gas) is
being produced in syngas and higher hydrocarbons production [13].
The feasible utilization of CH4 and CO2 to higher value-added products
such as higher hydrocarbons, syngas and liquid oxygenates are being
investigated [3,14,15]. Besides being exploited as fuel, higher
hydrocarbons form basic materials for industries including petrochemical, rubber and plastics. Syngas is the building block for liquid
fuel production via FischerTropsch process and also a major source of
hydrogen in the renery processes. Oxygenates such as methanol not
only represents an easy and safe way to store and transport the
energy, but methanol mixed with dimethyl ether (DME) yield an
excellent fuel for diesel engines with high cetane number and
valuable combustion properties [16].

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

Nomenclature
aik
Ak
g
fi
s
fi
fis
fig
tg
G(T,
P)
ts
G(T,
P)
g
Gl

Gsl
Gs
c
Gg
i
Gs
i
t
G(T,
P)
g
Gi, e

GgC
GSC
g

Gfi

GS
fC
Gr
ni
N
P
P
R
T
yi

Number of atoms of the kth element present in each


molecule of species i
Total mass of kth element in the feed
Fugacity of species i in gas system
Fugacity of species i in the solid phase
Standard-state of fugacity of pure component i in the
solid phase
Standard-state of fugacity of pure component i in the
gas phase
Total Gibbs free energy in gas phase (kJ)
Total Gibbs free energy in solid phase (kJ)
Partial molar Gibbs free energy of species i in a gas
phase (kJ/mol)
Partial molar Gibbs free energy of species i in a solid
phase (kJ/mol)
Standard Gibbs free energy of pure solid carbon (kJ/mol)
Standard Gibbs free energy of species i in a gas phase
(kJ/mol)
Standard Gibbs free energy of species i in a solid phase
(kJ/mol)
Total Gibbs free energy of two phase (kJ)
Standard Gibbs free energy of elements i in a gas phase
(kJ/mol)
Partial molar Gibbs free energy of carbon in a gas state
(kJ/mol)
Partial molar Gibbs free energy of carbon in a solid
state (kJ/mol)
Standard Gibbs free energy of formation for species i in
the gas phase (kJ/mol)
Standard Gibbs energy of formation for solid graphite
carbon (kJ/mol)
Gibbs free energy change of reaction (kJ/mol)
Mole of species i (mole)
Number of species in a reaction system
Standard-state pressure (1 atm)
Pressure of the reaction system (atm)
Molar gas constant (kJ/mole.K)
Temperature of the reaction system (K)
Mole fraction of species i in a gas phase

Greek Symbol
k
Lagrange multiplier
Si
Chemical potential of species i in a solid phase
gi
Chemical potential of species i in a gas phase
CS
Chemical potential of carbon in a gas phase
usc
Chemical potential of carbon in a solid phase
yi
Stoichiometric coefcient of species i
i

Fugacity coefcient of species i

Superscript
G
Gas phase
S
Solid phase
T
Total

Subscript
i
Component in the mixture
k
Element in each molecule

679

The major advantage of CO2 reforming of methane is given that


the H2/CO ratio being close to 1 makes it suitable for the synthesis of
oxygenated chemicals [15,16]. Extensive investigations have been
lured towards the development of active and selective catalysts for
the CO2 reforming of methane with Ni containing [1721] and noble
metal-supported catalyst such as Pt, Rh, Ru and Pd [2225] catalysts
being commonly used to produce syngas. However, carbon deposition
is still the main cause for catalyst deactivation due to deep cracking of
CH4 and CO2 as well as sintering of the metallic active phase.
Therefore, determination of the optimum operating condition by
considering carbon formation boundary provides a fundamental
remedy to prolong the catalyst lifetime and to surpass the performance of the reaction system. Thermodynamic analysis is usually
helpful to determine the constraints governed on a reaction system
and provide the range of operating conditions.
Research on thermodynamic behaviors of reaction systems by
calculating equilibrium compositions have been utilized in understanding
the feasibility of a vast variety of reactions. Garcia and Laborde [26]
analyzed the thermodynamics of steam reforming of ethanol in order to
investigate the possibility of superseding methanol by ethanol as a
feedstock without consideration on the existence of solid carbon in
equilibrium. After estimating the equilibrium product distributions of the
gaseous species, they checked for the formation of carbon. In another
research [27], the equilibrium compositions of CO and CO2 calculated by
the Gibbs free energy minimization method were further used to estimate
the presence of carbon deposition in all possible reactions. Maggio et al.
[28] used the same models to compare steam reforming of methane,
methanol and ethanol in a molten carbonate fuel cell, but carbon
formation was not considered in their calculations. Thermodynamic
calculations for obtaining equilibrium compositions of steam reforming of
DME was recently reported, but again carbon formation had not been
taken into account in the calculation [29]. In another development,
thermodynamic equilibrium computation of DME steam reforming in an
external reformer for fuel cell applications was performed [30] in which
Gibbs free energy of formation for solid carbon was considered in the total
Gibbs free energy minimization equation. Recently, the dry reforming of
glycerol for synthesis gas production by considering Gibbs free energy of
carbon in the total Gibbs energy minimization equation was investigated
[31].
Pertaining to CO2 reforming of methane, Gibbs free energy
minimization without considering carbon deposition was applied to
investigate the inuence of operating parameters on the methane
conversion and syngas production for reforming reactor design [12].
Also, Yaw and Amin [32] conducted a thermodynamic equilibrium
analysis on the syngas production from CO2 reforming and partial
oxidation of methane at the atmospheric pressure, but had not
considered carbon deposition. Apparently, previous investigators had
not considered carbon formation or often used the principle of
equilibrated gas to compute the carbon formation in reaction systems.
A thermodynamic analysis of CO2 reforming of methane to the cogeneration of C2 hydrocarbons and syngas has been reported [14]. The
effect of various conditions, i.e. temperature, CO2/CH4 feed ratio and
system pressure on chemical equilibrium using Chemkin Collection R3.7.1
were investigated. In the paper, the regions of carbon and no carbon
formation in the equilibrium system were demonstrated, but not the
numerical trends of carbon formation over the investigated range of
temperature, CO2/CH4 ratios and pressure. Therefore, in this paper
thermodynamic equilibrium compositions for the co-generation of C2
hydrocarbons, syngas and oxygenates (methanol and DME) from CH4 and
CO2 in view of coke formation were computed by employing total Gibbs
free energy minimization method with Aspen plus. Effects of CO2/CH4
feed ratio, temperature and pressure on carbon deposition, water
formation and H2/CO ratio in the reaction system as important parameters
for producing various products are discussed. Meanwhile, a thermodynamic analysis for oxidative CO2 reforming of methane is studied to
provide a deep insight into syngas production as the main product of

680

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

methane reforming. Appropriate temperature and CO2:CH4:O2 ratio for


syngas production from oxidative CO2 reforming of methane are also
presented.
2. Methodology
The accurate function to specify the equilibrium composition of a
reaction system can be dened by Gibbs free energy minimization.
The total Gibbs free energy of a single-phase only for a gas or a solid
system at certied temperature and pressure is a function of
equilibrium composition of compounds and can be respectively
represented as Eqs. (1) and (2):
g

-g

GT; P = i = 1 ni Ggl = i = 1 ni i = i = 1 ni Gi


-g
N
g
+ RTi = 1 ni ln f i = fi
tg

ts

-s

GT;P = i = 1 ni Gsl = i = 1 ni i = i = 1 ni Gi


-s
N
s
+ RTi = 1 ni ln f i = fi :

The equilibrium computations employing the R-GIBBS reactor


were accomplished with the Aspen plus, Aspen Tech with the
capability of simulating a maximum of 9 phases of multi-components
at equilibrium in the reaction system. The total moles of the reactants
including methane and carbon dioxide were 2. In order to simulate
both LFG and fossil NG with high CO2 content, the CO2/CH4 ratio was
varied between 0.5 and 3. The pressure was varied in the range
of 125 atm while the temperature in the range of 573 to 1473 K.
SoaveRedlichKwong (SRK) model was used as the equation of state,
since some polar components such as methanol and DME were
considered as products in the reaction system. Types of reactants,
products, temperature, pressure, moles of reactants, composition of
reactants and reaction phases have to be claried to perform the
equilibrium composition calculations. The possible products in CO2
reforming of methane were considered to be the following ten
components: CH4, C2H6, C2H4, CO, CO2, H2, DME, water and methanol
as gas species. Pure solid carbon (graphite) was in chemical and phase
equilibrium with the gas species.
The Gibbs free energy change of reaction (Gr) for each reaction at
different temperatures was calculated by Eq. (9):
-g

Hence, the total Gibbs free energy of a two-phase system can be


written as Eq. (3):
t
GT;P



-g
g
=
Gi +
ln fi = fi


-s
-s
N
N
s
+ i = 1 ni Gi + RTi = 1 ni ln fi = fi :
N
i = 1 ni

-g

N
RTi1 ni

Since the standard state is dened as the pure ideal gas state at
1 atm, fig = P = 1 atm and since Gi,ge equals to zero for each element
g
in its standard state; hence, Gg
i = G [33]. Meanwhile, the solid
phase is assumed as a pure solid carbon. As the reference state for the
solid phase is at atmospheric pressure and 25 C, the partial fugacity
can be written as shown in Eq. (4):
-s

f i = fi :
s

Considering equality of chemical potential of carbon in an


equilibrated gas to solid phase and applying Eq. (2) along with the
above assumption for solid carbon species, the equation for solid
phase obtained is shown in Eq. (5) [33]:
g

-s

-s

C = C = GgC = GSC = GC = GfC 0:


s

g
g
By substituting fi = yi
i P; fi = P = 1 and Gi = G in Eq. (1)
and using the Lagrange multiplier method, the Gibbs free energy
minimization of gaseous species can be expressed as Eq. (6):

g
fi

fig = P = 1,

= yi i P,
In the same manner, by substituting
-g
Gg
i = G and Eq. (5) into Eq. (3) the Gibbs free energy minimization with considering solid carbon can be expressed as Eq. (7):

 

 
N
i P
-g
y i
+ aik k + nC GfC S = 0: 7
ni Gfi + RT ln

i=1

P-

The above equations (Eqs. (6) and (7)) along with the constraint
shown in Eq. (8) are simultaneously solved to obtain the equilibrium
composition of the reaction system.
i ni aik = Ak

where i is the stoichiometric coefcient of species i [34]. The


equilibrium constant (K), obtained from Eq. (10), determined the
feasible range of the spontaneous occurrence of the reactions.
g
Thermodynamic data such as Gibbs free energy of formation G

for all compounds were obtained elsewhere [35].




K = exp Gr = RT

10

3. Results and discussion


3.1. Feasible reactions
The main reactions which may occur in CO2 reforming of methane
are considered in Table 1. The equilibrium constants of all reactions
that are supposed to occur are exhibited as a function of temperature
in Fig. 1. According to the thermodynamic principles [36], when the
Gibbs free energy change of reaction (Gr) is negative, the reaction is
spontaneous. Conversely, for positive Gr, the reaction is thermodynamically limited [36]. The equilibrium constant (K) (Eq. (10))
determines the extent to which the reaction occurs. The reactions
cannot be shifted to the opposite side by changing the molar ratio of

 
 
i P
-g
y i
Gfi + RT ln
+ aik k = 0:
P-

Gr = i Gfi :

Table 1
Reactions in CO2 reforming of methane.
Reaction number

Reaction

H298 (kJ/mol)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

CH4 + CO2 2CO + 2H2


CO2 + H2 CO + H2O
2CH4 + CO2 C2H6 + CO + H2O
2CH4 + 2CO2 C2H4 + 2CO + 2H2O
C2H6 C2H4 + H2
CO + 2H2 CH3OH
CO2 + 3H2 CH3OH + H2O
CH4 C + 2H2
2CO C + CO2
CO2 + 2H2 C + 2H2O
H2 + CO H2O + C
CH3OCH3 + CO2 3CO + 3H2
3H2O + CH3OCH3 2CO2 + 6H2
CH3OCH3 + H2O 2CO + 4H2
2CH3OH CH3OCH3 + H2O
CO2 + 4H2 CH4 + 2H2O
CO + 3H2 CH4 + H2O

247
41
106
284
136
90.6
49.1
74.9
172.4
90
131.3
258.4
136
204.8
37
165
206.2

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

681

H2O), namely methanation can occur at lower temperature (b800 K)


with positive magnitude of Ln(K), but are restricted at the high
temperature due to their negative Ln(K) and both reactions being
exothermic.
3.2. Effect of temperature and CO2/CH4 ratio on equilibrium reactant
conversion and product distribution

Fig. 1. Equilibrium constants of reactions involving in CH4CO2 reaction at different


temperatures and atmospheric pressure.

reactants when K is much higher than 1. But for K in the vicinity of 1,


varying the molar ratio of the reactants has considerable inuence on
the distribution of the products [31]. Whenever Gr is negative, a
larger Ln(K) indicates a spontaneous reaction is more feasible to
occur. As shown in Fig. 1, it can be deduced that CO2 reforming of
methane (reaction 1: CH4 + CO2 2CO + 2H2) to form syngas is a
favorable reaction, particularly at a temperature N1000 K, consistent
with the suggested temperature range in the previous study
[14]. Reverse water gas-shift (RWGS) as numbered reaction 2
(CO2 + H2 CO + H2O) is much affected by equilibrium within the
entire investigated temperature range. In general, CO2 reforming of
methane is typically accompanied by simultaneous occurrence of
RWGS reaction.
The high negative values of Ln(K) for CO2 OCM (Oxidative coupling
of methane) reactions (reactions 3 and 4: 2CH4 + CO2 C2H6 + CO +
H2O and 2CH4 + 2CO2 C2H4 + 2CO + 2H2O) illustrate that these
reactions are not feasible to occur except at a very high temperature.
Dehydrogenation of ethane (reaction 5: C2H6 C2H4 + H2) has the
enough tendency to occur at higher temperature for ethylene
production, although the reaction can be also affected by equilibrium
limitations. Meanwhile, reaction 5 takes place along with reactions 3
and 4 [14]. Hydrogenation of CO2 and CO (reactions 6 and 7: CO +
2H2 CH3OH and CO2 + 3H2 CH3OH + H2O) is much favorable
towards the reverse side, especially at high temperatures, as their Ln
(K) are negative.
Carbon may be formed via methane decomposition (reaction 8:
CH4 C + 2H2), disproportionation (reaction 9: 2CO C + CO2) [37],
hydrogenation of carbon dioxide (reaction 10: CO2 + 2H2 C +
2H2O) and hydrogenation of carbon monoxide (reaction 11: H2 +
CO H2O + C), although these reactions are affected by the change in
molar ratio of reactants due to their low Ln(K) within the investigated
temperature range. Reaction 8 is more plausible for carbon formation
at the higher temperature, whereas all three reactions [911] tend to
generate carbon at lower temperature (b800 K) and can be inuenced
by equilibrium limitations at the higher temperature. Reactions 12
(CH3OCH3 + CO2 3CO + 3H2), 13 (3H2O + CH3OCH3 2CO2 + 6H2)
and 14 (CH3OCH3 + H2O 2CO + 4H2) however, can be improved
towards the right-hand side within the whole considered temperature range, while reaction 15 (2CH3OH CH3OCH3 + H2O) is easily
inuenced by equilibrium limitations. The composition of the
products can be greatly varied for those reactions whose Ln(K) are
in the vicinity of zero within the considered temperature range.
Reactions 16 (CO2 + 4H2 CH4 + 2H2O) and 17 (CO + 3H2 CH4 +

3.2.1. Methane conversion


The effect of operating temperature at atmospheric pressure on
the equilibrium state is shown in Fig. 2 for methane gas. For all CO2/
CH4 ratios, CH4 conversion almost quickly increases with increasing
temperature up to 1000 K, beyond which the conversion goes up
smoothly to attain unity. Meanwhile, CH4 conversion increases with
CO2/CH4 ratio implying the CO2 gas as a soft oxidant has a positive
effect on CH4 conversion in the temperature range of interest. The
positive effect is more pronounced at temperatures lower than 973 K;
therefore, adding more CO2 to CH4 as an active oxidant brings about a
higher activity for methane molecules. Nevertheless, exothermic
reactions 16 and 17 may be involved in decreasing methane
conversion at the lower temperature. For example, the equilibrium
conversion reported in [14] was about 42% when CO2/CH4 ratio and
temperature were 1 and 873 K, respectively, while our work reported
a value of 82%. It is supposed that methane decomposition (reaction 8)
is the main reason for the increase in methane conversion. At CO2/CH4
ratio being 0.5 to 1 for reaction 1, CO2 gas acts as a limiting reactant
and is not able to convert CH4 completely. Therefore, high conversion
of CH4 at higher temperatures can be ascribed to reaction 8 as the
predominant reaction (see Fig. 1) to form hydrogen and carbon.
From the experimental results reported by Khalesi et al. [38]
displayed in Table 2, it can be observed that the values of CH4
conversions using Sr0.8La0.2Ni0.3Al0.7O2.6 catalyst are less than those of
equilibrium CH4 conversions at lower than 873 K. Owing to the fact
that the thermodynamic equilibrium calculations were performed
considering solid carbon, the values of CH4 conversions less than that
of equilibrium conversions may imply CH4 decomposition cannot
improve easily; therefore, the resistance of the catalyst to carbon
deposition. Although, for temperatures N923 K, the presence of the
catalyst caused a higher CH4 conversion compared to equilibrium
condition, it may be due to the small susceptibility of the catalyst to
carbon deposition at higher temperatures, evident by the presence of
solid carbon on the surface of the catalyst. For the experimental
results of Jablonski et al. [39], all CH4 conversions were below
equilibrium, suggesting higher resistance to carbon deposition
compared to the results of Khalesi et al. [38] in the whole investigated
temperature range.

Fig. 2. CH4 equilibrium conversion as a function of temperature and CO2/CH4 ratio at


1 atm n (CH4 + CO2) = 2 mol.

682

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

Table 2
Reactant conversions (X) and product yield (Y) for CO2 reforming of methane with CO2/CH4 ratio of 1 at atmospheric pressure.
Temperature (K)

XExpCH4 (%)

XEqCH4 (%)

XExpCO2 (%)

XEqCO2 (%)

YExpH2 (%)

YEqH2 (%)

YExpCO (%)

773

52.5a

71a
20.4b
81.9a
40.6b
90.7a
58b
95.4a
72.4b
97.9a
82.6b
100a
85.1b

70.9

37.1b

59.1b

78.4b

91.7b

95.7b

96.5b

45.9

47a

64a

78.8a

87.7a

91.1a

91.5a

92.8a

28.5

31.1a

45.3a

61.6a

75.6a

80.6a

83.2a

84.4a

823
873
923
973
1023
1073

76.6
82.3
87.4
91.5
94.4
96.3

44.8
47.2
54.5
66.3
79.1
88.6

40.1
52.5
64.9
75.9
84.6
90.6

YEqCO (%)
3.5
8.3
17.4
31.9
50.8
69.3
82.9

H2/COExp

H2/COEq

1.5a

1.4a
0.41b
1.28a
0.49b
1.16a
0.55b
1.13a
0.6b
1.1a
0.63b
1.1a
0.65b

8.1
4.8
3
2
1.5
1.2
1.1

Exp and Eq are representatives of experimental data by the other researchers and equilibrium values calculated in our work, respectively.
a
The values are corresponding to Sr0.8La0.2Ni0.3Al0.7O2.6 taken from a research by Khalesi et al. [38].
b
The values are corresponding to Pt/Na(0.3 wt.%)Al2O3 taken from a research by E. L. Jablonski et al. [39].

3.2.2. Carbon dioxide conversion


Since CO2 as a soft oxidant has a positive effect on CO2 reforming of
methane, conversion of CO2 needs to be considered and discussed.
Although methane conversion increases within the considered
temperature range and CO2/CH4 ratios, Fig. 3 depicts two trends of
CO2 conversion versus temperature for all CO2/CH4 ratios. Initially,
CO2 conversion gradually decreases with temperature beginning from
573 K to about 823873 K (depending on the CO2/CH4 ratios). The
rst decreasing trend can be mainly described by reaction 10, which
converts CO2 and hydrogen to a large quantity of carbon and water.
The exothermic reaction spontaneously occurs at the low temperature, but diminishes as the equilibrium constant decreases and
reduces CO2 conversion. Reaction 16 that is exothermic and favorable
at the lower temperature may be also attributed to the same
decreasing trend. The trend can also be veried by decreasing moles
of water in the mentioned temperature range (Fig. 8). None of the
other reactions are signicantly involved in CO2 conversion, since
they have a negative value of Ln(K) within the mentioned temperature range.
The following trend illustrates that CO2 conversion begins to
enhance, as endothermic reactions 1 and 2 are favorable at higher
temperatures. This result is consistent with what has been obtained
from previous calculations regardless of the amount of the CO2
conversion which is due to the reactions involved in carbon formation
[12,14]. Temperatures above 873 K favor the conversion of CO2 as

Fig. 3. CO2 equilibrium conversion as a function of temperature and CO2/CH4 ratio at


1 atm for n (CH4 + CO2) = 2 mol.

depicted in Fig. 3. Equilibrium conversion of CO2 reaches a maximum


between 1273 K and 1473 K for all CO2/CH4 ratios as similarly
reported [40]. CO2 is completely converted at CO2/CH4 ratio of 0.5
and temperature of 1273 K better than when CO2/CH4 has a larger
proportion since CO2 is the limiting reactant. Whenever the CO2/CH4
ratio is greater, CO2 conversion is less, as methane more intensively
plays the role of a limiting reactant to the extent that at CO2/CH4 ratios
higher than 1, equilibrium conversion of CO2 cannot be completed.
Table 2 displays that the values of CO2 conversions reported by
Jablonski et al. [39] are higher than the equilibrium ones for all
investigated temperatures, except for 823 K where experimental CO2
conversion is lower than its equilibrium value. Besides, equilibrium
conversions of CO2 are lower than those of CH4 within the whole
investigated temperature range. Conversely, the experimental CO2
conversions were greater than CH4 conversions. It is worth mentioning that the variation trends of experimental conversions of CO2 and
CH4 are closer to the results of the previous researches [12,14]
obtained without considering the solid carbon formation, suggesting a
lower carbon deposition in this catalytic reaction compared to the
equilibrium condition. The dissociation of CO2 followed by CO and O is
thought to be more preferable than CH4 decomposition by the catalyst
leading to higher conversion of CO2 than that of CH4. This result
presents a good agreement with the research by Wisniewski et al. [41]
in which conversions of CO2 exceeded those of CH4 while no carbon
deposition was observed in CO2 reforming of methane over Ir/CGO
catalysts.
3.2.3. Syngas production
The region for H2 and CO production with respect to CO2/CH4
ratios can be divided to two: CO2/CH4 N1 and CO2/CH4 b1.
Fig. 4 depicts the production of H2 gas as a function of different
temperatures and CO2/CH4 ratios at the atmospheric pressure. When
CO2/CH4 ratios b1, the amount of H2 produced enhances within the
whole investigated temperature, as CO2 is the limiting reactant and
the RWGS reaction cannot simultaneously improve along with
reaction 1, as much as when the CO2/CH4 is higher than 1. Meanwhile,
the number of H2 moles produced decreases with increasing CO2/CH4
ratio from 0.5 to 1 at a specied temperature. As RWGS is not being
much involved, whenever CO2/CH4 ratio is larger, reaction 1 proceeds
better and faster suppressing reaction 8, which causes the lower
amount of H2 production.
In the case of CO2/CH4 ratios N1, the number of moles of H2 steps
up with increasing temperature, attains a maximum around 973
1023 K (depending on the CO2/CH4 ratio), and then slightly decreases
at higher temperatures. The declining trend of H2 either for specied
CO2/CH4 ratios (N1) versus different temperatures or for specied

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

683

Fig. 4. Moles of H2 as a function of temperature and CO2/CH4 ratio at 1 atm for


n (CH4 + CO2) = 2 mol.

Fig. 5. Moles of CO as a function of temperature and CO2/CH4 ratio at 1 atm for


n (CH4 + CO2) = 2 mol.

temperature (N973 K) versus different CO2/CH4 ratios (N1) are


presumably ascribed to RWGS reaction (reaction 2) in which H2
produced reacts with CO2 to form water and CO. Generally, H2
production becomes lower with increasing CO2/CH4 ratio due to CH4
being a more intensive limiting reactant restricted the source of
hydrogen atoms. This result is compatible with the results of the
thermodynamic analysis reported previously [12,14]. In comparison,
the number of moles of H2 produced in CO2 reforming of methane at
CO2/CH4 ratio of 1 equals to those reported in the past [42]. Moreover,
from an investigation on CO2 reforming of methane using Ni/Al2O3 at
1000 K, H2 mole fractions decreased as the feed ratio increased from
0.5 to 3, which had a good agreement with the thermodynamic
calculations without considering carbon deposition [43]. Consequently, this catalyst seems to be carbon resistant as the values of H2 mole
fractions were tted by the equilibrium curve. The H2 mole fractions
obtained from our thermodynamic calculations also were almost
identical to the previous thermodynamic calculation without considering carbon formation except for CO2/CH4 ratio of 0.5. The H2 mole
fraction reached 0.6 for CO2/CH4 ratio of 0.5 with respect to our
calculations, but 0.4 for thermodynamic calculations without considering carbon deposition (the mole fractions are not shown)
[12,14,43]. The greater amount of H2 mole fractions obtained from
thermodynamic calculations considering solid carbon can be attributed to the occurrence of CH4 decomposition. Comparing the
experimental results of Khalesi et al. [38] with the equilibrium
calculations in Table 2, the experimental H2 yields exceeded the
equilibrium yields of H2 at all investigated temperatures. The higher
values of H2 yields than the equilibrium ones at lower 873 K cannot be
justied by the occurrence of CH4 decomposition as the CH4
conversions attained lower values than thermodynamic calculations,
as mentioned before. It may be due to the signicant ability of the
catalyst in improving CO2 reforming of methane (reaction 1).
Fig. 5 depicts the number of moles of CO versus CO2/CH4 ratio and
temperature at atmospheric pressure. Higher temperatures favor CO
production for any CO2/CH4 ratios, since all the reactions involved in
CO production are endothermic (see Table 1). At CO2/CH4 ratio b1 and
specied temperature, CO production enhances along with increasing
CO2/CH4 ratio, as CO2 is the limiting reactant. However, increasing
CO2/CH4 ratio N1 at higher temperatures brings about a decrease in CO
production, as CH4 becomes the limiting reactant and CO2 conversion
attains a lower quantity. These variation trends of CO production are
consistent with the experimental results of CO2 reforming of methane
on Ni/Al2O3 at 1000 K by Takano et al. [43].
Contrary to the decreasing trend of H2 production for a specied
CO2/CH4 ratio (N1) versus temperature (N973), CO production is

favorable at the above mentioned range due to RWGS reaction


(reaction 2). In RWGS reaction, H2 reacts with CO2 to produce CO and
water. Hence, it is inevitable to see the opposite trends in H2 and CO
production at temperatures above 9731023 K and CO2/CH4 ratios N1.
With reference to Table 2, higher experimental CO and H2 yields than
their equilibrium ones along with high methane conversions could
provide evidence for the signicant ability of the catalyst in
dissociation of CO2 followed by CO and O production as well as
improving CO2 reforming of methane. Furthermore, both ascending
experimental CO and H2 yields with temperature conrm that RWGS
reaction is not involved when CO2/CH4 ratio is 1.
Comparing this work with the preceding study which did not
consider methane decomposition (reaction 8) in the reaction system
[14], contradict trends of moles of CO and H2 for CO2/CH4 b1 at a
specied temperature is observed. The number of moles of CO
increases opposite to that of H2 with increasing CO2/CH4 ratio up to 1
in our work, but both increased in the previous study [14]. As a matter
of fact, this contradict arises from reaction 8 which just as mentioned,
is preferable along with reaction 1 and plays a key role in H2 (and solid
carbon) production when CO2/CH4 b1, or in other words, when CO2 is
a limiting reactant. Just due to this reason, CH4 conversions reported
in our work attain higher magnitudes than previous study at the same
temperature when CO2/CH4 b1 [14]. Besides, in the present study due
to the same said reason, moles of CO and H2 achieved their maximum
at CO2/CH4 being 1 and 0.5, respectively, but both at 1 in the former
work. It is interesting to note that according to the experimental
results of Takano et al. [43], both CO and H2 productions reached to
their maximum at CO2/CH4 ratio of 1, while tting the equilibrium
curve without considering solid carbon deposition.
Fig. 6 portraits the H2/CO ratios produced from CO2 reforming of
methane as a function of temperature and CO2/CH4 ratio at
atmospheric pressure. Furthermore, this gure provides a broad
range of operating parameters for production of syngas for various
industrial applications. The desirable amount of H2/CO ratio varies
with different industrial applications. The ratio of H2/CO decreases
with increasing either temperature or CO2/CH4 ratio. As can be seen in
Fig. 6, the values for H2/CO ratios are high; especially at lower CO2/CH4
ratios and temperatures less than 823 K. Hydrogen produced in this
area is more suitable for application in industries with demand for
more concentrated H2.
A H2/CO ratio of unity is needed for FTS to produce liquid fuel
such as DME [17,44]. From Fig. 6, a H2/CO ratio in the order of 1 can
be obtained at higher than 1173 K for CO2/CH4 ratio being 1 at which
about 4 mol of syngas can be produced from 2 mol of reactants
(CO 2 + CH 4 ) with a CO 2 conversion of more than 98%. The

684

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

Fig. 6. H2/CO ratio as a function of temperature (573823 K) and CO2/CH4 ratio at 1 atm
for n (CH4 + CO2) = 2 mol.

experimental results taken from a research by Khalesi et al. [38] as


shown in Table 2, revealed a H2/CO ratio close to unity in all
temperature range studied, demonstrating the considerable activity
and selectivity of the catalyst towards CO2 reforming of methane
(reaction 1). It is more pronounced at temperatures higher than 923 K
where H2/CO ratio remained at the constant value of 1.1, representing
the preference of CO2 reforming of methane. Furthermore, equilibrium H2/CO ratios greater than the experimental ones at a lower
temperature are a sign of more carbon formed due to the preferable
occurrence of CH4 decomposition. In another research by Jablonski et
al. [39] higher CO2 conversions along with lower H2/CO ratios
compared to the corresponding equilibrium ones are associated
with the RWGS reaction (see Table 2). In other words, the exploited
catalyst in their work was selective towards RWGS reaction, although
the CO2/CH4 was 1.
A H2/CO ratio of about 2 is needed to produce methanol from
syngas. Up to now, Figs. 46 reveal that the suitable condition for
syngas to methanol production is a temperature higher than 1123 K,
CO2/CH4 ratios of 0.5 at which about 3.5 mol of syngas can be
produced. In the next section, it will be discussed this condition
cannot be acceptable due to a signicant amount of carbon deposition
in the reaction system.
3.2.4. Carbon production
All reactions involve in carbon formation (reactions 811) can be
relatively affected by operational parameters due to their low
equilibrium constants. From Fig. 7, carbon formation decreases with
increasing temperature, as higher temperature is a barrier for
improving the exothermic reactions of 9, 10 and 11 (2CO C + CO2,
CO2 + 2H2 C + 2H2O and H2 + CO H2O + C) which involved carbon formation. When the CO2/CH4 is higher than 1, moles of carbon
disappear at the lower temperature. However, a considerable and
nearly constant amount of carbon still remained for CO2/CH4 ratio
equals to 0.5 at temperatures higher than 1073 K as reaction 8
becomes more plausible. Moreover, it can be explained by reverse of
disproportionation (reverse of reaction 9: C + CO2 2CO) which is
favorable at the higher temperature (especially at CO2/CH4 N1), but
the rate reduces when CO2 is the limiting reactant [45,46]. This is
consistent with the existence of less amount of CO at CO2/CH4 ratio of
0.5 compared to higher CO2/CH4 ratios. In the case of increasing
carbon formation for CO2/CH4 ratio of 0.5 at low temperatures up to
873 K while CO2 is a limiting reactant, it can be probably due to
endothermic reaction of methane decomposition (reaction 8) which
improves with increasing temperature. Meanwhile, the number of
moles of carbon for CO2/CH4 ratio of 0.5 is lower than that of CO2/CH4

Fig. 7. Moles of carbon as a function of temperature and CO2/CH4 ratio at 1 atm for
n (CH4 + CO2) = 2 mol.

ratio of 1 at temperatures less than 723 K. It may be mainly due to


reaction 10, which is not plausible when CO2 is the limiting reactant.
Also, carbon formation decreases with increasing CO2/CH4 ratio (N1)
at a constant temperature, since CH4 becomes more intensively a
limiting reactant and the amount of H2 available for reactions 10 and
11 are less, which results in a decrease of carbon formation. Moreover,
in this case, reaction 8 is not much involved in carbon formation due
to CH4 being a limiting reactant.
According to thermodynamic calculations presented in Figs. 6 and
7, negligible amount of carbon would be formed to achieve a H2/CO
ratio of unity at temperature higher than 1173 K for CO2/CH4 ratio of
1. However, about 0.7 mol solid carbon is produced in order to obtain
a H2/CO ratio of about 2 for methanol production at a temperature
higher than 1123 K and CO2/CH4 ratio of 0.5. Although CO2 reforming
of methane seemed to be a suitable reaction system for production of
syngas to methanol production signicant amount of carbon formation lowers the process efciency and increases the operational
cost. Moreover, CO2 reforming of methane would be a useful method
for syngas production with a negligible amount of carbon in DME
production, since production of DME needs a H2/CO ratio of unity [18].
Choudhary et al. [25] compared the rate of carbon deposition
over different Ni-, Co-, or noble metal containing catalysts at 1123 K
during the CO2 reforming of methane with CO2/CH4 ratio of 1.1. The
Ru (5 wt.%) /Al2O3 achieved methane conversion of 97.8%, 1.9 mol of
H2, and H2/CO ratio of 0.96. They reported 0.106 mol of solid carbon,
approximately identical to their corresponding equilibrium quantities. The moles of the products have been calculated based on 2 mol
of reactants containing CO2 and CH4 for comparison with our work.
Meanwhile, no carbon was formed when CoNdOx (Co/Nd = 1) was
used as a catalyst [25]. Resistance of this catalyst to carbon deposition
was attributed to the strong interaction between metallic Co and
Nd2O3 particles, which assisted in the uniform dispersion of the
metallic Co on the Nd2O3 support [25]. However, NiMgOx (Ni/Mg = 1)
showed a higher amount of carbon deposition in the order of 0.35 mol
at the same temperature (1123 K). Practically, it may suggest applying
a higher temperature than the equilibrium limiting temperature (the
temperature above which carbon formation is avoided from the reaction equilibrium) to prevent carbon deposition. On the other hand,
using higher operation temperature leads to consuming more energy
and even catalyst sintering. It has been proven that graphitic carbon is
inevitable to form using the nickel catalyst, especially for CO2/CH4
ratio b1. The greater carbon deposition on the latter catalyst may be
ascribed to the inherence of the kinetic mechanism for CO2 reforming
of methane on Ni catalysts. Ni catalysts kinetically catalyze CO

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

685

disproportionation, CH4 decomposition and CO dissociation [47,48].


As a matter of fact, carbon deposition arises from both CH4 decomposition and intensive CO2 dissociation on the catalyst surface
[46,47], which causes excess carbon formation compared to thermodynamic equilibrium quantity. The initial step of CO2 reforming of
methane is methane decomposition to solid carbon (and H2) which is
then oxidized with CO2 (via reverse disproportionation reaction) in
the reaction system or kinetically by the reducible oxides on the
catalyst surface to produce CO [46,4850]. In the absence of sufcient
reducible oxides on the catalyst structure, carbon would be deposited.
In the following, CO2 dissociates to CO and O on the catalyst surface;
therefore, O reoxidizes the catalyst surface and enables it to be competent for ceaseless redox reactions to remove solid carbon. When the
rate of methane decomposition surpasses the CO2 dissociation, solid
carbon would be deposited [48]. Hence, the presence of CO2 in the
extent to remove solid carbon formed can be determined by the redox
capability of the catalyst structure in addition to the carbon-free
temperature of the equilibrium reaction system.
Noble metals such as Pt, Ru, Rh and Pd are reported to be superior
resistant towards carbon formation [23,48], in some cases even for
CO2/CH4 b1 [51]. For example, no carbon formation was seen on the
Rh/-Al2O3 monolithic catalyst for CO2/CH4 ratio of 1 and 0.71;
however, solid carbon was observed over the investigated temperature range based on our thermodynamic equilibrium calculations.
Moreover, the carbon deposition mechanism on Rh-containing
catalysts may be different from Ni-containing catalyst [47]. Therefore,
different types and preparation methods for CO2 reforming catalysts
make diverse type of deviation from the thermodynamic equilibrium
prole.

which is not plausible when CO2 is the limiting reactant, just as was
discussed for carbon formation. Increasing CO2/CH4 ratio ( N1) at
specied temperature (b973 K) causes a decrease of water formation
since fewer hydrogen atoms would be available to form water and
even H2 as CH4 is the limiting reactant. This can be corroborated by
Fig. 4 where H2 production decreases at CO2/CH4 ratios larger than 1
at the identical temperatures.
Raising the CO2/CH4 ratio from 0.5 to 3 increases the moles of
water at higher temperature (N1000 K). It is in good agreement with
the results of equilibrium composition of water that did not consider
solid carbon irrespective of the amount of water formed [14,43]. The
moles of water formed considering carbon deposition exceeded the
equilibrium ones when carbon formation was not taken into account
due to participation of CH4 decomposition, carbon monoxide and
carbon dioxide hydrogenation reactions associated with solid carbon
and water production while carbon deposition is considered. From
Fig. 8, the increasing number of moles of water at CO2/CH4 ratios being
2 and 3 within the temperature range of higher than 9731023 K
(depending on the reactant molar ratios) is reasonably owed to RWGS
reaction (reaction 2) which is conrmed by enhanced moles of CO,
declined moles of H2 and smooth increase of CO2 conversion. These
outcomes are compatible with preceding research regardless of a little
difference in the value of the temperature at which water formation
increased [14,44]. However, exploiting Ir/CGO catalyst in CO2
reforming of methane reduced water production at higher than
700800 C depending on the CO2/CH4 ratios due to improving
preferable reaction between water formed and methane (steam
reforming of methane) which lowered water concentration [41].

3.2.5. Water production


The number of moles of water produced at different temperatures
and CO2/CH4 ratios are exhibited in Fig. 8. Although similar reducing
trends can be seen before 9731023 K (depending on considered CO2/
CH4 ratio), two different trends are observed after that. According
to the above discussion, exothermic reactions 10 and 11 (CO2 +
2H2 C + 2H2O and H2 + CO H2O + C) are associated with carbon
formation as well as water production at lower temperatures. Hence,
with increasing temperature and decreasing equilibrium constants for
reactions 10 and 11 (see Fig. 1); the number of moles of water reduces
like carbon as shown in Fig. 8. Furthermore, exothermic reactions 16
and 17 limit water production when temperature increases. Meanwhile, the amount of water formation for CO2/CH4 equals to 0.5 was
lower than that for CO2/CH4 ratio of 1, possibly due to reaction 10,

3.2.6. Ethylene and ethane production


The number of moles of C2H6 and C2H4 produced at different
temperatures and CO2/CH4 ratios are depicted in Figs. 9 and 10. These
gures illustrate that the number of moles of C2H6 and C2H4 formed in
the reaction system decreases at higher CO2/CH4 ratio as fewer H
atoms from methane are available to react for C2H6 and C2H4
production.
As can be seen from Figs. 9 and 10, moles of C2H6 and C2H4 initially
enhance with increasing temperature as both reactions 3 and 4
are endothermic. Moles of C2H6 go through a maximum around
723773 K (depending on reactants molar ratio) but decreases at the
higher temperature as reaction 5 consumes moles of C2H6 to form
C2H4. The temperature at which moles of C2H6 start to reduce almost
coincides with the temperature at which reaction 5 occurs (see Fig. 1).

Fig. 8. Moles of water as a function of temperature and CO2/CH4 ratio at 1 atm for
n (CH4 + CO2) = 2 mol.

Fig. 9. Moles of C2H6 as a function of temperature and CO2/CH4 ratio at 1 atm for
n (CH4 + CO2) = 2 mol.

686

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

Fig. 10. Moles of C2H4 as a function of temperature and CO2/CH4 ratio at 1 atm for
n (CH4 + CO2) = 2 mol.

From Fig. 10, two different trends for C2H4 formation are observed
depending on the reactant molar ratio. When CO2/CH4 ratio b1, the
number of moles of C2H4 increases with increasing temperature. This
result resembles that of previous work [14] regardless of the values for
C2H4 formed. In this case, considering CO2 as the limiting reactant,
there are sufcient H atoms available from CH4 that can contribute in
the reactions leading to C2H4 production. Moreover, moles of C2H4
increases with a swift incline owing to reaction 5 which multiplies
C2H4 formation. From Fig. 10, a considerable increase of moles of C2H4
is exhibited relative to that of C2H6 (see Fig. 9) at temperatures higher
than 1073 K due to the higher equilibrium constant in reaction 4
compared to reaction 3 (see Fig. 1).
From Fig. 10 the trends for CO2/CH4 ratios being 2 and 3, are similar
to those for CO2/CH4 ratio b1, except at a temperature above
923973 K (depending on the reactant molar ratio) at which the
number of moles of C2H4 decreases. The decreasing trend of C2H4 for
CO2/CH4 ratios being 2 and 3 reveals that there is a relation between
reaction temperature and contribution of H atom from methane at
equilibrium system leading to a notable effect on C2H4 formation
[14]. As a result, temperatures lower than 1073 K and CO2/CH4 ratios
below 1 is more in favor of C2H6 production than C2H4 although
temperatures above 1073 K and similar CO2/CH4 ratio are more
favorable to C2H4 production as the predominant product relative to
C2H6. As CO2/CH4 ratio b1 produces a large amount of carbon (see
Fig. 7) which causes clogging in the reactor, CO2/CH4 ratio of 1 and
temperature of 1273 K are suggested for C2H4 production at which
carbon deposition is negligible. Meanwhile, due to production of the
large amount of solid carbon at CO2/CH4 ratio b1, it seems CO2/CH4
ratio of about 2 and temperature of about 1000 K is suitable for higher
C2H6 production while eliminating solid carbon. Since the number of
moles of C2H6 and C2H4 formed are very low, an active and selective
catalyst with decreasing operational temperature is strongly necessary. The catalysts performance using CaO/MnO/CeO2 to produce C2
(ethylene and ethane) has shown the optimal factors of CO2/CH4 ratio
and reactor temperature at 2 and 1127 K, respectively, corresponding
to maximum CH4 conversion of 5.1% and C2 hydrocarbons yield of 3.9%
[52]. However, the value of equilibrium CH4 conversion is 99.5%, while
an only negligible amount of C2 equilibrium yield is calculated. These
large differences from the thermodynamic equilibrium values arise
from the synergistic effect between the catalyst basicity and
reducibility, enhancing the activity towards CO2 OCM (reactions 3
and 4) [52,53]. Some preceding researches for C2 production from
CO2 reforming of methane over different metal oxide catalysts such

Fig. 11. Moles of methanol as a function of temperature and CO2/CH4 ratio at 1 atm for
n (CH4 + CO2) = 2 mol.

as CeO2ZnO, MnO2SrCO3 and MnO/CaO/CeO2 have been reported


[5457]. Although the conditions were similar with the simulation, a
maximum yield of about 4.5% was achieved revealing a more efcient
catalyst is needed for the process.

3.2.7. Methanol and DME production


Moles of methanol and DME produced at different temperatures
and CO2/CH4 ratios are depicted in Figs. 11 and 12, respectively. These
gures illustrate that the amount of methanol and DME in the reaction
system increases with enhancing temperature and reaches a
maximum around 923973 K (depending on the reactant molar
ratio) before declining.
From Fig. 1 and reactions 6 and 7, methanol can be produced from
syngas (CO, H2 and CO2) at the lower temperature than our
investigated temperature range. Furthermore, reaction 1 is a slow
reaction at the low temperature. Only small amounts of H2 and CO
present are in the reaction system and therefore, a negligible amount
of methanol can be produced at low temperatures as shown in Fig. 11.
Also, occurrence of quite exothermic methanation reactions (reactions 16 and 17) restrict methanol production at the low temperature.
As temperature increases, due to increasing syngas production from

Fig. 12. Moles of DME as a function of temperature and CO2/CH4 ratio at 1 atm for
n (CH4 + CO2) = 2 mol.

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

687

reaction 1, the number of moles of methanol produced from


reactions 6 and 7 increases to a maximum. RWGS, an endothermic
reaction is much faster at higher temperatures and seems to be the
main competitive reaction for methanol formation [56]. Hence, RWGS
reaction dominates and decreases methanol at a temperature N900 K.
The amount of methanol for CO2/CH4 ratio equals to 0.5 is higher
relative to that for greater ratios, since CO2 would be the limiting
reactant and more H atoms are available owing to the existence of
more CH4 than CO2. Meanwhile, methanol production is diminished
with the increase of CO2/CH4 ratio, especially at higher than 1 as CH4
more intensively plays the role of a limiting reactant. Although Fig. 11
shows that moles of methanol have the highest value at CO2/CH4 ratio
of 0.5 and temperature in the order of 923 K, moles of carbon formed
are signicant. Therefore, a temperature in the order of 1000 K and
CO2/CH4 ratio of 2 seem to be appropriate for methanol production
with negligible carbon deposition. However, the amount of methanol is
still very low for the mentioned temperature and a highly selective and
active catalyst is desirable for improving the methanol production.
Similar trends for DME and methanol production as a function of
temperature revealed in Figs. 11 and 12 are elucidated by reaction 15
(an exothermic reaction) at which methanol is consumed to form DME.
DME production is supported by reaction 15 at the lower temperature,
reach a maximum and start to decline at a temperature of about 973 K,
almost the same at which methanol begins to decrease. Reactions 12, 13
and 14 are active in the whole investigated temperature range and
reduce the number of moles of DME. Similar to methanol production,
CO2/CH4 ratio being 0.5 and temperature around 973 K is not a suitable
condition for DME production due to the presence of a considerable
amount of carbon. Fig. 12 reveals that temperature about 1173 K and
CO2/CH4 ratio being in unity are suitable for DME production without
considerable carbon formation provided that an active and selective
catalyst is available to overcome the thermodynamic limitation on DME
formation.
3.3. Effect of pressure on reactant conversion and product distribution
during CO2 reforming of CH4
Effect of pressure is presented only for the main products: H2, CO,
water and solid carbon, as the analysis of the equilibrium data for CO2
reforming of CH4 showed a slight quantity of C2H6, C2H4, DME and
methanol as byproducts. Fig. 13a portraits the effect of system
pressure on CH4 and CO2 conversions, the main products distribution
and H2/CO ratio at 1173 K for CO2/CH4 ratio of 1, where syngas
production and reactant conversion attained the optimum with a
negligible amount of carbon at the atmospheric pressure. CO2 and CH4
conversions are always higher at lower pressures than those at higher
pressures. This suggests that at such a high temperature, greater
pressures can suppress the effect of temperature on increased
reactant conversion. These decreased trends can be expressed by
endothermic CO2 reforming of methane (reaction 1), which tends to
shift to the left (reactant) side, according to LeChatelier's principles.
Besides, methane decomposition and CO disproportionation assist in
lowering CH4 and CO2 conversions, as well as decreasing CO and H2
formation at the higher pressures. Moreover, reactions (16) and (17)
contribute to lower CH4 conversion, the moles of H2 and CO, as well as
to augment water formation. At 1173 K, water disappears at
atmospheric pressure, but increases as the pressure increases. The
greater quantity of water at higher pressures can be mainly expressed
by RWGS [45] and in less degrees by reactions (10), (11), (16) and
(17).
Contrary to the preceding study [14] which did not consider solid
carbon in the equilibrium reaction system, the presence of solid
carbon in our thermodynamic equilibrium calculations resulted in
greater moles of H2 than those of CO within the whole investigated
range of pressure; hence, H2/CO ratio N1. As one can see in Fig. 13a,
H2/CO ratio increases from 1 at 1 atm to 1.15 at 25 atm, but in

Fig. 13. The effect of pressure on a) equilibrium conversion of reactants and products
distribution for CO2/CH4 = 1, 1173 K and n (CH4 + CO2) = 2 mol and on b) carbon
deposition as a function of temperature.

previous work [14], H2/CO ratio was less than 1. This contradict
mainly arises from methane decomposition (reaction 8) which assists
CO2 reforming of CH4 to produce the higher amount of H2. These
equilibrium data are compatible with the thermodynamic equilibrium
calculations by S-W. Hong et al. [40].
Fig. 13b shows the effect of pressure on carbon deposition at
variable temperatures during CO2 reforming of CH4. The moles of solid
carbon go up with increasing pressure to attain the maximum. Shamsi
and Johnson [45], performing thermodynamic calculations on the CO
disproportionation and methane decomposition (as the two major
reactions of carbon formation) showed that the carbon deposited by
methane decomposition decreased with increasing the pressure,
while increased by CO disproportionation. These results suggest that
disproportionation reaction is the most favorable contributor to
carbon formation at a higher pressure, particularly when the CO2/CH4
ratio is higher than 1. Meanwhile, due to disproportionation reaction,
CO decreases with a steeper incline than H2 when increasing the
pressure. On the contrary, as said before methane decomposition is a
favorable reaction when CO2 is a limiting reactant (CO2/CH4 b1).

688

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

3.4. The effect of O2 addition in CO2 reforming of CH4


As discussed in the previous section, a carbon-free CO2 reforming
of methane is practically possible by increasing the temperature
higher than 1173 K at the atmospheric pressure, yielding a considerable amount of syngas and H2/CO ratio of 1 suitable for downstream
FischerTropsch synthesis of different chemical products. However,
producing such a high temperature for endothermic CO2 reforming of
methane requires intensive energy consumption. In order to reduce
the energy requirements and to control the thermal behavior of the
reaction system, coupling exothermic reaction of methane oxidation
with the endothermic CO2 reforming of methane has attracted much
attention as a promising solution for methane reforming to syngas in
recent years [46]. To obtain an understanding of syngas production
with no carbon deposition and with minimum loss of syngas at a
lower temperature, a thermodynamic analysis was performed for
oxidative CO2 reforming of methane with an identical amount of CH4
and CO2, since the optimum condition for syngas production with
negligible carbon deposition was achieved for this ratio in CO2
reforming of methane. The main reactions involved in methane
oxidation are considered in Table 3 numbered as reactions 1 and 2.
The effects of adding different O2 contents to CO2 reforming of
methane and temperature on the equilibrium reactant conversions,
moles of H2, CO, H2O and H2/CO ratio are depicted in Fig. 14 for CO2/
CH4 ratio of 1. In comparing Fig. 2 and 14a, introducing different O2
contents to CO2 reforming of methane caused higher CH4 conversions
relative to those of CO2 reforming of methane within the whole
considered temperature range (N973 K). The increment was more
pronounced at higher O2/CH4 ratios since CH4 became more intensive
limiting reactant while being consumed in CO2 reforming, partial and
total oxidation of methane (reactions 1 and 2 in Table 3). In the
meantime, O2 was fully consumed for all O2/CH4 ratios, while CH4
conversion via partial or total oxidation of methane depending on O2/
CH4 ratios was augmented. Against CH4 conversion, increasing O2/CH4
reduced CO2 conversion due to contribution of O2 in the total
oxidation of methane leading to more CO2 production.
It can be deduced from previous research [46] total oxidation of
methane is signicantly predominant reaction over dry reforming of
methane when O2/CH4 ratio reaches minimum 0.75 at 973 K and even
greater than that at higher temperatures. Therefore, total oxidation of
methane did not prevailingly govern on our reaction system with the
dened operational conditions: O2/CH4 ratio b0.4 and temperature
N973 K. These results are consistent with the results obtained by the
other researchers [46,58,59].
According to Fig. 14a and b, the moles of CO and H2 decreased with
augmenting O2/CH4 ratios, as more moles of O2 would participate in
the partial and total oxidation of methane resulting in plausible
burning H2 and CO into CO2 and water. Although the variation of
equilibrium moles of H2 presents a descending trend with increasing
O2 contents and is in good agreement with the previous studies [32],
practically, it may differ in some cases depending on various catalysts.
For example, according to a research on oxidative CO2 reforming of
methane over CoNi/Al2O3 catalysts [46], initial increase appeared for
overall H2 production before its decrement, probably due to oxidation
of carbon residue (CxH1 x species) produced from CO2 reforming of
methane to H2 and CO.
From Fig. 14c, d and e, greater temperatures (N1023 K) favor lower
H2, higher CO and water production implying the considerable
Table 3
Oxidation reactions in oxidative CO2 reforming of methane.
No

Name of reaction

Reaction formulae

H 298 (kJ/mol)

1
2
3

Partial oxidation of methane


Total oxidation of methane
Carbon oxidation

CH4 + 1/2O2 CO + 2H2


CH4 + 2O2 CO2 + 2H2O
C (s) + 1/2O2 CO

36
802
110

contribution of RWG in the combined oxidation and CO2 reforming


of methane. Avila-Neto et al. [60] have proven that for CO2 reforming
of methane with CO2/CH4 ratio of 1, WGS reaction is the main
responsible for increasing the production of H2 and fading that of
water higher than 973 K. However, it seems that addition of O2 to CO2
reforming of methane reverses the WGS direction and reinforces
RWGS. It is more pronounced at higher O2/CH4 ratios and temperatures, which can be substantiated by a steeper decrease of H2
simultaneously, along with a steeper increase of water and CO. It may
be attributed to the availability of more CO2 produced from total
oxidation of methane.
With reference to Fig. 14e, the decreasing trend of water production
below 1023 K may be ascribed to exothermic reactions 10 and 11
associated with carbon formation as well as reactions 16 and 17
(methanation reactions). Nevertheless, the variation trend of water
production changed to a linear functionality over O2/CH4 ratio of 0.3,
since total oxidation of methane proceeds rather in the reaction system
with burning more H2 formed to produce a signicant amount of water.
The effect of different O2 contents on H2/CO ratios as a function of
temperature has been depicted in Fig. 14f. H2/CO ratios decreased as low
as unity at a temperature in the order of 1050 K and almost did not vary
from 1050 to 1273 K. Comparing the H2/CO ratios prole of our work
with those of Amin and Yaw [32] revealed a descending trend for H2/CO
ratios as O2/CH4 was incremented at lower 1073 K in our work, whereas
an ascending trend was observed for the results of Amin and Yaw [32].
This is due to not considering solid carbon formation by Amin and Yaw
in the reaction system. In fact, solid carbon formed at lower
temperatures would be burned by oxygen via reaction 3 (mentioned
in Table 3) (C + 1/2O 2 CO) and reverse disproportionation
(C+ CO2 2CO) leading to a higher amount of CO and lower H2/CO,
particularly at greater O2/CH4 ratios. The increment of the moles of CO at
greater O2 contents at lower temperature (b1000 K) in Fig. 14d as an
evidence for lower H2/CO ratio at less than 1000 K can be considered.
Fig. 15 presents the moles of carbon formed in the reformate
product of oxidative CO2 reforming of methane in chemical
equilibrium as a function of temperature with a pressure of 1 atm
and CO2/CH4 ratio of 1. Carbon formation decreased while raising the
temperature in oxidative CO2 reforming of methane just as justied
for decreasing carbon deposition regarding CO2 reforming of methane
in previous section. However, the employment of O2 in CO2 reforming
of methane caused a steeper diminish of carbon deposition compared
to CO2 reforming, which is due to the effective role of co-oxidant of O2
in burning solid carbon. When the temperature increased over
1073 K, no carbon deposition was formed for all O2/CH4 ratios
studied. The O2/CH4 ratio of 0.1 at the temperature of 973 K in
oxidative CO2 reforming of methane decreased the carbon deposition
only 28% less than in CO2 reforming of methane. Moreover, conversion
of CH4, that of CO2 and the amount of syngas obtained under this
condition were as low as 92%, 61% and 2.5 mol, respectively, but the
amount of water was as great as 0.31 mol. However, the employment
of the same value of O2/CH4 at higher temperature, 1073 K, caused a
decrement in carbon formation by 93% less than in CO2 reforming of
methane, while conversion of CH4, that of CO2 and the amount of
syngas reached to 96.5%, 87% and 3.5 mol with H2/CO ratio of 1,
respectively. Meanwhile the water formation attained its minimum in
the whole range of temperature and O2/CH4 studied. It is also
observed that although the temperature higher than 1073 K favors
the reactants conversion, caused a slight decrease of H2 production
and an increase of water formation.
Although higher O2 contents eliminated solid carbon at lower
temperatures, there is a signicant decrease in the amount of syngas
production and an increase in water formation. Previous studies [32]
proposed O2/CH4 ratios from 0.1 to 0.2 were suitable for oxidative CO2
reforming of methane at 1073 K, But referring to Fig. 14 in the present
work, ratios greater than 0.1 caused about 7% decrease of H2 and 59%
increment of water productions.

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

689

Fig. 14. Equilibrium conversion of reactants and product distribution as a function of temperature and O2/CH4 ratio at 1 atm and CO2/CH4 = 1 for n (CH4 + CO2 + O2) = 2 mol: (a) CH4
conversions; (b) CO2 conversions; (c) moles of H2; (d) moles of CO; (e) moles of water; and (f) H2/CO ratios.

In a previous literature [32], feed ratio of CH4:CO2:O2 = 1:1:0.2 was


reported as the optimal condition at a temperature below 1000 K
when solid carbon deposition was not considered in thermodynamic
calculations. However, referring to Figs. 14 and 15, a minimum value
of solid carbon in the order of 0.16 mol and that of water in the order
of 0.25 mol still remained in the oxidative CO2 reforming reaction
system. Accordingly, based on our results only the optimal operating
temperature of 1073 K and feed ratio of CO2:CH4:O2 = 1:1:0.1 have
been justied to obtain reactant conversions and syngas yields of

higher than 90% as well as to maintain H2/CO ratio of unity, while


minimizing water formation and eliminating solid carbon.
4. Conclusion
The thermodynamic chemical equilibrium analysis of CO2 reforming of methane to form syngas, C2, methanol and DME considering
the reactions involved in carbon deposition was studied using Gibbs
free energy minimization. The CO2/CH4 reactant ratio and reaction

690

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691

Fig. 15. Moles of solid carbon as a function of temperature for different O2/CH4 ratios at
1 atm and n (CH4 + CO2 + O2) = 2 mol.

temperature had a considerable inuence on the equilibrium of the


reactants conversion, equilibrium composition of the products and
solid carbon formation. Four reactions regarding carbon formation
were considered: hydrogenation of carbon dioxide, hydrogenation of
carbon monoxide, methane decomposition and disproportionation.
Methane decomposition, an endothermic reaction, is the key reaction
for carbon deposition when CO2/CH4 b1 leading to increased moles of
H2 and constant moles of solid carbon even at a very high
temperature. Adversely, hydrogenation of carbon dioxide is not
much involved in carbon deposition when CO2/CH4 b1, and it can be
attested by lower carbon deposition for CO2/CH4 in the order of 0.5
relative to that of 1 at a temperature less than 723 K. At CO2/CH4 N1,
three hydrogenation reactions, hydrogenation of carbon dioxide,
hydrogenation of carbon monoxide, and disproportionation are more
pertinent to carbon formation than methane decomposition, as CH4 is
the limiting reactant. This was veried by a gradual decrease of water,
hydrogen, solid carbon formation, and a decrease of CO2 conversion at
higher CO2/CH4 (N1). Meanwhile, the number of moles of carbon for
CO2/CH4 ratio of 0.5 is lower than that of CO2/CH4 ratio being 1 at
temperatures less than 723 K. It may be mainly due to reaction 10,
which is not plausible when CO2 is the limiting reactant. RWGS
reaction was not fast and active in low temperature and CO2/CH4 b1,
but favorable at temperature N973 and CO2/CH4 ratio N1 attested by
the decreasing trend of H2 production contrary to increasing trend of
CO production within the mentioned ranges. From the foregoing
thermodynamic equilibrium calculations and analysis, we have found
a considerable range of H2/CO ratios regarding CO2 reforming of
methane that can facilitate industrial downstream application of
syngas. An optimal working condition for syngas (with H2/CO in the
order of 1) using FTS can be chosen at a temperature higher than
1173 K for CO2/CH4 ratio being 1 at which about 4 mol of syngas can
be produced from 2 mol of reactants (CO2 + CH4) with a CO2
conversion of more than 98% and only a negligible amount of carbon
formation. Choosing CO2/CH4 ratio b1 is not suggested as the solid
carbon formation is quite high within the entire considered
temperature range (5731473 K). Thermodynamic chemical equilibrium analysis and standard Gibbs free energy calculations reveal that
reactions 3, 4 and 5 are less favorable to ethylene and ethane
formation. Therefore, the yields of C2H4 and C2H6 production may be
increased using a highly active and selective catalyst. CO2/CH4 ratio of
1 and temperature of about 1273 K are suggested for C2H4 production
at which carbon deposition is negligible. Meanwhile, CO2/CH4 ratio
being 2 and temperature 1000 K are appropriate for higher C2H6
production while eliminating solid carbon. Numerical results show

that a temperature around 1000 K and CO2/CH4 ratio of 2 are


suggested for methanol production with considerable carbon elimination. Meanwhile, the temperature about 1173 K and CO2/CH4 ratio
of 1 seem to be suitable for DME production with a negligible amount
of solid carbon. Since the numbers of moles of methanol and DME
formed were very small, an active and selective catalyst is highly
desired to improve the yield of methanol and DME production. Effect
of pressure on CO2 reforming of methane revealed that methane
decomposition and CO disproportionation assisted in lowering CH4
and CO2 conversions, as well as decreasing CO and H2 formation at
higher pressures. At a higher pressure, disproportionation reaction is
the most favorable contributor to carbon formation especially when
the CO2/CH4 ratio N1, but methane decomposition was the main
contributor when CO2/CH4 b1. Increasing O2 contents in CO2
reforming of methane had increased CH4 conversion while the risk
of carbon formation at lower temperatures was reduced. Additionally,
it has a signicant effect on decreasing CO2 conversion, CO and H2
yield but water yield was observed to increase. From thermodynamic
equilibrium analysis, optimal operating temperature of 1073 K and
feed ratio of CO2:CH4:O2 = 1:1:0.1 at atmospheric pressure gave
conversions and syngas yields higher than 90%. At the optimum
condition, H2/CO ratio was maintained in unity, water formation
minimized and interestingly carbon deposition eliminated.
Acknowledgements
The authors express their sincere gratitude to Professor Dr. Jahanmiri
(Shiraz University-Iran), Professor Dr. Zainuddin Manan (Director of
PROSPECT, UTM) and Dr. Sirus Ghotbi (Sharif University-Iran) for their
kind advice during the simulation process. Calculations were performed at
the Simulation Lab, Department of Chemical Engineering in UTM. The
authors also thank the Ministry of Science, Technology and Innovation
(MOSTI) Malaysia and Exxon-Mobil (Malaysia) for sponsoring this
research under Project no: 03-01-06-SF0210 and Vot 68729, respectively.
References
[1] T. Suhartanto, A.P.E. York, A. Hanif, Potential utilisation of Indonesia's Natuna
natural gas eld via methane dry reforming to synthesis gas, Catal. Lett. 71 (2001)
4954.
[2] E.M. Galimov, A.M. Rabbani, Geochemical characteristics and origin of natural gas
in southern Iran, Geochem. Int. 39 (2001) 780792.
[3] B. Eliasson, C.J. Liu, U. Kogelschatz, Direct conversion of methane and carbon
dioxide to higher hydrocarbons using catalytic dielectric-barrier discharges with
zeolites, Ind. Eng. Chem. Res. 39 (2000) 12211227.
[4] T. Oberreuther, C. Wolff, A. Behr, Volumetric plasma chemistry with carbon
dioxide in an atmospheric pressure plasma using a technical scale reactor, IEEE
Trans. Plasma Sci. 31 (2003) 7478.
[5] S.C. Bhumkar, L.L. Lobban, Diffuse reectance infrared and transient studies of
oxidative coupling of methane over Li/MgO catalyst, Ind. Eng. Chem. Res. 31
(1992) 18561864.
[6] K. Moradi, K.C. Depecker, J.J. Barbillat, Diffuse reectance infrared spectroscopy:
an experimental measure and interpretation of the sample volume size involved
in the light scattering process, Spectrochim Acta A 55 (1999) 4364.
[7] A.M. O'Connor, J.R.H. Ross, F.C. Meunier, An in-situ DRIFTS study of the mechanism
of the CO2 reforming of CH4 over a Pt/ZrO2 catalyst, Stud. Surf. Sci. Catal. 119
(1998) 819824.
[8] J.M.C. Comier, I. Rusi, Syngas production via methane steam reforming with
oxygen: plasma reactors versus chemical reactors: the future of technological
plasmas, J. Phys. D Appl. Phys. 34 (2001) 27982803.
[9] T.V. Choudhary, V.R. Choudhary, Energy-efcient syngas production through catalytic
oxy-methane reforming reactions, Angew. Chem. Int. Ed. 47 (2008) 18281847.
[10] W-Ch. Fong, R.F. Wilson, inventor; Texaco Inc., assignee. Gasication process
combined with steam methane reforming to produce syngas suitable for
methanol production, United States Patent US No 5, 496, 859, March 5, 1996.
[11] J. Zhu, D. Zhang, K.D. King, Reforming of CH4 by partial oxidation thermodynamic
and kinetic analyses, Fuel 80 (2001) 899905.
[12] L. Yanbing, J. Baosheng, X. Rui, Carbon dioxide reforming of methane with a free
energy minimization approach, Korean J. Chem. Eng. 24 (2007) 688692.
[13] Z.W. Liu, H.S. Roh, K.W. Jun, Carbon dioxide reforming of methane over Ni/La2O3/
Al2O3, J. Ind. Eng. Chem. 9 (2003) 267274.
[14] Istadi, N.A.S. Amin, Co-generation of C2 hydrocarbons and synthesis gases from
methane and carbon dioxide: a thermodynamic analysis, J. Nat. Gas Chem. 14
(2005) 140150.

M.K. Nikoo, N.A.S. Amin / Fuel Processing Technology 92 (2011) 678691


[15] U. Olsbye, T. Wurzel, L. Mleczko, Kinetic and reaction engineering studies of dry
reforming of methane over a Ni/La/Al2O3 catalyst, Ind. Eng. Chem. Res. 36 (1997)
51805188.
[16] G.A. Olah, G.K. Surya Prakash, inventor; WINSTON & STRAWN LLP., assignee.
Conversion of carbon dioxide to dimethyl ether using bi-reforming of methane or
natural gas, Patent Application Publication US No 2008/0319093A1 Dec 25, 2008.
[17] K. Seshan, J.H. Bitter, J.A. Lecher, Design of stable catalysts for methane carbon
dioxide reforming, Stud. Surf. Sci. Catal. 113 (1998) 187191.
[18] K. Asami, Dry reforming of methane over ceria supported metal catalysts,
Proceeding of the 13th Saudi-Japanese Catalyst Symposium, Dhahran, Saudi
Arabia; Dec, 2003, 2003, pp. 144152.
[19] L. Qian, Z.F. Yang, Study on the reaction mechanism for carbon dioxide reforming of
methane over supported nickel catalysts, Chin. Chem. Lett. 14 (2003) 10811084.
[20] Y. Lwin, Wan W.R. Daud, A.B. Mohamad, Z. Yaakob, Hydrogen production from
steammethanol reforming: thermodynamic analysis, Int. J. Hydrogen Energy 25
(2000) 4753.
[21] S. Jankhah, N. Abatzoglou, F. gitzhofer, Thermal and catalytic dry reforming and
cracking of ethanol for hydrogen and carbon nano-laments' production, Int. J.
Hydrogen Energy 33 (2008) 47694779.
[22] C. Michael, J. Bradford, M.A. Vannice, CO2 reforming of CH4 over supported Pt
catalysts, J. Catal. Today 173 (1998) 157171.
[23] A. Shamsi, Partial oxidation and dry reforming of methane over Ca/Ni/K(Na)
catalysts, Catal. Lett. 109 (2006) 189193.
[24] P. Gronchi, D. Fumagalli, R. Del Rosso, P. Centola, Carbon deposition in methane
reforming with carbon dioxide, J. Therm. Anal. 47 (1996) 227234.
[25] V.R. Choudhary, K.C. Mondal, A.S. Mamman, U.A. Joshi, Carbon free dry reforming
of methane to syngas over NdCoO3 perovskite-type mixed metal oxide catalyst,
Catal. Lett. 100 (2005) 271276.
[26] E.Y. Garcia, M.A. Laborde, Hydrogen production by steam reforming of ethanol:
thermodynamic analysis, Int. J. Hydrogen Energy 16 (1991) 307312.
[27] K. Vasudeva, N. Mitra, P. Umasankar, S.C. Dhingra, Steam reforming of ethanol for
hydrogen production: thermodynamic analysis, Int. J. Hydrogen Energy 21 (1996)
1318.
[28] G. Maggio, S. Freni, S.O. Cavallar, Light alcohols/methane fuelled molten carbonate
fuel cells: a comparative study, J. Power Sources 74 (1998) 1721.
[29] T.A. Semelsberger, R.L. Borup, Thermodynamic equilibrium calculations of
dimethyl ether steam reforming and dimethyl ether hydrolysis, J. Power Sources
152 (2005) 8796.
[30] F. Kajornsak, K. Ryuji, E. Koichi, Thermodynamic analysis of carbon formation
boundary and reforming performance for steam reforming of dimethyl ether, J.
Power Sources 164 (2007) 7379.
[31] X. Wang, M.M. Wang Li, H. Wang, S. Li, S. Wang, X. Ma, Thermodynamic analysis of
glycerol dry reforming for hydrogen and synthesis gas production, Fuel 88 (2009)
21482153.
[32] N.A.S. Amin, T.C. Yaw, Thermodynamic equilibrium analysis of combined carbon
dioxide reforming with partial oxidation of methane to syngas, Int. J. Hydrogen
Energy 32 (2007) 17891798.
[33] K. Faungnawakij, R. Kikuchi, K. Eguchi, Thermodynamic evaluation of methanol
steam reforming for hydrogen production, J. Power Sources 161 (2006) 8794.
[34] S.I. Sandler, Chemical and Engineering Thermodynamics, third ed.John Wiley &
Sons, New York, 1999.
[35] C.L. Yaws, Yaws Handbook of Thermodynamic Properties for Hydrocarbons and
Chemicals, 2nd ed.Gulf Publishing Company, Houston, Texas, 2006.
[36] J.M. Smith, H.C. Van Ness, M.M. Abbort, Introduction to Chemical Engineering
Thermodynamics, 7th ed.Mc Graw-Hill Book Company, New York, 2005.
[37] G.F. Froment, Production of synthesis gas by steam- and CO2-reforming of natural
gas, J. Mol. Catal. A Chem. 163 (2000) 147156.
[38] A. Khalesi, H.R. Arandiyan, M. Parvari, Effects of lanthanum substitution by
strontium and calcium in LaNiAl perovskite oxides in dry reforming of
methane, Chin. J. Catal. 29 (2008) 960968.

691

[39] E.L. Jablonski, I. Schmidhalter, S.R. de Miguel, O.A. Scelza, A.A. Castro, Dry
reforming of methane on Pt/Al2O3alkaline metal catalysts, 2nd Mercosur
Congress on Chemical Engineering, Rio de Janeiro, 2005, pp. 113.
[40] S.-W. Hong, Oh. S-M, D.-W. Park, G.-J. Kim, The methane reforming with carbon
dioxide on Ni-catalyst activated by a DC-pulsed corona discharge, J. Ind. Eng.
Chem. 7 (2001) 410416.
[41] M. Wisniewski, A. Boreave, P. Gelin, Catalytic CO2 reforming of methane over Ir/
Ce0.9Gd0.1O2 x, Catal. Commun. 6 (2005) 596600.
[42] H.-L. Tsai, Ch.-S.h. Wang, Thermodynamic equilibrium prediction for natural gas dry
reforming in thermal plasma reformer, J. Chin. Inst. Chem. Eng. 31 (2008) 891896.
[43] A. Takano, T. Tagawa, S. Goto, Carbon dioxide reforming of methane on supported
nickel catalysts, J. Chem. Eng. Jpn 27 (1994) 727731.
[44] Y. Cai, L. Chou, B. Zhang, J. Zhao, Selective conversion of methane to C2 hydrocarbon
using carbon dioxide over MnSrCO3 catalysts, Catal. Lett. 86 (2003) 191195.
[45] A. Shamsi, Ch.D. Johnson, Effect of pressure on the carbon deposition route in CO2
reforming of CH4, Catal. Today 84 (2003) 1725.
[46] S.Y. Foo, C.K. Cheng, T.H. Nguyen, A.A. Adesina, Oxidative CO2 reforming of
methane on Alumina-supported Co-Ni catalyst, Ind. Eng. Chem. Res. 49 (2010)
1045010458.
[47] J. Rostrup-Neilsen, Production of synthesis gas, Catal. Today 18 (1993) 305324.
[48] L.V. Mattos, E. Rodino, D.E. Resasco, F.B. Passos, F.B. Noronha, Partial oxidation and
CO2 reforming of methane on Pt/Al2O3, Pt/ZrO2, and Pt/CeZrO2 catalysts, Fuel
Process. Technol. 83 (2003) 147161.
[49] J. Wei, E. Iglesia, Isotopic and kinetic assessment of the mechanism of reactions of
CH4 with CO2 or H2O to form synthesis gas and carbon or nickel catalysts, J. Catal.
224 (2004) 370383.
[50] M. Maestri, D.G. Vlachos, A. Beretta, G. Groppi, E. Tronconi, Steam and dry
reforming of methane on Rh: microkinetic analysis and hierarchy of kinetic
models, J. Catal. 259 (2008) 211222.
[51] M.p. Kohn, M.J. Castaldi, R.J. Farrauto, Auto-thermal and dry reforming of landll
gas over a Rh/-Al2O3 monolith catalyst, Appl. Catal. B 94 (2010) 125133.
[52] Istadi, N.A.S. Amin, Optimization of process parameters and catalyst compositions
in carbon dioxide oxidative coupling of methane over CaOMnO/CeO2 catalyst
using response surface methodology, Fuel Process. Technol. 87 (2006) 449459.
[53] Y. Wang, Y. Ohtsuka, CaOZnO catalyst for selective conversion of methane to C2
hydrocarbons using carbon dioxide as the oxidant, J. Catal. 192 (2000) 252255.
[54] Y. He, B. Yang, G. Cheng, On the oxidative coupling of methane with carbon
dioxide over CeO2/ZnO nanocatalysts, Catal. Today 98 (2004) 595600.
[55] N.A.S. Amin, Istadi, Selective conversion of methane to C2 hydrocarbons using
carbon dioxide as an oxidant over CaOMnO/CeO2 catalyst, Stud. Surf. Sci. Catal.
159 (2006) 213216.
[56] B. Eliasson, U. Kogelschatz, B. Xue, L.M. Zhou, Hydrogenation of carbon dioxide to
methanol with a discharge-activated catalyst, Ind. Eng. Chem. Res. 37 (1998)
33503357.
[57] S. Wang, Z.H. Zhu, Catalytic conversion of alkanes to olens by carbon dioxide
oxidative dehydrogenation: a review, Energy Fuels 18 (2004) 11261139.
[58] S.L. Liu, G.X. Xiong, H. Dong, W.S. Yang, Effect of carbon dioxide on the reaction
performance of partial oxidation of methane over a LiLaNiO/-Al2O3 catalyst,
Appl. Catal. A 202 (2004) 141146.
[59] J.F. Mnera, C. Carrara, L.M. Cornaglia, E.A. Lombardo, Combined oxidation and
reforming of methane to produce pure H2 in a membrane reactor, Chem. Eng. J.
161 (2010) 204211.
[60] C.N.A. vila-Neto, S.C. Dantas, F.A. Silva, T.V. Franco, L.L. Romanielo, C.E. Hori, A.J. Assis,
Hydrogen production from methane reforming: thermodynamic assessment and
autothermal reactor design, J. Nat. Gas Sci. Eng. 1 (2009) 205215.

You might also like