Fuel 107 (2013) 254260

Contents lists available at SciVerse ScienceDirect

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

Modeling of xed bed methanation reactor for syngas production: Operating

window and performance characteristics
Naren Rajan Parlikkad a,1, Stphane Chambrey a, Pascal Fongarland a,d,, Nouria Fatah b, Andrei Khodakov a,
Sandra Capela c, Olivier Guerrini c
a

Unit de Catalyse et de Chimie du Solide (UCCS), UMR CNRS 8181, Ecole Centrale de Lille, Cite Scientique, BP 42, 59651 Villeneuve dAscq Cedex, France
Unit de Catalyse et de Chimie du Solide (UCCS), UMR CNRS 8181, Ecole Nationale Suprieure de Chimie de Lille, Ecole Centrale de Lille, Cite Scientique, BP 42, 59651
Villeneuve dAscq Cedex, France
c
GDF-Suez, Direction de la Recherche, 361, avenue du Prsident Wilson, BP 33, 93211 Saint-Denis La Plaine Cedex, France
d
Universit Lyon 1, UMR CNRS 5256, IRCELYON, Institut de recherches sur la catalyse et lenvironnement de Lyon, 2 avenue Albert Einstein, F-69626 Villeurbanne, France
b

h i g h l i g h t s
" Developed phenomenological reactor engineering model for methanation.
" Pseudo-homogeneous and heterogeneous reactor models were developed.
" Simulated xed bed methanator to study the effect of inlet feed conditions.
" Water present in feed results in decrease in catalyst inventory.
" Pore diffusion plays important role in computing actual catalyst inventory.

a r t i c l e

i n f o

Article history:
Received 11 June 2012
Received in revised form 9 January 2013
Accepted 10 January 2013
Available online 31 January 2013
Keywords:
Substitute natural gas
Methanation
Fixed bed reactor
1D pseudo-homogeneous model
1D heterogeneous model

a b s t r a c t
The present work focuses on the development of phenomenological model for the bio-syngas to methane
conversion process. One dimensional heterogeneous and pseudo-homogeneous model were simulated
for a typical pilot plant scale xed bed methanator processing 55 mol/h of CO (total molar ow rate of
310 mol/h) with inlet composition of H2/CO = 3, CO2/CO = 1, CH4/CO = 0.5 at 550 K and 1 atm. Performance of the xed bed reactor at different operating conditions like CO2/CO ratio, H2/CO ratio, effect
of H2O in the feed was studied. It was found that for feeds that were not pre-enriched with hydrogen,
presence of water and water gas shift activity was found to decrease the catalyst inventory substantially.
CO2 in the inlet feed stream would help to decrease the temperature due to dilution effect and more
importantly, can be chosen to maximize methane yield per mole of CO converted. Further, the model
was simulated to predict the performance characteristics of reactor with a mixture containing two types
of catalyst, one of them being specically added to increase H2/CO ratio in feed through water gas shift
reaction. The work also laid the importance of incorporating pore diffusion and external mass transfer
locally in the computation of actual catalyst inventory and reactor volume. The work was useful in selection of operating window and assessing the various viable options for an industrial reactor. The model
developed will serve in selection of operability window for commercialization of substitute natural gas
synthesis (SNG) process.
2013 Elsevier Ltd. All rights reserved.

1. Introduction
The development of novel technologies for production of alternative fuels has gained high interest in recent years, owing to the
frequent uctuations seen in the price of fossil fuels and concerns
Corresponding author.
E-mail address: pascal.fongarland@ircelyon.univ-lyon1.fr (P. Fongarland).
Present address: School of Chemical and Biotechnology, SASTRA University,
Thanjavur 613 401, Tamilnadu, India.
1

0016-2361/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.fuel.2013.01.024

about their uninterrupted supply. One possibility to make of use of

biomass is to convert biomass via a thermo-chemical process into
valuable hydrocarbons. Syngas produced from biomass gasication
can further be converted into higher hydrocarbons through the
FischerTropsch route, or to methane via the methanation route
to produce substitute natural gas (SNG). The main interest to produce bio-SNG is the possibility to use the existing natural gas grid
distribution network to distribute the gas for domestic heating
purposes. More so, bio-SNG research is of specic interest in European countries [13], which meets a large part of its natural gas

N.R. Parlikkad et al. / Fuel 107 (2013) 254260

255

Nomenclature
E
k
k0
Ki
K 0i
Kj
p
P
ri
r 00i
R
T
ug
X
yi
Y
z

activation energy for the chemical reaction, J/mol

intrinsic reaction rate constant, (units as given in Table 1)
pre exponential factor for the intrinsic rate constant,
(unit as given in Table 1)
adsorption coefcient for the component i, bar
pre exponential factor for the adsorption rate coefcient
of component i, bar
equilibrium constant for the reaction j, (in appropriate
units according to reaction)
partial pressure, Pa
pressure, Pa
rate of reaction of component i per unit volume of reactor, mol/m3 s
rate of reaction of component i per unit mass of catalyst,
mol/kgcat s
universal gas constant, J/mol K
temperature, K
supercial gas velocity, m/s
percentage conversion, dimensionless
mole fraction of the component i, dimensionless
percentage yield, dimensionless
axial position, m

requirement through imports. Despite the fact that methanation

reaction is known since several decades, very few studies have focused on the production of SNG from syngas obtained from biomass and on the development of robust engineering model to
simulate the methanation reactor. The processes developed in
the past were mainly concerned with the coal gasication to methane and are very different from biomass gasier units where the
plant are smaller due to the lower amount of the feedstock. This
is why a better optimization is required for bio-SNG production
unit.
CO methanation is a highly exothermic and a reversible reaction, specically at higher temperatures. The reaction is typically
catalyzed by transition metals like Nickel and Ruthenium, with
Nickel being most favored catalyst industrially because its lower
cost [4]. CO methanation is often accompanied by water gas shift
(WGS) and carbon deposition. Elucidation of CO kinetics is difcult
owing to the occurrence of these simultaneous reactions and the
high activity reported on Nickel catalysts leading to possible
mass/heat transfer control even at the laboratory scale. The number of kinetic studies and mechanisms reported in open literature
are testimony for the same [48]. The industrial process design and
development of methanator is complex in nature owing to the issues of exothermicity, mass transfer, equilibrium limitation, occurrence of secondary reactions like water gas shift and the effect of
feed composition. For instance, Kopyscinski [4] through experiments and simulation studies on uidized bed methanation reactor
showed that capturing the characteristic features of fast exothermic methanation reaction was rather difcult and hence remains
a challenge. They observed that most of the reaction was completed within few mm from inlet (roughly in about 10 mm) and
was accompanied by steep increase in reactor temperature.
In this background, the prime focus of the present work was set to
develop a simple yet robust lower order phenomenological reactor
engineering model for the methanator that would essentially capture the fundamental operations of the methanation process and
help to provide guidelines for the industrial operation of methanator. The objective of the work was to study the performance aspects
and key features of a xed bed methanator and subsequently be able
to synergize the model outputs with the biomass to SNG process.

Greek letters
DHi
heat of adsorption of component i, J/mol
g
effectiveness factor to account for pore diffusion with
reaction, dimensionless
gglobal
global effectiveness factor including external mass
transfer, dimensionless
Subscripts
A, B, . . . component index
a, b, . . . stoichiometric coefcients
i
component index (CO, CO2, CH4, H2, H2O)
j
reaction index (methane reforming, WGS)
s
solid phase or catalyst particle
Abbreviations
1D
one dimensional
2D
two dimensional
FBR
xed bed reactor
SNG
substitute or synthetic natural gas
WGS
water gas shift reaction
rWGS
reverse water gas shift reaction

2. Mathematical model
In the present work, xed bed reactor (FBR) for the methanation
process was modeled following one dimensional approach. The
one dimensional model ignores the presence of radial gradients
in the bed. However a-priori calculations do not exclude signicant
external mass transfer, interparticle heat transfer limitation in the
xed bed methanation reactor. For instance, the observed Mears
criterion [9,10] for interphase mass transfer (external mass transfer criteria) was found to be greater than 1 (hence not satised)
for a typical 2.5 mm xed bed catalyst at 550 K. Similarly the interparticle heat transfer criterion was also found to be not satised at
the typical operating conditions (Table 2). Nonetheless, given the
mandate to develop reactor engineering model with an industrial
perspective and the laboratory constraints to measure and obtain
reliable local concentration, temperature data (2D data); it was
thought judicious to proceed initially with one dimensional approach and improvise the same in later stages of the project cycle.
Both 1D pseudo-homogeneous and 1D heterogeneous framework
were adopted to simulate the performance of the FBR. In the heterogeneous model, the effectiveness factor for the catalyst was computed locally along the reactor and coupled with the mass and
energy balance equation for the gas phase. In contrast, in the pseudo-homogenous model, the effectiveness factor was assumed at a
xed value (a user input) and was invariant within the model simulation. The pressure drop in the reactor was obtained from Ergun
relation [11]. Detailed description of the 1D heterogeneous and 1D
pseudo-homogeneous model can be found from Froment and Bischoff [12] and is not presented here for sake of brevity. The nal
conservation equations are provided in the Supplementary
material.

2.1. Methanation kinetics

In this work, the kinetic expression used for the CO methanation was based on Xu and Froment [6]. Experiments were conducted in a laboratory catalytic test facility equipped with a
micro-Berty reactor (Autoclave Engineers) and using a proprietary

256

N.R. Parlikkad et al. / Fuel 107 (2013) 254260

Table 1
Kinetic and adsorption constants for the methanation and WGS rate expressions.
Quantity
0
kSMR
K 0WGS
K 0CO
K 0H2

Value
0.5

49,216,232,299,804,600 (mol bar

(Fitted for UCCS data)

K CH4
K H2 O

/g h)

Quantity

Value

ESMR

240 (kJ/mol)

1.955E6 (mol/g h bar)

Ewgs

8.23E5 (bar)

DHCO

70.65 (kJ/mol)

67.13E3 (kJ/mol)

6.12E9 (bar)

DHH2

82.90 (kJ/mol)

6.65E4 (bar)
1.77E5 (bar)

DHCH4
DHH2 O

38.28 (kJ/mol)
88.68 (kJ/mol)

Table 2
Operating parameters and inlet conditions for the base case: Fixed bed reactor.
Reactor diameter (D)
Reactor height (H)
Particle size (dp)
Particle density (pp)
Effective thermal conductivity of catalyst
Effective molecular diffusivity (Deff)
Gas properties

Inlet pressure (P)

Inlet temperature (T)
Inlet supercial gas velocity (ug) at reactor
temperature and pressure
Inlet composition in terms of mol fraction (Ratio
with CO in parenthesis)

0.10 m
Adjusted variable
0.005 m
1700 kg/m3
0.2 W/m K
1E5 m2/s
Gas properties like viscosity (l), thermal conductivity (k) and specic heat capacity (Cp) are computed as function of
temperature. Molar mixture property are computed from individual component values by applying mol fraction
weighted mixture rule
101,325 Pa
550 K
0.5 m/s
yCO: 0.18 (1)
yCO2 : 0:18 (1)
yH2 : 0:54 (3)
yCH4 : 0:09 (0.5)
yN2 : 0 (0)
yH2 O : 0:01 (0, but altered to make Ry = 1, therefore 0.06)

Reaction accounted
Methanation
Water gas shift
Heat transfer condition

Industrial catalyst rate model based on Xu and Froment [6] (Eq. (1))
Xu and Froment [6] (Eq. (2))
Adiabatic reactor (No external heat transfer)

Ni based industrial catalyst [13]. More details on the experimental

work can be had from Naren et al. [13] or upon request from corresponding author. The kinetic parameter of Xu and Froment [6]
were then tuned to t the UCCS experimental data, as the original
parameter underestimated the observed rate. The nal rate expressions for the in-house kinetic models (called hereafter as industrial
catalyst rate model) are as follows
Reactions

CO methanation :

CO 3H2
CH4 H2 O

Industrial catalyst rate model

pCO p3H

K SMR

 pCH4 pH2 O

kSMR


p2:5
H2
1 K CO pCO K H2 pH2 K CH4 pCH4 K H2 O pH2 O p1
H2

The water gas shift kinetics (Eq. (2)) was taken from Xu and Froment [6]. The adsorption coefcients (Ki, i = CO, H2, CH4, H2O) for
the industrial catalyst rate model were taken from Xu and Froment
[6] and are tabulated in Table 1. KSMR and KWGS denote the equilibrium constant for the steam methane reforming and water gas shift
reaction.


pCO p 
2 H2
pCO pH2 O  K WGS
kWGS
r
2
pH2 1 K CO pCO K H2 pH K CH4 pCH K H2 O pH O p1
H2
2
4
2
where



Ei
;
RT

K i K 0i exp



DH
;
RT

i SMR; WGS

i CO;CH4 ; H2 ; H2 O

3. Results and discussion

The xed bed reactor model was simulated for a typical pilot
plant xed bed reactor of 10 cm diameter. The inlet conditions
and operating parameters for the base case are tabulated in Table 2.
The intrinsic kinetic rates were computed using Eq. (1) for CO
methanation and Eq. (2) for water gas shift reaction.

WGS reaction : CO H2 O
H2 CO2


ki ki exp

3.1. Signicance of transport limitations: heterogeneous model

predictions
In order to elucidate the signicance of external mass transfer
and internal pore diffusion resistance in catalyst, 1D heterogeneous model was simulated where in the global effectiveness factor was computed locally along the length of the reactor. The global
effectiveness factor denotes the combined resistance due to
intraparticle pore diffusion and external interphase mass transfer
resistance.
The axial conversion prole and temperature prole in the reactor predicted from heterogeneous model approach is shown in
Figs. 1a and 1b. The simulated proles from 1D pseudo homogeneous model at different xed values of effectiveness factor are

257

N.R. Parlikkad et al. / Fuel 107 (2013) 254260

35

1000

6000

Pseudo homogeneous: eta = 1

Pseudo homogeneous: eta = 0.1

900

5000

Pseudo homogeneous: eta = 0.01

Heterogeneous model

30
25
20
15
10

800
4000
700
3000

Methanation

WGS

600
500

Inlet conditions
P: 1 atm FCO: ~2 mol/m2s ug: 0.5 m/s
Inlet mol composition:
yCO: 0.18 yCO2:0.18 yH2: 0.54
yCH4: 0.09 yH2O: 0.01

2000
1000

400
300

0
200
-1000

100

-2000
0.0001

0
0.0001

0.001

0.01

0.1

0.001

0.01

0.1

Axial location, (m)

10

Axial location, (m)

Temperature, (K)

Percentage conversion of CO, (%)

40

Inlet conditions
2
P: 1atm T: 550 K FCO : ~2 mol/m s
u g: 0.5 m/s
Inlet composition: (mol fraction)
y CO : 0.18 yCO2 : 0.18 yH2: 0.54
yCH4 : 0.09 y H2O : 0.01

Reaction rate, mol/m3 s

45

Fig. 1c. Axial prole of volumetric reaction rate in an adiabatic xed bed reactor
from 1D pseudo-homogeneous model.

Mass of catalyst at equilibrium conversion, (kg)

Fig. 1a. Percentage conversion of CO along the length of the FBR predicted from 1D
pseudo homogeneous and heterogeneous model approaches.

1150
Inlet conditions

Pseudo homogeneous: eta = 1

2

Temperature, (K)

1050

P: 1atm T: 550 K F CO: ~2 mol/m s

ug: 0.5 m/s
Inlet composition: (mol fraction)
yCO: 0.18 yCO2: 0.18 yH2: 0.54
yCH4: 0.09 yH2O: 0.01

Pseudo homogeneous: eta = 0.1

Pseudo homogeneous: eta = 0.01
Heterogeneous model

950

850

750

650

100
With diffusion and external mass tranfer limitation

10

0.1

No diffusion and
external mass tranfer
limitation

0.01
1

550
0.0001

0.1

0.01

Varies along length

Global effectiveness factor

0.001

0.01

0.1

Axial location, (m)

Fig. 1b. Axial prole of temperature in the FBR predicted from 1D pseudo
homogeneous and heterogeneous model approaches.

also shown in Figs. 1a and 1b. At the base case conditions, the exit
conversion from FBR or the nal conversion (at reactor exit) upon
reaching equilibrium is roughly 10% in an adiabatic xed bed reactor. A steep increase in CO conversion to 28% was noted initially at
a distance of about 1.8 mm from inlet and the reactor temperature
increased correspondingly to 883 K. Thermodynamic reaction
equilibrium sets in and reverses the reaction and results in net production of CO against CO consumption. To further elucidate the
same, the axial variation in volumetric rate of methanation and
WGS in an adiabatic FBR from 1D pseudo-homogenous model for
a typical case with effectiveness factor value of 1, is shown in
Fig. 1c. It was evident that with increase in temperature, reaction
rate for methanation increases, however, higher temperatures
would result in reverse water gas shift reaction. The rWGS reaction
is endothermic and would result in decrease in reactor temperature. The coupled effect eventually lead to decrease in CO consumption through methanation, a net CO production and thus CO
conversion decreased from 28% to 10%. Under this circumstance,
the temperature drops from 883 K to 869 K. Thus, CO conversion
was found to be limited by thermodynamic equilibrium in an adiabatic reactor and was as expected.
1D heterogeneous model simulations showed that the global
effectiveness factor was much greater than one at the entrance of

Fig. 1d. Catalyst inventory required for different conditions of adiabatic FBR.

the reactor. However, with the increase in temperature, severe

external mass transfer and pore diffusion limitation was found
and as a consequence the global effectiveness factor drops rapidly
to low value (about 1.7 at 550 K and z = 0 to say 0.2 at 615 K at
z = 2 mm). This steep drop in the global effectiveness factor in
the reactor retards the system from attaining the equilibrium
and results in larger reactor volume. For instance, if there was no
diffusional and mass transfer resistance (g = 1), the system attained the equilibrium conversion at about z = 2 mm, however,
with transport resistance, this takes more than 1 m to attain the
equilibrium point (Fig. 1a). Though the overall key features and
characteristics of the reactor were kept intact, the actual reactor
size or the catalyst inventory needed was increased by a factor of
about 3 with transport resistances (Fig. 1d).
Thus, it would be advisable to investigate the relative merits of
various process options from 1D pseudo homogenous model
neglecting diffusional and external mass transfer. However, actual
reactor sizing would be very much sensitive to the value of local
effectiveness factor in the reactor. Reactor sizing should be obtained only after incorporating the effect of external mass transfer
and pore diffusion limitations. Attempts to specify effectiveness
factor at a xed value other than 1, (say for instance gglobal = 0.1,
0.01) were also found not satisfactory to predict the actual reactor
size for the adiabatic xed bed reactor (Fig. 1d). Hence, in this
work, the performance of the reactor was studied with the focus
on percentage conversion and relative catalyst inventory and not

N.R. Parlikkad et al. / Fuel 107 (2013) 254260

3.2. Effect of CO2 dilution

In most cases, the inlet stream into the Methanator would contain CO2. Also the amount of CO2 would increase within the reactor
due to recycle. From kinetic studies reported, CO2 methanation is
inhibited in the presence of CO in the Nickel catalyst and thus often
not accounted [1416]. Given this circumstance, it was found
worthwhile to assess the effect of CO2 in the performance of the
methanator. Moreover, diluting of the reacting CO and H2 gaseous
mixture with CO2 would also help to control the temperature rise
in the reactor. The 1D pseudo homogeneous model was simulated
to study the effect of CO2 dilution in the inlet gas stream of the FBR.
The importance of inclusion of transport resistances through global
effectiveness factor was already discussed in the foregoing section.
In light of the observations aforementioned previously, the present
simulations were done to assess the relative merits of dilution of
inlet feed gas with CO2 and not focused on reactor sizing. The inlet
gas composition is given in Table 3. The inlet supercial gas velocity was adjusted to keep the molar feed ow rate of CO constant in
all the cases corresponding to the base case value. Other input
parameters are set as per Table 2 and the methanation kinetics given by industrial catalyst rate model (Eq. (1)).
The percentage conversion of CO along the length of the reactor
for different inlet CO2/CO ratios is shown in Fig. 2a. The equilibrium
temperature (exit temperature) attained in the reactor at various
CO2 inlet ratios is shown in Fig. 2b. It was observed that with the
increase in the CO2 ratio at inlet at constant molar feed ow rate
of CO, the reactor temperature decreased monotonically from
930 K for feed with no CO2 to 630 K for feed with CO2/CO ratio
of 35.5.
Another distinguishable characteristic was the monotonic increase in the active reactive volume of the reactor with the increase in the CO2 feed ratio. The active reactive volume would
signify the point up to which the reaction takes place and thereafter equilibrium is said to prevail in the system. The equilibrium
limitation or the quench point was found to move across the reactor length with the increase in the CO2 concentration at the inlet.
The active reactive volume increased from near about 1 mm to
7 cm with increase in CO2/CO ratio from 0 to 35.5. Thus increase
in CO2 in the feed mixture could eventually result in larger reactors
or increased catalyst inventory. However, it should be worthy to

70
60

Percentage conversion of CO, (%)

on the exact catalyst inventory or reactor volume. Any attempt to

study and report catalyst inventory and/or reactor volume without
considering the local variation in effectiveness factor is incomplete.
Thus, further performance studies, like effect of inlet feed composition, are discussed based on the simulated results from 1D pseudo-homogeneous model, with focus on relative performance
criteria and not based on their absolute values.

50
40
30

0.5

2.2

5.5

8.8

15.5

20.5

28.8

35.5
CO2 /CO: 35.5

Legend: CO 2 /CO ratio at inlet

Operating conditions:
FCO : 2 mol/m 2 s T: 550 K P: 1 atm
Inlet composition: Mol ratio with CO
yCO2/CO : 0 - 35.5 y H2/CO : 3
yCH4/CO: 0.5 yH2O/CO : ~0
ug: 0.41 - 3.6 m/s

28.8
0
20.5

15.5

20

0.5

10
0

0.00001

0.0001

0.001

0.01

0.1

Axial location, (m)

Fig. 2a. Percentage conversion of CO along the length of FBR at different CO2 to CO
ratios at the inlet.

1000
950

Exit Temperature, (K)

258

900
850
800
750
700
650
600
550
0

0.5

2.16667 5.5

8.83333 15.5

20.5 28.8333 35.5

Inlet CO 2 /CO mol ratio, (-)

Fig. 2b. Exit temperature from the reactor operated at different inlet CO2 to CO mol
ratios.

reinstate here that the actual reactor volume should include the effect of interphase transport and pore diffusional resistance. The
numbers quoted above just signify the relative contribution of
CO2 ratio on reactor size or catalyst inventory and not the true
value.
It is important to look into the effect of CO2 dilution on percentage conversion of CO at exit (Fig. 2c). A local minimum in the CO

Table 3
Input operating conditions for the CO2 dilution study.
Inlet supercial gas velocity (m/s)
0.409091

0.45

0.5 (Base case)

0.6

0.9

1.2

1.8

2.25

3.6

Inlet mol fraction ()

CO
0.22
CO2
0
H2
0.66
CH4
0.11
H2O
0.01

0.2
0.1
0.6
0.1
0

0.18
0.18
0.54
0.09
0.01

0.15
0.325
0.45
0.075
0

0.1
0.55
0.3
0.05
0

0.075
0.6625
0.225
0.0375
0

0.05
0.775
0.15
0.025
0

0.04
0.82
0.12
0.02
0

0.03
0.865
0.09
0.015
0

0.025
0.8875
0.075
0.0125
0

Mol ratio with respect to CO ()

CO/CO
1
CO2/CO
0
H2/CO
3
CH4/CO
0.5
H2O/CO
0.045455

1
0.5
3
0.5
0

1
1
3
0.5
0.056

1
2.17
3
0.5
0

1
5.5
3
0.5
0

1
8.83
3
0.5
0

1
15.5
3
0.5
0

1
20.5
3
0.5
0

1
28.83
3
0.5
0

1
35.5
3
0.5
0

259

50

1000
CO conversion

45
Yield to conversion ratio

Y CH4 /x CO at exit, (-)

40
100
35
30
25

10

20
15
1
10
5
0

0.1
0.00

0.50

1.00

2.17

5.50

8.83 15.50 20.50 28.83 35.50

Percentage conversion of CO at exit, (% )

N.R. Parlikkad et al. / Fuel 107 (2013) 254260

Inlet CO2/CO mol ratio, (-)

Fig. 2c. Consolidated plot of percentage CO conversion and methane yield to CO
conversion ratio at exit for FBR operated at different inlet CO2/CO mol ratios.

conversion was observed with the CO2 dilution. This would suggest
that for a given FBR system, there is a certain CO2 dilution ratio at
which the reactor performance would be not so good in terms of
CO conversion. At the condition corresponding to lowest CO conversion in the reactor, the yield of methane per mol of CO converted Y CH4 =xCO was found to be very high (about 120, see
Fig. 2c) suggesting that at a particular CO2 dilution ratio (in this instance CO2/CO = 2.17), very high yield of methane is obtained per
mol of CO consumed in the reactor. The stoichiometric limit for
the yield of methane per mol of CO consumed Y CH4 =xCO is 1, in
line with the reaction stoichiometry for CO methanation. The value
of Y CH4 =xCO more than 1 denotes that more methane is produced
than expected from reaction stoichiometry. The value of Y CH4 =xCO
less than 1 denotes that CO in the inlet feed stream undergoes side
reactions (for instance here water gas shift) other than the primary
reaction of CO methanation, thus resulting in the decrease of methane produced per mol of CO consumed. The presence of CO2 resulted in the net production of CO towards the exit of the reactor
by reverse water gas shift (rWGS) reaction and thus eventually
increasing the yield of CH4 produced to the net CO consumed in
the reactor. Thus presence of CO2 in the inlet feed was found to
have benecial effect of lowering of reactor temperature and also
increasing the yield of methane per mol of CO consumed in the
reactor. Alternatively, when no CO2 was present in the inlet feed
stream, the yield of methane per mol of CO converted dropped below the stoichiometric limit of 1, signifying that CO was also consumed in WGS reaction in addition to the methanation reaction.

3.3. Importance of syngas feed composition

It was thought important to assess effect of syngas feed composition on reactor performance and CO conversion. From equilib-

rium computations at base case conditions, it was observed that

the maximum attainable CO conversion for the base case feed
was >90% at temperature less than 650 K. To assess the effect of inlet composition, an isothermal FBR without pore diffusion and
external mass transfer was simulated for different choices of inlet
feed composition varying from H2/CO = 1 to 3. Also in few cases,
H2O was also included in the inlet feed stream to capture the effect
of water gas shift reaction and its effects on the reactor performance. Table 4 summarizes the key output performance characteristics of the isothermal FBR operated with different inlet feed
compositions. All other input parameters for the model were kept
as given in Table 2.
It was clearly observed that decrease in the stoichiometric ratio
of H2 in the inlet feed stream would result in the obvious decrease
in the methane yield. However, the maximum attainable CO conversion was not appreciable affected by the H2/CO mol ratio at inlet. This was attributed to the consumption of CO in the water gas
shift reaction resulting in the production of CO2 and H2. Another
striking feature observed from Table 4 is that for feeds that were
not enriched with stoichiometric H2/CO ratio of 3, presence of
water in the inlet feed stream would resulted in substantial decrease in the reactor size or catalyst inventory to achieve the same
CO conversion or methane yield. This was due to the fact that presence of water in the inlet stream would initiate the water gas shift
sooner than the case wherein water was not present in the inlet
stream and was produced only as byproduct of CO methanation.
3.4. Simulation of FBR with mixed catalysts
It is known that the syngas obtained from biomass is not enriched with H2 and hence there is a need to look out for means
to increase the H2/CO ratio from 1.5 to about 3 for the CO methanation. Typical industrial processes employ WGS shift reactor before the methanator and increase the H2/CO in the feed to the
methanator. It was thought to use mixed catalysts in the same
FBR reactor with the aim to produce H2 in situ in the methanator.
Accordingly, simulations were done with two catalysts mixed
intermittently in the xed bed methanator. One catalyst was the
proprietary Ni based industrial catalyst (Cat I) and the other was
a typical WGS shift catalyst based on Iron (Cat II). Both CO methanation and WGS was assumed to occur on the sites of the industrial
Ni catalyst whereas only WGS reaction was assigned for the Fe
based WGS catalyst (Cat II). The WGS kinetics for Cat II was taken
from Wang et al. [17]. The FBR was simulated to see the effect of
mixed catalyst on the performance of a lean H2/CO feed. The feed
composition consists of yCO: 0.3, yH2 : 0:3 and yH2 O : 0:4. The inlet
molar ow rate of CO was kept constant at 55 mol/h as per base
case. The objective was to see whether in situ hydrogen generation
will help to improve the CH4 yield and or decrease catalyst inventory in the reactor. All other parameters were kept as given in
Table 1.
It was seen that the in situ hydrogen generation was not found
to substantially change the CO conversion and/or methane yield in

Table 4
Effect of inlet feed composition on reactor performance in an isothermal FBR at 550 K.
Inlet feed composition in terms of mol ratio with CO
CO

CO2

H2

CH4

H2O

1
1
1
1
1
1
1

1
0
0
1
0
1
1

3
3
1
1
1
1
2

0.5
0
0
1
1
1
0.5

0.056
0
0
0
1
1
0.5

Max attainable %CO conversion

Yield of CH4 (%)

Relative catalyst inventory

99.90
99.99
99.47
97.29
99.94
99.86
99.89

98.5
98
49.8
49.22
48.54
48.6
73.7

1
1
25
30
8
8
4

260

N.R. Parlikkad et al. / Fuel 107 (2013) 254260

sion ratio rather than on actual reactor size or catalyst inventory.

Dilution of inlet feed stream with carbon-dioxide was found to increase the yield of methane. Presence of water in the inlet feed
stream was found to be benecial, especially for feeds that were
not pre-enriched with stoichiometric H2/CO ratio. Operating conditions selected based on one dimensional model could further be
tested with a more rigorous two dimensional models for their
accurate performance characteristics like local hot spots in the
reactor.

Percentage conversion of CO, (%)

120
100
0.75

80
60
1

40
Legend: Volume fraction of catalyst I
in the FBR
Operating condition
FCO: 56 mol/hr T: 550 P: 1 atm
Feed composition: (mol fraction)
yCO: 0.3 yH2 : 0.3 yH2O: 0.4

20

0.25

0.5

0.75

Acknowledgements

0
0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Axial location, (m)

Fig. 3. Percentage conversion of CO along the length of isothermal FBR operated
with different volume fraction of catalyst I (Industrial Ni catalyst).

the reactor. For instance, in an adiabatic reactor, it was found that

the methane yield in the reactor increased very marginally from
8.24 to 8.3 with inclusion of 75% volume fraction of WGS cat
II in the reactor. Thus, the in situ hydrogen generation was not
found be very attractive for the said feed condition in a mixed catalyst FBR system. The same was observed in case of the isothermal
reactor (Fig. 3). However, the promising feature of incorporation of
mixed catalysts was the net reduction in the catalyst inventory
and/or reactor size. As seen from Fig. 3, even with about 25% of
WGS cat II in the reactor, there was substantial reduction in the relative catalyst inventory. For instance, for the given feed condition,
the catalyst inventory (or the reactor size/length) reduces by about
1/5th in comparison that of the case with no mixed catalyst. Thus
having specic catalysts for WGS reaction in the same reactor as of
methanator was found to be benecial in terms of reducing the
catalyst inventory to achieve the same conversion. However, care
should be noted that such an observation was for lean feed and
hence, dependent on the feed composition. Work should be extended to generalize this observation for other feed conditions.
4. Conclusions
The present work was a purposeful and focused attempt to
showcase the ability of lower order phenomenological reactor
engineering models in predicting and analyzing the performance
of the methanation reactor under various operating conditions.
The work brought out a simple and robust model to judiciously
analyze and decide the operating conditions of a xed bed methanator. Incorporation of the local effectiveness factor was found to
be critical in actual reactor sizing. Hence, cautious attempt was
made to discuss and highlight the outcome of the present simulations based on extensive parameters like methane yield to conver-

The authors thank Agence Nationale de la Recherche (ANR for

the grant of program ANR BIOE 2009 Vegaz under which this
work was carried out). One of the authors, N.R.P. thanks UCCS,
Ecole Centrale de Lille for grant of contract research fellowship.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.fuel.2013.01.024.
References
[1] Mozaffarian M, Zwart RWR. Feasibility of biomass/waste-related SNG
production technologies. Energy research Centre of the Netherlands, ECN-C03-066; 2003.
[2] Zwart RWR, Boerrigter H, Deurwaarder EP, van der Meijden CM, van Paasen
SVB. Production of synthetic natural gas (SNG) from biomass. Energy research
Centre of the Netherlands, ECN-E-06-018; 2006.
[3] http://www.biosng.com/home/: 05 June 2012.
[4] Kopyscinski J. Production of synthetic natural gas in a uidized bed reactor.
PhD dissertation, ETH Zurich; 2010.
[5] Yadav R, Rinker RG. Steady state methanation kinetics over a Ni/Al2O3 catalyst.
Can J Chem Eng 1993;71:2028.
[6] Xu J, Froment GF. Methane steam reforming, methanation and water gas shift:
I intrinsic kinetics. AIChE J 1989;35:8896.
[7] Klose J, Baerns M. Kinetics of the methanation of carbon monoxide on an
alumina supported nickel catalyst. J Catal 1984;85:10516.
[8] Schoubye P. Methanation of CO on some Ni catalysts. J Catal 1969;14:23846.
[9] Mears DE. Test for transport limitations in experimental catalytic reactors. Ind
Eng Chem Process Des Dev 1971;10:5417.
[10] Mears DE. Diagnostic criteria for heat transport limitations in xed bed
reactors. J Catal 1971;20:12731.
[11] Ergun S. Fluid ow through packed columns. Chem Eng Prog 1952;48:8991.
[12] Froment GF, Bischoff KB. Chemical reactor analysis and design. New York: John
Wiley and Sons; 1979.
[13] Naren PR, Fongarland P, Fatah N, Khodakov A, Capela S, Kara Y, et al.
Methanantion of syngas: Reactor choices and operating window. CAMURE-8,
Naantali, Finland; 2011.
[14] van Herwijnen T, Van Doesburg H, De Jong WA. Kinetics of the methanation of
CO and CO2 on a nickel catalyst. J Catal 1973;28:391402.
[15] van Doesburg H, De Jong WA. Transient behaviour of an adiabatic xed bed
methanator II methanation of mixtures of carbon monoxide and carbon
dioxide. Chem Eng Sci 1976;31:538.
[16] Yaccato K, Carhart R, Hagemeyer A, Lesik A, Strasser P, Volpe Jr AF, et al.
Competitive CO and CO2 methanation over supported noble metal catalysts in
high throughput scanning mass spectrometer. Appl Catal A 2005;296:3048.
[17] Wang YN, Xu YY, Xiang HW, Li YW, Zhang BJ. Modeling of catalyst pellets for
Fischer Tropsch synthesis. Ind Eng Chem Res 2001;40:432435.