Professional Documents
Culture Documents
CATALYST
Y.-J. Wu,1 J.C. Santos,1 Ping Li,2 J.-G. Yu,2 A.F. Cunha1 and A.E. Rodrigues1 *
1. Laboratory of Separation and Reaction Engineering (LSRE), Associated Laboratory LSRE/LCM, Faculty of Engineering, Department of
Chemical Engineering, University of Porto, Rua Dr. Roberto Frias s/n, 4200-465 Porto, Portugal
2. State Key Laboratory of Chemical Engineering, College of Chemical Engineering, East China University of Science and Technology (ECUST),
Shangai 200237, China
Ethanol steam reforming is a promising method for hydrogen production. Kinetic studies are carried out over a nickel-based catalyst from 200
to 600 C. A simplified LangmuirHinshelwoodHougenWatson (LHHW) kinetic model was proposed with nine parameters, where the surface
decomposition of methane is assumed as the rate determining step (RDS), and while all other reaction steps are set as reversible. It is postulated
that the velocity of the backward reaction step of the dissociative methane adsorption CH4 + 2 ) ) CH3 + ) H is similar to all other equilibrium
reactions proposed, while the forward reaction step is set as the RDS. In addition, the formation of adsorbed carbon species is excluded due to the
fact that SRE experiments were performed with excess water, suggesting that OH species, excluding O species formation, stick better on active
Ni surface than methane. In addition, a power law kinetic model was used for analysis; the activation energy was 31.8 kJ/mol, and the reaction
order of ethanol pressure was 1.52. Experimental results were successfully demonstrated by both kinetic models. The purpose of this work was to
develop a simplified mechanistic model to replace empirical models (power rate model) and other proposed kinetic models that are too complex
for an applied kinetic process. It was found that the simplified LHHW kinetic model has a better fit than that of the power law model. The proposed
kinetic model works well over a wide temperature range (200600 C).
Keywords: hydrogen, ethanol steam reforming, kinetic studies, kinetic model
INTRODUCTION
[1,2]
(1)
KINETIC MODELS
[911]
it is
Since research efforts have primarily focused on SRE,
clear that several reaction pathways are possible in SRE (Figure 1),
depending on the catalysts used.[10] The efforts are mainly focused
on Co, Cu, Pt and Ni based catalysts.[912] As typical transition
metals, the active outer layer electrons and associated valence
states make it possible for these metals to be candidates for SRE. It
is worth noting that the conversion of the reactants used, ethanol
and water and the product distribution are strongly dependent
on the nature of catalysts employed (active phase, promoters and
support material). Therefore, it is necessary to carry out a comprehensive kinetic study on SRE to understand the nature of the
reaction.
In this work, SRE reaction is studied in a large range of temperature (200600 C) and space time (0.0086.731 s) in which
the ethanol conversion goes from 0% to 100%. The goal is to
116
The kinetic behaviour of the SRE reaction has been studied over
various catalysts in recent years to describe the reaction rate using
a power law rate expression[11,1322] or an equation based on certain mechanistic assumptions.[13,2329] However, these reports on
kinetics show significant differences. This might be related with
rSRE = k0 exp
EA
RT
pEtOH
pH2 O
(2)
117
118
96
51
85
18.4
r = k pEtOH
r = kN3.64
EtOH
r = kCEtOH a
r = k pEtOH
pH
NiAl-LDH
Y 2 O3
After heat treated at 500 C
Al2 O3
La2 O3
Al2 O3
Support/precursor
Pt (15 wt.%)
RhPt (4 wt.%)
Ni (20.6 wt.%)
Ru (5 wt.%)
Ni (15 wt.%)
Ni (25 wt.%)
CeO2
Monolith (BASF:SR10D)
n/a
Al2 O3
Al2 O3
MgOAl2 O3
Ni (43.6 wt.%)
Ni (16.1 wt.%)
Ni (15.3 wt.%)
Ni
Ni (20.6 wt.%)
Active metal
300450
564700
400590
600700
320520
400600
250300
550650
400
130
T ( C)
1.56
310
4.3
10
4.3
318
5.5
4.3
RH2 O/EtOH
(mol/mol)
RDS
) C2 H5 OH CO + ) CH4 + H2
) C2 H5 OH + ) H2 O ) CH4 + CO2 + 2H2 + )
) CH4 + ) H2 O CO + 3H2 + 2 )
) CH4 + 2 ) H2 O CO2 + 4H2 + 3 )
) C2 H5 OH + ) H2 O 2CO + 4H2 + 2 )
) C2 H5 OH + 3 ) H2 O 2CO + 6H2 + 4 )
) CHO + ) ) CO + ) H
) CHO + ) OH ) CO2 + ) H2
) CHO + ) OH ) CO2 + ) H2
) CO2 + ) H2 ) CO + ) OH
) C2 H5 OH + ) ) CH4 O + ) CH2
C2 H5 OH + ) ) C2 H5 OH
) C2 H5 OH + ) ) CH4 O + ) CH2
) CH3 CH2 O + ) ) CH3 CHO + ) H
) CH3 CHO + ) ) CH4 + ) CO
) HCOO + ) ) CO + ) OH
Reaction scheme
C2 H5 OH CH4 + CO + H2
C2 H5 OH + H2 O CH4 + CO2 + 2H2
CH4 + H2 O CO + 3H2
CH4 + 2H2 O CO2 + 4H2
C2 H5 OH + H2 O 2CO + 4H2
C2 H5 OH + 3H2 O 2CO2 + 6H2
C2 H5 OH CH4 + CO + H2
C2 H5 OH + H2 O CH4 + CO2 + 2H2
CH4 + 2H2 O CO2 + 4H2
CO2 + H2 CO + H2 O
C2 H5 OH + 3H2 O 2CO2 + 6H2
C2 H5 OH + 3H2 O 2CO2 + 6H2
C2 H5 OH + 3H2 O 2CO2 + 6H2
C2 H5 OH + 3H2 O 2CO2 + 6H2
C2 H5 OH CH4 + CO + H2
CO + H2 O CO2 + H2
279
235
124
214
146
113
123
196
174
166
21.0
59.7
4.4
82.7
43.6
71.3
Active energy
(kJ/mol)
Ni (25 wt.%)
Ni
Ni (15 wt.%)
Co (1020 wt.%)
Ni
Active metal
MgOAl2 O3
n/a
Al2 O3
Al2 O3
NiAl-LDH precursor
Support
12
4.3
4.3
38
5.5
600650
400600
400590
320520
400700
3.510
RH2 O/EtOH
(mol/mol)
Mathure et al.[17]
Akpan et al.[14]
Akande et al.[13]
Sahoo et al.[29]
Llera et al.[25]
Mas et al.[16,27]
Refs.
Ciambelli et al.[15]
Simson et al.[19]
Akpan et al.[14]
Akande et al.[13]
Mathure et al.[17]
Mas et al.[16]
Therdthianwong et al.[22]
Sun et al.[20,21]
Refs.
550650
T ( C)
Table 2a. Reported LangmuirHinshelwood and EleyRideal kinetics with rate determining steps (RDS) and activation energies for Ni and Co based catalysts
4.4
r = kN0.43
EtOH
2O
23
0.711 2.71
r = k pEtOH
pH O
149
r = kCEtOH
144
7.0
5.9
16.9
1.9
n/a
EA
(kJ mol1 )
0.8
r = k pEtOH
2.52 7.0
r = k pEtOH
pH O
r = kNEtOH
Expression
119
110
77.5
162
90.9
85.9
418
151
107
C2 H5 OH+ ) ) C2 H5 OH
) CH4 + H2 O CO + 3H2 + )
) CO4 CO2 + )
No RDS
) C2 H5 OH + ) H2 O ) C2 H5 OOH + ) H2
) C2 H5 OH ) C2 H4 O + H2
) CH4 + ) H2 O ) CH4 O + ) H2
) CO + ) H2 O ) CO2 + ) H2
) CHO + ) ) CO + ) H
) CHO + ) OH ) CO2 + H2 + )
) CH3 + ) OH ) CO + 2H2 + )
) CO + ) OH ) CO2 + ) H
C2 H5 OH CH4 + CO + H2
CH4 + H2 O CO + 3H2
CO + H2 O CO2 + H2
C2 H5 OH + H2 O CH4 + CO2 + 2H2
CH4 + H2 O CO + 3H2
CO + H2 CO2 + H2
C2 H5 OH + H2 O 2CO + 4H2
C2 H5 OH CH4 + CO + H2
CH4 + H2 O CO + 3H2
CO + H2 O CO2 + H2
C2 H5 OH CH4 + CO + H2
C2 H5 OH + H2 O CH4 + CO2 + 2H2
CH4 + H2 O CO + 3H2
CO + H2 CO2 + H2
RDS
) CH2 O+ ) ) CHO + ) H
) CHO+ ) O ) CO2 + ) H
) CO+ ) O ) CO2 + )
CH4 +2 ) ) CH3 + ) H
CH4 + 2 ) ) CH3 + ) H
Reaction
CH4 + H2 O CO + 3H2
CO + H2 O CO2 + H2
CH4 + 2H2 O CO2 + 4H2
CH4 + 2H2 O CO2 + 4H2
240.1
67.1
243.9
53.1
86.7
102
Active energy
(kJ/mol)
MgAl2 O4
K/MgAl2 O4
MgO
Ni
Ni
MgAl2 O4
Ni
Active metal
550600
275550
500575
T ( C)
MgAl2 O4 Al2 O3
Rh (1 wt.%)
Support
Al2 O3
CeO2
Support
Ru (2 wt.%)
Ru (5 wt.%)
Rh (5 wt.%)
Active metal
Table 2c. SMR kinetics with rate determining steps (RDS) and activation energies for comparison purposes
558
400
496
110
RDS
Reaction scheme
Active energy
(kJ/mol)
11.25
48
10
48
Graschinsky et al.[24]
Lu et al.[26]
Gorke et al.[23]
Refs.
Rostrup-Nielsen et al.[69]
Xu and Froment[63]
Refs.
RH2 O/EtOH
(mol/mol)
RH2 O/CH4
(mol/mol)
500600
450550
600700
350660
T ( C)
Table 2b. Reported LangmuirHinshelwood and EleyRideal kinetics with rate determining steps (RDS) and activation energies for Ru and Rh based catalysts
0
H298
K
(kJ/mol)
Reactions
Chemical description
Dehydration
Dehydrogenation
Acetaldehyde
decomposition
Direct ethanol
decomposition
Watergas-shift
Latent
watergas-shift
Acetic acid
formation
Acetic acid
decomposition
Steam methane
reforming
Overall steam
methane
reforming
CH3 CH2 OH C2 H4 + H2 O
CH3 CH2 OH CH3 CHO + H2
CH3 CHO CH4 + CO
+45.4
+68.4
18.8
+49.7
CO + H2 O CO2 + H2
CH3 CHO + H2 O CO2 + H2 + CH4
41.1
59.9
26.7
33.3
CH4 + H2 O CO + 3H2
+205.9
+164.7
120
Chemical description
Number
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
ACD
( 9)
(11)
(12)
(13)
(14)
WGS
) + CO ) CO
H2 O + ) ) H2 O
) H2 O + ) ) H + ) OH
) CO + ) OH ) COOH + )
) COOH + ) ) CO2 + ) H
) CO2 ) + CO2
(14)
(15)
(16)
(17)
(18)
(19)
2 ) H 2 ) + H2
) CH3 + ) H 2 ) + ) CH4
) CH3 + ) ) CH2 + ) H
) CH2 + ) ) CH + ) H
) CH + ) ) C + ) H
(10)
(13)
(20)
(21)
(22)
H2 O + ) ) H2 O
) H2 O + ) ) H + ) OH
) C + ) OH ) CHO + )
) CHO + ) ) CO + ) H
(14)
(16)
(23)
(12)
2 ) H 2 ) + H2
) CO ) + CO
(10)
(14)
ETD
ACD
WGS
SMR
Overall SMR
Overall SRE
ETD
SMR
(3)
kSMR pCH4
1 + apH2 O pH12 + bpCO
(4)
Allen et al.[61] also carried out SMR experiments using a commercial nickel catalyst and concluded that the RDS was the
desorption of products. Later, De Deken et al.[62] determined the
SMR kinetics on a Ni/Al2 O3 catalyst. Hydrogen was added to the
feed stream. The kinetic model was composed of two expressions
that accounted for the formation of CO and CO2 .
The proposed SMR kinetics[5860,62,63] were compared by
Elnashaie et al.[64] The most important conclusion was the different dependences for the partial pressure of steam: negative,[5860]
positive De Deken et al.[62] and positive or negative.[63] The conclusion was that the work of Xu and Froment[63] is the most generic
kinetic model. They postulated that carbon intermediates react
with adsorbed oxygen, which should be the RDS.
Another mechanism for SMR was proposed later by RostrupNielsen.[65] Hou and Hughes[66] later developed a similar model
to fit the kinetics on a Ni/-Al2 O3 catalyst with more parameters. Reports in the literature for SMR kinetics indicate that
the key step for a successful hydrogen production must be the
sufficient supply of activation energy to crack the CH bonds
in the methane molecule. It has been also reported that the
theoretical methane activation in SMR may occur at temperatures higher than 300 C.[67] It is well-known that the XH bond
strength in covalent hydrides depends on the electronegativity
and element size (X = H, C, N, O). When increasing the atomic
number, the bond force decreases in a group. Bonds between
hydrogen, carbon, oxygen or nitrogen atoms are the most stable covalent bonds. The dissociation energy of methane is very
high (diss H = 435.3 kJ/mol). Also, the electronic configuration
of methane is 1s2 2s2 2p2 , with two unpaired electrons present
in two 2p orbitals. The transfer of one electron from the 2s to
the 2p orbitals results in four single-filled orbitals (1s2 2s1 2p3 )
for bonding. Thus, the arrangement of the four CH bonds in the
methane molecule is the result of a tetrahedral symmetry, which
minimises electron repulsion.
Wei and Iglesia[68] found that the reaction rates are only proportional to the partial pressure of CH4 . This observation leads
to the conclusion that the dissociation of methane is the RDS
for SMR. The methane dissociation is limited kinetically by the
initial activation of the CH bonds and unaffected by the H2 O
concentration.
The current understanding of methane dissociation[69] leads to
the conclusion that the reaction does not proceed via an adsorbed
precursor state. It has been found that the adsorbed methyl
species may decompose so far that an adsorbed carbon species
is obtained.[70]
The first reaction step is the dissociative adsorption of CH4 :
CH4 + 2 ) ) CH3 + ) H
(5)
The ) CH3 species are subsequently dissociated by many possible surface reactions until chemisorbed carbon atoms are formed
on the metal surface, as shown in Figure 2a.
DFT calculations of these steps on Ni surface, which are reproduced in an energy diagram given in Figure 2b, were performed by
Bengaard et al.[71] They proved that the energies of the CHx species
are higher than those for methane, while the adsorbed methyl
species CH3 show a similar energy to CH2 and CH species.
In addition, the energies of the chemisorbed carbon C* species
are lower than for CHx species and CH4 , with higher activation
energy for decomposition of CHx species to C* then to the hydrogenation of CHx species to methane. The dissociative adsorption
of methane Equation (5) seems to be the RDS.
121
(6)
) H3 COH + ) ) H2 COH + ) H
(7)
) H2 COH + ) ) H2 CO + ) H
(8)
) H2 CO + ) ) CHO + ) H
(9)
and
) CHO + ) ) CO + ) H
(10)
A commercial 15 wt.% Ni/-Al2 O3 catalyst (Octolyst 1001), supplied by Degussa AG in form of extrudates, was used. The
experiments were performed from 200 to 600 C in 100 C intervals
and atmosphere pressure with a water-to-ethanol molar ratio of
10 in the feed. Details on the characterisation of the catalyst and
reaction units can be found in a previous work.[33] In addition,
a Dycor ProLine Process Mass Spectrometer (AMETEK, Berwyn)
was used to measure all the partial pressures of products in real
time. The schematic representation of the experimental units can
be found in the Supplementary Information.
The conversion of the reactants ethanol and water were calculated as follow:
n EtOH,0 n EtOH
100,
n EtOH,0
n H O,0 n H2 O
XH2 O (%) = 2
100
n H2 O,0
XEtOH (%) =
where n EtOH,0 and n H2 O,0 represents the molar flow rate of ethanol
and water in the feed, respectively. n EtOH and n H2 O is the outlet
molar flow rate of unreacted ethanol and water. The partial pressures were obtained from all species (molar distribution) present
in the outlet goes stream:
n
i
yi =
n i
122
pi
ptotal
ptotal =
n
(12)
pi
n i
where n i represents the molar flow rate of one species and
represents the total molar flow rate of species in the outlet stream.
The yield of species i was calculated from:
Proposed Mechanism
Based on findings in the literature, Table 4 depicts the possible
elementary reaction steps based on a LangmuirHinshelwood
HougenWatson (LHHW) model. This model is associated with
Figure 3, where a SRE mechanism with excess water is proposed.
It is clear that the proposed SRE mechanism with excess water is
too complex; therefore, the establishment of a new simple kinetic
model is described in the following section (Table 5), associated
with the corresponding LH expressions and using the HW methodology where the dissociative adsorption of methane (Equation 5)
is set as the RDS.
In the first step (1s), ethanol reacts with a nickel site directly
into adsorbed acetaldehyde and releases molecular hydrogen
simultaneously.
(11)
Yi (%) =
n i n i,0 vEtOH
100
n EtOH,0 vi
(13)
Yi
(14)
XEtOH
mcat 1
VR
=
b V
V
(15)
dXi
d(mcat /n i,0 )
Variable
r=
(16)
Simplified model
(1s)
(2s)
(3s)
(4s)
(RDS)
(5s)
(6s)
(7s)
(8s)
Operative conditions
T ( C)
p (kPa)
mcat (kg)
V inert (N cm3 /min)
(s)
yEtOH,0 (kPa/kPa)
yH2 O,0 (kPa/kPa)
RS/E (mol/mol)
Rinert/EtOH,0 (mol/mol)
200600
100
0.00010.05
0200
0.0086.731
0.0560.0901
0.5600.901
10:1
06.9
123
200 C
300 C
400 C
500 C
600 C
40
20
0
0
[s]
30
24
m exp
(ri ricalc )2
i
=
m
exp
2
(18)
r )
XH O [mol%]
(ri
60
XEtOH [mol%]
where k0 is the pre-exponential factor, EA is the energy of activation, pi is the partial pressure of the reactant ethanol and n in the
superscript is the reaction order with regard to the reactant. These
parameters can be determined with a nonlinear regression of the
experimental results.
In a following step, the establishment of a simplified LHHW
kinetic model for SRE was prepared; details can be found from
the supplementary information.
In order to estimate kinetic parameters, experimental data
points obtained were fitted with the use of the power rate model
and the LHHW model. The sums of squared residuals were used
to obtain the objective function Fobj :
100
80
n
rSRE = k0 e(EA /(RT)) pEtOH
Fobj
18
200 C
300 C
400 C
500 C
600 C
12
0
0
[s]
124
1.2x10-2
6.0x10-1
200 C
300 C
400 C
500 C
600 C
1.0x10-2
4.5x10-1
yH [kPa/kPa]
yAc [kPa/kPa]
8.0x10-3
6.0x10-3
3.0x10-1
4.0x10-3
200 C
300 C
400 C
500 C
600 C
1.5x10-1
2.0x10-3
0.0
0.0
[s]
5.0x10-2
200 C
300 C
400 C
500 C
600 C
200 C
300 C
400 C
500 C
600 C
1.2x10-1
yCO [kPa/kPa]
3.0x10-2
8.0x10-2
2.0x10-2
4.0x10-2
0.0
[s]
1.5x10-1
200 C
300 C
400 C
500 C
600 C
1.2x10
9.0x10
6.0x10-2
3.0x10-2
0.0
1
[s]
-2
[s]
-1
yCH [kPa/kPa]
[s]
0.0
1.6x10-1
1.0x10-2
4.0x10-2
yCO [kPa/kPa]
125
126
a
40
200 C
300 C
400 C
500 C
600 C
rSRE[molkg-1h-1]
30
20
10
0
0
pEtOH [kPa]
40
200 C
300 C
400 C
500 C
600 C
30
rSRE [molkg-1h-1]
Gorke
et al.[23] reported that the RDS steps in the model were
CO2 desorption, dissociative ethanol adsorption and reaction of
adsorbed methane with steam from the gas phase for WGS, ETD
and SMR, respectively. Since a micro reactor has been used, no
heat or mass transfer limitations were detected.
Mathure et al.[17] used acquired experimental data from SRE
to fit them in the power-law kinetic model. A deviation of 10.2%
was calculated, and considerable amounts of coke formation were
observed during the process as well, reducing catalytic activity. Hereafter an EleyRideal mechanism postulated by Akande
et al.[13] was proposed for a better fitting and description of reactions involved in SRE. Four RDS are proposed: adsorption of
ethanol on active sites, dissociative adsorption of ethanol, surface
reaction of the adsorbed oxygenated hydrocarbon fraction with no
adsorbed steam, and surface reaction of the adsorbed hydrocarbon
fraction with non-adsorbed steam.
The values of kinetic parameters and the fitness value of objective function for power rate law model and LHHW model obtained
20
10
0
0
pEtOH [kPa]
Figure 7. Comparison between the mathematical model (lines) and
experimental results (filled spots): power rate model (a) and LHHW
model (b).
from fitting are summarised in Table 7. The coefficients of determination R2 is obtained from:
R2 = 1 Fobj
(19)
Table 7. Calculated results for rSRE with 200 N cm3 /min inert for the kinetic models, pre-exponential factor, energies of activation, reaction
enthalpies, heat of adsorptions, reaction order and coefficients of determination
Model
ri (mol/h)
Relation
k0,j (mol/g h)
Power law
LHHW
rSRE
rSRE
KCO
KOH
KH
KCH
R2
31.8 2.1
1.52 0.11
0.918
31.8 1.6
n/a
Hj (kJ/mol)
Sj,673.15 a (J/mol K) 0.981
1
(7.20 0.58) 10
22.2 1.0
(6.55 0.41) 101
100.9 10.5
(2.33 0.28) 101
244.8 34.4
(5.53 0.38) 100
15.8 1.9
(4.43 0.32) 100
6.7 0.8
(4.82 0.35) 101
20.2 2.6
(1.73 0.24) 101
84.0 14.2
(1.23 0.18) 102
36.8 5.6
Hi (kJ/mol)
Si,673.15 a (J/mol K)
1
(6.55 0.41) 10
100.9 6.0
(5.53 0.38) 100
15.8 1.9
(4.22 0.88) 100
35.4 8.2
(8.65 0.12) 100
42.0 1.7
(1.01 0.21) 102
11.2 2.5
When S is positive and H is negative, a process is always spontaneous. When S is positive and H is positive, a process is spontaneous at high
temperatures. When S is negative and H is negative, a process is spontaneous at low temperatures. When S is negative and H is positive, the reverse
process is spontaneous.
EA (kJ/mol)
CONCLUSIONS
In this work, the kinetic behaviour of SRE over a commercial
nickel-based catalyst has been investigated from 200 to 600 C
with a water-to-ethanol mole ratio of 10 in the feed. The purpose
of this work was to develop a simplified mechanistic model to
replace empirical models (power rate model) and other proposed
kinetic models that are too complex for an applied kinetic process. The following conclusions can be drawn from the analysis
and discussion above.
(1) Both the power rate law and LHHW kinetic models employed
in this work can successfully demonstrate the catalytic SRE
process in a fixed-bed reactor among a large temperature
range.
(2) The energy of activation for SRE reaction was 31.8 kJ/mol,
and the reaction order of the ethanol pressure obtained from
the power rate law model was 1.52, which are in good agreement with other reported results.
(3) According to the comparison of experimental data between
the results from two kinetic models, the LHHW model with
assumption of the surface decomposition of methane as the
RDS has a better fitting than that from the power rate law
model.
(4) The values of the parameters for LHHW kinetic model indicate that the acetaldehyde decomposition reaction is much
faster than the ethanol decomposition reaction, while the
elementary reaction between CH3 and OH is also a strong
endothermic reaction as the SMR.
127
Figure 8. Parity plot for rates of reaction calculated using the power rate model (a) and LHHW-model (b) versus the experimental reaction rates.
NOMENCLATURE
ACD
EA
ETD
EtOH
LHHW
m
mcat
128
n
n i
n i,0
rSRE
r
RS/E
RDS
Si
SMR
acetaldehyde decomposition
activation energy (kJ/mol)
ethanol decomposition
ethanol
LangmuirHinshelwoodHougenWatson
total number of experimental measurements
catalyst mass (kg)
SR
SRE
T
p
pi
VR
WGS
XEtOH
XH2 O
yi
Yi
steam reforming
steam reforming of ethanol
temperature ( C)
total pressure (kPa)
partial pressure for component i (kPa)
reactor volume (m3 )
watergas shift
ethanol conversion (%)
water conversion (%)
mole fraction for component i in the gas phase
(kPa/kPa)
yield of product i (%)
Greek Symbols
b
i
Subscripts
0
t
standard states
total
ACKNOWLEDGEMENTS
Y.J.W. gratefully acknowledges doctoral grant from China Scholarship Council (2010674011).A.F.C. is gratefully to Fundaca o
para a Ciencia e Tecnologia (FCT), Portugal, for providing
financial support for this research program and postdoctoral
grant (SFRH/BPD/62968/2009).The authors also thank for LSRE
financing by FEDER/POCI/2010.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
129
130