Experimental Investigation on Upgrading of Lignin-Derived Bio-Oils: Kinetic Analysis of Anisole Conversion on Sulfided CoMo/Al2O3 Catalyst

Article abstract of DOI:10.1002/kin.21026

Kinetics of the hydroprocessing of anisole, a compound representative of lignin-derived bio-oils, catalyzed by a commercial sulfided CoMo/Al₂O₃, was determined at 8–20 bar pressure and 573–673 K with a once-through flow reactor. The catalyst was sulfided in an atmosphere of H₂ + H₂S prior to the measurement of its performance. Selectivity-conversion data were used as a basis for determining an approximate, partially quantified reaction network showing that hydrodeoxygenation (HDO), hydrogenolysis, and alkylation reactions take place simultaneously. The data indicate that these reactions can be stopped at the point where HDO is virtually completed and hydrogenation reactions (and thus H₂ consumption) are minimized. Phenol was the major product of the reactions, with direct deoxygenation of anisole to give benzene being kinetically almost insignificant under our conditions. We infer that the scission of the C_methyl–O bond is more facile than the scission of the C_aromatic–O bond, so that the HDO of anisole likely proceeds substantially through the reactive intermediate phenol to give transalkylation products such as 2-methylphenol. The data determine rates of formation of the major primary products. The data show that if oxygen removal is the main processing goal, higher temperatures and lower pressures are favored.

Full text of DOI:10.1002/kin.21026

Experimental Investigation
on Upgrading of
Lignin-Derived Bio-Oils:
Kinetic Analysis of Anisole
Conversion on Sulfided
CoMo/Al O Catalyst
2
3
1
2
1
3
HAMID REZA RAHIMPOUR, MAJID SAIDI, PARISA ROSTAMI, BRUCE C. GATES,
MOHAMMAD REZA RAHIMPOUR1
,3
1Department, of Chemical Engineering, Shiraz University, Shiraz, 71345, Iran
2Faculty of Engineering, Shahrekord University, Shahrekord, Iran
3Department of Chemical Engineering, University of California, Davis, CA 95616
Received 9 October 2015; revised 22 June 2016; accepted 22 June 2016
DOI 10.1002/kin.21026
Published online in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: Kinetics of the hydroprocessing of anisole, a compound representative of lignin-
derived bio-oils, catalyzed by a commercial sulfided CoMo/Al2O3, was determined at 8–20
bar pressure and 573–673 K with a once-through flow reactor. The catalyst was sulfided in an
atmosphere of H2 + H2S prior to the measurement of its performance. Selectivity-conversion
data were used as a basis for determining an approximate, partially quantified reaction network
showing that hydrodeoxygenation (HDO), hydrogenolysis, and alkylation reactions take place
simultaneously. The data indicate that these reactions can be stopped at the point where HDO
is virtually completed and hydrogenation reactions (and thus H2 consumption) are minimized.
Phenol was the major product of the reactions, with direct deoxygenation of anisole to give
benzene being kinetically almost insignificant under our conditions. We infer that the scission
of the Cmethyl–O bond is more facile than the scission of the Caromatic–O bond, so that the
HDO of anisole likely proceeds substantially through the reactive intermediate phenol to give
Correspondence to: M. R. Rahimpour; e-mail: rahimpor@
shirazu.ac.ir, mrahimpour@ucdavis.edu. M. Saidi; e-mail: m.saidi@
eng.sku.ac.ir.
Supporting Information is available in the online issue at
www.wileyonlinelibrary.com.
ꢀC
2016 Wiley Periodicals, Inc.

2
RAHIMPOUR ET AL.
transalkylation products such as 2-methylphenol. The data determine rates of formation of the
major primary products. The data show that if oxygen removal is the main processing goal,
higher temperatures and lower pressures are favored. ꢀC 2016 Wiley Periodicals, Inc. Int J Chem
Kinet 1–12, 2016
INTRODUCTION
compound such as anisole) and the other involving a
hydrogenation–hydrogenolysis route whereby the aro-
matic ring is hydrogenated prior to the rupture of the
carbon–oxygen bond [15, 17–23].
The environmental concerns related to the burning of
fossil fuels motivate interest in the development of
renewable raw materials as feedstocks for the produc-
tion of fuels and even chemicals. The depolymerization
of biomass and the upgrading the resulting fragments
into compounds with properties that approximate those
of petroleum represent substantial challenges [1–5].
The compositions of liquids formed from biomass de-
pend on the raw materials and the methods of con-
version [4,6–11]. Generally, these liquids include high
concentrations of oxygen-containing compounds, and
improvements in the stability of these liquids and their
compatibility with petroleum products require a reduc-
tion in their oxygen contents. Thus, catalytic hydropro-
cessing to convert the compounds primarily into hy-
drocarbons and water (hydrodeoxygenation, HDO) is
a candidate process for upgrading of biomass-derived
liquids, and such processes are usually catalytic, per-
formed at high temperatures and in the presence of
high partial pressures of H2.
Among the various polymers present in lignocellu-
losic biomass, lignin, with its polyaromatic structure,
is the one that is chemically most similar to petroleum.
Nonetheless, lignin contains much more oxygen than
crude oil [12–14], and lignin-derived bio-oils are good
candidates for hydroprocessing for oxygen removal.
Only little information characterizing the kinetics
of HDO of lignin-derived bio-oil compounds has been
reported. HDO processes are similar to the hydrodesul-
furization (HDS) and hydrodenitrogenation processes
used widely in petroleum refining [15,16], and the cat-
alysts of choice for the latter processes are metal sul-
fides, exemplified by molybdenum sulfide promoted
with cobalt and supported on porous γ -Al2O3. These
catalysts have also been investigated for HDO, and we
chose to investigate anisole HDO in the presence of
this catalyst, referred to simply as CoMo/Al2O3. Be-
cause such catalysts revert from the sulfided to oxide
forms in the absence of sulfur-containing compounds
and H2, we carried out our investigation with the re-
actant in the presence of presulfided catalyst that had
been pretreated with H2S. Earlier work has shown that
HDO proceeds by two reaction paths on different kinds
of active sites on the sulfided catalyst, one involving
direct hydrogenolysis (breaking of the bond between
the oxygen atom and the aromatic carbon atom in a
The presence of cobalt improves the activity of
molybdenum sulfide catalysts for hydroprocessing
reactions [24–28]. Bui et al. [24,25] considered the
promoting effect of cobalt on molybdenum sulfide cat-
alysts for lignin-derived bio-oil conversion. They ob-
served that the addition of cobalt increased the rates
of the direct demethoxylation and direct deoxygena-
tion pathways [25]. Lee et al. [29] concluded that de-
oxygenation, demethoxylation, and demethylation re-
actions following saturation of the aromatic ring in
guaiacol were essential to account for the product dis-
tributions observed with sulfided CoMo/Al2O3 and sul-
fided NiMo/Al2O3 catalysts. Furimsky and cowork-
ers [27,30] investigated HDO of phenolic compounds
catalyzed by sulfided CoMo/Al2O3, finding that pre-
sulfiding of the catalyst improved the conversion of
phenols to benzene and cyclohexane. In related work,
Vuori et al. [28] reported that sulfiding of catalysts had
a pronounced effect on the conversion of guaiacol.
H2S plays a role in addition to that of maintain-
ing the catalyst in a sulfided state. H2S is also an in-
hibitor of HDO reactions, as illustrated for a mixture
of lignin-derived bio-oil compounds by Laurent and
Delmon [19,20,31], who investigated the performance
of CoMo/Al2O3 catalysts under various conditions and
found that oxygen removal reactions are inhibited by
both sulfur- and nitrogen-containing compounds. Fer-
rari et al. [32] inferred that increasing the H2S partial
pressure led to inhibition of both the hydrogenolysis
and hydrogenation pathways, and Senol et al. [33] ob-
served that the formation of both benzene and cyclo-
hexene in the conversion of phenol was inhibited by
H2S. Gutierrez et al. [34] observed that in guaiacol
HDO with a sulfided CoMo/Al2O3 catalyst the prod-
ucts were contaminated with sulfur and the catalyst
underwent deactivation associated with carbonaceous
deposits.
Because of the complexity and variability of struc-
tures of compounds in bio-oils and the variations in
molecular weight and concentration of these com-
pounds, only little quantitative information character-
izing the kinetics of bio-oils hydroprocessing is avail-
able [6]. Massoth et al. [35] investigated the kinet-
ics of HDO of methyl-substituted phenols catalyzed
International Journal of Chemical Kinetics DOI 10.1002/kin.21026

EXPERIMENTAL INVESTIGATION ON UPGRADING OF LIGNIN-DERIVED BIO-OILS
3
Figure 1 Schematic representation of a fixed-bed tubular microflow reactor system.
by a sulfided CoMo/Al2O3 and represented their data
with Langmuir–Hinshelwood expressions, but the re-
sults are too few data to justify any generalizations.
Bui et al. [24,25] investigated HDO of guaiacol in
the presence of similar CoMo catalysts in a fixed-bed
reactor and represented their data with simple first-
order kinetics. In related work, Saidi et al., [8,12,14],
Runnebaum et al. [9,17,36,37], and Nimmanwudipong
et al. [38] used first-order kinetics to approximate their
data and provided a basis for comparison of the reac-
tivities of various lignin-derived compounds in HDO,
but their catalyst was an unsulfided supported noble
metal, Pt/Al2O3.
Runnebaum’s investigation is one of the few to
include quantitative kinetics characterizing catalytic
HDO. They elucidated an approximate reaction net-
work characterizing anisole conversion catalyzed by
Pt/Al2O3 and determined pseudo–first-order rate con-
stants for the formation of individual products. The cat-
alytic hydrotreating of anisole with sulfided hydrotreat-
ing catalysts was also investigated by Bredenberg
et al. [39] and by Hurff and Klein [40], but reaction
networks were either not reported or simplified to in-
clude only a few main products.
values of approximate rate constants and activation en-
ergies.
EXPERIMENTAL
Catalytic reaction experiments were performed with
a once-through fixed-bed tubular micro flow reactor.
The flow system is shown schematically in Fig. 1, and
details are represented in Table I. The catalyst used in
all experiments was American Cyanamid HDS 16A.
The catalyst properties before sulfiding are as follows:
2
3
surface area, 176 m /g; pore volume, 0.50 cm /g; Co
content, 4.4 wt%; and Mo content, 7.5 wt%. Parti-
cles of the catalyst were placed on the porous plate in
the once-through reactor operated in down-flow mode,
Table I Flow Reactor System Specification and
Operating Conditions
Specification of Equipment
Components
Property
Reactor length, mm
305
Reactor external diameter, mm
Reactor internal diameter, mm
Reactor material
14.5
9
The literature of catalytic HDO of lignin-derived
bio-oils has been reviewed in detail [6], but there is
still a lack of fundamental understanding of the re-
actions, networks, kinetics, and mechanisms. Specifi-
cally, there is a lack of data characterizing kinetics and
reaction networks for HDO with sulfided hydropro-
cessing catalysts.
316 L stainless steel
20
Reactor volume, mL
Reactor operating temperature
range, K
573–673 (± 1.5)
Reactor operating pressure, bar
Carrier gas (N2) flow rate, L/h
H2 flow rate, L/h
8–20 (± 0.1)
20
20
Our goals in the work summarized below were to
investigate the HDO of a compound representative of
lignin-derived bio-oils, namely, anisole, and to assess
the performance of sulfided CoMo/Al2O3 by determin-
ing an approximate reaction network including some
Liquid feed (anisole) flow rate
range, mL/min
Catalyst (sulfided
CoMo/Al2O3 - American
Cyanamid HDS 16A) mass, g
0.03–0.5 (± 1%)
0.25–2.00
International Journal of Chemical Kinetics DOI 10.1002/kin.21026

4
RAHIMPOUR ET AL.
mixed with about 2 g of inert, nonporous α-Al2O3 par-
ticles. Sulfiding was achieved by passing 10% H2S in
H2 through the bed at a rate of 20 L/h, with the catalyst
being heated at a rate of about 10°C/min to 500°C.
After 2 h at 500°C, the H2S/H2 stream was replaced by
N2, and the reactor was cooled to the operating temper-
ature. A sample of catalyst powder (less than 100 mesh
particles) was charged to the reactor. A fresh catalyst
zene, phenol, 2-methylphenol, 2,6-dimethylphenol,
2,4,6-trimethylphenol, 2,3,5,6-tetramethylphenol, and
2-tert-butyl-4-methylphenol.
Conversion (X) and selectivity to various products i
Si) are defined as follows:
(
(Moles of anisole)in − (Moles of anisole)out
X =
× 100%
1)
(Moles of anisole)
in
(0.25–2.00 g) was used for each experiment, and each
(
experiment was typically run for 6 h of continuous
operation.
The liquid reactant (anisole 99.8%; Merck-
Billerica, Massachusetts, USA) was introduced into
the reactor at room temperature by means of a high-
pressure liquid chromatography pump (HPLC). N2 was
used as the carrier gas, and H2, when present as a reac-
tant, was mixed into the carrier gas stream. Liquid and
gas streams were introduced into a hot-box system that
included an electric forced-convection heater that held
the components at a temperature of 453 K to avoid pos-
sible condensation of reactants or products. Preheated
liquid was vaporized, and the vapor stream was mixed
with the gas feed stream and flowed to the reactor. At
the reactor outlet, the product stream flowed out of the
hot box to a condenser, which operated at 275–280
K. Liquid product samples were taken periodically at
intervals of 30 min downstream of the condenser for
analysis.
Molar flow rate of product i
Molar flow rate of main products formed
Si =
(2)
The catalytic conversion of anisole was carried out
at temperatures in the range of 573–673 K and pres-
sures in the range of 8–20 bar; the liquid anisole flow
rates were in the range of 0.03–0.5 mL/min, and the
N2 and H2 flow rates were each 20 L/h. The value of
the weight hourly space velocity (WHSV) was in the
range of 3–120 (g of anisole)/(g of catalyst × h), varied
by changing the catalyst mass and liquid flow rate.
RESULTS
The conversion of anisole catalyzed by commercial
sulfided CoMo/Al2O3 was investigated under HDO
conditions. The catalyst was presulfided, but the sul-
fiding agent H2S was not added to the feed mixture
because it was expected that it would have inhibited
the reactions involving anisole and H2. GC analy-
sis of the liquid reaction products demonstrated the
formation of complex reaction mixtures in which the
main products were benzene, phenol, 2-methylphenol,
2,4-dimethylphenol, 2,4,6-trimethylphenol, 2,3,5,6-
tetramethylphenol, and 2-tert-butyl-4-methylphenol.
GC-MS analyses of the product also gave evidence
the minor products methane, methanol, and water. The
conversion data demonstrate a lack of dependence on
the mass of catalyst (in the range 0.25–2.00 g) at fixed
values of the WHSV in the range of 3–120 (g of
anisole)/(g of catalyst × h), consistent with the infer-
ence that extraparticle mass transfer limitations were
negligible. Calculations representing the intraparticle
mass transfer resistance indicate that it was also neg-
ligible: The calculated values of the modified Thiele
modulus (Weisz modulus) were less than 0.4, indicat-
ing that the anisole conversion rate was not limited by
internal diffusion and therefore that the catalytic reac-
tion experiments determined intrinsic chemical kinet-
ics. The mass balance closures were generally between
95% and 100% (Table II).
Product analysis was performed with a Bruker 450
gas chromatograph (GC) equipped with a flame ion-
ization detector. Samples were diluted with acetone
to make them single phase. The GC parameters were
as follows: sample volume, 0.5 µL; injector tempera-
ture, 573 K; temperature program starting temperature,
323 K; temperature ramp, 20 K/min, followed by a
soak of 35 min. The GC was equipped with an OP-
TIMA 5 Macherey–Nagel capillary column (30 m ×
0.25 mm, with a 0.25-µm film thickness). The detector
temperature was set at 548 K, and the gas flow rates
were as follows: H2, 30 mL/min; air, 300 mL/min. The
carrier gas flow rate was constant within the column,
with a makeup flow to give a combined flow rate of
30 mL/min; the makeup gas was argon. Moreover, for
identification of trace compounds, a few liquid sam-
ples were analyzed qualitatively with a Shimadzu QP
50/50 gas chromatograph-mass spectrometer (GC-MS)
equipped with a SGE BPX5 capillary column (0.32 mm
×
30 m × 0.25 µm), with helium (50 mL/min) used as
the carrier gas. Chromatographic peaks were identified
with individual compounds on the basis of mass spec-
tra by matching to a Willy library. The identifications
of the most abundant products were verified by com-
parison with the analysis of authentic standards pur-
chased from Sigma Aldrich; these standards were ben-
International Journal of Chemical Kinetics DOI 10.1002/kin.21026

EXPERIMENTAL INVESTIGATION ON UPGRADING OF LIGNIN-DERIVED BIO-OILS
5
Table II Overall Mass Balance Closure Data
Approximate Mass
of Liquid
Remaining in
Condenser (g)
Time on Liquid Flow
Mass of Liquid Mass of Reactant
Collected from Leaving Condenser
Experiment Mass Balance Stream
Rate
Mass of
Number
Closure (%)
(min)
(mL/min) Liquid Fed (g) Condenser (g)
as Vapor (g)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
96.2
98.7
95.7
98.5
98.1
95.8
96.6
99.7
95.4
95.9
97.4
97.3
96.9
95.5
96.8
96.0
97.6
97.4
96.5
98.1
150
150
150
150
150
150
150
180
180
180
180
180
180
180
180
360
360
360
360
360
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.03
0.03
0.03
0.03
0.03
14.93
14.93
14.93
14.93
14.93
14.93
14.93
8.96
8.96
8.96
8.96
8.96
8.96
8.96
8.96
10.75
10.75
10.75
10.75
10.75
13.07
13.52
12.98
13.49
13.36
12.98
13.18
7.78
7.39
7.4
7.58
7.56
7.51
7.36
7.44
9.14
9.28
9.25
9.19
9.35
0.29
0.21
0.3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.21
0.28
0.31
0.24
0.15
0.15
0.19
0.14
0.15
0.17
0.19
0.23
0.18
0.21
0.22
0.18
0.19
0
1
2
3
4
5
6
7
8
9
0
Approximate Reaction Network
the formation of phenol derivatives remained approxi-
mately unchanged with increasing conversion, but that
of benzene decreased.
The complexity of the product distribution makes it
challenging to determine an exact reaction network to
summarize the anisole conversion data. Thus, one of
our goals was to establish a good approximation of the
network—on the basis of selectivity–conversion data
determining which products are primary and which
are nonprimary. The evolution of the selectivities of
the main products as a function of anisole conversion
in the presence of H2 at 8 bar, 573−673 K, and 0.008–
The formation of these products in the hydropro-
cessing of anisole suggests that both HDO (indicated,
e.g., by the formation of benzene) and hydrogenoly-
sis by O-demethylation (indicated, e.g., by the forma-
tion of phenol and methane) are kinetically significant
reaction pathways and that they are accompanied by
transalkylation (indicated, e.g., by the formation of 2-
methylphenol). Thus, a partial statement of the network
for conversion of anisole by the sulfided CoMo/Al2O3
catalyst under our conditions, accounting for just the
primary products and the other major products, is that
represented in Fig. 3.
As indicated in Fig. 3, we infer two routes for
the conversion of anisole to benzene: One, the direct
deoxygenation (DDO) pathway whereby benzene is
formed via loss of the methoxy functionality and, an-
other, demethylation of the methoxy group followed
by HDO. The observation of methane, methanol, and
water as products is consistent with the occurrence of
these reactions. The absence of substituted cyclohex-
anes with oxygen-containing groups among the main
products shows that the demethylation and HDO reac-
tions that take place on the CoMo catalyst dominate
0.667 (g of catalyst × h)/(g of anisole) is shown in
Figs. 2, S-1, and S-2 (the latter two are in the Sup-
porting Information). The lines shown in these fig-
ures were used for extrapolation to estimate the selec-
tivities at zero conversion. The selectivity–conversion
plots provide the following information about the re-
action network: The main products phenol, benzene,
and 2-methylphenol appeared with positive intercepts
on the plots (consistent with the values determined
by linear regression), indicating primary products. In
contrast, the other main products, 2,6-dimethylphenol,
2,4,6-trimethylphenol, 2,3,5,6-tetramethylphenol, and
2-tert-butyl-4-methylphenol, appeared with approxi-
mately zero intercepts, indicating nonprimary prod-
ucts. The maximum selectivity is that for formation
of phenol. As indicated in Fig. 2, the selectivities for
International Journal of Chemical Kinetics DOI 10.1002/kin.21026

6
RAHIMPOUR ET AL.
Figure 2 Selectivity for the formation of the following products in the conversion of anisole catalyzed by sulfided CoMo/Al2O3
in the presence of H2 at temperature = 573 K and pressure = 8 bar. Data for each of the main products were fitted with a
straight line and extrapolated to zero conversion; intercepts of the lines that are significantly different from zero selectivity at
zero conversion indicate primary products.
kinetically over the formation of hydrogenation
products—thus the data indicate that the reactions can
be essentially stopped at the point where HDO is virtu-
ally completed and H2 consumption is minimized. We
infer that sequential methyl group transfer from com-
pounds including anisole to methylphenols produces
bond. O-Demethylation is inferred to be accompanied
by methylation of the aromatic ring, a class of reaction
that is often associated with support acidity. Forma-
tion of 2-methylphenol as one of the main products,
in contrast to 4-methylphenol, indicates that the o-
position is strongly favored over the p-position in the
alkylation.
2,6-dimethylphenol, 2,4,6-trimethylphenol, 2,4,5,6-
tetramethylphenol, and 2-tert-butyl-4-methylphenol.
According to the selectivity–conversion data, the se-
lectivity for one route may decline in favor of
another as conversion increases. Nonetheless, it is ev-
ident that the conversion of the anisole at various op-
erating conditions on the sulfided CoMo/Al2O3 cata-
lyst leads to the formation of phenolic products via a
combination of demethylation/alkylation steps that are
known to take place on acidic surfaces such as that
of the alumina support [6,17,36,41]. The high selec-
tivities to phenol suggest that the demethylation path-
way is the major pathway in the anisole conversion,
as might be expected because of the higher strength
of the H3CO–Caromatic bond relative to the O–Cmethyl
Kinetics
In investigations of catalytic reaction kinetics, it is
common to report initial reaction rates as measures
of catalytic activity, and we report such data be-
low. In investigations of catalytic hydroprocessing, it
is common to report kinetics in terms of rate con-
stants that are pseudo–first order in the organic re-
actant [35–37,42–45], and we report such data also,
with results showing that the approximation is good
at the intermediate conversions we have investigated.
Because catalyst deactivation was observed to be neg-
ligible during 6 h of continuous operation in our
International Journal of Chemical Kinetics DOI 10.1002/kin.21026

EXPERIMENTAL INVESTIGATION ON UPGRADING OF LIGNIN-DERIVED BIO-OILS
7
Figure 3 Approximate reaction network for the conversion of anisole catalyzed by sulfided CoMo/Al2O3 at 573–673 K and
–20 bar.
8
at a given WHSV, and the rate of formation of each of
the products, increase with increasing temperature in
the observed range, from 573 to 673 K, and, at a given
temperature, the conversion increases with increasing
inverse WHSV, as expected. The rates of anisole con-
version at 573, 623, and 673 K were found to be 5.26
−
3
−3
−2
×
10 , 8.54 × 10 , and 1.28 × 10 (g of anisole/g
of catalyst × h), respectively. The initial rates of ap-
pearance of the various products of anisole conversion
(Table III) show that the primary product formed at the
highest initial rate is phenol, and it is formed more than
10 times faster than benzene.
To estimate rate constants for the primary reactions,
we made plots of (1 – Xi) (where Xi is the conversion
of anisole to species i) on logarithmic scales versus
inverse space velocity (Figs. 6, S-4 and S-5 (in the
Supporting Information)). The pseudo–first-order rate
constants for the three primary products shown in the
reaction network of Fig. 5 were determined accord-
ing to Eqs. (3)–(7). As indicated in these figures, the
slopes of the linear plots are consistent with pseudo–
first-order reactions, but the conversions are too low to
determine the reaction orders with confidence. These
are composite rather than fundamental rate constants
because the primary products were consumed in non-
primary reactions.
Figure 4 Conversion of anisole at various temperatures and
bar.
8
experiments, we have used average conversions over
the entire 6-h period of time on stream to determine
the reported rate constants.
Conversions of anisole were determined from
changes in the anisole concentration, and conversions
to individual products were determined from analy-
ses of those products. Data showing the conversion of
anisole as a function of the inverse WHSV and the con-
versions to individual major products in a typical run
are presented in Figs. 4 and 5, S-3, and S-4 (the latter
two figures in the Supporting Information). The data
fall near straight lines that pass through the origin, so
that the slopes of the lines determine the rates of the
reactions. As shown in Fig. 4, the conversion of anisole
d ci
ri =
= kicAnisole
(3)
d(cAnisole,0/WHSV)
ci = cAnisole,0Xi
(4)
International Journal of Chemical Kinetics DOI 10.1002/kin.21026

8
RAHIMPOUR ET AL.
ture are those for the formation of phenol. These values
of the apparent rate constants were used to determine
values of the apparent activation energies, calculated
on the basis of the Arrhenius equation (Eq. 8). The cor-
responding plots are shown in Fig. 7, and the estimated
values of apparent activation energies are summarized
in Table IV. The values of these apparent activation en-
ergies are the following: for the formation of benzene,
approximately 18 kJ/mol; for phenol, approximately
2
1
4 kJ/mol; and for 2-methylphenol, approximately
4 kJ/mol. We emphasize that these kinetics parame-
ters are composite values and do not represent a quan-
titative resolution of the approximate reaction network
of Fig. 3.
ꢀ−
ꢁ
Ea
k = k0 exp
(8)
RT
Effects of Operating Conditions
The effects of operating conditions on the anisole con-
version were investigated at a constant WHSV (3 (g
of anisole)/(g of catalyst × h)). As shown in Fig. 8,
the anisole conversion at 8 bar increased from 17%
to 57% as the temperature increased from 573 to
673 K. The selectivities for formation of phenol and 2-
methylphenol decreased with increasing temperature,
whereas the selectivities to benzene and other phenol
derivatives increased (Fig. 9).
At 623 K, the conversion increased from 57%
to 69% as the pressure increased from 8 to 20 bar
(Fig. 10). The effect of pressure on selectivity for
the formation of the major products at conversions
in the range of 55%–70% is shown in Fig. 11. Un-
der our conditions, increasing the pressure increased
the selectivity for formation of phenol, but it de-
creased the selectivity for formation of benzene and
other products formed from phenol. We infer that al-
though increasing the pressure improved conversion
and selectivity to the hydrogenolysis product phenol,
it led to decreased selectivity to HDO products; the
selectivity to benzene was reduced from 0.12 to 0.08
as the pressure increased from 8 to 20 bar. We in-
fer that if oxygen removal is the principal process-
ing goal, higher temperatures and lower pressures are
more favorable, within the range of our operating
conditions.
Figure 5 Conversion to main products at 573 K and 8 bar;
the term Xi represents the conversion of anisole to product i.
cAnisole = cAnisole,0(1 − Xi)
(5)
(6)
(7)
d Xi
ri =
= ki(1 − Xi)
d(cAnisole,0/WHSV)
DISCUSSION
ln(1 − Xi) = −ki(cAnisole,0/WHSV)
The results presented here are more complete than
reported data [20,46–49] for anisole conversion cat-
alyzed by sulfided CoMo/Al2O3, including kinetics
As shown in Table IV, the maximum rate constants
among those determined in this work at each tempera-
International Journal of Chemical Kinetics DOI 10.1002/kin.21026

EXPERIMENTAL INVESTIGATION ON UPGRADING OF LIGNIN-DERIVED BIO-OILS
9
Table III Initial Rates of Formation of the Main Products
Rate (mol/(g of catalyst × h)) at T
Product
573 K
623 K
673 K
−3
−4
−3
−4
−4
−5
−5
−3
−3
−4
−3
−3
−4
−4
−4
Phenol
Benzene
2.31 × 10
2.19 × 10
1.76 × 10
2.15 × 10
1.89 × 10
8.14 × 10
3.24 × 10
4.20 × 10
7.25 × 10
5.36 × 10
3.64 × 10
2.10 × 10
5.69 × 10
5.12 × 10
1.81 × 10
−4
3.52 × 10
−3
2
2
2
2
2
-Methylphenol
,6-Dimethylphenol
,4,6-Trimethylphenol
,3,5,6-Tetramethylphenol
-tert-Butyl-4-methylphenol
2.76 × 10
−3
1.09 × 10
−4
3.00 × 10
−4
2.42 × 10
−5
7.31 × 10
Figure 6 Conversion of anisole in the presence of H2 at 573 K and 8 bar to give the main products, where Xi is conversion of
anisole to product i on a logarithmic scale.
data that provide a preliminary quantitative basis for
determining the initial product distributions at several
temperatures and pressures and a basis for identify-
ing the reaction classes to help identify major reaction
pathways.
The selectivity–conversion data represented in
Figs. 2, S-1, and S-2 (latter two figures in the Sup-
porting Information) were used to identify primary
and nonprimary products, with phenol, benzene, and
2-methylphenol being primary products and other
International Journal of Chemical Kinetics DOI 10.1002/kin.21026

1
0
RAHIMPOUR ET AL.
Table IV Pseudo–First-Order Rate Constants and
Apparent Activation Energies for Reactions in the
Network Representing Anisole Conversion in the
Presence of H2 and Sulfided CoMo/Al2O3 Catalyst
ki (L/(g of
Ea
Product
Phenol
T (K) catalyst × h)) (kJ/mol)
573
3.03
3.56
6.56
24.3
18.4
14.4
56.0
24.3
51.6
49.2
6
6
23
73
Benzene
573
0.23
0.29
0.41
6
6
23
73
2
2
2
2
2
-Methylphenol
573
1.92
2.25
3.02
6
6
23
73
,6-Dimethylphenol
,4,6-Trimethylphenol
,3,5,6-Tetramethylphenol
573
0.22
0.90
1.23
6
6
23
73
573
0.19
0.23
0.41
6
6
23
73
573
0.08
0.19
0.40
6
6
23
73
-tert-Butyl-4-methylphenol 573
0.03
0.06
0.14
6
6
23
73
Figure 7 Arrhenius plots for formation of main products in
the reaction of anisole catalyzed by CoMo/γ Al2O3, where k
is rate constant and T is the reaction temperature.
phenol derivatives being nonprimary products. But
we emphasize that if a secondary product is formed
faster than primary product, it may appear to be
primary by our criterion of a positive intercept on
a selectivity–conversion plot indicating a primary
product.
The results demonstrate the occurrence of HDO, hy-
drogenolysis, and alkylation as the main reaction path-
ways for anisole conversion catalyzed by our sulfided
CoMo/Al2O3. The data show that under our conditions
these reactions dominated over hydrogenation.
Our data are consistent with the premise that CoMo
sites are active for HDO reactions, and we thus con-
sider it likely that Co-Mo catalyzes (a) hydrogenoly-
sis, to form phenol, and (b) HDO to form benzene,
which are kinetically significant reactions under our
conditions. Alkylation reactions are expected to oc-
cur in the presence of acidic sites on the catalyst sup-
port (γ -Al2O3). Consistent with results observed previ-
ously [21,24,39,46,47,49], no evidence was observed
in our work for hydrogenated phenols. Ring hydro-
genation may require the aromatic ring to adsorb al-
most flat on the catalyst surface, and the substituents
Figure 8 Conversion of anisole at 8 bar and WHSV = 3.6
g of anisole)/(g of catalyst × h) and various temperatures.
(
on the ring may have led to nonplanar coordination and
hampering of hydrogenation pathways.
We suggest that the kinetics parameters presented
here may be of value for preliminary design predic-
tions for the conversion of anisole and perhaps related
lignin-derived bio-oils components. The data summa-
rized in Table II show that among the various reactions
on sulfided CoMo/Al2O3, hydrogenolysis (in our case,
International Journal of Chemical Kinetics DOI 10.1002/kin.21026

EXPERIMENTAL INVESTIGATION ON UPGRADING OF LIGNIN-DERIVED BIO-OILS
11
Figure 10 Conversion of anisole at 623 K and WHSV =
.6 (g of anisole)/(g of catalyst × h) and various pressures.
3
Figure 9 Temperature dependence of selectivity for the
formation of the main products of the reaction of anisole with
H2 at 8 bar and WHSV = 3.6 (g of anisole)/(g of catalyst ×
h).
to form phenol) is characterized by the highest rate
and that selectivities for oxygen removal are favored
by operation at higher temperatures and lower pres-
sures. Increasing the temperature enhances the selec-
tivity for HDO and alkylation products but reduces the
selectivity for hydrogenolysis. But we emphasize that
because of the complexity of bio-oils feedstocks and
the limitations of available data characterizing reaction
networks and kinetics, much work is required for a firm
foundation to allow even rough design predictions.
Figure 11 Pressure dependence of selectivity for the for-
mation of the main products of the reaction of anisole with
H2 at 623 K and WHSV = 3.6 (g anisole)/(g of catalyst ×
h).
CONCLUSIONS
The catalytic upgrading of anisole catalyzed by sul-
fided CoMo/Al2O3 was investigated in the tempera-
ture range of 573–673 K and the pressure range of
network accounts for hydrogenolysis leading to re-
moval of methoxy groups in anisole to form phe-
nol, followed by deoxygenation to give benzene or
transalkylation to give other phenol derivatives; ben-
zene was the only fully deoxygenated organic prod-
uct. We infer that the Co–Mo sites are active for
8–20 bar. The primary observed products were phe-
nol, benzene, and 2-methylphenol, with major non-
primary products being 2,6-dimethylphenol, 2,4,6-
trimethylphenol, 2,3,5,6-tetramethylphenol, and 2-
tert-butyl-4-methylphenol. An approximate reaction
International Journal of Chemical Kinetics DOI 10.1002/kin.21026

1
2
RAHIMPOUR ET AL.
hydrogenolysis, HDO, (and by inference, hydrogena-
tion) and that the acidic alumina support is respon-
sible for the transalkylation reactions. The rates of
anisole conversion at 573, 623, and 673 K and 8 bar
22. Oasmaa, A.; Czernik, S. Energy Fuels 1999, 13, 914–
921.
23. Wang, Y.; Lin, H.; Zheng, Y. Catal Sci Technol 2014, 4,
109–119.
−3
−3
24. Bui, V. N.; Laurenti, D.; Afanasiev, P.; Geantet, C. Appl
Catal B: Environ 2011, 101, 239–245.
were found to be 5.3 × 10 , 8.5 × 10 , and 1.3 ×
−
2
10
(g anisole/g of catalyst × h), respectively. The
2
2
2
2
2
5. Bui, V. N.; Laurenti, D.; Delich e` re, P.; Geantet, C. Appl
Catal B: Environ 2011, 101, 246–255.
6. Lee, W.-S.; Wang, Z.; Wu, R. J.; Bhan, A. J Catal 2014,
data suggest that oxygen removal of lignin-derived bio-
oils is favored by operation at higher temperatures and
lower pressures.
3
19, 44–53.
7. Furimsky, E. Appl Catal A: Gen 2000, 199, 147–
90.
8. Vuori, A.; Helenius, A.; Bredenberg, J. B. S. Appl Catal
989, 52, 41–56.
1
BIBLIOGRAPHY
1
9. Lee, C. R.; Yoon, J. S.; Suh, Y.-W.; Choi, J.-W.; Ha,
J.-M.; Suh, D. J.; Park, Y.-K. Catal Commun 2012, 17,
54–58.
1
. Chan, J. M. W.; Bauer, S.; Sorek, H.; Sreekumar, S.;
Wang, K.; Toste, F. D. ACS Catal 2013, 3, 1369–1377.
. Zaheer, M.; Kempe, R. ACS Catal 2015, 5, 1675–1684.
. Wahyudiono; Sasaki, M.; Goto, M. Chem Eng Process:
Process Intensification 2008, 47, 1609–1619.
2
3
30. Furimsky, E.; Massoth, F. E. Catal Today 1999, 52, 381–
495.
31. Taimoor, A. A.; Favre-Reguillon, A.; Vanoye, L.;
Pitault, I. Catal Sci Technol 2012, 2, 359–363.
32. Ferrari, M.; Maggi, R.; Delmon, B.; Grange, P. J Catal
2001, 198, 47–55.
4
5
6
. Werhan, H.; Assmann, N.; Rudolf von Rohr, P. Chem
Eng Process: Process Intensification 2013, 73, 29–37.
. Macfarlane, A. L.; Farid, M. M.; Chen, J. J. J. Chem Eng
Process: Process Intensification 2009, 48, 864–870.
. Saidi, M.; Samimi, F.; Karimipourfard, D.;
Nimmanwudipong, T.; Gates, B. C.; Rahimpour,
M. R. Energy Environ Sci 2014, 7, 103–129.
˙
33. S¸ enol, O. I.; Ryymin, E. M.; Viljava, T. R.; Krause,
A. O. I. J Mol Catal A: Chem 2007, 277, 107–112.
34. Gutierrez, A.; Kaila, R. K.; Honkela, M. L.; Slioor, R.;
Krause, A. O. I. Catal Today 2009, 147, 239–246.
35. Massoth, F. E.; Politzer, P.; Concha, M. C.; Murray,
J. S.; Jakowski, J.; Simons, J. J Phys Chem B 2006, 110,
14283–14291.
36. Runnebaum, R. C.; Nimmanwudipong, T.; Limbo,
R. R.; Block, D. E.; Gates, B. C. Catal Lett 2012, 142,
7–15.
7
8
9
. Sharma, R. K.; Bakhshi, N. N. Can J Chem Eng 1991,
69, 1071–1081.
. Saidi, M.; Rostami, P.; Rahimpour, M. R.; Gates, B. C.;
Raeissi, S. Energy Fuels 2014, 29, 191–199.
. Runnebaum, R. C.; Nimmanwudipong, T.; Block, D. E.;
Gates, B. C. Catal Sci Technol 2012, 2, 113–118.
1
0. Xiong, H.; Pham, H. N.; Datye, A. K. Green Chem 2014,
6, 4627–4643.
37. Runnebaum, R.; Nimmanwudipong, T.; Block, D.;
Gates, B. Catal Lett 2011, 141, 817–820.
1
1
1. Garc ´ı a, A.; Erdocia, X.; Gonz a´ lez Alriols, M.; Labidi,
J. Chem Eng Process: Process Intensification 2012, 62,
38. Nimmanwudipong, T.; Runnebaum, R.; Block, D.;
Gates, B. Catal Lett 2011, 141, 779–783.
1
50–158.
2. Saidi, M.; Rahimpour, M. R.; Raeissi, S. Energy Fuels
015, 29, 3335–3344.
39. Bredenberg, J. B. s.; Huuska, M.; R a¨ ty, J.; Korpio, M. J
Catal 1982, 77, 242–247.
40. Hurff, S. J.; Klein, M. T. Ind Eng Chem Fundam 1983,
22, 426–430.
41. Nimmanwudipong, T.; Runnebaum, R.; Tay, K.; Block,
D.; Gates, B. Catal Lett 2011, 141, 1072–1078.
42. Yang, Y. Q.; Tye, C. T.; Smith, K. J. Catal Commun
2008, 9, 1364–1368.
1
1
1
2
3. Xu, H.; Wang, K.; Zhang, H.; Hao, L.; Xu, J.; Liu, Z.
Catal Sci Technol 2014, 4, 2658–2663.
4. Saidi, M.; Rostami, P.; Rahimpour, H. R.; Roshanfekr
Fallah, M. A.; Rahimpour, M. R.; Gates, B. C.; Raeissi,
S. Energy Fuels 2015, 29, 4990–4997.
1
1
1
5. Girgis, M. J.; Gates, B. C. Ind Eng Chem Res 1991, 30,
43. Whiffen, V. M. L.; Smith, K. J.; Straus, S. K. Appl Catal
A: Gen 2012, 419–420, 111–125.
44. Zhao, H. Y.; Li, D.; Bui, P.; Oyama, S. T. Appl Catal A:
Gen 2011, 391, 305–310.
45. Centeno, A.; Laurent, E.; Delmon, B. J Catal 1995, 154,
288–298.
2
021–2058.
6. Mohan, D.; Pittman, C. U.; Steele, P. H. Energy Fuels
006, 20, 848–889.
2
7. Runnebaum, R. C.; Lobo-Lapidus, R. J.;
Nimmanwudipong, T.; Block, D. E.; Gates, B. C.
Energy Fuels 2011, 25, 4776–4785.
46. Huuska, M.; Rintala, J. J Catal 1985, 94, 230–238.
47. Jongerius, A. L.; Jastrzebski, R.; Bruijnincx, P. C. A.;
Weckhuysen, B. M. J Catal 2012, 285, 315–323.
48. Loricera, C. V.; Pawelec, B.; Infantes-Molina, A.;
1
1
2
2
8. Gevert, B. S.; Otterstedt, J. E.; Massoth, F. E. Appl Catal
1
987, 31, 119–131.
9. Laurent, E.; Delmon, B. Ind Eng Chem Res 1993, 32,
516–2524.
0. Laurent, E.; Delmon, B. Appl Catal A: Gen 1994, 109,
7–96.
1. Elliott, D. C. Energy Fuels 2007, 21, 1792–1815.
´
2
Alvarez-Galv a´ n, M. C.; Huirache-Acu n˜ a, R.; Nava, R.;
Fierro, J. L. G. Catal Today 2011, 172, 103–110.
49. Viljava, T. R.; Komulainen, R. S.; Krause, A. O. I. Catal
Today 2000, 60, 83–92.
7
International Journal of Chemical Kinetics DOI 10.1002/kin.21026