Zeolite-Beta-supported platinum catalysts for hydrogenation/ hydrodeoxygenation of pyrolysis oil model compounds

Article abstract of DOI:10.1016/j.cattod.2012.08.015

Phenol, cresols, guaiacol and eugenol were studied to describe the effect of substituent position and type on the reactivity of these phenolic compounds. Conversion of cresols under deoxygenation/hydrogenation reactivities increased in order m- > o- > p-cresol over Pt/Beta zeolite with Si/Al = 12.7 and o- > m- > p-cresol over the Pt/Beta with lower Al content (Si/Al = 20.7). Major products of cresols hydrogenation were methyl cyclohexanols. Phenol was converted into a mixture of cyclohexane, cyclohexanol and cyclohexanone. An increase in Si/Al ratio of the parent zeolite-based catalysts resulted in a decrease in the overall conversion of cresols and phenol. On the other hand, an increase in Si/Al ratio of catalysts prepared by dealumination led to accelerated conversion of phenol to cyclohexane. Using the lab-prepared mesoporous Beta catalyst with a comparable Si/Al ratio to the commercial one (ca. 20), higher yields of cyclohexanol were obtained in conversion of phenol.

Full text of DOI:10.1016/j.cattod.2012.08.015

Catalysis Today 204 (2013) 38–45
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Zeolite-Beta-supported platinum catalysts for
hydrogenation/hydrodeoxygenation of pyrolysis oil model compounds
∗
ˇ
Jan Horácˇek, Gabriela St’ávová, Vendula Kelbichová, David Kubicˇka
Research Institute of Inorganic Chemistry, UniCRE-RENTECH, Chempark Litvínov, 43670 Litvínov, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 May 2012
Received in revised form 3 August 2012
Accepted 4 August 2012
Available online 18 September 2012
Phenol, cresols, guaiacol and eugenol were studied to describe the effect of substituent position
and type on the reactivity of these phenolic compounds. Conversion of cresols under deoxygen-
ation/hydrogenation reactivities increased in order m- > o- > p-cresol over Pt/Beta zeolite with Si/Al = 12.7
and o- > m- > p-cresol over the Pt/Beta with lower Al content (Si/Al = 20.7). Major products of cresols
hydrogenation were methyl cyclohexanols. Phenol was converted into a mixture of cyclohexane, cyclo-
hexanol and cyclohexanone. An increase in Si/Al ratio of the parent zeolite-based catalysts resulted in
a decrease in the overall conversion of cresols and phenol. On the other hand, an increase in Si/Al ratio
of catalysts prepared by dealumination led to accelerated conversion of phenol to cyclohexane. Using
the lab-prepared mesoporous Beta catalyst with a comparable Si/Al ratio to the commercial one (ca. 20),
higher yields of cyclohexanol were obtained in conversion of phenol.
Keywords:
Hydrogenation
Deoxygenation
Phenols
Pyrolysis oils
Zeolite Beta
Platinum
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
production can become important source of chemicals. These mate-
rials are usually used traditionally for process heat recovery (lignin
Lignocellulose is the most abundant renewable source of chemi-
cals. While cellulose is primarily used for pulp and paper production
(5–36 × 108 tons/year [1]), it can be also hydrolyzed to fermentable
sugars that can be either fermented to produce ethanol or butanol,
or transformed for instance by hydrogenation or dehydration to
yield intermediates for chemical syntheses [2–7]. A similar strat-
egy relying on initial hydrolysis can be also adopted for upgrading
of hemicelluloses [5,8]. On the other hand, the chemical nature of
lignin differs significantly from cellulose and hemicelluloses and
different strategies for its upgrading need to be employed [9,10].
In addition, the origin of lignocellulosic biomass plays an impor-
tant role in selecting the most suitable upgrading technology. For
instance, large scale pulping or derived technologies are suitable
for wood processing [11], while agricultural residues, such as straw,
might require the use of pyrolysis technologies in order to enhance
the volumetric energy density (the volumetric energy density of a
straw bale is only 2 GJ/m3) and make the transport to an upgrading
facility economically feasible [12].
burning). Lignin, typically formed by methoxy-substituted phenyl
propanoid units (basic precursors: sinapyl alcohol, coniferyl alco-
hol and p-coumaryl alcohol) is a potential source for gasoline and
special chemicals production [9].
Thermal or so-called catalytic pyrolysis makes possible the
transformation of solid lignin or lignocellulose into a liquid phase
that can be processed much more easily by industrial catalytic tech-
nologies [13–15]. Typically, the liquid products consist of phenolic
compounds (originating from lignin), various acids, aldehydes and
ketones (attributable to cellulose and hemicelluloses), and water.
The exact composition of pyrolysis oil depends on feedstock com-
position and reaction (pyrolysis) conditions, such as temperature,
residence time or catalyst presence [16]. In particular phenolics
present in pyrolysis oils are interesting compounds for subsequent
hydrothermal processing (upgrading) as potential fuel and renew-
able chemicals sources. Due to a high reactivity of pyrolysis oil, i.e.
their tendency to polymerize under increased temperature condi-
tions, pyrolysis oils have to be generally stabilized first, e.g. by mild
hydrotreating [17–20].
Consequently, large attention is focused on transformation
of relevant model compounds. For instance, many studies were
focused on hydrothermal conversion of phenol. This model com-
pound can be deoxygenated over sulfided NiMo and CoMo catalysts
[21]. Typical reaction pathway described for phenol conversions
encompasses parallel and consecutive reactions, e.g. hydrodeoxy-
genation of hydroxyl group (benzene formation) followed by
Particularly wood is being exploited in millions of cubic meters
for pulp, paper and ethanol production. Together with agricul-
tural residuals (straw), which can be transformed into ethanol
using hydrolysis method, waste materials from pulp and paper
∗
Corresponding author.
E-mail address: david.kubicka@vuanch.cz (D. Kubicˇka).
0920-5861/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2012.08.015

J. Horácˇek et al. / Catalysis Today 204 (2013) 38–45
39
aromatic ring hydrogenation (cyclohexane) [22]. Alternatively, the
OH
OH
HO
OH
d)
c)
b)
a)
aromatic ring can be hydrogenated first followed by cyclohexanol
dehydration and subsequent hydrogenation of the resulting cyclo-
hexene to cyclohexane [22]. Sulfided catalysts are able to transform
phenol at temperatures around 250 ◦C. Their disadvantage is the
contact with sulfur and possible product contamination, because
of sulfur removal from active sites during the reaction [23]. More-
over, sulfur needs to be continuously replenished otherwise the
catalyst will be desulfided and lose gradually its activity In addi-
tion the presence of sulfur-containing compounds has a negative
impact on the reaction rate of deoxygenation due to competitive
adsorption of sulfur- and oxygen-containing compounds.
Presence of water, which is a typical component of pyrolysis
oils (water content in pyrolysis oil can reach up to approx. 30 wt.%
[24]), can also negatively affect catalytic activity. Nevertheless, the
deactivation caused by water is not as severe as in case of ammonia
and hydrogen sulfide [23,25].
CH3
CH3
H3C
O
CH3
CH2
HO
f)
e)
O
CH3
HO
Fig. 1. Model compounds used in the hydrogenation/deoxygenation study: (a) phe-
nol, (b) o-cresol, (c) m-cresol, (d) p-cresol, (e) guaiacol, and (f) eugenol.
Nickel and molybdenum oxides were described as suitable
catalysts for deoxygenation of alkyl and methoxyphenols [26].
Alumina-supported NiMo oxide catalyst (3% NiO, 15% MoO3)
was found to deoxygenate and hydrogenate dihydroxybenzenes,
cresols and guaiacol at temperatures between 350 and 500 ◦C. Dihy-
droxybenzenes were predominantly converted into phenol (ortho
isomer) and products with saturated ring (meta and para isomers),
cresols were converted to toluene, dimethylbenzenes and methyl
cyclohexane. Guaiacol was transformed into a mixture of phe-
nol, benzene, toluene and cyclohexane. The yields of benzene and
toluene increased with increasing reaction temperature. Pt/Al2O3
(1 wt.% of Pt) catalyst was reported to convert guaiacol at 300 ◦C into
phenol, catechol and methylcatechol without ring saturation [27].
In comparison with Pt/Al2O3 (1 wt.% of Pt), higher guaiacol con-
version can be obtained at the same reaction conditions with HY
zeolite without platinum [27], in both cases, catechols and dihy-
droxybenzenes were described as main products, other detected
products were identified as methyl guaiacols. On the other hand,
selective conversion of phenol to cyclohexanol was described over
Pd/C catalyst at a relatively low temperature 80 ◦C. Cyclohexanone
was detected as a reaction intermediate [28].
analysis. According to literature, only non-skeletal aluminum
should be removed at these conditions [30]. The last support used
in the study was mesoporous zeolite Beta prepared in VUAnCh and
denoted 3MF (Si/Al = 10.4, specific surface area 602 m2/g, micropo-
res volume 0.155 cm3/g, mesopores volume 0.361 cm3/g, medium
pore diameter 3.7 nm). The sample 3MF was prepared in a 5 l
autoclave by hydrothermal synthesis with tetramethylammoni-
umhydroxide as a template. A secondary template, a renewable raw
material, enabling formation of mesopores during hydrothermal
synthesis was used (for details see [31]). The removal of this
template during calcination led to creation of mesopores in the
structure.
Zeolites
were
then
impregnated
by
tetraammine-
platinum(II)nitrate to obtain catalyst with Pt content of 0.5 wt.%.
The impregnation was performed by a conventional impregnation
method using aqueous solution of the precursor. The support was
suspended in the aqueous solution of [Pt(NH3)4](NO3)2 and stirred
in a rotator–evaporator for 1 h. Then the solution was heated under
constant stirring until evaporated to dryness. After impregnation,
the samples were dried at 120 ◦C for 1 h.
Because of their crystalline structure and defined pore shapes
and sizes, zeolites are interesting catalyst supports as they allow
tuning catalytic properties and performance. Platinum is well-
known as highly active metal for hydrogenation reactions at
significantly lower temperatures than for example sulfided cata-
lysts. Hence zeolite Beta, having large pores formed by 12 rings
[29], was selected for low temperature transformation of pyrolysis
oil model compounds. The following phenolic model compounds
were selected to study the effects of the substituent types and
their mutual positions: phenol, o-, m- and p-cresols, guaiacol and
eugenol. Apart from studying the effects of substituents on hydro-
genation and deoxygenation, the influence of catalyst properties
(Si/Al ratio, porosity) and reaction conditions on catalyst activities
and selectivities were investigated as well.
2.2. Chemicals
Phenol (S. Aldrich, >98%), o-cresol (S. Aldrich, >98%), m-cresol
(S. Aldrich, >98%) and p-cresol (S. Aldrich, >98%), guaiacol (S.
Aldrich, >98%) and eugenol (S. Aldrich, >98%) were used as
lignin pyrolysis oil model compounds (Fig. 1) for hydrodeoxy-
genation/hydrogenation experiments. Hydrogen was supplied by
Air Products (99.9%). Isooctane (>98%) was supplied by Lachner
Chemicals.
2.3. Catalyst characterization
Si/Al ratios of the four zeolite Beta samples were obtained using
XRF analysis (Table 1). The same method was also used to confirm
Pt concentration on the catalysts (Table 1). Specific surface was
determined using nitrogen physisorption using BET and Dubinin
calculation methods (Table 2). Catalyst acidity was measured with
FTIR/pyridine adsorption (Table 3). Distributions of catalyst particle
2. Experimental
2.1. Catalysts
Table 1
Two commercial zeolites Beta (Zeolyst, CP 814E and CP 814C)
were used for studying the effect of Si/Al ratio on phenol and cresols
hydrogenation. They were denoted as Z-20 (CP 814 E; Si/Al = 12.7)
and Z-23 (CP 814C; Si/Al = 20.7). The Z-20 zeolite Beta has been
dealuminated using 1 M hydrochloric acid for 15 min at labora-
tory temperature to obtain zeolite Beta with a similar Si/Al ratio
to that of Z-23. The dealuminated zeolite Beta was denoted Z-
20D and had Si/Al ratio equals to 21.5 as determined by XRF
Si/Al ratios and platinum content measured using XRF technique and platinum
dispersion with platinum particle diameter – pD (by CO chemisorption).
Catalyst
Si/Al
Pt (wt.%)
Pt dispersion (%)
pD (nm)
Pt/Z-20
Pt-Z-23
Pt/3MF
Pt/Z-20D
12.7
20.7
10.4
21.5
0.50
0.48
0.48
0.43
20.6
15.5
34.6
11.4
5.5
7.3
3.3
9.9

40
J. Horácˇek et al. / Catalysis Today 204 (2013) 38–45
Table 2
Specific surface areas of catalysts.
Catalyst
Specific surface area (BET) (m2/g)
Specific surface area (Dubinin) (m2/g)
Micropore volume (cm3/g)
Pt/Z-20
Pt/Z-20D
Pt/Z-23
Pt/3MF
514
548
539
508
751
788
879
722
0.267
0.280
0.312
0.256
Table 3
Brønsted (B) and Lewis (L) sites concentration determined by FTIR/pyridine adsorp-
tion followed by desorption at 200 ◦
(CB concentration of Brønsted sites, CL
concentration of Lewis sites, and CAl concentration of aluminum in structure was
determined from concentration of acid sites).
3. Results and discussion
C
Direct comparison of catalysts for phenol hydrogenation and
deoxygenation is depicted in Fig. 2. The comparison of two par-
ent zeolites Beta (Z-20, Z-23) modified with platinum reveals a
higher catalytic activity of Pt/Z-23 with higher Si/Al ratio and hence
lower Brønsted as well as Lewis acidity (Table 3), as compared with
Pt/Z-20. As both catalysts have the same Pt content (Table 1), the
difference could be attributed to the smaller size of zeolite crystal
aggregates of Pt/Z-23 (2.8 vs. 10.2 ␮m, Table 4), i.e. a better acces-
sibility of active sites in Pt/Z-20. Furthermore, Pt/Z-23 has also a
higher specific surface area than Pt/Z-20 (Table 2). The mild dea-
lumination of Z-20 resulted in a significantly increased activity of
the Pt/Z-20D catalyst as it afforded virtually complete phenol con-
version in less than 15 min (Fig. 2). The physico-chemical data are
comparable with Pt/Z-23 and thus do not provide a clear evidence of
the cause of this substantial improvement, but it could be plausibly
explained by partial removal of aluminum due to dealumination,
particularly in extra framework positions [30], and hence improved
mass transfer of phenol in the pore structure. On the other hand, the
mesoporous Beta (Pt/3MF) exhibits catalytic activity comparable
with Pt/Z-20 (Fig. 2) even though the lowest mass transfer limita-
tions can be expected due to the mesoporosity. However, as can be
seen in Table 4, the mean catalyst particle size exceeds dramatically
the size of the other three catalysts (72.8 vs. <15 ␮m) and could be
the reason for the apparently lower catalyst activity. Moreover, it
can be observed that the initial catalytic activity (Fig. 2) is lower for
the catalysts with Si/Al ratio 10–12 than for those with Si/Al ratio
about 20–22 (Table 1). This might indicate that low acidity of a cata-
lyst is preferred. The Pt dispersion (Table 1) shows that the particle
size of Pt crystallites might have an impact on the observed con-
version as well. However, there is no obvious correlation between
the Pt dispersion on the one hand and phenol conversion on the
other hand. In fact, the highest phenol conversion after 20 min was
observed over Pt/Z-20D having the lowest dispersion (Fig. 2), while
the lowest conversion after the same time was found when using
the parent zeolite (Pt/Z-20) having the second lowest Pt dispersion
(Table 1).
Catalyst
CAl (mmol/g)
CB (mmol/g)
CL (mmol/g)
CB (%)
CL (%)
Pt/Z-20D
Pt/Z-20
Pt/Z-23
Pt/3MFa
0.40
0.55
0.40
0.33
0.16
0.24
0.20
0.19
0.12
0.16
0.10
0.07
57
60
66
72
43
40
34
28
a
Catalyst activation prior pyridine adsorption at 400 ◦C, the other catalysts were
activated only at 200 ◦C.
sizes were analyzed using laser scattering method (Mastersizer
2000) and the values are reported in Table 4.
2.4. Experiments
Hydrogenation/deoxygenation experiments were carried out
in a 500 ml batch stirred reactor (PPI). The reactor vessel was
filled with 1.00 g of not calcined catalyst and flushed with nitro-
gen and hydrogen. Before each experiment, catalyst was heated at
200 ◦C in presence of hydrogen. After reactor cooling to ambient
temperature, the vessel was filled with 300 ml of isooctane con-
taining 0.125 mol of substrate. Reactor free space was gradually
flushed first with nitrogen and then with hydrogen and pressur-
ized to 1 MPa. Reactor was then heated under slow stirring (less
than 50 rpm) for promoting heat transfer until the desired reac-
tion temperature (180 ◦C) was reached. After reaching the reaction
temperature, hydrogen pressure was set at 5 MPa and the stir-
rer rotation was increased to 1500 rpm. The reaction was carried
out under the set reaction conditions for 90 min. Samples of reac-
tion products were taken at defined reaction times and analyzed
off-line by GC. As constant amount of reactant and catalyst were
used in each experiment, the determined conversions correspond
to catalyst activity, i.e. rate of disappearance of reactant per mass
of catalyst.
In terms of selectivity, discernible differences can be found
among the zeolite-Beta-based platinum catalysts. Three main
products were observed that can be classified as hydrogenation
products (cyclohexanone and cyclohexanol) and a deoxygen-
ation product (cyclohexane). The other plausible deoxygenation
product, benzene, was detected only in trace amounts (<0.1%).
The presence of cyclohexanone in product samples and partic-
ularly the cyclohexanone yield profile (Fig. 2C) as a function of
reaction time or conversion indicate reaction pathway of ring
saturation with a parallel hydroxyl group into ketone change,
i.e. cyclohexenol to cyclohexanone rearrangement (Fig. 3). The
same reaction pathway has been described for phenol conversion
over sulfided catalysts [21]. The results indicate that cyclohex-
ane is a product of cyclohexanol deoxygenation, i.e. dehydration
over acid sites followed by hydrogenation of the olefin formed
as suggested recently by Lercher and coworkers [32], rather than
benzene hydrogenation (direct hydrodeoxygenation is a plausi-
ble pathway, but not a significant one in the case of phenol
deoxygenation).
For catalyst activity study, o-, m- and p-cresols were used for
studying the effect of isomer type, i.e. of the methyl position relative
to the hydroxyl group, on substrate hydrogenation and deoxygen-
ation. Eugenol was used as a model compound for pyrolysis oil
stabilization (allyl hydrogenation). Guaiacol represents a model
compound for studying the effect of methoxy substitution that is
typical in products of lignin pyrolysis. Samples of products and
substrates were analyzed by gas chromatography, using HP-PONA
column with temperature ramp starting at 35 ◦C for 15 min and
5 ◦C/min up to 250 ◦C.
Table 4
Catalyst mean particle size.
Catalyst particle size (␮m)
Pt/Z-20
10.20
Pt/Z-23
2.79
Pt/Z-20D
13.20
Pt/3MF
72.82

J. Horácˇek et al. / Catalysis Today 204 (2013) 38–45
41
A)
B)
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
0
15
30
45
60
75
90
105
0.0
0.2
0.4
0.6
0.8
1.0
Reaction time [min]
Conversion
0.5
0.4
0.3
0.2
0.1
0.0
1.0
0.8
0.6
0.4
0.2
0.0
C)
D)
0.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
Conversion
Conversion
Fig. 2. Conversion (A) and selectivities to cyclohexanol (B), cyclohexanone (C) and cyclohexane (D) over Pt/Z-23 (ꢀ) and Pt/Z-20 ( ), Pt/Z-20D ( ) and Pt/3MF ( ).
The high initial selectivity to cyclohexane (at ca. 5% conversion
about 40%), which is the ultimate product of phenol hydrogenation
and deoxygenation is rather surprising and can be attributed to
quick initial deactivation of the most active deoxygenation sites
(Fig. 2D). Cyclohexanol was the main reaction product for all the
Pt/Beta catalysts, but Pt/Z-20D, which yielded almost exclusively
cyclohexane.
The deoxygenation level (calculated as relative oxygen content
decrease; based on product composition) increased gradually with
increasing phenol conversion for all tested catalysts (Fig. 4). Pt/Z-23
converted phenol almost exclusively into cyclohexanol, deoxygen-
ation selectivity at 60% conversion was ca. 6%. The lower Al content
of Pt/Z-23, i.e. lower concentration of Brønsted and Lewis acid sites
(Table 3), resulted in very low deoxygenation selectivity in com-
parison with Pt/Z-20 over the whole conversion range (ca. 10 vs.
ca. 40%). The deoxygenation level did not exceed 15% in the case of
Pt/Z-23 even at complete conversion of phenol (Fig. 4). On the con-
trary, the data in Fig. 2D demonstrate that Pt/Z-20 exhibited fairly
stable formation of cyclohexane over the whole range of achieved
conversions over this catalyst (0–60%). This can be attributed to the
higher acidity of Pt/Z-20 in comparison with Pt/Z-23 together with
the larger size of zeolite crystal aggregates. Due to a higher dif-
fusional resistance the secondary reactions (i.e. deoxygenation of
cyclohexanol to cyclohexane) would be preferred over the primary
ones (i.e. aromatic ring saturation). This conclusion is further sup-
ported by the lower concentration of cyclohexanone in the product
mixture over Pt/Z-20 in comparison with Pt/Z-23 at comparable
phenol conversion, e.g. 4% vs. 13% at 60% conversion (Fig. 2C).
Dealumination of Z-20 causing an increase in Si/Al ratio (Table 1)
to a similar value of Si/Al as the one of Pt/Z-23 resulted in faster
ring hydrogenation accompanied by high deoxygenation activity
OH
O
OH
+ H2
+2 H2
+ H2
- H2O
+ H2
R
R
R
+3 H2
R
R
R= H or CH3
Fig. 3. Reaction pathways leading to (alkyl)naphthenes from (alkyl)phenols.

42
J. Horácˇek et al. / Catalysis Today 204 (2013) 38–45
100
80
60
40
20
0
0.0
0.2
0.4
0.6
Conversion
0.8
1.0
Fig. 4. Relative deoxygenation as function of phenol conversion. Pt/Z-23 (ꢀ) and Pt/Z-20 ( ), Pt/Z-20D ( ) and Pt/3MF ( ).
(Fig. 2). As a result of dealumination slight specific surface area
increase (Table 2) and decrease in concentration of Brønsted and
Lewis acid sites (Table 3) in comparison with the parent mate-
rial were observed. Despite the final values are relatively similar
to those of Pt/Z-23, Pt/Z-20D outperformed Pt/Z-23 both in terms
of activity and selectivity. In particular a striking difference was
observed in selectivity as at complete conversion Pt/Z-23 yielded
almost exclusively cyclohexanol (selectivity > 95%), while Pt/Z-20D
afforded only cyclohexane. Such a difference cannot be explained
by the relatively minor differences in acidity and can be thus
attributed to higher porosity of the dealuminated catalyst resulting
in better accessibility of active centers.
Platinum supported on the in-house prepared mesoporous zeo-
lite Beta support (Pt/3MF) with 75% conversion after 90 min of
reaction (Fig. 2A) exhibited a medium hydrogenation activity in
comparison with Pt/Z-23 (99% conversion) and Pt/Z-20 (61% con-
version). It also showed a higher ring hydrogenation activity than
Pt/Z-23 and a higher deoxygenation activity than Pt/Z-20. Based on
its physico-chemical properties, a performance closer to the Pt/Z-
20D catalyst would be expected. The significant difference from
this catalyst can be most probably assigned to the dramatically
larger size of catalyst particles and the resulting diffusion limitation
effects (72.8 vs. <15 ␮m, Table 4).
and Pt/Z-20 than the reactivity of phenol. Predominant reac-
tion products were the corresponding methylcyclohexanols and
methylcyclohexanones. In contrast to phenol conversion, direct
hydrodeoxygenation yielding toluene was observed in addition to
the aromatic ring hydrogenation (Fig. 6). The selectivity to toluene
was pronounced particularly at low conversions suggesting that
this reaction pathway is parallel to the hydrogenation route and the
active centers might be quickly deactivated. Toluene formation can
be explained as result direct hydrodeoxygenation without aromatic
ring hydrogenation as described in literature for high tempera-
ture conversion over aluminum supported catalysts [21,26]. The
difference between cresols and phenol could be attributed to the
presence of methyl substituent, which makes the carbon–oxygen
bond weaker and facilitates thus its rupture, i.e. direct deoxygen-
ation. Moreover, the electronic effects of the methyl substituent are
further reflected in the initial selectivities to toluene. The lowest
selectivity to toluene was observed in the case of m-cresol (Fig. 6),
where the methyl substituent has a weaker electronic effect on the
hydroxyl group than the methyl substituent in p- or o-position.
The lower initial selectivity to toluene in o-cresol deoxygenation
(in comparison with p-cresol) can be attributed to steric effects
affecting the adsorption of o-cresol, but not of p-cresol (Fig. 6).
Only insignificant amounts of the ultimate product, i.e. methyl-
cyclohexane, were obtained. This indicates that hydrogenation of
toluene was very limited. Moreover, it also suggests that methylcy-
clohexanols were not deoxygenated to afford methylcyclohexane.
Similar behavior over Pt/Z-23 was observed also for phenol as it
yielded predominantly cyclohexanol. While Pt/Z-20 showed higher
The effect of o-, m- and p-substitution on hydrogenation and
deoxygenation activity of zeolite-Beta-supported platinum cata-
lysts was studied using individual cresol isomers. The reactivity
of cresol isomers decreased in the order m- > o- > p- over Pt/Z-23
(Fig. 5). In general, the reactivity of cresols was lower over Pt/Z-23
0.60
0.60
A)
B)
0.50
0.40
0.30
0.20
0.10
0.00
0.50
0.40
0.30
0.20
0.10
0.00
0
15
30
45
60
75
90
105
0
15
30
45
60
75
90
105
Reaction time [min]
Reaction time [min]
Fig. 5. Effect of methyl position on compound reactivity (A) (Pt/Z-23); ( ) p-cresol, ( ) m-cresol, ( ) o-cresol and conversion into methylcyclohexanols (B).

J. Horácˇek et al. / Catalysis Today 204 (2013) 38–45
43
A)
B)
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
0
0.2
0.4
0.6
0
0.2
0.4
0.6
Conversion
conversion
C)
D)
0.6
0.5
0.3
0.2
0.0
0.6
0.5
0.3
0.2
0.0
0.0
0.2
0.4
0.6
0.0
0.2
0.4
0.6
Conversion
Conversion
E)
F)
0.6
0.5
0.3
0.2
0.0
0.6
0.5
0.3
0.2
0.0
0.0
0.2
0.4
0.6
0.0
0.2
0.4
0.6
Conversion
Conversion
Fig. 6. Effect of methyl position on selectivity to methylcyclohexanol over Pt/Z-23 (A) and Pt/Z-20 (B), to toluene over Pt/Z-23 (C) and Pt/Z-20 (D), and to methylcyclohexanones
over Pt/Z-23 (E) and Pt/Z-20 (F). Feedstocks: ( ) p-cresol, ( ) m-cresol, and ( ) o-cresol.
selectivity to deoxygenation products in the case of phenol, the
same was not observed in the case of cresols. However, their con-
versions were rather low ca. 10%. Using the analogy with phenol
hydrodeoxygenation discussed above, it can be assumed that the
ketones were formed along a similar pathway as cyclohexanone
from phenol. This is supported by the experimental data.
The comparison of phenol and guaiacol over Pt/Z-23 has shown
an important stabilization effect of the methoxy group sub-
stitution for deoxygenation, as the only detected product was
2-methoxycyclohexanol. Moreover, the ring hydrogenation was
significantly suppressed by the methoxy group in comparison
with phenol (no substitution) and o-cresol (methyl substitution)
as depicted in Fig. 7. Presence of another substituent on aro-
matic ring hinders the aromatic ring saturation due to affecting
its adsorption on active Pt sites. At the same time, deoxygenation
rate is also affected and with increasing substituent molecular
weight lower deoxygenation level can be awaited. Crucial for
reactivity of differently substituted phenols in Pt-modified zeo-
lites are adsorption modes and transport parameters affected by
their molecular dimensions. Estimates of the maximum molecular
dimension (kinetic diameter) of the investigated model compounds
are given in Table 5. Molecule size effect on aromatic ring hydro-
genation can be important particularly in zeolites. The maximum
diameter (dmax) increases in the order phenol < o-cresol < guaiacol

44
J. Horácˇek et al. / Catalysis Today 204 (2013) 38–45
100
80
60
40
20
0
2
1.6
1.2
0.8
0.4
0
0
15
30
45
60
75
90
105
Reaction time [min]
Fig. 7. Aromatic ring hydrogenation affected by 2-phenol substitution, ( ) cyclohexanol (from phenol), ( ) 2-methylcyclohexanol (+methyl substitution, from o-cresol) (
) 2-methoxycyclohexanol (+methoxy group, from guaiacol).
100
75
50
25
0
0
15
30
45
60
75
90
105
Reaction time [min]
Fig. 8. Allyl group hydrogenation (eugenol (ꢁ) to 2-methoxy-4-propylphenol ( ))).
which agrees with their observed reactivity, i.e. the reactivity over
Pt/Z-23 decreased with the increasing molecule diameter (dmax).
The increasing molecule diameter affects the mobility of molecules
in catalyst pores.
hydrogenation of allyl group in eugenol can be attributed to pore-
mouth catalysis (the allyl chain entering platinum at pore mouth
where the aromatic ring cannot adsorb properly). Alternatively it
can indicate that the three substituents of the aromatic ring prevent
its adsorption on platinum crystallites which ultimately excludes
the aromatic ring saturation.
Stabilization of allyl groups of methoxy compounds, such
as eugenol is essential for pyrolysis oil stabilization before its
hydrogenation/deoxygenation treatment. In agreement with the
experiments performed with guaiacol the aromatic ring of eugenol
was not hydrogenated over Pt/Z-23. Only the allyl group was hydro-
genated to yield 2-methoxy-4-propylphenol. The hydrogenation
of the double bond was very facile and selective, affording 100%
yield of 2-methoxy-4-propylphenol in just 10 min (Fig. 8). In con-
trast to cresols and phenol, aromatic ring hydrogenation was not
observed, which can be explained by the limited access of eugenol
or 2-methoxy-4-propylphenol to active platinum sites due to their
bulkiness. This is in line with guaiacol transformation where the
yield of products having saturated aromatic ring was <2%. The
4. Conclusions
Beta-zeolites-supported platinum catalysts were found to be
suitable for hydrogenation and deoxygenation of phenol. The
catalysts yielded either cyclohexanol (Pt/Z-23) or cyclohexane
(Pt/Z-20D) depending on their preparation. The mildly dealumi-
nated catalyst allowed better accessibility of active sites and as
a result fast conversion of phenol was achieved in comparison
with the not-dealuminated catalysts. Cresols, methyl substituted
phenols, were less reactive than phenol and also exhibited lower
selectivity to the ultimate product of combined hydrogenation and
deoxygenation. Methoxy group was shown to hinder both hydro-
genation and deoxygenation.
Table 5
Maximum molecular dimension of studied phenolics as estimated by using Chems-
ketch (ACD Labs) software.
Compound
dmax (Å)
Acknowledgments
Phenol
6.06
6.10
6.48
6.94
6.68
8.97
o-Cresol
m-Cresol
p-Cresol
Guaiacol
Eugenol
ˇ
The financial support by the Czech Science Foundation (GACR,
project P106/10/1733) is gratefully acknowledged. The project
(P106/10/1733) is being carried out in the UniCRE centre
(CZ.1.05/2.1.00/03.0071).

J. Horácˇek et al. / Catalysis Today 204 (2013) 38–45
45
References
[17] A. Gayubo, A. Aguayo, A. Atutxa, B. Valle, J. Bilbao, Journal of Chemical Technol-
ogy and Biotechnology 80 (2005) 1244–1251.
[18] S.R. Katikaneni, J. Adjaye, N. Bakhshi, Energy and Fuels 9 (1995) 1065–1078.
[19] W.W. Zmierczak, J.D. Miller, US Patent Application Publication US (2008)
2008/0050792 A1.
[20] J.S. Shabtai, W.W. Zmierczak, E. Chornet, US Patent (1999) 5 959 167.
[21] O.I. Senol, E.-M. Ryymin, T.-R. Viljava, A.O.I. Krause, Journal of Molecular Catal-
ysis A: Chemical 277 (2007) 107–112.
[22] E.-M. Ryymin, L. Maija, M.L. Honkela, T.-R. Viljava, A.O.I. Krause, Applied Catal-
ysis A: General 389 (2010) 114–121.
[23] E. Furimsky, F.E. Massoth, Catalysis Today 52 (1999) 381–495.
[24] W. Baldauf, U. Balfanz, M. Rupp, Biomass and Bioenergy 7 (1994) 237–244.
[25] E. Laurent, B. Delmon, Industrial and Engineering Chemistry Research 32 (1993)
2516–2524.
[26] R.K.M.R. Kallury, W.M. Restivo, T.T. Tidwell, D.G.B. Boocock, A. Crimi, J. Douglas,
Journal of Catalysis 96 (1985) 535–543.
[27] T. Nimmanwudipong, R.C. Runnebaum, D.E. Block, B.C. Gates, Energy and Fuels
25 (2011) 3417–3427.
[28] C. Zhao, Y. Kou, A.A. Lemonidou, X. Li, J.A. Lercher, Angewandte Chemie Inter-
national Edition 48 (2009) 3987–3990.
[29] M.W. Deem, J.M. Newsam, Journal of the American Chemical Society 114 (1992)
7189–7198.
[1] G. Gellerstedt, G. Henriksson, in: M. Belgacem, A. Gandini (Eds.), Monomers,
Polymers and Composites from Renewable Resources, Elsevier, Amsterdam,
2008, pp. 201–224.
[2] L. Petrus, M. Noordermeer, Green Chemistry 8 (10) (2006) 861–867.
[3] A. Corma, S. Iborra, A. Velty, Chemical Reviews 107 (6) (2007) 2411–2502.
[4] P. Mäki-Arvela, B. Holmbom, T. Salmi, D. Murzin, Catalysis Reviews – Science
and Engineering 49 (3) (2007) 197–340.
[5] P. Gallezot, Catalysis Today 121 (2007) 76–91.
[6] M.J. Climent, A. Corma, S. Iborra, Green Chemistry 13 (3) (2011) 520–540.
ˇ
[7] D. Kubicˇka, I. Kubicˇková, J. Cejka, Catalysis Reviews – Science and Engineering
(2012), http://dx.doi.org/10.1080/01614940.2012.685811.
[8] X. Tong, Y. Ma, Y. Li, Applied Catalysis A: General 385 (1–2) (2010) 1–13.
[9] J. Zakzeski, P. Bruijnincx, A. Jongerius, B. Weckhuysen, Chemical Reviews 110
(6) (2010) 3552–3599.
[10] J. Horácˇek, D. Kubicˇka, F. Homola, I. Kubicˇková, Catalysis Today 179 (2012)
191–198.
[11] P. Mäki-Arvela, I. Anugwom, P. Virtanen, R. Sjöholm, J. Mikkola, Industrial Crops
and Products 32 (3) (2010) 175–201.
[12] G. Schaub, A. Vetter, Chemical Engineering and Technology 31 (2008) 721–729.
[13] S. Yaman, Energy Conversion and Management 45 (5) (2004) 651–671.
[30] C. Márquez-Alvarez, A.B. Pinar, R. García, M. Grande-Casas, J. Pérez-Pariente,
Topics in Catalysis 52 (2009) 1281–1291.
[31] V. Tokarová, et al., submitted to Catalysis letters.
[32] C. Zhao, J. He, A.A. Lemonidou, X. Li, J.A. Lercher, Journal of Catalysis 280 (2011)
8–16.
[14] A.
http://dx.doi.org/10.1016/j.biombioe.2011.01.048.
[15] D. Bulushev, J. Ross, Catalysis Today 171 (1) (2011) 1–13.
Bridgwater,
Biomass
and
Bioenergy
(2011),
[16] M.F. Demirbas, M.F. Balat M., Journal of Scientific and Industrial Research 66
(2007) 797–804.