Hydrodeoxygenation of Phenol over Zirconia-Supported Catalysts: The Effect of Metal Type on Reaction Mechanism and Catalyst Deactivation

Article abstract of DOI:10.1002/cctc.201700047

This work aims at investigating the effect of the type of metal (Pt, Pd, Rh, Ru, Cu, Ni, Co) on the performance of ZrO₂-supported catalysts for the hydrodeoxygenation of phenol in the gas phase at 573 K and 1 atm. Two different reaction pathways take place depending on the type of the metal. For Pt/ZrO₂ and Pd/ZrO₂ catalysts, phenol is mainly tautomerized, followed by hydrogenation of the C=C bond of the tautomer intermediate formed, producing cyclohexanone and cyclohexanol. By contrast, the direct dehydroxylation of phenol followed by hydrogenolysis might also occur over more oxophilic metals such as Rh, Ru, Co, and Ni. In addition to the metals, the oxophilic sites of this support represented by Zr⁴⁺ and Zr³⁺ cations near the perimeter of the metal particles also increased the selectivity to deoxygenated products. All catalysts were significantly deactivated mainly owing to the growth of metal particle size and the decrease in the density of oxophilic sites.

Full text of DOI:10.1002/cctc.201700047

Accepted Article
Title: Hydrodeoxygenation of phenol over zirconia supported catalysts.
The effect of metal type on reaction mechanism and catalyst
deactivation
Authors: Camila Abreu Teles, Raimundo Crisostomo Rabelo-Neto,
Gary Jacobs, Burtron H Davis, Daniel E Resasco, and Fabio
Bellot Noronha
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201700047
Link to VoR: http://dx.doi.org/10.1002/cctc.201700047
A Journal of
www.chemcatchem.org

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
Hydrodeoxygenation of phenol over zirconia supported catalysts.
The effect of metal type on reaction mechanism and catalyst
deactivation
Camila A. Teles[a],[b], Raimundo C. Rabelo-Neto[b], Gary Jacobs[c], Burtron H. Davis[c], Daniel E. Resasco
[d], Fábio B. Noronha*[a],[b]
Abstract: This work aims at investigating the effect of the type of
metal (Pt, Pd, Rh, Ru, Cu, Ni, Co) on the performance of ZrO2
supported catalysts for the hydrodeoxygenation of phenol in the gas
phase at 573 K and 1 atm. Two different reaction pathways take place
investigated for the HDO reaction of different model molecules.[2]
In spite of the numerous studies on HDO of phenol, there have
been only few studies about the performance of different metals
for this reaction under the same conditions.
depending on the type of the metal. For Pt/ZrO2 and Pd/ZrO2 catalysts, Mortensen et al.[2e] studied the performance of Ru/C, Pd/C and
phenol is mainly tautomerized, followed by hydrogenation of the C=C
bond of the tautomer intermediate formed, producing cyclohexanone
and cyclohexanol. By contrast, the direct dehydroxylation of phenol
followed by hydrogenolysis might also occur over more oxophilic
metals such as Rh, Ru, Co and Ni. In addition to the metals, the
oxophilic sites of this support represented by Zr4+ and Zr3+ cations near
the perimeter of the metal particles also increased the selectivity to
deoxygenated products. All catalysts significantly deactivated mainly
due to the growth of metal particle size and the decrease in the density
of oxophilic sites.
Pt/C catalysts for the HDO of phenol in liquid phase and
correlated, using the work of Norskov et al.[3] as a basis, the
oxophilicity of the metals (Ru > Pd > Pt) with their deoxygenation
activity. Moreover, selectivity depended on the metal as well, with
cyclohexane favored on Ru, and cyclohexanol mainly produced
on Pd and Pt.
Teles et al. [4] investigated different metals supported on silica for
HDO of phenol in the gas phase and obtained a correlation
between the oxophilicity of the metal and deoxygenation activity.
Silica supported Pt, Pd, and Rh tended to produce more
hydrogenated products such as cyclohexanone and
cyclohexanol; on the other hand, Co, Ni, and Ru favored the
production of methane as well as C5-C6 hydrocarbons, due to their
greater hydrogenolysis activity. From these results, the authors
proposed two different reaction pathways to explain the
differences in reactivity. With Pt, Pd, and Rh, the keto tautomer
of phenol was hydrogenated at the C=C bond to produce
Introduction
The HDO of phenol reaction has been considered a bifunctional
mechanism,
requiring
a
metallic
site
for
hydrogenation/dehydrogenation combined with an acid site of the
support for dehydration.[1] Thus, understanding the contribution of
each type of site in the reaction mechanism is essential for the
design of appropriate catalysts.
cyclohexanone
and
cyclohexanol.
However,
direct
dehydroxylation of phenol and hydrogenolysis were deemed to
occur over the more oxophilic metals (i.e., Ru, Co and Ni).
Several catalysts based on Ni, Co, Fe, Cu, Ni-Cu, Pt, Rh, Pd and
Ru dispersed on different supports (e.g., Al2O3, SiO2, CeO2, ZrO2,
CeZrO2, V2O5, C, Mg2Al2O4, TiO2, HZSM5 and HY) have been
The type of the support also plays an important role in the activity,
selectivity and stability of the catalysts for HDO of phenol. For
example, the role of support on Ni catalysts was investigated for
liquid phase HDO of phenol by Mortensen et al.[2e] Some metal
oxide supports investigated included Al2O3, CeO2, MgAl2O4, SiO2,
ZrO2, as well as binary systems like V2O5/ZrO2 and CeO2-ZrO2.
Carbon, which is typically a relatively inactive support, was also
examined. Interestingly, ZrO2 as a support for Ni particles offered
the highest selectivity to cyclohexane. Defect sites (e.g.,
coordinatively unsaturated zirconia cation sites in the oxide
surface) were suggested to stabilize the phenoxide ion in
interaction with Ni, thus promoting hydrogenation of the aromatic
ring. Cyclohexanone formation was favored, which rapidly
converted to cyclohexanol and subsequently dehydrated to
[a]
C. A. Teles, R. C. Rabelo-Neto, F. B. Noronha*, Catalysis Division,
National Institute of Technology, Av. Venezuela 82, 20081-312, Rio de
Janeiro, RJ 20081-312, Brazil. Email: fabio.bellot@int.gov.br
[b]
C. A. Teles, F. B. Noronha, Chemical Engineering Department,
Military Institute of Engineering, Praça Gal. Tiburcio 80, 22290-270, Rio de
Janeiro, RJ, Brazil
[c]
University of Kentucky, 2540 Research Park Dr., Lexington, KY 40511, USA.
[d] D. E. Resasco, Center for Biomass Refining, School of Chemical,
G. Jacobs, B. H. Davis, Center for Applied Energy Research,
Biological, and Materials Engineering, University of Oklahoma, 100 East
Boyd St., Norman, OK 73019, USA.
cyclohexene.
Cyclohexene, in turn, was hydrogenated to
cyclohexane. We have also studied the effect of the support on
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
three reduction peaks at 370, 403 and 428 K, which originates
from the reduction of different RuOx species.[8] The Cu/ZrO2
catalyst showed two reduction peaks at 423 and 473 K
corresponding to the reduction of CuO dispersed species in
weaker interaction with the support and crystalline CuO
respectively.[9] The H2-TPR profile of the Ni/ZrO2 catalyst showed
a small peak at 463 K which was attributed to residual nitrate
decomposition. The peak centered at about 588 K and the
shoulder at 673 K corresponded to the reduction of bulk NiO and
species in strong interaction with the support. [10] Two peaks were
observed in the TPR profile of Co/ZrO2, which are due to the two
step reduction of cobalt oxide (Co3O4  CoO Co).[11]
Hydrogen uptakes determined from the TPR profiles are reported
in Table 2. Ru/ZrO2, Cu/ZrO2 and Ni/ZrO2 catalysts showed a
complete reduction of the respective metal oxide, whereas
Co/ZrO2 catalyst had a reduction degree of only 40%. Pt/ZrO2,
Pd/ZrO2 and Rh/ZrO2 catalysts exhibited a hydrogen uptake
higher than that corresponding to the complete reduction of the
supported metal oxide, indicating that zirconia can be partially
reduced, consistent with previously reported XANES results. [12]
The dehydration of cyclohexanol has been used as a model
reaction for the titration of oxophilic sites. [12,13] In our work, the
results obtained for zirconia supported catalysts are reported in
Table 1. The rate of cyclohexanol dehydration followed the order:
Pd/ZrO2  Rh/ZrO2  Pt/ZrO2 > Ni/ZrO2 > Ru/ZrO2 > Co/ZrO2.
Recently, the number of acid sites of Pd/ZrO2 catalysts with
different morphologies was determined by temperature-
programmed desorption of NH3 (NH3-TPD) and the cyclohexanol
dehydration reaction. Both techniques revealed that Pd supported
over ZrO2 with tetragonal structure exhibited the highest density
of acid sites. Therefore, the result obtained in this work agrees
with the fraction of tetragonal phase present in the zirconia
support.
the mechanism of HDO of phenol in the gas phase over Pd
[5]
catalysts supported on SiO2, Al2O3 and ZrO2. Selectivity was
found to depend heavily upon the type of support. Over Pd/SiO2
and Pd/Al2O3 catalysts, the formation of cyclohexanone was
favored, whereas benzene was promoted over Pd/ZrO2 catalyst.
These selectivity differences were ascribed to the oxophilicity of
the support, whereby coordinatively unsaturated cations near the
periphery of metal particles played an important role. The oxygen
atom of the keto tautomer intermediate of phenol interacted more
strongly with the support,
a
configuration that favors
hydrogenation of the carbonyl group by the metal at the metal-
support junction. This, in turn, increased the selectivity to
benzene once the hydrogenated intermediate was dehydrated.
This work aims at investigating the effect of the metal type on the
HDO of phenol over supported catalysts. A systematic study was
performed by carrying out the HDO of phenol over ZrO2 supported
metals such as Co, Cu, Ni, Pd, Pt, Rh, and Ru. Zirconia was
selected as one of the best supports for deoxygenation due to the
presence of oxophilic sites. Catalytic testing and diffuse
reflectance infrared Fourier transform spectroscopy (DRIFTS)
measurements were carried out to investigate the main reaction
pathways on the different metals. The mechanism of catalyst
deactivation was also studied using a combination of Raman
spectroscopy and model reactions.
Results and Discussion
Catalyst characterization
Table 1 shows the metal loadings, surface areas, metal dispersion
and rates of HDO of phenol for the various samples. The metal
loading of the catalysts was close to the nominal values. The
surface areas of all samples were quite similar and around 68 –
78 m2 g-1. The metal dispersions of the catalysts varied from 2 to
62% depending on the metal. Table 1 also lists the metal particle
size calculated from the dispersion values considering d = 1/D.
X-ray diffraction patterns for allcatalysts are shown in Fig. S1. The
diffractograms of all catalysts showed lines characteristic of a
mixture of monoclinic (JCPDS 37-1484) and tetragonal zirconia
(JCPDS 17-0923). The procedure outlined by Khaodee et al. [6]
was utilized to estimate the fraction of each crystalline phase, and
the results are listed in Table 1. The monoclinic phase is the
dominant one, ranging from 82 - 89% for all catalysts. For Pt, Pd,
Rh and Ru catalysts, the lines characteristic of the respective
metal oxides were not detected, which is likely due to either their
superposition with diffraction lines of the zirconia phases or the
low metal loading of the samples. The diffractograms of Cu/ZrO2,
Ni/ZrO2 and Co/ZrO2 catalysts also exhibited lines corresponding
to CuO, NiO and Co3O4.
Fig. S2 shows the TPR profiles of zirconia support and catalysts.
The Pt/ZrO2, Pd/ZrO2 and Rh/ZrO2 catalysts showed H2-
consumption at 441, 351 and 380 K respectively, due to reduction
of PtO2, PdO and Rh2O3 oxides.[7] The Ru/ZrO2 catalyst exhibited
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
Table 1. Pd content, specific surface area, calculated Pd dispersion, reaction rate for dehydration of cyclohexanol (DHA) and for HDO of phenol over zirconia
supported catalysts.
Tetragonal
Rate of DHA[c]
wt%
BET
dp (nm)
Metal
Reaction rate[e]
catalysts
Metal[a]
dispersion[d] (%)
phase (%)[b]
(µmolg-1catmin-1)
(m2g-1)
(mmol g-1Me min-1)
72
78
72
71
81
68
77
17
18
18
13
12
16
10
28.8
29.6
29.2
9.2
2.7
37
41.7
28.6
104.4
3.5
Pt/ZrO2
Pd/ZrO2
Rh/ZrO2
Ru/ZrO2
Cu/ZrO2
Ni/ZrO2
Co/ZrO2
0.96
0.96
0.91
5.3
2.6
38
1.6
62
8 (41)f
12.5 (2.4)
7.1
-
-
0.0
4.9
17.2
6.0
25
50
4
2
4.4
9.7
1.0
9.4
[a] Measured by XRF
[b] Calculated by using the data from XRD
[c] Reaction rate for dehydration of cyclohexanol
[d] Metal dispersion calculated by H2 chemisorption (Pt, Rh, Ru, Ni, Co) and cyclohexane dehydrogenation reaction (Pd)
[e] Reaction rate calculated from a phenol conversion of around 10%
[f] Ru dispersion of 1%Ru/ZrO2 catalyst
significantly increased in the presence of zirconia for all catalysts.
For instance, the reaction rate for Rh/ZrO2 was 28-fold higher than
that for Rh/SiO2. It is important to stress that the zirconia support
did not exhibit activity under this reaction conditions. These
results are in agreement with our previous work that revealed an
important role of the support on the activity of Pd supported
catalysts for HDO of phenol. [5] These selectivity differences were
explained in terms of the oxophilicity of the support, whereby
coordinatively unsaturated cations near the perimeter of metal
particles participated in the reaction. Therefore, the activity of
zirconia supported catalyst involves not only the intrinsic activity
of the metal but also the type of the support.
The product distributions were also remarkably different
depending on the metal (Fig. 1). For Pt/ZrO2, Pd/ZrO2 and Ni/ZrO2
catalysts, cyclohexanone (ONE) and benzene were the main
products formed at W/F < 0.15 h. Significant yields of benzene
were only obtained at higher W/F. Small amounts of cyclohexanol
(OL) and C12 hydrocarbons such as biphenyl and
cyclohexylbenzene were also observed on these catalysts. For
Ni/ZrO2 catalyst, a significant formation of CH4 was also observed.
On the other hand, benzene was the dominant product over the
entire range of W/F over the Rh/ZrO2 catalyst, with only small
amounts of CH4 and C5-C6 hydrocarbons (n-hexane, 2-metyl-
butane, cyclohexene, cyclohexane) being formed. Besides
benzene, significant formation of methane and C5-C6
hydrocarbons was also observed for the Ru/ZrO2 catalyst. For
Co/ZrO2 catalyst, methane, benzene, o-cresol and C12
hydrocarbons were the products obtained.
Table 2. Hydrogen consumption and reduction degree calculated from the TPR
profiles
Theoretical H2
consumption
(cm3 g-1)
Experimental
H2 consumption
(cm3 g-1)
Reduction
degree (%)
Catalyst
2.21
3.15
142
184
118
93
Pt/ZrO2
Pd/ZrO2
Rh/ZrO2
Ru/ZrO2
Cu/ZrO2
Ni/ZrO2
Co/ZrO2
2.02
5.94
3.72
7.06
32.83
17.27
37.01
47.65
35.24
14.41
34.16
19.25
100
92
40
HDO of phenol over Me/ZrO2 catalysts
The phenol conversion and product yield as a function of W/F at
573 K over all catalysts are shown in Fig. 1. All catalysts were
active for HDO of phenol except Cu/ZrO2, which did not display
any activity at the highest W/F used. The zirconia support was
also tested and was not found to be active under the reaction
conditions used. Thus, the catalysis likely occurs on the surface
of metal particles and/or at the junction between metal particles
and the support. Table 1 lists the deoxygenation reaction rates
calculated using the data obtained under differential reaction
conditions (i.e., conversion of around 10%). Rh/ZrO2 catalyst
exhibited the highest reaction rate, which followed the order:
Rh/ZrO2 > Pt/ZrO2 > Pd/ZrO2 > Ni/ZrO2 > Ru/ZrO2 > Co/ZrO2.
Recently, we studied the effect of metals for HDO of phenol on
SiO2 supported catalysts. [4] Silica was used as an inert support
that does not exhibit activity for this reaction. The following order
of reaction rates (mmol.g-1Me.min-1) was observed: Rh/SiO2 (3.7)
> Ru/SiO2 (2.1) > Pd/SiO2 (1.3)  Pt/SiO2 (1.0) > Co/SiO2 (0.6) ~
Ni/SiO2 (0.5). This reinforces that the HDO of phenol is strongly
influenced by the choice of the metal. However, the comparison
between the reaction rates for the catalysts supported on silica
and zirconia reveals that ZrO2 changed the activity order reported
for SiO2 supported catalysts. Furthermore, the reaction rates
In order to compare the remarkable differences in product
distribution over all catalysts, the product selectivities obtained at
a similar level of conversion (around 10%) are reported in Table
3. Cyclohexanone and cyclohexanol were the main products
formed over Pt/ZrO2 and Pd/ZrO2 catalysts, with significant
formation of benzene. The highest selectivity to benzene was
observed for Rh/ZrO2 catalyst that also produced C5-C6
hydrocarbons and methane. Ru/ZrO2, Ni/ZrO2 and Co/ZrO2
catalysts show high selectivities to methane, with the Ru catalyst
also producing C5-C6 hydrocarbons. Significant formation of o-
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
cresol and small amounts of C12 hydrocarbons were also
observed for Co/ZrO2. The selectivity to deoxygenated products
(benzene, methane and C5-C6 hydrocarbons) followed the order:
Ru/ZrO2 > Rh/ZrO2 >> Ni/ZrO2 ≈ Pd/ZrO2 > Pt/ZrO2 ≈ Co/ZrO2.
(D)
Conversion
Benzene
ONE
70
60
50
40
30
20
10
0
OL
C12 Hydrocarbons
C5-C6 Hydrocarbons
CH4
Conversion
Benzene
ONE
(A)
70
60
50
40
30
20
10
0
OL
C12 Hydrocarbons
C5-C6 Hydrocarbons
0.0
0.1
0.2
0.3
0.4
0.5
W/F (h)
Conversion
Benzene
ONE
70
(E)
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16
60
50
40
30
20
10
0
OL
W/F (h)
C12 Hydrocarbons
C5-C6 Hydrocarbons
CH4
70
60
50
40
30
20
10
0
Conversion
Benzene
ONE
(B)
OL
C12 Hydrocarbons
C5-C6 Hydrocabons
0.00
0.05
0.10
0.15
0.20
0.25
0.30
W/F (h)
Conversion
Benzene
ONE
70
(F)
0.00 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24
60
50
40
30
20
10
0
OL
W/F (h)
o-Cresol
C12 Hydrocarbons
C5-C6 Hydrocarbons
CH4
Conversion
Benzene
ONE
(C)
70
60
50
40
30
20
10
0
OL
C12 Hydrocarbons
C -C Hydrocarbons
5
6
CH
4
0.0
0.1
0.2
0.3
0.4
0.5
0.6
W/F (h)
Figure 1. Phenol conversion and product yield as a function of W/F over (A)
Pt/ZrO2, (B) Pd/ZrO2, (C) Rh/ZrO2, (D) Ru/ZrO2, (E) Ni/ZrO2 and (F) Co/ZrO2.
Reaction conditions: 573 K, atmospheric pressure, H2/phenol molar ratio= 60.
ONE: Cyclohexanone; OL: Cyclohexanol.
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
W/F (h)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
Table 3. Product distributions for HDO of phenol at 573 K and atmospheric
on the Lewis acid sites of the support. Only small amounts of
benzene were detected for all catalysts regardless of the metal.
pressure over zirconia supported catalysts.
DRIFTS experiments of HDO of phenol over zirconia
supported
Pt/ZrO2 Pd/ZrO2 Rh/ZrO2
Ru/ZrO2
Ni/ZrO2 Co/ZrO2
0.01
9.3
0.02
8.2
0.01
8.8
0.07 (0.15) 0.02
0.08
10.0
W/F
Conversion
(%)
8.3 (12.7)
11.4
Fig. 2 shows the DRIFTS spectra obtained upon exposure of the
zirconia supported catalysts to a phenol/H2 mixture at different
temperatures. At 323 K, adsorption of phenol led to bands
characteristic of adsorbed phenoxy species in either the
monodentate or bidentate mode [16], including: (CHring) 3024-
3068 cm-1, (CCring) 1593-1595 cm-1, (CCring) 1489-1491 cm-1,
and (CO) 1292-1294 cm-1 for monodentate, with a shoulder(s) at
1225-1265 cm-1 for bidentate. These bands were formed by
adsorption of phenol at metal oxide cationic sites. Intensities were
highest for the Pt, Pd, and Ni supported catalysts, followed by Rh,
and then Ru. Based on the literature [5], bands corresponding to
adsorbed cyclohexanol were also observed as follows: a(CH2)
aliphatic CH, 2933-2937 cm-1; s(CH2) stretching, 2856-2860 cm-
1, CH2 scissoring, 1452-1454 cm-1; CH2 wagging in the alcohol,
1344-1365 cm-1; and (CO) 1070-1167 cm-1. These were most
significant for the Pt, Rh, and Ru catalysts, followed by the Pd
catalyst, with only low intensity bands being observed for Ni.
Features between 1620 and 1740 cm-1 were detected for Pt, Pd,
Rh, and Ru catalysts, consistent with (C=O) stretching bands,
suggesting a small amount of adsorbed cyclohexanone species
was also present. Interestingly, the sum of all band intensities was
significantly lower for the Ru catalyst, and somewhat lower for Ni
as well.
Selectivity (%)
56.6
12.8
27.4
52.6
1.7
22.0
--
5.6 (6.9)a
0.8 (1.7)
37.7
10.9
16.0
5.7
ONE
OL
43.2
63.4
49.7 (74.2) 26.7
16.5
Benzene
o-cresol
C5-C6
hydrocarbons
C12
--
1.8
--
0.1
--
9.1
1.1 (-)
14.9 (8.8)
5.6
0.8
38.0
3.1
1.4
--
2.4
--
0.3
5.2
0.6 (-)
2.8
5.4
hydrocarbons
CH4
27.3 (8.4)
15.4
15.4
ONE: cyclohexanone; OL: cyclohexanol
C5-C6 hydrocarbons: cyclohexane; cyclohexene; 2-methyl-butane; n-hexane
C12 hydrocarbons: biphenyl; dibenzofurane; cyclohexylbenzene; pentylbenzene
aSelectivities for HDO of phenol over 1%Ru/ZrO2 catalyst with smaller Ru
particle size.
In order to determine the reaction mechanism of HDO of phenol,
tests for the conversion of intermediate products (e.g.,
cyclohexanol and cyclohexanone) over all catalysts and supports
under reaction conditions were performed.
Cyclohexanol and cyclohexanone conversion over Me/ZrO2
catalysts
At 373 K, for all catalysts, the relative intensities of the bands
corresponding to adsorbed phenoxy species and cyclohexanone
decreased while those of cyclohexanol increased. At 473 K,
phenoxy and cyclohexanol bands decreased for Pt, Rh, and Ru
catalysts, while phenoxy bands continued to decrease and
cyclohexanol bands continued to increase for Pd and Ni catalysts.
The formation of Pt carbonyl bands was also significant at this
temperature. At 573K, all band intensities decreased for Pt, Rh,
and Ru catalysts. However, while cyclohexanol band diminished,
residual phenoxy bands were still present over Pd and Ni
catalysts. At 673 K and 773 K, only traces of bands were present.
One exception was Pd, where phenoxy species (albeit at
significantly reduced intensity) and a band at 1545 cm-1 were
observed even at 673 K.
Product yields for each feed at the same conversion level are
reported in Table 4. For all catalysts, cyclohexanol was
preferentially converted to cyclohexanone, indicating that this
reaction is rapid. Pt/ZrO2, Ni/ZrO2 and Co/ZrO2 catalysts exhibited
the highest yields of cyclohexanone. The formation of benzene
was quite low for most of the catalysts, suggesting that the
reaction of cyclohexanol dehydration is not relevant. Indeed, the
cyclohexanol conversion over the zirconia support at the highest
W/F (used when feeding phenol over the supported catalysts) was
very low, confirming that the contribution of dehydration catalyzed
by the zirconia support is negligible. Therefore, benzene cannot
be produced by this reaction pathway.
The cyclohexanone was mainly converted to cyclohexanol and
bicyclic compounds (typically C12 hydrocarbons) such as biphenyl,
2-phenylphenol, cyclohexylbenzene, 2-cyclohexylcyclohexan-1-
one and 2-cyclohexylphenol. The formation of these bicyclic
products has been reported in the literature [14] stemming from the
The results of DRIFTS are cast in terms of two mechanisms, as
supported by prior DFT and experimental investigations: (1) direct
dehydroxylation of phenol to a partially unsaturated hydrocarbon
surface species (e.g., C6H5*), with subsequent hydrogenation to
benzene [by analogy with [17]], which is favored by metal surfaces
such as Ru(0001) or (2) a mechanism involving a keto tautomer
intermediate, the latter of which may undergo selective
hydrogenation of the carbonyl to 2,4-cyclohexadienone
(depending on the support), followed by dehydration to benzene.
This was suggested to occur on Pd/ZrO2 catalysts by
experimental observations [5], as well as on Pt (111). [18]
alkylation of phenolic and aromatic rings by cyclohexanone.
[15]
Nimmanwudipong et al.
observed the formation of C10-C12
hydrocarbons when they studied the conversion of
cyclohexanone over Pt/Al2O3 catalysts. Different bicyclic C12
products were detected by the bimolecular reactions between
cyclohexanone and the different products formed during the
reaction. This reaction is generally catalyzed by acid sites such
as Lewis acid sites. In our work, these reactions could take place
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
Table 4. Comparison of performance of catalysts using different feeds: phenol, cyclohexanol (OL) or cyclohexanone (ONE) at 573 K and 1 atm.
Feed
Conversion
(%)
Yield (%)
o-cresol
ONE
OL
Benzene
Phenol
CH4
C5-C6
C12
hydrocarbons
0.8
hydrocarbons
0.9
27.1
46.4
28.4
23.0
20.8
15.5
19.3
24.5
21.8
32.4
19.0
22.8
21.5
33.3
20.3
28.4
28.5
25.4
10.0
42.4
-
3.4
-
12.0
0.8
1.1
11.2
3.1
0.1
12.8
4.9
0.4
14.0
0.1
-
-
2.5
4.5
-
-
-
-
Phenol
OL
-
0.5
0.4
0.0
1.2
0.1
1.5
2.0
0.1
4.7
0.4
-
0.2
4.2
Pt/ZrO2
Pd/ZrO2
Rh/ZrO2
Ru/ZrO2
Ni/ZrO2
Co/ZrO2
18.2
0.5
-
-
-
ONE
Phenol
OL
9.7
12.8
-
-
-
1.6
3.3
0.7
-
-
-
0.4
3.0
0.1
-
-
-
11.6
0.1
ONE
Phenol
OL
2.6
15.8
-
-
2.2
-
1.5
1.1
-
-
0.3
6.1
0.3
-
-
-
14.1
0.3
ONE
Phenol
OL
1.1
16.8
-
0.3
11.7
1.0
-
0.5
-
-
0.2
5.6
2.2
-
-
17.2
2.9
ONE
Phenol
OL
4.3
26.1
-
6.3
0.3
0.2
4.8
0.1
-
-
1.2
4.4
0.3
0.5
7.5
0.2
-
0.2
0.1
0.1
1.0
0.6
0.2
6.0
1.1
-
-
-
0.5
7.0
0.9
-
11.4
2.4
ONE
Phenol
OL
2.5
26.0
-
9.3
-
1.3
-
0.3
6.4
-
18.8
ONE
interlocking of the carbonyl of the keto tautomer intermediate (i.e.,
2,4-cyclohexadienone) with defect sites of zirconia, predisposing
the molecule to attack of the carbonyl functional group to
hydrogenation to 2,4-cyclohexadienol. Formation of the latter
facilitates formation of benzene by dehydration.
DRIFTS spectra of the Pt catalyst bear many similarities, but
some differences, to Pd. The carboxylate-like band as previously
described for Pd was detected at 573 K (~1537 cm-1). However,
the ratio of cyclohexanol bands to phenoxy bands was higher for
Pt versus Pd at least up to 473 K, and broad low-lying bands in
the range of (C=O) (e.g., 1620 – 1800 cm-1) were detected up to
573 K in the case of Pt. These results are consistent with the
higher selectivities to cyclohexanone (and its subsequent
hydrogenation product, cyclohexanol) and lower selectivity to
benzene for Pt relative to Pd (Table 3).
In comparison with Pd and Pt, the cyclohexanol to phenoxy band
ratio is even greater for Rh up to 473 K. However, at 573 K,
cyclohexanol bands are virtually absent for Rh, whereas they are
clearly present for Pd and detected at low intensity with Pt.
Likewise, phenoxy bands were barely discernible, as was a broad
band peaking at 1614 but extending up to 1680 cm-1, which may
be ascribed to (C=O). The latter is consistent with the formation
of cyclohexanone (Table 3). In comparison with Pt and Pd, results
for Rh at 573 K suggest that it is more reactive than the other two
metals. Further evidence for this conclusion is the higher methane
selectivity, as well as the significantly higher C5-C6 hydrocarbon
selectivities (Table 3).
For Ru (0001), an important side reaction was proposed to occur,
from prior DFT and experimental work, C-C breaking of the
dehydroxylated intermediate, leading to C1-C5 hydrocarbons,
consistent with the current reaction testing results (e.g., Figure 3
and Table 3). With Pd and Pt catalysts, and depending on the
support utilized, hydrogenation of the 2,4-cyclohexadienone
intermediate to cyclohexanone may also occur, with further
hydrogenation of the carbonyl to cyclohexanol. This is the reason
that oxophilic supports are important for improving selectivity, as
preferential adsorption of the keto intermediate by its O atom can
facilitate the selective hydrogenation of the carbonyl bond. A
screening of metals in Table 3 suggests that Pd and Pt likely
follow the second pathway, as selectivities to cyclohexanone (and
cyclohexanol, in the case of Pt) side product were higher. Ru had
low cyclohexanone and cyclohexanol selectivities, and the
highest light side product selectivities, in agreement with the first
pathway. The other metals – Rh, Ni, and Co – appear to exhibit
aspects of both mechanisms.
The DRIFTS results lend some support to this. Based on intensity
of the bands, the coverage of phenoxy species at 573 K is the
highest for Pd, while only traces of bands are detected for Ru.
Moreover, although the bands of cyclohexanol were diminished
from the 473 K condition, the coverage of cyclohexanol remains
significant at least up to 573 K with Pd. In contrast, with Ru, the
intensity of cyclohexanol decreases to below detection limits in
moving from 473 K to 573 K. The results thus suggest that a more
active bond-breaking mechanism may be operating for Ru – for
example, path 1, while the less reactive path 2 may be operating
for Pd. Moreover, a band at 1545 cm-1 is present at 573 K and
673 K for Pd, which is consistent with the earlier proposal that a
(OCO) mode of a carboxylate-like species forms from the
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
dissociation energy of the C-O bond of the aromatic ring in the
phenol molecule and thus, it may occur depending on reaction
temperature and catalyst used; and (iii) the phenol
tautomerization to a 2,4-cyclohexadienone intermediate.[5] At this
stage, 2,4-cyclohexadienone may be hydrogenated at the ring to
produce 2-cyclohexen-1-one, which is further hydrogenated (e.g.,
to cyclohexanone and cyclohexanol), or with the right catalyst
surface structure be preferentially hydrogenated at the carbonyl
group. This produces 2,4-cyclohexadienol, which is subsequently
dehydrated to benzene.
The results obtained in the present work revealed that the reaction
pathway for HDO of phenol strongly depends on the type of metal.
Similar results have been reported in the literature for HDO of
different model molecules. [2e,f,j,l,17,21]
Silica supported metal catalysts (i.e., Ni, Pd, and Pt) were
[40]
investigated for m-cresol HDO
The main products formed at
573K were toluene and 3-methylcyclohexanone over Pt- and Pd-
based catalysts. However, a high selectivity to methane and
phenol was also observed for Ni/SiO2 catalyst. Direct
deoxygenation producing toluene and hydrogenation (forming
methylcyclohexanone
and/or
methylcyclohexanol)
were
suggested by Chen et al.[19d] to be the major reactions, with
selectivity depending on the metal. They suggested that
deoxygenation occurs through a mechanism that may involve
direct C-O hydrogenolysis or tautomerization, because there was
insufficient acidity to catalyze 3-methylcyclohexanol for
a
hydrogenation/deoxygenation pathway. Scission of the C-C bond
by hydrogenolysis of the m-cresol methyl group over Ni catalyst
yielded methane and phenol. However, the reasons for the higher
hydrogenolysis activity on Ni/SiO2 catalyst were not given.
A detailed investigation about the HDO of m-cresol in gas phase
over silica supported catalysts (Pd, Pt, Ru, Ni, Fe, NiFe) is
available in the literature. [5,17,22] Products of hydrogenation (e.g.,
3-methylcyclohexanone and 3-methylcyclohexanol) were mainly
formed over Pt/SiO2, Pd/SiO2 and Ni/SiO2 catalysts whereas
toluene was the main product on Fe and Ni–Fe bimetallic
catalysts. Ru/SiO2 catalyst also exhibited a high selectivity to
toluene, CH4 and C2-C6 hydrocarbons. Differences in product
distributions were attributed to a complex reaction pathway that
varied depending on the type of metal. For Pt/SiO2, Pd/SiO2,
Ni/SiO2, Fe/SiO2, NiFe/SiO2 catalysts, the formation of a keto-
tautomer intermediate (3-methyl-3,5-cyclohexadienone) is a key
step, followed by the hydrogenation of the ring of this tautomer,
producing 3-methylcyclohexanone, or the hydrogenation of the
carbonyl group, originating 3-methyl-3,5-cyclohexadienol. The
mechanism depends on metal type. With an oxophilic metal, such
as the partially reduced Fe species in NiFe/SiO2 catalysts,
hydrogenation of the carbonyl group of the tautomer intermediate
was favored. In contrast, the direct dehydroxylation of m-cresol
occurs over Ru based catalyst. Computational studies involving
DFT indicated a higher reaction energy and energy barrier for
carbonyl group hydrogenation of the keto tautomer; this is due to
the greater strength of the interaction between O of the carbonyl
and Ru than Pt. The direct dehydroxylation was suggested to
produce partially unsaturated hydrocarbon species; which may
further react to toluene via hydrogenation or undergo C-C scission
(yielding light hydrocarbons like methane). Thus, the oxophilicity
Figure 2. DRIFTS spectra obtained under reaction mixture containing phenol
and hydrogen (H2/phenol molar ratio = 60) at different temperatures (323, 373,
473, 573, 673 K) over Pt/ZrO2; Pd/ZrO2; Rh/ZrO2; Ru/ZrO2 and Ni/ZrO2 catalysts.
As in the case of Ru, the summation of band intensities for Ni
reveals a lower coverage, which may indicate that Ni is carrying
out more active bond breaking reactions. In agreement with this,
and consistent with path 1, the selectivity to methane was quite
high (albeit approximately half that of Ru), and C5-C6
hydrocarbons were formed. However, judging from the infrared
band intensities, there remains a low coverage of adsorbed
cyclohexanol and phenoxy bands on the Ni catalyst at 573 K.
These species are likely responsible for the cyclohexanone,
cyclohexanol, and possibly benzene that is produced from
pathway 2.
Proposed reaction pathway for hydrodeoxygenation of
phenol
The reaction mechanism for HDO of phenol has been debated in
[2a,5,18,19]
the open literature.
Different reaction pathways have
been proposed for the HDO of phenol such as: (i) successive
metal-catalyzed steps for hydrogenating (HYD) the aromatic ring
and acid-catalyzed dehydration of the cyclohexanol formed. This
route depends strongly on the acidity of the support. [14b,20]
Zirconia does not offer a significant extent of dehydration due to
insufficient acidity, as revealed by the catalytic tests with the pure
support; (ii) the direct deoxygenation (DDO) that involves high
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
of the metal surface plays a key role in determining which reaction
pathway is favored. In summary, oxophilic metals such as Ru and
higher hydrogenolysis rates due to the weaker bonding of H*
atoms to terrace sites.
Fe
promote
the
deoxygenation
whereas
In order to demonstrate that direct deoxygenation takes place
over Ru-based catalysts, a 1%Ru/ZrO2 catalyst with smaller Ru
particle size was tested for HDO of phenol and the results are
presented in Table 3. The decrease in Ru particle size was
accompanied by a significant reduction in the formation of
methane and C5-C6 hydrocarbons as well as by an increase in the
selectivity to benzene. This result agrees very well with the work
of Newman et al.[2g] They reported that a Ru/TiO2 catalyst with
higher dispersion exhibited high selectivity toward direct
deoxygenation. Therefore, the selectivity to methane and C5-C6
hydrocarbons could be used to measure the extent of the direct
deoxygenation reaction pathway for HDO of phenol reaction. For
the catalysts that do not exhibit significant hydrogenolysis
reaction (Pt/ZrO2, Pd/ZrO2, Rh/ZrO2), the changes in product
distribution are not due to the different metal dispersion. For these
catalysts, the hydrogenation was the main reaction, which is
considered an insensitive structure-reaction. Recently, the effect
of Pd particle size on the product distribution for HDO of phenol
in gas phase over Pd/ZrO2 catalysts with different Pd dispersions
was investigated. [5] Similar selectivities were observed with both
catalysts; thus, the dispersion of the metal does not significantly
affect the HDO product distribution. But the effect of the metal
dispersion on HDO of phenol reaction is controversial in the
literature. In addition, When Co/ZrO2 is compared with the Ni/ZrO2
catalyst with approximately the same dispersion, it is clear that
the differences in product distribution are due to the type of the
metal.
tautomerization/hydrogenation is favored over less oxophilic
metals (e.g., Pt, Pd). Note that this reaction route may be further
extended to explain the formation of hydrogenolysis products on
some catalysts.
Concerning the HDO of phenol, there are only two studies in the
literature investigating the effect of the metal on the reaction
mechanism.[2e,4] Recently, we investigated the HDO of phenol
reaction over a series of silica supported metal (Pd, Pt, Rh, Ru,
Ni, Co) catalysts.[4] The formation of hydrogenated products
(cyclohexanone and cyclohexanol) was favored over Pt/SiO2,
Pd/SiO2 and Rh/SiO2 catalysts, whereas significant formation of
hydrogenolysis products (C5-C6 hydrocarbons and methane) was
observed over Co/SiO2, Ni/SiO2 and mainly Ru/SiO2 catalysts.
These results showed that the reaction pathway depends on the
type of the metal. With silica supported Pt, Pd, and Rh, phenol
undergoes tautomerization, with subsequent hydrogenation of the
ring. The direct dehydroxylation of phenol followed by
hydrogenolysis is favored over more oxophilic metals (Ru, Co and
Ni). To establish a relation between HDO activity and oxophilicity
of the metals in terms of the ability to bind with O from the carbonyl
functional group, the selectivity to deoxygenated products was
investigated as a function of the binding energy of atomic oxygen
calculated by Lee et al.[23] The following order for selectivity to
deoxygenated products was observed (in parentheses are
reported the binding energies of atomic oxygen): Ru (-5.90) > Co
(-5.58) > Ni (-5.40) > Rh (-5.04) > Pd (-4.48) > Pt (-4.20). This
result reveals that the more oxophilic the metal, the higher the
selectivity to deoxygenated products. As observed by Tan et al.
However, the order observed for the selectivity to deoxygenated
products for zirconia supported catalysts was not the same as that
obtained for the silica supported catalysts from our previous study
[17]
and Hensley et al. [24], the stronger interaction between the
[4]
oxygen from carbonyl group to metal favors the formation of
deoxygenated products.
In the present work, both reaction pathways (tautomerization and
direct deoxygenation) might take place and the importance of
each one will depends on the type of the metal. For Pd/ZrO2 and
(Fig. 3). For silica supported catalysts, Ni exhibits a higher
selectivity to deoxygenated products than Rh, which is
reasonable since Ni is more oxopholic than Rh. However, this is
not observed for Ni/ZrO2 and Rh/ZrO2 catalysts. Even Pt and Pd
metals, which did not produce benzene when supported on silica,
Pt/ZrO2 catalysts, tautomerization is the reaction pathway favored, showed significant formation of deoxygenated products in the
as recently demonstrated by the formation of tautomer
presence of zirconia. This result reveals the important role the
[5]
[27]
intermediate during the HDO of phenol reaction.
The high
support for the HDO of phenol reaction. Griffin et al.
also
formation of methane and C5-C6 hydrocarbons over Ru/ZrO2,
Ni/ZrO2 and Co/ZrO2 catalysts could be attributed to the direct
dehydroxylation of phenol followed by hydrogenolysis of the
intermediate formed. According to the mechanism proposed by
Tan et al. [17], the direct deoxygenation of m-cresol produces an
unsaturated hydrocarbon surface species that undergoes
hydrogenolysis to C1-C5 hydrocarbons; or it can be hydrogenated,
forming toluene. The hydrogenolysis reaction proceeds by
cleaving C-C bonds to form CxHy species and CH species. The
hydrogenolysis of alkanes has been considered a structure-
sensitive reaction and a decrease in activity with decreasing
particle size has been observed. [25] Iglesia’s group performed a
detailed study about the hydrogenolysis of n-alkanes on Ir, Rh and
Pt clusters. [25a,26] They used statistical mechanics and transition
state theory to determine the rates and selectivities of C-C bond
cleavage in C2-C10 n-alkanes to describe the hydrogenolysis of
hydrocarbons. They reported that large metal clusters exhibit
observed a high selectivity to toluene for HDO of m-cresol when
Pt was supported on TiO2. They proposed that tautomerization
and direct deoxygenation to toluene were energetically favorable
routes over TiO2 (101) surfaces. In this case, the reaction takes
place at oxygen vacancy defects. For the Pt(111) surface, 3-
methylcyclohexanone and 3-methylcyclohexanol were mainly
formed by the hydrogenation of the ring, which explains the low
selectivity to toluene for the Pt/C catalyst used.
In the present work, in addition to the metal, the support also
provides oxophilic sites that will take part in the reaction
mechanism. This explains the higher formation of deoxygenated
products for the HDO of phenol over zirconia supported catalysts.
There are some studies in the literature that report the promoting
effect of oxophilic sites for the deoxygenation activity. For
instance, the high selectivity to benzene for the HDO of phenol in
the liquid phase over Ru/TiO2 catalyst was attributed to the strong
interaction between the Ti3+ sites (created during the reduction)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
with the oxygen from the phenol molecule that weakened the C-
O bond.
Ni/SiO2
Ni/ZrO2
63.3
88
37.7
Pt/SiO2
Pt/ZrO2
26.7
20
16.2
10.9
56.6
8.4
5.6
4.8
2.8
2
1.2
.
.
l
o
s
e
E
N
d
d
y
H
2
n
OL
y
e
e
r
c
-
o
O
H
z
n
6
e
1
C
-
B
C
5
C
27.6
+
4
H
C
12.8
11
1.8
1.4
.
.
d
e
E
d
n
N
OL
y
y
H
e
O
H
z
Co/SiO2
Co/ZrO2
n
6
2
e
1
C
-
B
C
5
C
+
4
H
C
93.7
Pd/SiO2
Pd/ZrO2
38
35.9
18.6
18.5
16.5
17
16
52.6
12
11.7
43.2
5.7
5.4
4.8
.
.
l
o
s
e
E
d
d
y
H
2
n
N
OL
y
e
e
r
c
-
o
O
H
z
n
6
e
1
C
-
B
C
5
C
+
4
H
C
3.3
3
2.4
1.7
Figure 3. Selectivity for HDO of phenol in gas phase at 573 K and 1 atm over
silica [16] and zirconia [this work] supported catalysts.
.
d
.
d
y
H
e
E
n
N
OL
y
e
O
z
H
n
6
2
e
1
C
-
B
C
5
C
+
4
H
C
Recently, we investigated the effect of the support on HDO of
phenol in gas phase over Pd-based catalysts. [28] Pd/TiO2 and
Pd/ZrO2 catalysts exhibited high selectivity to benzene, which was
assigned to the oxophilic sites of these supports represented by
incompletely coordinated Ti4+/Ti3+ and Zr4+/Zr3+ cations near the
perimeter of the metal particles. The selective hydrogenation of
the carbonyl function of the tautomer intermediate produced was
favoured by a stronger interaction between the oxygen atom in
phenol molecule with the metal oxide cation.
Rh/SiO2
85.3
Rh/ZrO2
63.4
In addition, Co/ZrO2 catalyst exhibited a high selectivity to o-
cresol. This product is likely formed from the decomposition of
bicyclic compounds formed over Lewis acid sites. Kallury et al.
[14a] reported the formation of cresols from o-cyclohexylphenol, a
bicyclic compound that can be formed from alkylation of phenolic
and aromatic rings by cyclohexanol or cyclohexanone. Recently,
22
14.3
7.2
4.2
3.1
1.7
0.2 0.3
.
.
d
e
E
d
n
N
OL
y
y
e
O
H
H
z
n
6
2
e
1
C
-
B
C
5
C
+
4
H
C
we reported a high selectivity to o-cresol for HDO of phenol over
[4]
Co/SiO2 catalyst.
This result was attributed the presence of
unreduced Co species as Lewis acid sites. From Table 3, it is
noticed that Co/ZrO2 catalyst showed the highest formation of
bicyclic compounds. In addition, the TPR experiment revealed a
low degree of reduction of this catalyst, which could be
responsible for the high selectivity towards o-cresol.
Ru/SiO2
Ru/ZrO2
71
49.7
42.2
19.4
5.6
5.8
3.4
0.8
0.4 0.6
.
.
d
e
E
d
n
N
OL
y
y
e
O
H
H
z
n
6
2
e
1
C
-
B
C
5
C
+
4
H
This article is protected by copyright. All rights reserved.
C

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
Stability of supported metallic catalysts on HDO of phenol
The conversion of phenol and product distribution as a function of
time on stream (TOS) for all catalysts are shown in Fig. 4. All
100
90
80
70
60
50
40
30
20
10
0
Conversion
ONE
(D)
OL
Benzene
C5-C6 hydrocarbons
catalysts deactivated during 24
h of TOS but the main
deactivation occurs at the beginning of reaction (before 200 min.).
In addition, the deactivation degree depended on the metal. The
stability of the catalysts defined by Eq. (3) was calculated and the
following order was obtained: Pd/ZrO2 (0.58) > Pt/ZrO2 (0.41) >
Co/ZrO2 (0.39) > Rh/ZrO2 (0.35) > Ru/ZrO2 (0.24) > Ni/ZrO2 (0.19).
Furthermore, the selectivities significantly changed with TOS. The
selectivity to cyclohexanone increased and the formation of
benzene decreased as the phenol conversion decreased for Pt,
Pd, Rh, Ni catalysts. The selectivity to C12 hydrocarbons also
decreased with TOS. For Ru/ZrO2, the selectivity to CH4 strongly
decreased at the beginning of reaction, whereas it slightly
diminished for the other catalysts.
C12 Hydrocarbons
CH4
0
200
400
600
800
1000
1200
1400
Time (min)
100
90
80
70
60
50
40
30
20
10
0
100
Conversion
ONE
(E)
Conversion
(A)
90
ONE
OL
OL
80
Benzene
o-cresol
Benzene
C12 Hydrocarbons
-C hydrocarbons
70
C12 Hydrocarbons
-C hydrocarbons
C
5
6
C
60
50
40
30
20
10
0
5
6
CH
4
0
200
400
600
800
1000
1200
1400
0
200
400
600
800
1000
1200
1400
Time (min)
Time (min)
100
90
80
70
60
50
40
30
20
10
0
Conversion
ONE
100
90
80
70
60
50
40
30
20
10
0
(F)
Conversion
ONE
(B)
OL
Benzene
o-cresol
OL
Benzene
C12 Hydrocarbons
-C hydrocarbons
C
C12 Hydrocarbons
-C hydrocarbons
5
6
C
CH
4
5
6
0
200
400
600
800
1000
1200
1400
0
200
400
600
800
1000
1200
Time (min)
Time (min)
Figure 4. Effect of time on stream on of conversion of phenol and selectivity of
products over (A) Pt/ZrO2; (B) Pd/ZrO2; (C) Rh/ZrO2; (D) Ru/ZrO2; (E) Ni/ZrO2
and (F) Co/ZrO2. Reaction conditions: 573 K, atmospheric pressure, H2/phenol
molar ratio= 60. ONE: Cyclohexanone; OL: Cyclohexanol.
100
90
80
70
60
50
40
30
20
10
0
(C)
Conversion
ONE
OL
Benzene
C12 Hydrocarbons
Recently, we studied the deactivation of Pd-based catalysts for
C
-C hydrocarbons
6
5
HDO of phenol. Pd/SiO2, Pd/Al2O3, Pd/TiO2 and Pd/ZrO2 catalysts
CH
4
[28]
significantly deactivated with TOS.
However, Pd/CeO2 and
0
200
400
600
800
1000
1200
1400
Time (min)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
Pd/CeZrO2 were more stable and only slight losses in activity
were observed. Carbon deposits were not detected by Raman
spectroscopy after reaction. DRIFTS experiments under reaction
conditions revealed a build-up of phenoxy and intermediate
species during reaction. These species remained adsorbed on
the oxophilic sites, blocking those sites and inhibiting further
reactant adsorption.
1284
(B)
1488
1589
The growth of Pd particle size and a decrease in the density of
oxophilic sites during the HDO of phenol reaction were the main
causes of catalyst deactivation.
2934
1448
1159
3069
Therefore, in the present study, DRIFTS experiments under
reaction conditions were carried out in order to monitor the
evolution of adsorbed species.
(f)
(e)
(d)
Fig. 5 show the DRIFTS spectra for HDO of phenol at 573 K.
Bands of adsorbed phenoxy species were detected for all
catalysts. In all cases, except that of Ru, bands of adsorbed
cyclohexanol were also evidenced. In addition, the (C=O) band
of cyclohexanone was observed for the Pt and Rh catalysts as a
high wavenumber shoulder; furthermore, the Pt-carbonyl band
was identified at 1984 cm-1. These key bands tended to increase
with TOS. The build-up of an inventory of intermediates on the
catalyst surface suggests that the turnover rate of these species
is slowly becoming hindered with TOS. The sharpest increases
in inventory followed the trend: Ru > Ni > Rh > Pt > Pd. These
results are consistent with the initial deactivation trends observed
in the long-term catalytic tests, with Ru offering the steepest initial
decline, and Pd offering the greatest stability.
(c)
(b)
(a)
3500
3000
2500
2000
1500
1000
Wavenumber (cm-1)
1280
1587
1473
(C)
(A)
1587
1616
1538
1280
1407
1427
1984
3040
1168
2934
2932
2862
(f)
(f)
(e)
(d)
(e)
(d)
(c)
(c)
(b)
(a)
(b)
(a)
3500
3000
2500
2000
1500
1000
3500
3000
2500
2000
1500
1000
Wavenumber (cm-1)
Wavenumber (cm-1)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
240; (e) 300; (f) 360 over (A) Pt/ZrO2; (B) Pd/ZrO2; (C) Rh/ZrO2; (D) Ru/ZrO2;
(E) Ni/ZrO2.
(D)
Therefore, DRIFTS experiments revealed the build-up of
intermediate species on the surface of all catalysts. This could be
caused by the loss of the Pd-support interaction, which could be
associated with: (i) carbon deposition resulting in site blocking at
the interface; (ii) metal sintering, which decreases the interface
between metal particles and support; or (iii) changes in the acid
sites of the support (e.g., decreases in Lewis acid sites due to site
blocking by strong adsorbates).
1284
1484
1448
1583
(f)
(e)
(d)
In the present work, Raman spectra of the used catalysts did not
detect the formation of carbon deposits on the surface of all
catalysts.
In order to investigate the changes in metal particle size during
the reaction, the dehydrogenation of cyclohexane reaction was
carried out before and after the HDO of phenol reaction. The ratio
between the reaction rates of dehydrogenation of cyclohexane at
543 K over fresh and used (after 24 h of TOS for HDO of phenol
reaction) catalysts is reported in Table 5. The results showed a
significant reduction in reaction rate, which suggest that metal
particles growth contributed to catalyst deactivation. However, the
degree of sintering varied depending on the metal. This was more
important for Pt, Pd and Co supported catalysts. The lowest
decrease in the reaction rate of dehydrogenation of cyclohexane
was observed for the Ru/ZrO2 and Rh/ZrO2 catalysts.
(c)
(b)
(a)
3500
3000
2500
2000
1500
1000
Wavenumber (cm-1)
Table 5. Ratio between the reaction rate of dehydrogenation of cyclohexane at
543 K over fresh and used catalysts and the reaction rate of dehydration of
cyclohexanol at 543 K over fresh and used catalysts (after 23 h of HDO of
phenol reaction).
1284
(E)
1484
Catalyst
Dehydrogenation of
cyclohexane ratio[a]
Dehydration of
cyclohexanol ratio[b]
1585
Pt/ZrO2
Pd/ZrO2
Rh/ZrO2
Ru/ZrO2
Ni/ZrO2
Co/ZrO2
18.5
10.1
2.9
6.2
2.5
8.8
0.9
7.7
0.9
1440
1160
(f)
1.4
5.5
(e)
24.0
[a] Ratio between the reaction rate of dehydrogenation of cyclohexane at 543 K
over fresh and used (after 23 h of HDO of phenol reaction) catalysts.
[b] Ratio between the reaction rate of dehydration of cyclohexanol at 543 K over
fresh and used (after 23 h of HDO of phenol reaction) catalysts.
(d)
(c)
The changes in the density of oxophilic sites were measured by
the dehydration of cyclohexanol reaction. Table 5 lists the ratio
between the reaction rates for dehydration of cyclohexanol before
and after HDO reaction. A significant decrease in the rate of
cyclohexanol dehydration is observed for Pt/ZrO2, Rh/ZrO2, and
Ni/ZrO2 catalysts. For Ru/ZrO2, and Co/ZrO2 catalysts, the
reaction rate only slightly decreased after 20 h of TOS. The
comparison between the variation of metal dispersion and the
density of acid sites during HDO of phenol over all catalysts
suggest that metal sintering is the main cause of catalyst
deactivation. The growth of metal particle size decreases the
(b)
(a)
3500
3000
2500
2000
1500
1000
Wavenumber (cm-1)
Figure 5. DRIFTS spectra obtained at 573 K and under the reaction mixture
containing phenol and hydrogen during 6 h TOS: (a) 60; (b) 120; (c) 180; (d)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
metal particle-support interface that affects the ability of the
adsorbed species to turnover, leading to an accumulation of
phenoxy species during reaction, as revealed by DRIFTS
experiments. However, the decrease in the density of acid sites is
significant for Pt, Rh and Ni-based catalysts. Considering that the
HDO of phenol is a bifunctional mechanism, the decrease of both
the metal particle-support interface (Pd sintering) as well as the
number of Lewis acid sites (site blocking) might contribute to the
deactivation observed for zirconia supported catalysts during the
HDO of phenol reaction.
adsorption of intermediates) during the HDO of phenol reaction were
the main causes of catalyst deactivation.
Experimental Section
Catalyst synthesis
ZrO2 support was synthesized by the precipitation method. A solution of
2.0 mol L-1 zirconyl nitrate (35 wt.% ZrO(NO3)2 in dilute nitric acid, > 99%,
Sigma-Aldrich) was added slowly to a solution of 4.0 mol L-1 ammonium
hydroxide (NH4OH, Vetec) at room temperature, and kept under vigorous
stirring for 30 min. The resulting precipitate was filtered and washed with
distilled water until a pH of 7 was reached. Then, the solid was dried at
383 K for 12 h and calcined under a dry air flow at 773 K (heating ramp 5
K min-1) for 6 h.
Conclusions
ZrO2 supported metal catalysts with a nominal Pt, Pd and Rh loading of
1.0 wt. %, Ru and Cu of 5.0 wt. %, Ni and Co of 10.0 wt. % were prepared
by incipient wetness impregnation of the supports with aqueous solution
of respective precursor salts (Pt(NH3)4(NO3)2 (Aldrich), Pd(NO3)2.2H2O
(Umicore), RhCl3.H2O (Aldrich), Ru(NO)(NO3)3 (Alfa Aesar),
Ni(NO3)2.6H2O (Acros), Co(NO3)2.6H2O (Acros) and Cu(NO3)2.3H2O
(Aldrich)). After impregnation, the powder was dried in air at 293 K for 12
h and then calcined in air at 673 K for 3 h (2 K min-1).
This work investigated the effect of metal type (Pt, Pd, Rh, Ru, Cu, Ni,
Co) for the HDO of phenol in gas phase over zirconia supported
catalysts. Activity and product distribution were significantly affected
by the type of metal. While zirconia supported Pt and Pd tended to
produce hydrogenated products, such as cyclohexanone and
cyclohexanol, zirconia supported Rh, Co, Ni, and Ru catalysts
produced significant fractions of C5, C6, and methane, as the
products of hydrogenolysis. Cu/ZrO2 catalyst was inactive under the
reaction conditions used. Two reaction pathways (tautomerization
and direct deoxygenation) might take place, the importance of each
one will depends on the type of the metal. Phenol tautomerization is
the main reaction pathway for Pd/ZrO2 and Pt/ZrO2 catalysts. The high
formation of methane and C5-C6 hydrocarbons over Ru/ZrO2, Ni/ZrO2
and Co/ZrO2 catalysts could be attributed to the direct dehydroxylation
of phenol that produces an unsaturated hydrocarbon surface species.
This intermediate underwent hydrogenolysis to C1-C5 hydrocarbons
or it can be hydrogenated, forming benzene. The selectivity to
methane and C5-C6 hydrocarbons could be used to measure the
extent of direct deoxygenation reaction pathway for HDO of phenol.
Catalyst characterization
The chemical composition of each sample was determined using a
Wavelength Dispersive X-Ray Fluorescence Spectrometer (WD-XRF) S8
Tiger (Bruker) with a rhodium tube operated at 4 kW. The analyses were
performed with the samples (300 mg) in powder form using a semi-
quantitative method (QUANT-EXPRES/Bruker). Specific surface areas of
the samples were measured on a Micromeritics ASAP 2020 analyzer by
N2 adsorption at the boiling temperature of liquid nitrogen. The X-ray
powder diffraction (XRD) patterns were obtained using
a Rigaku
diffractometer with Cu K radiation ( = 1.5406 Å) over a 2θ range of 10-
80º at a scan rate of 0.04 o/step and a scan time of 1 s/step. TPR
experiments were performed in a TPR/TPD 2900 Micromeritics system
equipped with a thermal conductivity detector (TCD). The catalyst was
pretreated at 673 K for 1 h under a flow of air prior to the TPR experiment
to remove adsorbed species from the catalyst surface. The reducing
mixture (10.0 % H2/N2) was passed through the sample (100 mg) at a flow
rate of 30 mL min-1 and the temperature was increased to 1273 K at a
heating rate of 10 K min-1. The metal dispersion was measured by H2
chemisorption at 298 K for Rh, Pt, Ru, Ni, Co-based catalysts using a
Micromeritics apparatus. Before adsorption, the samples were reduced
under pure H2 (60 ml min-1) at 773 K for 1h (10 K.min-1), followed by
evacuation for 30 min and cooling to room temperature. Irreversible
uptakes at 298 K were determined from the dual isotherms measured for
hydrogen. First, the total adsorption was determined at 298 K and the
catalyst was outgassed for 30 min, after which a new isotherm was
measured. Then, the difference between both isotherms was extrapolated
to zero pressure to calculate the amount of hydrogen adsorbed,
considering a stoichiometry H/Me = 1. This method was not used for the
determination of Pd dispersion due to the formation of hydride in palladium.
In this case, the dehydrogenation of cyclohexane was used as an
insensitive structure reaction to determine the metal dispersion of
supported Pd catalysts. [5,12] This procedure enables the measurement of
In addition to the metals, the oxophilic sites of this support
represented by incompletely coordinated Zr4+ and Zr3+ cations near
the perimeter of the metal particles also played an important role on
the selectivity to deoxygenated products. This explains the higher
formation of deoxygenated products for the HDO of phenol over
zirconia supported catalysts in comparison to silica supported ones.
All catalysts significantly deactivated during TOS, regardless the type
of the metal. Carbon deposits were not detected by Raman
spectroscopy after reaction. DRIFTS experiments under reaction
conditions revealed a build-up of phenoxy and intermediate species
during reaction. These species remained adsorbed on the oxophilic
sites, blocking those sites and inhibiting further reactant adsorption.
The growth of metal particle size decreases the metal particle-support
interface that affects the ability of the adsorbed species to turnover,
leading to an accumulation of phenoxy species during reaction, as
revealed by DRIFTS experiments. The growth of metal particle size
and a decrease in the density of oxophilic sites (site blocking by strong
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
the metal dispersion of fresh and used catalysts (after HDO of phenol for
24 h). This reaction was performed in a fixed bed reactor at atmospheric
pressure. The fresh samples were first reduced at 773 K for 1 h and then
cooled to the cyclohexane dehydrogenation reaction temperature (543 K).
The reaction mixture was fed to the reactor after bubbling H2 through a
saturator containing cyclohexane kept at 285 K (H2/C6H12 = 13.2). For the
used catalysts, after HDO of phenol for 24 h, the saturator containing
phenol was bypassed and hydrogen flowed through the catalyst for 30 min
at 573 K, which removed hydrocarbons that remained adsorbed on the
surface. The reactor was cooled to 543 K under hydrogen. Then, a
cyclohexane/H2 mixture was passed through the reactor. The exit gases
were analyzed by an Agilent Technologies 7890A/5975C CGMS, using a
HP-Innowax capillary column and a flame-ionization detector (FID).
The titration of oxophilic sites of the catalysts before and after HDO of
phenol was measured by conducting the cyclohexanol dehydration
reaction. [13] The reaction was performed using a fixed-bed quartz reactor
at atmospheric pressure and 543 K. Prior to reaction, the fresh catalyst
was reduced in situ under pure hydrogen (60 mL min-1) at 573 K for 1 h.
The used catalysts were not exposed to air before starting the
cyclohexanol reaction. The reactant mixture was obtained by flowing He
(30 mL min-1) through a saturator containing cyclohexanol, which was
maintained at 336 K. The reaction products were analyzed by GCMS
(Agilent Technologies 7890A/5975C) using an HP-Innowax capillary
column and a flame-ionization detector (FID). The dehydration rate was
calculated by the sum of cyclohexene, cyclohexane and benzene yields.
To investigate the reaction mechanism, experiments were performed in an
in situ DRIFTS cell (Thermo Spectra-Tech high P/high T with ZnSe
windows) using a Nicolet Nexus 870 spectrometer equipped with a DTGS-
TEC detector under similar conditions to those employed in the HDO
reaction. Scans were taken at a resolution of 4 cm-1 to give a data spacing
of 1.928 cm-1. The number of scans taken was 1024. The amount of
catalyst was ~ 40 mg. The sample was reduced in H2 at 773 K for 1 h and
cooled to 323 K in He, and a background spectrum was recorded. Pure
H2 was then flowed through a bubbler containing phenol at 351 K and the
temperature was raised to 373, 473, 573, 673 and 773 K. To study the
deactivation of the catalysts, several spectra were recorded during the
steady-state HDO reaction at 573K for 6 h using a H2 flow of 60 ml min-1
through the phenol saturator.
mol of product produced
mol of phenol fed
(1)
(2)
yield(%) 
100
mol of product produced
mol of phenol consumed
Selectivity (%) 
100
Stability  Ln

1 X   / Ln

1 X  
t f
t0
(3)
Acknowledgements
The authors thank CAPES (Coordenação de Aperfeiçoamento de
Pessoal de Ensino Superior) and CNPq (Conselho Nacional de
Desenvolvimento Científico e Tecnológico) for supporting this
research and the scholarship received. CAER researchers would
like to thank the Commowealth of Kentucky for support.
Keywords: Dehydroxylation • Hydrodeoxygenation • Phenol •
Tautomerization • ZrO2
[1]
[2]
X. Zhu, L. Nie, L. L. Lobban, R. G. Mallinson, D. E. Resasco, Energy
Fuels 2014, 28, 4104-4111.
a) J. He, C. Zhao, J. A. Lercher, J. Catal. 2014, 309, 362-375; b) C. R.
Lee, J. S. Yoon, Y. W. Suh, J. W. Choi, J. M. Ha, D. J. Suh, Y. K. Park,
Catal. Comm. 2012, 17, 54-58; c) D. H. Hong, S. J. Miller, P. K. Agrawal,
C. W. Jones, Chem. Comm. 2010, 46, 1038-1040; d) J. Horacek, G.
St’ávová, V. Kelbichová, D. Kubicka, Catal. Today 2013, 204, 38-45; e)
P. M. Mortensen, J-D. Grunwaldt, P. A. Jensen, A. D. Jensen, ACS Catal.
2013, 3, 1774-1785; f) T. Mochizuki, S. Chen, M. Toba, Y. Yoshimura,
Appl. Catal. B: Environ. 2014, 146, 237-243; g) C. Newman, X. Zhou, B.
Goundie, I. T. Ghampson, R. A. Pollocck, Z. Ross, M. C. Wheeler, R.
Meulenberg, R. N. Austin, B. G. Frederick, App. Catal. A: Gen. 2014, 477,
64-74; h) J. Filley, C. Roth, J. Mol. Catal. A: Chem. 1999, 139, 245-252;
i) C. Zhao, J. He, A. A. Lemonidou, X. Li, J. Lercher, J. Catal. 2011, 280,
8-16; j) J. Sun, A. M. Karim, H. Zhang, L. Kovarik, X. S. Li, A. J. Hensley,
J. S. Mcewen, J. Wang, J. Catal. 2013, 47-57; k) X. Zhang, T. Wang, L.
Ma, Q. Zhang, Y. Yu, Q. Liu, Catal. Comm. 2013, 33, 15-19; l) B.
Guvenatam, O. Kursun, E. H. J. Heeres, E. A. Pidko, E. J. M. Hensen,
Catal. Today 2013, 233, 83-91; m) S. Echeandia, B. Pawelec, V. L. Barrio,
P. L. Arias, J. F. Cambra, C. V. Loricera, J. L. G. Fierro, Fuel 2014, 117,
1061-1073.
Catalytic activity
The vapor-phase conversion of the oxygenate compound of interest
(phenol, cyclohexanol, or cyclohexanone) was carried out in a fixed-bed
quartz reactor, operating at atmospheric pressure and 573 K. Prior to
reaction, the catalyst was reduced in situ under pure hydrogen (60 mL min-
1) at 773 K for 1 h. The catalysts were diluted with inert material (SiC mass/
catalyst mass = 3.0) to avoid hot-spot formation. The reactant mixture was
obtained by flowing H2 through the saturator containing organic compound,
which was kept at the specific temperature required to obtain the desired
H2/organic compound molar ratio (about 60). To avoid condensation, all
lines were heated at (523 K). The reaction products were analyzed by GC-
MS (Agilent Technologies 7890A), using an HP-Innowax capillary column
and a flame-ionization detector (FID). Each of the catalysts investigated
was evaluated at different W/F by varying the catalyst amount in the range
of 2.5-120 mg. The W/F is defined as the ratio of the catalyst mass (g) to
the mass flow rate of the organic feed (g/h). The product yield and
selectivity for each product were calculated as follows:
[3]
[4]
[5]
J. K. Norskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, J Phys Chem B
2007, 108: 17886-17892.
C. A. Teles, R. C. R. Neto, J. R. Lima, L. V. Mattos, D. E. Resasco, F. B.
Noronha, Catal. Lett. 2016, doi:10.1007/s10562-016-1815-5.
P. M. De Souza, R. C. Rabelo-Neto, L. E. P. Borges, G. Jacobs, B. H.
Davis, T. Sooknoi, D. E. Resasco, F. B. Noronha, ACS Catal. 2015, 5,
1318-1329.
[6]
[7]
W. Khaodee, B. Jongsomjit, P. Praserthdam, S. Goto, S. Assabumrungrat,
J. Mol. Catal. A: Chem. 2008, 280, 35-42.
a) L. V. Mattos, E. R. De Oliveira, P. D. Resende, F. B. Noronha, Catal.
Today 2002, 77, 246-246; b) S. Sitthisa, D. E. Resasco, Catal. Lett. 2011,
141, 784-791; c) X. Karatzas, K. Jansson, A. Gonzalez, J. Dawody, L.
Pettersson, J. App. Catal. B: Environ. 2011, 106, 476-487.
a) L. Chen, H. Zhu, H. Zheng, C. Zhang, Y. Li, App. Catal. A: Gen 2012,
411-412, 95-104; b) Y. H. Kim, E. D. Park, H. C. Lee, D. Lee, App. Catal.
A: Gen. 2009, 366, 363-369.
[8]
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
[9]
X. Zhang, T. Wang, L. Ma, Q. Zhang, Y. Yu, Q. Liu, Catal. Comm. 2013,
33, 15-19.
[10] T. L. Lai, Y. Y. Shu, G. L. Huang, C. C. Lee, C. B. Wang, J. All. Comp.
2008, 450 (1-2), 318-322.
[11] H. Song, L. Zhang, U. S. Ozkan, Green Chem. 2007, 9, 686-694.
[12] P. M. De Souza, R. C. R. Neto, L. E. P. Borges, G. Jacobs, B. H. Davis,
U. M. Graham, D. E. Resasco, F. B. Noronha, ASC Catal. 2015, 5, 7385-
7398.
[13] D. Martin, D. Duprez, J. Mol. Catal. A: Chem 1997, 118, 113-128.
[14] a) R. K. M. R. Kallury, T. T. Tidwell, Can. J. Chem. 1984, 62, 2540-2545;
b) C. Zhao, D. M. Camaioni, J. A. Lercher, J. Catal. 2012, 288, 92.
[15] T. Nimmanwudipong, R. C. Runnebaum, K. Tay, D. E. Block, B. C. Gates,
Catal. Lett. 2011, 141, 1072-1078.
[16]
a) B-Q. Xu, T. Yamaguchi, K. Tanabe, Mater. Chem. Phys. 1988, 19,
291-297; b) A. Popov, E. Kondratieva, J. M. Goupil, L. Mariey, P. Bazin,
J-P. Gilson, A. Travert, F. Maugé, Journal Phys. Chem. C 2016, 114,
15661-15670; c) W. Roth, P. Imhof, M. Gerhards, S. Schumm, K.
Kleinermanns, Chem. Phys. 2000, 252, 247-256.
[17] Q. Tan, G. Wang, L. Nie, A. Dinse, C. Buda, J. Shabaker, D. E. Resasco,
ACS Catal. 2015, 5, 11, 6271-6283.
[18] F. E. Massoth, P. Politzer, M. C. Concha, J. S. Murray, J. Jakowski, J.
Simons, J. Phy. Chem. B 2006, 110, 14283-14291.
[19] a) C. Zhao, Y. Kou, A. A. Lemonidou, X. Li, J. Lercher, Ang. Chem. 2009,
121, 4047-4050; b) X. Zhang, T. Wang, L. Ma, Q. Zhang, X. Huang, Y.
Yu, App. Energy 2013, 112, 533; c) J. Matos, A. Corma, App. Catal. A:
Gen. 2011, 404, 103-112; d) C. Chen, G. Chen, F. Yang, H. Wang, J.
Han, Q. Ge, X. Zhu, Chem. Eng. Sci. 2015, 135, 145-154; e) K. L.
Deutsch, B. H. Shanks, App. Catal. A: Gen. 2012, 447, 144-150.
[20] A. F. Foster, P. T. M. Do, R. F. Lobo, Top. Catal. 2012, 55, 118-128.
[21]
a) D, Gao, C. Schweitzer, H. T. Wang, H. A. Varma, Ind. Eng. Chem.
Res. 2014, 53, 18658-18667; b) Y. K. Hong, D. W. Lee, H. J. Eom, Y. K.
Lee, App. Catal B: Environ. 2014, 150-151, 438-445.
[22] L. Nie, P. M. De Souza, F. B. Noronha, W. Na, T. Sooknoi, D. E. Resasco,
J. Mol. Catal. A: Chem. 2013, 388, 47-55.
[23]
J. Lee, Y. Xu, G. W. Huber, App. Catal. B: Environ. 2013, 140-141, 98-
107.
[24] A. J. R. Hensley, Y. Wang, J-S. McEwen, ACS Catal. 2015, 5, 523-536.
[25]
a) D. W. Flaherty, E. Iglesia, J. Am. Chem. Soc. 2013, 135 (49), 18586-
18599; b) D. E. Resasco, G. L. Haller, J. Catal. 1983, 82, 279-288.
[26] a) D. W. Flaherty, A. Uzun, E. Iglesia, J. Phys. Chem. C 2015, 119, 2597-
2613; b) D. D. Hibbitts, D. W. Flaherty, E. Iglesia, J. Phys. Chem. C 2016,
120, 8125-8138; c) Flaherty, D. W.; Hibbitts, D. D.; E. Iglesia, J. Am.
Chem. Soc. 2014, 136 (27), 9664-9676.
[27] M. B. Griffin, G. A. Ferguson, D. A. Ruddy, M. J. Biddy, G. T. Beckham,
J. A. Schaidle, ACS Catal. 2016, 6, 2715-272.
[28] P. M. De Souza, R. C. R. Neto, L. E. P. Borges, G. Jacobs, B. H. Davis,
D. E. Resasco, F. B. Noronha, submitted to ACS Catal. 2016.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
FULL PAPER
Product distribution as a function of
the type of the metal
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700047
ChemCatChem
FULL PAPER
This article is protected by copyright. All rights reserved.