Hydrogenation of benzoic acid to benzyl alcohol over Pt/SnO2

Article abstract of DOI:10.1016/j.apcata.2020.117420

Hydrogenation of benzoic acid in the liquid phase over Pt/SnO₂ has afforded excellent selectivity to benzyl alcohol under mild reaction conditions (97 % at 98 % conversion of benzoic acid). DFT calculations have shown that the favoured mode of

Full text of DOI:10.1016/j.apcata.2020.117420

Journal Pre-proof
Hydrogenation of benzoic acid to benzyl alcohol over Pt/SnO2
Xiaohan Chen, Ziyun Wang, Helen Daly, Richard Morgan, Haresh
Manyar, Conor Byrne, Alex S. Walton, S.F. Rebecca Taylor, Matthew
Smith, Robbie Burch, Peijun Hu, Christopher Hardacre
PII:
S0926-860X(20)30013-2
DOI:
https://doi.org/10.1016/j.apcata.2020.117420
APCATA 117420
Reference:
To appear in:
Applied Catalysis A, General
Received Date:
Revised Date:
Accepted Date:
4 October 2019
8 January 2020
17 January 2020
Please cite this article as: Chen X, Wang Z, Daly H, Morgan R, Manyar H, Byrne C, Walton
AS, Taylor SFR, Smith M, Burch R, Hu P, Hardacre C, Hydrogenation of benzoic acid to
benzyl alcohol over Pt/SnO2, Applied Catalysis A, General (2020),
doi: https://doi.org/10.1016/j.apcata.2020.117420
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2020 Published by Elsevier.

Hydrogenation of benzoic acid to benzyl alcohol over
Pt/SnO2
Xiaohan Chena,b, Ziyun Wanga,b, Helen Dalya, Richard Morganb, Haresh Manyarb,
Conor Byrnec,d, Alex S. Waltonc,d, S.F.Rebecca Taylor,a Matthew Smithe, Robbie
Burchb, Peijun Hub and Christopher Hardacrea*
a School of Chemical Engineering and Analytical Science, University of Manchester,
Manchester, M13 9PL, UK
b School of Chemistry and Chemical Engineering, Queen's University Belfast, Belfast,
BT9 5AG, UK
c Photon Science Institute, University of Manchester, Manchester, M13 9PL, UK
d School of Chemistry, University of Manchester, Manchester, M13 9PL, UK
e School of Materials, University of Manchester, Manchester, M13 9PL, UK
Graphical abstract
1

Highlights
 Highly selective hydrogenation of benzoic acid to benzyl alcohol over
Pt/SnO2
 Aromatic ring hydrogenation hindered by alloy phases/migration of SnOx
onto Pt
 Acid carbonyl group activated by oxidised tin species at periphery of Pt
particles
 The interaction between Pt and SnO2 depends on temperature of reduction
 DFT modelling: switch in adsorption mode of benzoic acid with support
reduction
Abstract
Hydrogenation of benzoic acid in the liquid phase over Pt/SnO2 has afforded excellent
selectivity to benzyl alcohol under mild reaction conditions (97% at 98% conversion
of benzoic acid). DFT calculations have shown that the favoured mode of adsorption
of benzoic acid switched from the aromatic ring to the acid carbonyl group with
2

partial reduction of the SnO2 support. The switch in adsorption mode corroborates the
experimentally observed selectivity with acid carbonyl group adsorption and
hydrogenation occurring over Pt/SnO2. Reduction of Pt/SnO2 was observed by in-situ
XPS forming a degree of metallic tin and complete reduction of Pt which could lead
to Pt-Sn alloy phases and/or migration of SnOx onto Pt blocking sites for ring
hydrogenation. No hydrogenation of benzene or the benzene ring of benzoic acid was
observed over Pt/SnO2 with oxidic tin species at the interface of Pt particles proposed
to activate the acid carbonyl group for hydrogenation.
Keywords:
Benzoic acid, benzyl alcohol, selective hydrogenation, DFT, ATR-IR
1 Introduction
There is continuing interest in the development of heterogeneous catalysts for
the selective hydrogenation of carboxylic acids to form alcohols as platform
chemicals and intermediates in the production of fine chemicals and
pharmaceuticals.[1] Hydrogenation of saturated carboxylic acids and fatty acids in the
liquid phase to the corresponding alcohols has been reported over Ru, Pt and Re based
catalysts predominantly supported on reducible metal oxides such as TiO2.[1-4]
However, to chemoselectively hydrogenate the carbonyl group in an acid containing
additional and more reactive functionality presents an additional challenge.
Pd-Re/graphite catalysts have shown ability for hydrogenation of a range of
acid substrates to either alcohols or alkanes through control of the reaction conditions.
Reactions at 130 °C/20 bar H2 formed the corresponding alcohol while reaction at 160
°C/30 bar H2 lead to an alkane product[5]. This catalyst showed high yields of
alcohols from aliphatic acids and diols from diacids; however, for unsaturated acids
containing aromatic rings or C=C bonds (benzoic acid, furoic acids, oleic acid
(isolated C=C) and 9-decyncaeboxylic acid (isolated C≡C)) reaction of the
unsaturated functionality in addition to the acid carbonyl group occurred. Of note is
that separation of the aromatic ring from the acid functionality in the substrate, as in
phenylacetic acid, provided high selectivity to the aromatic alcohol product.
3

Phenylethanol was formed with a yield of 73% with no aromatic ring hydrogenation
observed. In addition, cinnamic acid showed reaction of the C=C and acid carbonyl
groups to form phenylpropanol with the aromatic ring left intact. This was attributed
to rapid reaction of the C=C group to yield phenylpropanoic acid which as for the
non-conjugated phenylacetic acid gave only reaction of the carboxylic acid group.
Unsaturated carboxylic acid hydrogenation has also been reported over
Re/TiO2 at 140°C/50 bar H2 where acid hydrogenations proceeded without reaction of
the aromatic ring for substrates such as phenylpropanoic acid [4, 6]. Interestingly, it
was found that Re/TiO2 did have an ability to hydrogenate an aromatic ring as was
demonstrated with the reaction of propylbenzene to propyl cyclohexane; however,
this reaction did not occur when in the presence of an acid or alcohol. This suggests
stronger adsorption of acids/alcohols compared to aromatic rings on Re which was
also evidenced from DFT calculations on Re(001) and ReOx surfaces which showed
relatively weaker adsorption of benzene and stronger adsorption of acetic acid
compared to Pd(111) and Rh(111) surfaces where benzene was adsorbed more
strongly [4]. For aromatic carboxylic acids, the higher affinity of Re and ReOx
surfaces towards adsorption of the acid carboxyl group provides a highly selective
reaction for acids such as phenylpropanoic acid. However, of note is that high
selectivity towards acid carbonyl hydrogenations for substrates where the acid
carbonyl group was conjugated with an aromatic ring, as in benzoic acid, was not
observed over PdRe/graphite (not tested for Re/TiO2). Hydrogenation of benzoic acid
over PdRe/graphite formed cyclohexylmethanol with a 72% yield showing the ease of
aromatic ring hydrogenation of this substrate compared to phenylacetic acid.
The selective hydrogenation of benzoic acid to cyclohexane carboxylic acid
has been shown to proceed with excellent selective towards the hydrogenation of the
aromatic ring being achieved under very mild reaction conditions, [7-12] whilst
PdReTiO2 (130 °C and 20 bar H2) and Rh/Al2O3 based catalysts (150 °C and 100 bar
H2) resulted in complete reduction to the saturated alcohol. [5, 13] (Scheme 1). High
selectivity to benzyl alcohol has been achieved homogeneously [14, 15] and in the gas
phase over Au supported on CeO2 and Ce0.62Zr0.38O2 catalysts at 250-300 °C with the
concentration of oxygen vacancies in the supports influencing direct or sequential
pathways to benzyl alcohol.[16] The nature of the metal and support influences the
activity and selectivity in the hydrogenation of benzoic acid [17]; however, to date,
there are few studies where both high activity and high selectivity to benzyl alcohol
4

have been reported showing that selective hydrogenation of the benzoic acid to benzyl
alcohol is still a challenging reaction over heterogeneous catalysts.
Scheme 1: Hydrogenation pathways of benzoic acid.
The highest selectivity reported, to date, for the liquid phase hydrogenation
of benzoic acid was over a Ru based catalyst with the addition of tin as a
promoter of selectivity [18]. Tin has also been used as a promoter of selectivity
for Ru and Rh based catalysts for the chemoselective hydrogenation of the
unsaturated carboxylic acid oleic acid[19, 20] and for selective C=O
hydrogenation in α,β-unsaturated aldehydes[21-23]
.
For benzoic acid, Ru-Sn/Al2O3, with 5% Ru with 11.7% Sn, prepared by
sequential impregnation (Sn addition to alumina with calcination before
addition of the Ru precursor) with optimal activity found for the catalyst
following high temperature calcination (973 K) and reduction (723 K), afforded
benzyl alcohol in 94% yield for reaction at 260 °C and 98 bar H2 (selectivity
quoted after 6 h of reaction). The interaction between Pt and Sn has been
reported to enhance catalytic activity and selectivity for Pt-Sn based catalysts
with metal precursor, catalyst preparation method and reduction temperature all
shown to influence the nature and degree of interaction formed between Pt and
Sn[24].
Herein, we report the chemoselective hydrogenation of benzoic acid over a
Pt/SnO2 catalyst prepared by a standard wet impregnation method. Pt supported
on SnO2 catalysts have been shown to improve the selectivity to C=O
5

hydrogenation for αβ-unsaturated aldehyde crotonaldehyde where formation of
PtSn alloy phases were proposed to be important for activity and selectivity of
the catalyst [21] Increased activity for crotonaldehyde hydrogenation was also
.
observed for PtSn/C catalysts where alloy formation and oxidised tin species
were proposed to provide the selectivity enhancement with alloy formation /
dilution of the Pt surface proposed to hinder C=C bond hydrogenation while
oxidised tin species activated the C=O group for hydrogenation [22].
In this study we report the selective hydrogenation of benzoic acid to benzyl
alcohol over Pt/SnO2 under significantly milder conditions, 190 °C and 30 bar H2,
than has been reported, to date, with no ring hydrogenation products observed.
2 Materials and methods
2.1 Catalyst preparation and characterisation
Both 5%Pt supported on alumina and tin(IV) oxide catalysts were prepared by wet
impregnation. Platinum (IV) nitrate (Alfa Aesar) precursor was added to γ-aluminium
oxide (255 m2g-1, Alfa Aesar) or SnO2 (6.70 m2g-1, Aldrich) in a methanol solution
and the mixture stirred for 12 h. After the solvent was evaporated, the paste was dried
in air for 4 h at 120 °C and then calcined in air for 4 h at 500 °C. Pt/C (10 wt% Pt)
was purchased from Sigma.
The surface area, total pore volume and average pore diameter of the synthesised
catalysts were measured by N2 adsorption-desorption isotherms at 77 K using
Micromeritics ASAP 2020. The pore size was calculated on the adsorption branch of
the isotherms using Barrett-Joyner-Helenda (BJH) method and the surface area was
calculated using the Brunauer-Emmett-Teller (BET) method. The composition of
these catalysts was determined using a Perkin Elmer Optima 4300 DV Inductively
Coupled Plasma Optical Emission Spectrometer (ICP-OES). The metals were
digested from the oxide supports using nitric acid and hydrochloric acid.
The powder X-ray diffraction technique was used for the characteristics of the
crystalline structure of catalysts, with a PANalytical X'Pert Pro MPD X-ray
diffractometer using Cu Kα irradiation. The diffractograms were measured from 20 to
80° (2), with a step size of 0.033° (2).
6

H2-TPR measurements were carried out using a Quantachrome ChemBet Pulsar
TPR/TPD, equipped with a thermal conductivity detector (TCD). Prior to the TPR
measurement, approximately 20 mg of the respective catalysts was loaded into a
quartz U-tube plugged with quartz wool. The sample was pre-treated under 10%
O2/He at 350 ºC for 60 min at a total flow of 40 cm3·min-1 and the temperature was
then decreased to -50 ºC, where 5% H2/Ar was introduced at the same flow rate. The
temperature was then increased from -50 to 800 °C at 10 °C·min-1.
CO chemisorption measurements were carried out using a Quantachrome ChemBet
Pulsar TPR/TPD, equipped with a thermal conductivity detector (TCD). Prior to the
adsorption of CO, the Pt/Al2O3 or Pt/SnO2 (30 mg) was reduced at either 200 or 350
°C under 5%H2/Ar at a total flow of 40 cm3·min-1. After reduction, the system was
purged with He and the temperature lowered. The adsorption of CO was performed at
-60 °C with the catalysts exposed to pulses of 10% CO/He repeatedly until no further
uptake was observed.
Transmission electron microscope (TEM) images, scanning transmission electron
microscope (STEM) images and energy dispersive X-ray spectroscopy (EDS)
spectrum images were obtained with an FEI Talos F200X microscope equipped with
an X-FEG electron source and Super-X SDD EDS detectors. The experiment was
performed using an acceleration voltage of 200 kV and a beam current of
approximately 1 nA. TEM images were recorded with an FEI CETA 4k × 4k CMOS
camera. STEM images were acquired with a HAADF detector.
Near ambient pressure (NAP) X-ray photoelectron spectroscopy (XPS) was carried
out on a Specs DeviSim NAP system (base pressure 1x10-9 mbar). The photoelectrons
were excited using a monochromated Al-Kα (1486.6 eV) source which was micro
focused to a 300 μm spot with analysis of the photoelectrons carried out using a Specs
Phoibos 150 differentially pumped electron analyser operating at 30 eV pass energy.
The 5% Pt-SnO2 powder was mounted to a gold foil substrate support prior to
loading into the vacuum environment. NAP-XPS analysis, H2 exposure and
temperature treatments were carried out in a highly controlled fashion within the
Specs DeviSim NAP cell which was docked onto the first differential aperture of the
Phoibos analyser during analysis.
7

XPS spectral fitting and analysis was carried out using Aanalyzer software in
conjunction with CasaXPS. Spectra were fitted using Voigt profiles (with the
exception of metallic species which were fitted using asymmetrical line shapes)
utilising Shirley backgrounds. All spectra were charge corrected to the gold foil
substrate signal (Au4f7/2 at 84 eV) with the minor adventitious carbon signal also
noted to remain at the accepted value of 284.8 eV throughout the course of the
analysis.
Reduction of the metal oxide species was carried out by flowing hydrogen into the
NAP cell up to a pressure of 1 mbar and subsequently raising the sample temperature
from room temperature (~300 K) to 475 K during hydrogen flow where it was held at
this target temperature for 15 minutes.
DRIFT spectra were recorded over the catalyst using a Bruker Tensor 27 FTIR
spectrometer with a Spectra Tech Collector II DRIFTS accessory with environmental
chamber allowing spectra to be recorded under gas environments and at elevated
temperatures.[25] The Pt/Al2O3 and Pt/SnO2 catalysts were reduced at reduction
temperatures between 100 and 350 °C for 1 h in 10%H2 in Ar. The gas feed was then
switched to Ar to purge the cell before cooling to room temperature for the adsorption
of CO (2% CO in Ar, 50 mL·min-1). Spectra were recorded with 128 scans at 4 cm-1
and the background was the fresh catalyst before reduction.
ATR-IR spectra were recorded over alumina and Pt/Al2O3 layers slurried onto a
ZnSe crystal in water. Spectra were recorded using a PIKE ATRMaxII accessory
housed in a Bruker Tensor II spectrometer. All layers were initially heated to 90 °C
under a flow of hydrogen saturated dioxane for 30 min before cooling and switching
to 0.1 M benzoic acid in dioxane solution. Spectra were recorded with 8 scans at 4
cm-1 resolution and the background was the fresh catalyst layer before introduction of
solvent.
2.3 Hydrogenation reactions
All hydrogenation reactions were performed in a 100 mL Parr batch reactor and the
typical reaction conditions were 0.002 mol substrate (benzoic acid or benzene) in 40
mL 1,4-dioxane, 190 °C and 30 bar H2 (30 bar total reaction pressure for reaction at
8

190 °C with a partial pressure of H2 of 20 bar) over 100 or 200 mg of the respective
catalysts with the reaction stirred at 1500 rpm. In addition, reactions at varied H2
partial pressures (5-20 bar) and concentration of benzoic acid (0.001-0.003 mol) were
also performed. Before each reaction, the catalyst was pre-reduced in situ for 1 h at
200 °C at 30 bar H2 in dioxane. Additional reactions following a gas phase reduction
of the Pt/SnO2 catalyst under a flow of H2 at 200 °C and 400 °C were performed. The
flow reduction at 200 °C was performed in the Parr autoclave with no solvent for 1
hour after which 40 ml of benzoic acid solution (0.05 M) was syringed into the reactor
through the liquid sampling port while still under H2 flow. The reactor was then
purged with H2 and the reaction performed (100 mg Pt/SnO2, 30 bar H2 and 190 °C).
The reduction at 400 °C for 1 hour was carried out in a furnace under a flow of H2.
After the reduction, the furnace temperature was lowered and the system purged with
N2 before rapid transfer to the autoclave for reaction again performed at 30 bar H2 and
190 °C.
Samples were taken from the reactor at different time intervals and analysed by gas
chromatography to monitor the compositional change with reaction time and a mass
transfer study was performed to ensure that reactions were under intrinsic kinetic
control. The gas chromatograph used for composition analysis of the reaction samples
was a Perkin Elmer Clarus 500 with a Phenomenex Zebron ZB-WAX capillary
column, utilising an FID. Calibration curves for benzoic acid and products
benzaldehyde, benzyl alcohol, benzyl benzoate and cyclohexylcarboxylic acid were
performed to allow quantitative analysis of the reactions with determination of the
reaction mass balance.
The recycle experiment was performed with 0.002 mol substrate in 40 mL 1,4-
dioxane, 30 bar H2 and 190 °C, over 200 mg of the catalyst with the reaction stirred at
1500 rpm. The catalyst was recovered by filtration, washed with dioxane and dried in
air. Before each reaction, the recovered catalyst was re-reduced as for the first
reaction (200 °C for 1 h under 30 bar H2). In addition, a recycle experiment whereby
the catalyst did not undergo reductive regeneration between reactions was performed.
The first reaction ran for 4 hours before the reaction was stopped and the liquid
removed before addition of dioxane and 0.002 mol of benzoic acid and a reaction at
190 °C and 30 bar H2 performed for a second time.
9

2.4 DFT calculations
DFT calculations were carried out with a periodic slab model using the Vienna ab
initio simulation program (VASP).[26-29] The generalized gradient approximation
(GGA) was used with the Perdew-Burke-Ernzerhof (PBE)[30] exchange-correlation
functional. The projector-augmented wave (PAW) method [31, 32] was utilized to
describe the electron-ion interactions, and the cut-off energy for the plane-wave basis
set was 450 eV. Furthermore, to illustrate the long-range dispersion interactions
between the adsorbates and catalyst surfaces, we employed the van der Waals density
functional (vdW-DF) method, which was originally proposed by Dion et al.[33] and
further developed by Thonhauser [34], Michaelides, [35, 36] Román-Pérez [37] and
some others. Exchange functional, optB88-vd, [36, 37] were chosen, which were
reported to be reasonably accurate for surface adsorption calculations.[38, 39]
Brillouin zone integration was accomplished using a 2×2×1 Monkhorst-Pack k-point
mesh. All the adsorption geometries were optimized using a force-based conjugate
gradient algorithm, while transition states (TSs) were located with a constrained
minimisation technique.[40-42]
Three different surfaces were considered in this work, namely the close-packed
platinum surface (Pt(111)), platinum on tin oxide (Pt/SnO2), and platinum on reduced
tin oxide (Pt/SnO2-x).
For the modelling of Pt(111), a typical surface model with a periodic four-layer
p(4×4) model with the 2 lower layers fixed and 2 upper layers relaxed were used. For
the modelling of platinum on tin oxide, we used a global optimization method,
namely the particle swarm optimization (PSO), to obtain the most stable structures of
cluster-oxide structures. Comparing to some previous modelling of bi-component
systems,[43, 44] this global optimization algorithm can provide a much more realistic
and accurate illustration of cluster and interface structures, which may be of great
importance in the selective hydrogenation of benzoic acid. In this work, the PSO were
carried out using the CALYPSO code,[45-47] which is a power tool to get the stable
structures at given external conditions. The unreduced tin oxide was modelled with a
six-layer model with 3 upper layers relaxed and 3 lower layers fixed, while the
reduced surface was modelled by removing the bridge oxygen of unreduced tin oxide.
For both Pt/SnO2 and Pt/SnO2-x, Pt clusters with twelve platinum atoms were chosen,
and the structures of cluster and interface were obtained using PSO. Twenty
10

optimization generations were calculated with sixteen surfaces in each generation.
Thus, overall 320 structures were examined for both systems, and the most stable
surfaces were chosen.
3 Results and discussion
3.1 Hydrogenation of benzoic acid
The hydrogenation of benzoic acid over Pt/SnO2 gave excellent selectivity to
benzyl alcohol with selectivity of 97% at 98% conversion of benzoic acid
(Figure 1). This reaction was performed at 190 °C and 30 bar H2 which are
considerably milder reaction conditions for hydrogenation of benzoic acid than
have been previously reported.[18]
Trace amounts of intermediate
hydrogenation product benzaldehyde was also detected (Figure 1b) with a
maximum of 16% selectivity to benzaldehyde observed only at the very start of
the reaction while esterification product benzyl benzoate was only formed in
trace amounts <1% after 24 hour of reaction. Under these reaction conditions,
Pt/SnO2 was found to be reusable with no significant change in the conversion
of benzoic acid or the selectivity to benzyl alcohol observed for three recycle
reactions with a reduction at 200 °C, 30 bar H2 for 1 hour performed before
each recycle reaction (Figure S1). There was no activity observed when the
catalyst was recycled without the reduction step.
Interestingly, hydrogenation of the aromatic ring was not observed over Pt/SnO2
under any of the reactions conditions used in this study. In addition, when benzene
was used as the substrate, no hydrogenation was observed. Adsorption/reaction of
aromatic rings was hindered on Pt when supported on SnO2. The promoting effect of
SnO2 as the support was also highlighted by the differing reactivities/selectivities that
were observed for Pt supported on carbon and alumina. For a commercial Pt/C
catalyst, lower activity compared to Pt/SnO2 was observed with only 24% conversion
after 24 hours of reaction. The major product of the reaction over Pt/C was
cyclohexane carboxylic acid with selectivities of 81.4% to cyclohexane carboxylic
acid, 14.9% to benzyl alcohol and 3.7% to cyclohexylmethanol (Figure S2). Even
lower activity was observed for reaction over a 5%Pt/Al2O3 catalyst where only 3%
conversion was observed after 24 h of reaction at 190 °C and 30 bar H2 (Figure S3)
11

and reaction over the SnO2 support alone gave only 2% conversion in 24 h of reaction
likely due to lack of activation of hydrogen without platinum (reaction at 190 °C and
30 bar H2). The activity for acid carbonyl hydrogenation is clearly promoted by an
interaction between Pt and SnO2.
Figure 1: (a) reaction profiles for the hydrogenation of benzoic acid over Pt/SnO2 at
190 °C and 30 bar H2, catalyst (200 mg) pre-reduced in situ at 200 °C and 30 bar H2
for 1 h before reactions; benzoic acid,  benzyl alcohol and  benzaldehyde with
 showing the sum of the concentration of reagent and products through the reaction
(b) %selectivity vs. %conversion for hydrogenation of benzoic acid for reactions at
190 °C/30 bar (closed symbols) and 150°C/15 bar (hollow symbols).
The acid carbonyl group activation ability of this catalyst was further highlighted by
the formation of benzyl alcohol in a reaction performed at only 150 °C and 15 bar H2.
However, at these milder conditions, an effect of adsorption was observed with
comparable initial rates of removal of the benzoic acid observed for the reactions
under H2 and N2. However, the formation of benzyl alcohol was only observed for
the reaction under H2 (Figure S4) and at a lower rate than the removal of benzoic acid
indicating that the latter is dominated by adsorption at these milder conditions.
Variation of the reaction pressure partial pressures of H2 of 5, 10 and 20 bar
(total reaction pressures of 15, 20 and 30 bar) showed an increase in initial rate of
production of benzyl alcohol from 0.028 to 0.048 mmol dm-3 min-1 with increasing
pressure. The initial rates of reaction were determined from the formation of benzyl
alcohol as opposed to the conversion of benzoic acid due to the strong adsorption
12

observed in the initial period of the reaction. The order of reaction with respect to
hydrogen was determined from a plot of ln(initial rate) against ln(H2 pressure) and
was found to be of fractional order with a value of 0.4 (Figure 2).
The initial rate of reaction was also found to vary with the concentration of
benzoic acid in the reactor (0.001 to 0.003 moles) with the rate increasing with
concentration. The order of reaction with respect to benzoic acid was found to be 0.8
(Figure 3).
0.050
0.045
0.040
0.035
0.030
0.025
-9.8
-10.0
-10.2
-10.4
-10.6
y = 0.40x - 11.13
R2 = 0.9992
0
5
10
15
20
25
1.5
2.0
2.5
Ln (pressure)
3.0
3.5
Partial pressure H2 (bar)
Figure 2: (a) initial rate of reaction (formation of benzyl alcohol) over Pt/SnO2 at 190
°C and 5 - 20 bar partial pressure of H2, catalyst (100 mg), 0.002 mol benzoic acid, 40
ml dioxane; and (b) ln(initial rate) against ln(H2 partial pressure). The catalyst was
pre-reduced in situ at 200 °C and 30 bar H2 for 1 h before reactions.
-9.0
-9.5
0.08
0.06
0.04
0.02
0.00
y = 0.83x - 7.40
R2 = 0.9975
-10.0
-10.5
-4.0
-3.5
-3.0
-2.5
0.00
0.02
0.04
0.06
0.08
Concentration (mol/dm-3)
Ln(concentration)
13

Figure 3: (a) initial rate of reaction (formation of benzyl alcohol) over Pt/SnO2 at 190
°C and 20 bar partial pressure of H2, catalyst (100 mg), varying the concentration of
benzoic acid between 0.001 and 0.003 mol, 40 ml dioxane and (b) ln(initial rate)
against ln(benzoic acid concentration). The catalyst was pre-reduced in situ at 200 °C
and 30 bar H2 for 1 h before reactions.
3.2 Catalyst characterisation
Characterisation of the properties of the Pt/SnO2 catalyst (Table S1) was
performed and compared to that of Pt/Al2O3 which was inactive for the
hydrogenation of benzoic acid. While Pt/SnO2 has a significantly lower surface
area than Pt/Al2O3 (9.3 m2 g-1 and 235 m2 g-1 respectively) TEM analysis
showed comparable Pt particle sizes of 1-2 nm for both catalysts (Figures S5-
S6). XRD of the catalysts showed only peaks due to the supports which
corroborates the small size of the Pt particles observed by TEM (Figure S7).
This indicates that the surface area and particle size effects/dispersion are not
the main influence on the activity of these catalysts and again suggests that an
interaction between the metal and the support is important for the activity of
Pt/SnO2.
TPR of Pt/Al2O3 and PtSnO2 catalysts are shown in Figure 4. For PtSnO2,
the main reduction features occurred between 200-400 °C due to reduction of
PtOx strongly interacting with the support and also possibly some partial
reduction of SnO2 most likely at the interface with the Pt nanoparticles (Figures
4 and S8).[48, 49] A small reduction feature at much lower reduction
temperatures (below 100 °C) was also observed which could be due to
reduction of PtOx weakly interacting with the Pt/SnO2 support. Bulk SnO2
reduction occurred above 500 °C for Pt/SnO2 while for the support alone, bulk
reduction of SnO2 occurred above 650 °C (Figure S8). The reduction of PtOx
for the Pt/Al2O3 catalyst was observed to be complete at temperatures below
300 °C with the main reduction peak occurring at 110 °C, showing the weaker
interaction of Pt with the alumina compared to the tin dioxide support.
14

Figure 4: TPR of Pt/SnO2 (dashed line) and Pt/Al2O3 (solid line).
To probe the oxidation state of the Pt and Sn in Pt/SnO2, XPS analysis was
performed on the fresh Pt/SnO2 catalyst and after reduction at 200 °C (1 mbar
of H2 pressure for 15 min). XP spectra of the Pt4f region for Pt/Al2O3 showed
only a single Pt species with Pt4f7/2 peak at 73.4 eV due to Pt(II) species (Figure
S9). For Pt/SnO2, Pt was predominantly in an oxidised state with Pt4f7/2 peaks
at 74.7 eV and 72.8 due to Pt(IV) and Pt(II) species with a minor contribution
due to metallic Pt at 71.2 eV also observed (Figure 5)[50, 51] which has also
been reported for PtSn/C catalysts.
In order to investigate the effect of the reduction procedure on the Pt/SnO2
catalyst, NAP-XPS was employed to simulate the reduction procedure utilising
an in-situ approach preventing ambient exposure of the sample prior to analysis
which can alter the true chemistry of the reduced surface. Figure 5 displays the
Pt4f regions following the reduction where a dramatic change in peak shape
was noted for the Pt4f profile with only metallic Pt observed in the sample
surface after reduction.
In addition to a change in the Pt4f profile following reduction, a change was
also noted in the Sn3d profile as displayed in Figure 6. A single doublet within
the Sn3d window was noted for the fresh catalyst upon loading, consistent with
a SnO2 state (binding energy for Sn3d5/2 of 486.9 eV). Following the reduction,
15

two key features are noted for the Sn3d spectrum. Firstly, a shift to higher
binding energy is noted for the oxidised Sn peaks (~ 0.3 eV) relative to the as
loaded state. This shift in B.E. could be explained as a result of band bending,
as seen in previous SnO2 studies via XPS [52].
Secondly, a shoulder feature emerges (binding energy for Sn3d5/2 of 485.6
eV) from the main peak doublet with a separation of 1.4 eV on the lower
binding energy side from the main feature, consistent with the shift separation
expected between SnO2 and Sn metal. Previous studies of metallic Pt deposited
on a SnO2 surface by Körber et al have shown that metallic platinum can reduce
the SnO2 surface to metallic Sn [53] and formation of metallic tin has also been
observed for a range of Pt-Sn catalysts with Sn 3d5/2 peaks at 485.6 eV reported
to be in the range for metallic tin in an alloy phase [54]. The proportion of
metallic tin after reduction was below 10% highlighting that tin is
predominantly in an oxidised state even after reduction at 200 °C.
The presence of both metallic tin and metallic platinum could suggest
formation of alloy phases which have been reported for Pt/SnO2 and Pt-
Sn/metal oxide catalysts. XRD of Pt/SnO2 after reduction at 170/200 °C
showed formation of a PtSn phase; [21] however, reduction of a 20%Pt/SnO2
catalyst at 200 °C did not show formation of intermetallic compounds by XRD
[48]. Interestingly both these studies showed higher reduction temperatures
(250 and 400 °C respectively) to generate tin rich alloy phases (PtSn2 and
PtSn4) with lower activity observed following this high temperature reductive
treatment. However, study of crotonaldehdye hydrogenation over pure Pt(111)
and model Sn/Pt(111) alloy surfaces showed only an enhancement in the
activity for the alloy surfaces compared to the Pt(111) surface with no change
in selectivity [55]. This suggests that alloy formation alone cannot account for
the enhancements in selectivity for C=O hydrogenations
Pt-support
interactions as well as Pt-Sn interactions required for control of the activity and
selectivity which are likely influenced by catalyst preparation method as well as
the temperature of reduction. The activation of C=O bonds by ionic (oxidic) tin
species has been reported for PtSn catalysts prepared by surface organometallic
reactions which have been shown to be active and selective for hydrogenation
of furfural to furfural alcohol[56] and cinnamaldehyde to cinnamyl alcohol
16

[57]. For these catalysts, Pt-Sn alloy phases and ionic tin species were
observed on the catalyst with ionic tin proposed to activate the C=O bond of the
aldehydes to hydrogenation.
Figure 5: XP spectra of the Pt 4f region for Pt/SnO2 as loaded and after
reduction.
17

Figure 6: XP spectra of the Sn 3d region for Pt/SnO2 as loaded and after
reduction.
CO pulse chemisorption and was performed to study the effect of reduction
temperature on the CO adsorption capacity of Pt. For Pt/Al2O3 reduced at 200
°C, CO pulse chemisorption gave a dispersion of 55% which decreased to 16%
following reduction at 350 °C which is likely due to sintering of the Pt
particles. For the Pt/SnO2 catalyst, significantly lower adsorption of CO was
observed (calculated dispersion was only 0.72%) following reduction at 200 °C
while no CO uptake occurred when reduced at 350 °C. The lower
chemisorption ability of Pt when supported on SnO2 compared to Al2O3 is
indicative of the interaction between Pt and Sn which could be due to dilution
effects from alloy formation and/or migration of SnOx species onto the Pt
surface.
DRIFT spectroscopy was also utilised to probe the electronic state of Pt in
the Pt/SnO2 catalyst. The adsorption of CO at room temperature following
reduction at temperatures between 100 and 350 °C was performed over Pt
supported on SnO2 and compared to Pt supported on Al2O3.
For Pt/SnO2, in the temperature region where reduction of PtOx and partial
reduction of interfacial SnO2 occurs (200-400 °C from TPR), the chemisorption
ability of Pt for CO was significantly modified (Figure 7). For reduction below
200 °C, weak Pt-CO bands were observed with a band at 2086 cm-1 for Tred of
100 °C, 2080 cm-1 for Tred of 150 °C shifting to 2075 cm-1 for Tred of 250 °C.
These bands are assigned to CO linearly adsorbed on terrace sites of Pt
particles.[58] No bands due to CO adsorption on the support were observed
under any conditions used herein.
The shifting of the Pt-CO band from 2086 - 2075 cm-1 could be a result of
restructuring of the Pt nanoparticle with more coordinatively unsaturated sites
forming or the generation of electron rich Pt sites (increased backbonding into
CO antibonding orbitals weakening the CO bond and red-shifting the Pt-CO
band) following reduction at 200 °C.[59] Alternatively, the red shift (2086 to
2075 cm-1) could be a result of a geometric effect where alloy formation or
migration of SnO2/SnO2-x species onto the Pt could block sites for CO
18

adsorption giving a lower coverage of CO and a reduction in dipole-dipole
coupling which shifts the band position to lower wavenumbers.
Figure 7: DRIFT spectra recorded under 2% CO in Ar at room temperature over
Pt/SnO2 following reduction in 10% H2/Ar between (a) 100 and 200 °C and (b) 200
and 350 °C.
Following reduction at 200 °C, the intensity of the linear Pt-CO band at 2075
cm-1 increased and a new linear Pt-CO band was observed at 2040 cm-1 suggesting
modification of the platinum surface with reduction of the catalyst. Similar
19

spectra were observed for Pt-Sn/TiO2 catalysts with bands at 2072 cm-1 with a
shoulder at 2040 cm-1 [60]. A band at 2050 cm-1 has also been observed for Pt-
Sn alloyed catalysts[58, 60] and assigned to CO adsorbing on Pt sites belonging
to a Pt-Sn phase. In the study of Moscu et al. it was highlighted that the PtSn
phases could be intermetallic compounds such as PtSn or Pt3Sn or a random
alloy and interestingly, they showed that the alloy phases could segregate and
be regenerated with variation in the temperature. Under the CO/H2 feed at
temperatures above 175 cm-1, a single band for CO adsorbed on Pt-Sn species
was observed between 2054-2045 cm-1 while at lower temperatures (<175 °C)
segregation of the Pt-Sn alloy phases occurred with the Pt-CO band position
shifting to 2071 cm-1 indicating adsorption of CO on essentially pure Pt sites.
Herein, the formation of the band at 2040 cm-1 between 150-200 °C
reduction temperature, could be due to CO adsorption on a Pt-Sn alloy phase
which corroborates the study of Liberková and Touroude [21]where Pt-Sn alloy
phases were observed to form at 170 °C
.
The Pt-CO band at 2075 cm-1 (CO adsorbed on essentially pure Pt sites)
continued to shift to lower wavenumbers upon increasing the temperatures until
a reduction of 350 °C wherein, Pt-CO bands were no longer detected. DRIFT
spectra of the adsorption of CO on Pt/SnO2 show that both Pt and Pt-Sn sites
are present on the surface of the catalyst with reduction of the catalyst above
200 °C resulting in significant perturbation of the chemisorption ability of the
catalyst which is likely due to encapsulation of platinum by SnO2/SnO2-x
species or formation of tin-rich alloy phases.[48, 61].
The characterisation of the catalysts highlights that the temperature of
reduction for Pt/SnO2 can significantly alter the interaction between Pt and Sn.
However, all the techniques used to characterise the catalysts were performed
under a flow of H2 (gas phase reduction) which differs from the in-situ
reduction in the liquid phase performed in the catalytic activity tests. As the
reduction is essential for the state of the catalyst, the hydrogenation of benzoic
acid after gas phase flow reductions at 200 °C (performed in the autoclave
without solvent) and at 400 °C (ex-situ reduction in a furnace) were also
performed.
20

For the flow reduction at 200 °C, the conversion of benzoic acid was slightly
lower which could be due to mass transfer limitation of H2 to the catalyst for
flow in the Parr reactor (no solvent/mixing) (Figure S10). Importantly, the
same selective formation of benzyl alcohol was observed which suggests that
the active state of Pt/SnO2 was comparable following both reductive treatments
at 200 °C.
For the reduction of Pt/SnO2 at 400 °C, the hydrogenation activity of
Pt/SnO2 was significantly reduced with only 3.5% conversion after 24 h of
reaction at 190 °C and 30 bar H2 forming benzyl alcohol, benzaldehyde and
benzyl benzoate. Figure 11 shows the %Selectivity vs. %Conversion profile for
this reaction where slow conversion of benzaldehyde to benzyl alcohol is
observed (%S = 73%, 16% and 11% to benzyl alcohol, benzaldehyde and
benzyl benzoate after 24 h of reaction). Reduction at 400 °C has significantly
impaired the activity of the catalyst compared to the standard 200 °C reduction
which is likely related to decreased hydrogen dissociation when Pt is
encapsulated by SnOx.
The Pt assisted reduction of SnO2 is essential in providing the activity of this
catalyst. The formation of an alloy phase/ migration of SnOx onto Pt could
block sites for hydrogenation of the aromatic ring with oxidised tin species
(SnOx) likely at the periphery of the Pt nanoparticle proposed to activate the
acid carbonyl group for hydrogenation.
3.3 DFT calculations
To probe whether partially reduced (SnOx) species can activate the C=O
bond for hydrogenation, DFT calculations of the mode and energy of
adsorption of benzoic acid on Pt(111) and Pt supported on tin dioxide were
performed.
To model the platinum on tin dioxide, a global optimization method was
used, namely particle swarm optimization (PSO), to obtain the most stable Pt-
SnO2 cluster-oxide structures (Table S2 and Figure S12). In comparison with
previous modelling of bi-component systems,[43, 44] this global optimization
21

algorithm can provide a much more realistic and accurate illustration of cluster
and interface structures, which is required in order to study catalytic systems
where the activity is related to support-metal interactions.
For each system, the adsorption energy was calculated for benzoic acid
adsorption though either the aromatic ring or the acid carbonyl group (Table 1).
For the Pt(111) surface, π adsorption through the aromatic ring was found to be
considerably more favoured than σ adsorption through the acid carbonyl group
with adsorption energies of -1.25 and -0.89 eV respectively (Figure 8). The
adsorption energy of benzoic acid on Pt/SnO2 showed a comparable trend to
that of benzoic acid on Pt(111) with the π interaction favoured with adsorption
energies of -1.25 and –0.89 eV for Pt(111), and -2.02 and -1.28 eV for Pt/SnO2
(Figure 8 and S13). The favoured π mode of adsorption indicates that
hydrogenation of the aromatic ring would be the expected pathway over both
these surfaces; however, this was not observed experimentally. This suggests
that reduction to the metal alone cannot produce an active catalyst therefore
reduction of the support was also considered in the modelling of benzoic acid
adsorption in this catalytic system.
The adsorption of benzoic acid on a reduced Pt/SnO2-x surface was modelled
by removing the bridging oxygen from tin dioxide. This partial support
reduction resulted in a change in the most stable Pt cluster structure and
interestingly, a switch in the favoured mode of adsorption of benzoic acid with
the σ adsorption mode now more favourable than the π interaction with binding
energies of -0.98 and -0.74 eV, respectively (Figures 8 and S13). Reduction of
the tin dioxide led to a significant weakening of the π mode of adsorption of
benzoic acid compared to when adsorbed on the Pt(111) and Pt/SnO2
(unreduced tin dioxide) surfaces. Modelling of benzoic acid adsorption on
platinum with a strong interaction with the tin dioxide support resulted in
activation of the acid carbonyl group in favour of the aromatic ring and showed
that partial reduction of SnO2 could influence the reaction selectivity in line
with the activity data and catalyst characterisation results.
22

Table 1: DFT calculated adsorption energy for benzoic acid on Pt(111),
unreduced Pt/SnO2 and partially reduced Pt/SnO2-x
Adsorption energy / eV
σ adsorption
π adsorption
Pt(111)
-0.89
-1.25
Pt/SnO2
-1.28
-0.98
-2.02
-0.74
Pt/SnO2-x
Figure 8: Favoured mode of adsorption of benzoic acid over (a) Pt(111), (b)
Pt/SnO2, unreduced catalyst, and (c) Pt/SnO2-x reduced surface. Oxygen, carbon
and hydrogen atoms are in red, grey, and white, respectively.
3.4 Hydrogenation of benzoic acid over Pt/Al2O3
DFT calculations on Pt(111) and Pt on SnO2 (without reduction) showed that
benzoic acid adsorption occurs through the aromatic ring which has also been
reported experimentally for Pt(111).[62] It is interesting therefore, that no
hydrogenation activity was observed over the Pt/Al2O3 catalyst as a degree of
activity towards cyclohexane carboxylic acid could be expected as was
observed over Pt/C (Figure S2). To investigate this, in-situ ATR spectroscopy
was used to study the adsorption of benzoic acid over Pt/Al2O3.
23

A Pt/Al2O3 layer, pre-reduced under a flow of H2-saturated dioxane at 90 °C
was exposed to a 0.1 M benzoic acid in dioxane solution at room temperature
(Figure 9 and S15). In the spectra over both the catalyst (Figure 9), and the
alumina support alone (Figure S16) adsorption of benzoic acid formed
bidentate benzoate species from dissociative adsorption of benzoic acid on
alumina. Bands due to the benzoate species were observed at 1604, 1552 and
1397 cm-1 and assigned to a ring stretch of the benzoate and the carboxylate
asymmetric and symmetric stretches respectively.[63] Bands at 1723 cm-1 and
1239 cm-1 are due to liquid phase benzoic acid dissolved in the dioxane.[64] In
addition, a band at 2053 cm-1 was observed and assigned to CO linearly
adsorbed on metallic Pt. A band at 2053 cm-1 was also observed following
adsorption of CO on a pre-reduced Pt/Al2O3 catalyst layer (Figure S17). The
formation of CO during benzoic acid adsorption over Pt/Al2O3 could result
from decomposition or decarbonylation reactions. Decarbonylation of benzyl
alcohol[65, 66] and benzaldehyde[67] has been reported over Pd/Al2O3 in the
liquid phase at comparable temperatures to those used in this ATR study. Ferri
et al. reported the desired oxidation of benzyl alcohol to benzaldehyde to be
insensitive to specific sites on the Pd particles; however, decarbonylation was
found to occur over hollow sites on terraces of the Pd nanoparticles
(Pd(111))[68] with the presence of oxygen in the feed reducing the deactivating
effect of CO adsorbed on Pd for this reaction.
Under the hydrogenation conditions used within this study, CO (and possibly
hydrocarbon fragments from decomposition of benzoic acid/benzaldehyde)
could remain adsorbed on the Pt surface poisoning the catalyst for subsequent
adsorption of benzoic acid resulting in the low activity observed experimentally
over Pt/Al2O3. The reducible nature of the tin dioxide support is able to
promote both the activity and selectivity to the acid carbonyl hydrogenation of
benzoic acid suppressing hydrogenation of the aromatic ring and side
reactions/decomposition.
24

Figure 9: In-situ ATR spectra recorded during adsorption of 0.1 M benzoic acid in dioxane over
Pt/Al2O3 at room temperature (3 min of exposure). Pt/Al2O3 was pre-reduced in a flow of H2-
saturated dioxane at 90 °C. Bands due to dioxane solvent have been subtracted. * Bands due to
bidentate benzoate species adsorbed on alumina:
4 Conclusions:
In conclusion, platinum supported on tin dioxide is active and highly
selective for the formation of benzyl alcohol from benzoic acid under extremely
mild reaction conditions in the liquid phase. The interaction between Pt and
SnOx provides activation of the acid carbonyl group with the catalysts showing
no activity for aromatic ring hydrogenation as further evidenced by the
inactivity for hydrogenation of benzene. Characterisation of the catalyst shows
a degree of reduction of the SnO2 support following treatment under H2 at 200
°C however, tin is present predominantly as an oxidised species although with
the formation of both metallic Pt and Sn, alloy phases could also be generated.
DFT suggests that interaction between the Pt and SnOx can weaken the strength
of adsorption of the aromatic ring of benzoic acid rendering its adsorption
unfavourable relative to the acid carbonyl group. The switch in adsorption
mode with partial reduction of the tin dioxide corroborates the observed
25

reaction selectivity with ring hydrogenation observed over Pt/C, ring adsorption
and decomposition observed over Pt/Al2O3 and acid carbonyl group adsorption
and hydrogenation over Pt/SnO2 with excellent selectivity to benzyl alcohol.
Author contribution
The contribution of the authors is as follows:
Preparation and characterisation of catalysts (BET, XRD, ICP, TPR, CO chemisorption):
Xiaohan Chen and Richard Morgan
Activity testing: Xiaohan Chen and Helen Daly
DFT calculations: Ziyun Wang and Peijun Hu
XPS and NAP-XPS data collection and analysis: Rebecca Taylor, Conor Byrne and Alex Walton
DRIFTS and ATR data collection and analysis: Helen Daly
TEM: Mathew Smith
Project ideas and discussion of results: Chris Hardacre, Helen Daly, Haresh Manyar and
Robbie Burch
Writing the paper: Xiaohan Chen, Ziyun Wang, Helen Daly and Chris Hardacre
Acknowledgements
The authors would like to acknowledge EPSRC for funding as part of the CASTech
grant (EP/G011397/1). XC would like to acknowledge the Postgraduate Research
Studentship from Queen’s University Belfast and the Research Scholarship from The
University of Manchester. This work was supported by the Henry Royce Institute for
Advanced Materials, funded through EPSRC grants EP/R00661X/1 and
EP/P025021/1. The authors acknowledge the use of the School of Materials X-ray
Diffraction Suite at the University of Manchester and is grateful for the technical
support of Mr Gary Harrison and Dr John E. Warren.
References
26

[1] J. Pritchard, G.A. Filonenko, R. van Putten, E.J. Hensen, E.A. Pidko, Chem. Soc. Rev., 44 (2015)
3808-3833. https://doi.org/10.1039/c5cs00038f
[2] H.G. Manyar, C. Paun, R. Pilus, D.W. Rooney, J.M. Thompson, C. Hardacre, Chem. Commun., 46
(2010) 6279-6281. https://doi.org/10.1039/c0cc01365j
[3] B. Rozmyslowicz, A. Kirilin, A. Aho, H. Manyar, C. Hardacre, J. Warna, T. Salmi, D.Y. Murzin, J
Catal, 328 (2015) 197-207. https://doi.org/10.1016/j.jcat.2015.01.003
[4] T. Toyao, S.M. Siddiki, A.S. Touchy, W. Onodera, K. Kon, Y. Morita, T. Kamachi, K. Yoshizawa,
K.I. Shimizu, Chemistry, 23 (2017) 1001-1006. https://doi.org/10.1002/chem.201604762
[5] J. Ullrich, B. Breit, ACS Catal., 8 (2018) 785-789. https://doi.org/10.1021/acscatal.7b03484
[6] T. Toyao, K.W. Ting, S.M.A.H. Siddiki, A.S. Touchy, W. Onodera, Z. Maeno, H. Ariga-Miwa, Y.
Kanda, K. Asakura, K.-i. Shimizu, Catal. Sci. Technol.,
https://doi.org/10.1039/c9cy01404g
9
(2019) 5413-5424.
[7] P. Maki-Arvela, J. Kuusisto, E.M. Sevilla, I. Simakova, J.P. Mikkola, J. Myllyoja, T. Salmi, D.Y.
Murzin, Appl. Catal., A, 345 (2008) 201-212. https://doi.org/10.1016/j.apcata.2008.04.042
[8] M. Tang, S. Mao, M. Li, Z. Wei, F. Xu, H. Li, Y. Wang, ACS Catal., 5 (2015) 3100-3107.
https://doi.org/10.1021/acscatal.5b00037
[9] M. Tang, S. Mao, X. Li, C. Chen, M. Li, Y. Wang, Green Chem., 19 (2017) 1766-1774.
https://doi.org/10.1039/c7gc00387k
[10] X. Xu, M. Tang, M. Li, H. Li, Y. Wang, ACS Catal.,
https://doi.org/10.1021/cs500859n
4
(2014) 3132-3135.
[11] R. Nie, H. Jiang, X. Lu, D. Zhou, Q. Xia, Catal. Sci. Technol., 6 (2016) 1913-1920.
https://doi.org/10.1039/C5CY01418B
[12] R. Nie, M. Miao, W. Du, J. Shi, Y. Liu, Z. Hou, Appl. Catal., B, 180 (2016) 607-613.
https://doi.org/10.1016/j.apcatb.2015.07.015
[13] D.H. He, N. Wakasa, T. Fuchikami, Tetrahedron Lett., 36 (1995) 1059-1062.
https://doi.org/10.1016/0040-4039(94)02453-I
[14] T. vom Stein, M. Meuresch, D. Limper, M. Schmitz, M. Holscher, J. Coetzee, D.J. Cole-Hamilton, J.
Klankermayer,
W.
Leitner,
J.
Am.
Chem.
Soc.,
136
(2014)
13217-13225.
https://doi.org/10.1021/ja506023f
[15] T.J. Korstanje, J.I. van der Vlugt, C.J. Elsevier, B. de Bruin, Science, 350 (2015) 298-302.
https://doi.org/10.1126/science.aaa8938
[16] N. Perret, X. Wang, J.J. Delgado, G. Blanco, X. Chen, C.M. Olmos, S. Bernal, M.A. Keane, J. Catal.,
317 (2014) 114-125. https://doi.org/10.1016/j.jcat.2014.06.010
[17] S.B. Shinde, R.M. Deshpande, Catalytic Hydrogenation of Benzoic Acid, New Advances in
Hydrogenation Processes - Fundamentals and Applications2017, pp. 103-109.
[18] K. Tahara, E. Nagahara, Y. Itoi, S. Nishiyama, S. Tsuruya, M. Masai, Appl. Catal., A, 154 (1997) 75-
86. https://doi.org/10.1016/s0926-860x(96)00361-4
[19] K.Y. Cheah, T.S. Tang, F. Mizukami, S.-i. Niwa, M. Toba, Y.M. Choo, J. Am. Oil Chem. Soc., 69
(1992) 410-416. https://doi.org/10.1007/bf02540940
[20] C.A. Fonseca Benítez, V.A. Mazzieri, M.A. Sánchez, V.M. Benitez, C.L. Pieck, Appl. Catal., A, 584
(2019). https://doi.org/10.1016/j.apcata.2019.117149
[21] K. Liberková, R. Touroude, J. Mol. Catal. A: Chem., 180 (2002) 221-230.
https://doi.org/10.1016/S1381-1169(01)00439-3
[22] F. Coloma, A. Sepúlveda-Escribano, J.L.G. Fierro, F. Rodríguez-Reinoso, Appl. Catal., A, 136
(1996) 231-248. https://doi.org/10.1016/0926-860X(95)00259-6
[23] T. Ohno, T. Tsubota, Y. Nakamura, K. Sayama, Appl. Catal., A, 288 (2005) 74-79.
https://doi.org/10.1016/j.apcata.2005.04.035
[24] K. Balakrishnan, J. Schwank, J. Catal., 127 (1991) 287-306. https://doi.org/10.1016/0021-
9517(91)90227-U
[25] F. Meunier, A. Goguet, S. Shekhtman, D. Rooney, H. Daly, Appl. Catal., A, 340 (2008) 196-202.
https://doi.org/10.1016/j.apcata.2008.02.034
[26]
https://doi.org/10.1103/PhysRevB.54.11169
[27] G. Kresse, J. Hafner, Phys. Rev. B, 47 (1993) 558-561. https://doi.org/10.1103/PhysRevB.47.558
[28] G. Kresse, J. Hafner, Phys. Rev. B, 49 (1994) 14251-14269.
https://doi.org/10.1103/PhysRevB.49.14251
G.
Kresse,
J.
Furthmuller,
Phys.
Rev.
B,
54
(1996)
11169-11186.
[29] G. Kresse, J. Furthmuller, Comp. Mater. Sci., 6 (1996) 15-50. https://doi.org/10.1016/0927-
0256(96)00008-0
[30] J.P. Perdew, K. Burke, M. Ernzerhof, 77 (1996) 3865-3868. 10.1103/PhysRevLett.77.3865
[31]
G.
Kresse,
D.
Joubert,
Phys.
Rev.
B,
59
(1999)
1758-1775.
https://doi.org/10.1103/PhysRevB.59.1758
[32] P.E. Blöchl, Phys. Rev. B, 50 (1994) 17953-17979. https://doi.org/10.1103/PhysRevB.50.17953
[33] M. Dion, H. Rydberg, E. Schroder, D.C. Langreth, B.I. Lundqvist, Phys. Rev. Lett., 92 (2004)
246401. https://doi.org/10.1103/PhysRevLett.92.246401
27

[34] T. Thonhauser, V.R. Cooper, S. Li, A. Puzder, P. Hyldgaard, D.C. Langreth, Phys. Rev. B, 76 (2007).
https://doi.org/10.1103/PhysRevB.76.125112
[35]
J.
Klimeš,
D.R.
Bowler,
A.
Michaelides,
Phys.
Rev.
B,
83
(2011).
https://doi.org/10.1103/PhysRevB.83.195131
[36] J. Klimes, D.R. Bowler, A. Michaelides, J. Phys. Condens. Matter, 22 (2010) 022201.
https://doi.org/10.1088/0953-8984/22/2/022201
[37]
https://doi.org/10.1103/PhysRevLett.103.096102
[38] J. Bjork, S. Stafstrom, Chem.
https://doi.org/10.1002/cphc.201402063
G.
Roman-Perez,
J.M.
Soler,
Phys.
Rev.
Lett.,
103
(2009)
096102.
Phys.
Chem.,
15
(2014)
2851-2858.
[39] F. Mittendorfer, A. Garhofer, J. Redinger, J. Klimeš, J. Harl, G. Kresse, Phys. Rev. B, 84 (2011).
https://doi.org/10.1103/PhysRevB.84.201401
[40] A. Michaelides, Z.P. Liu, C.J. Zhang, A. Alavi, D.A. King, P. Hu, J. Am. Chem. Soc., 125 (2003)
3704-3705. https://doi.org/10.1021/ja027366r
[41] Z.P. Liu, P. Hu, J. Am. Chem. Soc., 125 (2003) 1958-1967. https://doi.org/10.1021/ja0207551
[42] A. Alavi, P.J. Hu, T. Deutsch, P.L. Silvestrelli, J. Hutter, Phys. Rev. Lett., 80 (1998) 3650-3653.
https://doi.org/10.1103/PhysRevLett.80.3650
[43] Z. Wang, H.F. Wang, P. Hu, Chem. Sci., 6 (2015) 5703-5711. https://doi.org/10.1039/c5sc01732g
[44] R. Burch, C. Paun, X.M. Cao, P. Crawford, P. Goodrich, C. Hardacre, P. Hu, L. McLaughlin, J. Sá,
J.M. Thompson, J. Catal., 283 (2011) 89-97. https://doi.org/10.1016/j.jcat.2011.07.007
[45] X. Luo, J. Yang, H. Liu, X. Wu, Y. Wang, Y. Ma, S.H. Wei, X. Gong, H. Xiang, J. Am. Chem. Soc.,
133 (2011) 16285-16290. https://doi.org/10.1021/ja2072753
[46] S. Lu, Y. Wang, H. Liu, M.S. Miao, Y. Ma, Nat. Commun.,
https://doi.org/10.1038/ncomms4666
5 (2014) 3666.
[47] Y. Wang, M. Miao, J. Lv, L. Zhu, K. Yin, H. Liu, Y. Ma, J. Chem. Phys., 137 (2012) 224108.
https://doi.org/10.1063/1.4769731
[48] T. Okanishi, T. Matsui, T. Takeguchi, R. Kikuchi, K. Eguchi, Appl. Catal., A, 298 (2006) 181-187.
https://doi.org/10.1016/j.apcata.2005.09.035
[49] P.A. Sermon, V.A. Self, E.P.S. Barrett, J. Mol. Catal., 65 (1991) 377-384. 10.1016/0304-
5102(91)85072-A
[50] J.Z. Shyu, K. Otto, Appl. Surf. Sci., 32 (1988) 246-252. https://doi.org/10.1016/0169-4332(88)90085-
2
[51] L.K. Ono, B. Yuan, H. Heinrich, B.R. Cuenya, J. Phys. Chem. C, 114 (2010) 22119-22133.
https://doi.org/10.1021/jp1086703
[52] J. Szuber, G. Czempik, R. Larciprete, D. Koziej, B. Adamowicz, Thin Solid Films, 391 (2001) 198-
203. https://doi.org/10.1016/s0040-6090(01)00982-8
[53] C. Körber, S.P. Harvey, T.O. Mason, A. Klein, Surf. Sci., 602 (2008) 3246-3252.
https://doi.org/10.1016/j.susc.2008.08.015
[54] M.M. Schubert, M.J. Kahlich, G. Feldmeyer, M. Hüttner, S. Hackenberg, H.A. Gasteiger, R.J. Behm,
Phys. Chem. Chem. Phys., 3 (2001) 1123-1131. https://doi.org/10.1039/B008062O
[55]
D.I.
Jerdev,
A.
Olivas,
B.E.
Koel,
J.
Catal.,
205
(2002)
278-288.
https://doi.org/10.1006/jcat.2001.3452
[56] A.B. Merlo, V. Vetere, J.F. Ruggera, M.L. Casella, Catal. Commun., 10 (2009) 1665-1669.
https://doi.org/10.1016/j.catcom.2009.05.005
[57] J.M. Ramallo-López, G.F. Santori, L. Giovanetti, M.L. Casella, O.A. Ferretti, F.G. Requejo, J. Phys.
Chem. B, 107 (2003) 11441-11451. https://doi.org/10.1021/jp030579y
[58] A. Moscu, Y. Schuurman, L. Veyre, C. Thieuleux, F. Meunier, Chem. Commun., 50 (2014) 8590-
8592. https://doi.org/10.1039/C4CC03208J
[59] O.S. Alexeev, S.Y. Chin, M.H. Engelhard, L. Ortiz-Soto, M.D. Amiridis, J. Phys. Chem. B, 109
(2005) 23430-23443. https://doi.org/10.1021/jp054888v
[60] J. Ruiz-Martínez, F. Coloma, A. Sepúlveda-Escribano, J.A. Anderson, F. Rodríguez-Reinoso, Catal.
Today, 133-135 (2008) 35-41. https://doi.org/10.1016/j.cattod.2007.11.053
[61] T. Matsui, T. Okanishi, K. Fujiwara, K. Tsutsui, R. Kikuchi, T. Takeguchi, K. Eguchi, Sci. Technol.
Adv. Mat., 7 (2016) 524-530. https://doi.org/10.1016/j.stam.2006.02.020
[62] D.A. Stern, L. Laguren-Davidson, D.G. Frank, J.Y. Gui, C.H. Lin, F. Lu, G.N. Salaita, N. Walton,
D.C. Zapien, A.T. Hubbard, J. Am. Chem. Soc., 111 (1989) 877-891. https://doi.org/10.1021/ja00185a015
[63] J. Kondo, N. Ding, K.-i. Maruya, K. Domen, T. Yokoyama, N. Fujita, T. Maki, Bull. Chem. Soc.
Jpn., 66 (1993) 3085-3090. https://doi.org/10.1246/bcsj.66.3085
[64] D.M. Meier, A. Urakawa, A. Baiker, Phys. Chem. Chem. Phys., 11 (2009) 10132-10139.
https://doi.org/10.1039/B910535B
[65] A.M. Robinson, L. Mark, M.J. Rasmussen, J.E. Hensley, J.W. Medlin, J. Phys. Chem. C, 120 (2016)
26824-26833. https://doi.org/10.1021/acs.jpcc.6b08415
[66] C. Keresszegi, D. Ferri, T. Mallat, A. Baiker, J. Catal., 234 (2005) 64-75.
https://doi.org/10.1016/j.jcat.2005.05.019
[67] D. Shi, J.M. Vohs, Surf. Sci., 650 (2016) 161-166. https://doi.org/10.1016/j.susc.2015.10.015
[68] D. Ferri, C. Mondelli, F. Krumeich, A. Baiker, J. Phys. Chem. B, 110 (2006) 22982-22986.
https://doi.org/10.1021/jp065779z
28

29