On the role of water in selective hydrogenation of cinnamaldehyde to cinnamyl alcohol on PtFe catalysts

Article abstract of DOI:10.1016/j.jcat.2018.05.008

A series of carbon nanotube (CNT)-supported bimetallic PtFe nanoparticles were synthesized and employed as catalysts for hydrogenation of cinnamaldehyde in pure water. A synergy between water and bimetallic PtFe catalysts has allowed the efficiently selective production of cinnamyl alcohol. With the aid of water, an initial reaction rate of >1200 h^?1 and a high selectivity of >97%, as well as a good cycling stability, were achieved with a Pt₃Fe/CNT catalyst under mild reaction conditions. Isotopic labeling studies and theoretic calculation results demonstrated that the water-involved hydrogen-exchange pathway occurred with a lower energy barrier, which coexisted with the pathway of direct H₂ dissociation–hydrogenation. This work also suggested that water participated in the catalytic hydrogenation reaction by serving as a hydrogen-exchange bridge.

Full text of DOI:10.1016/j.jcat.2018.05.008

Journal of Catalysis 364 (2018) 192–203
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
On the role of water in selective hydrogenation of cinnamaldehyde to
cinnamyl alcohol on PtFe catalysts
a
a
a
a
a
a
Yihu Dai a, , Xing Gao , Xiaofang Chu , Chunyang Jiang , Yao Yao , Zhen Guo , Chunmei Zhou ,
⇑
Chuan Wang a, Hongming Wang b, , Yanhui Yang a,
⇑
⇑
a Institute of Advanced Synthesis (IAS), School of Chemistry and Molecular Engineering, Jiangsu National Synergetic Innovation Center for Advanced Materials, Nanjing Tech
University, 211816 Nanjing, China
b Institute of Advanced Study and College of Chemistry, Nanchang University, 330031 Nanchang, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of carbon nanotube (CNT)-supported bimetallic PtFe nanoparticles were synthesized and
employed as catalysts for hydrogenation of cinnamaldehyde in pure water. A synergy between water
and bimetallic PtFe catalysts has allowed the efficiently selective production of cinnamyl alcohol. With
the aid of water, an initial reaction rate of >1200 hꢀ1 and a high selectivity of >97%, as well as a good
cycling stability, were achieved with a Pt3Fe/CNT catalyst under mild reaction conditions. Isotopic label-
ing studies and theoretic calculation results demonstrated that the water-involved hydrogen-exchange
pathway occurred with a lower energy barrier, which coexisted with the pathway of direct H2
dissociation–hydrogenation. This work also suggested that water participated in the catalytic hydrogena-
tion reaction by serving as a hydrogen-exchange bridge.
Received 24 March 2018
Revised 6 May 2018
Accepted 7 May 2018
Keywords:
Water
Selective hydrogenation
a,b-Unsaturated aldehyde and alcohol
Ó 2018 Elsevier Inc. All rights reserved.
Bimetallic catalyst
Hydrogen exchange
1. Introduction
orientation of ligands with different chain lengths on the metal
surfaces can be used to tune the adsorption geometry of the alde-
The chemoselective hydrogenation of carbonyls in multiunsatu-
rated hydrocarbons, particularly ,b-unsaturated aldehydes and
hyde molecule and boost the hydrogenation selectivity [14]. The
confinement of metallic Pt or Pd sites in porous metal–organic
frameworks also promotes the selectivity of the C@O activation
due to steric hindrance of the C@C group as it approaches the
active sites [15–17]. In addition, the catalytic performance depends
on the manipulation of kinetic parameters (e.g., temperature, pres-
sure, and humidity).
The use of water as a promoter or inhibitor to enhance the
activity and selectivity of a catalyst has been demonstrated in
many types of reactions, including CO oxidation, methanation, Fis-
cher–Tropsch synthesis, oxygen evolution reactions, and various
processes used in organic synthesis [18–23]. The extensive interest
in the utilization of water is due to its substantial economic and
ecological advantages, as well as its various functions, including
solvation effects, carbon removal, and surface reconstruction
a
ketones, is a longstanding challenge, since both thermodynamics
and kinetics favor the hydrogenation of C@C bonds over C@O
bonds [1]. A variety of heterogeneous metallic catalysts have been
developed to control the selectively of the reaction by means of
alloying, ligand modification, and support optimization [2–10].
The elucidation of geometric and electronic effects on Pt-
catalyzed hydrogenation reactions has been attempted. The selec-
tivity of C@O hydrogenation exhibits a strong dependence on the
particle size distribution, and the reaction rate is also sensitive to
the structure of the Pt nanoparticles [11–13]. The electronic mod-
ification of Pt in bimetallic catalysis is an effective strategy for
improving catalytic performance. However, the lack of structural
control limits further enhancement because the second metal
promoter unselectively blocks the Pt active sites. Surface science
studies have suggested that direct adsorption of substances on
the self-assembled layers of ligands is critical to adjusting the
selectivity of the hydrogenation. The arrangement and the
[24–27].
A reaction pathway followed by a water-mediated
Mars–Van Krevelen reaction has been reported in single-atom
Pt1/CeO2-catalyzed CO oxidation. The intermediate is formed via
the interaction of CO and the hydroxyl group from the dissociation
of water rather than the direct reaction of CO with the lattice oxy-
gen [28] The Au-catalyzed oxidation of CO has also been enhanced
by water-promoted decomposition and conversion of the reaction
intermediates [29]. Recently, water-assisted hydrogen diffusion
⇑
Corresponding author.
E-mail addresses: ias_yhdai@njtech.edu.cn (Y. Dai), hongmingwang@ncu.edu.cn
(H. Wang), yhyang@njtech.edu.cn (Y. Yang).
https://doi.org/10.1016/j.jcat.2018.05.008
0021-9517/Ó 2018 Elsevier Inc. All rights reserved.

Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
193
across the surface of a metal oxide was directly confirmed by scan-
ning tunneling microscopy (STM) movies, suggesting that catalytic
hydrogenation, hydrogen evolution, or the reformation of hydro-
gen may be accelerated by the presence of water [30]. For instance,
it was found that water could take part in the synthesis of C2+ alco-
hols via CO2 hydrogenation by serving as the hydrogen source [31].
Despite the great progress and valuable insights from the above
investigations, the details of the role of water in heterogeneous
catalytic hydrogenation processes remain elusive.
For conjugated C@O hydrogenation, the replacement of organic
solvent with water has been proven available to improve both the
activity and selectivity of Au and Pt catalysts [32,33]. Trace
amounts of water have also been shown to enhance the activity
on Pt-based catalysts [34]. The orientable chemisorption of the
hydrophilic C@O moiety in dipolar water solvent leads to satisfac-
tory selectivity to unsaturated alcohol, according to the kinetic
characterizations and adsorption spectra data [35–37]. Further-
more, isotopic-exchange studies have clarified the evolution of cat-
alytically active sites in the pathway of direct hydrogenation by the
dissociated H2 [38,39]. The essential effect of water merits further
investigation for selective hydrogenation on heterogeneous
catalysts.
oxygen-containing groups to enhance the metal dispersion. The
treated CNTs were washed with deionized water several times
and dried in a 100 °C oven overnight. The CNT supports were kept
dry before use.
2.2.2. Preparation of supported Pt-based catalysts
In a typical synthesis of the supported Pt catalyst, Pt(acac)2
(100 mg, 0.25 mmol), 40 mL of benzyl alcohol and 800 mg of dry
support were mixed together at room temperature and then stirred
for 10 min. The resulting homogeneous yellow suspension was
transferred to a glass pressure vessel. The sealed vessel was
charged with 1 bar of CO gas and then heated for 3 h with vigorous
stirring in an oil bath that had been preheated to 180 °C. After the
vessel was cooled to room temperature, the catalyst was collected
by centrifugation and washed several times with ethanol. For sup-
ported PtFe catalysts, the same preparation procedure was used
and certain amounts of Pt(acac)2 and Fe(acac)3 were employed as
metal precursors to obtain the desired Pt/Fe mole ratio. The Pt
loadings were fixed at approximately 6.2 wt% for all Pt-based cat-
alysts and confirmed by ICP-OES measurements. The identified real
Pt/Fe ratios were 3.3, 8.5, and 58.7 for the Pt3Fe/CNT, Pt9Fe/CNT,
and Pt60Fe/CNT catalyst, respectively.
In this work, we employ carbon nanotube (CNT)-supported
bimetallic PtFe nanoparticles (NPs) to efficiently achieve the
chemoselective hydrogenation of cinnamaldehyde (CALD) to cin-
namyl alcohol (CALA) with water as the only solvent. We illustrate
the role of water as both a hydrogen carrier and a polar solvent by
isotopic labeling experiments and theoretical calculations. The
coexistence of two reaction pathways for water-mediated hydrogen
exchange and dissociative hydrogenation with H2 is also confirmed.
2.2.3. Acid washing of the Pt3Fe/CNT catalyst
The reference catalyst was prepared via acid washing. The acid-
washed sample without surface Fe species (Pt3Fe/CNT-H2SO4) was
obtained as follows: Pt3Fe/CNT (100 mg) was soaked in 20 mL of
H2SO4 (0.5 M) for 30 min, and the slurry was filtered and repeat-
edly washed with deionized water. Based on ICP-OES analysis,
approximately 10% of the Fe has been removed, and the estimated
Pt/Fe ratio of the Pt3Fe/CNT-H2SO4 sample was 3.7.
2. Experimental
2.3. Characterizations
2.1. Materials
The metal loadings for all catalysts were quantified using induc-
tively coupled plasma optical emission spectrometry (ICP-OES,
Avio 200, Perkin Elmer) by measuring the metal concentrations
in solutions of the complexes dissolved in 40% hydrofluoric acid.
X-ray diffraction (XRD) patterns were recorded on a Bruker AXS
Pt(acac)2 (97.0%), Fe(acac)3 (97.0%), benzyl alcohol (99.0%), cin-
namaldehyde (CALD, 99.0%), phenylpropyl aldehyde (HCALD,
95.0%), cinnamyl alcohol (CALA, 98.0%), 3-phenylpropanol (HCALA,
99.0%), isopropanol (99.9%), n-heptane (98.0%), ethyl acetate
(99.8%), cyclohexane (99.5%), and n-hexane (97.0%) were obtained
from Sigma-Aldrich (Shanghai, China). HNO3 (69%) and H2SO4
(98%) were provided by Aladdin Reagent Corp. (Shanghai, China).
The deuterated solvent D2O (deuteration degree > 99.8%) was pro-
vided by J&K Scientific (Shanghai, China). CO (99.99%), H2
(99.999%), and Ar (99.999%) were supplied by Nanjing Special
Gas Factory (Nanjing, China). Multiwalled carbon nanotubes (pur-
ity 97.1%, SBET 241 m2/g, bulk density 0.05 g/cm3) were purchased
from CNano Technology (Zhenjiang, China).
D8 Focus diffractometer using Ni-filtered CuKa radiation (k =
0.154 nm) at 40 kV and 40 mA. Diffraction data were collected
between 10° and 70° with a resolution of 0.02° (2h). Transmission
electron microscopy (TEM) images were obtained on a JEOL JEM
3010 instrument operated at 300 kV. Scanning transmission elec-
tron microscopy (STEM) and elemental line scanning were carried
out on an FEI Tecnai G2 F20 S-TWIN system operated at 200 kV. X-
ray photoelectron spectroscopy (XPS) measurements were
acquired using a VG Scientific ESCALAB Mark II spectrometer
equipped with two ultra-high-vacuum (UHV) chambers. All bind-
ing energies were referenced to the C1s peak at 284.6 eV.
2.2. Catalyst preparation
2.4. Catalytic tests
CNTs were chosen as the support due to their anchoring effect;
the CNT surface with abundant oxygen-containing groups can sta-
bilize the metallic sites via strong interactions. Supported Pt-based
nanoparticle (NP) catalysts were prepared according to a modified
procedure using benzyl alcohol and carbon monoxide as the co-
reductant. The method affords a moderate size distribution of
bimetallic NPs that have clean surfaces and do not contain residual
organic ligands [40]. Consequently, the effects of the ligand can be
ruled out, and the intrinsic activity of the Pt-based catalysts in
water is truly reflected in the selective hydrogenation reaction.
Selective hydrogenations of CALD were carried out in a stainless
steel reactor with a Teflon liner (Parr 4950, controller 4843). A 10-
mg portion of catalyst, a certain amount of CALD (200, 400, or
1000 mL), and 10 mL of the solvent were added into the reactor.
The molar ratio of CALD to total Pt atoms in the catalysts was
approximately 500, 1000, or 2500, based on the CALD sampling
amount. The residual air inside the reactor was expelled by pres-
surizing and releasing hydrogen several times. The reaction was
performed at 60 °C under 20 bar of hydrogen. The stirring speed
was set at 700 rpm for all the experiments in this study to elimi-
nate the external mass transfer limitation. The equivalents of
CALD, hydrogen pressure, and reaction temperature, as well as
reaction time, were controlled in the tests of catalytic performance.
2.2.1. Pretreatment of CNT support
The pristine CNTs were pretreated with concentrated HNO3 at
120 °C for 4 h to remove impurities and to introduce surface

194
Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
After the reactions were carried out in organic solvents, the cata-
lyst powder was removed by centrifugation. For the reactions in
water, in consideration of insoluble substances in water phase,
ethyl acetate was added into the liner to thoroughly extract
organic products and the solid catalyst was removed by centrifuga-
tion. All the products in the solution were analyzed using a gas
chromatograph (GC, Agilent, 6890N) equipped with a CP-WAX 52
selected Fe6O18H15 as a model for Fe(OH)x. Our calculations
showed that Fe6O18H15 can strongly anchor to the stepped surfaces
of the Pt(1 1 1) planes. First-principle calculations were carried out
using the Vienna ab initio simulation package (VASP) based on the
all-electron projected augmented wave (PAW) method [41,42]. The
k-point sampling was generated following the Monkhorst–Pack
procedure with a 2 ꢁ 3 ꢁ 1 mesh. The Perdew–Wang 91 form of
the generalized gradient approximation was used to calculate non-
local gradient corrections to the correlation and exchange energies
[43,44]. The GGA + U approximation with a Ueff of 3.0 eV for Fe
atoms (antiferromagnetic) was used [44,45]. During structural
optimization, the bottom layers of the slab were fixed at the bulk
truncated position, while the top two layers and the adsorbates
were fully relaxed. To prevent artificial interaction between the
repeated slabs along the z-direction, a 15 Å vacuum was intro-
duced with correction of the dipole moment. The reaction barriers
were calculated using the climbing image nudged elastic band
(CINEB) method [46,47]. The final transition state (TS) structures
were refined using a quasi-Newton algorithm until the forces were
less than 0.03 eV/Å.
CB column (30 m ꢁ 0.25 mm ꢁ 0.25
lm) and a flame ionization
detector (FID). Tetradecane was added as an internal standard.
The byproducts were identified by gas chromatography–mass
spectrometry (GCMS, Agilent, GC 6890N, MS 5973). The carbon
balance was >98% in the hydrogenation reactions. We computed
Conversion ¼ f½NinðCALDÞ ꢀ NoutðCALDÞꢂ=NinðCALDÞg ꢁ 100%;
X
Selectivityðp ꢀ iÞ ¼ Noutðp ꢀ iÞ= iNoutðp ꢀ iÞ ꢁ 100%;
X
Carbon balance ¼ ½NoutðCALDÞþ
Specific reaction rate ¼ ½V ꢁ
where p-i represents hydrogenation products; N(p-i) is the molar
iNoutðpꢀiÞꢂ=NinðCALDÞꢁ100%;
q
ꢁ Conversion=MWꢂ=½Time ꢁ NtotalðPtÞꢂ;
3. Results and discussion
amount of the corresponding product i; V, , and MW are the sam-
q
pling volume amount, density, and molecular weight for CALD,
respectively; and Ntotal(Pt) is the molar number of total Pt atoms
in employed catalyst.
The experimental CALD concentrations versus time are plotted
according to a rate law that is first-order with respect to CALD,
and the rate constant k is calculated from a regression fit of the
experimental data:
3.1. Catalyst structure
The XRD patterns (Fig. 1A) confirmed that the metallic NPs were
well dispersed on the surfaces of CNT support. For all the catalysts,
the (0 0 2) main diffraction at 25.8° suggested that the hexagonal
graphite structure had been retained in the CNT support after the
loading of the metals and pretreatment procedure of acid washing
[48]. The diffraction peaks corresponding to the Pt(1 1 1) and (2 0
0) planes at 39.6° and 46.1°, respectively, were observed for the Pt/
CNT catalyst. For the bimetallic PtFe catalysts, these peaks gradu-
ally shifted to higher angles with increasing Fe loadings. Specifi-
cally, the peaks were located at 40.1° and 46.8° for the Pt3Fe/CNT
catalyst. This indicated the formation of the PtFe alloy phase and
was consistent with the unit cell shrinkage due to the smaller lat-
tice parameter of Fe [49]. In addition, the alloy structure of the PtFe
NPs in the Pt3Fe/CNT–H2SO4 catalyst was maintained during acid
washing with dilute H2SO4.
The XPS studies confirmed that the Pt in the Pt-based catalysts
was mainly reduced to metallic Pt0 (Fig. 1B). The binding energies
of 71.4 and 74.8 eV for the Pt4f7/2 and 4f5/2 orbitals, respectively,
were observed on the CNT-supported Pt and PtFe catalysts. Neither
modifying Fe nor removing the oxygen-containing groups on the
surfaces of the CNTs had a notable impact on the Pt binding ener-
gies. The signals from the Fe2p3/2 orbitals were relatively weak;
however, the coexistence of Fe2+ and Fe3+ species (55% vs. 45%)
was observed for the Pt3Fe/CNT catalyst (Fig. 1C) [50].
lnðCt=C0Þ ¼ ꢀk ꢁ t;
where Ct represents the CALD concentration at reaction time t and
C0 is the initial CALD concentration.
Isotope studies were performed under similar reaction condi-
tions except that deuterated solvent (D2O) was used. For control
experiments with/without the Pt3Fe/CNT catalyst, the substances
for the standard hydrogenation reaction, including CALD and three
kinds of main products (CALA, HCALD, and HCALA), were mixed
with D2O and stirred in a Parr reactor at 25 °C under 20 bar of N2
for 24 h. The deuterated products were analyzed by GCMS and
1H NMR (Bruker Advance 400 M).
2.5. Theoretical calculations
The structural model for the FePt surface in the present calcula-
tions was built based on previously reported characterization
results for the Pt3Fe NP catalysts [21]. Briefly, we used three-
layered Pt(1 1 1) surfaces to model the Pt substrate, and we
Fig. 1. (A) XRD patterns and (B, C) XPS spectra of Pt4f and Fe2p orbitals of various Pt and PtFe catalysts.

Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
195
Furthermore, the disappearance of the Fe-related peaks in the XPS
spectrum of the Pt3Fe/CNT–H2SO4 catalyst confirmed the possibil-
ity of targeted removal of surface Fe species by acid washing pre-
treatment [51].
Pt60Fe, Pt9Fe, and Pt3Fe particles, as illustrated in Fig. 2I–K. Accord-
ingly, the ratio of the signal strength of Fe to Pt lines increased with
the increasing Fe/Pt ratio from 1/60 to 1/3. The HRTEM images
directly showed a lattice spacing of 0.23 nm, which corresponds
to the (1 1 1) faces of Pt and PtFe alloys (Fig. 3A and B). Based on
these characterization data, the CNT-supported PtFe NPs were
composed of a PtFe alloy core and a shell comprising Fe oxide
species.
The TEM images of the Pt/CNT catalyst revealed that the CNT-
supported monometallic Pt NPs are 2.7 0.6 nm in particle size
(Fig. 2A and B). With the modification of the second Fe metal, the
sizes of PtFe NPs on CNT surfaces increased slightly in comparison
with that of monometallic Pt NPs. As shown in Fig. 2C–H, the size
distributions of PtFe NPs for Pt60Fe, Pt9Fe, and Pt3Fe catalysts are
5.4 1.3, 5.4 0.9, and 6.7 1.7 nm, respectively. STEM-EDX ele-
mental line-scanning analysis was performed on various PtFe cat-
alysts to further identify the formation of PtFe alloy structure. The
compositional line profiles of Pt and Fe clearly cross individual
3.2. Effects of the modification with Fe species
The selective hydrogenation of cinnamaldehyde (CALD) to cin-
namyl alcohol (CALA) was performed on CNT-supported Pt and
PtFe catalysts under relatively mild conditions (Fig. 4). The
Fig. 2. TEM images and particle size distributions of CNT-supported Pt and PtFe catalysts (A–H); (I–K) STEM-EDX elemental line scanning profiles of PtFe NPs in PtFe/CNT
catalysts.

196
Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
catalytic performances in isopropanol, a common solvent for this
reaction, are presented in Fig. 4A. To better judge the selectivity
of the catalytic hydrogenation, a consistent level of CALD conver-
sion (approximately 60%) over various catalysts was achieved by
regulating the reaction times. The monometallic Pt/CNT showed
an unsatisfactory CALA selectivity of 50.5%. The modification of
Pt with Fe as a second metal dramatically enhanced the chemose-
lectivity for the C@O hydrogenation, and the addition of a sufficient
amount of Fe was necessary for the enhancement. Specifically, the
selectivity for CALA reached 95.5% at 60.9% conversion over the
Pt3Fe/CNT catalyst. The relationship between the specific reaction
rate and the Pt/Fe ratio demonstrated that the alloyed PtFe could
also boost the rate of the hydrogenation reaction as compared with
the monometallic Pt/CNT catalyst. Furthermore, the variations in
the performances with the Pt/Fe ratio indicated that the catalytic
process was very sensitive to the structure of the PtFe NPs and
the Pt–Fe interactions on the surface. On one hand, the electropos-
itive Fe species could increase the electron density on the Pt to
lower the probability of C@C activation and polarize the C@O
group for nucleophilic attack by the H species dissociated on the
Pt site [52]. On the other hand, the oxidized Fe species may act
as Lewis acidic sites and interact with the C@O group in CALD, pro-
moting the selective conversion of CALD to CALA [34,53]. For
instance, our previous work demonstrated that iron oxide
nanoisland-decorated Pt NPs displayed improved catalytic perfor-
mances in the selective hydrogenation of CALD [54]. The specific
Pt–FeOx interfaces could also be generated by using atomic layer
deposition, and the high selectivity for CALA was believed to be
due to the precise blocking of low-coordinated Pt sites with Fe oxi-
des [55].
It was reported that isopropanol could serve as a hydrogen
donor in the transfer catalytic hydrogenation reactions [56,57].
A H2-free blank hydrogenation reaction test in isopropanol was
performed over a Pt3Fe/CNT catalyst. CALD conversion less than
1% and CALA selectivity 48.9% were obtained under N2. Such
low selectivity suggested that the hydrogenation of CALD on Pt3-
Fe/CNT catalyst may not follow the transfer hydrogenation path-
way. The huge conversion gap also indicated that the direct
Fig. 3. HRTEM images of Pt/CNT and Pt3Fe/CNT catalysts.
Fig. 4. Reaction network of CALD hydrogenation and the catalytic performances of the CNT-supported Pt-based catalysts in (A) isopropanol (i-PrOH) and (B) water. Reaction
conditions: 10 mg of catalyst, 400 mL of CALD, 10 mL of solvent, 20 bar of H2, 60 °C, 700 rpm, and reaction time (a) 1.5 h; (b) 6 h; (c) 10 h; (d) 0.5 h; (e) 1 h; (f) 4 h; and (g) 8 h.
The CALD conversions remained approximately 60% for all hydrogenation reactions.

Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
197
hydrogenation overwhelmingly dominates the reaction process
3.3. Activity in water
and isopropanol plays a solo role of solvent rather than hydrogen
donor.
The optimized Pt3Fe/CNT catalyst was employed to explore the
catalytic performance in CALD hydrogenation with pure water as
the solvent. The activities and selectivities of Pt3Fe/CNT catalyst
in water and other types of organic solvents were first compared.
Those solvents (e.g., cyclohexane and ethyl acetate) are frequently
used in hydrogenation reactions of a,b-unsaturated aldehydes and
ketones (Table S1 in the Supporting Information). As shown in
Fig. 5, satisfactory CALD conversion and CALA selectivity were
achieved in water as the solvent. It should be mentioned that the
reaction rate of hydrogenation of the
a,b-unsaturated aldehydes
is not correlated with the hydrogen solubility, because the CALD
conversion in isopropanol was less than half of that in water even
though the solubility of H2 in isopropanol is four times higher than
that in water [58]. With the aid of water, the CALA selectivity of the
Pt3Fe/CNT catalyst further increased to 97.2%, while the conversion
reached 62.1% in 0.5 h (Fig. 6). Such high selectivity (95.9%) was
still observed at nearly complete conversion (98.1% in 1 h). The
decline in the CALA selectivity in favor of the formation of the fully
hydrogenated hydrocinnamyl alcohol (HCALA) suggested that the
sequential hydrogenation of the C@O and C@C bonds occurred on
the Pt3Fe/CNT catalyst in water.
Fig. 5. The catalytic hydrogenation performance of Pt3Fe/CNT catalysts in various
solvents. The calculated specific reaction rate of CALD is based on the total molar
amount of Pt atoms in Pt3Fe/CNT catalyst. Reaction conditions: 10 mg of catalyst,
10 mL of solvent, 400 mL of CALD, 20 bar of H2, 60 °C, and 0.5 h. C6H12, EA, C6H14, and
C7H16 represent cyclohexane, ethyl acetate, n-hexane, and n-heptane, respectively.
The facilitation effect of water was also observed in the hydro-
genation reactions over Pt, Pt9Fe, and Pt60Fe catalysts (Fig. 4B). For
instance, the CALA selectivity in water increased to 72.8% while the
selectivity in isopropanol was 50.5% at the same level of conversion
for the monometallic Pt/CNT catalyst. Slightly increased selectivity
of CALA was also obtained in water on Pt9Fe/CNT and Pt60Fe/CNT
catalysts, compared with that in isopropanol. In general, the polar
water can substantially promote the adsorption and activation of
hydrophilic C@O groups over C@C groups, thus producing CALA
with a similar polarity [59]. However, we expect that this may
not be the only role of water in this reaction process.
3.4. Water-involved hydrogen exchange
Fig. 6. The catalytic performance (CALD conversion and product selectivity) of
Pt3Fe/CNT as a function of reaction time. Reaction conditions: 10 mg of catalyst, 10
mL of H2O, 400 mL of CALD, 20 bar of H2 and 60 °C.
To further study the effects of water, pure deuterium oxide
(D2O) was used instead of H2O to determine its effect on the
Fig. 7. The mass spectra of CALD and hydrogenation products in D2O-labeling reactions with/without Pt3Fe/CNT catalyst at different temperatures and H2 pressures. (A–D)
reference test without catalyst, (E–H) reference test with catalyst, (I–L) Pt3Fe/CNT catalyzed reaction at 25 °C and 1 bar of H2, (M–P) reaction at 25 °C and 20 bar of H2, (Q–T)
reaction at 60 °C and 20 bar of H2.

198
Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
hydrogenation reaction catalyzed by Pt3Fe/CNT. The exchange
between D2O and the related substances in the CALD hydrogena-
tion reaction should be one of the first concerns. The reference test
for the exchange process was carried out in the absence of the cat-
alyst and H2. The H–D exchanges between the D atom in D2O and
the H atom in CALD and three kinds of main hydrogenation prod-
ucts (CALA, HCALD, and HCALA) were not favored without the cat-
alyst, because only the regular molecular ion peak was observed on
Table 1
The performance of Pt3Fe/CNT in H2O/D2O in the reaction of CALD hydrogenation.a
Catalyst
Solvent
Temp. (°C)
P(H2) (bar)
20
CALD (mL)
Time (h)
Conv. (%)
Selec. CALA (%)
Selec. HCALD (%)
Selec. HCALA (%)
Pt3Fe/CNT
H2O
D2O
D2O
D2O
60
400
0.5
1
1
62.1
63.7
63.8
68.3
97.2
94.2
95.2
64.7
2.7
3.6
2.6
15.7
0.1
2.2
2.2
19.6
25
25
20
1
200
200
24
a
10 mg catalyst, 10 mL solvent.
Scheme 1. The diagram of product distributions in the reference tests of substance–D2O exchange with/without Pt3Fe/CNT catalyst and the catalytic hydrogenation reaction
in D2O.

Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
199
GCMS spectra for the individual substances (Fig. 7A–D). The H2-
free exchange test was also performed in the presence of the Pt3-
Fe/CNT catalyst. No deuterated CALD and HCALD were detected
by GCMS spectra, indicating that the alkenyl group and the alde-
hyde groups could not interact with molecular D2O on the surface
of the PtFe catalyst (Fig. 7F and H). The peak of m/z 135 for CALA
and the peak of m/z 137 for HCALA were observed on GCMS spec-
tra, which suggested that the H atom in the hydroxyl group pre-
ferred to exchange with the D atom in D2O with the action of the
PtFe catalyst (Fig. 7E and G). However, it should be noted that
although the H–D exchange process was nearly completed, only
one D atom was deuterated into the hydroxyl groups for all CALA
and HCALA molecules. This suggested that
a-C-H could not be
exchanged with D2O under the reaction condition.
The Pt3Fe/CNT catalyzed hydrogenation reaction of CALD in D2O
was then performed and the details of the isotope studies are pro-
vided in Table 1. Analysis of the GCMS spectra revealed that the
Fig. 8. Yields of gaseous D2 and HD in H2–D2O exchange reaction and deuteration
degree of CALA in hydrogenation reaction catalyzed by Pt3Fe/CNT.
Fig. 9. The energetic profiles for the production of (A) CALA and (B) HCALD in the selective hydrogenation of CALD over the PtFe catalyst model with/without the assistance of
water.

200
Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
hydrogenation products were deuterated during the catalytic reac-
tion (Fig. 7I–T). For the reaction at 25 °C and 1 bar of H2, the molec-
ular ion peaks at m/z 136 and 135 have clearly emerged in the
GCMS spectrum of the CALA product, as shown in Fig. 7I, which
suggested that both ends of the C@O bond in CALD were deuter-
ated during the hydrogenation reaction. This result confirmed that
it was a D2O-involved reaction process rather than a simple H–D
exchange process. Such a D2O-involved process also occurred in
the C@C hydrogenation reaction, which was supported by the fact
that the peaks at m/z 138 and 137 and the coexisting peaks at m/z
140, 139, and 138 were observed on the GCMS spectra of HCALD
and HCALA products, respectively (Fig. 7J and K). This result is
summarized in Scheme 1, which strongly proves that D2O was
involved as a hydrogen exchange bridge in the selective hydro-
genation of C@O and C@C bonds in CALD [60].
The core role of water was examined by density functional the-
ory (DFT) calculations for CALD hydrogenation over the PtFe sur-
face model (details in Figs. S1–S3). As shown in Fig. 9, the
assistance of water significantly decreases the energy barrier of
the overall reaction, especially the energy barrier of the rate-
determining step, for both the CALA and HCALD formation pro-
cesses. However, even in the presence of water, noticeably lower
activation energy barriers favored C@O hydrogenation over C@C
hydrogenation on the PtFe surface (14.84 vs. 18.88 kcal/mol). This
highlights the excellent synergetic effect between the water and
the bimetallic PtFe NP catalysts in enhancing both the conversion
of CALD and the selectivity for C@O hydrogenation. In contrast to
the water-free formation of CALA, the water-mediated pathway
had a much lower energy barrier (14.84 vs. 21.49 kcal/mol for
WIN1-WTS1-WIN2 and IN1-TS1-IN2, respectively, in Fig. 9A). After
In consideration that the deuteration of products using HD or D2
generated in situ from H2–D2O exchange on Pt catalyst may be one
of the possible reaction pathways [61], the H2–D2O exchange reac-
tions in the absence of CALD were also performed over Pt3Fe/CNT
catalyst. The yields of gaseous HD and D2 produced from the H2–
D2O exchange reaction and the deuteration degrees (the relative
abundances for m/z 136, 135, and 134) of the CALA product in
the D2O-involved hydrogenation reaction could be compared
under the same reaction conditions (e.g., temperatures, H2 pres-
sures, samplings, and times), as shown in Fig. 8. For H2–D2O
exchange, similarly low yields of D2 (approximately 0.3%) were
observed under different reaction conditions, and the yield of HD
at 25 °C and 1 bar of H2 was much lower than that at 60 or 25 °C
and 20 bar of H2. This suggested that the reaction rate of exchange
between H2 and D2O was mainly improved by the higher H2 pres-
sure in this temperature range. For the D2O-involved hydrogena-
tion reaction under a H2 pressure of 20 bar, the ratio of the
relative abundances of the peaks at m/z 135 and 136 to that at
m/z 134 at 25 °C was obviously higher than that at 60 °C for CALA
product (Fig. 7 M and 7Q). When the reaction temperature was 25
°C, the peak at m/z 135 became dominant and the degree of deuter-
ation was significantly increased under ambient H2 pressure, in
marked contrast to that under 20 bar (Fig. 7I and M). This sug-
gested that much more D2O actively and rapidly participated in
the hydrogenation reaction at lower H2 pressure and lower tem-
perature. The reaction rate of H2–D2O exchange and the deutera-
tion degree of CALA product in the hydrogenation reaction
revealed the opposite trend with varied H2 pressures and temper-
atures. In addition, the lower yield of HD and D2 could not guaran-
tee adequate supplies of D atoms to achieve such a higher
deuteration degree of hydrogenation products. Therefore, as illus-
trated in Scheme 1, the D2O-involved hydrogenation was neither
a simple exchange process between D2O and substances nor a tan-
dem reaction process consisting of H2–D2O exchange reaction and
deuteration reaction by HD or D2 generated in situ.
Fig. 10. The catalytic performance of the Pt3Fe/CNT catalyst in mixtures of (A) n-
heptane–water and (B) isopropanol–water. Reaction conditions: 10 mg of catalyst,
400 mL of CALD, 60 °C, 20 bar H2, 10 mL solvent, 0.5 h.
Fig. 11. The reaction rate constants (k) of Pt3Fe/CNT-catalyzed CALD hydrogenation
in H2O and D2O. C0 represents the initial molar concentration of CALD while Ct
represents the molar concentration of CALD varied with reaction time. The reaction
conditions: 10 mg of catalyst, 1.0 mL of CALD, 10 mL solvent, 20 bar H2, 60 °C. For
all reaction tests, the conversion and the CALA selectivity stayed at <25% and >92%,
respectively.
Scheme 2. The schematic diagram of the reaction pathway of water-involved
hydrogen exchange in the Pt3Fe/CNT-catalyzed CALD hydrogenation.

Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
201
the co-adsorption of CALD and dissociated H atoms on the surface
of the catalyst, the water initially attached to the Pt–H surface and
the O atom of the CALD species via two hydrogen bonds, acting as a
hydrogen bridge. The water-assisted hydrogen exchange process
was then achieved. Complete abstraction of the H atom from the
Pt surface generated a bond with the O atom in H2O, while the
O–H bond on the other side of the H2O broke and the H atom trans-
ferred into the hydrogenation product, producing H-exchanged
H2O and CALA, as shown in Scheme 2.
Water-involved hydrogen exchange is apparently affected by
the hydrogen bonding interactions between the H2O, CALD, and
Pt sites. To elucidate their mechanism of influence, we created
two mixed solvent systems, H2O–n-heptane and H2O–isopropanol,
with different combinations of hydrogen bonds. Fig. 10A shows
that the conversion rate of CALD increased proportionally with
the ratio of water to heptane over the Pt3Fe/CNT catalyst. However,
the reaction rate initially decreased and then increased as the
water/isopropanol ratio increased (Fig. 10B). This clearly indicated
that the interactions between polar solvent molecules could com-
petitively interrupt the hydrogen bonding interactions in the
water-involved hydrogen exchange and thus influence the CALD
conversion.
For Pt3Fe/CNT-catalyzed CALD hydrogenation, the reaction rate
constants (k) in H2O and D2O were analyzed in the initial reaction
stages. ln(Ct/C0) was plotted as a first-order function of reaction
time with respect to CALD, as shown in Fig. 11. kH2O and kD2O are
8.2 ꢁ 10ꢀ3 minꢀ1 and 5.8 ꢁ 10ꢀ3 minꢀ1, respectively, and the ratio
of kH2O to kD2O is only 1.4. This result suggested that the kinetic iso-
tope effect existed in the catalytic hydrogenation reaction with the
involvement of D2O; however, the cleavage of OAH bonds in H2O
and subsequent hydrogen exchange were not the rate-
determining steps. This accords with the result from DFT calcula-
tions that the H2O-bridged hydrogen-exchange process easily
occurs with a relatively low energy barrier.
3.5. Reusability of Pt3Fe/CNT in water
The presence of stable Fe3+ and Fe2+ species on the surface of the
Pt3Fe NPs was of great importance for both the catalytic activity
and the hydrogenation selectivity. When oxidized Fe species were
eliminated from the surfaces of the NPs by acid washing with
dilute H2SO4, the CALD conversion after the same reaction time fell
to half of its initial value (Fig. S4). At a similar level of CALD conver-
sion, the selectivity for CALA had decreased slightly with the Pt3Fe/
CNT–H2SO4 catalyst, unlike that with the fresh Pt3Fe/CNT catalyst.
However, the Pt3Fe/CNT–H2SO4 catalyst still showed a higher CALA
selectivity and a conversion twice as high as what was achieved
with the monometallic Pt/CNT catalyst, and this may have been
due to the stability of the PtFe alloy in the bulk NPs [53,62].
The stability and reusability tests for the Pt3Fe/CNT catalyst in
water were performed repeatedly, and the results are shown in
Fig. 12. After each reaction run, the catalyst was separated from
Fig. 12. Recycling stability of the Pt3Fe/CNT catalyst in the selective hydrogenation
of CALD in water as the solvent. Reaction conditions: 10 mg catalyst, 10 mL H2O,
200 mL CALD, 20 bar H2, 60 °C, 1 h.
Fig. 13. (A) XRD pattern, (B, C) XPS Pt4f and Fe2p spectra, (D, E) TEM and HRTEM images, (F) particle size distribution, and (G) STEM-EDX elemental line-scanning profiles of
the spent Pt3Fe/CNT catalyst after six-cycle hydrogenation reaction.

202
Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
the mixture, washed with ethanol several times, and then dried
overnight. There was a slight fluctuation in both the CALD conver-
sion and the CALA selectivity in the recycling runs. A CALD conver-
sion of 87.5% and a CALA selectivity of 90.3% were maintained even
after six runs, indicating good recyclability of the Pt3Fe/CNT cata-
lyst for the selective hydrogenation of CALD in water.
vent, the main molecular ion peak in MS spectra from the CALA
product was m/z 135 rather than m/z 134, which strongly con-
firmed the participation of D2O in the reaction system. In the
fourth run, the intensity of the peak at m/z 135 did not change, sug-
gesting that the water-involved hydrogen exchange was a stable
reaction pathway in the hydrogenation of CALD.
The structure of the recycled Pt3Fe/CNT catalyst was examined
by a series of characterization techniques, and no marked varia-
tions were observed. In addition, no leaching of Fe into the reaction
solution was detected. Both the alloy structure and the moderate
crystal size of the PtFe NPs were maintained, which was reflected
in the XRD pattern of the spent catalyst (Fig. 13A). The XPS spectra
suggested that Pt was kept in the metallic Pt0 state, as shown in
Fig. 13B. Surface Fe species were still present as both Fe2+ and
Fe3+; however, the Fe3+/Fe2+ ratio increased during the hydrogena-
tion reaction (31% Fe2+ and 69% Fe3+, Fig. 13C). The transformation
of Fe2+ to Fe3+ suggested that the electron-donating process
occurred on the Fe2+ sites in the hydrogenation reaction. The
TEM and HRTEM images revealed that the Pt3Fe NPs on the CNT
surface slightly grew to 7.3 2.2 nm from 6.7 1.7 nm after six
reaction runs (Fig. 13D–F). The STEM-EDX elemental line-
scanning profiles (Fig. 13G) confirmed that the core structure of
Pt3Fe alloy NPs was well preserved.
4. Conclusions
In summary, we demonstrated that supported PtFe NPs can effi-
ciently catalyze the hydrogenation of cinnamaldehyde to cinnamyl
alcohol with high selectivity (>97%) under relatively mild reaction
conditions with pure water as the solvent. By conducting a combi-
nation of experimental and theoretical studies, we confirmed that
the pathway of water-involved hydrogen exchange was active dur-
ing the catalytic reaction, and this pathway coexisted with direct
hydrogenation with dissociated H2. The water molecule acts as a
bridge to facilitate hydrogen exchange between the CALD species
and the Pt sites with lower thermodynamic energy barriers. We
expect that the strategy of exploiting the synergistic effects of
water and metallic catalysts could be adopted to explore novel cat-
alytic systems for the development of selective hydrogenation
processes.
To further study whether water-involved hydrogen exchange
occurs steadily in the hydrogenation of CALD, alternating reactions
were run in H2O and D2O. The reactions were performed at 25 °C
and 1 bar of H2, and the products were analyzed by GC–MS. As
illustrated in Fig. 14, when D2O was used instead of H2O as the sol-
See Figs. S1–S4 and Table S1 for DFT calculation details and
comparison of catalytic performances.
Acknowledgments
We are grateful for financial support from the Jiangsu Provincial
Department of Education and Natural Science Foundation
(17KJB150021 and BK20170986) and partial support from the
National Natural Science Foundation of China (21503187). We also
appreciate the support from Nanjing Tech University and a SICAM
Fellowship from the Jiangsu National Synergetic Innovation Center
for Advanced Materials. We thank Dr. Shihui Zou and Professor Jie
Fan for assistance with the STEM-EDX analysis.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.jcat.2018.05.008.
References
[1] P. Gallezot, D. Richard, Selective hydrogenation of
Catal. Rev. 40 (1998) 81–126.
a, b-unsaturated aldehydes,
[2] C. Vriamont, T. Haynes, E. McCague-Murphy, F. Pennetreau, O. Riant, S.
Hermans, Covalently and non-covalently immobilized clusters onto
nanocarbons as catalysts precursors for cinnamaldehyde selective
hydrogenation, J. Catal. 329 (2015) 389–400.
[3] K.-H. Dostert, C.P. O’Brien, F. Ivars-Barceló, S. Schauermann, H.-J. Freund,
Spectators control selectivity in surface chemistry: acrolein partial
hydrogenation over Pd, J. Am. Chem. Soc. 137 (2015) 13496–13502.
[4] G. Kennedy, L.R. Baker, G.A. Somorjai, Selective amplification of CO bond
hydrogenation on Pt/TiO2: catalytic reaction and sum-frequency generation
vibrational spectroscopy studies of crotonaldehyde hydrogenation, Angew.
Chem. Int. Ed. 53 (2014) 3405–3408.
[5] B. Wu, H. Huang, J. Yang, N. Zheng, G. Fu, Selective hydrogenation of
a, b-
unsaturated aldehydes catalyzed by amine-capped platinum-cobalt
nanocrystals, Angew. Chem. Int. Ed. 51 (2012) 3440–3443.
[6] Y. Zhu, H. Qian, B.A. Drake, R. Jin, Atomically Precise Au25(SR)18 nanoparticles
as catalysts for the selective hydrogenation of
a, b-unsaturated ketones and
aldehydes, Angew. Chem. Int. Ed. 49 (2010) 1295–1298.
[7] E. Castillejos, P.-J. Debouttière, L. Roiban, A. Solhy, V. Martinez, Y. Kihn, O.
Ersen, K. Philippot, B. Chaudret, P. Serp, An efficient strategy to drive
nanoparticles into carbon nanotubes and the remarkable effect of
confinement on their catalytic performance, Angew. Chem. Int. Ed. 48 (2009)
2529–2533.
[8] H. Lin, J. Zheng, X. Zheng, Z. Gu, Y. Yuan, Y. Yang, Improved chemoselective
hydrogenation of crotonaldehyde over bimetallic AuAg/SBA-15 catalyst, J.
Catal. 330 (2015) 135–144.
Fig. 14. Mass spectra of CALA product in Pt3Fe/CNT-catalyzed hydrogenation
reaction runs using H2O and D2O as solvent. Reaction conditions: 10 mg catalyst,
200 mL CALD, 10 mL solvent, 1 bar H2, 25 °C, and 1 h.

Y. Dai et al. / Journal of Catalysis 364 (2018) 192–203
203
[9] H. Yin, S. Zhao, K. Zhao, A. Muqsit, H. Tang, L. Chang, H. Zhao, Y. Gao, Z. Tang,
Ultrathin platinum nanowires grown on single-layered nickel hydroxide with
high hydrogen evolution activity, Nat. Commun. 6 (2015) 6430.
[10] Y. Dai, Y. Wang, B. Liu, Y. Yang, Metallic nanocatalysis: an accelerating
seamless integration with nanotechnology, Small 11 (2015) 268–289.
[11] W.O. Oduro, N. Cailuo, K.M.K. Yu, H. Yang, S.C. Tsang, Geometric and electronic
effects on hydrogenation of cinnamaldehyde over unsupported Pt-based
nanocrystals, Phys. Chem. Chem. Phys. 13 (2011) 2590–2602.
[12] S.C. Tsang, N. Cailuo, W. Oduro, A.T.S. Kong, L. Clifton, K.M.K. Yu, B. Thiebaut, J.
Cookson, P. Bishop, Engineering preformed cobalt-doped platinum
nanocatalysts for ultraselective hydrogenation, ACS Nano 2 (2008) 2547–2553.
[13] Y. Zhu, F. Zaera, Selectivity in the catalytic hydrogenation of cinnamaldehyde
promoted by Pt/SiO2 as a function of metal nanoparticle size, Catal. Sci.
Technol. 4 (2014) 955–962.
[36] H.G. Manyar, D. Weber, H. Daly, J.M. Thompson, D.W. Rooney, L.F. Gladden, E.
Hugh Stitt, J. Jose Delgado, S. Bernal, C. Hardacre, Deactivation and
regeneration of ruthenium on silica in the liquid-phase hydrogenation of
butan-2-one, J. Catal. 265 (2009) 80–88.
[37] S.-I. Fujita, Y. Sano, B.M. Bhanage, M. Arai, Supported liquid-phase catalysts
containing ruthenium complexes for selective hydrogenation of
a, b-
unsaturated aldehyde: Importance of interfaces between liquid film, solvent,
and support for the control of product selectivity, J. Catal. 225 (2004) 95–104.
[38] A. Chambers, S. David Jackson, D. Stirling, G. Webb, Selective Hydrogenation of
cinnamaldehyde over supported copper catalysts, J. Catal. 168 (1997) 301–
314.
[39] H. Yoshitake, Y. Iwasawa, Cooperative behavior of two kinds of reaction sites
and reaction mechanisms for deuteration of acrolein on SMSI-Pt/Nb2O5
catalyst, J. Catal. 125 (1990) 227–242.
[14] K.R. Kahsar, D.K. Schwartz, J.W. Medlin, Control of metal catalyst selectivity
through specific noncovalent molecular interactions, J. Am. Chem. Soc. 136
(2014) 520–526.
[40] G. Chen, Y. Tan, B. Wu, G. Fu, N. Zheng, Carbon monoxide-controlled synthesis
of surface-clean Pt nanocubes with high electrocatalytic activity, Chem.
Commun. 48 (2012) 2758–2760.
[15] Z. Guo, C. Xiao, R.V. Maligal-Ganesh, L. Zhou, T.W. Goh, X. Li, D. Tesfagaber, A.
Thiel, W. Huang, Pt nanoclusters confined within metal-organic framework
[41] G. Kresse, J. Furthmüller, Efficient iterative schemes for ab initio total-energy
calculations using a plane-wave basis set, Phys. Rev. B 54 (1996) 11169–
11186.
[42] G. Kresse, J. Hafner, Ab initio, Phys. Rev. B 48 (1993) 13115–13118.
[43] J.P. Perdew, J.A. Chevary, S.H. Vosko, K.A. Jackson, M.R. Pederson, D.J. Singh, C.
Fiolhais, Atoms, molecules, solids, and surfaces: applications of the generalized
gradient approximation for exchange and correlation, Phys. Rev. B 46 (1992)
6671–6687.
[44] L. Giordano, G. Pacchioni, J. Goniakowski, N. Nilius, E.D.L. Rienks, H.-J. Freund,
Interplay between structural, magnetic, and electronic properties in a FeO/Pt
(111) ultrathin film, Phys. Rev. B 76 (2007) 075416.
[45] W. Zhang, Z. Li, Y. Luo, J. Yang, First principles study on the geometric and
electronic structures of the FeO/Pt(111) surface, J. Phys. Chem. C 113 (2009)
8302–8305.
cavities for chemoselective cinnamaldehyde hydrogenation, ACS Catal.
(2014) 1340–1348.
4
[16] M. Zhao, K. Yuan, Y. Wang, G. Li, J. Guo, L. Gu, W. Hu, H. Zhao, Z. Tang, Metal-
organic frameworks as selectivity regulators for hydrogenation reactions,
Nature 539 (2016) 76–80.
[17] M. Zhao, K. Deng, L. He, Y. Liu, G. Li, H. Zhao, Z. Tang, Core-shell palladium
nanoparticle@Metal-organic frameworks as multifunctional catalysts for
cascade reactions, J. Am. Chem. Soc. 136 (2014) 1738–1741.
[18] M.-O. Simon, C.-J. Li, Green chemistry oriented organic synthesis in water,
Chem. Soc. Rev. 41 (2012) 1415–1427.
[19] J. Fester, M. García-Melchor, A.S. Walton, M. Bajdich, Z. Li, L. Lammich, A.
Vojvodic, J.V. Lauritsen, Edge reactivity and water-assisted dissociation on
cobalt oxide nanoislands, Nat. Commun. 8 (2017) 14169.
[20] A.M. Abdel-Mageed, S. Eckle, R.J. Behm, High selectivity of supported Ru
catalysts in the selective CO methanation-water makes the difference, J. Am.
Chem. Soc. 137 (2015) 8672–8675.
[21] G. Chen, Y. Zhao, G. Fu, P.N. Duchesne, L. Gu, Y. Zheng, X. Weng, M. Chen, P.
Zhang, C.-W. Pao, J.-F. Lee, N. Zheng, Interfacial effects in iron-nickel
hydroxide-platinum nanoparticles enhance catalytic oxidation, Science 344
(2014) 495–499.
[22] Y. Yang, C.A. Mims, D.H. Mei, C.H.F. Peden, C.T. Campbell, Mechanistic studies
of methanol synthesis over Cu from CO/CO2/H2/H2O mixtures: the source of C
in methanol and the role of water, J. Catal. 298 (2013) 10–17.
[46] G. Henkelman, B.P. Uberuaga, H. Jónsson, A climbing image nudged elastic
band method for finding saddle points and minimum energy paths, J. Chem.
Phys. 113 (2000) 9901–9904.
[47] M. Ritter, W. Ranke, W. Weiss, Growth and structure of ultrathin FeO films on
Pt(111) studied by STM and LEED, Phys. Rev. B 57 (1998) 7240–7251.
[48] Z.-T. Liu, C.-X. Wang, Z.-W. Liu, J. Lu, Selective hydrogenation of
cinnamaldehyde over Pt-supported multi-walled carbon nanotubes: Insights
into the tube-size effects, Appl. Catal. A Gen. 344 (2008) 114–123.
[49] Q. Fu, W.-X. Li, Y. Yao, H. Liu, H.-Y. Su, D. Ma, X.-K. Gu, L. Chen, Z. Wang, H.
Zhang, B. Wang, X. Bao, Interface-confined ferrous centers for catalytic
oxidation, Science 328 (2010) 1141–1144.
[23] B.N. Zope, D.D. Hibbitts, M. Neurock, R.J. Davis, Reactivity of the gold/water
interface during selective oxidation catalysis, Science 330 (2010) 74–78.
[24] C.-R. Chang, Z.-Q. Huang, J. Li, The promotional role of water in heterogeneous
catalysis: Mechanism insights from computational modeling, WIREs Comput.
Mol. Sci. 6 (2016) 679–693.
[25] D.D. Hibbitts, B.T. Loveless, M. Neurock, E. Iglesia, Mechanistic role of water on
the rate and selectivity of Fischer-Tropsch synthesis on ruthenium catalysts,
Angew. Chem. Int. Ed. 52 (2013) 12273–12278.
[26] L. Yang, Y. Choi, W. Qin, H. Chen, K. Blinn, M. Liu, P. Liu, J. Bai, T.A. Tyson, M. Liu,
Promotion of water-mediated carbon removal by nanostructured barium
oxide/nickel interfaces in solid oxide fuel cells, Nat. Commun. 2 (2011) 357.
[27] S. Liu, R. Ohnishi, M. Ichikawa, Promotional role of water added to methane
feed on catalytic performance in the methane dehydroaromatization reaction
on Mo/HZSM-5 catalyst, J. Catal. 220 (2003) 57–65.
[28] C. Wang, X.-K. Gu, H. Yan, Y. Lin, J. Li, D. Liu, W.-X. Li, J. Lu, Water-mediated
Mars-Van Krevelen mechanism for CO oxidation on ceria-supported single-
atom Pt1 catalyst, ACS Catal. 7 (2017) 887–891.
[50] T.-C. Lin, G. Seshadri, J.A. Kelber, A consistent method for quantitative XPS peak
analysis of thin oxide films on clean polycrystalline iron surfaces, Appl. Surf.
Sci. 119 (1997) 83–92.
[51] C. Zhou, Y. Chen, Z. Guo, X. Wang, Y. Yang, Promoted aerobic oxidation of
benzyl alcohol on CNT supported platinum by iron oxide, Chem. Commun. 47
(2011) 7473–7475.
[52] S. Galvagno, A. Donato, G. Neri, R. Pietropaolo, D. Pietropaolo, Hydrogenation
of cinnamaldehyde over platinum catalysts: influence of addition of metal
chlorides, J. Mol. Catal. 49 (1989) 223–232.
[53] N. Mahata, F. Gonçalves, M.F.R. Pereira, J.L. Figueiredo, Selective hydrogenation
of cinnamaldehyde to cinnamyl alcohol over mesoporous carbon supported Fe
and Zn promoted Pt catalyst, Appl. Catal. A Gen. 339 (2008) 159–168.
[54] Z. Guo, C. Zhou, D. Shi, Y. Wang, X. Jia, J. Chang, A. Borgna, C. Wang, Y. Yang,
Toward the decoration of Pt nanoparticles supported on carbon nanotubes
with Fe oxides and its effect on the catalytic reaction, Appl. Catal. A Gen. 435–
436 (2012) 131–140.
[55] Q. Hu, S. Wang, Z. Gao, Y. Li, Q. Zhang, Q. Xiang, Y. Qin, The precise decoration of
Pt nanoparticles with Fe oxide by atomic layer deposition for the selective
hydrogenation of cinnamaldehyde, Appl. Catal. B Environ. 218 (2017) 591–599.
[56] D. Wang, D. Astruc, The Golden Age of Transfer Hydrogenation, Chem. Rev. 115
(2015) 6621–6686.
[29] M. Daté, M. Okumura, S. Tsubota, M. Haruta, Vital role of moisture in the
catalytic activity of supported gold nanoparticles, Angew. Chem. Int. Ed. 43
(2004) 2129–2132.
[30] L.R. Merte, G. Peng, R. Bechstein, F. Rieboldt, C.A. Farberow, L.C. Grabow, W.
Kudernatsch, S. Wendt, E. L
mediated proton hopping on an iron oxide surface, Science 336 (2012) 889–
893.
æ
gsgaard, M. Mavrikakis, F. Besenbacher, Water-
[57] Q. Zhang, S.-S. Li, M.-M. Zhu, Y.-M. Liu, H.-Y. He, Y. Cao, Direct reductive
amination of aldehydes with nitroarenes using bio-renewable formic acid as a
hydrogen source, Green Chem. 18 (2016) 2507–2513.
[31] Z. He, Q. Qian, J. Ma, Q. Meng, H. Zhou, J. Song, Z. Liu, B. Han, Water-enhanced
synthesis of higher alcohols from CO2 hydrogenation over a Pt/Co3O4 catalyst
under milder conditions, Angew. Chem. Int. Ed. 55 (2016) 737–741.
[32] M.-M. Wang, L. He, Y.-M. Liu, Y. Cao, H.-Y. He, K.-N. Fan, Gold supported on
mesostructured ceria as an efficient catalyst for the chemoselective
hydrogenation of carbonyl compounds in neat water, Green Chem. 13
(2011) 602–607.
[58] J. Hájek, N. Kumar, P. Mäki-Arvela, T. Salmi, D.Y. Murzin, I. Paseka, T. Heikkilä,
E. Laine, P. Laukkanen, J. Väyrynen, Ruthenium-modified MCM-41 mesoporous
molecular sieve and
Y zeolite catalysts for selective hydrogenation of
cinnamaldehyde, Appl. Catal. A Gen. 251 (2003) 385–396.
[59] P. Mäki-Arvela, L.-P. Tiainen, A.K. Neyestanaki, R. Sjöholm, T.-K. Rantakylä, E.
Laine, T. Salmi, D.Y. Murzin, Liquid phase hydrogenation of citral: suppression
of side reactions, Appl. Catal. A Gen. 237 (2002) 181–200.
[33] X. Xiang, W. He, L. Xie, F. Li, A mild solution chemistry method to synthesize
hydrotalcite-supported platinum nanocrystals for selective hydrogenation of
cinnamaldehyde in neat water, Catal. Sci. Technol. 3 (2013) 2819–2827.
[34] Z. Guo, Y. Chen, L. Li, X. Wang, G.L. Haller, Y. Yang, Carbon nanotube-supported
[60] Y. Li, X. Xu, P. Zhang, Y. Gong, H. Li, Y. Wang, Highly selective Pd@mpg-C3N4
catalyst for phenol hydrogenation in aqueous phase, RSC Adv. 3 (2013) 10973–
10982.
[61] T. Kurita, F. Aoki, T. Mizumoto, T. Maejima, H. Esaki, T. Maegawa, Y. Monguchi,
H. Sajiki, Facile and convenient method of deuterium gas generation using a
Pd/C-catalyzed H2–D2 exchange reaction and its application to synthesis of
deuterium-labeled compounds, Chem. Eur. J. 14 (2008) 3371–3379.
[62] P. Beccat, J.C. Bertolini, Y. Gauthier, J. Massardier, P. Ruiz, Crotonaldehyde and
methylcrotonaldehyde hydrogenation over Pt(111) and Pt80Fe20(111) single
crystals, J. Catal. 126 (1990) 451–456.
Pt-based bimetallic catalysts prepared by
a microwave-assisted polyol
reduction method and their catalytic applications in the selective
hydrogenation, J. Catal. 276 (2010) 314–326.
[35] J. Hidalgo-Carrillo, M.A. Aramendía, A. Marinas, J.M. Marinas, F.J. Urbano,
Support and solvent effects on the liquid-phase chemoselective hydrogenation
of crotonaldehyde over Pt catalysts, Appl. Catal. A Gen. 385 (2010) 190–200.