Selective Hydrogenation of Biomass-Derived Furfural: Enhanced Catalytic Performance of Pd?Cu Alloy Nanoparticles in Porous Polymer

Article abstract of DOI:10.1002/cplu.202000383

Here, the development of a new catalyst is reported for the selective furfural (FF) hydrogenation to furfuryl alcohol (FA) based on about 7 nm sized Pd?Cu alloy nanoparticles (NPs) formed in inexpensive, commercially available micro/mesoporous hypercrosslinked polystyrene (HPS). A comparison of the catalytic properties of as-synthesized and reduced (denoted “r”) catalysts as well as Pd?Cu alloy and monometallic palladium NPs showed a considerable enhancement of the catalytic performance of Pd?Cu/HPS-r compared to other catalysts studied, resulting in about 100 percent FF conversion, 95.2 percent selectivity for FA and a TOF of 1209 h^?1. This was attributed to the enrichment of the NP surface with copper atoms, disrupting the furan ring adsorption, and to the presence of both zerovalent and cationic palladium and copper species, resulting in optimal hydrogen and FF adsorption. These factors along with exceptional stability of the catalyst in ten consecutive catalytic cycles make it highly promising in practical applications.

Full text of DOI:10.1002/cplu.202000383

A Multidisciplinary Journal Centering on Chemistry
Accepted Article
Title: Selective Hydrogenation of Biomass-Derived Furfural: Enhanced
Catalytic Performance of Pd-Cu Alloy Nanoparticles in Porous
Polymer
Authors: Ksenia Salnikova, Yurii Larichev, Esther Sulman, Alexei
Bykov, Alexander Sidorov, Galina Demidenko, Mikhail
Sulman, Lyudmila M Bronstein, and Valentina Matveeva
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemPlusChem 10.1002/cplu.202000383
Link to VoR: https://doi.org/10.1002/cplu.202000383
0
1/2020

ChemPlusChem
10.1002/cplu.202000383
FULL PAPER
Selective Hydrogenation of Biomass-Derived Furfural: Enhanced
Catalytic Performance of Pd-Cu Alloy Nanoparticles in Porous
Polymer
Ksenia E. Salnikova,[a,b] Yurii V. Larichev,[c,d] Esther M. Sulman, Alexey V. Bykov, Alexander I.
[a]
[a]
Sidorov, Galina N. Demidenko, Mikhail G. Sulman, Lyudmila M. Bronstein,* [e,f,g] Valentina G.
[
a]
[a]
[a]
Matveeva,* [a,b]
[
a]
Ms. K. E. Salnikova, Prof. E. M. Sulman, Dr. A. V. Bykov, Dr. A. I. Sidorov, Dr. G. D. Demidenko, Prof. M. G. Sulman, Prof. V. G. Matveeva
Department of Biotechnology and Chemistry
Tver State Technical University
A.Nikitin str., 22, 170026 Tver, Russia
E-mail: matveeva@science.tver.ru
Ms. K. E. Salnikova, Prof. V. G. Matveeva
Regional technological centre
[
b]
Tver State University
Zhelyabova str., 33, 170100 Tver, Russia
Dr. Y. V. Larichev
Boreskov Institute of Catalysis
[
[
[
c]
d]
e]
5
Academician Lavrentiev Avenue, 630090 Novosibirsk, Russia
Dr. Y. V. Larichev
Novosibirsk State University
2
Pirogov Street, 630090 Novosibirsk, Russia
Dr. L. M. Bronstein
Department of Chemistry
Indiana University
800 E. Kirkwood Av., Bloomington, IN, 47408, USA
E-mail: lybronst@indiana.edu
[
[
f]
Dr. L. M. Bronstein
A.N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences
28 Vavilov St., Moscow, 119991 Russia
g]
Dr. L. M. Bronstein
Faculty of Science, Department of Physics
King Abdulaziz University
Jeddah, 21589 Saudi Arabia
Supporting information for this article is given via a link at the end of the document.
Abstract: Here, we report the development of a new catalyst for
selective furfural (FF) hydrogenation to furfuryl alcohol (FA) based
on ~ 7 nm Pd-Cu alloy nanoparticles (NPs) formed in inexpensive,
biomass ingredients as precursors for syntheses of value-added
chemicals. Furfural (FF) is a high-output chemical obtained by
dehydration of xylose (a byproduct of farming and wood
processing)[1] and utilized for hydrogenation to furfuryl alcohol
(FA), tetrahydrofurfuryl alcohol (THFA), cyclopentanone, etc. [
FA is generally used in a paint production, as solvent and
precursor in the production of glass fiber and polymer
concrete.[ Heterogeneous catalysts are commonly employed in
the FF hydrogenation due to their high stability, facile separation
from the reaction medium, and recyclability.[4]
commercially
available
micro/mesoporous
hypercrosslinked
2]
polystyrene (HPS). A comparison of catalytic properties of as-
synthesized and reduced ( denoted “r”) catalysts as well as Pd-Cu
alloy and monometallic Pd NPs showed
a
considerable
3]
enhancement of the catalytic performance of Pd-Cu/HPS-r
compared to other catalysts studied, resulting in ~100% FF
-
1
conversion, 95.2% of FA selectivity and TOF of 1209 h . This was
attributed to the enrichment of the NP surface with Cu atoms,
disrupting the furan ring adsorption, and the presence of both
zerovalent and cationic Pd and Cu species, resulting in optimal
hydrogen and FF adsorption. These factors along with exceptional
stability of the catalyst in ten consecutive catalytic cycles make it
highly promising in practical applications.
The most frequently utilized noble metal catalysts in FF
hydrogenation are based on Pd nanoparticles (NPs) stabilized
by various supports[ such as carbons (non-modified and
5]
chemically modified)[6] and metal oxides (SiO
2 2 3
, Al O , MgAl
oxides, etc.), including mesoporous solids and those modified
with transition metals (Ti, Zr, Sn).[2b, 6b] The support can control
the interactions between the metal surface and the furan ring,
thus, controlling selectivity. Pd NPs formed on amorphous
supports (SiO
2 2 3
, Al O ) were found to be quite efficient for the FF
Introduction
hydrogenation, while doped ordered mesoporous supports
(
TiMCM-41, ZrMCM-41, SnMCM-41) allowed for higher
In recent years, a great deal of attention has been paid to the
development of environmentally friendly processes using
[2b]
selectivity to FA and THFA.
Moreover, Pd NPs placed in
1
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.202000383
FULL PAPER
mesopores were more stable than those on the support surface,
allowing higher reusability. The support porosity is definitely
advantageous for catalysis, but commercially available
mesoporous silica and alumina have low attrition resistance
which jeopardizes the catalyst stability.[7]
main products are furan and 2-methylfuran (Scheme 1b) with no
selectivity to any product. This was surprizing considering high
selectivity and conversion of Cu-containing catalysts on metal
oxide supports[ and could be tentatively attributed to the HPS
influence. For Pd/HPS catalysts (Table S2), the conversion
increases with the increase of the Pd content and for all the
samples, two products are observed: FA and THFA. The best
selectivity to FA was obtained for the catalyst containing 3 wt.%
of Pd. Moreover, for all the samples, both conversion and
selectivity are higher for the catalysts where Pd-containing NPs
were reduced by hydrogen vs. as-synthesized catalysts, whose
oxidized Pd species were reduced in situ during the catalyst
activation and the catalytic reaction.
10]
Porous crosslinked polymers have shown multiple
advantages as catalyst supports, including controlled porosity
and a high tensile strength, i.e., enhanced attrition resistance.[
In our preceding work, Pd NPs were formed in commercially
available micro/mesoporous hypercrosslinked polystyrene (HPS)
8]
using
Pd(CH
two
different
precursors,
PdCl
2
(CH
3
CN)
2
2a]
and
[
3
COO) , and studied in FF hydrogenation to FA. It was
2
shown that FF conversion and selectivity to FA were higher for
smaller Pd NPs, but no other structural features were explored.
It is well-established that modification of Pd NPs with a
second metal is an excellent way to control a reaction pathway
and selectivity in the FF hydrogenation via weakening the H
2
adsorption on Pd and altering the FF adsorption.[ Poisoning of
Pd with Bi allowed the FF hydrogenation with a high selectivity
to FA and cyclopentanone.[ It is believed that in the Pd-Bi
catalyst there is an active site isolation, which is accountable for
the increase of selectivity to target products, while eliminating
9]
9d]
oligomeric byproducts. Copper is another metal of choice for Pd
NP modification.[
1a, 4a, 4c, 5a, 6c, 9c, 10]
Alloying of Pd and Cu leads to
higher reactivity.[6c, 9c] In the case of Cu@Pd core-shell bimetallic
catalysts, it was demonstrated that an enhanced activity can be
assigned to decreased hydrogen surface coverage, making
more catalytic sites accessible for FF adsorption.[
1a]
Taking into account successes of Pd-Cu alloys in the FF
hydrogenation to FA and advantages of the micro/mesoporous
HPS support in a number of hydrogenation reactions,[11] in this
work we developed novel heterogeneous catalysts with Pd-Cu
alloy NPs in the HPS pores and compared their properties with
those of monometallic Pd NPs. We demonstrate that the
formation of Pd-Cu alloy NPs, whose surface is enriched with Cu
atoms and contains a fraction of cationic Pd and Cu species,
leads to high FF conversion and FA selectivity as well as an
enhanced activity and reusability.
Scheme 1. Schematic representation of the formation of Pd-Cu NPs in the
HPS pores (a) and FF hydrogenation (b). The red rectangle in (b) shows the
target product, FA.
For bimetallic catalysts with the Pd:Cu=1:1.7 molar ratio (Table
1), the increase of the metal loading leads to the FF conversion
increase for both as-synthesized and reduced catalysts. The
maximum FF conversion of 99.9% is achieved for reduced
catalysts containing 3 or 5 wt.% of each metal. At the same time,
the selectivity varies. For all bimetallic catalysts, furan and 2-
methylfuran are present in the reaction mixture due to the
presence of Cu (Table S2). The best selectivity to FA (95.2%) is
obtained for 3 wt.% of metal loading, while at the higher content
Results and Discussion
Catalytic properties
The catalyst synthesis was carried out by impregnation of HPS
with the solution containing both palladium and copper acetates,
2 3
followed by the Na CO treatment to precipitate NPs in the HPS
pores (Scheme 1a). This results in the formation of mixed Pd-Cu
oxides as reported elsewhere.[ These as-synthesized samples
are denoted “as”. The reduction of Pd and Cu species was
performed prior to the catalytic reaction in the hydrogen flow at
(5 wt.%) the selectivity to FA drops to 87.3% due to the
12]
increased THFA amount, indicating an acceleration of the C=C
hydrogenation of the furan ring. An increase or decrease of the
Pd:Cu molar ratio to 1:0.33 or 1:8.4 (Pd:Cu weight ratios of 5:1
and 1:5, Table 1) results in the decrease of the selectivity to FA
and the increase of fractions of other compounds: THFA, furan,
and 2-methylfuran.
Thus, the best conversion (99.9%) and selectivity (95.2%) were
obtained for the reduced Pd-Cu catalyst containing 3 wt.% of
each metal. For this catalyst, a carbon balance was calculated
275 °C. The reduced samples are denoted “r”. The Pd and Cu
contents in the catalysts were varied between 1 and 5 wt.% by
loading, keeping the Pd:Cu molar ratio unchanged at 1:1.7
(Table 1). The molar ratio was varied for the catalysts containing
1
and 5 wt.% of Pd or Cu. The elemental analysis data for key
samples are presented in Table S1.
Before studying catalytic properties of bimetallic Pd-Cu/HPS
catalysts in the FF hydrogenation, we tested monometallic
Cu/HPS and Pd/HPS catalysts synthesized in a similar way. The
data presented in Table S2 demonstrate that although Cu
catalyzes the FF hydrogenation, the conversion is low and the
using an overall carbon yield to verify fidelity of the product
analysis.[ . The overall carbon yield (%) expressed as (5 × C
13]
5
moles + 4 × C
9.78%. Taking into account cleavage of CO (not measured)
upon the furan formation, the total carbon yield is 100%.
4 5
moles)products/ (5 × C moles)FF × 100% is equal
9
2
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.202000383
FULL PAPER
Table 1. Catalytic properties of Pd-Cu bimetallic catalysts in the FF hydrogenation.[a]
Molar ratio,
FF
Catalyst
Pd/Cu, wt.% (by
XRF)
Selectivity,%
FA
conversion,
THFA
Furan
2-methyl-furan
Pd:Cu
%
Pd-Cu/HPS-as
Pd-Cu/HPS-r
3/3
3/3
1/1
1/1
5/5
5/5
1/5
1/5
5/1
5/1
3/0
3/0
1:1.7
1:1.7
1:1.7
1:1.7
1:1.7
1:1.7
1:8.4
1:8.4
1:0.33
1:0.33
-
88.5
86.2
95.2
76.2
78.5
77.2
87.3
67.9
73.3
61.6
79.2
74.2
84.6
7.7
2.3
1.1
7.4
4.1
5.4
2.3
3.8
0.5
5.2
3.6
4.7
1.9
10.0
8.8
7.6
1.9
-
99.9
69.2
78.4
94.9
99.9
66.2
75.5
96.5
98.2
63.9
79.9
3.2
Pd-Cu/HPS(1)-as
Pd-Cu/HPS(1)-r
Pd-Cu/HPS(2)-as
Pd-Cu/HPS(2)-r
Pd-Cu/HPS(3)-as
Pd-Cu/HPS(3)-r
Pd-Cu/HPS(4)-as
Pd-Cu/HPS(4)-r
Pd/HPS-as
11.2
13.8
12.7
8.5
8.9
13.2
7.7
10.5
3.3
-
10.2
20.3
15.6
25.8
15.4
Pd/HPS-r
-
-
-
[
a]
2
Reaction conditions: temperature 120 °C, pressure (H ) 6 MPa, furfural volume 2 mL, isopropanol volume 48 mL, catalyst amount 0.1 g, reaction time 180 min.
Monometallic Pd/HPS-r with the same Pd content showed much
lower conversion (79.9%) and selectivity (84.6%) than those for
Pd-Cu/HPS-r.
The dependencies of the conversion and selectivity on time,
hydrogen pressure, and the reaction temperature for Pd/HPS
and Pd-Cu/HPS samples with 3 wt.% of each metal loading
(Table S1) are presented in Figure 1. The data show that for all
Influence of FF hydrogenation conditions on catalysis
the samples, the highest conversion is achieved for 180 min and
at the 6 MPa hydrogen pressure, while the temperatures for
reaching the highest conversion differ. For Pd-Cu/HPS-r, it is
already reached at 120 °C, while for three other samples, a
higher temperature (140 °C) is needed. As for selectivity, the
highest value is obtained at the beginning of the reaction which
is consistent with literature data.[14] For Pd-Cu-HPS-r, the
selectivity remains nearly unchanged until the end of the
reaction, while for other samples, it noticeably drops. For all
samples, the selectivity is the highest at 6 MPa pressure and
a
Pd-Cu/HPS-as
Pd-Cu/HPS-r
Pd/HPS-as
Pd/HPS-r
b
100
1
00
8
6
4
2
0
0
0
0
0
80
60
40
20
0
Pd-Cu/HPS-as
Pd-Cu/HPS-r
Pd/HPS-as
PD/HPS-r
0
20 40 60 80 100 120 140 160 180
Time, min
0
20 40 60 80 100 120 140 160 180
Time, min
120 °C, which determined our choice of optimal reaction
d
c
100
conditions. It is noteworthy that in all reaction conditions, the
conversion and selectivity are higher for Pd-Cu bimetallic
catalysts compared to Pd/HPS.
100
90
90
80
70
60
50
40
30
20
80
70
Pd-Cu/HPS-as
Pd-Cu/HPS-r
Pd/HPS-as
Pd/HPS-r
Pd-Cu/HPS-as
Pd-Cu/HPS-r
Pd/HPS-as
Pd/HPS-r
Structure, morphology, and textural properties
60
50
To better understand the influence of the synthetic procedure
(prior to reduction) and the Cu presence on the behavior of
bimetallic Pd-Cu/HPS catalysts, we studied structure,
morphology, and textural properties of four Pd-Cu/HPS and
Pd/HPS samples listed in Table S2.
Porosity data for bimetallic Pd-Cu and monometallic Pd
samples are presented in Table S3 and Figure S1. Liquid
nitrogen adsorption-desorption isotherms (Figure S1a,b) are
typical for micro-mesoporous hypercrosslinked polymers.[15] The
4
5
6
7
8
4
5
6
7
8
Pressure, MPa
Pressure, MPa
e
f100
1
00
90
80
70
60
50
40
30
90
8
0
0
7
Pd-Cu/HPS-as
Pd-Cu/HPS-r
Pd/HPS-as
Pd/HPS-r
Pd-Cu/HPS-as
Pd-Cu/HPS-r
Pd/HPS-as
Pd/HPS-r
60
50
changes in surface areas and pore volumes (Table S3) show a
similar trend for HPS samples with both bimetallic and
monometallic NPs: upon precipitation of NPs in the HPS pores
100
110
120
130
C
140
100
110
120
130
C
140
Temperature,
o
Temperature,
o
Figure 1. Dependencies of the conversion (a, c, e) and selectivity (b, d, f) on
time (a, b), pressure (c, d), and temperature (e, f).
(
as-synthesized samples), the polymer porosity decreases and
then further decreases upon the NP reduction. The size of the
3
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.202000383
FULL PAPER
majority of mesopores (4 nm) remains unchanged but their pore
volume significantly decreases after NP incorporation. Larger
mesopores with diameters in the range of 4-20 nm display a
similar behavior. These data indicate that NPs are formed in the
pores, leading to the decrease of the pore volume and pore
surface area, however, they are not blocking pore junctions
because the porosity decrease is moderate.
6-7 nm which noticeably exceeds the typical pore size (4 nm)
(Figure S4). This reveals that NPs are mainly located in pore
junctions, but without blocking the pore system as about ~60%
of pore volume remains available.
The X-ray diffraction (XRD) patterns of Pd-Cu/HPS-r and
Pd/HPS-r and their high resolution (HR) X-ray photoelectron
spectroscopy (XPS) Pd3d spectra are presented in Figure 3.
To evaluate NP sizes, we used small-angle X-ray scattering
(
SAXS) and transmission electron microscopy (TEM). It is
Pd Pd
0
2
+
b
noteworthy that for NP-containing HPS samples, SAXS is a
more reliable technique because NPs can overlap in HPS
a
HPS
Pd/HPS-r
Pd-Cu/HPS-r
Pd/HPS-r
(Figure 2a, b), giving irregular, non-spherical projections and
making the NP size evaluation untrustworthy, unless very high
NP populations are examined for better fidelity. On the other
(111)
(
200)
Pd-Cu/HPS-r
332
(
220)
hand, in some cases SAXS data can result in the NP size
overestimation.[
16]
10
20
30
40
50
60
70
344
340
336
Thus, the combination of both techniques is
2
Θ, degree
Binding energy, eV
the most trustworthy approach. The NP diameter distributions
obtained by measuring 709 NPs for Pd-Cu/HPS-r and 841 NPs
for Pd/HPS-r on the respective TEM images are shown in Figure
S2. The experimental SAXS data and the best fits for spherical
NPs are presented in Figure S3. The radii of gyration and NP
diameters are shown in Table S4.
Figure 3. XRD patterns (a) of Pd-Cu/HPS-r and Pd/HPS-r and their HR XPS
Pd3d spectra (b).
The XRD pattern of Pd/HPS-r shows typical reflections of Pd0
(JCPDS-ICDD 46-1043), whose line broadening allowed us to
determine the crystallite size of 6 nm, indicating that Pd NPs are
most likely single crystals. For bimetallic Pd-Cu NPs in Pd-
Cu/HPS-r, the XRD pattern is nearly identical, but its reflections
are shifted towards larger angles and the peak positions are in
between those characteristic of Pd (JCPDS-ICDD 46-1043) and
Cu (JCPDS-ICDD 4-9-2090) phases. According to the literature
data,[
17]
this peak positioning reveals the formation of the
bimetallic Pd-Cu alloy. The absence of the reflections from Cu
metal as well as its oxides or hydroxides once more confirms the
alloy formation. The crystallite size of Pd-Cu alloy NPs is also ~6
nm which could be underestimation due to NP heterogeneity. It
is worth noting that as-synthesized mono- and bimetallic
samples are completely amorphous and their XRD patterns (not
shown) resemble that of HPS. This was anticipated, considering
mild conditions of the sample preparation (70 °C). Thus, the
hydrogen treatment (at 275 °C) of a mixed Pd-Cu oxide is
accompanied by metal reduction and crystallization. According
to Vegard's rule, ideal Pd-Cu alloys should show a linear
increase of lattice parameters with the Pd content increase.[ In
the case of Pd-Cu/HPS-r (37 at.% Pd by elemental analysis), the
calculated lattice parameter should be 3.75 Å, while in fact, it is
C
Pd-Cu/HPS-as
Pd-Cu/HPS-r
d
100
Pd/HPS-as
Pd/HPS-r
100
80
80
60
40
20
0
60
40
20
0
0
5
10
15
20
25
0
5
10
15
20
25
NP sizes, nm
NP sizes, nm
f
100
TEM data for Pd/HPS-r
SAXS data for Pd/HPS-r
e
TEM data for Pd-Cu/HPS-r
SAXS data for Pd-Cu/HPS-r
1
00
18]
80
80
60
40
20
0
60
40
.87 Å, revealing heterogeneity of alloy NPs,[19] which is further
3
20
revealed by XPS.
0
For both samples, the deconvolution of the HR XPS Pd3d
spectra results in two doublets in the Pd3d5/2 and Pd3d3/2
regions. The Pd3d5/2 peak positions at 335.0-335.1 eV and
336.5-336.9 eV (Table S5) are typical for Pd0 and Pd (most
0
5
10
15
20
0
5
10
15
20
NP sizes, nm
NP sizes, nm
2+
Figure 2. TEM images (a, b), volume NP size distribution from the SAXS data
c, d), and a comparison of the NP sizes from TEM and SAXS data (e, f) for
Pd-Cu/HPS-r (a, c, e) and Pd/HPS-r (b, d, f).
(
likely PdO),[20] respectively, with nearly the same Pd :Pd ratios
0
2+
(
Table S5). The HR XPS Cu2p spectrum of Pd-Cu/HPS-r (Figure
0
S5) shows that the major component is Cu with a smaller
2+
0
fraction of Cu (Table S6). The BE of Cu (932.2 eV) is typical
for Cu NPs.[ The presence of Cu2+ could be due to incomplete
reduction of CuO in the as-synthesized sample or oxidation of
For both bimetallic (Pd-Cu) and monometallic (Pd) samples
Figure 2 c, d), as-synthesized NPs are smaller than those after
21]
(
the hydrogen reduction. The increase of NP sizes upon
reduction indicates that there is a migration of metal species
between NPs, i.e., ripening, which is not surprising due to an
elevated reduction temperature (275 °C). The NP sizes obtained
from TEM and SAXS data are in good agreement, especially for
bimetallic NPs. The mean NP size from both methods is about
0
Cu species under air. A comparison of the Pd:Cu atomic ratio
from XPS (surface method) and the elemental analysis data
(bulk method) demonstrate an enrichment of the NP surface by
copper, i.e., heterogeneity. Both theory and experiment
demonstrate that the metal with the lowest surface energy has a
4
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.202000383
FULL PAPER
tendency to segregate to the surface of alloys.[22] In our case,
Cu with the average surface energy of 113.9 meVÅ- favors the
surface enrichment compared to Pd, whose average surface
chemical reactivity, which could be assigned to weaker
2
[6c,
interactions between metal sites and FA (for easy desorption).
9c]
This is well illustrated for Pd-Cu alloy NPs in Pd-Cu/HPS-r,
-
2 [22]
energy is 131 meVÅ .
Nonetheless, XRD demonstrates the
allowing for 99.9% of the FF conversion and the 95.2%
selectivity to FA (Table 1). To the best of our knowledge, this is
one of the highest values of the FA selectivity at the 100% FF
conversion, reported to date.[ Although a higher selectivity
(98.6%) has been described for Pd-Cu/MgO[ , a double catalyst
amount was required in that case. The presence of Cu in
bimetallic NPs most probably disrupts the weaker C=C bond
adsorption on neighboring Pd atoms, leading to the higher yield
of the target product. However, the question arises what degree
of disruption is needed and what electronic state of metal
species is favored?
formation of Pd-Cu alloy which enhances interactions between
[
23]
Pd and Cu atoms, boosting the NP catalytic performance.
14]
9c]
Influence of Pd-Cu interactions in alloy NPs on the catalytic
action
To evaluate the influence of Pd-Cu interactions in Pd-Cu/HPS-r
on the selective hydrogenation of FF to FA, we assessed the
functions of Cu and Pd in monometallic catalysts. The data
presented in Table S2 demonstrate that for all Cu/HPS catalysts,
the FF hydrogenation products are FA, furan, and 2-methylfuran
(
Scheme 2a). This is in agreement with the literature data,[5a]
It was established that a combination of Pd and Cu may
change the electronic structure of Pd species compared to pure
indicating that copper based catalysts facilitate hydrogenation of
the C=O bond due its adsorption on the catalytic site, while the
C=C bonds in the furan ring remain unchanged.[
Pd, decreasing the decarbonylation rate, while increasing the
1c, 4c, 5a, 10a]
[14a]
hydrogenation rate.
In the reaction mechanism study of
bimetallic Pd-Cu NPs, it was discovered that the presence of Cu
improves the selectivity to FA, while Pd mainly increases the FF
conversion most likely due to optimized hydrogen adsorption
14b]
(
Scheme 2c),[
preventing hydrogen oversaturation on the
surface and allowing for the better FF adsorption.[ Moreover,
for dissociative hydrogen adsorption on metal, the presence of
zerovalent and cationic metal species is essential.[9c, 14b] The
data presented in Tables S5 and S6 demonstrate that in Pd-Cu
1a]
alloy NPs (the Pd-Cu/HPS-r sample), both Pd -Pd and Cu0-
0
2+
2+
Cu pairs exist with approximately 20% of oxidized species.
This distribution of oxidized-reduced Pd-Cu species seems to be
beneficial for optimal hydrogen adsorption and high FF
conversion, while the disruption of Pd atom coverage on the NP
surface due to alloying and enrichment by Cu (Table S6) may
lead to suppression of the furan ring adsorption and its
hydrogenation, thus, high FA selectivity. The TOF value for Pd-
Cu/HPS-r, calculated per moles of NP surface atoms (taking into
account Pd:Cu ratios on the surface and in bulk, Table S6) is
1
209 h , which is comparable (1148 h in ref.[24]) or exceeds the
values reported for many other heterogeneous catalytic systems,
for example 204 h-1 in ref. [25], 21 h-1 in ref.[26], 96 h-1 in ref.[10b]
400 h-1 in ref.[27], 952 h-1 in ref.[28], etc. The only exception is a
customised catalytic membrane reactor with the Ru–
-
1
-1
,
~
-
1 [3b]
polyethersulfone membrane, allowing TOF of 48000 h , but a
special equipment is required in this case and the conversion is
only 26.13%.
In view of high conversion and selectivity of Pd-Cu/HPS-r,
we studied reusability of this catalyst in ten consecutive
reactions. Figure 4 demonstrates that both conversion and
selectivity to FA decrease by only a couple of percent after ten
cycles, indicating enhanced stability of alloy NPs in the HPS
pores as well as practically unchanged catalytic performance
which exceeds the catalytic performance upon reusability of
Scheme 2. Schematic representation of FF hydrogenation on Cu (a), Pd (b),
and Pd-Cu NPs.
[9c, 10a]
other catalysts.
Alternatively, for Pd/HPS catalysts (Table S2), the FF
[9c]
hydrogenation products are only FA and THFA, revealing that
on the Pd surface both the C=O group and the furan ring are
adsorbed due to planarity of the FF molecule. The С=С bonds
interact with Pd via π-π interactions (Scheme 2b).[5a, 14a]
It was demonstrated that when two metals, Cu and Pd, are
combined in a bimetallic NP, the NP surface properties as well
as the structure of active sites change, thus, affecting the
5
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.202000383
FULL PAPER
After that the catalyst was washed with distilled water until neutral pH
and dried at 70 °C for 2 h. The catalyst was denoted “as-synthesized”
Conversion
Selectivity
(
as). Reduction of Pd and Cu species in the catalyst was carried out in a
1
00
hydrogen flow at 275 °C for 2 h. The catalyst was denoted “reduced” (r).
Monometallic Cu- and Pd-containing catalysts were prepared in a similar
way using a single precursor.
8
6
4
2
0
0
0
0
0
Catalytic testing
FF hydrogenation was performed in a Parr reactor (Series 5000 Multiple
Reactor, stainless steel). In a typical experiment, the catalyst (0.1 g) was
mixed with 24 mL of isopropanol and treated in a hydrogen flow for 30
min at 120 °C and 6.0 MPa pressure under vigorous stirring (1000 rpm)
for the catalyst activation. After that 2 mL of FF and another 24 mL of
isopropanol were added to the reactor. Samples of the reaction mixture
were taken for analysis regularly and analyzed with GC-MS (Shimadzu
GCMS-QP2010S) equipped with a capillary column HP-1MS (30 m ×
1
2
3
4
5
6
7
8
9
10
Number of cycles
0
.25 mm i.d., 0.25 μm film thickness) (see details in the Supporting
Information). The experiments were performed for 180 min. The TOF
value for Pd-Cu/HPS-r was calculated as moles of the substrate
consumed per moles of NP surface atoms (calculated knowing the NP
size and composition) per hour.
Figure 4. Conversion and selectivity to FA in repeated uses of Pd-Cu/HPS-r.
The reaction time for all reuse experiments remained 180 min.
For reusability experiments, the reaction mixture was placed in glass
tubes and the catalyst was separated by centrifugation using a bench
centrifuge at 4,000 rpm. The supernatant was removed and the catalyst
was washed with isopropanol three times, following the same isolation
procedure. Finally, the catalyst was utilized in the repeated experiments
without drying.
Conclusion
We developed
a
novel catalyst for the selective FF
hydrogenation based on Pd-Cu alloy NPs and the commercially
available micro/mesoporous HPS support. The advantages of
this inexpensive polymer support such as high porosity and
mechanical robustness, were fully realized by forming well-
defined 6-7 nm NPs in the pore junctions, providing a sufficient
pore system (~60% by volume) for transport of reacting and
target molecules. XRD confirmed the formation of alloy Pd-Cu
NPs, while XPS showed the enrichment of the NP surface with
Cu atoms as well as the presence of both zerovalent and
cationic Pd and Cu species, i.e., NP heterogeneity. Such a
structure of Pd-Cu alloy NPs allowed for nearly 100%
conversion and excellent selectivity to FA (95.2%) along with a
Acknowledgements
The research leading to these results has received funding from
the Russian Science Foundation (project 19-19-00490),
supporting physiochemical studies and the manuscript
preparation. K.S. and A.S. thank the Russian Foundation for
Basic Research, which supported catalytic testing (grant 19-38-
90049) and studies of the catalyst reuse (grant 20-08-00433).
-
1
s
significant activity of the catalyst (TOF = 1209 h ). This
Y.L. acknowledges the resource center “VTAN” (Novosibirsk
State University) for the access to instrumentation.
exceptional performance was attributed to the prevention of the
furan ring adsorption on Pd due to neighboring Cu species and
facilitated FA desorption, both leading to higher selectivity.
Controlled hydrogen and FF adsorption due to mixed valence
states of Pd and Cu species lead to higher conversion. These
factors as well as remarkable catalyst reusability in ten
consecutive reactions makes this catalyst promising for practical
applications.
Keywords: furfural • hydrogenation • hypercrosslinked
polystyrene • nanostructures • Pd-Cu alloys
[
1]
a) P. Liu, W. Qiu, C. Zhang, Q. Tan, C. Zhang, W. Zhang,
Y. Song, H. Wang, C. Li, ChemCatChem 2019, 11, 3296-
3306; b) J.-P. Lange, E. van der Heide, J. van Buijtenen,
Experimental Section
R. Price, ChemSusChem 2012, 5, 150-166; c) K. Gupta, R.
K. Rai, S. K. Singh, ChemCatChem 2018, 10, 2326-2349;
d) X. Li, P. Jia, T. Wang, ACS Catal. 2016, 6, 7621-7640;
e) J. Parikh, S. Srivastava, G. C. Jadeja, Ind. Eng. Chem.
Res. 2019, 58, 16138-16152; f) D. Vargas-Hernandez, J.
M. Rubio-Caballero, R. Moreno-Tost, J. M. Merida-Robles,
J. Santamaria-Gonzalez, A. Jimenez-Lopez, M. A. Perez-
Cruz, R. Hernandez-Huesca, P. Maireles-Torres, Mod.
Res. Catal. 2016, 5, 85-94.
a) K. E. Salnikova, V. G. Matveeva, Y. V. Larichev, A. V.
Bykov, G. N. Demidenko, I. P. Shkileva, M. G. Sulman,
Catal. Today 2019, 329, 142-148; b) Z. Yi, H. Xu, D. Hu, K.
Yan, J. Alloys Compd. 2019, 799, 59-65.
The description of materials, analysis of catalytic samples, and
characterization methods are presented in the Supporting Information.
Synthesis of catalysts
The Pd-Cu-containing HPS-based catalyst was synthesized via a wet-
impregnation method according to the procedure described elsewhere for
Pd/HPS using Pd acetate as precursor.[2a] In a typical experiment, 1 g of
HPS granules (pretreated, dried and crushed to the size <60 µm) were
[
2]
3]
impregnated with
Pd(CH COO) and 0.090 g of Cu(CH
dried at 70 °C for 1 h and treated with 2.7 mL of the 0.3 M Na
3
mL of THF solution containing 0.071
COO) ×H O. The product was
CO
g of
[
a) A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107,
3
2
3
2
2
2411-2502; b) G. Bagnato, A. Figoli, C. Ursino, F. Galiano,
2
3
A. Sanna, J. Mater. Chem. A 2018, 6, 4955-4965.
[
6b, 14b]
solution for precipitation of oxides/hydroxide NPs.
6
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.202000383
FULL PAPER
[4]
[5]
[6]
a) C. Nguyen-Huy, H. Lee, J. Lee, J. H. Kwak, K. An, Appl.
Catal., A 2019, 571, 118-126; b) Y. Nakagawa, M. Tamura,
K. Tomishige, ACS Catal. 2013, 3, 2655-2668; c) S. Chen,
R. Wojcieszak, F. Dumeignil, E. Marceau, S. Royer, Chem.
Rev. 2018, 118, 11023-11117.
[22]
[23]
[24]
[25]
[26]
R. Ferrando, J. Jellinek, R. L. Johnston, Chem. Rev. 2008,
108, 845-910.
B. T. Meshesha, N. Barrabes, J. Llorca, A. Dafinov, F.
Medina, K. Foettinger, Appl. Catal., A 2013, 453, 130-141.
G. Wang, R. Yao, H. Xin, Y. Guan, P. Wu, X. Li, RSC Adv.
2018, 8, 37243-37253.
C. K. P. Neeli, Y.-M. Chung, W.-S. Ahn, ChemCatChem
2017, 9, 4570-4579.
Y. Feng, G. Yan, T. Wang, W. Jia, X. Zeng, J. Sperry, Y.
Sun, X. Tang, T. Lei, L. Lin, Green Chem. 2019, 21, 4319-
4323.
a) S. Sitthisa, D. E. Resasco, Catal. Lett. 2011, 141, 784-
791; b) C. Li, G. Xu, X. Liu, Y. Zhang, Y. Fu, Ind. Eng.
Chem. Res. 2017, 56, 8843-8849; c) J. Lee, J. Woo, C.
Nguyen-Huy, M. S. Lee, S. H. Joo, K. An, Catal. Today
2020, 350, 71-79.
a) R. Kosydar, D. Duraczynska, J. Gurgul, J. Krysciak-
Czerwenka, A. Drelinkiewicz, React. Kinet., Mech. Catal.
[27]
[28]
M. Li, L. Collado, F. Cardenas-Lizana, M. A. Keane, Catal.
Lett. 2018, 148, 90-96.
W. S. Putro, T. Kojima, T. Hara, N. Ichikuni, S. Shimazu,
Catal. Sci. Technol. 2017, 7, 3637-3646.
2019, 126, 417-437; b) R. M. Mironenko, V. P. Talsi, T. I.
Gulyaeva, M. V. Trenikhin, O. B. Belskaya, React. Kinet.,
Mech. Catal. 2019, 126, 811-827; c) M. Hronec, K.
Fulajtarova, I. Vavra, T. Sotak, E. Dobrocka, M. Micusik,
Appl. Catal., B 2016, 181, 210-219.
[
7]
a) N. Cherkasov, V. Jadvani, J. Mann, Y. B. Losovyj, Z. B.
Shifrina, L. M. Bronstein, E. V. Rebrov, Fuel Process.
Technol. 2017, 167, 738-746; b) Y. Wang, Y. S. Lin, J.
Sol-Gel Sci. Technol. 1998, 11, 185-195.
[
[
8]
9]
a) Z. B. Shifrina, V. G. Matveeva, L. M. Bronstein, Chem.
Rev. 2020, 120, 1350-1396; b) J. Huang, S. R. Turner,
Polym. Rev. (Philadelphia, PA, U. S.) 2018, 58, 1-41.
a) N. Pino, S. Sitthisa, Q. Tan, T. Souza, D. Lopez, D. E.
Resasco, J. Catal. 2017, 350, 30-40; b) Y. Nakagawa, K.
Takada, M. Tamura, K. Tomishige, ACS Catal. 2014, 4,
2
718-2726; c) K. Fulajtarova, T. Sotak, M. Hronec, I.
Vavra, E. Dobrocka, M. Omastova, Appl. Catal., A 2015,
02, 78-85; d) N. Cherkasov, A. J. Exposito, M. S. Aw, J.
5
Fernandez-Garcia, S. Huband, J. Sloan, L. Paniwnyk, E. V.
Rebrov, Appl. Catal., A 2019, 570, 183-191.
[
[
10]
11]
a) H. Du, X. Ma, P. Yan, M. Jiang, Z. Zhao, Z. C. Zhang,
Fuel Process. Technol. 2019, 193, 221-231; b) M. Lesiak,
M. Binczarski, S. Karski, W. Maniukiewicz, J. Rogowski, E.
Szubiakiewicz, J. Berlowska, P. Dziugan, I. Witonska, J.
Mol. Catal. A: Chem. 2014, 395, 337-348.
a) V. G. Matveeva, E. M. Sulman, O. V. Manaenkov, A. E.
Filatova, O. V. Kislitza, A. I. Sidorov, V. Y. Doluda, M. G.
Sulman, E. V. Rebrov, Catal. Today 2017, 280, 45-50; b)
V. N. Sapunov, M. Y. Grigoryev, E. M. Sulman, M. B.
Konyaeva, V. G. Matveeva, J. Phys. Chem. A 2013, 117,
4073-4083.
[
[
[
12]
13]
14]
J.-Y. Luo, M. Meng, J.-S. Yao, X.-G. Li, Y.-Q. Zha, X.
Wang, T.-Y. Zhang, Appl. Catal., B 2009, 87, 92-103.
C. J. Barrett, J. N. Chheda, G. W. Huber, J. A. Dumesic,
Appl. Catal., B 2006, 66, 111-118.
a) S. Sitthisa, T. Pham, T. Prasomsri, T. Sooknoi, R. G.
Mallinson, D. E. Resasco, J. Catal. 2011, 280, 17-27; b) J.
Du, J. Zhang, Y. Sun, W. Jia, Z. Si, H. Gao, X. Tang, X.
Zeng, T. Lei, S. Liu, L. Lin, J. Catal. 2018, 368, 69-78.
M. G. Schwab, I. Senkovska, M. Rose, N. Klein, M. Koch,
J. Pahnke, G. Jonschker, B. Schmitz, M. Hirscher, S.
Kaskel, Soft Matter 2009, 5, 1055-1059.
[
15]
[16]
[17]
[18]
[19]
T. Li, A. J. Senesi, B. Lee, Chem. Rev. 2016, 116, 11128-
1
1180.
G. Behmenyar, A. N. Akin, J. Power Sources 2014, 249,
39-246.
A. Peles, M. Shao, L. Protsailo, Catalysts 2015, 5, 1193-
201.
2
1
T. Kawaguchi, W. Cha, V. Latyshev, S. Vorobiov, V.
Komanicky, H. You, arXiv.org, e-Print Arch., Condens.
Matter 2019, 1-14.
[
[
20]
21]
a) L. Gregor, A. K. Reilly, T. A. Dickstein, S. Mazhar, S.
Bram, D. G. Morgan, Y. Losovyj, M. Pink, B. D. Stein, V. G.
Matveeva, L. M. Bronstein, ACS Omega 2018, 3, 14717-
14725; b) M. Brun, A. Berther, J. C. Bertolini, J. Electron
Spectrosc. Relat. Phenom. 1999, 104, 55-60.
a) T. Kruk, K. Szczepanowicz, J. Stefanska, R. P. Socha,
P. Warszynski, Colloids Surf., B 2015, 128, 17-22; b) Y.
Song, D. Cho, S. Venkateswarlu, M. Yoon, RSC Adv.
2017, 7, 10592-10600; c) Y. Bai, Q. Wang, C. Du, T. Bu, Y.
Liu, X. Sun, W. Luo, R. Li, Y. Zhao, X. Zheng, L. Wang, J.
Colloid Interface Sci. 2019, 553, 768-777.
7
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.202000383
FULL PAPER
Entry for the Table of Contents
Biomass-derived furfural hydrogenation: We developed novel Pd-Cu alloy nanoparticles (NPs) in pores of inexpensive,
commercially available hypercrosslinked polystyrene (HPS) and tested in selective furfural (FF) hydrogenation to furfuryl alcohol (FA).
The controlled surface composition of Pd-Cu NPs ensures high FF converson and FA selectvity, while the NP location in HPS pores
allows for high stability upon reuse in ten consecutive reactions.
8
This article is protected by copyright. All rights reserved.