Hydrogenolysis of 5-hydroxymethylfurfural to 2,5-dimethylfuran over supported Pt-Co bimetallic catalysts under mild conditions

Article abstract of DOI:10.1039/c8gc00716k

Highly dispersed Pt-Co bimetallic catalysts were deposited on multi-walled carbon nanotubes (MWCNTs) by atomic layer deposition. High-resolution TEM and TPR analyses verified the formation of Pt-Co bimetallic particles. Catalysts were applied for the hydrogenolysis of 5-hydroxymethyfurfural (HMF) to 2,5-dimethyfuran (DMF). A high yield of DMF (>90%) was achieved in the hydrogenolysis of HMF over the optimized Pt-Co/MWCNTs catalyst after 8 h of reaction time under 10 bar H₂ at 160 °C. Through a series of experiments and comparison, the synergistic effect among Pt, Co, and MWCNTs was investigated. The results revealed that the synergistic effect between Pt-Co and MWCNTs played an important role in the improvement of selectivity to DMF for Pt-Co/MWCNTs bimetallic catalysts. In addition, steric hindrance appeared when Co loading in Pt-Co/MWCNTs was high and it affected the activity of the Pt-Co bimetallic catalysts. However, moderate activity can inhibit the production of byproducts and thereby improve the yield of DMF.

Full text of DOI:10.1039/c8gc00716k

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: X. Wang, Y. Liu
and X. Liang, Green Chem., 2018, DOI: 10.1039/C8GC00716K.
This is an Accepted Manuscript, which has been through the
Volume 18 Number
7 7 April 2016 Pages 1821–2242
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Accepted Manuscripts are published online shortly after
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Please do not adjust margins
Green Chemistry
Page 1 of 10
View Article Online
DOI: 10.1039/C8GC00716K
Journal Name
ARTICLE
Hydrogenolysis of 5-hydroxymethylfurfural to 2,5-dimethylfuran
over supported Pt-Co bimetallic catalysts under mild conditions
Xiaofeng Wang,a Yuzi Liu b and Xinhua Liang *a
Received 00th January 20xx,
Accepted 00th January 20xx
DOI: 10.1039/x0xx00000x
Highly dispersed Pt-Co bimetallic catalysts were deposited on multi-walled carbon nanotubes (MWCNTs) by atomic layer
deposition. High-resolution TEM and TPR analysis verified the formation of Pt-Co bimetallic particles. Catalysts were
applied for hydrogenolysis of 5-hydroxymethyfurfural (HMF) to 2,5-dimethyfuran (DMF). High yield of DMF (> 90%) was
achieved in hydrogenolysis of HMF over the optimized Pt-Co/MWCNTs catalyst after 8 hr of reaction time under 10 bar H2
at 160 ºC. Through a series of experiments and comparison, the synergistic effect among Pt, Co, and MWCNTs was
investigated. The results revealed that the synergistic effect between Pt-Co and MWCNTs played an important role in the
improvement of selectivity to DMF for Pt-Co/MWCNTs bimetallic catalysts. In addition, steric hindrance appeared when
the Co loading in Pt-Co/MWCNTs was high and it affected the activity of Pt-Co bimetallic catalysts. However, the moderate
activity can inhibit the production of byprodcts and thereby improve the yield of DMF.
www.rsc.org/
group, but to avoid the further hydrogenation of a furan ring, are
the main challenge in the selective hydrogenolysis of HMF to DMF.
So far, many researchers have focused on the improvement of the
1. Introduction
Nowadays, fossil fuels, such as petroleum, coal, and natural gas,
play principal roles as source of energy and chemicals.1 However,
due to the gradual exhaust of fossil resource and greenhouse effect
for the consumption of fossil fuels, the transformation of renewable
biomass into biofuel has become a hotspot research area to replace
oil-based energy. With the advances in conversion technology,
abundant biomass resources have the potential as feedstock for
biofuels. 2,5-dimethylfuran (DMF), a product from hydrogenolysis
of biomass-derived 5-hydroxymethylfurfural (HMF), has a high
hydrogenolysis process and some heterogeneous catalysts have
5
been used for this reaction.3, For example, Nishimura et al.
prepared PdxAuy/C catalysts with various Pd/Au molar ratios (x/y) to
examine the hydrogenolysis of HMF to DMF in the presence of
hydrochloric acid (HCl) under an atmospheric hydrogen pressure.3
The highest conversion of HMF (99%) and yield of DMF (96%) was
obtained for Pd50Au50/C after a reaction time of 12 hr. Wang et al.
prepared PtCo bimetallic nanoparticles (NPs), and 98% DMF yield
with 100% HMF conversion was observed at 180 ℃ after 2 hr under
10 bar H2.5a Yang et al. synthesized Ni/Co3O4 catalysts, and the yield
of DMF was up to 76% at 130 °C and 1.0 MPa.5b
energy density (30 MJL-1),
a high research octane number
(RON=119), and a low volatility (boiling point= 92-94 ℃), which are
close to the values of gasoline (34 MJL-1, RON=89-96, and boiling
point= 96.3 ℃, respectively).2 Besides, a low solubility in water (2.3
gL-1) is superior to miscible ethanol, which makes DMF more
suitable than ethanol as the blend fuel.3 These excellent properties
make DMF a more appropriate, ideal, and promising biomass-
derived liquid biofuel for transportation.
Although excellent DMF yields (as high as 98%) were gained by
some studies, most of these catalytic systems required strict
reaction conditions, including a high reaction temperature (200-300
℃),5c, 5d a high hydrogen pressure (20-40 bar),5c-e, 5g, 5h, 5l or corrosive
hydrochloric acid (HCl or H2SO4),3 which is unfavorable for the
practical production of DMF. Thus, in this study, we prepared
supported Pt-Co bimetallic catalysts with high activity and high
selectivity under mild conditions. In addition, the role of interface
between bimetallic NPs (e.g., Pt-Co) and substrates on catalytic
performance was not explored. There is no study on the synergistic
effect between metal NPs and substrates on the hydrogenolysis of
HMF and the mechanism is not clear as to how they work
collectively to improve the activity and selectivity to DMF. It is vital
to investigate the synergistic effect systematically in order to
further optimize the bimetallic catalysts and improve the yield of
DMF.
It is a promising and reliable approach to obtain DMF from the
hydrogenolysis of HMF. HMF, containing an aldehyde group, a
hydroxyl group, and
a furan ring, is very reactive, and its
hydrogenation and hydrogenolysis products are kind of
complicated, as shown in Scheme 1.4 Therefore, how to ensure the
hydrogenation priorities of an aldehyde group and a hydroxyl
In this study, supported Pt-Co bimetallic NPs were prepared and
studied for HMF hydrogenolysis. First, Pt NPs was deposited on
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 2 of 10
View Article Online
DOI: 10.1039/C8GC00716K
ARTICLE
Journal Name
Scheme. 1 Proposed reaction pathways for the conversion of HMF, and the desired product is encircled. Compounds:
5-(hydroxymethyl)furfural (HMF), 2,5-dimethylfuran (DMF), 2,5-bis(hydroxymethyl)furan (DHMF), 5-mmethyl furfural
(5-MF),
5-methyl
furfuryl
alcohol
(MFA),
5-bis(hydroxymethyl)tetrahydrofuran
(DHTHF),
5,5-
(oxybis(methylene))bis(2-methylfuran) (OMBM), and tetrahydro-5-methyl 2-furammethanol (MTHFA).
different substrates (i.e., multi-walled carbon nanotubes reduction (H2-TPR) analysis. The ALD process was performed in a
(MWCNTs), silica gel particles, and γ-Al2O3 NPs) using atomic layer fluidized bed reactor, as reported in detail previously.6 All catalysts
deposition (ALD). After screening examination in HMF were directly used in the following catalytic experiments without
hydrogenolysis reactions, MWCNTs were chosen as the support to any further treatment. The details of Pt ALD, Co ALD, and Al2O3 ALD
prepare Pt-Co bimetallic catalysts, since Pt/MWCNTs showed an are described in the Electronic Supplementary Information (ESI).
excellent catalytic performance. Then, 10 and 25 cycles of Co were
2.2. Catalyst characterization
deposited on Pt/MWCNTs via ALD to obtain different mass ratios of
Pt, Co, and Al mass fractions of all catalysts were analyzed by
Pt to Co for Pt-Co/MWCNTs bimetallic catalysts (Fig. 1, Path A), and
inductively coupled plasma atomic emission spectroscopy (ICP-AES).
these two Pt-Co bimetallic catalysts were used in hydrogenolysis of
The Brunauer–Emmett–Teller (BET) surface area and pore size
HMF to DMF with different reaction temperatures to further
distribution of substrates and catalysts were measured by
improve the yield of DMF. In order to investigate the synergetic
Quantachrome Autosorb-1. The Pt and Pt-Co NPs were visualized
effect among Pt, Co, and MWCNTs on the catalytic performance of
with transmission electron microscopy/energy dispersive
Pt-Co/MWCNTs bimetallic catalysts, Al2O3 ALD was applied for the
spectroscopy (TEM/EDS) and high-resolution TEM (HRTEM). The
preparation of Pt-Co/Al2O3/MWCNTs catalysts to separate Pt-Co
reduction properties of the MWCNTs supported Pt, Pt-Co, Co, and
from MWCNTs, as schematically shown in Fig. 1 (Path B). In
Pt-Co/Al2O3 catalysts were analyzed by H2-TPR. CO-chemisorption
addition, a mixture of Pt/MWCNTs and Co/MWCNTs monometallic
was used to determine the Pt dispersion of Pt monometallic
catalysts was prepared to avoid the contact between Pt NPs and Co
catalysts. X-ray photoelectron spectroscopy (XPS) analysis was
NPs (Fig. 1, Path C), and it was also used in HMF hydrogenolysis
performed to verify the bimetallic structure of Pt-Co catalysts. The
reactions to investigate the synergetic effect between Pt and Co
details of characterization are described in ESI.
NPs through comparing with Pt-Co/MWCNTs bimetallic catalysts.
2.3. Catalyst performance
Hydrogenolysis of HMF was performed in a stainless steel Parr
2. Experimental section
reactor (50 mL) with an external temperature and stirring
controller, as described in detail elsewhere.7 Briefly, 0.5 g HMF (~
2.1. Catalyst preparation
0.4 mmol) and 30 mL 1-butanol were added into the reactor, and
All MWCNTs supported catalysts were synthesized as presented in
then certain amount of catalyst was added. Due to the different
Fig. 1. MWCNTs supported Pt and Pt-Co catalysts were obtained via
contents of Pt on the catalyst supports, the mass of catalysts added
path A. The Pt-Co bimetallic NPs were isolated from MWCNTs via
in the reactor was different in order to ensure an identical Pt
coating 5 cycles of Al2O3 ALD on MWCNTs (path B), and a mixture of
loading (4.35 mg) added in the reaction. The reactor was first
Pt/MWCNTs and Co/MWCNTs catalysts was prepared, as shown in
pressurized to 10 bar with H2 and the valve was opened to air to
path C. x, y and z are the number of Pt, Co, and Al2O3 ALD cycles,
depressurize the reactor to atmospheric pressure. This process was
respectively. 1 cycle of Pt ALD was also deposited on silica gel and
repeated 10 times to make sure 99.99% of the air in the reactor was
porous γ-Al2O3 particles to obtain Pt/SiO2 and Pt/γ-Al2O3 catalysts.
replaced by H2. Then the reactor was pressurized with H2 at 10 bar
In addition, 15 cycles of Co ALD were applied on MWCNTs directly
and the temperature increased to certain temperature (140, 160 or
to get 15c-Co/MWCNTs for hydrogen temperature programmed
180 °C). The volume of
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 3 of 10
View Article Online
DOI: 10.1039/C8GC00716K
Journal Name
ARTICLE
Fig. 1 Scheme for the synthesis of catalysts (x, y and z are the number of Pt, Co, and Al2O3 ALD cycles, respectively, and the
order of x, y, and z represents the order of ALD deposition).
Table 1 Properties of the MWCNTs and MWCNTs supported Pt and Pt-Co catalysts (Path A and Path C).
Sample
BET surface area, m2/g
144
Pore volume, cm3/g a
2.1
Pt, wt.% b
0
Co, wt.% b
0
Average particle size, nm c
-
MWCNTs
3c-Pt/MWCNTs
133
130
1.6
1.4
8.7
8.5
0
1.5
1.6
10c-Co/3c-
2.4
Pt/MWCNTs
25c-Co/3c-
125
140
1.3
1.9
8.2
0
6.3
3.5
1.6
1.3
Pt/MWCNTs
15c-Co/MWCNTs
a Pore volume was obtained from the adsorption curve of the N2 isotherm. b Pt and Co loadings were obtained from ICP-AES. c The average
metal particle size was obtained from TEM images.
H2 was around 20 mL, and there was 8.2 mmol H2 in the reactor areas of MWCNTs, silica gel, and γ-Al2O3 were 144, 272, and 91
initially based on ideal gas law. The slurry was stirred at a rate of m2/g, respectively. The pore volume of MWCNTs was the highest
500 rpm to eliminate the external diffusion limitations. A small and reached 2.1 cm3/g. As shown in Fig. S1, the structure of all
volume of sample (1 mL) was periodically collected and analyzed by substrates were mesoporous. MWCNTs showed two broad peaks
using a gas chromatography with flame ionization detector system between 2 to 100 nm, and silica gel presented a sharp peak
(GC-FID, Agilent 6890) according to internal standard (toluene). centered at 15 nm with a narrow pore size distribution. The size of
After reaction, the catalysts were washed with 1-butanol and most pores of γ-Al2O3 was between 5 to 30 nm. Based on ICP-AES
collected by centrifugation for cycling tests.
results, Pt loadings on silica gel and γ-Al2O3 were 4.0 and 2.4 wt.%,
respectively. The average Pt particle size and Pt dispersion on γ-
Al2O3 were 1.3 nm and 83% through CO-chemisorption analysis,
respectively, and the Pt dispersion on silica gel was lower than that
on γ-Al2O3, as shown in Table S1. These results were consistent with
our previous report.7
3. Results and discussion
3.1. Catalyst characterization
Some physical-chemical properties of different substrates
(MWCNTs, silica gel, and γ-Al2O3) are shown in Table 1 and S1. The
The physical-chemical properties of the MWCNTs and MWCNTs-
supported catalysts are summarized in Table 1. After Pt and Co ALD,
BET
surface
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 4 of 10
View Article Online
DOI: 10.1039/C8GC00716K
ARTICLE
Journal Name
size of Co was smaller and it was 1.3 nm. To verify the structure of
Pt-Co bimetallic NPs, HRTEM was carried out. As presented in Fig. 2,
some metal NPs were very close and formed “twinned” Pt-Co NPs
(red circle). The crystal fringes are clearly displayed (the inset from
Fig. 2) and the d-spacing was determined to be 0.230 nm and 0.206
nm, corresponding to the (111) planes of the face centered cubic
(fcc) Pt crystals and the (111) planes of Co crystals, respectively.8
Thus, the formation of Pt-Co bimetallic NPs was confirmed and the
“twinned” Pt-Co NPs were defined as bimetallic NPs in this study.
Pt, Co, and Al contents in the Al2O3 coated Pt and Pt-Co catalysts
are summarized in Table 2. After 5 cycles of Al2O3 ALD and 2 cycles
of Pt ALD, the Al and Pt loadings in all catalysts were similar and
they were about 4.7 and 5.0 wt.%, respectively. It is noted that after
Al2O3 ALD coating, the Co content increased fast on 5c-
Al2O3/MWCNTs with the increasing number of Co ALD cycles in
comparison to that on MWCNTs, as shown in Fig. S4, which could
be due to the fact that Co preferably grew on Al2O3 surface during
the ALD process. According to previous report, the growth rate of
Al2O3 film was 0.10-0.15 nm/cycle via ALD.9 Thus, the Al2O3 film
thickness, after five cycles of Al2O3 ALD, was ~0.6 nm in this work.
Fig. 2 HRTEM image of 25c-Co/3c-Pt/MWCNTs catalyst.
Based on theoretical calculation,
1 g of 5c-Al2O3/MWCNTs
the BET surface areas and total pore volume of all catalysts contained about 0.089 g Al2O3 (8.9 wt.%, and density of 3.95
decreased slightly as compared to those of MWCNTs, but the pore g/cm3), and the thickness of Al2O3 should be around 0.16 nm
size distribution of all catalysts almost did not change (Fig. S2). (0.089/3.95/140*1000 nm), if Al2O3 formed as a uniform film on the
Thus, the structure of catalysts should not affect the catalytic MWCNTs (~ 140 m2/g). This number was much lower than 0.6 nm,
performance of MWCNTs supported catalysts. The ICP−AES results which indicates that the Al2O3 film covered on MWNCTs was not
in Table 1 showed the Pt and Co contents on MWCNTs. After 3 uniform. This could be attributed to the fact that Al2O3 only grew on
cycles of Pt ALD, the Pt content was 8.7 wt.% for the 3c- specific defective sites rather than forming a uniform film on the
Pt/MWCNTs sample. The Pt content decreased slightly with the MWCNTs through ALD, since the surface of MWCNTs lacked
increase of the number of Co ALD on the 3c-Pt/MWCNTs sample. functional groups (e.g., hydroxyl groups) and the structure was
The TEM images of MWCNTs supported Pt and Pt-Co catalysts, near-perfect.
along with their particle size distribution were shown in Fig. S3. The previously.10
NPs were highly dispersed on the surface of MWCNTs for all
A
similar phenomenon has been observed
As shown in Table 2, the BET surface areas of all Al2O3 coated
samples decreased from ~140 to ~95 m2/g, which could be
attributed to the fact that some pores were blocked by Al2O3.
However, the Pt and Pt-Co NPs were still highly dispersed on the
MWCNTs, and the
catalysts. The average size of the Pt NPs were 1.5 nm for the 3c-
Pt/MWCNTs catalyst (Table 1). Compared with Pt NPs, the size of
Pt-Co NPs was slightly larger and their particle size distribution was
broader (Fig. S3). Compared to that of Pt NPs, the average particle
Table 2 Pt, Co, and Al loading of Al2O3-coated catalysts (Path B).
Sample
BET surface area, m2/g
95
Pt, wt.%a
5.5
Co, wt.%a
0
Al, wt.%a
4.8
Average particle size, nmb
1.3
2c-Pt/5c-
Al2O3/MWCNTs
3c-Co/2c-Pt/5c-
Al2O3/MWCNTs
92
5.3
3.4
4.7
1.5
5c-Co/2c-Pt/5c-
Al2O3/MWCNTs
98
94
5.2
5.2
5.5
6.4
4.6
4.5
1.7
1.7
8c-Co/2c-Pt/5c-
Al2O3/MWCNTs
a Pt, Co and Al loadings were obtained from ICP-AES. b The average metal particle size was obtained from TEM images.
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 5 of 10
View Article Online
DOI: 10.1039/C8GC00716K
Journal Name
ARTICLE
samples (Fig. S6). HRTEM image of the 5c-Co/2c-Pt/5c-
Al2O3/MWCNTs catalyst is shown in Fig. S7, and some metal
“twinned” NPs were formed, which were similar to those in Fig. 2.
TPR and XPS analysis were performed to further verify the
bimetallic structure of Pt-Co catalysts.
Fig. 3 shows the reducibility of MWCNTs supported catalysts
based on TPR analysis. There was almost no H2 consumption peak
observed for the 3c-Pt/MWCNTs and 2c-Pt/5c-Al2O3/MWCNTs
catalysts, which indicates that most of Pt NPs kept metallic state in
the air after ALD process. This was consistent with previous
publication.11 While a peak at around 422 ºC appeared in Fig. 3c,
which could be attributed to the reduction of CoOx. It implies that
Co NPs were partly oxidized in the air after Co ALD for the 15c-
Co/MWCNTs sample. The TPR profile of 25c-Co/3c-Pt/MWCNTs
showed two intense peaks observed at 298 and 475 ºC (Fig. 3d),
which were attributed to the reduction of CoOx that was in contact
with Pt and the isolated CoOx, respectively.12 Affected by hydrogen
spillover, the reduction of CoOx shifted to lower temperatures, as
the reduced Pt atoms were able to dissociate hydrogen molecules
into hydrogen atoms, resulting in the accelerated reduction rate of
CoOx.13 Thus, it indicates the electron transfer between Pt and Co in
e
d
c
b
a
100
300
500
700
900
Temperature, ºC
Fig. 3 TPR profiles of different catalysts: (a) 3c-Pt/MWCNTs, (b) 2c- Pt-Co bimetallic NPs. This H2-TPR analysis further proved the
Pt/5c-Al2O3/MWCNTs, (c) 15c-Co/MWCNTs, (d) 25c-Co/3c- formation of Pt-Co “twinned” NPs, which was consistent with the
Pt/MWCNTs, and (e) 5c-Co/2c-Pt/5c- Al2O3/MWCNTs.
HRTEM results. For the 5c-Co/2c-Pt/5c-Al2O3/MWCNTs catalyst,
two-stage reduction also showed at 388 ºC and 480 ºC. Compared
with that of 25c-Co/3c-Pt/MWCNTs, the reduction temperatures of
5c-Co/2c-Pt/5c-Al2O3/MWCNTs increased (Fig. 3e), which implied
that Pt-Co NPs were deposited on Al2O3 instead of MWCNTs and
average particle sizes of Pt and Pt-Co NPs were ~1.3 and ~1.6 nm,
respectively, as shown in Fig. S5. They were close to the average
particle sizes of Pt and Pt-Co NPs on MWCNTs, which implied that
the decrease of surface area did not affect the Pt dispersion. In
order to verify this, the Pt dispersion of the 3c-Pt/MWCNTs and 2c-
Pt/5c-Al2O3/MWCNTs catalysts was measured via CO-
chemisorption, and they were 77.5% and 79.3%, respectively, which
were similar and consistent with TEM analysis. In addition, since the
Al2O3 film was ultrathin, no Al2O3 films were observed in the TEM
images (Fig. S5), although the EDS spectra verified the existence of
Al2O3 on the
there was
a stronger interaction between Co and Al2O3 in
comparison with that between Co and MWCNTs.
XPS spectra of Pt, Co, and Pt-Co catalysts were shown in Figs. 4,
S8, and S9. Fig. 4 presents that the Pt 4f7/2 and 4f5/2 binding energies
were 71.1 and 74.5 eV on the 3c-Pt/MWCNTs catalyst, respectively,
which were assigned to zero-valence Pt0. After depositing Co, the
positions of Pt 4f7/2 and 4f5/2 (Pt0) positively shifted to 71.4 and 74.7
eV for the 25c-Co/3c-Pt/MWCNTs catalyst, respectively. It has been
reported that the binding energy is strongly correlated with the
adsorption/desorption capability of reaction species on the catalyst
surface, and such a tendency indicates that the d-band center of Pt-
Co bimetallic nanostructure downshifted in comparison to that of Pt
monometallic structure,14 implying a charge transfer from Co to Pt.
(a)
Pt°
PtᴵᴵO
As shown in Fig. S9a, the Pt 4f binding energy of the 5c-Co/2c-
Pt/5c-Al2O3/MWCNTs catalyst also positively shifted compared to
that of the 2c-Pt/5c-Al2O3/MWCNTs catalyst, which verified the
formation of Pt-Co bimetallic structure as well. In addition,
compared to the catalysts without Al2O3 coating, the intensity of Pt
4f binding energy was stronger, though Pt loading in the catalysts
without Al2O3 coating was higher, which could be attributed to the
overlap of Al 2p and Pt 4f peaks. The high resolution XPS spectra of
Co 2p for the catalysts are shown in Fig. S9b. The positions of Co
2p3/2 are 779.5, 777.9, and 780.4 eV for 15c-Co/MWCNTs, 25c-
Co/3c-Pt/MWCNTs,
(b)
(e)
81
79
77
75
73
71
69
Binding Energy (eV)
Fig. 4. XPS spectra of Pt 4f for (a) 3c-Pt/MWCNTs and (b) 25c-Co/3c-
Pt/MWCNTs.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 6 of 10
View Article Online
DOI: 10.1039/C8GC00716K
ARTICLE
Journal Name
Table 3 Hydrogenolysis of HMF to DMF over Pt monometallic catalysts with different supports. a
Selectivity distribution (%)
Temperature/time
Entry
Catalyst
Conversion (%)
DMF
-
DHMF
MFA
OMBM
-
Others b
(ºC/hr)
1
2
3
4
5
Blank c
MWCNTs d
3c-Pt/MWCNTs
Pt/SiO2
180/10
180/8
180/8
180/8
180/8
0
-
-
-
-
-
-
-
7.3
17.4
59.6
8.3
-
trace
28.1
-
-
100
68.6
84.6
-
-
10.7
91.7
87.2
Pt/γ-Al2O3
12.8
trace
trace
a Reaction conditions: 0.5 g HMF, 30 mL 1-butanol, 4.35 mg Pt, and 10 bar H2. b Other byproducts include 5-MF, DHTHF, and MTHFA. c No
catalyst was added in the reaction. d 50 mg MWCNTs was added in the reaction.
Table 4 Hydrogenolysis of HMF to DMF over MWCNTs-supported Pt-Co bimetallic catalysts. a
Selectivity distribution (%)
Temperature/time Conversion
Entry
Catalyst
(
º
C/hr)
(%)
DMF
69.0
84.2
92.3
77.2
DHMF
-
MFA
OMBM
20.2
12.3
5.7
Others a
10.0
3.2
1
2
3
4
10c-Co/3c-Pt/MWCNTs
25c-Co/3c-Pt/MWCNTs
25c-Co/3c-Pt/MWCNTs
25c-Co/3c-Pt/MWCNTs
180/8
180/8
160/8
140/10
100
100
100
79.3
-
trace
trace
10.1
-
-
2.0
trace
4.1
7.1
3c-Co/2c-Pt/5c-
Al2O3/MWCNTs
5
6
7
8
160/8
160/8
160/8
160/8
41.7
56.9
39.8
100
14.2
13.5
13.1
65.3
trace
-
-
-
-
7.2
10.2
9.3
78.6
76.3
77.6
12.6
5c-Co/2c-Pt/5c-
-
trace
-
Al2O3/MWCNTs
8c-Co/2c-Pt/5c-
Al2O3/MWCNTs
A mixture of 3c-Pt/MWCNTs
and 15c-Co/MWCNTs b
22.1
a Reaction conditions: 0.5 g HMF, 30 mL 1-butanol, 4.35 mg Pt, and 10 bar H2. a Other byproducts include 5-MF, DHTHF, and
MTHFA. b 53 mg 3c-Pt/MWCNTs and 95 mg 15c-Co/MWCNTs were used in hydrogenolysis of HMF reaction, which were the same
as the Pt and Co masses of 25c-Co/3c-Pt/MWCNTs catalyst, respectively.
and 5c-Co/2c-Pt/5c-Al2O3/MWCNTs, respectively. It implies that 5,5-(oxybis(methylene))bis(2-methylfuran) (OMBM) was 59.6% and
both Co0 and CoIIO existed in all catalysts, since the standard Co0 28.1%, respectively (entry 3), indicating the effectiveness of
2p3/2 and CoIIO 2p3/2 binding energy are 778.2 eV, and 780.0 eV, Pt/MWCNTs for selective hydrogenation of the formyl group in
respectively.15 This is consistent with the TPR results.
HMF. In contrast, the selectivity to DMF was very low over both
Pt/SiO2 and Pt/γ-Al2O3 catalysts (entries 4 and 5). It demonstrates
that MWCNTs played an important role in enhancing the activity
and selectivity to DMF for Pt catalysts in the HMF hydrogenolysis
reaction. It can be due to a transfer of the π-electrons from the
graphene layer of MWCNTs to the Pt NPs.16 In this way the charge
density on the Pt increases, thus improving the probability for the
C=O bond activation.17 MWCNTs are the material with excellent
electrical
3.2. Catalyst performance
The performance of all Pt catalysts supported on different
substrates for the hydrogenolysis of HMF to DMF reactions at 180
ºC and 10 bar H2 is presented in Table 3. No reaction occurred
without catalyst (entry 1), and the conversion of HMF over
MWCNTs was very low (entry 2). When using 3c-Pt/MWCNTs as a
catalyst, the conversion of HMF was 100% within 6 hr (Fig. S10a)
and,
the
yield
of
DMF
and
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 7 of 10
View Article Online
DOI: 10.1039/C8GC00716K
Journal Name
ARTICLE
Fig.
5 Adsorption models of reactant on different catalysts: (a) yc-Co/xc-Pt/MWCNTs, (b) yc-Co/xc-Pt/zc-
Al2O3/MWCNTs, and (c) a mixture of xc-Pt/MWCNTs and yc-Co/MWCNTs.
conductivity, but silica gel and γ-Al2O3 are semiconductor and out at different temperatures over the 25c-Co/3c-Pt/MWCNTs
insulated substrate, respectively. Hence, they cannot enhance the catalyst (entries 2-4 in Table 4). The highest yield of DMF (92.3%)
charge density of Pt NPs, and thereby MWCNTs were chosen as the achieved at 160
ºC (entry 3 in Table 4). With the reaction
support to prepare Pt-Co bimetallic catalysts. temperature decreasing, the selectivity to OMBM decreased,
In search of the desired catalyst, first the reaction was conducted indicating that the reaction of OMBM production occurred easier at
over 10c-Co/3c-Pt/MWCNTs and 25c-Co/3c-Pt/MWCNTs at 180 °C. high temperature. Though the selectivity to OMBM was lower at
Compared to the selectivity to DMF over the 3c-Pt/MWCNTs 140 ºC than that at 160 ºC, the selectivity to DMF was not high,
catalyst (59.6%, entry 3 in Table 3), with the increase of Co loading since HMF did not react completely (Fig. S11) and other byproducts
in the bimetallic catalysts, the selectivity to DMF increased greatly formed (entry 4, Table 4), especially intermediate byproducts. Thus,
(entries 1 and 2 in Table 4), and 84.2% yield of DMF was obtained high yield of DMF (>90%) was obtained over 25c-Co/3c-Pt/MWCNTs
over the 25c-Co/3c-Pt/MWCNTs catalyst, indicating the presence of under mild conditions.
Co can promote the selectivity of HMF hydrogenolysis to DMF and
The product distribution for the 25c-Co/3c-Pt/MWCNTs catalyst
hinder the production of OMBM. The increase of selectivity to DMF at different reaction temperatures is shown in Fig. S12. DHMF and
could be due to the fact that the deposition of Co on Pt NPs MFA were the main intermediates during the HMF hydrogenolysis
modified the electronic structure of Pt atoms, and destabilized the reaction, and thereby the proposed reaction pathway is shown in
adsorption of the C=C bond.18 Thereby, the selectivity to C=O Fig. S13. The selectivity to main byproduct OMBM also increased
improved greatly. The proposed hypothesis and adsorption models along with reaction time. At first HMF was hydrogenated to DHMF,
will be discussed later.
which then underwent dehydroxylation and hydrogenation to DMF.
It is interesting to note that the 10c-Co/3c-Pt/MWCNTs catalyst Since Pt is a good catalyst for hydrogenation of C=O bond, the first
showed the highest activity, followed by the 3c-Pt/MWCNTs and step should be hydrogenation of HMF to DHMF. Then, since the
25c-Co/3c-Pt/MWCNTs samples, since the conversion of HMF over barrier of dehydroxylation process is low in comparison to that of
10c-Co/3c-Pt/MWCNTs reached 100% in the shortest time (Fig. ring opening,5m dehydroxylation process occurred over the Pt-
S10a). As reported in a previous work, the presence of Co could Co/MWCNTs bimetallic catalysts. In addition, CoOx and some other
improve the activity of Pt-Co catalysts due to the synergistic effect metal oxides could enhance the activation of C-O bonds, which has
of Pt-Co bimetallic NPs.18 However, compared to the 3c- been reported previously.5b, 19 Thus, in this study, the existence of
Pt/MWCNTs and 10c-Co/3c-Pt/MWCNTs catalysts, the low activity CoOx would effectively break the C-O bond during dehydroxylation
of 25c-Co/3c-Pt/MWCNTs should be attributed to the fact that too reaction, and it would accelerate the dehydroxylation process and
many Co NPs attached to Pt NPs and it led to steric hindrance. It thus improve the selectivity of DMF partly.
would hinder the adsorption of reactant onto the 25c-Co/3c-
In order to verify the proposed synergistic effect among Pt, Co,
Pt/MWCNTs catalyst due to lack of enough space and thereby affect and MWCNTs, an ultra-thin Al2O3 film (~0.6 nm thick) deposited by
the catalytic activity of the catalyst. But the moderate activity is 5 cycles of Al2O3 ALD was used to modify the interfaces of catalysts.
good for improving the selectivity to DMF, since OMBM is relatively All Al2O3 coated catalysts were used for the reaction of HMF
easy to be produced over catalysts with very high activity. It is also hydrogenolysis and the results are presented in Table 4 (entries 5-
noted that the selectivity to DMF kept the same between 8 hr and 7). After depositing Co on 2c-Pt/5c-Al2O3/MWCNTs, all Pt-Co/5c-
10 hr of reaction time over all three catalysts (Fig. S10b), indicating Al2O3/MWCNTs catalysts presented a much lower selectivity to
that there was no more intermediate byproducts (e.g., DHMF, 5- DMF than the 25c-Co/3c-Pt/MWCNTs catalyst did. A mixture of 3c-
MF, and MFA) produced in the solution after 8 hr. In addition, DMF Pt/MWCNTs and 15c-Co/MWCNTs monometallic catalysts was also
was stable in this catalytic system and further hydrogenation of used in the hydrogenolysis of HMF reaction to investigate the
furan ring did not occur, which is crucial in the large-scale synergistic effect between Pt and Co NPs. As shown in Table 4
production of DMF in industry. In order to obtain a higher (entry 8), the HMF conversion and selectivity to DMF were 100%
selectivity to DMF, the reaction of HMF hydrogenolysis was carried and 65.3%, respectively. Compared with that over 25c-Co/3c-
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 8 of 10
View Article Online
DOI: 10.1039/C8GC00716K
ARTICLE
Journal Name
Pt/MWCNTs, the selectivity to DMF was lower over the mixture the leaching of Pt-Co bimetallic NPs in the catalyst. In contrast, the
catalyst, though the conversion of HMF reached 100% for both conversion and selectivity almost remained the same over 3c-
cases.
Pt/MWCNTs after 3 cycling tests, which is consistent with our
Fig. 5a shows the reactant adsorption mode over the Pt- previous cycling results of Pt/MWCNTs.7 Thus, the deactivation of
Co/MWCNTs catalysts. The contact between Pt and Co favored the the 25c-Co/3c-Pt/MWCNTs catalyst was mainly due to the leaching
electronic transfer from Co0 towards Pt0,20 which resulted in a of Co during the cycling reactions. It implied that the interaction
positive charge density in the Co atoms and a negative charge between Co and MWCNTs was relatively weak, compared to that of
density in the Pt atoms. The increased charge density on the metal Pt and MWCNTs. We are working on this topic and trying to
NPs would help Cδ+-Oδ- adsorb over a Pt δ--Co δ+ site and decrease improve the cycling stability of the Pt-Co/MWCNTs catalyst, and the
the probability of adsorption via the furan group, so the selectivity results will be reported later.
to DMF was higher over the 25c-Co/3c-Pt/MWCNTs catalyst. The
electron can be transferred from Co to Pt directly or through
Conclusions
MWCNTs. Since MWCNTs are a kind of material with excellent
electrical conductivity, they could enhance the electronic transfer
Highly dispersed Pt-Co bimetallic NPs were deposited on
and thereby enhance the charge density on metal NPs, as
MWCNTs for hydrogenolysis of HMF to DMF. Both the HRTEM
mentioned above, which would further promote the yield of DMF.
and H2-TPR analysis confirmed the formation of Pt-Co
Thus, the synergistic effect among Pt, Co, and MWCNTs improved
bimetallic NPs on MWCNTs. High yield of DMF (> 90%) was
the yield of DMF over the Pt-Co/MWCNTs bimetallic catalysts, as
achieved in hydrogenolysis of HMF over the 25c-Co/3c-
compared to the 3c-Pt/MWCNTs catalyst.
Pt/MWCNTs catalyst after 8 hr of reaction under mild
For the Al2O3-coated Pt-Co bimetallic catalysts, as
conditions. More importantly, through a series of experiments,
diagrammatically showed in Fig. 5b, since Pt-Co NPs were isolated
it was revealed that the synergistic effect between the Pt-Co
from MWCNTs via Al2O3 film, the charge transfer can only occur
bimetallic NPs and MWCNTs dramatically improved the yield
from Co to Pt directly, instead of through MWCNTs in the Pt-Co/5c-
of DMF in the HMF hydrogenolysis reactions, since MWCNTs
Al2O3/MWCNTs catalyst. Thus, the charge density on the bimetallic
with excellent electrical conductivity enhanced the electronic
Pt-Co NPs was not strong enough to hinder hydrogenation of furan
transfer and thereby enhanced the charge density on Pt-Co
ring, though the charge density increased slightly. Thus, the
NPs.
selectivity to DMF decreased greatly and more byproducts were
produced. The increased charge density on the bimetallic Pt-Co NPs
would promote the selectivity to C=O. In other words, the
Conflicts of interest
interaction between Pt-Co bimetallic NPs and MWCNTs worked as
The authors declare no competing financial interest.
an important role in the selectivity to DMF. It can be explained that
MWCNTs worked as a “bridge”, connected Pt with Co, and helped
electronic transfer from Co to Pt, which enhanced the charge
Acknowledgements
density on metal NPs (Pt and Co). At the same time, the conversion
This work was supported in part by the ACS Petroleum
of HMF over the Pt-Co/5c-Al2O3/MWCNTs catalysts decreased as
Research Fund. The authors thank Dr. Jessica TerBush and Dr.
well, compared with that over the 25c-Co/3c-Pt/MWCNTs catalysts.
Jingjing Qing at the Materials Research Center at Missouri
It could be attributed to the fact that the enhanced charge density
University of Science and Technology for TEM analysis. Use of
of bimetallic NPs would boost the adsorption and dissociation rate
of C=O over Pt δ--Co δ+ sites in the reaction of HMF hydrogenolysis.
the Center for Nanoscale Materials, Office of Science User
Facilities operated for the U.S. Department of Energy (DOE)
Thus, the synergistic effect between Pt-Co bimetallic NPs and
Office of Science by Argonne National Laboratory, was
MWCNTs promoted both the selectivity to DMF and activity of Pt-
supported by the U.S. DOE under Contract No. DE-AC02-
Co bimetallic catalysts greatly. For the mixture of 3c-Pt/MWCNTs
06CH11357.
and 15c-Co/MWCNTs catalysts, as diagrammatically presented in
Fig. 5c, the reactant was mainly adsorbed on the 3c-Pt/MWCNTs
instead of 15c-Co/MWCNTs, because the activity of the 3c-
Notes and references
Pt/MWCNTs catalyst was much higher than that of the 15c-
1
2
3
4
5
Y. Zu, P. Yang, J. Wang, X. Liu, J. Ren, G. Lu and Y. Wang, Applied
Catalysis B: Environmental, 2014, 146, 244.
B. Saha, C. M. Bohn and M. M. Abu-Omar, ChemSusChem, 2014,
7, 3095.
S. Nishimura, N. Ikeda and K. Ebitani, Catalysis Today, 2014, 232,
89.
Y. Nakagawa, M. Tamura and K. Tomishige, ACS Catalysis, 2013,
3, 2655.
(a) G. H. Wang, J. Hilgert, F. H. Richter, F. Wang, H. J. Bongard,
B. Spliethoff, C. Weidenthaler and F. Schüth, Nature Materials,
2014, 13, 293; (b) P. Yang, Q. Cui, Y. Zu, X. Liu, G. Lu and Y.
Wang, Catalysis Communications, 2015, 66, 55; (c) L. Hu, X.
Co/MWCNTs catalyst. Thus, this revealed that the interaction
between Pt and Co in the Pt-Co bimetallic NPs had played an
important role in the hydrogenolysis of HMF reaction and improved
the selectivity to DMF.
In order to evaluate the stability and lifetime of the catalysts, the
cycling experiments of HMF hydrogenolysis were carried out using
the 3c-Pt/MWCNTs and 25c-Co/3c-Pt/MWCNTs catalysts at 160 ºC.
The results are shown in Fig. S14. 3 cycling tests were applied for
25c-Co/3c-Pt/MWCNTs and a decrease in selectivity to DMF was
observed with an increase in the number of reaction cycles, though
the conversion kept 100% after 8 hours reaction time. It indicates
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 9 of 10
View Article Online
DOI: 10.1039/C8GC00716K
Journal Name
ARTICLE
Tang, J. Xu, Z. Wu, L. Lin and S. Liu, Industrial and Engineering
Chemistry Research, 2014, 53, 3056; (d) J. Zhang, L. Lin and S.
Liu, Energy and Fuels, 2012, 26, 4560; (e) J. Luo, L. Arroyo-
Ramírez, R. J. Gorte, D. Tzoulaki and D. G. Vlachos, AIChE
Journal, 2015, 61, 590; (f) D. Scholz, C. Aellig and I. Hermans,
ChemSusChem, 2014, 7, 268; (g) Y. B. Huang, M. Y. Chen, L.
Yan, Q. X. Guo and Y. Fu, ChemSusChem, 2014, 7, 1068; (h) J.
Jae, W. Zheng, R. F. Lobo and D. G. Vlachos, ChemSusChem,
2013, 6, 1158; (i) S. De, S. Dutta and B. Saha, ChemSusChem,
2012, 5, 1826;
(j) M. Chidambaram and A. T. Bell, Green
Chemistry, 2010, 12, 1253; (k) J. B. Binder and R. T. Raines,
Journal of the American Chemical Society, 2009, 131, 1979; (l)
J. Luo, J. D. Lee, H. Yun, C. Wang, M. Monai, C. B. Murray, P.
Fornasiero and R. J. Gorte, Applied Catalysis B: Environmental,
2016, 199, 439;
(m) J. Luo, H. Yun, A. V. Mironenko, K.
Goulas, J. D. Lee, M. Monai, C. Wang, V. Vorotnikov, C. B.
Murray and D. G. Vlachos, ACS Catalysis, 2016, 6, 4095.
(a) Z. Shang, R. L. Patel, B. W. Evanko and X. Liang, Chemical
Communications, 2013, 49, 10067; (b) X. Wang, H. Zhao, T.
Wu, Y. Liu and X. Liang, Catalysis Letters, 2016, 146, 2606.
X. Wang, W. Hu, B. Deng and X. Liang, Journal of Nanoparticle
Research, 2017, 19, Article No. 153.
(a) L. Xin, F. Yang, S. Rasouli, Y. Qiu, Z.-F. Li, A. Uzunoglu, C.-J.
Sun, Y. Liu, P. Ferreira and W. Li, ACS Catalysis, 2016, 6, 2642;
(b) Y. Su, Y. Zhu, H. Jiang, J. Shen, X. Yang, W. Zou, J. Chen and
C. Li, Nanoscale, 2014, 6, 15080.
6
7
8
9
M. Groner, F. Fabreguette, J. Elam and S. George, Chemistry of
Materials, 2004, 16, 639.
10 S. M. George, Chemical Reviews, 2010, 110, 111.
11 T. Aaltonen, M. Ritala, T. Sajavaara, J. Keinonen and M. Leskelä,
Chemistry of Materials, 2003, 15, 1924.
12 D. O. Silva, L. Luza, A. Gual, D. L. Baptista, F. Bernardi, M. J.
Zapata, J. Morais and J. Dupont, Nanoscale, 2014, 6, 9085.
13 (a) G. Jacobs, T. K. Das, P. M. Patterson, J. Li, L. Sanchez and B.
H. Davis, Applied Catalysis A: General, 2003, 247, 335; (b) S.
Vada, A. Hoff, E. Ådnanes, D. Schanke and A. Holmen, Topics in
Catalysis, 1995, 2, 155.
14 B. Y. Xia, H. B. Wu, N. Li, Y. Yan, X. W. D. Lou and X. Wang,
Angewandte Chemie International Edition, 2015, 127, 3868.
15 (a) N. McIntyre and M. Cook, Analytical Chemistry, 1975, 47,
2208;
(b)
J. Haber and L. Ungier, Journal of Electron
Spectroscopy and Related Phenomena, 1977, 12, 305.
16 H. Ma, L. Wang, L. Chen, C. Dong, W. Yu, T. Huang and Y. Qian,
Catalysis Communications, 2007, 8, 452.
17 M. L. Toebes, F. F. Prinsloo, J. H. Bitter, A. J. van Dillen and K. P.
de Jong, Journal of catalysis, 2003, 214, 78.
18 S. C. Tsang, N. Cailuo, W. Oduro, A. T. Kong, L. Clifton, K. K. Yu,
B. Thiebaut, J. Cookson and P. Bishop, ACS Nano, 2008, 2, 2547.
19 (a) S. Koso, N. Ueda, Y. Shinmi, K. Okumura, T. Kizuka and K.
Tomishige, Journal of Catalysis, 2009, 267, 89; (b) W. Xu, H.
Wang, X. Liu, J. Ren, Y. Wang and G. Lu, Chemical
Communications, 2011, 47, 3924; (c) S. Srivastava, G. Jadeja
and J. Parikh, Chinese Journal of Catalysis, 2017, 38, 699.
20 N. M. Bertero, A. F. Trasarti, B. Moraweck, A. Borgna and A. J.
Marchi, Applied Catalysis A: General, 2009, 358, 32.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Green Chemistry
Page 10 of 10
View Article Online
DOI: 10.1039/C8GC00716K