Hydrogenation of dimethyl malonate to 1,3-propanediol catalyzed by a Cu/SiO2 catalyst: The reaction network and the effect of Cu+/Cu0 on selectivity

Article abstract of DOI:10.1039/c6nj03960j

1,3-Propanediol, a vital monomer for the manufacture of commodity polytrimethylene-terephthalate, is commercially produced nowadays through either hydration of acrolein or hydroformylation of ethylene oxide. Herein, for the first time, we present the investigation of an alternative route for 1,3-propanediol production from vapor-phase catalytic hydrogenation of syngas-derived dimethyl malonate on a Cu/SiO₂ catalyst. The catalytic reaction network has been disclosed for the Cu/SiO₂ catalyst, and the reaction proceeds through sequential hydrogenation with methyl 3-hydroxypropionate as the primary product, which can be further converted into 1,3-propanediol or methyl propionate. Excessive hydrogenation of 1,3-propanediol or methyl propionate leads to the formation of n-propanol. Meanwhile, a small amount of dimethyl malonate cracks into methyl acetate. The structural and textural properties of Cu/SiO₂ catalysts with varied copper loadings were extensively characterized by X-ray diffraction, Fourier transform infrared spectroscopy, H₂-temperature programmed reduction, X-ray photoelectron spectroscopy, N₂ physisorption, CO chemisorption, N₂O titration, and transmission electron microscopy. A correlation of the areal activity to copper components suggests that metallic copper modified by the co-present Cu⁺ species may be the active site for the hydrogenation. The highest 1,3-propanediol selectivity was achieved on a catalyst with a maximum Cu⁺/(Cu⁰ + Cu⁺) ratio of 0.41. The revelation of catalytic networks and insights into the active species can provide guidance for future rational design of catalysts for regioselective hydrogenation of CO bonds in dimethyl malonate.

Full text of DOI:10.1039/c6nj03960j

View Article Online
View Journal
NJC
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: S. Zheng, K. Zhu,
W. Li and Y. Ji, New J. Chem., 2017, DOI: 10.1039/C6NJ03960J.
This is an Accepted Manuscript, which has been through the
Volume 40 Number
1 January 2016 Pages 1–846
Royal Society of Chemistry peer review process and has been
accepted for publication.
NJC
Accepted Manuscripts are published online shortly after
New Journal of Chemistry
www.rsc.org/njc
A journal for new directions in chemistry
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 1144-0546
PAPER
Jason B. Benedict et al.
The role of atropisomers on the photo-reactivity and fatigue of
diarylethene-based metal–organic frameworks
rsc.li/njc

Please do not adjust margins
New Journal of Chemistry
Page 1 of 14
View Article Online
DOI: 10.1039/C6NJ03960J
Journal Name
ARTICLE
Hydrogenation of Dimethyl Malonate to 1,3-Propanediol
Catalyzed by Cu/SiO2 Catalyst: Reaction Network and Effect of
Cu+/ Cu0 on Selectivity
Received 00th January 20xx,
Accepted 00th January 20xx
Sainan Zhenga, Kake Zhua, Wei Lia*, and Yang Jib
DOI: 10.1039/x0xx00000x
Abstract: 1,3-propanediol, a vital monomer for the manufacture of commodity poly trimethylene-terephthalate, is
commercially produced nowadays through either hydration of acrolein or hydroformylation of ethylene oxide. Herein, for
the first time, we present the investigation of an alternative route for 1,3-propanediol production from vapor-phase
catalytic hydrogenation of syngas-derived dimethyl malonate on Cu/SiO2 catalyst. The catalytic reaction network has been
disclosed for Cu/SiO2 catalyst, the reaction proceeds through sequential hydrogenation with methyl 3-hydroxypropionate
as the primary product, which can be further converted into 1,3-propanediol or methyl propionate. Excessive
hydrogenation of 1,3-propanediol or methyl propionate leads to the formation of n-propanol. Meanwhile, a small amount
of dimethyl malonate cracks into methyl acetate. The structural and textural properties of Cu/SiO2 catalysts with varied
copper loadings were extensively characterized by X-ray diffraction, Fourier transform infrared spectroscopy, H2-
temperature programmed reduction, X-ray photoelectron spectroscopy, N2 physisoprtion, CO chemisorption, N2O
titration, and transmission electron microscopy. A correlation of areal activity to copper components suggests that
metallic copper modified by the co-present Cu+ specie may be the active site for the hydrogenation. The highest 1,3-
propanediol selectivity was achieved on a catalyst with maxima Cu+/(Cu0+Cu+) ratio of 0.41. The revelation of catalytic
network and insight into the active species can provide guidance for future rational design of catalysts for regioselevtive
hydrogenation of C=O bonds in dimethyl malonate.
www.rsc.org/
Keywords: dimethyl malonate, 1,3-propanediol, vapor-phase hydrogenation, Cu0/Cu+ active sites, synergetic effect
between production and the global market will widen
progressively.12 Therefore, there is a driven force to cultivate
other ways to meet the increasing demand for 1,3-PDO
1. Introduction
Poly trimethylene-terephthalate (PTT) is an attractive polymer
which finds broad applications in fiber and textile industries
due to its better stretching and stretch-recovering
characteristics, lower dyeing temperature, and higher light
stability than other polyesters. In particular, its
biodegradability offers great competitive environmental and
economic advantages.1,2 One of the monomer for PTT, 1,3-
propanediol (1,3-PDO), is nowadays produced chiefly via two
routes: hydration of acrolein by the Degussa process and
hydroformylation of ethylene oxide by the Shell process.3
Besides, microbial production has recently been reported as an
alternative for 1,3-PDO synthesis.4-6 Moreover, hydrogenolysis
of glycerol also potentially contributes the commercial
productivity of 1,3-PDO.7-9 Recently, Ding et al. reported
another approach to synthesize 1,3-PDO involving
hydrogenation of diethyl malonate.10,11 Since the global market
for PTT production is predicted to rise continuously, the gap
production. Dimethyl malonate (DMM) hydrogenation to 1,3-
PDO is a possible alternative route for large scale production,
as DMM can potentially be generated from both syngas- and
petroleum-derived
products,
for
instance,
via
carboxymetylation of dihalo methane and carbonylation of
methyl chloroacetate.13,14 More importantly, DMM can be
synthesized via a Claisen condensation process involving
methyl carbonate and dimethyl carbonate, both of which are
commercially produced through low-cost syngas-based routes.
This approach for preparing DMM has been investigated in our
previous work.15 In this contribution, DMM hydrogenation to
1,3-PDO via a vapor-phase catalytic process over a Cu/SiO2
catalyst is tentatively explored.
Copper-based catalysts are optional for their remarkable role
in vapor-phase hydrogenation, such as the hydrogenation of
esters to alcohols and glycerol to propanediols.16-22 Cu based
catalysts are effective for the chemoselective hydrogenation of
C-O bonds, while fairly deficient for the hydrogenolysis of C-C
bonds.20 Previous studies for catalytic vapor-phase
hydrogenation show that silica-supported copper catalysts are
a.State Key Laboratory of Chemical Engineering, School of Chemical Engineering,
East China University of Science and Technology, 130 Meilong Road, Shanghai
200237, P. R. China. Email: liwei@ecust.edu.cn; Tel: +86-21-64253118.
b.Shanghai Pujing Chemical Industry Co. Ltd, 1305 Huajing Road, Shanghai 200231,
P. R. China.
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
New Journal of Chemistry
Page 2 of 14
View Article Online
DOI: 10.1039/C6NJ03960J
ARTICLE
Journal Name
active for di-esters to di-alcohols, like dimethyl oxalate (DMO)
to ethyl glycerol (EG),23,24 dimethyl maleate to 1,4-
butanediol24-26. Furthermore, the nature of active sites in
carbonyl hydrogenation has been intensively investigated. For
instance, in an earlier study, San et al.27 considered Cu0 species
to act as the only active centers in the hydrogenation process.
Moreover, the synergetic effect between Cu0 and the oxide
components has been proposed to improve the catalytic
performance toward the hydrogenation of methyl acetate
(MA),28 dimethyl oxalate (DMO),29,30 furfural,31,32 and CO2.33
Although the cooperation of Cu0 and Cu+ species in C=O
hydrogenation has been suggested, their respective
contribution to the catalytic activity remains debated. Inspired
by the known regioslectivity for Cu/SiO2 in catalytic
hydrogenation, we have adopted a homogeneous deposition
precipitation method using urea as precipitator to prepare
Cu/SiO2 catalyst for DMM hydrogenation to 1,3-PDO. Such a
protocol can effectively disperse Cu species on silica by
forming strong metal-support interactions, thereby
guaranteeing the co-existence of both Cu0 and Cu+ species in
the reduced catalysts.23,34 The purpose of this study is to
explore the catalytic hydrogenation of DMM to 1,3-PDO on
Cu/SiO2 to understand how the reaction proceeds and what
constitute the active sites. A reaction network for DMM
hydrogenation on Cu/SiO2 catalysts will be proposed and the
formation and nature of active sites will be explored. The
effects of copper loading will also be estimated by taking
resultant catalyst precursor was denoted as xCu/SiO2, where x
stands for the preset mass loading of copper. The catalyst was
crushed to sieve 40-60 mesh for the activity test in DMM
hydrogenation to 1, 3-PDO.
For comparison, a reference catalyst containing 20 wt. %
copper was prepared by incipient wetness of silica support
with an aqueous solution of Cu(NO3)2·3H2O, followed by drying
(393K, 12h), and calcination at 723 K for 4 h in flowing air. The
catalyst was denoted as 20Cu/SiO2-IM.
2.2 Characterization Techniques
The actual copper amount in the samples was determined by
elemental analysis using an inductively coupled plasma atomic
emission spectrometer (ICP-AES, Perkin Elmer Optima 7300
DV).
XRD measurements were carried out using a Rigaku C/max-
2500 diffractometer employing the graphite filtered Cu Kα
radiation (λ=1.5406 Å) with a scanning angle (2θ) range of 10-
80°. The average diameter of Cu particles was calculated on
the basis of the Cu0 (111) peak broadening according to
Scherrer equation (after subtracting the instrumental
broadening).
The Fourier-transform infrared (FT-IR) spectra of catalyst
precursors and CO chemisorption on reduced catalysts were
collected on a Nicolet 6700 spectrometer. Briefly, the samples
were finely grounded, dispersed in KBr, and pelletized. The
spectral resolution was 4 cm-1, and 64 scans were recorded for
each spectrum. For CO-chemisorption IR experiments, 30 mg
of catalyst was compressed into a self-supporting wafer and
advantage of
a
systematic characterization on the reduced in a flow of 10% H2-90%Ar (30 ml min-1) at 623 K for 4
h before evacuating for 1h to remove the chemisorbed surface
contaminants and cooling down to room temperature. Next,
the reduced sample was exposed to highly pure CO (99.999%)
for 30 min and evacuated for 30 min.
physicochemical properties of these catalysts by X-ray
diffraction, Fourier transform infrared spectroscopy, H2-
temperature programmed reduction, X-ray photoelectron
spectroscopy, N2 physisoprtion, CO chemisorption, N2O
titration, and transmission electron microscopy. The catalysis-
structure relationship will be proposed by correlating of
catalyst activity to distinct curious species.
X-ray photoelectron spectroscopy (XPS) and Auger electron
spectroscopy (XAES) measurements were performed on a
Quantum 2000 Scanning ESCA Microprobe instrument
(Physical Electronics) equipped with an Al Kα X-ray radiation
source (hν = 1486.6 eV). The reduced catalyst was carefully
collected and sealed under the protection of Ar after
reduction. Binding energies were calibrated using the C1s peak
at 284.8 eV as a reference with an uncertainty of ±0.2 eV.
2. Experimental
2.1 Catalyst Preparation
The
H2
temperature-programmed
reduction
(TPR)
All Cu/SiO2 catalysts were prepared by urea-assisted
deposition-precipitation (UDP) of metal nitrate salts in the
presence of colloidal silica, as described previously.35 Briefly,
25 wt.% colloidal silica (JA-25, Qingdao Haiyang Chemical Ltd.)
was dispersed in a given amount of aqueous solution
containing Cu(NO3)2·3H2O (AR, Shanghai Lingfeng Chemical
Reagent Co. Ltd.) and urea (A.R., Shanghai Lingfeng Chemical
Reagent Co. Ltd.) in a round-bottomed flask. The suspension
was then heated to 363 K under vigorously stirring, allowing a
steady and homogeneous increase of pH value by the
hydrolysis of urea which triggered the precipitation of a fine
copper species on silica. After reaching pH 6~7, the lump was
collected after hot filtration of the obtained precipitation,
which was washed thrice with deionized water, dried
overnight at 393 K, and calcined in air at 723 K for 4h. The
measurements were operated on a Micromeritics Autochem II
2920 instrument. Typically, catalyst of 100 mg was dried at
473K under argon (highly pure, 99.999%) (30 ml·min-1) for 1 h,
then cooled down to room temperature in pure argon. After
that, the catalyst was then heated in 30 ml·min-1 of 10%H2-
90% Ar at a ramp of 10 K·min-1 up to 973 K. The amount of H2
consumed was monitored by a thermal conductivity detector
(TCD). The specific surface area of metallic copper and copper
dispersion of the catalysts were determined by N2O titration
on a Micromeritics Autochem II 2920, following a method
reported before.36 Briefly, after pretreatment with He at 473 K
for 1 h, 100 mg of the catalysts was reduced under a flow of 10
vol. % H2/Ar (30 ml min-1), and the temperature was raised
from room temperature to 623 K with a ramp of 10 K min-1
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
New Journal of Chemistry
Page 3 of 14
Journal Name
View Article Online
DOI: 10.1039/C6NJ03960J
ARTICLE
then cooled to 323 K. The pure N2O (30 ml min-1) was charged
to the catalyst at 323 K for 0.5 h to oxidize surface Cu0 into
Cu2O. A subsequent reduction of surface Cu2O to Cu0 by H2
was conducted, and the temperature was raised to 773 K with
a ramp of 10 K min-1. The consumed amount of hydrogen was
valued by subtracting the amount of residual H2 calculated
from the areas shown in TCD signals from the total amount of
H2 pulsed. From that, the surface metallic copper areas of
catalysts could be calculated. In combination with the Cu
loading measurements from ICP, the metallic Cu dispersion can
be obtained as well.
♥
(A)
♥ CuO
♦ SiO2
♥
♦
♥
♥
♥
♥
♥
(g)
♥
(f)
(e)
(d)
(c)
(b)
(a)
N2 adsorption-desorption analysis was conducted at 77 K
employing a Micromeritics ASAP 2020 Analyzer instrument.
Pore size distribution was evaluated from the desorption
branches of the adsorption-desorption isotherms by the
Barrett-Joyner-Halenda (BJH) method, and the specific surface
area (SBET) was calculated using the Brunauer-Emmet-Teller
(BET) method.
10
20
30
40
50
60
70
80
2θ
/degree
•
(B)
• Cu0
♦ Cu2O
Transmission electron microscopy (TEM) micro-images and
High angle annular dark field scanning transmission electron
microscopy (HADDF-STEM) were obtained on a JOEL JEM-
2100F system electron microscope equipped with a field
emission electron gun with the operation voltage of 200 kv. To
prepare an appropriate sample for TEM observation, catalyst
powder was ultrasonically dispersed in absolute ethanol at
room temperature for 30 min and transferred onto nickel grid
by dipping.
•
•
♦
(g)
(f)
(e)
(d)
(c)
(b)
(a)
10
20
30
40
50
60
70
80
2.3 Evaluation of the Catalytic Performance
2θ
/degree
The catalytic performance was assessed in a continuous fixed-
bed microreactor. Samples (40-60 mesh, 2.0 g in all cases)
were packed into the tubular reactor with an inner diameter of
10 mm. The catalyst samples were pre-reduced in H2 at 623 K
for 4 h with a ramping rate of 2 K min-1, and cooled to the
Fig. 1. XRD patterns of Cu/SiO2 catalysts: (a) 10Cu/SiO2, (b) 15Cu/SiO2, (c)
20Cu/SiO2, (d) 25Cu/SiO2, (e) 30Cu/SiO2, (f) 40Cu/SiO2, (g) 20Cu/SiO2-IM.
==================================================================
could be observed in all the samples, indicating the formation
of highly dispersed amorphous copper species at low
loadings.29 Weak reflections at 35.4° and 38.7° from CuO
(tenorite) (JCPDS 05-0661) can be observed for 40Cu/SiO2,
showing copper oxide agglomerated at this loading. In striking
constant, the 20Cu/SiO2-IM consists of well-developed CuO, as
exhibited by the sharp diffractions in Fig. 1(A). The diffraction
lines of CuO for UDP-derived catalysts are less intense and
much broader that the 20Cu/SiO2-IM, indicating that UDP is
advantageous in promoting the metal dispersion in precursors.
The broad feature at 2θ of approximately 22° from amorphous
silica support. The enhanced dispersion of copper in the
catalyst precursors is reckoned as a result of formation of
copper phyllosilicate phase (JCPDS 00-003-0219), which has
been identified by the diffraction peaks at 2θ of 31.0, 34.8,
57.2, 63.3, and 71.2° (Supporting Information, Fig. S1). It also
should be noted that the peaks from copper phyllosilicate
diminished without notable development of the peaks from
copper oxide after calcination (Fig. S1). Consequently, a well
dispersed CuO on the calcined Cu/SiO2 samples can emerge as
a result of copper phyllosilicate thermal decomposition. The
phenomenon could arise from the preparation method, as
homogeneous precipitation favors complexing of Cu2+ with the
preset reaction temperature. Afterwards, 15 wt.
%
DMM/methanol solution and H2 were fed into the reactor.
Typical test conditions were set to reaction temperature of
463 K, a total pressure of 5.0 MPa, and H2/DMM molar ratio of
200, and the weight liquid hourly space velocity of DMM
(LHSVDMM) of 0.4 g gcat-1 h-1. The products were condensed with
a cold trap, collected and then analyzed using a Huaai GC 9560
with DB-200 capillary column and a flame ionization detector
(FID). The major byproducts include methanol, methyl acetate,
n-propanol,
methyl
propionate,
and
methyl
3-
hydroxypropionate.
3. Results
3.1 XRD patterns
The Cu/SiO2 catalysts with varied Cu loadings were prepared
following a previously established homogeneous precipitation
UDP method.35 Fig. 1(A) illustrates the XRD profiles for the
calcined xCu/SiO2 precursors prepared from UDP protocol and
a contrast 20Cu/SiO2-IM sample derived from an impregnation
method. The XRD patterns of all the UDP derived samples
exhibited an amorphous pattern except 40Cu/SiO2, no obvious
diffraction peak of bulky CuO or related cupreous substances
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
New Journal of Chemistry
Page 4 of 14
View Article Online
DOI: 10.1039/C6NJ03960J
ARTICLE
Journal Name
surface silanols by forming =SiOCuII monomers and their
subsequent polymerization leads to the formation of
phyllosilicate.24,35,37,38 Toupance et al.39 found that a fraction of
copper phyllosilicate tends to decompose during calcination at
723 K. Van der Grift et al.40 also revealed that a fraction of
copper phyllosilicate decomposes when calcined, bringing
about well dispersed CuO accompanied by the intact copper
phyllosilicate.
10 40
6 70
800
A
(g)
(f)
e
c
n
a
b
r
(e)
(d)
(c)
o
s
b
A
After reduction (Fig. 1(B)), the Cu/SiO2 catalysts demonstrate
an obvious diffraction peak at 2θ of 43.3° along with two weak
ones at 50.4 and 74.1°, which are characteristics of Cu (110)
(JCPDS 04-0836).29 The intensity of these diffraction lines grow
with Cu contents in the UDP-derived samples, and becomes
only visible at loadings higher 25 wt.%. Below this loading, no
obvious diffraction peak can be observed, indicating that the
Cu is well dispersed and does not form crystalline structure to
be detected by XRD technique. Apart from that, a broad and
weak peak at 36.4° which is ascribable to the Cu2O (111) plane
(JCPDS 05-0667),41 is presented in the high-loading reduced
catalysts. These findings show that the Cu0 and Cu2O coexist in
the functioning catalysts. Unlike samples derived from UDP
protocol, copper species aggregated to larger clusters (ca. 34.2
nm) after reduction in the 20Cu/SiO2-IM catalyst, due to its
weak interaction with the silica support and the aggregation of
metallic copper.42 In short, XRD results demonstrate that UDP-
derived sample are made up of well dispersed Cu species on
SiO2 support, owing to the strong interaction between the Cu2+
with Si-OH on silica surface that lead to the formation of a
precursor copper phyllosilicate. Partial reduction of copper
phyllosilicate in H2 atmosphere produces metallic Cu species
together with Cu2O in the silica matrix.
(b)
(a)
4000
3
500
3000
2500
2000
1500
1
1000
500
-
Wave N umbe rs/cm
(B)
0.16
0.12
0.08
0.04
0.00
5
10
15
20
25
30
35
40
45
Copper Loading / wt. %
Fig. 2. (A) FTIR spectra of calcined samples with different copper loadings. (B) The
I670/I800 intensity ratio. (a) 10Cu/SiO2, (b) 15Cu/SiO2, (c) 20Cu/SiO2, (d) 25Cu/SiO2,
(e) 30Cu/SiO2, (f) 40Cu/SiO2, (g) 20Cu/SiO2-IM.
3.2 Fourier-transform infrared spectroscopy (FTIR)
==================================================================
Fourier-transform infrared spectroscopy (FTIR) has been used
to detect the formed species in catalyst preparation, as
displayed in Fig. 2. FTIR is employed to probe copper
phyllosilicate (Cu2Si2O5(OH)2), because the fingerprint δOH
bands in the structure are clearly distinguishable.35,43 The
broad adsorption bands centered at approximately 3460 cm-1
in the spectra for calcined samples (Fig. 2(A)) are attributed to
the OH stretching vibration of water and surface silanols.44
Similarly, the bands around 1630 cm-1 are ascribed to the
bending mode of OH groups of adsorbed water.44 Bands at
approximately 1110 cm-1 are ascribed to stretching vibration of
Si-O bond; bands at 800 and 470 cm-1 are associated with
bending vibration modes of the Si-O bond in amorphous
silica.45 The δOH vibration at 670 cm-1 together with the νSi-O
shoulder peak at 1040 cm-1 suggests that the existence of
copper phyllosilicate35 in the calcined samples derived from
UDP method. In contrast, such features are not detected for
the 20Cu/SiO2-IM sample, showing that no copper
phyllosilicate is produced by impregnation method. Weak
vibration bands at 575, 500, and 460 cm-1 belonging to CuO
phase46,47 (Fig. 2(A)) have not been observed, as the intense
and broad band at 470 cm-1 from silica overlaps with them.
The relative amount of copper phyllosilicate in the calcined
Cu/SiO2 catalysts can be evaluated by I670/I80035 which is
defined as the ratio between the integrated intensity of the
δOH band at 670 cm-1 and the integrated intensity of the νSi-O
symmetric stretching band of SiO2 at 800 cm-1. The I670/I800
ratio gives semi-qualitative amount of copper phyllosilicate, as
the extinction coefficients of the corresponding IR bands are
not known. Fig. 2(B) demonstrates the relative amount of
copper phyllosilicate in calcined Cu/SiO2 samples as a function
of copper loading. The copper phyllosilicate content is found
to increase with increasing Cu loading for samples derived
from UDP method, whereas no copper phyllosilicate is
detected for impregnated sample 20Cu/SiO2-IM, which is in
line with XRD result.
The structural evolution during catalyst preparation for a
representative 20Cu/SiO2 is shown in Fig. S2. Copper
phyllosilicate is only detectable for as-prepared and calcined
samples, and H2 reduction has caused its complete
decomposition.
3.3 H2-TPR profiles of the calcined catalysts
H2-TPR has been applied to characterize reducibility of calcined
catalysts to obtain some surface species information. TPR
profiles (Fig. 3) of all the UDP-derived catalysts present a single
reduction peak centered at ca. 483 K. This reduction pattern
can be attributed to the presence of well dispersed Cu2+
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
New Journal of Chemistry
Page 5 of 14
Journal Name
View Article Online
DOI: 10.1039/C6NJ03960J
ARTICLE
species with small particle size (< 10.5 nm), according to S3. The Cu 2p3/2 peak at 935.7 eV with a shoulder on the high
previous assignment.48 The absence of reduction peaks below BE side, is characteristic of Cu2+ species34,56 in the calcined
473 K and 528-573 K suggests the lack of copper oxocations samples. The asymmetry of Cu 2p3/2 envelope allows the
(Cu-O-Cu)2+ 49,50 and bulk copper oxide,51-53 respectively. Unlike calcined sample peak to be deconvoluted into two
UDP-derived samples, there are three adjacent reduction contributions entering around 933.4 and 935.9 eV,
peaks can be identified, with reduction temperatures centered corresponding to well dispersed CuO and copper phyllosilicate,
at 470, 500, and 560 K, belonging to the reduction of CuO with respectively.51,57 Therefore, the rather large positive shift of
diverse sizes. The 470 and 500 K temperature reduction peaks the Cu2p binding energy implies that the charge from the
are often ascribed to the reduction of fine CuO nanoparticles, metal ions transfer toward the support matrix, meaning that
whose varied size can cause the divergence of reduction the metal ions interact with the support rather strongly.58
temperature. The reduction at 560 K is generated by the Apparently, the co-existence of copper phyllosilicate and well
reduction of bulky CuO.54 The presence of bulky CuO phase is dispersed CuO contribute to the observed asymmetric line-
also confirmative to the XRD pattern for the same sample (Fig. shape of the Cu 2p3/2 peaks of the samples in Fig. S3, which is
1(A)). As the reduction of well dispersed CuO to Cu0 occurs at in line with the XRD, FTIR results.
the same temperature for copper phyllosilicate reduction to In order to discriminate Cu0 and Cu+ species, which are nearly
Cu+ the two species cannot be differentiated by using H2- identical in their XPS patterns, AES spectra are measured (Fig.
,
TPR.42,50 Briefly, H2-TPR measurements demonstrate that 4(B)).55 It is clear that each Cu LMM spectrum involves multiple
samples derived from UDP method comprise mainly well components as indicated by the asymmetric and broad shape
dispersed Cu species, whereas impregnation-derived sample for the LMM peak. The broad envelope can be deconvoluted
with both dispersed and bulk CuO.
into two individual peaks cantered at 916.0 and 912.0 eV,
corresponding to Cu0 and Cu+, respectively.24 Henceforth,
there coexists Cu0 and Cu+ species on silica-supported copper
catalysts derived from UDP protocol.23
3.4 XPS measurements
To elucidate the chemical states of the copper species in the
reduced samples, surface sensitive XPS and AES spectra have
been collected, as manifested in Fig. 4. To exclude the
adventitious oxidation of Cu0 by contacting with air, the as-
reduced samples were sealed in protection of N2 before
measurements. For the reduced samples (Fig. 4(A)), the
binding energy of Cu 2p3/2 core levels appearing at 932.4 -
932.9 eV could be attributed to either Cu+ or Cu0.55 In the case
of Cu 2p3/2, there should be one main peak at ca. 935.8 eV,
with an accompanied satellite lines.24 The absence of these
characteristics imply that no Cu2+ presents in the reduced
samples derived from UDP method. The discrimination was
also verified by the presence of Cu2O and Cu phases found in
XRD measurements at high loadings (Fig. 1(B)). To determine
==================================================================
932.8
(A)
952.6
(f)
(e)
(d)
(c)
(b)
(a)
==================================================================
960
955
950
945
940
935
930
925
Binding Energy/ eV
(g)
9 16 0.
9 120.
(B)
(f)
(f)
(e)
(e)
y
t
i
s
n
(d)
(c)
e
t
n
I
(d)
(c)
(b)
(a)
(b)
(a)
400
500
600
700
800
900
1000
Temperature /K
90
5
9
10
9
15
9
20
9 25
K niet icEne rgye/
V
Fig. 3. TPR profiles of the Cu/SiO2 catalysts. (a) 10Cu/SiO2, (b) 15Cu/SiO2, (c)
20Cu/SiO2, (d) 25Cu/SiO2, (e) 30Cu/SiO2, (f) 40Cu/SiO2, (g) 20Cu/SiO2-IM.
Fig. 4. (A) Cu 2p XPS spectra of the reduced catalysts; (B) Cu LMM XAES
profiles of the reduced catalysts with different copper loadings. (a)
10Cu/SiO2, (b) 15Cu/SiO2, (c) 20Cu/SiO2, (d) 25Cu/SiO2, (e) 30Cu/SiO2, (f)
the structure change during H2 reduction, the XPS patterns for
typical as-prepared and reduced 20Cu/SiO2 are laid out in Fig.
40Cu/SiO2
.
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
New Journal of Chemistry
Page 6 of 14
View Article Online
DOI: 10.1039/C6NJ03960J
ARTICLE
Journal Name
Table 1 Cu LMM deconvolution results of the surface Cu components in 1846.0 eV can be attributed to Cu0 and Cu+, respectively. The
the reduced catalysts
smaller α’ for Cu0 and Cu+ than the bulk values stem from the highly
dispersed state and the strong interaction with silica, respectively.
In short, the strong metal-support interaction leads to the partial
reduction precursor and co-presence of both Cu0 and Cu+ species.
3.5 TEM images
Cu
XCu+
(%)c
KE (eV)a
Sample
AP (eV)b
Cu+
2p3/2
BE (eV)
Cu+
Cu0
Cu0
Fig. 5 demonstrates the HRTEM images of the representative
20Cu/SiO2 samples after calcination and reduction. The HRTEM
images clearly reveal that numerous whisker-shaped copper
phyllosilicate are present in the calcined sample, and no obvious
CuO particles can be observed since they are either well dispersed
or incorporated into silica substrate. After reduction at 623 K, the
whisker-shaped copper phyllosilicate vanished, with the formation
of Cu0 (< 5nm in size) densely covering the surface of the cross-
linked SiO2 spheres. In addition, surface characterization by
HAADF-STEM was also used to determine the surface
distributions of Cu species on the Cu/SiO2 catalyst (Fig. 5(D)),
demonstrating that via urea-assisted deposition precipitation, the
Cu species are evenly distributed onto the silica texture. Fig. S5
compares the representatives for the IM- and UDP-series of the
calcined and reduced samples. As for the IM-derived one, copper
particles can be clearly observed on the silica surface without any
whisk-like phyllosilicate after calcination, and they aggregated
resulting a poorer copper dispersion after reduction at 623 K. For
the calcined UDP- samples, the corresponding amount of copper
phyllosilicate with the increase in copper loading can be directly
observed. This agrees well with the FTIR observation that content of
10Cu/SiO2 913.2
15Cu/SiO2 913.4
20Cu-SiO2 913.0
25Cu-SiO2 912.6
30Cu-SiO2 912.9
916.5
916.5
916.2
916.4
916.5
1845.8 1849.2
1846.0 1849.1
1845.7 1848.9
1845.2 1849.0
1845.3 1848.9
1845.1 1849.0
932.6
932.6
932.8
932.6
932.5
24.3
35.7
41.0
31.4
30.1
26.3
40Cu-SiO2 912.5
916.3
932.7
a
b
c
Note: Kinetic Energy. Modified Auger Parameter. Intensity ratio
between Cu+ and (Cu0 + Cu+) by deconvolution of Cu LMM XAES spectra.
=============================================================
Table 2 Characterization of Cu Species on reduced Cu/SiO2 catalysts
+
0
+
SCu
/m2 g-1 a
SCu
/m2 g-1c
ACu
/area g-1 d
Sample
XCu+ /%b
10Cu/SiO2
15Cu/SiO2
20Cu/SiO2
25Cu/SiO2
30Cu/SiO2
21.6
24.3
35.7
41.0
31.4
30.1
26.3
6.9
62
26.0
14.4
22.7
13.7
11.4
8.7
145
32.6
190
30.0
126
26.5
111
40Cu/SiO2
a
24.3
79
b
Note:
Cu+/(Cu++Cu0) calculated from Cu LMM XAES spectra. Cu+ surface area
Metallic Cu surface area determined by N2O titration.
c
=============================================================
0
estimated on the basis of SCu and XCu+ under the assumption that the Cu+
ions and Cu0 atoms occupy identical areas and have identical atomic
d
sensitivity factors. Integral area under the peaks of CO adsorption on
active Cu+ on the surface of catalysts after evacuation (Figure S4).
=============================================================
An influence of the copper loading on the components of the
Cu+ and Cu0 species could be inferred from AES measurement.
Table 1 compiles the deconvolution results of the original Cu
LMM peaks, as well as the peak positions and their distinct
contributions. The Cu+/ (Cu+ + Cu0) ratio collected from the
deconvolution of Cu LMM spectra increased with the copper
loading, and passes a maximium at 20 wt.% with a ratio of
41.0%. It has been recognized that CuO weakly bonding to the
support can be readily reduced to Cu0 at 513 K.59,60 It is
reckoned that the strong metal-support interaction in Cu2+-
SiO2 has hindered the complete reduction of Cu2+ into Cu0, and
henceforth some Cu+ has formed after H2 reduction. At high Cu
loadings, the formation of excessive CuO phase results in the
formation of more Cu0.
40
Mean size=4.0 nm
(C)
30
20
10
0
The modified Auger parameter α’, defined as the summation of the
kinetic energy (KE) of the Cu LMM Auger electron and the binding
energy of the Cu 2p3/2 photoelectron, was performed to quantify
the contribution from Cu0 and Cu+ species.51 Generally, α’ is ca.
1851.0 eV for Cu0 and 1849.0 eV for Cu+.61 In addition, it can be 2-3
eV lower when the copper (in +2, +1, or 0 valence) is highly
dispersed and interacts intimately with the support. 62 Therefore, as
demonstrated in Table 2, the α’ values at ca. 1849.0 eV and ca.
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
Particle size /nm
Fig. 5. HRTEM images of 20Cu/SiO2: (A) as-calcined; (B) as-reduced; (C) particle
size distribution of the as-reduced catalyst; (D) HAADF-STEM image of the as-
reduced catalyst.
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
New Journal of Chemistry
Page 7 of 14
Journal Name
View Article Online
DOI: 10.1039/C6NJ03960J
ARTICLE
phyllosilicate grows with the copper loading, as demonstrated by derived from the desorption branch using the BJH algorithm
the I670/I800 ratio increase with increasing Cu loading. For the (Fig. 1B) show that at elevated copper loadings, the
reduced catalysts, the sizes of copper particles are notably small (<5 contribution of pores at ca. 12.4 nm to the total pore volume
nm), which is consistent with the estimation using Scherrer’s increased considerably at the expense of pores at ca. 7.6 nm.
equation (Table 3). Besides, the TEM images shows that copper is The appearance of the larger pore sizes and improvements of
well dispersed on support for all the reduced catalysts, and is BET surface area in the calcined samples may originate from
confirmative to absence of XRD diffraction lines.
the abundant formation of copper phyllosilicate which is
reported to enhance the BET surface area,48 whose formation
is also confirmed by FTIR and TEM. On account of the fact that
the copper content in the precipitates never exceeded that of
bulk chrysocolla (ca. 40 wt. %), all copper species in the
catalyst precursors could be accommodated in a well-
dispersed state,48 which has been identified by XRD, XPS, and
TEM. The hysteresis loops vary from H1-type for 20Cu/SiO2 to
H3 type for 25Cu/SiO2 implying that the pore shapes of these
Cu/SiO2 samples have altered from spherical to silt-like
channels.63 The hysteresis loop demonstrated in Fig. 6(A) is
typically acquired with aggregated platy particles with slit-
shaped mesopores,30 which is corroborated also by the TEM
3.6 N2O titration
The exposed active surface areas for metallic Cu0 has been
deduced from a N2O titration measurements, as is compiled in
Table 3. The metallic copper surface area increases through a
maximum of 32.6 m2 g-1 at the copper loading of 20 wt. %,
then decreases to 26.5 m2 g-1. In contrast, the Cu0 surface area
of 20Cu/SiO2-IM is 3.4 m2 g-1. The efficacious dispersion of
copper species on the catalyst surface can be ascribed to the
special textural structure originating from that copper
phyllosilicate.
3.7 Physiochemical properties of catalyst
Textural properties have been derived from N2 adsorption images (Fig. S5).
branch of the physisorption isotherms, as shown in Fig. 6. The 3.8 Catalytic tests
physicochemical properties of the calcined Cu/SiO2 samples
The catalytic performance of the above prepared catalysts was
are tabulated in Table 3. Fig. 6(A) and Fig. 6(B) demonstrates
the N2 adsorption-desorption isotherms and the corresponding
pore size distribution curves of the calcined samples,
respectively. It is found that the BET surface area increased
from 287 to 422 m2 g-1 when the copper loading increases
from 5 to 25 wt.%, then decreases to 388 m2 g-1 when copper
loading further increases. All these samples display type IV
isotherms, and the abrupt uptake of N2 at relative pressure of
P/P0 at 0.8 originates from a capillary effect due to the
presence of mesopores. The inferred pore-size-distribution is
shown in Fig. 6(B), and mesopores centered at nm can be
derived, showing that UDP method could create a porous
matrix during preparation. The pore size distribution curves
pre-reduced in H2 before catalytic DMM hydrogenation tests.
The tests were carried out at 463 K, 5.0 MPa, LHSVDMM 0.4 g
-1
gcat h-1, DMM/H2 malar ratio 200, in a fixed bed reactor.
Besides 1,3-PDO, sides products, such as meyhyl 3-
hydroxypropionate (3-HPM), n-propanol (NPA), methyl
propionate (MP), and methyl acetate (MA) are also detected in
the products. Catalytic activity and selectivity as a function of
Cu loading are shown in Fig. 7. The catalytic activity in terms of
DMM conversion is found to increase hand-in-hand with Cu
loading, but more significant at loadings below 20 wt.%. Above
20 wt.%, the conversion reaches > 90% and increase in activity
becomes less prominent with Cu loading increase. The
================================================================================================================================
Table 3 Physicochemical properties of the UDP-derived Cu/SiO2 catalysts compared with 20Cu/SiO2-IM catalyst
Metallic Crystallite size (nm) b
Cu
d
SBET
VP
DP
(nm)
DCu
(%)
SCu0
TOFDMM
(h-1)
Catalyst
loadinga
(wt. %)
(m2 g-1)
(cm3 g-1)
(m2 g-1)
by XRD
NDc
by N2O titration
3.2
10Cu/SiO2
15Cu/SiO2
20Cu/SiO2
25Cu/SiO2
30Cu/SiO2
40Cu/SiO2
20Cu/SiO2-IM
9.8
287
313
361
422
406
388
148
0.50
0.57
0.78
0.82
0.81
0.67
0.28
7.0
7.2
8.6
8.8
7.9
6.9
7.5
2.4
3.6
5.8
5.5
4.6
4.0
2.0
33.3
21.6
14.7
19.8
24.6
29.4
39.5
20.2
ND
ND
4.0
4.3
26.7
25.1
18.5
14.5
26.0
32.6
30.0
28.3
5.9
6.3
7.9
34.2
5.8
7.4
10.5
31.5
10.2
2.6
26.5
5.4
Note: a Determined by ICP-AES analysis. b Diameter of Cu particles obtained from the XRD data by the Scherrer equation. c ND= not detected. d The TOF value
was calculated as moles of ester reacted in the initial 1h per mole of surface copper calculated from the copper dispersion.
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
New Journal of Chemistry
Page 8 of 14
View Article Online
DOI: 10.1039/C6NJ03960J
Journal Name
ARTICLE
preferred method for catalyst preparation. In contrast, the
DMM conversion of 70.4% and the 1,3-PDO selectivity of
20.6% were obtained on the IM catalyst with the same copper
loading. These results evidently showed the UDP catalyst is
efficacious in the chemo-selective hydrogenation of DMM to
1,3-PDO, which may be ascribed to the enhanced Cu0 surface
area (Fig. S6).
(A)
+60
+50
70
60
50
40
30
20
10
0
(g)
(f)
e
m
+40
+30
(e)
(d)
u
l
o
V
n
o
i
t
p
r
The analyses of products distribution allow the identification
of the different side reactions and the mapping of the reaction
network. As shown in Scheme 1, DMM hydrogenation is a
consecutive reaction, according to the gas kinetic theory, the
possibility of the simultaneous hydrogenation of two C=O
functional groups is very low. Therefore, the main reaction of
DMM hydrogenation is believed to involve two serial
reactions: the hydrogenation of DMM to form methyl 3-
+20
+10
(c)
(b)
(a)
o
s
e
D
2
N
0.0
0.2
0.4
0.6
0.8
1.0
P/P
0
hydroxypropionate
(3-HPM)
and
the
subsequent
(B)
3.2
7.6
12.4
hydrogenation of 3-HPM to produce 1,3-PDO. Besides the
hydrogenation to 1,3-PDO, 3-HPM also undergoes a parallel
dehydration-hydrogenation to methyl propionate (MP) based
on the hydrogenation of different functional groups in 3-HPM.
Furthermore, excessive hydrogenation of both 1,3-PDO and
MP can lead to the formation of n-propanol (NPA). In parallel
to hydrogenation reaction, a small fraction (< 5%) of DMM
decomposes to form methyl acetate (MA) and CO2. The
comparable hydrogenation selectivities for 1,3-PDO and MP
(g)
(f)
(e)
(d)
(c)
(b)
(a)
40
0
10
20
30
50
==================================================================
Pore Diameter /nm
Fig. 6. (A) N2 physisorption isotherms of the calcined catalysts; (B) BJH
mesopore size distribution for the calcined catalysts. (a) 10Cu/SiO2, (b)
15Cu/SiO2, (c) 20Cu/SiO2, (d) 25Cu/SiO2, (e) 30Cu/SiO2, (f) 40Cu/SiO2, (g)
100
50
20Cu/SiO2-IM
.
←
→
90
DMM
PA
MA
40
N
M
M
==================================================================
MP
13,- PDO
3-H PM
D
f
o
y
t
i
30
v
i
t
80
70
n
selectivity for target product 1,3-PDO increases with Cu
loading until a maximum (34.1%) is attainted at 20 wt.% Cu
loading, and selectivity drops with further loading increase.
The 20Cu/SiO2 catalyst demonstrated a highest activity (90.5%
conversion for DMM hydrogenation) with 34.1% of 1,3-PDO
selectivity. The selectivity for side-products, both methyl
propionate and n-propanol increases with increasing Cu
loading, with analogous trend to that of DMM conversion. The
selectivity for 3-HPM is found to decrease with increasing
loading and DMM conversion. Selectivity for MA is
independent of Cu loading, and also remains low (< 5%) during
the tests. For 20Cu/SiO2-IM and 20Cu/SiO2 with identical Cu
loading, both higher activity and higher 1,3-PDO selectivity
have been manifested by the latter, showing that UDP is the
c
o
i
e
l
s
r
e
e
v
20
S
n
o
C
10
0
60
10
20
30
40
IM
Coppe r Loadni gw/ t . %
Fig. 7. Catalytic performance of Cu/SiO2 catalyst for DMM hydrogenation as
a function of copper loading. The 20Cu/SiO2-IM sample has been shifted to
0 on the x-axial. Reaction conditions: T=463 K, P=5.0 MPa, H2/DMM=200
(mol/mol), and LHSVDMM =0.4 g gcat-1 h-1.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 8
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 9 of 14
Journal Name
View Article Online
DOI: 10.1039/C6NJ03960J
ARTICLE
suggest that the two parallel reaction leading to their time, showing it is the primary hydrogenation product. Selectivity
formation have close kinetic features. Their selectivities reach for 1,3-PDO and MP increases with increasing contact time, as a
maximum at medium conversion levels, together with the result of 3-HPM hydrogenation. 1,3-PDO selectivity is found to be
continuous increase in NPA with DMM conversion, strongly higher than that of MP, implying that the formation of the former is
suggest that both 1,3-PDO and MP are intermediates for the kinetically favourable. Selectivity for excessive hydrogenation
formation of NPA. This is for the very first time that DMM product, NPA, increase with contact time and exceeds selectivity for
-1
hydrogenation network has been disclosed. The finding 1,3-PDO at LHSVDMM of 0.4 g gcat h-1. This suggests that the longer
demonstrates that future efforts should be placed on the control contact time favours deeper hydrogenation, which is in consistent
over catalyst hydrogenation chemo-selectivity and prevent with the reaction network. The maximum of 1,3-PDO selectivity is
excessive hydrogenation to unwanted NPA, and simultaneously obtained at LHSVDMM of 0.4 g gcat-1 h-1. Meanwhile, the selectivity for
lower the decomposition of DMM.
decomposition product MA is found to be unaffected by the change
The influence of operation conditions, including liquid hourly space in contact time.
velocity of dimethyl malonate (LHSVDMM), temperature, pressure, The effect of reaction temperature over the range of 443 to
and H2/DMM ratio, on DMM catalytic hydrogenation was also 483 K was also assessed, as shown in Fig. 8(B). DMM
explored for 20Cu/SiO2, as displayed in Fig. 8. For a given reaction conversion
increases
monotonically
with
reaction
temperature (463 K), pressure (5.0 MPa), and H2/DMM ratio (200 temperature, and a simultaneous drop in 3-HPM selectivity is
mol/mol), the conversion of DMM undergoes a significant decrease observed, showing a fast transformation of the reactant and
-1
-1
when increasing the LHSVDMM from 0.2 g gcat h-1 to 0.6 g gcat h-1, primary product. The 1,3-PDO selectivity increases steadily
translating to contact variation from 0.70 min to 0.23 min. The with
selectivity of 3-HPM increases prominently with shorter contact
========================================================================================================================================
100
90
80
70
60
50
100
90
80
70
60
50
(A)
(B)
←
←
DMM
PAl
MA
MP
13,- PDO
3-H PM
→
DMM
PA
MA
MP
13,- PDO
3-H PM
N
N
M
M
M
M
→
D
f
o
D
f
o
40
40
y
y
t
i
t
i
v
i
v
i
70
70
t
t
n
n
c
c
o
i
o
i
e
e
30
30
l
l
s
r
s
r
e
e
e
v
e
v
S
S
60
50
60
50
n
o
n
o
20
20
C
C
10
0
10
0
40
40
0
.2
0 3.
0
.4
0
.5
0 6.
440
450
460
470
480
-
1
Te mpK/
LH SVh/
100
90
80
70
60
50
100
90
80
70
60
50
DMM
PA
MA
MP
13,- PDO
3-H PM
(C)
N
←
←
(D)
DMM
N
PA
MA
MP
13,- PDO
→
M
M
M
M
D
f
o
D
f
o
3-H PM
40
40
y
y
t
i
t
i
v
i
v
i
70
70
t
t
n
n
c
c
o
i
o
i
e
l
e
l
→
30
30
s
r
s
r
e
e
e
v
e
v
S
S
60
50
60
50
n
o
n
o
20
20
C
C
10
0
10
0
40
40
2
3
4
5
6
100
150
200
250
300
Per ssuer /MPa
H E/
Fig. 8. Effect of reaction conditions on reaction reactivity. (A) Effect of DMM liquid hourly space velocity (LHSVDMM), T = 463 K, P = 5.0 MPa, H2/DMM = 200
mol/mol. (B) Effect of temperature, LHSVDMM = 0.4 g gcat-1 h-1, P = 5.0 MPa, H2/DMM = 200 mol/mol. (C) Effect of pressure, LHSVDMM = 0.4 g gcat-1 h-1, T = 463
K, H2/DMM = 200 mol/mol. (D) Effect of molar ratio of H2/DMM, LHSVDMM = 0.4 g gcat-1 h-1, T = 463 K, P =5.0 MPa.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 10 of 14
View Article Online
DOI: 10.1039/C6NJ03960J
Journal Name
ARTICLE
temperature before passing the maximum at 463 K, whereby Among all the important chemical approaches above-
an optimized selectivity of 34.1% is attained. Further mentioned, the yield is no more than 80% and 40% for AC and
3
hydrogenation of 1,3-PDO to NPA predominates at higher EO method, respectively. As for the glycerol hydrogenolysis
temperatures. Unlike selectivity for 1,3-PDO, the MP selectivity method, the yield of 1,3-PDO is 28.1% (glycerol conversion
increases with reaction temperature constantly, implying that 54.3%, 1,3-PDO selectivity of 52.0%),9 the yield of 29.3% is
3-HPM
dehydration-hydrogenation
overrates
MP obtained via the hydrogenation of diethyl malonate method
hydrogenation at high temperatures. The selectivity for NPA (90.7% conversion of dimethyl malonate, 32.3% selectivity of
increases with temperature, and becomes significant as a 1,3-PDO).10,11 Compared with all these approaches, the new
result of 1,3-PDO and MP conversion.
alternative, hydrogenation of dimethyl malonate to 1,3-PDO
The DMM hydrogenation products vary significantly with over Cu/SiO2, achieving a yield of 41.7%, has its promising
pressure change for reactions involving molecule number industrial application.
The deactivation and lifespan of the Cu/SiO2 catalysts influence the
practical industrialization of vapour-phase hydrogenation of
dimethyl malonate to 1,3-propanediol. Therefore, the catalytic
performance of the optimal 20Cu/SiO2 catalysts was evaluated by a
time-on-stream stability test under optimal reaction conditions, i.e.,
change. Fig. 8(C) shows the corresponding selectivity for each
product as a function of operating pressure. It is seen that high
pressure favours the formation of 1,3-PDO, but disfavours the
formation of 3-HPM and MP, reflecting that formation of 1,3-
PDO is the kinetically preferred reaction at elevated pressures.
It also significantly influences DMM conversion. It shows that
pressure increase leads to the increase of hydrogenation
products that involves a reduced molecule number, while
reactions without molecular number change does not vary
prominently. A high molar ratio of H2/DMM in the feeding is
also favourable for the hydrogenation of DMM, as delineated
in Fig. 8(D), both DMM conversion and 1,3-PDO selectivity
increase with H2/DMM ratio under the given H2/DMM range,
and achieve their maximum of 97.1% and 43.3%, respectively
at the H2/DMM ratio of 300.
-1
463 K, 5.0 MPa, H2/DMM = 300, and LHSVDMM = 0.4 g gcat h-1 (Fig.
9). The result shows that the catalyst maintains its high catalytic
activity, i.e. ca. 97% conversion of DMM and ca. 43% selectivity to
1,3-PDO, for the duration of 60 h, indicating that the 20Cu/SiO2
catalyst exhibits a satisfactory level of catalytic activity and long-
term stability under the optimal operation condition.
4. Discussion
Based on the proposed reaction network, the catalytic activity
and selectivity has been correlated with Cu0 and Cu+
components in the samples (Fig. 10), with the aim to shed light
on a catalysis-structure relationship. The hydrogenation
activity for H2 consumption increases almost linearly with
exposing Cu0 surface area; the abrupt deviation to the linear
relationship can be ascribed to high DMM conversion
indicating too fast reaction rate of DMM hydrogenation over
the catalysts under higher copper loadings (>30 wt. %), mass
transfer resistance may become important and the kinetics
obtained would not be intrinsic. And the yield of 1,3-PDO is in
a significantly linear relationship with Cu0 surface area (Fig.
S7), both convincing that Cu0 is the active site for
hydrogenation reactions. This agrees well with previous
reports on similar hydrogenation reaction.23,29,34 Deviation
from linearity can be observed for the overall process may be
controlled by mass transport at conversions higher than 90%.
On the other hand, 1,3-PDO selectivity shows no clear trend
with Cu0 content (Fig. S8), but increases with Cu+/(Cu++Cu0)
ratio (Fig. 10(B)),
=============================================================
100
DMM Conv.
1,3-PDO Sel.
90
50
40
0
10
20
30
40
50
60
Time on stream /h
Fig. 9. The long-term catalytic performance of the DMM hydrogenation over
20Cu/SiO2. Reaction conditions: T = 463 K, P =5.0 MPa, H2/DMM =300, LHSVDMM
=
0.4 g gcat-1 h-1.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 10
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 11 of 14
Journal Name
View Article Online
DOI: 10.1039/C6NJ03960J
ARTICLE
be directed towards generating and maintaining an
appropriate Cu+/ (Cu0 + Cu+) ratio for DMM hydrogenation.
5. Conclusions
In summary, dimethyl malonate vapor-phase hydrogenation to
produce 1,3-PDO has been examined on Cu/SiO2 catalyst. The
catalytic hydrogenation network has been disclosed by
analyzing the catalytic products distribution. Dimethyl
malonate hydrogenation takes place via 3-HPM as the primary
product, which can be further hydrogenated into 1,3-PDO or
MP through parallel reactions. 1,3-PDO or methyl propionate
can be further hydrogenated into n-propanol that reduces the
selectivity of 1,3-PDO. An insignificant amount of DMM thermal
decomposition can also reduce the selectivity. Manipulation of
reaction temperature, pressure and feeding conditions to optimize
1,3-PDO selectivity shows that a maximized 1,3-PDO yield (41.7%)
can be achieved at LHSVDMM = 0.4 g gcat-1 h-1, T = 463 K, P = 5.0 MPa,
H/E =300.
Scheme 1 Catalytic reaction network for the hydrogenation of DMM over
Cu/SiO2 catalyst.
=============================================================
implying Cu+ modification is important for the selective
hydrogenation to 1,3-PDO. We tentatively hypothesize that
the catalytic activity is associated solely to the number of
Structural characterization and catalytic performance
evaluation show that co-presence of Cu+ is crucial for 1,3-PDO
exposed surface Cu0, but the selectivity is strongly affected by ==================================================================
the co-presence of Cu+, despite that the connection between
Cu+ density and selectivity is still not well understood thus far.
As far as the nature of active sites are concerned, there is no
0.015
consensus for the hydrogenation of acyl compounds. On the
one hand, some authors believe that only Cu0 acts as the
(A)
0.014
active site, on which H2 is disassociately activated, and the
reactant is adsorbed through the acyl group and cleaved.26,64
0.013
0.012
On the other hand, in methyl acetate hydrogenation, Poels and
Brands,65 Wang et al.55 reported that the Cu+ sites contribute
0.011
to the adsorption of methoxy and acyl species of esters, while
Cu0 species are not capable of activating methoxy at the
0.010
catalytic temperature (483 K), as sponsored by their in situ
FTIR spectra and DFT modeling results. Chen et al.24 also
postulated that H2 is dissociatively activated on Cu0 sites and
0.009
0.008
20
22
24
26
28
30
32
34
SCu0 /m2 g-1
the C=O bond in DMO molecules are polarized and activated
on Cu+ sites.
In the current DMM hydrogenation, the correlations shown in
Fig. 10 and Fig. S7 suggests that Cu0 is likely the active site
where all hydrogenation reactions occur, and the Cu+ is crucial
for the selectivity of 1,3-PDO. These speculations are
supported by the XRD and AES results, as well as the clear
correlation with respect to hydrogenation activity and
selectivity. Cu+ may work as electrophilic or Lewis acidic sites
to polarize the C=O bond via the electron lone pair in oxygen,66
thus improving the reactivity of the ester group in dimethyl
malonate. Thus, it is highly reasonable to suppose that Cu0
functions as the primary active site, and the Cu+ facilitates
conversion of substrate and intermediates, as has been
36
34
32
30
28
26
24
22
20
(B)
24
28
32
36
40
44
Cu+/(Cu0+Cu+) /%
suggested in the hydrogenation reaction of dimethyl
furfural,68
64
oxalate,34,67
crotonaldehyde.
dimethyl
69
maleate,
and
Such a relationship suggests that the
Fig. 10. Synergetic effect of Cu0 and Cu+ on the catalytic performance: (A)
correlation of H2 consumption with Cu0 surface areas; (B) correlation of 1,3-PDO
selectivity with Cu+/(Cu0+Cu+)
synergetic effect of Cu0 and Cu+ is important to the
maximization of 1,3-PDO selectivity, and future efforts should
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 11
Please do not adjust margins

Please do not adjust margins
New Journal of Chemistry
Page 12 of 14
View Article Online
DOI: 10.1039/C6NJ03960J
ARTICLE
Journal Name
selectivity and Cu0 is relevant to the overall hydrogenation
activity. The reaction network and promoting role of Cu+ are
important to guide catalyst design in the future.
28 E. K. Poels, D. S. Brands, Appl. Catal. A-Gen. 2000, 191, 83-96.
29 A.Y. Yin, X.Y. Guo, W.L. Dai, K.N. Fan, J. Phys. Chem. C. 2009,
113, 11003-11013.
30 S. Wang, X. Li, Q. Yin, L. Zhu, Z. Luo, Catal. Commun. 2011,
12, 1246-1250.
31 K. L. Deutsch, B. H. Shanks, J. Catal. 2012, 285, 235-241.
32 C. Sun, P. Zeng, M. He, X. He, X. Xie, Catal. Commun. 2016,
86, 5-8.
Acknowledgements
Support from the Research Centre of Pujing Chemical Industry
Co. Ltd is gratefully acknowledged. The assistance of
Yongzheng Li in performing the N2O titration and Jingyu Zhou
in performing CO-adsorption estimates is greatly appreciated.
33 S. Natesakhawat, J. W. Lekse, J. P. Baltrus, P. R. Ohodnicki, B.
H. Howard, X. Deng, C. Matranga, ACS Catal. 2012,
1676.
2, 1667-
34 J. Gong, H. Yue, Y. Zhao, S. Zhao, L. Zhao, J. Lv, S. Wang, X.
Ma, J. Am. Chem. Soc. 2012, 134, 13922-13925.
35 T. Toupance, M. Kermarec, J.F. Lambert, C. Louis, J. Phys.
Chem. B. 2002, 106, 2277-2286.
36 E.D. Guerreiro, O.F. Gorriz, J.B. Rivarola, L.A. Arrua, App.
Catal. A-Gen. 1997, 165, 259-271.
37 X. Guo,; A. Yin, W.L. Dai, K. Fan, Catal. Lett. 2009, 132, 22-27.
References
1
2
3
J. V., Kurian, J. Polym. Environ. 2005, 13, 159-167.
G. A. Kraus, CLEAN - Soil, Air, Water. 2008, 36, 648-651.
R. K. Saxena, P. Anand, S. Saran, J. Isar, Biotechnol adv. 2009,
27, 895-913.
38 H. Yue, Y. Zhao, X. Ma, J. Gong, Chem. Soc. Rev., 2012, 41
4218-44.
,
39 T. Toupance, M. Kermarec, C. Louis, J. Phys. Chem. B. 2000,
104, 965-972.
4
5
H. Biebl, Appl. Microbiol. Biotechnol, 1991, 35, 701-705.
R. A. Urban, B. R. Bakshi, Ind. Eng. Chem. Res. 2009, 48
,
40 C. J. G. Van der Grift, A. F. H. Wielers, A. Mulder, J. W. Geus,
Thermochim. Acta. 1990, 171, 95-113.
41 E. D. Guerreiro, O. F. Gorriz, G. Larsen, L. A. Arrúa, App. Catal.
A-Gen. 2000, 204, 33-48.
42 A.J. Marchi, L.L.G. Fierro, J. Santamaría, A. Monzón, App.
Catal. A-Gen. 1996, 142, 375-386.
43 Z.W. Huang, F. Cui, J.J. Xue, J.L. Zuo, J. Chen, C.G. Xia, J. Phys.
Chem. C. 2010, 114, 16104-16113.
8068-8082.
H. S. Wu, Y. J. Wang, Ind. Eng. Chem. Res. 2012, 51, 10930-
10935.
S. Zhu, X. Gao, Y. Zhu, J. Cui, H. Zheng, Y. Li, Appl. Catal. B.
2014, 158-159, 391-399.
E.S. Vasiliadou, T.M. Eggenhuisen, P. Munnik, P.E. de Jongh,
K.P. de Jong, A.A. Lemonidou, Appl. Catal. B. 2014, 145, 108-
119.
6
7
8
44 G. Cordoba, R. Arroyo, J.L.G. Fierro, M. Viniegra, J. Solid State
Chem. 1996, 123, 93-99.
45 T. López, J. Hernandez-Ventura, M. Asomoza, A. Campero, R.
Gómez, Mater. Lette. 1999, 41, 309-316.
46 Y. X. Zhang, M. Huang, F. Li, Z. Q. Wen, Int. J. Electrochem.
9
S. Zhu, X. Gao, Y. Zhu, Y. Zhu, H. Zheng, Y. Li, J. Catal. 2013,
303, 70-79.
10 T. Ding, H. Tian, J. Liu, W. Wu, J. Yu, Chinese J. Catal. 2016,
37, 484-493.
11 T. Ding, H. Tian, B. Zhao, React. Kinet. Mech. Cat. 2016, 118
497-508.
,
Sci. 2013, 8, 8645-8661.
47 G. Díaz, R. Pérez-Hernández, A. Gómez-Cortés, M. Benaissa,
R. Mariscal, J.L.G. Fierro, J. Catal. 1999, 187, 1-14.
48 C. J. G. Van der Grift, P. A. Elberse, A. Mulder, J. W. Geus,
Appl. Catal. 1990, 59, 275-289.
49 L.Y. Chen, T. Horiuchi, T. Osaki, T. Mori, App. Catal. B-
Environ. 1999, 23, 259-269.
12 http://www.grandviewresearch.com/industry-analysis/1-3-
propanediol-pdo-market
13 H. Yagita ， H. Kuwahara, K. Omata, K. Fujimoto,
ChemInform. 1990, 43, 335-338.
14 K. Maeda, H. Yagita, K. Omata, K. Fujimoto, J. Mol. Catal.
1992, 71, 347-355.
50 J.A. Anderson, C. Márquez-Alvarez, M.J. López-Munóz, I.
Rodríguez-Ramos, A. Guerrero-Ruiz, App. Catal. B-Environ.,
1997, 14, 189-202.
51 A. Gervasini, M.M.G. Martra, A. Ponti, N.R.L. Sordelli, F.
Zaccheria, J. Phys. Chem. B. 2006, 110, 7851-7861.
52 C. J. G. Van der Grift, A. Mulder, J. W. Geus, Appl. Catal.
1990, 60, 181-192.
53 Q. Sun, Y.L. Zhang, H.Y. Chen, J.F. Deng, D. Wu, S.Y. Chen, J.
Catal. 1997, 167, 92-105.
54 F. Boccuzzi, A. Chiorino, G. Martra, M. Gargano, N. Ravasio,
B. Carrozzini, J. Catal. 1997, 165, 129-139.
55 Y. Wang, Y. Shen, Y. Zhao, J. Lv, S. Wang, X. Ma, ACS Catal.
15 X. Wang, W. Li, , R.C. Shen, Nat. Gas Chem. Ind. 2016, 41, 38-
40.
16 M. A. N. Santiago, M. A. Sánchez-Castillo, Cortright, R. D.;
Dumesic, J. A., J. Catal. 2000, 193, 16-28.
17 S. Wang, W. Guo, H. Wang, L. Zhu, S. Yin, K. Qiu, New J.
Chem. 2014, 38, 2792-2800.
18 J. Ding, J. Zhang, C. Zhang, K. Liu, H. Xiao, F. Kong, J. Chen,
App. Catal. A-Gen. 2015, 508, 68-79.
19 H. Lin, X. Zheng,; Z. He, J. Zheng, X. Duan, Y. Yuan, App. Catal.
A. 2012, 445-446, 287-296.
20 D.S. Brands, E.K. Poels, A. Bliek, Appl. Catal. A-Gen. 1999,
184, 279-289.
21 S. Zhu, X. Gao, Y. Zhu, W. Fan, J. Wang, Y. Li, Catal. Sci.
2015, 5, 6200-6208.
56 G. Petrini, F. Garbassi, J. Catal. 1984, 90, 113-118.
57 B. Peplinski, W. E. S. Unger, I. Grohmann, Appl. Surf. Sci.
1992, 62, 115-129.
Technol., 2015, 5, 1169-1180.
22 Z. Huang, F. Cui, H, Kang, J. Chen, C. Xia, Appl. Catal. A. 2009,
366, 288-298.
58 G. G. Jernigan, G. A. Somorjai, J. Catal. 1994, 147, 567-577.
59 Z. Liu, M.D. Amiridis, Y. Chen, J. Phys. Chem. B. 2005, 109
23 B. Zhang, S. Hui, S. Zhang, Y. Ji, W. Li, D.Y. Fang, J. Nat. Gas
Chem. 2012, 21, 563-570.
,
1251-1255.
60 E. L. Rodrigues, A. J. Marchi, C. R. Apesteguia, J. M. C. Bueno,
App. Catal. A-Gen. 2005, 294, 197-207.
61 K. Sun, W. Lu, F. Qiu, S. Liu, X. Xu, App. Catal. A-Gen. 2003,
252, 243-249.
24 L. Chen, P. Guo, M. Qiao, S. Yan, H. Li, W. Shen, H. Xu, K. Fan,
J. Catal. 2008, 257, 172-180.
25 C. Ohlinger, B. Kraushaar-Czarnetzki, Chem. Eng. Sci. 2003,
58, 1453-1461.
26 L.F. Chen, P.J. Guo, L.J. Zhu, M.H. Qiao, W. Shen, H.L. Xu, K.N.
Fan, App. Catal. A-Gen. 2009, 356, 129-136.
27 X.G. San, Y. Zhang, W.J. Shen, N. Tsubaki, Energy Fuel. 2009,
23, 2843-2844.
62 M. A. Kohler, H. E. Curry-Hyde, A. E. Hughes, B. A. Sexton, N.
W. Cant, J. Catal. 1987, 108, 323-333.
12 | 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
New Journal of Chemistry
Page 13 of 14
Journal Name
View Article Online
DOI: 10.1039/C6NJ03960J
ARTICLE
63 R. Schmidt, E. W. Hansen, M. Stoecker, D. Akporiaye, O.H.
Ellestad, J. Am. Chem. Soc. 1995, 117, 4049-4056.
64 M. Mokhtar, C. Ohlinger, J. H. Schlander, T. Turek, Chem.
Eng. Technol. 2001, 24, 423-426.
65 E. K. Poels, D. S. Brands, App. Catal. A-Gen. 2000, 191, 83-96.
66 S. Sitthisa, T. Sooknoi, Y. Ma, P. B. Balbuena, D. E. Resasco, J.
Catal. 2011, 277, 1-13.
67 S. Hui, B. Zhang, S. Zhang, W. Li, J. Nat. Gas Chem. 2012, 21
753-758.
,
68 C. P. Jiménez-Gómez, J. A. Cecilia, D. Durán-Martín, R.
Moreno-Tost, J. Santamaría-González, J. Mérida-Robles, R.
Mariscal, P. Maireles-Torres, J. Catal. 2016, 336, 107-115.
69 A. Dandekar, R.T.K. Baker, M.A. Vannice, J. Catal. 1999, 184
421-439.
,
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 13
Please do not adjust margins

New Journal of Chemistry
Page 14 of 14
View Article Online
DOI: 10.1039/C6NJ03960J
Contents Entry
1,3-propanediol was synthesized via the hydrogenation of dimethyl malonate over Cu/SiO2
catalyst. The reaction network and active sites were revealed for the first time.