Continuous Hydrogenation of Ethyl Levulinate to 1,4-Pentanediol over 2.8Cu-3.5Fe/SBA-15 Catalyst at Low Loading: The Effect of Fe Doping

Article abstract of DOI:10.1002/cssc.201901198

Bimetallic Cu–Fe catalysts with low loading were prepared for hydrogenation of ethyl levulinate (EL) to 1,4-pentanediol (1,4-PDO). Among them, 2.8Cu-3.5Fe/SBA-15 (Cu/Fe molar ratio of 1:1.5) performed best, capable of converting EL to the key intermediate γ-valerolactone (GVL) at 140 °C with 97 % yield. It can also be used to hydrogenate GVL to 1,4-PDO with 92.6 % selectivity or convert EL to 1,4-PDO in one pot. The high activity of the catalyst at such a low loading was attributed to the highly dispersed metal species and the Fe doping effect. Various characterization methods indicated that Fe acted as both structural and electronic modifier to promote the chemical properties of the Cu species. Besides, the incorporation of Fe provided abundant Lewis acid sites and accelerated the reaction process. CuFeO₂ was detected by energy-dispersive X-ray spectroscopy, X-ray photoelectron spectroscopy, and XRD. On the basis of a combination of characterization and reaction kinetics, synergistic catalysis by Cu⁰ and CuFeO₂ is considered to be responsible for the excellent performance of the Cu–Fe catalysts.

Full text of DOI:10.1002/cssc.201901198

Accepted Article
Title: Continuous Hydrogenation of Ethyl Levulinate to 1,4-Pentanediol
over 2.8Cu-3.5Fe/SBA-15 Catalyst at Low Loading: The Effect of
Fe Doping
Authors: Tianyu Deng, Long Yan, Xinglong Li, and Yao Fu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.201901198
Link to VoR: http://dx.doi.org/10.1002/cssc.201901198
A Journal of
www.chemsuschem.org

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
Continuous Hydrogenation of Ethyl Levulinate to 1,4-Pentanediol
over 2.8Cu-3.5Fe/SBA-15 Catalyst at Low Loading: The Effect of
Fe Doping
Tianyu Deng, Long Yan, Xinglong Li and Yao Fu*[a]
Abstract: Bimetallic Cu-Fe catalysts with low loading were prepared
for hydrogenation of ethyl levulinate (EL) to 1,4-pentanediol (1,4-
PDO). Among them, 2.8Cu-3.5Fe/SBA-15 (Cu/Fe molar ratio of
1:1.5) performed best, capable of converting EL into key -
Valerolactone (GVL) intermediate at 140 °C with 97% yield. It can
also be used to hydrogenate GVL to 1,4-PDO with 92.6% selectivity,
or convert EL to 1,4-PDO in one-pot. The high activity of catalyst at
such a low-loading was attributed to the highly dispersive metal
species and the Fe-doping effect. Various characterizations
indicated that Fe acted as both structural and electronic modifier to
promote the chemical properties of Cu-species. Besides, the
incorporation of Fe provided abundant Lewis acid sites and
accelerated the reaction process. A CuFeO2 conformation was
detected by EDX, XPS and XRD spectra. Combining
characterizations with reaction kinetics, the synergistic catalysis of
Cu0 and CuFeO2 is considered to be responsible for the excellent
performance of Cu-Fe catalysts.
performance on partial hydrogenation of C=O bonds, in which
Cu0 promotes hydrogen decomposition, while Cu+ sites activates
acyl groups. [2c, 4a, 5] Patankar et al. synthesized Cu/ZrO2 catalyst
with 79 wt% of copper loading to reach 70% selectivity of GVL
and 25% selectivity of 1,4-PDO at 200 °C.[2b] Iker et al. used
Cu/Al2O3 catalyst with 35 wt% of copper loading to obtain 64%
yield of GVL at 250 °C.[6] However, monometallic Cu-based
catalysts suffered the drawbacks of harsh reaction conditions
[5a, 5d, 7]
[5d]
(exceed 200 °C),
unsatisfactory diol selectivity,
the
and
[5b]
deactivation owing to the instability of Cu+ species
severe leaching owing to the high metal loading[2b, 6], which
hinder their industrial development. Introducing a metal to
synthesize bimetallic catalyst may be a feasible approach to
remedy defects.[2b, 6]
Introduction
Renewable biomass energy, which is expected to replace the
depleted fossil fuel, can be transferred into various building
blocks.[1] Ethyl levulinate (EL) derived from lignocellulosic
biomass is an attractive platform to produce fine chemicals, such
as -valerolactone (GVL) and 1,4-pentanediol (1,4-PDO), both of
[1-2]
which are valuable industrial reagents.
GVL is a potential
candidate for food flavor, fuel additive and green solvent owing
[3]
to its high reactivity and low toxicity,
and 1,4-PDO together
with other bio-based diols have been regarded as intermediates
for synthesis of specialized reagents and promising monomers
[2]
Scheme 1. The hydrogenation routes of ethyl levulinate.
of biodegradable polymeric materials.
Hence, the production
of GVL and 1,4-PDO from EL is of great significance in recent
years.
Researches on heterogeneous catalysts for hydrogenation of
EL to 1,4-PDO focused on noble metals and cheap copper in the
past decades. [2a, 2b, 4] The synergistic effect of the coexisting Cu0
and Cu+ species in Cu-based catalysts is the key for excellent
Fe is the most abundant metal resource in nature, capable of
[8]
reducing unsaturated bonds.
Some works indicate that the
incorporation of Fe into noble metal catalysts (Pd, Pt .etc) could
enhance the dispersion and/or stability of metal particles and
[8-9]
improve the catalytic activity.
Pino et al. reported the Fe-
doping effect on Pd-based catalyst enhanced the interaction of
oxygen-containing groups with the metal surface due to the
oxophilic nature of Fe, which would promote the hydrogenation
of C=O bonds and the cleavage of C-O bonds. [8a] However, the
existing research on Fe-modified catalysts is far less extensive
than Co, Cr, and Ni, and dominated by noble metal catalysts. In
particular, the study on low-content Fe-doping effect on low
loading cheap metal-based catalysts is relatively limited.
[a]
T. Deng, Dr. L. Yan, Dr. X. Li, Prof. Dr. Y. Fu
Hefei National Laboratory for Physical Sciences at the Microscale,
iChEM, CAS Key Laboratory of Urban Pollutant Conversion, Anhui
Province Key Laboratory of Biomass Clean Energy
University of Science and Technology of China
Hefei, Anhui, 230026, P.R. China.
E-mail: fuyao@ustc.edu.cn
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
The multiple reaction routes employing EL as feedstock and
H2 as hydrogen donor were presented in Scheme 1.[2, 4c, 7, 10]
The hydrogenation of EL will generate unstable ethyl 4-
hydroxypentanoate (4-HPE), which might be undetected in
several systems due to the subsequent rapid lactonization to
GVL.[4a, 11] The pathway including the initiating hydrolysis of EL,
the following dehydration of levulinic acid and the eventual
hydrogenation of angelica lactone afford GVL as well.
Subsequent hydrogenation of the stable GVL to the target
product 1,4-PDO is relatively demanding. The ring-opening
isomerization and hydrogenation of GVL will allow the formation
of byproducts valeric acid, meanwhile 1,4-PDO can be further
transferred into dehydrated and over-hydrogenated byproducts
such as n-pentanol (n-PeOH) and 2-methyltetrahydrofuran (2-
MTHF).[5d, 12] The diversity and complexity of the reaction routes
for hydrogenation of EL make it challenging to develop an
efficient catalyst for controllable production of GVL or 1,4-PDO
in the varying reaction conditions.
Figure 1. (A) TEM image and (B) SEM-EDX elemental mapping of Cu and Fe
of 2.8Cu-3.5Fe/SBA-15.
3.5Fe/SBA-15 was 2.60 wt%, and the iron was 2.58 wt%, which
were both lower than the theoretical values.
Accordingly, a series of SBA-15 supported bimetallic Cu-Fe
catalysts at low loading were prepared by co-impregnation
method for continuous hydrogenation of EL to 1,4-PDO with
molecule H2 as hydrogen donor. Because of the stability and the
versatility of GVL intermediate, the conversion of EL to GVL was
preferentially investigated, the researches on hydrogenation of
GVL to 1,4-PDO and the combination of above two steps into
one-pot synthesis ensued posteriorly. In this work, the structure
and microscopic properties of this highly active catalysts at such
a low loading were investigated by various characterization
methods, and the low-content Fe-promoting effect was explored
by adjusting the Cu/Fe ratio. Based on reaction kinetics, the
synergistic catalysis of Cu0 and CuFeO2 was presumed to be
responsible for the superior catalytic performance.
The microscopic morphology of catalyst showed in TEM and
SEM images (Figure 1) was similar to as-synthesized SBA-15
by Zhao et al.[13] Figure 1A exhibited the regular mesoporous
structure of SBA-15 with a spacing of 9.43 nm. To judge whether
the small shadow in TEM image was metal particles or not, the
whole macroscopic structure in Figure 1A and three regions
with different topography in it were characterized by TEM-EDS.
The EDS spectrogram shown in Figure S1† displayed a similar
ratio of atomic abundances in four selected regions of Figure 1A,
which demonstrated that the shadow was fragments of catalyst.
The stable distribution of Cu/Fe abundance ratio indicated that a
mixed phase may be formed.
XPS spectra showed in Figure 2 was used to analyze the
metal species on the surface of 2.8Cu-3.5Fe/SBA-15 and
2.8Cu/SBA-15. The deconvolution of Cu 2p3/2 spectra of 2.8Cu-
3.5Fe/SBA-15 showed three peaks. The highest-intensity peak
at 933.7 eV corresponded to Cu0/Cu+ species. The peaks at
935.5 eV and 944.7 eV corresponded to the 2p3/2 orbital binding
energy and satellite peak of Cu2+ species.[10d, 14]
Results and Discussion
Catalyst and species characterization
A series of SBA-15 supported bimetallic Cu-Fe catalysts were
synthesized by co-impregnation method. The actual loadings of
metals in the bulk catalyst were analyzed by ICP and the results
are listed in Table S1†. Actual copper loading of 2.8Cu-
The Cu 2p3/2 spectra of 2.8Cu/SBA-15 displayed a same peak
of Cu2+ species at 935.5 eV, while the peak corresponding to
Cu0/Cu+ species shifted to lower binding energy (933.7→933
eV) compared with 2.8Cu-3.5Fe/SBA-15.
Figure 2. XPS spectra in Cu 2p3/2 and Fe 2p regions and Cu LMM XAES spectra of 2.8Cu-3.5Fe/SBA-15, 2.8Cu/SBA-15 and 3.5Fe/SBA-15.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
Figure 3. Wide angle XRD patterns of (A) SBA-15 and precursors of catalysts (B) Precursor and reduced 28Cu-35Fe/SBA-15.
Cu LMM Auger map (Figure 2) was studied to define Cu0 and
Cu+ species. The calcined 2.8Cu-3.5Fe/SBA-15 showed one
peak of Cu+ species at 914 eV, and the reduced 2.8Cu-
3.5Fe/SBA-15 showed two overlapping peaks at 913.8 eV and
916.1 eV, corresponding to Cu+ species and Cu0, respectively.
After reduction, Cu0 increased the electron density around the
Cu+ species, result in a lower binding energy of Cu+ species.[4a]
According to the deconvolution results of XPS and XAES
spectra listed in Table S2† and Table S3†, 0.4% of the Cu+
species on 2.8Cu-3.5Fe/SBA-15 was converted to Cu0 during
the reduction, while 4.4% was reduced on 2.8Cu/SBA-15. This
data highlighted the Fe-doping effect on strengthening the
stability of Cu+ species.
of 3.5Fe/SBA-15 moved toward lower binding energy
(712.2→710.8 eV).
The higher binding energy of Cu+ species and Fe3+ species on
bimetallic catalyst than monometallic catalysts suggested that
the incorporation of Fe created a metal interaction to alter the
chemical state of Cu species.
The wide angle XRD patterns of SBA-15 supported catalysts
were showed in Figure 3. The broad diffraction peak around 23°
was assigned to silicon-based support. The precursor of 5Cu-
6.56Fe/SBA-15 exhibited weak diffraction peaks at 31.5° and
35.8° in Figure 3A, belonging to the (006) and (012) crystal
planes of CuFeO2, respectively (delafossite, PDF, 00-012-
0752).[16]
A 28Cu-35Fe/SBA-15 sample with higher metal
The deconvolution of Fe 2p spectra of catalysts (Figure 2)
exhibited two main peaks at about 712 eV and 725 eV,
corresponding to the 2p3/2 peak and 2p1/2 peak of Fe3+ species,
respectively, accompanied with the satellite peaks at 718 eV and
733 eV.[14-15] The absent characteristic peaks of Fe2+ species
indicated that there was no Fe3O4 on neither iron-based catalyst.
While the characteristic peaks of Fe3+ species on Fe 2p spectra
loadings was prepared by the same method to better identify the
possible species on catalysts, and distinct diffraction peaks
attributed to CuFeO2 were exhibited in Figure 3B. Given lack of
the peak near 31° on standard XRD patterns of other oxides, it
could be used to recognize CuFeO2.
The HRTEM images of 2.8Cu-3.5Fe/SBA-15 were showed in
Figure 4. The clear lattice fringes were obtained as 0.232 nm,
Figure 4. (A-F) HRTEM images of reduced 2.8Cu-3.5Fe/SBA-15. (G) HADDF-STEM image of same domain in Figure 4F. (H) EDS line scan showing Cu and Fe
distributions along the arrow marked in Figure 4G.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
0.244 nm, 0.367 nm (Figure 4A-C), corresponding to the d-
spacing of (111) plane of CuO,[17] (111) plane of Cu2O[17b] and
(012) plane of Fe2O3.[18] The interplanar spacings of 0.250 nm,
0.224 nm and 0.284 nm (Figure 4D-F) belonged to (012) (104)
and (006) planes of CuFeO2.[16] Besides, no clear lattice fringe of
Cu0 was found in images. The high-angle annular dark-field
scanning TEM (HAADF-STEM) image in Figure 4G was taken
from the same domain in Figure 4F, and the energydispersive
X-ray spectroscopy (EDS) line scan across the individual crystal
lattice structure certificated that Cu and Fe elements coexisted
uniformly throughout the one same nanocrystal lattice and the
bimetallic phase was certainly formed.
SEM-EDX elemental mapping of Cu and Fe on 2.8Cu-
3.5Fe/SBA-15 were presented in Figure 1B. It was distinct that
the high-concentration areas of copper and iron (hi-lite parts of
the image) were almost completely coincident. The combination
of the high coincidence of Cu and Fe elements on EDX-element
mapping, the migrations of standard binding energies in XPS
diagram, the coexistence of Cu and Fe elements on one same
Figure 5. Wide angle XRD patterns of reduced catalysts. SI: Catalyst
prepared by step impregnation method.
lattice proved by EDS line scan and the identification of species
on XRD and HRTEM spectra, it could be determined that the
CuFeO2 conformation existed indeed on the bimetallic Cu-Fe
catalysts.
Table 1. Performance of various catalysts for hydrogenation of EL to GVL by
H2. [a]
Performances of Catalysts
Because of the stability and the versatility of GVL intermediate,
the conversion of EL to GVL was preferentially investigated, and
the catalyst performances on this process were showed in Table
1. 2.8Cu/SBA-15 exhibited excellent EL conversion with only
55.5% yield of GVL and plenty of 4-HPE intermediate left. (entry
1, Table 1). Only 0.4% yield of GVL was obtained on
3.5Fe/SBA-15 indicated that Fe2O3 was not active sites (entry 2,
Table 1). 2.8Cu-3.5Fe/SBA-15 gave the best performance with
90.8% conversion of EL and 89.0% yield of GVL (entry 3, Table
1). While the physical mixture of two monometallic catalysts
performed worse than the former, only 58.2% yield of GVL and
12.8% yield of 4-HPE were observed (entry 4, Table 1). The
result confirmed that the intimate contact of Cu-Fe species was
critical to the catalytic performance. The precursor catalyst gave
only 29.1% yield of GVL owing to lack of Cu0 sites (entry 5,
Table 1). 81.8% conversion of EL and 75% yield of GVL was
provided by 5Cu-6.56Fe/SBA-15, which was worse than 2.8Cu-
3.5Fe/SBA-15 (entry 6, Table 1). The agglomeration of particles
indicted by the appearance of CuFeO2 characteristic peaks in
XRD pattern (Figure 3A) were responsible for the weakened
performance.
Entry
Catalysts
Conv.
(%)
GVL
4-HPE
(%) [b]
C (%)
[c]
(%) [b]
1
2
3
4
2.8Cu/SBA-15
89.4
6.0
55.5
0.4
29.8
——
——
12.8
95.9
94.4
98.2
89.1
3.5Fe/SBA-15
2.8Cu-3.5Fe/SBA-15
90.8
81.9
89.0
58.2
2.8Cu/SBA-15
3.5Fe/SBA-15 [d]
+
5
2.8Cu-3.5Fe/SBA-
15(Precursor)
35.5
29.1
——
93.6
6
7
8
5Cu-6.5Fe/SBA-15
81.8
63.8
48.4
75.0
61.7
36.7
——
——
——
93.2
97.9
88.3
2.8Cu-3.5Fe/SBA-15(SI) [e]
2.8Cu-3.5Fe/SiO2(Quartz
sand)
9
2.8Cu-3.5Fe/MCM-41
67.5
1.9
65.3
0.0
——
——
97.8
98.1
10 [f]
——
The wide angle XRD pattern of reduced catalysts was showed
in Figure 5. Except the diffraction peaks of SiO2 exhibited on
quartz sand supported catalyst, none of catalysts displayed any
apparent peak, indicating that Cu and Fe were uniformly and
continuously dispersed throughout the carrier. Therefore, cost-
effective low-loading catalysts which offered greater atom
efficiency have the ability to achieve higher activity than
conventional high-loading catalysts.
[a] Reaction conditions: EL 72 mg, 10 mL hexane, 130 °C, 3 MPa H2, 50 mg
catalyst, 3 h. [b] Yield of GVL and 4-HPE. [c] Carbon balance: Formula is
shown in supplementary information. [d] Physical mixture of 50 mg of
2.8Cu/SBA-15 and 50 mg of 3.5Fe/SBA-15. [e] SI: Catalyst prepared by step
impregnation method. [f] Blank: same conditions except 3 MPa N2 pressure.
N2 adsorption-desorption experiments were carried out to
investigate the textural properties of various catalysts, and the
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
Figure 6. Effect of Cu/Fe mole ratio on (A) conversion of EL to GVL. Reaction conditions: 72 mg EL, 10 mL hexane, 130 °C, 3 MPa H2, 50 mg catalyst, 3 h. (B)
Conversion of GVL and Selectivity of products. Reaction conditions: 50 mg GVL, 10 mL hexane, 210 °C , 7 MPa H2, 80 mg catalyst, 10 h.
results were showed in Table 2. The BET specific surface area
of SBA-15 up to 718.0 m2 g−1 reflected its abundant mesoporous
structure.[13] Compared with the bare SBA-15, a significant
decrease in BET data and total mesopore volume was observed
on SBA-15 supported catalysts (2.8Cu/SBA-15, 3.5Fe/SBA-15
and 2.8Cu-3.5Fe/SBA-15) (entries 1-4, Table 2), implying that
metals were successfully loaded.[19] The average pore diameters
of 2.8Cu/SBA-15 and 3.5Fe/SBA-15 decreased to 5.4 nm and
5.3 nm respectively compared with bare SBA-15 (6.1 nm).
Similar to the situation reported by Yue et al.,[19] this result
indicated that independent Cu or Fe preferred being filled in the
larger-size pores of the carrier, which reduced the number of
larger-size pores, resulting in a decrease in the average pore
size.[19] While the average pore size of 2.8Cu-3.5Fe/SBA-15
increased to 6.4 nm (SBA-15 6.1nm), even though the total
mesopore volume decreased to 0.89 cm3 g−1. This result
indicated the aggregated bimetallic particles were more likely to
be filled in smaller-size pores, resulting in a decrease in total
mesopore volume but an increase in average pore size. The
pore diameter of the catalyst prepared via step impregnation
method exhibited a similar decrease as those of monometallic
catalysts, which meant the particles of 2.8Cu-3.5Fe/SBA-15(SI)
tended to aggregate and augment (entry 8, Table 2). The above
results further demonstrated the promoting effect of Fe on
reducing particle size and increasing metals dispersion.
2.8Cu-3.5Fe/SBA-15(SI) gave 63.8% conversion of EL and
61.7% yield of GVL, worse than 2.8Cu-3.5Fe/SBA-15 (entry 7,
Table 1). The significant increase in particle size and a lower
dispersion detected by BET method was responsible for the
weakened performance. Catalyst supported by quartz sand with
low specific surface area gave only 48.4% EL conversion (entry
8, Table 1). None of metal-related peak on XRD pattern of
2.8Cu-3.5Fe/Quartz sand indicted that metals remained well
dispersed under low loading, thence 36.7% yield of GVL was still
available. MCM-41 supported catalyst performed worse than
SBA-15, giving 67.5% EL conversion and 65.3% GVL yield
Table 2. Textural properties of SBA-15 and transition metal supported
catalysts.
Entry
Sample
SBET
(m2
)
Pore
Pore
size
(nm) [b]
volume (cm3
g−1
g−1
)
[a]
1
2
3
4
5
6
7
8
SBA-15
718.0
639.6
583.4
571.4
3.2
1.06
0.75
0.71
0.89
0.02
0.68
0.72
0.72
6.1
5.4
5.3
6.4
20.2
2.8
7.1
5.1
2.8Cu/SBA-15
3.5Fe/SBA-15
2.8Cu-3.5Fe/SBA-15
2.8Cu-3.5Fe/SiO2(Quartz sand)
2.8Cu-3.5Fe/MCM-41
Used 2.8Cu-3.5Fe/SBA-15 [c]
2.8Cu-3.5Fe/SBA-15(SI) [d]
800.2
373.1
603.7
(entry 9, Table 1), which forecast that the robust structure of
[13, 20]
SBA-15 was more suitable for this reaction.
In addition,
none of bimetallic catalysts left any intermediates, indicating the
Fe-doping effect on promoting the intramolecular
transesterification reaction and accelerating the reaction process.
The Fe-doping effect.
[a] BJH desorption cumulative volume of pores between 17.0 Å and 3000.0 Å
diameter. [b] BJH desorption average pore diameter (4V/A). [c] after 4 cycles.
[d] SI: Catalyst prepared by step impregnation method.
To further explore the Fe-doping effect on catalyst performances,
four bimetallic catalysts with different Cu/Fe mole ratios were
prepared with the same total metal loading as 6.3 wt% to apply
to the hydrogenation of EL to GVL and conversion of GVL to
1,4-PDO. As shown in Figure 6A, increasing the Fe content
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
Cu/SBA-15 (Not included in figure), which was considered to be
the ring-opening product of GVL.[2c]
Based on the above results, it was obvious that reduction of
Cu-species affected the hydrogenation activity of catalysts, while
the incorporation of iron significantly improved the selectivity of
hydrogenation of carbonyl groups and accelerated the reaction
process. To investigate how the introduction of Fe affected the
reaction, the Fe-doping effect on reduction properties of Cu-
species and the distinctions on bimetallic catalysts with different
Cu/Fe ratios were characterized by TPR and XPS spectra.
The Fe-doping effect on reduction properties of Cu-species
was investigated by H2-TPR tests (Showed in Figure 7). The
curve of 3.5Fe/SBA-15 showed distinct hydrogen adsorption
peaks at 400 °C and 750 °C, representing the reduction of Fe3+
species to Fe0.[21] The curve of 2.8Cu/SBA-15 exhibited two
overlapping peaks between 265 °C and 400 °C, relating to the
continuous reduction of Cu2+ species to Cu+ species and Cu0.[22]
Three peaks arose on the curve of 2.8Cu-3.5Fe/SBA-15, the
peaks at 220 °C and 710 °C represented the reduction of
scattered Cu2+ species and FeOx respectively. The broad peak
Figure 7. H2-TPR profiles of 2.8Cu-3.5Fe/SBA-15, 2.8Cu-3.5Fe/SBA-15 (SI),
2.8Cu/SBA-15 and 3.5Fe/SBA-15. SI: Catalyst prepared by step impregnation
method.
from 1:0.5 to 1:1.5, the EL conversion and GVL yield increased
gradually and reached the highest of 90.8% and 89% at 1:1.5.
Continuously increasing the ratio to 1:2, the conversion of EL
decreased sharply to 15% with 100% selectivity of GVL.
at 460 °C was assigned to the reduction of CuFeO2 to copper
[21-22]
and FeOx.
Compared with 3.5Fe/SBA-15, the intimate
metal interaction on 2.8Cu-3.5Fe/SBA-15 made the reduced Cu0
provide spillover H to reduce FeOx at milder temperature. While
compared with 2.8Cu/SBA-15, the reduction of Cu+ species on
2.8Cu-3.5Fe/SBA-15 moved toward higher temperature, and the
Cu2+ species was opposite. Therefore, the Fe-doping effect
enhanced and suppressed the reduction of Cu2+ species and
Cu+ species, respectively, and the formation of CuFeO2
improved the stability of Cu+ species. Besides, the retained
CuFeO2 and Cu0 sites under the reduction of 300 °C were key
factors for super performance.
Hydrogenation of GVL over bimetallic catalysts with different
Cu/Fe ratio was conducted with Cu/SBA-15 as a reference. As
shown in Figure 6B, the conversion of GVL was significantly
enhanced by the incorporated Fe, and reached the highest
(86%) at Cu/Fe ratio of 1:1.5. Continue to increase to 1:2, the
catalytic activity dropped sharply, but still higher than Cu/SBA-15
(23.6%). The distribution of products was strongly influenced by
the ratio of Cu/Fe. Similar to the tendency of conversion, the
selectivity of 1,4-PDO increased with the growing Fe content,
achieved the highest over 1:1.5 (2.8Cu-3.5Fe/SBA-15), and
reduced suddenly at 1:2. The selectivity of 2-MTHF on bimetallic
catalysts increased slightly compared with Cu/SBA-15, and
underwent an insignificant fluctuation with the varying iron
contents. The selectivity of n-PeOH achieved the highest on a
special ratio of 1:0.5, and decreased gradually with the increase
of iron content. Besides, 12% selectivity of 5-hydroxy-2-
pentanone (HPN) intermediate was detected in the reaction over
XPS spectra of bimetallic catalysts with different Cu/Fe ratios
was shown in Figure 8. The labeled characteristic peaks of Cu+
species or Fe3+ species of all bimetallic catalysts appeared
significant skewing compared with monometallic catalysts. As
the ratio of incorporated iron improved, the binding energies of
both Cu+ species and Fe3+ species was enhanced gradually,
which indicated that the content of CuFeO2 was gathering. The
deconvolution results showed in the Cu 2p3/2 map demonstrated
Figure 8. XPS spectra in Cu 2p3/2 and Fe 2p regions of calcined samples with different Cu/Fe ratios.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
Table 3. Effect of solvents on hydrogenation of EL to GVL. [a]
Entry
Solvents
Conv.
(%)
Yield (%)
C
(%)
1
2
3
4
5
6
7
Isopropanol
Ethanol
8.79
7.84
8.32
13.0
15.3
34.8
97.0
4.2
95.4
95.5
95.1
91.3
94.8
94.9
98.8
3.3
Tetrahydrofuran
Ethyl acetate
Water
3.4
4.3
10.1
29.7
95.8
Figure 9. Effect of catalyst loading on EL conversion (blue) and GVL yield
(red). Reaction conditions: 72 mg EL, 10 mL hexane, 140 °C, 3 MPa H2, 50
mg 2.8Cu-3.5Fe/SBA-15, 3 h.
Water + Hexane
Hexane
[a] Reaction conditions: 72 mg EL, 10 mL hexane, 140 °C, 3 MPa H2, 50 mg
2.8Cu-3.5Fe/SBA-15, 3 h.
Table 4. Effect of temperature and H2 pressure on hydrogenation of EL to
GVL. [a]
Entry
Temperature/°C
P(H2)/MPa
Conv. (%)
99.2
Yield (%)
97.0
1
2
3
4
5
6
7
8
150
140
130
120
110
140
140
140
3
3
97.0
95.8
3
90.8
89.0
3
58.2
55.9
Figure 10. Effect of reaction time on EL conversion, GVL yield and C-balance.
Reaction conditions: 72 mg EL, 10 mL hexane, 140 °C, 3 MPa H2, 50 mg
2.8Cu-3.5Fe/SBA-15, 3 h.
3
31.6
29.6
2.5
2
90.5
88.9
83.9
81.6
in pure hexane (entries 5-7, Table 3). Water addition promoted
H-spillover on the catalyst surface,[10a] but H2 has much lower
solubility in water than organic solvents, and the conformation of
SBA-15 is easier to be damaged in aqueous solution due to
surface hydroxyl groups,[7, 10a, 23] resulting in more serious metal
leaching. Reducing the water content in system could reserve
1
61.7
60.1
[a] Reaction conditions: 72 mg EL, 10 mL hexane, 50 mg 2.8Cu-3.5Fe/SBA-
15, 3 h.
the active sites maximally. In addition, the stronger adsorption of
[24]
oxygen-containing groups
and the promotion of hydrogen
that the Fe-doping effect promoted the formation and increased
the proportion of Cu+ species.
Effect of reaction conditions
activation on silica supports in alkane solvents than water or
[25]
alcohol
reported in several works were critical factors for
obtaining the better GVL yields in n-hexane solvent.
Hydrogenation of EL to GVL. Because of the stability and the
versatility of GVL intermediate, the effect of reaction conditions
on the conversion of EL to GVL was preferentially investigated,
and the solvent effect was showed in Table 3. Among various
agents evaluated, the yield of GVL was no more than 5% in
isopropanol, ethanol, tetrahydrofuran and ethyl acetate (entries
1-4, Table 3). 15.3% conversion of EL and a low yield of GVL
were obtained in pure water, increased to 34.8% and 29.7% in
50 vol% of hexane-water biphasic solvent, and reached highest
Temperature and H2 pressure influences were investigated
and showed in Table 4. As increasing the temperature, the
conversion of EL gradually improved from 31.6% at 110 °C, and
reached a maximum of 99.2% at 150 °C (entries 1-5, Table 4).
Since the dissociation of hydrogen on Cu0 sites is
thermodynamically endothermic, higher temperature helped to
stride the energy barrier of H2 activation.[26] In addition, the only
product GVL and a considerable carbon balance remained
stable in the range of 110-150 °C. This Cu-Fe catalyst is an
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
Figure 11. Effect of temperature or reaction time on conversion of GVL and selectivity of products. Reaction conditions: 50 mg GVL, 10 mL hexane, 7 MPa H2,
80mg catalyst. Optional 210 °C or 10 h.
excellent option for production of GVL in lower temperature
range, giving 95.8% yield after 3 hours at 140 °C.
increase from 50 mg to 70 mg, the conversion of EL stopped
growing, while the yield of GVL decreased slightly.
The conversion of EL and the yield of GVL rapidly improved
from 61.7% and 60.1% with the increasing H2 pressure from 1
MPa to 3 MPa (entries 7-9, Table 4). None of by-products were
detected during the process. The instant hydrogenation of
carbonyl groups was affected by the concentration of H2
dissolved in the solvent, leading to a weaker activity at
pressures below 2 MPa.
The effect of catalyst loading on reaction efficiency was
investigated and showed in Figure 9. When increasing catalyst
loading from 30 mg to 50 mg, the conversion of EL and the yield
of GVL raised significantly from 56.1% and 43.6%, respectively,
and achieved the maximum of 96.6% and 96%. Continue to
Reaction kinetics was explored and showed in Figure 10, the
conversion of EL was only 45.6% after 10 minutes and
increased rapidly to 92% in the next 50 minutes. Continue
extending to 240 minutes, the gradually increased EL
conversion reached the maximum of 97.4%. Under the rich
carbon balance, GVL was the only product throughout the period,
which emphasized that the Fe-doping effect on promotion of
intramolecular transesterification to accelerate the reaction
process.
Production of 1,4-PDO. Effect of reaction conditions on the
production of 1,4-PDO was explored and showed in Table 5. As
increasing the catalyst loading from 65 mg to 95 mg, the
Table 5. Effect of catalyst loading and H2 pressure on hydrogenation of EL or GVL to 1,4-PDO. [a]
Entry
Substrate
Catalyst
P(H2)/MPa
Conv. (%)
Selectivity (%)
C (%)
loading / mg
1,4-PDO
91.1
n-PeOH
0.0
2-MTHF
2.2
GVL
——
——
——
——
——
19.4
1
2
3
4
5
6
GVL
GVL
GVL
GVL
GVL
EL
65
80
95
80
80
80
7
7
7
6
8
7
73.9
85.8
91.2
70.5
86.4
99.3
95.0
96.6
95.8
94.7
97.0
91.9
92.6
1.0
2.4
88.0
2.2
5.2
89.7
0.4
2.7
91.3
3.1
2.3
64.9
1.5
7.0
[a] Reaction conditions: 0.5 mmol substrate, 2.8Cu-3.5Fe/SBA-15 catalyst, 10 mL hexane, 210 °C, 10 h.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
Figure 12. (A) NH3-TPD profile of SBA-15 and reduced catalysts. (B) FTIR of pyridine desorption at 150 °C and 250 °C on SBA-15 and 2.8Cu-3.5Fe/SBA-15.
conversion of GVL along with the selectivity of n-PeOH and 2-
MTHF were improved to 91.2%, 2.2% and 5.2%, respectively.
While the selectivity of 1,4-PDO rose to the maximum of 92.6%
at 80 mg of catalyst loading and decreased to 88% at an
excessive loading of 90 mg (entries 1-3, Table 5). Both the
selectivity of n-PeOH and catalytic activity were enhanced to
3.1% and 86.4%, respectively by the higher H2 pressure, while
the selectivity of 2-MTHF was impervious to the changing
pressures and kept in 2.5% or so. The selectivity of 1,4-PDO
was 89.7% at 6 MPa, and achieved the maximum at 7 MPa,
while suffered a decrease to 91.3% at 8 MPa (entries 2, 4-5,
Table 5). The most suitable conditions of 80 mg of catalyst
loading and
7 MPa H2 pressure were applied to the
Scheme 2. Synergistic catalytic mechanism for hydrogenation of EL to 1,4-
PDO by Cu0 and CuFeO2.
hydrogenation of EL to 1,4-PDO (entry 6, Table 5). Upon an
exhaustive conversion of EL, 19.4% selectivity of remaining GVL
intermediate coupled with a sharp increased selectivity of 2-
MTHF led to a significant decrease in 1,4-PDO selectivity to
64%. While the change in the selectivity of n-PeOH was not
obvious.
[27]
to the hydrogen-bonded pyridine desorption.
These two
broad bands disappeared at 250 °C , and rare acidity was
detected on SBA-15. The characteristic bands at 1457 cm-1 and
1618 cm-1 on the curve of 2.8Cu-3.5Fe/SBA-15 were attributed
to Lewis acid sites. The weak bands appeared at 1540 cm-1 and
1640 cm-1 corresponding to the adsorption of pyridine on the
Brønsted acid sites. The medium band at 1490 cm-1 was
associated with the both acid sites.
An objective view on the reaction pathway can be inferred
from the kinetic analysis. Given the conclusion that abundant
Lewis acid sites were generated on 2.8Cu-3.5Fe/SBA-15 due to
the Fe-doping effect, a synergistic catalytic mechanism for
hydrogenation of EL to 1,4-PDO by Cu0 and CuFeO2 was
presumed and showed in Scheme 2: First, H2 was adsorbed
and dissociated into independent H atoms on Cu0 sites. The
adjacent CuFeO2 sites acted as Lewis acid sites to activate the
carbonyl groups of EL. H atoms overflowed from Cu0 to the
adsorbed oxygen atom of carbonyl groups, where a σ bond was
formed to yield an intermediate 4-HPE. Then the 4-HPE was
rapidly cyclized to afford GVL. The carbonyl groups of GVL were
next adsorbed and reduced on CuFeO2 sites, the obtained 2-
hydroxy-5-methyltetrahydrofuran was further converted to HPN
through ring-opening isomerization. Finally, the carbonyl groups
of HPN were hydrogenated to 1,4-PDO on CuFeO2 sites.
Effect of temperature and reaction time on the hydrogenation
of GVL to 1,4-PDO was showed in Figure 11. The escalating
trend of conversion of GVL was similar to that of the selectivity
of 2-MTHF and n-PeOH, which grew gradually with increasing
temperature or reaction time, and achieved their maximums at
220 °C or 13 h. The selectivity of 1,4-PDO kept exceeding 93%
at temperatures below 210 °C , and declined to 88% at 220 °C .
During the first hour of an entire reaction, 13% selectivity of HPN
intermediate was detected and the selectivity of 1,4-PDO was
only 70%. As the reaction progressed, HPN was reduced
continuously and completely consumed after 6 h. While the
selectivity of 1,4-PDO reached 94% simultaneously, kept the
dominant position within 7-11 hours, and then declined after 11
h due to the rapid accumulation of by-products.
Synergistic effect of Cu0 and CuFeO2
To investigate how the introduction of Fe to Cu-based catalysts
affected the catalytic process, the surface acidity of catalysts
was detected by NH3-TPD and FTIR spectra of pyridine
desorption. As showed in Figure 12B, when samples were
outgassed at 150 °C , both SBA-15 and 2.8Cu-3.5Fe/SBA-15
exhibited two strong bonds at 1445cm-1 and 1595 cm-1 relating
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
reflecting the (100) crystal plane of the microstructure of SBA-15.
Meanwhile, two smaller diffraction peaks were observed at 1.5°
and 1.7° corresponding to the (110) and (200) crystal plane of
secondary structure, and the d(100) value remained fixed and
matched well with Figure 1A. While the intensity of three
diffraction peaks decreased after recycling, especially the hardly
distinguishable (110) and (200) peaks. TEM image of used
catalyst showed in Figure 13 revealed that the internal
mesoporous structure in carrier remained, while the port of
channels had amorphized, which matched the result that a
substantial drop in the BET specific surface area of used
catalyst (Table 1).
Approximately 9.6% copper loss of used catalyst was
determined by ICP (Table S1), while iron was much weaker,
only 1.9% iron loss had been checked. The copper loss and the
partially destroyed mesoporous structure answered for the
sharply decrease in conversion. While another noteworthy was
that the H2 reactivation after the 2nd run significantly regenerated
the catalyst activity and brought about an uptrend of EL
conversion in the 3rd run, with addition to that scanty surface Cu0
sites existed on used catalyst shown in the deconvolution results
of Cu LMM XAES spectra (Table S2†), it could be inferred that
the decline in catalyst activity attributed to the loss of Cu0 sites.
XPS was used to analyze the changes in metal species.
Table S2† exhibited a much higher content of Cu2+ species and
a lower Cu0 content on the used catalyst compared with the
fresh. As showed in Figure 14, in contrast to the fresh catalyst,
the characteristic peaks of both Cu+ and Fe3+ species on used
catalyst shifted toward lower binding energy, getting closer to
the monometallic catalysts, which revealed that the partial
dissociation of CuFeO2 weakened the intimate metals interaction.
Figure 13. Small angle XRD patterns of SBA-15, reduced and used 2.8Cu-
3.5Fe/SBA-15. The insert was TEM image of used catalyst.
Recycling of catalyst
To evaluate the stability of this bimetallic catalyst, 2.8Cu-
3.5Fe/SBA-15 was used for 4 cycles of hydrogenation of EL to
GVL (50 mg Cat, 140 °C, 3 MPa, 3 h). A detailed description of
recycling was showed in Supplementary Information. Briefly, the
catalyst was separated and charged into reactor without H2
activation except before the 3rd run. As showed in Figure S4†,
the conversion declined sharply from 96% to 70% in the 2nd run
and rose to 88.7% after the 3rd run because of activation again,
then fell to 62.8% in the 4th run. Meanwhile, the selectivity of
GVL underwent a slight decline to 88.5%, but still much higher
than the single run of 2.8Cu/SBA-15 (62%).
Accordingly, the used bimetallic catalyst approximated
a
physical mixture of two monometallic catalysts, resulting in the
slight decrease in selectivity.
The used catalyst after 4 cycles was calcined at 550 °C for 3 h
to remove the potential carbon deposits. Besides the catalyst
loss during recovery, the eventual changes in features were
assessed by various characterizations.
Conclusions
The changes in structure were characterized by N2 adsorption
desorption, TEM and XRD. The small angle XRD pattern
showed in Figure 13 exhibited a strong diffraction peak at 0.9°,
A series of bimetallic Cu-Fe catalysts with low metal loading
were prepared by co-impregnation method for the continuous
Figure 14. XPS spectra in Cu 2p3/2 and Fe 2p regions of 2.8Cu/SBA-15, 3.5Fe/SBA-15 and used 2.8Cu-3.5Fe/SBA-15.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
Step impregnation method was used for a reference. In a typical
procedure to synthesize 2.8Cu-3.5Fe/SBA-15 (SI) (Cu/Fe ratio=1:1.5):
Fe(NO3)3·9H2O (0.404 g) was dissolved in acetone (30 mL), then add
SBA-15 (1.5 g) into the solution and stirring at room temperature
overnight. After removing the acetone, calcined in air at 350 °C for 6 h.
hydrogenation of EL to 1,4-PDO. Among them, 2.8Cu-
3.5Fe/SBA-15 with a Cu/Fe molar ratio of 1:1.5 exhibited the
best performance, capable of converting EL into a key GVL
intermediate at 140 °C with a yield of up to 97%. It is also
capable of continuously hydrogenating GVL to 1,4-PDO at
210 °C with a selectivity of up to 92.6%, and it can be used to
convert EL to 1,4-PDO and GVL in one-pot with a total yield of
85%. In this work, the high activity of cheap copper at such a low
loading to rival those of noble metal catalysts can be attributed
to the Fe doping effect and the high dispersion of the active
metal species. The doping effect of low content Fe on structure
and microscopic properties of catalyst was considered by
various characterization methods. The results indicated that Fe
in the bimetallic system can act as a structural promoter to
reduce particle size and increase metal dispersion, and can also
be used as an electronic modifier to promote and inhibit the
reduction of Cu2+ species and Cu+ species, respectively, as well
as increasing the stability and proportion of Cu+ species. In
addition, the incorporation of Fe provided plenty of Lewis acid
sites to enhance the adsorption of carbonyl groups, and
promoted intramolecular transesterification to accelerate the
reaction process. The presence of the bimetallic oxide phase
CuFeO2 under the condition of low metal loading was detected
by the high coincidence of Cu and Fe elements on EDX-element
mapping, the migration of binding energy in XPS diagram, the
coexistence of Cu and Fe elements on the one same lattice
proved by EDS line scan and identification of species on XRD
and HRTEM spectra. Combining characterizations with reaction
kinetics, the synergistic catalysis between Cu0 and CuFeO2
species is considered to be the key to ensure excellent
performance of the bimetallic catalyst. Our work will provide
ideas for the design and preparation of low-loading, high-activity
hydrogenation base metal catalysts.
Then put it into
a acetone solution with appropriate amount of
Cu(NO3)2·3H2O. Stirred overnight, calcined in air at 350 °C for 6 h and
reduced at 300 °C for 2 h. The catalyst obtained was labeled as 2.8Cu-
3.5Fe/SBA-15 (SI).
Characterization techniques: N2 adsorption-desorption experiments
were performed on the Micromeritics ASAP 2020 analyzer. ICP-OES was
measured on Varian Vista-MPX analyzer. XRD patterns were obtained
on the PANalytical X'pert diffractometer. TEM and HRTEM images were
obtained from a JEOL-2010 electron microscope. EDS analysis was
done on JEOL X-Ray TEM 200 micro-analysis system. SEM graphs and
EDX-elemental mapping were obtained using
a GeminiSEM 500
electron microscope. XPS and XAES were carried out on the Thermo
Scientific Escalab 250 X-ray photoelectron spectrometer. TPR and NH3-
TPD were measured on a home-built adsorption instrument coupled with
a gas chromatograph. FTIR of pyridine adsorption was performed on a
FTIR spectrometer (Perkin-Elmer Model 170-SX). More information was
showed in ESI.
Catalytic test and the analysis of products: Substrate (0.5 mmol),
catalyst and solvent (10 mL) were added into the autoclave (25 mL,
Kemi) and then purged by hydrogen. After the reaction, the mixture was
collected by ethyl acetate. The composition of product was determined
by Shimadzu GC-2010 plus gas chromatograph equipped with RTX-65
column, and anisole was subjoined as an internal standard.
Recycling and Reactivation of catalyst: After the 1st cycle, the catalyst
was separated by centrifugation and washed several times with ethyl
acetate and hexane, then directly charged into the fresh feed without
reactivation. After the 2nd run, the recovered catalyst was dried at 90 °C
for 12 h, and reduced at 300 °C for 2 h in a hydrogen-nitrogen mixed
atmosphere for a reactivation. Then the catalyst was reused for the 3rd
run. After a completed 3rd run, the catalyst was washed and put into the
4th run without reactivation.
Experimental Section
Synthesis of SBA-15. According to the co-condensation method
introduced by Ref. [11]. 4 g of P123 (Sigma-Aldrich) was added into 120
mL of HCl solution (2 mol L-1) under stirring. After maintaining at 40 °C
for 4 h, add into 8.5 g of TEOS (Sinopharm). Stirring at 40 °C for 24 h,
the emulsion was transferred to a polytetrafluoroethylene hydrothermal
reactor (Kemi) and crystallized at 100 °C for 24 h. After filtration and
washing, the solid was dried at 50 °C and calcined at 550 °C for 6 h.
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (21572212, 51821006, 51961135104), the
National Key R&D Program of China (2018YFB1501604), the
Major Science and Technology Projects of Anhui Province
(18030701157), the Strategic Priority Research Program of the
CAS (XDA21060101), and the Local Innovative and Research
Teams Project of Guangdong Pearl River Talents Program
(2017BT01N092). The authors thank the Hefei Leaf Co., Ltd and
Anhui Kemi Machinery Technology Co., Ltd for free samples and
equipment that helped us conduct this study.
Synthesis of SBA-15 supported catalysts.
Co-impregnation method applied to all supported catalysts. For a
typical procedure to synthesize 2.8Cu-3.5Fe/SBA-15 (Cu/Fe mole ratio of
1:1.5): Cu(NO3)2·3H2O (0.1904 g) and Fe(NO3)3·9H2O (0.404 g)
dissolved in acetone (30 mL), then put in SBA-15 (1.5 g) and stirred at
room temperature. After rotary evaporation, the solid was calcined at
350 °C for 6 h and reduced at 300 °C for 2 h. The catalysts were labeled
as xCu-yFe/SBA-15, in which x and y represented the mass fractions of
element copper and iron, respectively (based on catalyst). xCu-
yFe/MCM-41, xCu-yFe/SiO2 were prepared as a reference using above-
mentioned method.
Keywords: biomass • Fe-doping effect • heterogeneous
catalysis • hydrogenation of carbonyl groups • 1,4-pentanediol
[1]
a) P. Gallezot, Chem. Soc. Rev. 2012, 41, 1538-1558; b) C. H.
Zhou, X. Xia, C. X. Lin, D. S. Tong, J. Beltramini, Chem. Soc.
Rev. 2011, 40, 5588-5617.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
[2]
a) T. Mizugaki, Y. Nagatsu, K. Togo, Z. Maeno, T. Mitsudome,
S. K. Bhargava, A. Venugopal, Appl. Catal. A Gen. 2017, 531,
169-176.
K. Jitsukawa, K. Kaneda, Green Chem. 2015, 17, 5136-5139;
b) S. C. Patankar, G. D. Yadav, ACS Sustainable Chem. Eng.
2015, 3, 2619-2630; c) D. Sun, T. Saito, Y. Yamada, X. Chen,
S. Sato, Appl. Catal. A Gen. 2017, 542, 289-295.
a) D. M. Alonso, S. G. Wettstein, J. A. Dumesic, Green Chem.
2013, 15, 584-595; b) K. Yan, Y. Yang, J. Chai, Y. Lu, Appl.
Catal. B Environ. 2015, 179, 292-304.
a) J. Wu, G. Gao, P. Sun, X. Long, F. Li, ACS Catal. 2017, 7,
7890-7901; b) K.-i. Shimizu, S. Kanno, K. Kon, Green Chem.
2014, 16, 3899-3903; c) L. Corbel-Demailly, B. K. Ly, D. P.
Minh, B. Tapin, C. Especel, F. Epron, A. Cabiac, E. Guillon, M.
Besson, C. Pinel, ChemSusChem 2013, 6, 2388-2395.
a) Z. Li, M. Zuo, Y. Jiang, X. Tang, X. Zeng, Y. Sun, T. Lei, L.
Lin, Fuel 2016, 175, 232-239; b) A. Yin, X. Guo, W. Dai, K.
Fan, J. Phys. Chem. C 2009, 113, 11003–11013; c) Y. Wang,
Y. Shen, Y. Zhao, J. Lv, S. Wang, X. Ma, ACS Catal. 2015, 5,
6200-6208; d) I. Obregón, I. Gandarias, A. Ocio, I. García-
García, N. A. d. Eulate, P. L. Arias, Appl. Catal. B Environ.
2017, 210, 328-341.
[13]
[14]
D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F.
Chmelka, G. D. Stucky, Science 1998, 279, 548-552.
S.-Y. Zhao, B. Zhang, H. Su, J.-J. Zhang, X.-H. Li, K.-X. Wang,
J.-S. Chen, X. Wei, P. Feng, J. Mater. Chem. A 2018, 6, 4331-
4336.
C. Pereira, A. M. Pereira, P. Quaresma, P. B. Tavares, E.
Pereira, J. P. Araujo, C. Freire, Dalton Trans. 2010, 39, 2842-
2854.
a) Y. Dong, C. Cao, Y. S. Chui, J. A. Zapien, Chem. Commun.
2014, 50, 10151-10154; b) Y. Oh, W. Yang, J. Tan, H. Lee, J.
Park, J. Moon, Nanoscale 2018, 10, 3720-3729.
a) K. Panigrahi, S. Das, S. Saha, B. Das, D. Sen, P. Howli, K.
K. Chattopadhyay, CrystEngComm 2016, 18, 3389-3398; b) Z.
Wang, S. Zhao, S. Zhu, Y. Sun, M. Fang, CrystEngComm
2011, 13, 2262-2267.
[3]
[4]
[15]
[16]
[17]
[5]
[18]
[19]
[20]
[21]
[22]
S. Zhu, M. Chen, W. Ren, J. Yang, S. Qu, Z. Li, G. Diao, New
J. Chem. 2015, 39, 7923-7931.
H. Yue, Y. Zhao, S. Zhao, B. Wang, X. Ma, J. Gong, Nat.
Commun. 2013, 4, 2339-2345.
H.-P. Lin, C.-Y. Tang, C.-Y. Lin, J. Chin. Chem. Soc. 2002, 49,
981-988.
[6]
I. Obregon, I. Gandarias, N. Miletic, A. Ocio, P. L. Arias,
ChemSusChem 2015, 8, 3483-3488.
[7]
[8]
A. M. Hengne, C. V. Rode, Green Chem. 2012, 14, 1064-1072.
a) N. Pino, S. Sitthisa, Q. Tan, T. Souza, D. López, D. E.
Resasco, J. Catal. 2017, 350, 30-40; b) T. Zhang, J. Liu, D.
Wang, Z. Zhao, Y. Wei, K. Cheng, G. Jiang, A. Duan, Appl.
Catal. B Environ. 2014, 148-149, 520-531; c) A. V.-H. Soares,
G. Perez, F. B. Passos, Appl. Catal. B Environ. 2016, 185, 77-
87; d) M. Tamura, D. Yonezawa, T. Oshino, Y. Nakagawa, K.
Tomishige, ACS Catal. 2017, 7, 5103-5111.
Y. H. Choi, Y. J. Jang, H. Park, W. Y. Kim, Y. H. Lee, S. H.
Choi, J. S. Lee, Appl. Catal. B Environ. 2017, 202, 605-610.
M. P. Yeste, H. Vidal, A. L. García-Cabeza, J. C. Hernández-
Garrido, F. M. Guerra, G. A. Cifredo, J. M. González-Leal, J.
M. Gatica, Appl. Catal. A Gen. 2018, 552, 58-69.
M. Sudhakar, V. V. Kumar, G. Naresh, M. L. Kantam, S. K.
Bhargava, A. Venugopal, Appl. Catal. B Environ. 2016, 180,
113-120.
Y. Nakagawa, K. Mori, K. Chen, Y. Amada, M. Tamura, K.
Tomishige, Appl. Catal. A Gen. 2013, 468, 418-425.
A. P. Palermo, S. Zhang, S. J. Hwang, D. A. Dixon, B. C.
Gates, A. Katz, Dalton Trans. 2018, 47, 13550-13558.
B. Bridier, J. Pe´rez-Ramı´rez, J. Am. Chem. Soc. 2010, 132,
4321–4327.
a) M. L. Martinez, M. B. Gomez Costa, O. A. Anunziata,
Nanotechnology 2019, 30, 065703; b) L. Chmielarz, P.
Kuśtrowski, M. Drozdek, M. Rutkowska, R. Dziembaj, M.
Michalik, P. Cool, E. F. Vansant, J. Porous Mater. 2010, 18,
483-491; c) J. M. Ramos, J. A. Wang, S. O. Flores, L. F. Chen,
N. Nava, J. Navarrete, J. M. Domínguez, J. A. Szpunar, Catal.
Today 2018.
[23]
[9]
J. K. Kim, J. K. Lee, K. H. Kang, J. C. Song, I. K. Song, Appl.
Catal. A Gen. 2015, 498, 142-149.
[24]
[25]
[26]
[27]
[10]
a) J. Tan, J. Cui, T. Deng, X. Cui, G. Ding, Y. Zhu, Y. Li,
ChemCatChem 2015, 7, 508-512; b) Y. Kuwahara, W.
Kaburagi, Y. Osada, T. Fujitani, H. Yamashita, Catal. Today
2017, 281, 418-428; c) I. Obregón, E. Corro, U. Izquierdo, J.
Requies, P. L. Arias, Chin. J. Catal. 2014, 35, 656-662; d) B.
Cai, X.-C. Zhou, Y.-C. Miao, J.-Y. Luo, H. Pan, Y.-B. Huang,
ACS Sustainable Chem. Eng. 2016, 5, 1322-1331.
Q. Xu, X. Li, T. Pan, C. Yu, J. Deng, Q. Guo, Y. Fu, Green
Chem. 2016, 18, 1287-1294.
a) F. M. Geilen, B. Engendahl, A. Harwardt, W. Marquardt, J.
Klankermayer, W. Leitner, Angew. Chem. Int. Ed. 2010, 49,
5510-5514; b) V. V. Kumar, G. Naresh, S. Deepa, P. G.
Bhavani, M. Nagaraju, M. Sudhakar, K. V. R. Chary, J. Tardio,
[11]
[12]
This article is protected by copyright. All rights reserved.

10.1002/cssc.201901198
ChemSusChem
FULL PAPER
FULL PAPER
Selective production of 1,4-
Tianyu Deng, Long Yan, Xinglong Li and
Yao Fu*
pentanediol over Cu-Fe/SBA-15: an
efficient, inexpensive and selective
heterogeneous bimetallic catalyst,
which can flexibly convert biomass-
derived ethyl levulinate into high-value
-valerolactone or 1,4-pentanediol in
varying conditions. The Fe-promoting
effect and the formed CuFeO2
Page No. – Page No.
Continuous Hydrogenation of Ethyl
Levulinate to 1,4-Pentanediol over
2.8Cu-3.5Fe/SBA-15 Catalyst at Low
Loading: The Effect of Fe Doping
conformation are responsible for the
high reactivity.
This article is protected by copyright. All rights reserved.