Hydrogenation of γ-Butyrolactone to 1,4-Butanediol over CuCo/TiO2 Bimetallic Catalysts

Article abstract of DOI:10.1021/acscatal.7b03016

Titania-supported monometallic and bimetallic Cu-Co catalysts were prepared using (co)impregnation and studied for the hydrogenation of γ-butyrolactone (GBL) to 1,4-butanediol (BDO) at temperatures from 100 to 180 °C and a hydrogen pressure of 3.4 MPa. The highest catalytic activity occurred at a Cu:Co atomic ratio of 1:9 (Cu_0.1Co_0.9/TiO₂), and a 95% yield of BDO was obtained. Characterization results showed mainly small nanoparticles (average size 2.6 nm) for pure Cu/TiO₂, large particles (～19.8 nm) for pure Co/TiO₂, and a bimodal particle size distribution of both small (～2.3 nm) and large (～16.5 nm) particles for the bimetallic catalyst with a Cu:Co ratio of 1:1. The addition of ～10 mol % Cu to Co/TiO₂ increased the reducibility of the Co and resulted in the formation of core-shell CuCo bimetallic nanoparticles with a Co-rich core and Cu-rich shell. GBL hydrogenation in liquid ethanol and water produced an ester (ethyl 4-hydroxybutanoate) and a carboxylic acid (4-hydroxybutanoic acid) as the major products, respectively. GBL hydrogenation in 1,4-dioxane likely went through a 2-hydroxytetrahydrofuran (2-HTHF) intermediate. The 2-HTHF underwent facile ring-opening tautomerization to 4-hydroxybutanal (4-HB), followed by rapid hydrogenation to BDO at a reaction rate up to 700 times faster than GBL hydrogenation. The Cu_0.1Co_0.9/TiO₂ catalyst maintained the BDO selectivity and about 80% of initial activity for GBL hydrogenation after 150 h time on stream in a continuous flow reactor.

Full text of DOI:10.1021/acscatal.7b03016

Subscriber access provided by READING UNIV
Article
Hydrogenation of #-butyrolactone to 1,4-
butanediol over CuCo/TiO2 bimetallic catalysts
Zhiwei Huang, Kevin J. Barnett, Joseph Paul Chada, Zachary James
Brentzel, Zhuoran Xu, James A. Dumesic, and George W. Huber
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b03016 • Publication Date (Web): 31 Oct 2017
Downloaded from http://pubs.acs.org on November 1, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
Hydrogenation of γꢀbutyrolactone to 1,4ꢀbutanediol over CuCo/TiO2
bimetallic catalysts
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Zhiwei Huang#, Kevin J. Barnett#, Joseph P. Chada, Zachary J. Brentzel, Zhuoran Xu, James A. Dumesic,
George W. Huber
*
Department of Chemical and Biological Engineering, University of WisconsinꢀMadison, Madison, WI 53706, USA
#These authors contributed equally to this work
*Corresponding author，Eꢀmail: gwhuber@wisc.edu
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 29
1
2
3
4
Abstract:
5
6
7
8
Titaniaꢀsupported monometallic and bimetallic Cu–Co catalysts were prepared using
(co)impregnation and studied for hydrogenation of γꢀbutyrolactone (GBL) to 1,4ꢀbutanediol (BDO)
o
9
at temperatures from 100 to 180 C and a hydrogen pressure of 3.4 MPa. The highest catalytic
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
activity occurred at a Cu:Co atomic ratio of 1:9 (Cu0.1Co0.9/TiO2) and a 95% yield of BDO was
obtained. Characterization results showed mainly small nanoparticles (average size 2.6 nm) for pure
Cu/TiO2, large particles (~19.8 nm) for pure Co/TiO2, and a bimodal particle size distribution of both
small (~2.3 nm) and large (~16.5 nm) particles for the bimetallic catalyst with Cu:Co ratio of 1:1.
The addition of ~10 mol% Cu to Co/TiO2 increased the reducibility of the Co and resulted in the
formation of coreꢀshell CuCo bimetallic nanoparticles with a Coꢀrich core and Cuꢀrich shell. GBL
hydrogenation in liquid ethanol and water produced an ester (ethyl 4ꢀhydroxybutanoate) and a
carboxylic acid (4ꢀhydroxybutanoic acid) as the major products, respectively. GBL hydrogenation in
1,4ꢀdioxane likely went through a 2ꢀhydroxytetrahydrofuran (2ꢀHTHF) intermediate. The 2ꢀHTHF
underwent facile ringꢀopening tautomerization to 4ꢀhydroxybutanal (4ꢀHB), followed by rapid
hydrogenation to BDO at a reaction rate up to 700 times faster than GBL hydrogenation. The
Cu0.1Co0.9/TiO2 catalyst maintained the BDO selectivity and about 80% of initial activity for GBL
hydrogenation after150 h timeꢀon stream in a continuous flow reactor.
Keywords：CuCo bimetallic catalyst, hydrogenation, γꢀbutyrolactone, 1,4ꢀbutanediol, reaction
pathway and mechanism
1. Introduction
1,4ꢀButanediol (BDO) is an industrially important chemical that can be used to produce many basic
chemicals such as tetrahydrofuran (THF), polybutylene terephthalate (PBT), polybutylene succinate
(PBS), polyurethane, and polyester polyols.1,2 The global BDO market was valued at 6.19 billion
USD in 2015 and is expected to grow at a Compound Annual Growth Rate of 7.7% in the next 8
years.1 BDO is commercially produced by three major routes: 1) hydrogenation of butynediol
produced via carbonylation of acetylene with formaldehyde (Reppe process), 2) hydrogenation of
2
ACS Paragon Plus Environment

Page 3 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
maleic anhydride (MA), and 3) acetoxylation of butadiene.3,4 Hydrogenation of MA (made via
oxidation of nꢀbutane) is attracting growing interest due to its safe operational advantages and the
low cost of nꢀbutane feedstock.3,5,6 The hydrogenation of MA or MAꢀderived dialkyl esters to
produce BDO has been reported to proceed via the reaction intermediate γꢀbutyrolactone (GBL) over
various noble metal and copper catalysts at temperatures and pressures between 170ꢀ260˚C and 3ꢀ7
MPa, respectively.5ꢀ10 Combined BDO and GBL selectivities above 94% are obtained from
MAꢀderived dialkyl esters with Cuꢀbased catalysts.5, 9,10
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1 Reaction route for the production of BDO from petroleumꢀ and biomassꢀ derived GBL.
BDO can also be produced from renewable platform molecules including succinic acid (SA) and
furfural, as shown in Scheme 1.2,4,11ꢀ16 The annual market volume for BDO (2000 kton/yr) is larger
than the current volume for SA (60 kton/yr) and furfural (604 kton/yr), therefore it would be highly
desirable to develop efficient technologies for production of BDO from SA and furfural. The
hydrogenation of SA and the oxidationꢀhydrogenation of furfural to BDO proceed via the same GBL
intermediate (Scheme 1).2,4,11ꢀ17 Many noble metals, including Rh, Pt, Pd and Ru, have been studied
for the direct SA hydrogenation into BDO.12,13,15,16 Rh/C, Pt/C, and Pd/C gave high selectivities to
BDO (70ꢀ90%) at above 60% conversions, while Ru produced large amounts of THF.15 Re has been
used as an additive for Ru and Pd catalysts in SA hydrogenation to promote the conversion of the
GBL intermediate to BDO, increasing the rate of BDO production by 29 times.12,17,18 Furfural can be
oxidized to form SA,19 MA20,21 and 2(5H)ꢀfuranone,4,19 all of which can be hydrogenated to GBL19
and further converted to BDO.4,14 For example, Yuan et al.4 achieved a 85% BDO yield from furfural
over a Pt/TiO2ꢀZrO2 catalyst via the 2ꢀstep oxidationꢀhydrogenation of furfural.
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 29
1
2
3
4
5
6
7
8
9
Cuꢀ and Pdꢀbased heterogeneous catalysts have been studied for GBL hydrogenation due to their
high C=O hydrogenation activity.2 Hydrogenation of GBL under batch conditions over a
heterogeneous Pd/C catalyst using ethanol as solvent at 180˚C and 10 MPa H2 pressure resulted in
88.6% BDO selectivity at 98% conversion22, and reaction over a FeOx modified Pd/C catalyst using
water as solvent at 200 °C and 5 MPa H2 gave a 70.8% BDO selectivity at 76.4% conversion.13 Over
a MgSiO3 supported multicomponent CuꢀPdꢀKOH catalyst, GBL was hydrogenated to BDO in the
liquidꢀphase with 99.0% selectivity at 96.5% conversion under 160˚C and 6.2 MPa H2.23 In the
liquidꢀphase hydrogenation of a 70:30 GBLꢀdimethyl succinate (DMS) mixture over Cu/ZnO at
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
180ꢀ220 C and 8 MPa H2, 90.5% BDO and 9.0% THF selectivities were obtained at 90% GBL
conversion (98% DMS conversion).24 Vaporꢀphase GBL hydrogenation over CuCrOx at 210°C
produced BDO selectively (87ꢀ97%) at 60ꢀ70% conversions, with 1ꢀbutanol (1ꢀBol) and THF
selectivities up to 1.1% and 5.5%, respectively.25 In addition, homogeneous Ru catalysts have been
studied for GBL hydrogenation, achieving a 99% BDO yield over a Ru complex of tetradentate
bipyridine ligand at 25°C.26
Clearly, it is highly desirable to develop efficient earthꢀabundant heterogeneous catalysts to
catalyze the hydrogenation of GBL at lower temperature and pressure to produce BDO. In the
present work, up to 95% BDO yield was achieved over a bimetallic Cu0.1Co0.9/TiO2 catalyst at
o
relatively mild conditions of 140 C and 3.4 MPa H2. This Crꢀfree nonꢀnoble CuCo/TiO2 bimetallic
catalyst had over a two times higher BDO selectivity than Pt/C, Pd/C, and Ru/C catalysts. However,
the reaction rate on a total mole of metal basis for GBL hydrogenation on Ru/C was 2.5 times greater
than on the Cu0.1Co0.9/TiO2 catalyst. The CuCo/TiO2 bimetallic catalyst had an activity up to 11 times
higher than that of a commercial copper chromite catalyst. The correlations between the catalyst
structure and catalytic performance of the CuꢀCo/TiO2 catalysts with different Co contents are
discussed to gain insights into the probable active sites and reaction mechanism.
Despite previous research on GBL hydrogenation, the mechanism of GBL
ringꢀopening/hydrogenation to BDO remains unclear. Recent research on the production of C5 and C6
diols from biomassꢀderived hemiacetals provides insight into the mechanism of GBL hydrogenation
to BDO. Brentzel et. al.27,28 reported a 3ꢀstep dehydrationꢀhydrationꢀhydrogenation (DHH) pathway
4
ACS Paragon Plus Environment

Page 5 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
for the production of 1,5ꢀpentanediol (PDO) from furfural. The final step in this process involves
hydrogenation of the hemiacetal 2ꢀhydroxytetrahydropyran (2ꢀHTHP) to PDO, which occurs by
tautomerization of the 2ꢀHTHP to the ringꢀopened aldehyde, 5ꢀhydroxyvaleraldehyde.27 The same
chemistry was shown to proceed in the analogous C6 route from tetrahydropyranꢀ2ꢀmethanol to
1,6ꢀhexanediol.29 It will be shown in this paper that the C4 hemiacetal analog of 2ꢀHTHP,
2ꢀhydroxytetrahydrofuran (2ꢀHTHF), is likely formed by hydrogenation of GBL (Scheme 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accordingly, 2ꢀHTHF can similarly undergo
a
ringꢀopening tautomerization to form
4ꢀhydroxybutanal (4ꢀHB)30, the latter of which has been reported to be a reaction intermediate in the
production of GBL by BDO dehydrogenation.31,32 Hamminga et al.33 used in situ ATRꢀFTIR
spectroscopy to study the mechanism of liquidꢀphase hydrogenation of GBL over CuꢀZnO. Neither
2ꢀHTHF nor 4ꢀHB was observed in their study. Thus, another key objective of this work is to
elucidate the reaction pathways for GBL hydrogenation over CuꢀCo/TiO2 bimetallic catalyst based
on the effect of reaction parameters, catalytic performance of different model compounds and 13C
variable temperature NMR.
2. Experimental
2.1 Catalyst preparation
CuꢀCo/TiO2 catalysts with various Co molar percentages (0, 25%, 50%, 75%, 90%, 95%, 100%)
were prepared by incipient wetness (co)impregnation. A commercial P25 TiO2 (Aldrich) was used as
the support. We chose TiO2 as support because it is a stable support that we have used for other
aqueous phase reactions and it has low Brønsted surface acidity.34 The support was calcined at 750
o
oC (4 C/min) for 4 h before (co)impregnating with an aqueous solution containing the calculated
amounts of Co(NO3)2·6H2O and Cu(NO3)2·3H2O (Aldrich, ≥98%). The total metal loading was set at
1.8 mmol/g. After impregnation, the materials were kept at room temperature for 2 h, dried at 110 oC
overnight, and then calcined at 450 oC for 3 h in a muffle furnace with a ramp rate of 3 oC/min. The
Cuꢀrich (Cu mol% ≥75%) CuꢀCo/TiO2 catalysts were reduced under H2 flow (80 ml/min, >99.99%,
Airgas) at 300 oC (ramping rate: 1 oC /min) for 2 h. Coꢀrich (Co mol% ≥50%) catalysts were reduced
at 450oC (ramping rate: 2oC /min) for 2 h. The calcined and reduced catalysts are denoted as
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 29
1
2
3
4
5
6
7
8
Cu1ꢀxCox/TiO2ꢀC and Cu1ꢀxCox/TiO2ꢀR, respectively, where x refers to the molar fraction of Co in
decimal form. The Cu and Co content of the catalysts measured from the ICP results were similar to
the amount expected from impregnation (Table 1).
9
Copper chromite (CuCrOx) and 5%Ru/C were purchased from Aldrich. 5%Pd/C and 5%Pt/C
were supplied by Strem Chemicals Inc. The copper chromite was reduced at the same conditions as
monometallic Cu/TiO2. The Pd, Pt and Ru catalysts were reduced under H2 flow (80
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
o
ml/min, >99.99%, Airgas) at 300 C (ramping rate: 4 C /min) for 2 h. All the above reduced
catalysts were stored in a glove box in Ar (>99.99%, Airgas) to avoid oxidation.
2.2 Characterization
The elemental compositions of the synthesized samples were analyzed by ICPꢀOES (Varian
VistaꢀMPX CCD Simultaneous ICPꢀOES) after the samples (50 mg) were digested with a mixture of
2 mL nitric acid (Fisher, 69.2 wt%), 6 mL hydrochloric acid (Fisher, 37.4 wt%), and 10 mL
hydrofluoric acid (Acros Organics, 48–51%) in a Teflon beaker at 100 oC for 24 h. Cobalt and copper
standards for the ICP analysis were prepared from copper (Fluka, 1000
(Fluka, 1000 2 mg/L) ICP standards.
Xꢀray powder diffraction (XRD) patterns of the samples were obtained in the scanning angle (2θ)
range of 20ꢀ80° on a Bruker D8 Discover diffractometer using Cu Kα1 radiation ( = 1.5406 Å)
) of CuO and Cu in the samples were
is the DebyeꢀScherrer constant (0.90),
± 2 mg/L) and cobalt
±
λ
operated at 50 kV and 1000 mA. The crystallite sizes (
calculated by the Scherrer equation: kλ L cos , where
is the diffraction angle and
d
d
=
/
β
θ
k
θ
β
L is the full width at half maximum.
Thermogravimetric analysis (TGA) of the spent catalyst was performed with a TA Instruments
Q500 system. Approximately 12 mg of the sample was loaded onto a Pt pan in 100 mL/min N2 flow,
ramped at 10 °C/min from room temperature to 100 °C, held for 20 min, and then ramped at
10 °C/min to 800 °C.
The Brunauer–Emmett–Teller (BET) surface area was calculated from nitrogen adsorption data at
liquid nitrogen temperature obtained on a Micromeritics ASAP 2020 system. Prior to the BET
measurements, samples were degassed under vacuum at 250 °C for 4 h. The BET surface areas of the
calcined samples were in the range of 12.8 to 15.6 m2/g (Table 1).
6
ACS Paragon Plus Environment

Page 7 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
Scanning transmission electron microscopy (STEM) was performed using a Csꢀcorrected FEI
Titan microscope operated at 200 kV. Prior to STEM analysis, the samples were passivated in 1%
oxygen after reduction. Catalyst powders were dispersed in ethanol and ultrasonicated for 30 min
and deposited on either carbonꢀcoated Cu or Au grids and plasma cleaned prior to insertion into the
microscope. Images were collected with highꢀangle annular darkꢀfield (HAADF) STEM, with
collection angles of 54ꢀ270 mrad, probe convergence angle of 24.5 mrad, and probe current of 25 pA.
Energyꢀdispersive xꢀray spectra (EDS) and electron energy loss spectra (EELS) were collected on the
same microscope with a EDAX SiLi and Gatan GIF 865 detector, respectively. Spectral images (SI)
were collected in EFSTEM with an energy resolution of 0.8 eV. The SIs were deꢀnoised by a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
weighted principal component analysis (PCA) using temDM software,
a
plugin for
DigitalMicrograph.35,36 12 principle components were retained. Relative composition maps were
generated from the SI by standard background extrapolation and subtraction techniques for the
coreꢀloss Ti, Co, and Cu Lꢀedges. Approximately 100ꢀ200 particles were analyzed for each particle
size distribution. In the case of the bimetallic samples, bimodal distribution of particle sizes was
observed. Since it was very difficult for us to quantify the exact ratio of small (<4 nm) to large (>4
nm) nanoparticles, an equally number of both sizes were counted.
Temperatureꢀprogrammed reduction (TPR) was carried out on a Micromeritics AutoChem II
2920 unit. Before TPR, the asꢀprepared calcined samples (100 mg) were pretreated in helium at 150
oC for 30 min. A TPR run was carried out with a 10% H2/Ar mixture gas at a flow rate of 50 mL/min
o
o
and the temperature was increased to 800 C at a temperature ramp rate of 10 C/min. A dry
ice/acetone cooling bath was used to remove moisture from the TPR effluent stream. The H2
consumption was monitored by a thermal conductivity detector (TCD) and quantified based on H2
consumption of a TPR standard (Ag2O).
NH3ꢀTPD measurements were carried out using a Micromeritics Autochem II 2920 unit. Prior to
NH3ꢀTPD analysis, samples were reduced at 300 °C for Cu1/TiO2 and CuCrOx and 450 °C for
CuCo/TiO2 bimetallic catalysts for two hours (2 °C/min) under a flow of 10% H2 in Ar. Then
samples were cooled to 100 °C for NH3 adsorption. Adsorption was performed under a flow of
15%NH3–He (Airgas, 50 mL/min) for 30 min. NH3 was desorbed under a He flow of 50 mL/min
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 29
1
2
3
4
5
6
7
8
9
with 10° C/min ramp rate to 700 °C. The desorbed NH3 molecules were continuously monitored by a
MKS Cirrus 2 quadrupole mass spectrometer recording the signals m/z = 17 and 16 (corresponding
to NH3). The density of acid sites was quantified by calibrating the MS signals with NH3/He gas
flows at several different concentrations.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A 10 wt% 2ꢀhydroxytetrahydrofuran (2ꢀHTHF) (Accela ChemBio, 95%) in 1,4ꢀdioxaneꢀd8
(Aldrich,
≥ 99 atom% deuterium) sample was prepared for nuclear magnetic resonance (NMR)
spectroscopy. Variable temperature NMR experiments were performed on a Bruker AvanceꢀIII 500
MHz spectrometer with a broadband fluorine observe probe. The 2H signal from 1,4ꢀdioxaneꢀd8 was
2
used as the fieldꢀfrequency lock. The magnetic field shimming was performed on the H solvent
1
13
signals. H and C chemical shifts were referenced against solvent. Spectra were collected from 24
to 82 °C. The sample was equilibrated at each temperature for 15 minutes before collecting the
1
spectrum. H experiments were performed with a Bruker standard 1D pulse sequence, ‘zg30’. The
interꢀscan relaxation delay, sweep width, acquisition time, and number of scans were 2 s, 20 ppm
centered at 10 ppm, 3.3 s, and 8 scans, respectively. 13C spectra were collected with a quantitative 1H
decoupled pulse sequence, ‘zgig30’. The interꢀscan relaxation delay, sweep width, acquisition time,
and number of scans were 2 s, 284 ppm centered at 110 ppm, 2 s, and 32 scans, respectively.
2.3 Reaction kinetics measurements
GBL hydrogenation batch reactions were carried out in a hastelloy autoclave (50 ml) at a stirring
speed of 750 rpm. The reduced active catalysts were loaded into the reactor in the glove box. In a
typical run, 20 g of 10 wt% GBL (Aldrich, 99%) in 1,4ꢀdioxane (Acros, >99%, stabilized with BHT)
solution and 0.4 g of reduced catalyst were introduced into the autoclave. The reactor was then
purged with H2 three times, pressurized with H2 to 3.4 MPa, and heated to the desired reaction
temperature, which was held constant for the duration of the reaction. The reactor was quenched in
an ice bath to terminate the reaction. THF (Acros, 99.9%, stabilized with BHT), ethanol (Decon Labs
Inc., 100%), and deionized water were used to study the effect of solvents.
Reaction kinetics studies using a continuous flow reactor were carried out with the
o
Cu0.1Co0.9/TiO2 catalyst in an upꢀflow configuration at 140 C and 3.4 MPa. The calcined catalyst
sample (0.6 g) was loaded into a ¼” OD (0.18” ID) 316 stainless steel tube and secured in place by
8
ACS Paragon Plus Environment

Page 9 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
alternating layers of quartz wool (Grace) and granular silicon dioxide (Sigma, 99.9%, 4ꢀ20 mesh).
The reactor tube was contained inside a tube furnace equipped with aluminum filler rods for uniform
heat distribution. Both ends of the furnace were sealed with quartz wool. The reactor was pressurized
with H2 gas to 3.4 MPa. The reactor furnace controller was set to the desired reaction temperature.
The catalyst was reduced at the same conditions as for batch reactor. After the temperature of the
catalyst bed was decreased to the reaction temperature, a solution of 10 wt% GBL in 1,4ꢀdioxane
was pumped into the reactor with an HPLC pump (Lab Alliance Series I) at a speed of 1.2 mL/h. The
H2 to GBL mole ratio was 50. Liquid products accumulated in a stainlessꢀsteel gas–liquid separator
(300 mL) above the reactor outlet. Liquid samples were analyzed by gas chromatography (Shimadzu
GC2010) equipped with a flame ionization detector (FID) and a ZBꢀ5MS capillary column (30 m ×
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.25 mm × 0.5
ꢀm). The products were also confirmed by GCꢀMS (Shimadzu GCMSꢀQP2010S)
equipped a Restek RTXꢀVMS capillary column (30 m × 0.25 mm × 1.4
ꢀm). The gaseous products
from batch reactions were collected using a gasꢀbag and analyzed using a gas chromatograph
(Shimadzu GCꢀ2014) with both TCD (fitted with a GSꢀCarbonPLOT column) and FID (fitted with a
RtꢀQꢀBond capillary column). Only negligible amounts of CH4, CO and CO2 (in total <0.1%) were
detected in gas products. GBL conversion, product selectivities and product production rates were
calculated on carbon basis and are defined as follows
：
mol of GBL charged ꢀ mol of GBL left
mol of GBL charged
100
×
Conversion (%) =
x
mol of product C atoms in product
x
Selectivity (%) =
100
(mol of GBL charged ꢀ mol of GBL left) x C atoms in GBL
mol of GBL converted × Product selectivity
mol of active metal × reaction time
Reaction rate (molprod/molmetal hꢀ1) =
In
the
hydrogenation of 10 wt% GBL in 1,4ꢀdioxane, 4ꢀhydroxybutyl,4ꢀhydroxybutanoate (4ꢀHHB) was
formed by the esterification of GBL with the BDO product. The formation of 4ꢀHHB was also
confirmed by separate reaction studies using equal amounts (5 wt%) of GBL and BDO in
1,4ꢀdioxane catalyzed by either acid (eg. HY zeolite) or base (eg. CaO) in an oil bath at 60 °C. Due
to the unavailability of the standard for 4ꢀHHB, its GC response factor was assumed to be the sum of
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 29
1
2
3
4
that for GBL and BDO. In most cases, the carbon balances were 100
±4%. Because small
5
6
7
8
concentrations of products were detected when ethanol and water were separately used as solvent,
the selectivity (%) for ethyl 4ꢀhydroxybutanoate and 4ꢀhydroxybutanoic acid was estimated by
subtracting the sum of the known products from 100%.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
Results and discussion
The XRD patterns of calcined and reduced monometallic and bimetallic Cu/Co catalysts are
shown in Figure 1. The monometallic Co1/TiO2ꢀC displayed the characteristic peaks of Co3O4 (PDF
# 43ꢀ1003) and rutile TiO2, while the pattern of the monometallic Cu1/TiO2ꢀC showed the
characteristic peaks of CuO (PDF # 05ꢀ0661) and rutile TiO2 (Figure 1A). The crystallite sizes of
Co3O4 and CuO were estimated to be 18.0 and 21.7 nm, respectively, from the Scherrer equation
(Table 1). The intensity of the Co3O4 diffractions decreased for the Cu0.1Co0.9/TiO2ꢀC sample with the
(311) peak at 36.8° possibly decomposing into two peaks centered at 36.7 and 36.8° (Figure 1B). The
shift of the Co3O4 (311) peak to lower 2θ reflects the possible expansion of Co3O4 lattice by addition
of Cuinto the lattice37, with the lattice parameter increased from 2.437 to 2.444 Å. Only a broad
shoulder at around 2θ = 36.7° was seen for Cu0.5Co0.5/TiO2ꢀC (Figure 1B). Note that due to the
overlapping of the diffraction peaks for Co3O4 and CuCo2O4, it is difficult to distinguish the two
components by XRD patterns.37,38 No diffraction peaks corresponding to CuO could be seen in
bimetallic Cu0.5Co0.5/TiO2ꢀC and Cu0.1Co0.9/TiO2ꢀC samples (Figure 1B). This result indicates that
large CuO crystallites only existed in the monometallic Cu1/TiO2ꢀC sample, and Cu species were
highly dispersed in the bimetallic samples. This behavior is in agreement with the work of Wang et
al.38
o
After reduction at 450 C, the diffractions of Co3O4 for both monometallic Co1/TiO2ꢀR and
bimetallic CuCo/TiO2ꢀR samples disappeared and no other peaks related to CoO could be observed
(Figure 1C). Only a shoulder peak at 2θ = 44.3°, assigned to Co0, could be observed in Co1/TiO2ꢀR
(Figure 1D) because of the overlapping of the intense diffraction peaks of the TiO2 support. The
broad shoulder peak at 2θ = 44.3° in Cu0.1Co0.9/TiO2ꢀR is either metallic Co or a Co phase that has
small amounts of Cu in the lattice. Diffraction peaks assigned to Cu0 were clearly observed in
10
ACS Paragon Plus Environment

Page 11 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
monometallic Cu1/TiO2ꢀR. The intensity of these peaks decreased with increasing Co content, and
almost disappeared in Cu0.1Co0.9/TiO2ꢀR. The slight shift of 2θ value for Cu (111) from 43.3° for
Cu1/TiO2ꢀR to 43.4° in Cu0.5Co0.5/TiO2ꢀR, probably corresponded to a Cu phase that has some Co in
the lattice.37, 39,40
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(Co,Cu)Co2O4
CuO
(B)
(A)
Cu1/TiO ꢀC
2
(Cu,Co)Co2O4
Cu1/TiO2ꢀC
CuO
Cu0.5Co0.5/TiO2ꢀC
Cu0.1Co0.9/TiO ꢀC
2
Cu0.5Co0.5/TiO2ꢀC
Cu0.1Co0.9/TiO2ꢀC
Co1/TiO2ꢀC
Co1/TiO2ꢀC
TiO2
TiO2
Co3O PDF#43ꢀ1003
4
CuOPDF#05ꢀ0661
20 25 30 35 40 45 50 55 60 65 70 75 80
35
36
37
38
2 Theta (degree)
2 Theta (degree)
Cu0
Cu0
Co0/CuCo alloy
(C)
(D)
Cu1/TiO2ꢀR
Cu0.5Co0.5/TiO2ꢀR
Cu0.1Co0.9/TiO2ꢀR
Co1/TiO2ꢀR
Cu1/TiO2ꢀR
Cu0.5Co0.5/TiO2ꢀR
Cu0.1Co0.9/TiO2ꢀR
Co1/TiO2ꢀR
TiO2
Co PDF#15ꢀ0806
TiO2
Cu PDF#04ꢀ0836
42
43
44
46
2 Theta (degree) 45
20 25 30 35 40 45 50 55 60 65 70 75 80
2 Theta (degree)
Figure 1 (A,B) XRD patterns of calcined catalyst samples with (B) magnified (2θ = 35–38°); (C, D)
XRD of reduced catalysts with (D) magnified (2θ = 42–46°).
Table 1 Metal loadings, textural properties and reduction degrees of CuCo/TiO2 catalyst samples
with different Co contents
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 29
1
2
3
4
5
6
7
8
9
CuO/Co3O4
Composition
(wt%) a
Extent of
reduction
(%)e
SBET
Particle size
(nm) c
crystallite size
(nm) b
Sample
(m2/g)
Cu
11.93
5.78
1.20
ꢀ
Co
d
Cu1/TiO2ꢀC
Cu0.5Co0.5/TiO2ꢀC
ꢀ
12.8
21.7/ꢀ
N.D/N.D
N.D/16.3
ꢀ/18.0
2.6, >20
62.7
5.19
9.36
15.6
14.6
13.3
2.3, 16.5d
1.6, 21.3 d
19.8
50.2
48.1
34.6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cu0.1Co0.9/TiO2ꢀC
Co1/TiO2ꢀC
10.53
a Obtained from ICP analysis.
b The crystallite sizes calculated by the Scherrer equation. N.D: Not detected.
c Particle size measured by STEM
d Bimodal size distribution of particles measured by STEM
e From H2ꢀTPR.
0.12
Co1/TiO2-C
359
0.08
0.04
313
295
500
0.00
0.12
0.08
0.04
0.00
0.12
0.08
0.04
0.00
0.12
0.08
0.04
0.00
0.12
0.08
0.04
0.00
231
Cu0.1Co0.9/TiO2-C
236
Cu0.5Co0.5/TiO2-C
Cu1/TiO2-C
TiO2
195
150
186
220
112
100
200
300
400
500
600
700
800
Temperature (oC)
Figure 2 Temperatureꢀprogrammed reduction profiles of calcined catalyst samples and TiO2 support.
Figure 2 shows the H2ꢀTPR profiles of the calcined catalyst precursors. No H2 consumption was
seen for the TiO2 support demonstrating that a metal is needed to reduce the TiO2 consistent with
previous studies.41,42 The TPR profile of monometallic Co1/TiO2 gave an initial reduction peak at 313
oC followed by a broad peak at 359 C. This TPR behavior has been described previously for the
o
twoꢀstep reduction of Co3O4 to Co0.38, 41, 43,44 The shoulder peak at ~500 C is likely associated with
o
reduction of TiOx overcoated CoOx particles.34 The monometallic Cu1/TiO2ꢀC showed three peaks at
12
ACS Paragon Plus Environment

Page 13 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
112, 186 and 220oC. The reduction of CuO occurs at lower temperature than the reduction of CoOx,
which is consistent with the works of ÁvilaꢀNeto et al.4 and Wang et al.38 The addition of a small
amount of Cu (10 mol%) to the Co led to a large decrease in the reduction temperature for the
o
9
Cu0.1Co0.9/TiO2ꢀC sample (down to 231 and 295 C). The TPR of the Cu0.5Co0.5/TiO2ꢀC sample had
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
three peaks at even lower temperatures (150, 195 and 236 oC) than the Cu0.1Co0.9/TiO2ꢀC sample. It is
likely that Cu0 species initiates the reduction process by activation of H2 and the adsorbed hydrogen
is transferred to cobalt oxides by a spillover mechanism.44,45
The extent of reduction of the catalysts is shown in Table 1. Only 34.6% of the cobalt was
reduced in the Co1/TiO2ꢀR catalyst sample while 62.7 % of the copper was reduced in the
Cu1/TiO2ꢀR. The low reduction degree of Co1/TiO2ꢀR is in agreement with our previous studies.44
The extent of reduction of Cu0.1Co0.9/TiO2ꢀR and Cu0.5Co0.5/TiO2ꢀR was 48.1% and 50.2%,
respectively. If we assume that 62.7% of the Cu is reduced in the bimetallic catalysts, then the extent
of reduction of Co in Cu0.1Co0.9/TiO2ꢀR and Cu0.5Co0.5/TiO2ꢀR is 45.0 and 30.8%, respectively. It is
likely that the extent of reduction under reaction conditions is different than the extent of reduction
reported in Table 1 because of differences in the hydrogen partial pressure.
Figure 3 shows the STEM micrographs of CuCo/TiO2ꢀR catalysts with different Co contents.
The majority of the Cu particles in the monometallic Cu1/TiO2ꢀR catalyst were quasiꢀcircular with an
average diameter of 2.6 nm (Figure 3a). Small amounts of large Cu aggregates (> 20 nm) were
observed in this catalyst (Figure 3a and Figure S1a). The presence of some large Cu particles in
Cu1/TiO2ꢀR catalyst explains the much larger crystallite size (25.4 nm) determined by XRD (Figure
1c). A bimodal distribution of dispersed small particles (average particle sizes ~ 2.3 nm) and large
particles (average particle sizes ~16.5 nm) was observed for Cu0.5Co0.5/TiO2ꢀR. Larger particles were
observed in both Cu0.1Co0.9/TiO2ꢀR (21.3nm) and Co1/TiO2ꢀR (19.8nm) catalysts compared to the
catalysts which had more Cu. A few small particles (< 3 nm) were also observed in Cu0.1Co0.9/TiO2ꢀR
(Figure S1d); however, this catalyst had fewer of these smaller particles than the Cu0.5Co0.5/TiO2ꢀR
catalyst. Accordingly, Cu0.1Co0.9/TiO2ꢀR has less of a bimodal character than Cu0.5Co0.5/TiO2ꢀR. The
Co1/TiO2 sample consisted exclusively of larger (19.8 nm average size) particles.
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 29
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3 STEM micrographs of CuCo/TiO2 catalysts with different Co contents: (a) Cu1/TiO2ꢀR; (b)
Cu0.5Co0.5/TiO2ꢀR; (c) Cu0.1Co0.9/TiO2ꢀR and (d) Co1/TiO2ꢀR.
14
ACS Paragon Plus Environment

Page 15 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4 Spectral image analysis of Cu0.1Co0.9/TiO2ꢀR bimetallic nanoparticles: (a) STEMꢀHAADF
image; (b) RGB composite map of twoꢀdimensional EELS (red: Ti L map, green: Co L map, and
blue: Cu L map); (c) elemental projection of Ti; (d) elemental projection of Co and (e) elemental
projection of Cu.
EELS and EDS studies were conducted on several nanoparticles to investigate the elemental
composition and distribution of elements in the Cu0.1Co0.9/TiO2ꢀR catalyst. Accurate elemental
mapping of the small nanoparticles (<3 nm) was not possible due to damage by the electron beam before a
sufficient signalꢀtoꢀnoise could be obtained, although Cu and Co both appear to exist to some extent in
these particles. EELS of the large (~20 nm) nanoparticles revealed that Co and Cu formed bimetallic
nanoparticles and were in close contact (Figure 4b). The composition maps showed highly concentrated
Co signals in the core and more concentrated Cu signals on the outer surfaces of the nanoparticles (Figure
4b, d), suggesting that Cu prefers to deposit on the outer surface of the Co nanoparticles. EDS spot scans
of several large nanoparticles were in agreement with the observation from EELS (Figure S2). The
coꢀexistence of Cu and Co, especially at the outer surface of the bimetallic nanoparticles (Figure 4b, d,
e), suggests the formation of a CuCo alloy shell structure. Note that similar coreꢀshell CuCo
bimetallic nanoparticles with a Coꢀrich core and Cuꢀrich shell were also determined in Cu0.5C0.5/TiO2
by EELS (Figure S3) and EDS spot scans (Figure S4). The formation of such Coꢀrich core and CuCo
alloy nearꢀsurface shell particles is driven by the lower surface energy of Cu compared with Co (1.93
J/m2 vs. 2.71 J/m2).38, 46ꢀ48. The positive segregation energy (+0.73 eV) for Coꢀcore and Cuꢀshell
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 29
1
2
3
4
5
6
7
8
alloy nanoparticles calculated by DFTꢀPW91 also indicates the preferable formation of such
coreꢀshell configuration.49 A similar Coꢀrich core and a Cuꢀdominated outer shell catalyst has also
been recently prepared by Xiang et al. using a coꢀprecipitation method.50
9
Table 2 Effect of different catalysts and solvents on GBL hydrogenation at 100oC
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reaction rate
(mmolBDO
molmetal·hꢀ1)
Carbon Selectivity (%)
Conversion
(%)
Catalyst
Solvent
/
BDO Byproducts
Blank
1,4ꢀDioxane
1,4ꢀDioxane
1,4ꢀDioxane
1,4ꢀDioxane
1,4ꢀDioxane
1,4ꢀDioxane
THF
0
0
0
0
0
0
0
0
5%Pd/C
5%Pt/C
2.2
22.4
75.2
5.7
30.1
1.2
5.9
0
56.9
34.5
100
100
100
23.7
0
1ꢀBol (3.6)
5%Ru/C
23.1
8.8
1ꢀProl (7.5)
CuCrOx
0
Cu0.1Co0.9/TiO2ꢀR
Cu0.1Co0.9/TiO2ꢀR
Cu0.1Co0.9/TiO2ꢀR
Cu0.1Co0.9/TiO2ꢀR
11.5
0.9
0
0
Ethanol
9.3
EHB (76.3)
4ꢀHBA (100)
H2O
16.8
Reaction conditions: 20 g 1 wt% GBL in different solvents, 0.20 g catalyst, 100 °C, 3.4 MPa H2, 24 h.
The reaction rate is defined as mmol of BDO generated per mol of active metal per h. BDO:
1,4ꢀbutanediol; 1ꢀBol: 1ꢀbutanol; 1ꢀProl: 1ꢀpropanol; EHB: ethyl 4ꢀhydroxybutanoate; 4ꢀHBA:
4ꢀhydroxybutanoic acid.
Table 2 shows the catalytic performance of different catalysts for hydrogenation of 1 wt% GBL
at 100 °C in a batch reactor. Carbonꢀsupported noble metal catalysts either presented low activity
(rates <30 mmolBDO/molmetal·hꢀ1 for Pd/C and Pt/C) or low BDO selectivity (<57% for Ru/C, Pd/C,
and Pt/C). Conversion to undetected humins or GBL adsorption onto Pt/C and Ru/C (similar as with
GBL analog γꢀvalerolactone on Ru/C51) may have caused the low observed BDO selectivities over
these catalysts. A commercial copperꢀchromite catalyst showed moderate GBL hydrogenation
activity (5.7 mmolBDO/molmetal·hꢀ1) and up to 100% BDO selectivity; however, the use of
Crꢀcontaining catalysts are undesirable because of the high toxicity of Cr(VI) 52,53
The
.
Cu0.1Co0.9/TiO2ꢀR bimetallic catalyst exhibited high BDO production rate (30.1 mmolBDO/molmetal·hꢀ1
)
while maintaining 100% BDO selectivity.
16
ACS Paragon Plus Environment

Page 17 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
A range of commonly used solvents was studied for GBL hydrogenation with the
Cu0.1Co0.9/TiO2ꢀR catalyst, as shown in Table 2. Almost no reaction took place (GBL conversion
<1%) when the reaction was carried out in THF. In ethanol and water, low selectivities to BDO
(<24%) were obtained due to the formation of ester (ethyl 4ꢀhydroxybutanoate, EHB) or acid
(4ꢀhydroxybutanoic acid, 4ꢀHBA) byproducts, respectively. The highest activity and selectivity to
BDO (100%) was attained when 1,4ꢀdioxane was used as solvent. Previous studies also found
1,4ꢀdioxane as a suitable solvent for 1,4ꢀdiols production from GBL,33 SA17 and levulinate esters54.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5 Effect of cobalt content on the catalytic activity and selectivity of CuCo/TiO2ꢀR catalysts at
fixed Cu + Co molar amount (1.8 mmol/g) for GBL hydrogenation. The Co content is the amount of
Co in the Cu+Co bimetallic catalyst. (Reaction conditions: 20 g 10 wt% GBL in 1,4ꢀdioxane, 0.40 g
catalyst (0.80 g for Cu1/TiO2ꢀR), 140°C, 3.4 MPa H2, 5ꢀ24 h, ~20% conversion).
Figure 5 shows the reaction rate and selectivity to major products for hydrogenation of 10 wt%
GBL in 1,4ꢀdioxane over CuCo/TiO2ꢀR catalysts as a function of Co content. The BDO production
rate increased from 0.068 molBDO/molmetal·hꢀ1 for monometallic Cu1/TiO2 to a maximum of 1.36
molBDO/molmetal
·
hꢀ1 at 90 mole% Co content for Cu0.1Co0.9/TiO2ꢀR and then decreased to 0.55
hꢀ1 for monometallic Co1/TiO2ꢀR. The BDO selectivity did not vary significantly
molBDO/molmetal
·
(maintained in the range of 87.2ꢀ93.4%) with the change in Co content, with the CuCo/TiO2
bimetallic catalysts having a slightly higher selectivity than their Cu and Co monometallic
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 29
1
2
3
4
5
6
7
8
counterparts. In contrast to experiments performed at low GBL concentrations (1 wt%; Table 2), the
higher GBL concentrations (10 wt%) used in this experiment resulted in formation of
4ꢀhydroxybutyl,4ꢀhydroxybutanoate (4ꢀHHB) with selectivities up to 12.5%. Other byproducts
9
detected were small amounts of THF (
≤
0.3%) and 1ꢀBol (
≤
0.2%) (not shown).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12
10
100
90
80
70
60
50
40
30
20
10
0
ꢀ
ꢁ
ꢂ
Cu1Co9/TiO2 BDOselectivity
CuCrOx BDOselectivity
Cu1Co9/TiO2 reaction rate
8
6
4
2
0
ꢃ
CuCrO reaction rate
x
120
130
140
150
160
170
180
Temperature ( )
oC
Figure 6 Effect of reaction temperature for hydrogenation of GBL on Cu0.1Co0.9/TiO2ꢀR and
commercial CuCrOx catalysts. (Reaction conditions: 20 g 10 wt% GBL in 1,4ꢀdioxane, 0.40 g
catalyst, 3.4 MPa H2, ~20% conversion, 0.6ꢀ28 h reaction).
Figure 6 shows the reaction rate and BDO product selectivity of GBL hydrogenation over
Cu0.1Co0.9/TiO2ꢀR and a commercial CuCrOx catalyst as a function of reaction temperature from 120
to 180 °C. At a temperature of 120 °C, the rate of BDO production over Cu0.1Co0.9/TiO2ꢀR (0.412
molBDO/molmetal·hꢀ1) is 11.4 times higher than over CuCrOx (0.036 molBDO/molmetal·hꢀ1) on a mol
active metal basis (for CuCrOx is based on mol of Cu). At 180 °C the rate of BDO production for
Cu0.1Co0.9/TiO2ꢀR (8.711 molBDO/molmetal·hꢀ1) is 4.8 times higher than over CuCrOx (1.803
molBDO/molmetal·hꢀ1). The apparent activation energy (Ea) for Cu0.1Co0.9/TiO2ꢀR is 75 kJ/mol, versus
93 kJ/mol for CuCrOx. The BDO selectivity for Cu0.1Co0.9/TiO2ꢀR slightly declined from 93.4 to 92.3%
with increasing temperature from 120 to 180 °C. The BDO selectivity for Cu0.1Co0.9/TiO2ꢀR was
higher than the BDO selectivity for CuCrOx (86ꢀ89.9%) over the entire temperature range studied.
The rather low surface acidity of the Cu0.1Co0.9/TiO2 bimetallic catalyst (~1/9 of CuCrOx catalyst,
18
ACS Paragon Plus Environment

Page 19 of 29
ACS Catalysis
1
2
3
4
5
6
Figure S5) explains its higher BDO selectivity for GBL hydrogenation, as acid sites catalyzed the
dehydration of BDO to form THF and 1ꢀBol byproducts.13,31,55,56
7
8
9
Figure 7 shows the time course of GBL hydrogenation over Cu0.1Co0.9/TiO2ꢀR catalyst at 140 °C
and 3.4 MPa H2, with detailed GBL conversion and products selectivity at each time point presented
in Table S1. The concentration of GBL decreased significantly in the first 24 h ꢀ with a
corresponding increase in the concentration of BDO ꢀ before stabilizing at 0.06 mol/L C after 36 h.
The concentration of 4ꢀHHB went through a maximum at 8 h, after which it was likely converted to
BDO. A maximum in BDO yield (95.2%) was achieved at 36 h, followed by a decrease inBDO
concentration from 4.60 at 36 h to 4.38 mol/L C at 48 h due to overꢀconversion of BDO to THF,
1ꢀBol, and 1ꢀProl through a series of secondary reactions. THF, 1ꢀBol, and 1ꢀProl concentrations
were minimal even after 48 h: 0.22, 0.05 and 0.08 mol/L C, respectively. No 2ꢀHTHF or 4ꢀHB was
observed in this reaction even though they are likely reaction intermediates.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1.0
5
0.8
4
GBL
0.6
BDO
3
4ꢀHHB
THF
1ꢀButanol
0.4
2
1
0
1ꢀPropanol
0.2
0.0
0
10
20
30
40
50
Time (h)
Figure 7 GBL hydrogenation over Cu0.1Co0.9/TiO2ꢀR as a function of reaction time (Reaction
conditions: 30 g 10 wt% GBL in 1,4ꢀdioxane, 1.2 g catalyst, 140 °C and 3.4 MPa H2).
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 29
1
2
3
4
5
6
7
8
9
25
20
15
10
5
100
90
80
70
60
50
40
30
20
10
0
GBLconversion
BDOselectivity
4ꢀHHBselectivity
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
20
40
60
80
100
120
140
160
Time (h)
Figure 8 GBL hydrogenation with Cu0.1Co0.9/TiO2ꢀR in a continuous flow reactor. (Reaction
conditions: 10 wt% GBL in 1,4ꢀdioxane, 0.60 g catalyst, 140°C , 3.4 MPa H2, WHSV of GBL = 0.2
hꢀ1, and H2/GBL mole ratio = 50.
The longꢀterm stability of the Cu0.1Co0.9/TiO2ꢀR bimetallic catalyst for GBL hydrogenation was
carried out using a fixedꢀbed flow reactor as shown in Figure 8. The GBL conversion remained
constant between 16.5ꢀ18.2% for 98 h before slowly declining to 14.2% after 150 h timeꢀonꢀstream.
The BDO (81.2ꢀ87.8%) and 4ꢀHHB intermediate (11.7ꢀ17.2%) selectivities were stable from 30ꢀ150
h timeꢀonꢀstream. The total selectivity to the THF, 1ꢀBol, and 1ꢀProl byproducts remained below 0.6%
during the entire reaction.
The spent catalyst after the 150 h reaction was characterized by XRD, TGA, ICP, BET, and
STEM. No perceptible structure changes were observed by XRD as compared with the reduced fresh
catalyst (Figure 9). The BET surface area of the spent catalyst (12.0 m2/g) was nearly unchanged
from the fresh catalyst (11.6 m2/g). ICP analysis showed no leaching of Cu or Co during the reaction.
In addition, no detectable weight loss due to the desorption/decomposition of coke can be seen from
the TG profile (Figure S6). While it is possible that a small amount of sintering occurred, there were
no large particle size changes observed by the STEM between fresh and spent Cu0.1Co0.9/TiO2ꢀR
(Figure S7). It is possible that slight restructuring of the surface morphology of Cu and Co occurred
and caused deactivation.
20
ACS Paragon Plus Environment

Page 21 of 29
ACS Catalysis
1
2
3
4
5
6
)
.
7
u
.
a
(
8
y
t
i
9
s
n
e
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
t
n
I
Spent
Reduced
20
30
40
50
60
70
80
2 Theta (degree)
Figure 9 XRD spectrum of the Cu0.1Co0.9/TiO2ꢀR catalyst before and after the 150 h timeꢀon stream
GBL hydrogenation reaction
Table 3 Reaction results of various substrates hydrogenation over Cu0.1Co0.9/TiO2ꢀR catalyst at 100
oC
Time
(h)
Catalyst
amount
(g)
Rate
Carbon Selectivity (%)
BDO GBL 1ꢀButanol
Conversion
(%)
Substrates
(molprod
/molmetal·hꢀ1)
GBL
24
0.2
11.5
18.2
15.4
1.4
0.03
100
95.5
0
ꢀ
4.5
0
0
2ꢀHTHF
Butyraldehyde
Ethyl butyrate
Butyric acid
0.5
1
0.02
0.02
0.2
21.1
11.7
0.004
0
0
100
100
0
24
24
0
0
0.2
0
0
0
Reaction conditions: 20 g 1 wt% substrate in 1,4ꢀdioxane, 100 °C, 3.4 MPa H2
Table 3 shows the reaction rate for hydrogenation of different model compounds over the
o
Cu0.1Co0.9/TiO2ꢀR catalyst at 100 C. Butyraldehyde, ethyl butyrate, and butyric acid were used as
the corresponding model compounds for 4ꢀHB, 4ꢀHHB, and 4ꢀHBA due to the unavailability of the
latter compounds. The reaction rate for hydrogenation of 2ꢀHTHF and butyraldehyde is ~700 and
~400 times higher than the reaction rate for hydrogenation of GBL, explaining why these
intermediates were not detected by GC. The similar conversion rates of 2ꢀHTHF and butyraldehyde
are further evidence that conversion of 2ꢀHTHF to BDO proceeds via hydrogenation of the aldehyde
group of 4ꢀHB. Note that a small amount (4.5%) of GBL was detected at the beginning (conversion
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 29
1
2
3
4
5
6
7
8
< 20%) of 2ꢀHTHF hydrogenation, which further confirms that 2ꢀHTHF is a reaction intermediate
for GBL hydrogenation. The reaction rate of the other molecules was much lower than 2ꢀHTHF and
butyraldehyde and decreased in the order GBL > ethyl butyrate > butyric acid.
9
In situ NMR spectroscopy measurements were collected to determine the equilibrium of 2ꢀHTHF
and its ringꢀopened tautomer, 4ꢀHB, in 1,4ꢀdioxane at varying temperature (24 to 82 °C), as shown in
Figure 10. At 24 °C 5.8% of the cyclic hemiacetal, 2ꢀHTHF, exists in its ringꢀopen form (4ꢀHB). The
equilibrium shifts towards the ringꢀopened product at elevated temperatures as a result of
endothermic tautomerization, with up to 20% of 2ꢀHTHF transformed to 4ꢀHB at 82 °C.
Extrapolation of the NMR ringꢀopening equilibrium data suggests that the equilibrium is 28.4% and
65% ringꢀopened to 4ꢀHB at a reaction temperature of 100 and 140 °C, respectively.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10 Variable temperature NMR of 2ꢀHTHF in 1,4ꢀDioxane.
22
ACS Paragon Plus Environment

Page 23 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2 Proposed reaction pathway for the production of BDO and byproducts (The blue arrows
show that 4ꢀHHB is a reaction intermediate that can be further converted to BDO).
Based on the timeꢀcourse reaction studies, variable temperature NMR measurements, and probe
reaction studies, we propose the reaction pathway for the production of BDO and byproducts shown
in Scheme 2. GBL is initially hydrogenated to 2ꢀHTHF, which then equilibrates with its ringꢀopened
tautomer, 4ꢀHB. This equilibration to the ringꢀopened form is followed by the rapid hydrogenation to
BDO over metallic active sites. This reaction pathway is similar to the hydrogenation of other
hemiacetals including the hydrogenation of the hemiacetal 2ꢀhydroxytetrahydropyran (2ꢀHTHP) to
1,5 pentanediol27 and an analogous C6 route from tetrahydropyranꢀ2ꢀmethanol to 1,6ꢀhexanediol.29
Higher feedstock concentrations lead to the formation of the ester 4ꢀHHB from the GBL reactant and
the BDO product. The ester can be further hydrogenated to BDO. The dehydration and
hydrogenolysis of BDO generates the byproducts THF and 1ꢀBol, respectively.13,31,55,56 The
1ꢀpropanol (1ꢀProl) byproduct is likely formed via the decarbonylation of 4ꢀHB. 4ꢀHBA is formed
when water is a solvent but not in 1,4ꢀdioxane.
Although Cuꢀbased catalysts have been widely used for the hydrogenation of GBL and ester
derivatives of SA and MA to BDO,2,5,6,8,10,33,57 our findings show that Co1/TiO2ꢀR is more than 8
times active than Cu1/TiO2ꢀR for GBL hydrogenation. The reaction rate is further enhanced by
2.1ꢀ2.5 times by the addition of small amount of Cu (5~25 mol%) into the Co catalyst (Figure 5). Cu
has a low solubility in Co with a maximum of 9 at% Cu that can be dissolved in Co at 1050 °C,
according to the binary CuꢀCo phase diagram.58 The amount of Cu that can form a CuꢀCo solution
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 29
1
2
3
4
5
6
7
8
9
decreases with decreasing temperature. However, recent studies indicate that CuCo alloy particles
can be formed in nanoparticles below 100 nm. 46,47,50,59ꢀ61 A variety of experimental techniques,
including STEMꢀEDS(EDX) chemical analysis,46,59ꢀ61,63,64 highꢀresolution TEM,46,63 Atom probe
tomography (APT)50, and Xꢀray photoelectron spectroscopy (XPS) with Ar+ ions etching47,62 have
been applied to confirm the formation of CuCo nanoalloy particles. In the present work,
STEMꢀEELS (Figure 4) and STEMꢀEDS (Figure S2) showed the formation of Coꢀrich core and
CuCo alloy nearꢀsurface shell nanoparticles in the Cu0.1Co0.9/TiO2ꢀR bimetallic catalyst. XRD results
(Figure 1) also provided evidence for the formation of CuCo bimetallic phase in the CuCo/TiO2
bimetallic catalysts. CuCo bimetallic nanoparticles have been reported to have higher activity than
their monometallic counterparts in many catalytic reactions.50, 60ꢀ63,65 Prieto et al.60 ascribed the high
activity of a CuCo/MoOx catalyst in synthesis gas hydrogenation to the formation of a nanosized
CuCo alloy phase at a Co/(Cu + Co) ratio of 0.65. Therefore, the higher reaction rate and BDO
selectivity obtained with the Coꢀrich CuCo/TiO2 bimetallic catalysts is likely due to the formation of
CuCo alloy shell nanoparticles.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4 Conclusions
It was demonstrated that GBL can be selectively hydrogenated into BDO in a 1,4ꢀdioxane solvent
with CuꢀCo/TiO2 catalysts. The rate of GBL hydrogenation of GBL increases with increasing Co
content with a maximum at around 90 mol% Coꢀ10 mol% Cu. The Cu0.1Co0.9/TiO2 bimetallic
catalyst had ~11, 2.5 and 20 times higher activity than a commercial copper chromite, monometallic
Co and monometallic Cu catalysts, respectively. A small amount of Cu (~10 mol%) in the bimetallic
catalyst increased the reducibility of the Co and resulted in the formation of Coꢀrich core and CuCo
alloy nearꢀsurface shell nanoparticles, leading to higher activity and selectivity of the catalyst. GBL
hydrogenation in ethanol and water produced EHB and 4ꢀHBA, respectively, as the major products.
GBL hydrogenation in 1,4ꢀdioxane produced up to 95% yield of BDO with Cu0.1Co0.9/TiO2 in a batch
reactor under relatively mild conditions of 140˚C and 3.4 MPa H2. The Cu0.1Co0.9/TiO2 also exhibited
good stability for GBL hydrogenation in a continuous flow reactor, maintaining about 80% of its
activity after 150 h timeꢀon stream. The rate of hydrogenation of 2ꢀHTHF and butyraldehyde (a
24
ACS Paragon Plus Environment

Page 25 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
model of 4ꢀHB) in 1,4ꢀdioxane is ~700 and ~400 times greater than the rate of hydrogenation of
GBL, respectively, indicating GBL hydrogenation likely goes through 2ꢀHTHF and 4ꢀHB
intermediates. In 1,4ꢀdioxane, 2ꢀHTHF undergoes facile ring opening tautomerization to 4ꢀHB, with
approximately 65% of 2ꢀHTHF converting to 4ꢀHB at 140˚C.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Supporting Information
STEM images of CuCo/TiO2ꢀR catalysts at different magnifications; EDS elemental spot scans of
selected areas on Cu0.1Co0.9/TiO2ꢀR and Cu0.1Co0.9/TiO2ꢀR; Spectral image analysis of Cu0.5C0.5/TiO2
bimetallic nanoparticles by STEMꢀHAADF and EELS mapping; NH3ꢀTPD of CuCo/TiO2ꢀR
catalysts with different Co contents and the reference of a commercial CuCrOx catalyst; Effect of
reaction time on the GBL conversion and products selectivity over Cu0.1Co0.9/TiO2ꢀR; TGA of fresh
and spent Cu0.1Co0.9/TiO2 after 150 h timeꢀonꢀstream; STEM image of fresh and spent
Cu0.1Co0.9/TiO2 after 150 h timeꢀon steam.
Acknowledgements
This work is supported by the Department of Energy, Office of Energy Efficiency and Renewable
Energy (EERE), under Award Number DEꢀEE0006878, and the National Science Foundation
Engineering Research Center for Biorenewable Chemicals (CBiRC), under Award No. EECꢀ0813570.
The facilities and instrumentation for microscopy at UWꢀMadison were supported by the University
of Wisconsin Materials Research Science and Engineering Center (DMRꢀ1121288). Zhiwei Huang
acknowledges the supports from China Scholarship Council for the Visiting Scholars Program (File
No. 201604910003), the National Natural Science Foundation of China (Grant No. 21473224) and
the Youth Innovation Promotion Association, CAS.
References
1. Grand View Research, Inc. 1,4ꢀButanediol (BDO) Market Analysis By Application
(Tetrahydrofuran (THF), Polybutylene Terephthalate (PBT), GammaꢀButyrolactone (GBL),
Polyurethane (PU)), By Region (North America, Europe, Asia Pacific, CSA, MEA), And Segment
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 29
1
2
3
4
5
6
Forecasts, 2014ꢀ2025. http://www.grandviewresearch.com/industryꢀanalysis/1ꢀ4ꢀbutanediolꢀmarket
(accessed March 2017).
7
8
9
2. Sun, D.; Sato, S.; Ueda, W.; Primo, A.; Garcia, H.; Corma, A. Green Chem. 2016, 18,
2579ꢀ2597.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3. Haas, T.; Jaeger, B.; Weber, R.; Mitchell, S. F.; King, C. F. Appl. Catal. A 2005, 280, 83ꢀ88.
4. Li, F.; Lu, T.; Chen, B.; Huang, Z.; Yuan, G. Appl. Catal. A 2014, 478, 252ꢀ258.
5. Chen, L. F.; Guo, P. J.; Zhu, L. J.; Qiao, M. H.; Shen, W.; Xu, H. L.; Fan, K. N. Appl. Catal. A
2009, 356, 129ꢀ136.
6. Ohlinger, C.; KraushaarꢀCzarnetzki, B. Chem. Eng. Sci. 2003, 58, 1453ꢀ1461.
7. Meyer, C. I.; Regenhardt, S. A.; Marchi, A. J.; Garetto, T. F. Appl. Catal. A 2012, 417, 59ꢀ65.
8. Schlander, J. H.; Turek, T. Ind. Eng. Chem. Res. 1999, 38, 1264ꢀ1270.
9. Pillai, U. R.; SahleꢀDemessie, E.; Young, D. Appl. Catal. B 2003, 43, 131ꢀ138.
10. Ding, G.; Zhu, Y.; Zheng, H.; Zhang, W.; Li, Y. Catal. Commun. 2010, 11, 1120ꢀ1124.
11. Besson, M.; Gallezot, P.; Pinel, C. Chem. Rev. 2014, 114, 1827ꢀ1870.
12. Kang, K. H.; Han, S. J.; Lee, J. W.; Kim, T. H.; Song, I. K. Appl. Catal. A 2016, 524, 206ꢀ213.
13. Liu, X.; Wang, X.; Xu, G.; Liu, Q.; Mu, X.; Liu, H. J. Mater. Chem. A 2015, 3, 23560ꢀ23569.
14. Li, X.; Jia, P.; Wang, T. ACS Catal. 2016, 6, 7621ꢀ7640.
15. Luque, R.; Clark, J. H.; Yoshida, K.; Gai, P. L. Chem. Commun. 2009, 5305ꢀ5307.
16. CorbelꢀDemailly, L.; Ly, B. K.; Minh, D. P.; Tapin, B.; Especel, C.; Epron, F.; Cabiac, A.;
Guillon, E.; Besson, M.; Pinel, C. ChemSusChem 2013, 6, 2388ꢀ2395.
17. Kang, K. H.; Hong, U. G.; Bang, Y.; Choi, J. H.; Kim, J. K.; Lee, J. K.; Han, S. J.; Song, I. K.
Appl. Catal. A 2015, 490, 153ꢀ162.
18. Minh, D. P.; Besson, M.; Pinel, C.; Fuertes, P.; Petitjean, C. Top. Catal. 2010, 53, 1270ꢀ1273.
19. Li, X.; Lan, X.; Wang, T. Green Chem. 2016, 18, 638ꢀ642.
20. AlonsoꢀFagúndez, N.; Ojeda, M.; Mariscal, R.; Fierro, J. L. G.; López Granados, M. J. Catal.
2017, 348, 265ꢀ275.
21. Li, X.; Hoa, B.; Zhang, Y. Green Chem. 2016, 18, 2976ꢀ2980
22. Fuchigami, T.; Wakasa, N.; Iwai, N. J.P. Patent 7082189A, 1995.
26
ACS Paragon Plus Environment

Page 27 of 29
ACS Catalysis
1
2
3
4
23. Thomas, D.; Taylor, P. D. U.S. Patent 4797382, 1989.
5
6
24. Bertola, A. U.S. Patent 6191322, 2001.
7
8
9
25. Cawse, J. N.; Johnson, N. E.; Whitaker, M. T. U.S. Patent 4652685, 1987.
26. Li, W.; Xie, J. H.; Yuan, M. L.; Zhou, Q. L. Green Chem. 2014, 16, 4081ꢀ4085.
27. Brentzel, Z. J.; Barnett, K. J.; Huang, K.; Maravelias, C. T.; Dumesic, J. A.; Huber, G. W.
ChemSusChem 2017, 10, 1351ꢀ1355.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
28. Huang, K.; Brentzel, Z. J.; Barnett, K. J.; Dumesic, J. A.; Huber, G. W.; Maravelias, C. T. ACS
Sustainable Chem. Eng. 2017, 5, 4699ꢀ4706.
29. Burt, S. P.; Barnett, K. J.; McClelland, D. J.; Wolf, P.; Dumesic, J. A.; Huber, G. W.; Hermans, I.
Green Chem. 2017, 19, 1390ꢀ1398.
30. Leite, L.; Lebedevs, A.; Stonkus, V.; Fleisher, M. J. Mol. Catal. A 1999, 144, 323ꢀ328.
31. Zhang, B.; Zhu, Y.; Ding, G.; Zheng, H.; Li, Y. Appl. Catal. A 2012, 443, 191ꢀ201.
32. Ichikawa, N.; Sato, S.; Takahashi, R.; Sodesawa, T.; Inui, K. J. Mol. Catal. A 2004, 212,
197ꢀ203.
33. Hamminga, G. M.; Mul, G.; Moulijn, J. A. Chem. Eng. Sci. 2004, 59, 5479ꢀ5485.
34. Lee, J.; Burt, S. P.; Carrero, C. A.; AlbaꢀRubio, A. C.; Ro, I.; O’Neill, B. J.; Kim, H. J.; Jackson,
D. H. K.; Kuech, T. F.; Hermans, I.; Dumesic, J. A.; Huber, G. W. J. Catal. 2015, 330, 19ꢀ27.
35. Bosman, M.; Watanabe, M.; Alexander, D. T. L.; Keast, V. J. Ultramicroscopy 2006, 106,
1024ꢀ1032.
36. Potapov, P.; Longo, P.; Okunishi, E. Micron 2017, 96, 29ꢀ37.
37. Su, J.; Mao, W.; Xu, X. C.; Yang, Z.; Li, H.; Xu, J.; Han, Y. F. AIChE J. 2014, 60, 1797ꢀ1809.
38. Wang, J.; Chernavskii, P. A.; Khodakov, A. Y.; Wang, Y. J. Catal. 2012, 286, 51ꢀ61.
39. Liu, G.; Pan, D.; Niu, T.; Cao, A.; Yue, Y.; Liu, Y. RSC Adv. 2015, 5, 31637ꢀ31647.
40. Xiao, K.; Qi, X.; Bao, Z.; Wang, X.; Zhong, L.; Fang, K.; Lin, M.; Sun, Y. Catal. Sci. Technol.
2013, 3, 1591ꢀ1602.
41. Bertella, F.; Concepción, P.; Martínez, A. Catal. Today 2017, 289, 181ꢀ191.
42. Jongsomjit, B.; Wongsalee, T.; Praserthdam, P. Catal. Commun. 2005, 6, 705ꢀ710.
43. Srivastava, S.; Jadeja, G. C.; Parikh, J. RSC Adv. 2016, 6, 1649ꢀ1658.
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 29
1
2
3
4
5
6
44. ÁvilaꢀNeto, C. N.; Zanchet, D.; Hori, C. E.; Ribeiro, R. U.; Bueno, J. M. C. J. Catal. 2013, 307,
222ꢀ237.
7
8
9
45. Anton, J.; Nebel, J.; Song, H.; Froese, C.; Weide, P.; Ruland, H.; Muhler, M.; Kaluza, S. Appl.
Catal. A 2015, 505, 326ꢀ333.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
46. Gao, W.; Zhao, Y.; Chen, H.; Chen, H.; Li, Y.; He, S.; Zhang, Y.; Wei, M.; Evans, D. G.; Duan, X.
Green Chem. 2015, 17, 1525ꢀ1534.
47. Ahmed, J.; Ganguly, A.; Saha, S.; Gupta, G.; Trinh, P.; Mugweru, A. M.; Lofland, S. E.;
Ramanujachary, K. V.; Ganguli, A. K. J. Phys. Chem. C 2011, 115, 14526ꢀ14533.
48. Liu, G.; Niu, T.; Pan, D.; Liu, F.; Liu, Y. Appl. Catal. A 2014, 483, 10ꢀ18.
49. Wang, L. L.; Johnson, D. D. J. Am. Chem. Soc. 2009, 131, 14023ꢀ14029.
50. Xiang, Y.; Chitry, V.; Liddicoat, P.; Felfer, P.; Cairney, J.; Ringer, S.; Kruse, N. J. Am. Chem. Soc.
2013, 135, 7114ꢀ7117.
51. Ding, D.; Wang, J.; Xi, J.; Liu, X.; Lu, G.; Wang, Y. Green Chem. 2014, 16, 3846ꢀ3853.
52. Scancar, J.; Milacic, R. J. Anal. At. Spectrom. 2014, 29, 427ꢀ443.
53. Ma, Y.; Hooda, P. S. Chromium, Nickel and Cobalt. In Trace Elements in Soils; Hooda, P. S.,
Ed.; Blackwell Publishing, Ltd: Hoboken, NJ, 2010; pp 470ꢀ472.
54. Ren, D.; Wan, X.; Jin, F.; Song, Z.; Liu, Y.; Huo, Z. Green Chem. 2016, 18, 5999ꢀ6003.
55. Reddy, K. H. P.; Anand, N.; Prasad, P. S. S.; Rao, K. S. R.; Raju, B. D. Catal. Commun. 2011, 12,
866ꢀ869.
56. Guo, P. J.; Chen, L. F.; Yan, S. R.; Dai, W. L.; Qiao, M. H.; Xu, H. L.; Fan, K. N. J. Mol. Catal.
A 2006, 256, 164ꢀ170.
57. Christian, R. V.; Brown, H. D.; Hixon, R. M. J. Am. Chem. Soc. 1947, 69, 1961ꢀ1963.
58. Rocha, A. L.; Solórzano, I. G.; Vander Sande, J. B., Mater. Sci. Eng. C 2007, 27, 1215ꢀ1221.
59. Carenco, S.; Tuxen, A.; Chintapalli, M.; Pach, E.; Escudero, C.; Ewers, T. D.; Jiang, P.;
Borondics, F.; Thornton, G.; Alivisatos, A. P.; Bluhm, H.; Guo, J.; Salmeron, M., J. Phys. Chem. C
2013, 117, 6259ꢀ6266.
60. Prieto, G.; Beijer, S.; Smith, M. L.; He, M.; Au, Y.; Wang, Z.; Bruce, D. A.; de Jong, K. P.;
Spivey, J. J.; de Jongh, P. E., Angew. Chem. Int. Ed. 2014, 53, 6397ꢀ6401.
28
ACS Paragon Plus Environment

Page 29 of 29
ACS Catalysis
1
2
3
4
5
6
61. Bulut, A.; Yurderi, M.; Ertas, Đ. E.; Celebi, M.; Kaya, M.; Zahmakiran, M., Appl. Catal. B 2016,
180, 121ꢀ129.
7
8
9
62. Li, J.; Zhu, Q. L.; Xu, Q., Catal. Sci. Technol. 2015, 5, 525ꢀ530.
63. Wen, C.; Cui, Y.; Yin, A.; Fan, K.; Dai, W. L., ChemCatChem 2013, 5, 138ꢀ141.
64. Cao, A.; Liu, G.; Yue, Y.; Zhang, L.; Liu, Y., RSC Adv. 2015, 5, 58804ꢀ58812.
65. Chen, B.; Li, F.; Huang, Z.; Yuan, G., Appl. Catal. B 2017, 200, 192ꢀ199.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table of Contents
29
ACS Paragon Plus Environment