Highly Selective Silica-supported Copper Catalysts Derived from Copper Phyllosilicates in the Hydrogenation of Adipic Acid to 1,6-hexanediol

Article abstract of DOI:10.1002/cctc.201801580

Hydrogenation of adipic acid (AA) is a potential way to prepare 1,6-hexanediol (HDOL). Herein, silica-supported copper catalysts derived from copper phyllosilicates were synthesized, characterized, and tested in the hydrogenation of AA to HDOL. In a full conversion of AA, a high yield of HDOL (approximately 90 %) was obtained through the use of each Cu-based catalyst. The turnover frequency calculated according to the consumption rate of AA was discovered to be dependent on the Cu⁺/(Cu⁰+Cu⁺) ratio and the reduction condition. Moreover, recycling tests showed that a catalyst derived from a copper phyllosilicate had better stability compared with Cu/SiO₂ made through impregnation due to a stronger interaction between Cu and silica. The kinetic analyses based on the global rate expression and Langmuir?Hinshelwood?Hougen?Watson (LHHW) formalism of AA conversion were conducted. The possible competitive adsorption between AA and its derivatives, including hydroxycaproic acid and ?-caprolactone, was thereby revealed.

Full text of DOI:10.1002/cctc.201801580

Accepted Article
Title: Highly selective silica-supported copper catalysts derived from
copper phyllosilicates in the hydrogenation of adipic acid to 1,6-
hexanediol
Authors: Jia-Wei Jiang, Cheng-Chieh Tu, Chao-Huang Chen, and Yu-
Chuan Lin
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: ChemCatChem 10.1002/cctc.201801580
Link to VoR: http://dx.doi.org/10.1002/cctc.201801580
A Journal of
www.chemcatchem.org

ChemCatChem
10.1002/cctc.201801580
Highly selective silica-supported copper catalysts
derived from copper phyllosilicates in the
hydrogenation of adipic acid to 1,6-hexanediol
1
1
2
1, *
Jia-Wei Jiang , Cheng-Chieh Tu , Dr. Chao-Huang Chen , and Prof. Dr. Yu-Chuan Lin
1
Department of Chemical Engineering, National Cheng Kung University, Tainan 70101, Taiwan
2
Division of Chemical Engineering, Material and Chemical Research Laboratories, Industrial
Technology Research Institute, Hsinchu 30011, Taiwan
*
Corresponding author’s Tel.: (886) 6 275 7575 ext. 62668; Fax: (886) 6 234 4496; E-mail:
yclin768@mail.ncku.edu.tw
1
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Abstract
Hydrogenation of adipic acid (AA) is a potential way to prepare 1,6-hexanediol (HDOL).
Herein, silica-supported copper catalysts derived from copper phyllosilicates were synthesized,
characterized, and tested in the hydrogenation of AA to HDOL. In a full conversion of AA, a
high yield of HDOL (approximately 90%) was obtained through the use of each Cu-based
catalyst. The turnover frequency calculated according to the consumption rate of AA was
+
0
+
discovered to be dependent on the Cu /(Cu +Cu ) ratio and the reduction condition. Moreover,
recycling tests showed that a catalyst derived from a copper phyllosilicate had better stability
2
compared with Cu/SiO made through impregnation due to a stronger interaction between Cu
and silica. The kinetic analyses based on the global rate expression and Langmuir–Hinshelwood–
Hougen–Watson (LHHW) formalism of AA conversion were conducted. The possible
competitive adsorption between AA and its derivatives, including hydroxycaproic acid and ε-
caprolactone, was thereby revealed.
Keywords: adipic acid; copper; 1,6-hexandiol; hydrogenation; phyllosilicate
2
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Introduction
Adipic acid (AA) is an important chemical used in polymer synthesis and food
[1]
manufacturing. The production of AA through oxidation of ketone-alcohol (KA) oil with nitric
[
1]
[2]
acid is industrialized and with gaseous oxygen is under development. Lately, the method for
AA to derive from renewable feedstocks such as 5-hydroxymethylfurfural and lignin fragments
[
3]
has been explored. The productivity of AA grows steadily due to the blooming coal-based
[4]
industry.
The increasing productivity and decreasing market price of AA carves a niche for AA as
a raw material to synthesize high-value chemicals. 1,6-Hexanediol (HDOL), a versatile chemical
being used in polyurethanes and coatings, can be synthesized from AA. The global market of
[
5]
HDOL is expected to grow from USD 727.8 million in 2016 to USD 1,042.1 million by 2021.
The conventional way to prepare HDOL from AA is through a two-step process: esterification of
AA and methanol for preparing dimethyl adipate (DMA) and selective hydrogenation of DMA
[
6]
for producing HDOL. However, energy-intensive operations such as separation of DMA and
[
7]
[8]
water and distillation of crude products for purifying HDOL are required, making the two-
step process economically unattractive.
Hydrogenation of AA to HDOL has the potential to replace the integrated esterification-
hydrogenation process. The major obstacle of hydrogenation of AA is the rigid (low
[
9]
electrophilicity) carbonyl carbon, which hinders the polarization of carbonyl groups.
[7, 10]
[6, 11]
[12]
[6,
Bifunctional catalysts, i.e., combinations of noble metals (such as Ru,
Rh,
Pd, or Pt
1
2-13]
[7, 10a, 12]
[6, 10b, 11, 13]
[6, 11a]
[6, 11a]
)
and oxophilic oxides (such as SnO
x
,
MoO
x
,
WO
x
,
or ReO
x
) are
commonly used. The intimacy between a metal and an oxide promoter weakens the strength of
[9, 14]
[10a]
the C=O bond and facilitates hydride attack on the adsorbed carboxylate.
synthesized RuSn/Al catalysts with different methods including sol-gel and impregnation.
They claimed that the Ru-enriched surface is favorable to enhance the HDOL yield (ca. 71%) in
Toba et al.
2 3
O
[15]
AA hydrogenation. Takeda et al.
reported that the presence of AA affected the degree of
reduction of PdRe/SiO
production.
2
and thus resulted in different hydrogenation activities in HDOL
3
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Precious-metal-based catalysts have not yet been commercialized in the fatty alcohol
[16]
[6-7, 10a]
industry.
Costly prices, unsatisfying HDOL yields (mostly lower than 80%
), and
[
17]
possible corrosion for stainless reactors could be the hurdles. Copper chromite catalysts have
[
16]
been widely deployed in the production of alcohols from hydrogenation of esters. Copper-
[18]
[19]
[20]
based catalysts such as CuZn, CuMn, and CuFe have also been developed, and some of
them displayed better performances than copper chromite did. Surprisingly, limited efforts have
been dedicated to utilize Cu-based catalysts in hydrogenation of carboxylic acids to alcohols.
[
16, 21]
Formation of copper soap and sintering of copper could be the challenges.
On the premise
of enhanced stability, Cu-based catalysts should be promising in direct hydrogenation of AA to
+
0
HDOL. The advantage of using Cu-based catalysts is that the Cu -Cu pair site is highly
[
22]
+
selective in hydrogenation of the C=O bond: Cu acts as an electrophilic site to polarize the
0
C=O bond while Cu dissociates H
insertion.
2
to form adsorbed hydrides for the subsequent hydride
This study investigated silica-supported Cu catalysts derived from copper phyllosilicates
CuPSs) in the hydrogenation of AA to HDOL, and then compared the CuPSs-derived catalysts
(
with Cu/SiO
had a high yield of HDOL at approximately 90%. However, the use of catalyst derived from
CuPS displayed a better stability (reusability) than the use of Cu/SiO made through
impregnation. The strong interaction between Cu and silica formed by CuPSs not only improves
2
synthesized by impregnation. In a full conversion of AA, each Cu-based catalyst
2
+
0
+
catalyst stability, but also increases the surface Cu /(Cu +Cu ) ratio, which is likely to enhance
hydrogenation activity. In addition, a plausible reaction network of AA hydrogenation was
established and a kinetic analysis was conducted. To our knowledge, this is the first-time report
on deploying CuPS-derived catalysts in hydrogenation of AA to HDOL, and this report could
pave the way for hydrogenation of carboxylic acids to their respective alcohols.
4
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Experimental
Materials
[22b, 23]
CuPSs were prepared by the ammonia evaporation hydrothermal method.
5 wt% of ammonia aqueous solution and appropriate amount of Cu(NO •3H O were mixed in
0 mL of deionized water under vigorous stirring. 2.5 gram of colloid silica (LUDOX® HS30,
Briefly,
2
7
3
)
2
2
Aldrich) was then added into the mixture with controlled pH value in the range of 11 to 12 and
o
stirred for 24 h. The yielded suspension was heated in a water batch at 80 C to allow ammonia
evaporation gradually. The yielded mixture was transferred into a hydrothermal bomb and
o
treated at 150 C for 24 h. The precipitates were filtered, washed with deionized water three
o
o
times, dried at 120 C for 4 h, and calcined at 400 C in air (50 mL/min) for 4 h. The yielded
CuPSs, corresponding to 10 wt%, 20 wt%, and 30 wt% of Cu loadings, were denoted as 10CuPS,
2
2 2
0CuPs, and 30CuPS, respectively. For comparison, 20 wt% Cu/SiO (denoted as Cu/SiO ) was
prepared by incipient wetness method. Prior to the activity evaluation, each catalyst was reduced
o
in a fix-bed system at 350 C in an 80% H
2
/N
2
stream for 3 h. After cooling to ambient
temperature in a He stream (100 mL/min), the reduced sample was passivated in a 2% O
2
/He
stream for 4 h. The passivated catalysts were suffixed with “-red” in the following text.
Characterizations
The compositions of Cu and Si were quantified by the inductively coupled plasma-atomic
emission spectrometry (ICP-AES, Kontron S-35). The crystallographic structures were verified
by X-ray diffraction (XRD) patterns obtained by a Rigaku D/Max-IIB diffractometer using Cu
2
Kα radiation. The porosities were determined by the N adsorption-desorption isotherms
acquired using an automated N physisorption analyzer (Micromeritics ASAP 2010 Plus).
2
Surface morphology was observed by a high-resolution transmission electron microscopy (HR-
TEM, JEOL JEM-2010). X-ray photoelectron spectra (XPS) of Cu 2p level were obtained by
using a Kratos Axis Ultra DLD equipped with a focused monochromatic Al-Kα X-ray (1486.7
eV) source. Binding energy shift was corrected by using the C 1s signal at 285.0 eV.
Temperature-programmed reduction of H
2
2
(H -TPR) was performed on a chemisorption analyzer
(
(
Micromeritics, AutoChem II), and the signals were recorded by a thermal conductivity detector
TCD). The amount of hydrogen consumption was calibrated by using the H
2
-TPR profile of
5
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Ag
2
O at the same condition of tested catalysts. Detailed pretreatments and operating conditions
[24]
were reported in earlier studies.
The infrared (IR) spectra of CO-adsorbed samples were
recorded by using a Thermo Scientific Nicolet iS50 spectrometer equipped with a transmission
o
mode in situ quartz cell at 30 C. The cell was connected to a vacuum system which can be
-3
operated with a residual pressure at approximately 2.5*10 Pa. Each sample was pelletized into a
5-mm diameter disk of 0.2 mm thickness, weighting approximately 40 mg. The absorbance
1
-1
spectrum was recorded by collecting 64 scans at 4 cm resolution using a mercuric cadmium
telluride (MCT) detector. Detailed procedures of IR study of CO adsorption can be found in
Appendix A of the Supporting Information.
Reactivity Evaluation
A batch-type reactor (Parr model 4561) was used to evaluate the catalytic performance.
The mixture containing 0.2 g of AA, 60 mL of 1,4-dioxane, and 0.1 g of passivated catalyst with
a particle size ranged from 0.125 mm to 0.180 mm (120 to 80 mesh sizes) was charged and then
2
sealed in the reactor. Residual air inside the reactor was expelled by a N stream (100 mL/min)
for 10 min for three times, followed by purging with a H
2
stream (100 mL/min) for 10 min for
o
three times. The reactor was then pressurized to 58 bar by H
2
and heated to 240 C, yielding an
o
autogenous pressure at approximately 90 bar. Once the system attends 240 C, this point was set
as the outset (t = 0) and the stirring system was started with an agitating rate of 500 rpm. After
attending the proper reaction time, the reactor was immediately quenched in ice water. For the
recycling test, each run was performed for 16 h. The recycled catalyst was washed with 1,4-
o
dioxane three times, dried at 100 C overnight, and subjected to the above-mentioned reduction
and passivation treatment prior to the next recycling test. The quantification of the reactant and
products was conducted by an external standard method by using a gas chromatograph (GC,
Shimadzu 2010 Plus) with a 0.5 μm x 60 m SH-Rtx-Wax capillary column. The definitions of
conversion of AA and product yield are as below:
Moles of AA consumed
Conversion of AA (%) =
100%
Moles of AA fed in
(
Eq. 1)
Moles of product i formed
Yield of product i (%) =
100%
Moles of AA fed in
(
Eq. 2)
6
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
To reveal the intrinsic activity of AA hydrogenation, the TOF was calculated as:
Moles of AA consumed
TOF =
0
+
(
Time, h)(Moles of the sum of Cu and Cu )
(
Eq. 3)
Internal and external mass transfer limitations were evaluated by varying the catalyst
article size (0.425 mm to 0.075 mm) and the agitating rate (400 rpm to 600 rpm), and both were
found to be trivial (Figures S1 and S2). Kinetic parameters were estimated by the nonlinear
[25]
regression using Athena Visual Studio based on the criterion of the least residual sum of the
squares (RSS).
Results and discussion
Physicochemical Properties
bulk
Table 1 lists the Cu contents and the ratios of bulk Cu/Si ((Cu/Si) ) of CuPSs and
2
Cu/SiO measured by ICP-AES. The actual Cu loading is close to its nominal value.
Table 1. Compositions and textural properties of tested samples
bulk
2
3
Sample
Cu (wt%)
(Cu/Si)
S
BET (m /g)
p
V (cm /g)
1
2
3
0CuPS / 10CuPS-red
0CuPS / 20CuPS-red
0CuPS / 30CuPS-red
10.6
20.5
29.5
19.8
0.16
0.36
0.56
0.23
316 / 197
348 / 309
470 / 408
218 / 197
0.98 / 0.83
1.03 / 0.84
1.29 / 0.81
1.14 / 0.98
2 2
Cu/SiO / Cu/SiO -red
2
Figure 1 (a) shows the XRD patterns of CuPSs and reduced CuPSs. Cu/SiO and
o
Cu/SiO
2
-red were included for reference. For Cu/SiO
2
, diffraction peaks of CuO at 35.6 and
estimated with the Scherrer
o
3
8.7 can be speculated. The CuO crystallite size of Cu/SiO
2
7
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
o
o
o
o
equation was approximately 19.2 nm. For CuPSs, weak signals at 30.8 , 35.0 , 57.5 , and 62.4 ,
corresponding to the diffraction pattern of Cu Si (OH) , were identified. All samples had a
broad signal in the 2θ range from 20 to 30 , suggesting the presence of amorphous SiO
2
2
O
5
2
o
o
2
.
Figure 1 (b) displays the XRD patterns of the passivated samples. The peaks of metallic
o
o
o
Cu at 43.2 , 50.4 , and 74.2 can be seen. The estimated Cu sizes were approximately 5.8 nm for
20CuPS-red, 6.9 nm for 30CuPS-red, and 17.2 nm for Cu/SiO
2
-red. The crystallite size of Cu of
10CuPS-red could not be estimated due to the limitation of XRD technique in characterizing Cu
[
26]
particles smaller than 4 nm. The diffraction of the (1 1 1) plane of Cu
2
O can be observed at 2θ
o
=
36.3 for each reduced CuPS. The estimated crystallite size of Cu
2
O of 20CuPS-red was
approximately 5.4 nm; 30CuPS-red, 5.6 nm. The signals of CuO were undetectable, implying
little amounts of CuO and/or small CuO clusters were existed in the reduced catalysts.
8
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Figure 1. XRD patterns of as-synthesized copper phyllosilicates and their respective passivated
Cu-based catalysts derived from reduction of copper phyllosilicates. Cu/SiO and passivated
Cu/SiO -red were included for references.
2
2
9
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Table 1 lists the textural properties, and Figure 2 shows the N
2
adsorption–desorption
isotherms and pore size distributions (PSD) calculated by the Barrett-Joyner-Halenda (BJH)
equation using the desorption branches. CuPSs possessed relatively higher surface areas than
Cu/SiO
2
did, and the surface areas of CuPSs displayed an ascending trend with an increase of Cu
2
2
loading from 316 m /g (10CuPS) to 470 m /g (30CuPS). Measured pore volumes were similar,
3
3
ranging from 0.98 cm /g to 1.29 cm /g.
[
27]
Each sample displayed a type IV isotherm.
The bimodal pore structure showed the
mesopore diameter centered at 3.2 nm and 8.9 nm for 10CuPS; at 3.1 nm and 14.6 nm for
0CuPS; at 3.1 nm and 9.1 nm for 30CuPS. The smaller mesopores at approximately 3 nm were
related to the narrow parallel plates of CuPS, while the larger mesopores could be attributed to
2
[
28]
the void space of agglomerated particles. Apparently, 10CuPS contains more large mesopores
(
approximately 96% of the mesoporosity) than 20CuPS (approximately 80%) and 30CuPS
approximately 65%) do.
(
The reduced samples displayed smaller surface areas and pore volumes compared with
their calcined counterparts. A considerable loss of surface area (approximately 38%) was
observed for 10CuPS-red compared with 10CuPS. 10CuPS-red also displayed a different type of
hysteresis loop compared with 10CuPS; the difference was possibly due to the collapse of
2
lamellar structures after the treatment of reduction. The PSD of Cu/SiO -red and 10CuPS-red
displayed an average mesopore diameter of 35 nm and 25 nm, respectively. The average pore
diameter of the smaller mesopores of 20CuPS-red and 30CuPs-red was approximately 2 nm, and
the average pore sizes of the larger mesopores of these two catalysts were nearly unaltered. The
decrease of the mean size of the smaller pores implied the shrinkage of interlayer distance;
however, the lamellar structure could still hold. Similar porosities of 20CuPS and 30CuPS and
the reduced forms of both CuPSs suggest that their phyllosilicate structures are more rigid than
that of 10CuPS.
10
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Figure 2. N
2
adsorption–de
sorption iso
therms and
pore size distribution
of tested cat
alysts.
Figure 3 exhibits the TEM images and the histograms of Cu particle size distribution of
[
29]
reduced catalysts. The randomly oriented filamentous structure was observed for 20CuPS-red
and 30CuPS-red. The dark, non-symmetrical Cu particles of Cu/SiO -red, 10CuPS-red, 20CuPS-
2
red, and 30CuPS-red showed their average diameters of 8.3 nm, 3.3 nm, 4.0 nm, and 5.3 nm,
respectively. The lattice spacing value of 0.2 nm, indexing the Cu (1 1 1) plane, evidences the
presence of Cu particles. Figure S3 shows the patterns of energy dispersive spectroscopy (EDS)
mapping of O, Si, and Cu of reduced catalysts. Homogeneously distributed O, Si, and Cu could
be speculated. Among reduced CuPSs, well-dispersed Cu could be found on 10CuPS-red,
whereas agglomerated Cu could be observed on 20CuPS-red and 30CuPS-red.
11
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
(
a)
(
b)
(
c)
(
d)
Figure 3. TEM images and histograms of Cu particle size distribution of (a) Cu/SiO
0CuPS-red, (c) 20CuPS-red, and (d) 30CuPS-red.
2
-red, (b)
1
12
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
The TEM images of CuPSs were also recorded (Figure S4); however, Cu particle could
not be identified clearly. Formation of immobilized Cu during the reduction treatment of CuPSs
(
known as exsolution) is likely to cause the differences of resolutions of Cu particles between
[
30]
CuPSs and reduced CuPSs.
Figure 4 (a) exhibits the XPS spectra of Cu 2p level of calcined samples. The surface
surf
Cu/Si ratio ((Cu/Si) ) estimated by the intensity ratio of Cu 2p and Si 2p peaks showed an
2
ascending order as follows: Cu/SiO (0.02) < 10CuPS (0.15) < 20CuPS (0.40) < 30CuPS (0.49).
surf
bulk
The differences between (Cu/Si)
and (Cu/Si)
of CuPSs were small, suggesting
surf
homogeneously dispersed Cu in phyllosilicate structures. In contrast, the value of (Cu/Si) of
bulk
Cu/SiO
2
is approximately ten times smaller than that of (Cu/Si) , indicating a low dispersion of
[
31]
Cu on the surface.
13
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
2 2
Figure 4. Cu 2p XPS spectra of (a) Cu/SiO and CuPSs and (b) reduced Cu/SiO and CuPSs
The spectra of calcined samples showed the shakeup satellite peaks at approximately 943
eV together with the signals of Cu 2p2/3 core level in the range of 933 to 937 eV, indicating the
2
+ [32]
presence of Cu .
The peak of Cu 2p3/2 displayed an ascending order as follows: Cu/SiO
2
14
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
(
933.1 eV) < 10CuPS (935.7 eV) < 20CuPS (936.0 eV) < 30CuPS (936.2 eV). The chemical
shift of the Cu 2p3/2 level to higher binding energy implied a charge transfer from the metal ion
[
33]
toward the support oxide. That is, CuPSs, particularly 30CuPS, should have stronger Cu-SiO
2
[22d]
interactions than that of Cu/SiO
2
.
0
Figure 4 (b) exhibited the XPS spectra of reduced catalysts. The binding energies of Cu
+
and Cu are nearly identical, showing two peaks located at approximately 933 eV and 953 eV of
[34]
2+
Cu 2p2/3 and Cu 2p1/2 core levels. The absence of the satellite peak of Cu suggests that the
2
+
surface concentrations of Cu of reduced catalysts should be negligible. The estimated
surf
(
2
Cu/Si) ratios, listed in Table 2, displayed an ascending order as follows: Cu/SiO -red (0.02) <
surf
1
0CuPS-red (0.04) < 20CuPS-red (0.09) < 30CuPS-red (0.12). The (Cu/Si) ratios of Cu/SiO
2
and Cu/SiO
2
-red were the same, implying the extents of sintering/agglomeration of Cu are trivial.
surf
Considerable decreases of (Cu/Si) ratios were found for reduced CuPSs compared to those of
surf
CuPSs before reduction. The decreases of (Cu/Si) ratios of reduced CuPSs indicated that
exsolution of Cu from phyllosilicate structure could be promoted during reduction. Exsoluted Cu
[
30]
formed immobilized Cu particles on the surface,
resulting in lower Cu-to-Si XPS intensity
ratios of reduced CuPSs than those of CuPSs. The schematic diagram of the formation of
exsoluted Cu particles through the reduction of CuPS was shown in Figure S5 of the Supporting
Information.
Figure 5 displays the H
2
-TPR profile of each calcined sample and the peak area is
-TPR was then exposed to N O to oxidize
the surface of metallic Cu. The sample with the Cu-oxidized surface then went through the
second H -TPR test, and the area of the resulting profile was denoted as A2. The H -TPR profile
of calcined Cu/SiO displayed three responses at 231 C, 256 C, and 300 C. The multi-response
of Cu/SiO is due to different Cu particle sizes. In comparison, CuPSs had major responses at
approximately 250 C and shoulders in the range of 275-350 C. The single TPR peak indicates a
denoted as A1. The sample that went through the H
2
2
[
35]
2
2
o
o
o
2
[36]
2
o
o
[29]
relatively narrow distribution of Cu particle sizes in CuPSs.
15
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Figure 5. The f
irst H
2
-TP
R profiles (solid line)
and the sec
ond H
2
-TP
after N
2
O oxidation of calcined catalysts.
o
The second H
2
-TPR profile showed a broad peak at approximately 170 C for each
[
37]
sample. The H
2
uptake of A2 can be used to calculate the concentration and surface area of
0
19
accessible Cu with the assumption that the specific surface area of Cu is 1.46 * 10
2
[22b]
atoms/m ,
listed in Table 2. Table 2 also presents the dispersion (DCu) of accessible Cu
[37]
calculated by 2*A2/A1*100%. The values of DCu showed a descending trend as the following
2
order: 10CuPS-red (55.3%) > 30CuPS-red (27.4%) > 20CuPS-red (25.3%) > Cu/SiO -red
0
(
11.1%). Apparently, CuPSs derived catalysts had better dispersions of Cu than that of Cu/SiO
2
-
red. Among reduced CuPSs, an outstanding high DCu was achieved by 10CuPS-red, while those
of 20CuPS-red and 30CuPS-red were similar.
16
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
a
Table 2. Chemical properties of Cu species of tested catalysts
surf
+
0
+
Sample
(Cu/Si)
C
Cu0
S
Cu0
D
Cu
C
Cu+
Cu /(Cu +Cu )
(
mmol/g)
(%)
(mmol/g)
2
(
m /g)
37.7
49.6
57.4
18.2
1
2
3
0CuPS-red
0CuPS-red
0CuPS-red
0.04
0.09
0.12
0.02
0.87
1.15
1.33
0.42
55.3
25.3
27.4
11.1
0.15
0.27
0.61
-
0.15
0.19
0.31
-
2
Cu/SiO -red
a
surf
(
Cu/Si) was estimated from the intensities of XPS spectra of Cu 2p and Si 2p levels; surface CCu0 and
S
Cu0 were estimated from the second H
the second H -TPR peak area) / A1 (the first H
was estimated from the IR spectrum of adsorbed CO of each reduced and N
2
-TPR profile after N
2
O oxidation; DCu was estimated by 2* A2
+
(
2
2
-TPR peak area) * 100%; surface concentration of Cu
O-oxidized catalyst.
2
Figure 6 shows the IR spectra of CO adsorption of reduced catalysts and their N
2
O-
-3
oxidized forms at an equilibrium pressure of approximately 2.5*10 Pa. According to the
-1
literature, the IR signals of CO adsorption at approximately 2110 cm can be attributed to CO
0
adsorbed linearly on metallic Cu (forming CO-Cu carbonyls), while the signal at approximately
-1
+
+
[38]
2
130 cm can be assigned to CO adsorbed linearly on Cu (forming CO-Cu carbonyls). The
2
+
-1
unstable CO-Cu species in the interval of 2150-2240 cm can hardly been seen at room
[
39]
temperature.
0
+
After evacuation, the unstable CO-Cu species were removed and CO-Cu carbonyls
[
28, 39a]
+
were preserved on reduced and N
absent on Cu/SiO -red, suggesting negligible amounts of Cu -O-Si units. The Cu cations of
Cu/SiO -red should be mainly in Cu
2
O-oxidized CuPSs.
Noted that the CO-Cu band was
+
+
2
[40]
2
2
O clusters bounded weakly with silica support.
The
+
+
integrated absorbance band of CO-Cu was used to estimate the surface concentration of Cu .
+
The area of CO-Cu band of each reduced sample (AIR1) was smaller than the area (AIR2) of its
0
+ [28, 41]
subsequent N
2
O-treated form due to N
2
O-induced oxidation of Cu to Cu .
Hence, the
O-
difference (ΔAIR) by subtracting AIR1 from AIR2 can be correlated to the concentration of N
2
17
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
0
+
oxidized Cu . The concentration of Cu (CCu+, see Table 2) of each catalyst was then calculated
from the equation: (CCu+) = (AIR1/ΔAIR) (CCu0).
2
Figure 6. IR spectra of CO adsorbed on reduced (solid) and on N O-oxidized (dash) samples at
o
-3
3
0 C with a residual pressure of 2.5*10 Pa.
+
0
+
+
0
+
Table 2 also presents the ratios of Cu /(Cu +Cu ). The Cu /(Cu +Cu ) ratio exhibited an
ascending order as follows: 10CuPS-red (0.15) < 20CuPS-red (0.19) < 30CuPS-red (0.31).
+
Presumably, the surface concentration of Cu can be enhanced by improving the strength of the
Cu-O-Si bond, and the catalysts derived from CuPSs (particularly 30CuPS-red) have stronger
interaction between Cu and silica.
Activity Evaluation
Table 3 presents conversions of AA and the yields of detected products, including
hydroxycaproic acid (HCA), ε-caprolactone (ε-CPL), HDOL, and 1-hexanol (HNOL) by using
18
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
reduced catalysts at different reaction times. Most trials of tested catalysts had mass balance of
carbon greater than 90%. Conversions of AA increased with time, but different catalysts showed
different rates. At low conversions of AA (< 30%), HCA, -CPL and trace HDOL were detected.
With increased reaction time, the yields of HCA and ε-CPL increased. When AA was totally
consumed, HDOL was the major product for each catalyst. A sudden increase of HDOL together
with abrupt declines in HCA and ε-CPL in a full conversion of AA occurred was observed for
each catalyst, and will be discussed later in the section of Kinetic Analysis. The highest HDOL
yield (96.6%) was obtained by using 20CuPS-red for 6 h reaction time. Scheme 1 exhibits the
proposed reaction network
Table 3. Catalytic performances of tested catalysts in AA hydrogenation
Catalyst
Time CB
X
Y (%)
TOF
-1
(
h) (%) (%) HCA -CPL HDOL HNOL (h )
Cu/SiO
2
-red
1
3
6
99.1 12.3
8.1
3.1
8.4
0.2
0
0
0
0.36
99.3 34.2 24.6
77.8 78.4 25.2
0.5
2.8
-
28.2
-
1
6
92.2 100
98.8 9.7
0
0
91.8
0.1
0.4
0
-
10CuPS-red
20CuPS-red
30CuPS-red
1
5.9
2.5
8.5
35.5
0
0.12
3
97.2 27.2 15.7
90.5 79.5 31.0
0.2
0
-
1
2
0
3.5
0
-
-
2
94.8 100
0
94.5
0.1
0.3
0
0.5 94.2 27.1 18.2
3.0
6.5
23.0
0.1
4.8
11.2
20.3
0.1
0.47
1
2
6
93.5 39.7 26.2
88.0 71.8 34.6
0.5
0
2.2
0
-
97.4 100
0
96.6
1.2
0.7
0
-
0.1 93.3 27.8 15.1
.5 98.8 54.9 41.0
91.4 81.0 49.6
88.2 100
1.05
0
1.5
0
-
-
-
1
2.5
0
3
0
87.2
0.9
19
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
3
3
0CuPS-red-300 0.1 98.3 11.3
0CuPS-red-400 0.1 91.2 17.3
8.5
7.5
1.1
1.0
0
0
0
0
0.37
0.73
Scheme 1. Proposed reaction network of hydrogenation of AA by using Cu-based catalysts
[
10a]
(
adapted from Toba et al.
).
Table 3 also presents the TOF values at initial conditions. The TOF showed a descending
-1
-1
trend as the following order: 30CuPS-red (1.05 h ) > 20CuPS-red (0.47 h ) > Cu/SiO
2
-red (0.36
-1
-1
h ) > 10CuPS-red (0.12 h ). 30CuPS-red had the highest TOF, even though it had a DCu close to
+
0
+
that of 20CuPS-red. A possible explanation is that 30CuPS-red had a higher Cu /(Cu +Cu ) ratio
+
0
(
0.31) than that of 20CuPS-red (0.19), yielding a higher concentration of Cu -Cu pairs for the
contact synergy of C=O bond hydrogenation. Contrastively, 10CuPS-red had a significantly low
-1
+
0
+
TOF value (0.12 h ), even though the Cu /(Cu +Cu ) ratio of 10CuPS-red (0.15) was
comparable to that of 20CuPS-red (0.19). Presumably, the outstandingly high DCu (55.3%) of
+
0
1
0CuPS-red leads to a finely dispersed but isolated Cu and Cu sites, resulting in a low
+
0
proximity of Cu and Cu on the surface. A similar catalytic behavior was observed using a low
Cu-loading (8 wt%) catalyst derived from CuPS in furfural hydrogenation to 2-methylfuran. The
low Cu-loading catalyst derived from CuPS displayed an activity lower than that of the catalyst
[
34b]
containing a moderate Cu loading (24 wt%).
2
It should be noted that Cu/SiO -red showed
+
considerable TOF value, even though it contained negligible amounts of Cu species. The
2
noticeable TOF value of Cu/SiO -red implied that the C=O bond of AA can be activated not only
+
0
0
[42]
2
by Cu -Cu pairs, but also by interfaces of adjacent Cu and Cu O.
20
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
To investigate the influence of reduction condition, 30CuPS subjected to different
o
o
reduction treatments (300 C and 400 C for 3 h, denoted as 30CuPS-red-300 and 30CuPS-red-
4
00, respectively) were tested in the hydrogenation of AA at an initial condition. The estimated
0
+
concentrations of Cu (Figure S6) and Cu (Figure S7) are 1.18 mmol/g and 1.08 mmol/g of
30CuPS-red-300 and are 1.43 mmol/g and 0.32 mmol/g of 30CuPS-red-400. The conversion of
-1
-1
AA and TOF of 30CuPS-red-300 were 11.3% and 0.37 h , and were 17.3% and 0.74 h of
-1
3
0CuPS-red-400. These values are much lower than those (27.8% and 1.05 h ) of 30CuPS-red.
Moreover, 30CuPS-red produced higher yields of products, particularly HDOL, than those
generated by using 30CuPS-red-300 and 30CuPS-red-400. The different activities underlined the
importance of reduction treatment, which influences the nature and the contents of exsoluted Cu.
+
0
+
It should be noted that 30CuPS-red-300 had the highest Cu /(Cu +Cu ) ratio at 0.48 but
displayed the lowest TOF value, in contradict to the aforementioned claim that a high TOF lies
+
0
+
in a high surface Cu /(Cu +Cu ) ratio. This mismatch may be attributed to different chemical
+
+
states of copper species: various copper species such as Cu
2
O, Cu -O-Si units, Cu -metal alloy,
0
[43]
and Cu particles could be generated under different reduction conditions. Nevertheless, the
interplay of these copper species and their catalytic performances is still unclear.
Catalyst Recycling
Table 4 presents the three sequential reactions of tested catalysts. The conversion of AA
st
nd
2
using Cu/SiO -red decreased significantly from 64.0% (the 1 cycle) to 26.6% (the 2 cycle) and
rd
st
then to 8.3% (the 3 cycle). The yields of products also varied significantly from the 1 cycle to
rd
the 3 cycle: HCA decreased from 24.8% to 1.0%; ε-CPL decreased from 22.6% to 1.9%;
HDOL decreased from 1.1% to undetectable level. Comparing to the fresh Cu/SiO
XRD pattern (Figure S8) of used Cu/SiO -red showed increased crystallite sizes of metallic Cu
and Cu O from 17.2 nm to 41.5 nm and from < 4 nm to 14.1 nm, respectively. The TPR profiles
Figure S9) displayed an approximately 50% decrease of reducible copper species. Moreover, the
2
-red, the
2
2
(
ICP analysis showed that the Cu loading is approximately 8.9 wt%, which indicated an
approximately 55% loss of Cu content compared with its pristine form. This indicates sintering
2
and Cu leaching of recycled Cu/SiO -red.
21
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
o
2 2
Table 4. Recycling tests of Cu/SiO -red and 20CuPS-red at 240 C with a H pressure of 90 bar
Catalyst
Run
Time
CB
X
Y (%)
-CPL
22.6
6.9
(
h)
(%)
(%)
HCA
HDOL
Cu/SiO
2
-red
1
2
3
1
2
3
1
2
3
1
2
3
5
84.5
90.1
93.6
85.6
86.8
82.0
90.0
88.6
87.5
92.5
89.2
87.8
64.0
26.6
8.3
24.8
9.8
1.1
0
5
5
1.0
1.9
0
1
2
3
0CuPS-red
0CuPS-red
0CuPS-red
12
12
12
2
77.6
67.6
55.1
74.5
69.2
60.1
80.5
77.5
70.6
28.5
31.7
22.0
34.4
30.4
28.1
47.9
43.3
38.4
32.3
22.6
15.1
25.8
24.4
18.2
21.9
20.5
17.6
2.9
0.1
0.1
4.3
3.0
1.3
3.2
2.8
2.4
2
2
1
1
1
st
Comparatively, a little decrease of conversion of AA from 77.6% (the 1 cycle) to 55.1%
rd
st
rd
st
(
the 3 cycle), 74.5% (the 1 cycle) to 60.1% (the 3 cycle), 80.5% (the 1 cycle) to 70.6% (the
rd
3
cycle) were observed by using 10CuPS-red, 20CuPS-red, and 30CuPS-red, respectively.
Moreover, the yields of products were similar for each run by using a CuPS-red. The small
variation of yields of products implied that most of the active phases were unaltered after three
sequential reactions. Comparing to the results of Figure 1(b), the XRD patterns (Figure S8) of
2
recycled 10CuPS-red showed little increases in the crystallite diameters of metallic Cu and Cu O
22
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
from < 4 nm to 8.5 nm and from < 4 nm to 5.2 nm; recycled 20CuPS-red, from 5.8 to 8.6 nm and
from 5.4 nm to 5.5 nm; recycled 30CuPS-red, from 6.9 nm to 7.3 nm and from 5.6 nm to 7.9 nm.
Figure S9 shows the TPR profiles of used CuPSs-red. Compared to the profiles of their freshly
o
prepared counterparts, the peaks of used CuPS-red shifted downward from approximately 250 C
o
to 200 C. The low-temperature shifted peaks indicated destruction of some Cu-O-Si units.
However, the peak areas of fresh and recycled catalysts were similar. The Cu loadings of used
CuPSs-red determined by ICP analyses showed less than 15% losses compared to that of fresh
CuPSs. This highlighted that CuPSs-red had better stability than Cu/SiO
2
-red. To sum up, the
Cu-O-Si bonding of CuPS-red stronger than that of Cu/SiO
stability.
2
is responsible for improving the
Kinetic Analysis
A global kinetic analysis of AA hydrogenation was conducted by using 20CuPS-red.
Detailed derivations (Appendix B), experimental data (Tables S1 and S2), and the comparisons
of observed and model-calculated results (Figure S10) can be found in the Supporting
Information. The differences between experimental and predicted concentrations are
considerable, indicating that a more comprehensive analysis is needed.
Table S3 lists the possible steps of conversion of AA to HDOL and the derived rate
expressions according to the Langmuir–Hinshelwood–Hougen–Watson (LHHW) formalism.
[14b, 44]
[45]
Detailed derivations, assumptions,
and simplifications
we made can be found in the
Appendix C of the Supporting Information. The resultant rate expressions were then fitted with
experimental data (Table S4) via nonlinear regression. Table 5 presents obtained kinetic
parameters, Figure S11 shows the experimental and the predicted concentration profiles of
reactant and products, and Figure S12 exhibits the parity plot. The experimental and the
estimated values are in a good agreement, suggesting that the proposed steps and the
assumptions are rational.
23
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Table 5. Estimated kinetic parameters obtained from the LHHW formalism
Obtained kinetic parameters 20CuPS-red
6
2
k
1
(dm /mol h)
0
1
1
0
0
1
0
.12  0.02
6.3  1.8
3
K
K
AA (dm /mol)
3
HCA (dm /mol)
4.1  1.2
-1
k
k
2
(h )
.04  0.003
.01  0.004
0.0  3.2
6
2
3
(dm /mol h)
3
K
-CPL (dm /mol)
3
K
HDOL (dm /mol)
Through a close inspection of these parameters (see Table 5), we can find that k
1
is three
times higher than k and two orders of magnitude higher than k , implying a faster reaction rate
2
3
of AA conversion to HCA. Moreover, the equilibrium constant of the adsorption and desorption
of AA (KAA, 16.3) is larger than that of HCA (KHCA, 14.1) and that of ε-CPL (Kε-CPL, 10.0),
while KHDOL is negligible. The similar equilibrium constants of AA, HCA, and ε-CPL indicate
that these species are competing for active sites (competitive adsorption) when a reaction
[46]
proceeds. In a pressurized liquid phase system, a full coverage of active sites is common.
Since AA is the reactant, it should occupy most of the available sites initially. Once a high
conversion of AA is achieved, some active sites are then available to adsorb HCA and ε-CPL for
further transformation. The global kinetic analysis does not consider the influence of competitive
adsorption, resulting in a substantial disagreement between the experimental and simulated
results. Furthermore, the negligible KHDOL indicates that HDOL is weakly adsorbed and is
desorbed immediately once it is formed, thereby leading to a high HDOL productivity.
24
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Conclusions
Cu-based catalysts derived from CuPSs were effective in direct hydrogenation of AA to
HDOL. In a full conversion of AA, merely a trace of byproducts and remarkable high HDOL
yields were generated by using CuPS-red catalysts. In addition, better reusabilities were
2
discovered through the tests of CuPSs-red rather than Cu/SiO -red, and the differences are
related to the stronger strengths of the Cu-O-Si bonds of CuPSs-red. The stronger Cu-silica
+
0
+
+
0
+
interaction can also elevate the Cu /(Cu +Cu ) ratio. The Cu /(Cu +Cu ) ratio, the amounts of
0
+
Cu and Cu sites, and the reduction condition could affect the intrinsic activity of hydrogenation
of AA. The outcomes of kinetic analysis based on the LHHW formalism revealed that the
competitive adsorption of AA and its derivatives, i.e., HCA and ε-CPL, could influence the
catalytic behaviors of hydrogenation of AA to HDOL.
Acknowledgements
This study was supported by the Ministry of Science and Technology (Projects 106-2221-E-006-
188-MY3), the Ministry of Economic Affairs (Project H354DP2120), and Taiwan’s Deep
Decarbonization Pathways toward a Sustainable Society (Project AS-KPQ-106-DDPP).
References
[1]
[2]
[3]
A. Castellan, J. C. J. Bart, S. Cavallaro, Catal. Today 1991, 9, 237-254.
A. Mazzi, S. Paul, F. Cavani, R. Wojcieszak, ChemCatChem 2018, 10, 3680-3682.
a
S. Xiaoyan, B. H. M., Z. Xiao, A. B. K., W. Yong, ChemSusChem 2011, 4, 1071-1073;
b
c
S. Van de Vyver, Y. Roman-Leshkov, Catal. Sci. Technol. 2013, 3, 1465-1479; D. R.
Vardon, M. A. Franden, C. W. Johnson, E. M. Karp, M. T. Guarnieri, J. G. Linger, M. J.
Salm, T. J. Strathmann, G. T. Beckham, Energy Environ. Sci 2015, 8, 617-628.
IHS Market. Adipic acid, chemical economics handbook; 2017.
[
4]
5]
[
1,6-Hexanediol Market by Application (Polyurethanes, Coatings, Acrylates, Adhesives,
Polyester Resins, Plasticizers), and Region (Europe, North America, Asia-Pacific, South
25
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
America, and Middle East & Africa) - Global Forecasts to 2021; Markets and Markets:
2017.
[
[
[
[
[
6]
V. J. Murphy, E. L. Dias, J. A. W. Shoemaker, Production of 1,6-hexanediol from adipic
acid. US Patent 20160200646A1, 2015.
7]
Y. Hara, K. Endou, S. Nishimura, Process for preparing 1, 6-hexanediol. US Patent
5
969194A, 1997.
E.L. Dias, V.J. Murphy, J.A.W. Shoemaker, Process For production of adipic acid from
,6-hexanediol. US 2013/0331606 A1, 2013.
8]
1
9]
J. Pritchard, G. A. Filonenko, R. van Putten, E. J. M. Hensen, E. A. Pidko, Chem. Soc.
Rev. 2015, 44, 3808-3833.
a
10]
M. Toba, S.-i. Tanaka, S.-i. Niwa, F. Mizukami, Z. Koppány, L. Guczi, K.-Y. Cheah, T.-
b
S. Tang, Appl. Catal., A 1999, 189, 243-250; Y. Takeda, T. Shoji, H. Watanabe, M.
Tamura, Y. Nakagawa, K. Okumura, K. Tomishige, ChemSusChem 2015, 8, 1170-1178.
a
b
[11]
T. Fuchigami, T. Ga, K. Tawara, N. Wakasa, Japan, 1995; M. Tamura, R. Tamura, Y.
c
Takeda, Y. Nakagawa, K. Tomishige, Chem. Commun. 2014, 50, 6656-6659; M. Tamura,
R. Tamura, Y. Takeda, Y. Nakagawa, K. Tomishige, Chem. Eur. J. 2015, 21, 3097-3107.
[
[
[
[
12] R. H. Fischer, R. Pinkos, F. Stein, Method for producing 1,6-hexanediol and 6-
hydroxycaproic acid or their esters. US Patent 6426438B1, 1997.
13] T. Mizugaki, Y. Nagatsu, K. Togo, Z. Maeno, T. Mitsudome, K. Jitsukawa, K. Kaneda,
Green Chem. 2015, 17, 5136-5139.
a
14]
D. Sun, S. Sato, W. Ueda, A. Primo, H. Garcia, A. Corma, Green Chem. 2016, 18, 2579-
b
2
597; K. Tomishige, Y. Nakagawa, M. Tamura, Green Chem. 2017, 19, 2876-2924.
15] Y. Takeda, M. Tamura, Y. Nakagawa, K. Okumura, K. Tomishige, Catal. Sci. Technol.
016, 6, 5668-5683.
2
[
16] D. S. Thakur, A. Kundu, J. Am. Oil Chem. Soc. 2016, 93, 1575-1593.
17] D. R. Vardon, A. E. Settle, V. Vorotnikov, M. J. Menart, T. R. Eaton, K. A. Unocic, K. X.
Steirer, K. N. Wood, N. S. Cleveland, K. E. Moyer, W. E. Michener, G. T. Beckham,
ACS Catal. 2017, 7, 6207-6219.
[
a
b
[18]
D. S. Thakur, B. D. Roberts, T. J. Sullivan, A. L. Vichek, 1992; L. Jürgen, R. Tiberius,
in Studies in Surface Science and Catalysis, Vol. 121 (Eds.: H. Hattori, K. Otsuka),
Elsevier, 1999, pp. 215-220.
26
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
a
[19]
M. Schneider, K. Kochloefl, G. Maletz, Catalyst and process for hydrogenation of
b
carboxylic acid alkyl esters to higher alcohols, US Patent 5386066A, 1995; J. P. Chen,
Preparation and use of non­chrome catalysts for Cu/Cr catalyst applications, US Patent
6916457, 2005.
[
[
[
20] Y. Hattori, K. Yamamoto, J. Kaita, M. Matsuda, S. Yamada, J. Am. Oil Chem. Soc. 2000,
77, 1283-1288.
a
b
21]
22]
M. V. Twigg, M. S. Spencer, Appl. Catal., A 2001, 212, 161-174; M. Besson, P.
Gallezot, Catal. Today 2003, 81, 547-559.
a
b
A. Yin, X. Guo, W.-L. Dai, K. Fan, J. Phys. Chem. C 2009, 113, 11003-11013; J. Gong,
H. Yue, Y. Zhao, S. Zhao, L. Zhao, J. Lv, S. Wang, X. Ma, J. Am. Chem. Soc. 2012, 134,
c
1
3922-13925; L. He, X. Li, W. Lin, W. Li, H. Cheng, Y. Yu, S.-i. Fujita, M. Arai, F.
d
Zhao, J. Mol. Catal. A., Chem. 2014, 392, 143-149; Y. Zhu, X. Kong, J. Yin, R. You, B.
Zhang, H. Zheng, X. Wen, Y. Zhu, Y.-W. Li, J. Catal. 2017, 353, 315-324.
[23] T. Toupance, M. Kermarec, J.-F. Lambert, C. Louis, J. Phys. Chem. B 2002, 106, 2277-
2286.
[
[
[
24] P.-J. Hsu, J.-W. Jiang, Y.-C. Lin, ACS Sustainable Chem. Eng. 2018, 6, 660-667.
25] W. E. Stewart, M. Caracotsios, J. P. Sørensen, AIChE J. 1992, 38, 641-650.
a
26]
F. Boccuzzi, A. Chiorino, G. Martra, M. Gargano, N. Ravasio, B. Carrozzini, J. Catal.
b
1
997, 165, 129-139; F. Dong, G. Ding, H. Zheng, X. Xiang, L. Chen, Y. Zhu, Y. Li,
Catal. Sci. Technol. 2016, 6, 767-779.
[
[
[
27] K. S. W. Sing, Pure Appl. Chem. 1985, 57, 603-619.
28] W. Di, J. Cheng, S. Tian, J. Li, J. Chen, Q. Sun, Appl. Catal., A 2016, 510, 244-259.
29] L.-F. Chen, P.-J. Guo, M.-H. Qiao, S.-R. Yan, H.-X. Li, W. Shen, H.-L. Xu, K.-N. Fan, J.
Catal. 2008, 257, 172-180.
[30] X. Gong, M. Wang, H. Fang, X. Qian, L. Ye, X. Duan, Y. Yuan, Chem. Commun. 2017,
53, 6933-6936.
[
[
[
31] J. W. Niemantsverdriet, in Spectroscopy in Catalysis, 2007.
32] A. Guerrero-Ruiz, I. Rodriguez-Ramos, J. L. G. Fierro, Appl. Catal. 1991, 72, 119-137.
33] A. Auroux, A. Gervasini, C. Guimon, J. Phys. Chem. B 1999, 103, 7195-7205.
27
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
a
[34]
J. Ghijsen, L. H. Tjeng, J. van Elp, H. Eskes, J. Westerink, G. A. Sawatzky, M. T.
b
Czyzyk, Phys. Rev. B 1988, 38, 11322-11330; F. Dong, Y. Zhu, H. Zhao, Z. Tang, Catal.
Sci. Technol. 2017, 7, 1880-1891.
a
b
[
35]
36]
T. J. Osinga, B. G. Linsen, W. P. van Beek, J. Catal. 1967, 7, 277-279; J. J. F. Scholten,
J. A. Konvalinka, Trans. Faraday Soc. 1969, 65, 2465-2473.
a
[
M. A. Kohler, H. E. Curry-Hyde, A. E. Hughes, B. A. Sexton, N. W. Cant, J. Catal.
b
1
987, 108, 323-333; M. A. Kohler, J. C. Lee, D. L. Trimm, N. W. Cant, M. S.
Wainwright, Appl. Catal. 1987, 31, 309-321.
37] S. Sato, R. Takahashi, T. Sodesawa, K.-i. Yuma, Y. Obata, J. Catal. 2000, 196, 195-199.
[
[
a
b
38]
J. Pritchard, T. Catterick, R. K. Gupta, Surf. Sci. 1975, 53, 1-20; I. A. Fisher, A. T. Bell,
c
J. Catal. 1998, 178, 153-173; J. Schumann, J. Kröhnert, E. Frei, R. Schlögl, A.
Trunschke, Top. Catal. 2017, 60, 1735-1743.
a
b
[
39] K. Hadjiivanov, H. Knözinger, Phys. Chem. Chem. Phys. 2001, 3, 1132-1137; G. Busca,
J. Mol. Catal. 1987, 43, 225-236.
[
40] A. N. Pestryakov, A. A. Davydov, Appl. Surf. Sci. 1996, 103, 479-483.
41] C. J. G. Van Der Grift, A. F. H. Wielers, B. P. J. Jogh, J. Van Beunum, M. De Boer, M.
Versluijs-Helder, J. W. Geus, J. Catal. 1991, 131, 178-189.
[
[42] L. Wu, B. Li, C. Zhao, ChemCatChem 2018, 10, 1140-1147.
[
43] R.-P. Ye, L. Lin, Q. Li, Z. Zhou, T. Wang, C. K. Russell, H. Adidharma, Z. Xu, Y.-G.
Yao, M. Fan, Catal. Sci. Technol. 2018, 8, 3428-3449.
[44] Y. Takeda, M. Tamura, Y. Nakagawa, K. Okumura, K. Tomishige, ACS Catal. 2015, 5,
7034-7047.
a
[45]
E. Crezee, B. W. Hoffer, R. J. Berger, M. Makkee, F. Kapteijn, J. A. Moulijn, Appl.
b
Catal., A 2003, 251, 1-17; J. Zhang, B. Hou, A. Wang, Z. Li, H. Wang, T. Zhang, AIChE
c
J. 2015, 61, 224-238; A. S. Piskun, H. H. van de Bovenkamp, C. B. Rasrendra, J. G. M.
Winkelman, H. J. Heeres, Appl. Catal., A 2016, 525, 158-167.
[46] G. F. Froment, K. B. Bischoff, J. D. Wilde, in Chemical Reactor Analysis and Design,
John Wiley & Sons INC., USA, 2010.
28
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801580
Table of Contents
Silica-supported copper catalysts derived from copper phyllosilicates are selective and stable in
the hydrogenation of adipic acid to 1,6-hexandiol
29
This article is protected by copyright. All rights reserved.