Ti3+ Tuning the Ratio of Cu+/Cu0 in the Ultrafine Cu Nanoparticles for Boosting the Hydrogenation Reaction

Article abstract of DOI:10.1002/smll.202008052

Hydrogenation of diesters to diols is a vital process for chemical industry. The inexpensive Cu⁺/Cu⁰-based catalysts are highly active for the hydrogenation of esters, however, how to efficiently tune the ratio of Cu⁺/Cu⁰ and stabilize the Cu⁺ is a great challenge. In this work, it is demonstrated that doped Ti ions can tune the ratio of Cu⁺/Cu⁰ and stabilize the Cu⁺ by the Ti-O-Cu bonds in Ti-doped SiO₂ supported Cu nanoparticle (Cu/Ti–SiO₂) catalysts for the high conversion of dimethyl adipate to 1,6-hexanediol. In the synthesis of the catalysts, the Ti⁴⁺-O-Cu²⁺ bonds promote the reduction of Cu²⁺ to Cu⁺ by forming Ti³⁺-O_V-Cu⁺ (O_V: oxygen vacancy) bonds and the amount of Ti doping can tune the ratio of Cu⁺/Cu⁰. In the catalytic reaction, the O vacancy activates C=O in the ester by forming new Ti^3+δ-O_R-Cu^1+δ bonds (O_R: reactant oxygen), and Cu⁰ activates hydrogen. After the products are desorbed, the Ti^3+δ-O_R-Cu^1+δ bonds return to the initial state of Ti³⁺-O_V-Cu⁺ bonds. The reversible Ti-O-Cu bonds greatly improve the activity and stability of the Cu/Ti–SiO₂ catalysts. When the content of Ti is controlled at 0.4?wt%, the conversion and selectivity can reach 100% and 98.8%, respectively, and remain stable for 260 h without performance degradation.

Full text of DOI:10.1002/smll.202008052

ReseaRch aRticle
www.small-journal.com
Ti3+ Tuning the Ratio of Cu+/Cu0 in the Ultrafine Cu
Nanoparticles for Boosting the Hydrogenation Reaction
Ziyang Zhang, Zhong-Li Wang,* Kang An, Jiaming Wang, Siran Zhang, Pengfei Song,
Yoshio Bando, Yusuke Yamauchi,* and Yuan Liu*
1. Introduction
Hydrogenation of diesters to diols is a vital process for chemical industry. The
Diols are high value industrial raw
material, such as ethylene glycol (EG),
of esters, however, how to efficiently tune the ratio of Cu+/Cu0 and stabilize
1,6-hexanediol (HDO). Especially, HDO
inexpensive Cu+/Cu0-based catalysts are highly active for the hydrogenation
the Cu+ is a great challenge. In this work, it is demonstrated that doped Ti
is widely used in pharmaceuticals, adhe-
+
0
+
sives, dyes, and plasticizers.[1–6] At pre-
sent, diols are produced by hydrogenating
carboxylic acids or their esters.[1] How-
ever, the hydrogenation reaction of the
carboxylic acids usually involves harsh
operating conditions (T = 523–623 K and
P(H2) = 10–20 MPa) and severe corrosion
of the reactors. In contrast, the reaction
of ester is relatively easier than acid, so
hydrogenation of dimethyl adipate (DMA)
has received researchers’ attention. To
achieve high yield of HDO from hydro-
genation of DMA, noble metallic-based
(Pt, Pd, Rh, and Rh) catalysts supported
on different supports (Al2O3, La2O3, TiO2,
SiO2, and Nb2O5) were studied in this cat-
alytic process and various Co, Sn, and Zn
metal elements are doped into the noble
metals to improve the activities of these
 
ions can tune the ratio of Cu /Cu and stabilize the Cu by the Ti O Cu
bonds in Ti-doped SiO2 supported Cu nanoparticle (Cu/Ti–SiO2) catalysts for
the high conversion of dimethyl adipate to 1,6-hexanediol. In the synthesis of
the catalysts, the Ti4+ O Cu2+ bonds promote the reduction of Cu2+ to Cu+
 
by forming Ti3+ OV Cu (OV: oxygen vacancy) bonds and the amount of Ti
+


doping can tune the ratio of Cu+/Cu0. In the catalytic reaction, the O vacancy
OR Cu1+δ bonds

3+δ
activates C O in the ester by forming new Ti

(OR: reactant oxygen), and Cu0 activates hydrogen. After the products
are desorbed, the Ti3+δ OR Cu1+δ bonds return to the initial state of


Ti3+ OV Cu bonds. The reversible Ti O Cu bonds greatly improve the
activity and stability of the Cu/Ti–SiO2 catalysts. When the content of Ti
is controlled at 0.4 wt%, the conversion and selectivity can reach 100%
and 98.8%, respectively, and remain stable for 260 h without performance
degradation.
+


 
catalysts.[1–4,7] Specifically, it demonstrated that the presence of
Sn4+ species acting as Lewis acid sites are responsible for the
carbonyl activation and the addition of cobalt could stabilize
Sn4+ species. Based on this hypothesis, Li and co-workers[8]
reported a Ru–Sn–Co/Al–O(OH) catalyst for hydrogenation of
DMA. The DMA conversion and the selectivity to HDO were as
high as 98% and 95%, respectively, under 5 MPa of H2 at 493 K
for 10 h. They also found that the synergistic effect between
cobalt and tin species could promote the reduction of ruthe-
nium oxide. However, the high cost of noble metal catalysts and
the harsh operation condition limited the further development
for DMA hydrogenation.
Z. Zhang, Prof. Z.-L. Wang, K. An, J. Wang, S. Zhang, P. Song,
Prof. Y. Liu
Tianjin Key Laboratory of Applied Catalysis Science and Technology
School of Chemical Engineering and Technology
Tianjin University
Tianjin 300072, China
E-mail: wang.zhongli@tju.edu.cn; yuanliu@tju.edu.cn
Prof. Y. Bando
Institute of Molecular Plus
Tianjin University
Tianjin 300072, China
Prof. Y. Bando, Prof. Y. Yamauchi
School of Chemical Engineering and Australian Institute for
Bioengineering and Nanotechnology (AIBN)
The University of Queensland
Brisbane, QLD 4072, Australia
E-mail: y.yamauchi@uq.edu.au
The inexpensive Cu-based catalysts have been reported
to exhibit high activities for hydrogenation of esters to alco-
hols.[9–13] The Cu sites can selectively activate C O bond


without catalyzing the hydrogenation of C C bond, thus
Prof. Y. Yamauchi
[14]

JST-ERATO Yamauchi Materials Space-Tectonics Project
International Center for Materials Nanoarchitectonics (WPI-MANA)
National Institute for Materials Science (NIMS)
1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan
avoiding the breaking of C C bond. The synergistic effect
between Cu0 and Cu+ active sites, in which Cu0 species acti-
+

vate hydrogen and Cu species activate C O, are widely applied
to enhance the catalytic performance toward hydrogenation
of esters.[15–17] Therefore, it is crucial to tune the right ratio
of Cu+/Cu0 and maintain it in the whole reaction process
for improving the activity of catalyst. Ma and co-workers[15]
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/smll.202008052.
DOI: 10.1002/smll.202008052
2008052 (1 of 10)
© 2021 Wiley-VCH GmbH
Small 2021, 2008052

www.advancedsciencenews.com
www.small-journal.com
reported that when the accessible metallic Cu surface area is
below a certain value, the catalytic activity of hydrogenation
linearly increases with increasing Cu0 surface area. While the
amount of Cu0 species is sufficient, the catalytic performance
increases with Cu+ surface area. Tuning of the surface copper
species of Cu/SiO2 and increasing the amount of Cu+ can
enable catalysts with high hydrogenation performance.[11] Ge
and co-workers[18] reported that the Cu0 can be oxidized to Cu+
by the ester and the obtained Cu+ can be reduced back to Cu0
by H2 under reaction conditions. It is a dynamic redox cycle
between Cu+ and Cu0, but the balance of copper species cannot
be maintained. In order to obtain high activity and stability of
Cu-based catalysts, doping trace metal or nonmetallic elements
such as B,[19,20] La,[21] In,[12] and Zn[9,22–24] into SiO2 support or
mixed oxide support are used to control the valence state of Cu.
Although some progress has been made, it is still a challenge
to realize the controllable adjustment and maintain the ratio
of Cu+/Cu0 and prevent the aggregation of Cu nanoparticles in
the Cu-based catalysts.
2. Results and Discussion
2.1. Synthesis of Cu/Ti(x)–SiO2 Catalysts
As shown in Figure 1, the synthesis process mainly includes
two parts: the first part is the synthesis of Ti-doped SiO2 sup-
port, and the second part is the loading of Cu catalyst. The
Ti-doped SiO2 supports are synthesized by dual-template
hydrothermal method. Polystyrene spheres (PSs) with diam-
eter of 160 nm and F127 triblock copolymers are used as tem-
plates and dispersed in the HCl solution, and then tetrabutyl
titanate (TBOT) predissolved in alcohol and tetraethyl orthosili-
cate (TEOS) are injected into above acid solution (Figure S1a,
Supporting Information). The mixture solution reacts for 96 h
at 80 °C in a sealed beaker. The solid product is obtained by
filtering and washing three times with deionized water. After
drying at 120 °C and calcining at 550 °C of the solid product,
the porous Ti-doped SiO2 support with pore size of ≈140 nm
was obtained (Figures S1b–d and S2, Supporting Information).
And then, the Ti-doped SiO2 support is added into the aqueous
solution of Cu(NO3)2·3H2O and ammonia. The derived
[Cu(NH3)4](OH)2 will react with Ti–SiO2 during the ammonia
evaporation (AE) process at 90 °C and the copper phyllosilicate
phase is formed after the further calcination in air at 450 °C, in
which the copper phyllosilicate phase is demonstrated by the
X-ray diffraction (XRD) patterns and Fourier-transform infrared
(FT-IR) spectra (Figures S3 and S4, Supporting Information),
and the calcined Cu/Ti(x)–SiO2-C (C represents calcination)
catalysts are obtained. Finally, the Cu2+ ions trapped in the
cupric silicate are gradually reduced to Cu+/Cu0 under H2 flow
at 300 °C. The reduced catalyst is denoted as Cu/Ti(x)–SiO2 (x
refers to the wt% dopant of Ti, x = 0%, 0.2%, 0.4%, and 0.6%).
In this work, we demonstrate that doped Ti ions can tune
the ratio of Cu+/Cu0 and stabilize the Cu+ by the reversible
 
Ti O Cu bonds in the Ti-doped SiO2 supported Cu nanopar-
ticles (Cu/Ti–SiO2) catalysts for the high conversion of DMA
to HDO. The dual-template hydrothermal method is applied
to synthesize the Ti–SiO2 supports with macro–mesoporous
structure, in which macropores can increase the diffusion
of reactants and products,[25–29] and mesopores are used to
anchor catalyst nanoparticles.[1,30] And then, the Cu nano-
particles are loaded on the support by the ammonia evapo-
ration method, in which copper salt reacts with Ti–SiO2 to
form copper silicate as the precursor of Cu+/Cu0. A series of
characterizations show that doped Ti ions can tune the ratio
+
0
+
 
of Cu /Cu and stabilize the Cu by the reversible Ti O Cu
bonds in Cu nanoparticles. In the synthesis of the catalysts,
the Ti4+ O Cu bonds promote the reduction of Cu2+ to
2.2. Texture Properties of Cu/Ti(x)–SiO2 Catalysts
2+
 
Cu+ by the formation of Ti3+ OV Cu (OV: oxygen vacancy)
bonds and the amount of Ti doping can tune the ratio of
Cu+/Cu0. In the catalytic reaction, the O vacancy acti-
+


The morphology and structure of the reduced catalysts
Cu/Ti(x)–SiO2 are characterized by transmission electron
microscopy (TEM). The porous structure of Cu/Ti(0.4)–SiO2
catalyst can be observed with the average pore size of 140 nm
(Figure 2a). Compared to the Ti–SiO2 support (Figures S1 and S2,
Supporting Information), the surface of Cu/Ti(0.4)–SiO2
catalyst becomes rough due to the reaction of Cu salts with
Ti–SiO2 during the AE process. In the high-magnification
TEM image (Figure 2b,c), small Cu nanoparticles with average
size of 4.6 nm are observed on the surface of support. The
3+δ
1+δ



vates C O in the ester by forming a new Ti
OR Cu
(OR: reactant oxygen) bond, and Cu0 activates hydrogen. After
the products are desorbed, the Ti3+δ OR Cu bonds return
1+δ


to the initial state of Ti3+ OV Cu bonds. The reversible
+


 
Ti O Cu bonds greatly improve the activity and stability of
the Cu/Ti–SiO2 catalysts, with conversion of 100% and selec-
tivity of 98.8% during 260 h hydrogenation for the optimized
0.4 wt% Ti sample.
Figure 1. The synthesis process of Cu/Ti(x)–SiO2 catalysts.
Small 2021, 2008052
2008052 (2 of 10)
© 2021 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
Figure 2. Morphology of reduced Cu/Ti(0.4)–SiO2 catalysts. a) The macroporous of Cu/Ti(0.4)–SiO2 catalysts, b) TEM image of Cu/Ti(0.4)–SiO2 cata-
lyst, c) particle size distribution, d) HRTEM of copper species, e) HAADF-STEM image of Cu/Ti(0.4)–SiO2 catalyst, and f–i) EDX-mapping of reduced
catalyst.
high-resolution TEM (HRTEM) image (Figure 2d) shows the
lattice fringes of the Cu metal and Cu2O nanoparticles. The
distribution of Cu and Ti species are characterized by high-
angle annular dark-field scanning transmission electron micro-
scopy (HAADF-STEM) combined with an energy dispersive
X-ray (EDX) spectroscopy (Figure 2e–i; Figure S5, Supporting
Information). The mapping images of Cu and Ti elements in
the Cu/Ti(0.4)–SiO2 catalyst overlap each other, demonstrating
that Cu and Ti species are highly distributed on the support.
Other samples without or with different Ti doping amounts are
also characterized by TEM. As shown in Figure S6 in the Sup-
porting Information, the Cu particles size of Ti-doped catalysts
(about 5 nm) is smaller and distribution is narrower, compared
to that of Ti free sample (about 6 nm). However, the specific
surface areas and pore volumes of the catalysts significantly
decrease relative to the supports (Table 1; Figure S7 and Table
S1, Supporting Information), suggesting that Cu nanoparticles
occupy mesopores, and interestingly, the sizes of mesopore and
nanoparticles are similar. So, as expected, mesopores play a role
in anchoring nanoparticles with the controllable sizes. The Cu0
dispersion and surface area of Ti-doped samples measured by
N2O titration exhibit larger values than that of Ti-free samples
(Table 1), suggesting that the Ti doping slightly increase the dis-
persion of Cu nanoparticles. The Cu and Ti contents of the Cu/
Ti(x)–SiO2 catalysts are detected by inductively coupled plasma
optical emission spectroscopy (ICP-OES) (Table 1), which are
close to the designed values.
Table 1. The texture and physicochemical properties of synthesized
Cu/Ti(x)–SiO2 catalysts.
b)
c)
c)
d)
e)
Catalysts
Cu
Ti
SBET
VP
DP SACu
Cu DCu
loadinga) loadinga) [m2 g−1
]
[cm3 g−1
]
[nm] [m2] Dispd) [nm]
[%]
[%]
[%]
0
Cu/SiO2
19.94
19.81
112
193
0.317
0.410
3.76 32.5
3.68 38.4
25.1 5.8
31.1 4.9
Cu/
0.15
Ti(0.2)–SiO2
Cu/
Ti(0.4)–SiO2
20.52
20.61
0.35
0.56
191
160
0.372
0.271
3.42 38.9
3.79 37.8
29.9 4.6
Cu/
28.3
5.1
Ti(0.6)–SiO2
a)Determined by ICP-OES analysis; b)Specific surface area measured by BET method;
c)Mesopore volume determined by BJH method; d)Cu dispersion is detected by N2O
titration; e)Copper nanoparticle size is calculated by Scherrer equation.
2008052 (3 of 10)
© 2021 Wiley-VCH GmbH
Small 2021, 2008052

www.advancedsciencenews.com
www.small-journal.com
2.3. Chemical Compositions of Cu/Ti(x)–SiO2 Catalysts
increases significantly (Figure S8, Supporting Information).
There are two reduction peaks in Ti-doped samples indicating
that part of the Cu2+ ions is preferentially reduced at low tem-
perature under the activation of Ti ions.[35] Obviously, there is
an interaction between Cu2+ and Ti4+ ions. As discussed above,
Ti ions are uniformly doped in the SiO2 matrix. Therefore, the
The chemical compositions of the reduced Cu/Ti(x)–SiO2
catalysts are characterized by the XRD patterns. As shown in
Figure 3a, the broad diffraction peaks centered at 2θ = 36.73°
and 61.50° are ascribed to the Cu2O phase (JCPDS 34-1354) or
the 6CuO·Cu2O phase (JCPDS 03-0879).[31] The characteristic
diffraction peaks at 2θ = 43.30°, 50.43°, and 74.13° are assigned
to the Cu0 (JCPDS 04-0836). In the sample of Cu/SiO2 without
Ti doping, the diffraction peaks intensity of Cu oxides is slightly
stronger than that of Cu0. In contrast, for the Ti-doped sam-
ples, the phenomenon is reverse and the peaks of Cu0 is obvi-
ously stronger than that of Cu oxides, indicating the Ti doping
enhances the reducibility of cupric silicate. Moreover, with
the increase of Ti content, the peak of Cu0 gradually becomes
sharp, indicating the enhancement of crystallinity. The absence
of TiO2 diffraction peaks in the XRD patterns, no lattice fringe
of TiO2 in the HRTEM images, and uniform Ti element dis-
tribution in the EDX mapping indicate that the Ti4+ ions are
uniformly doped in the SiO2 matrix.[32,33]
To study the reducibility of calcined samples, the H2 temper-
ature program reduction (H2-TPR) is conducted. As shown in
Figure 3b, the Ti(0.4)–SiO2 support shows no reduction peaks
and the Cu-based catalysts of temperature-programmed reduc-
tion (TPR) profiles exhibit a main and nearly symmetric reduc-
tion peak centered at 265 °C belongs to the reduction of Cu2+ to
Cu0 and/or Cu+ species.[34] Although there is no obvious differ-
ence among the Cu/SiO2-C, Cu/Ti(0.2)–SiO2-C, and Cu/Ti(0.4)–
SiO2-C, a shoulder peak centered at 228 °C can be clearly
observed in the Cu/Ti(0.6)–SiO2-C and further increasing Ti
content to 3.0 wt%, the intensity of this lower temperature peak
interaction between Cu and Ti ions can only be realized by the
2+
Ti4+ O Cu bond at the interface between Cu and Ti–SiO2
 
matrix. The oxygen atoms at the interface may be easier to lose
than those in the lattice due to the asymmetric structure, which
2+
leads to the preferential reduction of Ti4+ O Cu bonds into
 
Ti3+ OV Cu bonds. Accordingly, the Ti O Cu2+ bonds pro-
mote the reducibility of partial Cu2+ to Cu+ at low temperature
(Figure S9, Supporting Information).[35–37]
+
4+


 
The surface chemical states of the reduced Cu/Ti(x)–SiO2 cat-
alysts are further characterized by X-ray photoelectron spectra
(XPS). The high-resolution Cu 2p spectra display two peaks
at 932.8 and 952.6 eV (Figure S10a, Supporting Information),
which are assigned to Cu 2p3/2 and Cu 2p1/2, respectively.[1] In
the Ti-doped samples, the whole peaks can be assigned to the
Cu+ and/or Cu0 species, but in the Cu/SiO2 sample, except the
main peak at 932.8 eV, an additional peak at 934.8 eV can be
fitted and there is also a satellite peak at 943.3 eV, and both of
two peaks are attribute to the Cu2+ species. Obviously, the Ti
species promoted the reduction of Cu2+ species to lower valence
states under the applied reduction condition, which is con-
sistent with the H2-TPR results. Figure 3c shows the Ti 2p XPS
spectra in the Ti-doped samples. After fitting, one set of peaks
at 459.1 and 464.5 eV can be assigned to Ti3+ ions and another
set of peaks at 459.9 and 465.9 eV are assigned to Ti4+ ions.[38]
The coexistence of Ti3+ and Cu+ from the XPS of reduced
Figure 3. The characterization of calcined and reduced Cu/Ti(x)–SiO2 catalysts. a) XRD patterns of reduced catalysts, b) H2-TPR profiles of calcined
Cu/Ti(x)–SiO2-C catalysts, c) Ti 2p XPS spectra of reduced catalysts, and d) Cu LMM XAES spectra of the reduced catalysts.
Small 2021, 2008052
2008052 (4 of 10)
© 2021 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
Ti-doped samples further suggest the formation of
that when the amount of Cu+ increases to a certain extent, it
will begin to decrease, which indicates that the excessive Ti spe-
cies promote the reduction of Cu+ to Cu0.[35,37]
+
2+
Ti3+ OV Cu bonds during the reduction of Ti4+ O Cu


 
bonds.[39–41] Moreover, the ratio of Ti3+/(Ti4++Ti3+) in the three
Ti-doped samples are similar, 41.2%, 39.6%, and 40.4% for Cu/
Ti(0.2)–SiO2, Cu/Ti(0.4)–SiO2, and Cu/Ti(0.6)–SiO2, respec-
tively (Table S2, Supporting Information). Due to the similar
sizes and dispersion of Cu nanoparticles among the three
samples, the contact areas between Cu nanoparticles and the
2.4. Catalytic Performance in DMA Hydrogenation
The gas-phase DMA hydrogenation is conducted in a fixed bed
reactor to study the catalytic performance of Cu/Ti(x)–SiO2 cata-
lysts. The reaction condition is T = 210–235 °C, P(H2) = 3.0 MPa,
H2/DMA = 170, and WHSVDMA = 0.53 h−1. Figure 4a,b shows
the catalytic performance for hydrogenation of DMA as a
function of reaction temperature and each temperature main-
tained for 12 h (Figure S11, Supporting Information). It can be
seen that the conversions of DMA over Cu/Ti(0.2)–SiO2 and
Cu/Ti(0.4)–SiO2 are almost 100% in the whole temperature
range (Figure 4a). However, for the Cu/SiO2 catalyst without Ti
doping, the DMA conversion is very low, only 78% at 210 °C,
while for the Cu/Ti(0.6)–SiO2 catalyst with the high Ti content,
the conversion is also below the Cu/Ti(0.2)–SiO2 and Cu/Ti(0.4)–
SiO2. For the selectivity of HDO over Cu/Ti(x)–SiO2 catalysts
(Figure 4b), they change obviously with the Ti doping amount.
All the samples exhibit a volcanic curve and the maximum of
98.8% was achieved at 220 °C over the Cu/Ti(0.4)–SiO2 catalyst.
From Cu/SiO2 to Cu/Ti(0.6)–SiO2, the maximum of selectivity
for each sample first increase and decrease, depending on the Ti
doping amount. The HDO yields of all samples also exhibit sim-
ilar variation trend (Figure S12a, Supporting Information). We
also calculated the turnover frequency (TOF) value of 31.5 h−1
for the Cu/Ti(0.4)–SiO2 catalyst, which are higher than the
supports are similar, which leads to the similar proportion
2+
of Ti4+ O Cu bonds to be reduced. Additionally, the O 1s
 
XPS is shown in Figure S10b in the Supporting Information.
The observed OH peak at 531.4 eV in O 1s can be assigned to
oxygen vacancies induced by hydrogenation, and the variation
trend of intensity is similar to that of Ti3+ species.[42,43] The peak
of 532.5 eV attribute to the O 1s of supports.[44]
Due to the overlapping of XPS peaks of Cu+ and Cu0, it is dif-
ficult to distinguish them. So, the X-ray-induced Cu LMM Auger
electron spectra (XAES) of the reduced Cu/Ti(x)–SiO2 catalysts
are carried out (Figure 3d). The Cu LMM XAES spectra exhibit
a broad and asymmetrical peak, suggesting that Cu+ and Cu0
species coexist on the surface of catalysts. The peak is decon-
voluted into two peaks located at 916.0 and 913.2 eV, which can
be assigned to Cu0 and Cu+, respectively.[10] In addition, the Cu0
and Cu2+ coexist in Cu/SiO2 catalyst at 915.9 eV.[45] Interest-
ingly, the ratio of Cu+/(Cu0+Cu+) first increases from 36.0% to
48.8% as the Ti content raises from Cu/SiO2 to Cu/Ti(0.4)–SiO2
and then decreases to 43.4% for the Cu/Ti(0.6)–SiO2 (Table S2,
Supporting Information). According to the analysis of H2-TPR
and Ti XPS spectra, the proportion of Cu+ should increase with
the increase of Ti content, but the Cu LMM XAES results show
Figure 4. DMA hydrogenation over Cu/Ti(x)–SiO2 catalysts as a function of temperature. a) Conversion of DMA; b) selectivity of HDO; c) correlation
between surface area of Cu+ species and the STY of HDO; d) DMA hydrogenation over Cu/Ti(0.4)–SiO2 and Cu/SiO2 catalysts as a function of time
on stream. Reaction condition: T = 220 °C, P(H2) = 3.0 MPa, H2/DMA = 170, and WHSVDMA = 0.53 h−1.
2008052 (5 of 10)
© 2021 Wiley-VCH GmbH
Small 2021, 2008052

www.advancedsciencenews.com
www.small-journal.com
reported TOF value of 15.1 h−1 for CuZn catalyst.[46] The
H2/DMA ratio affects the selectivity of the product and varying
the H2/DMA ratio can produce the product with one ester
group hydrogenated (Figure S13, Supporting Information).
Moreover, except DMA, the Cu/Ti(0.4)–SiO2 can also catalyze
the dimethyl oxalate (DMO) hydrogenation to EG (Figure S14,
Supporting Information).
oxides in the Cu/SiO2 catalyst increases obviously, compared
with that of reduced catalyst. In the Cu XPS spectrum of Cu/
SiO2 catalyst (Figure S17a, Supporting Information), the char-
acteristic peak of Cu2+ disappeared after the performance test,
implying that Cu2+ species was reduced to Cu+ or Cu0.
The Ti XPS and Cu LMM XAES spectra of Cu/Ti(x)–SiO2
catalysts after the performance tests are fitted and analyzed
quantitatively (Figure S16c,d and Table S4, Supporting Informa-
tion). An interesting discovery is that the contents of Ti3+ and
Cu+ almost remain unchanged in the Cu/Ti(0.4)–SiO2 catalyst,
while the changes are obvious in Cu/SiO2 catalyst without Ti
Generally, Cu+ ions are considered to be the primary sites for

activation of C O bonds by polarizing the electron lone pair
in oxygen atoms and Cu0 acts as adsorption sites of activating
H2.[11] The ratio of Cu+/(Cu0+Cu+) and the synergistic effect
between Cu0 and Cu+ play a key role in the ester hydrogenation
process.[14] Assuming that the Cu+ ions and Cu0 atoms occupy
identical area, and have identical atomic sensitivity factors, the
Cu+ surface areas can be estimated according to the Cu0 surface
areas from the N2O titration and the ratios of Cu+/(Cu++Cu0)
from the Cu LMM XAES.[1] These values are calculated and
are given in Table S2 in the Supporting Information. Interest-
ingly, the space time yield (STY) of HDO is linearly related
to both surface areas of Cu+ and Cu0 (Figure 4c; Figure S12b,
Supporting Information),[16] indicating that Cu+ and Cu0 are
almost equally important in the DMA hydrogenation reaction.
Coincidentally, the ratio of Cu+/Cu0 in the best catalyst of Cu/
Ti(0.4)–SiO2 is close to 1:1. Therefore, appropriate Ti doping
amount can optimize the ratio of Cu+/Cu0, thus improve the
performance of the catalyst. In addition to the high activity, the
Cu/Ti(0.4)–SiO2 catalyst also shows excellent stability, as shown
in Figure 4d. The DMA conversion and HDO selectivity of Cu/
Ti(0.4)–SiO2 catalyst have no obvious decline during hydro-
genation for 260 h. In sharp contrast, for the Cu/SiO2 catalyst
without Ti doping, the DMA conversion drops to 84% and the
HDO selectivity drops to 76% after 260 h. This demonstrates
that the appropriate Ti species also enhance the stability of the
catalyst by the strong interaction between Cu and Ti species.
or Cu/Ti(0.6)–SiO2 catalyst with high Ti content. For the Cu/
2+
SiO2 catalyst, due to the weak reducibility of Si4+ O Cu
 
bonds, the reduction process is long (Cu2+ to Cu+ to Cu0), so
the composition has been changing. For the Cu/Ti(0.6)–SiO2
2+
catalyst, with high Ti content, too many Ti4+ O Cu bonds
 
lead to a large amount of Cu+, which promotes the transition
to Cu0, leading to the strong Cu0 diffraction peak (Figure S16b,
Supporting Information). However, for the Cu/Ti(0.4)–SiO2
catalyst, the content of Ti3+ and Cu+ can be maintained, which
means that the formed Ti3+ OV Cu bonds are stable in the
reaction process.[47] The O 1s XPS spectra were also carried out
as shown in Figure S17b in the Supporting Information. It is
well-known that the oxygen vacancy is highly active sites for
catalytic reactions.
+


To explore more detailed structural information, the cata-
lysts are further investigated by the X-ray absorption near-edge
structure (XANES) and extended X-ray absorption fine struc-
ture (EXAFS) spectroscopies of the Cu K-edge and Ti K-edge.
As shown in the XANES spectra of the Cu K-edge (Figure 5a),
the absorption edges of the Cu/Ti(0.4)–SiO2 and Cu/SiO2
located between those of pristine Cu foil and Cu2O, suggesting
the averaged valence state of both Cu species is between 0 and
+1, indicating the coexistence of Cu0 and Cu+. But the curve of
Cu/Ti(0.4)–SiO2 slightly shifts to higher energy compared
to that of Cu/SiO2, indicating the higher Cu+ content in Cu/
Ti(0.4)–SiO2. However, the absorption edge of Cu/Ti(0.6)–SiO2
with high Ti content shift back to lower energy, close to that
of Cu/SiO2, indicating that excessive Ti species enhance the
reduction of Cu+ to Cu0 (Figure S18a, Supporting Information).
The bonding environments of Cu atoms are analyzed by the
EXAFS spectra (Figure 5b). The peaks at ≈1.55 Å (uncorrected)
are assigned to the Cu–O path, and the peaks at ≈2.25 Å (uncor-
rected) are attributed to the Cu–Cu path. The fitted data are
shown in Table S3 in the Supporting Information, and interest-
ingly, the Cu–O coordination numbers of Cu/Ti(0.4)–SiO2 and
Cu/SiO2 are ≈1.2 and 0.46, respectively. Obviously, compared
to the Cu/SiO2, the existence of Ti in Cu/Ti(0.4)–SiO2 induces
2.5. Chemical Properties of Cu/Ti(x)–SiO2 Catalysts after
Performance Test
To further gain insight into the interaction between Cu and Ti
species, the structure and composition of the catalysts are char-
acterized after the performance tests. According to the TEM
images, the Cu nanoparticles in the Cu/Ti(0.4)–SiO2 catalyst
have no obvious change compared with the reduced catalyst
(Figure S15, Supporting Information), and the HRTEM image
of Cu/Ti(0.4)–SiO2 also shows the lattice fringes of Cu2O(111)
and Cu(111) as shown in Figure S16a in the Supporting Infor-
mation. But for the Cu/SiO2 catalyst, the average sizes of Cu
particles increase from 5.8 to 6.9 nm (Figure S15a, Supporting
Information) and for the Cu/Ti(0.6)–SiO2 catalyst, the Cu par-
ticle sizes also increase slightly from 5.1 to 5.9 nm and the size
distribution becomes wide (Figure S15d, Supporting Informa-
tion). The chemical component is further detected by XRD pat-
terns. Figure S16b in the Supporting Information shows that
diffraction peaks at 36.42° and 43.30 ° are ascribed to Cu2O and
Cu0 species, respectively. The copper species of Cu/Ti(0.4)–
SiO2 catalyst show almost no change during the reaction pro-
cess, which indicates the Ti species stabilize Cu+ species in
Cu/Ti(0.4)–SiO2, while the diffraction peak intensity of Cu
 
the formation one-coordinated Cu O Ti bonds. From the Ti
K-edge absorption curves (Figure S18b, Supporting Informa-
tion), the averaged valence state of Ti species is between 0 and
+4 and close to +4, suggesting the possible existence of Ti3+ spe-
 
cies due to the formation of Cu O Ti bonds. Therefore, the
results of XAFS are consistent with other experimental data
 
such as XAES and XPS, and further demonstrate the Cu O Ti
bonds.
The strong interaction between Ti3+ and Cu+ is further veri-
fied through the different characteristics of Cu/SiO2 and Cu/
Ti(0.4)–SiO2 catalysts after the stability test. The performance
Small 2021, 2008052
2008052 (6 of 10)
© 2021 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
Figure 5. a) XANES spectra and b) Fourier transformed EXAFS spectra of the Cu K-edge for the Cu/Ti(0.4)–SiO2 and Cu/SiO2 catalysts after perfor-
mance test.

test lasted only 60 h, while the stability test lasted 260 h. As
shown in Figure S19 in the Supporting Information, the severe
aggregation of Cu nanoparticles is observed in the Cu/SiO2 cat-
alyst. The particle size increases from 5.8 nm of reduced catalyst
to 9.6 nm of the catalyst after stability test. By contrast, the Cu
nanoparticles in the Cu/Ti(0.4)–SiO2 catalyst only increased
from 4.6 to 6.1 nm. The almost disappeared diffraction peak of
Cu2O species at 36.42° and the sharp peak of Cu0 species at
43.30° in the XRD pattern of Cu/SiO2 imply the Cu+ species
are reduced to Cu0 species, as well as confirming the sintering
of Cu nanoparticles (Figure S20a, Supporting Information). For
the Cu/Ti(0.4)–SiO2 catalyst, although the two diffraction peaks
of Cu2O and Cu0 species are stronger than the reduced sample,
the intensity of the peaks still keeps a certain proportion, sim-
ilar to that of Ti3+/Ti4+ species (Figure S20b,S21a,b, Supporting
Information). The Cu LMM XAES spectra (Figure S21c, Sup-
porting Information) show that the reduction of Cu+ to Cu0 spe-
cies results to the decreased Cu+/(Cu0+Cu+) ratio by about 10%
in Cu/SiO2 during the stability test process, compared to that
of after performance test. Whereas, the Cu+/(Cu0+Cu+) ratio
of Cu/Ti(0.4)–SiO2 merely decreased by 2% (Figure S21c and
Table S4, Supporting Information). Additionally, the O 1s XPS
of Cu/Ti(0.4)–SiO2 catalyst also shows no significant change
after stability test (Figure S21d, Supporting Information). These
results confirm that Ti3+ stabilize Cu+ and prevent the aggrega-
of C O group by polarizing the lone pair electron on oxygen
1+δ
atoms. The Ti3+δ OR Cu bonds (site B) are formed under


reaction conditions. IV) The active H that spilled over from Cu0

directly attack the nearby C O bond, followed by formation of
CH3OH and acyl intermediate that absorb on Ti3+ OV Cu
+


bond. V) The acyl intermediate reacts with other H atoms
and complete the hydrogenation of C O bond. VI) The C O
group is reduced completely and the products are desorbed, as


well as, the Ti3+ OV Cu bond is exposed again and the next
circle starts again. There are two main reaction paths for diester
hydrogenation to synthesize alcohol.[48,49] One is that methoxy
group is preferentially hydrogenated to produce methanol, then
carbonyl group, in which C and O are hydrogenated in turn;
the other is that carbonyl group is hydrogenated first, and
then the methoxy group. Density functional theory (DFT) cal-
culations show that the activation energy barriers (ΔG⧧) of the
first and second ester groups of the former pathway are 19.2
and 24.7 kcal mol−1, respectively, which are lower than those
of the latter pathway (ΔG⧧ = 19.9 and 26.2 kcal mol−1, respec-
tively), indicating that the former reaction path is more likely
to occur.[48] Interestingly, the reaction mechanism based on our
experimental results is consistent with the former one. Methoxy
+


group is hydrogenated first and then carbonyl group. Moreover,
the formation of Ti3+δ OR Cu bond activates methoxy and
carbonyl groups at the same time and accelerates the hydro-
genation reaction.
1+δ


 
tion of Cu nanoparticles by the Ti O Cu bonds.
2.6. Discussion of the Reaction Mechanism
3. Conclusion
Based on above experimental results, we can reasonably specu-
late the reaction process of DMA hydrogenation to HDO over
Cu/Ti(0.4)–SiO2 catalyst and provide the most likely reaction
model. As shown in Figure 6: I) The precursor of the cata-
lyst is reduced under H2 flow at 300 °C for 3 h. The Cu2+ on
the surface of Cu nanoparticles are reduced to Cu0, and the
In this work, a series of Ti-doped SiO2 supported Cu (Cu/
Ti–SiO2) catalysts with different Ti doping amount have been
synthesized by dual-templating hydrothermal and ammonia
evaporation method for the high conversion of DMA to HDO.
Doped Ti ions can tune the ratio of Cu+/Cu0 and stabilize
+
 
the Cu by the reversible Ti O Cu bonds. A series of char-
2+
+
Ti4+ O Cu bonds are reduced to Ti3+ OV Cu bonds (site
A) by H atoms that spill over from Cu0. II) H2 and DMA gas
flow is introduced to the reaction system. H2 is activated by Cu0
acterizations show that in the synthesis of the catalysts, the
 


Ti4+ O Cu bonds promote the reduction of Cu2+ to Cu+
2+
 
by the formation of Ti3+ OV Cu bonds and the amount of
+


species, meanwhile, the C O group in DMA is absorbed by
Ti doping can tune the ratio of Cu+/Cu0. In the catalytic reac-

+


Cu species, and the oxygen in C O group occupy the oxygen
tion, the O vacancy activates C O in the ester by forming new
vacancy between Cu+ and Ti3+ species. III) The synergistic
effect between Cu+ and Ti3+ species enhance the activation
Ti3+δ OR Cu
bonds, and Cu0 activates hydrogen. After
1+δ


the products are desorbed, the Ti3+δ OR Cu bonds return
1+δ


2008052 (7 of 10)
© 2021 Wiley-VCH GmbH
Small 2021, 2008052

www.advancedsciencenews.com
www.small-journal.com
Figure 6. The mechanism of DMA hydrogenation to HDO over Cu/Ti(x)–SiO2 catalysts.
+
to the initial state of Ti3+ OV Cu bonds, and then begin to
bond new reaction molecules to achieve continuous reversible
cycle. When the content of Ti is controlled at 0.4 wt%, the ratio
of Cu+/Cu0 is close to 1:1, and the performance reached the
highest value, with conversion of 100% and selectivity of 98.8%,
running stably for 260 h without performance degradation. The
F127 (3 g) was first dissolved in alcohol (25 mL) and then the solution was
injected in HCl solution (75 mL, 2 m) under stirring at 35 °C. Then, PS
(4 g) was added to the above solution and the solution was stirred for 1 h
to uniformly disperse PS. And then, TEOS (7 mL) was introduced into the
solution. The calculated TBOT was dissolved in alcohol (15 mL), and the
resulting solution is added to the above reaction solution drop by drop and
stirred at 35 °C for 24 h. The obtained mixture solution was reacted at 80 °C
for 96 h in a sealed beaker under static conditions. The solid products after
reaction were washed for three times with deionized water, dried at 120 °C
for 12 h, and calcined in air at 350 °C for 1 h and 550 °C for 6 h with heating
rate of 0.5 °C min−1. By varying the amount of TBOT, a series of macro–
mesoporous supports (Ti(x)–SiO2) were obtained with different Ti contents
(x represents the Ti content in Cu/Ti(x)–SiO2: 0, 0.2, 0.4, and 0.6 wt%).
Catalyst Preparation: Cu/Ti(x)–SiO2 catalysts were prepared by using
ammonia evaporation method as reported in ref. [51]. The detailed process
is described as follows. Cu (NO3)2·3H2O (1.236 g) was dissolved in
deionized water (25 mL). And then, ammonia aqueous solution (10 mL,
25 wt%) was added to adjust the pH of the solution to 11. After the
solution was stirred for 30 min, Ti(x)–SiO2 supports (1.3 g) were added,
maintaining the pH value of 11 by adding ammonia aqueous (about 2 mL).
The obtained suspension was stirred under room temperature for 4 h and
then ammonia evaporation process was carried out at 90 °C in air under
ambient pressure. When the pH value of suspension decreased to 6–7,
the evaporation process was terminated. The solid products were filtered,
washed with deionized water three times, dried at 120 °C overnight, and
calcined in air at 450 °C for 4 h. The obtained products were the precursor
of the catalyst Cu/Ti(x)–SiO2-C. After H2 reduction at 300 °C for 3 h, the
final products of Cu/Ti(x)–SiO2 catalysts were obtained.


 
reversible Ti O Cu bonds greatly improve the activity and sta-
bility of the Cu/Ti(x)–SiO2 catalysts. This work provides a new
strategy to highly active and inexpensive Cu-based catalysts for
the hydrogenation reactions.
4. Experimental Section
Synthesis Protocol of Polystyrene Spheres: PS with a diameter of 160 nm
was prepared by the surfactant-assisted emulsion anionic polymerization
technique.[50] Styrene (St) and divinylbenzene (DVB) were first washed
three times with NaOH solution (0.1 m) and then washed three times
with deionized water to remove polymerization inhibitors. In the detailed
synthesis, deionized water (240 mL) was added to 500 mL three neck
bottom flask, and nitrogen flow was injected to flush out the dissolved
oxygen at least 30 min. And then, St (25 mL) and DVB (4.75 mL) were
introduced in three neck bottom, and sodium dodecyl sulfate (SDS,
0.3 g) was added the above solution under the continuous nitrogen
bubbling and vigorous stirring. After heating up to 80 °C, 0.187 g of
potassium persulfate (KPS) predissolved in deionized water (20 mL)
was injected to the reaction system and the reaction lasted 5 h. After
centrifugation, washing with deionized water and drying at 80 °C for 12 h,
the PS with the suitable size was obtained.
Measurement of Catalytic Activity and Stability: The activity test was
carried out in a continuous flow unit equipped with a stainless-steel
fixed-bed tubular reactor. For the activity test, the catalyst precursor
was in situ reduced in pure H2 flow (flow rate of 30 mL min−1) at
300 °C under ambient pressure for 3 h and then was directly applied
Synthesis Protocol of Hierarchical Porous Ti(x)–SiO2 Support: Macro–
mesoporous Ti–SiO2 support was synthesized according to the literature.[50]
Small 2021, 2008052
2008052 (8 of 10)
© 2021 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
to the catalytic reaction. Catalyst precursor (0.4 g, 40–60 meshes) was
filled into tubular reactor with inner diameter of 8 mm. After reduction
where NA is Avogadro’s constant, MCu is the relative atomic weight, and
W
Cu is the content of copper (wt%).
The TOF calculation based on the following Equation (3)
process, the reactant (1 mol L−1 DMA in methanol) was injected by
−1
||
HPLC pump (elite HPLC P230 ) with a flow rate of 0.02 mL min . The
reaction was conducted at a system pressure of 3.0 MPa, temperature
range of 210–235 °C, and a H2/DMA molar ratio of 170. The liquid
products were collected in cold trap with ice-water bath and analyzed by
gas chromatograph (Agilent 7820 GC) with a DB-WAX capillary column
(30 m × 0.32 mm × 0.25 µm) and a TCD detector.
nDMA ×CDMA
TOF =
(h−1
)
(3)
WCu
MCu
DispCu × mCatal
×
where nDMA is the number of DMA molecules fed per hour; CDMA is the
conversion of DMA; DispCu is the dispersion of copper species; mCatal is
the dosage of catalyst (g); WCu is the content of copper (wt%); MCu is
the relative atomic weight.
Catalyst Characterization: N2 physisorption analysis was carried out at
−196 °C using Tristar 3000 Micromeritics instrument. The surface area
of catalysts was calculated from Barrett–Joyner–Halenda (BJH) method
cumulative desorption surface area. The size distribution of mesopores
was calculated by BJH method based on desorption isotherms and the
pore volume obtained from BJH method cumulative desorption pore
volume. ICP-OES was conducted to obtain the content of Cu (WCu) and
Ti (WTi) with Perkin Elmer Optima 5300DV instrument. The solid catalyst
was dissolved in the HNO3 and HF mixed acid by HBO3 neutralizing.
TPR was carried out on TP-5070 Multifunction automatic adsorber.
50 mg catalyst precursor was loaded into a quartz tube and reduced in
5% H2/Ar (flow rate of 30 mL min−1) at a temperature range of 20 to
900 °C with a heating rate of 5 °C min−1. The hydrogen consumption
was detected by a thermal conductivity detector (TCD). X-ray diffraction
was conducted on a Bruker D8 Focus X-ray diffractometer using Cu Kα
radiation (λ = 0.15 418 nm). The tube voltage and current were 40 kV
and 40 mA, respectively. The diffraction angle of samples was recorded
from 20° to 80° (2θ) with scan rate of 8° min−1 (The calcined samples
were recorded from 20° to 80° (2θ) with a scan rate of 2° min−1). The
FT-IR measurements were conducted on a Thermo Nicolet Nexus 470
FT-IR with a spectral resolution of 4 cm−1 from 400 to 4000 cm−1. The
standard KBr method was used to press sample wafers. TEM images
of samples were obtained by JEOL F200 system electron microscope
at 200 kV. The samples were dispersed in ethanol and ultrasonic
treatment for 20 min, and then dropped the samples onto the Mo
grid ultrathin carbon film. XPS and XAES were conducted on the
Thermo ESCALAB 250Xi instrument. X-ray radiation source was Al Ka
(hν = 1486.6 eV) with X-ray power of 150 W and spot size was 500 µm.
The pass energy was 30 eV. The XPS was calibrated by C 1s binding
energy of 284.8 eV. X-ray absorption spectroscopy (XAS) measurements
were performed on the copper K-edges (Transsmition) and Ti K-edges
(Lytle) at the hard X-ray Beamline BL11B at Shanghai Synchrotron
Radiation Facility (SSRF).
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
This work was supported by the National Natural Science Foundation of
China (NSFC) (grant numbers 22075201, 21962014, 21872101). This work
was performed in part at the Queensland node of the Australian National
Fabrication Facility Queensland Node (ANFF-Q), a company established
under the National Collaborative Research Infrastructure Strategy to
provide nano- and micro-fabrication facilities for Australia’s researchers.
The XAS measurement was performed at the SSRF beamline (BL11B).
Conflict of Interest
The authors declare no conflict of interest.
Data Availability Statement
Research data are not shared.
The number of surface metallic copper was measured by N2O
titration method using TP-5080 Multifunction automatic adsorber with
TCD by N2O chemisorption and H2 reduction based on the reaction of
2Cu(s) + N2O → Cu2O(s) + N2.[1] 50 mg calcined catalyst was flushed with
He flow at 200 °C for 1 h to remove the air. Catalyst was first reduced in 5%
H2/He at 300 °C for 2 h (Cu2+ to Cu0). The amount of hydrogen
consumption in first reduction was detected by TCD detector and
denoted as A1, which represents the total amount of copper species in
the catalyst. And then, the catalyst bed was purged with He and cooled to
90 °C. Surface copper atoms were oxidized in 10% N2O/N2 (20 mL min−1)
at 90 °C for 1.5 h, ensuring the surface metallic copper was oxidized to
Cu2O. Finally, the sample was flushed with He flow to blow off the oxidant
and cooled to room temperature to start another reduction process. The
second reduction process was performed at 300 °C in 5% H2/He flow, and
the H2 consumption was detected by TCD and denoted as A2 that reflects
the number of surface metallic copper. The dispersion of metallic copper
(DispCu) can be calculated with Equation (1)
Keywords
+
0


Cu bonds,
Cu /Cu tuning, hydrogenation, reversible conversion, Ti
ultrafine Cu nanoparticles
O
Received: December 22, 2020
Revised: March 7, 2021
Published online:
[1] Y. Zhao, Z. Guo, H. Zhang, B. Peng, Y. Xu, Y. Wang, J. Zhang, Y. Xu,
S. Wang, X. Ma, J. Catal. 2018, 357, 223.
[2] F. C. A. Figueiredo, E. Jordão, W. A. Carvalho, Catal. Today 2005,
107–108, 223.
[3] J. Fontana, C. Vignado, E. Jordao, F. C. A. Figueiredo, W. A. Carvalho,
Catal. Today 2011, 172, 27.
[4] J. Fontana, C. Vignado, E. Jordão, W. A. Carvalho, Chem. Eng. J.
2010, 165, 336.
[5] S. M. Dos Santos, A. M. Silva, E. Jordão, M. A. Fraga, Catal. Today
2005, 107–108, 250.
2A2
A1
(1)
DispCu
=
×100%
The metallic copper surface area (SCu) was calculated by Equation (2)
based on an atomic copper surface density of 1.46 × 1019 Cu atoms m−2
[6] A. M. Silva, M. A. Morales, E. M. Baggio-Saitovitch, E. Jordão,
M. A. Fraga, Appl. Catal., A 2009, 353, 101.
2A2NAWCu
1.46 ×1019 A1MCu
m2 g−1
(2)
SCu
=
(
)
2008052 (9 of 10)
© 2021 Wiley-VCH GmbH
Small 2021, 2008052

www.advancedsciencenews.com
www.small-journal.com
[7] S. Santos, A. Silva, E. Jordao, M. Fraga, Catal. Commun. 2004, 5,
377.
[8] H.-B. Jiang, H.-J. Jiang, K. Su, D.-M. Zhu, X.-L. Zheng, H.-Y. Fu,
H. Chen, R.-X. Li, Appl. Catal., A 2012, 447–448, 164.
[9] Y. Zhao, B. Shan, Y. Wang, J. Zhou, S. Wang, X. Ma, Ind. Eng. Chem.
Res. 2018, 57, 4526.
[32] Q. Wu, C. Zhang, B. Zhang, X. Li, Z. Ying, T. Liu, W. Lin, Y. Yu,
H. Cheng, F. Zhao, J. Colloid Interface Sci. 2016, 463, 75.
[33] W. Dong, Y. Sun, C. W. Lee, W. Hua, X. Lu, Y. Shi, S. Zhang, J. Chen,
D. Zhao, J. Am. Chem. Soc. 2007, 129, 13894.
[34] T. Song, W. Chen, Y. Qi, J. Lu, P. Wu, X. Li, Catal. Sci. Technol. 2020,
10, 5149.
[10] T. Popa, Y. Zhang, E. Jin, M. Fan, Appl. Catal., A 2015, 505, 52.
[11] Y. Zhao, S. Li, Y. Wang, B. Shan, J. Zhang, S. Wang, X. Ma, Chem.
Eng. J. 2017, 313, 759.
[12] Y. Zhang, C. Ye, C. Guo, C. Gan, X. Tong, Chin. J. Catal. 2018, 39, 99.
[13] W. Wang, H. Wang, J. Zhang, L. Kong, H. Huang, W. Liu, S. Wang,
X. Ma, Y. Zhao, ChemCatChem 2020, 12, 3849.
[35] C. Wen, A. Yin, Y. Cui, X. Yang, W.-L. Dai, K. Fan, Appl. Catal., A
2013, 458, 82.
[36] Y. Bao, C. Huang, L. Chen, Y. Zhang, L. Liang, J. Wen, M. Fu, J. Wu,
D. Ye, J. Energy Chem. 2018, 27, 381.
[37] Z. Wang, D. Brouri, S. Casale, L. Delannoy, C. Louis, J. Catal. 2016,
340, 95.
[14] Y. Yao, X. Wu, O. Y. Gutiérrez, J. Ji, P. Jin, S. Wang, Y. Xu, Y. Zhao,
S. Wang, X. Ma, J. A. Lercher, Appl. Catal., B 2020, 267, 118698.
[15] Y. Wang, Y. Shen, Y. Zhao, J. Lv, S. Wang, X. Ma, ACS Catal. 2015, 5,
6200.
[16] J. Gong, H. Yue, Y. Zhao, S. Zhao, L. Zhao, J. Lv, S. Wang, X. Ma, J.
Am. Chem. Soc. 2012, 134, 13922.
[17] R.-P. Ye, L. Lin, J.-X. Yang, M.-L. Sun, F. Li, B. Li, Y.-G. Yao, J. Catal.
2017, 350, 122.
[18] X. Ma, Z. Yang, X. Liu, X. Tan, Q. Ge, RSC Adv. 2015, 5, 37581.
[19] Z. He, H. Lin, P. He, Y. Yuan, J. Catal. 2011, 277, 54.
[20] S. Zhao, H. Yue, Y. Zhao, B. Wang, Y. Geng, J. Lv, S. Wang, J. Gong,
X. Ma, J. Catal. 2013, 297, 142.
[21] X. Zheng, H. Lin, J. Zheng, X. Duan, Y. Yuan, ACS Catal. 2013, 3,
2738.
[22] Q. Wang, L. Qiu, D. Ding, Y. Chen, C. Shi, P. Cui, Y. Wang,
Q. Zhang, R. Liu, H. Shen, Catal. Commun. 2018, 108, 68.
[23] B. Zhang, Y. Chen, J. Li, E. Pippel, H. Yang, Z. Gao, Y. Qin, ACS
Catal. 2015, 5, 5567.
[24] Y. Zhu, L. Shi, J. Ind. Eng. Chem. 2014, 20, 2341.
[25] Y. Ai, L. Liu, C. Zhang, L. Qi, M. He, Z. Liang, H.-B. Sun, G. Luo,
Q. Liang, ACS Appl. Mater. Interfaces 2018, 10, 32180.
[26] D. Yao, Y. Wang, K. Hassan-Legault, A. Li, Y. Zhao, J. Lv, S. Huang,
X. Ma, ACS Catal. 2019, 9, 2969.
[27] R. Kanwar, R. Bhar, S. K. Mehta, ChemCatChem 2018, 10, 2087.
[28] Y. Ai, Z. Hu, L. Liu, J. Zhou, Y. Long, J. Li, M. Ding, H.-B. Sun,
Q. Liang, Adv. Sci. 2019, 6, 1802132.
[38] P. Zhang, Y. Sun, M. Lu, J. Zhu, M. Li, Y. Shan, J. Shen, C. Song,
Energy Fuels 2019, 33, 7696.
[39] S. Zhu, X. Chen, Z. Li, X. Ye, Y. Liu, Y. Chen, L. Yang, M. Chen,
D. Zhang, G. Li, H. Li, Appl. Catal., B 2020, 264, 118515.
[40] C. Liu, S. L. Nauert, M. A. Alsina, D. Wang, A. Grant, K. He,
E. Weitz, M. Nolan, K. A. Gray, J. M. Notestein, Appl. Catal., B 2019,
255, 117754.
[41] L. Yan, F. Yang, C. Y. Tao, X. Luo, L. Zhang, Ceram. Int. 2020, 46,
9455.
[42] T. Billo, F.-Y. Fu, P. Raghunath, I. Shown, W.-F. Chen, H.-T. Lien,
T.-H. Shen, J.-F. Lee, T.-S. Chan, K.-Y. Huang, C.-I. Wu, M. C. Lin,
J.-S. Hwang, C.-H. Lee, L.-C. Chen, K.-H. Chen, Small 2018, 14,
1702928.
[43] Z. Dong, D. Ding, T. Li, C. Ning, Appl. Surf. Sci. 2019, 480, 219.
[44] A. Carnera, P. Mazzoldi, A. Boscolo-Boscoletto, F. Caccavale,
R. Bertoncello, G. Granozzi, I. Spagnol, G. Battaglin, J. Non-Cryst.
Solids 1990, 125, 293.
[45] I. Platzman, R. Brener, H. Haick, R. Tannenbaum, J. Phys. Chem. C
2008, 112, 1101.
[46] V. Pospelova, J. Aubrecht, O. Kikhtyanin, K. Pacultova, D. Kubicka,
ChemCatChem 2019, 11, 2169.
[47] M. Xu, S. Yao, D. Rao, Y. Niu, N. Liu, M. Peng, P. Zhai, Y. Man,
L. Zheng, B. Wang, B. Zhang, D. Ma, M. Wei, J. Am. Chem. Soc.
2018, 140, 11241.
[48] J. Sun, J. Yu, Q. Ma, F. Meng, X. Wei, Y. Sun, N. Tsubaki, Sci. Adv.
2018, 4, eaau3275.
[29] D. Yao, Y. Wang, Y. Li, Y. Zhao, J. Lv, X. Ma, ACS Catal. 2018, 8,
1218.
[49] C. Xu, G. Chen, Y. Zhao, P. Liu, X. Duan, L. Gu, G. Fu, Y. Yuan,
N. Zheng, Nat. Commun. 2018, 9, 3367.
[30] Y. Ai, M. He, F. Zhang, Y. Long, Y. Li, Q. Han, M. Ding, H.-B. Sun,
Q. Liang, J. Mater. Chem. A 2018, 6, 16680.
[50] A. Zaki, J. Xu, G. Stoclet, S. Casale, J.-P. Dacquin, P. Granger,
Microporous Mesoporous Mater. 2015, 208, 140.
[31] J. Medina-Valtierra, J. Ramírez-Ortiz, V. M. Arroyo-Rojas, P. Bosch,
J. A. De Los Reyes, Thin Solid Films 2002, 405, 23.
[51] 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.
Small 2021, 2008052
2008052 (10 of 10)
© 2021 Wiley-VCH GmbH