Insight into the role of hydroxyl groups on the ZnCr catalyst for isobutanol synthesis from syngas

Article abstract of DOI:10.1016/j.apcata.2017.08.022

A series of ZnO, ZnCr catalysts were prepared by a sol-gel method and ammonia solution was used to adjust the pH of the sol solution. The catalysts were characterized by XRD, in-situ FR-IR, NH₃-TPD, in-situ XPS, HRTEM. The results show that lower calcination temperature is helpful to reduce the crystal size and crystallinity of ZnCr nanocrystal, as well as forming a certain amount of structure defects and hydroxyl groups. The hydroxyl groups could be consumed by CO under the interaction between ZnO and ZnCr spinel, resulting in more exposed oxygen vacancies. We find the proper calcination temperature and Zn/Cr molar ratio for the ZnCr catalysts preparation are 400 °C and 1.0, respectively. The Zn₁Cr₁–400 ～ 2.0 catalyst prepared with the pH value of 2 shows more hydroxyl groups and small particle size, exhibiting the best catalytic performance both on the CO conversion (20.9%) and isobutanol selectivity (24.2 wt%).

Full text of DOI:10.1016/j.apcata.2017.08.022

Accepted Manuscript
Title: Insight into the role of hydroxyl groups on the ZnCr
catalyst for isobutanol synthesis from syngas
Authors: Xiaofeng Gao, Yingquan Wu, Guohui Yang, Tao
Zhang, Xiaoli Li, Hongjuan Xie, Junxuan Pan, Yisheng Tan
PII:
S0926-860X(17)30402-7
DOI:
Reference:
http://dx.doi.org/10.1016/j.apcata.2017.08.022
APCATA 16377
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
31-5-2017
11-8-2017
15-8-2017
Please cite this article as: Xiaofeng Gao, Yingquan Wu, Guohui Yang, Tao Zhang,
Xiaoli Li, Hongjuan Xie, Junxuan Pan, Yisheng Tan, Insight into the role of hydroxyl
groups on the ZnCr catalyst for isobutanol synthesis from syngas, Applied Catalysis A,
Generalhttp://dx.doi.org/10.1016/j.apcata.2017.08.022
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Insight into the role of hydroxyl groups on the ZnCr catalyst for isobutanol
synthesis from syngas
Xiaofeng Gao a,b, Yingquan Wu a, Guohui Yang a, Tao Zhang a, Xiaoli Li a,b, Hongjuan Xie a, Junxuan Pan a and
Yisheng Tan a,c*
a State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences,
Taiyuan 030001, China. Email: tan@sxicc.ac.cn. Fax: +86 351 4044287.Tel: +86 351 4044287.
b University of Chinese Academy of Sciences, Beijing 100049, China
c
National Engineering Research Center for Coal-Based Synthesis, Institute of Coal Chemistry, Chinese
Academy of Sciences, Taiyuan 030001, China
1

Graphical Abstract
The hydroxyl groups could be consumed by
CO under the interaction between ZnO and
ZnCr spinel, resulting in more exposed
oxygen vacancies for C-O activation.
Highlights
1. ZnCr catalysts are prepared by sol-gel method to explore role of hydroxyl groups.
2. Lower calcination temperature causes more structure defects and hydroxyl groups.
3. The ZnCr prepared at pH=2 exhibits high total alcohol and isobutanol selectivity.
4. The hydroxyl groups could be consumed by CO, resulting in more oxygen vacancies.
Abstract
A series of ZnO, ZnCr catalysts were prepared by a sol-gel method and ammonia solution.
The catalysts were characterized by XRD, in-situ FR-IR, NH3-TPD, in-situ XPS, HRTEM. The
2

results show that lower calcination temperature is helpful to reduce the crystal size and
crystallinity of ZnCr nanocrystal, as well as forming a certain amount of structure defects and
hydroxyl groups. The hydroxyl groups could be consumed by CO under the interaction between
ZnO and ZnCr spinel, resulting in more exposed oxygen vacancies. We find the proper
o
calcination temperature and Zn/Cr molar ratio for the ZnCr catalysts preparation are 400 C and
1.0, respectively. The Zn1Cr1-400~2.0 catalyst prepared with the pH value of 2 shows more
hydroxyl groups and small particle size, exhibiting the best catalytic performance both on the CO
conversion (20.9%) and isobutanol selectivity (24.2 wt%).
Keywords: ZnCr2O4, ZnO, hydroxyl groups, interaction, oxygen vacancies
1. Introduction.
Syngas (CO + H2), one of the most important industrial chemical products, is widely used as raw
material for other chemicals synthesis or directly as fuel [1-4]. Syngas conversion to other
chemicals, Fischer-Tropsch synthesis, methanol synthesis, etc., has attracted extensive attentions
with the development of modern catalysis science. Among lots of processes for syngas
conversion, the higher alcohols synthesis has been widely investigated due to its broader
prospects in our society and huge market demand as fuel blends, additive agent or in
intermediates for fine chemicals synthesis [5-9]. Isobutanol is one important alcohol product. It
could be used directly as an additive to improve the quality of gasoline, on the other hand, it is
also an essential intermediate material to produce antioxidants, plasticizer, synthetic rubber,
esters, etc [10-12]. However, isobutanol is mainly obtained as the by-product of butanol
3

synthesis during propylene carbonyl reaction. This general isobutanol synthesis method is unable
to meet the growing demand. Considering the increasing price of petroleum and the requirement
of the environmental friendly policy, a low cost and non-oil route for isobutanol synthesis is
necessary and has roused considerable research interests.
For isobutanol synthesis from syngas, recently, researchers mainly focus on three types catalysts:
alkali-modified Cu-based methanol synthesis catalysts [13-15], and alkali-modified Cr-based
methanol synthesis catalysts [16, 17] and ZrO2 based catalyst [18]. The ZnCr nanospinel catalyst
has been studied extensively since 1980s due to its high activity and selectivity for isobutanol
synthesis at high-temperature and high-pressure. The products obtained over the ZnCr catalysts
are mainly composed of methanol and isobutanol with a negligible amount of other components,
which make the industrial-scale isobutanol synthesis from syngas possible. The mechanism of
isobutanol synthesis from syngas on ZnCr catalyst is still not explicitly established. The catalyst
active sites and the reaction mechanism are always the focus of researchers in the past years.
Now, the majority debates of the active sites on ZnCr catalyst for isobutanol/methanol synthesis
can be grouped into the following three explanations: (a) the excess ZnO modified by alkali
metals on ZnCr spinel, (b) the non-stoichiometry phase (ZnCr) with cation disorder, (c) the
oxygen vacancies on ZnCr based catalysts. Epling et al. [19-21] had reported that the ZnCr
spinel was not important for isobutanol synthesis and acted only as a high-surface-area support.
As the near-surface region contained primarily ZnO with very little Cr, thus they drew a
conclusion that the alkali-modified ZnO phase was the active catalyst site. Meanwhile, they
pointed out that the hydroxyl groups played an important role in the HAS (higher alcohol
synthesis) reaction. Removal of hydroxyl groups could expose more of the active catalyst sites.
Nevertheless, Piero et al. [22] reported that the non-stoicheiometric spinel-like phase (ZnCr)
4

must be the active sites for methanol synthesis from syngas at high temperature. An unusual
co-ordination could be formed when the Zn atoms occupied the octahedral sites of ZnCr spinel,
which facilitated methanol synthesis. There are also some other studies presenting the similar
opinions [23,17]. In our previous studies, our group has performed a lot of works on the reaction
mechanism and catalyst active sites for isobutanol synthesis from syngas. Wu et al. [24-26]
indicated that COH→CCOH was the slow initial growth step for isobutanol synthesis on
ZrO2-based catalyst based on chemical enrichment method. The isobutanol was derived from the
C2+ alcohols undergoing an aldo-condensation reaction at high temperature. Tian et al. [27-30]
demonstrated that the non-stoichiometric spinel (ZnxCr2/3(1-x)O), presenting variation of cations
on the tetrahedral and octachedral coordination sites, was the active catalytic sites for isobutanol
formation. And the isobutanol productivity increased linearly with the increase of the cation
disorder degree. In the recent years, more and more researchers consider the oxygen vacancies of
ZnCr based catalysts are the active sites for methanol/isobutanol synthesis from syngas. The
oxygen vacancies could improve the activation ability of C-O band on surface of different spinel,
such as ZnCr2O4 [31], Fe3O4 [32], Cu/ZnO [33] and ZnCa2O4 [34]. Lunkenbein et al. [35] showed
that the sintering of ZnO and the loss of oxygen vacancies caused a decrease in the catalytic
activity of Cu-based catalysts for methanol synthesis.
Although the ZnO phase shows almost no catalytic activity for isobutanol synthesis, however,
we found that the addition of ZnO phase could obviously enhance the catalytic performance of
ZnCr spinel catalysts both in CO conversion and isobutanol selectivity. The similar results had
also been reported by Kuld et al. for the methanol synthesis reaction over the ZnO promoted Cu
catalysts [36], and they deemed that the improved catalyst performance was related to the Zn
coverage and the size-dependent thermodynamic activities of the Cu and ZnO nanoparticles. Our
5

previous study [27] has proved that there exists interaction between ZnO and ZnCr spinel, but
how the interaction works is not explicit. In order to further explore the roles of ZnO and ZnCr
spinel for isobutanol synthesis from syngas, in this report we designed a series of ZnCr based
catalysts via a sol-gel method, discovering the function of hydroxyl groups under the interaction
between ZnO and ZnCr spinel. In addition, the chemical and physical properities of catalysts
constructed by different preparation methods could significantly affect catalysts performance.
Compared with the previous studies [27-31], in which the ZnCr nanospinel catalysts used for
isobutanol synthesis are mainly prepared by co-precipitation method, the sol-gel method in this
paper could achieve molecular level mixing which facilitates the formation of polycrystalline
homogeneous particles with improved properties.
2. Experimental Method
2.1 Catalyst preparation.
All the catalysts were prepared by sol-gel method and the used chemicals were analytic grade.
The pure ZnO was prepared by using Zn(NO3)2·6H2O as precursor and oxalic acid as
complexing agent. 59.5 g Zn(NO3)2·6H2O was dissolved in 200 mL ethanol under vigorous
stirring at 50oC, and then 200 mL ethanol solution containing 29.0 g oxalic acid was added into
the above solution by peristaltic pump with the flow rate of 20 mL/min under vigorous stirring.
The obtained solution was aged for 20 h at room temperature to form a sol. Subsequently, the sol
was directly heated at 70 oC for 2 h and 120 oC for another 10 h to evaporate the ethanol solvent
to form a gel. Finally, the gel was calcined at 250 oC for 2 h to decompose the organic chemicals
completely, followed by calcination at 400 oC for 6 h to obtain the pure ZnO sample.
We also prepared a series of ZnCr catalysts ZnxCr1-y (x=0.5; y=800, 600，400) and ZnxCr1-y
(x=1, 1.5 or 3; y=400) by the similar method to the above ZnO catalysts preparation.
6

Zn(NO3)2·6H2O and Cr(NO3)3·9H2O were employed as Zn and Cr resources, respectively. The
amount of oxalic acid is excess 15 mol% in order to ensure all the metal ions completely
chelating with oxalic anion. After aging, evaporation of the ethanol solvent and calcination, the
ZnxCr1-y (x=0.5; y=800, 600，400) and ZnxCr1-y (x=1, 1.5 or 3; y=400) catalysts were obtained,
where the “x” in the catalyst name stands for Zn/Cr molar ratio, the “y” means ultima calcination
temperature. As the Zn/Cr molar ratio was 0.5, only ZnCr2O4 phase formed. When the Zn/Cr
molar ratio exceeded 0.5, besides ZnCr2O4 phase, the ZnO phase also formed.
For the sol-gel method, after the addition of oxalic acid ethanol solution into the metal nitrate
solution, the pH value of the sol was less than 1. Since the pH value can influence the structure
and morphology of the ZnCr catalysts, we used ammonia solution (28 wt%) to adjust the pH
value of the sol, by which to get a series of ZnCr spinel catalysts. Required amount of ammonia
solution was added to keep the pH value of the final sol solutionat 1.0, 2.0, 3.0 and 4.0. After
aging, evaporation of the ethanol solvent and calcination, a series of catalysts (Zn/Cr=1.0, 400
oC) named Zn1Cr1-400~z were obtained, where the “z” denotes the pH value of sol solution.
All the catalysts were impregnated with potassium carbonate through an incipient wetness
method to get the final 3 wt% K2O promoted catalysts for isobutanol direct synthesis from
syngas. The size of all the catalysts were in 30~40 mesh.
2.2 Catalytic performance measurement.
o
All the catalysts were first reduced with a 10% hydrogen-in-nitrogen mixture for 4 h at 400 C
on-line before reaction. Then, these catalysts were tested for isobutanol catalytic activity in a
fixed bed using a feed gas of CO:H2 = 2.6:1 at a space velocity of 3,000 h-1. The reactions were
o
performed at 400 C and 10 MPa. Products were analyzed by four on-line GC during the
reaction. Inorganic gas products of CO, H2, CH4 and CO2 were detected on-line by thermal
7

conductivity detector using a GC4000A (carbon molecular sieves column). The organic gas
products of hydrocarbons and methanol were detected on-line by flame ionization detector using
a GC4000A (GDX-403 column). The alcohol products in liquid phase were detected by flame
ionization detector using a GC-7AG (Chromsorb101). The H2O and methanol products in liquid
phase were detected by thermal conductivity detector using a GC4000A (GDX-401column).
2.3 Catalyst characterization.
The X-ray diffraction (XRD) data of the catalysts were collected using a D8 Advance X-ray
diffractometer in a 2θ range from 10o to 80o with Cu Kα radiation. The morphology and
microstructure of catalysts were investigated using a field emission transmission electron
microscope (HRTEM, JEM-2100F) operated at 400kV. The (in-situ) XPS patterns were recorded
using an AXIS ULTRA DLD X-ray photoelectron spectrometer equipped with a multichannel
detector. Charge referencing was done against adventitious carbon (C 1s, 284.8 eV). A
Shirley-type background was subtracted from the signals. The recorded spectra were fitted using
Gauss–Lorentz curves to determine the surface composition of different samples. FT-IR Spectra
were obtained with a TENSOR-27 in the range from 4000 to 600 cm−1 with 4 cm−1 resolution.
In-situ IR Spectra of H2 reduction were recorded by a TENSOR-27 from beginning introduction
o
of hydrogen at 400 C. In-situ IR Spectra of CO adsorption and desorption were obtained by a
TENSOR-27 in the range from 600 to 4000cm−1 with 4 cm−1 resolution after H2 reduction for 0.5
o
o
h at 400 C on line. CO adsorption was taken at 400 C and after 0.5 h of pure Ar flow at the
same temperature and the adsorbates existing on the surface of different ZnCr spinel catalysts
would be measured. The NH3 Temperature Programmed Desorption (NH3-TPD) was recorded on
o
TP-5050 automatic chemical adsorption instrument in the range from 50oC to 600 C. Pure
standard mixtures (zeolite + activated carbon) were used during preliminary TPD measurements.
8

It was saturated with NH3 flow at 50oC after pretreatment at 400 oC for 180 min in reducing gas
of 10% H2 in Ar. Then pure Ar flowed for 120 min at the same temperature to eliminate the
adsorption of physically adsorbed water and impurities. Finally, the temperature was increased at
a constant rate of 5oC min-l under an appropriate amount of Ar flow. The desorbed species were
analyzed on GC4000A by thermal conductivity measurements. The chemical compositions of the
catalysts were measured by inductively coupled plasma–atomic emission spectroscopy (ICP–
AES) using a Thermo iCAP 6300 instrument.
3. Results and discussion.
The catalytic performance of ZnO, ZnxCr1-y and Zn1Cr1-400~z catalysts for isobutanol
synthesis from syngas is shown in Table 1. The ZnO catalyst shows the CO conversion of 16.8%,
but the selectivity for alcohol is only 21.6%. The main product of the alcohol is methanol (78.7
wt%) and only a little amount of isobutanol (2.9 wt%) is produced. Therefore, the alkali metal
promoted ZnO is not a good catalyst candidate for isobutanol synthesis. For the Zn0.5Cr1-y
(y=800, 600 or 400, ZnCr2O4 phase) catalysts, the selectivity of alcohol increases from 11.4% to
o
o
60.1% with decreasing the calcination temperature from 800 C to 400 C and the total alcohol
production ratio also reaches a maximum value for Zn0.5Cr1-400 of 0.051 g.mL-1.h-1. Meanwhile,
the weight percentage of isobutanol increases clearly from 1.6 wt% to 14.7 wt%. These data
indicate that the catalyst with lower calcination temperature has better catalytic performance both
on higher catalyst activity and higher isobutanol selectivity. As shown by the XRD patterns in
Fig. 1 and HRTEM images in Fig. 2, the lower calcination temperature results in lower
crystallinity and smaller particle size, which maybe contribute to the formation of defects or
activity sites for the isobutanol synthesis from syngas. Because the reaction temperature of this
9

conversion process is 400 oC, there is no point in decreasing the calcination temperature further.
If the Zn/Cr molar ratio in ZnxCr1-400 catalysts exceeds 0.5, the formation of ZnO phase can
be confirmed clearly. Compared with Zn0.5Cr1-400 catalyst, the increased Zn/Cr molar ratio of
1.0 in Zn1Cr1-400 catalyst results in promoted CO conversion from 11.3% to 16.8% and
enhanced isobutanol selectivity from 14.7 wt% to 20.9 wt%. It must be stressed that though the
ZnCr spinel (Zn0.5Cr1-400) has a certain catalytic activity for isobutanol synthesis, its activity is
substantially boosted by the interaction with ZnO (Zn1Cr1-400). The in-situ CO FT-IR in Fig. 6
and in-situ XPS in Fig. 9 have verified that the hydroxyl groups could be consumed by CO under
the effect of synergy between ZnO and ZnCr spinel, whereby to increase the amount of oxygen
vacancies for isobutanol synthesis from syngas. The Zn1.5Cr1-400 catalyst gives similar CO
conversion and isobutanol selectivity to that of Zn1Cr1-400 catalyst, but lower total alcohol
selectivity of 60.7%. Further increasing the Zn/Cr molar ratio to 3.0, however, results in a sharp
decrease on isobutanol selectivity (11.8 wt%) and CO conversion (13.1%). Higher Zn/Cr molar
ratio like 3.0 will decrease the amount of ZnCr2O4 phase in ZnCr catalyst, and the ZnO phase is
the predominant component, therefore resulting in lower activity. The catalytic performance of
ZnCr catalysts with varied Zn/Cr molar ratios is shown in Table S1and Fig. S1a.
In order to adjust the amount of hydroxyl groups and the interaction between ZnO and ZnCr
spinel, ammonia solution (28 wt%) was used to change the pH value of the sol solution for
Zn1Cr1-400~z catalysts preparation. Compared with Zn1Cr1-400 catalyst, the Zn1Cr1-400~1.0
and Zn1Cr1-400~2.0 catalysts show a great increase both in the CO conversion and the
isobutanol selectivity. A maximum total alcohol production rate of 0.103 g.mL-1.h-1 is achieved at
pH=2.0. Increasing the pH value to 4.0, however, results in a sharp decrease both on the total
alcohol production rate and isobutanol selectivity. The catalytic performance of ZnCr catalysts
10

with varied pH value of sol solution is shown in Table S2 and Fig. S1b.
The stability of the Zn1Cr1-400~2.0 catalyst is further measured in Fig S2. The results
demonstrate that the Zn1Cr1-400~2.0 catalyst shows excellent stability over 300 h on stream. The
CO conversion could stably maintain at 22 ± 2 % and the total alcohol selectivity is only reduced
by 4% throughout the test. In the first 160 h on stream, the weight percentage of isobutanol in
liquid products keeps above 24 wt%, and slightly decreases in the rest of the test. Thus, the
Zn1Cr1-400~2.0 catalyst is a suitable catalyst candidate for isobutanol synthesis directly from
syngas.
The XRD patterns of different catalysts are presented in Fig. 1. The ZnO catalyst in Fig, 1a
displays the typical peaks at 31.8o, 34.4o, 36.3o, 47.5o, 56.6o and 62.9o related to ZnO phase
(JCPDS cards 36-1451), indicating that ZnO crystal formed well in the sol-gel method. For the
catalysts of Zn0.5Cr1-400, Zn0.5Cr1-600 and Zn0.5Cr1-800 in Fig. 1a, only some typical peak
appeared at 30.3o, 35.7o, 43.4o, 57.5o and 63.1o, which is ascribed to ZnCr2O4 (JCPDS cards
22-1107). It should be noted, here, the diffraction intensity of different catalysts is different. By
decreasing the calcination temperature, the diffraction peaks become broader and the intensity
decreases linearly, illustrating the smaller crystal particle size obtained at lower calcination
temperature. In our previous study, we had proved that the calcination temperature employed for
ZnCr2O4 spinel preparation greatly influenced the crystal structure [29, 30]. Lower calcination
temperature would cause low crystallinity of ZnCr2O4 spinel, meanwhile, leading to serious
cation disorder and more structure defects. This kind of ZnCr2O4 spinel with cation disorder was
called non-stoichiometric spinel (ZnxCr2/3(1-x)O), as reported by other articles [37, 23]. The
11

particle size calculated by Scherrer equation decreases from 24.7 nm to 4.8 nm with decreasing
o
o
the calcination temperature from 800 C to 400 C. The ZnCr spinel catalysts with excess ZnO,
ZnxCr1-y (x=1.0, 1.5, 3.0; y=400), exhibit two kinds of typical diffraction peaks corresponding to
ZnCr2O4 and ZnO respectively. Among these catalysts, the increasing trend of the intensity of
ZnO diffraction peaks is more obvious with Zn/Cr molar ratio increasing from 1.0 to 3.0,
however, the ZnCr2O4 diffraction peaks change little.
We also used ammonia solution to adjust the pH value of sol solution to get the catalysts with
varied crystallinity. The XRD patterns of the prepared catalysts are shown in Fig. 1b. It is shown
that the intensity of diffraction peaks of both ZnO and ZnCr2O4 phases is much lower as the pH
value is in the range of 1~3. Accordingly, the ZnO and ZnCr2O4 particle size of Zn1Cr1-400~1.0,
Zn1Cr1-400~2.0, Zn1Cr1-400~3.0 catalysts should be much smaller than those of Zn1Cr1-400~4.0
catalyst. By careful observation, the crystallinity of Zn1Cr1-400~2.0 is lowest among these
catalysts. It is worth noting that there is no diffraction peaks of potassium species in the XRD
pattern of all catalysts, indicating the potassium species would be amorphous or dispersed well
so that they could not be detected by XRD.
The HRTEM was employed to characterize the morphology and microstructure of ZnO and
ZnxCr1-y catalysts. The TEM images are shown in Fig. 2. The TEM images of ZnO catalyst in
Fig. 2a show that the particles are well-separated, and the particles size is in the range of 20~60
nm, peaking at 30~40 nm, much bigger than that of other catalysts. For the Zn0.5Cr1-y catalysts
in Fig. 2b-d, the mean particle sizes of Zn0.5Cr1-800, Zn0.5Cr1-600 and Zn0.5Cr1-400 catalysts are
obtained as about 25 nm, 11 nm and 4.5 nm, respectively. This decline trend should be attributed
to the lower calcination temperature, which is helpful to reduce the crystal size. Furthermore, the
12

typical lattice spacing of 4.81 Å, 2.95 Å, 2.51 Å, are corresponded to the (111), (220), (311)
planes of ZnCr2O4 nanocrystal. ZnO phase is not observed as the Zn/Cr molar ratio is 0.5. If the
Zn/Cr molar ratio exceeds 0.5 (Fig. 2e-g), the TEM images clearly demonstrate that, besides
ZnCr2O4 nanocrystal, the ZnO phase also formed with a typical lattice spacing of 2.48Å
corresponded to (101) plane of ZnO crystal. These results are in good agreement with XRD
results in Fig. 1. It should also be noted that the particle size of ZnO on Zn1Cr1-400 catalyst is
much smaller than pure ZnO sample in Fig. 1a that makes a close connection between ZnO and
ZnCr2O4 phase and is helpful to improve the interaction between them. With increasing the
Zn/Cr molar ratio from 1.0 to 3.0, the particle size of ZnO crystal increases obviously, but the
particle size of ZnCr2O4 crystal changes slightly. For Zn3Cr1-400 catalyst, the amount of ZnO
phase with the particle size in the range of 20~30 nm increases to be the predominant component
of the sample. The aggregation degree of ZnO particles on this Zn3Cr1-400 catalyst is higher than
other ZnxCr1-400 catalysts, finally resulting in poor catalytic performance both in CO conversion
(13.1%) and isobutanol selectivity (11.8 wt%).
The TEM images of Zn1Cr1-400~z catalysts prepared with different pH value of sol solution
are shown in Fig. 3. Compared with Zn1Cr1-400 catalyst in Fig. 2e, the TEM images of
Zn1Cr1-400~z catalysts exhibit different morphology. As the pH value is in the range of 1~3, the
particles are relative small and the smallest particle size is obtained at pH = 2 which is helpful to
improve the isobutanol selectivity (24.2 wt%) for Zn1Cr1-400~2.0. Further increasing the pH
value of sol solution can change the morphology of ZnO and ZnCr2O4 crystal at the same time.
For Zn1Cr1-400~4.0 catalysts, the morphology and particle size of ZnO and ZnCr2O4 increase
clearly. It could be easily observed that the ZnO phase and the ZnCr2O4 phase separate from
each other with a particle size in the range of 20~50 nm when the pH value is 4 (Fig. 3d). The
13

poor catalytic performance of Zn1Cr1-400~4.0 in CO conversion (13.0%) and isobutanol
selectivity (4.6 wt%) should be contributed to the high crystallinity and poor interaction between
ZnO and ZnCr2O4.
The FT-IR spectrum of ZnO, ZnxCr1-y and Zn1Cr1-400~z catalysts is shown in Fig. 4. The
broad band at around 3480 cm-1 should be attributed to the H-O band vibration modes of
hydroxyl groups [38-41]. For the ZnO catalyst, the amount of hydroxyl groups is more than other
catalysts. It is clear that, with the decrease of calcination temperature from 800 oC to 400 oC, the
FT-IR bands of H-O vibration modes of the ZnCr2O4 spinel catalysts increase obviously. As the
cation disorder could be more serious at low calcination temperature, the Zn2+ ions are located in
octahedral sites in a random substitution of Cr3+ions, leaving some of the tetrahedral vacant sites
[22]. Then, the oxygen atoms around theses vacant sites would tend toward to form hydroxyl
groups. Additionally, increasing the Zn/Cr molar ratio of ZnxCr1-400 catalyst from 0.5 to 3.0
further increases the content of hydroxyl groups. This fact might be explained by the formation
of ZnO phase on the surface of ZnCr spinel, which is consistent with the X-ray diffraction results
in Fig. 1. As the ammonia solution (28wt%) is used to adjust the pH value of sol solution, the
intensity of H-O band vibration modes become stronger at pH =2 and 3 and weaker at other pH
values. Especially, as the pH value of the sol solution is 4, the amount of hydroxyl groups of
Zn1Cr1-400~4.0 catalyst is much less than that of other catalysts.
The band at 1637 cm-1 is assigned to the surface adsorbed oxygen species (Oads) [42, 43]. We
found that the non-stoichiometric spinel with disordered cations was not stable [28]. Large
numbers of dangling bonds (Oads+ OOH-) existing in the surface of ZnxCr1-400 and ZnxCr1-400~z
catalysts would stabilize this metastable state. However, the band signals at 1637 cm-1 for
Zn0.5Cr1-800 and Zn1Cr1-400~4.0 catalysts are hardly observed, because the high calcination
14

temperature and high pH value of sol solution cause the catalysts with higher crystallinity. This
result also led to the low reaction activity of Zn0.5Cr1-800 and Zn1Cr1-400~4.0 catalysts during
the reaction of isobutanol synthesis from syngas as shown in Table 1. The band at 1380 cm-1 and
-
830 cm-1 should be assigned to NO3 ion that originated from nitrate precursor [44-46]. The band
at 1100 cm-1 corresponded to O=C=O vibration is derived from oxalate precursor [38, 39]. For
-
Zn0.5Cr1-800 catalyst, the bands related to NO3 ion and O=C=O vibration are much weaker than
other catalysts, suggesting that the residues from precursor materials for this catalyst preparation
had been removed completely by high temperature calcination. The broad peak near 940 cm-1
should be assigned to the deformation mode of Zn-O at 945 cm-1 merged with the band of Cr6+ at
935 cm-1 [47]. The band at 1583 cm-1 is ascribed to the asymmetric stretching vibration of C=O
mode [45], which is only observed for ZnO catalyst.
The acidic properties of the catalysts were evaluated by NH3-TPD. The profiles are shown in
Fig. 5. and the acid sites distribution is shown in Table S3. The acidity pattern shows that there
are two main types of desorption peaks on catalysts, indicating two kinds of acid sites: a large
o
amount of weak acid sites and a small amount of strong acid sites in range of 150 ~ 300 C and
o
400 ~ 550 C respectively. For ZnO and Zn0.5Cr1-800 catalysts, the amount of both weak and
strong acid sites is small, even hardly being detected. But by decreasing calcination temperature,
the amount of weak acid sites of ZnCr spinel (Fig. 5b-d) significantly increases and a certain
amount of strong acid sites appear. When the Zn/Cr molar ratio exceeds 0.5 (Fig. 5d-g), the area
of weak acid sites increases and the peak becomes wide. Meanwhile, the low temperature peaks
shifts towards higher temperature, but the strong acid peaks do not change obviously. Hoflund et
al. [48] had used Mn atom to replace a part of Cr atom in ZnCr spinel. Since the acidity of Mn is
15

weaker than that of Cr, the catalyst showed low hydrocarbon selectivity. In our study [29], we
have found that the existed weak acid sites is beneficial to improve the catalytic activity for the
isobutanol synthesis from syngas, on the contrary, the strong acid sites on the catalysts would
depress the formation of alcohol and isobutanol.
Fig. 5h-k shows the acidic properties of ZnCr catalysts prepared with the sol solution that had
been adjusted on its pH value. Compared with Zn1Cr1-400~1.0, the amount of both weak acid
sites and strong acid sites of Zn1Cr1-400~2.0 catalyst increase obviously, but further increasing
pH value leads to smaller acid site amount, even no acid peaks for Zn1Cr1-400~4.0. As the
results shown in Table S3, the amount of weak acid sites reaches a maximum of 40.59 umol/g at
pH=2.0. It is worth noting that the weak acid peaks of Zn1Cr1-400~z catalysts shift to lower
temperature with increasing the pH value, indicating that the intensity of weak acid sites become
weaker. The proper acidic properties of Zn1Cr1-400~2.0 may be an important reason for its high
alcohol yield during the isobutanol synthesis.
The in-situ CO FT-IR was employed to detect the change of species after adsorbing CO
o
molecule at 400 C, and the result profiles are shown in Fig. 6. The inversus peak at 3480 cm-1
shows the consumption of hydroxyl groups, and the bands at 1560 cm-1 and 1375 cm-1
-
correspond to the vibration of νas(CO3 ) and σ(CH) originating from formate species [49, 50].
The peak at 1330 cm-1 could be attributed to the carbonate species [49, 50]. For the ZnO catalyst,
only a little amount of formate species formed after adsorbing CO molecule. For the Zn0.5Cr1-y
catalysts (Fig. 6b-d), a large number of formate and carbonate species are produced by
o
o
calcination from 800 C to 400 C. In general, the formate species is considered as an important
intermediate for alcohol synthesis. Waugh et al. [51] had found that the formate species was a
stable intermediate during the conversion of CO/CO2 to methanol, while the carbonate species
16

was not essential to alcohol synthesis beacuse it was easily converted to CO2. Tian [28, 30] and
Wu [26] reported that the formate species was a very significant C1 intermediate for isobutanol
from syngas and its formation was associated with consuming hydroxyl groups and adsorbed CO
on catalyst surface. In this study, however, formate species could be formed on the Zn0.5Cr1-y
catalysts without consuming hydroxyl groups, although there is a certain amount of hydroxyl
group verified by FT-IR spectrum in Fig. 4.
It needs to be stressed that, though a large amount of formate species are formed for the
Zn0.5Cr1-800 catalyst, its catalytic activity is much lower than others as shown in Table 1. This
means that besides the formate species (one C1 intermediate), there must exist another C1
intermediate (denoted as C1*) which is difficult to be formed for the catalysts calcinated at high
o
o
temperature. With decreasing calcination temperature from 800 C to 400 C, the isobutanol
distribution in the liquid products of Zn0.5Cr1-y catalysts increases from 1.6 wt% to 14.7 wt%. A
lot of defects and oxygen vacancies will exist on catalyst surface at lower calcination
temperature that may be essential to the formation of C1* intermediate.
Fig. 6d-g shows the in-situ CO FT-IR analysis results of ZnxCr1-400 catalysts as the Zn/Cr
molar ratio increases from 0.5 to 3.0. For Zn1Cr1-400, the carbonate species signal at 1300 cm-1
gets minimized among the ZnxCr1-400 catalysts. However the signal of formate species at 1560
cm-1 increased slightly with increasing the Zn/Cr molar ratio. For the inversus peak of hydroxyl
groups consumption at 3480 cm-1, an obvious hydroxyl consumption peak formed when the
Zn/Cr molar ratio exceeds 0.5, suggesting that the interaction between ZnO and ZnCr spinel
could improve the utilization of hydroxyl species. Moreover, the Zn1Cr1-400 shows the biggest
hydroxyl consumption peak compared with other ZnxCr1-400 catalysts. The interaction between
ZnO and ZnCr spinel, possibly, was more effective or stronger when the Zn/Cr molar ratio is 1.0.
17

As the reaction results exhibited in Table 1, the Zn1Cr1-400 gives excellent catalytic performance,
with total alcohol production rate of 0.098 g.mL-1.h-1 and isobutanol distribution of 20.9 wt%,
better than other ZnxCr1-400 catalysts. Epling et al. [19-21, 52, 53] showed that the hydroxyl
groups could interact with or cover the active catalytic sites in the near-surface region, and the
reductive pretreatment preformed on the catalysts before reaction in HAS (higher alcohol
synthesis) reactor resulted in the removal of hydroxyl groups, thereby exposing more of the
active catalyst sites. Bao and his co-wokers [31] reported that Density functional Theory (DFT)
calculations using the ZnCr2O4 spinel (111) surface as a model revealed that it was reductive,
resulting in a surface with a number of oxygen vacancies. Meng et al [54] found that the
enrichment of the acid sites could be related to the formation of oxygen vacancy by the reduction
of MoO3 and V2O5. Other studies had also reported that the oxygen vacancies could improve the
activation ability of C-O bond on surface of different spinels, for example Fe3O4 [32], Cu/ZnO
[33] and ZnCa2O4 [34]. The hydroxyl groups are not conducive to isobutanol synthesis, and
removing them would lead to the exposure of oxygen vacancies. The in-situ H2 reduction FT-IR
of Zn1Cr1-400 was also employed and the spectrum is shown in Fig. S3. The results reveal that,
although the catalysts are reduced by a 10% hydrogen-in-nitrogen mixture for 4 h at 400oC
before reaction, this reductive pretreatment could not remove the hydroxyl groups. On the
contrary, the amount of hydroxyl groups increases.
From the discussion above, some conclusions could be drawn cautiously that the amount of
hydroxyl groups increases with the formation of the non-stoichiometric spinel (calcinated at low
temperature) and the hydroxyl groups could be removed by CO under the synergy between ZnO
and ZnCr spinel to form more oxygen vacancies at reaction conditions. Except of formate
species as a C1 intermediate, another one C1 intermediate (denoted as C1*) exist potentially for
18

isobutanol synthesis from syngas. The formation of this new C1* is closely related with the
amount of oxygen vacancies and the interaction between ZnO and ZnCr spinel. Further
investigation on uncovering this new intermediate of C1* should be performed in future study.
For the Zn1Cr1-400~z catalysts shown in Fig. 6h-k, the inversus peaks of hydroxyl group
consumption decrease with increasing the pH value of sol solution from 1.0 and 2.0 to 3.0 and
4.0, and the carbonate species peaks shows similar tendency. While, the signal of formate species
gets maximized at pH at 2.0. Isobutanol synthesis from syngas is not determined by one factor,
as we known, and both of the oxygen vacancies and formate species play important roles in this
process. Adjusting of pH value of sol solution for catalyst preparation could change the
crystallinity and particle size of both ZnO and ZnCr2O4 spinel (Fig.1b and Fig.3), whereby to
influence the interaction between them and the amount of hydroxyl groups. The analysis results
here showed that the optimal pH value of sol solution is 2.0.
The x-ray photoelectron spectrometer was employed to characterize the composition of the
near-surface region of ZnxCr1-y and Zn1Cr1-400~z catalysts and the results are shown in Fig. S4.
Epling and his co-workers [52, 53] have reported that Zn is rich, but Cr is not, in the near-surface
region of ZnCr catalysts. They drew the conclusion that the alkali-promoted ZnO was the active
phase for HAS, and the ZnCr spinel only acted as a high-surface-area support. But in this study,
it shows that the predominant peaks corresponded to Zn, O and Cr are readily distinguished in
Fig. S4, and small amount of K features could be also observed. A clear change is about the Zn
features that become bigger with increasing the Zn/Cr molar ratio. The Zn 2p and Cr 2p XPS
analysis results of the fresh ZnxCr1-y and Zn1Cr1-400~z catalysts are shown in Fig. 7. The
19

binding energy of Zn 2p at 1022.2 eV and 1045.2 eV should be attributed to Zn2+ from ZnO and
ZnCr2O4 spinel [28, 55]. The binding energy of Cr 2p at 576.4 eV and 586.0 eＶ are derived
from Cr3+ of ZnCr2O4 spinel [28, 55, 56]. For the Zn0.5Cr1-y catalysts (y=800, 600 or 400), the
binding energy of Zn 2p3/2 constantly increases with decreasing the calcination temperature from
800 oC to 400 oC. This change is due to the decrease of electron density of Zn cation in ZnCr2O4
spinel [57], which verifies that the cation disorder is more serious and more defects or oxygen
vacancies could be formed for the catalyst prepared by lower calcination temperature. With the
Zn/Cr molar ratio increasing from 0.5 to 3.0, the amount of ZnO phase increases and the binding
energy of Zn 2p3/2 decreases. For the Zn1Cr1-400~z catalysts (z=1.0, 2.0, 3.0 or 4.0), the Zn 2p3/2
binding energy of Zn1Cr1-400~2.0 is higher than other catalysts, which may be owing to the
stronger interaction between ZnO and ZnCr2O4 phase at pH =2.0. The binding energy of Cr 2p of
different catalysts exhibits similar tendency with Zn 2p. The spectrum also shows a Cr6+ peak at
578.2 eV, indicating that the catalyst probably has been over oxidized.
The O 1s XPS analysis results of the fresh ZnxCr1-y and Zn1Cr1-400~z catalysts is shown in
Fig. 8. The main chemical state of O in the near-surface region of all the catalysts is the lattice
oxygen at 529.4 eV contributed by ZnO and ZnCr2O4 spinel [55, 56]. The O of hydroxyl species
at 531.7 eV are derived from surface hydroxyl groups [58, 59]. The molar ratio of (OOH-)/(Olatt
+
OOH-) obtained by quantitative calculation on the corresponding peaks areas is shown in Table 2.
-
For the ZnxCr1-y (x=0.5; y=800, 600 or 400) catalyst, the molar ratio of (OOH-)/(Olatt + OOH
)
increases from 12.37 % to 19.14 % along with decreasing the calcination temperature from 800
o
oC to 400 C. Because the ZnCr spinel (by low calcination temperature) with defect sites and
cations disorder is unstable, more hydroxyl groups as dangling bonds are formed to stabilize this
20

metastable state. Varying the Zn/Cr molar ratio for ZnxCr1-y (x=0.5, 1.0, 1.5 or 3.0; y=400)
catalysts preparation, obvious changes on the molar ratio of (OOH-)/(Olatt + OOH-) have also been
found. The hydroxyl groups signal increases when the Zn/Cr molar ratio is increased from 0.5 to
1.5, which is consistent with the FT-IR results in Fig. 4. For the Zn3Cr1-400 catalyst, a certain
amount of chemisorbed O is observed at 526.6 eV, as reported by previous studies [19, 52, 53],
which may be related to the chemical absorption on ZnO phase. Because the chemisorbed O
could be eliminated by the H2 reductive pretreatment preformed on the catalysts before the
reaction [19, 53], so it is helpless for the isobutanol synthesis. For Zn1Cr1-400~z catalysts, the
amount of hydroxyl groups has a maximum amount at pH=2 (OOH-/(Olatt + OOH-) = 25.52 %).
Thus more oxygen vacancies could be formed for Zn1Cr1-400~2.0 catalyst after the introduction
of CO at reaction temperature. A lowest value of OOH-/(Olatt + OOH-) appears on the
Zn1Cr1-400~4.0, possibly due to the high crystallinity of both ZnO and ZnCr2O4 obtained with
the pH value of 4.0.
The in-situ H2 reduction and in-situ CO O 1s XPS spectrum of Zn1Cr1-400 catalyst at reaction
temperature is shown in Fig.9, and the molar ratio of (OOH-)/(Olatt + OOH-) is shown in Table 2.
The in-situ H2 reduction O 1s XPS in Fig. 9b shows the amount of hydroxyl groups increases
apparently, which is consistent with the results of in-situ H2 reduction FT-IR in Fig. S3.
Nevertheless, the amount of hydroxyl groups significantly decreases with the introduction of CO
at reaction temperature as indicated by Fig. 9c. These results indicate that the hydroxyl groups
could not be consumed by the H2 pretreatment, and they could react with CO under the
interaction between ZnO and ZnCr spinel, which is consistent with the results of in-situ CO
FT-IR in Fig. 6. Untill now, there is no an effective quantitative analysis tool to measure the
oxygen vacancies on the surface of oxide catalyst during the reaction process. Therefore, the
21

surface compositions of Zn1Cr1-400 obtained by in-situ XPS analysis under different treatment
conditions in Table 3 are employed to indirectly analyze the change of surface O amount on the
ZnCr catalyst. It can be observed that the surface predominant elements of Zn, O, Cr and K
change little after the H2 treatment compared with fresh Zn1Cr1-400. However, the surface molar
ratio of O decreases from 51.97 % to 48.14 % with CO introduction after H2 pretreatment at 400
oC for 2 h. This phenomenon verifies that hydroxyl groups could be removed by CO,
consequently resulting in more exposed oxygen vacancies to improve the catalyst performance.
In addition, the molar ratio of Cr and K increases slightly after CO treatment, which should be
attributed to that the alkali metal is bound to Cr under reduction treatment [60].
The bulk compositions of Zn1Cr1-400, Zn1Cr1-400~2.0 and used Zn1Cr1-400~2.0 catalysts
(Zn, Cr, K) measured using inductively coupled plasma–atomic emission spectroscopy (ICP–
AES) are presented in Table S4. The results show that the element compositions of Zn1Cr1-400,
Zn1Cr1-400~2.0 catalyst are in agreement with recipe calculated from the element concentration
presented in the synthetic solution. This means that the expected Zn/Cr molar ratio could be
realized by sol-gel method with directly evaporating the ethanol solvent. For Zn1Cr1-400 catalyst,
compared with surface composition in Table 3, the Zn/Cr molar ratio in bulk is a little higher
than that at surface. The used Zn1Cr1-400~2.0 catalyst exhibits the similar element compositions
to the fresh Zn1Cr1-400~2.0 catalyst that may be the main reason for its excellent stability (Fig
S2).
The proposed mechanism of isobutanol synthesis is shown in Fig. 10. Based on the XRD and
HRTEM analysis results, a relative perfect crystal could be formed when the ZnCr spinel is
o
calcined at high temperature of 800 C; meanwhile, the particle size is much bigger than other
catalysts. Decreasing the calcination temperature could lead to small crystal particle size and low
22

crystallinity. On the other hand, lower calcination temperature will also cause the formation of
structure defects including inherent oxygen vacancies. This kind of ZnCr spinel with cation
disorder is also called non-stoichiometric spinel (ZnxCr2/3(1-x)O) that usually is in unstable state.
Then, the oxygen atoms around the vacant sites and structure defects would tend to form
hydroxyl groups to stabilize the metastable state, as proved by the FT-IR results (Fig. 4) that the
amount of hydroxyl groups increases with decreasing the calcination temperature. The in-situ H2
reduction FT-IR (Fig. S3) and in-situ H2 reduction XPS (Fig. 9b) reveal that the H2 pretreatment
before the reaction could not remove the hydroxyl groups, on the contrary, the amount of
hydroxyl groups increases with H2 introduction. As the hydroxyl groups could interact with or
cover the active catalytic sites, removing hydroxyl groups is helpful to expose these active sites
in the near-surface region of catalysts. According to the results of in-situ CO FT-IR in Fig. 6 and
the in-situ CO XPS in Fig. 9c, the hydroxyl groups could be consumed by CO molecule,
resulting in the increase of surface oxygen vacancies for C-O activation. And the interaction
between ZnO and ZnCr spinel could improve the utilization of hydroxyl groups. That may be
one important reason why the activity of ZnCr spinel (by low calcination temperature) could be
boosted by the interaction with ZnO for isobutanol synthesis, though the ZnO phase shows
almost no catalytic activity. Because the formate species could be easily formed (Fig. 6b), we
could draw a conclusion that there must exist another intermediate C1* whose formation is
related to the oxygen vacancies. But what the intermediate C1* is needs to be further
investigated in future study. The oxygen vacancies could be regenerated at the actual operation
conditions, which is very important to obtain a long catalyst lifetime for the ZnCr based catalysts
during the process of isobutanol synthesis from syngas [31]. A proposed pathway of isobutanol
synthesis is shown in Fig. S5. The C1→C2 process is rate-determining-step for isobutanol
23

synthesis with coupling of two C1 intermediates [25, 61-64,]. After the C2 alcohol is formed, the
isobutanol could quickly synthesis by aldo-condensation reaction of C2+ alcohols [65, 66]. As
stereo-hindrance effect is stronger for side reactions, the products obtained over the ZnCr
catalysts are mainly composed of methanol and isobutanol with a negligible amount of other
components [25].
Conclusion.
In this study, a series of ZnO, ZnxCr1-y and ZnxCr1-y~z catalysts were prepared by sol-gel
method. With decreasing the calcination temperature for catalyst preparation, the crystallinity of
ZnCr2O4 became lower with a certain amount of structure defects (including oxygen vacancies),
and the particle size of catalysts also decreased obviously from the TEM images. Meanwhile, the
amount of hydroxyl groups increased according to the FI-IR and XPS analysis results. The
hydroxyl groups could be consumed by CO instead of H2 under the interaction between ZnO and
ZnCr spinel. Removing hydroxyl groups was helpful to expose more oxygen vacancies (the
active sites for C-O activation). For the reaction of isobutanol synthesis directly from syngas,
besides formate species (an important C1 intermediate), there must exist another intermediate
C1* whose formation was closely related to the oxygen vacancies. The proper calcination
temperature and Zn/Cr molar ratio for ZnCr catalysts preparation to realize efficient isobutanol
o
synthesis from syngas were 400 C and 1.0, respectively. Varied pH value of sol solution for
catalyst preparation could obviously affect the amount of hydroxyl groups and nanocrystal
morphology of the obtained ZnCr catalysts. The Zn1Cr1-400~2.0 catalyst (the pH value of sol
solution is 2.0) showed more hydroxyl groups and a relative small particle size of ZnO and
ZnCr2O4, realizing excellent interaction between ZnO and ZnCr spinel, whereby to exhibit the
best catalytic performance for isobutanol synthesis from syngas: the CO conversion of 20.9%
24

and the isobutanol selectivity of 24.2 wt%.
Conflict of interest
The authors declare no competing financial interest.
Acknowledgements
This work was supported by the National Natural Science Foundation of China
(21573269); The Key Research and Development Program of Shanxi Province,
Special Research Foundation of Shanxi Province for Coal-based Synthesis of Fine
Chemicals.
Notes and references
[1] I. Wender, Fuel Process. Technol. 48 (1996) 189-297.
[2] Y. Xiang, V. Chitry, P. Liddicoat, P. Felfer, J. Cairney, S. Ringer, N. Kruse, J. Am. Chem.
Soc. 135 (2013) 7114-7117.
[3] H. M. Torres Galvis, K. P. de Jong, Acs Catal. 3 (2013) 2130-2149.
[4] V. Navarro, M. A. van Spronsen, J. W. M. Frenken, Nat. Chem. 8 (2016) 929-934.
[5] A. J. Medford, A. C. Lausche, F. Abild-Pedersen, B. Temel, N. C. Schjødt, J. K. Nørskov,
F. Studt, Top. Catal. 57 (2014) 135-142.
[6] V. Subramani, S. K. Gangwal, Energy & Fuels 22 (2008) 814-839.
[7] J. G. Nunan, C. E. Bogdan, K. Klier, K. J. Smith, C.W. Young, R. G. Herman, J. Catal. 116
(1989) 195-221.
[8] J. Anton, J. Nebel, H. Song, C. Froese, P. Weide, H. Ruland, M. Muhler, S. Kaluza, J.
Catal. 335 (2016) 175-186.
[9] X. Mo, Y.-T. Tsai, J. Gao, D. Mao, J. G.Goodwin Jr, J. Catal. 285 (2012) 208-215.
25

[10] L. Yu, S. Huang, S. Zhang, Z. Liu, W. Xin, S. Xie, L. Xu, Acs Catal. 2 (2012) 1203-1210.
[11] J. D. Taylor, M. M. Jenni, M. W. Peters, Top. Catal. 53 (2010) 1224-1230.
[12] G. Busca, Chem. Rev. 107 (2007) 5366-5410.
[13] N.-Y. Topsøe, H. Topsøe, J. Mol. Catal. A-Chem. 141 (1999) 95-105.
[14] J. G. Nunan, C. E. Bogdan, K. Klier, K. J. Smith, C.-W. Young, R. G. Herman, J. Catal. 113
(1988) 410-433.
[15] M. Xu, E. Iglesia, J. Catal. 188 (1999) 125-131.
[16] W. Keim, W. Falter, Catal. Lett. 3 (1989) 59-63.
[17] L. Tan, G. Yang, Y. Yoneyama, Y. Kou, Y. Tan, T. Vitidsant, N. Tsubaki, Appl. Catal.
A-Gen. 505 (2015) 141-149.
[18] D. He, Y. Ding, H. Luo, C. Li, J. Mol. Catal. A-Chem. 208 (2004) 267-271.
[19] W. S. Epling, G. B. Hoflund, W. M. Hart, D. M. Minahan, J. Catal. 169 (1997) 438-446.
[20] G. B. Hoflund, W. S. Epling, D. M. Minahan, Catal. Lett. 45 (1997) 135-138.
[21] D. M. Minahan, W. S. Epling, G. B. Hoflund, Appl. Catal. A-Gen. 166 (1998) 375-385.
[22] G. D. Piero, F. Trifiro, A. Vaccari, Chem. Commun. (1984) 656-658.
[23] E. Errani, F. Trifiro, A. Vaccari, M. Richter, G. Del Piero, Catal. Lett. 3 (1989) 65-72.
[24] Y. Wu, H. Xie, Y. Kou, L. Tan, Y. Han, Y. Tan, J. Fuel Cell Sci. Technol. 41 (2013)
868-874.
[25] Y. Wu, H. Xie, Y. Kou, N. Tsubaki, Y. Han, Y. Tan, Korean J. Chem. Eng. 32 (2015)
406-412.
[26] Y. Wu, H. Xie, S. Tian, N. Tsubaki, Y. Han, Y. Tan, J. Mol. Catal. A-Chem. 396 (2015)
254-260.
[27] S. Tian, L. Tan, Y. Wu, Y. Kou, H. Xie, N. Tsubaki, Y. Tan, Appl. Catal. A-Gen. 536 (2017)
26

57-66.
[28] S. Tian, S. Wang, Y. Wu, J. Gao, P. Wang, H. Xie, G. Yang, Y. Han, Y. Tan, Catal. Sci.
Technol. 6 (2016), 4105-4115.
[29] S. Tian, S. Wang, Y. Wu, J. Gao, Y. Bai, P. Wang, H. Xie, Y. Han, Y. Tan, J. Mol. Catal.
A-Chem. 404–405 (2015) 139-147.
[30] S. Tian, S. Wang, Y. Wu, J. Gao, H. Xie, X. Li, G. Yang, Y. Han, Y. Tan, RSC Adv. 5
(2015) 89273-89281.
[31] F. Jiao, J. Li, X. Pan, J. Xiao, H. Li, H. Ma, M. Wei, Y. Pan, Z. Zhou, M. Li, S. Miao, J. Li,
Y. Zhu, D. Xiao, T. He, J. Yang, F. Qi, Q. Fu, X. Bao, Science 351 (2016) 1065-1068.
[32] Y. Tamaura,; M. Tahata, Nature 346 (1990) 255-256.
[33] J. Xiao, T. Frauenheim, J. Phys. Chem. Lett. 3 (2012) 2638-2642.
[34] C. Jia, W. Fan, X. Cheng, X. Zhao, H. Sun, P. Li, N. Lin, Phys. Chem. Chem. Phys. 16
(2014), 7538-7547.
[35] T. Lunkenbein, F. Girgsdies, T. Kandemir, N. Thomas, M. Behrens, R. Schlögl, E. Frei,
Angew. Chem.-Int. Edit. 128 (2016) 12900-12904.
[36] S. Kuld, M. Thorhauge, H. Falsig, C. F. Elkjær, S. Helveg, I. Chorkendorff, J. Sehested,
Science 352 (2016) 969-974.
[37] S. Chen, Y. Wu, P. Cui, W. Chu, X. Chen, Z. Wu, J. Phys. Chem. C 117 (2013)
25019-25025.
[38] X. Kong, X. Zhang, J. Chen, Catal. Commun. 65 (2015) 46-50.
[39] L. Shi, Tan, Y. N. Tsubaki, ChemCatChem 4 (2012) 863-871.
[40] H. Li, J. Wang, H. Liu, C. Yang, H. Xu, X. Li, H. Cui, Vacuum 77 (2004) 57-62.
[41] R. Wahab, S. G. Ansari, Y.-S. Kim, H.-K. Seo, H.-S. Shin, Appl. Surf. Sci. 253 (2007)
27

7622-7626.
[42] Z.-X. Wei,; L. Wei,; L. Gong,; Y. Wang, C.-W. Hu, J. Hazard. Mater. 177 (2010) 554-559.
[43] L. Wu, J. C. Yu, L. Zhang, X. Wang, S. Li, J. Solid State Chem. 177 (2004) 3666-3674.
[44] Z. Yue, L. Li, J. Zhou, H. Zhang, Z. Gui, Mater. Sci. Eng. B 64 (1999) 68-72.
[45] X. L. Cheng, H. Zhao, L. H. Huo, S. Gao, J. G. Zhao, Sens. Actuator B-Chem. 102 (2004)
248-252.
[46] L. Shi, K. Tao, R. Yang, F. Meng, C. Xing, N. Tsubaki, Appl. Catal. A-Gen. 401 (2011)
46-55.
[47] A. Venugopal, R. Sarkari, S. N. Kumar, M. K. Kumar, S. S. John, J. K. Reddy, A. H.
Padmasri, J. Chem. Sci. 126 (2014) 387-393.
[48] G. B. Hoflund, W. S. Epling, D. M. Minahan, Catal. Lett. 62 (1999) 169-173.
[49] A. Riva, F. Trifiro, A. Vaccari, L. Mintchev, G. Busca, J.Chem. Soc. Faraday Transactions
1: Physical Chemistry in Condensed Phases 84 (1988) 1423-1435.
[50] K. Pokrovski, K. T. Jung, A. T. Bell, Langmuir 17 (2001) 4297-4303.
[51] K. C.Waugh, Catal. Lett., 142 (2012) 1153-1166.
[52] W. S. Epling, G. B. Hoflund, D. M. Minahan, React. Kinet. Catal. Lett. 67 (1999) 225-232.
[53] W. S. Epling, G. B. Hoflund, J.Catal. 172 (1997) 13-23.
[54] Y. Meng, T. Wang, S. Chen, Y. Zhao, X. Ma, J. Gong, Appl. Catal. B-Environ. 160–161
(2014) 161-172.
[55] D. Gingasu, I. Mindru, D. C. Culita, L. Patron, J. M. Calderon-Moreno, Preda, S. O. Oprea,
P. Osiceanu, E. Morena Pineda, Mater. Res. Bull. 49 (2014) 151-159.
[56] S. A. Hosseini, M. C. Alvarez-Galvan, J. L. G. Fierro, A. Niaei, D. Salari, Ceram. Int. 39
(2013) 9253-9261.
28

[57] Z. Zhang, Q. Zhang, L. Jia, W. Wang, T. Zhang, Y. Han, N. Tsubaki, Y. Tan, Catal. Sci.
Technol. 6 (2016) 2975-2983.
[58] N. Liu, Z. Yuan, C. Wang, S. Wang, C. Zhang, S. Wang, Fuel Process. Technol. 89 (2008)
574-581.
[59] H. Zhao, P. Jiang, Y. Dong, M. Huang, B. Liu, New J. Chem. 38 (2014) 4541-4548.
[60] D. M. Minahan, W. S. Epling, G. B. Hoflund, J. Catal. 179 (1998) 241-257.
[61] R. M. Agny, C. G. Takoudis, Ind. Eng. Chem. Res. 24 (1985) 50-55;
[62] J. Saussey, J. C. Lavalley, J. Mol. Catal. 50 (1989) 343-353.
[63] S. Fujita, M. Usui, H. Ito, N. Takezawa, J. Catal. 157 (1995) 403-413
[64] K. C. Waugh, D. A. Whan, J. Chem. Soc.-Faraday Trans. 83 (1987) 2193-2212.
[65] K. J. Smith, R. B. Anderson, Can. J. Chem. Eng. 61 (1983) 40-45.
[66] K. J. Smith,; R. B. Anderson, J. Catal. 85 (1984) 428-436.
ZnCr
O
ZnO
2
4
ZnCr
O
ZnO
a
2
4
b
Zn Cr -400
3
1
Zn Cr -400
1.5
1
Zn Cr -400
Zn Cr -400~4.0
1
1
1
1
Zn Cr -400
0.5
1
Zn Cr -600
0.5
1
Zn Cr -400~3.0
1
1
Zn Cr -800
0.5
1
Zn Cr -400~2.0
1
1
ZnO
Zn Cr -400~1.0
1
1
10
20
30
40
50
60
70
80
90
10
20
30
40
50
60
70
80


2
2
Fig. 1 XRD patterns of the ZnO, ZnxCr1-y and Zn1Cr1-400~z catalysts
29