Role of Ga3+promoter in the direct synthesis of iso-butanolviasyngas over a K-ZnO/ZnCr2O4catalyst

Article abstract of DOI:10.1039/d0cy01688h

The direct synthesis of iso-butanol is an important reaction in syngas (composed of CO and H₂) conversion. K-ZnO/ZnCr₂O₄(K-ZnCr) is a commonly used catalyst. Here, Ga³⁺is used as an effective promoter to boost the efficiency of the catalyst and retard the production of CO₂. X-ray diffraction, X-ray photoelectron spectroscopy, ultraviolet-visible diffuse reflection spectroscopy and electron microscopy were used to characterize the structural variations with different amounts of Ga³⁺, the results showed that the particle size of the catalyst decreases with the addition of Ga³⁺. The temperature-programmed desorption of NH₃and CO₂, and diffuse reflectance infrared Fourier-transform spectroscopy (DRIFTs) analysis of the CO adsorption revealed that the acidity and basicity were altered owing to the different forms of Ga³⁺adoption. X-ray photoelectron spectroscopy and density functional theory (DFT) calculations revealed that the formation of Ga clusters that are coordinated on the exposed surfaces of ZnCr₂O₄, and undergo a tetra-coordinated Ga³⁺exchange with one of the Zn in ZnCr₂O₄(ZG) and ZnGa₂O₄, probably depends on the amount of Ga added. The structural evolution of the Ga³⁺promoted K-ZnO/ZnCr₂O₄catalysts can be described as follows: (i) the main forms are ZG and Ga coordinated ZnCr₂O₄, in which the amount of Ga³⁺is below 1.10 wt%; and (ii) the Ga³⁺containing compound is gradually changed from ZG to ZnGa₂O₄and the amount of gallium clusters increased when the amount of Ga³⁺was higher than 1.10 wt%. The catalytic performance evaluation results show that K-Ga_1.10ZnCr exhibits the highest space time yield and selectivity of alcohols, in which the three compounds play different roles in syngas conversion: ZG is the main active site that boosts the efficiency of the catalysts, owing to the intensified CO adsorption and decreased activation energy of CHO formation through CO hydrogenation; ZnGa₂O₄only modifies the surface basicity and acidity on the catalyst, thereby impacting the carbon chain growth after the CO is adsorbed. The effects of Ga coordinated with ZnCr₂O₄shows little impact on the CO adsorption owing to the weak electron donating effects of Ga.

Full text of DOI:10.1039/d0cy01688h

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
PAPER
Role of Ga3+ promoter in the direct synthesis of
iso-butanol via syngas over a K–ZnO/ZnCr2O4
catalyst†
Cite this: Catal. Sci. Technol., 2021,
11, 1077
ac
Tao Zhang,a Chunyang Zeng,b Yingquan Wu,a Nana Gong,
Jiaqian Yang,ac
a
Guohui Yang,a Noritatsu Tsubaki
d and Yisheng Tan
*
*
The direct synthesis of iso-butanol is an important reaction in syngas (composed of CO and H2)
conversion. K–ZnO/ZnCr2O4 (K–ZnCr) is a commonly used catalyst. Here, Ga3+ is used as an effective
promoter to boost the efficiency of the catalyst and retard the production of CO2. X-ray diffraction, X-ray
photoelectron spectroscopy, ultraviolet-visible diffuse reflection spectroscopy and electron microscopy
were used to characterize the structural variations with different amounts of Ga3+, the results showed that
the particle size of the catalyst decreases with the addition of Ga3+. The temperature-programmed
desorption of NH3 and CO2, and diffuse reflectance infrared Fourier-transform spectroscopy (DRIFTs)
analysis of the CO adsorption revealed that the acidity and basicity were altered owing to the different
forms of Ga3+ adoption. X-ray photoelectron spectroscopy and density functional theory (DFT) calculations
revealed that the formation of Ga clusters that are coordinated on the exposed surfaces of ZnCr2O4, and
undergo a tetra-coordinated Ga3+ exchange with one of the Zn in ZnCr2O4 (ZG) and ZnGa2O4, probably
depends on the amount of Ga added. The structural evolution of the Ga3+ promoted K–ZnO/ZnCr2O4
catalysts can be described as follows: (i) the main forms are ZG and Ga coordinated ZnCr2O4, in which the
amount of Ga3+ is below 1.10 wt%; and (ii) the Ga3+ containing compound is gradually changed from ZG
to ZnGa2O4 and the amount of gallium clusters increased when the amount of Ga3+ was higher than 1.10
wt%. The catalytic performance evaluation results show that K–Ga1.10ZnCr exhibits the highest space time
yield and selectivity of alcohols, in which the three compounds play different roles in syngas conversion:
ZG is the main active site that boosts the efficiency of the catalysts, owing to the intensified CO adsorption
and decreased activation energy of CHO formation through CO hydrogenation; ZnGa2O4 only modifies
the surface basicity and acidity on the catalyst, thereby impacting the carbon chain growth after the CO is
adsorbed. The effects of Ga coordinated with ZnCr2O4 shows little impact on the CO adsorption owing to
the weak electron donating effects of Ga.
Received 28th August 2020,
Accepted 26th November 2020
DOI: 10.1039/d0cy01688h
rsc.li/catalysis
important chemical among all of the downstream products
of syngas and can be used as a gasoline additive to improve
Introduction
The increasing demand for clean fuel has ignited research
on converting carbon containing substances to fuel or
valuable chemicals.1 Syngas, mainly composed of H2 and
CO, is usually obtained through the gasification of coal or
biomass or is produced by methane and CO2 reforming, and
is an important platform material for producing gasoline,
diesel oil and many valuable chemicals.2 Iso-butanol is an
the octane number.3,4 In addition, iso-butanol can also be
used to purify rare earth or produce important chemicals.5
However, until now, the main production route of iso-
butanol is the carbonylation of propylene and fermentation
of sugar.4,5 Hence, it is urgent to produce iso-butanol using
an alternative approach.
Direct synthesis of iso-butanol through syngas has been
proven to be an effective method owing to the wide usage of
carbon containing substances and the high selectivity of iso-
butanol.4,6,7 Among all of the catalysts, the K–ZnO/ZnCr2O4
(K–ZnCr) catalyst is considered to be a good candidate owing
to its longer life time and reduced CO2 emissions.4,8,9 To
determine the reaction mechanism of the K–ZnCr catalyst,
researchers performed a lot of work to disclose the structure–
property relationships. For instance, Tian et al. proposed that
a State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese
Academy of Science, Taiyuan, Shanxi 030001, China. E-mail: tan@sxicc.ac.cn
b China Petroleum Chemical Industry Federation, Beijing 100723, China
c University of Chinese Academy of Science, Beijing 100049, China
d Department of Applied Chemistry, School of Engineering, University of Toyama,
Gofuku 3190, Toyama, 930-8555, Japan. E-mail: tsubaki@eng.u-toyama.ac.jp
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0cy01688h
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 1077–1088 | 1077

View Article Online
Paper
Catalysis Science & Technology
the non-stoichiometry ZnCr spinel is the active site that
produces iso-butanol, and this so-called non-stoichiometry
structure is constructed by the disorder of Zn2+ and Cr3+ in a
spinel crystal cell.8,10,11 Gao et al. enhanced the performance
of the catalyst by modulating the configuration and the
amount of ZnO.12–14 In addition to the studies on the
structure activity performance of the ZnCr catalyst, other
researchers also attempted to explore the reaction
mechanism of alcohol formation. For example, K. J. Smith
et al. revealed that the selectivity of methanol and iso-butanol
is controlled by the activity of an α- or β-carbon atom of the
growing alcohol.15 Lietti et al. investigated the complex
reaction network of higher alcohol synthesis by studying the
temperature-programmed reaction of C3, C4, and C5
oxygenates over the K-promoted ZnCr catalyst.16 Inspired by
this research, Wu et al. explored the rate-determining step of
iso-butanol formation and found that the formation of C2
oxygenates through CO and formyl (CHO) is the key step to
controlling the reaction atmosphere and enriching the
intermediate alcohols.5,17
Based on the reaction mechanism, some researchers have
made numerous efforts to further promote the efficiency of
the ZnCr catalyst by adding promoters. For example, Epling
et al. added Pd and Mn to boost the selectivity of iso-butanol
and alcohols under the reaction conditions of 10 MPa and
430 °C on a Cs-promoted ZnCr catalyst.18–21 However, until
now, a systematic study concentrated on the influence of the
third composition on the structure of the ZnCr iso-butanol
catalyst has not been reported. Moreover, among all these
investigations, the amount of CO2 in the tail gas is relatively
high. Therefore, reducing the amount of CO2 emissions
during iso-butanol synthesis is also an important issue.
In recent years, Ga3+ has been used as a promoter to
modulate the electron properties, textural properties,
configurations and distribution of the active site composition
of Co, Co/AC, CuAl, and CuZnAl catalysts to further modify
the activation of CO during the synthesis of alcohols.6,22–25 In
addition, Ga3+ can also be used to modulate the efficiency of
the water–gas-shift reaction or to improve the performance of
CO2 conversion by altering the acid–base properties of the
catalyst.26–28 However, when Ga was introduced into the Zn
containing catalyst, the existing form of Ga and its effect on
the performance of the catalyst was still conflicting. For
example, Meng-Jung Li et al. proposed that the formation of
ZnGa2O4 creates an electronic heterojunction with excess
ZnO to facilitate the reduction of Zn2+ to Zn in the CuZnO
methanol catalyst.27 Ham et al. confirmed that the phase
transformation of gallium oxide from tetrahedral Ga–O sites
to octahedral Ga–O sites suppressed the formation of
dimethyl ether (DME).29 At present, whether the catalytic
performance of K–ZnCr is promoted by the introduction of
Ga3+ is not clear based on our knowledge. In addition, the
form of Ga3+, its impact on the structure of the catalyst and
the explicit role it plays during the reaction are still
unknown. To unravel the above-mentioned problems, Ga3+
was added into a K–ZnCr iso-butanol catalyst to modulate the
surface structure through the formation of novel compounds
to further improve the performance of the catalyst.
Experimental section
2.1 Catalyst preparation
Ga3+ promoted K–ZnCr catalysts were prepared by
precipitation–impregnation. The precursors for Zn, Cr and
Ga are zinc nitrate hexahydrate, chromium nitrate nine-
hydrate and gallium nitrate hydrate, respectively. In a typical
procedure, aqueous solutions of Zn(NO3)2·6H2O and
Cr(NO3)3·9H2O (molar ratio of Zn : Cr = 1 : 1) were precipitated
with (NH4)2CO3 aqueous solution. The detailed procedure is
described in our previous report.30 The precipitate was dried
at 100 °C for 12 h, then milled into
a powder and
impregnated with the aqueous solution of gallium nitrate
hydrate, then dried at 100 °C for 12 h and calcined at 400 °C
in a muffle furnace under a steady atmosphere for 4 h to
obtain the GaZnCr samples. Finally, the GaZnCr samples
were impregnated with the aqueous solution of K2CO3, then
dried and calcined according to the process of preparing
GaZnCr to obtain the final catalysts, the samples can be
abbreviated as K–GaZnCr. The amount of Ga (mole fraction)
is changed from 0, 0.55 1.10, 1.64, to 2.18 wt%, respectively.
The final catalysts were denoted as K–GaxZnCr, in which x
represents the loading amount of Ga.
2.2 Catalyst evaluation for iso-butanol synthesis
The performance of the catalysts was evaluated in a fixed-bed
tubular reactor, composed of a titanium alloy tube (Φ10 × 2 ×
400 mm) inserted into a furnace. 1.0 g of the catalyst with a
particle size of 50 to 60 mesh was packed, and the two ends
were stuffed with quartz sand to support the catalyst. Before
introducing syngas, the catalyst was reduced using 10% H2 in
N2 with a flow rate of 100 mL h−1 at 400 °C for 6 h. After
reduction, syngas with a molar ratio of H2/CO = 2.4 was
introduced into the reactor. The reaction conditions were as
follows: pressure 10.0 MPa, gas hourly space velocity (GHSV)
3300 h−1 and a reaction temperature of 400 °C. The detailed
procedures for sample collection, the composition analysis of
syngas, tail gas and the liquid phase are described in our
previous report.11,30
2.3 Characterization
The specific surface face areas were obtained according to
the Brunauer–Emmett–Teller (BET) method using nitrogen
adsorption and desorption isotherms at −196 °C on ASAP
2020 V4.03 equipment. Before the test, each sample was
degassed at 200 °C for
4 h. The morphology and
microstructure of the catalysts were investigated using high
resolution transmission electron microscopy (HRTEM, JEM-
2100F) operated at 200 kV (produced by Japan Electronics
Co., Ltd). The X-ray diffraction (XRD) data for the K–GaxZnCr
catalysts were recorded using
a
D8 Advance X-ray
diffractometer in the 2θ range from 10° to 80° with Cu Kα
1078 | Catal. Sci. Technol., 2021, 11, 1077–1088
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
radiation. To obtain a spectrum with high quality, the Results and discussion
scanning step size and speed were set to 0.01 and 1° min−1,
respectively. The X-ray photoelectron spectroscopy (XPS)
3.1 BET surface areas of the catalysts
The textural parameters of the reduced K–GaxZnCr catalysts
are listed in Table 1. As shown, the BET areas of the samples
regularly changed with the addition of Ga3+, that is 63.86
m2 g−1 for K–Ga0ZnCr, 78.43 m2 g−1 for K–Ga0.55ZnCr, 75.96
m2 g−1 for K–Ga1.10ZnCr, 66.54 m2 g−1 for K–Ga1.64ZnCr and
61.71 m2 g−1 for K–Ga2.18ZnCr, respectively. This
phenomenon is also observed when investigating the external
surface area. However, as for the change in the average pore
diameter and its distributions (Fig. S1†), one can see that the
smaller pores were created by Ga3+ addition, which is in
accordance with the studies performed by Ham et al. and
Kim et al. who discovered an increased surface area and
decreased pore diameter when Ga was added to the Cu/m-
Al2O3 catalyst or CuO–Cr2O3 water gas shift reaction
catalyst.26,29 With the increase in the BET area, more active
sites were created. Therefore a superior performance will be
expected from the catalyst. The change in the catalyst texture
may be caused by the interaction between the Ga species and
ZnCr oxides which alters the morphology of the catalyst. The
exact images of the Ga3+ doped catalyst are further
characterized using HRTEM in the following section.
patterns were recorded using an AXIS ULTRA DLD X-ray
photoelectron spectrometer equipped with a multichannel
detector. Charge referencing was performed against
adventitious carbon (C 1s, 284.8 eV).
A
Shirley-type
background was subtracted from the signals. The recorded
spectra were fitted using Gaussian–Lorentzian curves to
determine the composition of the surface of the different
samples. CO2 and NH3 temperature programmed desorption
profiles were recorded on a TP-5050 automatic chemical
adsorption instrument.
The test temperature is in the range from 50 to 800 °C.
The catalyst samples were reduced using a mixture of 10%
H2 and 90% N2. The reduction was performed at 400 °C for
2.5 h with a heating rate of 10 °C min−1. After reduction, the
samples were cooled to 50 °C under the protection of 10% H2
and 90% N2. Then, the catalysts were treated in a constant
flow of N2 with a rate of 30 mL min−1 to eliminate the
physically adsorbed water and contaminates. After that, NH3
or CO2 was introduced into the system for 0.5 h to achieve
saturated adsorption. The samples were purged with N2 for
about 1 h to remove the physically absorbed NH3 or CO2.
Finally, the temperature was increased at a rate of 10 °C
min−1 under an N2 atmosphere. The signals of CO2 and NH3
were recorded by a thermal conductivity detector respectively.
X-ray fluorescence (XRF) spectra were recorded using an E3
Tiger XRF spectrometer with a silicon drift detector (SDD).
Raman spectra and solid-state UV-visible spectra were also
recorded to reveal the influence of the Ga species on the K–
GaxZnCr catalyst structure. A Raman test was performed
using a HORIBA 800 with a 532 nm excitation wavelength.
Solid state UV-visible spectrums were recorded in a SU3900
spectrophotometer, the scanning range is from 200 to 800
nm by using barium sulfate as the base.
3.2 HRTEM characterization
The HRTEM images of the K–ZnCr catalyst with and without
Ga modification are shown in Fig. S2 and S3.† When
comparing the photos of the two samples, one can see that
after adding Ga3+ to the K–ZnCr catalyst, the particles
become smaller. As shown in the histogram, the particle size
of the K–Ga0ZnCr sample is mostly in the range of 5.5 to 6.5
nm, while for the K–Ga1.10ZnCr sample, the particle is in the
range of 4.5 to 6.5 nm with a uniform distribution. The
average particle size decreased from 6.98 nm to 6.11 nm after
Ga3+ introduction, implying that the microstructure of the K–
ZnCr catalyst was altered by Ga3+ addition, which is also
observed for the Ga doped CoZnAl catalyst.22 When
considering the exposed surface of the K–ZnCr catalyst with
or without Ga addition, one can see that the (311), (310) and
(202) surfaces of ZnCr2O4 are the most exposed surfaces after
Ga modification. The alteration of the particle sizes and
distribution of the crystal surface caused by Ga addition may
be attributed to the interaction between Ga3+ and ZnCr
oxides or the interactions between the Ga cluster and ZnCr
For the above analysis, BET, XPS, Raman spectra and solid
state ultraviolet-visible diffuse reflectance spectrums were
performed on the reduced catalyst samples. The composition
of the reduction gas and the procedure used was identical to
the procedure used before catalyst evaluation.
In addition, the FT-IR spectra of CO adsorption were
obtained on a TENSOR-27 in the range of 4000 to 600 cm−1
with a 4 cm−1 resolution, the detailed procedures for CO
adsorption were described in our earlier reports.30
Table 1 BET areas, pore volume and average pore diameter of the K–GaxZnCr catalysts (x = 0, 1.5, 3.0, 4.5 and 6.0)
Catalyst
BET areaa (m2 g−1
)
External surface areab (m2 g−1
)
Pore volumec (cm3 g−1
)
Average pore diameterd (nm)
K–Ga0ZnCr
63.86
78.43
75.96
66.08
61.71
62.89
76.67
71.90
65.77
57.42
0.2197
0.2490
0.2388
0.1924
0.2052
12.08
11.48
11.35
10.46
12.75
K–Ga0.55ZnCr
K–Ga1.10ZnCr
K–Ga1.64ZnCr
K–Ga2.18ZnCr
a
b
c
BET area was calculated by applying the multi-point BET equation in the linear range. t-Plot external surface area. Single point adsorption
total pore volume at P/P0 = 0.99. d Desorption average pore diameter (4 V/A by BET).
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 1077–1088 | 1079

View Article Online
Paper
Catalysis Science & Technology
oxides. The explicit form of the Ga3+ doped species was
verified using XPS analysis and density functional theory
(DFT) calculations in the following section. In addition,
although the change in the particle sizes is in accordance
with the observation of the BET test, the TEM images only
gives part of the morphologies of the catalyst samples, the
exact particle sizes of the K–GaxZnCr samples should be
analysed by characterizing the bulk structure of the samples,
which will be investigated by analysing the XRD patterns in
the subsequent section.
the weight fraction of ZnCr2O4 increases with the increasing
amount of Ga3+, while this trend is opposite for ZnO, indicating
that the composition of the K–ZnCr catalyst was altered by the
introduction of Ga3+. Furthermore, after 48 h of time-on-
stream, it can be observed from the corresponding catalysts
that the diffraction peaks of both ZnO and ZnCr2O4 were
intensified (Fig. 1B), implying the growth of the particle size of
these two species during the reaction. Gao et al. illustrated that
the configurations and absorption energy of CO can be altered
by the morphology of ZnO owing to its different interactions
with the ZnCr2O4 (111) surface.12 Based on this discovery, we
can deduce that the interaction between ZnO and ZnCr2O4 can
be altered by the addition of Ga3+. This may be caused by the
interaction between Ga and ZnCr2O4, or the formation of other
compounds such as ZnGa2O4 through the respective oxides of
gallium, chromium and zinc. In the following section, a
thermodynamic calculation was performed to verify the
possible transformation process of the Ga species.
3.3 XRD characterization
The XRD patterns of the catalyst samples with different
amounts of added Ga3+ are shown in Fig. 1. As shown, all the
K–GaxZnCr catalysts presented diffraction peaks that were
assigned to ZnO and ZnCr2O4, and no new peaks were observed
when Ga3+ was introduced, implying the high dispersion of the
Ga3+ compounds. The particle size of the (310) surfaces of
ZnCr2O4 were calculated using the Scherrer equation and are
listed in Fig. 1A. As can be seen, the ZnCr2O4 particles become
much smaller when Ga3+ is introduced. This trend is in
accordance with the observations from the HRTEM analysis. In
addition, to give a better understanding of the impact of Ga3+
on ZnO and ZnCr2O4, a Rietveld refinement was performed
using MUAD software.31–34 The results of the refinement are
shown in Fig. S4.† The mass fractions of ZnO and ZnCr2O4
according to the refinement are listed in Table S1.† As shown,
3.4 Possible locations of Ga3+ calculated using DFT
To deduce the possible locations of Ga3+ in the K–ZnCr iso-
butanol catalyst,
a
thermodynamic calculation was
performed. When considering interactions between the Ga3+
and ZnCr oxides, the following compounds can be formed: (i)
gallium oxide; (ii) Ga3+ doped ZnO (ZGO); (iii) one of the
metals in ZnCr2O4 is replaced by Ga3+ to form a four
coordination or six coordination spinel; and (iv) ZnGa2O4.
DFT calculations were performed to deduce the possible
locations of Ga3 (the detailed procedures are listed in the
ESI†). If the Zn2+ located at the tetrahedral coordination sites
in ZnO and ZnCr2O4 was replaced by Ga3+, the reduced cell
parameters of a and c would be expected owing to the smaller
ionic radius (0.47 Å for Ga3+ vs. 0.6 Å for Zn2+). A similar
deduction can also be expected if the Cr3+ located at the six
coordinated sites of ZnCr2O4 was replaced by Zn2+(0.74 Å for
Cr3+ vs. 0.62 Å for Ga3+). However, an adverse consequence
was obtained here (Table S1†). The differences in the cell
parameter, c, of ZnCr2O4 between the speculated result and
the results of Rietveld may originate from the cation
disordered ZnCr spinel. To further deduce the location of
Ga3+, the binding energy EB and formation energy Ef were
calculated based on the optimized structure using a DFT
calculation. The configuration of the possible Ga3+ doped
species and the reaction are shown in Table S2 (ESI†). As
shown, the EB of these compounds follows the order below:
ZG > ZnCr2O4 > ZnGa2O4 > CG > ZGO > ZnO, implying that
ZG is the most stable Ga3+ doped ZnCr compound. When
Ga3+ was added into a ZnCr catalyst, ZG may be one of the
most likely to exist forms of the Ga species, rather than ZGO
and CG, owing to its higher stability.
In addition to the doping of Ga3+ in ZnCr2O4, other
structures may be formed. For example, the unsaturated sites
of the ZnCr2O4 surfaces were coordinated by Ga3+ or clusters
of Ga species. To verify this speculation, the adsorption of Ga
on the most exposed surfaces of ZnCr2O4, such as (311), (310)
Fig. 1 XRD patterns of the (A) fresh and (B) spent K–GaxZnCr catalysts.
1080 | Catal. Sci. Technol., 2021, 11, 1077–1088
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
and (202) (obtained through HRTEM) were calculated. Based
on the observation and speculation, the interaction between
Ga and ZnCr2O4 was investigated by calculating the
adsorption of Ga on the coordination unsaturated sites of the
surfaces. The most stable adsorption configurations are listed
in Fig. S5.† Bader charge distributions before and after Ga
adsorption are also listed in Table S3.† As shown, all the
surfaces accept electrons from Ga, the trend is 310 (0.2940) <
CrO stretching vibration in the chromate compound.36 The
peak located at 350 nm belongs to the vibration mode
associated with the multiple-phonon scattering processes of
ZnO.38 Based on the observation, one can see that the
vibrations of ZnO, ZnCr2O4 and chromate were all weakened
after the introduction of Ga3+. Furthermore, for all the
samples, the peaks located at 770 and 500 nm, corresponding
to the asymmetric stretching of the Ga–O bond in the
tetrahedral sites, shows almost the same intensity with an
enhanced amount of Ga3+.39 However, the typical Raman
vibration mode that is ascribed to β-Ga2O3 was not
observed.39 Moreover, the peaks at 601 and 712 nm that were
assigned to the T2g and A1g modes of ZnGa2O4 were
intensified by increasing the amount of Ga3+.40 This may be
the consequence of the interactions between the ZnCr oxides
311 (0.4092)
electrons, a stronger adsorption of CO will be created if the
surface is electron deficient. Therefore, reverse CO
< 202 (0.7449). Generally, CO is rich in
a
adsorption could be expected. The calculated adsorption
energies and configurations of CO (Table S4 and Fig. S19†)
support this conjecture.
Furthermore, except for the above observed surfaces, there
may be lots of possible step surfaces owing to the fact that
the surfaces of the particles are very uneven, which could be
attributed to the other reaction sites for CO hydrogenation.
To better analyse the structure and their effects on syngas
with Ga. With the gradual increase in the amount of Ga3+
,
the extra ZnO was reacted with Ga2O3 to form ZnGa2O4, Zn
located in the tetrahedral site of ZnCr2O4 was partly replaced
by Ga to form a ZG structure which exhausted both ZnO and
conversion,
more
detailed
characterization
and
ZnCr2O4. In addition, by increasing the amount of Ga3+
,
computational work should be carried out. Owing to the
complexity of the catalyst surface, these configurations are
not easy to investigate using DFT calculations. Therefore the
DFT calculation shown above gives only part of the functions
of the particular structure. However, this approach could be
used as a simplified method to study the real surface. In
future studies, we will attempt to investigate the structure
and function of these step surfaces.
chromate was consumed so the intense absorption of
chromate vanished. Based on the above mentioned analysis,
we may speculate that when a small amount of Ga3+ is
introduced into the K–ZnCr catalyst, ZG is the dominant
structure. However, upon further increasing the amount of
Ga3+, the ZnO will react with Ga2O3 to form ZnGa2O4. The
surface composition and properties were altered by the
different forms of Ga3+ species, which need to be confirmed
by XPS and chemisorption characterization.
3.5 Raman analysis of the K–GaxZnCr catalysts
3.6 Ultraviolet-visible diffuse reflectance spectra
Raman peaks for the reduced K–GaxZnCr samples with an
excitation wavelength of 532 nm are shown in Fig. 2. The
peaks located at 392 and 900 nm are the characteristic bands
for the normal vibration modes of the ZnO and Cr atoms in
the tetrahedral and octahedral sites formed by the oxygen
atoms.35 The peaks located at 511 and 605 nm represent the
vibration of F2g for ZnCr2O4 and 680 nm is assigned to the
symmetric Cr–O stretching vibration of the A1g symmetry
originating from the CrO6 groups of the spinel.36,37 The
bands at around 846, 867 and 933 nm are ascribed to the
An interesting phenomenon for the reduced catalyst samples is
the change in the colour of the catalyst, for example K–Ga0ZnCr
is a blue-green, upon gradually increasing the amount of Ga3+,
the colour becomes increasingly darker. This change was
verified by ultraviolet visible diffuse reflectance analysis. As
shown in Fig. 3, all the samples absorbed in the range of 200–
800 nm, the absorbance of the samples was intensified by
increasing the amount of Ga3+. In addition, the diffuse
Fig.
3 Ultraviolet-visible diffuse reflectance spectrums of the K–
Fig. 2 Raman spectra of the K–GaZnCr catalysts.
GaxZnCr catalysts.
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 1077–1088 | 1081

View Article Online
Paper
Catalysis Science & Technology
reflectance spectrum gives evidence of the charge transfer of
The BE of Zn 2p was barely changed with the altered
concentration of Ga3+, indicating the similar chemical
environment of Zn with and without the introduction of the
Ga3+ promoter. To further explore the influence of Ga3+ addition
on the composition of the K–GaZnCr catalyst, XPS analysis of
Ga 2p and Ga 3d were performed, and are shown in Fig. 4. The
O2− → Zn2+ in ZnO at 370 nm and in ZnCr2O4 at 300 nm.41 The
absorption bands at 370 and 260 nm, attributed to the O2−
→
Cr6+ charge transfer, were also observed.41,42 Upon gradually
increasing the amount of Ga3+, the absorbance located at 300
nm become overlapped by the emergence of new peaks that
were located at 287 and 353 nm, indicating the formation of a
new compound, which may be caused by the growth of ZG or
Ga3+ coordinated on the ZnCr2O4 surfaces.
BE at 1116.23–1117.27 eV is the characteristic peak of Ga 2p3/2
,
49
and 1443.06–1143.20 eV is the characteristic peak of Ga 2p1/2
.
These values are much smaller than the peaks in K–ZnGa2O4
and Ga2O3 (Fig. S9†), indicating the chemical environment is
quite different from pure Ga2O3 and K–ZnGa2O4. In addition, it
is obvious that the BE of Ga 2p1/2 increased from 1143.06 (K–
Ga1.10ZnCr) to 1143.20 eV (K–Ga2.18ZnCr) upon an enhancement
of the amount of Ga3+, suggesting the chemical environment of
Ga3+ was altered by the amount of its usage. To further study
the coordination states of Ga3+ on the K–GaZnCr catalyst, the
Ga 2p3/2 peaks were deconvoluted into two peaks, the results
are shown in Fig. 4D. It can be seen that the Ga3+ ions occupy
two different coordination sites in the K–GaZnCr: peaks located
at 1116–1117 eV, which can be assigned to Ga3+ at the
tetrahedral sites,50–52 and the peak at 1117.1–1117.6 eV can be
assigned to the Ga3+ located at the octahedral sites. In addition,
for K–Ga1.10ZnCr, the Ga3+ located at the tetrahedral sites
reaches the maximum value, 74.31%. Another interesting
finding is the assignment of Ga 2p1/2 shown at 1143.06–1143.20
eV. According to Delichere et al. the Ga 2p1/2 for ZnGa2O4 is
located at 1144.3 eV, while for Ga2O3 it is 1144.15 eV.39,48,50,53
The decreased BE of Ga 2p1/2 may be caused by the formation
of the ZG structure or the coordination of the Ga cluster on the
surface of ZnCr2O4, because the electronegativity of Zn and Cr
are smaller than that of Ga.54
Moreover, the indirect band gap energy Eg for the K–Gax-
ZnCr samples was calculated using the equation (αhν)2
=
C(hν − Eg) to investigate the influence of Ga addition on the
electron structure of the catalysts.38,43 In the equation, α is
the absorption coefficient, hν is the photon energy with
frequency ν, Eg is the direct band gap energy, and C is a
constant. Fig. S6† shows the plot of (αhν)2 versus hν. The
calculated direct band gap energies for the samples are 2.51
eV for K–Ga0ZnCr, 2.49 eV for K–Ga0.55ZnCr, 1.75 eV for K–
Ga1.10ZnCr, 2.07 eV for K–Ga1.68ZnCr, and 2.03 eV for K–
Ga2.18ZnCr respectively, which indicates that when Ga was
introduced into the K–ZnCr catalyst, the band gap was
reduced. Tajizadegan et al. observed an Eg value of 1.8 eV for
a nanoparticle of ZnCr2O4 with a particle size of about 17
nm.44 Huang et al. discovered that the Eg of TiO2 is size
dependent, the band gap decreased from 3.239 to 3.173 eV
when the particle size decreased from 29 to 17 nm and then
increased from 3.173 to 3.289 eV as the particle decreased
from 17 to 3.8 nm.45 Based on the phenomenon and
published research, we deduce that the lower Eg after Ga
addition may be caused by the smaller particle size of
ZnCr2O4 observed using TEM and XRD. In addition, the
coordination of Ga on the exposed oxygen of the ZnCr2O4
surfaces or the formation of ZG may be another reason that
caused the reduced Eg owing to the fact that Ga can donate
electrons to ZnCr2O4.
From the above analysis, no remarkable peaks
corresponding to Ga2O3 were observed owing to the
mismatched peaks of the Ga 2p3/2 and Ga 2p1/2 spectra for
the Ga2O3 samples. To further confirm the existing form of
Ga3+, the 3d spectra of Ga was performed and the results are
shown in Fig. 4E–H. According to Yang et al.,55 Ga is mainly
located at the tetrahedral (BE around 21.62) and octahedral
sites (BE around 19.93 eV) for all the catalyst samples. The
amount of six-coordinated Ga decreased gradually with the
3.7 Analysis of the surface composition of K–GaxZnCr using
XPS spectra
The XPS spectra were investigated to disclose the structure
evolution process of the K–ZnCr catalyst promoted by Ga3+. The
Cr 2p3/2 and Cr 2p1/2 spectra of all the samples are listed in Fig.
S7.† Four peaks were obtained for each of the samples after
deconvoluting the recorded peaks. The binding energy (BE) of
575.75–574.86 eV is the characteristic peak for Cr metal,46 more
specifically, 585.27–579.08 eV and 575.88–576.00 eV are the
characteristic peaks of Cr3+ (ref. 8 and 11) and 578.49–579.08 eV
is the peak of Cr6+.47 Analysis of the concentration of surface
chromium showed that the main composition of surface
chromium is Cr3+, the amount of metallic Cr and Cr6+ species
are only changed in the range of 5.32–12.27% and 5.71–21.21%,
respectively (Table S5†), implying that the addition of Ga3+ has
an obvious impact on the valence state of Cr. In addition, XPS
spectra of Zn 2p were also measured. As is shown in Fig. S8,†
the BE values of 1020.46–1020.60 eV and 1043.57–1043.68 eV
are characteristic of Zn 2p1/2 and Zn 2p 3/2, respectively.37,48,49
increasing amount of Ga added (from 32.52% for K–Ga0.55
-
ZnCr to 22.71% for K–Ga2.18ZnCr, Table S6†), and the
amount of four-coordinated Ga shows a reversible trend,
indicating that different forms of gallium species were
formed with different amounts of added Ga3+. This trend is
in accordance with the observation of Ga 2p. In addition to
the peaks that denote to Ga located at different coordination
sites, peaks located at 16.94 to 17.06 eV and 23.37 to 23.85 eV
were also observed. The former may be caused by the
formation of ZG, while the latter may be the result of gallium
clusters or the coordination of Ga to the unsaturated ZnCr2O4
surfaces.
By fitting the XPS spectrum of Cr 2p, Zn 2p and Ga 2p, the
surface concentration of these elements were calculated. As a
comparison, the compositions of bulk catalysts were also
investigated by XRF analysis. The results are shown in
1082 | Catal. Sci. Technol., 2021, 11, 1077–1088
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
Fig. 4 Ga 2p (A)–(D) and 3d (E)–(H) of the K–GaxZnCr samples (x = 0.55, 1.10, 1.64 and 2.18).
Table 2, the molar ratios of Zn/Cr in the bulk for each
catalyst are in the range of 0.89 to 0.91, which is smaller than
the feed rate. However, on the surface of the catalysts, a
remarkable enhancement of Zn was observed. Furthermore,
when comparing the ratios of Zn/Ga and Cr/Ga for the bulk
and surface, an enhancement of Ga was observed. The
enhancement of Zn and Ga on the surface of the catalyst may
be ascribed to the interaction of Ga3+ with ZnCr2O4 as
discussed in the above section.
To further investigate the structural evolution caused by
Ga3+ addition, especially the surface oxygen species over Ga
promoted catalysts, the O1s XPS spectra of the catalysts were
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 1077–1088 | 1083

View Article Online
Paper
Catalysis Science & Technology
alcohol.14,17,56 If the basicity of the catalyst is too strong, the
adsorption of the unsaturated precursors for branched
carbon chain growth, for example, propanol or/and formyl
group, is so strong that the formation of higher branched
carbon chains can not take place.5,15,16
Table 2 Bulk and surface compositions of Zn, Cr, and Ga for the K–
GaZnCr catalysts
Bulka
Zn/Cr
On the surfaceb
Catalysts
Zn/Ga
Cr/Ga
Zn/Cr
Zn/Ga
Cr/Ga
K–Ga0ZnCr
0.94
0.91
0.91
0.89
0.90
—
—
1.15
1.13
1.12
1.14
1.04
—
—
K–Ga0.55ZnCr
K–Ga1.10ZnCr
K–Ga1.64ZnCr
K–Ga2.18ZnCr
79.82
34.39
22.71
10.19
55.88
23.95
16.12
7.20
28.03
26.13
17.48
14.94
26.90
23.23
15.64
13.11
3.9 NH3-TPD analysis of the catalysts
Fig. S12† shows the NH3-TPD profiles of the K–GaxZnCr
samples. As shown, all the samples give two major peaks
that can be assigned to the weak acidity (around 210 °C)
and medium–strong acidity (300–600 °C). The profiles were
fitted into five peaks to further investigate the detailed
distribution of the acidity. According to the NH3-TPD
profiles, the ratios of weak acidity to medium–strong acidity
increased with the increasing amount of Ga3+ promoter. As
for the medium–strong acidity, K–Ga1.10ZnCr shows an
identical ratio of medium acidity to strong acidity. The
proper ratios of weak to medium–strong acidity and
medium to strong acidity play a major part in activating CO
and boosting the carbon chain growth during synthesis of
higher alcohols.1 In general, the strong acidity is beneficial
to the dissociative absorption of CO, while the weak and
medium acidity contribute to the molecular adsorption of
CO.57 Therefore, the moderate acidity is a key role in syngas
conversion.
a
The bulk composition is obtained by the analysis of the XRF of
b
the catalysts. The surface composition is obtained by fitting the
spectra of XPS.
recorded. As shown in Fig. S10 and Table S7,† the surface
oxygen species are mainly composed of lattice oxygen (Olatt
)
located at 529.5 eV, surface absorbed oxygen (Oads) located at
531.1 eV and absorbed OH groups (OOH) located at 532 eV.8
Oads and OOH created by the disordered ZnCr catalyst are
considered to be the active sites for CO activation and these
are advantageous to the formation of a formate group, which
is an important C1 precursor to form C2 oxygenates.
Therefore, the higher the concentration of Oads and OOH, the
better the performance of the K–ZnCr catalyst. As listed in
Table S7,† the molar ratio of (Oads + OOH)/(Olatt + Oads + OOH
)
ranged from 36.72% to 46.94%. In addition, the proportion
of absorbed OH groups show irregular changes upon
increasing the amount of Ga3+, which may be caused by the
deduced particle size or the altered interactions between ZnO
and ZnCr2O4. Furthermore, the amount of Ga clusters, ZG
and ZnGa2O4 also changed with the amount of Ga added.
This will also influence the basicity and acidity of the
catalysts, which is very important for CO activation and iso-
butanol formation. The detailed basicity-acidity information
about the K–GaZnCr catalyst was investigated using CO2 and
NH3 TPD and is reported in the following section.
3.10 DRIFTs of CO adsorption on K–GaxZnCr catalysts
The adsorption of CO was performed using in situ diffuse
reflectance infrared Fourier transform spectroscopy (DRIFTS)
at 400 °C after H2 reduction of the K–GaxZnCr catalysts. As
shown in Fig. S13,† all the samples give the absorption of
carbonates located at 1045, 1310 and 1560 cm−1 and the
absorption of formates located at 1590 and 1610 cm−1
respectively.30,58,59 Moreover, the anti-symmetric stretching
−
vibration of CO2 and the symmetric stretching vibration of
CH located at 2955, 2865, and 2745 cm−1 were also
observed.59 By comparing the intensity of the formate and
carbonates located at 1590, 1607, 1045 and 1033 cm−1, one
can see that K–Ga0.55ZnCr and K–Ga1.10ZnCr give a much
stronger absorption than the other three samples, indicating
that more active sites exist on these two samples. Formate is
an important precursor of the C1 intermediate CHO for the
growth of the C–C chain, meanwhile it is also the precursor
for forming the OCH3 group which is the key intermediate to
produce methanol.60 As a result, the higher the concentration
of formate, the higher the space time yield (STY) of alcohols
that can be expected.
3.8 CO2-TPD analysis of the basicity of the catalyst samples
In this section, the basicity of the K–GaZnCr iso-butanol
catalysts was tested using a CO2-TPD study, and the results
are shown in Fig. S11.† As can be seen from Fig. S11,† all the
samples show a remarkable desorption of CO2 at 170–204 °C,
260–288 °C, 350–367 °C, 440–484 °C and above 600 °C,
respectively, corresponding to the different basicity of the
catalysts, that is weak basicity, medium basicity, and strong
basicity. The CO2-TPD results demonstrated that the ratios of
weak basicity to strong basicity and medium basicity are in
the range 0.57–0.47 and 0.64–0.76 for K–GaxZnCr (x = 0.0,
0.55, 1.10, and 1.64) and a sharp increase for K–Ga2.18ZnCr
(0.68 and 1.25), indicating that an excessive amount of Ga3+
is harmful to the density of the weak basicity site, but is
advantageous to the density of the medium and strong
basicity sites, especially to the strong basicity sites. Generally,
moderate basicity is important for the carbon chain growth
through β-addition during the synthesis of higher
3.11 Performance of the K–GaZnCr catalysts
The typical catalytic performances of the K–GaxZnCr catalysts
are listed in Table 3. All the evaluation results for the
catalysts were collected after 22 h of time on stream in order
to minimize the influence of the disturbance by frequent
sampling. As can be seen in Table 3, by gradually increasing
the amount of Ga3+, the selectivity of CO2 was decreased from
1084 | Catal. Sci. Technol., 2021, 11, 1077–1088
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
Table 3 Typical catalytic performance of the K–GaxZnCr iso-butanol catalysts
Selectivity, C mol%
Distribution of alcohols, wt%
CO
STY g
Catalysts
conversion, %
h−1 gcat
CH4
CO2
CHx
DMEb
Alcohol
MEOHc
EtOHd
PrOHe
i-BuOHf
C5+OHg
−1
a
K–Ga0.0ZnCr
K–Ga0.55ZnCr
K–Ga1.10ZnCr
K–Ga1.64ZnCr
K–Ga2.18ZnCr
K–ZnGa2O4
23.59
23.75
23.17
21.24
24.16
14.18
0.0695
0.0996
0.1103
0.1101
0.0819
0.0427
0
42.55
38.73
34.64
37.23
34.36
19.53
2.95
6.70
5.46
7.40
19.96
2.91
0
51.50
52.87
59.03
53.72
41.99
17.13
57.98
57.23
59.53
58.94
69.44
87.84
0.93
0.92
1.02
0.87
0.99
1.89
4.38
3.97
4.53
3.96
4.09
0.53
33.13
36.03
31.23
31.58
23.21
5.84
3.57
1.84
3.69
4.64
2.19
1.36
1.69
0.85
1.81
1.26
1.73
0.01
0.01
0.03
2.43
58.70
a
b
c
d
e
Summary of the amount of ethane, propane and butane in the tail gas. Dimethyl ether. Methanol. Ethanol. n-Propanol and
isopropanol. f Iso-butanol. g Amyl alcohol. Reaction conditions: temperature = 400 °C, pressure = 10.0 MPa, GHSV = 3300 h−1
.
42.55% for K–Ga0ZnCr to 34.64% for KGa1.10ZnCr, while the
selectivity of the alcohols was increased from 51.50% to
59.03%. In addition, when the amount of Ga3+ is lower than
1.64%, the selectivity of methanol and iso-butanol are higher
than 57 wt% and 31 wt%, respectively. Upon further
increasing the amount of Ga3+, the isobutanol selectivity in
the total alcohols was reduced to only 23.21 wt%, while
methanol (69.44 wt%) turned out to be the major component
of the alcohols. The amount of CHx and DME were also
increased sharply under a higher doping concentration of
Ga3+. The evaluation results indicate that the higher
concentration of Ga3+ has an adverse effect on the
performance of the iso-butanol catalyst, which may be
attributed to the altered structure of the catalyst. In other
words, the main composition of the catalyst is ZnO and
ZnCr2O4 without adding Ga3+. When Ga3+ was added, ZG or
saturated ZnCr2O4 were formed. With the gradual addition of
Ga3+, the extra ZnO will react with Ga2O3 to form ZnGa2O4
during calcination. Furthermore, the K–ZnGa2O4 sample was
prepared to verify the structure of ZnGa2O4 and its catalytic
performance. The evaluation results show that K–ZnGa2O4 is
probably a DME synthesis catalyst, and this was validated
using a K–Ga2.18ZnCr sample and only 2.43% DME was
obtained, which was much higher than that obtained using
the other four catalysts. Thus, the increased amount of DME
in K–Ga2.18ZnCr indicates the formation of the ZnGa2O4
species. Other possible reasons, such as the formation of Ga
clusters, the interaction of Ga with the coordinately
unsaturated ZnCr2O4 surface will result in the loss of
performance of the K–ZnCr catalyst, which was verified by
the reduced CO adsorption energy. Moreover, the well-
performing K–Ga1.10ZnCr catalyst was evaluated using a long-
period stability test. The results are shown in Table 4. During
102 h of time on stream, the selectivity of the total alcohol
and the fraction of iso-butanol values were found to be above
59% and 31.23 wt% respectively. The STY of the alcohols is
in the range of 0.0875–0.1103 g h−1 gcat−1, which indicates
that the K–Ga1.10ZnCr catalyst exhibits a good stability during
the reaction.
3.12 Effects of Ga3+ on the performance of the K–ZnCr catalyst
In this section, we will further investigate the relationship
between the structure of the Ga3+ doped catalysts and their
performance. To begin with, the possible location of Ga3+
was deduced by calculating the EB and Ef values, as shown in
Table S2.† The EB values for ZnCr2O4, ZnGa2O4, ZG, CG, and
ZGO follow the order: ZG (348.8783 eV) > ZnCr2O4 (334.9394
eV) > ZnGa2O4 (309.8619 eV) > CG (250.3525 eV) > ZGO
(19.1836 eV), suggesting that ZG and ZnCr2O4 are more stable
than the other compounds. In addition, the Er value
calculated using their most stable oxides follow the order:
ZnCr2O4 (254.7572 eV) > ZG (243.7237 eV) > ZnGa2O4
(229.6797 eV) > CG (162.7697 eV), indicating that ZnCr2O4
and ZG are easier to form through their oxides. As suggested
by Li et al., the Er determines the driving force of a formation
Table 4 The catalytic performance of the K–Ga1.10ZnCr iso-butanol catalysts with time on stream
Selectivity, C mol%
Distribution of alcohols, wt%
CO
STY g
TOS, h
conversion, %
h−1 gcat
CH4
CO2
CHx
DMEb
Alcohol
MEOHc
EtOHd
PrOHe
i-BuOHf
C5+OHg
−1
a
22
31
46
55
78
102
23.17
24.91
24.04
23.44
29.46
21.34
0.1103
0.1000
0.0955
0.1069
0.0875
0.1009
0.85
0.04
0
0.43
0
34.64
34.75
33.77
32.13
32.57
35.47
5.46
5.18
5.13
5.42
4.47
5.15
0.01
0.02
0.01
0.01
0.01
0.01
59.03
60.00
61.08
62.00
62.95
59.37
59.53
57.28
56.75
59.22
58.37
60.75
1.02
0.97
1.01
1.06
1.02
1.35
4.53
4.62
4.52
4.57
4.56
4.77
31.23
33.44
31.92
32.83
33.00
33.13
3.69
3.70
5.80
2.42
2.96
0
0
a
b
c
d
e
Summary of the amount of ethane, propane and butane in the tail gas. Dimethyl ether. Methanol. Ethanol. n-Propanol and
isopropanol. f Iso-butanol. g Amyl alcohol. Reaction conditions: temperature = 400 °C, pressure = 10.0 MPa, GHSV = 3300 h−1
.
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 1077–1088 | 1085

View Article Online
Paper
Catalysis Science & Technology
reaction, a larger Er will lead to the formation of a more
stable compound.61 By combining the Er with the EB, we
deduced that the most stable form of the Ga3+ doped ZnCr
spinel is ZG. The detailed configuration of ZG is listed in the
ESI.† The cell parameters of a, b, and c for ZG are 8.3490,
8.3466, and 8.3495 Å, respectively, which are larger than the
un-doped ZnCr spinel (a = 8.2300 Å, b = 8.2302 Å, c = 8.32304
Å). The calculated cell parameters are larger than the values
that were deduced from the ionic radius of the tetrahedral
coordinated Zn2+ and Ga3+ (their radii are 0.74 and 0. 61 Å,
respectively62). The difference may be caused by the higher
valance of Ga3+. If Zn2+ is partly replaced by Ga3+, excessive
positive charges need to be neutralized to remain electrically
neutral meaning that re-distributions of all cations can be
expected, which will induce enlarged cell parameters.
Furthermore, when Ga3+ was embed in the tetrahedral site of
the ZnCr spinel, the Ga–O bond is only about 1.89 Å, which
is much shorter than the Ga–O bond located in the Ga2O3
crystal (1.94 to 2.10 Å). The calculated Bader charge
distributions of Ga for Ga2O3, ZnGa2O4, and ZG are +1.87,
+1.83, and +1.77, respectively, suggesting that the electron
density of Ga in ZG is higher than the other gallium
compounds which may explain the lower BE of Ga 2p in the
XPS patterns of the K–GaZnCr catalysts.
Secondly, when ZG was formed, the surface composition of
the catalyst was altered. The altered surface composition
significantly affected the interaction between ZnO and
ZnCr2O4 which will further impact the ratio of Zn/Cr, and the
basicity and acidity of the catalysts. To determine the
different surface properties between ZnCr2O4 and ZG, the
ZnCr2O4 (111) and ZG (111) surfaces were constructed, the
adsorption of CO, H2, H2O and CO2 were investigated using
DFT calculations. The most stable configurations and
adsorption energies of the species are shown in Fig. S14–S17
and Table S8.† The results show that Cr is the most favourable
absorption site (Fig. S11†) for CO absorption on both the
ZnCr2O4 (111) and ZG (111) surfaces. The absorption energies
are −0.74 and −2.99 eV respectively, which indicate that Ga3+
adoption is helpful to the absorption of CO. In addition,
remarkably decreased adsorption energies for CO2 and H2O
on the ZG (111) surface were also observed (Table S6†),
suggesting that CO2 and H2O are not favourable for
absorption on the ZG (111) surface. During syngas conversion,
the water gas shift reaction (WGSR) is an inevitable reaction
that consumes a large amount of CO and thus reduces the
efficiency of carbon resource usage. Hence, it is important to
reduce or suppress this reaction. The smaller absorption
energy of H2O indicates that it is not easy for H2O to
participate in the WGSR to produce CO2. This is firmly
verified by the evaluation results of the K–GaxZnCr catalyst.
Thirdly, the productivity of hydrocarbons such as ethane
and propane increased with the increasing amount of Ga3+,
the gas product are mainly caused by the presence of the ZG
compound. The reason for this lies in the following two
aspects: (i) the ratios of Zn/Cr on the surface of the samples
are almost the same for the K–GaxZnCr (x = 0, 0.55, 1.10,
1.64, and 2.18) samples, as detected by XPS, therefore the
effects of extra chromium on the selectivity of the
hydrocarbons can be excluded. (ii) CO is easier to dissociate
on the ZG (111) surface than on the ZnCr2O4 (111) surface,
which can be verified using the CI-NEB calculation on the
elementary reaction COC + O. The results show that the
activation energy for CO dissociation is 3.32 eV lower on the
ZG (111) surface than on the ZnCr2O4 (111) surface (shown in
Fig. S18A†). The dissociated CO is considered to be the origin
of the CHx species, leading to the formation of hydrocarbons
on the samples with a higher amount of added Ga3+.
Finally, the formation of CHO by CO hydrogenation was
also studied on the different catalyst surfaces to investigate
the influence of the surface structure on the CHO formation.
CHO is regarded as an important intermediate to participate
in the formation of the first C2 oxygenates, which is the rate
determining step for higher alcohol synthesis. In addition,
CHO is also the precursor of the OCH3 group, and then by
step by step hydrogenation, methanol will be produced.
Therefore, the higher the concentration of CHO, the higher
the selectivity of alcohol is expected. Lower values for the
activation energy and reaction energy were obtained using
DFT calculations, for example 0.96 and 0.24 eV on the ZnCr
spinel (111) surface versus 0.78 and −1.05 eV on the ZG (111)
surface (Fig. S18B†). The reduced activation energy and
reaction energy for CHO formation indicate that Ga3+
adsorption is advantageous to the formation of CHO, which
will further boost the selectivity of the alcohols.
Conclusions
Ga3+ promoted K–ZnCr catalysts were prepared using the
precipitation–impregnation method. The optimal K–Ga1.10
-
ZnCr catalyst shows a 58.70% increase in the STY and a
14.62% promotion on the selectivity of the alcohols under
the reaction conditions of 10.0 MPa, 400 °C and 3300 h−1
GHSV. Spectral analysis and first principles calculations
revealed that the Zn2+ ions, located at the tetrahedral site
of the ZnCr spinel, were replaced by Ga3+ ions to form
the ZG structure, which promoted the direct dissociation
of CO and the formation of the CHO group. The
coordinately unsaturated ZnCr2O4 surfaces interacted with
Ga or gallium clusters which reduced the adsorption of
CO along with this process. However, upon further
increasing the amount of Ga3+
, ZnGa2O4 was formed
which would modulate the morphologies, surface acidity
and basicity of the final catalysts, eventually inducing the
poor performance of the catalysts.
that is from 2.95% for K–Ga0.0ZnCr to 19.96% for K–Ga2.18
-
ZnCr. Although Cryder et al. proposed that the excess of
surface chromium oxides gives rise to the amount of CO2
and hydrocarbons,63 we conclude that the hydrocarbons in
Conflicts of interest
The authors declare no competing financial interests.
1086 | Catal. Sci. Technol., 2021, 11, 1077–1088
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
Acknowledgements
22 Z. An, X. Ning and J. He, J. Catal., 2017, 356, 157–164.
23 A. Venugopal, J. Palgunadi, J. K. Deog, O.-S. Joo and C.-H.
Shin, J. Mol. Catal. A: Chem., 2009, 302, 20–27.
24 S.-H. Kang, J. W. Bae, P. S. S. Prasad, J.-H. Oh, K.-W. Jun,
S.-L. Song and K.-S. Min, J. Ind. Eng. Chem., 2009, 15,
665–669.
25 K. Sun, Y. Wu, M. Tan, L. Wang, G. Yang, M. Zhang, W.
Zhang and Y. Tan, ChemCatChem, 2019, 11, 2695–2706.
26 A. R. Kim, B. Lee, M. J. Park, D. J. Moon and J. W. Bae,
Catal. Commun., 2012, 19, 66–69.
This work is supported by the National Natural Science
Foundation of China (21573269). Financial aid from the Fund
Program for the Scientific Activities of Selected Returned
Overseas Professionals in Shanxi Province, China is also
greatly appreciated. The DFT calculations were performed on
the Tianhe-II HPC at the Lvliang HPC Centre, China. The
authors give special thanks for the foundations and
computational resource support provided by the government
and the company.
27 M. M.-J. Li, Z. Zeng, F. Liao, X. Hong and S. C. E. Tsang,
J. Catal., 2016, 343, 157–167.
28 Y.-x. Pan, P. Kuai, Y. Liu, Q. Ge and C.-j. Liu, Energy Environ.
Sci., 2010, 3, 1322–1325.
References
1 M. Ao, G. H. Pham, J. Sunarso, M. O. Tade and S. Liu, ACS
Catal., 2018, 8, 7025–7050.
29 H. Ham, S. W. Baek, C.-H. Shin and J. W. Bae, ACS Catal.,
2019, 9, 679–690.
2 H. Guo, S. Li, H. Zhang, F. Peng, L. Xiong, J. Yang, C. Wang,
X. Chen and Y. Chen, Ind. Eng. Chem. Res., 2014, 53,
123–131.
3 K. A. N. Verkerk, B. Jaeger, C.-H. Finkeldei and W. Keim,
Appl. Catal., A, 1999, 186, 407–431.
4 H. T. Luk, C. Mondelli, D. C. Ferre, J. A. Stewart and J. Perez-
Ramirez, Chem. Soc. Rev., 2017, 46, 1358–1426.
5 Y. Wu, N. Gong, M. Zhang, W. Zhang, T. Zhang, J. Zhang, L.
Wang, H. Xie and Y. Tan, Catal. Sci. Technol., 2019, 9,
2592–2600.
6 S. Gao, X. Li, Y. Li, H. Yu, F. Zhang, Y. Sun, H. Fang, X.
Zhang, X. Liang and Y. Yuan, Fuel, 2018, 230, 194–201.
7 S. Tian, L. Tan, Y. Wu, Y. Kou, H. Xie, N. Tsubaki and Y.
Tan, Appl. Catal., A, 2017, 536, 57–66.
8 S. Tian, S. Wang, Y. Wu, J. Gao, P. Wang, H. Xie, G. Yang, Y.
Han and Y. Tan, Catal. Sci. Technol., 2016, 6, 4105–4115.
9 L. Tan, G. Yang, Y. Yoneyama, Y. Kou, Y. Tan, T. Vitidsant
and N. Tsubaki, Appl. Catal., A, 2015, 505, 141–149.
10 S. Tian, S. Wang, Y. Wu, J. Gao, Y. Bai, P. Wang, H. Xie, Y.
Han and Y. Tan, J. Mol. Catal. A: Chem., 2015, 404, 139–147.
11 X. Gao, Y. Wu, T. Zhang, L. Wang, X. Li, H. Xie and Y. Tan,
Catal. Sci. Technol., 2018, 8, 2975–2986.
30 T. Zhang, Y. Wu, X. Gao, H. Xie, G. Yang, N. Tsubaki and Y.
Tan, Fuel, 2019, 237, 1021–1028.
31 L. Lutterotti, M. Bortolotti, G. Ischia, I. Lonardelli and H.-R.
Wenk, Z. Kristallogr. Suppl., 2007, 26, 125–130.
32 L. Lutterotti, Nucl. Instrum. Methods Phys. Res., Sect. B,
2010, 268, 334–340.
33 L. Lutterotti, D. Chateigner, S. Ferrari and J. Ricote, Thin
Solid Films, 2004, 450, 34–41.
34 L. Lutterotti, S. Matthies, H. R. Wenk, A. S. Schultz and J. W.
Richardson, J. Appl. Phys., 1997, 81, 594–600.
35 C. Vargas-Hernández, O. Almanza and J. F. Jurado, J. Phys.:
Conf. Ser., 2009, 167, 012037.
36 S. Tian, S. Wang, Y. Wu, J. Gao, Y. Bai, P. Wang, H. Xie, Y.
Han and Y. Tan, J. Mol. Catal. A: Chem., 2015, 404–405,
139–147.
37 H. Song, D. Laudenschleger, J. J. Carey, H. Ruland, M. Nolan
and M. Muhler, ACS Catal., 2017, 7, 7610–7622.
38 T. Pandiyarajan, M. L. Baesso and B. Karthikeyan, Eur. Phys.
J. D, 2014, 68(28), 1–8.
39 J. Wang, L. Ye, X. Wang, H. Zhang, L. Li, C. Kong and W. Li,
J. Alloys Compd., 2019, 803, 9–15.
40 X. Lu, Z. Du, B. Quan, W. Bian, H. Zhu and Q. Zhang,
J. Mater. Chem. C, 2019, 7, 8261–8268.
12 X. Gao, T. Zhang, Y. Wu, G. Yang, M.-H. Tan, L. Xiaoli, H.
Xie, J. Pan and Y. Tan, Fuel, 2018, 217, 21–30.
41 C. Cristiani, P. Forzatti and M. Bellotto, J. Chem. Soc.,
Faraday Trans. 1, 1989, 85(4), 895–906.
42 M. M. Can, G. H. Jaffari, S. Aksoy, S. I. Shah and T. Firat,
J. Alloys Compd., 2013, 549, 303–307.
43 R. S. Weber, J. Catal., 1995, 151, 470–474.
44 H. Tajizadegan, A. Heidary, O. Torabi, M. H. Golabgir and A.
Jamshidi, Int. J. Appl. Ceram. Technol., 2015, 13(2), 289–294.
45 H. Lin, C. P. Huang, W. Li, C. Ni, S. I. Shah and Y.-H. Tseng,
Appl. Catal., B, 2006, 68, 1–11.
13 L. Wang, X. Gao, Y. Bai, M. Tan, K. Sun, T. Zhang, Y. Wu, J.
Pan, H. Xie and Y. Tan, Fuel, 2019, 253, 1570–1577.
14 K. J. Smith and R. B. Anderson, J. Catal., 1984, 85, 428–436.
15 K. J. Smith and R. B. Anderson, J. Catal., 1984, 85, 428–436.
16 L. Lietti, E. Tronconi and P. Forzatti, J. Catal., 1992, 135,
400–419.
17 Y. Wu, H. Xie, Y. Kou, N. Tsubaki, Y. Han and Y. Tan, Korean
J. Chem. Eng., 2015, 32, 406–412.
18 W. S. Epling, G. B. Hoflund and D. M. Minahan, Appl. Catal.,
A, 1999, 183, 335–343.
46 O. Nishimura, K. Yabe and M. Iwaki, J. Electron Spectrosc.
Relat. Phenom., 1989, 49, 335–342.
19 D. M. Minahan, W. S. Epling and G. B. Hoflund, J. Catal.,
1998, 179, 241–257.
47 R. Merryfield, M. McDaniel and G. Parks, J. Catal., 1982, 77,
348–359.
20 W. S. Epling, G. B. Hoflund and D. M. Minahan, J. Catal.,
1998, 175, 175–184.
48 P. Delichere, S. Daniele and L. G. Hubert-Pfalzgraf, Surf. Sci.
Spectra, 2001, 8, 303–311.
21 W. S. Epling, G. B. Hoflund and D. M. Minahan, React. Kinet.
Catal. Lett., 1999, 67, 225–232.
49 M. Chen, P. Wu, Q. Wei, Y. Zhu, S. Yang, L. Ju, N. Zhu and Z.
Lin, Environ. Chem., 2018, 15, 226–235.
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 1077–1088 | 1087

View Article Online
Paper
Catalysis Science & Technology
50 H. Fan, G. Wang and L. Hu, Solid State Sci., 2009, 11,
2065–2070.
57 B. T. Loveless, C. Buda, M. Neurock and E. Iglesia, J. Am.
Chem. Soc., 2013, 135, 6107–6121.
51 X. Duan, F. Yu and Y. Wu, Eur. J. Inorg. Chem.,
2013, 1287–1293, DOI: 10.1002/ejic.201201004.
52 N. Li, P. Zhu and X. Duan, Phys. B, 2018, 530, 290–294.
53 C. Han, W. Mao, K. Bao, H. Xie, Z. Jia and L. Ye, Int. J.
Hydrogen Energy, 2017, 42, 19913–19919.
54 N. C. Jeong, J. S. Lee, E. L. Tae, Y. J. Lee and
K. B. Yoon, Angew. Chem., Int. Ed., 2008, 47,
10128–10132.
58 M. D. Rhodes, K. A. Pokrovski and A. T. Bell, J. Catal.,
2005, 233, 210–220.
59 A. Riva, F. Trifirò, A. Vaccari, L. Mintchev and G. Busca,
J. Chem. Soc., Faraday Trans. 1, 1988, 84(5), 1423–1435.
60 L. C. Grabow and M. Mavrikakis, ACS Catal., 2011, 1, 365–384.
61 Y. Li, F. Lian, N. Chen, Z.-J. Hao and K.-C. Chou, Int. J.
Miner., Metall. Mater., 2015, 22, 524–529.
62 J. U. Brehm, M. Winterer and H. Hahn, J. Appl. Phys.,
2006, 100, 9.
63 D. S. Cryder and P. K. Frolich, Ind. Eng. Chem., 1929, 21,
867–871.
55 W. Yang, J. Li, X. Zhang, C. Zhang, X. Jiang and B. Liu, Inorg.
Chem., 2019, 58, 549–556.
56 T. J. Mazanec, J. Catal., 1986, 98, 115–125.
1088 | Catal. Sci. Technol., 2021, 11, 1077–1088
This journal is © The Royal Society of Chemistry 2021