Promotion effect of iron addition on the structure and CO2 hydrogenation performance of Attapulgite/Ce0.75Zr0.25O2 nanocomposite supported Cu-ZnO based catalyst

Article abstract of DOI:10.1016/j.mcat.2021.111820

A series of CZF_xK/ATP-CZO catalysts (x= 0, 0.3, 0.5, 1.0, 1.5 and 2.0) are applied to clarify the effects of the iron addition on the catalytic performance of CO₂ hydrogenation to CH₃OH. The physicochemical properties and catalytic mechanism were investigated by N₂ adsorption/desorption, XRD, TEM, N₂O chemisorption, XPS, H₂-TPR, CO₂-TPD and in-situ DRIFT techniques. The best catalytic performance is achieved over CZF_0.5K/ATP-CZO catalyst, exhibiting X_CO2 = 17.5%, STY_CH3OH = 0.108 g/g_cat.?h and STY_CO = 0.146 g/g_cat.?h (T = 320°C, P = 6 MPa). The formation of dispersed surface metallic Cu species and larger number of surface adsorbed and lattcie oxygen species and ZnO-CZO interfaces are detected over CZF_0.5K/ATP-CZO due to stronger interaction between dispersed metallic Cu particles on ZnO-Fe nano-cluster and ATP-CZO composite, resulting in the superior activation ability for H₂ and CO₂ respectively. Additionally, the evidence is provided by in-situ DRIFTS under the activity test temperature (320°C) that HCOO? and CO* species are preferable for accumulating over CZK/ATP-CZO catalyst without Fe addition while medium Fe-modified CZF_xK/ATP-CZO catalysts (CZF_0.3K/ATP-CZO, CZF_0.5K/ATP-CZO) catalysts are benefitial to promote the transformation of HCOO? species to CH₃OH. These excessive Fe-modified CZF_xK/ATP-CZO catalysts (CZF_1.0K/ATP-CZO, CZF_2.0K/ATP-CZO) are more easily to produce CH₄ via formate pathway (CO₂* → HCOO* → HCO* → CH* → CH₄*). The abundant population and high transformation activity of formate intermediate species over CZF_0.5K/ATP-CZO give a strong positive effect on the CO₂ hydrogenation to methanol performance.

Full text of DOI:10.1016/j.mcat.2021.111820

Molecular Catalysis 513 (2021) 111820
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.journals.elsevier.com/molecular-catalysis
Research Paper
Promotion effect of iron addition on the structure and CO2 hydrogenation
performance of Attapulgite/Ce0.75Zr0.25O2 nanocomposite supported
Cu-ZnO based catalyst
,
b
,
,
b
c
,
,
c
,
,
b
,
c
,
,
b
,
c
,
,b,c,d
d, Fen Penga
Haijun Guoa
d, Shuai Dinga e, Hairong Zhanga
d, Can Wanga
,
Shimiao Yaoa
d, Lian Xionga
d, Xinde Chena
,
,
b,
c,
,
b
,
c
,
d
,
*
a Guangzhou Institute of Energy Conversion, Chinese Academy of Sciences, Guangzhou 510640, PR China
b CAS Key Laboratory of Renewable Energy, Guangzhou 510640, PR China
c Guangdong Provincial Key Laboratory of New and Renewable Energy Research and Development, Guangzhou 510640, PR China
d R&D Center of Xuyi Attapulgite Energy and Environmental Materials, Xuyi 211700, PR China
e Nano Science and Technology Institute, University of Science and Technology of China, Suzhou 215123, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
A series of CZFxK/ATP-CZO catalysts (x= 0, 0.3, 0.5, 1.0, 1.5 and 2.0) are applied to clarify the effects of the iron
addition on the catalytic performance of CO2 hydrogenation to CH3OH. The physicochemical properties and
catalytic mechanism were investigated by N2 adsorption/desorption, XRD, TEM, N2O chemisorption, XPS, H2-
TPR, CO2-TPD and in-situ DRIFT techniques. The best catalytic performance is achieved over CZF0.5K/ATP-CZO
catalyst, exhibiting XCO2 = 17.5%, STYCH3OH = 0.108 g/gcat.∙h and STYCO = 0.146 g/gcat.∙h (T = 320◦C, P = 6
MPa). The formation of dispersed surface metallic Cu species and larger number of surface adsorbed and lattcie
oxygen species and ZnO-CZO interfaces are detected over CZF0.5K/ATP-CZO due to stronger interaction between
dispersed metallic Cu particles on ZnO-Fe nano-cluster and ATP-CZO composite, resulting in the superior acti-
vation ability for H2 and CO2 respectively. Additionally, the evidence is provided by in-situ DRIFTS under the
activity test temperature (320◦C) that HCOO* and CO* species are preferable for accumulating over CZK/ATP-
CZO catalyst without Fe addition while medium Fe-modified CZFxK/ATP-CZO catalysts (CZF0.3K/ATP-CZO,
CZF0.5K/ATP-CZO) catalysts are benefitial to promote the transformation of HCOO* species to CH3OH. These
excessive Fe-modified CZFxK/ATP-CZO catalysts (CZF1.0K/ATP-CZO, CZF2.0K/ATP-CZO) are more easily to
produce CH4 via formate pathway (CO2* → HCOO* → HCO* → CH* → CH4*). The abundant population and high
transformation activity of formate intermediate species over CZF0.5K/ATP-CZO give a strong positive effect on
the CO2 hydrogenation to methanol performance.
CO2 hydrogenation
Attapulgite
Ce0.75Zr0.25O2 solid solution
Fe promoter
In-situ DRIFTS
1. Introduction
from the competitive reverse water-gas shift (RWGS) reaction [6, 7].
Therefore, an efficient catalyst should be developed to improve the
With the increasing demand for reducing CO2 emissions all over the
world, CO2 hydrogenation to methanol has become a promising route
that can converting the greenhouse gas into high-value added liquid
energy-carrier [1-4]. Methanol is an excellent alternative fuel for vehi-
cles and important chemical feedstock for producing formaldehyde,
acetic acid, olefins and aromatics etc. However, due to the thermody-
namic stability and kinetic inertness of CO2 molecule, CO2 conversion is
hard to be improved in methanol synthesis from CO2 hydrogenation [5].
Another challenge is the easy formation of undesired byproducts CO
catalyst activity and methanol selectivity in CO2 hydrogenation to
methanol.
In the past decades, a large number of catalysts have been developed
for CO2 hydrogenation to methanol. Among those catalysts, Cu-ZnO-
ZrO2 system (CZZ) is the most popular type due to its prominent cata-
lytic activity and methanol selectivity which are directly determined by
metallic Cu dispersion and surface basicity of CZZ catalyst [8-11].
Recent years, some unconventional routes have been reported to pre-
pare CZZ catalyst, mainly aiming to improve the copper surface area and
* Corresponding author at: No.2 Nengyuan Road, Tianhe District, Guangzhou 510640, PR China.
E-mail address: cxd_cxd@hotmail.com (X. Chen).
https://doi.org/10.1016/j.mcat.2021.111820
Received 24 June 2021; Received in revised form 28 July 2021; Accepted 4 August 2021
Available online 11 August 2021
2468-8231/© 2021 Elsevier B.V. All rights reserved.

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
dispersion. Bonura et al. [12] found that the CZZ catalyst prepared from
gel-oxalate coprecipitation procedure showed much higher copper sur-
face area and dispersion than those samples obtained from coprecipi-
tation with sodium bicarbonate and complexation with citric acid
methods. Arena et al. [8, 13, 14] developed a reverse co-precipitation
method under ultrasound irradiation to prepare CZZ catalysts. They
found that the increase of Zn/Cu ratio improved the dispersion of
metallic Cu acting as hydrogenation active site and enhanced the surface
oxides basic site for adsorbing and activating CO2 molecules. Witoon
et al.’s study [15] showed that increasing Zn/Cu ratio of CZZ catalyst
prepared via a reverse coprecipitation method provided a better
inter-dispersion of metal components (Cu, Zn and Zr), leading to a
smaller CuO crystallite size, a higher porosity and Cu dispersion of the
catalysts. Furthermore, the interaction between CO2 molecules and the
catalyst surface was also alleviated which was beneficial to the forma-
tion of methanol at a lower reaction temperature (240 and 250 ◦C).
and Ce0.75Zr0.25O2 solid solution) in CZFK/ATP-CZO-SC-P400 catalyst.
However, the effects of iron promoter on the structure and catalytic
performances of Cu-ZnO-K2O catalyst (CZK) are not clear. Therefore, in
the present work, a series of ATP-CZO-SC-P400 supported CZFK cata-
lysts were synthesized with theoretical Fe/Cu molar ratio from 0 to 2.0
and Cu/Zn molar ratio of 1.0. The physicochemical properties of pre-
pared catalysts with different Fe content were characterized by X-ray
fluorescence (XRF), X-ray diffraction (XRD), N2 adsorption/desorption,
N2O chemisorption, Transmission electron microscopy (TEM), X-ray
photoelectron spectroscopy (XPS), H2-Temperature programmed
reduction (H2-TPR), CO2-Temperature programmed desorption
(CO2-TPD) and in-situ diffuse reflectance infrared Fourier transform
spectroscopy (DRIFTS) techniques. The effects of Fe promoter on the
physicochemical properties and catalytic performances of the catalysts
would be importantly addressed.
Dong et al. [16] prepared
a
series of CZZ catalysts by
2. Experimental
precipitation-reduction method of which the exposed Cu surface area
and the ratio of Cu0/Cu+ could be adjusted by changing the content of
NaBH4 reducing agent. Guo et al. synthesized CZZ catalyst by urea–ni-
trate combustion method [17] and glycine–nitrate combustion method
[18]. The metallic copper surface and dispersion of CuO and the phase
state of ZrO2 could be effectively improved by adopting suitable fuel
content. The above reports clearly indicated that the nature of the active
site and fine structure of CZZ catalyst could be effectively tuned by using
appropriate preparation method and changing catalyst composition.
Besides Cu-ZnO-ZrO2 catalyst, Cu-ZnO-CeO2 [6] and Cu-ZnO--
Ce1ꢀ xZrxO2 [19] catalysts have also been applied in CO2 hydrogenation
to methanol. The Ce1ꢀ xZrxO2 solid solution contained more oxygen
vacancies which provided a larger adsorption and an easy spillover of
active hydrogen species to speed up the hydrogenation of formate in-
termediate to form methanol [19, 20].
2.1. Materials and chemicals
Attapulgite powders were provided by ZHONGKE New Energy
Technological Development Co., Ltd (Huai-An, China) and the acid-
activated ATP was prepared by 21 wt% H2SO4 activation [34]. Ce
(NO3)3•6H2O (AR) and ZrO(NO3)2•2H2O (AR) are purchased from
Tianjin Kemiou Chemical Reagent Co., Ltd., China. Ethylene glycol and
PEG 400 are purchased from Sinopharm Chemical Reagent Co., Ltd,
China. These reagents were used without further purification.
2.2. Catalyst preparation
ATP-CZO-SC-P400 nanocomposite was prepared according to our
previous work and abbreviated as ATP-CZO [26]. 10.2
g Ce
In order to further improve the activity and stability of CZZ catalyst,
some porous carriers such as silica [21], carbon nanotubes [22] and
Mg-Al layered double hydroxides (LDHs) [23] have been used to modify
or support it. The large surface area of these materials is beneficial for
improving the copper dispersion of catalyst, meanwhile, the adsorption
capacity of CO2 can be enhanced due to the increase of surface basicity
of catalyst [21, 24]. Attapulgite (ATP) is another promising natural
porous nanomaterial, with an ideal chemical formula of Mg5Si8O20(O-
H)2(OH2)4•4H2O, has large surface area and special porous structure
[25]. Recently, we developed a series of ATP/Ce0.75Zr0.25O2 (ATP-CZO)
nanocomposites by different methods and applied to support Cu-ZnO--
Fe2O3-K2O catalyst (CZFK) for CO2 hydrogenation reaction [26]. The
CZFK catalyst supported on ATP-CZO nanocomposite which is prepared
by polyethylene glycol (PEG 400) modified solution combustion method
(ATP-CZO-SC-P400) showed the largest BET surface area (SBET) and
more strong basic sites, thus obtaining the highest yield of methanol and
CO from CO2 hydrogenation. The exposed copper surface area and
dispersion of metallic copper of CZFK/ATP-CZO-SC-P400 catalyst
exhibited a volcano variation trend with the increase of PEG 400 addi-
tion content.
(NO3)3•6H2O and 2.1 g ZrO(NO3)2•2H2O were sufficiently dissolved in
20 mL de-ionized water/ethylene glycol mixed solvent (v/v=1/1). 10 g
acid-activated ATP powder (100 mesh) and 3g PEG 400 were homoge-
nously dispersed in the above solution by ultrasonification for 20 min,
and then magnetically stirred for 4.5 h. The resulting gel was combusted
in air from room temperature to 500 ◦C at a rate of 2.5 ◦C/min and
remained for 6 h.
These ATP-CZO-SC-P400 supported CZFK catalysts were synthesized
by excessive impregnation method. The nitrates precursors of Cu, Zn and
Fe and anhydrous potassium carbonate were solubilized in H2O by
stirring, and then was impregnated into ATP-CZO-SC-P400 powder
(>100 mesh) for 4.5 h. Excessive solution was removed by vacuum
filtration. The obtained catalyst precursor was dried in air at 120 ◦C
overnight, and then calcined in air at 350 ◦C for 5.5 h. These catalysts
were denoted as CZFxK/ATP-CZO (x= 0, 0.3, 0.5, 1.0, 1.5 and 2.0).
2.3. Catalyst characterization
The catalyst composition was determined by XRF on an AxiosmAX
-
Petro XRF (PANalytical, Netherlands). The composition was expressed
in percentage by weight of oxides. XRD analysis was conducted using a
X’Pert Pro MPD diffractometer (PANalytical, Netherlands) in the 2θ
As mentioned above, the catalytic hydrogenation performances of
Cu-ZnO based catalysts are also been largely affected by catalyst
composition. Some metallic promoters, like La, Ce, Nd, Pr, Mg, Mn, were
added to improve the metallic copper surface and dispersion of CuO [27,
28], which can increase the catalytic activity of methanol synthesis. Fe,
as typical Fischer–Tropsch (FT) element, was usually applied to increase
the selectivity of higher alcohols (C2+OH) from CO or CO2 hydrogena-
tion reaction due to a strong carbon chain growth ability [29-32], while
its application in methanol synthesis from CO2 hydrogenation is rarely
reported. In fact, Fe effectively inhibited the sintering and oxidation of
Cu and elevated the catalytic stability of Cu-based catalysts at high
temperatures for the RWGS reaction [33]. In our previous work [26], the
iron promoter improved the interdispersion of mixed metal oxides and
the interactions between Cu species and other metal oxides (ZnO, Fe2O3
range from 5 o to 80o with Cu K
α radiation (λ = 0.15406 nm) at 40 kV
and 40 mA. The textural properties of catalyst were determined by N2
adsorption/desorption at liquid N2 temperature using an ASIQMO002-2
analyzer (Quantachrome, US). The adsorption isotherms branch was
applied to calculate the surface area according to Brunauer–Emmett–-
Teller (BET) method. The pore size distribution (PSD) was obtained
using Barrett–Joyner–Halenda (BJH) method. TEM observation was
performed on a JEM-2100F microscope (JEOL, Japan) operating at 200
kV. Before testing, a little sample was dispersed in anhydrous ethanol by
ultrasonification and dripped onto a carbon film-coated cooper grid. XPS
detection of fresh catalysts was conducted on an ESCALAB 250Xi spec-
trometer (Thermo Fisher Scientific Inc, US) with Al Kα radiation (1486.8
2

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
eV). The binding energy (BE) was calibrated using the C1s peak (284.6
eV) as the reference.
mm × 0.25
μ
m) and a hydrogen flame ionization detector (FID). CO2
conversion and products selectivity were calculated by mass-balance
method [32]. The space time yield (STY) of CH3OH and CO was calcu-
lated according to Dong et al. [16].
H2-TPR experiments of these catalysts were carried out by using a
ChemBET Pulsar-1 automated chemisorption analyzer (Quantachrome,
US) in a U shape quartz tube reactor. Prior to reduction, the catalyst (120
mg) was pretreated in Ar gas flow at 120 ◦C for 60 min. After that, the
reduction experiment was started from 50 to 800 ◦C at a heating rate of
10 ◦C/min in 10%H2/Ar flow of 30 mL/min. The hydrogen consumption
was monitored by a Thermal Conductivity Detector (TCD) and quanti-
fied by five pulses calibration.
3. Results and discussion
3.1. Structural and Textural properties
The N2 adsorption/desorption isotherms and the corresponding PSD
curves of the ATP-CZO composite and these CZFxK/ATP-CZO catalysts
are displayed in Fig. 1. As shown in Fig. 1a, the type IV isotherms with a
big H3-type hysteresis loop are observed for all the samples, indicating
the presence of meso/macropores [15, 35]. However, the adsorbed
volume of ATP-CZO support is much higher than that of catalysts. From
Fig. 1b, it can be seen that all PSD plots derived from desorption curve
show a common peak centered at about 3.8 nm which is caused by an
abrupt desorption of N2 at a relative pressure (P/P0) of about 0.4. The
rising branch curve of the peaks located at the range of 5-100 nm and the
highest N2 desorption changes remarkably with increasing the content
of Fe promoter.
The Cu surface area and dispersion of these catalysts were deter-
mined by N2O chemisorption measurement on the same reactor as H2-
TPR detection [26]. Firstly, the fresh catalysts (105 mg) were reduced in
10%H2/Ar mixture at 400 ◦C for 2 h and the amount of hydrogen con-
sumption was denoted as X. And then, the reactor was purged with Ar to
50 ◦C. The 20%N2O/N2 gas was introduced to reoxidize surface copper
atoms to Cu2O for 1 h. Subsequently, the catalyst was flushed with Ar to
remove the oxidant. Finally another TPR experiment was performed as
the same as the first reduction and the amount of consumed H2 was
denoted as Y. By quantifying the amount of consumed H2, the dispersion
of Cu and exposed Cu surface area of the catalyst were calculated by the
following equations, respectively.
From the textural properties as shown in Table 1, it can be seen that
the BET surface area (SBET) and pore volume (VP) of these catalysts are
much smaller than ATP-CZO composite, indicating the surface of ATP-
CZO composite was covered by metallic oxides. The CZK/ATP-CZO
catalyst has a SBET of 67.9 m2/g and Vp of 0.249 cm3/g. The introduc-
tion of Fe promoter leaded to a slight decrease of SBET but increase of Vp
and average pore diameter (DP) for the CZF0.3K/ATP-CZO catalyst.
However, the further increase of Fe content brought the increase of SBET
DCu = 2Y/X × 100(%)
/ꢀ
)
SACu = (2 × Y × Nav) X × MCu × 1.4 × 1019
/ ꢀ
/
)
≈ 1353 × Y X m2 ꢀ Cu g ꢀ Cu
Where, DCu is dispersion of Cu, SACu is exposed Cu surface area, Nav is
Avogadro’s constant (6.02 × 1023 atoms/mol), MCu is the relative
atomic mass (63.456 g/mol), and 1.4 × 1019 is the number of Cu atoms
per square meter.
CO2-TPD experiments of these catalysts were conducted in the same
reactor as H2-TPR experiment. The catalyst was firstly reduced at 400 ◦C
for 1 h in a continuous 10%H2/Ar flow of 30 mL/min. After cooling to 50
◦C, the catalyst was saturated in CO2 flow for 1 h and then flushed with
Ar to remove the physical adsorbed molecules. Then, the CO2 desorption
process was performed from 50 to 800 ◦C with a heating rate of 10 ◦C/
min in Ar gas flow and the desorbed CO2 was recorded by a mass
spectrometer.
In-situ DRIFTS analysis was applied to detect the surface adsorbed
species after CO2 adsorption. All spectra were recorded by a TENSOR 27
spectrometer (Bruker, US) equipped with a liquid N2 cooled MCT de-
tector. The spectra were obtained by 32 scans and each scan was
collected ranging from 800 to 4000 cmꢀ 1 at a resolution of 4 cmꢀ 1. Each
sample was first reduced at 400 ◦C for 1 h in 10%H2/Ar and then cooled
down to 50 ◦C. And then the pre-reduced samples were exposed to H2/
CO2 for 1 h and purged with N2 for 5 min. After collecting the back-
ground spectra, the DRIFTS tests were carried out at 320 ◦C in H2/CO2
(3:1) gas flow of 60 mL/min under pressure of 0.2 MPa.
2.4. Catalytic performances evaluation
CO2 hydrogenation performances of these catalysts were evaluated
in a stainless steel tubular, continuous-flow, fixed-bed reactor (internal
diameter, 8 mm; length, 280 mm) [32]. Catalyst particles (60-80 mesh,
2.7 g) diluted with quartz sand (40-60 mesh, 9 g) was firstly reduced in
pure H2 under pressure of 0.5 MPa at 400 ◦C for 4 h. After reduction, the
reactor was cooled down to 100 ◦C and then the reactant gas flow
(H2/CO2 with molar ratio of 3/1) was introduced into the reactor,
raising the pressure to 6.0 MPa and adjusting the gas hourly space ve-
locity (GHSV) to 5000 hꢀ 1. When the reactor temperature increased to
320 ◦C, CO2 hydrogenation reaction was timed to start. The gaseous
products (H2, CO2, CO and CH4) were off-line analyzed by gas chro-
matographs (GC) with a TDX-01 column (1 m × 3 mm) and a TCD. The
liquid products were analyzed by GC with a FFAP column (30 m × 0.25
Fig. 1. (a) N2 adsorption-desorption isotherms and (b) BJH pore size distri-
bution for ATP-CZO support and CZFxK/ATP-CZO catalysts.
3

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
Table 1
Chemical composition and textural properties of ATP-CZO support and CZFxK/ATP-CZO catalysts.
Sample
Chemical composition
Metallic molar ratio b
SBET c (m2/g)
VP d (cm3/g)
DP e (nm)
dCu f (nm)
SACu h (m2/g)
DCu i (%)
(wt%) a
CuO
ZnO
Fe2O3
Fe/
Cu
Fe/
b
(Cu+Zn)
-
ATP-CZO
CZK/ATP-CZO
CZF0.3K/ATP-
CZO
-
-
2.4
0
-
158.4
67.9
58.0
0.348
0.249
0.253
8.8
-
-
-
10.2
10.5
9.3
10.1
0.00
0.25
0.00
0.13
14.7
17.4
71.5
42.4
40
84
5.9
12.4
2.7
CZF0.5K/ATP-
CZO
9.2
8.5
5.3
8.5
7.8
5.0
4.1
0.44
0.87
1.85
0.23
0.46
0.97
65.0
87.4
118.9
0.236
0.265
0.277
14.5
12.1
9.3
28.7
19.6
17.9
125
177
242
18.4
26.2
35.8
CZF1.0K/ATP-
CZO
7.5
CZF2.0K/ATP-
CZO
10.0
a
b
Detected by XRF.
Calculated by deducting the Fe2O3 content in the ATP-CZO support itself.
SBET: BET surface area.
c
d
e
f
VP: Pore volume.
DP: Average pore diameter.
dCu: Crystallite size of Cu calculated using Scherrer equation based on the reflection of (111) lattice plane at 2θ = 43.3◦ in those reduced catalysts.
gSACu: Surface area of Cu.
DCu: Dispersion of Cu
h
and Vp, which is consistent with the unsupported CZFK catalysts pre-
pared by co-precipitation method [32]. Fe has been determined as a
structural promoter for Cu-based catalyst to improve the dispersion of
active oxide particles [29, 30]. N2O chemisorption was used to detect
the dispersion of metallic Cu (DCu) and surface area of Cu (SACu). The
calculated DCu and SACu according to these equations reported by Van
Der Grift et al. [36] are showed in Table 1. A higher DCu and SACu are
obtained in these catalysts with higher Fe content, indicating that the
addition of Fe promoter improved the dispersion of metallic Cu.
reduction showed a decrease from 76.6% to 45.4% (Table 2) due to the
decrease of CuO content in these catalysts (Table 1), the peak area of
α
CZF0.5K/ATP-CZO catalyst is the highest among these CZFxK/ATP-CZO
catalysts, implying the presence of more dispersed CuO species (Fig. 4).
It is worth noting that a pair of shoulder peaks (β peak) in the range of
320 ~ 400 ◦C are accompanied with the reduction peak of CuO species
only on the CZF0.5K/ATP-CZO catalyst. It can be assigned to the
reduction of composited copper oxides [26, 32], indicating some extra
active copper species are present in the catalyst. Because of the intrinsic
Fe2O3 in ATP-CZO itself (Table 1), the consecutive reduction peak (γ
peak) of Fe2O3→Fe3O4→FeO→Fe with lower intensity is also displayed
at 400 ~ 750◦C on the CZK/ATP-CZO catalyst [38]. The increase of Fe
addition promoted the increase of the total H2 consumption proportion
of Fe2O3 reduction from 23.4% to 54.6% (Table 2), showing that more
iron oxides were formed and reduced [29]. According to the N2O
chemisorption detection results (Table 1), the increase of Fe addition
continually promoted the dispersion of metallic Cu. However, the
highest total H2 consumption was obtained on the CZF0.5K/ATP-CZO
catalyst. Therefore, there are some other factors for enhancing H2
chemisorption (Table 2) [26]: (1) the decrease of interaction between
dispersed CuO and ATP; (2) the intimate contact between CuO and other
oxides (Ce-Zr-O solid solution, ZnO and Fe2O3) for increased in-
teractions; (3) the improved dispersion of CuO upon ZnO sheets to form
much more Cu-ZnO interface.
The XRD patterns of ATP-CZO composite and fresh and reduced
CZFxK/ATP-CZO catalysts are shown in Fig. 2. The diffraction peaks of
palygorskite, quartz and Ce0.75Zr0.25O2 solid solution are observed all
samples with similar 2θ value and peak intensity (Fig. 2A), implying the
loading of Cu, Zn and Fe did not bring obvious change to the crystal
structure of ATP-CZO composite. As can be noted from Fig. 2A, with
regard to all the samples, no characteristic diffraction peaks of ZnO and
Fe2O3 are present, which may be resulted from the lower loading con-
tent of CZFK particles. The diffraction peaks at 2θ = 35.7o, 38.9o, 49.0o,
58.5o and 61.7o are corresponded to CuO (JCPDS, No. 45-0937), of
which the intensity decreases gradually with the increase of Fe addition
[32]. Similarly, the peak intensity of formed metallic Cu after reduction
at 400 ◦C also decreases companying with the peak intensity increase of
metallic Fe (Fig. 2B). The crystallite size of metallic Cu (2θ = 43.3◦) on
these CZFxK/ATP-CZO samples determined by Scherrer equation
(Table 1) decreases from 71.5 nm (x = 0) to 17.9 nm (x = 2.0), showing
that incorporation of Fe facilitates the dispersion of metallic Cu [29],
which is consistent with the N2O chemisorption results. The HRTEM
images of fresh CZFxK/ATP-CZO catalysts were also observed as shown
in Fig. 3. All samples exhibit spherical particles dispersing on the cata-
lyst surface. Furthermore, more rod-like crystals are exposed when the
Fe/Cu molar ratio is much higher, while the average particle size of CuO
particles gradually decreased which is consistent with the XRD results.
3.3. Effect of Fe addition on the surface chemical state and formation of
oxygen vacancies
XPS measurements of CZK/ATP-CZO, CZF0.5K/ATP-CZO and
CZF2.0K/ATP-CZO catalysts were carried out to investigate the surface
chemical state and the assignment of oxygen species. As shown in Fig. 5,
a peak at lower BE of 932.7 ~ 933.2 eV and another peak at higher BE of
934.6 ~ 935.3 eV are observe on the XPS spectra of Cu 2p3/2, which is
2+
3.2. Effect of Fe addition on the reducibility of catalysts
assigned to the CuA2+ in tetrahedral CuO (A) species [39] and the CuB
in octahedral composite metal oxides (B) [30], respectively. Further-
more, a pair of satellite peaks at higher BE of 940 ~ 945 eV ascribed to
The TPR profiles of CZFxK/ATP-CZO samples are shown in Fig. 4. For
all catalysts, the reduction peaks of CuO (
α
peak) are centered in the
the shake-up transitions by ligand-to-metal 3d charge transform are also
2+
range of 220 ~ 350 ◦C and divided into two peaks (250 ~ 280 ◦C and
270 ~ 310 ◦C) which are corresponded to the reduction of CuO species
with small and large crystalline particle size, respectively [37]. With the
addition of Fe promoter, the reduction peak of CuO gradually transfers
towards lower temperature, demonstrating the reduction of CuO was
promoted [29, 32]. Though the total H2 consumption proportion of CuO
present [40]. From the peak area proportion of CuB2+/(CuA2++CuB
)
in Table 3, it can be seen that the CZF0.5K/ATP-CZO catalyst owns the
highest proportion of octahedral CuB2+ which has also been determined
by H2-TPR (Fig. 4 and Table 2). The XPS spectra of Fe 2p3/2 shows
similar XPS peak information with Cu 2p3/2, on which a peak at lower BE
of 711.0 ~ 711.9 eV ascribed to FeA3+ in Fe2O3 phase and another peak
4

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
Fig. 3. TEM images of fresh CZFxK/ATP-CZO catalysts: (a-b) CZK/ATP-CZO, (c-
d) CZF0.3K/ATP-CZO, (e-f) CZF0.5K/ATP-CZO, (g-h) CZF1.0K/ATP-CZO and (i-j)
CZF2.0K/ATP-CZO.
hydrogenation to produce alcohols, the CZF0.5K/ATP-CZO catalyst
showed the optimal activity level, as would be expected. For the XPS
spectra of Zn 2p, the characteristic peak at BE of 1045 eV and 1021 eV is
ascribed to Zn 2p1/2 and Zn 2p3/2, respectively, indicating the surface
phase of Zn species is ZnO [48]. The increase of iron addition leads to the
decline of XPS peak intensity, implying that ZnO was covered by other
metallic species with a higher extent at higher iron addition.
Fig. 2. XRD patterns of (a) ATP-CZO support and (A) fresh, (B) reduced and (C)
used CZFxK/ATP-CZO catalysts: (b) CZK/ATP-CZO, (c) CZF0.3K/ATP-CZO, (d)
CZF0.5K/ATP-CZO, (e) CZF1.0K/ATP-CZO and (f) CZF2.0K/ATP-CZO.
The formation and accumulation of adsorbed oxygen species has
been proved to be favorable for enhancing CO2 adsorption and activa-
tion in CO2 hydrogenation reactions [37, 49-51]. The chemical state of
oxygen species over the ATP-CZO supported Cu-ZnO based catalysts
surface by fitting the XPS curves is shown in Fig. 5 and the relative
proportion of surface adsorbed oxygen species (Oβ) at 531.7 eV and
surface lattice oxygen species (Oα) at 529.6 eV is calculated on basis of
the XPS data as shown in Table 3. It has been reported that the adsorbed
oxygen species are related to the groups including ꢀ OH in water and
Cꢀ O in CO32ꢀ which is derived from carbonate species trapped by ox-
ygen vacancies [49, 52]. Therefore, the amount of oxygen vacancies can
be obtained from the relative proportion of adsorbed oxygen species
(Oβ) [52]. From Table 3, it is noted that the Oα/Oβ ratio shows a gradual
at higher BE of 713.8 ~ 714.9 eV assigned to FeB3+ in composite metal
3+
oxide phase are obviously observed [30]. The FeB
in
CZF0.5K/ATP-CZO catalyst shows the highest BE, while the peak area
proportion of FeB3+/(FeA3++FeB3+) is the lowest, implying the presence
of the strongest interaction between Cu and Fe species. Much more
Cu-Fe dual-functional active site can be formed due to the enhancement
of Cu-Fe interaction, thus improving the equilibrium capacity between
Cꢀ O bond insertion and Cꢀ C chain growth to produce more alcohols
from CO hydrogenation [38, 41-47]. Therefore, when performing CO2
5

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
Fig. 4. H2-TPR profiles of CZFxK/ATP-CZO catalysts: (a) CZK/ATP-CZO, (b) CZF0.3K/ATP-CZO, (c) CZF0.5K/ATP-CZO, (d) CZF1.0K/ATP-CZO and (e) CZF2.0K/
ATP-CZO.
Table 2
Reduction peak temperature and hydrogen consumption over CZFxK/ATP-CZO catalysts.
H2 consumption (%) a
Total H2 consumption (
μ
mol gcat.
)
ꢀ 1
Sample
Temperature (◦C)
α
β
γ
α
β
γ
CZK/ATP-CZO
269.1/309.5
277.4/294.5
260.4/284.9
261.5/278.5
256.1/269.6
568.0/693.2
518.2/690.5
472.7/577.5/713.1
515.7/715.8
539.6/719.9
47.8/28.8
51.3/18.8
20.2/32.6
40.8/9.6
39.2/6.2
-
17.9/5.5
25.1/4.8
6.5/24.0/4.0
43.7/5.9
48.3/6.3
2145
2586
5202
2375
1947
CZF0.3K/ATP-CZO
CZF0.5K/ATP-CZO
CZF1.0K/ATP-CZO
CZF2.0K/ATP-CZO
-
-
340.0/377.8
11.1/1.6
-
-
-
-
a
Relative proportion of hydrogen consumption.
increase with the increase of iron addition, indicating much more sur-
face lattice oxygen species are formed due to the stronger interaction
between the oxygen atoms and other metallic atoms. In addition, a
fitting peak at 532.2 eV is attributed to the surface oxidized metal ions
(Oγ) in the catalyst nanoparticles [53]. The CZF0.5K/ATP-CZO catalyst
shows the highest Oα+Oβ proportion in the entire surface oxygen spe-
cies, which means that the adsorption and activation of CO2 can be
improved to enhance the CO2 hydrogenation activity.
6

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
Fig. 5. XPS patterns of CZFxK/ATP-CZO catalysts: (a) CZK/ATP-CZO, (b) CZF0.5K/ATP-CZO and (c) CZF2.0K/ATP-CZO.
Table 3
Surface atomic ratio estimated by XPS for CZFxK/ATP-CZO catalysts.
Sample
BE (eV)
Area (%)
2+
2+
3+
3+
2+
3+
CuA
CuB
FeA
FeB
Oα
Oβ
Oγ
CuB
/
FeB
/
Oα/
Oβ
(Oα+Oβ)/
(Oα+Oβ+Oγ)
79.4
(CuA2++CuB
)
(FeA3++FeB
)
2+
3+
CZK/ATP-CZO
CZF0.5K/ATP-
CZO
932.7
932.8
934.6
934.7
711.9
711.6
714.9
715.2
529.4
529.6
531.6
531.7
532.2
532.3
67.4
50.5
6.9
7.5
79.2
38.9
85.3
CZF2.0K/ATP-
CZO
933.2
935.3
711.0
713.8
529.6
531.7
532.4
45.1
42.3
11.0
72.3
3.4. Effect of Fe addition on the adsorption behavior of catalysts
support. The β peak intensity also shows a decrease due to the H2
reduction of catalyst but the γ peak intensity obviously increases,
especially for the CZF0.5K/ATP-CZO catalyst which shows the highest
CO2 desorption proportion of 65.9% (Table 4). Our previous work has
determined that the γ peak was caused by the formation of ZnO-CZO
interfaces active sites for CO2 adsorption and conversion due to the
enhancement of interaction between ZnO and Ce-Zr-O oxide [26].
Therefore, the decrease of γ peak intensity (CO2 desorption proportion)
with the increase of Fe addition when the actual Fe/(Cu+Zn) molar ratio
is more than 0.46 (Table 1), implies that the formation of ZnO-CZO
interfaces might be restrained because of the coverage of ZnO by
excessive iron species which has been proved by XPS results.
CO2-TPD was performed from 50 ◦C to 800 ◦C to detect the basicity
site of H2-reduced CZFxK/ATP-CZO catalysts. As shown in Fig. 6, two
broad CO2 desorption bands are displayed ranging from 60 ◦C and 300
◦C to 250 ◦C and 700 ◦C (labeled as LT and HT peaks, respectively),
which is similar with our previous report [26]. For the ATP-CZO sup-
port, the LT peak (α
peak) centered at 127.1 ◦C is assigned to weakly
basic hydroxyl groups on the ATP and Ce-Zr-O oxide [16, 54]. The HT
peak is fitted to form a main peak (β peak) at 429.7 ◦C accompanying
with a overlapped peak at 610.4 ◦C (γ peak), which is attributed to the
existence of low coordination oxygen atoms [55] and the bidentate
carbonate species derived from the strong basic sites of Ce-Zr mixed
oxides [54], respectively. For these H2-reduced CZFxK/ATP-CZO cata-
The mode of CO2 adsorption and intermediate species of CO2 hy-
drogenation are studied using in-situ DRIFTS. Fig. 7a-e represent the in-
situ DRIFT spectra over these H2-reduced CZFxK/ATP-CZO catalysts in
the reaction gas (H2+CO2) at the actual CO2 hydrogenation temperature
(320 ◦C). As shown, three key surface absorbed species, including car-
bonate species (1535, and 1495 cmꢀ 1) [56], formate species (2906,
1633, 1419, and 1351 cmꢀ 1) [11, 31, 57], and adsorbed CO2 (1246
cmꢀ 1) [58], were mainly detected on both other catalysts except for
CZK/ATP-CZO catalyst. Moreover, the peaks at 2145 and 2107 cmꢀ 1 for
lysts, the peak intensity of
α peak shows different extent of decrease due
to the loss of weakly basic hydroxyl groups during the calcination pro-
cess of the catalyst. However, the peak temperature is much higher
compared to ATP-CZO support which may be related with the enhanced
interaction between metallic species and ATP-CZO support. On the
contrary, the introduction of CZFK metallic particles lead to the decrease
of desorption peak temperature of β and γ peaks on the ATP-CZO
7

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
CZF2.0K/ATP-CZO catalyst also showed similar changing trend. More-
over, two novel peaks at 2886 cmꢀ 1 and 2973 cmꢀ 1 ascribed to the
ethoxy species (CH3CH2O*) were present after 30 min (Fig. 7e), sug-
gesting the formation of CH3CH2O* due to the enhancement of carbon
chain growth ability [31]. Given above analysis, we believe that meth-
anol was synthesized from CO2 hydrogenation over ATP-CZO supported
Cu-ZnO based catalysts via formate intermediate species, and the
introduction of Fe promoter facilitates the conversion of CO and formate
intermediates.
3.6. Catalyst performances
3.6.1. Evaluation of CO2 hydrogenation over different catalysts
The CO2 hydrogenation performances over these CZFxK/ATP-CZO
catalysts were evaluated. As shown in Table 5, the content of Fe addition
plays an important influence on the catalytic performance. As well
known, CO2 is thermodynamically stable and difficult to be activated.
The presence of oxygen vacancies is liable to enhance the adsorption of
CO2 [11, 37, 50, 60-62]. The dispersed CuO species are responsible for
catalytic activity of Cu-based catalysts in CO2 hydrogenation reaction
[15, 18, 37]. Because of less ZnO-CZO interfaces (Fig. 6a) and lower
dispersion of metallic Cu (Table 1) over the CZK/ATP-CZO catalyst, the
CO2 conversion is only 5.9% over the CZK/ATP-CZO catalyst. The
introduction of Fe promoter obviously improved the CO2 conversion to
the highest value of 17.5% over the CZF0.5K/ATP-CZO catalyst. From
Table 1, it can be seen that the surface area SACu and dispersion of
metallic Cu (DCu) were continually improved from 40 m2/g and 5.9% to
242 m2/g and 35.8% with the increase of Fe addition, respectively.
However, the further increase of Fe content leaded to the decrease of
CO2 conversion to 12.8% over the CZF2.0K/ATP-CZO catalyst (Table 5),
which is related with its lower reduction degree during H2-TPR detec-
tion process due to the restriction of hydrogen spillover (Fig. 4e) and the
restriction of CO2 adsorption over ZnO-CZO interfaces (Fig. 6f).
Fig. 6. CO2-TPD patterns of ATP-CZO (a) and CZFxK/ATP-CZO catalysts: (b)
CZK/ATP-CZO, (c) CZF0.3K/ATP-CZO, (d) CZF0.5K/ATP-CZO, (e) CZF1.0K/ATP-
CZO and (f) CZF2.0K/ATP-CZO.
Table 4
Desorption peak temperature and CO2 desorption over CZFxK/ATP-CZO
catalysts.
Sample
Temperature (◦C)
CO2 desorption (%) a
α
β
γ
α
β
γ
ATP-CZO
127.1
137.5
128.0
157.6
139.7
137.8
429.7
407.9
389.6
392.5
395.3
403.6
610.4
521.3
464.5
484.1
452.7
457.9
41.6
33.9
30.8
22.3
27.7
26.3
44.2
43.0
20.1
11.8
14.7
20.3
14.2
23.1
49.1
65.9
57.6
53.4
CZK/ATP-CZO
CZF0.3K/ATP-CZO
CZF0.5K/ATP-CZO
CZF1.0K/ATP-CZO
CZF2.0K/ATP-CZO
The products selectivity was also affected by the content of Fe
addition (Table 5). CO was mainly produced from CO2 hydrogenation
over all catalysts through the endothermic RWGS reaction (CO2 + H2 →
CO + H2O) [63, 64]. CH4 was also generated with different levels by the
methanation reaction of CO2 (CO2 + 4H2 → CH4 + 2H2O) [65]. Mixed
alcohols (marked as ROH, including methanol-C1OH and higher alco-
hols-C2+OH) could be formed by the direct CO2 hydrogenation and
hydrogenation of CO intermediate [66]. Therefore, to a certain degree,
the product selectivity reflects the CO2 hydrogenation reaction routes.
As shown in Table 5, with the increase of x, the selectivity of CH4 and CO
decrease at first and then increase through the minima of 4.2% and
5.1%, respectively. ROH shows an opposite change trend due to the
competitive reaction route of CO2 hydrogenation and achieves the
maxima of 35.1% over the CZF1.0K/ATP-CZO catalyst. When x is 2.0, the
formation of ROH were significantly restrained which seems to promote
the production of CH4, while C2+OH with fraction of 5.4% in alcohols
product suggested the formation of CH3CH2O*.
a
Relative proportion of CO2 desorption.
adsorbed CO* appear on these Fe-modified Cu-ZnO based catalysts,
suggesting the formation of CO at high temperature (320 ◦C), which is
similar with the study of Xu et al. [31]. The bands at 1633 and 1351
cmꢀ 1 were assigned to asymmetric and symmetric Oꢀ Cꢀ O stretching
vibrations respectively, and the peaks at 2906 and 1419 cmꢀ 1 were
assigned to the stretching vibration ν(CH) and bending vibration δ(CH),
which are commonly belonging to the adsorbed HCOO* species [59].
For the CZK/ATP-CZO catalyst, the intensity of adsorbed CO2 (CO2*)
band at 1246 cmꢀ 1 and formate bands at 2906, 1633, and 1419 cmꢀ 1
showed gradual increase with the increase of exposure time in reaction
gas from 1 to 30 min, while their intensity almost kept unchanged from
30 to 60 min (Fig. 7a). Therefore, CO may be accumulated as the main
CO2 hydrogenation product over CZK/ATP-CZO catalyst. After adding
Fe promoter, even if the exposure time in reaction gas is only 1 min, the
intensity of formate bands at 1633 and 1419 cmꢀ 1 on the
CZF0.3K/ATP-CZO catalyst is enough high and shows slight decrease
with the increase of exposure time to 60 min (Fig. 7b), implying that the
formate species (HCOO*) might be converted into methoxy species
(CH3O*) at 2831 cmꢀ 1 [60-62]. Meanwhile, the band intensity of
adsorbed CO2 showed similar changing trend. When x increased 0.5, the
peak intensity of all bands on the CZF0.5K/ATP-CZO catalyst increased at
first and then decreased (Fig. 7c), indicating the carbonate species,
formate species and gaseous CO had been effectively transformed which
is beneficial for the improvement of CO2 conversion. However,
compared to the CZF0.5K/ATP-CZO catalyst, the peak intensity of all
bands changed sharply lower for the CZF1.0K/ATP-CZO catalyst as seen
from the smaller proportional scale (Fig. 7d) and showed gradual in-
crease with the increase of exposure time in reaction gas. The
The effects of the addition content of Fe promoter on the STY of
methanol and CO formation over CZFxK/ATP-CZO catalysts are shown
in Fig. 8a. As can be seen, the addition of Fe promoter enhanced the
production of methanol and CO when x is less than 0.5 because of the
improvement of metallic Cu dispersion (Table 1). However, the further
increase of x to 2.0 leaded to the obvious decrease of STY of methanol
which might be resulted from the improved formation of CO and CH4. To
sum up, the CZF0.5K/ATP-CZO sample presents superior catalytic per-
formance in terms of CO2 conversion and CH3OH production. The
turnover frequency (TOF) which represents the number of CO and
methanol produced on a unit site of a copper atom per second (sꢀ 1) on
the catalyst surface, was calculated from the SCu for various catalysts to
compare the catalytic efficiency. As shown in Fig. 8b, the TOFMethanol
showed gradual increase with the increase of Fe addition content until
reaching the maximum value of 2.4 × 10ꢀ 3 over CZF0.5K/ATP-CZO
catalyst, and then declined to 0.5 × 10ꢀ 3 over CZF2.0K/ATP-CZO
8

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
Fig. 7. In-situ DRIFT spectra of CZFxK/ATP-CZO catalysts: (a) CZK/ATP-CZO, (b) CZF0.3K/ATP-CZO, (c) CZF0.5K/ATP-CZO, (d) CZF1.0K/ATP-CZO and (e) CZF2.0K/
ATP-CZO. CO2 hydrogenation reaction conditions: 320◦C, 20 mL/min CO2 flow +60 mL/min H2 flow, 0.2 MPa.
Table 5
Catalytic performance for CO2 hydrogenation over CZFxK/ATP-CZO catalysts. a
Catalyts
CO2 conversion
Products selectivity (C-mol%)
Alcohols distribution (wt%)
(%)
CH4
8.5
CO
ROH b
19.0
30.7
33.4
35.1
20.2
12.2
HOC c
1.0
C1OH
98.8
99.3
99.3
99.1
96.9
94.6
C2+OH
CZK/ATP-CZO
5.9
71.5
59.1
51.1
52.7
58.2
62.1
1.2
0.7
0.7
0.9
3.1
5.4
CZF0.3K/ATP-CZO
CZF0.5K/ATP-CZO
CZF1.0K/ATP-CZO
CZF1.5K/ATP-CZO
CZF2.0K/ATP-CZO
11.1
17.5
14.3
13.6
12.8
4.2
6.0
4.7
10.8
5.0
7.3
16.3
25.6
5.3
0.1
a
Reaction condition: T= 593 K, P= 6.0 MPa, n(H2):nCO2= 3:1 and GHSV= 5000 hꢀ 1
.
b
c
ROH stands for mixed alcohols.
HOC stands for other oxygenates.
9

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
species (Oβ) (Table 3) was present than other catalysts (Fig. 5). CO2-TPD
result proved that weakly basic hydroxyl groups (α peak) and low co-
ordination oxygen atoms (β peak) occupied the main CO2 adsorption
sites over CZK/ATP-CZO catalyst (Fig. 6), thus forming much more
gaseous CO intermediate species on the in-situ DRIFT spectra (Fig. 7a).
To sum up, as shown in Fig. 9a, a Cu0/Cuox redox mechanism for RWGS
reaction may occur at the boundary layers of metal copper in intimate
contact with the surface of ZnO over CZK/ATP-CZO catalyst [26, 67].
Methanol has been extensively synthesized from CO2 hydrogenation
over different metallic oxides (Al2O3, CeO2, ZrO2, etc.) or their com-
posites supported and modified Cu-ZnO based catalysts [8, 10, 11,
14-16, 18, 19, 21, 22, 28, 68-70]. Some crucial factors for active sites,
such as a Cu step with a nearby Zn serving as adsorption site for
oxygen-bound intermediates [71], a synergy of Cu and ZnO at the
interface [72, 73], and an adequate balance between metal (Cu0) and
oxide surface sites (ZnO and ZrO2) [12], were proposed to form and
stabilize the intermediates for methanol synthesis. Metallic copper
species of which the exposed surface area [18], dispersion, particle size
[74] and the ratio of Cu0/Cu+ [16], is the active site for H2 adsorption
and activation to control the CO2 conversion and methanol selectivity.
Oxides carriers of which the crystalline form [70] and basic sites [21, 68,
69], is the active site for the adsorption and activation of CO2 to affect
the reaction routes over Cu-ZnO based catalyst system by adjusting
catalyst texture and exposed metal surface [19]. Therefore,
a
metal-oxide dual site nature on metal/oxides interface is crucial for the
efficient synthesis of methanol from CO2 hydrogenation [26].
In our previous work, it has been suggested that the ATP-CZO-SC-
P400 support owns significant interactions between ATP and CZO ox-
ides due to better dispersion of CZO particles and directionally arranged
ATP rodlike crystals [26]. The introduction of moderate content of Fe
promoter into CZK/ATP-CZO catalyst would bring positive effects for
metallic copper species and ATP-CZO support. On one hand, according
to the XRD (Fig. 2b and 2c) and XPS (Fig. 5) patterns, the generated
CuFe2O4 species could act as a textural promoter to inhibit the oxidation
of metallic Cu [33]. Meanwhile, the dispersion of metallic Cu (Table 1)
was also improved to maintain lower particle size which provides more
opportunities for the dissociative adsorption of H2 to produce atomic
hydrogen and spillover onto the intimate contacted ZnO or CZO com-
posite (Fig.9b) [16, 70, 72]. In addition, the interactions between Cu
species and other metal oxides (ZnO, Fe2O3 and CZO composite) were
also largely enhanced as shown in XPS (Fig. 5) and CO2-TPD results
(Fig. 6). On the other hand, more surface adsorbed and lattcie oxygen
species (Fig. 5 and Table 3) and ZnO-CZO interfaces active sites (Fig. 6
and Table 4) could be generated to enhance CO2 adsorption, activation
and conversion into carbonate species (CO3*, HCO3*) and formate
species (HCOO*) (Fig. 8). According to the mechanism of CO2 hydro-
genation to methanol, the larger exposed copper surface area is favor-
able to provide large interfacial contact between copper and metal
oxides [21], thus enhancing the formation and transformation of
formate species into methoxy species (CH3O*). As a result, the syner-
gistic interactions between dispersed metallic Cu particles on ZnO-Fe
nano-cluster and ATP-CZO composite favored the production of meth-
anol over CZFxK/ATP-CZO catalysts with x less than 0.5 [21].
Fig. 8. Effect of Fe/Cu molar ratio in CZFxK/ATP-CZO catalysts on the
dispersion of Cu (DCu) and STY of methanol and CO (a) and surface area of Cu
(SCu) and TOF of methanol and CO (b). Reaction conditions: T= 320◦C, P= 6.0
MPa, n(H2):nCO2= 3:1 and GHSV= 5000 hꢀ 1
.
catalyst. However, the highest TOFCO of 5.0 × 10ꢀ 3 was obtained over
CZK/ATP-CZO catalyst, which further indicating the RWGS reaction
happened dominantly. The increase of Fe promoter leaded to the
decrease of TOFCO to 2.6 × 10ꢀ 3 over CZF1.0K/ATP-CZO and CZF2.0K/
ATP-CZO catalysts. But overall, CO is more likely to be produced than
methanol by CO2 hydrogenation over CZFxK/ATP-CZO catalysts due to
the promotion of C=O bond scission by reducible CeO2 [26].
3.6.2. Structure-properties-catalytic performance correlations
Some previous work reported that RWGS reaction over Cu-ZnO
based catalysts conforms the surface redox mechanism involving the
oxidation and reduction of the Cu surface, Cu(0) ⇌ Cu(I) [64, 67]. CO2
was adsorbed and dissociated to give CO and the surface oxygen species.
Surface Cu(I) oxide was formed by the reaction with CO2. The oxygen
species were hydrogenated to H2O and the surface Cu(I) oxide was
reduced to metallic Cu. According to the XRD spectra of reduced and
used CZK/ATP-CZO catalyst (Fig. 2b and 2c), the diffraction peak in-
tensity of metallic Cu sharply declined and the diffraction peak of Cu(I)
oxide was obviously displayed, implying that the Cu(I) oxide was
formed by the oxidation of metallic Cu. H2-TPR result showed that much
higher H2 consumption proportion (Table 2) was assigned to the
With the further increase of Fe content to x = 2.0, the dispersion of
metallic Cu could be increasingly improved (Table 1). However, exces-
sive Fe content leaded to the increase of particle size of metallic Cu due
to the surrounding of metallic Cu by some small metallic Fe particles.
Since the formation ability of weakly basic hydroxyl groups and low
coordination oxygen atoms was recovered (Table 4), more oxygen atoms
produced from the reaction process on metallic Cu migrated to iron
surfaces through a spillover process [33], thus effectively prevents sin-
tering and oxidation of Cu. According to the in-situ DRIFT spectra
(Fig. 7d), the peak intensity of formate species showed gradual increase
with the increase of exposure time in reaction gas over
CZF1.0K/ATP-CZO catalyst, implying that the formate pathway by CO2
associative methanation (CO2* → HCOO* → HCO* → CH* → CH4*) was
reduction of dispersed CuO (
α peak) over CZK/ATP-CZO catalyst
(Fig. 4a). XPS result showed that much more surface adsorbed oxygen
10

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
Fig. 9. Schematic of the tailoring of product distribution in CO2 hydrogenation over CZFxK/ATP-CZO catalysts.
the most plausible reaction scheme for CH4 formation [50, 65, 75, 76].
As shown in Fig. 9c, beside CO2 methanation, the RWGS reaction can
occur simultaneously over CZF2.0K/ATP-CZO catalyst. The peak in-
tensity of formate species started to decline after adsorption for 40 min
in reaction gas (Fig. 7e) provided an evidence for CH4 production
through formate pathway.
original draft, Funding acquisition.
Shuai Ding: Investigation, Formal analysis.
Hairong Zhang: Methodology, Validation, Resources, Funding
acquisition.
Can Wang: Investigation, Validation.
Fen Peng: Investigation, Formal analysis, Funding acquisition.
Shimiao Yao: Investigation, Formal analysis, Funding acquisition.
Lian Xiong: Conceptualization, Writing - Review & Editing.
Xinde Chen: Supervision, Conceptualization, Project administration.
4. Conclusions
The work explores the effects of Fe promoter on structure and cata-
lytic performances of ATP-CZO nanocomposite supported Cu-ZnO based
catalysts. The strong interaction effect over CZF0.5K/ATP-CZO catalyst
with medium Fe addition contributes to the remarkable catalytic activity
for CO2 hydrogenation to CH3OH, which leads to higher surface
dispersion of metallic Cu and larger number of surface adsorbed and
lattcie oxygen species and ZnO-CZO interfaces active sites, promoting H2
and CO2 adsorption and conversion ability, thus achieving high catalytic
performance (STYCH3OH = 0.108 g/gcat.∙h, STYCO = 0.146 g/gcat.∙h),
compared with other catalysts. Additionally, in-situ DRIFTS suggest that
CZK/ATP-CZO catalyst without Fe addition is much easier to enrich
HCOO* and CO* species while medium Fe-modified CZFxK/ATP-CZO
catalysts (x ≤ 0.5) are more benefit to promote the hydrogenation of
HCOO* species to generate the HCO3* intermediate species for meth-
anol formation, besides excessive Fe-modified CZFxK/ATP-CZO catalysts
(1.0 ≤ x ≤ 2.0) are more liable to produce CH4 via formate pathway
(CO2* → HCOO* → HCO* → CH* → CH4*). The decrease of larger
number of active intermediate species HCOO* is responsible for the
higher activity of CO2 hydrogenation to CH3OH over CZF0.5K/ATP-CZO
sample.
Declaration of Competing Interest
There are no conflicts to declare.
Acknowledgments
This work was financially supported by the Guangdong Basic and
Applied Basic Research Foundation (2020A1515010516), the Project of
Guangdong Provincial Natural Science Foundation (2018A030310126,
2018A030313150), the Project of Jiangsu Province Science and Tech-
nology (BE2018342), the Key-Area Research and Development Program
of Jiangsu Province (BE2020394) and the Project of Key Laboratory
Foundation of Renewable Energy, Chinese Academy of Science (CAS)
(E1290104).
References
[1] M.D. Porosoff, B. Yan, J.G. Chen, Energ. Environ. Sci. 9 (2016) 62–73.
[2] S. Saeidi, N.A.S. Amin, M.R. Rahimpour, J. CO2 Util 5 (2014) 66–81.
[3] X. Jiang, X. Nie, X. Guo, C. Song, J.G. Chen, Chem. Rev. 120 (2020) 7984–8034.
[4] J. Zhong, X. Yang, Z. Wu, B. Liang, Y. Huang, T. Zhang, Chem. Soc. Rev. 49 (2020)
1385–1413.
Creadit author statements
[5] J. Schneider, H. Jia, J.T. Muckerman, E. Fujita, Chem. Soc. Rev. 41 (2012)
2036–2051.
Haijun Guo: Methodology, Formal analysis, Investigation, Writing
[6] X. Hu, W. Qin, Q. Guan, W. Li, ChemCatChem 10 (2018) 4438–4449.
11

H. Guo et al.
Molecular Catalysis 513 (2021) 111820
¨
[7] E.L. Kunkes, F. Studt, F. Abild-Pedersen, R. Schlogl, M. Behrens, J. Catal. 328
[43] K. Xiao, Z. Bao, X. Qi, X. Wang, L. Zhong, K. Fang, M. Lin, Y. Sun, J. Mol. Catal. A:
(2015) 43–48.
Chem. 378 (2013) 319–325.
´
´
[8] F. Arena, K. Barbera, G. Italiano, G. Bonura, L. Spadaro, F. Frusteri, J. Catal. 249
(2007) 185–194.
[44] H.T. Luk, C. Mondelli, S. Mitchell, S. Siol, J.A. Stewart, D. Curulla Ferre, J. Perez-
Ramírez, ACS. Catal. 8 (2018) 9604–9618.
[9] C. Li, X. Yuan, K. Fujimoto, Appl. Catal. A. 469 (2014) 306–311.
[10] P. Gao, F. Li, H. Zhan, N. Zhao, F. Xiao, W. Wei, L. Zhong, H. Wang, Y. Sun, J. Catal.
298 (2013) 51–60.
[45] S. He, W. Wang, Z. Shen, G. Li, J. Kang, Z. Liu, G.-C. Wang, Q. Zhang, Y. Wang,
Mol. Catal. 479 (2019) 151–163.
[46] Y. Li, W. Gao, M. Peng, J. Zhang, J. Sun, Y. Xu, S. Hong, X. Liu, X. Liu, M. Wei,
B. Zhang, D. Ma, Nat. Commun. 11 (2020) 61.
[11] Y. Wang, S. Kattel, W. Gao, K. Li, P. Liu, J.G. Chen, H. Wang, Nat. Commun. 10
(2019) 1166.
[47] W.U. Khan, L. Baharudin, J. Choi, A.C.K. Yip, ChemCatChem 13 (2021) 111–120.
[48] R.F. Mulligan, A.A. Iliadis, P. Kofinas, J. Appl. Polym. Sci. 89 (2003) 1058–1061.
[49] M. Zhang, J. Zhang, Y. Wu, J. Pan, Q. Zhang, Y. Tan, Y. Han, Appl. Catal. B. 244
(2019) 427–437.
[12] G. Bonura, M. Cordaro, C. Cannilla, F. Arena, F. Frusteri, Appl. Catal. B. 152–153
(2014) 152–161.
[13] F. Arena, G. Italiano, K. Barbera, G. Bonura, L. Spadaro, F. Frusteri, Catal. Today.
143 (2009) 80–85.
[50] X. Jia, X. Zhang, N. Rui, X. Hu, C.-j. Liu, Appl. Catal. B. 244 (2019) 159–169.
[51] W. Wang, Z. Qu, L. Song, Q. Fu, J. Energ. Chem. 40 (2020) 22–30.
[52] N. Wang, K. Shen, L. Huang, X. Yu, W. Qian, W. Chu, ACS. Catal. 3 (2013)
1638–1651.
[14] F. Arena, G. Italiano, K. Barbera, S. Bordiga, G. Bonura, L. Spadaro, F. Frusteri,
Appl. Catal. A. 350 (2008) 16–23.
[15] T. Witoon, N. Kachaban, W. Donphai, P. Kidkhunthod, K. Faungnawakij,
M. Chareonpanich, J. Limtrakul, Energ. Convers. Manage. 118 (2016) 21–31.
[16] X. Dong, F. Li, N. Zhao, F. Xiao, J. Wang, Y. Tan, Appl. Catal. B. 191 (2016) 8–17.
[17] X. Guo, D. Mao, G. Lu, S. Wang, G. Wu, Catal. Commun. 12 (2011) 1095–1098.
[18] X. Guo, D. Mao, G. Lu, S. Wang, G. Wu, J. Catal. 271 (2010) 178–185.
[19] G. Bonura, F. Arena, G. Mezzatesta, C. Cannilla, L. Spadaro, F. Frusteri, Catal.
Today. 171 (2011) 251–256.
[53] V. Deerattrakul, P. Dittanet, M. Sawangphruk, P. Kongkachuichay, J. CO2 Util 16
(2016) 104–113.
[54] F. Qu, Y. Wei, W. Cai, H. Yu, Y. Li, S. Zhang, C. Li, Energ. Fuel. 32 (2018)
2104–2116.
[55] Y. Sun, L. Chen, Y. Bao, G. Wang, Y. Zhang, M. Fu, J. Wu, D. Ye, Catal. Today. 307
(2018) 212–223.
[20] K. Pokrovski, A. Bell, J. Catal. 241 (2006) 276–286.
[21] T. Phongamwong, U. Chantaprasertporn, T. Witoon, T. Numpilai, Y. Poo-arporn,
W. Limphirat, W. Donphai, P. Dittanet, M. Chareonpanich, J. Limtrakul, Chem.
Eng. J. 316 (2017) 692–703.
[56] K. Pokrovski, K.T. Jung, A.T. Bell, Langmuir 17 (2001) 4297–4303.
[57] G.J. Millar, C.H. Rochester, K.C. Waugh, Cat. Let. 14 (1992) 289–295.
[58] K. Kahler, M.C. Holz, M. Rohe, J. Strunk, M. Muhler, ChemPhysChem 11 (2010)
2521–2529.
[22] Q. Zhang, Y.-Z. Zuo, M.-H. Han, J.-F. Wang, Y. Jin, F. Wei, Catal. Today. 150
(2010) 55–60.
[59] S. Kattel, B. Yan, Y. Yang, J.G. Chen, P. Liu, J. Am. Chem. Soc. 138 (2016)
12440–12450.
[23] X. Fang, Y. Men, F. Wu, Q. Zhao, R. Singh, P. Xiao, T. Du, P.A. Webley, J. CO2 Util
29 (2019) 57–64.
[60] T. Liu, X. Hong, G. Liu, ACS. Catal. 10 (2019) 93–102.
[61] S.S. Dang, B. Qin, Y. Yang, H. Wang, J. Cai, Y. Han, S.G. Li, P. Gao, Y. Sun, Sci. Adv.
6 (2020) eaaz2060.
[24] N.D. Hutson, S.A. Speakman, E.A. Payzant, Chem. Mater. 16 (2004) 4135–4143.
[25] W.L. Haden, I.A. Schwint, Ind. Eng. Chem. Res. 59 (1967) 58–69.
[26] H. Guo, Q. Li, H. Zhang, F. Peng, L. Xiong, S. Yao, C. Huang, X. Chen, Mol. Catal.
476 (2019), 110499.
[62] Y.H. Wang, W.G. Gao, H. Wang, Y.E. Zheng, W. Na, K.Z. Li, RSC Adv 7 (2017)
8709–8717.
[63] C.-S. Chen, W.-H. Cheng, S.-S. Lin, Appl. Catal. A. 238 (2003) 55–67.
[64] S.-I. Fujita, M. Usui, N. Takezawa, J. Catal. 134 (1992) 220–225.
[65] B. Miao, S.S.K. Ma, X. Wang, H. Su, S.H. Chan, Catal. Sci. Tchnol. 6 (2016)
4048–4058.
[27] H. Ban, C. Li, K. Asami, K. Fujimoto, Catal. Commun. 54 (2014) 50–54.
[28] F. Meshkini, M. Taghizadeh, M. Bahmani, Fuel 89 (2010) 170–175.
[29] M. Ding, M. Qiu, T. Wang, L. Ma, C. Wu, J. Liu, Appl. Energy. 97 (2012) 543–547.
[30] H. Guo, H. Zhang, F. Peng, H. Yang, L. Xiong, C. Wang, C. Huang, X. Chen, L. Ma,
Appl. Catal. A. 503 (2015) 51–61.
[66] D.L.S. Nieskens, D. Ferrari, Y. Liu, R. Kolonko, Catal. Commun. 14 (2011) 111–113.
[67] H. Guo, S. Li, F. Peng, H. Zhang, L. Xiong, C. Huang, C. Wang, X. Chen, Catal. Let.
145 (2015) 620–630.
[31] D. Xu, M. Ding, X. Hong, G. Liu, S.C.E. Tsang, ACS. Catal. (2020) 5250–5260.
[32] S. Li, H. Guo, C. Luo, H. Zhang, L. Xiong, X. Chen, L. Ma, Catal. Lett. 143 (2013)
345–355.
[68] F. Arena, G. Mezzatesta, G. Zafarana, G. Trunfio, F. Frusteri, L. Spadaro, J.Catal
300 (2013) 141–151.
[33] C.-S. Chen, W.-H. Cheng, S.-S. Lin, Appl. Catal. A. 257 (2004) 97–106.
[34] H.J. Guo, H.R. Zhang, F. Peng, H.J. Yang, L. Xiong, C. Huang, C. Wang, X.D. Chen,
L.L. Ma, Appl. Clay Sci. 111 (2015) 83–89.
[69] F. Arena, G. Mezzatesta, G. Zafarana, G. Trunfio, F. Frusteri, L. Spadaro, Catal.
Today. 210 (2013) 39–46.
[70] T. Witoon, J. Chalorngtham, P. Dumrongbunditkul, M. Chareonpanich,
J. Limtrakul, Chem. Eng. J. 293 (2016) 327–336.
[35] M. Nazari, R.M. Behbahani, G. Moradi, A.S. Lemraski, J. Porous Mater. 23 (2016)
1037–1046.
¨
[71] M. Behrens, F. Studt, I. Kasatkin, S. Kühl, M. Havecker, F. Abild-Pedersen,
[36] C.J.G. Van Der Grift, A.F.H. Wielers, B.P.J. Jogh, J. Van Beunum, M. De Boer,
M. Versluijs-Helder, J.W. Geus, J. Catal. 131 (1991) 178–189.
[37] W. Wang, Z. Qu, L. Song, Q. Fu, J. Energ. Chem. 47 (2020) 18–28.
[38] H.T. Luk, C. Mondelli, S. Mitchell, D.C. Ferre, J.A. Stewart, J. Perez-Ramirez,
J. Catal. 371 (2019) 116–125.
S. Zander, F. Girgsdies, P. Kurr, B.-L. Kniep, M. Tovar, R.W. Fischer, J.K. Nørskov,
¨
R. Schlogl, Sci. 336 (2012) 893–897.
[72] S. Kattel, P.J. Ramirez, J.G. Chen, J.A. Rodriguez, P. Liu, Sci. 355 (2017)
1296–1299.
[73] S.D. Senanayake, P.J. Ramírez, I. Waluyo, S. Kundu, K. Mudiyanselage, Z. Liu,
Z. Liu, S. Axnanda, D.J. Stacchiola, J. Evans, J.A. Rodriguez, J. Phys. Chem. C. 120
(2016) 1778–1784.
´
[39] J. Haber, T. Machej, L. Ungier, J. Ziołkowski, J. Solid State Chem. 25 (1978)
207–218.
[40] B. Wallbank, I.G. Main, C.E. Johnson, J. Electron. Spectrosc. Relat. Phenom. 5
(1974) 259–266.
[74] F. Frusteri, M. Cordaro, C. Cannilla, G. Bonura, Appl. Catal. B. 162 (2015) 57–65.
[75] H.L. Huynh, J. Zhu, G. Zhang, Y. Shen, W.M. Tucho, Y. Ding, Z. Yu, J. Catal. 392
(2020) 266–277.
[41] Y. Lu, R. Zhang, B. Cao, B. Ge, F.F. Tao, J. Shan, L. Nguyen, Z. Bao, T. Wu, J.
W. Pote, B. Wang, F. Yu, ACS. Catal. 7 (2017) 5500–5512.
[42] Y. Lu, F. Yu, J. Hu, J. Liu, Appl. Catal. A. 429–430 (2012) 48–58.
¨
[76] T. Burger, F. Koschany, O. Thomys, K. Kohler, O. Hinrichsen, Appl. Catal. A. 558
(2018) 44–54.
12