Design of highly stable MgO promoted Cu/ZnO catalyst for clean methanol production through selective hydrogenation of CO2

Article abstract of DOI:10.1016/j.apcata.2021.118239

The synergistic interaction between small Cu particles and MgO/ZnO-supported catalysts, synthesized by the hydrothermal method, show a very high methanol production rate (0.0063 mol g_Cu^?1 h^?1). High Cu dispersion and large Cu surface area in the hydrothermal synthesized Cu/MgO/ZnO catalyst postulated to be the reason for high activity. The formation of defected ZnO crystals with Mg atoms provided a better adsorption site for CO₂ (near Mg atom), whereas Cu-ZnO interface sites are responsible for the activation of CO₂. 20 wt% loaded MgO catalyst showed preference to selective CO₂ hydrogenation pathway producing clean methanol with > 99 % selectivity. In addition, Density Functional Theory (DFT) studies revealed that the basic nature of the MgO support can be the probable reason for the higher CO₂ adsorption at the Cu-MgO interface compared to the Cu-ZnO interface. Cu₁₃/MgO/ZnO (100) surface model is studied to understand the promoting effect of MgO on CO₂ adsorption.

Full text of DOI:10.1016/j.apcata.2021.118239

Applied Catalysis A, General 623 (2021) 118239
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Design of highly stable MgO promoted Cu/ZnO catalyst for clean methanol
production through selective hydrogenation of CO2
,
,
,c
Sachin Kumar Sharmaa b, Tuhin Suvra Khana b, Rajib Kumar Singhaa, Bappi Paula
,
Mukesh Kumar Poddara, Takehiko Sasakid, Ankur Bordoloia, Chanchal Samantae,
Shelaka Guptaf, Rajaram Bala
*
,b,
a Light Stock Processing Division, CSIR-Indian Institute of Petroleum, Dehradun, 248005, India
b Academy of Scientific and Innovative Research (AcSIR), Ghaziabad, 201002, India
c Department of Chemistry, National Institute of Technology Nagaland, Dimapur, Nagaland, 797103, India
d Department of Complexity Science and Engineering, Graduate School of Frontier Science, The University of Tokyo, Kashiwanoha, Kashiwa-shi, Chiba, 277-8561, Japan
e Bharat Petroleum Corporation Ltd., Greater Noida, Uttar Pradesh, 201306, India
f Department of Chemical Engineering, Indian Institute of Technology Hyderabad, Kandi, 502285, Sangareddy, Telangana, India
A R T I C L E I N F O
A B S T R A C T
Keywords:
The synergistic interaction between small Cu particles and MgO/ZnO-supporteꢀd1catalysts, synthesized by the
hydrothermal method, show a very high methanol production rate (0.0063 mol gCu hꢀ 1). High Cu dispersion and
large Cu surface area in the hydrothermal synthesized Cu/MgO/ZnO catalyst postulated to be the reason for high
activity. The formation of defected ZnO crystals with Mg atoms provided a better adsorption site for CO2 (near
Mg atom), whereas Cu-ZnO interface sites are responsible for the activation of CO2. 20 wt% loaded MgO catalyst
showed preference to selective CO2 hydrogenation pathway producing clean methanol with > 99 % selectivity.
In addition, Density Functional Theory (DFT) studies revealed that the basic nature of the MgO support can be
the probable reason for the higher CO2 adsorption at the Cu-MgO interface compared to the Cu-ZnO interface.
Cu13/MgO/ZnO (100) surface model is studied to understand the promoting effect of MgO on CO2 adsorption.
Cu-Nanoparticles
MgO promotor
CO2 hydrogenation
Methanol
DFT-study
1. Introduction
ways that may offer a solution to utilize CO2 and thereby mitigate its
emission into the natural environment [2]. Because of the high ther-
Major industrialized nations are now more committed than ever to
drastically reducing greenhouse gas emissions after the Paris agreement
on Climate change. The Parris agreement’s main goal is to limit global
warming to well below 2 ◦C, preferably to 1.5 ◦C, compared to pre-
industrial levels. Carbon dioxide (CO2) is a greenhouse gas and
emitted as a by-product o from power plants, steel industries, oil re-
fineries, chemical industries, and other energy production processes.
Thus, excessive anthropogenic CO2 emission led to climate change and
global warming, which is one of the biggest problems nature and hu-
manity is facing in the 21st century [1,2]. Concurrently, CO2 is a high
potential, cheap, non-toxic, and abundant C1 feedstock that can be
utilized for the synthesis of fuels and value-added products. The cata-
lytic transformation of carbon dioxide into valuable products like
methanol, ethanol, syngas (CO + H2), dimethyl ether (DME), urea,
formaldehyde, carbonates, and hydrocarbons are the most attractive
modynamic stability of CO2, the successful utilization of CO2 is a real
challenge, and selective catalytic hydrogenation of CO2 into methanol
has attracted appreciable attention as one of the possible paths for CO2
fixation. Methanol can be used as a fuel (such as in methanol fuel cell),
as a fuel additive (such as gasoline additive or MTBE production), or as
an initial feedstock in the chemical industries (mainly downstream
processes), and as a hydrogen supplier in direct methanol fuel cell
(DMFC) [3].
Researchers around the world have been exploring different catalytic
pathways like photocatalysis, electrocatalysis, and thermal heteroge-
neous/homogeneous catalysis for the utilization of CO2 [2,4]. The main
obstacle of CO2 hydrogenation to methanol and other value-added
chemicals by thermal heterogeneous catalysts is reverse water gas
shift (RWGS) reaction [3] because the copper-based catalyst used in the
methanol synthesis may also favour the RWGS reaction
* Corresponding author.
E-mail address: raja@iip.res.in (R. Bal).
https://doi.org/10.1016/j.apcata.2021.118239
Received 2 March 2021; Received in revised form 30 May 2021; Accepted 1 June 2021
Available online 4 June 2021
0926-860X/© 2021 Elsevier B.V. All rights reserved.

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
[31] reported the significant role of alkaline-earth metal oxide (MgO,
BaO, SrO, and CaO) on the copper-based catalyst, which enhanced the
number of CO2 and H2 adsorption active sites and also increased the
metal-support interaction with high metallic Cu surface area. There are
reports that MgO inhibits RWGS reaction [32,33], so a catalyst with
properties of inhibiting RWGS reaction and capable of CO2 activation at
low temperatures could be ideal for methanol formation via CO2 hy-
drogenation. MgO as a catalyst component could lead to higher CO2
adsorption because of its basic nature and also improve metal dispersion
by increasing surface area due to its low density [33]. Different pa-
rameters like particle size, surface area, Cu surface area, and composi-
tion of the catalyst influence the catalytic activity and methanol
selectivity in a CO2 hydrogenation reaction. Although there are several
reports in the literature for the catalytic hydrogenation of CO2 to
methanol over Cu based catalysts applying different preparation
methods but to the best of our knowledge, there is no report till date
where a Cu-based catalyst is showing very high selectivity and stability
for a longer run. So, the development of highly stable and selective
catalysts is highly inevitable.
CO2 + 3H2 ↔ CH3OH + H2O ΔH298 = -49.47 kJ/m ol
CO2 + H2 ↔ CO + H2O ΔH298 = 41.17 kJ/m ol
(1)
(2)
From the above two equations, it can be observed that with
increasing temperature, the RWGS (Eq. (2) shown above) reaction be-
comes predominant. So, hydrogenation of CO2 to methanol should be
carried out at lower temperatures. Industrially methanol is produced
from synthesis gas mixture over Cu/ZnO/Al2O3 catalyst. The commer-
cial Cu-Zn-Al-based methanol synthesis catalysts from syn-gas were also
used for the production of methanol from pure CO2, but the selectivity is
poor due to the formation of excess CO and methane [5–7]. Although
several research groups found that Cu-Zn-Zr catalysts are effective for
the production of methanol from CO2 hydrogenation [4,8,9]. but the
industrial application is far away due to poor selectivity and rapid
deactivation due to the sintering of Cu particles [10–12]. ZnO appears to
be an important component as it prevents the agglomeration of Cu
particles leading to the high Cu surface area required for methanol
production [11,12]. It is reported that graphitic like ZnOx over-layers on
Cu nanoparticles is the active site over the industrial Cu-Zn-Al catalyst
[13]. When H2 dissociates over metallic Cu hydrogen, spill-over is
inevitable for high CO2 conversion. So, to get high methanol selectivity,
it is required to use a catalyst that can adsorb and activate CO2 without
Here, we report the highly dispersed sinter resistant Cu-
nanoparticles supported on MgO-ZnO catalyst prepared by hydrother-
mal method, and the catalyst showed superior activity and > 99 %
methanol selectivity at low temperature (200 ◦C) and low pressure (30
bar).
–
breaking both the C O bonds of the molecule [14]. For industrial
Cu-Zn-Al catalysts, they do not cleave both bonds in the molecule, i.e.,
no methane formation, but the big problem is RWGS. High temperature
increases the RWGS reaction, produces more CO, and results the
decrease of selectivity of methanol. In order to increase the CO2 con-
version and to reduce the CO selectivity, the reaction is generally carried
out at high pressure (> 50 bar) in industrial methanol production [15,
16]. The techno-economic analysis of this process was conducted, and it
was found that the cost of the compressor to increase the system pressure
is almost 45 % of the total equipment cost. The required energy con-
sumption for compressing the gas is almost 66 % of the total electricity
cost for the methanol plant [17]. So, from the industrial point of view,
moderate pressure is economically favorable, but the performances of
the Cu-based catalysts are very poor in terms of selectivity and stability
at 30 bar and the temperatures below 277 ◦C [18–20]. Nakamura et al.
[21] reported a six-fold increase in Cu/ZnO catalyst activity compared
to the bare Cu surface for CO2 conversion to methanol. They also re-
ported that the exposed faces of Cu also play a pivotal role in CO2 hy-
drogenation [21]. The same group also reported that the stability and
activity of Cu-based ZnO catalyst are strongly associated with small
Ga2O3 particles, which led to the formation of Cu◦ and Cu+2 species for
the superior catalytic activity for CO2 reduction to methanol [22]. It is
also reported in the literature that H2O produced during the CO2 hy-
drogenation to methanol reaction accelerates the crystallization of Cu
and ZnO, leading to catalyst deactivation [23]. It is also reported that
high dispersion of Cu particles, which will be stable against sintering, is
necessary to present on the catalysts [24,25]. Mureddu et al. [26] re-
ported that ZnO plays an essential role in the Cu/ZnO catalyst because it
inhibits the agglomeration and sintering of Cu nanoparticles and also
leading a large surface area, which enhanced the activity of the catalyst
for methanol formation. The strong synergy between Cu and ZnO has
great inherent stability for CO2 hydrogenation. It is generally accepted
that the nanocrystalline metallic Cu species in catalysts are active phases
for the reduction of CO2 [27]. Researchers also reported that the con-
version of CO2 is affected by the metallic surface area, and the selectivity
of methanol highly depends on the dispersion of basic sites on the
catalyst surface [24,25]. So, the presence of high copper surface area
and basicity of the catalyst is the key factors for selecting the methanol
synthesis catalyst via CO2 hydrogenation [28]. Alkaline-earth oxides are
known as a solid base and also employed in different organic trans-
formation reactions as a promoter or catalyst [29]. Additionally, the
presence of alkaline earth oxide with the catalyst prevents agglomera-
tion and increases the surface area of the catalyst [30]. Dasireddy et al.
2. Experimental
2.1. Catalyst preparation
MgO promoted 5 wt% Cu on ZnO catalysts were synthesized by the
hydrothermal method. A typical synthesis process for 20 (wt.%), MgO
promoted Cu/ZnO catalyst is as follows: 1.8 g of low molecular weight
poly (diallydimethylammonium chloride) (PDADMAC), and 1.07 g
cetyltrimethylammonium bromide (CTAB) dissolved in 20 mL water by
vigorous stirring. Then 13.7 g Zn(NO3)2.6H2O and 6.3 g of Mg
(NO3)2.6H2O of the precursor salts were dissolved in 80 mL water, and
the mixing solution was added dropwise to the polymer solution. The
whole solution mixture was stirred continuously for 2 h, and 0.9 g Cu
(NO3)2⋅2.5H2O dissolved in 10 mL water was added to the final mixture
solution. Na2CO3 (2 M) solution was used as a precipitant and main-
tained the pH of the solution around 9–10 with vigorous stirring for 2 h
continuously. After complete precipitation, the resulting solution was
put into a Teflon-lined stainless-steel autoclave and treated for 24 h at
180 ◦C in an oven. After cooling down to room temperature, the product
obtained after filtration were sequentially washed with distilled water to
remove ions possibility of the remnant in the products. Finally, the
catalyst was dried at 110 ◦C overnight and further calcined at 450 ◦C in
the air for 6 h with a 1 ◦C/minute ramping rate. The catalyst is desig-
nated as CMZ-XHT, where X (5, 10, 15, and 20) is the weight % of MgO in
the catalyst.
2.2. Catalyst characterization
The catalyst was characterized by XRD, N2-physisorption, ICP-AES,
XPS, SEM, HR-TEM, TPR, Pulse Chemisorption, EXAFS, and N2O titra-
tion techniques. The details of characterization methods are given in the
supporting information (Table S1).
2.3. Catalytic activity test
The activity measurement of the CO2 to methanol transformation
reaction over the synthesized catalysts was carried out in a continuous
downflow fixed bed high-pressure reactor (Fig. 1), where typically 0.5 g
catalyst (40–60 mesh) was mixed with 1.0 g porcelain bead, which was
placed at the halfway point of a 7.92 mm stainless steel tube reactor
between two quartz wools. The temperature of the furnace was
2

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
Table 1
Physicochemical properties of different catalysts.
Catalyst
Surface
Metal
Cu
Cu particleb
size
Number
of the
active
Cu
areaa (m2/g)
Fresh Spent
(After
Dispersion
(MD; %)
Fresh Spent
(After
surface
area
dCu(nm)
S
Cu (m2
calcination)
gꢀ 1
)
atoms /g
calcination)
ZnOHT
CMHT
50.3
41.2
ND
NA
NA
7.3
NA
NA
9.1
ND
35.7
11.4
12.4
14.7
19.0
21.3
24.2
30.2
1.74 ×
1015
CZHT
52.8
54.3
57.8
60.4
64.2
43.2
46.8
50.4
57.7
62.6
13.3
16.6
18.8
22.5
29.5
9.8
7.2
5.6
4.9
4.7
3.5
1.58 ×
1016
CMZ-
5HT
13.7
15.8
19.3
29.1
5.82 ×
1016
CMZ-
10HT
CMZ-
15HT
CMZ-
20HT
6.78 ×
1016
7.92 ×
1016
Fig. 1. Schematic diagram of continuous fixed bed reactor set-up for methanol
3.06 ×
1017
production from CO2 hydrogenation.
NA: Not available; and ND- Not determined.
measured by a thermocouple fitted in the middle of the furnace, and
also, the temperature of the reactor was measured by the thermocouple
exposed to the thermowell at the bottom of the reactor (Supporting in-
formation). At the beginning of the reaction, the catalyst was initially
reduced (heating rate 2 ◦C/min) in the presence of 10 % H2 balanced He
a
N2 physisorption method.
b
N2O decomposition method using Cu metal dispersion.
◦
gas at 350 C for 2 h under atmospheric pressure condition. After the
is also presented in Table 1. CZHT catalyst shows the SCu of 14.7 m2/g,
and the Cu dispersion was 13.3 %. With the addition of MgO, dispersion
of Cu particles and copper surface area (SCu) increases, and CMZ-20HT
(MgO = 20 wt%) catalyst showed maximum SCu value of 30.2 m2/g. The
increase in copper surface area (SCu) with the addition MgO may be due
to the low density of MgO and the formation of small Cu species [33,34].
The variation of Cu dispersion (DCu) also showed a similar trend with the
reactor was cooled down to reaction temperature, the feed mixture (H2:
CO2: N2 = 3:1:1) with the total gas flow 60 mL/min) was introduced into
the reactor, maintaining the weight hourly space velocity (WHSV) of
7200 mL. gꢀca1t. hꢀ 1 and reaction pressure of 30 bar. Kinetic analysis of the
reaction was carried out between the temperature range of 200ꢀ 300 ◦C.
The products were analyzed using online GC (gas chromatography,
Agilent 7890A) fitted with two detectors thermal conductivity detector
(TCD) and flame ionization detector (FID) using porapack-Q column (for
analyzing H2, N2, CO, CO2, and H2O), and HP-Plot Q (for analyzing
CH3OH, CH4, and other hydrocarbons).
addition of MgO, and it is the inverse trend of the Cu particle size (dCu
)
measured by N2O titration. The maximum Cu dispersion value of 29.5 %
was obtained in the CMZ-20HT catalyst. The Cu-specific surface area and
Cu dispersion of the CMZ-20HT catalyst are higher compared to the other
catalysts (CMZ-15HT, CMZ-10HT, CMZ-5HT, CZHT, and CMHT). Several
studies also supported the observation of increasing surface area with
increasing metal loading [33–35].
The CO2 conversion and product selectivity for CH3OH, CH4, and CO
data were calculated by both internal standard normalization and mass-
balanced method. The carbon balanced and material balance was con-
ducted with an accuracy of ±3% in between 97%–103 %. The conver-
sion of CO2 (XCO2) and the selectivity of CH3OH (SCH3OH) were
determined based on the following equations:
3.1.2. Metal dispersion
The N2O decomposition analysis was carried out to calculate the Cu
dispersion, an average number of copper species present on the catalyst
surface, copper surface area, and the results are tabulated in Table 1.
From the analysis, it was observed that the dispersion of copper species
markedly depends on both the amount of magnesium loading and the
preparation method. The catalysts synthesized by the hydrothermal
method exhibited higher dispersions, presence of smaller copper parti-
cles with high copper surface area. The addition of CTAB as a
morphology controlling agent and PDADMAC as the size controlling
agent produces very small, highly dispersed copper particles supported
on nanocrystalline MgO-ZnO compared to the catalyst prepared in the
absence of MgO. Copper dispersion increases with increasing the MgO
loading over the Cu/ZnO catalyst. The CMHT, CZHT, CMZ-5HT, CMZ-
10HT, CMZ-15HT, and CMZ-20HT catalysts showed Cu dispersion values
of 11.4 %, 13.3 %, 16.6 %, 18.8 %, 22.5 % and 29.5 %, respectively. The
number of active particles presents estimated from metal dispersion
analysis is reported in Table1. It was found that the specific surface area
(64.2 m2/g) was highest for CMZ-20HT, and the number of active Cu
atoms present on the catalyst is 3.06 × 1017 Cu atoms/gcat. The number
of active Cu atoms present on other catalyst are 1.58 × 1016, 5.82 ×
F
, in ꢀ F
CO2
CO2, out
XCO
=
× 100%
(3)
2
F
CO2, in
F
CH3OH, out
SCH
=
× 100%
(4)
3OH
F
,in ꢀ F
CO2
CO2,out
3. Result & discussion
3.1. Characterization
3.1.1. Physicochemical properties and N2 physiosorption study
The estimation of Cu was examined by ICP-AES, and the results are
given in Table S2, supporting information. BET surface area analysis by
N2 adsorption-desorption of the prepared catalysts is also provided in
Table 1. The surface area of the support ZnO (ZnOHT) synthesized by the
hydrothermal method in the presence of CTAB and PDADMAC showed a
surface area value of 50.3 m2/g. The surface areas of the Cu/ZnOHT
catalyst prepared by the hydrothermal method were 52.8 m2/g
(Table 1). With the addition of MgO, Cu particles size decreases and
dispersion increases, so surface to volume ratio of the catalyst increases,
and the catalyst showed higher surface area. The specific surface area of
the CMZHT catalyst was increased with increasing MgO loading, and it
follows the order: CMZ-20HT > CMZ-15HT > CMZ-10HT > CMZ-5HT
(Table 1). Copper surface area (SCu) determined by N2O titration method
1016, 6.78 × 1016 and 7.92 × 1016 Cu atoms/gcat, respectively for CZHT
,
CMZ-5HT, CMZ-10HT, CMZ-15HT. The number of active Cu atoms in-
creases with increasing MgO loading, and the number of active Cu atoms
present on the CMZ-20HT catalyst is higher (3.06 × 1017 Cu atoms/gcat
)
compare to the other catalysts. We have also measured the Cu dispersion
3

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
using N2O titration method of the spent Cu/MgO/ZnO catalysts. N2O
titration method of the spent CMZ-20HT catalyst was found that Cu
dispersion (29.1 %) was almost same as that of the fresh CMZ-20HT
catalyst which confirms that the Cu particles present in the catalyst are
very stable during CO2 hydrogenation reaction. On the other hand, spent
CMHT, CZHT, CMZ-5HT, CMZ-10HT and CMZ-15HT catalysts showed lower
Cu dispersion (7.3, 9.8, 13.7, 15.8 and 19.3 %) compare to the fresh
CMZ-20HT catalyst and this may be the reason for the deactivation of this
catalyst during CO2 hydrogenation reaction (Table 1).
of metallic Cu particles. XRD peak at 2θ value of 50.45 are corre-
sponding to metallic Cu (JCPDS Card no. 89-2838). The absence of XRD
peaks for any copper species for CMZ-20HT catalyst indicates that most
probably highly dispersed Cu-species (< 5 nm) [36,37] are present, and
the size of the Cu species also remains unaffected during catalysis. The
observation also revealed that the interaction of the metal with support
present in the CMZ-20HT catalyst also played a vital role in preventing
the agglomeration of the Cu-species and maintained its size to < 5 nm
during the catalysis.
3.1.3. X-ray diffraction (XRD)
3.1.4. SEM analysis
Crystallinity of the prepared catalysts was analyzed by powder XRD,
where all the synthesized samples showed very high crystallinity. The
CZHT and CMHT catalysts showed the characteristic peaks for ZnO and
MgO, respectively, and also a peak at 38.72◦, which is assigned for CuO
(JCPDS Card no. 89-5896). CMZ-5HT, CMZ-10HT, and CMZ-15HT cata-
lysts also showed the characteristic peaks for CuO peaks at 38.72◦ for
CuO species using Scherrer’s equation (Table S3, supporting informa-
tion), whereas it was also found that the fresh CMZ-20HT catalyst did not
show any peak of Cu-species, indicating that the catalyst might contain
very small Cu-species (< 5 nm), which was not detected by XRD
(Fig. 2Af) [36,37]. The observation also indicates that CuO species have
a strong interaction with ZnO crystals in the presence of MgO, which led
to the high dispersion of Cu throughout the catalyst surface, forming
very small Cu-species [38,39].
The morphology of the prepared CMZ-20HT catalyst was carried out
by SEM analysis. As shown in Fig. 3A, the CMZ-20HT catalyst showed
platelet likes morphology, which can be seen on the surface of micro-
spherical species. The catalyst showed
a uniform distributed
morphology with a particle size between 20ꢀ 70 nm. The observations
also indicated the strong metal-support interaction between Cu-
nanoparticles on supported MgO/ZnO catalyst. The morphology of the
spent catalysts was also examined by the SEM analysis, which is shown
in Fig. 3B. The spent CMZ-20HT catalysts exhibited almost unchanged
morphology after reaction catalysts, which showed thermal stability.
3.1.5. TEM analysis
Morphological properties of the synthesized catalysts were carried
out by doing TEM analysis, and Fig. 4 represents the TEM images of fresh
and used (spent) CMZ-20HT catalysts. Fig. 4a–e show the overall
morphological image of the fresh CMZ-XHT catalyst. The fresh CMZ-20HT
catalyst showing the size of the CuO particles ~ 5 nm. The lattice fringes
with d-spacing values of 2.32 Å and 2.47 Å, corresponding to CuO (111)
and ZnO (101) planes, respectively, can also be seen in Fig. 4e. It is clear
from the TEM images that the morphology of the spent catalyst is almost
the same as that of the fresh catalyst, and the lattice fringes with d-
spacing values of 2.32 Å for CuO (111) and 2.08 Å for metallic Cu (111)
planes for the spent catalysts can also be seen in Fig. 5e. The presence of
both metallic Cu and CuO species in the fresh and spent CMZ-20HT
catalysts also supported by the XPS and EXAFS analysis (discussed later).
The observation from TEM analysis revealed that the CMZ-20HT catalyst
is very stable against sintering, and the Cu particle size remains ~ 5 nm
during CO2 hydrogenation. On the other hand, the Cu catalyst prepared
hydrothermally on supported ZnO (CZHT) showed agglomerated parti-
cles for the fresh catalyst as well as for the spent catalysts (Fig. 5a). This
could be due to larger Cu-species particles with lower metal-support
interactions, and these Cu particles are not stable under reaction tem-
perature and pressure [41,42]. The TEM images of the different fresh
With the increasing of MgO loading, it was observed that the in-
tensity of the XRD peaks for MgO was increasing, and it was also
influencing the intensity of ZnO peaks. It was revealed that the addition
of a small amount of MgO in the catalyst influencing the exposed planes
of the ZnO crystals. The peak at a 2θ value of 34.44◦ for ZnO in the
synthesized CMZ-XHT catalysts is shifted to a higher value. It was also
observed that the peak at a 2θ value of 34.39◦ for ZnO in the CZHT
catalyst shows the normal value as the JCPDS value. So, in the case of
CMZ-XHT catalysts, ZnO lattice size decreased (shown in Table S2,
Supporting information) as Mg atoms partially replaced Zn atoms in the
ZnO crystals [40]. We observed that there is a formation of a solid so-
lution between ZnO and MgO, which highly influences the catalytic
activity, showing very high methanol selectivity for the CMZ-XHT cata-
lysts. The unit cell parameter value ‘a’ of the ZnO lattice decreases from
normal value 0.3507 nm to 0.3505 nm after substitution of Zn2+ ion
(rZn2+ = 0.088 nm) by small Mg2+ ion (rMg2+ = 0.086 nm), which is in
agreement with the Vegard’s rule.
XRD patterns of all the spent catalysts are shown in Fig. 2B, where
except the CMZ-20HT catalyst, all other catalysts indicated the presence
Fig. 2. Powder XRD patterns of (A) fresh (After calcination) and (B) Spent catalysts.
4

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
Fig. 3. SEM images (A) fresh (After calcination), (B) spent and (C) EDS analysis of CMZ-20HT catalyst.
and spent catalysts (CMZ-5HT, CMZ-10HT, and CMZ-15HT) exhibited
mostly the agglomerated particles, as shown in Fig. 5 (5b, 5c and 5d).
The Cu dispersion on nanocrystalline MgO/ZnO was determined by the
elemental mapping of nanostructured catalysts, indicating a homoge-
nous dispersion of Cu on the nanocrystalline MgO/ZnO support (Fig. S1,
supporting information).
the dependence of the methanol production rate of all the catalysts on
copper particle size is shown in Fig. 6B, where a high methanol pro-
duction rate was achieved with smaller Cu particles. The CZHT
,
CMZ-5HT, CMZ-10HT, CMZ-15HT, CMZ-20HT catalysts showed the
methanol production rates of 0.0033, 0.0045, 0.0050, 0.0052 and
0.0063 mol gꢀCu1 hꢀ 1, and the Cu particles size was 7.2, 5.6, 4.9, 4.7 and
3.5 nm respectively.
3.1.6. Structure-activity correlations
In general, the properties of low-index (110) and (100) planes, cor-
ners, edges are influenced by Cu particle size, which in turn influences
both surface structure and electronic properties [47]. Low-index facets
get primarily exposed by the surface of larger particles, with fewer edge
or defect sites [48]. On the other hand, a larger number of open planes,
edge/defect sites with coordinately unsaturated atoms are present in
smaller particles. Hence, they are more reactive than fully coordinated
species. We believe that in our case, the activity of the Cu/ZnO catalyst
promoted with MgO is superior since low coordinated Cu-atom sites of
smaller nanoparticles are present. These sites stabilize HCOO, H2COO,
H2CO, key intermediates species of the CO2 reduction process. This kind
of interaction and stabilization of these species lowered the activation
energy barrier for the hydrogenation step. Smaller Cu particles will also
have more interfacial area with supported metal oxide, indicating that
metal-support interaction may also play a crucial role during catalysis.
Structure-activity correlations were established by using the steady-
state activity (time on stream 3 h) data. It is evident that the activity is
strongly correlated with the copper surface area and Cu particle size
(shown in Fig. 6A and B). It was observed that catalysts with high Cu
surface area as well as high Cu dispersion exhibited with very high
methanol production rate. Cu surface area and dispersion followed the
order: CMZ-20HT (S.A. = 30.2 m2/g, dispersion = 29.5) > CMZ-15HT (S.
A. = 24.2 m2/g, dispersion = 22.5) > CMZ-10HT (S.A. = 21.3 m2/g,
dispersion = 18.8) > CMZ-5HT (S.A. = 19.0 m2/g, dispersion = 16.6) >
CZHT (S.A. = 14.7 m2/g, dispersion = 13.3) > CMHT (S.A. = 12.4 m2/g,
dispersion = 11.4 and the methanol production rate also followed the
same order: CMZ-20HT (0.0063 mol g hꢀ 1) > CMZ-15HT (0.0052 mol
ꢀ 1
Cu
Cu
ꢀ 1
ꢀ 1
ꢀ 1
gꢀ 1
h
) > CMZ-10HT (0.0050 mol g hꢀ 1) > CMZ-5HT (0.0045 mol g
Cu
Cu
hꢀ 1) > CZHT (0.0033 mol g hꢀ 1) as shown in Fig. 6A. Earlier re-
ꢀ 1
Cu
searchers also observed a similar linear correlation between Cu surface
area and activity over different metal oxide support [43–46]. In addtion,
5

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
Fig. 4. TEM images of fresh (After calcination) (a) CZHT, (b) CMZ-5HT, (c) CMZ-10HT, (d) CMZ-15HT, and (e) CMZ-20HT catalysts.
Fig. 5. TEM images of spent (a) CZHT, (b) CMZ-5HT, (c) CMZ-10HT, (d) CMZ-15HT, and (e) CMZ-20HT catalysts.
3.1.7. H2-TPR analysis
the synthesized catalysts [49,50]. Tmax for different catalysts follows the
Reducibility of the synthesized catalysts was tested by H2-TPR
analysis (Fig. 7). All the catalysts showed similar kinds of reduction
patterns with the differences in reduction temperatures (Tmax). Cu-oxide
species in the CMZ-20HT catalyst were reduced at the lowest tempera-
order: CMZ-20HT (Tmax: =220 ◦C) < CMZ-15HT (Tmax: =232 ◦C) <
CMZ-10HT (Tmax: =237 ◦C) < CMZ-5HT (Tmax: =242 ◦C) < CZHT (Tmax
:
=251 ◦C) < CMHT (Tmax: =268 ◦C), so the Cu particle size also follows a
similar order [51]. It can be observed that all the catalysts showed board
Cu reduction peaks, and these reduction peaks could be due to the
presence of larger Cu-oxide nanoparticles with different sizes at the
◦
ture (Tmax =220 C), indicating that the catalyst has the smallest Cu-
oxide particles that exist on the surface of the ZnO support amongst
6

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
Fig. 6. (A) Correlations between methanol formation rate and copper surface area and, (B) Correlations between methanol formation rate and Cu particle size over
the catalysts.
during catalysis, and the activity decreases with time for all the catalysts
except CMZ-20HT. It was also found that the surface area of the spent
catalyst also decreases over all the CMZ catalyst except CMZ-20HT
,
indicating the sintering of the catalyst. So, in the case of CMZ-20HT, SMSI
plays a very vital role in stabilizing the very small Cu particles, and the
interface between the Cu-particles and Mg-incorporated ZnO support
(confirm from XRD analysis) facilitates the formation of reactive
intermediates.
3.1.8. XPS measurement
X-ray photoelectron spectroscopic (XPS) analysis was carried out to
examine the chemical state or oxidation state of Cu elements present on
the fresh and spent catalysts. Both fresh and spent CMZ-5HT, CMZ-10HT
,
CMZ-15HT, and CMZ-20HT catalysts were analyzed, and comparative
analysis of CZHT catalyst was also carried out (Fig. 8A and B). Fig. 8A
exhibited the core level Cu2p XPS spectra of the fresh CZHT catalyst.
Cu2p3/2 XPS peaks with binding energy values of 933.6 eV indicated
CuO, and the binding energy value of 934.4 eV attributed to Cu(OH)2 on
the surfaces of fresh CZHT catalyst. However, the Fresh CMZ-20HT
catalyst showed characteristic peaks for Cu-species, and after deconvo-
lution, two peaks at binding energy values of 932.7 eV and 933.8 eV
were observed. The first peak is due to metallic Cu or Cu2O, as both the
species show very close binding energies [54,55]. EXAFS analysis of the
fresh CMZ-20HT confirms the presence of metallic Cu and Cu2O
(Table 2). Deconvolution of the XPS of spent CMZ-20HT catalyst also
showed two peaks at 932.2 eV and 933.3 eV, and these peaks are
assigned for metallic Cu and CuO species, respectively. For comparison,
XPS analysis of CZHT catalysts (fresh and spent) was also carried out, and
it showed that the spent catalyst contains three types of Cu-species as
Cu2O (932.7 eV), CuO (933.6 eV), and Cu(OH)2 (934.4 eV). For spent
catalysts, the Cu 2p3/2 and Cu 2p1/2 peaks are followed by extended
shake-up satellite appearance at 942 and 962 eV confirmed the presence
of Cu2+ species [56]. These satellites were due to the charge transfer
between the transition metal 3d and surrounding ligand oxygen 2p or-
bitals, and the absence of satellites in the spectra confirmed that no
considerable quantity of Cu2+ species is left on the surface [56]. The Cu
Fig. 7. H2-TPR patterns of fresh (After calcination) catalysts.
surface or encapsulated Cu-nanoparticles present in bulk [49], as the
reduction of ZnO [52], and MgO is not possible by H2 at this temperature
range [53]. The Cu peak of CMZ-20HT catalyst showed single reduction
peaks at 220 ◦C, which revealed that the catalyst has smaller and well
dispersed Cu-oxide particles, which are easily reducible. As the catalyst
(CMZ-20HT) does not contain large Cu particles (confirmed by HRTEM,
high metal dispersion, and STEM elemental mapping analysis in Fig. S1),
it indicates the presence of strong metal-support interaction (SMSI). Due
to the presence of SMSI, the interface of the small active Cu and nano-
crystalline ZnO-MgO support binds the reactive intermediates very
strongly and reduced the activation energy barrier and shows superior
activity. SMSI is the main driving force to keep the Cu particle size intact
during catalysis by preventing against sintering in the presence of H2,
and the catalyst does not deactivate even after the 120 h time-on-stream
test. It is also believed that the synergy between very small
Cu-nanoparticles and nanocrystalline ZnO-MgO support also favours the
high activity of the catalyst. The absence of SMSI or the presence of very
XPS spectra of the higher Cu loading fresh catalysts (CMZ-5HT
,
CMZ-10HT, and CMZ-15HT) exhibited the presence of CuO and Cu(OH)2,
whereas the XPS spectra of spent catalysts showed the presence of
metallic Cu and CuO and Cu(OH)2 species, as shown in Fig. 8.
weak metal support interaction over other catalysts (CZHT, CMZ-5HT
,
CMZ-10HT, and CMZ-15HT) makes the Cu particles vulnerable for sin-
tering (agglomeration) during catalysis in the presence of H2 (Fig. 7). It
has to be noted that the agglomeration of the metal nanoparticles is
energetically favoured due to minimizing the surface areas by saturating
the binding and co-ordination sites. It was found that the Cu particle size
increases as agglomeration taking place (confirmed by TEM analysis)
3.1.9. EXAFS analysis
EXAFS spectra of fresh and spent CMZ-20HT catalysts are shown in
Fig. 9A and B, respectively, and the curve fitting results of both catalysts
are summarized in Table 2. For the fresh catalyst (Fig. 9A), EXAFS
–
spectra exhibited the presence of Cu O bond length of 0.1962 ± 0.0019
7

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
Fig. 8. XPS analysis of fresh (After calcination) (A) and (B) spent catalysts.
Table 2
Summary of fitted results for Cu k-edge EXAFS analysis.
Catalysts
Path
R
CN
DW(10ꢀ 5 nm2)
Δk
ΔR (10ꢀ
ΔE0
Rf
(10ꢀ 1nm)
(10 nmꢀ 1
)
1 nm)
(eV)
(%)
CMZ-20HT
fresh#
Cu-O
1.962 ± 0.019
2.641 ± 0.046
1.905 ± 0.037
2.554 ± 0.037
2.1 ± 0.5
6.5 ± 1.0
0.6 ± 0.4
10.0 ± 3.0
2.6 ± 2.2
26.7 ± 3.9
1.0 (fixed)
14.5 ± 2.1
2.0 – 11.0 1.2– 2.7 -1.4 ± 2.9 2.43
Cu-Cu
Cu-O
CMZ-20HT
spent
2.0 – 12.0 1.2– 2.7 -9.9 ± 6.1 3.99
Cu-Cu
*1: The reducing factor for Cu-O S20 was assumed as 0.95. The reducing factor for metallic Cu phase S20 was 0.694, determined by fitting with Cu foil data. #After
calcination.
Fig. 9. k3-weighted Fourier transforms of Cu k-edge EXAFS for the (A) fresh (After calcination) CMZ-20HT and (B) spent (After reaction) CMZ-20HT catalyst.
Imaginary and amplitude part is traced by dotted and solid curves, respectively. Ascertained data are indicated with solid lines, and fitting data are indicated with
dotted lines.
nm with the co-ordination number of 2.1 ± 0.5, which indicates the
4. Results and discussions
–
presence of Cu (II) species. The Cu Cu bond length of 0.2641 ± 0.0046
with a co-ordination number of 6.5 ± 1.0 confirms the presence of
4.1. Catalyst activity
metallic Cu species. These results also supported by the XPS analysis,
EXAFS analysis for the spent catalysts showed the Cu O bond length of
The activity of the CMZ-20HT catalyst was measured at different re-
action temperatures. The effect of temperature on CO2 conversion,
methanol selectivity, and yield was tested in the range of 200ꢀ 300 ◦C,
and it was found that CMZ-20HT catalyst showed a continuous increase
of CO2 conversion with an increase in temperature. CMHT and CZHT
catalyst also showed conversion of CO2 at 200 ◦C with 3.5 %, 5.5 % CO2,
and 66 %, 72 % methanol selectivity, respectively. CMZ-20HT catalyst
showed the 8.7 % CO2 conversion with the methanol selectivity of ~100
%. From the reaction results, it is evident that the presence of both MgO
and ZnO played a significant role in CO2 activation and higher methanol
–
0.1905 ± 0.0037 nm with a coordination number of 0.6 ± 0.4 indicates
–
the presence of Cu (II) species, whereas Cu Cu bond length was 0.2554
± 0.0037 with the coordination number of 10.0 ± 3.0 confirms the
presence of metallic copper species (Fig. 9B). EXAFS analysis also con-
firms that the Cu species is stable during the CO2 hydrogenation
reaction.
8

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
selectivity. The surface area of CZHT and CMZ-20HT is 52.8 m2/g and
64.2 m2/g, respectively, and from XRD, it was observed that the size of
Cu particles of CZHT is much larger (> 5 nm) than the size of Cu particles
(~ 5 nm) in CMZ-20HT. Metal dispersion analysis of CMZ-20HT catalyst
also revealed the presence of Cu particles with size ~ 3.5 nm and for
CZHT catalyst, it is ~ 7.2 nm. The influence of Cu crystallite size in-
dicates that the CO2 transformation to methanol reaction over MgO
promoted Cu-ZnO is structure sensitive. The addition of Mg increases the
strain in the ZnO lattice is also responsible for the higher activity. The
increase in rates with smaller Cu crystallites may be attributed to the
higher number of exposed crystal planes, edges/corners/defect sites
present in the catalyst, which contains co-ordinately unsaturated atoms,
which are responsible for lowering the activation energy by strongly
binding the reactive intermediates. The incorporation of MgO improved
the Cu dispersion of the CMZ-20HT catalyst. However, it did not increase
the total surface area of the catalyst, which is the key factor in showing
the high activity of the catalyst. The Cu surface area measured by the
N2O decomposition method was 30.2 m2/g for the CMZ-20HT catalyst,
which is very high compared to the other catalysts (Table 1). The
number of active Cu atoms presents in the CMZ-20HT catalyst (3.06 ×
1017 Cu atoms/gcat) is also very high compared to the other catalyst
(Table 1). With the presence of a very high Cu surface area and a number
of active Cu atoms on the surface of CMZ-20HT, this catalyst showed very
good catalytic activity for CO2 hydrogenation to methanol trans-
formation with very high methanol selectivity and high catalyst stabil-
ity. On the other hand, the low basicity of ZnO and the presence of larger
Cu particles of CZHT is the reason for showing low activity of the CZHT
catalyst. The addition of MgO to the Cu-ZnO catalyst improved the basic
nature of the catalyst, where adsorption of CO2 molecules increased. It is
also reported that the interface between Cu and ZnO plays the most
significant role for methanol formation from CO2 over the Cu-ZnO
catalyst [7,57–59]. In the case of CMZ-20HT catalyst, the presence of
smaller Cu particles with high Cu dispersion and the presence of
metal-support interaction increases CO2 conversion and methanol
selectivity compared to other catalysts. The effect of temperatures over
other CMZ-HT (CMZ-5HT, CMZ-10HT, and CMZ-15HT) catalysts are shown
in Fig. 10. It was observed that the conversion of carbon dioxide in-
creases with increasing temperatures, as shown in Fig. 9A. It was also
noticed that with increasing temperature, the CO selectivity increases,
as shown in Fig. 9C. All the catalysts (CMZ-5HT, CMZ-10HT, CMZ-15HT,
and CMZ-20HT) showed a higher CO formation with increasing the
temperature due to the reverse water-gas-shift reaction (RWGS). It was
observed that in the case of the CMZ-20HT catalyst, there was no CO
formation at 200 ◦C (Fig. 9C). So, we conclude that at low temperatures,
the selectivity for methanol is higher. Sun et al. reported that the for-
mation of higher CO as a by-product via reverse water gas shift reaction
(RWGS) led to the moderate deactivation of the catalyst as CO acts as a
reducing agent [60] Due to the higher reducing potential of CO, it can
reduce the surface of ZnO supported particles. Cu is an active metal for
promoting methanol production from CO2, but the size of the copper
particles plays an important role, which affects the catalyst activity [61,
62]. As the size of the active species increases, the reducibility decreases
[49,50], which directly influences the methanol selectivity. For
CMZ-20HT catalyst, TEM images show the presence of Cu particles with
size ~ 5 nm, but the Cu particle size calculated from metal dispersion
analysis is ~3.5 nm. The difference in the active species particle sizes
can be explained by the fact that in TEM images, only a portion of the
catalyst is analyzed, and very small Cu particles were not detected by the
technique. It was found that the strong metal-support interaction (SMSI)
is present (confirmed by H2-TPR) in the case of CMZ-20HT, which resist
the very small Cu-particles against sintering during catalysis, whereas
the absence of SMSI for the other catalysts (CMZ-5HT, CMZ-10HT, and
CMZ-15HT) make the Cu particles easy to sinter in the presence of H2
during catalysis.
Fig. 10. Catalytic activity catalysts (A) CO2 conversion, (B) CH3OH selectivity, (C) CO selectivity, and (D) CH3OH yield. Reaction Conditions: Temperature (200 ◦C),
Pressure (30 bar), WHSV (7200 mL.gꢀ 1. hꢀ 1), Feed ratio (H2: CO2: N2 = 3:1:1).
9

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
Time-on-stream study of the CMZ-20HT catalyst is shown in Fig. 11,
where this catalyst does not show any deactivation, whereas the CZHT
catalysts deactivate very rapidly. The stability of the CMZ-20HT catalyst
arises due to the presence of strong metal-support interaction and the
presence of small Cu particles with very high dispersion over ZnO sup-
port in the presence of MgO, which helps to inhibit the sintering of the
catalyst, showing no deactivation. Whereas the absence of strong metal
supports interaction, the presence of big Cu particles with low dispersion
favours sintering for the CZHT catalyst, showing deactivation of the
catalyst. CZHT catalyst loses almost 80 % activity after 120 h time-on-
stream due to the sintering of Cu-particles during catalysis. TEM im-
ages and EXAFS analysis also support the fact that the size of Cu-
particles remains constant for the CMZ-20HT, whereas the particle size
increases (agglomeration took place) in the case of the CZHT catalyst. In
general, the deactivation of the copper-based catalysts occurred due to
Cu sintering, decrease in catalytic reducibility, presence of excess sur-
face hydroxyls, and absence of strong metal-support interaction. It was
also found that the stability of the catalyst depends on active particle
size, presence of high metal dispersion with a high surface area of the
support (which also improves the dispersion of Cu-species). It is clear
that the CMZ-20HT catalyst is more stable during CO2 hydrogenation
compare to CMZ-5HT, CMZ-10HT, and CMZ-15HT catalysts. The reason
for the deactivation of these catalysts may be due to the sintering in Cu-
species due to the absence of strong metal-support interaction.
Fig. 12. The DFT optimized geometry of the active catalyst surfaces; (a) MgO
(100); (b) ZnO(100). Color code: Zn (grey), O (red), Mg (green).
a small Mg6O7 cluster at the Cu13-ZnO(100) surface at the Cu-ZnO
interface, as shown in Fig. 13(c). The Cu13/MgO/ZnO(100) surface
model (Fig. 13(c)) represents the interface between the Cu particle and
MgO/ZnO support. To represent the large Cu cluster, a Cu-strip/ZnO
(100) model was created by adsorbing a two-layer of Cu-strip over the
ZnO(100) surface, as shown in Fig. 13(d). The Cu-strip retains its overall
geometry even after the adsorption at the ZnO(100) surface, as shown in
Fig. 13(d).
4.1.1. DFT results
To understand the promotional effect of MgO in the Cu/ZnO catalyst,
DFT calculations were performed to calculate the CO2 adsorption energy
over different model surfaces of the catalyst. The DFT optimized ge-
ometry of the ZnO(100) and MgO(100) surfaces were shown in Fig. 12.
The Cu13/ZnO(100) catalyst surface is obtained by adsorbing a Cu13
nanocluster over the ZnO(100) surface, as shown in Fig. 13(a). The ge-
ometry optimized Cu13/ZnO(100) surface shows a strong interaction
between the ZnO(100) surface and the Cu13 nanoparticle. Similarly, the
Cu13/MgO(100) catalyst surface is obtained by adsorbing a Cu13 nano-
cluster over the MgO(100) surface, as shown in Fig. 13(b). The
adsorption energy of the Cu13 nanoparticle over the MgO(100) surface
was found to be -2.1 eV, which is 0.6 eV higher compared to the ZnO
(100) surface. Two more models of active catalyst surfaces, Cu13/MgO/
ZnO(100) and Cu-strip/ZnO(100), were shown in Figs. 13(c) and 11 (d),
respectively. The Cu13/MgO/ZnO(100) surface was obtained by adding
Adsorption of CO2 is studied over all the catalytically active model
surfaces described above using the DFT method. The ZnO(100) surface
was found to be not active for the CO2 activation as the CO2 does not get
adsorbed at the ZnO(100) surface, as shown in Fig. 14(a). The CO2 only
gets physically adsorbed at a distance of 3.6 Å from the surface. Whereas
the MgO(100) surface actively binds the CO2 molecule forming a strong
Mg-O(CO) bond (2.81 Å), as can be seen in Fig. 14(b). The adsorption
energy of the CO2 molecule to the MgO(100) surface was calculated to
be - 0. 38 eV.
The interface of the Cu13/ZnO(100) catalyst surface was found to be
active for CO2 activation, as can be seen from the CO2 adsorption ge-
ometry in Fig. 15(a). The CO2 molecule adsorbs strongly at the Cu13 and
ZnO(100) interface, forming bonds to the ZnO surface and Cu13 nano-
cluster. The Zn-O, Cu-C, and Cu-O bonds were measured to be 2.04 Å,
2.05 Å, and 2.05 Å, respectively. The adsorption energy of CO2 at the
Cu13/ZnO(100) catalyst surface was calculated to be -0.69 eV. Similar to
the Cu13/ZnO(100) catalyst surface, the Cu13/MgO(100) catalyst sur-
face also activates the CO2 by adsorbing at the Cu13/MgO interface, as
shown in Fig. 15(b). The CO2 molecule forms bonds to both the MgO
(100) surface and the Cu13 nanocluster. The Mg-O, Cu-C, and Cu-O
bonds were measured to be 2.12 Å, 2.05 Å, and 2.08 Å, respectively.
The adsorption energy of CO2 at the Cu13/MgO(100) catalyst surface
was calculated to be -1.25 eV, which is 0.56 eV stronger compared to the
Cu13/ZnO(100) surface. The stronger adsorption energy at the Cu-MgO
interface compared to the Cu-ZnO interface will enhance the CO2
adsorption, as shown in the CO2 chemisorption experiment. The basic
nature of the MgO support can be the probable reason for the higher CO2
adsorption at the Cu-MgO interface compared to the Cu-ZnO interface.
To understand the promoting effect of MgO on the CO2 adsorption Cu13
/
MgO/ZnO(100) surface model is studied as shown in Fig. 15(c). The
addition of the Mg6O7 cluster at the Cu13/ZnO interface drastically
enhance the CO2 binding from -0.69 eV in Cu13/ZnO(100) to -1.04 eV in
Cu13/MgO/ZnO(100), indicating the promotional effect of MgO in the
CO2 conversion over conventional Cu/ZnO catalysts, as also observed
experimentally. At the Cu13/MgO/ZnO(100) surface, the CO2 binds to
the Cu/MgO interface forming Mg-O (2.11 Å) and Cu-C (2.05 Å) bonds,
Fig. 11. Time on stream (TOS) results for CO2 hydrogenation to methanol.
Reaction Condition: Weight of Catalyst (0.5 g), Temperature (200 ◦C), Pressure
(30 bar), WHSV (7200 mL.gꢀ 1. hꢀ 1), Feed ratio (H2: CO2: N2 = 3:1:1).
10

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
Fig. 13. The DFT optimized geometry of the active catalyst surfaces, (a) Cu13-MgO(100); (b) Cu13-ZnO(100); (c) Cu13-MgO-ZnO(100) and (d) Cu-strip-ZnO(100).
Color code: Zn (grey), O (red), Mg (green), Cu (orange), C (black).
Fig. 14. The DFT optimized geometry of the CO2 adsorption at the active catalyst surfaces; (a) MgO(100) and (b) ZnO(100). Color code: Zn (grey), O (red), Mg
(green), Cu (orange), C (black).
as shown in Fig. 15(c). CO2 adsorption study at the Cu-strip/ZnO(100)
catalyst surface, representing the larger Cu particles at the ZnO sup-
port, show no chemical adsorption of CO2 (Cu-C bond distance ~ 3.78
Å), as shown in Fig. 15(d), indicating the low activity of the larger Cu
particles compared to the small metal cluster. This is also in accordance
with the experimental findings where catalysts with high dispersion and
small particle size showed higher CO2 to methanol conversion.
5. Conclusion
MgO promoted Cu/ZnO catalysts prepared by hydrothermal method
using cetyltrimethylammonium bromide as morphology controlling
agent and poly (diallydimethylammonium chloride) as a structure-
directing agent produce small Cu particles with high Cu surface area,
and dispersion, which directly influence the catalytic performances. The
addition of basic MgO decreased the surface density of Cu during syn-
thesis, which increases the dispersion of Cu species and also favours the
adsorption of CO2 and 20 % MgO loading showed superior catalytic
11

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
Fig. 15. The DFT optimized geometry of the CO2 adsorption at the active catalyst surfaces; (a) Cu13/ZnO(100); (b) Cu13/MgO(100); (c) Cu13/MgO/ZnO(100) and (d)
Cu-strip/ZnO(100). Color code: Zn (grey), O (red), Mg (green), Cu (orange), C (black).
activity, high stability, and > 99 % selectivity. XRD confirms that the
addition of MgO decreases the lattice parameters of ZnO, where some Zn
atom is replaced by Mg atom in the ZnO lattice, which favours the for-
mation of reactive intermediates. H2-TPR also shows that the metal
support interaction is present for 20 wt% MgO promoted Cu/ZnO (CMZ-
20HT) catalyst, which favours the stability of active Cu particles against
sintering. The average Cu particle size was < 5 nm, and the presence of
highly disperse surface-active Cu sites led to maximum CO2 conversion
of 8.7 % at 200 ◦C with > 99 % methanol selectivity, and the catalyst
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgments
SKS acknowledges the Petrotech Society of India for providing
research fellowship. CS and RB acknowledge Petrotech Society of India
for providing support for a collaborative research project between
Bharat Petroleum Corporation Ltd. (BPCL) and CSIR-IIP, Dehradun. CS
acknowledges higher management of BPCL, Corporate research, and
development center for granting permission to carry out this collabo-
rative research work. B.P acknowledges SERB-DS-T for research funding
(RJF/2020/000042). The XAFS measurements were performed at KEK-
IMSS-PF with the approval of the Photon Factory Advisory Committee
(project 2017G190). The Director, CSIR-IIP, is acknowledged for his
help and encouragement. The authors thank the Analytical Science Di-
vision, Indian Institute of Petroleum, for analytical services.
showed the methanol production rate of 0.0063 mol g hꢀ 1. With
ꢀ 1
Cu
◦
increasing temperature, CO2 conversion reached to 16.0 % at 300 C
with 62 % methanol selectivity. CMZ-20HT catalyst is highly stable
during CO2 hydrogenation, and the catalyst did not show any deacti-
vation even after 120 h time-on-stream. DFT calculation showed that the
adsorption energy of CO2 at the Cu13/MgO/ZnO(100) catalyst surface
was calculated to be -1.04 eV, which is 0.35 eV stronger compared to the
Cu13/ZnO(100) surface. The stronger adsorption energy at the Cu/MgO/
ZnO interface compared to the Cu-ZnO interface will enhance the CO2
adsorption, which in turn increases the CO2 conversion over the Cu/
MgO/ZnO catalyst. The activity and high methanol selectivity of the
CMZ-20HT catalyst is the combination of the presence of large Cu surface
area, very small Cu particles with high Cu dispersion, and the synergistic
interaction between small Cu particles and ZnO support in the presence
of MgO.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2021.118239.
CRediT authorship contribution statement
References
Sachin Kumar Sharma: Design and performing experiment of the
research work, manuscript writing and correction. Tuhin Suvra Khan:
Design and performing DFT studies, manuscript writing and correction.
Rajib Kumar Singha: manuscript writing and correction. Bappi Paul:
manuscript writing and correction. Mukesh Kumar Poddar: catalyst
characterization, manuscript writing and correction. Takehiko Sasaki:
catalyst characterization, manuscript writing and correction. Ankur
Bordoloi: Manuscript writing and correction. Chanchal Samanta:
Manuscript writing and correction. Shelaka Gupta: Design DFT studies,
manuscript writing and correction. Rajaram Bal: Conceptualization and
Design the whole study, design of experiment of the research work,
manuscript writing and correction.
[1] P. Gao, S. Li, X. Bu, S. Dang, Z. Liu, H. Wang, L. Zhong, M. Qiu, C. Yang, J. Cai, Nat.
Chem. 9 (2017) 1019–1024.
[2] B.M. Tackett, E. Gomez, J.G. Chen, Nat. Catal. 2 (2019) 381–386.
[3] O. Martin, A.J. Martín, C. Mondelli, S. Mitchell, T.F. Segawa, R. Hauert,
´
´
C. Drouilly, D. Curulla-Ferre, J. Perez-Ramírez, Angew. Chem. Int. Ed. 55 (2016)
6261–6265.
[4] G. Bonura, M. Cordaro, C. Cannilla, F. Arena, F. Frusteri, Appl. Catal. B: Environ.
152 (2014) 152–161.
[5] H. Ban, C. Li, K. Asami, K. Fujimoto, Catal. Commun. 54 (2014) 50–54.
[6] X. Guo, D. Mao, G. Lu, S. Wang, G. Wu, J. Catal. 271 (2010) 178–185.
[7] B. Hu, Y. Yin, G. Liu, S. Chen, X. Hong, S.C.E. Tsang, J. Catal. 359 (2018) 17–26.
´
´
[8] R. Ladera, F.J. Perez-Alonso, J.M. Gonzalez-Carballo, M. Ojeda, S. Rojas, J.L.
G. Fierro, Appl. Catal. B: Environ. 142 (2013) 241–248.
[9] F. Arena, G. Mezzatesta, G. Zafarana, G. Trunfio, F. Frusteri, L. Spadaro, J. Catal.
300 (2013) 141–151.
12

S.K. Sharma et al.
Applied Catalysis A, General 623 (2021) 118239
[10] J. Wang, G. Li, Z. Li, C. Tang, Z. Feng, H. An, H. Liu, T. Liu, C. Li, Sci. Adv. 3
(2017), e1701290.
[38] A. Scarabello, D. Dalle Nogare, P. Canu, R. Lanza, Appl. Catal. B: Environ. 174
(2015) 308–322.
[11] C.H. Bartholomew, Appl. Catal. A 212 (2001) 17–60.
[12] M.V. Twigg, M.S. Spencer, Top. Catal. 22 (2003) 191–203.
[13] H. Lei, R. Nie, G. Wu, Z. Hou, Fuel 154 (2015) 161–166.
[14] C. Tisseraud, C. Comminges, T. Belin, H. Ahouari, A. Soualah, Y. Pouilloux, A. Le
Valant, J. Catal. 330 (2015) 533–544.
[39] V.A. Kondratenko, C. Berger-Karin, E.V. Kondratenko, ACS Catal. 4 (2014)
3136–3144.
[40] M. Rouchdi, E. Salmani, B. Fares, N. Hassanain, A. Mzerd, Res. Phys. 7 (2017)
620–627.
[41] S. Li, Y. Xu, Y. Chen, W. Li, L. Lin, M. Li, Y. Deng, X. Wang, B. Ge, C. Yang, Angew.
Chem. Int. Ed. 56 (2017) 10761–10765.
[15] W.-C. Liu, J. Baek, G.A. Somorjai, Top. Catal. 61 (2018) 530–541.
[16] I. Ganesh, Renewable Sustainable Energy Rev. 31 (2014) 221–257.
[42] W. Li, G. Zhang, X. Jiang, Y. Liu, J. Zhu, F. Ding, Z. Liu, X. Guo, C. Song, ACS Catal.
9 (2019) 2739–2751.
´
¨
[17] M. Perez-Fortes, J.C. Schoneberger, A. Boulamanti, E. Tzimas, Appl. Energy 161
(2016) 718–732.
[43] G. Chinchen, K. Waugh, D. Whan, Appl. Catal. 25 (1986) 101–107.
[44] W. Pan, R. Cao, D. Roberts, G. Griffin, J. Catal. 114 (1988) 440–446.
[18] F. Liao, Y. Huang, J. Ge, W. Zheng, K. Tedsree, P. Collier, X. Hong, S.C. Tsang,
Angew. Chem. Int. Ed. 50 (2011) 2162–2165.
´
[45] C. Baltes, S. Vukojevic, F. Schüth, J. Catal. 258 (2008) 334–344.
[19] C. Li, X. Yuan, K. Fujimoto, Appl. Catal. A 469 (2014) 306–311.
[20] M.D. Porosoff, B. Yan, J.G. Chen, Energy Environ. Sci. 9 (2016) 62–73.
[21] J. Nakamura, I. Nakamura, T. Uchijima, T. Watanabe, T. Fujitani, Model studies of
methanol synthesis on copper catalysts. Stud. Surf. Sci. Catal., Elsevier, 1996,
pp. 1389–1399.
[46] S. Natesakhawat, J.W. Lekse, J.P. Baltrus, P.R. Ohodnicki Jr., B.H. Howard,
X. Deng, C. Matranga, ACS Catal. 2 (2012) 1667–1676.
[47] D. Uzio, G. Berhault, Catal. Rev. 52 (2010) 106–131.
[48] R. Rioux, H. Song, J. Hoefelmeyer, P. Yang, G. Somorjai, J. Phys. Chem. B 109
(2005) 2192–2202.
́
[22] T. Fujitani, I. Nakamura, Bull. Chem. Soc. J. 75 (2002) 1393–1398.
[23] J. Wu, M. Saito, M. Takeuchi, T. Watanabe, Appl. Catal. A 218 (2001) 235–240.
[24] L. Wang, L. Yang, Y. Zhang, W. Ding, S. Chen, W. Fang, Y. Yang, Fuel Process.
Technol. 91 (2010) 723–728.
[49] M. Zabilskiy, P. Djinovic, E. Tchernychova, O.P. Tkachenko, L.M. Kustov, A. Pintar,
ACS Catal. 5 (2015) 5357–5365.
`
[50] L. Pino, A. Vita, F. Cipitì, M. Lagana, V. Recupero, Appl. Catal. B: Environ. 104
(2011) 64–73.
[25] M. Mureddu, I. Ferino, A. Musinu, A. Ardu, E. Rombi, M.G. Cutrufello, P. Deiana,
M. Fantauzzi, C. Cannas, J. Mater. Chem. A 2 (2014) 19396–19406.
[26] M. Mureddu, F. Ferrara, A. Pettinau, Appl. Catal. B: Environ. 258 (2019), 117941.
[51] R.K. Singha, A. Shukla, A. Yadav, S. Sain, C. Pendem, L.S.K. Konathala, R. Bal, Mol.
Catal. 432 (2017) 131–143.
[52] S. Kuld, C. Conradsen, P.G. Moses, I. Chorkendorff, J. Sehested, Angew. Chem. Int.
Ed. 53 (2014) 5941–5945.
ˇ
ˇ
ˇ ˇ
[27] M. Hus, V.D. Dasireddy, N.S. Stefancic, B. Likozar, Appl. Catal. B: Environ. 207
´
(2017) 267–278.
[53] I. Melian-Cabrera, M.L. Granados, J. Fierro, J. Catal. 210 (2002) 273–284.
[28] X. Guo, D. Mao, G. Lu, S. Wang, G. Wu, J. Mol. Catal. A: Chem. 345 (2011) 60–68.
[29] W. Zhang, H. Wang, W. Wei, Y. Sun, J. Mol. Catal. A: Chem. 231 (2005) 83–88.
[54] B. Sarkar, C. Pendem, L.S. Konathala, R. Tiwari, T. Sasaki, R. Bal, Chem. Commun.
50 (2014) 9707–9710.
´
[30] L. Madeira, R. Martın-Aranda, F. Maldonado-Hodar, J. Fierro, M. Portela, J. Catal.
[55] N. Siddqui, B. Sarkar, C. Pendem, L.S. Konthala, T. Sasaki, A. Bordoloi, R. Bal,
Catal. Sci. Technol. 7 (2017) 2828–2837.
169 (1997) 469–479.
ˇ
ˇ
[31] V.D. Dasireddy, S.S. Neja, L. Blaz, J. CO2 Util. 28 (2018) 189–199.
[32] L. Xu, H. Song, L. Chou, Appl. Catal. B: Environ. 108 (2011) 177–190.
[33] R.K. Singha, A. Yadav, A. Agrawal, A. Shukla, S. Adak, T. Sasaki, R. Bal, Appl.
Catal. B: Environ. 191 (2016) 165–178.
[56] M.N. Barroso, M.F. Gomez, J.A. Gamboa, L.A. Arrúa, M.C. Abello, J. Phys. Chem.
Solids 67 (2006) 1583–1589.
[57] M.M.-J. Li, Z. Zeng, F. Liao, X. Hong, S.C.E. Tsang, J. Catal. 343 (2016) 157–167.
[58] F. Liao, Y. Huang, J. Ge, W. Zheng, K. Tedsree, P. Collier, X. Hong, S.C. Tsang,
Angew. Chem. 123 (2011) 2210–2213.
[34] V. Tiwari, J. Jiang, V. Sethi, P. Biswas, Appl. Catal. A 345 (2008) 241–246.
[35] H. Ren, C.-H. Xu, H.-Y. Zhao, Y.-X. Wang, J. Liu, J.-Y. Liu, J. Ind. Eng. Chem. 28
(2015) 261–267.
[59] Y. Sun, C. Huang, L. Chen, Y. Zhang, M. Fu, J. Wu, D. Ye, J. CO2 Util. 37 (2020)
55–64.
[36] G. Delahay, D. Valade, A. Guzman-Vargas, B. Coq, Appl. Catal. B: Environ. 55
(2005) 149–155.
[60] J.T. Sun, I.S. Metcalfe, M. Sahibzada, Ind. Eng. Chem. Res. 38 (1999) 3868–3872.
[61] C. Liu, B. Yang, E. Tyo, S. Seifert, J. DeBartolo, B. von Issendorff, P. Zapol, S. Vajda,
L.A. Curtiss, J. Am. Chem. Soc. 137 (2015) 8676–8679.
[37] B.M. Abu-Zied, W. Schwieger, A. Unger, Appl. Catal. B: Environ. 84 (2008)
277–288.
[62] G.K. Ramesha, J.F. Brennecke, P.V. Kamat, ACS Catal. 4 (2014) 3249–3254.
13