New and revisited insights into the promotion of methanol synthesis catalysts by CO2

Article abstract of DOI:10.1039/c3cy00573a

CO hydrogenation, CO₂ hydrogenation, and water-gas shift (WGS) reactions have been simultaneously investigated over industry-like catalysts based on Cu-ZnO-Al₂O₃, under methanol synthesis conditions (513 K, 5.0 MPa). For this, a novel methodology has been applied: the concentration of carbon dioxide in the syngas feed was consecutively increased (R = CO₂:(CO + CO₂) = 0-100) resulting in a volcano-type plot of the rate of methanol formation and forming a hysteresis loop when decreasing the CO₂ concentration again. H₂O co-feeding experiments revealed that the enhancement of activity can be correlated with the WGS activity linking both hydrogenation paths of CO and CO₂. On the other hand, excessive amounts of surface hydroxyls seem to inhibit methanol production, explaining the drop in activity at high CO₂ concentrations. An investigation of the catalytic performance was accompanied by an extensive characterisation of the fresh and used catalytic materials by X-ray diffraction, temperature-programmed reduction by H₂, N ₂O pulse chemisorption, X-ray photoelectron spectroscopy, and Auger electron spectroscopy. It was shown that the copper surface area affects the CO₂ hydrogenation; however, this parameter is unambiguously not the key descriptor for CO₂-promoted methanol synthesis, which is a consequence of the synergistic interaction of zinc oxide and copper. This structural feature is further promoted by Al₂O₃ through stabilisation of the surface. The position of the activity maximum is determined by the surface ratio Cu:Zn. The hysteresis behaviour is a result of the continuous decrease of Cu dispersion and the fixation of copper species in its monovalent oxidation state, both detrimental for CO₂ hydrogenation. CO hydrogenation is strongly affected by the Cu:Zn bulk ratio and thus the reducibility of the catalyst. These facts could be substantiated by the use of impregnated model catalysts. The Royal Society of Chemistry.

Full text of DOI:10.1039/c3cy00573a

View Article Online
View Journal
Catalysis
Science&Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: O. Martin and J. Pérez-Ramírez, Catal.
Sci. Technol., 2013, DOI: 10.1039/C3CY00573A.
This is an Accepted Manuscript, which has been through the RSC Publishing peer
review process and has been accepted for publication.
Catalysis
Science&Technology
Accepted Manuscripts are published online shortly after acceptance, which is prior
www.rsc.org/catalysis
Volume 1 | Number 1 | 2011 | Pages 0001–0100
to technical editing, formatting and proof reading. This free service from RSC
Publishing allows authors to make their results available to the community, in
citable form, before publication of the edited article. This Accepted Manuscript will
be replaced by the edited and formatted Advance Article as soon as this is available.
To cite this manuscript please use its permanent Digital Object Identifier (DOI®),
which is identical for all formats of publication.
More information about Accepted Manuscripts can be found in the
Information for Authors.
Please note that technical editing may introduce minor changes to the text and/or
graphics contained in the manuscript submitted by the author(s) which may alter
content, and that the standard Terms & Conditions and the ethical guidelines
that apply to the journal are still applicable. In no event shall the RSC be held
responsible for any errors or omissions in these Accepted Manuscript manuscripts or
any consequences arising from the use of any information contained in them.
ISSN 2044-4753
www.rsc.org/catalysis
Registered Charity Number 207890

Page 1 of 10
Catalysis Science & Technology
Dynamic Article Links ►
Journal Name
View Article Online
DOI: 10.1039/C3CY00573A
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
New (revisited) insights into the promotion of methanol synthesis
catalysts by CO2
Oliver Martina and Javier Pérez-Ramírez*a
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
5
DOI: 10.1039/b000000x
CO hydrogenation, CO2 hydrogenation, and waterꢀgas shift (WGS) reactions have been simultaneously
investigated over industrialꢀlike catalysts based on CuꢀZnOꢀAl2O3 under methanol synthesis conditions
(513 K, 5.0 MPa). For this, a novel methodology has been applied: the concentration of carbon dioxide in
the syngas feed was consecutively increased (R = CO2:(CO+CO2) = 0–100) resulting in a volcanoꢀtype
10 plot of the rate of methanol formation and forming a hysteresis loop when decreasing the CO2
concentration again. H2O coꢀfeeding experiments revealed that the enhancement of activity can be
correlated with the WGS activity linking both hydrogenation paths of CO and CO2. On the other hand,
excessive amounts of surface hydroxyls seem to inhibit methanol production explaining the drop in
activity at high CO2 concentrations. The investigation of the catalytic performance was accompanied by
15 an extensive characterisation of the fresh and used catalytic materials by Xꢀray diffraction, temperatureꢀ
programmed reduction by H2, N2O pulse chemisorption, Xꢀray photoelectron spectroscopy, and Auger
electron spectroscopy. It was unveiled that the copper surface area affects the CO2 hydrogenation;
however, this parameter is unambiguously not the key descriptor for CO2ꢀpromoted methanol synthesis,
which is a consequence of the synergistic interaction of zinc oxide and copper. This structural feature is
20 further promoted by Al2O3 through stabilization of the surface. The position of the activity maximum is
determined by the surface ratio Cu:Zn. The hysteresis behaviour is a result of the continuous decrease of
Cu dispersion and the fixation of copper species in its monovalent oxidation state, both detrimental for
CO2 hydrogenation. CO hydrogenation is strongly affected by the Cu:Zn bulk ratio and thus the
reducibility of the catalyst. These facts could be substantiated by the use of impregnated model catalysts.
25 1. Introduction
CO+2H2 ꢀ CH3OH
ꢁRH493 K = −97.4 kJ mol−1 (1)
ꢁRH493 K = −57.4 kJ mol−1 (2)
ꢁRH493 K = −40.0 kJ mol−1 (3)
Methanol (MeOH) belongs to the top ten petrochemicals in the
CO2+3H2 ꢀ CH3OH+H2O
world.1,2 It is further seen as a promising bridging technology for
replacing fuels or serving as a chemical feedstock besides oil
since syngas, which is needed for the production of methanol, can
CO+H2O ꢀ CO2+H2
50 equilibrium towards an unfavourable regime.6 However,
operating at pure CO conditions does not result in high methanol
yields due to kinetic reasons.7 A certain amount of CO2
(commonly (H2−CO2):(CO+CO2) ~ 2) is crucial in the feed in
order to reach a reasonable conversion.8 The promotional effect
55 by carbon dioxide has been demonstrated in several studies and
has been speculatively assigned to the WGS reaction linking both
hydrogenation routes.7ꢀ9 Strong adsorption of reactants,8,10,11
inhibition of CO hydrogenation by CO2,12 or the dependence on
the oxygen coverage13 have been controversially discussed in
60 literature as reason for the drop in activity at high CO2
concentrations. The investigations of relationships between the
structural (surface) properties of the catalyst and the CO2
promotion are even more speculative.
30 be obtained from different sources, e.g. coal or biomass.3
Additionally, methanol synthesis offers currently the only
possibility of utilising CO2 emissions on large scale.4 In the
standard industrial process, a combined feed of CO, H2, and CO2
is converted into methanol over CuꢀZnOꢀAl2O3 catalysts
35 (Cu:Zn:Al ~ 6:3:1) at temperatures of 493–573 K and total
pressures of 5–10 MPa offering a selectivity higher than 99.9%.1
Accordingly, the equilibria (eqn. (1–3)), i.e. CO and CO2
hydrogenation and waterꢀgas shift (WGS), are required for
describing the methanol synthesis.5
40
The most relevant limitation in the commercial process is the
thermodynamic one and thus high concentrations of CO2 push the
a Institute for Chemical and Bioengineering, Department of Chemistry
and Applied Biosciences, ETH Zurich, WolfgangꢀPauliꢀStrasse 10,
45 CHꢀ8093 Zurich, Switzerland. Fax: +41 44 633 14 05; Eꢀmail:
Extensive studies have been conducted in methanol synthesis
65 tackling primarily how to obtain catalysts with high Cu surface
area and how to stabilise such surfaces preventing sintering of the
jpr@chem.ethz.ch
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

Catalysis Science & Technology
Page 2 of 10
View Article Online
DOI: 10.1039/C3CY00573A
copper nanoparticles,14 the main reason for deactivation.1,2,15ꢀ17 In
contrast, the research aiming at understanding the CuꢀZnOꢀAl2O3
catalyst remains highly contentious due to material and pressure
Cu(NO3)2ꢂ3 H2O (SigmaꢀAldrich, 98%), Zn(NO3)2ꢂ6 H2O (Acros
Organics, 98%) and/or Al(NO3)3ꢂ9 H2O (Alfa Aesar, 98%), in the
desired molar ratio reaching a final pH 8.0 of the slurry. The
gaps usually existing in those studies: model catalysts have been 60 precipitates were directly filtered without ageing and
5
mostly used and the reaction has been operated at pressure
conditions far below 5 MPa. Overcoming both deficiencies is
crucial since the CuꢀZnOꢀAl2O3 system changes dynamically in
structure depending on the process conditions, especially on the
subsequently the filter cake was washed by deionised H2O until
pH 7.0 was reached in the filtrate. Finally, it was dried at 338 K
overnight. Ternary and binary catalyst compositions, i.e. Cuꢀ
ZnOꢀAl2O3 and accordingly CuꢀZnO, CuꢀAl2O3, ZnOꢀAl2O3,
syngas composition.15,18ꢀ20 This fact entails that static kinetic 65 have been synthesised corresponding to this method and are
10 models are not sufficiently characterising the nature of the
methanol synthesis mechanism.21 Monitoring these alterations
becomes consequently essential since it was proven for industrialꢀ
like catalysts that methanol synthesis is sensitive towards
denoted expressing their metal composition, e.g. CuZnAl for Cuꢀ
ZnOꢀAl2O3. Suffix figures refer to the molar ratio of Cu and Zn
in the bulk, e.g. CuZnꢀ2 for CuꢀZnO with ratio Cu:Zn = 2.
For impregnation, neutral γꢀAl2O3 (MP Biomedicals, 99%)
different facets of copper.22 This valuable result was supported by 70 was dispersed in deionised water together with the metal
15 density functional theory calculations within the same study
revealing in addition that Cu surfaces, decorated by Zn atoms,
offer reaction pathways of lower energy for both CO and CO2
hydrogenation (vide Liu and coꢀworkers23). The role of ZnO in
precursors, Cu(CH3COO)2ꢂH2O and/or Zn(CH3COO)2ꢂ2 H2O
(both from SigmaꢀAldrich, 99%). The stirred slurry was fed to a
Büchi mini spray dryer Bꢀ290 applying a liquid flow rate of
5.3 cm3 STP min−1,
a
flow rate of spraying air of
the CO2 promotion effect has not been unravelled in this work, 75 450 cm3 STP min−1 and a flow rate of heated air stream of
20 nevertheless.
Klier et al.8 investigated in 1982, for the first time, CO2ꢀ
0.6 m3 STP min−1 and heating the inlet to 493 K. All asꢀ
synthesised materials were finally calcined at 573 K for 2 h in
static air, ramping by 2 K min−1.
promoted methanol synthesis over a CuꢀZnO catalyst by
consecutively changing the concentration of carbon dioxide in the
syngas feed from R = CO2:COx = 0 to 100 (COx = CO+CO2),
25 which evidenced a huge increase of carbon oxide conversion for
low contents of CO2 compared to pure feeds of CO/H2 or CO2/H2,
respectively. This approach, following the catalyst history
depending on R, has never again been seized extensively except
in a few kinetic modelling studies.12,13,24ꢀ27 However, none of
30 these studies attempted to correlate the catalytic performance
with the structure of the catalyst or went beyond the speculations
by Klier et al.8 who claimed the ratio R determines the redox
properties of the copper surface.
We have found that this experimental approach discloses
35 worthwhile insights when altering the composition of the catalyst
and subsequently conducting complementary changes of R,
resulting in a reversible hysteresis phenomenon. It consequently
offers the possibility to tackle the complexity of the methanol
synthesis process, i.e. three interconnected equilibria proceeding
40 over the ternary catalyst CuꢀZnOꢀAl2O3. H2O coꢀfeeding
experiments revealed the mechanistic nature of CO2ꢀpromoted
methanol synthesis. From the application of binary catalyst
compositions (i.e. CuꢀZnO, CuꢀAl2O3, and ZnOꢀAl2O3) we could
gain valuable structureꢀperformance relationships: ZnO is not
45 solely a good disperser of copper nanoparticles, but it is required
in Cuꢀbased catalysts in order to observe the promotional effect
by CO2. This synergistic interaction of Cu and ZnO has been
further investigated by varying the Cu:Zn ratio and by applying
impregnated model catalysts.
2.2. Characterisation
80 All catalysts were characterised by powder Xꢀray diffraction
(XRD) at any stage of the preparation, i.e. asꢀsynthesised,
calcined, reduced, and after the catalytic test, on an X’Pert Pro
MPD from PANalytical utilising CuꢀKα radiation (λ = 0.1542 nm)
with an angular step size of 0.033° 2θ and a counting time of 8 s
85 per step. Particle sizes of reduced copper were estimated from
Cu(1 1 1) reflection using the Scherrer equation. No other
crystalline phases than Cu metallic or ZnO were detected by
XRD after reduction or reaction. N2 isotherms were obtained at
77 K on a Quadrasorb SI Quantachrome instrument, calculating
90 the total surface area by BET method. Samples were degassed
before the measurement by heating them up to 473 K in vacuum
for 3 h.
The metal content (Cu, Zn, and Al) of the calcined samples
was determined by ICPꢀOES (inductively coupled plasma –
95 optical emission spectroscopy) on a Horiba Ultima2 applying an
Ar plasma. Each sample was previously heated to 373 K in a
mixture of aqua regia and HF (volumetric ratio = 1:6) for
completely dissolving the catalysts. Both, temperatureꢀ
programmed reduction by H2 (H2ꢀTPR) and N2O pulse
100 chemisorption, were run on a Thermo TPDRO 1100 setꢀup
equipped with a TCD (thermal conductivity detector). Prior to
analysis, the catalysts were dried in He at 373 K for 30 min.
Subsequently, the H2ꢀTPR experiments were conducted ramping
from 323 to 673 K by 2 K min−1 in 5 molꢀ% H2 in N2 at
105 atmospheric pressure. The copper surface area was identified by
pulsing N2O at 363 K over the catalyst that was dried and reduced
under H2ꢀTPR conditions (vide supra), however, levelling the
temperature off at 503 K for 2 h.28
50 2. Materials and Methods
2.1. Catalyst preparation
Xꢀray photoelectron spectroscopy (XPS) and Auger electron
110 spectroscopy (AES) were performed on a VGꢀMicrotech Multilab
3000 spectrometer featuring a hemispheric electron analyser with
9 channeltrons and nonꢀmonochromatised MgꢀKα radiation at
1253.6 eV. The used catalyst was cooled down to room
temperature under nitrogen atmosphere before transferring the
All catalysts have been synthesised either via coꢀprecipitation, the
standard industrial approach of preparing methanol synthesis
catalysts,1 or impregnation. Coꢀprecipitation was conducted by
55 mixing two 1 M solutions containing either Na2CO3 (Acros
Organics, 99.8%) or the corresponding mixture of metal nitrates,
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 2

Page 3 of 10
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C3CY00573A
sample under ultraꢀhigh vacuum conditions (residual pressure of
as CuZnAl (Fig. 1). The parameter R = CO2:COx expresses the
CO2 concentration in the syngas feed. The experiment is named
CO or CO2 cycle when starting the alteration of R from 0 or 100,
ca. 5 10–7 Pa) into the analysis chamber of the spectrometer. The
spectra were collected at a pass energy of 50 eV. The intensities
were estimated by calculating the integral peak, after subtracting 60 respectively. For a clear assignment, the branch to be referred to
5
the Sꢀshaped background, and by fitting the experimental curve to
a linear combination of Lorentzian and Gaussian functions scaled
by a ratio of 30:70. The binding energy scale was referenced to
the C 1s level of the carbon overlayer at 284.6 eV. 933.6 eV and
932.1 eV have been used as binding energies to distinguish Cu2+
at
Table 1 Catalytic materials that have been synthesised und tested within
this study.
Denotation
Method SCud / m2 g−1 Molar Cu:Zn:Ale
10 and Cured, the latter species combines copper in reduced state, i.e.
a
CuZnAl or CP
CuZn or CuZnꢀ2
CuZnꢀ0.5
CuZnꢀ10
CuAl
ZnAl
SD
Cu@Zn
15.6±0.5
13.7±0.4
3.8±0.1
5.8±0.2
5.4±0.2
–
6.0:2.8:1.4
6.0:2.4:0
6.0:12.4:0
6.0:0.6:0
6.0:0:1.4
0:3.0:1.5
Cu+ and Cu0, which cannot be distinct on the basis of XPS.29
a
a
a
a
a
b
c
2.3. Catalytic evaluation
A homeꢀmade fixedꢀbed reactor setꢀup was built tackling
industrial methanol synthesis conditions, i.e. T ≤ 573 K,
15 P ≤ 10.0 MPa. For liquid injection, an HPLC (highꢀperformance
liquid chromatography) pump 307 (Gilson) was connected to the
heated part on top of the reactor. All gases were obtained from
Pangas/Linde with a purity ≥99.99995%. Online analysis of the
outlet gas stream was carried out by gas chromatography (GC,
2.2±0.1
1.1±0.1
6.0:3.1:163
3.0:6.1:187
a Coꢀprecipitation. b Simultaneously sprayꢀdried Cu(CH3COO)2 and
65 Zn(CH3COO)2 on Al2O3. c Cu(CH3COO)2 sprayꢀdried on ZnO/Al2O3.
d Determined by N2O pulse chemisorption on the fresh catalysts. e ICPꢀ
OES.
20 Agilent 7890A), using
a GSꢀCARBONPLOT column for
80
separating CO, CO2, N2, and H2 coupled to a TCD and a DBꢀ1
column to analyse all organic compounds by a flame ionisation
detector (FID).
(a)
3
1
60
Methanol synthesis could potentially form hotspots in the
25 catalyst bed due to its elevated exothermic nature. However, we
have not diluted the bed since Villa et al.30 have calculated the
external heat transfer to be as low as ~1 K. It was anticipated that
the internal heat transfer is even lower for the small particles
applied (125–300 ꢃm sieved fraction) corroborated by the fact
30 that no difference between the temperature of the oven and the
bed could be determined. The activation of the catalysts was done
in situ and consisted of a drying step at 433 K (2 K min−1) for 30
min in a flow of N2 and of a reduction step at 503 K (2 K min−1)
for 2 h in 5 molꢀ% H2 in N2 at 0.5 MPa total pressure, referred to
35 as fresh sample. Finally, desired reaction conditions were applied;
typically 513 K, 5.0 MPa total pressure, H2:COx = 7,
WHSV = 7.5 h−1. These conditions were used to ensure working
below thermodynamic limitation. N2 was used as internal
standard at a fixed mole fraction of x(N2) = 0.28. Waterꢀgas shift
40 activity was conducted at same conditions feeding a gas stream of
x(CO) = 0.09, H2O:CO = 1–5, GHSV = 16,000 h−1, N2ꢀbalanced.
After each reaction, the catalyst was passivated by flushing with
2 molꢀ% O2 in N2 at 323 K for 1 h.
2
CO cycle
4
40
20
60
40
20
0
1
2
CO2 cycle
0
10
100
R = CO2:COx / –
30
20
10
0
15
10
5
Carbon balance was determined for each experiment to be less
45 than 5% in deviation. All coꢀprecipitated catalysts performed
with a selectivity of methanol >99% in terms of organic products.
The only byꢀproduct, which was formed in the case of the sprayꢀ
dried catalysts, was determined to be dimethyl ether (DME). The
relative uncertainty of the formation rate of MeOH and DME has
50 been estimated from an error propagation calculation to be 6% on
average. All catalytic tests were reproduced within error range.
(b)
3. Results and discussion
0
Fresh CO/H2 0(1)
6(1)
100(1)
0(2) = R(branch)
3.1. The source of CO2 promoted methanol synthesis
CO cycle
The basic characterisation of the catalysts utilised in this study is
55 given in Table 1. CO2 promotion was initially studied over
conventionally coꢀprecipitated, ternary CuꢀZnOꢀAl2O3, denoted
Fig. 1 (a) Rate of methanol formation per mol of exposed Cu (rMeOH) over
70 CuZnAl at 513 K, 5.0 MPa, and H2:COx = 7: consecutively changing the
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

Catalysis Science & Technology
Page 4 of 10
View Article Online
DOI: 10.1039/C3CY00573A
CO2 concentration starting from R = 0 (circles, CO cycle) or R = 100
(diamonds, CO2 cycle). Branch 3 and 4 (top) refer to a second subsequent
CO cycle. (b) Corresponding evolution of copper surface areas (SCu, black
bars) and particle sizes of Cu(1 1 1) facet (dCu(1 1 1), white bars) along the
CO cycle. Additionally, the characterisation of the sample is shown which
latter aspect has been tested by starting the reaction at R = 0 and
consecutively varying the concentration of coꢀfed H2O steam
(Fig. 2), referred to as H2O cycle, akin to the CO cycle (Fig. 1a).
A promotional effect and a hysteresis were observed, too. This
5
has been operated at R = 0 (CO/H2) maintaining the same timeꢀonꢀstream 50 result corroborates the hypothesis that WGS is mechanistically
that was necessary to run one CO cycle (~60 h).
the source of CO2 promotion by converting carbon monoxide into
CO2, which is supposed to be the main source in methanol
synthesis.7,13,36 H2O plays consequently an ambivalent role as
80
dCu(1 1 1) / nm
CO2 in methanol synthesis: promotional when small
55 concentrations are
Fresh
8.7
60
40
20
0
H2O:CO = 0(1) 8.9
H2O:CO = 0(2) 10.7
1
80
2
Reꢀactivated
H2O cycle
CO cycle
0
0.2
0.4
0.6
Molar feed ratio H2O:CO / –
Fig. 2 H2O cycle over CuZnAl at 513 K, 5.0 MPa, H2:CO = 7: water
10 steam has been coꢀfed to the syngas and the H2O:CO ratio was
consecutively altered. Copper particle sizes are given for the fresh catalyst
and the catalyst before and after H2O coꢀfeeding.
400
420
440
460
480
500
T / K
a particular value of R is given additionally in brackets as
subscript.
Fig. 3 H2ꢀTPR profiles of CuZnAl recorded after one CO cycle (vide Fig.
1a, top) and after the H2O cycle (vide Fig. 2). Reꢀactivated refers to the
sample used in the following sequence: CO cycle ꢁ reꢀactivation ꢁ CO
60 cycle (vide Fig. 4a).
15
The observations that CO2 hydrogenation is favoured over CO
hydrogenation (compare R = 0(1) and 100(1) of CO and CO2 cycle,
respectively, in Fig. 1a) and that a maximum of methanol
production occurs by increasing or decreasing R coincide with the
results from literature.7,8,13,31 A novel result is the presence of a
20 hysteresis when the CO2 concentration is decreased again in the
CO cycle. Methanol synthesis catalysts are thus highly affected
by their history.20 This sensitivity was exploited to gather
performanceꢀmechanismꢀstructure relationships.
It is thus necessary to discuss the former dogma that copper
25 surface area is the key parameter of methanol synthesis
catalysts17 taking the several studies into account, which
indicated other aspects being relevant such as CuꢀZnO
interactions.32ꢀ35 First of all, CO hydrogenation and CO2
hydrogenation must be clearly distinct since MeOH production
30 from CO2 depends on the catalyst history whereas CO
hydrogenation obviously does not within the cycle experiment.
Copper surface area and particle size according to the Cu(1 1 1)
facet show the continuous decrease or increase, respectively,
monitored over the CO cycle (Fig. 1b). CO hydrogenation (R = 0)
35 results always in the same reaction rate independent of Cu
dispersion. In contrast, CO2 hydrogenation (R = 100) exhibits
only half of the initial rate when starting from CO or at the end of
the CO2 cycle. It appears to be therefore sensitive to the Cu
dispersion.
present and deactivating by excessive coꢀfeeding.37
One reason for the deactivation at high concentrations of H2O
or CO2 is probably that the increased coverage by hydroxyl
species inhibits the CO2 hydrogenation (vide Fig. SI1), e.g.
65 through blocking of active sites.13,38,39 A second point is the
irreversible loss of Cu surface area leading to an increased Cu
particle size. The sintering is amplified when H2O is introduced
in the feed (Fig. 2). This substantiates the conclusion that CO2
hydrogenation is sensitive towards Cu dispersion. Apesteguía et
70 al.40 have shown that also the WGS itself is enhanced by
increased dispersion of copper. This would in turn be one
explanation for the hysteresis behaviour.
The fact that a higher molar ratio of H2O:CO = 0.3 is
necessary for reaching the maximum activity, compared to
75 CO2:CO = 0.06 in the dry cycle, relies most probably on CO2
being a molecule that itself is quickly converted into methanol.
Additionally, the standard reduction potential of H2O is much
more negative than the one of CO231 and thus the strength of their
influence on the catalyst surface will differ, too; however, the
80 reducible surface sites of the catalyst are the same after both
experiments according to the H2ꢀTPR profiles (Fig. 3).
The correlation of WGS activity and CO2 promotion is
unequivocally demonstrated by Fig. 4: CuZnAl was reꢀcalcined
(2 molꢀ% O2 in N2, 0.5 MPa, 573 K, 2 K min−1, 2 h) and reꢀ
85 reduced before conducting WGS reaction or a second CO cycle,
respectively. Both MeOH synthesis performance (Fig. 4a) and
WGS activity (Fig. 4b) are significantly lowered after reꢀ
activating the catalyst. This is most likely to be caused by
morphological changes since 90% of copper surface area is lost
40
The effect of CO2 promotion cannot be correlated with the
dispersion of copper which follows a different trend than the
methanol production does along R. The enhancement of activity
must be therefore the consequence of (i) either the formation of
special catalytically active sites at particular syngas compositions
45 and/or (ii) the beneficial involvement of the WGS reaction. The
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 4

Page 5 of 10
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C3CY00573A
compared to the fresh catalyst. Waterꢀgas shift must be therefore
to CO/H2 converts all copper surface species into Cu metallic.
seen as an essential step in CO2ꢀpromoted methanol synthesis 30 CO2 hydrogenation conditions induce the formation of (partially)
rather than being a side reaction.
oxidised copper, Cu+ and Cu2+. Monovalent copper is also the
major species at the maximum of CO2 promotion (R = 6).
Interestingly, not all of the Cu+ species can be reꢀconverted to
metallic copper anymore as soon as the catalyst has been treated
35 by CO2 hydrogenation conditions; clearly demonstrated by AES
of the catalyst after the CO cycle (Fig. 5b, R = 0(2)). This finding
gives an explanation for the hysteresis behaviour in addition to
the drop of Cu surface area. The slight enhancement of Cu0 after
the cycle likely stems from the reduction of Cu2+ which was
40 formed along the cycle. The partially reꢀreduced Cu surface after
the first cycle
3.2. Evolution of copper surface upon reaction
5
The presence of hysteresis in the CO cycle revealed that the
deactivation effects cannot be solely a result of the mechanism,
i.e. the interplay between CO2 promotion and hydroxyl coverage.
It is also unlikely that the behaviour can be generally explained
by a nonꢀsteady state of the catalyst since the hysteresis is
10 preserved in the second CO cycle (Fig. 1a). It turned out that
these effects have
80
(a)
100
(a)
1
60
80
Zn
2
40
20
0
60
Reꢀactivation
40
20
0
Cu0
+
Cu+
3
4
Cu2+
0
10
R = CO2:COx /
100
Fresh CO/H2 CO2/H2 R = 6(1) 0(2)
0(4)
–
Reꢀ
activated
CO cycle
100
80
60
40
20
0
(b)
Cu+
Cu0
(b)
R = 0(2)
R = 6(1)
CO2/H2
CO/H2
Fresh
0
1
2
3
4
5
910
915
920
925
Molar feed ratio H2O:CO / –
Kinetic energy / eV
Fig. 4 (a) CO cycle profiles (conditions as before) and (b) waterꢀgas shift
activity at 513 K, 5.0 MPa, H2O:CO = 1−5 in terms of yield of carbon
15 dioxide (YCO2) over fresh CuZnAl (circles) and CuZnAl that was operated
for one CO cycle and subsequently reꢀactivated (diamonds), i.e. reꢀ
calcined and reꢀreduced.
Fig. 5 Monitoring of the surface changes of CuZnAl upon the CO cycle
(vide Fig. 4a) by XPS and AES: (a) surface composition regarding Zn,
45 Cu0/Cu+, and Cu2+, estimated from the deconvolution of the XPS spectra;
(b) corresponding Auger Cu LMM spectra. Both figures include the data
of the catalyst exposed to fixed syngas conditions (CO/H2 or CO2/H2,
respectively).
to be correlated with the composition of the catalyst surface. This
is indicated by the H2ꢀTPR profiles recorded after the
20 experiments (Fig. 3): the bimodal profile is the same after
conducting the CO or the H2O coꢀfeeding cycle, but it is
significantly changed for reꢀactivated CuZnAl. This alludes to the
assumption that the morphology41ꢀ43 and/or the redox properties
of surface Cu/ZnOx are significantly altered over reꢀactivation.
could then be the reason why branch 3 (second cycle) exceeds
50 branch 2 (first cycle).
The issue has been controversially debated for decades which
copper site is the active one in methanol synthesis. Monnier et
al.45 found a linear correlation of the amount of Cu+ in CuꢀCr2O3
catalysts and their activity for CO hydrogenation. They
55 demonstrated subsequently that this type of catalyst does not
exhibit a promotional effect by CO2 at 543 K and 5.2 MPa like
ZnOꢀcontaining catalysts.46 They concluded that ZnO and Cr2O3
25
In order to unveil the influence of the syngas composition on
the catalyst surface, XPS and AES have been carried out (Fig. 5).
The Cu species clearly respond to the syngas composition as also
stated by other authors:8,29,44 prolonged exposition of the catalyst
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 5

Catalysis Science & Technology
Page 6 of 10
View Article Online
DOI: 10.1039/C3CY00573A
possess a different ability to stabilise Cu+ species under reaction 45 promotional effect. Obviously, this structural interaction reveals
conditions: Cr2O3 locks the oxidation state of copper whereas
ZnO allows dynamic alteration of it.
the key for understanding methanol synthesis catalysts.
Table 2 Evolution of catalysts properties as a function of Cu:Zn bulk
CO2 hydrogenation proceeds over similar surface compositions
with different activities (compare Table SI1, entry 4 and 8). It is
therefore questionable whether Cu+ is an active site for this
reaction. Especially the required scission of the CꢀO bond might
occur preferentially on Cu metallic. In contrast, CO
hydrogenation results in a similar performance irrespective of the
ratio.
5
−1
SCub / m2 gCu
Fresh
Cu:Zna
SCu,relb,c / %
SZnO,reld / %
After cycle
0.5
2
10
15
36
38
85
64
62
16.0±0.5
22.5±0.7
6.6±0.2
5.9±0.2
10.2±0.3
1.6±0.1
10 alteration of the surface. We conclude from this result that CO is
converted to methanol over both Cu0 and Cu+, but monovalent
copper has no promotional effect on the CO hydrogenation itself.
Monovalent copper species might consequently only account for
the WGS
a
Determined by ICPꢀOES on the basis of the calcined materials.
50 b Analysed by N2O pulse chemisorption on the fresh catalyst. c Surface
ratios are referred to the BET N2 surface area as total area of the fresh
catalyst. d Estimated from the difference of BET N2 surface area and SCu
;
this approach is reliable since ZnO does not chemisorb oxygen
irreversibly from N2O.15
80
60
30
20
10
30
60
20
40
10
0
40
0
1
CuZnAl CuZn CuAl
20
1
1
10
20
Cu:Zn ratio / –
2
0
2
0
10
100
R = CO2:COx / –
1
15
Fig.
6 CO cycle over binary CuAl (conditions as before). The
40
20
corresponding ZnAl possesses only poor activity of approximately 0.1%
conversion of COx without indication of hysteresis behaviour. Inset:
copper surface areas of the ternary and the binary coꢀprecipitated catalysts
20 fresh (black) and after one CO cycle (white).
400
450
500
2
T / K
reaction in the CO2 promotion.
XPS also disclosed that reꢀactivating CuZnAl results in a
catalyst surface with a high fraction (~75%) of inactive Cu2+ (Fig.
5a). The fact that this Cu species cannot be converted into
25 (partially) reduced ones, despite of the strong reducing conditions
at R = 0, shows that morphological changes most likely occurred,
which also influenced the stabilisation of Cu in its different
oxidation states.
40
20
0
1
3.3. Impact of metal oxides on CO2 promotion
30 The monitored dynamics of copper oxidation states lead to the
assumption that a strong interaction with the metal oxides, ZnO
and Al2O3, is present in methanol synthesis catalysts and is
supposed to dictate the performance. This is unveiled on the basis
of extending the cycle approach towards binary catalyst
35 compositions, particularly CuꢀZnO (CuZn), CuꢀAl2O3 (CuAl),
and ZnOꢀAl2O3 (ZnAl). The low activity of CuAl (Fig. 6) and the
inactivity of ZnAl coincide with reported results.2 CO2 promotion
is only observed for the CuꢀZnO catalyst and is a consequence of
strong synergism between Cu and ZnO (Fig. 7). This is a clear
40 evidence of ZnO being an essential partner in CO2ꢀpromoted
methanol synthesis. The unique collaboration of Cu and ZnO was
already observed by other authors18,47,48 and has been
controversially discussed.17 However, we could demonstrate that
the CuꢀZnO synergy is necessary in order to observe the
2
0
10
R = CO2:COx / –
Fig. 7 CO cycle (conditions as before) of CuZn catalysts of different
Cu:Zn ratios = 0.5 (−ꢁ−), 2 ( ), and 10 (—). Insets: (top) CO
100
55
hydrogenation activity at R = 0(2) as a function of Cu:Zn ratio; (middle)
respective H2ꢀTPR profiles (signals normalised on catalyst mass).
60 Contradictory, Chinchen et al.49 concluded that the CuZnAl
catalyst is solely the sum of its components. Indeed, this is true in
terms of copper surface area (Fig. 6, inset): both metal oxides are
necessary to highly disperse copper nanoparticles. However, it
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 6

Page 7 of 10
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C3CY00573A
has been observed that a physical mixture of Cu/SiO2 and 45 strong reducing conditions (CO) accompanied by a migration of
ZnO/SiO2 is much more active than the sum of both single
components utilising CO/CO2/H2 mixtures, respectively.50 This
fact and the absence of CO2 promotion over CuAl elucidate that
there is a synergistic effect between Cu and ZnO, which has been
a matter of hot debate in literature.
It could be thus anticipated that the Cu:Zn ratio will influence
the performance of the catalyst. The position of the maximum
could be correlated with the Cu:Zn surface ratio: the higher the
Zn to the catalyst surface.18,60,61 Indeed, we could monitor such
surface enrichment of Zn (Fig. 5a) and a shift of the zinc Auger
signal at R = 0 (Fig. SI2) indicating a minor abundance of Zn0.
However, a clear proof for these sites was not found and their
50 impact on CO2ꢀpromoted MeOH synthesis remains therefore
speculative.
Another theory involves the oxygen vacancies on ZnO as
active sites.59,62,63 The result of CO2 inhibiting methanol
production over ZnOꢀAl2O3 was interpreted to be the reason why
5
10 bulk ratio, the greater the surface concentration of Cu (Table 2),
and the more the maximum is shifted towards greater R (Fig. 7); 55 a certain ratio R is needed for the reaction keeping the oxygen
in line with the higher Cu:Zn surface ratio of the ternary catalyst
(Fig. 5a) and its maximum at higher CO2 concentration (Fig. 1a).
The amount of Cu on the catalyst surface is levelled off at a ratio
15 of Cu:Zn ~ 40:60, comparing the samples Cu:Zn = 2 and 10
(Table
vacancies highly abundant through the reduction by CO. It
emerges thereof that electron deficient zinc sites could have a
significant impact on methanol synthesis. However, the inactivity
of ZnOꢀAl2O3 in our experiments leads us to the conclusion that
60 Zn might not participate directly in all reaction steps. Other
authors concluded that the oxygen vacancies on ZnO influence
the Cu/ZnOx interface energy and therefore the morphology of
the Cu nanoparticles.19,41
10
1
8
6
100
(a)
(b)
2
80
60
40
20
0
1
Al
Reꢀ
activated
4
2
0
2
After
CO cycle
Zn
Cu0/+
Fresh
10
R = CO:COx / –
0
100
Fresh
After
Reꢀ
1015 1020 1025 1030
Binding energy / eV
CO cycle activated
Fig. 8 CO cycle performance (conditions as before) over sprayꢀdried
catalysts as carbonꢀbased product formation: (i) simultaneously deposited
20 Cu and Zn precursor on Al2O3, Cu:Zn = 2 (SD, diamonds) and (ii) Cu
precursor impregnated on ZnO/Al2O3, Cu:Zn = 0.5 (Cu@Zn, circles).
65 Fig. 9 (a) Surface composition and (b) Zn 2p3/2 spectra of CuZnAl fresh,
after one CO cycle, and after reꢀactivation and a subsequent CO cycle
(vide Fig. 4a).
2). ZnO seems to exhibit a strong affinity to accumulate at the
surface. This surface ZnO is very likely to influence the
stabilisation of copper species in their different oxidation state.
25 We consequently deduce that the dynamics in Cu+ formation is a
measure for the interaction of Cu and ZnO.
Besides, the alteration in Cu:Zn bulk ratio also affects strongly
the CO hydrogenation activity (Fig. 7, inset top). It appears that
the reducibility of the catalyst influences the CO hydrogenation
30 activity comparing the H2ꢀTPR profiles recorded after the CO
cycles (Fig. 7, inset middle): the lower the copper content in the
catalyst the easier the reducibility (shift to lower temperature) and
the higher the MeOH yield. Accordingly, the redox properties of
the catalyst surface and so the interaction of Cu with ZnO seem to
35 be important for CO hydrogenation, too.50,51
The uniqueness of the CuꢀZnO composition remains
controversial. The interaction of ZnO with the adsorbed reactants
of methanol synthesis (CO, CO2, and H2) was formerly studied by
several authors.52ꢀ54 In addition to the impact on the oxidation
40 state of Cu by ZnO,35,55 it is obvious that zinc oxide itself is
involved in the reaction mechanism as an active site.48,56ꢀ59
Otherwise the low activity of CuAl even at R = 0 could be not
explained (Fig. 6). Topsøe et al. proposed a model of CuꢀZn alloy
as active sites which are supposed to be formed in situ under
Consequently, this morphology alters dynamically according to
the syngas composition. This structural feature seems to be
70 unique for ZnO since other supports like SiO2 do not exhibit such
an interplay.
The type of interaction between Cu and ZnO will
unambiguously determine the performance of the catalyst. This is
affirmed by conducting the CO cycle over impregnated Al2O3
75 catalysts (Fig. 8): CO2 promotion is only observed when the
copper precursor is deposited on ZnO/Al2O3 (Cu@Zn). This
catalyst offers easier reducible copper sites which indicates a
better interaction of Cu and ZnO (vide H2ꢀTPR profiles in Fig.
SI3) and is also responsible for the enhanced CO hydrogenation
80 performance; in line with the correlation found for the coꢀ
precipitated CuZn catalysts (Fig. 7, inset top). The much higher
dispersion of the Cu nanoparticles in Cu@Zn explains the higher
CO2 hydrogenation activity (vide R = 100(1) in Fig. 8). The
combination of both structural properties seems to be the preꢀ
85 requisite for CO2 promotion. The flattened shape of the cycle
might be a consequence of the low copper content in the
impregnated catalysts, and therefore they are less affected by the
detrimental hydroxyl coverage at higher R.
Finally, we have found that Al2O3 acts as important structural
90 promoter in methanol synthesis catalysts. It is only a minor
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 7

Catalysis Science & Technology
Page 8 of 10
View Article Online
DOI: 10.1039/C3CY00573A
compound in the bulk composition; however, it is highly
abundant on the catalyst surface (Fig. 9a). Alumina enhances the
dispersion of Cu nanoparticles and stabilises them (Fig. 6, inset).
It has also impact on the CuꢀZnO interactions which explains its
importance in terms of CO2 promotion: the amount of Zn on the
surface is doubled at the expense of Al after the CO cycle over reꢀ
activated CuZnAl (Fig. 9a). This increase of ZnO fraction can be
assigned to isolated islands evidenced by the second peak
appearing at ~1024 eV binding energy in the Zn 2p3/2 spectrum
(5) The observed hysteresis appears to be the consequence of
two effects: (i) the continuous decrease of Cu dispersion and
(ii) the enhanced fixation of copper in its monovalent
oxidation state. Both are disadvantageous for CO2
hydrogenation (the first aspect is also detrimental for WGS).
60
5
Acknowledgements
We thank Total Research & Technology Feluy for permission to
65 publish these results. Thomas Soltermann is acknowledged for
his contribution to the preparation of the catalysts.
10 (Fig. 9b).
The strong influence of Al2O3 on the CuꢀZnO interaction is
further evidenced by the CO cycles over the impregnated
catalysts (Fig. 8): the hysteresis loop is not closing. The excess of
Al2O3 appears to influence the reducibility of the catalysts and
15 thus the CO hydrogenation performance which could be caused
by the fixation of particular structural arrangements of Cu and
ZnO formed along the cycle. These results substantiate former
discussions on the importance of the microstructure of Cuꢀbased
catalysts and the influence by Al2O3 on it.64,65 It is finally the
20 interplay of cascading stabilisation of differently active Cu
species by the metal oxides, ZnO and Al2O3, and the ability of
dynamic alteration of the active sites which determines the
catalyst behaviour in CO2ꢀpromoted methanol synthesis.
References
1
J. B. Hansen and P. E. H. Nielsen, in Handbook of Heterogeneous
Catalysis, eds. G. Ertl, H. Knözinger, F. Schüth and J. Weitkamp,
WileyꢀVCH Verlag GmbH & Co. KGaA, Weinheim, 2nd edn., 2008,
vol. 6, 2920.
70
75
80
85
2
3
4
K. Klier, Adv. Catal., 1982, 31, 243.
G. A. Olah, Angew. Chem., Int. Ed., 2005, 44, 2636.
B. Tidona, C. Koppold, A. Bansode, A. Urakawa and P. R. von Rohr,
J. Supercrit. Fluids, 2013, 78, 70.
5
6
7
8
9
Reaction enthalpies have been calculated using equation of state for
ideal gases by software HSC Chemistry 7.1.
S. Lee, Methanol Synthesis Technology, CRC Press, Boca Raton,
1990.
4. Conclusions
25 The dynamic nature of CuꢀZnOꢀAl2O3 catalysts in CO2ꢀpromoted
methanol synthesis could be clearly unveiled by means of
consecutively altering the CO2 concentration in the syngas feed
over various catalyst compositions and samples prepared via
different routes. Kinetic descriptions therefore need to take into
30 account that the active sites are not necessarily the same for all
syngas compositions and that these sites are changed depending
on the catalyst history. Consequently, turn over frequencies solely
based on the copper surface area do not appear to be sufficient for
investigating this reaction. Novel key findings of this study and
35 revisited controversial results from literature are summarised in
the following:
Y. L. Zhang, Q. Sun, J. F. Deng, D. Wu and S. Y. Chen, Appl. Catal.,
A, 1997, 158, 105.
K. Klier, V. Chatikavanij, R. G. Herman and G. W. Simmons, J.
Catal., 1982, 74, 343.
K. G. Chanchlani, R. R. Hudgins and P. L. Silveston, J. Catal., 1992,
136, 59.
10 J. A. Brown Bourzutschky, N. Homs and A. T. Bell, J. Catal., 1990,
124, 73.
11 G. Liu, D. Willcox, M. Garland and H. H. Kung, J. Catal., 1985, 96,
251.
90 12 C. J. Schack, M. A. McNeil and R. G. Rinker, Appl. Catal., 1989, 50,
247.
(1) The performance of Cuꢀbased catalysts in CO2ꢀpromoted
methanol synthesis depends on the syngas composition and
13 J. S. Lee, K. H. Lee, S. Y. Lee and Y. G. Kim, J. Catal., 1993, 144,
414.
the history of the catalyst resulting in
a hysteresis
14 G. Prieto, J. Zečević, H. Friedrich, K. P. de Jong and P. E. de Jongh,
40
phenomenon. This behaviour is less affected by transient
changes since a quasiꢀequilibrated catalyst exhibited a
similar performance.
95
Nature Mater., 2013, 12, 34.
15 J. C. J. Bart and R. P. A. Sneeden, Catal. Today, 1987, 2, 1.
16 G. C. Chinchen, P. J. Denny, J. R. Jennings, M. S. Spencer and K. C.
Waugh, Appl. Catal., 1988, 36, 1.
(2) It turned out unambiguously that the surface area of copper
is not the key parameter for understanding the promotional
effect since CO hydrogenation does not show a dependence
on the copper surface area whereas CO2 hydrogenation does.
In line with reported results by several authors, the
microstructure of the catalyst needs to be considered
thoroughly.
45
17 K. C. Waugh, Catal. Lett., 2012, 142, 1153.
100 18 J. D. Grunwaldt, A. M. Molenbroek, N.ꢀY. Topsøe, H. Topsøe and B.
S. Clausen, J. Catal., 2000, 194, 452.
19 P. L. Hansen, J. B. Wagner, S. Helveg, J. R. RostrupꢀNielsen, B. S.
Clausen and H. Topsøe, Science, 2002, 295, 2053.
50 (3) The promotional effect by CO2 is coupled to the WGS
activity. The deactivation at higher concentrations of CO2 is
thus a result of the excessive presence of surface hydroxyl
species accompanied by the decrease in Cu dispersion which
is detrimental for CO2 hydrogenation.
55 (4) A strong evidence was found that the interaction of copper
and zinc oxide dictates essentially the CO2 promotion
substantiating results formerly reported.
20 H. Wilmer and O. Hinrichsen, Catal. Lett., 2002, 82, 117.
105 21 H. Topsøe, C. V. Ovesen, B. S. Clausen, N.ꢀY. Topsøe, P. E. H.
Nielsen, E. Törnqvist and J. K. Nørskov, in Dynamics of Surfaces
and Reaction Kinetics in Heterogeneous Catalysis, eds. G. F.
Froment and K. C. Waugh, Elsevier Science Bv, Amsterdam, 1997,
vol. 109, 121.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 8

Page 9 of 10
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C3CY00573A
22 M. Behrens, F. Studt, I. Kasatkin, S. Kühl, M. Hävecker, F. Abildꢀ
52 R. Naumann d'Alnoncourt, X. Xia, J. Strunk, E. Löffler, O.
Hinrichsen and M. Muhler, Phys. Chem. Chem. Phys., 2006, 8, 1525.
53 D. L. Roberts and G. L. Griffin, Appl. Surf. Sci., 1984, 19, 298.
54 C. Wöll, Prog. Surf. Sci., 2007, 82, 55.
Pedersen, S. Zander, F. Girgsdies, P. Kurr, B.ꢀL. Kniep, M. Tovar, R.
Fischer, J. K. Nørskov and R. Schlögl, Science, 2012, 336, 893.
23 Y. Yang, J. Evans, J. A. Rodriguez, M. G. White and P. Liu, Phys.
Chem. Chem. Phys., 2010, 12, 9909.
5
10
15
60 55 J. Skrzypek, M. Lachowska, M. Grzesik, J. Sloczynski and P.
Nowak, Chem. Eng. J. Biochem. Eng. J., 1995, 58, 101.
24 C. Kuechen and U. Hoffmann, Chem. Eng. Sci., 1993, 48, 3767.
25 H.ꢀW. Lim, M.ꢀJ. Park, S.ꢀH. Kang, H.ꢀJ. Chae, J. W. Bae and K.ꢀW.
Jun, Ind. Eng. Chem. Res., 2009, 48, 10448.
56 V. E. Ostrovskii, Catal. Today, 2002, 77, 141.
57 M. S. Spencer, Top. Catal., 1999, 8, 259.
26 K. L. Ng, D. Chadwick and B. A. Toseland, Chem. Eng. Sci., 1999,
54, 3587.
58 X. M. Liu, G. Q. Lu, Z. F. Yan and J. Beltramini, Ind. Eng. Chem.
65
Res., 2003, 42, 6518.
27 K. M. Vanden Bussche and G. F. Froment, J. Catal., 1996, 161, 1.
28 G. C. Chinchen, C. M. Hay, H. D. Vandervell and K. C. Waugh, J.
Catal., 1987, 103, 79.
59 S. Polarz, J. Strunk, V. Ischenko, M. W. E. van den Berg, O.
Hinrichsen, M. Muhler and M. Driess, Angew. Chem., Int. Ed., 2006,
45, 2965.
29 W. L. Dai, Q. Sun, J. F. Deng, D. Wu and Y. H. Sun, Appl. Surf. Sci.,
2001, 177, 172.
60 N.ꢀY. Topsøe and H. Topsøe, J. Mol. Catal. A: Chem., 1999, 141, 95.
70 61 N.ꢀY. Topsøe and H. Topsøe, Top. Catal., 1999, 8, 267.
62 M. Kurtz, J. Strunk, O. Hinrichsen, M. Muhler, K. Fink, B. Meyer
and C. Wöll, Angew. Chem., Int. Ed., 2005, 44, 2790.
30 P. Villa, P. Forzatti, G. BuzziꢀFerraris, G. Garone and I. Pasquon,
Ind. Eng. Chem. Process Des. Dev., 1985, 24, 12.
31 G. Liu, D. Willcox, M. Garland and H. H. Kung, J. Catal., 1984, 90,
139.
63 G. Rossmüller, V. Kleinschmidt, J. Kossmann and C. Hättig, J. Phys.
Chem. C, 2009, 113, 1418.
20 32 F. Arena, K. Barbera, G. Italiano, G. Bonura, L. Spadaro and F.
Frusteri, J. Catal., 2007, 249, 185.
75 64 I. Kasatkin, P. Kurr, B. Kniep, A. Trunschke and R. Schlögl, Angew.
Chem., Int. Ed., 2007, 46, 7324.
65 M. Kurtz, H. Wilmer, T. Genger, O. Hinrichsen and M. Muhler,
Catal. Lett., 2003, 86, 77.
33 S. Kaluza, M. Behrens, N. Schiefenhövel, B. Kniep, R. Fischer, R.
Schlögl and M. Muhler, ChemCatChem, 2011, 3, 189.
34 A. Karelovic, A. Bargibant, C. Fernández and P. Ruiz, Catal. Today,
25
2012, 197, 109.
35 J. Nakamura, T. Uchijima, Y. Kanai and T. Fujitani, Catal. Today,
1996, 28, 223.
36 Y. Yang, C. A. Mims, D. H. Mei, C. H. F. Peden and C. T. Campbell,
J. Catal., 2013, 298, 10.
30 37 M. Sahibzada, I. S. Metcalfe and D. Chadwick, J. Catal., 1998, 174,
111.
38 M. Saito, T. Fujitani, M. Takeuchi and T. Watanabe, Appl. Catal., A,
1996, 138, 311.
39 M. Peter, M. B. Fichtl, H. Ruland, S. Kaluza, M. Muhler and O.
35
Hinrichsen, Chem. Eng. J., 2012, 203, 480.
40 M. J. L. Ginés, N. Amadeo, M. Laborde and C. R. Apesteguía, Appl.
Catal., A, 1995, 131, 283.
41 C. V. Ovesen, B. S. Clausen, J. Schiotz, P. Stoltze, H. Topsøe and J.
K. Nørskov, J. Catal., 1997, 168, 133.
40 42 M. Peter, J. Fendt, S. Pleintinger and O. Hinrichsen, Catal. Sci.
Technol., 2012.
43 P. C. K. Vesborg, I. Chorkendorff, I. Knudsen, O. Balmes, J. Nerlov,
A. M. Molenbroek, B. S. Clausen and S. Helveg, J. Catal., 2009, 262,
65.
45 44 G. U. Kulkarni and C. N. R. Rao, Top. Catal., 2003, 22, 183.
45 G. R. Apai, J. R. Monnier and M. J. Hanrahan, J. Chem. Soc., Chem.
Commun., 1984, 212.
46 J. R. Monnier, G. Apai and M. J. Hanrahan, J. Catal., 1984, 88, 523.
47 T. Fujitani and J. Nakamura, Catal. Lett., 1998, 56, 119.
50 48 R. G. Herman, K. Klier, G. W. Simmons, B. P. Finn, J. B. Bulko and
T. P. Kobylinski, J. Catal., 1979, 56, 407.
49 G. C. Chinchen, K. Mansfield and M. S. Spencer, Chemtech, 1990,
20, 692.
50 R. Burch and R. J. Chappell, Appl. Catal., 1988, 45, 131.
55 51 M. Amara, M. Bettahar and D. Olivier, Appl. Catal., 1989, 51, 141.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 9

Catalysis Science & Technology
Page 10 of 10
View Article Online
DOI: 10.1039/C3CY00573A
Table of content
The CO2ꢀpromoted methanol synthesis over binary and ternary
Cuꢀbased catalysts has been assessed by using cyclic tests in
CO:H2:CO2:H2O mixtures.
5
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 10