High-pressure advantages in stoichiometric hydrogenation
of carbon dioxide to methanol

Rohit Gaikwad, Atul Bansode and Atsushi Urakawa*

Institute of Chemical Research of Catalonia (ICIQ), The Barcelona Institute of Science and
Technology, Av. Països Catalans 16, 43007 Tarragona, Spain

*Corresponding author. Email: aurakawa@iciq.es

1

Abstract

Interplay between three important reaction parameters (pressure, temperature, and
space velocity) in stoichiometric hydrogenation of carbon dioxide (CO2:H2=1:3) was
systematically investigated using a commercial Cu/ZnO/Al2O3 catalyst. Their impacts on
reaction performance and important ranges of process conditions towards full one-pass
conversion of CO2 to methanol at high yield were rationalized based on the kinetics and
thermodynamics of the reaction. Under high-pressure condition above a threshold
temperature, the reaction overcomes kinetic control, entering thermodynamically controlled
regime. Ca. 90% CO2 conversion and >95% methanol selectivity was achieved with a very
good yield (0.9-2.4 gMeOH gcat-1 h-1) at 442 bar. Such high-pressure condition induces the
formation of highly dense phase and consequent mass transfer limitation. When this
limitation is overcome, the advantage of high-pressure conditions can be fully exploited and
weight time yield as high as 15.3 gMeOH gcat-1 h-1 could be achieved at 442 bar. Remarkable
advantages of high-pressure conditions in the terms reaction kinetics, thermodynamics, and
phase behavior in the aim to achieve better methanol yield are discussed.

Keywords:

CO2

hydrogenation;

methanol

thermodynamics; Cu/ZnO/Al2O3

2

synthesis;

high-pressure;

kinetics;

1. Introduction
The ever-increasing energy demand to sustain the industrialization and modern
lifestyle has led to depletion consumption of world’s current primary energy supply, finite
and non-renewable fossil fuels. In parallel, their irreversible consumption resulted in
accumulation of carbon dioxide (CO2) in the atmosphere, causing the climate to change. For
sustainable development of mankind, the carbon cycle has to be closed, and conversion of
CO2 into chemical fuels and feedstocks serves as an effective strategy to cope with the
interrelated energetic and environmental problems [1]. Heterogeneous catalytic conversion of
CO2 to fuels and industrially important chemicals, such as methanol by hydrogenation
reaction, offers a unique path to transform a large amount of CO2 in a short span of time by
the high reaction rates. The vital roles of methanol as chemical energy carrier and starting
material or chemical intermediates are well recognized [2]. However, CO2 is a
thermodynamically stable and relatively inert molecule. Its activation typically requires
energy inputs, e.g. by the use of elevated pressure and temperature as well as effectual
strategies such as innovative catalytic processes [3-6].
CO2 hydrogenation to methanol is exothermic (reaction 1), while the competing
reaction, reverse water-gas shift (RWGS, reaction 2), is endothermic [7]. Moreover, CO
produced by RWGS may undergo exothermic hydrogenation to form methanol (reaction 3).
CO2 + 3H2 ⇄ CH3OH + H2O

ΔH298K,5MPa = −40.9 kJ / mol …………………(1)

CO2 + H2 ⇄ CO + H2O

ΔH298K,5MPa = +49.8 kJ / mol ..………………(2)

CO + 2H2 ⇄ CH3OH

ΔH298K,5Mpa = −90.7 kJ / mol …………………(3)

In accordance with Le Chàtelier’s principle, high-pressure and low-temperature reaction
conditions are favorable to achieve high CO2 conversion and methanol selectivity [8]. In fact,

3

the advantage and necessity of high-pressure conditions in the synthesis of methanol from
syngas (CO and H2 mixture typically containing some fraction of CO2) have been known
over the last 90 years [9]. Since 1966, the trend has shifted to lower pressure methanol
synthesis (<100 bar) using highly active and economic Cu-ZnO based catalysts [10]. For this
family of catalysts which are most common for methanol synthesis nowadays, high-pressure
advantages in methanol synthesis by the hydrogenation of CO and particularly CO2 had not
been explored and documented for a long time, except the excellent work reported by Ipatieff
and Monroe in 1945 for Cu-based catalysts [5]. Recently, we reported a range of highpressure reaction conditions, yielding remarkable almost-full one-pass conversion of CO2 to
methanol with high selectivity using Cu/ZnO/Al2O3 catalysts and also to methanol-derived
products such as dimethyl ether (DME) by co-presence of an acidic zeolite [11]. The elevated
H2 partial pressure (molar ratio, CO2:H2=1:>10), higher than the stoichiometric one
(CO2:H2=1:3, reaction 1), was found kinetically as well as thermodynamically beneficial for
methanol synthesis. Employing the reaction pressure of 360 bar (reactants pressure of 331 bar
due to the presence of Ar for GC analysis), outstanding CO2 conversion (>95%) and
methanol selectivity (>98%) were achieved at 260 °C at relatively high gas hourly space
velocity (GHSV) of ca. 10000 h-1. With a commercial Cu-ZnO based methanol synthesis
catalyst, exceptionally high methanol yield of 7.7 gMeOH gcat-1 h-1 was attained at the expense
of lower CO2 conversion (65.8%) and methanol selectivity (77.3%). Energy-demanding highpressure conditions are not necessarily disadvantageous in this reaction because of smaller
geometrical requirements for the reactor and plant area, which lowers the capital cost and
possibly improve safety aspects [4, 11]. Moreover, the energetic cost associated with
compression of the reactants is less significant than that required for hydrogen production in
the overall process of CO2 hydrogenation to methanol when water electrolysis is assumed as
the mean to produce H2 [12].

4

Despite the exceptionally high CO2 conversion and methanol selectivity under highpressure conditions and also high process viability in terms of costs and methanol
productivity, the reported reaction condition requires recycling or further conversion of
unreacted H2 fed in excess. In addition, CO produced by RWGS should be recycled if
methanol selectivity is not sufficiently high. Recycling of precious H2 can only be avoided by
achieving its full conversion. In other words, the challenge in this respect is to achieve
complete conversion of both CO2 and H2 with high methanol productivity. This goal naturally
requires the operation of the reaction at the stoichiometric CO2 to H2 ratio.
Herein, we have thoroughly examined high-pressure reaction conditions (50-480 bar;
in reactants pressure of 46-442 bar) at 220-300 °C for stoichiometric CO2 hydrogenation, in
the aim to identify reaction conditions maximizing CO2 and H2 conversions with high
methanol selectivity and/or productivity. A commercial Cu/ZnO/Al2O3 catalyst, optimized for
the conversion of syngas to methanol, was employed as catalyst due to its high activity in
CO2 hydrogenation to methanol [11]. Tendency and effects of kinetic and thermodynamic
controls over the reaction performance are discussed along with the trends in theoretical
thermodynamic equilibria to critically evaluate what is achievable with the optimized
Cu/ZnO/Al2O3 catalyst and to highlight possible directions of future research in CO2
hydrogenation to methanol.

2. Experimental
The details of high-pressure fixed-bed reactor and analytical systems are described
elsewhere [4]. Specificity of this work about reactor system, catalyst material, activation
procedure and product analysis by gas chromatography is described in Supplementary
Material. Continuous CO2 hydrogenation to methanol was tested at four different pressure

5

conditions of 50, 100, 200, 360, 480 bar (considering 8% Ar in feed composition, actual total
pressure of CO2 and H2 was 46, 92, 184, 331, and 442 bar, respectively). In this work, GHSV
is defined by the volumetric flow rate of inlet stream at normal pressure divided by the
reactor volume where the catalyst is packed (including the catalyst volume). A wide range of
GHSV conditions (650-100000 h-1) were examined. GHSV is also shown in catalyst-massnormalized unit, in which the value ranges 0.37-49.85 NL gcat-1 h-1. For the GHSV calculation
in both units, the total flow rate at normal pressure including Ar was used. The vaporized
outlet stream were injected to GC every ca. 12 min for 3 h at each condition of temperature,
pressure and GHSV and an averaged values was taken. The standard deviation of the
quantification is presented in Figures S3-S5 in Supplementary Materials. No catalyst
deactivation was detected for the duration of catalytic tests performed.
Equilibrium conversion and product selectivity were calculated by the Soave-RedlichKwong (SRK) equation of state (EOS) using Aspen HYSYS V8.6. Modified SRK-EOS
binary interaction parameters for CO, CO2, H2, methanol and water were taken from the
optimized values reported by Heeres and coworkers for methanol synthesis [13]. The
calculations were performed by minimization of Gibbs free energy. Methane was not
considered in all calculations.

3. Results and discussion
3.1 Effects of temperature under high-pressure conditions
First, the effects of temperature on CO2 conversion and methanol selectivity were
examined at the reactants pressure of 46, 92, 184, 331, and 442 bar (Figure 1). The catalytic
tests were performed at a constant GHSV of 10000 h-1, although, as discussed in section 3.2,
this reaction parameter can directly influence the residence time of the reactants in the reactor

6

and thus catalytic performance. CO2 conversion and methanol selectivity are presented in
comparison to the thermodynamic equilibrium values.
Advantages of high-pressure conditions are obvious according to the thermodynamic
calculations (Figure 1, dotted lines). At 46 bar, CO2 version varies between 25-30% with
rapidly decreasing methanol selectivity from ca. 90 to 20% in the temperature window of
220-300 °C. At 92 bar, CO2 conversion varies from roughly 50% (220 °C) to 30% (300 °C)
with very good to moderate methanol selectivity (96.5% at 220 °C and 53.4% at 300 °C),
whereas at the highest examined pressure of 442 bar, theoretically CO2 can be effectively
converted to methanol (98.7% at 220 °C and 86.1% at 300 °C) with very high selectivity for
the entire temperature range (>99.9% at 220 °C and 99.0% at 300 °C). At 184 and 331 bar,
there was a sudden change in CO2 equilibrium conversion at ca. 230 and 280 °C, respectively
(Figure S1 in Supplementary Materials; this change also takes place at 46 and 92 bar but at
much lower temperatures at ca. 110 (not shown in the figure) and 160 °C, respectively.). This
is due to enhanced CO2 conversion induced by the phase transition and separation (formation
of liquid phase) associated with the condensation of the products when the reaction
temperature is lower than the transition point. Such phase separation allows CO2 conversion
to methanol beyond one-phase equilibrium, as precisely described and demonstrated by
Heeres and coworkers [6]. The positive impact of such phase separation on CO2 conversion
should become less prominent at higher pressures as noticeable from the equilibrium CO2
conversion curves of 184 and 331 bar. At 442 bar the impact becomes even unnoticeable.
This tendency is attributed to the highly dense reactant/product mixture whose density only
slightly differs from that of the liquid products and/or it indicates that they are simply
miscible at the high-pressure condition.
Experimentally, the general advantages of high-pressure conditions in CO2 conversion,
methanol selectivity, and thus methanol yield were confirmed with better catalytic

7

performance at higher pressures (Figure 1). Besides methanol, CO was found as the only
major product arising from RWGS reaction. Another product observed was methane with a
minor quantity (<0.8%). In comparison to the theoretical equilibrium values, larger
deviations were observed at lower temperatures for both CO2 conversion and methanol
selectivity. These two key indicators of reaction performance showed the maxima at 260-280
°C, except methanol selectivity at 46 and 331 bar, and then decreased at higher temperatures.
The performance deterioration above the optimum temperature of 260-280 °C is in
accordance with the trend expected by the thermodynamic equilibrium. In the range of 220300 °C there were smaller deviations between experimental and theoretical CO2 conversion
and methanol selectivity above the optimum temperature, whereas larger deviations were
found below the optimum temperature. This implies that thermodynamic equilibrium has
been reached or, at least, has significant influence on the reaction performance at the
temperatures higher than the optimum temperature at every pressure condition. In other
words, at the temperatures below the maxima in catalytic performance, the reaction is
kinetically controlled due to poor reaction rates determined by the catalyst at the low
temperatures. Theoretically, CO2 conversion can be drastically boosted below 230 °C at 184
bar. However, such performance enhancement was not observed and a very poor value was
obtained at 220 °C. This is a clear indication that the reaction is kinetically controlled at the
temperature. To fully benefit from the phase separation, the reaction has to be performed at
lower GHSVs to achieve high conversion at low temperatures. Also, it is important to remark
that the advantageous phase separation is expected to occur theoretically at higher
temperatures under higher pressure conditions. Therefore, high-pressure conditions can be
greatly beneficial in this respect to achieve phase separation under kinetically favorable hightemperature conditions.

8

The best catalytic performance in terms of CO2 conversion and methanol selectivity was
obtained at 260 °C at 331 bar and at 280 °C at 46, 92, 184, and 442 bar. Maximally
performing reaction temperatures were examined at higher and lower GHSV conditions at
331 bar. Interestingly, it was found that the optimum temperature remained the same
irrespective of different GHSV conditions (data are not shown). Therefore, we have taken the
optimum temperatures at the respective pressures for the following study where the influence
of GHSV on the catalytic performance is investigated.

3.2 Effects of GHSV under high-pressure conditions
The reaction performance under the high-pressure conditions at the optimum temperature
was further evaluated in a wide range of GHSV (650-100000 h-1, equivalent to 0.37-49.85
NL gcat-1 h-1). Figure 2 presents CO2 conversion and methanol selectivity as a function of
GHSV at 46, 92, 184, 331, and 442 bar, and Figure 3 shows corresponding methanol
productivity in terms of weight time yield (WTY) expressed in the unit of gMeOH gcat-1 h-1. In
Figure 2, equilibrium CO2 conversion and methanol selectivity values are indicated by dotted
lines.
Clearly, the catalytic performance approaches the thermodynamic limit at the low range
of GHSV (i.e. longer residence time). It is, however, not beneficial to over-reducing GHSV
as the catalytic performance, especially methanol selectivity, becomes worse. This is mainly
due to the formation of side products like methane and ethanol (Tables S2-S6, Supplementary
Material). Also, under the very low GHSV conditions, methanol yield is consequently very
low. Thus, such reaction conditions are not practically relevant for large-scale industrial
operation. The decreased CO2 conversion towards the lowest examined GHSV at 184 bar
may be due to additional chemical equilibria involving methane and ethanol, but further
investigation is required to identify the exact cause.

9

What is striking from the dependence of methanol WTY on GHSV (Figure 3, filled
symbols for 100-300 µm catalyst particle size) is that there are reaction conditions giving
high CO2 conversion and methanol selectivity with methanol WTY close to 1.0 gMeOH gcat-1 h1

, which is generally considered as an excellent one. At 442 bar, the WTY reached the value

of 0.92 gMeOH gcat-1 h-1 at 4000 h-1 with 88.5% CO2 conversion and 97.2% methanol
selectivity (Table S2). 0.89 gMeOH gcat-1 h-1 was obtained at 331 bar also at 4000 h-1 with
83.3% CO2 conversion and 96.8% methanol selectivity (Table S3). Similar methanol WTY
can be attained at lower pressure, but this requires increasing GHSV due to lower CO2
conversion and methanol selectivity. For example, at 184 bar 0.88 gMeOH gcat-1 h-1 was
obtained at 8000 h-1 with 47.0% CO2 conversion and 84.8% methanol selectivity (Table S4).
At 92 and 46 bar (Tables S5 and S6), high GHSVs (30000 or 100000 h-1) was necessary to
achieve >1.0 gMeOH gcat-1 h-1 with poor CO2 conversion (28.6 or 20.2%, respectively) and
moderate-poor methanol selectivity (53.6 or 19.7 %, respectively).
In practice, high CO2 conversion and high methanol selectivity may not be the most
critical performance indicator when unreacted CO2, CO, and H2 can be efficiently recycled.
Although larger molar and thus volumetric flow (i.e. high GHSV) demands for higher
energetic requirement for the recycling process due to low CO2 conversion, such conditions
can greatly improve methanol WTY as discussed above. This was clearly demonstrated under
the high GHSV conditions of this work (Figure 3, filled symbols). At 100000 h-1 even at the
moderate pressure of 92 bar, a very high WTY of ca. 3 gMeOH gcat-1 h-1 was achieved and
overall excellent WTYs above 4.5 gMeOH gcat-1 h-1 could be attained above 184 bar.
Interestingly, the high-pressure benefit in CO2 conversion was less pronounced at high
GHSV, and the conversion values converged to roughly 20-30% at 100000 h-1 for all
examined pressure conditions. In contrast, high-pressure advantage in methanol selectivity

10

remained (70.0% at 331 bar, 47.7% at 92 bar, 19.7% at 46 bar) although methanol selectivity
decreased remarkably at higher GHSV at 442 bar).
Furthermore, there were clear differences of the GHSV dependency of CO2 conversion at
the different pressures. The drop in CO2 conversion was more prominent at higher pressure
conditions (331 and 442 bar) upon increasing GHSV, whereas methanol selectivity was not
affected by the GHSV variation as much except methanol selectivity at 442 bar. According to
the thermodynamic calculation (Figure 1, dotted lines and Figure S1, Supplementary
Materials), only under the two high-pressure conditions (331 and 442 bar) product
condensation and phase separation (or formation of highly dense phase of the reactants and
products at 442 bar due to the rather smooth and continuously changing CO2 conversion
profile with increasing temperature) are expected to occur at the reaction temperatures
examined. The more significant drop in CO2 conversion at higher GHSVs may be related to
the phase behavior. For example, a more severe mass transfer limitation may be induced at
higher GHSV conditions, resulting in hindered diffusion of the reactants and products
through the catalyst body by the dense/liquid phase formation. In order to verify if there is
internal mass transfer limitation or not, we have performed the reaction using the catalyst
with the particle size one order of magnitude smaller (10-20 µm) than those screened and
reported above (100-300 µm) at representative pressure (92, 331, and 442 bar) and GHSV
(10000-100000 h-1) conditions. External mass transfer limitation was neglected because the
drop in catalytic performance occurs at high GHSV conditions which are favorable for
external mass transfer.
Figures 2 and 3 present the effects of catalyst particle size on the catalytic performance
and WTY obtained at different GHSVs with smaller catalyst particles (empty symbols and
dotted lines) and by larger catalyst particles (filled symbols and solid lines). The reaction
performance using the smaller catalyst particles was almost identical to that of the larger ones
11

at 92 and 331 bar, but there was a remarkable enhancement of CO2 conversion and methanol
selectivity observed at 442 bar. Even at the highest examined GHSV (100000 h-1) high CO2
conversion (65.3%) and methanol selectivity (91.9%) were achieved, giving outstanding
WTY of 15.3 gMeOH gcat-1 h-1. At lower GHSV of 30000 h-1, CO2 conversion was 80.0% with
96.7% methanol selectivity, giving WTY of 6.7 gMeOH gcat-1 h-1. Compared to the theoretical
WTY limit (7.7 gMeOH gcat-1 h-1) defined by the equilibrium conversion and selectivity, the
value is very high. At 10000 h-1, WTY of 2.4 gMeOH gcat-1 h-1 was almost the same as the
theoretical value of 2.6 gMeOH gcat-1 h-1.
These large effects of particle size on the catalytic performance clearly prove that there
was a severe internal mass transfer limitation especially at 442 bar as hinted by the great
decrease in CO2 conversion at higher GHSVs using the larger catalyst particles. The degree
of internal mass transfer was quantitatively evaluated by means of Thiele modulus,
effectiveness factor, and Weisz-Prater criterion (Tables S8 and S9, Figure S6, Supplementary
Materials). The Weisz-Prater criterion clearly shows the values much larger than 1 at higher
GHSVs at 442 bar with the larger catalyst particles. This indicates severe internal mass
transfer limitation at 442 bar. The effectiveness factor was 0.3 at 100000 h-1 at 442 bar,
showing poor utilization of catalyst surfaces within the particle. The effectiveness of catalyst
utilization improves at the lowest space velocity (10000 h-1) of the study and the value of
0.92 was obtained with the catalyst of the larger particle size. When the smaller size catalyst
was used, much smaller values of the Weisz-Prater criterion and high value (above 0.95) of
the effectiveness factor were obtained, evidencing the effective use of the whole catalyst
body for the reaction when the size is reduced by one order of magnitude. At 331 bar, the
availability of the catalytic sites are improved as shown by lower values of the Weisz-Prater
criterion and by the high values of effectiveness factor. These results strongly suggest that
condensation of reactants/products takes place within the catalyst body at the very high12

pressure conditions examined, inducing the mass transfer limitation. As demonstrated by the
extraordinary WTY above 15 gMeOH gcat-1 h-1, this product condensation can be extremely
beneficial if mass transfer limitation can be overcome and/or absent. A higher pressure drops
observed at 331 and 442 bar (ΔP = ca. 2-5 bar) than at 92 bar (ΔP = ca. 1 bar) also indicate
the formation of dense phase under higher pressure conditions.
The reaction mechanisms of methanol synthesis via CO2 hydrogenation, namely via CO2
or CO, are widely debated [14], although a recent study as represented by Studt and
coworkers concluded that the major carbon source of methanol is CO2, promoted by the
synergetic functions of Cu and ZnO [15]. In this study, CO selectivity increased consistently
at higher GHSV (Figure 2 and Tables S2-S5). Detailed mechanistic discussion is out of the
scope of this work, but the results indicate that longer residence time enhances methanol
selectivity, and that methanol synthesis proceeds via CO produced by RWGS. The same
conclusion had been drawn in over-stoichiometric CO2 hydrogenation where excess hydrogen
was used (CO2:H2=1:10) [11]. Still, there is one point which has not been discussed widely,
which is the exothermicity of methanol synthesis which can create local hot spots and
temperature gradients along the axial and radial directions of the catalyst bed. The reactor we
have used has high surface to volume ratio as a kind of microreactor and in principle the
geometry is well suited for heat management. However, the generated heat may not be
sufficiently removed when WTY of methanol is very high and thus a large heat is generated
within the reactor and large temperature increase may occur especially at the catalyst bed
where reactants concentration is higher, i.e. close to the inlet of the reactor. The reaction
enhanced under such conditions would be endothermic one, i.e. RWGS, thus CO formation
could be pronounced in such cases. The trend is indeed what we observed; the higher the
GHSV, i.e. higher the WTY consequently in most cases, the higher the CO selectivity. These
aspects will be investigated further, but this may be a possible explanation of apparent
13

reaction path of methanol synthesis via CO produced by RWGS under high-pressure
conditions because of existence of local hot spots, besides the scenario that CO2
hydrogenation indeed proceeds via CO at high-pressures.

4. Conclusion
Relationships among reaction temperature, pressure, and GHSV in stoichiometric CO2
hydrogenation to methanol over a well-established commercial Cu/ZnO/Al2O3 catalyst were
systematically investigated in the aim to understand the advantages given by high-pressure
reaction conditions (46-442 bar) and to achieve as high as possible CO2 conversion and
methanol selectivity with high methanol productivity towards full conversion to methanol. A
strong interplay among kinetics, thermodynamics and phase behavior on the reaction
performance was evidenced. At kinetically favorable high temperatures (>260 °C) especially
at lower GHSV, it was possible to enter the regime where thermodynamic equilibrium plays
dominant roles in determining the catalytic activity. In this regime, high-pressure advantages
can be conveniently predicted based on the equilibrium conversion and selectivity. A good
WTY of 0.9 gMeOH gcat-1 h-1 could be achieved at 442 bar with 88.5% CO2 conversion and
97.2% methanol selectivity using our standard, larger size of catalyst particles (100-300 µm).
At high-pressure conditions above 331 bar, the dense phase formation by product
condensation limits the overall reaction rate by internal mass transfer. When smaller catalyst
particles (10-20 µm) are used instead, the limitation can be effectively removed. Thusobtained catalytic performance fully benefits from the high-pressure advantage of high
reaction rate (kinetics), high equilibrium conversion (thermodynamics) and enhanced
conversion (phase separation). Under these conditions of negligible mass transfer limitations,
at 442 bar a very good WTY of 2.4 gMeOH gcat-1 h-1 could be observed with 87.7% CO2

14

conversion and 97.6% methanol selectivity. At a very high GHSV (100000 h-1), an
extraordinary WTY of 15.3 gMeOH gcat-1 h-1 could be achieved.
This work clearly shows favorable reaction conditions towards full one-pass conversion
in stoichiometric CO2 hydrogenation to methanol. Development of highly active, new
generation catalysts is mandatory to reach this goal by entering to the thermodynamically
controlled regime at lower temperature. Another practically important operation condition
identified was high GHSV. Even at lowered pressure of 184 bar, a remarkable WTY of 4.5
gMeOH gcat-1 h-1 could be obtained. Hence, high GHSV conditions at relatively high-pressure
were found also beneficial in practice for high-yield methanol synthesis when unreacted CO2,
H2, and formed CO are recycled. In summary, this work has demonstrated that both kinetic
and thermodynamic factors play decisive roles in methanol synthesis and also that
thermodynamically favorable high-pressure conditions allow reaching the reactivity in the
thermodynamically controlled regime and/or with outstanding methanol productivity.

Acknowledgements
The authors thank the financial support from the ICIQ Foundation and MINECO (CTQ201234153) and also thank MINECO for support through Severo Ochoa Excellence Accreditation
2014−2018 (SEV-2013-0319). Recognition and support by the Catalan government (2014
SGR 893) are greatly acknowledged. RG acknowledges MINECO for FPI predoctoral
fellowship (EEBB-I-15-10528).

Appendix A. Supplementary Material
Supplementary data associated with this article can be found, in the online version, at
http://dx.doi.org/XXXX.
15

References
[1] M. Mikkelsen, M. Jorgensen, F.C. Krebs, Energ. Environ. Sci., 3 (2010) 43-81.
[2] G.A. Olah, Angew. Chem. Int. Ed., 44 (2005) 2636-2639.
[3] O.d.Q.F. Araújo, J.L.d. Medeiros, R.M.B. Alves, CO2 Utilization: A Process Systems
Engineering Vision, 2014.
[4] A. Bansode, B. Tidona, P.R. von Rohr, A. Urakawa, Catal. Sci. Technol., 3 (2013) 767778.
[5] V.N. Ipatieff, G.S. Monroe, J. Am. Chem. Soc., 67 (1945) 2168-2171.
[6] J.G. van Bennekom, R.H. Venderbosch, J.G.M. Winkelman, E. Wilbers, D. Assink, K.P.J.
Lemmens, H.J. Heeres, Chem. Eng. Sci., 87 (2013) 204-208.
[7] J.A. Dumesic, G.W. Huber, M. Boudart, Principles of Heterogeneous Catalysis, in:
Handbook of Heterogeneous Catalysis, Wiley-VCH Verlag GmbH & Co. KGaA, 2008.
[8] J. Skrzypek, M. Lachowska, M. Grzesik, J. Słoczyński, P. Nowak, Chem. Eng. J. and
Biochem. Eng. J., 58 (1995) 101-108.
[9] BASF, German Patents 415 686, 441 433, 462 837, 1923.
[10] J.B. Hansen, P.E. Højlund Nielsen, Methanol Synthesis, in: Handbook of
Heterogeneous Catalysis, Wiley-VCH Verlag GmbH & Co. KGaA, 2008.
[11] A. Bansode, A. Urakawa, J. Catal., 309 (2014) 66-70.
[12] B. Tidona, C. Koppold, A. Bansode, A. Urakawa, P. Rudolf von Rohr, J. Supercrit.
Fluids, 78 (2013) 70-77.
[13] J.G. van Bennekom, J.G.M. Winkelman, R.H. Venderbosch, S.D.G.B. Nieland, H.J.
Heeres, Ind. Eng. Chem. Res., 51 (2012) 12233-12243.
[14] E.L. Kunkes, F. Studt, F. Abild-Pedersen, R. Schlögl, M. Behrens, J. Catal., 328 (2015)
43-48.

16

[15] F. Studt, M. Behrens, E.L. Kunkes, N. Thomas, S. Zander, A. Tarasov, J. Schumann, E.
Frei, J.B. Varley, F. Abild-Pedersen, J.K. Nørskov, R. Schlögl, ChemCatChem, 7 (2015)
1105-1111.

17

Figure 1. Effects of reaction temperature and pressure on CO2 conversion (XCO2) and
methanol selectivity (SMeOH) in high-pressure stoichiometric CO2 hydrogenation using
commercial Cu/ZnO/Al2O3 catalyst at constant GHSV of 10000 h-1 (5.87 NL gcat-1 h-1).
Dotted lines show the theoretical equilibrium CO2 conversion and methanol selectivity.

18

Figure 2. CO2 conversion (XCO2), methanol selectivity (SMeOH) and methanol weight time
yield (WTYMeOH) in high-pressure stoichiometric CO2 hydrogenation at different GHSV
conditions (650-100000 h-1, equivalent to 0.37-49.85 NL gcat-1 h-1) at 280 °C (46 bar (black),
92 bar (red), 184 bar (blue), and 442 bar (magenta)) and at 260 °C (331 bar, green) using
commercial Cu/ZnO/Al2O3 catalyst. The filled symbols correspond to the catalytic results
obtained with the catalyst of 100-300 µm size fraction, while the empty symbols correspond
to those obtained with the catalyst of 10-20 µm size fraction. The arrows on the right indicate
the thermodynamic equilibrium values at the respective temperature and pressure.

19

Figure 3. (top) Methanol weight time yield (WTYMeOH) in high-pressure stoichiometric CO2
hydrogenation at different GHSV conditions (650-100000 h-1, equivalent to 0.37-49.85 NL
gcat-1 h-1) at 280 °C (46 bar (black), 92 bar (red), 184 bar (blue), and 442 bar (magenta)) and at
260 °C (331 bar, green) using commercial Cu/ZnO/Al2O3 catalyst. The filled symbols
correspond to the catalytic results obtained with the catalyst of 100-300 µm size fraction,
while the empty symbols correspond to those obtained with the catalyst of 10-20 µm size
fraction. (bottom) WTYMeOH at equilibrium conversion and selectivity at the different
GHSVs.

20