The influence of the precipitation/ageing temperature on a Cu/ZnO/ZrO 2 catalyst for methanol synthesis from H2 and CO 2

Article abstract of DOI:10.1002/cctc.201300665

For heterogeneous catalysts, the constitution of the precursor is an important parameter to adjust the properties of the active catalyst. Therefore, we examined the influence of the temperature during the precipitation process and during the ageing time in the mother liquor for a Cu/ZnO/ZrO₂ catalyst system obtained through a coprecipitation route. The variation of the temperature affects the ratio and crystallinity of the precursor phases zincian malachite and aurichalcite, as detected by powder XRD (phase and line width) and FTIR spectroscopy (characteristic asymmetric C-O stretching modes of the carbonate anions at ν=1600-1100 cm^-1). Therefore, the precatalyst surface area (A_s,BET) and pore distribution are adjustable (i.e., A_s,BET of 190 m² g^-1 was reached). The influence of the synthesis conditions on the catalysts activity for methanol production was analyzed and discussed up to the level of productivity/activity testing at 413/513 K and 40 bar total H ₂/CO₂ pressure. The best catalyst showed a methanol productivity of 9.16 mmol g_cat^-1 h^-1 (513 K, 40 bar, and gas hourly space velocity=8000) and is better than an industrial catalyst tested under the same conditions (8.34 mmol g_cat ^-1 h^-1). However, despite considerable differences in the precursor and precatalyst structure and morphology, their influence on the methanol productivity is only small. This demonstrates that the active catalyst is formed under reaction conditions. Magic for methanol: For heterogeneous catalysts, the constitution of the precursor is an important parameter to adjust the properties of the active catalyst. Therefore, we examined the influence of the temperature during the precipitation process and during the ageing time in the mother liquor for a Cu/ZnO/ZrO₂ catalyst system. The influence of the synthesis conditions on the catalyst activity for methanol production was analyzed.

Full text of DOI:10.1002/cctc.201300665

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201300665
The Influence of the Precipitation/Ageing Temperature on
a Cu/ZnO/ZrO2 Catalyst for Methanol Synthesis from H2
and CO2
Elias Frei,[a] Achim Schaadt,[b] Thilo Ludwig,[a] Harald Hillebrecht,[a] and Ingo Krossing*[a]
For heterogeneous catalysts, the constitution of the precursor
is an important parameter to adjust the properties of the
active catalyst. Therefore, we examined the influence of the
temperature during the precipitation process and during the
ageing time in the mother liquor for a Cu/ZnO/ZrO2 catalyst
system obtained through a coprecipitation route. The variation
of the temperature affects the ratio and crystallinity of the pre-
cursor phases zincian malachite and aurichalcite, as detected
by powder XRD (phase and line width) and FTIR spectroscopy
(characteristic asymmetric CꢀO stretching modes of the car-
bonate anions at n˜ =1600–1100 cmꢀ1). Therefore, the precata-
lyst surface area (As,BET) and pore distribution are adjustable
(i.e., As,BET of 190 m2 gꢀ1 was reached). The influence of the syn-
thesis conditions on the catalysts activity for methanol produc-
tion was analyzed and discussed up to the level of productivi-
ty/activity testing at 413/513 K and 40 bar total H2/CO2 pres-
sure. The best catalyst showed a methanol productivity of
9.16 mmolgcatꢀ1 hꢀ1 (513 K, 40 bar, and gas hourly space veloci-
ty=8000) and is better than an industrial catalyst tested under
the same conditions (8.34 mmolgcatꢀ1 hꢀ1). However, despite
considerable differences in the precursor and precatalyst struc-
ture and morphology, their influence on the methanol produc-
tivity is only small. This demonstrates that the active catalyst is
formed under reaction conditions.
Introduction
ð1Þ
ð2Þ
COþ2 H2 Ð CH3OH
DrH298 K ¼ ꢀ90:52 kJ molꢀ1
Global warming and the CO2 concentration in the atmosphere
are correlated to each other.[1,2] This implies that to solve the
CO2 problem is to solve the climate change problem. Many
concepts to cope with CO2 have been proposed, for example,
carbon capture and storage (CCS) or carbon capture and
usage (CCU).[3,4] One promising approach in the CCU concept
is the hydrogenation of CO2 to methanol.[5] Methanol can be
used as a fuel, base chemical, in fuel cells, or to store hydrogen
and energy from regenerative sources through the electrolysis
of water.[6–8]
CO2þ3 H2 Ð CH3OHþH2O
DrH298 K ¼ ꢀ48:97 kJ molꢀ1
Although Equation (1) leads to a 55–60% methanol yield
even at 523 K, reactions according to Equation (2) only lead to
equilibria with 15–18% methanol for the most active systems
known/published. Already, a reduction of the reaction temper-
ature from 523 to 473 K at 40 bar would imply a thermodynam-
ic doubling of the methanol yield available under equilibrium
conditions. Therefore, catalysts that are more active and work
at lower, thermodynamically more favored temperatures have
to be developed. For methanol synthesis directly from H2 and
CO2, the Cu/ZnO synergy is evidently important to obtain an
active system.[11–15] The elements that increase the activity of
this binary system the most are still under discussion. The
Handbook of Heterogeneous Catalysis ranks the supports as in
the order ZrO2 >TiO2 >SiO2, but Al2O3, Cr2O3, Ga2O3, MnOx, and
some noble metals also act as promoters and increase the
methanol yield.[16–21] Multicomponent catalysts, which consist
of four or five different metal oxide species, have been devel-
oped and implemented into a pilot plant by Saito et al.[22–24]
Based on the results of Arena et al. and others, who identified
ZrO2 as an excellent support, we worked on the active and se-
lective ternary Cu/ZnO/ZrO2 catalyst system.[25–30] An under-
standing of the chemistry of the precursors allows the control
of the performance of the catalysts, and the precipitation and
ageing conditions are crucial to control the structure and prop-
erties of the precursors.[31] Therefore, our main motivation was
to investigate this hitherto less explored but reportedly[25–27]
Currently, the industrial synthesis of methanol is performed
with a Cu/ZnO/Al2O3 catalyst at 513–543 K, 40–70 bar, and
fossil syngas (a mixture of CO and H2 with the addition of
a few percent of CO2).[9,10] Both the hydrogenation of CO and
CO2 are exothermic reactions [Eqs. (1) and (2)], which perform
best at low temperatures to give a high product yield. Howev-
er, for kinetic reasons, the reactions have to be performed at
high temperatures.
[a] E. Frei, Dr. T. Ludwig, Prof.Dr. H. Hillebrecht, Prof.Dr. I. Krossing
Institute for Inorganic and Analytical Chemistry
Freiburg Materials Research Center (FMF) and
Freiburg Institute for Advanced Studies (FRIAS)
Section Soft Matter Science, Albert-Ludwig University of Freiburg
Albertstr. 21, 79104 Freiburg (Germany)
E-mail: krossing@uni-freiburg.de
[b] Dr. A. Schaadt
Fraunhofer-Institute for Solar Energy Systems ISE
Heidenhofstraße 2, 79110 Freiburg (Germany)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201300665.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&1
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
very suitable and active ternary system and to learn if, com-
pared to a binary Cu/ZnO or ternary Cu/ZnO/Al2O3 catalysts,
the same phases in the precursors influence the catalyst struc-
ture, morphology, and performance.[32–35] The aim of this study
was to understand the relationship between catalyst synthesis
(for instance, the (micro-)structural properties of the catalyst
precursors and the final catalyst) and reactivity.
Results and Discussion
Earlier investigations have addressed the precipitation behavior
of a variety of metal salts with a focus on the pH and tempera-
ture dependence with respect to the precipitation behavior of
single-metal solutions such as copper and zinc nitrates[36] or
the influence of the ageing time on the formation of zincian
malachite and aurichalcite structures that constitute a Cu-Zn
catalyst system.[37–39] Some aspects of the temperature and pH
influence on the precipitation process for a Cu/ZnO/Al2O3
system were investigated.[35] It is common knowledge that the
aurichalcite and even better the zincian malachite phase gen-
erate these important Cu–ZnO contacts that induce very active
systems for methanol synthesis.[12,40,41]
Figure 1. Molar ratios of the metals in the precatalysts from EDX measure-
ments. The straight lines represent the targeted ratios of the metals applied
in the solutions. Every value has an average error bar of approximately
ꢂ1 atom%.
are very close to the ideal target values aimed at during the
catalysts synthesis.
The parent malachite Cu2(CO3)(OH)2 structure contains Cu2+
cations, CO32ꢀ, and OHꢀ anions. This sustains the uptake of Zn
ions, and changes the name of the material to zincian mala-
chite (Cu1ꢀxZnx)(CO3)(OH)2. Malachite crystallizes in the mono-
clinic space group P21/a and has two independent Jahn–Teller-
distorted metal positions, one with higher and one with lower
local symmetry. The Zn2+ (d10) cations preferentially occupy the
less distorted positions. As a consequence, an upper Zn uptake
of ꢁ30% results for the malachite structure.[42] The aurichalcite
Zn3Cu2(OH)6(CO3)2 structure allows the formation of a more Zn-
rich phase, which crystallizes in the P21m1 space group and is
structurally related to the pure zinc mineral hydrozincite with
the composition Zn5(OH)6(CO3)2.
Powder XRD measurements
The precursor structures were investigated with ex situ powder
XRD (PXRD) measurements. As a result of their more amor-
phous character, it was often very difficult to obtain well-re-
solved diffraction patterns. A special sample preparation tech-
nique was developed in which a collodium/solvent treatment
of the powder and the insertion of the resulting paste be-
tween two plastic films enhanced the quality of the diffracto-
grams reliably. Remarkably, the variation of the crystallinity of
the precursors was influenced very strongly by the preparation
conditions (Figure 2).
An increase of the temperature during the precipitation and
ageing time resulted in a higher crystallinity of the samples.
The PXRD patterns of these precursor materials may be in-
Here, we were interested to study the influence of the tem-
perature on a ternary Cu/Zn/Zr catalyst system with respect to
the precipitation and ageing phases of the precursors. Another
important aspect was the influence of the temperature on the
formation of the obtained phases as well as their crystallinity,
surface area, and pore distributions. In the following, the sam-
ples are named according to the experimental preparation
conditions, that is, a sample labeled as 70/70 corresponds to
a precipitation temperature of 343 K (708C) and an ageing
temperature of 343 K (708C) maintained for 24 h.
Structural and physicochemical characterization
Composition according to energy-dispersive X-ray spectro-
scopy
The stoichiometric ratio of the elements present in the pre-
pared samples, according to the method outlined in the Exper-
imental Section, was determined by EDX spectroscopy. The
mean values of the precursors and the calcined catalysts are
shown in Figure 1. Independent of the given temperatures
during the precipitation and ageing, the observed metal ratios
Figure 2. PXRD patterns of the dry catalyst precursors. The diffraction pat-
terns develop and the peaks get sharper with increasing temperature during
the precipitation process. All reflections can be indexed to either zincian
malachite (Cu1ꢀxZnx)2(CO3)(OH)2 or aurichalcite Zn3Cu2(OH)6(CO3)2
phases.[29,30,32]
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
2
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
dexed to either zincian malachite or aurichalcite phases.[32–34]
However, the diffractograms of the room temperature RT/RT
and 40/RT samples were too amorphous and thus could not
be studied reliably. The more amorphous character of these
materials was confirmed additionally by the color of these pre-
cursors after the ageing process:[37] they are greener/more tur-
quoise and not light green as observed for the other more
crystalline samples. Surprisingly, we found no reflections that
belong to any Zr phase, although EDX measurements showed
that Zr is present. It appears that an amorphous and nondif-
fracting precipitate of a zirconium hydroxide oxide mixture
was formed. This situation does not change during the calcina-
tion process, apparently because crystalline ZrO2 only forms at
much higher temperatures than 573 K. However, phase trans-
formations of Zr ceramics in steam were observed under these
mild conditions.[43–45] Therefore, rearrangement or reorganiza-
tion between these more amorphous and crystalline states is
conceivable.
for the integration into the malachite structure, and the shift
of the reflections is less pronounced. For more details on these
structures and their substitution effects see the work by Beh-
rens et al.[32,33,39,41] or others.[41]
Vibrational spectroscopy
The conclusion that a higher temperature influences the phase
development, was further supported by the FTIR spectra of the
precursor samples, in which both the crystalline and more
amorphous parts of the samples may be detected (Figure 4).
The top and the bottom spectra show the pure aurichalcite
and malachite phases, respectively, and the catalyst precursors
are depicted in between. All discussed modes and bands are
in agreement with the assignments in the literature.[33,46–48]
The OH band at n˜ =3500–3200 cmꢀ1 typically splits into two
Interestingly, the ratio between the zincian malachite and
the aurichalcite changed as indicated through the intensities
of the reflection at 2q=138 ([200] aurichalcite) as well 2q=32
and 338 (zincian malachite, cf. Figure 3). The intense reflection
at 2q=248 ([310] aurichalcite and [220] zincian malachite) is
observed for both phases and not further discussed. The 85/RT
sample favors the formation of a product that is enriched in
the aurichalcite phase (Figure 3). This appears to be independ-
ent of the molar ratio of the metals in the solution chosen for
this set of experiments as Cu/Zn/Zr=4.2:2:1.
Through the incorporation of Zn into the malachite struc-
ture, the b axis increases and the a and c axes decrease. Over-
all, the [20ꢀ1]/[21ꢀ1] reflections are shifted to higher 2q
angles, which results in a cell contraction (Figure 3). For the
85/RT sample, these reflections are less shifted to higher angles
than those of the 70/70 sample, apparently because of a lower
Zn content in the malachite structure. This is an indication that
with higher precipitation temperatures, the Zn-rich aurichalcite
phase is favored (kinetically), although zincian malachite
should form preferentially with the metal ratios used for these
experiments.[32] Probably, not enough Zn2+ ions are available
Figure 4. Stack plot of the IR spectra of the catalyst precursors. For compari-
son, the pure aurichalcite and malachite phases are given as top and
bottom traces. The region around n˜ =3350 cmꢀ1 shows the vibrational fre-
quencies of the hydroxyl groups, and the modes of the carbonate groups
appear between n˜ =1700–500 cmꢀ1
.
signals for the malachite struc-
ture because of the crystallo-
graphically different OH groups.
The spectra of the pure aurichal-
cite and the 85/RT precursors
show only a broad single band
at approximately n˜ =3350 cmꢀ1
.
The spectra of the 70/RT and 70/
70 samples also show shoulders
at approximately n˜ =3280 cmꢀ1
.
The asymmetric CꢀO stretching
modes of the carbonate anions
n˜ =1600–1100 cmꢀ1 lead to four
bands (n˜ =1558, 1511, 1415, and
1370 cmꢀ1; Figure 4) in the spec-
tra of the aurichalcite or auri-
Figure 3. Left: The PXRD patterns of the 70/70 and 85/RT samples. At the bottom the reflections as indexed from
the ICDD database are given (pure malachite 41-1390; pure aurichalcite 82-1253). Right: Magnification of
2q=30.5–33.58: Illustration of the effect of the incorporation of Zn into the malachite structure.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
3
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
chalcite-dominated (85/RT) samples. In particular, the outer
modes are characteristic for aurichalcite. The spectrum of the
pure malachite phase shows only two broad peaks at n˜ =1490
and 1379 cmꢀ1. The spectra of the more malachite-rich precur-
sors in between show two broad bands (RT/RT and 40/RT) and,
with increasing temperature, also a progressive splitting of the
vibrational modes (70/RT and 70/70), which indicates the pres-
ence of the aurichalcite phase and verifies that the precursors
exist as a mixture of these phases. The MꢀOꢀH deformation vi-
bration in the fingerprint region (n˜ =1201 and 969 cmꢀ1) is
also a feature of the aurichalcite phase and is not visible in the
malachite spectra, and the reverse is true of the vibrations at
n˜ =1097 and 1043 cmꢀ1. Other carbonate deformation bands
(in- and out-of-plane bending), OH vibrations, and lattice
modes can be seen at lower energy.
porting Information). The results indicate that malachite is the
more populated phase in these two low-crystallinity samples.
Electron microscopy and PXRD measurements
To prepare the metal oxide precatalysts, the precursors were
calcined at 573 K for 4 h. Through this thermal treatment, CO2
and H2O are released from the carbonate and the hydroxyl
groups, which changes the morphology of the precatalysts
drastically. A concomitant weight loss of approximately 20–
25% was observed.[33] This behavior is shown through electron
microscopy images (Figure 5) recorded with SEM (left) and
TEM (right). With both methods, the pictures of all samples
show an increasing porosity after the calcination process. The
morphology itself, and its change with the calcination is very
well documented by the SEM image: crystalline parts (especial-
ly in sample 40/RT) and regions (especially in sample 70/70)
are totally absent or, at least, less present after the calcination.
The effect of the temperature during the precipitation process
of the samples is clearly visible: the developing crystallinity
with the increasing temperature, mentioned during the discus-
Selected electron diffraction data
To gain more structural information to complement the PXRD
patterns, electron diffraction measurements of the almost XRD-
amorphous samples RT/RT and 40/RT were performed (Sup-
Figure 5. Electron microscopy images of the catalysts precursors and the calcined samples obtained from SEM (left) and TEM (right).
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
4
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Table 1. Volumetric characterization of the calcined precatalysts.
[a]
[b]
[c]
[c]
[c]
Precatalyst
As,BET
rp,TEM
Vp
Dav
Dmax
[m2 gꢀ1
]
[nm]
[cm3 gꢀ1
]
[nm]
[nm]
RT/RT
40/RT
70/RT
85/RT
70/70
99
106
121
146
190
19.5/3.8
15.4/6.1
12.2/8.0
7.6/2.4
1.2
25.06
22.78
10.02
9.85
19.9
23.44
12.02
6
0.58
0.56
0.58
0.6
8.6/3.2
12.94
1.78
[a] Specific surface area; [b] mean particle size, obtained from TEM statis-
tics. More information is given in the Supporting Information; [c] pore
volume (cumulative), average pore diameter and maximal pore diameter
are obtained by BJH analysis, as outlined in the Experimental Section.
Figure 6. PXRD patterns of the precatalysts. The more crystalline precursor
samples and the corresponding calcined materials give diffractograms with
sharper reflections. Only CuO (red columns) and ZnO (blue columns) phases
are assignable, and no ZrO2 is visible.
tion: a more statistical, noncrystalline arrangement of cations
and anions forms at lower temperatures (RT/RT and 40/RT sam-
ples). These large particles and their huge difference to the
smaller particles may be attributed to separate aggregation
mechanisms, which have their origin during the calcination
process. This leads to less interspersed (larger) CuO and (small-
er) ZnO/ZrO2 particles (Supporting Information).[38] It may be
suggested that the smaller particles, which are obtained for
every sample (left values in Table 1, 8.0–2.4 nm) are more
amorphous residues of separated metal oxides that change
with the precipitation protocol.
sion of the PXRD results, is observed as well (Figure 6). The RT/
RT and 40/RT samples are more amorphous, only small and less
ordered particles and structures are visible. The 70/RT sample
marks the transition to well-ordered and crystallite-dominated
samples such as 85/RT and 70/70. Although the 85/RT sample
includes very small particles and thin rods in an ordered
manner that are assigned to aurichalcite platelike structures,
the 70/70 sample contains thicker rods or column-like struc-
tures that originate from the zincian malachite (for images of
the pure aurichalcite and malachite phases see Supporting In-
formation).
It is evident from the data given in Figure 5 (TEM) and
Table 1 that the rp decreases with increasing temperature from
19.5 to 7.6 nm for the larger particles in sample 85/RT (Table 1).
The 70/70 sample includes slightly larger particles (8.6 nm), in-
duced possibly through the slightly different crystallinity. The
platelike structure of the aurichalcite (uncalcined sample 85/
RT) may be the reason for this discrepancy in the rp. The rp of
the 70/RT sample lies in between (12.2–8.0 nm) and shows
a narrow rp distribution.
The PXRD patterns of the calcined precatalysts are in good
agreement with the electron microscopy images. The more
crystalline precursor samples, which begin with the 70/RT
sample, give also sharper diffraction patterns, which are never-
theless rather broad and indicate small particle sizes with the
development of high porosity and surface area. Thus, with the
loss of H2O and CO2 through calcination, a decrease of crystal-
linity is also observed in parallel to the formation of the small
metal oxide particles. The 40/RT sample is even as a precatalyst
almost completely PXRD amorphous.
The porous character of all of the samples can be classified
as mesoporous, although some pores of the RT/RT sample lie
in the macroporous and some of the 70/70 sample in the mi-
croporous region. The pore volume (Vp) of all of the samples is
approximately 0.6 cm3 gꢀ1, which excludes the least crystalline
RT/RT sample that has a Vp of 1.2 cm3 gꢀ1 (Table 1). According
to the nitrogen isotherm and the desorption behavior (Sup-
porting Information), this may be attributed to a higher
number of pores. The average and maximal pore diameters
(Dav and Dmax) give information about the pore distribution.
The Dmax value shows in which region the pores are most
populated. Both values are comparable for the low-tempera-
ture samples (20–25 nm). For the RT/RT sample, the difference
between Dav and Dmax is approximately D=5 nm, which is
slightly larger than that of the 40/RT sample that has a less dis-
tributed pore structure (D=1 nm). The 70/RT sample (D=
2 nm) shows the same behavior, but the pore size itself is
halved (10–12 nm). The 85/RT sample shows the smallest Dav of
all samples (9.85 nm). This rp distribution is reasonable, if one
regards the nanostructures of the catalysts and that the pores
result from the size of these particles. The platelike structure of
this aurichalcite-rich sample may again be the reason for the
Surface areas and pore sizes
Most astonishingly, with the increasing crystallinity, the specific
surface area (As,BET) of the calcined samples also increased
(Table 1) and almost doubles from 99 (RT/RT) to 190 m2 gꢀ1
(70/70). A second trend gives additional information on the
particle size (rp) and the pore distributions and diameters (Dav
and Dmax). Every sample consists of two different kinds of parti-
cles, smaller (2.4–8.0 nm) and larger ones (7.6–19.5 nm). Ac-
cording to PXRD and EDX measurements, both particles are
constituted from CuO and/or ZnO (Supporting Information).
ZrO2 is too amorphous to be visible in the PXRD/ED diffraction
measurements.
The structural characteristics of the differently sized particles
in the samples may be attributed to their mode of precipita-
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
5
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
larger Dmax, if compared to the columnar malachite-rich 70/70
sample (6.0–1.96 nm). Through the calcination process, the
structure of the 70/70 sample stays almost unaffected, and
a porous columnar morphology was obtained. The volume
inside these columns corresponds almost to a microporous
character (inner pores). The higher Dav value describes the ag-
glomeration of these porous columns and their stacking (outer
pores). The discrepancy between the Dav and the Dmax of ap-
proximately D=10 nm, is caused by the higher weighting of
larger pores in the statistics of the Dav value.
Table 2. TPR data for the precatalysts: Tonset and Tmax and the H2
consumption.
[a]
[a]
[a]
Precatalyst
Tonset
Tmax,1
Tmax,2
H2 consumption
[mmolH2 gꢀ1
]
RT/RT
40/RT
70/RT
85/RT
70/70
402
393
389
388
388
468
471
451
449
459
8.4
8.5
8.2
6.9
8.7
439
423
[a] Temperature in K (error bar ꢂ1 K). See the TPR part of the Experimen-
It appears that the pores of the nanostructured catalysts are
introduced by two different mechanisms: larger pores that
originate from the agglomeration of particles during the pre-
cipitation (outer pores), and smaller pores are introduced
during the calcination process (inner pores).
tal Section for further information.
ture, Tonset decreases from 402 K (RT/RT) to values below 400 K.
The H2 consumption during the reduction process to Cu0 was
also determined (Table 2). The RT/RT, 40/RT, and 70/RT samples
show
a comparable H2 consumption between 8.2 and
Chemical characterization of the precatalysts
8.5 mmol. Below and above this range, the H2 consumption of
the 85/RT and 70/70 samples were measured at 6.9 and
8.7 mmol, respectively. The huge H2 consumption is also the
reason for the higher Tmax,2 value of the 70/70 sample. It seems
that the crystallinity of the sample has no significant influence
on the amount of H2 that reacts but it affects the temperature-
dependent behavior of the catalysts strongly with its different
constitution and different kinds of Cu species. The results of
TPO experiments on the catalysts are shown in Figure 7(right).
As a result of the low sensitivity of these measurements, the
information that can be extracted is rather qualitative. All of
the samples show the same curve progression with two
maxima at more or less 380 and 470 K. This excludes the 70/70
sample that has a third maximum at 420 K. This suggests the
presence of two different kinds of species, which vary in activi-
ty to be oxidized. This is also in accordance with the TPR pro-
files, in which a shoulder observed in several cases suggests
the same.
The measurement of the reduction behavior of the precatalysts
as a function of temperature gives valuable information to
create an efficient activation protocol suitable for successful
catalyst testing in the reactor. Additionally, an indication of if
different kinds of active species are present is obtained. Tem-
perature-programmed oxidation (TPO) experiments complete
these measurements. The Tonset value gives the starting region
of the reaction, and the Tmax,n values are at a shoulder and on
the top of the curve.
The temperature-programmed reduction (TPR) measure-
ments of the catalysts are shown in Figure 7(left). With increas-
ing precipitation temperature (RT to 858C), Tmax (TPR) shifts to
lower values. The maximum at 468 K of the RT/RT sample is la-
beled for comparison. The curves of the 40/RT and 85/RT sam-
ples show in addition to Tmax,2, shoulders (Tmax,1) at 439 and
423 K, respectively (Table 2). The maximum of the 70/70
sample is reached at a higher temperature (459 K). The Tonset
describes the start of the reduction of the catalysts and shows
the same tendency. With the increasing precipitation tempera-
Figure 7. Temperature dependence of the reduction (left) and oxidation (right) behavior of all calcined precatalysts tested by TPR (left) and TPO (right). The
experiments were performed with a heating rate of 2 Kminꢀ1, heated to 573 K and held for 15 min. As a reducing agent 5% H2 in Ar and as an oxidizing
agent 5% O2 in He were used.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
6
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Active Cu surface area
through the smallest value to obtain the activity differences
[%].
The precatalyst surface area (As,BET) gives information about the
general physical surface properties of the catalyst (Table 1) but
there is no correlation with the catalyst performance, which is
often related to the specific surface area, for example, that of
Cu0. Through controlled pulse experiments with N2O, the re-
duced catalysts become partially oxidized from Cu0 to CuI.
Therefore, only the Cu surface area (As,Cu) is determined
(Table 3). It has been suggested in the literature that the As,Cu
All catalysts investigated in this study were tested at least
twice. Between each test, they were gradually oxidized upon
standing in air for more than 30 days. After a second identical
activation procedure, their productivity was as high as before
without any loss of activity. Additionally, a long-term test was
performed, an example of which is shown in Figure S7 in the
Supporting Information for the 70/70 sample at different tem-
peratures. This test included reactions at doubled flow rates
(8000 NmLhꢀ1 > 75% H2 and 25% CO2) to adjust to more dif-
ferential-like testing conditions.
Table 3. Cu surface area and the performance of the catalysts in the
reactor.
The productivity of the catalysts appears to be directly relat-
ed to the As,Cu. Therefore, the RT/RT and 40/RT samples with
the highest As,Cu values also show the highest methanol
yields of 8.11 and 9.16 mmolMeOH gcatꢀ1 hꢀ1, respectively. The
other samples with only approximately half of the As,Cu value
show a lower productivity, but not as low as one would have
expected (76–87% of that of the best entry or 6.93–
7.94 mmolMeOH gcatꢀ1 hꢀ1).
[a]
Catalyst As,Cu
Productivity[b]
[mmolMeOH gcatꢀ1 hꢀ1
Intrinsic activity[c]
Onset
activity[d]
[m2 gꢀ1
]
]
[mmolMeOH
m
ꢀ2 hꢀ1
]
RT/RT
40/RT
70/RT
85/RT
70/70
27.34
29.03
18.00
14.30
16.59
8.11
9.16
7.94
6.93
7.80
0.30
0.32
0.44
0.48
0.47
1.00
1.00
1.14
1.06
1.19
PXRD data were collected for all catalysts after the per-
formance tests. The samples were prepared under as inert con-
ditions as possible, and the patterns are astonishingly uniform
compared to the very different precursor patterns shown in
Figure 6. All measurements show only Cu and ZnO as detect-
[a] Surface area of Cu from the N2O pulse technique at 373 K, as outlined
in the Experimental Section. [b] 513 K, 40 bar, 75% H2, 25% CO2, GHSV=
4000. [c] Intrinsic activity=Productivity/As,Cu. [d] relative activity measured
at 415 K and 40 bar, value for the “onset activity”.
influences the catalyst performance.[40,23] Interestingly, the RT/
RT and 40/RT samples with the lowest crystallinity show the
highest As,Cu values of 27 and 29 m2 gꢀ1. The most crystalline
samples 85/RT and 70/70 only achieve half of this (14 and
16 m2 gꢀ1). The 70/RT sample with a value of 18 m2 gꢀ1 rather
adheres to the lower values (Table 3). It is surprising that the
samples with a high As,BET (precatalysts: 85/RT: 146 m2 gꢀ1 and
70/70: 190 m2 gꢀ1) show a comparably low As,Cu (reduced cata-
lysts). This trend might be because the reducible sites of the
more crystalline samples are less easily accessible or that the
reducible sites of the more amorphous samples are better dis-
persed and the Cu parts of these metal oxides react more
readily.
Performance of the catalysts
Figure 8. PXRD patterns of the catalysts in a reduced state after the reactor
test. The Cu (red columns) and ZnO (blue columns) phases are the only
ones detectable.
The performance of the catalysts may be documented by two
values: the productivity, which gives the quantity of methanol
produced under the chosen conditions, and the intrinsic activi-
ty, which shows the catalyst behavior related to As,Cu. The pro-
ductivity was determined by using a thermal conductivity de-
tector (TCD) with a calibrated GC. The onset activity was as-
sessed as follows: starting at 393 K, a 30 min catalyst test was
performed. The temperature was then increased in steps of 5 K
and held for 30 min. For each temperature, three GC measure-
ments were performed. For a better comparison and better
signal-to-noise ratio in the measurements, we discuss the GC
areas at 415 K and not the “starting temperature” at which the
first methanol signal was detected. Here, the methanol produc-
tion was determined by a more sensitive GC method. The inte-
grals were compared directly to each other and divided
able phases (Figure 8). Only the 70/RT sample shows a small re-
flection at 2q=388, which is assigned to CuO that results from
the trace oxidation of Cu0 during sample preparation. In con-
trast to the precursor or precatalyst patterns, these diffracto-
grams are very similar with respect to the reflection intensity
and width. Therefore, under these physical (temperature and
pressure) and chemical (gas mixture) conditions, the structure
and morphology of the catalysts seem to adapt and become
more uniform. This would account for the rather small differen-
ces in the activity and productivity versus the large differences
in the physical and chemical properties of the precatalysts.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
7
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
a hint as to how the catalyst
properties are connected to
Table 4. Comparison to published catalyst productivities, which includes their testing conditions.
Entry Gas ratios
Gas flow
[NmLhꢀ1
T
[K]
p
Productivity
Ref.
their
performance
generated
(Table 3).
from
]
[bar] [mmolMeOH gcatꢀ1 hꢀ1
]
Catalysts
precursors with a higher crystal-
linity have a lower methanol
productivity and As,Cu but surpris-
ingly show the highest intrinsic
1
2
3
4
5
6
7
75% H2, 25% CO2
18000
3000
8800
50000
–
523 50
493 10
473 10
513 50
503 30
513 40
513 40
20.78
11.6
2.03
36.88
9.5
[23]
[36]
[27]
[55]
72% H2, 10% CO, 4% CO2 in He
69% H2, 23% CO2, 8% N2
66% H2, 33% CO, 1% CO2
72% H2, 24% CO2, 4% Ar
75% H2, 25% CO2
[56]
activity
of
more
than
4000
4000
9.16
8.34
This work, Table 3
Industrial catalyst[a]
0.44 mmolMeOH mꢀ2 hꢀ1. The more
amorphous samples are intrinsi-
cally less active with values of
0.30 and 0.32 mmolMeOH mꢀ2 hꢀ1
(maximum difference 37%). This
75% H2, 25% CO2
[a] Industrial Cu/ZnO/Al2O3 catalyst system.
For a better comparison, an industrial Cu/ZnO/Al2O3 catalyst
was also tested in our setup under the same conditions as the
samples described in Table 3 (513 K, 40 bar, 75% H2, 25% CO2,
and 4000 NmLhꢀ1). A productivity of 8.34 mmolMeOH gcatꢀ1 hꢀ1
was obtained for this system (Table 4), which is in the region
of the second best value (RT/RT, 8.11 mmolMeOH gcatꢀ1 hꢀ1) and
approximately 10% inferior to our best system (40/RT,
9.16 mmolMeOH gcatꢀ1 hꢀ1).
confirms the conclusion of the TPR experiments, which sug-
gest the existence of different Cu sites with high and attenuat-
ed activities. The onset activity values measured at 415 K sup-
port these arguments and show a higher activity for the cata-
lysts prepared from the precursors of higher crystallinity at low
temperature (Table 3). Additionally, the catalysts derived from
the more aurichalcite-rich 85/RT sample are less active than
those derived from zincian malachite (70/RT, 70/70). Informa-
tion on the “early activity” of the catalysts, in the context of
their structural properties, gives valuable hints to develop cata-
lytic systems, which should work at temperatures that are
more thermodynamically favored. The materials described in
entries 6 and 7 (Table 4) will be used as benchmarks for com-
parison with independent systems currently under develop-
ment by our group.
Our productivity results obtained from catalyst particles of
0.5–0.8 mm diameter (Table 3), are competitive compared to
the best literature values given in Table 4. The difficulty in the
interpretation of these data lies in the variation of the parame-
ters, such as pressure, temperature, and feed gas rate, by other
groups (Table 4). In the literature, mostly percent values of CO2
conversion and selectivity to methanol are discussed or the
productivity per milliliter of catalyst is mentioned, but without
the material density this is hard to compare. Another problem
is the addition of CO to the gas feed, which changes the reac-
tion mechanism and the thermodynamics. The values that are
the most comparable to our system and the conditions we
have chosen are entry 1 from Saito et al. and entry 5 from
Behrens et al. in Table 4. The productivity of entry 1 of
20.78 mmolMeOH gcatꢀ1 hꢀ1 was reached at a higher pressure of
50 bar at 523 K. Additionally, a much higher gas flow of
18000 NmLhꢀ1 was used to enable a higher productivity. The
value of entry 5 of 9.5 mmolMeOH gcatꢀ1 hꢀ1 was achieved at
a lower pressure of 30 bar at 503 K, but information on the gas
flow used is not given so a classification and direct comparison
is not possible.
Conclusion
The influence of the temperature during the precipitation and
ageing phase of a coprecipitation process on the structure and
morphology of a precursor up to the level of catalyst per-
formance was investigated for a Cu/ZnO/ZrO2 system. The
main findings for the precursors are:
*
By increasing the precipitation temperature up to 358 K,
the crystallinity of the precursor strongly increased (powder
XRD, ED, and electron microscopy techniques). More amor-
phous samples are formed at room temperature.
*
The precipitation temperature guides which precursor
phases will prevail. At 358 K, the formation of aurichalcite is fa-
vored, and at lower temperatures, zincian malachite formation
dominates at the chosen metal ratios (powder XRD and FTIR
spectroscopy measurements).
As an example, we investigated the possibility of tuning the
catalysts productivity with higher flow rates to obtain more
differential-like conditions for the catalyst 70/70 (Table 3).
Thus, if we used a feed of 8000 NmLhꢀ1, a productivity of
*
The typical aurichalcite/malachite phases and structural
13.65 mmolMeOH gcatꢀ1 hꢀ1
at
523 K
(before
tuning:
peculiarities known for the Cu/Zn system were also observed
in the presence of zirconia. Only small changes in the catalyst
system geneses led to considerable alterations of their mor-
phology and catalytic properties, also compared to systems
discussed in the literature.[41,33]
8.20 mmolMeOH gcatꢀ1 hꢀ1) and 12.20 mmolMeOH gcatꢀ1 hꢀ1 at 513 K
(before tuning: 7.80 mmolMeOH gcatꢀ1 hꢀ1) was measured. This
demonstrates the competitiveness of this catalytic system to
all comparable systems mentioned in Table 4. Entries 2 and 4
are not directly comparable, as they use CO in the feed gas
and should only be seen as an orientation.
The structural differences found for the precursors also man-
ifested themselves in the crystallinity of the precatalyst and
bulk structure. Distinct and precipitation-temperature-depen-
dent trends for the particle size, pore volume and size, specific
surface areas, and redox behavior were observed. Surprisingly,
Although all the prepared catalysts have the same chemical
composition (from EDX spectroscopy), their behavior varies
strongly. The intrinsic activities with respect to the As,Cu give
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
8
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
tape preparations were less suitable. All diffraction patterns shown
are background corrected.
these large differences influenced the performances of the cat-
alysts only slightly. Overall, the productivity of the catalysts
generated from the more amorphous precursors was 10–25%
higher than that of the more crystalline materials. This might
be because of their higher As,Cu values by a factor of almost
two, although the favorable intrinsic activity would favor cata-
lysts generated from more crystalline samples. Yet the distribu-
tion of the productivity and activity results are less scattered
than one would have anticipated from the rather different
properties of the precatalysts. This notion is supported by
powder XRD measurements of the active catalysts after the
testing experiments, which show very similar diffraction pat-
terns for all samples (Figure 8). Probably, structural reorganiza-
tions take place during the reduction to the active catalyst and
the following test period under 40 bar H2/CO2 atmosphere. It
appears that this heterogeneous catalyst system transforms to
a more uniform morphology by contact with the feed gas
under the reaction conditions, under which it develops its
active state. This also implies that predictions, for instance, on
whether a certain precursor morphology is favorable for an op-
timal catalyst performance, are very difficult without actual cat-
alyst testing.
TPO and TPR measurements
The oxidation and reduction behavior of the calcined catalysts
were studied by using a ThermoFisher (TPDRO 1100) plant with
TCD. For the oxidation process, a mixture of 5% O2 in He was
used. The reduction was realized with a mixture of 5% H2 in Ar.
Each reaction was heated to 573 K (heating rate 2 Kminꢀ1) and
held at that temperature for 15 min.
N2O titration measurements
The metal surface area of the active catalyst, in our case As,Cu was
determined by using a ThermoFisher (TPDRO 1100) unit with TCD
using the N2O pulse chemisorption technique.[50–52]
Surface area and pore volume measurements
[53]
The specific surface area (As,BET
)
and the pore volumes (Vp, by the
Barrett–Joyner–Halenda (BJH) approach[54]) of the samples were de-
termined by the measurement of nitrogen adsorption/desorption
isotherms at 77 K by using a ThermoFisher Sorptomatic instrument.
For sample pretreatment, the catalysts were degassed (10ꢀ3 mbar)
and dried (5 h at 423 K).
Experimental Section
In the upper part and the following, we addressed the dried pre-
cipitate as precursor, the calcined metal oxide samples as precata-
lysts and the active, reduced ones as catalysts.
EDX spectroscopy
The elemental constitutions of the calcined samples were mea-
sured through EDX spectroscopy under high vacuum by using an
Oxford Inca X-act system.
Preparation of the calcined precatalysts
SEM
Samples of Cu/ZnO/ZrO2 precatalysts with a metal ratio of Cu/Zn/
Zr=4.2:2:1 were prepared by a coprecipitation method. From two
syringes on perfusors, one with an aqueous Na2CO3 solution
(1.1m) and the other with an aqueous metal nitrate solution
(1.0m), the reactants were added dropwise into water (250 mL) at
different well controlled constant temperatures between 298 and
358 K. The pH in the precipitation vessel was kept constant at 7
(Methrom Titrino DMS 716). After the precipitation process, the
precursors were aged in the mother liquor for 24 h under vigorous
stirring at different well-controlled temperatures, filtered, washed
(five times with 200 mL distilled water), dried overnight at 373 K,
and finally calcined in a constant nitrogen flow (starting at RT with
a heating rate of 1.5 Kminꢀ1 then holding at 573 K for 4 h).
The morphology and the surface texture were studied by SEM by
using a QUANTA FEG 250 instrument. The samples were fixed on
a carbon grid, charged on an aluminum carrier and analyzed under
high vacuum.
TEM and electron diffraction (ED)
The particle size and distribution were investigated by TEM by
using a Zeiss LEO 912 Omega instrument using the GATAN soft-
ware. The mean particle size was determined by statistical averag-
ing over a few hundred particles to eliminate spherical deviations.
The samples were dispersed in a solvent (e.g., acetone) and
dipped on a Cu grid. Different particle species and particles with
different sizes were characterized by ED.
Characterization of the samples
XRD analysis
FTIR spectroscopy
PXRD patterns of the precursor powders were recorded by using
The FTIR spectra were measured by using a Nicolet/Thermo
Magna-IR 760 spectrometer with a diamond attenuated total re-
flectance (ATR) unit (180–4000 cmꢀ1).
a
Stoe Transmission-Diffractometer (STADI P, Modell II) with
a Ge(111) monochromator and CuKa radiation (constant slit width).
Qualitative phase analysis was performed by comparison with ref-
erence patterns from the International Centre for Diffraction Data
(ICDD) database using the EVA software.[49] The samples were treat-
ed with a collodium solvent mixture (ethanol/diethyl ether/iso-pen-
tylacetate) to form a paste, which was fixed between two foils.
With this procedure, it was possible to obtain reliable and well-de-
fined diffraction patterns. Other techniques such as capillary or
Catalyst testing
The catalyst performance was tested at 40 bar total pressure and
at different temperatures (413–543 K) by using a continuous-flow
tube reactor system from Miniplant with an inner diameter of
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
9
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[22] M. Saito, T. Fujitani, M. Takeuchi, T. Watanabe, Appl. Catal. A 1996, 138,
311.
[23] M. Saito, T. Fujitani, I. Takahara, T. Watanabe, M. Takeuchi, Y. Kanai, K.
Moriya, T. Kakumoto, Energy Convers. Manage. 1995, 36, 577.
[24] K. Ushikoshi, K. Mori, T. Kubota, T. Watanabe, M. Saito, Appl. Organomet.
Chem. 2000, 14, 819.
[25] F. Arena, G. Italiano, K. Barbera, S. Bordiga, G. Bonura, L. Spadaro, F.
Frusteri, Appl. Catal. A 2008, 350, 16.
[26] F. Arena, G. Italiano, K. Barbera, G. Bonura, L. Spadaro, F. Frusteri, Catal.
Today 2009, 143, 80.
[27] F. Arena, K. Barbera, G. Italiano, G. Bonura, L. Spadaro, F. Frusteri, J.
Catal. 2007, 249, 185.
[28] Y. Nitta, O. Suwata, Y. Ikeda, Y. Okamoto, T. Imanaka, Catal. Lett. 1994,
26, 345.
[29] Y.-W. Suh, S.-H. Moon, H.-K. Rhee, Catal. Today 2000, 63, 447.
[30] R. Raudaskoski, E. Turpeinen, R. Lenkkeri, E. Pongracz, R. L. Keiski, Catal.
Today 2009, 144, 318.
6 mm. The catalyst samples were fixed in the middle of the tube
reactor with a metal plate (reaction inert) and high-temperature
fibers in contact with a thermocouple. The catalyst powder
(ꢁ0.7 g for each test) was pelletized, crushed, and sieved to obtain
defined particles of 0.5–0.8 mm diameter. The gas flows (H2, N2 and
CO2) and pressures were controlled by mass flow controllers
(MFCs, Bronkhorst) and magnetic valves (Bꢁrkert). H2/CO2 reaction
mixtures with molar ratios of 75% H2 and 25% CO2 at gas hourly
space velocities (GHSV) of 8000 NmL were fed. The catalysts were
activated at different temperatures (373–453 K) and H2 concentra-
tions (4–100%), see Supporting Information. The product stream
was directly analyzed by GC (Agilent 6890 N) with TCD. Details of
the catalyst test station and accompanying analytical data are de-
posited with the Supporting Information.
[31] R. Raudaskoski, M. Niemelae, R. Keiski, Top. Catal. 2007, 45, 57.
[32] M. Behrens, F. Girgsdies, Z. Anorg. Allg. Chem. 2010, 636, 919.
[33] M. Behrens, F. Girgsdies, A. Trunschke, R. Schlçgl, Eur. J. Inorg. Chem.
2009, 1347.
[34] B. Bems, M. Schur, A. Dassenoy, H. Junkes, D. Herein, R. Schlçgl, Chem.
Eur. J. 2003, 9, 2039.
Acknowledgements
We thank the DBU for funding this project (Az: 29629-31). Addi-
tionally we want to acknowledge Dr. Ralf Thomann for helping
with the SEM, TEM, EDX and ED data.
[35] J. L. Li, T. Inui, Appl. Catal. A 1996, 137, 105.
[36] M. Behrens, D. Brennecke, F. Girgsdies, S. Kißner, A. Trunschke, N. Nasru-
din, S. Zakaria, N. F. Idris, S. B. A. Hamid, B. Kniep, R. Fischer, W. Busser,
M. Muhler, R. Schlçgl, Appl. Catal. A 2011, 392, 93.
[37] S. Zander, B. Seidlhofer, M. Behrens, Dalton Trans. 2012, 41, 13413.
[38] B. L. Kniep, F. Girgsdies, T. Ressler, J. Catal. 2005, 236, 34.
[39] M. Behrens, I. Kasatkin, S. Kꢁhl, G. Weinberg, Chem. Mater. 2010, 22,
386.
Keywords: copper
catalysis · methanol synthesis · zinc · zirconium
· energy conversion · heterogeneous
[1] http://www.wri.org/chart/global-co2-emissions-growth-select-sectors-
2000-2005.
[2] http://www.ipcc.ch/publications_and_data/publications_and_data_re-
ports.shtml.
[40] M. Saito, M. Takeuchi, T. Fujitani, J. Toyir, S. Luo, J. Wu, H. Mabuse, K.
Ushikoshi, K. Mori, T. Watanabe, Appl. Organomet. Chem. 2000, 14, 763.
[41] M. Behrens, J. Catal. 2009, 267, 24.
[42] N. Perchiazzi, S. Merlino, Eur. J. Mineralogy Vol. 18, 2006, 787.
[43] J. Chevalier, B. Cales, J. M. Drouin, J. Am. Ceram. Soc. 1999, 82, 2150.
[44] J. R. Kelly, I. Denry, Dent. Mater. 2008, 24, 289.
[45] R. Nielsen in Ullmann’s Encyclopedia of Industrial Chemistry, Wiley-VCH,
Weinheim, 2000.
[3] http://www.umweltdaten.de/publikationen/fpdf-k/k3074.pdf.
[4] http://www.greenpeace-energy.de/fileadmin/docs/sonstiges/Green-
peace_Energy_Gutachten_Windgas_Fraunhofer_Sterner.pdf.
[5] G. A. Olah, G. K. S. Prakash, A. Goeppert, J. Am. Chem. Soc. 2011, 133,
12881.
[6] G. Olah, A. Goeppert, G. K. S. Prakash, Beyond Oil and Gas: The Methanol
Economy, 2005.
[7] F. Asinger, Methanol—Chemie und Energierohstoff, 1986.
[8] G. A. Olah, Angew. Chem. Int. Ed. 2013, 52, 104; Angew. Chem. 2013,
125, 112.
[46] M. Schmidt, H. D. Lutz, Phys. Chem. Miner. 1993, 20, 27.
[47] D. Stoilova, V. Koleva, V. Vassileva, Spectrochim. Acta Part A 2002, 58,
2051.
[48] F. Stone, D. Waller, Top. Catal. 2003, 22, 305.
[9] K. Klier, D. D. Eley, H. Pines, P. B. Weisz, Adv. Catal. 1982, 31, 243.
[10] J. C. J. Bart, R. P. A. Sneeden, Catal. Today 1987, 2, 1.
[11] M. S. Spencer, Top. Catal. 1999, 8, 259.
[12] T. Fujitani, J. Nakamura, Appl. Catal. A 2000, 191, 111.
[13] Y. Kanai, T. Watanabe, T. Fujitani, M. Saito, J. Nakamura, T. Uchijima,
Catal. Lett. 1994, 27, 67.
[49] D. S.-. Diffrac(plus)Brucker 2002.
[50] O. Hinrichsen, T. Genger, M. Muhler, Chem. Eng. Technol. 2000, 23, 956.
[51] J. W. Evans, M. S. Wainwright, A. J. Bridgewater, D. J. Young, Appl. Catal.
1983, 7, 75.
[52] G. C. Chinchen, C. M. Hay, H. D. Vandervell, K. C. Waugh, J. Catal. 1987,
103, 79.
[14] J. Nakamura, I. Nakamura, T. Uchijima, Y. Kanai, T. Watanabe, M. Saito, T.
Fujitani, Catal. Lett. 1995, 31, 325.
[15] J. Bailie, C. Rochester, G. Millar, Catal. Lett. 1995, 31, 333.
[16] J. Toyir, P. R. de La Piscina, J. L. G. Fierro, N. Homs, Appl. Catal. B 2001,
29, 207.
[53] K. S. W. Sing, Pure Appl. Chem. 1985, 57, 603.
[54] E. P. Barrett, L. G. Joyner, P. P. Halenda, J. Am. Chem. Soc. 1951, 73, 373.
[55] Y. Zhang, Q. Sun, J. Deng, D. Wu, S. Chen, Appl. Catal. A 1997, 158, 105.
[56] S. Zander, E. L. Kunkes, M. E. Schuster, J. Schumann, G. Weinberg, D.
Teschner, N. Jacobsen, R. Schlçgl, M. Behrens, Angew. Chem. 2013, 125,
6664.
[17] M. Kilo, J. Weigel, A. Wokaun, R. A. Koeppel, A. Stoeckli, A. Baiker, J. Mol.
Catal. A 1997, 126, 169.
[18] I. A. Fisher, H. C. Woo, A. T. Bell, Catal. Lett. 1997, 44, 11.
[19] M. Sahibzada, D. Chadwick, I. S. Metcalfe, Catal. Today 1996, 29, 367.
[20] C. Frçhlich, R. A. Kçppel, A. Baiker, M. Kilo, A. Wokaun, Appl. Catal. A
1993, 106, 275.
Received: August 9, 2013
Revised: January 17, 2014
[21] J. B. Hansen, H. Knoezinger, G. Ertl, J. Weitkamp, Handbook of Heteroge-
neous Catalysis, Wiley-VCH, Weinheim, 1997.
Published online on && &&, 0000
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&10
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
Magic for methanol: For heterogene-
ous catalysts, the constitution of the
precursor is an important parameter to
adjust the properties of the active cata-
lyst. Therefore, we examined the influ-
ence of the temperature during the pre-
cipitation process and during the
ageing time in the mother liquor for
a Cu/ZnO/ZrO2 catalyst system. The in-
fluence of the synthesis conditions on
the catalyst activity for methanol pro-
duction was analyzed.
E. Frei, A. Schaadt, T. Ludwig,
H. Hillebrecht, I. Krossing*
&& – &&
The Influence of the Precipitation/
Ageing Temperature on a Cu/ZnO/
ZrO2 Catalyst for Methanol Synthesis
from H2 and CO2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&11
&
ÞÞ
These are not the final page numbers!