Calcium oxide supported on monoclinic zirconia as a highly active solid base catalyst

Article abstract of DOI:10.1002/cctc.201300676

Calcium oxide supported on ZrO₂ is a highly active catalyst for base-catalyzed reactions such as aldol-type reactions and transesterification reactions. The role of key parameters during preparation, that is, impregnation versus precipitation, heat treatment, and metal oxide loading on the basicity and catalytic activity were investigated for CaO supported on ZrO₂. An impregnation of 10 wt % CaO on monoclinic zirconia followed by heat treatment at 600°C resulted in high activity for the self-condensation reaction of acetone. An evaluation of a series of CaO/ZrO₂ samples with different loadings showed that the activity increased for impregnated amounts per gram catalyst of 0-10 wt % CaO, and at higher loading the activity decreased as a result of a decrease in dispersion. The number of strong base sites (calculated from CO₂ desorbed at temperatures higher than 625°C) correlated with the activity. For MgO, CaO, SrO, and BaO on zirconia the catalytic activity increased as the ionic radius of the metal cation increased, suggesting the impact of base strength on catalytic performance. Look at the strong base! If prepared in an optimized way, CaO/ZrO₂ is a highly active solid base catalyst. Calcination temperature and the preparation method (impregnation or deposition) have a strong influence on the number of strong base sites, which correlate positively with the activity of the material. Copyright

Full text of DOI:10.1002/cctc.201300676

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201300676
Calcium Oxide Supported on Monoclinic Zirconia as
a Highly Active Solid Base Catalyst
Anne Mette Frey, Tomas van Haasterecht, Krijn P. de Jong, and Johannes Hendrik Bitter*[a]
Calcium oxide supported on ZrO2 is a highly active catalyst for
base-catalyzed reactions such as aldol-type reactions and trans-
esterification reactions. The role of key parameters during
preparation, that is, impregnation versus precipitation, heat
treatment, and metal oxide loading on the basicity and catalyt-
ic activity were investigated for CaO supported on ZrO2. An im-
pregnation of 10 wt% CaO on monoclinic zirconia followed by
heat treatment at 6008C resulted in high activity for the self-
condensation reaction of acetone. An evaluation of a series of
CaO/ZrO2 samples with different loadings showed that the ac-
tivity increased for impregnated amounts per gram catalyst of
0–10 wt% CaO, and at higher loading the activity decreased as
a result of a decrease in dispersion. The number of strong base
sites (calculated from CO2 desorbed at temperatures higher
than 6258C) correlated with the activity. For MgO, CaO, SrO,
and BaO on zirconia the catalytic activity increased as the ionic
radius of the metal cation increased, suggesting the impact of
base strength on catalytic performance.
Introduction
Base catalysis is currently gaining in interest owing to its rele-
vance in the field of biomass conversions such as transesterifi-
cation and aldol condensation reactions.[1–4] Compared to ho-
mogeneous base catalysts solid base catalysts offer several ad-
vantages such as easy regeneration and separation and lower
corrosivity.[5–7] One drawback of bulk solid bases is, however,
that the number of active sites per unit mass is generally low.
This drawback can be overcome by dispersing the active
phase as nanosized particles on a support.[8,9]
thus expected to be more catalytically active in reactions re-
quiring a strong base.[23–25]
Previously it has been demonstrated that supporting CaO
on carbon nanofibers (CNF) resulted in small particles of ap-
proximately 3 nm that are active as a solid base catalyst for
aldol reactions and transesterification reactions.[26] The advant-
age of using a carbon support is that the interaction between
the support and the active phase is kept as low as possible.
Therefore, it is expected that the catalysis of nanosized CaO
will not be affected by the support (see below).
Alkaline earth metal oxides (MgO, CaO, SrO, and BaO) have
been investigated as base catalysts for multiple different reac-
tion such as isomerization reactions,[10,11] Claisen–Schmidt con-
densations,[12,13] Knoevenagel condensation,[14] Michael addi-
tions,[15] Tishchenko reactions,[16] aldol condensations,[17] and
transesterification reactions.[18] For more examples and details
we refer to a landmark review by Corma and Iborra.[1]
The inert nature of carbon support makes it possible to gain
fundamental insight on the properties of nanoparticles. It is,
for example, possible to investigate the role of the cation in
basic oxides. By using CNF as a support, nanosized alkaline
earth metal oxide particles containing the same number of
active sites could be prepared,[26] and the catalytic activity of
such nanosized particles correlated directly with the base
strength, which again correlates with, for example, the ionic
radius of the alkaline earth metal ion.
It is well known that not only the number of actives sites
but also the base strength of the sites are important for the
catalytic performance of bulk solid base catalysts.[19–22] As MgO
and CaO are readily available, cheap, and environmentally
benign, these are most promising candidates among the alka-
line earth metal oxide for catalytic applications. CaO is gaining
a lot of interest because it is a stronger base than MgO and
If using traditional oxide supports, there is a possibility of
strong interaction between the support and the active phase,
that is, formation of a mixed oxide that often exhibits lower
basicity than the active phase itself.[27,28] The influence of
mixed oxide formation on catalysis has, to the best of our
knowledge, not been reported in the literature for solid base
catalysts. Therefore, we performed an initial screening study
on the role of the support (CNF, SiO2, TiO2, Al2O3, and ZrO2) on
the activity of supported CaO for the base-catalyzed self-con-
densation of acetone (Scheme 1).
[a] Dr. A. M. Frey, T. van Haasterecht, Prof. K. P. de Jong, Prof. Dr. J. H. Bitter+
Department of Inorganic Chemistry and Catalysis
Debye Institute
Utrecht University
Universiteitsweg 99, 3584 CG, Utrecht (The Netherlands)
[+] Present address:
From this screening, the ZrO2 support provided the most
active catalysts and, therefore, this support was selected for
further studies. Zirconia has already been shown to be a good
choice as a catalyst support material in the processing of bio-
mass in liquid-phase reactions.[29,30] Zirconia as a support for
Wageningen University and Research Center
(The Netherlands)
E-mail: harry.bitter@wur.nl
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201300676.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 9
&
1
&
ÞÞ
These are not the final page numbers!

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Scheme 1. Self-condensation of acetone to diacetone alcohol (DAA).
CaO has been reported in the literature as a suitable catalyst
for transesterification reactions in preliminary studies.[31–33]
Wang et al.[31] compared precipitated CaO–ZrO samples with
samples in which the calcium was introduced on tetragonal
zirconia, t-ZrO2, by impregnation, or in which CaO and t-ZrO2
were physically mixed. For the precipitated sample, no calci-
um-containing phase could be observed in XRD analyses,
which was taken as an indication that a solid solution was
formed. For the impregnated sample, a mixed-oxide CaZrO3
with very low basicity was formed. A CaO phase was only ob-
served for the physical mixture of CaO and t-ZrO2. The precipi-
tated sample was found to be the most active catalyst.
Figure 1. DAA concentration from the self-condensation of 100 g acetone at
08C using 1 g of 5 wt% CaO/support-600, support=SiO2, TiO2, Al2O3, CNF,
and m-ZrO2.
Herein, we investigated in detail the role of the preparation
methods of calcium oxide–zirconia samples on their catalytic
performance and their base properties. Catalysts prepared by
impregnation on a monoclinic (m) ZrO2 support were com-
pared with precipitated samples, tetragonal (t) ZrO2. The calci-
nation temperature and calcium oxide loading was evaluated
for CaO on m-ZrO2. Amount and strength of the base sites
were evaluated by using temperature-programmed desorption
(TPD) of CO2. The catalytic activity correlated with the amount
of strong base sites (desorption of CO2 >6258C). For different
alkaline earth metal oxides on zirconia, the catalytic activity
paralleled the increase of the ionic radius within the group
Table 1. CO2 uptake, specific surface area, and activity of supports and
catalysts.
Catalysts
CO2 uptake
Specific surface
Initial rate
[mmolDAA gÀ1
À1
[mmolgÀ1
]
area [m2 gÀ1
]
h ]
cat
CNF-600
CaO/CNF-600
Al2O3-600
CaO/Al2O3-600
m-ZrO2-600
CaO/ZrO2-600
CaO/ZrO2-800
t-ZrO2-600
p-CaO–ZrO2-600
p-CaO–ZrO2-800
MgO/ZrO2-600
SrO/ZrO2-600
BaO/ZrO2-600
9
154
76
419
459
504
423
224
371
–
190
143
150
147
91
69
–
111
106
–
0
33
0
10
0
149[a]/152[b]
from Mg2+ to Ba2+
.
26
0
0.1
0.1
10
Results and Discussion
411
609
450
–
–
–
175
265
Screening of support materials for calcium oxide based
catalysts
[a] Catalyst (1 g) used in test per 100 g of acetone. [b] Catalyst (0.5 g)
used in test per 100 g of acetone.
In a screening study, preparations of CaO on different supports
(CNF, Al2O3, m-ZrO2, SiO2, and TiO2) prepared by incipient wet-
ness impregnation and activated by heat treatment at 6008C
in N2 were compared for their performances as a base catalyst
in the self-condensation reaction of acetone to diacetone alco-
hol (DAA). The catalysts were handled in inert atmosphere to
avoid exposure to CO2 in the air. The results are shown in
Figure 1 and Table 1.
change in concentration of DAA, it can be concluded that the
observed performance of CaO/ZrO2-600 was solely the result of
heterogeneous catalysis. The zirconia supports, m-ZrO2 and t-
ZrO2, showed negligible activities.
As CaO/ZrO2-600 displayed a very high activity in the screen-
ing study, only half the catalyst amount was used in the rest of
the study to avoid comparing catalysts at conversions too
close to those predicted by thermodynamics.
CaO on TiO2 or SiO2 displayed no activity whereas CaO/ZrO2-
600 showed superior activity compared to the other catalysts,
with the activity following the order CaO/ZrO2-600>CaO/CNF-
600>CaO/Al2O3-600. Only with the CaO/ZrO2-600 catalyst the
thermodynamic maximum concentration of DAA (1.4m, at
08C) was reached within 7 h. TEM images of the three active
samples, before reaction, can be found in the Supporting
Information (S1).
Zirconia as a support for CaO
The role of the preparation method (impregnation vs. precipi-
tation) and heat-treatment temperature (600 vs. 8008C) on the
catalytic activity was studied first (Figure 2, Table 1).
To rule out homogeneous catalytic activity, as a result of
CaO leaching, an experiment was performed in which the
CaO/ZrO2-600 was filtrated off after 1 h. The continuation of
the reaction in the filtrate (if present) was monitored for an ad-
ditional 6 h. As the filtrate exhibited no activity, that is, no
The impregnated CaO/ZrO2-600 was far more active than
precipitated p-CaO–ZrO2-600. The catalytic activity after activa-
tion at 6008C was compared to activities after heat treatment
at 8008C. The precipitated p-CaO–ZrO2 sample exhibited simi-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 9
&
2
&
ÞÞ
These are not the final page numbers!

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
the Figure). The impregnated CaO/ZrO2 and the m-ZrO2 after
heat treatment at 6008C exhibited diffractions from a monoclin-
ic zirconia phase (marked with m in the Figure) along with
traces of tetragonal zirconia, identified by the low intensity
peak at 2q=368. If heat treating the impregnated sample at
higher temperature, CaO/ZrO2-800, the peak at 2q=368 in-
creased whereas the peaks originating from monoclinic ZrO2
decreased. This shows that a partial phase transformation of
monoclinic zirconia into tetragonal zirconia took place. This in-
crease in the amount of tetragonal zirconia (Figure 3) corre-
sponds to a decrease in the catalytic activity as is observed in
Figure 2. This suggests that either the monoclinic zirconia
phase is advantageous for catalytic performance, or that base
sites from the calcium oxide are lost, for example, by incorpo-
ration in the support during the transformation from mono-
clinic to tetragonal ZrO2.
Figure 2. Impact of the preparation method on the catalytic activity in the
self-condensation reaction of 100 g acetone at 08C with 0.5 g of catalysts
CaO/ZrO2 and p-CaO–ZrO2, after heat treatment at 6008C and 8008C.
The CaO/ZrO2-600 and p-CaO–ZrO2-600 were examined with
SEM (Figure 4a and 4b, respectively) and with TEM (Figure 4c
and 4d, respectively) to gain insight in their morphology.
lar low activity after activation at 8008C to after activation at
6008C. The impregnated samples CaO/ZrO2 were significantly
less active after heat treatment at 8008C than after heat treat-
ment at 6008C.
The catalytic performances of CaO/ZrO2 and p-CaO–ZrO2
after heat treatment at 6008C was also evaluated in the trans-
esterification reaction between triacetine and methanol. The
product formation was evaluated after 30 min reaction time.
CaO/ZrO2-600 exhibited 100% conversion of the triacetin
whereas the p-CaO–ZrO2-600 converted less than 1%. This
shows that the trend in performance for the two catalysts is
the same for both the aldol coupling and the transesterifica-
tion reaction. Furthermore it demonstrates the applicability of
CaO/ZrO2-600 as a catalyst for biomass-related conversions.
The samples were characterized to understand the differen-
ces in catalytic performance. XRD patterns of the materials
after different preparation methods were evaluated (Figure 3)
to gain insight into the structure of the different catalysts.
No peaks originating from CaO were observed in any of the
XRD patterns of the materials displayed in Figure 3. The pre-
cipitated t-ZrO2 and coprecipitated p-CaO–ZrO2 exhibited only
diffractions from a tetragonal zirconia phase (marked with t in
Figure 4. SEM images of a) CaO/ZrO2-600 and b) p-CaO–ZrO2-600, and TEM
images of c) CaO/ZrO2-600 and d) p-CaO–ZrO2-600.
The impregnated CaO/ZrO2-600 sample consisted of 1–5 mm
lumps, built up from clusters of approximately 10 nm primary
particles (Figure 4a). The precipitated sample p-CaO–ZrO2-600
was more inhomogeneous and consisted of both large lumps
(ꢀ50 mm) with flat surfaces and smaller (1–5 mm) lumps on
top. In Figure 4b an example of one such large lump is dis-
played with smaller parts of the material deposited on the sur-
face. Electron dispersive X-ray (EDX) analysis revealed that the
composition of Zr and Ca is the same for both the large and
the small lumps in the precipitated sample. The morphologies
of m-ZrO2 and t-ZrO2 resembled those of CaO/ZrO2-600 and p-
CaO–ZrO2-600, respectively (see the Supporting Information,
S2). The presence of calcium was verified with EDX but individ-
ual CaO particles were not observed. In TEM the two materials,
CaO/ZrO2-600 and p-CaO–ZrO2-600 (Figure 4c–d) looked com-
parable, and again the materials were found to be indistin-
guishable in morphology to zirconia reference samples. In TEM
the calcium oxide could thus also not be observed directly but
Figure 3. XRD patterns of p-CaO–ZrO2-600, t-ZrO2-600, CaO/ZrO2-800, CaO/
ZrO2-600, and m-ZrO2-600.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 9
&
3
&
ÞÞ
These are not the final page numbers!

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
459 mmolgÀ1) has more base sites than the p-CaO–ZrO2-600
(371 mmolgÀ1; t-ZrO2 224 mmolgÀ1), whereas the specific sur-
face area follows the opposite trend, larger for p-CaO–ZrO2-
600 (106 m2 gÀ1) than for CaO/ZrO2-600 (69 m2 gÀ1). The differ-
ence in number of base sites is consistent with the XPS results
revealing that more calcium is present on the surface for the
impregnated sample than on the precipitated sample. The
number of base sites alone cannot account for the difference
in catalytic activity between the impregnated and coprecipitat-
ed samples, indicating that base strength also plays an impor-
tant role. The base strengths of CaO/ZrO2-600 and p-CaO–
ZrO2-600 were thus evaluated by using CO2 TPD as displayed
in Figure 6 and compared to those of m-ZrO2-600 and
t-ZrO2-600.
Figure 5. SEM and element mapping of Ca, Zr, and O for CaO/ZrO2-600.
EDX analysis confirmed that for both preparation methods the
element ratios corresponded to the expected loadings of
calcium.
In Figure 5 element mapping by SEM of CaO/ZrO2-600 is
shown. From this it can be seen that the Ca is well distributed
over the zirconia within the boundaries of the resolution ob-
tainable with SEM. The same held true for the coprecipitated
sample (see the Supporting Information, S3). This is in accord-
ance with EDX line profile scans in TEM analysis, which
showed that the calcium and zirconium were always in close
proximity. The fact that the calcium could not be visualized in-
dicated that the CaO is present either in form of a thin film
over the surface, as very small particles (<1 nm), incorporated
in the structure of the zirconia, or forming a solid solution as
suggested in the literature.[31,32]
Figure 6. CO2 TPD results for CaO/ZrO2-600, p-CaO–ZrO2-600, m-ZrO2-600,
and t-ZrO2-600.
From the CO2 TPD data it can be seen that both zirconia
samples (m-ZrO2 and t-ZrO2) display one broad desorption
peak with a maximum at approximately 2008C. This peak can
be ascribed to weak basic sites intrinsically present on a zirco-
nia surface. For the CaO/ZrO2-600 this low-temperature peak is
shifted to higher temperature with a desorption maximum at
3758C. This suggests that part of the calcium has been incor-
porated in the structure in form of a solid solution with stron-
ger sites than those intrinsic to the zirconia support. Addition-
ally, a peak with a maximum at approximately 7758C is ob-
served, indicating that the sample contains some very strong
base sites most likely associated to the presence of highly dis-
persed CaO on the ZrO2 surface. For the precipitated p-CaO–
ZrO2-600 sample, the low-temperature peak is also shifted to-
wards higher temperature, 2508C, but to a lesser extent that
for the impregnated sample, which could be explained by part
of the calcium dissolving in the zirconia lattice structure. A
very broad but low-intense peak is observed at 600–9008C for
p-CaO–ZrO2-600. The fact that this high temperature peak has
so low intensity indicates that only few strong base sites are
present, compared to in CaO/ZrO2-600. The p-CaO–ZrO2-600
and CaO/ZrO2-600 have thus very different properties with re-
spect to base strength, dependent on the preparation method
and the calcium-support interaction. We propose that the su-
To gain more insight into the location of calcium in CaO/
ZrO2-600 and p-CaO–ZrO2-600, XPS was performed. A differ-
ence was found between CaO/ZrO2-600 samples having a Ca/
Zr ratio of 0.29 and p-CaO–ZrO2-600 samples having a Ca/Zr
ratio of only 0.17 (see the Supporting Information, S4). As the
two samples have the same nominal Ca/Zr composition this
shows that part of the calcium has been incorporated in the
structure in the precipitated sample. The higher calcium con-
tent on the surface of the impregnated sample might play
a role in the higher activity displayed by this sample. It sug-
gests that CaO is well dispersed (e.g., as a film) on the zirconia
surface meaning that there are more sites available but poten-
tially also sites of a different nature, for example, different base
strength compared to the precipitated sample.
The specific surface area and number of base sites were
evaluated by using BET and CO2 chemisorption techniques, re-
spectively; these results are summarized in Table 1. The num-
bers of base sites are difficult to compare, which is owing to
the amount of base sites already present in zirconia itself.
However, the CaO/ZrO2-600 sample (504 mmolgÀ1; m-ZrO2-600
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 9
&
4
&
ÞÞ
These are not the final page numbers!