Aldol condensation of furfural and acetone over MgAl layered double hydroxides and mixed oxides

Article abstract of DOI:10.1016/j.cattod.2013.09.022

Liquid phase aldol condensation of furfural and acetone catalyzed by solid base catalysts (MgAl hydrotalcites and MgAl mixed oxides) has been investigated as a method to valorize short chain ketones obtainable from biomass pyrolysis for the production of higher-molecular-weight products usable as fuel components. The Mg/Al molar ratios of the investigated catalysts was varied in the range from 2 to 4 and their catalytic activity was tested at different reaction temperatures (20-100 C) using several activation methods. MgAl hydrotalcites were used either after calcination as mixed oxides or after subsequent contact with liquid water or steam as rehydrated materials. Both furfural-acetone condensation products, i.e. C₈ and C₁₃, were the desired products of aldol condensation reaction. Higher reaction temperature facilitated the dehydration step and enhanced the catalyst selectivity to both dehydrated products (C₈ and C₁₃). The best results were achieved with calcined catalyst sample having Mg/Al molar ratio equal to 3 at 100 C (>95% furfural conversion and >90% selectivity to the desired products). The selectivity to the main by-product, diacetone alcohol, did not exceed 5% in any experiment. The ex situ rehydration of the calcined samples resulted in catalysts with a significantly lower activity, except the catalyst with Mg/Al molar ratio equal to 2. On the other hand, in situ rehydration caused catalyst activity improvement only in the case of the sample with Mg/Al molar ratio 3.

Full text of DOI:10.1016/j.cattod.2013.09.022

Catalysis Today 223 (2014) 138–147
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Aldol condensation of furfural and acetone over Mg Al layered
double hydroxides and mixed oxides
Lukásˇ Horaa,∗, Vendula Kelbichováa, Oleg Kikhtyanina,
Oleg Bortnovskiyb, David Kubicˇkaa
a Research Institute of Inorganic Chemistry, UniCRE-RENTECH, Areál Chempark Záluzˇí, 436 70 Litvínov, Czech Republic
b Eurosupport Manufacturing Czechia, Areál Chempark Záluzˇí, 436 70 Litvínov, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 February 2013
Received in revised form 26 August 2013
Accepted 5 September 2013
Available online 14 October 2013
Liquid phase aldol condensation of furfural and acetone catalyzed by solid base catalysts (Mg Al hydro-
talcites and Mg Al mixed oxides) has been investigated as a method to valorize short chain ketones
obtainable from biomass pyrolysis for the production of higher-molecular-weight products usable as
fuel components. The Mg/Al molar ratios of the investigated catalysts was varied in the range from 2
to 4 and their catalytic activity was tested at different reaction temperatures (20–100 ◦C) using several
activation methods. Mg Al hydrotalcites were used either after calcination as mixed oxides or after
subsequent contact with liquid water or steam as rehydrated materials.
Both furfural-acetone condensation products, i.e. C8 and C13, were the desired products of aldol conden-
sation reaction. Higher reaction temperature facilitated the dehydration step and enhanced the catalyst
selectivity to both dehydrated products (C8 and C13). The best results were achieved with calcined cata-
lyst sample having Mg/Al molar ratio equal to 3 at 100 ◦C (>95% furfural conversion and >90% selectivity
to the desired products). The selectivity to the main by-product, diacetone alcohol, did not exceed 5% in
any experiment. The ex situ rehydration of the calcined samples resulted in catalysts with a significantly
lower activity, except the catalyst with Mg/Al molar ratio equal to 2. On the other hand, in situ rehy-
dration caused catalyst activity improvement only in the case of the sample with Mg/Al molar ratio 3.
Keywords:
Aldol condensation
Hydrotalcites
Mixed oxides
Furfural
Renewables
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
of sugar cane bagasse, a residue from sugar cane processing, fol-
lowed by extraction [9]. To exploit the full potential of furfural
The extensive use of fossil fuels has been a cornerstone of
the rapid development during twentieth century, but it has also
resulted in many adverse environmental impacts. Consequently,
enormous research focus has recently been directed toward the
sustainable utilization of renewable resources in an attempt to
replace at least partially the finite fossil fuels resources [1,2]. The
primary objective of these efforts is to maximize the sustain-
able utilization of biomass-derived feedstocks while minimizing
environmental impacts (CO2 emissions, loss of diversity, etc.) and
avoiding competition with food and fodder production. It is partic-
ularly desirable to produce liquids fungible with conventional fuels
and chemicals production from non-food crops [3–5].
as a platform chemical its transformation by aldol conden-
sation with acetone followed by hydrogenation/deoxygenation
has been proposed to afford hydrocarbons, namely C8 and C13
alkanes [10].
Typically, aldol condensation is catalyzed by homogenous base
catalysts, such as sodium and calcium hydroxide. However, these
processes generate significant waste water streams that must be
treated (neutralized and cleaned) [10]. Apart from the negative
environmental impact, the production cost is affected adversely.
Consequently, many researchers focus on developing a new active
and stable solid base catalyst (such as MgO [11], MgO ZrO [10,12],
ZnO, TiO2 [13]) to replace the generally used homogenous catalysts
[14,15]. Among the solid base catalysts, layered double hydroxides
(LDH) also known as anionic clays or hydrotalcite like materials
(HTC) have been also reported to be promising aldolization cata-
lysts [16–19]. The main advantages of heterogeneous catalysts over
homogeneous ones include easier catalyst recovery and recycling
resulting in mitigation of the adverse environmental impacts, such
as waste water treatment, and elimination of corrosion problems
to name the most significant ones.
A
promising route to using the chemical potential of
cellulose-rich feedstocks is their hydrolysis yielding furfural or
hydroxymethylfurfural that are often denoted as platform chemi-
cals [6–8]. For example, furfural can be obtained by acid hydrolysis
∗
Corresponding author.
E-mail address: lukas.hora@vuanch.cz (L. Hora).
0920-5861/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2013.09.022

L. Hora et al. / Catalysis Today 223 (2014) 138–147
139
Table 1
The name “hydrotalcites” is reserved for solids having struc-
List of the catalysts used for catalytic tests.
ture closely related to that of the mineral hydrotalcite, that is,
rhombohedral Mg6Al2(OH)16CO3·4H2O. Hydrotalcites consist of a
brucite-like [Mg(OH)2] network wherein isomorphous substitution
of Mg2+ ion by a trivalent cation M3+ occurs and the excess posi-
tive charge is compensated by gallery anions which are located in
the interlayer along with water molecules [20–22]. The Mg Al LDH
can be used as catalysts either in the form of mixed oxides Mg(Al)O
obtained by the calcination of an Mg Al hydrotalcite precursor or
in the rehydrated form of these calcined materials [23].
The aim of this work is to describe the effect of pretreatment
of the hydrotalcite-based catalysts on their activity and selectivity
in aldol condensation of furfural with acetone. Besides calcination
of the hydrotalcite precursor, different rehydration procedures (ex
situ vs. in situ and liquid vs. gas phase) have been performed and
their impact on the yield of the desired aldol condensation products
has been determined. Moreover, the effect of the Mg/Al ratio has
been studied as well.
Catalyst name
Activation procedure
Cal. HTC-feed
Cal. HTC 2:1
Cal. HTC 3:1
Calcination
Calcination
Calcination
Calcination
Cal. HTC 4:1
ExWR HTC-feed
ExWR HTC 2:1
ExWR HTC 3:1
ExWR HTC 4:1
ExVR HTC-feed
ExVR HTC 2:1
ExVR HTC 3:1
ExVR HTC 4:1
InWR HTC-feed
InWR HTC 2:1
InWR HTC 3:1
InWR HTC 4:1
Full ex situ rehydration in liquid water
Full ex situ rehydration in liquid water
Full ex situ rehydration in liquid water
Full ex situ rehydration in liquid water
Full ex situ rehydration in water vapor
Full ex situ rehydration in water vapor
Full ex situ rehydration in water vapor
Full ex situ rehydration in water vapor
Partial in situ rehydration in liquid water
Partial in situ rehydration in liquid water
Partial in situ rehydration in liquid water
Partial in situ rehydration in liquid water
by X-ray powder diffraction using a Philips MPD 1880, working
with the Cu K␣ line (ꢀ = 0.154 nm) in the 2ꢁ range of 5◦–70◦ at a
scanning rate of 2ꢁ of 2.4◦/min. The elemental composition was
determined by X-ray fluorescence using Philips PW 1401 equipped
with an Rh RTG lamp. The UniQuant program was used to evaluate
the results. The textural properties of the catalysts (specific surface
area and pore volume) were measured by nitrogen physisorption at
−196 ◦C using a Micromeritics ASAP 2020 surface area and poros-
ity analyzer. The plausible leaching of the main metals (Mg, Al)
from the catalysts into the reaction mixture during the reaction
was evaluated by ICP–OES equipment.
2. Experimental and materials
2.1. Materials
Mg Al layered double hydrotalcites (LDH) with Mg/Al molar
ratio ranging from 2:1 to 4:1 were provided by Eurosupport Manu-
facturing Czechia (ESMC). One sample, denoted HTC-feed, was used
as a reference material. Their properties and activation procedures
are described below. Furfural (Acros Organics, 99%) and acetone
(Lach:Ner, p.a.) were used in the experiments.
Thermogravimetric analysis of the dried LDH catalysts was per-
formed using a TA Instruments TGA Discovery series equipment
and operating at heating ramp of 10 ◦C/min from room temperature
to 900 ◦C in flowing nitrogen (20 mL/min, Linde 3.0). Approximately
15 mg of sample was heated in an open alumina crucible.
2.2. Catalysts activation
Three different activation procedures were applied to transform
the as-supplied LDH materials into catalysts–calcination, ex situ
rehydration of the calcined material and in situ rehydration of the
calcined material. The calcination of LDH was performed at 450 ◦C
in air and mixed Mg Al oxide catalysts were obtained. The tem-
perature during calcination was raised at the rate of 10 ◦C/min to
reach 450 ◦C and maintained at the final temperature for 16 h. The
calcined mixed oxides were used as catalysts in aldol condensation
reaction directly or after rehydration.
The ex situ rehydration was carried out either in liquid or gas
phase. The liquid phase ex situ rehydration was performed in a
stirred flask using demineralized water at 20 ◦C for 2 h under inert
atmosphere (N2). After the activation, the catalyst was separated
by filtration and dried under vacuum at 60 ◦C for 2 h. The gas phase
ex situ rehydration was carried out in a glass tube at 50 ◦C under
flow of nitrogen gas saturated with water vapor for 16 h. The flow
of wet nitrogen of 6 L/h was maintained for the specified period
(16 h for 5 g of catalyst). Both types of ex situ treatments lead to full
rehydration of the calcined materials (mixed oxides). The amount
of water needed for the full rehydration of the mixed oxides was
approximately 1 g of water per 1 g of mixed oxides.
2.4. Reaction studies
The experiments were carried out in a 100 mL stirred batch
reactor (a glass flask reactor for experiments at up to 50 ◦C, an
autoclave for experiments at temperatures above 50 ◦C) in liq-
uid phase. Before the start of an experiment, 2 g of catalyst was
mixed together with a stirred mixture of 39.5 g acetone and
6.5 g furfural (acetone/furfural molar ratio 10:1, pre-heated to
the desired reaction temperature in range from 20 to 100 ◦C)
and kept at this temperature for 2 or 24 h under intensive stir-
ring.
The experiments were carried out at the strictly controlled con-
ditions. Every time the same heating rate of 5 ◦C per minute was
used. The experiments carried out at up to 50 ◦C were performed
with the catalyst added into the reaction mixture after reaching of
the reaction temperature. The experiments carried out at 100 ◦C
were performed with the presence of catalyst in reaction mixture
from the beginning of the heating. No overshooting of the target
temperature was observed.
The in situ rehydration was carried out only in liquid phase. It
was achieved by addition of a small amount of demineralized water
(0.2 mL) into the reaction mixture before immersing the calcined
catalyst into the reactor. In this case, approximately 10% wt. of the
mixed oxides was rehydrated.
The used acetone/furfural molar ratio and the absolute amounts
of the reactants could result, in the case of complete conver-
sion of furfural, in the formation of approximately 0.8 g of water.
This amount of water equals approximately to 40% of the amount
needed for a full rehydration of the mixed oxides (calcined hydro-
talcites).
The list of all tested catalysts including their names and activa-
tion procedure is shown in Table 1.
The samples of reaction mixtures were withdrawn from the
reactor during the experiment at certain reaction times, filtered,
and analyzed by an Agilent 7890A gas chromatograph equipped
with a flame ionization detector (FID), using a HP 5 capillary column
(30 m/0.32 mm ID/0.25 ␮m).
2.3. Catalysts characterization
The catalysts were characterized by several techniques to assess
their structure, composition, surface area and pore volume. The
crystallographic structures of dried LDH catalysts were determined

140
L. Hora et al. / Catalysis Today 223 (2014) 138–147
Table 2
Chemical composition of the catalysts.
Parameter/sample
Unit
HTC-feed
HTC 2:1
HTC 3:1
HTC 4:1
MgO
Al2O3
SiO2
CaO
TiO2
wt.%
wt.%
wt.%
wt.%
wt.%
wt.%
mol/mol
50.5
27.2
0.19
<0.01
<0.01
0.10
2.35
59.07
37.42
0.23
1.09
0.31
1.18
2.0
70.86
27.78
0.37
0.43
0.05
0.14
3.25
75.25
23.45
0.25
0.52
0.29
0.14
4.1
Fe2O3
Mg: Al
Table 3
Main textural characteristics of the catalysts.
Catalyst form
Dried
Calcined
Spent
Parameter
Sample/unit
Specific area (BET)
m2/g
Pore volume
Specific area (BET)
m2/g
Pore volume
Specific area (BET)
m2/g−1
cm3/g
cm3/g
HTC-feed
HTC 2:1
HTC 3:1
HTC 4:1
149.0
137.2
128.8
121.2
0.42
0.31
0.32
0.31
185.3
170.9
164.3
159.1
0.40
0.30
0.33
0.31
–
63.9
28.2
71.3
Table 4
TGA analysis of dried LDH catalysts.
3. Results and discussion
1st peak (◦C) 2nd peak (◦C) Weight loss
Residue
3.1. Characterization of the catalyst
Sample
30–450 ◦
C
450–900 ◦
C
The catalytic materials were characterized in the as-obtained
form to determine their composition (Table 2) and textural prop-
erties (Table 3). The main impurities in most of the samples were Fe,
Ca, Ti, Si, particularly in the HTC 2:1 sample. On the other hand, the
reference sample denoted HTC-feed was virtually free of all impu-
rities, besides traces of Si and Fe. The pore volume of the HTC 2:1 to
4:1 samples was almost identical, around 0.31 cm3/g, regardless the
variation in Mg/Al ratio while the sample HTC-feed exhibited about
30% higher pore volume, 0.42 cm3/g (Table 3). The specific surface
areas of all dried samples were within the range 120–150 m2/g,
while the calcined samples exhibited specific surface areas in the
range 159–186 m2/g. Surprisingly, the sample HTC 2:1 exhibited
about 30% higher BET surface area than HTC 3:1 and HTC 4:1 despite
the fact that they all showed the same pore volume. This indicates
that HTC 2:1 has a smaller particle size than the other two catalysts.
The structural data were obtained both for as-obtained (dried)
and calcined catalysts – Figs. 1 and 2. A clear difference among
the studied samples can be observed in case of the dried sam-
ples. In particular, the sample HTC-feed differs from the other
samples as it has fewer distinguishable diffraction lines which
indicate a lower degree of crystallinity. On the other hand, the
HTC 2:1 shows a well-developed and characteristic hydrotalcite
diffractogram (Fig. 1). With the increasing Mg/Al ratio, the clarity
of diffractograms decreases which can be attributed to a lower crys-
tallinity of these samples. Moreover, the XRD data indicate also the
presence of other crystalline phases, namely MgO and Mg(OH)2.
The shape of the diffraction line belonging to MgO (at 43◦) is virtu-
ally identical for HTC 2:1–4:1 indicating that the MgO crystallites
are approximately of the same size. On the other hand, the line is
significantly broadened in HTC-feed, which suggests that the MgO
crystallites are smaller in this sample than in the other hydrotal-
cites.
HTC feed 205
388
330
396
388
21.80%
17.15%
29.47%
23.43%
0.41%
2.35%
3.48%
3.14%
77.79%
80.50%
67.05%
73.43%
HTC 2:1
HTC 3:1
HTC 4:1
105
105
113
with two main events that are reflected as peaks in the DTA. In gen-
eral, for all the hydrotalcite samples, the first step involves loss of
interlayer water that occurs in between ambient temperature and
about 125 ◦C. Dehydration takes place in two minor steps followed
by a step at around 200 ◦C. The third step corresponds to the loss of
OH− groups and decomposition of CO3 in the LDH structure and
2−
can be observed at around 380–400 ◦C [26].
In our case the shapes of the TGA curves and the positions of
the DTA peaks demonstrate differences among the used materials.
The similar shape of the curves of the samples HTC 3:1 a HTC 4:1
suggest their similar structure (XRD patterns of these LDH materi-
als are also similar, Fig. 1), differences are only in the total weight
loss. The indistinctive course of the thermal analysis of the sample
HTC 2:1 (as clearly visualized by the DTA peaks) was not expected,
because its XRD patterns suggested the best hydrotalcite structure.
This sample had the lowest weight loss from all the tested samples.
Although the sample HTC feed did also not show a very indistinctive
TGA curve, the DTA curves exhibit the presence of the two peaks
characteristic for LDH.The catalysts were also analyzed by the XRD
and TGA after their use in aldol condensation. The XRD results of
calcined catalysts used at low reaction temperatures did not show
any significant differences in the catalyst structures. The only crys-
talline phase observed was the MgO crystalline phase. The amount
of water that can be produced in the reaction may be estimated as
0.4 g (furfural conversion about 50%), which is 20% of the catalyst
weight and full re-hydration necessitates the water amount to be
100% of the catalyst weight. In line with this, the TGA results con-
firmed partial re-hydration of catalysts by the reaction water. The
TGA results also indicated that there were not any carbonaceous
deposits on the spent catalysts. Nonetheless, the specific surface of
the spent catalysts significantly decreased (below 100 m2/g). The
leaching of the main elements, Mg and Al, was checked by ICP anal-
ysis of the reaction products mixtures. In all cases the content of
Mg and Al in the obtained liquid products was below the detection
limit of 20 mg/kg which corresponds to 0.15% of the mass of the
The XRD patterns of the calcined samples (Fig. 2) indicate the
presence of MgO crystalline phase only. While the shapes of the
diffraction lines of HTC 2:1–4:1 are practically the same, reference
sample HTC-feed shows dramatically wider and lower peaks. This
indicates, in accordance with XRD patterns of dried sample, that
HTC-feed is less crystalline than the other samples.
The TGA–DTA peaks obtained from the thermal analysis studies
of the dried LDH catalysts are presented in Fig. 3 and Table 4. The
course of the TGA curves exhibited a gradual weight loss process

L. Hora et al. / Catalysis Today 223 (2014) 138–147
141
Fig. 1. X-ray diffraction patterns of the dried hydrotalcites.
used catalyst being present or dissolved in the total liquid reaction
product.
The data also confirm the consecutive nature of several reaction
steps (Fig. 4, Scheme 1). In particular, the intermediate product,
C8 alcohol (see Scheme 1), is preferentially formed in the initial
period of the reaction as a primary condensation product. It is sub-
sequently dehydrated into the first condensation C8 product (FAc).
A higher reaction temperature facilitates this dehydration process.
FAc successively reacts with another furfural molecule and the
second condensation C13 product (FA2c) is formed, which is also
a dehydrated product. It can be clearly seen that higher reaction
temperature favors the formation of FAc and F2Ac, respectively,
resulting thus in an increased selectivity to these desired prod-
ucts. It can be concluded that the selectivity to alcohols decreases
at the expense of the selectivity to formation of ketones with
the increasing reaction temperature. The selectivity to diacetone
3.2. Aldol condensation of furfural and acetone
3.2.1. Basic reaction behavior
The influence of the reaction temperature and time on furfural
conversion and catalyst selectivity in the range 25–56 ◦C (reflux
temperature) is shown in Fig. 4. Both reaction temperature and
time have significant effect on the course of furfural conversion.
As expected higher furfural conversions were obtained at higher
temperatures. It can be inferred that the conversion of furfural is
not limited by thermodynamic equilibrium.
Fig. 2. X-ray diffraction patterns of the calcined hydrotalcites (mixed oxides).

142
L. Hora et al. / Catalysis Today 223 (2014) 138–147
Fig. 3. TGA–DTA curves of dried of dried LDH catalysts.
alcohol (DAA), the main by-product, remains practically constant
during the course of the reaction, i.e. it is not favored by the
increased reaction temperature.
can be revealed at 20 ◦C. The highest conversion, about 30%, was
achieved with catalyst HTC 3:1, while the conversion over HTC 2:1
and HTC 4:1 are significantly lower, ca. 20% and ca. 15%, respec-
tively. It can be thus concluded that the Mg/Al ratio has an impact
on the catalyst activity and that there is an optimum at a molar
ratio Mg/Al around 3 (HTC 3:1).
3.2.2. Influence of the Mg/Al molar ratio of the calcined catalysts
and reaction temperature
There is a significant difference in product distribution depend-
ing on the reaction temperature regardless the mixed-oxide
catalyst used (Fig. 5). While C8 alcohol is formed almost exclusively
(selectivity >90% at conversion <20%) at reaction temperature 20 ◦C,
only minor concentrations of C8 alcohol were observed over cata-
lysts that afforded conversion >90% at 100 ◦C. On the other hand,
over HTC 4:1 conversion of only about 60% was reached after 2 h and
consequently significant selectivity to C8 alcohol was still observed
at the expense of selectivity to the desired FAc and F2Ac products
(Fig. 2). It can be concluded that at 20 ◦C the dehydration of C8
alcohol to the desired C8 product FAc is very slow as is the sub-
sequent second condensation step to C13 product F2Ac (Fig. 5). A
The performance of the calcined LDH catalysts without any
rehydration, i.e. mixed oxides catalysts, was investigated using cat-
alysts with Mg/Al molar ratio between 2 and 4 at two reaction
temperatures, 20 and 100 ◦C. The results are summarized in Fig. 5
where a comparison of furfural conversion and product distribu-
tions is given.
The results in the whole range of the investigated Mg/Al ratios
are in accordance with the above described reaction behavior. Fur-
fural conversions more than 95% were achieved at 100 ◦C with all
catalysts except the catalyst denoted HTC 4:1 (Fig. 5). In this case
the conversion was only around 62%. While the performance of
HTC-feed, HTC 2:1 and HTC 3:1 is very similar at 100 ◦C, differences

L. Hora et al. / Catalysis Today 223 (2014) 138–147
143
Fig. 4. Profile of furfural conversion and catalyst selectivity (products distribution) during reaction temperature switching between 25 and 56 ◦C in a 28-h experiment.
OH
O
O
O
O
O
H3C
CH3
O
base
CH3
CH3
+
O
-H2O
Acetone
(Ac)
Furfural
(F)
4-(2-furyl)-3-buten-2-one
(FAc)
4-(2-furyl)-4-hydroxy-butan-2-one
(C8 alcohol)
O
O
O
O
O
O
base
-H2O
O
CH3
+
1,4-pentadien-3-one, 1,5-di-2-furanyl
(F2Ac)
Scheme 1. Reaction scheme of the aldol condensation of furfural and acetone.
comparison of the product distributions with the phase com-
position of the catalysts indicates that at low temperature the
formation of FAc and F2Ac can be related to the presence of
MgO and Mg(OH)2 phases (although the dependence could not be
expressed explicitly). A more detailed structural investigation is,
however, needed to reach an unequivocal conclusion. The best
results were achieved with HTC 3:1 catalyst that yielded almost
exclusively FAc and F2Ac products at nearly complete conversion
Fig. 5. Comparison of furfural conversions obtained over calcined catalysts with different Mg/Al molar ratio using reaction temperature 20 and 100 ◦C.

144
L. Hora et al. / Catalysis Today 223 (2014) 138–147
Fig. 6. Comparison of furfural conversion and product selectivity achieved over calcined and rehydrated catalysts with different Mg/Al molar ratio using reaction temperatures
20 (A) and 100 ◦C (B).
of furfural. The selectivity to diacetone alcohol (DAA) did not exceed
5% in any of the experiments. Moreover, the formation DAA was not
detected at 20 ◦C (Fig. 5).
over HTC 2:1 stayed practically constant after all three activation
procedures (Fig. 6). It is also interesting to note that the conver-
sion over HTC 3:1 dropped less than over HTC feed and again
the steam rehydrated catalysts afforded higher furfural conver-
sions than the liquid phase rehydrated ones. A comparison with
the structural data suggests that the catalytic materials exhibiting
a lower relative crystallinity (HTC 3:1 and particularly HTC feed)
are significantly affected by the rehydration procedure while HTC
2:1 catalyst with rather well-developed hydrotalcite structure is
more resistant to any of the applied ex situ rehydration procedures.
However, the plausible influence of the crystallinity of the LDH
phase in the prepared catalyst samples on their catalytic behav-
ior in aldol condensation requires additional investigation. Another
explanation can be provided based on the analysis of the phase
composition of the dried samples according to their XRD patterns
(Fig. 1). All samples are a mixture of two main phases, hydrotal-
cite and MgO/Mg(OH)2. Both hydrotalcite and MgO possess high
initial activity for aldol condensation. Nevertheless, water which is
formed in the course of the reaction may affect the properties of
these two main phases differently. While the presence of water in
the reaction mixture can even increase the activity of HTC materials
in the condensation of aldehydes and ketones [26], MgO gradually
loses its activity due to its continuous hydration and transformation
to inactive Mg(OH)2. As a consequence, the activity of the different
materials in the study is determined in a great extent by the ratio
between HTC and MgO phases, which is the highest in case of HTC
3.2.3. Influence of ex situ rehydration on the catalyst activity and
selectivity
The activation of the calcined hydrotalcites (mixed oxides) with
water or steam is often reported in the literature [24,25]. Hence, the
three best calcined catalysts (HTC-feed, HTC 2:1 and HTC 3:1) were
investigated after rehydration under the same reaction conditions
as the calcined catalysts (20 and 100 ◦C, 2 h). Fig. 6 compares con-
version of furfural and the product distribution over the calcined
catalysts with the conversion of furfural and the product distribu-
tion over ex situ rehydrated catalysts, both liquid phase rehydrated
(Ex-RW) and steam rehydrated (Ex-VR).
It can be clearly observed that at 20 ◦C the rehydration led
to significant decrease in conversion for both ex situ rehydration
methods. Moreover, the decrease in conversion for all investigated
catalysts was more pronounced in the case of ex situ liquid phase
rehydration than in the case of steam rehydration (Fig. 6). Due to
the decrease in the conversion, the selectivity to the desired prod-
ucts (FAc and F2Ac) dropped as well (Fig. 6). In contrast, the ex
situ rehydration affected the catalyst performance at 100 ◦C differ-
ently. While over HTC-feed and HTC 3:1 the conversion of furfural
dropped dramatically due to the rehydration, HTC 2:1 maintained
its activity even after rehydration (Fig. 6). The conversion of furfural

L. Hora et al. / Catalysis Today 223 (2014) 138–147
145
Fig. 7. Comparison of furfural conversion and catalyst selectivity achieved over calcined and in situ rehydrated catalysts with different Mg/Al molar ratio using reaction
temperatures 20 and 50 ◦C for (A) 2 h and (B) 24 h.
2:1. In terms of the product distribution (Fig. 6), the rehydrated
catalysts (except the HTC 2:1) show a higher selectivity to the
intermediate C8 alcohol not only at 20 ◦C, but also at 100 ◦C in com-
parison with the calcined catalysts. Due to the consecutive nature of
the reactions, this could be explained by the lower conversion over
the rehydrated catalysts. This conclusion is supported comparing
the results obtained over the calcined HTC 4:1 (furfural conversion
ca. 60%, selectivity to C8 alcohol ca. 32%, Fig. 5) and over rehydrated
HTC 3:1 (furfural conversion ca. 50%, selectivity to C8 alcohol ca.
28%, Fig. 6). It is known that calcined HTC materials possess strong
Lewis basicity, and their rehydration results in a Lewis-to-Brønsted
sites transformation [27]. The obtained results show that rehydra-
tion, regardless the ex situ rehydration procedure, does not affect
the reaction pathway and results in a comparable product composi-
tion at equal furfural conversion. In this case, the activity of both the
calcined and the rehydrated catalyst samples as well as the com-
position of the obtained reaction products are determined by the
amount of the accessible active sites (both Brønsted and Lewis) with
the strength sufficient to activate the particular reactant molecule.
during catalyst handling, or by water dissolved in the raw materials.
To verify this assumption a series of experiments were carried out
at temperature 20 and 50 ◦C. Due to the relatively low reaction tem-
peratures the reaction time was extended up to 24 h. A comparison
of furfural conversions is given in Fig. 7. A rather peculiar behav-
ior can be observed, the conversion over HTC 3:1 has increased in
comparison with the calcined sample (Fig. 5 and Fig. 7), while it
did drop in the case of liquid phase rehydration Fig. 6. The effect on
the other catalysts was very similar to the one observed for the ex
situ water rehydration. Another interesting observation is related
to the conversion of furfural at 20 and 50 ◦C where a lower con-
version was observed at the higher temperature in the case of HTC
3:1 and HTC 4:1 catalysts after 2 h (Fig. 7). This can be attributed to
the influence of water that is inhibiting the catalyst activity, partic-
ularly at low reaction temperatures (Fig. 6) at which the intrinsic
reaction rate is significantly lower.
To explain this behavior, it is helpful to look at the conversion
of furfural as a function of reaction time (Fig. 8). When the calcined
catalyst is not pre-hydrated, it behaves at 20 ◦C as a calcined cat-
alyst and the effect of water addition can be observed only by the
gradual decrease in the reaction rate. On the other hand, at 50 ◦C
the catalyst activity is apparently affected at the very beginning of
the experiment (due to the higher rate of inhibition by water) and
3.2.4. Influence of catalysts activation by in situ rehydration
The effect of ex situ rehydration indicates that the cata-
lyst activity could be negatively influenced by air moisture, e.g.

146
L. Hora et al. / Catalysis Today 223 (2014) 138–147
Fig. 8. Time evolution of furfural conversion and catalysts selectivity at reaction temperature (A) 20 ◦C and (B) 50 ◦C over catalyst HTC 3:1 rehydrated in situ.
consequently an induction period is observed (Fig. 8). When the
reaction continues for a longer time, the final furfural conversion
is higher at 50 ◦C than at 20 ◦C due to the higher reaction rate at
the higher reaction temperature. It can be also seen that the reac-
tion rate drops significantly with time (Fig. 8). After about 6 h the
increase in conversion of furfural is a small one despite full conver-
sion was not yet reached. The product distributions (Figs. 7 and 8) at
20 ◦C and after 2 and 24 h are very similar, dominant product being
C8 alcohol. In agreement with the previous results, the extent of
dehydration is very limited at 20 ◦C.
furfural fully into condensation products. The selectivity to the
main by-product (DAA) again did not exceed 5%.
The differences of catalysts behavior after in situ rehydration
can be also explained with the help of TGA analysis results. As it
was presented above in the Fig. 3 and Table 4, HTC 3:1 and 4:1
show a more typical thermal decomposition profile of LDH than
HTC-feed and HTC 2:1. This means, that after in situ rehydration
(by a small amount of water) large amount additional active basic
sites can be formed. As a result, the in situ rehydration might lead
to improvement of the catalysts activity.
Analogously with other experiments, higher reaction tempera-
ture favors the dehydration step and higher yields of the desired
C8 and C13 products (FAc and F2Ac) were formed during reaction
at 50 ◦C, particularly over HTC 3:1 catalyst. This tendency can be
observed clearly when the product distributions are plotted as a
function of reaction time (Fig. 8). While the apparent rate of for-
mation of the C8 alcohol is significantly higher than the apparent
rate of formation of FAc and F2Ac, the apparent rates of forma-
tion of these three products are virtually equal at 50 ◦C (Fig. 8). It
can be also seen that at 50 ◦C the C8 alcohol intermediate is vir-
tually completely consumed in the successive reactions, while at
20 ◦C an equilibrium-like state is established, i.e. the selectivities
to the products do not change (Fig. 8). It can be concluded that
higher reaction temperatures, at least 50 ◦C, are needed to convert
It can be concluded that in addition to the inherent structure of
the HTC catalysts their pretreatment determines their final activity
and selectivity. The observed peculiarities indicate that in fact the
activity is very closely related to the rehydration process. Hence the
process should be carefully monitored and investigated in more
detail than it was done so far. We have already started working
along this research line.
4. Conclusions
Mg Al layered double hydrotalcites (LDH) with different Mg/Al
molar ratio have been investigated as potential catalysts in aldol
condensation of furfural with acetone at different reaction tem-
peratures. Catalytic activity of the parent LDH materials has been

L. Hora et al. / Catalysis Today 223 (2014) 138–147
147
studied in the form of mixed oxides (after calcination) and ex situ
References
or in situ rehydrated samples. Their catalytic activity was affected
both by reaction temperature and catalyst Mg/Al molar ratio. A
significant effect of reaction temperature on the product distribu-
tion was observed. While C8 alcohol formation was prevailing at
the lower temperatures, higher temperatures supported C8 alcohol
dehydration and the formation of the desired FAc and F2Ac prod-
ucts. The optimum Mg/Al molar ratio was found to be around 3. The
best results were achieved with this catalyst at 100 ◦C (>95% furfural
conversion and only a minor selectivity to C8 alcohol). Selectivity
to the main by-product DAA did not exceed 5%.
Ex situ rehydration (liquid and steam phase) led to a signifi-
cant decrease in furfural conversion, which caused also a decrease
of selectivity to the desired product (FAc and F2Ac) in comparison
with the calcined hydrotalcites, i.e. mixed oxides. This phenomenon
was observed not only at 20 ◦C, but also at 100 ◦C. Only one cata-
lyst, the one with Mg/Al molar ratio 2, preserved its high activity
(furfural conversion more than 90%) and the desired product selec-
tivity after rehydration treatment. The reason lies probably in the
highest relative crystallinity of the as prepared (dried) hydrotalcite
catalyst in comparison with the other catalysts that is transformed
to more stable and active catalysts, i.e. catalyst with a lower relative
content MgO active phase that is more sensitive to water.
[1] G. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
[2] D. Kubicˇka, I. Kubicˇková, J. Cejka, Catal. Rev. Sci. Eng. 55 (2013) 1–78.
[3] L.D. Schmidt, P.J. Dauenhauer, Nature 447 (2007) 914.
[4] J.R. Rostrup-Nielsen, Science 308 (2005) 1421–1422.
[5] C. Somerville, H. Youngs, C. Taylor, S.C. Davis, S.P. Long, Science 329 (2010)
790–792.
[6] X. Hu, L. Wu, Y. Wang, Y. Song, D. Mourant, R. Gunawan, M. Gholizadeh, C.-Z.
Li, Bioresour. Technol. 133 (2013) 469–474.
[7] J.N. Chheda, G.W. Huber, J.A. Dumesic, Angew. Chem. Int. Ed. 46 (2007)
7164–7183.
[8] D.R. Dodds, R.A. Gross, Science 318 (2007) 1250–1251.
[9] S. Gámez, J.J. González-Cabriales, J.A. Ramírez, G. Garrote, M. Vázquez, J. Food
Eng. 74 (2006) 78–88.
[10] C.J. Barrett, J.N. Chheda, G.W. Huber, J.A. Dumesic, Appl. Catal. B: Environ. 66
(2006) 111–118.
[11] A.S. Ndou, N. Plint, N.J. Coville, Appl. Catal. A: Gen. 251 (2003) 337–345.
[12] W. Shen, G.A. Tompsett, K.D. Hammond, R. Xing, F. Dogan, C.P. Grey, W.C. Conner
Jr., S.M. Auerbach, G.W. Huber, Appl. Catal. A: Gen. 392 (2011) 57–68.
[13] P.A. Zapata, J. Faria, M.P. Ruiz, D.E. Resasco, Top. Catal. 55 (2012) 38–52.
[14] R. Xing, A.V. Subrahmayam, H. Olcay, W. Qi, P. van Walsum, G. Pendse, G.W.
Huber, Green Chem. 12 (2010) 1933–1946.
[15] R.M. West, Z.Y. Liu, M. Peter, C.A. Gartner, J.A. Dumesic, J. Mol. Catal. A: Chem.
296 (2008) 18–27.
[16] D. Tichit, M.N. Bennani, F. Figueras, R. Tessier, J. Kervennal, Appl. Clay Sci. 13
(1998) 401–415.
ˇ
[17] F. Prinetto, D. Tichit, R. Teissier, B. Coq, Catal. Today 55 (2000) 103–116.
[18] B.F. Sels, D.E. de Vos, P.A. Jacobs, Catal. Rev. Sci. Eng. 43 (2001) 443–488.
[19] F. Figueras, Top. Catal. 29 (2004) 189–196.
[20] V. Rives (Ed.), Layered Double Hydroxides: Present and Future, Nova Science,
New York, 2001, pp. 251–411.
[21] S.M. Auerbach, K.A. Carrado, P.K. Dutta, Handbook of Layered Materials, Marcel
Dekker, New York, 2004.
Acknowledgments
Financial support from the Ministry of Industry and Trade of the
Czech Republic (project FR-TI3/327) and Czech Science Foundation
(P106/11/0773) is gratefully acknowledged. This publication was
created in connection with the project Unipetrol research and edu-
cation center Reg. CZ.1.05/2.1.00/03.0071, which is funded through
the Operational Programme for Research and Innovation Develop-
ment of the Structural Funds (specifically the European Regional
Development Fund) and the state budget of the Czech Republic.
[22] X. Duan, D.G. Evans (Eds.), Layered Double Hydroxides, Structure and Bonding
Series, vol. 119, Springer, Berlin, 2006.
[23] D. Tichit, D. Lutic, B. Coq, R. Durand, R. Teissier, J. Catal. 219 (2003) 167–175.
[24] H. Liu, W. Xu, X. Liu, Y. Guo, Y. Guo, G. Lu, Y. Wang, Kinet. Catal. 51 (2010)
75–80.
[25] M. Campanati, S. Franceschini, O. Piccolo, A. Vaccari, A. Zicmanis, Catal. Com-
mun. 5 (2004) 145–150.
[26] C. Xu, Y. Gao, X. Liu, R. Xin, Z. Wang, RSC Adv. 3 (2013) 793–801.
[27] D.P. Debecker, E.M. Gaigneaux, G. Busca, Chem. Eur. J. 15 (2009) 3920–3935.