Upgrading of furfural to biofuel precursors: Via aldol condensation with acetone over magnesium hydroxide fluorides MgF2- x(OH)x

Article abstract of DOI:10.1039/c9cy01259a

Wastes from lignocellulosic materials, especially hemicellulose, are extremely promising resources to produce fuels from renewable raw materials. Furfural, resulting from the depolymerization of hemicellulose, is often considered as an extremely interesting platform molecule. Particularly, new biofuels containing molecules with 8 and 13 carbon atoms can be produced from aldol condensation of furfural and acetone followed by a deoxygenation reaction. In this work, several magnesium hydroxide fluorides MgF_2-x(OH)_x were prepared by a sol-gel method with various F/Mg ratios (0 to 2) at 100 °C. All solid samples were fully characterized by several techniques (nitrogen adsorption-desorption, TEM, IR, XRD and ICP). MgF_2-x(OH)_x were mainly composed of an intimate mixture of MgF₂ and Mg(OH)_2-x(OCH₃)_x and exhibited both acid-base properties and high surface areas. From CO₂ adsorption experiments, a basicity scale corresponding to basic sites with moderate strength was established: MgF_1.5(OH)_0.5 > MgF(OH) ～ MgF_1.75(OH)_0.25 > MgF_0.5(OH)_1.5 > Mg(OH)₂ ? MgF₂. It was proposed that the presence of fluorine allowed stabilization of the basic sites with moderate strength at ambient atmosphere. The aldol condensation of furfural and acetone was carried out under mild reaction conditions (50 °C, P_atm) over MgF_2-x(OH)_x. These catalysts were involved in this reaction without using a classical activation step for basic solid catalysts, which constitutes a major advantage of energy conservation and thus, economic efficiency. The solid with a F/Mg ratio equal to 1.5 (MgF_1.5(OH)_0.5) exhibited the highest activity, the furanic dimer (1,5-di(furan-2-yl)penta-1,4-dien-3-one) being the main product. A good correlation between the catalytic activity and the basicity scale was highlighted. Based on these results, the nature of active sites was proposed: a combination of a Lewis acid site (coordinatively unsaturated metal site) in the vicinity of a basic site (hydroxyl groups of Mg(OH)_2-x(OCH₃)_x). The effect of the furfural/acetone ratio on the catalytic properties of MgF_1.5(OH)_0.5 was also investigated.

Full text of DOI:10.1039/c9cy01259a

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Xu, S. Celerier,
J. Comparot, J. Rousseau, M. Corbet, F. Richard and J. Clacens, Catal. Sci. Technol., 2019, DOI:
10.1039/C9CY01259A.
Volume
Number
21 January 2017
Pages 313-534
7
2
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
rsc.li/catalysis
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2044-4761
PAPER
Chaoqiu Chen, Yong Qin et al.
Highly dispersed Pt nanoparticles supported on carbon
nanotubes produced by atomic layer deposition for hydrogen
generation from hydrolysis of ammonia borane
rsc.li/catalysis

Please do not adjust margins
Catalysis Science & Technology
Page 1 of 11
View Article Online
DOI: 10.1039/C9CY01259A
ARTICLE
Upgrading of furfural to biofuel precursors via aldol condensation
with acetone over magnesium hydroxide fluorides MgF2-x(OH)x
Minrui Xu a, Stéphane Célerier *a, Jean-Dominique Comparot a, Julie Rousseau a, Matthieu Corbet b,
Received 00th January 20xx,
Accepted 00th January 20xx
Frédéric Richard *a, Jean-Marc Clacens a
DOI: 10.1039/x0xx00000x
Wastes from ligno-cellulosic material, especially hemicellulose, are extremely promising resources to produce fuels from
renewable raw materials. Furfural, resulting from the depolymerization of hemicellulose, is often considered as an extremely
interesting platform molecule. Particularly, new biofuels containing molecules with 8 and 13 carbon atoms can be produced
from aldol condensation of furfural and acetone followed by a deoxygenation reaction. In this work, several magnesium
hydroxide fluorides MgF2-x(OH)x were prepared by a sol-gel method with various F/Mg ratios (0 to 2) at 100 °C. All solid
samples were fully characterized by several technics (nitrogen adsortion-desorption, TEM, IR, XRD and ICP). MgF2-x(OH)x
were mainly composed of an intimate mixture of MgF2 and Mg(OH)2-x(OCH3)x and exhibited both acid-base properties and
high surface area. From CO2 adsorption experiments, a basicity scale corresponding to basic sites with moderate strength
was established: MgF1.5(OH)0.5 > MgF(OH) ~ MgF1.75(OH)0.25 > MgF0.5(OH)1.5 > Mg(OH)2 >> MgF2. It was proposed that the
presence of fluorine allowed to stabilize the basic sites with moderate strength at ambient atmosphere. The aldol
condensation of furfural and acetone was carried out under mild reaction conditions (50 °C, Patm) over MgF2-x(OH)x. These
catalysts were involved in this reaction without using classical activation step for basic solid catalysts, which constitutes a
major advantage for energy saving and thus, economical efficiency. The solid with the F/Mg ratio equal to 1.5 (MgF1.5(OH)0.5
)
exhibited the highest activity, the furanic dimer (1,5-di(furan-2-yl)penta-1,4-dien-3-one) being the main product. A good
correlation between the catalytic activity and the basicity scale was highlighted. Based on these results, the nature of active
sites was proposed: a combination of a Lewis acid site (coordinatively unsaturated metal site) in the vicinity of a basic site
(hydroxyl groups of Mg(OH)2-x(OCH3)x). The effect of the furfural/acetone ratio on the catalytic properties of MgF1.5(OH)0.5
was also investigated.
reactions play a crucial role in the upgrading of furfural since this
approach increases the carbon chain length by forming new C-C
bonds. Especially, the aldol condensation of furfural and acetone
leads to products containing either 8 (4-(furan-2-yl)but-3-en-2-one,
FAc) or 13 carbon atoms (1,5-di(furan-2-yl)penta-1,4-dien-3-one,
F2Ac), as described in scheme 1.11 In addition, their
hydrodeoxygenation were studied over metallic based catalysts to
upgrade these compounds as renewable fuels.17–20
Introduction
Nowadays environment-friendly development arouses more and
more attention to the public. For instance, extensive efforts are
made to replace petroleum resources by renewable biomass-derived
chemicals.1,2 Among the main components of agricultural wastes,
lignocellulose remains the most abundant biomass.3 Hemicellulose,
representing about 25% of lignocellulose, can be considered as a
precursor of furfural,
a promising biomass-derived platform
O
O
molecule, mainly obtained by acid-catalyzed dehydration of xylose.4–
O
O
O
OH
H
O
O
6
O
O
O
O
O
H
-H2O
-H2O
Since the research of bio-based building block synthesis from furfural
continues to burgeon, multifarious value-added compounds were
produced from furfural.7–11 The synthesis of new types of biofuels
can also be considered from furfural, involving aldol condensation
and hydrodeoxygenation reactions.12–16 Aldol condensation
FAcOH
FAc
F2Ac
Scheme 1: Schematic illustration of adol condensation between furfural and
acetone
Generally, sodium hydroxide is a well-known homogeneous basic
catalyst for aldol condensation. For example, Fakhfakh et al.21
obtained a mixture containing FAc and F2Ac in equimolar amounts
when a total conversion of furfural was reached under mild
experimental conditions (40 °C, atmospheric pressure).
Nevertheless, the use of such catalyst leads to many drawbacks,
especially due to its high corrosiveness. In addition, homogeneous
catalytic processes commonly encounter disadvantages, including
a. Institut de Chimie des Milieux et Matériaux de Poitiers, UMR 7285 Université de
Poitiers - CNRS, 4 rue Michel Brunet, BP633, 86022 Poitiers Cedex, France. E-mail :
stephane.celerier@univ-poitiers.fr, frederic.richard@univ-poitiers.fr
b. Solvay Research & Innovation center of Lyon, 85 rue des Frères Perret, BP 62
69192, Saint Fons, France
Electronic Supplementary Information (ESI) available: [details of any supplementary
information available should be included here]. See DOI: 10.1039/x0xx00000x
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 2 of 11
ARTICLE
Journal Name
the neutralization of reaction mixture producing wastes and the atmospheric pressure)33 and for condensationViewoArfticle2O,3nl,in6e-
DOI: 10.1039/C9CY01259A
difficulty of catalysts separation. This reaction was also studied over trimethylhydroquinione with isophytol (100 °C, atmospheric
solid catalysts possessing basic properties such as metal oxides.22–25 pressure).34 On the contrary, magnesium oxide (hydroxide) fluorides
For example, Shen et al.22 obtained a high conversion of furfural MgF2-x(OH)x possess basic properties. Such solids were successfully
(90%) with a selectivity on furanic dimer (F2Ac) close to 65% over used as either basic catalysts for the Michael addition of 2-
MgO-ZrO2 in presence of a furfural/acetone molar ratio equal to 1 methylcyclohexane-1,3-dione to methyl vinyl ketone (25 °C,
under drastic experimental conditions (120 °C and 5.2 MPa). Under atmospheric pressure)35 or catalytic support.36–40 Moreover, the use
milder conditions (60 °C, atmospheric pressure) and in presence of of solid basic catalysts involves generally an activation step at high
an excess of acetone (furfural/acetone molar ratio close to 0.1), a temperature to active basic sites by removal of carbonate species.
high conversion of furfural (about 70%) was also reached over MgO- Find an active catalyst which does not require this activation step
ZrO2 mixed oxide catalysts.23 The alcohol intermediate (4-(furan-2- could be interesting for energy saving and economic point of view
yl)-4-hydroxybutan-2-one, FAcOH) was always the main product, when the reaction is performed at low temperature.
whereas the selectivity of F2Ac was always less than 10 mol%,
In this work, various hydroxide fluorides MgF2-x(OH)x solid samples
with different F/Mg ratio were prepared by a sol-gel method
according to the procedure developed by Scholz et al. 41. The easy
control of the F/Mg ratio of these materials allows to fine tune their
acid-base properties. These solid materials were evaluated as
catalyst without preliminary thermal activation for aldol
condensation of furfural and acetone under mild conditions (50 °C,
atmospheric pressure) in a batch reactor. Based on these results, the
nature of active sites was discussed and a reaction mechanism was
proposed.
irrespective to the furfural conversion. Over the same type of catalyst
at 50 °C but under nitrogen pressure (1 MPa), a conversion of furfural
close to 80% was obtained by using an equimolar mixture of acetone
and furfural.24 In this case, the reaction mixture was mainly
composed by F2Ac (75 mol% in selectivity). In presence of a
mesoporous TiO2 catalyst exhibiting both basic and acid sites, a low
conversion of furfural was obtained (25%) at 50 °C.25 Probably due to
the high amount of acetone compared to furfural (furfural/acetone
ratio equal to 0.2), the monomer (FAc) was the main detected
product, its selectivity being higher than 70 mol%. In this case, the
alcohol intermediate (FAcOH) was observed in traces due to the
presence of acid sites favouring its dehydration in FAc. Due to this,
solids having acidic properties were also evaluated as catalysts for
this reaction, such as zeolites,26,27 acids metal chlorides28 and metal-
organic frameworks.29,30 For instance, a HBEA zeolite was used as
catalyst for aldol condensation of furfural and acetone even if the
furfural conversion was only 38% at 100 °C (furfural/acetone molar
ratio equal to 0.1) as shown by Kikhtyanin et al.26 Under these
experimental conditions, as discussed above, an excess of acetone
and the presence of acid sites explained that FAc was the main
product detected in the reaction mixture. Nevertheless, an
undesired reaction between two molecules of F2Ac leading to a
heavy product, noted (FAc)2, was observed due to the presence of
Brønsted acid sites. When Sn was added in the HBEA zeolite in order
to promote its Lewis acidity, a very high furfural conversion (more
than 90%) was reached at 160 °C (furfural/acetone molar ratio equal
to 0.1).27 FAc was the main product and the presence of (FAc)2 was
not mentioned. Interestingly, when acidic catalysts were used, the
monomer (FAc) was always much more favored than the dimer
(F2Ac).26–30 As a consequence, in order to favor the C13 furanic dimer
(F2Ac) instead of FAc, during the aldol condensation of furfural and
acetone, the use of basic catalysts seems to be more appropriate.
Experimental
Synthesis of catalysts
The magnesium hydroxide fluorides MgF2-x(OH)x with x = 2, 1.5, 1,
0.5, 0.25 and 0 were prepared by a sol-gel method, partly based on
the work of Scholz et al.41 In the first step, magnesium metal (3.23 g,
Aldrich, 99.98%) was treated with methanol in excess (100 mL,
Sigma-Aldrich, 99.8%) under reflux conditions for 6 h to form a
Mg(OCH3)2 metal alkoxide solution. Various volumes of aqueous HF
(48 wt% HF in water, Sigma-Aldrich, 99.99%) were added
progressively under stirring conditions to obtain several HF/Mg
molar ratios, ranging from 0.5 to 2. The amount of water (provided
from aqueous HF or/and additional water) was adjusted to obtain in
each case a H2O/Mg ratio of 2 to favor the hydrolysis of remaining
OCH3 groups. A highly exothermic reaction occurred to obtain a sol.
This sol was stirred for 24 h, aged at ambient temperature for 24 h
and dried at 100 °C for 24 h. According to this synthesis procedure,
five solids named MgF2, MgF1.75(OH)0.25, MgF1.5(OH)0.5, MgF(OH) and
MgF0.5(OH)1.5 were obtained with the initial HF/Mg molar ratio of 2,
1.75, 1.5, 1 and 0.5, respectively. To synthesize the sample named
Mg(OH)2, the protocol was the same except that the aqueous HF
solution was replaced by water (initial H2O/Mg molar ratio equal to
2.5).
In the past decade, inorganic oxide hydroxide fluorides with high
specific surface area prepared by soft chemistry were developed as
a new and performant class of catalyst and/or (active) support.31,32
Such solids are known to exhibit different types of acid-base
properties. As an example, aluminium hydroxide fluorides AlF3-x(OH)x
exhibit tuneable acid properties, with neighbouring Lewis and
Brønsted acid sites. These fluorides were successfully used as
catalyst for acylation of 2-methylfuran by acetic anhydride (50 °C,
Catalyst characterizations
The magnesium composition of each catalyst was determined by
inductively coupled plasma optical emission spectrometry (ICP-OES)
using a PerkinElmer Optima 2000DV instrument. The fluorine
content was measured by ionic chromatography (DIONEX ICS-2000).
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 11
Catalysis Science & Technology
Please do not adjust margins
Journal Name
ARTICLE
X-ray diffraction (XRD) patterns were collected between 2θ = 5 and Heptane, inert in these conditions, was used as an internal standard
View Article Online
DOI: 10.1039/C9CY01259A
75° using a PANalytical EMPYREAN powder diffractometer using Cu- for quantitative analysis carried out by Gas Chromatography. Each
liquid sample obtained after various reaction times (between 30 and
360 min) was analysed by using a Varian 430 chromatograph
equipped with CPSIL–5CB capillary column (length: 50 m; diameter:
0.25 mm, film thickness: 0.4 m) and a flame ionization detector. The
oven temperature was programmed from 70 °C to 150 °C (5 °C min-
1), then from 150 °C to 250 °C (10 °C min-1) and maintained at 250 °C
during 10 minutes. All detected products were identified by co-
injection of commercial samples and/or by using Varian 3800
chromatograph coupled with a 1200 TQ mass spectrometer.
Kα monochromatized radiation source (K = 1.54059 Å) with a 0.1°
step and 600s dwell time. Phase identification was performed by
comparison with the ICDS database reference files.
The textural properties were determined from N2 adsorption-
desorption isotherms, which were measured using a TRISTAR 3000
gas adsorption system at -196 °C. Prior to N2 adsorption, the samples
were degassed overnight under secondary vacuum at 90 °C. The
specific surface area (SBET in m2 g-1) was calculated from the
adsorption isotherm (P/P0 between 0.05 and 0.20) using the
Brunauer-Emmett-Teller (BET) method. The total pore volume was The conversion of furfural (XF in %) and the yield in each product (Y
calculated from the adsorbed volume of nitrogen at P/P0 equal to in mol%) were calculated using the following equations, according to
0.99. The average mesopore-size distribution was calculated from O’Neill et al. :42
the desorption isotherm branch using the Barret-Joyner-Halenda
(BJH) method.
퐶퐹,0 ― CF
푋퐹 =
. 100
(1)
(2)
(3)
퐶퐹,0
The measurement of the basicity was obtained by adsorption of CO2
followed by FTIR spectroscopy using a ThermoNicolet NEXUS 5700
spectrometer (resolution of 2 cm-1 and 128 scans per spectrum).
Solid samples were pressed into thin pellets (15-30 mg) with
diameter of 16 mm under a pressure of 1-2 t cm-2 and activated in
situ during two hours under secondary vacuum (2 10-6 bar) at 90 °C.
After cooling down the samples until room temperature, CO2 was
introduced in the cell, which was placed under vacuum and heated
at 50 °C to remove physisorbed CO2. Finally, IR spectrum was
recorded and normalized at the same mass for each sample. The
measurement of the acidity of all solid samples was obtained by
adsorption of CO followed by FTIR spectroscopy. Experiments were
carried out with the same equipment and the same activation
procedure (90°C) such as those used for CO2 adsorption experiments.
The cell was cooled down with liquid nitrogen to -173 °C. Then
successive doses of CO were introduced quantitatively and a
spectrum was recorded after each adsorption until saturation. The
final spectrum was recorded with 1 Torr of CO at equilibrium
pressure (saturation). All spectra were normalized at the same mass
of magnesium hydroxide fluoride samples (20 mg).
CFAcOH
. 100
퐶퐹,0
푌퐹퐴푐푂퐻
=
CFAc
푌퐹퐴푐
=
. 100
퐶퐹,0
2.CF2Ac
푌퐹 퐴푐
=
. 100
(4)
2
퐶퐹,0
with CF,0 being the initial concentration of furfural, CF, CFAcOH, CFAC and
CF2Ac being the concentrations of various compounds determined at
a given reaction time. Whatever the catalysts, the molar balance
carbon is between 95 and 100%.
Results and discussion
Characterization of the catalysts
The so-called Mg(OH)2 sample was prepared by a sol gel method
without added HF aqueous solution during the synthesis. The XRD
pattern of this solid exhibited broad peaks highlighting a low rate of
crystallinity (Fig. 1).
Transmission Electronic Microscopy (TEM) experiments were carried
out using a JEOL 2100 instrument (operated at 200 kV with a LaB6
source and equipped with a Gatan Ultra scan camera) coupled with
an Energy Dispersive X-ray Spectrometer (EDS) allowing the
determination of F, O, C and Mg contents. For EDS measurements,
10-15 analyses were performed for each studied samples (MgF2,
Mg(OH)2 and MgF(OH)) to obtain an average value. SAED (Selected
area electron diffraction) experiments were also performed.
Catalytic procedure for aldol condensation of furfural and acetone
The reaction was carried out in a batch reactor fitted with a magnet
stirrer and heated at 50 °C. The catalyst (200 mg) was introduced in
a mixture composed of furfural (5 mmol, Sigma–Aldrich, 99%,
preliminarily distilled), acetone (10 mmol, 5 mmol or 2.5 mmol,
Sigma–Aldrich, >99%), heptane (7.5 mmol, Sigma–Aldrich, 99%) and
2-methyltetrahydrofuran as solvent (5 mL, Alfa Aesar, 99%).
Fig. 1 X-ray diffraction patterns of MgF2-x(OH)x samples. Dotted lines: mean diffraction
peaks of MgF2 (JCPDS file N° 01-070-2268); solid lines: main diffraction peaks of
Mg(OH)1.3(OCH3)0.7 (JCPDS file N°00-022-1788)
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 4 of 11
ARTICLE
Journal Name
As expected, the formation of layered-methoxide of magnesium with When an initial HF/Mg ratio equal to 2 was used during the synthesis,
View Article Online
DOI: 10.1039/C9CY01259A
the Mg(OH)2-x(OCH3)x formula was observed attesting that the the solid corresponds to MgF2 crystallographic phase as shown by
hydrolysis of magnesium alkoxide by water was incomplete even if XRD (Fig 1). The F/Mg molar ratio was determined by a combination
water was added in excess, in agreement with Rywak et al. 43 A weak of ICP-OES for Mg content, and ionic chromatography for F content.
shift towards lower angles of the peaks was observed in comparison The value, given in Table 1, is in agreement with the expected one.
with the peaks of the JCPDS file of the Mg(OH)1.3(OCH3)0.7 material. This solidsample can be defined by a homogeneous porous structure
This composition is probably not entirely the same as our synthetized composed of small crystallites of 4-7 nm as observed by TEM (Fig. 2).
material explaining this shift. This solid sample exhibited
a
The F/Mg ratio determined by EDS was close to 2.1 whatever the
microstructure composed of crumpled sheets as shown by analysed region attesting the formation of MgF2. From the
transmission electron microscopy (Fig. 2). The observation of sheets corresponding energy-filtered SAED pattern, d spacing of 3.4, 2.2 and
is in accordance with the formation of layered-methoxide 1.7 Å were determined and corresponded to (110), (111) and (121)
magnesium as observed by XRD (Fig. 1). As expected from the diffraction planes of MgF2.
stoichiometry of this sample (Mg(OH)2-x(OCH3)x), the O/Mg ratio
Several magnesium hydroxide fluorides solids MgF2-x(OH)x, with x
equal to 0.25, 0.5, 1, 1.5 and 1.75, were also prepared by a sol-gel
method as described in the experimental part. For each solid sample,
the experimental F/Mg ratio was close to the target value (Table 1).
In these conditions, a competition between fluorolysis (reaction with
HF) and hydrolysis (reaction with H2O) occurred. As the rate of the
former reaction being higher, the fluorine content in the magnesium
hydroxide fluoride is easily tuned and reasonably controlled by the
initial HF/Mg molar ratio thanks to this sol-gel method and increases
determined by EDS was close to 2.0. Using the same technic, the
carbon content of this material was also evaluated. Indeed, a value
of x corresponding to the number of OCH3 groups was approximately
ranged between 0.5 and 0.9, close to the reference indicated in XRD
patterns (equal to 0.7). The presence of such methoxy groups was
confirmed by FT-IR (Fig. 3). Indeed, the bands between 2800 and
2950 cm-1 can be assigned to OCH3 groups.44
linearly with the amount of HF added, in agreement with results
35
reported by Prescott et al.
In these samples, at least two
crystallographic phases can be detected (Fig. 1). For example, the
XRD pattern of MgF(OH) sample exhibited the peaks of both MgF2
and Mg(OH)2-x(OCH3)x phases. As expected, higher the initial HF/Mg
molar ratio used in solid synthesis is, higher the contribution of the
MgF2 phase in the considered pattern is at the expense of Mg(OH)2-
x(OCH3)x phase.
Table 1: Physico-chemical properties of magnesium hydroxide fluoride MgF2-x(OH)x
samples
F/Mg
ratio1
0
0.4
1.1
1.6
1.8
2.0
SBET
Vpore
dpore
Basicity5
(a.u.)
0.005
0.039
0.069
0.174
0.068
0
2
3
4
Catalyst
(m2 g-1)
313
630
599
406
(cm3 g-1)
0.56
0.73
0.75
0.75
(nm)
6.1
4.5
5.2
7.2
6.0
4.4
Fig. 2 TEM images of Mg(OH)2 (a and b), MgF(OH) (c and d) and MgF2 (e and f).
Corresponding SAED pattern (insert image) of picture e
Mg(OH)2
MgF0.5(OH)1.5
MgF(OH)
MgF1.5(OH)0.5
MgF1.75(OH)0.25
MgF2
387
280
0.55
0.26
[1] Fluor content measured by ionic chromatography and Mg content
measured by ICP-OES; [2] Specific surface area measured by the BET method;
[3] Total pore volume calculated at P/P0 equal to 0.99; [4] Average pore size
deduced from the BJH method using the desorption branch; [5] Determined
from the area of the band at 1225 cm-1 obtained after activation at 90 °C under
vacuum and adsorption-desorption of CO2 at 50 °C (values deduced from
spectra presented in Fig. 4)
The formation of this intimate mixture of phases was confirmed by
TEM analysis (Fig. 2). As an example, two well separated zones were
visible in the TEM image for the MgF(OH) sample (Fig. 2c). The zone
“A” presents a microstructure comparable to MgF2 phase given in
Fig. 2f. The determined F/Mg ratio of 1.9 by EDS was close to the
expected value (equal to 2). The corresponding SAED (not shown in
Fig 2d) was equivalent to the one inserted in Fig. 2e. The zone “B”
Fig. 3: IR spectra of MgF2-x(OH)x samples after activation at 90 °C under vacuum
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 11
Catalysis Science & Technology
Please do not adjust margins
Journal Name
ARTICLE
with crumpled sheets is similar to the Mg(OH)2-x(OCH3)x phase shown that the presence of fluorine seems to favour the retention of
View Article Online
DOI: 10.1039/C9CY01259A
in Fig. 2b. In addition, a O/F/Mg ratio of 1.2/0.9/1 measured by EDS, adsorbed water. This is probably linked to the increase of the Lewis
indicates the presence of an intimate mixture of MgF2 and Mg(OH)2- acidity of solid samples with the fluorine content as confirmed below
x(OCH3)x in the zone “B”. Nevertheless, the insertion of a fraction of with the characterization of acidity. It can be noted that the
fluorine in Mg(OH)2-x(OCH3)x and of hydroxyl group in MgF2 cannot activation step used for IR analysis (degassing at 90°C under vacuum)
be obviously excluded to form MgF2-x-y(OH)x(OCH3)y mixed phase as was insufficient to remove all the water adsorbed on these samples.
already proposed.35 Moreover, taking into account the low
crystallinity of these samples showed by XRD (Fig. 1), each phase is
composed of randomly-dispersed small domain leading to a very
close proximity and a high amount of interfaces between them.
The adsorption-desorption isotherms of all solids are shown in Fig.
S1. A type IV isotherm according to the IUPAC classification was
observed for all samples whatever the fluorine content showing the
formation of mesoporous solids with a pore volume between 0.3 and
0.7 cm3 g-1 (Table 1). The average pore size determined by the BJH
method is between 4 and 7 nm for all materials. The specific surface
areas of these solids are very high (from 280 m2 g-1 for MgF2 to 630
m2 g-1 for MgF0.5(OH)1.5), in accordance with the low crystallinity
observed by XRD (Fig. 1). These surface areas are relatively close to
the values obtained for the magnesium hydroxide fluorides prepared
Fig. 4 Difference of IR spectra obtained after CO2 adsorption at 50 °C and after activation
at 90 °C under vacuum of MgF2-x(OH)x samples with 0  x  2
in similar conditions by Scholz et al.41 Interestingly, the specific
surface areas of the MgF2-x(OH)x samples were drastically higher than
those of MgF2 and Mg(OH)2 pure materials. The synthesis of an
According to the results reported by Wuttke et al.,45 the OH groups
intimate mixture of MgF2 and Mg(OH)2-x(OCH3)x allowed hence to
present on such MgF2-x(OH) x materials exhibit either acidic or basic
obtain very high specific surface areas. This can be explained by a
properties. This is obviously due to the stronger inductive effect of
delay in the crystallization processes due to the presence of a second
fluorine atoms in comparison to oxygen atoms. Thus, the easy
phase which affects the rate of crystallization of the first one.36 This
control of the F/Mg ratio of these materials by sol-gel process allows
phenomenon is in favour of the nanoparticles formation, as observed
to fine tune their acid-base properties. When x is low (about less than
by TEM (Fig. 2).
0.1) corresponding to the so-called MgF2 in the present study, it was
proposed that OH groups in minor quantities are Brønsted acid sites
After their activation under secondary vacuum at 90 °C, all catalysts,
and this solid is an acidic material, exhibiting mainly Lewis acidity
initially synthesized at 100°C, were characterized by FT-IR before (Fig.
attributed to the presence of under coordinated Mg2+ species.46,47 It
3) and after CO2 adsorption (Fig. 4). Concerning the spectra shown in
can explain the absence of IR bands in the carbonate and hydrogen
carbonate region (between 1350 and 1640 cm-1) as shown in Fig. 3.
Fig. 3, the sharp band observed around 3730 cm-1 is assigned to υ(MgO-
41
.
As expected, the intensity of this band decreases with the
H)
On the contrary, with x higher than 0.1, the OH groups can be
considered as basic sites. Consequently, several bands between 1350
and 1640 cm-1 can be observed due to reaction between hydroxyl
groups and atmospheric CO2. Besides, their intensity increased with
the decrease of the fluorine content.
increase of the fluorine content, due to the lower amount of OH
groups. According to Prescott et al.,35 a weak band centred at 3683
cm-1 can be attributed to the bridging of OH groups in MgF2-x(OH)x
samples, especially observed when x  1, i.e. for MgF(OH),
MgF1.5(OH)0.5 and MgF1.75(OH)0.25 samples. The bands at 2930, 2870
and 2810 cm-1 can be attributed to υ(CH) of the methoxy groups
present in these magnesium hydroxide fluoride samples.41 The
decrease of their intensity with the increase of F content can be
explained by the lower amount of Mg(OH)2-x(OCH3)x phase in the
solid samples, in accordance with XRD analysis. The strong bands
between 1650 and 1400 cm-1 originate from carbonate and hydrogen
carbonate species formed by the reaction of atmospheric CO2 with
the MgF2-x(OH)x materials basic sites.35,41 The strong intensity of
these bands observed for the Mg(OH)2 solid can be explained by the
highest basicity of such solid. In addition, adsorbed water can lead to
a broad band located between 3400-3600 cm-1 and a sharp band
To deeper understand the acid-base properties of the as-synthesized
MgF2-x(OH)x samples, additional experiments using CO2 adsorption-
desorption procedure were carried out at 50 °C under vacuum. The
spectra presented in Fig. 4 were obtained through subtraction
between the spectrum realized after CO2 adsorption followed by
heating at 50 °C under vacuum and those measured after the
activation procedure at 90 °C (presented in Fig. 3). As expected, no
bands were observed for MgF2 highlighting the absence of basic sites
on this solid. For all other samples, several bands appeared between
1100 – 1900 cm-1, showing that some basic sites remained available
to adsorb CO2 and to form carbonate and bicarbonate species.48–50
The decrease of the fluorine content is linked to an increase of
several bands, including notably those located at 1358 and 1460 cm-
43
around 1659 cm-1, as reported by Rywak et al. These bands were
particularly intense on IR spectra of solids having x  0.5, indicating
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 6 of 11
ARTICLE
Journal Name
1, which could be attributed to the presence of carbonate species. increases with the fluorine content as showed by the iVniecwreAartsiceleoOfntlhinee
-1
DOI: 10.1039/C9CY01259A
These species can be formed on strong basic sites. On the contrary, surface of the IR band located at 2180 cm .
the band at 1225 cm-1 which corresponds to the δ(OH) vibration of
Catalytic properties of magnesium hydroxide fluorides
hydrogen carbonate species, allows to evaluate the presence of basic
sites with moderate strength. The area of this band was reported in
Table 1 for all solids. Surprisingly, the amount of these sites increases
with the fluorine content in MgF2-x(OH)x series (x in the range of 2 –
0.5), whereas the inductive effect of fluorine is rather known to
decrease the basicity of the solids.45 The higher amount of basic
hydroxyl group with low strength can be explained by the higher
amount of fluorine in the vicinity of this hydroxyl group leading to
decrease the strength of basicity. In the case of low-fluorine-
containing magnesium hydroxide fluorides, the majority of hydroxyl
groups are probably basic with high strength which react with CO2 of
ambient air. Thus, the fluorine atoms included in these materials act
as stabilizer of basic hydroxyl groups with low strength explaining
why MgF1.5(OH)0.5 retains higher amount of basic hydroxyl groups.
The activity of magnesium hydroxide fluorides were evaluated in
aldol condensation of furfural (Furf) and acetone (Ac) at 50 °C under
atmospheric pressure. Except for the MgF2 sample which was totally
inactive in this reaction, furfural conversion increased with the
increase of the experiment duration (Fig. 6). The Mg(OH)2 sample,
which was synthesized in absence of fluorine, led to only a
conversion of furfural close to 13% after 6 h of reaction. Interestingly,
the presence of fluorine in solids favoured their catalytic properties
for aldol condensation of furfural. Indeed, the conversions of furfural
ranged between 19 and 84% after 6 hours, depending on the F/Mg
ratio used, the most active being the MgF1.5(OH)0.5 sample. It can be
mentioned that the difference in the specific surface area between
catalysts seems not to be a key parameter to explain the difference
in the catalytic properties of the magnesium hydroxide fluorides.
Indeed, MgF1.5(OH)0.5 exhibited a lower specific surface area than
MgF0.5(OH)1.5, 406 and 630 m2 g-1, respectively (Table 1), but the
former was about 2.7 times more active than the latter (Fig. 7). In the
same way, while the specific surface areas of MgF1.5(OH)0.5 and
MgF1.75(OH)0.5 are very close (406 and 387 m2 g-1, respectively) , the
former is much more active than the latter (Fig. 7). Such a finding was
already observed for several mixed-oxides using as catalysts for aldol
condensation of furfural and acetone 24. On the other hand,
commercial MgO basic catalyst was also evaluated for this reaction
under the same experimental conditions (without activation step).
The obtained conversion was only 13% after 6h of reaction showing
the interest of using hydroxide fluoride as catalyst for the aldol
condensation of furfural by acetone.
Fig. 5 Difference of IR spectra obtained after CO adsorption at saturation at -173 °C and
after activation at 90 °C under vacuum of MgF2-x(OH)x samples with 0  x  2.
The initial reaction rate of each catalyst can be evaluated by taking
the tangent at the origin of each curve depicted in Fig. 6.
Interestingly, a linear correlation was found between the initial
reaction rate and the F/Mg molar ratio until 1.5 (Fig. 7). Indeed, an
increase of the F/Mg ratio in the catalyst, between 0 and 1.5, favored
the reaction rate in aldol condensation. Nevertheless, the two
samples with the highest F/Mg ratios (1.75 and 2) exhibited a very
limited activity. The initial reaction rate of MgF1.75(OH)0.25 was about
7 times lower than the one of MgF1.5(OH)0.5, and MgF2 was totally
inactive. It was already reported that this type of catalyst exhibited
virtually only acidity.40,47 This is here confirmed by the
characterization of acid-base properties using both CO and CO2
probe molecules. As proposed by Kikhtyanin et al.,26 such reaction
was rather catalyzed by Brønsted acid sites explaining, at least in part,
the poor catalytic properties of MgF2 for aldol condensation of
furfural. Moreover, acidic catalysts were often used in more drastic
experimental conditions, between 100 °C29,30 and 140 °C28, i.e. at
higher temperature than the one used in the present study (50 °C).
The acid properties were also determined by adsorption of CO
followed by IR. CO is
a suitable probe molecule for the
characterization of Lewis acidity because the (CO) frequency is very
sensitive to the local cationic environment.51 After CO adsorption,
this band shifts more or less depending on the strength of Lewis
acidity: the stronger the Lewis acidity, the higher the shift. The
spectra obtained after CO saturation for each catalyst are reported
in Fig. 5. Except for Mg(OH)2 and MgF0.5(OH)1.5, the main (CO) band
is observed at 2180 cm-1 and corresponds to CO adsorbed on Lewis
acid sites with moderate strength.52,53 Taking into account that no
significant shift was observed in the OH region (around 3700 cm-1),
which is characteristic of CO adsorption on Brønsted acid sites (not
showed here),54 we can reasonably assigned the band at 2180 cm-1
to an interaction of CO with Lewis acid sites. The low amount of
Brønsted acidity on MgF2 prepared in the same conditions was
already observed.47 Conversely, Mg(OH)2 and MgF0.5(OH)1.5 have not
or few Lewis acid sites. Only a band at 2138 cm-1 is observed
corresponding to the weak and unspecific CO adsorption
(physisorbed CO species).53 As expected, the amount of Lewis acidity
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 11
Catalysis Science & Technology
Please do not adjust margins
Journal Name
ARTICLE
magnesium stabilized by the presence of fluorine) in the vicinity of a
View Article Online
DOI: 10.1039/C9CY01259A
basic site (hydroxyl group). In order to validate this hypothesis, a
mechanical mixture containing 25 wt% of Mg(OH)2 and 75 wt% of
MgF2 was evaluated as catalyst for this reaction. The initial rate of
this mechanical mixture was about 17 times lower than the one of
the corresponding MgF1.5(OH)0.5 confirming the necessity to have
neighboring sites obtained thanks to the intimate mixture of MgF2
and Mg(OH)2-x(OCH3)x prepared by the sol-gel method used in our
work.
Fig. 6 Aldol condensation of furfural and acetone over magnesium hydroxide fluorides
(MgF2-x(OH)x) at 50 °C under atmospheric pressure (furfural/acetone ratio equal to 0.5).
Influence of the reaction time on the conversion of furfural. over Mg(OH)2 (),
MgF0.5(OH)1.5 (), MgF(OH) (), MgF1.5(OH)0.5 (), MgF1.75(OH)0.25 () and MgF2()
A correlation was also found between the F/Mg molar ratio and the
surface of the IR band at 1225 cm-1 measured after adsorption of CO2
as shown in Fig. 7. As indicated in the previous part, such band
corresponds to (OH) of hydrogen carbonate species and allows to
quantify the number of basic sites with moderate strength.48–50
Consequently, the aldol condensation activity can be directly
connected to the number of OH groups involved in the formation of
hydrogen carbonate species deduced from CO2 adsorption
experiments. Such correlation was already reported between the
activity for carbon oxysulfide hydrolysis of various mixed oxides
(TiO2-Al2O3, TiO2-ZrO2 and ZrO2-Al2O3) and the intensity of the band
at 1225 cm-1.48,49 As observed in Fig. 4, other basic sites are present
in our samples characterized by the bands at 1358 and 1460 cm-1
attributed to the presence of carbonates species produced by
reaction with CO2. The intensity of these bands increases with the
decrease of fluorine in the magnesium hydroxide fluoride (Fig 4),
which is the opposite of the catalytic properties evolution (Fig. 7).
Nevertheless, we cannot totally exclude their involvement in the
aldol condensation reaction as active sites, even if their influence is
probably minor.
Fig. 7 Aldol condensation of furfural and acetone over magnesium hydroxide fluorides
MgF2-x(OH)x and over a mechanical mixture composed by 25 wt% of Mg(OH)2 and 75 wt%
of MgF2 at 50 °C under atmospheric pressure (furfural/acetone ratio equal to 0.5). Effect
of the F/Mg molar ratio (i) on the initial reaction rate of aldol condensation of furfural by
acetone (red) and (ii) on the number of medium basic sites evaluated by the area of the
band determined at 1225 cm-1 after CO2 adsorption-desorption at 50 °C (blue).
The basic site can lead to the formation of carbanion species from
the abstraction of the -proton from acetone (Scheme 2). The Lewis
acid site could favor the nucleophile addition onto furfural through
its adsorption by the oxygen atom of the carbonyl group. Such
assumption, i.e. the participation of a Lewis acid site located in the
vicinity of a basic site with medium strength, to explain the catalytic
properties of different mixed-oxide catalysts (Mg-Zr, Mg-Al and Ca-
Zr) in aldol condensation of furfural, has been already suggested by
Faba et al. 24 In addition, these authors proposed that the activity of
mixed-oxide catalysts was promoted by medium-strength base sites,
in accordance with our proposals.
For this reaction, the involvement of two types of active sites has
been suggested: electron-donating sites (hydroxyl ions and/or oxide-
ion defect) acting as reduced sites and Lewis acidic sites
(coordinatively unsaturated metal site).55 Similarly, we can propose
the participation of two types of sites to explain the catalytic
properties of the MgF2-x(OH)x solid materials:
a vacancy on
magnesium present on the MgF2 phase, acting as Lewis acid site, and
hydroxyl groups with moderate strength present on Mg(OH)2 phase,
acting as basic site (Scheme 2). The reaction was then produced at
the interface between both phases where basic and acid sites are
close and it was facilitated by the probable high concentration of
interfaces in these nanoscopic materials as discussed previously.
Moreover, since the formation of amorphous MgF2-x-y(OH)x(OCH3)y
mixed phase cannot be excluded, some acid-pair sites can be also
present on the same phase: a Lewis acid site (undercoordinated
Scheme 2: Proposals concerning the active sites involved during the aldol condensation
of furfural and acetone
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 8 of 11
ARTICLE
Journal Name
It is very important to mention that all MgF2-x(OH)x samples were (furan-2-yl)but-3-en-2-one) and F2Ac (1,5-di(furan-2-yl)penta-1,4-
View Article Online
DOI: 10.1039/C9CY01259A
used without any treatment. Indeed, as basic solids are very sensitive dien-3-one), as depicted in Scheme 1. Surprisingly, FAc-OH, expected
to ambient atmosphere by reaction with several molecules such as as the only primary product, was detected in low quantities even at
H2O or CO2, it is known that their use as basic catalysts requires a pre- low level of furfural conversion, its yield being only equal to 4 mol%
treatment at a high temperature (300-800 °C depending on the solid at 22 mol% of conversion (after 30 min of reaction). An increase of
composition and the basic site strength) before reaction in order to the reaction time led to an increase of the yield of the two other
reveal basic sites on the catalytic surface.56 For example, the products: FAc obtained by dehydration of FAc-OH, and F2Ac arising
calcination of a Mg-Al hydrotalcite material at 450 °C resulted in a from interaction of FAc-OH with another furfural molecule.
doubling of the furfural conversion for its aldol condensation.25 In Irrespective the reaction time and thus the furfural conversion, the
addition, this activation step can lead to a drastic decrease of the furanic dimer (F2Ac) was always the main product. This result can be
specific surface area of the solid and thus of the amount of active explained by the fact that the kinetic constant of the final reaction
sites. The preparation of basic solid catalysts allowing to retain basic (condensation of FAc and Furf) is much greater than the
sites without a calcination step is hence very attractive for, at least, condensation of acetone and furfural yielding C8 units (initial
both reasons: (i) to avoid costly heat-treatment for an industrial reactions), as demonstrated by Faba et al.24 In addition, as shown in
process, particularly for a reaction performed at low temperature Table 2, the selectivity of the four catalysts containing fluoride was
(lower than 100 °C) which not requires devices adapted for high practically independent on their fluorine content. Indeed, the main
temperature, (ii) to retain high specific surface area which will be product was always the furanic dimer, its amount being close to 50
favourable to maintain high active site concentration.
mol% irrespective of the catalyst. As F2Ac is a more bulky molecule
compared to FAc, the fact that the former was favoured compared
to the latter over magnesium hydroxide fluoride catalysts seems to
indicate the absence of diffusion limitations over these mesoporous
catalysts. It is important to emphasize that diacetone alcohol
produced by aldol condensation of acetone was never detected in
our experimental conditions.
Table 2: Aldol condensation of furfural and acetone at 50 °C under atmospheric pressure
over magnesium hydroxide fluorides. Effect of the catalyst on the product distribution
determined at the same level of conversion of furfural (close to 30%) except for Mg(OH)2
Conv.
Furfural
(%)
Distribution of products (%)
Catalyst
FAc-OH
FAc
F2Ac
Mg(OH)2
MgF0.5(OH)1.5
MgF(OH)
MgF1.5(OH)0.5
MgF1.75(OH)0.25
13
34
28
34
30
16
19
16
17
35
34
37
36
33
35
50
44
48
50
27
The ability to reuse the MgF1.5(OH)0.5 sample catalyst was also
investigated. For that, after 6 h of aldol condensation, the spent
catalyst was filtered and dried at 90 °C during a night. The recovered
catalyst was introduced in a new reaction mixture (second run).
Unfortunately, the furfural conversion obtained after 6 hours during
this second run was only close to 10%. This value was thus much
lower than the one obtained during the first run (equal to 84%, Fig.
6) showing a strong deactivation of this kind of catalyst. As an
evidence, further investigations are needed to understand the
deactivation mode that occurs and to investigate experimental
procedures allowing to restore their catalytic properties. Obviously,
this procedure should not include heat treatment at high
temperatures in order to keep the advantages of using these
catalysts without activation step. This is, however, outside the topic
of the present study and hence we will discuss this research topic in
a forthcoming publication.
Fig. 8 Aldol condensation of furfural and acetone over MgF1.5(OH)0.5 at 50 °C under
atmospheric pressure. Influence of the furfural/acetone ratio on the distribution of
products at iso-conversion of furfural (close to 20%)
The influence of the composition of reactants feed on the catalytic
properties of MgF1.5OH0.5 was also investigated. It can be observed
that the molar ratio of reactants for aldol condensation plays a
significant role in both the activity and the selectivity of the catalyst.
Indeed, Fig. S3 (see ESI) shows that the presence of an excess of
acetone favours the consumption of furfural by aldol condensation.
Actually, with the same initial molar quantity of furfural (5 mmol),
the use of a furfural/acetone ratio equal to 0.5 allowed to convert
more than 4 mmol of furfural after 6 hours while the use of a
The effect of the reaction time on the conversion of furfural and on
the yields of all detected products over MgF1.5(OH)0.5 is given in Fig.
S2 (see ESI). As expected, three products were observed, their
quantity depending on the reaction time. Accordingly with the
accepted reaction route, aldol condensation of furfural and acetone
over magnesium hydroxide fluorides results in the successive
formation of FAc-OH (4-(furan-2-yl)-4-hydroxybutan-2-one), FAc (4-
furfural/acetone molar ratio equal to
2 led only to the
transformation of 1.3 mmol of furfural. As an evidence, the
conversion of furfural is limited to 50% in the latter case.
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 11
Catalysis Science & Technology
Please do not adjust margins
Journal Name
ARTICLE
3
4
5
6
7
8
9
H. Kobayashi and A. Fukuoka, Green Chem., 2013, 15, 1740–
Consequently
a
higher furfural/acetone ratio diminishes
View Article Online
1763.
considerably the initial reaction rate (from 10.7 to 2.5 mmol g-1 h-1)
since an excess of furfural may engender unnecessary adsorption on
Lewis acid sites and thereby could inhibit the access of acetone to
the neighbour basic sites. As a result, less active carbanions species
can be formed when an excess of furfural was used and hence
restraining its transformation. However, an excess of furfural led to
a better selectivity on furanic dimer (F2Ac) as highlighted in Fig. 8.
Indeed, when the molar ratio of furfural/acetone was increased from
0.5 to 2, the selectivity for the formation of dimer species (F2Ac)
increased by 38%.
DOI: 10.1039/C9CY01259A
S. Dutta, S. De, B. Saha and M. I. Alam, Catal. Sci. Technol.,
2012, 2, 2025–2036.
M. J. Climent, A. Corma and S. Iborra, Green Chem., 2014, 16,
516–547.
I. Delidovich, K. Leonhard and R. Palkovits, Energy Environ.
Sci., 2014, 7, 2803–2830.
W. Xu, H. Wang, X. Liu, J. Ren, Y. Wang and G. Lu, Chem.
Commun., 2011, 47, 3924–3926.
L. Bui, H. Luo, W. R. Gunther and Y. Román-Leshkov, Angew.
Chem. Int. Ed., 2013, 52, 8022–8025.
J. Wu, G. Gao, J. Li, P. Sun, X. Long and F. Li, Applied Catalysis
B: Environmental, 2017, 203, 227–236.
10 Z. J. Brentzel, K. J. Barnett, K. Huang, C. T. Maravelias, J. A.
Dumesic and G. W. Huber, ChemSusChem, 2017, 10, 1351–
1355.
11 H. Li, A. Riisager, S. Saravanamurugan, A. Pandey, R. S.
Sangwan, S. Yang and R. Luque, ACS Catal., 2018, 8, 148–187.
12 R. M. West, Z. Y. Liu, M. Peter and J. A. Dumesic,
ChemSusChem, 2008, 1, 417–424.
13 R. Xing, A. V. Subrahmanyam, H. Olcay, W. Qi, G. P. van
Walsum, H. Pendse and G. W. Huber, Green Chem., 2010, 12,
1933–1946.
14 J. Yang, N. Li, S. Li, W. Wang, L. Li, A. Wang, X. Wang, Y. Cong
and T. Zhang, Green Chem., 2014, 16, 4879–4884.
15 Q. Xia, Y. Xia, J. Xi, X. Liu and Y. Wang, Green Chem., 2015, 17,
4411–4417.
16 R. Ramos, Z. Tišler, O. Kikhtyanin and D. Kubička, Catal. Sci.
Technol., 2016, 6, 1829–1841.
17 L. Faba, E. Díaz and S. Ordóñez, Applied Catalysis B:
Environmental, 2014, 160–161, 436–444.
18 L. Faba, E. Díaz and S. Ordóñez, Catal. Sci. Technol., 2015, 5,
1473–1484.
19 L. Faba, E. Díaz, A. Vega and S. Ordóñez, Catalysis Today, 2016,
269, 132–139.
20 R. Ramos, Z. Tišler, O. Kikhtyanin and D. Kubička, Applied
Catalysis A: General, 2017, 530, 174–183.
21 N. Fakhfakh, P. Cognet, M. Cabassud, Y. Lucchese and M. D.
de Los Ríos, Chemical Engineering and Processing: Process
Intensification, 2008, 47, 349–362.
22 W. Shen, G. A. Tompsett, K. D. Hammond, R. Xing, F. Dogan,
C. P. Grey, W. C. Conner, S. M. Auerbach and G. W. Huber,
Applied Catalysis A: General, 2011, 392, 57–68.
23 I. Sádaba, M. Ojeda, R. Mariscal, J. L. G. Fierro and M. L.
Granados, Applied Catalysis B: Environmental, 2011, 101,
638–648.
Conclusions
It was demonstrated that magnesium hydroxide fluorides MgF2-
x(OH)x can be used as efficient catalysts without preliminary thermal
activation for aldol condensation of furfural and acetone under mild
conditions (50 °C, under atmospheric pressure). A strong effect of the
F/Mg molar ratio used during the synthesis of fluorides on their
catalytic properties was highlighted. An optimum was determined
for a F/Mg ratio equal to 1.5. This optimum was attributed to the
presence of an intimate mixture of MgF2 and Mg(OH)2 phases
allowing to create active sites combined an under-coordinated
magnesium in the MgF2 phase, acting as Lewis acid site, in the vicinity
of hydroxyl groups with moderate strength in Mg(OH)2-x(OCH3)x
phase. Moreover, it was highlighted that the presence of fluorine
allows to stabilize the basic sites with moderate strength at ambient
temperature, allowing to use these catalysts without pre-treatment
of activation, a great benefit for an industrial point of view. Such
hydroxide fluoride catalyst favored the C13 dimer species,
particularly in excess of furfural. The present work confirms the
potential of nanosized magnesium hydroxide fluorides as
heterogeneous acid-base catalysts for aldol condensation reactions.
The easy tuning of their acid-base properties and their high specific
surface area make them attractive for other reactions involved in
catalytic processes.
24 L. Faba, E. Díaz and S. Ordóñez, Applied Catalysis B:
Environmental, 2012, 113, 201–211.
25 D. Nguyen Thanh, O. Kikhtyanin, R. Ramos, M. Kothari, P.
Ulbrich, T. Munshi and D. Kubička, Catalysis Today, 2016, 277,
97–107.
Conflicts of interest
There are no conflicts to declare.
26 O. Kikhtyanin, V. Kelbichová, D. Vitvarová, M. Kubů and D.
Kubička, Catalysis Today, 2014, 227, 154–162.
27 M. Su, W. Li, T. Zhang, H. Xin, S. Li, W. Fan and L. Ma, Catal.
Sci. Technol., 2017, 7, 3555–3561.
28 H. Li, Z. Xu, P. Yan and Z. C. Zhang, Green Chem., 2017, 19,
1751–1756.
Acknowledgements
Minrui Xu acknowledges a PhD fellowship from Solvay.
Stéphane Célerier, Frédéric Richard and Jean-Marc Clacens
acknowledge financial support from the European Union (ERDF)
and “Region Nouvelle Aquitaine”.
29 O. Kikhtyanin, D. Kubička and J. Čejka, Catalysis Today, 2015,
243, 158–162.
30 S. Rojas-Buzo, P. García-García and A. Corma, Green Chem.,
2018, 20, 3081–3091.
31 S. Célérier and F. Richard, Catalysis Communications, 2015,
67, 26–30.
32 E. Kemnitz, Catal. Sci. Technol., 2015, 5, 786–806.
33 F. Frouri, S. Célérier, P. Ayrault and F. Richard, Applied
Catalysis B: Environmental, 2015, 168, 515–523.
Notes and references
1
A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411–
2502.
2
G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106,
4044–4098.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 10 of 11
ARTICLE
Journal Name
34 S. M. Coman, S. Wuttke, A. Vimont, M. Daturi and E. Kemnitz,
Advanced Synthesis & Catalysis, 2008, 350, 2517–2524.
35 H. A. Prescott, Z.-J. Li, E. Kemnitz, J. Deutsch and H. Lieske, J.
Mater. Chem., 2005, 15, 4616–4628.
View Article Online
DOI: 10.1039/C9CY01259A
36 M. Wojciechowska, A. Wajnert, I. Tomska-Foralewska, M.
Zieliński and B. Czajka, Catal Lett, 2009, 128, 77–82.
37 I. Tomska-Foralewska, M. Zieliński, M. Pietrowski, W.
Przystajko and M. Wojciechowska, Catalysis Today, 2011, 176,
263–266.
38 M. Zieliński, I. Tomska-Foralewska, M. Pietrowski, W.
Przystajko and M. Wojciechowska, Catalysis Today, 2012, 191,
75–78.
39 M. Zieliński, A. Kiderys, M. Pietrowski, I. Tomska-Foralewska
and M. Wojciechowska, Catalysis Communications, 2016, 76,
54–57.
40 S. Célerier, S. Morisset, I. Batonneau-Gener, T. Belin, K.
Younes and C. Batiot-Dupeyrat, Applied Catalysis A: General,
2018, 557, 135–144.
41 G. Scholz, C. Stosiek, M. Feist and E. Kemnitz, Eur. J. Inorg.
Chem., 2012, 2012, 2337–2340.
42 R. E. O’Neill, L. Vanoye, C. De Bellefon and F. Aiouache,
Applied Catalysis B: Environmental, 2014, 144, 46–56.
43 A. A. Rywak, J. M. Burlitch and T. M. Loehr, Chem. Mater.,
1995, 7, 2028–2038.
44 Y. Diao, W. P. Walawender, C. M. Sorensen, K. J. Klabunde and
T. Ricker, Chem. Mater., 2002, 14, 362–368.
45 S. Wuttke, A. Vimont, J.-C. Lavalley, M. Daturi and E. Kemnitz,
J. Phys. Chem. C, 2010, 114, 5113–5120.
46 I. Agirrezabal-Telleria, Y. Guo, F. Hemmann, P. L. Arias and E.
Kemnitz, Catal. Sci. Technol., 2014, 4, 1357–1368.
47 F. Richard, S. Célérier, M. Vilette, J.-D. Comparot and V.
Montouillout, Applied Catalysis B: Environmental, 2014, 152–
153, 241–249.
48 C. Lahousse, A. Aboulayt, F. Maugé, J. Bachelier and J. C.
Lavalley, Journal of Molecular Catalysis, 1993, 84, 283–297.
49 C. Lahousse, F. Maugé, J. Bachelier and J.-C. Lavalley, J. Chem.
Soc., Faraday Trans., 1995, 91, 2907–2912.
50 J. Szanyi and J. H. Kwak, Phys. Chem. Chem. Phys., 2014, 16,
15117–15125.
51 K. I. Hadjiivanov and G. N. Vayssilov, in Advances in Catalysis,
Academic Press, 2002, vol. 47, pp. 307–511.
52 T. Krahl, A. Vimont, G. Eltanany, M. Daturi and E. Kemnitz, J.
Phys. Chem. C, 2007, 111, 18317–18325.
53 A. Astruc, C. Cochon, S. Dessources, S. Célérier and S. Brunet,
Applied Catalysis A: General, 2013, 453, 20–27.
54 S. Wuttke, S. M. Coman, J. Kröhnert, F. C. Jentoft and E.
Kemnitz, Catalysis Today, 2010, 152, 2–10.
55 M. Akimoto and I. G. D. Lana, Journal of Catalysis, 1980, 62,
84–93.
56 H. Hattori, Applied Catalysis A: General, 2015, 504, 103–109.
10 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 11 of 11
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C9CY01259A
Highlight: MgF2-x(OH)x catalysts involving neighboring Lewis acid sites and moderate
strength basic sites were used successfully in aldol condensation of furfural.