Catal Lett (2013) 143:356–363

DOI 10.1007/s10562-013-0970-1

Synthesis of Monoglycerides by Esterification of Oleic Acid

with Glycerol in Heterogeneous Catalytic Process Using
Tin–Organic Framework Catalyst
Lik H. Wee • Tristan Lescouet • Julia Fritsch • Francesca Bonino •
Marcus Rose • Zhijun Sui • Eva Garrier • Dirk Packet • Silvia Bordiga •
Stefan Kaskel • Moti Herskowitz • David Farrusseng • Johan A. Martens

Received: 23 November 2012 / Accepted: 26 January 2013 / Published online: 13 February 2013
Ó Springer Science+Business Media New York 2013

Abstract Selective synthesis of monoglycerides by frameworks and, especially, tin–organic frameworks were
esterification of glycerol with fatty acids is a difficult found promising. A tin–organic framework (Sn–EOF) was
reaction because of immiscibility of reagents and the for- most active and achieved C98 % monoglyceride selectivity
mation of di- and tri-glyceride by-products. In this work a at 40 % conversion in catalyzing esterification of oleic acid
heterogeneous catalytic process was conceived in which with glycerol at a low reaction temperature of 150 °C.
the reactant mixture was homogenized using tert-butanol Leaching of tin from Sn–EOF catalyst was suppressed by
solvent. Candidate catalysts were screened in the reaction limiting the amount of oleic acid in the starting mixture.
of oleic acid with glycerol. While under such reaction Characterization of the acid sites of Sn–EOF by pyridine-
conditions zeolites were rather inactive, metal–organic chemisorption and FTIR revealed Lewis acidity to be
responsible for the catalytic activity.

L. H. Wee  J. A. Martens (&) Keywords Tin–organic framework  Esterification 

Centre for Surface Chemistry and Catalysis, KU Leuven,
Oleic acid  tert-Butanol  Monoglycerides
Kasteelpark Arenberg 23, Heverlee B3001, Belgium
e-mail: [email protected]

T. Lescouet  D. Farrusseng 1 Introduction

Institut de Recherche sur la Catalyse et l’Environnement
de Lyon (IRCELYON), CNRSC, University Lyon 1,
2 Av. Albert Einstein, 69626 Villeurbanne, France Monoglycerides are renewable surfactant molecules find-
ing application in food and pharmaceutical formulations,
J. Fritsch  M. Rose  S. Kaskel cosmetics and detergents [1]. The synthesis of monogly-
Department for Inorganic Chemistry, Dresden University
cerides on industrial scale is achieved mainly via two
of Technology, 01062 Dresden, Germany
routes: (1) direct esterification of glycerol with fatty acid
F. Bonino  S. Bordiga [2] and (2) transesterification of tri-glycerides with glyc-
Department of Chemistry, NIS Centre of Excellence, erol, both catalyzed by homogeneous strong acid or base
University of Turin, Via P. Giuria 7, 10125 Turino, Italy
catalysts such as H2SO4, H3PO4, KOH, NaOH, or
F. Bonino  S. Bordiga Ca(OH)2. The reaction typically is carried out at relatively
Department of Chemistry, INSTM Reference Center, high temperatures of 220–250 °C. These conventional
University of Turin, Via Quarello 11, 10135 Turino, Italy processes involving the use of homogeneous catalysts
typically lead to low monoglycerides selectivity (40–50 %)
Z. Sui  M. Herskowitz
Blechner Center for Industrial Catalysis and Process due to di- and tri-glyceride side products formation and
Development, Ben-Gurion University of the Negev, undesired formation of soap. For monoglyceride synthesis,
Beer-Sheva, Israel dissolved transition metal complexes with Lewis acid
properties based on Sn are highly active and more selective
E. Garrier  D. Packet
Novance, Rue les rives de l’Oise, Venette, [3], but the elimination of the transition metal from the
B.P. 20609, 60206 Compiegne Cedex, France monoglyceride product is problematic. In addition to

123
Synthesis of Monoglycerides 357

chemical contamination by the catalyst, for human con- vast range of catalytic applications with variable success
sumption such as in food and pharmaceutical products, [18–21]. MOFs show great promise for esterification
limitation of by-product formation is essential. High-tem- reactions involving short alcohols with acetic acid and fatty
perature processes also bear the risk of deterioration of acids performed in monophasic liquid phase [17, 22, 23],
taste, aroma and color. In addition, techniques for separa- but to our knowledge the potential of MOFs for mono-
tion of monoglycerides from di- and tri-glycerides via glycerides synthesis has not yet been explored. Here, we
distillation are technically complicated, and energy con- report a convenient reaction procedure enabling screening
suming. For all these reasons the development of an effi- of MOFs for monoglycerides synthesis through esterifica-
cient monoglyceride production process taking advantage tion of oleic acid with glycerol. An amorphous tin based
of heterogeneous catalysis under mild reaction conditions organic framework with pore size of ca. 1 nm enabled
with convenient catalyst recovery and recycling is a sci- synthesis of monoglycerides with high selectivity and
entific challenge of practical interest. For the targeted minimum leaching of tin into the reaction product.
reaction several types of heterogeneous catalysts have been
investigated in literature [4–13]. For example, 3-mercap-
topropyltrimethoxysilane functionalized-MCM-41 has 2 Experimental
been tested for the esterification of glycerol with lauric or
oleic acid [4, 5]. Oleic acid conversion up to 89 % was 2.1 Preparation of MOFs
achieved with 40 % monoglycerides selectivity at 120 °C
after 8 h. The use of short chain fatty acid (lauric acid) Cu3(BTC)2 encapsulated Keggin phosphotungstic acid
resulted in higher monoglyceride selectivity (59 % at 85 % (HPW) or silicotungstic acid (HSiW) denoted as HPW/
lauric acid conversion) [5]. Corma et al. [9] investigated Cu3(BTC)2 or HSiW/Cu3(BTC)2 (BTC benzene-1,3,5-tri-
the glycerolysis of the triolein or rapeseed oil with glycerol carboxylic acid), respectively, was prepared from 12 g of
over solid basic catalysts such as sepiolite-Cs, MCM-41- copper(II) nitrate trihydrate (Cu(NO3)23H2O) (99–104 %,
Cs, MgO and hydrotalcites. MgO was found to be the Sigma-Aldrich) and 8.0 g of HPW (H3PW12O40nH2O) (for
better catalyst achieving 97 % triolein conversion and microscopy, Fluka) dissolving in 100 ml of distilled water.
75 % monoglyceride selectivity after 5 h reaction at 5.8 g of BTC was dissolved in 140 ml of absolute ethanol
240 °C. Zeolites with different framework topology cata- (BDH). Both the synthesis solutions were mixed under
lyzed esterification of lauric acid with glycerol [10]. vigorous stirring for 5 min and transferred into a 1 l round
Monolaurate selectivity up to 69 % at 50 % lauric acid bottom flask and heated to 110 °C for 24 h under reflux.
conversion was reached using an acid Y type zeolite at The solid product was washed with distilled water: ethanol
112 °C [11]. Generally the monoglyceride selectivity of the mixture (1:1, v/v), recovered by filtration and dried in an
investigated heterogeneous catalysts were not significantly oven at 60 °C. Nanosized HPW/Cu3(BTC)2 was synthe-
better than in the homogeneous catalytic processes. The sized via the freeze drying approach as previously reported
development of a heterogeneous catalytic process for [22]. The synthesis of Sn–EOF was performed according to
monoglyceride synthesis is complicated by the immisci- the previous reported method by Fritsch et al. [24] 7.8 g
bility of glycerol and fatty acid, and the autocatalytic (25 mmol) of 4,40 -dibromobiphenyl (99 %, Acros Organ-
activity of the fatty acid [4]. ics) was dissolved in 500 ml of dry tetrahydrofuran (THF)
In the 1990s, a new class of nanoporous materials (VWR) under argon atmosphere. After cooling the solution
known as metal–organic frameworks (MOFs) emerged to - 78 °C, 20 ml (50 mmol, 2.5 M, Sigma-Aldrich) of
[14]. MOFs are organic–inorganic hybrid materials con- n-butyllithium (CAUTION: n-butyllithium is explosive
structed from organic linkers of metal centers. MOFs offer when exposed to air and water) was added slowly under
outstanding physical properties such as high surface area, vigorous stirring with controlling the temperature via a
tunable porosity, and functionality [15]. An ever increasing syringe pump (40 ml/h). Afterwards, 3.26 g of
number of MOFs containing Al, Mg, Mn, Fe, Ti, Zn, Cr or (12.5 mmol, 1.44 ml) SnCl4 (98 %, Fisher Scientific) was
Zr with potential Lewis type acidity are being reported. added dropwise under vigorous stirring. The mixture was
The metal atoms in the framework can serve as active slightly warmed to room temperature and stirring under
Lewis acid catalytic centers. Moreover, the organic linkers argon was continued overnight. The resulting precipitate
of MOF structures are amenable and can be foreseen with was filtered and washed twice with THF, distilled water
amine or amide basic catalytic functions. MOFs can and ethanol. After drying under vacuum at room temper-
encapsulate guest species with catalytic activity such as ature the product was obtained as light beige powder. The
polyoxometalates or porphyrins [16] and such species can specific surface area of Sn–EOF was determined using
be released and recovered again by encapsulation in the nitrogen physisorption. The sample was outgassed at 80 °C
MOF after reaction [17]. MOFs have been evaluated for a overnight under vacuum. The BET surface area was

123
358 L. H. Wee et al.

consistent with earlier work [24]. Al-MIL-53-NH2 was inlet tube. The reaction mixture was stirred at 400 rpm.
prepared and purified according to the synthesis protocols Conversion of oleic acid and selectivity of monoglyceride
introduced by Stock et al. [25]. Suspension containing were determined using gas chromatography (GC). GC
stoichiometric amount of aluminum chloride hexahydrate analyses were performed on a high-temperature apolar
(965.7 mg, 4 mmol) and 2-aminobenzene-1,4-dicarboxyl- column DB5 (Agilent J&W, 30 m 9 0.53 mm, 1.50 lm)
ate (ABDC) (724.6 mg, 4 mmol) in 20 ml deionized H2O with an HP-6890 instrument equipped with flame ioniza-
was introduced in a 23 ml Teflon-lined autoclave. The tion detector (FID). The content of leached Sn was deter-
mixture was heated at 150 °C for 9 h. After cooling the mined by inductively coupled plasma optical emission
product was separated from water by centrifugation. In spectroscopy (ICP-OES) (VARIAN 725 S instrument).
order to remove the incorporated ABDC, the as-synthe-
sized product was placed in an autoclave with anhydrous 2.3 Characterization of Acid Sites by Pyridine
dimethylformamide (DMF) and heated at 150 °C for a day. Adsorption Monitored by Fourier Transform
Then the solvent was substituted by fresh DMF and the Infrared (FTIR) Spectroscopy
solvothermal treatment was repeated twice. Afterwards the
DMF was evacuated by Soxhlet-extraction using dichlo- FTIR spectra were collected on a Vertex 70 Bruker spec-
romethane (DCM) for 20 h. Finally, solid was dried under trophotometer equipped with a MCT detector, at 2 cm-1
reduced pressure for 9 h to obtain activated, fine powdered resolution, on a thin self-supported wafer. The sample was
Al-MIL-53-NH2. In-MIL-68-NH2 was synthesized and activated in a home-made cell under high vacuum (residual
activated following the published procedures [26, 27]. pressure \10-4 mbar) at 120 °C for one night. At RT
Indium nitrate (1,203.3 mg, 4 mmol) and ABDC pyridine vapor was dosed on the sample and then pro-
(724.6 mg, 4 mmol) were dissolved in DMF (30 ml). The gressively outgassed.
reaction mixture was stirred for 5 min, and then a solution
of 4-diazabicyclo[2.2.2] octane (DABCO) (897.4 mg,
8 mmol) in DMF (10 ml) was added. The reaction mixture 3 Results and Discussion
was stirred for 120 min at room temperature. The obtained
precipitate was washed with DMF in an autoclave at 3.1 Esterification of Glycerol with Oleic Acid
150 °C overnight followed by Soxhlet extraction with in Absence of Solvent
DCM for 24 h. UiO-66-NH2 and UiO-66 were synthesized
by solvothermal method in a Teflon-lined autoclave [28]. Oleic acid and glycerol are immiscible liquids. The
ZrCl4 (600 mg), ABDC or benzene-1,4- dicarboxylic acid monoglyceride reaction product acts as an emulsifier and,
(BDC) (480 mg) and water (50 ll) were added to 40 ml of upon stirring, the two liquid layers become an emulsion
DMF. The mixture was heated at 120 °C for 6 h. The and after sufficient conversion, a homogeneous liquid
obtained precipitate was washed with DMF in an autoclave phase was obtained. The reaction of an equimolar mixture
at 150 °C overnight followed by a DCM Soxhlet for 24 h. of oleic acid and glycerol was run first in absence of cat-
Zinc triazolate with amine substituants ZnF(TAZ-NH2) alyst in a closed glass vessel at 120 °C (blank, Table 1).
was synthesized by solvothermal method in a Teflon-lined After 3 h, 30 % oleic acid conversion was reached, and
autoclave [29]. ZnF24H2O (1.05 g) and 3,5-diamino-1,2, 75 % monoglyceride selectivity. After 8 h the conversion
4-triazole (0.6 g) were added to 30 ml of water. The mix- was increased to 45 % while the monoglyceride selectivity
ture was heated at 160 °C for 3 days. The obtained pre- was decreased to 63 %. Diglycerides and triglyceride were
cipitate was washed with water and dried at 90 °C the observed as side products. The esterification reaction
overnight. Ultrastable Y zeolite (CBV-720) was supplied by under these conditions is catalyzed by the oleic acid itself.
Zeolyst International and used without further treatment. Similar autocatalytic effects have been previously reported
[4].
2.2 Esterification Reactions All the MOFs used in this studies were synthesized
according to the previously reported methods and the
The screening of MOF catalysts for esterification of oleic synthesis procedures were presented in details in the
acid with glycerol was performed using glass vials and experimental section. Phase purity, crystallinity and
mini stainless steel autoclaves (both 12 ml) magnetically microporosity of the MOFs were verified and confirmed by
stirred at 1,000 rpm and heated in a temperature controlled XRD and N2 adsorption (not shown). The esterification
copper block. Alternatively the esterification reaction was reaction in absence of solvent was performed using HPW/
carried out using a reflux system consisting of a 1 l four- Cu3(BTC)2 catalyst, a copper benzene tricarboxylate MOF
necked flask equipped with Teflon shaft stirrer, a thermo- having encapsulated Keggin HPW in the pores of HKUST-
couple, air and water condensers and an inert gas (nitrogen) 1 framework structure [30]. HKUST-1 framework has two

123
Synthesis of Monoglycerides 359

Table 1 Esterification of oleic

Catalysts Catalyst loading Reaction Oleic acid Selectivity (%)
acid with glycerol in the
(wt%) time (h) conversion (%)
absence of solvent Mono Di Tri

Blank – 3 30 75 24 1
– 8 45 63 34 3
HPW/Cu3(BTC)2 1 3 30 77 22 1
1 8 45 62 35 3
2 3 28 76 23 1
5 3 31 75 24 1
Ultrastable Y zeolite 1 3 33 78 21 1
Molar ratio of oleic acid: 1 8 45 63 34 3
glycerol = 1:1, 120 C, copper 2 3 32 77 22 1
block reactor. Catalyst loading 5 3 29 76 23 1
was based on added oleic acid

types of pores with free diameters of 1.3 and 1.0 nm, reaction conditions, a catalytic effect was superimposed on
respectively, connected through accessible windows of 1.1 an autocatalytic contribution. HPW converted the glycerol
and 0.9 nm. The largest cage accommodate the Keggin into polyglycerol and produced little monoglyceride.
anions [30]. HPW is well known for its strong Brønsted While polyglycerol formation was also observed when
acidity and HPW/Cu3(BTC)2 MOF has been reported as using HWP encapsulated in nanopowder of Cu3(BTC)2, it
active heterogeneous catalyst for ester hydrolysis [30], and was absent on larger particles of this catalyst formulation.
esterification of acetic acid with 1-propanol [23, 31]. Ul- Visual inspection revealed a color change indicating that
trastable Y zeolite was used as a referece. It has micropores the MOF catalysts were degraded after 3 h at 180 °C
with 0.7 nm windows and a secondary mesoporosity that (photographs of Fig. 1). It was clear such reaction condi-
should be wide enough to accommodate the reagent and tions are less suitable for Cu3(BTC)2 type catalysts. The
product molecules [32]. Ultrastable Y zeolite has been highest conversion of oleic acid (100 %) was obtained
reported to be superior among other zeolite catalysts in the using Sn–EOF catalyst. Sn–EOF is an elemental-organic
investigated reaction [33]. framework (EOF) material constructed from Sn4? ions
Esterification of oleic acid with glycerol in 1:1 molar connected to 4,40 -dibromobiphenyl organic linker via
ratio was performed in closed glass vessels at 120 °C using element-carbon bondings. Sn–EOF is an amorphous
catalyst contents of 1, 3 and 5 wt% based on oleic acid, and microporous material with a BET surface area of
reaction times of 3 or 8 h (Table 1). The catalytic activity 445 m2 g-1 and pore size around 1 nm [24]. It is ther-
of HPW/Cu3(BTC)2 and ultrastable Y zeolite in this reac- mally stable up to 200 °C in air [24]. A full character-
tion was negligible since the conversion obtained in pres- ization of the material could be obtained from ref. 24. The
ence and in absence of these catalysts was similar monoglyceride selectivity in these experiments at high
(Table 1). Increasing the catalyst content from 1 to 5 wt% conversion was below 50 %. Such experimental setup with
did not improve conversion. The autocatalytic effect of the continuous elimination of water by evaporation appeared
oleic acid reagent was responsible for the conversion. unsuited.
In an alternative experimental setup water produced by
the esterification reaction was evacuated. The reflux 3.2 Esterification of Glycerol with Oleic Acid
reactor consisted of a glass stirring shaft with a Teflon in Presence of Solvent
stirring paddle for efficient mixing of the reaction mixture
and a condenser connected to the reactor for collecting When esterification of oleic acid with glycerol is performed
water vapor formed during reaction. The esterification at high reaction temperature (200–260 °C) and pressure
reaction of oleic acid with glycerol was carried out at (30–100 psig) a monophasic reaction phase is obtained
180 °C for 3 h departing from an equimolar mixture of with water addition [34]. Such conditions likely are not
oleic acid and glycerol. The tested MOF catalysts were suitable for MOFs. Alternatively, for homogenizing the
nanosized and micron-sized HPW/Cu3(BTC)2, HSiW/ starting mixture a solvent can be used. The selection of an
Cu3(BTC)2, homogeneous HPW, Sn–EOF and ultrastable appropriate solvent was not obvious, however. The solvent
Y zeolite. The results (Table 2) show that even without should have high boiling point to limit vaporization, and
catalyst a high oleic acid conversion (89 %) was obtained. provide high solubility of both oleic acid and glycerol. Five
In presence of the majority of tested catalysts the oleic candidate solvents emerged from our search: diglyme,
acid conversion was further enhanced. Under these dimethylsulfoxide (DMSO), DMF, monoolein and

123
360 L. H. Wee et al.

Table 2 Esterification of oleic

Catalysts Reaction Oleic acid Selectivity (%)
acid with glycerol with water
time (h) conversion
evacuation in reflux reactor Mono Di Tri Polyglycerol
(%)

Blank 3 89 51 40 9
HPW 2 95 2 12 13 73
Ultrastable Y zeolite 3 92 47 43 10
HPW/Cu3(BTC)2 3 97 39 49 12
Nano HPW/ 3 99 21 35 9 35
Molar ratio of oleic acid: Cu3(BTC)2
glycerol = 1:1, 180\C, reflux HSiW/Cu3(BTC)2 3 88 50 43 7
reactor. 1 wt% catalyst loading Sn–EOF 3 100 48 42 10
based on added oleic acid

were found less active for catalyzing esterification of oleic

acid with glycerol. Sn–EOF was a superior catalyst
achieving 26 and 40 % conversion after 8 and 20 h,
respectively, which was higher than in the absence of
catalyst (16 and 23 % conversion, respectively). The
monoglyceride selectivity exceeded 98 %. A drawback
was that Sn–EOF catalyst was partially dissolved. To trace
the origin of the solubility, Sn–EOF catalyst was suspended
in glycerol, monoglyceride, oleic acid or tert-butanol and
the suspension heated. Oleic acid was found responsible for
degradation of the Sn–EOF catalysts upon heating above
Fig. 1 Photograph showing initial biphasic reaction mixture becomes
monophasic. HPW/Cu3(BTC)2 degradation gives rise to colour 150 °C. Thus, subsequent esterification reactions were
change from greenish to brownish conducted using a lower concentration of oleic acid.
Experiments were run with 1, 0.4, 0.3 and 0.2 mol oleic
acid per mol glycerol in the starting mixture. The stability
tert-butanol. Except for diglyme both oleic acid and glyc- of Sn–EOF catalyst was improved at reduced concentration
erol were soluble in these solvents. of oleic acid as evidenced from visual inspection of the
The esterification reaction was performed using DMSO, amount of catalyst that precipitated after reaction (Fig. 2).
DMF, and monoolein solvent. The solvent was added in a A further esterification experiment was performed at
weight proportion of solvent to reaction mixture of 2:1. Sn– 150 °C for 20 h using tert-butanol solvent and an oleic acid
EOF was chosen for further investigation as its catalytic to glycerol molar ratio of 0.1:1 in the starting mixture
activity was already superior in absence of solvent (Fig. 3). The oleic acid conversion reached up to 40 and
(Table 2). In the presence of DMSO and DMF solvent, 98 % monoglyceride selectivity was achieved. High
high oleic acid conversion (70–80 %) together with higher selectivity clearly is reached with the presence of a
selectivity for monoglyceride ([90 %) were reached. homogeneous liquid phase, independent of catalyst. Under
Furthermore, conversion and selectivity in presence and similar reaction conditions, the selectivity of monoolein
absence of catalyst were similar, showing that the auto- achieved was limited at 60 % at 40 % oleic acid conver-
catalytic effect predominated. The same observation was sion in the absence of tert-butanol. The Sn–EOF catalyst
made with ultrastable Y zeolite catalyst. Furthermore, Sn– could be recovered and recycled. However, a gradual loss
EOF catalyst was unstable in these solvents, leading to of the catalytic activity was noted in the second run, where
dissolution at the end of the experiment. The use of 30 % oleic acid conversion and [90 % monoglyceride
monoolein solvent was also investigated. Since monoolein selectivity were obtained. Chemical analysis of the reaction
is a monoglyceride, the consecutive reaction to diglyce- product from which Sn–EOF catalyst was removed through
rides was enhanced, which is undesired. centrifugation revealed a residual tin concentration of
In another series of experiments, tert-butanol solvent 700 ppm. Although the solubility of the Sn–EOF catalyst
was used (Table 3). MOF candidate catalysts grafted with was significantly suppressed, some leaching could not be
amino functions (Al-MIL-53-NH2, In-MIL-68-NH2, UiO- avoided. The presence of solvent is a handicap if the
66-NH2 and ZnF (TAZ-NH2), UiO-66 in addition to Sn– monoglyceride products were to be used in food or phar-
EOF were investigated. These esterification reactions were maceutical application. Given the large different in vola-
performed at 150 °C for 20 h. MOFs with amino group tility, the monoglyceride could be easily purified by

123
Synthesis of Monoglycerides 361

Table 3 Esterification of oleic acid with glycerol in tert-butanol

solvent
Catalysts Reaction Oleic acid Selectivity to
time (h) conversion (%) Mono (%)

Blank 8 16 100
Blank 20 23 100
Sn–EOF 8 26 99
Sn–EOF 20 40 98
Al-MIL-53-NH2 8 11 100
Al-MIL-53-NH2 20 23 100
In-MIL-68-NH2 8 14 100
In-MIL-68-NH2 20 22 99
UiO-66 8 17 100 Fig. 3 Esterification of oleic acid with glycerol catalyzed by Sn–
EOF. Reaction conditions: temperature of 150 °C, starting mixture
UiO-66 20 29 99 composed of 0.175 g oleic acid, 0.57 g glycerol, 4.64 g tert-butanol
UiO-66-NH2 20 18 100 and 0.2 g Sn–EOF catalyst
ZnF(TAZ-Nh2) 20 25 100
Molar ratio of oleic acid: glycerol = 1:1, 150 °C, stainless steel
autoclave reactor. Total reaction mixture: solvent = 1:2(w/w). 1 wt% SnEOF activated
catalyst loading based on added oleic acid pyridine vp
pumping

1.0
Absorbance (a.u.)

8b:hb,ph
8a:ph 19b:Lewis 19b:hb,ph
? 19a
8a:hb

8a:Lewis

3500 3000 1600 1550 1500 1450 1400

-1
Wavenumber (cm )

Fig. 4 FTIR spectra of Sn–EOF catalyst after pyridine vapor

adsorption (blue curve, upper part); Sn–EOF activated in vacuo at
120 °C for 1 night (black curve, upper part). Bottom part reports
Fig. 2 Photographs of reaction product with sedimented Sn–EOF difference spectra (in respect to the activated sample) obtained at
catalyst obtained using different starting glycerol:oleic acid molar decreasing coverages (black curves) in respect to the highest one
ratio’s indicated on the vials. The increasing amount of catalyst (blue curve)
precipitate and reduction of coloration from left to right reveal the
decrease of catalyst solubility at lower oleic acid concentration.
Reaction conditions: tert-butanol solvent; reaction temperature of difference spectra (bottom part of Fig. 4) showing positive
150 °C; reaction time of 24 h, 4.64 g of tert-butanol added to 0.57 g bands due to species formed upon pyridine adsorption, and
of glycerol, 0.2 g of Sn–EOF negative bands due to consumed species. Bands in the
3,100–3,000 cm-1 range were ascribed to pyridine m(C–H)
distillation. There exist applications where the presence of stretching modes, while the negative band at 3,580 cm-1 was
residual solvent is less critical. due to the consumption of some Sn–OH species, assumed to
be present in defects. In the 1,625–1,400 cm-1 range based
3.3 In Situ FTIR Spectroscopy Analysis by Pyridine on literature on pyridine adsorption on different materials
Adsorption [35–39], all the emerging bands were attributed to pyridine
interacting with a Lewis site, hydrogen-bonded (hb) or phy-
The acid sites in Sn–EOF were characterized by FTIR and sorbed (ph). In particular the assignment can be done in the
pyridine adsorption (Fig. 4). Sn–EOF itself has some vibra- following way: 1610 8a Lewis, 1597 8a hb, 1580 8a ph or 8b
tional modes overlapping with pyridine (e.g. ring modes) hb, 1573 8b ph, 1480 19a ph or hb (probable artifact due to
[24]. The spectral features ascribed to pyridine appear in the subtraction hb (probable artifact due to subtraction marked

123
362 L. H. Wee et al.

with a question mark), 1450 19b Lewis, 1443 19b hb, 1437 4. Bossaert WD, De Vos DE, Van Rhijn WM, Bullen J, Grobet PJ,
19b ph. It is evident that in some cases there is the superpo- Jacobs PA (1999) J Catal 182:156
5. Dı́az I, Mohino F, Pérez-Pariente J, Sastre E (2003) Appl Catal A
sition of some components due to the frequency closeness of 242:161
the different species. In particular, for instance, for some 6. Pérez-Pariente J, Dı́az I, Mohino F, Sastre E (2003) Appl Catal A
modes the perturbation induced by Lewis sites is definitely 254:173
smaller and these bands are, therefore, superimposed to those 7. Dı́az I, Mohino F, Pérez-Pariente J, Sastre E (2001) Appl Catal A
205:19
of hydrogen-bonded species. The identification of Lewis 8. Dı́az I, Márquez-Alvarez C, Mohino F, Pérez-Pariente J, Sastre E
moieties is supported by comparison with IR experiments of (2000) J Catal 193:295
pyridine dosed on c-Al2O3 (a solid characterized by strong 9. Corma A, Iborra S, Miquel S, Primo J (1998) J Catal 173:315
Lewis acid sites), where four bands, respectively at 1620 (8a), 10. da Machadoa MS, Pérez-Pariente J, Sastre E, Cardoso D, de
Guereñu AM (2000) Appl Catal A 203:321
1578 (8b), 1492 (19a), and 1450 (19b) cm-1, are observed 11. Heykants E, Verrelst WH, Parton RF, Jacobs PA (1997) Progress
[38, 39]. Sn-beta zeolite is known as a Sn-based Lewis acid, in zeolite and microporous materials. In: Chon H, Ihm S-K,
catalyzing a variety of green chemical reactions [40, 41]. We Uh YS (eds) Studies in surface science and catalysis. Elsevier,
tested a reference Sn-beta zeolite under the same reaction Amsterdam, p 1277
12. Abro S, Pouilloux Y, Barrault J (1997) Heterogeneous catalysis
conditions as Sn–EOF and observed a black colouring. The and fine chemicals IV. In: Blaser HU, Baiker A, Prins R (eds)
Sn-beta zeolite was inactive with an oleic acid conversion of Studies in surface science and catalysis. Elsevier, Basel, p 539
4 % only after 20 h, even below the conversion in absence of 13. Pouilloux Y, Abro S, Vanhove C, Barrault J (1999) J Mol Catal
catalyst (20 %). A: Chem 149:243
14. Férey G (2008) Chem Soc Rev 37:191
15. Farrusseng D, Aguado S, Pinel C (2009) Angew Chem Int Ed
48:7502
4 Conclusions 16. Wee LH, Alaerts L, Martens JA, de Vos D (2010) Metal–organic
framework: applications from catalysis to gas storage. In: Far-
russeng D (ed) Metal–organic frameworks as catalysts for organic
Selective esterification of oleic acid with glycerol to reactions. Wiley-Vch, Weinheim, p 191
monoglyceride is a demanding reaction. The catalytic 17. Janssens N, Wee LH, Bajpe S, Breynaert E, Kirschhock CEA,
activity of MOFs including HPW/Cu3(BTC)2, HSiW/ Martens JA (2012) Chem Sci 3:1847
Cu3(BTC)2, Al-MIL-53-NH2, In-MIL-68-NH2, UiO-66, 18. Corma A, Garcı́a H, Llabrés i Xamena FX (2010) Chem Rev
110:4606
UiO-66-NH2 and ZnF(TAZ-NH2) hardly exceeds the 19. Dhakshinamoorthy A, Alvaro M, Garcia H (2012) Chem Com-
autocatalytic activity of the fatty acid. On top of limited mun 48:11275
activity, there is a MOF stability issue, oleic acid being 20. Yoon M, Srirambalaji R, Kim K (2012) Chem Rev 112:1196
responsible for the degradation of MOFs. Solvents in 21. Juan-Alcañiz J, Gascon J, Kapteijn F (2012) J Mater Chem
22:10102
which both glycerol and oleic acid can be dissolved in high 22. Wee LH, Bajpe SR, Janssens N, Kirschhock CEA, Martens JA
quantity were identified: tert-butanol, DMF, DMSO and- (2010) Chem Commun 46:8186
monoolein. tert-Butanol was identified as the optimum 23. Juan-Alcañiz J, Ramos-Fernandez EV, Lafont U, Gascon J,
solvent. In this solvent MOF catalysts and especially Sn– Kapteijn F (2010) J Catal 269:229
24. Fritsch J, Rose M, Wollmann P, Böhlmann W, Kaskel S (2010)
EOF are catalytically active. In a catalytic experiment Materials 3:2447
using Sn–EOF catalyst departing from a mixture with 25. Ahnfeldt T, Gunzelmann D, Loiseau T, Hirsemann D, Senker JR,
molar ratio of oleic acid to glycerol of 0.1:1 in tert-butanol Férey G, Stock N (2009) Inorg Chem 48:3057
solvent, Sn–EOF was most effective and showed limited 26. Savonnet M, Bazer-Bachi D, Bats N, Perez-Pellitero J, Jeanneau E,
Lecocq V, Pinel C, Farrusseng D (2010) J Am Chem Soc 132:4518
tin leaching. The use of tert-butanol solvent and reaction 27. Savonnet M, Kockrick E, Camarata A, Bazer-Bachi D, Bats N,
conditions with low concentration of oleic acid are rec- Lecocq V, Pinel C, Farrusseng D (2011) New J Chem 35:1892
ommended for further MOF catalyst screening programs. 28. Silva AG, Luz I, Llabrés i Xamena FX, Corma A, Garcı́a H
(2010) Chem Eur J 16:11133
Acknowledgments The authors gratefully acknowledge financial 29. Savonnet M, Aguado S, Ravon U, Bazer-Bachi D, Lecocq V,
support from the EU FP-7 Framework NANOMOF project under Bats N, Pinel C, Farrusseng D (2009) Green Chem 11:1729
grant agreement CP-IP 228604-2. JAM acknowledges the Flemish 30. Sun A-Y, Liu S-Y, Liang D–D, Shao K-Z, Ren Y-H, Su Z-M
government for long-term structural funding (Methusalem). (2009) J Am Chem Soc 131:1883
31. Wee LH, Janssens N, Bajpe SR, Kirschhock CEA, Martens JA
(2011) Catal Today 171:275
32. Teyssier L, Thomas M, Bouchy C, Martens JA, Guillon E (2007)
References Micropor Mesopor Mater 100:6
33. Aracil J, Martinez M, Sanchez N, Corma A (1992) Zeolites
1. Falbe J (1987) Surfactants in consumer products: theory, tech- 12:233
nology and applications. Springer, New York 34. Logan, RL (2010) US Patent 0,148,117
2. Hartman L (1958) Chem Rev 58:845 35. Kline AH, Turkevich J (1944) J Chem Phys 12:300
3. Guner FS, Sirkecioglu A, Yilmaz S, Erciyes AT, Erdem-Senat- 36. Buzzoni R, Bordiga S, Ricchiardi G, Lamberti C, Zecchina A,
alar A (1996) J Am Oil Chem Soc 73:347 Bellussi G (1966) Langmuir 12:930

123
Synthesis of Monoglycerides 363

37. Bonino F, Damin A, Bordiga S, Lamberti C, Zecchina A (2003) 40. Corma A, Nemeth LT, Renz M, Valencia S (2001) Nature
Langmuir 19:2155 412:423
38. Morterra C, Magnacca G (1996) Catal Today 27:497 41. Moliner M, Román-Leshkov Y, Davis ME (2010) Proc Natl Acad
39. Yamamoto T, Tanaka T, Matsuyama T, Funabiki T, Yoshida S Sci 107:6164
(2001) J Phys Chem B 105:1908

123