Catalytic acetylation of glycerol with acetic anhydride

Article abstract of DOI:10.1016/j.catcom.2010.05.007

We studied the acetylation of glycerol with acetic anhydride using different solid acid catalysts. The results indicated that at 60 °C, zeolite Beta and K-10 Montmorillonite showed 100% selectivity to triacetin within 20 min, with a molar ratio of 4:1. Amberlyst-15 acid resin yielded 100% triacetin after 80 min, whereas niobium phosphate gave diacetin and triacetin in 53% and 47% selectivity, respectively. All catalysts were more selective to triacetin than the uncatalyzed reaction. By contrast, zeolite Beta gave poor yield of triacetin when acetic acid was used as acetylating agent. The different behavior was explained in terms of the stabilization of the acylium ion intermediate.

Full text of DOI:10.1016/j.catcom.2010.05.007

Catalysis Communications 11 (2010) 1036–1039
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Catalytic acetylation of glycerol with acetic anhydride
Leonardo N. Silva, Valter L.C. Gonçalves, Claudio J.A. Mota ⁎
Universidade Federal do Rio de Janeiro, Instituto de Química. Av Athos da Silveira Ramos 149, CT Bloco A, 21941-909, Rio de Janeiro, Brazil
INCT de Energia e Ambiente, UFRJ, 21941-909, RJ, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
We studied the acetylation of glycerol with acetic anhydride using different solid acid catalysts. The results
indicated that at 60 °C, zeolite Beta and K-10 Montmorillonite showed 100% selectivity to triacetin within
Received 16 March 2010
Received in revised form 4 May 2010
Accepted 7 May 2010
20 min, with a molar ratio of 4:1. Amberlyst-15 acid resin yielded 100% triacetin after 80 min, whereas
niobium phosphate gave diacetin and triacetin in 53% and 47% selectivity, respectively. All catalysts were
more selective to triacetin than the uncatalyzed reaction. By contrast, zeolite Beta gave poor yield of triacetin
when acetic acid was used as acetylating agent. The different behavior was explained in terms of the
stabilization of the acylium ion intermediate.
Available online 16 May 2010
Keywords:
Glycerol
Triacetin
© 2010 Elsevier B.V. All rights reserved.
Zeolites
Acid catalysis
1
. Introduction
Glycerol is a good feedstock for other chemicals. Hydrogenolysis to
1
,2 and 1,3-propanediol over metal catalysts [9–11] has been
Glycerol is a byproduct of transesterification of vegetable oil to
extensively studied. Dehydration to acrolein, a useful intermediate
for plastic production, has been reported to occur over solid acid
catalysts [12–14]. Anaerobic fermentation of glycerol is a potential
route for the production of butanol and ethanol [15], which can be
used as biofuels. Glycerol reforming to synthesis gas, a mixture of
carbon monoxide and hydrogen, has been studied over noble metal
catalysts [16], and could be a possible pathway for the production of
hydrocarbons in the diesel and gasoline range. Glycerol acetates are
also good candidates to drain part of the glycerol produced by the
biodiesel industry. Traditionally, this chemical is used as a plasticizer
in cigarette filters, but in recent years it has been tested as a biodiesel
additive [17,18], improving the cold flow properties. This application
is extremely interesting, because the glycerol produced from
transesterification could be used in the biodiesel chain, as an additive
to improve the properties of this biofuel.
Triacetin can be produced in the acid-catalyzed reaction of glycerol
with acetic acid. There are many studies [19–22] using different solid
acid catalysts and operational conditions, but the selectivity to
triacetin is normally limited. The presence of water, shifting
equilibrium and weakening the catalyst acid strength, as well as the
sequential acetylation of the hydroxyl groups contributes for this
result. Recently, triacetin was produced in 100% selectivity from
glycerol with the use of a two-step process [23]. Initially, glycerol is
reacted with nine fold molar excess of acetic acid in the presence of
Amberlyst-35 acid resin. After 4h at 105 °C, acetic anhydride is
introduced in the reaction medium to complete the acetylation,
yielding 100% triacetin. Calculations suggest [24] that acetylation with
acetic acid is an endothermic process, requiring high energy demand
for the introduction of the third acetyl group to form triacetin. In
produce biodiesel [1]. It is normally produced in 10 wt.% mass balance
from the transesterification reaction, and its global production is
estimated to reach 1.2 million tons by 2012 [2]. This enormous
amount of glycerol must find an economical destination. The chemical
transformation [3–5] of glycerol into more valuable compounds is one
of the most promising applications. Glycerol acetates, namely mono,
di and triacetin, have major applications in cryogenics [6], plastics, [7]
and fuel additives [8]. However, the efficient production of triacetin
(
glycerol triacetate) requires large excess of the acetylation reactant,
normally acetic acid, as well as long reaction times or expeditious
water removal to shift equilibrium. We wish to report in this
communication, the fast and easy production of triacetin using acetic
anhydride and a solid acid catalyst.
Biodiesel is presently one of the major biofuels used worldwide. It
is normally produced from the transesterification of triglycerides with
methanol, under base catalysis conditions. This reaction affords the
fatty acid methyl esters, the biodiesel themselves, and glycerol. Today,
one of the most challenging aspects of the biodiesel technology is
related with the economical utilization of the glycerol, or glycerin,
formed. The major uses of this compound are in personal care
products, cosmetics and soaps, but these sectors will not be able to
drain the tons of glycerol coming from biodiesel production.
⁎
Corresponding author. Universidade Federal do Rio de Janeiro, Instituto de Química.
Av Athos da Silveira Ramos 149, CT Bloco A, 21941-909, Rio de Janeiro, Brazil. Fax: +55
1 25627106.
E-mail address: cmota@iq.ufrj.br (C.J.A. Mota).
2
1
566-7367/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.catcom.2010.05.007

L.N. Silva et al. / Catalysis Communications 11 (2010) 1036–1039
1037
Scheme 1. Acetylation of glycerol with acetic anhidride.
comparison, acetylation with acetic anhydride is exothermic, favoring
formation of triacetin. These studies prompted us to show our results
of glycerol acetylation with acetic anhydride in the presence of solid
acid catalysts (Scheme 1).
Reactions with acetic acid as acylating reagent were also carried
out. The procedure was similar as the one reported previously [20],
with a molar ratio acid/glycerol of 4.
3
. Results and discussion
2
. Experimental
In all reactions with acetic anhydride, no glycerol was detected
among the products, indicating 100% conversion. However, the
formation of di and triacetin involves consecutive acetylations of the
remaining hydroxyl groups and although we express the results in
terms of selectivity, they are indirectly related with catalyst activity.
Table 1 shows the results of glycerol acetylation with acetic
anhydride. One can see that 100% selectivity to triacetin is achieved
within 20 min with the use of zeolite H-Beta and K-10 Montmoril-
lonite as catalyst, at 60 °C and 4:1 molar ratio of anhydride to glycerol,
respectively. Reduction of the molar ratio to the stoichiometric value
decreases the selectivity to triacetin, even after 2 h of reaction time,
but triacetin is still the major product formed. Amberlyst-15 acid resin
The reactions were carried out in batch conditions, mixing 2.0 g of
glycerol, 8.2 mL of acetic anhydride (4:1 molar ratio anhydride to
glycerol) and a mass of catalyst corresponding to 2.0 mmol of acid
sites. The acidity was obtained from n-butylamine termodesorption
experiments, and the procedure was reported elsewhere [25]. In some
experiments, a molar ratio of 3:1 of anhydride to glycerol was used.
Reactions were carried out at 60 °C and, in some cases, at 120 °C. Blank
experiments, without adding catalysts, were also carried out for
comparison purposes. The product distribution was determined,
withdrawing samples from the reaction mixture at specific time
intervals and analyzing them by gas chromatography coupled with
mass spectrometry. In all cases, 1,4-dioxane was used as external
standard.
Table 1
Selectivity to glycerol acetates in reactions of glycerol with acetic anhydride at different
a
conditions.
Catalyst
Molar
ratio
Temperature Time
Selectivity to
acetins %
(°C)
(min)
anhydride:
glycerol
Mono
Di
Tri
H-Betab
4:1
3:1
4:1
3:1
4:1
4:1
60
60
60
60
60
60
60
120
60
120
20
120
20
120
80
20
120
80
–
–
–
100
62
100
78
100
90
47
100
34
94
H-Beta
38
–
K-10c
–
K-10
–
22
–
d
Amberlyst-15
Amberlyst-15
–
–
10
53
–
Niobium phosphatee 4:1
–
Niobium phosphate
Blank
Blank
4:1
4:1
4:1
–
Fig. 1. Glycerol conversion ( ) and selectivity to monoacetin ( ), diacetin ( ), triacetin
120
120
10
–
56
6
(
) and acetol ( ) in the reaction with acetic acid at 120 °C, catalyzed by zeolite Beta.
a
Reaction carried out with 2.0 g of glycerol. In all cases, glycerol conversion was
1
00%.
b
c
2
Si/Al=16, area=633 m /g, acidity=1.6 mmol/g. Pre-treatment at 500 °C/30 min.
2
Si/Al=6.6, area=240 m /g, acidity=0.53 mmol/g. Pre-treatment at 150 °C/30 min.
Area=50 m /g, acidity=4.7 mmol/g. Pre-treatment at 120 °C/30 min.
Area=187 m /g, acidity=0.4 mmol/g. No pre-treatment temperature.
d
e
2
2
Table 2
a
Selectivity for glycerol acetates in reactions of glycerol and acetic acid at 120 °C.
Catalystb
Conversion Selectivity (%)
(
%)
Monoacetin Diacetin Triacetin Acetol
H-Beta
K-10
Niobium phosphate 100
Amberlyst-15
Blank
94
100
48
36
38
18
50
39
52
49
55
40
4
6
7
24
4
9
6
5
3
7
100
91
a
Molar ratio acetic acid to glycerol of 4:1. Results after 120 min of reaction.
Mass of catalyst used corresponds to 2.0 mmol of acid sites.
Fig. 2. Glycerol conversion ( ) and selectivity to monoacetin ( ), diacetin ( ), triacetin
) and acetol ( ) in the reaction with acetic acid at 120 °C, catalyzed by Amberlyst-15.
b
(

1038
L.N. Silva et al. / Catalysis Communications 11 (2010) 1036–1039
Scheme 2. AAC2 mechanism with formation of a tetrahedral intermediate.
is slightly less active, yielding 100% selectivity to triacetin only after
0 min of reaction and a molar ratio of 4:1. Within the first 20 min, the
selectivity to triacetin with this catalyst is 90%, with the remaining
reactions with acetic anhydride (only traces were observed in some
reactions, especially at higher temperatures).
8
It is not completely clear why the acetylation with acetic
anhydride performs well on zeolite Beta, whereas it does not with
acetic acid. We have shown that zeolite Beta is a good catalyst for the
acetalyzation of glycerol with formaldehyde [27], which is sensitive to
the presence of water. The explanation was that, due to the high Si/Al
ratio of this zeolite, its pore environment is hydrophobic [28] and does
not favor the adsorption of the water released during the reaction on
the active sites. It might also prevent or minimize the reverse reaction,
because the water tends to diffuse out of the pores. Thus, we cannot
explain the low selectivity to triacetin in the glycerol acetylation with
acetic acid on zeolite Beta to the effect of water, by shifting
equilibrium or weakening the acid sites.
1
6
1
8
0% as diacetin. Niobium phosphate is considerably less active at
0 °C. After 120 min, the selectivity to triacetin was only 47%. At
20 °C, this catalyst showed 100% selectivity to triacetin within
0 min of reaction time. By contrast, the uncatalyzed (blank reaction)
showed a considerably lower selectivity at both temperatures. At
6
1
0 °C, the selectivity to triacetin was 34% after 120 min, whereas at
20 °C the selectivity was about 94% within the same period. These
results show that triacetin can be obtained in excellent selectivity at
0 °C with the use of zeolite H-Beta or K-10 Montmorillonite as
6
catalysts and a molar ratio of anhydride to glycerol of 4 to 1.
Table 2 shows the results of glycerol acetylation with acetic acid at
1
20 °C, using the same solid acid catalysts. One can see that in all
One can envisage two possible mechanisms [29] for acetylation in
strong acidic medium: the first one is the normal AAC2 mechanism,
involving protonation of the carbonyl oxygen atom and nucleophilic
attack in the carbonyl to form a tetrahedral intermediate, presenting a
quaternary carbon atom (Scheme 2); the second mechanism is the
cases, the selectivity to triacetin is significantly lower when compared
with reactions with acetic anhydride. The best selectivity to triacetin
(
24%) was achieved with use of Amberlyst-15. The other catalysts
showed selectivity in the range of 4 to 7%, similar to the uncatalyzed
reaction. Zeolite Beta showed the worse performance among the
catalysts tested, presenting glycerol conversion lower than 100% and
the highest selectivity to monoacetin (Fig. 1). These results are similar
to what was found [20] in the acetylation of glycerol with acetic acid
catalyzed by HUSY and HZSM-5 zeolites, which did not show good
conversion and selectivity to triacetin either. Exclude by contrast,
Amberlyst-15 was the best catalysts tested (Fig. 2). The acid resin is
the strongest acid catalysts tested [25] and this might partly explain
the results. However, neither K-10, nor niobium phosphate have
higher acid strength than zeolites [25] and their performance might
be associated with other factors. A possible explanation is that their
structure would be capable of adsorbing the water formed, probably
shifting the equilibrium or preventing catalyst deactivation. It is
interesting to note that K-10 Montmorillonite also showed a good
performance in glycerol etherification with benzyl alcohol [26], a
reaction that also releases water as byproducts. Acetol (hydroxi-
acetone), arisen from glycerol dehydration, was observed over all
catalysts tested. This compound was not significantly observed in the
A
AC1, where protonation takes place in the oxygen atom attached to the
carbonyl group, followed by formation of an acylium ion (Scheme 3).
The first pathway is normally less energetic, because of the higher
stability of the intermediate formed upon protonation in the carbonyl
oxygen atom. On the other hand, formation of the tetrahedral
intermediate is space demanding, and the second mechanism,
involving the acylium ion, prevails in situations of steric constraints
[30]. Zeolites are known for their shape selectivity properties [31],
especially concerning the formation of bulk transition states. It is
possible that in acetylation with acetic anhydride, formation of the
tetrahedral intermediate inside the zeolite pores might be prevented,
due to transition state shape selectivity. This might be especially
difficult for the third acetylation to form triacetin. Therefore, acylation
takes place through the AAC1 mechanism. By contrast, it seems that the
same mechanism is difficult to operate in the case of acetylation with
acetic acid, even at higher temperature. This point is not completely
clear and cannot be answered with the data of this study. We propose
an explanation based on the interaction of the protonated acetic acid
Scheme 3. AAC1 mechanism with formation of an acylium ion.

L.N. Silva et al. / Catalysis Communications 11 (2010) 1036–1039
1039
compound from acetic acid, structural arrangements do not permit
assistance from the framework of oxygen atoms of the zeolite.
Nevertheless, formation of the acylium ion from the anhydride allows
the participation of the zeolite framework, stabilizing the formation of
the intermediate.
Acknowledgements
Scheme 4. Proposed structure of the protonated acetic acid on the zeolite surface and its
dehydration to the acylium ion (no zeolite assistance).
Authors thank financial support from FINEP, CNPq and FAPERJ.
References
[
[
[
1] F.R. Ma, M.A. Hanna, Bioresour. Tech. 70 (1999) 1–15.
2] C.H. Zhou, J.N. Beltramini, Y.X. Fan, G.Q. Lu, Chem. Soc. Rev. 37 (2008) 527–549.
3] A. Behr, J. Eilting, K. Irawadi, J. Leschinski, F. Lindner, Green Chem. 10 (2008)
1
3–30.
4] M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi, C.D. Pina, Angew. Chem. Int. Ed. 46
2007) 4434–4440.
[
(
[
5] F. Jérôme, Y. Pouilloux, J. Barrault, ChemSusChem 1 (2008) 586–613.
[
6] A.V. Nikolenko, A.M. Kompaniets, V.I. Lougovoy, Probl. Kriobiologii 4 (1995)
36–41.
Scheme 5. Proposed structure of the protonated acetic anhydride and its transforma-
[7] [7] Y. Taguchi, A. Oishi, Y. Ikeda, K. Fujita, T. Masuda, JP patent 298099, 2000.
tion to the acylium ion (zeolite-assisted).
[8] R. Wessendorf, Erdoel & Kohle, Erdgas Petrochemie 48 (1995) 138–142.
[
9] M.A. Dasari, P.P. Kiatsimkul, W.R. Sutterlin, G.J. Suppes, Appl. Catal. A 281 (2005)
25–231.
10] Y. Kusunoki, T. Miyazawa, K. Kunimori, K. Tomishige, Catal. Commun. 6 (2005)
45–649.
2
[
[
6
and protonated acetic anhydride with the zeolite surface, prior to the
acyl cation formation. In the case of acetic acid, the two hydrogen atoms
are interacting with the zeolite structure through hydrogen bonds
11] I. Furikado, T. Miyazawa, S. Koso, A. Shimao, K. Kunimori, K. Tomishige, Green
Chem. 9 (2007) 582–588.
[12] S.H. Chai, H.P. Hang, Y. Liang, B.Q. Xu, Green Chem. 10 (2008) 1087–1093.
13] H. Atia, U. Armbruster, A. Martin, J. Catal. 258 (2008) 71–82.
14] E. Tsukuda, S. Sato, R. Takahashi, T. Sodesawa, Catal. Comm. 8 (2007) 1349–1353.
15] S.S. Yazdani, R. Gonzalez, Curr. Opin. Biotech. 18 (2007) 213–219.
[
[
[
(
Scheme 4). This type of structure has already been proposed in other
zeolite-catalyzed reactions involving protonation of hydroxyl groups
32–34], and might impair the formation of the acylium ion, because it
[
[16] R.R. Soares, D.A. Simonetti, J.A. Dumesic, Angew. Chem. Int. Ed. 45 (2006)
would be away from the zeolite surface and would not be well
stabilized. On the other hand, in the case of acetic anhydride, formation
of the acylium ion might be assisted by the zeolite surface (Scheme 5),
favoring its formation. It is well recognized that the zeolite surface
assists the formation of cationic species [35], as well as stabilizes them
through covalent bonding.
The use of acetic anhydride in the preparation of triacetin from
glycerol may become feasible with zeolite Beta, K-10 Montmorillonite
or Amberlyst acid resin as catalysts. The reaction is exothermic and
can be carried out at mild temperatures and short reaction times with
use of these catalysts. This permits a better control of the system for
scaling up purposes.
3982–3985.
[
17] E. Garcia, M. Laca, E. Pérez, A. Garrido, J. Peinado, Energy & Fuel 22 (2008)
274–4280.
18] J. Delgado, Es Pat. 2201894 (2002).
4
[
[19] D. Gelosa, M. Ramaioli, G. Valente, M. Morbidelli, Ind. Eng. Chem. Res. 42 (2003)
536–6544.
20] V.L.C. Gonçalves, B.P. Pinto, J.C. Silva, C.J.A. Mota, Catal. Today 133–135 (2008)
73–677.
[21] J.A. Melero, R. van Grieken, G. Morales, M. Paniagua, Energy & Fuel 21 (2007)
782–1791.
22] P. Ferreira, I.M. Fonseca, A.M. Ramos, J. Vital, J.E. Castanheiro, Appl. Catal. B 91
2009) 416–422.
[23] X. Liao, Y. Zhu, S.G. Wang, Y. Li, Fuel Proc. Tech. 90 (2009) 988–993.
6
[
6
1
[
(
[
[
24] X. Liao, Y. Zhu, S.G. Wang, H. Chen, Y. Li, Appl. Catal. B 94 (2010) 64–70.
25] V.L.C. Gonçalves, R.C. Rodrigues, R. Lorençatto, C.J.A. Mota, J. Catal. 248 (2007)
158–164.
[
26] C.R.B. da Silva, V.L.C. Gonçalves, E.R. Lachter, C.J.A. Mota, J. Braz. Chem. Soc. 20
(2009) 201–204.
4
. Conclusions
[
[
27] C.X.A. da Silva, V.L.C. Gonçalves, C.J.A. Mota, Green Chem. 11 (2009) 38–41.
28] T. Okuhara, Chem. Rev. 102 (2002) 3641–3666.
The use of zeolite Beta or K-10 Montmorillonite as catalysts in the
[29] T.H. Lowry, K.S. Richardson, Mechanism and Theory in Organic Chemistry, 2nd Ed.
acetylation of glycerol with acetic anhydride produces triacetin in
00% selectivity at 60 °C within 20 min of reaction. Amberlyst-15 acid
resin requires longer reaction times, whereas niobium phosphate
gives 100% selectivity only at higher temperatures.
By contrast, zeolite Beta showed the lowest selectivity to triacetin
when acetic acid is used as catalyst. This result was explained in terms
of the stability of the acylium ion intermediate. Upon formation of this
Harper & Row, New York, 1981 pp. 652–658.
30] M.L. Bender, H. Ladenheim, M.C. Chen, J. Am. Chem. Soc. 83 (1961) 123–127.
31] B. Smit, T.L.M. Maessen, Nature 451 (2008) 671–678.
32] A. Thursfield, M.W. Anderson, J. Phys. Chem. 100 (1996) 6698–6707.
[33] S.R. Blaszkowski, R.A. van Santen, J. Am. Chem. Soc. 118 (1996) 5152–5153.
34] T. Maihom, B. Boekfa, J. Sirijaraensre, T. Nanok, M. Probst, J. Limtrakul, J. Phys,
Chem. C 113 (2009) 6654–6662.
35] J.F. Haw, J.B. Nicholas, T. Xu, L.W. Beck, D.B. Ferguson, Acc. Chem. Res. 29 (1996)
259–267.
[
[
[
1
[
[