Etherification of Glycerol by Li–Mg/SBA-15 Catalyst
215
Fig. 4 Shape selective effect of 10Li10Mg/SBA-15 catalyst on the
distribution of linear DG isomers. (T = 240 °C, 2 wt% of catalyst
loading, at conversion level of 50 %)
Fig. 5 Shape selective effect of 10Li10Mg/SBA-15 catalyst on the
distribution of linear DG isomers at different reaction times.
(T = 240 °C, 2 wt% of catalyst loading)
effect in 10Li10Mg/SBA-15 catalyst could not be the only
governing mechanism responsible for the dimers distribu-
tion. Additionally, this catalyst had an average pore size of
up to 11.95 nm as obtained in surface analysis. This size
was large enough to allow the formation of any of DG
isomers or even the formation of linear higher oligomers
such triglycerol. However, the formation of cyclic com-
pound was not detected in this reaction system.
4 Conclusions
10Li10Mg/SBA-15 could provide high diglycerol yield of
up to 63 % under the reaction conditions with a diglycerol
selectivity of 82 %. The catalyst had large pore size of up
to 11.95 nm which could allow minimization of internal
diffusion effects while inhibiting subsequent etherification
reaction steps to form higher oligomers. Besides, the cat-
alyst predominantly yielded ab-DG isomer in contrast to
homogeneous LiOH catalyst that favored aa0-DG isomer.
Improved understanding on the influence of mesoporosity
of the catalyst on DG isomers distribution was achieved.
Homogeneous LiOH used in this glycerol etherification
reaction resulted in a DG yield of 40 % at a glycerol
conversion of 50 % after 1.5 h. In contrast with heteroge-
neous catalysis system, the aa0-DG isomer (45 %) pre-
dominated in this reaction system. Thus, the negative
charge of an isolated glycerol anion in aqueous environ-
ment was preferentially localized on the b-oxygen com-
pared to a-oxygen. However, in the presence of Li cations,
the a-oxygen position was much more preferred due to its
coordination to Li? [14]. Hence, a-oxygen substituted
products were more favored to be formed.
Acknowledgments A Research University Grant (814144) to sup-
port this work is acknowledged.
References
1. Anuar MR, Abdullah AZ, Othman MR (2013) Catal Commun
32:67
Another possible explanation might be related with the
nature of isomers itself. It was observed that the proportion
of ab-DG was maintained for all time from 8 to 18 h in the
products mixture (Fig. 5). However, the aa-DG slightly
decreased while bb0 gradually increased with the course of
the reaction. It was suggested that the aa0-DG isomer,
being the most reactive isomer as revealed in the reaction
was preferentially consumed in the consecutive reactions
with increasing reaction time. As compared to ab-DG
molecule, aa0-DG molecule is expected to interact more
effectively with the Li sites. This phenomenon led to a
decrease in the aa0-DG desorption rate which resulted in
the distribution products mixture. Otherwise, the bb0-DG
formation was not favored, probably due to preferred a-
oxygen activation on Li-sites [14]. It was expected that the
bb0-DG formation was still low inside the mesopores of the
catalyst but it could instantaneously desorb from the porous
solid and quickly removed from the equilibrium [5].
2. Gholami Z, Abdullah AZ, Lee KT (2013) J Taiwan Inst Chem
Eng 44:117
3. Clacens JM, Pouilloux Y, Barrault J (2002) Appl Catal A 227:181
4. Martin A, Richter M (2011) Eur J Lipid Sci Technol 113:100
5. Hees U, Bunte R, Hachgenei JW, Kuhm P, Harris EG (1993) PCT
Int Appl WO9325511
6. Ayoub M, Abdullah AZ (2013) Catal Commun 34:22
7. Krisnandi YK, Eckelt R, Schneider M, Martin A, Richter M
(2008) ChemSusChem 1:835
8. Fulvio PF, Pikus S, Jaroniec M (2005) J Colloid Interf Sci
287:717
9. Sweeley CC, Bentley R, Makita M, Wells WW (1963) J Am
Chem Soc 85:2497
10. Sun LB, Kou JH, Chun Y, Yang J, Gu FN, Wang Y, Zhu JH, Zou
ZG (2008) Inorg Chem 47:4199
11. Ayoub M, Abdullah AZ (2011) J Appl Sci 11:3510
12. Thielemann JP, Girgsdies F, Schlogl R, Hess C (2011) J Nano-
technol 2:110
´
13. Ruppert AM, Meeldijk JD, Kuipers BWM, Erne BH, Weckhuy-
sen BM (2008) Chem Eur J 14:2016
14. Smit B, Maesen TLM (2008) Nature 451:671–678
123