1
56
E. Battistel et al. / Applied Catalysis A: General 394 (2011) 149–157
than TBD and consequently it has a lower reactivity. The addition
of MeOH did not improve the DBU performance. This is a surprising
result if one considers that the catalytic active species, the methox-
ide ion, should be easily derived from methanol in the presence of
a strong base.
two esters in our case) have to adsorb on closely neighbouring free
reactive sites. Therefore, a reasonable hypothesis, although based
on evidences not yet fully established, is that the surface-mediated
reaction step between two adsorbed species, more the adsorption
of a single species from solution, as in the case of acid catalysis
[
27], may become the rate determining step [26,30]. This implies
3
.5. Heterogeneous base catalysis
that effective catalysis critically could depend on geometric and
spatial distribution of the surface catalytic sites other than on the
basicity strength of the metal oxides.
The results obtained in the presence of various basic heteroge-
neous catalysts are presented in Table 6. Hydrotalcites are double
2
+
3+
layered Mg /Al hydroxides whose regular spatial arrangement
is lost during calcination at high temperatures. After thermal treat-
4. Conclusions
ment, which induces the evolution of H O and CO , a porous
2
2
The double ester exchange between tributyrin, a model triglyc-
amorphous mixed oxide is formed. This material has been suc-
cessfully used to catalyze the transesterification of vegetable oil
with methanol [16]. The intrinsic basicity of the oxide surface may
be tuneable by changing the Mg/Al ratio. As shown in Table 2, the
increase of the Mg/Al ratio induced a concurrent increase of the base
sites (determined by CO2 adsorption calorimetry). Unfortunately,
this did not afford a parallel significant increase of the catalytic
activity as suggested by the missed detection of TAG formation.
It was suggested that enhancement of the transesterification
catalytic activity of the mixed oxide was apparently improved after
the frame insertion of Fe3 ion due to the increase of the active site
basicity (results based on colorimetric evidences) [16]. Therefore,
mixed oxides containing Fe3 ion were prepared with a similar syn-
thesis protocol and characterized by standard XRD and ICP surface
analysis that showed the correct structural insertion of the iron
eride, and methyl acetate has been carried out in the presence of
either acid or basic catalysts. Base catalysis is characterized by fast
reaction rates (few minutes) and moderate temperature require-
◦
ment (60–80 C), whereas acid catalysis achieves similar yields in
longer reaction time (several hours) and at higher temperature
◦
(130–140 C). These results are similar to those obtained in the case
of transesterification of vegetable oils with methanol [19].
Interestingly, under optimized conditions, both types of cata-
lysts yielded a similar amount of reaction products, reaching the
same final TGA selectivity of about 70% even in those cases in
which a quantitative conversion of the tributyrin was observed.
This implies that the third acetylation step from DAG to form TAG
did not go to completion with either type of catalysts. In order to
achieve a complete conversion is probably necessary to consider
adding a side reaction able to force the equilibrium towards the
end product, TAG.
The weight percent composition of the final products mixture
is: BuMe 55%, TAG 28%, DAG 15%, MAG 2%, TriBu 0%, out of which
only MAG (2%) may be considered a truly unwanted impurity. The
two main products add up to more than 80% (up to 98% including
DAG). This mixture can be taken into consideration as a component
of fossil fuels blends (Table 1) if a long fatty acid chain triglyceride
+
+
(
data not shown). The properties of the iron doped oxides are pre-
3+
sented in Table 2. The enrichment of the mixed oxide with Fe
ion did not alter significantly surface area, porosity and base sites
concentration (as assessed by adsorption calorimetry).
As shown in Table 6, the mixed oxide containing a Mg–Al ratio
◦
◦
of 3:1 is almost inactive at 80 C and slightly active at 140 C. A
further enhancement of the conversion was observed by increas-
ing the catalyst concentration up to 15% (w/v), suggesting that the
distribution and concentration of the actual active sites per unit of
oxide surface plays a relevant role on the overall catalytic activity.
On the other hand, long reaction times (20 h, data not shown) did
not change the conversion and distribution of the intermediates.
The increase of the Mg/Al ratio (up to 6:1) and the insertion of Fe3
ions did not alter significantly the results obtained with the mixed
oxide at 15% (w/v) concentration.
(a vegetable oil) were used instead of tributyrin, a triglyceride with
a very short chain fatty acid. The study of this reaction by using
vegetable oils as substrates could be a further step forward in the
search of suitable processes for the synthesis of green diesel com-
ponents without constrained glycerol accumulation as in the case
of the traditional biodiesel synthesis.
+
Acknowledgments
The commercial hydrotalcite Pural Mg 70 (Mg/Al 3:1) calcined
to the corresponding oxides showed an activity even lower than
that observed with the other oxides at the same Mg–Al molar ratio.
Other basic catalysts, such as titanium-silicalite ETS-10 and sodium
aluminateKatalco, althoughefficient in the transesterificationreac-
tions [29], were almost inactive. It should be noted that, unlike
ETS-10, Katalco and Pural oxides have average pore diameters that
should not significantly limited the accessibility of the tributyrin
molecule inside the pores (Table 2).
The activity of these heterogeneous base catalysts is signif-
icantly lower than that observed in homogenous conditions. A
possible explanation may be related to the intrinsic basicity
strength of the material surface which, although enough to cat-
alyze other transesterification reactions, may not sufficient to carry
on the ester exchange reaction. On the other hand, the dependence
on the concentration of the activity (see Table 6) may suggest that
the activity also depends critically on the amount and distribution
of the catalytic sites and/or on the surface properties (porosity) of
the material. As a matter of fact, in heterogeneous catalysis the
interaction between the incoming reactants and the surface active
sites is a critical prerequisite for the reaction. Kinetic studies of
the transesterification mechanism in the presence of strong base
sites suggested that the reacting molecules (which correspond to
Dr. R. Buzzoni is gratefully acknowledged for the synthesis of
doped mixed oxides and for providing fully characterized sam-
ples of the heterogeneous catalysts. Dr. Daniela Meloni and Franca
Sini (Department of Chemical Sciences, University of Cagliari, Italy)
are specially acknowledged for the synthesis of mixed oxides
at different molar ratios and for the carefully surface area and
microcalorimetric measurements, respectively. E.M. Usai thanks
the Regione Autonoma della Sardegna for financial support through
PO Sardegna FSE 2007-2013, L.R. 7/2007 “Promozione della ricerca
scientifica e dell’innovazione tecnologica in Sardegna”.
References
[
1] C.-H. Zhou, J.N. Beltramini, Y.-X. Fan, G.Q. Lu, Chem. Soc. Rev. 37 (2008)
527–549.
[2] J.A. Melero, R. van Grieken, G. Morales, M. Paniagua, Energy Fuels 21 (2007)
782–1791.
1
[
[
3] D. Fabbri, V. Bevoni, M. Notari, F. Rivetti, Fuel 86 (2007) 690–697.
4] K. Klepá cˇ ová, D. Mravec, M. Bajus, App. Catal. A: Gen. 294 (2005) 141–147.
[5] E. García, M. Laca, E. Pérez, A. Garrido, J. Peinado, Energy Fuels 22 (2008)
274–4280.
4
[
[
6] S. Saka, Y. Isayama, Fuel 88 (2009) 1307–1313.
7] D. Gelosa, M. Ramaioli, G. Valente, M. Morbidelli, Ind. Eng. Chem. Res. 42 (2003)
6536–6544.