Chemistry Reference
In-Depth Information
phenomenon and catalysts that had been passivated
by trimethylsilylation gave even better performance.
The best yields obtained were 90% with a passivated
piperidine catalyst (Fig. 7.36).
A second group led by Jaenicke also has published
results in this area [143]. They investigated a range
of supported bases, including supported primary
and tertiary amines, a supported guanidine and a
supported pyrimidine. Potentiometric titrations
showed that the pyrimidine was only weakly basic,
as expected, and that the guanidine was somewhat
more basic than the primary and tertiary amines (but
by less than 1 p
K
unit, indicating an interaction with
the surface that reduces the basic strength of the cat-
alyst). As Brunel found, the catalysts improve with
reuse and good results were obtained for all the cat-
alysts investigated. Yields ranged from 84% (second
reuse) to 95%, with the guanidine catalyst being the
best (Fig. 7.36).
The use of supported guanidines has been explored
in a number of reactions, including the Knoevenagel
(where little real advantage appears over the simpler
and cheaper aminopropyl materials), the Michael
addition, base-catalysed epoxidation, aldol conden-
sations and transesterifications. These materials,
which are close to being sufficiently basic to be a
direct replacement for metal hydroxides in synthesis,
are thus promising and versatile catalysts. However,
some evidence is available that suggests that the
presence of the silica surface can reduce their basi-
city [143] compared to free guanidines, but leaving
them significantly stronger bases than aminopropyl
systems for example [69]. How much surface treat-
ments and enhanced catalyst design can reduce this
interaction remains to be seen.
Transesterifications can be carried out under acidic
or basic conditions. Brunel's group have shown that
the TBD guanidine catalyst (see Fig. 7.25) is a selec-
tive and efficient catalyst for the transesterification
of ethyl propionate [111]. This catalyst carried out
the reaction somewhat more slowly than a homoge-
neous equivalent, but reasonable conversions were
obtained. It should be noted that without separation
of the released alcohol high conversions are ham-
pered by re-equilibration in this type of reaction, and
that the rate and efficiency of separation of the
evolved alcohol often can limit the apparent reaction
rate.
Schuchardt
et al
. also have published work relat-
ing to the transesterification of soybean oil with
methanol using MCM-supported guanidines [144].
Rates were lower than unsupported catalysts but
similar conversions were obtained. Solid catalysts for
this reaction are particularly valuable because the
longer chain esters that can be produced are valu-
able as synthetic lubricants and therefore must have
very low levels of acid or base present to minimise
corrosion of engine parts under conditions of high
temperature and mechanical pressure. Water
washing is difficult due to phase separation problems
and the risk of slight levels of hydrolysis.
Corma
et al
. [145] have published details on a
novel supported hydroxide that is a useful base
catalyst. They supported trimethoxysilyl-
propyltrimethylammonium chloride onto MCM-41
by grafting. Subsequent ion-exchange with tetra-
methylammonium hydroxide in methanol gave the
supported trimethylammonium hydroxide. They
showed that this material was an excellent catalyst
for the Knoevenagel reaction of benzaldehyde and
ethyl cyanoacetate, as well as for the correspond-
ing reaction using phenylsulphonylacetonitrile as
the carbon acid. The catalyst also was active in the
Michael addition.
The Michael addition of carbon acids to enones is
an atom-efficient transformation that can be carried
O
CO
2
H
+
R
NH
2
N
N
N
OH
N
Si
Si
Si
O
Si
O
Si
O
O
OR
O
OR
O
OR
O
O
O
COR
OH
HO
Fig. 7.36
Monoglyceride formation by
various catalysts.
Brunel 62% Brunel 90% Jaenicke 95%
