5200
M. Gruttadauria et al. / Tetrahedron Letters 42 (2001) 5199–5201
In a typical preparation procedure 0.1 mol of tetra-
ethoxysilane was added to 10 mL of an aqueous solu-
tion of 0.01N HCl and mixed with 5 mL of ethanol as
co-solvent. The mixture was stirred for 30 min to
obtain a sol containing principally hydrolyzed tetra-
ethoxysilane. In order to complete the hydrolysis step,
the ethanol was removed by distillation under vacuum
at 35°C. Next, the sol solution was cooled at 0°C to
avoid gelation that would be promoted at room tem-
perature. Chromium(III) chloride hexahydrate (2.66 g,
0.01 mol) was dissolved in 10 mL of bi-distilled water
and cooled at 0°C. Finally, the solution containing
chromium chloride was added slowly to the sol and
the pH was raised from 2 to 4 (Scheme 1).
The spent Cr(VI)–SiO2 reagent was washed with more
dichloromethane, then dried in the oven for a few
minutes. Chromium(VI) was regenerated by packing
the reagent in a column and again letting the substrate
stand in contact with a continuous ozone flow as
before [usually 50 mg/h within 48 h, ca. 1.3 g of spent
Cr(VI)–SiO2]. After several cycles, the Cr(VI)–SiO2
showed no change in the pore size distribution
(Scheme 2).
Several benzylic alcohols were oxidized to the corre-
sponding aldehydes in >99% selectivity and in 84 to
>99% conversion in 5–24 h (Table 1).
In the first three cycles, the reaction time was longer,
probably because the entrapped reduced chromium
was oxidized with ozone for a shorter time (30 h).
After these cycles the entrapped reduced chromium
was oxidized for 48 h. Both benzylic alcohols with
electron-withdrawing or electron-donating groups were
promptly oxidized. It has been proposed that the role
of the silica is to adsorb the lower valent chromium
species produced during the oxidation.11 At this stage
we have made no attempt to characterize fully both
the oxidized and reduced species bound to the silica.
The poor conversion (38%) of in the oxidation of
p-chloro-benzyl alcohol, using a molar ratio Cr/alco-
hol of 2/3 (entry 8), could be an indication that the
reduced species is Cr(IV).
The gelation occurred in about 20 h. The resulting
oxide ground in granules (<425 m) was washed with
water and sonicated for 30 min in the same solvent in
order to remove any chromium compound that
adhered onto the outer surface of the silica matrix.
The resulting material was heated to 343 K until a
constant weight was achieved (24 h). The Cr(III) in
the washing solution was determined spectrophotomet-
rically as CrO42−. The calculated content of the silica
entrapped chromium was 4% wt. Before use the result-
ing oxide was dried at 25°C at 1 mmHg for 12 h.
In order to have an oxidizing material, the sol-gel
entrapped chromium(III) was oxidized to Cr(VI) by
reaction with ozone. The Cr(VI)–SiO2 reagent was
obtained by packing the sol-gel Cr(III) in a column,
letting the substrate stand in contact with a continu-
ous ozone flow (10 Nl/h, ca. 50 mg O3/h) for several
hours. At the end of the process the starting green
silica turned to a red-brown material. Oxidation by
heating in air at 550°C for 5 h was inefficient.
In summary the most interesting feature of this mate-
rial is the recyclability (at least up to 16 times) without
loss in activity; moreover, we believe that the fact that
the chromium remains firmly entrapped before and
The sample showed a very high surface area (650
m /g) with a pore size distribution centered at 10–20 A
O3
2
,
as determined by nitrogen adsorption/desorption mea-
surements at 77 K. This pore size should be suitable
for the reaction with the entrapped Cr(VI). The X-ray
diffraction pattern showed only the background of the
silica amorphous glass. No reflections due to any
chromium oxide phase were detected. This could be an
indication that the chromium species are highly dis-
persed.
Cr(VI)
Cr(III)
RCH2OH
RCHO
Scheme 2. Oxidation cycles.
Oxidation of primary alcohols was carried out using a
molar ratio Cr/alcohol of 2.5/1. The reactions were
performed in dichloromethane/diethyl ether 3/1 at
room temperature in a Pyrex-glass reactor, equipped
with magnetic stirring, working in batch conditions.19
The solvent mixture was chosen in order to minimize
leaching of chromium in solution, stressing the incom-
patibility between solvent and Cr-species, as chromium
trioxide is insoluble in dichloromethane.11 The reac-
tions were monitored by GC–MS; all the products
were identified and quantified by comparison with
known samples. The Cr(VI)–SiO2 reagent was found
to lead to a very easy work-up, which became reduced
to a mere filtration. No leaching of chromium in solu-
tion was observed (spectrophotometric measurements).
Table 1. Oxidation reactions of benzylic alcohols with
Cr(VI)–SiO2
Entry
Alcohol
t (h)
Aldehyde (%)
Cycles
1
2
3
4
5
6
7
8a
PhCH2OH
PhCH2OH
48
23
7
22
5
17
24
24
98
\99
96
1–3
4–5
6–7
8–10
11–12
13–14
15–16
17
p-Cl-C6H4CH2OH
o-Br-C6H4CH2OH
p-MeO-C6H4CH2OH
o-Me-C6H4CH2OH
o-OEt-C6H4CH2OH
p-Cl-C6H4CH2OH
84
\98
\98
97
38
a Molar ratio Cr/alcohol 2/3.