Reduction of carbonyl compounds with polyethylsiloxane in the presence of titanium compounds

Article abstract of DOI:10.1023/B:RUJO.0000044534.92992.fc

Reduction of carbonyl compounds with polyethylsiloxane in the presence of titanium alkoxides gives the corresponding alcohols in high yields.

Full text of DOI:10.1023/B:RUJO.0000044534.92992.fc

Russian Journal of Organic Chemistry, Vol. 40, No. 6, 2004, pp. 759–762. Translated from Zhurnal Organicheskoi Khimii, Vol. 40, No. 6, 2004,
pp. 801–804.
Original Russian Text Copyright © 2004 by Kozhukhova, Yatluk, Suvorov, Koryakova.
Dedicated to Full Member of the Russian Academy of Sciences
O.N. Chupakhin on his 70th Anniversary
Reduction of Carbonyl Compounds with Polyethylsiloxane
in the Presence of Titanium Compounds
V. V. Kozhukhova, Yu. G. Yatluk, A. L. Suvorov, and O. V. Koryakova
Institute of Organic Synthesis, Ural Division, Russian Academy of Sciences,
ul. S. Kovalevskoi 20, Yekaterinburg, 620219 Russia
e-mail: eop@ios.uran.ru
Received January 23, 2004
Abstract—Reduction of carbonyl compounds with polyethylsiloxane in the presence of titanium alkoxides
gives the corresponding alcohols in high yields.
In the recent years, reduction of esters with silicon
hydrides acquires increasing importance from the
preparative viewpoint. Presumably, the most con-
venient reducing agents are polymeric silicon hydrides,
for these compounds are widely used in industry
and, unlike monomeric analogs, are not prone to
undergo disproportionation leading to formation of
explosive silane.
Preliminary experiments showed that during the
reduction the reaction solution turns dark blue, which
suggests formation of tervalent titanium compounds.
We presumed that the presence of Ti(III) derivatives
favors the reduction process and examined the
catalytic activity of other titanium compounds. The
results are summarized in Table 1. It is seen that
decrease of the amount of butyl titanate from 100 to
5 mol % leads to considerable reduction of the yield.
However, transformation of Ti(IV) derivative into
Ti(III), regardless of the nature of Ti(III) compound,
increases the yield to nearly quantitative.
Breeden and Lawrence [1] studied the reduction of
esters with excess polymethylsiloxane at a molar ratio
of 1:10 in the presence of an equimolar amount of
titanium isopropoxide or zirconium ethoxide on
heating in boiling THF for 5 to 80 h. The yields of the
reduction products ranged from 65 to 98%. Ridding
and Buchwald [2] reported on the reduction of esters
with polymethylsiloxane (molar ratio 1:2.5) in the
presence of 25–100 mol % of titanium isoropoxide
without a solvent at 23 to 70°C. The reaction time was
1.25 to 24 h, and the yields ranged from 8 to 99%. The
reduction of some ketones was also examined in [1].
The yield of the reduction products depends on the
ester structure (Table 2). It decreases with increase in
steric hindrance in the alkyl fragment of lauric acid
esters. However, the above system ensures successful
reduction of various esters derived from lower
alcohols. As follows from the data given for ethyl
esters, the lowest yield is observed for ethyl benzoate
and ethyl 3-phenylpropionate. In order to estimate the
In the present work we effected reduction of esters
with more accessible and more stable polyethyl-
siloxane in the presence of different titanium com-
pounds at 100°C. Initially, we compared the results of
reduction of butyl laurate with polymethylsiloxane
(PMS) and polyethylsiloxane (PES). In the presence
of an equimolar amount of butyl titanate as catalyst,
the yields of lauryl alcohol were 95 and 88%, respec-
tively. Although in the reduction with polyethyl-
siloxane the yield is somewhat lower, it nevertheless
remains fairly high.
Table 1. Reduction of butyl laurate with polyethylsiloxane
in the presence of titanium compounds (5 mol %)
Titanium compound
Ti(OBu)4
Yield of lauryl alcohol, %
30
88
95
94
87
Ti(OBu)4–Li–anthracene
Ti(OEt)4–Li–naphthalene
LiTi(OPr-i)4
LiTiOC(CH3)2CH(CH3)2
1070-4280/04/4006-0759 © 2004 MAIK “Nauka/Interperiodica”

KOZHUKHOVA et al.
760
Table 2. Reduction of esters and lactones with polyethyl-
siloxane in the presence of 5 mol % of LiTi(OPr-i)4
compounds could be formed only by self-condensation
of the corresponding aldehydes. Therefore, the reduced
yields of alcohols from aryl-substituted acetic acid
esters may result not only from steric hindrances but
also from side condensation processes. On the other
hand, these findings indicate that the reduction
involves intermediate formation of the corresponding
aldehyde which is generally reduced at a higher rate
and hence it does not accumulate in the system.
Ester or lactone
Me(CH2)10COOEt
Yield of alcohol, %
93
88
57
50
41
57
74
49
95
39
61
26
85
73
52
37
Me(CH2)10COOBu
Me(CH2)10COOPr-i
Me(CH2)10COOPh
Me3CCOOEt
H(CF2)4COOMe
Me(CH2)7CH=CH(CH)2COOEt
Heteroauxin ethyl ester
Naproxen ethyl ester
PhCOOEt
While discussing the reduction mechanism [1, 2],
the authors tend to conclude that silicon hydride reacts
with titanium alkoxide to give an intermediate which is
then converted into titanium hydride. The latter was
considered in [2] to be the active reducing species.
PhCH2COOEt
H
Ph(CH2)2COOEt
Ph(CH2)3COOEt
Butyrolactone
+
Si
H
RO Ti
Si
Ti
O
R
Valerolactone
+
Caprolactone
Si OR
H
Ti
Breeden and Lawrence [1] believe that the observed
color (typical of tervalent titanium compounds)
originates from another side process:
effect of aromatic radical in the acid fragment we
examined a series of esters in which the benzene ring
is located at different distances from the carboxy
group. Provided that ethyl 3-phenylpropionate is
excluded, the yield of the reduction product increases
as the benzene ring becomes more distant from the
carboxy group, i.e., conjugation between the carboxy
group and benzene ring inhibits the process. However,
the determining factor is steric hindrances. Actually, in
keeping with the Newman rule, substituents at the
β-carbon atom create greater steric hindrances than do
α-substituents, and the yield in the reduction of ethyl
3-phenylpropionate is the lowest. As a rule, the yields
given in Table 2 are not optimal. For example,
prolonging the reduction of ethyl benzoate from 24
to 48 h leads to a quite satisfactory yield of benzyl
alcohol (75%).
2(RO)3TiH
2(RO)3Ti
+
H2
Taking into account the results of our experiments
where Ti(III) derivatives showed the highest catalytic
activity, we believe that just Ti(III) compounds act as
reducing agents.
Reaction of titanium alkoxides with alkali metals
was studied in detail by Vyshinskaya and co-workers
[3], and the proposed mechanism is beyond doubt:
THF
(RO)4Ti
+
Li
LiTi(OR)4
The alkoxide complex thus formed is reduced to
hydride complex, and the latter is the active reducing
species (i.e., catalyst):
According to [1], the reduction of ethyl phenyl-
acetate is accompanied by formation of 2,4-diphenyl-
2-buten-1-ol as by-product.
+
+
RO Si
LiTi(OR)4
H
Si
LiTi(OR)3H
PhCH2COOMe
PhCH2CH
PhCH2CH2OH
C(Ph)CH2OH
+
Analogous trialkoxyaluminum and trialkoxyboron
hydrides are excellent reducing agents; for example,
sodium trialkoxyboron hydrides are known to be more
reactive than sodium tetrahydridoborate. Addition of
anthracene (or naphthalene) accelerates dissolution of
Analogous compound was isolated in the reduction
of ethyl 2-thienylacetate [2] (here, the yield of the
target product was as poor as 8%). Obviously, such
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 40 No. 6 2004

REDUCTION OF CARBONYL COMPOUNDS WITH POLYETHYLSILOXANE
761
Table 3. Reduction of ketones with polyethylsiloxane in the
presence of Ti(OBu)4
lithium, but the yield of the reduction product remains
essentially unaffected.
In order to compare the reduction patterns for
aromatic and aliphatic esters we examined the reduc-
tion process by IR spectroscopy using Me(EtO)2SiH as
model reducing agent. In the reduction of ethyl ben-
zoate, the carbonyl absorption intensity decreased at
a lower rate than in the reduction of ethyl valerate.
This suggests some chemoselectivity of the reducing
agent. By reduction of a 1:1 mixture of ethyl benzoate
and ethyl laurate we obtained benzyl alcohol and lauryl
alcohol at a ratio of 12:80.
As noted above, reduction of aldehydes can be
accompanied by their self-condensation catalyzed by
titanium alkoxides [4]. Therefore, using cyclohexa-
none as an example, we preliminarily examined how
the amount of the catalyst affects the yield of the
reduction product. The yield of cyclohexyl alcohol was
46, 41, 40, and 25%, respectively, in the presence of
25, 50, 75, and 100 mol % of the catalyst. The low
yield of the alcohol generally results from the con-
current self-condensation of cyclohexanone:
Ketone
Yield, %
Me(CH2)5COMe
Me2CHCH2COMe
Me3CSOMe
67
44
44
47
30
42
38
40
68
PhCOMe
PhCOPh
(CF3)2CFCOCF2CF3
Cyclopentanone
Cyclohexanone
Tetrahydronaphthalen-1-one
Table 4. Reduction of aldehydes with polyethylsiloxane in
the presence of Ti(OBu)4
Aldehyde
Me2CHCH2CHO
Me(CH2)7CHO
PhCHO
Yield, %
35
40
51
p-O2NC6H4CHO
PhCH2CHO
16
O
O
14–17
00
2-Furaldehyde
The low yield of the corresponding alcohol
obtained from p-nitrobenzaldehyde is likely to result
just from side reactions (it is known that nitro group is
not reduced with silicon hydrides). In the reduction of
furaldehyde, the reaction mixture solidified, presum-
ably as a result of polycondensation.
According to the GLC data, the still residue
obtained after distillation of the reaction mixture con-
tained 2-(1-cyclohexenyl)cyclohexanone and products
of more profound condensation. The results showed
that the use of an equimolar amount of titanium
alkoxide is the least favorable. In the other cases, the
yield may be regarded as constant within the experi-
mental error. Therefore, in further experiments we
used an average (50%) amount of the catalyst. No
appreciable increase in the yield was observed in the
presence of a complex catalyst. On the whole, satisfac-
tory to good yields can be achieved (Table 3).
Aldehydes are reduced more readily than ketones,
and the reaction successfully occurs at 50°C (Table 4).
The yields of the reduction products are not high. In
addition, various by-products are formed. In the reduc-
tion of benzaldehyde, the alkaline extract contained
benzoic acid. Presumably, the process is accompanied
by the Tishchenko reaction:
IR spectroscopic study of the reduction of aceto-
phenone and methyl isobutyl ketone with
Me(EtO)2SiH (following the change in the carbonyl
absorption intensity) showed that the aromatic ketone
is reduced at a lower rate. Analogous study of the
reduction of isovaleraldehyde revealed that decrease in
the carbonyl absorption intensity is accompanied by
appearance of a new absorption band at 1692 cm–1,
which is typical of α,β-unsaturated compounds; its
intensity increases with time. These data confirm
the above assumptions concerning side condensation
processes.
EXPERIMENTAL
Ti(OR)4
GLC analysis was performed on a Tsvet-100
chromatograph equipped with a thermal conductivity
detector; carrier gas helium, inlet pressure 2 atm;
2PhCHO
NaOH
PhCOOCH2Ph
H+
PhCOONa
PhCOOH
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 40 No. 6 2004

KOZHUKHOVA et al.
762
100×0.6-cm column packed with 5% of tris(cyano-
ethoxy)propane on Chromaton AW (grain size 0.200–
0.250 mm). The products were identified using
authentic samples. The IR spectra were recorded on
a Perkin–Elmer Spectrum-One spectrometer.
extract was subjected to fractional distillation. The
silicon hydride-to-NaOH molar ratio was ≥1:3. The
products were identified by comparing their physical
constants with published data.
REFERENCES
A mixture of the coresponding ester, silicon
hydride, and titanium compound at a molar ratio of
1:3:(0.05–1) was heated for 24 h at 100°C (the
reaction flask was plugged with wool). A 10% solution
of sodium hydroxide was added (the mixture can be
transferred into another flask by washing with THF),
the mixture was heated for 3 h at the boiling point,
cooled, and extracted with diethyl ether, and the
1. Breeden, W. and Lawrence, J., Synlett, 1994, p. 833.
2. Ridding, M.T. and Buchwald, S.L., J. Org. Chem., 1995,
vol. 60, p. 7884.
3. Vyshinskaya, L.I., Drobotenko, V.V., and Samarina, T.P.,
Zh. Obshch. Khim., 1988, vol. 58, p. 1332.
4. Feld, R. and Cowe, P.L., The Organic Chemistry of
Titanium, London, Butterworths, 1965.
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 40 No. 6 2004