330 J. CHEM. RESEARCH (S), 1997
J. Chem. Research (S),
1997, 330–331†
Regioselective Reductive Cleavage of Terminal Epoxides
with Polymer-supported Chloroaluminium
Tetrahydroborate†
Bahman Tamami,* M. Mansour Lakouraj and Hamid Yeganeh
Department of Chemistry, Shiraz University, Shiraz, Iran
Epoxides are reduced exclusively to the less substituted alcohols with regenerable polyvinylpyridine-supported
chloroaluminium tetrahydroborate in high yields.
The reductive cleavage of epoxides to alcohols is one of the
most useful reactions in organic synthesis.1 Cleavage of
unsymmetrical substituted epoxides with most conventional
reducing agents generally results in the more substituted
alcohols.2 Considerable effort has focused on the develop-
ment of methods for regioselective epoxide cleavage to the
less substituted alcohols, and for this purpose several types of
reagents have been used.2,3 However, in most cases mixtures
of alcohols have been obtained. The most satisfactory results
have been achieved only with a limited number of reagents
such as sodium cyanotrihydroborate–BF3 etherate,3b potas-
sium triphenylhydroborate–Ph3B3c system and recently silica
gel-supported zinc tetrahydroborate reported by Ranu3e,f and
the sodium tetrahydroborate–amine system via a photo-
chemical approach.3h
Polymer-supported reducing agents have received con-
siderable attention in recent years and a variety of them,
while exhibiting the advantages of polymeric reagents, have
been used in the mild and selective reduction of organic
compounds.4 The only reported polymer-supported tetrahy-
droborate reagents are Amberlyst anion exchange resin-
supported tetrahydroborate and cyanotrihydroborate5 as well
as polymer-supported zinc6 and zirconium7 tetrahydrobor-
ates which we recently reported as stable, efficient and select-
ive reducing agents for a variety of organic functional
groups.8 With these polymer-supported metal tetrahydrobor-
ates, however, the regioselective reduction of epoxides was
not possible and a mixture of alcohols was obtained with the
more substituted alcohols being the major products.8
We now report that in the course of our studies on a new
polymer-supported metal tetrahydroborate, namely, poly-
vinylpyridine-supported chloroaluminium tetrahydroborate,
we found that it can reduce epoxides exclusively to their
lesser substituted alcohols in almost quantitative yields.
Aluminium chloride was supported on crosslinked poly-
(4-vinylpyridine) by adding a solution of the metal salt to an
alcoholic suspension of the polymer. The polymeric reagent
was then obtained by an exchange reaction between the poly-
mer-supported aluminium chloride and lithium tetrahydro-
borate. It was a white-cream, stable and non-hygroscopic
powder. The capacity of the reagent was 1.8 mmol BH4 per
gram. Data obtained on the Al and Cl content by atomic
absorption and potentiometric titration techniques showed,
within experimental error, that the reducing species are in
the form of AlCl2(BH4) supported on the polymeric ligand
(such species are known to be formed as unstable inter-
mediates in the formation of aluminium tetrahydroborate
from the reaction of AlCl3 and LiBH4).9 Attempts to prepare
polymer-supported aluminium tetrahydroborate, Al(BH4)3,
were not successful.
using this reagent and the results are shown in Table 1. The
reductions were performed in solvents such as diethyl ether,
tetrahydrofuran, dichloromethane, acetonitrile, methanol
and ethanol under reflux. The reagent was most efficient in
absolute ethanol. No solvolytic ring opening of the epoxides
was detected in this solvent. Identification and analysis of the
1
products were made on the basis of IR, H NMR and GLC
evidence compared with the authentic samples. Regenera-
tion of the reagent was achieved by washing with acid and
then base to obtain the original polymer followed by com-
plexation and exchange reactions. The regenerated poly-
meric reagent, except for a small weight loss in each cycle,
had almost the same capacity and performance as the origi-
nal reagent.
In conclusion, this regenerable polymer-supported reduc-
ing agent provides an efficient methodology for excellent
regioselective cleavage of terminal epoxides to the less substi-
tuted alcohols. Moreover, the mildness, convenience and
high yield make this reagent an attractive reagent for such an
epoxide cleavage. Further utilization of this new stable poly-
mer-supported metal tetrahydroborate reducing agent for
different reductive transformations will be reported in due
course.
Experimental
Preparation of Poly(4-vinylpyridine)-supported Chloroaluminium
Tetrahydroborate.sCrosslinked poly(4-vinylpyridine) (2% divinyl
benzene, Fluka AG) (5.0 g) was suspended in methanol (50 ml)
and stirred for 1 h. A solution of AlCl3 (20 g, 0.15 mol) in THF–
MeOH (4:1, 150 ml) was added to the polymer suspension. The
mixture was then stirred for 12 h at room temperature and the
precipitate filtered off, washed with excess methanol and diethyl
ether and then dried in vacuum at 40 °C to yield 7.39 g polymer-
supported aluminium chloride as a light yellow powder.
To a solution of LiBH4 (3.0 g, 0.14 mol) in dry THF (100 ml), the
above supported polymer was added portionwise at 0–5 °C and
stirred for 3 h. The resulting material was separated and washed
several times with THF and diethyl ether and then dried under
vacuum to produce a white-cream stable and non-hygroscopic
powder. The capacity of the reagent was determined by iodometric
titration method to be 1.8 mmol BHꢀ4 per gram. The amount of
aluminium determined by atomic absorption spectroscopy and the
Clꢀ content determined by potentiometric titration were 1.90 and
3.85 mmol per gram of polymer respectively.
General Procedure for the Reductive Cleavage of Epoxides.sIn a
flask (100 ml) equipped with a condenser and a magnetic stirrer, a
solution of epoxide (1 mmol) in absolute ethanol (10 ml) was
prepared. The supported reagent (1.5 g) was added and the reac-
tion mixture stirred while being refluxed. Progress of the reaction
was monitored by TLC (CCl4–diethyl ether as 5:1) or GC. On
completion of the reaction, water (100 ml) was added and the
solution stirred for 5 min. The mixture was then filtered and the
filter cake washed several times with diethyl ether. The organic
layer was separated and dried over MgSO4. On evaporation of
solvent the pure product was obtained in high yield.
Regioselective reduction of various epoxides to their
corresponding less substituted alcohols were performed
Regeneration of the Reagent.sThe spent reagent (5 g) was
treated with hydrochloric acid (20 ml, 0.1 ), stirred for 0.5 h and
M
then filtered. The resulting filter cake was washed first with sodium
) and then with distilled water, acetone
*To receive any correspondence.
hydroxide (3Å10 ml, 0.1
M
†This is a Short Paper as defined in the Instructions for Authors,
Section 5.0 [see J. Chem. Research (S), 1997, Issue 1]; there is there-
fore no corresponding material in J. Chem. Research (M).
and diethyl ether. The solid was dried in vacuum at 60 °C to give
the original polymer (4.8 g) which could then be supported and
used again.