The 'Hirao reduction' revisited: A procedure for the synthesis of terminal vinyl bromides by the reduction of 1,1-dibromoalkenes

Article abstract of DOI:10.1016/S0040-4039(00)00353-1

A convenient procedure for the synthesis of vinyl bromides is described, which involves the selective reduction of the corresponding 1,1- dibromoalkenes with dimethylphosphite and triethylamine. The 1,1- dibromoalkenes are obtained in excellent yields fro

Full text of DOI:10.1016/S0040-4039(00)00353-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 3215–3219
The ‘Hirao reduction’ revisited: a procedure for the synthesis of
terminal vinyl bromides by the reduction of 1,1-dibromoalkenes
Sahar Abbas, Christopher J. Hayes ∗ and Stephen Worden
The School of Chemistry, The University of Nottingham, University Park, Nottingham NG7 2RD, UK
Received 17 December 1999; revised 14 February 2000; accepted 28 February 2000
Abstract
A convenient procedure for the synthesis of vinyl bromides is described, which involves the selective reduction
of the corresponding 1,1-dibromoalkenes with dimethylphosphite and triethylamine. The 1,1-dibromoalkenes
are obtained in excellent yields from the corresponding aldehydes via olefination with carbon tetrabromide and
triphenylphosphine. © 2000 Elsevier Science Ltd. All rights reserved.
(E)- and (Z)-Vinyl bromides are extremely useful intermediates in organic chemistry and the develop-
ment of methods for their stereoselective synthesis is of considerable importance. Their use as precursors
to vinyl anions (vinyllithiums, Grignards, etc.) and as coupling partners in a wide range of transition
metal-mediated coupling reactions has stimulated a great deal of interest in their synthesis. A number
of excellent one- or two-step procedures exist for the stereoselective synthesis of (Z)-vinyl bromides
from aldehydes, with this usually being accomplished via a Wittig olefination using bromomethylene
triphenylphosphorane.1 More recently, a two-step procedure, involving the stereoselective palladium-
catalysed reduction of intermediate 1,1-dibromoalkenes, has been reported.2 Unfortunately, however,
similar methodology for the efficient synthesis of (E)-vinyl bromides seems to be somewhat underdevel-
oped, although the bromoform/CrCl2 conditions of Takai perform this transformation well in a number
of favourable cases.3 In 1981 Hirao et al. reported a procedure for the reduction of 1,1-dibromoalkenes to
the corresponding vinyl bromides using diethylphosphite and triethylamine.4a Almost two decades have
passed since this initial report, and during this time very little attention has been paid to this potentially
valuable procedure.5 In this letter we wish to report our own studies in this area which have led to the
synthesis of a range of vinyl bromides via the reduction of 1,1-dibromoalkenes using dimethylphosphite
and triethylamine (Fig. 1).
Our interest in this particular transformation was stimulated by a purely serendipitous discovery. As
part of a research program examining the synthesis of oligonucleic acid analogues bearing modified
backbones,6 we had reason to study the palladium(0)-catalysed coupling reaction of the thymidine-
derived 1,1-dibromoalkene 1 with commercially available dimethylphosphite. Surprisingly, we found
that when the coupling was performed at 70°C the major product formed was the (E)-vinyl phosphonate
∗
Corresponding author. Tel: +44 (0)115 951 3045; fax: +44 (0)115 951 3564; e-mail: chris.hayes@nottingham.ac.uk (C. J.
Hayes)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00353-1
tetl 16633

3216
Fig. 1.
2 (Scheme 1). To our knowledge, only one previous attempt to achieve this particular transformation
has been reported in the literature and this was unsuccessful.4b We were intrigued by this result and
began to speculate about the nature of the reaction mechanism involved. The overall transformation
requires two processes to occur: (i) reduction of a carbon–bromine bond and (ii) cross-coupling with
dimethylphosphite. Initially, the order of these transformations was not known, but the isolation of small
amounts of the (E)-vinyl bromide 3 gave us circumstantial evidence that the reduction process occurred
first, followed by the palladium-catalysed cross-coupling reaction of the newly formed (E)-vinyl bromide
3. Confirmation of this hypothesis was secured when the coupling reaction was performed at room
temperature. Under these conditions the major product was the (E)-vinyl bromide 3, accompanied by
a small amount of the corresponding (Z)-isomer and no cross coupling was seen after 24 h.
Scheme 1.
Further studies revealed that the reduction reaction (viz 1–3) could be performed in the absence of the
palladium catalyst by treating 1 with dimethylphosphite and triethylamine in DMF at room temperature
for 24 h. This resulted in the production of the (E)-vinyl bromide 3, along with its (Z)-isomer, in 84% yield
as a 3:1 mixture of separable stereoisomers. This procedure is very similar to that described previously
by Hirao and co-workers, with the only major difference being that ours uses DMF as a co-solvent.4a
Due to the potential utility of this procedure for the synthesis of vinyl bromides, we began a study to
test the scope and limitations of this transformation using a more extensive series of 1,1-dibromoalkenes.
The 1,1-dibromoalkenes were synthesised in high yield from the corresponding aldehydes by reaction
with carbon tetrabromide and triphenylphosphine in methylene chloride.7 Each of these compounds was
then treated with dimethylphosphite (4 equivalents) and triethylamine (4.5 equivalents) in DMF at 70°C.
The reduction reactions were all complete after 16 h, and following aqueous workup, the desired vinyl
bromides were purified by column chromatography (Table 1). In general, the reduction products could
be isolated in moderate to high yield with a fair degree of E:Z selectivity.
In a few cases, however (entries 9, 10 and 11), we found that the isolated yields of the vinyl bromides
were quite low even though TLC and 1H NMR analysis of the crude mixtures indicated good conversions
and generally clean reactions. Further analysis of the reaction mixture obtained from the reduction of 11
revealed that in this case a major by-product (48%) was produced in the reaction, which was identified
1
as being the corresponding terminal acetylene 11c. Closer inspection of the H NMR spectra of the
crude reaction mixtures of 9 and 10 also revealed trace amounts of the corresponding terminal acetylenic

3217
Table 1
products. In these latter cases, however, the acetylene-containing by-products were not isolated, as they
were extremely volatile.
A mechanism for the reduction was proposed by Hirao et al. in their original publication,4a which
involves halophilic attack8,9 of the 1,1-dibromoalkene by a dialkylphosphite anion. Protonation of the
resulting vinyl anion then gives the corresponding vinyl bromide. Initially, we were a little sceptical

3218
about certain aspects of this mechanism, especially the formation of an extremely basic vinyl anion, but
it does adequately account for the formation of the acetylenic by-products observed in our work (cf. the
Corey–Fuchs procedure).10
Hirao’s proposal:
Obvious alternative mechanisms could involve single electron transfer and/or radical
intermediates,11,12 so we decided to investigate the effects of single electron and radical traps on
the course of the reduction. We soon found that the presence of an excess of 1,3-dinitrobenzene, a
known single electron trap,8 had little effect on the course of the reduction, with no significant change in
yield, reaction time or product distribution being observed. On the basis of this observation we believe
that single electron transfer is not an important process in this transformation. To test for the presence
of radical intermediates, we next performed the reduction under an oxygen atmosphere. Oxygen was
bubbled through the reaction mixture, and the oxygen atmosphere (balloon) was maintained for the
duration of the reduction. Once again, no noticeable difference in the course of the reaction was observed
and no new products were detected. We also performed a series of experiments where the reaction
mixture was shielded from light (a potential radical initiator), and like the previous experiments this
also failed to inhibit the reduction process. We feel that these data combine to support a mechanism
involving anionic intermediates. These findings are consistent with the mechanism proposed by Hirao,
but another possibility exists, not involving vinyl anions, which involves an initial Michael-type addition
of dimethylphosphite anion, followed by protonation and elimination. Both mechanisms adequately
describe the formation of the vinyl bromide and acetylene products13 but at present we are unable to
distinguish between these two pathways and studies are continuing in this area.
In summary, this methodology should prove useful for the syntheses of a range of valuable vinyl
bromide building blocks, especially when the use of organometallic intermediates is prohibited. At this
point in time this procedure is only a partial solution to the problem of vinyl bromide synthesis, but it
should be regarded as a simple, inexpensive and complementary alternative to existing methods.

3219
Acknowledgements
We would like to thank The University of Nottingham and Pfizer Central Research for financial
support.
References
1. Matsumoto, M.; Kuroda, K. Tetrahedron Lett. 1980, 21, 4021 and references cited therein.
2. Uenishi, J.; Kawahama, R.; Shiga, Y.; Yonemitsu, O. Tetrahedron Lett. 1996, 37, 6759.
3. Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986, 108, 7408.
4. (a) Hirao, T.; Masunaga, T.; Ohshiro, Y.; Agawa, T. J. Org. Chem. 1981, 46, 3745; (b) Hirao, T.; Masunaga, T.; Yamada, N.;
Ohshiro, Y.; Agawa, T. Bull. Chem. Soc. Jpn. 1982, 55, 909.
5. Rossi, R.; Carpita, A.; Lippolis, V. Synth. Commun. 1991, 21, 333.
6. Abbas, S.; Hayes, C. J. Synlett 1999, 1124.
7. Ramirez, F.; Desai, N. B.; McKelvie, N. J. Am. Chem. Soc. 1962, 84, 1745.
8. Meijs, G.; Doyle, I. R. J. Org. Chem. 1985, 50, 3713.
9. Zefirov, N. S.; Makhon’kov, D. I. Chem. Rev. 1982, 82, 615.
10. Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769.
11. These possibilities were also proposed, and subsequently disproved in the analogous reduction of gem-dibromo-
cyclopropanes (see Ref. 8).
12. For the use of compounds containing P–H bonds in radical processes, see: Graham, S. R.; Murphy, J. A.; Coates, D.
Tetrahedron Lett. 1999, 40, 2415 and references cited therin.
13. When the vinyl bromide products were re-subjected to the reaction conditions, no formation of the corresponding acetylene
was observed. We therefore conclude that the acetylene products are formed from an intermediate formed during the course
of the reaction.