Angewandte
Chemie
DOI: 10.1002/anie.200905167
Radical Reactions
Radical Cyclization of a-Bromo Aluminum Acetals: An Easy
Approach to g-Lactols**
Anne Boussonniꢀre, Fabrice Dꢁnꢀs,* and Jacques Lebreton*
Aluminum acetals are well-known intermediates whose
potential in organic synthesis is a subject of growing interest.
The most common access to aluminum acetals involves the
reduction of carboxylic acid esters with diisobutylaluminium
hydride (DIBAL-H) at À788C. Their stability at low temper-
ature allows the selective reduction of carboxylic acid esters
into the corresponding aldehydes with limited over-reduction
into alcohols.[1,2] Since the 1980s, several applications of
aluminum acetals have been reported. The reduction of esters
using DIBAL-H at low temperature has been used to form
highly unstable or epimerizable aldehydes in situ. The latter
are formed either upon warming or by the addition of a protic
source or Lewis acid. The resulting aldehydes have been used
in carbon–carbon bond-forming reactions with various nucle-
ophiles such as allylstannanes,[3,4] silyl ketene acetals,[5–7] or
lithium enolate.[8] Olefination[9,10] and the preparation of
alkynes[11] have also been described. The addition of Grignard
reagents to aluminum acetals obtained from a-amino esters
through reduction with DIBAL-H or DIBAL-H/iBu3Al have
been reported to give 1,2-amino alcohols with a high level of
stereoselectivity.[12] A SN2-like displacement of the aluminum
acetal, or alternatively a displacement of a tight ion pair, has
been proposed to explain the chiral induction.[12] Aluminum
acetals have been successfully trapped by silyl triflates and
silyl imidazole or acetic anhydride to give the corresponding
monosilyl acetals[13–15] and a-acetoxy acetals,[16,17] respectively.
Acid fluorides have also been reported as efficient reagents to
trap aluminum acetals.[18] Monosilyl acetals have been
employed in condensation reactions with allyl trimethylsi-
lane.[14] In addition, a-acetoxy acetals have proven to be
useful precursors of the oxacarbenium ion, which can be
trapped by various nucleophiles, including allylstannanes
reagents,[16,19] trimethylsilylcyanide,[16] dialkylaluminum (tri-
methylsilyl)acetylide,[16] triethylsilane,[17] thiophenol,[16] and
phenylselenol.[20] Moreover, the oxacarbenium ion has been
employed in Prins[21] and oxonia-Cope[22] rearrangements to
give polysubstituted tetrahydropyrans. Note that contrary to
the Prins rearrangement of a-acetoxy acetals derived from
homoallylic alcohols, which gave excellent results both in
terms of regio- and stereoselectivity, the related allyl alcohols
failed to give the corresponding tetrahydrofuran deriva-
tives.[21c]
In spite of the aforementioned applications of aluminum
acetals in ionic processes, their use as intermediates in radical
reactions has thus far remained completely unexplored.
Herein, we report the formation of a-bromo aluminum
acetals derived from O-allyl-a-bromo esters and their cycli-
zation in situ under radical conditions to give a variety of
polysubstituted g-lactols.
Radical reactions have been intensively investigated
during the last two decades.[23] The new synthetic methods
that have been developed from this work are characterized by
their mildness and their complementarity to ionic processes.
The cyclization of a-haloacetals (the Ueno–Stork reaction)[24]
was developed independently by Ueno[25] and Stork[26] in the
1980s, and has become a very popular approach for the
cyclization of related a-bromo esters under reductive con-
ditions.[27,28] The resulting cyclic acetals have proven to be
useful precursors for the preparation of corresponding
lactones—even though this transformation often implies the
formation of the hemiacetal intermediate under strongly
acidic conditions, which limits this approach to substrates
without acid-sensitive functional groups. Moreover, the syn-
thesis of the a-haloacetal precursors, as well as the enol ethers
employed in their preparation, is not always straightforward
and can be somewhat tricky—we have experienced difficulty
during the course of our research on the synthesis of
biologically relevant heterocyclic compounds. For instance,
attempts to prepare a-bromo acetal 3 from propargylic
alcohol 1 and enol ether 2 (contaminated with significant
amounts of acetal iPrCH(OEt)2, which was used for its
preparation) under standard reaction conditions gave irre-
producible results. A mixture of three compounds was
typically obtained and, beside the expected a-bromo acetal
3, which was usually obtained as the minor compound, bromo
ether 4 and acetal 5 were also isolated (Scheme 1).
[*] A. Boussonniꢀre, Dr. F. Dꢁnꢀs, Prof. J. Lebreton
Universitꢁ de Nantes, Laboratoire CEISAM
UMR 6230, UFR des Sciences et des Techniques
2, rue de la Houssiniꢀre
BP 92208, 44322 Nantes Cedex 3 (France)
Fax: (+33)2-5112-5402
E-mail: fabrice.denes@univ-nantes.fr
These observations prompted us to develop a more
practical approach to g-lactols based on a one-pot reaction
involving the formation of an a-bromo aluminum acetal and
its cyclization under reductive radical conditions at low
temperature. We carried out a feasibility study in which
various a-bromo aluminum acetals were generated by
reduction of the parent a-bromo esters using DIBAL-H,
and were subsequently subjected to radical cyclization con-
ditions.
[**] This work was supported by the French Ministry of Science. We
would like to thank Nathali Henriques, Christian Duchamp, and Dr.
Denis Bouchu (Centre Commun de Spectroscopie de Masse—
Universitꢁ Claude Bernard Lyon 1—France) for HRMS measure-
ments.
Supporting information for this article is available on the WWW
Angew. Chem. Int. Ed. 2009, 48, 9549 –9552
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9549