sequence through an allenolate formation, further 1,3-O-
transposition, and enolate coupling leading to unsaturated
â-hydroxy ketones 5 as the main products. In this sequence,
an oxonia-[3,3]-sigmatropic rearrangement over the oxocar-
benium ion 2, as competitive alternative pathway to the Prins
cyclization, leads to the key allenolate 3 (Scheme 1).5 The
Table 1. Silylalkyne-Prins Cyclization of Secondary
Trimethylsilyl Homopropargylic Alcohols and Aldehydes Using
FeX3 as a Promoter
Scheme 1. Coupling of Homopropargylic Alcohols and
Aldehydes Catalyzed by Iron(III) Halides
entry
R1
Me
Me
Me
Me
Me
Et
Bn
R2
R3
X
6/7
yield (%)
1
2
3
4
5
6
7
8
9
H
H
H
H
H
H
H
H
H
i-Bu
Cl
50:50
80
65
c-C6H11
Ph
Cl >99:1c
Cl
s-Bu
s-Bu
Cl >99:1c
Br >99:1c
Cl >99:1c
Cl >99:1c
Cl >99:1c
Cl >99:1c
82
75
75
57
55
61
60
62
c-C6H11
c-C6H11
c-C6H11
Bn
c-C6H11
c-C6H11
Me
10a
11b
Me Br2CdCH(CH2)- Cl >99:1c
Me c-C6H11
Cl >99:1c
Me
a Racemic (2R*,3R*)-3-methyl-5-(trimethylsilyl)pent-4-yn-2-ol was used
as 1. b Racemic (2S*,3R*)-3-methyl-5-(trimethylsilyl)pent-4-yn-2-ol was
used as 1. c Within the NMR detection limit, 7 was not observed.
reaction works well with a wide range of aldehydes except
benzaldehyde (entry 3). However, other aldehydes containing
aromatic rings, although located on a distal position relative
to the carbonyl group, proceeded satisfactorily (entry 9).
The presence of the silyl group at the alkyne is essential
to achieve the reaction since when the acetylene is substituted
with a methyl group the process is inhibited. In addition,
the size of the substituent at the silicon atom is also a critical
factor. For instance, when the triple bond bears a triisopropyl
silyl group instead of TMS, the reaction does not take place.
The alkyne silyl-Prins reaction proceeds at room temper-
ature and with complete diastereocontrol, obtaining exclu-
sively the cis-2,6-dihydropyran product. An exception was
found for 3-methylbutanal (entry 1).
With this methodology in hand, we can access tetra- and
pentasubstituted dihydropyrans in one step using the suitable
secondary trimethylsilyl homopropargylic alcohol. The alkyne-
Prins cyclization worked in good yields, leading to the
pentasubstituted dihydropyrans, when 3-methyl-5-(trimeth-
ylsilyl)pent-4-yn-2-ol was used as starting material (entries
10 and 11).
To account for the stereochemical course of the reaction,
“ab initio” theoretical calculations at the B3LYP/6-31G(d)
level were performed for some cases of 6 and 7 (Table 2).
The calculations showed that the cis stereoisomers 6 are
always more stable than trans isomers 7. The small energetic
difference for both stereoisomers when R3 ) i-Bu accounts
for the only exception (50:50) (Table 1, entry 1) relative to
the general stereochemical course of the reaction.
As pointed out above, the presence of the trimethylsilane
is essential to generate the corresponding dihydropyran.
Surprisingly, the obtained halosilylalkene proved to be a very
unreactive system. For instance, all standard conditions
course of the reaction (rearrangement or Prins type cycliza-
tion) depends directly on the stability of the species involved
in this rearrangement.
Considering the well-documented silyl-modified Prins
cyclization,6 we decided to explore the introduction of a
trimethylsilyl group in the triple bond (R4 ) TMS), as a
way to minimize the 2-oxonia-[3,3]-sigmatropic rearrange-
ment in the above-mentioned equilibrium.7
In this paper, we describe a new, general, and stereose-
lective method to obtain tetra- and pentasubstituted dihy-
dropyrans based on an alkyne silyl-Prins cyclization cata-
lyzed by iron(III) halides. Ab initio theoretical calculations
of the species involved in the rearrangement support our
results and the proposed mechanism.
First, to test the scope of FeX3 in this coupling reaction,
we carried out the reaction between secondary trimethylsilyl
propargyllic alohols (1, R4 ) TMS, R2 * H) and several
8
aldehydes using FeCl3 and FeBr3 as promoters.9 Table 1
summarizes the results obtained in this study.
The methodology produced exclusively and in good yield
the desired six-membered rings. No traces of the products
resulting from the oxonia rearrangement were observed. The
(5) Miranda, P. O.; Ram´ırez, M. A.; Padro´n, J. I.; Mart´ın, V. S.
Tetrahedron Lett. 2006, 47, 283-286.
(6) Silyl-modified Prins cyclization: (a) Semeyn, C.; Blaauw, R. H.;
Hiemstra, H.; Speckamp, W. N. J. Org. Chem. 1997, 62, 3426-3427. (b)
Viswanathan, G.; S.; Yang, J.; Li, C.-J. Org. Lett. 1999, 1, 993-995.(c)
Dobbs, A. P.; Martinovic, S. Tetrahedron Lett. 2002, 43, 7055-7057. (d)
Dobbs, A. P.; Guesne, S. J. J.; Martinovic, S.; Coles, S. J.; Hursthouse, M.
B. J. Org. Chem. 2003, 68, 7880-7883. (c) Meilert, K.; Brimble, M. A.
Org. Lett. 2005, 7, 3497-3500.
(7) Lee, K.-C.; Lin, M.-J.; Loh, T.-P. Chem. Commun. 2004, 2456-
2457.
(8) FeCl3 and FeBr3 were purchased form the Aldrich Chemical Co.
(9) For availability or synthesis of 1, see the Supporting Information.
1634
Org. Lett., Vol. 8, No. 8, 2006