736
S. J. Connon, A. F. Hegarty / Tetrahedron Letters 42 (2001) 735–737
SiMe3
1. LDA, THF, -78oC
2. TMSCl, 4.0 eq.
4
N
Cl
Scheme 3.
pyridines. The decrease in adduct yield as the C-4
substituent is changed from methoxy to the relatively
weaker electron donating phenoxy and thiophenoxy
groups indicates that electron donation into the pyri-
dine ring system is the dominant factor in pyridyne
stabilisation by these substituents. These reactions were
relatively clean in that the only other products detected
were 2-diisopropylamino-4-alkoxypyridines (derived
from attack of diisopropylamine on the 2,3-pyridyne
intermediate) and starting material. To our knowledge
the strong stabilisation exhibited by the p-methoxyphe-
noxy group resulting in a 58% yield of 1413 represents a
significant improvement over present literature methods
and is unexpected when compared with the results
obtained when the methoxy group is directly attached
to the 4-position (as is the case with 5). This could be
explained in terms of the electron-rich 4-methoxy sub-
stituted benzoid ring encouraging preferential electron
donation towards the ring nitrogen by the pyridyl-ether
oxygen, as depicted by 15.
Scheme 2.
Synthesis of the required precursors was possible from
simple inexpensive starting materials. Chlorination of
commercially available 4-picoline-N-oxide with phos-
phorous oxychloride at 100°C overnight gave 2-chloro-
4-methylpyridine 4 in 64% yield. 2-Chloro-4-alkoxy-
and 4-thiophenoxypyridines 5–8 were conveniently pre-
pared by treating 2-chloro-4-nitropyridine11 with
sodium alkoxide or thiophenoxide at 100°C (Scheme 2).
Attempted generation of substituted 2,3-pyridynes from
precursors 3–8 using n-, sec-, or tert-butyllithium failed
due to competing substitution and halogen-exchange
reactions. For example, treatment of 6 with tert-butyl-
lithium at −78°C followed by quenching with water
gave 4-phenoxypyridine in quantitative yield, pre-
sumably via halogen–metal exchange at C-2. This prob-
lem could be avoided using the sterically hindered
lithium diisopropylamide (LDA) as the lithiating
reagent, which does not undergo halogen–metal
exchange reactions in these systems. The results of
lithiation of these precursors with LDA at −78°C in
THF and subsequent trapping with furan12 are pre-
sented in Table 1.
OMe
N
O
15
The failure of 2-chloropyridine 3 to give trapped adduct
is not surprising, the main product arises from nucle-
ophilic attack on 1 by its lithiated precursor, giving a
resinous black tar. The unsuitability of 4 as a 2,3-pyri-
dyne precursor is due to the acidity of the methyl group
protons, this was confirmed by trapping the intermedi-
ate lithio species with trimethylsilylchloride (TMSCl),
giving silylation of the methyl group as the only
product (Scheme 3).
The novel use of a sulphur based stabilising group 7
holds particular promise given the possibility of cleav-
age of the sulphide functionality after cycloaddition.14
Further ‘tuning’ of the 4-aryloxy and thiophenoxy sub-
stituents and the use of more sterically hindered bases
should further increase the synthetic utility of these
species. Work along these lines is currently in progress
in our laboratory.
Reasonable to good yields of cycloadduct were
obtained using 2-chloro-4-alkoxy- and 4-thiophenoxy-
Acknowledgements
Table 1. [4+2] cycloadditions of 4-substituted 2,3-pyridynes
with furan
S.C. is grateful to Bristol Myers-Squibb Corporation,
BOC gases and the Irish–American Partnership for
financial support.
Entry
Adduct
Yield (%)a
3
4
5
6
7
8
9
10
11
12
13
14
0
0
37
25
29
58
References
1. Hart, H. Supplement C2, The Chemistry of Triple Bonded
Functional Groups; Wiley: Chichester, 1994; p. 1113–
1134; Reinecke, M. G. Tetrahedron 1982, 38, 427–498.
a Refers to isolated yields after chromatography.