S. J. Pridmore et al. / Tetrahedron Letters 48 (2007) 5115–5120
5117
Table 1 (continued)
Entry Starting material
OH
Product
Conversionb (%) Diketone 3b (%) Furan 4b (%) Pyrrole 5b (%)
F
10
N
100
100
9
0
25
66
91
Ph
F
OH
OH
CN
11
N
9
Ph
NC
OH
OH
CO2Et
N
12
100
100
100
0
20
0
37
10
14
63
70
86
Ph
EtO2C
OH
OH
13
N
Ph
OH
OH
O
14
N
O
Ph
OH
a Reaction conditions: 1,4-Alkynediol (1 mmol), Ru(PPh3)3(CO)H2 (2.5 mol %) and Xantphos (2.5 mol %) were dissolved in dry PhMe (1 mL) and
heated to reflux. After 30 min, amine (2 mmol) was added to the reaction mixture. After 24 h, the reaction mixture was cooled, diluted with
MeOH:PhMe (1:1, 10 mL) and injected into the GC–MS without further purification.
b Determined by GC–MS.
c PhCH2N@CHPh also formed as a minor by-product.
nor enhanced reactivity. We had hoped that in those
cases where furan formation was a significant problem,
that the presence of base may deter this process, but it
had essentially no effect on the product distribution,
and we therefore did not add base (or acid) to these
reactions.
We also wished to investigate the range of amines that
could be used in these pyrrole-forming reactions. In
reactions to form the 2,5-dimethyl substituted pyrroles,
we were pleased to find that aniline could be used as
the amine (Table 2, entry 1), although some diketone
remained uncyclised under these conditions, probably
reflecting the lower nucleophilicity of anilines with
respect to aliphatic amines. The aniline containing the
strongly electron-withdrawing p-nitro group failed to
provide any of the corresponding pyrrole product
(Table 2, entry 3). However, as expected, the aniline
containing the electron-donating p-methoxy group (Table
2, entry 4) reacted with excellent selectivity towards
pyrrole formation. 2-Aminopyridine (Table 2, entry 8)
was also a problematic amine, giving an unsatisfactory
conversion into pyrrole under these reaction conditions.
Optimisation of the reaction conditions showed that
addition of amine after 30 min of heating the catalyst,
ligand and starting material in toluene was ideal. There-
fore the 1,4-alkynediols, shown in Table 1, were treated
with Ru(PPh3)3(CO)H2 (2.5 mol %) and Xantphos
(2.5 mol %) and heated at reflux in toluene for 30 min,
prior to the addition of two equivalents of the amine
(benzylamine or 2-phenylethylamine). Heating at reflux
was then continued to give a total reaction time of
24 h.10
Formation of pyrroles possessing a 2-(p-chloro)-substi-
tuent was generally less selective towards pyrrole forma-
tion, although reasonable conversions were obtained in
favourable cases, where the amine was an unbranched
aliphatic primary amine (Table 2, entries 12, 13, 16).
The use of a branched aliphatic amine provided an
acceptable conversion for 1-phenylethylamine (Table 2,
entry 14) but not in the case of isopropylamine, which
may be due to an issue of volatility (bp 34 °C). The
The reactions were highly selective for pyrrole formation
from unhindered substrates, leading to the formation of
2,5-dimethylpyrroles (Table 1, entries 1 and 2). More
hindered substrates still afforded pyrroles selectively,
although in some cases, the corresponding furan was a
significant by-product (Table 1, entries 5–7, 10, 12).
However, many functional groups were tolerated,
including halide, nitrile, ester and furyl.