Generation and contrasting gas-phase reactivity of 2-(2-alkenylpyrrol-1-yl) phenoxyl and thiophenoxyl radicals

Article abstract of DOI:10.1039/b604727k

The pyrrolylacrylates 9 and 10 were synthesised and subjected to flash vacuum pyrolysis (FVP) at 650-700 °C to generate the radicals 11 and 18, respectively. The phenoxyl 11 underwent hydrogen capture to give a mixture of the phenol 12 and the pyrrolobenzoxazine 13 in low yields, which were also obtained by a Wittig reaction of the 2-formylpyrrole 14. The thiophenoxyl 18 gave a single major product in 41% yield which was identified as the pyrrolo[1,2-a]quinoline 17 by a sequence of NMR experiments. A mechanism for the formation of 17 by a rearrangement-sulfur extrusion sequence is proposed. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b604727k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Generation and contrasting gas-phase reactivity of
2-(2-alkenylpyrrol-1-yl)phenoxyl and thiophenoxyl radicals
J. I. G. Cadogan,*a Chandralal M. Hewage,b Hamish McNab,*b Andrew D. MacPherson,b Iain S. Nicolson,b
David Reedb and Ian H. Sadlerb
Received 31st March 2006, Accepted 25th April 2006
First published as an Advance Article on the web 18th May 2006
DOI: 10.1039/b604727k
The pyrrolylacrylates 9 and 10 were synthesised and subjected to flash vacuum pyrolysis (FVP) at
650–700 ◦C to generate the radicals 11 and 18, respectively. The phenoxyl 11 underwent hydrogen
capture to give a mixture of the phenol 12 and the pyrrolobenzoxazine 13 in low yields, which were also
obtained by a Wittig reaction of the 2-formylpyrrole 14. The thiophenoxyl 18 gave a single major
product in 41% yield which was identified as the pyrrolo[1,2-a]quinoline 17 by a sequence of NMR
experiments. A mechanism for the formation of 17 by a rearrangement–sulfur extrusion sequence is
proposed.
chosen rather than the corresponding O-allyl derivative in an
attempt to minimise the hydrogen atom flux during the FVP
Introduction
In previous work, we have shown that phenoxyl1 and
thiophenoxyl2 radicals can interact with adjacent acrylate groups
under flash vacuum pyrolysis (FVP) conditions to give benzofuran
(X = O) and benzothiophene (X = S) ring systems respectively in
good yield (Scheme 1). The loss of the entire ester function as
a thermal radical leaving group is apparently highly favourable
in these reactions and so we have explored the scope and
limitations of this process by incorporating acrylate groups and
radical generators into other molecular architectures. In this
paper we report examples in which the radical site (phenoxyl or
thiophenoxyl) and the acrylate unit are situated within an N-
arylpyrrole framework which we hoped would lead to the creation
of new seven-membered rings. In the event, these pyrolyses did not
give the products expected by analogy with Scheme 1 but instead
the behaviour of the phenoxyl and thiophenoxyl proved to be very
different (cf. ref. 3 and 4) and an efficient sulfur extrusion—rather
than ester extrusion—has been identified.
experiment.1 In both cases the Vilsmeier reaction gave two
products (5 and 7, and 6 and 8 respectively) due to competitive
substitution at the 2- and 3-positions of the pyrrole ring. These
products were easily separated by chromatography and the re-
quired 2-formylated products 5 and 6 were isolated in >70% yield.
The position of the formyl group was determined by comparison
of the 1H NMR spectra of the products with those of the known
spectra of 2- and 3-formyl-1-phenylpyrrole.5 The chemical shifts
of the 2-formyl protons at dH ca. 9.4–9.5 and those of the 3-formyl
protons at dH ca. 9.7–9.8 ppm are particularly characteristic.
Although the Wittig reactions were slow and required extended
reaction times, 9 and 10 were obtained in 45% and 60% yields
respectively after chromatography, exclusively as the E-isomers.
Scheme 1
Results and discussion
Our chosen substrates 9 and 10 were synthesised from the 1-
arylpyrroles 1 and 2 by successive O- or S-alkylation, Vilsmeier
formylation and Wittig olefination. The O-benzyl product 3 was
The electron impact (EI) mass spectrum of the benzyloxy
compound 9 is dominated by cleavage of the benzyl group (m/z 91,
100%) but there is also evidence of ionisation at the ester function
giving small peaks at M − 31 and M − 59. The corresponding
spectrum of the allylthio compound 10 shows initial loss of the
allyl group (M − 41, 81%) followed by loss of a fragment of m/z
aDepartment of Chemistry, University of Wales Swansea, Swansea, UK SA2
3PP
bSchool of Chemistry, The University of Edinburgh, West Mains Road,
Edinburgh, UK EH9 3JJ
2446 | Org. Biomol. Chem., 2006, 4, 2446–2451
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
32 to give an ion at m/z 226 (100%). This may be rationalised by
loss of the sulfur atom or by cleavage of methanol; the latter is
more likely since the subsequent loss of CO (m/z 198, 69%) is then
readily explained (Scheme 2).
The phenol 12 is probably formed by hydrogen atom capture by
the phenoxyl radical 11; such reactions are common in the gas-
phase chemistry of phenoxyl radicals.4 The source of the ether 13
is more ambiguous. It could be obtained by intramolecular con-
jugate addition of the radical centre of 11 onto the acrylate unit,
followed by hydrogen atom capture. Alternatively, the phenol 12
may be the sole primary product of the pyrolysis, but may exist in
equilibrium with the ether 13 by a heterolytic conjugate addition
mechanism. A control experiment (Scheme 3) showed that the
same mixture of 12 and 13 was formed when the aldehyde 14
was subjected to a Wittig reaction (see Experimental section),
which suggests that 12 and 13 can indeed exist in equilibrium
in solution, though they are stable to chromatography and can
be separated. It therefore appears that intermolecular hydrogen
capture4 by the phenoxyl is the most likely product-forming route
open to the radical derived from 9 and that processes related to the
efficient cyclisation observed in Scheme 1 cannot take place with
this system. In this context, the use of an O-benzyl (rather than an
O-allyl) derivative as the radical precursor may have contributed
to the low yields, since a benzyl radical leaving group is known to
minimise the hydrogen atom flux under FVP conditions.1
Scheme 2
In contrast, pyrolysis of the S-allyl derivative 10 at 650 ◦C (0.001
Torr) gave a single major product in 41% yield after dry-flash
chromatography on silica. It was clear from its 1H and 13C NMR
spectra that the 1,2-disubstituted pyrrole (AMX spin system) and
1,2-disubstituted aromatic (AKQX spin system) systems together
with the ester function were still present but only one of the
two alkene protons of the precursor 10 appeared in the product.
Remarkably, its mass spectrum (m/z 225, M+) suggested that the
sulfur atom had been lost in the pyrolysis process as well as the
allyl group and a hydrogen atom. This represents a very unusual
example of a pyrolysis in which the atom bearing the original
radical species is not present in the ultimate product. In addition,
NOE experiments showed that the methyl group of the ester
function was close in space to the pyrrole ring and not adjacent
to the aromatic system. The two structures, 16 and 17, which fulfil
these requirements, are both unknown compounds and could not
be distinguished from their routine NMR data or by comparison
with the NMR spectra of model compounds. For example, the
set of NOE data could be equally interpreted in terms of either
structure (Fig. 1).
Flash vacuum pyrolysis of the O-benzyl compound 9 at
650 ◦C (0.005 Torr) was disappointing; apart from the inevitable
formation of bibenzyl, only two significant products were obtained
in low yield (<10%) and these proved to be the phenol 12 and the
cyclic ether 13 (Scheme 3). Compound 12 could not be obtained
in pure form, but was identified by comparison with an authentic
sample (see below). The ether 13 was identified by its spectra;
in particular the 1H NMR spectrum showed three signals due to
aliphatic protons (in addition to the seven due to the pyrrole and
the benzene rings) which were shown to be a CH (dC 70.26), a CH3
(dC 51.88) and a CH2 (dC 38.72) by a 13C NMR DEPT experiment.
1
The H and 13C NMR chemical shifts of the CH group suggest
that it is adjacent to an oxygen atom, as required for structure 13.
The carbon connectivity is however different in these two
isomers and so a 2D-INADEQUATE spectrum was obtained
at natural abundance (40 mg) at 150 MHz, and optimized to
give responses from single and double bonded pairs of carbon
atoms. Interpretation was aided by prior identification of the
methine carbon resonances, and thus the quaternary carbon
resonances, from an HMQC 13C–1H correlation spectrum, and
by the symmetrical disposition of coupled pairs of carbon-13
doublets about the diagonal of the 2-D INADEQUATE spectrum.
The connectivity sequence obtained (Fig. 2) is consistent only with
structure 17.
Clear connectivities were observed for all carbons from the six-
membered ring through to carbon C although that from E to the
carbonyl carbon P was only identifiable from a very weak signal
for E at the expected position. The connectivity from C to B was
not observable; since the B and C carbon resonances are almost
superimposed the anti-phase signals of the central lines of the AB
system would effectively cancel. Structure 16 is clearly excluded
Scheme 3 Reagents and conditions: (i) FVP (650 ◦C, 0.005 Torr); (ii)
=
DMF–POCl3; (iii) Ph3P CHCO2Me.
This journal is The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2446–2451 | 2447
©

View Article Online
Fig. 1 NOE data (% enhancements) for the pyrolysis product of 10,
interpreted as structures 16 and 17.
Fig. 2 Carbon connectivity of 17 as revealed by an INADEQUATE
experiment.
Scheme 4 Reagents and conditions: (i) FVP (650 ◦C, 0.001 Torr).
an acrylate chain (as in Scheme 1) is sensitive to the structure of
the precursor and that deviations from the optimum examples
of benzofuran or benzothiophene formation is likely to result in
reduced efficiency of the cyclisation process.
since it requires an uninterrupted sequence of four quaternary
carbon atoms. The full assignment of the NMR spectra of 17 is
given in the Experimental section.
The formation of 17 from10 requires an unusualand unexpected
(yet efficient) rearrangement sequence, and a possible mechanism
is shown in Scheme 4. The favoured 6-exo-trig cyclisation of
the thiophenoxyl 18 to give 19 is followed by attack at the
pyrrole system and a neophyl-type rearrangement to transform
the connectivity of the precursor 10 into that of the product.
Neophyl-type rearrangements are well known in aromatic systems
(e.g. under FVP conditions6) yet apparently none of this type have
been observed before in the sparse radical chemistry of the pyrrole
ring system.7 We believe that 20 is then transformed directly into
21 as a single step. By analogy with Scheme 1, our previous
work suggests that the alternative formation of radicals with a
b-carbomethoxy group such as 22 would be expected to suffer
loss of the ester function leading to other products. For a similar
reason, we favour the loss of the sulfur atom as the HS radical via
23 and 24 (Scheme 4), rather than as elemental sulfur,8 because
this latter process might be expected to generate the radical 25
which should again lead to loss of the ester function.
Experimental
1H and 13C NMR spectra were recorded at 250 (or 200) and
63 (or 50) MHz respectively for solutions in [2H]chloroform
unless otherwise stated. Coupling constants are quoted in Hz.
13C NMR signals refer to CH resonances unless otherwise stated;
in most cases assignments were confirmed by appropriate DEPT
experiments. Mass spectra were obtained under electron impact
conditions.
The results of this study show that the intramolecular radical
reactions of pyrrol-2-ylacrylate species such as 11 and 18 under
FVP conditions are quite different from those of corresponding
benzenoid systems. The behaviour of the phenoxyl and thio-
phenoxyl species is quite different and one unusual case of
rearrangement and sulfur extrusion has been identified to give
17. We conclude that cyclisation of radicals onto the 2-position of
N-Arylpyrroles 1 and 2
A
mixture of the appropriate 2-substituted aminophenol
(33 mmol), 2,5-dimethoxytetrahydrofuran (3.96 g, 30 mmol) and
glacial acetic acid (15 cm3) in dioxane (30 cm3) was heated under
reflux for 4 h. The volatiles were then removed on a rotary
evaporator and the residue was partitioned between ether (60 cm3)
2448 | Org. Biomol. Chem., 2006, 4, 2446–2451
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
and aqueous sodium hydroxide (3%, 90 cm3). The aqueous phase
was separated, acidified (pH 4) and extracted with chloroform
(3 × 50 cm3). The combined extracts were washed with sodium
hydrogen carbonate (1 M, 50 cm3), dried (MgSO4) and the solvent
was removed on a rotary evaporator. The crude product was then
purified by bulb to bulb distillation.
continued for 1 h. The mixture was then poured onto crushed
ice, hydrolysed with sodium hydroxide solution (2 M, 25 cm3)
and then acidified to pH 6–7 with hydrochloric acid (2 M). The
mixture was then extracted with ether (3 × 25 cm3), the organic
extracts were washed with water (3 × 50 cm3) and dried (MgSO4).
TLC showed that formylation had occurred at both the 2- and
3-positions of the pyrrole ring, so the products were pre-adsorbed
onto silica and separated by dry-flash chromatography (10% ethyl
acetate–hexane: 5% gradient).
2-Aminophenol (3.60 g, 33 mmol)◦ gave N-(2-hydroxy-
phenyl◦)pyrrole 1 (2.95 g, 61%), bp 138–140 C (0.03 Torr) (lit.,9 mp
45–47 C) (found: M+ 159.0687. C10H9NO requires M 159.0684);
3
dH 7.35–7.26 (2H, m), 7.11–6.98 (2H, m), 6.94 (2H, t, J 2.1)
N-[2-(Benzyloxy)phenyl]pyrrole 3 (0.75 g, 3 mmol) gave 2-
form◦yl-N-[2-(benzyloxy)phenyl]pyrrole 5 (0.64 g, 77%), bp 114–
118 C (3 Torr) (found: M+ 277.1103. C18H15NO2 requires M
277.1103); dH 9.49 (1H, s), 7.43–6.99 (11H, m), 6.42 (1H, dd, 3J 4.1
and 2.6) and 5.05 (2H, s); dC 178.97, 153.44 (quat), 136.17 (quat),
133.02 (quat), 130.98, 129.61, 128.27 (2CH), 127.99, 127.61,
126.55 (2CH), 120.87, 120.55 (quat), 113.61, 110.32 and 70.32
(CH2) (one CH overlapping); m/z 277 (M+, 15%), 248 (23), 158
(16) and 91 (100) and 3-formyl-N-[2-(benzyloxy)phenyl]pyrrole
3
and 6.44 (2H, t, J 2.1); dC 150.16 (quat), 128.70, 128.22 (quat),
126.58, 121.86 (2CH), 120.79, 116.81 and 110.12 (2CH); m/z
159 (M+, 100%), 158 (11), 131 (27), 130 (41), 103 (10) and
51 (21).
2-Aminothiophenol (6.25 g, 50 mmol) gave N-(2-mercap-
tophenyl)pyrrole 2 (6.28 g, 72%), bp 110–115 ◦C (0.05 Torr)
[lit.,10 bp 119–121 ◦C (1.0 Torr)]; dH 7.41–7.20 (4H, m), 6.85 (2H,
t, 3J 2.1), 6.39 (2H, t, 3J 2.2) and 3.41 (1H, s) spectrum consistent
with literature data.10
◦
7 (0.12 g, 15%), bp 145–150 C (2 Torr), (found: M+ 277.1103.
C18H15NO2 requires M 277.1103); dH 9.81 (1H, s), 7.61 (1H, t, 3J
3
1.7), 7.41–7.25 (7H, m), 7.14–6.98 (3H, m), 6.76 (1H, dd, J 3.1
N-[2-(Benzyloxy)phenyl]pyrrole 3, and
N-[2-(allylthio)phenyl]pyrrole 4—general method
and 1.6) and 5.12 (2H, s); dC 185.32, 151.48 (quat), 135.84 (quat),
130.74, 129.13 (quat), 128.74, 128.42 (2CH), 127.87, 126.92 (quat),
126.68 (2CH), 125.52, 124.85, 121.40, 114.26, 107.60 and 70.68
(CH2); m/z 277 (M+, 75%), 186 (26), 158 (32), 92 (20), 91 (100)
and 65 (30).
A suspension of potassium carbonate (1.1 equiv.) in DMF (10 cm3
per gram) was stirred for 10 min. The appropriate phenol or
thiophenol (1.0 equiv.) and the appropriate alkyl bromide (1.1
equiv.) were added and the mixture was stirred until TLC showed
the disappearance of the N-arylpyrrole. Water (2 cm3 per cm3
DMF) was added and the mixture was extracted with ether [3 ×
(volume of water/3)]. The combined organic extracts were washed
with water [3 × (volume of ether × 2/3)] and dried (MgSO4). The
solvent was removed on a rotary evaporator to yield the crude
product which was purified by bulb to bulb distillation.
N-[2-(Allylthio)phenyl]pyrrole 4 (0.75 g, 3 mmol) gave 2-for◦myl-
N-[2-(allylthio)phenyl]pyrrole 6 (0.57 g, 77%), bp 124–128 C (2
Torr) (found: M+ 243.0713. C14H13NOS requires M 243.0718); dH
9.42 (1H, s), 7.42–7.39 (2H, m), 7.28–7.24 (2H, m), 7.13 (1H, dd, 3J
4.0 and 1.6), 6.95 (1H, m), 6.43 (1H, dd, 3J 4.0 and 2.6), 5.73 (1H,
m), 5.16–5.02 (2H, m) and 3.40–3.35 (2H, m); dC 178.70, 137.99
(quat), 135.02 (quat), 132.84 (quat), 132.64, 130.84, 129.26, 129.10,
128.05, 126.29, 120.64, 118.20 (CH2), 110.63 and 35.65 (CH2);
m/z 243 (M+, 43%), 214 (71), 174 (97), 173 (59), and 170 (100)
and 3-formyl-N-[2-(allylthio)phenyl]pyrrole 8 (0.18 g, 23%), bp
136–140 ◦C (3 Torr) (found: M+ 243.0726. C14H13NOS requires M
243.0718); dH 9.79 (1H, s), 7.46–7.23 (5H, m), 6.84 (1H, m), 6.73
(1H, dd, 3J 2.7 and 1.5), 5.68 (1H, m), 5.10–4.98 (2H, m) and 3.32
N-(2-Hydroxyphenyl)pyrrole 1 (0.80 g, 50 mmol) with benzyl
bromide (1.03 g, 55 mmol) gave N-[2-(benzyloxy)phenyl]pyrrole
3 (1.05 g, 85%), bp 160–165 ◦C (0.005 Torr) (found C, 82.0;
H, 6.2; N, 5.65. C17H15NO requires C, 81.9; H 6.05; N, 5.6%);
3
3
dH 7.42–7.05 (9H, m), 7.15 (2H, dd, J 2.1), 6.42 (2H, dd, J
2.1) and 5.13 (2H, s); dC 151.62 (quat), 136.54 (quat), 130.95
(quat), 128.38 (2CH), 127.69, 127.18, 126.85 (2CH), 125.69, 121.97
(2CH), 121.46, 114.71, 108.70 (2CH) and 70.81 (CH2); m/z 249
(M+, 15%), 172 (20), 158 (42), 91 (100), 77 (23), 65 (28), 63 (12),
51 (22) and 39 (26).
3
(2H, d, J 6.8); dC 185.29, 139.00 (quat), 132.74 (quat), 132.46,
130.46, 130.27, 128.82, 127.16 (quat), 126.85, 126.69, 124.94,
118.30 (CH2), 108.00 and 36.03 (CH2); m/z 243 (M+, 19%), 202
(84), 175 (19), 174 (100) and 173 (51).
N-(2-Mercaptophenyl)pyrrole 2 (2.00 g, 11 mmol) gave N-[2-
(allylmercapto)phenyl]pyrrole 4 (2.46 g, 100%), bp 118–120 ◦C
(0.03 Torr) (found C, 72.3; H, 6.2; N, 6.35. C13H13NS requires C,
72.5; H, 6.1; N, 6.5%); dH 7.47–7.26 (4H, m), 6.91 (2H, t, 3J 2.2),
Wittig reactions—general method
The appropriate 2-formyl-N-arylpyrrole (2 mmol) was dissolved in
dry methylene chloride (50 cm3). Methyl (triphenylphosphoranyli-
dene)acetate (0.736 g, 2.2 mmol) was added and the mixture was
then heated under reflux until TLC showed that all the aldehyde
had been consumed. The mixture was pre-adsorbed onto silica and
purified by dry-flash chromatography (10% ethyl acetate–hexane:
10% gradient).
3
6.36 (2H, t, J 2.2), 5.86 (1H, m), 5.17–5.03 (2H, m) and 3.35–
3.30 (2H, m); dC 133.00, 132.78 (quat), 130.06, 127.50, 126.91,
126.53, 121.98 (2CH), 117.87 (CH2), 108.96 (2CH), 108.80 (quat)
and 35.86 (CH2); m/z 215 (M+, 8%), 175 (16), 174 (100), 173 (20),
45 (9), 41 (8) and 39 (18).
2-Formyl-N-[2-(benzyloxy)phenyl]pyrrole 5 (0.554 g, 2 mmol)
(48 h) gave methyl 3-{N-[2-(benzyloxy)phenyl]pyrrol-2-yl}pro-
penoate 9 (0.300 g, 45%), bp 150–155 ◦C (0.05 Torr) (found: M+
333.1374. C21H19NO3 requires M 333.1364); dH 7.42–7.20 (8H, m),
7.10–7.06 (2H, m), 6.92 (1H, m), 6.84 (1H, m), 6.38 (1H, m), 6.03
Formylation of N-arylpyrroles—general method
A solution of the appropriate N-arylpyrrole (3 mmol) in DMF
(5 cm3) was added to a solution of phosphoryl chloride (0.60 g,
3.9 mmol) in DMF (10 cm3). After stirring for 1 h a further portion
of phosphoryl chloride (0.60 g, 3.9 mmol) was added and stirring
3
(1H, m, J 15.8), 5.05 (2H, s) and 3.70 (3H, s); dC 167.97 (quat),
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2446–2451 | 2449
©

View Article Online
153.71 (quat), 136.25 (quat), 133.74, 130.22 (quat), 129.71, 128.93,
128.29 (2CH), 128.05 (quat), 127.61, 127.30, 126.58 (2CH), 121.13,
114.10, 112.03, 111.73, 110.05, 70.31 (CH2) and 51.14 (CH3); m/z
333 (M+, 35%), 274 (12), 170 (34), 168 (23), 154 (23) and 91 (100).
2-Formyl-N-[2-(allylthio)phenyl]pyrrole 6 (0.486 g, 2 mmol)
(8 h) gave methyl 3-{N-[2-(allylthio)phenyl]pyrrol-2-yl}prope-
Torr) (found: M+ 187.0644. C11H9NO2 requires M 187.0633); dH
9.42 (1H, s), 7.39–6.94 (4H, m) and 6.46–6.25 (3H, m) (OH not
apparent); dC 178.40, 150.39 (quat), 131.16, 129.29, 127.26, 120.13,
116.98, 113.92 and 110.78 (two quaternaries not apparent); m/z
187 (M+, 100%), 170 (28), 159 (93), 158 (50), 131 (23) and 130 (55),
and 3-formyl-N-(2-hydroxyphenyl)pyrrole 15 (0.058 g, 16%), bp
150–155 ◦C (0.5 Torr), (found: M+ 187.0625. C11H9NO3 requires M
187.0633); dH 9.69 (1H, s), 7.72 (1H, t, 4J 1.8), 7.30–6.91 (5H, m)
and 6.76 (1H, dd, 3J 3.0 and 4J 1.8) (OH not apparent); dC 186.45,
149.91 (quat), 131.11, 129.03, 128.88 (quat), 128.06 (quat), 125.42,
124.78, 120.51, 117.41 and 108.68; m/z 187 (M+, 92%), 170 (40),
159 (100), 158 (60), 130 (32) and 94 (27).
◦
noate 10 (0.342 g, 60%), bp 145–150 C (0.05 Torr) (found: M+
299.0984. C17H17NO2S requires M 299.0980); dH 7.41 (2H, m),
7.27–7.14 (3H, m), 6.85–6.80 (2H, m), 6.36 (1H, dd, 3J 3.3),
5.93 (1H, d, 3J 15.9), 5.81–5.64 (1H, m), 5.18–5.03 (2H, m),
3
3.67 (3H, s) and 3.40 (2H, d, J 6.6); dC 167.85 (quat), 137.36
(quat), 135.97 (quat), 133.08, 132.62, 129.88 (quat), 129.23, 128.88,
128.75, 127.04, 126.21, 118.24 (CH2), 112.48, 112.13, 110.33,
51.20 (CH3) and 35.40 (CH2); m/z 299 (M+, 41%), 258 (81),
226 (100), 199 (44), 198 (69), 197 (31), 186 (53), 167 (23) and
41 (22).
Methyl 3-{N-(2-hydroxyphenyl)pyrrol-2-yl}propenoate 12 and
methyl (4H-5-oxa-9b-aza-cyclopenta[a]naphthalen-4-yl)acetate 13
2-Formyl-1-(2-hydroxyphenyl)pyrrole 14 (0.070 g, 0.37 mmol)
was dissolved in dry THF (50 cm3). Methyl (triphenylphos-
phoranylidene)acetate (0.38 g, 1.12 mmol) was added and the
mixture was then heated under reflux until TLC showed that
all the aldehyde had been consumed (36 h). The mixture was
pre-adsorbed onto silica (1 g) and purified by dry-flash chro-
matography (20% ethyl acetate–hexane: 5% gradient). This gave
two products: methyl (4H-5-oxa-9b-aza-cyclopenta[a]naphthalen-
4-yl)acetate 13 (0.031 g, 34%), bp 130–135 ◦C (0.1 Torr) (found: M+
243.0883. C14H13NO3 requires M 243.0895); dH 7.33 (1H, m), 7.15
Flash vacuum pyrolysis experiments
The substrate was volatilised under vacuum through an electrically
heated empty silica tube (35 × 2.5 cm) and the products were
collected in a liquid nitrogen cooled U-tube, situated at the
exit point of the furnace. Conditions for the pyrolyses were
established in small-scale experiments in which the product(s) were
dissolved in a deuterated solvent and analysed immediately by 1H
NMR spectroscopy. The precursor, pyrolysis conditions [quantity
of precursor, furnace temperature (Tf), inlet temperature (Ti),
pressure range (P) and pyrolysis time (t)] and, where appropriate,
approximate yields are given.
3
3
(1H, dd, J 2.9 and 1.4), 7.09–7.00 (3H, m), 6.32 (1H, t, J 3.4),
3
4
3
6.02 (1H, dt, J 3.5 and J 1.3), 5.62 (1H, ddd, J 7.0 and 6.0,
4J 0.7), 3.77 (3H, s) and 3.00 (2H, m); dC 170.24 (quat), 144.85
(quat), 126.24 (quat), 126.07 (quat), 124.95, 122.28, 118.12, 114.91,
114.58, 110.51, 104.15, 70.25, 51.83 (CH3) and 38.72 (CH2); m/z
243 (M+, 30%), 171 (15) and 170 (100), and methyl 3-[N-(2-
hydr◦oxyphenyl)pyrrol-2-yl]propenoate 12 (0.024 g, 27%), bp 145–
150 C (0.1 Torr) (found: M+ 243.0887. C14H13NO3 requires M
243.0895); dH 7.33–7.12 (3H, m), 7.01–6.97 (2H, m), 6.87 (1H, m),
6.80 (1H, d, 3J 3.8), 6.38 (1H, t, 3J 3.0), 6.16 (1H, br), 5.97 (1H,
FVP of methyl 3-{N-[2-(benzyloxy)phenyl]pyrrol-2-yl}propenoate
9
Methyl 3-{N-[2-(benzyloxy)phenyl]pyrrol-2-yl}propenoate
9
(0.204 g, 6 mmol) (Tf 650 ◦C, Ti 140–160 ◦C, P 0.005 Torr,
t 20 min), gave a number of products, only three of which
could be separated by dry-flash chromatography. The first
to elute was bibenzyl. The second was methyl (4H-5-oxa-9b-
aza-cyclopenta[a]naphthalen-4-yl)acetate 13 (0.006 g, 4%), bp
3
d, J 15.7) and 3.66 (3H, s); m/z 234 (M+, 10%), 241 (13), 227
(21), 226 (78), 213 (86), 198 (100), 197 (75) and 183 (56). These
data are consistent with those reported above for the pyrolysis
of 9.
◦
120–125 C (0.05 Torr) (found: M+ 243.0879. C14H13NO3
requires M 243.0895); dH 7.33 (1H, m), 7.14 (1H, dd, 3J 2.9
4
3
and J 1.3), 7.05–7.00 (3H, m), 6.31 (1H, apparent t, J 3.2),
3
4
3
6.01 (1H, dt, J 3.5 and J 1.3), 5.61 (1H, apparent t, J 6.3),
3.76 (3H, s) and 2.99 (2H, m); dC 170.26 (quat), 144.88 (quat),
126.25 (quat), 126.09 (quat), 124.98, 122.31, 118.16, 114.90,
114.57, 110.48, 104.18, 70.26, 51.88 (CH3) and 38.72 (CH2);
m/z 243 (M+ 30%), 171 (13) and 170 (100). The third product
could not be isolated in a pure form but was identified as methyl
3-[1-(2-hydroxyphenyl)pyrrol-2-yl]propenoate 12 by comparison
with authentic data (see below); dH 7.06–6.99 (2H, m), 6.87 (1H,
m), 6.80 (1H, m), 6.35 (1H, t), 5.99 (1H, d, 3J 15.8) and 3.68 (3H,
s) two aryl protons, one alkenyl proton and OH not assigned; m/z
243 (M+).
FVP of methyl 3-{N-[2-(allylthio)phenyl]pyrrol-2-yl}propenoate
10
Methyl 3-{N-[2-(allylthio)phenyl]pyrrol-2-yl}propenoate 10
◦
◦
(0.250 g, 8 mmol) (Tf 650 C, Ti 140–160 C, P 0.001 Torr, t 20
min), gave one major product which was purified by dry-flash
chromatography (1% ethyl acetate–hexane: 10% gradient) and
identified as methyl pyrrolo[1,2-a]quino◦line-4-carboxylate 17 (see
discussion) (0.084 g, 41%) bp 125–130 C (0.2 Torr) (found: M+
225.0784. C14H11NO2 requires M 225.0790); dH (600 MHz,
[2H6]acetone, see Fig. 2 for atom labels) 8.11 (1H, dd, 3J 2.9 and
4J 1.5, proton B), 8.08 (1H, br. d, 3J 8.5, proton D), 7.87 (1H, s,
3
4
n
proton H), 7.83 (1H, ddt, J 7.8, J 1.4, J 0.7 and 0.7, proton
2-Formyl-N-(2-hydroxyphenyl)pyrrole 14
3
4
K), 7.63 (1H, ddd, J 8.5 and 7.2, J 1.4, proton L), 7.36 (1H,
Application of the general formylation procedure described above
to N-(2-hydroxyphenyl)pyrrole 1 gave, after dry flash chromatog-
raphy (15% ethyl acetate–hexane: 5% gradient), 2-formyl-1-(2-
hydroxyphenyl)pyrrole 14 (0.092 g, 25%), bp 134–139 ◦C (0.2
ddd, 3J 7.8 and 7.2, 4J 1.1, proton G), 7.21 (1H, dd, 3J 3.9 and 4J
3
1.5, proton A), 6.83 (1H, dd, J 3.9 and 2.9, proton C) and 3.95
(3H, s); dC (150 MHz, [2H6]acetone, see Fig. 2 for atom labels)
165.5 (quat P), 134.8 (quat M), 131.1 (L), 130.8 (K), 127.9 (quat
2450 | Org. Biomol. Chem., 2006, 4, 2446–2451
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
J), 125.2 (H), 124.5 (G), 122.3 (quat F), 121.0 (quat E), 115.0
(D), 113.7 (C), 113.6 (B), 105.6 (A) and 52.3 (CH3); m/z 225 (M+
100%), 194 (7), 168 (7), 167 (63), 166 (49), 140 (13) and 139 (12).
exponential window functions in both dimensions with zero filling
of the F1 data from 128 W to 512 W.
Acknowledgements
NMR spectroscopic analysis of 17
We are grateful to British Petroleum for a Research Studentship (to
A.D. MacP.) and to the EPSRC for the provision of the 600 MHz
NMR spectrometer.
The NMR spectra were measured on [2H6]acetone solutions using
a Varian INOVA 600 MHz spectrometer operating at 599.9 MHz
for protons and 150.9 MHz for 13C nuclei.
The 2-D proton detected one-bond 1H–13C correlation
(HMQC) spectra were obtained using the sequence:11 Dl–90◦(1H)–
D2–180◦(1H); 180◦ (13C)–D2–90◦(1H)–D3–90◦(1H)–D2–90◦(13C)–
t1/2–180◦(1H)–t1/2–90◦(13C)–D2–AQ. The delays used were Dl =
1.5 s, D2 = 3.7 ms (1/2 1JCH) and D3 = 1 s (to minimise
signals from protons bonded to 12C nuclei). The experiment was
preceded by 64 dummy scans to establish thermal equilibrium. A
4-step phase cycle (hypercomplex acquisition) was used with 13C
broad band decoupling during acquisition of the proton signals.
Other parameters were SW(1H) = 5000 Hz; 2 K data points;
400 increments; SW(13C) = 20000 Hz, AQ = 0.205 s. The data
were processed using shifted sine-bell squared functions in both
dimensions with zero filling of the F1 data from 400 W to 1024 W
before transformation.
The 2-D INADEQUATE spectrum was obtained on a 40 mg
sample of 17 over a period of 64 h using the pulse sequence:12
Dl–90◦–D2–180◦–D2–90◦-t1–90◦–AQ, with a 16 step phase cycle
(absolute value mode) and where Dl = 1.3 s (relaxation delay),
D2 = 5.55 ms (1/4 1JCC), AQ = 0.3 s (acquisition time) and t1 is the
incremented delay. WALTZ-16 broad band proton decoupling was
employed during AQ. Other parameters were SW2 = 23000 Hz;
13 K data points; SWl = 40000 Hz; 128 FIDs each with 144
transients. The data were processed using optimized decreasing
References
1 M. Black, J. I. G. Cadogan, H. McNab, A. D. MacPherson, V. P.
Roddam, C. Smith and H. R. Swenson, J. Chem. Soc., Perkin Trans. 1,
1997, 2483–2493.
2 J. I. G. Cadogan, A. D. MacPherson, H. McNab and A. A. Milligan,
unpublished work.
3 J. I. G. Cadogan, H. S. Hutchison and H. McNab, J. Chem. Soc., Perkin
Trans. 1, 1988, 2875–2879.
4 J. I. G. Cadogan, H. S. Hutchison and H. McNab, J. Chem. Soc., Perkin
Trans. 1, 1991, 385–393.
5 D. G. Durham and A. H. Rees, Can. J. Chem., 1971, 49, 136–138.
6 J. I. G. Cadogan, C. L. Hickson and H. McNab, Tetrahedron, 1986, 42,
2135–2165.
7 For leading references on radical reactions of pyrroles, see: T. C. T. Ho
and K. Jones, Tetrahedron, 1997, 53, 8287–8294; F. Aldabbagh, W. R.
Bowman, E. Mann and A. M. Z. Slawin, Tetrahedron, 1999, 55, 8111–
8128; L. D. Miranda, R. Cruz-Almanza, A. Alvarez-Garc´ıa and J. M.
Muchowski, Tetrahedron Lett., 2000, 41, 3035–3038.
8 cf. J. I. G. Cadogan, J. B. Husband and H. McNab, J. Chem. Soc.,
Perkin Trans. 2, 1983, 697–701.
9 M. Artico, G. C. Porretta and G. De Martino, J. Heterocycl. Chem.,
1971, 8, 283–287.
10 D. K. Bates, R. T. Winters and B. A. Sell, J. Heterocycl. Chem., 1986,
23, 695–699.
11 M. F. Summers, L. G. Marzilli and A. Bax, J. Am. Chem. Soc., 1986,
108, 4285–4294.
12 T. H. Mareci and R. Freeman, J. Magn. Reson., 1982, 48, 158–163.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2446–2451 | 2451
©