An exploratory study of ring closures of aryl radicals onto cyclopropyl- and oxiranyl-isocyanate acceptors

Article abstract of DOI:10.1039/b405735j

The idea that ring closures of C-centred radicals onto isocyanates could be made permanent by designing the cyclised radical to undergo a rapid onward β-scission, was investigated for the 2-(2-isocyanato)cyclopropylphenyl and 2-(2-isocyanato)oxiranylphenyl radicals. The radical precursors, trans- and cis-l-bromo-(2-isocyanatocyclopropyl)benzene and (2-bromophenyl)-3- isocyanatooxirane, were prepared from the corresponding bromophenylcyclopropane and bromophenyloxirane carboxylic acids via Curtius rearrangements of the derived azides. The structure of the trans-2-(2-isocyanato)cyclopropylphenyl radical prevents cyclization, however, it was shown that isomerisation to the analogous cis-radical occurred, probably by scission of the disubstituted cyclopropane bond followed by internal rotation of the resulting resonance stabilised diradical. It was found, however, that the main product from homolytic reactions of both trans- and cis-isocyanatocyclopropyl compounds, with tributyltin hydride and tris(trimethylsilyl)silane, was the direct reduction product, trans-(2-isocyanatocyclopropyl)benzene. Only traces of cyclised products, that were probably 4,5-dihydrobenzo[c]azepin-1-one from the cyclopropane precursor and 5H-6-oxa-8-azabenzocyclohepten-9-one from the oxirane precursor, were detected. We conclude, therefore, that the rate of cyclization onto isocyanate acceptor groups must be slower in these systems than hex-5-enyl cyclization or that the reverse ring-opening process must be faster than for analogous radicals.

Full text of DOI:10.1039/b405735j

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
An exploratory study of ring closures of aryl radicals onto
cyclopropyl- and oxiranyl-isocyanate acceptors
Patricia L. Minin and John C. Walton*
University of St. Andrews, School of Chemistry, St. Andrews, Fife, UK KY16 9ST.
E-mail: jcw@st-and.ac.uk; Fax: +(0)1334 463808; Tel: +(0)1334 463864
Received 16th April 2004, Accepted 15th June 2004
First published as an Advance Article on the web 4th August 2004
The idea that ring closures of C-centred radicals onto isocyanates could be made permanent by designing the cyclised
radical to undergo a rapid onward -scission, was investigated for the 2-(2-isocyanato)cyclopropylphenyl and 2-(2-
isocyanato)oxiranylphenyl radicals. The radical precursors, trans- and cis-1-bromo-(2-isocyanatocyclopropyl)benzene
and (2-bromophenyl)-3-isocyanatooxirane, were prepared from the corresponding bromophenylcyclopropane and
bromophenyloxirane carboxylic acids via Curtius rearrangements of the derived azides. The structure of the trans-2-(2-
isocyanato)cyclopropylphenyl radical prevents cyclization, however, it was shown that isomerisation to the analogous
cis-radical occurred, probably by scission of the disubstituted cyclopropane bond followed by internal rotation of the
resulting resonance stabilised diradical. It was found, however, that the main product from homolytic reactions of both
trans- and cis-isocyanatocyclopropyl compounds, with tributyltin hydride and tris(trimethylsilyl)silane, was the direct
reduction product, trans-(2-isocyanatocyclopropyl)benzene. Only traces of cyclised products, that were probably 4,5-
dihydrobenzo[c]azepin-1-one from the cyclopropane precursor and 5H-6-oxa-8-azabenzocyclohepten-9-one from the
oxirane precursor, were detected. We conclude, therefore, that the rate of cyclization onto isocyanate acceptor groups must
be slower in these systems than hex-5-enyl cyclization or that the reverse ring-opening process must be faster than for
analogous radicals.
amidyl radical 8a is more thermodynamically stabilised. Radical
Introduction
8a contains a cyclopropylaminyl structural unit that is expected to
Scarcely any preparations of N-heterocycles via cyclizations of -
ring open very rapidly by -scission of its inter-ring bond to afford
isocyanato-C-centered radicals have been reported. The reason for
the dihydrobenzoazepinonyl radical 9a. Radical 9a is a resonance
this neglect can be traced to the rapid reverse ring opening process
stabilised benzyl type radical and therefore its formation will be
which counteracts the cyclisation. For example, the 2-(isocyanato-
favoured over that of the alternative six-membered ring dihydro-
carbonyl)ethyl radical cyclised rapidly to the succinimidyl radical
isoquinolinonylmethyl radical that would result from scission of
but the rate of ring opening was almost as fast.1,2 Similarly, the 3-(iso
the outer cyclopropane bond.4,5 It was anticipated that the rapid
cyanatocarbonyl)propyl radical afforded the glutarimidyl radical; but
formation of 9 would ensure that formation of dihydrobenzo-
this 6-endo-type process was reversible.2 We showed recently that the
azepinones would supersede the reverse C6n step.
2-(2-isocyanatophenyl)ethyl radical (1) underwent ring closure at the
central C-atom, in the endo mode, to afford acylaminyl radical 2 and,
to a minor extent, at the N-terminus to produce aminoacyl radical 4
(Scheme 1).3 The resonance stabilisation of radical 2 reinforced the
ring closure and inhibited the reverse ring opening so that substantial
amounts of cyclised products could be isolated (Scheme 1).
Scheme 2 Domino cyclisation and ring opening of 2-(2-isocyanato)-
cyclopropylphenyl radicals.
To test the practicality of this idea we prepared 1-bromo-(2-
isocyanatocyclopropyl)benzene 13 and (2-bromophenyl)-3-
isocyanatooxirane 26 as radical precursors and examined their
Scheme 1 Cyclisation of the 2-(2-isocyanatophenyl)ethyl radical.
reactions with several radical generating reagents.
Another potential way of avoiding reversal of ring closure
onto –NCO groups might be to offer the cyclised species an alter-
native, more rapid, onward process. For example, if the radical
Results and discussion
centre in the cyclised species was created adjacent to a three-
membered ring, opening of the strained ring should supervene,
thus preserving the initial cyclised structure. The system shown in
Scheme 2 was designed to test out this idea. The 2-(2-isocyanato)
cyclopropylphenyl radical 6 might ring close to afford carbamoyl
radical 7a. However, literature precedents1,3 indicated that the main
cyclisation mode would be 6-endo (C6n) to generate the dihydro-
isoquinolinonyl radical 8a. Aryl radical 6 will be electronically
neutral, unlike the nucleophilic alkyl radicals. Ring closure onto
the electrophilic C-atom of NCO should also be favoured because
Trans-bromophenylpropenoate 10 was made by palladium-
catalysed coupling of methyl acrylate with 2-bromoiodobenzene
under phase transfer conditions with tetra-n-butyl ammonium
chloride.6 Cyclopropanation of 10 with diazomethane in the
presence of 0.005 equivalents of palladium(II) acetate afforded the
trans-ester 11 in high yield after purification on a silica column
(Scheme 3). The first attempt at this cyclopropanation gave a
mixture of trans-11 together with some cis-isomer. However, the
reaction was subsequently repeated many times and gave solely
trans-11. The observation of some cis-isomer was probably due
T h i s j o u r n a l i s
©
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 4 7 1 – 2 4 7 5
2 4 7 1

View Article Online
are thermodynamically stabilised. Internal rotation about one of
the CH2–CH˙ bonds will convert 14 to the cis-diradical 15, ring
closure of which will provide the cis-precursor 16.Asimilar thermal
interconversion is feasible for the trans- and cis-radicals 6a and
17. 6-endo-ring closure, and/or 5-exo-ring closure, of cis-radical
17 would then be favourable. The interconversion process might
lead to thermodynamic control of the cyclization. To encourage
this isomerization, a photolysis was carried out using 13 in neat n-
Bu3SnH at 150 °C. At this temperature the main product was again
21 from simple reduction. However, the GC-MS analysis showed
that the residual 13 had isomerised to a mixture of trans- and cis-
isomers, i.e. that the cyclopropane rearrangement of Scheme 3 had
occurred to some extent. The product analysis also showed a minor
amount of 2-phenylcyclopropylamine which suggested that some
hydrolysis of 21 occurred under these forcing conditions.
The presence of residual cis-16 in this experiment suggested
that the cyclization step did not compete effectively with hydrogen
atom transfer from the tin hydride. To examine this conclusion more
thoroughly we set out to prepare the cis-cyclopropyl-isocyanate 16
directly. cis-Bromophenylpropenoate 23 was prepared in low yield
by reaction of trifluoroethylphosphonoester 2211 with 2-bromo-
benzaldehyde using the optimal base system KN(TMS)2/18-crown-
6-acetonitrile (Scheme 4).12 The cis/trans mixture was converted
to the corresponding cyclopropylisocyanate, following the same
procedure as for the trans-isomer, and a 2.5:1 mixture of cis:trans
cyclopropanes was obtained. The radical cyclization of this cis/
trans mixture was attempted with n-Bu3SnH but showed no cyclised
products. Reaction with TTMSS and 1,1′-azobiscyclohexane-
carbonitrile was carried out at 103 °C. The GC showed 21 as the
main product together with residual starting bromide. During this
reaction the cis/trans ratio of 13:16 changed from 2.5:1 to 1:5,
i.e. isomerization took place to favour the thermodynamically more
stable trans-isomer. Probably cyclization could not compete with
rapid isomerization and H-atom transfer reactions.
Scheme 3 Preparation and homolytic reactions of 1-bromo-(2-iso-
cyanatocyclopropyl)benzene.
to thermal isomerisation (see below). Alkaline hydrolysis of 11
provided acid 12 which was treated with sodium azide and ethyl
chloroformate under anhydrous conditions to produce trans-iso-
cyanate 13 via a Curtius rearrangement. The easily degraded iso-
cyanate was used without purification for the radical reactions.
Trans-bromo-isocyanate 13 and azobisisobutyronitrile (AIBN)
were photolysed with tributyltin hydride in toluene at ambient
temperature (ca. 40 °C). The sole product obtained under these
conditions was the direct reduction product trans-(2-isocyanato-
cyclopropyl)benzene (21). No other products were observed when
the reaction was carried out at 70 °C or when hexamethylditin was
used in place of tributyltin hydride. A reaction was also carried
out in which the Bu3SnH was added with a syringe pump over
3 hours during continuous photolysis. The main product under
these circumstances was again 21. However, GC-MS analysis of
the product mixture showed a minor amount of a product having
M+ = 159 [C10H9NO]. The fragmentation pattern was in better
agreement with the 4,5-dihydrobenzo[c]azepin-1-one structure
20 for this product rather than 4-methyl-4H-isoquinolin-1-one;
which is the main alternative. Reactions were also carried out with
tris(trimethylsilyl)silane (TTMSS) and 1,1′-azobiscyclohexane-
carbonitrile in benzene at 70 °C and also gave 21 as the only
significant product. A reaction initiated with ethyl piperidine
hypophosphite (EPHP)7 at 70 °C in benzene yielded only a complex
intractable mixture none of whose components could be identified.
The lack of ring closure could result from the trans structure
of compound 13 (and that of its derived trans-aryl radical 6a).
The distance between the radical centre and the isocyanate
moiety will be large and the structure will be quite rigid so that
the stereoelectronically favoured transition state geometry cannot
be attained. However, we expected that trans-13 would readily
interconvert with the cis-isomer 16 via a stabilised diradical
intermediate 14 (Scheme 3). Heating 13 should lead to rupture
of the weak disubstituted cyclopropane bond with production of
diradical 14 which is stabilised by resonance delocalisation of both
unpaired electrons. Analogous cyclopropane isomerisations are
well documented8–10 and can occur rapidly when the intermediates
Scheme 4 Preparation of methyl cis-bromophenylpropenate.
Replacement of the cyclopropane ring with an oxiranyl ring was
also investigated. Methyl 3-(2-bromophenyl)-2-oxirane carboxylate
25 was made in 24% yield by treating alkenyl ester 10 with m-
chloroperbenzoic acid following a literature procedure.13 Because
of the poor yield, the method of Wunsch,14 involving reaction of
2-bromobenzaldehyde with methyl chloroacetate mediated by
NaOMe was carried out and afforded 25 in a much improved yield
of 82% (Scheme 5). Oxirane ester 25 was then converted to the
corresponding trans-isocyanate 26 using the same method as for
the corresponding cyclopropane derivative. Treatment of 26 with
n-Bu3SnH gave a mixture of products. The main component of
which, 2-bromophenyl-3-aminooxirane, was probably formed by
hydrolysis of the isocyanate. TTMSS was also employed, with 1,1′-
azobiscyclohexanecarbonitrile as initiator, in reactions at 83 °C
1
and at 103 °C. Analysis by H NMR and GC-MS showed much
unreacted 26 at both temperatures together with (2-bromophenyl)-
3-aminooxirane (from hydrolysis) and the reduced compound, 2-
isocyanato-3-phenyloxirane. The GC-MS showed the presence of
a minor product having M+ = 161 [C9H7NO2]. Attempts to isolate
and characterise this were unsuccessful but it was probably due to
traces of the cyclised product 5H-6-oxa-8-azabenzocyclohepten-9-
one, 29.
Conclusions
Analysis of the residual 1-bromo-(2-isocyanatocyclopropyl)-
benzene showed that isomerization of 16 to 13 occurred rapidly
at 150 °C (and likely also at lower temperatures) probably by the
2 4 7 2
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 4 7 1 – 2 4 7 5

View Article Online
The filtrate was collected, and the aqueous layer was extracted with
petroleum ether (3 × 30 cm3). The combined organic layers were
dried over magnesium sulfate, filtered, and concentrated. The resi-
due was purified on a silica column (ether/hexane, 1:3) to afford
pure 10 as a clear yellow oil (4.9 g, 81%). H 3.85 (3H, s, Me), 6.43
(1H, d, CHCOOR, J 16.0), 7.25–7.40 (2H, m, ArH), 7.65–7.75 (2H,
m, ArH), 8.83 (1H, d, CHAr, J 16.0); C 51.8, 120.6, 125.2, 127.6,
127.7, 131.1, 133.3, 134.4, 143.1, 166.7.
Methyl trans-2-(2-bromophenyl)cyclopropanecarboxylate 116
A solution of p-tolylsulfonylmethylnitrosamide in ether (12 g in
70 cm3) was added over 30 min to a solution of potassium hydroxide
(3 g) in water (5 cm3), monoethyl ether of diethylene glycol (13 cm3)
and ether (5 cm3) at 75–80 °C.As soon as all the nitrosamide solution
had been added, additional ether (60 cm3) was added at the previous
rate until the distillate was colourless. The resulting ether solution
of diazomethane was continuously distilled into a stirred, cooled
(−10 °C) solution of methyl trans-3-(2-bromophenyl)propenoate
(1.0 g, 5.6 mmol) and palladium acetate (0.006 g, 0.003 mmol) in
dichloromethane/ether (1:2, 75 cm3). The reaction was quenched
by addition of 2 cm3 of acetic acid after 2 h. The resulting mixture
was washed with saturated aqueous sodium hydrogen carbonate
(3 × 5 cm3), dried over magnesium sulfate, filtered, and concen-
trated. The crude product was purified on a silica column eluted with
ether/hexane (1:1) to afford pure 11 (2.0 g, 98%); H 1.35–1.40 (1H,
m, CH), 1.60–1.70 (1H, m, CH), 1.80–1.70 (1H, m, CH), 2.70–2.80
(1H, m, CH), 3.80 (3H, s, Me), 7.00–7.20 (2H, m, ArH), 7.25–7.35
(1H, m, ArH), 7.60–7.65 (1H, m, ArH); C 15.8, 22.9, 27.0, 51.9,
126.2, 127.4, 127.5, 128.2, 132.6, 138.9, 173.7.
Scheme 5 Preparation and reaction of an oxiranyl-isocyanate.
mechanism outlined in Scheme 3. The analogous cis- and trans-
aryl radicals are -radicals with localised radical centres. They
are therefore structurally very similar to their parent bromides and
probably equilibrate via a similar rearrangement. Cyclization of
the trans-radical is structurally forbidden. However, the cis-radical
17 appears capable of ring closure preferably in the 6-endo mode
onto the C-atom of the isocyanate moiety. Although 17, and the
analogous oxiranyl radical 27, were generated under a variety of
experimental conditions, only traces of cyclised products were
detected. Previous work with -isocyanato-C-centered radicals has
shown that they cyclise with rate constants comparable to that of the
archetype hex-5-enyl radical.1,3 However, it appears that for 17 (and
cis-27) the cyclization cannot compete with H-atom transfer from
n-Bu3SnH (or TTMSS) and/or the reverse ring opening reaction.
Cyclizations of other hex-5-enyl type radicals can compete with
H-atom transfer from both metal hydrides under similar experimen-
tal conditions. We conclude, therefore, that the rate of cyclization
of 17 (and cis-27) must be slower than hex-5-enyl cyclization or that
the reverse ring-opening process must be faster than for analogous
radicals. The presence of the cyclopropane ring (or oxirane ring)
and near-linear NCO group in the chain of 17 (and cis-27) affects
their conformational preferences significantly in comparison with
hex-5-enyl. The chains are likely to be stiffer and much less flexible
so that approach from above the -system of the isocyanate group in
either the 5-exo or 6-endo-modes may be more difficult. This may
explain their reluctance to cyclise.
trans-2-(2-Bromophenyl)cyclopropane carboxylic acid 126
Ester 11 (1 g, 5.2 mmol) was dissolved in methanol (10 cm3), and
2 M aqueous sodium hydroxide (5 cm3) was added. The solution
was stirred at room temperature for 2 h. The methanol was evapo-
rated, and the remaining solution was diluted with water (15 cm3),
washed with ether (20 cm3), acidified with 5 M aqueous hydro-
chloric acid, and extracted with ether (3 × 30 cm3). The combined
organic layers were dried over magnesium sulfate, filtered, and
concentrated to afford pure 12 (0.5 g, 53%); H 1.25–1.40 (1H, m,
CH), 1.50–1.65 (1H, m, CH), 1.65–1.80 (1H, m, CH), 2.60–2.80
(1H, m, CH), 6.90–7.10 (2H, m, ArH), 7.15–7.20 (1H, m, ArH),
7.40–7.60 (1H, m, ArH).
Experimental
trans-1-Bromo-(2-isocyanatocyclopropyl)benzene 13
1H, and 13C, NMR spectra were obtained using Bruker AM300,
Bruker Avance 300 and Varian Gemini 2000 spectrometers. All
samples were dissolved in deuterated chloroform, unless other-
wise stated, using Me4Si as an internal standard. Chemical shifts
are given in ppm. IR spectra were obtained using a Perkin-Elmer
1710 Infrared FT spectrometer. Frequencies are given in cm−1.
Mass spectra and GC-MS spectra were obtained using a Finnigan
Incos 50 quadrupole instrument coupled to a Hewlett Packard HP
5890 chromatograph fitted with a 25 m HP 17 capillary column
(50% phenyl methyl silicone). All melting points were determined
in open capillary tubes and are reported uncorrected. TLC was
performed on pre-coated plates of silica gel G-60 F-254 (Merck).
Elemental analyses were recorded with an Agilent 7500 Series
ICP-MS spectrometer that had built in laser ablation capability
or on a Carlo Erba CHNS analyser. Column chromatography was
performed using BDH silica gel (40–63 m) eluting with the given
solvent mixture.
A mixture of trans-3-(2-bromophenyl)cyclopropanecarboxylic
acid 12 (0.5 g, 2.1 mmol), triethylamine (0.3 g, 2.9 mmol), and
ethyl chloroformate (0.3 g, 3.1 mmol) in dry acetone (20 cm3) was
stirred at −10 °C. A solution of sodium azide (0.2 g, 3.6 mmol) in
H2O (10 cm3) was added after 2.5 h. The stirring was discontinued
after an additional 30 min. The resulting suspension was poured into
cold H2O (22 cm3) and was extracted with toluene. The combined
organic layers were dried over magnesium sulfate, filtered and con-
centrated to about 50% of the volume to remove remaining traces
of H2O. The resulting solution was heated (90 °C bath tempera-
ture) until the evolution of nitrogen ceased, and the solution was
concentrated to give the resulting isocyanate as an oil. The NMR
spectra showed that the isocyanate has been obtained as a mixture
of the trans and cis isomers; H 1.20–1.38 (4H, m, cis + trans),
1.40–1.58 (0.5H, m, cis), 1.65–1.80 (0.5H, m, cis), 2.30–2.40 (1H,
m, trans), 2.80–2.90 (2H, m, cis + trans), 6.90–7.10 (4H, m, ArH,
cis + trans), 7.15–7.20 (2H, m, ArH, cis + trans), 7.40–7.60 (2H,
1
m, ArH, cis + trans). H NMR (trans-isomer); H 1.20–1.38 (2H,
Methyl trans-2-(2-bromophenyl)propenoate 106
m), 2.30–2.40 (1H, m), 2.80–2.90 (1H, m), 6.95 (1H, d, ArH, J 8.8),
7.10 (1H, t, ArH, J 8.1), 7.20 (1H, t, ArH, J 8.1), 7.55 (2H, d, ArH,
J 8.8); C (trans-isomer) 17.5, 26.0, 29.4, 126.5, 127.6, 127.9, 128.8,
132.9, 138.4, 179.5; NCO = 2273 cm−1; m/z (%), 238/236 (M+ 2),
208/210 (8), 182/184 (6), 158 (100), 130 (28), 115 (29), 103 (23),
89 (7), 77 (7), 63 (8), 51 (16). (Found: C, 50.35; H, 3.36; N, 5.78.
C10H8BrNO requires C, 50.45; H, 3.39; N 5.88%).
A mixture of 2-bromoiodobenzene (7.0 g, 24.7 mmol), methyl
acrylate (2.8 cm3, 31.9 mmol), palladium acetate (0.1 g, 0.5 mmol),
tetrabutylammonium chloride (6.9 g, 24.7 mmol), and finely
ground potassium carbonate (8.6 g, 62.0 mmol) in DMF (26 cm3)
was stirred at 50 °C for 19 h. Petroleum ether (150 cm3) and brine
(40 cm3) were added and the mixture was filtered under suction.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 4 7 1 – 2 4 7 5
2 4 7 3

View Article Online
Reaction of trans-1-bromo-(2-isocyanatocyclopropyl)benzene
same conditions to afford 23 (0.2 g, 13%); H 3.23 (1H, s, CH), 3.75
(13) with tributyltin hydride
(3H, s, Me), 3.94 (1H, s, CH), 6.05 (1H, d, J 13.2), 6.13 (1H, d,
J 13.2), 7.2–8.0 (4H, m). The trans-isomer was also observed in the
NMR spectrum in a 1.0:2.5 ratio (trans:cis).
Procedure 1. trans-1-Bromo-(2-isocyanatocyclopropyl)benzene
13 (10 mg, 0.04 mmol) and tributyltin hydride (11 mg, 0.04 mmol)
in solution in toluene (5 cm3) were photolysed for 4.5 h in the
presence of 10% ofAIBN. GC-MS: major product; peak 198, trans-
(2-isocyanatocyclopropyl)benzene (21, library fit 95%), m/z (%)
159 (M+ 13), 115 (2), 91 (100), 65 (11), 51 (3). Several additional
minor components were present on the chromatogram.
cis-2-(2-Bromophenyl)cyclopropane carboxylic acid
An ether solution of diazomethane from p-tolylsulfonylmethyl-
nitrosamide (7.2 g) was continuously distilled into a stirred, cooled
(−10 °C) solution of methyl cis-3-(2-bromophenyl)propenoate
(200 mg 0.8 mmol) and palladium acetate (0.089 g) in
dichloromethane/ether (1:2, 75 cm3). The reaction was quenched
by addition of 1 cm3 of acetic acid after 2 h. The resulting mixture
was washed with saturated aqueous sodium hydrogen carbonate
(3 × 5 cm3), dried over magnesium sulfate, filtered, and concen-
trated to afford 0.2 g of crude product. The product was purified on
a silica column eluted with ether/hexane (2:1) to afford the cis- plus
trans-esters as a yellow oil (0.06 g, 30%); H 1.35–1.40 (1H, m,
CH), 1.60–1.70 (1H, m, CH), 1.80–1.70 (1H, m, CH), 2.70–2.80
(1H, m, CH), 3.80 (3H, s, Me), 7.00–7.20 (2H, m, ArH), 7.25–7.35
(1H, m, ArH), 7.60–7.65 (1H, m, ArH).
The methyl 3-(2-bromophenyl)cyclopropenoate isomer mixture
(0.06 g, 0.24 mmol) was dissolved in ethanol (10 cm3), and 2 M
aqueous potassium hydroxide (5 cm3) was added. The solution was
stirred at room temperature for 2 h. The ethanol was evaporated, and
the remaining solution was diluted with water (5 cm3), washed with
ether (20 cm3), acidified with 5 M aqueous hydrochloric acid, and
extracted with ether (3 × 10 cm3). The combined organic layers were
dried over magnesium sulfate, filtered, and concentrated to afford
the cis- and trans-3-(2-bromophenyl)cyclopropane carboxylic acids
(0.057 g, 98%). 1H NMR (acetone) cis + trans: H 1.07–1.12 (2H,
m, CH), 1.40–1.55 (2H, m, CH), 1.75–1.81 (1H, m, CH), 2.03–2.07
(1H, m, CH), 7.09–7.86 (8H, m, ArH).
Procedure 2. Tributyltin hydride (11 mg, 0.04 mmol) was added
slowly into a solution of 13 (10 mg, 0.04 mmol) in toluene (5 cm3)
in the presence of 10% of AIBN. The solution was photolysed for
3 h. GC-MS: peak 215; trans-(2-isocyanatocyclopropyl)benzene
21, peak 238; trans-1-bromo-(2-isocyanatocyclopropyl)benzene.
Several additional minor components were present on the
chromatogram.
Procedure 3. A quartz tube was heated in an oil bath at 150 °C
then 13 (100 mg, 0.42 mmol) was added and, after 5 min, tri-
butyltin hydride (147 mg, 0.50 mmol). The mixture was photolysed
and monitored by GC-MS. After 30 min the product mixture
was subjected to column chromatography (hexane/EtOAc, 1:3).
Fraction 3, trans-(2-isocyanatocyclopropyl)benzene 21, fraction
4 contained trans- + cis-2-phenyl-cyclopropylamine (library fit
96%), m/z (%) M+ 132 (M+ 100), 115 (40), 104 (14), 91 (20),
77 (18), 63 (8), 56 (39), together with residual 13 and 16.
Reactions of trans-1-bromo-(2-isocyanatocyclopropyl)benzene
(13) with hexamethylditin and with EPHP
Trans-isocyanate 13 (10 mg, 0.04 mmol) and hexamethylditin
(12.4 mg, 0.04 mmol) in solution in toluene (5 cm3) were
photolysed for 3 h in presence of 10% of AIBN. GC-MS analysis
showed a complex mixture, but no trace of cyclised products.
Trans-isocyanate 13 (0.50 g, 2 mmol) and EPHP (10 eq.) in
solution in benzene (30 cm3) were heated for 1 h under a nitrogen
atmosphere at 70 °C. AIBN (0.04 eq.) was added in two portions
over 30 min and reflux was continued for 3 days. On cooling, the
reaction was diluted with petroleum ether and washed successively
with sodium hydrogen carbonate, aqueous hydrochloric acid
(2 M), sodium hydrogen carbonate and brine. The organic layer
was then dried over magnesium sulfate, filtered and concentrated
in vacuo. GC-MS showed only a complex mixture of unidentified
components.
cis-1-Bromo-(2-isocyanatocyclopropyl)benzene 16
The mixture of cis- and trans-3-(2-bromophenyl)-cyclopropane
carboxylic acids (0.55 g, 0.24 mmol), triethylamine (0.03 g,
1.4 eq.), and ethyl chloroformate (0.04 g, 1.5 eq.) in dry acetone
(10 cm3) was stirred at −10 °C and a solution of sodium azide
(0.03 g, 1.7 eq.) in H2O (0.1 cm3) was added after 2.5 h. The stirring
was discontinued after an additional 30 min. The resulting suspen-
sion was poured into cold H2O (5 cm3) and was extracted with
toluene. The combined organic layers were dried over magnesium
sulfate, filtered and concentrated to about 50% of the volume to
remove remaining traces of H2O. The resulting solution was heated
(90 °C bath temperature) until the evolution of nitrogen ceased,
and the solution was concentrated to give a mixture of 13 and 16
(ratio 1.0:2.5) (0.055 g, 98%). NCO 2272 cm−1; 1H NMR (benzene)
cis-isomer: H 1.20–1.40 (2H, m, CH), 1.50–1.65 (0.5H, m, CH),
1.65–1.80 (0.5H, m, CH), 2.60–2.80 (1H, m, CH), 6.90–7.10 (2H,
m, ArH), 7.15–7.20 (1H, m, ArH), 7.40–7.60 (1H, m, ArH).
Reaction of trans-1-bromo-(2-isocyanatocyclopropyl)benzene
(13) with TTMSS
trans-Isocyanate 13 (60 mg, 0.25 mmol) and tris(trimethylsilyl)-
silane (75 mg, 0.30 mmol) in solution in C6D6 (1 cm3) were
heated for 2 h at 103 °C in the presence of 10% of 1,1′-azodicyclo-
hexanecarbonitrile (0.006 g). The 1H NMR spectrum, showed
mainly starting material plus several resonances due to minor
unidentified components. GC-MS showed trans-(2-isocyanato-
cyclopropyl)benzene 21 and unreacted 13. Reaction at 83 °C gave
a similar result.
Reaction of cis-1-bromo-(2-isocyanatocyclopropyl)benzene
(16) with TTMSS
The mixture of 13 and 16 (20 mg, 0.08 mmol) and tris(trimethyl-
silyl)silane (25 mg, 0.10 mmol) in solution in C6D6 (1 cm3)
was heated at 103 °C for 2 h in the presence of 10% of 1,1′-
Methyl cis-2-(2-bromophenyl)propenoate 23
1
azodicyclohexanecarbonitrile (0.002 g). Analysis by H NMR and
A
solution of bis(trifluoroethyl)phosphonoacetate11 (1.92 g,
GC-MS showed: peak tR 12.89, trans-(2-isocyanatocyclopropyl)-
benzene (21); peak tR 15.50, cis-16 and, peak tR 15.65, trans-13
(cis:trans ratio: 1:5). Traces of several unidentified components
were also observed.
7 mmol) and 18-crown-6-acetonitrile complex12 (11.62 g, 5 eq.) in
anhydrous THF (15 cm3) was cooled to −78 °C under nitrogen and
treated with KN(TMS)2 (1.34 g, 1 eq., 0.6 M in toluene). 2-Bromo-
benzaldehyde (1.24 g, 7 mmol) was then added and the resulting
mixture was stirred for 30 min at −78 °C. Saturated NH4Cl was
added and the product was extracted into ether (3 × 30 cm3). The
organic phases were dried over magnesium sulfate and evaporated.
Flash chromatography was performed (7:3 EtOAc:petroleum
ether) and the product was eluted at the same time as the aldehyde.
The solution was treated with sodium bisulfite to remove the
bromobenzaldehyde and a second column was then run using the
Methyl 3-(2-bromophenyl)-2-oxiranecarboxylate 2514,15
Method 1. A solution of methyl cis-2-(2-bromophenyl)-
propenoate (2.5 g, 10.4 mmol) in dichloromethane was treated with
m-CPBA (2.15 g, 12.5 mmol) and refluxed for 48 h. The reaction
mixture was cooled with ice, filtered, and the filtrate was extracted
with saturated NaHSO3, aq NaHCO3 and brine. The resulting
2 4 7 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 4 7 1 – 2 4 7 5

View Article Online
organic layers were dried over anhydrous sodium sulfate and
evaporated in vacuo. The product was purified by column chromato-
graphy to afford 25 (24%).
183/185 (60), 153/155 (18), 132 (15), 115 (20), 89 (25), 75 (20),
50 (15).
Reaction of 26 with tributyltin hydride
Method 214. 2-Bromobenzaldehyde (9.25 g, 50 mmol) and
methyl chloroacetate (6.6 cm3, 75 mmol) were added dropwise at
the same time to Na (1.8 g, 78.5 mmol) in dry methanol (75 cm3).
The solution was stirred for 1 h at −10 °C, then for 16 h at rt. HCl
(60 cm3 of 0.3 M) was added and the solution was extracted with di-
chloromethane. The organic phases were then dried over magnesium
sulfate and the solvent evaporated under reduced pressure. The
product was purified by fractional distillation (bp 112–120 °C)
to afford epoxide 25 as a clear oil which crystallised on standing
(10.5 g, 82%); H 3.29 (1H, s, CH), 3.76 (3H, s, CH3), 4.27 (1H,
s, CH), 7.08–7.24 (3H, m, ArH), 7.41–7.43 (1H, m, ArH); m/z (%)
257/249 (M+ 43), 225 (20), 199 (100), 169 (29), 89 (9). (Found: C,
46.26; H, 3.59. C10H9BrO3 requires: C, 46.72; H, 3.53%).
Oxirane 26 (50 mg, 0.21 mmol) and tributyltin hydride (60 mg,
0.21 mmol) in solution in C6D6 (0.5 cm3) were photolysed at 40 °C
for 3 h. The GC-MS showed no trace of the cyclised product.
Only the product of reduction was identified. Several traces of
unidentified components were observed. GC-MS, peak 373,
3-phenyloxiranylamine, m/z (%) 135 (M+ 100), 118 (17), 107 (68),
91 (67), 79 (49), 65 (16), 51 (14).
Reaction of 26 with TTMSS
Oxirane 26 (40 mg, 0.17 mmol) and TTMSS (50 mg, 0.20 mmol)
in solution in C6D6 (1 cm3) were heated for 2 h at 103 °C in the
presence of 1,1′-azodicyclohexanecarbonitrile (0.004 g). The same
procedure was repeated at 83 °C for 7 h. The results were essentially
1
(2-Bromophenyl)-2-oxirane carboxylic acid14
identical. The H NMR spectrum showed mainly the unreacted
starting materials. GC-MS: peak tR 18.4: m/z (%) 161 (M+ 100),
144 (3), 117 (15), 105 (70), 91 (4), 77 (36), 51 (13) [C9H7NO2,
probably 5H-6-oxa-8-azabenzocyclohepten-9-one, 29], together
with residual 26 and several minor unidentified components.
Methyl-3-(2-bromophenyl)-2-oxiranecarboxylate (2.1 g, 8.2 mmol)
was dissolved in ethanol (20 cm3), and 2 M aqueous potassium
hydroxide (15 cm3) was added. The solution was stirred at room
temperature for 2 h. The methanol was evaporated, and the
remaining solution was diluted with water (15 cm3), washed with
ether (20 cm3), acidified with 5 M aqueous hydrochloric acid, and
extracted with ether (3 × 30 cm3). The combined organic layers
were dried over magnesium sulfate, filtered, and concentrated to
afford pure (2-bromophenyl)-2-oxirane carboxylic acid (1.96 g,
98%), mp 108 °C; H 3.42 (1H, s, CH), 4.45 (1H, s, CH), 5.45
(1H, br, OH), 7.21–7.41 (3H, m, ArH), 7.56–7.65 (1H, m, ArH); C
55.9, 58.5, 123.1, 126.7, 128.2, 130.6, 132.9, 134.7, 173.1; m/z (%)
243/245(M+, 76), 227 (31), 199 (100), 185 (140), 168 (28), 119 (7),
57 (20). (Found: C, 44.48; H, 2.68. C9H7BrO3 requires: C, 44.47;
H, 2.90%).
References
1 G. Merenyi, J. Lind and L. Eberson, Acta Chem. Scand., 1998, 52, 62;
R. L. Tlumak, J. C. Day, J. P. Slanga and P. S. Skell, J. Am. Chem. Soc.,
1982, 104, 7257; T. Koenig and A. Wielesek, Tetrahedron Lett., 1975,
2007.
2 P. Kaushal and B. P. Roberts, J. Chem. Soc., Perkin Trans. 2, 1989,
1559.
3 P. L. Minin and J. C. Walton, J. Org. Chem., 2003, 68, 2960.
4 For a review see: J. C. Walton, Houben-Weyl, Carbocyclic Three-
Membered Ring Compounds, vol. E17c, ed. A. de Meijere, Thieme,
Stuttgart, 1997, p. 2438.
5 M. Cook, O. Hares, A. Johns, J. S. Murphy and C. W. Patterson,
J. Chem. Soc., Chem. Commun., 1986, 1419; A. Johns and J. A.
Murphy, Tetrahedron Lett., 1988, 29, 837.
trans-2-(2-Bromophenyl)-3-isocyanatooxirane 26
6 J. Vallgarda, J. Med. Chem., 1996, 39, 1485.
A mixture of (2-bromophenyl)-2-oxirane carboxylic acid (1.0 g,
4.0 mmol), triethylamine (0.5 g, 5.0 mmol), and ethyl chloro-
formate (0.6 g, 5.5 mmol) in dry acetone (30 cm3) was stirred
at −10 °C. A solution of sodium azide (0.45 g, 6.9 mmol) in H2O
(2 cm3) was added after 2.5 h. The stirring was discontinued after
an additional 30 min. The resulting suspension was poured into cold
H2O (12 cm3) and was extracted with toluene. The combined organic
layers were dried over magnesium sulfate, filtered and concentrated
to about 50% of the volume to remove remaining traces of H2O.
The resulting solution was heated (90 °C bath temperature) until the
evolution of nitrogen ceased, and the solution was concentrated to
give 26 as a clear oil (83%); H 3.08 (1H, s, CH), 4.46 (1H, s, CH),
6.60–6.83 (1H, m, ArH), 7.01–7.29 (2H, m, ArH), 7.71–7.77 (1H,
m, ArH); NCO 2260 cm−1; m/z (%) 239/241 (M+ 100), 195/197 (3),
7 J. A. Murphy, Pure Appl. Chem., 2000, 72, 1327; C. G. Martin,
J. A. Murphy and C. R. Smith, Tetrahedron Lett., 2000, 41, 1833;
S. R. Graham, J. A. Murphy and A. R. Kennedy, J. Chem. Soc., Perkin
Trans. 1, 1999, 3071.
8 R. Merenyi, in Substituent Effects in Radical Chemistry, ed. H. G.
Viehe, Z. Janousek and R. Merenyi, D Reidel, Louvain-la-Neuve, 1986,
p. 301.
9 B. S. Rabinovitch, E. W. Schlag and K. B. Wiberg, J. Phys. Chem.,
1958, 28, 504.
10 J. A. Berson and J. M. Balquist, J. Am. Chem. Soc., 1968, 90, 7343.
11 C. W. Still and C. Gennari, Tetrahedron Lett., 1983, 24, 4405.
12 G. W. Gokel and D. J. Cram, J. Org. Chem., 1974, 39, 2445.
13 H. Honig, P. Seufer-Wasserthal and H. Weber, Tetrahedron, 1990, 46,
3841.
14 B. Wunsch, Arch. Pharm., 1990, 323, 493.
15 D. Yang, M.-K. Wong and Y.-C. Yip, J. Org. Chem., 1995, 60, 3887.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 4 7 1 – 2 4 7 5
2 4 7 5