From dioxime oxalates to dihydropyrroles and phenanthridines via iminyl radicals

Article abstract of DOI:10.1039/b808625g

Dioxime oxalates are useful precursors for the clean generation of iminyl radicals by sensitised UV photolysis and can be adapted for serviceable preparations of 3,4-dihydro-2H-pyrroles and phenanthridines. The Royal Society of Chemistry.

Full text of DOI:10.1039/b808625g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
From dioxime oxalates to dihydropyrroles and phenanthridines via iminyl
radicalsw
a
Fernando Portela-Cubillo,a Eoin M. Scanlan,z Jackie S. Scottb and
John C. Walton*a
Received (in Cambridge, UK) 21st May 2008, Accepted 7th July 2008
First published as an Advance Article on the web 1st August 2008
DOI: 10.1039/b808625g
Dioxime oxalates are useful precursors for the clean generation
of iminyl radicals by sensitised UV photolysis and can be
adapted for serviceable preparations of 3,4-dihydro-2H-pyrroles
and phenanthridines.
preparation and photochemical reactions of a range of sym-
metrical and unsymmetrical dioxime oxalates.
Jochims and co-workers made dioxime oxalates by conden-
sation of an oxime with oxalyl chloride to give an oxime oxalyl
chloride which was then reacted with a second mole of the
oxime to afford the dioxime oxalate in good yield.11 We used
this method to prepare both symmetrical and unsymmetrical
dioxime oxalates. Purification was usually impracticable be-
cause most dioxime oxalates hydrolysed and degraded rapidly
on exposure to air and during chromatography (SiO2 or
Al2O3). By using fresh oxalyl chloride, and by careful control
of the reactant quantities, almost pure dioxime oxalates could
be made quantitatively and used immediately without further
purification. In a few cases solid dioxime oxalates resulted
which were purified by low temperature recrystallisation. Most
of the dioxime oxalates were obtained as mixtures of E/Z
stereoisomers about their CQN bonds.12
Photochemical dissociations of 1a and 1b were studied by 9
GHz EPR spectroscopy. Deaerated solutions of each dioxime
oxalate in tert-butylbenzene were photolysed with unfiltered
light from a 500 W Hg lamp directly in the EPR resonant
cavity. Signals were weak unless 4-methoxyacetophenone
(MAP) (1 or 2 equiv.) was included as a photosensitiser. In
that case good spectra of dimethyl- and diphenyliminyl radi-
cals (Me2CQNꢀ and Ph2CQNꢀ), with EPR parameters iden-
tical to those reported in the literature,13 were observed. In
initial experiments with 1a,b, and with some of the dioxime
oxalates reported below, spectra of iminoxyl radicals
(R2CQNOꢀ) accompanied the iminyl radical spectra. How-
ever, iminoxyls were not observed when a completely pure
dioxime oxalate was employed. We believe, therefore, that the
iminoxyl spectra result from H-atom abstraction from traces
of oxime impurities and are not due to UV induced scission of
the O–C bonds in 1.14 The EPR spectra established that iminyl
radicals are cleanly released from dioxime oxalates on photo-
lysis, although the signal intensities suggested this process was
more efficient for dioxime oxalates with phenyl substituents on
their CQN bonds.
The rate constant for iminyl radical (R2CQNꢀ) cyclisation is
only about a factor of 20 less than that of the archetype hex-5-
enyl radical,1 so it is not surprising the reaction is useful for
N-heterocycle preparations.2 An important objective is to find
ways of generating iminyls that are simple, ‘‘clean’’ and
flexible. Zard et al. made significant progress towards this
goal with N-hydroxypyridine-2-thione and ketoxime xanthate
precursors.3 A somewhat specialised thermal route involves
addition of imidoyl radicals onto cyanoalkyl groups.4 A direct
thermal route, with promise of flexibility, employs microwave
irradiation of O-phenyl oxime ethers.5 Oxime esters have been
enlisted for release of iminyl radicals6 and acyl types are good
precursors for photochemical syntheses of phenanthridines
and isoquinolines.7 Anion radicals generated by one-electron
transfer to O-aryl oxime ethers and acyl oximes act as iminyl
radical equivalents in N-heterocycle preparations.8
We already knew that oxime oxalate amides
[R2CQNOC(O)C(O)NR02] were good photochemical precur-
sors for iminyl and carbamoyl radicals.9 It seemed possible,
therefore, that the structurally related dioxime oxalates 1
could function as particularly clean and atom-efficient sources
of iminyl radicals because the only by-product would be CO2:
R2CQNOC(O)C(O)ONQCR2 - 2 R2CQNꢀ + 2CO2
1, a; R = Me, b; R = Ph
The only previous attempt to generate radicals from dioxime
oxalates was described in a couple of brief mentions in papers
from Forrester et al.10 They concluded that both iminyl and
iminoxyl radicals [R2CQNOꢀ] were formed during UV photo-
lyses. In this communication we report our study of the
a University of St. Andrews, School of Chemistry, EaStChem,
St. Andrews, Fife, UK KY16 9ST. E-mail: jcw@st-and.ac.uk;
Fax: +44 (0)1334 463808; Tel: +44 (0)1334 463864
b GlaxoSmithKline, New Frontiers Science Park, Third Avenue,
Harlow, Essex, UK CM19 5AW
w Electronic supplementary information (ESI) available: Preparations
of a model ketone, oxime and dioxime oxalate; also dihydropyrroles
and phenanthridines with NMR spectra of new compounds. EPR
spectra of iminyl and iminoxyl radicals. See DOI: 10.1039/b808625g
z Current address: School of Chemistry, University of Dublin, Trinity
College, Dublin 2.
A set of dioxime oxalates containing pent-4-enyl chains 4a–d
was prepared to assess the usefulness of these precursors in
preparations of dihydropyrroles. Individual dioxime oxalates
were dissolved in toluene along with 2 equivalents of MAP and
the solutions were photolysed for 4 h in quartz tubes with light
from a 400 W medium pressure Hg lamp. After removal of
MAP by chromatography the products were shown to be
5-aryl-3,4-dihydro-2H-pyrroles 7a–d.15 A plausible mechanism
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4189–4191 | 4189

View Article Online
is outlined in Scheme 1. By analogy with mono-oxime esters, the
weak N–O bond of the dioxime oxalate should break first,
yielding an iminyl radical 5 and an acyloxyl radical. Rapid
dissociation of the latter is expected to release a second iminyl
radical along with two CO2 molecules. The iminyl radicals 5
then undergo ring closure in the favoured 5-exo-trig mode to
afford dihydropyrrolomethyl radicals 6 that abstract a hydrogen
atom from the toluene solvent with production of the hetero-
cycles 7. The imines (or their ketone hydrolysis products)
resulting from H-atom abstraction by uncyclised radicals 5,
were at most minor by-products, so the iminyl cyclisations must
be fast in comparison with this process. The yield of 7d was
lower than that of 7a, in agreement with the expected slower ring
closure at the more substituted –C(Me)Qatom. The yield of 7c
was also comparatively low, i.e. no support was forthcoming for
gem-dimethyl enhancement of aryliminyl cyclisation.
Fig. 1 Symmetrical and unsymmetrical dioxime oxalates from ester-
containing precursors. Yields are mol% isolated product.
radical will afford a resonance-stabilised benzyl type radical
for which H-atom abstraction from solvent will be much
slower. Instead, the pyrrolobenzyl radical loses an H-atom to
produce 9. Possibly the mechanism involves electron transfer
from the pyrrolobenzyl radical to MAP with production of the
pyrrolobenzyl cation which then transfers a proton (see below).
Dihydropyrroles of type 12 were of interest because they can
serve as precursors for biologically active pyrrolizines.16 The
symmetrical dioxime oxalate 10 was therefore prepared and
photolysed in toluene. However, no 12 was obtained on
photolysis. Instead, the starting material and its hydrolysis
products were recovered.
An interesting contrast was provided by dioxime oxalate 8
obtained from 5-phenylpent-4-en-1-al. After photolysis under
similar conditions, the product isolated was 2-benzylidene-3,4-
dihydro-2H-pyrrole 9, instead of the expected benzyldihydro-
pyrrole. In this case cyclisation of the intermediate iminyl
Compound 10 contains no aryl ring and, in agreement with
the conclusion mentioned above, iminyl generation was much
less efficient. To overcome this problem unsymmetrical dioxime
oxalates 11a–c, with one half containing an aromatic oxime,
were examined. They were prepared by reacting the oxime oxalyl
chloride from methyl 5-methyl-4-oxooct-7-enoate with the oxi-
mes of benzaldehyde, 2,4-dimethoxybenzaldehyde and benzo-
phenone respectively. Photolyses of these precursors with MAP
Scheme 1 Preparation of dihydropyrroles. Yields are mol% isolated
product except those marked # which were determined by NMR
spectra of the total product mixtures. * Yield over 3 steps from
5-phenylpent-4-en-1-ol.
Scheme 2 Preparation of phenanthridines.
ꢁc
This journal is The Royal Society of Chemistry 2008
4190 | Chem. Commun., 2008, 4189–4191

View Article Online
did indeed afford 12 along with the aryl-imines (plus ketones
from hydrolysis) produced by hydrogen transfer to the aryl-
iminyl radicals. The somewhat greater yields of 12 from 11b and
11c (Fig. 1) suggested there was an advantage in using di-
methoxy-substitution of the aromatic part or in employing
benzophenone. We conclude that dihydropyrroles like 12, with
no aryl substituents, can be made in good yields by employing
unsymmetrical dioxime oxalates but, of course, this reduces the
atom-efficiency of the method.
4 D. Nanni, P. Pareschi, C. Rizzoli, P. Sgarabotto and A. Tundo,
Tetrahedron, 1995, 51, 9045; R. Leardini, H. McNab, M. Minozzi
and D. Nanni, J. Chem. Soc., Perkin Trans. 1, 2001, 1072.
5 (a) F. Portela-Cubillo, J. S. Scott and J. C. Walton, Chem.
Commun., 2007, 4041.
6 A. J. McCarroll and J. C. Walton, J. Chem. Soc., Perkin Trans. 2,
2000, 2399.
7 R. Alonso, P. J. Campos, B. Garcia and M. A. Rodriguez, Org.
Lett., 2006, 8, 3521; R. Alonso, P. J. Campos, M. A. Rodrıguez
and D. Sampedro, J. Org. Chem., 2008, 73, 2234.
8 T. Mikami and K. Narasaka, Chem. Lett., 2000, 338; K. Narasaka,
K. Uchiyama, A. Ono and Y. Hyashi, Bull. Chem. Soc. Jpn., 1998,
71, 2945; K. Narasaka, K. Uchiyama and Y. Hyashi, Tetrahedron,
1999, 55, 8915; K. Narasaka and T. Mikami, C. R. Acad. Sci., Ser.
IIc: Chim., 2001, 477; K. Narasaka, M. Kitamura and M. Yoshida,
Chem. Lett., 2002, 144; M. Kitamura and K. Narasaka, Bull.
Chem. Soc. Jpn., 2008, 81, 539.
9 E. M. Scanlan and J. C. Walton, Helv. Chim. Acta, 2006, 89
2133; E. M. Scanlan and J. C. Walton, Chem. Commun., 2002,
2086.
We also prepared dioxime oxalates 15a–d from 2-formylbi-
phenyl derivatives 13a–d as shown in Scheme 2. After UV
irradiation in acetonitrile, phenanthridine derivatives 18a–d
were obtained. In this case, the iminyl radicals released on
photolysis of 15 preferentially underwent 6-endo cyclisation
onto the phenyl acceptors because this yielded the resonance-
stabilised cyclohexadienyl type radicals 16. The latter are too
thermodynamically stabilised to abstract H-atoms from the
solvent. Instead they lost an H-atom, re-aromatised and
afforded phenanthridines 18 (Scheme 2). The reaction was
tolerant of Me and Ph substituents on the iminyl radical and a
methylenedioxy substituent in the base aryl ring.
10 A. R. Forrester, M. Gill, C. J. Meyer, J. S. Sadd and R. H.
Thomson, J. Chem. Soc., Perkin Trans. 2, 1979, 606; A. R.
Forrester, R. J. Napier and R. H. Thomson, J. Chem. Soc., Perkin
Trans. 1, 1981, 984.
11 J. C. Jochims, S. Hehl and S. Herzberger, Synthesis, 1990, 1128.
12 For our purpose this was not important because each Z/E isomer
gave the same iminyl radical on fission of the N–O bond.
13 A. R. Forrester and F. A. Neugebauer, in Landolt–Bornstein,
Magnetic Properties of Free Radicals, ed. H. Fischer and K.-H.
Hellwege, Springer-Verlag, Berlin, vol. II9c1, 1979, p. 115.
14 Iminoxyl radicals are persistent and have much longer lifetimes
than iminyls so their concentration can build up to levels detectable
by EPR spectroscopy even when only minute traces of R2CQNOH
are present. The observation of iminoxyls by Forrester et al.10 was
probably due to the same cause.
The mechanism of the final oxidation may involve electron
transfer from the cyclohexadienyl radical 16 to MAP yielding
the corresponding delocalised cation, together with MAPꢂ
ꢀ
radical anion. Proton transfer from 16 to MAPꢂ would then
yield the phenanthridine 18 together with MAPHꢀ which
would pick up hydrogen in solution to give 1-(4-methoxyphe-
nyl)ethanol 19. This alcohol was detected by NMR spectro-
scopy and MS in several reactions.
ꢀ
15 Experimental details for the preparation of 5-(2,4-dimethoxy-
phenyl)-2-methyl-3,4-dihydro-2H-pyrrole (7b) are typical of the
methodology. A solution of 1-(2,4-dimethoxyphenyl)pent-4-en-1-
one oxime 2b (500 mg, 2.1 mmol) in dry ether (10 cm3) was added
Dioxime oxalates are easily and efficiently prepared from a
wide variety of oximes and can be used immediately without
purification for UV generation of iminyl radicals. The process
works best with precursors having aryl substituents attached
to their CQN bonds. The advantage over other precursors is
that the symmetrical variety cleanly release just one type of
iminyl radical. The method is useful for spectroscopic work
and can also be adapted for serviceable preparations of
3,4-dihydro-2H-pyrroles and phenanthridines.
dropwise to
a stirred solution of oxalyl chloride (133 mg,
1.05 mmol) in ether (10 cm3) at ꢂ40 1C. The mixture was allowed
to reach rt and then stirred at rt for 3 h. Evaporation of solvent
yielded the dioxime oxalate as a red oil (96%). A solution of the
dioxime oxalate (400 mg, 0.57 mmol) and 4-methoxyacetophenone
(171 mg, 1.14 mmol) in toluene (25 cm3) in a quartz tube was
photolysed for 4 h at rt by light from a 400 W medium pressure UV
lamp. After this time the toluene was evaporated to dryness to give
a yellow oil. The oil was purified by column chromatography (10%
EtOAc–hexane) giving 7b as a red oil (67%); 1H NMR (400 MHz,
CDCl3) dH 1.28 (3H, d, J = 6.7 Hz, CH3), 1.44 (1H, m, CH2), 2.12
(1H, m, CH2), 2.88 (1H, m, CH2), 3.04 (1H, m, CH2), 3.77 (3H, s,
CH3), 3.78 (3H, s, CH3), 4.12 (1H, m, CH), 6.43 (2H, m , CH), 7.68
(1H, d, J = 8.5 Hz, CH); 13C NMR dC 21.9 (CH3), 30.1, 38.1
(CH2), 55.5 (CH3 ꢃ 2), 66.4 (CH), 98.6, 105.1 (CH), 116.6 (C),
131.7 (CH), 159.0, 165.8, 172.3 (C); IR 3018, 2964,1609 cmꢂ1
HRMS (CI+) calcd for C13H18NO2 (MH+); 220.1338. Found:
220.1337.
We thank GSK and EaStChem for financial support.
Notes and references
1 M.-H. Le Tadic-Biadatti, A.-C. Callier-Dublanchlet, J. H. Horner,
B. Quiclet-Sire, S. Z. Zard and M. Newcomb, J. Org. Chem., 1997,
62, 559.
2 A. G. Fallis and I. M. Brinza, Tetrahedron, 1997, 53, 17543.
3 J. Boivin, E. Fouquet and S. Z. Zard, Tetrahedron Lett., 1991, 32,
4299; J. Boivin, A.-M. Schiano and S. Z. Zard, Tetrahedron Lett.,
1992, 33, 7849.
16 See for example: M. Vargas-Sanchez, F. Couty, G. Evano, D. Prim
and J. Marrot, Org. Lett., 2005, 7, 5861.
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4189–4191 | 4191