Aryl pyrrolidinones via radical 1,4-aryl migration and 5-endo-trig cyclisation of N-(2-bromoallyl)arylcarboxamides

Article abstract of DOI:10.1016/j.tetlet.2006.09.040

Radical reaction of a series of N-(2-bromoallyl)arylcarboxamides led to the production of 4-arylpyrrolidin-2-ones and directly reduced materials in comparable yields. A cascade process, involving sequential 5-exo-trig spirocyclisation, β-scission, and 5-endo-trig cyclisation of the resulting acyl radical, is proposed to explain the pyrrolidinone products.

Full text of DOI:10.1016/j.tetlet.2006.09.040

Tetrahedron Letters 47 (2006) 8423–8425
Aryl pyrrolidinones via radical 1,4-aryl migration and 5-endo-trig
cyclisation of N-(2-bromoallyl)arylcarboxamides
Matthew J. Palframan, Kirill Tchabanenko* and Jeremy Robertson
Department of Chemistry, University of Oxford, Chemistry Research Laboratory, Mansfield Road, Oxford, OX1 3TA, UK
Received 26 July 2006; revised 31 August 2006; accepted 7 September 2006
Available online 5 October 2006
Abstract—Radical reaction of a series of N-(2-bromoallyl)arylcarboxamides led to the production of 4-arylpyrrolidin-2-ones and
directly reduced materials in comparable yields. A cascade process, involving sequential 5-exo-trig spirocyclisation, b-scission,
and 5-endo-trig cyclisation of the resulting acyl radical, is proposed to explain the pyrrolidinone products.
ꢀ 2006 Elsevier Ltd. All rights reserved.
ipso-Type radical cyclisations onto aromatic rings can
result in the formation of spirocycles.1 For example,
since our initial communication on dearomatising radi-
cal spirocyclisations onto benzofuran and indole nuclei,2
radical cyclisations of benzamides, employing samarium
iodide3 and aerobic conditions,4 and spirocyclisation
onto furan derivatives5 have been reported. In this letter,
we present a reaction of vinyl radicals tethered to aro-
matic rings via an amide that helps to define the scope
and limitations of the use of such systems in synthesis.
O
O
Bn
iv
i-iii
Ar
N
Ar
OH
Br
Ar = 2-naphthyl
O
1
O
Me
Ar
N
+
N Bn
Ph
Ar
Naphthalene-2-carboxamide derivative 1 was prepared
according to our previously established procedure,2
although with a disappointingly poor yield for the final
alkylation step. When this compound was submitted to
standard radical cyclisation conditions,6 none of the
expected spirocycle was isolated; instead, the reaction
yielded a substituted pyrrolidinone 2 and a product of
phenyl transfer 3 (Scheme 1).
2, 10%
3, 28%
Scheme 1. Reagents and conditions: (i) (COCl)2, DMF, CH2Cl2; (ii)
BnNH2, NEt3, CH2Cl2, 92% for two steps; (iii) NaH, 2,3-dibromo-
propene, DMF, 10%; (iv) Bu3SnH, AIBN, benzene, reflux.
with the production of amidoyl radical 48 which can cyc-
lise in an overall 5-endo-trig sense9 to give pyrrolidinone
derivative 2. An alternative 4-exoC@C/3-exoC@O/b-scis-
sion pathway from 4 ! 2 is discounted on the basis
of a lack of precedent for radical ring expansion of
lactams10 and because none of the b-lactam 5, a likely
product of radical 6, was observed.
Our mechanistic explanation for the formation of prod-
ucts 2 and 3 is presented in Scheme 2. The initially
formed vinylic radical can undergo 5-exo ipso-type cycli-
sation either onto the naphthalene nucleus or onto the
phenyl ring of the N-benzyl group. The latter-formed
spirocycle can then ring-open with rearomatisation7 to
generate a nitrogen-stabilised methylene radical, and
then amide 3 following hydrogen atom abstraction. b-
Scission of the spirocyclic radical arising from addition
to the naphthalene ring would result in rearomatisation
In order to investigate this process further we decided to
synthesise a selection of aromatic amides, this time hav-
ing a methyl group in place of benzyl, thus removing the
possibility of competing phenyl transfer. Also, the strat-
egy for amide formation was modified so as to avoid the
low-yielding secondary amide alkylation with 2,3-di-
bromopropene. Thus, the use of 2-bromoallyl amine,
*
Corresponding author. Tel.: +44 0 1865 275690; e-mail: kirill.
tchabanenko@chem.ox.ac.uk
0040-4039/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.09.040

8424
M. J. Palframan et al. / Tetrahedron Letters 47 (2006) 8423–8425
O
O
O
5-exoPh
CH2
Bu3SnH
Bu3SnH
Bu3Sn
Ar
N
Ph
Ar
N
Ar
N
3
2
1
Ph
5-exoAr
Bn
N
O
O
O
5-endo
N Bn
N
Ar
Bn
Ar
4
4-exo
O
O
3-exoAr
;
O
3-exoC=O
H2C
Ar
N Bn
N Bn
N Bn
β-scission;
Bu3SnH
Ar
Ar
5
6
Scheme 2. (Ar = 2-naphthyl).
prepared following Bottini’s procedure,11 in the
amide-forming step was followed by N-methylation
(!7) with good overall yields. The yields for these two
steps and the results of the radical reactions are summa-
rised in Table 1.
ucts 9a–h. Attempts to optimise the reaction conditions
(e.g., by varying the rate of Bu3SnH addition or reaction
temperature12) did not improve the outcome signifi-
cantly and the ratio of rearranged (8) to reduced (9)
products was generally approximately 1:1. As an expla-
nation for this insensitivity to the reaction conditions,
we envisaged possible formation of the reduced products
by 1,4-hydrogen abstraction13 from the amide methyl
followed by reduction of the N-methylene radical. How-
ever, in the radical reaction of amide 7a with Bu3SnD,
the 2H NMR spectrum of the reduced product indicated
deuterium incorporation only at the olefinic methine
which effectively eliminated this explanation. On this
basis, we conclude that the product profile is dictated
largely by the relative population of s-cis and s-trans
amide rotameric radicals (Fig. 1) that interconvert
slowly relative to their lifetimes under the reaction
conditions.14
Formation of the 4-substituted pyrrolidinones 8a–h was
observed with examples of ortho-, meta- and para-substi-
tuted benzamides; in all cases, these lactams were
formed in competition with the directly reduced prod-
Table 1. Isolated yields for the synthesis and radical reactions of aryl
carboxamides 7a–h
O
Me
N
Ar
O
8a–h
The observation that the highest yield of the rearrange-
ment product 8 was obtained in the reaction of 4-(tri-
fluoromethyl)benzamide derivative 7d may indicate a
favourable polarity effect on the rate of the first 5-exo
cyclisation, however, no clear trend arises from the reac-
tions of analogous substrates 7a–c to support this.
Finally, it is noteworthy that the radical reaction of
2-pyrrolecarboxamide 7h produced the pyrrolidinone
8h in spite of an expectation that a spirocyclic product
Me
O
i
ii
Ar
N
Br
O
Ar
Cl
Me
Ar
N
7a–h
9a–h
Reagents and conditions: (i) 2-bromoallylamine, aq. NaOH.
CH2Cl2; then NaH, MeI, DMF; (ii) Bu3SnH, AIBN,
benzene, reflux.
Entry Ar
7 (%) 8 (%) 9 (%)
O
a
b
c
d
e
f
p-Anisyl
p-Tolyl
Phenyl
p-(Trifluoromethyl)phenyl 69
o-Tolyl
m-Tolyl
71
77
63
37
35
30
46
28
36
38
33
41
38
35
O
Me
slow
Ar
N
Ar
N
Me
27
66
64
52
69
Not isolated
s-cis
(→ 9 only)
s-trans
(→ 8 + 9)
34
35
30
g
h
m-Anisyl
N-Methyl-2-pyrrolyl
Figure 1.

M. J. Palframan et al. / Tetrahedron Letters 47 (2006) 8423–8425
8425
chromatography to yield the pyrrolidinone 8a as a pale
yellow oil (63 mg, 37%); Rf 0.07 (petrol/ethyl acetate, 2:1);
m
would result based on the similar cases described by
Jones et al.15
max(thin film)/cmꢀ1 3448w, 2927s, 2360w, 1689s, 1515s,
1463s, 1400s; dH (400 MHz, CDCl3) 2.49 (1H, dd, J 17.0,
8.0 Hz, H3), 2.78 (1H, dd, J 17.0, 8.0 Hz, H3), 2.89 (3H, s,
NCH3), 3.35 (1H, dd, J 9.5, 7.0 Hz, H5), 3.48–3.57 (1H, m,
H4), 3.71 (1H, dd, J 9.5, 8.0 Hz, H5), 3.79 (3H, s, OCH3),
6.87 (2H, d, J 8.5 Hz, ArH), 7.14 (2H, d, J 8.5 Hz, ArH);
dC (100 MHz, CDCl3) 29.5 (NCH3), 36.5 (C4), 38.9 (C3),
55.3 (OCH3), 56.9 (C5), 114.2 (2 · ArCH), 127.7
(2 · ArCH), 134.5 (ArC), 158.5 (ArC–O), 173.9 (C@O);
m/z (CI+) 206 (MH+, 100%), 134 (45); HRMS found:
206.1188, C12H16NO2 (MH+) requires 206.1181.
7. Review: Studer, A.; Bossart, M. Tetrahedron 2001, 57,
9649–9667.
In summary, we have demonstrated the operation of
sequential radical 1,4-aryl migration and formal 5-
endo-trig cyclisation of 1-(arylcarboxamido)prop-2-
en-2-yl radicals in which the amide nitrogen bears an
alkyl group. In a future report, we will describe our
study of the synthesis and radical reactions of substrates
designed to address the limitations caused by rotameric
preferences about the amide group.
Acknowledgements
8. For a related example of b-scission of a spirocyclic lactam
to generate an amidoyl radical, see: Wang, S.-F.; Chuang,
C.-P.; Lee, J.-H.; Liu, S.-T. Tetrahedron 1999, 55, 2273–
2288.
We are grateful to the Chemistry Research Laboratory
NMR and MS staff for their assistance in structure
determination.
9. Review: (a) Ishibashi, H.; Sato, T.; Ikeda, M. Synthesis
2002, 695–713; see also the discussion in (b) Chatgilialo-
glu, C.; Ferreri, C.; Guerra, M.; Timokhin, V.; Froudakis,
G.; Gimisis, T. J. Am. Chem. Soc. 2002, 124, 10765–10772.
10. For pertinent studies see: (a) Bowman, W. R.; Westlake,
P. J. Tetrahedron 1992, 48, 4027–4038; (b) McNab, H. J.
Chem. Soc., Chem. Commun. 1990, 543–545; see also: (c)
Dowd, P.; Zhang, W. Chem. Rev. 1993, 93, 2091–2115.
11. Bottini, A. T.; Dev, V. J. Org. Chem. 1962, 27, 968–973.
12. Carrying out the reactions in refluxing benzene (80 ꢁC) or
chlorobenzene (136 ꢁC) did not change the ratios of
rearranged/reduced starting material.
References and notes
1. Early examples: (a) Hey, D. H.; Rees, C. W.; Todd, A. R.
J. Chem. Soc. (C) 1967, 1518–1525; (b) Hey, D. H.; Jones,
G. H.; Perkins, M. J. J. Chem. Soc. (C) 1971, 116–122.
2. Kyei, A. S.; Tchabanenko, K.; Baldwin, J. E.; Adlington,
R. M. Tetrahedron Lett. 2004, 45, 8931–8934.
3. Ohno, H.; Iwasaki, H.; Eguchi, T.; Tanaka, T. Chem.
Commun. 2004, 2228–2229.
13. 1,4-Hydrogen atom abstraction in both photochemical
and thermal radical processes is most often reported as an
unwanted side-reaction; an exception is Gulea, M.; Lopez-
Romero, J. M.; Fensterbank, L.; Malacria, M. Org. Lett.
2000, 2, 2591–2594.
4. De Turiso, F. G.-L.; Curran, D. P. Org. Lett. 2005, 7, 151–
154.
5. Guindeuil, S.; Zard, S. Z. Chem. Commun. 2006, 665–667.
6. General conditions for radical cyclisation exemplified by
the preparation of 4-(4-methoxyphenyl)-1-methylpyrrolid-
inone (8a). To a stirred, degassed solution of amide 7a
(210 mg, 0.73 mmol) and AIBN (12 mg, 0.07 mmol) in
benzene (40 ml) at reflux was added a solution of Bu3SnH
(0.27 ml, 0.88 mmol) and AIBN (32 mg, 0.20 mmol) in
benzene (10 ml) over 4 h and the resulting mixture was
heated at reflux for a further 2 h and then concentrated in
vacuo. The crude product was eluted with dichlorometh-
ane through a stationary phase of KF (10%) impregnated
silica (Ref. 16) to remove tin residues followed by flash
14. Cohen, T.; McMullen, C. H.; Smith, K. J. Am. Chem. Soc.
1968, 90, 6866–6867.
15. (a) Jones, K.; Ho, T. C. T.; Wilkinson, J. Tetrahedron Lett.
1995, 36, 6743–6744; (b) Ho, T. C. T.; Jones, K.
Tetrahedron 1997, 53, 8287–8294; (c) Escolano, C.; Jones,
K. Tetrahedron Lett. 2000, 41, 8951–8955; (d) Escolano,
C.; Jones, K. Tetrahedron 2002, 58, 1453–1464.
16. Harrowven, D. C.; Guy, I. L. Chem. Commun. 2004, 1968–
1969.