Radical cyclisation onto nitriles

Article abstract of DOI:10.1016/S0040-4039(00)01596-3

Iminyl radicals, generated by 5-exo cyclisation of alkyl, vinyl and aryl C-centred radicals onto nitriles, undergo β-scission (nitrile translocation), reduction or tandem cyclisation onto alkenes depending on the nature of the α-substituent. 5-exo Cyclisations of aryl radicals onto nitriles undergo nitrile translocation when the α-substituent is CN, CO₂R, SO₂Ph or CONMe₂. The rate of translocation is faster than 5- or 6-exo cyclisation onto alkenes or 1,5-hydrogen abstraction of allylic hydrogens. When the α-substituents are alkyl, the intermediate iminyl radicals do not undergo nitrile translocation. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01596-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 8989–8994
Radical cyclisation onto nitriles
W. Russell Bowman,* Colin F. Bridge and Philip Brookes
Department of Chemistry, Loughborough University, Loughborough, Leics LE11 3TU, UK
Received 2 August 2000; revised 5 September 2000; accepted 14 September 2000
Abstract
Iminyl radicals, generated by 5-exo cyclisation of alkyl, vinyl and aryl C-centred radicals onto nitriles,
undergo b-scission (nitrile translocation), reduction or tandem cyclisation onto alkenes depending on the
nature of the a-substituent. 5-exo Cyclisations of aryl radicals onto nitriles undergo nitrile translocation
when the a-substituent is CN, CO2R, SO2Ph or CONMe2. The rate of translocation is faster than 5- or
6-exo cyclisation onto alkenes or 1,5-hydrogen abstraction of allylic hydrogens. When the a-substituents
are alkyl, the intermediate iminyl radicals do not undergo nitrile translocation. © 2000 Elsevier Science
Ltd. All rights reserved.
Keywords: radical cyclisation; nitriles; iminyl radicals; b-scission.
Radical cyclisation onto nitriles remains enigmatic1 with both successes2 and failures3
reported. In our studies towards the synthesis of bicyclic nitrogen heterocycles we sought to use
tandem radical reactions involving cyclisation onto nitriles to yield intermediate iminyl radicals
followed by cyclisation of the iminyl radicals onto suitably placed alkenes (Scheme 1). Although
this type of tandem reaction has not been previously reported, iminyl radicals generated by
cyclisation onto nitriles have been shown to cyclise onto arenes by an oxidative mechanism.4
The cyclisation of iminyl radicals onto alkenes has been studied by Zard5 and reviewed.1,5 The
rate of cyclisation of 2-methyl-6,6-diphenyl-5-hexeniminyl has been measured6 as 2.2×106 s−1 at
25°C. We report our initial results as a guide to further understanding the reactivity of radical
cyclisation onto nitriles.
Scheme 1. Tandem cyclisation of iminyl radicals generated by cyclisation onto nitriles
* Corresponding author. Tel: +44 1509 222 569; fax: +44 1509 223 925; e-mail: w.r.bowman@lboro.ac.uk
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01596-3

8990
The use of radical cyclisation onto nitriles has been hindered by the slow rate and is not
generally synthetically favourable, e.g. the rate of cyclisation of 5-cyanobutyl is 4×104 s−1 at
80°C.7 In order to accelerate the rate of cyclisation of nucleophilic C-centred radicals onto
nitriles we studied the radical cyclisation of precursors with electron withdrawing a-substituents
(Scheme 2). We were mindful of the known radical translocation of nitriles facilitated by the
stability of the nitrile triple bond. Reported examples indicate that these translocations are
favoured by ring strain8–10 and the formation of stabilised radicals.9,11–14
Scheme 2. Nitrile translocation via alkyl radical cyclisation. Bu3SnH (1.3 equiv., syringe pump addition over 3 h),
Z=CN (9 h reflux, AIBN, PhMe); Z=CO2Me (6 h, cyclohexane, AMBN)
The cyclisation was fast as predicted with no traces of 6-exo cyclisation onto the alkene, i.e.
the rate of cyclisation onto the nitrile was faster than the rate of 6-exo cyclisation (4.1×104 s−1
at 80°C).15 Both rates would be enhanced by Thorpe–Ingold effects because of the gem-disubsti-
tution. The electron withdrawing effect of the a-substituent should increase the electrophilicity
of the nitrile and the intermediate iminyl radical and therefore enhance both the rate of
cyclisation onto the nitrile and of the iminyl radical onto the alkene. However, the rate of
b-scission is clearly faster and no traces of tandem products were observed. In general the rate
of cyclisation of iminyl radicals is one order of magnitude less rapid than the equivalent carbon
radical,6 i.e. ca. 2.5×104 s−1 indicating that the rate of nitrile translocation is reasonably fast even
when little ring strain is present. GCMS analysis indicated near quantitative yields of translo-
cated products and lower isolated yields reflect problems of separation from tin residues.
Similar translocation has been reported12 for cyclisation of 4-cyano-4-(ethoxycarbonyl)-1-
’
buten-1-yl radicals [(EtO2C)(CN)CHCH2CHꢀCH ]. Therefore, we also tested whether transloca-
tion was faster than 5-exo cyclisation of the intermediate iminyl radical with precursor 1
(Scheme 3). The vinyl bromide 1 also gave only nitrile translocation with the nitrile 2 as the only
detectable product. After some difficulty of separation, 2 was isolated in 23% yield. Again, the
rate of cyclisation onto the nitrile is faster than 6-exo cyclisation of the vinyl radical onto the
Scheme 3. Nitrile translocation via vinyl radical cyclisation. Bu3SnH (1.3 equiv., syringe pump addition over 3 h),
(6 h reflux, AMBN, cyclohexane)

8991
alkene and the rate of translocation is faster than 5-exo cyclisation of the iminyl radical onto the
alkene.
The planned tandem cyclisation using aryl radicals from 2a and 2b also failed and only
translocation (3a (70%) and 3b (42%), respectively) in high yield (by GCMS) was detected
(Scheme 4). Again, b-scission of the intermediate iminyl radicals was faster than cyclisation onto
the pendant alkenes and no products resulting from 5- or 6-exo cyclisation or from 1,5-hydrogen
abstraction of the allylic hydrogen in 2b were observed. Similarly, only translocation (5, 53%)
was obtained for precursor 4 and the possible 5-exo or 6-endo cyclisation onto the aryl ring was
not observed.
Scheme 4. Nitrile translocation via aryl radical cyclisation. Bu3SnH (1.3 equiv., syringe pump addition over 6 h),
(9 h reflux, AMBN, cyclohexane)
We decided to use this system as a model for studying the effect of the a-substituent on
translocation of the nitrile group and the results are shown in Scheme 5 and Table 1. When the
a-substituent was an electron withdrawing group (CN, CONMe2, CO2Et, SO2Ph) and phenyl,
only translocation to 12 and no cyclised imine 11 was observed. When the a-substituent was an
Scheme 5. Effect of a-substituent on nitrile radical translocation

8992
Table 1
Cyclisation of aryl radicals onto nitriles
Precursor 6
% yield 10
% yield 11
% yield 12
Reaction conditionsa
Z=CN, R=H
0
7
12
5
7
0
0
0
0
0
72
41
Cyclohexane, AMBN,b 9 h, syringe pump
Toluene, AIBN, 9 h, syringe pump
Z=CONMe2, R=H
10Z=CO2Et, R=H
Z=Ph, R=H
Z=SO2Ph, R=H
Z=o-BrBn, R=H
Z=Me, R=H
59
24 (24c)
Toluene, AIBN, 2 hd
0
16
0
0
Toluene, AIBN, 2 hd
15e
Toluene, AIBN, 7 h, syringe pump
Toluene, AIBN, 7 h, syringe pump
Toluene, AIBN, 7 h, syringe pump
0
0
15e
Z=Me, R=Me
57e (4814)e
0
a Bu3SnH (1.3 equiv.), reflux, under nitrogen.
b AMBN (azobismethylisobutyronitrile, or 2-(1-cyano-1-methyl-propylazo)-2-methylbutyronitrile) was used because
AIBN is insoluble in cyclohexane.
c 8,10-Dihydrophenanthrene-1-carbonitrile.
d All reagents added at the beginning of the reaction.
e Isolated as the corresponding ketone after hydrolysis of the imine.
electron donating group (alkyl, Bn), no translocation to 12 and only cyclised imine 11
(hydrolysed, as the respective ketone) was observed. For the latter group, the yields were poor
except when disubstituted. The reactions for 6 (R=H, Z=SO2Ph and Ph) were problematic and
reactions using syringe pump addition of the Bu3SnH were inhibited and gave only unaltered
starting material. We suggest that this inhibition is caused by disproportionation between the
intermediate aryl radical 7 and an intermediate biaryl p-radical, generated by addition of the
initial nucleophilic aryl radical 7 to the electrophilic arene in the a-substituent in a bimolecular
reaction.
We propose that the results are explained by the a-substituents, as shown in Scheme 5. When
the a-substituent is electron withdrawing, the rate of b-scission is faster than trapping of the
weakly nucleophilic iminyl radicals with Bu3SnH. The equilibrium between intermediate iminyl
radicals 8 and ring-opened radicals 9 is also aided by rapid reaction between the strongly
electrophilic ring-opened radical 9 and the nucleophilic Bu3SnH, i.e. kH(open)ꢀkH(iminyl) and
the slow rate of cyclisation between the electrophilic radical and the electrophilic carbon of the
’
nitrile group. Weakening of the Cꢁ(CꢀN ) bond and stabilisation of the ring-opened radicals by
the a electron withdrawing, and a-phenyl, groups also plays an important role. For a electron
donating groups the rate of H-transfer between the nucleophilic Bu3SnH and the weakly
nucleophilic iminyl or ring-opened radicals will be similar and the rate of cyclisation by attack
of the nucleophilic a-alkyl radical on the electrophilic centre of the nitrile will be faster than
’
b-scission. Likewise, the weakening of the Cꢁ(CꢀN ) bond will be less than for electron-with-
drawing groups. The literature reports conflicting results for a-(2-bromophenylamino)nitriles,
which yield selective translocation14a or largely imine formation.14b
The conclusion is that tandem cyclisation will best be facilitated by use of a-alkyl substituents
and initial study has shown this to be the case. The a-propenyl nitrile 13 readily gave tandem
cyclisation to yield 17, apparently by a 5-exo, 6-endo cyclisation, and small amounts (<5%) of
the 5-exo, 5-exo tandem product (Scheme 6). We propose that the iminyl radical 14 undergoes
a 5-exo cyclisation to 15 which undergoes rearrangement via the stable a-aminyl benzylic radical

8993
Scheme 6. Tandem cyclisation via intermediate iminyl radicals
16 rather than 6-endo cyclisation. The tricycle 17 could also be arise from thermodynamic
control of the second cyclisation, i.e. 14 to 15 is reversible. However, reversibility is unlikely
because iminyl radicals (e.g. 14) are not stabilised and thermodynamic control is only observed
for stable radicals.
Our results have shown the further synthetic potential of radical cyclisation onto nitrile
groups. Choice of the a-substituent can facilitate nitrile translocation, cyclisation to imines and
tandem cyclisation of the intermediate iminyl radical.
Acknowledgements
We thank the EPSRC for a Research Associate grant (C.F.B.), Loughborough University for
a Postgraduate Research Studentship (P.B.) and the EPSRC Mass Spectrometry Unit, Swansea
University, Wales for mass spectra.
References
1. Review: Fallis, A. G.; Brinza, I. M. Tetrahedron 1997, 53, 17543–17594.
2. Representative papers: Curran, D. P.; Liu, W. Synlett 1999, 117–119; Yamamoto, Y.; Matsumi, D.; Itoh, K. J.
Chem. Soc., Chem. Commun. 1998, 875–876; Alonso, R. A.; Burgey, C. S.; Rao, B. V.; Vite, G. D.; Vollerthun,
R.; Zottola, M. A.; Fraser-Reid, B. J. Am. Chem. Soc. 1993, 115, 6666–6672 and references cited therein; Snider,
B. B.; Buckman, B. O. J. Org. Chem. 1992, 57, 5322–5426; Clive, D. L.; Beaulieu.; Set, L. J. Org. Chem. 1984,
49, 1314–1316; Corey, E. J.; Pyne, S. G. Tetrahedron Lett. 1983, 24, 2821–2814; Molander, G. A.; Wolfe, C. N.
J. Org. Chem. 1998, 63, 9031–9036.
3. Kilburn, J. Tetrahedron Lett. 1990, 31, 2193–2196; Yeung, B.-W. A.; Contelles, J. L. M.; Fraser-Reid, B. J.
Chem. Soc., Chem. Commun. 1989, 1160–1162; Chenera, B.; Chuang, C.-P.; Hart, D. J.; Hsu, L.-Y. J. Org. Chem.
1985, 50, 5409–5410.
4. Montevecchi, P. C.; Navacchia, M. L.; Spagnolo, P. Tetrahedron 1998, 54, 8207; Curran, D. P.; Liu, H.; Josien,
H.; Ko, S.-B. Tetrahedron 1996, 52, 11385–11404; Curran, D. P.; Liu, H. J. Am. Chem. Soc. 1991, 113,
2127–2132; Camaggi, C. M.; Leardini, R.; Nanni, D.; Zanardi, G. Tetrahedron 1998, 54, 5587–5598; Nanni, D.;
Pareschi, P.; Rizzoli, C.; Sgarabotto, P.; Tundo, A. Tetrahedron 1995, 51, 9045–9062; Leardini, R.; Nanni, D.;
Pareschi, P.; Tundo, A.; Zanardi, G. J. Org. Chem. 1997, 62, 8394–8395 and Tetrahedron Lett. 1998, 39,
2441–2442.
5. Review: Zard, S. Z. Synlett 1996, 1148–1154.
6. Le Radic-Biadatti, M.-L.; Callier-Dublanchet, A.-C.; Horner, J. H.; Quiclet-Sire, B.; Zard, S. Z.; Newcomb, M.
J. Org. Chem. 1997, 62, 559–563.
7. Griller, D.; Schmid, P.; Ingold, K. U. Can. J. Chem. 1979, 57, 831–833.

8994
8. Quiclet-Sire, B.; Callier, A. C.; Zard, S. Z. Tetrahedron Lett. 1994, 35, 6109–6112; Callier-Dublanchet, A.-C.;
Quiclet-Sire, B.; Zard, S. Z. Tetrahedron Lett. 1997, 38, 2463–2466; Boivin, J.; Callier-Dublanchet, A.-C.;
Quiclet-Sire, B.; Scaiano, A.-M.; Zard, S. Z. Tetrahedron 1995, 51, 6517–6528;
9. Curran, D. P.; Seong, C. M. Tetrahedron 1992, 48, 2175–2190.
10. Boivin, J.; Fouquet, E.; Zard, S. Z. Tetrahedron 1994, 50, 1757–1768.
11. Kim, S.; Jon, S. Y. J. Chem. Soc., Chem. Commun. 1998, 815–816.
12. Beckwith, A. L. J.; O’Shea, D. M.; Gerba, S.; Westwood, S. W. J. Chem. Soc., Chem. Commun. 1987, 666–667;
Beckwith, A. L. J.; O’Shea, D. M.; Westwood, S. W. J. Am. Chem. Soc. 1988, 110, 2565–2575.
13. Rychnovsky, S. D.; Swenson, S. S. Tetrahedron 1997, 53, 16489–16502.
14. (a) Cossy, J.; Poitevin, C.; Pardo, D. G.; Peglion, J. L. Synthesis 1995, 1368–1370; (b) Sulsky, R.; Gougoutas, J.
Z.; DiMarco, J.; Biller, S. A. J. Org. Chem. 1999, 64, 5504–5510.
15. Beckwith, A. L. J.; Moad, G. J. Chem. Soc., Chem. Commun. 1974, 472–473.
.