Radical synthesis of guanidines from N-Acyl cyanamides

Article abstract of DOI:10.1002/anie.200907237

Chemical Equation Presented Center stage: Additions of nitrogen-centered radicals to cyanamide compounds provided the first radical synthesis of aromatic polycyclic guanidine derivatives (see scheme). Modular assembly of the substrates allows for a rapid increase of the molecular complexity of scaffolds, which have potential applications for medicinal chemistry.

Full text of DOI:10.1002/anie.200907237

Communications
DOI: 10.1002/anie.200907237
Guanidines
Radical Synthesis of Guanidines from N-Acyl Cyanamides**
Marie-Hꢀlꢁne Larraufie, Cyril Ollivier, Louis Fensterbank,* Max Malacria,* and
Emmanuel Lacꢂte*
Guanidines—especially those embedded into polyclic frame-
works—are an important structural unit of valuable synthetic
intermediates and/or natural products.[1] Therefore guani-
dines are appealing targets for total synthesis,[2] bio-inspired
molecular recognition,[3] organocatalysis,[4] and coordination
chemistry.[5] The development of innovative, efficient, and
flexible methods to access these compounds thus remains an
important goal.[6]
Radical cascade cyclization reactions have become an
important tool used to construct polycyclic structures, in
particular nitrogen-containing heterocycles.[7] Our research
group has introduced N-acyl cyanamides as novel radical
partners for the preparation of quinazolinone systems such as
luotonin A, through a radical domino sequence.[8]
Our approach to luotonin A included a retrosynthetic
disconnection featuring the cyclization of a 2-quinolyl radical
to an acylcyanamide A (Scheme 1). We reasoned that switch-
ing the initial carbon radical to a nitrogen-centered one (as in
B) would provide an entry to cyclic guanidines after aromatic
substitution of iminyl radical C, via tricyclic radical D
(path a). To the best of our knowledge, a radical synthesis
of guanidines is unprecedented in the literature. Iminyl
Scheme 1. Access to luotonin A and proposed route to guanidine
derivatives.
radical C could also lead to a competing and unproductive
b-elimination of an amidyl radical and deliver E, where the
cyano group of the starting cyanamide has translocated to
form a nitrogen-centered radical (path b). Nonetheless, our
previous results with carbon-centered radicals made us
confident that this would, at worst, be a minor path.
selected to validate our approach and was assembled in a very
modular fashion from the corresponding amine, cyanogen
bromide, and benzoyl chloride.
We initially used the reaction conditions developed in our
previous work; Bu3SnH (2 equiv) and AIBN (1.5 equiv) were
slowly added (0.2 molhÀ1) to 1a in benzene at reflux.[8]
Gratifyingly, the desired tricyclic guanidine 2a was isolated,
but in a modest 41% yield (Table 1, entry 1).
Spagnolo and co-workers have shown that the stannyla-
minyl radicals obtained from reactions of tin radicals with
alkyl azides add efficiently to the electrophilic cyano group.[9]
We thus decided to follow the same strategy, even though the
reactivity of cyanamides may differ from that of nitriles
because of the added nitrogen substituent. Substrate 1a was
Replacement of benzene by toluene or tBuOH reduced
the yields (Table 1, entries 2 and 3). Slow addition of Bu3SnH
from a syringe pump was required and resulted in the yields
increasing from 20% (addition of Bu3SnH in one batch;
Table 1, entry 4), to 41% (0.2 mmolhÀ1; Table 1, entry 1), and
then to 76% (0.06 mmolhÀ1; Table 1, entry 5). Lowering the
amount of tin was not helpful (43% yield with 1.2 equiv of
Bu3SnH; Table 1, entry 6). Switching to [(CH3)3Si]3SiH or
running the reaction at room temperature with initiation by
light led to a near complete shutdown of the reaction (Table 1,
entries 7 and 8). Therefore, the best yield was obtained by
slowly adding Bu3SnH (0.06 molhÀ1, 2 equiv) and AIBN
(1.5 equiv) to a solution of w-azido N-acyl cyanamide in
benzene at reflux (Table 1, entry 5).
[*] M.-H. Larraufie, Dr. C. Ollivier, Prof. L. Fensterbank,
Prof. M. Malacria, Dr. E. Lacꢀte
UPMC Univ Paris 06, Institut Parisien de Chimie Molꢁculaire (UMR
CNRS 7201)
4 place Jussieu, C. 229, 75005 Paris (France)
Fax: (+33)1-4427-7360
E-mail: louis.fensterbank@upmc.fr
max.malacria@upmc.fr
emmanuel.lacote@upmc.fr
[**] We thank the UPMC, the CNRS, the IUF (M.M., L.F.), and the ANR
(grant no. BLAN0309, Radicaux Verts) for financial support. M.-H.L.
has been awarded a graduate fellowship by the Rꢁgion Ile-de-France,
which is gratefully acknowledged. Technical assistance was gen-
erously offered through FR 2769.
With these optimized reaction conditions in hand, we next
examined the scope for the radical synthesis of guanidine
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200907237.
2178
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 2178 –2181

Angewandte
Chemie
Table 1: Optimization of the guanidine radical synthesis.
Table 2: Scope of the guanidine synthesis.
Entry Hydride (equiv)
vadd [mmolhÀ1
]
Solvent, T[a]
Yield [%][b]
Entry
1
Substrate
Guanidine
Yield [%][a]
76
1
2
3
4
5
6
7
8
Bu3SnH (2)
Bu3SnH (2)
Bu3SnH (2)
Bu3SnH (2)
Bu3SnH (2)
Bu3SnH (1.2)
0.2
0.2
0.2
one batch
0.06
0.06
benzene, D
toluene, D
tBuOH, D
benzene, D
benzene, D
benzene, D
benzene, D
41
30
31
20
76
1b
1c
2b
2c
43
<10
[(CH3)3Si]3SiH (2) 0.06
Bu3SnH (2) 0.06
2
3
4
75
77
59
benzene, RT[c] <10
[a] Reaction heated to reflux unless otherwise noted. [a] Yield of isolated
product. [b] Reaction was initiated by irradiation. AIBN=2,2’-azobisiso-
butyronitrile.
1d
1e
2d
2e
(Table 2). The substitution on the azide side of the molecule
was examined first. The 5,6,6-tricyclic guanidine derivatives
were obtained in good yields (Table 2, entries 1–3). The yields
decreased gradually when the size of ring A was increased
from five- to six- and to seven-membered rings (76% for 2a,
Table 1, entry 5; 59% for 2e, and 24% for 2 f, Table 2,
entries 4, and 5, respectively). In this latter case, as the
uncyclized amino cyanamide was observed in the crude
reaction mixture. Therefore access to 7,6,6-tricyclic guani-
dines may be affected by a competitive 1,5-hydrogen transfer.
The introduction of substituents in the para position of the
aromatic group did not affect the radical cascade. Substrates
1g (electron-withdrawing group) and 1h (electron-donating
group) behaved similarly, and delivered the desired adducts in
82% yield (Table 2, entries 6 and 7). Naphthyl and pyridyl
derivatives 1i and 1j delivered 2i and 2j, respectively, in good
yields (Table 2, entries 8 and 9). On the other hand, a
thiophene substituent proved more fragile, and 2k was
isolated in the somewhat diminished yield of 49% (Table 2,
entry 10). Finally, cyanamide 1l led to desmethyl adduct 2l via
a final radical aromatization through extrusion of a methyl
radical.[8b]
5
1 f
1g
1h
1i
2 f
2g
2h
2i
24
82
82
81
71
49
69
6
7
8
9
1j
2j
10
11
1k
1l
2k
2l
In all the cases where the final rearomatization proceeded
À
by hydrogen abstraction, the process is formally a radical C
H activation.[10] The reaction is thus complementary to metal-
À
mediated approaches, in particular the rhodium-catalyzed C
[a] Yield of isolated product.
H insertion of guanidines developed by Du Bois and co-
workers.[2c,e] Also, we never observed any sign of the cyano-
translocated products.
Replacement of the final aryl acceptor with an olefin
delivered new insights into the cascade process (Scheme 2).
Azide 3a, which features an a-methyl cinnamyl substituent,
provided two bicyclic guanidines; one with a 5,5 framework
(4a, 26%), and another with a 5,6 framework (5a, 57%).[11]
These compounds arose from a final 5-exo-trig cyclization
(4a), or a 6-endo-trig cyclization and subsequent diastereo-
selective hydrogen transfer (5a). The single diastereomer of
guanidine 5a was determined to be cis by examination of its
coupling constants (see the Supporting Information for
details). Introduction of an additional substituent on the
b carbon atom (as in 3b) blocked the 6-endo-trig process—no
diastereoselectivity was observed in that case.
In conclusion, bi- and tricyclic guanidines were efficiently
prepared for the first time via a radical domino process
involving cyanamides as relay partners. The adducts obtained
are a new class of molecules, whose architecture should prove
attractive to the chemical community at large.
Angew. Chem. Int. Ed. 2010, 49, 2178 –2181
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2179

Communications
5727; Angew. Chem. Int. Ed. 2008, 47, 5641 – 5645; for reviews,
see: d) T. Ishikawa, T. Kumamoto, Synthesis 2006, 737 – 752;
e) D. Leow, C.-H. Tan, Chem. Asian J. 2009, 4, 488 – 507.
[5] M. P. Coles, Dalton Trans. 2006, 985 – 1001.
[6] a) R. G. S. Berlinck, M. H. Kossuga, A. M. Nascimento in
Science of Synthesis, Vol. 18 (Ed.: J. G. Knight), Thieme,
Stuttgart, 2005, pp. 1077 – 1116; in particular, hydroamination
of cyanamides to generate guanidine-based heterocycles by
using an ionic strategy parallel to our own has been reported:
b) R. L. Giles, J. D. Sullivan, A. M. Steiner, R. E. Looper,
Angew. Chem. 2009, 121, 3162 – 3166; Angew. Chem. Int. Ed.
2009, 48, 3116 – 3120.
[7] a) D. J. Hart in Radicals in Organic Synthesis, Vol. 2 (Eds. P.
Renaud, M. P. Sibi), Wiley-VCH, Weinheim, 2001, pp. 279 – 302;
b) M. Albert, L. Fensterbank, E. Lacꢁte, M. Malacria, Top. Curr.
Chem. 2006, 264, 1 – 62; c) W. R. Bowman, J. M. D. Storey,
Chem. Soc. Rev. 2007, 36, 1803 – 1822; for recent examples in
alkaloid synthesis, see: d) P. Schꢂr, P. Renaud, Org. Lett. 2006, 8,
1569 – 1571; e) A. Biechy, S. Hachisu, B. Quiclet-Sire, L. Ricard,
S. Z. Zard, Angew. Chem. 2008, 120, 1458 – 1460; Angew. Chem.
Int. Ed. 2008, 47, 1436 – 1438; f) T. Taniguchi, H. Ishibashi, Org.
Lett. 2008, 10, 4129 – 4131.
Scheme 2. Extension to N-enoyl cyanamides ds=diastereoselectivity.
[8] a) A. Servais, M. Azzouz, D. Lopes, C. Courillon, M. Malacria,
Angew. Chem. 2007, 119, 582 – 585; Angew. Chem. Int. Ed. 2007,
46, 576 – 579; b) A. Beaume, C. Courillon, E. Derat, M.
Malacria, Chem. Eur. J. 2008, 14, 1238 – 1252; for references of
domino reactions involving cyclization onto nitriles and subse-
quent cyclization onto arenes, see: c) D. P. Curran, H. Liu, H.
Josien, S.-B. Ko, Tetrahedron 1996, 52, 11385 – 11404; d) R.
Leardini, D. Nanni, P. Pareschi, A. Tundo, G. Zanardi, J. Org.
Chem. 1997, 62, 8394 – 8399; e) D. Nanni, G. Calestani, R.
Leardini, G. Zanardi, Eur. J. Org. Chem. 2000, 707 – 711; f) W. R.
Bowman, C. F. Bridge, P. Brookes, M. O. Cloonan, D. C. Leach,
J. Chem. Soc. Perkin Trans. 1 2002, 58 – 68; g) A. Gansꢂuer, F.
Piestert, I. Huth, T. Lauterbach, Synthesis 2008, 3509 – 3515;
radical additions onto nitriles have been pioneered by Clive, see:
h) D. L. J. Clive, P. L. Beaulieu, L. Set, J. Org. Chem. 1984, 49,
1313 – 1316; nitriles can also translocate, through b elimination
of the intermediate iminyl radical (not observed in our case, see:
i) R. W. Freerksen, W. E. Pabst, M. L. Raggio, S. A. Sherman,
R. R. Worble, D. S. Watt, J. Am. Chem. Soc. 1977, 99, 1536 –
1542; j) A. L. J. Beckwith, D. M. OꢃShea, S. W. Westwood, J. Am.
Chem. Soc. 1988, 110, 2565 – 2575; k) D. P. Curran, C. M. Seong,
Tetrahedron 1992, 48, 2175 – 2190; l) J. Cossy, C. Poitevin, D.
Gomez Pardo, J.-L. Peglion, Synthesis 1995, 1368 – 1370; m) S. D.
Rychnovsky, S. S. Swenson, Tetrahedron 1997, 53, 16489 – 16502;
n) S. Kim, S. Y. Jon, Chem. Commun. 1998, 815 – 816; o) R.
Sulsky, J. Z. Gougoutas, J. DiMarco, S. A. Biller, J. Org. Chem.
1999, 64, 5504 – 5510; p) W. R. Bowman, C. F. Bridge, P. Brookes,
Tetrahedron Lett. 2000, 41, 8989 – 8994; q) D. Crich, A. A.
Bowers, Org. Lett. 2006, 8, 4327 – 4330; r) D. Crich, A. Banerjee,
J. Am. Chem. Soc. 2006, 128, 8078 – 8086.
Experimental Section
Bu3SnH (0.4 mmol, 2 equiv, 108 mL) and AIBN (0.24 mmol, 1.2 equiv,
39 mg) in benzene (3 mL) were added over 6.5 h (0.06 mmolhÀ1
relative to Bu3SnH) to a degassed solution of 1b (0.20 mmol,
58 mg) and AIBN (0.06 mmol, 0.3 equiv, 10 mg) in benzene (9 mL)
at reflux. The resulting solution was heated at reflux for an additional
hour, cooled to RT and concentrated in vacuo. The residue was
purified by flash column chromatography on silica gel (eluent:
pentane/EtOAc/EtOH = 5:5:1) to afford the desired guanidine 2b
~
(40 mg, 76%) as a white solid. m.p. = 237–2398C; IR (neat): n = 1681,
1650, 1608, 758, 693 cmÀ1; 1H NMR (400 MHz, CDCl3): d = 3.98–4.14
(m, 1H, NCHH), 4.66 (dd, J = 11.8, 9.4 Hz, 1H, NCHH), 5.17 (t, J =
8.3 Hz, 1H, PhCH), 6.86 (d, J = 8.2 Hz, 1H, arom), 7.15 (t, J = 7.5 Hz,
1H, arom), 7.34–7.50 (m, 6H, arom), 7.57 (bs, 1H, NH), 8.11 ppm (d,
J = 8.0 Hz, 1H, arom); 13C NMR (100 MHz, CDCl3): d = 51.2
(NCH2), 56.3 (PhCH), 118.1 (C, arom), 123.1 (CH, arom), 124.2
(CH, arom), 126.5 (CH, arom), 126.8 (CH, arom), 129.1 (CH, arom),
129.5 (CH, arom), 134.5 (CH, arom), 140.2 (C, arom), 150.3 (C,
=
=
arom), 154.3 (NC N), 161.0 ppm (C O); HRMS: m/z calcd for
C16H14N3O ( [M+H]+): 264.1131; found: 264.1130.
Received: December 22, 2009
Published online: February 24, 2010
Keywords: azides · cyanamides · cyclization · heterocycles ·
.
radicals
[1] R. G. S. Berlinck, A. C. B. Burtoloso, M. H. Kossuga, Nat. Prod.
Rep. 2008, 25, 919 – 954.
[9] a) L. Benati, G. Bencivenni, R. Leardini, M. Minozzi, D. Nanni,
R. Scialpi, P. Spagnolo, G. Zanardi, C. Rizzoli, Org. Lett. 2004, 6,
417 – 420; b) L. Benati, G. Bencivenni, R. Leardini, D. Nanni, M.
Minozzi, P. Spagnolo, R. Scialpi, G. Zanardi, Org. Lett. 2006, 8,
2499 – 2502; c) H. Zhai, M. Zlotorzynska, G. Sammis, Chem.
Commun. 2009, 5716 – 5718; azides as source of N-centered
radicals were first introduced by Kim, see: d) S. Kim, G. H. Joe,
J. Y. Do, J. Am. Chem. Soc. 1993, 115, 3328 – 3329.
[2] a) F. Cohen, L. E. Overman, J. Am. Chem. Soc. 2001, 123,
10782 – 10783; b) M. A. Arnold, K. A. Day, S. G. Durꢀn, D. Y.
Gin, J. Am. Chem. Soc. 2006, 128, 13255 – 13260; c) J. J. Fleming,
M. D. McReynolds, J. Du Bois, J. Am. Chem. Soc. 2007, 129,
9964 – 9975; d) P. A. Evans, J. Qin, J. E. Robinson, B. Bazin,
Angew. Chem. 2007, 119, 7561 – 7563; Angew. Chem. Int. Ed.
2007, 46, 7417 – 7419; e) J. V. Mulcahy, J. Du Bois, J. Am. Chem.
Soc. 2008, 130, 12630 – 12631.
[3] P. Blondeau, M. Segura, R. Perez-Fernandez, J. De Mendoza,
Chem. Soc. Rev. 2007, 36, 198 – 210.
[4] Selected references: a) E. J. Corey, M. J. Grogan, Org. Lett.
1999, 1, 157 – 160; b) M. Terada, M. Nakano, H. Ube, J. Am.
Chem. Soc. 2006, 128, 16044 – 16045; c) D. Leow, S. Lin, S. K.
Chittimalla, X. Fu, C.-H. Tan, Angew. Chem. 2008, 120, 5723 –
[10] For a discussion of the radical rearomatization process which is
À
at the root of the formal radical C H activation, see: a) D. P.
Curran, H. Liu, J. Chem. Soc. Perkin Trans. 1 1994, 1377 – 1393;
b) D. P. Curran, H. Yu, H. Liu, Tetrahedron 1994, 50, 7343 – 7366;
c) S. M. Allin, W. R. S. Barton, W. R. Bowman, T. McInally,
Tetrahedron Lett. 2001, 42, 7887 – 7890; d) A. L. J. Beckwith,
V. W. Bowry, W. R. Bowman, E. Mann, J. Parr, J. M. D. Storey,
2180
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 2178 –2181

Angewandte
Chemie
Angew. Chem. 2004, 116, 97 – 100; Angew. Chem. Int. Ed. 2004,
43, 95 – 98; e) D. Crich, D. Grant, V. Krishnamurthy, M. Patel,
Acc. Chem. Res. 2007, 40, 453 – 463; f) S. M. Allin, W. R. S.
Barton, W. R. Bowman, E. Bridge (nꢄe Mann), M. R. J. Else-
good, T. McInally, V. McKee, Tetrahedron 2008, 64, 7745 – 7758;
when aryl iodides are involved in intramolecular radical
additions to quinolines, a single electron transfer process may
operate and only a catalytic amount of AIBN is required; see:
g) D. C. Harrowven, B. J. Sutton, S. Coulton, Tetrahedron 2002,
58, 3387 – 3400; h) D. C. Harrowven, I. L. Guy, L. Nanson,
Angew. Chem. 2006, 118, 2300 – 2303; Angew. Chem. Int. Ed.
2006, 45, 2242 – 2245.
[11] The corresponding cinnamyl substituted cyanamide decomposed
under the reaction conditions.
Angew. Chem. Int. Ed. 2010, 49, 2178 –2181
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2181