Synthesis of fluoroazaindolines by an uncommon radical ipso substitution of a carbon-fluorine bond

Article abstract of DOI:10.1021/ol101240f

(Equation Presented). Rare examples of a synthetically useful radical ipso substitution of a carbon-fluorine bond are reported. Highly functionalized fluoroazaindoline structures have thus been prepared with use of cheap and readily available substrates and reagents.

Full text of DOI:10.1021/ol101240f

ORGANIC
LETTERS
2010
Vol. 12, No. 15
3426-3429
Synthesis of Fluoroazaindolines by an
Uncommon Radical ipso Substitution of
a Carbon-Fluorine Bond
Yann Laot, Laurent Petit, and Samir Z. Zard*
Laboratoire de Synthe`se Organique, UMR 7652 CNRS/Ecole Polytechnique,
91128 Palaiseau, France
zard@dcso.polytechnique.fr
Received May 29, 2010
ABSTRACT
Rare examples of a synthetically useful radical ipso substitution of a carbon-fluorine bond are reported. Highly functionalized fluoroazaindoline
structures have thus been prepared with use of cheap and readily available substrates and reagents.
In chemistry, minor and seemingly harmless modifications
can have a profound effect on reactivity. We recently made
Scheme 1. Divergent Radical Cyclization on a Pyridine Ring
such a remarkable and very surprising observation while
studying a new approach to azaindolines and related sub-
stances.1 We found that whereas N-acetyl-protected xanthate
1a underwent the expected ring-closure to give azaindoline
2a in good yield, the corresponding N-Boc derivative 1b gave
only a very small amount of the analogous azaindoline 2b.
The major product turned out to be pyridinone 3, which was
formed by a hitherto undocumented radical attack on the
pyridine nitrogen (Scheme 1).2 While further work indicated
that the Boc group may be replaced by another carbamate
motif and the chlorine in the pyridine ring could be
exchanged for a fluorine, no completely satisfactory rationale
for this curious reactivity has yet emerged.3 In an attempt
to expand the scope of the process and perhaps gain some
additional mechanistic insight, we examined the behavior
of fluorinated derivative 6 (Scheme 2). This compound was
readily obtained by radical addition of xanthate 5a to Boc-
protected 2-N-allylamine pyridine 4.4 The latter was acces-
sible from commercially available pentafluoropyridine through
substitution, first with dimethylamine then with N-allylamine,
followed by reaction with Boc anhydride.5
(4) For reviews on xanthate chemistry, see: (a) Quiclet-Sire, B.; Zard,
S. Z. Top. Curr. Chem. 2006, 264, 201. (b) Quiclet-Sire, B.; Zard, S. Z.
Chem.sEur. J. 2006, 12, 6002. (c) Zard, S. Z. In Radical in Organic
Chemistry; Renaud, P., Sibi, M. P., Eds.; Wiley-VCH: Weinheim, Germany,
2001; Vol. 1, pp 90-108. (d) Quiclet-Sire, B.; Zard, S. Z. Phosphorus Sulfur
Silicon 1999, 153-154, 137–154. (e) Quiclet-Sire, B.; Zard, S. Z. J. Chin.
Chem. Soc. 1999, 46, 139. (f) Zard, S. Z. Angew. Chem., Int. Ed. Engl.
1997, 36, 672.
(1) Bacque´, E.; El Qacemi, M.; Zard, S. Z. Org. Lett. 2004, 6, 3671.
(2) El Qacemi, M.; Ricard, L.; Zard, S. Z. Chem. Commun. 2006, 4422.
(3) We wish to acknowledge in this respect our collaboration on the
theoretical aspects with Dr. Michelle Coote at the Australian National
University.
(5) Pentafluoropyridine usually reacts with amines first at the 4-position
then at the 2-position. See for example: Hargreaves, C. A.; Sanford, G.;
Slater, R.; Yufit, D. S.; Howard, J. A. K.; Vong, A. Tetrahedron 2007, 63,
5204. The reactivity of pentafluoropyridine and other fluorinated derivatives
is extensively discussed in ref 6j.
10.1021/ol101240f  2010 American Chemical Society
Published on Web 07/16/2010

The possibility of an ipso closure on carbon bearing
fluorine on the pyridine ring raised the question of the fate
of the cyclized radical in the absence of a second nitrogen
substituent to stabilize the corresponding cation. We therefore
examined the reaction of xanthate 14 derived from N-Boc-
N-allyl-4-aminotetrafluoropyridine 13. Upon treatment with
DLP in refluxing ethyl acetate, a new trifluoroazaindoline 17
was produced in about 40% crude NMR yield (Scheme 4).
Scheme 2. Unexpected Radical ipso Cyclization-Demethylation
Scheme 4
.
Azafluoroindoline by a Radical ipso Cyclization and
Fluorine Atom Elimination
Our anticipation was that the ring-closure on such a
structure could only take place on the pyridine nitrogen to
give 7 since the other position was blocked by a fluorine
atom. In the event, none of compound 7 could be observed
upon treatment of xanthate 6 with lauroyl peroxide (DLP).
Instead, azaindoline 8 was isolated in modest yield. Not only
did the ring-closure take place on carbon, with loss of a
fluorine atom, but a methyl group was also lost from the
dimethylamino group on C-4.
A plausible mechanism for this unexpected transformation
is pictured in Scheme 3. Radical 9, derived from xanthate
Scheme 3
.
Plausible Mechanism for Radical ipso
Cyclization-Demethylation
The formation of an azaindoline implies apparently the
loss of a fluorine atom by a homolytic rupture of a Very
strong C-F bond in intermediate 16. As far as we know,
such a fragmentation under comparably mild conditions is
quite rare. The decomposition of peroxides in perfluorinated
aromatic derivatives (normally used as the solvent) has been
reported to give products which could arise by a homolytic
ipso substitution of a carbon-fluorine bond. The yields were
generally low and considerable amounts of dimers of the
intermediate perfluorocyclohexadienyl radicals were ob-
(6) (a) Brooke, G. M. J. Fluorine Chem. 1997, 86, 1, and references
cited therein. We thank one of the referees for pointing out this reference
to us. ꢀ-Scission of chlorides, bromides, and iodides is well known: (b)
Struss, J. A.; Sadeghipour, M.; Tanko, J. M. Tetrahedron Lett. 2009, 50,
2119. (c) Maddess, M. L.; Mainetti, E.; Harrak, Y.; Brancour, C.; Devin,
P.; Dhimane, A.-L.; Fensterbank, L.; Malacria, M. Chem. Commun. 2007,
936. (d) Kim, S.; Kim, N.; Chung, W.-J.; Cho, C. H. Synlett 2001, 937. (e)
Tanko, J. M.; Sadeghipour, M. Angew. Chem., Int. Ed. 1999, 38, 159. (f)
Huval, C. C.; Singleton, D. A. Tetrahedron Lett. 1993, 34, 3041. (g) Kraus,
G. A.; Andersh, B.; Su, Q.; Shi, J. Tetrahedron Lett. 1993, 34, 1741. (h)
Knapp, S.; Gibson, F. S.; Choe, Y. H. Tetrahedron Lett. 1990, 31, 5397.
(i) Meijs, G. F.; Rizzardo, E.; Tang, S. H. Polym. Bull. 1990, 24, 501. (j)
For a review on C-F bond activation, see: Amii, H.; Uneyama, K. Chem.
ReV. 2009, 109, 2119.
6, undergoes an ipso substitution on C-5 to give intermediate
10, which is then converted into the corresponding allylic
cation 11 by electron transfer to the peroxide. This species
is destabilized by the adjacent electron-withdrawing fluorine
atoms but stabilized by the two exocyclic nitrogen substit-
uents. Loss of hydrogen fluoride then furnishes iminium
derivative 12. This species is readily hydrolyzed upon
workup to furnish novel difluoroazaindoline 8.
Org. Lett., Vol. 12, No. 15, 2010
3427

served.6 In our case, a favorable entropic term and the
rearomatization of the pyridine ring may constitute a sufficient
driving force for expelling a fluorine atom. Furthermore, the
nucleophilic character of the adduct radical and the electrophilic
character of the pyridine ring are complementary and may aid
the radical cyclization. Previous studies, especially by Walton
and Studer, have demonstrated the possibility of breaking strong
C-C and C-O bonds under similar circumstances.7 Cleavage
in the same manner of the stronger C-F bond, while obviously
more difficult, is not therefore unrealistic. An alternative
pathway can be envisaged involving first a solvolysis to give
radical-cation 18-18′ and a fluoride anion, followed by quench-
ing with a nucleophile in a medium such as water or lauric
acid (R′OH) leading to radical 19.8 This is then followed by
ꢀ-scission to give the observed azaindoline 17. Even if it cannot
yet be discounted, this variant seems unlikely, however, in view
of the poor leaving group ability of the fluoride anion and the
low dielectric constant of the solvent, which disfavors normally
a heterolytic process.9 Yet another pathway involves abstraction
by intermediate radical 16 of a hydrogen atom from the solvent
to give derivative 20, which can then ionically lose hydrogen
fluoride to furnish the observed product 17. However, radical
16 is highly stabilized (it is not only both tertiary and allylic,
but further stabilized by the lone pair of the nitrogen) and its
reaction with the solvent would be expected to be very
endothermic and in fact quite unlikely. The possibility of a prior
ionic elimination of HF from 16 followed by hydrogen
abstraction from the solvent is even less likely. Ionic elimination
of HF from 16 in the absence of a base in a nonpolar solvent
is too slow in comparison with the lifetime of the radical;
moreover, the resulting radical species would still be too
stabilized and incapable of hydrogen abstraction from the
solvent.
Table 1. Preparation of Fluoroazaindolines 23
From a synthetic standpoint, we found this reaction capri-
cious, giving variable and often low yields (hence the crude
NMR yield given above for azaindoline 17). The observation
of unidentified side products where the Boc group had appar-
ently been lost indicated a likely detrimental effect of the
hydrogen fluoride produced by either mechanistic pathway. We
therefore replaced the Boc group in 14 with the more robust
acetyl group. Working with analogue 22a allowed us to operate
at the much higher temperature of refluxing o-dichlorobenzene
(175 °C) and thus improve the chances of the intrinsically slow
cyclization, as well as the high entropy fragmentation step.10
At this temperature, it is necessary to replace the lauroyl
peroxide with di-tert-butyl peroxide, which has a much longer
lifetime. Under these conditions, the reaction was more repro-
ducible and cleaner, furnishing the desired trifluoroazaindoline
(7) For a review, see: Walton, J. A.; Studer, A. Acc. Chem. Res. 2005,
38, 794.
(8) (a) Beckwith, A. L. J.; Crich, D.; Duggan, P. J.; Yao, Q. Chem.
ReV. 1997, 97, 3273. (b) Crich, D.; Brebion, F.; Suk, D.-H. Top. Curr.
Chem. 2006, 263, 1.
(9) Shoute and Mittal generated a radical-anion from hexafluorobenzene
through radiolysis and noted the relative difficulty of eliminating a fluoride
anion. Eliminating fluoride from a simple free radical should therefore be
even more difficult. See: Shoute, L. C. T.; Mittal, J. P. J. Phys. Chem.
1993, 97, 379. Interestingly, these authors also remarked on the “consider-
able resistance” of hexafluorobenzene to radical addition. Both of these
observations make our present cyclization and extrusion of a fluorine atom
all the more unique.
23a in 50% yield along with 25% of prematurely reduced
material (24, R ) Me).
3428
Org. Lett., Vol. 12, No. 15, 2010

In the same manner, various trifluoroazaindolines were
prepared as shown by the results collected in Table 1.
Xanthates 5a-j bearing various functional groups added
readily to the allyl group in 21 allowing ultimately access
to a broad assortment of novel trifluoroazaindolines. Fur-
thermore, the presence of easily replaceable fluorines at
positions C-2 and C-6 of the pyridine ring opens countless
opportunities for further modification of these structures.11
Azaindolines and related derivatives are emerging as im-
portant motifs in medicinal chemistry, especially as kinase
inhibitors in the treatment of cancer.12 This is reflected by
the constantly increasing flux of recent publications dealing
with their synthesis. More generally, fluorine-containing
heterocycles represent a particularly interesting class for the
pharmaceutical and agrochemical industry.13
The present work complements existing methods and
highlights the hitherto unsuspected possibility of using a
fluorine atom as a radical leaving group. It also documents
an example of an unusual demethylation reaction in the
conversion of 6 into 8. Even if the yields are still relatively
modest because of the competing premature reduction of the
intermediate radical, it must be realized that such cyclizations
would be very hard, if not impossible, to accomplish with
existing methodology.
(10) For an example of the use of high temperature to force a difficult
cyclization on an aromatic ring, see: Quiclet-Sire, B.; Zard, S. Z. Chem.
Commun. 2002, 2306.
(11) The use of ring-fused trifluoropyridine systems in further synthetic
transformations has recently been described. See for example: (a) Sandford,
G.; Slater, R.; Yufit, D. S.; Howard, J. A. K.; Vong, A. J. Org. Chem.
2005, 70, 7208. (b) Cartwright, M. W.; Convery, L.; Kraynck, T.; Sandford,
G.; Yufit, D. S.; Howard, J. A. K.; Christopher, J. A.; Miller, D. D.
Tetrahedron 2010, 66, 519.
Acknowledgment. We respectfully dedicate this paper to
Professor A. Vasella (ETH, Switzerland). Y.L. and L.P. thank
Ecole Polytechnique and Laboratoires Servier, respectively,
for a studentship.
(12) For some very recent references, see: (a) Jeanty, M.; Blu, J.;
Suzenet, F.; Guillaumet, G. Org. Lett. 2009, 11, 5142. (b) Huestis, M. P.;
Fagnou, K. Org. Lett. 2009, 11, 1357. (c) Echalier, A.; Bettayeb, K.;
Ferandin, Y.; Lozach, O.; Cle´ment, M.; Valette, A.; Liger, F.; Marquet,
B.; Morris, J. P.; Endicott, J. A.; Joseph, B.; Meijer, L. J. Med. Chem.
2008, 51, 737. (d) Jeanty, M.; Suzenet, F.; Guillaumet, G. J. Org. Chem.
2008, 73, 7390. (e) Wipf, P.; Maciejewski, J. P. Org. Lett. 2008, 10, 4383.
(f) Ma, Y.; Breslin, S.; Keresztes, I.; Lobkovsky, E.; Collum, D. B. J. Org.
Chem. 2008, 73, 9610. (g) Fang, Y.-Q.; Yuen, Y.; Lautens, M. J. Org.
Chem. 2007, 72, 5152. For some recent reviews, see: (h) Popowycz, F.;
Routier, S.; Joseph, B.; Me´rour, J.-Y. Tetrahedron 2007, 63, 1031. (i)
Popowycz, F.; Joseph, B.; Me´rour, J. Y. Tetrahedron 2007, 63, 8689. (j)
Song, J. J.; Reeves, J. T.; Gallou, F.; Tan, Z.; Yee, N. K.; Senanayake,
C. H. Chem. Soc. ReV. 2007, 36, 1120.
Supporting Information Available: Experimental pro-
1
cedures, characterization, and copies of H and 13C NMR
spectra for all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL101240F
(13) Fluorinated Heterocyclic Compounds: Synthesis, Chemistry and
Applications; Petrov, V. A., Ed.; J. Wiley & Sons, Inc: Hoboken, NJ, 2009.
Org. Lett., Vol. 12, No. 15, 2010
3429