From a remarkable manifestation of polar effects in a radical fragmentation to the convergent synthesis of highly functionalized ketones

Article abstract of DOI:10.1021/ja400831w

A new radical addition/C-C bond fragmentation process is reported. Vinyl carbinols derived from 2-methyl-2-phenylpropanal react with radicals generated from xanthates to give the corresponding ketones. The radical cleavage reaction proceeds under mild conditions, in good to high yield, and in the presence of the unprotected carbinol. Highly functionalized 1,5-diketones and pyridines are readily available using this approach.

Full text of DOI:10.1021/ja400831w

Communication
pubs.acs.org/JACS
From a Remarkable Manifestation of Polar Effects in a Radical
Fragmentation to the Convergent Synthesis of Highly Functionalized
Ketones
Laurent Debien* and Samir Z. Zard*
̀
Laboratoire de Synthese Organique, CNRS UMR 7652 Ecole Polytechnique, 91128 Palaiseau Cedex, France
S
* Supporting Information
Scheme 1. Polar Effects in C−C Bond versus C−O Bond
Homolytic Cleavages
ABSTRACT: A new radical addition/C−C bond frag-
mentation process is reported. Vinyl carbinols derived
from 2-methyl-2-phenylpropanal react with radicals
generated from xanthates to give the corresponding
ketones. The radical cleavage reaction proceeds under
mild conditions, in good to high yield, and in the presence
of the unprotected carbinol. Highly functionalized 1,5-
diketones and pyridines are readily available using this
approach.
adical fragmentations represent powerful synthetic pro-
R
cesses that have been used for the creation of structurally
diverse alkenes. In practice, these are limited to scissions of C−
Sn,1 C−S,2 and C−O bonds.3 While homolytic cleavage of C−C
bonds is quite common for strained ring structures, especially
cyclopropanes and cyclobutanes,3a synthetically useful C−C
bond fragmentations triggered by ordinary carbon radicals (in
contrast to highly energetic alkoxy radicals) in open chain,
unstrained derivatives are rare.4
We recently reported that xanthates 2 react with various
fluoropyridyl ethers 1 via a radical addition/C−O bond
homolytic scission to yield the corresponding alkenes.3e−i In
the context of our study on trifluoromethyl-alkenes,3i we were
surprised to find that the reaction between fluoropyridyl ether 1a,
derived from 2-methyl-2-phenylpropanal, and xanthate 2a gave
enol ether 5a in good yield instead of the expected alkene 3a
(Scheme 1, Path B). This unexpected result prompted us to
search for the origin of this unusual reactivity. We thus decided to
study the effect of the R substituent on the fragmentation
process. We found that substitution of the trifluoromethyl
moiety by a methyl group in 1a completely modified the course
of the reaction and delivered the corresponding olefin 3b
(Scheme 1, Path A). A similar divergence of behavior was
observed with sulfone 1c and sulfide 1d, with the former leading
to C−C bond cleavage to give 5c and the latter undergoing C−O
rupture to yield 3d (Scheme 1). This sharp difference of
reactivity emphasizes the dramatic effect of an electron-
withdrawing group on the β-fragmentation of the intermediate
radical 4. Indeed, σ-bond acceptors such as CF3 and SO2Ph lower
the SOMO energy level of the intermediate radical 4 and
therefore increase the interaction between the SOMO of the
intermediate radical 4 and the HOMO of the cumyl benzylic C−
C bond. This enhanced interaction causes the preferential
scission of the comparatively weak benzylic C−C bond and the
extrusion of a stabilized cumyl radical to yield enol ethers 5a and
5c. Such a clear-cut distinction between two radical leaving
groups is very rare at best and has interesting synthetic
implications.5 We have indeed explored these observations to
develop a new route to synthetically useful ketones, and our
preliminary results are described herein.
We first examined the possibility of replacing the fluoropyr-
idoxyl moiety with an acetoxy group that is unable to act as a
radical leaving group under normal circumstances, so as to force
the extrusion of the cumyl radical, whatever the electronic nature
of the substituent. Thus, as depicted in Scheme 2, addition of
xanthate 2b to allylic acetate 6 resulted in the formation of the
expected enol acetate 7, but it was heavily contaminated with
indane 8 formed by cyclization of the intermediate radical onto
the phenyl ring.6 Moreover, the separation of these two products
proved to be difficult by standard purification methods. This
undesired cyclization is no doubt favored by the Thorpe−Ingold
effect exerted by the germinal dimethyl substituents. Clearly, the
Received: January 26, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja400831w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
fragmentation process. A broad variety of functional groups and
substitution patterns are tolerated, and both starting materials are
readily available.9,10 Noteworthy is the possibility of forming
complex trifluoromethyl ketones such as 10n and 10o, which
would be quite tedious to make by more conventional
approaches. The presence of a sulfonyl group in example 10h
is also synthetically interesting, since it allows the modulation of
the acidity of the various acidic protons in the molecule and can
later be reductively removed.11 In most of the examples, 2 equiv
of alkene were used, as this generally allowed reactions to go to
completion and simplified the purification, but an excess of
xanthate (2 equiv) may also be employed if necessary, without
adversely impacting the reaction yield (e.g. 10k, Figure 1).
As highlighted by adducts 10f, 10i, and 10n, our method is well
suited for the modification of the steroid side chains. In
particular, product 10i stemming from 9-fluorohydroxycortisone
derived xanthate testifies to the wide functional group tolerance
of the reaction. Xanthates 12a and 12b used to make 10f and 10n
were respectively prepared from bile acids 11a and 11b, using a
method we developed some years ago (Scheme 3).12 The steroid
Scheme 2. Design of an Adequate Olefinic Partner
β-scission of the benzylic C−C bond is a sluggish process that
competes only poorly with the radical cyclization onto the phenyl
ring. Unfortunately, the solution consisting of blocking the
unwanted cyclization by substituting the aromatic ring at the
ortho-positions was not practical.
Hence, we decided to accelerate the desired fragmentation by
weakening further the benzylic bond by increasing the
interaction between the lone pair of the oxygen atom and the
σ* orbital of the benzylic C−C bond (an anomeric type effect).
In practice, we found that removing the acetyl group and simply
using the parent allylic alcohol 9a as the olefinic partner eliminate
the side reaction leading to indanes by significantly increasing the
rate of the desired fragmentation. Thus, reaction of xanthate 2b
with olefin 9a using stoichiometric amounts of lauroyl peroxide
furnished the corresponding ketone 10a via its transient enol. In
a similar fashion, aldehyde 10b and ketone 10c were obtained
from vinyl sulfones 9b and 9c. The sharp contrast between the
reactivity of olefin 1b (Scheme 1) and olefin 9a (Scheme 2) is
noteworthy.
With these successful preliminary results in hand, we explored
the scope of the reaction. As indicated by the examples displayed
in Figure 1, this process is particularly adapted for the synthesis of
highly functionalized 1,5-diketones.7 Indeed, α-keto-xanthates
are particularly competent and versatile partners in radical
reactions8 and are easily accessible from the corresponding α-
haloketones.9 One of the keto groups is thus present in the
xanthate partner, and the second is generated through the
Scheme 3. Preparation of Bile Acid Derived Xanthates
examples, as well as some of the other examples of 1,5-diketones
in Figure 1, contain functional groups that would not be
compatible with the conditions needed for the ionic Michael type
additions commonly employed to assemble 1,5-diketones.
Complexity may be introduced by taking advantage of the fact
that the “normal” addition of a xanthate to an alkene leads to
another xanthate, which could then be subjected to this new
ketone forming reaction. One example of this modular approach
is depicted in Scheme 4. Thus, addition of xanthate 2c to N-
vinylphthalimide13 furnishes xanthate 13, and this, in turn, can be
Figure 1. Scope of the radical addition/C−C bond fragmentation reaction. Reaction conditions: Olefin (2.0 equiv), xanthate (1.0 equiv), addition of
lauroyl peroxide 30 mol %/h with respect to the xanthate (1.8−2.1 equiv of lauroyl peroxide, 6−7 h). For 10i, a 1:1 mixture of trifluoroethanol and 1,2-
dichloroethane was used as solvent. Reaction conditions for 10k: Olefin (1.0 equiv), xanthate (2.0 equiv), addition of lauroyl peroxide 30 mol %/h with
respect to the olefin (2.1 equiv of lauroyl peroxide, 7 h).
B
dx.doi.org/10.1021/ja400831w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Scheme 4. Modular Construction of Complex Ketones by Sequential Radical Reactions
made to react with vinyl sulfone 9c to give densely functionalized
phenylsulfonyl ketone 14 in good yield. An alternative route to
complex polyketones, also shown in Scheme 4, is to start with
chloroacetonyl xanthate 2d. After the first addition−fragmenta-
tion to alkene 9d, the chlorine atom may be substituted by
potassium O-neopentyl xanthate to give a new xanthate 15,
which in turn can be engaged in an addition−fragmentation on
another alkene 9e to finally give triketone 16 (Scheme 4).
The mildness of the experimental conditions and the general
tolerance of the process are further demonstrated by the
Figure 2. Synthesis of trisubstituted pyridines.
synthesis of ynones. Thus, reaction of xanthate 2e with
propargylic alcohol 9f furnished the expected product 10p,
accompanied by lactone 17 in a good overall yield (Scheme 5).
Scheme 6. Proposed Mechanism
Scheme 5. Preparation of Ynones
The latter is the result of ring closure of the enol of 10p onto the
activated anilide. Indeed, N-mesyl-3,5-dichloroaniline is a
respectable leaving group.14 On the other hand, when xanthate
2f was used as a partner, ynone 10q was readily obtained in good
yield (Scheme 5).
As a class, 1,5-diketones are of special importance since they
are immediate precursors for a variety of carbo- and hetero-
cycles.15 Pyridines, in particular, are readily obtained from 1,5-
diketones by a number of classical procedures. In the present
case, pyridines with unusual substitution patterns become easily
accessible. This is illustrated by the examples in Figure 2, where
1,5-diketones 10j, 10n, and 10g are converted into pyridines 18a,
18b, and 18c respectively upon treatment with excess
ammonium acetate in refluxing acetic acid.7e
reaction mixture showed that ester 21 was not formed under the
present conditions.17 2,3-Dimethyl-2,3-diphenylbutane 22 re-
sulting from the dimerization of two cumyl radicals 20 could
nevertheless be isolated from the crude mixture. This allows the
formulation of the mechanism depicted in Scheme 6. Radical R3
generated from xanthate 2 and lauroyl peroxide (step a) reacts
with olefin 9 to give intermediate radical 19 (step b), which
undergoes β-fragmentation (step c) to give enol 10′ and a
stabilized cumyl radical 20. Enol 10′ tautomerizes to the desired
ketone 10 (step d) while the cumyl radical combines to give the
observed dimer 22 (step e, Scheme 6).18
During our study, we wondered about the fate of the cumyl
radical 20 after its extrusion from intermediate 19 (Scheme 6,
step c). Based on our experience,16 we first envisioned that
lauroyl peroxide would act as an oxidant for the cumyl radical 20
to give undecylate ester 21. However, careful analysis of a typical
In summary, we have established a powerful and convergent
route to ketones which involves a rare cleavage of a C−C bond in
a nonstrained open chain structure. This new radical addition/
elimination process is tolerant of numerous functional groups
and provides a particularly easy access to 1,5-diketones.
C
dx.doi.org/10.1021/ja400831w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(5) Polar effects in radical reactions are extensively documented,
especially in additions to alkenes but less so in fragmentations: (a) De
Vleeschouwer, F.; Van Speybroeck, V.; Waroquier, M.; Geerlings, P.; De
Proft, F. Org. Lett. 2007, 9, 2721. (b) Fischer, H.; Radom, L. Angew.
Chem., Int. Ed. 2001, 40, 1340. (c) Laird, E. R.; Jorgensen, W. L. J. Org.
Chem. 1990, 55, 9. (d) Tedder, J. M. Tetrahedron 1982, 38, 313.
(6) The preparation of indanes by a radical cyclization onto aromatic
rings is a known but limited method: Ly, T.-M.; Quiclet-Sire, B.; Sortais,
B.; Zard, S. Z. Tetrahedron Lett. 1999, 40, 2533.
(7) 1,5-Diketones are generally prepared by ionic Michael additions;
see: (a) Narasaka, K.; Soai, K.; Aikawa, Y.; Mukaiyama, T. Bull. Chem.
Soc. Jpn. 1976, 49, 779. (b) Wei, L.-L.; Loh, T.-P. Tetrahedron 1998, 54,
6715. For recent enantioselective versions, see: (c) Li, W.; Wu, W.;
Yang, J.; Liang, X.; Ye, J. Synthesis 2011, 1085. (d) Chi, Y.; Gellman, H.
Org. Lett. 2005, 7, 4253. For a xanthate mediated formation of 1,5-
diketones, see: (e) Boivin, J.; Carpentier, F.; Jrad, R. Synthesis 2006, 10,
1664.
(8) For examples using α-keto-xanthates and general reviews of the
xanthate transfer, see: (a) Zard, S. Z. Angew. Chem., Int. Ed. Engl. 1997,
36, 672. (b) Quiclet-Sire, B.; Zard, S. Chem.Eur. J. 2006, 12, 6002.
(c) Quiclet-Sire, B.; Zard, S. Z. Top. Curr. Chem. 2006, 264, 201.
(d) Quiclet-Sire, B.; Zard, S. Z. Pure Appl. Chem. 2011, 83, 519. (e) El
Qacemi, M.; Petit, L.; Quiclet-Sire, B.; Zard, S. Z. Org. Biomol. Chem.
2012, 10, 5707.
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures as well as a compilation of spectral and
analytical data of all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
zard@poly.polytechnique.fr; debien@dcso.polytechnique.fr
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This article is dedicated to Professor Keith U. Ingold. Financial
support provided by the ANR is gratefully acknowledged.
REFERENCES
■
(1) (a) Rosenstein, I. J. In Radical Fragmentation Reactions in Radical in
Organic Synthesis; Renaud, P., Sibi, M. P., Eds.; Wiley-VCH: Weinheim,
2001; Vol. 1, Chapter 1.4, pp 50−71 and references cited therein. For
some applications of the use of C−Sn bond radical fragmentation, see:
(b) Mizuno, K.; Ikeda, M.; Toda, S.; Otsuji, Y. J. Am. Chem. Soc. 1988,
110, 1288. (c) Ryu, I.; Yamazaki, H.; Ogawa, A.; Sonoda, N. J. Am. Chem.
Soc. 1993, 115, 1187. (d) Oh, H.-S.; Cha, J. K. Tetrahedron: Asymmetry
2003, 2911. (e) Siegel, D. R.; Danishefsky, S. J. J. Am. Chem. Soc. 2006,
128, 1048.
(2) For examples of C−S bond fragmentation in the sulfide series, see:
(a) Lewis, S. N.; Winstein, S. J. Org. Chem. 1972, 37, 1478. (b) Curran,
D. P.; Yoo, B. Tetrahedron Lett. 1992, 33, 6931. (c) Barton, D. H. R.;
Crich, D. Tetrahedron Lett. 1984, 25, 2787. (d) Barton, D. H. R.; Crich,
D. J. Chem. Soc., Perkin Trans. 1 1986, 1603. For examples of C−S bond
fragmentation in the sulfone series, see: (e) Lin, P.; Hitham, G. J. Chem.
Soc., Chem. Commun. 1983, 1102. (f) Padwa, A.; Bullock, W. H.;
Dyszlewski, A. D. Tetrahedron Lett. 1987, 28, 3193. (g) James, P.; Felpin,
F.-X.; Landais, Y.; Schenl, K. J. Org. Chem. 2005, 70, 7985. (h) Quiclet-
Sire, B.; Zard, S. Z. J. Am. Chem. Soc. 1996, 118, 1209. (i) Kim, S.; Lim,
K.-C.; Kim, S.; Ryu, I. Adv. Synth. Catal. 2007, 349, 527. (j) Gong, J.;
Fuchs, P. L. J. Am. Chem. Soc. 1996, 118, 4486.
(9) For a recent review on the synthesis of α-haloketones, see: Erian, A.
W.; Sherif, S. M.; Gaber, H. M. Molecules 2003, 8, 793.
(10) All starting carbinols were prepared from commercially available
2-methyl-2-phenylpropanal and functionalized vinyl-organolithium. For
the Shapiro reaction, see: (a) Shapiro, R. H.; Lipton, M. F.; Kolonko, K.
J.; Buswell, R. L.; Cupuano, L. A. Tetrahedron Lett. 1975, 1811. For a
review, see: (b) Chamberlin, A. R.; Bloom, S. H. Org. React. 1990, 39, 1.
For the preparation of functionalized vinyl bromide from 2,3-
dibromopropene, see: (c) Bigot, A.; Breuninger, D.; Breit, B. Org.
Lett. 2008, 10, 5321. For the preparation of vinyl iodide from alkyne,
see: (d) Sugiyama, H.; Yokokawa, F.; Shioiri, T. Org. Lett. 2000, 2, 2149.
(e) Zheng, Z.-S.; Cheng, H.-N.; Liu, L.; Wang, F.-L. J. Am. Chem. Soc.
2011, 133, 11080. (f) Kawaguchi, S.; Ogawa, A. Org. Lett. 2010, 12,
1893. (g) Gao, F.; Hoveyda, A. H. J. Am. Chem. Soc. 2010, 132, 10961.
For the preparation of (1-(phenylthio)vinyl)lithium, see: (h) Foubelo,
F.; Gutierrez, A.; Yus, M. Tetrahedron Lett. 1999, 40, 8173.
(11) For a review of the reduction of α-sulfonyl-carbonyls, see: Naj
́
era,
C.; Yus, M. Tetrahedron 1999, 55, 10547.
(3) For reactions involving radical opening of epoxides, see:
(12) Zard, S. Z. Synlett. 2009, 3, 333. See the Supporting Information
for more details.
(a) Gansauer, A.; Pierobon, M. Rearrangements of Cyclopropanes and
̈
Epoxides in Radicals in Organic Synthesis; Renaud, P., Sibi, M. P., Eds.;
Wiley-VCH: Weinheim, 2001; Vol. 2, Chapter 3.3, pp 207−220 and
references cited therein. (b) Ichinose, Y.; Oshima, K.; Utimoto, K. Chem.
Lett. 1988, 1437. (c) Tanaka, S.; Nakamura, T.; Yorimitsu, H.; Oshima,
K.; Nakamura, T.; Yorimitsu, H.; Oshima, K. Synlett 2002, 4, 569.
(d) Charrier, N.; Gravestock, D.; Zard, S. Z. Angew. Chem., Int. Ed. 2006,
45, 6520. For the synthesis of various olefins from allylic alcohols
involving a C−O bond fragmentation, see: (e) Charrier, N.; Quiclet-
Sire, B.; Zard, S. Z. J. Am. Chem. Soc. 2008, 130, 8898. (f) Braun, M.-G.;
Quiclet-Sire, B.; Zard, S. Z. J. Am. Chem. Soc. 2011, 133, 15954.
(g) Debien, L.; Quiclet-Sire, B.; Zard, S. Z. Org. Lett. 2011, 13, 5676.
(h) Brioche, J.; Michalak, M.; Quiclet-Sire, B.; Zard, S. Z. Org. Lett. 2011,
13, 6296. (i) Debien, L.; Quiclet-Sire, B.; Zard, S. Z. Org. Lett. 2012, 14,
5118.
(13) Quiclet-Sire, B.; Revol, G.; Zard, S. Z. Tetrahedron 2010, 66, 6656.
(14) For precedents, see: (a) Fuchs, J. R.; Funk, R. L. J. Am. Chem. Soc.
2004, 126, 5068. (b) Nishigushi, A.; Ikemoto, T.; Tomimatsu, K.
Tetrahedron 2007, 63, 4048.
(15) For the preparation of carbocycles from 1,5-diketones, see for
example: (a) McMurry, J. E.; Flemming, M. P. J. Am. Chem. Soc. 1974,
96, 4708. (b) Rapson, W. S.; Robinson, R. J. Chem. Soc. 1935, 1285.
(c) Hajos, Z. G.; Parrish, D. R. J. Org. Chem. 1974, 39, 1612. (d) Brown,
H. C.; Racherla, U. S.; Singh, S. M. Synthesis 1984, 922. For the
́
preparation of heterocycles, see: (e) Seoane, C.; Soto, J. L.; Quintero, M.
Heterocyles 1980, 14, 337. For a review, see: (f) Kharchenko, V. G.;
Markova, L. I.; Fedotova, O. V.; Pchelintseva, N. V. Chem. Heterocycl.
Compd. 2003, 39, 1121.
(16) Lauroyl peroxide oxidizes intermediate radicals generated from
radical cyclizations onto aromatic rings: See ref 8.
(4) (a) Heng, R.; Zard, S. Z. Chem. Commun. 2011, 47, 3296.
(b) Kessabi, F. M.; Winkler, T.; Luft, J. A. R.; Houk, K. N. Org. Lett.
2008, 10, 2255. (c) Luft, J. A. R.; Winkler, T.; Murphy, F.; Kessabi, F. M.;
Houk, K. N. J. Org. Chem. 2008, 73, 8175. (d) Cholleton, N.; Gauthier-
Gillaizeau, I.; Six, Y.; Zard, S. Z. Chem. Commun. 2000, 535. For radical
(17) Undecylate ester 21 was prepared from 2-phenylpropan-2-ol by
standard methods. TLC and 1H NMR analyses showed no formation of
21 during the reaction. See Supporting Information for more details.
(18) The dimerization of such cumyl radicals has been reported
previously: Resendiz, M. J. E.; Garcia-Garibay, M. A. Org. Lett. 2005, 7,
371. In our case, a certain amount of disproportionation must have also
taken place leading to cumene and to 2-phenylpropene, both of which
are volatile and lost upon evaporation of the solvent.
́
́
based depolymerizations, see: (e) Fodor, C.; Bozi, J.; Blazsο, M.; Ivan, B.
Macromolecules 2012, 45, 8953. (f) Kruse, T. M.; Woo, O. S.; Wong, H.-
V.; Khan, S. S.; Broadbelt, L. J. Macromolecules 2002, 35, 7830. Cleavage
of radical cations leading to benzylic radicals have been described:
(g) Floreancig, P. E.; Tu, W. Org. Lett. 2007, 9, 2389. (h) Siders, J. R., II;
Wang, L.; Floreancig, P. E. J. Am. Chem. Soc. 2003, 125, 2406. (i) Kumar,
V. S.; Floreancig, P. E. J. Am. Chem. Soc. 2001, 123, 3842.
D
dx.doi.org/10.1021/ja400831w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX