Synthesis of new sulfonyloximes and their use in free-radical olefin carbo-oximation

Article abstract of DOI:10.1021/acs.orglett.5b00672

New bifunctional reagents for free-radical carbo-oximation of olefins have been developed. In this process, a single reagent can act both as a trap for nucleophilic radicals as well as a source of electrophilic radical via an α-scission of an alkylsulfonyl radical. This strategy involving the addition of a C-centered electrophilic radical and an oxime across the double bond of an electron-rich alkene is initiated with a t-BuO radical following an unusual mechanism, supported by both experiments and density functional theory calculations.

Full text of DOI:10.1021/acs.orglett.5b00672

Letter
pubs.acs.org/OrgLett
Synthesis of New Sulfonyloximes and Their Use in Free-Radical Olefin
Carbo-oximation
Benjamin Ovadia, Frederic Robert, and Yannick Landais*
́ ́
́
University of Bordeaux, Institute of Molecular Sciences, UMR-CNRS 5255, 351 cours de la liberation, Talence, 33405 Cedex, France
*
S Supporting Information
ABSTRACT: New bifunctional reagents for free-radical carbo-
oximation of olefins have been developed. In this process, a single
reagent can act both as a trap for nucleophilic radicals as well as a
source of electrophilic radical via an α-scission of an alkylsulfonyl
radical. This strategy involving the addition of a C-centered electrophilic radical and an oxime across the double bond of an
electron-rich alkene is initiated with a t-BuO radical following an unusual mechanism, supported by both experiments and density
functional theory calculations.
ddition of C-centered radicals to sulfonyloximes such as 1
A
(Figure 1) constitutes a straightforward one-carbon chain
elongation strategy and an efficient manner to incorporate a
masked aldehyde group. This methodology, first reported by
1
Kim and co-workers in 1996, has since been applied to various
2
transformations, including multicomponent carbonylation
3
reactions. Sulfonyloxime 1 was also used as a radical trap in
the carbo-oximation of olefins, leading to new functionalized
4
oximes such as 2 (Figure 1). This methodology allows the
incorporation of two new functional groups on the carbon
backbone of a large variety of olefins with the formation of two
σ-bonds. The carbo-oximation reaction relies on the addition of
an electrophilic radical species I, generated from a xanthate or
an alkyl halide, to the less hindered end of an electron-rich
olefin, generating a nucleophilic radical II, which can then be
trapped by the electrophilic sulfonyl acceptor 1. The PhSO2
radical released during the addition−β-elimination sequence
does not fragment under the above reaction conditions (80 °C)
to furnish a phenyl radical which could sustain the radical chain.
Hexabutylditin is thus added and upon reaction with the PhSO2
radical generates a tin radical, which is able to propagate the
chain. Unlike arylsulfonyl radicals, certain alkylsulfonyl radicals
Figure 1. Free-radical three-component carbo-oximation of olefins.
Scheme 1. Bifunctional Sulfonyloximes for Olefin Carbo-
oximation Reactions
can undergo (reversibly) an α-scission to afford an alkyl radical
5
and SO . Zard et al. have elegantly used this property to
2
develop tin-free allylation and vinylation of alkyl iodides and
6
7
xanthates. Based on these premises, Renaud and co-workers
recently developed an efficient carboazidation of olefins using
ethyl-2-(azidosulfonyl)ethanoate. This reagent was reported to
transfer to the olefin both an azido group and an electrophilic
radical of type I (electron-withdrawing group (EWG) =
CO Me) generated upon α-scission of the corresponding
2
alkylsulfonyl radical. To overcome the use of a stoichiometric
amount of hexabutylditin required in the three-component
olefin carbo-oximation depicted in Figure 1, it was envisioned
to develop bifunctional sulfones such as 3 (Scheme 1), which
would incorporate both an alkylsulfonyl group and an oxime.
Addition of a nucleophilic radical onto 3 would afford the
desired carbo-oximation product along with an alkylsulfonyl
radical, precursor of the electrophilic radical I. Herein, we
report on the synthesis of sulfones 3 and their reactions with
olefins under free-radical conditions. The scope and limitation
Received: March 6, 2015
Published: March 31, 2015
©
2015 American Chemical Society
1958
DOI: 10.1021/acs.orglett.5b00672
Org. Lett. 2015, 17, 1958−1961

Organic Letters
Letter
of the methodology has been studied, and a mechanism, which
includes an unusual initiation process supported by exper-
imental evidence and density functional theory (DFT)
calculations, is discussed.
using nucleophilic alkyl radicals, generated from di-tert-
butyldiazene (DTBD) or dilauroyl peroxide (DLP). Surpris-
ingly, while the former was reported to be very efficient in the
7
carboazidation reaction, no trace of the desired compound 8
The preparation of bifunctional sulfones was first carried out
extending Kim’s methodology, which includes a sulfanylation
was observed (Table 2, entry 1). Similarly, the ethyl radical
1
produced upon reaction of Et B with oxygen was unable to
3
9
of a chloro-oxime with a thiol, followed by the oxidation of the
initiate the process. More encouraging results were obtained
8
corresponding thioether into the desired sulfone 3a. However,
with DLP leading to 8, albeit in moderate yield (entry 3). A
catalytic amount (10 mol %) of an ethyl xanthate was then
added with DLP to generate the electrophilic radical precursor
while this approach was reliable to access 3a (EWG = CO Et;
2
see Supporting Information), it was found to be difficult to
generalize to other sulfones as the required thiols are not
readily available.
An alternative strategy was thus devised based on the
alkylation of O-benzylthiohydroxamate 4, with commercially
available alkyl halides 5a−d (Table 1). Compound 4 was easily
10
I (EWG = CO Et) (entry 4). The yield was effectively
2
improved compared to the same reaction without xanthate
(entry 3), but our efforts to further increase the reaction yield
unfortunately failed. Surprisingly, when the reaction was
achieved in the presence of di-tert-butyl hyponitrite
(
DTBHN) (0.15 equiv), a source of electrophilic alkoxy
Table 1. Preparation of Bifunctional Sulfonyloximes 3a−e
radicals, the oxime was totally consumed after only 3 h and the
carbo-oximation product 8 was formed, albeit with a modest
yield (entry 5). To avoid excessive consumption of reagent 3a,
0
.05 equiv of DTBHN was added first at 70 °C, with two
additional portions later (2 × 0.05). This effectively improved
the yield of 8, indicating that the bifunctional oxime 3a was able
to propagate the radical chain (entry 6). Addition of a catalytic
amount of ditin (10 mol %) was also tested to increase the
11
chain length. Under these conditions, 8 was obtained in good
yield (entry 7). These conditions, although not tin-free, were
considered optimal, the purification of the product being much
easier than in the original three-component process which
a
a
4
entry
sulfone
alkylation yield (%)
oxidation yield (%)
required 1.5 equiv of (Bu Sn) and 5 equiv of olefin.
3
2
1
2
3
4
3b
3c
3d
80
50
58
45
74
With the optimized experimental conditions in hand, various
olefins 9 were submitted to the carbo-oximation process in the
presence of acceptors 3a−e (Scheme 2). The reactions were
carried out starting from 1 equiv of oximes 3a−e and 2 equiv of
olefins 9, in the presence of 0.15 equiv of DTBHN in benzene
b
32
75
75
3e
b
a
Isolated yields. Conditions: K CO , TBAI, THF, 60 °C.
2
3
(
Scheme 2). Oximes 3a and 3b, leading to electrophilic radicals,
prepared by treatment of the corresponding O-benzylhydrox-
amate with Lawesson’s reagent (THF, reflux, 1 h, 90%; see
Supporting Information). Alkylation of 4 led to the desired
sulfanyloximes 6a−e, which were then oxidized with m-CPBA
into the desired bifunctional sulfonyl oximes 3a−e in moderate
to good overall yields.
Preliminary carbo-oximation experiments were carried out
using sulfonyloxime 3a (1 equiv) and allylsilane 7 (2 equiv) as
the olefinic partner (Table 2). Initiation was first performed
generally led to reasonable yields, although a slightly lower
efficiency was observed with less activated olefins, emphasizing
3b
the role of polar effects in these reactions. In contrast, yields
dropped dramatically with oximes 3c and 3e, which may be
related to the lower rate for the α-scission of the sulfonyl radical
in these cases. C-centered radicals α to a sulfonyl or a Weinreb
amide group were found to add efficiently onto electron-rich
4b
olefins in three-component processes, so that the poor yields
here cannot be associated with their low reactivity. The process
was also found to be sensitive to steric hindrance, as indicated
by the modest yields observed for 10m and 10n issued from the
corresponding sterically congested allylsilane and allylic ester.4
Finally, in contrast with the three-component process,
which affords exclusively the (Z)-oxime, a mixture of Z/E
isomers, in the favor of the former, is observed here. A tentative
reaction mechanism, closely related to that proposed by
Table 2. Olefin Carbo-oximation: Preliminary Studies
7
Renaud for the carboazidation process, was proposed, as
depicted in Scheme 3. This would involve an attack of the
electrophilic radical species I onto the olefin to generate a new
nucleophilic radical II (1), which could then add onto the
sulfonyloximes 3a−e to afford the desired product, along with
an alkylsulfonyl radical III (2), which after α-scission would
a
entry
initiator
DTBD
equiv
temp (°C)
additives
yield (%)
1
2
3
4
5
6
7
2 × 0.05
3 × 0.1
2 × 0.1
2 × 0.1
0.15
reflux
20
0
0
Et B/O2
3
DLP
reflux
reflux
70
39
50
34
60
79
,
bc
regenerate I and SO (3). Although this mechanism appeared
DLP
Xanth
2
plausible, the nature of the initiation step remained obscure.
Propagation of the radical chain was found to be effective
without tin additive and in the presence of a low amount of
initiator. The low efficiency of alkyl radicals generated from
DTBHN
DTBHN
DTBHN
3 × 0.05
3 × 0.05
70
c
2
70
(Bu Sn)
3
a
b
c
Isolated yields. Xanth: EtO CCH S(CS)OEt. 10 mol %.
DTBD, DLP, and Et B in the initiation step may be attributed
2
2
3
1
959
DOI: 10.1021/acs.orglett.5b00672
Org. Lett. 2015, 17, 1958−1961

Organic Letters
Letter
Scheme 2. Scope and Limitations of the Carbo-oximation Reaction between Oximes 3a−e and Olefins 9
to both steric and electronic effects. The absence of reactivity of
nucleophilic t-butyl radicals generated from DTBD may result
from the important steric hindrance around the oxime carbon
center, as suggested by the low yields obtained with sterically
congested olefins (e.g., 10m,n). In parallel, the low reactivity of
less nucleophilic primary alkyl radicals may be explained by the
low electrophilicity of oxime ethers (vide infra). In contrast, the
reactivity of DTBHN releasing an electrophilic alkoxy radical
was somehow surprising as 0.15 equiv of DTBHN is sufficient
to promote the reaction (Table 2, entry 6). t-BuO radicals are
known to decompose at high temperature (140 °C) into
From these computational studies, two kinetically favored
pathways emerged, leading to intermediates with the lowest
energy. The most favored route (blue) involved the formation
of a benzylic radical i′ resulting from the abstraction of a
hydrogen of 3a by the t-BuO radical. The second route (green)
resulted from an addition of the alkoxy radical onto the oxime
3a, generating an alkoxyaminyl radical i. This pathway, which is
only slightly more energetic, was less predictable but suggests
that oxime ethers are also prone to react with electrophilic
radical species as they form highly stabilized alkoxyaminyl
14
12
radicals. Three other possible pathways involving a hydrogen
abstraction of 3a by the t-BuO radical were considered to be
too high in energy and thus were discarded. The benzylic
radical i′ then led to benzaldehyde along with an iminyl radical
ii′, which upon fragmentation led to HCN and the radical
species III. In the alternative route, the alkoxyaminyl radical i
afforded III and an alkoxyoxime ii through a single step,
exhibiting a very low energy barrier. The observation of
benzaldehyde in our experiments suggests that the initiation
pathway proposed in Scheme 3 and supported by the
calculations below is likely the favored one. However, although
the alkoxyoxime ii could not be detected, the second more
favorable (green) route is on the computation error margin and
cannot be totally ruled out based on our calculations. The role
of the catalytic amount of tin is not yet clear, but an attack of
acetone and methyl radical. Although our reactions are
performed at 70 °C, one cannot totally rule out an initiation
step resulting from a nucleophilic attack of a small amount of
reactive methyl radical onto oximes 3. However, identification
1
of traces of benzaldehyde from the crude H NMR and by GC-
MS analysis led us to propose an alternative (or concomitant)
mechanism where benzylic hydrogens of the oxime ether would
13
be abstracted first by a t-BuO radical (Scheme 3). The
Scheme 3. Tentative Mechanism
the nucleophilic Bu Sn radical onto the sulfonyl oxime,
3
followed by a β-fragmentation step, appears to be an additional
and plausible pathway to generate radical III then I.
In summary, we have developed efficient access to new
bifunctional sulfonyloximes that exhibit moderate to good
reactivity in radical olefin carbo-oximation reactions. This
strategy allows such processes to be carried out with only a
catalytic amount of tin and only 2 equiv of olefin, thus
facilitating final product purification. An original initiation
pathway was also discovered, which was supported by the
isolation of a fragmentation product and DFT calculations.
Further investigations on this unusual initiation process are
under scrutiny in our laboratory and will be reported in due
course.
resulting benzylic radical would then β-fragment, releasing
benzaldehyde, cyanhydric acid, and SO2 to generate the
electrophilic radical I. To get better insights into this initiation
step, DFT calculations at the M062X/6-31G+(d,p) level were
carried out using oxime 3a and t-BuO radical as model
substrate and reagent respectively. Several pathways were
envisioned, and the energies are depicted in Figure 2.
1
960
DOI: 10.1021/acs.orglett.5b00672
Org. Lett. 2015, 17, 1958−1961

Organic Letters
Letter
Figure 2. Computational studies on the initiation step at the M062X/6-31G+(d,p) level.
ASSOCIATED CONTENT
Supporting Information
(c) Le Guyader, F.; Quiclet-Sire, B.; Seguin, S.; Zard, S. Z. J. Am.
Chem. Soc. 1997, 119, 7410. (d) Bertrand, F.; Quiclet-Sire, B.; Zard, S.
Z. Angew. Chem., Int. Ed. 1999, 38, 1943. (e) Kim, S.; Lim, C. J. Angew.
Chem., Int. Ed. 2002, 41, 3265. (f) Kim, S.; Kim, S. Bull. Chem. Soc. Jpn.
2007, 80, 809.
■
*
S
Experimental procedures and product characterization data.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(
7) Weidner, K.; Giroult, A.; Panchaud, P.; Renaud, P. J. Am. Chem.
Soc. 2010, 132, 17511.
8) A small amount of the oxime below is also formed from 3a, likely
as a result of a Ramberg−Backlund rearrangement. Ramberg, L.;
Backlund, B. Ark. Kemi, Mineral. Geol. 1940, 13A, 50.
(
AUTHOR INFORMATION
Corresponding Author
■
̈
̈
*E-mail: y.landais@ism.u-bordeaux1.fr.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(
(
(
(
9) Darmency, V.; Renaud, P. Top. Curr. Chem. 2006, 263, 71.
10) Quiclet-Sire, B.; Zard, S. Top. Curr. Chem. 2006, 264, 201.
11) Curran, D. P.; Yoo, B. Tetrahedron Lett. 1992, 33, 6931.
12) Denisov, E. T.; Denisova, T. G.; Pokidova, T. S. Handbook of
ANR (MCR-RAD, 11-BS07-010-01) and the CNRS are
gratefully acknowledged for financial support. We thank Prof.
L. Feray (University of Aix-Marseille) for helpful discussions,
and A. Jacquet (University of Bordeaux) for preliminary
experiments.
Free Radical Initiators; Wiley & Sons: Hoboken, NJ, 2003.
(13) (a) Russell, G. A. J. Org. Chem. 1958, 23, 1407. (b) Gilliom, R.
D.; Ward, B. F. J. Am. Chem. Soc. 1965, 87, 3944. (c) Gilliom, R. D. J.
Org. Chem. 1985, 50, 4336.
REFERENCES
■
(14) (a) Bella, A. F.; Slawin, A. M. Z.; Walton, J. C. J. Org. Chem.
(
1) (a) Kim, S.; Lee, I. Y.; Yoon, J.-Y.; Oh, D. H. J. Am. Chem. Soc.
996, 118, 5138. (b) Kim, S.; Yoon, J.-Y. J. Am. Chem. Soc. 1997, 119,
982.
2
004, 69, 5926. (b) Godineau, E.; Schenk, K.; Landais, Y. J. Org. Chem.
1
2
008, 73, 6983.
5
(
2) (a) Kim, S.; Lim, K.-C.; Kim, S. Chem. Commun. 2007, 4507.
(
b) Schaffner, A.-P.; Darmency, V.; Renaud, P. Angew. Chem., Int. Ed.
2
(
006, 45, 5847.
3) (a) Ryu, I. Multicomponent Radical Reactions. In Multicomponent
Radical Reactions; Zhu, J., Bienayme, H., Eds.; Wiley-VCH: Weinheim,
́
Germany, 2005; p 169. (b) Liautard, V.; Landais, Y. Free-Radical
Multicomponent Processes. In Multicomponent Reactions in Organic
Synthesis, 2nd ed.; Zhu, J., Wang, Q., Wang, M. X., Eds.; Wiley-VCH:
Weinheim, Germany, 2014; p 401.
(
4) (a) Godineau, E.; Landais, Y. J. Am. Chem. Soc. 2007, 129, 12662.
b) Landais, Y.; Robert, F.; Godineau, E.; Huet, L.; Likhite, N.
Tetrahedron 2013, 69, 10073.
5) (a) Chatgilialoglu, C. In The Chemistry of Sulfones and Sulfoxides;
Patai, S., Rappoport, Z., Stirling, C. J. M., Eds.; Wiley: Chichester,
988; p 1089. (b) Bertrand, M. P. Org. Prep. Proced. Int. 1994, 26, 257.
6) (a) Bertrand, F.; Le Guyader, F.; Liguori, L.; Ouvry, G.; Quiclet-
Sire, B.; Seguin, S.; Zard, S. Z. C. R. Acad. Sci. 2001, 4, 547.
b) Quiclet-Sire, B.; Zard, S. Z. J. Am. Chem. Soc. 1996, 118, 1209.
(
(
1
(
(
1
961
DOI: 10.1021/acs.orglett.5b00672
Org. Lett. 2015, 17, 1958−1961