Rate constants for the β-elimination of tosyl radical from a variety of substituted carbon-centered radicals

Article abstract of DOI:10.1021/jo026870b

The rate constants for the β-elimination of tosyl radical (Ts^?) from a series of carbon-centered radicals have been determined by using the radical clock methodology. Depending on the substituents R in Ts-CH₂-CH(?)R radicals, the rate constants at 293 K vary by more than 2 orders of magnitude in the range of 10³-10⁶ s^-1. The lowest values were found for the 2-naphthyl and carbamoyl substituents, whereas the benzyl substituent is located at the other extremity. The effect of the substituent upon the stabilization of the starting radical exerts a predominant influence in this reaction in decreasing the rate of fragmentation.

Full text of DOI:10.1021/jo026870b

Ra te Con sta n ts for th e â-Elim in a tion of Tosyl Ra d ica l fr om a
Va r iety of Su bstitu ted Ca r bon -Cen ter ed Ra d ica ls
Vitaliy I. Timokhin,†,‡ Ste´phane Gastaldi,† Miche`le P. Bertrand,*,† and
Chryssostomos Chatgilialoglu§
Laboratoire de Chimie Mole´culaire Organique, UMR 6517, Boite 562, Faculte´ des Sciences St J e´roˆme,
Universite´ d’Aix-Marseille III, Av. Escadrille Normandie-Niemen, 13397 Marseille Cedex 20, France, and
ISOF, Consiglio Nazionale delle Ricerche, Via P. Gobetti 101, 40129 Bologna, Italy
michele.bertrand@univ.u-3mrs.fr
Received December 18, 2002
The rate constants for the â-elimination of tosyl radical (Ts•) from a series of carbon-centered radicals
have been determined by using the radical clock methodology. Depending on the substituents R in
Ts-CH2-CH(•)R radicals, the rate constants at 293 K vary by more than 2 orders of magnitude in
the range of 103-106 s-1. The lowest values were found for the 2-naphthyl and carbamoyl
substituents, whereas the benzyl substituent is located at the other extremity. The effect of the
substituent upon the stabilization of the starting radical exerts a predominant influence in this
reaction in decreasing the rate of fragmentation.
SCHEME 1
In tr od u ction
Owing to the reversible character of their addition to
•
double bonds, sulfonyl radicals (RSO2 ) are versatile
intermediates that can be used either as entering or as
leaving groups, or both sequentially in complex cascade
processes.1 These radicals play a major role in the
technologically important free-radical copolymerization
of olefins with SO2 and offer routes to a great variety of
sulfones.1 Over the past decade, the number of publica-
tions exploiting the synthetic potential of these short-
lived species has increased considerably. During the
course of our studies on tosyl radical mediated cycliza-
tions of 1,6-dienes, we frequently came to the conclusion
that the rates of â-elimination of sulfonyl radicals might
have as much influence as the rates of the addition of
the tosyl radical to olefins on the course of these reac-
tions.2
However, despite their practical importance, there is
a lack of kinetic data for the â-elimination of sulfonyl
radicals from carbon-centered radicals. The rationaliza-
tion of their chemistry is based on the isolated report of
1978 by Wagner and co-workers,3 who have measured
rate constants for the â-cleavage of photogenerated
diradical 1 (Scheme 1). In particular, estimated kf values
of 7.8 × 105 s-1 for BuSO2 and 1.1 × 107 s-1 for PhSO2
•
•
based on ks ) 1 × 107 s-1 were reported. The kf value for
•
PhSO2• was 15 times faster than that for BuSO2 , which
at that time even sounded reasonable from the thermo-
chemistry of sulfonyl radicals. Indeed, it was reported
† Universite´ d’Aix-Marseille III.
§ Consiglio Nazionale delle Ricerche.
‡ Present address: Department of Chemistry, University of Wis-
consin-Madison, 1101 University Ave., Madison, WI 53706.
(1) (a) Bertrand, M. P.; Ferreri, C. In Radical in Organic Synthesis;
Renaud, P., Sibi, M., Eds.; Wiley-VCH: Weinheim, 2001; Vol. 2, pp
485-503. (b) Chatgilialoglu, C.; Bertrand, M. P.; Ferreri, C. In
S-Centered Radicals; Alfassi, Z. B., Ed.; Wiley: Chichester, 1999; pp
312-354. (c) Crich, D. In Organosulfur Chemistry; Page, P., Ed.;
Academic Press: London, 1995; Vol. 1, pp 49-88. (d) Bertrand, M. P.
Org. Prep. Proc. Int. 1994, 26, 257-290. (e) Chatgilialoglu, C.; Guerra,
M. In Supplement S: The Chemistry of Sulphur-Containing Functional
Groups; Patai, S., Rappoport, Z., Eds.; Wiley: Chichester, 1993; pp 363-
394. (f) Chatgilialoglu, C. In The Chemistry of Sulfones and Sulfoxides;
Patai, S., Rappoport, Z., Stirling, C. J . M., Eds.; Wiley: Chichester,
1988; pp 1089-1113.
(2) (a) De Riggi, I.; Gastaldi, S.; Surzur, J .-M.; Bertrand, M. P.;
Virgili, A. J . Org. Chem. 1992, 57, 6118-6125. (b) De Riggi, I.;
Nouguier, R.; Surzur, J .-M.; Bertrand, M. P.; J aime, C.; Virgili, A. Bull.
Soc. Chim. Fr. 1993, 130, 229-235. (c) Bertrand, M. P.; De Riggi, I.;
Lesueur, C.; Gastaldi, S.; Nouguier, R.; J aime, C.; Virgili, A. J . Org.
Chem. 1995, 60, 6040-6045. (d) Bertrand, M. P.; Gastaldi, S.;
Nouguier, R. Tetrahedron Lett. 1996, 37, 1229-1232. (e) Lesueur, C.;
Nouguier, R.; Bertrand, M. P.; Hoffmann, P.; De Mesmaeker, A.
Tetrahedron 1994, 50, 5369-5380. (f) Bertrand, M. P.; Lesueur, C.;
Nouguier, R. Carbohydrate Lett. 1995, 1, 393-398. For closely related
studies, see: (g) Wang, C.; Russell, G. A. J . Org. Chem. 1999, 64, 2346-
2352.
(3) Wagner, P. J .; Sedon, J . H.; Lindstrom, M. J . J . Am. Chem. Soc.
1978, 100, 2579-2580.
10.1021/jo026870b CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/28/2003
3532
J . Org. Chem. 2003, 68, 3532-3537

â-Elimination of Tosyl Radical
SCHEME 2
than the analogous bromides. The starting compounds 2
were synthesized by the standard reactions (eq 1) fol-
lowing literature procedure.2a,10
On the other hand the products 5 derived from the
reaction of radical 3 (Scheme 2) were prepared as
authentic samples by reaction of an appropriate olefin
with the p-toluenethiol followed by oxidation with oxone
(eq 2) and compared with the reaction products.11,12
that bond dissociation energies of PhSO2-Me were 14
kcal/mol lower than in MeSO2-Me, implying a strong
delocalization of the unpaired electron into the phenyl
group.4
Later EPR1f,5 and optical absorption6 spectra of sulfonyl
radicals indicated they are σ-type species with similar
spin density of the unpaired electron on the SO2 moiety.
Theoretical calculation supports these findings and shows
that the spin distribution for the unpaired electron in
Kin etics. According to the radical clock methodology,8,9
the relative rate constant kH/kf can be obtained by the
chain reaction of bromide or selenide 2 with Bu3SnH,
provided that conditions can be found under which the
radicals 3 are partitioned between the two reaction
channels, i.e., hydrogen transfer from Bu3SnH and
â-fragmentation (Scheme 2).
In addition, the trapping of tosyl radicals by Bu3SnH
must be efficient to ensure the irreversibility of the
fragmentation process.13 This scenario can be achieved
under pseudo-first-order conditions at an appropriate
temperature depending on the starting material. Under
these conditions eq 3 holds:
•
MeSO2 is 42% on sulfur and 44% on the oxygens,
•
whereas for PhSO2 there is very little difference, with
the corresponding figures being 43% and 39%.6 Moreover,
the DH298(RSO2-Cl) were measured for R ) Me or
Ph by photoacoustic calorimetry and were found to be
equal (70.5 kcal/mol) and it was also suggested that
DH298(RSO2-R′) are independent of R for a given R′.7
These findings are in contrast with the different kf value
•
•
for PhSO2 and BuSO2 in Scheme 1.
In the present study, we have made use of the radical
clock methodology8,9 to determine rate constants for the
•
â-elimination of p-CH3C6H4SO2 radical (Ts•) from a
series of carbon-centered radicals 3 bearing different
types of substituents in the R position (Scheme 2).
kH
kf
[5]
)
[Bu3SnH]0
(3)
[4]
Resu lts
A series of experiments was conducted in which the
bromides and/or selenides 2 were treated with a large
excess of tin hydride in known concentrations at various
temperatures, i.e., [Bu3SnH]0 = 10 × [2]. The concentra-
Sta r tin g Ma ter ia ls a n d Rea ction Refer en ces. The
choice of bromides and/or phenylselenides as the precur-
sors of radicals 3 (Scheme 2) for the present study was
guided by literature reports on the reactivity of TsBr and
TsSePh toward alkenes and on the stability of their
adducts.1,2 Bromides were chosen when possible, al-
though the selenide precursors are much more stable
1
tions of 5 and 4 were determined by H NMR analysis,
using pentamethylbenzene as an internal standard. The
[5]/[4]ratio varied in the expected manner with the
change in the Bu3SnH concentration at each tempera-
ture. The plots of [5]/[4] vs [Bu3SnH]0 provided values of
kH/kf, and the results obtained for radicals 3 are sum-
(4) Benson, S. W. Chem. Rev. 1978, 78, 23-35. Also, see: Herron, J .
T. In The Chemistry of Sulfones and Sulfoxides; Patai, S., Rappoport,
Z., Stirling, C. J . M., Eds.; Wiley: Chichester, 1988; pp 95-106.
(5) Chatgilialoglu, C.; Gilbert, B. C.; Norman, R. O. C. J . Chem. Soc.,
Perkin Trans. 2 1979, 770-775. Chatgilialoglu, C.; Gilbert, B. C.;
Norman, R. O. C. J . Chem. Soc., Perkin Trans. 2 1980, 1429-1436.
(6) Chatgilialoglu, C.; Griller, D.; Guerra, M. J . Phys. Chem. 1987,
91, 3747-3750.
(7) (a) Chatgilialoglu, C.; Griller, D.; Kanabus-Kaminska, J . M.;
Lossing, F. P. J . Chem. Soc., Perkin Trans. 2 1994, 357-360. (b) See,
also: Laarhoven, L. J . J .; Mulder, P.; Wayner, D. D. M. Acc. Chem.
Res. 1999, 32, 342-349.
(10) Gancarz, R. A.; Kice, J . L. J . Org. Chem. 1981, 46, 4899-4906.
(11) Hauser, F. M.; Ellenberg, S. R.; Glusker, J . P.; Smart, C. J .;
Carrell, H. L. J . Org. Chem. 1986, 51, 50-57.
(12) Trost, B. M.; Curran, D. Tetrahedron Lett. 1981, 22, 1287-1290.
(13) The rate constants for hydrogen abstraction by RSO2• radicals
(R ) Me, n-Pr, and n-Bu) from cyclohexane were estimated to be
ca. 2 × 102 M-1 s-1 at 393 K.14 The bond dissociation energies of
c-C6H11-H and Bu3Sn-H are 95.515 and 78 kcal/mol,7b respectively.
One would expect the reaction of tosyl radical with Bu3SnH to be much
faster because it is 17 kcal/mol more exothermic, and polar effects are
more favorable. Therefore, it is reasonable to assume that the
â-elimination of radical 3 is irreversible under the above-mentioned
experimental conditions.
(8) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317-323.
(9) Newcomb, M. Tetrahedron 1993, 49, 1151-1176. Newcomb, M.
In Radicals in Organic Synthesis; Renaud, P., Sibi, M., Eds; Wiley-
VCH: Weinheim, 2001; Vol. 1, pp 317-335.
J . Org. Chem, Vol. 68, No. 9, 2003 3533

Timokhin et al.
TABLE 1. Rela tive Ra te Con sta n ts for â-elim in a tion of Tosyl Ra d ica l (kf) vs Hyd r ogen Abstr a ction fr om Bu 3Sn H (kH
for Ra d ica ls 3a -e
)
a
b
Errors correspond to one standard deviation of at least three independent measurements. Taken from ref 21; see text for details.
c Calculated from the Arrhenius expression in eq 4. Calculated by taking the rate constant at 403 K and assuming log(A/s-1) ) 13; see
text.
d
marized in Table 1. The values of the intercepts obtained
from the plots are very small, near zero, as is expected
from eq 3.
The kH/kf ratios for radical 3b were also obtained at
different temperatures. Linear regression analysis of a
plot of log(kH/kf) vs 1/T yields the relative Arrhenius
parameters given in eq 4, where θ ) 2.3RT kcal/mol and
the errors correspond to one standard deviation.
possible source of experimental errors resides in the
partial consumption of the olefin via hydrostannation.18
Hydrostannation would lead to overestimating the kH/kf
ratio, and as a consequence the rate constant for the
fragmentation reaction would be underestimated. Al-
though the experiments were conducted in the presence
of a large excess of reducing agent compared to selenide
or bromide, we took care to use as low as possible an
absolute concentration of Bu3SnH, short reaction course,
and rate of conversion superior to 40%. Usually prepara-
tive hydrostannylations are conducted at high tin hydride
concentrations and long runs because the addition of tin
radical is reversible and high concentrations of tin
hydride favors the trapping of the adduct.19 However,
blank experiments were carried out in the case of olefin
CH2dCHC(O)N(i-Pr)2 (4b) that derived from the frag-
mentation of radical 3b. This olefin was selected because
kH
9.1 ( 0.5
log
(M-1) ) - (4.3 ( 0.3) +
(4)
kf
θ
It is gratyfing to see that the kH/kf value for radical 3a
is independent of whether the starting material is
bromide (2a ) or selenide (2a ′) and that the formation of
the undesired PhSeH byproduct is unimportant.16,17
A
(14) Horowitz, A.; Rajenbach, I. A. J . Am. Chem. Soc. 1975, 97, 10-
13.
(17) Simakov, P. A.; Martinez, F. N.; Horner, J . H.; Newcomb, M.
J . Org. Chem. 1998, 63, 1226-1232. Crich, D.; J iao, X.-Y.; Yao, Q.;
Harwood, J . S. J . Org. Chem. 1996, 61, 2368-2373.
(15) Lias, S. G.; Bartmess, J . E.; Liebman, J . F.; Holmes, J . L.; Levin,
R. D.; Mallard, W. G. J . Phys. Chem. Ref. Data 1988, 17, Suppl. 1.
(16) These results indicate that under our conditions the formation
of PhSeH as byproduct is unimportant.17 Indeed, the involvement of
PhSeH, which reacts with alkyl radicals approximately 3 orders of
magnitude faster than does Bu3SnH, would lead to an overestimated
kH/kf ratio.
(18) For the discussion of a similar situation, see: Halgren, T. A.;
Roberts, J . D.; Horner, J . H.; Martiner, F. N.; Tronche, C.; Newcomb,
M. J . Am. Chem. Soc. 2000, 122, 2988-2994.
(19) Davies, A. G. In Organotin Chemistry; VCH: Weinheim, 1997
and references therein.
3534 J . Org. Chem., Vol. 68, No. 9, 2003

â-Elimination of Tosyl Radical
the rate constant for the addition of Bu3Sn• radical is ca.
108 M-1 s-1 at 25 °C20 and the reverse fragmentation is
expected to be relatively slow, i.e, ideal situation for
hydrostannylation. When 4b was treated at 100 °C for 1
h in a 0.129 M solution of Bu3SnH in toluene ([Bu3SnH]0/
[4b] ) 173), 17% of the olefin was consumed. The same
experiment conducted at 80 °C on a 0.074 M solution of
Bu3SnH in benzene ([Bu3SnH]0/[4b] ) 70) led to 11%
consumption of 4b. These experiments lead to the conclu-
sion that the partial consumption of 4b via hydrostan-
nation did not interfere substantially in the determina-
tion of the kH/kf value. Therefore, potential loss of 4
through hydrostannation might induce a small error in
the determination of the kH/kf value.
between the values determined for kf and the IE of the
resulting olefin. The polar effect seems to play a minor
role in the â-elimination of the reputed electrophilic tosyl
radical. It is worth noting that the â-elimination of tosyl
radical from 3e is ca. 5 times slower than from 3a .
If polar contributions of the type 6 play an important role
in the transition state, one would expect any substituent
stabilizing the development of a positive charge on the
carbon atom (like the benzoyloxyl group in 3e) to acceler-
ate the â-fragmentation, and carbonyl groups would slow
it down even more efficiently than the aromatic substitu-
ent does. The resonance stabilization of the starting
radical exerts a major influence on the rate of â-frag-
mentation of tosyl radical.26 The extent of contribution
of polar effects and steric effects is difficult to evaluate.
The kf value for radical 3a is 1.5 × 106 s-1 as an
average of the two sets of experiments (RBr and RSePh).
This value is two times higher and seven times smaller
than the kf value reported by Wagner et al.3 for diradical
1, RSO2 being a BuSO2 and PhSO2 group, respectively
(Scheme 1). On the basis of the structural characteris-
tics5,6 and thermochemical data7 of sulfonyl radical (see
introduction for more details), the rate constant for the
â-elimination of RSO2• from the carbon-centered radical
7 should be independent of the nature of R, being aryl or
alkyl, for a given substituent X. Therefore, we suggest
that the kf values in Table 1 can be safely extended to
the â-elimination of alkanesulfonyl radicals.
Discu ssion
Table 1 also shows the kH values for the reaction of a
specific alkyl radical with Bu3SnH.21 It can be seen that
these values are very similar and in the range of 1-3 ×
106 M-1 s-1 at 293 K with the exception of benzyl-type
radical, which is 2 orders of magnitude slower. Indeed,
•
the kH values for the reaction of PhCH2 with Bu3SnH
are 3.0 × 104 at 293 K and 4.2 × 105 M-1 s-1 at 403 K.
Then kf values at 293 K (403 K for radical 3c) can be
calculated and are reported in the last column of Table
1. The temperature dependence of kf values for radical
3b can also be calculated from eq 4 because the Arrhenius
parameters for the reaction of R-amide secondary alkyl
radicals with Bu3SnH are reported by Newcomb and co-
workers to be log(A/M-1 s-1) ) 7.6 and Ea ) 1.5 ( 0.4
kcal/mol.22 The combination of these data yields the
temperature dependence: log(kf/s-1) ) (11.9 ( 0.6) - (10.6
( 0.9)/θ, where θ ) 2.3 RT kcal/mol. However, this
expression should be treated with caution since Newcomb
and co-workers22 also suggested caution for the Arrhenius
parameters of the reference reaction. They noted that the
log A is smaller than that found for the tin hydride
trapping of secondary alkyl radicals (log A ) 8.7).23 If
this last value is used with eq 4 then a log(A/s-1) ) 13 is
obtained for the â-elimination. However, there is no doubt
about the accuracy of kH values at 293 K and, conse-
quently, of kf in this temperature.24
Con clu sion
Radical clock methodology has been used to provide
the first set of â-elimination rate constants for tosyl
radical. In particular, bromides and/or selenides of type
2 react with Bu3SnH in the radical chain reaction shown
in Scheme 2 and afford a mixture of desired products (4
and 5). By the choice of starting radicals 3, we were able
to evaluate the influence of the nature of the substituent
on the â-elimination. The values are in the range of 103-
106 s-1 at room temperature. Radicals 3 having substit-
The kf values in Table 1 are strongly affected by the
nature of the substituent. For the fragmentation of
radical 3c a kf ) 4.5 × 103 s-1 at 293 K can be calculated
by taking the rate constant value at 403 K and assuming
a log(A/s-1) ) 13. The â-elimination of tosyl radical from
radical 3b, 3c, and 3d is ca. 2 orders of magnitude slower
than from radical 3a . This suggests that the carbonyl
groups, like the aromatic, substantially increase the
activation energy due to resonance stabilization of the
starting radical.25 No correlation can be established
(25) It is worth mentioning that the rate constants for the analogous
â-elimination of PhS• are in the range of 105-108 s-1 depending on
the stabilizing effect of the R-substituent in the carbon-centered radical;
see: Ito, O. In S-Centered Radicals; Alfassi, Z. B., Ed.; Wiley:
Chichester, 1999; pp 193-224. Chatgilialoglu, C.; Altieri, A.; Fischer,
H. J . Am. Chem. Soc. 2002, 124, 12816-12823.
(26) According to Timberlake (see: Timberlake, J . W. In Substituent
Effects in Radical Chemistry; Viehe, H. G., J anousek, Z., Mere´nyi, R.,
Eds; NATO ASI Series; Reidel: Dordrecht, 1986; Vol. 189, pp 271-
281) the stabilization of radicals (Me)2C•-X varies slightly in the series
C(dO)NH2 < C(dO)OMe < Ph (the rate constants for the decomposi-
tion of the corresponding azoalcanes (free from polar effects) differ by
1 order of magnitude from one to the next). The 2-naphthyl group is
likely to stabilize the carbon-centered radical more or less as efficiently
as a phenyl group. Of course, some polar contribution is not completely
excluded in the case of carbonyl substituents.
(20) Ingold, K. U.; Lusztyk, J .; Scaiano, J . C. J . Am. Chem. Soc.
1984, 106, 343-348.
(21) Chatgilialoglu, C.; Newcomb, M. Adv. Organomet. Chem. 1999,
44, 67-112.
(22) Musa, O. M.; Choi, S.-Y.; Horner, J . H.; Newcomb, M. J . Org.
Chem. 1998, 63, 786-793.
(23) Chatgilialoglu, C.; Ingold, K. U.; Scaiano, J . C. J . Am. Chem.
Soc. 1981, 103, 7739-7742.
(24) The kf values reported in Table 1 are associated with relatively
large errors (up to 30%) because the kH/kf were calculated on the basis
of the 1H NMR integration and the kH contain ca. 10% uncertainties.
When possible, it is better to use the relative rate constants for
planning synthesis.
J . Org. Chem, Vol. 68, No. 9, 2003 3535

Timokhin et al.
2-Meth yl-2-p h en ylsela n yl-3-(tolu en e-4-su lfon yl)-p r op i-
on ic a cid ter t-bu tyl ester (2d ) was prepared from 4d (116
mg, 0.816 mmol) and TsSePh. Column chromatography (EtOAc/
pentane, 0:100 to 10:90) gave 2d (88 mg, 0.194 mmol, 48%).
1H NMR: δ 1.39 (s, 9H), 1.67 (s, 3H), 2.34 (s, 3H), 3.48 (d, 1H,
J ) 13.8), 3.88 (d, 1H, J ) 13.8), 7.19-7.25 (m, 4H), 7.29-
7.36 (m, 1H), 7.46-7.49 (m, 2H), 7.66 (d, 2H, J ) 8.3). 13C
NMR: δ 22.0, 22.3, 28.1, 45.6, 65.3, 82.4, 126.7, 128.1, 129.5,
130.3, 130.4, 138.6, 138.8, 145.0, 170.8.
uents such as 2-naphthyl or carbamoyl exhibit the lowest
tendency to eliminate the tosyl radical as a result of
resonance stabilization.
Exp er im en ta l Section
Gen er a l. NMR spectra of CDCl3 solutions were recorded
at 300 MHz (1H) and 75 MHz (13C); J values are given in Hz.
Column chromatographies were performed on silica gel 60.
Initiators azobisisobutyronitrile, benzoyl and tert-butyl per-
oxides were commercially available and used without ad-
ditional purification. Pentamethylbenzene was used as inter-
nal standard for 1H NMR analysis of the crude product after
reaction. Tributyltin hydride was distilled and stored under
argon. Alkenes 4a , 4c, 4d , and 4e were commercially available
and used without additional purification. Amide 4b was
prepared from the appropriate acyl chloride2g and the corre-
sponding amine according to a standard procedure. TsBr and
TsSePh were prepared according to known procedures10,27 and
were dried under vacuum before use. Starting compounds 2
(bromides and phenylselenides) were synthesized by standard
methods 2a,10 and were purified by column chromatography on
silica gel. Spectral data for new compounds are listed below.
Gen er a l P r oced u r e for P r ep a r a tion of Br om o- or
P h en ylsela n yla lk yl Su lfon es (2). All reactions were carried
out under argon atmosphere. A solution of alkene 4 (1 mmol),
TsSePh (or TsBr) (1.3 mmol), and AIBN (0.13 mmol) in
degassed benzene (2 mL) was refluxed for 3-5 h (TLC
monitoring) (every 90 min additional portions of AIBN (5%)
were added). The solvent was evaporated, and the crude
product was purified by flash chromatography on silica gel
(EtOAc/pentane).
Ben zoic a cid 1-br om o-2-(tolu en e-4-su lfon yl)-eth yl es-
ter (2e) was prepared from 4e (105 mg, 0.709 mmol) and TsBr.
Column chromatography (EtOAc/pentane, 5:95 to 15:85) gave
1
2e (210 mg, 0.547 mmol, 77%). H NMR: δ 2.24 (s, 3H), 3.95
(dd, 1H, J ) 14.8, 1.5), 4.30 (dd, 1H, J ) 14.8, 10.4), 7.11-
7.18 (m, 3H), 7.35-7.43 (m, 2H), 7.56-7.61 (m, 1H), 7.67-
7.72 (m, 4H). 13C NMR: δ 21.9, 64.2, 67.0, 128.5, 128.8, 130.5,
130.6, 134.5, 136.3, 145.8, 163.3.
Gen er a l P r oced u r e for P r ep a r a tion of Red u ction
P r od u cts (5), a s Au th en tic Sa m p les. A solution of alkene
4 (1 mmol), TolSH (1.3 mmol) and AIBN (0.2 mmol) in
degassed benzene (2 mL) was refluxed for 3-5 h (TLC
monitoring) (every 90 min additional portions of AIBN (5%)
were added). The solvent was evaporated, the crude product
was dissolved in ethanol (10 mL) and cooled at 0°C, and then
a solution of Oxone (3.5 equiv) in water (9 mL) was added.
After stirring for 4 h, the mixture was diluted with water and
extracted three times with Et2O. The combined organic phases
were dried over sodium sulfate, concentrated, and purified by
flash chromatography (EtOAc/pentane).
1-P h en yl-3-(tolu en e-4-su lfon yl)-p r op a n e (5a ). Following
the general procedure, the title compound was prepared in 9%
yield. Column chromatography (EtOAc/pentane, 0:100 to 10:
90).1H NMR: δ 1.98-2.08 (m, 2H), 2.45 (s, 3H), 2.69 (t, 2H,
J ) 7.5), 3.03-3.08 (m, 2H), 7.09-7.11 (m, 2H), 7.17-7.29 (m,
3H), 7.34 (d, 2H, J ) 7.9), 7.76 (d, 2H, J ) 8.3). 13C NMR: δ
22.0, 24.7, 34.5, 55.9, 126.8, 128.5, 128.8, 129.0, 130.3, 136.5,
140.3, 145.0.
N ,N -D iis o p r o p y l-3-(t o lu e n e -4-s u lfo n y l)-p r o p io n a -
m id e (5b). Following the general procedure, the title com-
pound was prepared in 59% yield. Column chromatography
(EtOAc/pentane, 0:100 to 30:70).1H NMR: δ 1.19 (d, 6H, J )
6.8), 1.30 (d, 6H, J ) 6.8), 2.45 (s, 3H), 2.77-2.85 (m, 2H),
3.40-3.48 (m, 3H), 3.94 (sept, 1H, J ) 6.8), 7.36 (d, 2H, J )
8.3), 7.80 (d, 2H, J ) 8.3). 13C NMR: δ 20.5, 20.8, 21.6, 27.6,
45.9, 48.4, 52.2, 127.9, 129.9, 136.3, 144.8, 167.1.
1-[2-(Tolu en e-4-su lfon yl)-eth yl]-n a p h th a len e (5c). Fol-
lowing the general procedure, the title compound was prepared
in 65% yield. Column chromatography (EtOAc/pentane, 0:100
to 15:85).1H NMR: δ 2.41 (s, 3H), 3.16-3.21 (m, 2H), 3.40-
3.45 (m, 2H), 7.18-7.25 (m, 1H), 7.32 (d, 2H, J ) 7.9), 7.39-
7.46 (m, 2H), 7.52 (s, 1H), 7.70-7.78 (m, 3H), 7.81 (d, 2H, J )
8.1). 13C NMR: δ 22.0, 29.5, 57.9, 126.2, 126.7, 126.9, 127.2,
127.9, 128.0, 128.5, 128.9, 130.4, 132.7, 133.9, 135.3, 136.4,
145.2.
2-Meth yl-3-(tolu en e-4-su lfon yl)-p r op ion ic Acid ter t-
Bu tyl Ester (5d ). Following the general procedure, the title
compound was prepared in 64% yield. Column chromatography
(EtOAc/pentane, 0:100 to 10:90).1H NMR: δ 1.27 (d, 2H, J )
7.2), 1.40 (s, 9H), 2.44 (s, 3H), 2.82-2.93 (m, 1H), 3.00 (dd,
1H, J ) 14.3, 5.4), 3.65 (dd, 1H, J ) 14.2, 7.2), 7.36 (d, 2H,
J ) 7.9), 7.79 (d, 2H, J ) 8.3).13C NMR: δ 18.4, 22.0, 28.2,
36.0, 59.1, 81.7, 128.5, 130.3, 136.8, 145.2, 173.1.
Ben zoic a cid 2-(tolu en e-4-su lfon yl)-eth yl ester (5e) was
prepared as described in the literature.28 1H NMR: δ 2.35 (s,
3H), 3.59 (t, 2H, J ) 5.9), 4.65 (t, 2H, J ) 5.9), 7.26-7.29 (m,
2H), 7.33-7.38 (m, 2H), 7.45-7.64 (m, 1H), 7.70-7.73 (m, 2H),
7.81 (d, 2H, J ) 8.1). 13C NMR: δ 22.0, 55.7, 58.8, 128.5, 128.6,
128.9, 129.4, 130.0, 130.4, 130.6, 133.7, 134.1, 136.9, 145.3,
166.2.
1-P h en yl-2-br om o-3-(tolu en e-4-su lfon yl)-p r op a n e (2a )
was prepared from 4a (99 mg, 0.838 mmol) and TsBr. Column
chromatography (EtOAc/pentane, 0:100 to 10:90) gave 2a (266
1
mg, 0.754 mmol, 90%). H NMR: δ 2.44 (s, 3H), 3.17 (dd, 1H,
J ) 14.5, 8.3), 3.49 (dd, 1H, J ) 14.5, 5.1), 3.60-3.70 (AB part
of an ABX spectra, 2H, J AB ) 13.8), 4.45-4.55 (m, 1H), 7.18-
7.37 (m, 7H), 7.79 (d, 2H, J ) 8.3). 13C NMR: δ 22.1, 44.6,
45.2, 63.3, 127.7, 128.5, 129.0, 129.9, 130.5, 136.7, 137.1, 145.7.
1-P h en yl-2-p h en ylsela n yl-3-(t olu en e-4-su lfon yl)-p r o-
p a n e (2a ′) was prepared from 4a (119 mg, 1.003 mmol) and
TsSePh. Column chromatography (EtOAc/pentane, 0:100 to 10:
1
90) gave 2a ′ (373 mg, 0.869 mmol, 87%). H NMR: δ 2.41 (s,
3H), 3.05 (dd, 1H, J ) 14.4, 8.3), 3.35 (dd, 1H, J ) 14.4, 3.4),
3.42-3.51 (m, 2H), 3.61-3.72 (m, 1H), 7.12-7.33 (m, 12H),
7.62 (d, 2H, J ) 8.3). 13C NMR: δ 22.1, 38.5, 40.2, 60.7, 127.4,
128.3, 128.5, 128.9, 129.7, 129.9, 130.4, 135.1, 136.5, 138.3,
145.2.
N,N-Diisopr opyl-2-ph en ylselan yl-3-(tolu en e-4-su lfon yl)-
p r op ion a m id e (2b) was prepared from 4b (100 mg, 0.64
mmol) and TsSePh. Column chromatography (EtOAc/pentane,
1
1:99 to 25:75) gave 2b (230 mg, 0.49 mmol, 77%). H NMR: δ
1.14 (d, 3H, J ) 6.8), 1.22-1.29 (m, 9H), 2.39 (s, 3H), 3.34
(sept, 1H, J ) 6.8), 3.45 (d, 1H, J ) 12.7), 4.12 (sept, 1H, J )
6.6), 4.29-4.50 (m, 2H), 7.20-7.38 (m, 5H), 7.48-7.52 (m, 2H),
7.70 (d, 2H, J ) 8.2). 13C NMR: δ 20.0, 20.3, 20.9, 21.6, 33.5,
46.2, 49.4, 59.3, 126.3, 128.0, 129.1, 129.4, 129.7, 135.5, 135.6,
136.4, 144.6, 166.4.
1-[1-Br om o-2-(t olu e n e -4-su lfon yl)-e t h yl]-n a p h t h a -
len e (2c) was prepared from 4c (111 mg, 0.721 mmol) and
TsBr. Column chromatography (EtOAc/pentane, 0:100 to 20:
1
80) gave 2c (169 mg, 0.435 mmol, 60%). H NMR: δ 2.13 (s,
3H), 4.11 (dd, 1H, J ) 14.6, 5.0), 4.24 (dd, 1H, J ) 14.6, 9.9),
5.53 (dd, 1H, J ) 9.9, 5.0), 6.85 (d, 2H, J ) 8.1), 7.26 (dd, 1H,
J ) 8.2, 2.0), 7.37 (d, 2H, J ) 8.3), 7.44-7.50 (m, 2H), 7.58-
7.61 (m, 2H), 7.67-7.73 (m, 2H). 13C NMR: δ 21.7, 44.8, 64.2,
124.8, 127.0, 127.4, 127.44, 127.9, 128.3, 128.5, 129.3, 129.8,
133.1, 133.7, 135.8, 136.1, 145.1.
(27) Truce, W. E.; Heuring, D. L.; Wolf, G. C. J . Org. Chem. 1974,
39, 238-244.
(28) Miller, A. W.; Stirling, C. J . M. J . Chem. Soc. C 1962, 2612-
2617.
3536 J . Org. Chem., Vol. 68, No. 9, 2003

â-Elimination of Tosyl Radical
Kin etic Exp er im en ts. Benzene, toluene, or o-xylene con-
taining a small amount of pentamethylbenzene as an internal
standard was used as solvent. The solution containing the
radical precursor 2 (0.001-0.01M) and the initiator was
flushed with a stream of argon. The flask was placed in a bath
maintained at a constant temperature. After thermal equili-
bration, ca. 10 equiv of Bu3SnH was injected with a syringe
through a rubber septum, and the reaction was initiated
thermally or irradiated with a water-cooled 150-W high-
pressure mercury lamp at a distance of ca. 0.1 m for a period
of 5-60 min. The crude mixture was analyzed by 1H NMR.
The products were identified by comparison of their 1H NMR
spectra with authentic samples.
Ack n ow led gm en t. We thank the Centre National
de la Recherche Scientifique for financial support.
Su p p or tin g In for m a tion Ava ila ble: 1H, 13C, and DEPT
HMR spectra and kinetic data. This material is available free
of charge via the Internet at http://pubs.acs.org.
J O026870B
J . Org. Chem, Vol. 68, No. 9, 2003 3537