Photochemical alkylation of ketene dithioacetal S,S-dioxides. An example of captodative olefin functionalization

Article abstract of DOI:10.1021/jo990817e

Radical alkylation of some ketene dithioacetal S,S-dioxides failed through the tin hydride promoted chain process but was successfully performed through stoichiometric photochemical initiation, either by electron transfer or hydrogen abstraction. In the first case, alkyl radicals were produced from tetralkylstannanes (t-Bu-, i-Pr-, n-Bu-SnR₃) via radical cation fragmentation, while in the second case these were produced from alkanes (cyclohexane, adamantane) by benzophenone triplet. When bulky radicals (t-Bu, adamantyl) were involved, the addition occurred with complete diastereoselectivity.

Full text of DOI:10.1021/jo990817e

J . Org. Chem. 2000, 65, 297-303
297
P h otoch em ica l Alk yla tion of Keten e Dith ioa ceta l S,S-Dioxid es. An
Exa m p le of Ca p tod a tive Olefin F u n ction a liza tion
Ana Maria Gonza´lez-Cameno, Mariella Mella, Maurizio Fagnoni, and Angelo Albini*
Dipartimento di Chimica Organica, Universita`, V. Taramelli 10, 27100 Pavia, Italy
Received May 18, 1999
Radical alkylation of some ketene dithioacetal S,S-dioxides failed through the tin hydride promoted
chain process but was successfully performed through stoichiometric photochemical initiation, either
by electron transfer or hydrogen abstraction. In the first case, alkyl radicals were produced from
tetralkylstannanes (t-Bu-, i-Pr-, n-Bu-SnR3) via radical cation fragmentation, while in the second
case these were produced from alkanes (cyclohexane, adamantane) by benzophenone triplet. When
bulky radicals (t-Bu, adamantyl) were involved, the addition occurred with complete diastereo-
selectivity.
Sch em e 1
Captodative olefins have been shown to be convenient
substrates for C-C bond formation via cycloaddition and
addition reactions.1 Ketene dithioacetal S,S-dioxides, as
an example, are easily prepared;2 â-alkylation of such
derivatives may lead to highly substituted R-thiosulfones.
These, in turn, are convenient synthetic intermediates
in view of several possible elaborations, e.g., C-alkylation
and -arylation,3 substitution of an allyl for a sulfone
group,4 desulfurization,5,6 pyrolysis to yield a carbonyl
derivative,7 Pummerer rearrangement.8 Ketene dithio-
acetal S,S-dioxides have been shown to act as efficient
acceptors of carbenes9 as well as of 1-hydroxy- (or
alkoxy-)alkyl radicals.5,10 Furthermore, intramolecular
radical addition of â-(ω-iodoalkyl) derivatives has been
shown to occur under classic radical chain conditions
yielding five-, six-, and also four-membered cyclic deriva-
tives, a result rationalized on the basis of the peculiar
stability of the adduct radical due to the captodative
effect.11 Somewhat surprisingly, no corresponding inter-
molecular radical alkylation has been reported except for
the above-mentioned case of hydroxyalkyl radicals. It
appeared worthwhile to test the generality of such a
reaction since this would widen the scope of thiosulfones
synthesis. We thus explored the radical alkylation of
these substrates and found that, while standard tin
hydride methods were ineffective, a satisfactory addition
could be obtained by photochemical means. Bulky radi-
cals could be added under this condition with excellent
diastereoselectivity.
* To whom correspondence should be addressed.
(1) (a) Viehe, H. G.; J anousek, Z.; Merenyi, R. Acc. Chem. Res. 1985,
18, 148. (b) Viehe, H. G.; Merenyi, R.; J anousek, Z. Pure Appl. Chem.
1988, 60, 1635. (c) Beckhaus, H. D., Ruchardt, C. Ang. Chem., Int. Ed.
Engl. 1987, 26, 770. (d) Boucher, J . L.; Stella, L. Tetrahedron 1986,
3871. (e) Doepp, D.; Pies, M. J . Chem. Soc., Chem. Commun. 1987,
1734. (f) Lahousse, F.; Merenyi, R.; Desmurs, J . R.; Allaime, H.;
Borghese, A.; Viehe, H. G. Tetrahedron Lett. 1984, 25, 3823. (g) Park,
S. U.; Chung, S. K.; Newcomb, M. I. J . Am. Chem. Soc. 1986, 108,
240. (h) Baldock, R. W.; Hudson, P.; Katritzky, A. R.; Soti, F. J . Chem.
Soc., Perkin Trans. 1 1974, 1422. (i) Katritzky, A. R.; Zerner, M. C.;
Karelson, M. M. J . Am. Chem. Soc. 1986, 108, 7213.
(2) Ogura, K.; Iihama, T.; Kiuchi, S.; Kajiki, T.; Koshikawa, O.;
Takahashi, K.; Iida, H. J . Org. Chem. 1986, 51, 700.
(3) Tsunoda, T.; Nagaku, M.; Nagino, C.; Kawamura, Y.; Ozaki, F.;
Hioki, H.; Ito, S. Tetrahedron Lett. 1995, 36, 2531. Haberman, A. K.;
J ulia, M.; Verpeaux, J . N.; Zhang, D. Bull. Soc. Chim. Fr. 1994, 131,
965. Torisawa, Y.; Satoh, A.; Ikegami, S. Tetrahedron Lett. 1988, 29,
1729.
(4) Simpinks, N. S. Tetrahedron 1991, 47, 323.
(5) Ogura, K.; Kayano, A.; Sumitani, N.; Akazome, M.; Fujita, M.
J . Org. Chem. 1995, 60, 1106.
(6) Kayano, A.; Akazome, M.; Fujita, M.; Ogura, K. Tetrahedron
1997, 53, 12101.
(7) Wladislaw, B.; Marzorati, L.; Uchoa, R. B. Synthesis 1986, 964;
Wladislaw, B.; Marzorati, L.; Uchoa, R. B. Phosphorus Sulfur 1987,
33, 87.
Resu lts
Attem p ted Tin Hyd r id e In d u ced Alk yla tion . We
focused our attention on three ketene dithioacetal S,S-
dioxides, viz. 1-(methylthio)-1-[(4-methylphenyl)sulfonyl)]-
ethene (1a ) and (E)-1-(methylthio)-1-[(4-methylphenyl)-
sulfonyl)]propene (1b) (both known as useful dienophiles
in cycloaddition reaction)9 as well as (E)-1-(2-methylbu-
tylthio)-1-[(4-methylphenyl)sulfonyl)]propene (1c, see
Scheme 1 for the synthesis). Alkene 1b was treated with
t-BuI in the presence of Bu3SnH/AIBN under standard
conditions12 at 110 °C for 7 h. The alkene was not
significantly consumed. Likewise, with Bu3SnCl/NaBH4
under UV irradiation, no reaction occurred (see Experi-
(8) Schank, K.; Buegler, S.; Schott, N. Phosphorus, Sulfur, Silicon,
Relat. Elem. 1994, 95&96, 435.
(9) Schank, K.; Abdel Wahab, A. A.; Buegler, S.; Eigen, P.; J ager,
J .; J ost, K. Tetrahedron 1994, 50, 3721.
(10) Ogura, K.; Yanagisawa, A.; Fujino, T.; Takahashi, K. Tetrahe-
dron Lett. 1988, 29, 5387.
(11) Ogura, K.; Sumitani, N.; Kayano, A.; Iguchi, H.; Fujita, M.
Chem. Lett. 1992, 1487.
(12) (a) Giese, B. Radicals in Organic Synthesis. Formation of
Carbon-Carbon Bonds; Pergamon: Oxford, 1986. (b) Burke, D. S.;
Fobare, F. W.; Armistead, M. D. J . Org. Chem. 1982, 47, 3348.
10.1021/jo990817e CCC: $19.00 © 2000 American Chemical Society
Published on Web 01/05/2000

298 J . Org. Chem., Vol. 65, No. 2, 2000
Gonza´lez-Cameno et al.
Sch em e 2
Sch em e 3
Sch em e 4
Ta ble 1. Alk yla tion Yield by Ir r a d ia tion of Alk en es 1
w ith Sta n n a n e 5 in th e P r esen ce of DCB/P h en
isomer distribution
irradn
product
(% on the total
adduct yield)
alkene stannane time (h)
(% yield)a
1a
1b
1b
1b
1c
5x
5x
5y
5z
5x
6
6
6
20
18
6a x (40), 7 (8)
6bx (52)
6by + 6′by (60)
6bz + 6′bz (59)
6cx (52)b
6 (100)
6/6′ (69/ 31)
6/6′ (50/50)
6 (100)
addition of the tosyl radical to the starting material (see
Scheme 3 for a rationalization). This side path caused a
decrease in the yield of 6a x (40%, compare with 50% with
1b as the substrate). Addition to alkene 1b led to the
formation of two diastereomers 6 and 6′ (two stereogenic
centers were formed). The stereoselectivity of the reaction
depended on the steric hindrance of the attacking alkyl
radical. With a tertiary radical (t-Bu), a single diastereo-
isomer was isolated and identified as the u (1R,2S/1S,2R)
diastereoisomer (6bx) with the NMR data. The presence
of bulky substituents made the rotation around C1-C2
bond hindered, so that the most favored rotamer was
observed. In the proton spectrum, a close to zero coupling
constant between H-1 and H-2 was observed, resulting
from a gauche arrangement with a dihedral angle ca. 90°.
The most stable rotamer had the t-Bu group near to H-1
(Scheme 4 shows the Newman projections along C1-C2
bond). The spatial arrangement of substituents was
derived from 1D-NOE difference and 2D-NOESY spectra
(see Experimental Section). A computational study was
carried out for the two t-Bu derivatives and confirmed
that 6bx was the low energy isomer. A conformational
search showed that the two diastereoisomers (6bx and
6′bx) exhibited four and six minima, respectively. The
most stable conformer of 6bx (Scheme 4) was 3.22 kcal/
mol lower in energy than the most stable conformer of
the other isomer (6′bx). The large computed difference
ensured that isomer 6bx was more stable than 6′bx and
confirmed the above NMR experimental evidence. The
situation changed with secondary or primary radicals.
Addition of an i-Pr radical gave both stereoisomers (6by
and 6′b y) with a low selectivity (6/6′ 2.2/1). As in the
previous case, the rotation around C1-C2 bond was
hindered for the major one (6by); the coupling constant
between H-1 and H-2 (1.5 Hz) was consistent with a
dihedral angle of ca. 60° (Scheme 4). The configuration
of both diastereisomers was attributed on the basis of
NMR data (see Experimental Section). With the n-butyl
radical, no stereoselection was observed (6bz/6′bz 1/1).
Introducing a bulky group R′′ in the sulfide moiety caused
a
b
Isolated yields. Mixture of diastereoisomers (in 1 to 1 ratio)
due to the presence of a further stereogenic center with R′′ group
() 2-methylbutyl).
mental Section), showing that the alkylation of ketene
dithioacetal S,S-dioxides by using radical-chain methods
in the intermolecular mode is ineffective. We thus turned
to stoichiometric methods for producing the key alkyl
radicals. These were obtained in two ways, through the
recently developed photoinduced electron transfer (PET)
method and by photochemical hydrogen transfer.
Alk yla tion via P h otoin d u ced Electr on Tr a n sfer .
This is based upon the oxidation of a donor by an excited
acceptor and the fragmentation of the resulting radical
cation.13 Tetraalkylstannanes were chosen as donors, and
various acceptors were tested. These were 1,4-dicy-
anobenzene (DCB), 1,4-dicyanonaphthalene, 1,2,4,5-tet-
racyanobenzene, and tetramethyl pyromellitate. All of
these acceptors worked to some extent, but the best
results were obtained by using DCB as electron-acceptor
and phenanthrene (Phen) as a secondary donor. This
allowed us to carry out the reaction by using Pyrex-
filtered light which is absorbed by Phen. The reactivity
of alkene 1 was tested by employing three stannanes as
a radical source: t-BuSnMe3 (5x), i-PrSnMe3 (5y), and
Bu4Sn (5z). Under these conditions, the alkylation was
successful, and the results are shown in Scheme 2 and
Table 1.
The isolated yields of the thiosulfones 6 were moderate
(40-60%), in part due to the fact that precipitation of
tin-containing byproducts during the irradiation made
it difficult to carry out the reaction to completion. The
reaction generally led cleanly to adducts 6, as shown by
GC and TLC, except in the case of 1a , where disulfone 7
was formed as a side product. This resulted from the
(13) (a) Saeva, F. D. Top. Curr. Chem. 1991, 156, 59 (b) Albini, A.;
Mella, M.; Freccero, M. Tetrahedron 1994, 50, 575.

Captodative Olefin Functionalization
J . Org. Chem., Vol. 65, No. 2, 2000 299
Ta ble 2. Alk yla tion Yield by Ir r a d ia tion of Alk en es 1
w ith Hyd r oca r bon s in th e P r esen ce of Ben zop h en on e
Sch em e 6
isomer distribution
hydro- irradn
alkene carbon time (h)
product
(% on the total
adduct yield))
(% yield)a
1b
1b
8j
8k
15
18
9bj + 9′bj (89)
9bk (90)
9/9′ (82/18)
9 (100)
a
Isolated yields.
Sch em e 5
ence of an electron-donating and an electron-withdrawing
group (“captodative effect” or “merostabilization”).1h,i
From the synthetic point of view, this has been exploited
in the easy addition of various radicals to alkenes such
as R-amino or R-thio nitriles or esters.15a-d
no change in the reaction course, and addition of the tert-
butyl radical to 1c (R′′ ) 2-methylbutyl) occurred with
the same selectivity as in the case of 1b.
The presently studied R-thiosulfones 1a -c are known
to pertain to this category. However, our attempt to
alkylate these molecules through classical tin hydride
methods, both thermal or photochemical, led to no result
(in contrast vinyl sulfones undergo easy radical chain
alkylation under these conditions).15e We feel that this
can be rationalized by the peculiar captodative stabiliza-
tion of the adduct radical.11 It has previously been shown
that captodative radicals are poor hydrogen abstractors
and easily dimerize. In the present case, the stabilization
of the adduct radical makes hydrogen abstraction from
Bu3SnH too slow and precludes propagation of the
radical-chain process (Scheme 6).
Stoichiometric methods were successful, however. The
first method we used is based on photoinduced electron
transfer. When a light-absorbing acceptor A is excited
(A*), it becomes a strong oxidant [Ered (A*) ) Ered (A) +
Eexc (A)]. This oxidizes a donor RD,where D is a donating
moiety (eq 1), and the radical cation thus formed frag-
ments to generate an alkyl radical (R•) (eq 2).13
Th e Hyd r ogen Tr a n sfer P a th w a y. Another sto-
ichiometric method for generating a radical is photo-
chemical hydrogen abstraction by means of an nπ*
triplet, e.g., benzophenone.14 Ogura reported the high
yield functionalization of ketene thioacetal S,S-dioxides
with various hydroxy (alkoxy)alkyl radicals generated
from alcohols or ethers by irradiation with an equimo-
lecular amount of benzophenone.5,10 We explored whether
the same scheme could be applied to hydrocarbons
possessing no activated C-H bond. As it appears from
Table 2 and Scheme 5 the result was positive. The
reaction was first tested with the cyclohexyl radical. The
alkylation in neat cyclohexane occurred with a surpris-
ingly high yield (89%), and the stereoselectivity was
larger than with the isopropyl radical (9bj/9′bj 4.5/1,
compared to 6by/6′by 2/1, see Tables 1 and 2). Further-
more the radical coupling product cyclohexyldiphenyl-
methanol was detected only in traces. The reaction with
adamantane was also carried out. In this case the
selectivity of the alkylation (at the bridged or bridgehead
position) was a further interesting issue. The reaction
was best carried out in benzene (1 M, in acetonitrile the
yield was lower), and a single isomer, the 1-adamantyl
derivative 9bk , was isolated by column chromatography
(90%). Most of the adamantyl radicals were trapped by
the olefin, and only traces of side products (phenylada-
mantane, bisadamantane) were detected (see Experi-
mental Section). Traces of 1-adamantanol and, to an even
lesser extent, of 2-adamantanone were also detected and
arose from the reaction of the radicals with residual
oxygen.
A* + RD f A-• + RD+•
RD+• f R• + D+
(1)
(2)
(3)
A
-• + RD+• f A + RD
This method produces radicals from unconventional
donors [RD may be a tetraalkylstannane, a tetralkylsi-
lane, a 2,2-dialkyl (or 2-alkyl-2-aryl)dioxolane] under
mild conditions since the oxidant is the in situ generated
A*, and the use of a strong inorganic oxidant normally
employed in ET methods for radical reactions is avoided.
(15) (a) Stella, L.; J anousek, Z.; Merenyi, R.; Viehe, H. G. Angew.
Chem., Int. Ed. Engl. 1978, 17, 691. (b) Mignani, S.; J anousek, Z.;
Merenyi, R.; Viehe, H. G.; Riga, J .; Verbist, J . Tetrahedron Lett. 1984,
25, 1571. (c) Mignani, S.; Merenyi, R.; J anousek, Z.; Viehe, H. G.; Bull.
Soc. Chim. Belg. 1984, 93, 991. (d) Beaujean, M.; Mignani, S.; Merenyi,
R.; J anousek, Z.; Viehe, H. G.; Kirch, M. Lehn, J . M. Tetrahedron 1984,
40, 4395. (e) Mase, N.; Watanabe, Y.; Toru, T. Tetrahedron Lett. 1999,
40, 2797.
Discu ssion
A large body of theoretical and experimental evidence
supports the stabilization of a radical site by the copres-
(14) Wagner, P.; Park, B. S. Org. Photochem. 1991, 11, 227.

300 J . Org. Chem., Vol. 65, No. 2, 2000
Gonza´lez-Cameno et al.
Sch em e 7
presence of 0.1% D2O, no deuteration was observed in
the product by NMR spectroscopy (see Experimental
Section). This is in contrast with what observed in the
alkylation of R,â-unsaturated esters and nitriles by the
same method16 and is in accordance with the fact that
captodative carbanions are difficult to obtain and are
quite unstable. Indeed, such anions are known to be
readily oxidized to captodative radicals.1a Thus, we infer
that path a is preferred. Apparently, despite the mark-
edly negative reduction potential of DCB (Ered A/A-•
)
-1.6 V vs SCE), ET from the corresponding radical anion
to the adduct radical is slow. It may be that the radical
adduct accumulates as the dimer (see Scheme 6) and then
is slowly reduced to the final products 6.
The formation of product 7 from 1a is likely to result
from the liberation of tosyl radicals from the adduct
radical, again an indication that pathways different from
hydrogen abstraction are significant for such a species.
There is previous evidence for the elimination of tosyl
radicals from R-thiosulfones under radical conditions.21
The addition gives exclusively the u (1R,2S/1S,2R)
diastereoisomer with tertiary radicals, although a 1 to 1
mixture is obtained with primary radicals. This supports
that the selectivity depends on the energy difference
between the isomers 6 and 6′ as determined by the
repulsion between the entering radical and the p-tosyl
group. Thus, despite the different mechanism in the PET
pathway (the educt radical is generated through an
oxidative step, and then a reduction step completes the
sequence), the regio- and stereochemistry of the reaction
(the latter one determined in the final hydrogen abstrac-
tion step) are the same as in the classical radical-chain
alkylation.
The second method for the generation of radicals
involves triplet benzophenone. Hydrogen abstraction
from cyclohexane under these conditions in fluid media
has been thoroughly studied,22-24 but no intermolecular
addition of cyclohexyl radical produced by this pathway
has been hitherto reported. The addition occurs with
some stereoselectivity (4.5 to 1) with cyclohexane and
with complete selectivity when the adamantyl radical is
involved. The bulkiness of the latter species is obviously
responsible for this result (as in the t-Bu case). Notewor-
thy, no adduct containing the 2-adamantyl radical has
been obtained, despite the fact that hydrogen abstraction
from adamantane by benzophenone, as in general by nπ*
ketone triplets, is known to be rather unselective.25 It
thus appears that the selectivity arises at the addition
step.
Aromatic nitriles and esters are used as acceptors.
Recently we described the synthetic application of this
process for the addition of photogenerated radicals to
electron-poor olefins.16,17 When a synthesis of this type
is planned, some requirements must be met: (1) Ered (A*)
must be high enough to oxidize the donor; (2) Ered (A)
must be low enough to allow the reduction of the adduct
radical by A-• (See Scheme 7); (3) back electron transfer
(eq 3) must be controlled. Choosing the right donor-
acceptor couple according to the olefin used as a trap (R,â-
unsaturated ketones,18 esters and nitriles16,19) determines
the outcome. In the present case the best results were
obtained by using the DCB/Phen pair.
The mechanism of the reaction is depicted in Scheme
7. When Phen is excited, it transfers an electron to the
acceptor, and the radical cation thus formed oxidizes the
stannane, finally yielding an alkyl radical after fragmen-
tation.20 The use of a secondary donor (Phen in this case)
favors radical ion diffusion and thus offers an enhanced
chance of radical cation fragmentation (eq 2) vs back
electron transfer (eq 3); in turn, the radical adds to the
captodative olefin. The process is terminated by reduction
of the radical, either through hydrogen abstraction (from
the solvent or the stannane; path a) or by electron
transfer from the radical anion and proton addition16
(path b). When the reaction was carried out in the
(16) Fagnoni, M.; Mella, M.; Albini, A. J . Org. Chem. 1998, 63, 4026.
(17) Mella, M.; Fagnoni, M.; Freccero, M.; Fasani, E.; Albini, A.
Chem. Soc. Rev. 1998, 27, 81.
(18) Fagnoni, M.; Mella, M.; Albini, A. J . Phys. Org. Chem. 1997,
10, 777.
Con clu sion s
Captodative olefins are known as radicophiles, due to
the stabilization of the adduct radical. However, this very
fact may undermine the success of an alkylation reaction
when this depends, as it is the case in a chain process,
on the rate of hydrogen abstraction by this radical (see
(19) Fagnoni, M.; Mella, M.; Albini, A. J . Am. Chem. Soc. 1995, 117,
7877.
(20) (a) ∆GET (S1) for electron transfer from phenanthrene (EOX
)
1.50 V vs SCE)20b to DCB is negative (-7.6 kcal/mol).20c ∆GET (S1)for
the electron-transfer process between DCB and Bu4Sn is -20 kcal/
mol as calculated by means of the Weller equation (EOX Bu4Sn ) 1.75
V vs SCE, converted from the IP value20d by means of the Miller
equation).20e KSV measured from fluorescence quenching is 80 M-1
.
19
Under the present conditions, however, the absorbing species is
phenanthrene and oxidation of the stannane 5 is accomplished by
Phe+•, a slightly endothermic process. Cosentisization by this method
has been previously demonstrated to be particularly effective.16,20c,f (b)
Pysh, E. S.; Yang, N. C. J . Am. Chem. Soc. 1963, 85, 2124. (c) Borg, R.
M.; Arnold, D. R.; Cameron, T. S. Can. J . Chem. 1984, 62, 1785. (d)
Wong, C. L.; Kochi, J . K. J . Am. Chem. Soc. 1979, 101, 5593 (e) Miller,
L. L.; Nordblom, G. D.; Mayeda, E. A. J . Org. Chem. 1972, 37, 916. (f)
Schaap, A. P.; Lopez, L.; Gagnon, S. D. J . Am. Chem. Soc. 1983, 105,
663.
(21) Fox, J . M.; Morris, C. M.; Smyth, G. D.; Whitam, G. H. J . Chem.
Soc., Perkin Trans 1 1994, 731.
(22) Topp, M. R. Chem. Phys. Lett. 1975, 32, 144.
(23) Cuering, L.; Berger, M.; Steel, C. J . Am. Chem. Soc. 1974, 96,
953.
(24) Inbar, S.; Linschitz, H.; Cohen, S. G. J . Am. Chem. Soc. 1981,
103, 1048.
(25) (a) Mella, M.; Freccero, M.; Albini, A. Tetrahedron 1996, 52,
5549. (b) Minisci, F.; Fontana, F.; Zhao, L.; Banfi, S.; Quici, S.
Tetrahedron Lett. 1994, 35, 8033.

Captodative Olefin Functionalization
J . Org. Chem., Vol. 65, No. 2, 2000 301
7.35-7.85 (AA′BB′ system, 4H). IR, (neat) ν/cm-1 1316, 1149.
Anal. Calcd for C13H20S2O2: C, 57.32; H, 7.40. Found: C, 57.4;
H, 7.3.
Scheme 6), hence, the failure of the tin hydride method
with ketene dithioacetal S,S-dioxides. Stoichiometric
methods are effective, however, and we reported here two
photoinitiated methods for the radical alkylation of such
alkenes. These procedures involve unusual radical pre-
cursors and are based on the activation of the C-Sn bond
in tetraalkylstannanes via SET oxidation and, even more
remarkably, of the C-H bond in alkanes by H abstrac-
tion. The reaction is carried out under mild conditions
and, at least with bulky groups, occurs with a degree of
selectivity (100% with tertiary radicals) not often docu-
mented with open-chain derivatives. This contrasts with
the failure of the tributyltin hydride method and enlarges
the scope of the synthesis of substituted R-thiosulfones,
affording an alternative to direct alkylation of R-thiosul-
fones via the enolate,2 which is complementary to that
method when the required alkyl halide is not easily
available.
A biphasic mixture of 2 (1.1 g, 4.03 mmol) and Aliquat (1.63
g, 4.03 mmol) in toluene (20 mL) and 50% aqueous NaOH
solution (23 mL) was stirred at room temperature for 1 h. Then
freshly distilled ethyl bromide (600 µL, 8.07 mmol) in toluene
(1 mL) was slowly dropped during 2 h, and the mixture was
stirred overnight. The toluene phase was washed with water
and separated, while the aqueous phase was extracted with
CH2Cl2. The combined organic phase was dried and evaporated
under reduced pressure to give a syrup. Purification on a silica
gel column (cyclohexane/ethyl acetate 96:4 as eluant) gave 1.02
g of 1-(2-methylbutylthio)-1-[(4-methylphenyl)sulfonyl)]pro-
pane (3) (84% yield) as a colorless syrup (mixture of diaste-
reoisomers).
3: 1H NMR δ 0.85 (t, 3H, J ) 7 Hz), 0.95 (d, 3H, J ) 7 Hz),
1,10 (t, 3H, J ) 7 Hz), 1.1-1.6 (m, 4H), 2.15-2.35 (m, 1H),
2.45 (s, 3H), 2.4-2.8 (m, 2H), 3.65 (m, 1H), 7.35-7.85 (AA′BB′
system, 4H). IR, (neat) ν/cm-1 1313, 1143. Anal. Calcd for
C15H24S2O2: C, 59.96; H, 8.05. Found: C, 60.0; H, 8.0.
To a stirred solution of 3 (300 mg, 1.0 mmol) in CH2Cl2 (1.0
mL) under argon atmosphere, was added sulfuryl chloride (120
µL, 1.5 mmol) in CH2Cl2 (1.0 mL) dropwise at -15 °C, and
the mixture was stirred at the same temperature for 30 min.
The organic phase was then washed with water and separated,
and the aqueous layer was extracted with CH2Cl2. The
combined organic extracts were dried (anhydrous Na2SO4).
After removal of the volatile compounds under reduced pres-
sure, the residue was purified on a silica gel column (cyclo-
hexane/ethyl acetate 97:3 as eluant) to give 315 mg of 1-chloro-
1-(2-methylbutylthio)-1-[(4-methylphenyl)sulfonyl)]propane (4)
(94% yield, mixture of diastereoisomers) as a colorless oil.
4: 1H NMR δ 0.9 (t, 3H, J ) 7 Hz), 1.0 (d, 3H, J ) 7 Hz),
1,20 (t, 3H, J ) 7 Hz), 1.15-1.35 (m, 1H), 1.35-1.55 (m, 1H),
1.55-1.7 (m, 1H), 2.1-2.3 (m, 2H), 2.45 (s, 3H), 2.85-3.1 (AB
part of an ABX system, 2H), 7.35-7.85 (AA′BB′ system, 4H).
IR, (neat) ν/cm-1 1324, 1149. Anal. Calcd for C15H23ClS2O2:
C, 53.79; H, 6.92. Found: C, 53.7; H, 6.9.
Exp er im en ta l Section
Gen er a l. 1H, 13C, 13C-DEPT and 2D-Correlated NMR
spectra were recorded on a 300 MHz spectrometer in CDCl3
solutions, and chemical shifts are reported in ppm downfield
from TMS. 1-(Methylthio)-1-[(4-methylphenyl)sulfonyl)]ethene
1a ,26 (E)-1-(methylthio)-1-[(4-methylphenyl)sulfonyl)]propene
1b,2 and stannanes 5x18 and 5y18 were prepared according to
literature methods. Tetrabutylstannane, chloromethyl p-tolyl
sulfone, 2-methylbutanthiol, and 1,4-dicyanobenzene were
commercial samples and were used without purification.
Acetonitrile was refluxed over CaH2 and distilled just before
use; benzene and toluene were distilled from sodium-ben-
zophenone ketyl radical. The photochemical reactions were
performed by using nitrogen-purged solution and a multilamp
reactor fitted with six 15-W phosphor-coated lamps (maximum
of emission, 320 or 360 nm) for the irradiation. The reaction
course was followed by TLC (cyclohexane-ethyl acetate) and
GC. Workup of the photolytes involved concentration in vacuo
and chromatographic separation using Merck 60 silica gel.
Calculations reported in this paper were performed with the
Gaussian 94 suite program.27 Computational study was carried
out using semiempirical PM3 method.28
In a round-bottom flask fitted with a reflux condenser with
a HCl trap on top was placed 4 (425 mg, 1.26 mmol) in toluene
(40 mL). The solution was refluxed under nitrogen for 48 h
while its color changed to brown. Water was added, and the
toluene phase separated. The aqueous phase was extracted
with CH2Cl2, and the combined organic phases were dried over
Na2SO4. After solvent removal, the residue was purified on a
silica gel column (cyclohexane/ethyl acetate 96:4 as eluant)
obtaining 308 mg of 1-(2-methylbutylthio)-1-[(4-methylphenyl)-
sulfonyl)]propene (1c) (E isomer) as a colorless oil (82% yield).
1c: 1H NMR δ 0.85 (t, 3H, J ) 7 Hz), 0.90 (d, 3H, J ) 7
Hz), 1.1-1.55 (m, 3H), 2.05 (d, 3H, J ) 7 Hz), 2.45 (s, 3H),
2.55-2.75 (AB part of an ABX system, 2H), 7.35-7.85 (AA′BB′
system, 4H), 7.60 (q, 1H, J ) 7 Hz). IR, (neat) ν/cm-1 1600,
1321, 1153. Anal. Calcd for C15H22S2O2: C, 60.37; H, 7.43.
Found: C, 60.3; H, 7.4.
Syn th esis of (E)-1-(2-Meth ylbu tylth io)-1-[(4-m eth yl-
p h en yl)su lfon yl)]p r op en e (1c). This compound was pre-
pared according to Lissel’s29 and Ogura’s2 procedures as
detailed in the following. To a biphasic mixture of chloromethyl
p-tolyl sulfone (500 mg, 2.44 mmol) and methyltrioctylammo-
nium chloride (Aliquat) (99 mg, 0.24 mmol) in toluene (6 mL)
and a 17.6% aqueous NaOH solution (5 mL) was added
2-methylbutanthiol in toluene (2 mL). The resulting mixture
was stirred at room temperature for 17 h. The toluene layer
was separated, and the aqueous one was extracted with further
toluene. The combined organic phase was dried over anhydrous
Na2SO4 and evaporated. The residue was purified on a silica
gel column (cyclohexane/ethyl acetate 98:2 as eluant) to give
359 mg of 2-(methylbutylthio)-methyl p-tolyl sulfone (2) (54%
yield) as a colorless syrup.
Attem p ted Alk yla tion of 1b by Usin g Bu 3Sn H. Th er -
m a l Sta n d a r d P r oced u r e. To a stirred solution of 1b (50
mg, 0.206 mmol), AIBN (2.4 mg, 0.015 mmol), and tert-butyl
iodide (30 µL, 0.25 mmol) in dry toluene (7 mL) at reflux
temperature under argon atmosphere was slowly added tribu-
tyltin hydride (67 µL, 0.25 mmol) in dry toluene (1.2 mL) via
a dropping funnel in 3 h. Then another aliquot of AIBN was
added (2.4 mg), and further tributyltin hydride (67 µL, 0.25
mmol) in dry toluene (1.2 mL) was added in 3,h. The reflux
was continued for a further 1 h. The alkene was recovered
almost quantitatively (consumption < 5%) at the end of the
reaction.
2: 1H NMR δ 0.85 (t, 3H, J ) 7 Hz), 0.95 (d, 3H, J ) 7 Hz),
1.1-1.3 (m, 1H), 1.3-1.5 (m, 1H), 1.5-1.65 (m, 1H), 2.45 (s,
3H), 2.6-2.75 (AB part of an ABX system, 2H), 3.9 (s, 2H),
(26) Kayano, A.; Yajima, Y.; Akazome, M.; Fujita, M.; Ogura, K. Bull.
Chem. Soc. J pn. 1995, 68, 3599.
(27) Gaussian 94, Revision E.3, Frisch, M. J .; Trucks, G. W.;
Schlegel, H. B.; Gill, P. M. W.; J ohnson, B. G.; Robb, M. A.; Cheeseman,
J . R.; Keith, T.; Petersson, G. A.; Montgomery, J . A.; Raghavachari,
K.; Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.;
Cioslowski, J .; Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.;
Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J . L.;
Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.;
Defrees, D. J .; Baker, J .; Stewart, J . P.; Head-Gordon, M.; Gonzalez,
C.; Pople, J . A. Gaussian, Inc., Pittsburgh, PA, 1995.
P h otoch em ica l Sta n d a r d P r oced u r e. A solution of 145
mg (0.6 mmol, 0.12 M) of 1b, 57 µL (0.48 mmol, 0.1 M) of t-BuI,
and 27 mg (0.71 mmol, 0.14 M) of NaBH4 in dry EtOH (5 mL)
was purged under argon for 10 min, 26 µL (0.1 mmol, 0.02 M)
of Bu3SnCl were added, and the resulting mixture was
irradiated at 320 nm for 45 min. A further aliquot of Bu3SnCl
(28) Stewart, J . J . P. J . Comput. Chem. 1989, 10, 209.
(29) Lissel, M. J . Chem. Res. (S) 1982, 286.

302 J . Org. Chem., Vol. 65, No. 2, 2000
Gonza´lez-Cameno et al.
(26 µL) was added, and irradiation was continued for 45 min.
GC analysis showed only a small consumption of the substrate
(<5%).
neighboring between the ortho-hydrogens of the aromatic ring
and H-2 as well as with Me-2. In the minor one (6′by), the
coupling constant between H-1 and H-2 was 6 Hz; this value
corresponded to an average coupling between different rota-
mers. In the NOE difference experiment, the irradiation of H-1
caused enhancement only of Me-2. Thus 6by was the 1R,2S/
1S,2R isomer and 6′by the 1R,2R/1S,2S isomer.
Alk yla tion of Alk en es 1 by Usin g Sta n n a n es 5. Gen er a l
P r oced u r e. The alkene (1) (0.1 M), a stannane (5) (0.05 M),
Phen (0.05 M), and DCB (0.05 M) were dissolved in MeCN
(35 mL), and the solution was divided into two portions that
were irradiated (320 nm) in quartz tubes until the stannane
was consumed. The solvent was then removed and the residue
purified on a chromatographic column by using cyclohexanes-
ethyl acetate mixtures as eluants.
6by: 1H NMR (C6D6) δ 0.81 (d, 6H, J ) 7 Hz), 1.1 (d, 3H,
J ) 7 Hz, Me-2), 1.75 (m, 1H, H-3), 1.9 (s, 3H, SMe), 1.95 (s,
3H, CH3), 2.4 (m, 1H, H-2), 3.92 (d, 1H, J ) 1.5 Hz, H-1), 6.8
(m, 2H), 8.0 (m, 2H). IR, (neat) ν/cm-1 1311, 1143. 13C NMR
(C6D6) δ 12.6 (CH3), 17.3 (CH3), 19.4 (CH3), 20.2 (CH3), 20.8
(CH3), 32.1 (CH), 38.8 (CH), 75.7 (CH), 129.1 (CH), 129.7 (CH),
135.7, 143.8. 6′by: 1H NMR (C6D6) δ 0.85 (d, 3H, J ) 7 Hz),
0.9 (d, 3H, J ) 7 Hz), 1.3 (d, 3H, J ) 7 Hz, Me-2), 1.8 (s, 3H,
SMe), 2.0 (s, 3H, CH3), 2.25 (m, 1H, H-2), 2.5 (m, 1H, H-3),
3.6 (d, 1H, J ) 6 Hz, H-1), 6.8 (m, 2H), 8.0 (m, 2H). 13C NMR
(C6D6) δ 12.8 (CH3), 16.2 (CH3), 16.8 (CH3), 20.8 (CH3), 21.8
(CH3), 28.7 (CH), 39.4 (CH), 76.2 (CH), 128.9 (CH), 129.4 (CH),
136.9, 143.6. IR, (neat) ν/cm-1 1301, 1140.
P h otoch em ica l Rea ction betw een 1b a n d 5z. A solution
of 1b (425 mg, 0.05 M), 5z (575 µL, 0.05 M), Phen (315 mg,
0.05 M), and DCB (225 mg, 0.05 M) in MeCN (35 mL) was
irradiated for 20 h. After workup the two diastereoisomers of
1-(methylthio)-1-[(4-methylphenyl)sulfonyl)]-2-methylhexane
were isolated (about 165 mg each) (6bz and 6′bz) (59% yield,
6/6′ 50/50) both as colorless oils.
P h otoch em ica l Rea ction betw een 1a a n d 5x. A solution
of 1a (400 mg, 0.05 M), 5x (390 mg, 0.05 M), Phen (315 mg,
0.05 M), and DCB (225 mg, 0.05 M) in MeCN (35 mL) was
irradiated for 6 h. Workup gave 200 mg of 1-(methylthio)-1-
[(4-methylphenyl)sulfonyl)]-3,3-dimethylbutane (6a x) (40%) as
a white solid (mp 108-109 °C) and 49 mg of 1-(methylthio)-
1,2-(bis-p-tolylsulfonyl)ethane (7) (8%) as a white solid (mp
155-157 °C dec).
6a x: 1H NMR δ 0.95 (s, 9H), 1.4 (dd, 1H, J ) 8 Hz and 14
Hz), 2.05 (dd, 1H, J ) 2 Hz and 14 Hz), 2.25 (s, 3H), 2.45 (s,
3H), 3.6-3.7 (dd, 1H, J ) 2 and 8 Hz), 7.35-7.85 (AA′BB′
system, 4H). IR, (KBr) ν/cm-1 1300, 1140. Anal. Calcd for
C
14H22S2O2: C, 58.70; H, 7.74. Found: C, 58.6; H, 7.7.
7: 1H NMR δ 2.15 (s, 3H), 2.49 (s, 3H), 2.5 (s, 3H), 3.4 (dd,
1H, J ) 11 and 14 Hz), 3.9 (dd, 1H, J ) 2 and 14 Hz), 4.2 (dd,
1H, J ) 2 and 11 Hz), 7.37 and 7.85 (AA′BB′ system, 4H),
7.40 and 7.85 (AA′BB′ system, 4H). IR, (KBr) ν/cm-1 1304,
1139. Anal. Calcd for C17H20S3O4: C, 53.10; H, 5.24. Found:
C, 53.2; H, 5.3.
6bz: 1H NMR δ 0.85 (t, 3H, J ) 7 Hz), 0.9 (d, 3H, J ) 7
Hz), 1.2 (m, 4H), 1.45 (m, 2H), 2.05 (s, 3H), 2.4 (dq, 1H, J )
1.5 and 7 Hz), 2.5 (s, 3H), 3.7 (d, 1H, J ) 1.5 Hz), 7.35-7.85
(AA′BB′ system, 4H). 13C NMR δ 13.9 (CH3), 15.0 (CH3),18.1
(CH3), 21.5 (CH3), 22.3 (CH3), 29.0 (CH2), 32.6 (CH2), 35.2
(CH2), 77.6 (CH), 129.4 (CH), 129.5 (CH), 135.0, 144.6. IR,
(neat) ν/cm-1 1313, 1145. Anal. Calcd for C15H24S2O2: C, 59.96;
H, 8.05. Found: C, 60.0; H, 7.9.
P h otoch em ica l Rea ction betw een 1b a n d 5x. A solution
of 1b (425 mg, 0.05 M), 5x (390 mg, 0.05 M), Phen (315 mg,
0.05 M), and DCB (225 mg, 0.05 M) in MeCN (35 mL) was
irradiated for 6 h. Workup gave 274 mg of [1R,2S/1S,2R]-
1-(methylthio)-1-[(4-methylphenyl)sulfonyl)]-2,3,3-trimethylbu-
tane (6bx) (52% yield) as a colorless oil. The stereochemistry
of this diastereoisomer was attributed on the basis of the
following spectroscopic evidences. The coupling constant be-
tween H-1 and H-2 in the proton spectrum was close to zero
(J < 1 Hz), and this was consistent with a dihedral angle ca.
90°. The relative configuration was derived from NOE experi-
ments. Thus, the irradiation of H-1 caused enhancements for
H-2 and t-Bu resonances in the corresponding 1D-NOE dif-
ference spectrum. A selective irradiation of t-Bu group was
not possible owing to the small chemical shift difference
between t-Bu and Me-2 signals. Thus, the spatial arrangement
of the other substituents was derived from the NOESY
experiment in C6D6 solution (a better chemical shifts separa-
tion was found). A NOE correlation between SMe and t-Bu
groups was found as well as a correlation between Me-2 and
the ortho-aromatic protons.
6′bz: 1H NMR δ 0.85 (t, 3H, J ) 7 Hz), 1.1 (d, 3H, J ) 7
Hz), 1.2 (m, 4H), 1.45 (m, 2H), 2.05 (s, 3H), 2.4 (dq, 1H, J ) 2
and 7 Hz), 2.5 (s, 3H), 3.6 (d, 1H, J ) 2 Hz), 7.35-7.85 (AA′BB′
system, 4H). 13C NMR δ 13.8 (CH3), 18.3 (CH3), 18.6 (CH3),
21.5 (CH3), 22.5 (CH3), 29.6 (CH2), 30.4 (CH2), 33.4 (CH2), 79.4
(CH), 129.4 (CH), 129.4 (CH), 135.0, 144.6. IR, (neat) ν/cm-1
1313, 1145. Anal. Calcd for C15H24S2O2: C, 59.96; H, 8.05.
Found: C, 59.9; H, 8.1.
P h otoch em ica l Rea ction betw een 1c a n d 5x. A solution
of 1c (1.05 g, 0.1 M), 5x (390 mg, 0.05 M), Phen (315 mg, 0.05
M), and DCB (225 mg, 0.05 M) in MeCN (35 mL) was
irradiated for 18 h. Workup gave 325 mg of 1-(2-methylbu-
tylthio)-1-[(4-methylphenyl)sulfonyl)]-2,3,3-trimethylbutane
(6cx) (52%, mixture of two diastereoisomer, ratio 1:1) as a
white solid (mp 48-50 °C). Diastereoisomerism arose from the
presence of a chiral center in the sulfide moiety (1R2S2′R/
1S2R2′S and 1S2R2′R/1R2S2′S isomers). These were indis-
The same experiment was carried out in the presence of
0.1% D2O, and the 1H NMR spectrum of 6bx showed no
significant deuteration on H-1.
1
tinguishable by H NMR, while most signals were split in 13C
NMR (as indicated in parentheses).
6bx: 1H NMR δ 0.8 (d, 3H, J ) 7 Hz), 0.9 (s, 9H), 2.10 (s,
3H), 2.20 (dq, 1H, J ) 0.7 and 7 Hz), 2.45 (s, 3H), 3.85 (d, 1H,
J ) 0.7 Hz), 7.35-7.85 (AA′BB′ system, 4H). 13C NMR δ 10.7
(CH3), 17.8 (CH3), 21.5 (CH3), 27.3 (3 CH3), 33.9, 41.0 (CH),
74.7 (CH), 129.4 (CH), 128.8 (CH), 134.1, 144.6. IR, (neat)
ν/cm-1 1301, 1140. Anal. Calcd for C15H24S2O2: C, 59.96; H,
8.05. Found: C, 59.9; H, 8.0.
6cx 1H NMR δ 0.75-1 (m, 18H), 1.0-1.5 (m, 3H), 2.15-
2.25 (m, 1H), 2.35-2.65 (m, 2H), 2.45 (s, 3H), 3.9 (bs, 1H),
7.35-7.85 (AA′BB′ system, 4H). 13C NMR δ 11.1 (10.8) (CH3),
18.7 (18.9) (CH3), 21.5 (21.5) (CH3), 27.4 (27.4) (CH3), 28.7
(28.5), 34.0 (34.0) (CH2), 34.5 (34.4) (CH2), 41.0 (40.9) (CH),
41.9 (42.1) (CH) 73.1 (73.2) (CH), 129.3 (129.3) (CH), 129.9
(129.9) (CH), 134.2 (134.2), 144.5 (144.5).
P h otoch em ica l Rea ction betw een 1b a n d 5y. A solution
of 1b (425 mg, 0.05 M), 5y (360 mg, 0.05 M), Phen (315 mg,
0.05 M), and DCB (225 mg, 0.05 M) in MeCN (35 mL) was
irradiated for 6 h. Workup gave 300 mg of 1-(methylthio)-1-
[(4-methylphenyl)sulfonyl)]-2,3-dimethylbutane as a diaster-
eomeric mixture (6by + 6′by) (60% yield, 6/6′ 69/31) as a
colorless oil. In the proton spectrum, the resonances of each
diastereoisomer were separated in C6D6 solution and were
identified with COSY spectra. As in the case above, the major
one (6by) showed a small coupling constant between H-1 and
H-2 (1.5 Hz), thus a value which corresponded to a gauche
arrangement with a torsion angle close to 60°. In the NOE
difference experiment, the irradiation of H-1 caused enhance-
ments for H-2 and H-3. The NOESY spectrum revealed a close
Alk yla tion of Alk en e 1b by Usin g Alk a n es. P h oto-
ch em ica l r ea ction of 1b in Cycloh exa n e (8j) in th e
P r esen ce of Ben zop h en on e. A solution of 1b (127 mg, 0.015
M) and benzophenone (96 mg, 0.015M) in cyclohexane (35 mL)
was irradiated for 15 h (λ ) 360 nm). After the usual workup,
the residue was purified on a silica gel column (cyclohexane/
ethyl acetate 95:5 as eluant) to give 153 mg of 2-cyclohexyl-
1-(methylthio)-1-[(4-methylphenyl)sulfonyl)]propane as a di-
astereomeric mixture 9bj + 9′bj (89% yield, 9/9′ 82/18) as a
colorless oil.
9bj: 1H NMR (from the mixture) δ 0.9 (d, 3H), 0.75-1.8
(m, 11H), 2.05 (s, 3H), 2.15 (dqui, 1H, J ) 2 and 7 Hz), 2.45
(s, 3H), 3.85 (d, 1H, J ) 2 Hz), 7.35-7.85 (AA′BB′ system, 4H).
IR, (neat) ν/cm-1 1312, 1143.

Captodative Olefin Functionalization
J . Org. Chem., Vol. 65, No. 2, 2000 303
9′bj: 1H NMR (from the mixture) δ 0.9 (d, 3H), 0.75-1.8
(m, 12H), 1.95 (s, 3H), 2.45 (s, 3H), 3.65 (d, 1H, J ) 6 Hz),
7.35-7.85 (AA′BB′ system, 4H). IR, (neat) ν/cm-1 1312, 1143.
P h otoch em ica l Rea ction of 1b w ith Ad a m a n ta n e (8k )
in th e P r esen ce of Ben zop h en on e. A solution of 1b (850
mg, 0.1 M), adamantane (4.77 g, 1.0 M), and benzophenone
(638 mg, 0.1 M) in benzene (35 mL) was irradiated for 18 h
(λ ) 360 nm). After the usual workup, the residue was purified
on a silica gel column (cyclohexane/ethyl acetate 96:4 as
eluant) to give 1.18 g of 2-(1′-adamantyl)-1-(methylthio)-1-[(4-
methylphenyl)sulfonyl)]propane (90% yield) as a white solid
(mp 109-110 °C, from methanol). Traces of 1-adamantanol,
2-adamantanone, 1-phenyladamantane,30 and bisadaman-
tane31 were detected by GC and identified through their MS
spectra and by comparison with authentic samples (com-
mercial samples, Aldrich, for the first two).
9bk : 1H NMR δ 0.8 (d, 3H, J ) 7 Hz), 1.4-1.9 (m, 12H),
1.95 (bs, 3H), 2.0 (q, 1H, J ) 7 Hz), 2.10 (s, 3H), 2.45 (s, 3H),
3.90 (bs, 1H), 7.35-7.85 (AA′BB′ system, 4H). 13C NMR δ 9.2
(CH3), 17.5 (CH3), 21.5 (CH3), 28.2 (CH), 35.4, 36.7 (CH2), 39.3
(CH2), 41.6 (CH), 73.3 (CH), 129.2 (CH), 129.7 (CH), 134.0,
144.5. IR, (KBr) ν/cm-1 1309, 1143. Anal. Calcd for C21H30S2-
O2: C, 66.62; H, 7.99. Found: C, 66.7; H, 7.9.
Ackn owledgm en t. A.M.G.C. thanks EU (ERB CHRX
CT93 0151) for a fellowship. We are greatly indebted
to Dr. Mauro Freccero for the semiempirical PM3
calculation. Partial support of this work by Consiglio
Nazionale delle Ricerche, Rome and MURST, Rome, is
gratefully acknowledged.
(30) Hachiya, I.; Moriwaki, M.; Kobayashi, S. Bull. Chem. Soc. J pn.
1995, 68, 2053.
(31) Son, V. V.; Son, T. V. J . Org. Chem. USSR 1991, 27, 2081.
J O990817E