Nucleophilic dechlorocarbomethoxylation of (E)-methyl β-chloro-α-methyl-β-(3-bromo-2,4,6-trimethylphenyl)acrylate

Full text of DOI:10.1021/jo9600565

J . Org. Chem. 1996, 61, 3553-3556
3553
(a ) MeS-. (i) At a 1:1 [Nu -]:[1] ratio (ca. 0.1 mol/L)
after 7 h reflux in MeCN, 12% conversion to 2 was ob-
served, and the rest of the compound was unreacted 1.
Nu cleop h ilic Dech lor oca r bom eth oxyla tion
of (E)-Meth yl â-Ch lor o-r-m eth yl-â-(3-br om o-
2,4,6-tr im eth ylp h en yl)a cr yla te
(ii) At a 5:1 [Nu -]:[1] ratio, all the ester had reacted
after 6 h reflux in MeCN, giving ca. 90% of 2 and ca. 10%
of a compound that by GC/MS displayed isotopomeric
molecular peaks at m/z 388, 390, 392 with intensity
distribution consistent with the presence of one chlorine
and one bromine atom. The compound was not identi-
fied, but by its molecular weight it cannot be an adduct
of MeSH to the double bond.
Michal Beit Yannai and Zvi Rappoport*
Department of Organic Chemistry, The Hebrew University,
J erusalem 91904, Israel
Received J anuary 8, 1996
Vinylic substitution of alkyl â-halocinnamates by nu-
cleophiles was extensively studied.1 When the R-position
is substituted, e.g., with an R-carboalkoxy or an R-cyano
group, the reaction usually proceeds by the addition-
elimination route1f,g (eq 1), whereas with an R-hydrogen
The acetylene was isolated from a large scale experi-
ment.
(b) p-TolS-. (i) With a 1.5:1 [Nu -]:[1] ratio, the
reaction is slow, and after 44 h reflux in MeCN only 4%
conversion of 1 to 2 was observed. Di-p-tolyl disulfide
[1H NMR δ (CDCl3) 2.31 (6H, s, Me), 7.25 (8H, AA′BB′q,
J ) 8 Hz, Ar-H); mass spectrum m/z (relative abundance,
assignment) 246 (93, M), 123 (B, TolS+)] is formed in a
4-fold excess over 2.
ArC(X)dC(R)CO2Me + Nu-
X ) Cl, Br; Nu- ) SR′, OR′, F-
R ) CN, CO2Me
f
ArC(Nu)dC(R)CO2Me (1)
(ii) With a 10:1 [Nu -]:[1] ratio, 48 h reflux in MeCN
led to 88% conversion. The main product was 2, but some
1 still remained and di-p-tolyl disulfide was also formed.
GC/MS showed the formation of three additional com-
pounds (I-III) in very low percentages.
I. Formed in <3% and assigned by its mass spectrum
base peak and some fragments as methyl S-tolyl thiol-
carbonate 3 [m/z (relative abundance, assignment) 182
(64, M), 152 (4, M - 2Me), 105 (B, PhCO), 77 (50, Ph)].
The most abundant fragment at m/z 105 is most likely
the benzoyl cation PhCO+, whose formation requires a
substantial rearrangement and cleavage. However, an
extensive rearrangement is required if the benzoyl frag-
ment arises from any other precursor.
ArC(X)dCHCO2Me + Nu- f ArCtCCO2Me (2)
X ) Cl; Nu- ) R′O-
derivative substitution via addition-elimination1d or
elimination to the acetylene1b,e may take place (eq 2). We
have found that when the aryl group is bulky, a different
reaction course, i.e., nucleophilic dechlorocarbomethoxy-
lation with formation of an acetylene, takes place.
Resu lts
(E)-Methyl R-methyl-â-chloro-â-(3-bromo-2,4,6-trimeth-
ylphenyl)acrylate (1) was prepared by a configuration-
retaining esterification either with diazomethane or with
thionyl chloride and then methanol of the (E)-acid, which
was prepared in turn according to Adams and Miller.2
II. A compound whose main fragment is a pair of base
peaks with identical intensities at m/z 225 and 227.
Hence, this is bromine-containing, and it is consistent
with 3-Br-2,4,6-Me3C6HCO+(ArCO+). However, a parent
peak is apparently missing. Other fragments are at m/z
199, 197 (26, 26, ArCO-2Me), and 117 (27, ArCO+H -
Br - 2Me). Again, formation of this fragment requires
a structural rearrangement.
1
The ester obtained by both methods was identical by H
NMR and GC/MS.
In an attempted substitution of the vinylic chlorine,
the ester was reacted with several oxygen and sulfur
nucleophiles. Whereas on reflux of 1 in MeCN for 15.5
h in the absence of a nucleophile 1 was recovered
unchanged, with all the nucleophiles a dechlorocar-
bomethoxylation rather than a substitution process took
place with formation of (3-bromo-2,4,6-trimethylphenyl)-
methylacetylene (2) (eq 3).
III. A bromine-containing compound formed in ca.
4%, with the highest m/z values at 360, 362 (M, 12, 12)
whose cleavage pattern [m/z 233, 237 (31, 31, M - TolS),
158 (B, M - TolS - Br), 143 (44, M - TolS - Br - Me),
128 (26, M - TolS - Br - 2Me) is consistent with
structure 4.
(c) Nu - ) P h O-. With a 5-fold excess of nucleophile
after 12 h reflux in MeCN 20% of 2 (identified by 1H NMR
and GC/MS) and 80% of 1 were observed.
(d ) Nu - ) i-P r O-. With a 5-fold excess of nucleophile,
reflux for 70 h in MeCN gave 97% conversion to two
products: 57% of the trans-esterification product 5 and
40% of the acetylene 2. The structure of 5 was cor-
roborated by the identity of its spectra with those of an
authentic sample prepared from the acyl chloride and
2-propanol. Due to the atropisomerism of 5 resulting
from restricted rotation around the Ar-C bond, two
isopropyl-Me doublets were observed for this chiral
molecule.
Two thio nucleophiles and three oxygen nucleophiles
have been studied:
(1) (a) Rappoport, Z.; Topol, A. J . Chem. Soc., Perkin Trans. 2, 1972,
1823. (b) Youssef, A.-H. A.; Abdel-Maksoud, H. M. J . Org. Chem. 1975,
40, 3227. (c) Youssef, A.-H. A.; Abdel-Reheim, A. G. Indian J . Chem.,
Sect. B 1976, 14B, 101. (d) Youssef, A.-H. A.; Sharaf, S. M.; El-Sadany,
S. K.; Hamed, A. E. J . Org. Chem. 1981, 46, 3813. (e) Indian J . Chem.,
Sect. B 1982, 21B, 359. (f) Avramovitch, B.; Rappoport, Z. J . Am. Chem.
Soc. 1988, 110, 911; J . Org. Chem. 1982, 47, 1397. (g) Rappoport, Z.;
Gazit, A. J . Org. Chem. 1986, 51, 4112; J . Am. Chem. Soc. 1987, 109,
6698.
(2) Adams, R.; Miller, M. W. J . Am. Chem. Soc. 1940, 62, 53.
S0022-3263(96)00056-4 CCC: $12.00 © 1996 American Chemical Society

3554 J . Org. Chem., Vol. 61, No. 10, 1996
Notes
Cl2CdC-(CO2Me) ClCtCCl
ArC(STol-p)dCHMe
p-TolSCOOMe
3
4
10
9
ArC(Cl)dC(Me)COOPr-i
XCtCCO2Me
11a : X ) CO2Me
b: X ) CN
XC(Cl)dC(CO2Me)2
12a : X ) CO2Me
b: X ) CN
5
Ar ) 3-Br-2,4,6-Me3C6H
(e) Nu - ) t-Bu O-. With a 4.4-fold excess of the nu-
cleophile, after 18 h reflux in MeCN all precursor 1 had
disappeared, but the product was not identified. The GC/
MS displayed only one compound with the highest signal
at m/z 123, suggesting the presence of a nitrogen, and
signals at m/z 96 and 83, suggesting the loss of HCN and
CH from it. NMR showed several signals, including two
Me signals at 2.25 and 2.41 ppm and a H singlet at 6.07
ppm. We speculate that cleavage of 1 took part and one
of the fragments reacted with the solvent.
Several mechanistic points and questions arise in
relation to the observed dechlorocarbomethoxylation. In
the catalysis by a thio-nucleophile it is possible that the
reaction is initiated by a low conversion to the vinylic
substitution product and Cl-, and the latter serves as the
reagent in an autocatalytic dechlorocarbomethoxylation
process which is faster than further substitution by the
RS-. We therefore conducted a control experiment in
which benzyltrimethylammonium chloride reacted for 7
h in refluxing MeCN with 1. Since no reaction was
observed, this route is excluded.
A second question is what is the detailed pathway for
the loss of the ester group? J ones and co-workers10
pyrolytically decomposed methyl â-halovinyl carboxy-
lates, 13 and 14, which are structurally related to 5, in
the presence or the absence of halide ions (eq 4). Al-
1
Discu ssion
Decarbalkoxylation of esters, especially malonates and
â-keto carboxylates, in water containing aprotic solvents
such as DMSO-H2O was studied extensively by Krapcho
and co-workers3 and by others. The reactions were
accelerated by added metal (M) salts, and halides salts
and cyanides (MCl, MBr, MI, MCN) were the catalysts
used most often. In a mechanistic study of the reaction
Krapcho et al.4 investigated the catalytic efficiency of
other salts including NaOAc, Na3PO4, Na2SO4, and KF.
Decarbalkoxylations with NaOEt5a and Me4NOAc5b or
with amines6a,b or a thiolate ion6c,d are also known.
Solvent isotope effects suggested that the “uncatalyzed”
reaction, i.e., with water as the nucleophile, proceeds by
water attack on the ester carbonyl (a BAc2 mechanism)
whereas the nucleophilic anions react either by this BAc2
mechanism forming a tetrahedral intermediate4 or by
attack on the alkyl group (a BAL2 route).4,6b Evidence for
the latter route is the isolation of RCN in the KCN-
catalyzed reaction of CO2R esters and the trapping of the
carboxylate anion formed from a malonate ester when
the reaction is conducted with LiCl/HMPA.7 However,
a concerted loss of RNu and CO2 initiated by attack on
the alkyl group is an alternative route in other systems.4
It was suggested that in the general case the BAc2 and
the BAL2 routes are competitive.
RC(X)dCHCO2Me
8 MeX + CO2 (4)
13: R ) CO2Me; X ) Cl, Br,
14: R ) Ph; X ) Br
though the main organic product was usually not
monitored, 14 gave in PhNO2 without additives both
phenylacetylene and R-bromostyrene, whereas with
NaBr/crown-6 in MeCN the accelerated reaction gave
only phenylacetylene. Hence, at least part of the reaction
involves the loss of only the ester group. A similar
conclusion is derived from the work of Ykman and
Hall.9
The two feasible BAL-type routes suggested for the
reaction are demonstrated below for 1. (a) The nucleo-
phile attacks the ester methyl group, generating the
carboxylate ion, which by loss of CO2 generates the vinyl
anion which then expels the â-halogen or the loss of the
CO2 and X- may be concerted, i.e., a Grob-type fragmen-
tation. Indeed, Grob and co-workers obtained acetylenes
by heating â-bromocarboxylates11 (eq 5).
The number of decarbalkoxylations of vinylic esters is
smaller than in saturated systems,8 and R,â-dechloro-
carbalkoxylation is rare. The first work is that of Ykman
and Hall,9 who were able to substitute the chlorine of
ester 6 with KF/crown ether, whereas the analogous
ClCHdC(CO2Me)2 Cl2CdC(CO2Me)2
6
7
Cl2CdC(Cl)CO2Me
(5) (a) Thole, F. B.; Thorpe, J . F. J . Chem. Soc. 1911, 99, 2183, 2187.
Ingold, C. K.; Thorpe, J . F. Ibid. 1919, 115, 143. Cope, A. C.; McElvain,
S. M. J . Am. Chem. Soc. 1932, 54, 4311, 4319. (b) J ohnson, W. S.;
Harbert, C. A.; Ratcliffe, B. A.; Stipanovich, R. D. J . Am. Chem. Soc.
1976, 98, 6188. Trost, B. M.; Verhoeven, T. R. J . Am. Chem. Soc. 1980,
102, 4743.
(6) (a) Huang, B. S.; Parish, E. J .; Miles, D. H. J . Org. Chem. 1974,
39, 2647. Miles, D. H.; Huang, B. S. Ibid. 1976, 41, 208. Parish, E. J .;
Huang, B. S.; Miles, D. H. Synth. Commun. 1975, 5, 341. (b) Saunier,
Y. M.; Danion-Bougot, R.; Danion, D.; Carrie´, R. Bull. Soc. Chim. Fr.
1976, 1963. (c) Keinan, E.; Eren, D. J . Org. Chem. 1986, 51, 3165. (d)
Bartlett, P. A.; J ohnson, W. S. Tetrahedron Lett. 1979, 4459.
(7) Asaoka, M.; Miyake, K.; Takei, H. Chem. Lett. 1975, 1149.
(8) Venkateswaran, R. V.; Ghosh, V.; Sarkar, A. Tetrahedron 1979,
35, 553.
8
reaction of ester 7 gave 8 presumably via carbanion 9. If
9 is formed from 7, then the dechlorocarbomethoxylation
is not concerted. The two options of nucleophilic attack
at the Me or the CdO were not distinguished. Formation
of 10 via anion analogous to 9 is an overall dechlorocar-
bomethoxylation, and the same applies to formation of
11a , b from 12a , b with KCl/crown in sulfolane.
(3) For a review see: (a) Krapcho, A. P. Synthesis 1982, 805; (b)
893.
(4) Krapcho, A. P.; Weimaster, J . F.; Eldridge J . M.; J ahngen, E. G.
E., J r.; Lovey, A. J .; Stephens, W. B. J . Org. Chem. 1978, 43, 138.
(9) Ykman, P.; Hall, H. K., J r. Tetrahedron Lett. 1975, 2429. See
also: Hall, H. K., J r.; Ykman, P. Macromolecules 1977, 10, 464.
(10) J ones, G., II; Fantina, M. E.; Pachtman, A. H. J . Org. Chem.
1976, 41, 329.

Notes
A priori, this pathway seems preferable over a (b) BAL
route where nucleophilic attack on the ester carbonyl
results in expulsion of the CO2Me group either concert-
edly with the halide ion (e.g., eq 6) or stepwise with initial
formation of the vinyl anion. Our work supplies some
J . Org. Chem., Vol. 61, No. 10, 1996 3555
(E)-Meth yl r-Meth yl-â-ch lor o-3-br om o-â-(2,4,6-tr im eth -
ylph en yl)acr ylate (1). (a) (E)-r-Meth yl-â-ch lor o-â-(3-br om o-
2,4,6-tr im eth ylp h en yl)a cr ylic Acid . The acid was prepared
according to Adams and Miller2 in a synthesis which involves a
Friedel-Crafts acylation of bromomesitylene with propionic
anhydride reported to give bromomesityl ethyl ketone. However,
we found that mesitylene and dibromomesitylene are also formed
in this reaction, presumably by disproportionation of bro-
momesitylene under the Friedel-Crafts conditions.15 The mix-
ture formed was chromatographed on silica with 1:1 CH2Cl2:
petroleum ether eluent. The first fraction eluted was the known
dibromomesitylene, mp 64 °C (lit.16 mp 64 °C). Anal. Calcd for
C9H10Br2: C, 38.88; H, 3.63. Found: C, 38.52, H, 3.68). The
bromomesityl ethyl ketone required for the large scale synthesis
was separated from the other compounds by distillation. The
fraction boiling at 103 °C at 0.3 mm was nearly pure bro-
momesityl ethyl ketone, which was used for continuation of the
synthesis of the acid.
(b) Ester 1. (i) To a solution of (E)-R-methyl-â-chloro-â-(3-
bromo-2,4,6-trimethylphenyl)acrylic acid (1 g, 3.1 mmol) in ether
(24 mL) was added a solution of diazomethane in ether (ca. 0.3
g CH2N2, prepared from Diazald) at room temperature. After
15 min, AcOH (20 mL) was added to neutralize the unreacted
diazomethane, the solution was washed with 10% aqueous
NaHCO3 solution, the phases were separated, and the organic
phase was washed with water and dried (CaCl2). After evapora-
tion of the ether, the methyl ester was obtained as an oil (400
mg, 39%).
information concerning this question. First, the reaction
with i-PrO- shows clearly that anion exchange, i.e., OMe
f OPr-i, of the ester takes place, either before or
concurrently with acetylene formation. Second, and
unfortunate since the compound formed only in a very
low yield and was assigned only on its base peak in the
mass spectrum, the fragment formed by attack on the
carbonyl group with loss of the whole ester group is
observed in the reaction with TolS-. The thiolcarbonate
ester 3 could be formed only by attack of TolS- on the
ester carbonyl, as in eq 6, followed by cleavage of the
C-Cd bond. In contrast, TolSMe, which was not found,
is the expected product of eq 5.
Third, the question arises why, in spite of the many
reactions of the exclusive vinylic substitution of â-ha-
lovinylic esters,1 the dechlorocarbomethoxylation reaction
with thionucleophiles and the oxygen (i-PrO-) nucleo-
philes was not observed so far. We believe that the
reason for a reaction at the ester group rather than at
the vinylic carbon of the electrophilic vinyl halide is that
the approach to the latter is hindered, especially by the
bulky substituted mesityl ring. Apparently, the competi-
tive attack on the less hindered carbonyl is faster. This
is corroborated by our attempted substitution of 15 with
TolS- and TolO- anions which gave reaction at the
tosylate rather than at the vinylic carbon.12
(ii) A solution of the acid (5.56 g, 17.5 mmol) in dry toluene
(35 mL) was heated to reflux, and afterwards thionyl chloride
(4.5 mL, 62 mmol) was added dropwise. The mixture was
refluxed for 2 h, and the IR spectrum showed that the acid
absorption at 1688 cm-1 was replaced by that of the acyl halide
at 1774 cm-1. The solvent and excess SOCl2 were evaporated,
toluene (20 mL) was added, and the liquid was further evapo-
rated, leaving a yellow oil. Methanol (25 mL) and pyridine (0.25
mL) were added, and the mixture was refluxed for 3.5 h. IR
spectra showed that 1774 cm-1 absorption was replaced by that
of the ester at 1712 cm-1. The methanol was evaporated, leaving
5.15 g of the crude ester. Chromatography on silica with 80:20
petroleum ether:CH2Cl2 as eluent gave 3.5 g (60%) of the pure
ester as an oil, which did not form crystals even after standing
for a few months. 1H NMR δ (CDCl3): 2.16, 2.23, 2.27, 2.33 (4
× 3H, 4s, Me), 3.49 (3H, s, OMe), 6.94 (1H, s, Ar-H). Mass
spectrum m/z (relative abundance, assignment): 334, 332, 330
(25, 100, 76, M [37Cl81Br, 37Cl79Br/35Cl81Br, 35Cl 79Br]), 302, 300,
298 [21, 75, 55, M - 2Me - 2H) 297, 295 (8, 8, M - Cl), 267,
265 (3, 9, M - Cl - 2Me), 237, 235 (59, 57, M - Cl - 4Me), 193,
191 (24, 67, M - Br - 4Me), 157 (27, ArC+dCMe - Br), 141
(32, ArCtC+ - Br), 115 (11, Me2C6H3C). Anal. Calcd for
MesC(OTs)dC(CO2Me)2
15
Finally, although the formation of 4 can be ascribed
to a substitution-decarbomethoxylation sequence (i), an
addition of TolS- to acetylene 2 (ii) which comprises the
“addition” part in the nucleophilic elimination-addition
process is also possible. However, reaction of 2 with
p-TolS- with or without sonication under the conditions
that gave 4 showed no traces of 4, suggesting that it is
formed by route i.
C
14H16O2BrCl: C, 50.43; H, 4.99; Br, 23.05; Cl, 11.09. Found:
C, 50.70; H, 4.86; Br, 24.09; Cl, 10.69.
(E) Isop r op yl r-Met h yl-â-ch lor o-â-(3-b r om o-2,4,6-t r i-
m eth ylp h en yl)a cr yla te (5). To (E)-R-methyl-â-chloro-â-(3-
bromo-2,4,6-trimethylphenylacryloyl chloride (1 g, 3 mmol)
prepared as described in method ii above were added isopropyl
alcohol (20 mL) and pyridine (0.25 mL). The mixture was
refluxed for 18 h, after which the only CdO absorption was at
1726 cm-1. The solvent was evaporated, and chromatography
of the oil obtained on silica using 4:1 petroleum ether:CH2Cl2
eluent gave pure (E)-isopropyl R-methyl-â-chloro-â(3-bromo-
2,4,6-trimethylphenyl)acrylate, 5, as an oil (800 mg, 74%). 1H
NMR δ (CDCl3): 0.78 (3H, d, J ) 6.2 Hz, i-PrMe), 0.86 (3H, d,
J ) 6.2 Hz, i-PrMe), 2.16, 2.21, 2.34, 2.38 (4 × 3H, 4s, Me), 4.71-
4.81 (1H, hep, i-PrH), 6.93 (1H, s, ArH). Mass spectrum m/z
(relative abundance, assignment): 362, 360, 358 (3, 14, 11, M
Exp er im en ta l Section
The salts p-TolSNA (Merck) and MeSNa, t-BuOK (Aldrich)
were commercial samples. PhONa was prepared according to
Elkobaisi et al.13 and i-PrONa according to Seubold.14 The
MeCN was HPLC grade. All the reactions were conducted under
argon with oxygen filter. The reactions were conducted with
200 mg of 1 and the appropriate ratio of nucleophile given above
in 8-15 mL of MeCN. NMR spectra were recorded with a
Bruker AMX-400 pulsed FT spectrometer operating at 400.13
MHz for 1H and at 100.62 MHz for 13C, and IR spectra were
recorded with a Nicolet Impact 400 spectrometer.
[
37Cl81Br, 37Cl79Br/35Cl81Br, 35Cl79Br]), 302, 300, 298 (6, 17, 14,
M - 4Me), 202 (18, M - Cl - Br - i-Pr), 194, 192 (9, 24, M -
Br - CO2Pr-i), 158 (51, M - Br - Cl - CO2Pr-i), 141 (45,
MesCdCH), 115 (23), 43 (B, i-Pr+). Anal. Calcd for C16H20
-
BrClO: C, 53.58; H, 5.34; Br, 22.28; Cl, 9.88. Found: C, 53.84,
H, 5.60; Cl, 21.71; Br, 9.46.
(11) Grob, C. A.; Csapilla, J .; Cseh, G. Helv. Chim. Acta 1964, 47,
1590.
(12) Gazit, A.; Rappoport, Z. Unpublished results.
(13) Elkobaisi, F. M.; Hickinbottom, W. J . J . Chem. Soc. 1958, 2431.
(14) Seubold, F. H., J r. J . Org. Chem. 1956, 21, 156.
(15) Adams reported that a similar reaction of acetic anhydride with
3-bromomesitylene gave a poor yield due to “an excessive amount of
rearrangement to tribromomesitylene” (Adams, R.; Theobald, C. W.
J . Am. Chem. Soc. 1943, 65, 2383).
(16) Sussenguth, H. J ustus Liebigs Ann. Chem. 1882, 215, 248.

3556 J . Org. Chem., Vol. 61, No. 10, 1996
Notes
(3-Br om o-2,4,6-tr im eth ylp h en yl)m eth yla cetylen e (2). A
solution of ester 1 (300 mg, 0.9 mmol) and MeSNa (300 mg, 4.3
mmol) in MeCN (15 mL) was refluxed for 9 h under argon. The
solvent was evaporated, the remainder was dissolved in CH2-
Cl2, the solution was filtered from the salt residues and then
evaporated, and the residue was chromatographed over silica
with petroleum ether (40-60 °C) as eluent. The first fraction
collected was acetylene 2, mp 50 °C (60 mg, 28%). 1H NMR δ
(CDCl3): 2.15, 2.33, 2.36, 2.56 (4 × 3H, 4s, Me), 6.92 (1H, s,
Ar-H). 13C NMR (CDCl3) δ: 4.8 (Me), 77.4 (one CtC carbon;
the other overlaps the solvent signal), 122.8 (C3), 124.4 (C1),
128.8 (C5), 138.7 (C4) 139.6 (C-2,6), (assignment based on
additivity scheme). Mass spectrum m/z (relative abundance,
assignment): 238, 236 (93, 97, M [81Br, 79Br]), 223, 221 (4, 4, M
- Me), 157 (47, M - Br), 142 (B, M - Br - Me), 115 (26, M -
Br - 2Me - C?). Anal. Calcd for C12H13Br: C, 60.89; H, 5.77;
Br, 33.40. Found: C, 60.78; H, 5.52; Br, 33.40.
Attem p ted Rea ction of 2 w ith Sod iu m p-Tolu en eth io-
la te. A solution of acetylene 2 (20 mg, 0.084 mmol) and sodium
p-toluenethiolate (60 mg, 0.41 mmol) in MeCN (8 mL) was
refluxed with stirring under argon for 17 h. Only 2 and traces
of di-p-tolyl disulfide were observed. The mixture was sonicated
for 4 days at room temperature. TLC analysis during the
reaction and NMR and GC/MS of the solid obtained after
filtering the mixture and evaporating the solvent showed only
2 and the disulfide and no traces of compound 4.
Ack n ow led gm en t. This work was supported by the
U.S.A.-Israel Binational Science Foundation (BSF),
J erusalem, Israel, to which we are indebted.
J O9600565