Adducts of ArSCl with vinyl ethers or esters as synthones in geminal alkylation

Article abstract of DOI:10.1070/MC2003v013n01ABEH001710

The in situ-prepared adducts of arylsulfenyl chloride and vinyl ethers (esters) were employed as synthetic equivalents of 1,1-bis-electrophiles in the Lewis acid-promoted reaction sequence with two different carbon nucleophiles resulting in the formation of geminal bisalkylation products.

Full text of DOI:10.1070/MC2003v013n01ABEH001710

, 2003, 13(1), 21–23
Adducts of ArSCl with vinyl ethers or esters as synthones in geminal alkylation
William A. Smit,*a Alexei V. Gromova,b and Elisey A. Yagodkina,b
a N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 119991 Moscow, Russian Federation.
Fax: +7 095 135 5328; e-mail:smt@ioc.ac.ru
b Higher Chemical College, Russian Academy of Sciences, 125047 Moscow, Russian Federation
10.1070/MC2003v013n01ABEH001710
The in situ-prepared adducts of arylsulfenyl chloride and vinyl ethers (esters) were employed as synthetic equivalents of 1,1-bis-
electrophiles in the Lewis acid-promoted reaction sequence with two different carbon nucleophiles resulting in the formation of
geminal bisalkylation products.
The synthetic merits of controlled geminal alkylation, which
might secure an opportunity to introduce two different groups
SiMe3
OMe
OMe
Me
OMe
Cl
Me
at the same carbon atom, are obvious. Most typically, this result
is achieved using bifunctional reagents such as malonic ester or
1,3-dithiane, which are employed as synthetic equivalents of
1,1-bisnucleophiles in sequential transformations which involve
the generation of a carbanion and reactions with carbon electro-
philes (EC).1 At the same time, an opportunity to employ for this
purpose an alternative (in a way, an inverse polarity) approach
using synthetic equivalents of 1,1-biselectrophiles in sequential
reactions with two different carbon nucleophiles (NuC) seems
promising. To the best of our knowledge, no systematic attempts
have been undertaken to explore the viability of this option.
Our interest in this problem is based on the reactivity of easily
preparable adducts of vinyl ethers with arylsulfenyl chloride.
Thus, we found that under the action of Lewis acids (L.a.) these
adducts are converted into the episulfonium ion (ESI-I)-like
electrophiles capable of reacting with π-donors such as vinyl silyl
ethers, silyl ketene acetals and allylsilanes (stannanes) (NuC-I)
to give the respective products of ternary coupling in prepara-
tive yields (mono-adduct, Scheme 1).2 It was tempting to specu-
late that, due to the presence of a β-methoxyalkylaryl sulfide
moiety in the adducts thus prepared, the latter can be used as
the precursors of electrophiles (ESI-II). Hence, the second alkyla-
tion at the same carbon is considered as a feasible reaction.†
This pathway, which leads eventually to the 1,1-bisadduct, is
shown in Scheme 1. Retrosynthetically, it corresponds to the
sequential coupling of a β-arylthio biscationic synthone with two
nucleophilic synthones.
ArSCl
4a
ArS
ArS
ArS
ArS
TiCl4,
– 70 ° C, 10 min
Me
CH2Cl2,
– 70 °C,
15 min
, 85%
5
6
1
2
SnBu3
OMe
OMe
Ph
OMe
ArSCl
4b
Cl
Ph
Ph
CH2Cl2,
– 50 °C,
2 min
AgOTf,
– 40 °C, 1h
, 85%
Me
SnBu3
OAc
Cl
OAc Me
, 85%
OAc
ArS
ArSCl
ArS
7
CH2Cl2,
room
temperature,
AgOTf,
8
3
10 °C, 40 min.
5 min
L. a.
Me
O
?
O
ArS
Ar = p-Tol
ArS
O
(– MeCO+)
+
Scheme 2
styrene 2 and vinyl acetate 3 were chosen as model compounds.‡
The reaction of 1 with allyltrimethylsilane 4a in the presence of
TiCl4 (or another Lewis acid) proceeded uneventfully to give
adduct 5 in a good yield (Scheme 2).3 Unexpectedly, the reaction
of adduct 2 with 4a proceeded very slowly in the presence of
Lewis acids (TiCl4, EtAlCl2, TMSOTf and AgSbF6).§ Required
product 6 was prepared by the reaction of 2 with more reactive
tributylallylstannane 4b in the presence of AgSbF6 or AgOTf.
The possibility to generate episulfonium ion-like electrophilic
species from adduct 3 seemed questionable because of the de-
activating effect of the acetoxy group. Moreover, it is believed
that in this case the ESI-like intermediate once formed would
immediately undergo fragmentation, as shown in Scheme 2.
Much to our reward, we found that the generation of the cationoid
intermediate from 3 and its reaction with tributylmethallyl-
stannane 7 could be carried out under the action of AgOTf to
form required adduct 8 in a nearly quantitative yield.¶
The adducts of p-TolSCl with 2-methoxypropene 1, α-methoxy-
OR
OR
L.a.
Cl
+
ArSCl
ArS
(– Cl –
)
R = alkyl
OR
L.a.
MR3
OR
(– OR – )
X
X
+
SAr
ArS
(NuC-I)
X = O, CH2
for X = CH2
mono-adduct
ESI-I
†
It seems obvious that this assumption could be considered as plausible
X
only for the adducts formed with the use of allylsilanes (stannanes) as
NuC (X = CH2, see Scheme 1), since the carbonyl-containing adducts
(X = O) most certainly would undergo methoxy group elimination under
the action of a Lewis acid, leading to the formation of α,β-unsaturated
carbonyl derivatives.
Adducts 1– 3 were prepared using the equimolar quantities of the reac-
tants in a CH2Cl2 solution under the conditions specified in Scheme 2. The
identity of 1– 3 was ascertained by H NMR data. The in situ-prepared
solutions of these adducts were used in further reactions.
TLC-monitored experiments ascertained the easiness of the formation
of an ESI intermediate upon the treatment of 2 with a Lewis acid. Thus,
it seems likely that the sluggishness of the overall transformation 2 ® 6
could be explained by the ‘hyperstabilization effect’ of the intermediately
formed electrophile due to the presence of the phenyl group at the reacting
carbocationic centre.
X
ArS
+
R3M
ArS
(NuC-II)
‡
X = O, CH2
ESI-II
ArS
1,1-bis-adduct
1
X
_
_
§
++
+
+
Nuc-I
Scheme 1
Nuc-II
– 21 –

, 2003, 13(1), 21–23
OTMS
OMe
9
O
H
ArS
ArS
Me
, 58%
10
Me
OTMS
OMe
Me
Me
OMe
ArS
Me
ii, NuC-II =
11
i, TMSOTf/Et2AlCl
(– MeO– )
O
ArS
+
0 ° C → 20 °C, 1 h
Me
Me
ESI-II
Me
5
, 76%
12
Me
Me
TMS
13
ArS
Me
Ar = p-Tol
14, 40%
Me
Me
SnBu3
7
ArS
ArS
H
Ph
OMe
ArS
ii, NuC-II =
, 78%
15
i, Et2AlCl
(– MeO– )
ArS
+
SnBu3
Ph
– 30 C, 3 min
°
Ph
ESI-II
.
6
16
Ph
17, 78%
ii,
O
Me
OTMS
(NuC-II)
Me
OAc Me
ArS
9
i, TMSOTf/Et2AlCl
(– AcO– )
+
ArS
ArS
– 10 C, 12 h
°
8
ESI-II
Scheme 3
18, 45%
Previously,4 it was found that the cleavage of a methoxy
group from β-methoxyalkylaryl sulfides leading to the formation
of ESI-like intermediates can be achieved under the action of
Lewis acids.4 The substrates employed in all these cases con-
tained the properly positioned double bond or aromatic residue.
Hence, the cationic species formed in situ were able to undergo
an intramolecular reaction with these π-donors to give cycliza-
tion products. One might have anticipated that a similar ESI-like
intermediate generated from 5 (ESI-II, Scheme 3) would also
be useful as a carbon electrophile (EC) in intermolecular reac-
tions with various π-donors. However, the initial results of our
experiments to employ adduct 5 as the precursor of electrophilic
species were discouraging since in the presence of Lewis acids
(TiCl4, Et2AlCl, TMSOTf and SnCl4) no desired reaction occurred
under mild conditions (at – 70 to – 20 °C) with even such a reactive
carbon nucleophile as trimethylsiloxy cyclopent-1-ene 9 and only
trace amounts of expected product 10 were formed under more
severe conditions. We found that the desired transformation can
be carried out under the action of an excess of the mixed Lewis
acid TMSOTf– Et2AlCl at 0– 20 °C to give product 10 in a satis-
factory yield (Scheme 3). The same reaction conditions are
applicable to the Lewis acid-induced coupling of 5 with other
π-donors like trimethylsilylketene acetal 11 or methallylsilane
13, which furnished respective adducts 12 and 14.†† As would
be expected, the conversion of 6 into ESI-like electrophilic species
and interaction of the latter with methallylstannane 7 or allenyl-
stannane 16 proceeded under milder conditions (– 30 °C, Et2AlCl)
and resulted in the formation of products 15 or 17 in good yields.
The above examples referred to the generation of ESI-II inter-
mediates via the Lewis acid-induced removal of the methoxy
group from the tertiary carbon atom. This approach was found
††
The typical experimental procedure is described for the conversion of
adduct 5 into 1,1-bisadduct 12: TMSOTf (0.4 ml, 2.2 mmol) was added
to a stirred solution of Et2AlCl (1.2 ml, 1.8 M solution, 2.2 mmol) in
CH2Cl2 (5 ml) at 0 °C. After 5 min, a solution of ketene acetal 11 (150 mg,
0.86 mmol) and adduct 5 (110 mg, 0.43 mmol) in CH2Cl2 (5 ml) was added.
After stirring for 30 min at room temperature, TLC data revealed the
complete disappearance of the starting compound. The reaction mixture
was quenched with NaHCO3– H2O– Et2O; the organic layer was separated
and dried over MgSO4; the solvent was evaporated in a vacuum; the
residue was subjected to preparative TLC to give analytically pure product
12 in 76% yield. 1H NMR (250 MHz, CDCl3) d: 7.42 (d, 2H, J 8.3 Hz),
7.23 (d, 2H, J 8.3 Hz), 6.08– 5.92 (m, 1H), 5.33– 5.19 (m, 2H), 3.82 (s,
3H), 3.25 (s, 2H), 2.62– 2.42 (m, 2H), 2.47 (s, 3H), 1.43 (s, 6H), 1.22
(s, 3H). Found (%): C, 70.38; H, 8.52; S, 10.40. Calc. for C18H26O2S
(%): C, 70.54; H, 8.55; S, 10.46.
¶
The identity of all the mono- and 1,1-bisadducts was unambiguously
established by 1H and 13C NMR spectroscopy and microanalysis. Yields
refer to the isolated and purified products.
– 22 –

, 2003, 13(1), 21–23
ineffective when we used an analogue of adduct 5 bearing a
methoxy group at the secondary carbon atom (easily preparable
from the ArSCl adducts with methyl vinyl ether following the
general sequence shown in Scheme 2). Therefore, it was rather
rewarding to find out that this goal could be achieved for acetoxy
adduct 8 containing a better leaving group. The mixed Lewis
acid TMSOTf– Et2AlCl was found to be a preferable agent for
the generation of ESI-II species. The reaction of the latter formed
in situ with model carbon nucleophile 9 afforded expected
adduct 18.¶
Finally, we found that the transformations tentatively outlined
in Scheme 1 could be carried out as a one-pot four-component
coupling. Thus adduct 10 was formed in 36% yield as a result
of the consecutive Lewis acid-induced reactions of in situ formed
adduct 1 with silane 4 (NuC-I) followed by the treatment of
arising adduct 5 with silyl enol ether 9 (NuC-II). Similarly, a one-
pot procedure was also effective for the preparation of adducts
15 and 17 directly from α-methoxystyrene (via intermediate
adducts 2 and 6) in 65– 70% overall yields.‡‡
The above results, preliminary as they are, nevertheless re-
present a ‘proof-of-principle’ evidence attesting to the promise
of further studies in this area. Our current research efforts are
aimed at the condition optimization and broadening the scope of
the preparative application of the suggested novel 1,1-bisalkyla-
tion protocol with the help of cationic reactions as a potentially
promising method of the multi-component coupling. It is also
noteworthy that the described approach might be especially
useful for the ready preparation of the polyfunctional adducts
capable of serving as substrates for a number of intramolecular
transformations such as Pauson– Khand alkyne– alkene– carbonyl
cycloaddition (e.g. adduct 17), ene reaction (adducts 10 or 18)
or cationic cyclization (adducts 14 or 15).
This work was supported by the Russian Foundation for Basic
Research (grant no. 00-03-32790a) and the US CRDF (grant
no. RC 2-2207).
References
1 (a) W. L. Meyer, T. E. Goodwin, R. G. Hoff and C. W. Sigel, J. Org.
Chem., 1977, 42, 2761; (b) D. Seebach, B. Seuring, H.-O. Kalinowsky
and W. Lubosch, Angew. Chem., Int. Ed. Engl., 1977, 16, 264.
2 W. A. Smit, R. Caple and I. P. Smolyakova, Chem. Rev., 1994, 94, 2359.
3 (a) R. R. P. Alexander and I. Paterson, Tetrahedron Lett., 1983, 24, 5911;
(b) I. P. Smolyakova and W. A. Smit, Izv. Akad. Nauk SSSR, Ser. Khim.,
1985, 485 (Bull. Acad. Sci. USSR, Div. Chem. Sci., 1985, 34, 443).
4 (a) D. S. Checkmarev, M. I. Lazareva, G. V. Zatonsky, A. V. Maskaev,
R. Caple and W. A. Smit, J. Org. Chem., 2002, 67, 7957; (b) S. R.
Harring and T. Livinghouse, Tetrahedron Lett., 1989, 30, 1499.
‡‡
The typical experimental procedure is described for the one-pot prepara-
tion of 17: β-methoxystyrene (45 mg, 0.33 mmol), allyltributyltin 4b
(120 mg, 0.37 mmol) and AgOTf (103 mg, 0.44 mmol) were added to a
stirred solution of p-TlSCl (53 mg, 0.33 mmol) in CH2Cl2 (3 ml) at – 50 °C.
The reaction mixture was stirred for 1 h at – 40 °C and additionally for
30 min at room temperature. After that the mixture was cooled down to
– 30 °C and tributylallenyltin 16 (91 mg, 0.26 mmol) and Et2AlCl (0.56 ml,
1.8 M solution, 1 mmol) were added. TLC data revealed the complete con-
version of adduct 6 into target compound 17. The reaction mixture was
quenched with NaHCO3– H2O– Et2O; the organic layer was separated and
dried over MgSO4 and the solvent was evaporated in a vacuum. The
residue was subjected to preparative TLC to give analytically pure product
17 in 60% yield. 1H NMR (250 MHz, CDCl3) d: 7.40– 7.00 (m, 9H),
5.55– 5.40 (m, 1H), 5.15– 5.00 (m, 2H), 3.45 (s, 2H), 2.92 (dd, 1H,
J 16.1 Hz, J 2.4 Hz), 2.79 (dd, 1H, J 16.1 Hz, J 2.4 Hz), 2.65– 2.75 (m,
2H), 2.30 (s, 3H), 2.05 (t, 1H, J 2.4 Hz). Found (%): C, 82.37; H, 7.24;
S, 10.41. Calc. for C21H22S (%): C, 82.30; H, 7.24; S, 10.46.
Received: 8th January 2003; Com. 03/2036
– 23 –