β-arylthio-α-chloroalkyl ethers - Novel 1,1-bis-electrophiles for geminal alkylation

Article abstract of DOI:10.1002/ejoc.200500689

A novel protocol for geminal alkylation is suggested based upon the consecutive generation of two cationoid intermediates from the easily prepared adducts of arylsulfenyl chlorides and vinyl ethers. Wiley-VCH Verlag GmbH & Co. KGaA, 2006.

Full text of DOI:10.1002/ejoc.200500689

FULL PAPER
DOI: 10.1002/ejoc.200500689
β-Arylthio-α-chloroalkyl Ethers − Novel 1,1-Bis-electrophiles for Geminal
Alkylation
Alexey V. Gromov[a] and William A. Smit*[a]
Keywords: Episulfonium ion / Arylthio-stabilized electrophiles / Geminal bis-electrophiles / Alkylation / Geminal alky-
lation / Synthetic methods
A novel protocol for geminal alkylation is suggested based
upon the consecutive generation of two cationoid intermedi-
ates from the easily prepared adducts of arylsulfenyl chlo-
rides and vinyl ethers.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
Introduction
with allylsilanes or stannanes followed by Friedel–Crafts al-
kylation in the presence of HfCl4 or Cl2Si(OTf)2 as Lewis
acid.[5] An entirely different and synthetically promising
protocol for the sequential one-pot introduction of two dif-
ferent carbon nucleophiles at the same carbon atom was
elaborated by Suga et al.[6] As was shown in these studies,
methyl 2,2-bis(trimethylsilyl)pyrrolidine-1-carboxylate could
be employed as a convenient precursor for the stepwise
electrochemical generation of N-acyliminium ions in a con-
trollable way that secured a highly selective substitution of
both silyl groups with two different nucleophiles.
The ArS group is known to anchimerically facilitate the
substitution of a nucleofuge at the β-position as a result of
the stabilization of the developing carbocationic center via
the formation of a episulfonium ion (ESI) type intermedi-
ate.[7] Once formed these intermediates exhibit rather high
electrophilic activity and their reaction with a number of
carbon nucleophiles (NuC), like silicon- or tin-capped π do-
nors, is well-documented as a useful preparative method
(see below). Hence one might have suggested that placing
two leaving groups (LGs) at the β-position to an ArS group,
as shown in the fragment 1, would provide an opportunity
for employing the latter as an equivalent of geminal bis-
electrophilic synthon 1Ј provided effective control could be
executed over the selectivity of the consecutive reactions, as
shown in Scheme 1.
The availability of diverse and reliable methods for per-
forming a sequence of alkylation reactions leading to the
introduction of two different substituents at the same car-
bon atom is of obvious importance for organic synthesis.
Most commonly this goal could be achieved by using
various types of synthetic equivalents of 1,1-bis-nucleo-
philic synthons. The list of compounds which could be used
in this role includes such classical reagents as malonic and
acetoacetic esters,[1a] nitromethane[1b] and various types of
thioacetals.[1c] Ample experimental evidence attests to the
versatility and preparative usefulness of the general proto-
col based upon a sequential one-pot 1,1-bis-alkylation of
the carbanionic species derived from these types of precur-
sors.[2]
There is much less data referring to the alternative option
based upon the sequential alkylation of a 1,1-bis-electro-
philic center with two different carbon nucleophiles (NuC).
Cahiez et al. employed various acyl chlorides as synthetic
equivalents of 1,1-electrophilic synthons in an efficient one-
pot preparation of unsymmetrical tertiary alcohols which
involves consecutive treatment of these electrophiles first
with organomanganese and then with organomagnesium
(or lithium) compounds.[3] Mayr and Gorath described the
formation of the products of 1,1-geminal allylation in the
BHal3-initiated reaction of aromatic aldehydes or acetals
with allylsilanes which was shown to proceed via the se-
quential generation of cationoid species at the benzylic car-
bon atom.[4] A similar use of aromatic aldehydes as 1,1-bis-
electrophiles was described recently by Shiina et al. who
have shown that a series of 4,4-diarylbut-1-enes could be
prepared by the successive allylation of aromatic aldehydes
Here we wish to report the results of our study attesting
to the validity of this concept and its potential for the elab-
oration of a new methodology of controlled geminal bis-
alkylation.
β-Arylthio-α-chloroalkyl ethers of the general formula 2
(Scheme 2) were chosen as promising candidates for the role
of synthetic equivalents to the synthon 1Ј, both because of
the obvious difference in the nucleofugacity of the Cl and
MeO groups and their easy availability through the AdE
reaction of ArSCl to vinyl ethers.[8]
[a] N. D. Zelinsky Institute of Organic Chemistry, Russian Acad-
emy of Sciences,
Leninsky prospect 47, 119991 Moscow, Russian Federation
Fax: +7-095-135-53-28
E-mail: smt@ioc.ac.ru
Eur. J. Org. Chem. 2006, 1317–1322
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1317

A. V. Gromov, W. A. Smit
FULL PAPER
Results and Discussion
In order to evaluate the ease of generation of ESI-II in-
termediates, the interaction of adduct 3a[11] with trimethyl-
silyl enolate 4a in the presence of Lewis acids within the
temperature range of –78 and 20 °C was studied as a model
reaction. Initially TiCl4 was employed as the Lewis acid and
it was found that no reaction occurred at low temperatures
(TLC monitoring data). However, an increase in tempera-
ture up to 20 °C resulted in fast and complete conversion of
the adduct 3a. However, instead of the expected alkylation
product 5, the Cl adduct 6a (evidently as a result of the
substitution of MeO by Cl) was exclusively formed
(Scheme 3).
Scheme 1.
Scheme 3. *6b was obtained upon hydrolysis of 6a on silica gel
(see Exp. Sect.).
The clean formation of the adduct 6a prompted us to
study its potential as an alternative precursor for the desired
transformation. It turned out that Lewis acids like TiCl4,
BF3, TMSOTf, BF3 + TMSOTf, or AgSbF6 could not pro-
mote the reaction of 6a (formed in situ from 3a) with 4a.
Scheme 2.
Earlier studies revealed that the interaction of adducts 2 No reaction occurred at low temperatures and complex
with a Lewis acid leads to the generation of a cationoid mixtures of unidentified products were formed at ambient
intermediate of ESI-I type which can be effectively used as temperatures. However, the use of Et2AlCl resulted in the
the electrophile in C–C bond-forming reactions with a wide formation of the target product 5, albeit in a rather modest
range of carbon nucleophiles (NuC) such as vinyl silyl ethers yield.
and ketene acetals, allylsilanes or stannanes resulting in the
This Lewis acid was shown to be useful for promoting
formation of the corresponding adducts 3 in good-to-excel- the reactions of in situ formed chloro adduct 6a with a
lent yields (Scheme 2).[9] Regardless of the structure of the number of other carbon nucleophiles such as silyl enolates
substrate and the nature of the Lewis acid, in all cases the 4b,c and silyl ketene acetal 4d, giving the expected alky-
formation of the electrophilic ESI-I involved the selective lation products 7–9 in moderate yields (see procedure A,
removal of chlorine (LG-I) due to the better nucleofugacity Expt. Sect.) (Scheme 4).
of this group and the enhanced stabilization of the interme-
diate ESI-I by the methoxy substituent at the developing acid Et2AlCl·TMSOTf[12,13] allowed us to perform the di-
cationoid center. rect reaction of the methoxy adduct 3a with a series of car-
Further studies revealed that use of the stronger Lewis
While the presence of the β-methoxyalkyl arylthio frag- bon nucleophiles thus bypassing the necessity to obtain the
ment in adducts 3 makes them potentially useful precursors above-mentioned chloro adduct 6a. As is shown in
for the Lewis acid induced generation of the next cationoid Scheme 5, under these conditions the alkylation of siloxy-
intermediate (ESI-II, Scheme 1), the viability of such a alkenes 4c,d as well as methallylsilane 4e furnishes the cor-
transformation is far from self-evident in view of the rather responding adducts 8–10 in satisfactory-to-good yields (see
poor nucleofugacity of the methoxy group. In fact, previous procedure B, Expt. Sect.). Finally, we have also demon-
literature data pertinent to the ability of the methoxy group strated the possibility of preparing the adduct 8 starting
to serve as a nucleofuge in related transformations are from 2-methoxypropene by a one-pot sequence of three
rather scarce and refer exclusively to intramolecular cycliza- AdE reactions, as is also shown in Scheme 5 (see procedure
tion reactions initiated by the generation of ESI-like inter- C, Expt. Sect.).
mediates from substrates bearing appropriately positioned
The possibility of extending the scope of the above-
shown transformations was also studied for [α-chloro-α-
double bond(s).[10a,10b]
1318
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1317–1322

1,1-Bis-electrophiles for Geminal Alkylation
FULL PAPER
Scheme 5. *An inseparable mixture of 10 and p-
TolSCH2C(CH3)=CH2 (20% of the total, NMR spectroscopic
data) was formed.[14]
fore it was not surprising to find that Et2AlCl was suffic-
iently active as the Lewis acid in the reaction of 13 with
methallylstannane 14 or allenylstannane 15, employed as
NuC-II, to give the expected adducts 16 or 17 in good yields
(Scheme 6) (see procedure D, Expt. Sect.). We also found
that the consecutive transformation of α-methoxystyrene
first to the adduct 11, then to the product 13 and finally to
the 1,1-bis-alkylation products 16 or 17 could be carried
out as a one-pot sequence with a good overall yield (see
Scheme 4.
methoxy-β-(p-tolylthio)ethyl]benzene (11), readily prepared procedure E, Expt. Sect.).
from α-methoxystyrene and p-TolSCl.[15] No data were
It is to be emphasized that in the latter sequence as well
available concerning the Lewis acid initiated reactions of 11 as in the one-pot preparation of adduct 8 (Scheme 5) the
with π donors. After some preliminary experimentation it initially formed adducts of p-TolSCl and the respective vinyl
was found that the interaction of 11 with allylstannane 12 ethers act as the precursors of 1,1-bifunctional electrophiles
proceeds smoothly under the action of AgSbF6 at –50 °C following the mechanism which was tentatively outlined in
to furnish the corresponding adduct 13 in a good yield.
Scheme 1.
The nucleofugacity of the methoxy group in the adduct
The functionalities present in the adducts 5, 7–10, 16 and
13 should clearly be higher than that in the adduct 3. There- 17 suggest quite a number of options for their further use
Scheme 6.
Eur. J. Org. Chem. 2006, 1317–1322
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1319

A. V. Gromov, W. A. Smit
FULL PAPER
organic layers were dried with MgSO4 and solvents were removed
in vacuo. The resulting oil was subjected to column chromatog-
raphy (ethyl acetate/hexane, gradient 0:100 to 1:60) or preparative
TLC (ethyl acetate/hexane, 1:20). The Rf (ethyl acetate/hexane,
10:1) value, the method of purification, yield and physical data for
each product (5, 7–9) are given below.
as substrates in various intramolecular cyclization reactions,
such as the carbonyl-ene reaction, electrophilic cyclization
of 1,5-dienes or intramolecular Pauson–Khand cyclization
reaction.
In order to illustrate the usefulness of the described syn-
thetic protocol, the feasibility of the Pauson–Khand trans-
formation[16] was studied for the 1,6-enyne 17 (Scheme 7).
It was found that treatment of 17 with Co2(CO)8 and then
trimethylamine N-oxide led to the formation of the ex-
pected bicyclo[3.3.0]octenone 18 in a satisfactory yield
(nonoptimized). Note that this product was eventually pre-
pared from three simple precursors in only two separate
steps in 40% overall yield.
Procedure B: TMSOTf (0.4 mL, 2.2 mmol) was added to a cold
(0 °C) solution of Et2AlCl (2.2 mmol, 1.2 mL of a 1.8  solution
in toluene) in dry CH2Cl2 (5 mL) and the resulting mixture was
stirred for 5 min. A solution of 3a (101 mg, 0.43 mmol) and silyl
enolate (0.86 mmol) in CH2Cl2 (5 mL) was added, the cooling bath
removed and the reaction mixture stirred at room temp. until (ca.
30 min) TLC monitoring (ethyl acetate/hexane, 10:1) showed com-
plete consumption of 3a (Rf = 0.38). The reaction mixture was
quenched with a saturated solution of NaHCO3 and extracted with
diethyl ether. The combined organic layers were dried with MgSO4
and solvents were removed in vacuo. The resulting oil was subjected
to column chromatography (ethyl acetate/hexane, gradient 0:100 to
1:60, unless otherwise mentioned) or preparative TLC (ethyl ace-
tate/hexane, 1:20). The Rf (ethyl acetate/hexane, 10:1) value,
method and conditions of purification, yield and physical data for
the products 8–10 are given below.
Scheme 7.
Procedure C: 2-Methoxypropene (63 mg, 0.88 mmol) was added
dropwise to a cold (–70 °C) orange solution of p-TolSCl (126 mg,
0.79 mmol) in dry CH2Cl2 (3 mL). After 10 min the reaction mix-
ture turned colourless. Allyltrimethylsilane (94 mg, 0.82 mmol) and
Et2AlCl (1.0 mmol, 0.55 mL of a 1.8  solution in toluene) were
added and in 5 min TLC monitoring (ethyl acetate/hexane, 10:1)
showed complete formation of 3a (Rf = 0.38). (Trimethylsiloxy)
cyclopentene (4c) (475 mg, 3.1 mmol) followed by Et2AlCl
Conclusions
The results shown above provide “proof-of-principle”
evidence for controlled geminal alkylation as a novel proto-
col for the easy assembly of polyfunctional substrates from
simple starting compounds. Further studies aimed at opti- (3.8 mmol, 2.13 mL of a 1.8  solution in toluene) and TMSOTf
(0.7 mL, 3.8 mmol) were added and the reaction mixture was
stirred at room temp. for 15 min until TLC monitoring (ethyl ace-
tate/hexane, 10:1) showed complete consumption of 3a and forma-
tion of 8 (Rf = 0.22). The reaction mixture was quenched with a
saturated solution of NaHCO3 and extracted with diethyl ether.
The combined organic layers were dried with MgSO4 and solvents
were removed in vacuo. The residue was purified by column
chromatography (ethyl acetate/hexane, 0:100 to 1:40) to give 82 mg
(36%) of 8 (dr = 1:1).
mizing the reaction conditions and broadening the prepara-
tive scope of the disclosed reaction sequence are most cer-
tainly warranted.
Experimental Section
General: 1H and 13C NMR spectra were recorded on a 200, 250,
300 or 500 MHz instrument at 22 °C unless otherwise noted. The
chemical shifts (δ) are measured in ppm relative to internal CHCl3
(δ = 7.26 ppm for 1H) or CDCl3 (δ = 77 ppm for 13C). All reactions,
unless otherwise noted, were conducted under argon. Liquid rea-
gents were transferred to reaction flasks through a rubber septum
using a hypodermic syringe.
Procedure D: Et2AlCl (0.79 mmol, 0.44 mL of a 1.8  solution in
toluene) was added to a cold solution (–30 °C) of 13 (130 mg,
0.44 mmol) and 14 or 15 (0.48 mmol) in dry CH2Cl2 (5 mL). The
resulting mixture was stirred for 5 min until TLC monitoring (for
14: ethyl acetate/hexane, 10:1, 13 Rf = 0.38; for 15: toluene, 13 Rf
= 0.73) showed complete consumption of 13. The reaction mixture
was quenched with a saturated solution of NaHCO3 and extracted
with diethyl ether. The combined organic layers were dried with
MgSO4, solvents were removed in vacuo and the residue was puri-
fied by column chromatography (ethyl acetate/hexane, 0:100 to
1:50) to give the product 16 or 17. The physical data for 16 and 17
are given below.
Materials: Unless otherwise noted, materials were obtained from
commercial suppliers and reagent grade materials were used with-
out further purification. CH2Cl2 was freshly distilled over CaH2
prior to use. THF was commercially available and used as received.
pTolSCl was prepared following a previously described pro-
cedure.[17] Analytical TLC was performed on precoated silica gel
60 F254. Column chromatography was performed on silica gel 60–
120 µ.
Procedure E: α-Methoxystyrene (45 mg, 0.33 mmol) was added
dropwise to a cold orange (–50 °C) solution of p-TolSCl (53 mg,
0.33 mmol) in dry CH2Cl2 (3 mL). After 1 min the reaction mixture
turned colourless. Allyltributylstannane (12) (122 mg, 0.37 mmol)
and AgOTf (103 mg, 0.4 mmol) were added and the resulting mix-
ture was stirred at –40 °C for 1 h until TLC monitoring (toluene)
showed complete formation of 13. The cooling bath was removed
and stirring was continued at room temp. for 30 min (to make an
excess of allyltributylstannane react with AgOTf) followed by cool-
ing to –30 °C and addition of 14 or 15 (0.43 mmol) and Et2AlCl
Procedure A: Titanium tetrachloride (0.1 mmol, 19 mg in 0.12 mL
of CH2Cl2) was added to a cold solution (0 °C) of 3a (46 mg,
0.2 mmol) in dry CH2Cl2 (2 mL). The reaction mixture turned yel-
low. After 3 min TLC monitoring (ethyl acetate/hexane, 10:1)
showed complete conversion of 3a (Rf = 0.38) into chloride 6a (Rf
= 0.55). Corresponding silyl enolate (0.8 mmol) was added followed
by Et2AlCl (0.6 mmol, 0.33 mL of a 1.8  solution in toluene). Af-
ter 1.5 h the reaction mixture was quenched with a saturated solu-
tion of NaHCO3 and extracted with diethyl ether. The combined
1320
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1317–1322

1,1-Bis-electrophiles for Geminal Alkylation
FULL PAPER
(1 mmol, 0.56 mL of a 1  solution in toluene). The resulting mix-
ture was stirred for 5 min until TLC monitoring (for 14: ethyl ace-
tate/hexane, 10:1, 13 Rf = 0.38; for 15: toluene, 13 Rf = 0.73)
showed complete consumption of 13. The reaction mixture was
quenched with a saturated solution of NaHCO3 and extracted with
diethyl ether. The combined organic layers were dried with MgSO4,
the solvent was removed in vacuo and the residue was purified by
column chromatography to give the product 16 or 17. The physical
data for 16 and 17 are given below.
NMR (250 MHz, CDCl3): δ = 7.32–7.08 (m, 4 H, Ar), 5.93–5.26
(m, 1 H, CH2=CHCH2), 5.18–5.06 (m, 2 H, CH2=CH), 4.93, 4.79
(2 s, 2 H, CH3–C=CH2), 2.93 (s, 2 H, CH2S), 2.35 (s, 3 H, CH3–
Ar), 2.30–2.10 (m, 2 H, CH2–C=CH2), 1.84 (s, 3 H, CH3–C=C),
1.03 (s, 3 H, CH3–Cq) ppm. GC-MS: m/z = 260 [M]+.
1-(2-Chloro-2-methoxy-2-phenylethylsulfanyl)-4-methylbenzene (11).
NMR-Monitored
Preparation:
α-Methoxystyrene
(23 mg,
0.17 mmol) was added to a cold (–50 °C) orange solution of p-
TolSCl (26 mg, 0.17 mmol) in dry CD2Cl2 (1 mL). After 1 min the
reaction mixture turned colourless. A syringe cooled in dry ice was
used to transfer 0.5 mL of the prepared solution to a NMR tube
closed with a septum and placed in a bath at –78 °C. 1H NMR
(500 MHz, CD2Cl2, 203°K): δ = 7.60–7.06 (m, 9 H, Ar), 3.92 (d, J
= 3 Hz, 1 H, CHaS), 3.78 (d, J = 3 Hz, 1 H, CHbS), 3.43 (s, 3 H,
OCH3), 2.28 (s, 3 H, CH3–Ar) ppm. 13C NMR (75 MHz, CD2Cl2):
δ = 140.8, 137.9, 132.3, 11.6, 130.5, 129.7, 129.1, 127.6, 111.0, 54.2,
49.4, 21.8 ppm. The instability of the adduct 11 precluded its isola-
tion and it was used, prepared in situ, as described above.
4-Methyl-4-(p-tolylsulfanylmethyl)hept-6-en-2-one (5): Procedure A:
yield 21% (11.0 mg); purification: column chromatography; Rf =
0.29. 1H NMR (250 MHz, CDCl3): δ = 7.30–7.27 (d, 2 H, Ar),
7.10–7.07 (d, 2 H, Ar), 5.85–5.68 (m, 1 H, CH2=CHCH2), 5.11–
5.04 (m, 2 H, CH2=CH), 3.14 (d, J = 12 Hz, 1 H, CHaS), 3.07 (d,
J = 12 Hz, 1 H, CHbS), 2.52 (s, 2 H, CH2C=O), 2.36–2.17 (m, 2
H, CH2=CHCH2), 2.32 (s, 3 H, CH3–Ar), 2.04 (s, 3 H, CH3–C=O),
1.10 (s, 3 H, CH3–Cq) ppm. 13C NMR (60 MHz, CDCl3): δ =
207.8, 135.9, 133.9, 133.6, 130.1, 129.6, 118.4, 49.6, 44.5, 43.4, 37.7,
31.7, 24.4, 20.9 ppm. C16H22OS (262.41): C 73.23, H 8.45, S 12.22;
found C 73.25, H 8.56, S 12.00.
1-[(2-Methoxy-2-phenyl-4-pentenyl)sulfanyl]-4-methylbenzene (13):
α-Methoxystyrene (317 mg, 2 mmol) was added dropwise to a cold
(–50 °C) orange solution of p-TolSCl (317 mg, 2 mmol) in dry
CH2Cl2 (5 mL). After 1 min the reaction mixture turned colourless.
The temperature was lowered to –70 °C and allyltributylstannane
(12) (795 mg, 2.4 mmol) followed by AgSbF6 (756 mg, 2.2 mmol in
1 mL of CH2Cl2) were added. The resulting mixture was stirred for
1 h (TLC monitoring, ethyl acetate/hexane, 1:10, showed complete
formation of 13, Rf = 0.38), quenched with a saturated solution of
NaHCO3 and extracted with diethyl ether. The combined organic
layers were dried with MgSO4, the solvent removed in vacuo and
the residue purified by chromatography (ethyl acetate/hexane, 0:100
to 1:60) to give 466 mg (79%) of 13. 1H NMR (250 MHz, CDCl3):
δ = 7.41–7.05 (m, 9 H, Ar), 5.70–5.54 (m, 1 H, CH2=CHCH2),
5.14–5.06 (m, 2 H, CH2=CH), 3.50 (d, J = 12.5 Hz, 1 H, CHaS),
3.47 (d, J = 12.5 Hz, 1 H, CHbS), 3.17 (s, 3 H, OCH3), 2.97–2.77
(m, 2 H, CH2–C=CH2), 2.33 (s, 3 H, CH3–Ar) ppm. 13C NMR
(50 MHz, CDCl3): δ = 142.2, 13.1, 133.3, 132.9, 130.6, 129.6, 129.4,
128.2, 127.4, 126.7, 118.7, 80.6, 50.5, 42.5, 40.9, 21.0 ppm.
C19H22OS (298.44): C 76.46, H 7.43, S 10.74; found C 76.35, H
7.79, S 10.76.
3-Methyl-1-phenyl-3-(p-tolylsulfanylmethyl)hex-5-en-1-one (7): Pro-
cedure A: yield 26% (16.9 mg); purification: preparative TLC; Rf
1
= 0.42. H NMR (250 MHz, CDCl3): δ = 8.03–7.01 (m, 9 H, Ar),
5.90–5.73 (m, 1 H, CH2=CHCH2), 5.12–5.06 (m, 2 H, CH2=CH),
3.26, (d, J = 15 Hz, 1 H, CHaS), 3.17 (d, J = 15 Hz, 1 H, CHbS),
3.06 (s, 2 H, CH2C=O), 2.48–2.33 (m, 2 H, CH2=CHCH2), 2.27 (s,
3 H, CH3–Ar), 1.17 (s, 3 H, CH3–C) ppm. 13C NMR (60 MHz,
CDCl3): δ = 199.3, 138.3, 136.1, 134.1, 132.8, 130.5, 129.6, 128.7,
128.0, 125.7, 118.7 ppm. C21H24OS (324.48): C 77.73, H 7.46, S
9.88; found C 77.95, H 7.78, S 9.51.
2-[1-Methyl-1-(p-tolylsulfanylmethyl)but-3-enyl]cyclopentanone (8):
Procedure A: yield 43% (24.8 mg); purification: preparative TLC;
procedure B: yield 56% (69.5 mg); purification: column chromatog-
raphy; one-pot procedure C: yield 36%; purification: column
chromatography; Rf = 0.22; NMR spectroscopic data are given for
1
a mixture of diastereomers (4:5). H NMR (250 MHz, CDCl3): δ
= 7.32–7.08 (m, 4 H, Ar), 5.88–5.70 (m, 1 H, CH2=CHCH2), 5.14–
5.08 (m, 2 H, CH2=CH), 3.28 (s, 2 H, CH2S), 3.22 (d, J = 12 Hz,
1 H, CHaS), 2.92 (d, J = 12 Hz, 1 H, CHbS), 2.58–1.62 (m, 9 H,
CH2=CHCH2, CH2CH2CH2CHC=O), 2.33 (s, 3 H, CH3–Ar),
1.04, 1.02 (2 s, 3 H, CH3–Cq) ppm. 13C NMR (60 MHz, CDCl3):
δ = 219.8, 219.7, 135.9, 135.7, 134.1, 133.7, 130.1, 129.9, 129.6,
118.5, 118.4, 54.2, 43.9, 42.9, 41.6, 40.7, 40.3, 39.8, 39.6, 25.9, 21.7,
20.9, 20.1 ppm. C18H24OS (288.45): C 74.95, H 8.39, S 11.12; found
C 74.66, H 8.33, S 10.80.
2-Allyl-4-methyl-2-phenylpent-4-enyl 4-Methylphenyl Sulfide (16):
Procedure D: yield 80% (113.5 mg); procedure E: yield 73%
(77.7 mg); Rf = 0.35 (ethyl acetate/hexane, 10:1). 1H NMR
(300 MHz, CDCl3): δ = 7.39–7.07 (m, 9 H, Ar), 5.70–5.56 (m, 1 H,
CH2=CHCH2), 5.30–5.02 (m, 2 H, CH2=CH), 4.79 (s, 1 H, CH3–
C=CHa), 4.69 (s, 1 H, CH3–C=CHa), 3.40 (s, 2 H, CH2S), 2.74 (d,
J = 7 Hz, 2 H, CH2–C=CH2), 2.64 (d, J = 15 Hz, 1 H, CHa–
C=CH2), 2.53 (d, J = 15 Hz, 1 H, CHb–C=CH2), 2.34 (s, 3 H, CH3–
Ar), 1.24 (s, 3 H, CH3–C=C) ppm. 13C NMR (75 MHz, CDCl3): δ
= 144.8, 142.2, 135.9, 134.0, 130.3, 129.6, 128.1, 126.8, 126.2, 118.6,
115.1, 47.0, 44.8, 43.2, 41.5, 24.7, 21.1 ppm. C22H26S (322.51): C
81.93, H 8.13, S 9.94; found C 81.59, H 8.26; S 9.96.
Methyl 2,2,3-Trimethyl-3-(p-tolylsulfanylmethyl)hex-5-enoate (9):
Procedure A: yield 45% (27.6 mg); purification: preparative TLC;
procedure B: yield 76% (100.2 mg); purification: column
chromatography; Rf = 0.67. 1H NMR (300 MHz, CDCl3): δ =
7.28–7.07 (m, 4 H, Ar), 5.93–5.77 (m, 1 H, CH2=CHCH2), 5.11–
5.03 (m, 2 H, CH2=CH), 3.66 (s, 3 H, OCH3), 3.13 (d, J = 12 Hz,
1 H, CHaS), 3.12 (d, J = 12 Hz, 1 H, CHbS), 2.44–2.32 (m, 2 H,
CH2=CHCH2), 2.32 (s, 3 H, CH3–Ar), 1.26 [s, 6 H, C-(CH3)2], 1.05
(s, 3 H, CH3–Cq) ppm. 13C NMR (60 MHz, CDCl3): δ = 177.2,
136.0, 135.4, 134.8, 130.0, 129.7, 117.9, 51.6, 51.6, 49.1, 22.4, 22.2,
21.3, 21.1 ppm. C18H26O2S (306.46): C 70.54, H 8.55, S 10.46;
found C 70.22, H 8.45, S 10.19.
1-Methyl-4-(2-phenyl-2-prop-2-ynylpent-4-enylsulfanyl)benzene (17):
Procedure D: yield 78% (105.2 mg); procedure E: yield 60%
1
(60.7 mg); Rf = 0.88 (toluene). H NMR (250 MHz, CDCl3): δ =
7.41–7.06 (m, 9 H, Ar), 5.57–5.43 (m, 1 H, CH2=CHCH2), 5.17–
5.03 (m, 2 H, CH2=CH), 3.48 (s, 2 H, CH2S), 2.92 (dd, J1 = 17, J2
= 3 Hz, 1 H, CHa–CϵC), 2.82 (dd, J1 = 17, J2 = 3 Hz, 1 H, CHb–
CϵC), 2.74–2.70 (m, 2 H, CH2=CHCH2), 2.33 (s, 3 H, CH3–Ar),
2.02 (dd, J1 = J2 = 3 Hz, 1 H, CϵCH) ppm. 13C NMR (75 MHz,
CDCl3): δ = 143.0, 136.1, 133.5, 133.4, 130.8, 129.5, 128.2, 126.5,
1-(2-Allyl-2,4-dimethylpent-4-enylsulfanyl)-4-methylbenzene
(10):
Procedure B: yield 40% (60.0 mg of a mixture of products); purifi-
cation: column chromatography (hexane); adduct 10 could not be
separated from ca. 50 mol.% of p-TolSCH2C(Me)=CH2 (evaluated 118.6, 81.2, 71.3, 45.1, 44.9, 42.8, 26.4, 21.1 ppm. C21H22S (306.46):
from the integration of the 1H NMR spectrum); Rf = 0.55. 1H
C 82.30, H 7.24, S 10.46; found C 81.92, H 7.22, S 10.48.
Eur. J. Org. Chem. 2006, 1317–1322
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1321

A. V. Gromov, W. A. Smit
5-[(4-Methylphenyl)sulfanylmethyl]-5-phenyl-3a,4,5,6-tetrahydro- (RFBR, Grant Nos. 03-03-32884 and 03-03-04001) for financial
FULL PAPER
2(3H)-pentalenone (18): A solution of 17 (200 mg, 0.65 mmol) and
Co2(CO)8 (220 mg, 0.78 mmol) in THF (10 mL) was stirred at
room temp. for 1 h until CO liberation stopped and the solution
turned cherry-red. TLC monitoring (ethyl acetate/hexane, 1:10)
showed complete consumption of 17 (Rf = 0.38) and the formation
of a brown-coloured complex (Rf = 0.75). Me3NO·H2O (360 mg,
3.9 mmol) was added and the reaction mixture stirred for an ad-
ditional 16 h. TLC monitoring showed complete consumption of
the complex and formation of 18 (Rf = 0.17). The reaction mixture
was filtered through a plug of silica gel (4 cm) using diethyl ether
as eluent. Solvents were removed in vacuo and the residue was
purified by silica gel column chromatography (ethyl acetate/hexane,
1:10 to 1:5) to give 143 mg (66%) of 18 (inseparable mixture of
diastereomers, dr = 1:1). NMR spectroscopic data is given for the
support of this work.
[1] For a general overview and examples, see: a) J. C. Stowell,
Carbanions in Organic Synthesis, Wiley Interscience, New York,
1979, chapter 6, pp. 192, 198; b) J. C. Stowell, Carbanions in
Organic Synthesis, Wiley Interscience, New York, 1979, chapter
6, p. 212; c) J. C. Stowell, Carbanions in Organic Synthesis,
Wiley Interscience, New York, 1979, chapter 4, p. 103.
[2] For example, see: A. B. Smith III, A. M. Boldi, J. Am. Chem.
Soc. 1997, 119, 6925–6926, and references therein.
[3] G. Cahiez, J. Rivas-Enterrios, H. Granger-Veyron, Tetrahedron
Lett. 1986, 27, 4441–4444.
[4] H. Mayr, G. Gorath, J. Am. Chem. Soc. 1995, 117, 7862–7868.
[5] I. Shiina, M. Suzuki, K. Yokoyama, Tetrahedron Lett. 2002,
43, 6395–6398.
1
mixture of diastereomers. H NMR (500 MHz, CDCl3): δ = 7.48–
[6] S. Suga, M. Watanabe, J.-i. Yoshida, J. Am. Chem. Soc. 2002,
7.00 (m, 9 H, Ar), 5.95, 5.92 (2 s, 1 H, =CH–C=O), 3.47–3.04 (m,
4 H, CH2S, CH2C=O), 2.87–2.60 (m, 3 H, CH2–C=C, CH–C=C),
2.39 (s, 3 H, CH3–Ar), 2.14–2.18, 1.76–1.70 (br. d, m, 2 H, Ph–C–
CH2–CH) ppm. 13C (125 MHz, CDCl3): δ = 210.0, 209.7, 188.1,
146.8, 145.7, 136.3, 133.2, 130.3, 129.6, 128.5, 128.3, 126.7, 126.6,
126.5, 126.3, 125.3, 125.3, 53.4, 53.2, 49.7, 49.5, 44.7, 44.5, 42.8,
42.6, 42.5, 42.2, 40.5, 39.6, 21.0 ppm. C22H22OS (334.47): C 79.00,
H 6.63, S 9.59; found C 79.15, H 6.57, S 9.94.
124, 14824–14825.
[7] For a discussion, see the review: C. J. M. Stirling, “Sulfonium
Salts”, in Organic Chemistry of Sulphur (Ed.: S. Oae), Plenum
Press, New York, 1977, chapter 9, pp. 474–525.
[8] a) K. Toyoshima, T. Okuyama, T. Fueno, J. Org. Chem. 1978,
43, 2789–2792; b) M. A. Ibragimov, M. I. Lazareva, W. A.
Smit, Synthesis 1985, 880–884.
[9] For reviews, see: a) W. A. Smit, R. Caple, I. P. Smolyakova,
Chem. Rev. 1994, 94, 2359–2382; b) W. A. Smit, M. I. Lazar-
eva, I. P. Smolyakova, R. Caple, Russ. Chem. Bull. 2001, 50,
1949–1970.
[10] For examples, see the data in: a) S. R. Harring, T. Livinghouse,
Tetrahedron Lett. 1989, 30, 1499–1502; b) D. S. Chekmarev,
M. I. Lazareva, G. V. Zatonsky, A. V. Maskaev, R. Caple, W. A.
Smit, J. Org. Chem. 2002, 67, 7957–7967, and references
therein.
1-(2-Chloro-2-methylpent-4-enylsulfanyl)-4-methylbenzene (6a): Ti-
tanium tetrachloride (0.22 mmol, 42 mg in 0.24 mL of CH2Cl2) was
added to a cold solution (0 °C) of 3a (100 mg, 0.42 mmol) in dry
CH2Cl2 (4 mL). The reaction mixture turned yellow. In 3 min TLC
monitoring (ethyl acetate/hexane, 10:1) showed complete conver-
sion of 3a (Rf = 0.38) into 6a (Rf = 0.55). The reaction mixture
was quenched with a saturated solution of NaHCO3 and extracted
with diethyl ether. The combined organic layers were dried with
MgSO4, filtered through a plug of Celite and the solvent was re-
moved in vacuo. The residue was dissolved in hexane (30 mL) and
filtered through a plug of silica gel (3 mm). Evaporation of hexane
in vacuo gave 83 mg (75%) of 6a. Chloride 6a reacts with AgOTf
in DCM to give a precipitate of AgCl. Subjection of 6a to column
chromatography (ethyl acetate/hexane, 1:10) partly converts 6a into
the corresponding alcohol 6b probably due to hydrolysis on silica
gel. 1H NMR (200 MHz, CDCl3): δ = 7.35–7.09 (m, 4 H, Ar), 5.99–
5.80 (m, 1 H, CH2=CHCH2), 5.19–5.10 (m, 2 H, CH2=CH), 3.35
(s, 2 H, CH2S), 2.66–2.61 (m, 2 H, CH2–C=CH2), 2.33 (s, 3 H,
CH3–Ar), 1.61 (s, 3 H, CH3–C–Cl) ppm.
[11] Adduct 3a was prepared in 85% yield following the procedure
described in: W. A. Smit, I. P. Smolyakova, Izvestia Acad.
Nauk, Ser. Khim. 1985, 485–488.
[12] Compare with the data on the enhancement of catalytic activity
of TMSOTf in the presence of the organoaluminium Lewis
acid co-catalyst in: M. Oishi, S. Aratake, H. Yamamoto, J. Am.
Chem. Soc. 1998, 120, 8271–8272; see also the results presented
in ref.[10b]
.
[13] Attempts to use Et2AlCl or TMSOTf separately, or other Lewis
acids such as BF3·Et2O, BF3·Et2O + TMSOTf, Bu2BOTf, or
BBr3, in this reaction turned out to be fruitless.
[14] The formation of this product is evidently due to the attack of
a carbon nucleophile at the sulfur atom of the ESI intermedi-
ate; cf. a discussion in the review: G. H. Schmid, “Thiiranium
and Thiirenium Ions” (Eds.: A. Senning, P. S. Magee), George
Thieme Publishers, Stuttgart, 1977, vol. 3, chapter 3, pp. 103–
117.
[15] The adduct 11 was used as prepared in situ as its instability
precluded its isolation. The identity and purity of the adduct
was ascertained by consistent 1H NMR spectroscopic data (see
Exp. Sect.).
2-Methyl-1-(p-tolylsulfanyl)pent-4-en-2-ol (6b): Rf = 0.1 (ethyl ace-
tate/hexane, 10:1). 1H NMR (300 MHz, CDCl3): δ = 7.33–7.10 (m,
4 H, Ar), 5.90–5.71 (m, 1 H, CH2=CHCH2), 5.17–5.08 (m, 2 H,
CH2=CH), 3.12 (d, J = 12 Hz, 1 H, CH2S), 3.05 (d, J = 12 Hz, 1
H, CH2S), 2.43–2.27 (m, 2 H, CH2–C=CH2), 2.33 (s, 3 H, CH3–
Ar), 1.27 (s, 3 H, CH –C–OH) ppm. IR (CDCl ): ν = 3560 cm–1.
˜
3
3
[16] For a review, see: N. E. Shore, Org. React. 1991, 40, 1–90; for
more recent data, see: O. Geiss, H.-G. Schmaltz, Angew. Chem.
Int. Ed. 1998, 37, 911–914.
Acknowledgments
[17] E. Kuhle, Synthesis 1970, 561–580.
We are indebted to the Deutsche Forschungsgemeinschaft (Grant
No. MA673/19) and to the Russian Foundation of Basic Research
Received: September 9, 2005
Published Online: December 27, 2005
1322
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1317–1322