Different Reactivity Modes of Cis and Trans Di-tert-butylthiiranium Tetrafluoroborates with Water. A New Insight in the Electrophilic Additions to Z and E Di-tert-butylethylenes

Full text of DOI:10.1021/jo970024x

7018
J . Org. Chem. 1997, 62, 7018-7020
sulfonium tetrafluoroborate12 in liquid SO2 at -78 °C.
Differ en t Rea ctivity Mod es of Cis a n d
After evaporation of the solvent the salts have been
dissolved in CH2Cl2 and crystallized by addition of
pentane at -20 °C.
Tr a n s Di-ter t-bu tylth iir a n iu m
Tetr a flu or obor a tes1 w ith Wa ter . A New
In sigh t in th e Electr op h ilic Ad d ition s to Z
a n d E Di-ter t-bu tyleth ylen es†
Vittorio Lucchini,‡ Giorgio Modena,§ Manuela Pasi,§ and
Lucia Pasquato*,§
Centro CNR Meccansimi di Reazioni Organiche and
Dipartimento di Chimica Organica, Universita` di Padova,
via Marzolo 1, 35131 Padova, Italy, and Dipartimento di
Scienze Ambientali, Universita` di Venezia, Dorsoduro 2137,
30123 Venezia, Italy
Received J anuary 3, 1997
Our continuing interest in the chemistry of thiiranium
and thiirenium ions2-6 and in the electrophilic additions
to carbon-carbon double and triple bonds2 prompted us
to investigate the reactivity of stable thiiranium ions with
nucleophiles. Thiiranium and thiirenium ions (and the
selenium analogues7) are the most stable cyclic interme-
diates in the electrophilic addition to alkenes and alkynes,
and considering that the attack of a nucleophile to the
sulfonium sulfur of the thiiranium ion is the inverse
reaction of the electrophilic addition to the olefin, we may
acquire mechanistic information on this still unclear
process.2,8 Furthermore, in the last decade a growing
interest on the use of thiiranium ions as key intermedi-
ates in many processes with a high degree of stereochem-
ical control have been developed.9 Therefore studies on
the relative reactivities of nucleophiles at sulfur and
carbon atoms of thiiranium ions may be helpful in organic
synthesis.6,10
The cis isomer 1, treated with an excess (11 equiv) of
water in CD2Cl2 at 25 °C, undergoes a quantitative
formation of threo-2,2,5,5-tetramethyl-4-(methylthio)-3-
1
hexanol (5),13 as monitored by H NMR. No trace of the
Z di-tert-butylethylene 3, which would result from nu-
cleophilic attack of water on sulfur, could be detected.
The reaction is described by the first two equations of
Scheme 1.
In this paper we report the different reactivity modes
of cis and trans thiiranium tetrafluoroborates 1 and 2
toward the nucleophile water.11
Sch em e 1
Salts 1 and 2 are obtained from the reaction of the
appropriate olefins, 3 and 4, with methylbis(methylthio)-
† This paper is dedicated to Prof. George A. Olah in the occasion of
his 70th birthday.
‡ University of Venezia.
§ University of Padova.
(1) For easier reading we use the trivial names cis and trans di-
tert-butylthiiranium ions for the t-2,t-3- and c-2,t-3-di-tert-butyl-r-1-
methylthiiranium ions 1 and 2, respectively .
(2) Capozzi, G.; Modena, G.; Pasquato, L. The Chemistry of Sulfenic
Acids and Their Derivatives; Patai, S., Ed.; Wiley: Chichester, U.K.,
1990; Chapter 10.
(3) (a) Lucchini, V.; Modena, G.; Pasquato, L. J . Am. Chem. Soc.
1988, 110, 6900. (b) Lucchini, V.; Modena, G.; Pasquato, L. J . Am.
Chem. Soc. 1991, 113, 6600.
(4) Modena, G.; Pasquato, L.; Lucchini, V. Phosphorous, Sulfur,
Silicon 1994, 95-96, 265.
(5) Lucchini, V.; Modena, G.; Pasquato, L. J . Am. Chem. Soc. 1993,
115, 4527. Lucchini, V.; Modena, G.; Pasquato, L. J . Am. Chem. Soc.
1995, 117, 2297.
(6) Lucchini, V.; Modena, G.; Pasquato, L. J . Chem. Soc., Chem.
Commun. 1994, 1565.
(7) Back, T. G. The Chemistry of Organic Selenium and Tellurium
Compounds; Patai, S., Ed.; Wiley: Chichester, U.K., 1987; Vol 2,
Chapter 3.
(8) For bromination, see: Cossi, M.; Persico, M.; Tomasi, J . J . Am.
Chem. Soc. 1994, 116, 5373. Ruasse, M.-F. Adv. Phys. Org. Chem. 1993,
28, 207.
(9) Harring, S. R.; Edstrom, E. D.; Livinghouse, T. Adv. Heterocycl.
Nat. Prod. Synth. 1992, 2, 319. Rayner, C. M. Organosulfur Chemistry.
Synthetic Aspects; Page, P., Ed.; Academic Press: London, 1995;
Chapter 3.
(10) For selenium, see, for example: Toshimtsu, A.; Nakano, K.;
Mukai, T.; Tamao, K. J . Am. Chem. Soc. 1996, 118, 2756.
(11) We used the tetrafluoroborate anion because preliminary
experiments with the hexachloroantimonate ion showed a possible
attack of water on the anion.
Under the same conditions, but using 0.7 equiv of
water, the reaction is more complex, as revealed by
accurate analysis of the 1H NMR spectra taken at regular
time intervals. The decrease of the cis thiiranium ion 1
was monitored, accompanied by the formation of the
threo alcohol 5, the trans thiiranium ion 2, and the
thietanium ion 7.14 Also in this case no traces of the
olefin 3 could be detected. Considering the higher proton
activity with respect to the previous case, and in light of
(12) Capozzi, G.; DaCol, L.; Lucchini, V.; Modena, G. J . Chem. Soc.,
Perkin Trans. 2 1980, 68.
(13) Raynolds, P.; Zonnebelt, S.; Bakker, S.; Kellogg, R. M. J . Am.
Chem. Soc. 1974, 96, 3146.
(14) No traces of the E olefin 4, eventually formed from the trans
ion 2, generated in turn from the cis isomer 1, could be detected.
S0022-3263(97)00024-8 CCC: $14.00 © 1997 American Chemical Society

Notes
J . Org. Chem., Vol. 62, No. 20, 1997 7019
Ta ble 2. In itia l Con cen tr a tion s a n d Ra te Con sta n tsa for
th e Rea ction of Tr a n s Th iir a n iu m Ion 2 w ith Wa ter , in
CD2Cl2 a t 25 °C (cf. Sch em e 2)
the known rearrangement of thiiranium ion 2 to thieta-
nium ion 7,3 the reaction shown in Scheme 1 can be
suggested. The attack of water on a ring carbon atom of
the cis thiiranium ion 1 gives the threo alcohol 5+, in
equilibrium with its conjugate base 5; both species will
readily assume the most stable conformation (with an-
tiperiplanar tert-butyl groups) which prevents the an-
chimeric assistance by the nucleophilic methylthio group
to the detachment of the hydroxyl group. The threo
alcohol 5 undergoes the acid-catalyzed nucleophilic sub-
stitution by a water molecule, yielding the epimeric
erythro alcohol 6. This latter gives, in an acid-catalyzed
process, trans thiiranium ion 2, which then rearranges
to the stable thietanium ion 7.15
[2], M
[H2O], M
kS, M-1 s-1
kR, s-1
1
2
1.4 × 10-2 5.7 × 10-2 3.6((1.1) x 10-5 10.4((0.5) x 10-6
1.2 × 10-2 1.9 × 10-2 5.9((1.0) x 10-5 11.2((0.8) x 10-6
a
Average and statndard deviation from three determinations.
in Table 2. The rate constants kR are in good agreement
with those measured in the case of the reactions of the
cis thiiranium ion 1 (Table 1).
A further point deserves a comment. No traces of E
olefin 4 could be detected in the reaction of thiiranium
ion 1, which also generates thiiranium ion 2, although
the reaction of authentic 2 gives rise to a small amount
of the same olefin. We propose the following rationale.
The attack of water to the ring carbons of 1 is about 2
orders of magnitude faster that the attack to sulfonium
sulfur of 2. Thus, during the greatest part of the reac-
tion, water is effectively sequestered by 1 in the form of
threo alcohol 5 or is deactivated as nucleophile by
protonation so that the attack to the newly formed
thiiranium ion 2 is minimized. The E olefin 4 might be
generated, but at such a low concentration to escape
detection.
The proposed mechanism is supported by an indepen-
dent experiment. The addition of CF3SO3H to a solution
of the threo alcohol 5 in CD2Cl2, obtained using an excess
of water, shows indeed that alcohol 5 is converted to trans
thiiranium ion 2 and, subsequently, to thietanium ion
7.
As the key point in the reaction Scheme 1 is the acid-
catalyzed interconversion of alcohols 5 and 6, the behav-
ior of cis thiiranium ion 1 with a large excess of water
may be understood: under these conditions the generated
threo alcohol 5 is not effectively protonated.
The differential equations related to the consecutive
and competitive reactions shown in Scheme 1 have been
numerically integrated16 and fitted (with a Simplex
procedure17) to the concentrations of 1, 2, and 7 and the
sum of the concentration of 5 and 5+. The most relevant
rate constants kC, describing the consumption of cis
thiiranium ion 1 by reaction with water, and kR, for the
rearrangement of the trans thiiranium ion 2 to thieta-
nium ion 7, are reported in Table 1.
The results herein reported outline the different reac-
tivity modes of cis and trans thiiranium ions 1 and 2,
even though the study is limited to only one nucleophile
and to a small range of concentrations. The reactivity
modes seem to be highly specific, and within the detection
1
limits of H NMR spectroscopy, the competitive attacks
at sulfur in cis thiiranium ion 1 and at the ring carbon
in trans thiiranium ion 2 may be considered at least 2
orders of magnitude slower than the observed ones.
Ta ble 1. In itia l Con cen tr a tion s a n d Ra te Con sta n tsa for
th e Rea ction of Cis Th iir a n iu m Ion 1 w ith Wa ter , in
CD2Cl2 a t 25 °C (cf Sch em e 1)
If the different behavior of thiiranium ions 1 and 2
toward the nucleophile water holds for other nucleo-
philes, and particularly for halide ions, then these results
call for a reconsideration in interpreting the reactivities
of Z and E di-tert-butylethylene 3 and 4 toward sulfenyl
chlorides and other electrophiles. The relevant rate ratio
of 1.6 × 105 measured for the addition of 4-chloroben-
zenesulfenyl chloride to isomers 3 and 418 is to be
compared with the much reduced rate ratios of 13.6 for
the addition of bromine19 and of 0.37 for the addition of
chlorine.20
-1
[1], M
[H2O], M
kC, M-1 s
kR, s-1
1
2
7.5 × 10-3 8.4 × 10-2
9.6((0.4) x 10-4
6.6 × 10-3 4.7 × 10-3 11.9((1.5) x 10-4 8.9((0.5) x 10-6
a
Average and standard deviation from three determinations.
Sch em e 2
The huge reactivity ratio for the addition of the sulfenyl
chloride was explained by invoking the spatial charac-
teristic of the olefins, i.e., by the greater steric hindrance
opposed to the approaching electrophile by two trans tert-
butyl groups than by two cis group.18 This rationalization
cannot account for the smaller reactivity ratios for the
addition of the halogens. Rather, the different spatial
characteristics of the electrophiles sulfenyl chloride and
bromine have been invoked.19 In thiiranium ions 10
(Scheme 3) the S-R′ bond is nearly perpendicular to the
three-membered ring;21 if this orientation is already
present in the transition state, then a nonbonding
interaction between R′ of the sulfenyl chloride 9 and the
cis-oriented tert-butyl group may lower the addition rate
to olefin 4.
Under the same reaction conditions we monitored the
behavior of trans thiiranium ion 2 in the presence of
water. We could follow the formation of thietanium ion
7 and of different amounts of the E di-tert-butylethylene
4, depending on water concentration. No trace of erythro
alcohol 6 could be observed. The reaction is therefore
described by Scheme 2. The related differential equa-
tions, integrated and fitted to the concentrations of 2, 4,
and 7, give the selected rate constants kS and kR reported
(15) This hypothesis is in agreement with the formation of thiira-
nium ion 2 from erythro-2,2,5,5-tetramethyl-4-(methylthio)-3-chloro-
hexane in liquid SO2, as reported in ref 3.
(16) Press, W. H.; Flannery, B. P.; Teukolsky, S. A.; Vetterling, W.
T. Numerical Recipes. The Art of Scientific Computing; Cambridge
University Press: Cambridge, U.K., 1986; p 547.
(17) Nash, J . C. Compact Numerical Methods for Computers; Adam
Hilger Ltd.: Bristol, U.K., 1979; p 141.
(18) Dean, C. L.; Garratt, D. G.; Tidwell, T. T.; Schmid, G. H. J .
Am. Chem. Soc. 1974, 96, 4958.
(19) Ruasse, M. F., Argile, A., Bienvenue-Goe¨tz, E., Dubois, J . E. J .
Org. Chem. 1979, 44, 2758.
(20) Fahey, R. C. J . Am. Chem. Soc. 1966, 88, 4681.

7020 J . Org. Chem., Vol. 62, No. 20, 1997
Notes
Sch em e 3
standard chemical suppliers or prepared according to literature
procedures and purified to match the reported physical and
spectral data. Solvents were purified according to standard
procedures.
The different steric hindrances presented by alkenes
3 and 4 to the approaching electrophile may contribute
to the reactivity ratios, but only as a minor factor. Indeed
our results suggest an alternative rationalization. Con-
sidering the general reaction scheme for the addition of
sulfenyl chlorides to alkenes (Scheme 3) and assuming
that the steady state approximation can be applied to
the intermediates thiiranium ion 10 and chloride ion,
which have never been observed in reactions carried out
in chlorinated solvents,2 eq 1 expresses the rate of
product formation. In the case of the cis thiiranium ion
1, kC > kS and eq 1 simplifies to eq 2, while in the case of
the trans isomer 2, kC < kS and eq 1 becomes eq 3.
Kin etic Mea su r em en ts. The decrease in the concentrations
of cis and of trans thiiranium ions 1 and 2 and the formation of
cumulated threo-5 and threo-5+ alcohols were followed by
measuring the 1H NMR integrated area of the S-methyl reso-
nances; the formation of the thietanium ion 7 was followed by
measuring the integrated area of the two singlets of the geminal
methyl groups in the reactions of cis-1 and the doublet of the
methylene proton geminal to the tert-butyl in the reaction of
trans-2; the formation of E alkene was followed by monitoring
the integrated area of the tert-butyl resonance. The initial
concentrations of thiiranium ions 1 and 2 have been determined
by weighing. These concentrations and the concentration of
water in CD2Cl2 have been verified by comparison of the
integrated areas of selected resonances with that of the proton
impurity of the solvent. The concentration of the impurity has
been previously measured by comparison with the signal of 1,4-
dinitrobenzene, at a concentration determined by weighing. The
reactions have been monitored, at appropriate intervals of time,
using NMR tubes equipped with air-tight screw caps. In order
to compensate for the varying spectrometer conditions, the
monitored intensities were normalized against their sum. The
differential equations relative to Schemes 1 and 2 have been
numerically integrated16 and fitted to the normalized concentra-
tions with the Simplex procedure.17
t-2,t-3-Di-ter t-b u t yl-r -1-m et h ylt h iir a n iu m Tet r a flu or o-
bor a te (1). In a 25 mL three necked flask containing methyl-
bis(methylthio)sulfonium tetrafluoroborate12 (0.155 g, 0.68 mmol),
SO2 was condensed to 10 mL at -78 °C under nitrogen
atmosphere and the Z alkene 33b (0.095 g, 0.68 mmol), dissolved
in 1 mL of dry methylene chloride, was added in one portion.
The reaction mixture was left 40 min at -78 °C, and then, after
addition of 5 mL of dry methylene chloride, the SO2 was allowed
to evaporate. The salt was precipated with pentane and
recrystallized from methylene chloride/pentane at -20 °C in 63%
yield. 1H NMR (200 MHz, CD2Cl2): δ 1.32 (9 H, s), 2.77 (3 H,
s), 4.44 (2 H, s). 13C (62.9 MHz, CD2Cl2): δ 24.22, 29.97, 33.71,
78.12. Anal. Calcd for C11H23BF4S: C, 48.19; H, 8.45. Found:
C, 48.04; H, 8.38.
In this second instance the observed reaction rate is
reduced by the presence of a high value of kS at the
denominator. It may then be proposed that the low rate
for the addition of sulfenyl chlorides to the E alkene 4 is
mainly due to the reversibility of the first step of the
reaction, while the equivalent step for the addition to the
Z isomer 3 is substantially irreversible. The rational for
this different behavior is straightforward: the presence
of the substituent at sulfur in thiiranium ions constrains
the approach of the nucleophile Cl- to sulfur from a
direction which is strongly hindered in cis thiiranium ion
1.
The difference in hindrance cannot explain the very
reduced rate ratios measured for the additions of bromine
or chlorine to 3 and 4. Rather, the addition rates are
similar because the first steps for the additions are
equally reversible: the approach of the nucleophile Br-
or Cl- to the halogen center in cis and trans haloiranium
ions 11 and 12, the proposed intermediates, is hindered
to the same extent in both isomers, or even more
hindered in the trans isomer.
c-2,t-3-Di-ter t-bu tyl-r -1-m eth ylth iir a n iu m Tetr a flu or o-
bor a te (2). This salt was prepared as described for isomer 1
using the E alkene 43b and crystallized from methylene chloride/
pentane at -20 °C, 68% yield. 1H NMR (200 MHz, CD2Cl2): δ
1.14 (9 H, s), 1.32 (9 H, s), 2.80 (3 H, s), 3.65 (1 H, d, J ) 13.4
Hz), 4.38 (1 H, d, J ) 13.4 Hz). 13C (62.9 MHz, CD2Cl2): δ 17.86,
26.76, 28.77, 33.09, 34.72, 68.44, 73.34. Anal. Calcd for C11
23BF4S: C, 48.19; H, 8.45. Found: C, 48.08; H, 8.40.
t-4-ter t-Bu tyl-r -1,2,2,c-3-tetr a m eth ylth ieta n iu m tetr a flu -
-
H
or obor a te (7)3b was obtained from the rearrangement of 2 in
methylene chloride at 25 °C and recrystallized from methylene
chloride/pentane at -20 °C. 1H NMR (250 MHz, CD2Cl2): δ 1.08
(9 H, s), 1.23 (3 H, d, J ) 6.8 Hz), 1.72 (3 H, s), 1.74 (3 H, s),
3.00 (3 H, s), 3.03 (1 H, dq, J ) 11.3, 6.8 Hz), 3.91 (1 H, d, J )
11.3 Hz). 13C (62.9 MHz, CD2Cl2): δ 15.87, 18.04, 22.76, 26.56,
28.92, 33.64, 44.29, 65.32, 72.49. Anal. Calcd for C11H23BF4S:
C, 48.19; H, 8.45. Found: C, 48.12; H, 8.35.
Exp er im en ta l Section
Gen er a l. Melting points are uncorrected. NMR spectra were
recorded at 200 or 250 MHz. Commercial reagents and known
compounds used in this research were either purchased from
(21) Huang, X.; Batchelor, R. J .; Einstein, F. W. B.; Bennet, A. J . J .
Org. Chem. 1994, 59, 7108.
J O970024X