Kinetics of the [2+ + 4]-cycloaddition reactions of 1,3-dithian-2-ylium ions with 1,3-dienes

Article abstract of DOI:10.1002/(sici)1099-0690(199809)1998:9<1919::aid-ejoc1919>3.0.co;2-e

The kinetics of the [2⁺ + 4] cycloadditions of 1,3-dithian-2-yhum ions (1) with 1,3-dienes was investigated photometrically in dichloromethane. The second-order rate constants determined for the reactions of 1 with 2,3-dimethyl-1,3-butadiene (2a) and isoprene (2b) were identical to those calculated for the first step of a stepwise cycloaddition pathway by the correlation Ig k = s (E + N). Though a concerted cycloaddition pathway is not excluded by this finding, it is obvious that the transition states of these reactions are not noticeably stabilized by the simultaneous formation of two new σ bonds.

Full text of DOI:10.1002/(sici)1099-0690(199809)1998:9<1919::aid-ejoc1919>3.0.co;2-e

FULL PAPER
؉
Kinetics of the [2 ؉ 4]-Cycloaddition Reactions of 1,3-Dithian-2-ylium Ions
with 1,3-Dienes
Herbert Mayr* and Joachim Henninger
Institut für Organische Chemie, Ludwig-Maximilians-Universität München,
Karlstraße 23, D-80333 München, Germany
Fax: (internat.) ϩ 49(0)89/5902254
E-mail: hmy@org.chemie.uni-muenchen.de
Received April 2, 1998
Keywords: Heterocycles / Carbocations / Dienes / Cycloadditions / Kinetics
The kinetics of the [2+ + 4] cycloadditions of 1,3-dithian-2-
cycloaddition pathway by the correlation lg k = s (E + N).
ylium ions (1) with 1,3-dienes was investigated
Though a concerted cycloaddition pathway is not excluded
photometrically in dichloromethane. The second-order rate by this finding, it is obvious that the transition states of these
constants determined for the reactions of 1 with 2,3-
dimethyl-1,3-butadiene (2a) and isoprene (2b) were identical
reactions are not noticeably stabilized by the simultaneous
formation of two new σ bonds.
to those calculated for the first step of
a stepwise
Introduction
As discussed previously, Equation 1 holds for reactions
of cationic electrophiles with neutral nucleophiles when one
1
,3-Dithian-2-ylium ions are useful acylium equivalents.
[2]
new bond is formed in the rate-determining step . Pe-
ricyclic reactions in which the transition state profits from
the simultaneous formation of more than one new bond,
can be expected to proceed faster than predicted by Equa-
Their synthetic potential has been specified by the determi-
[1]
nation of their electrophilicity parameters E , which can
be used to estimate the rate constants k of their reactions
with uncharged nucleophiles (with known nucleophilicity
parameter N and slope parameter s)[ by Equation 1
[
5]
2]
tion 1 .
Corey and Walinsky[6] described hetero DielsϪAlder re-
(
Table 1).
Ϫ
actions of 1,3-dithian-2-ylium tetrafluoroborate (1ϪBF4
)
with 1,3-dienes (2). We have now determined the kinetics of
these reactions and compared the rate constants with those
calculated from Equation 1 in order to obtain information
on the concertedness of these cycloadditions.
lg k ϭ s (E ϩ N)
(1)
Table 1. Comparison of calculated and observed rate constants
for the reactions of the 1,3-dithian-2-ylium tetrafluoroborate
Ϫ
4
(1ϪBF ) with nucleophiles at 20°C
Results
Suspensions of the 1,3-dithian-2-ylium tetrafluoroborate
Ϫ
(
1ϪBF4 ) in dichloromethane react with 2,3-dimethyl-1,3-
butadiene (2a), isoprene (2b), and 1,3-butadiene (2c) to give
the 3,4,8,8a-tetrahydro-2H,5H-1-thia-4a-thionianaphthalene
Ϫ
tetrafluoroborates (3ϪBF4 , Scheme 1). Since the spectral
data of these compounds have not been published by Corey
nucleophile
N[a]
s[a]
k
calcd. (Eq.1)
k
obsd.
Ϫ1 Ϫ1
[l mol sϪ1
]
Ϫ1
[
l mol
s
]
Scheme 1. Reactions of the 1,3-dithian-2-ylium tetrafluoroborate
Ϫ
2.69 ϫ 102
1.82 ϫ 102[b]
4
(1ϪBF ) with the 1,3-dienes 2aϪc
4
3
1
.90
.68
.62
0.89
0.92
2.45 ϫ 101
2.57 ϫ 101[b]
1.01
0.72
2.78 ϫ 10Ϫ1
7.56 ϫ 100
4.04 ϫ 10Ϫ1[b]
1.07 ϫ 100[b]
2
HSiMe Ph
3.39
.5[c]
0
1.0[c]
2.1 ϫ 10Ϫ2
2.1 ϫ 10Ϫ2[d]
diene
R
RЈ
T [°C]
time
yield (%)
2
a
Me
H
H
Me
Me
H
25
15 min
60 min
6 d
82
64
84
[
a]
Ref. . Ϫ Ref.[1]. Ϫ Ref.[3]. Ϫ [d] The measured rate constant 2b
[2]
[b]
[c]
25
Ϫ2
Ϫ1 Ϫ1
[4]
k ϭ 3.8 ϫ 10 l mol
2
s
(CD
3
CN, 28°C) was converted to 2c
Ϫ17 to 25
Ϫ1
Ϫ1
0°C by assuming an activation entropy of Ϫ100 J mol
K .
Eur. J. Org. Chem. 1998, 1919Ϫ1922

WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ193X/98/0909Ϫ1919 $ 17.50ϩ.50/0
1919

H. Mayr, J. Henninger
FULL PAPER
Table 2. 13C-NMR chemical shifts of compounds (3aϪc)ϪBF Ϫ CN (75.5 MHz)[a]
in CD
4
3
compound
C-2
C-3
C-4
C-5
C-6
C-7
C-8
C-8a
CH
3
33.60[
31.62
31.61
b]
115.86
109.60
115.67
127.19
135.65
127.12
33.46
31.87
27.39
[b]
43.69
43.99
43.80
19.39, 20.03
3a
3b
3c
25.40
25.19
25.26
21.19
21.84
21.84
34.57
33.19
33.60
24.20
Ϫ
[
a]
Assignments corroborated by (H,H)-COSY-45 and INV-(C,H)-correlated 2D-NMR experiments. Ϫ [b] Assignments may be reverse.
[
6]
and Walinsky , a complete characterization of the com- ceed by attack of 1 at the s-trans conformers of 2aϪc to
Ϫ
pounds 3ϪBF4 will be given in this article (Table 2 and yield the allyl cations 4aϪc.
1
Experimental). Though complex H-NMR resonances for
Since attempts to optimize the structures 4aϪc with the
all ring hydrogen atoms can be expected, triplets were ob- AM1 method sometimes immediately yielded the bicyclic
served for 8a-H in all compounds 3 and for 4-H in 3a and cations 6aϪc, one can conclude that the barrier for the
3
b.
cyclization 4 Ǟ 6 is negligible. Rotation of a CϪC single
The DielsϪAlder adducts 3 do not show any absorption bond may yield 7aϪc, which may undergo a [1,3]-shift of
at λmax of the 1,3-dithian-2-ylium ion (1) (311 nm). It was sulfur to give the cycloadducts 3aϪc probably via the allyl
possible, therefore, to monitor the rates of the reactions by cations 5aϪc.
UV spectroscopy using a workstation as described pre-
[
7]
Scheme 2. Concerted and stepwise cycloaddition of the 1,3-di-
viously . In analogy to other reactions of carbocations
with π nucleophiles, the reactions of the carbenium ion 1
with 2,3-dimethyl-1,3-butadiene (2a) and isoprene (2b) fol-
low second-order kinetics, first-order with respect to the nu-
cleophile and first-order with respect to the carbocation
f
thian-2-ylium cation (1) with the 1,3-dienes 2aϪc (∆H °
Ϫ1
[kJ mol ] calculated by the semiempirical AM1 me-
thod)[
8]
(Table 3).
The second-order rate law is not exactly obeyed in the
reaction with 1,3-butadiene (2c). Since the formal second-
order rate constant appears to be larger in the initial phase
of the reaction than at higher degree of conversion, one can
only determine an approximate value of k (Table 3).
Table 3. Kinetics of the reactions of the 1,3-dithian-2-ylium ion (1)
2 2
with 1,3-butadienes (2) in CH Cl
[
c]
diene
T
k
obsd.
k
calcd.
[l mol sϪ1
Ϫ1
]
(1 ϩ 2 Ǟ 4)
1.6 ϫ 10Ϫ1
[
°C]
2a
20
(3.10 ± 0.12) ϫ 10Ϫ1
2
b
c
20
0
(1.23 ± 0.05) ϫ 10Ϫ1 [a] 9.4 ϫ 10Ϫ2
(2 to 13) ϫ 10Ϫ3 [b] 7.4 ϫ 10Ϫ5[d]
2
[
a]
°
Ϫ1
°
Ϫ1
∆
H ϭ 46.32 ± 0.81 kJ mol ; ∆S ϭ Ϫ104.2 ± 2.9 J mol
[b]
Ϫ1
K
. Ϫ
The second-order rate law is not exactly obeyed (see
From Eq. (1) with E(1) ϭ Ϫ2.17 , N(2a) ϭ 1.37,
[
c]
[1]
text). Ϫ
[2]
s(2a) ഠ 1, N(2b) ϭ 1.12, s(2b) ϭ 0.98, N(2c) ϭ Ϫ1.15, s(2c) ഠ 1
.
[
d]
Ϫ4
Ϫ1 Ϫ1
Ϫ
∆
From kcalcd. (20°C) ϭ 4.79 ϫ 10 l mol
ϭ Ϫ104 J mol
s
with estimated
°
Ϫ1
Ϫ1
S
K .
Discussion
For the DielsϪAlder reactions of 1 with the butadienes
Because of the favorable pathway 4 Ǟ 6 Ǟ 7 Ǟ 5, the
Ϫ1
2
, which are exothermic by Ϫ133.6 to Ϫ140.9 kJ mol
direct isomerizations of 4 to 5, for which rotational barriers
[8]
Ϫ1
according to AM1 calculations (Scheme 2), a concerted of 45.6 (4a), 53.4 (4b), and 78.5 (4c) kJ mol can be as-
[9]
or a stepwise mechanism has to be considered. While the sumed (Eq. 19 in ref. ), do not play a role.
less stable s-cis conformer of the dienes 2 [the calculated
These considerations which are summarized in the free-
higher stability of s-cis compared to s-trans-2a (Scheme 2) enthalpy profile of Scheme 3 suggest that in case of a step-
is an artifact of the AM1 method] is required for the con- wise cycloaddition the formation of the allyl cation 4 can
certed [4 ϩ 2] cycloaddition, the stepwise process can pro- be assumed to be rate-determining. Since the formation of
1920
Eur. J. Org. Chem. 1998, 1919Ϫ1922

Kinetics of the [2ϩ ϩ 4]-Cycloaddition Reactions of 1,3-Dithian-2-ylium Ions with 1,3-Dienes
FULL PAPER
4
from 1 and 2 is mechanistically analogous to the rate-
The observation that in none of the cycloadditions stud-
determining step of the reactions described in Table 1, ied in this work a large difference between kobsd. and kcalcd.
Equation 1 should be applicable to calculate its rate.
was observed, implies that Equation 1, which has been pa-
rametrized for reactions of carbocations with non-charged
nucleophiles, may also be used to estimate the rates of the
Scheme 3. Free-enthalpy profile (20°C) for the stepwise DielsϪAl-
der reaction of the 1,3-dithian-2-ylium cation (1) with
isoprene (2b)
ϩ
[2 ϩ 4] cycloadditions of 1 with conjugated dienes. It is
thus for the first time possible to predict absolute rate con-
stants of such cycloadditions with reasonable accuracy.
Generous support by the Deutsche Forschungsgemeinschaft and
the Fonds der Chemischen Industrie is gratefully acknowledged. We
thank Nathalie Hampel for experimental assistance.
Experimental Section
1
General: NMR: Bruker WM 300 or Bruker ARX 300. H NMR
1
3
(
(
300 MHz) and C NMR (75.5 MHz) refer to [D
3
]acetonitrile
δ
H
ϭ 1.93 and δ ϭ 1.30) as internal standard. Ϫ IR: IR spectro-
C
photometer 325 (PerkinϪElmer). Ϫ MS: Varian MAT 311A. Ϫ
Melting points (uncorrected): Büchi 530. Ϫ All reactions were car-
ried out under dry nitrogen. Dichloromethane was freshly distilled
from CaH
2
before use. The 1,3-butadienes 2aϪc are commercially
[
a]
From MopacϪThermo (AM1) calculations[10]: ∆G°(20°C)/kJ
Ϫ1
available. 2a,b were freshly distilled before use. 1,3-Dithian-2-ylium
mol ϭ 580.8 (1), 4.0 (2b), 501.5 (3b), 655.2 (4b), 658.2 (5b), 598.3
[
b]
Ϫ
(
6b), 600.5 (7b). Ϫ
Barrier derived from Eq. 1 (see Table 3). Ϫ tetrafluoroborate (1ϪBF ) was prepared from 1,3-propanedithiol,
4
[
c]
[d]
[12]
Unknown barrier assumed to be very small (see text). Ϫ Bar-
ethyl orthoformate and HBF as described in ref.
.
4
[11]
[e]
rier of interconversion of 1-butene conformers . Ϫ Rotational
barrier of the 1,1-dimethylallyl cation in superacid solution (Eq. 19
General Procedure: The 1,3-butadienes 2 were added dropwise to
(or condensed into) a suspension of 1,3-dithian-2-ylium tetra-
fluoroborate (1ϪBF
until the solid component had disappeared (except for the reaction
with 2c because 3c is also poorly soluble in CH Cl ). The solvent
[
9]
in ref. ).
Ϫ
4 2 2
) in dry CH Cl . The mixture was stirred
According to Table 3, the observed cycloaddition rate
constants for 2,3-dimethyl-butadiene (2a) and isoprene (2b)
and those calculated for the rate-determining steps of the
stepwise processes are almost identical. This coincidence
suggests stepwise cycloaddition mechanisms. Though one
cannot exclude concerted pathways on this basis, it is obvi-
ous that the transition states are not noticeably stabilized
by the simultaneous formation of two new σ bonds.
2
2
and the excess of the dienes were removed in vacuo to yield the
solid products which were purified by recrystallization.
6
,7-Dimethyl-3,4,8,8a-tetrahydro-2H,5H-1-thia-4a-thionianaph-
Ϫ
Ϫ
thalene Tetrafluoroborate (3aϪBF
4
): Reaction of 1ϪBF
4
(0.13 g,
0
.63 mmol) with 2a (0.28 g, 3.4 mmol) at ambient temperature for
The cycloaddition of 1 with 1,3-butadiene (2c) was ob-
served to be 10 times faster than calculated for the forma-
tion of the intermediate 4c. Though a small degree of con-
2
Table 4. Kinetics of the reactions of 1,3-dithian-2-ylium tetrafluo-
Ϫ
roborate (1ϪBF
4
) with 2,3-dimethyl-1,3-butadiene (2a)
certedness is suggested by this deviation of kobsd. and kcalcd.
,
[a]
T
[1]
o
[2a]
o
conversion k
[
°C]
[mol lϪ1]
[mol lϪ1]
(%)
[l molϪ1 sϪ1
]
the poor quality of the observed rate constants prevents
definite conclusions.
Ϫ4
Ϫ4
Ϫ4
Ϫ4
Ϫ2
Ϫ3
Ϫ2
Ϫ3
Ϫ1
Ϫ1
Ϫ1
Ϫ1
2
1
2
2
0.0
9.9
0.0
0.0
2.82 ϫ 10
2.49 ϫ 10
2.52 ϫ 10
2.81 ϫ 10
1.46 ϫ 10
7.70 ϫ 10
1.04 ϫ 10
5.70 ϫ 10
51
82
70
71
3.25 ϫ 10
3.03 ϫ 10
3.18 ϫ 10
2.94 ϫ 10
The failure to unequivocally assign the mechanism of
these reactions illuminates an important feature of this
mechanistic tool. In principle, the comparison of observed
cycloaddition rate constants with those calculated for step-
wise processes by Equation 1 provides a means to estimate
the degree of concertedness, i.e., the energetic preference of
[a]
k(20°C) ϭ (3.10 ± 0.12) ϫ 10Ϫ1 l molϪ1 sϪ1
.
the concerted over a (hypothetical) stepwise process. How- Table 5. Kinetics of the reactions of 1,3-dithian-2-ylium tetrafluo-
Ϫ
roborate (1ϪBF
4
) with isoprene (2b)
ever, in this way, we can neither differentiate borderline
cases nor exclude concerted mechanisms with highly un-
symmetrical transition states. It should furthermore be
noted that this method of determining the degree of con-
certedness is not applicable to cycloadditions of two non- 20.0
charged components, since one cannot presently predict the
rates of formation of dipolar or diradical intermediates
a]
T
[1]
o
[2b]
o
conversion k[
[mol lϪ1]
[mol lϪ1]
[l molϪ1 sϪ1
]
[°C]
(%)
Ϫ4
Ϫ4
Ϫ4
Ϫ4
Ϫ4
Ϫ4
Ϫ4
Ϫ2
Ϫ2
Ϫ2
Ϫ3
Ϫ2
Ϫ2
Ϫ2
Ϫ1
Ϫ1
Ϫ1
Ϫ1
Ϫ2
Ϫ2
Ϫ3
2.33 ϫ 10
1.91 ϫ 10
2.06 ϫ 10
3.80 ϫ 10
1.90 ϫ 10
2.29 ϫ 10
4.31 ϫ 10
1.45 ϫ 10
1.98 ϫ 10
1.09 ϫ 10
4.03 ϫ 10
1.95 ϫ 10
2.66 ϫ 10
5.01 ϫ 10
90
80
83
83
84
87
52
1.19 ϫ 10
1.19 ϫ 10
1.25 ϫ 10
1.23 ϫ 10
2.97 ϫ 10
1.42 ϫ 10
2.22 ϫ 10
2
1
2
0.0
9.9
0.0
which might be involved in such reactions. In cycload- 0.0
Ϫ9.9
Ϫ28.5
ditions of closed-shell ions with neutral reaction partners,
on the other hand, the potential intermediates are closed-
shell ions like one of the reactants and like the cycloadduct,
which greatly facilitates the mechanistic analysis.
[a]
Ϫ1
Ϫ1 Ϫ1
°
k(20°C) ϭ (1.23 ± 0.05) ϫ 10 l mol
s
; ∆H ϭ 46.32 ±
Ϫ1
°
Ϫ1
Ϫ1
0.81 kJ mol ; ∆S ϭ Ϫ104.23 ± 2.92 J mol
K .
Eur. J. Org. Chem. 1998, 1919Ϫ1922
1921

H. Mayr, J. Henninger
4
) with 1,3-butadiene (2c)
FULL PAPER
Ϫ
Table 6. Kinetics of the reactions of 1,3-dithian-2-ylium tetrafluoroborate (1ϪBF
[
a]
[a]
T
[1]
o
[2c]
o
conversion 1
(%)
k
max
conversion 2
(%)
k
min
Ϫ1
[l mol sϪ1
]
Ϫ1
[
°C]
[mol lϪ1]
[mol lϪ1]
[l mol sϪ1
]
Ϫ4
Ϫ4
Ϫ4
Ϫ4
Ϫ4
Ϫ2
Ϫ2
Ϫ2
Ϫ3
Ϫ2
Ϫ2
Ϫ3
Ϫ3
Ϫ3
Ϫ3
Ϫ3
Ϫ3
Ϫ3
Ϫ3
Ϫ3
Ϫ0.1
6.82 ϫ 10
3.04 ϫ 10
5.90 ϫ 10
4.10 ϫ 10
1.84 ϫ 10
1.41 ϫ 10
1.26 ϫ 10
2.44 ϫ 10
9.86 ϫ 10
3.55 ϫ 10
0 to 5
0 to 7
0 to 7
0 to 4
0 to 11
1.27 ϫ 10
7.14 ϫ 10
3.28 ϫ 10
4.19 ϫ 10
8.45 ϫ 10
21 to 50
9 to 43
7 to 56
11 to 33
41 to 81
3.76 ϫ 10
3.74 ϫ 10
1.91 ϫ 10
2.20 ϫ 10
1.91 ϫ 10
0
0
0
0
.1
.1
.4
.1
[
a]
k(0°C) ϭ (2 to 13) ϫ 10Ϫ3 l molϪ1 Ϫ1; the second-order rate law is not exactly obeyed in this reaction; kmax is determined from the
s
regression line in the initial phase of the reaction (range evaluated for the determination of kmax: conversion 1), kmin is determined from
the regression line in a latter phase of the reaction (range evaluated for the determination of kmin: conversion 2).
1
5 min and recrystallization of the product from butanone/diethyl 2. Ϫ MS (FD, 0Ϫ15 mA); m/z (%): 435, 434, 433, 432 (1, 1, 11, 2)
ϩ
Ϫ
ϩ
Ϫ
ether yielded 0.15 g (0.52 mmol, 83%) of 3a as colorless needles
[2 M Ϫ BF
4 4
], 175, 174, 173 (10, 27, 100) [M Ϫ BF ]. Ϫ
with m.p. 79.5Ϫ81°C (ref.[ : 96%, 85Ϫ86°C). Ϫ H NMR
6]
1
C H13BF S (260.1): calcd. C 36.94, H 5.04; found C 36.98, H 4.98.
8 4 2
(
2
3
CD
3
CN, 300 MHz): δ ϭ 1.72, 1.76 (2 s, 2 ϫ 3 H, 6-CH
3
, 7-CH
3
),
),
Kinetic Investigations: The reactions were monitored with a
.05Ϫ2.35 (m, 2 H, 3-H ), 2.62Ϫ3.01 (m, 4 H, 8-H and 2-H
.40 (t, J ϭ 5.9 Hz, 2 H, 4-H
2
2
2
[7]
workstation as described in ref. . The spectrometric measurements
were performed with a fiber optic system: The light of a deuterium
lamp (CLD 300, Fa. Zeiss) was conducted in a quartz fiber to a
probe (Hellma GmbH and Co. Müllheim) and was led back to the
spectrometer (MCS 220, Fa. LOT).
3
2
2
), 3.73, 3.94 (2 d, J ϭ 16.8 Hz, 2
), 4.80 (t, J ϭ 6.2 Hz, 1 H, 8a-H). Ϫ 13C NMR: Table 2.
2
3
H, 5-H
Ϫ MS (FD, 0Ϫ12 mA); m/z (%): 204, 203, 202, 201 (2, 12, 11, 100)
ϩ
Ϫ
[M
Ϫ BF
4 4 2
]. Ϫ C10H17BF S (288.2): calcd. C 41.68, H 5.95;
found C 41.84, H 5.95.
7
-Methyl-3,4,8,8a-tetrahydro-2H,5H-1-thia-4a-thionianaphthalene
[1]
H. Mayr, J. Henninger, T. Siegmund, Res. Chem. Intermed.
Ϫ
Ϫ
Tetrafluoroborate (3bϪBF
4
): Reaction of 1ϪBF
4
(0.41 g, 2.0
1996, 22, 821Ϫ838.
[2] [2a]
H. Mayr, M. Patz, Angew. Chem. 1994, 106, 990Ϫ1010; An-
mmol) with 2b (0.27 g, 4.0 mmol) at ambient temperature for 1 h
and recrystallization of the product from butanone/ethyl acetate
yielded 0.35 g (1.3 mmol, 64%) of 3b as colorless prisms with m.p.
[
2b]
gew. Chem. Int. Ed. Engl. 1994, 33, 938Ϫ957. Ϫ
H. Mayr,
O. Kuhn, M. F. Gotta, M. Patz, J. Phys. Org. Chem., in press.
[3] [3a]
H. Mayr, G. Lang, M. Roth, M. Patz, Polym. Prepr. 1996,
[
6]
1
[3b]
1
34Ϫ135°C (ref. : 95%, 131Ϫ132°C). Ϫ H NMR (CD
MHz): δ ϭ 1.79 (m , 3 H, 7-CH ), 2.07Ϫ2.36 (m, 2 H, 3-H
.63Ϫ2.92 (m, partly superimposed, 2 H, 8-H ), 2.83Ϫ3.04 (m,
partly superimposed, 2 H, 2-H ), 3.27Ϫ3.45 (m, 2 H, 4-H ),
.81Ϫ3.98 (m, 2 H, 5-H ), 4.82 (t, J ϭ 6.4 Hz, 1 H, 8a-H), 5.49
, 1 H, 6-H). Ϫ 13C NMR: Table 2. Ϫ IR (KBr): ν˜ ϭ 1680 cm
3
CN, 300
37, 353Ϫ354. Ϫ
H. Mayr, G. Lang, M. Roth in Cationic
Polymerisation Ϫ Fundamentals and Applications, ACS Sym-
posium Series 665 (Ed.: R. Faust, T. D. Shaffer), American
Chemical Society, Washington, 1997, pp. 25Ϫ40.
I. Stahl, Chem. Ber. 1985, 118, 4857Ϫ4868.
M. Hartnagel, K. Grimm, H. Mayr, Liebigs Ann. 1997, 71Ϫ80.
E. J. Corey, S. W. Walinsky, J. Am. Chem. Soc. 1972, 94,
8932Ϫ8933.
H. Mayr, R. Schneider, C. Schade, J. Bartl, R. Bederke, J. Am.
Chem. Soc. 1990, 112, 4446Ϫ4454.
c
3
2
),
2
2
[4]
[5]
2
2
3
3
(
2
[6]
[7]
[8]
Ϫ1
m
C
.
ϩ
Ϫ MS (FD, 0Ϫ12 mA); m/z (%): 189, 188, 187 (10, 27, 100) [M
Ϫ BF
9.65, H 5.57.
Ϫ
4 9 4 2
]. Ϫ C H15BF S (274.2): calcd. C 39.43, H 5.52; found C
3
C. J. Cramer, G. D. Hawkins, G. C. Lynch, D. J. Giesen, I.
Rossi, J. W. Storer, D. G. Truhlar, D. A. Liotard, AMSOL-ver-
sion 5.0, Quantum Chemistry Program Exchange Program 606,
based in part on AMPAC-version 2.1 by D.A. Liotard, E. F.
Healy. J. M. Ruiz, M. J. S. Dewar.
3
,4,8,8a-Tetrahydro-2H,5H-1-thia-4a-thionianaphthalene Tetra-
Ϫ
4
Ϫ
4
fluoroborate (3cϪBF
): Reaction of 1ϪBF
(1.50 g, 7.50 mmol)
with 2c (2.36 g, 43.6 mmol) at Ϫ17°C for 2 d and at ambient tem-
perature for 4 d and recrystallization of the product from butanone
yielded 1.59 g (6.11 mmol, 82%) of 3c in colorless prisms with m.p.
[9]
H. Mayr, W. Förner, P. v. R. Schleyer, J. Am. Chem. Soc. 1979,
1
01, 6032Ϫ6040.
[10]
AM1 calculations were performed with the program MOPAC
6.01 by J. P. Stewart, University of Texas at Austin, Austin, TX.
E. E. Eliel, S. H. Wilen, L. N. Mander, Stereochemistry of Or-
ganic Compounds, John Wiley and Sons, New York 1994.
J. Klaveness, F. Rise, K. Undheim, Acta Chem. Scand. Ser.B
[
6]
1
1
3
H
H
5
14.0Ϫ114.5°C (ref. : 85%, 118Ϫ119°C). Ϫ H NMR (CD
00 MHz): δ ϭ 2.10Ϫ2.38 (m, 2 H, 3-H ), 2.71Ϫ3.03 (m, 4 H, 8-
), 3.85Ϫ4.00 (m, 2 H, 5-
3
CN,
[11]
2
2
2 2
and 2-H ), 3.33Ϫ3.47 (m, 2 H, 4-H
[12]
3
2
), 4.82 (t, J ϭ 6.4 Hz, 1 H, 8a-H), 5.71Ϫ5.79 (m, 1 H, 6-H),
1
986, 40, 398Ϫ399.
.94Ϫ6.01 (m, 1 H, 7-H). Ϫ 13C NMR(CD
[98140]
3
CN, 75.5 MHz): Table
1922
Eur. J. Org. Chem. 1998, 1919Ϫ1922