Reactions of cyclohexenyl iodonium tetrafluoroborate with bromide ion: Retardation due to the formation of λ3-bromoiodane

Article abstract of DOI:10.1021/jo0523013

The reaction of 4-tert-butylcyclohex-1-enyl(phenyl)iodonium tetrafluoroborate (1a) and the 4-chlorophenyl derivative (1b) with bromide ion was examined in methanol, acetonitrile, and chloroform. Products include those derived from the intermediate cyclohexenyl cation as well as 1-bromocyclohexene. Kinetic measurements show that the reaction of 1 is strongly retarded by the added bromide. The curved dependence of the observed rate constant on the bromide concentration is typical of a pre-equilibrium formation of the intermediate adduct with a fast bromide-independent reaction (solvolysis of the iodonium ion). The formation of the adduct, λ³-bromoiodane, was also confirmed by the UV spectral change. The relative reactivity of the iodonium ion and λ³-bromoiodane is evaluated to be on the order of 10². The bromide substitution product forms both via the S_N1 reaction of the free iodonium ion and via the ligand coupling of the iodane.

Full text of DOI:10.1021/jo0523013

Reactions of Cyclohexenyl Iodonium Tetrafluoroborate with
Bromide Ion: Retardation Due to the Formation of λ3-Bromoiodane
Tadashi Okuyama,* Shohei Imamura,† and Morifumi Fujita
Graduate School of Material Science, Himeji Institute of Technology, UniVersity of Hyogo, Kamigori,
Hyogo 678-1297, and Graduate School of Engineering Science, Osaka UniVersity, Toyonaka,
Osaka 560-8531, Japan
okuyama@sci.u-hyogo.ac.jp
ReceiVed NoVember 7, 2005
The reaction of 4-tert-butylcyclohex-1-enyl(phenyl)iodonium tetrafluoroborate (1a) and the 4-chlorophenyl
derivative (1b) with bromide ion was examined in methanol, acetonitrile, and chloroform. Products include
those derived from the intermediate cyclohexenyl cation as well as 1-bromocyclohexene. Kinetic
measurements show that the reaction of 1 is strongly retarded by the added bromide. The curved
dependence of the observed rate constant on the bromide concentration is typical of a pre-equilibrium
formation of the intermediate adduct with a fast bromide-independent reaction (solvolysis of the iodonium
ion). The formation of the adduct, λ3-bromoiodane, was also confirmed by the UV spectral change. The
relative reactivity of the iodonium ion and λ3-bromoiodane is evaluated to be on the order of 102. The
bromide substitution product forms both via the SN1 reaction of the free iodonium ion and via the ligand
coupling of the iodane.
Introduction
of 1 to give cyclohexyne, and the reaction with an amine was
concluded unexpectedly to occur via the E1 mechanism, in
contrast to the reaction with acetate proceeding via the E2
mechanism.3 That is, cyclohexenyl cation 6 is readily generated
thermally even in aprotic solvents (Scheme 1).
A secondary vinylic iodonium salt, cyclohex-1-enyl(aryl)-
iodonium tetrafluoroborate, 1, undergoes SN1-type solvolysis
in alcoholic solvents via a vinylic cation as an intermediate (eq
1).1 Thermolysis of 1 in chloroform to give 1-fluorocyclohexene,
The reaction of 1 with bromide ion provides 1-bromocyclo-
hexene 7 in a high yield in chloroform, and this product was
suggested to form by ligand coupling within the hypervalent
adduct, λ3-bromoiodane 8 (eq 3), instead of a possible SN1-
type mechanism.3 Reactions of simple alk-1-enyl iodonium salts
with a halide ion other than fluoride give 1-haloalk-1-ene with
(1) Okuyama, T.; Takino, T.; Sueda, T.; Ochiai, M. J. Am. Chem. Soc.
1995, 117, 3360-3367.
(2) Okuyama, T.; Fujita, M.; Gronheid, R.; Lodder, G. Tetrahedron Lett.
2000, 41, 5125-5129.
(3) Fujita, M.; Kim, W. H.; Sakanishi, Y.; Fujiwara, K.; Hirayama, S.;
Okuyama, T.; Ohki, Y.; Tatsumi, K.; Yoshioka, Y. J. Am. Chem. Soc. 2004,
126, 7548-7558.
4, was also suggested to follow the SN1-type mechanism (eq
2).2 Weak bases such as acetate and amines induce â-elimination
† Osaka University.
10.1021/jo0523013 CCC: $33.50 © 2006 American Chemical Society
Published on Web 01/07/2006
J. Org. Chem. 2006, 71, 1609-1613
1609

Okuyama et al.
SCHEME 1
SCHEME 2
FIGURE 1. Observed rate constants for the reaction of 1b in methanol
in the presence of tetrabutylammonium perchlorate (0), tetrabutylam-
monium tetrafluoroborate (∆), and tetrabutylammonium bromide (O)
at 50 °C.
complete inversion of configuration,4 which are concluded to
occur via the vinylic SN2 (SNVσ) mechanism.5 In this and other
reactions, a halide is known first to form a λ3-haloiodane
intermediate like 8,8-11 and this intermediate as well as the free
iodonium ion undergoes ensuing reactions such as SNVσ in the
case of alk-1-enyl iodonium salts (Scheme 2).6,7 When the SNVσ
and other reactions become sluggish, the ligand-coupling (or
SNVπ)7 reaction may show up, as in the case of 2-haloalk-1-
enyliodonium salt.8,9
The nucleophilic substitution of 1 can occur via the SN1 or
the ligand-coupling mechanism, as mentioned above (the SNVσ
mode of reaction is not possible with the cyclic substrate), and
kinetic analysis allows the differentiation between the two
mechanisms.9 One purpose of this study is to address this point
for the reaction of 1 with bromide ion. The reactivity of λ3-
iodane in comparison with that of the iodonium ion is also an
interesting issue to identify the true reactive species to afford
the products. The effects of an added bromide salt on the
reaction rates for 1 in different solvents are examined in this
study, showing that the λ3-bromoiodane is much less reactive
than the free iodonium ion in polar solvents.
FIGURE 2. Initial absorbance (b) at 255 nm and observed rate
constants (O) for the reaction of 1b in methanol at 50 °C in the presence
of tetrabutylammonium bromide and at the constant ionic strength of
0.20 mol dm-3, maintained with tetrabutylammonium perchlorate. The
solid curves are those calculated with the parameters given in Table 2.
50-60 °C in the presence of tetrabutylammonium bromide and
other salts. The yields of the enol ether derivatives were
determined as their hydrolysis product, 4-tert-butylcyclohex-
anone (2′). The product distributions for the reactions of 1a
under varied conditions are summarized in Table 1.
Kinetic measurements were performed with 1a and 1b by
monitoring the decrease in UV absorption. When the tetrafluo-
roborate salt of 1 is dissolved in a solution containing bromide
(or chloride ion), a new absorption band immediately develops
at a longer wavelength, as observed for other vinyl iodonium
salts.5,6,9-11 This absorbance decays following the pseudo-first-
Results
order kinetic law to provide the observed rate constants kobsd
.
The reactions of 4-tert-butylcyclohex-1-enyl(phenyl)iodonium
tetrafluoroborate (1a) and the 4-chlorophenyl derivative (1b)
were carried out in methanol, acetonitrile, and chloroform at
The extrapolated initial absorbance A0 can also be measured in
the same way as described previously.5,6 The results obtained
in methanol, acetonitrile, and chloroform are presented in
Figures 1-4.
(4) Ochiai, M.; Oshima, K.; Masaki, Y. J. Am. Chem. Soc. 1991, 113,
7059-7061.
(5) Okuyama, T.; Takino, T.; Sato, K.; Oshima, K.; Imamura, S.;
Yamataka, H.; Asano, T.; Ochiai, M. Bull. Chem. Soc. Jpn. 1998, 71, 243-
257.
Discussion
Solvolysis in Methanol. The solvolysis of 1 in various
alcoholic solvents affords the recombination products 3 in
addition to the expected enol derivative 2 (or cyclohexanone
2′), and the rate is only slightly dependent on solvent polarity,
which is consistent with the reaction of an ionic substrate giving
an ionic product (intermediate).1 The recombination products
3 are of the Friedel-Crafts type and are always abundant in
the ortho (about 80%) of the three isomers, to imply recombina-
(6) Okuyama, T.; Takino, T.; Sato, K.; Ochiai, M. J. Am. Chem. Soc.
1998, 120, 2275-2282.
(7) Okuyama, T.; Lodder, G. AdV. Phys. Org. Chem. 2002, 37, 1-56.
(8) Ochiai, M.; Oshima, K.; Masaki, Y. Chem. Lett. 1994, 871-874.
(9) Okuyama, T.; Takino, T.; Sato, K.; Ochiai, M. Chem. Lett. 1997,
955-956.
(10) Okuyama, T.; Oka, H.; Ochiai, M. Bull. Chem. Soc. Jpn. 1998, 71,
1915-1921.
(11) Okuyama, T.; Sato, K.; Ochiai, M. Bull. Chem. Soc. Jpn. 2000, 73,
2341-2349.
1610 J. Org. Chem., Vol. 71, No. 4, 2006

Reactions of Iodonium and λ3-Bromoiodane
TABLE 1. Product Yields of the Reactions of 1a under Varied Conditions
product yield (%)
temp
(°C)
time
(h)
[Bu4NBr]
other salt
3a
(o/m/p)
no.
solv.
(mol dm-3
)
(concn; mol dm-3
)
2′
5
7
PhI
1
2
3
4
5
6
7
8
9
10
11
12
13
14
CHCl3
CHCl3
CHCl3
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeOH
MeOH
MeOH
MeOH
60
60
60
50
50
50
50
50
50
50
60
60
60
60
1.0
25
25
0
0.10
0
Bu4NClO4 (0.10)
none
28 (n.d.)
42
nd
90
80
72
75
78
83
88
84
75
57
75
73
62
1.5 (86:6:8)
1.8 (86:6:8)
7.9 (81:6:13)
8.7 (79:9:12)
6.8 (86:7:7)
4.5 (86:7:7)
5.1 (84:7:9)
2.5 (87:8:5)
3.8 (79:9:12)
6.7 (81:6:13)
9.7 (81:9:10)
5.8 (79:12:9)
9.0 (81:7:12)
98
Bu4NCl (0.10)
Bu4NClO4 (0.20)
Bu4NClO4 (0.19)
Bu4NClO4 (0.18)
Bu4NClO4 (0.15)
Bu4NClO4 (0.10)
none
Bu4NCl (0.20)
Bu4NClO4 (0.20)
Bu4NClO4 (0.10)
none
98
1.0
1.5
2.5
4.0
4.0
6.0
11.5
1.5
2.0
3.0
3.0
0
39
31
30
23
18
17
18
36
42
47
49
0.01
0.02
0.05
0.10
0.20
0
21
40
62
76
80
70
0
0.10
0.20
0
7.8
13
Bu4NCl (0.20)
7.4
SCHEME 3
More significant deceleration by added bromide is apparent
when the ionic strength is kept constant at 0.2 mol dm-3 with
Bu4NClO4 in methanol (Figure 2). The initial absorbance A0 at
255 nm increases with increasing bromide concentration [Bu4-
NBr], as shown in Figure 2. Both the rate and the absorbance
curves show saturation with [Bu4NBr]. These curves may be
Scheme 3 and Table 2 explained by a reaction scheme with a
pre-equilibrium formation of the adduct (iodane 8 as well as
iodate 9, if applicable), as shown in Scheme 3.5,6,10,11 The
absorbance increase, ∆A0 ) A0 - A0 at [Br-] ) 0, and the rate
constant kobsd can be represented by eqs 4 and 5, respectively,
if all three species contribute to the reaction.5,6,10,11
TABLE 2. Kinetic Parameters Used for the Simulation of the
Kinetic Curves of Figures 2-4 for the Reaction of 1b with Bromide
Iona
MeOH
MeCNb
CHCl3
MeCNc
∆A0 ) (∆A1K1[Br-] + ∆A2K1K2[Br-]2)/
(1 + K1[Br-] + K1K2[Br-]2) (4)
K1 (mol-1 dm3)
K2 (mol-1 dm3)
5.0
∼0
690 (370)
∼0 (0.7)
30 (40)
0.6 (2)
1600
3.3
7160
15.1
104 k1 (s-1
104 k2 (s-1
k1/k2
)
)
6.8
8.1
1.5
5.4
(0.30)d
(0.028)d
(11)d
kobsd ) (k1 + k2K1[Br-] + k3K1K2[Br-]2)/
(1 + K1[Br-] + K1K2[Br-]2) (5)
∼0
large
50 (20)
a Reactions were carried out at 50 °C and with an ionic strength of 0.20
mol dm-3 in methanol and acetonitrile or 0.10 mol dm-3 in chloroform,
maintained with Bu4NClO4. b Values in parentheses are those obtained for
the reaction of 1a at 60 °C. c Data taken from ref 6 for the reaction of
dec-1-enyl(phenyl)iodonium tetrafluoroborate with chloride at 25 °C and
The absorbance and rate curves are simulated by a nonlinear
least-squares method to give solid curves shown in Figure 2
with the parameters K1 ) 5.0 mol-1 dm3 and k1 ) 6.8 × 10-4
s-1; contributions from 9 (K2) and the reaction of 8 (k2) can be
neglected here. Reasonable curve fittings with these kinetic
parameters show that the reaction of the free iodonium ion 1 is
retarded by the formation of iodane 8 and that the reactivity of
8 is negligibly small (probably more than two orders of
magnitude smaller than 1) in methanol. The formation of the
bromide-substitution product 7 (up to 13% at [Br-] ) 0.2 mol
dm-3) must occur via the SN1 mechanism from the iodonium
ion 1 (the rate-determining step does not involve the bromide
ion). Intermediate cation 6 is trapped by bromide as well as by
the methanol solvent.
Reaction in Acetonitrile. The reaction products of 1a in
acetonitrile are those of the solvolysis type, 2′ and 3a after
workup, in the absence of an added nucleophile (no. 4 in Table
1). The primary product must be enamide, a product of the Ritter
reaction, which is readily hydrolyzed to cyclohexanone 2′. When
bromide is added to the reaction mixture, bromocyclohexene 7
is formed up to nearly 80% at [Br-] ) 0.2 mol dm-3 (nos.
5-9, Table 1). Similar results were obtained with chloride (no.
10, Table 1).
with an ionic strength of 0.10 mol dm-3
.
d Second-order rate constant in
mol dm-3 s-1 for the SN2 (SNVσ) reaction.
tion occurring within the tight ion-molecule pair.1 This product,
3, was always observed in the thermal reactions of 1 in the range
of the solvents given in Table 1, suggesting intermediary
formation of the cyclohexenyl cation 6 at least partially under
any conditions employed.
The effects of added salts on the reaction rate for the 4-chloro
substrate 1b were examined in methanol at 50 °C, and the results
are shown in Figure 1. The addition of tetrabutylammonium
bromide and chloride (not shown) decelerates the reaction, while
neutral salts, perchlorate and tetrafluoroborate, accelerate it. The
effects of perchlorate and fluoroborate are much the same. The
increased ionic strength must enhance the polarity of the medium
and accelerate heterolytic reactions. Halides provide some
substitution products, 7 (or 5), but the neutral salts do not much
affect the products, just a small decrease in the fractions of the
recombination products 3, which may be ascribed to the
dissociation of the ion-molecule pair. Because the ionic strength
increases significantly the reaction rate of the iodonium salts
1, all the measurements hereafter were undertaken at the constant
The remarkable change in the rate was observed with added
bromide (Figures 3 and 4). Figure 3 shows the results for the
ionic strength of 0.1 or 0.2 mol dm-3
.
J. Org. Chem, Vol. 71, No. 4, 2006 1611

Okuyama et al.
previously with other iodonium salts,5,6,10,11 but the contribution
from 9a to the reaction (k3) may be negligible in the concentra-
tion range studied. The reactivity of the iodate should be still
lower than that of the iodane.
The reaction rates for 1b in acetonitrile at 50 °C, shown in
Figure 4a, are also simulated by eq 5, with parameters given in
Table 2. The formation of iodate 9b (K2) cannot be evaluated
only from the rate measurements as a result of it being too small
of a contribution. A satisfactory fitting of the curve was obtained
with K2 ) 0 (Figure 4a) to give the apparent iodonium/iodane
reactivity (k1/k2) of about 50.
These kinetic analyses show that both λ3-bromoiodane 8 and
free iodonium ion 1 are reactive species of the reaction in the
presence of bromide ion. The contribution from each species
can be calculated from eq 5, with the parameters given in Table
2, and trial calculations show that the fraction of the bromide
product 7 formed is always greater than the calculated contribu-
tion from 8. This indicates that 7 forms not only from 8 but
also from free 1. Although the formation of 7 from 8 should
occur via ligand coupling, the formation of 7 from 1 must
proceed through the SN1 mechanism. The latter mechanism
accompanies the formation of 2′ and 3.
FIGURE 3. Initial absorbance (b) at 270 nm and observed rate
constants (O) for the reaction of 1a in acetonitrile at 60 °C in the
presence of tetrabutylammonium bromide and at the constant ionic
strength of 0.20 mol dm-3, maintained with tetrabutylammonium
perchlorate. The solid curves are those calculated with the parameters
given in Table 2.
Reaction in Chloroform. The thermolysis of 1a in chloro-
form affords fluoride- and chloride-substitution products, 4 and
5, as well as the recombination product 3a (eq 2).2 When 0.1
mol dm-3 of perchlorate, Bu4NClO4, was added, the formation
of fluoride 4 was essentially inhibited and chloride 5 (42%)
and 3a (28%) were obtained as the main products (no. 1, Table
1). The addition of excess perchlorate leads to the dissociation
of the tetrafluoroborate of 1a, which prohibits the intermediate
cation 6 from picking up fluoride. The reaction of 1a with
bromide in chloroform was found to proceed efficiently to give
a high yield of the substitution product 7.3 However, a careful
reexamination of this reaction showed the formation of a small
amount (1.5% at [Bu4NBr] ) 0.1 mol dm-3) of 3a in addition
to 7 (no. 2, Table 1). Similar results are also seen with chloride
(no. 3, Table 1). These observations suggest the formation of 7
via the SN1 mechanism, if only in a small fraction. This is
compatible with the results of the kinetic analysis.
Although the bromide (and the chloride) ion induces an
effective substitution reaction of 1 in chloroform, the reaction
rate sharply decreases with the addition of a bromide salt (Figure
4b). The rate curve can be reproduced satisfactorily by eq 5,
with the parameters of Table 2. In this case, a rough value of
K2 could be evaluated, but the contribution from 9 to the reaction
seems to be too small in the range of the [Br-] studied. The k3
is neglected in the simulation of the kinetic curve. The relative
rate constant, k1/k2, is rather small (5.4) in this reaction.
FIGURE 4. Observed rate constants for the reaction of 1b in the
presence of tetrabutylammonium bromide at 50 °C in acetonitrile (a)
at the ionic strength of 0.20 mol dm-3 and in chloroform (b) at the
ionic strength of 0.10 mol dm-3, adjusted with tetrabutylammonium
perchlorate. The solid curves are those calculated with the parameters
given in Table 2.
Iodonium and Iodane Reactivity. The relative reactivity of
the free iodonium ion 1 and iodane 8 evaluated from the k1/k2
values is very dependent on the solvent employed, as seen in
the last line of Table 2. Both of the processes responsible for
k1 and k2 are unimolecular, but the reactions involved are not
the same: the former involving only spontaneous heterolysis
of 1, while the latter involves ligand coupling as well as
spontaneous heterolysis. The ease of unimolecular heterolysis
should be dependent on the polarity of the solvent, but an
intramolecular ligand-coupling reaction would not depend much
on the solvent. The relatively larger contribution of the latter
reaction in a less polar solvent must give rise to the apparently
smaller relative reactivities of 1 and 8. So, the intrinsic relative
reactivities of 1 and 8 compared for the same reaction, for
reaction of 1a in acetonitrile at the ionic strength of 0.2 mol
dm-3, maintained with Bu4NClO4 at 60 °C. The absorbance at
270 nm increases with [Br-], while kobsd sharply decreases at
very low concentrations of Br-. These curves are simulated
according to eqs 4 and 5, respectively. The solid curves in Figure
3 are calculated with K1 ) 370 mol-1 dm3, K2 ) 0.7 mol-1
dm3, k1 ) 4.0 × 10-3 s-1, and k2 ) 2 × 10-4 s-1 (Table 2).
The relative rate for the reaction of free 1a and λ3-bromoiodane
8a is evaluated to be about 20 (k1/k2) under these conditions.
The absorbance change observed seems to show the formation
of iodate 9a at a higher concentration of bromide, as found
1612 J. Org. Chem., Vol. 71, No. 4, 2006

Reactions of Iodonium and λ3-Bromoiodane
example, spontaneous heterolysis, should be large, as observed
in methanol and could be evaluated at least to the order of 102.
The value obtained in chloroform may be the result of
competition of different reactions, spontaneous heterolysis and
ligand coupling. From the previous kinetic analysis of the SNVσ
reaction of dec-1-enyliodonium salt6 (the last column of Table
2), the relative reactivities of iodonium and iodane are evaluated
to be 11. This rather small value may be a result of the nature
of the transition state of the bimolecular substitution.
The other problem is the relative reactivities of the two
species. In the beginning of this study, we assumed that the
iodanyl group (PhIX-) could be a better leaving group than
the iodonio group (PhI+-) because the hypervalent C-I bond
of iodane is longer than the C-I bond involved in the iodonium
ion,19,20 and, thus, the former may be weaker than the latter.
However, this hypothesis is clearly disproved by the present
kinetic results. The leaving ability of the iodanyl group is about
102 times smaller than that of the iodonio group. Our previous
conclusion that the leaving ability of the phenyliodonio group
(PhI+-) is about 106 times greater than that of triflate was
obtained from the solvolysis in aqueous ethanol,1 and this refers
to the iodonio group. The iodanyl group (PhIBr-) is only 104-
fold better than triflate as a leaving group.
Iodonium salts are often called iodane12,13 because species
represented by R2IX can exist either in an ionic or in a covalent
form; an electronegative ligand X may be anionic or may form
a hypervalent (covalent) bond to the iodine. There is a strong
argument in favor of the iodane nomenclature13 on the basis of
the structure of the iodonium salts in a crystalline form, where
the iodine moiety is T-shaped and the counteranion locates on
the apical direction of the polyvalent iodine.13-15 There is
hypervalent interaction, but the separation between the iodine
and the stable anion, like tetrafluoroborate or triflate, is quite
large, suggesting the ionic character of this bonding.14,15 The
triflate in iodonium crystals has a weak interaction with the
iodine through one of the three oxygen atoms of the sulfonate
group, but these oxygen atoms are apparently equivalent,16
characteristic of the triflate anion. Furthermore, the iodane
obviously dissociates in solution, as the UV spectrum suggests.
Molecular weight17 and conductivity measurements18 show that
the dissociation in solution results in the formation of the
iodonium ion. That is, both iodonium and iodane forms do exist
in solution in a proportion that is dependent on the concentration
and the dissociation constant, which is very dependent on the
counteranion and solvent.
Experimental Section
Substrates 1a and 1b, solvents, and tetrabutylammonium salts
were obtained as described previously.1,5,6 All the products were
characterized by NMR and MS spectra and compared with reported
data.1,3
Reactions were carried out on a scale of a 1-5 mL solution
containing 1-10 mg of the substrate 1, and products were
determined by vapor-phase chromatography in the same way as
before.1,5,6 The UV absorbance changes were recorded on a
spectrophotometer Shimadzu UV-2200. The measurements of the
initial absorbance and the reaction rates were also done in the same
way as described for similar iodonium salts.1,5,6
Acknowledgment. The initial part of this work was started
in collaboration with Professor Masahito Ochiai during the study
on the solvolysis reported in ref 1. We are grateful for his
valuable suggestions on this manuscript. Also acknowledged
is financial support by Grant-in-Aid for Scientific Research on
Priority Area, Reaction Control of Dynamic Complexes, from
the Ministry of Education, Culture, Sports, Science and Tech-
nology, Japan.
(12) Varvoglis, A. HyperValent Iodine in Organic Synthesis; Academic
Press: San Diego, California, 1997; p 3.
(13) (a) Ochiai, M. J. Organomet. Chem. 2000, 611, 494-508. (b) Ochiai,
M. Hypervalent Iodine Chemistry. In Topics in Current Chemistry; Wirth,
T., Ed.; Springer-Verlag: Berlin, 2003; Vol. 224, pp 5-68.
(14) Stang, P. J. Angew. Chem., Int. Ed. Engl. 1992, 31, 274-285.
(15) Koser, G. F. In The Chemistry of Functional Groups; Supplement
D2; Patai, S., Rappoport, Z., Eds.; Wiley: Chichester, 1995; Chapter 21.
(16) Stang, P. J.; Arif, A. M.; Crittell, C. M. Angew. Chem., Int. Ed.
Engl. 1990, 29, 287-288.
(17) Ochiai, M.; Kida, M.; Sato, K.; Takino, T.; Goto, S.; Donkai, N.;
Okuyama, T. Tetrahedron Lett. 1999, 40, 1559-1562.
(18) Kline, E. R.; Kraus, C. A. J. Am. Chem. Soc. 1947, 69, 814-816.
(19) Zhdankin, V. V.; Stang, P. J. In Chemistry of HyperValent
Compounds; Akiba, K.-y., Ed.; Wiley-VCH: New York, 1999; p 329.
JO0523013
(20) The apical C-S bond of sulfurane (1.926 Å in o,o′-bisbi-
phenylylenesulfurane)20a is also obviously longer than the sulfonium C-S
bond (1.80 Å).20b (a) Ogawa, S.; Matsunaga, Y.; Sato, S.; Iida, I.; Furukawa,
N. J. Chem. Soc., Chem. Commun. 1992, 1141-1142. (b) Perozzi, E. F.;
Paul, I. C. In The Chemistry of the Sulphonium Group; Part 1; Stirling, C.
J. M., Ed.; Wiley: Chichester, England, 1981; Chapter 2.
J. Org. Chem, Vol. 71, No. 4, 2006 1613