Kinetic investigation on the hydrolysis of aryl(fluoro)(phenyl)-λ6-sulfanenitriles

Article abstract of DOI:10.1246/bcsj.74.945

A kinetic investigation on the hydrolysis of aryl(fluoro)(phenyl)-λ⁶-sulfanenitriles was carried out in some aqueous and mixed aqueous-organic solutions. The pH-rate profiles showed that the hydrolysis consists of pH-independent, acidcatalyzed

Full text of DOI:10.1246/bcsj.74.945

.
.
© 2001 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 74, 945—954 (2001)
945
Chemical
iety of Ja-
Chemical
iety of Ja-
Kinetic Investigation on the Hydrolysis of
λ6
Aryl(fluoro)(phenyl)- -sulfanenitriles
Hydrolysis of Fluoro-λ6-sulfanenitriles
T. Dong et al.
Tiaoling Dong, Takayoshi Fujii,* Satoro Murotani, Huagang Dai, Shin Ono, Hiroyuki Morita,
Choichiro Shimasaki, and Toshiaki Yoshimura*
01
Department of Material Systems Engineering and Life Science, Faculty of Engineering, Toyama University,
Gofuku, Toyama 930-8555
SJA8
9-2673
53
(Received October 5, 2000)
ober
0
1
1
A kinetic investigation on the hydrolysis of aryl(fluoro)(phenyl)-λ6-sulfanenitriles was carried out in some aqueous
and mixed aqueous-organic solutions. The pH-rate profiles showed that the hydrolysis consists of pH-independent, acid-
catalyzed and base-catalyzed reactions. The neutral hydrolysis of fluoro-λ6-sulfanenitriles was found to proceed via an
SN1 or an S-nitrilosulfonium cation-like transition state, which is characterized by a large negative Hammett ρ-value (ρ
ꢀ ꢁ1.76 in water; ꢁ1.85 in CH3CN/H2O(1/4, v/v); ꢁ2.35 in TFE/H2O (1/1, v/v)), relatively large m-values (ca. 0.83 for
fluoro(diphenyl)-λ6-sulfanenitrile; ca. 0.82 for fluoro(p-nitrophenyl)(phenyl)-λ6-sulfanenitrile against the solvent ioniz-
ing power YOTs-values in acetonitrile–water), a common ion effect in TFE/H2O, and a small salt effect. The large nega-
tive activation entropies (ꢁ60–ꢁ101 J Kꢁ1 molꢁ1) were presumed to be due to strong solvation of Fꢁ with H2O in the re-
action systems. The ease of ion dissociation of the S–F bond was examined by a theoretical calculation in a DFT
method, to show that the SN1-like transition state is caused by a facile tendency of dissociation of the S–F bond of fluoro-
λ6-sulfanenitriles. The acid-catalyzed hydrolysis was found to proceed via a more cation-like transition state involving a
concerted proton transfer to the fluorine atom and breaking of the sulfur-fluorine bond in the λ6-sulfanenitrile. The alka-
line hydrolysis probably takes place via an SN2 mechanism.
Aryl(fluoro)(phenyl)-λ6-sulfanenitriles,1 ArPhFS(N) 1 bear-
ing an SN triple bond are interesting unusual organic com-
pounds. We previously reported2 that the spectral data of fluo-
ro(diphenyl)-λ6-sulfanenitrile reported by Clifford et al.3 are
doubtful because of their improper isolation and large differ-
ence from those of our sample synthesized by a new method.
In recent years, we have indirectly confirmed the structure of
the compound by derivation to triphenyl-4 and diphenyl(pro-
poxy)-λ6-sulfanenitriles,5 and we have also prepared other λ6-
sulfanenitriles by using the fluoro-λ6-sulfanenitriles as starting
materials.6–8 An X-ray crystallographic analysis of the struc-
ture of fluoro(p-nitrophenyl)(phenyl)-λ6-sulfanenitrile has just
been carried out.9 We have been further studying their reactiv-
ities of conversion to various other new λ6-sulfanenitriles. It is
evident that these derivatives are formed by nucleophilic sub-
stitution at the sulfur atom. At the same time, these fluoro-λ6-
sulfanenitriles were found to be readily hydrolyzed to give the
corresponding sulfoximides, unlike tetracoordinated sulfonyl
fluorides, which are usually unreactive toward hydrolysis un-
der acidic or neutral conditions.10
fur compounds, such as sulfones and sulfoximides. Nucleo-
philic substitutions at sulfur are usually suggested to proceed
with or without a hypervalent intermediate (sulfurane) via an
addition–elimination mechanism or a concerted SN2-like
mechanism.12 However, there has been no mechanistic investi-
gation on the nucleophilic reaction at the sulfur atom of λ6-sul-
fanenitriles. Since it is expected that we can find new reactions
and some new information from the unusual SN bonding
through a clarification of the mechanism for the hydrolysis of
fluoro-λ6-sulfanenitriles 1, kinetic investigations on the hydrol-
ysis were carried out in some aqueous and mixed aqueous–or-
ganic solutions by an UV spectroscopic method.
Results and Discussion
Hydrolysis Products: Product analyses were carried out
for the hydrolysis of aryl(fluoro)(phenyl)-λ6-sulfanenitriles
1a–e in an aqueous acetonitrile solution under neutral, acidic
or basic conditions. Without addition of halide salts, the reac-
tions were found to afford the corresponding sulfoximides 2a–
e quantitatively (Scheme 1). However, in the presence of po-
Our main interest in organic λ6-sulfanenitriles is how the SN
triple bond activates other sulfur bonds. More recently, we re-
ported that alkoxy-λ6-sulfanenitriles undergo a facile hydroly-
sis through an SN1 or SN1-like mechanism at the alkyl carbon
atom, in which the sulfoximide moiety becomes a very good
leaving group upon protonation of alkoxy-λ6-sulfanenitriles.11
In our previous paper,8 it was reported that alkoxy-λ6-sulfanen-
itriles exceptionally undergo a facile Ei reaction, although λ6-
sulfanenitriles are classified as tetracoordinated hexavalent sul-
Scheme 1.

946 Bull. Chem. Soc. Jpn., 74, No. 5 (2001)
Hydrolysis of Fluoro-λ6-sulfanenitriles
tassium bromide, the hydrolysis of 1d was also found to give
N-bromo-S-p-nitrophenyl-S-phenylsulfimide and p-nitrophe-
nyl phenyl sulfide together with the corresponding sulfoximide
2d.
pH-Rate Profiles and Kinetic Data: The rates of the hy-
drolysis of fluoro-λ6-sulfanenitriles 1 in some aqueous solu-
tions, aqueous acetonitrile, and aqueous 2,2,2-trifluoroethanol
(TFE) were determined by following the changes in the UV
absorbance of the reaction mixture at a proper wavelength. In
all cases, the reactions were found to follow nicely the pseudo-
first-order kinetics.
Typical pH-rate profiles for the hydrolysis of 1a, 1c, and 1d
are shown in Fig. 1. As can be seen from the pH-rate profiles,
the hydrolysis is catalyzed by both acid and base, while in the
pH range of 4.57–10.96 for 1a and 1c and in the pH range of
4.57–9.60 for 1d, the rate of hydrolysis is almost pH-indepen-
dent. The linear dependences (Figs. 2 and 3) of kobs on [Hꢂ]
and [OHꢁ] in the acidic and basic pH ranges afforded the sec-
Fig. 3.
tions at 20.4 ˚C.
Dependence of kobs on [OHꢁ] in dilute NaOH solu-
ond-order rate constants, kH and kOH, respectively, as summa-
rized in Table 1. The intercepts in the linear plots of kOH vs.
[OHꢁ] were found to be consistent with the observed rate con-
stants in water in Table 1.
Alkaline Hydrolysis: Substituent Effect. A Hammett
plot for the second-order rate constants, kOH, of the alkaline hy-
drolysis afforded a positive ρ-value of 1.12 (Table 1), indicat-
ing that the bond formation of sulfur-nucleophile is kinetically
more important than fission of the leaving group. Thus, the al-
kaline hydrolysis reaction is considered to proceed via a direct
SN2 displacement (Scheme 2) or via an addition-elimination
mechanism involving a hypervalent intermediate (sulfurane).
The rates of typical SN2 reactions are not very sensitive to the
substituents on the phenyl group on the displacement center at-
om, since there is a cancellation of electronic demands for the
bond-making and bond-breaking components. Meanwhile,
two-step addition-elimination reactions which proceed via
rate-determining bond-making usually give larger positive
Hammett ρ-values. For example, the Hammett ρ-values are
around 2.2 for the alkaline hydrolysis of benzoate esters pro-
ceeding via a tetrahedral intermediate.13 For the alkaline hy-
drolysis of substituted benzenesulfonyl fluorides in dioxane–
water (45/55, w/w), the Hammett ρ-value is as high as 2.79.14
On the contrary, a smaller ρ-value (1.546) was reported for the
alkaline hydrolysis of substituted benzenesulfonyl chlorides,15
in which a synchronous displacement is assumed from the fact
of no 18O incorporation in the sulfonyl group in the alkaline
hydrolysis of phenyl benzenesulfonate in H218O–dioxane. The
ρ-value observed for the alkaline hydrolysis of 1 is even small-
er than this, suggesting that there is more development of
bond-breaking at the transition state, and that the present alka-
line hydrolysis likely occurs via the SN2 mechanism.
Fig. 1.
pH-Rate profiles for the hydrolysis of 1a, 1c and 1d
at 20.4 ˚C. 2.81 ≤ pH ≤ 10.55: various buffer solutions at
ionic strength 0.01; pH < 2.81: aqueous HClO4; pH >
10.55: aqueous NaOH solutions.
Activation Parameters. The activation enthalpy (∆H‡)
and entropy (∆S‡) for the alkaline hydrolysis of 1d were deter-
mined to be 43 kJ molꢁ1 and ꢁ93 J Kꢁ1 molꢁ1 at 25 ˚C, re-
spectively (see Table 1). The activation entropy is more nega-
tive than those of the alkaline hydrolysis of substituted
benzenesulfonyl chloride and bromide,15,16 which are in the
range of ∆S‡ values (ꢁ20–ꢁ40 J Kꢁ1 molꢁ1) for the usual SN2
reactions of neutral substrates with anion nucleophiles.17 The
Fig. 2.
Dependence of kobs on [Hꢂ] in dilute perchloric
acid solutions containing 35.5% (v/v) of CH3CN at 10.9
˚C.

T. Dong et al.
Bull. Chem. Soc. Jpn., 74, No. 5 (2001) 947
Table 1.
Kinetic Data for the Hydrolysis of 1
kH/sꢁ1 molꢁ1 dm3
a)
[1]
X
Temp.
˚C
k
obsꢃ105/sꢁ1
kOH
∆H‡
∆S‡
H2O
(D2O)
CH3CN
/H2Ob)
84.9
TFE/
H2Oc)
13.5
CH3CN/H2O
(D2O)d)
3.01 (3.39)
H2O
s
ꢁ1 molꢁ1
kJ
J Kꢁ1
molꢁ1
(D2O)e)
dm3
molꢁ1
a
a
H
H
10.9
20.4
2260
(1510)
202
37.0
0.278
60ꢄ1f)
73ꢄ1g)
ꢁ87ꢄ1f)
ꢁ60ꢄ2g)
a
a
b
b
b
c
c
c
d
d
H
H
30.0
39.1
10.9
20.4
30.0
10.9
20.4
30.0
10.9
20.4
468
102
244
p-Me
p-Me
p-Me
p-Cl
p-Cl
p-Cl
249
600
1310
33.9
84.4
192
6.61 (7.83)
1.09
130
60ꢄ1f)
ꢁ84ꢄ4f)
996
11.2
0.363
63ꢄ1f)
ꢁ91ꢄ3f)
p-NO2
p-NO2
0.117 (0.142) 0.371
1.09
1.97
65ꢄ1f)
60ꢄ1h)
ꢁ101ꢄ5f)
ꢁ97ꢄ2h)
101
(60.5)
225
10.2
0.781
3.10
0.910
(1.11)
2.31
d
d
e
p-NO2
p-NO2
m-Cl
30.0
39.1
10.9
20.4
25.1
53.0
3.66
66ꢄ2i)
43ꢄ1j)
ꢁ20ꢄ6i)
ꢁ93ꢄ4j)
471
0.459
e
m-Cl
369
31.9
0.418
ρk)
r
ꢁ1.76
0.990
ꢁ1.85
0.986
ꢁ2.35
0.986
ꢁ1.87
0.995
1.12
0.958
a) In dilute NaOH of 0.001–0.01 mol dmꢁ3. b) CH3CN/H2O ꢀ 1/4 (v/v). c) TFE/H2O ꢀ 1/1 (v/v). d) In dilute perchloric acid
solution of 0.001–0.3 mol dmꢁ3 containing 35.5% (v/v) CH3CN. e) In dilute perchloric acid solution of 0.001–0.02 mol dmꢁ3. f)
For the hydrolysis in CH3CN/H2O (1/4, v/v) at 25 ˚C. g) For the hydrolysis in TFE/H2O (1/1, v/v) at 25 ˚C. h) For the hydrolysis
in water at 25 ˚C. i) For kH in dilute perchoric acid at 25 ˚C. j) For kOH in dilute NaOH solution at 25 ˚C. k) For the rate of hydrol-
ysis at 20.4 ˚C except for kH at 10.9 ˚C.
Scheme 2.
present larger negative ∆S‡ is probably due to the strong solva-
tion of the fluoride ion leaving group.
Solvent Effect. The solvent effect on rate was examined
by using mixed CH3CN–H2O solvents. The m-values for the
solvent effects obtained by the correlations (the Winstein equa-
tion) of the rate constants with YOTs-values18 are given in Table
2. For both 1a and 1d, the solvent effect on kOH showed a pos-
itive m-value (0.35) for bimolecular reactions with a hydroxide
anion. Usually, the reaction of an uncharged molecule with an
anion would be expected to show no change or decrease in rate
Table 2.
Solvent Effect on the Hydrolysis of 1 under Neutral and Basic Conditions at 25.2 ˚C
Aq CH3CN
YOTs
k
obsꢃ104/sꢁ1 b)
kOH/sꢁ1 molꢁ1 dm3 c)
a)
% v/v (w/w)a)
40.8 (35)
35.5 (30)
29.9 (25)
24.3 (20)
12.5 (10)
0
1a
3.99
1d
0.183
1a
1d
0.392
1.8
1.9
2.5
2.7
3.6
4.1
0.044
0.056
0.076
0.104
0.193
0.310
10.6
19.3
78.5
0.535
1.03
4.05
0.569
0.840
1.37
347
15.4
2.61
m
r
0.83
0.992
0.82
0.995
0.35
0.995
0.35
0.986
a) Ref. 18. b) Neutral. c) Dilute NaOH of 0.001–0.01 mol dmꢁ3
.

948 Bull. Chem. Soc. Jpn., 74, No. 5 (2001)
Hydrolysis of Fluoro-λ6-sulfanenitriles
with increasing the solvent polarity, since an initial charge is
dispersed in the transition state.19 The present opposite solvent
effect, i.e., the 7-fold increase in rate with decreasing the ace-
tonitrile concentration from 40.8% (v/v) to zero, reflects a rela-
tively large degree of development of fission of the sulfur-fluo-
rine bond in the transition state. There is an extreme example
of large m-values (0.75–0.80), in spite of the SN2 reaction, for
the reaction of 1-phenylethyl derivatives with azide ion pro-
ceeding through an open, “exploded” transition state that
closely resembles a carbocation.20
Neutral Hydrolysis: Substituent Effect. The rates of
hydrolysis of 1 under neutral conditions were nicely correlated
with σ-values and gave considerably large negative ρ-values
under all three conditions (ρ ꢀ ꢁ1.76 (0.990) in water; ρ ꢀ
ꢁ1.85 (0.986) in CH3CN/H2O (1/4, v/v); ρ ꢀ ꢁ2.35 (0.986) in
TFE/H2O (1/1, v/v)) (see Table 1). The behavior of the substit-
uent effect suggests that there is a large development of posi-
tive charge on the central sulfur atom in the transition state of
the nucleophilic reaction. There is an example of SN1 hydroly-
sis involving a carbocation intermediate having an unsaturated
carbon heteroatom bond, i.e., hydrolysis of p-dimethylami-
nobenzoyl fluoride.21 However, the magnitude of the negative
ρ-value might be too small for an SN1 mechanism, although
the standard ρ-value for SN1 solvolysis on the sulfur atom is
not established, and the ρ-value for a reaction involving an S-
nitrilosulfonium cation may become less negative by electron
movement from the nitrogen atom toward the central sulfur
atom by polarization or resonance in Scheme 3. The charge on
the sulfur atom in Scheme 3 shows a contribution of dication.
However, the magnitude of charge is not important for the sta-
bility of the ion, as can be seen in the extreme examples of sul-
fonium ions; instead, the charge difference between the initial
state and the transition state is rather important. There are
mechanistic variations for the hydrolysis of acyl halides de-
pending on the leaving ability. Most substituted benzoyl fluo-
rides undergo hydrolysis in aqueous solution through an asso-
ciative mechanism with ρ ꢀ 1.7, while most substituted
benzoyl chloride through a dissociative one with nucleophilic
ꢂ
participation by solvent with ρ ꢀ ꢁ3.12.21 On the other
hand, the substituent effects for the neutral hydrolysis of tetra-
coordinated sulfur (VI) compounds, benzenesulfonyl chlo-
ride,15 and thiophene-2-sulfonyl chloride and bromide22 were
reported to give upward curved or U-shaped Hammett plots.
For this reason, the mechanism of the hydrolysis of sulfonyl
halides was considered to involve a loose SN2 transition state,
which changes from the one in which the bond-breaking of
sulfonyl-halogen is more advanced to the one in which the
bond-making of sulfur-nucleophile is more advanced, depend-
ing on the electron-donating ability of the substituents and the
ability of the leaving group. Meanwhile, a Hammett plot for
the hydrolysis of benzenesulfonyl fluoride gives almost a
straight line with a positive slope of 1.8,23 suggesting an ad-
vanced bond-making transition state. All these facts reveal that
the sulfonyl sulfur–fluorine bond is especially hard to be ion-
ized. In the present case, a good linearly correlated Hammett
plot with a large negative slope shows that even the sulfonyl
sulfur–fluorine bond in fluoro-λ6-sulfanenitriles has a dissocia-
tive character, as shown in Scheme 4. Though the ρ-value for
the reaction in CH3CN/H2O (1/4, v/v) is essentially not differ-
ent from that in water, the ρ-value in TFE/H2O (1/1, v/v) is
more negative than that in water, suggesting that the transition
state for the reaction in less nucleophilic solvent is shifted to a
Scheme 3.
Scheme 4.
Table 3.
[1] Salt
Salt Effect on the Hydrolysis of 1a and 1d in Water and TFE/H2O at 20.4 ˚C
obsꢃ103/sꢁ1
Solvent
water
k
k
saltꢃ102 b)
[Salt]a)
0
0.05
22.0
1.04
0.1
0.2
0.3
0.4
0.5
1
1.5
s
ꢁ1 molꢁ1
dm3
a
a
d
d
d
d
d
KF
LiClO4
KF
KCl
KBr
22.6
22.6
1.01
22.1
23.4
23.3
24.0
1.16
25.2
33.1
27.8
1.19
0.48
0.08
0.65
1.39
0
24.5
1.26
1.33
1.01
1.01
1.01
1.01
1.71
2.48
2.43
3.60
3.10
5.13
3.64
6.61
LiClO4
NaClO4
0.989 0.977 0.950 0.945
0.953 0.950
0.980
TFE/H2O
(l/l, v/v)
a
a
KF
LiClO4
0.370
0.370
0.359
0.417
0.333 0.286 0.222
0.537 0.860 1.19
a) [Salt]: mol dmꢁ3. b) ksalt is the slope in the plot of kobs vs [salt].

T. Dong et al.
Bull. Chem. Soc. Jpn., 74, No. 5 (2001) 949
more cation-like one. This is supported by the common ion ef-
fect of KF and the stronger effect of the ionic strength of
LiClO4 on the hydrolysis of 1a in TFE/H2O, shown later in Ta-
ble 3.
sociative character, as proposed by Song and Jencks for the
hydrolysis of benzoyl chlorides.21 On the other hand, the
present solvent isotope effects are close to those for the hydrol-
ysis of sulfonyl chlorides, which was suggested a looser and
more ionic transition state.26
Activation Parameters. The activation enthalpies and en-
tropies for the neutral hydrolysis of 1a–d in CH3CN/H2O (1/4,
v/v) at 25 ˚C are in the range of 60–65 kJ molꢁ1 and ꢁ84–
ꢁ101 J Kꢁ1 molꢁ1, respectively, as shown in Table 1. The val-
ues of ∆H‡ are found to decrease, and ∆S‡ to increase slightly
with increasing electron-donating ability of the substituents.
This may suggest that the electron-donating substituents facili-
tate the dissociation of the sulfur–fluorine bond, leading to a
positive contribution to ∆S‡. Also, the activation enthalpy (73
kJ molꢁ1) and entropy (ꢁ60 J Kꢁ1 molꢁ1) for the hydrolysis of
1a in TFE/H2O (1/1, v/v) at 25 ˚C in Table 1 are larger than
those in CH3CN/H2O (1/4, v/v), showing that the transition
state of the reaction in the former mixed solvent is more disso-
ciative than that in the latter. The more negative ∆S‡ values
than that for the usual SN1 process are probably due to the
strong solvation of the leaving fluoride ion in the transition
state with water, as explained by Jencks et al.24 Although the
present values are still small, they are obviously larger than
that (ꢁ171 J Kꢁ1 molꢁ1) for the hydrolysis of p-nitrobenzene-
sulfonyl fluoride in dioxane/water (40/60, v/v),23 suggesting
the dissociative character in the present system.
Solvent Effect. The solvent effect using CH3CN–H2O af-
forded large m-values (m ꢀ 0.83 for 1a and 0.82 for 1d) for the
neutral hydrolysis as given in Table 2. The large m-values sug-
gest further that the dissociation of the leaving group is impor-
tant for the neutral hydrolysis, although the smaller m-values
than unity likely reflect incomplete dissociation of the fluoride
anion in the transition state caused by some nucleophilic solva-
tion of the cationic center as proposed by Bentley et al.25
Thus, an S-nitrilosulfonium cation-like transition state with an
advanced bond cleavage and little bond formation is the case
for this reaction. In the present case, almost the same m-values
for the compounds with very different substituents (H and p-
NO2) do not suggest a continuous change in the nature of the
transition state upon changing the substituents to the more
electron-attracting, unlike in the case of the hydrolysis of sul-
fonyl halides.26 This is consistent with a linear Hammett cor-
relation for the neutral hydrolysis in several aqueous solutions.
Therefore, the present transition state is suggested to be looser
than that for the sulfonyl halides.
Salt Effect. In order to find further evidence supporting
the dissociative transition state, salt effect was also examined.
In the case of the reaction of either 1a or 1d in water, addition
of potassium fluoride does not exhibit a common ion inhibi-
tion, but slightly increases the rate of hydrolysis, as shown in
Table 3. Potassium chloride and bromide were found to obvi-
ously accelerate the hydrolysis of 1d, which is attributable to
an attack of the halide ion on the nitrogen atom, because the
formation of N-bromo-S-p-nitrophenyl-S-phenylsulfimide was
confirmed by UV and NMR spectra. The rate-enhancing effect
of fluoride ion may also be due to the formation of a more eas-
ily hydrolyzable intermediate, N-fluorosulfimide, although the
formation of any intermediate was not observed in the UV
spectral change of the reaction mixture in the presence of po-
tassium fluoride. The rate of hydrolysis of N-fluorosulfimide
may be estimated as being sufficiently faster than other N-ha-
losulfimides from our previous results concerning the halide
effect on the alkaline hydrolysis of N-halosulfimides,27 but it is
difficult to discuss the rate of the unknown N-fluorosulfimide.
Furthermore, there is another possibility of a general base ca-
talysis of fluoride ion toward the attacking water molecule.
On the other hand, the salt effect in Table 3 shows that per-
chlorate salts slightly accelerate the hydrolysis of 1a in water,
but not the hydrolysis of 1d. The small positive salt effect of
perchlorate for 1a and the almost no salt effect for 1d are con-
sistent with those reported by Manege et al. for SN1–SN2 inter-
mediates, isopropyl bromide and benzyl chloride.28
It is interesting that the addition of potassium fluoride to
TFE/H2O (1/1, v/v) exhibited an obvious common ion inhibi-
tion, and lithium perchlorate more strongly accelerated the hy-
drolysis of 1a in the mixed solvent than in water, suggesting a
substantial or complete dissociation of fluoride anion in the
transition state. The results show that the hydrolysis of 1a in a
less nucleophilic solvent, TFE/H2O (1/1, v/v), leads to SN1 or
more SN1-like transition state than in water. The close ρ-value
for the hydrolysis in water to the one in TFE–H2O suggests
that the transition state in the reaction in water is also loose
enough.
Density Function Theory Calculation of S-Nitrilosulfo-
nium Cation. In order to confirm the ease of ionic dissocia-
tion of fluoro-λ6-sulfanenitriles, the heterolytic dissociation
energy of a model compound of fluoro-λ6-sulfanenitriles was
calculated and compared with that of formyl fluoride, methyl
fluoride, and sulfonyl fluoride at the B3LYP/6-311ꢂꢂG (3df,
2pd) level. A zero-point energy correction was performed at
the B3LYP/6-31G*//B3LYP/6-31G* level using the scaling
factor 0.9804.29 The results are compiled in Table 4. The het-
erolytic bond dissociation energies include information about
both the bond energy and the stabilization energy of the cation
by substituents. The large positive endothermic dissociation
energies are actually compensated by solvation to make ioniza-
tion possible. The heterolytic dissociation energy of formyl
fluoride is expectedly smaller than that of methyl fluoride, but
the energy for the fluoro-λ6-sulfanenitriles is further 105 kJ
Solvent Isotope Effect. According to the kinetic data in
Table 1, the ratios of
/
for the neutral hydrolysis of
k
D2O
k
H2O
1a and 1d were found to be 1.50 and 1.67 at 20.4 ˚C, respec-
tively. Song and Jencks reported that an SN1 mechanism for
the hydrolysis of p-dimethylaminobenzoyl fluoride had a sol-
vent isotope effect,
/
, of 1.1.21 Meanwhile, the hy-
kD O
2
k
H2O
drolysis of p-methoxybenzoyl and benzoyl fluorides was sug-
gested to occur via an associative mechanism involving a
general base catalysis, in which the solvent isotope effects are
2.3. In our case, the associative mechanism is not consistent
with the large negative ρ-value for the neutral hydrolysis of
fluoro-λ6-sulfanenitriles. However, the present solvent isotope
effects are larger than 1.1 for p-dimethylaminobenzoyl fluo-
ride, which may be attributed to a weak nucleophilic interac-
tion with a water molecule in the transition state having a dis-

950 Bull. Chem. Soc. Jpn., 74, No. 5 (2001)
Hydrolysis of Fluoro-λ6-sulfanenitriles
Table 4. Heterolytic Dissociation Energies (∆E) of R–F
Bond in Various Model Compoundsa)
b)
c)
c)
R
Em
ꢁ367046 102
Em0
Ecb)
Ec0
∆Ed)
1074
ꢁ103693
81
ꢁ561475
54
ꢁ298233
37
967
ꢁ1454371
ꢁ1704430
66
57
ꢁ1191236
ꢁ1441134
51
43
862
1024
a) The total energies (kJ molꢁ1) were calculated at the
B3LYP/6-311ꢂꢂG (3df, 2pd) level. The zeropoint ener-
gies were calculated at the B3LYP/6-31G*//B3LYP/6-31G*
level and corrected by the scaling factor 0.9804 (Ref. 29).
b) Em, Ec are the total energies of molecule and cation,
respectively. c) Em0, Ec0 are the zero-point energies of the
molecule and the cation, respectively. d) The total energy
(Ef) of fluoride ion is ꢁ262258 kJ molꢁ1
(EcꢂEc0)ꢂEfꢁ(EmꢂEm0).
.
c) ∆E ꢀ
molꢁ1 lower than that of formyl fluoride and 162 kJ molꢁ1
lower than that of sulfonyl fluoride. This means that the ion-
ization of fluoro-λ6-sulfanenitriles is essentially easier than
that of acyl fluoride, although the stabilization energies by sub-
stituted phenyl groups of the present system cannot be com-
pared with those of the p-dimethylaminobenzoyl fluoride and
p-methoxybenzoyl fluoride.21 The optimized geometry of the
S-nitrilosulfonium cation is planar with a longer SN bond
length of 1.472 Å than that of 1.438 Å of the fluoro-λ6-sul-
fanenitrile, as shown in Scheme 5, which is in contrast to the
results for the C–O bond length in formyl fluoride, although
the change in the S–O bond length is very small in sulfonyl flu-
oride. NBO natural population analyses were performed30 at
the B3LYP/6-311ꢂꢂG (3df, 2pd)//B3LYP/6-311ꢂꢂG (3df,
2pd) level. The change in the natural charge on the central at-
oms and some other atoms and fluorine are shown in Scheme
5. The increase in the positive natural charge on the sulfur
atom is more negative in fluoro-λ6-sulfanenitrile than that in
sulfonyl fluoride, although the opposite change does not match
to the reaction image and cannot be compared with the change
on the different atoms. The electron movement from the nitro-
gen to the sulfur atom in the dissociation process is found to be
very large. The negative charge on the fluorine atom of fluoro-
λ6-sulfanenitrile is the largest of the four. The NBO analysis
for fluoro-λ6-sulfanenitrile shows a strong ionic character for
the S–F bond, which is characterized as lone-pair occupancies
(1.996, 1.972, 1.958, 1.616), suggesting that the negative
hyperconjugation31 from the lone electron-pair on the nitrogen
atom to the S–F σ* bond activates the S–F bond to make easier
the S–F ionization. The relatively small change in the charge
on the sulfur atom is the result of resonance stabilization of the
cation. It is therefore probable that the strong stabilization by
resonance between the nitrogen and the cationic sulfur atom
makes it possible to dissociate even the poorly leaving fluoride
ion, so that the substitution reaction of 1 with a weak nucleo-
Scheme 5.
phile, such as a water molecule, takes place via a loose transi-
tion state having a dissociative character, and with a further
weak nucleophilic solvent, TFE–H2O, via an SN1 process.
Acidic Hydrolysis: Substituent Effect. The acidic hy-
drolysis is also facilitated by electron-donating substituents.
The Hammett plot of kH for the hydrolysis of 1 in dilute per-
chloric acid containing 35.5% (v/v) CH3CN at 10.9 ˚C afford-
ed a large negative ρ-value of ꢁ1.87, close to those for the
neutral hydrolysis (see Table 1), indicating that the positive
charge at the central sulfur atom in the transition state is simi-

T. Dong et al.
Bull. Chem. Soc. Jpn., 74, No. 5 (2001) 951
ate value, considering a positive contribution by the proton
transfer from hydronium ion to the fluoride ion leaving group
relative to ∆S‡ (ꢁ97 J Kꢁ1molꢁ1) for the neutral hydrolysis,
and for paths a and d, the initial protonation should make a fur-
ther positive contribution.
Solvent Isotope Effect. The solvent isotope effects, kH/kD,
for 1a, 1b, and 1d in dilute perchloric acid containing 35.5%
(v/v) CH3CN at 10.9 ˚C were found to be smaller than those in
the neutral hydrolysis, and even a little smaller than unity (0.89
for 1a, 0.84 for 1b, and 0.82 for 1d, respectively, see Table 1).
The hydrolysis of 1d in aqueous perchloric acid at 20.4 ˚C also
afforded the same isotope effect (0.82) as in the aqueous aceto-
nitrile solution at 10.9 ˚C. The values are not in the range of
0.3–0.45 for typical A1 mechanisms,34 nor in the range of 0.5–
0.6 for acid-catalyzed ester hydrolyses,35 showing paths c and
a to be unlikely. These specific acid-catalyzed mechanisms
contribute largely to the reverse isotope effect, and are not con-
sistent with the present results. For path d, the large reverse
isotope effect is cancelled by some secondary nomal effect due
to the attack of water, depending on the position of the transi-
tion state. Meanwhile, a general acid-catalyzed mechanism
shows a large normal isotope effect, depending on the position
of the transition state. However, the present results show that
some contribution of protonation should be involved in the
mechanism of the neutral hydrolysis. There are examples of
reverse isotope effect (0.5–1) for the general acid-catalyzed
hydrolysis of ortho esters in which the Brønsted α-values are
large.36
Buffer Effect. In order to further clarify the acid-cata-
lyzed mechanism, the buffer effect was examined, as shown in
Table 5. Typical buffer dependences of the hydrolysis of 1d in
several buffers are shown in Fig. 4. In almost all cases, the ob-
served rate constants, kobs, are linearly dependent on the total
buffer concentration, [B]t: kobs ꢀ k0 ꢂ kB[B]t. k0 is the rate
constant extrapolated to zero buffer concentration, and kB is the
second-order rate constant calculated from the buffer-depen-
dent term. The k0 values are almost equal to the values in neu-
tral hydrolysis or in acidic hydrolysis without a buffer. In the
range of neutral pH, phosphate and imidazole buffers have
some buffer effects, but MES, MOPS, and MOPSO buffers do
not show any buffer effects, likely suggesting a nucleophilic
catalysis by a conjugate base of a buffer, depending on the pos-
sibility of bond formation between the sulfur atom in the sub-
strate and the buffer base. In the low-pH range, plots of the
buffer-catalyzed constant, kB, vs. acid fraction of buffer gave
straight lines with intercepts, as shown in Fig. 5, suggesting
that both the conjugate acid of the buffer and the base compo-
nent catalyze this hydrolysis. Since there is no tendency to
curve involving second-order dependency of the buffer con-
centration at the present range in Fig. 4, the two kinds of catal-
ysis take place in two parallel reaction processes.
Scheme 6.
lar to, or a little larger than, that in the neutral hydrolysis. The
acidic hydrolysis may be considered to take place by several
possible pathways, as shown in Scheme 6. Path a is an acid-
catalyzed addition–elimination mechanism involving an initial
protonation on the nitrogen atom, followed by a rate-determin-
ing addition of a water molecule on the sulfur atom to form a
hypervalent intermediate, resembling acid-catalyzed carboxyl-
ic ester hydrolysis. The initial protonation cannot be the rate-
determining step, because proton transfer between heteroat-
oms without other concerted bond cleavage or bond formation
is commonly very fast. Paths c and d involve an initial pre-
equilibrium protonation on the fluorine atom, followed by a
rate-determining cleavage of the sulfur–fluorine bond unassist-
ed and assisted by a nucleophilic attack of a water molecule on
the sulfur atom, respectively. Path b involves a general acid-
catalyzed cleavage of the sulfur–fluorine bond. Of these, path
a is not expected to afford a large negative ρ-value due to the
result of a cancellation of the negative ρ-value for the pre-equi-
librium protonation by a positive one for the rate-determining
nucleophilic reaction, as is the case for ester hydrolysis.32 Path
c would give a large negative ρ-value (probably ꢁ2.35 esti-
mated from the SN1 solvolysis in TFE–H2O) due to the net
positive charge on the sulfur atom of the intermediate. Paths b
and d would give a moderate ρ-value, depending on the degree
of the S–F bond cleavage and the nucleophilic O–S bond for-
mation. The protonated fluorine atom in path d is a much bet-
ter leaving group than the fluoride ion in the neutral hydrolysis,
and thus the transition state in path d should be more cation-
like to give a more negative ρ-value than that in the neutral hy-
drolysis. The present ρ-value is too close to that in the neutral
hydrolysis to consider path d, and is therefore most consistent
with path b.
According to the above experimental results, the total rate
equation is as follows, considering an equilibrium protonation
at the nitrogen atom in terms of the basicity of the nitrogen
atom of λ6-sulfanenitrile:4,6
Activation Parameters. The activation enthalpy and en-
tropy for kH of the hydrolysis of 1d in dilute perchloric acid so-
lutions ([Hꢂ]: 0.001–0.02 mol dmꢁ3) are 66 kJ molꢁ1 and ꢁ20
J Kꢁ1molꢁ1 at 25 ˚C, respectively (see Table 1). Path c of an
A1 mechanism is not compatible to the present system, be-
cause it should afford a large positive activation entropy.33
Meanwhile, the activation entropy for path b may be a moder-
KaSN (k ꢂ kH[Hꢂ])
KaSN
aSN ꢂ[Hꢂ]
H2O
aSN ꢂ[Hꢂ]
kobs ꢀ
ꢂ
(kHA [HA]ꢂkA [Aꢁ]),
(1)
K
K
where
is the pseudo first-order rate constant for the reac-
k
H2O

952 Bull. Chem. Soc. Jpn., 74, No. 5 (2001)
Table 5.
Hydrolysis of Fluoro-λ6-sulfanenitriles
Buffer Effect on the Hydrolysis of 1da)
Bufferb)
Fraction of
buffer acidc)
Temp
pH
k0ꢃ103
sꢁ1
8.32ꢄ0.08
3.75ꢄ0.28
2.09ꢄ0.02
kBꢃ102
ꢁ1 molꢁ1 dm3
2.0ꢄ0.1
1.5ꢄ0.2
0.92ꢄ0.03
k
HAꢃ102
kAꢃ102
ꢁ1 molꢁ1 dm3
˚C
s
s
ꢁ1 molꢁ1 dm3
s
Phosphate
Phosphate
Phosphate
0.67
0.50
0.33
10.9
10.9
10.9
1.72
2.12
2.35
3.0
0
Glycine
0.50
10.9
2.66
1.22ꢄ0.03
0.18ꢄ0.03
Chloroacetate
Chloroacetate
Chloroacetate
0.67
0.50
0.33
10.9
10.9
10.9
2.42
2.78
3.01
1.77ꢄ0.08
1.13ꢄ0.07
0.71ꢄ0.07
0.41ꢄ0.04
0.43ꢄ0.05
0.36ꢄ0.08
0.48
0.27
0.29
0.33
0.68
0.82
Formate
Formate
Formate
0.67
0.50
0.33
10.9
10.9
10.9
3.31
3.62
3.92
0.59ꢄ0.04
0.48ꢄ0.03
0.42ꢄ0.04
0.40ꢄ0.02
0.47ꢄ0.02
0.54ꢄ0.04
Acetate
Acetate
0.50
0.33
25.2
25.2
4.66
4.98
1.51ꢄ0.01
1.50ꢄ0.02
0.56ꢄ0.01
0.65ꢄ0.02
MES
MES
MOPSO
MOPS
Phosphate
Imidazole
0.83
0.50
0.50
0.50
0.50
0.50
25.2
25.2
25.2
25.2
25.2
10.9
5.40
6.12
6.93
7.20
6.87
7.50
1.51ꢄ0.01
1.50ꢄ0.01
1.44ꢄ0.02
1.43ꢄ0.03
1.48ꢄ0.01
0.42ꢄ0.02
0
0
0
0
0.39ꢄ0.02
0.32ꢄ0.01
a) The buffers at the ionic strength 0.10 maintained with LiClO4 were used. b) MES: 2-morpholinoethanesulfonate (pKa 6.15, 20
˚C), MOPSO: 2-hydroxy-3-morpholinopropanesulfonate (pKa 6.90), MOPS: 3-morpholinopropanesulfonate (pKa 7.20, 20 ˚C). c)
[Buffer acid]/total buffer concentration.
Fig. 5.
Plots of kB for the hydrolysis of 1d vs. fraction of
buffer acid in buffer at ionic strength 0.1 maintained with
LiClO4 at 10.9 ˚C.
Fig. 4.
Buffer effect on the hydrolysis of 1d in phosphate
(pH ꢀ 2.12), chloroacetate (pH ꢀ 2.78), and formate (pH
ꢀ 3.62) buffer solutions with ionic strength 0.1 maintained
with LiClO4 at 10.9 ˚C.
respectively. The protonation on the nitrogen atom is consid-
ered to depress the dissociation of fluoride anion. Since no sat-
uration was observed at low pH in the pH-rate profile, no sig-
nificant protonation on the nitrogen atom is considered to
occur in the present pH-range, suggesting that pKa of fluoro-
λ6-sulfanenitrile is smaller than 1 for 1d and 2 for 1a. Thus,
the buffer-dependent second-order rate constant, kB, can be de-
scribed as follows from the second term of Eq. 1:
tion with a water molecule, and kH, kHA, and kA are the second-
order rate constants, depending on the concentration of proton,
buffer acid, and buffer base, respectively. KaSN is the dissocia-
tion constant of the N-protonated λ6-sulfanenitrile. [HA] and
[Aꢁ] represent the concentration of the buffer acid and base,

T. Dong et al.
Bull. Chem. Soc. Jpn., 74, No. 5 (2001) 953
133.3, 133.6, 135.6, 143.0, 143.2, 145.2, 145.4. Found: C, 56.72;
H, 3.62; N, 5.17%. Calcd for C12H9ClFNS: C, 56.81; H, 3.58, N,
5.52%.
[HA]
[B]t
[A]
[B]t
kB ꢀ kHA
ꢂkA
,
(2)
Hydrolysis Product Analysis: To an CD3CN solution of 1
(0.023 mmol, 0.19 mL) in an NMR tube was added 0.37 mL of
D2O and NMR spectrum was immediately taken. After two days
at room temperature, the mixture was analyzed by NMR again.
Besides the corresponding sulfoximides being formed, no other
products were found.
On the other hand, an acetonitrile solution of 1 (0.14 mmol, 0.5
mL) was pipetted into 4 mL of CH3CN/H2O (1/4, v/v) under neu-
tral, acidic (0.02 mol dmꢁ3 of HClO4) or basic (0.01 mol dmꢁ3 of
NaOH) conditions and stirred rapidly to start the reaction at room
temperature. When the reaction finished, to the mixture solution
was added 50 mL of 0.05 mol dmꢁ3 of NaOH; it was then extract-
ed with CHCl3. The obtained organic layer was dried over anhy-
drous magnesium sulfate. After the solvent was evaporated, the
residue was identified by NMR as corresponding sulfoximides
only.
where [B]t is the total buffer concentration. Since [HA] ꢂ
[Aꢁ] ꢀ [B]t, the equation can be rearranged as
[HA]
(3)
kB ꢀ kAꢂ(kHAꢁkA )
,
[B]t
and the catalytic rate constants, kA and kHA, are determined
from the intercepts extrapolated at that buffer acid fraction is
zero and unity, respectively, in the plot of kB vs. acid fraction
of buffer (Fig. 5). There is an apparent tendency that the stron-
ger is the acidity of the buffer acid, the larger is the kHA value,
though the α-value cannot be estimated because of large ex-
perimental errors and too few data. The observation of general
acid catalysis is consistent with the small reverse isotope ef-
fect, and thus path b. The buffer effect is attributable to the
ease of ionic dissociation of the sulfur-fluorine bond in fluoro-
λ6-sulfanenitriles.
Although the question whether the acid-catalyzed hydroly-
sis of fluoro-λ6-sulfanenitrile is SN1 or not is hard to answer,
the ρ-value of ꢁ1.87 is too small a negative value compared to
that of ꢁ2.35 for the neutral hydrolysis in TFE–H2O, suggest-
ing that some nucleophilic assistance of water cannot be ne-
glected in the transition state.
The hydrolysis products of 1 in TFE/H2O (1/1, v/v) mixed sol-
vent were analyzed in the similar experimental procedure in
CH3CN/H2O (1/4, v/v) under neutral conditions. Besides the cor-
responding sulfoximide, no other products were found by NMR.
λ6
The Hydrolysis of Fluoro(p-nitrophenyl)(phenyl)- -sul-
fanenitrile (1d) in the Presence of KBr. To a 12.2% (v/v) ace-
tonitrile aqueous solution (114 mL) of 2 mol dm–3 of KBr was
gradually added 1 mL of an acetonitrile solution of fluoro(p-nitro-
phenyl)(phenyl)-λ6-sulfanenitrile (21.3 mg). After 30 min, the so-
lution was diluted with an NaHCO3 solution and extracted by
CHCl3. The organic layer was dried over anhydrous magnesium
sulfate. After the solvent was evaporated, the residue was identi-
fied by NMR to contain N-bromo-S-p-nitrophenyl-S-phenyl-
sulfimide (40%), p-nitrophenyl phenyl sulfide (20%), besides the
corresponding hydrolysis product, S-p-nitrophenyl-S-phenylsul-
foximide (2d) (40%).
Kinetic Measurement: The hydrolysis reaction was started
by adding a 2 µL sample of the acetonitrile solution of 1 (0.14 mol
dmꢁ3) to 3.0 mL of a reaction solution in a 10 mm quartz cell
equilibrated at 20.4 ꢄ 0.1 ˚C in a cell compartment of a spectro-
photometer. The variation of the UV absorbance of λ6-sulfaneni-
triles at a given wavelength (220 nm for 1a, 225 nm for 1b and 1c,
240 nm for 1d, and 229 nm for 1e) was automatically recorded at
10-second intervals. The reactions were followed over 10 half-
lives. The pseudo-first-order rate constants were calculated by a
least-squares method using more than 8 points accumulated dur-
ing the first 70% of the reaction.
Conclusion
The S–F bond in aryl(fluoro)(phenyl)-λ6-sulfanenitriles 1
easily undergoes heterolytic dissociation, probably due to
strong resonance stabilization between the nitrogen and the
sulfur atom. It results in that the neutral hydrolysis of aryl(flu-
oro)(phenyl)-λ6-sulfanenitriles 1 proceeds via an S-nitrilosul-
fonium cation intermediate or cation-like transition state with a
large amount of dissociative character, even the sulfone-type
compounds, and even having a poorer leaving group fluoride
ion than other halide ions. The alkaline hydrolysis takes place
more possibly via an SN2 mechanism. The acid-catalyzed hy-
drolysis may involve a concerted proton transfer to the fluorine
atom and breaking of the sulfur-fluorine bond in the λ6-sul-
fanenitrile. All three mechanisms under the respective condi-
tions reflect the ease of the ionic cleavage of the S–F bond of
fluoro-λ6-sulfanenitriles.
Experimental
An examination of the solvent effect was carried out by mea-
suring the rates of hydrolysis of 1a and 1d in mixed solvents of
several compositions of CH3CN/H2O under neutral and basic con-
ditions. The m-values of the solvent effects, as shown in Table 2,
General: All reagents and solvents were obtained commer-
cially, and were further purified by general methods when neces-
sary. The pH value of a buffer solution was determined by a Hori-
ba F-13 pH meter. UV spectra were measured by a HITACHI U-
3000 spectrophotometer. Density function theory calculations and
natural bond orbital calculations were performed using the GAUS-
SIAN 9437 program package in an IBM RS/6000-SP2 computer at
Toyama University Computing and Network Services Center.
λ6
were obtained from plots of the logarithms of various rate con-
18
stants against the solvent ionizing power, YOTs
,
of 2-adamantyl
tosylate.
The salt effect was examined by measuring the rate constants of
hydrolysis in several concentrations of salt solutions. The solvent
isotope effect was determined by using the corresponding deuteri-
um solvent or reagent.
The buffer effect was examined by measuring the rates of hy-
drolysis in five concentrations of buffer solutions with the same
pH and the same ionic strength, which were prepared by diluting a
stock concentrated buffer solution with a LiClO4 solution.
The activation parameters were obtained from a least-squares
Aryl(fluoro)(phenyl)- -sulfanenitriles:
The aryl(fluoro)-
(phenyl)-λ6-sulfanenitriles 1a–e were prepared according to the
methods reported in our previous papers.2,8
λ6
m-Chlorophenyl(fluoro)(phenyl)- -sulfanenitrile (1e).
Yield 55%; mp 118–120 ˚C (decomp); IR (KBr) 1366 cmꢁ1(νSN);
1H NMR (400 MHz, CDCl3) δ 7.47–7.51 (m, 1H), 7.55–7.59 (m,
3H), 7.63–7.67 (m, 1H), 7.78–7.81 (m, 1H), 7.90–7.95 (m, 3H);
13C NMR (100 MHz, CDCl3) δ 125.0, 127.0, 127.1, 129.5, 130.6,

954 Bull. Chem. Soc. Jpn., 74, No. 5 (2001)
Hydrolysis of Fluoro-λ6-sulfanenitriles
treatment of ln k against Tꢁ1 at three or four temperatures of 10–
40 ˚C.
16 E. Ciuffarin, L. Senatore, and M. Isola, J. Chem. Soc., Per-
kin Trans. 2, 1972, 468.
17 N. S. Isaacs, “Physical Organic Chemistry,” Longman Sci-
entific & Technical (1987), p. 379.
18 T. W. Bentley and G. Llewellyn, Prog. Phys. Org. Chem.,
17, 121 (1990).
19 J. March, “Advanced Organic Chemistry,” 4th ed, John
Wiley & Sons, New York (1992), p. 358.
20 J. P. Richard and W. P. Jencks, J. Am. Chem. Soc., 106,
1383 (1984).
This work was supported in part by a Grant-in-Acid for Sci-
entific Research on Priority Areas (A) of the Chemistry of In-
ter-element Linkage (No. 09239218, 10133218, and
11120219) from the Ministry of Education, Science, Sports
and Culture. We thank Prof. Tadashi Okuyama of the Himeji
Institute of Technology for helpful discussions.
21 B. D. Song and W. P. Jencks, J. Am. Chem. Soc., 111, 8470
(1989).
22 a) A. Arcoria, F. P. Ballistreri, G. Musumarra, and G. A.
Tomaselli, J. Chem. Soc., Perkin Trans. 2, 1981, 221. b) A.
Arcoria, F. P. Ballistreri, E. Spina, and G. A. Tomaselli, J. Chem.
Soc., Perkin Trans. 2, 1988, 1793.
References
1
This type of compound was first named as “thiazyne” by
Clifford et al. (Ref. 3) previously, and we have also used this
name. But because of confusing with heterocyclic didehydrothi-
azine etc., the name, “λ6-sulfanenitrile,” is used in this paper based
on IUPAC recommendation.
23 M. E. Aberlin and C. A. Bunton, J. Org. Chem., 35, 1825
(1970).
2
T. Yoshimura, H. Kita, K. Takeuchi, E. Takata, K.
24 B. D. Song and W. P. Jencks, J. Am. Chem. Soc., 109, 3160
(1987).
Hasegawa, C. Shimasaki, and E. Tsukurimichi, Chem. Lett., 1992,
1433.
25 T. W. Bentley, C. T. Bowen, D. H. Morten, and P. V. R.
Scheyer, J. Am. Chem. Soc., 103, 5466 (1981).
3
A. F. Clifford, J. L. Howell, and D. L. Wooton, J. Fluorine
Chem., 11, 433 (1978).
26 O. Rogne, J. Chem. Soc. B, 1969, 663.
4
T. Yoshimura, K. Hamada, M. Imado, K. Hamata, K.
27 T. Yoshimura, E. Tsukurimichi, H. Kita, H. Fujii, and C.
Shimasaki, Bull. Chem. Soc. Jpn., 63, 1764 (1990).
28 a) L. C. Manege, T. Ueda, and M. Hojo, Bull. Chem. Soc.
Jpn., 71, 589 (1998). b) L. C. Manege, T. Ueda, M. Hojo, and M.
Fujio, J. Chem. Soc., Perkin Trans. 2, 1998, 1961.
29 M. W. Wong, Chem. Phys. Lett., 256, 391 (1996).
30 A. E. Reed, L. A. Cartiss, and F. Weinhold, Chem. Rev., 88,
899 (1988).
31 A. E. Reed and P. von R. Schleger, J. Am. Chem. Soc., 112,
1434 (1990).
32 N. S. Isaacs, “Physical Organic Chemistry,” Longman Sci-
entific & Technical (1987), p. 475.
Tomoda, T. Fujii, H. Morita, C. Shimasaki, S. Ono, E.
Tsukurimichi, N. Furukawa, and T. Kimura, J. Org. Chem., 62,
3802 (1997).
5
N. Furukawa, E. Horn, T. Yoshimura, T. Fujii, M. Ohkubo,
S. Ono, H. Morita, and E. R. T. Tiekink, Z. Kristallogr. NCS, 213
(1), 53 (1998).
6
T. Yoshimura, E. Takata, T. Miyake, C. Shimasaki, K.
Hasegawa, and E. Tsukurimichi, Chem. Lett., 1992, 2213.
M. Ohkubo, T. Fujii, S. Ono, H. Morita, T. Yoshimura, E.
Horn, and S. Sato, Chem. Lett., 1997, 153.
T. Yoshimura, M. Ohkubo, T. Fujii, H. Kita, Y. Wakai, S.
7
8
Ono, H. Morita, C. Shimasaki, and E. Horn, Bull. Chem. Soc.
Jpn., 71, 1629 (1998).
33 N. S. Isaacs, “Physical Organic Chemistry,” Longman Sci-
entific & Technical (1987), p. 340.
9
E. Horn, T. Dong, T. Fujii, T.Yoshimura, and C. Shimasaki,
34 N. S. Isaacs, “Physical Organic Chemistry,” Longman Sci-
entific & Technical (1987), p. 351.
Z. Kristallogr. NCS, 215(3), 357 (2000).
10 M. E. Aberlin and C. A. Bunton, J. Org. Chem., 35, 1825
(1970).
11 T. Yoshimura, T. Dong, T. Fujii, M. Ohkubo, M. Sakuta, Y.
Wakai, S. Ono, H. Morita, and C. Shimasaki, Bull. Chem. Soc.
Jpn., 73, 957 (2000).
12 T. Okuyama, H. Takano, and K. Senda, Bull. Chem. Soc.
Jpn., 69, 2639 (1996).
13 N. S. Isaacs, “Physical Organic Chemistry,” Longman Sci-
entific & Technical (1987), p. 470.
35 R. L. Schowen, Prog. Phys. Org. Chem., 9, 275 (1972).
36 E. H. Cordes and H. G. Bull, Chem. Rev., 74, 581 (1974).
37 Gaussian 94 (Revision E. 1), M. J. Frisch, G. W. Trucks, H.
B. Schlegel, P. M. W. Gill, B. G. Johnson, M. A. Robb, J. R.
Cheeseman, T. A. Keith, G. A. Petersson, J. A. Montgomery, K.
Raghavachari, M. A. Al-Laham, V. G. zakrzewski, J. V. Ortiz, J. B.
Foresman, J. Cioslowski, B. B. Stefanov, A. Nanayakkara, M.
Challacombe, C. Y. Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L.
Andres, E. S. Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S.
Binkley, D. J. Defrees, J. Baker, J. P. Stewart, M. Head-Gordon, C.
Gonzalez, and J. A. Pople, Gaussian, Inc., Pittsburgh PA, 1995.
14 E. Ciuffarin and L. Senatore, Tetrahedron Lett., 17, 1635
(1974).
15 O. Rogne, J. Chem. Soc. B, 1968, 1294.