Structure and mechanism of hydrolysis of diaryl(acylamino)(chloro)-λ4-sulfanes and diaryl(acylamino)sulfonium salts

Article abstract of DOI:10.1039/b008156f

Aryl (methylaminocarbonylaryl) sulfides were converted by t-BuOCl to diaryl(acylamino)(chloro)-λ⁴-sulfanes or the corresponding diaryl(acylamino)sulfonium chlorides depending on the substituent of the S-aryl group. 1H NMR data showed that chloro-λ⁴-sulfanes exist only in CDCl3 and DMSO-d6 solvents, whereas in CD3OD complete ionic dissociation takes place, leading to sulfonium chlorides. Both the chemical shifts of 1H NMR signals and NOE data suggest that chloro-λ⁴-sulfanes and sulfonium salts having an o-MeO, o-Cl or o-Me substituent on the phenyl ring assume a skew conformation, whereas the aryl ring in compounds without an ortho-substituent can rotate practically free about the S-C(1') axis. In o-MeO-substituted derivatives there exists an equatorial 1,4 type S...O close contact. Sulfonium salts with axial 1,5 type S...O close contacts involving neighbouring COOMe, CONHMe, COMe or NO2 groups occur in butterfly conformation, like spiro-λ⁴-sulfanes. There is a correlation between the 15N chemical shift of the amide-nitrogen and the elongation of the S-N covalent bond, by which the interdepending S-N, S-Cl and S...O bonds can be characterized. Effective intermolecular S...O interaction was detected between the sulfonium centre and solvent molecules having a negatively polarized oxygen atom. The hydrolysis of sulfonium salts yielding sulfoxides was investigated by a kinetic method in 98:2 (v/v) dioxane-water mixture and in water. On the basis of medium, substituent (ρ + 1.03), steric, salt and kinetic isotope effects detailed mechanisms involving a hydroxy-λ⁴-sulfane intrmediate are proposed. The more reactive sulfonium salts with a five-membered hetero ring are hydrolyzed by water, whereas the sulfonium centre of the less reactive analogues with a six-membered ring is attacked only by OH^- ions.

Full text of DOI:10.1039/b008156f

View Article Online / Journal Homepage / Table of Contents for this issue
Structure and mechanism of hydrolysis of diaryl(acylamino)-
(chloro)-ꢀ4-sulfanes and diaryl(acylamino)sulfonium salts†
Péter Nagy,a Antal Csámpai,b Dénes Szabó,a Jeno´´ Varga,a Veronika Harmat,c Ferenc Ruff*a
and Árpád Kucsmana
2
a Department of Organic Chemistry, b Department of General and Inorganic Chemistry,
c Department of Theoretical Chemistry, L. Eötvös University, PO Box 32, H-1518,
Budapest 112, Hungary. E-mail: szabod@szerves.chem.elte.hu; Fax: 36 1 3722 620;
Tel: 36 1 2090 555
Received (in Cambridge, UK) 10th October 2000, Accepted 13th December 2000
First published as an Advance Article on the web 26th January 2001
Aryl (methylaminocarbonylaryl) sulfides were converted by t-BuOCl to diaryl(acylamino)(chloro)-λ4-sulfanes or the
corresponding diaryl(acylamino)sulfonium chlorides depending on the substituent of the S-aryl group. 1H NMR
data showed that chloro-λ4-sulfanes exist only in CDCl3 and DMSO-d6 solvents, whereas in CD3OD complete ionic
dissociation takes place, leading to sulfonium chlorides. Both the chemical shifts of 1H NMR signals and NOE data
suggest that chloro-λ4-sulfanes and sulfonium salts having an o-MeO, o-Cl or o-Me substituent on the phenyl ring
assume a skew conformation, whereas the aryl ring in compounds without an ortho-substituent can rotate practically
free about the S–C(1Ј) axis. In o-MeO-substituted derivatives there exists an equatorial 1,4 type S ؒ ؒ ؒ O close contact.
Sulfonium salts with axial 1,5 type S ؒ ؒ ؒ O close contacts involving neighbouring COOMe, CONHMe, COMe or
NO2 groups occur in butterfly conformation, like spiro-λ4-sulfanes. There is a correlation between the 15N chemical
shift of the amide-nitrogen and the elongation of the S–N covalent bond, by which the interdepending S–N, S–Cl and
S ؒ ؒ ؒ O bonds can be characterized. Effective intermolecular S ؒ ؒ ؒ O interaction was detected between the sulfonium
centre and solvent molecules having a negatively polarized oxygen atom. The hydrolysis of sulfonium salts yielding
sulfoxides was investigated by a kinetic method in 98:2 (v/v) dioxane–water mixture and in water. On the basis of
medium, substituent (ρ ϩ1.03), steric, salt and kinetic isotope effects detailed mechanisms involving a hydroxy-λ4-
sulfane intermediate are proposed. The more reactive sulfonium salts with a five-membered hetero ring are
hydrolyzed by water, whereas the sulfonium centre of the less reactive analogues with a six-membered ring is
attacked only by OHϪ ions.
expected for (acylamino)(chloro)-λ4-sulfanes (e.g. 2a 3a; see
Introduction
Scheme 1).
In the late 1970s cyclic diaryl(acyloxy)(chloro)-λ4-sulfanes (e.g.
We found earlier5 that ortho- (or peri-) methoxycarbonyl-
2i with O instead of NMe; see Scheme 1) were described as
substituted (acylaminoaryl) aryl sulfides were not converted to
possible reactive intermediates formed from the corresponding
(acylamino)(chloro)-λ4-sulfanes by chlorinating reagents, but
diaryl sulfides by treatment with chlorinating agents.1 Later,
to ionized (acylamino)sulfonium chlorides (e.g. 1f→3f, 6→7,
several stable representatives of (alkoxy)(chloro)-λ4-sulfanes2–4
9f→11f; see Schemes 1–3), in which the positive sulfonium
and diaryl(acylamino)(chloro)-λ4-sulfanes5 were prepared and
centre is stabilized by S ؒ ؒ ؒ O close contact involving the
their structures which exhibited trigonal bipyramidal geometry,
carbonyl-oxygen of the neighbouring substituent (i.e. the dis-
sociation of chloride ion may be attributed to neighbouring-
group participation).
In the first part of this paper we report on how electronic
were determined by X-ray diffraction. As is expected from
earlier studies on “nonsymmetric” λ4-sulfanes,2–11 the axial
bonds in chloro-λ4-sulfanes proved to be of a different char-
acter. For example,5 the S–N bonds are practically covalent
factors, i.e. the different ortho, para and peri substituents,
single bonds in 2a and 10a (1.727 and 1.675 Å, respectively; see
Schemes 1 and 3), whereas the S–Cl bonds correspond to weak
and various solvents control the formation of cyclic diaryl-
(acylamino)(chloro)-λ4-sulfanes and corresponding diaryl-
and strongly polarized hypervalent bonds. The latter are signifi-
(acylamino)sulfonium chlorides from the precursor sulfides.
cantly longer (2.690 and 3.101 Å) than the covalent S–Cl bonds
1
Using H and 15N NMR spectroscopy data including DNOE
(2.05–2.20 Å), but shorter than the sum of the van der Waals
we attempt to determine the constitution and the conform-
radii (3.65 Å). Data also indicate that the relative strengths of
S–N and S–Cl bonds are influenced by steric factors, e.g. by the
ation of products formed in the chlorination reaction. In
the second part we describe kinetic investigations on the
nature of the S-aryl substituents.
hydrolysis of chloro-λ4-sulfanes and sulfonium chlorides
Owing to the polarized structure, chloro-λ4-sulfanes may
obtained, and propose a detailed mechanism for reactions
yielding sulfoxides.
readily undergo ionic dissociation yielding sulfonium chlorides.
1H NMR investigations showed that compounds exhibiting
an (alkoxy)(chloro)-λ4-sulfane structure in the solid state may
dissociate in protic solvents (e.g. in CD3OD) to give (alkoxy)-
Results and discussion
Model compounds
sulfonium chlorides.2–4 A similar dissociation may also be
Diaryl (acylamino)(chloro)-λ4-sulfanes (2, 10) and diaryl(acyl-
amino)sulfonium chlorides (3, 7, 11) or perchlorates (4, 12)
† 1H and 13C NMR data are available as supplementary data. For direct
electronic access see http://www.rsc.org/suppdata/p2/b0/b008156f/
DOI: 10.1039/b008156f
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349
This journal is © The Royal Society of Chemistry 2001
339

View Article Online
Scheme 1 Synthesis and hydrolysis of diaryl(acylamino)(chloro)-λ4-sulfanes 2a–2h and 2j–2q, and diaryl(acylamino)sulfonium salts 3a–3h, 3j–3q,
4a and 4b.
NMR studies
Product analysis. NMR investigations of products obtained
from sulfides 1a–1q, 6, 9a and 9f by chlorination were carried
out in CDCl3, DMSO-d6 and CD3OD solvents. The obtained
NMR parameters were compared to those of the nonsymmetric
and symmetric spiro-λ4-sulfanes 14 and 15 (Fig. 1) to establish
the solvent-dependent transformation between λ4-sulfanes and
sulfonium salts (2 3, 10 11) and the conformation of these
compounds. Detailed assignations for all novel compounds are
provided as supplementary material.
Previous observations2,4,5 showed that the characteristic
downfield shift of the H-6 ortho-proton lying near to the hyper-
valent S–Cl bond may be an indicator of the existence of the
chloro-λ4-sulfane structure (see Fig. 1). Data in Table 1 suggest
that chlorination products obtained from sulfides 1a–1d, 1j–1q
and 9a exhibit chloro-λ4-sulfane structures 2a–2d, 2j–2q and
10a, respectively, in dipolar aprotic DMSO-d6 [δH-6 (or δH-2
for 10a) = 8.9–9.2 ppm].
The H-6 signal was significantly upfield shifted in DMSO-d6
for other products obtained from sulfides 1e–1h, 6 and 9f (δH-6
or δH-2 = 8.1–8.8 ppm), for spiro-λ4-sulfane 14 (formed from
1i) and its N-symmetric analogue 15 (prepared from 1g through
3g) (δH-6 = 8.3 and 7.5 ppm, respectively). Thus the former
products were assigned as sulfonium chlorides 3e–3h, 7 and 11f,
respectively, stabilized by axial S ؒ ؒ ؒ O close contact. The
upfield shift of H-6 signals can be explained by the fact that in
sulfonium structures the H-6 atom (in contrast with the corre-
sponding chloro-λ4-sulfanes) is not deshielded by an axial S–Cl
hypervalent bond (Fig. 1). A similar upfield shift was observed
in DMSO-d6 for sulfonium perchlorates 4a–4b and hydroxy-
sulfonium chloride 5bؒHCl (δH-6 = 8.4, 8.5 and 7.9 ppm,
respectively), all having a positive sulfur atom but a somewhat
different conformation (see later).
The same results were obtained when CDCl3 as a less polar
solvent was used (Table 1), with the exception that product
formed from 1e carrying an ortho-NO2 substituent proved to be
chloro-λ4-sulfane 2e (δH-6 = 9.6 ppm) instead of sulfonium
chloride 3e. That deviation may be ascribed to the ortho-NO2
neighbouring group which is less effective at forming an S ؒ ؒ ؒ O
close contact than the carbonyl group. Products formed from
1m–1q could not be investigated in CDCl3 because instant-
aneous hydrolysis yielding sulfoxide took place, caused by the
water impurity of the solvent.
Scheme 2 Synthesis and hydrolysis of diaryl(acylamino)sulfonium
salts 7.
were obtained from aryl (methylaminocarbonylaryl) sulfides
substituted differently on the phenyl (1, 9) or naphthyl ring (6),
as shown in Schemes 1–3.
The following types of precursor sulfides were chosen: (i)
derivatives of phenyl (2-methylaminocarbonylphenyl) sulfide
(1a) carrying an “inert” (OMe, Me, Cl) or an “active” (NO2,
COOMe, CONHMe, COMe, COOH) ortho-substituent (1b–1d
and 1e–1i, respectively); the “active” groups may be involved in
the formation of intramolecular S ؒ ؒ ؒ O close contact with a
five-membered “ring”; (ii) the para-isomers of the former com-
pounds (1j–1q); (iii) compound 6 in which the COOMe sub-
stituent is in the peri-position of a naphthalene ring (possible
formation of a six-membered “ring” with S ؒ ؒ ؒ O close con-
tact); (iv) phenyl (8-methylaminocarbonylnaphthyl) sulfide (9a)
and its derivative carrying an o-COOMe group (9f). Com-
pounds of (i)–(iii) and (iv) type were converted in a chlorination
reaction to cyclic products (chloro-λ4-sulfanes or sulfonium
chlorides) having
a five-membered 1,2-thiazoline or six-
membered 1,2-thiazine ring, respectively. Under similar condi-
tions sulfide 1i yielded the spiro-λ4-sulfane 149 (see Fig. 1).
Spiro-λ4-sulfane 15 was prepared from 3g by treatment with
KH.12 All λ4-sulfanes and sulfonium salts gave sulfoxides (5, 8,
13) on hydrolysis.
In dipolar protic CD3OD solvent only sulfonium chlorides
3a–3h, 3j–3q (δH-6 = 8.2–8.5 ppm) and 7, 11a, 11f [δH-6 (or
δH-2 for 11a and 11f) = 8.7–8.9 ppm] were detected, owing to
the ionic dissociation of chloro-λ4-sulfanes formed primarily
(Table 1).
340
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349

View Article Online
Table 1 Selected data obtained by 1H NMR,a,b 15N NMRa,c and X-ray analysis for products prepared from sulfides 1a–1q by chlorination
δH-6
δH-6Ј
δ15N
Sulfide
Product
(C)
(D)
(M)
(C)
(D)
(M)
(C)
(D)
(M)
S–N/Å
1a
1a
1a
1b
1b
1b
1b
1c
1c
1d
1d
1e
1e
1f
1g
1g
1h
1i
1j
1j
2a
3a
4a
2b
3b
4b
5bؒHCl
2c
9.66
9.07
7.77
7.75
125.0
125.1
1.7275
1.6785
8.50
8.26
7.82
7.81
109.9
111.3
8.19
9.56
8.41
8.87
7.71
7.12
7.80
7.39
110.8
121.0
116.0
118.8
8.35
8.24
7.73
7.55
7.54
7.11
104.5
104.5
8.33
9.71
9.68
9.63
8.46
7.89
9.13
7.75
6.39
6.47
6.69
7.42
7.19
6.63
100.8
130.0
127.8
128.8
113.3
126.4
125.7
3c
8.51
8.54
7.04
7.12
110.2
110.2
2d
3d
2e
3e
3f
9.21
6.70
8.64
8.41
8.28
7.54
8.12
8.29
8.86
8.31
8.33
8.31
7.57
8.51
8.38
7.44
7.56
7.63
7.54
7.65
7.47
7.70
7.80
7.68
7.72
7.57
7.69
7.48
114.3
109.5
111.9
169.2
108.0
129.3
122.2
101.2
101.5
101.2
170.6
101.9
122.7
9.06
8.43
7.52
8.74
8.39
9.43
6.78
7.34
7.52
6.97
7.22
7.72
116.2
114.9
169.2
115.7
131.7
122.7
3g
15
3h
14
2j
1.6855
1.899d
1.7349
3j
8.28
8.32
8.35
8.43
8.35
8.36
8.33
7.77
7.75
7.82
8.08
8.27
8.29
8.18
110.2
110.4
110.1
110.9
110.6
110.9
110.6
1k
1k
1l
2k
3k
2l
9.52
9.49
8.92
9.08
9.22
9.02
9.15
9.05
9.01
7.68
7.71
7.57
7.75
7.99
7.87
7.89
7.80
7.85
121.6
126.0
122.5
123.6
127.1
124.5
123.8
124.3
124.1
1l
3l
1m
1m
1n
1n
1o
1o
1p
1p
1q
1q
6
2m
3m
2n
3n
2o
3o
2p
3p
2q
3q
7e
8.35
8.72
8.26
7.72
109.6
114.5
8.74
9.91
8.51
9.03
7.10
7.98
7.74
7.71
122.5
92.8
116.1
92.5
1.6855
1.6755
9a
9a
9f
10af
11af
11ff
8.67
8.95
7.67
7.70
88.8
86.7
9.23
8.81
7.41
7.62
88.1
89.8
1.6625
a Solvents for NMR measurements: (C) CDCl3, (D) DMSO-d6, (M) CD3OD. b The H-6 and H-6Ј signals are doublets with coupling constants
of 7.9 0.5 Hz. c For N(Me) in hetero ring. d For the N,NЈ-diisopropyl analogue of 15 from ref. 13. e The numbering of atoms is changed in 7
(from H-6Ј to H-2Ј). f The numbering of atoms is changed in 10a, 11a and 11f (from H-6 to H-2).
Scheme 3 Synthesis and hydrolysis of diaryl(acylamino)(chloro)-λ4-sulfane 10a and diaryl(acylamino)sulfonium salts 11a, 11f and 12a.
Conformation. 1H NMR data suggest that chloro-λ4-sulfanes
2c and 2d with o-Me and o-Cl substituents, respectively have a
skew conformation (Fig. 1), meaning that the phenyl ring is
nearly coplanar with the C(1)–S–C(1Ј) plane, whereas the con-
densed aromatic ring is perpendicular to it. The o-Me and o-Cl
substituents hinder the rotation about the S–C(1Ј) bond, there-
fore the anisotropic shielding effect of the aromatic ring con-
densed to the heteroring gives rise to a considerable upfield shift
of the H-6Ј signals as compared to those of the p-substituted
counterparts 2k and 2l (∆δH-6Ј = 1.24–1.29 ppm in CDCl3 and
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349
341

View Article Online
Table 2 NOE data obtained for diaryl(acylamino)(chloro)-λ4-sulfanes (2 and 10), diaryl(acylamino)sulfonium salts (3, 4, 5 and 11) and spiro-λ4-
sulfanes (14 and 15)a
NOE (%)
Torsion angle
Proposed
Compound
Solvent
H(6Ј)–NMe
H(6Ј)–H(6)
N–S–C(1Ј)–C(6Ј)/Њ
conformation
b
2a
C/D
C/D
C/D
C/D
C
D
M
M
M
M
D/M
C/D/M
M
C/D/M
C/D/M
D/M
C/D/M
C/D
M
1.7/1.4
1.1/0.8
1.2/1.0
No effect
2.5/1.7
4.8/4.3
5.4/5.0
89 (ref. 5)
2b
Skew
Skew
Skew
2c
2d
1.0/0.8
c
2e
Skewd
b
2j–2q
3a
0.8–1.5
1.1
0.8
0.9
0.8
5.9/6.2
7.3–7.6/6.8–7.4/5.9–6.2
0.7–1.0
1.1/1.5/0.9
1.2/0.8/0.9
No effect
9.4/7.4/6.7
1.5/1.2
No effect
No effect
1.4
1.2
1.3
No effect
No effect
No effect
No effect
2.5/2.0/1.6
2.7/1.8
No effect
No effect
No effect
No effect
No effect
No effect
b
3b
Skew
Skew
Skew
Butterfly
3c
3d
3e
3f–3h
3j–3q
4a
11 in 3g (ref. 5)
Butterfly
b
b
37 (ref. 5)
4b
Skew
Skew
5bؒHCl
95.5e
31 (ref. 5)
63 (ref. 5)
7f
Butterfly-like
b
10ag
11ag
11fg
14
b
1.1
C/D/M
C/D/M
C/D/M
9.3/8.8/7.8
7.2/6.5/6.5
8.2/8.0/8.1
3 (ref. 5)
Butterfly
Butterfly
Butterfly
11 (ref. 9)
15
6.2 (ref. 13)h
a Solvents for measurements: (C) CDCl3, (D) DMSO-d6, (M) CD3OD. b Practically free rotation about the S–C(1Ј) bond supported by the isotropy of
H(2Ј) and H(6Ј) signals. c NOE cannot be measured because of decomposition. d Established only on the basis of the considerable upfield shift of
H(6Ј) signal (6.69 ppm, see Table 1). e Torsion angle O(2)–S–C(1Ј)–C(6Ј). f The numbering of atoms is changed in 7 (from H-6Ј to H-2Ј). g The
numbering of atoms is changed in 10a, 11a and 11f (from H-6 to H-2). h For the N,N-diisopropyl analogue of 15.
butterfly conformation (Fig. 1). This means that both aromatic
rings are nearly perpendicular to the C(1)–S–C(1Ј) plane.
Suggestions about the different conformations of chloro-λ4-
sulfanes and sulfonium chlorides, based on δ(H-6) and δ(H-6Ј)
shifts, were confirmed by studies on NOE between H(6Ј)–NMe
and H(6)–H(6Ј) protons (Table 2).
NOE was found between H-6Ј and NMe for compounds with
a butterfly conformation, whereas in the case of a skew
conformation NOE was found between H-6 and H-6Ј. The
observed NOE values vary according to the actual orientation
of the non-condensed aromatic ring or the freedom of the
rotation about the S–C(1Ј) bond. Data indicate that sulfonium
salts 4a and o-methoxy-substituted 3b, 4b and 5bؒHCl take
up a skew shape as shown in Figs. 1 and 2. All these results are
in agreement with those obtained by X-ray diffraction. The
N–S–C(1Ј)–C(6Ј) torsion angle data indicative of H(6Ј)–N(Me)
proton distance are listed in Table 2 for some compounds
investigated also in the solid state.
NOE was not observed between H-6 and 2Ј-Me protons in
2c, and very small NOEs could be detected between H-6Ј and
NMe protons in the case of 2c and 2d. These observations
together with the upfield shielded H-6Ј signals and the marked
interaction of H-6 and H-6Ј (Table 2) give further support for
Fig. 1 The conformation of diaryl(acylamino)(chloro)-λ4-sulfanes 2b–
2e (for X and Y, see Scheme 1), diaryl(acylamino)sulfonium salts 3b
(ZϪ = ClϪ), 3e (NO2 instead of QCO), 3f (Q = OMe), 3g (Q = NHMe),
3h (Q = Me), 4b (ZϪ = ClO4Ϫ), 5bؒHCl, and spiro-λ4-sulfanes 14 and 15.
the rigid skew geometry of chloro-λ4-sulfanes with o-Me and
o-Cl substituents in the exo position, and suggest less hindered
or practically free rotation about the S–C(1Ј) bond for com-
pounds without ortho-substituents (e.g. 2a, 2j–2q, 4a, 10a).
In the case of chloro-λ4-sulfane 2b with an o-MeO group
(similarly to the analogous sulfonium chloride 3b, perchlorate
4b and hydroxysulfonium chloride 5bؒHCl) a smaller NOE
between H-6 and H-6Ј protons (1.7–2.5%) was observed than
for 2c and 2d (4.5–5.4%, Table 2), and the H-6Ј protons in the
former compounds (in contrast with 2c and 2d) do not experi-
ence a shielding effect from the condensed aromatic ring (see
Table 1). These phenomena may be explained by the polariz-
ation of the S–Cl bond, i.e. the enhanced sulfonium character
of 2b (caused by an equatorially directed, 1,4 type S ؒ ؒ ؒ O close
contact between the positive sulfur atom and the methoxy-
oxygen atom, see Figs. 1 and 2), associated with the widening of
1.04–1.05 ppm in DMSO-d6). In the para-substituted com-
pounds the rotation about the S–C(1) bond may be regarded
as practically free. A much smaller effect was observed for
the 2b–2j pair carrying an MeO group in the o/p position
(∆δH-6Ј = 0.60 ppm in CDCl3 and 0.31 ppm in DMSO-d6; see
explanation later).
The similarity of δH-6 and δH-6Ј values found for spiro-λ4-
sulfane 14 and sulfonium salts 3e–3h and 7 with S ؒ ؒ ؒ O inter-
action (Table 1) reveals that the sulfur atom in the latter com-
pounds should also exhibit trigonal bipyramidal geometry with
N and O hetero atoms in axial positions, and they assume a
342
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349

View Article Online
The correlation works also in the case of (acylamino)-
sulfonium salts without S ؒ ؒ ؒ O interaction. δ15N = 116.0 and
88.8 ppm were measured for 4a and 11a with S–N bond lengths
1.6785 and 1.651 Å5 (for 12a perchlorate), respectively. It is also
remarkable that the relatively low δ15N shift in CDCl3 (121.0
ppm) observed for o-MeO-substituted 2b, as compared to those
found for 2c–2e with o-Me, o-Cl and o-NO2 groups (130.0,
127.8 and 128.8 ppm, respectively), points to an enhanced
sulfonium character of 2b, as discussed above.
Data from X-ray diffraction and Table 1 also show that the
shorter the axial S–N bond (i.e. the smaller the δ15N shift) is,
the longer the axial S–Cl hypervalent bond and axial S ؒ ؒ ؒ O
contact incorporated in a five-membered “ring” may appear;
e.g. S–Cl is 2.6905 and 3.101 Å5 in 2a and 10a (with
δ15N = 125.1 and 92.5 ppm, respectively), and in a similar way
S ؒ ؒ ؒ O is 2.1329 (weak hypervalent S–O bond), 2.373,9 and
2.736 Å5 in 14, 3g and 11f (with δ15N = 129.3, 111.9 and 89.8
ppm, respectively). The not quite regular behaviour of com-
pound 7 (S ؒ ؒ ؒ O 2.408 Å,5 δ15N = 116.1 ppm) may be attributed
to the S ؒ ؒ ؒ O contact built in a six-membered “ring”.
We may conclude that axial hypervalent S–Cl bond and
S ؒ ؒ ؒ O close contact may also be characterized by using δ15N
data. Thus it seems very likely that ortho- or para-substitution
in the aromatic ring of chloro-λ4-sulfanes does not affect the
S–Cl bond (except for the o-MeO group). It is also revealed that
there is no significant difference in S ؒ ؒ ؒ O contacts involving
o-COOMe, o-CONHMe or o-COMe neighbouring groups
(Table 1).
Fig. 2 Perspective representation of sulfoxide-hydrochloride 5bؒHCl
molecule. Selected bond lengths (Å) and angles (Њ): S(1)–O(1) 1.587(4),
S(1) ؒ ؒ ؒ O(2) 2.411(4), S(1) ؒ ؒ ؒ O(3) 2.764(4), S(1)–C(1) 1.770(6),
S(1)–C(1Ј) 1.777(5), C(7)–O(2) 1.229(6), C(2Ј)–O(3) 1.357(7),
θ(O1–S1 ؒ ؒ ؒ O2) 178.9(2), θ1(O1–S1 ؒ ؒ ؒ O3) 97.9(2), θ2(O1–S1 ؒ ؒ ؒ C1)
100.4(2), θ3(O1–S1 ؒ ؒ ؒ C1Ј) 98.1(2), θ4(C1–S1–C1Ј) 102.8(3), θ5(S1–
C1Ј–C2Ј) 115.2(4), θ6(C1Ј–C2Ј–O3) 117.0(5), φ1(S1–C1–C2–C7) 5.8(6),
φ2(C1–C2–C7–O2) 7.7(7), φ3(C2–C7–O2 ؒ ؒ ؒ S1) Ϫ12.8(5), φ4(C7–
O2 ؒ ؒ ؒ S1–C1) 12.7(4), φ5(O2 ؒ ؒ ؒ S1–C1–C2) Ϫ9.2(4), φ6(O1–S1–C1–
C2) 171.0(4), φ7(O1–S1–C1Ј–C6Ј) 83.4(5), φ8(C1Ј–S1–C1–C2) Ϫ88.1(4),
φ9(C1–S1–C1Ј–C6Ј) Ϫ19.2(5), φ10(S1–C1Ј–C2Ј–O3) Ϫ177.5(5).
The neighbouring-group effect of the carbamoyl group may
be better understood if we compare compounds 3g and 3o with
a CONHMe substituent in ortho and para positions, respect-
ively. The significant downfield shift of the 15N(H) signal in 3g
as compared to that in 3o (108.9 and 103.6 ppm, respectively,
in DMSO-d6) points to a more effective delocalization of the
N(H) lone pair towards the carbonyl-oxygen in the former case,
promoting the formation of S ؒ ؒ ؒ O close contact.
the C(1)–S–C(1Ј) bond angle characteristic of sulfonium struc-
ture (cf. data 82Њ, 102Њ and 104Њ in 2a, 4a and 5bؒHCl, respect-
ively). Nevertheless, the small NOEs between H-6Ј and NMe
protons (Table 2) point to a skew conformation of 2b and the
other o-methoxy-substituted compounds (4b and 5bؒHCl).
S–N, S–Cl and S ؒ ؒ ؒ O bonding. Further information about
the axial NS–Cl and NSϩ ؒ ؒ ؒ O bond systems was gained from
the comparison of the 15N-chemical shifts of amide nitrogen in
the investigated models with that of the reference spiro-λ4-
sulfanes 14 and 15. Earlier X-ray diffraction studies revealed5,9
that the strength of axial S–N bonds (reflected in different
bond lengths) in spirocyclic (acylamino)-λ4-sulfanes, mono-
cyclic (acylamino)(chloro)-λ4-sulfanes and in the corresponding
(acylamino)sulfonium salts (stabilized by S ؒ ؒ ؒ O close contact)
depends on the relative electronegativities (nucleophilicities) of
the axial heteroatoms. The weakest (longest) S–N (hypervalent)
bond was found in the N,N-diisopropyl analog of symmetric
bis(acylamino)spiro-λ4-sulfane 1513 (1.898 Å) and the strongest
(shortest) S–N (covalent) bond in the sulfonium salt 11f5 (1.662
Å). The S–N bond length data of compounds 14,9 2a,5 7,5 3g9
and 10a5 lie between the former extreme values (1.734, 1.727,
1.685, 1.685, 1.675 Å, respectively). We may assume that the
more pronounced is N(amide) lone pair delocalization toward
the carbonyl oxygen (O᎐CR–NR ↔ϪO–CR᎐NR ϩ) in these
compounds the more the N–S bond is weakened. It is known,
however, from NMR studies14 that the shielding of the nitrogen
attached to an unsaturated carbon is highly influenced by the
degree of lone pair delocalization. Consequently, the increase
in S–N bond lengths should be reflected in the downfield shift
of 15N signal, as compared to that of NH3 (δ = 0). δ15N data in
Table 1 (169.2, 129.3, 125.1, 116.1, 111.9, 92.5 and 89.8 ppm
in DMSO-d6 for 15, 14, 2a, 7, 3g, 10a and 11f, respectively)
support the correlation outlined above which may be recom-
mended to characterize the S–N bonds in (acylamino)-λ4-
sulfanes and (acylamino)sulfonium salts. Here it should be
mentioned that the observed 15N chemical shifts represent a
wide region (~90 ppm), whereas the measured 13C resonances
of the lactam carbonyls lie in a substantially narrower range
(~4 ppm).
Solvent effect. We discussed above that the actual structure
of products formed from sulfides by chlorination depends
on the solvent used. In addition, the solvent effect observed for
15N signals gives further insight into the factors governing the
N–S–Cl and N–Sϩ ؒ ؒ ؒ O bonding systems.
δ15N data in Table 1 show that CDCl3 solvent is the most
favourable for stabilizing both the S–Cl hypervalent bond in
chloro-λ4-sulfanes (2a–2e and 2j–2l) and intramolecular S ؒ ؒ ؒ O
close contact in sulfonium chlorides having an ortho or peri
neighbouring group (3f–3h and 7). Thus S–N bonds in these
compounds are somewhat weaker in apolar CDCl3 than in
polar DMSO-d6. The latter solvent promotes the polariz-
ation of S–Cl and S ؒ ؒ ؒ O bonding explaining e.g. that the
chlorination product of o-NO2-substituted sulfide 1e occurs in
DMSO-d6 as a dissociated salt (3e) instead of a neutral chloro-
λ4-sulfane (2e). For the same reason, the o-CONHMe group in
3g is less effective at forming an S ؒ ؒ ؒ O close contact in DMSO-
d6 than in CDCl3 [δ15N(H) = 108.9/118.4 ppm, respectively].
The 15N-resonances of perchlorates 4a and 4b in CDCl3
(110.8 and 100.8 ppm, respectively) suggest that in apolar
medium even the perchlorate anion can be involved in S ؒ ؒ ؒ O
contact, elongating the S–N bond in 4a. As a consequence
of the S ؒ ؒ ؒ O interaction from the equatorial direction, the
o-methoxy group in 4b keeps the perchlorate anion away
from the positive sulfur atom (see Fig. 1), and the enhanced
sulfonium character pertains to the shortening of the S–N
bond. On the other hand, the considerably downfield shifted
15N-resonances of 4a and 4b measured in DMSO-d6 (116.0 and
113.3 ppm, respectively) point to a marked interaction of the
sulfonium centre with DMSO-d6 molecules having a nega-
tively polarized oxygen atom. In o-methoxy-substituted 4b the
equatorial S ؒ ؒ ؒ O close contact may be responsible for
the slightly weaker interaction with DMSO-d6 molecules.
᎐
᎐
2
2
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349
343

View Article Online
Owing to the dissociation of the S–Cl bond, chloro-λ4-
sulfanes 2a–2e and 2j–2q occur in the dipolar protic solvent
CD3OD as sulfonium chlorides 3a–3e and 3j–3q, respectively.
The 15N-resonances of the latter salts and 4a in CD3OD lie in
a narrow region at about 110 ppm. In these compounds the
sulfonium centre is not involved in intramolecular S ؒ ؒ ؒ O
interaction, so it is easily accessible to the solvent molecules
with negatively polarized oxygen atoms. On the other hand, the
comparison of 15N-NMR shifts of o-methoxy-derivatives 3b
and 4b having the same sulfonium cation (104.5 ppm) with that
of p-methoxy-derivative 3j (110.2 ppm) reveals that the 1,4 type
intramolecular S ؒ ؒ ؒ O contact in the equatorial position is
preserved in CD3OD, too, shielding the sulfonium centre from
the solvent molecules.
Table 3 Effects of the initial concentration of the reactant and that
of the water content of the solvent on the rate constant of hydrolysis of
diaryl(acylamino)sulfonium salts 3a, 4a and 7 at 25 ЊC
Initial reactant
concentration/
mol dmϪ3
k/10Ϫ4 sϪ1
Dioxane–H2O
(v/v)
3a
4a
7
98:2
98:2
98:2
97:3
96:4
7.5 × 10Ϫ4
1.5 × 10Ϫ4
3.0 × 10Ϫ5
1.5 × 10Ϫ4
1.5 × 10Ϫ4
4.69
7.77
10.5
20.2
—
0.868
1.41
1.81
9.67
—
—
1.02
—
4.04
9.26
Table 4 Effects of TsOH and LiClO4 on the rate constant of
hydrolysis of diaryl(acylamino)sulfonium chloride 3f in 98:2 (v/v)
dioxane–water solvent at 25Њ C
Similar but larger differences in δ15N shifts (about 101–110
ppm) were measured in CD3OD for the ortho–para isomeric
pairs with substituents forming an axial 1,5 type intramolecular
S ؒ ؒ ؒ O close contact in ortho-position (3e–3m, 3f–3n, 3g–3o
and 3h–3p) that may also be explained by the shielding of the
sulfonium centre from the solvent molecules in the ortho-
substituted derivatives. One may conclude that the interaction
of the sulfonium centre with the nucleophilic solvent molecules
weakens the S–N bond more effectively than intramolecular
S ؒ ؒ ؒ O close contact does in the axial position.
[TsOH]/mol dmϪ3
k/10Ϫ4 sϪ1
0
10Ϫ4
3.38a
6.10
8.90
9.90
12.1
13.4
5 × 10Ϫ 4
10Ϫ3
5 × 10Ϫ3
10Ϫ2
A much smaller solvent effect measured for naphthalene
derivatives with a six-membered 1,2-thiazine ring (10a and 11f)
may be attributed to steric factors. It is remarkable that the
nonsymmetric spiro-λ4-sulfane 14 and sulfonium salt 7 with
strong S ؒ ؒ ؒ O close contacts are slightly affected by a solvent
effect, whereas the structure of symmetric 15 is invariant to
solvent change.
a Rate constants 1.61 × 10Ϫ 4 and 1.52 × 10Ϫ4 sϪ1 were measured when
solvent contained 5 × 10Ϫ5 and 10Ϫ4 mol dmϪ3 LiClO4, respectively.
The rate of reaction of sulfonium chloride 3a is faster than
that of perchlorate 4a. The difference in reactivities may be
explained by the formation of ion aggregates in apolar solvents
containing only a small amount of water. Ion pairs are known
to have lower reactivity than free ions.18 In the given media the
ion-pairs of sulfonium chloride 3a can dissociate more easily
than those of 4a, because chloride ions are solvated better with
water than perchlorate ions in 4a. Thus the concentration of
free ions for 3a is higher in the solution than that for 4a. The
separation of ion-pairs is also promoted by the increase in
water content of the solvent. The rate constants for compound
4a increase therefore much faster with the concentration of
water than in the case of 3a (Table 3).
All (acylamino)sulfonium salts with a five-membered ring
(3a–3h, 3j–3q, 4a and 7) are hydrolysed only by water molecules.
The rate of the reaction of 3f is not proportional to the concen-
tration of the hydroxide ions, but increases slightly with the
increase in the concentration of acid added to the solvent. On
the other hand rate constants decrease when LiClO4 is added to
the reaction mixture (Table 4).
X-Ray diffraction studies
The molecular structure of diaryl(hydroxy)sulfonium chloride
5bؒHCl as determined by single-crystal X-ray diffraction is
shown in Fig. 2 together with selected interatomic distances and
angles. Some details can be found in the Experimental section.
It is, however, noteworthy to mention here that the skew-like
conformation of 5bؒHCl provides a means of forming both a
remarkably short, 1,5 type axial S ؒ ؒ ؒ O close contact [S ؒ ؒ ؒ O
2.411 Å, θ(O–S ؒ ؒ ؒ O) 179Њ] and an unusual 1,4 type equatorial
S ؒ ؒ ؒ O interaction [S ؒ ؒ ؒ O 2.764 Å, θ(O–S ؒ ؒ ؒ O) 98Њ]. The
value of the θ(Car–S–Car) bond angle (103Њ) resembles more a
tetrahedral (sulfonium) geometry than a trigonal-bipyramidal
(λ4-sulfane) arrangement.
Kinetics and mechanism of hydrolysis reactions
NMR measurements showed that (acylamino)(chloro)-λ4-
sulfanes are converted immediately to (acylamino)sulfonium
chlorides (e.g. 2a→3a), when dissolved in a protic solvent like
methanol. As a continuation of our earlier work,15–17 we
investigated by a kinetic method the hydrolysis of (acylamino)-
sulfonium chlorides (3, 7, 11) and perchlorates (4, 12) leading
to sulfoxides (5, 8, 13) in solvents containing water (cf. Schemes
1–3). The rates of hydrolysis were measured by a UV spectro-
photometric method in dioxane–water mixtures or in water
where water as nucleophile was always in a great excess
compared to the reactants (for details see the Experimental
section).
The solvation of lithium ions decreases the activity and the
nucleophilicity of water, and the number of ion aggregates
increases with increasing ionic strength, which may cause the
decrease of the rate constants. Acids may have two opposing
effects on the rate of the reactions. On the one hand the
solvation of protons, like that of lithium ions, may decrease
the activity of water and thus the rate of the reaction, on the
other the protonation of the acylamino group in the substrate
can increase the rate constants.
The activation parameters calculated from the temperature
dependence of the rate constants for 3a, 4a, 3c, 3f and 7 suggest
a bimolecular rate-determining step, especially as the entropies
of activation have high negative values. Significant primary
kinetic isotope effects were also observed in solvents containing
D2O, meaning that one of the O–H bonds in water molecules is
broken in the slow step when the transition state is developed
(Table 5).
Diaryl(acylamino)sulfonium salts with five-membered 1,2-
thiazoline ring. The hydrolyses of compounds 3a–3h, 3j–3q, 4a
and 7 proved to be so fast that their rates could be measured
by conventional methods only at a low concentration of water
[max. 2–4% (v/v)] in dioxane–water mixtures. The reactions fol-
lowed first-order kinetics up to 50–70% conversion, but the rate
constants increased significantly if the initial concentration of
the reactants decreased by more than one order of magnitude
(Table 3).
The substituent effect was investigated by comparing the rate
constant of 3a with those of para-substituted derivatives 3j–3q.
Electron-withdrawing substituents (e.g. NO2, Cl) increase the
rate, whereas electron-releasing groups (e.g. MeO, Me) exert an
opposite effect (Table 6).
344
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349

View Article Online
Table 5 Temperature dependence of the rate constants, activation parameters and deuterium solvent isotope effect for the hydrolysis of diaryl-
(acylamino)sulfonium chlorides 3a, 4a, 3c, 3f and 7 in dioxane–water mixtures
Temperature/ЊC
3aa
4ab
3ca
3fa
7b
k/10Ϫ4 sϪ1
20.0
22.5
25.0
25.0
27.5
30.0
6.15
6.86
7.77
2.31c
8.56
9.63
8.52
9.11
9.67
3.99c
10.2
10.7
3.89
4.60
5.38
—
6.35
7.47
2.42
2.88
3.38
1.09c
4.07
4.81
2.88
3.44
4.04
1.41c
4.79
5.68
∆H ‡/kJ molϪ1
30.5
Ϫ202
3.4
14.3
Ϫ255
2.4
45.5
Ϫ154
48.3
Ϫ149
3.1
47.3
Ϫ151
2.9
∆S ‡/J molϪ1 KϪ1
d
kH O/kD O
—
2
2
a Measured in 98:2 (v/v) dioxane–H2O. b Measured in 97:3 (v/v) dioxane–H2O. c Measured in dioxane–D2O mixtures of the same composition as in
dioxane–H2O. d Solvent isotope effects 3.0, 2.6, 2.8 and 2.6 were calculated from the rate constants given in Table 6 for compounds 3b, 3h, 3j and 3q,
respectively.
Table 6 Substituent effect for the hydrolysis of ortho- and para-
substituted diaryl(acylamino)sulfonium chlorides (3a–q) in 98:2 (v/v)
dioxane–water mixture at 25 ЊC
by the weak axial S ؒ ؒ ؒ O interaction developed between the
sulfonium centre and the nitro-oxygen atom20 (see the butterfly
conformation of 3e in Fig. 1) by which the nucleophilic attack
of water molecules on the sulfonium centre is somewhat
hindered.
k/10Ϫ4 sϪ1
As is known,5,11,20 more effective sulfur–oxygen close con-
tact can be formed in sulfur compounds having a carbonyl
group in the ortho-position of an aromatic ring bonded to the
sulfur atom. In the (acylamino)sulfonium salts studied the
carbonyl-oxygen of the o-COQ group (Q = OMe, NHMe, Me,
OϪ) approaches the sulfonium centre from the axial direction
(see the butterfly conformation of 3f, 3g, 3h and 14 in Fig. 1;
in the latter case the S ؒ ؒ ؒ O close contact is regarded as a
weak hypervalent bond). As a result, the positive sulfur atom
is more or less hindered from the nucleophilic attack of
water, depending on the relative nucleophilicity of the neigh-
bouring group. The stronger the S ؒ ؒ ؒ O close contact is, the
smaller the value which may be expected for the kortho/kpara
ratio. Indeed, data in Table 6 indicate that o-COOCH3,
Substituent
3b–h (ortho)
3j–q (para)
kortho/kpara
MeO
Me
Cl
NO2
3b
3c
3d
3e
3f
3g
3h
3.30a
5.38
32.1
31.3
3j
3k
3l
2.61a
4.97
13.3
1.3
1.1
2.4
3m
3n
3o
3p
3q
3ad
39.1
20.3
13.8
18.3
18.5a
7.7
0.80
0.17
0.005
0.15
—
COOMe
CONHMe
COCH3
COOH
3.38
0.066b
2.66a
CO2
14
0.0081c
0.001d
Ϫ
a Rate constants 1.11 × 10Ϫ4, 1.02 × 10Ϫ4, 9.22 × 10Ϫ5 and 7.08 × 10Ϫ4
s
Ϫ1 were measured in 98:2 (v/v) dioxane–D2O at 25 ЊC for 3b, 3h, 3j and
3q, respectively. b Calculated from the solvent dependence of the rate
constants given in ref. 17. c Calculated from the solvent dependence of
the rate constants given in ref. 16. d Rate constant for 3a was used to
Ϫ
o-COCH3, o-CONHCH3 and o-CO2 groups decrease the
Ϫ
calculate the kortho/kpara ratio. The σp Hammett constant of p-CO2 is
rate of the reaction to 1/6, 1/7, 1/200 and 1/1000 parts, respect-
ively, of those measured for the the para-substituted
compounds.
zero; the rate constant should be equal to that of the unsubstituted
compound 3a.
Starting from the above results we propose a detailed mech-
anism shown in Scheme 4 for the hydrolysis of (acylamino)-
sulfonium salts containing a five-membered 1,2-thiazoline ring
The Hammett equation gave a good correlation with the
σ constants taken from ref. 19 [eqn. (1)].
log k = (1.03 0.07)σ Ϫ (3.19 0.03) (r = 0.986) (1)
The positive value of the reaction constant (ρ ϩ1.03) is in
accordance with the nucleophilic attack of water on the posi-
tively charged sulfur atom in the rate-determining step, which
is promoted by electron-withdrawing substituents.
In order to study the steric effect of the neighbouring ortho-
substituents, we compared the rate constants of the hydrolysis
obtained for ortho- and para-substituted (acylamino)sulfonium
salts listed in Table 6. Because the electronic effect of the
ortho-substituents can be regarded as similar to that of the
para-analogues, the ratio of the rate constants of ortho- and
para-substituted compounds (kortho/kpara) should characterize
the steric effect of the ortho-substituent. If the ortho-substitu-
ents do not interact effectively with the sulfonium centre
(e.g. MeO, Me), the kortho/kpara ratio is near to one (3b/3j, 3c/3k),
showing that hydrolysis is not hindered sterically. (The
equatorial S ؒ ؒ ؒ O interaction in the skew-like conformation of
3b, as shown in Fig. 1, does not influence the rate of the
hydrolysis.) Surprisingly, compound 3d with an o-Cl group is
twice as reactive as the para-derivative 3l. This observation may
be interpreted by the strong electron-withdrawing effect of the
ortho-chlorine atom. Although the o-NO2 group in 3e is also
strongly electron-withdrawing, the kortho/kpara ratio is somewhat
less than unity in this case. The contradiction may be explained
Scheme 4 Mechanism of hydrolysis of diaryl(acylamino)sulfonium
salts 3a–3h and 3j–3q.
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349
345

View Article Online
Table 7 Rate constants for the hydrolysis of diaryl(acylamino)-
sulfonium salts 11a, 12a and 11f at different acid concentrations in
water at 25 ЊC
Table 8 pH dependence of the rate constant for the hydrolysis of
diaryl(acylamino)sulfonium chloride 11f in 0.2 mol dmϪ3 aqueous
acetate buffer (I = 0.074) at 25 ЊC; NaCl was added to the solvent to
maintain constant ionic strength
k/10Ϫ4 sϪ1
pH
k/10Ϫ4 sϪ1
[TsOH]/mol dmϪ3
11a
12a
11f
3.60
3.80
4.00
4.20
4.40
1.39
2.15
3.42
4.72
7.06
5 × 10Ϫ4
7.5 × 10Ϫ4
10Ϫ3
6.34
4.81
3.75
0.384
8.74
5.28
3.70
0.348
4.48a/0.746a,b
—
0.528
—
10Ϫ2
a Solvent contained 10Ϫ4 mol dmϪ3 TsOH. b In D2O.
Table 9 Effects of the buffer concentration and ionic strength on the
rate constant of the hydrolysis of diaryl(acylamino)sulfonium chloride
11f in aqueous acetate buffer (pH 4.00), at 25 ЊC; ionic strength was
adjusted by the addition of NaCl to the solvent
(formed occasionally from the corresponding chloro-λ4-sulfane
by fast ionic dissociation in a protic solvent).
In the rate-determining first step the sulfonium centre is
attacked by a water molecule from which a proton is split off
to give an (acylamino)(hydroxy)-λ4-sulfane intermediate (16).
Sulfoxide is produced in subsequent fast steps by O-deproton-
ation, N-protonation and splitting of the S–N bond. The attack
of the water molecule is obviously hindered by the ortho-
carbonyl group, which occupies the axial position about the
central sulfur atom of trigonal bipyramidal geometry.
k/10Ϫ4 sϪ1
[Buffer]/
mol dmϪ3
I = 0.074
I = 0.036
I = 0.009
0.2
0.1
3.42
—
3.88
3.77
3.84
—
—
4.17
0.05
—
Diaryl(acylamino)sulfonium salts with six-membered 1,2-
thiazine ring. Sulfonium salts having an acylamino group
incorporated in a six-membered hetero ring hydrolyze much
more slowly than their analogues with a five-membered ring.9
In the case of compounds 11a, 12a and 11f rates can be meas-
ured with conventional methods only in water. The hydrolysis
reactions, however, do not follow first-order kinetics in this
solvent; the calculated rate constants decrease with the progress
of the reaction.
The hydrolyses do not deviate from first-order kinetics if acid
is added to the reaction mixture in great excess. In water the
equilibrium between chloro-λ4-sulfane 10a and the correspond-
ing sulfonium salt 11a is obviously shifted completely to the
latter, and both chloride 11a and perchlorate 12a are dissoci-
ated to free ions. As expected, 11a and 12a show practically the
same reactivity; their rate constants are equal within experi-
mental error (Table 7).
Both chloride and perchlorate anions are solvated well, and
therefore the reversible formation of ion pairs need not to be
taken into account. The rate constants of the hydrolysis of
compounds 11a and 12a decrease when the concentration of
acid is increased, i.e. when the concentration of hydroxide ions
is decreased (Table 7). The log k versus pH plot is linear with a
slope near to unity [eqn. (2)].
hydroxide ions, and this is the reason why the hydrolysis does
not follow first-order kinetics in non-buffered media.
At constant ionic strength the rate constants obtained for the
hydrolysis of 11f are independent of the concentration of
the acetate ions in buffered media (Table 9), meaning that the
hydrolysis is not general base-catalysed.
If NaCl was added to the buffered reaction mixture the rate
constant of the hydrolysis decreased with the increase of the
ionic strength (Table 9). These data point to a reaction between
two oppositely charged ions, i.e. to that of the sulfonium and
hydroxide ions.
The values of the deuterium solvent isotope effect found for
sulfonium salts 11a, 11f and 12a with a six-membered hetero
ring are higher than six (Table 10). The primary kinetic isotope
effect observed suggests that proton-transfer is involved in the
rate-determining step.
The activation parameters calculated from the temper-
ature dependence of the rate constants (Table 10) are in accord-
ance with an unimolecular rate-determining step for the
hydrolysis of the above (acylamino)sulfonium salts. The
values of both activation enthalpies and entropies are higher
than those obtained for analogues with a five-membered hetero
ring.
On the basis of experimental data we propose the detailed
mechanism shown in Scheme 5 for the hydrolysis of acylamino-
sulfonium salts containing a six-membered 1,2-thiazine ring.
The first step is the fast nucleophilic addition of hydroxide ion
to the sulfonium centre, i.e. the formation of an (acylamino)-
(hydroxy)-λ4-sulfane (17) intermediate. We surmise that the
opening of the hetero ring is started by the rate-determining
deprotonation of the S-hydroxy group in 17. The electronic
shift from the negatively charged oxygen in 18, i.e. the form-
ation of a sulfinyl group with a strong S–O bond in sulfoxides
13a and 13f, can displace the acylamino group from the sulfur
atom, which is assisted by N-protonation.
log k = (0.96 0.03)pH Ϫ (6.32 0.09) (r = 0.999) (2)
The rate of the hydrolysis of o-COOMe-substituted 11f,
which reacts more slowly than 11a, cannot be measured in
acidic solutions. In solvent containing 10Ϫ2 mol dmϪ3 TsOH the
reaction hardly shows any measurable progress. In acetate
buffer, however, the rate constants can be determined, showing
an increase with the concentration of hydroxide ions (Table 8);
log k is a linear function of the pH [eqn. (3)].
log k = (0.88 0.03)pH Ϫ (7.00 0.12) (r = 0.998) (3)
Experimental
Data indicate that (acylamino)sulfonium salts with a six-
membered hetero ring react only with hydroxide ions which are
efficient enough to attack the sulfonium centre to cause the
loosening of the strong S–N covalent bond (see Table 1 and
ref. 9) in these compounds. (The N-methylacylamino group
incorporated into a six-membered ring is a poorer leaving
group than in a five-membered ring.) Acid formed as co-
product during the reaction changes the concentration of the
The preparation of chloro-λ4-sulfanes 2a and 10a, sulfonium
perchlorates 4a and 12a, sulfonium chlorides 3g, 7 and 11f, and
sulfides 1a, 1g, 1i, 6, 9a and 9f has been published previously.5,9
The synthesis of spiro-λ4-sulfanes 14 and 15 has also been
described.9,12 In the synthesis of diaryl sulfides 1b–1q
2,2Ј-dithiobenzoic acid (19) was first converted into bis-N-
methylcarbamoyl derivative 20 through the corresponding acid
346
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349

View Article Online
Table 10 Temperature dependence of the rate constants, activation parameters and kinetic isotope effects for the hydrolysis of diaryl(acylamino)-
sulfonium salts 11a, 12a and 11f in water
Temperature/ЊC
11aa
12aa
11fc
k/10Ϫ4 sϪ1
20.0
22.5
25.0
27.5
30.0
2.79
3.05
2.21
3.73
4.81/0.746b
6.40
3.98
5.28/0.796b
6.94
3.03
3.88
5.86
7.88
8.35
8.90
∆H ‡/kJ molϪ1
78.7
Ϫ45.9
6.5
77.1
Ϫ48.9
6.6
95.9
∆S ‡/J molϪ1 KϪ1
12.1
d
kH O/kD O
6.0e
2
2
a Water contained 7.5 × 10Ϫ4 mol dmϪ3 TsOH. b In D2O. c Measured in 0.2 mol dmϪ3 acetate buffer at pH 4.00. d Solvent contained 10Ϫ4 mol dmϪ3
TsOH. e Rate constants are given in Table 7.
Scheme 5 Mechanism of hydrolysis of diaryl(acylamino)sulfonium
Scheme 6 Synthesis of substituted diphenyl sulfides 1b–1q (for X and
salts 11a, 11f and 12a.
Y, see Scheme 1).
chloride, then disulfide 20 was reduced with zinc in AcOH to
give thiol 21. Compound 21 was coupled with the correspond-
ing aryl iodide (22b–22q) yielding sulfides 1b–1q (Scheme 6).
3-oxo-1,2-benzisothiazole 2n, yield: 51%; IR (KBr): 1715vs
᎐
(C᎐O) cmϪ1; 2,3-Dihydro-1-chloro-1-[4-(N-methylcarbamoyl)-
phenyl]-2-methyl-3-oxo-1,2-benzisothiazole 2o, yield: 57%; mp
Diaryl(acylamino)(chloro)-ꢀ4-sulfanes and (acylamino)sulfonium
chlorides
245–255 ЊC; IR (KBr): 3300m (N–H), 1700vs, 1630vs (C᎐O)
᎐
cmϪ1; 2,3-Dihydro-1-chloro-1-(4-acetylphenyl)-2-methyl-3-oxo-
1,2-benzisothiazole 2p, yield: 39%; mp 155–159 ЊC; IR (KBr):
General procedure. To the mixture of an N-methylcarbamoyl-
substituted diaryl sulfide (0.01 mol) in dry CH2Cl2 (30 cm3) was
added dropwise tert-butyl hypochlorite (0.12 cm3, 0.01 mol)
under stirring at 20 ЊC. After 1 h further stirring pentane
(5–20 cm3) was added and the mixture was allowed to stand
overnight at 20 ЊC. The crystals were filtered off, washed with
pentane and dried.
1695vs, 1660vs (C᎐O) cmϪ1; 2,3-Dihydro-1-chloro-1-(4-carboxy-
᎐
phenyl)-2-methyl-3-oxo-1,2-benzisothiazole 2q, yield: 57%; mp
238–243 ЊC; IR (KBr): 3600–2300br (O–H), 1720vs, 1710sh
(C᎐O) cmϪ1; 2,3-Dihydro-1-(2-nitrophenyl)-2-methyl-3-oxo-1,2-
᎐
benzisothiazol-1-ium chloride 3e, yield: 50%; mp 153–159 ЊC;
IR (KBr): 1710vs (C᎐O), 1528vs (NO ) cmϪ1; 2,3-Dihydro-1-(2-
᎐
2
methoxycarbonylphenyl)-2-methyl-3-oxo-1,2-benzisothiazol-1-
2,3-Dihydro-1-chloro-1-(2-methoxyphenyl)-2-methyl-3-oxo-
1,2-benzisothiazole 2b, yield: 40%; mp 177–188 ЊC; IR (KBr):
ium chloride 3f, yield: 57%, mp 317–320 ЊC; IR (KBr): 1713vs,
1688vs (C᎐O) cmϪ1; 2,3-Dihydro-1-(2-acetylphenyl)-2-methyl-
᎐
1703vs (C᎐O) cmϪ1; 2,3-Dihydro-1-chloro-1-(2-methylphenyl)-
᎐
3-oxo-1,2-benzisothiazol-1-ium chloride 3h, yield: 64%; mp 149–
2-methyl-3-oxo-1,2-benzisothiazole 2c, yield: 50%; mp 200–
152 ЊC; IR (KBr): 1728vs, 1650vs (C᎐O) cmϪ1
.
᎐
218 ЊC; IR (KBr): 1710vs (C᎐O) cmϪ1; 2,3-Dihydro-1-chloro-
᎐
1-(2-chlorophenyl)-2-methyl-3-oxo-1,2-benzisothiazole
2d,
;
Diaryl sulfides
yield: 42%; mp 178–189 ЊC; IR (KBr): 1702vs (C᎐O) cmϪ1
᎐
2,3-Dihydro-1-chloro-1-(4-methoxyphenyl)-2-methyl-3-oxo-1,2-
benzisothiazole 2j, yield: 60%; mp 124–128 ЊC; IR (KBr):
General procedure. The mixture of an aryl iodide (22b–22q,
0.1 mol), thiol 21 (16.7 g, 0.1 mol) and Cu2O (7.15 g, 0.05 mol)
in dry pyridine (150 cm3) was refluxed under nitrogen for 2 h
and after cooling to room temperature was poured into a mix-
ture of ice (600 g) and c. HCl (600 cm3). The precipitate was
filtered off, washed with water and dried then boiled with EtOH
(200 cm3) for 15 min. The insoluble part was filtered off then
the filtrate was saturated with H2S. The precipitate was removed
by filtration, then the filtrate was evaporated and the solid crude
product was recrystallized from a suitable solvent.
1705vs (C᎐O) cmϪ1; 2,3-Dihydro-1-chloro-1-(4-methylphenyl)-
᎐
2-methyl-3-oxo-1,2-benzisothiazole 2k, yield: 59%; mp 119–
124 ЊC; IR (KBr): 1710vs (C᎐O) cmϪ1; 2,3-Dihydro-1-chloro-1-
᎐
(4-chlorophenyl)-2-methyl-3-oxo-1,2-benzisothiazole 2l, yield:
40%; mp 145–149 ЊC; IR (KBr): 1700 cmϪ1; 2,3-Dihydro-1-
chloro-1-(4-nitrophenyl)-2-methyl-3-oxo-1,2-benzisothiazole
2m, yield: 66%; mp 182–183 ЊC; IR (KBr): 1700vs (C᎐O) cmϪ1
.
᎐
2,3-Dihydro-1-chloro-1-[4-(methoxycarbonyl)phenyl]-2-methyl-
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349
347

View Article Online
N-Methyl-2-sulfanylbenzamide (21). To boiling thionyl chlor-
ide (330 cm3, 4.55 mol) 2,2Ј-dithiobenzoic acid 19 (180 g, 0.587
mol) was added in small portions, then the reaction mixture was
refluxed for 1 h. After evaporation of the excess thionyl chloride
in vacuo, the residue was dissolved in dry benzene (800 cm3) and
added dropwise to the stirred solution of methylamine (109 g,
3.52 mol) in dry THF (900 cm3) at 0 ЊC. The reaction mixture
was allowed to stand overnight, then the solvent was removed in
vacuo, the residue triturated with water (2000 cm3), filtered and
dried. The crude product was washed thoroughly with acetone
(500 cm3) to give N,NЈ-dimethyl-2,2Ј-dithiobenzamide (20) (147
g, 75%); mp 208–216 ЊC; IR (KBr): 3285vs (N–H), 1632vs,
1550vs (amide I, II) cmϪ1. The mixture of disulfide 20 (100 g,
0.301 mol), zinc dust (79 g, 1.21 mol) and AcOH (900 cm3) was
refluxed for 6 h, then the solvent was removed in vacuo. After
adding water (180 cm3) and c. HCl (600 cm3) to the residue, the
mixture was filtered into water (1800 cm3), the crystals were
filtered off, washed with water, then dried to yield thiol 21 (40
g, 40%); mp 90–92 ЊC; IR (KBr) 3305s (N–H), 1618vs, 1540vs
1015vs (S᎐O) cmϪ1; 2-(N-methylcarbamoyl)phenyl 2-methyl-
᎐
phenyl sulfoxide 5c, yield: 80%; mp 159–161 ЊC; IR (KBr):
3290s (N–H), 1640vs, 1545s (amide I, II), 987vs (S᎐O) cmϪ1
;
᎐
2-(N-methylcarbamoyl)phenyl 2-chlorophenyl sulfoxide 5d,
yield: 78%; mp 175–179 ЊC; IR (KBr): 3320s (N–H), 1635vs,
1548s (amide I, II), 990vs (S᎐O) cmϪ1
; 2-(N-methyl-
᎐
carbamoyl)phenyl 2-nitrophenyl sulfoxide 5e, yield: 69%; mp
198–199 ЊC; IR (KBr): 3150–3400br (N–H), 1630vs, 1550vs
(amide I, II), 990vs (S᎐O) cmϪ1; 2-(N-methylcarbamoyl)phenyl
᎐
2-(methoxycarbonyl)phenyl sulfoxide 5f, yield: 87%; mp 190–
204 ЊC; IR (KBr): 3280w (N–H), 1712vs (C᎐O), 1665vs, 1550s
᎐
(amide I, II), 1014vs (S᎐O) cmϪ1; bis[2-(N-methylcarbamoyl)-
᎐
phenyl]sulfoxide 5g, yield: 78%; mp 212–214 ЊC (212–214 ЊC
in ref. 11); IR (KBr): 3290m (N–H), 1646vs, 1532 (amide I,
II), 1022vs, 1000s (S᎐O) cmϪ1; 2-(N-methylcarbamoyl)phenyl 2-
᎐
acetylphenyl sulfoxide 5h: yield: 76%; mp 192–194 ЊC; IR (KBr):
3220s (N–H), 1668vs (C᎐O), 1645vs, 1560s (amide I, II), 1000vs
᎐
(S᎐O) cmϪ1; 2-(N-methylcarbamoyl)phenyl 2-carboxyphenyl
᎐
sulfoxide 5i, yield: 77%; mp 186–193 ЊC (182–193 ЊC in ref. 9);
(amide I, II) cmϪ1
.
IR (KBr): 3480–2300br (O–H), 1682vs (C᎐O), 1648vs, 1560s
᎐
2-(N-Methylcarbamoyl)phenyl 2-methoxyphenyl sulfide 1b,
yield: 70%; mp 132–133 ЊC (109–110 ЊC in ref. 21); IR (KBr):
3260s (N–H), 1632vs, 1550s (amide I, II) cmϪ1; 2-(N-
methylcarbamoyl)phenyl 2-methylphenyl sulfide 1c, yield: 67%;
mp 95–96 ЊC; IR (KBr): 3280s (N–H), 1628vs, 1540vs (amide I,
II) cmϪ1; 2-(N-methylcarbamoyl)phenyl 2-chlorophenyl sulfide
1d, yield: 66%; mp 125–127 ЊC; IR (KBr): 3270s (N–H), 1642vs,
(amide I, II), 1028vs (S᎐O) cmϪ1; 2-(N-methylcarbamoyl)phenyl
᎐
4-methoxyphenyl sulfoxide 5j, yield: 86%; mp 147–149 ЊC; IR
(KBr): 3300s (N–H), 1640vs, 1555s (amide I, II), 1015vs (S᎐O)
᎐
cmϪ1; 2-(N-methylcarbamoyl)phenyl 4-methylphenyl sulfoxide
5k, yield: 70%; mp 137–139 ЊC; IR (KBr): 3320s (N–H), 1628vs,
1545vs (amide I, II), 1035vs (S᎐O) cmϪ1
; 2-(N-methyl-
᎐
carbamoyl)phenyl 4-chlorophenyl sulfoxide 5l, yield: 82%; mp
1550s (amide I, II) cmϪ1
; 2-(N-methylcarbamoyl)phenyl
164–167 ЊC; IR (KBr): 3380s (N–H), 1635vs, 1535vs (amide I,
2-nitrophenyl sulfide 1e, yield: 80%; mp 82–83 ЊC; IR (KBr):
3225s (N–H), 1632vs (amide I), 1510vs (NO2) cmϪ1; 2-(N-
methylcarbamoyl)phenyl 2-(methoxycarbonyl)phenyl sulfide 1f,
yield: 70%; mp 125–128 ЊC; IR (KBr): 3265m (N–H), 1715vs
II), 990vs (S᎐O) cmϪ1; 2-(N-methylcarbamoyl)phenyl 4-nitro-
᎐
phenyl sulfoxide 5m, yield: 77%; mp 210–216 ЊC; IR (KBr):
3290vs (N–H), 1650vs, 1562vs (amide I, II), 1017vs (S᎐O) cmϪ1
;
᎐
2-(N-methylcarbamoyl)phenyl 4-(methoxycarbonyl)phenyl sulf-
(C᎐O), 1640vs, 1565s (amide I, II) cmϪ1; 2-(N-methylcarb-
oxide 5n, yield: 73%; mp 160–168 ЊC; IR (KBr): 3250s (N–H),
᎐
amoyl)phenyl 2-acetylphenyl sulfide 1h, yield: 56%; mp 133–
1725vs (C᎐O), 1650vs, 1548s (amide I, II), 1010s (S᎐O) cmϪ1
;
᎐ ᎐
135 ЊC; IR (KBr): 3260s (N–H), 1668vs (C᎐O), 1630vs, 1552vs
2-(N-methylcarbamoyl)phenyl 4-(N-methylcarbamoyl)phenyl
᎐
(amide I, II) cmϪ1; 2-(N-methylcarbamoyl)phenyl 4-methoxy-
phenyl sulfide 1j, yield: 74%; mp 140–141ЊC (140–140.5 ЊC in
ref. 21); IR (KBr): 3300s (N–H), 1640vs, 1550vs (amide I, II)
cmϪ1; 2-(N-methylcarbamoyl)phenyl 4-methylphenyl sulfide 1k,
yield: 74%; mp 158–159 ЊC; IR (KBr): 3370s (N–H), 1645vs,
sulfoxide 5o, yield: 80%; mp 239–241 ЊC; IR (KBr): 3320vs
(N–H), 1665vs, 1635vs, 1545vs (amide I, II), 990vs (S᎐O) cmϪ1
;
᎐
2-(N-methylcarbamoyl)phenyl 4-acetylphenyl sulfoxide 5p,
yield: 77%; mp 122–130 ЊC; IR (KBr): 3290s (N–H), 1685vs
(C᎐O), 1638vs, 1550s (amide I, II), 988s (S᎐O) cmϪ1; 2-(N-
᎐
᎐
1534vs (amide I, II) cmϪ1
;
2-(N-methylcarbamoyl)phenyl
methylcarbamoyl)phenyl 4-carboxyphenyl sulfoxide 5q, yield:
66%; mp 244–247 ЊC; IR (KBr): 3350–2300br (O–H), 3250s
(N–H), 1715vs (C᎐O), 1620vs, 1580vs (amide I, II), 993vs (S᎐O)
4-chlorophenyl sulfide 1l, yield: 74%; mp 140–141ЊC; IR (KBr):
3300s (N–H), 1640vs, 1550vs (amide I, II) cmϪ1; 2-(N-
methylcarbamoyl)phenyl 4-nitrophenyl sulfide 1m, yield: 77%;
mp 189–192 ЊC; IR (KBr): 3260s (N–H), 1632vs, 1570vs (amide
I,II)cmϪ1;2-(N-methylcarbamoyl)phenyl4-(methoxycarbonyl)-
phenyl sulfide 1n, yield: 61%; mp 133–135 ЊC; IR (KBr): 3295s
᎐
᎐
cmϪ1; 8-methoxycarbonyl-1-naphthyl 2-(N-methylcarbamoyl)-
phenyl sulfoxide 8, yield: 66%; mp 211–214 ЊC; IR (KBr): 3372s
(N–H), 1710vs (C᎐O), 1650vs, 1538s (amide I, II), 1032s (S᎐O)
᎐
᎐
cmϪ1; 8-(N-methylcarbamoyl)-1-naphthyl phenyl sulfoxide 13a,
yield: 70%; mp 204–206 ЊC; IR (KBr): 3310s (N–H), 1650vs,
(N–H), 1722vs (C᎐O), 1640vs, 1550s (amide I, II) cmϪ1; 2-(N-
᎐
methylcarbamoyl)phenyl 4-(N-methylcarbamoyl)phenyl sulfide
1o, yield: 70%; mp 197–201 ЊC; IR (KBr): 3330s, 3265s (N–H),
1632 vs, 1550vs (amide I, II) cmϪ1; 2-(N-methylcarbamoyl)-
phenyl 4-acetylphenyl sulfide 1p, yield: 54%; mp 151–153 ЊC; IR
1555vs (amide I, II), 1021vs, 1015vs (S᎐O) cmϪ1; 8-(N-methyl-
᎐
carbamoyl)-1-naphthyl 2-(methoxycarbonyl)phenyl sulfoxide
13f, yield: 70%, mp 208–237 ЊC; IR (KBr) 3300s (N–H), 1713vs
(C᎐O), 1630s, 1570s (amide I, II), 1030vs (S᎐O) cmϪ1
.
᎐
᎐
(KBr): 3325vs (N–H), 1680vs (C᎐O), 1632vs, 1548s (amide I, II)
᎐
cmϪ1; 2-(N-methylcarbamoyl)phenyl 4-carboxyphenyl sulfide
1q, yield: 70%; mp 167–168 ЊC; IR (KBr): 3300s (N–H), 3450–
[2-(N-Methylcarbamoyl)phenyl](2-methoxyphenyl)(hydroxy)-
sulfonium chloride 5bؒHCl. The product obtained from 1b by
chlorination was crystallized from dichloromethane–ether and
the solid product separated was filtered off. After standing a
week at room temperature, 5bؒHCl could be isolated from the
mother liquor, owing to the hydrolysis of sulfonium salt 3b
brought about by a trace of water in the solvent. Mp 156–
161 ЊC, IR (KBr) 3240s (N–H), 1625vs, 1565vs (amide I, II),
2550–1730br (O–H).
2300br (O–H), 1695vs (C᎐O), 1640vs, 1548s (amide I, II) cmϪ1
.
᎐
Diaryl sulfoxides
General procedure. To the solution of a diaryl(acylamino)-
(chloro)-λ4-sulfane or a diaryl(acylamino)sulfonium chloride
(0.01 mol) in dichloromethane (10 cm3) was added water (10
cm3) and the mixture was stirred for 6 h at room temperature.
The organic phase was separated, dried with MgSO4, then the
solvent was removed in vacuo.
NMR Measurements
1
2-(N-Methylcarbamoyl)phenyl phenyl sulfoxide 5a, yield:
The H, 13C and 15N NMR spectra were recorded in CDCl3,
88%; mp 63–65 ЊC; IR (KBr): 3310vs, 3560s (N–H), 1632vs,
DMSO-d6 and CD3OD in 5 mm tubes at RT on a Bruker DRX
500 spectrometer at 500.13 (1H), 125.76 (13C) and 50.58 (15N)
MHz, with the deuterium signal of the solvent as the lock and
TMS as internal reference for 1H and 13C NMR. The standard
1552s (amide I, II), 1030s (S᎐O) cmϪ1; 2-(N-methylcarbamoyl)-
᎐
phenyl 2-methoxyphenyl sulfoxide 5b, yield: 80%; mp 165–
167 ЊC; IR (KBr): 3265s (N–H), 1650vs, 1560s (amide I, II),
348
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349

View Article Online
Bruker microprogram NOEMULT.AU to generate NOE was
used with a selective preirradiation time. The 2D COSY, HSQC
Acknowledgements
This work was supported by the Foundation of the Hungarian
Ministry of Education (FKFP No. 0.165/1997) and the Hun-
garian Scientific Research Foundation (OTKA Nos. T029799
and T033086).
1
and HMBC (1H–13C and H–15N) spectra were obtained by
using the standard Bruker gradient pulse programs COSYGS,
INV4GSTP and INV4GSLRNDSW, respectively. In the 2D
1H–15N HMBC spectra the resolution in the 15N dimension
was 4.9 Hz/point (SW = 50 ppm and TD = 512) to give 15N-
resonances (downfield from the signal of liquid ammonia as
external reference) with the accuracy of one decimal number.
References
1 I. Kapovits, J. Rábai, F. Ruff and Á. Kucsman, Tetrahedron, 1979,
35, 1869.
2 M. Balthazor and J. C. Martin, J. Am. Chem. Soc., 1975, 97, 5634;
J. C. Martin and M. Balthazor, J. Am. Chem. Soc., 1977, 99, 152.
3 P. D. Livant, J. Org. Chem., 1986, 51, 5384.
4 K. Onogi, M. Kido, M. Hori, T. Katakao and H. Shimizu,
Tetrahedron Lett., 1983, 4337; J. Zhang, S. Saito and T. Koizumi,
J. Org. Chem., 1998, 63, 5265.
5 D. Szabó, I. Kapovits, Á. Kucsman, P. Nagy, Gy. Argay and
A. Kálmán, J. Mol. Struct., 1999, 476, 157.
6 Á. Kucsman and I. Kapovits, in Organic Sulfur Chemistry:
Theoretical and Experimental Advances, ed. F. Bernardi, I. G.
Csizmadia and A. Mangini, Elsevier, Amsterdam, 1985, pp. 191–
245, and references therein.
7 R. A. Hayes and J. C. Martin, in Organic Sulfur Chemistry:
Theoretical and Experimental Advances, ed. F. Bernardi, I. G.
Csizmadia and A. Mangini, Elsevier, Amsterdam, 1985, pp. 408–
483, and references therein.
8 P. Livant and J. C. Martin, J. Am. Chem. Soc., 1977, 99, 5761.
9 D. Szabó, I. Kapovits, Á. Kucsman, P. Huszthy, Gy. Argay,
M. Czugler, V. Fülöp, A. Kálmán, T. Koritsánszky and L. Párkányi,
J. Mol. Struct., 1993, 300, 23.
10 I. Kapovits, J. Rábai, D. Szabó, K. Czakó, Á. Kucsman, Gy. Argay,
V. F ülöp, A. Kálmán, T. Koritsánszky and L. Párkányi, J. Chem.
Soc., Perkin Trans. 2, 1993, 847.
X-Ray diffraction study of 5bؒHCl
A single crystal of 5bؒHCl was mounted on a glass fiber and
transferred to the diffractometer. Data were collected on a
Rigaku RAXIS-II imaging plate detector using graphite-
monochromated Mo-Kα radiation at 293 K. Data processing
was carried out by using the software supplied with the
diffractometer. Structure solutions with direct methods were
carried out with the teXsan package.22 The refinements were
carried out by using the SHELXL-93 program23 with the full
matrix least squares method on F2. All non-hydrogen atoms
were refined anisotropically. Hydrogen atoms were generated
based upon geometric evidence and their positions were refined
by the riding model. The crystal data are as follows:
¯
C15H16ClNO3S, M 325.80, triclinic, space group P1, a =
10.162(4), b = 10.872(4), c = 8.234(2) Å, α = 104.94(3), β =
93.172(13), γ = 63.989(11)Њ, V = 788.2(4) Å3, Z = 2, Dc = 1.373
Mg mϪ3, 1688 independent reflections. At the end of the
refinement R = 0.0676, wR2 = 0.1931 for the reflections with
I > 2σI and R = 0.0948, wR2 = 0.2578 for all data. An extinc-
tion coefficient was refined to the value of 0.21(4). The maximal
residual peak and hole in the final difference electron density
11 D. Szabó, I. Kapovits, Á. Kucsman, V. Fülöp, M. Czugler and
A. Kálmán, Struct. Chem., 1990, 1, 305; D. Szabó, I. Kapovits,
Á. Kucsman, M. Czugler, V. Fülöp and A. Kálmán, Struct. Chem.,
1990, 2, 529; D. Szabó, I. Kapovits, Gy. Argay, M. Czugler,
A. Kálmán and T. Koritsánszky, J. Chem. Soc., Perkin Trans. 2,
1997, 1045.
map are 0.287 and Ϫ0.327 e ÅϪ3
.
Fractional atomic coordinates, bond lengths and angles
as well as thermal parameters have been deposited at the
Cambridge Crystallographic Data Centre.‡
12 D. Szabó, T. Ádám and I. Kapovits, Sulfur Lett., 1997, 21, 21 and
refs. therein.
Kinetics
13 L. J. Adzima, C. C. Chiang, I. C. Paul and J. C. Martin, J. Am.
Chem. Soc., 1978, 100, 953.
Kinetic measurements were performed with a Cary 3E (Varian)
UV–VIS spectrophotometer in tempered quartz cells of 1 cm
path length, in the wavelength range 286–340 nm, where the
absorption of the reactants and products differed significantly.
Dioxane–water mixtures [2–4% (v/v) of water] were used as
solvents for compounds 2a–2h, 2j–2q, 4a and 7, and water for
11a, 11f and 12a. The starting concentration of the reactants
was 1.5 × 10Ϫ4 mol dmϪ3 if not otherwise stated. The measuring
temperature and the concentrations of added acid, salt and
buffer are given in Tables 3–10. The reaction was started
by dissolving the reactants quickly in the solvent containing
H2O, D2O, acid, salt and/or buffer as required. The rate of
dissolution was fast; measurements were started immediately as
the solution became homogenous. First order rate constants
were calculated by using the kinetic computer programs of the
Cary 3E spectrophotometer. Standard deviations of the rate
constants are 5%.
14 G. J. Martin, J. P. Gouesnard, J. Dorie, C. Rabiller and M. L.
Martin, J. Am. Chem. Soc., 1977, 99, 1381; W. Schwotzer and
W. v. Philipsborn, Helv. Chim. Acta, 1977, 60, 1501; M. I. Burgar,
T. E. St. Amour and D. Fiat, J. Phys. Chem., 1981, 85, 502;
B. Thomas, W. Bieger and G. Grossmann, Z. Chem., 1981, 21, 292.
15 E. Vass, F. Ruff, I. Kapovits, J. Rábai and D. Szabó, J. Chem. Soc.,
Perkin Trans. 2, 1993, 855.
16 E. Vass, F. Ruff, I. Kapovits, D. Szabó and Á. Kucsman, J. Chem.
Soc., Perkin Trans. 2, 1997, 2061.
17 T. Ádám, F. Ruff, I. Kapovits, D. Szabó and Á. Kucsman, J. Chem.
Soc., Perkin Trans. 2, 1998, 1269.
18 F. Ruff and I. G. Csizmadia, Organic Reactions, Equilibria, Kinetics
and Mechanism, Elsevier, Amsterdam, 1994, pp. 283–289.
19 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91, 165.
20 Á. Kucsman, I. Kapovits, I. Kövesdi, A. Kálmán and L. Párkányi,
J. Mol. Struct., 1985, 127, 135; Á. Kucsman, I. Kapovits, L. Párkányi
and A. Kálmán, J. Mol. Struct., 1986, 140, 141.
21 J. Jilek, K. Sindelar, J. Pomykacek, V. Kmonicek and Z. Sedivy,
Collect. Czech. Chem. Commun., 1989, 12, 3294.
22 teXsan Crystal Structure Analysis Package, Molecular Structure
Co., Houston, TX, 1985, 1992.
‡ CCDC reference number 188/283. See http://www.rsc.org/suppdata/
p2/b0/b008156f/ for crystallographic files in .cif format.
23 G. M. Sheldrick, SHELXL-93 Program for the refinement of crystal
structures, University of Göttingen, 1993.
J. Chem. Soc., Perkin Trans. 2, 2001, 339–349
349