Ring-opening reactions of 5-(aryl)thianthrenium bromides with aryl thiolates

Article abstract of DOI:10.1002/poc.462

A series of 5-(aryl)thianthrenium bromides (6a-e) with aryl (Ar) groups phenyl (a), p-tolyl (b), p-anisyl (c), p-chlorophenyl (d) and p-bromophenyl (e) was prepared by reaction of thianthrene 5-oxide with the appropriate Grignard reagent ArMgBr. Reactions of 6a-e with aryl thiolates (Ar′SNa, Ar′ = phenyl, p-tolyl and p-chlorophenyl, 7a-c) were carried out in MeCN at room temperature. Products from 6a-c were a small amount of arene [benzene (8a) and toluene (8b)], and substantial amounts of ArSAr′ (9), thianthrene (Th) and a trisulfide, namely a 2-(ArS)-2′-(Ar′S)-diphenyl sulfide (10). Products from reactions of 6d,e were smaller amounts of 9 and 10 but substantial amounts of 1,4-di(Ar′S)benzene (11) and a tetrasulfide (12a-c). The reactions that lead to products 9-12 are attributed to ligand coupling (LC) pathways in sulfuranes formed by attack of Ar′S^- at the sulfonium S atom of 6. In the formation of 11 and 12 the halogen atom (CI, Br) is first displaced from 6d,e by Ar′S^-, giving a new thianthrenium ion (14) from which sulfurane formation (15) follows. Products 10 and 12 result from opening of the thianthrenium ring of sulfuranes 13 and 15 through LC. Products were assayed with a combination of GC and isolation with TLC, and were identified with a combination of GC (authentic compounds), x-ray crystallography (10a), elemental analyses and high-resolution mass spectrometry. The reactions of 6 are compared with earlier reactions of 5-(alkoxy)-and 5-(alkyl)thianthrenium salts (1 and 2). Copyright

Full text of DOI:10.1002/poc.462

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2002; 15: 139–147
DOI:10.1002/poc.462
Ring-opening reactions of 5-ꢀaryl)thianthrenium bromides
with aryl thiolates
Ding-Quan Qian,1 Bo Liu,1 Henry J. Shine,1* Ilse Y. Guzman-Jimenez2 and Kenton H. Whitmire2
1Department of Chemistry and Biochemistry, Texas Tech University, Lubbock, Texas 79409, USA
2Department of Chemistry, Rice University, Houston, Texas 77005, USA
Received 14 May 2001; revised 3 October 2001; accepted 5 October 2001
ABSTRACT: A series of 5-(aryl)thianthrenium bromides (6a–e) with aryl (Ar) groups phenyl (a), p-tolyl (b), p-anisyl
epoc
(c), p-chlorophenyl (d) and p-bromophenyl (e) was prepared by reaction of thianthrene 5-oxide with the appropriate
Grignard reagent ArMgBr. Reactions of 6a–e with aryl thiolates (Ar'SNa, Ar' = phenyl, p-tolyl and p-chlorophenyl,
7a–c) were carried out in MeCN at room temperature. Products from 6a–c were a small amount of arene [benzene
(8a) and toluene (8b)], and substantial amounts of ArSAr' (9), thianthrene (Th) and a trisulfide, namely a 2-(ArS)-2'-
(Ar'S)-diphenyl sulfide (10). Products from reactions of 6d,e were smaller amounts of 9 and 10 but substantial
amounts of 1,4-di(Ar'S)benzene (11) and a tetrasulfide (12a–c). The reactions that lead to products 9–12 are attributed
to ligand coupling (LC) pathways in sulfuranes formed by attack of Ar'SÀ at the sulfonium S atom of 6. In the
formation of 11 and 12 the halogen atom (Cl, Br) is first displaced from 6d,e by Ar'SÀ, giving a new thianthrenium ion
(14) from which sulfurane formation (15) follows. Products 10 and 12 result from opening of the thianthrenium ring
of sulfuranes 13 and 15 through LC. Products were assayed with a combination of GC and isolation with TLC, and
were identified with a combination of GC (authentic compounds), x-ray crystallography (10a), elemental analyses and
high-resolution mass spectrometry. The reactions of 6 are compared with earlier reactions of 5-(alkoxy)-and 5-
(alkyl)thianthrenium salts (1 and 2). Copyright  2002 John Wiley & Sons, Ltd.
Additional material for this paper is available from the epoc website at http://www.wiley.com/epoc
KEYWORDS: 5-(aryl)thianthrenium; nucleophilic reactions; aryl thiolates
INTRODUCTION
In earlier publications we reported on the reactions of
nucleophiles suchasiodide andthiophenoxide ionswith 5-
(alkoxy)- (1) and 5-(alkyl)thianthrenium ions (2). A clear
distinction was found, particularly in the reactions of
thiophenoxide ion. Reaction of this nucleophile with 1
occurred exclusively at the 5-thianthrenium sulfur atom.
The alkoxy group was displaced (to become an alkanol),
thianthrene (Th) was formed correspondingly, while
thiophenoxide was converted into diphenyl disulfide
(DPDS). A mechanism was proposed which invoked the
participationofasulfurane(3)anda5-(thiophenoxy)thian-
threnium ion (4) (Scheme 1).1 In contrast, reactions of
thiophenoxide with 2 were deduced to follow a mixture of
SN2 and elimination (E2C2) pathways. The latter pathway
was especially pronounced with 5-(cyclohexyl)thianthre-
nium ions.3 It was deduced, furthermore, that if, in analogy
*Correspondence to: H. J. Shine, Department of Chemistry and
Biochemistry, Texas Tech University, Lubbock, Texas 79409, USA.
Contract/grant sponsor: Robert A. Welch Foundation; Contract/grant
number: D-0028; Contract/grant number: C-0976.
Contract/grant sponsor: National Science Foundation; Contract/grant
number: CHE-9983352.
Scheme 1
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 139–147

140
D.-Q. QIAN ET AL.
5-(aryl) group is phenyl, p-tolyl and p-anisyl (6a–c) and
the second in which it is p-chlorophenyl and p-
bromophenyl (6d,e).
RESULTS
As far as we are aware, the 5-(aryl)thianthrenium
bromides 6 are not to be found in the literature. They
were prepared by reaction of the appropriate arylmagne-
sium bromide with thianthrene 5-oxide (ThO), according
to the method for preparing triarylsulfonium bromides
from diaryl sulfoxides.6 The identities of 6a–e, of
reactants 7a–c and of products 8a,b, 9a–1, 10a–1, 11a–
c and 12a–c are shown in Scheme 3. Products 8, 9 and 11
were assayed with gas chromatography (GC). Where
necessary, authentic products were prepared (9b–l, 11a);
others (11b,c) were identified by isolation and compari-
son with literature information. Products 10 and 12 were
isolated with column and preparative thin-layer chroma-
tography (TLC). They were identified with x-ray crystal-
lography (10a), and either elemental analysis (10f,h,i,l)
or high-resolution mass spectrometry (10b–e,g,j,k and
12a–c). The 1H NMR spectrum of each of the products 9–
12 was consistent with its structure.
Scheme 2
with reactions of 1, sulfurane (5) formation occurred in the
reactions of 2, it was reversible and was not responsible for
product formation via ligand coupling (LC4,5) (Scheme 2).
We have now studied the reactions of 5-(aryl)thian-
threnium bromides (6a–e) with sodium arylthiolates (7a–
c) in acetonitrile (MeCN) at room temperature. The
reactions are divided into two series, the first in which the
Reactions of 6a±c with 7a±c
The products of these reactions were Th, 8a,b, 9a–d,f,h,i
Table 1. Products of reactions of 6a±c with 7a±ca
Reactants
Products (%)
Sum (%) of
Run
6
7
8
9
Th
10
(Th ‡ 10)/6
1
2
3
4
5
6
7
8
9
a
a
a
b
b
b
c
a
b
c
a
b
c
a
b
c
a, 2.5
a, 1.2
a, 2.2
b, 4.2
a, 7.9
b, 5.7
d, 9.5
b, 7.6
17
6.5
21
14
a, 79
b, 91
d, 75
b, 82
l, 99
f, 79
c, 84
h, 94
i, 89
96
98
96
96
99
96
88
96
94
b, 4.2
f, 4.8
c, 2.3
h, 1.6
i, 1.6
17
3.6
1.6
5.4
c
c
a
In MeCN at room temperature.
Table 2. Products of reactions of 6d,e with 7a±ca
Reactants
Products (%)
Sum (%) of
(Th ‡ 10 ‡ 12)/6
Run
6
7
9
11
Th
10
12
10
11
12
13
14
15
d
d
d
e
a
b
c
a
b
c
d, 3.1
f, 3.4
a, 43
b, 33
c, 70
a, 42
b, 34
c, 60
47
37
65
48
33
60
d, 16
f, 17
k, 6.7
e, 17
g, 18
j, 11
a, 33
b, 41
c, 22
a, 32
b, 48
c, 25
96
98
96
96
99
96
e, 2.2
g, 2.6
j, 1.1
e
e
a
In MeCN at room temperature.
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 139–147

RING-OPENING REACTIONS OF THIANTHRENIUM BROMIDES
141
Scheme 3
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 139–147

142
D.-Q. QIAN ET AL.
runs. Also, the sum of the yields of Th, 10 and 12 should
equal the amount of 6 that was used. This is also borne
out and is expressed as a pecentage of 6 in Table 2.
DISCUSSION
Few reactions of thiolate ions with sulfonium salts are to
be found in the literature. Oae and Khim reported
reactions of aryl thiolates with triarylsulfonium ions in
boiling ethanol.7 Displacement of an aryl group with the
formation of diaryl sulfides occurred in high yield [Eqn.
(1)] and was attributed to ipso-attack on the displaced
aryl group. An alternative route to products is addition of
thiolate to sulfonium sulfur, forming a sulfurane, which
would collapse by LC4,5 to the observed products. This
route was ruled out because reaction at sulfonium sulfur
was expected also to cause ejection of an aryl group by
ligand exchange, with formation of an arene, either
benzene or toluene, and none of these arenes was found.
Scheme 4
More closely related to our work is that of Mori et al.
on reactions of thiophenoxide ion with 5-(aryl)diben-
zothiophenium ions in boiling methanol.8 In these
reactions a diaryl sulfide (ArSPh) was formed and ring
opening also occurred (Scheme 4). On the basis of the
effect of the methoxy group’s position on the relative
yields of diaryl sulfide and ring opening among the three
(anisyl)dibenzothiophenium ions, Mori et al. concluded
that the products were formed by ipso-attack reactions.
Oae, on the other hand, later quoted these reactions as
typical examples how products could be formed by LC
within a sulfurane.4,5
and 10a–d,f,h,i,l. Their yields are listed in Table 1. A
detailed example is given for reaction of 6a with 7a
(Experimental). The major product by far in all reactions
was a trisulfide (10), the certification of whose structure is
given with the Ortep diagram of 10a. Lesser amounts of a
diaryl sulfide (9) and of Th were obtained. In the reaction
of 6b with 7b the trisulfide (10l) was, in fact, the only
product found. The arenes 8a,b were obtained from 6a,b
in small amounts that varied, however, from one
experiment to another. Products 10 arise from the
opening of the thianthrenium ring by the thiolate
nucleophile. Therefore, the sum of the yields of 10 and
Th should equal the amount of reactant 6 that was used.
That this is the case can be seen in Table 1, where the sum
is expressed as a percentage of the amount of 6. In
principle, the sum of the yields of 8 and 9 should equal
the yield of Th, but this is not borne out well in Table 1.
These examples are illustrative of the dilemma
common to reactions involving sulfonium compounds
with nucleophiles, that is, in distinguishing between
direct reactions (such as SN2 and ipso-attack) and LC
Reactions of 6d,e with 7a±c
Yields of products of these reactions are listed in Table 2.
There are notable differences from the distribution of
products in Table 1. A diaryl sulfide (9d–g,j,k) and a
trisulfide (10d–g,j,k) were formed but not in as large
amounts as in reactions of 6a–c. Instead, the dominant
products were a 1,4-di(arylthio)benzene (11a–c), in
which the arylthio groups have been supplied by the
nucleophile, and a tetrasulfide (12a–c), which again
contains two arylthio groups from the nucleophile.
Products 8 (X = Cl, Br) were not found. In these
reactions, the sum of the yields of 9 and 11 should equal
the yield of Th, and this is borne out reasonably well in all
Scheme 5
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 139–147

RING-OPENING REACTIONS OF THIANTHRENIUM BROMIDES
143
Scheme 7
by ligand exchange. An alternative source of benzene and
toluene was thought to be possibly the decomposition of
unreacted 6a,b in the GC inlet. However, although 6a,b
did decompose when their solutions were injected into
the GC column, neither benzene nor toluene was found.
Consequently, exchange of ligands appears to have
occurred, albeit in small amounts. That exchange (path
e) would lead subsequently to the formation of a diaryl
disulfide, a product that was obtained and was assayed in
a few runs. However, the diaryl disulfide is also formed
by inadvertent oxidation of the thiolate nucleophile, and
the two routes cannot be distinguished. This possible
product of reaction, therefore, does not play a further role
in our discussion.
In the reactions of 6d,e, although 9 and 10 are formed,
the dominant products are 11 and 12. We thought,
initially, that these may have been formed by aromatic
nucleophilic substitution after 9 and 10 had been formed
[Eqns (2) and (3)]. We tested this possibility with
reaction between 9f and 7b [Eqn. (4)], under the
conditions typical of reactions of 6 (room temperature,
Scheme 6
within sulfurane intermediates.9–11 In our earlier work we
argued against LC and for SN2/E2C reactions with 2,3 but
for reaction at sulfonium sulfur with 1.1 In the reactions
of 6a–c with 7a–c, the products 9 and 10 could arise from
ipso-pathways as shown in Scheme 5 (paths a and b).
These products could also arise from thiophilic addition
to sulfonium sulfur and LC within the sulfurane (13),
Scheme 6 (paths c and d). Thiophilic addition would also
account for the formation of products 8a,b (path e);
products that, being diagnostic of sulfurane formation,
were sought but not found by Oae and Khim.7 At first
sight, it seems improbable that an arylthiolate (whose
conjugate acid has pKa in the range 6–8) could displace,
for example, phenyl anion (with pKa analogously 4312)
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 139–147

144
D.-Q. QIAN ET AL.
24 h, MeCN). Had reaction occurred, 11b would have
been formed, but none was detected with GC. In support
of this finding of unreactivity, we note that the
nucleophilic aromatic substitution reactions that we did
have to achieve in order to prepare authentic products
9b–l had to be carried out at high temperature in ethylene
glycol solution (Experimental). Our conclusion is that the
aromatic substitution needed in the formation of 11 and
12 must occur in 6d,e themselves. That reaction is
promoted by the sulfonium group, as shown in abbre-
viated form in Eqn. (5), much in the way that nitro groups
facilitate substitution in halobenzenes.13 We summarize
our interpretation of the reactions of 6d,e in Scheme 7,
wherein we propose that aromatic nucleophilic substitu-
tion (14) and LC reactions within sulfuranes (13 and 15)
account for all products.
out notable differences. 5-(Alkoxy) groups in 1 are
displaced entirely, 5-(alkyl) groups in 2 undergo SN2/
E2C reactions, while 5-(aryl) groups in 6 lead mainly to
diaryl sulfides and ring opening. Ring opening was not
observed at all in reactions of 1 and 2, and appears, in
reactions with thiolates, to be limited to 5-(aryl)thian-
threnium ions. These differences reflect the relative ease
of displacing an alkoxy group, as compared with alkyl
and aryl anions, through ligand exchange. Although
resistant to ligand exchange, the alkyl groups of 2 are
subject to easy SN2 and E2C reactions. In contrast, aryl
groups, unactivated toward substitution and ligand
exchange, appear to direct reactions into LC paths.
EXPERIMENTAL
The solvent MeCN was dried as described earlier.1 Aryl
thiols used in preparing 7a–c and aryl halides used in
preparing Grignard reagents for 6a–e and in preparing
diaryl sulfides 9b–l were from commercial sources; 9a is
commonly available. GC was carried out with columns A
and B described earlier.1 Silica gel of 60–200 mesh was
used for column chromatography. The plates used in
preparative TLC were commercial pre-coated silica gel
60 F254, 20 Â 20 cm. Mass spectrometric data were
provided by Dr Terry Marriott, Rice University. Ele-
mental analyses were performed by Desert Analytics
(Tucson, AZ, USA).
5-ꢀAryl)thianthrenium bromides ꢀ6a±e). These were
prepared by reaction of ThO with an arylmagnesium
bromide. An example is given with 6a. Phenylmagne-
sium bromide was prepared from 4.12 g (26.2 mmol) of
bromobenzene and 600 mg (24.7 mmol) of Mg in 30 ml
of diethyl ether. To the solution was added 30 ml of
benzene and the ether was distilled off. A solution of
1.15 g (4.96 mmol) of ThO in 15 ml of benzene was
added and the mixture was boiled under reflux in N2 for
18 h. To the cooled solution was added 2 ml of 48% HBr
in 2 ml of water. The benzene layer was washed with
2 Â 20 ml of 5% HBr solution. All aqueous portions
combined were extracted with 3 Â 60 ml of CHCl3. The
combined benzene and CHCl3 solution was evaporated to
give 1.38 g (3.70 mmol, 74.7% based on ThO) of 6a, m.p.
264–265°C. This was purified by precipitation from
solution in CHCl3–acetone (1:5) with diethyl ether, m.p.
269–270.5°C.
Anal. Calcd for C18H13S2Br (6a): C, 57.9; H, 3.51; S,
17.2; Br, 21.4. Found: C, 57.9; H, 3.41; S, 16.9; Br,
21.7%.
Preparation of 5-(p-tolyl)thianthrenium bromide (6b)
from p-tolylmagnesium bromide gave 49%, m.p. 248–
249°C.
Anal. Calcd for C19H15S2Br (6b): C, 58.9; H, 3.90; S,
16.6; Br, 20.6. Found: C, 58.7; H, 3.83; S, 16.4; Br,
20.6%.
Our present and recent work1,3 with reactions between
5-(substituted)thianthrenium salts and thiolates brings
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 139–147

RING-OPENING REACTIONS OF THIANTHRENIUM BROMIDES
145
5-(p-Anisyl)thianthrenium bromide (6c), 69%, had
m.p. 225–226.5°C. 5-(p-Chlorophenyl)thianthrenium
bromide (6d), 48%, had m.p. 258–259°C. 5-(p-Bromo-
phenyl)thianthrenium bromide (6e), 63%, had m.p. 268–
269.5°C. When first isolated, 6c–e were yellow solids.
Reprecipitation gave slightly yellow or white products.
Anal. Calcd for C18H12S2Br2 (6e): C, 47.8; H, 2.67; S,
14.2; Br, 35.3. Found: C, 47.9; H, 2.67; S, 13.8; Br,
35.1%.
9g (from p-dibromobenzene and 7b, 170°C, 16 h),
37%, m.p. 83–84°C. Lit.18 m.p. 82.5°C. 1H NMR: 7.365,
d, 2H, 8.41; 7.303, d, 2H, 8.08; 7.153, d, 2H, 8.29; 7.091,
d, 2H, 8.25; 2.354, s, 3H.
9h (from p-iodoanisole and 7b, 140°C, 20 h), 89%,
m.p. 45–46°C. Lit.19 m.p. 43–45°C. 1H NMR: 7.361, d,
2H, 8.84; 7.133, d, 2H, 8.48; 7.059, d, 2H, 8.32; 6.866, d,
2H, 8.84; 3.806, s, 3H; 2.299, s, 3H.
9i (from p-iodoanisole and 7c, 150°C, 2 d), 87%, m.p.
1
60.5–61.5°C. Lit.20 m.p. 60–61°C. H NMR: 7.405, d,
Sodium arylthiolates ꢀ7a±c). An example is given with
7b. Sodium (820 mg, 35.7 mmol) and 40 ml of dry
diethyl ether were placed in a three-necked flask
equipped with a dropping funnel, reflux condenser and
stirrer bar. A solution of an excess of p-thiocresol (6.5 g,
54 mmol) in 40 ml of diethyl ether was added with
stirring and heating to reflux until all of the sodium had
disappeared. The white solid was filtered, washed several
times with hexane and dried under vacuum to give 5.1 g
(34.9 mmol, 97%) of 7b.
2H, 8.95; 7.194, d, 2H, 8.52; 7.068, d, 2H, 8.87; 6.904, d,
2H, 8.87; 3.828, s, 3H.
9j (from p-bromoiodobenzene and 7c, 175°C, 22 h),
56%, m.p. 102–103°C. Lit.21 m.p. 106–107°C.
9k (from p-bromochlorobenzene and 7c, 170°C, 24 h),
61%, m.p. 94.5–95.5°C. Lit.22 m.p. 93–94°C.
9l was prepared and reported earlier, m.p. 58–
58.5°C.23
11a was isolated in the preparation of 9e, 43%, m.p.
81–82°C. Lit.14 m.p. 80°C.
11b was isolated in the preparation of 9g, 17%, m.p.
101–101.5°C. Lit.14 m.p. 95°C.
Diaryl sul®des ꢀ9a±l) and 1,4-bisꢀarylthio)benzenes
ꢀ11a±c). These compounds were prepared by heating
the appropriate halobenzene with an aryl thiolate at a
temperature in the range of 150–200°C in ethylene glycol
solution containing a small amount of the catalyst 1,2-
ethylbis(diphenylphosphino)nickel(II) bromide for 15–
24 h. 1H NMR spectra (ꢀ ppm, J in Hz) were obtained in
CDCl3 at 200 MHz. An example is given with 9f.
11c was not prepared directly, but was isolated from
the reaction of 6e with 7c and had m.p. 139.5–140.5°C.
Lit.24 m.p. 140–141°C.
Reactions of5-ꢀaryl)thianthrenium bromides ꢀ 6a±e)
with sodium thiolates ꢀ7a±c). Reactions in general were
carried out in MeCN containing a GC standard or
standards and at room temperature. That is, biphenyl was
used as a standard in all reactions, but was accompanied
by 2-butanone when product benzene or toluene was to
be assayed on column B. At the beginning of all reactions
the solution became orange or reddish–orange in color
and the color faded to pale yellow within about 2 h. An
attempt to characterize the orange color spectroscopically
was made with the reaction of 6a with 7b but was not
informative; a broad shoulder in the region of 380–
425 nm was obtained which disappeared within 60 min.
The assay of products depended on whether a trisulfide
(10) was formed (from 6a–c) or both 10 and a tetrasulfide
(12) were formed (from 6d,e). In all reactions, the
products 8, 9, 11 and Th were assayed with GC. Also
assayed with GC were the diaryl disulfides YC6H4SS-
C6H4Y, but those assays are not tabulated because we
could not distinguish between disulfide formed as a
reaction product and by inadvertent oxidation of 7.
Products 10 were assayed by weight after column
chromatography in reactions of 6a–c and by a combina-
tion of column chromatography and TLC in reactions of
6d,e. Products 12 were assayed after column chroma-
tography and TLC in these reactions, too. Examples of
each type procedure are given.
In a Pyrex pressure tube were placed 200 mg
(1.04 mmol)
of
p-bromochlorobenzene,
150 mg
(1.03 mmol) of 7b, 10 mg of catalyst and 3 ml of
ethylene glycol. The sealed tube was heated at 175°C
in a silicone oil bath for 15 h. To the cooled solution was
added 50 ml of water. The mixture was extracted with
3 Â 30 ml of CH2Cl2. The combined CH2Cl2 solution
was dried (MgSO4) and concentrated under vacuum. The
residue was placed on a 12 in column of silica gel and
eluted with light petroleum (b.p. 35–60°C) to give
140 mg (0.597 mmol, 58%) of 9f, m.p. 67–69°C. This
was chromatographed again twice, m.p. 71.5–73°C.
Lit.14 m.p. 73°C.
Analogous reactions gave 9b (from p-bromotoluene
and 7a, 200°C, 24 h), 51%, as an oil with a single GC
peak. 1H NMR: 7.322–7.114, m, 9H; 2.342, s, 3H. Lit.15
b.p. 145–147°C (2 mmHg).
9c (from p-iodoanisole and 7a, 160°C, 22 h), 66%, b.p.
222–225°C (75 mmHg), m.p. 30–30.5°C. Lit.16 b.p.
138–140°C (2 mmHg). 1H NMR: 7.417, d, 2H, 8.70; 7.3–
7.1, m, 5H; 6.897, d, 2H, 8.81; 3.820, s, 3H.
9d (from p-bromochlorobenzene and 7a, 200°C, 24 h),
47%, as an oil with single GC peak. Lit.17 b.p. 120°C
(2mmHg). 1H NMR: 7.361–7.229, m, 5H; 7.216, m, 4H.
9e (from p-dibromobenzene and 7a, 180°C, 17 h),
1
41%, as an oil. Lit.15 b.p. 153–155°C (3 mmHg). H
Reaction of5-ꢀphenyl)thianthrenium bromide ꢀ 6a)
with sodium thiophenoxide ꢀ7a). A solution of
52.6 mg (0.729 mmol) of 2-butanone and 82.9 mg
NMR: 7.395, d, 2H, 8.63; 7.361–7.253, m, 5H; 7.159, d,
2H, 8.54.
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 139–147

146
D.-Q. QIAN ET AL.
Figure 1. Ortep diagram of 2, 2'-bis-phenylthio)diphenyl sul®de -10a)
(0.537 mmol) of biphenyl in 20 ml of MeCN was placed
in a flask containing a stirrer bar. To this were added
294 mg (0.788 mmol) of 6a and 273 mg (2.07 mmol) of
7a. The flask was capped with a rubber septum. When the
reactants were mixed, the solution became orange in
color, which faded to yellow in 1.5 h. The mixture was
stirred for 2 d. Three GC assays were made on columns A
and B, to give 0.0195 Æ 0.0003 mmol (2.5%) of benzene
(8a), 0.0626 Æ 0.0008 mmol (7.9%) of diphenyl sulfide
(9a), 0.1306 Æ 0.004 mmol (16.6%) of Th and
0.127 Æ 0.001 mmol of diphenyl disulfide (DPDS).
Following the GC assays, the mixture was poured into
100 ml of water. The aqueous mixture was extracted with
3 Â 30 ml of diethyl ether and the dried (MgSO4) ether
solution was concentrated in a rotary evaporator. The
residue was transferred to a column of silica gel. Elutions
with light petroleum, which were monitored with TLC,
removed the earlier components of the mixture (9a, Th,
DPDS). Elution with petroleum ether/ether (40:1 by
volume) then removed 250 mg (0.621 mmol, 79%) of
10a, m.p. 90–91°C. A single crystal was grown
successfully for X-ray crystallography (Figure 1).
with petroleum ether/ether (40:1). Separated bands were
scraped from the plates and extracted with acetone in a
filter funnel. Work up gave 20.7 mg of 10d and 36.3 mg
of 12a. Thus, the total of 10d, obtained as an oil, was
52.1 mg (0.119 mmol, 15.9%) and of 12a, obtained as an
oil, was 127.5 mg (0.246 mmol, 32.9%). Crystallization
of 10d, obtained in reaction of 6a with 7c, from isopropyl
alcohol gave m.p. 69–70°C.
Identi®cation oftrisul®des ꢀ 10a±l) and tetrasul®des
ꢀ12a±c). Each of 10a–l had a 1H NMR spectrum
consistent with its structure. Trisulfides were crystalline
solids except 10e,g,j,k, which were oils. A single crystal
of 10a was grown and an Ortep diagram was obtained.
We were unsuccessful in attempts to do the same with
other solid trisulfides. Elemental analyses were obtained
of 10f,h,i,l. The remaining trisulfides were characterized
with high-resolution mass spectrometry masses. The
1
tetrasulfides had acceptable H NMR spectra. Each was
characterized with its high-resolution mass spectrum
mass.
Elemental analyses. 10f, calcd for C25H19S3Cl: C, 66.6;
H, 4.25: S, 21.3; Cl, 7.86. Found: C, 66.4; H, 4.11; S,
21.2; Cl, 7.60%.
10h, calcd for C26H22S3O: C, 69.9; H, 4.96; S, 21.5.
Found: C, 70.1; H, 5.02; S, 21.3%.
10i, calcd for C25H19S3ClO: C, 64.3; H, 4.10; S, 20.6;
Cl, 7.59. Found: C, 64.3, H, 4.06; S, 20.4; Cl, 7.42%.
10l, calcd for C26H22S3: C, 72.5; H, 5.15; S, 22.3.
Found: C, 72.3; H, 5.14; S, 22.4%.
Reaction of5-ꢀp-chlorophenyl)thianthrenium bromide
ꢀ6d) with sodium thiophenoxide ꢀ7a). The reactants
were 305 mg (0.748 mmol) of 6d and 396 mg
(3.00 mmol) of 7a. The GC standard was biphenyl
(95.8 mg, 0.622 mmol). After stirring overnight the
orange color had become yellow. GC assays gave
0.0233 Æ 0.0005 mmol (3.1%) of 9d, 0.350 mmol Æ
0.0034 mmol (47%) of Th and 0.0323 Æ 0.0038 mmol
(43%) of 11a. The residue from pouring into water and
extracting with ether was chromatographed with petro-
leum ether to remove biphenyl, Th, 9d and DPDS. Then,
elution with petroleum ether/ether (40:1) gave 31.4 mg of
10d, 91.2 mg of 12a and a mixture of 10d, 11a and 12a.
This mixture was streaked on TLC plates and developed
Mass spectrometry: molecular masses. The error in
ppm is given in parentheses.
10b, calcd for C25H20S3: 416.072719. Found: 416.
071504 (2.99).
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 139–147

RING-OPENING REACTIONS OF THIANTHRENIUM BROMIDES
147
10c, calcd for C26H20S3O: 432.067633. Found:
432.067237 (0.92).
10d, calcd for C24H17S3Cl: 436.018097. Found:
436.017406 (1.7).
www.wiley.com/epoc. The crystal structure has been
deposited at the Cambridge Crystallographic Data Centre
under the deposition number CCDC 172558.
10e, calcd for C24H17S3Br: 479.967592. Found:
479.967775 (0.38).
10g, calcd for C25H19S3Br: 495.981313. Found:
495.981357 (0.89).
Acknowledgements
10j, calcd for C24H16S3BrCl: 513.928620. Found:
513.928112 (0.99).
10k, calcd for C24H16S3Cl2: 469.979125. Found:
469.980243 (2.2).
12a, calcd for C30H22S4: 510.060442. Found:
510.060196 (0.48).
We thank the Robert A. Welch Foundation for support
under Grants D-0028 (H.J.S) and C-0976 (K.H.W) and
the National Science Foundation under Grant CHE-
9983352 (K.H.W).
12b, calcd for C32H26S4: 538.091742. Found:
538.091241 (0.93).
12c, calcd for C30H20S4Cl2: 577.982498. Found:
577.983218 (1.2).
REFERENCES
1. Liu B, Shine HJ, Zhao W. J. Phys. Org. Chem. 1999; 12: 827–836.
2. (a) McLennan DJ. Tetrahedron 1975; 31: 2999–3010; (b) Ford
WT. Acc. Chem. Res. 1973; 6: 410–415.
3. Liu B, Shine HJ. J. Phys. Org. Chem. 2001; 14: 81–89.
4. Oae S. Croat. Chem. Acta 1986; 59: 129–151.
5. Oae S. Phosphorus Sulfur 1986; 27: 13–29.
6. Wildi BS, Taylor SW, Potratz HA. J. Am. Chem. Soc. 1951; 73:
1965–1967.
7. Oae S, Khim YH. Bull. Chem. Soc. Jpn. 1969; 42: 3528–3535.
8. Hori M, Kataoka T, Shimizu H, Miyagaki M, Takagi T. Chem.
Pharm. Bull. 1974; 22: 2030–2041.
9. Trost BM, LaRochelle R, Atkins RC. J. Am. Chem. Soc. 1969; 91:
2175–2176.
10. LaRochelle RW, Trost BM. J. Am. Chem. Soc. 1971; 93: 6077–
6086.
11. Pryor WA, Smith K. J. Am. Chem. Soc. 1970; 92: 2731–2738.
12. March J. Advanced Organic Chemistry (3rd edn). John Wiley &
Sons: New York, 1985; 221–222.
13. Bruice PY. Organic Chemistry (3rd edn). Prentice Hall: Upper
Saddle River, NJ, 2001; 660–664.
14. Uhlenbroek JH. Recl. Trav. Chim. 1961; 80: 1057–1065.
15. Voronkov MG, Deryagina EN, Klochkova LG, Ivanova GM. Russ.
J. Org. Chem. 1977; 13: 2391–2399.
X-ray crystallographic data. A suitable crystal was
encapsulated in a capillary tube and mounted on the tip
of a glass-fibre with epoxy cement. Data were collected
with a Bruker AXS automated CCD diffractometer using
the Bruker AXS package and using Mo Ka radiation
˚
(l = 0.71069 A). The structure was solved by direct
methods. All non-hydrogen atoms were refined with
anisotropic displacement parameters, and hydrogen
atoms were calculated in ideal positions (riding model).
Refinement of F2 was performed using all reflections.
The weighted R-factor wR and goodness of fit S are based
on F2, conventional R-factors R are based on F, with F
set to zero for negative F2. R-factors based on F2 are
statistically about twice as large as those based on F, and
R-factors based on all data will be even larger. All
software used is contained in the SHELXTL 5.10
program library.25
16. Adams R, Reifschneider W, Nair MD. Croat. Chem. Acta 1957;
29: 277–285.
17. Cristau HJ, Chabaud B, Chene A, Christol H. Synthesis 1981; 11:
892–894.
18. Bourgeois E, Fouassin A. Bull. Soc. Chim. Fr. 1911; 9: 938–944.
19. Still IWJ, Toste FD. J. Org. Chem. 1996; 61: 7677–7680.
20. Fujisawa T, Kobori T, Ohtsuka N, Tsuchihashi G. Tetrahedron
Lett. 1968; 49: 5071–5074.
21. Wilson HF, Tarbell DS. J. Am. Chem. Soc. 1950; 72: 5200–5205.
22. Ogata Y, Kamei T. Tetrahedron 1970; 26: 5667–5674.
23. Shine HJ, Rahman M, Seeger H, Wu GS. J. Org. Chem. 1967; 32:
1901–1908.
Supplementary material
Five tables giving crystal data and structural refinements,
atomic coordinates and equivalent isotropic displacement
parameters, bond lengths and angles, anisotropic dis-
placement parameters, and hydrogen coordinates and
equivalent isotropic displacement parameters for
C24H18S3 are available at the epoc website at http://
24. Schmidt U, Kabitzke K, Markau K, Mueller A. Justus Liebigs Ann.
Chem. 1964; 672: 78–90.
25. Sheldrick G. SHELXTL 5.10. Bruker: Madison, WI, 1997.
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 139–147