Synthesis and decomposition of some dialkyl oxide derivatives of organotrisulfides

Article abstract of DOI:10.1021/jo950826x

The isolation or detection of sulfenic sulfonic thioanhydrides 1 (e.g., 6), sulfenyl vic-disulfoxides 4 (e.g., 9, 14), and sulfinic thioanhydrides 5 (e.g., 20) has been carried out by oxidative procedures at various temperatures. The decomposition of these compounds has been investigated and is shown to be consistent with the mechanism proposed for the decomposition of trisulfide monoxides.

Full text of DOI:10.1021/jo950826x

J . Org. Chem. 1996, 61, 991-997
991
Syn th esis a n d Decom p osition of Som e Dia lk yl Oxid e Der iva tives
of Or ga n otr isu lfid es
G e´ rard Derbesy and David N. Harpp*
Department of Chemistry, McGill University, Montreal, Quebec H3A 2K6, Canada
Received May 2, 1995X
The isolation or detection of sulfenic sulfonic thioanhydrides 1 (e.g., 6), sulfenyl vic-disulfoxides 4
(e.g., 9, 14), and sulfinic thioanhydrides 5 (e.g., 20) has been carried out by oxidative procedures at
various temperatures. The decomposition of these compounds has been investigated and is shown
8
to be consistent with the mechanism proposed for the decomposition of trisulfide monoxides.
The preparation, characterization, and applications of
preparation by direct oxidation. In addition, little has
1
2
thiosulfonates and vic-disulfoxides (R-disulfoxides) has
been reported on the trisulfide dioxide derivatives 4 and
been extensively reviewed.3 In contrast, only a few
5.
4a,c,9
Therefore, it was decided to further investigate
dioxide derivatives of alkyl trisulfides (RS
3
O
2
R′) have
this class of compounds, focusing on the di-tert-butyl
derivatives.
been isolated and characterized.4 A variety of sulfenic
-6
sulfonic thioanhydrides 1 have been reported and were
7
prepared by nonoxidative, synthetic procedures. Sulfen-
ter t-Bu tylsu lfen ic ter t-Bu tylsu lfon ic
ic sulfonic thioanhydrides 1 are relatively stable and are
isolated as decomposition products of various polysul-
fide-polyoxides.3
Th ioa n h yd r id e (6)
,8
tert-Butylsulfenic tert-butylsulfonic thioanhydride (6)
The electrophilic oxidation of trisulfide 2 or sulfenic
sulfinic thioanhydride 3 has been postulated to take place
at the central sulfur giving the corresponding sulfenyl
vic-disulfoxide 4 that would rapidly rearrange to the
sulfenic sulfonic thioanhydride 1 derivatives.4 However,
no clear evidence of this mechanism has been reported.
To our knowledge, dioxides 4 have never been detected.
The more electron-rich sulfur atom of 3 is believed to
be the external sulfenyl sulfur which should preferen-
tially react with the oxidizing agent to give the disulfinic
has been isolated from the decomposition of various
oxidized derivatives of di-tert-butyl trisulfide probably
8
because 6 is one of the most stable of this class of
10
compounds. Bass and Evans reported that the peracid
c
oxidation of di-tert-butyl thiosulfinate 7a in the presence
of anhydrous tungsten(VI) oxide affords 6 in high yield
(eq 1).
9
thioanhydride 5 analog. Such compounds were first
postulated by Steudel,4 and clear evidence of diaster-
a,b
eomeric 5 has been found.3
,9
By analogy with the oxidation of disulfides,1a,11 the
nucleophilic oxidation of tert-butylsulfenic tert-butyl-
sulfinic thioanhydride (7b) (t-BuS(O)SSBu-t) should have
given 6. In our hands, these nucleophilic oxidation
experiments were unsuccessful. However, the oxidation
of di-tert-butyl dithiosulfite (8) gives a high yield of
dioxide 6 using dimethyldioxirane (DMD), and moderate
yields were observed using m-CPBA (eq 2).
Although various dioxides 1 have been reported,3-5,7,8
little is known about their chemical properties or their
X
Abstract published in Advance ACS Abstracts, J anuary 1, 1996.
(1) (a) Zefirov, N. F.; Zyk, N. V.; Beloglazkina, E. K.; Kutateladze,
A. G. Sulfur Rep. 1993, 14, 223. (b) Farng, L.-P. O.; Kice, J . L. J . Am.
Chem. Soc. 1981, 103, 1137. (c) Freeman, F.; Angeletakis, C. N. J . Org.
Chem. 1982, 47, 4194.
(2) Highlights of the most interesting results can be found in: (a)
Freeman, F. Chem. Rev. 1984, 84, 117. (b) Folkins, P. L.; Harpp, D.
N. J . Am. Chem. Soc. 1991, 113, 8998. (c) Folkins, P. L.; Harpp, D. N.
J . Am. Chem. Soc. 1993, 115, 3066 and references cited therein.
(3) (a) Derbesy, G.; Harpp, D. N. J . Org. Chem. 1995, 60, 1044. (b)
For NMR and structural information on these and related compounds
see: Derbesy, G.; Harpp, D. N. Sulfur Rep. 1995, 16, 363. (c) Derbesy,
G. Ph.D. Thesis, McGill University, 1994.
(4) (a) Steudel, R.; Latte, J . Chem. Ber. 1977, 110, 423. (b) Steudel,
R. Phosphorus Sulfur 1985, 23, 33. (c) Freeman, F.; Ma, X.-B.; Lin, R.
I.-S. Sulfur Lett. 1993, 15, 253.
Such a conversion provides further demonstration of
the formation of the vic-disulfoxide 9 which rapidly
rearranges to the tert-butylsulfenic tert-butylsulfonic
thioanhydride (6). Compound 6 was fully characterized
by NMR and mass spectrometry. Recrystallization from
(
5) Freeman, F.; Lee, C. Magn. Reson. Chem. 1988, 26, 813.
(6) The naming of these polysulfide-polyoxide derivatives has
caused concern in that the IUPAC names do not permit the reader to
visualize the molecule. As a consequence, a variety of simpler names
have survived. For example, one literature reference for compound
3
2
,4,5-trithia-4-oxotricyclo[5.2.1.0]decane names it as
-oxide derivative (ref 20a). We have used the thioanhydride approach,
a trithiolane
(8) Derbesy, G.; Harpp, D. N. J . Org. Chem. 1995, 60, 4468.
(9) Derbesy, G.; Harpp, D. N. Sulfur Lett. 1995, 18, 167.
(10) Bass, F. W.; Evans, S. A., J r. J . Org. Chem. 1980, 45, 710.
(11) Kim, Y. H.; Takata, T.; Oae, S. Tetrahedron Lett. 1978, 2305.
0
022-3263/96/1961-0991$12.00/0 © 1996 American Chemical Society

9
92 J . Org. Chem., Vol. 61, No. 3, 1996
Derbesy and Harpp
decomposes to di-tert-butyl disulfide (16a ) after 12 h at
9
0 °C (eq 4). Strong evidence of the evolution of sulfur
dioxide was given by the change of color (yellow to red)
of the pH paper that was placed over the reaction
mixture.
ter t-Bu tylsu lfen yl ter t-Bu tyl vic-Disu lfoxid e (9)
The formation of tert-butylsulfenic tert-butylsulfonic
thioanhydride (6) by oxidation of the corresponding
dithiosulfite 8 implies the formation of tert-butylsulfenyl
tert-butyl vic-disulfoxide (9) as an intermediate. By
analogy with the low temperature detection of vic-
disulfoxides,2 we felt it might be possible to detect this
trisulfide analog 9.
F igu r e 1. ORTEP drawing of 6.
Sch em e 1
,3,8
The low temperature oxidation of di-tert-butyl dithio-
sulfite (8) was carried out according to our established
procedure3 (-60 °C for 12 h). The reaction mixture was
filtered and transferred to the NMR spectrometer at the
same temperature. The 13C and H NMR were recorded
at various temperatures from -60 °C to room tempera-
ture (Figure 2). Although the spectra obtained are not
completely clean, significant amounts of 9 can be detected
at -60 °C. Compound 9 is barely stable at this temper-
ature and rapidly decomposes at -50 °C; only traces can
be observed at -40°C. As expected, compound 9 is far
,8
1
hexane gave colorless crystals that were suitable for
X-ray analysis. The ORTEP drawing is represented by
Figure 1. It is interesting to note that the presence of
the two oxygen atoms has a great effect on the S-S bonds
and there is a difference of 0.11 Å between the length of
less stable than its disulfide analog 17, which was very
3,8
stable at -40 °C.
However, clear evidence for such an
intermediate could be obtained by comparing its NMR
signals and its behavior with the ones of 17. The
13
C
the S(O) -S bond and the S-S bond. Full details on this
2
NMR signals of 9 could be clearly identified as they are
and related molecules are available in ref 3b.
consistent with those of the various di-tert-butyl trisul-
An unsymmetrical sulfenic sulfonic thioanhydride 10
has also been prepared by the reaction of tert-butyl
hydrodisulfide12 (11) with 2-propanesulfonyl chloride
3,8,9
fide-polyoxides
as well as with the ones of 17.
The presence of the two diastereoisomers of tert-
butylsulfenyl tert-butyl vic-disulfoxides 9a ,b could not be
defined with certainty due to the small quantities of 9
formed and the complexity of the reaction mixture. The
analysis of the spectrum obtained at -60 °C revealed that
some starting material 8 (∼30%) was not oxidized under
these conditions. Large quantities (∼45%) of unknown
compounds could also be detected. The study of the
decomposition of 9 from -60 to -40 °C clearly shows that
these unknown compounds correspond to the decomposi-
tion products of 9.
13
(
12) in the presence of pyridine. Pure tert-butylsulfenic
isopropylsulfonic thioanhydride (10) was isolated in low
yield by chromatography; it was found to be relatively
stable at room temperature (>2 months) (eq 3).
Table 1 clearly shows that the signals of the di-tert-
butyl vic-disulfoxide 17 and the trisulfide analog 9 are
very similar. In both cases, the tert-butyl group attached
to the sulfinyl sulfur has its chemical shifts at slightly
higher field than the ones in the corresponding oxides
The direct oxidation of tert-butyl isopropyl dithiosulfite
13) using 1 equiv of DMD was not regioselective. A
careful analysis of the reaction mixture (vide infra)
suggests the formation of the two possible vic-disulfoxides
14a ,b) which rearrange to give a mixture of the two
(
(
7
a and 7b. Finally, the chemical shifts observed for the
symmetric compounds 6 and 15b and the two unsym-
metric sulfenic sulfonic thioanhydrides 10 and 15a
tert-butyl group bonded to the sulfenyl sulfur of 9 are
slightly lower field than the ones of the corresponding
(
Scheme 1).
monoxide 8. These results are consistent with other
As we found,3,8 the rearrangement of vic-disulfoxide
NMR studies.3
,14
The 1H NMR were very difficult to
derivatives is solvent and concentration dependent. The
cleanest results were observed using DMD as oxidizing
agent.
A decomposition study of 6 at varying temperatures
reveals that it is very stable at 25 °C. However, it readily
interpret owing to problems with overlapping signals.
In order to further understand these spectra and to
optimize the reaction conditions, this low temperature
(14) (a) Freeman, F.; Angeletakis, C. N. J . Org. Chem. 1982, 47,
4
194. (b) Mieloszynski, J . L.; Weber, J . V.; Schneider, M.; Paquer, D.;
(
12) Derbesy, G.; Harpp, D. N. Sulfur Lett. 1992, 14, 199.
Boen, M.; Pare, G. Sulfur Lett. 1988, 8, 27. (c) Freeman, F.; Lee, C.
Magn. Reson. Chem. 1988, 26, 813.
(13) Douglass, I. B.; Norton, R. V. J . Org. Chem. 1968, 33, 2104.

Dialkyl Oxide Derivatives of Organotrisulfides
J . Org. Chem., Vol. 61, No. 3, 1996 993
F igu r e 3. 1H and 13C NMR spectra of intermediates 18 and
1
9.
the lack of reactivity of the substrate at this temperature.
At higher temperatures (>-60 °C), no vic-disulfoxide 9
can be observed, which emphasizes the low stability of
9
. As mentioned previously, 9 is barely stable at -60
°C, and several decomposition products could be observed.
To avoid this stability problem and increase the reactivity
of the substrate, the low temperature experiment was
attempted using methylene chloride-d
2
which has a lower
freezing point (-93 °C) and is more polar than chloroform-
d. The reaction was carried out at -80 °C but did not
afford good results because of low conversion (<30%),
variation of the chemical shifts, and overlapping of the
solvent and the reaction product peaks. From these
results, it was concluded that the temperatures either
were too low to allow a suitable conversion of the starting
material 8 or too high for 9 to be stable enough to detect.
The study of the decomposition of tert-butylsulfenyl
tert-butyl vic-disulfoxide 9 is complicated by the remain-
ing starting material and the impossibility of detecting
undecomposed 9. However, the main, final decomposi-
tion product (spectrum at room temperature) is 6, which
is consistent with the preparation of 6 (vide supra).
Analysis of the spectra from -60 °C to room temperature
revealed the presence of eight carbon and four proton
signals corresponding to four different tert-butyl groups.
The chemical shifts observed are consistent with an
unsymmetrical di-tert-butyl diastereomeric derivative or
two very closely related unsymmetrical di-tert-butyl
analogs 18 and 19. These signals were observable in
various ratios up to room temperature (Figure 3).
F igu r e 2. 13C NMR spectra at several temperatures of the
product mixture of the m-CPBA oxidation of 8.
Ta ble 1. 13C a n d 1H NMR Ch em ica l Sh ifts of 9 a n d Its
a ,b
Rela ted Der iva tives
An NMR comparison of the different classes of di-tert-
butyl polysulfide-polyoxides3 clearly reveals that the
chemical shifts of a given tert-butyl group can be unam-
biguously interpreted depending on which sulfur it is
bonded to (sulfenyl, sulfinyl, sulfonyl, or sulfinate). This
carbon NMR study was used to assign the unknown
signals observed in the decomposition of 9. These
unknown chemical shifts and the ranges of the different
types of di-tert-butyl derivatives are presented in Table
2.
,8,9
a
Recorded using deuterated chloroform (CDCl3) as NMR sol-
vent. b Relaxation time (t1) used: t1 ) 2s. The spectra were
obtained at low temperature and do not represent a specific
diastereoisomer.
c
This comparison clearly shows that intermediates 18
and 19 present two different tert-butyl groups bonded to
a sulfenyl sulfur (∼49 and ∼29.5 ppm), another one
bonded to a sulfinate sulfur (∼60 and 21.5 ppm), and
finally corresponding to a moiety in between a tert-
butylsulfinyl and a tert-butyl sulfinate (∼60 and ∼23.5
ppm). Considering that 6 is the main decomposition
product and that no diastereoisomer should be observed,
the unknown intermediates are believed to be two closely
related unsymmetrical di-tert-butyl analogs 18 and 19.
experiment was repeated several times using excess
m-CPBA (2 equiv), at different temperatures, in a dif-
ferent solvent (CD
4 h to 4 days).
The analysis of these results reveals that the oxidation
of monoxide 8 was not even complete using 2 equiv of
m-CPBA at -60 °C for 4 days. This result points out
2 2
Cl ), and for various reaction times
(

9
94 J . Org. Chem., Vol. 61, No. 3, 1996
Derbesy and Harpp
Ta ble 2. 1H a n d 13C NMR of 18 a n d 19 a n d Ch em ica l
Sch em e 4
Sh ifts Ra n ges (δ p p m ) of th e ter t-Bu tyl Gr ou p s Bon d ed
a
to Differ en tly Oxid ized Su lfu r
a
All the spectra were recorded in CDCl3. b The spectra were
obtained at -60 °C.
Sch em e 2
These results can be rationalized by proposing that,
under these reaction conditions, both sulfenyl vic-disul-
foxides 14a and 14b are formed. The formation of large
quantities of both symmetrical dioxides 6 and 15b is
consistent with the previously proposed cleavage of the
S(O)-S(O) bonds,3 and random recombination of ions
,8
(
less steric hindrance here because of the extra sulfur)
should eventually give the four possible sulfinic thio-
sulfenic anhydrides 18a -d which should rearrange to the
four possible 6, 15b, 10, and 15a (Scheme 4).
Sch em e 3
It is surprising that small quantities of di-tert-butyl
vic-disulfoxide (17) and tert-butylsulfinic thioanhydride
(
(
20) were detected because the di-tert-butyl dithiosulfite
8) used did not contain any traces of the corresponding
polysulfides or polysulfide-monoxides 7a or b. However,
the remaining 8 was not stable under these conditions
and probably participates in the decomposition of 9.
Considering the complexity of the mechanism reported,
it is not unusual that many different di-tert-butyl deriva-
tives are observed.
Although the structure of these intermediates cannot
be determined with certainty, it is clear that 9 is formed
under these low temperature conditions and that the
final decomposition product is mainly 6. The structure
1
3
of 9 is completely consistent with related C NMR
signals3b as well as a logical series of likely intermediates
To our knowledge, this is the first detection of a
sulfenyl vic-disulfoxide. However, the rearrangement of
as shown in Scheme 2 and displayed in Figure 2.
Therefore, considering the mechanism proposed3,8 for the
9
to 6 could not be fully understood because under these
decomposition of the di-tert-butyl vic-disulfoxide (17) and
according to the previous NMR study, a consistent
intermediate would be tert-butylsulfinic tert-butylthio-
sulfenic anhydride (18a ), which is the expected inter-
mediate of the rearrangement of 9 to 6 (Scheme 2).
The only unsymmetrical tert-butyl derivative closely
related to intermediate 18a that would match the re-
quired chemical shifts would be the tert-butylsulfenic tert-
butyl thiosulfinic anhydride (19).
conditions the detection of 9 is temperature controlled.
Therefore, it considerably limits the decomposition study
as other derivatives are always present in the reaction
mixture. Low temperature experiments using stronger
oxidizing agents such as DMD should afford better
results. Unfortunately, we were unable to develop a
suitable technique for the low temperature DMD oxida-
tion.
ter t-Bu tylsu lfin ic Th ioa n h yd r id e (20)
9
As we have found, the 1-equiv electrophilic oxidation
of tert-butylsulfenic tert-butylsulfinic thioanhydride (7b)
and the 2-equiv electrophilic oxidation of di-tert-butyl
trisulfide (16b) afforded a high yield of the diastereo-
meric di-tert-butylsulfinic thioanhydrides (20) (eq 5).
It is conceivable that intermediate 19 rearranges to
give back the vic-disulfoxide 9. At this temperature, 9
would immediately redecompose to a mixture of 18a and
1
6
9. Considering that intermediate 18a gives the dioxide
, the final decomposition mixture should be mainly
composed of 6 (Scheme 3).
The 1-equiv DMD oxidation of tert-butyl isopropyl
dithiosulfite (13) reported above (Scheme 1), afforded an
almost equimolar mixture of the two symmetrical and
two unsymmetrical possible dioxides (6, 15b, 10, 15a ).
tert-Butylsulfinic thioanhydride (20) is barely stable
at room temperature and cannot be kept in the freezer
for more than 1 week. The decomposition of 20 was also
investigated and afforded a complex mixture. Various

Dialkyl Oxide Derivatives of Organotrisulfides
J . Org. Chem., Vol. 61, No. 3, 1996 995
Sch em e 5
Sch em e 7
Sch em e 6
Sch em e 8
attempts under slightly different conditions have shown
that the two main decomposition products were the
corresponding thiosulfinate 7a and sulfenic sulfonic
thioanhydride 6. Significant quantities of tert-butyl-
sulfinic tert-butylsulfonic thioanhydride (21) were de-
tected in most cases as well as other di-tert-butyl polysul-
fide-polyoxides (Scheme 5).
In a separate experiment, the decomposition of 20 was
carried out in the presence of 1 equiv of isopropylsulfinyl
chloride (22) as trapping agent. Although the final
decomposition mixture was even more complex, ca. 40%
of tert-butylsulfinyl chloride (23) could be clearly identi-
fied by 13C and H NMR. The detection of tert-butylsulfi-
Sch em e 9
1
3
,8
nyl chloride (23) confirms our current hypothesis of
ionic cleavage of 20 at the S-S(O) bond. Under these
conditions, the tert-butyl thiosulfinate anion (24) formed
can react with isopropylsulfinyl chloride (22) to give the
unsymmetrical tert-butylsulfinic isopropylsulfinic thio-
anhydride (25) and the tert-butyl sulfinyl cation (26a )
can then react with the remaining chloride ion to afford
the observed tert-butylsulfinyl chloride (23) (Scheme 6).
Although compound 25 could not be detected, signifi-
cant amounts of its decomposition products, isopropyl-
sulfenic tert-butylsulfonic thioanhydride (15b) and iso-
propyl tert-butylthiosulfonate (27a ),3 were identified as
well as other derivatives issued from the normal and
mixed decompositions.
would also account for the formation of thiosulfinate 7a .
The formation of tert-butylsulfinic dithioanhydride (28)
is also possible (low intensity NMR signals in the region
of 28 could be detected) (Scheme 8).
,13
As mentioned earlier, tert-butylsulfinic thioanhydride
(
20) is not stable at -30 °C when stored in the freezer.
The results observed in this decomposition study of 20
are even more complex than other systems we have
studied.3 Therefore, it is very difficult to give a detailed
decomposition mechanism. However, using the mecha-
The determination of the decomposition products of 20
at this temperature confirmed the previous results. After
,8
3
weeks in the freezer, the decomposition mixture
presented 30% of the thermodynamic diastereoisomer
0b, 35% of 7a , 22% of tert-butylsulfinic tert-butylsulfonic
3
,8
nistic principles developed in the previous cases, these
results can be reasonably rationalized.
2
3
b
dithioanhydride (29), and 13% of the trisulfide analog
First, good evidence of the cleavage of the S-S(O) bond
is given by the formation of tert-butylsulfinyl chloride
2
1 (Scheme 9).
This low temperature study allowed the detection of
(
23) (Scheme 6). The presence of 6 (Scheme 7) can be
the decomposition intermediates; the ratios and products
observed are completely consistent with the mechanism
previously proposed. It is clear that the decomposition
of 20a ,b proceeds through the ionic cleavage of the
S(O)-S bond. The reaction of the anion 24 formed with
another molecule of 20 affords 28 and the tert-butylsulfi-
nyl anion (26b). The ambident sulfinyl ions 26a ,b
formed in the two previous steps react with 28 or 20 by
explained by the recombination of these ions to form
intermediate 19. Rearrangement of such an intermedi-
ate could give back the starting material 20 or eventually
afford the corresponding sulfenyl vic-disulfoxide 9 that
would immediately be converted to 6. This theory is
suggested because chemical shifts in the region of this
type of intermediate such as 19 were detected by NMR
(Scheme 7).
8
Finally, the formation of tert-butylsulfinic tert-butyl-
an oxygen transfer reaction affording the corresponding
sulfonic thioanhydride (23) suggests that an oxygen
trioxides 29 or 21, respectively. The difference in reac-
tivity of 28 (22%) and 20 (13%) can be explained by steric
transfer reaction3 also takes place in this case. This
,8

9
96 J . Org. Chem., Vol. 61, No. 3, 1996
Derbesy and Harpp
NaIO 11
according to literature procedures. In all cases, the
Sch em e 10
4
final products were a mixture of di-tert-butyl di-, tri-, and
tetrasulfides that were characterized by NMR spectrometry
and GC analysis.
m -CP BA Oxid a tion of Di-ter t-bu tyl Dith iosu lfite (8).
A solution of m-CPBA (990 mg, 5.75 mmol, 1.4 equiv) in
methylene chloride (25 mL) was added dropwise to a cooled
solution (-40 °C) of 8 (0.93 g, 4.11 mmol) in CH
during 0.5 h under nitrogen. After a stirring period of 5 h at
40 °C, the mixture was concentrated to 10 mL by rotoevapo-
2 2
Cl (15 mL)
-
ration. The solution was cooled to -78 °C, and the m-CBA
that crystallized was collected. The solvent was removed in
vacuo to give an oily residue. Silica gel column chromatog-
raphy using a 10% ethyl acetate/hexanes solution afforded the
desired 6 (0.48 g, 1.97 mmol, 48%) as a colorless solid: mp
5
8-65 °C (lit.19 56.5-62.5 °C) (it is likely that 6 decomposes
1
near the melting point); H NMR (CDCl
3
) δ 1.465 (s, 9H), 1.395
1
3
(s, 9H), ppm; C NMR (CDCl
3
) δ 70.06, 49.95, 29.86, 24.18
2
0
•+
ppm; MS (EI, 70 eV, 30 °C) m/z (rel int) 242 (M , 0.4), 178
•
+
•+
•+
(
(
M
- S(O)
•+
3 5
S(O) , 20), 57 (t-Bu , 100), 41 (C H , 46).
2 2 4 10
, 6.6), 122 (t-BuS(O) , 14), 90 (C H S , 16), 64
+
2
P er a cetic Acid Oxid a tion of Di-ter t-bu tyl Dith iosu lfite
8). The oxidation of 8 (1.09 g, 4.82 mmol) was carried out
(
3 3
using CH CO H (40%) (-20 °C, 5 h, 1.4 equiv) according to
the procedure described above. The solvent was removed in
vacuo, and silica gel column chromatography using a 10% ethyl
acetate/hexanes solution gave 6 (0.41 g, 1.69 mmol, 35%) as a
colorless solid. Analytical data were consistent with previously
reported values.
and electronic effects as the tetrasulfide derivative is less
hindered and has an extra sulfur atom (Scheme 10).
A number of different mechanisms can take place, and
it is probably a combination of all of them that delivers
this rather complex mixture. In addition, products such
as 20 and 28 can participate in the decomposition and
even further complicate the study. As a consequence, the
more stable compounds, dioxide 6 and thiosulfinate 7a ,
are the major products observed at room temperature
because they are thermodynamically favored.15
DMD Oxid a tion of Di-ter t-bu tyl Dith iosu lfite (8). A
0.05 M solution of DMD in acetone (104 mL, 5.20 mmol, 1.2
equiv) was added dropwise to a cooled solution (-78 °C) of 8
(0.98 g, 4.33 mmol) in acetone (10 mL) during 30 min under
nitrogen. After a stirring period of 2 h at -78°C, the solvent
was removed in vacuo to give 6 (0.89 g, 3.68 mmol, 85%) as a
colorless solid. Analytical data were consistent with previously
reported values.
Decom p osition of ter t-Bu tylsu lfen ic ter t-Bu tylsu lfon ic
Exp er im en ta l Section 8
Th ioa n h yd r id e (6). The decomposition of 6 (87 mg, 0.36
mmol) in CDCl
3
was followed by 1H and C NMR at room
13
X-r a y Cr ysta llogr a p h ic Da ta for 6. Intensity data were
collected at room temperature on a AFC6S Rigaku diffracto-
meter using graphite-monochromated Cu KR (λ ) 1.54056 Å)
temperature. After 6 months under these conditions no
decomposition occurred. The decomposition of 6 (198 mg, 0.82
mmol) at 80 °C in CCl gave di-tert-butyl disulfide (16a ) as
4
the sole product after 15 h (identified by NMR and GC). The
condenser employed was equipped with a trap of wet pH paper.
Clear evidence of sulfur dioxide evolution was given by the
strongly acidic coloration of the pH paper by the end of the
decomposition.
1
6
radiation using the θ/2θ scan mode.
Oxid ized Der iva tives of Disu lfid es. The preparation
and characterization of the oxidized derivatives of disulfides
3
,5,8
mentioned here were previously reported.
Oxid ized Der iva tives of Tr isu lfid es. The preparation
and characterization of the sulfenic sulfinic thioanhydrides,
dithiosulfites, and sulfinic sulfonic thio- and dithioanhydrides
Syn th esis of ter t-Bu tylsu lfin ic Th ioa n h yd r id e9 (20).
m -CP BA Oxid a tion of Di-ter t-bu tyl Tr isu lfid e (16b).
Compound 16b (0.53 g, 2.52 mmol) was oxidized using m-
CPBA (12 h, -40 °C, 2.5 equiv) according to the procedure
previously described. The solvent was removed at -10 °C in
vacuo (2.5 mmHg) using a dry ice condenser rotoevaporator
to give an oily residue. Compound 20 (0.58 g, 2.42 mmol, 96%)
crystallized from n-pentane in the freezer: mp 60-85 °C (it
3
,8
mentioned here were previously reported.
Syn th esis of ter t-Bu tylsu lfen ic ter t-Bu tylsu lfon ic Th io-
a n h yd r id e (6). m -CP BA Oxid a tion of 7b. The oxidation
of 7b (1.09 g, 4.82 mmol) was carried out using m-CPBA (-40
C, 5 h, 1.4 equiv) according to the procedure in ref 8.
Compound 6 was formed in 48% yield, mp 58-65 °C (lit.
°
2
0
5
6.5-62.5 °C).
1
Attem p ted Nu cleop h ilic Oxid a tion of ter t-Bu tylsu lfen -
ic ter t-Bu tylsu lfin ic Th ioa n h yd r id e (7b). The oxidation
of 7b (0.4-0.6 g, 1.77-2.65 mmol) was attempted using H /1
and
is likely that 20 decomposes near the melting point). H and
13C NMR chemical shifts are reported in the literature.
9
2
O
2
Compound 20 was not stable enough to give consistent results
on MS.
1
7
17
17
18
N NaOH, t-BuOOH/1 N NaOH, KO
2
,
4
KMnO ,
Syn th esis of ter t-Bu tylsu lfon ic ter t-Bu tylsu lfin ic Th io-
a n h yd r id e (21). Compound 21 was prepared according to
the general oxidizing conditions summarized in ref 8 using
m-CPBA. The solid was crystallized from pentane as colorless
needles, mp 80-100 °C (probably decomposed before its
(
15) We thank a reviewer for a thoughtful suggestion concerning a
mechanism to explain the complex product formation of 8 and 13; it
involved a bimolecular process. A “Steven’s”-type of reaction was
portrayed involving a series of 1,2 shifts and a cross thioalkylation
using RSO
however, for a related molecule (the 1-oxide), the kinetics suggest a
unimoleulcar process (ref 8) and the formation of two oppositely
charged, ambident anions. That such anions are involved is not without
literature precedent; see ref 3a as well as: Schreiner, P. R.; Schleyer,
P. v. R; Hill, R. K. J . Org. Chem. 1993 58, 282. Schreiner, P. R.;
Schleyer, P. v. R.; Hill, R. K. J . Org. Chem. 1994, 59, 1849.
-
2
S . We cannot easily carry out kinetics on these systems;
(17) Adams, W.; Hass, W.; Lohray, B. B. J . Am. Chem. Soc. 1991,
113, 6202.
(18) (a) Ghosh, F.; Bartlett, P. D. J . Am. Chem. Soc. 1988, 110, 7499.
(b) Yomoji, N.; Takahashi, S.; Chida, S.-I.; Ogawa, S.; Sato, R. J . Chem.
Soc., Perkin Trans. 1 1993, 1995.
(16) The authors have deposited atomic coordinates for 6 with the
Cambridge Crystallographic Data Centre. The coordinates can be
obtained, on request, from the director of the Cambridge Crystal-
lographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ, UK.
(19) Block, E.; O’Connor, J . J . Am. Chem. Soc. 1974, 96, 3929.
(20) For comparative NMR values of related structures see refs 3b,
8, and 10.

Dialkyl Oxide Derivatives of Organotrisulfides
J . Org. Chem., Vol. 61, No. 3, 1996 997
melting point). MS (CI (NH
3
), 70 eV, 100 °C) m/z (rel int) 276
mmol, 95%) as a pale yellow solid that was recrystallized in
+
(
M(NH
4
) , 4.6; X-ray crystallographic analysis confirmed the
the freezer from a 5% CH
°C (28 probably decomposed before reaching its melting point)
2 2
Cl /n-pentane solution: mp 70-80
2
1
structure.
4
1
DMD Oxid a tion of Di-ter t-bu tyl Tr isu lfid e (16b). Com-
pound 16b (0.53 g, 2.52 mmol) was oxidized using DMD (0.05
M) (1.5h, -78 °C, 2.1 equiv) according to the procedure
described above. The solvent was removed at 0 °C in vacuo
(lit. mp 68-78 °C); H NMR (CDCl
3
) δ 1.41 (s, 9H) 1.37 (s,
9H) ppm; 13C NMR (CDCl
) δ 61.51, 61.20, 23.77, 23.50 ppm.
3
P r ep a r a tion of ter t-Bu tylsu lfen ic Isop r op ylsu lfon ic
Th ioa n h yd r id e (10). Isopropyl sulfonyl chloride (12) was
most conveniently prepared according to the Douglass proce-
dure13 using 4 equiv of acetic anhydride. A solution of
isopropylsulfonyl chloride (12) (0.257 g, 1.81 mmol) in 15 mL
of ether was added dropwise over a 1 h period under nitrogen
(2.5 mmHg) to give di-tert-butyl trisulfide 1,3-dioxide (20) as
a solid (0.60 g, 2.47 mmol, 98%): mp 60-85 °C (20 probably
decomposed before it reached its melting point). Analytical
data were consistent with previously reported values.
Decom p osition of ter t-Bu tylsu lfin ic Th ioa n h yd r id e
1
2
to an ice-cooled solution of tert-butyl hydrodisulfide (11)
0.221 g, 1.81 mmol) and pyridine (0.160 g, 2 mmol) in 25 mL
(
(
20). The decomposition of 20 (87 mg, 0.36 mmol) in CDCl
3
1
13
of ether. After a stirring period of an additional 5 h at room
temperature, the reaction mixture was washed with 2 × 25
mL portions of water, 3 × 25 mL portions of 1 N NaOH
solution, and 25 mL portions of water until neutral to pH
was followed by H and C NMR at room temperature. After
2 h no traces of 20 could be detected. The decomposition
1
1
13
products were identified by H and C NMR as a mixture of
3
b
6
(45%), 7a (40%), 21 (10%), and other undefined tert-butyl
4
paper. The organic phase was separated, dried with MgSO ,
derivatives (3%).
filtered, and evaporated. The solvent was removed in vacuo
to give an oily residue that was chromatographed on silica gel
using a 13% ethyl acetate/hexanes solution to give 10 (0.090
P r ep a r a tion of ter t-Bu tylsu lfen ic ter t-Bu tylsu lfin ic
Dith ioa n h yd r id e (7c). The oxidation of di-tert-butyl tetra-
sulfide (16c) (1.05 g, 4.34 mmol) was carried out using
m-CPBA (0 °C, 2 h, 1.2 equiv) according to the procedure
described above. The solvent was removed in vacuo to give
an oily residue that was chromatographed on alumina using
a 13% ethyl acetate/hexanes solution to give 7c (0.82 g, 3.17
g, 0.39 mmol, 22%) as a liquid. The first fraction was tert-
1
butyl tetrasulfide (16c) (65%). 10: H NMR (CDCl
(
6
(rel int) 246 (M + NH
intensity less than 3%.
3
) δ 3.52
) δ
1.85, 50.17, 30.34, 17.02 ppm; MS (CI, 70 eV, 100 °C) m/z
h, 1H), 1.42 (d, 6H), 1.38 (s, 9H) ppm; 13C NMR (CDCl
3
+
4
) , 100.0), all the other peaks have an
4
1
mmol, 73%) as a solid: mp 65-68 °C (lit. 68-78 °C); H NMR
1
3
(
6
CDCl
0.60, 49.91, 29.76, 23.82 ppm.
DMD Oxid a tion of ter t-Bu tyl Isop r op yl Dith iosu lfite8
3
) δ 1.33 (s, 9H) 1.32 (s, 9H) ppm; C NMR (CDCl ) δ
3
Decom p osition of ter t-Bu tylsu lfin ic Th ioa n h yd r id e
(
20) in th e P r esen ce of 1 Equ iv of Isop r op yl Su lfin yl
Ch lor id e (22). The decomposition of a solution 20 (300 mg,
.24 mmol) and isopropylsulfinyl chloride (22) (156 mg, 1.24
mmol, 1 equiv) in 10 mL of CHCl was carried out at room
(
13). The DMD (0.05 M, -78 °C, 1.2 equiv) oxidation of 13
1
(1.053 g, 4.97 mmol) was carried out according to procedure
8
3
3
.
After evaporation of the solvent, the crude mixture was
temperature for 12 h. The 1H and C NMR of the crude
mixtures clearly revealed the formation of tert-butylsulfinyl
chloride (23)13 (∼40%). The partial separation of the reaction
mixtures by silica gel chromatography and the careful analysis
of the crude NMR spectra allowed the detection of the
decomposition products of 20 and the mixed decomposition.
Low Tem p er a tu r e Detection of ter t-Bu tylsu lfen yl ter t-
Bu tyl vic-Disu lfoxid e (9). The oxidation of di-tert-butyl
dithiosulfite (8) (170 mg, 0.75 mmol) dissolved in 0.5 mL of
13
1
13
analyzed by H and C NMR and presented a close to
equimolar mixture of the four possible symmetrical and
unsymmetrical sulfenyl sulfonyl thioanhydrides, 15a (26%),
6
(25%), 10 (26%), and 15b (23%). These percentages were
estimated from the crude NMR spectrum according to the
calibration experiments reported earlier. Silica gel chroma-
tography using 10% ethyl acetate/hexanes as eluent gave a
mixture of 15a (85%) and 6 (15%) as a first fraction and 10
(
83%) and 15b (17%) as second fraction. The similarity of the
C NMR signals between the symmetrical and unsymmetrical
dry CDCl
drous m-CPBA (>99%) (194 mg, 1.12 mmol, 1.5 equiv) partially
dissolved in 2.5 mL of CDCl
3
was achieved by the slow addition of pure anhy-
1
3
trisulfide-dioxides and the previous characterization of one
of the two derivatives in each fraction allowed a clear NMR
characterization of the unknown trisulfide-dioxides. 15a :
3
. After 4 days at -60 °C the
8
reaction mixture was treated as described in the literature.
The NMR spectra are given in Table 2.
1
H
NMR (CDCl
3
) δ 3.38 (h, 1H), 1.48 (s, 9H), 1.39 (d, 6H) ppm;
C NMR (CDCl
) δ: 70.28, 43.32, 24.02, 21.92 ppm. 15b: 1
NMR (CDCl ) δ 3.36 (h, 1H), 3.58 (s, 1H), 1.41 (d, 6H), 1.36
1
3
3
H
Ackn owledgm en t. We thank Elf Aquitaine (France),
Atochem North America, King of Prussia, and the
Natural Sciences and Engineering Research Council of
Canada for financial support of this work.
3
1
3
3
(d, 6H) ppm; C NMR (CDCl ) δ 60.30, 42.88, 21.92, 16.36
ppm. Compound 15a was also isolated from the decomposition
of 20 in the presence of isopropylsulfinyl chloride (22).8
P r epar ation of ter t-Bu tylsu lfin ic Dith ioan h ydr ide (28).
The oxidation of di-tert-butyl tetrasulfide (16c) (0.50 g, 2.06
mmol) was carried out using m-CPBA (-40 °C, 6 h, 2.2 equiv)
according to the procedure described above. The solvent was
removed at low temperature (-20 °C) under high vacuum (2.5
mmHg) to give a diastereomeric mixture of 28 (0.53 g, 1.96
Su p p or tin g In for m a tion Ava ila ble: NMR spectra of
compound 10 as well as X-ray data of compound 6 (4 pages).
This material is contained in libraries on microfiche, im-
mediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
(21) Derbesy, G.; Harpp, D. N. submitted for publication.
J O950826X