Thermolysis of 3,3,5,5-tetramethyl-1,2,4-trithiolane 1-oxide: First matrix isolation of the HOSS· radical

Article abstract of DOI:10.1002/ejoc.201200146

Flash vacuum pyrolysis of 3,3,5,5-tetramethyl-1,2,4-trithiolane 1-oxide performed at 700 °C yields the 1-oxatrisulfan-3-yl radical (HOSS ^·) along with disulfur monoxide (S₂O) and diisopropyl sulfide, which were isolated in argon matrices at 10 K. Upon irradiation with UV light, the 1-oxatrisulfan-3-yl radical undergoes isomerization to the 1-oxatrisulfan-1-yl radical (HSSO^·). Both radicals were identified by comparison of their computed and experimental IR and UV/Vis spectra. In addition, density functional theory (DFT) computations offer a plausible explanation of the most likely reaction mechanism, suggesting that the initial step is a 1,3-H shift with simultaneous ring opening. A 1-oxatrisulfane derivative formed thereby undergoes fragmentations via a radical and a competitive concerted pathway leading to the observed final products. The same mechanism also governs the thermal fragmentation of di-tert-butyl disulfide S-oxide. Its pyrolysis at 700 °C affords an analogous set of products, including the 1-oxatrisulfan-3-yl radical (HOSS^·) as the key intermediate. 3,3,5,5-Tetramethyl-1,2,4-trithiolane 1-oxide was pyrolyzed in vacuo (FVP) yielding the 1-oxotrisulfan-1-yl radical, which was isolated in an Ar matrix at 10 K and subsequently characterized by spectroscopic methods (IR and UV/Vis). The same radical was obtained after vacuum pyrolysis of di-tert-butyl disulfide S-oxide.

Full text of DOI:10.1002/ejoc.201200146

FULL PAPER
DOI: 10.1002/ejoc.201200146
Thermolysis of 3,3,5,5-Tetramethyl-1,2,4-trithiolane 1-Oxide:
First Matrix Isolation of the HOSS· Radical
Hans Peter Reisenauer,*[a] Grzegorz Mloston,*[b] Jaroslaw Romanski,[b] and
Peter R. Schreiner[a]
Dedicated to Professor Dr. Helge Willner on the occasion of his 65th birthday
Keywords: Matrix isolation / Flash pyrolysis / High-temperature chemistry / Radicals / Reactive intermediates / Sulfur
Flash vacuum pyrolysis of 3,3,5,5-tetramethyl-1,2,4-trithiol-
ane 1-oxide performed at 700 °C yields the 1-oxatrisulfan-3-
yl radical (HOSS·) along with disulfur monoxide (S2O) and
diisopropyl sulfide, which were isolated in argon matrices at
10 K. Upon irradiation with UV light, the 1-oxatrisulfan-3-
yl radical undergoes isomerization to the 1-oxatrisulfan-1-yl
radical (HSSO·). Both radicals were identified by comparison
of their computed and experimental IR and UV/Vis spectra.
In addition, density functional theory (DFT) computations of-
fer a plausible explanation of the most likely reaction mecha-
nism, suggesting that the initial step is a 1,3-H shift with si-
multaneous ring opening.
A 1-oxatrisulfane derivative
formed thereby undergoes fragmentations via a radical and a
competitive concerted pathway leading to the observed final
products. The same mechanism also governs the thermal
fragmentation of di-tert-butyl disulfide S-oxide. Its pyrolysis
at 700 °C affords an analogous set of products, including the
1-oxatrisulfan-3-yl radical (HOSS·) as the key intermediate.
S-sulfide (thiosulfine) (4a) and thioformaldehyde (5a). The
parent thiosulfine 4a was converted into dithiirane (6a)
upon photolysis with λ Ͼ 385 nm. Prolonged irradiation
eventually led to a mixture s-cis and s-trans rotamers of
dithioformic acid (7) (Scheme 2).[2a] A similar pathway was
observed for the thermolysis of 3,3,5,5-tetramethyl-1,2,4-
trithiolane (1b) in the gas phase at 500–700 °C. In this case,
the initially formed thioacetone S-sulfide (4b) underwent
cyclization yielding 3,3-dimethyldithiirane (6b), which was
found together with a mixture of isomeric products iden-
tified as s-cis- and s-trans-isopropenyl disulfane (8)
(Scheme 2).[2b]
Introduction
Sulfur-rich heterocycles with diverse ring sizes are com-
mon in nature. In this group of compounds, 1,2,4-trithiol-
anes as well as their S-oxides were isolated from Shitake
mushrooms,[1a,1b] red algae Hondria californica,[1c] hyper-
thermophilic archea,[1d] and other sources.[1e–1g] In a series
of recent papers we described thermal reactions of 1,2,4-
trithiolanes 1a,b as well as their 4-oxides 2a,b and 1-oxides
3a,b (Scheme 1).
Scheme 1. 1,2,4-Trithiolanes 1 and their S-oxides 2 and 3.
We demonstrated that gas-phase thermolysis of 1a occurs
through [3+2] cycloreversion leading to thioformaldehyde
[a] Justus Liebig University, Institute of Organic Chemistry,
Heinrich-Buff-Ring 58, 35392 Giessen, Germany
Fax: +49-641-99-34209
E-mail: Hans.P.Reisenauer@org.chemie.uni-giessen.de
[b] University of Lodz, Section of Heteroorganic Compounds,
Tamka 12, 91-403 Lodz, Poland
Scheme 2. Thermal fragmentation of 1,2,4-trithiolanes 1 leading to
thiosulfines 4 and dithiiranes 6.
When 1,2,4-trithiolane 4-oxide (2a) was used as a sub-
strate for the thermolysis reaction performed in the gas
phase at 650 °C, the products collected in the matrix were
identified as a mixture of thioformaldehyde S-oxide (9a)
Fax: +48-42-665-51-62
E-mail: gmloston@uni.lodz.pl
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200146.
3408
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 3408–3415

Thermolysis of 3,3,5,5-Tetramethyl-1,2,4-trithiolane 1-Oxide
and thiosulfine 4a as the major components. These results
can also be explained by invoking a [3+2] cycloreversion of
the starting material 2a. In this case, 9a can be recognized
as a dipolarophile, and 4a is the corresponding 1,3-di-
pole.[2c] The same reaction pathway was observed for
3,3,5,5-tetramethyl-1,2,4-trithiolane 4-oxide (2b). The ma-
trix-isolated thioacetone S-oxide (9b) was subsequently
used as a substrate to generate 3,3-dimethyloxathiirane
(10b) (Scheme 3).[2d]
Scheme 5. Formation of 1-oxatrisulfane through thermolysis of di-
sulfane S-oxide 12.
means of vacuum pyrolysis in combination with matrix iso-
lation. Moreover, there are no reports available on the ther-
mal fragmentations of this type of thiosulfinates by using
matrix-isolation spectroscopy. This approach is a powerful
tool for studying reactive intermediates involved in this type
of thermal transformation of thiosulfinates.
Results and Discussion
Oxidation of 3,3,5,5-tetramethyl-1,2,4-trithiolane (1b)
with m-chloroperbenzoic acid (m-CPBA) in dichlorometh-
ane solution at room temperature leads to a mixture of S-
oxides 2b and 3b that can be easily separated by chromatog-
raphy.[5] Analytically pure 3b was used for the pyrolysis ex-
periments described in this paper.
Scheme 3. Thermal fragmentation of 1,2,4-trithiolane 4-oxides 2
leading to sulfines 9 and formation of oxathiiranes 10.
However, when 1,2,4-trithiolane 1-oxide (3a) was
thermolyzed in the gas phase at 700 °C, a different reaction
was observed. Spectroscopic analysis of the matrix-isolated
products led to the identification of thioformaldehyde S-
oxide (sulfine) (9a) formed together with dithio-
formic acid (7) (Scheme 4). The reaction mechanism was
elucidated by DFT computations at the B3LYP/6-
311+G(3df,3pd) level of theory (see Experimental Section
for details), and – in this case – the initial step is an allowed
1,4-H shift leading to open-chain intermediate 11, which
spontaneously splits into a mixture of final products 7 and
9a (Scheme 4).[2c] Structure 3a can also be considered as
a cyclic thiosulfinate, and the initial step of the observed
fragmentation corresponds to the typical reactivity of S-
alkyl alkanethiosulfinates under pyrolytic conditions.[3]
In a typical experiment, an evaporated sample of 3b was
pyrolyzed in high vacuum at 500–800 °C (empty quartz
tube; inner diameter: 8 mm; heated zone: 5 cm). The pyroly-
sis products, diluted with a large excess of argon, were con-
densed on the cooled spectroscopic window (BaF2 or CsI)
of the cryostat at 10 K. The matrices formed thereby were
investigated by means of FTIR (4000–300 cm–1) and UV/
Vis spectroscopy (850–200 nm). The FTIR spectrum re-
corded after a typical pyrolysis experiment carried out at
700 °C is presented in Figure 1. Two absorption bands lo-
cated at 1157 and 673 cm–1 are attributed to disulfur oxide
S2O (14).[6] Another band at 1275 cm–1 indicates the pres-
ence of a moderate amount of thioacetone (5b).[7] Another
set of bands with the most intense absorption located at
893 cm–1 is attributed to diisopropenyl sulfide (15). Because
this compound is unknown, we also independently pre-
pared it by conducting large-scale vacuum pyrolysis and
subsequently identified it by high-resolution mass spec-
trometry and NMR spectroscopy. A sample of 15 was con-
densed with Ar at 10 K, and the recorded IR bands were
identical to those found in the original pyrolysate of 3b.
Four other bands observed in the IR spectrum at 3540,
Scheme 4. Mechanism of thermal fragmentation of 1,2,4-trithiol-
ane 1-oxide (3a).
The goal of this study was to investigate the behavior of 1092, 729, and 659 cm–1 can be assigned to 1-oxatrisulfanyl
3,3,5,5-tetramethyl-1,2,4-trithiolane 1-oxide (3b), which can radical [HO–S–S·] (16) by comparison with the computed
also be considered as a type of an S-alkyl alkanethiosulfin- IR spectrum (vide infra).
ate with β-hydrogen atoms. In addition, di-tert-butyldisul-
The corresponding UV/Vis spectrum revealed two strong
fane S-oxide (12) was included in this study as an easily absorption bands located at λmax = 290 and 385 nm (Fig-
available model representative for thiosulfinates with a sim- ure 2). The first absorption displays a resolved vibrational
ilar substitution pattern (Scheme 5).
structure that belongs to 14, whereas the second band can
Although experimental and theoretical studies on the be assigned to radical 16, the existence of which in the gas
fragmentation of 12 leading to 1-oxatrisulfane (13) have phase was previously proven by neutralization-reionization
been reported,[4] this process has never been investigated by mass spectrometric measurements.[8] The absorption band
Eur. J. Org. Chem. 2012, 3408–3415
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3409

H. P. Reisenauer, G. Mloston, J. Romanski, P. R. Schreiner
FULL PAPER
Figure 1. Matrix-isolated pyrolysis products of 3,3,5,5-tetramethyl-1,2,4,trithiolane 1-oxide (3b) (argon matrix, 10 K, pyrolysis temperature
700 °C).
Figure 2. UV/Vis spectra of matrix-isolated pyrolysis products of 3,3,5,5-tetramethyl-1,2,4,trithiolane 1-oxide (3b) (argon matrix, 10 K,
pyrolysis temperature 700 °C) before (full line) and after (dashed line) photolysis (λ = 405 nm). Inset: Computed [TD-UB3LYP/6-
311+G(3df,3pd)] electronic transitions for 16.
of 16 observed experimentally was in good agreement with IR absorption bands of 14 and the parallel formation of
the most intense computed electronic transitions at λmax
=
new IR absorption bands of the cyclic S2O isomer 17.[6c]
377 nm (f = 0.0294).
Photolysis of the matrix with wavelengths corresponding
To gain more information on the photochemical reactiv- to the broad absorption band of 16 either at λ = 366 or at
ity of the matrix-isolated pyrolysis products, the mixture 405 nm induced a 1,3-H shift leading to the isomeric radical
was photolyzed at 10 K with monochromatic light of dif- [HS–S–O·] (18). This photoisomerization is reversible, and
ferent wavelengths. Thus, irradiation with λ = 290 nm re- radical 16 can be regenerated upon irradiation at λ =
sulted in the disappearance of the corresponding UV and 546 nm (Scheme 6).
3410
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 3408–3415

Thermolysis of 3,3,5,5-Tetramethyl-1,2,4-trithiolane 1-Oxide
Scheme 6. Thermolysis of 3,3,5,5-tetramethyl-1,2,4-trithiolane 1-oxide (3b) and subsequent photolysis of matrix-isolated products.
Comparison of computed and experimental IR absorp- 18 were computed at the UB3LYP/6-311+G(3df,3pd) and
tion bands of isomeric radicals 16 and 18 allowed their un- CCSD(T)-FC/cc-pVTZ level of theory (Figure 4).
ambiguous identification (Figure 3). Whereas in the case of
In accordance with earlier theoretical studies of the iso-
16 the O–H stretching frequency appears at 3540 cm–1, 18 meric radicals 16 and 18,[8] the CCSD(T)-FC/cc-pVTZ-op-
displays a weak S–H stretching band at 2560 cm–1. More- timized geometry shows a planar (Cs) cisoid structure for
over, the strongest absorption at 730 cm–1 is attributed to 16 as an energy minimum. Computation of the internal ro-
the S–O stretching vibration in 16, and the corresponding tation energy profile identified the planar transoid confor-
S–O stretching band in 18 was found at 1091 cm–1. The mation as a transition structure associated with a barrier of
observed large wavenumber difference of 361 cm–1 proves a ca. 2 kcalmol–1 (see the Supporting Information, Fig-
significant enhancement of the SO bond order in 18. The ure S2). In contrast to 16, computations indicate a C1 mini-
latter absorption overlaps with the SOH deformation mode mum structure for 18 with a torsional angle of 63°. In ad-
in isomer 16, resulting in diminished intensities of both dition, the computed S–O bond in radical 18 is shorter, and
bands in the experimental difference spectrum. However, the S–S bond is longer than in the case of 16. The latter
replacement of argon as the matrix material with nitrogen displays an approximate 2:1 spin density distribution be-
leads to a characteristic shift of the corresponding absorp- tween the two sulfur atoms. Approximately the same
tion bands, and both bands can be observed undisturbed in amount of delocalization of the unpaired electron between
the difference spectrum (see the Supporting Information, the S and O atoms is computed for radical 18 (Figure 4).
Figure S1).
To extend the study of 3b, an analogous high-vacuum
The characteristic absorption attributed to the S–S pyrolysis experiment was carried out with di-tert-butyldi-
stretching vibration in 16 appears at 653 cm–1. In the case sulfane S-oxide (12) at 500 and 700 °C. The 500 °C experi-
of 18, the position of the corresponding absorption was ment led to the formation of 3-tert-butyl-1-oxatrisulfane
found at 458 cm–1. Structures of isomeric radicals 16 and (19), which was postulated in an earlier paper based on
Figure 3. Comparison of the computed [UB3LYP/6-311+G(3df,3pd)] and experimental IR spectra of radicals 18 (top) and 16 (bottom).
The experimental spectrum refers to a difference of spectra taken before and after radiation with 366 nm light.
Eur. J. Org. Chem. 2012, 3408–3415
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3411

H. P. Reisenauer, G. Mloston, J. Romanski, P. R. Schreiner
FULL PAPER
Figure 4. Geometry and spin densities (numbers in italic) of isomeric radicals 16 and 18 computed at the UCCSD(T)-FC/cc-pVTZ level
of theory [UB3LYP/6-311+G(3df,3pd) in parentheses]. Bond lengths are given in Å, angles in °.
trapping experiments with methyl propiolate.[4a] Its exis- values.[4] In addition, the IR absorption bands of matrix-
tence was also postulated as an intermediate that undergoes isolated 13 correspond closely to the computed IR spec-
further conversion in the gas phase yielding 1-oxatrisulfane trum (Figure 5). The computed IR-band intensities of 14,
(13). The structure of 13 was elucidated by analysis of its 16, 21, and 13 allow an approximate molar ratio of these
rotational spectrum as well as that of its isotopologues.[4b,4c] compounds in the matrix to be estimated as 6:2:2:1.
Recently, the mechanism of thermolysis of 12, leading to
The IR absorption bands of matrix-isolated parent 1-
13 and S2O (14), was studied by means of computational oxatrisulfane (13), which are very close to those of radical
methods.[9] In accordance with earlier reported results,[4a,4b] 16, deserve a short comment. In the case of 13, the OH
the experiment conducted at 500 °C leads to elimination of absorption band is found at 3560 cm–1, which is close to the
isobutene (20) and the formation of 1-oxatrisulfane deriva- analogous absorption of 16 (3540 cm–1). The S–O stretch-
tive 19 (Scheme 7).
ing vibration (13: 736 cm–1) is also only slightly shifted to
a higher value (16: 730 cm–1). Furthermore, the S–S stretch-
ing in 13 (511 cm–1) is significantly lower than that of 16
(653 cm–1). In addition, two new absorption bands attrib-
uted to the torsional mode of the HOSS fragment were ob-
served at 386 and 416 cm–1. However, due to the very low
intensity, the S–H stretching absorption band in 13 could
not be detected.
The presented data show that the distributions of prod-
ucts 16, 13, and 14 are very similar in both experiments
performed either with 3b or with 12. This important finding
allows us to conclude that, in both cases, the reaction path-
ways are governed by analogous mechanisms in which sul-
fur radicals play a key role. The appearance and role of the
sulfur radical 16 was not considered in either experimental
Scheme 7. Thermolysis of disulfane S-oxide 12 at different tem-
peratures.
The experimental and computed IR spectra are in good or computational works focusing on the thermolyses of
agreement, thus lending confidence to the correct assign- thiosulfinates.[4,9]
ment of the postulated structure 19. It is worth noting that,
Based on the experimental findings and computational
in the experimental IR spectrum, bands of the radical 16 results, we suggest a fragmentation mechanism of 3,3,5,5-
and disulfur oxide (14), which were formed in trace tetramethyl-1,2,4-trithiolane 1-oxide (3b), as presented in
amounts, were also observed (see the Supporting Infor- Scheme 8. The first step is an electrocyclic ring opening
mation, Figure S3).
with a simultaneous H-shift yielding 1-oxatrisulfane deriva-
When the thermolysis experiment was carried out at tive 22. It corresponds to the mechanistic pathway pos-
700 °C, only a very small amount of 19 formed. Instead, tulated on the basis of computations carried out for 12.[9]
the most intense absorption bands evidenced the formation In both cases, the computed activation energies are below
of 14 and radical 16 as the major products. In addition, 30 kcalmol–1. The reaction is slightly endothermic, with
tert-butyl radical (21) and 1-oxatrisulfane (13) were present ΔHr = 3.3 kcalmol–1. In contrast to 19, formation of 22
in the mixture. Radical 21 could be identified by its charac- results from an intramolecular reversible process, which is
teristic IR absorption bands.[10] The absorption bands as- very likely the reason why the latter intermediate cannot be
signed to 13 are in agreement with their reported gas-phase detected in the matrix. We assume that 22 undergoes very
3412
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 3408–3415

Thermolysis of 3,3,5,5-Tetramethyl-1,2,4-trithiolane 1-Oxide
Figure 5. Matrix-isolated pyrolysis products of 12 (700 °C, Ar, 11 K); bands of isobutene (20) have been subtracted for clarity. Top:
Computationally simulated IR spectrum [B3LYP/6-311+G(3df,3pd)] of a mixture consisting of S2O (14), radical 16, tert-butyl radical
(21), and 1-oxatrisulfane (13) in a molar ratio of 6:2:2:1.
Scheme 8. Proposed mechanism of the thermal fragmentation of 3,3,5,5-tetramethyl-1,2,4-trithiolane 1-oxide (3b) based on the results of
B3LYP/6-311+G(3df,3pd) computations.
fast fragmentation through two competitive pathways pre- undergoes conversion into 1-oxatrisulfane (13), which was
sented as “A” and “B” (Scheme 8).
found in the matrix in trace amounts. The route through
Whereas pathway A leads to the formation of radical 16 which disulfur oxide (14) is formed is not completely clear.
and very likely to its cofragment radical 23, the IR bands One of the possible pathways could be a synchronous H2
of which – according to DFT computations – overlap with elimination from cis-conformer of 13. However, the com-
those of sulfide 15 and therefore could not be assigned un- puted activation energy for this type of H2 elimination is
ambiguously, pathway B follows a concerted H-shift of a rather high (ΔH‡ = 61.0 kcalmol–1). Alternative formation
retro-ene fragmentation, yielding diisopropyl sulfide (15) of radical 16 either from 24 or from 1-oxotrisulfane (13)
and parent thiosulfinic acid (24). Both processes require by homolytic cleavage of the corresponding S–H bonds are
comparable activation energies and reaction enthalpies energetically less favorable processes, with bond dissoci-
(Scheme 8). However, product 24 could not be detected in ation energies (BDEs) of 40.4 and 62.8 kcalmol–1, respec-
the matrix and, very likely, in the gas phase, immediately tively.
Eur. J. Org. Chem. 2012, 3408–3415
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3413

H. P. Reisenauer, G. Mloston, J. Romanski, P. R. Schreiner
FULL PAPER
some cases, matrices were subjected to irradiation with selected
wavelengths (see the Results and Discussion).
It should be noted that the alternative pathway B corre-
sponds to that computed for the fragmentation of 19.[9]
However, competitive homolytic fragmentation correspond-
ing to pathway A (Scheme 8) has not been considered in
this report.[9] The results of the experiment conducted at
700 °C (Scheme 7) underscores the importance of the “radi-
cal” pathway A.
Semipreparative Flash Vacuum Pyrolysis of 3,3,5,5-Tetramethyl-
1,2,4-trithiolane 1-Oxide (3b): Compound 3b (30 mg) was sublimed
at a pressure of 10–2 mbar from a storage vial at room temperature
through an empty quarz tube (outer diameter 12 mm, length of
the heated zone 60 mm) heated to 700 °C. The pyrolysis tube was
connected to a vacuum line by a U-shaped trap that was cooled to
liquid nitrogen temperature during the pyrolysis. After a pyrolysis
time of 2 h, the trap was warmed, and the volatile reaction products
were transferred to an IR gas cell and checked by IR spectroscopy.
At –40 °C, a small amount of acetylene was released. The volatile
products at room temperature consisted of mainly diisopropenyl
disulfide (15). A part of this fraction was mixed with argon in a
ratio of 1:500 and checked by matrix-isolation IR spectroscopy.
Another part was checked by high-resolution mass spectrometry;
the rest was mixed with CDCl3 and analyzed by NMR spec-
troscopy.
Conclusions
The present study shows that thermal fragmentations of
3,3,5,5-tetramethyl-1,2,4-trithiolane 1-oxide (3b) and di-
tert-butyldisulfane S-oxide (12) occur in a similar manner
to other S-alkyl alkanethiosulfinates with β-hydrogen
atoms. The initial step of the observed fragmentation is a
synchronous H-shift, resulting in the formation of the cor-
responding 1-oxatrisulfane derivative together with an ole-
finic fragment. The intermediates formed thereby undergo
further fragmentation, yielding 1-oxatrisulfan-3-yl radical
(16) in both cases. For the first time, vacuum pyrolysis in
combination with matrix isolation allowed the detection
and spectroscopic characterization of this type of sulfur-
containing radicals. These findings lead to the final conclu-
sion that, besides a concerted reaction pathway postulated
in earlier reports,[4,9] a competitive radical process does play
an important role. For the first time, the key intermediate
16 was isolated in an argon matrix and identified spectro-
scopically.
Diisopropenyl Sulfide (15): IR (gas phase): ν = 3114 (m), 2983–
˜
1
2870 (m), 1606 (m), 1447 (m), 1380 (m), 894 (vs) cm–1. H NMR
(400 MHz, CDCl3): δ = 5.04 (s, 2 H), 4.96 (s, 2 H), 1.90 (s, 6 H)
ppm. 13C NMR (150.9 MHz, CDCl3): δ = 143.4 (C), 113.0 (CH2),
20.6 (CH3) ppm. HRMS: calcd. for C6H10S 114.0503; found
114.0503.
Computational Methods: All geometries were fully optimized and
characterized as minima or transition structures by means of ana-
lytical harmonic vibrational frequency computations at the B3LYP/
6-311+G(3df,3pd) level of theory.[12] In addition, for radicals 16
and 18, full geometry optimization and harmonic frequency com-
putations were performed at the CCSD(T)-FC/cc-pVTZ level. The
Gaussian Program Suite was used for all computations.[13]
Supporting Information (see footnote on the first page of this arti-
cle): Spectroscopic data, tables of observed and computed IR spec-
tra, computational results.
Experimental Section
Materials: 3,3,5,5-Tetramethyl-1,2,4-trithiolane 1-oxide (3b) was
prepared according to known protocols by oxidation of the corre-
sponding 1,2,4-trithiolane 1b using m-CPBA.[5b] An analogous pro-
tocol was used for the preparation of 12 from commercially avail-
able di-tert-butyldisulfane.[11]
Acknowledgments
This study was supported by the Deutscher Akademischer Aus-
tausch-Dienst (DAAD) within the partnership between the Univer-
sity of Lodz and the Justus-Liebig University.
Matrix Isolation Experiments: The cryostat used for the matrix iso-
lation studies was an APD Cryogenics HC-2 closed-cycle refrigera-
tor system fitted with CsI windows for IR and BaF2 windows for
UV/Vis measurements. The matrix temperature was measured and
controlled by a Scientific Instruments 9600-1 silicon diode tempera-
ture controller. For irradiations, a mercury high-pressure lamp
(HBO 200, Osram) with a monochromator (Bausch & Lomb) was
used (bandwidth ca. 10 nm). IR spectra were recorded with a
Bruker IFS 55 FTIR spectrometer (4500–300 cm–1, resolution
0.7 cm–1), and UV/Vis spectra were recorded with a JASCO V-670
spectrophotometer. For the combination of high-vacuum flash pyr-
olysis (HVFP) with matrix isolation, we employed a home-built,
water-cooled oven directly connected to the vacuum shroud of the
cryostat. The pyrolysis zone consisted of an empty quartz tube (in-
ner diameter 8 mm, length of heating zone 50 mm) resistively
heated by a thermo-coax wire. The temperature was controlled by
an Ni/CrNi thermocouple. In a typical FVP experiment, a sample
of 3,3,5,5-tetramethyl-1,2,4-trithiolane 1-oxide (3b) or di-tert-butyl
disulfide S-oxide (12) was evaporated from a precooled (3b: 0 °C,
12: –15 °C) storage vessel and pyrolyzed at 500–700 °C. Pyrolysis
products were condensed together with a large excess of argon or
nitrogen on a cold spectroscopic window. The obtained matrices
were analyzed by means of FTIR and UV/Vis spectroscopy. In
[1] a) K. Morita, S. Kobayashi, Chem. Pharm. Bull. 1967, 15, 988–
993; Ch.-Ch. Chen, Ch.-T. Ho, J. Agric. Food Chem. 1986, 34,
830–833; b) S. J. Wratten, D. J. Faulkner, J. Org. Chem. 1976,
41, 2465–2467; c) M. Ritzau, M. Keller, P. Wessels, K. O.
Stetter, A. Zeeck, Liebigs Ann. Chem. 1993, 871–876; d) A. W.
Brinkmann, H. Copier, J. J. M. De Leuw, S. B. Tjan, J. Agric.
Food Chem. 1972, 20, 177–181; e) R. Gmelin, R. Susilo, G. R.
Fenwick, Phytochemistry 1981, 20, 2521–2523; f) R. Kubec, J.
Velisˇek, M. Dolezal, V. Kubelka, J. Agric. Food Chem. 1997,
ˇ
45, 3580–3585; g) E. A. Shelaby, N. F. Nasr, S. M. El Sherief,
J. Med. Plants Res. 2011, 5, 922–931.
[2] a) G. Mloston´, J. Roman´ski, H. P. Reisenauer, G. Maier, An-
gew. Chem. 2001, 113, 401; Angew. Chem. Int. Ed. 2001, 40,
393–395; b) G. Maier, H. P. Reisenauer, J. Roman´ski, H. Pet-
zold, G. Mloston´, Eur. J. Org. Chem. 2006, 3721–3729; c) G.
Mloston, J. Romanski, M. L. McKee, H. P. Reisenauer, P. R.
Schreiner, Eur. J. Org. Chem. 2010, 2132–2137; d) H. P. Re-
isenauer, G. Mloston, J. Romanski, P. R. Schreiner, Eur. J. Org.
Chem. 2011, 6269–6275.
[3] a) T. Takata, T. Endo, “Thiosulphinc acids and esters” in The
chemistry of functional groups, vol. The chemistry of sulphinic
acids, esters and their derivatives (Ed.: S. Patai), J. Wiley & Sons,
3414
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 3408–3415

Thermolysis of 3,3,5,5-Tetramethyl-1,2,4-trithiolane 1-Oxide
Chichester, 1990, chapter 18, pp. 528–576; b) E. Block, Angew.
Chem. 1992, 104, 1158; Angew. Chem. Int. Ed. Engl. 1992, 31,
1135–1178.
[11]
[12]
T. Waag, Ch. Gelhaus, J. Rath, A. Stich, M. Leippe, T. Schir-
meister, Bioorg. Med. Chem. Lett. 2010, 20, 5541–5543.
a) A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100; b) C. Lee,
W. Yang, R. G. Parr, Phys. Rev. B 1993, 37, 785–789; c) B.
Miehlich, A. Savin, H. Stoll, H. Preuss, Chem. Phys. Lett. 1989,
157, 200–206.
[4] a) E. Block, J. Am. Chem. Soc. 1972, 94, 147–151; b) A.
Königshofen, M. Behnke, M. Hoverath, J. Halm, Z. Anorg.
Allg. Chem. 1999, 625, 1779–1786; c) M. Körber, U. Baum, J.
Halm, J. Gauss, T. F. Giesen, S. Schlemmer, J. Mol. Spectrosc.
2009, 257, 34–39; d) M. Körber, U. Baum, T. F. Giesen, S.
Schlemmer, J. Gauss, Inorg. Chem. 2009, 48, 2269–2272.
[5] a) W. Weigand, S. Bräutigam, G. Mloston, Coord. Chem. Rev.
2003, 245, 167–175; b) H. Petzold, S. Bräutigam, H. Görls,
W. Weigand, U. Uhlemann, R. Gessner, W. Kiefer, J. Popp, A.
Majchrzak, G. Mloston, Inorg. Chim. Acta 2004, 357, 1897–
1908.
[13] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-
gawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Starov-
erov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Strat-
mann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski,
G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D.
Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski,
D. J. Fox, Gaussian 09, revision B.01, Gaussian, Inc., Wall-
ingford, CT, 2010.
[6] a) A. G. Hopkins, F. P. Daly, C. W. Brown, J. Phys. Chem.
1975, 79, 1849–1851; b) S. Y. Tang, C. W. Brown, Inorg. Chem.
1975, 14, 2856–2858; c) W.-J. Lo, Y.-J. Wu, Y. P. Lee, J. Chem.
Phys. 2002, 117, 6655–6661; d) L. F. Phillips, J. J. Smith, B.
Meyer, J. Mol. Spectrosc. 1969, 29, 230–243; e) W.-J. Lo, Y. P.
Lee, J. Phys. Chem. A 2003, 107, 6944–6947.
[7] F. Bourdon, J.-L. Ripoll, Y. Valée, Tetrahedron Lett. 1990, 31,
6183–6184.
[8] G. de Petris, M. Rosi, A. Troiani, J. Phys. Chem. A 2007, 111,
6526–6533.
[9] B. Mondal, D. Mandal, A. K. Das, J. Phys. Chem. A 2011, 115,
Received: February 7, 2012
Published Online: May 16, 2012
3068–3078.
[10] J. Pacansky, J. S. Chang, J. Chem. Phys. 1981, 74, 5539–5546;
B. Schrader, J. Pacansky, J. Phys. Chem. 1984, 88, 4069–4073.
Eur. J. Org. Chem. 2012, 3408–3415
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3415