Novel Type Elimination Reactions of Sulfoxides Bearing Several Heteroaromatics: Trapping of Sulfines with 2,3-Dimethyl-1,3-butadiene

Article abstract of DOI:10.1021/jo990610l

Previously we reported the novel thioaldehydes generation via thermolyses of phenacyl sulfoxides bearing some heteroaromatics. Thermolysis of sulfoxides (1a,b and 2a-4a) bearing other heterocycles such as thiadiazole, triazole, and tetrazole in the presence of 2,3-dimethyl-1,3-butadiene in dioxane at 100 °C led to the unexpected products 6-substituted-5,6-dihydro-3,4-dimethyl-2H-thiapyran 1-oxide (5a,b). These products were considered to be formed by the Diels-Alder reaction of the diene with the sulfines formed initially by the thermal decomposition of the sulfoxides. The rate acceleration and the improvement of the yield by addition of 1.5 equiv of triethylamine, especially in the case of ethoxycabonylmethyl sulfoxide 1c, was observed. The cis-trans selectivity for sulfine cycloadducts was also studied by NMR spectrometry. The reactions of α-substituted phenacyl sulfoxides 1d-f bearing a phenyl-substituted tetrazolyl group in the presence of the same diene were also studied.

Full text of DOI:10.1021/jo990610l

6730
J . Org. Chem. 1999, 64, 6730-6737
Novel Typ e Elim in a tion Rea ction s of Su lfoxid es Bea r in g Sever a l
Heter oa r om a tics: Tr a p p in g of Su lfin es w ith
2,3-Dim eth yl-1,3-bu ta d ien e
Hiroyuki Morita,*,† Masahiro Takeda, Toshiaki Yoshimura, Takayoshi Fujii, Shin Ono, and
Choichiro Shimasaki
Department of System Engineering of Materials and Life Science, Faculty of Engineering,
Toyama University, 3190 Gofuku, Toyama 930-8555, J apan
Received April 8, 1999
Previously we reported the novel thioaldehydes generation via thermolyses of phenacyl sulfoxides
bearing some heteroaromatics. Thermolysis of sulfoxides (1a ,b and 2a -4a ) bearing other
heterocycles such as thiadiazole, triazole, and tetrazole in the presence of 2,3-dimethyl-1,3-butadiene
in dioxane at 100 °C led to the unexpected products 6-substituted-5,6-dihydro-3,4-dimethyl-2H-
thiapyran 1-oxide (5a ,b). These products were considered to be formed by the Diels-Alder reaction
of the diene with the sulfines formed initially by the thermal decomposition of the sulfoxides. The
rate acceleration and the improvement of the yield by addition of 1.5 equiv of triethylamine,
especially in the case of ethoxycabonylmethyl sulfoxide 1c, was observed. The cis-trans selectivity
for sulfine cycloadducts was also studied by NMR spectrometry. The reactions of R-substituted
phenacyl sulfoxides 1d -f bearing a phenyl-substituted tetrazolyl group in the presence of the same
diene were also studied.
In tr od u ction
foxides is interesting from the synthetic and mechanistic
aspect, and hence, we synthesized several â-ketosulfox-
ides and further studied their behavior under similar
conditions.
Recently, we have reported that the thermolytic reac-
tion of phenacyl sulfoxide bearing a heterocycle such as
the 2-N-oxypyridyl, 2-pyridyl, 4-pyridyl, or 2-benzothia-
zolyl group in the presence of 2,3-dimethyl-1,3-butadiene
at 100 °C for 24 h afforded 6-benzoyl-5,6-dihydro-3,4-
dimethyl-2H-thiapyran.1 This product was considered to
be formed by the hetero Diels-Alder reaction of the diene
with the thioaldehyde, which was probably formed by
elimination of the sulfenate ester intermediate generated
by rearrangement of the starting sulfoxide. To gain
further details and limitations of this reaction, the
thermolytic reactions of phenacyl sulfoxide derivatives
bearing another heterocycle, the 5-(1-phenyl)tetrazolyl
group, under the same conditions were studied. The
products in the case of the thermal reaction of 5-(1-
phenyl)-1,2,3,4-tetrazolyl phenacyl sulfoxide (1a ) in the
presence of 2,3-dimethyl-1,3-butadiene at 100 °C for 3 h
were 1-phenyl-1,2,3,4-tetrazole (6) and 6-benzoyl-5,6-
dihydro-3,4-dimethyl-2H-thiapyran 1-oxide (5a ), in con-
trast to the previous results.1 The product 5a is consid-
ered to be formed by the Diels-Alder reaction of sulfine
and butadiene.2
Resu lts a n d Discu ssion s
Rea ction of P h en a cyl Su lfoxid es 1a -4a Bea r in g
Heter oa r om a tics w ith 2,3-Dim eth yl-1,3-bu ta d ien e.
Thermolytic reactions of phenacyl sulfoxides having
1-phenyltetrazolyl, 1-methyltetrazolyl, 2-methylthiadia-
zolyl, and 4-methyl-1,2,4-triazolyl groups were carried out
in dioxane at 100 °C for 3 h in the presence of 2,3-
dimethyl-1,3-butadiene. The results are summarized in
Table 1. The thermolytic reaction of 5-(1-phenyl)-1,2,3,4-
tetrazolyl phenacyl sulfoxide (1a ) in the presence of diene
afforded 6-benzoyl-5,6-dihydro-3,4-dimethyl-2H-thia-
pyran 1-oxide (5a ) and 1-phenyl-1,2,3,4-tetrazole (6) in
quantitative and 60% yields, respectively (entry 1). In
the case of 5-(1-methyl)-1,2,3,4-tetrazolyl phenacyl sul-
foxide (2a ), the reaction also afforded 5a and 1-methyl-
1,2,3,4-tetrazole (7) in quantitative yields (entry 2). The
reaction of 5-(2-methyl)-1,3,4-thiadiazolyl phenacyl sul-
foxide (3a ) afforded 5a and 2-methyl-1,3,4-thiadiazole (8)
in rather poor yields of 20 and 30%, respectively (entry
3). This reaction for 16 h afforded the corresponding
products in quantitative yields (entry 4). In the case of
3-(4-methyl)-1,2,4-triazolyl phenacyl sulfoxide (4a ), 5a
and 4-methyl-1,2,4-triazole (9) were obtained quantita-
tively (entry 5). The yields of these reactions were
Sulfines are generally formed by the oxidation of
thioketones3 or the elimination reactions of sulfinyl halide
derivatives.4 Therefore, the facile direct formation of
sulfine under mild thermal conditions from â-ketosul-
† Phone: +81-76-445-6851. FAX: +81-76-445-6703. e-mail: morita@
eng.toyama-u.ac.jp.
(1) Morita, H.; Takeda, M.; Kamiyama, H.; Hashimoto, T.; Yoshimu-
ra, T. Shimasaki, C.; Tsukurimichi, E. Tetrahedron Lett. 1996, 37,
3739. (b) Proceedings of the Seventeenth International Symposium on
the Organic Chemistry of Sulfur (ISOCS-17): Takeda, M.; Yoshimura,
T.; Shimasaki, C.; Morita, H. Phosphorus, Sulfur, and Silicon 1997,
120 and 121, 409. (c) Morita, H.; Takeda, M.; Fujii T.; Yoshimura, T.;
Shimasaki, C. J . Org. Chem. 1997, 62, 9018.
(2) (a) Boger, D. L.; Weinreb, S. M. Hetero Diels-Alder Methodology
in Organic Synthesis; Academic Press Inc.: London, 1987; pp 120-
145. (b) Zwanenburg, B. Recl. Trav. Chim. Pays-Bas 1982, 101, 1.
(3) (a) Kitamura, R. J . Pharam. Soc. J pn. 1938, 58, 246. (b)
Kitamura, R. Chem. Zentralbl. I 1939, 4606. (c) Kitamura, R. Chem
Abstr. 1960, 33, 1726. (d) Walter, W. J ustus Liebigs Ann. Chem. 1969,
633, 35. (e) Zwanenburg, B.; Thijs, L.; Strating, J . Tetrahedron Lett.
1969, 51, 4461.
(4) (a) Sheppard, W. A.; Diekmann, J . J . Am, Chem. Soc. 1964, 86,
1891. (b) Strating, J .; Thijs, L.; Zwanenburg, B. Recl. Trav. Chim. Pays-
Bas 1962, 83, 1. (c) Lenz, B. G; Regeling, H.; van Rozendaal, H. L. M.;
Zwanenburg, B. J . Org. Chem. 1985, 50, 2930.
10.1021/jo990610l CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/06/1999

Trapping of Sulfines with 2,3-Dimethyl-1,3-butadiene
J . Org. Chem., Vol. 64, No. 18, 1999 6731
Ta ble 1. Th er m olysis of P h en a cyl Su lfoxid e Bea r in g
Heter oa r om a tics in th e P r esen ce of 10 equ iv of
Bu ta d ien e in Dioxa n e a t 100 °C
benzene; however, the trans isomer could not be obtained
in a pure form and was shown to contain ca. 40% of cis
5b by NMR spectrometry. The determination of cis 5b
was easily performed by the comparison of the spectra
of cis and trans 5a , while assignment of trans 5b was
possible by subtraction of the signals of the pure cis
isomer from the spectrum of the impure sample. In the
case of 5c, both cis and trans isomers could not be
obtained as pure material. Therefore, assignment of the
cis and trans isomers was established by comparing the
NMR spectra of pure cis and trans 5a and an 1H-1H
COSY experiment of this mixture. In the COSY spectrum
the set of correlation peaks facilitated finding the set of
peaks of both cis and trans 5c, and finally the ratio
determination of cis and trans isomers was performed
by the integration of both C-6 protons. Thus, the cis-
trans ratio of 5a obtained in each entry was found to be
1:5 in entry 1, 1:3.7 in entry 2, 1:30.0 in entry 4, and
2.1:1 in entry 5 by NMR spectrometry (cf. Table 1).5
In general, sulfines are known to have two geometrical
isomers, i.e., E and Z,6 with a significant barrier of
interconversion between them.7 The preferred retention
of configuration of the starting sulfines is also known to
be common during the formation of cycloadducts in the
Diels-Alder reaction.2a,b,5b For example, the reaction of
(Z)-chloro phenyl thioketone S-oxide with 2,3-dimethyl-
1,3-butadiene afforded mainly cis cycloadduct (cis:trans
) 3.9: 1),8 while in the case of (E)-chloro phenyl thio-
ketone S-oxide cis and trans cycloadducts are reported
to be formed in 12 and 67% yields (cis:trans ) 1:5.6),
respectively.7 The kinetic studies of these cyclizations
were also reported: activation parameter ∆Hq )18.4 kcal/
mol and ∆Sq ) -15 eu for (E)-chloro phenyl thioketone
S-oxide; ∆Hq) 20 kcal/mol, ∆Sq ) -15 eu for (Z)-chloro
phenyl thioketone S-oxide. The negative ∆Sq values for
both cases, although not very large, are suggestive of a
concerted cycloaddition process.8
More recently, for the monoaryl and monoalkyl sulfines,
Barbaro et al. reported that the rate of isomerization of
Z to E isomer was competitive with that of [4 + 2]
cycloaddition with a diene at a low diene/sulfine ratio,5b
and the Z isomer is reported to be more stable than the
E isomer because of a “syn effect”, i.e., the interaction
between the R-hydrogen in the alkyl group or the o-
hydrogen in the benzene ring and sulfinyl oxygen.9
However, in benzoyl-substituted sulfine, the E isomer is
considered to be more stable than the Z isomer, probably
due to the steric repulsion between the benzoyl and
sulfinyl groups because the “syn effect” is not present in
this case. Furthermore, according to Zwanenburg and co-
workers, the cycloaddition of benzoyl-substituted sulfine
F igu r e 1. Coupling constants between H-5 and H-6 of cis and
trans 5a .
determined by NMR spectrometry of the reaction mixture
directly. Compound 5a has two stereoisomers, cis and
trans forms, which are incidentally separable by recrys-
tallization; cis 5a was obtained as a pure form by
repeated recrystallization from benzene, while trans 5a
was only obtained as a syrup, which contained a slight
amount of cis isomer 5a (approximately less than 5%).
The relative stereochemistry of cis and trans 5a was
1
established by vicinal H-1H coupling constants. H-6 of
trans 5a exhibits a rather small coupling as J ) 7.6 and
6.4 Hz to either of H-5 and H-5′, respectively, implying
that the dihedral angles between H-6 and H-5 and
between H-6 and H-5′ are both around 40-60°; these two
coupling constants mean obviously that the relationship
between the methyne proton at C-6 and the sulfinyl
group is cis (cf. left side of Figure 1). In contrast to trans
5a , a set of large coupling (J ) 11.2 Hz) and rather small
coupling (J ) 4.4 Hz) in the NMR spectrum of cis 5a
revealed the trans relationship between the methyne
proton at C-6 and the sulfinyl group, indicating that the
former large value and the latter small value are due to
the difference in the two dihedral angles (ca. 150° and
60°, respectively, estimated by the molecular model
study) between H-6 and H-5 and H-6 and H-5′, respec-
tively (cf. right side of Figure 1). The cis isomer of 5b
was also purified by repeated recrystallization from
(5) (a) Bonini, B. F.; Mazzanti, G.; Zani, P.; Giorgianni, P.; Barbaro,
G.; Maccagnani, G.; Battaglia, A.; Giorgianni, P. J . Chem. Soc., Chem.
Commun. 1986, 964. (b) Barbaro, G.; Battaglia, A.; Giorgianni, P. J .
Org. Chem. 1991, 50, 2512.
(6) (a) Walsh, A. D. J . Chem. Soc. 1953, 2260. Walsh, A. D. J . Chem.
Soc. 1953, 2266. Walsh, A. D. J . Chem. Soc. 1953, 2288. (b) Walsh, A.
D. J . Chem. Soc. 1953, 2296. (c) Walsh, A. D. J . Chem. Soc. 1953, 2301.
(d) Zwanenburg, B.; Thijs, L.; Strating, J . Tetrahedron Lett. 1967, 3453.
(e) Zwanenburg, B.; Thijs, L.; Strating, J . Tetrahedron Lett. 1968, 2871.
(f) Zwanenburg, B.; Thijs, L.; Strating, J . Recl. Trav. Chim. Pays-Bas
1971, 91, 443.
(7) (a) King, J . F.; Durst, T. J . Am. Chem. Soc. 1963, 85, 2976. (b)
King, J . F.; Durst, T. Can. J . Chem. 1966, 44, 819. (c) Virtanen, A. I.
Angew Chem. 1963, 74, 374.
(8) (a) Zwanenburg, B.; Thijs, L.; Broens, J . B.; Strating, J . Recl.
Trav. Chim. Pays-Bas 1972, 91, 614.
(9) Block, E.; Revelle, L. K.; Bazzi, A. A. Tetrahedron Lett. 1980,
21, 1277.

6732 J . Org. Chem., Vol. 64, No. 18, 1999
Morita et al.
Sch em e 1
Sch em e 2
with butadiene was reported to be fast, even at temper-
atures lower than -78 °C.4c
respectively, as shown in more detail in Scheme 2. If the
elimination reaction of tetrazole from the intermediate
or transition state 1a 1-4a 1 is similar to that of the usual
E2 or E1-like E2 elimination mechanism and, hence, the
carbanion lobes or hydrogen atom on the R-position are
assumed to be fixed in an antiperiplanar position to the
As a result, the stereoselectivity of the cis and trans
cycloadducts is considered to be controlled by the sulfine
formation step, i.e., the elimination step of nitrogen-
containing heteroaromatics and/or isomerization of (Z)-
to (E)-sulfine. In the thermolytic reaction of 1a , the
formation of benzoyl-substituted sulfine was not ob-
served; this means the cycloaddition step is rather fast
in our case, the same as in Zwanenburg’s results.5c
Furthermore, the interconversion between cis and trans
5a at 100 °C in dioxane was also studied independently
and it was revealed that no detectable change in cis to
trans isomer ratio was observed by NMR spectroscopy.
In view of these results and discussions, all the
reactions in Table 1 are considered to afford (E)-ben-
zoylthioformaldehyde S-oxide predominantly or exclu-
sively (cf. entries 1-4). The speculative mechanism is
illustrated in Scheme 1. The abstraction of a carbonyl-
substituted R-methyl hydrogen of the starting sulfoxide
takes place easily to afford the ionic intermediate 1a 4-
4a 4 or enol 1a 5-4a 5, which leads to (Z)- or (E)-sulfine,
tetrazolyl group in the transition state, apparently 1a 1a
-
4a 1a are less stable than 1a 1b-4a 1b in view of the larger
steric repulsion existing between sulfinyl and the benzoyl
group and, reversely, 1a 5-4a 5 may be more stable than
1a 4-4a 4, if the stabilization effect due to hydrogen
bonding between the enol hydrogen and the sulfinyl
group is working. Therefore, the cycloaddition reaction
via (Z)- or (E)-sulfine thus formed initially resulted in
the formation of the cycloadducts cis or trans 5a stereo-
selectively with the diene. Consequently, the results in
the Table 1 seem to indicate the importance of the path
via 1a 1b-4a 1b (or 1a 2b-4a 2b; syn-elimination) as shown
in entries 1-4.
However, in the case of the 4a (heteroaryl ) 3-(4-
methyl)-1,2,4-triazolyl, entry 5 in Table 1), the cis-trans
selectivity of cycloadduct 5 was reversed. This result is,

Trapping of Sulfines with 2,3-Dimethyl-1,3-butadiene
J . Org. Chem., Vol. 64, No. 18, 1999 6733
Ta ble 2. Effect of Exter n a l Ba se for Cis-Tr a n s
Selectivity of 5a on th e Rea ction of 1a -4a in th e
P r esen ce of 10 equ iv of 2,3-Dim eth yl-1,3-bu ta d ien e
presumably, considered to be the difference of the acidity
of the R-methylene protons and the basicity and the
leaving ability of the heteroaromatic ring.
Th e H-D Exch a n ge Stu d y a t Meth ylen e Hyd ogen
of P h en a cyl Su lfoxid e bea r in g Tetr a zole, Tr ia zole,
a n d Th ia d ia zole. Among the nature of the mother
nuclei of heteroaromatics, such as diazole, triazole,
tetrazole, thiazole, thiadiazole, etc., no basicity or acidity
data under the same conditions are available. Therefore,
to obtain further clues for the acidity of the R-methylene
hydrogens, the kinetic experiments of the H-D exchange
reaction of 1a , 2a , 3a , and 4a were carried out in CDCl3/
CD3OD at 37 °C (NMR probe temperature) by monitoring
the decrease of the integration value of methylene
hydrogens by NMR spectroscopy.
The calculated first-order kinetic constants for the
H-D exchange reaction of R-methylene hydrogens of 1a ,
2a , 3a , and 4a under these conditions were 7.77 × 10-3
,
9.14 × 10-3, 9.78 × 10-3, and 4.7 × 10-2 min-1, respec-
tively, and the relative rates were 1.0, 1.2, 1.3, and 6.0,
respectively. Compound 4a has a rate ca. 6 times faster
than that of the other three sulfoxides, 1a , 2a , and 3a ,
which, unexpectedly, showed almost the same reactivi-
ties. These results clearly indicates that the acidity of
the methylene hydrogens of triazolyl phenacyl sulfoxide
is the highest among the tetrazolyl- and thiadiazolyl-
substituted sulfoxides 1a , 2a , and 3a . This increasing
order of H-D exchange rate constants also seems to
imply the tendency of the increasing basicity of the
heteroaromatic moieties, because these parts apparently
catalyzed the H-D exchange reactions either intra- or
intermolecularly, although no such values are reported
in the literature for their mother nuclei, namely 1-phenyl-
1,2,3,4-tetrazole, 1-methyl-1,2,3,4-tetrazole, 2-methyl-
1,3,4-thiadiazole, and 4-methyl-1,2,4-triazole, under the
same conditions. Concerning the leaving ability for these
heteroaromatic moieties, no useful data exists using the
same conditions; however, the 1-phenyl- or 1-methyl-
1,2,3,4-tetrazolyl group is apparently much better than
the 4-methyl-1,2,4-triazolyl or 2-methyl-1,3,4-thiadiazolyl
group, in view of number of electron-withdrawing nitro-
gen atoms in the five-membered ring. Therefore, in the
case of 4a , the rate-determining step is not the R-meth-
ylene hydrogen abstraction step but the elimination step
of the 4-methyl-1,2,4-triazolyl group; this means that the
possible mechanism for 4a will be via an E1cB or an
E1cB-like transition state and, consequently, the pos-
sibility of existing as enol transition state 1a 5a -4a 5a will
be increased, stabilized by hydrogen bonding between the
hydroxy and sulfinyl groups, successively, to form (Z)-
sulfine predominantly. In the case of 1a , and 2a , the
R-methylene hydrogen abstraction step is, probably, the
rate-determining step in view of the good leaving nature
of the phenyl-substituted tetrazolyl group; this means
that the possible mechanism will be via an E2 or an E1-
like transition state.
more preferable transition state for 3a is 1a 1b-4a 1b or
1a 2b-4a 2b, which leads to the (E)-sulfine preferentially
and, successively to trans 5a .
Effect of a n Exter n a l Ba se on th e Cis-Tr a n s
Selectivity of 5a . As shown in Schemes 1 and 2, the
formation of the ionic intermediate 1a 4-4a 4 or enol 1a 5-
4a 5 is expected to be accelerated by the abstraction with
an external base added, if the R-hydrogen abstraction
step is related to the rate-determining step. Therefore,
to obtain further mechanistic details, the reactions of 1a -
4a with 2,3-dimethyl-1,3-butadiene in the presence of
triethylamine or sodium methoxide as an external base
were studied and these results are summarized in Table
2 (the results of Table 1 are included for comparison).
First of all, in all cases of addition of the external bases,
the reaction was apparently accelerated substantially
(lower reaction temperature and shoter reaction time) to
afford 5a and, interestingly, the formation of product 10
was observed in trace to low yields in every case. Product
10 was apparently formed via a different mechanism
which involves thioaldehyde formation by 1,2-elimination
of heteroaryl phenacyl sulfenate generated by the rear-
rangement of the starting sulfoxide as reported in our
precedent study.1c In the case of 3a in the presence of
triethylamine (entry 8) the yield of 10 is extreamely high.
The reason is not certain at present; however, the
relatively lower leaving ability of the 2-methyl-1,3,4-
thiadiazolyl group of 3a seems to be important. The
tendency of trans predominace of 5a in the reaction of
1a -2a in the presence or absence of triethylamine or
sodium methoxide is unchanged, although the selectivity
seems to become lower with the stronger base (entries
1-3 and 4-6). In the case of 3a in the presence of an
external base (entries 7-9), the trans-selectivity became
very low compared with that in the absence of base,
suggesting that the elimination mechanism shifts to an
E1cB-like or E1cB mechanism which will raise the
possibility of cis-selectivitiy. As mentioned previously,
because of the high acidity of the methylene proton of
In view of the rather low reactivity of 3a compared with
that of the other sulfoxides, 1a , 2a , and 4a (compare the
yields of 5a in Table 1), the leaving ability of the
2-methyl-1,3,4-thiadiazolyl group does not appear to be
substantially high. In addition, the acidity of the R-me-
thylene hydrogen is also not as high as that of 4a .
Therefore, in the case of 3a , the elimination reaction
would be shifted to via an ideal E2 mechanism (anti-
elimination, external base catalysis route, or syn-elimina-
tion, internal base catalysis route). Consequently, the

6734 J . Org. Chem., Vol. 64, No. 18, 1999
Morita et al.
Ta ble 4. Th er m olytic Rea ction of
5-(1-P h en yl)-1,2,3,4-tetr a zolyl r-Su bstitu ted P h en a cyl
Su lfoxid e in th e P r esen ce of 10 equ iv of
2,3-Dim eth yl-1,3-bu ta d ien e
Ta ble 3. Th er m olysis of 1-P h en yl-1,2,3,4-tetr a zolyl
Su lfoxid e in th e P r esen ce of 10 equ iv of Bu ta d ien e
4a , which apparently was derived from the higher
basicity of the triazolyl moiety, the reaction of 4a ,
probably, procceds via an E1cB or E1cB-like mechanism
and, hence, the formation of the enol 4a 5a is involved both
in the absence and the presence of triethylamine. In
contrast to this result, when a strong base is used, sodium
methoxide as an external base (entry 12), the mechanism
sibility of the thioketone oxide formation, the thermolyses
of R-methyl or R-substituted-methyl phenacyl sulfoxide
(1d -f) in the presence of 2,3-dimethyl-1,3-butadiene
using bases such as triethylamine or sodium methoxide
were also carried out, and the results are summarized
in Table 4. The reaction of 5-(1-phenyl)-1,2,3,4-tetrazolyl
R-methylphenacyl sulfoxide (1e) in the presence of 2,3-
dimethyl-1,3-butadiene without base afforded 4-benzoyl-
1,2-dimethyl-1-cyclohexene (15) and 6 in 85 and 29%
yields, respectively (entry 1 in Table 4). Product 15 is
considered to be formed by the Diels-Alder reaction of
the diene with 12 which was initially formed by the
â-elimination of the starting sulfoxide 1d together with
the corresponding unstable sulfenic acid (13) (cf. Scheme
3). Therefore, next, in the absence of the diene and a base,
the reaction of 1d in dioxane at 100 °C for 4.8 h was
carried out to afford 5-(1-phenyl)-1,2,3,4-tetrazole (6),
phenyl vinyl ketone (12), and 5-(1-phenyl)-1,2,3,4-tetra-
zolyl benzoylethyl sulfoxide (14) in 60, 12, and 30% yields,
respectively. Product 14 is considered to be formed by
the Michael addition of sulfenic acid 13 to 12. This result,
apparently, revealed that the â-elimination of sulfoxide
took place to afford 12 and the corresponding unstable
sulfenic acid 13 which probably led to complex reaction
mixtures. Further, to confirm this consideration a trap-
ping experiment of the corresponding sulfenic acid with
methyl propiolate was carried out under similar condi-
tions; however, the corresponding trapped product of
sulfenic acid, the methyl 1-tetrazolylsulfinylacrylate
derivative, was not obtained, and products 6 and 12 were
formed in 73 and 45% yields, respectively.
The reaction of 1d with triethylamine at 70 °C afforded
6 in 91% yield (entry 2). However, no Diels-Alder
cycloadduct of sulfine with butadiene was obtained under
these conditions, in contrast with the result of entry 2 in
Table 3. This is rationalized by the lower R-methyne
hydrogen acidty of 1d than that of 1a by introduction of
an electron-releasing methyl group to the sulfoxide 1a .
Therefore, in this case the â-elimination reaction path
to 12 and 13 may be competing with the elimination
reaction path to the sulfine and 6 as shown in Scheme 3.
Consequently, in this case the complex product mixture
was obtained under these conditions. Using a stronger
base, sodium methoxide, than triethylamine, the reaction
shifts to the ideal E1cB, in which case the enol form 4a 5a
,
stabilized by internal hydrogen bonding, cannot exist
because of the absence of protons (a strong basic media),
resulting in predominant formation of the (E)-sulfine
mainly via the transition state 4a 4b rather than 4a 4a
.
Th e Rea ction of â-Keto Su lfoxid es 1b,c Bea r in g
5-(1-P h en yl)-1,2,3,4-tetr a zole w ith 2,3-Dim eth yl-1,3-
bu ta d ien e. The reactions of 1b,c with 2,3-dimethyl-1,3-
butadiene in the presence of triethylamine were further
studied, and the results are summerized in Table 3. In
the case of 5-(1-phenyl)-1,2,3,4-tetrazolyl acetylmethyl
sulfoxide (1b), the reaction in dioxane was accelerated
by the addition of triethylamine to afford 5b and 6 in 90
and 70% yields, respectively (see entries 3 and 4). In both
cases, whether a base is present or not, the cis-trans
isomer ratios of 5b are almost identical. This means that
the mechanism is considered to be the same or quite
similar in both the presence and absence of a base; if
other mechanisms are operating as illustrated in Scheme
2, the cis-trans product ratios will be changed signifi-
cantly. In entry 5, the reaction of 5-(1-phenyl)-1,2,3,4-
tetrazolyl ethoxycarbonylmethyl sulfoxide 1c with 2,3-
dimethyl-1,3-butadiene without a base did not afford the
desired cycloadduct product. However, the addition of
triethylamine was found to accelerate the reaction dra-
matically to afford 6-ethoxycarbonyl-5,6-dihydro-3,4-di-
methyl-2H-thiapyran 1-oxide (5c)11 and 6 in 73 and 68%
yields, respectively (entry 6). In entry 6, the cis-trans
isomer selectivity of 5c was completely lost. The loss of
the cis-trans selectivity in entries 3, 4, and 6, compared
to the results in the case of the sulfoxide 1a , is probably
due to the sterical difference between the benzoyl and
acetyl or ethoxycarbonyl groups.
Rea ction of r-Su bstitu ted P h en a cyl Su lfoxid e
w ith 2,3-Dim eth yl-1,3-bu ta d ien e. To study the pos-
(10) (a) Block, E.; Wall. A. J . Org. Chem. 1987, 52, 809. (b) Kice, J .
L.; Kupczyk-Subotkowska, L. J . Org. Chem. 1991, 56, 1424.
(11) Freer, A. A.; Kirby, G. W.; Lewis, R. A J . Chem. Soc., Chem.
Commun. 1987, 718.

Trapping of Sulfines with 2,3-Dimethyl-1,3-butadiene
J . Org. Chem., Vol. 64, No. 18, 1999 6735
Sch em e 3
125.7-127.0 °C (dec). Spectral data: 1H NMR δ 4.37 (s, 3H),
4.83 (d, J ) 15.2 Hz, 1H), 5.00 (d, J ) 15.2 Hz, 1H), 7.49-7.53
(m, 2H), 7.63-7.67 (m, 1H), 7.93-7.96 (m, 2H); 13C NMR δ
35.3, 62.8, 128.6, 129.1, 134.8, 134.9, 155.9, 190.9; IR (KBr)
1670, 1070 cm-1. Anal. Calcd for C10H10N4O2S: C, 47.99; H,
4.03; N, 22.39. Found: C, 48.04; H, 3.98; N, 22.62.
2-(5-Meth yl)-1,3,4-th ia d ia zolyl p h en a cyl su lfoxid e (3a ):
72% yield; white crystals from CH2Cl2/hexane; mp 118.5-129.7
°C (dec). Spectral data: 1H NMR δ 2.88 (s, 3H), 5.05 (d, J )
16.4 Hz, 1H), 5.41 (d, J ) 16.4 Hz, 1H), 7.50-7.55 (m, 2H),
7.65-7.69 (m, 1H), 7.92-7.95 (m, 2H); 13C NMR δ 16.2, 65.0,
128.7, 129.0, 134.6, 135.4, 169.7, 176.9, 190.5; IR (KBr) 1650,
1030 cm-1. Anal. Calcd for C11H10N2O2S2: C, 49.61; H, 3.78;
N, 10.52. Found: C, 49.74; H, 3.75; N, 10.48.
5-(4-Met h yl)-4H -1,2,4-t r ia zolyl p h en a cyl su lfoxid e
(4a ): 75% yield; white crystals from CH2Cl2/hexane; mp
127.0-134.3 °C (dec). Spectral data: 1H NMR δ 4.03 (s, 3H),
5.10 (d, J ) 16.0 Hz, 1H), 5.55 (d, J ) 1.6, 16.0 Hz, 1H), 7.50-
7.54 (m, 2H), 7.64-7.68 (m,1H), 7.97-8.00 (m, 2H), 8.29 (s,
1H); 13C NMR δ 31.9, 61.6, 128.5, 128.9, 134.5, 135.0, 146.8,
154.0, 191.7; IR (KBr) 1640, 1050 cm-1. Anal. Calcd for
afforded the corresponding product 5d in 41% yield (entry
3). Similarly, the reaction of 2-(5′-(1′-phenyl)-1,2,3,4-
tetrazolylsulfinyl)-1-indanone (1e) using sodium meth-
oxide afforded the corresponding cycloadduct 5e in 33%
yield (entry 4). Zwanenburg and co-workers also have
reported the formation of 5e via the sulfine which was
generated by reaction of the corresponding silyl enol ether
with thionyl chloride.12 In the case of no possibility of
â-elimination, such as with 1f, which has no â-hydrogens,
expectedly, this reaction gave the corresponding cyclo-
adduct 5f in rather high yield, 73% (entry 5).
Exp er im en ta l Section
All melting points are uncorrected. 1H NMR (400 MHz) and
13C NMR (100 MHz) spectra were recorded in CDCl3 using
TMS as an internal standard. The elemental analyses were
performed by the Microanalytical Laboratory of the Depart-
ment of Chemical and Biochemical Engineering of Toyama
University. All the reactions were monitored by TLC using
Silica Gel 60 F254 TLC plates, and products were separated
by column chromatography using Silica Gel 60, also by
preparative layer chromatography using 60 PF254 with UV
detection. All reagents were of highest quality and were
further purified by distillation or recrystallization. Solvents
were further purified by general methods.
Gen er a l P r oced u r e for th e P r ep a r a tion of Su lfoxid e
(1a -4a , 1b-f). To a stirred solution of sulfide in chloroform
or methylene chloride was added m-chloroperbenzoic acid (m-
CPBA, >70%) in chloroform or methylene chloride at 0 °C. This
reaction mixture was stirred at 0 °C until starting sulfide
disappeared by TLC monitoring. The reaction mixture was
washed with saturated aqueous NaHCO3. The aqueous layer
was extracted well with chloroform. This combined organic
layer was washed with brine, dried over anhydrous MgSO4,
filtered, and condensed to afford the crude product. Purifica-
tion was performed by chromatography on silica gel and/or
recrystallization.
C
11H11N3O2S: C, 53.00; H, 4.45; N, 16.86. Found: C, 52.79;
H, 4.49; N, 16.82.
5-(1-P h en yl)-1,2,3,4-tetr a zolyl a cetylm eth yl su lfoxid e
(1b): 66% yield; white crystals from CH2Cl2/hexane; mp 119.8
°C. Spectral data: 1H NMR δ 2.37 (s, 3H), 4.68 (d, J ) 16.4
Hz, 1H), 5.17 (d, J ) 16.4 Hz, 1H), 7.63-7.66 (m, 2H), 7.73-
7.75 (m, 3H); 13C NMR δ 30.6, 64.5, 124.9, 130.1, 131.3, 200.4;
IR (KBr) 1710, 1050 cm-1. Anal. Calcd for C10H10N4O2S: C,
47.99; H, 4.03; N, 22.39. Found: C, 47.97; H, 4.06; N, 22.73.
5-(1-P h en yl)-1,2,3,4-tetr a zolyl eth oxyca r bon ylm eth yl
su lfoxid e (1c): 35% yield; white crystals from CH2Cl2/hexane;
mp 68.2-69.4 °C. Spectral data: 1H NMR δ 1.26 (t, J ) 7.2
Hz, 3H), 4.17-4.23 (m, 2H), 4.50 (d, J ) 15.6 Hz, 1H), 4.92 (d,
J ) 15.6 Hz, 1H), 7.62-7.66 (m, 3H), 7.71-7.74 (m, 2H); 13C
NMR δ 19.9, 56.5, 62.8, 125.0, 130.1, 131.4, 132.8, 156.3, 164.4;
IR (KBr) 1740, 1070 cm-1. Anal. Calcd for C11H12N4O3S: C,
47.14; H, 4.31; N, 19.99. Found: C, 46.83; H, 4.30; N, 20.41.
2-[5′-(1′-P h en yl)-1,2,3,4-t et r a zolylsu lfin yl]p r op iop h e-
n on e (1d ): 78% yield; white crystals from CH2Cl2/hexane; mp
119.8-123.4 °C (dec). Spectral data: 1H NMR (CDCl3) δ 1.96
(d, J ) 7.6 Hz, 3H), 5.97 (t, J ) 7.6 Hz, 1H), 7.51-7.62 (m,
2H), 7.65-7.73 (m, 4H), 7.76-7.80 (m, 2H), 7.82-7.94 (m, 2H);
13C NMR (CDCl3) δ 13.9, 67.1, 124.9, 128.8, 129.2, 130.0, 131.2,
5-(1-P h en yl)-1,2,3,4-tetr a zolyl p h en a cyl su lfoxid e (1a ):
89% yield; white crystals from EtOAc/hexane; mp 105.9 °C.
Spectral data: 1H NMR δ 5.22 (d, J ) 16.4 Hz, 1H), 5.81 (d, J
) 16.4 Hz, 1H), 7.51-7.55 (m, 2H), 7.65-7.70 (m, 4H), 7.80-
7.82 (m, 2H), 7.93-7.95 (m, 2H); 13C NMR δ 62.1, 124.9, 128.6,
129.1, 130.1, 131.3, 132.8, 134.6, 135.0, 157.0, 192.3; IR (KBr)
1670, 1080 cm-1. Anal. Calcd for C15H12N4O2S: C, 57.68; H,
3.87; N, 17.94. Found: C, 57.66; H, 3.89; N, 17.96.
132.8, 133.6, 134.8, 156.9, 198.3. Anal. Calcd for C16H14
-
N4O2S: C, 58.88; H, 4.32; N, 17.17. Found: C, 58.76; H, 4.54;
N, 17.16.
5-(1-Meth yl)-1,2,3,4-tetr a zolyl p h en a cyl su lfoxid e
(2a ): 75% yield; white crystals from CH2Cl2/hexane; mp
2-[5′-(1′-P h en yl)-1,2,3,4-tetr azolylsu lfin yl]in dan on e (1e):
69% yield; white crystals from CH2Cl2/hexane; mp 103.2-131.5
°C (dec). Spectral data: 1H NMR (CDCl3) δ 3.80-3.86 (m, 1H),
4.03-4.09 (m, 1H), 4.85-4.89 (m, 1H), 7.42-7.46 (t, J ) 7.2
Hz, 1H), 7.58-7.73 (m, 7H), 7.79 (d, J ) 7.6 Hz, 1H); 13C NMR
(12) Lenz, B. G.; Regeling, H.; Zwanenburg, B. Tetrahedron Lett.
1984, 25, 5947.

6736 J . Org. Chem., Vol. 64, No. 18, 1999
Morita et al.
(CDCl3) δ 25.6, 67.5, 124.7, 125.0, 126.8, 128.3, 130.1, 131.4,
integration values of residual methylene protons were moni-
tored by NMR spectroscopy. The pseudo-first-order rate con-
stants (k1) for the H-D exchange reaction of R-methylene
hydrogens were correlated nicely (correlation coefficients:
0.998, 0.999, 0.978, and 0.997 for 1a , 2a , 3a , and 4a ,
respectively.) to the following first-order equation and the rate
constants were calculated graphically:
132.9, 135.4, 136.4, 153.8, 197.6. Anal. Calcd for C16H12
-
N4O2S: C, 59.25; H, 3.73 N, 17.27. Found: C, 59.17; H, 3,78;
N, 17.29.
2-[5′-(1′-P h en yl)-1,2,3,4-t et r a zolylsu lfin yl]-2-p h en yl-
a cetop h en on e (1f): 79% yield; white crystals from CH2Cl2/
hexane; mp 150.0-161.1 °C (dec). Spectral data: 1H NMR
(CDCl3) δ 6.91 (s, 1H), 6.97-7.00 (m, 2H), 7.16-7.28 (m, 5H),
7.46 (m, 4H), 7.52-7.61 (m, 2H), 8.00-8.01 (m, 2H); 13C NMR
(CDCl3) δ 79.1, 124.8, 126.2, 129.0, 129.9, 129.32, 129.7, 129.8,
130.2, 131.1, 132.3, 134.4, 134.7, 157.0, 192.0. Anal. Calcd for
C21H16N4O2S: C, 64.93; H, 4.15; N, 14.42. Found: C, 64.72;
H, 4.19; N, 14.32.
k1 ) (log(At/A0 ))/t
t is the time (in min) from the addition of CD3OD and At
and A0 are the integration values of CH2 peaks of sulfoxides
at t ) t and 0 min, respectively. The calculated first-order
kinetic constants for the H-D exchange reaction of R-meth-
ylene hydrogens of 1a , 2a , 3a , and 4a under these conditions
Th er m olysis of 5-(1-P h en yl)-1,2,3,4-tetr a zolyl P h en a -
cyl Su lfoxid e (1a ) in th e P r esen ce of Dien e. To a dioxane
solution (3.0 mL) of 1a (150 mg, 0.48 mmol) was added freshly
distilled 2,3-dimethyl-1,3-butadiene (0.54 mL, 10 equiv). This
mixture in a Pyrex tube was degassed thoroughly in vacuo at
dry ice-acetone temperature, and the glass tube was sealed.
Then the mixture was heated at 100 °C for 3 h. The reaction
mixture was chromatographed on a silica gel preparative plate
using EtOAc/hexane (1:5) to afford 115 mg of 5a (cis:trans )
1:5) and 52 mg of 6 in 99 and 60% yields, respectively. Spectral
data of 5a : trans and cis isomers were separated by recrys-
tallization from benzene. IR (neat) 1680, 1050 cm-1. Trans
isomer: 1H NMR δ 1.74 (s, 3H), 1.78 (s, 3H), 2.68 (s, 2H), 3.47
(d, J ) 15.6 Hz, 1H), 3.59 (d, J ) 15.6 Hz, 1H), 4.83 (dd, J )
6.4, 7.6 Hz, 1H), 7.49-7.53 (m, 2H), 7.61-7.65 (m, 1H), 8.02-
8.04 (m, 2H); 13C NMR δ 19.3, 10.0, 32.1, 53.5, 65.5, 117.6,
127.6, 128.7, 128.8, 134.1, 135.8, 195.5. Cis isomer: mp 128.8-
are 7.77 × 10-3, 9.14 × 10-3, 9.78 × 10-3, and 4.7 × 10-2 min-1
,
respectively, and the relative rates are 1.0, 1.2, 1.3, and 6.0,
respectively.
Gen er a l P r oced u r e of Th er m olysis of Su lfoxid e (1a -
4a ) in th e P r esen ce of Dien e w ith Ba se (Et3N or MeONa ).
To a solution of the starting sulfoxide in 2 mL of dioxane were
added 10 equiv of freshly distilled 2,3-dimethyl-1,3-butadiene
(10 equiv) and 1.5 equiv of base. This mixture was heated at
70 °C. After cooling, this reaction mixture was evaporated and
dried completely in vacuo to give the crude residue, which was
used for the direct determination of the cis-trans isomer ratio
of 5a by NMR spectroscopy. Pure 5a was obtained by a
preparative thin-layer chromatography using EtOAc/hexane
(1:5) as eluent.
Rea ction of 1a in th e p r esen ce of Et 3N: yield of 5a , 67%;
cis-trans ratio of 5a , 1:5.0.
Rea ction of 1a in th e p r esen ce of MeONa : yield of 5a ,
82%; cis-trans ratio of 5a , 1:3.0.
Rea ction of 2a in th e p r esen ce of Et 3N: yield of 5a , 81%;
cis-trans ratio of 5a , 1:1.1.
Rea ction of 2a in th e p r esen ce of MeONa : yield of 5a ,
98%; cis-trans ratio of 5a , 1:2.0.
Rea ction of 3a in th e p r esen ce of Et 3N: yield of 5a , 27%;
cis-trans ratio of 5a , 1:14.
Rea ction of 3a in th e p r esen ce of MeONa : yield of 5a ,
56%; cis-trans ratio of 5a , 1:4.3.
Rea ction of 4a in th e p r esen ce of Et 3N: yield of 5a , 99%;
cis-trans ratio of 5a , 49.5:1.
Rea ction of 4a in th e p r esen ce of MeONa : yield of 5a ,
98%; cis-trans ratio of 5a , 1:2.3.
1
138.5 °C. H NMR δ 1.78 (s, 3H), 1.81 (s, 3H), 2.32-2.37 (m,
1H), 3.10-3.17 (m, 1H), 3.35 (d, J ) 17.6 Hz, 1H), 3.48 (d, J )
17.6 Hz, 1H), 4.44 (dd, J ) 4.4, 11.2 Hz, 1H), 7.48-7.52 (m,
2H), 7.60-7.64 (m, 1H), 7.92-7.94 (m, 2H); 13C NMR δ 19.6,
19.9, 26.0, 52.3, 59.1, 155.3, 127.6, 128.7, 128.8, 133.8, 135.6,
194.4. Anal. Calcd for C14H16O2S: C, 67.71; H, 6.49. Found:
C, 67.57; H, 6.51.
Gen er a l P r oced u r e for Th er m olysis of Su lfoxid es (2a-
4a ). To a dioxane solution of a sulfoxide in a Pyrex tube was
added 10 equiv of freshly distilled 2,3-dimethyl-1,3-butadiene.
This mixture was degassed thoroughly in vacuo at dry ice-
acetone temperature, and the glass tube was sealed and then
was heated at 100 °C for 3 h. After cooling, solvent and
butadiene were removed in vacuo and the NMR spectra of the
residual products were measured directly to determine the
cis-trans product ratio.
Th er m olysis of 5-(1-Meth yl)-1,2,3,4-tetr a zolyl P h en a -
cyl Su lfoxid e (2a ) in th e P r esen ce of Dien e for 3 h . The
above procedure was employed by using 29 mg of 2a , 0.13 mL
of 2,3-dimethyl-1,3-butadiene, and 1.0 mL of dioxane to afford
5a (cis:trans ) 1:3.7) and 7 in quantitative yield by NMR
spectroscopy.
Th er m olysis of 2-(5-Meth yl)-1,3,4-th ia d ia zolyl P h en a -
cyl Su lfoxid e (3a ) in th e P r esen ce of Dien e for 3 h . The
above procedure was employed using 39 mg of 3a , 0.18 mL of
2,3-dimethyl-1,3-butadiene, and 1.2 mL of dioxane to afford
the recovered 3a , 5a (only trans), and 8 in 70, 20, and 30%
yields by NMR spectroscopy.
Th er m olysis of 2-(5-Meth yl)-1,3,4-th ia d ia zolyl P h en a -
cyl Su lfoxid e (3a ) in th e P r esen ce of Dien e for 16 h . The
above procedure was employed using 34 mg of 3a , 0.15 mL of
2,3-dimethyl-1,3-butadiene, and 1.0 mL of dioxane to afford
5a (cis:trans ) 1:30.0) and 8 in quantitative yield by NMR
spectroscopy.
Th er m olysis of 5-(1-P h en yl)-1,2,3,4-tetr a zolyl Acetyl-
m eth yl Su lfoxid e (1b) in th e P r esen ce of Dien e. To a
dioxane solution (3.0 mL) of 1b (100 mg, 0.40 mmol) was added
10 equiv of freshly distilled 2,3-dimethyl-1,3-butadiene (0.45
mL, 10 equiv). This mixture in a Pyrex tube was degassed
thoroughly in vacuo at dry ice-acetone temperature, and the
glass tube was sealed. Then the mixture was heated at 100
°C for 3 h. The reaction mixture was chromatographed on a
silica gel preparative plate using EtOAc/hexane (1:5) to afford
61 mg of 5b (cis:trans ) 1:1.2) and 35 mg of 6 in 82 and 66%
yields, respectively. The cis isomer of 5b was successfully
separated as white crystals by recrystallization of the obtained
cis-trans mixture from benzene. Spectral data of 5b: IR (neat)
1710, 1030 cm-1. Trans isomer: 1H NMR δ 1.72 (s, 3H), 1.73
(s, 3H), 2.38 (s, 3H), 2.57 (d, J ) 7.0 Hz, 1H), 3.50 (d, J ) 16.8
Hz, 1H), 3.28-3.32 (m, 2H), 3.95 (t, J ) 7 Hz, 1H); 13C NMR
δ 19.3, 19.7, 30.6, 31.2. 54.1, 69.3, 117.7, 127.4, 203.6. Cis
1
isomer: mp 120.2-123.2 °C; H NMR δ 1.75 (s, 3H), 1.78 (s,
Th er m olysis of 3-(4-Meth yl)-4H-1,2,4-tr ia zolyl P h en a -
cyl Su lfoxid e (4a ) in th e P r esen ce of Dien e 3 h . The above
procedure was employed using 39 mg of 4a , 0.18 mL of 2,3-
dimethyl-1,3-butadiene, and 1.2 mL of dioxane to afford 5a
(cis:trans ) 2.1:1) and 9 in quantitative yield by NMR
spectroscopy.
Th e Kin etic Stu d y of H-D Exch a n ge of Meth ylen e
P r oton of 1a -4a in CD3OD/CDCl3. 1a -4a (0.05 mmol) was
dissolved in CDCl3 (0.015 mL). To this solution in an NMR
tube was added 0.002 mL of dioxane as an internal standard.
To this mixture was added 0.035 mL of CD3OD, and then, the
3H), 2.28-2.34 (m, 1H), 2.40 (s, 3H), 2.87-2.94 (m, 1H), 3.28-
3.38 (m, 2H), 3.51 (dd, J ) 5, 10 Hz, 1H); 13C NMR δ 19.8,
19.9, 25.5, 29.7, 51.9, 64.1, 115.4, 127.0. 202.4.
Th er m olysis of 5-(1-P h en yl)-1,2,3,4-tetr a zolyl Acetyl-
m et h yl Su lfoxid e (1b) in t h e P r esen ce of Dien e w it h
Ba se. The above procedure was employed using 100 mg of 1b,
0.41 mL of 2,3-dimethyl-1,3-butadiene, 0.085 mL of triethy-
lamine, and 3.0 mL of dioxane to afford 5b (cis:trans ) 1:1)
and 6 in 90 and 70% yields, respectively.
Th er m olysis of 5-(1-P h en yl)-1,2,3,4-tetr a zolyl Eth oxy-
ca r bon ylm eth yl Su lfoxid e (1c) in th e P r esen ce of Dien e

Trapping of Sulfines with 2,3-Dimethyl-1,3-butadiene
J . Org. Chem., Vol. 64, No. 18, 1999 6737
w ith Ba se. In a Pyrex tube, 89 mg of 1c (0.32 mmol), 0.36
mL of freshly distilled 2,3-dimethyl-1,3-butadiene, and 0.066
mL of triethylamine in 2.7 mL of dioxane were degassed
thoroughly in vacuo at dry ice-acetone temperature, and the
glass tube was sealed and then was heated at 100 °C for 3 h.
The reaction mixture was chromatographed on a silica gel
preparative plate using EtOAc/hexane (1:5) to afford 47 mg of
5c (cis:trans ) 1:1) and 34 mg of 6 in 68 and 73% yields,
Th er m olysis of 1d -f w ith 2,3-Dim eth yl-1,3-bu ta d ien e
in th e P r esen ce of Ba se. To a solution of the starting
sulfoxide 1d -f in 3 mL of dioxane were added 1.5 equiv of
2,3-dimethyl-1,3-butadiene and 1.5 equiv of a base, sodium
methoxide or triethylamine, at 70 °C with stirring. After
cooling, the reaction mixture was evaporated and dried in
vacuo. The residual product was separated on a silica gel
preparative plate using AcOEt/hexane (1/1) as eluent to give
the corresponding cycloadducts 5d -f and 6.
respectively. Spectral data of 5c: IR (neat) 1730, 1030 cm-1
.
Trans isomer: 1H NMR δ 1.30 (t, J ) 7.2 Hz, 3H), 1.72-1.79
(m, 6H), 2.56-2.62 (m, 1H), 2.71-2.77 (m, 1H), 3.31-3.32 (bs,
2H), 3.82 (dd, J ) 5.6, 8.0 Hz, 1H), 4.29 (q, J ) 7.2 Hz, 2H).
Cis isomer: 1H NMR δ 1.33 (t, J ) 7.2 Hz, 3H), 1.73-1.79 (m,
6H), 2.40-2.43 (m, 1H), 2.90-2.99 (m, 1H), 3.2 8-3.32 (m,
1H), 3.39 (dd, J ) 4.8, 12.0 Hz, 1H), 3.50 (d, J ) 16.4 Hz, 1H),
4.26 (q, J ) 7.2 Hz, 2H); 13C NMR (cis-trans mixture) δ 14.0,
19.4, 19.5, 19.8, 19.9, 24.9, 29.9, 52.0, 52.4, 57.2, 61.9, 62.0,
62.1, 115.2, 117.2, 126.8, 126.9, 168.1, 168.5.
6-Ben zoyl-5,6-d ih yd r o-3,4,6-t r im et h yl-2H -t h ia p yr a n
1-oxid e (5d ): 1H NMR δ 1.50 (s, 3H), 1.59 (s, 3H), 1.78 (s,
3H), 2.57 (d, J ) 17.6 Hz, 1H), 2.81 (d, J ) 17.6 Hz, 1H), 3,12
(d, J ) 18 Hz, 1H), 3.32 (d, J ) 18 Hz, 1H), 7.45 (t, J ) 8.0
Hz, 2H), 7.55 (t, J ) 7.2 Hz, 1H), 7.76 (d, J ) 7.6 Hz, 2H); 13
C
NMR δ 19.1, 20.1, 20.8, 33.4, 50.2, 67.0, 117.0, 127.2, 127.9,
128.5, 132.8, 136.2, 201.5.
5,6-Dih yd r o-3,4-d im et h yl-2H -t h ia p yr a n -6-sp ir o-2′-in -
1
d a n on e 1-oxid e (5e): H NMR δ 1.69 (s, 3H), 1.78 (s, 3H),
Th er m olysis of 5-(1-P h en yl)-1,2,3,4-tetr a zolyl r-Meth -
ylp h en a cyl Su lfoxid e (1d ) in th e P r esen ce of Dien e. To
a dioxane solution (3.0 mL) of 1d (106 mg) was added 10 equiv
of freshly distilled 2,3-dimethyl-1,3-butadiene (0.45 mL, 10
equiv). This mixture in a 10-mL Pyrex tube was degassed
thoroughly in vacuo at dry ice-acetone temperature, and the
glass tube was sealed. Then, the mixture was heated at 100
°C for 4 h. The reaction mixture was chromatographed on a
silica gel preparative plate using EtOAc/hexane (1/5) as eluent
to give 59 mg of 4-benzoyl-1,2-dimethyl-1-cyclohexene (15) and
14 mg of 1-phenyl-1,2,3,4-tetrazole (6) in 85 and 29% yields,
respectively. Spectral data of 15: colorless oil; 1H NMR δ 1.65
(s, 6H), 1.67-1.74 (m, 1H), 1.92-2.34 (m, 5H), 3.46-3.54 (m,
1H), 7.46 (t, J ) 7.2 Hz, 2H), 7.53-7.57 (m, 1H), 7.96 (d, J )
6.8 Hz, 2H); 13C NMR δ 18.8, 19.0, 26.4, 31.4, 34.2, 42.5, 124.4,
125.3, 128.2, 128.6, 132.8, 136.4, 203.6; IR (neat) cm-1; MS
(m/z) 214.
2.34 (d, J ) 18.4 Hz, 1H), 1.79 (d, J ) 18.4 Hz, 1H), 3.02 (d,
J ) 17.6 Hz, 1H), 3.33 (d, J ) 16.0 Hz, 1H), 3.68 (d, J ) 14.4
Hz, 1H), 3.93 (d, J ) 17.2, 1H), 7.40 (t, J ) 7.6 Hz, 1H), 7.51
(d, J ) 7.6 Hz, 1H), 7.64 (t, J ) 8.0, 1H), 7.78 (d, J ) 7.6 Hz,
1H); 13C NMR δ 19.4, 19.8, 29.2, 31.1, 39.1, 52.2, 118.5, 124.6,
126.5, 126.8, 128.0, 135.4, 135.8, 152.6, 202.4.
6-B e n zo y l-5,6-d ih y d r o -3,4-d im e t h y l-6-p h e n y l-2H -
th ia p yr a n 1-oxid e (5f): mp 141.3-142.9 °C; 1H NMR δ 1.25
(s, 3H), 1.65 (s, 3H), 2.75 (d, J ) 17.2 Hz, 1H), 3.04 (d, J )
17.6 Hz, 1H), 3.52 (d, J ) 16.8 Hz, 1H), 3.57 (d, J ) 18.4 Hz,
1H), 2.29 (t, J ) 8.0 Hz, 2H), 7.38-7.51 (m, 8H); 13C NMR δ
19.1, 20.4, 30.8, 52.2, 74.1, 120.4, 126.3, 127.6, 128.3, 128.6,
129.11, 129.5, 132.6, 133.8, 186.0, 199.6; IR (neat) 1669, 1050
cm-1. Anal. Calcd for C20H20O2S: C, 74.04; H, 6.21. Found:
C, 73.70; H, 6.23.
J O990610L