The chemistry of the thiosulfinyl group: Preparation, structure, and spectroscopic and chemical properties of cyclic thionosulfites

Article abstract of DOI:10.1002/hlca.200590117

Treatment of the tetrahydrothiophene-3,4-diol 5 with 1,1′-thiobis- (1H-benzimidazole) (6) furnished two diastereoisomers of the novel cyclic thionosulfite 4 with different configurations at the pseudo-tetrahedral center of the thiosulfinyl (S=S) group. The configuration of the S=S group of the major diastereoisomer (isolated in 45% yield) was established to be syn to the thiolane ring, as determined by X-ray crystallographic analysis, while that of the minor diastereoisomer (isolated in 10% yield) was anti. ¹H-NMR Spectroscopic analysis clarified that the shielding and deshielding zones of the S=S group are similar to those of the well-documented S=O group. Characteristic absorptions of the S=S group in the IR, Roman, and UV/VIS spectra were assigned on the basis of calculations at the B3LYP/6-31G* level. The reactivity of the S=S group toward thermolysis, hydrolysis, and oxidation was examined. The S=S group is more resistant toward oxidation than the divalent sulfide S-atom, but is oxidatively converted to the S=O group.

Full text of DOI:10.1002/hlca.200590117

Helvetica Chimica Acta ± Vol. 88 (2005)
1451
The Chemistry of the Thiosulfinyl Group: Preparation, Structure, and
Spectroscopic and Chemical Properties of Cyclic Thionosulfites
by Juzo Nakayama*, Sanae Yoshida (nee Tanaka), Yoshiaki Sugihara, and Akira Sakamoto
Department of Chemistry, Faculty of Science, Saitama University, Sakura-ku, Saitama, Saitama 338-8570, Japan
(phone and fax: ‡ 8148-858-3390; e-mail: nakaj@post.saitama-u.ac.jp)
Dedicated to Professor Rolf Huisgen on the occasion of his 85th birthday
Treatment of the tetrahydrothiophene-3,4-diol 5 with 1,1'-thiobis-(1H-benzimidazole) (6) furnished two
diastereoisomers of the novel cyclic thionosulfite 4 with different configurations at the pseudo-tetrahedral
center of the thiosulfinyl (SˆS) group. The configuration of the SˆS group of the major diastereoisomer
(isolated in 45% yield) was established to be syn to the thiolane ring, as determined by X-ray crystallographic
analysis, while that of the minor diastereoisomer (isolated in 10% yield) was anti. 1H-NMR Spectroscopic
analysis clarified that the shielding and deshielding zones of the SˆS group are similar to those of the well-
documented SˆO group. Characteristic absorptions of the SˆS group in the IR, Raman, and UV/VIS spectra
were assigned on the basis of calculations at the B3LYP/6-31G* level. The reactivity of the SˆS group toward
thermolysis, hydrolysis, and oxidation was examined. The SˆS group is more resistant toward oxidation than the
divalent sulfide S-atom, but is oxidatively converted to the SˆO group.
1. Introduction. ± Much effort has been devoted to developing the chemistry of the
thiosulfinyl group, a highly reactive functional group, but much still remains to be
explored1). F2SˆS (1) is a long-known compound. Its SˆS bond length was
determined to be 1.860Š by microwave spectroscopy in 1963 [2]. The first synthesis
of thionosulfites (RO)2SˆS was reported in 1965 [3a]. In this report, the isomeric
1
dialkoxy disulfide structure ROSSOR was ruled out on the basis of H-NMR analysis
of cyclic derivatives. Later, a more-stable thionosulfite, 2, was synthesized, and its
molecular structure was determined by X-ray single-crystal structure analysis by Harpp
et al. [4a,c]. The SˆS bond length (1.901 Š) of 2 is clearly shorter than those of
common disulfides (2.03 ± 2.07 Š) [5], and is of the same order as that of 1 (1.860Š),
diatomic sulfur 134S2 (1.890Š) [6], and N-(thiosulfinyl)amines (ÀNˆSˆS) (1.898,
1.918 Š) [7]. On the other hand, although thiosulfoxides (R ˆ alkyl or aryl for R2SˆS)
have been proposed occasionally as transient intermediates, they still elude detection
even by spectroscopic methods [1]. This suggests that the thiosulfinyl group is more
stabilized by heteroatom substituents such as Fand RO than by alkyl or aryl groups [3].
1
)
For a leading review, see [1].
ꢀ 2005 Verlag Helvetica Chimica Acta AG, Zürich

1452
Helvetica Chimica Acta ± Vol. 88 (2005)
The S-atom in sulfites acts as a stereogenic center because of the stable pseudo-
tetrahedral geometry. Thus, enantiomerically pure sulfites have been prepared and
applied to asymmetric synthesis [8]. In this connection, the structure and reactions of
thionosulfites are also of much interest. Recently, the inversion energy at the sulfur
center of 3 was predicted to be 32.3 kcal/mol by calculations at the MP2/6311G(3d)
level [4b]. However, reactions of the thiosulfinyl group of thionosulfites have been
scarcely studied.
Herein, we report the preparation of the five-membered cyclic thionosulfite 4 by
treatment of cis-3,4-di-(tert-butyl)tetrahydrothiophene-3,4-diol (5) [9] with 1,1'-thio-
bis(1H-benzimidazole) (6) [10], and the isolation of the two diastereoisomers 4a and 4b
(see Scheme 1 below)2). The structures of 4a and 4b are discussed on the basis of X-ray
crystallographic analyses, spectral data, and density functional theory (DFT) calcu-
lations. Also reported are the diamagnetic anisotropy and chemical properties of the
SˆS group.
2. Results and Discussion. ± 2.1. Preparation of Thionosulfites 4a and 4b. Initially,
we thought that 4 could be derived from the corresponding sulfite 7 by replacement of
its O-atom by an S-atom. Thus, 7 was prepared by condensation of 5 [9] with SOCl2 as a
mixture of the diastereoisomers 7a and 7b [12][13]. Disappointingly, however, the
attempted conversion of 7 to 4 by treatment with Lawessonꢁs reagent was unsuccessful
(Scheme 1).
Previously, Harpp et al. showed that thionosulfites are produced from 1,2-diols by
treatment with 6 [10]. We applied this method to 5 to obtain 4 directly [4a] (Scheme 1).
Thus, the reaction of 5 with 6 was studied by varying the solvent, the amount of 6, the
temperature, and the reaction time. As a result, it was found that the yield of 4 was
optimal when 5 was treated with 4 mol-equiv. of 6 in MeCN for 72 h at room
temperature. Under these conditions, the reaction led to a mixture of the diaster-
eoisomers 4a and 4b in a ratio of 82 :18. Separation by HPLC afforded 4a and 4b in
45% and 10% yield, respectively. Similar results were obtained when the reaction was
conducted at 408 for 48 h. When 4 mol-equiv. of 1,2,4-triazole were used as an additive,
the yield of 4 slightly increased to afford 4a and 4b in 50% and 11% yield, respectively.
However, we cannot explain how the additive improves the yield. Pretreatment of 5
with a base such as t-BuOK or BuLi in THF and subsequent treatment with 6 gave rise
to byproducts, with formation of 4 in decreased yield.
The use of 1,1'-thiobis(1H-1,2,4-triazole) (8) [10] instead of 6 gave 4 in low yield
(14% for 4a, and 6% for 4b). Although the use of 1,1'-dithiobis(1H-benzimidazole) (9)
[10] gave a small amount of 4, its formation might be due to the presence of 6 as a
contaminant. The use of sulfur dichloride (SCl2) or sulfur monochloride (S2Cl2) did not
afford 4 even in trace amounts.
A probable mechanism for the formation of 4 is shown in Scheme 2. The initial
reaction would involve a sulfur transfer from 6 to 5 to produce the dioxathiolane 10,
with elimination of two molecules of benzimidazole. The S-atom of 10, which is
activated by repulsive lone-pairÀlone-pair interactions with the adjacent O-atoms,
2
)
Part of this work was preliminarily reported [11].

Helvetica Chimica Acta ± Vol. 88 (2005)
1453
Scheme 1
Scheme 2

1454
Helvetica Chimica Acta ± Vol. 88 (2005)
would further react with 6 to give the ionic intermediate 11. Finally, the elimination of
1,1'-bi-1H-benzimidazole (12) [14] from 11 leads to the formation of 4a and 4b.
2.2. Molecular Structures of 4a and 4b. The configuration of 4a and 4b was
established by X-ray crystallographic analysis (Fig. 1; for crystal data, see Table 3 in the
Exper. Part). Table 1 shows the relevant bond angles, bond lengths, and dihedral angles
of 4a and 4b, and also those of 4a and 4b obtained by DFT calculations (B3LYP/6-
31G* level)3). The optimized molecular structures of 4a and 4b were in good
agreement with the experimental structures, as is evident from Figs. 1 and 2 (see also
Table 1) [15].
Fig. 1. ORTEP Representations of the molecular structures of 4a and 4b
Fig. 2. Optimized structures of 4a and 4b calculated at the B3LYP/6-31G* level
3
)
The calculations were performed with the Gaussian 98 (revision A.7) program on personal computers
running RedHat Linux 6.0. All of the geometry optimization, vibrational-frequency calculations, and time-
dependent DFT calculations were carried out at the B3LYP/6-31G* level. The calculated frequencies are
scaled by 0.9613.

Helvetica Chimica Acta ± Vol. 88 (2005)
1455
Table 1. Observed and Calculated (B3LYP/6-31G*) Bond Lengths [Š], Bond Angles [8], and Dihedral Angles
[8] of Compounds 4
Observed
Calculated
4a
4b
4a
4b
S(1)ÀS(2)
S(2)ÀO(1)
S(2)ÀO(2)
S(3)ÀC(3)
S(3)ÀC(4)
O(1)ÀC(1)
O(2)ÀC(2)
C(1)ÀC(2)
C(1)ÀC(4)
C(1)ÀC(5)
C(1)ÀC(3)
C(2)ÀC(6)
C(5)ÀC(7)
C(5)ÀC(8)
C(5)ÀC(9)
C(1)ÀC(10)
C(6)ÀC(11)
C(6)ÀC(12)
1.9154(6)
1.6399(12)
1.6432(12)
1.802(2)
1.800(2)
1.464(2)
1.484(2)
1.598(2)
1.557(2)
1.564(2)
1.527(2)
1.572(2)
1.546(2)
1.535(3)
1.527(2)
1.541(3)
1.534(3)
1.534(3)
1.8964(13)
1.639(2)
1.644(2)
1.750(2)
1.787(4)
1.475(3)
1.457(3)
1.607(4)
1.527(4)
1.574(4)
1.556(5)
1.575(5)
1.545(5)
1.540(5)
1.530(6
1.936
1.706
1.704
1.836
1.827
1.454
1.472
1.617
1.566
1.587
1.541
1.595
1.557
1.546
1.542
1.555
1.547
1.547
1.929
1.714
1.708
1.825
1.833
1.470
1.446
1.625
1.541
1.595
1.573
1.588
1.556
1.548
1.543
1.538(6)
1.530(6)
1.526(6)
1.558
1.545
1.543
S(1)ÀS(2)ÀO(1)
S(1)ÀS(2)ÀO(2)
O(1)ÀS(2)ÀO(2)
C(3)ÀS(3)ÀC(4)
S(2)ÀO(1)ÀC(1)
S(2)ÀO(2)ÀC(2)
O(1)ÀC(1)ÀC(2)
O(1)ÀC(1)ÀC(4)
C(2)ÀC(1)ÀC(4)
O(2)ÀC(2)ÀC(1)
O(2)ÀC(2)ÀC(3)
C(1)ÀC(2)ÀC(3)
S(3)ÀC(3)ÀC(2)
S(3)ÀC(4)ÀC(1)
111.88(5)
106.28(5)
93.43(6)
111.69(10)
105.92(9)
92.74(11)
95.3(2)
113.1
107.0
91.3
113.9
107.2
90.9
94.25(8)
92.9
93.5
112.83(10)
113.60(9)
103.13(12)
107.82(12)
103.73(12)
98.59(11)
104.09(13)
101.82(13)
109.82(11)
109.84(11)
116.1(2)
109.4(2)
100.6(2)
105.2(2)
107.7(2)
102.8(2)
108.4(3)
105.0(2)
112.3(3)
109.1(2)
112.0
114.0
104.1
108.3
103.3
99.2
103.8
108.8
110.4
109.0
116.1
109.7
100.3
104.6
108.5
103.8
108.3
104.2
111.1
110.6
S(1)ÀS(2)ÀO(1)ÀC(1)
S(1)ÀS(2)ÀO(2)ÀC(2)
O(1)ÀS(2)ÀO(2)ÀC(2)
O(2)ÀS(2)ÀO(1)ÀC(1)
O(1)ÀC(1)ÀC(2)ÀO(2)
C(3)ÀC(2)ÀC(1)ÀC(4)
C(5)ÀC(1)ÀC(2)ÀC(6)
98.37(10)
132.94(10)
18.99(10)
10.65(10)
41.86(11)
37.66(14)
43.7(2)
115.8(2)
145.8(2)
32.3(2)
7.4(2)
36.4(2)
33.0(3)
38.0(3)
98.5
133.8
19.1
10.7
42.6
37.6
43.1
111.6
144.8
29.5
2.3
40.0
36.1
41.2
For comparison, the bond angles and bond lengths around the SˆS group of 4a, and
those around the SˆO group of 7a are shown in Fig. 3 [13]. The sum of the two
OÀSÀS bond angles and the OÀSÀO bond angle of 4a is 311.58, which is comparable
with that of 7a (309.58), while it is smaller than the sum of three HÀCÀH bond angles
of methane (328.58). Incidentally, the sum of the corresponding angles is 310.48 for 4b.
The SˆS bond lengths of 4a and 4b are 1.9154(6) and 1.8964(13) Š, respectively, and

1456
Helvetica Chimica Acta ± Vol. 88 (2005)
Fig. 3. Bond angles and bond lengths around the SˆS group of 4a and the SˆO group of 7a
are comparable with that of 2 (1.901(2) Š). The two t-Bu groups of 4a and 4b are
twisted, with dihedral angles of 43.7(2) and 38.0(3)8, respectively, to reduce steric
repulsion.
2.3. Spectroscopy of 4a and 4b. As to the diamagnetic anisotropy of the SˆS group,
1
no information has been available to date. Fig. 4 shows the H-NMR chemical-shift
data of the CH2 groups of 4a and 4b, and those of 7a and 7b. Each H-atom of these
compounds appeared as a doublet, with J ˆ 13 to 14 Hz due to geminal coupling [13].
The 1H-NMR spectrum of 4a shows two doublets (J ˆ 13.4 Hz) at d(H) 3.41 and 3.90,
and that of 4b shows two doublets (J ˆ 14.1 Hz) at 3.13 and 3.42. For 4a, the doublet at
d(H) 3.41 is assigned to Hb due to the observation of an 8% NOE on irradiation of the
t-Bu H-atoms, while, for 4b, the doublet at d(H) 3.42 is assigned to Hb (13% NOE; Hb of
4a and 4b are cis to the bulky t-Bu groups). Thus, Ha of 4a appears at a lower field than
does Hb, while Ha of 4b appears at higher field than Hb, similar to 7a and 7b.
The analysis of these data leads to the conclusion that the shielding and deshielding
zones of the SˆS group are similar to those of the SˆO group. The shielding and
deshielding zones of the SˆS group are, therefore, assigned as depicted in Fig. 5 by
analogy of the corresponding well-documented zones of the SˆO group [16].
Fig. 4. 1H-NMR Chemical shifts of 4a,b and 7a,b

Helvetica Chimica Acta ± Vol. 88 (2005)
1457
Fig. 5. Magnetic-Shielding (‡) and -deshielding (À) zones of the SˆS
group
The SˆS bond-stretching vibration would be the greatest concern in the IR and
Raman spectra of 4a and 4b. The observed and DFT-calculated3) IR and Raman
spectra of 4a are shown in Figs. 6 and 7, respectively, and those of 4b are given in Figs. 8
and 9, respectively [15]. Both types of spectra of 4a showed a strong absorption band at
653 and 650cm À1, respectively. The calculations on 4a predicted that the strong IR
absorption band and the medium-sized Raman band appear at 639 cmÀ1, with a scaling
factor of 0.9613 mainly due to the contribution of the SˆS stretching vibration.
Therefore, the above-mentioned strong IR and Raman bands at 653 and 650cm À1
,
respectively, can be assigned to the SˆS stretching vibration. Similarly, the strong
Raman band assignable to the SˆS stretching vibration was observed at 666 cmÀ1 for
4b. The corresponding IR absorption band was observed at 665 cmÀ1 as a shoulder of
the strong 670-cmÀ1 band. The DFT calculations of 4b predicted the medium-sized IR
Fig. 6. Observed (top) vs. calculated (bottom) IR spectra of 4a

1458
Helvetica Chimica Acta ± Vol. 88 (2005)
Fig. 7. Observed (top) vs. calculated (bottom) Raman spectra of 4a
absorption band and the strong Raman band at 647 cmÀ1, which are assignable to the
SˆS stretching vibration. For 2, the SˆS stretching IR and Raman bands are observed
at 650and 652 cm À1, respectively [4b].
The UV/VIS spectrum of 4a is shown in Fig. 10. Thus, 4a showed the two absorption
maxima at 253 (e ˆ 2790) and 324 (142) nm, whereas time-dependent DFT calculations
predicted the appearance of two strong absorptions at 249 and 263 nm, and a weak
absorption at 361 nm that originate from the SˆS group. This might be rationalized by
the overlap of the two expected absorptions at 249 and 263 nm, which became a single
absorption at 253 nm. Similarly, 4b showed the two absorption maxima at 245 (e ˆ
3060) and 313 (203) nm, although the calculations predicted the appearance of the two
strong absorptions at 241 and 262 nm, and the weak absorption at 351 nm. Also, for 2,
only two absorptions, a strong one at 250nm, and a weak one at 311 nm, were observed
[4b].
2.4. Reactions of 4a and 4b: Thermolysis, Hydrolysis, and Oxidation. No thermal
isomerization was observed between 4a and 4b, even when each isomer was heated at
1208 in [D8] toluene, though some decomposition took place (Scheme 3). Thus, when
4a was heated at 1208 for 96 h, 13, 14 [9], and 4a were present in a ratio of 39 :13 :48.
On the other hand, 4b decomposed completely to produce 13 and 14 in a ratio of 6 :94,
when heated to 1208 for 24 h. The progress of the decomposition of 4b was monitored

Helvetica Chimica Acta ± Vol. 88 (2005)
1459
Fig. 8. Observed (top) vs. calculated (bottom) IR spectra of 4b
Scheme 3
by 1H-NMR analysis. Thus, we concluded that the pseudo-tetrahedral geometry around
the thiosulfinyl group is rigid enough to induce a stereogenic center [4b]. Hence, the
observed product ratio 4a/4b of 82 :18 is kinetically controlled.
Compounds 13 and 14 would be produced through two independent competitive
pathways (Scheme 4). For 4a, the terminal S-atom and one of the CH2 H-atoms can
come close together, thus enabling an intramolecular H-atom abstraction, which gives
rise to the thiosulfurous acid ester 15. Then, another intramolecular H-shift takes place
in 15 to form 13 as the major product, with elimination of HOS(O)SH (H2S ‡ SO2).
Meanwhile, thermal extrusion of diatomic sulfur (S2) of 4a would result in the
formation of 14, with simultaneous CÀC bond cleavage. The formation of 14 through

1460
Helvetica Chimica Acta ± Vol. 88 (2005)
Fig. 9. Observed (top) vs. calculated (bottom) Raman spectra of 4b
Fig. 10. UV/VIS Spectrum of 4a
the dioxetane intermediate 16, which is formed by loss of S2, would be disfavorable. On
the other hand, for 4b, the foregoing intramolecular H-shift is difficult to occur, because
the terminal S-atom is remote from the CH2 H-atoms. Therefore, the decomposition of
4b would take place predominantly with liberation of S2 to give 14 as the major product.
The minor product 13 from 4b might be rationalized as intermolecular H-atom

Helvetica Chimica Acta ± Vol. 88 (2005)
1461
abstraction, by which two molecules of 4b produce two molecules of 15. The formation
of 13, through isomerization of 4b to 4a by pyramidal inversion followed by
intramolecular H-atom abstraction, would be least probable, because 4a is thermally
more stable than 4b. The DFT calculations also predicted that 4a is thermodynamically
more stable than 4b by 1.69 kcal/mol.
Scheme 4
Thionosulfite 4a is susceptible to alkaline hydrolysis. Thus, 4a was hydrolyzed to the
diol 5 in 93% yield in the presence of NaHCO3 in a 1:1 mixture of H2O and THF,
whereas it remained unchanged in the absence of NaHCO3 for several days.
Next, the reactivity of the thiosulfinyl group toward oxidants was examined.
Oxidation of 4a with an equimolar amount of dimethyldioxirane (DMD) gave 17a in
91% yield, and oxidation of 4b gave 17b in 94% yield (Scheme 5). The structure of 17a
was determined by X-ray crystallographic analysis (Fig. 11), while the structure of 17b
was determined on the basis of its spectroscopic properties. The crystal data of 17a are
summarized in the Exper. Part.
Oxidation of 4a with 1.1 mol-equiv. of ꢂm-chloroperbenzoic acidꢁ (MCPBA) gave
17a and 18b in a ratio of 94 :6; 17a was isolated in 77% yield. The use of 3.3 mol-equiv.
of MCPBA gave 18b and 19b in a ratio of 90:10. The structures of 18b and 19b were
determined by comparison of their spectroscopic data with those of authentic samples
prepared by oxidation of 7b [13]. These results show that the thiosulfinyl group is
more-resistant toward oxidation than the S-atom in the CH2ÀSÀCH2 moiety of the
thiolane.
The above results also show that the oxidation of both 4a and 4b takes place
exclusively anti (exo side) to the sulfite ring, i.e., syn to the bulky t-Bu groups, and that
the resulting configuration of each is the same as that observed previously for the

1462
Helvetica Chimica Acta ± Vol. 88 (2005)
Scheme 5
Fig. 11. ORTEP Representation of the molecular structure of 17a. Relevant bond lengths [Š], bond angles [8],
and dihedral angles [8]: S(1)ÀS(2), 1.8928(11); S(2)ÀO(1), 1.644(2); S(2)ÀO(2), 1.656(2); S(3)ÀO(3),
1.489(2); S(1)ÀS(2)ÀO(1), 111.08(7); S(1)ÀS(2)ÀO(2), 107.48(7); O(1)ÀS(2)ÀO(2), 97.73(8);
O(1)ÀS(1)ÀS(2)ÀO(2), 43.47(14); C(3)ÀC(2)ÀC(1)ÀC(4), 37.3(2); C(5)ÀC(1)ÀC(2)ÀC(6), 43.1(2).
oxidation of 7a and 7b [13]. This would be best rationalized by steric hindrance:
analysis of the molecular models based on X-ray crystallographic analyses of 4a and 4b
suggests that the exo side is less hindered, in spite of the bulky t-Bu groups. The
difference in the electron densities of the S-atom between anti- and syn-sides to the

Helvetica Chimica Acta ± Vol. 88 (2005)
1463
sulfite ring would not be in accord with the results, because the DFT calculations
predicted that there is no distinct difference for either 4a and 4b in the HOMO electron
densities of the S-atoms between these two sides.
The above oxidation also converted the thiosulfinyl group to the sulfinyl group. We
then investigated the oxidation of 17a and 17b with MCPBA, DMD, H2O2, and t-
BuOCl to get more insight into the mechanism of this conversion (Scheme 6) [17]. The
oxidations gave mixtures of 18a and 18b, with the ratios being dependent on the
substrates and oxidants, as summarized in Table 2. Further oxidations of 18a and 18b to
19a and 19b, respectively, also took place (Scheme 6 and Entries 1 ± 3 and 4 in Table 2).
The structures of 18a and 19a were determined by comparison of the spectroscopic data
Scheme 6
Table 2. Oxidations of 17a and 17b
Entry Substrate Oxidizing agent Solvent
Product ratioa)
17a/17b/18a/18b/19a/19b (18a ‡ 19a)/(18b ‡ 19b)
1
2
3
4
5
6
7
8
17a
17b
17a
17b
17a
17b
17a
17b
MCPBAb)
MCPBAb)
DMDb)
CH2Cl2
CH2Cl2
8 : 0 : 14 : 66 : 0 : 12
0 : 8 : 34 : 37: 9 : 12
CH2Cl2/Me2CO 13 : 0: 58 : 21 : 5 : 3
15 : 85
47: 53
72 : 28
67: 33
7: 93
DMDb)
CH2Cl2/Me2CO
Me2CO
Me2CO
CHCl3/THF
CHCl3/THF
0: 15 : 50: 23 : 7: 5
46 : 0 : 4 : 50 : 0 : 0
0: 59 : 16 : 25 : 0: 0
8 : 0: 74 : 18 : 0: 0
10: 3 : 71 : 16 : 0: 0
c
H2O2
H2O2
)
)
c
39 : 61
80: 20
82 : 18
tBuOCld)
tBuOCld)
a) Determined by 1H-NMR analysis. b) With 2.2 mol-equiv. at 08 for 3 h. c) With excess H2O2 (30% soln.) at r.t.
for 15 h. d) With 2.2 mol-equiv. at À 508 for 15 min, then without cooling for 6 h.

1464
Helvetica Chimica Acta ± Vol. 88 (2005)
with those of authentic samples prepared by oxidation of 7a [13]. Note that 2 mol-
equiv. of the oxidants are required for the conversion of 17 to 18. This conversion shows
a poor stereoselectivity (see the ratio (18a ‡ 19a)/(18b ‡ 19b) in Table 2). Thus, for
Entries 3, 6, and 7, the configuration at the S-atom is mostly retained, for Entries 1, 4, 5,
and 8, the configuration at the S-atom is mostly inverted, and for Entry 2, no selectivity
was observed. In addition, a separate experiment showed that, when the oxidation of
17a with MCPBA was performed in the presence of 3,4-dimethylbuta-1,3-diene,
compound 20, which is the adduct of the diene with sulfur monoxide (SO), was
obtained in 27% yield (Scheme 7) [18].
Scheme 7
For the oxidation with MCPBA and DMD, three mechanisms should be taken into
consideration, as illustrated in Scheme 8 for 17a as the substrate. In the case of Path A,
initial oxidation takes place at the tetravalent S-atom from the exo face to produce a
Scheme 8

Helvetica Chimica Acta ± Vol. 88 (2005)
1465
thionosulfate (21a). According to Path B, it takes place at the terminal S-atom to
produce an intermediate (22a) with a new functional group (SˆSˆO). Route C is
based on a three-membered-ring intermediate (23a), which corresponds to the epoxide
formation from alkenes (see also Scheme 9 below).
Scheme 9
Mechanism A seems to be less probable because no sign of the formation of 21a was
observed, although 21a might be more stable than 17a and isolable (see the results of
DFT calculations discussed later). In addition, if 18b were formed through loss of an S-
atom from 21a, the reaction should take place with inversion at the S-center, and an
equimolar amount of the oxidant should be enough to complete the oxidation to 18b.
However, no clean inversion of configuration was observed, and 2 mol-equiv. of the
oxidants were required to complete the conversion. Furthermore, Mulliken population
analysis predicted that the tetravalent S-atom of 17a,b is considerably electron-
deficient and, thus, hard to oxidize (Fig. 12).
Fig. 12. Mulliken population analyses of 17a,b and 18a,b at the B3LYP/6-31G* level
For mechanism B, if 22a once is formed, it might extrude SO to give a sulfoxylate
(24). Although the mechanism is compatible with the formation of SO, which was

1466
Helvetica Chimica Acta ± Vol. 88 (2005)
trapped by 3,4-dimethylbuta-1,3-diene, it does not rationalize the stereochemical
outcome. If 24 exists as the intermediate, both oxidation of 17a with MCPBA and that
of 17b should produce 18a ‡ 19a and 18b ‡ 19b in the same ratio. The same is true for
the oxidation with DMD.
We will then consider mechanism C (Scheme 9). Mulliken population analysis3)
predicted that the SˆS bond of 17a,b is less polarized than the SˆO bond of 18a,b
(Fig. 12) [15]. Therefore, the three-membered-ring formation may take place to give
23. Thus, oxidation of 17a and 17b gives the spiro compounds 23a and 23b, respectively.
DFT Calculations predicted that these compounds are placed in the local energy
minima, although they are thermodynamically less stable than 21a and 22a (Fig. 13)
[15].
Fig. 13. Predicted structures of the presumed intermediates 21a ± 23a
Compounds 23a and 23b might be interconvertible through pseudo-rotation4) [19].
The next oxidation takes place at the divalent S-atom to give 25a from 23a, and 25b
from 23b, before the equilibrium is completed by pseudo-rotation. The interconversion
between 25a and 25b by pseudo-rotation might also take place. Extrusion of SO from
25a and 25b gives 18b and 18a, respectively. Therefore, the ratio of 18a to 18b, i.e., the
ratio of (18a ‡ 19a) to (18b ‡ 19b), depends on the rate of oxidation of 23a and 23b,
when there is no interconversion of 25a and 25b, i.e., no pseudo-rotation. As a result,
the ratio (18a ‡ 19a)/(18b ‡ 19b) becomes dependent on the substrates and oxidants.
Thus, mechanism C, which initiates by formation of 23, nicely accommodates the
experimental observations and, thus, seems to be most likely, although we cannot rule
out another mechanism, because the interconversion between 23a and 23b, whose
trigonal-bipyramidal (TBP) structure is highly distorted, might have a high energy
barrier5).
4
)
We thank Dr. S. Sato of Tsukuba University for discussions on the pseudo-rotation of the presumed
intermediates.
A pseudo-rotation that takes place through the biradical intermediates formed by a hemolytic cleavage of
the SÀO bond of the three-membered ring would be least likely, since the hemolytic cleavage of, e.g., 23a
should result in the formation of 21a.
5
)

Helvetica Chimica Acta ± Vol. 88 (2005)
1467
For the t-BuOCl oxidation, compounds 26a and 26b would form from 17a and 17b,
respectively, through the formation of three-membered-ring intermediates (intermedi-
‡
ates correspond to halonium ions for alkenes) by addition of Cl to the SˆS bond,
followed by addition of H2O (Scheme 10). Compounds 26a and 26b interconvert
quickly by pseudo-rotation to give an equilibrium mixture. Hydrolysis of 26a and 26b
will produce 18b and 18a, respectively. This rationalizes why the ratio of (18a ‡ 19a)/
(18b ‡ 19b) is nearly equal for the oxidation of 17a and 17b.
Scheme 10
This work was supported by Grants-in-Aid for Scientific Research (Nos. 13029016 and 16350019) from the
Ministry of Education, Science, Culture and Technology, Japan. We thank Prof. A. Ishii of Saitama University for
fruitful discussions.
Experimental Part
1. General. Solvents were purified and dried in the usual manner. Column chromatography (CC) was
performed on Silica Gel 7734 (70± 230mesh; Merck) or Silica Gel 60 N (63 ± 210mesh; Kanto). Melting points
(m.p.) were determined on a Mel-Temp capillary-tube apparatus and are uncorrected. UV Spectra were
determined on a JASCO V-560 spectrometer; lmax (e) in nm. IR Spectra were recorded on a Perkin-Elmer
System-2000 FT-IR spectrometer; in cmÀ1. Raman spectra were taken on a Perkin-Elmer System-2000 FT-
Raman spectrophotometer; in cmÀ1. 1H- and 13C-NMR Spectra were recorded on Bruker ARX-400, Bruker AM-
400, or Bruker AC-300P spectrometers in CDCl3 soln., with Me4Si as internal standard; d in ppm, J in Hz.
Electron-impact mass spectra (EI-MS) were recorded on a JEOL JMS-DX303 spectrometer operating at 70eV;
in m/z. Elemental analyses were performed by the Material and Life Science Research Center of Saitama
University.
2. Preparation of (3aR,6aS)-3a,6a-Di(tert-butyl)tetrahydrothieno[3,4-d][1,3,2]dioxathiole 2-Sulfide (4).
2.1. Reaction with 1,1'-Thiobis(1H-benzimidazole) (6) as Sulfur Transfer Agent. 2.1.1. Reaction at Room
Temperature. A mixture of cis-3,4-di(tert-butyl)tetrahydrothiophene-3,4-diol (5; 465 mg, 2.0mmol) and
6
(2.34 g, 8.8 mmol) in MeCN (30ml) was stirred for 72 h at r.t. The insoluble materials were removed by
filtration, and the filtrate was evaporated. The resulting oily residue was stirred in hexane (ca. 30ml), and the
insoluble materials were removed again. The filtrate was evaporated to give a crude mixture of the syn- and anti-

1468
Helvetica Chimica Acta ± Vol. 88 (2005)
isomers of 4. The mixture was subjected to CC (SiO2), and then to HPLC, to furnish 265 mg (45%) of 4a (syn)
and 59 mg (10%) of 4b (anti).
Data of 4a. Colorless plates. M.p. 110± 111 8 (hexane). UV/VIS (hexane): 253 (2790), 324 (142). IR (KBr):
942, 817, 804, 791, 736, 693, 653 (SˆS). Raman (neat): 816, 650(S ˆS), 631, 594, 383, 287, 228. 1H-NMR
(400 MHz): 1.30 (s, 18 H); 3.41 (d, J ˆ 13.4, 2 H); 3.90( d, J ˆ 13.4, 2 H). 13C-NMR (100.6 MHz): 28.4 (br.); 35.5;
.
‡
39.4; 111.1. EI-MS: 294 (M ), 157. Anal. calc. for C12H22O2S3: C 48.94, H 7.53; found: C 49.05, H 7.57.
Data of 4b. Colorless plates. M.p. 97 ± 998 (hexane). UV/VIS (hexane): 245 (3060), 313 (203). IR (KBr):
967, 949, 810, 797, 735, 692, 670 (SˆS). Raman (neat): 816, 666 (SˆS), 623, 577, 330, 263. 1H-NMR (400 MHz):
1.33 (s, 18 H); 3.13 (d, J ˆ 14.1, 2 H); 3.42 (d, J ˆ 14.1, 2 H). 13C-NMR (100.6 MHz): 28.8 (br.); 38.6; 39.6; 107.0.
.
‡
EI-MS: 294 (M ), 157. Anal. calc. for C12H22O2S3: C 48.94, H 7.53; found: C 49.16, H 7.57.
2.1.2. Reaction at a Temperature of 408. Treatment of 5 (465 mg, 2 mmol) with 6 (2.13 g, 8 mmol) for 48 h at
408 in MeCN (30ml) gave 4a (236 mg, 40%) and 4b (46 mg, 8%).
2.1.3. Reaction in the Presence of 1,2,4-Triazole. Treatment of 5 (58 mg, 0.25 mmol) with 6 (293 mg,
1.1 mmol) in the presence of 1,2,4-triazole (76 mg, 1.1 mmol) for 48 h at 408 in MeCN (4 ml) gave 4a (37 mg,
50%) and 4b (8 mg, 11%).
2.2. Reaction with 1,1'-Thiobis(1H-1,2,4-triazole) (8) as Sulfur Transfer Agent. Treatment of 5 (23 mg,
0.1 mmol) with 8 (74 mg, 0.4 mmol) for 50 h at r.t. in MeCN (2 ml) gave 4a (4 mg, 14%) and 4b (2 mg, 6%).
3. Thermolyses of Compounds 4a and 4b. A soln. of 4a (1 mg, 3 mmol) in [D8]toluene (0.5 ml) was heated at
1208 for 96 h in a sealed NMR tube. The reaction led to 4a, 13, and 14 in a ratio of 48 :39 :13. The products were
identified by comparison of their NMR chemical shifts with those of authentic samples. Although the progress of
the reaction was monitored by 1H-NMR, the formation of 4b by isomerization of 4a was not observed during
this period. Similarly, a soln. of 4b (1 mg, 3 mmol) in [D8]toluene was heated at 1208 in a sealed NMR tube. The
reaction produced 13 and 14 in a ratio of 6 :94, with complete consumption of 4b after heating for 24 h.
Although the progress of the reaction was monitored by 1H-NMR, the formation of 4a by isomerization of 4b
was not observed during this period.
4. Hydrolysis of Compound 4a. A mixture of 4a (3.9 mg, 0.013 mmol) and NaHCO3 (4.5 mg, 0.05 mmol) in
THF/H2O 1:1 (4 ml) was stirred at r.t. for 45 h. The resulting mixture was washed with H2O, dried (MgSO4), and
evaporated. 1H-NMR Analysis showed that the residue (3 mg) was only composed of 5 and 4a in a ratio of
96 :4, corresponding to yields of 93 and 4%, resp.
5. Oxidation Reactions. 5.1. Oxidation of 4a with Dimethyldioxirane (DMD). A 52-mm acetone soln. of
DMD (3.1 ml, 0.16 mmol) was added to a soln. of 4a (52.7 mg, 0.17 mmol) in CH2Cl2 (5 ml) at 08, and the
resulting mixture was stirred for 1.5 h. The mixture was evaporated, and the residue was purified by CC (SiO2)
to furnish 17a (48.2 mg, 91%). Colorless plates. M.p. 148 ± 1508 (CH2Cl2/hexane; dec.). UV/VIS (hexane): 254
(2400), 320 (208). IR (KBr): 1042, 961, 952, 818, 799, 722, 677, 661. 1H-NMR (300 MHz): 1.30 (s, 18 H); 3.05 (d,
.
‡
J ˆ 14.7, 2 H); 4.60( d, J ˆ 14.7, 2 H). 13C-NMR (75 MHz): 26.2 (br.); 39.0; 59.1; 107.1. EI-MS: 310 ( M ), 174,
111. Anal. calc. for C12H22O3S3: C 46.42, H 7.14; found: C 46.62, H 7.17.
5.2. Oxidation of 4b with DMD. A 66.4-mm acetone soln. of DMD (1.5 ml, 0.10 mmol) was added to a soln.
of 4b (31.2 mg, 0.10 mmol) in CH2Cl2 (1.5 ml) at 08, and the resulting mixture was stirred for 1.5 h. The mixture
was evaporated, and the residue was purified by CC (SiO2) to furnish 17b (29.2 mg, 94%). Colorless plates. M.p.
146 ± 1488 (CH2Cl2/hexane; dec). UV/VIS (MeCN): 242 (2691), 312 (46). IR (KBr): 1042 (SˆO), 973, 726, 686,
670(S ˆS). 1H-NMR (400 MHz): 1.35 (s, 18 H); 3.13 (d, J ˆ 15.0, 2 H); 4.02 (d, J ˆ 15.0, 2 H). 13C-NMR
.
‡
(100 MHz): 27.0 (br.); 38.4; 61.2; 103.2. EI-MS: 310 (M ), 174, 111. Anal. calc. for C12H22O3S3: C 46.42, H 7.14;
found: C 46.65, H 7.16.
5.3. Oxidation of 4a with 3-Chloroperbenzoic acid (MCPBA). 5.3.1. With 1.1 Equiv. of Oxidant. A mixture
of 4a (42.6 mg, 0.14 mmol) and MCPBA (27.5 mg, 0.16 mmol) in CH2Cl2 (3 ml) was stirred at À 138 for 1 h. The
1
resulting mixture was washed with H2O, dried (MgSO4), and evaporated. H-NMR Analysis showed that the
residue was composed of 17a and 18b in a ratio of 94 :6. Crystallization from hexane/CH2Cl2 afforded syn,anti-
(3aR,6aS)-3a,6a-di(tert-butyl)tetrahydrothieno[3,4-d][1,3,2]dioxathiole 5-Oxide 2-Sulfide (17a; 33.4 mg, 77%).
5.3.2. With 3.3 Equiv. of Oxidant. To a soln. of 4a (35.8 mg, 0.12 mmol) in CH2Cl2 (3.5 ml) was added
MCPBA (69.2 mg, 0.40 mmol) at 08. The mixture was warmed to r.t. and stirred for 17 h. Then, the mixture was
worked up as described above to afford a solid residue, which, according to 1H-NMR analysis, was composed of
18b and 19b in a ratio of 90:10. The spectroscopic data of 18b and 19b agreed with those of authentic samples
obtained by oxidation of 7b [13].
5.4. Oxidation of syn,anti-(3aR,6aS)-3a,6a-Di(tert-butyl)tetrahydrothieno[3,4-d][1,3,2]dioxathiole 5-Oxide
2-Sulfide (17a). 5.4.1. With MCPBA as Oxidant. A mixture of 17a (5.1 mg, 0.016 mmol) and MCPBA (6.0 mg,
0.035 mmol) in CH2Cl2 (1.5 ml) was stirred at 08 for 3 h. The resulting mixture was washed with H2O, dried

Helvetica Chimica Acta ± Vol. 88 (2005)
1469
(MgSO4), and evaporated. 1H-NMR Analysis of the residue (5.5 mg) indicated 17a, 18a, 18b, and 19b in a ratio
of 8 :14 :66 :12. The spectroscopic data of 18a agreed with those of an authentic sample obtained by oxidation of
7a [13].
5.4.2. With DMD as Oxidant. A soln. of 17a (12.2 mg, 0.039 mmol) in CH2Cl2 (1 ml) and a 51-mm acetone
soln. of DMD (1.7 ml, 0.087 mmol) were combined at 08 and stirred for 3 h. The resulting mixture was
evaporated. 1H-NMR Analysis of the residue indicated 17a, 18a, 18b, 19a, and 19b in a ratio of 13 :58 :21 :5 :3.
The spectroscopic data of 19a agreed with those of an authentic sample obtained by oxidation of 7a [13].
5.4.3. With H2O2 as Oxidant. A mixture of 17a (7.6 mg, 0.024 mmol) and an aq. 30% soln. of H2O2 (15.0mg,
0.13 mmol) was stirred at 08 in acetone (2.0ml), and then at r.t. for 15 h. 1H-NMR Analysis of the crude product
showed that the residue was composed of 17a, 18a, and 18b in a ratio of 46 :4 :50.
5.4.4. With t-BuOCl as Oxidant. An 88-mm soln. of t-BuOCl in CHCl3 (0.9 ml, 0.079 mmol) was added to a
soln. of 17a (11 mg, 0.035 mmol) in a binary mixture of CHCl3 (0.6 ml) and THF (0.5 ml) at À 508. The mixture
was stirred for 15 min, and then treated with ice-cold H2O (2.5 ml). The resulting mixture was stirred without
cooling for 6 h, diluted with CH2Cl2, washed with H2O, dried (MgSO4), and evaporated. 1H-NMR Analysis of
the residue (9.5 mg) indicated 17a, 18a, and 18b in a ratio of 8 :74 :18.
5.5. Oxidation of anti,anti-(3aR,6aS)-3a,6a-Di(tert-butyl)tetrahydrothieno[3,4-d][1,3,2]dioxathiole 5-Ox-
ide 2-Sulfide (17b). 5.5.1. With MCPBA as Oxidant. A mixture of 17b (6.3 mg, 0.020 mmol) and MCPBA
(7.6 mg, 0.044 mmol) in CH2Cl2 (1.5 ml) was stirred at 08 for 3 h. The resulting mixture was washed with H2O,
dried (MgSO4), and evaporated. 1H-NMR Analysis of the residue (6.5 mg) indicated 17b, 18a, 18b, 19a, and 19b
in a ratio of 8 :34 :37:9 :12.
5.5.2. With DMD as Oxidant. A soln. of 17b (5.8 mg, 0.018 mmol) in CH2Cl2 (1 ml) and a 66-mm acetone
soln. of DMD (0.6 ml, 0.040 mmol) were combined at 08 and stirred for 3 h. The resulting mixture was
evaporated. 1H-NMR Analysis of the residue indicated 17b, 18a, 18b, 19a, and 19b in a ratio of 15 :50:23 :7 :5.
5.5.3. With H2O2 as Oxidant. A mixture of 17b (12.7 mg, 0.040 mmol) and an aq. 30% soln. of H2O2
(25.0mg, 0.22 mmol) was stirred at 0 8 in acetone (2.0ml), and then at r.t. for 15 h. 1H-NMR Analysis of the
crude product indicated 17b, 18a, and 18b in a ratio of 59 :16 :25.
5.5.4. With t-BuOCl as Oxidant. An 88-mm soln. of t-BuOCl in CHCl3 (0.3 ml, 0.026 mmol) was added to a
soln. of 17b (3.6 mg, 0.012 mmol) in a mixture of CHCl3 (0.5 ml) and THF (0.3 ml) at À 508. The mixture was
stirred for 15 min, and then treated with ice-cold H2O (2.5 ml). The resulting mixture was stirred without
external cooling for 3 h, diluted with CH2Cl2, washed with H2O, dried (MgSO4), and evaporated. 1H-NMR
Analysis of the residue (3.5 mg) indicated 17a, 17b, 18a, and 18b in a ratio of 10:3 :71:16.
6. Trapping of Sulfur Monoxide. A soln. of MCPBA (20.7 mg, 0.12 mmol) in CH2Cl2 (2.0ml), and a soln. of
the trapping agent 2,3-dimethylbuta-1,3-diene (0.045 ml, 0.4 mmol) in CH2Cl2 (3.0ml) were added separately
and slowly to a stirred soln. of 17a (12.4 mg, 0.040 mmol) in CH2Cl2 (3.0ml) at r.t. The mixture was stirred for
24 h and then evaporated. The residue was purified by CC (SiO2; CH2Cl2/Et2O 1:1) to afford a mixture
(12.7 mg) composed of 17a, 18a, and 18b in a ratio of 34 :3 :63. Further elution of the column (MeOH) gave an
oily mixture (1.3 mg), which was purified by GPC (gel-permeation chromatography) [18] to afford 20 (0.9 mg,
27%).
7. X-Ray Crystal Structures of 4a, 4b, and 17a. Crystal data were recorded on a Mac Science DIP-3000
diffractometer equipped with a graphite monochromator. Oscillation and Weissenberg photographs were
recorded on the imaging plates of the diffractometer with MoKa radiation (l ˆ 0.71073 Š), and data reduction
was performed with the MAC DENZO program system. The cell parameters were determined and refined with
MAC DENZO for all observed reflections. The structures were solved by direct methods using SIR97 [20], and
refined with full-matrix least-squares (SHELXL-97) methods [21] using all independent reflections. Absorption
corrections for 4b and 17a were made by a multi-scan method (SORTAV) [22]. The non-H-atoms were refined
anisotropically. The data are summarized in Table 3.
Crystallographic data for the structural analysis have been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication numbers CCDC-218636 (4a), -218637 (4b), and -218638 (17a).
Copies of the data can be obtained, free of charge, from the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK
(fax: ‡ 44-1223-336033; e-mail: data_request@ccdc.cam.ac.uk; internet: http://www.ccdc.cam.ac.uk/data_
request/cif).

1470
Helvetica Chimica Acta ± Vol. 88 (2005)
Table 3. Crystal Data of 4a, 4b, and 17a
4a
4b
17a
Chemical formula
Formula weight
Crystal shape
Crystal size [mm]
Crystal system
Space group
a [Š]
C12H22O2S3
.51 294.50294.50310
Cube
C12H22O2S3
C12H22O3S3
Cube
Cube
0.20 Â 0.14 Â 0.14
Monoclinic
P21/c
8.660(1)
13.241(1)
13.140(1)
109.381(2)
1424.16(12)
4
1.374
3241
3103
2500
0.20 Â 0.20 Â 0.16
Monoclinic
P21/c
8.820(1)
8.425(1)
19.442(1)
90.198(2)
1444.70(13)
4
1.354
3043
2861
2267
0.32 Â 0.26 Â 0.14
Orthorhombic
P212121
8.622(1)
13.806(1)
24.820(1)
b [Š]
c [Š]
b [8]
V [Š3]
2954.5(2)
8
Z
Dcalc (Mg mÀ3
)
1.396
3511
3106
2357
251
No. of measured refl.
No. of independent refl.
No. of observed refl. (I > 2s(I))
No. of parameters
242
242
R1
wR2
Goodness-of-fit
T (K)
Dpmax
0.034
0.084
1.033
153
0.061
0.178
1.069
153
0.046
0.123
0.984
153
0.41
1.11
0.47
Dpmin
À 0.37
À 1.17
À 0.57
REFERENCES
[1] G. W. Kutney, K. Turnbull, Chem. Rev. 1982, 82, 333.
[2] R. L. Kuczkowski, E. B. Wilson Jr., J. Am. Chem. Soc. 1963, 85, 2028; R. L. Kuczkowski, J. Am. Chem. Soc.
1963, 85, 3047; R. L. Kuczkowski, J. Am. Chem. Soc. 1964, 86, 3617.
[3] a) Q. E. Thompson, M. M. Crutchfield, M. W. Dietrich, J. Org. Chem. 1965, 30, 2696; b) G. K. Abdullaev,
I. A. Mamedov, M. M. Mamedov, Azerb. Khim. Zh. 1973, 5 ± 6, 43; c) K. Miaskiewicz, R. Steudel, J. Chem.
Soc., Dalton Trans. 1991, 2395; d) R. Steudel, Y. Drozdova, K. Miaskiewicz, R. H. Hertwig, W. Koch, J. Am.
Chem. Soc. 1997, 119, 1990.
[4] a) D. N. Harpp, K. Steliou, C. J. Cheer, J. Chem. Soc., Chem. Commun. 1980, 825; b) J. P. Snyder, N. Nevins,
S. L. Tardif, D. N. Harpp, J. Am. Chem. Soc. 1997, 119, 12685; c) E. Zysman-Colman, C. B. Abrams, D. N.
Harpp, J. Org. Chem. 2003, 68, 7059.
[5] F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen, R. Taylor, J. Chem. Soc., Perkin Trans. 2
1987, S1.
[6] S. C. Abrahams, Acta Crystallogr. 1955, 8, 661.
[7] F. Iwasaki, Acta Crystallogr., Sect. B 1979, 35, 2099; C. Tamura, K. Aiba, S. Sato, T. Hata, S. Morimura, T.
Yoshioka, Acta Crystallogr., Sect. B 1977, 33, 3918.
[8] F. Rebiere, O. Samuel, L. Ricard, H. B. Kagan, J. Org. Chem. 1991, 56, 5991.
[9] J. Nakayama, S. Yamaoka, M. Hoshino, Tetrahedron Lett. 1988, 29, 1161; J. Nakayama, R. Hasemi, K.
Yoshimura, Y. Sugihara, S. Yamaoka, J. Org. Chem. 1998, 63, 4912.
[10] D. N. Harpp, K. Steliou, T. H. Chan, J. Am. Chem. Soc. 1978, 100, 1222.
[11] S. Tanaka, Y. Sugihara, A. Sakamoto, A. Ishii, J. Nakayama, J. Am. Chem. Soc. 2003, 125, 9024.
[12] G. Mitchell, in ꢂComprehensive Heterocyclic Chemistry IIꢁ, Ed. R. C. Storr, Pergamon Press, Oxford, 1996,
Vol. 4, Chapt. 4.15.
[13] S. Tanaka, Y. Sugihara, A. Sakamoto, A. Ishii, J. Nakayama, Heteroat. Chem. 2003, 14, 587.
Á
[14] T. Benincori, E. Brenna, F. Sannicolo, L. Trimarco, P. Antognazza, E. Cesarotti, F. Demartin, T. Pilati, G.
Zotti, J. Organomet. Chem. 1997, 529, 445.

Helvetica Chimica Acta ± Vol. 88 (2005)
1471
[15] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski,
J. A. Montgomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin,
M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S.
Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A.
Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L.
Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle, J. A. Pople, Gaussian, Inc., Pittsburgh, PA, USA,
1998.
[16] C. H. Green, D. G. Hellier, J. Chem. Soc., Perkin Trans. 2 1972, 458; C. H. Green, D. G. Hellier, J. Chem.
Soc., Perkin Trans. 2 1973, 243; J. G. Pritchard, P. C. Lauterbur, J. Am. Chem. Soc. 1961, 83, 2105; C. H.
Green, D. G. Hellier, J. Chem. Soc., Perkin Trans. 2 1975, 190; G. W. Buchanan, D. G. Hellier, Can. J. Chem.
1976, 54, 1428.
[17] C. R. Johnson, D. McCants Jr., J. Am. Chem. Soc. 1965, 87, 1109; G. Barbieri, M. Cinquini, S. Colonna, F.
Montanari, J. Chem. Soc. C 1968, 659; W. O. Siegl, C. R. Johnson, J. Org. Chem. 1970, 35, 3657; E. Block,
E. J. Corey, R. E. Penn, T. L. Renken, P. F. Sherwin, J. Am. Chem. Soc. 1976, 98, 5715; I. Jalsovszky, F. Ruff,
Â
M. Kajtar-Peredy, A. Kucsman, Synthesis 1990, 1037; D. V. Deubel, J. Org. Chem. 2001, 66, 2686; T.
Â
Â
Â
Patonay, W. Adam, A. Levai, P. Köver, M. Nemeth, E.-M. Peters, K. Peters, J. Org. Chem. 2001, 66, 2775.
[18] I. A. Abu-Yousef, D. N. Harpp, J. Org. Chem. 1997, 62, 8366; S. L. Tardif, A. Z. Rys, C. B. Abrams, I. A.
Â
Abu-Yousef, P. B. F. Leste-Lasserre, E. K. V. Schultz, D. N. Harpp, Tetrahedron 1997, 53, 12225.
[19] R. S. Berry, J. Chem. Phys. 1960, 32, 933; S. Oae, M. Yokoyama, M. Kise, N. Furukawa, Tetrahedron Lett.
1968, 4131; P. Gillespie, P. Hoffman, H. Klusacek, D. Marquarding, S. Pfohl, F. Ramirez, E. A. Tsolis, I. Ugi,
Angew. Chem., Int. Ed. 1971, 10, 687; S. Oae, T. Kawai, N. Furukawa, F. Iwasaki, J. Chem. Soc., Perkin
Trans. 2 1987, 405; N. Furukawa, T. Shibutani, H. Fujihara, Tetrahedron Lett. 1987, 5845.
[20] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni,
G. Polidori, R. Spagna, J. Appl. Crystallogr. 1999, 32, 115.
[21] G. M. Sheldrick, SHELXL 97, Program for the Refinement of Crystal Structures, 1997.
[22] R. H. Blessing, Acta Crystallogr., Sect. A 1995, 51, 33.
Received January 18, 2005