182 Bull. Chem. Soc. Jpn., 74, No. 1 (2001)
© 2001 The Chemical Society of Japan
tacks the nucleophilic oxygen of 6 in the reaction process. In
OTf anion. Therefore, the transfer of the trimethylsiloxysulfo-
nyl cation from the oxonium ion to OTf– did not occur in the re-
action (Eq. 4).
the second study, the desulfation reaction of decyl trimethyl-
1
silyl sulfate 7 with 2a in acetonitrile-d3 was monitored by H
and 29Si NMR spectroscopies.8 In the resulting reaction mix-
ture, bis(trimethylsilyl)sulfate and acetonitrile were detected
along with decyl trimethylsilyl ether as the main products:
C10H21OSO3SiMe3 +SiMe3OTf −→
7
2a
−OTf
C10H21O+ SO3SiMe3
(4)
C10H21OSO3SiMe3
−−−−−−−−−−→
CD3CN,100◦C,2h
SiMe3
2g
7
C10H21OSiMe3 +CH3CN+(Me3SiO)2SO2.
(2)
In summary, we developed an efficient method for the prepa-
ration of secondary alkyl sulfates having a primary hydroxy
group by the regioselective desulfation of bis(trimethylsilyl)
ester of 1-O-decyl glycerol-2,3-disulfate. The oxomium salts
might play an important role in proceeding the reaction.
From these results, we postulate a possible mechanism of
the desulfation reaction, illustrated in Scheme 2. As shown,
since the trimethylsilyl group seems to attack the alcoholic ox-
ygen of 7, oxonium salt 8 would be generated. The oxonium
salt should be liable to release the trimethylsiloxy sulfonyl cat-
ion, and the released cation would be trapped by the imidate
anion to give 9, which would subsequently decompose to ace-
tonitrile and bis(trimethylsilyl)sulfate. Because the oxonium
salt 8 would be the intermediate, a steric hindrance in the vicin-
ity of the oxygen atom of 8 would cause a difference in the ox-
onium formation between the primary alkyl sulfate and sec-
ondary alkyl sulfate, and the difference would affect the selec-
tivity of the reaction.
Experimental
Preparation of 1a. After a mixture of 1-O-decyl glycerol
(10.8 g, 50.0 mmol) with chloro sulfuric acid (14.0 g, 120.0 mmol)
in dichloromethane (40.0 ml) was stirred at r.t. for 0.5 h , chlorotri-
methylsilane (19.6 g, 180.0 mmol) was added. The solution was
stirred for 0.5 h at 50 °C and then concentrated under reduced
pressure to give the blown amorphous of 1a (26.0 g) quantitatively
without any impurities. 1a: 1H NMR(CDCl3) δ 0.42 (s, 18H), 0.85
(t, 3H), 1.24 (m, 14H), 1.53 (m, 2H), 3.45 (t, 2H), 3.69 (d, 2H),
4.38 (dd, 1H), 4.48 (dd, 1H), 4.83 (m, 1H).
General Procedure. A solution of 1a (2.6 g, 5.0 mmol) and
2a (2.03 g, 10.0 mmol) in acetonitrile (5.0 ml) was heated at 100
°C for 2 h. A NaOH solution was then added, and the solution was
evaporated. The yield was determined by 1H NMR of this residue.
The residue was easily purified by ODS gel chromatography or re-
1
crystallization. 3: H NMR (D2O) δ 0.90 (t, 3H), 1.31 (m, 14H),
1.61 (m, 2H), 3.52 (m, 2H), 3.71−3.89 (m, 3H), 3.85 (dd, 1H), 4.51
(m, 1H). 4: 1H NMR (D2O) δ 0.89 (t, 3H), 1.31 (m, 14H), 1.62 (m,
2H), 3.50−3.64 (m, 4H), 3.98−4.13 (m, 3H).
Scheme 2.
References
The postulated mechanism requires an electrophilic attack
of the trimethylsiloxysulfonyl cation on the imidate anion to
form bis(trimethylsilyl)sulfate. Evidence for this electrophilic
reaction process was gained in the reaction of 7 with 2b. In this
reaction, Me2NSO3SiM3 was obtained as a by-product along
with decyl trimethylsilyl ether (Eq. 3). This result shows that
an electrophilic attack of the trimethylsiloxysulfonyl cation to
Me2N– occurred in the reaction.9
1
M. Nakamura, T. Takeda, H.Yamada, M. Miyamato, and A.
Hasegawa, Kagaku To Kougyou, 66, 400 (1992).
B. Gruber, B. Fabry, B. Giesen, and W. F. Muller, Tenside
Surf. Det., 30, 422 (1993).
2
3
E. Gilbert, “Sulfation and Related Reactions,” Interscience
Publishers, NewYork (1965), Chap. 6.
4
5
E. Gilbert, Chem. Rev., 62, 549 (1962).
“Methoden der Organishen Chemie Band E11/Teil12,”ed
by Houben Weyl, Georg Thieme Verlag, Stuttgart (1985), p. 997.
R. Takano, M. Matsuno, K. Kamei-Hayashi, and S. Hirase,
Biosci. Biotech. Biochem., 56, 1577 (1992).
M. Matsuno, R. Takano, K. Kamei-Hayashi, and S. Hara,
Carbohydr. Res., 241, 209 (1993).
29Si NMR(CDCl3) δ 15.9 (C10H21OSiMe3), 33.9 ((Me3-
SiO)2SO2), 35.6 (C10H21OSO3SiMe3).
29Si NMR(CDCl3) δ 31.6 (Me2NOSO3SiMe3).
C10H21OSO3SiMe3 +Me2NSiMe3
6
7
2b
−−−−−−−−−−→
C10H21OSiMe3 +Me2NSO3SiMe3. (3)
7
CD3CN,100◦C,2h
In the case of 2g, which did not give a secondary sulfate in
the reaction with 1a, it seems reasonable to assume that the
formed oxonium salt is very stable because of the stability of
8
9