M. Hayashi et al. / Tetrahedron Letters 45 (2004) 1409–1411
1411
phosphine, and tert-butyl chloride under N2 at rt. A
distinct red color of lithium diphenylphosphide soon
faded. After the mixture was refluxed for 30 min, the
solvent was removed under reduced pressure, and the
residue was directly distilled in vacuo affording 1b (9.1 g,
80%); physical and spectral data for 1b: bp 152 °C/
1.5 mmHg. Anal. Calcd for C18H25PSi: C, 71.96; H, 8.39.
Found: C, 72.00; H, 8.52. 1H NMR (C6D6) d 7.73–7.68
(m, 4H), 7.19–7.11 (m, 6H), 1.03 (s, 9H), 0.21 (s, 3H), 0.20
(s, 3H); 13C NMR (C6D6) d 136.30 (d, JC–P ¼ 15.7 Hz),
135.01 (d, JC–P ¼ 18.5 Hz), 128.33 (d, JC–P ¼ 7.2 Hz), 127.64
(d, JC–P ¼ 0.8Hz), 27.38 (d, JC–P ¼ 2.9 Hz), 18.79 (d,
JC–P ¼ 13.1 Hz), )4.98, )5.07; 31P{1H} NMR (C6D6)
d )61.2. 1c: bp 140 °C/0.7 mmHg. 1H NMR (C6D6)
d 7.78–7.67 (m, 4 H), 7.24–7.13 (m, 6 H), 1.04 (t,
J ¼ 6.4 Hz, 9 H), 0.84 (dq, J ¼ 3.2, 6.4 Hz, 6 H); 13C
NMR (C6D6) d 135.11 (d, JC–P ¼ 15.7 Hz), 134.01 (d,
JC–P ¼ 18.5 Hz), 128.17 (d, JC–P ¼ 7.3 Hz), 127.64 (d,
JC–P ¼ 0.8Hz), 27.38 (d, JC–P ¼ 2.4 Hz), 14.42 (d,
JC–P ¼ 11.1 Hz); 31P{1H} NMR (C6D6) d )60.7. 1d: bp
165 °C/0.6 mmHg. Anal. Calcd for C21H31PSi: C, 73.64; H,
afforded in only 30–58% yields. Introduction of TIPS
group also required slightly excess reagents for satis-
factory results (entries 14, 16, and 17). No silylation
occurred even under more drastic conditions with excess
reagents in the reactions of tertiary alcohol 9 with 1b
(entry 4) and secondary alcohol 5 with 1d (entry 15). A
primary hydroxyl group was rapidly silylated in the
presence of a secondary hydroxyl group in quite selec-
tive manner (entries 3, 12, and 16). In contrast, silylation
of diol 7 by using common methods in short period
resulted in low selectivity (TBDMSOTf/2,6-lutidine/
DMF: 8b 18%; bis-silyl ether 51%) or poor yield
(TBDMSCl/Et3N/DMAP/CH2Cl2: 8b 33%) under simi-
lar conditions (1.2 equiv of each silylating agents, rt,
5 min). These results clearly showed the peculiar reac-
tivity of the present reagents that differed from common
silylating agents; silylation of a primary hydroxyl group
proceeded quite rapidly as a silyl triflate with high
selectivity.
1
9.12. Found: C, 73.80; H, 9.21. H NMR (C6D6) d 7.80–
7.76 (m, 4 H), 7.19–7.11 (m, 6 H), 1.51–1.42 (dsept,
J ¼ 7.6, 2.0 Hz, 3 H), 1.21 (d, J ¼ 7.6 Hz, 18H); 13C NMR
In conclusion, we have discovered that silylphosphines
were instantly activated by means of DEAD to form the
reactive silyl cation equivalents. The present reaction
provides a novel utility of silylphosphines toward
organic synthesis as an acidic silylation procedure for
hydroxyl groups. Further application of silylphosphines
is now under investigation.
(C6D6)
d
136.05 (d, JC–P ¼ 15.8Hz), 134.91 (d,
JC–P ¼ 18.6 Hz), 128.37 (d, JC–P ¼ 7.1 Hz), 127.48, 19.48
(d, JC–P ¼ 4.7 Hz), 13.38(d, JC–P ¼ 10.5 Hz); 31P{1H} NMR
(C6D6) d )57.8.
7. These silylphosphines are still somewhat sensitive to
oxygen and should avoid contact with the air.
8. It was difficult to determine the accurate structure of the
adduct 2 because it showed very broad signals in its NMR
spectra. The FABMS spectrum of 2 showed a parent peak
at m=z ¼ 475, which corresponded to a simple sum of the
molecular weights of 1b and DEAD. In addition, 2 gave
the corresponding desilylated phosphinohydrazinedicarb-
oxylate product quantitatively by an acidic hydrolysis.
These facts clearly indicated that 2 is a 1:1 adduct of 1b
and DEAD. The 1H, 13C, and 31P NMR spectra of 2 at rt
showed the presence of three isomers, which probably
corresponded to one of the E/Z isomer and the rotamers
of the other. It is in accord with the observation of the
coalescence of three peaks in 31P NMR to two peaks on
heating. Thus, we represented tentatively the most plau-
sible structure of 2 as in Scheme 2 speculating from the
results described above and the mechanistic consideration
of the Mitsunobu reaction.9
Acknowledgements
The authors thank Venture Business Laboratory of
Ehime University for their financial support. We also
thank Integrated Center for Science, Center for Coop-
erative Research and Development, Ehime University
for measurement of mass spectra and elemental analysis.
This work was supported by the Fujisawa Foundation
and Saneyoshi Scholarship Foundation.
References and notes
1. For review, see: Fritz, G.; Scheer, P. Chem. Rev. 2000, 100,
3314–3401.
2. (a) Abel, E. W.; Sabherwal, I. H. J. Chem. Soc. (A) 1968,
1105–1108; (b) Thottathil, J. In Handbook of Organophos-
phorus Chemistry; Engel, R., Ed.; Marcel Dekker: New
York, 1992; pp 84–85.
9. These betaines were speculated from the mechanism
accepted for the Mitsunobu reaction, see: Hughes, D. L.
Org. React. 1992, 42, 335.
10. A typical procedure of the silylation by a silylphosphine
and DEAD is depicted as follows: to a mixture of 1b
(54 mg, 0.18mmol), 3-phenyl-1-propanol
3
(20 mg,
3. (a) Tunney, S. E.; Stille, J. K. J. Org. Chem. 1987, 52, 748–
753; (b) Kunzek, H.; Braun, M.; Nesener, E.; Ruehlmann,
K. J. Organometal. Chem. 1973, 49(1), 149–156.
4. Bordachev, A. A.; Kabachnik, M. M.; Novikova, Z. S.;
Beletskaya, I. P. Izv. Akad. Nauk, Ser. Khim. 1994;(4), 756.
5. Luther, G. W.; Beyerle, G. Inorg. Synth. 1977, 17, 186.
6. Representative procedure for preparation of 1b: To a
solution of tert-butyldimethylchlorosilane (6.3 g, 42 mmol)
in THF was added a THF solution of lithium diphenyl-
phosphide (38mmol) prepared from lithium, triphenyl-
0.15 mmol), and PPTS (45 mg, 0.18mmol) in CH 2Cl2
(1.5 mL) was added DEAD (31 mg, 0.18mmol) dropwise
at rt. A distinct yellow color of DEAD immediately
dissappeared upon addition. After the addition was
completed, the reaction mixture was concentrated and
purified by a column chromatography on silica gel to
afford the silyl ether 4b (36 mg, 95%).
11. Greene, T. W.; Wuts, P. G. M. Protective Groups in
Organic Synthesis. 3rd ed.; John Wiley: New York, 1999;
pp 113–148.