not effective for secondary allyl ethers such as 1b (R1 = H, R2
= Ph) (entry 1 in Table 1). Similarly to the treatment with
hydrogen, various metal hydrides were found to activate the Ir-
cod complexes, presumably by removing the cod ligand via
hydrometalation. The addition of catecholborane or DIBAL-H
(1 equiv.) generated the more active catalyst solution than
treatment with hydrogen, but the arylphosphine derivatives did
not give good results due to their large steric hindrance (entries
2 and 3). Finally, Ir–trialkylphosphine complexes 3 obtained by
hydrogenation of a mixture of [Ir(cod)2]PF66 and R3P (R = Et,
Pr, Bu) (2 equiv.) gave an excellent catalyst for the isomeriza-
tion of secondary allyl ethers with (Z)-4 predominating (Z =
88–92%) (entries 5–7). Trimethylphosphine, tricyclohexyl-
phosphine (PCy3) and bidentate phosphine ligands such as dppe
are not effective (entries 4, 8 and 9).
R1
R1
OSiR3
3
R2
L2Ir+
OSiR3
3
(E)-4
R2
L2Ir+
H
H
5
R1
6
R1
R2
OSiR3
R2
OSiR3
3
L2Ir+
(Z)-4
3
L2Ir+
H
H
7
8
Scheme 2
The isomerization of the representative allyl silyl ethers with
3 (R = Pr) is summarized in Table 2. In an acetone solution, the
isomerization of primary allyl ethers was completed within 30
min, provided (E)-4 in high yields and with high selectivity,
results which were comparable to those obtained for reactions
catalyzed by 2 (entries 1–5). The reaction was further
accelerated in a mixed solvent of acetone and CH2Cl2 giving
(Z)-4 predominantly (E:Z = 24:76) (entry 1), although all
attempts for the stereoselective preparation of (Z)-4 were
unsuccessful. The Z-selectivity further improved for secondary
allyl trimethylsilyl ethers (entries 6–10), especially when R2
was a secondary alkyl unit (entries 8 and 9). Both catalysts
reported by Felkin (2) and 3 worked well for primary allyl ethers
(entry 4), but 3 demonstrated a higher catalytic efficiency for
secondary allyl ethers.
reacted in acetone, the kinetic products [(E)-4] were obtained
through the syn-p-allyl intermediate [5 ? 6 ? (E)-4]. On the
other hand, high Z-selectivity was achieved for secondary allyl
ethers under conditions leading to equilibration, where the steric
difference between the R2 and R3Si groups controls the
stereochemistry of the products [5 ? 6 ? 8 or 7 ? 8 ?
(Z)-4].
Notes and References
† E-mail: miyaura@organ.hokudai.ac.jp
1 H. M. Colquhoun, J. Holton, D. J. Thompson and M. V. Twigg, New
Pathways for Organic Synthesis—Practical Applications of Transition
Metals, Plenum Press, London, 1984, pp. 173–193.
2 H. Suzuki, Y. Koyama, Y. Moro-oka and T. Ikawa, Tetrahedron Lett.,
1979, 1415.
3 D. Baudry, M. Ephritikhine and H. Felkin, J. Chem. Soc., Chem.
Commun., 1978, 694.
4 T. Moriya, A. Suzuki and N. Miyaura, Tetrahedron Lett., 1995, 36,
1887.
5 P. L. Hall, J. H. Gilchrist and D. B. Collum, J. Am. Chem. Soc., 1991, 113,
9571.
6 M. Green, T. A. Kuc and S. H. Taylor, J. Chem. Soc. (A), 1971, 2334.
7 R. H. Crabtree, M. F. Mellea, J. M. Mihelcic and J. M. Quirk, J. Am.
Chem. Soc., 1982, 104, 107.
The iridium-catalyzed isomerization of allyl ethers proceeds
through the oxidative addition of an allylic C–H bond to the
iridium(i) metal center giving a syn-p-allyliridium complex 6
which selectively led to (E)-4 (Scheme 2).1–3
1
The H NMR study reveled that the isomerization involves
two process: the first and selective formation of (E)-4
(kinetically controlled process) is followed by equilibration to a
mixture of (E)- and (Z)-4 through the anti-p-allyl intermediate
(thermodynamically controlled process) which is slow in a
solvent coordinating to the iridium metal center such as
acetone.7 Thus, the stereochemistry of 4 is highly dependent on
the solvents and the substrates. When primary allyl ethers were
Received in Cambridge, UK, 14th April 1998; 8/02760I
1338
Chem. Commun., 1998