J. Am. Chem. Soc. 1996, 118, 1577-1578
1577
Scheme 1
Zirconium-Catalyzed Enantioselective
Alkylalumination of Monosubstituted Alkenes
Proceeding via Noncyclic Mechanism
Denis Y. Kondakov and Ei-ichi Negishi*
Department of Chemistry, Purdue UniVersity
West Lafayette, Indiana 47907
ReceiVed October 31, 1995
We recently reported a Zr-catalyzed enantioselective methyl-
alumination of monosubstituted alkenes.1 In contrast, the initial
outlook for achieving a similar Zr-catalyzed enantioselective
alkylmetalation with ethyl-, propyl-, and higher alkylmetals was
not promising. The reaction of monosubstituted alkenes with
EtMgX, where X ) Cl or Br, in the presence of a catalytic
amount of Cp2ZrCl2 was known to give (2-ethylalkyl)magne-
sium halides.2 The reaction was later shown to proceed via a
cyclic mechanism,3,4 and its successful application to the
development of a cyclic enantioselective carbometalation-
elimination tandem reaction involving cyclic, allylic ethers and
amines has been recently reported.5 However, attempts to
develop Zr-catalyzed enantioselective conversion of monosub-
stituted alkenes into (2-ethylalkyl)magnesium halides have led
only to very disappointing results. Also known was the reaction
of monosubstituted alkenes with Et3Al, catalyzed by Cp2ZrCl2,
producing aluminacyclopentanes, e.g., 1,6 but the reaction of
1-decene with 1 equiv of Et3Al in the presence of 8 mol % of
dichlorobis(neomenthylindenyl)zirconium (2)7 in hexanes pro-
duced, after oxidation, a 65% yield of 2-(n-octyl)-1,4-butanediol
(3) in only 33% ee (Scheme 1).
68% ee. In view of the dramatic solvent effects discussed above,
various halogenated solvents were screened, and it was found
that the use of CH3CHCl2 or CH2Cl2 in place of (CH2Cl)2
boosted the % ee figures to the 80s, and they were further
increased to the 90-95% range for various monosubstituted
alkenes by merely lowering the reaction temperature to 0 °C.
Some representative results are summarized in Table 1. The
% ee figures were determined from the 1H NMR spectra of the
(+)- and (-)-MTPA esters. In most cases, the CH2 group
directly bonded to OH showed distinct signals for the two
diastereomers. In the cases of (R)-2-ethyl-1-hexanol and (R)-
2-ethyl-1-decanol, the CH3 signals for the ethyl substituent were
used for this purpose. As in the methylalumination with Me3-
Al and 2, alkylmetalation takes place selectively and uniformly
on the re face of alkenes.9 The optimized % ee figures range
from 90 to 96% except for one case involving (n-Oct)3Al, where
the product was 85% ee. Here again, the presence of remote
hydroxy and amino groups may not significantly affect the
course of reaction, and diallyldimethylsilane underwent an
inter-intra tandem carbometalation exhibiting >92% de and
96% ee stereoselectivity figures. On the other hand, ethylalu-
mination of styrene gave an intractable product, and that of
cyclohexylethene did not proceed over 12 h at 25 °C.
The highly favorable results observed in ethylalumination in
CH3CHCl2 prompted us to reexamine the previously reported
methylalumination1 in this solvent. Under otherwise the same
conditions, the reaction of 1-octene with Me3Al in the presence
of 8 mol % of 2 at 25 °C in CH3CHCl2 gave, after oxidation, an
83% yield of (R)-2-methyl-1-octanol in 81% ee, corresponding
to an increase by roughly 10% in % ee.1 Evidently, % ee figures
for ethylalumination are 10-15% higher than the corresponding
figures for methylalumination under comparable conditions. On
the other hand, the yields of ethylalumination are slightly but
unmistakably lower than those of methylalumination. A detailed
analysis of the products of the above reaction indicated the
presence of 3-methylundecane (17%), decane (2%), and the
unreacted 1-decene (2%). No more than traces, if any, of
2-ethyl-1-decene and dimeric products were present. Clearly,
the lower yield is not due to competitive hydrometalation. Since
the results of deuterolysis with DCl-D2O closely parallel those
of oxidation, the origin of 3-methylundecane does not appear
to be due to incomplete oxidation. It must have been formed
during the carboalumination itself. The use of (CH2Cl)2 and
CH2Cl2 led to its formation in 12 and 11%, respectively. The
fact that the reaction run in CD2Cl2 followed by protonolysis
with 3 N HCl does not incorporate D suggests that the solvents
Interestingly, the reaction of 1-decene with 1 equiv of Et3Al
in the presence of 8 mol % of Cp2ZrCl2 in (CH2Cl)2 in place of
hexanes produced, after deuterolysis, a 37% yield of 3-(deute-
riomethyl)undecane (4), which contained D in the C-1 position
only to the extent of 9%. The extent of D incorporation in the
deuteriomethyl group was >90%. 2-Ethyl-1-decene (5) and
1-deuteriodecane (6) were also obtained in 20% yield each8 (eq
1). Although synthetically unattractive, these results clearly
indicated that, in polar solVents, e.g., (CH2Cl)2, noncyclic
ethylalumination similar to methylalumination,1 partially ac-
companied by competitiVe hydroalumination inVolVing a (2-
ethyldecyl)alane, can take place in preference to the preViously
reported cyclic carbometalation processes.2-6 Following this
intriguing lead, we treated 1-decene with Et3Al in (CH2Cl)2 at
25 °C using 8 mol % of 2 as a catalyst and obtained, after
oxidation with O2, a 65% yield of (R)-2-ethyl-1-decanol (7) in
(1) Kondakov, D. Y.; Negishi, E. J. Am. Chem. Soc. 1995, 117, 10771.
(2) Dzhemilev, V. M.; Vostrikova, O. S.; Sultanov, R. M. IzV. Akad.
Nauk SSSR, Ser. Khim. 1983, 219.
(3) Takahashi, T.; Seki, T.; Nitto, Y.; Saburi, M.; Rousset, C. J.; Negishi,
E. J. Am. Chem. Soc. 1991, 113, 6266.
(4) For related suggestions and studies, see: (a) Hoveyda, A. H.; Xu, Z.
J. Am. Chem. Soc. 1991, 113, 5079. (b) Knight, K. S.; Waymouth, R. M.
J. Am. Chem. Soc. 1991, 113, 6268. (c) Lewis, D. P.; Muller, D. M.; Whitby,
R. J.; Jones, R. V. H. Tetrahedron Lett. 1991, 33, 6797.
(5) Morken, J. P.; Didiuk, M. T.; Hoveyda, A. H. J. Am. Chem. Soc.
1993, 115, 6997.
(6) Dzhemilev, V. M.; Ibragimov, A. G.; Zolotarev, A. P.; Muslukhov,
R. R.; Tolstikov, G. A. IzV. Akad. Nauk SSSR, Ser. Khim. (Engl. Transl.)
1989, 194; 1991, 2570.
(7) Erker, G.; Aulbach, M.; Knickermeier, M.; Wingbermuhle, D.;
Kru¨ger, C.; Nolte, M.; Werner, S. J. Am. Chem. Soc. 1993, 115, 4590.
(8) The amounts of possible dimeric products, such as 2-(n-octyl)-
dodecane and 2-(n-octyl)-1-dodecene, were e1-2%, if any.
(9) The absolute configurations of (R)-2-ethyl-1-hexanol (Barth, S.;
Effenberger, F. Tetrahedron Asymmetry 1993, 4, 823), (R)-5-methyl-1-
octanol (Sonnet, P. E.; Gazzillo, J. A.; Dudley, R. L.; Boswell, R. T. Chem.
Phys. Lipids 1990, 54, 205), and (R)-5-methyl-1-heptanol (Chattopadhyay,
S.; Mamdapur, V. R.; Chadha, M. S. Synth. Commun. 1990, 20, 825) were
assigned by comparing the observed optical rotation signs and magnitudes
with those reported in the literature. The other assignments are more tentative
and are based, in part, on empirical predictions [(a) Brewster, J. H. J. Am.
Chem. Soc. 1959, 81, 5475. (b) Marker, R. E. J. Am. Chem. Soc. 1936, 58,
976)] and experimental observation of optical rotation signs and, in part,
on an assumption that the uniform sign of the observed optical rotation is
an indication that the face selection remains the same for all cases herein
reported.
0002-7863/96/1518-1577$12.00/0 © 1996 American Chemical Society