Control of ansa-zirconocene stereochemistry by reversible exchange of cyclopentadienyl and chloride ligands

Article abstract of DOI:10.1021/ja0681523

The reaction of Li₂[Me₂Si(3-R-C₅H₃)₂] salts with Zr{Me₃SiN(CH₂)₃NSiMe₃}Cl₂(THF)₂ in THF quantitatively afforded rac-Me₂Si(3

Full text of DOI:10.1021/ja0681523

Published on Web 03/01/2007
Control of ansa-Zirconocene Stereochemistry by Reversible Exchange of
Cyclopentadienyl and Chloride Ligands
Richard M. Buck, Nawaporn Vinayavekhin, and Richard F. Jordan*
Department of Chemistry, The UniVersity of Chicago, 5735 South Ellis AVenue, Chicago, Illinois 60637
Received November 14, 2006; E-mail: rfjordan@uchicago.edu
Scheme 1 a
The development of efficient routes to chiral ansa-zirconocenes
is important owing to the utility of these complexes in catalysis.1-3
We report that the substitution of Zr-Cl ligands by cyclopentadienyl
ligands (Cp-) is reversible and that this property can be exploited
in the predictable synthesis of racemic ansa-zirconocenes.
We reported the stereoselective synthesis of ansa-zirconocenes
by the reaction of ansa-bis-Cp- reagents (1) with Zr{RN(CH2)3-
NR}Cl2(THF)2 complexes (R ) SiMe3 (2), Ph (3)).2b As shown in
Scheme 1, the reaction of Li2[Me2Si(3-tBu-C5H3)2] (1a) with 2 in
THF affords pure rac-Me2Si(3-tBu-C5H3)2Zr{Me3SiN(CH2)3NSiMe3}
(rac-4a); metallocene products are not formed in Et2O because of
the insolubility of the reactants. In contrast, reaction of 1a with 3
in Et2O affords pure meso-Me2Si(3-tBu-C5H3)2Zr{PhN(CH2)3NPh}
(meso-5a), whereas rac/meso-5a mixtures are formed in THF. We
studied the scope and mechanism of these reactions to understand
these results.
The reaction of 1b-g with 2 in THF affords rac-Me2Si(3-R-
C5H3)2Zr{Me3SiN(CH2)3NSiMe3} (rac-4b-g) in quantitative iso-
lated yield (Scheme 1). In contrast, the reaction of 1d,e with 3 in
Et2O affords meso-Me2Si(3-R-C5H3)2Zr{PhN(CH2)3NPh} (meso-
t
a R ) Bu (a), SiMe3 (b), cyclohexen-1-yl (c), 1-Me-Cy (d), 1-Ph-Cy
(e), 1-Me-cyclo-C12H22 (f), CMe2Ph (g)
5d,e) in >95% NMR yield and 71-91% isolated yield. These
results show that the behavior of 1a in Scheme 1 is characteristic
for this class of ligands. The reaction of rac-4a-g with HCl gives
the corresponding rac-Me2Si(3-R-C5H3)2ZrCl2 complexes (rac-6a-
g) with retention of stereochemistry. Reaction of meso-5d with HCl
gives 6d with a slight loss in stereochemistry (rac/meso ) 1/16).
To probe the mechanism of stereocontrol in the formation of
rac-metallocenes in Scheme 1, the reaction of 1c and 2 in THF-d8
at 60 °C was monitored by NMR. These experiments showed that
1c and 2 are completely converted within 5 min to a 2/1 rac/meso-
4c mixture, which in turn converts to pure rac-4c in 6 h. No
precipitates or intermediates were observed, and the sum of the
concentrations of rac- and meso-4c remained constant after the
consumption of 1 and 2 was complete. The conversion of meso-4c
to rac-4c displays first-order behavior in metallocene (Figure 1,
Figure 1. Time dependence of the concentrations of rac-4c (upper curves)
run i). Similar observations were made for the reaction of 1b with
2. These results show that the formation of rac-metallocenes by
the reaction of 1 and 2 in THF is thermodynamically controlled.
The meso to rac isomerization requires cleavage of a Zr-Cp
bond and re-coordination of the Cp through the opposite face.
Several mechanisms for such Cp enantioface exchange processes
have been identified in metallocenes, including photochemical,
thermal, or radical-induced M-Cp bond homolysis, silatropic
rearrangement, reversible amine elimination, heteroatom-assisted
enantioface exchange, and LiCl-induced M-Cp bond heterolysis.3-5
A series of experiments was performed to probe the mechanism in
the present system. As shown in Figure 1, conversion of the 2/1
rac/meso-4c mixture (initially formed from 1c and 2 in THF-d8) to
pure rac-4c occurs at the same rate in ambient fluorescent light
(run i) and in the dark (run ii), which is inconsistent with a
photochemical meso/rac isomerization. To probe the role of the
and meso-4c (lower curves) measured relative to an internal standard starting
from a 2/1 rac/meso-4c mixture (THF-d8, 60 °C). Run i (black, squares), 2
equiv LiCl; run ii (violet, diamonds), 2 equiv LiCl and dark; run iii (blue,
diamonds), no additive; run iv (green, squares), 2 equiv Li[B(C6F5)4]; run
v (red, squares), 2 equiv [nBu4N]Cl.
LiCl byproduct, which is soluble in THF, the LiCl was removed
from the 2/1 rac/meso-4c mixture (see Supporting Information for
details), and the sample was monitored by NMR. In this case,
essentially no rac/meso isomerization occurred (run iii). Addition
of Li[B(C6F5)4] as a Li+ source to the LiCl-free rac/meso-4c mixture
had no effect (run iv). However, addition of [nBu4N]Cl to the LiCl-
free rac/meso-4c mixture resulted in rapid conversion (<5 min) to
pure rac-4c (run v). Similar results were obtained for rac/meso-
4b. These results show that the isomerization is catalyzed by
chloride ion.6 [nBu4N]Cl is a more effective rac/meso isomerization
catalyst than LiCl because it is less strongly ion-paired.
9
3468
J. AM. CHEM. SOC. 2007, 129, 3468-3469
10.1021/ja0681523 CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2
The kinetics of isomerization of rac-6d, and of a 1/16 rac/meso-
6d mixture, catalyzed by [nBu4N]Cl in THF-d8 were studied in
detail. These reactions both afford a 1/2 equilibrium mixture of
rac/meso-6d (2 d, 60 °C) and exhibit clean first-order approach-
to-equilibrium kinetics. Identical kinetics are observed in ambient
room light and in the dark, and no reaction occurs in the absence
of chloride. These results are consistent with a mechanism
analogous to that in Scheme 2.
To probe if the SiMe2 bridge is required for facile displacement
of Cp- by chloride, a nonbridged system was investigated. The
reaction of a 1/1 mixture of (C5H5)2ZrCl2 and (C5H4Me)2ZrCl2 with
[nBu4N]Cl in THF-d8 afforded a 1/2/1 mixture of (C5H5)2ZrCl2,
(C5H5)(C5H4Me)ZrCl2, and (C5H4Me)2ZrCl2 after 1 h at 60 °C. An
identical dark reaction yielded the same 1/2/1 mixture. No reaction
occurs in the absence of chloride.4d
Several conclusions emerge from these studies. (i) Cyclopenta-
dienyl ligands are easily displaced from zirconocene species by
chloride ion under mild conditions. (ii) As a result, the generation
of zirconocenes by Cp-/Cl- substitution is reversible under
conditions where the displaced Cl- remains in solution. (iii) In the
case of ansa-zirconocene synthesis via the reaction of ansa-bis-
Cp- reagents with Zr{RN(CH2)3NR}Cl2(THF)2 or enantiopure Zr-
{RNCHMeCH2CHMeNR}Cl2(THF)2 compounds,2c N-R groups
that deliver the desired {ansa-bis-Cp}Zr(bis-amide) stereoisomer
in high yield can be chosen in adVance based on the relative
energies of the {ansa-bis-Cp}Zr(bis-amide) products, which can
be computed (e.g., by DFT).2e Thus ansa-zirconocenes can now
be made with a high degree of predictability. (iv) Facile loss of
metallocene stereochemistry can occur under conditions where free
chloride or other nucleophilic species are present, which has
important implications for stereoselective catalysis.
The solubility of LiCl is very low in Et2O, which should disfavor
Cl--catalyzed rac/meso isomerization in this solvent. NMR moni-
toring of the reaction of 1a with 3 in Et2O-d10 at 22 °C showed
that the starting materials are completely converted to meso-5a
within 2 h. No intermediates or further reaction were observed. In
contrast, NMR monitoring of the same reaction in THF-d8 at 0 °C
revealed the initial formation of a 1/3 rac/meso-5a mixture within
4 h and subsequent conversion to an equilibrium 3/1 rac/meso-5a
mixture. Complex meso-5a is stable in THF, but addition of LiCl
or [nBu4N]Cl to a solution of meso-5a in THF-d8 results in
conversion to the equilibrium 3/1 rac/meso-5a mixture. These
results show that the formation of meso-metallocenes by the reaction
of 1 and 3 in Et2O is kinetically controlled.
The kinetics of isomerization of meso-5a to the equilibrium rac/
meso-5a mixture in the presence of LiCl or [nBu4N]Cl in THF-d8
were measured by NMR and exhibit clean first-order approach-to-
equilibrium kinetics (eq 1,2). kobs is the sum of the forward (k1,
meso to rac) and reverse (k-1, rac to meso) rate constants, and Keq
) k1/k-1. A series of approach-to-equilibrium experiments using
varying concentrations of LiCl established that the isomerization
is first order in [Cl-]. The mechanism in Scheme 2, in which rac
and meso interconvert via a transient “mono-Cp” η5,η0-Me2Si(3-
R-C5H3)2Zr{Me3SiN(CH2)3NSiMe3}Cl- intermediate (A), is con-
sistent with these results.
Acknowledgment. This work was supported by the National
Science Foundation (Grant CHE-0212210).
Supporting Information Available: Experimental procedures,
kinetic analyses, and data for new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
References
(1) (a) Resconi, L.; Cavallo, L.; Fait, A.; Piemontesi, F. Chem. ReV. 2000,
100, 1253. (b) Hoveyda, A. H. Chiral Zirconium Catalysts for Enantio-
selective Synthesis. In Titanium and Zirconium in Organic Synthesis;
Marek, I., Ed.; Wiley-VCH Verlag GmbH: Weinheim, Germany, 2002;
pp 180-229.
(2) (a) Zhang, X.; Zhu, Q.; Guzei, I. A.; Jordan, R. F. J. Am. Chem. Soc.
2000, 122, 8093. (b) LoCoco, M. D.; Jordan, R. F. Organometallics 2003,
22, 5498. (c) LoCoco, M. D.; Jordan, R. F. J. Am. Chem. Soc. 2004, 126,
13918. (d) LoCoco, M. D.; Zhang, X.; Jordan, R. F. J. Am. Chem. Soc.
2004, 126, 15231. (e) Dunn, A. R.; Sweet, L. E.; Wiser, D. C.; LoCoco,
M. D.; Jordan, R. F. Organometallics 2004, 23, 5671.
meso-5a yk1z rac-5a
(1)
(2)
k-1
[meso-5a] - [meso-5a]∞
ln
) -kobst
(3) Damrau, H.-R. H.; Royo, E.; Obert, S.; Schaper, F.; Weeber, A.;
Brintzinger, H.-H. Organometallics 2001, 20, 5258.
(
)
[meso-5a]0 - [meso-5a]∞
(4) (a) Wild, F. R. W. P.; Zsolnai, L.; Huttner, G.; Brintzinger, H. H. J.
Organomet. Chem. 1982, 232, 233. (b) Collins, S.; Hong, Y.; Taylor, N.
J. Organometallics 1990, 9, 2695. (c) Schmidt, K.; Reinmuth, A.; Rief,
U.; Diebold, J.; Brintzinger, H. H. Organometallics 1997, 16, 1724. (d)
Brubaker, C. H., Jr.; Peng, M. H. J. Organomet. Chem. 1977, 135, 333.
(5) (a) Ringwald, M.; Stu¨rmer, R.; Brintzinger, H. H. J. Am. Chem. Soc. 1999,
121, 1524. (b) Miyake, S.; Henling, L. M.; Bercaw, J. E. Organometallics
1998, 17, 5528. (c) Diamond, G. M.; Rodewald, S.; Jordan, R. F.
Organometallics 1995, 14, 5. (d) Curnow, O. J.; Fern, G. M.; Hamilton,
M. L.; Zahl, A.; van Eldik, R. Organometallics 2004, 23, 906. (e) Hollis,
T. K.; Wang, L.; Tham, F. J. Am. Chem. Soc. 2000, 122, 11737. (f) Yoder,
J. C.; Day, M. W.; Bercaw, J. E. Organometallics 1998, 17, 4946. (b)
Lin, R. W. U.S. Patent 5,965,759, 1999.
To probe if a bis-amide ligand is required for chloride-catalyzed
rac/meso isomerization, several Me2Si(η5-3-R-C5H3)2ZrCl2 com-
plexes were examined. Reaction of rac-6c with [nBu4N]Cl under
the conditions used for isomerization of rac/meso-4c (Figure 1, run
v) afforded an equilibrium 0.9/1 rac/meso-6c mixture.7 The
isomerization of 6c followed first-order approach-to-equilibrium
kinetics and k1 (meso to rac) was >25 times slower than the value
estimated for 4c. Similarly, the isomerization of 6b is much slower
than that of 4b. These results show that the bis-amide ligand
accelerates but is not required for rac/meso isomerization. The
strong donor ability of the bisamide ligand may stabilize the electron
deficient intermediate A.
(6) For NMe2H-catalyzed rac/meso isomerization of Me2Si(3-tBu-C5H3)2Zr-
(NMe2)2 see: Diamond, G. M.; Jordan, R. F.; Petersen, J. L. Organome-
tallics 1996, 15, 4045.
(7) No reaction was observed in the absence of [nBu4N]Cl.
JA0681523
9
J. AM. CHEM. SOC. VOL. 129, NO. 12, 2007 3469