Diastereomerically-specific zirconium complexes of chiral salan ligands: Isospecific polymerization of 1-hexene and 4-methyl-1-pentene and cyclopolymerization of 1,5-hexadiene

Article abstract of DOI:10.1021/ja0647654

Chiral Salan ligands were found to wrap in a highly diastereoselective manner around zirconium leading to C2-symmetric complexes of predetermined chirality at the metal. These complexes led to active polymerization of higher olefins, their activity and is

Full text of DOI:10.1021/ja0647654

Published on Web 09/15/2006
Diastereomerically-Specific Zirconium Complexes of Chiral Salan Ligands:
Isospecific Polymerization of 1-Hexene and 4-Methyl-1-pentene and
Cyclopolymerization of 1,5-Hexadiene
Adi Yeori,† Israel Goldberg,† Michael Shuster,‡ and Moshe Kol*,†
School of Chemistry, Raymond and BeVerly Sackler Faculty of Exact Sciences, Tel AViV UniVersity, Ramat AViV,
Tel AViV 69978, Israel, and Research and DeVelopment Group, Carmel Olefins, Ltd., P.O.B. 1468,
Haifa 31014, Israel
Received July 5, 2006; E-mail: moshekol@post.tau.ac.il
Scheme 1. The Possible fac-fac Wrapping Modes of Chiral Salan
Ligands around Octahedral Group 4 Metal Centers
Chirality is an essential requirement of isospecific polymerization
catalysts of R-olefins that operate by an enantiomorphic-site control
mechanism. While conventional polymerizations employ racemic
catalysts, there are several notable exceptions in which enantio-
merically pure catalysts are required. These include the chiral
cyclopolymerization of 1,5-hexadiene,1 chiral oligomerization,2 and
kinetic resolution via polymerization.3 Such catalysts may be
prepared by resolution of the racemic complexes, as is achieved in
the case of ansa-metallocenes,4 or, preferably, by employing an
enatiomerically pure ligand that would wrap in a well-defined way
around the metal. Herein we describe the first highly diastereo-
selective formation of zirconium complexes of chiral Salan ligands
and their application in polymerization of R-olefins leading to novel
materials.
Scheme 2. Ligand Precursors Employed in This Work and
Synthesis of Their Corresponding Dibenzylzirconium Complexes
Recently, we introduced the family of Salan complexes of group
4 metals as R-olefin polymerization catalysts.5 The dianionic
tetradentate [ONNO]-type Salan ligands tend to wrap around an
octahedral metal center in a fac-fac mode, so that the overall
symmetry of the [ONNO]-MX2 complex is C2-chiral and the two
labile X groups are cis-related, an essential requirement for polym-
erization catalysis. While achiral Salan ligands lead to racemic
mixtures, introduction of chirality to the Salan backbone may lead
to two C2-diastereomers of fac-fac geometry. We have shown that
chiral Salan ligands assembled around a trans-1,2-diaminocyclo-
hexane bridge lead to titanium complexes with varying degrees of
diastereoselectivity: several unsubstituted Salan ligands (bearing
N-H groups) led to single diastereomers, whereas those bearing
N-Me groups led to diastereomer mixtures (Scheme 1).6 Unfortu-
nately, unsubstituted Salan ligands are not inert enough for
polymerization catalysis. We therefore turned to investigate the
wrapping of chiral N-Me substituted Salan ligands around zirco-
nium.
The ligand precursors described in this work were synthesized
either in an enantiomerically pure form (Lig1-4H2) or as racemates
(Lig1-5H2) following two possible synthetic schemes.7 Lig1,2H2
featuring electron-withdrawing substituents and Lig4H2 featuring
the bulky tert-Bu groups were described previously. Lig3H2
featuring nonbulky ortho-Me groups, and Lig5H2 featuring the very
bulky ortho-adamantyl groups8 are described herein for the first
time. The ligand precursors were reacted with tetrabenzylzirconium
in a toluene elimination reaction to produce the corresponding di-
benzyl complexes Lig1-5Zr(Bn)2 in typical yields of 90% (Scheme
2).
1H NMR spectra. The number of peaks observed for all complexes
supported the formation of a C2-symmetric diastereomer. X-ray
quality crystals of Lig3Zr(Bn)2 synthesized from rac-Lig3H2 were
obtained from toluene, and an ORTEP representation is shown in
Figure 1. The noncentrosymmetric space-group indicates that a
conglomerate rather than a racemate crystallization had occurred.6
The crystal structure of rac-Lig5Zr(Bn)2 that had crystallized in a
centrosymmetric space group was of lower quality owing to
disordered solvent molecules. Both structures indicated that the fac-
fac isomers have formed and, most notably, that the wrapping of
the ligands in both structures followed the same pattern; that is, an
(R,R)-Salan ligand leads to a ∆-binding. This is the same wrapping
preference that was observed for the titanium complexes of the
nonsubstituted ligands. Furthermore, an X-ray structure of the di-
(tert-butoxide) zirconium complex rac-Lig3Zr(O-tert-Bu)2 that had
also formed as a single diastereomer indicated an identical wrapping
mode.7 A possible explanation for the considerably higher diaste-
reoselectivity found for zirconium in comparison to titanium is the
higher covalent radius of zirconium enabling a less hindered fitting
of the N-Me substituted ligands.
All the dibenzylzirconium complexes were found to be active
catalysts for 1-hexene polymerization at room temperature upon
activation with tris(pentafluorophenyl)borane (eq 1). Their activities
were found to follow those of the analogous catalysts derived from
the nonchiral ligands: The three catalysts featuring nonbulky groups
(Lig1-3Zr(Bn)2) led to atactic polymerization, whereas Lig4,5Zr-
In a sharp contrast to the titanium coordination chemistry that
gave diastereomer mixtures for the N-Me ligands, all the zirconium
complexes formed as single diastereomers, as evident from their
† Raymond and Beverly Sackler Faculty of Exact Sciences.
‡ Carmel Olefins, Ltd.
9
13062
J. AM. CHEM. SOC. 2006, 128, 13062-13063
10.1021/ja0647654 CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
thermal effects associated with the helix racemization on melting-
recrystallization were observed. Cyclopolymerization of 1,5-hexa-
diene may lead to main-chain chiral poly(methylene-1,3-cyclopen-
tane) if the stereocontrol is trans-diisotactic.1 All the complexes
Lig1-5Zr(Bn)2 led to active catalysts for 1,5-hexadiene polymeri-
zation, their activity following the trend observed above, that is,
highly active catalysis by the complexes featuring electron with-
drawing phenolate substituents, low activity by the catalysts bearing
bulky groups, and intermediate activity by Lig3Zr(Bn)2. While most
samples were highly insoluble, 13C NMR spectra at high temperature
of a more soluble sample (prepared from (R,R)-(∆)-Lig2Zr(Bn)2)
featured only cyclopolymerization peaks, and no vinyl groups peaks
or peaks attributed to cross-linking were observed. Peak integration
aided by spectrum simulation of the C4,5 peaks region of the 13C
NMR spectrum supported a preference for trans substituted rings
(1 - σ ) 0.75), and a mild isotactic preference (R ) 0.75).1 The
combination of these two properties should render this polymer
chiral; however, its insolubility prevents a direct optical rotation
assessment. Polymers prepared from Lig2Zr(Bn)2 and Lig3Zr(Bn)2
show clear melting peaks at ca. 100 °C and glass transition at ca.
+5 °C (second heating thermograms), indicating a semicrystalline
structure. In contrast, the polymers derived from Lig4Zr(Bn)2 and
Lig5Zr(Bn)2 showed strong glass transitions at ca. -4 °C,7 and no
melting, indicating lack of crystallinity, which, combined with their
insolubility, may suggest a higher degree of cross-linking.
Figure 1. An ORTEP representation of rac-Lig3Zr(Bn)2, 50% probability.
(Bn)2 led to highly isotactic poly(1-hexene).9 Lig5Zr(Bn)2 is
somewhat more active than Lig4Zr(Bn)2, and the 13C NMR spectrum
of its polymer shows no traces of any stereo-errors (e.g., mmmr),
signifying a very high degree of isotacticity.7,5,10 These results are
consistent with those obtained with the analogous zirconium
complex of the nonchiral ligand in propylene polymerization.8 As
expected, the two catalysts featuring the electron withdrawing
groups (Lig1,2Zr(Bn)2) exhibited very high activities. In contrast to
their analogues based on the nonchiral ligands,5b they led to high-
molecular-weight polymers rather than to oligomers. This is
tentatively attributed to their higher rigidity that may restrict
conformations that favor termination events.
In conclusion, we have introduced the first zirconium complexes
of chiral Salan ligands. They were found to form as single
diastereomers of predetermined chirality around the metal and led
to active polymerization catalysts of higher R-olefins whose
isotacticity induction depended on the ligands’ bulk. We are
currently investigating further polymerizations as well as other
enantioselective transformations with these types of catalysts.
Acknowledgment. We thank the Israel Science Foundation and
the United States-Israel Binational Science Foundation for financial
support. We thank Dr Limor Frish for helpful discussions.
Lig1-5Zr(Bn)2
1-hexene
8 poly(1-hexene)
(1)
B(C6F5)3, rt
Supporting Information Available: Details of the syntheses and
characterization of the complexes, polymerization procedures, and
polymer characterization; crystallographic data in text format for rac-
Lig3Zr(Bn)2, rac-Lig5Zr(Bn)2, and rac-Lig3Zr(O-tert-Bu)2. This material
is available free of charge via the Internet at http://pubs.acs.org.
These catalysts were also found suitable for polymerization of
the somewhat bulkier monomer, 4-methyl-1-pentene. This polym-
erization followed the trend observed for 1-hexene polymeriza-
tion: The two complexes featuring electron-withdrawing phenolate
substituents (Lig1,2Zr(Bn)2) gave rise to highly active catalysts, and
to high-molecular weight stereoirregular polymers. On the other
hand, the two complexes featuring the bulky phenolate substituents
(Lig4,5Zr(Bn)2) led to catalysts of low activity and to crystalline
polymers insoluble in common solvents at room temperature. 13C
NMR at an elevated temperature of these crystalline polymers
indicated the formation of highly isotactic poly(4-methyl-1-pentene).
Differential scanning calorimetry results for polymer samples
synthesized from (rac)-Lig4Zr(Bn)2, (R,R)-(∆)-Lig4Zr(Bn)2, and
(rac)-Lig5Zr(Bn)2 supported their crystalline structure as evident
from the presence of exothermal (crystallization) peaks on the
cooling thermograms and endothermal (melting) peaks on the
heating thermograms. The melting peak (second heating thermo-
grams) temperatures are at 218.0, 216.7, and 222.7 °C, respectively.7
The somewhat higher melting point of the polymer derived from
(rac)-Lig5Zr(Bn)2 may signify a higher isotacticity. It is noteworthy
that the first and the second heating thermograms of the polymer
samples derived from the enantiomerically pure and the racemic
varieties of Lig4Zr(Bn)2 are almost identical. Thus, although a
presumable enantiomorphic-site control mechanism is expected to
produce chains of single helicity from (R,R)-(∆)-Lig4Zr(Bn)2
(assuming no helix-racemization prior to precipitation), no expected
References
(1) (a) Coates, G. W.; Waymouth, R. M. J. Am. Chem. Soc. 1991, 113, 6270.
(b) Coates, G. W.; Waymouth, R. M. J. Am. Chem. Soc. 1993, 115, 91.
(c) de Ballesteros, O. R.; Venditto, V.; Auriemma, F.; Guerra, G.; Resconi,
L.; Waymouth, R.; Mogstad, A.-L. Macromolecules 1995, 28, 2383.
(2) (a) Kaminsky, W.; Ahlers, A.; Mo¨ller-Lindenhof, N. Angew. Chem., Int.
Ed. Engl. 1989, 28, 1216. (b) Pino, P.; Cioni, P.; Wei, J. J. Am. Chem.
Soc. 1987, 109, 6189.
(3) Baar, C. R.; Levy, C. J.; Min, E. Y.-J.; Henling, L. M.; Day, M. W.;
Bercaw, J. E. J. Am. Chem. Soc. 2004, 126, 8216.
(4) Wild, F. R. W. P.; Zsolnai, L.; Huttner, D. S.; Brintzinger, H. H. J.
Organomet. Chem. 1982, 232, 233.
(5) (a) Tshuva, E. Y.; Goldberg, I.; Kol, M. J. Am. Chem. Soc. 2000, 122,
10706. (b) Segal, S.; Goldberg, I.; Kol, M. Organometallics 2005, 24,
200.
(6) Yeori, A.; Groysman, S.; Goldberg, I.; Kol, M. Inorg. Chem. 2005, 44,
4466. See also: Balsells, J.; Carroll, P. J.; Walsh, P. J. Inorg. Chem. 2001,
40, 5568.
(7) See Supporting Information for further details.
(8) A zirconium complex of the nonchiral ligand introduced by Busico and
co-workers led to highly isotactic polypropene: Busico, V.; Cipullo, R.;
Friederichs, N.; Ronca, S.; Talarico, G.; Togrou, M.; Wang, B. Macro-
molecules 2004, 37, 8201.
(9) Under the conditions employed, these catalysts did not lead to living
polymerization.
(10) (a) van der Linden, A.; Schaverian, C. J.; Meijboom, N.; Ganter, C.; Orpen,
A. G. J. Am. Chem. Soc. 1995, 117, 3008. (b) Babu, G. N.; Newmark, R.
A.; Chien, J. C. W. Macromolecules 1994, 27, 3383.
JA0647654
9
J. AM. CHEM. SOC. VOL. 128, NO. 40, 2006 13063