Syndiospecific living propylene polymerization catalyzed by titanium complexes having fluorine-containing phenoxy-imine chelate ligands

Article abstract of DOI:10.1021/ja029560j

The propylene polymerization behavior of a series of Ti complexes featuring fluorine-containing phenoxy-imine chelate ligands is reported. The Ti complexes combined with methylalumoxane (MAO) can be catalysts for living and, at the same time, stereospecific polymerization of propylene at room temperature or above. DFT calculations suggest that the attractive interaction between a fluorine ortho to the imine nitrogen and a β-hydrogen of a growing polymer chain is responsible for the achievement of room-temperature living propylene polymerization. Although the Ti complexes possess C₂ symmetry, they are capable of producing highly syndiotactic polypropylenes. ¹³C NMR is used to demonstrate that the syndiotacticity is governed by a chain-end control mechanism and that the polymerization is initiated exclusively via 1,2-insertion followed by 2,1-insertion as the principal mode of polymerization. ¹³C NMR spectroscopy also elucidated that the polypropylenes produced with the Ti complexes possess regio-block structures. Substitutions on the phenoxy-imine ligands have profound effects on catalytic behavior of the Ti complexes. The steric bulk of the substituent ortho to the phenoxy oxygen plays a decisive role in achieving high syndioselectivity for the chain-end controlled polymerization. Over a temperature range of 0-50 °C, Ti complex having a trimethylsilyl group ortho to the phenoxy oxygen forms highly syndiotactic, nearly monodisperse polypropylenes (94-90% rr) with extremely high peak melting temperatures (T_m = 156- 149 °C). The polymerization behavior of the Ti complexes can be explained well by the recently proposed site-inversion mechanism for the formation of syndiotactic polypropylene by a Ti complex having a pair of fluorine-containing phenoxy-imine ligands.

Full text of DOI:10.1021/ja029560j

Published on Web 03/18/2003
Syndiospecific Living Propylene Polymerization Catalyzed by
Titanium Complexes Having Fluorine-Containing
Phenoxy-Imine Chelate Ligands
Makoto Mitani, Rieko Furuyama, Jun-ichi Mohri, Junji Saito, Seiichi Ishii,
Hiroshi Terao, Takashi Nakano, Hidetsugu Tanaka, and Terunori Fujita*
Contribution from the R & D Center, Mitsui Chemicals, Inc., 580-32 Nagaura, Sodegaura,
Chiba, 299-0265, Japan
Received December 3, 2002; E-mail: Terunori.Fujita@mitsui-chem.co.jp
Abstract: The propylene polymerization behavior of a series of Ti complexes featuring fluorine-containing
phenoxy-imine chelate ligands is reported. The Ti complexes combined with methylalumoxane (MAO)
can be catalysts for living and, at the same time, stereospecific polymerization of propylene at room
temperature or above. DFT calculations suggest that the attractive interaction between a fluorine ortho to
the imine nitrogen and a â-hydrogen of a growing polymer chain is responsible for the achievement of
room-temperature living propylene polymerization. Although the Ti complexes possess C2 symmetry, they
are capable of producing highly syndiotactic polypropylenes. 13C NMR is used to demonstrate that the
syndiotacticity is governed by a chain-end control mechanism and that the polymerization is initiated
exclusively via 1,2-insertion followed by 2,1-insertion as the principal mode of polymerization. 13C NMR
spectroscopy also elucidated that the polypropylenes produced with the Ti complexes possess regio-block
structures. Substitutions on the phenoxy-imine ligands have profound effects on catalytic behavior of the
Ti complexes. The steric bulk of the substituent ortho to the phenoxy oxygen plays a decisive role in achieving
high syndioselectivity for the chain-end controlled polymerization. Over a temperature range of 0-50 °C,
Ti complex having a trimethylsilyl group ortho to the phenoxy oxygen forms highly syndiotactic, nearly
monodisperse polypropylenes (94-90% rr) with extremely high peak melting temperatures (Tm ) 156-
149 °C). The polymerization behavior of the Ti complexes can be explained well by the recently proposed
site-inversion mechanism for the formation of syndiotactic polypropylene by a Ti complex having a pair of
fluorine-containing phenoxy-imine ligands.
Introduction
performance catalysts for the living polymerization of ethylene,
propylene, 1-hexene, and others at relatively high temperatures.2-15
Living olefin polymerization enables consecutive enchainment
of monomer units without termination. Therefore, it can be used
for the preparation of precisely controlled polymers such as
monodisperse polymers, end-functionalized polymers, and block
copolymers, all of which are expected to display new or
enhanced properties and thus expanded utility. However, living
olefin polymerization is normally conducted at very low to
subambient temperatures to reduce the rates of chain termination
and transfer reactions, and accordingly, low activities and
relatively low molecular weights are typically observed. In
addition, it is generally difficult to polymerize R-olefins in a
living fashion with control of polymer stereochemistry.
For instance, Brookhart et al. developed Ni and Pd complexes
with diimine ligands that initiate living polymerization of
(2) For a recent review, see: Coates, G. W.; Hustad, P. D.; Reinartz, S. Angew.
Chem., Int. Ed. 2002, 41, 2236-2257.
(3) (a) Brookhart, M.; DeSimone, J. M.; Grant, B. E.; Tanner, M. J.
Macromolecules 1995, 28, 5378-5380. (b) Gottfried, A. C.; Brookhart,
M. Macromolecules 2001, 34, 1140-1142.
(4) (a) Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc. 1995,
117, 6414-6415. (b) Killian, C. M.; Tempel, D. J.; Johnson, L. K.;
Brookhart, M. J. Am. Chem. Soc. 1996, 118, 11664-11665.
(5) (a) Scollard, J. D.; McConville, D. H. J. Am. Chem. Soc. 1996, 118, 10008-
10009. (b) Scollard, J. D.; McConville, D. H.; Vittal, J. J.; Payne, N. C. J.
Mol. Catal. A 1998, 128, 201-214.
(6) (a) Baumann, R.; Davis, W. M.; Schrock, R. R. J. Am. Chem. Soc. 1997,
119, 3830-3831. (b) Baumann, R.; Schrock, R. R. J. Organomet Chem.
1998, 557, 69-75. (c) Liang, L. C.; Schrock, R. R.; Davis, W. M.;
McConville, D. H. J. Am. Chem. Soc. 1999, 121, 5797-5798. (d) Baumann,
R.; Stumpf, R.; Davis, W. M.; Liang, L. C.; Schrock, R. R. J. Am. Chem.
Soc. 1999, 121, 7822-7836. (e) Schrock, R. R.; Baumann, R.; Reid, S.
M.; Goodman, J. T.; Stumpf, R.; Davis, W. M. Organometallics 1999, 18,
3649-3670. (f) Schrock, R. R.; Bonitatebus, P. J.; Schrodi, Y. Organo-
metallics 2001, 20, 1056-1058. (g) Schrodi, Y.; Schrock, R. R.; Boni-
tatebus, P. J. Organometallics 2001, 20, 3560-3573. (h) Goodman, J. T.;
Schrock, R. R. Organometallics 2001, 20, 5205-5211. (i) Mehrkhodavandi,
P.; Bonitatebus, P. J.; Schrock, R. R. J. Am. Chem. Soc. 2000, 122, 7841-
7842. (j) Mehrkhodavandi, P.; Schrock, R. R. J. Am. Chem. Soc. 2001,
123, 10746-10747.
Enormous advances in the rational design and synthesis of
well-defined transition-metal complexes for olefin polymeriza-
tion1 have spurred the development of quite a few high-
(1) (a) Kaminsky, W.; Ku¨lper, K.; Brintzinger, H. H.; Wild, F. R. W. P. Angew.
Chem., Int. Ed. Engl. 1985, 97, 507-508. (b) Kaminsky, W. Catal. Today
1994, 20, 257-271. (c) Mo¨hring, P. C.; Coville, N. J. J. Organomet. Chem.
1994, 479, 1-29. (d) Kaminsky, W.; Arndt, M. AdV. Polym. Sci. 1997,
127, 144-187. (e) Britovsek, G. J. P.; Gibson, V. C.; Wass, D. F. Angew.
Chem., Int. Ed. 1999, 38, 428-447. (f) Ittel, S. D.; Johnson, L. K.;
Brookhart, M. Chem. ReV. 2000, 100, 1169-1203. (g) Gibson, V. C.;
Spitzmesser, S. K. Chem. ReV. 2003, 103, 283-315.
(7) Jeon, Y. M.; Park, S. J.; Heo, J.; Kim, K. Organometallics 1998, 17, 3161-
3163.
9
10.1021/ja029560j CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 4293-4305
4293

A R T I C L E S
Mitani et al.
ethylene, propylene, 1-hexene, etc. and produce R-olefin-based
block copolymers.3,4 Alternatively, McConville,5 Schrock,6 and
Kim7 described Ti and Zr complexes featuring diamide-based
ligands that form nearly monodisperse atactic poly(1-hexene)s
at ambient temperatures. Additionally, Sita et al. reported on
Zr complexes containing amidinate donors that are capable of
combining living enchainment of 1-hexene and vinylcyclohex-
ane with the control of polymer stereochemistry to create
isotactic poly(1-hexene)s and poly(vinylcyclohexane)s.8 More-
over, Kol and Goldschmidt described Zr complexes containing
phenoxy-amine ligands that generate isotactic poly(1-hexene)s
with narrow polydispersity.9 Very recently, Shiono et al. reported
that [tBuNSiMe2Flu]TiMe2 combined with dried methylalu-
moxane (MAO) produced prevailingly syndiotactic polypropy-
lene (sPP; rr 63%).10d There are, however, a limited number of
examples of catalysts that polymerize propylene in a living
fashion at elevated temperatures such as above room temper-
ature. Moreover, highly stereospecific living polymerization of
propylene has not yet been achieved, despite the fact that nearly
perfect stereospecific nonliving propylene polymerization has
been performed with appropriately designed group 4 metallocene
catalysts.16 Consequently, there are crucial goals still remaining
in the field of the living polymerization of propylene.
R-olefins and dienes. For example, Zr-FI Catalysts are capable
of producing vinyl-terminated low molecular weight polyeth-
ylenes (PEs)20j and ultrahigh molecular weight ethylene/propy-
lene copolymers with high efficiency.20k In addition, Ti-FI
Catalysts favor 2,1-insertion of 1-hexene and form high mo-
lecular weight atactic poly(1-hexene)s with ca. 50 mol %
regioirregular units.20d Surprisingly, fluorinated Ti-FI Catalysts
are capable of mediating highly controlled living ethylene
polymerization at a high temperature of 50 °C by the attractive
interaction between the ligand and a growing polymer chain.21
Furthermore, we discovered that these fluorinated Ti-FI Catalysts
can be catalysts for the living and, at the same time, highly
syndiospecific polymerization of propylene, as introduced in
our patent filed in January 2000 and a series of subsequent
papers.19b,22 Thus, fluorinated Ti-FI Catalysts are the first
examples of olefin polymerization catalysts capable of initiating
the living polymerization of both ethylene and propylene,
allowing the synthesis of a variety of block copolymers such
as PE-b-poly(ethylene-co-propylene), PE-b-poly(ethylene-co-
propylene)-b-PE, sPP-b-poly(ethylene-co-propylene), PE-b-sPP,
and PE-b-poly(ethylene-co-propylene)-b-sPP. Many of these
block copolymers are previously unobtainable from conventional
(18) (a) Makio, H.; Kashiwa, N.; Fujita, T. AdV. Synth. Catal. 2002, 344, 477-
493. (b) Yoshida, Y.; Matsui, S.; Takagi, Y.; Mitani, M.; Nitabaru, M.;
Nakano, T.; Tanaka, H.; Fujita, T. Chem. Lett. 2000, 1270-1271. (c)
Yoshida, Y.; Matsui, S.; Takagi, Y.; Mitani, M.; Nakano, T.; Tanaka, H.;
Kashiwa, N.; Fujita, T. Organometallics 2001, 20, 4793-4799. (d) Yoshida,
Y.; Saito, J.; Mitani, M.; Takagi, Y.; Matsui, Ishii, S.; S.; Nakano, T.;
Kashiwa, N.; Fujita, T. Chem. Commun. 2002, 1298-1299. (e) Inoue, Y.;
Nakano, T.; Tanaka, H.; Kashiwa, N.; Fujita, T. Chem. Lett. 2001, 1060-
1061. (f) Suzuki, Y.; Kashiwa, N.; Fujita, T. Chem. Lett. 2002, 358-359.
(19) (a) Fujita, T.; Tohi, Y.; Mitani, M.; Matsui, S.; Saito, J.; Nitabaru, M.;
Sugi, K.; Makio, H.; Tsutsui, T. Europe Patent, EP-0874005, 1998 (filing
date, Apr 25, 1997) (Chem. Abstr. 1998, 129, 331166). (b) Mitani, M.;
Yoshida, Y.; Mohri, J.; Tsuru, K.; Ishii, S.; Kojoh, S.; Matsugi, T.; Saito,
J.; Matsukawa, N.; Matsui, S.; Nakano, T.; Tanaka, H.; Kashiwa, N.; Fujita,
T. WO Pat. 01/55231 A1, 2001 (filing date, Jan 26, 2000) (Chem. Abstr.
2001, 135, 137852).
(20) (a) Matsui, S.; Tohi, Y.; Mitani, M.; Saito, J.; Makio, H.; Tanaka, H.;
Nitabaru, M.; Nakano, T.; Fujita, T. Chem. Lett. 1999, 1065-1066. (b)
Matsui, S.; Mitani, M.; Saito, J.; Tohi, Y.; Makio, H.; Tanaka, H.; Fujita,
T. Chem. Lett. 1999, 1263-1264. (c) Matsui, S.; Mitani, M.; Saito, J.;
Matsukawa, N.; Tanaka, H.; Nakano, T.; Fujita, T. Chem. Lett. 2000, 554-
555. (d) Saito, J.; Mitani, M.; Matsui, S.; Kashiwa, N.; Fujita, T. Macromol.
Rapid Commun. 2000, 21, 1333-1336. (e) Matsui, S.; Fujita, T. Catal.
Today 2001, 66, 63-73. (f) Matsukawa, N.; Matsui, S.; Mitani, M.; Saito,
J.; Tsuru, K.; Kashiwa, N.; Fujita, T. J. Mol. Catal. A 2001, 169, 99-104.
(g) Matsui, S.; Mitani, M.; Saito, J.; Tohi, Y.; Makio, H.; Matsukawa, N.;
Takagi, Y.; Tsuru, K.; Nitabaru, M.; Nakano, T.; Tanaka, H.; Kashiwa,
N.; Fujita, T. J. Am. Chem. Soc. 2001, 123, 6847-6856. (h) Saito, J.; Mitani,
M.; Matsui, S.; Tohi, Y.; Makio, H.; Nakano, T.; Tanaka, H.; Kashiwa,
N.; Fujita, T. Macromol. Chem. Phys. 2002, 203, 59-65. (i) Ishii, S.; Saito,
J.; Mitani, M.; Mohri, J.; Matsukawa, N.; Tohi, Y.; Matsui, S.; Kashiwa,
N.; Fujita, T. J. Mol. Catal. A 2002, 179, 11-16. (j) Ishii, S.; Mitani, M.;
Saito, J.; Matsuura, S.; Kojoh, S.; Kashiwa, N.; Fujita, T. Chem. Lett. 2002,
740-741. (k) Ishii, S.; Saito, J.; Matsuura, S.; Suzuki, Y.; Furuyama, R.;
Mitani, M.; Nakano, T.; Kashiwa, N.; Fujita, T. Macromol. Rapid Commun.
2002, 23, 693-697. (l) Saito, J.; Onda, M.; Matsui, S.; Mitani, M.;
Furuyama, R.; Tanaka, H.; Fujita, T. Macromol. Rapid Commun. 2002,
23, 1118-1123. (m) Nakayama, Y.; Bando, H.; Sonobe, Y.; Kaneko, H.;
Kashiwa, N.; Fujita, T. J. Catal. In press. (n) Tohi, Y.; Matsui, S.; Makio,
H.; Onda, M.; Fujita, T. Macromolecules 2003, 36, 523-525. (o) Furuyama,
R.; Saito, J.; Ishii, S.; Mitani, M.; Matsui, S.; Tohi, Y.; Makio, H.;
Matsukawa, N.; Tanaka, H.; Fujita, T. J. Mol. Catal. A, in press. (p) Saito,
J.; Suzuki, Y.; Fujita, T. Chem. Lett. 2003, 32, 236-237.
Previously, based on ligand-oriented catalyst design research
focusing on nonsymmetric [O, N], [N, N], and [O, O] chelate
ligands,13,17,18 we found a new family of group 4 transition-
metal complexes featuring phenoxy-imine chelate ligands,
named FI Catalysts.19,20 FI Catalysts on activation with MAO
or iBu3Al/Ph3CB(C6F5)4 exhibit unique catalytic properties for
polymerization of ethylene and/or R-olefins, including higher
(8) (a) Jayaratne, K. C.; Sita, L. R. J. Am. Chem. Soc. 2000, 122, 958-959.
(b) Jayaratne, K. C.; Keaton, R. J.; Henningsen, D. A.; Sita, L. R. J. Am.
Chem. Soc. 2000, 122, 10490-10491. (c) Keaton, R. J.; Jayaratne, K. C.;
Henningsen, D. A.; Koterwas, L. A.; Sita, L. R. J. Am. Chem. Soc. 2001,
123, 6197-6198. (d) Jayaratne, K. C.; Sita, L. R. J. Am. Chem. Soc. 2001,
123, 10754-10755.
(9) (a) Tshuva, E. Y.; Goldberg, I.; Kol, M.; Goldschmidt, Z. Inorg. Chem.
Commun. 2000, 3, 611-614. (b) Tshuva, E. Y.; Goldberg, I.; Kol, M. J.
Am. Chem. Soc. 2000, 122, 10706-10707. (c) Tshuva, E. Y.; Goldberg,
I.; Kol, M.; Goldschmidt, Z. Chem. Commun. 2001, 2120-2121.
(10) (a) Hagihara, H.; Shiono, T.; Ikeda, T. Macromolecules 1998, 31, 3184-
3188. (b) Hasan, T.; Ioku, A.; Nishii, K.; Shiono, T.; Ikeda, T. Macro-
molecules 2001, 34, 3142-3145. (c) Hagimoto, H.; Shiono, T.; Ikeda, T.
Macromol. Rapid Commun. 2002, 23, 73-76. (d) Hagimoto, H.; Shiono,
T.; Ikeda, T. Macromolecules 2002, 35, 5744-5745.
(11) (a) Fukui, Y.; Murata, M.; Soga, K. Macromol. Rapid Commun. 1999, 20,
637-640. (b) Fukui, Y.; Murata, M. Appl. Catal., A 2002, 237, 1-10.
(12) (a) Yasuda, H.; Furo, M.; Yamamoto, H.; Nakamura, A.; Miyake, S.;
Kibino, N. Macromolecules 1992, 25, 5115-5116. (b) Mashima, K.;
Fujikawa, S.; Nakamura, A. J. Am. Chem. Soc. 1993, 115, 10990-10991.
(c) Mashima, K.; Fujikawa, S.; Urata, H.; Tanaka, E.; Nakamura, A. J.
Chem. Soc., Chem. Commun. 1994, 1623-1624. (d) Mashima, K.;
Fujikawa, S.; Tanaka, Y.; Urata, H.; Oshiki, T.; Tanaka, E.; Nakamura, A.
Organometallics 1995, 14, 2633-2640.
(13) (a) Matsugi, T.; Matsui, S.; Kojoh, S.; Takagi, Y.; Inoue, Y.; Fujita, T.;
Kashiwa, N. Chem. Lett. 2001, 566-567. (b) Matsugi, T.; Matsui, S.; Kojoh,
S.; Takagi, Y.; Inoue, Y.; Fujita, T.; Kashiwa, N. Macromolecules 2002,
35, 4880-4887.
(14) (a) Doi, Y.; Keii, T. AdV. Polym. Sci. 1986, 73/74, 201-248. (b) Doi, Y.;
Tokuhiro, N.; Soga, K. Makromol. Chem., Macromol. Chem. Phys. 1989,
190, 643-651 and references therein.
(15) Turner, H. W.; Hlatky, G. G. Exxon, PCT International Application
9112285, 1991 (Chem. Abstr. 1992, 116, 61330t).
(21) (a) Saito, J.; Mitani, M.; Mohri, J.; Yoshida, Y.; Matsui, S.; Ishii, S.; Kojoh,
S.; Kashiwa, N.; Fujita, T. Angew. Chem., Int. Ed. 2001, 40, 2918-2920.
(b) Mitani, M.; Mohri, J.; Yoshida, Y.; Saito, J.; Ishii, S.; Tsuru, K.; Matsui,
S.; Furuyama, R.; Nakano, T.; Tanaka, H.; Kojoh, S.; Matsugi, T.; Kashiwa,
N.; Fujita, T. J. Am. Chem. Soc. 2002, 124, 3327-3336. (c) Mitani, M.;
Mohri, J.; Furuyama, R.; Ishii S.; Fujita, T. Chem. Lett. 2003, 32, 238-
239. (d) Mitani, M.; Nakano, T.; Fujita, T. Chem. Eur. J. In press.
(22) (a) Saito, J.; Mitani, M.; Mohri, J.; Ishii, S.; Yoshida, Y.; Matsugi, T.;
Kojoh, S.; Kashiwa, N.; Fujita, T. Chem. Lett. 2001, 576-577. (b) Kojoh,
S.; Matsugi, T.; Saito, J.; Mitani, M.; Fujita, T.; Kashiwa, N. Chem. Lett.
2001, 822-823. (c) Saito, J.; Mitani, M.; Onda, M.; Mohri, J.; Ishii, S.;
Yoshida, Y.; Nakano, T.; Tanaka, H.; Matsugi, T.; Kojoh, S.; Kashiwa,
N.; Fujita, T. Macromol. Rapid Commun. 2001, 22, 1072-1075. (d) Mitani,
M.; Furuyama, R.; Mohri, J.; Saito, J.; Ishii, S.; Terao, H.; Kashiwa, N.;
Fujita, T. J. Am. Chem. Soc. 2002, 124, 7888-7889.
(16) (a) Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt, R.; Rieger, B.; Waymouth,
R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143-1170. (b) Coates, G.
W. Chem. ReV. 2000, 100, 1223-1252. (c) Coates, G. W. J. Chem. Soc.,
Dalton Trans. 2002, 467-475.
(17) Pioneering research in the field of group 4 transition-metal complexes with
nonsymmetric ligands. For example: (a) Cozzi, P. G.; Gallo, E.; Floriani,
C.; Chiesi-Villa, A.; Rizzoli, C. Organometallics 1995, 14, 4994-4996.
(b) Tjaden, E. B.; Swenson, D. C.; Jordan, R. F. Organometallics 1995,
14, 371-386. (c) Bei, X.; Swenson, D. C.; Jordan, R. F. Organometallics
1997, 16, 3282-3302. (d) Tsukahara, T.; Swenson, D. C.; Jordan, R. F.
Organometallics 1997, 16, 3303-3313. (e) Kim, I.; Nishihara, Y.; Jordan,
R. F.; Rogers, R. D.; Rheingold, A. L.; Yap, G. P. A. Organometallics
1997, 16, 3314-3323.
9
4294 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Syndiospecific Living Propylene Polymerization
A R T I C L E S
Ziegler-Natta catalysis. Recently, Coates et al. have worked
on Ti-FI Catalysts, and based on combinatorial methods, they
have independently obtained a modified Ti-FI Catalyst that
forms highly syndiotactic polypropylene with narrow MWD and
makes sPP-b-poly(ethylene-co-propylene) and sPP-b-poly(m-
ethylenecyclopentane-co-vinyl tetramethylene).23 Therefore, FI
catalysts and related early-transition-metal complexes have been
the subject of active investigation due to their high potential as
olefin polymerization catalysts that produce unique polymers.24
In this report, we describe the detailed catalytic behavior of
Ti complexes with fluorine-containing phenoxy-imine chelate
ligands (fluorinated Ti-FI Catalysts) for the polymerization of
propylene. These complexes are capable of promoting chain-
end controlled, thermally robust, living propylene polymerization
via a 2,1-insertion and produce highly syndiotactic nearly
monodisperse polypropylenes with very high peak melting
temperatures.
Figure 1. Molecular structure of complex 4 with thermal ellipsoids at 50%
probability level.
Results and Discussion
Syntheses and X-ray Analyses of Titanium Complexes
Having Fluorine-Containing Phenoxy-Imine Chelate Ligands.
A general preparation route to titanium complexes utilized in
this study is shown in Scheme 1. The fluorine-containing
phenoxy-imine ligands of structures a-h were conveniently
synthesized in high yields by the Schiff base condensation of
the appropriate aniline and salicylaldehyde derivatives. The
titanium complexes 1-8 were obtained in moderate to good
yields by the reaction of TiCl4 with 2 equiv of the lithium salt
of the corresponding phenoxy-imine ligands.
Scheme 1. Synthetic Procedure for Complexes 1-8
Figure 2. Molecular structure of complex 6 with thermal ellipsoids at 40%
probability level.
Table 1. Selected Bond Distances (Å) and Angles (deg) for
Complexes 1, 4, and 6
1
4
6
Distances
Ti-Cl(1)
Ti-Cl(2)
Ti-O(1)
2.2876(7)
2.2578(8)
1.841(1)
1.845(1)
2.234(2)
2.217(2)
2.245(1)
2.284(1)
1.860(2)
1.854(2)
2.268(3)
2.253(3)
1.893(4)
1.890(4)
2.253(2)
2.297(2)
1.840(3)
1.859(3)
2.265(4)
2.213(4)
Ti-O(2)
Complexes 4 and 6 were analyzed by single-crystal X-ray
diffraction. The molecular structures are depicted in Figures 1
and 2; selected bond distances and angles are presented in Table
1, which also includes those for complex 1 as a comparison.
Both complexes 4 and 6 possess six-coordinate Ti metal centers
in an approximately octahedral structure with a trans-O, cis-N,
Ti-N(1)
Ti-N(2)
Si(1)-C(2)
Si(2)-C(18)
Angles
Cl(1)-Ti-Cl(2)
O(1)-Ti-O(2)
N(1)-Ti-N(2)
96.42(3)
163.61(6)
86.94(7)
99.19(4)
165.6(1)
85.30(10)
96.49(7)
162.1(2)
88.0(1)
(23) (a) Tian, J.; Coates, G. W. Angew. Chem., Int. Ed. 2000, 39, 3626-3629.
(b) Hustad, P. D.; Tian, J.; Coates, G. W. J. Am. Chem. Soc. 2001, 123,
5134-5135. (c) Tian, J.; Hustad, P. D.; Coates, G. W. J. Am. Chem. Soc.
2002, 124, 3614-3621. (d) Hustad, P. D.; Coates, G. W. J. Am. Chem.
Soc. 2002, 124, 11578-11579. (e) Fujita, M.; Coates, G. W. Macromol-
ecules 2002, 35, 9640-9647.
(24) For example: (a) Corden, J. P.; Errington, W.; Moore, P.; Wallbridge, M.
G. H. Chem. Commun. 1999, 323-324. (b) Woodman, P. R.; Munslow, I.
J.; Hitchcock, P. B.; Scott, P. K. J. Chem. Soc., Dalton Trans. 1999, 4069-
4076. (c) Strauch, J.; Warren, T. H.; Erker, G.; Frohlich, R.; Saarenketo,
P. Inorg. Chim. Acta 2000, 300-302, 810-821. (d) Gibson, V. C.;
Mastroianni, S.; Newton, C.; Redshaw, C.; Solan, G. A.; White, A. J. P.;
Williams, D. J. J. Chem. Soc., Dalton Trans. 2000, 1969-1972. (e) Emslie,
D. J. H.; Piers, W. E.; MacDonald, R. J. Chem. Soc., Dalton Trans. 2002,
293-294. (f) Knight, P. D.; Clarke, A. J.; Kimberley, B. S.; Jackson, R.
A.; Scott, P. Chem. Commun. 2002, 352-353. (g) Jones, D. J.; Gibson, V.
C.; Green, S. M.; Maddox, P. J. Chem. Commun. 2002, 1038-1039.
and cis-Cl disposition, which is the same as that of complex 1.
Thus, complexes 4 and 6, in addition to complex 1, are revealed
to have structures with approximate C2 symmetry.
The bond distances of Ti-N (2.268, 2.253 Å) and Ti-O
(1.860, 1.854 Å) for complex 4 are slightly longer compared
with those for complex 1 (Ti-N, 2.234, 2.217 Å; Ti-O, 1.841,
1.845 Å). These facts suggest that the electron donation from
the phenoxy-imine ligand to the Ti center is reduced by the
introduction of a trimethylsilyl group in the place of the tert-
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4295

A R T I C L E S
Mitani et al.
Figure 3. Top views of X-ray structures for complexes 1, 4, and 6 and distances between chlorine atoms and nearest carbon atoms in the ligands. The
pentafluorophenyl groups are omitted for clarity.
Table 2. Results of Propylene Polymerization with Complex
butyl group, probably due to less electron-donating nature of
the trimethylsilyl group relative to the tert-butyl group.
To compare the steric congestion of complexes 1, 4, and 6
in close proximity to the chlorine-bound sites (potential po-
lymerization sites), the top views of the complexes and the
distances between a chlorine atom and its nearest carbon atom
in the ligand are shown in Figure 3. The distances for complex
6 (5.070, 5.155 Å) are much longer than those for complex 1
(4.027, 4.022 Å) and complex 4 (3.886, 3.981 Å). Therefore,
among the complexes complex 6 probably has the most open
structure of the catalyst active site, followed by complex 1 and
complex 4. These structural differences among complexes 1,
4, and 6 may influence their catalytic behavior for olefin
polymerization.
1/MAOa
b
d
P
(MPa)
T
(°C)
time
(h)
yield
(g)
TOF
(h-1
M c
Tm
n
entry
)
(×103)
M /M c
(°C)
w
n
1
2
3
4
5
6
7
8
0.1
0.1
0.1
0.1
0.6
0.6
0.6
0.6
25
25
0
50
25
25
25
25
3
5
5
5
1
2
3
5
0.096
0.183
0.144
0.148
0.158
0.312
0.460
0.713
76
87
68
20.5
28.5
23.6
16.4
30.9
52.8
73.8
108.0
1.09
1.11
1.05
1.37
1.07
1.10
1.10
1.14
137
137
136
130
135
70
375
371
364
339
135
135
a Conditions: complex 1 (10 µmol), cocat. MAO (2.5 mmol). b Turnover
frequency. c Determined by GPC using polypropylene calibration. d Melting
temperature of produced PP determined by DSC.
may also exhibit living behavior for propylene polymerization.
A single-crystal X-ray analysis revealed that complex 1 has a
six-coordinate center in a distorted octahedral geometry with a
trans-O, cis-N, and cis-Cl arrangement and thus possesses
approximate C2 symmetry. This molecular structure implies that
1 combines living enchainment with isospecific control of
polymer stereochemistry.
Propylene polymerization behavior of complex 1 was inves-
tigated with MAO cocatalyst in toluene solvent. The results are
collected in Table 2. The activities of complex 1 under 0.1 MPa
of propylene are moderate (TOF 68-87 h-1) and much lower
than those observed with ethylene polymerization. This may
be attributed to the stronger binding of propylene to the
electrophilic Ti center and the larger molecular size of propylene
compared to ethylene, impeding the rate of chain growth relative
to that of ethylene polymerization. GPC analyses revealed that
the polypropylenes produced under 0.1 MPa of propylene at
25 °C (Table 2, entries 1, 2) possess extremely narrow molecular
weight distributions (Mw/Mn 1.09, 1.11), suggesting that complex
1/MAO catalyst system may have the characteristics of a living
propylene polymerization under the given conditions.
Propylene Polymerization Behavior of Complex 1. As
reported, the complex 1/methylalumoxane (MAO) catalyst
system is capable of initiating highly controlled living ethylene
polymerization by the unique interaction of a fluorine ortho to
the imine nitrogen with a â-hydrogen of a growing polymer
chain and produces high molecular weight polyethylenes (Mn
> 400 000) with extremely narrow molecular weight distribu-
tions (Mw/Mn < 1.2). We reasoned that the interaction may be
extended to the living polymerization of propylene because a
growing polypropylene chain also possesses a â-hydrogen.
Indeed, DFT calculations indicate that the ortho-fluorine of an
active species generated from 1, for propylene polymerization,
interacts with a â-hydrogen of a growing polypropylene chain
(polymer chain models: sec-butyl and isobutyl) derived either
from 1,2-insertion or 2,1-insertion25 (1,2-insertion, ortho F-Hâ
distance 2.251 Å, electrostatic energy -46.8 kJ/mol; 2,1-
insertion, ortho F-Hâ distance 2.262 Å, electrostatic energy
-30.7 kJ/mol; for ethylene polymerization, ortho F-Hâ distance
2.276 Å, electrostatic energy -27.1 kJ/mol), suggesting that 1
(25) For detailed results, see the Supporting Information. Significantly, Chan
et al. have recently reported the first NMR and X-ray structural evidence
for an attractive interaction between a fluorine in the ligand and a hydrogen
on a benzyl group attached to the central metal for group 4 transition-
metal complexes, potentially indicating the generality of the attractive
interaction between the functionalized ligand and the polymer chain. Kui,
S. C. F.; Zhu, N.; Chan, M. C. W. Angew. Chem., Int. Ed. In press.
To gain further insight into propylene polymerization behavior
of the catalyst system, Mn and Mw/Mn values-polymerization
time relationships were plotted for the polymerization conducted
under 0.6 MPa of propylene at 25 °C (Table 2, entries 5-8).
9
4296 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Syndiospecific Living Propylene Polymerization
A R T I C L E S
Figure 6. 13C NMR spectra of the highly syndiotactic PP produced by 1.
pylenes formed from complex 1/MAO catalyst system exhibited
peak melting temperatures (Tm), indicating the formation of
stereoregular polymers.
Figure 4. Plots of Mn and Mw/Mn as a function of polymerization time for
propylene polymerization with complex 1/MAO (25 °C, 0.6 MPa).
Pentad Distribution Analysis and End-Group Analysis of
the Polypropylenes Arising from Complex 1/MAO Catalyst
System. The nearly monodisperse polypropylene having an Mn
value of 28 500 (Table 2, entry 2) displays a peak melting
temperature (Tm) of 137 °C, indicative of a high degree of
stereocontrol. To our surprise, microstructural analysis using
13C NMR spectroscopy revealed that the polymer is highly
syndiotactic polypropylene, which contains 87% rr triads. The
unexpected formation of syndiotactic polypropylene with com-
plex 1 having C2 symmetry indicates that the catalyst symmetry
is not a rigid requirement for determining polymer stereochem-
istry. The pentad distribution analysis of the 13C NMR of the
syndiotactic polypropylene (Figure 6) shows the presence of
single m stereodefects, rrrm (9.0%) and rrmr + mmrm (5.6%),
as the predominant source of stereoirregularity in the polymer
microstructure. These results suggest that the highly syndiospe-
cific polymerization proceeds via a chain-end control mecha-
nism. Complex 1 is unique in its ability to produce a highly
stereoregular polymer through a chain-end control mechanism
at room temperature since chain-end control operates well at
very low to subambient temperatures and rapidly loses its
stereoregulating ability at elevated temperatures.14,26 In fact, the
rr triad, 87%, obtained at 25 °C, is exceptionally high for a
polypropylene produced by a chain-end control mechanism. On
the basis of combinatorial methods, Coates et al. have recently
obtained a modified version of complex 1, which has an
additional tert-butyl group in the phenoxy-benzene ring23b
(complex 2). This complex exhibits a similar catalytic perfor-
mance compared to complex 1 for living propylene polymeri-
zation according to our data.22d
Figure 5. GPC profile of polypropylenes obtained by complex 1/MAO
(25 °C, 0.6 MPa): (a) 1 h, Mn 30 900, Mw/Mn 1.07; (b) 2 h, Mn 52 800,
Mw/Mn 1.10; (c) 3 h, Mn 73 800, Mw/Mn 1.10; (d) 5 h, Mn 108 000, Mw/Mn
1.14.
As shown in Figure 4, a linear relationship between Mn and
polymerization time as well as narrow Mw/Mn values for all
runs (Mw/Mn 1.07-1.14) was found. Moreover, GPC peaks of
the produced polypropylene shifted to higher molecular weight
region with an increase in polymerization time, retaining
unimodal shape during the course of the polymerization (Figure
5). These results clearly show that the propylene polymerization
is indeed living. Considering that MAO used as the cocatalyst
is a potential chain transfer agent, the room-temperature living
propylene polymerization exhibited by complex 1/MAO is of
great significance. Complex 1 is a unique olefin polymerization
catalyst capable of inducing the living polymerization of both
ethylene and propylene, resulting in the creation of ethylene-
based as well as propylene-based block copolymer.19b,21,22
With these polymerization results along with the computa-
tional studies, the attractive interaction between a fluorine in
the ligand and a â-hydrogen of a growing polymer chain is
thought to provide a new strategy for the design of living
polymerization catalysts. Catalyst efficiency for the polymeri-
zation calculated based on polymer yield, catalyst amount, and
the molecular weight of the resulting polymer lies in the range
of 51-66%. It is worth noting that at 50 °C the catalyst system
is capable of producing polypropylene with a fairly narrow
molecular weight distribution (Mw/Mn 1.37), as listed in Table
2, entry 4, though chain termination or transfer and/or catalyst
deactivation are not negligible at this temperature. The polypro-
Considering that chain-end control is enhanced by 2,1-
insertion due to the increased steric influence of the chain-end
on enchainment relative to 1,2-insertion27 and that the complex
1/iBu3Al/Ph3CB(C6F5)4 catalyst system is known to prefer 2,1-
insertion of 1-hexene, it seems reasonable that the high level
of stereoregularity in the polypropylene formed with complex
1 is the result of a 2,1-insertion process.
(26) (a) Pellecchia, C.; Zambelli, A.; Oliva, L.; Pappalardo, D. Macromolecules
1996, 29, 6990-6993. (b) Pellecchia, C.; Zambelli, A. Macromol. Rapid
Commun. 1996, 17, 333-338. (c) Pellecchia, C.; Zambelli, A.; Mazzeo,
M.; Pappalardo, D. J. Mol. Catal. A 1998, 128, 229-237. (d) Milano, G.;
Guerra, G.; Pellecchia, C.; Cavallo, L. Organometallics 2000, 19, 1343-
1349.
(27) Zambelli, A.; Allegra, G. Macromolecules 1980, 13, 42-49 and references
therein.
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4297

A R T I C L E S
Mitani et al.
Figure 7. Chain-end analysis of low molecular weight syndiotactic PP produced by 1.
To obtain information on mechanistic details, 13C NMR
analysis of low Mn syndiotactic polypropylene (Mn 2000, Mw/
Mn 1.08)28a was performed. Figure 7 shows the 13C NMR
spectrum of the low Mn polypropylene, which exhibits peaks
at 14.2, 20.6, 22.4-22.8, 23.3-23.6, 25.6, 30.2, 35.5, 36.5, and
40.4 ppm relating to chain-end structures in addition to three
major sets of peaks derived from the main chain carbons.
The following assignments are based on a comparison with
literature data.29 The peaks at 14.2 and 23.3-23.6 ppm were
assigned to the methyl carbons of n-propyl and isobutyl chain-
end groups, respectively. Additionally, the peaks at 22.4-22.8
ppm were attributed to the methyl carbons of isobutyl and
isopentyl chain-end groups. Regarding the CH and CH2 carbons
associated with chain-end groups, the peaks at 35.5 and 36.5
ppm and 25.6 ppm were assigned to the SRâ CH2 carbon of
isopentyl chain-end group and the CH carbon of isobutyl chain-
end group, respectively. These results further confirmed the
presence of isopentyl and isobutyl chain-end groups. Thus, the
polypropylene contains three kinds of chain-end groups: n-
propyl, isobutyl, and isopropyl groups. No peaks assignable to
other possible chain-end groups, namely ethyl, n-butyl, sec-
butyl, and isopropyl groups, were observed in the NMR
spectrum.
found in the polypropylene. This indicates unambiguously that
the first propylene is inserted into the Ti-Me bond of an initial
active species derived from complex 1 and MAO in a 1,2-
fashion. On the other hand, the ratio of the isopentyl group (an
initiation chain-end; 32.6%) and the isobutyl group derived from
initiation chain-end (15.2%; estimated based on the amount of
the isopentyl group) is 70 to 30. In addition, the amount of the
n-propyl chain-end group is 34.8%, which is the termination
chain-end derived from the termination after two consecutive
2,1-insertions, and the amount of the isobutyl group originating
from termination chain-end is 17.4%, which is the termination
chain-end derived from the termination after two consecutive
1,2-insertions. These results suggest that 2,1-insertion is pre-
dominant for chain propagation (ca. 70%). The absence of
n-butyl and isopropyl chain-end groups, which originate from
the termination after 1,2-2,1- or 2,1-1,2-insertion, may indicate
that the polymer has a blocklike structure consisting of
consecutive 2,1-insertion and 1,2-insertion segments (regio-block
structure). Moreover, the syndiotactic polypropylenes having
Mns of 3300 (Mw/Mn 1.08)28b and 9700 (Mw/Mn 1.05)28c also
possess isopentyl, n-propyl, and isobutyl chain-end groups with
practically the same content as the living polypropylene
described above (Mn 2000), indicating that chain-end structures
are independent of polymer molecular weights. These facts
further confirm the chain propagation mechanism discussed
above. Therefore, the propylene polymerization was demon-
strated to be initiated exclusively by a 1,2-insertion, followed
by a 2,1-insertion (ca. 70%) as the principal mode of polym-
erization. To our knowledge, this is the first example of a
predominant 2,1-insertion mechanism for chain propagation
displayed by a group 4 metal-based catalyst,22c though V- and
Fe-based catalysts are known to prefer 2,1-insertion of
propylene.29d,30 Predominant 2,1-insertion may be derived from
the steric congestion near the polymerization reaction center,
which places the methyl of the reacting propylene in the opposite
direction from a polymer chain. Recently, the generality of this
highly unusual and unprecedented mechanism of insertion in
The mechanism for the formation of possible chain-end
groups is summarized in Scheme 2. On the basis of the peak
intensities of the methyl carbons, the amounts of the above-
mentioned chain-end groups are suggested to be 1.6 mol % (n-
propyl group, a termination chain-end, 34.8%), 1.5 mol %
(isobutyl group, an initiation or a termination chain-end, 32.6%),
and 1.5 mol % (isopentyl group, an initiation chain-end, 32.6%),
respectively.
If the first propylene is enchained by a 2,1-insertion, the
resulting polymer possesses an ethyl or a sec-butyl chain-end
group. However, no ethyl or a sec-butyl chain-end group is
(28) Polymerization conditions: (a) cat. 1 (100 µmol), MAO (5 mmol), 25 °C,
15 min; (b) cat. 1 (100 µmol), MAO (5 mmol), 25 °C, 30 min; (c) cat. 1
(30 µmol), MAO (5 mmol), 25 °C, 90 min.
(29) (a) Corradini, P.; Guerra, G.; Pucciariello, R. Macromolecules 1985, 18,
2030-2034. (b) Chen, H. N.; Bennett, M. A. Macromol. Chem. 1987, 188,
135-148. (c) Hagiwara, H.; Shiono, T.; Ikeda, T. Macromolecules 1997,
30, 4783-4785. (d) Zambelli, A.; Sessa, I.; Grisi, F.; Fusco, R.; Accomazzi,
P. Macromol. Rapid Commun. 2001, 22, 297-310.
(30) (a) Small, B. L.; Brookhart, M. Macromolecules 1999, 32, 2120-2130.
(b) Pellecchia, C.; Mazzeo, M.; Pappalardo, D. Macromol. Rapid Commun.
1998, 19, 651-655.
9
4298 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Syndiospecific Living Propylene Polymerization
A R T I C L E S
Scheme 2. Mechanism of Initiation and Chain Propagation for Propylene Polymerization Using Complex 1
Table 3. Results of Propylene Polymerization with Complexes
2-8/MAOa
Table 4. Microstructural Analysis of PP Obtained by Complexes
1, 2, 4, 6-8/MAO
b
d
T
(°C)
yield
(g)
TOF
M c
Tm
microstructural data
n
entry
complex
(h-1
)
(×103)
M /M c
(°C)
w
n
a
T
(°C)
Tm
rr
(%)
mr
(%)
mm
(%)
H−T
(%)
H−H
(%)
T−T
(%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
2
2
2
3
3
3
4
4
4
5
5
5
6
7
8
0
25
50
0
25
50
0
25
50
0
25
50
25
25
25
0.109
0.219
0.263
0.101
0.115
0.114
0.151
0.293
0.237
0.089
0.174
0.132
1.534
3.440
1.555
52
19.2
29.8
31.3
16.8
16.5
15.7
24.7
47.0
35.1
11.9
24.4
20.4
189.0
260.2
153.7
1.05
1.15
3.19
1.04
1.11
1.48
1.08
1.08
1.23
1.08
1.16
1.23
1.51
1.22
1.16
141
135
127
143
140
124
156
152
149
152
151
148
nde
entry
complex
(°C)
104
125
48
55
54
1
2
3
4
5
6
7
8
1
2
4
4
4
6
7
8
25
25
0
25
50
25
25
25
137
135
156
152
149
ndb
ndb
ndb
87
86
94
93
90
43
50
75
10
10
4
4
7
46
42
22
3
4
2
3
3
11
8
3
92
89
94
94
90
81
85
84
4
4
3
3
6
10
9
8
4
7
3
3
4
9
7
8
72
139
113
42
83
63
729
1635
739
a Melting temperature of produced PP determined by DSC. b Not
detected.
nde
of the produced polymers, the propylene polymerization ability
of complexes 2-8 having a series of substituents at the R1 and
R2 positions was investigated. Propylene polymerizations were
conducted under atmospheric pressure using MAO as a cocata-
lyst. The polymerization results are tabulated in Table 3, and
the data for 13C NMR microstructural analyses of the produced
polypropylenes are summarized in Table 4.
Regarding the R2 position, the data indicated that the R2
substituent had no significant effect on the catalytic performance
of the complexes, probably because the R2 position is far from
the active site. Thus, complexes 2 (R2 ) tert-butyl) and 3 (R2
) methyl) exhibited similar catalytic properties to complex 1
(R2 ) H) and produced nearly monodisperse syndiotactic
polypropylenes with Tms ranging from 141 to 127 °C (complex
2) and 143 to 124 °C (complex 3) over a temperature range of
0-50 °C. At 50 °C, complexes 2 and 3 generated polypropyl-
enes with broadened molecular weight distributions the same
as complex 1 (Mw/Mn: 1, 1.37; 2, 3.19; 3, 1.48), suggesting
that chain termination or transfer and/or catalyst deactivation
are not negligible for these complexes at this temperature.
It was revealed that the R1 substituent significantly affected
the catalytic properties of the complexes. Complex 4 bearing a
nde
a Conditions: complex (10 µmol), cocat. MAO (2.5 mmol), 0.1 MPa, 5
h. b Turnover frequency. c Determined by GPC using polypropylene calibra-
tion. d Melting temperature of produced PP determined by DSC. e Not
detected.
propylene polymerization with Ti-FI Catalysts has been con-
firmed by Pellecchia31 and Coates.23c More recently, Talarico,
Cavallo, and Busico have reported32 that based on a theoretical
approach a 1,2-insertion is slightly favored for a 1,2-inserted
polymer chain, whereas a 2,1-insertion is preferred for a 2,1-
inserted polymer chain, which is consistent with our experi-
mental data.
The Effects of Ligand Structures on Catalytic Behavior
and Polymer Microstructures. To examine the effects of
substitutions on the phenoxy-imine ligands on catalytic per-
formance of the resulting complexes and on the microstructures
(31) Lamberti, M.; Pappalardo, D.; Zambelli, A.; Pellecchia, C. Macromolecules
2002, 35, 658-663.
(32) (a) Cavallo, L.; Talarico, G. Natl. Meet.-Am. Chem. Soc., DiV. Polym.
Mater.: Sci. Eng. 2002, 87, 38. (b) Talarico, G.; Busico, V.; Cipullo, R.;
Cavallo, L. Proceedings of the 1st Blue Sky Conference on Catalytic Olefin
Polymerization, Sorrento, Italy, June 20-21, 2002.
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4299

A R T I C L E S
Mitani et al.
trimethylsilyl at the R1 position yielded (at 25 °C) highly
syndiotactic narrow MWD polypropylene (Mw/Mn 1.08, rr 93%)
with an extremely high Tm of 152 °C. Likewise, complex 5 (R1
) triethylsilyl) produced highly syndiotactic polypropylene with
a very high Tm (151 °C). The pentad distribution analysis
revealed that the syndiospecific polymerizations proceed via a
chain-end control mechanism. Such an extremely high level of
chain-end control at room temperature is unprecedented in a
propylene polymerization. The syndiotactic polypropylene
produced by complex 4 at 0 °C displayed an exceptionally high
Tm of 156 °C (rr 94%), which represents one of the highest Tm
values among syndiotactic polypropylenes ever known.33 Re-
markably, at 50 °C, complexes 4 and 5 furnished syndiotactic
polypropylenes with very high Tms of 149 and 148 °C,
respectively. These are of great significance because generally,
as described, chain-end control works well at very low to
subambient temperatures and rapidly loses its stereoregulating
capability at elevated temperatures.
Figure 8. Plots of rr triad values as a function of calculated R1-H volume
for propylene polymerization with complexes 1, 4, 6-8/MAO (25 °C, 5
h).
The reduction in the steric bulk of the R1 substituent resulted
in the enhancement in catalytic activity and, interestingly,
marked decrease in polymer stereoregularity, producing amor-
phous polypropylenes. Complex 8, bearing an isopropyl group
at the R1 position, formed polypropylene having lower syndio-
tacticity (rr 75%) with a much higher TOF of 739 h-1, which
is a factor of ca. 8 higher than that of complex 1. The effect of
the steric bulk of the R1 substituent was more pronounced for
complex 7 having a methyl, which exhibited a far higher TOF
of 1635 h-1 and produced polypropylene possessing still lower
syndiotacticity (rr 50%). The enhanced activities probably result
from the structurally open nature of the complexes. Complex 6
with a hydrogen at the R1 position displayed a lower TOF of
729 h-1 compared to complex 7 (R1 ) methyl) and formed
amorphous polypropylenes (rr 43%) with a broadened molecular
weight distribution. The lower activity and broadened molecular
weight distribution value displayed by complex 6 (R1 ) H) are
probably ascribed to the catalyst decay due to the extremely
reduced steric bulk of the R1 substituent, which provides steric
protection to the phenoxy oxygen from the attack of aluminum
species in the reaction medium.
The decisive role of the R1 substituent in realizing the highly
syndiospecific polymerization indicated that the R1 substituent
is situated in close proximity to the active site, and the R1
substituent can have steric repulsion against the methyl of the
reacting propylene, leading to syndiospecific polymerization.
In fact, a correlation exists between the syndioselectivity and
the size of the R1 substituent. The plots of rr triad values vs
the size of the R1 substituent34 are displayed in Figure 8, where
a linear relationship was found. These results further confirmed
that the steric bulk of the R1 substituent controls the syndiose-
lectivity of the polymerization. The Ti complexes are unique
in their ability to form chain-end-controlled stereoregular
polymers whose stereoregularity is governed by their ligand
structures, and thus the way of controlling the stereoselectivity
is similar to that for a site-control mechanism.
Propylene polymerization governed by a chain-end control
typically gives low to moderate tacticity polypropylenes, and
changing the ligand structure of a catalyst can have influence
on tacticity but not significantly. However, the Ti complexes
introduced herein are capable of promoting highly syndiospecific
propylene polymerizations. Moreover, the ligand structure has
a dramatic effect on tacticity, and the sterically bulky substituent
ortho to the phenoxy oxygen results in highly stereospecific,
thermally robust, chain-end controlled living propylene polym-
erization. Therefore, we have given the name “ligand-directed
chain-end control” to this highly controlled polymerization that
enables us to have highly syndiotactic polypropylenes with
extremely high Tms.
The regiochemistry of the polypropylenes produced with these
complexes was investigated on the basis of the 13C NMR peaks
in the regions of 13-18 and 30-45 ppm, attributable to
regioirregular units (Table 4 and Figure 9).35 The data revealed
that the polypropylenes formed with the Ti complexes contain
a considerable amount of regioirregularly arranged propylene
units (regioinversion: 1, 8%; 4, 6%; Table 4, entries 1 and 4)
in marked contrast to the syndiotactic polypropylene obtained
with the Ewen-type ansa-metallocene Ph2C(Cp, 2,7-di-tBu-Flu)-
ZrCl2 (complex 9, regioinversion 1%).36 Although V- and Ni-
based catalysts are known to polymerize propylene in a
syndiotactic fashion with significant regioinversion, these are
rare examples of group 4 metal-based catalysts that produce
syndiotactic polypropylenes with considerable regioirregular
units.
Further analyses of the spectra indicate that the polypropy-
lenes arising from the Ti complexes possess blocklike structures,
which are classified into two types; one involves consecutive
regioirregular units (type I, Figure 10) and the other includes
isolated regioirregular units (type II, Figure 10). Thus, the
syndiotactic polypropylenes, formed from complexes 1 and 4,
with a significant amount of regioirregular units possess a
blocklike structure involving long regioregular units and short
consecutive-regioirregular units (type I). On the other hand, the
(33) (a) Veghini, D.; Henling, L. M.; Burkhardt, T. J.; Bercaw, J. E. J. Am.
Chem. Soc. 1999, 121, 564-573. (b) Grisi, F.; Longo, P.; Zambelli, A.;
Ewen, J. A. J. Mol. Catal. A 1999, 140, 225-233.
(34) A volume of substituent was calculated as follows. The geometry of the
substituent, which is capped with a hydrogen atom, was optimized using
the 6-311G(d, p) basis set at the Hartree-Fock level, and its volume was
calculated on the basis of the van der Waals radii of each atom. The
geometry and its volume was calculated using the Spartan’02 program
(Spartan’02, Wavefunction, Inc. Irvine, CA).
(35) The percentage of regioinversion units (H-H, T-T) is calculated on the
basis of NMR peak integrals: H-T, 27.0-29.5 ppm (Tââ); H-H, 14.0-
18.0 ppm (PRâ); T-T, 30.5-31.5 ppm (Tâγ).
(36) The polymer was synthesized under the same conditions (25 °C): Mn, 159.7
× 103; Mw/Mn, 1.86; Tm, 148 °C; rr/mr/mm (%), 95/4/1; H-T/H-H/T-T
(%), 99/1/0.
9
4300 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Syndiospecific Living Propylene Polymerization
A R T I C L E S
Figure 9. 13C NMR spectrum of PPs obtained by complexes 1, 4, 6, and 9 at 25 °C.
Figure 10. Regiosequence patterns of PPs obtained by fluorinated Ti-FI Catalysts.
syndiotactic-rich polypropylenes, produced with complexes 6-8
(see the Supporting Information for the 13C NMR spectra of
the polypropylenes formed with complexes 7 and 8), with a
higher amount of regioirregular units, also have a blocklike
structure, but it contains relatively short regioregular units and
isolated regioirregular units (type II). Therefore, the polypro-
pylenes obtained with the Ti complexes are revealed to possess
unique molecular architectures.
It is noted that the Tms of syndiotactic polypropylenes formed
with the Ti complexes are higher than the values expected from
the syndiotacticity of the polymers obtained with metallocene
catalysts. For example, the syndiotactic polypropylene contain-
ing 95% rr triads, formed with complex 9, displays a Tm of
148 °C under the same DSC measurement conditions. The
differences in Tm probably originate from the blocklike structures
of the polymers arising from the Ti complexes.
The Origin of Syndioselectivity. On the basis of the C2
symmetric structures of the Ti complexes 1, 4, and 6 demon-
strated by X-ray analyses, the Ti complexes were initially
targeted as living polymerization catalysts capable of forming
isotactic polypropylenes via an enantiomorphic site control
mechanism, as described. Consequently, the production of the
syndiotactic polypropylenes was unexpected and surprising. The
unexpected production of the syndiotactic polypropylenes
deserves investigation.
The results discussed herein show that the Ti complexes
promote chain-end controlled, syndiospecific propylene polym-
erization via a 2,1-insertion mechanism and that the steric bulk
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4301

A R T I C L E S
Mitani et al.
of the catalyst can also be explained by the proposed mechanism,
because stereoselectivity is virtually governed by the ligand
structure. Thus, the predicted catalytic behavior derived from
the site-inversion mechanism and our experimental results
(ligand-directed chain-end control) match well. Therefore, the
site-inversion mechanism is most probable and reasonable for
the unexpected formation of highly syndiotactic polypropylenes
at the present time.
The site-inversion mechanism gives the idea that if propylene
insertion is faster than the site-inversion, a Ti complex with a
sterically hindered substituent ortho to the phenoxy oxygen can
be a catalyst for the highly isospecific polymerization of
propylene. Reduction in the site-inversion rate may be achieved
by ligand modifications such as bridging the ligands and the
introduction of steric congestion, which suppresses the site-
inversion.9,20l Such phenoxy-imine ligated complexes could be
catalysts for the isospecific and living polymerization of
propylene, resulting in the production of “living isotactic
polypropylenes”. Detailed mechanistic studies including the
elucidation of the structure of catalytically active species are
underway, which may provide a new stereoregulating strategy
for R-olefin polymerization.
Figure 11. Proposed mechanism for syndiospecific propylene polymeri-
zation catalyzed by Ti-FI Catalysts.
of the substituent ortho to the phenoxy oxygen plays a pivotal
role in determining the syndioselectivity of the polymerization.
It should be pointed out that the principal behavior of the Ti
complexes for propylene polymerization (chain-end controlled
syndiospecific polymerization via a 2,1-insertion) is the same
as that of V-based catalysts, though the V-based catalysts suffer
from low stereoselectivity. There may be a similarity between
the polymerization mechanisms displayed by the Ti and V
catalysts. In fact, recently, on the basis of a QM/MM theoretical
study, Cavallo and Pellecchia have proposed a site-inversion
mechanism37,38 (Figure 11), which was originally introduced
for the explanation of syndiospecific propylene polymerization
exhibited by V-based catalysts. The bottom line of the proposed
mechanism is that a site-inversion between diastereomeric Λ
and ∆ configurations, which may occur via a five-coordinate
Ti cationic species, is much faster than propylene insertion.
According to the proposed mechanism, the octahedral active
species undergoes fluxional isomerization between Λ and ∆
forms governed by the stereochemistry of R-carbon in the last-
inserted propylene unit. Thus, propylene insertion to the Ti-
alkyl bond of a cationic species with Λ configuration, having a
lower energy than that with ∆ configuration, leads to a new
cationic species with ∆ configuration, which undergoes a site-
inversion to form a cationic species with Λ configuration prior
to propylene insertion.
Experimental Section
General Comments. Syntheses. Ligand syntheses were carried out
under nitrogen using oven-dried glassware. All manipulations of
complex syntheses were performed with the exclusion of oxygen and
moisture under argon using standard Schlenk techniques using oven-
dried glassware.
Materials. Dried solvents [toluene, tetrahydrofuran (THF), diethyl
ether, dichloromethane, and n-hexane] used for complex syntheses were
purchased from Wako Pure Chemical Industries, Ltd., and used without
further purification. Toluene used as a polymerization solvent (Wako
Pure Chemical Industries, Ltd.) was dried over Al2O3 and degassed by
the bubbling of dried nitrogen gas. Phenol derivatives and aniline
derivatives for ligand synthesis were obtained from Aldrich Chemical
Co., Inc., Wako Pure Chemical Industries, Ltd., Acros Organics, Kanto
Chemicals Co. Inc., or Tokyo Kasei Kogyo Co., Ltd. Trimethylsilyl
chloride was purchased from Wako Pure Chemical Industries, Ltd.,
and 2-triethylsilylphenol was obtained from Shin-Etsu Chemicals Co.
A 1.0 M trichloroborane in CH2Cl2 solution and a n-butyllithium in
n-hexane solution were purchased from Aldrich Chemical Co., Inc.
and Wako Pure Chemical Industries, Ltd., respectively. TiCl4 (Wako
Pure Chemical Industries, Ltd.) was used as received. Bis[N-(3-tert-
butylsalicylidene)-2,3,4,5,6-pentafluoroanilinato]titanium(IV) dichloride
(1) was synthesized according to our previous report.21b Complex 9
was synthesized according to the patent.40 Propylene was obtained from
Mitsui Chemicals, Inc.
Assuming that the site-inversion is operating in the propylene
polymerization and the catalytically active species possesses an
octahedral geometry with a trans-O and cis-N disposition,39 the
steric bulk of the substituent ortho to the phenoxy oxygen should
have profound effects on the syndioselectivity of the polymer-
ization, which is exactly what we observed (Figure 8). In
addition, the highly syndiospecific and thermally robust behavior
Cocatalyst. Methylalmoxane (MAO) was purchased from Albemarle
as a 1.2 M toluene solution, and the remaining trimethylaluminum was
evaporated in vacuo prior to use.
1
Ligand and Complex Analysis. H NMR spectra were recorded
1
on a JEOL270 at ambient temperatures. Chemical shifts for H NMR
were referenced to the resonance of tetramethylsilane. FD-MS spectra
were recorded on an SX-102A from Japan Electron Optics Laboratory
Co., Ltd. Elemental analysis for CHN was carried out by a CHNO
type from Helaus Co.
Polymer Characterization. 13C NMR data for polypropylenes were
obtained using o-dichlorobenzene with 20% benzene-d6 as a solvent at
120 °C using an ECP500 spectrometer from Japan Electron Optics
Laboratory Co., Ltd., at 125 MHz. The spectra were referenced vs r
pentad methyl peak shifts, δ ) 19.97 ppm. Mn and Mw/Mn values of
(37) (a) Lamberti, M.; Milano, G.; Cavallo, L.; Guerra, G.; Pellecchia, C. Natl.
Meet.-Am. Chem. Soc., DiV. Polym. Mater.: Sci. Eng. 2002, 87, 44-46.
(b) Milano, G.; Cavallo, L.; Guerra, G. J. Am. Chem. Soc. 2002, 124,
13368-13369.
(38) Brookhart et al. have successfully prepared a stereoblock polyketone with
a Pd-based catalyst through a ligand exchange. Brookhart, M.; Wagner,
M. I. J. Am. Chem. Soc. 1996, 118, 7219-7220.
(39) Cationic complexes described herein can take more than one configuration
as a consequence of the different binding geometries of the phenoxy-
imine ligands. Therefore, the possibilities cannot be ruled out that a
catalytically active species possesses configurations other than a trans-O
and cis-N arrangement. If a catalytically active species has an octahedral
geometry with a cis-O and cis-N disposition, the mechanism for the
production of the syndiotactic polypropylene seems very complicated.
(40) Ewen, J. A.; Reddy, B. R.; Elder, M. J. Europe Patent, EP-0577581.
9
4302 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Syndiospecific Living Propylene Polymerization
A R T I C L E S
polypropylenes were determined using a Waters GPC2000 gel perme-
ation chromatograph equipped with four TSKgel columns (two sets of
TSKgelGMHHR-H(S)HT and two sets of TSKgelGMH6-HTL) at 140
°C using polypropylene calibration. o-Dichlorobenzene was employed
as a solvent at a flow rate of 1.0 mL/min. Transition melting temper-
atures (Tm) of the polypropylenes were determined by DSC with a Shi-
madzu DSC-60 differential-scanning calorimeter, measured upon re-
heating the polymer sample to 200 °C at a heating rate of 10 °C/min.
Preparation of Bis[N-(3,5-di-tert-butylsalicylidene)-2,3,4,5,6-pen-
tafluoroanilinato]titanium(IV) Dichloride (2).
1.47 (s, 3H, Me), 7.04 (d, J ) 1.9 Hz, 1H, aromatic-H), 7.27 (d, J )
1.9 Hz, 1H, aromatic-H), 8.76 (s, 1H, CHdN), 12.71 (s, 1H, OH).
Bis[N-(3-tert-butyl-5-methylsalicylidene)-2,3,4,5,6-pentafluoroa-
nilinato]titanium(IV) Dichloride, Complex 3: reddish brown crystals;
1H NMR (CDCl3); δ 1.33 (s, 18H, tBu), 1.35 (s, 6H, Me), 7.06 (d, J )
1.9 Hz, 2H, aromatic-H), 7.42 (d, J ) 1.9 Hz, 2H, aromatic-H), 8.14
(s, 2H, CHdN); FD-MS, 830 (M+). Anal. Found: C, 51.91; H, 3.51;
N, 3.30. Calcd. for TiC36H30N2O2F10Cl2: C, 52.01; H, 3.64; N, 3.37.
Preparation of Bis[N-(3-trimethylsilylsalicylidene)-2,3,4,5,6-pen-
tafluoroanilinato]titanium(IV) Dichloride (4).
Ligand Synthesis: 3,5-Di-tert-butylsalicylaldehyde. To a stirred
solution of 3.0 M ethylmagnesium bromide in dried THF (40 mL) was
added a solution of 2,4-di-tert-butylphenol (20.63 g, 100 mmol) in dried
THF (40 mL) dropwise over a 10-min period at 0 °C and the mixture
was stirred for 0.5 h at room temperature. To the resulting mixture,
dried toluene (250 mL) was added and then a mixture of triethylamine
(19.9 mL, 143 mmol) and paraformaldehyde (7.73 g, 94% purity, 242
mmol) was added. The mixture was heated at 75 °C for 40 min with
stirring and cooled to 0 °C. To the resulting mixture was added a
mixture of concentrated HCl (20 mL) and H2O (150 mL) at 0 °C. The
organic phase was separated, and the aqueous phase was extracted with
n-hexane (150 mL × 2). The combined organic phase was washed
with saturated aqueous NaHCO3 (100 mL) and brine (100 mL) and
then dried over MgSO4. Evaporation of the solvent in vacuo gave a
yellow oil, which was purified by column chromatography on silica
gel using n-hexane/AcOEt (100/1) as eluent to give 3,5-di-tert-
butylsalicylaldehyde (18.61 g, 77.0 mmol) as a pale yellow solid in
Ligand Synthesis: 2-Trimethylsilylanisole. To a stirred solution
of N,N,N′,N′-tetramethylethylenediamine (5.75 g, 49.5 mmol) in dried
diethyl ether (36 mL) at 20 °C was added a 1.6 M n-butyllithium in
n-hexane solution (31 mL, 49.5 mmol) dropwise over a 20-min period.
To the mixture was added a solution of anisole (4.87 g, 45.0 mmol) in
dried diethyl ether (36 mL) dropwise over a 5-min period and the
resulting mixture stirred at 50 °C for 7 h. Chlorotrimethylsilane (6.85
mL, 54.0 mmol) in dried diethyl ether (18 mL) was added dropwise
over a 10-min period at 0 °C to the mixture and the resulting mixture
stirred for 1.5 h. To the mixture was added H2O (60 mL) at 0 °C. The
organic phase was separated, and the aqueous phase was extracted with
n-hexane (50 mL × 2). The combined organic solution was washed
with aqueous NH4Cl (50 mL) and brine (100 mL) and dried over
MgSO4. Concentration of the organic solution in vacuo gave a crude
product as yellow oil, which was purified by column chromatography
on silica gel using n-hexane as eluent to give 2-trimethylsilylanisole
(6.59 g, 36.4 mmol) as colorless oil in 81% yield: 1H NMR (CDCl3)
δ 0.26 (s, 9H, Si(CH3)3), 3.80 (s, 3H, OCH3), 6.83 (t, J ) 7.8 Hz, 1H,
aromatic-H), 6.94 (dd, J ) 7.8, 1.8 Hz, 1H, aromatic-H), 7.34 (t, J )
7.8 Hz, 1H, aromatic-H), 7.37 (dd, J ) 7.8, 1.8 Hz, 1H, aromatic-H).
3-Trimethylsilylsalicylaldehyde Methyl Ether. To a stirred solution
of N,N,N′,N′-tetramethylethylenediamine (21.18 g, 182.3 mmol) in dried
diethyl ether (40 mL) at 20 °C was added a 1.6 M n-butyllithium in
n-hexane solution (113.9 mL, 182.3 mmol) dropwise over a 30-min
period. To the mixture was added a solution of 2-trimethylsilylanisole
(30.18 g, 99% purity, 165.7 mmol) in dried diethyl ether (100 mL)
dropwise over a 10-min period. The mixture was stirred at 50 °C for
8 h. N,N-dimethylformamide (15.4 mL, 198.8 mmol) in dried diethyl
ether (70 mL)was added to the mixture at 0 °C over a 25-min period
and the mixture was stirred for 1 h at room temperature. To the mixture
was added H2O (200 mL) at 0 °C. The organic phase was separated,
and the aqueous phase was extracted with diethyl ether (100 mL × 2).
The combined organic solution was washed with aqueous NH4Cl (100
mL) and brine (150 mL) and dried over MgSO4. Concentration of the
organic solution in vacuo gave the crude product as an orange oil, which
was purified by column chromatography on silica gel using n-hexane/
AcOEt (20/1) as eluent to give 3-trimethylsilylsalicylardehyde methyl
ether (19.95 g, 92.4 mmol) as a yellow oil in 56% yield: 1H NMR
(CDCl3) δ 0.34 (s, 9H, Si(CH3)3), 3.92 (s, 3H, OCH3), 7.23 (t, J ) 7.3
Hz, 1H, aromatic-H), 7.66 (dd, J ) 7.8, 1.8 Hz, 1H, aromatic-H), 7.85
(dd, J ) 7.8, 1.8 Hz, 1H, aromatic-H), 10.34 (s, 1H, CHO).
t
t
77% yield: 1H NMR (CDCl3) δ 1.33 (s, 9H, Bu), 1.43 (s, 9H, Bu),
7.33 (d, J ) 2.7 Hz, 1H, aromatic-H), 7.59 (d, J ) 2.7 Hz, 1H, aromatic-
H), 9.86 (s, 1H, CHO), 11.64 (s, 1H, OH).
Ligand b. To a stirred mixture of 3,5-di-tert-butylsalicylaldehyde
(2.42 g, 97% purity, 10.0 mmol) and 2,3,4,5,6-pentafluoroaniline (2.01
g, 11.0 mmol) in toluene (40 mL) was added p-toluenesulfonic acid
(ca. 10 mg) at room temperature. The mixture was heated at reflux
temperature for 6 h. The mixture was allowed to cool to room
temperature and concentrated in vacuo to afford a crude imine
compound. The crude imine compound was purified by column
chromatography on silica gel using n-hexane/AcOEt (100/1) as eluent.
Evapolation of the solvent gave N-(3,5-di-tert-butylsalicylidene)-
2,3,4,5,6-pentafluoroaniline (b) (2.99 g, 7.41 mmol) as a yellow-orange
solid in 74% yield: 1H NMR (CDCl3) δ 1.33 (s, 9H, tBu), 1.47 (s, 9H,
tBu), 7.19 (d, J ) 2.4 Hz, 1H, aromatic-H), 7.53 (d, J ) 2.4 Hz, 1H,
aromatic-H), 8.81 (s, 1H, CHdN), 12.71 (s, 1H, OH).
Complex Synthesis. To a stirred solution of N-(3,5-di-tert-butyl-
salicylidene)-2,3,4,5,6-pentafluoroaniline (1.21 g, 3.00 mmol) in dried
diethyl ether (25 mL) at -78 °C was added a 1.56 M n-butyllithium in
n-hexane solution (1.92 mL, 3.00 mmol) dropwise over a 5-min period.
The mixture was allowed to warm to room temperature and stirred for
3 h. The resulting mixture was added dropwise over a 5-min period to
a 0.5 M n-heptane solution of TiCl4 (3.00 mL, 1.50 mmol) in dried
diethyl ether (25 mL) at -78 °C. The mixture was allowed to warm to
room temperature and stirred for 19 h. Evapolation of the solvent affords
a crude product. To the crude product was added dried CH2Cl2 (40
mL), and the mixture was stirred for 15 min. The resulting mixture
was filtered, and the residue was washed with CH2Cl2 (5 mL × 2).
The combined organic filtrates were concentrated in vacuo to give a
dark red solid. Diethyl ether (20 mL) and n-hexane (40 mL) were added
to the solid and the mixture stirred for 30 min. The resulting mixture
was filtered, and the solid was washed with n-hexane (15 mL × 2)
and dried in vacuo to give complex 2 (0.536 g, 0.59 mmol) as brown
crystals in 37% yield:
3-Trimethylsilylsalicylaldehyde. To a stirred solution of 3-trimeth-
ylsilylsalicylardehyde methyl ether (19.90 g, 97% purity, 92.2 mmol)
in CH2Cl2 (150 mL) at -78 °C was added a 1.0 M trichloroborane in
CH2Cl2 solution (200 mL, 200 mmol) dropwise over a 2-h period. The
mixture was allowed to warm to room temperature and stirred for 5 h.
To the mixture was added H2O (200 mL) and then the mixture was
filtered. The organic phase was separated, and the aqueous phase was
extracted with n-hexane/AcOEt (9/1, 100 mL × 2). The combined
organic solution was washed with brine (200 mL × 2) and dried over
MgSO4. Concentration of the organic solution in vacuo gave the crude
product as a dark orange oil, which was purified by column chroma-
tography on silica gel using n-hexane/AcOEt (100/1) as eluent to give
3-trimethylsilylsalicylaldehyde (16.63 g, 83.9 mmol) as yellow oil in
91% yield: 1H NMR (CDCl3) δ 0.32 (s, 9H, Si(CH3)3), 7.00 (t, J )
7.3 Hz, 1H, aromatic-H), 7.55 (dd, J ) 7.8, 1.8 Hz, 1H, aromatic-H),
1H NMR (CDCl3) δ 1.32 (s, 18H, tBu), 1.36 (s, 18H, tBu), 7.20 (d,
J ) 2.2 Hz, 2H, aromatic-H), 7.67 (d, J ) 2.2 Hz, 2H, aromatic-H),
8.21 (s, 2H, CHdN); FD-MS, 914 (M+).
Ligand c, N-(3-tert-Butyl-5-methylsalicylidene)-2,3,4,5,6-pen-
tafluoroaniline: orange crystals; 1H NMR (CDCl3) δ 1.45 (s, 9H, tBu),
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4303

A R T I C L E S
Mitani et al.
Table 5. Summary of Crystallographic Data for Complexes
7.61 (dd, J ) 7.8, 1.8 Hz, 1H, aromatic-H), 9.88 (s, 1H, CHO), 11.30
(s, 1H, OH).
4 and 6
complex
formula
4
6
Ligand d. To a stirred mixture of 3-trimethylsilylsalicylaldehyde
(2.38 g, 98% purity, 12.0 mmol) and 2,3,4,5,6-pentafluoroaniline (3.30
g, 18.0 mmol) in toluene (50 mL) was added p-toluenesulfonic acid
(ca. 5 mg) at room temperature. The resulting mixture was stirred at
reflux temperature for 13 h, and concentration of the reaction mixture
in vacuo afforded a crude imine compound. Purification by column
chromatography on silica gel using n-hexane as eluent and recrystal-
lization from cold methanol gave N-(3-trimethylsilylsalicylidene)-
2,3,4,5,6-pentafluoroaniline (d) (3.28 g, 9.13 mmol) as yellow crystals
in 76% yield: 1H NMR (CDCl3) δ 0.36 (s, 9H, Si(CH3)3), 6.96 (t, J )
7.3 Hz, 1H, aromatic-H), 7.38 (dd, J ) 7.8, 1.8 Hz, 1H, aromatic-H),
7.73 (dd, J ) 7.8, 1.8 Hz, 1H, aromatic-H), 8.79 (s, 1H, CHdN), 12.41
(s, 1H, OH).
C32H26N2O2F10Si2TiCl2 C27H12N2O2F10TiCl4
835.53
formula weight
crystal color, habit
crystal size (mm)
crystal system
space group
a (Å)
776.10
red, block
0.20 × 0.20 × 0.20
monoclinic
C2/c (no. 15)
29.963(4)
16.29(1)
17.264(4)
115.82(1)
7582(5)
8
1.464
3376.00
5.10
0.710 69
-15
55.0
10380
8698
4618 (I > 3.00σ(I))
460
10.04
0.055, 0.098
0.040
1.30
0.00
red, needle
0.60 × 0.10 × 0.10
orthorhombic
Pbca (no. 61)
16.450(4)
b (Å)
c (Å)
24.882(5)
14.521(2)
â (deg)
V (Å3)
5943(2)
8
1.735
3072.00
7.41
0.710 69
-50
55.0
9807
6829
3042 (I > 3.00σ(I))
415
7.33
0.056, 0.065
0.048
Z
D
calcd (g cm-3
)
F000
µ (MoKR) (cm-1
λ (MoKR) (Å)
T (°C)
)
Complex Synthesis. To a stirred solution of N-(3-trimethylsilyl-
salicylidene)-2,3,4,5,6-pentafluoroaniline (1.44 g, 4.00 mmol) in dried
diethyl ether (20 mL) at -78 °C was added a 1.6 M n-butyllithium in
n-hexane solution (2.52 mL, 4.00 mmol) dropwise over a 5-min period.
The solution was allowed to warm to room temperature and stirred for
2 h. The resulting solution was added dropwise over a 5-min period to
a 0.5 M n-heptane solution of TiCl4 (4.00 mL, 2.00 mmol) in dried
diethyl ether (20 mL) at -78 °C. The mixture was allowed to warm to
room temperature and stirred for 19 h. Concentration of the reaction
mixture in vacuo gave a crude product. Dried CH2Cl2 (50 mL) was
added to the crude product, and the mixture was stirred and then filtered.
The solid residue was washed with dried CH2Cl2 (5 mL × 2), and the
combined organic filtrates were concentrated in vacuo to afford a dark
red solid. Diethyl ether (10 mL) and n-hexane (30 mL) were added to
the solid, and the mixture was stirred for 20 min and then filtered. The
resulting solid was washed with n-hexane (10 mL × 2) and dried in
vacuo to give complex 4 (0.764 g, 0.91 mmol) as orange crystals in
46% yield:
2θ max
no. of total reflns
no. of unique reflns
no. of observations
no. of variables
refln/parameter ratio
residuals: R, Rw
R1
GOF indicator
max shift/error in
final cycle
max and min peaks in
final diff map (e Å3)
1.88
0.00
0.58, -0.41
0.64, -0.83
(dd, J ) 7.3, 1.5 Hz, 1H, aromatic-H), 8.81 (s, 1H, CHdN), 12.49 (s,
1H, OH).
Bis[N-(3-methylsalicylidene)-2,3,4,5,6-pentafluoroanilinato]tita-
1
1H NMR (CDCl3) δ 0.31 (s, 18H, Si(CH3)3), 7.06 (t, J ) 7.3 Hz,
2H, aromatic-H), 7.40 (dd, J ) 7.6, 1.6 Hz, 2H, aromatic-H), 7.73
(dd, J ) 7.6, 1.6 Hz, 2H, aromatic-H), 8.19 (s, 2H, CHdN); FD-MS,
834 (M+). Anal. Found: C, 45.69; H, 2.90; N, 3.22. Calcd. for
TiC32H26N2O2F10Si2Cl2: C, 46.00; H, 3.14; N, 3.35.
nium(IV) Dichloride, Complex 7: reddish brown crystals; H NMR
(CDCl3) δ 2.08 (s, 6H, Me), 6.99 (t, J ) 7.6 Hz, 2H, aromatic-H),
7.27 (dd, J ) 7.3, 1.5 Hz, 2H, aromatic-H), 7.48 (dd, J ) 7.3, 1.5 Hz,
2H, aromatic-H), 8.24 (s, 2H, CHdN); FD-MS, 718 (M+). Anal. Found;
C, 46.59; H, 1.79; N, 3.58. Calcd. for TiC28H14N2O2F10Cl2: C, 46.76;
H, 1.96; N, 3.90.
Ligand h, N-(3-Isopropylsalicylidene)-2,3,4,5,6-pentafluoroani-
line: yellow crystals, 2.96 g, 8.98 mmol; 1H NMR (CDCl3) δ 1.28 (d,
J ) 7.0 Hz, 6H, CH3), 3.44 (sp, J ) 6.9 Hz, 1H, CH), 6.95 (t, J ) 7.6
Hz, 1H, aromatic-H), 7.24 (dd, J ) 8.2, 1.5 Hz, 1H, aromatic-H), 7.46
(dd, J ) 7.6, 1.6 Hz, 1H, aromatic-H), 8.81 (s, 1H, CHdN), 12.55 (s,
1H, OH).
Bis[N-(3-isopropylsalicylidene)-2,3,4,5,6-pentafluoroanilinato]ti-
tanium(IV) Dichloride, Complex 8: reddish brown crystals; 1H NMR
(CDCl3) δ 1.19 (d, J ) 7.0 Hz, 6H, CH3), 1.23 (d, J ) 6.8 Hz, 6H,
CH3), 2.90 (sp, J ) 6.8 Hz, 2H, CH), 7.05 (t, J ) 7.7 Hz, 2H, aromatic-
H), 7.29 (dd, J ) 7.8, 1.6 Hz, 2H, aromatic-H), 7.52 (dd, J ) 7.6, 1.4
Hz, 2H, aromatic-H), 8.26 (s, 2H, CHdN); FD-MS, 774 (M+). Anal.
Found: C, 49.36; H, 2.42; N, 3.58. Calcd. for TiC32H22N2O2F10Cl2:
C, 49.57; H, 2.86; N, 3.61.
X-ray Crystallography. Single crystals of complexes 4 and 6
suitable for an X-ray analysis were grown from a saturated pentane/
CH2Cl2 solution (complex 4) or Et2O/CH2Cl2 solution (complex 6).
The X-ray structural analysis data were collected using a Rigaku AFC7R
diffractometer. The structure was solved by direct method41 (complex
4) or heavy-atom Patterson methods42 (complex 6) and expanded using
Fourier techniques. The non-hydrogen atoms were refined anisotropi-
cally. Hydrogen atoms were included but not refined. All calculations
were performed using the teXan crystallographic software package of
Molecular Structure Corp.43 Experimental data for the X-ray diffraction
analyses of complexes 4 and 6 are summarized in Table 5.
Propylene Polymerization under Atmospheric Pressure. Propyl-
ene polymerization under atmospheric pressure was carried out in a
Ligand e, N-(3-Trimethylsilylsalicylidene)-2,3,4,5,6-pentafluo-
1
roaniline: a yellow oil; H NMR (CDCl3) δ 0.85-1.02 (m, 15H, Si-
(C2H5)3), 6.96 (t, J ) 7.4 Hz, 1H, aromatic-H), 7.38 (dd, J ) 7.6, 1.6
Hz, 1H, aromatic-H), 7.53 (dd, J ) 7.3, 1.6 Hz, 1H, aromatic-H), 8.77
(s, 1H, CHdN), 12.39 (s, 1H, OH).
Bis[N-(3-triethylsilylsalicylidene)-2,3,4,5,6-pentafluoroanilinato]-
titanium(IV) Dichloride, Complex 5: reddish brown crystals; 1H
NMR (CDCl3) δ 0.61-1.14 (s, 30H, Si(C2H5)3), 7.05 (t, J ) 7.4 Hz,
2H, aromatic-H), 7.39 (dd, J ) 7.6, 1.6 Hz, 2H, aromatic-H), 7.70
(dd, J ) 7.8, 1.6 Hz, 2H, aromatic-H), 8.16 (s, 2H, CHdN); FD-MS,
918 (M+). Anal. Found: C, 49.88; H, 3.87; N, 3.03. Calcd. for
TiC38H38N2O2F10Si2Cl2: C, 49.63; H, 4.16; N, 3.05.
Ligand f, N-(Salicylidene)-2,3,4,5,6-pentafluoroaniline: pale yel-
low crystals; 1H NMR (CDCl3) δ 6.98 (dt, J ) 1.1, 7.6 Hz, 1H,
aromatic-H), 7.05 (dd, J ) 8.6, 1.1 Hz, 1H, aromatic-H), 7.40 (dd, J
) 7.6, 1.6 Hz, 1H, aromatic-H), 7.46 (dt, J ) 8.6, 1.6 Hz, 1H, aromatic-
H), 8.82 (s, 1H, CHdN), 12.22 (s, 1H, OH).
Bis[N-(salicylidene)-2,3,4,5,6-pentafluoroanilinato]titanium(IV)
Dichloride, Complex 6: reddish brown crystals; 1H NMR (CDCl3) δ
6.63 (d, J ) 8.1 Hz, 2H, aromatic-H), 7.09 (dt, J ) 1.1, 7.6 Hz, 2H,
aromatic-H), 7.46 (d, J ) 8.1 Hz, 2H, aromatic-H), 7.60 (dt, J ) 1.6,
7.6 Hz, 2H, aromatic-H), 8.28 (s, 2H, CHdN); FD-MS, 690 (M+).
Anal. Found: C, 45.31; H, 1.53; N, 4.02. Calcd. for TiC26H10N2O2F10-
Cl2: C, 45.18; H, 1.46; N, 4.05.
Ligand g, N-(3-Methylsalicylidene)-2,3,4,5,6-pentafluoroaniline:
yellow crystals; 1H NMR (CDCl3) δ 2.32 (s, 3H, Me), 6.89 (t, J ) 7.6
Hz, 1H, aromatic-H), 7.25 (dd, J ) 7.3, 1.5 Hz, 1H, aromatic-H), 7.33
9
4304 J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003

Syndiospecific Living Propylene Polymerization
A R T I C L E S
500-mL glass reactor equipped with a propeller-like stirrer. Toluene
(250 mL) was introduced into the nitrogen-purged reactor and stirred
(600 rpm). The toluene was kept at a prescribed polymerization
temperature, and then propylene gas feed (100 L/h) was started. After
15 min, the propylene gas feed was changed to 30 L/h. Polymerization
was initiated by adding a 1.25 M MAO solution in toluene (2 mL, 2.5
mmol) and then a 0.002 M complex solution in toluene (5.0 mL, 10
µmol) was added into the reactor. After a prescribed time, sec-butyl
alcohol (10 mL) was added to terminate the polymerization. To the
resulting mixture were added methanol (1000 mL) and concentrated
HCl (2 mL). The polymer was collected by filtration, washed with
methanol (200 mL × 2), and dried in vacuo at 80 °C for 10 h.
ization of propylene at room temperature or above by the
attractive interaction between a fluorine in the ligand and a
â-hydrogen of a growing polymer chain. The attractive interac-
tion provides a conceptually new strategy for the achievement
of highly controlled living olefin polymerization.
These complexes can also promote chain-end controlled,
highly syndiospecific propylene polymerization via a 2,1-
insertion mechanism. Although chain-end control mechanism
is responsible for the stereoselectivity on the basis of the pentad
distribution analysis, the polymerization behavior is similar to
that directed by a site-control mechanism, and the steric bulk
of the substituent ortho to the phenoxy oxygen plays a crucial
role in determining the stereoselectivity of the catalyst. Thus,
we have given the name “ligand-directed chain-end control” to
this unique stereospecific polymerization behavior. “Ligand-
directed chain-end control” behavior can be explained well by
the recently proposed site-inversion mechanism.
Propylene Polymerization under Pressurized Conditions. Pres-
surized polymerization was carried out at 25 °C under 0.6 MPa
propylene pressure in a 1000 mL stainless steel reactor equipped with
a propeller-like stirrer. Toluene (380 mL) was introduced into the reactor
under propylene at atmospheric pressure. Propylene was pumped into
the reactor up to 0.6 MPa pressure with stirring (350 rpm) at 25 °C.
Polymerization was initiated by adding a 1.25 M MAO solution in
toluene (2 mL, 2.5 mmol) and then a 0.002 M complex solution in
toluene (5.0 mL, 10 µmol) into the reactor. After a prescribed time,
the polymerization was quenched by injection of methanol (5 mL).
The reactor was vented and the resulting mixture was added to acidified
methanol (1500 mL containing 5 mL of concentrated HCl). The polymer
was collected by filtration, washed with methanol (200 mL × 2), and
dried in vacuo at 80 °C for 10 h.
The Ti complexes reported herein (fluorinated Ti-FI Catalysts)
represent a new class of living propylene polymerization
catalysts capable of mediating highly stereospecific, thermally
robust propylene polymerization.
Acknowledgment. We would like to thank Dr. M. Mullins
for fruitful discussions and suggestions. We are also grateful
to Y. Yoshida, Y. Suzuki, N. Matsukawa, S. Matsuura, Y. Inoue,
S. Matsui, Y. Takagi, Y. Nakayama, H. Bando, Y. Tohi, H.
Makio, N. Urakawa, S. Kojoh, T. Matsugi, and J. S. Harp for
their research and technical assistance. We would also like to
thank M. Onda for NMR analysis and T. Abiru for GPC
analysis.
Conclusions
The catalytic properties of Ti complexes having fluorine-
containing phenoxy-imine chelate ligands (Ti-FI Catalysts) for
the polymerization of propylene have been described. These Ti
complexes (with MAO activation) initiate the living polymer-
Supporting Information Available: Detailed synthesis data
for complexes 3 and 5-8, full crystallographic data for
complexes 4 and 6, DFT calculation results, and polymer
analysis data (NMR, DSC). See any current masthead page for
ordering information and Web access instructions.
(41) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Burla, M.
C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435-435.
(42) PATTY: Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bosman, W. P.;
Garcia-Granda, S.; Gould, R. O.; Smits, J. M. M.; Smykalla, C. The DIRDIF
program system, Technical Report of the Crystallography Laboratory,
University of Nijmegen, The Netherlands, 1992.
(43) TeXsan; Molecular Structure Corp., 9009 New Trails Drive, The Woodlands,
TX 77381-5209, 1992.
JA029560J
9
J. AM. CHEM. SOC. VOL. 125, NO. 14, 2003 4305