Reaction of vinyl chloride with group 4 metal olefin polymerization catalysts

Article abstract of DOI:10.1021/ja028530d

The reactions of three types of group 4 metal olefin polymerization catalysts, (C₅R₅)₂ZrX₂/activator, (C₅Me₅)TiX₃/MAO (MAO = methylalumoxane), and (C₅Me₄SiMe₂N^tBu) MX₂/activator (M = Ti, Zr), with vinyl chloride (VC) and VC/propylene mixtures have been investigated. Two general pathways are observed: (i) radical polymerization of VC initiated by radicals derived from the catalyst and (ii) net 1,2 VC insertion into L_nMR⁺ species followed by β-Cl elimination, rac-(EBI)ZrMe(μ-Me)B(C₆F₅)₃. (EBI = 1,2-ethylenebis-(indenyl)) reacts with 2 equiv of VC to yield oligopropylene, rac-(EBI)ZrCl₂, and B(C₆F₅)₃. This reaction proceeds by net 1,2 VC insertion into rac-(EBI)ZrMe⁺ followed by fast β-Cl elimination to yield [rac-(EBI)- ZrCl][MeB(C₆F₅)₃] and propylene. Methylation of rac-(EBI)ZrCl⁺ by MeB(C₆F₅)₃- enables a second VC insertion/β-Cl elimination to occur. The evolved propylene is oligomerized by rac-(EBI)ZrR⁺ as it is formed. At high Al/Zr ratios, rac-(EBI)ZrMe₂/MAO catalytically converts VC to oligopropylene by 1,2 VC insertion into rac-(EBI)ZrMe⁺, β-Cl elimination, and realkylation of rac-(EBI)ZrCl⁺ by MAO; this process is stoichiometric in Al-Me groups. The evolved propylene is oligomerized by rac-(EBI)ZrR⁺. Oligopropylene end group analysis shows that the predominant chain transfer mechanism is VC insertion/β-Cl elimination/realkylation. In the presence of trace levels of O₂, rac-(EBI)ZrMe₂/MAO polymerizes VC to poly(vinyl chloride) (PVC) by a radical mechanism initiated by radicals generated by autoxidation of Zr-R and/or Al-R species. Cp*TiX₃/MAO (Cp* = C₅Me₅; X = OMe, Cl) initiates radical polymerization of VC in CH₂Cl₂ solvent at low Al/Ti ratios under anaerobic conditions; in this case, the source of initiating radicals is unknown. Radical VC polymerization can be identified by the presence of terminal and internal allylic chloride units and other radical defects in the PVC which arise from the characteristic chemistry of PCH₂CHCl^? macroradicals. However, this test must be used with caution, since the defect units can be consumed by postpolymerization reactions with MAO. (C₅Me₄SiMe₂N^tBu) MMe₂/[Ph₃C]][B(C₆F₅)₄] catalysts (M = Ti, Zr) react with VC by net 1,2 insertion/β-Cl elimination, yielding [(C₅Me₄SiMe₂N^tBu) MCl][B(C₆F₅)₄] species which can be trapped as (C₅Me₄SiMe₂N^tBu)MCI₂ by addition of a chloride source. The reaction of rac-(EBI)ZrMe₂/MAO or [(C₅Me₄- SiMe₂N^tBu)ZrMe][B(C₆ F₅)₄] with propylene/VC mixtures yields polypropylene containing both allylic and vinylidene unsaturated chain ends rather than strictly vinylidene chain ends, as observed in propylene homopolymerization. These results show that the VC insertion of L_nM(CH₂CHMe)_nR⁺ species is also followed by β-Cl elimination, which terminates chain growth and precludes propylene/VC copolymerization. Termination of chain growth by β-Cl elimination is the most significant obstacle to metal-catalyzed insertion polymerization/copolymerization of VC.

Full text of DOI:10.1021/ja028530d

Published on Web 12/19/2002
Reaction of Vinyl Chloride with Group 4 Metal Olefin
Polymerization Catalysts
Robert A. Stockland, Jr., Stephen R. Foley, and Richard F. Jordan*
Contribution from the Department of Chemistry, The UniVersity of Chicago,
5735 South Ellis AVenue, Chicago, Illinois 60637
Received September 12, 2002 ; E-mail: rfjordan@uchicago.edu
Abstract: The reactions of three types of group 4 metal olefin polymerization catalysts, (C5R5)2ZrX2/activator,
(C5Me5)TiX3/MAO (MAO ) methylalumoxane), and (C5Me4SiMe2NtBu)MX2/activator (M ) Ti, Zr), with vinyl
chloride (VC) and VC/propylene mixtures have been investigated. Two general pathways are observed:
(i) radical polymerization of VC initiated by radicals derived from the catalyst and (ii) net 1,2 VC insertion
into LnMR+ species followed by â-Cl elimination. rac-(EBI)ZrMe(µ-Me)B(C6F5)3 (EBI ) 1,2-ethylenebis-
(indenyl)) reacts with 2 equiv of VC to yield oligopropylene, rac-(EBI)ZrCl2, and B(C6F5)3. This reaction
proceeds by net 1,2 VC insertion into rac-(EBI)ZrMe+ followed by fast â-Cl elimination to yield [rac-(EBI)-
ZrCl][MeB(C6F5)3] and propylene. Methylation of rac-(EBI)ZrCl+ by MeB(C6F5)3 enables a second VC
-
insertion/â-Cl elimination to occur. The evolved propylene is oligomerized by rac-(EBI)ZrR+ as it is formed.
At high Al/Zr ratios, rac-(EBI)ZrMe2/MAO catalytically converts VC to oligopropylene by 1,2 VC insertion
into rac-(EBI)ZrMe+, â-Cl elimination, and realkylation of rac-(EBI)ZrCl+ by MAO; this process is
stoichiometric in Al-Me groups. The evolved propylene is oligomerized by rac-(EBI)ZrR+. Oligopropylene
end group analysis shows that the predominant chain transfer mechanism is VC insertion/â-Cl elimination/
realkylation. In the presence of trace levels of O2, rac-(EBI)ZrMe2/MAO polymerizes VC to poly(vinyl chloride)
(PVC) by a radical mechanism initiated by radicals generated by autoxidation of Zr-R and/or Al-R species.
Cp*TiX3/MAO (Cp* ) C5Me5; X ) OMe, Cl) initiates radical polymerization of VC in CH2Cl2 solvent at low
Al/Ti ratios under anaerobic conditions; in this case, the source of initiating radicals is unknown. Radical
VC polymerization can be identified by the presence of terminal and internal allylic chloride units and other
“radical defects” in the PVC which arise from the characteristic chemistry of PCH2CHCl• macroradicals.
However, this test must be used with caution, since the defect units can be consumed by postpolymerization
reactions with MAO. (C5Me4SiMe2NtBu)MMe2/[Ph3C]][B(C6F5)4] catalysts (M ) Ti, Zr) react with VC by net
1,2 insertion/â-Cl elimination, yielding [(C5Me4SiMe2NtBu)MCl][B(C6F5)4] species which can be trapped as
(C5Me4SiMe2NtBu)MCl2 by addition of a chloride source. The reaction of rac-(EBI)ZrMe2/MAO or [(C5Me4-
SiMe2NtBu)ZrMe][B(C6F5)4] with propylene/VC mixtures yields polypropylene containing both allylic and
vinylidene unsaturated chain ends rather than strictly vinylidene chain ends, as observed in propylene
homopolymerization. These results show that the VC insertion of LnM(CH2CHMe)nR+ species is also followed
by â-Cl elimination, which terminates chain growth and precludes propylene/VC copolymerization.
Termination of chain growth by â-Cl elimination is the most significant obstacle to metal-catalyzed insertion
polymerization/copolymerization of VC.
Introduction
lymerization of acrylates and vinyl ketones with ethylene and
propylene, a general solution to this problem remains elusive.3
Here, we report studies directed to the long-term goal of
developing metal catalysts for the insertion polymerization/
copolymerization of vinyl chloride (VC).
Poly(vinyl chloride) (PVC) and related copolymers are pre-
pared commercially by radical polymerization of VC and can
also be prepared using RLi or R2Mg initiators.4-6 PVC is an
important commercial material because of its low cost, useful
properties (e.g., good chemical resistance, good resistance to
The polymerization of olefins by insertion chemistry using
Ziegler-Natta, Cr-based or, more recently, single-site metal
catalysts provides a powerful approach to the synthesis of
polyolefins with a high degree of control of polymer composi-
tion and structure.1 A current challenge in this area is to develop
catalysts capable of polymerizing or copolymerizing polar
olefins, particularly CH2dCHX monomers with functional
groups directly bonded to the olefin unit, by insertion mecha-
nisms.2 While some success has been achieved in the copo-
(1) (a) Boor, J. Ziegler-Natta Catalysts and Polymerizations; Academic
Press: New York, 1979. (b) McDaniel, M. P. AdV. Catal. 1985, 33, 48.
(c) Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt, R.; Rieger, B.; Waymouth,
R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143. (d) Kaminsky, W.;
Arndt, M. AdV. Polym. Sci. 1997, 127, 143. (e) Britovsek, G. J. P.; Gibson,
V. C.; Wass, D. F. Angew. Chem., Int. Ed. 1999, 38, 428.
(2) Boffa, L. S.; Novak, B. M. Chem. ReV. 2000, 110, 3728.
(3) (a) Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J. Am. Chem.
Soc. 1998, 120, 888. (b) Johnson, L. K.; Mecking, S. F.; Brookhart, M. J.
Am. Chem. Soc. 1996, 118, 267. (c) Stibrany, R. T.; Schulz, D. N.; Kacker,
S.; Patil, A. O.; Baugh, L. S.; Rucker, S. P.; Zushma, S.; Berluche, E.;
Sissano, J. A. Polym. Mater. Sci. Eng. 2002, 86, 325.
9
796
J. AM. CHEM. SOC. 2003, 125, 796-809
10.1021/ja028530d CCC: $25.00 © 2003 American Chemical Society

Metal-Catalyzed Insertion Polymerization of Vinyl Chloride
A R T I C L E S
Scheme 1
involves H atom abstraction from a backbone methylene group
by the propagating radical and subsequent Cl transfer to the
monomer. Tertiary chloride units are formed from backbone
-CHCl- units by intramolecular (back-biting) or intermolecular
H atom abstraction by propagating radicals, followed by
monomer addition. PVC undergoes thermal degradation above
∼100 °C via autocatalytic dehydrochlorination and concomitant
formation of polyene units, necessitating the use of stabilizers
for processing and end use.4,7 The internal allylic and tertiary
chloride groups are the initiating sites for HCl loss. It has been
shown that the chemical derivatization of the defect sites by
alkylation with AlMe3 or other reactions provides PVCs with
enhanced thermal stability.9,10 Metal-catalyzed insertion polym-
erization of VC may provide a direct approach to “low defect”
PVC. It should also be noted that the defect sites, which can be
identified by NMR,6,8 are diagnostic for the free radical
polymerization mechanism.
As for most free radical polymers, manipulation of PVC
composition and microstructure is limited because it is difficult
to modify the reactivity of the propagating macroradicals.11 The
development of insertion polymerization routes to PVC and VC
copolymers may enable the control of tacticity, molecular
weight, branching, end group structures, and other polymer prop-
erties through the tailoring of catalyst structure, as in conven-
tional olefin polymerization. Target polymers of interest include
stereoregular PVCs,12 PVCs with tailored molecular weight
distributions,13 and linear copolymers of VC with ethylene or
styrene prepared by direct synthesis.
burning), compatibility with a wide range of modifiers, and vers-
atility in many applications. PVC homopolymer is an amor-
phous, rigid thermoplastic, and flexible PVCs can be prepared
by plasticization, typically using phthalates. Important VC
copolymers include VC-vinyl acetate, VC-acrylonitrile, and
vinylidene chloride-VC copolymers, which are prepared com-
mercially by radical copolymerization, and chlorinated poly-
ethylene, which is prepared by the chlorination of polyethylene.
Interest in developing nonradical syntheses of PVC materials
is driven by practical as well as fundamental reasons. A key
practical issue is the control of PVC “defect sites”.7 Chain
growth in radical VC polymerization occurs by the head-to-tail
addition of active macroradicals to the monomer. However,
several side reactions compete with chain growth and lead to
the formation of the branch and “defect” structures illustrated
in Scheme 1.8 As summarized by Starnes et al.,6,7 the head-to-
head addition of VC to the growing chain leads to the formation
of chloromethyl and 1,2-dichloroethyl branches. Allylic chloride
end groups are formed by head-to-head addition followed by a
1,2-Cl shift and Cl transfer to the monomer. Internal allylic
chloride units are generated by chain transfer to polymer, which
Two modes of insertion polymerization or copolymerization
of VC can be envisioned: 1,2 insertion or 2,1 insertion (Scheme
2). However, inspection of these reactions reveals several poten-
tial problems. A 1,2 insertion of VC into an LnMR active species
would generate an MCH2CH2ClR complex, which is anticipated
to be prone to â-Cl elimination.14 In fact, transition metal alkyls
containing â-halogens are extremely rare, being limited to a
few octahedral d6 complexes, such as PtCl5(CH2CH2Cl)2- 15
,
Pt(2,9-Me2-phen)Cl3(CH2CH2Cl),16 and Ir(PMe2Ph)2(CO)-
Br2(CH2CH2Br),17 several more highly halogenated complexes,18
(9) Rogestedt, M.; Hjertberg, T. Macromolecules 1993, 26, 60.
(10) (a) Pourahmady, N.; Bak, P. I.; Kinsey, R. A. J. Macromol. Sci., Chem.
1992, 29, 959. (b) Starnes, W. H., Jr.; Plitz, I. M. Macromolecules 1976,
9, 633. (c) Pi, Z.; Kennedy, J. P. J. Polym. Sci., Part A: Polym. Chem.
2001, 39, 312.
(11) For the living radical polymerization of VC, see: (a) Percec, V.; Popov,
A. V.; Ramirez-Castillo, E.; Monteiro, M.; Barboiu, B.; Weichold, O.;
Asandei, A. D.; Mitchell, C. M. J. Am. Chem. Soc. 2002, 124, 4940. (b)
Asandei, A. D.; Percec, V. J. Polym. Sci., Part A: Polym. Chem. 2001,
39, 3392. (c) Kaida, Y.; Yammamoto, H. JP 10130306 (Asahi Glass Co.).
Chem. Abstr. 1998, 129:41526. (d) Bak, P. I.; Bidinger, G. P.; Cozens, R.
J.; Klich, P. R.; Mayer, L. A. EP 617057 (Geon Co.) Chem. Abstr. 1994,
122:266335.
(12) (a) Free radical PVC is normally atactic, although a moderately syndiotactic
polymer is formed at low temperature and highly syndiotactic PVC has
been prepared by urea clathrate polymerization at low temperature. Highly
isotactic PVC is unknown. See ref 4 and (b) Cais, R. E.; Brown, W. L.
Macromolecules 1980, 13, 801. (c) King, J.; Bower, D. I.; Maddams, W.
F.; Pyszora, H. Makromol. Chem. 1983, 184, 879. (d) Talamani, G.; Vidotto,
G. Makromol. Chem. 1967, 100, 48. (e) Minagawa, M.; Narisawa, I.;
Suguisawa, H.; Hasegawa, S.; Yoshii, F. J. Appl. Polym. Sci. 1999, 74,
2820 and references therein. (f) White, D. M. J. Am. Chem. Soc. 1960, 82,
5678.
(13) (a) The molecular weight (Mn) of PVC is typically between 30 000 and
80 000 and is controlled by chain transfer to the monomer. High molecular
weight PVC cannot be prepared easily at high temperature. Xie, T. Y.;
Jamielec, A. E.; Wood, P. E.; Woods, D. R. Polymer 1991, 32, 537. (b)
Low molecular weight PVC can be produced using chain transfer agents.
Yamamoto, K.; Maehara, T.; Mitani, K.; Mizutani, Y. J. Appl. Polym. Sci.
1994, 51, 749.
(4) (a) Odian, G. Principles of Polymerization; Wiley & Sons: New York,
1991; pp 308-310 and pp 712-713. (b) Vinyl Chloride and Poly(Vinyl
Chloride). Kirk-Othmer Encyclopedia of Chemical Technology, 3rd ed.;
Wiley & Sons: New York, 1983; Vol. 23, pp 865-936. (c) Kirk-Othmer
Encyclopedia of Chemical Technology Home Page. http://www.mrw.
interscience.wiley.com/kirk (accessed July 2002). (d) Alger, M. Polymer
Science Dictionary, 2nd ed.; Chapman Hall: New York, 1997; pp 462-
463.
(5) (a) Endo, K.; Kaneda, N.; Waku, H. Polymer 1999, 40, 6883. (b) Jisova,
V.; Kolinsky, M.; Lim, D. J. Polym. Sci., Polym. Chem. Ed. 1970, 8, 1525.
(c) Guyot, A.; Mordini, J. J. Polym. Sci., Part C: Polym. Symp. 1971, 33,
65.
(6) (a) Benedikt, G. M.; Cozens, R. J.; Goodall, B. L.; Rhodes, L. F.; Bell, M.
N.; Kemball, A. C.; Starnes, W. H., Jr. Macromolecules 1997, 30, 10.
(7) Starnes, W. H., Jr.; Girois, S. Polym. Yearb. 1995, 12, 105 and references
therein.
(8) (a) Starnes, W. H., Jr.; Zaikov, V. G.; Chung, H.; Wojciechowski, B. J.;
Tran, H. V.; Saylor, K. Macromolecules 1998, 31, 1508. (b) Starnes, W.
H., Jr.; Chung, H.; Wojciechowski, B. J.; Skillicorn, D. E.; Benedikt, G.
M. AdV. Chem. Ser. 1996, 249, 3. (c) Starnes, W. H., Jr.; Wojciechowski,
B. J.; Chung, H. Macromolecules 1995, 28, 945. (d) Benedikt, G. M.;
Goodall, B. L.; Rhodes, L. F.; Kemball, A. C. Macromol. Symp. 1994, 86,
65. (e) Starnes, W. H., Jr.; Wojciechowski, B. J. Makromol. Chem.,
Macromol. Symp. 1993, 70/71, 1. (f) Starnes, W. H., Jr.; Wojciechowski,
B. J.; Velazquez, A.; Benedikt, G. M. Macromolecules 1992, 25, 3638.
(g) Llauro-Darricades, M. F.; Bensemra, N.; Guyot, A.; Petiaud, R.
Makromol. Chem., Macromol. Symp. 1989, 29, 171. (h) Starnes, W. H.,
Jr.; Schilling, F. C.; Plitz, I. M.; Cais, R. E.; Freed, D. J.; Hartless, R. L.;
Bovey, F. A. Macromolecules 1983, 16, 790. (i) Starnes, W. H., Jr.;
Schilling, F. C.; Abbas, K. B.; Cais, R. E.; Bovey, F. A. Macromolecules
1979, 12, 556.
(14) For a review on halogenated alkyl complexes, see Friedrich, H. B.; Moss,
J. R. AdV. Organomet. Chem. 1991, 33, 235.
(15) Halpern, J.; Jewsbury, R. A. J. Organomet. Chem. 1979, 181, 223.
9
J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003 797

A R T I C L E S
Stockland et al.
Scheme 2
MCHClCH2R species may be susceptible to nucleophilic
displacement of the R-chloride27 and, in non-d0-metal cases, may
undergo R-Cl elimination to generate carbene complexes.28
Additionally, the electron-withdrawing R-Cl will decrease the
nucleophilic character and migratory aptitude of the MCHClCH2R
group, inhibiting subsequent insertions.29 However, CO insertion
of Co(CO)3LCH2Cl (L ) CO, PPh3) has been observed.30
On the other hand, because VC is coordinated by the CdC
bond rather than the chlorine in known metal VC complexes,31
chlorocarbons are comparatively weak ligands,32 and sp2 C-X
bonds generally are not prone to nucleophilic substitution;33 side
reactions between VC and LnMR species leading to catalyst
poisoning are expected to be less important than for more
reactive monomers such as acrylates, vinyl esters, or acryloni-
trile. Numerous literature reports and patents describe the
polymerization of VC and the copolymerization of VC and
olefins, by Ziegler-Natta or single-site olefin polymerization
catalysts.34-37 In some cases, these reactions were recognized
as radical polymerizations, while, in other cases, nonradical
(26) (a) Hubbard, J. L.; Morneau, A.; Burns, R. M.; Nadeau, O. W. J. Am.
Chem. Soc. 1991, 113, 9180. (b) Ogino, H.; Shoji, M.; Abe, Y.; Shimura,
M.; Shioni, M. Inorg. Chem. 1987, 26, 2542. (c) Friedrich, H. B.; Moss, J.
R. J. Organomet. Chem. 1993, 453, 85. (d) Moss, J. R.; Niven, M. L.;
Stretch, P. M. Inorg. Chim. Acta 1986, 119, 177. (e) Haarman, H. F.;
Ernsting, J. M.; Kranenburg, M.; Kooijman, H.; Veldman, N.; Spek, A.
L.; van Leeuwen, P. W. N. M.; Vrieze, K. Organometallics 1997, 16, 887.
(f) Bergamini, P.; Bortolini, O.; Costa, E.; Pringle, O. G. Inorg. Chim.
Acta 1996, 252, 33. (g) Bradd, K. J.; Heaton, B. T.; Jacob, C.; Sampanthar;
J. T.; Steiner, A. J. Chem. Soc., Dalton Trans. 1999, 1109. (h) Leoni, P.
Organometallics 1993, 12, 2431. (i) Leoni, P.; Marchetti, F.; Paoletti, M.
Organometallics 1997, 16, 2146. (j) Mashima, K.; Fukumoto, A.; Nakano,
H.; Kaneda, Y.; Tani, K.; Nakamura, A. J. Am. Chem. Soc. 1998, 120,
12151. (k) Do¨hring, A.; Goddard, R.; Hopp, G.; Jolly, P. W.; Kokel, N.;
Kru¨ger, C. Inorg. Chim. Acta 1994, 222, 179.
(27) (a) McCrindle, R.; Ferguson, G.; McAlees, A. J.; Arsenault, G. J.; Gupta,
A.; Jennings, M. C. Organometallics 1995, 14, 2741. (b) Hubbard, J. L.;
McVicar, W. K. J. Organomet. Chem. 1992, 429, 369. (c) Werner, H.;
Wilfried, P.; Feser, R.; Zolk, R.; Thometzek, P. Chem. Ber. 1985, 118,
261. (d) Hubbard, J. L.; McVicar, W. K. Organometallics 1990, 9, 2683.
(28) (a) Wilhelm, T. E.; Belderrain, T. R.; Brown, S. N.; Grubbs, R. H.
Organometallics 1997, 16, 3867. (b) Belderrain, T.; Grubbs, R. H.
Organometallics 1997, 16, 4001. (c) Goldman, A. S.; Tyler, D. R.
Organometallics 1984, 3, 449. (d) Yang, M. X.; Sarkar, S.; Bent, B. E.;
Bare, S. R.; Holbrook, M. T. Langmuir 1997, 13, 229.
and perfluoroalkyl complexes.19 Early metal MCH2CHRX
species are particularly susceptible to â-X elimination.20 For
example, Caulton recently reported that Cp2ZrHCl reacts with
vinyl fluoride by 1,2 insertion and â-F elimination yielding
Cp2ZrFCl and ethylene,21 and Wolczanski showed that
(tBu3SiO)3TaH2 reacts with CH2dCHX (X ) F, Cl, Br) via
1,2 insertion and fast â-X elimination.22,23 Additionally, Boone
and co-workers recently reported that VC reacts with the active
Fe-alkyl species in (pyridine-bisimine)FeCl2/MAO-catalyzed
ethylene polymerization by 1,2 insertion followed by â-Cl
elimination.24
A 2,1 VC insertion would generate an MCHClCH2R species
which obviously cannot undergo â-Cl elimination without
prior rearrangement (Scheme 2). Metal alkyls containing
R-halogen substituents are, in fact, quite common.25,26 However,
(29) Axe, F. U.; Marynick, D. S. J. Am. Chem. Soc. 1988, 110, 3728.
(30) Galamb, V.; Pa´lyi, G.; Boese, R.; Schmid, G. Organometallics 1987, 6,
861.
(16) Fanizzi, F. P.; Maresca, L.; Pacifico, C.; Natile, G.; Lanfranchi, M.;
Tiripicchio, A. Eur. J. Inorg. Chem. 1999, 1351.
(31) (a) Alt, H. G.; Englehardt, H. E. J. Organomet. Chem. 1988, 346, 211. (b)
Oskam, A.; Shoemaker, G.; Rest, A. J. J. Organomet. Chem. 1986, 304,
C13. (c) Drouin, M.; Harrod, J. F. Can. J. Chem. 1985, 63, 353. (d) Jesse,
A. C.; Gijben, H. P.; Stufkens, D. J.; Vrieze, K. Inorg. Chim. Acta 1978,
31, 203. (e) Jesse, A. C.; Stufkens, D. J.; Vrieze, K. Inorg. Chim. Acta
1979, 32, 87. (f) Bigorgne, M. J. Organomet. Chem. 1977, 127, 55. (g)
Ashley-Smith, J.; Douek, Z.; Johnson, B. F. G.; Lewis, J. J. Chem. Soc.,
Dalton Trans. 1974, 128. (h) Tolman, C. A. J. Am. Chem. Soc. 1974, 96,
2780. (i) Quinn, H. W.; VanGilder, R. L. Can. J. Chem. 1971, 49, 1323.
(j) Cramer, R. J. Am. Chem. Soc. 1967, 89, 4621.
(17) Deeming, A. J.; Shaw, B. L. J. Chem. Soc. 1971, 376.
(18) Schmidt, E. K. G. J. Organomet. Chem. 1981, 204, 393.
(19) (a) Churchill, M. R.; Fennessey, J. P. Inorg. Chem. 1967, 6, 1213. (b)
Naumann, D.; Kaiser, M. Z. Anorg. Allg. Chem. 1995, 621, 812. (c) Hensley,
D. W.; Stewart, R. P., Jr. Inorg. Chem. 1978, 17, 905. (d) Burns, R. J.;
Bulkowski, P. B.; Stevens, S. C. V.; Baird, M. C. J. Chem. Soc., Dalton
Trans. 1974, 4, 415. (e) King, R. B.; Bond, A. J. Am. Chem. Soc. 1974,
96, 1334. (f) Hughes, R. P.; Overby, J. S.; Williamson, A.; Lam, K. C.;
Concolino, T. E.; Rheingold, A. L. Organometallics 2000, 19, 5190. (g)
Hughes, R. P.; Sweetser, J. T.; Tawa, M. D.; Williamson, A.; Incarvito, C.
D.; Rhatigan, B.; Rheingold, A. L.; Ross, G. Organometallics 2001, 20,
3800.
(32) For reviews and recent studies of halocarbon complexes, see (a) Kulawiec,
R. J.; Crabtree, R. H. Coord. Chem. ReV. 1990, 99, 89. (b) Huang, D.;
Bollinger, J. C.; Streib, W. E.; Folting, K.; Young, V., Jr.; Eisenstein, O.;
Caulton, K. G. Organometallics 2000, 19, 2281. (c) Huhmann-Vincent, J.;
Scott, B. L.; Kubas, G. J. Inorg. Chem. 1999, 38, 115. (d) Arndtsen, B. A.;
Bergman, R. G. Science 1995, 27, 1970.
(20) The reaction of (C5Me5)2ScMe with vinyl fluoride yields (C5Me5)2ScF and
propylene via insertion and subsequent â-F elimination. Burger, B. Thesis,
California Institute of Technology, 1987.
(21) Watson, L. A.; Yandulov, D. V.; Caulton, K. G. J. Am. Chem. Soc. 2001,
123, 603.
(22) Strazisar, S. A.; Wolczanski, P. T. J. Am. Chem. Soc. 2001, 123, 4728.
(23) For other reactions of VC with early metal complexes, see (a) Takahashi,
T.; Kotora, M.; Fischer, R.; Nishihara, Y.; Nadajima, K. J. Am. Chem.
Soc. 1995, 117, 11039. (b) Hanzawa, Y.; Kowase, N.; Taguchi, T.
Tetrahedron Lett. 1998, 39, 583.
(24) Boone, H. W.; Athey, P. S.; Mullins, M. J.; Philipp, D.; Muller, R.; Goddard,
W. A. J. Am. Chem. Soc. 2002, 124, 8790.
(25) (a) McCrindle, R.; Arsenault, G. J.; Farwaha, R.; McAlees, A. J.; Sneddon,
D. W. J. Chem. Soc,. Dalton Trans. 1989, 761. (b) McCrindle, R.; Arsenault,
G. J.; Farwaha, R.; Hampden-Smith, M. J.; Rice, R. E.; McAlees, A. J.
Chem. Soc., Dalton Trans. 1988, 1773. (c) McCrindle, R.; Arsenault, G.
J.; Gupta, A.; Hampden-Smith, M. J.; Rice, R. E.; McAlees, A. J. J. Chem.
Soc., Dalton Trans. 1991, 949. (d) Ferguson, G.; Li, Y.; McAlees, A. J.;
McCrindle, R.; Xiang, K. Organometallics 1999, 18, 2428. (e) Ferguson,
G.; Gallagher, J. F.; McAlees, A. J.; McCrindle, R. Organometallics 1997,
16, 1053.
(33) March, J. AdVanced Organic Chemistry; John Wiley and Sons: New York,
1985; pp 295-298.
(34) (a) Giannini, U.; Cesca, S. Chim. Ind. (Milan) 1962, 44, 371. (b) Yamazaki,
N.; Sasaki, K.; Kambara, J. J. Polym. Sci. B 1964, 2, 487. (c) Misono, A.;
Uchida, Y.; Yamada, K. Bull. Chem. Soc. Jpn. 1967, 40, 2366. (d) Higashi,
H.; Watabe, K.; Namikawa, S. J. Polym. Sci. B 1967, 5, 1125. (e) Misono,
A.; Uchida, Y.; Yamada, K.; Saeki, T. Bull. Chem. Soc. Jpn. 1968, 41,
2995. (f) Ulbricht, J.; Jurgen, G.; Manfred, G. Angew. Makromol. Chem.
1968, 69. (g) Guyot, A.; Rocaniere, P. J. Appl. Polym. Sci. 1969, 13, 2019.
(h) Pham, Q. T.; Rocaniere, P.; Guyot, A. J. Appl. Polym. Sci. 1970, 14,
1291. (i) Furukawa, J. Pure Appl. Chem. 1971, 26, 153. (j) Saito, M.;
Suzuki, Y. J. Polym. Sci., Polym. Chem. Ed. 1971, 9, 3639. (k) Haszeldine,
R. N.; Hyde, T. G.; Tait, P. J. T. Polymer 1973, 14, 215. (l) Mejzlik, J.;
Navratil, M.; Vilimova, L. Collect. Czech. Chem. Commun. 1973, 38, 3457.
(m) Haszeldine, R. N.; Hyde, T. G.; Tait, P. J. T. J. Polym. Sci., Polym.
Chem. Ed. 1974, 12, 1703. (n) Florjanczyk, Z.; Kuran, W.; Sitkowska, J.;
Ziolkowsko, A. J. Polym. Sci., Part A: Polym. Chem. 1987, 25, 343. (o)
Belov, G. P.; Solovyova, T. I.; Smirnov, V. I. Acta Polym. 1990, 41, 178.
(p) Yamazaki, N. Prog. Polym. Sci. Jpn. 1971, 2, 171.
9
798 J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003

Metal-Catalyzed Insertion Polymerization of Vinyl Chloride
A R T I C L E S
able rac-(EBI)Zr indenyl H2 or H3 resonances are present in
the ambient temperature 500 MHz H NMR spectrum of the
Chart 1
1
red solution after stage 1. However, low-temperature spectra
reveal that [{rac-(EBI)ZrCl}2(µ-Cl)][MeB(C6F5)3] (2, ∼10:1
mixture of diastereomers) is the major Zr-containing species
present (94% vs 1). The dinuclear cation of 2 is an adduct
of rac-(EBI)ZrCl+ and rac-(EBI)ZrCl2 (3). The identity of 2
was confirmed by treatment with 0.5 equiv of [NBu3Bn]Cl
(CD2Cl2, 25 °C, 1 min) to quantitatively generate 3; however,
mechanisms were suggested. However, to date, in no case has
an insertion mechanism been established.
1
2 could not be isolated in pure form. The H NMR signals for
2 are sharp below -50 °C but broaden into the baseline at
ambient temperature because of bridge/terminal Cl exchange.41
Another intermediate, assigned as rac-(EBI)Zr(Cl)(µ-Me)-
B(C6F5)3 (4), was detected when the reaction of 1 and VC at
25 °C was quenched after 1 min by rapid cooling to -78 °C
and then analyzed by low-temperature NMR.
This paper describes studies of the reaction of VC with three
important types of group 4 metal olefin polymerization cata-
lysts: zirconocene catalysts based on (C5R5)2ZrX2 complexes,
half-sandwich Ti catalysts based on Cp*TiX3 complexes (Cp*
) C5Me5), and “constrained geometry” catalysts based on
(C5Me4SiMe2NtBu)MX2 (M ) Ti, Zr) complexes (Chart 1). The
objectives of this work were to determine how these catalysts
react with VC and to identify the mechanism of VC polymer-
ization previously reported with these systems.38 In a subsequent
paper, we will describe parallel studies with representative late-
metal catalysts.39 As will be seen, while VC insertion polym-
erization has not been achieved, the key issues which must be
addressed in order to achieve this long-term goal have been
identified.
In the second stage of the reaction of 1 with VC, which is
complete after 24 h at 25 °C, the red solution of 2 evolves to a
yellow solution (eq 2). NMR studies establish that an additional
0.5 equiv of VC is converted to oligopropylene and that the
1
only significant organometallic products are 3 (100% by H
NMR, 81% isolated) and B(C6F5)3 (only species observed by
19F NMR). The final yield of the oligopropylene determined
Results and Discussion
Reaction of rac-(EBI)Zr(Me)(µ-Me)B(C6F5)3 with VC. The
reaction of rac-(EBI)ZrMe2 with B(C6F5)3 generates the ion pair
rac-(EBI)Zr(Me)(µ-Me)B(C6F5)3 (1, EBI ) 1,2-ethylenebis-
(indenyl)) which is an active olefin polymerization catalyst.40
Compound 1 was chosen for initial studies because it is easy to
generate and is soluble in hydrocarbon and chlorocarbon
solvents, which facilitates NMR studies, and the C2-symmetry
may lead to stereoregular polymer if insertion polymerization
were to occur. The reaction of 1 with VC proceeds in two stages.
In the first stage (eq 1), which is complete within 5 min at 25
°C in CD2Cl2 solution, 1 reacts with VC (16-fold excess) to
produce a dark red solution. NMR studies reveal that 1.5 equiv
of VC is consumed and a 1:1 mixture of B(C6F5)3 and
1
by H NMR is >95% versus consumed VC. Thus, the overall
reaction in stages 1 and 2 is that given in eq 3. Control
experiments establish that neither rac-(EBI)ZrMe2 nor B(C6F5)3
react separately with VC under these conditions. The reaction
of 1 and VC in benzene-d6 proceeds in a similar fashion,
although a higher temperature is required to convert 2 to 3 in
this solvent. This result confirms that the Cl ligands in 2 and 3
are not derived from the CD2Cl2 solvent.
-
MeB(C6F5)3 is formed. No PVC is observed, but atactic
These observations are consistent with the mechanism in
Scheme 3, although the precise series of steps is not known. In
stage 1, VC inserts into the Zr-Me bond of 1 to generate a
transient â-chloropropyl complex [rac-(EBI)ZrCH2CHClMe]-
[MeB(C6F5)3] (5, not observed), which undergoes rapid â-Cl
elimination to yield 4 and propylene (i, ii). Complex 4 undergoes
ligand redistribution to regenerate 1 (0.5 equiv maximum) and
form 3 and B(C6F5)3 (iii). The regenerated 1 consumes an addi-
tional 0.5 equiv of VC (iv) to produce 4 and propylene. Ulti-
mately 4 is trapped by 3 as the dinuclear species 2 (v). VC
polymerization does not occur because 5 undergoes â-Cl
oligopropylene is formed (95% vs consumed VC). No identifi-
(35) (a) Matsukawa, T.; Kiba, R.; Ikeda, T. Japn. Patent 08208736 (Chisso
Corp.). Chem Abstr. 1996, 125:276902. (b) For the use of (C5R5)2ZrCl2/
MAO as a photoactivated initiator for free-radical VC polymerization,
see: Nagy, S. M.; Krishnamurti, R.; Cocoman, M. K.; Opalinski, W. M.;
Smolka, T. F. WO 9923124 A1 (Occidental Chemical). Chem. Abstr. 130:
338530.
(36) (a) Endo, K.; Saitoh, M. Polym. J. 2000, 32, 300. (b) Endo, K.; Kaneda,
N.; Waku, H.; Saitoh, M.; Emori, N. J. Vinyl Addit. Technol. 2001, 7, 177.
(37) Taeji, K.; Uozumi, T.; Soga, K. Polymer Prepr. Jpn. 1995, 44, 2.
(38) Preliminary reports of some aspects of this work have appeared. (a)
Stockland, R. A., Jr.; Jordan, R. F. J. Am. Chem. Soc. 2000, 122, 6315. (b)
Jordan, R. F. Polym. Prepr. (Am. Chem. Soc., DiV. Polym. Chem.) 2001,
42, 829. (c) Stockland, R. A., Jr.; Shen, H.; Foley, S.; Jordan, R. F. Polym.
Mater. Sci. Eng. 2002, 87, 39.
(40) (a) Yang, X.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1994, 116,
10015. (b) Ewen, J. A.; Elder, M. Makromol. Chem., Macromol. Symp.
1993, 66, 179. (c) For generation of 1, see: Dagorne, S.; Rodewald, S.;
Jordan, R. F. Organometallics 1997, 16, 5541.
(41) The variable temperature (-73 °C to 25 °C) 1H NMR spectra of 2 are
unaffected by the addition of 3 (5 equiv), which implies that the fluxional
process involves intramolecular bridge/terminal chloride exchange rather
than reversible cleavage of {rac-(EBI)ZrCl}2(µ-Cl)+ to rac-(EBI)ZrCl2 and
rac-(EBI)ZrCl(CD2Cl2)+. Compound 2 decomposes above 40 °C to 3 and
other unidentified products which precluded high-temperature NMR studies.
(39) Foley, S. R.; Stockland, R. A., Jr.; Shen, H.; Jordan, R. F., manuscript in
preparation.
9
J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003 799

A R T I C L E S
Stockland et al.
Scheme 3 a
Scheme 4 a
a Zr ) rac-(EBI)Zr.
propylene and rac-(EBI)ZrCl+, and realkylation of the latter
species by MAO (or the AlMe3 therein). The propylene is
oligomerized by rac-(EBI)ZrR+. The mixed Al-Me/Cl product
was not characterized.
The oligopropylene produced in eq 4 and Scheme 4 is a low
molecular weight (Mn ) 500) atactic material which is isolated
as a colorless oil. No chlorine was detected in this material by
elemental analysis. 1H NMR established that the unsatu-
rated end groups of the oligopropylene are primarily allylic
(CH2dCHCH2s, 95%), although a small fraction of vinylidene
chain ends (CH2dCMeCH2s, 5%) was also detected. In
contrast, in the absence of VC, rac-(EBI)ZrMe2/MAO produces
polypropylene with strictly vinylidene unsaturated end groups
at low propylene pressures, as â-H elimination is the dominant
chain transfer process.42 These results indicate that chain transfer
in the propylene oligomerization in Scheme 4 occurs primarily
by VC insertion, â-Cl elimination, and realkylation; that is,
VC/MAO acts as a chain transfer agent. To confirm this point,
rac-(EBI)ZrMe2/MAO-catalyzed batch propylene polymeriza-
tions were carried out under identical conditions in the presence
and absence of VC. In the former case, the unsaturated end
groups of the polymer were exclusively vinylidene, while, in
the latter case, both allylic (76%) and vinylidene (24%) chain
ends were observed.
a Zr ) rac-(EBI)Zr; ArF ) C6F5.
elimination more rapidly than it inserts a second VC. Free
propylene is not observed by 1H NMR monitoring because it is
oligomerized by rac-(EBI)ZrMe+ as fast as it is formed (vi). It
is well established that rac-(EBI)ZrR+ produces low molecular
weight, atactic polypropylene under such monomer-starved
conditions.42 In Scheme 3, i-v sum to eq 1. In the second stage
(eq 2), 2 undergoes slow cleavage/ligand redistribution reactions
to regenerate rac-(EBI)ZrMe+, which in turn reacts with VC,
as in Scheme 1, until all of the Zr/B-Me groups are consumed;
this corresponds to the observed slow consumption of 0.5 equiv
of VC.
Reaction of rac-(EBI)ZrMe2/MAO with VC. The VC
insertion/â-Cl elimination reaction of rac-(EBI)ZrMe+, which
converts VC to propylene, becomes catalytic under conditions
where 4 is continually realkylated. This situation can be achieved
using methylalumoxane (MAO) as the activator. For example,
the reaction of rac-(EBI)ZrMe2/MAO (Al/Zr ) 17 000) with
VC in toluene (25 °C, 20 h) yields atactic oligopropylene (eq
4) with ∼2 000 equiv of VC consumed per Zr.
As VC is present in large excess relative to propylene in eq
4 and Scheme 4, the formation of oligopropylene implies that
rac-(EBI)ZrR+ reacts more rapidly with propylene than with
VC. This reactivity difference probably arises because propylene
coordinates more effectively than VC to rac-(EBI)ZrR+. In other
work, we have shown that propylene binds more strongly than
VC to the model system (C5H4Me)2Zr(OtBu)+.43
Reactions of Other Zirconocene Catalysts with VC. The
reaction of VC with (C5H5)2ZrMe2/B(C6F5)3, (indenyl)2ZrMe2/
(42) (a) Resconi, L.; Fait, A.; Piemontesi, F.; Colonnesi, M.; Rychlicki, H.;
Zeigler, R. Macromolecules 1995, 28, 6667. (b) Busico, V.; Cipullo, R. J.
Am. Chem. Soc. 1994, 116, 9329.
(43) Stoebenau, E. J., III; Jordan, R. F. Abstracts of Papers, 224th National
Meeting of the American Chemical Society, Boston, MA; American
Chemical Society: Washington, DC, 2002; INOR-284.
As shown in Scheme 4, this reaction proceeds by generation
of rac-(EBI)ZrMe+, VC insertion and â-Cl elimination to yield
9
800 J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003

Metal-Catalyzed Insertion Polymerization of Vinyl Chloride
A R T I C L E S
Figure 1. 1H NMR spectrum (400 MHz, THF-d8, 50 °C) of PVC prepared
with rac-(EBI)ZrMe2/MAO/O2 at 50 °C.
B(C6F5)3, or rac-(EBI)ZrMe2/[NHMe2Ph][B(C6F5)4] in CD2Cl2
or C6D5Cl (25 °C, 24 h) generates oligopropylene. No PVC is
formed in these reactions. Also, the reaction of VC with
(C5Me5)2ZrCl2/MAO (Al/Zr ) 1000; chlorobenzene), rac-(EBI)-
ZrMe2/dried MAO (Al/Zr ) 1000; C6D6), or rac-Me2Si(2-Me-
4,5-benz-indenyl)2ZrCl2/MAO (Al/Zr ) 2000; toluene) yields
oligopropylene with no PVC formed. Thus, net 1,2 VC insertion
and fast â-Cl elimination is a general feature of (C5R5)2ZrR+
species and precludes insertion polymerization of VC by
zirconocene catalysts.
Figure 2. 1H NMR spectra (400 MHz, THF-d8, 50 °C, defect region) of
PVC. (a) PVC prepared with rac-(EBI)ZrMe2/MAO/O2 at 50 °C. (b)
Commercial PVC prepared at 58 °C. Assignments: response a, sCHd
CHCH2Cl and sCH2CHdCHCHCls; response b and b′, cis and trans
sCHdCHCH2Cl; response c, sCHClCH2Cl; response d, sCH2Cl branch;
response e, sCH2CH2Cl; response f, 13C satellites from solvent (THF-d8);
response g, sCHCls main chain resonances.
Influence of Oxygen on the Reaction of VC with Zircono-
cene Catalysts. Our observation that (C5R5)2ZrR+ species react
with VC by net 1,2 insertion and fast â-Cl elimination under
anaerobic conditions suggests that previously reported VC
polymerizations mediated by metallocene catalysts are in fact
radical polymerizations.35a One likely source of initiating radicals
is autoxidation of Zr-R or Al-R species by adventitious
oxygen. To probe this possibility, we investigated the influence
of added O2 on the reaction of rac-(EBI)ZrMe2/MAO with VC.
and that AlR3/O2 combinations function as initiators of radical
polymerizations.47
It should be noted the PVC from eq 5 differs from the
conventional radical PVC in several respects: (i) the molecular
1
weight distribution is unusually broad,48 and (ii) the H NMR
The reaction of rac-(EBI)ZrMe2/MAO (Al/Zr ) 250) with
liquid VC at 50 °C in the presence of a trace amount of O2
results in the generation of PVC (∼1% isolated yield, eq 5).
spectrum contains broad peaks assignable to internal olefin units
(δ 5.6, Figure 2) and triplet resonances (δ 0.96, 1.05, and 1.15)
assignable to -CH2CH3 groups (Figure 3).9 These features
reflect postpolymerization reactions between the PVC defect
sites and Al-R species, as discussed below.
Reaction of VC with (C5Me5)TiX3/MAO Catalysts. One
strategy for circumventing the â-Cl elimination reaction of
MCH2CHClR species that precludes VC insertion polymeriza-
tion by zirconocene catalysts is to use a catalyst that inserts
VC in a 2,1 fashion (Scheme 2). Monocyclopentadienyl titanium
catalysts, that is, (C5R5)TiX3/MAO, are of potential interest in
this regard because these systems polymerize styrene to syn-
diotactic polystyrene by a 2,1 insertion mechanism.49 In fact,
NMR data show that the PVC in eq 5 is formed by a radical
mechanism. The tacticity of the PVC produced by rac-(EBI)-
ZrMe2/MAO/O2 determined by 13C{1H} NMR (0.32 rr, 0.51
mr, 0.17 mm; R ) mole fraction of r dyads ) 0.58) is nearly
identical to that of PVC prepared by radical polymerization at
1
58 °C (0.31 rr, 0.51 mr, 0.18 mm; R ) 0.56).44 The H NMR
spectrum of the PVC produced by rac-(EBI)ZrMe2/MAO/O2
is shown in Figure 1. An expansion of the “radical defect” region
(δ 3.5-6.0) is compared to that of PVC prepared by radical
polymerization at 58 °C in Figure 2. Resonances for the defect
sites which are characteristic of a radical polymerization are
clearly evident in both spectra in Figure 2. These results provide
strong evidence that the VC polymerization in eq 5 occurs by
a normal radical mechanism. The initiating radicals have not
been identified, but it is well established that the autoxidation
of (C5R5)2ZrRX, (C5R5)2ZrR2, and AlR3 compounds occurs by
(45) (a) Brindley, P. B.; Hodgson, J. C. J. Organomet. Chem. 1974, 65, 57. (b)
Brindley, P. B.; Scotton, M. J. J. Chem. Soc., Perkin Trans. 2 1981, 419.
(c) Blackburn, T. F.; Labinger, J. A.; Schwartz, J. Tetrahedron Lett. 1975,
3041. (d) Labinger, J. A.; Hart, D. W.; Seibert, W. E.; Schwartz, J. J. Am.
Chem. Soc. 1975, 97, 3851. (e) Lubben, T. V.; Wolczanski, P. T. J. Am.
Chem. Soc. 1987, 109, 424.
(46) (a) Davies, A. G.; Roberts, B. P. J. Chem. Soc. B 1968, 1074. (b) Davies,
A. G.; Roberts, B. P. Acc. Chem. Res. 1972, 5, 387. (c) Davies, A. G.;
Roberts, B. P. In Free Radicals; Kochi, J. K., Ed.: Wiley-Interscience:
New York, 1973; Vol I, pp 547-590. (d) Alexandrov, Y. A.; Chikinova,
N. V. J. Organomet. Chem. 1991, 418, 1.
(47) (a) Bhanu, V. A.; Kishore, K. Chem. ReV. 1991, 91, 99. (b) Milovskaya,
E. B.; Zamoiskaya, L. V.; Kopp, E. L. Russ. Chem. ReV. (Engl. Transl.)
1969, 38, 420.
(48) PVC from eq 5: Mw ) 180 000; Mn ) 36 000; Mw/Mn ) 5.0. Radical
PVC prepared at 58 °C: Mw ) 161 000; Mn ) 74 700; Mw/Mn ) 2.26.
(49) (a) Tomotsu, N.; Ishihara, N.; Newman, T. H.; Malanga, M. T. J. Mol.
Catal. A: Chem. 1998, 128, 167. (b) Ishihara, N.; Kuramoto, M.; Uoi, M.
Macromolecules 1988, 21, 3356. (c) Pellecchia, C.; Pappalardo, D.; Olivia,
L.; Zambelli, A. J. Am. Chem. Soc. 1995, 117, 6593.
a radical chain mechanism involving R• and RO2 radicals45,46
•
(44) PVC tacticity was determined by 13C NMR. See: Nakayama, N.; Aoki,
A.; Hayashi, T. Macromolecules 1994, 27, 63 and references therein.
9
J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003 801

A R T I C L E S
Stockland et al.
Figure 3. 1H NMR spectra (400 MHz, THF-d8, 50 °C, alkyl region) of
PVC. (a) PVC prepared with rac-(EBI)ZrMe2/MAO/O2 at 50 °C. (b) PVC
prepared with Cp*Ti(OMe)3/MAO (6/MAO) at 25 °C. (c) PVC prepared
with CpTiCl3/MAO (7/MAO) at 25 °C. (d) Commercial PVC prepared at
30 °C. Resonances at 1.29 and 0.89 are due to process oil from the MAO
and hexanes (a, b) and initiator residue (d).
Figure 4. 1H NMR spectra (400 MHz, THF-d8, 50 °C, defect region) of
PVC. (a) PVC prepared with Cp*Ti(OMe)3/MAO (6/MAO) at 25 °C. (b)
PVC prepared with Cp*TiCl3/MAO (7/MAO) at 25 °C. (c) Commercial
PVC prepared at 30 °C.
Endo and Saitoh recently reported that Cp*TiX3 catalysts (6,
X ) OMe; 7, X ) Cl) polymerize VC under MAO activation
conditions.36 The key requirements for polymerization were the
use of CH2Cl2 solvent (very low conversions were reported in
toluene) and low Al/Ti ratios (Al/Ti ) 10). The PVC produced
by 7/MAO contained low levels of allylic chloride defects, and
it was noted that some defect sites could have been consumed
by alkylation by MAO. However, 6/MAO produced PVC that
did not contain defects (at the NMR detection limit), and in
this case, “the polymerization was not inhibited although the
polymer yields [were] somewhat decreased” in the presence of
4-tert-butyl-catechol or TEMPO (molar ratio inhibitor/Ti
)1:1), suggesting the operation of a “nonradical” mechanism.
We have also investigated the reaction of Cp*TiX3/MAO
catalysts with VC.
However, our results differ from the previous results in several
respects. First, as illustrated in Figure 4, 1H NMR analysis shows
that the PVC produced by 6/MAO contains normal radical
defects, while that produced by 7/MAO does not. In the latter
case, broad resonances are observed at δ 5.4-5.7, which are
assigned to internal olefin units. Second, as shown in Figure 3,
for the PVC produced by 6/MAO, a triplet at δ 1.05 is observed,
which is assigned to -CH2CH3 groups,9 whereas, for the PVC
produced by 7/MAO, broad resonances at δ 0.85 and 1.25 are
present, which are assigned to -CH2CHMe- units formed by
methylation of main chain -CH2CHCl- units. Third, the
molecular weights are somewhat higher and the molecular
weight distributions are broader (for 6/MAO, Mn ) 27 200 and
Mw/Mn ) 3.3; for 7/MAO, Mn ) 20 200 and Mw/Mn ) 5.4)
than the values reported by Endo and Saitoh (Mn ) 11 000-
25 000; Mw/Mn ) 1.7-2.0), and the polydispersity is higher
for the PVC produced by 7/MAO versus 6/MAO. Fourth,
7/MAO does not produce PVC under these conditions in the
presence of excess benzoquinone or 4-tert-butyl-catechol (in-
hibitor/Al ) 3).50 On the other hand, the tacticity of the PVC
produced by 6/MAO and 7/MAO is in the range expected for
radical polymerization.51
The reaction of 6/MAO with VC in CH2Cl2 at 25 °C under
anaerobic conditions ([Ti] ) 0.0033 M, [Al] ) 0.033 M, [VC]
) 2.8 M; 24 h) yields PVC (eq 6, 36% conversion). 7/MAO
also produces PVC under similar conditions but in lower yield
(7% conversion). These results confirm that PVC is pro-
duced by 6/MAO or 7/MAO under the Endo and Saitoh
conditions.
The best explanation for these observations is that both
6/MAO and 7/MAO initiate radical VC polymerization, but
Al-Me species derived from the cocatalyst react with the PVC,
9
802 J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003

Metal-Catalyzed Insertion Polymerization of Vinyl Chloride
A R T I C L E S
Scheme 5 a
occur. For 6/MAO, the resulting [AlMey(OMe)1-y(O)]n and
AlMez(OMe)3-z species will be less reactive, while, for 7/MAO,
the resulting [AlMeyCl1-y(O)]n and AlMezCl3-z species will be
more reactive compared to [AlMe(O)]n and AlMe3, because of
the influence of -OMe versus -Cl substitution on the degree
of association and Lewis acidity of the Al species.57 Therefore,
substantially less methylation of the defect sites and main chain
units is expected for 6/MAO versus 7/MAO.
To test this explanation, the reaction of PVC with MAO was
investigated. The reaction of PVC (prepared by radical polym-
erization) with MAO in CH2Cl2 under conditions similar to those
of eq 6 ([Al] ) 0.05 M; [PVC] ) 0.44 M (monomer basis), 25
°C, 24 h) yields a dark red solution from which PVC was
isolated in >90% yield. 13C NMR analysis established that the
tacticity of the PVC is unchanged by the MAO treatment. How-
ever, as illustrated in Figure 5, 1H NMR analysis of the treated
PVC shows that allylic chloride units are completely removed
and the -CHClCH2Cl, -CH2Cl branch, and -CH2CH2Cl
resonances are reduced in intensity. Also, the 1H NMR spectrum
of the MAO-treated PVC contains broad resonances at δ 0.85
and 1.25 which are assigned to -CH2CHMe- units formed by
the methylation of main chain -CH2CHCl- units (cf. Figure
3c). GPC analysis established that the molecular weight
distribution of the PVC is broadened significantly by the MAO
treatment (before treatment, Mw ) 67 300, Mw/Mn ) 2.1; after
treatment, Mw ) 177 000, Mw/Mn ) 19.5). To model the post-
polymerization reaction in the 6/MAO case, the above experi-
ment was repeated except that 0.3 equiv of MeOH per Al was
added to convert a fraction of the Al-Me groups to Al-OMe
groups.58 As shown in Figure 5, 1H NMR analysis of the
MAO/MeOH-treated PVC showed that the level of allylic chlor-
ide defects was reduced by only 33% and the -CHClCH2Cl,
-CH2Cl branch, and -CH2CH2Cl resonances were unaffected.
These results confirm that the PVC radical defects react with
MAO under the conditions of eq 6 and that this reaction is
inhibited by the presence of Al-OMe groups.
The presence of radical defects in the PVC produced by
6/MAO, the control experiments which show that radical defects
in the PVC produced by 7/MAO will be consumed by
postpolymerization reactions with MAO, the absence of VC
polymerization by 7/MAO in the presence of radical inhibitors,
and the similarity of the tacticity of the PVC produced by
6/MAO and 7/MAO to that of PVC produced by radical
polymerization under similar conditions, leads us to conclude
that Cp*TiX3/MAO catalysts initiate radical polymerization of
VC at low Al/Ti ratios in CH2Cl2. The differences between our
results and those of previous investigators may reflect differ-
ences in the MAO used in the respective studies. As noted
above, the PVC prepared by rac-(EBI)ZrMe2/MAO/O2 also
exhibits features expected from postpolymerization reactions of
PVC and MAO.
a P ) polymer chain.
as illustrated in Scheme 5. Al-Me species are expected to
methylate the allylic chloride groups yielding internal olefin
units (i, ii)9,52 and to methylate or dehydrochlorinate tertiary
chloride units (iii, iv); in the latter case, the allylic chloride units
which are formed may be methylated in a subsequent step (v).53
Al-Me species can also methylate main chain (secondary)
-CH2CHCl- groups at a slower rate (vi).54 Carbocation inter-
mediates in these reactions can add to olefin units, leading to
chain coupling and resulting in the observed broad molecular
weight distributions.55
The differences in the PVCs prepared by 6/MAO versus
7/MAO are ascribed to two factors. First, because of the higher
PVC yield for 6/MAO versus 7/MAO, the molar ratio (Al-R
groups in cocatalyst)/(defect units in PVC product) is lower in
the former case, and therefore, consumption of the defect sites
is expected to be less complete.56 Second, and probably more
important, while the details of the reactions of 6 or 7 with MAO
and the AlMe3 therein are not known (vide infra), it is expected
that a substantial exchange of Ti-X and Al-R groups will
(50) An excess of inhibitor was used because the inhibitor may be consumed
by a reaction with 7 or MAO. The lack of complete inhibition in the
inhibitor experiments described in ref 36 (inhibitor/Ti/Al ) 1:1:10) can be
ascribed to the consumption of the inhibitor by such reactions. See: (a)
Mahanthappa, M. K.; Huang, K.-W.; Cole, A. P.; Waymouth, R. M. Chem.
Commun. 2002, 502. (b) Ziemkowska, W. Main Group Met. Chem. 2000,
23, 337. (c) Boucher, D. L.; Brown, M. A.; McGarvey, B. R.; Tuck, D. G.
J. Chem. Soc., Dalton Trans. 1999, 3445. (d) Granel, C.; Jerome, R.;
Teyssie, P.; Jasieczek, C. B.; Shooter, A. J.; Haddleton, D. M.; Hastings,
J. J.; Gigmes, D.; Grimaldi, S.; Tordo, P.; Greszta, D.; Matyjaszewski, K.
Macromolecules 1998, 31, 7133. (e) Carrera, M. A.; Mena, M.; Royo, P.;
Serrano, R. J. Organomet. Chem. 1986, 315, 329.
(51) The PVC produced by 6/MAO (0.36 rr, 0.49 mr, 0.15 mm, R ) 0.61) and
7/MAO (0.40 rr, 0.48 mr, 0.12 mm, R ) 0.63) is slightly more syndiotactic
than PVC produced by suspension polymerization at this temperature (0.33
rr, 0.50 mr, 0.18 mm, R ) 0.58; at 20 °C). Endo and Saitoh reported the
tacticity of PVC prepared by 6/MAO to be 0.34 rr, 0.50 mr, 0.16 mm, and
R ) 0.60. These small differences, which are near the limit of precision of
the NMR measurement, are better ascribed to solvent effects than to a
change in mechanism.
(52) (a) Gupta, S. N.; Kennedy, J. P.; Nagy, T. T.; Tudos, F.; Kelen, T. J.
Macromol. Sci., Chem. 1978, A12, 1407. (b) Kennedy, J. P.; Smith, R. R.
Polym. Eng. Sci. 1974, 14, 322. (c) Marshall, D. W.; Sorenson, W. R. U.S.
Patent 3,274,280 (Continental Oil Company). Chem. Abstr. 65:7216e.
(53) (a) Kennedy, J. P.; Desai, N. V.; Sivaram, S. J. Am. Chem. Soc. 1973, 95,
6386. (b) Miller, D. B. J. Org. Chem. 1966, 31, 908. (c) Kennedy, J. P. J.
Org. Chem. 1970, 35, 532.
(54) Pasynkiewicz, S.; Kuran, W. J. Organomet. Chem. 1969, 16, 43.
(55) Thame, N. G.; Lundberg, R. D.; Kennedy, J. P. J. Polym. Sci., Polym.
Chem. Ed. 1972, 10, 2507.
(56) Assuming 5 × 10-4 allylic chloride groups per monomer unit in PVC
prepared by radical polymerization at 25 °C, we estimate the ratio (mol Al
in cocatalyst)/(mol allylic chloride units in PVC) to be 65 for the 6/MAO
reaction and 420 for the 7/MAO reaction. The corresponding ratios (mol
total Al-Me in cocatalyst)/(mol allylic chloride units in PVC) are estimated
to be 84 and 580. 1H NMR analysis of a commercial PVC sample prepared
by radical polymerization at 20 °C established that the ratio (allylic chloride
units)/(monomer unit) ) 4.5 × 10-4
.
(57) (a) Mole, T.; Jeffery, E. A. Organoaluminium Compounds; Elsevier: New
York, 1972; p 271 and pp 349-354. (b) Pasynkiewicz, S.; Boleslawski,
M.; Sadownik, A. J. Organomet. Chem. 1976, 113, 303.
(58) Rogers, J. H.; Apblett, A. W.; Cleaver, W. M.; Tyler, A. N.; Barron, A. R.
J. Chem. Soc., Dalton Trans. 1992, 3179.
9
J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003 803

A R T I C L E S
Stockland et al.
erization that are formed from (C5R5)TiX3/activator catalysts
is a subject of current investigation in several laboratories.60 It
is clear that (C5R5)TiX3/activator systems are complicated and
that a variety of Ti(IV), Ti(III), and possibly Ti(II) species can
form, depending on the system and conditions. The current
consensus is that the active species for R-olefin polymerization
are Ti(IV) alkyls, and it is likely that a Ti(IV) species is
responsible for the oligomerization of the propylene in eq 7.
The identity and origin of the radicals that initiate VC
polymerization in eq 6 are unknown. For example, it is possible
that radicals are produced by autoxidation of Ti-R or Al-R
species by trace O2,45-47 homolysis of Ti-R species,61 reaction
of Cp*Ti(III) species with the solvent, or transfer of a radical
(R or Cl) from a Ti species to VC.
Constrained Geometry Catalysts. Another possible ap-
proach to favoring VC coordination and insertion over â-Cl
elimination is to use a sterically open LnM structure to minimize
the steric inhibition of monomer coordination. “Constrained
geometry” (C5Me4SiMe2NtBu)MMe+ catalysts (M ) Zr, Ti),
which exhibit high reactivity with R-olefins, styrene, and even
isobutylene in copolymerization reactions with ethylene, are of
interest in this regard.62,63 Soga reported VC polymerization with
constrained geometry catalysts, but no details were given.37 We
have also investigated the reaction of (C5Me4SiMe2NtBu)MMe+
species (M ) Zr, Ti) with VC.
The cationic complex [(C5Me4SiMe2NtBu)ZrMe(C6D6)]-
[B(C6F5)4] (8) was generated by the reaction of (C5Me4SiMe2-
NtBu)ZrMe2 with [Ph3C][B(C6F5)4] in C6D6 as described by
Marks.64 Compound 8 separates as liquid clathrate from
C6D6 and decomposes in CD2Cl2 at room temperature. The re-
action of 8 with VC (6 equiv) in C6D6 at 50 °C for 2 h results
in the consumption of 1 equiv of VC and formation of an
insoluble oil which is believed to be [(C5Me4SiMe2NtBu)ZrCl]-
[B(C6F5)4] (9) or its C6D6 solvate. No PVC is formed. Oligo-
propylene could not be unambiguously identified by NMR be-
cause resonances from catalyst species interfere with anticipated
oligopropylene resonances; however, a small amount of free
propylene was detected in the volatiles. Complex 9 could not
be characterized by NMR but was characterized by chemical
derivatization. Thus, the volatiles were removed under vacuum,
1 equiv of [NBu3Bn]Cl was added, and the mixture was taken
Figure 5. 1H NMR spectra (500 MHz, THF-d8, 50 °C, defect region) of
commercial PVC before and after treatment with MAO. (a) PVC treated
with MAO at 25 °C. (b) PVC treated with MAO/MeOH at 25°C. (c)
Commercial PVC prepared at 80 °C prior to treatment.
Reaction of Cp*TiCl3/MAO at High Al/Ti Ratios. The
reaction of 7/MAO with VC in C6D6 with MAO loadings more
typical of those used for olefin polymerization ([Ti] ) 7 mM,
Al/Ti ) 1000; [VC] ) 0.92 M, 80 °C) yields atactic oligopro-
pylene (eq 7). No PVC is formed under these conditions. NMR
(60) (a) Grassi, A.; Saccheo, S.; Zambelli, A.; Laschi, F. Macromolecules 1998,
31, 5588. (b) Ready, T. E.; Gurge, R.; Chien, J. C. W.; Rausch, M. D.
Organometallics 1998, 17, 5236. (c) Williams, E. F.; Murray, M. C.; Baird,
M. C. Macromolecules 2000, 33, 261. (d) Xu, G.; Cheng, D. Macromol-
ecules 2000, 33, 2825. (e) Mahanthappa, M. K.; Waymouth, R. M. J. Am.
Chem. Soc. 2001, 123, 12093.
(61) Borkowsky, S.; Baenziger, N. C.; Jordan, R. F. Organometallics 1993, 12,
486.
studies show that ∼80 equiv of VC are consumed per Ti in
this reaction. These results can be rationalized by a scheme
analogous to Scheme 4. Thus, an active Ti-Me species
undergoes 1,2 VC insertion to generate a TiCH2CHClMe
intermediate which undergoes rapid â-Cl elimination to yield
propylene and a Ti-Cl species. The propylene is oligomerized
by active Ti-R species, and the Ti-Cl product is realkylated
by the MAO. Catalysts derived from Cp*TiX3 precursors
generally produce atactic polypropylene.59
(62) (a) Stevens, J. C.; Timmers, F. J.; Wilson, D. R.; Schmidt, G. F.; Nickias,
P. N.; Rosen, R. K.; Knight, G. W.; Lai, S. (Dow Chemical Co.) European
Patent Application EP 0416815 A2, 1991; Chem. Abstr. 1991, 115:93163.
(b) Canich, J. M. (Exxon Chemical Co.) European Patent Application EP
0420436 A1, 1991; Chem. Abstr. 1991, 115:184145. (c) Stevens, J. C. In
Studies in Surface Science and Catalysis; Soga, K., Terano, M., Eds.;
Elsevier: Amsterdam, 1994; Vol. 89, pp 277-284.
(63) (a) McKnight, A. L.; Waymouth, R. M. Chem. ReV. 1998, 98, 2587. (b)
Soga, K.; Uozomi, T.; Nakamura, S.; Toneri, T.; Teranishi, T.; Sano, T.;
Arai, T. Macromol. Chem. Phys. 1996, 197, 4237. (c) Sernetz, F. G.;
Mu¨lhaupt, R.; Waymouth, R. M. Macromol. Chem. Phys. 1996, 197, 1071.
(d) Sernetz, F. G.; Mu¨lhaupt, R.; Amor, F.; Eberle, T.; Okuda, J. J. Polym.
Sci., Part A 1997, 35, 1571. (e) Shaffer, T. D.; Canich, J. M.; Squire, K.
R. Macromolecules 1998, 31, 5145.
(64) (a) Jia, L.; Yang, X.; Stern, C. L.; Marks, T. J. Organometallics 1997, 16,
842. (b) Chen, Y.-X.; Marks, T. J. Organometallics 1997, 16, 3649. (c)
[Ph3C][B(C6F5)4] was chosen as the activator for these studies because
the reaction of (C5Me4SiMe2NtBu)ZrMe2 with B(C6F5)3 yields a tight
[(C5Me4SiMe2NtBu)ZrMe][MeB(C6F5)4] ion pair which is essentially
inert for ethylene polymerization and was therefore expected to be inert to
VC.
Active Species in (C5Me5)TiCl3/MAO Catalysts. The
identity of the active species for R-olefin and styrene polym-
(59) (a) Sassmannshausen, J.; Bochmann, M.; Ro¨sch, J.; Lilge, D. J. Organomet.
Chem. 1997, 548, 23. (b) Park, J. R.; Shiono, T.; Soga, K. Macromolecules
1992, 25, 521. (c) See also: Quyoum, R.; Wang, Q.; Tudoret, M. J.; Baird,
M. C. J. Am. Chem. Soc. 1994, 116, 6435.
9
804 J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003

Metal-Catalyzed Insertion Polymerization of Vinyl Chloride
A R T I C L E S
Scheme 6 a
of atactic polypropylene. No PVC is formed. Similar results
were obtained over a range of reaction conditions (Al/Zr )
800-17 000; 25-80 °C, 2-24 h). These results can be
accounted for by a 1,2 insertion/â-Cl elimination/realkylation
process analogous to that observed for zirconocene/MAO and
Cp*TiCl3/MAO catalysts.
The titanium catalyst (C5Me4SiMe2NtBu)TiMe2/[Ph3C]-
[B(C6F5)4] reacts with VC in a similar manner as the Zr
analogue. The reaction of (C5Me4SiMe2NtBu)TiMe2 and [Ph3C]-
[B(C6F5)4] in C6D6 yields the dinuclear cation {(C5Me4SiMe2-
NtBu)TiMe}2(µ-Me)+.64b This species consumes 1 equiv of
VC per Ti, yielding a (C5Me4SiMe2NtBu)TiCl+ species which
is trapped by [NBu3Bn]Cl to afford (C5Me4SiMe2NtBu)TiCl2.
As for the Zr case, oligopropylene could not be definitively
observed but free propylene was detected. More strikingly,
(C5Me4SiMe2NtBu)TiMe2/[Ph3C][B(C6F5)4] polymerizes pro-
pylene in the absence of VC to high molecular weight atactic
polypropylene (Mn ) 462 000, unsaturated end groups not
a (CGC)Zr ) (C5Me4SiMe2NtBu)Zr.
Scheme 7 a
1
detectable by H NMR), whereas propylene polymerization in
the presence of VC yields low molecular weight polypropylene
(Mn ) 1610) with allylic chain ends. These results show that
(C5Me4SiMe2NtBu)TiR+ species react with VC by 1,2 insertion
followed by â-Cl elimination.
a (CGC)Zr ) (C5Me4SiMe2NtBu)Zr.
up in C6D6. NMR analysis established that the only significant
organometallic species present in the final solution was
(C5Me4SiMe2NtBu)ZrCl2, which was produced in 50% yield
versus that of 8.
Conclusions
The development of catalysts for insertion polymerization/
copolymerization of VC and related CH2dCHX monomers is
a challenging goal. To probe the chemical issues underlying
this problem, we have investigated the reactions of discrete
olefin polymerization catalysts with VC and VC/propylene
mixtures. In this study, three representative group 4 metal
catalyst classes, that is, zirconocene, monocyclopentadienyl
titanium, and constrained geometry Zr and Ti systems, were
studied. Two general reaction pathways were observed: (i)
radical polymerization of VC initiated by radicals derived from
the catalyst and (ii) net 1,2 VC insertion followed by fast â-Cl
elimination.
These results are consistent with 1,2 insertion of VC into 8
and fast â-chloride elimination to yield 9 and propylene, as
illustrated in Scheme 6. Polymerization of propylene by
(C5Me4SiMe2NtBu)MR+ catalysts affords atactic polypropylene,
but as noted above, the fate of the propylene in Scheme 6 could
not be unambiguously determined.63a,64,65
Control experiments confirm that [NBu3Bn]Cl acts simply
as a trap for 9 in Scheme 6. Complex 9 was generated as a
liquid clathrate in C6D6 by the initial generation of (C5Me4-
SiMe2NtBu)Zr(Me)Cl by comproportionation of (C5Me4SiMe2-
NtBu)ZrCl2 and (C5Me4SiMe2NtBu)ZrMe2, followed by treat-
ment with [Ph3C][B(C6F5)4] (Scheme 7). The reaction of 9 with
1 equiv of [NBu3Bn]Cl yields (C5Me4SiMe2NtBu)ZrCl2 in 63%
yield. In contrast, exposure of 8 to the conditions of Scheme 6
in the absence of VC, followed by reaction with 1 equiv of
[NBu3Bn]Cl, yields (C5Me4SiMe2NtBu)Zr(Me)Cl but not
(C5Me4SiMe2NtBu)ZrCl2.
Radicals capable of initiating VC polymerization can be
generated from group 4 metal catalysts in several ways. For
rac-(EBI)ZrMe2/MAO, trace O2 is required for VC polymeri-
zation, implicating the autoxidation of Zr-R and/or Al-R
species as the source of R• and RO2• radicals which can initiate
polymerization. For Cp*TiX3/MAO catalysts in CH2Cl2 solvent
at low Al/Ti ratios, radical VC polymerization occurs under
anaerobic conditions. In this case, the source of initiating radicals
is unknown. Radical polymerization of VC can be identified
by the presence of terminal and internal allylic chloride units
and other “radical defects” in the polymer, which arise from
the characteristic chemistry of PCH2CHCl• macroradicals.
However, the present study shows that this test must be used
with caution, since the radical defect units can be consumed by
postpolymerization reactions with the cocatalyst, as observed
most strikingly for Cp*TiCl3/MAO. Radical polymerization may
complicate the reactions of metal catalysts with VC but can be
avoided by using nonredox active metal catalysts, anaerobic
reaction conditions, and radical inhibitors.
To confirm that (C5Me4SiMe2NtBu)ZrR+ species react with
VC by 1,2 insertion and â-Cl elimination, as proposed in
Scheme 6, NMR scale batch reactions of 8 and propylene in
the absence and presence of VC were compared (23 °C, C6D6).
In the absence of VC, the propylene was completely polymer-
ized to low molecular weight atactic polypropylene containing
only vinylidene unsaturated end groups, consistent with chain
transfer by â-H elimination. In the presence of VC, however,
the propylene was only partially polymerized (75%), and both
allylic (13%) and vinylidene (87%) end groups were observed
in the polymer, consistent with competitive chain transfer by
â-H elimination and termination by 1,2 VC-insertion/â-Cl
elimination.
The reaction of (C5Me4SiMe2NtBu)ZrCl2/MAO (Al/Zr )
1000) with VC in toluene (80 °C, 2 h) yields a small amount
The second pathway observed in reactions of group 4
LnMR+ species and VC is the formation of LnMCl+ and
CH2dCHR products. While no intermediates have been ob-
served, this reaction most likely proceeds by 1,2 insertion to
(65) McKnight, A. L.; Masood, M. A.; Waymouth, R. M. Organometallics 1997,
16, 2879 and references therein.
9
J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003 805

A R T I C L E S
Stockland et al.
yield LnMCH2CHClR+ species which undergo rapid â-Cl
elimination. It is also possible that LnMCHClCH2R+ species
are formed by 2,1 VC insertion and rearrange to â-chloroalkyl
species prior to elimination. VC insertion of group 4 LnMR+
species can be facile, occurring readily, for example, at -10
°C for rac-(EBI)ZrMe+ and 50 °C for (C5Me4SiMe2NtBu)-
ZrMe+. However VC insertions are slower than analogous
propylene insertions, probably because VC binds more weakly
than propylene to d0 LnMR+ cations. Multiple VC insertions
leading to VC polymerization are not observed for group 4 metal
catalysts because LnMCH2CHClR+ species undergo â-Cl elimi-
nation faster than they coordinate and insert VC.
Experimental Section
General Procedures. All experiments were performed using stan-
dard Schlenk or vacuum line techniques or in a nitrogen-filled drybox.
Pentane, hexane, and toluene were distilled from sodium/benzophenone
or purified by passage through columns of activated alumina and a
BASF R3-11 oxygen removal catalyst. Dichloromethane, chloroben-
zene, dichloromethane-d2, and chlorobenzene-d5 were dried over CaH2
and distilled. Nitrogen was purified by passage through columns
containing activated 4-Å molecular sieves and a Q-5 oxygen scavenger.
MAO (30 wt % solution in toluene; 13.4 wt % Al) was obtained from
Albemarle; this material contained a small amount of a process oil that
was characterized as described below. B(C6F5)3, [Ph3C][B(C6F5)4], and
[NHMe2Ph][B(C6F5)4] were obtained from Boulder Scientific. Lecture
bottles of vinyl chloride (VC) were purchased from Aldrich and used
without further purification. VC containing 14 ppm H2O was obtained
from Geon and dried by passage through 4-Å molecular sieves.
Benzoquinone, 4-tert-butylcatechol, (indenyl)2ZrMe2, Cp*2ZrCl2, and
Cp*TiCl3 were obtained from Aldrich. Cp*Ti(OMe)3 was obtained from
Strem. rac-(EBI)ZrMe2, rac-(EBI)ZrCl2, rac-Me2Si(2-Me-4,5-benz-
indenyl)2ZrCl2, Cp2ZrMe2, and (C5Me4SiMe2NtBu)MMe2 (M ) Ti, Zr)
were prepared according to literature methods.62,64,69-71
The reaction of rac-(EBI)ZrMe2/MAO or [(C5Me4SiMe2-
NtBu)ZrMe][B(C6F5)4] with propylene/VC mixtures yields oli-
gopropylene containing both allylic and vinylidene chain ends
rather than strictly vinylidene (unsaturated) chain ends, as
observed in propylene homopolymerization by these catalysts.
These results show that VC insertion into LnM(CH2CHMe)nR+
active species is also followed by â-Cl elimination, which
terminates chain growth and precludes propylene/VC copolym-
erization.
Gel permeation chromatography (GPC) was performed on a Polymer
Labs PL-220 instrument using PLGel mixed B columns with refractive
index detection and calibration versus narrow polystyrene standards.
Elemental analyses were performed by Midwest Analytics. NMR
spectra were recorded at ambient temperature unless specified otherwise.
1H and 13C chemical shifts are reported relative to SiMe4 and were
determined by reference to the residual 1H and 13C solvent resonances.
11B chemical shifts are reported relative to BF3‚Et2O. 19F chemical shifts
are reported relative to CFCl3. All coupling constants are given in hertz.
Control Experiments. NMR experiments using internal standards
and product isolation experiments show that the following compounds
do not react with VC under the specified conditions: (i) B(C6F5)3 in
CD2Cl2 (25 °C, 86 h): (ii) AlMe3 in C6D6 (100 °C, sealed tube,
overnight);72 (iii) dried MAO in C6D6 (100 °C, sealed tube, 30 min);
(iv) AlMe3 and dried MAO in C6D6 (25 °C, overnight); (v) MAO in
toluene (25 °C, 20 h); (vi) rac-(EBI)ZrMe2 in C6D6 (100 °C, 3 h).
Process Oil in MAO. A Schlenk tube was charged with MAO (5.44
g of a 30 wt % solution in toluene; Al ) 26.8 mmol) and toluene (20
mL). The clear solution was added via pipet to acidified methanol (100
mL of a 1 M HCl solution) to quench the MAO. The resulting solution
was extracted with hexanes (3 × 30 mL). The extract was dried under
The ultimate products from VC insertion/â-Cl elimination
of group 4 LnMR+ species reflect the downstream chemistry
available to the system. For example, rac-(EBI)ZrMe2/B(C6F5)3
consumes 2 equiv of VC per Zr because the product of the first
insertion/â-Cl elimination, [rac-(EBI)ZrCl][MeB(C6F5)3] (4),
-
contains a reactive MeB(C6F5)3 anion which realkylates the
rac-(EBI)ZrCl+ cation, enabling a second VC insertion/â-Cl
elimination to occur. In this system, the evolved propylene is
oligomerized by rac-(EBI)ZrR+ as it is formed. For (C5R5)2-
ZrX2/MAO and Cp*TiCl3/MAO at high Al/Ti ratios, realkyla-
tion of the LnMCl+ species formed by â-Cl elimination by MAO
is efficient, and catalytic conversion of VC to oligopropylene
can be achieved; however, this process is stoichiometric in
Al-Me groups. In contrast, for [(C5Me4SiMe2NtBu)MMe]-
[B(C6F5)4] (M ) Ti, Zr), VC insertion/â-Cl elimination yields
[(C5Me4SiMe2NtBu)MCl][B(C6F5)4] species which do not react
further and can be trapped as (C5Me4SiMe2NtBu)MCl2 by
addition of a chloride source.
1
vacuum, yielding a small amount of an opaque pale yellow oil. H
NMR (THF-d8): δ 1.30 (br, 2H), 0.88 (m, 1H).
The facility of â-Cl elimination of group 4 LnMCH2CHClR+
species is not surprising, as this reaction is highly exothermic
because of the higher bond dissociation energies of MsCl
versus MsC bonds for early metals.66 Assuming a Zr(IV)s
olefin bond strength of 15 kcal/mol,67 we estimate the conversion
of (C5R5)2ZrCH2CHClR+ to (C5R5)2ZrCl(CH2dCHR)+ to be
exothermic by ∼42 kcal/mol.68 Additionally, the barrier to â-Cl
elimination is expected to be low because of the favorable
geometry of the transition state for syn elimination. Termination
of chain growth by â-Cl elimination is the most significant
obstacle to insertion polymerization/copolymerization of VC.
Possible approaches to avoiding this reaction include using late
metal catalysts for which MsR and MsCl bond strengths are
more comparable39 and designing catalysts for which 2,1 VC
insertion is favored.
Generation of rac-(EBI)ZrMe(µ-Me)B(C6F5)3 (1).40c An NMR tube
was charged with rac-(EBI)ZrMe2 (0.011 g, 0.029 mmol) and B(C6F5)3
(0.015 g, 0.029 mmol), and toluene (0.01 g, 0.1 mmol; internal standard)
and CD2Cl2 (0.5 mL) were condensed in at -196 °C. The tube was
sealed, warmed to -78 °C, and placed in an NMR probe that was
precooled to -73 °C. A 1H NMR spectrum was obtained which
1
confirmed that 1 had formed quantitatively. H NMR of 1 (CD2Cl2,
3
-63 °C): δ 7.53-7.38 (m, 5H, indenyl), 7.18 (t, JHH ) 3 Hz, 1H,
3
3
indenyl), 7.12 (d, JHH ) 3 Hz, 1H, indenyl), 6.69 (t, JHH ) 9 Hz,
1H, indenyl), 6.31 (d, 3JHH ) 3 Hz, 1H, indenyl), 6.22 (t, 3JHH ) 9 Hz,
3
3
1H, indenyl), 6.11 (d, JHH ) 3 Hz, 1H, indenyl), 5.92 (d, JHH ) 3
Hz, 1H, indenyl), 3.82-3.68 (m, 1H, CH2CH2), 3.60-3.45 (m, 1H,
CH2CH2), 3.45-3.31 (m, 2H, CH2CH2), -0.49 (s, 3H, ZrMe), -0.99
(s, 3H, MeB). 1 is stable in CD2Cl2 at -78 °C but decomposes
significantly within 1 h at 25 °C in this solvent.
(69) Diamond, G. M.; Jordan, R. F.; Petersen, J. L. J. Am. Chem. Soc. 1996,
118, 8024.
(70) Zhang, X.; Zhu, Q.; Guzei, I. A.; Jordan, R. F. J. Am. Chem. Soc. 2000,
122, 8093.
(71) Samuel, E.; Rausch, M. D. J. Am. Chem. Soc. 1973, 95, 6263.
(72) AlMe3 was reported to react with VC at 100 °C to yield propylene and
oligomers of propylene, but few details were given. Pasynkiewicz, S.;
Kuran, W. J. Organomet. Chem. 1968, 15, 307.
(66) Simoes, A. A. M.; Beauchamp, J. L. Chem. ReV. 1990, 90, 629.
(67) (a) Carpentier, J.-F.; Wu, Z.; Lee, C. W.; Stro¨mberg, S.; Christopher, J.
N.; Jordan, R. F. J. Am. Chem. Soc. 2000, 122, 7750. (b) Carpentier, J.-F.;
Maryin, V. P.; Luci, J.; Jordan, R. F. J. Am. Chem. Soc. 2001, 123, 898.
(68) Zr-C (67 kcal/mol) and Zr-Cl (116 kcal/mol) bond energies are taken
from: Schock, L. E.; Marks, T. J. J. Am. Chem. Soc. 1988, 110, 7701.
9
806 J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003

Metal-Catalyzed Insertion Polymerization of Vinyl Chloride
A R T I C L E S
1
3
Reaction of rac-(EBI)ZrMe(µ-Me)B(C6F5)3 (1) and VC in CD2Cl2.
A CD2Cl2 solution of 1 (0.029 mmol) containing toluene (0.09 mmol)
as an internal standard was prepared in an NMR tube as described
above. VC (20.3 mL at 419 mm and 25 °C; 0.46 mmol) was condensed
into the NMR tube at -196 °C. The tube was warmed to -78 °C and
inserted into an NMR spectrometer probe which had been precooled
to -88 °C. 1H NMR spectra were recorded from -88 °C to 25 °C. No
reaction between 1 and VC was detected below -10 °C. Above -10
°C, the VC resonances decreased in intensity and the resonances for 1
began to broaden into the baseline. At 25 °C, the resonances for 1
were absent and the solution was dark red. No PVC resonances were
observed. However, resonances for atactic oligopropylene were ob-
served. H NMR (CD2Cl2, -73 °C, C5 region): δ 7.09 (d, JHH ) 3
3
Hz, 1H, indenyl C5-H), 6.42 (d, JHH ) 3 Hz, 1H, indenyl C5-H),
3
3
6.34 (d, JHH ) 3 Hz, 1H, indenyl C5-H), 6.05 (d, JHH ) 3 Hz, 1H,
indenyl C5-H), -0.40 (br s, coordinated MeB(C6F5)3-).
Conversion of 2 to rac-(EBI)ZrCl2 (3). A CD2Cl2 solution of 1
(0.029 mmol) and toluene (0.01 g, 0.1 mmol; internal standard) was
generated in an NMR tube as described above. The tube was maintained
1
at -78 °C, and a H NMR spectrum was obtained at -73 °C. The
tube was removed from the spectrometer and cooled to -196 °C. VC
(0.46 mmol) was condensed into the tube at -196 °C. The tube was
warmed to 25 °C for 5 min, and the color changed from yellow to
dark red because of the formation of 2. The tube was then taken into
the drybox where [NBu3Bn]Cl (0.0045 g, 0.015 mmol) was added. The
color of the solution immediately changed from dark red to pale yellow.
The tube was removed from the drybox and placed in an NMR
spectrometer probe that had been precooled to -73 °C, and a 1H NMR
spectrum was obtained. The only Zr-containing species detectable in
the spectrum was 3. Comparison of the integrated intensities for 1, 3,
and the toluene internal standard established that 3 was formed
quantitatively versus 1.
1
served. H NMR (CD2Cl2, 25 °C): δ 1.60 (br, -CH), 1.2-0.7 (br,
CH2 and -Me).73 The tube was maintained at 25 °C for 24 h. The
color of the solution changed from red to pale yellow. NMR analysis
revealed that the only organometallic species present in the solution
were rac-(EBI)ZrCl2 (3, 100% by 1H NMR) and B(C6F5)3.74 1H NMR
integration established that 2 equiv of VC had been converted to
oligopropylene. In a similar experiment using 0.053 mmol of 1, benzene
was added to the tube at this point to induce precipitation of 3. The
solid was collected by filtration, washed with benzene and hexane, and
dried under vacuum to afford 3 as a pale yellow solid (0.018 g, 81%).
Quantification of VC Consumption and Formation of 2 in Stage
1. A CD2Cl2 solution of 1 (0.29 mmol) containing toluene (0.01 g,
0.10 mmol; internal standard) and VC (0.46 mmol) was generated in
an NMR tube as described above. The tube was warmed to -78 °C
and inserted into an NMR spectrometer probe that had been precooled
to -73 °C. A 1H NMR spectrum was recorded. The tube was warmed
to 25 °C for 5 min, and the color of the solution changed from pale
NMR Characterization of [{rac-(EBI)ZrCl}2(µ-Cl)][MeB(C6F5)3]
(2). A CD2Cl2 solution of 1 (0.029 mmol) was generated in an NMR
tube as described above. VC (20.3 mL at 419 mm and 25 °C, 0.46
mmol) was condensed in at -196 °C. The tube was warmed to 25 °C
for 5 min. The volatiles were removed under vacuum, fresh CD2Cl2
was added by vacuum transfer, and NMR spectra were recorded at
-73 °C. The 19F NMR spectrum established that a 1:1 mixture of
1
yellow to dark red. A H NMR spectrum was recorded at -73 °C.
Comparison of the integrated intensities of the resonances for 1, VC,
and the toluene internal standard before and after the 25 °C reaction
period revealed that 1.5 equiv of VC was consumed per Zr and that 2
had formed in >94% yield versus 1.
B(C6F5)3 and MeB(C6F5)3 was present. For B(C6F5)3, 19F NMR
-
3
(CD2Cl2, -73 °C): δ -127.5 (d, JFF ) 22 Hz, 6F, o-C6F5), -143.4
3
(t, JFF ) 21 Hz, 3F, p-C6F5), -160.4 (m, 6F, m-C6F5). For MeB-
3
3
(C6F5)3-: δ -132.5 (d, JFF ) 21 Hz, 6F, o-C6F5), -164.6 (t, JFF
)
Reaction of rac-(EBI)ZrMe2/MAO and VC. A 250 mL stainless
steel autoclave fitted with a glass insert was charged with rac-(EBI)-
ZrMe2 (0.001 g, 3 µmol) and MAO (10 g of a 30 wt % toluene solution;
51 mmol of Al; Al/Zr ) 17 000). The autoclave was exposed to a
constant feed of VC (1 atm) and was mechanically stirred for 20 h at
25 °C. The VC feed was terminated, the autoclave was vented, and
the reaction mixture was poured into acidified methanol (50 mL of a
1 M HCl solution). The resulting mixture was extracted with hexane.
The hexane extract was dried under vacuum, yielding 0.250 g of a
colorless oil. The 1H and 13C NMR spectra established that the colorless
oil was atactic oligopropylene that contained a small amount of the
process oil present in the MAO. 1H NMR (1,1,2,2-C2D2Cl4, 100 °C) δ
1.62 (m, sCH), 1.10 (m, CH2-anti-meso and CH2, racemic), and 0.96
(m, CH2-syn-meso and sMe). The unsaturated end groups observable
1
20 Hz, 3F, p-C6F5), -167.2 (m, 6F, m-C6F5). The H NMR spectrum
established that [{rac-(EBI)ZrCl}2(µ-Cl)][MeB(C6F5)3] (2, 10:1 isomer
ratio, >83% of the total rac-(EBI)Zr species) was the major rac-(EBI)-
-
Zr species present and confirmed that MeB(C6F5)3 and atactic
oligopropylene were present.75 For major isomer of 2, 1H NMR
(CD2Cl2, -73 °C): δ 7.70-7.10 (m, 8H, indenyl), 6.82 (d, 3JHH ) 3.0
3
3
Hz, 1H, indenyl), 6.48 (d, JHH ) 3.0 Hz, 1H, indenyl), 6.10 (d, JHH
) 3.0 Hz, 1H, indenyl), 5.91 (d, 3JHH ) 3.0 Hz, 1H, indenyl), 4.0-3.6
(m, 4H, CH2CH2). 13C NMR (CD2Cl2, -73 °C): δ 130.2 (s, ipso),
1
3
1
128.2 (dd, JCH ) 163 Hz, JCH ) 8 Hz, CH), 127.6 (dd, JCH ) 165
3
1
3
Hz, JCH ) 8 Hz, CH), 127.3 (dd, JCH ) 162 Hz, JCH ) 8 Hz, CH),
126.9 (s, ipso), 126.8 (dd, 1JCH ) 166 Hz, 3JCH ) 8 Hz, CH), 126.5 (s,
ipso), 126.2 (dd, 1JCH ) 166 Hz, 3JCH ) 8 Hz, CH), 125.4 (ipso), 123.9
(dd, 1JCH ) 151 Hz, 3JCH ) 7 Hz, CH), 123.8 (s, ipso), 123.6 (s, ipso),
122.7 (dd, JCH ) 149 Hz, JCH ) 7 Hz, CH), 120.8 (dd, JCH ) 162
Hz, JCH ) 7 Hz, CH), 115.1 (d, JCH ) 179 Hz, CH), 114.3 (d, JCH
) 171 Hz, CH), 113.7 (d, JCH ) 162 Hz, CH), 112.4 (d, JCH ) 175
Hz, CH), 29.3 (t, 1JCH ) 132 Hz, CH2), 27.3 (t, 1JCH ) 131 Hz, CH2).
For minor isomer of 2, 1H NMR (CD2Cl2, -73 °C): δ 7.70-7.10 (m,
in the H spectrum comprised 95% allylic (1H NMR δ 5.83 (m, 1H,
1
1
3
1
dCHR), 5.05 (m, 2H, dCH2)) and 5% vinylidene (1H NMR δ 4.79 (s,
1H, dCH2), 4.72 (s, 1H, dCH2)) chain ends. The number average
3
1
1
1
1
1
molecular weight of the polymer (Mn) was determined from the H
NMR spectrum, assuming one unsaturated end group per chain. For
the sample generated above, Mn ) 563 (average degree of polymeri-
zation ) 13). Elemental analysis of oligopropylene Anal. Calcd for
(CH2CHMe)n: C, 85.63; H, 14.37. Found: C, 86.13; H, 13.56; Cl, not
detectable.
3
3
8H, indenyl), 6.78 (d, JHH ) 3 Hz, 1H, indenyl), 6.38 (d, JHH ) 3
Hz, 1H, indenyl), 6.32 (d, 3JHH ) 3 Hz, 1H, indenyl), 5.96 (d, JHH
)
3
3 Hz, 1H, indenyl), 4.0-3.6 (m, 4H, CH2CH2). The 13C spectrum of
the minor isomer was not discernible because of poor signal/noise and
peak overlap with the major isomer.
Batch Polymerization of Propene by rac-(EBI)ZrMe2/MAO in
the Absence and Presence of VC. A 250 mL stainless steel autoclave
vessel fitted with a glass insert was charged with rac-(EBI)ZrMe2 (0.001
g, 3 µmol), MAO (10 g of a 30 wt % toluene solution; 51 mmol of Al;
Al/Zr ) 17 000), and propene (250 mL at 380 mm and 25 °C, 5.1
mmol). The mixture was stirred mechanically for 20 h at 25 °C. The
autoclave was vented, and the reaction mixture was poured into acidified
MeOH (50 mL of a 1 M HCl solution). A colorless precipitate formed
and was separated by filtration, washed with MeOH, and dried under
vacuum, yielding 0.135 g (63% based on propene) of a colorless solid.
1H and 13C NMR analysis established that the solid was isotactic
In another experiment designed to quench the reaction before stage
1 was complete, the NMR tube containing the CD2Cl2 solution of 1
and VC was maintained at 25 °C for 1 min and then cooled to -78 °C
and analyzed by low temperature 1H NMR. Resonances for an
intermediate assigned as rac-(EBI)ZrCl(µ-Me)B(C6F5)3 (4) were ob-
(73) Bovey, F. A. Chain Structure and Conformation of Macromolecules;
Academic Press: New York, 1982; p 78.
(74) Massey, A. G.; Park, A. J. J. Organomet. Chem. 1966, 5, 218.
(75) The isomer ratio of 2 varied from 10:1 to 10:2.
9
J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003 807

A R T I C L E S
Stockland et al.
polypropylene (Mn ) 7500). The unsaturated end groups were
exclusively vinylidene. 1H NMR (1,1,2,2-C2D2Cl4, 100 °C, olefin
region): δ 4.79 (s, 1H, dCH2); 4.72 (s, 1H, dCH2).
(3 × 50 mL), and dried under vacuum (0.256 g, 93%). GPC analysis:
Mw ) 176 770, Mn ) 9039, Mw/Mn ) 19.5. GPC analysis of the
untreated PVC: Mw ) 67 314, Mn ) 30 658, Mw/Mn ) 2.11. In a similar
experiment, PVC (prepared at 30 °C) was treated with MAO in the
same fashion. 13C NMR showed that the tacticity was essentially
unchanged (before treatment 0.31 rr, 0.51 mr, 0.18 mm, R ) 0.56;
after treatment 0.34 rr, 0.49 mr, 0.17 mm, R ) 0.58).
Reaction of PVC with MAO/MeOH. A Schlenk tube was charged
with PVC (0.550 g, 8.8 mmol of monomer) and CH2Cl2 (20 mL), and
MeOH (13 mg, 0.40 mmol, 0.3 equiv vs MAO, degassed and under
N2) was added by microsyringe. MAO (0.256 g of a 30 wt % solution
in toluene; Al ) 1.26 mmol) was added by syringe, and the solution
was stirred for 24 h at 23 °C. The solution turned pale yellow and then
colorless over the course of 1 h. The solution was added to acidified
methanol (100 mL of a 1 M HCl solution). The PVC was separated by
filtration, washed with methanol (3 × 30 mL), and dried under vacuum
(0.502 g, 91%).
The procedure described above was followed with the exception that
VC (250 mL at 380 mm and 25 °C, 5.1 mmol) was added in addition
to the propene. Workup yielded 0.085 g (40% based on propene) of
isotactic polypropylene (Mn ) 7200). The 1H NMR spectrum established
that unsaturated end groups consisted of 76% allylic (δ 5.83 (m, 1H,
dCH2), 5.05 (m, 2H, dCHR)) and 24% vinylidene (δ 4.79 (s, 1H,
dCH2) 4.72 (s, 1H, dCH2)) chain ends. Since VC is acting as a chain
transfer agent, the molecular weight of the polypropylene is expected
to decrease when VC is present. The similarity of the Mn values of the
polypropylene produced in the presence and absence of VC reflects
the difference in % propene conversion. The lower % conversion in
the presence of VC results in a higher average propene pressure which
results in a higher Mn value of the resulting polymer.42
Reaction of rac-(EBI)ZrMe2/MAO with VC in the Presence of
Oxygen. A 250 mL stainless steel autoclave fitted with a glass insert
was charged with rac-(EBI)ZrMe2 (0.8 mg, 2 µmol), MAO (0.10 g of
a 30 wt % toluene solution; 0.49 mmol of Al; Al/Zr ) 250), VC (110
mL, 1.88 mol), and air (0.04 mL). The autoclave was stirred at 50 °C
for 1 h and vented. The resulting mixture was poured into acidified
methanol (50 mL of a 1M HCl solution). A solid precipitated and was
separated by filtration, washed with methanol, and dried under vacuum
Reaction of Cp*TiCl3/MAO and VC (Al/Ti ) 1000). An NMR
tube was charged with Cp*TiCl3 (1.0 mg, 3.5 µmol) and dried MAO
(3.5 mmol of Al; Al/Ti ) 1000). C6D6 (0.5 mL) and toluene (0.06
mmol, internal standard) were added, and VC (0.46 mmol) was
condensed in at -196 °C. The tube was warmed to room temperature,
and a 1H NMR spectrum was recorded. The tube was maintained at 80
°C for 24 h. The solution was dark red. The tube was cooled to room
temperature, and a 1H NMR spectrum was recorded which established
the presence of atactic oligopropylene. No PVC was formed. Com-
parison of the integrated intensities of VC and toluene before and after
the reaction established that 60% of the VC was consumed (79 equiv
vs Ti).
1
to yield 1.1 g (1% conversion) of colorless PVC. H NMR (THF-d8,
50 °C): δ 4.59 (m, CHCl), 4.47 (m, CHCl), 4.34 (m, CHCl), 2.40-
2.10 (m, CH2). 13C NMR (C6D5Cl, 100 °C): δ 57.1 (m, CHCl, rr triad),
56.1 (m, CHCl, mr triad), 55.2 (m, CHCl, mm triad), 47.5 (m, CH2, rrr
tetrad), 47.1 (m, CH2, rmr tetrad), 46.7 (m, CH2, mrr tetrad), 46.1 (m,
CH2, mmr tetrad), 46.0 (m, CH2, mrm tetrad), 45.3 (m, CH2, mmm
tetrad); triad tacticity 0.32 rr, 0.51 mr, 0.17 mm (R ) 0.58). GPC: Mw
) 180 479; Mn ) 36 217; Mw/Mn ) 5.0. Use of higher amounts of
oxygen did not result in the formation of PVC.
Reaction of Cp*Ti(OMe)3/MAO and VC (Al/Ti ) 10). A 100
mL Fisher-Porter bottle was charged with Cp*Ti(OMe)3 (0.011 g,
0.040 mmol). VC (2.1 g, 34 mmol) was condensed in at -78 °C, and
MAO (0.080 g of a 30 wt % toluene solution diluted with 10 mL of
CH2Cl2; 0.40 mmol of Al) was injected by syringe. The bottle was
warmed to room temperature and stirred for 24 h at 25 °C. The bottle
was vented, and the reaction mixture was poured into acidified MeOH
(50 mL of a 1M HCl solution). A solid precipitated and was separated
by filtration, washed with MeOH, and dried under vacuum, yielding
0.76 g (36% conversion) of colorless PVC. GPC: Mw ) 91 066; Mn )
27 241; Mw/Mn ) 3.3. Tacticity by 13C{1H} NMR: 0.36 rr, 0.49 mr,
0.15 mm; R ) 0.61.
Reaction of Cp*TiCl3/MAO and VC (Al/Ti ) 10). A 100 mL
glass Fisher-Porter bottle was charged with Cp*TiCl3 (0.014 g, 0.048
mmol). VC (2.1 g, 34 mmol) was condensed in at -40 °C, and MAO
(0.100 g of a 30 wt % toluene solution diluted with 10 mL of CH2Cl2;
0.50 mmol of Al) was injected by syringe. The bottle was warmed to
room temperature and stirred for 24 h at 25 °C. The bottle was vented,
and the reaction mixture was poured into acidified MeOH (50 mL of
a 1M HCl solution). A solid precipitated and was separated by filtration,
washed with MeOH, and dried under vacuum, yielding 0.14 g (6.8%
conversion) of colorless PVC. GPC: Mw ) 117 928; Mn ) 21 950;
Mw/Mn ) 5.4. Tacticity by 13C{1H} NMR: 0.40 rr, 0.48 mr, 0.12 mm;
R ) 0.64. This procedure was repeated at 10 times the scale, and 0.81
g (4% conversion) of PVC was obtained (R ) 0.64). Similar results
were obtained when the reaction was conducted in the absence of light.
Reaction of PVC with MAO. A Schlenk tube was charged with
PVC (0.275 g, 4.4 mmol of monomer, commercial sample prepared at
80 °C), MAO (0.128 g of a 30 wt % solution in toluene; 0.63 mmol of
Al), and CH2Cl2 (10 mL). The solution was stirred for 24 h at 23 °C.
The solution turned dark red over 6 h. Acidified methanol (50 mL of
a 1 M HCl solution) was added to quench the reaction and precipitate
the PVC. The PVC was separated by filtration, washed with methanol
Reaction of [(C5Me4SiMe2NtBu)ZrMe(C6D6)][B(C6F5)4] and VC.
An NMR tube was charged with (C5Me4SiMe2NtBu)ZrMe2 (0.014 g,
0.038 mmol) and [Ph3C][B(C6F5)4] (0.035 g, 0.038 mmol), and C6D6
(0.5 mL) was added by vacuum transfer at -196 °C. The tube was
warmed to 23 °C, and [(C5Me4SiMe2NtBu)ZrMe(C6D6)][B(C6F5)4] (8)
was partially separated as an orange oil from the pale yellow solution.
The 1H NMR spectrum contained resonances for 8 and Ph3CMe along
with broad resonances assigned to the fraction of 8 present in the oil
phase. VC (0.23 mmol) was added by vacuum transfer at -196 °C.
The tube was maintained at room temperature for 2 h and then heated
1
to 50 °C for 2 h. H NMR analysis showed that 1 equiv of VC had
been consumed versus Ph3CMe, the Zr-Me resonances of 8 had
disappeared, and new (C5Me4SiMe2NtBu)Zr resonances had appeared.
The volatiles were removed under vacuum, leaving an orange oil
presumed to be [(C5Me4SiMe2NtBu)ZrCl][B(C6F5)4]. The NMR spec-
trum of the volatiles contained resonances for free propylene. Solid
[NBu3Bn]Cl (0.012 g, 0.038 mmol) was added to the orange oil, and
C6D6 (0.5 mL) was added by vacuum transfer at -196 °C. The tube
was warmed to room temperature to afford a mixture of an orange oil
([NBu3Bn][B(C6F5)4]) suspended in a pale yellow solution. A 1H NMR
spectrum showed that (C5Me4SiMe2NtBu)ZrCl2 had formed in 50%
yield versus Ph3CMe. Only resonances for free B(C6F5)4- were observed
in the 19F NMR spectrum.
Generation and Trapping of [(C5Me4SiMe2NtBu)ZrCl][B(C6F5)4].
An NMR tube was charged with (C5Me4SiMe2NtBu)ZrMe2 (0.011 g,
0.030 mmol) and (C5Me4SiMe2NtBu)ZrCl2 (0.012 g, 0.030 mmol), and
C6D6 (0.5 mL) was added by vacuum transfer at -196 °C. The tube
was warmed to 23 °C. The 1H NMR spectrum established that
conversion to (C5Me4SiMe2NtBu)Zr(Me)Cl was complete. 1H NMR
(C6D6): δ 2.04 (s, 3H), 1.96 (s, 3H), 1.94 (s, 3H), 1.85 (s, 3H), 1.35
(s, 9H), 0.44 (s, 3H), 0.40 (s, 3H), 0.25 (s, 3H). Solid [Ph3C][B(C6F5)4]
(0.055 g, 0.060 mmol) was added, and an orange oil (presumably
[(C5Me4SiMe2NtBu)Zr(Cl)][B(C6F5)4]) separated from the pale yellow
solution. A 1H NMR spectrum showed that Ph3CMe was present. The
tube was heated to 50 °C for 2 h. No change was observed by 1H NMR.
Solid [NBu3Bn]Cl (0.019 g, 0.060 mmol) was added, yielding a mixture
of an orange oil ([NBu3Bn][B(C6F5)4]) suspended in a pale yellow
9
808 J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003

Metal-Catalyzed Insertion Polymerization of Vinyl Chloride
A R T I C L E S
1
solution. A H NMR spectrum established that (C5Me4SiMe2NtBu)-
(0.5 mL) was added by vacuum transfer at -196 °C. The tube was
warmed to 23 °C, and a dark orange oil separated from the pale yellow
solution. The 1H NMR spectrum contained resonances for the dinuclear
complex [{(C5Me4SiMe2NtBu)TiMe}2(µ-Me)][B(C6F5)4],64b broad reso-
nances at similar chemical shifts assigned to the fraction of the dinuclear
complex that was present in the oil phase, and Ph3CMe. VC (0.22
mmol) was added by vacuum transfer at -196 °C. The tube was heated
ZrCl2 had formed in 63% yield versus Ph3CMe. The 19F NMR spectrum
contained only resonances for free B(C6F5)4
-
.
Reaction of [(C5Me4SiMe2NtBu)ZrMe(C6D6)][B(C6F5)4] with
[NBu3Bn]Cl. A suspension of [(C5Me4SiMe2NtBu)ZrMe(C6D6)]-
[B(C6F5)4] (0.033 mmol) in C6D6 was prepared in an NMR tube as
described above. The tube was heated to 50 °C for 2 h. Solid [BnBu3N]-
Cl (0.033 mmol) was added, yielding a mixture of an orange oil in a
pale yellow solution. A 1H NMR spectrum of the solution phase
established that (C5Me4SiMe2NtBu)Zr(Me)Cl had formed in 60% yield
versus Ph3CMe; no (C5Me4SiMe2NtBu)ZrCl2 was present.
1
to 50 °C for 1 h. H NMR analysis revealed that 1 equiv of VC had
been consumed versus Ph3CMe. The volatiles were removed under
vacuum leaving an orange solid presumed to be [(C5Me4SiMe2NtBu)-
TiCl][B(C6F5)4]. The 1H NMR spectrum of the volatiles contained weak
resonances for free propylene. Solid [NBu3Bn]Cl (0.011 g, 0.037 mmol)
was added to the orange solid, and C6D6 (0.5 mL) was added by vacuum
transfer at -196 °C. The tube was warmed to room temperature to
afford a mixture of an orange oil suspended in a pale yellow solution.
Reaction of [(C5Me4SiMe2NtBu)ZrMe(C6D6)][B(C6F5)4] with Pro-
pylene. A C6D6 solution of [(C5Me4SiMe2NtBu)ZrMe(C6D6)][B(C6F5)4]
(0.038 mmol) was prepared in an NMR tube as described above.
Propylene (0.76 mmol) was condensed into the tube at -196 °C. The
tube was warmed to 23 °C, at which point it became hot. A H NMR
spectrum obtained after 30 min at 23 °C established that the propylene
was completely polymerized to low molecular weight atactic polypro-
pylene. The unsaturated end groups were exclusively vinylidene. H
NMR (C6D6): δ 4.83 (s, 1H, dCH2), 4.78 (s, 1H, dCH2), 1.70 (br,
sCH), 1.28 (br, CH2-anti-meso), 1.13 (br, CH2-racemic), 0.92 (CH2-
syn-meso and sMe). Mn (by H NMR) ) 1030 (average degree of
1
A
1H NMR spectrum established that (C5Me4SiMe2NtBu)TiCl2 had
formed in 40% yield versus Ph3CMe. Only resonances for free
B(C6F5)4 were present in the 19F NMR spectrum.
-
1
Reaction of (C5Me4SiMe2NtBu)TiMe2/[Ph3C][B(C6F5)4] and Pro-
pylene. An NMR tube was charged with (C5Me4SiMe2NtBu)TiMe2
(0.046 mmol), [Ph3C][B(C6F5)4] (0.046 mmol), and C6D6 (0.5 mL).
Propylene (0.64 mmol) was condensed in at -196 °C. The tube was
warmed to 23 °C, at which point partial gelling of the solution phase
1
polymerization ) 24).
Reaction of [(C5Me4SiMe2NtBu)ZrMe(C6D6)][B(C6F5)4] with Pro-
pylene in the Presence of VC. A C6D6 solution of [(C5Me4SiMe2-
NtBu)ZrMe(C6D6)][B(C6F5)4] (0.048 mmol) was prepared in an NMR
tube as described above. VC (0.48 mmol) and propylene (0.48 mmol)
were condensed into the NMR tube at -196 °C. The tube was warmed
to 23 °C, at which point it became warm. A 1H NMR spectrum obtained
after 30 min at 23 °C established that 25% of the propene remained
1
was observed. A H NMR spectrum obtained after 30 min at 23 °C
established that the propylene was completely consumed and atactic
polypropylene had formed. No unsaturated end groups were discernible
by 1H NMR. The contents of the tube were added to acidified methanol
(50 mL of a 1 M HCl solution) to precipitate the polypropylene. The
polypropylene was separated by filtration, washed with methanol (3 ×
20 mL), and dried under vacuum (25 mg). 1H NMR (1,2-dichloroben-
zene-d4, 140 °C): δ 1.58 (br, -CH), 1.22 (br, CH2-anti-meso), 1.07
1
and atactic polypropylene had formed. No further change in the H
NMR spectrum was observed after 24 h. The unsaturated end groups
observable in the H NMR spectrum comprised 13% allylic (δ 5.78
(m, 1H, dCHR), 5.03 (m, 2H, dCH2)) and 87% vinylidene (δ 4.83 (s,
1H, dCH2), 4.78 (s, 1H, dCH2)). Aliphatic 1H NMR resonances
(C6D6): δ 1.73 (br, sCH), 1.28 (br, CH2-anti-meso), 1.13 (br, CH2-
racemic), 0.91 (CH2-syn-meso and sMe). Mn (by H NMR) ) 942
(average degree of polymerization ) 23).
1
(br, CH2-racemic), 0.84 (CH2-syn-meso and -Me). GPC: Mw
1 672 000, Mn ) 462 000, Mw/Mn ) 3.6.
)
Reaction of (C5Me4SiMe2NtBu)TiMe2/[Ph3C][B(C6F5)4] with VC
and Propylene. The above experiment was repeated except VC (0.46
mmol) and propylene (0.78 mmol) were condensed into the NMR tube
at -196 °C. The tube was warmed to 23 °C, at which point it became
warm to the touch. A 1H NMR spectrum obtained after 30 min
established that 6% of the propylene remained and atactic polypropylene
had formed. The unsaturated end groups were exclusively allylic. The
contents of the tube were added to acidified methanol (50 mL of a 1
M HCl solution); however, no polypropylene precipitate was observed.
The resulting mixture was extracted with hexanes (3 × 20 mL), and
the extract was dried under vacuum, leaving a small amount of an oil.
1H NMR (1,2-dichlorobenzene-d4, 140 °C): δ 5.73 (m, 1H, dCHR),
4.90 (m, 2H, dCH2), 1.62 (br, sCH), 1.25 (br, CH2-anti-meso), 1.11
(br, CH2-racemic), 0.87 (CH2-syn-meso and sMe). Mn determined by
1H NMR ) 1610 (average degree of polymerization ) 38).
1
Reaction of (C5Me4SiMe2NtBu)ZrCl2/MAO and VC. A 100 mL
Fisher-Porter bottle was charged with (C5Me4SiMe2NtBu)ZrCl2 (0.009
g, 0.02 mmol), MAO (4 g of a 30 wt % toluene solution; 20 mmol of
Al; Al/Zr ) 1000) and 20 mL of toluene. VC (2 mL, 1.8 g, 29 mmol)
was condensed into the Fisher-Porter bottle tube at -196 °C. The
bottle was heated to 80 °C for 2 h. The bottle was vented, and the
reaction mixture was poured into acidified methanol (100 mL of a 1
M HCl solution). The resulting mixture was extracted with hexanes (3
× 50 mL). The extract was dried under vacuum, yielding 0.070 g of a
pale yellow oil. The H and 13C NMR spectra established that the oil
1
was atactic oligopropylene and contained a process oil from the MAO.
The reaction was repeated with varying concentrations of MAO (800-
17 000 Al/Zr), temperatures (25-80 °C), and times (2-24 h) with
similar results.
Reaction of (C5Me4SiMe2NtBu)TiMe2/ [Ph3C][B(C6F5)4] and VC.
An NMR tube was charged with (C5Me4SiMe2NtBu)TiMe2 (0.012 g,
0.037 mmol) and [Ph3C][B(C6F5)4] (0.034 g, 0.037 mmol), and C6D6
Acknowledgment. This work was supported by the Edison
Polymer Innovation Corporation PVC Technology Consortium
and the Department of Energy (DE-FG02-00ER15036).
JA028530D
9
J. AM. CHEM. SOC. VOL. 125, NO. 3, 2003 809