Chiral epoxides by desymmetrizing deprotonation of meso-epoxides

REVIEW

Catalysts for the Living Insertion Polymerization of Alkenes:

Access to New Polyolefin Architectures Using Ziegler Natta Chemistry

Geoffrey W. Coates,* Phillip D. Hustad, and Stefan Reinartz

Dedicated to Professor Maurice Brookhart and Professor Robert H. Grubbs on the occasion of their 60th birthdays

Coordination insertion polymeriza-

tion systems have long been superior

to their anionic, cationic, and radical

polymerization counterparts with re-

gard to stereochemical control. How-

ever, until five years ago, these metal-

based insertion methods were inferior

to ionic and radical mechanisms in the

category of living polymerization,

which is simply a polymerization that

occurs with rapid initiation and negli-

gible chain termination or transfer. In

the last half decade, the living insertion

polymerization of unactivated olefins

has emerged as a powerful tool for the

synthesis of new polymer architectures.

Materials available today by this route

range from simple homopolymers such

as linear and branched polyethylene, to

atactic or tactic poly(a-olefins), to end-

functionalized polymers and block co-

polymers. This review article summa-

rizes recent developments in this rap-

idly growing research area at the inter-

face of synthetic and mechanistic

organometallic chemistry, polymer

chemistry, and materials science. While

special emphasis is placed on polymer

properties and novel polymeric archi-

tectures, most of which were inacces-

sible just a decade ago, important

achievements with respect to ligand

and catalyst design are also highlight-

ed.

Keywords: alkenes

¥ block copoly-

mers ¥ homogeneous catalysis ¥ living

polymerization ¥ Ziegler Natta catal-

ysis

as the synthesis of a wide array of polymer architectures.^[2]For

example, the initiation of multiple polymer chains from a

central core results in the formation of a star-branched

polymer, while the consecutive addition of two monomers to a

single initiator produces a diblock copolymer.^[3]Living

methods also allow the synthesis of end-functional polymers

if special initiation and/or quenching methods are employed.

Of course, living polymerizations have the liability that each

catalyst only forms one chain, in contrast to common alkene

polymerization catalysts that can produce thousands of chains

each as a result of periodic chain transfer or termination

events. (Note: in this review, we refer to living species for

alkene polymerization as ™catalysts∫, not ™initiators∫, to

emphasize the fundamental catalytic event of monomer

enchainment, not polymer chain formation.) The real value

of living polymerization methods is that they allow the

creation of virtually limitless types of new materials from a

basic set of available monomers.

1. Introduction

Polymeric materials are currently more indispensable to

modern society than at any other point in history. The

potential applications of a polymer are determined by its

physical and mechanical properties, which in turn are defined

by the morphology (solid-state arrangement) of the polymer.

Polymer morphology largely depends on the composition and

architecture of the polymer. Therefore, the development of

synthetic methods for the polymerization of a wide range of

monomers with control over the stereochemistry and molec-

ular weight of the resultant polymers is a long-standing

scientific challenge. A primary goal of synthetic polymer

chemistry that has existed for the last half century is the

development of chain-growth polymerization methods that

enable consecutive enchainment of monomer units without

termination. Such techniques, now known as living polymer-

izations,^[1]allow both precise molecular weight control as well

Based on annual production volume, polyolefins are by far

the most important commercial class of synthetic polymers.

Since the initial discoveries of Ziegler^[4]and Natta,^[5]remark-

able advances have been reported concerning the control of

comonomer incorporation as well as dramatic improvements

in activity. Homogeneous olefin polymerization catalysts now

exist that are unparalleled in all of polymer chemistry

[*] Prof. Dr. G. W. Coates, P. D. Hustad, Dr. S. Reinartz

Department of Chemistry and Chemical Biology

Baker Laboratory, Cornell University

Ithaca, New York 14853-1301 (USA)

Fax : (1)607-255-4137

E-mail: gc39@cornell.edu

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002

1433-7851/02/4112-2237 $ 20.00+.50/0

2237

G. W. Coates et al.

REVIEW

concerning the detailed control of macromolecular stereo-

chemistry.^[6]However, olefin insertion catalysts have always

been inferior to their other chain-growth counterparts in one

respect. While extraordinary advances in living/controlled

molecular weight to the number average molecular weight

(M_w/M_nꢀ 1);

6) block copolymers can be prepared by sequential monomer

addition;

polymerization have been discovered by using anionic,^[7]

7) end-functionalized polymers can be synthesized.^[13]

Few polymerization systems, whether ionic-, radical-, or

metal-mediated, that are claimed to proceed by a living

mechanism have been shown to meet all of these criteria. This

review will therefore include all systems that claim living

olefin polymerization, providing that a substantial number of

the key criteria have been met.

Common features of alkene polymerization catalyst sys-

tems are chain transfer and elimination reactions that

terminate the growth of a polymer chain and result in the

initiation of a new polymer chain by the catalyst (Scheme 1).

For example, in metallocene catalysts, consecutive alkene

insertion into the metal carbon bond connecting the catalyst

and polymer chain^[14]proceeds until b-hydrogen and/or b-

alkyl elimination occurs.^[15]When alkylaluminum cocatalysts

are employed, an additional termination route is chain

transfer to the aluminum center.^[15]In many systems, the

lifetime of chain formation is on the order of seconds,

rendering sequential monomer addition methods for block

copolymer synthesis futile. Several strategies have been

devised to decrease the rate of chain termination relative

to that of propagation such that living systems can be

formed. The first consideration in many cases is simply

lowering the polymerization temperature of an ordinary non-

living catalyst system to achieve living or at least controlled

11]

cationic,^[8]and radical-based^[9

polymerization, until very

recently there existed a comparative lack of living olefin

polymerization systems. A significant number of advances

have been reported in the last half decade, prompting us to

review the area of catalysts for alkene polymerization that

proceed without appreciable chain transfer or termination. In

this review, we only address the living polymerization of

unactivated alkenes by insertion methods; ring-opening

metathesis polymerization (ROMP), the polymerization of

conjugated dienes and acrylates, group transfer polymeriza-

tion, and CO/alkene copolymerization will not be addressed.

The seven generally accepted criteria for a living polymer-

ization are:

1) polymerization proceeds to complete monomer conver-

sion, and chain growth continues upon further monomer

addition;

2) number average molecular weight (M_n) of the polymer

increases linearly as a function of conversion;^[12]

3) the number of active centers remains constant for the

duration of the polymerization;

4) molecular weight can be precisely controlled through

stoichiometry;

5) polymers display narrow molecular weight distributions,

described quantitatively by the ratio of the weight average

Geoffrey W. Coates, born in Evansville,

Indiana in 1966, obtained a B.A. degree in

chemistry from Wabash College in 1989 and

a Ph.D. in organic chemistry from Stanford

University in 1994. In his thesis work, under

the direction of Robert M. Waymouth, he

investigated the stereoselectivity of metal-

locene Ziegler Natta catalysts. Following

his doctoral studies, he was an NSF Post-

doctoral Fellow with Robert H. Grubbs at

P. D. Hustad

G. W. Coates

S. Reinartz

the California Institute of Technology. Dur-

ing the summer of 1997, he joined the

Department of Chemistry at Cornell University where he is currently Professor of Chemistry. His main research interests

are the design, synthesis, characterization, and applications of polymers with an emphasis on catalytic transformations and

the control of stereochemistry.

Phillip D. Hustad, born in Bristol, Tennessee in 1975, obtained a B.S. degree in Chemistry in 1997 from Emory & Henry

College in Emory, Virginia. Inspired by undergraduate research with Prof. Koji Nakanishi at Columbia University, he then

entered the Ph.D. program in Chemistry at Cornell University. He is currently working under the guidance of Geoffrey W.

Coates on the development and applications of new catalysts for living olefin polymerization.

¬

Stefan Reinartz, born in Hilden, Germany in 1972, studied chemistry at Heinrich-Heine-Universit‰t D¸sseldorf, Universite

de Nantes (France), and the University of North Carolina at Chapel Hill (UNC). After obtaining his Diplom in 1998, he

returned to UNC, where he earned his Ph.D. in inorganic chemistry in 2001. His dissertation under the direction of

Joseph L. Templeton and Maurice Brookhart focused on organometallic platinum chemistry relevant to bond activation

processes. He is currently a DAAD Postdoctoral Fellow with Geoffrey W. Coates at Cornell University.

2238

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

Living Polymerization of Alkenes

REVIEW

2.1. Propylene Polymerization with Vanadium

Compounds

In the 1960s, Natta and co-workers discovered that

vanadium tetrachloride activated with diethylaluminum

chloride produced syndio-enriched polypropylenes at

À788C.^[18]The rate of polymerization was constant for long

reaction times (ca. 50 h), and molecular weight increase with

time was nearly linear over 25 h. The [Al]/[V] ratio had a

dramatic effect on the lifetime of active chains; an increase in

cocatalyst concentration was mirrored by a decrease in

lifetime, implicating chain transfer to aluminum centers as a

source of termination. The syndiospecificity of this catalyst

was found to be the result of regioregular secondary (2,1)

insertion of propylene.^[19]Anisole was also found to have a

dramatic effect on both catalytic activity and the average

lifetime of active chains.^[20]In later studies, polypropylenes

from this catalytic system were found to be unimodal with

narrow molecular weight distributions (M_w/M_n 1.4 1.9).^[21]

Although this catalytic system produced polypropylenes

that were nearly living, the first true example of living olefin

polymerization did not appear until over a decade later. In

1979, Doi et al. reported the first catalytic olefin polymer-

ization system to satisfy all the requirements for a living

polymerization. The catalyst, [V(acac)₃] (1a; Figure 1; acac

acetylacetonate) activated with Et₂AlCl, produced partially

Scheme 1. Mechanisms of propagation and chain transfer in Ziegler

Natta catalyzed olefin polymerization.

behavior. Since b-hydrogen and alkyl elimination processes

are unimolecular while propagation is bimolecular, a lowering

in temperature more adversely affects elimination processes

relative to enchainment. Since precipitation of polymers from

solution at low temperatures can hinder the controlled nature

of a living polymerization, it is generally advantageous to

perform reactions at ambient temperature. Therefore, a

second strategy for discovering living systems is to design

new catalysts through empirical modification and/or compu-

tational methods.^[16]By creating species that are incapable of

common termination reactions at room temperature, living

catalysts have been devised. A final consideration is to

eliminate the use of alkyl aluminum cocatalysts that give the

potential for chain-transfer reactions. In this regard, the

development of weakly coordinating anions has made sig-

nificant advances in living olefin polymerization possible.^[17]

O

V

O

V

O

1a

1b

Figure 1. Vanadium catalyst precursors for living olefin polymerization.

syndiotactic polypropylenes (81% r dyads; [r] 0.81)^[15]at

À788C with extremely narrow molecular weight distributions

(M_w/M_n 1.05 1.20; Scheme 2).^{[22, 23]}Molecular weight in-

creased linearly with time to values as high as 100000 gmol^À1,

and the number of polymer chains remained constant during

the course of the reaction. However, the catalyst exhibited

living behavior only at temperatures below À658C; reactions

at temperatures as low as À488C resulted in polypropylenes

with broadened molecular weight distributions (M_w/M_n

1.37 1.45). Also, catalytic activity (ca. 4 kg_PPmol_V^À1h^À1)

suffered from the fact that only about 4% of the vanadium

centers were active. The nature of the aluminum cocatalyst

also had a dramatic effect on polymerization activity, living

2. Vanadium Catalysts for Living Olefin

Polymerization

Olefin polymerization emerged in the 1950s as a principal

area of organometallic research when Ziegler and Natta and

co-workers discovered that titanium chloride in the presence

of alkylaluminum compounds was an efficient catalyst for

polymerization of ethylene and propylene.^{[4, 5]}Following this

breakthrough, vanadium compounds in combination with

alkylaluminum compounds were shown to be active catalysts

for the polymerization of olefins. Through modification of

ligands and activators, vanadium catalysts were also found to

polymerize olefins in a living fashion at low temperatures. The

utility of these living systems has been demonstrated through

the synthesis of a wide variety of end-functionalized poly-

propylenes (PPs) and PP-containing block copolymers. How-

ever, polymerizations with these systems are limited to

temperatures at or below À408C, and living behavior is

restricted to only a few monomers.

Scheme 2. Living polymerization of propylene with vanadium catalysts.

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

2239

G. W. Coates et al.

REVIEW

behavior, and stereoselectivity, suggesting a bimetallic active

species. Particularly, use of the more electron-deficient

cocatalyst EtAlCl₂led to an increase in chain transfer to

monomer, giving polymers with distributions approaching the

theoretical value for polymerizations involving chain transfer

(M_w/M_n 2.0).^{[23, 24]}The activating effect of anisole was also

demonstrated for the 1a/Et₂AlCl catalyst system.^[25]The

additive led to a threefold increase in the number of active

vanadium centers without adversely affecting livingness or

syndiospecificity.

hexadiene also displayed living behavior. The distribution of

the polymer microstructure varied with the mole fraction of

hexadiene in the copolymer; high diene content gave equal

amounts of the two repeat units, while lower incorporation

resulted in exclusive formation of tetramethylene-1-vinylenes.

However, the mechanism for formation of this unique

polymer is unclear.

2.2. End-Functionalized Polypropylenes from Vanadium

Compounds

To address the problem of activation, Doi and co-workers

studied a variety of vanadium complexes and found that

subtle ligand modifications had dramatic effects on catalytic

Doi and co-workers have demonstrated the utility of the

living vanadium catalysts through the synthesis of several

tailor-made polymers.^{[33 37]}By reacting active polymer chains

with additives, polypropylenes with a wide variety of func-

tional end groups can easily be prepared with these living

catalyst systems (Scheme 4). In addition to providing impor-

tant mechanistic information, these functional polymers

display unique properties and have also been used as macro-

initiators for the synthesis of block copolymers.

In an effort to synthesize new end-functionalized polymers,

a living vanadium-polypropylene species was quenched with

iodine at À788C to give a monodisperse iodine-functionalized

polypropylene (M_w/M_n 1.15).^[33]NMR spectroscopy re-

vealed that the structure of this end group resulted from

reaction of I₂with a secondary alkylvanadium compound,

providing evidence for a 2,1 insertion mechanism. The iodine

functionality was used to prepare an amine-terminated

polypropylene by reacting the polymer with excess ethyl-

enediamine in THF, followed by basic workup.^[35]These

polymers have also been used as macroinitiators for the

preparation of diblock copolymers (see Section 2.3).

By reacting active living vanadium centers with carbon

monoxide, Doi et al. have prepared aldehyde-terminated

polypropylenes (Scheme 4).^[38]Again, CO was found to insert

quantitatively into secondary vanadium carbon bonds. This

aldehyde functionality has also been used to prepare hydroxy-

functionalized polypropylenes by reduction of the aldehyde

with LiAlH₄in Et₂O, followed by acidic hydrolysis.^[36]

Hydroxy-terminated polypropylenes were also prepared with

moderate success by reaction of an active vanadium species

with propylene oxide.^[36]The insertion of the epoxide into the

growing polymer chain occurred with decent regioselectivity,

preferentially adding at the unsubstituted carbon atom (7/1)

to give the secondary alcohol. However, the reaction also

yielded a fraction of allyl-terminated polypropylene from an

undesirable elimination reaction (20%).

Polypropylene macromonomers containing methacryl func-

tionality were prepared by addition of ethylene glycol

dimethacrylate (EGDM) to a living chain end.^[39]Propylene

polymerization was conducted for 1 h at À608C with the

complex 1b/Et₂AlCl, giving a low molecular weight polypro-

pylene (M_n 3100 gmol^À1, M_w/M_n 1.13). Addition of an

excess of EGDM to this living chain end at À558C and

reaction for an additional 1 h resulted in quantitative capping

of the chain ends without formation of EGDM homopolymer

(M_n 4000 gmol^À1, M_w/M_n 1.10). Both IR and NMR spec-

troscopy revealed the presence of the methacryl unit in the

28]

behavior.^[26Through these studies, complex 1b was found

to be highly active (relative to 1a) for the living polymer-

ization of propylene (ca. 100 kg_PPmol_V^À1h^À1; Scheme 2). The

increase in activity was attributed to a greater number of

active vanadium centers in the reaction; the number of

polymer chains per metal center approached unity for this

catalytic system. Again, the polypropylene was partially

syndiotactic ([r] 0.80), consistent with a secondary insertion

mechanism. The catalyst also displayed living behavior at

temperatures as high as À408C, giving high molecular weight

polymers with low polydispersity (M_nup to 100000 gmol^À1,

M_w/M_n 1.2 1.4).

Despite the success of these vanadium catalysts with

propylene, living behavior is limited to only a few monomers

with this catalytic system. For example, ethylene polymer-

izations with 1a/Et₂AlCl at À788C gave high molecular

weight polyethylenes (PEs; >300000 gmol^À1) with molecular

weight distributions consistent with the probable distribution

for a single-site catalyst (M_w/M_n 2.0).^[29]However, ethylene/

propylene copolymerizations were living; reaction of ethylene

and propylene with 1b/Et₂AlCl at À608C rapidly produced

a

very high molecular weight copolymer (M_n

1020000 gmol^À1) with a narrow polydispersity (M_w/M_n

1.22).^[30]These catalysts were also found to be inactive for

polymerization of higher a-olefins. For example, reaction of

1-pentene with 1a/Et₂AlCl resulted in formation of heptane

and 3-methylhexane, the products of a single 1-pentene

addition to a vanadium ethyl species with 2,1- and 1,2-

regiochemistry, respectively.^[31]No further reaction was de-

tected, presumably due to the sterically congested nature of

the catalytic site following the insertion. However, the

vanadium system did produce living polymers with 1,5-

hexadiene; at À788C, the diene was polymerized to low

molecular weight polymer (M_n 6600 gmol^À1) with a narrow

molecular weight distribution (M_w/M_n 1.4).^[32]This polymer,

however, did not possess the typical methylene-1,3-cyclo-

pentane microstructure obtained with other insertion poly-

merization catalysts. By NMR analysis, the microstructure

was determined to contain tetramethylene-1-vinylene units

(46%) as well as methylene-1,3-cyclopentane structures

(54%; Scheme 3). Copolymerization of propylene and 1,5-

1a/Et₂AlCl

–78 °C

Scheme 3. Polymerization of 1,5-hexadiene with a vanadium catalyst.

2240

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

Living Polymerization of Alkenes

REVIEW

Alkenes

P

75%

25%

, H⁺

Methacryls

O

Phenyls

O

P

, H⁺

P

2

Styrene

H⁺

O

1.7

P

L_nV

1) LiAlH₄

2) EtOAc

ethylene

diamine

CO, H⁺

I₂

H

HO

P

H₂N

P

3) H₂SO₄

NaOH

I

N

H

O

, H⁺

Hydroxy compounds

Aldehydes

Iodides

Amines

P

HO

70%

10%

Hydroxy compounds

20%

Alkenes

Scheme 4. Synthesis of end-functional polypropylenes with a vanadium catalyst.

polymer; interestingly, an average of 1.7 molecules of EGDM

were reported to be present per chain.

vanadium catalysts. To this end, Doi and co-workers reported

the synthesis of both AB- and ABA-type block copolymers

from ethylene and propylene (Scheme 5). As described in

Section 2.1, the reaction of 1a/Et₂AlCl/anisole with propylene

at À788C generates a living polypropylene chain end.^[25]In

the presence of propylene, a small amount of ethylene was

added to this living chain, resulting in rapid formation of an

AB-type copolymer containing syndio-enriched polypropy-

Finally, additives that produce polypropylenes with alkenyl

and phenyl end groups were explored by Doi and co-

workers.^[36]By adding butadiene to living polypropylene

prepared with 1a/Et₂AlCl, a polymer containing alkene end

groups was formed after quenching. NMR spectroscopy

showed that the butadiene inserted with both 2,1 and 1,4

regiochemistry, producing polymers with both terminal

(75%) and internal (25%) olefins. Phenyl-terminated poly-

mers were also prepared in this fashion by addition of styrene

to a living polypropylene, which reacts quantitatively with the

growing polymer chain without formation of any styrene

homopolymer.

lene and ethylene/propylene rubber (EPR) domains.^[40]

A

sharp increase in both polymer yield and molecular weight

versus time was observed following the ethylene addition.

However, these rates quickly returned to values consistent

with propylene homopolymerization, indicating that ethylene

consumption was complete after a short time. The steady

increase in yield and molecular weight following the ethylene

addition indicated the continued formation of propylene

homopolymer to give an ABA-type triblock copolymer

(synPP-block-EPR-block-synPP).

2.3. Copolymers and Block Copolymers from Vanadium

Compounds

In addition to providing block copolymers based on non-

polar olefins by sequential monomer addition, the vanadium

catalysts have also been employed for the synthesis of block

copolymers from polar monomers by transforming the living

chain end to one capable of initiating a radical or cationic

One of the most attractive features of a living catalyst

system lies in its ability to produce well-defined block

copolymers by sequential monomer addition. Several meth-

ods have been described for block copolymer formation using

CO₂Me

1)

L_nV

P

CO₂Me

2)

3)

–78

25 °C

PP-block-EPR-block-PP

PP-block-PMMA

I₂

Ph

Na⁺

P

I

O

AgClO₄

P

50 °C

O

PP-block-PS

PP-block-PTHF

Scheme 5. Synthesis of block copolymers with a vanadium catalyst; EPR ethylene/propylene rubber.

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

2241

G. W. Coates et al.

REVIEW

polymerization (Scheme 5). Doi et al. discovered that the

polymerization of methyl methacrylate (MMA) with 1a/

Et₂AlCl at 258C displayed living character during the initial

stages of the reaction, giving low molecular weight poly-

(methyl methacrylate) (PMMA) with a narrow molecular

weight distribution (M_n 2400 gmol^À1, M_w/M_n 1.2).^[34]Co-

polymerizations of MMA with styrene suggested that the

reaction proceeded by a radical pathway. This new method for

MMA polymerization was employed for the synthesis of

polypropylene/PMMA diblock copolymers. By reacting pro-

pylene with 1a/Et₂AlCl at À788C, a living polypropylene

chain end was formed (M_n 16000 gmol^À1, M_w/M_n 1.2).

Addition of MMA to the reaction mixture and warming to

258C resulted in formation of a higher molecular weight

H

Me

Sm

M

thf

2a M = La

2b M = Nd

2c M = Lu

2d M = Sm

3

Me₃Si

M

SiMe₃

M

Si

H

Si

Me₃Si

SiMe₃

R

H

thf

Sm

Sm thf

H

Si

Me₃Si

SiMe₃

polymer containing

a syndio-enriched PMMA segment

(15 wt% PMMA, M_n 18000 gmol^À1, M_w/M_n 1.2).

4

5 M = Y, R = SiMe₂tBu

6 M = Sm, R = tBu

Another example of this strategy for block copolymer

formation was demonstrated in the synthesis of a copolymer

with a poly(tetrahydrofuran) (PTHF) domain. As described

in Section 2.2, Doi et al. synthesized a monodisperse iodine-

terminated polypropylene (M_n 16500 gmol^À1, M_w/M_n

1.15) by addition of I₂to a living polypropylene chain

end.^[33]This polymer was then dissolved in THF at 08C

and treated with AgClO₄, causing immediate precipita-

tion of AgI and concomitant formation of a cationic macro-

initiator capable of THF polymerization. After 96 h, a

monodisperse polymer (M_w/M_n 1.14) of higher molecular

weight was isolated. IR and NMR analysis revealed that

the polymer contained both polypropylene and PTHF do-

mains.

Me₃Si

tBu

Sm(thf)₂

Me₂Si

Me₃Si

tBu

7

Figure 2. Lanthanide catalysts for living olefin polymerization. THF

tetrahydrofuran.

polymerization at À788C with 2b showed that temperature

effects on possible chain termination reactions such as b-

hydride elimination were not important (T 258C, t 5 s,

M_n 590000 gmol^À1

;

M_w/M_n 1.81; T À788C, t 600 s,

M_n 648000 gmol^À1; M_w/M_n 1.95). Perfectly living behavior

of these hydride catalysts was believed to be impeded by

mass-transport effects and initiation-limiting dissociation of

the catalyst dimer.

Finally, tailor-made block copolymers can be prepared by

the coupling of iodine-terminated polypropylenes with mono-

functional or multifunctional living polymers. Doi et al.

exploited this strategy for the synthesis of a well-defined

polypropylene polystyrene diblock copolymer.^[29]Addition

of an iodine-terminated polypropylene (M_n 22000 gmol^À1,

Yasuda and co-workers reported in 1992 that complex 2d

was extremely active for the living, syndiospecific polymer-

ization of MMA ([r] 0.97 at À958C, [r] 0.91 at 08C)

to form high molecular weight polymers (M_nup to

560000 gmol^À1) with extremely narrow molecular weight

distributions (M_w/M_n< 1.05) via a coordination anionic poly-

merization mechanism.^[42]Immediately following this discov-

ery, the sequential addition copolymerization of ethylene and

polar monomers was reported to give block copolymers,

implicating a change of mechanism from insertion to coordi-

nation anionic during the course of the polymerization.^[43]

Since this material has been extensively reviewed elsewhere,

M_w/M_n 1.14) to

a

living polystyrene anion (M_n

13000 gmol^À1, M_w/M_n 1.27) resulted in formation of a higher

molecular weight polymer (M_n 35000 gmol^À1, M_w/M_n

1.25), consistent with the quantitative coupling of the two

component polymers.

3. Rare-Earth Metal Catalysts for Controlled

Alkene Polymerization

47]

Schumann, Marks, and co-workers showed as early as 1985

that organolanthanide complexes were promising candidates

for living olefin polymerization.^[41]Dimeric Cp*-based hy-

dride complexes 2a c (Figure 2; Cp* pentamethylcyclo-

pentadienyl) were extremely active for the polymerization of

ethylene, with turnover numbers of more than 1800 s^À1for 2a

at room temperature. Molecular weight distributions of

polymers from the lutetium catalyst 2c were reproducibly

less than two (M_n 96000 361000 gmol^À1; M_w/M_n 1.37

1.68), and the average number of polymer chains per

lanthanide center for catalysts 2a c was always less than

one. Based on these observations, it was speculated that these

systems were operating in a living fashion. Furthermore,

we will only highlight the general concept.^[44

In 1992, Yasuda et al. reported that samarium catalyst 3

(Figure 2) could effect the room-temperature block copoly-

merization of ethylene (insertion mechanism) and several

polar monomers (non-insertion mechanism) such as methyl

methacrylate, methyl acrylate, ethyl acrylate, d-valerolactone,

and e-caprolactone (CL).^[43]In a typical two-step procedure,

ethylene was first polymerized to a reactive polyethylene

(PE) (M_n 6600 27000 gmol^À1, M_w/M_n 1.39 2.01) in tol-

uene under atmospheric pressure, followed by the addition of

the respective polar monomer to form a linear diblock

copolymer with relatively narrow polydispersity (Scheme 6).

Reversal of monomer addition did not lead to block copoly-

2242

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

Living Polymerization of Alkenes

REVIEW

4. Cobalt, Niobium, and

Tantalum Catalysts for Living

Ethylene Polymerization

OMe

CO₂Me

3

Me

n

Cp*Sm O

2

Cp*Sm

2

m

n

20°C

CO₂Me

Polyethylene

M_n= 10 300 g mol^–1

M_w/M_n= 1.42

PE-block-PMMA

M_n= 24 200 g mol^–1

M_w/M_n= 1.37

In the beginning of the 1990s, a

number of catalyst systems for

ethylene polymerization were re-

ported in which the resulting poly-

ethylene displayed a surprisingly

narrow molecular weight distribu-

tion. The cobalt catalyst 8 (Ar'

3,5-(CF₃)₂C₆H₃), prepared by pro-

tonation of the alkene-bound pre-

cursor (Scheme 7), produced poly-

ethylenes with narrow polydisper-

sities for low molecular weight

samples (M_n 13600 gmol^À1; M_w/

OMe

CO₂Me

7

SmL_n

L_nSm

O

L_nSm

O

SmL_n

n

m

n

m

20°C

OMe

CO₂Me

Difunctional PE

PMMA-block-PE-block-PMMA

Scheme 6. Synthesis of polyethylene/poly(methyl methacrylate) diblock and triblock copolymers with

samarium catalysts.

mer formation. The resulting materials, such as PE-block-

M_n 1.17), while the polydispersity increased for samples of

higher molecular weight presumably due to mass transport

problems (M_n 48500 gmol^À1; M_w/M_n 1.71).^[53]In addition,

precipitation of polymer was regarded as a possible catalyst

deactivation route.

PMMA and PE-block-poly(CL) showed advantageous mate-

rials properties such as deep coloration with dyes when a short

polar block was present.

Recently, Yasuda and co-workers developed a binuclear

samarium complex 4 (Figure 2) which exhibited high activity

for the polymerization of ethylene (M_nup to 50000 gmol^À1

;

–

+

M_w/M_n 1.63 1.68) and efficiently formed diblock copoly-

mers such as PE-block-PMMA (M_nup to 70000 gmol^À1; M_w/

M_n 1.67 1.69) and polyethylene-block-poly(CL) (M_n

70000 gmol^À1; M_w/M_n 1.65).^[48]Structurally related com-

plexes 5 and 6 (Figure 2) were then applied in the first

controlled block copolymerization of 1-hexene and 1-pentene

with MMA and CL.^[49]Yasuda and co-workers have also

reported new divalent samarium complexes with bridging

bis(cyclopentadienyl) (Cp) ligands and applied them in the

polymerization of ethylene.^[50]In particular, racemic complex

7 provided encouraging results not only in the formation of

polyethylene but also in the polymerization of higher a-

olefins to give highly isotactic poly(a-olefins) (e.g. poly(1-

pentene): M_w 10600 gmol^À1, M_w/M_n 1.48, poly(1-hexene),

M_w 8300 gmol^À1, M_w/M_n 1.55). Cyclopolymerization of

1,5-hexadiene resulted in the formation of poly(methylene-

1,3-cyclopentane) with a cis-ring content of about 50% (M_w

28500 gmol^À1; M_w/M_n 1.88). Furthermore, since catalyst 7

operated by a mechanism involving coordination of ethylene

to two samarium centers followed by electron transfer to give

a telechelic ethylene-bridged dinuclear species, block copoly-

merization of ethylene and MMA resulted in formation

of the triblock PMMA-block-PE-block-PMMA copolymer

(Scheme 6).^[51]Corresponding trivalent lanthanide com-

pounds with bridging bis(Cp) ligands were not found to be

superior in polymerization activity.^[52]

BAr₄'

[H(OEt₂)₂][BAr' ]

4

Co

(MeO)₃P

Ar' = 3,5-(CF₃)₂C₆H₃

(MeO)₃P

R

H

R

R = H, Me

8 R = Me

Scheme 7. Synthesis of cobalt catalysts for olefin polymerization.

Based on this encouraging lead, Brookhart and co-workers

subsequently reported a slightly modified cobalt catalyst for

living ethylene polymerization and the synthesis of end-

functional polyethylenes.^[54]Highly electrophilic cobalt com-

plexes 9, when exposed to 1 atm of ethylene for 3 h at room

À

temperature followed by hydrogenolysis of the Co C_(alkyl)

bond, formed end-functional polyethylenes (M_nup to

20000 gmol^À1) with narrow polydispersities (M_w/M_n 1.11

1.16) (Scheme 8). ¹³C NMR analysis of the resulting polymer

revealed a polymer microstructure with no branching. The

presence of only one terminal methyl resonance signal

indicated that the initiating species was the b-aryl-substituted,

b-agostic complex (Scheme 8), and not its a-aryl-substituted,

b-agostic isomer, in which case the corresponding polymer

It should be mentioned that the organolanthanide com-

plexes discussed here do not entirely meet the generally

accepted criteria for living olefin polymerization, especially

with respect to molecular weight distribution. However,

through the combination of living anionic polymerization of

polar monomers with partially controlled olefin insertion

polymerization, these systems have found exciting applica-

tions in the synthesis of new materials with interesting

properties.

Scheme 8. Synthesis of end-functional polyethylenes with cobalt catalysts.

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

2243

G. W. Coates et al.

REVIEW

would have displayed two methyl resonance signals (i.e.

CH₃CH₂(CH₂CH₂)_nCH(CH₃)C₆H₄X). Furthermore, interac-

tion of the functional group with the cobalt center, which

could lead to catalyst deactivation, was prevented since the

metal center can not migrate past the aryl group. However,

broadening of the molecular weight distribution was again

observed for higher molecular weight polymers (M_n>

40000 gmol^À1).

ization upon addition of B(C₆F₅)₃. Furthermore, changing the

steric bulk of the catalyst system by replacing the Cp* ligand

of 11 with the less hindered Cp ligand resulted in a significant

broadening of the molecular weight distribution of the

resulting polyethylene (M_w/M_n 1.40), presumably due to

inferior catalyst stability.

Following these leads, Mashima and co-workers subse-

quently synthesized dimethyltantalum complexes 13a

c

Silyl-functionalized polyethylene could be generated

in an analogous fashion with the b-agostic com-

plex [Cp*{P(OMe)₃}CoCH₂CH(m-H)(CH₂)₄SiR₃][BAr'₄] (10)

(BAr'₄ B(3,5-(CF₃)₂C₆H₃)₄). Molecular weights and molec-

ular weight distributions for SiEt₃-functional polyethylenes

compared well with those of the aryl-substituted polymers.

However, Me₂SiCl-functional polyethylenes showed lower

molecular weight (M_n 10200 gmol^À1) and broader molec-

ular weight distribution (M_w/M_n 1.41), presumably due to

catalyst poisoning. While these cobalt complexes provided

routes to end-functional polyethylenes, there are a number of

drawbacks in these systems, namely the relatively low

molecular weight of the resulting polymer, rather elaborate

catalyst synthesis, and, especially in case of the complex 10,

their high moisture sensitivity.

(Figure 3). Interestingly, the corresponding niobium com-

plexes decomposed rapidly via carbene intermediates.^[57]

Although these tantalum species were inferior to niobium

complexes 11 and 12 with respect to catalytic activity and

molecular weight distribution of the resulting polymer, they

did polymerize olefins in a controlled fashion. Addition of

B(C₆F₅)₃to a solution of 13b at room temperature resulted in

formation of the ion-pair [Cp*Ta(h⁴-isoprene)Me]-

[MeB(C₆F₅)₃], as suggested by NMR experiments. The in situ

generated h⁴-butadiene congener 13a/B(C₆F₅)₃was found

to be active for the polymerization of ethylene

(1.38 kg_PEmol_Ta^À1h^À1), and the activity could be enhanced by

adding AlEt₃. When activated with MAO, dimethyl com-

plexes 13a c and the corresponding dichloro complexes

displayed comparable activities, indicating formation of

similar active species.

The next major advance towards living olefin polymer-

ization came in mid-1990s with the development of niobium

57]

diene based systems (Figure 3).^[55

Since these Group 5

systems are isoelectronic with Group 4 bis(Cp) complexes,

5. Nickel and Palladium Catalysts for Living Olefin

Polymerization

The search for catalyst systems for the living polymerization

of a-olefins gained additional momentum by the development

of diimine nickel and palladium catalysts in the mid-1990s.^[58]

The advantages of these late transition metal systems are

multifold. Late metal catalysts allow the copolymerization of

functional monomers because of the reduced oxophilicity of

the late metal center.^[59]Many of the late metal catalyst

precursors are easily accessible, and detailed mechanistic

studies have been undertaken since key species such as alkyl

olefin complexes can be cleanly generated.

Cl

Me

Nb

Ta

R²

R¹

11a R¹= R²= H

11b R¹= Me, R²= H

11c R¹= R²= Me

12

13a R¹= R²= H

13b R¹= Me, R²= H

13c R¹= R²= Me

Figure 3. Niobium- and tantalum catalyst precursors for living olefin

polymerization.

It is worth noting that more than a decade before the

discovery of late transition metal catalysts for living polymer-

ization of unactivated olefins, nickel allyl catalysts for living

Mashima and co-workers realized that these precursors

should be active in olefin polymerization. After an initial

paper disclosing the validity of this idea,^[55]living ethylene

polymerizations with the niobium diene systems 11a c/MAO

(MAO methylaluminoxane) and 12/MAO were reported in

1994.^[56]Polymerization below 08C occurred in a living fashion

to produce high molecular weight polyethylenes (M_nup to

40000 gmol^À1) with extremely narrow polydispersities (M_w/

M_nas low as 1.05). The catalyst precursors were synthesized in

low to moderate yields from [Cp*NbCl₄] and two equivalents

of the respective allyl Grignard reagent. The substitution

pattern of the diene moiety, which is essential for catalytic

activity, influenced catalytic performance. For example,

catalysts comprising 1,3-butadiene (11a) or 2,3-dimethyl-1,3-

butadiene (11c) displayed approximately equal polymeriza-

tion activities (38.7 versus 35.2 kg_PEmol_Nb^À1h^À1) at 208C),

while catalyst 11b containing the 2-methyl-1,3-butadiene

ligand was somewhat less active (19.2 kg_PEmol_Nb^À1h^À1). The

bis(diene) complex 12 was inactive for ethylene polymer-

polymerization of conjugated dienes already existed. In 1984,

3

¬

Teyssie and co-workers reported that bis[(h -allyl)(trifluor-

oacetato)nickel] (14) promoted the living polymerization of

1,3-butadiene to poly(1,4-butadiene) with predictable molec-

ular weight and narrow molecular weight distribution (M_w/

M_n 1.2 2.0) at or above room temperature.^[60]Even though

conversion of monomer was not complete (77% at best), the

linear increase in number-average molecular weight with

conversion was consistent with a living polymerization.

Deming and Novak, who utilized catalyst 14 in the polymer-

ization of tert-butyl-isocyanide to form the corresponding

helical polymers,^[61]took advantage of its living behavior in

the polymerization of butadiene to form poly(butadiene)/

poly(isocyanide) diblock copolymers (Scheme 9).^[62]Bimetal-

lic nickel initiators were then developed for the synthesis of

hydroxytelechelic poly(butadiene) and symmetric poly(iso-

cyanide)-block-poly(butadiene)-block-poly(isocyanide) tri-

2244

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

Living Polymerization of Alkenes

REVIEW

Diels Alder reaction, giving a poly(norbornene)-block-poly-

(acetylene) diblock copolymer.

CF₃CO₂

Ni

14

L_n

A major breakthrough in living olefin polymerization

occurred in 1995 with the discovery of a new family of late

transition metal catalysts by Brookhart and co-workers. Soon

after the initial report on the synthesis and polymerization

m

CF₃

RNC

R

O

Ni

O

Ni

activity of a-diiminepalladium and -nickel complexes,^[68]

a

O

N

subsequent paper appeared demonstrating the use of these

nickel systems for the living polymerization of a-olefins.^[69]

Catalyst precursor 17 (Figure 4), when activated with MAO in

CF₃CO₂

L_n

CF₃

14

Ni

n

m

Poly(isocyanide)-block-poly(butadiene)

Scheme 9. Nickel-catalyzed living copolymerization of butadiene and

isocyanides.

–

+ BAr₄

N

O

N

block copolymers.^[63]Allylnickel catalysts have also found

applications in the living polymerization of functionalized and

nonfunctionalized allenes.^[64]

Pd

N

Ni

MeO

Br Br

Risse and Mehler reported the living polymerization of

norbornene using [Pd(MeCN)₄][BF₄]₂(15). Poly(norbor-

nene)s with number-average molecular weights up to

30000 gmol^À1exhibited narrow molecular weight distribu-

tions at low monomer conversions; at higher conversions, the

distributions broadened.^[65]Using 15, Risse and Breunig

reported the polymerization of ester-functionalized norbor-

nenes. Depending on the nature of the ester substituent,

polymers with narrow molecular weight distributions were

obtained and block copolymers were synthesized.^[66]Novak

and Safir also demonstrated the potential of palladium

catalysts for living olefin polymerization in 1995. Hydro-

carbon poly(acetylene) block copolymers were synthesized

by living insertion polymerization using extremely robust, air-

and moisture-stable alkylpalladium(ii) complexes containing

s,p-alkyl ligands (16) (Scheme 10).^[67]Polymerization of

diethyl 7-oxabicyclo[2.2.1]hepta-2,5-diene-2,3-dicarboxylate,

which can be viewed as a protected acetylene monomer,

occurred quantitatively with 16 to yield a polymer with an

active palladium end group. Subsequent addition of norbor-

nene to this macroinitiator led to the formation of a block

copolymer. Upon heating, this material underwent a retro-

17

18

Ar

N

Ar =

tBu

N

Ni

Ar

Br Br

19

Figure 4. Nickel and palladium a-diimine catalysts and catalyst precursors

for living olefin polymerization.

toluene at room temperature, was shown to be very active for

the polymerization of a-olefins (turnover frequencies be-

tween 300 and 3000 h^À1), yielding high molecular weight

materials (M_nup to 190000 gmol^À1for polypropylene) with

relatively narrow polydispersities (1.4 < M_w/M_n< 1.8). At

lower temperatures (À108C) and low monomer concentra-

tions, polymers with very narrow molecular weight distribu-

tions were obtained because undesirable chain transfer

reactions were suppressed under these conditions. Accord-

ingly, complex 17/MAO in toluene under 1 atm of propylene

at À108C yielded high molecular weight polypropylene

(M_n 161000 gmol^À1, M_w/M_n 1.13; Scheme 11).

L_nPd

P

O

m

16

O

CO₂Et

EtO₂C

CO₂Et

MeO

Scheme 11. Living polymerization of a-olefins with a nickel diimine

catalyst system.

Cl

Pd

OMe

16

Since chain walking (consecutive b-hydride elimination

followed by olefin reinsertion with opposite regiochemistry) is

a distinguishing mechanistic feature of palladium and nickel

diimine catalysts,^{[68, 70, 71]}polymerization of ethylene results in

a highly branched polymer, while polymerization of higher a-

olefins leads to chain straightening, as indicated by a branch-

ing content that is lower than expected. The unique combi-

nation of living a-olefin polymerization behavior with the

formation of chain straightened poly(a-olefin)s provided the

EtO₂C

CO₂Et

1) HCl

m

O

n

L_nPd

2) 115 °C

m P

n

Poly(norbornene)-block-poly(acetylene)

Scheme 10. Synthesis of norbornene/acetylene block copolymers with a

palladium catalyst.

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

2245

G. W. Coates et al.

REVIEW

basis for a new strategy for the preparation of elastomeric

materials. While polymerization of propylene by 17 formed

an amorphous material (T_g À168C), polymerization of

octadecene led to a polymer with unbranched, crystalline

domains (T_m 568C). Based on these amorphous and crys-

talline polymer domains, several diblock and triblock poly-

mers were prepared. A poly(octadecene)-block-poly(octade-

cene-co-propylene)-block-poly(octadecene) triblock copoly-

mer, for example, consisting of semicrystalline poly(1-

octadecene) blocks and an amorphous propylene/octadecene

core, displayed elastomeric properties.^[69]Killian and Brook-

hart also exploited a similar strategy towards synthesis of

ethylene/1-hexene multiblock polymers (Scheme 12).^[72]Al-

though some chain transfer in ethylene homopolymerizations

independent of the polymerization conditions. However,

performing the polymerization at lower ethylene pressures

(1 atm) resulted in a somewhat broadened molecular weight

distribution (M_w/M_n 1.27). Lower ethylene pressures were

believed to retard the rate of initiation relative to the rate of

propagation since the palladium chelate complex 18 is favored

over the corresponding chelate-opened alkyl olefin catalyst

resting state. When polymerizations were performed at 278C,

catalyst decay led to a nonlinear increase of the molecular

weight over time and broadened molecular weight distribu-

tions of the resulting polyethylenes.

The full potential of the nickel and palladium a-diimine

systems has not yet been fully explored in the context of living

polymerization, and one can only imagine the exciting

possibilities for materials synthesis offered by these systems.

Possible future challenges involve control over stereochem-

istry in living a-olefin polymerization with C₂-symmetric

diimine catalysts, as well as the copolymerization of mono-

mers containing functional groups. The following two exam-

ples are given to illustrate this point. Non-living syndiospecific

polymerization of propylene by a-diimine nickel catalysts has

Scheme 12. Synthesis of polyethylene/poly(1-hexene) multiblock copoly-

mers with a nickel diimine catalyst.

77]

been reported.^[74Rieger and co-workers recently reported

the synthesis of extremely bulky C₂-symmetric nickel and

palladium a-diimine complexes and their application in the

formation of ultra high molecular weight polyethylenes at

ambient temperature. Linear polyethylene of narrow poly-

dispersity (M_w 4500000 gmol^À1; M_w/M_n 1.3) was obtained

with catalyst 19/MAO, providing an encouraging lead for

future investigations.^[78]Taking advantage of the living

behavior and functional group tolerance of catalyst 17, we

explored the copolymerization of 1-hexene with a function-

alized olefin capable of forming intermolecular hydrogen

bonds (Scheme 14).^[79]Ureidopyrimidone (UP) functional-

ized poly(1-hexene)s with low comonomer content (ca. 2%)

with nickel catalyst 17 occurs even at temperatures below

08C, a successful procedure for the synthesis of these unique

multiblock materials was developed. Catalyst 17 in a toluene/

1-hexene solution at À158C was combined with MMAO

(MMAO modified methylaluminoxane), and ethylene was

repeatedly added in short pulses. Since 1-hexene polymer-

ization was living under these conditions, the main reaction

product was expected to be a multiblock polymer despite

some chain transfer. Depending on the time intervals for both

ethylene and hexene polymerizations and the number of

cycles in which this procedure was repeated, the resulting

materials displayed excellent elastomeric properties (elonga-

tions of up to 1090%, tensile strength up to 860 psi).

While nickel diimine catalysts did not polymerize ethylene

in a living fashion, Gottfried and Brookhart recently reported

experimental conditions under which ethylene can be poly-

merized by palladium catalyst 18 to give amorphous polymers

with controlled molecular weights and narrow molecular

weight distributions (Scheme 13).^[73]Quenching the reaction

were synthesized in

a

living fashion (M_n 33000

104000 gmol^À1, M_w/M_n 1.2 1.4). These materials display

elastomeric properties, indicating the formation of hydrogen-

bond crosslinks between polymer chains. It should be pointed

out that the resulting polymer is certainly more complex than

shown in Scheme 14 because of chain walking.

À

y

x

mixture with triethylsilane, whereby the Pd C_(alkyl)bonds

y

+

x

were cleanly converted to saturated end groups, proved

crucial to obtaining monodisperse polyethylenes with this

catalytic system. When the polymerization was quenched with

acidified methanol, chain coupling frequently occurred and

the resulting polyethylene displayed a bimodal molecular

weight distribution. The polydispersity index of these poly-

ethylenes remained well under 1.1 up to molecular weights of

250000 gmol^À1. The resulting polymers were highly branched

(100 branches/1000 C), and the branching number was

1) Et₂AlCl

7

2) 17

H

O

N

O

N

nBu

N

O

N

H

N

O

H

Scheme 14. Synthesis of ureipyrimidone-functional poly(1-hexene) with a

nickel diimine catalyst.

6. Titanium, Zirconium, and Hafnium Catalysts for

Living Alkene Polymerization

Given the key advances in the control of polymer stereo-

chemistry and comonomer incorporation over the last two

decades,^[6]a significant amount of research in the quest for

Scheme 13. Living polymerization of ethylene with a palladium diimine

catalyst.

2246

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

Living Polymerization of Alkenes

REVIEW

living olefin polymerization catalysts has centered on com-

plexes based on the Group 4 metals. Metallocene catalyst

systems have been shown to exhibit living behavior at low

temperatures by suppressing undesirable b-hydride or b-alkyl

eliminations. Recently, non-metallocene systems based on

nitrogen and oxygen donor ligands have received consider-

able attention.^[80]In addition to providing pathways for living

olefin polymerization at ambient temperature, these systems

have also made possible considerable advances concerning

the control of stereochemistry in living olefin polymerization.

By combining these long-sought goals, these complexes

provide routes to the synthesis of polyolefin materials

inaccessible by conventional polymerization methods.

Me

Ti

Si

Ti

Me

N

21

22

Me

M

Zr

Me

Cl

23a M = Zr

23b M = Hf

24

Figure 5. Metallocene catalyst precursors for living olefin polymerization

at low temperatures.

6.1. Metallocene Complexes

Turner and Hlatky demonstrated the synthesis of a block

copolymer of propylene and ethylene using a cationic

hafnocene that was stable to chain transfer over the lifetime

of the polymerization reaction (Scheme 15), even though the

titanium centers operated in a living fashion. Polymers with

relatively narrow polydispersities (M_w/M_n 1.6) were ob-

tained at room temperature when polymerizations were

performed in light petroleum despite reduced catalyst pro-

ductivity in that solvent. Overall, the 21/B(C₆F₅)₃system

showed several characteristics of living behavior at room

temperature, such as linear increase of polymer yield with

both catalyst concentration and time. Addition of AlMe₃or

AliBu_3,even in substoichiometric amounts, dramatically

decreased catalytic activity and produced polymers with

bimodal molecular weight distributions. Polymerization of

1-hexene under similar conditions gave atactic poly(1-hex-

ene) with broader molecular weight distributions (M_nca.

10000 gmol^À1, M_w/M_nꢀ 2.4).^[83]

P

m

20

L_nHf

0°C

Atactic Polypropylene

+

NMe₂Ph

Me

Hf

–

B(C₆F₅)₄

P

m

L_nHf

n

20

PE-block-PP

M_w/M_n~ 1.7 – 1.9

Based on the notion that b-hydride elimination is negligible

in comparison with propagation in standard metallocene

MAO systems at temperatures below À408C,^[84]the research

groups of Shiono and Fukui investigated olefin polymer-

ization with metallocene or constrained geometry catalysts at

very low temperatures. To avoid chain transfer to aluminum,

Shiono and co-workers utilized borane activators in their

initial study.^[85]The titanium catalyst 22, when activated with

B(C₆F₅)₃at À508C, produced syndio-enriched polypropylene

([rrrr] 0.24) with narrow molecular weight distributions (M_n

up to 20000 gmol^À1, M_w/M_n 1.15 1.40), and polymerization

of 1-hexene under similar conditions also displayed living

behavior. Raising the temperature of the reaction to 08C

resulted in complete deactivation of the catalyst system. By

using rigorously dried MAO (free of AlMe₃impurities),

Shiono and co-workers then showed that chain transfer to

aluminum caused by residual AlMe₃could be efficiently

suppressed in this system. Accordingly, moderately syndio-

tactic polypropylene ([rrrr] 0.42) could be produced in a

Scheme 15. Synthesis of ethylene/propylene diblock copolymers with a

hafnocene catalyst.

obtained polymer molecular weight distributions were not

satisfying.^[81]Propylene was first added to a solution of

[Cp₂HfMe(PhNMe₂)][B(C₆F₅)₄] (20) at 08C. After consump-

tion of propylene, ethylene was added to the reactor. Based on

extraction studies, it was determined that 75% of the

polypropylene chains were incorporated into diblock copoly-

mers composed of atactic polypropylene and high-density

polyethylene segments. The copolymer could also be synthe-

sized by the reverse monomer addition, that is adding

ethylene first, but the molecular weight distribution was

higher in this case (1.89 versus 1.72) and the propylene content

was lower (30% versus 37%). Due to the relatively short

lifetimes of the living polymer chains, triblock copolymers

were not efficiently formed.

Bochmann and co-workers reported that [Cp*TiMe₃] (21;

Figure 5) activated with B(C₆F₅)₃was highly active for the

polymerization of propylene to form high molecular weight

atactic polypropylene (M_nup to 4000000 gmol^À1) that

displayed elastomeric properties.^[82]The presence of polymer

fractions with very narrow molecular weight distributions

(M_w/M_n 1.1) in this MAO-free polymerization system was

intriguing, and it was concluded that about half of the active

living fashion with 22/MAO even at 08C (M_n 10100 gmol^À1

M_w/M_n 1.35).^[86]

;

Fukui and co-workers demonstrated that the simple bis(Cp)

catalyst 23a activated with B(C₆F₅)₃was capable of the living

polymerization of propylene at À788C in the presence of

Al(nOct)₃as scavenger.^[87]The number of polymer chains was

found to be constant for the course of the reaction, indicating

that chain transfer did not occur at very low temperatures.

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

2247

G. W. Coates et al.

REVIEW

While the molecular weights of the resulting polymers were

rather low (M_w< 10000 gmol^À1), the molecular weight dis-

tributions were extremely narrow (M_w/M_n 1.06) and poly-

mer yields increased linearly with time. Carbon monoxide

also reacted quantitatively with the living polypropylene

chain ends, generating polypropylenes with terminal aldehyde

functionality.^[88]Similarly, the hafnium system 23b/B(C₆F₅)₃,

also displayed living behavior at À508C. Living polymer-

ization of 1-hexene was also observed with the C₂-symmetric

ansa-metallocene system 24/B(C₆F₅)₃at À788C, producing

highly isotactic monodisperse poly(1-hexene) (M_w/M_n 1.2

1.3). However, the catalytic activity for this system was low,

resulting in only very low molecular weight polymers (M_n<

5400 gmol^À1). Catalyst 24/B(C₆F₅)₃did not produce highly

isotactic polypropylene in a living fashion at À788C, presum-

ably due to b-alkyl elimination. Tritto and co-workers have

recently shown that 24 exhibits ™quasi-living∫ behavior for

ethylene/norbornene copolymerization when activated with

MAO.^[89]

Very recently, Fukui and Murata employed the mixed

metallocene catalyst system [Cp₂ZrMe₂]/B(C₆F₅)₃/[Cp*TiCl₃]

for the living polymerization of propylene at À508C.^[90]

Again, reaction of carbon monoxide with a living polypropy-

lene gave aldehyde-functionalized polypropylenes. Fukui and

Murata also employed the mixed metallocene catalyst system

[Ph₂C(Cp)(fluorenyl)ZrCl₂]/Al(nOct)₃/B(C₆F₅)₃/[Cp*TiCl₃]

for the living syndiospecific polymerization of propylene at

À508C.^[91]Finally, Brekner and co-workers described how

™quasi-living∫ metallocene/MAO systems could be used in the

synthesis of cycloolefin copolymers with narrow molecular

weight distribution when generated at temperatures between

08C and 408C (M_w/M_n 1.1 1.4).^[92]

R

N

Me

Ti

n

25/B(C₆F₅)₃

Me

R

CH₂Cl₂, 23 °C

Atactic Poly(1-hexene)

M_n> 120 000 g mol^–1

M_w/M_n= 1.07

25a R = iPr

25b R = Me

Scheme 16. Living polymerization of 1-hexene with McConville×s diami-

dotitanium catalysts.

bulk of the ligand by replacement of the isopropyl substitu-

ents with methyl groups (25b) did not substantially influence

catalyst performance. Polymerizations run in CH₂Cl₂led to a

dramatic increase in catalytic activity, yielding high molecular

weight atactic polymers (M_n> 120000 gmol^À1) with narrow

molecular weight distributions (M_w/M_n 1.07). This increase

in activity was reasoned to be the result of better separation of

the assumed catalyst ion-pair in the more polar solvent. On

the other hand, polymerization activity decreased in the

presence of toluene, presumably due to competitive binding

to the active site. Mechanistic studies, including isotopic

labeling and iodine quenching reactions, suggested that olefin

insertion occurred in a primary (1,2) fashion.^[94]

In a straightforward synthesis, the dimethyltitanium com-

plexes 25a, b were obtained by Grignard addition to the

corresponding dichlorides, which in turn were best synthe-

sized from the silylated diamines Me₃SiArN(CH₂)₃NArSiMe₃

and TiCl₄. Although attempts to isolate a cationic methyl-

titanium complex as a model for the presumed active species

failed, a few important mechanistic issues concerning catalyst

deactivation were addressed in these attempts. Addition of

B(C₆F₅)₃to a solution of the dimethyl complex 25a in pentane

led to the precipitation of a catalytically active borane adduct

26 (Scheme 17).^[95]Suspensions of this adduct slowly evolved

6.2. Diamido Complexes

Since the mid-1990s, several Group 4 catalysts with

ancillary amido ligands for living olefin polymerization have

been described. While these systems are of high academic

interest since they are amenable to detailed mechanistic

studies, their application in polymer synthesis has so far been

limited to atactic polymers of propylene and simple a-olefins.

Living polymerization of ethylene has not been reported with

these catalysts, and only one example of an olefin block

copolymer has been reported. In 1996, McConville and co-

workers introduced a new class of diamido catalysts for living

olefin polymerization. These tetrahedral dimethyltitanium

complexes 25a, b bearing ancillary propylene-bridged aryl-

substituted diamido ligands proved to be highly active

catalysts for polymerization of 1-hexene when activated by

MAO (Scheme 16). For example, treatment of precatalyst 25a

with MAO at room temperature gave atactic poly(1-hexene)

(M_n 47000 gmol^À1, M_w/M_n 1.73). Chain transfer to the

aluminum cocatalyst was implicated as the lone source of

termination since olefinic resonances were absent from the

NMR spectrum of the polymer. By activating the precatalyst

with B(C₆F₅)₃, chain transfer reactions were eliminated and

the polymerization of 1-hexene and higher a-olefins proceed-

ed in living fashion (Scheme 16).^[93]A decrease in the steric

B(C₆F₅)₃

Me

Me B(C₆F₅)₃

Me

CH₂B(C₆F₅)₂

C₆F₅

L_nTi

25a

26

27

Scheme 17. Deactivation pathway observed in McConville×s diamidotita-

nium complexes.

methane to form an inactive methylene-bridged derivative 27

that was characterized by single-crystal X-ray diffraction, thus

exemplifying

a possible catalyst deactivation pathway

(Scheme 17). Shiono and co-workers have recently reported

that 25a also produces polypropylene with low molecular

weight distribution using a modified MAO that is free of

trialkylaluminum residues.^[96]

Soon after McConville×s initial report, Schrock and co-

workers reported zirconium complexes with tridentate dia-

mido ligands based on the hypothesis that a propagating four-

coordinate cationic species would be more stable than a three-

coordinate species.^[97]Since 1997, Schrock and co-workers

have developed three different classes of compounds for use

in the living, aspecific polymerization of 1-hexene (Figure 6).

2248

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

Living Polymerization of Alkenes

REVIEW

R²

elimination as the primary mechanism of chain termination at

408C.^[98]The rate of b-hydride elimination was slow at 08C,

presumably due to steric crowding around the metal center

which prevents the ™backing up∫ of the bulky polymer chain

towards the ligand in an olefin hydride intermediate. There-

fore, at lower temperatures, the polymerization proceeded

without appreciable chain transfer. When the steric crowding

about the metal center was relaxed by replacing the tert-butyl

substituents with less bulky isopropyl or cyclohexyl groups,

the corresponding activated metal dialkyl complexes only

oligomerized 1-hexene.^[99]Interestingly, zirconium complexes

bearing the structurally-related sulfur donor ligand

[([D₆]tBu)N-o-C₆H₄)₂S]^2À(NSN) were not active for poly-

merization of 1-hexene.^[107]

A second class of compounds for living olefin polymer-

ization was introduced by Schrock and co-workers in 1999.

Replacing the oxygen in the ligand backbone with an amine

donor resulted in the more rigid tridentate complex 29a.^[102]

Activation of 29a with [Ph₃C][B(C₆F₅)₄] and subsequent

reaction with 1-hexene led to the formation of poly(1-hexene)

which displayed limited average molecular weight and broad-

er molecular weight distribution than the polymer obtained

with the NON system 28. Subsequent studies revealed that

this non-living behavior was caused by catalyst decomposition

R¹

tBu

Me

N

Me

tBu

1

M Me

R¹

R

N

Zr Me

O

R²

N

NMe

R¹

28a M = Ti

28b M = Zr

28c M = Hf

29a R¹= R²= Me

29b R¹= Cl, R²= H

R

N

Zr

R

Hf

R

N

30a R = Me

30b R = iBu

31 R =Et, nBu, iBu, nPr, iPr

Figure 6. Diamido catalyst precursors for living olefin polymerization.

In the first two classes of compounds 28^[97

and 29,^[102

101]

104]

À

by C H activation of one of the mesityl o-methyl substituents,

producing a dimeric decomposition product that was charac-

terized by X-ray crystallography (Scheme 18).^[103]A similar

the two amido donor atoms are connected to a central donor

(either oxygen or nitrogen), while in the third class 30/

31^{[105, 106]}all three donors are connected to a central carbon

atom. Even though living a-olefin polymerization has been

restricted almost exclusively to 1-hexene with these catalysts,

the resulting polymer molecular weight distributions are

among the narrowest known.

N

[Ph₃C][B(C₆F₅)₄]

Me

Zr

N

In 1997, Schrock and co-workers developed a very bulky

and robust tridentate diamido ligand with a central oxygen

donor atom.^[97]Group 4 catalyst precursors 28a c, incorpo-

rating the ligand [([D₆]tBu)N-o-C₆H₄)₂O]^2À(NON ligand)

were prepared by Grignard addition to the corresponding

dichloride complexes. The latter were best synthesized from

the ligand dianion and M(NMe₂)₂Cl₂and subsequent treat-

ment of the bis(dimethylamido) intermediates with Me₃SiCl.

The dimethyl complexes display trigonal-bipyramidal coordi-

nation geometry in the solid state with the oxygen donor and

one methyl substituent in the apical positions, as demonstrat-

ed by X-ray crystallography of 28a and 28b.^[100]Methide

abstraction from the dimethyl complex 28b by B(C₆F₅)₃

N

Zr Me

Zr

N

29a

C-H activated decomposition product

À

Scheme 18. C H activation as a catalyst deactivation pathway observed in

systems based on precursor 29a.

decomposition pathway of a cationic zirconium complex

À

through C H bond activation of an adjacent methyl group

had previously been uncovered by Horton and co-workers in a

structurally related triamido system.^[108]A simple change in

aryl substitution from 2,4,6-trimethylphenyl to 2,6-dichloro-

yielded

a reasonably stable ion pair, [(NON)ZrMe]-

À

[MeB(C₆F₅)₃] (32), which was characterized in the solid state

and by NMR spectroscopy. Both complex 32 as well as the

dimethylaniline adduct [(NON)ZrMe(PhNMe₂)][B(C₆F₅)₄]

(33) were active for the polymerization of ethylene at room

temperature (100 kg_PEmol_Zr^À1h^À1for 32, 800 kg_PEmol_Zr^À1h^À1

for 33). Catalyst 33 also polymerized 1-hexene to atactic

poly(1-hexene) (200 kg_PHmol_Zr^À1h^À1, M_n 45000 gmol^À1, M_w/

M_n 1.2). By lowering the temperature to 08C, chain transfer

reactions were suppressed, and the polymerization occurred

in a living fashion (M_w/M_n< 1.05 in the presence of more than

200 equivalents 1-hexene).

phenyl prevented C H activation of the ligand, and a living

1-hexene polymerization catalyst precursor 29b was ob-

tained.^[103]The polymer generated with catalyst 29b/

[Ph₃C][B(C₆F₅)₄] at 08C was of higher molecular weight than

that produced with the NON catalysts 28 (M_nup to

79000 gmol^À1), while the molecular weight distribution of

the polymer remained extremely narrow (M_w/M_n, 1.01 1.04).

In 2000, Schrock and co-workers reported a third class of

compounds for living olefin polymerization. Employing the

NNN framework in the form of diamidopyridine ligands

(MesNpy), in which the three donor atoms are connected to a

central carbon atom, the geometrically rigid zirconium

complexes 30a,b were prepared. An interesting initiator

effect was uncovered in 1-hexene polymerization by com-

Insights into the polymerization mechanism were gained in

a labeling study which established predominant 1,2 insertion

of the olefin (1-hexene or 1-nonene) as well as b-hydride

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

2249

G. W. Coates et al.

REVIEW

plexes 30a, b.^[105]Methide abstraction from the dimethyl

species 30a with [Ph₃C][B(C₆F₅)₄] was found to be a rather

complex process, leading to the formation of the putative

active species [(MesNpy)ZrMe][B(C₆F₅)₄] (34) in equilibrium

with 30a and a monocationic dimer [(MesNpy)₂Zr₂Me₃]-

species 36 decomposed in the absence of monomer by b-

hydride elimination, addition of 1-hexene led to a living

polymerization. In this case, heptenes were not observed,

suggesting that 1-hexene predominantly inserted into the

more sterically crowded Zr C_(isobutyl)bond in a primary (1,2)

fashion, yielding a 2-heptyl species that was relatively stable

towards b-hydride elimination. Polymers produced with 30b/

[Ph₃C][B(C₆F₅)₄] displayed molecular weights approximately

three times the theoretical value (based on Zr) and molecular

weight distributions were again extremely narrow (M_w/M_n

1.03).

Most recently, Schrock and co-workers reported the syn-

thesis, characterization, and polymerization activity of cat-

ionic hafnium complexes of the general formula [(MesN-

py)HfR][B(C₆F₅)₄] (31; R Et, nBu, iBu, nPr, iPr).^[106]These

compounds displayed a surprising stability towards b-hydride

elimination below 108C, and they promoted the living

polymerization of 1-hexene under these conditions. Insertion

[B(C₆F₅)₄]

(35)

(Scheme 19).

Solutions

of

30a/

[Ph₃C][B(C₆F₅)₄] were active for the polymerization of

Me

L_nZr

n

1-Hexene

(1,2)

Atactic Poly(1-hexene)

M_w/M_n= 1.02 – 1.08

1-Hexene

(1,2)

B(C₆F₅)₃

30a

[L_nZrMe][B(C₆F₅)₄]

[L_n2Zr₂Me₃][B(C₆F₅)₄]

L_nZrMe₂

35

30a

34

Dimeric Monocation

Putative Active Species

of 1-hexene into a Hf

C_(alkyl)bond was found to be about half

as fast as for the Zr analogue 30.

A final example of living olefin polymerization by a group

IV catalyst with an ancillary diamido ligand was reported by

Kim and coworkers.^[109]Activated versions of these bidentate,

aniline precursors 37a, b were active for the polymerization

of ethylene, propylene, and higher a-olefins (Scheme 20).

1-Hexene

(2,1)

L_nZr

Me

2-Heptenes

37b

L_nZr

B(C₆F₅)₃

x

0°C

n

B(C₆F₅)₃

1-Hexene

(1,2)

[L_nZriBu][B(C₆F₅)₄]

L_nZriBu₂

Atactic Poly(1-hexene)

1-octene

30b

36

Putative Active Species

iPr₃Si

Atactic Poly(1-hexene)

M_w/M_n= 1.02 – 1.08

N

R

Zr

R

L_nZr

Scheme 19. Initiator effect in diamidopyridine catalysts for living olefin

polymerization.

y

x

N

SiiPr₃

37a R = Cl

37b R = Me

1-hexene, but consumption of 1-hexene was not a first-order

process. The average molecular weight of the resulting

polymer was 10 times higher than expected assuming quanti-

tative initiation. Furthermore, NMR studies uncovered the

formation of 2-heptenes in the reaction, presumably originat-

ing from secondary (2,1) insertion of 1-hexene into the active

catalyst 34 to give a 3-heptyl species that is susceptible to b-

hydride elimination. The larger fraction of the 3-heptyl

species b-hydride eliminates and forms heptenes and a

catalytically inactive metal species, while the smaller fraction

propagates by primary (1,2) insertion. Despite these mecha-

nistic complications, the molecular weight distribution of the

resulting poly(1-hexene) was surprisingly narrow with this

system (M_w/M_n< 1.08).

When the initiating group was changed from methyl to

isobutyl (30b), activation with [Ph₃C][B(C₆F₅)₄] cleanly

generated a monomeric species [(MesNpy)ZriBu][B(C₆F₅)₄]

(36; Scheme 19). Since dimer formation like in the methyl

analogue 30a did not occur, all the zirconium centers were

available for polymerization and consequently no unreacted

dialkyl complex 30b was observed. Although the activated

Poly(1-octene)-block-poly(1-hexene)

M_n= 130 100 g mol^–1

M_w/M_n= 1.21

Scheme 20. 1-Hexene/1-octene block copolymer synthesis with a diamido

catalyst system.

For example, 37a/MAO showed an activity of

5300 kg_PEmol_Zr^À1h^À1under 1 atm of ethylene at room temper-

ature. However, the polymers produced with 37a/MAO at

room temperature were polydisperse, low molecular weight

materials (M_n< 9000 gmol^À1, M_w/M_n 2.09 2.35). Cooling

the polymerization to 08C reduced b-elimination reactions,

giving materials of higher molecular weight, but chain transfer

to the aluminum cocatalyst again gave polymers with broad

molecular weight distributions. However, employing B(C₆F₅)₃

as an activator for the dimethyl complex 37b at 08C gave high

molecular weight atactic polymers with relatively narrow

polydispersities. The catalyst system 37b/B(C₆F₅)₃at 08C

polymerized 1-hexene with a linear increase in molecular

2250

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

Living Polymerization of Alkenes

REVIEW

weight with yield to give atactic poly(1-hexene) with a narrow

molecular weight distribution (M_n 118300 gmol^À1, M_w/

M_n 1.23). The living nature of the polymerization was also

demonstrated by the sequential addition of two different a-

olefins to form a 1-hexene/1-octene block copolymer (M_n

108700 gmol^À1; M_w/M_n 1.21) (Scheme 20).

38b

[R₃NH][B(C₆F₅)₄]

n

–10°C, PhCl

Isotactic Poly(1-hexene)

[mmmm] > 0.95

M_n= 32 600 – 69 500 g mol^–1

M_w/M_n= 1.03 – 1.10

Scheme 21. Living isospecific polymerization of 1-hexene with zirconium

amidinate catalysts.

6.3. Zirconium Amidinates

Although the living vanadium catalysts developed by Doi

et al. gave syndio-enriched polypropylenes (Section 2.1), the

first catalyst to simultaneously achieve the highly sought goals

of livingness and stereoselectivity was reported by Jayaratne

and Sita in 2000.^[110]Using an amidinate zirconium catalyst,

the living isospecific polymerization of 1-hexene was achieved

with high stereoselectivity. By addition of carbodiimides to

with conversion. Furthermore, addition of a second portion of

1-hexene (180 equiv) to a living polymer chain end (M_n

20700 gmol^À1, M_w/M_n 1.03) resulted in formation of a higher

molecular weight polymer (M_n 35400 gmol^À1, M_w/M_n

1.13).

Although this catalytic system displayed excellent selectiv-

ity for production of isotactic poly(1-hexene), the mechanism

of stereocontrol was unclear. NMR spectroscopy revealed a

small resonance attributed to a mmmr-pentad as the only

visible stereoerror, consistent with an isotactic stereoblock

microstructure.^[110]The high selectivity demonstrated by this

system was remarkable considering the low barrier to

racemization of precursor 38b. Sita et al. suggested that the

barrier might be higher in the activated complex and the

structural defects in the polymer were a consequence of

amidinate ring flipping. To shed light on the origin of

stereocontrol in this unique system, crystallographic analyses

of 38b and 38b/[B(C₆F₅)₄] were performed.^[111]Indeed, the

more electrophilic metal center of the activated complex

À

the Zr C_(methyl)bond in [Cp*ZrMe₃], catalyst precursors 38a

d (Figure 7) were easily prepared. When R¹=R², the resulting

Me

R²

Me

R²

Zr

N

Zr

N

R¹

N

38a R¹= R²= cyclohexyl

38b R¹= tBu, R²= Et

38c R¹= tBu, R²= cyclohexyl

38d R¹= tBu, R²= 2,6-iPr₂C₆H₃

39a R¹= R²= cyclohexyl

39b R¹= R²= iPr

39c R¹= tBu, R²= cyclohexyl

Figure 7. Amidinate catalyst precursors for living polymerization of

1-hexene.

À

resulted in a shortening of the Zr N bonds, providing

evidence that suggests a higher barrier to racemization in

the active species. Also, coordination of diethyl ether to the

activated complex occurred on the more sterically encum-

bered face of the catalyst, suggesting that olefin coordination

might be preferred at that site. Crystallization in the absence

of Lewis bases at low temperatures (À108C) provided a dimer

with bridging methyl groups comprising a planar four-

membered ring, inside of which reside two bridging agostic

compounds are chiral but variable-temperature ¹H NMR

revealed low barriers to racemization. Upon activation with

[PhNMe₂H][B(C₆F₅)₄] in chlorobenzene at 08C, symmetric

complex 38a was active for the polymerization of 1-hexene,

giving a monodisperse polymer (M_n 11000 gmol^À1, M_w/

M_n 1.10). The narrow molecular weight distribution and

the lack of olefinic resonances characteristic of chain transfer

indicated the living nature of the polymerization. However,

microstructural analysis by ¹³C NMR revealed the lack of

stereochemical control in the polymerization.

À

hydrogen atoms (Scheme 22). The elongated Zr C_(methyl)bond

2+

1+

With this encouraging lead, the catalytic properties of C₁-

symmetric compounds 38b d were evaluated.^[110]Precursors

38c, d displayed poor activity toward 1-hexene, presumably

due to the sterically encumbered nature of the complex.

However, the activated complex 38b proved to be a superior

catalyst for the polymerization of 1-hexene at 258C

(Scheme 21). In addition to an increase in activity and

molecular weight, the polymer was also highly isotactic

([mmmm] > 0.95). The increase in activity was unfortunately

accompanied by a broadening of the molecular weight

distribution (M_w/M_n 1.50). However, lowering the polymer-

ization temperature to À108C led to higher molecular weight

polymers with extremely narrow polydispersities (M_w/M_n

1.03 1.10) and high stereoselectivities ([mmmm] > 0.95). The

living character of the polymerization was demonstrated by

the linear increase of number-average molecular weight (M_n)

Et

N

PhNMe₂

H₃

C

H₃

C

N

tBu

Zr

N

Zr

N

C

tBu

N

H₃

H₂

[PhNMe₂H]⁺

Et

Scheme 22. Potential deactivation pathway in amidinate catalyst systems.

length demonstrated the stabilizing effect of these a-agostic

interactions. It was suggested that these bridging a-agostic

interactions in dimeric cations might serve to lower the barrier

to migratory insertion, as has been proposed for mononuclear

complexes.^[14]On the other hand, crystallization of activated

complex 38b/[B(C₆F₅)₄] at 258C revealed the formation of a

m-CH₂, m-CH₃monocationic species, resulting from reduction

of one of the metal centers (Scheme 22). This species,

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

2251

G. W. Coates et al.

REVIEW

presumably the result of deprotonation of the dimer by

dimethylaniline introduced from the cocatalyst, revealed a

viable pathway for termination of active centers during the

polymerization with this catalytic system.

38b

[PhNMe₂H][B(C₆F₅)₄]

L_nZr

x

–10°C, PhCl

The enormous potential of this living and stereoselective

catalyst system lies in its potential for the synthesis of well-

defined olefin block copolymers with both crystalline and

amorphous domains. Due to the microphase separation of

such blocks, these materials have numerous applications as

compatibilizers and elastomers. In a demonstration of the

utility of this living catalyst system, Sita and co-workers

explored the cyclopolymerization of nonconjugated dienes to

give living polymers with high melting transitions.^[112]To this

end, 1,5-hexadiene was polymerized by the catalyst system

38a c/[PhNMe₂H][B(C₆F₅)₄] in chlorobenzene at À108C to

give poly(methylene-1,3-cyclopentane)s (PMCP) with narrow

iso-PH

L_nZr

z

y

x

y

x

iso-PH-block-PMCP-block-iso-PH

PMCP-block-iso-PH

molecular

weight

distributions

(M_w/M_n 1.03 1.09;

Scheme 24. Synthesis of poly(1-hexene)/PMCP diblock and triblock

copolymers with a zirconium amidinate catalyst.

Scheme 23). The lack of olefinic resonances in the polymers

demonstrated. A thin film of the material phase-separated

into a morphology composed of PMCP cylinders embedded in

a PH matrix.

38a-c

[PhNMe₂H][B(C₆F₅)₄]

n

–10°C, PhCl

In 2001, Sita et al. reported a modified class of amidinate

catalysts for living olefin polymerization (39a c; Fig-

ure 7).^[113]Employing the less bulky Cp ligand, the activated

complexes displayed enhanced activities relative to their Cp*

counterparts. For example, complex 39b/[PhNMe₂H]-

[B(C₆F₅)₄] polymerized 1-hexene (200 equiv) to 79% con-

version in only 2 min and proceeded to completion in under

10 min to give a monodisperse polymer (M_n 20800 gmol^À1,

M_w/M_n 1.03). A kinetic study of 1-hexene polymerization

with 39c revealed a linear relationship consistent with a living

polymerization, and GPC traces of all reported polymers

showed narrow distributions (M_w/M_n 1.03 1.09). While the

more exposed nature of the metal center was beneficial in

regard to catalytic activity, it had a detrimental effect on

stereoselectivity. Unlike complex 38b, which produced highly

isotactic poly(1-hexene)s, polymers from both the achiral

compounds 39a, b and the C₁-symmetric, chiral complex 39c

were atactic, demonstrating the importance of the bulky Cp*

ligand for stereodifferentiation.

Given the enhanced activities of these catalysts toward

polymerization of 1-hexene, Sita et al. reasoned that these

complexes might be capable of polymerization of more

sterically encumbered monomers such as vinylcyclohex-

ane.^[113]Indeed, activation of 39a, b at À108C in the presence

of vinylcyclohexane resulted in nearly complete conversion to

polymer (Scheme 25). Despite the achiral, C_s-symmetric

nature of catalyst precursors, the poly(vinylcyclohexane)

microstructures in each case were highly isotactic

Poly(methylene-1,3-cyclopentane)

64 – 82% trans-rings

M_n= 14 000 – 25 000 g mol^–1

M_w/M_n= 1.03 – 1.09

Scheme 23. Living cyclopolymerization of 1,5-hexadiene with zirconium

amidinate catalysts.

and the linear kinetic relationship (ln([M_o]/[M_t]) versus time)

were consistent with a living polymerization. Analysis of the

PMCP microstructures showed that the catalysts displayed

high selectivity for cyclization (>98%). Selectivity for

formation of trans rings was mediated by the steric bulk of

the catalyst precursor; increasing bulk of the amidinate

ligands in the series 38a !38b !38c was mirrored by an

increase in trans ring content and thus the degree of

crystallinity (T_m 98 1028C). Furthermore, the ligand struc-

ture had a huge influence on the tacticity of the PMCP. While

precursor 38a gave an atactic polymer, complex 38b dis-

played a high degree of stereocontrol, giving highly isotactic

PMCP.

Sita and co-workers exploited the ability of the amidinate

zirconium catalysts to produce living polymers with crystalline

domains through the synthesis of block copolymers from

1-hexene and 1,5-hexadiene (Scheme 24).^[112]Addition of

1-hexene to the activated complex 38b/[PhNMe₂H][B(C₆F₅)₄]

in chlorobenzene at À108C gave a living isotactic poly(1-

hexene) (M_n 12200 gmol^À1, M_w/M_n 1.03). Addition of 1,5-

hexadiene to this living chain end resulted in formation of a

higher molecular weight polymer (M_n 22800 gmol^À1, M_w/

M_n 1.05). The ¹³C NMR spectrum of the material was con-

sistent with both PH and PMCP and showed the highly

isotactic nature of the diblock material. This strategy was also

applied to the synthesis of a triblock copolymer (M_n

30900 gmol^À1, M_w/M_n 1.05) by addition of a second portion

of 1-hexene to a living diblock (Scheme 24). Through AFM

imaging, the block nature of this unique triblock material was

39a,b

[PhMe₂NH][B(C₆F₅)₄]

n

–10°C or 0°C, PhCl

Isotactic Poly(vinylcyclohexane)

Scheme 25. Isospecific living polymerization of vinylcyclohexane with a

zirconium amidinate catalyst.

2252

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

Living Polymerization of Alkenes

REVIEW

([mmmm] > 95%), indicative of a chain-end control mecha-

nism. The linear kinetic relationship and the narrow poly-

dispersities of the polymers (M_w/M_n 1.04 1.10) demonstrat-

ed the living nature of the polymerization. This living

polymerization was also applied to the synthesis of a new

triblock copolymer composed of poly(vinylcyclohexane) and

polyhexene domains through a sequential addition approach.

1500 gmol^À1, M_w/M_nꢀ 2). However, the polymerization of

1-hexene with 41b/B(C₆F₅)₃bearing the amine donor pro-

ceeded in a living fashion (M_n 14000 gmol^À1, M_n/M_n 1.18),

giving poly(1-hexene) with a linear increase in molecular

weight with time (Scheme 26). As anticipated, NMR spectro-

scopy revealed the absence of olefinic end groups and the

atactic microstructure of the polyhexene. In this case, the

presence of the extra donor arm served to suppress chain

transfer reactions.

6.4. Amine Bis(phenolate) Complexes

In 1999, a new family of Group 4 complexes bearing amine

bis(phenolate) [ONO]- and [ONNO]-type ligands was intro-

duced by Kol, Goldschmidt, and co-workers.^[114]Initial inves-

tigations of the C_s-symmetric zirconium compounds 40a and

41a (Figure 8) demonstrated the dramatic effect of the extra

41b

B(C₆F₅)₃

n

RT

Atactic Poly(1-hexene)

M_n= 14 000 g mol^–1

M_w/M_n= 1.18

42

B(C₆F₅)₃

RT

tBu

O

M

O

L_nTi

x

y

x

CH₂Ph

N

M

N

O

tBu

Atactic Poly(1-hexene)

M_n= 9 000 g mol^–1

M_w/M_n= 1.2

Poly(1-octene)-block-poly(1-hexene)

M_n= 11 600 g mol^–1

tBu

41a M = Zr

41b M = Ti

40a M = Zr

40b M = Ti

M_w/M_n= 1.2

Scheme 26. Living polymerization of 1-hexene and block copolymeriza-

tion of 1-hexene and 1-octene with amine bis(phenolate)-type catalysts.

R

tBu

R

O

With the successes of the extra donor arm of these

complexes, Kol, Goldschmidt, and co-workers recently re-

ported a new ligand in this family.^[117]Incorporating an extra

oxygen donor, this [ONOO]-type ligand gave the analogous

C_s-symmetric titanium species 42 (Figure 8). When activated

at room temperature by B(C₆F₅)₃, complex 42 proved

to be active for polymerization of 1-hexene (20

35 kg_PHmol_Ti^À1h^À1) (Scheme 26). The living behavior of this

system was evident by the narrow polydispersities (M_w/M_n

1.07 1.12) and the linear increase of molecular weight with

time. Amazingly, this linear relationship was still observed

after extremely long reaction times (31 h), giving a high

molecular weight poly(1-hexene) (M_n 445000 gmol^À1, M_w/

M_n 1.12). The polymerizations also had living character at

elevated temperatures, with reactions as high as 658C giving

relatively monodisperse polymers (M_n 22000 gmol^À1, M_w/

M_n 1.30). The livingness of this system was also demon-

strated through the synthesis of a block copolymer of

1-hexene and 1-octene (Scheme 26). Addition of 1-hexene

to 42/B(C₆F₅)₃in chlorobenzene at room temperature gave

poly(1-hexene) after 3.5 h (M_n 9000 gmol^À1, M_w/M_n 1.2).

After an additional 1.5 h, addition of 1-octene to this living

chain end resulted in formation of a higher molecular weight

polymer (M_n 11600 gmol^À1, M_w/M_n 1.2) with a ¹³C NMR

spectrum consistent with atactic poly(1-hexene-block-1-oc-

tene).

CH₂Ph

N

O

N

Ti

Zr

O

tBu

R

tBu

R

43a R = tBu

43b R = Me

42

Figure 8. Amine bis(phenolate) catalyst precursors for living olefin

polymerization.

donor arm on catalytic activity.^[115]When activated with

B(C₆F₅)₃at room temperature, complex 40a bearing the

tridentate [ONO] ligand displayed low activity toward

1-hexene (23 kg_PHmol_Zr^À1h^À1), producing only 1-hexene

oligomers (<20 units/chain). On the other hand, complex

41a/B(C₆F₅)₃based on the tetradentate [ONNO] ligand

was

extremely

active

for

the

polymerization

(15500 kg_PHmol_Zr^À1h^À1), giving high molecular weight

poly(1-hexene) with polydispersities consistent with a single-

site catalyst (M_n 170000 gmol^À1, M_w/M_n 2.2). The catalytic

performance of the titanium complexes 40b and 41b was also

investigated.^[116]In this case, the activated complex 40b

bearing the [ONO]-type ligand was slightly more active than

the [ONNO] complex 41b, but the polymerization quickly

attained a maximum value of molecular weight (M_n

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

2253

G. W. Coates et al.

REVIEW

With the promising lead of living behavior mediated by an

extra amine donor, Kol and co-workers targeted complexes

with a new type of [ONNO] ligand with similar functionality

but different connectivity, providing catalysts with C₂sym-

metry capable of stereocontrol in a-olefin polymerization.^[118]

The diamine bis(phenolate) ligands proved excellent for this

task, providing C₂-symmetric complexes (43a,b) analogous to

the ansa-metallocenes. Complex 43a, when activated at room

temperature by B(C₆F₅)₃, proved to be an active catalyst for

the polymerization of 1-hexene (18 kg_PHmol_Zr^À1h^À1)

(Scheme 27). The linear increase in molecular weight (M_n

up to 12000 gmol^À1) with conversion and the narrow molec-

ular weight distribution of the polymers (M_w/M_n 1.11 1.15)

R

tBu

N

O

Ti

O

Ar

N

Cl

N

Cl

N

Cl

N

Cl

Ti

N

Ar

tBu

R

45a Ar = C₆H₅

45b Ar = 2,6-F₂C₆H₃

45c Ar = 2,4,6-F₃C₆H₂

44a Ar = C₆H₅, R = tBu

44b Ar = F₅C₆, R = tBu

44c Ar = F₅C₆, R = H

Figure 9. Bis(phenoxyimine) and bis(indolide-imine) catalyst precursors

for living olefin polymerization.

mechanism.^{[19, 120, 121]}However, GPC analyses revealed the

lack of molecular weight control in this system (M_w/M_nꢀ 2).

In an effort to develop improved versions of these

phenoxyimine catalysts, Coates and coworkers discovered

that simple fluorination of the aniline moiety had a beneficial

effect on both catalytic activity and stereoselectivity.^[122]The

catalytic system 44b/MAO was an order of magnitude more

active than 44a and provided highly syndiotactic polypropy-

lene ([rrrr] 0.96; Scheme 28). The polymer exhibited a peak

43a

B(C₆F₅)₃

n

RT

Isotactic Poly(1-hexene)

[mmmm] > 0.95

M_n= 12 000 g mol^–1

M_w/M_n= 1.15

Scheme 27. Living isospecific polymerization of 1-hexene with a diamine

bis(phenolate) catalyst.

44b/c

demonstrated the living nature of the polymerization. Fur-

thermore, ¹³C NMR spectroscopy revealed that the polymer

possessed a highly isotactic microstructure (>95%). The

bulky tert-butyl substituents of 43a were crucial to both the

livingness and stereoselectivity of the catalytic system;

polymerization of 1-hexene with the methyl-substituted com-

plex 43b/B(C₆F₅)₃yielded atactic poly(1-hexene) with a

broader molecular weight distribution (M_n 23000 gmol^À1,

M_w/M_n 1.57).

MAO

0°C

Syndiotactic Polypropylene

[rrrr] up to 0.96

M_nup to 100 000 g mol^–1

M_w/M_n~ 1.1

Scheme 28. Highly syndiospecific and living polymerization of propylene

with bis(phenoxyimine) catalysts.

melting temperature of 1488C, among the highest values

reported for syndiotactic polypropylene. These polymers also

displayed narrow molecular weight distributions (M_w/M_nꢀ

1.1) and were free of olefinic end groups. Not only were b-

hydride and b-methyl eliminations suppressed, but the

polymerization also proceeded without transfer to the alumi-

num cocatalyst, a common source of chain transfer in catalytic

systems activated by MAO. The living nature of the reaction

was demonstrated by the linear relationship of molecular

weight and polymer yield. Although a slight deviation from

the initial linear relationship was observed at longer reaction

times, molecular weight distributions remained narrow (M_w/

M_n 1.11) to molecular weights approaching 100000 gmol^À1.

Furthermore, the molecular weight of the polymer was close

to the value calculated from the monomer/initiator ratio

obtained from the mass of the polymer produced, indicating

that each molecule of catalyst produces exactly one polymer

chain during the polymerization. Remarkably, polymers with

M_n> 300000 gmol^À1were synthesized with fairly narrow

molecular weight distributions (M_w/M_n 1.34), and the reac-

tions were also living at room temperature (M_w/M_n 1.13).

Fujita and co-workers concurrently reported that the fluorine-

containing catalyst system (44c/MAO) is also living for

propylene polymerization (M_nup to 108000 gmol^À1, M_w/

M_n 1.07 1.14).^[123]Polymers from this bis(phenoxyimine)

6.5. Bis(phenoxyimine) and Bis(indolideimine)

Complexes

Scientists at Mitsui have discovered that Group IV

complexes bearing phenoxyimine ligands are a remarkable

class of catalysts for olefin polymerization.^[119]Based on these

initial discoveries, several advances in the area of stereo-

selective as well as living olefin polymerization have been

reported. For example, Tian and Coates targeted Mitsui-type

complexes (44a) for isospecific propylene polymerization.

Due to their structural similarities with the ansa-metallocenes,

they reasoned that these C₂-symmetric complexes might be

suitable precursors for the isospecific polymerization of a-

olefins through a site-control mechanism. Surprisingly, com-

pound 44a/MAO (Figure 9) produced polypropylenes that

were substantially syndiotactic ([rrrr] 0.78).^[120]Microstruc-

tural analysis of the resulting polymer revealed that a chain-

end mechanism was responsible for the observed stereo-

control. Subsequent studies revealed that this extreme chain

end control was enhanced by an unusual secondary insertion

2254

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

Living Polymerization of Alkenes

REVIEW

catalyst, which incorporates a single tert-butyl substituent on

the phenoxy moiety, exhibited somewhat lower syndiotactic-

ity ([rrrr] 0.76) and peak melting temperature (T_m 1378C)

than polymers from 44b/MAO, although low molecular

weight oligomers (M_n 2000 g mol^À1) are nearly perfectly

syndiotactic.

Coates and co-workers demonstrated the utility of this

living and stereoselective catalyst system through the

synthesis of well-defined ethylene/propylene copolymers

with crystalline, syndiotactic polypropylene domains

(Scheme 29).^[122]Addition of 44b to a solution of propylene

molecular weight distributions (M_n 11000 41800 gmol^À1,

M_w/M_n 1.11 1.14). Increasing electrophilicity of the metal

center through fluorination of the ligand led to an increase in

activity, with 45c providing the most active precursor. The

living nature of the polymerization was demonstrated for 45c/

MAO by the linear increase in molecular weight with polymer

yield. Notably, polymerization with 45c/MAO at 508C also

produced polyethylene with a fairly narrow polydispersity

(M_w/M_n 1.24).

7. Outlook and Summary

44b

L_nTi

P

MAO

0°C

The last half-decade has witnessed a renaissance in living

olefin polymerization, building rapidly on the foundation set

solidly by Doi and co-workers over two decades ago. Today,

we have many efficient and selective catalysts available for

living olefin polymerization. Polyethylene, as well as atactic,

isotactic, and syndiotactic poly(a-olefins) can now be effi-

ciently synthesized in a living manner, allowing the creation of

unlimited new polymer architectures, such as block copoly-

mers and end-functional macromolecules. The ability to

synthesize such polymers will allow the detailed study of

structure property relationships and their influence on

mechanical and physical properties of this new class of

materials.

syn-PP

M

_n= 38 400 g mol^–1

M_w/M_n= 1.11

L_nTi

P

syn-PP-block-EPR

M_n= 145 000 g mol^–1

M_w/M_n= 1.12

T

_m= 131 °C, T_g= –45 °C

Perhaps the main challenge facing this new field is that

these sophisticated (and therefore usually expensive) metal

complexes only form one polymer chain during the polymer-

ization reaction, rendering the displacement of current

commodity polyolefins by these new materials economically

nonviable. Therefore, significant research must be conducted

to develop new strategies for the production of multiple

polymer chains per initiator. Such approaches will include the

development of agents that can cleave the growing polymer

chain from the metal but regenerate the catalyst in an active

form that can propagate a new polymer molecule. Alternately,

other methods for the synthesis of block copolymers from

non-living systems must also be pursued. These include: 1) the

development of oscillating catalysts that can change their

geometries during chain growth;^[127]2) the transfer of chains

between catalysts of differing stereospecificities;^[128]3) the

change of polymerization conditions that affect stereochem-

istry or comonomer incorporation on a time scale faster than

chain growth; and 4) the development of stopped-flow

techniques.^[129]

Scheme 29. Synthesis of ethylene/propylene diblock copolymers with a

bis(phenoxyimine) catalyst; EPR ethylene/propylene rubber.

and MAO at 08C resulted in the formation of a living

polypropylene after 2 h (M_n 38400 gmol^À1, M_w/M_n 1.11).

Addition to a slight overpressure of ethylene to this living

chain end resulted in rapid formation of a higher molecular

weight polymer (M_n 145100 gmol^À1, M_w/M_n 1.12). The

polymer, a syn-PP-block-EPR diblock copolymer, displayed

a T_mof 1318C, while the T_gof the EPR domain was À458C.

Fujita and co-workers have also reported the remarkable,

above room temperature living polymerization of ethylene^[124]

(M_nup to 400000 gmol^À1, M_w/M_n 1.05 1.13) and co-

polymerization of ethylene and propylene using the modi-

fied phenoxyimine system 44c.^[125]Monodisperse ethylene/

propylene copolymers with varying propylene content

(15 48 mol%) were prepared with 44c/MAO at 258C

(M_n> 80000 gmol^À1, M_w/M_n 1.07 1.13). Diblock and tri-

block copolymers composed of polyethylene, polypropylene,

and ethylene/propylene blocks have also been synthesized

with this catalyst system.^{[124, 125]}For example, a syn-PP-block-

EPR diblock similar to that described above by Coates and

co-workers was prepared from reaction of ethylene with a

living polypropylene chain (M_n 27000 gmol^À1, M_w/M_n

1.13). The resulting copolymer (M_n 161000 gmol^À1) dis-

played a broader molecular weight distribution (M_w/M_n

1.51).^[125]

Undoubtedly, the future will witness continued research at

this exciting interface of organometallic chemistry and

polymer science to the benefit of both fields. It is clear that

living systems will allow detailed mechanistic studies of

catalysts due to their discrete nature; in return these new

catalysts will allow the synthesis and study of a wide range of

new polymeric materials.

Fujita and co-workers have also developed a new class of

compounds bearing indolide-imine ligands for the living

polymerization of ethylene.^[126]When activated by MAO at

258C, complexes 45a c produced polyethylenes with narrow

G.W.C. gratefully acknowledges a Packard Foundation

Fellowship in Science and Engineering, a Dreyfus New Faculty

Award, an Alfred P. Sloan Research Fellowship, an Arnold and

Mabel Beckman Foundation Young Investigator Award, a NSF

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

2255

G. W. Coates et al.

REVIEW

[32] Y. Doi, N. Tokuhiro, K. Soga, Makromol. Chem. Macromol. Chem.

Phys. 1989, 190, 643 651.

[33] Y. Doi, Y. Watanabe, S. Ueki, K. Soga, Makromol. Chem. Rapid

Commun. 1983, 4, 533 537.

[34] Y. Doi, T. Koyama, K. Soga, Makromol. Chem. Macromol. Chem.

Phys. 1985, 186, 11 15.

CAREER Award, and a Camille Dreyfus Teacher-Scholar

Award. S.R. thanks the DAAD for a postdoctoral fellowship.

Received: November 7, 2001 [A499]

[35] Y. Doi, T. Keii, Adv. Polym. Sci. 1986, 73 4, 201 248.

[36] Y. Doi, G. Hizal, K. Soga, Makromol. Chem. 1987, 188, 1273 1279.

[37] Y. H. Doi, M. Nunomura, N. Ohgizawa, K. Soga, Makromol. Chem.

Rapid Commun. 1991, 12, 245 249.

[38] Y. Doi, M. Murata, K. Soga, Makromol. Chem. Rapid Commun.

1984, 5, 811 814.

[39] S. Ueki, H. Furuhashi, N. Murakami, M. Murata, Y. Doi, Stud. Surf.

Sci. Catal. 1995, 92, 359 362.

[40] Y. Doi, S. Ueki, T. Keii, Makromol. Chem. Rapid Commun. 1982, 3,

225 229.

[41] G. Jeske, H. Lauke, H. Mauermann, P. N. Swepston, H. Schumann,

T. J. Marks, J. Am. Chem. Soc. 1985, 107, 8091 8103.

[42] H. Yasuda, H. Yamamoto, K. Yokota, S. Miyake, A. Nakamura, J.

Am. Chem. Soc. 1992, 114, 4908 4910.

[43] H. Yasuda, M. Furo, H. Yamamoto, A. Nakamura, S. Miyake, N.

Kibino, Macromolecules 1992, 25, 5115 5116.

[44] H. Yasuda, E. Ihara, Macromol. Chem. Phys. 1995, 196, 2417 2441.

[45] H. Yasuda, E. Ihara, T. Hayakawa, T. Kakehi, J. Macromol. Sci. Pure

Appl. Chem. 1997, A34, 1929 1944.

[46] H. Yasuda, E. Ihara, Bull. Chem. Soc. Jpn. 1997, 70, 1745 1767.

[47] H. Yasuda, J. Polym. Sci. Part A 2001, 39, 1955 1959.

[48] G. Desurmont, Y. Li, H. Yasuda, T. Maruo, N. Kanehisa, Y. Kai,

Organometallics 2000, 19, 1811 1813.

[49] G. Desurmont, T. Tokimitsu, H. Yasuda, Macromolecules 2000, 33,

7679 7681.

[50] E. Ihara, M. Nodono, K. Katsura, Y. Adachi, H. Yasuda, M.

Yamagashira, H. Hashimoto, N. Kanehisa, Y. Kai, Organometallics

1998, 17, 3945 3956.

[51] G. Desurmont, M. Tanaka, Y. Li, H. Yasuda, T. Tokimitsu, S. Tone, A.

Yanagase, J. Polym. Sci. Part A 2000, 38, 4095 4109.

[52] E. Ihara, S. Yoshioka, M. Furo, K. Katsura, H. Yasuda, S. Mohri, N.

Kanehisa, Y. Kai, Organometallics 2001, 29, 1752 1761.

[53] M. Brookhart, A. F. Volpe, Jr., J. M. DeSimone, Polym. Prepr. Am.

Chem. Soc. Div. Polym. Chem. 1991, 32, 461 462.

[54] M. Brookhart, J. M. DeSimone, B. E. Grant, M. J. Tanner, Macro-

molecules 1995, 28, 5378 5380.

[1] M. Szwarc, J. Polym. Sci. Part A 1998, 36, IX XV.

[2] O. W. Webster, Science 1991, 251, 887 893.

[3] For leading references on block copolymers, see: a) F. S. Bates, G. H.

Fredrickson, Phys. Today 1999, 52, 32 38; b) I. W. Hamley, The

Physics of Block Copolymers, Oxford University Press, Oxford,

1998; c) F. S. Bates, Science 1991, 251, 898 905; d) M. Hillmyer,

Curr. Opin. Solid State Mater. Sci. 1999, 4, 559 564.

[4] K. Ziegler, E. Holzkamp, H. Breil, H. Martin, Angew. Chem. 1955,

67, 426.

[5] G. Natta, P. Pino, P. Corradini, F. Danusso, E. Mantica, G. Mazzanti,

G. Moraglio, J. Am. Chem. Soc. 1955, 77, 1708 1710.

[6] G. W. Coates, Chem. Rev. 2000, 100, 1223 1252.

[7] Anionic Polymerization: Principles and Practical Applications (Ed.:

H. L. Hsieh, R. P. Quirk), Marcel Dekker, New York, 1996.

[8] Cationic Polymerizations: Mechanisms, Synthesis, and Applications,

(Ed.: K. Matyjaszewski), Marcel Dekker, New York, 1996.

[9] C. J. Hawker, A. W. Bosman, E. Harth, Chem. Rev. 2001, 101, 3661

3688.

[10] M. Kamigaito, T. Ando, M. Sawamoto, Chem. Rev. 2001, 101, 3689

3746.

[11] K. Matyjaszewski, J. H. Xia, Chem. Rev. 2001, 101, 2921 2990.

[12] Molecular weights are most readily measured by using gel perme-

ation chromatography (GPC). This technique determines relative

number- and weight-average molecular weights unless the chromato-

graph is calibrated with monodisperse standards of the polymer

being examined. Since there are few commercially available poly-

olefin standards, especially of block copolymers most molecular

weight data in this review are measured relative to polystyrene

standards.

[13] R. P. Quirk, B. Lee, Polym. Int. 1992, 27, 359 367.

[14] For a leading reference on the mechanism of polymerization of

olefins by metallocene catalysts, see: R. H. Grubbs, G. W. Coates,

Acc. Chem. Res. 1996, 29, 85 93.

[15] L. Resconi, L. Cavallo, A. Fait, F. Piemontesi, Chem. Rev. 2000, 100,

1253 1345.

[16] L. Q. Deng, T. Ziegler, T. K. Woo, P. Margl, L. Y. Fan, Organo-

metallics 1998, 17, 3240 3253.

[17] E. Y. X. Chen, T. J. Marks, Chem. Rev. 2000, 100, 1391 1434.

[18] A. Zambelli, G. Natta, I. Pasquon, R. Signorini, J. Poly. Sci. Part C

1967, 2485 2499.

[55] K. Mashima, S. Fujikawa, H. Urata, E. Tanaka, A. Nakamura, J. Am.

Chem. Soc. 1993, 115, 10990 10991.

[56] K. Mashima, S. Fujikawa, H. Urata, E. Tanaka, A. Nakamura, J.

Chem. Soc. Chem. Commun. 1994, 1623 1624.

[57] K. Mashima, S. Fujikawa, Y. Tanaka, H. Urata, T. Oshiki, E. Tanaka,

A. Nakamura, Organometallics 1995, 14, 2633 2640.

[58] S. D. Ittel, L. K. Johnson, M. Brookhart, Chem. Rev. 2000, 100,

1169 1203.

[19] In a secondary insertion, the monomer inserts into the polymer chain

such that the more substituted carbon atom becomes bound to the

metal center and the polymer chain becomes attached to the

unsubstituted carbon of the alkene.

[59] L. S. Boffa, B. M. Novak, Chem. Rev. 2000, 100, 1479 1493.

[60] P. Hadjiandreou, M. Julemont, P. Teyssie, Macromolecules 1984, 17,

2455 2456.

[61] T. J. Deming, B. M. Novak, Macromolecules 1991, 24, 326 328.

[62] T. J. Deming, B. M. Novak, Macromolecules 1991, 24, 5478 5480.

[63] T. J. Deming, B. M. Novak, J. W. Ziller, J. Am. Chem. Soc. 1994, 116,

2366 2374.

[20] A. Zambelli, I. Pasquon, R. Signorini, G. Natta, Makromol. Chem.

Macromol. Chem. Phys. 1968, 112, 160 182.

[21] Y. Doi, M. Takada, T. Keii, Bull. Chem. Soc. Jpn. 1979, 52, 1802

1806.

[22] Y. Doi, S. Ueki, T. Keii, Macromolecules 1979, 12, 814 819.

[23] Y. Doi, S. Ueki, T. Keii, Makromol. Chem. Macromol. Chem. Phys.

1979, 180, 1359 1361.

[64] M. Suzuki, T. Takao, N. Sakamoto, I. Tomita, T. Endo, Polym. J. 1999,

31, 1021 1024.

[24] P. J. Flory, Principles of Polymer Chemistry, Cornell University Press,

Ithaca, 1953.

[65] C. Mehler, W. Risse, Macromolecules 1992, 25, 4226 4228.

[66] S. Breunig, W. Risse, Makromol. Chem. 1992, 193, 2915 2927.

[67] A. L. Safir, B. M. Novak, Macromolecules 1995, 28, 5396 5398.

[68] L. K. Johnson, C. M. Killian, M. Brookhart, J. Am. Chem. Soc. 1995,

117, 6414 6415.

[25] S. Ueki, Y. Doi, T. Keii, Makromol. Chem. Rapid Commun. 1981, 2,

403 406.

[26] Y. Doi, S. Suzuki, K. Soga, Makromol. Chem. Rapid Commun. 1985,

6, 639 642.

[27] Y. Doi, S. Suzuki, K. Soga, Macromolecules 1986, 19, 2896 2900.

[28] Y. Doi, S. Suzuki, G. Hizal, K. Soga, Transition Met. Catal. Polym.

1988, 182 194.

[69] C. M. Killian, D. J. Tempel, L. K. Johnson, M. Brookhart, J. Am.

Chem. Soc. 1996, 118, 11664 11665.

[70] V. M. Mˆhring, G. Fink, Angew. Chem. 1985, 97, 982 984; Angew.

Chem. Int. Ed. Engl. 1985, 24, 1001 1003.

[71] Z. Guan, P. M. Cotts, E. F. McCord, S. J. McLain, Science 1999, 283,

2059 2062.

[29] Y. Doi, S. Ueki, T. Keii in Coordination Polymerization (Eds.: C. C.

Price, E. J. Vandenberg), Plenum Press, New York, 1983.

[30] Y. Doi, N. Tokuhiro, S. Suzuki, K. Soga, Makromol. Chem. Rapid

Commun. 1987, 8, 285 290.

[72] C. M. Killian, Thesis, University of North Carolina, 1996.

[31] Y. Doi, S. Ueki, S. Tamura, S. Nagahara, T. Keii, Polymer 1982, 23,

258 262.

2256

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

Living Polymerization of Alkenes

REVIEW

[73] A. C. Gottfried, M. Brookhart, Macromolecules 2001, 34, 1140

1142.

[109] Y. M. Jeon, S. J. Park, J. Heo, K. Kim, Organometallics 1998, 17,

3161 3163.

[74] D. Pappalardo, M. Mazzeo, S. Antinucci, C. Pellecchia, Macro-

molecules 2000, 33, 9483 9487.

[75] C. Pellecchia, A. Zambelli, L. Oliva, D. Pappalardo, Macromolecules

1996, 29, 6990 6993.

[76] C. Pellecchia, A. Zambelli, Macromol. Rapid Commun. 1996, 17,

333 338.

[77] C. Pellecchia, A. Zambelli, M. Mazzeo, D. Pappalardo, J. Mol. Catal.

A Chem. 1998, 128, 229 237.

[110] K. C. Jayaratne, L. R. Sita, J. Am. Chem. Soc. 2000, 122, 958 959.

[111] R. J. Keaton, K. C. Jayaratne, J. C. Fettinger, L. R. Sita, J. Am. Chem.

Soc. 2000, 122, 12909 12910.

[112] K. C. Jayaratne, R. J. Keaton, D. A. Henningsen, L. R. Sita, J. Am.

Chem. Soc. 2000, 122, 10490 10491.

[113] R. J. Keaton, K. C. Jayaratne, D. A. Henningsen, L. A. Koterwas,

L. R. Sita, J. Am. Chem. Soc. 2001, 123, 6197 6198.

[114] E. Y. Tshuva, M. Versano, I. Goldberg, M. Kol, H. Weitman, Z.

Goldschmidt, Inorg. Chem. Commun. 1999, 2, 371 373.

[115] E. Y. Tshuva, I. Goldberg, M. Kol, H. Weitman, Z. Goldschmidt,

Chem. Commun. 2000, 379 380.

[78] M. Schmid, R. Eberhardt, M. Klinga, M. Leskela, B. Rieger,

Organometallics 2001, 20, 2321 2330.

[79] L. R. Rieth, R. F. Eaton, G. W. Coates, Angew. Chem. 2001, 113,

2211 2214; Angew. Chem. Int. Ed. 2001, 40, 2153 2156.

[80] G. J. P. Britovsek, V. C. Gibson, D. F. Wass, Angew. Chem. 1999, 111,

448 468; Angew. Chem. Int. Ed. Engl. 1999, 38, 428 447.

[81] H. W. Turner, G. G. Hlatky (Exxon), PCT Int. Appl. 9112285, 1991.

[82] J. Sassmannshausen, M. E. Bochmann, J. Rosch, D. Lilge, J. Organo-

met. Chem. 1997, 548, 23 28.

[83] M. C. Murray, M. C. Baird, J. Mol. Catal. A Chem. 1998, 128, 1 4.

[84] L. Resconi, F. Piemontesi, G. Franciscono, L. Abis, T. Fiorani, J. Am.

Chem. Soc. 1992, 114, 1025 1032.

[85] H. Hagihara, T. Shiono, T. Ikeda, Macromolecules 1998, 31, 3184

3188.

[86] T. Hasan, A. Ioku, K. Nishii, T. Shiono, T. Ikeda, Macromolecules

2001, 34, 3142 3145.

[87] Y. Fukui, M. Murata, K. Soga, Macromol. Rapid Commun. 1999, 20,

637 640.

[88] Y. Fukui, M. Murata, Macromol. Chem. Phys. 2001, 202, 1430 1434.

[89] J. C. Jansen, R. Mendichi, P. Locatelli, I. Tritto, Macromol. Rapid.

Commun. 2001, 22, 1394 1398.

[90] Y. Fukui, M. Murata, Macromol. Chem. Phys. 2001, 202, 1473 1477.

[91] Y. Fukui, M. Murata, Macromol. Chem. Phys. 2001, 202, 3205 3209.

[92] H. Cherdron, M.-J. Brekner, F. Osan, Angew. Makromol. Chem.

1994, 223, 121 133.

[93] J. D. Scollard, D. H. McConville, J. Am. Chem. Soc. 1996, 118,

10008 10009.

[94] J. D. Scollard, D. H. McConville, J. J. Vittal, N. C. Payne, J. Mol.

Catal. A Chem. 1998, 128, 201 214.

[95] J. D. Scollard, D. H. McConville, S. J. Rettig, Organometallics 1997,

16, 1810 1812.

[96] H. Hagimoto, T. Shiono, T. Ikeda, Macromol. Rapid. Commun. 2002,

23, 73 76.

[97] R. Baumann, W. M. Davis, R. R. Schrock, J. Am. Chem. Soc. 1997,

119, 3830 3831.

[98] R. Baumann, R. R. Schrock, J. Organomet. Chem. 1998, 557, 69 75.

[99] R. Baumann, R. Stumpf, W. M. Davis, L. C. Liang, R. R. Schrock, J.

Am. Chem. Soc. 1999, 121, 7822 7836.

[100] R. R. Schrock, R. Baumann, S. M. Reid, J. T. Goodman, R. Stumpf,

W. M. Davis, Organometallics 1999, 18, 3649 3670.

[101] J. T. Goodman, R. R. Schrock, Organometallics 2001, 20, 5205 5211.

[102] L. C. Liang, R. R. Schrock, W. M. Davis, D. H. McConville, J. Am.

Chem. Soc. 1999, 121, 5797 5798.

[116] E. Y. Tshuva, I. Goldberg, M. Kol, Z. Goldschmidt, Inorg. Chem.

Commun. 2000, 3, 611 614.

[117] a) E. Y. Tshuva, I. Goldberg, M. Kol, Z. Goldschmidt, Chem.

Commun. 2001, 2120 2121; b) E. Y. Tshuva, S. Groysman, I. Gold-

berg, M. Kol, Z. Goldschmidt, Organometallics 2002, 21, 662 670.

[118] E. Y. Tshuva, I. Goldberg, M. Kol, J. Am. Chem. Soc. 2000, 122,

10706 10707.

[119] a) S. Matsui, Y. Tohi, M. Mitani, J. Saito, H. Makio, H. Tanaka, M.

Nitabaru, T. Nakano, T. Fujita, Chem. Lett. 1999, 1065 1066; b) S.

Matsui, M. Mitani, J. Saito, Y. Tohi, H. Makio, H. Tanaka, T. Fujita,

Chem. Lett. 1999, 1263 1263; c) S. Matsui, M. Mitani, J. Saito, N.

Matsukawa, H. Tanaka, T. Nakano, T. Fujita, Chem. Lett. 2000, 554

555; d) T. Fujita, Y. Tohi, M. Mitani, S. Matsui, J. Saito, M. Nitabaru,

K. Sugi, H. Makio, T. Tsutsui, Eur. Pat. Appl. 874005 (Chem. Abstr.

1998, 129, 331166); for a patent regarding Mitsui×s research in living

alkene polymerization, see: e) M. Mitani, Y. Yoshida, J. Mohri, K.

Tsuru, S. Ishii, S. Kojoh, T. Matsugi, J. Saito, N. Matsukawa, S.

Matsui, T. Nakano, H. Tanaka, N. Kashiwa, T. Fujita, PCT Int. Appl.

WO2001055231 (Chem. Abstr. 2001, 135, 137852).

[120] J. Tian, G. W. Coates, Angew. Chem. 2000, 112, 3772 3775; Angew.

Chem. Int. Ed. 2000, 39, 3626 3629.

[121] a) J. Saito, M. Mitani, M. Onda, J. Mohri, S. Ishii, Y. Yoshida, T.

Nakano, H. Tanaka, T. Matsugi, S. Kojoh, N. Kashiwa, T. Fujita,

Macromol. Rapid Commun. 2001, 22, 1072 1075; b) M. Lamberti,

D. Pappalardo, A. Zambelli, C. Pellecchia, Macromolecules 2002, 35,

658 663; c) P. D. Hustad, J. Tian, G. W. Coates, J. Am. Chem. Soc.

2002, 124, 3614 3621.

[122] J. Tian, P. D. Hustad, G. W. Coates, J. Am. Chem. Soc. 2001, 123,

5134 5135.

[123] J. Saito, M. Mitani, J. Mohri, S. Ishii, Y. Yoshida, T. Matsugi, S. Kojoh,

N. Kashiwa, T. Fujita, Chem. Lett. 2001, 576 577.

[124] a) J. Saito, M. Mitani, J. Mohri, Y. Yoshida, S. Matsui, S. Ishii, S.

Kojoh, N. Kashiwa, T. Fujita, Angew. Chem., Int. Ed. Engl. 2001, 40,

2918 2920; Angew. Chem. 2001, 113, 3002 3004; b) M. Mitani, J.

Mohri, Y. Yoshida, J. Saito, S. Ishii, K. Tsuru, S. Matsui, R. Furuyama,

T. Nakano, H. Tanaka, S. Kojoh, T. Matsugi, N. Kashiwa, T. Fujita, J.

Am. Chem. Soc. 2002, 124, 3327 3336.

[125] S. Kojoh, T. Matsugi, J. Saito, M. Mitani, T. Fujita, N. Kashiwa,Chem.

Lett. 2001, 822 823.

[126] T. Matsugi, S. Matsui, S. Kojoh, Y. Takagi, Y. Inoue, T. Fujita, N.

Kashiwa, Chem. Lett. 2001, 566 567.

[103] R. R. Schrock, P. J. Bonitatebus, Y. Schrodi, Organometallics 2001,

20, 1056 1058.

[104] Y. Schrodi, R. R. Schrock, P. J. Bonitatebus, Organometallics 2001,

20, 3560 3573.

[105] P. Mehrkhodavandi, P. J. Bonitatebus, R. R. Schrock, J. Am. Chem.

Soc. 2000, 122, 7841 7842.

[106] P. Mehrkhodavandi, R. R. Schrock, J. Am. Chem. Soc. 2001, 123,

10746 10747.

[107] D. D. Graf, R. R. Schrock, W. M. Davis, R. Stumpf, Organometallics

1999, 18, 843 852.

[108] A. D. Horton, J. de With, A. J. van der Linden, H. van de Weg,

Organometallics 1996, 15, 2672 2674.

[127] G. W. Coates, R. M. Waymouth, Science 1995, 267, 217 219.

[128] For recent examples of stereoblock polymer synthesis using polymer

exchange pathways, see: a) J. C. W. Chien, Y. Iwamoto, M. D.

Rausch, W. Wedler, H. H. Winter, Macromolecules 1997, 30, 3447

3458; b) J. C. W. Chien, Y. Iwamoto, M. D. Rausch, J. Poly. Sci. Part

A 1999, 37, 2439 2445; c) S. Lieber, H. H. Brintzinger, Macro-

molecules 2000, 33, 9192 9199; d) K. C. Jayaratne, L. R. Sita, J. Am.

Chem. Soc. 2001, 123, 10754 10755.

[129] B. Liu, H. Matsuoka, M. Terano, Macromol. Rapid Commun. 2001,

22, 1 24.

Angew. Chem. Int. Ed. 2002, 41, 2236 2257

2257

Article Doi

Chiral epoxides by desymmetrizing deprotonation of meso-epoxides

DOI: 10.1002/1521-3773(20020703)41:13<2236::AID-ANIE2236>3.0.CO;2-3

Source and publish data:

Authors:

Article abstract of DOI:10.1002/1521-3773(20020703)41:13<2236::AID-ANIE2236>3.0.CO;2-3

Full text of DOI:10.1002/1521-3773(20020703)41:13<2236::AID-ANIE2236>3.0.CO;2-3

Products guided by the article

R&D Labs maybe for 57338-07-5

Relevant to this article

Hot Product