Mechanism of isotactic polypropylene formation with C 1-symmetric metallocene catalysts

3576

Organometallics 2006, 25, 3576-3592

Mechanism of Isotactic Polypropylene Formation with C₁-Symmetric

Metallocene Catalysts

Stephen A. Miller^†and John E. Bercaw*

Arnold and Mabel Beckman Laboratories of Chemical Synthesis, California Institute of Technology,

Pasadena, California 91125

ReceiVed September 29, 2005

The mechanism for isotactic polypropylene formation by the C₁-symmetric catalyst system [Me₂C(3-

tert-butyl-C₅H₃)(C₁₃H₈)]ZrCl₂/MAO (MAO ) methylaluminoxane, C₁₃H₈) fluorenylidene) has been

examined. Evidence supports an alternating mechanism, where both sites of the metallocene wedge are

utilized for monomer insertion, rather than the previously proposed site epimerization (inversion at Zr)

following each monomer insertion. As the polymerization temperature increases (0 to 60 °C) with lower

concentrations of propylene, the site epimerization mechanism begins to compete, as evidenced by an

increase in isotacticity. The alternating mechanism also accounts for polypropylene microstructures obtained

with Me₂C(3-R-C₅H₃)(Oct)ZrCl₂/MAO, where Oct ) octamethyloctahydrodibenzofluorenylidene and R

) methyl, cyclohexyl, diphenylmethyl, and with Me₂C(3-tert-butyl-4-Me-C₅H₂)(Oct)ZrCl₂/MAO. For

an Oct-containing catalyst system with R ) 2-methyl-2-adamantyl, unprecedentedly high (for a fluorenyl-

based metallocene catalyst) isotacticity ([mmmm] > 99%) is obtained; the polymer prepared at 0 °C has

T_m) 167 °C and M_w) 370 000.

Scheme 1

Introduction

Shortly following the report by Ewen, Razavi, et al.¹

demonstrating the MAO-cocatalyzed (MAO ) methylalumi-

noxane) formation of syndiotactic polypropylene with C_s-

symmetric fluorenyl-containing metallocenes of the type Me₂C-

(C₅H₄)(C₁₃H₈)MCl₂(e.g., M ) Zr, 1), several authors prepared

cyclopentadienyl-substituted variants of the parent metallocene.

Incorporation of a substituent at the 3 position of the cyclo-

pentadienyl ring effects desymmetrization of the metallocene

to C₁symmetry. As a consequence, the obtained polymers were

no longer syndiotactic, but displayed alternative tacticities

depending on the nature of the substituent (Scheme 1).

Ewen,²Spaleck,³and Razavi⁴have each reported on C₁-

symmetric fluorenyl-containing metallocenes and their behavior

in propylene polymerizations. The parent C_s-symmetric zir-

conocene 1 (Scheme 1) produces syndiotactic polypropylene

via a Cosse´e-Arlman⁵type chain migratory insertion mecha-

nism in which monomer insertions occur sequentially at

alternating sites of the metallocene.⁶Similarly, the hemiisotactic

polypropylene produced by 2 is best explained by the same

Cosse´e-Arlman type mechanism in which one site is enantio-

selective and the other site is aselective.⁷In contrast, 3 produces

isotactic polypropylene.⁸The mechanism of isotactic polypro-

pylene formation with this C₁-symmetric metallocene is still a

topic of debate.

* To whom correspondence should be addressed. E-mail:

bercaw@caltech.edu.

^†Current address: Department of Chemistry, Texas A&M University,

College Station, TX 77843-3255.

(1) Ewen, J. A.; Jones, R. L.; Razavi, A.; Ferrara, J. D. J. Am. Chem.

Soc. 1988, 110, 6255-6256.

(2) (a) Ewen, J. A.; Elder, M. J.; Jones, R. L.; Haspeslagh, L.; Atwood,

J. L.; Bott, S. G.; Robinson, K. Makromol. Chem., Macromol. Symp. 1991,

48/49, 253-295. (b) Ewen, J. A. Eur. Pat. Appl. EP 423101, 1989. (c)

Ewen, J. A. Macromol. Symp. 1995, 89, 181-196. (d) Ewen, J. A.; Elder,

M. J. In Ziegler Catalysts, Recent Scientific InnoVations and Technological

ImproVements; Fink, G., Mu¨lhaupt, R., Brintzinger, H.-H., Eds.; Springer:

Berlin, 1995; pp 99-109.

(4) (a) Razavi, A.; Atwood, J. L. J. Organomet. Chem. 1996, 520, 115-

120. (b) Razavi, A.; Vereecke, D.; Peters, L.; Den Dauw, K.; Nafpliotis,

L.; Atwood, J. L. In Ziegler Catalysts, Recent Scientific InnoVations and

Technological ImproVements; Fink, G., Mu¨lhaupt, R., Brintzinger, H.-H.,

Eds.; Springer: Berlin, 1995; pp 111-147. (c) Razavi, A.; Atwood, J. L.

J. Organomet. Chem. 1995, 497, 105-111. (d) Razavi. A.; Peters, L.;

Nafpliotis, L.; Vereecke, D.; Dendauw, K.; Atwood, J. L.; Thewald, U.

Macromol. Symp. 1995, 89, 345-367.

(5) (a) Cosse´e, P. J. Catal. 1964, 3, 80-88. (b) Arlman, E. J. J. Catal.

1964, 3, 89-98. (c) Arlman, E. J.; Cosse´e, P. J. Catal. 1964, 3, 99-104.

(6) For a detailed mechanistic analysis employing C_s-symmetric metal-

locenes, see: Veghini, D.; Henling, L.; Burkhardt, T.; Bercaw, J. E. J. Am.

Chem. Soc. 1999, 121, 564-573.

(3) (a) Dolle, V.; Rohrmann, J.; Winter, A.; Antberg, M.; Klein, R. Eur.

Pat. Appl. EP 399347, 1991. (b) Antberg, M.; Dolle, V.; Klein, R.;

Rohrmann, J.; Spaleck, W.; Winter, A. In Studies in Surface Science and

Catalysis (Catalytic Olefin Polymerization, Volume 56); Keii, T., Soga, K.,

Eds.; Kodansha: Tokyo, 1990; pp 501-515. (c) Spaleck, W.; Antberg, M.;

Aulbach, M.; Bachmann, B.; Dolle, V.; Haftka, S.; Ku¨ber, F.; Rohrmann,

J.; Winter, A. In Ziegler Catalysts, Recent Scientific InnoVations and

Technological ImproVements; Fink, G., Mu¨lhaupt, R., Brintzinger, H.-H.,

Eds.; Springer: Berlin, 1995; pp 83-97. (d) Spaleck, W.; Aulbach, M.;

Bachmann, B.; Ku¨ber, F.; Winter, A. Macromol. Symp. 1995, 89, 237-

247.

Publication on Web 06/16/2006

Mechanism of Isotactic Polypropylene Formation

Organometallics, Vol. 25, No. 15, 2006 3577

Scheme 2

model. One of the primary explanations offered in support of

the site epimerization mechanism is that the tert-butyl side of

the metallocene is too sterically crowded to accommodate the

growing polymer chain. Calculations by Morokuma¹²suggest

that the bulky substituent forbids the growing chain from

residing nearby, necessitating a site epimerization following

every insertion. Calculations by Fink¹³and Corradini,¹⁴however,

are more forgiving and allow the insertion with the polymer

chain proximal to the tert-butyl group.¹⁵

Also shown in Scheme 2 is a second limiting mechanism,

the alternating (blue pathway) mechanism. Following monomer

insertion at the more stereoselective site, the second site becomes

available for monomer coordination. In the transition state for

insertion at the less stereoselective site, the growing polymer

chain is directed competitively by both the tert-butyl group

(down) and the benzo substituent (up). In order for the resulting

polymer to be isotactic, the tert-butyl group must prevail, and

the growing polymer chain is preferentially directed toward the

benzo substituent. Insertion ensues with a trans arrangement

between the polymer chain and the methyl group of the inserting

monomer; this regenerates the original coordination site. In

contrast to the site epimerization mechanism, the alternating

mechanism employs two sites for monomer insertion, using

olefin coordination at both sides of the metallocene wedge, as

is currently accepted for the syndioselective (1) and hemiiso-

selective (2) relatives of isoselective 3.

Proposed Mechanisms for Isotactic Polypropylene Forma-

tion with C₁-Symmetric Metallocene Catalysts. There are two

limiting mechanisms possible for the formation of isotactic

polypropylene with C₁-symmetric metallocene catalysts.⁹These

are the site epimerization mechanism and the alternating

mechanism (Scheme 2).

The majority of published reports for isotactic polypropylene

formation with metallocenes based on 3 invoke the site

epimerization¹⁰mechanism (red pathway) to account for the

observed isoselectivity.^{2c,d,3c,4a,b,11}As shown in Scheme 2 for

one enantiomer of racemic 3, the growing polymer chain is on

the left side of the metallocene wedge directed up and away

from the benzo substituent of the fluorenyl ligand in the

transition state for monomer insertion at the more stereoselective

site. The methyl group of the incoming monomer is directed

down in a trans arrangement from the growing polymer chain.

Following migratory insertion, the growing polymer chain is

located on the more crowded, right side of the metallocene

wedge and moves away from the bulky tert-butyl substituent

in a unimolecular site epimerization process that inverts

stereochemistry at the metal center. These two consecutive

processes regenerate the original coordination site for monomer

insertion. Hence, only one transition structure is employed, one

that uses olefin coordination from one side of the metallocene

for monomer insertion, according to a strict site epimerization

Despite widespread support for the site epimerization mech-

anism, there has been little convincing evidence presented in

its defense. We sought to experimentally test which of the two

mechanisms, or combination thereof, most accurately describes

(10) The term “site epimerization mechanism” is used in preference to

other terms found in the literature. The term “epimerization” (Eliel, E. L.;

Wilen, S. H. Stereochemistry of Organic Compounds; Wiley-Interscience:

New York, 1994; p 1198) refers to the inversion of a single stereocenter

when two or more are present and can be correctly applied to metallocene

polymerizations when the stereochemistry of the metal is inverted while

the stereocenters present on the polymer chain remain unchanged. (a)

“Consecutive addition”: Farina, M.; Terragni, A. Makromol. Chem., Rapid

Commun. 1993, 14, 791-798. (b) “Skipped insertion”: ref 2a. (c) “Skipped-

out insertions” and “chain migratory catalyst isomerization”: ref 2c. (d)

“Back-skip mechanism”: ref 2d. (e) “Retention mechanism”: Di Silvestro,

G.; Sozzani, P.; Terragni, A. Macromol. Chem. Phys. 1996, 197, 3209-

3228. (f) “Isomerization without monomer insertion”: Fink, G.; Herfert,

N. Preprints of the International Symposium on AdVances in Olefin,

Cycloolefin, and Diolefin Polymerization; Lyon, 1992; p 15. (g) “Site-to-

site chain migration”: ref 3c. (h) “Side-to-side swing” and “back swing”:

ref 4a. (i) “Chain stationary insertion mechanism”: ref 4b. (j) “Hindered

chain migratory insertion”: Razavi, A.; Bellia, V.; De Brauwer, Y.;

Hortmann, K.; Lambrecht, M.; Miserque, O.; Peters, L.; Van Belle, S. In

Metalorganic Catalysts for Synthesis and Polymerization, Recent Results

by Ziegler-Natta and Metallocene InVestigations; Kaminsky, W., Ed.;

Springer: Berlin, 1999; pp 236-247.

(11) Hefert, N.; Fink, G. Makromol. Chem. Macromol. Symp. 1993, 66,

157-178.

(12) Yoshida, T.; Koga, N.; Morokuma, K. Organometallics 1996, 15,

766-777.

(13) (a) van der Leek, Y.; Angermund, K.; Reffke, M.; Kleinschmidt,

R.; Goretzki, R.; Fink, G. Chem. Eur. J. 1997, 3, 585-591. (b) Angermund,

K.; Fink, G.; Jensen, V. R.; Kleinschmidt, R. Macromol. Rapid Commun.

2000, 21, 91-97.

(14) Corradini, P.; Cavallo, L.; Guerra, G. In Metallocene-Based Poly-

olefins; Scheirs, J., Kaminsky, W., Eds.; John Wiley & Sons: New York,

2000; pp 3-36.

(7) For reviews of homogeneous, metallocene-mediated Ziegler-Natta

olefin polymerizations, see: (a) Brintzinger, H.-H.; Fischer, D.; Mu¨lhaupt,

R.; Rieger, B.; Waymouth, R. M. Angew. Chem., Int. Ed. Engl. 1995, 34,

1143-1170, and references therein. (b) Janiak, C. In Metallocenes:

Synthesis, ReactiVity, Applications; Togni, A.; Halterman, R. L., Eds.; Wiley-

VCH: Weinheim, 1998; pp 547-623. (c) Kaminsky, W., Ed. Metalorganic

Catalysts for Synthesis and Polymerization, Recent Results by Ziegler-

Natta and Metallocene InVestigations; Springer: Berlin, 1999. (d) Marks,

T. J.; Stevens, J. C., Eds. Top. Catal. 1999, 7, 1-208. (e) Alt, H. G.; Ko¨ppl,

A. Chem. ReV. 2000, 100, 1205-1221. (f) Coates, G. W. Chem. ReV. 2000,

100, 1223-1252. (g) Resconi, L.; Cavallo, L.; Fait, A.; Piemontesi, F. Chem.

ReV. 2000, 100, 1253-1345. (h) Angermund, K.; Fink, G.; Jensen, V. R.;

Kleinschmidt, R. Chem. ReV. 2000, 100, 1457-1470. (i) Scheirs, J.,

Kaminsky, W., Eds. Metallocene-Based Polyolefins: Preparation, Proper-

ties and Technology; John Wiley & Sons: New York, 2000. (j) Kaminsky,

W.; Scholz, V.; Werner, R. Macromol. Symp. 2000, 159, 9-17. (k)

Bochmann, M. J. Organomet. Chem. 2004, 689, 3982-3998.

(8) (a) Ewen, J. A.; Elder, M. J. Eur. Pat. Appl. EP 537130, 1993. (b)

Kaminsky, W.; Rabe, O.; Schauwienold, A.-M.; Schupfner, G. U.; Hanss,

J.; Kopf, J. J. Organomet. Chem. 1995, 497, 181-193.

(9) A comparison of mechanisms for isoselective propylene polymeri-

zation using C₁-symmetric metallocene precatalysts is given by: Busico,

V.; Cipullo, R.; Talarico, G.; Segre, A. L.; Chadwick, J. C. Macromolecules

1997, 30, 4786-4790.

(15) Straightforward examples of propylene insertion into sterically

encumbered sites do exist for C₂-symmetric metallocenes: (a) Resconi, L.;

Piemontesi, F.; Camurati, I.; Sudmeijer, O.; Nifant’ev, I. E.; Ivchenko, P.

V.; Kuz’mina, L. G. J. Am. Chem. Soc. 1998, 120, 2308-2321. (b) Resconi,

L.; Balboni, D.; Baruzzi, G.; Fiori, C.; Guidotti, S.; Mercandelli, P.; Sironi,

A. Organometallics 2000, 19, 420-429. This paper is concerned with the

competitive insertion at open and encumbered sites when both are present

in the same catalyst.

3578 Organometallics, Vol. 25, No. 15, 2006

Miller and Bercaw

mechanism, site epimerization or alternating, better predicts the

microstructure of the polypropylene produced. The enantio-

morphic site control model predicts that 3/MAO is employing

one site with a stereoselectivity of 95.2%. The alternating model

with R ) 1 predicts that 3/MAO employs two sites, one with a

stereoselectivity of 100.0% and the other with a stereoselectivity

of 90.6%. The alternating model with R < 1 predicts that

3/MAO employs two sites, each with a stereoselectivity of

95.2%. For this model the dependence of the RMS error on R

and â is somewhat shallow; for R ) 97.0% and â ) 93.0%,

the rms error ) 0.749, not significantly different from 0.687.

Related polymers subjected to these models have also provided

inconclusive results.²¹

Effect of Polymerization Temperature and Monomer

Concentration on Isotacticity. For the site epimerization

mechanism (Scheme 2), it is predicted that an increase in

polymerization temperature or a decrease in monomer concen-

tration will not significantly alter the polymer stereochemistry,

since this mechanism employs a single propagative transition

state with a stereoselectivity relatively independent of these

parameters. However, the polymer microstructure will be

sensitive to changes in polymerization temperature and monomer

concentration, if the alternating mechanism is the principal

pathway and the site epimerization pathway is accessible. As

the polymerization temperature increases or the monomer

concentration decreases, unimolecular site epimerization will

become increasingly likely, relative to bimolecular propagation.

To the extent that this occurs, the more stereoselective site will

be employed at the expense of the less stereoselective site.

Therefore, if the alternating mechanism is operating and the

site epimerization mechanism is accessible, one would anticipate

increased isotacticity with an increase in polymerization tem-

perature or a decrease in monomer concentration.

The literature reports that propylene polymerizations with

3/MAO conducted in liquid monomer at varying polymerization

temperatures⁴result in a shallow dependence of isotacticity on

polymerization temperature, as shown in Table 1. However, for

a similar series of experiments conducted in dilute monomer

(10 vol % in toluene), a more pronounced increase in isotacticity

is found with an increase in polymerization temperature.

Furthermore, a comparison between those polymerizations

conducted in liquid monomer and those conducted in 10%

monomer shows that higher isotacticity prevails under dilute

monomer conditions with ∆[mmmm] values of 4.7%, 10.8%,

and 11.9% for polymerization temperatures of 20, 40, and 60

°C, respectively. Although there are fewer runs to compare,

polymerization system 4/MAO reveals similar trends, albeit with

diminished magnitude (Table 1).

Figure 1. Statistical pentad analysis of a polypropylene sample

(from ref 4a) made with 3/MAO at 40 °C in liquid monomer.

the polymerization behavior of C₁-symmetric metallocene 3 and

related catalysts.¹⁶

Results and Discussion

Expected Pentad Distributions. Polymer stereochemistry

provides a permanent record of the stereochemical mechanism

for monomer enchainment. Therefore, observed polymer tac-

ticity can be compared to that predicted by each of the possible

mechanisms. Figure 1 presents such a comparison. The pentad

distribution reported for a polymer sample^4aprepared with

3/MAO in liquid monomer at 40 °C is subjected to three

statistical models.

The first is enantiomorphic site control,¹⁷which is predicted

by the site epimerization mechanism, since it employs a single

site with enantioselectivity R. The second is an alternating model

that is generally applicable to a catalyst that regularly alternates

insertions between a perfectly stereoselective site (R ) 1) and

a variably stereoselective site having a stereoselectivity equal

to â. This model employs a single independent parameter. The

third is an alternating model that is applicable to a catalyst that

regularly alternates insertions between two variably stereose-

lective sites. The stereoselectivity of one site is R, and the

stereoselectivity of the other site is â. This model employs two

independent parameters. Both alternating models assume that

no site epimerization is occurring. For a derivation of these

alternating models, see the Supporting Information.^18,19

Unfortunately, the RMS errors²⁰provided by these fits are

too similar to draw definitive conclusions concerning which

These results demonstrate that isotacticity can increase with

increasing polymerization temperature and decreasing monomer

concentration. The two-site alternating mechanism is likely

(19) A stochastic matrix method has been developed for C₁-symmetric

metallocene catalysts: (a) Busico, V.; Cipullo, R.; Monaco, G.; Vacatello,

M. Macromolecules 1997, 30, 6251-6263. (b) Busico, V.; Cipullo, R. Prog.

Polym. Sci. 2001, 26, 443-533. (c) Reference 7g. However, the models in

the Supporting Information section avoid matrix methods and are readily

applied with common software programs such as Microsoft Excel.

(20) The least-squares minimization was performed for the eight

measured intensities (mmrm, rrmr, and mrmr were combined) according

to RMS error ) ((∑(I_obs- I_calc)²)/8)^0.5. Cautionary note: Comparisons of

RMS errors and the resultant interpretations are always subject to the

limitations inherent in the statistical model. To the extent that stereochemical

behavior deviates from these idealized models, incorrect conclusions are

increasingly possible. Additionally, errors as high 1-3% are typical for

pentad distributions measured by ¹³C NMR for the polymers reported here.

(21) This includes polypropylenes made with metallocenes 3, 4, 6, 9,

13, 14, and 16.

(16) Perturbations introduced by a chain epimerization mechanism are

expected to be minimal and are neglected in the initial treatment presented

here. (a) Busico, V.; Cipullo, R. J. Am. Chem. Soc. 1994, 116, 9329-

9330. (b) Resconi, L.; Fait, A.; Piemontesi, F.; Colonnesi, M.; Rychlicki,

H.; Zeigler, R. Macromolecules 1995, 28, 6667-6676. (c) Janiak, C. In

Metallocenes: Synthesis, ReactiVity, Applications; Togni, A., Halterman,

R. L., Eds.; Wiley-VCH: Weinheim, 1998; p 596.

(17) (a) Farina, M. Top. Stereochem. 1987, 17, 1-111. (b) Ewen, J. A.

J. Am. Chem. Soc. 1984, 106, 6355-6364.

(18) Related models have been developed. (a) Reference 17a. (b) Farina,

M.; Di Silvestro, G.; Sozzani, P. Macromolecules 1993, 26, 946-950. (c)

Farina, M.; Terragni, A. Makromol. Chem., Rapid Commun. 1993, 14, 791-

798. (d) Farina, M.; Di Silvestro, G.; Terragni, A. Macromol. Chem. Phys.

1995, 196, 353-367. (e) Di Silvestro, G.; Sozzani, P.; Terragni, A.

Macromol. Chem. Phys. 1996, 197, 3209-3228. (f) Reference 13a.

Mechanism of Isotactic Polypropylene Formation

Organometallics, Vol. 25, No. 15, 2006 3579

Table 1. Dependence of Polymer Melting Temperature and

mmmm Pentad Content on Polymerization Temperature and

Monomer Concentration

Scheme 3

Table 2. Comparative Polymerizations between

Fluorenyl-Containing and Oct-Containing Metallocenes for

Various R Substituents (Table 5, entries 11-22)

^aSee ref 4b. ^bTable 5, entries 1-4. ^cTable 5, entries 7, 8. ^dTable 5,

entries 9, 10.

dominant, but can yield to the site epimerization mechanism

under certain conditions. These results and interpretations are

consistent with two reports: (1) that 3/MAO provides polymers

of increasing isotacticity as the monomer concentration is

progressively decreased (T_p) 50 °C; for liquid monomer, 1.7

and 0.9 M, [mmmm] ) 80%, 85%, and 87%, and T_m) 125,

129, and 134 °C, respectively)^22aand (2) that 3/MAO provides

polymers of increasing isotacticity with increasing polymeri-

zation temperature for reactions conducted in toluene with a

monomer pressure of 2.0 bar ([mmmm] ) 83.5% at 10 °C and

[mmmm] ) 87.8% at 70 °C).^22b

Steric Perturbation of the Fluorenyl Ligand. In addition

to performing polymerizations under various reaction conditions,

a powerful probative approach is steric modification of the

parent metallocene. To this end, 1,1,4,4,7,7,10,10-octamethyl-

1,2,3,4,7,8,9,10-octahydrodibenzo[b,h]fluorene²³has been in-

corporated as a more sterically hindered ligand (Oct) into various

C₁-symmetric metallocenes.²⁴

^aam. ) amorphous.

a substitution of the Flu ligand with the Oct ligand is expected

to result in decreased stereoselectivity at the less stereoselective

site (Scheme 3), because the Oct ligand is expected to compete

more favorably with the opposing R substituent for directing

the polymer chain during this propagative transition state.

Moreover, a larger change in stereoselectivity at the less

selective site might be anticipated. Therefore, if one observes

increased isotacticity upon substitution of Flu with Oct, the site

epimerization mechanism is likely operative, since it employs

only the more stereoselective site. Conversely, if one observes

decreased isotacticity upon substitution of Flu with Oct, the

alternating mechanism is likely operative, since it employs both

the more stereoselective site and the less stereoselective site.

Fluorenyl-containing metallocenes 2 (R ) methyl), 5 (R )

cyclohexyl), and 6 (R ) diphenylmethyl) were examined, as

were Oct-containing metallocenes 7 (R ) methyl), 8 (R )

cyclohexyl), and 9 (R ) diphenylmethyl) (Table 2). In each

case the isotacticity decreases substantially (∆[mmmm] ) 19.2%,

8.1%, and 11.7%, respectively for T_p) 0 °C) upon substitution

of the Flu ligand with the Oct ligand. This is especially

noteworthy for R ) diphenylmethyl; the isoselectivity of 6

([mmmm] ) 81.2% at T_p) 20 °C) is greater than that of 3 (R

) tert-butyl), whereas the isoselectivity of 9 ([mmmm] ) 76.4%

at T_p) 20 °C) is less than that of 3 ([mmmm] ) 79.2% at T_p

) 20 °C). The fact that the Oct-containing catalysts afford less

isotactic polypropylene relative to the Flu analogues, regardless

of R, implicates the alternating mechanism as the dominant

mechanism.

Substitution of the fluorenyl (Flu) ligand with the Oct ligand

is expected to result in increased stereoselectivity at the more

stereoselective site (Scheme 3), because the Oct ligand is

expected to be a better polymer chain directing substituent than

the Flu ligand during the propagative transition state. However,

(22) (a) Ewen, J. A.; Jones, R. L.; Elder, M. J. In Metalorganic Catalysts

for Synthesis and Polymerization, Recent Results by Ziegler-Natta and

Metallocene InVestigations; Kaminsky, W., Ed.; Springer: Berlin, 1999;

pp 150-169. (b) Kleinschmidt, R.; Reffke, M.; Fink, G. Macromol. Rapid

Commun. 1999, 20, 284-288. (c) Increasing isotacticity with decreasing

monomer concentration has also been identified in heterogeneous propylene

polymerization systems (ref 9).

(23) (a) Guilhemat, R.; Pereyre, M.; Petraud, M. Bull. Soc. Chim. Fr.

1980, 2, 334-344. (b) Gverdtsiteli, D. D.; Revazishvili, N. S.; Tsitsishvili,

V. G.; Kikoladze, V. S. Soobshch. Akad. Nauk. Gruz. SSR 1989, 133 (1),

77-80; Chem. Abstr. 1989, 111, 214206.

3580 Organometallics, Vol. 25, No. 15, 2006

Miller and Bercaw

Scheme 4

Table 3. Pentad and Dyad Distributions (%) Obtained for

Polypropylenes from 10/MAO (Table 5, entries 23-29)

neat C₃H₆

10% C₃H₆in toluene

T_p(°C) )

mmmm

mmmr

rmmr

mmrr

mmrm + rrmr

mrmr

rrrr

0

20

0

20

40

60

80

26.9

13.4

4.8

20.4

4.2

30.0

15.1

3.1

19.2

5.2

1.2

8.9

8.7

8.6

28.5

16.1

2.9

18.3

6.5

2.3

6.9

9.2

9.3

31.3

17.6

2.2

18.1

10.0

2.3

4.4

5.8

8.3

66.4

33.6

32.4

16.4

3.2

17.4

7.9

2.8

5.6

8.6

27.1

15.6

3.3

17.5

10.3

4.5

4.9

7.4

9.5

62.1

37.9

18.0

14.1

3.6

16.3

14.6

7.5

6.6

9.1

10.1

54.9

45.1

0.2

11.7

10.8

7.6

57.5

42.5

rrrm

mrrm

m

61.0

39.0

61.1

38.9

66.1

33.9

r

24), and Figure 4 illustrates the corresponding statistical fits

(see Supporting Information for details).²⁶Chain end control

and enantiomorphic site control are each single-site models.

Between these two, the latter provides a better statistical fit.

However, it poorly describes the pentad distribution and

correctly predicts the intensity ordering of only four out of the

nine resolved pentad peaks (#1, mmmm; #2, mmrr; #3, mmmr;

and #8, rmmr). The alternating models are each two-site models.

The alternating model with R ) 1 has only one independent

parameter, but correctly predicts the intensity ordering of all

seven allowed pentads. Not surprisingly, the alternating model

with R < 1, having two independent parameters, provides the

best fit of all (although this is technically an unfair comparison).

Figure 2. ORTEP drawing of the structure of Me₂C(3-tert-butyl-

4-Me-C₅H₂)(C₁₃H₈)ZrCl₂(10) with 50% probability ellipsoids.

Steric Perturbation of the Cyclopentadienyl Ligand. To

investigate the ability of the benzo substituent to direct the

growing polymer chain at the more stereoselective site of 3, a

methyl substituent on the cyclopentadienyl ligand opposed to

it was incorporated in zirconocene 10 (Scheme 4). Because the

condensation of Me₂C(3-Me-C₅H₄)(C₁₃H₉) with acetone occurs

selectively at the four position of the cyclopentadienyl ring, away

from the tertiary alkyl substituent, Me₂C(3-tert-butyl-4-Me-

C₅H₂)(C₁₃H₈)ZrCl₂(10) was obtained as a single regioisomer.

This structure was confirmed by an X-ray diffraction study

(Figure 2).²⁵

Figure 5 plots the RMS error of the statistical models as a

function of polymerization conditions for entries 23-29 (Table

5).²⁷In liquid propylene, the hemiisoselective alternating models

provide the best fits. In dilute monomer, the alternating model

with R < 1 excels the enantiomorphic site control model up to

80 °C.

Table 3 presents the pentad distributions measured for

polypropylenes obtained with 10/MAO under a variety of

polymerization conditions. The pentad distributions are mod-

erately dependent on polymerization temperature and monomer

concentration; the polymerization performed at 80 °C in 10%

monomer produced polypropylene with the most dissimilar

microstructure. As a representative example, the ¹³C NMR of

the methyl region for the polymer obtained in liquid monomer

at 20 °C (Table 5, entry 24) is displayed in Figure 3.

The statistical calculations are most consistent with the

operation of a two-site, alternating mechanism for 10/MAO in

liquid monomer. The single-site models cannot adequately

describe such polymers, which are essentially hemiisotactic with

a slight bias toward isotactic. However, the statistical results

indicate increased employment of the site epimerization mech-

anism by 10/MAO as the monomer concentration is decreased

and the polymerization temperature is increased.

The observed pentad distributions were subjected to a variety

of statistical models. Table 4 shows this analysis for the polymer

made with 10/MAO at 20 °C in liquid monomer (Table 5, entry

The parameters R and â derived from the alternating model

with R < 1 (Table 4) are the stereoselectivities of the two sites

and predict the stereochemical mechanism shown in Scheme

5. At the more stereoselective site, the growing polymer chain

is preferentially directed away from the tert-butyl group and

the enantiofacial selectivity is 94.5%, corresponding to ∆∆G^q

) 1.66 kcal/mol at 20 °C. At the less stereoselective site, benzo

and methyl are comparable in their abilities to direct the growing

(24) For reports of the Oct ligand see: (a) Miller, S. A.; Bercaw, J. E.

U.S. Patent 6,469,188, 2002. (b) Miller, S. A.; Bercaw, J. E. U.S. Patent

6,693,153, 2004. (c) Miller, S. A.; Bercaw, J. E. Organometallics 2004,

23, 1777-1789. (d) Irwin, L. J.; Reibenspies, J. H.; Miller, S. A. J. Am.

Chem. Soc. 2004, 126, 16716-16717. (e) Irwin, L. J.; Miller, S. A. J. Am.

Chem. Soc. 2005, 127, 9972-9973. (f) Irwin, L. J.; Reibenspies, J. H.;

Miller, S. A. Polyhedron 2005, 24, 1314-1324.

(25) Crystals of 10 were grown from benzene/n-heptane and were triclinic

P1h, a ) 9.286(3) Å, b ) 17.711(6) Å, c ) 17.903(6) Å, R ) 61.037(14)°,

â ) 81.471(14)°, γ ) 78.713(15)°, V ) 2521.2(13) Å³, Z ) 4, T ) 98 K.

Cambridge Crystallographic Data Center (CCDC, 12 Union Road, Cam-

bridge CB2 1EZ, UK) deposition number 137698.

(26) The least-squares minimization was performed for the nine measured

intensities according to RMS error ) ((∑(I_obs- I_calc)²)/9)^0.5

.

(27) The statistical analyses for the remaining polymers are found in

the Supporting Information.

Mechanism of Isotactic Polypropylene Formation

Organometallics, Vol. 25, No. 15, 2006 3581

Table 4. Statistical Pentad and Dyad (%) Comparison to Several Stereochemical Models for a Polypropylene Sample Made

with 10/MAO at 20 °C in Liquid Monomer (Table 5, entry 24)

chain end

control

enantiomorphic

site control

alternating

model R ) 1

alternating

model R < 1

observed

mmmm

mmmr

rmmr

30.0

15.1

3.1

29.1

21.0

3.8

30.1

16.7

2.8

31.6

14.6

4.4

29.6

15.3

4.0

mmrr

mmrm + rrmr

mrmr

19.2

5.2

1.2

7.6

23.8

7.6

16.7

11.3

5.7

23.5

0.0

21.0

4.7

2.3

rrrr

8.9

0.5

2.8

9.7

7.4

rrrm

8.7

2.8

5.7

8.8

8.0

mrrm

8.6

3.8

8.3

7.3

7.6

m

r

61.0

39.0

73.4

26.6

0.734

66.4

33.6

62.4

37.6

62.9

37.1

σ

R

0.786

3.53

1.000

0.624

2.44

0.945

0.646

1.02

σ

RMS error

8.73

Table 5. MAO-Cocatalyzed Propylene Polymerization Results with 3-16

zirconocene

M_w/M_n

entry

(mg)/MAO^a

T_p(°C)

toluene (mL)

C₃H₆(mL)

time (min)

yield (g)

T_m^b(°C)

[mmmm] (%)

M_w

1

2

3

4

5

6

7

8

9

3 (2.0)

4 (2.0)

4 (1.0)

2 (1.0)

5 (1.0)

6 (2.0)

0

20

40

60

0

20

0

20

0

20

0

20

0

20

0

20

0

20

0

20

0

20

0

20

0

20

40

60

80

0

20

0

20

0

30.0

2.0

30.0

2.0

1.0

30.0

1.0

2.0

3

60

10

5

20

3

0.84

0.56

0.50

0.38

0.93

1.01

0.45

1.77

0.30

1.71

1.43

4.95

0.71

4.01

0.32

0.47

0.37

0.86

0.36

5.53

0.05

0.14

1.23

1.12

1.90

1.82

1.16

0.47

0.10

1.50

1.08

0.18

1.62

0.16

0.36

0.41

0.83

3.88

2.13

1.38

0.87

0.50

0.03

0.02

0.29

0.70

129

134

135

128

126

125

120

118

121

131

n.o.

137

138

130

117

103

n.o.

125

135

n.o.

109

110

129

131

158

154

160

156

159

158

148

n.o.

167

163

82.2

83.9

88.8

89.4

79.5

81.5

74.1

77.0

78.1

77.6

21.6

18.3

13.2

14.5

86.1

81.2

2.4

3

30

3

15

5

431 000

252 000

1.74

1.88

30

15

10

20

15

5

20

3

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

3

30

3

80 000

1.81

7 (1.5)

8 (2.0)

9 (2.7)

2.4

5.1

7.3

74.4

76.4

26.9

30.0

28.5

31.3

32.4

27.1

18.0

28.4

31.4

60.2

57.5

78.6

80.0

>98

10 (1.0)

10 (0.5)

10 (1.0)

11 (0.5)^c

12 (1.0)

13 (2.0)

14 (1.0)

14 (2.0)

15 (3.0)

16 (2.0)

653 000

397 000

1.87

2.31

3

10

30

10

15

5

30

10

60

10

180

90

120

20

3

30

55

3

134 000

81 900

360 000

322 000

76 700

3.15

4.38

1.75

1.70

1.81

2.63

1.93

2.48

1.90

1.91

1.82

2.08

20

0

20

0

20

0

20

0

20

0

2.0

30.0

2.0

80 900

171 000

113 000

157 000

124 000

160 000

102 000

54 400

3

30

2.0

>99

370 000

425 000

1.39

1.77

20

^aMAO ) 1000 equiv unless otherwise specified. ^bn.o. ) melting temperature not observed. ^c2000 equiv of MAO used.

polymer chain. However, benzo is slightly more directing and

the enantiofacial selectivity is 64.6%, corresponding to ∆∆G^q

) 0.35 kcal/mol at 20 °C. Contrary to steric arguments^2c,4a,4b

that claim monomer insertion cannot occur while the growing

polymer chain is proximal to the tert-butyl group of 3/MAO,

the catalyst system 10/MAO readily inserts monomer at both

sites of the metallocene despite the presence of a bulky tert-

butyl group.

If indeed the alternating mechanism is operating for 3/MAO,

the two-parameter statistical model (Figure 1) suggests that the

3582 Organometallics, Vol. 25, No. 15, 2006

Miller and Bercaw

Figure 5. Dependence of the RMS error for a given statistical

model on polymerization temperature and monomer concentration

for polymerizations with 10/MAO.

Scheme 5

Figure 3. ¹³C NMR of the methyl region for polypropylene

obtained with 10/MAO in liquid monomer at 20 °C (Table 5, entry

24).

10/MAO (94.5%, tert-butyl > benzo) insofar as the methyl

group of 10 has no effect on the stereochemistry of insertion at

that site. Similarly, the stereoselectivity of the less stereoselective

site of 10/MAO (64.6%, benzo > methyl) should correspond

to that of the less stereoselective site of the known hemiiso-

selective catalyst system Me₂C(3-Me-C₅H₃)(C₁₃H₈)ZrCl₂/MAO

(2/MAO)^2a,4b(approximately 50%, benzo ≈ methyl) insofar the

tert-butyl group of 10 has no effect on the stereoselectivity at

that site. These stereoselectivity differences represent small

energy differences, and their magnitudes are proportional to the

steric perturbation: -1% for the addition of methyl to 3 and

+15% for the addition of tert-butyl to 2 (left side of Scheme

6). The magnitude of the stereoselectivity difference is markedly

larger when the steric perturbation occurs proximal to the

growing polymer chain: -31% for the addition of methyl to 3

and +94% for the addition of tert-butyl to 2 (right side of

Scheme 6).

Figure 4. Statistical pentad analysis of a polypropylene sample

made with 10/MAO at 20 °C in liquid monomer (Table 5, entry

24).

Steric Perturbation of the 3-Cyclopentadienyl Substituent.

Because 3/MAO likely employs the alternating rather than site

epimerization mechanism, steric perturbation of the 3-cyclo-

pentadienyl substituent should have an effect on polymer

stereochemistry much greater than previously thought. For this

reason, the zirconocenes shown in Scheme 7 were prepared.

coordination sites are operating with an average stereoselectivity

of 95.2% (benzo > H; tert-butyl > benzo). This stereoselectivity

should correspond to that of the more stereoselective site of

Mechanism of Isotactic Polypropylene Formation

Organometallics, Vol. 25, No. 15, 2006 3583

Scheme 6

stereochemistrysdespite their distal position relative to the metal

centersas evidenced by the comparatively meager isotacticity

reported for the isopropyl analogue Me₂C(3-isopropyl-C₅H₃)-

(C₁₃H₈)ZrCl₂/MAO: [mmmm] ) 15.4% for T_p) 10 °C and

[mmmm] ) 44.0% for T_p) 70 °C (2.0 bar of propylene in

toluene).^22b,31

The second largest cyclopentadienyl substituent employed,

2-methyl-2-adamantyl,³²is incorporated into metallocene 14.

This metallocene is capable of producing highly isotactic

polypropylene ([mmmm] > 98%) with a melting temperature

of 160 °C (Table 5, entry 38, T_p) 0 °C). The high isoselectivity

can be explained by one of two limiting scenarios. First, the

alternating mechanism is operating and the 2-methyl-2-ada-

mantyl substituent is an exceedingly good polymer-directing

substituent compared to benzo, rendering a catalyst with two

highly stereoselective sites. Second, the steric demands of the

2-methyl-2-adamantyl substituent prohibit use of that site, and

the alternating mechanism is no longer operating; the site

epimerization mechanism is now dominant, and metallocene 14

employs only one highly stereoselective site for monomer

insertion. In either case, it is difficult to rationalize the

comparatively poor isoselectivity of 3/MAO ([mmmm] ) 79.5%)

if one claims that 3/MAO always operates by the site epimer-

ization mechanismsfurther evidence supporting the alternating

mechanism for 3/MAO.

Scheme 7

The largest cyclopentadienyl substituent employed, 2-phenyl-

2-adamantyl, is incorporated into zirconocene 15. 15/MAO is

essentially inactive for propylene polymerization. Apparently,

the cyclopentadienyl substituent is so large that both the

alternating and site epimerization mechanisms have effectively

halted.³³

Formation of Highly Isotactic Polypropylene with C₁-

Symmetric Metallocene Catalysts. We sought to apply ev-

erything we have learned about isotactic polypropylene forma-

tion with C₁-symmetric zirconocene catalysts to design a

zirconocene precatalyst capable of producing very highly

isotactic polypropylene. Zirconocene dichloride precatalyst 16

is the result of this endeavor. Its structural features result from

three important factors. First, the cyclopentadienyl substituent

should be larger than tert-butyl, but not so large that it inhibits

polymerization altogether. Therefore, the 2-methyl-2-adamantyl

substituent has been incorporated into 16. Second, factors that

encourage site epimerization generally lead to polymers of

Each was subjected to MAO-cocatalyzed propylene polymeriza-

tions at 0 and 20 °C, as reported in entries 15, 16, and 30-44

of Table 5.

Catalyst system 11/MAO employs the alternating mech-

anism²⁸and affords essentially hemiisotactic polypropylene with

[mmmm] ) 28.4%. While 2-adamantyl is large, its steric effect

is not as great as a tertiary alkyl substituent, and apparently it

is far inferior to tert-butyl in its ability to direct the growing

polymer chain during the transition state for monomer insertion.

Zirconocene 12 is identical to 11 except that it bears a

dimethylsilylene bridge. Such silicon-containing metallocenes

often undergo site epimerization more readily than alkylidene

analogues.^2d,3d,29More rapid site epimerization for 12/MAO

relative to 11/MAO is indeed apparent, as [mmmm] doubles to

60.2%, indicative that this metallocene employs the more

stereoselective site to a greater degree.³⁰

Zirconocenes 13 and 6 produce polypropylene with isotac-

ticities ([mmmm] ) 78.6%, 80.0% for 13 and 86.1%, 81.2%

for 6) comparable to those produced by the parent tert-butyl-

substituted zirconocene 3 ([mmmm] ) 79.5% and 81.5%),

despite the fact that they contain secondary alkyl substituents

on the cyclopentadienyl ring. Thus, it is not necessary to have

a tertiary alkyl cyclopentadienyl substituent to obtain moderately

isotactic polypropylene. Moreover, atoms beyond the R and â

carbons of the substituent can greatly impact the polymer

(31) Yano, A.; Kaneko, T.; Sato, M.; Akimoto, A. Macromol. Chem.

Phys. 1999, 200, 2127-2135.

(32) For previous examples of metallocenes bearing the 2-methyl-2-

adamantyl substituent, see: Abrams, M. B.; Yoder, J. C.; Loeber, C.; Day,

M. W.; Bercaw, J. E. Organometallics 1999, 18, 1389-1401.

(33) 15/MAO is essentially inactive for ethylene polymerization as well

(activity ) 12 gP/(gZr‚h) at 20 °C). To test the suggestion that this

metallocene simply deactivates by aryl C-H activation, Me₂C(3-(2-(CH₂-

SiMe₃)-2-adamantyl)-C₅H₃)(C₁₃H₈)ZrCl₂(17) was prepared and tested for

its MAO-cocatalyzed propylene polymerization behavior.

(28) (a) Miller, S. A.; Bercaw, J. E. Organometallics 2002, 21, 934-

945. (b) Miller, S. A. Macromolecules 2004, 37, 3983-3995.

(29) Patsidis, K.; Alt, H. G.; Milius, W.; Palackal, S. J. J. Organomet.

Chem. 1996, 509, 63-71.

(30) Miller, S. A. Ph.D. Thesis, California Institute of Technology, 2000,

Chapter 2.

The small amount of amorphous polypropylene (activity ) 460 gP/(gZr‚h)

at 20 °C) that was obtained is attributed to impurities present in the

metallocene (likely Me₂C(3-(2-adamantyl)-C₅H₃)(C₁₃H₈)ZrCl₂based on the

synthetic procedure), and the lack of any isotactic polypropylene suggests

that the aliphatic cyclopentadienyl substituent 2-(CH₂SiMe₃)-2-adamantyl

effectively inhibits polymerization as well.

3584 Organometallics, Vol. 25, No. 15, 2006

Miller and Bercaw

cannot conclusively differentiate between a single-site model

employing enantiomorphic site control (site epimerization

mechanism) and a two-site model having one highly stereose-

lective site and one moderately stereoselective site (alternating

mechanism). Other approaches, therefore, were developed to

interrogate the two possible mechanisms.

The following observations suggest that the alternating

mechanism predominates, while the site epimerization mech-

anism can compete under certain conditions for 3/MAO and

other closely related C₁-symmetric zirconocenes/MAO. An

increase in isotacticity is observed for polymerization conditions

that favor unimolecular site epimerization over bimolecular

propagation. Incorporation of the bulky Oct ligand in place of

fluorenyl effects a decrease in isotacticity, a change consistent

with a slight increase in stereoselectivity at one site and a sub-

stantial decrease in stereoselectivity at a second site. The model

system Me₂C(3-tert-butyl-4-Me-C₅H₂)(C₁₃H₈)ZrCl₂/MAO (10/

MAO) produces essentially hemiisotactic polypropylene, sug-

gesting that both sites of the metallocene are readily employed

for insertion, despite the presence of a bulky tert-butyl group.

Finally, the use of a cyclopentadienyl substituent larger than

the tert-butyl group results in increased isotacticity. The

combination of these results implies operation of the alternating

mechanism with 3/MAO in which the two enantioselectivities

are somewhat different but have an average of 95.2% at 40 °C.

Three key elements contributed to the design of a highly

isoselective metallocene catalyst system, Me₂C(3-(2-Me-2-

adamantyl)-C₅H₃)(C₂₉H₃₆)ZrCl₂/MAO (16/MAO): incorpora-

tion of a cyclopentadienyl substituent larger than tert-butyl; con-

siderable steric bulk on one side of the metallocene to encourage

site epimerization; and exploitation of the enhanced ability of

the octamethyloctahydrodibenzofluorenyl (Oct) ligand to direct

the growing polymer chain away in the transition state for

monomer insertion. This catalyst system is capable of producing

highly isotactic polypropylene ([mmmm] > 99%, T_mup to 167

°C) and demonstrates how the correct understanding of the

mechanism can lead to the rational design of improved catalysts.

Figure 6. Highly isoselective polymerization catalyst 16/MAO

likely employs a single propagative transition state (shown), yielding

isotactic polypropylene largely devoid of stereoerrors as determined

by ¹³C NMR spectroscopy (methyl region depicted).

higher isotacticity, since the more stereoselective site is used

preferentially. This can be accomplished, in principle, by altering

polymerization conditions³⁴or by the inclusion of a dimethyl-

silylene bridge. For 16, it is plausible that site epimerization

predominates because of extreme steric crowding contributed

by both the 2-methyl-2-adamantyl substituent and the opposing

Oct ligand. Third, to the extent that the catalyst system utilizes

a given site for monomer insertion, enhancement of the

stereoselectivity at that site will lead to higher isotacticity. Since

the Oct ligand is a better polymer-directing group than fluorenyl,

incorporation of the Oct ligand in metallocene 16 is expected

to lead to greater isoselectivity at the more stereoselective site.

Experimental Section

General Considerations and Instrumentation. Unless other-

wise noted, all reactions and procedures are carried out under argon

or nitrogen using standard glovebox, Schlenk, and high-vacuum

line techniques.³⁶Solvents are dried according to standard proce-

dures. NMR spectra were recorded on a JEOL GX-400 (¹H, 399.78

MHz; ¹³C, 100.53 MHz) spectrometer interfaced with the Delta

software package. GC-MS were acquired with a Hewlett-Packard

5890 Series II gas chromatograph connected to a Hewlett-Packard

5989A mass spectrometer. The GC was equipped with a column

of dimensions 7.1 m × 0.1 µm having an HP-1 phase (cross-linked

methyl silicone gum). LC-MS were acquired with a Hewlett-

Packard 1090 Series II liquid chromatograph with a toluene phase

(solvent dried over sodium/benzophenone). The LC was connected

to a Hewlett-Packard 59980B particle beam interface, and this was

connected to a Hewlett-Packard 5989A mass spectrometer. The

elemental analysis of air-sensitive metallocenes routinely provides

numbers lower than those calculated. In some cases this can be

attributed to residual LiCl, but it is more likely a systemic problem

related to incomplete combustion.

The propylene polymerization results with 16/MAO are given

in entries 45 and 46 of Table 5. Highly isotactic polypropylenes

are obtained, as stereoerrors are virtually absent by ¹³C NMR

analysis ([mmmm] > 99%, Figure 6). The polymers have high

melting temperatures (167.0 and 162.7 °C, respectively) and

large enthalpies of melt (92.0 and 87.5 J/g, respectively).

Significantly, the highest previously reported melting temper-

ature for an isotactic polypropylene made via homogeneous

catalysis is 166 °C.³⁵The high isoselectivity of 16/MAO

suggests that it employs a single propagative transition state

for monomer insertion, as depicted in Figure 6.

Conclusions

Although a site epimerization mechanism for isotactic

polypropylene formation with Me₂C(3-tert-butyl-C₅H₃)(C₁₃H₈)-

ZrCl₂/MAO (3/MAO) has been invoked, experimental evidence

does not conclusively support it. The microstructure of this

material is sufficiently isotactic that stereochemical analyses

Zirconocene Dichloride Syntheses. Preparation of 2 and 3.

Metallocenes 2 and 3 were synthesized as described in the

literature.^4a,cFor 3: MS (LC-MS) m/z 488.6 (M⁺). Anal. Calcd

for C₂₅H₂₆Zr₁Cl₂: C, 61.46; H, 5.36. Found: C, 43.94; H, 4.31.

(34) In addition to low monomer concentration and high polymerization

temperatures, the addition of methylene chloride has been observed to

increase the relative rate of the site epimerization process for 1/MAO: Fink,

G.; Herfert, N.; Montag, P. In Ziegler Catalysts, Recent Scientific InnoVa-

tions and Technological ImproVements; Fink, G., Mu¨lhaupt, R., Brintzinger,

H.-H., Eds.; Springer: Berlin, 1995; pp 159-179.

(36) Burger, B. J.; Bercaw, J. E. In Experimental Organometallic

Chemistry: A Practicum in Synthesis and Characterization, Vol. 357;

Wayda, A. L., Darensbourg, M. Y., Eds.; American Chemical Society:

Washington D.C., 1987; pp 79-98.

(35) Ewen, J. A.; Elder, M. J.; Jones, R. L.; Rheingold, A. L.; Liable-

Sands, L. M.; Sommer, R. D. J. Am. Chem. Soc. 2001, 123, 4763-4773.

Mechanism of Isotactic Polypropylene Formation

Organometallics, Vol. 25, No. 15, 2006 3585

Fluorenyllithium/Diethyl Ether. A 500 mL flask was charged

with fluorene (47.00 g, 282.8 mmol) and attached to a swivel frit

before 200 mL of diethyl ether was condensed in. n-Butyllithium

solution (180.0 mL, 288 mmol, 1.6 M in hexanes) was syringed in

over 20 min at room temperature. After stirring for 18 h, the yellow

precipitate was collected and dried in vacuo: 50.64 g (72.7% based

on the mono diethyl ether adduct).

CH₀, not determined. Anal. Calcd for C₃₅H₃₀Zr₁Cl₂: C, 68.61; H,

4.93. Found: C, 64.90; H, 4.56.

6,6-(Pentamethylene)fulvene. (Synthesis modified from ref 37.)

Pyrrolidine (30.0 mL, 359 mmol) was slowly syringed into a

solution of cyclohexanone (150.0 mL, 1447 mmol) and cyclopen-

tadiene (100.0 mL, 1213 mmol) in 100 mL of methanol. The

reaction was stirred for 96 h before 40 mL of acetic acid was added,

followed by 300 mL of H₂O and 200 mL of diethyl ether. The

organic layer was isolated, and the aqueous layer was extracted

with diethyl ether (3 × 50 mL). The combined organic layers were

extracted with H₂O (3 × 30 mL) and 10% aqueous NaOH (3 × 30

mL). The organic layer was dried over MgSO₄, filtered, and

rotavapped to give 158.8 g of a yellow oil, which was subjected to

Kugelrohr distillation under high vacuum. The first 20 g of material

that distilled at 50 °C was discarded, and the product was obtained

from the second fraction that distilled at 80 °C: 110.13 g (61.1%).

Cyclohexylcyclopentadiene. 6,6-(Pentamethylene)fulvene (15.66

g, 107.1 mmol) was dissolved in 50 mL of tetrahydrofuran, and

this solution was added over 12 min to a stirred slurry of LiAlH₄

(4.500 g, 118.6 mmol) in 100 mL of tetrahydrofuran at 0 °C. After

15 h of stirring at room temperature, the reaction was cooled to 0

°C and quenched by slow addition of 20 mL of saturated NH₄Cl

solution. Then 300 mL of H₂O and 50 mL of diethyl ether were

added; the organic layer was isolated, and the aqueous layer was

extracted with additional diethyl ether (2 × 50 mL). The combined

organic layers were dried over MgSO₄, filtered, and rotavapped to

give the product, 2-cyclohexylcyclopentadiene, in quantitative yield

as a light yellow oil: 15.88 g.

3-tert-Butyl-6,6-diphenylfulvene. An argon-filled 1 L Schlenk

flask was charged with 6,6-dimethylfulvene (40.15 g, 378.2 mmol)

and 180 mL of diethyl ether. At 0 °C, methyllithium/lithium

bromide solution (420 mL, 630 mmol, 1.5 M in diethyl ether) was

syringed in over 25 min. The reaction was stirred for 7 days before

it was cooled to 0 °C, and 60 mL of aqueous NH₄Cl solution was

slowly added, followed by 120 mL of water. The organic layer

was isolated and the aqueous layer was extracted with diethyl ether

(3 × 50 mL). The combined organic layers were dried over MgSO₄,

filtered, and rotavapped to provide 38.41 g of tert-butylcyclopen-

tadiene (83.1%). 15.00 g of this material was combined with 100

mL of ethanol and 22.37 g of benzophenone. The solids were

dissolved before sodium methoxide (15.00 g, 278 mmol) was added.

The reaction was stirred for 27 days before 500 mL of water and

200 mL of diethyl ether were added. The organic layer was isolated,

and the aqueous layer was extracted with diethyl ether (3 × 50

mL). The combined organic layers were dried over MgSO₄, filtered,

and rotavapped to provide a red oil, which was subjected to

Kugelrohr distillation. Under high vacuum, 16.56 g was removed

and the next fraction was collected as product at 60 °C: 9.22 g of

red, viscous oil (26.2%).

3-Cyclohexyl-6,6-dimethylfulvene. To cyclohexylcyclopenta-

diene (15.88 g, 107.7 mmol) was added 100 mL of methanol,

acetone (20.0 mL, 272 mmol), and pyrrolidine (1.0 mL, 12 mmol).

After stirring for 21 h, 5 mL of acetic acid was injected, followed

by 150 mL of H₂O and 100 mL of diethyl ether. The organic layer

was isolated and the aqueous layer extracted with diethyl ether (3

× 50 mL). The combined organic layers were extracted with H₂O

(3 × 30 mL) and with 10% aqueous NaOH (3 × 30 mL), dried

over MgSO₄, filtered, and rotavapped. The product was obtained

in quantitative yield (20.17 g) as a yellow liquid and further purified

by passing the neat liquid through a short column of alumina.

Me₂C(3-cyclohexyl-C₅H₃)(C₁₃H₈)H₂. A 15.5 mL portion of an

n-butyllithium solution (24.8 mmol, 1.6 M in hexanes) was syringed

into a solution of fluorene (4.047 g, 24.35 mmol) in 60 mL of

tetrahydrofuran. After stirring for 45 min, 3-cyclohexyl-6,6-

dimethylfulvene (4.58 g, 24.3 mmol) was injected via syringe. After

stirring for 15 h, 60 mL of a saturated NH₄Cl solution was added,

the organic layer was isolated, and the aqueous layer was extracted

with diethyl ether (2 × 25 mL). The combined organic layers were

dried over MgSO₄, filtered, and rotavapped to give the product in

quantitative yield (8.63 g) as a yellow oil.

Me₂C(3-cyclohexyl-C₅H₃)(C₁₃H₈)Li₂. The dianion was prepared

by treating a solution of Me₂C(3-cyclohexyl-C₅H₃)(C₁₃H₈)H₂(8.63

g, 24.3 mmol) in 50 mL of diethyl ether with 32.0 mL of

n-butyllithium solution (51.2 mmol, 1.6 M in hexanes) at 0 °C.

After stirring for 20 h, the solvent was removed by vacuum transfer,

and 75 mL of petroleum ether was condensed in. The dilithio salt

was isolated by filtration and in vacuo drying in quantitative yield

as a red-orange powder.

Me₂C(3-cyclohexyl-C₅H₃)(C₁₃H₈)ZrCl₂(5). A 2.500 g portion

of Me₂C(3-cyclohexyl-C₅H₃)(C₁₃H₈)Li₂(6.82 mmol) and 1.59 g of

ZrCl₄(6.82 mmol) were combined in a swivel frit apparatus. Then

30 mL of petroleum ether was condensed in at -78 °C. This was

allowed to warm slowly to room temperature before solvent removal

after 17 h of stirring. Methylene chloride (40 mL) was condensed

in and removed in order to quench unreacted ligand. Then the

orange solid was extracted in the swivel frit with 50 mL of refluxing

diethyl ether. The volume was reduced to 20 mL, and two crops

Ph₂C(3-tert-butyl-C₅H₃)(C₁₃H₈)H₂. A 250 mL flask was charged

with waxy 3-tert-butyl-6,6-diphenylfulvene (9.22 g, 32.2 mmol)

and fluorenyllithium diethyl ether adduct (7.928 g, 32.19 mmol).

Diethyl ether (75 mL) was condensed in, and the homogeneous

reaction formed much precipitate after 16 days. After 20 days, 60

mL of aqueous NH₄Cl solution was slowly added, and the organic

layer was isolated. The aqueous layer was extracted with diethyl

ether (2 × 30 mL), and the combined organic layers were dried

over MgSO₄and filtered. The product crystallized from solution

at -78 °C, and the product was obtained as a white powder in two

crops: 6.58 g (45.2%). MS (GC-MS): m/z 452.5 (M⁺). Anal. Calcd

for C₃₅H₃₂: C, 92.87; H, 7.13. Found: C, 91.37; H, 6.56.

Ph₂C(3-tert-butyl-C₅H₃)(C₁₃H₈)Li₂. A swivel frit was charged

with Ph₂C(3-tert-butyl-C₅H₃)(C₁₃H₈)H₂(6.359 g, 14.05 mmol), and

75 mL of diethyl ether. n-butyllithium solution (20.0 mL, 32.0

mmol, 1.6 M in hexanes) was syringed in over 2 min at room

temperature. After 22 h, the orange precipitate was collected and

dried in vacuo to provide the product in theoretical yield (6.53 g).

Ph₂C(3-tert-butyl-C₅H₃)(C₁₃H₈)ZrCl₂(4). A 100 mL flask was

charged with Ph₂C(3-tert-butyl-C₅H₃)(C₁₃H₈)Li₂(3.987 g, 8.583

mmol) and ZrCl₄(2.000, 8.583 mmol) and equipped with a 180°

needle valve. Petroleum ether (60 mL) was condensed in at -78

°C and the cold bath removed. After 42 h, solvent was removed

from the pink slurry. The solid was extracted in a cellulose

extraction thimble with 150 mL of methylene chloride overnight.

The filtrate was attached to a swivel frit, filtered, and condensed

to 10 mL. The precipitate was collected and dried in vacuo: 1.841

g and a second crop of 0.614 g (46.7% yield for both crops). MS

1

(LC-MS): m/z 612.6 (M⁺). H NMR (CD₂Cl₂): δ 1.18 (s, 9H,

C(CH₃)₃), 5.61, 5.77, 6.22 (m, 3H, Cp-H), 6.39, 6.43, 8.18, 8.18

(d, ³J_HH) 8.8, 8.8, 8.4, 8.4 Hz, 4H, Flu-H), 6.96, 6.99, 7.85, 7.85

3

(t, J_HH) 7.0, 7.7, 7.3, 7.3 Hz, 4H, Flu-H), 7.33, 7.33, 7.95, 7.99

(d, 4H, phenyl-H), 7.30, 7.53, 7.46, 7.48, 7.54, 7.57 (t, 6H, phenyl-

H). ¹³C NMR (CD₂Cl₂, 40 °C): δ 29.90 (C(CH₃)₃), 33.20

(C(CH₃)₃), 101.33, 105.86, 115.11 (Cp-CH₁), 121.05, 121.51,

123.27, 124.00 (fluorenyl-CH₀), 123.60, 124.26, 124.46, 124.79,

125.40, 125.48, 126.63, 126.79, 127.19, 127.23, 127.94, 128.01,

129.03, 129.12, 129.12, 129.12, 129.26, 129.39 (benzo-CH₁and

Cp-CH₁), 145.03, 145.11 (ipso-C), 146.18 (9-fluorenyl-C), other

(37) Stone, K. J.; Little, R. D. J. Org. Chem. 1984, 49, 1849-1853.

3586 Organometallics, Vol. 25, No. 15, 2006

Miller and Bercaw

were obtained for a total of 1.261 g (35.9%) of 5 as an orange

powder following collection at 0 °C and in vacuo drying. MS (LC-

stirring, solvent was removed. Methylene chloride (50 mL) was

condensed in; the solution was warmed and stirred before solvent

removal. The solid was extracted for 64 h in a cellulose extraction

thimble with 150 mL of methylene chloride. The filtrate volume

was reduced to 50 mL, and the precipitated product was collected

on a swivel frit and dried in vacuo: 1.520 g of 6 (59.2%). MS

(LC-MS): m/z 598.5 (M⁺). ¹H NMR (C₆D₆): δ 1.59, 1.76 (s, 6H,

(CH₃)₂C-Flu-Cp), 5.24, 5.39, 5.77 (m, 3H, Cp-H), 5.92 (s, 1H,

CHPh₂), 6.90, 6.94, 7.29, 7.33 (t, ³J_HH) 7.0, 7.3, 7.7, 7.7 Hz, 4H,

Flu-H), 6.96-7.15 (m, 10H, phenyl-H), 7.39, 7.42, 7.82, 7.85 (d,

³J_HH) 8.8, 8.0, 8.4, 8.4 Hz, 4H, Flu-H). Anal. Calcd for C₃₄H₂₈-

Zr₁Cl₂: C, 68.21; H, 4.71. Found: C, 52.61; H, 3.82.

1

MS): m/z 514.7 (M⁺). H NMR (C₆D₆): δ 0.87-1.26 (m, 10H,

cyclohexyl-H), 1.81, 1.82 (s, 6H, C(CH₃)₂), 2.58 (m, 1H, 1-H-

3

cyclohexyl), 5.27, 5.40, 6.05 (t, J_HH) 2.6, 2.6, 2.6 Hz, 3H, Cp-

H), 7.01, 7.03, 7.30, 7.35 (t, ³J_HH) 7.0, 6.9, 8.4, 7.0 Hz, 4H, Flu-

3

H), 7.45, 7.47, 7.83, 7.83 (d, J_HH) 8.0, 8.1, 8.4, 8.4 Hz, 4H,

Flu-H). Anal. Calcd for C₂₇H₂₈Zr₁Cl₂: C, 63.01; H, 5.48. Found:

C, 57.74; H, 5.53.

6,6-Diphenylfulvene. (Synthesis modified from ref 38.) Sodium

methoxide (41.00 g, 759.0 mmol), ethanol (500 mL), and ben-

zophenone (125.00 g, 686.0 mmol) were added to a 1 L vessel.

Cyclopentadiene (100.0 mL, 1213 mmol) was poured in, giving a

red solution. After stirring for 7 days, the orange precipitate was

collected by filtration and rinsed with 50 mL of ethanol. The solid

was refluxed in 200 mL of methanol for 1 h. Upon cooling, the

solid was collected, rinsed with 75 mL of methanol, and dried in

vacuo for 48 h to provide the product as an orange powder: 136.18

g (86.2%). MS (GC-MS): m/z 230.3 (M⁺). Anal. Calcd for

C₁₈H₁₄: C, 93.87; H, 6.13. Found: C, 92.60, 92.59; H, 5.37, 5.19.

(Diphenylmethyl)cyclopentadiene. A 500 mL flask was charged

with LiAlH₄(4.50 g, 119 mmol) and 100 mL of tetrahydrofuran.

An addition funnel containing 6,6-diphenylfulvene (20.00 g, 86.84

mmol) dissolved in 100 mL of tetrahydrofuran was attached. The

vessel was cooled to 0 °C before dropwise addition over 45 min.

After 22 h of stirring at room temperature, the vessel was cooled

to 0 °C, and 60 mL of aqueous NH₄Cl solution was added dropwise.

Then, 300 mL of water and 20 mL of concentrated aqueous HCl

were added before the organic layer was isolated. The aqueous layer

was extracted with diethyl ether (3 × 50 mL). The combined

organic layers were dried over MgSO₄, filtered, rotavapped, and

dried in vacuo to provide the product in quantitative yield (20.17

g) as a light yellow oil.

3-(Diphenylmethyl)-6,6-dimethylfulvene. A 500 mL flask was

charged with (diphenylmethyl)cyclopentadiene (10.00 g, 43.0

mmol), 50 mL of methanol, acetone (20.0 mL, 272 mmol), and

pyrrolidine (5.0 mL, 60 mmol). After stirring for 67 h, the yellow

precipitate was collected by suction filtration, was washed with 20

mL methanol, and was dried in vacuo: 8.24 g (70.3%).

Me₂C(3-(diphenylmethyl)-C₅H₃)(C₁₃H₈)H₂. A 250 mL flask

was charged with fluorene (3.661 g, 22.03 mmol), evacuated, and

backfilled with argon before 50 mL of tetrahydrofuran and 14.0

mL of n-butyllithium solution (22.4 mmol, 1.6 M in hexanes) were

syringed in. The orange solution was stirred for 1 h before a solution

of 3-(diphenylmethyl)-6,6-dimethylfulvene (6.00 g, 22.03 mmol)

in 15 mL of tetrahydrofuran was syringed in. Following an

additional 16 h, the stirred reaction was quenched by slow addition

of 60 mL of aqueous NH₄Cl. The organic layer was isolated and

the aqueous layer extracted with diethyl ether (2 × 25 mL). The

combined organic layers were dried over MgSO₄, filtered, and

rotavapped to give the product in quantitative yield (9.66 g) as a

light yellow oil.

2,5-Dichloro-2,5-dimethylhexane. A 2 L argon-purged vessel

was charged with 2,5-dimethyl-2,5-hexanediol (200.00 g, 1.368

mol), and concentrated aqueous hydrochloric acid (1.00 L, 12.2

mol of HCl) was poured in. The white slurry was shaken and stirred

for 17 h. The white solid was collected by suction filtration and

rinsed with 500 mL of water. The solid was dissolved in 1.00 L of

diethyl ether, the small water layer was removed, and the organic

layer was dried over MgSO₄. The solution was forced through a

short column of alumina, solvent was removed from the filtrate by

rotary distillation, and the white crystalline solid was briefly (30

min) dried in vacuo to provide the product: 237.96 g (95.0%). ¹H

NMR (CDCl₃): δ 1.55 (s, 12H, CH₃), 1.90 (s, 4H, CH₂). ¹³C NMR

(CDCl₃): δ 32.59 (CH₃), 41.21 (CH₂), 70.13 (CH₀). Anal. Calcd

for C₈H₁₆Cl₂: C, 52.47; H, 8.81. Found: C, 52.65, 52.35; H, 9.74,

9.39.

Octamethyloctahydrodibenzofluorene. A 2 L argon-purged

vessel was charged with fluorene (36.00 g, 216.6 mmol) and 2,5-

dichloro-2,5-dimethylhexane (80.00 g, 436.9 mmol). The solids

were dissolved in 600 mL of nitromethane, and the vessel was

equipped with an addition funnel, which was charged with AlCl₃

(38.50 g, 289 mmol) dissolved in 100 mL of nitromethane. The

solution was added over 10 min, and the purple reaction was stirred

for 20 h before it was slowly poured into 700 mL of ice water.

The precipitate was collected by filtration and refluxed in 500 mL

of ethanol for 2 h. Upon cooling, the solid was collected by

filtration, and this was refluxed in 300 mL of hexanes for 2 h. After

cooling, the solid was collected by filtration and dried in vacuo,

giving the product as a white powder: 62.53 g (74.7%). MS (GC-

1

MS): m/z 386.5 (M⁺). H NMR (Cl₂DCCDCl₂): δ 1.38, 1.43 (s,

24H, CH₃), 1.77 (apparent s, 8H, CH₂), 3.82 (s, 2H, CH₂), 7.49,

7.71 (s, 4H, Flu-H). ¹³C NMR (Cl₂DCCDCl₂): δ 32.37, 32.53

(CH₃), 34.68, 34.71 (CH₀), 35.50, 35.55 (CH₂), 36.47 (CH₂), 117.48,

123.31 (CH₁), 139.20, 140.80, 143.50, 143.66 (CH₀). Anal. Calcd

for C₂₉H₃₈: C, 90.09; H, 9.91. Found: C, 89.07, 89.16; H, 8.94,

8.85.

3,6,6-Trimethylfulvene. A 1 L flask was charged with 400 mL

of methanol, methylcyclopentadiene (120.0 mL, 1.21 mol), acetone

(200 mL, 2.72 mol), and pyrrolidine (40.0 mL, 0.464 mol). After

stirring the orange solution for 71 h, 50 mL of acetic acid was

added, followed by 1200 mL of H₂O and 200 mL of diethyl ether.

The organic layer was isolated, and the aqueous layer was extracted

with diethyl ether (5 × 100 mL). The combined organic layers

were extracted with H₂O (3 × 30 mL) and 10% aqueous NaOH (3

× 30 mL). The organic layer was dried over MgSO₄, filtered, and

rotavapped to give 158.8 g of a red-orange oil, which was subjected

to Kugelrohr distillation under high vacuum. The first 15 g of

material that distilled at room temperature was discarded, and the

product was obtained from the second fraction that distilled at 50

°C: 136.58 g (94.0%).

Me₂C(3-(diphenylmethyl)-C₅H₃)(C₁₃H₈)Li₂. A round-bottom

flask containing 9.66 g (22.0 mmol) of Me₂C(3-(diphenylmethyl)-

C₅H₃)(C₁₃H₈)H₂was attached to a swivel frit and evacuated before

75 mL of diethyl ether was condensed in. At 0 °C, 28.0 mL of

n-butyllithium solution (44.8 mmol, 1.6 M in hexanes) was syringed

in over 2 min. After stirring for 18 h at room temperature, the red

precipitate was collected and dried in vacuo to provide the product

in quantitative yield (9.92 g).

Me₂C(3-(diphenylmethyl)-C₅H₃)(C₁₃H₈)ZrCl₂(6). In the glove-

box, 1.933 g of Me₂C(3-(diphenylmethyl)-C₅H₃)(C₁₃H₈)Li₂(4.29

mmol) was combined with ZrCl₄(1.00 g, 4.29 mmol) in a 100 mL

round-bottom flask. This was attached to a swivel frit, and 50 mL

of petroleum ether was condensed in by vacuum transfer at -78

°C. The vessel was allowed to warm slowly, and after 24 h of

Me₂C(3-methyl-C₅H₃)(C₂₉H₃₆)H₂. A 13.5 mL portion of an

n-butyllithium solution (21.6 mmol, 1.6 M in hexanes) was syringed

into a solution of octamethyloctahydrodibenzofluorene (8.00 g, 20.7

mmol) in 90 mL of tetrahydrofuran. After stirring for 90 min, 3,6,6-

trimethylfulvene (2.487 g, 20.7 mmol) was injected via syringe into

the red solution. After stirring for 22 h, 60 mL of a saturated NH₄-

(38) Alper, H.; Laycock, D. E. Synthesis 1980, 10, 799.

Mechanism of Isotactic Polypropylene Formation

Organometallics, Vol. 25, No. 15, 2006 3587

Cl solution was added, the organic layer was isolated, and the

aqueous layer was extracted with diethyl ether (2 × 25 mL). The

combined organic layers were dried over MgSO₄, filtered, and

rotavapped to give the product in quantitative yield (10.49 g) as a

light yellow oil.

Me₂C(3-methyl-C₅H₃)(C₂₉H₃₆)Li₂. The dianion was prepared

by treating a solution of Me₂C(3-methyl-C₅H₃)(C₂₉H₃₆)H₂(10.49

g, 20.7 mmol) in 75 mL of diethyl ether with 27.0 mL of

n-butyllithium solution (43.2 mmol, 1.6 M in hexanes) at 0 °C.

After stirring for 17 h, the precipitate was isolated by filtration

and in vacuo drying to provide the dianion as a yellow powder:

8.707 g (81.1%).

Me₂C(3-methyl-C₅H₃)(C₂₉H₃₆)ZrCl₂(7). Me₂C(3-methyl-C₅H₃)-

(C₂₉H₃₆)Li₂(3.34 g, 6.44 mmol) and 1.50 g of ZrCl₄(6.44 mmol)

were combined in a swivel frit apparatus. Then 50 mL of petroleum

ether was condensed in at -78 °C. This was allowed to warm

slowly to room temperature before solvent removal after 18 h of

stirring. Methylene chloride (20 mL) was condensed in and removed

in order to quench unreacted ligand. Then the orange solid was

extracted from a cellulose extraction thimble overnight with 150

mL of diethyl ether. The volume of the filtrate was reduced to 25

mL, and the precipitate was collected at 0 °C. A total of 1.051 g

(24.5%) of 7 as an orange-pink powder was obtained following in

vacuo drying. MS (LC-MS): m/z 666.6 (M⁺). ¹H NMR (C₆D₆): δ

1.20, 1.31, 1.31, 1.31, 1.31, 1.32, 1.50, 1.53 (s, 24H, Oct-CH₃),

1.65 (m, 8H, Oct-CH₂), 1.93 (s, 3H, Cp-CH₃), 2.03, 2.06 (s, 6H,

(CH₃)₂C-Oct-Cp), 5.21, 5.50, 5.89 (t, ³J_HH) 2.6, 2.9, 2.6 Hz, 3H,

Cp-H), 7.56, 7.70, 8.29, 8.30 (s, 4H, Oct-H). Anal. Calcd for C₃₈H₄₈-

Zr₁Cl₂: C, 68.44; H, 7.25. Found: C, 62.90; H, 6.97.

Me₂C(3-cyclohexyl-C₅H₃)(C₂₉H₃₆)H₂. An n-butyllithium solu-

tion (11.0 mL, 17.6 mmol, 1.6 M in hexanes) was syringed into a

solution of octamethyloctahydrodibenzofluorene (6.603 g, 17.08

mmol) in 60 mL of tetrahydrofuran. After stirring for 50 min,

3-cyclohexyl-6,6-dimethylfulvene (3.216 g, 17.08 mmol) was

injected via syringe into the red slurry. After stirring for 18 h, 60

mL of a saturated NH₄Cl solution was added, the organic layer

was isolated, and the aqueous layer was extracted with diethyl ether

(2 × 30 mL). The combined organic layers were dried over MgSO₄,

filtered, and rotavapped to give the product in quantitative yield

(9.82 g) as a light yellow wax.

7.729 mmol), evacuated, and backfilled with argon before 60 mL

of tetrahydrofuran and 5.2 mL of n-butyllithium solution (8.3 mmol,

1.6 M in hexanes) were syringed in. The orange solution was stirred

for 4 h before a solution of 3-(diphenylmethyl)-6,6-dimethylfulvene

(2.105 g, 7.728 mmol) in 25 mL of tetrahydrofuran was syringed

in. Following an additional 30 h, the stirred reaction was quenched

by slow addition of 60 mL of aqueous NH₄Cl. The organic layer

was isolated and the aqueous layer extracted with diethyl ether (2

× 30 mL). The combined organic layers were dried over MgSO₄,

filtered, and rotavapped to give the product in quantitative yield

(5.093 g) as a light yellow oil.

Me₂C(3-(diphenylmethyl)-C₅H₃)(C₂₉H₃₆)Li₂. A round-bottom

flask containing 5.093 g (7.728 mmol) of Me₂C(3-(diphenylmethyl)-

C₅H₃)(C₂₉H₃₆)H₂was attached to a swivel frit and evacuated before

50 mL of diethyl ether was condensed in. At 0 °C, 10.4 mL of

n-butyllithium solution (16.6 mmol, 1.6 M in hexanes) was syringed

in over 2 min. After stirring for 26 h at room temperature, the

solvent was removed by vacuum transfer and 50 mL of petroleum

ether was condensed in. The dilithio salt was isolated by filtration

and in vacuo drying in quantitative yield (5.185 g) as an orange

powder.

Me₂C(3-(diphenylmethyl)-C₅H₃)(C₂₉H₃₆)ZrCl₂(9). In the glove-

box, 2.879 g of Me₂C(3-(diphenylmethyl)-C₅H₃)(C₂₉H₃₆)Li₂(4.29

mmol) was combined with ZrCl₄(1.00 g, 4.29 mmol) in a 100 mL

round-bottom flask. This was attached to a swivel frit, and 50 mL

of petroleum ether was condensed in by vacuum transfer at -78

°C. The vessel was allowed to warm slowly, and after 17 h of

stirring, solvent was removed. Then 20 mL of methylene chloride

was condensed in; the solution was warmed and stirred before

solvent removal. The solid was extracted overnight in a cellulose

extraction thimble with 150 mL of diethyl ether. The filtrate volume

was reduced to 40 mL, and the precipitated product was collected

on a swivel frit and dried in vacuo: 0.793 g of 9 (22.6%). MS

1

(LC-MS): m/z 818.8 (M⁺). H NMR (C₆D₆): δ 1.08, 1.20, 1.23,

1.29, 1.30, 1.30, 1.51, 1.58 (s, 24H, Oct-CH₃), 1.63 (m, 8H, Oct-

CH₂), 1.90, 2.04 (s, 6H, (CH₃)₂C-Oct-Cp), 5.31, 5.55, 5.79 (t, ³J_HH

) 2.9, 3.0, 2.6 Hz, 3H, Cp-H), 5.86 (s, 1H, CHPh₂), 6.91-7.15

(m, 10H, phenyl-H), 7.46, 6.65, 8.29, 8.31 (s, 4H, Oct-H). Anal.

Calcd for C₅₀H₅₆Zr₁Cl₂: C, 73.32; H, 6.89. Found: C, 65.09; H,

6.86.

Me₂C(3-cyclohexyl-C₅H₃)(C₂₉H₃₆)Li₂. The dianion was prepared

by treating a solution of Me₂C(3-cyclohexyl-C₅H₃)(C₂₉H₃₆)H₂(9.82

g, 17.1 mmol) in 75 mL of diethyl ether with 22.0 mL of

n-butyllithium solution (35.2 mmol, 1.6 M in hexanes) at 0 °C.

After stirring for 18 h, the precipitate was isolated by filtration

and in vacuo drying to provide the dianion as an orange powder:

6.446 g (64.3%).

Me₂C(3-methyl-C₅H₃)(C₁₃H₈)H₂. A 500 mL round-bottom flask

was charged with fluorene (55.32 g, 332.8 mmol). This was

equipped with a 180° needle valve, evacuated, and backfilled with

argon before 240 mL of diethyl ether was added via syringe. Then

210.0 mL of n-butyllithium in hexanes (1.6 M, 336.0 mmol) was

syringed in at room temperature over 20 min. After shaking and

stirring the obtained yellow slurry for 1 h, 3,6,6-trimethylfulvene

(40.00 g, 332.8 mmol) was syringed in over 25 min, providing a

clear, red solution. After stirring for 17 h, the vessel was cooled to

0 °C, and 60 mL of aqueous NH₄Cl solution was added. The slurry

was filtered and the aqueous layer removed. The obtained solid

was extracted from a cellulose extraction thimble with 500 mL of

diethyl ether/hexanes for 2 days. The first crop was obtained by

filtration of the cooled filtrate: 28.45 g following in vacuo drying

(29.9%). The second and third crops were obtained by filtration of

the chilled (-78 °C) filtrate and massed 11.86 and 1.08 g,

respectively (43.4% for all three crops). MS (GC-MS): m/z 286.3

(M⁺). Anal. Calcd for C₂₂H₂₂: C, 92.26; H, 7.74. Found: C, 90.99,

90.92; H, 7.21, 7.21.

Me₂C(3-cyclohexyl-C₅H₃)(C₂₉H₃₆)ZrCl₂(8). Me₂C(3-cyclo-

hexyl-C₅H₃)(C₂₉H₃₆)Li₂(2.518 g, 4.29 mmol) and 1.00 g of ZrCl₄

(4.29 mmol) were combined in a swivel frit apparatus. Then 30

mL of petroleum ether was condensed in at -78 °C. This was

allowed to warm slowly to room temperature before solvent removal

after 18 h of stirring. Methylene chloride (40 mL) was condensed

in and removed in order to quench unreacted ligand. Then the

orange solid was extracted from a cellulose extraction thimble

overnight with 150 mL of diethyl ether. The volume of the filtrate

was reduced to 50 mL, and the precipitate was collected at 0 °C.

A total of 1.846 g (58.5%) of 8 as an orange powder was obtained

following in vacuo drying. MS (LC-MS): m/z 734.8 (M⁺). ¹H NMR

(C₆D₆): δ 0.99-1.25 (m, 10H, cyclohexyl-H), 1.28, 1.30, 1.30,

1.31, 1.32, 1.33, 1.51, 1.51 (s, 24H, Oct-CH₃), 1.63 (m, 8H, Oct-

CH₂), 2.08, 2.09 (s, 6H, (CH₃)₂C-Oct-Cp), 2.61 (m, 1H, 1-cyclo-

2,6,6-Trimethyl-4-(C(methyl)₂(9-fluorenyl))fulvene. Me₂C(3-

methyl-C₅H₃)(C₁₃H₈)H₂(11.86 g, 41.41 mmol) was combined with

200 mL of acetone (2720 mmol) and 15.0 mL of pyrrolidine (180

mmol). After stirring for 30 min, a homogeneous solution was

obtained and stirring was ceased. The product slowly crystallized

and after 30 days, the yellow crystals were collected by filtration.

These were combined with 100 mL of methanol, brought to a boil

for 4 h, and stirred overnight as the vessel cooled. Collection by

3

hexyl-H), 5.44, 5.60, 6.07 (t, J_HH) 2.9, 2.9, 2.6 Hz, 3H, Cp-H),

7.65, 7.71, 8.29, 8.30 (s, 4H, Oct-H). Anal. Calcd for C₄₃H₅₆Zr₁-

Cl₂: C, 70.26; H, 7.68. Found: C, 67.53; H, 7.76.

Me₂C(3-(diphenylmethyl)-C₅H₃)(C₂₉H₃₆)H₂. A 250 mL flask

was charged with octamethyloctahydrodibenzofluorene (2.988 g,

3588 Organometallics, Vol. 25, No. 15, 2006

Miller and Bercaw

suction filtration, rinsing with 25 mL of methanol, and in vacuo

drying afforded 8.15 g of the desired product (60.3%). MS (GC-

H). ¹³C NMR (CDCl₃): δ 28.30, 37.05, 37.35, 40.25 (adamantyl-

C), 119.47, 130.47 (fulvene-CH₁), 135.81, 167.38 (fulvene-CH₀).

Anal. Calcd for C₁₅H₁₈: C, 90.85; H, 9.15. Found: C, 90.20, 90.22;

H, 8.39, 8.50.

1

MS): m/z 326.5 (M⁺). H NMR (CDCl₃): δ 1.02, 1.02 (s, 6H,

C(CH₃)₂Flu), 2.16, 2.25, 2.53 (s, 9H, 2,6,6-CH₃-fulvene), 4.13 (s,

1H, 9-H-Flu), 5.96, 6.54 (s, 2H, 3,5-H-fulvene), 7.15, 7.31 (t, ³J_HH

2-Adamantylcyclopentadiene. 6,6-Adamantylidenefulvene (6.00

g, 30.3 mmol) was dissolved in 30 mL of tetrahydrofuran and this

solution added over 30 min to a stirred slurry of LiAlH₄(1.40 g,

0.0369 mol) at 0 °C. After 5 h of stirring at room temperature, the

reaction was cooled to 0 °C and quenched by slow addition of 20

mL of saturated NH₄Cl solution. Then 300 mL of H₂O, 25 mL of

concentrated HCl, and 50 mL of diethyl ether were added, the

organic layer was isolated, and the aqueous layer was extracted

with additional diethyl ether (3 × 50 mL). The combined organic

layers were dried over MgSO₄, filtered, and rotavapped to give the

product, 2-adamantylcyclopentadiene, in quantitative yield as a light

yellow oil. MS (GC-MS): m/z 200.3 (M⁺).

3

) 7.4, 7.4 Hz, 4H, Flu-H), 7.28, 7.70 (s, J_HH) 7.3, 7.7 Hz, 4H,

Flu-H). ¹³C NMR (CDCl₃): δ 19.04, 22.46, 24.53, 24.53, 25.18

(CH₃), 39.38 (CH₀), 55.66 (9-Flu-CH₁), 114.78, 130.54 (fulvene-

CH₁), 119.30, 119.30, 126.07, 126.07, 126.52, 126.52, 126.92,

126.93 (Flu-CH₁), 132.75, 133.98, 140.86, 151.75 (fulvene-CH₀),

142.04, 142.04, 145.54, 145.54 (Flu-CH₀). Anal. Calcd for

C₂₅H₂₆: C, 91.97; H, 8.03. Found: C, 90.83, 91.12; H, 7.33, 7.26.

Me₂C(3-tert-butyl-4-methyl-C₅H₂)(C₁₃H₈)H₂. A 250 mL round-

bottom flask was charged with 5.087 g of 2,6,6-trimethyl-4-

(C(methyl)₂(9-fluorenyl))fulvene (15.58 mmol). This was evacuated

before 100 mL of diethyl ether was condensed in. Then 75.0 mL

of methyllithium in diethyl ether (1.4 M, 105 mmol) was added by

syringe, giving an orange, homogeneous solution after 1 h. After

one month of stirring, a small amount of orange precipitate was

found. The amount slowly increased, and after 47 days total, the

orange slurry was cooled to 0 °C and slowly quenched with 60

mL of H₂O. The organic layer was isolated, and the aqueous layer

was extracted with diethyl ether (2 × 25 mL). The combined

organic layers were dried over MgSO₄, filtered, and rotavapped to

provide the product in quantitative yield (5.34 g) as a light yellow

oil, which slowly began to crystallize.

3-(2-Adamantyl)-6,6-dimethylfulvene. To 2-adamantylcyclo-

pentadiene (6.06 g, 30.3 mmol) was added 50 mL of methanol, 50

mL of ethanol, 20 mL of tetrahydrofuran, 36 mL of acetone (0.49

mol), and 0.5 mL of pyrrolidine (0.006 mol). After stirring for 48

h, 5 mL of acetic acid was injected, followed by 200 mL of H₂O

and 200 mL of diethyl ether. The organic layer was isolated and

the aqueous layer extracted with diethyl ether (3 × 40 mL). The

combined organic layers were extracted with H₂O (3 × 25 mL)

and with 10% aqueous NaOH (3 × 25 mL), dried over MgSO₄,

filtered, and rotavapped. The obtained yellow solid was further

purified by overnight Soxhlet extraction with 150 mL of methanol.

The precipitate in the filtrate was isolated by filtration at 0 °C and

in vacuo drying: 4.54 g (62.5%) of 3-(2-adamantyl)-6,6-dimeth-

ylfulvene, as a yellow powder. Anal. Calcd for C₁₈H₂₄: C, 89.94;

H, 10.06. Found: C, 82.23, 82.23; H, 8.78, 8.82.

Me₂C(3-tert-butyl-4-Me-C₅H₂)(C₁₃H₈)Li₂. A round-bottom flask

containing 5.34 g (15.6 mmol) of Me₂C(3-tert-butyl-4-methyl-

C₅H₂)(C₁₃H₈)H₂was attached to a swivel frit and evacuated before

75 mL of diethyl ether was condensed in. At 0 °C, 22.0 mL of

n-butyllithium in hexanes (1.6 M, 32.5 mmol) was syringed in over

1 min. After stirring for 15 h at room temperature, the orange

precipitate was collected and dried in vacuo: 5.37 g (97.3%).

Me₂C(3-(2-adamantyl)-C₅H₃)(C₁₃H₈)H₂. A 10.5 mL portion of

an n-butyllithium solution (1.6 M in hexanes, 0.0168 mol) was

syringed into a solution of fluorene (2.77 g, 0.0166 mol) in 60 mL

of tetrahydrofuran. After stirring for 5 h, a solution of 3-(2-

adamantyl)-6,6-dimethylfulvene (4.00 g, 0.0166 mol) in 40 mL of

tetrahydrofuran was injected over 2 min. After stirring for 20 h,

60 mL of a saturated NH₄Cl solution was added, the organic layer

isolated, and the aqueous layer extracted with diethyl ether (2 ×

25 mL). The combined organic layers were dried over MgSO₄,

filtered, and rotavapped to give the product in quantitative yield as

a yellow oil.

Me₂C(3-tert-butyl-4-Me-C₅H₂)(C₁₃H₈)ZrCl₂(10). In the glove-

box, 2.28 g of Me₂C(3-tert-butyl-4-methyl-C₅H₂)(C₁₃H₈)Li₂(6.44

mmol) was combined with ZrCl₄(1.50 g, 6.44 mmol) in a 100 mL

round-bottom flask. This was equipped with a 180° needle valve,

and 50 mL of petroleum ether was condensed in by vacuum transfer

at -78 °C. The vessel was allowed to warm slowly and after 23 h

of stirring, solvent was removed. Methylene chloride (30 mL) was

condensed in; the solution was warmed and stirred before solvent

removal; 30 mL of diethyl ether was condensed in; the slurry was

warmed and stirred before solvent removal. The obtained solid was

extracted overnight in a cellulose extraction thimble with 150 mL

of methylene chloride. The obtained solution was filtered through

a frit, all solvent was removed, and 50 mL of diethyl ether was

condensed in. The pink solid was broken up, stirred, collected on

the frit, and dried in vacuo to afford the product 10: 1.60 g (49.5%).

MS (LC-MS): m/z 502.3 (M⁺). ¹H NMR (CD₂Cl₂): δ 1.16 (s, 9H,

C(CH₃)₃), 2.07 (s, 3H, Cp-CH₃), 2.30, 2.32 (s, 6H, C(CH₃)₂), 5.43,

Me₂C(3-(2-adamantyl)-C₅H₃)(C₁₃H₈)Li₂. The dianion was pre-

pared by treating a solution of Me₂C(3-(2-adamantyl)C₅H₄)(C₁₃H₈)-

H₂(6.77 g, 16.6 mmol) in 75 mL of diethyl ether with 22.0 mL of

n-butyllithium solution (1.6 M in hexanes, 0.0352 mol) at 0 °C.

After stirring for 21 h, the solvent was removed by vacuum transfer

and 50 mL of petroleum ether was condensed in. The dilithio salt

was isolated by filtration and in vacuo drying in quantitative yield

as an orange powder.

3

5.52 (d, J_HH) 3.7, 3.7 Hz, 3H, Cp-H), 7.22, 7.23, 7.50, 7.53 (t,

Me₂C(3-(2-adamantyl)-C₅H₃)(C₁₃H₈)ZrCl₂(11). A 2.00 g

sample of Me₂C(3-(2-adamantyl)C₅H₄)(C₁₃H₈)Li₂(0.00478 mol)

and 1.114 g of ZrCl₄(0.00478 mol) were combined in a swivel frit

apparatus. Then 40 mL of petroleum ether was condensed in at

-78 °C. This was allowed to warm slowly to room temperature

before solvent removal after 14 h of stirring. Methylene chloride

(40 mL) was condensed in and removed in order to quench

unreacted ligand. Then the orange solid was extracted in the swivel

frit with 50 mL of refluxing diethyl ether. Two crops were obtained

for a total of 1.502 g (55.5%) of 11 as an orange powder following

collection at 0 °C and in vacuo drying. MS (LC-MS): m/z 566.5

(M⁺). ¹H NMR (C₆D₆): δ 1.84, 1.86 (s, 6H, CH₃), 1.36-2.04 (m,

14H, adamantyl-H), 3.32 (s, 1H, 2-adamantyl-H), 5.44, 5.48, 6.18

(m, 3H, Cp-H), 6.95, 7.03, 7.29, 7.34 (t, 4H, Flu-H), 7.41, 7.49,

7.84, 7.84 (d, 4H, Flu-H). ¹³C NMR (CD₂Cl₂): δ 28.58, 28.65 (C-

(CH₃)₂), 27.90, 27.93, 31.98, 32.41, 32.62, 32.66, 37.84, 38.50,

38.66, 43.83 (adamantyl-C), 102.56, 103.02, 116.65 (Cp-CH₁),

³J_HH) 7.3, 7.3, 8.4, 8.4 Hz, 4H, Flu-H), 7.79, 7.82, 8.10, 8.12 (d,

³J_HH) 9.2, 9.2, 8.4, 8.4 Hz, 4H, Flu-H). ¹³C NMR (CD₂Cl₂): δ

16.08, 28.24, 28.75 (CH₃), 29.17 (C(CH₃)₃), 33.52, 39.85 (CH₀),

78.40, 110.49, 121.76, 123.65, 123.79, 128.00, 140.84 (Cp and Flu

CH₀), 102.93, 108.11 (Cp-CH₁), 123.42, 123.64, 124.45, 124.55,

124.68, 124.96, 128.33, 128.80 (Flu-CH₁). Anal. Calcd for C₂₆H₂₈-

Zr₁Cl₂: C, 62.13; H, 5.61. Found: C, 60.88, 60.89; H, 4.90, 4.94.

6,6-Adamantylidene-fulvene. (Synthesis modified from ref 32.)

Pyrrolidine (10.0 mL, 0.116 mol) was syringed into a solution of

2-adamantanone (25.00 g, 0.1664 mol) and cyclopentadiene (30.0

mL, 0.364 mol) in 250 mL of methanol. The reaction was stirred

for 92 h before the yellow precipitate was collected by suction

filtration, rinsed with a small volume of methanol, and dried in

vacuo. Then 25.71 g (77.9%) of 6,6-adamantylidenefulvene was

1

isolated. MS (GC-MS): m/z 198.3 (M⁺). H NMR (CDCl₃): δ

1.93-2.08, 3.29 (m, 14H, adamantyl-H), 6.52, 6.60 (m, 4H, fulvene-

Mechanism of Isotactic Polypropylene Formation

Organometallics, Vol. 25, No. 15, 2006 3589

123.41, 123.67, 124.61, 124.67, 124.76, 124.83, 128.81, 128.81

(Flu-CH₁), 139.93 (9-Flu-C), CH₀not determined. Anal. Calcd for

C₃₁H₃₂Zr₁Cl₂: C, 65.70; H, 5.69. Found: C, 63.46, 61.93; H, 5.57,

5.42.

Fluorenyllithium. A Schlenk tube was charged with fluorene

(31.81 g, 191.4 mmol), evacuated, backfilled with argon, and

charged with 150 mL of toluene. n-Butyllithium solution (120.0

mL, 192 mmol, 1.6 M in hexanes) was syringed in, and the reaction

was stirred for 103 h before the yellow slurry was cannulated onto

a frit and the precipitate collected and dried in vacuo: 28.95 g

(87.9%).

9-(ClMe₂Si)-fluorene. A swivel frit was charged with fluore-

nyllithium (7.00 g, 40.66 mmol) and 80 mL of petroleum ether.

The vessel was cooled to -78 °C, and SiMe₂Cl₂(10.0 mL, 82.44

mmol) was syringed in. The cold bath remained as the vessel was

allowed to warm very slowly. After 48 h, the reaction was filtered

and the solvent was removed from the filtrate to provide the product

as an off-white powder: 8.10 g (77.0%). MS (GC-MS): m/z 258.3

(M⁺). Competing formation of Me₂Si(9-fluorenyl)₂(MS (GC-MS)

m/z 388.4 (M⁺)), as reported by ref 39, occurs to about 10% (GC),

but apparently does not affect the synthesis of 12.

and 200 mL of diethyl ether were added and the organic layer was

isolated. The aqueous layer was extracted with diethyl ether (5 ×

50 mL), and the combined organic layers were dried over MgSO₄,

filtered, and rotavapped. Under high vacuum, unreacted ketone was

removed by Kugelrohr distillation at 40 °C. Crude (95%) 6,6-

diisopropylfulvene was obtained from the next Kugelrohr fraction

1

at 60-80 °C; the orange oil massed 31.30 g (17.7%). H NMR

(CDCl₃): δ 1.28 (d, ³J_HH) 7.0 Hz, 12H, CH₃), 3.11, (septet, ³J_HH

) 7.0 Hz, 2H, i-Pr-H), 6.46, 6.64 (m, 4H, fulvene-H).

2,4-Dimethyl-3-pentylcyclopentadiene. A 500 mL Schlenk flask

was charged with 4.00 g of LiAlH₄(105 mmol) and 150 mL of

diethyl ether. An attached addition funnel was charged with 9.38 g

of 6,6-diisopropylfulvene (57.8 mmol) and 50 mL of diethyl ether.

The fulvene solution was added at 0 °C over 15 min and rinsed

down with an additional 50 mL of diethyl ether. The cold bath

was removed and the reaction was stirred for 48 h before it was

cooled to 0 °C, and 50 mL of H₂O was added dropwise via a

metered addition funnel. The ether layer was isolated, and the

remaining white solid was extracted with diethyl ether (3 × 50

mL). The combined organic layers were filtered and rotavapped to

yield 8.13 g of product (85.6%) as a light yellow oil.

2-Adamantylcyclopentadienyllithium. Adamantylcyclopenta-

diene (10.78 g, 53.81 mmol) was added to a swivel frit, and 75

mL of diethyl ether was added by vacuum transfer. At 0 °C,

n-butyllithium solution (34.0 mL, 54.4 mmol, 1.6 M in hexanes)

was syringed in over 5 min. After stirring for 15 h at room

temperature, the white solid was collected on the frit and dried in

vacuo. The product was isolated in quantitative yield (11.10 g).

Me₂Si(3-(2-adamantyl)-C₅H₃)(C₁₃H₈)Li₂. A swivel frit was

charged with 9-(ClMe₂Si)-fluorene (5.500 g, 21.25 mmol) and

adamantylcyclopentadienyllithium (4.383 g, 21.25 mmol). Tetrahy-

drofuran (40 mL) was condensed in and the reaction stirred at room

temperature for 19 h. Solvent was removed, and 50 mL of diethyl

ether was condensed in. Filtration and washing removed LiCl. To

the filtrate was added n-butyllithium solution (28.0 mL, 44.8 mmol,

1.6 M in hexanes) over 5 min at room temperature. Solvent was

removed after stirring for 20 h. Petroleum ether (50 mL) was

condensed in, and the material was broken up by stirring and

shaking. Solvent was decanted, and the red solid was dried in vacuo

to provide the product in quantitative yield (9.23 g).

3-(2,4-Dimethyl-3-pentyl)-6,6-dimethylfulvene. To 2,4-dimeth-

yl-3-pentylcyclopentadiene (8.13 g, 49.5 mmol) were added 50 mL

of methanol, 30 mL of acetone (409 mmol), and 10.0 mL of

pyrrolidine (120 mmol). After stirring for 10 days, 15 mL of acetic

acid was injected, followed by 300 mL of H₂O and 100 mL of

diethyl ether. The organic layer was isolated, and the aqueous layer

extracted with diethyl ether (3 × 50 mL). The combined organic

layers were extracted with H₂O (3 × 30 mL) and with 10% aqueous

NaOH (3 × 30 mL), dried over MgSO₄, filtered, and rotavapped.

This material was subjected to Kugelrohr distillation under high

vacuum. Then 3 mL was distilled at room temperature and

discarded. Product was obtained from the next fraction, obtained

at 80 °C: 9.39 g (92.9%) of an orange oil.

Me₂C(3-(2,4-dimethyl-3-pentyl)-C₅H₃)(C₁₃H₈)H₂. A 250 mL

flask was charged with fluorenyllithium diethyl ether adduct (6.026

g, 24.47 mmol). Diethyl ether (60 mL) was condensed in, and

3-(2,4-dimethyl-3-pentyl)-6,6-dimethylfulvene (5.00 g, 24.5 mmol)

was injected. After 6 days, 60 mL of aqueous NH₄Cl solution was

slowly added and the organic layer was isolated. The aqueous layer

was extracted with diethyl ether (2 × 25 mL), and the combined

organic layers were dried over MgSO₄, filtered, and rotavapped to

provide the product in quantitative yield (9.07 g) as a light yellow

oil.

Me₂Si(3-(2-adamantyl)-C₅H₃)(C₁₃H₈)ZrCl₂(12). A 100 mL

flask was charged with Me₂Si(3-(2-adamantyl)-C₅H₃)(C₁₃H₈)Li₂

(3.73 g, 8.58 mmol) and ZrCl₄(2.00 g, 8.58 mmol) and equipped

with a 180° needle valve. Petroleum ether (50 mL) was condensed

in at -78 °C, and the cold bath was removed. This was allowed to

warm slowly with stirring, and solvent was removed after 19 h.

The solid was placed in a cellulose extraction thimble and was

extracted overnight with 150 mL of methylene chloride in a Soxhlet

extractor. The filtrate was filtered on a swivel frit, and the volume

was reduced to 30 mL. The yellow-orange precipitate was collected

on the frit and dried in vacuo: 0.707 g (14.1%). MS (LC-MS):

Me₂C(3-(2,4-dimethyl-3-pentyl)-C₅H₃)(C₁₃H₈)Li₂. The dianion

was prepared by treating a solution of Me₂C(3-(2,4-dimethyl-3-

pentyl)-C₅H₃)(C₁₃H₈)H₂(9.07 g, 24.5 mmol) in 50 mL of diethyl

ether with 33.0 mL of n-butyllithium solution (52.8 mmol, 1.6 M

in hexanes) at 0 °C. After stirring for 25 h, the solvent was removed

by vacuum transfer, and 75 mL of petroleum ether was condensed

in. A red-orange powder was isolated in quantitative yield (9.36 g)

by filtration and in vacuo drying.

1

m/z 582.7 (M⁺). H NMR (CD₂Cl₂): δ 1.11, 1.13 (s, 6H, CH₃),

1.48-1.99 (m, 14H, adamantyl-H), 3.03 (s, 1H, 2-adamantyl-H),

5.49, 5.75, 6.34 (m, 3H, Cp-H), 7.27, 7.27, 7.58, 7.60 (t, 4H, Flu-

H), 7.51, 7.59, 8.11, 8.11 (d, 4H, Flu-H). ¹³C NMR (CD₂Cl₂, 35

°C): δ -1.12, -1.05 (Si-CH₃), 27.92, 32.49, 32.66, 37.86, 38.55,

38.71, 44.23 (adamantyl-C), 111.31, 111.86, 120.23 (Cp-CH₁),

123.52, 124.03, 124.20, 124.82, 126.23, 126.31, 128.54, 128.62

(Flu-CH₁), CH₀not determined. Anal. Calcd for C₃₀H₃₂Si₁Zr₁Cl₂:

C, 61.83; H, 5.53. Found: C, 58.63; H, 4.94.

6,6-Diisopropylfulvene. A 500 mL round-bottom flask was

charged with 150 mL of 2,4-dimethyl-3-pentanone (1060 mmol),

60.0 mL of cyclopentadiene (728 mmol), and 44.00 g of sodium

methoxide (815 mmol). The deep red slurry was placed on a

mechanical shaker for 12 days. Then 300 mL of aqueous NH₄Cl

Me₂C(3-(2,4-dimethyl-3-pentyl)-C₅H₃)(C₁₃H₈)ZrCl₂(13). Me₂C-

(3-(2,4-dimethyl-3-pentyl)-C₅H₃)(C₁₃H₈)Li₂(4.103 g, 10.73 mmol)

and 2.500 g of ZrCl₄(10.73 mmol) were combined in a 100 mL

round-bottom flask, equipped with a 180° needle valve. Then 40

mL of petroleum ether was condensed in at -78 °C. This was

allowed to warm slowly to room temperature before solvent removal

after 26 h of stirring. Then 30 mL of methylene chloride was

condensed in and removed in order to quench unreacted ligand.

The orange solid was extracted in a cellulose extraction thimble

for 48 h with 150 mL of diethyl ether. The filtrate volume was

reduced to 75 mL, and 0.491 g (8.6%) of 13 as an orange powder

was obtained following collection at 0 °C and in vacuo drying.

1

MS (LC-MS): m/z 530.7 (M⁺). H NMR (C₆D₆): δ 0.58, 0.90,

3

0.93, 1.01 (d, J_HH) 7.0, 7.0, 7.0, 7.0 Hz, 12H, CH₃), 1.84, 1.86

(39) Okuda, J.; Schattenmann, F. J.; Wocadlo, S.; Massa, W. Organo-

metallics 1995, 14, 789-795.

3

(s, 6H, C(CH₃)₂), 2.25, 2.25 (m, 2H, CH(CH₃)₂), 2.57 (t, J_HH

)

3590 Organometallics, Vol. 25, No. 15, 2006

Miller and Bercaw

3

2.4 Hz, 1H, 3-H-pentyl), 5.36, 5.53, 6.21 (t, J_HH) 2.6, 3.3, 2.6

Hz, 4H, Flu-H). ¹³C NMR (CD₂Cl₂): δ 26.69, 27.47, 27.77, 28.99,

33.25, 33.76, 34.84, 39.36, 40.18 (adamantyl-C), 28.28, 29.11,

(CH₃), 41.68 (2-C-adamantyl), 42.07 (2-CH₃-adamantyl), 102.54,

105.28, 120.28 (Cp-CH₁), 123.81, 124.17, 124.36, 124.46, 124.51,

125.25, 127.94, 129.10 (benzo-CH₁), 112.14, 120.68, 123.68,

125.72, 127.82, 129.55, 139.05, 145.94 (CH₀). Anal. Calcd for

C₃₂H₃₄Zr₁Cl₂: C, 66.18; H, 5.90. Found: C, 57.60; H, 5.23.

3

Hz, 3H, Cp-H), 6.98, 7.01, 7.29, 7.33 (t, J_HH) 7.0, 7.0, 7.0, 7.0

Hz, 4H, Flu-H), 7.47, 7.47, 7.82. 7.82 (d, ³J_HH) 8.8, 8.8, 8.4, 8.4

Hz, 4H, Flu-H). Anal. Calcd for C₂₈H₃₂Zr₁Cl₂: C, 63.37; H, 6.08.

Found: C, 56.61; H, 5.56.

6,6-Adamantylidenefulvene. (Synthesis modified from ref 32.)

2-Adamantanone (45.00 g, 299.6 mmol), methanol (200 mL),

cyclopentadiene (60.0 mL, 728 mmol), and pyrrolidine (20.0 mL,

240 mmol) were added to a 1 L round-bottom flask. After stirring

for 77 h, the yellow precipitate was collected by suction filtration

and washed with 50 mL of methanol. After in vacuo drying, 49.56

g of 6,6-adamantylidenefulvene was obtained (83.4%). MS (GC-

MS): m/z 198.3 (M⁺).

(2-Phenyl-2-adamantyl)cyclopentadiene. A 300 mL flask was

charged with 6,6-adamantylidenefulvene (10.00 g, 50.43 mmol),

and 75 mL of diethyl ether was condensed in. At -78 °C 60.0 mL

of phenyllithium solution (108 mmol, 1.8 M in cyclohexane/diethyl

ether) was injected and the cold bath removed. After 89 h, the vessel

was cooled to 0 °C, and 60 mL of aqueous NH₄Cl solution was

slowly added. The ether layer was isolated, and the aqueous layer

was extracted with diethyl ether (3 × 50 mL). The combined

organic layers were dried over MgSO₄, filtered, and rotavapped to

give the product in quantitative yield (13.94 g) as a tan-colored

solid.

3-(2-Methyl-2-adamantyl)-6,6-dimethylfulvene. A 500 mL

flask was charged with 6,6-adamantylidenefulvene (18.00 g, 90.77

mmol), equipped with a 180° needle valve, and charged with 120

mL of diethyl ether. At 0 °C, methyllithium lithium bromide

solution (150.0 mL, 225 mmol, 1.5 M in diethyl ether) was syringed

in over 10 min. Dimethoxyethane (10 mL) was syringed in and

the reaction was stirred at room temperature for 8 days when 60

mL of aqueous NH₄Cl solution was slowly added at 0 °C. The

organic layer was isolated, and the aqueous layer was extracted

with diethyl ether (3 × 25 mL). The combined organic layers were

dried over MgSO₄, filtered, rotavapped, and dried in vacuo to

provide 19.46 g of (2-methyl-2-adamantyl)cyclopentadiene as a light

yellow oil (theoretical yield). To this were added 30 mL of acetone

(409 mmol), 100 mL of methanol, and 10 mL of pyrrolidine (120

mmol). After stirring for 96 h, the yellow precipitate was collected

by filtration, rinsed with 50 mL of methanol, and dried in vacuo to

provide the product: 20.36 g (88.2%). MS (GC-MS): m/z 254.5

3-(2-Phenyl-2-adamantyl)-6,6-dimethylfulvene. To (2-phenyl-

2-adamantyl)cyclopentadiene (13.94 g, 50.4 mmol) were added 100

mL of methanol, 50 mL of acetone (680 mmol), and 10.0 mL of

pyrrolidine (120 mmol). After stirring for 4 days, 100 mL of

methanol was added, and the yellow precipitate was collected by

suction filtration. The product was washed with 100 mL of methanol

and dried in vacuo: 14.76 g (92.5%). MS (GC-MS): m/z 316.5

(M⁺). ¹H NMR (CDCl₃): δ 1.63-2.22, 2.94 (m, 14H, adamantyl-

H), 2.06, 2.10, (s, 6H, CH₃), 6.19 (s, 1H, 2-H-fulvene), 6.38, 7.48

(d, ³J_HH) 5.5, 4.8 Hz, 2H, 4,5-H-fulvene), 7.05 (t, ³J_HH) 7.0 Hz,

1H, 4-H-phenyl), 7.24, (t, ³J_HH) 8.0 Hz, 2H, 3,5-H-phenyl), 7.40

(d, ³J_HH) 7.7 Hz, 2H, 2,6-H-phenyl). Anal. Calcd for C₂₄H₂₈: C,

91.08; H, 8.92. Found: C, 90.66; H, 8.56.

(M⁺). H NMR (CDCl₃): δ 1.22, 2.17, 2.17 (s, 9H, CH₃), 1.56-

1

2.04 (m, 14H, adamantyl-H), 6.17, 6.52, 6.54 (m, 3H, fulvene-H).

¹³C NMR (CDCl₃): δ 22.88, 22.97, 27.92, 28.03 (CH₁), 27.92,

35.17, 35.17 (CH₃), 32.98, 32.98, 34.59, 34.59, 39.08 (CH₂), 41.47

(CH₀), 113.36, 121.04, 130.38 (fulvene-CH₁), 142.41, 146.12,

156.16 (fulvene-CH₀). Anal. Calcd for C₁₉H₂₆: C, 89.70; H, 10.30.

Found: C, 89.57; H, 10.04.

Me₂C(3-(2-phenyl-2-adamantyl)-C₅H₃)(C₁₃H₈)H₂. A 300 mL

flask was charged with fluorenyllithium diethyl ether adduct (3.113

g, 12.64 mmol) and 3-(2-phenyl-2-adamantyl)-6,6-dimethylfulvene

(4.000 g, 12.64 mmol). Diethyl ether (60 mL) was condensed in,

and the reaction was stirred for 42 h before 60 mL of aqueous

NH₄Cl solution was slowly added and the organic layer was

isolated. The aqueous layer was extracted with diethyl ether (2 ×

25 mL), and the combined organic layers were dried over MgSO₄,

filtered, and rotavapped to provide the product in quantitative yield

(6.10 g) as a light yellow oil.

Me₂C(3-(2-methyl-2-adamantyl)-C₅H₃)(C₁₃H₈)Li₂. A 250 mL

flask was charged with 3-(2-methyl-2-adamantyl)-6,6-dimethylful-

vene (8.000 g, 31.45 mmol) and fluorenyllithium diethyl ether

adduct (7.744 g, 31.45 mmol). Diethyl ether (75 mL) was condensed

in, and the reaction was stirred at room temperature for 4 days

before 60 mL of aqueous NH₄Cl was slowly added at 0 °C. The

organic layer was isolated, and the aqueous layer was extracted

with diethyl ether (2 × 25 mL). The combined organic layers were

dried over MgSO₄, filtered, and rotavapped to provide Me₂C(3-

(2-methyl-2-adamantyl)-C₅H₃)(C₁₃H₈)H₂in theoretical yield (13.23

g). The flask was attached to a swivel frit and charged with 50 mL

of diethyl ether before n-butyllithium solution (42.0 mL, 67.2 mmol,

1.6 M in hexanes) was syringed in over 4 min at 0 °C. After 23 h,

solvent was removed and 75 mL of petroleum ether was added by

vacuum transfer. The red solid was broken up, stirred, collected

on the frit, and dried in vacuo: 15.85 g (18.26 g theoretical yield

for the bis diethyl ether adduct).

Me₂C(3-(2-phenyl-2-adamantyl)-C₅H₃)(C₁₃H₈)Li₂. The dianion

was prepared by treating a solution of Me₂C(3-(2-phenyl-2-

adamantyl)-C₅H₃)(C₁₃H₈)H₂(6.10 g, 12.6 mmol) in 50 mL of

diethyl ether with 17.0 mL of n-butyllithium solution (27.2 mmol,

1.6 M in hexanes) at 0 °C. After stirring for 21 h, the solvent was

removed and 50 mL of petroleum ether was condensed in. The

product was isolated in quantitative yield (6.25 g) after filtration

and in vacuo drying.

Me₂C(3-(2-phenyl-2-adamantyl)-C₅H₃)(C₁₃H₈)ZrCl₂(15). A

100 mL flask was charged with Me₂C(3-(2-phenyl-2-adamantyl)-

C₅H₃)(C₁₃H₈)Li₂(2.653 g, 5.364 mmol) and ZrCl₄(1.250, 5.364

mmol) and equipped with a 180° needle valve. Petroleum ether

(50 mL) was condensed in at -78 °C and the cold bath removed.

After 22 h, solvent was removed from the pink slurry. The solid

was extracted in a cellulose extraction thimble with 150 mL of

methylene chloride overnight. The filtrate was attached to a swivel

frit and filtered. The solvent was removed, and 30 mL of diethyl

ether was condensed in. The yellow-orange solid was collected and

dried in vacuo: 1.993 g (57.8%). MS (LC-MS): m/z 642.6 (M⁺).

¹H NMR (C₆D₆): δ 1.36-3.13 (m, 14H, adamantyl-H), 1.77, 1.83

Me₂C(3-(2-methyl-2-adamantyl)-C₅H₃)(C₁₃H₈)ZrCl₂(14). A

100 mL flask was charged with Me₂C(3-(2-methyl-2-adamantyl)-

C₅H₃)(C₁₃H₈)Li₂(4.640 g, 10.73 mmol) and ZrCl₄(2.500, 10.73

mmol) and equipped with a 180° needle valve. Petroleum ether

(50 mL) was condensed in at -78 °C and the cold bath removed.

After 70 h, solvent was removed from the pink slurry. The solid

was extracted in a cellulose extraction thimble with 150 mL of

methylene chloride overnight. The filtrate was attached to a swivel

frit, filtered, and condensed to 40 mL. The precipitate was collected

and dried in vacuo: 3.246 g (52.1%). MS (LC-MS): m/z 580.5

(M⁺). ¹H NMR (C₆D₆): δ 1.32-2.62 (m, 14H, adamantyl-H), 1.73,

3

(s, 6H, C(CH₃)₂), 5.49, 5.66, 5.98 (t, J_HH) 3.3, 2.9, 2.9 Hz, 3H,

3

Cp-H), 6.92, 6.99, 7.25, 7.30 (t, J_HH) 7.0, 8.0, 7.7, 8.4 Hz, 4H,

3

Flu-H), 7.03, 7.18 (t, J_HH) 7.3, 7.0 Hz, 3H, phenyl-H), 7.38 (d,

3

1.85, 1.89 (s, 9H, CH₃), 5.73, 5.83, 6.14 (t, J_HH) 3.3, 2.9, 3.3

³J_HH) 8.8 Hz, 2H, phenyl-H), 7.40, 7.66, 7.73, 7.76 (d, J_HH

)

3

Hz, 3H, Cp-H), 6.98, 7.02, 7.28, 7.34 (t, J_HH) 7.0, 7.7, 7.0, 7.7

9.6, 7.3, 8.4, 8.0 Hz, 4H, Flu-H). ¹³C NMR (CD₂Cl₂): δ 28.17,

28.94 (C-(CH₃)₂), 26.43, 27.86, 32.27, 33.11, 34.25, 34.53, 37.85,

Hz, 4H, Flu-H), 7.46, 7.54, 7.76, 8.87 (d, ³J_HH) 9.2, 8.8, 8.4, 8.4

Mechanism of Isotactic Polypropylene Formation

Organometallics, Vol. 25, No. 15, 2006 3591

39.06, 40.03, 50.04 (adamantyl-C), 102.62, 105.10, 121.50 (Cp-

CH₁), 123.63, 123.89, 124.26, 124.50, 124.56, 125.04, 125.48,

127.42, 127.42, 127.87, 128.00, 128.84, 129.62 (phenyl- and Flu-

CH₁), 143.61, 144.27 (ipso-C and 9-Flu-C), CH₀not determined.

Anal. Calcd for C₃₇H₃₆Zr₁Cl₂: C, 69.13; H, 5.64. Found: C, 67.61;

H, 5.39.

acid was injected, followed by 200 mL of H₂O and 200 mL of

diethyl ether. The organic layer was isolated and the aqueous layer

extracted with diethyl ether (2 × 50 mL). The combined organic

layers were extracted with H₂O (4 × 25 mL), dried over MgSO₄,

filtered, and rotavapped. This material was subjected to Kugelrohr

distillation under high vacuum. Then 10.8 g was distilled at 60 °C

and discarded. Product was obtained from the next fraction, obtained

at 120-140 °C: 10.54 g (80.0%) of a yellow oil.

Me₂C(C₁₃H₈)(3-(2-(CH₂SiMe₃)-2-adamantyl)-C₅H₃)H₂. A 250

mL flask was charged with fluorenyllithium diethyl ether adduct

(5.482 g, 22.26 mmol) and 3-(2-(CH₂SiMe₃)-2-adamantyl)-6,6-

dimethylfulvene (7.270 g, 22.26 mmol). Diethyl ether (100 mL)

was condensed in, and the reaction was stirred for 16 h before 60

mL of aqueous NH₄Cl solution was slowly added and the organic

layer was isolated. The aqueous layer was extracted with diethyl

ether (3 × 30 mL), and the combined organic layers were dried

over MgSO₄, filtered, and rotavapped to provide the product in

quantitative yield (10.97 g) as a waxy solid.

Me₂C(3-(2-methyl-2-adamantyl)-C₅H₃)(C₂₉H₃₆)Li₂. A 250 mL

flask was charged with octamethyloctahydrodibenzofluorene (6.079

g, 15.72 mmol), equipped with a 180° needle valve, and charged

with 75 mL of diethyl ether before n-butyllithium solution (10.5

mL, 16.8 mmol, 1.6 M in hexanes) was syringed into the white

slurry over 10 min. After 20 h, solvent was removed from the

yellow slurry, and 3-(2-methyl-2-adamantyl)-6,6-dimethylfulvene

(4.000 g, 15.72 mmol) was added. Diethyl ether (75 mL) was

condensed in, and the reaction, which became homogeneous upon

warming, was stirred for 13 days before 60 mL of water was slowly

syringed in at 0 °C. The organic layer was isolated, and the aqueous

layer was extracted with diethyl ether (2 × 25 mL). The combined

organic layers were dried over MgSO₄, filtered, rotavapped, and

dried in vacuo to provide Me₂C(3-(2-methyl-2-adamantyl)-C₅H₃)-

(C₂₉H₃₆)H₂in theoretical yield (11.21 g). This flask was attached

to a swivel frit, evacuated, and charged with diethyl ether (75 mL)

by vacuum transfer. At room temperature, n-butyllithium solution

(21.0 mL, 16.8 mmol, 1.6 M in hexanes) was syringed in over 8

min. After 15 h, solvent was removed, and 50 mL of petroleum

ether was condensed in. The product slowly precipitated over 2 h

and was collected and dried in vacuo: 3.525 g (34.3%).

Me₂C(C₁₃H₈)(3-(2-(CH₂SiMe₃)-2-adamantyl)-C₅H₃)Li₂. The

dianion was prepared by treating a solution of Me₂C(C₁₃H₈)(3-(2-

(CH₂SiMe₃)-2-adamantyl)-C₅H₃)H₂(10.97 g, 22.26 mmol) in 50

mL of diethyl ether with 30.0 mL of n-butyllithium solution (48.0

mmol, 1.6 M in hexanes) at 0 °C. After stirring for 23 h, the solvent

was removed and 75 mL of petroleum ether was condensed in.

The product was isolated by decanting the solvent and drying the

residue in vacuo: 8.013 g (71.3%).

Me₂C(3-(2-(CH₂SiMe₃)-2-adamantyl)-C₅H₃)(C₁₃H₈)ZrCl₂(17).

A 100 mL flask was charged with Me₂C(C₁₃H₈)(3-(2-(CH₂SiMe₃)-

2-adamantyl)-C₅H₃)Li₂(4.331 g, 8.582 mmol) and ZrCl₄(2.000,

8.583 mmol) and equipped with a 180° needle valve. Petroleum

ether (50 mL) was condensed in at -78 °C and the cold bath

removed. After 22 h, solvent was removed. This was attached to a

swivel frit, 70 mL of toluene was condensed in, and the solution

was filtered. The filtrate was condensed to 10 mL, and the

precipitate was collected and dried in vacuo: 0.543 g (9.7%). MS

(LC-MS): m/z 652.6 (M⁺). ¹H NMR (C₆D₆): δ -0.03 (s, 9H, Si-

(CH₃)₃), 1.36, 1.37 (s, 2H, CH-₂Si(CH₃)₃), 1.36-2.15 (m, 14H,

Me₂C(3-(2-methyl-2-adamantyl)-C₅H₃)(C₂₉H₃₆)ZrCl₂(16). A

swivel frit apparatus was charged with Me₂C(3-(2-methyl-2-

adamantyl)-C₅H₃)(C₂₉H₃₆)Li₂(3.525 g, 5.399 mmol) and ZrCl₄

(1.258 g, 5.398 mmol). Petroleum ether (60 mL) was condensed

in at -78 °C, and the cold bath remained as the reaction was

allowed to warm very slowly. After 20 h, the reaction was filtered

and all solvent was removed from the filtrate. A red powder was

obtained following lyophilization from 30 mL of benzene. Hex-

amethyldisiloxane (30 mL) was condensed in, and the red slurry

was stirred for 4 h before the product was collected by filtration

and dried in vacuo: 0.614 g (14.2%). MS (LC-MS): m/z 800.9

(M⁺). ¹H NMR (CD₂Cl₂): δ 1.20, 1.22, 1.34, 1.36, 1.36, 1.38, 1.39,

1.39 (s, 24H, Oct-CH₃), 1.32, 1.70 (m, 14H, adamantyl-H), 1.48

(s, 3H, 2-CH₃-adamantyl), 1.72 (m, 8H, Oct-CH₂), 2.29, 2.31 (s,

adamantyl-H), 2.33, 2.35 (s, 6H, CH₃), 5.66, 5.88, 6.30 (t, ³J_HH

)

2.9, 3.3, 2.6 Hz, 3H, Cp-H), 7.23, 7.25, 7.52, 7.52 (m, 4H, Flu-H),

7.82, 7.89, 8.07, 8.10 (d, ³J_HH) 8.8, 8.8, 8.4, 8.4 Hz, 4H, Flu-H).

¹³C NMR (CD₂Cl₂): δ 1.51 (Si-CH₃), 27.44, 27.64 (CH₃), 28.22,

29.12, 31.24, 34.19, 34.27, 34.30, 34.52, 36.14, 37.76, 39.62

(adamantyl-C), 103.07, 104.52, 120.20 (Cp-CH₁), 123.68, 124.14,

124.36, 124.69, 124.77, 125.33, 128.19, 129.01 (Flu-CH₁), CH₀

not determined. Anal. Calcd for C₃₅H₄₂Si₁Zr₁Cl₂: C, 64.38; H, 6.48.

Found: C, 57.53; H, 5.51.

3

6H, (CH₃)₂C), 5.66 (m, 2H, Cp-H), 6.09 (t, J_HH) 2.6 Hz, 1H,

Cp-H), 7.60, 7.63, 7.98, 8.02 (s, 4H, Oct-H). ¹³C NMR (CD₂Cl₂):

δ 26.08, 27.55, 27.55, 27.55, 27.78, 27.78, 28.63, 28.86 (Oct-CH₃),

31.85, 32.37 (C(CH₃)₂), 31.81, 32.27, 33.24, 33.54 (adamantyl-

CH₁), 33.49, 33.86, 34.13, 34.34, 34.45, 34.92, 35.09, 35.23, 35.29,

38.95, 38.99, 39.26, 39.57 (adamantyl and OctCH₂and CH₀), 41.68

(2-C-adamantyl), 42.46 (2-CH₃-2-adamantyl), 74.50 (C(CH₃)₂),

101.17, 102.23, 116.91 (Cp-CH₁), 120.51, 120.91, 121.70, 121.84

(benzo-CH₁), 139.44 (9-fluorenyl-C), 109.97, 119.60, 122.35,

122.42, 143.91, 145.32, 145.39, 145.74, 146.84, 147.48 (Cp and

Oct CH₀). Anal. Calcd for C₄₈H₆₂Zr₁Cl₂: C, 71.96; H, 7.80.

Found: C, 71.62; H, 7.37.

Propylene Polymerization Procedures. CAUTION: All po-

lymerization procedures should be performed behind a blast

shield. All polymerization reactions were prepared in nitrogen-filled

gloveboxes. Methylaluminoxane (MAO) was purchased as a toluene

solution from Albemarle Corporation and used as the dry powder

obtained by in vacuo removal of all volatiles. Toluene was dried

over sodium and distilled. Propylene from Scott Specialty Gases

(>99.5%) was used following drying through a Matheson 6410

drying system equipped with an OXYSORB column. Polymeriza-

tions were conducted in a 3 oz. Lab Crest glass reaction vessel

(Andrews Glass Co.) and were stirred with a magnetic stir bar.

Monomer was condensed into the vessel over several minutes at 0

°C. The vessel was then equilibrated at either 0 or 20 °C with an

ice or water bath for 10 min. A given reaction commenced upon

injection of a toluene solution of the metallocene into the vessel

with a 2.5 mL Hamilton gastight syringe rated to 200 psi.

Temperature maintenance was monitored by an affixed pressure

gauge. Polymerization reactions were vented and quenched with a

small volume of methanol/concentrated HCl (12:1), and the

polymers were separated from hydrolyzed aluminoxanes by pre-

cipitation from methanol, followed by filtration. Residual amounts

(2-(CH₂SiMe₃)-2-adamantyl)cyclopentadiene. A 250 mL flask

was charged with 6,6-adamantylidenefulvene (8.000 g, 40.34 mmol)

and LiCH₂SiMe₃(8.000 g, 84.96 mmol). Then 100 mL of diethyl

ether was condensed in, and the reaction was stirred at room

temperature for 16 h when the vessel was cooled to 0 °C and 60

mL of aqueous NH₄Cl solution was slowly added. The organic layer

was isolated, and the aqueous layer was extracted with diethyl ether

(2 × 30 mL). The combined organic layers were dried over MgSO₄,

filtered, rotavapped, and dried in vacuo to provide the product in

quantitative yield (11.56 g).

3-(2-(CH₂SiMe₃)-2-adamantyl)-6,6-dimethylfulvene. To (2-

(CH₂SiMe₃)-2-adamantyl)cyclopentadiene (11.56 g, 40.3 mmol)

were added 100 mL of acetone (1360 mmol) and 10.0 mL of

pyrrolidine (120 mmol). After stirring for 4 days, 10 mL of acetic

3592 Organometallics, Vol. 25, No. 15, 2006

Miller and Bercaw

of toluene and methanol were removed from the obtained polymers

by in vacuo drying. Polymerization reactions are further described

in Table 5.

Representative Polymerization Procedures. Entry 36. A 100

mL Lab Crest glass pressure reactor was charged with MAO (0.100

g, 1.72 × 10^-3mol [Al]). Propylene (30 mL) was condensed in at

0 °C. A solution of 14 (0.001 g, 1.7 × 10^-6mol) in toluene (2.0

mL) was injected and the reaction stirred in a 0 °C ice/water bath

for 10 min. The reaction was vented and quenched with dilute HCl/

methanol.

Entry 37. A 100 mL Lab Crest glass pressure reactor was

charged with MAO (0.100 g, 1.72 × 10^-3mol [Al]). Propylene

(30 mL) was condensed in at 0 °C. A solution of 14 (0.001 g, 1.7

× 10^-6mol) in toluene (2.0 mL) was injected and the reaction

stirred in a 20 °C water bath for 10 min. The reaction was vented

and quenched with dilute HCl/methanol.

Entry 38. A 100 mL Lab Crest glass pressure reactor was

charged with MAO (0.200 g, 3.44 × 10^-3mol [Al]). Propylene

(60 mL) was condensed in at 0 °C. A solution of 14 (0.002 g, 3.4

× 10^-6mol) in toluene (2.0 mL) was injected and the reaction

stirred in a 0 °C ice/water bath for 60 min. The reaction was vented

and quenched with dilute HCl/methanol.

Entry 39. A 100 mL Lab Crest glass pressure reactor was

charged with MAO (0.200 g, 3.44 × 10^-3mol [Al]). Propylene

(55 mL) was condensed in at 0 °C. A solution of 14 (0.002 g, 3.4

× 10^-6mol) in toluene (2.0 mL) was injected and the reaction

stirred in a 20 °C water bath for 10 min. The reaction was vented

and quenched with dilute HCl/methanol.

Entry 40. A 100 mL Lab Crest glass pressure reactor was

charged with MAO (0.200 g, 3.44 × 10^-3mol [Al]). Propylene

(55 mL) was condensed in at 0 °C. A solution of 14 (0.002 g, 3.4

× 10^-6mol) in toluene (2.0 mL) was injected and the reaction

stirred in a 0 °C ice/water bath for 10 min. The reaction was vented

and quenched with dilute HCl/methanol.

Entry 45. A 100 mL Lab Crest glass pressure reactor was

charged with MAO (0.145 g, 2.50 × 10^-3mol [Al]). Propylene

(30 mL) was condensed in at 0 °C. A solution of 16 (0.002 g, 2.5

× 10^-6mol) in toluene (2.0 mL) was injected and the reaction

stirred in a 0 °C ice/water bath for 20 min. The reaction was vented

and quenched with dilute HCl/methanol.

Entry 46. A 100 mL Lab Crest glass pressure reactor was

charged with MAO (0.145 g, 2.50 × 10^-3mol [Al]). Propylene

(30 mL) was condensed in at 0 °C. A solution of 16 (0.002 g, 2.5

× 10^-6mol) in toluene (2.0 mL) was injected and the reaction

stirred in a 20 °C water bath for 20 min. The reaction was vented

and quenched with dilute HCl/methanol.

Polymer Characterization. Polymer melting temperatures were

determined by differential scanning calorimetry (Perkin-Elmer DSC

7). The second scan (from 50 to 200 °C at 10 °C/min) was used

when subsequent scans were similar. The polymer pentad distribu-

tions were determined by integration of the nine resolved peaks in

the methyl region (19-22 ppm) of the ¹³C NMR spectra obtained.⁴⁰

Spectra were acquired at 124 °C with tetrachloroethane-d₂as

solvent. A 90 degree pulse was employed with broadband decou-

pling. A delay time of 3 s and a minimum of 1000 scans were

used.

Acknowledgment. This work has been supported by US-

DOE Office of Basic Energy Sciences (Grant No. DE-FG03-

85ER13431), Exxon Chemicals America, and the National

Science Foundation (Grant No. CHE-0131180). A DOD Na-

tional Defense Science and Engineering Fellowship awarded

to S.M. is also acknowledged. The authors thank Dr. Michael

Day and Mr. Lawrence Henling for obtaining the X-ray crystal

structure, as well as Dr. Terry Burkhardt at ExxonMobil and

Dr. Andreas Ernst at BP-Amoco for assistance in obtaining

molecular weight data.

Entry 41. A 100 mL Lab Crest glass pressure reactor was

charged with MAO (0.200 g, 3.44 × 10^-3mol [Al]) and 28.0 mL

of toluene. Propylene (3 mL) was condensed in. A solution of 14

(0.002 g, 3.4 × 10^-6mol) in toluene (2.0 mL) was injected and

the reaction stirred in a 0 °C ice/water bath for 180 min. The

reaction was vented and quenched with dilute HCl/methanol.

Entry 42. A 100 mL Lab Crest glass pressure reactor was

charged with MAO (0.200 g, 3.44 × 10^-3mol [Al]) and 28.0 mL

of toluene. Propylene (3 mL) was condensed in. A solution of 14

(0.002 g, 3.4 × 10^-6mol) in toluene (2.0 mL) was injected and

the reaction stirred in a 20 °C water bath for 90 min. The reaction

was vented and quenched with dilute HCl/methanol.

Supporting Information Available: Derivation of the alternat-

ing models and complete statistical analysis for polypropylenes

made by 10/MAO (entries 23-29). Details of the X-ray structure

determination for 10, including selected bond distances and angles.

This material is available free of charge via the Internet at

http://pubs.acs.org.

OM050841K

(40) (a) Busico, V.; Cipullo, R.; Corradini, P.; Landriani, L.; Vacatello,

M.; Segre, A. L. Macromolecules 1995, 28, 1887-1892. (b) Reference 19a.

Article Doi

Mechanism of isotactic polypropylene formation with C 1-symmetric metallocene catalysts

DOI: 10.1021/om050841k

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om050841k

Full text of DOI:10.1021/om050841k

Products guided by the article

R&D Labs maybe for 473174-30-0

Relevant to this article

Hot Product