Highly Stereoregular Syndiotactic Polypropylene Formation with Metallocene Catalysts via Influence of Distal Ligand Substituants

Article abstract of DOI:10.1021/om030333f

Highly stereoregular syndiotactic polypropylene is obtained with the catalyst systems Ph₂C-(Oct)(C₅H₄)ZrCl ₂/MAO (8/MAO) (Oct = octamethyloctahydrodibenzofluorenyl; MAO = methylaluminoxane) and Me₂C(Oct)(C₅H₄)ZrCl ₂/MAO (12/MAO). The syndiotactic polypropylenes obtained are largely devoid of stereoerrors by ¹³C NMR analysis ([r] > 98%), and melting temperatures as high as 153 or 154°C (from 8 and 12, respectively) are found for the thermally quenched polymers (without annealing). A related hafnium catalytic system, Ph₂C-(Tet)(C₅H ₄)HfCl₂/MAO (11/MAO) (Tet = tetramethyltetrahydrobenzofluorenyl), was found to be the most syndioselective of the hafnocenes tested (T_m = 141°C). The metallocene dichloride precatalysts represent the first examples of transition metal complexes containing the Oct or Tet ligands. Reported are the solid state crystal structures of 8, the diprotio ligand precursor of 8 (Ph₂C(OctH) (C₅H₅)), and the zirconium analogue of 11, Ph ₂C(Tet)(C₅H₄)ZrCl₂ (10). Distal ligand influences are thus demonstrated to have a dramatic effect on polymer stereochemistry.

Full text of DOI:10.1021/om030333f

Organometallics 2004, 23, 1777-1789
1777
High ly Ster eor egu la r Syn d iota ctic P olyp r op ylen e
F or m a tion w ith Meta llocen e Ca ta lysts via In flu en ce of
Dista l Liga n d Su bstitu en ts
Stephen A. Miller† and J ohn E. Bercaw*
Arnold and Mabel Beckman Laboratories of Chemical Synthesis,
California Institute of Technology, Pasadena, California 91125
Received May 6, 2003
Highly stereoregular syndiotactic polypropylene is obtained with the catalyst systems Ph2C-
(Oct)(C5H4)ZrCl2/MAO (8/MAO) (Oct ) octamethyloctahydrodibenzofluorenyl; MAO ) meth-
ylaluminoxane) and Me2C(Oct)(C5H4)ZrCl2/MAO (12/MAO). The syndiotactic polypropylenes
obtained are largely devoid of stereoerrors by 13C NMR analysis ([r] > 98%), and melting
temperatures as high as 153 or 154 °C (from 8 and 12, respectively) are found for the
thermally quenched polymers (without annealing). A related hafnium catalytic system, Ph2C-
(Tet)(C5H4)HfCl2/MAO (11/MAO) (Tet ) tetramethyltetrahydrobenzofluorenyl), was found
to be the most syndioselective of the hafnocenes tested (Tm ) 141 °C). The metallocene
dichloride precatalysts represent the first examples of transition metal complexes containing
the Oct or Tet ligands. Reported are the solid state crystal structures of 8, the diprotio ligand
precursor of 8 (Ph2C(OctH)(C5H5)), and the zirconium analogue of 11, Ph2C(Tet)(C5H4)ZrCl2
(10). Distal ligand influences are thus demonstrated to have a dramatic effect on polymer
stereochemistry.
Sch em e 1
In tr od u ction
The first highly efficient syndiotactic polypropylene
polymerization system was reported by Ewen et al. in
1988.1 The original design employed ansa fluorenyl-
cyclopentadienyl precatalysts of the type Me2C(C13H8)-
(C5H4)MCl2 (M ) Zr, Hf; C13H8 ) fluorenyl) activated
by methylaluminoxane (MAO). Perfectly syndiotactic
polypropylene arises from a regularly alternating inser-
tion mechanism that employs sequentially opposite
enantiofaces of the propylene monomer.2 The Me2C-
(C13H8)(C5H4)MCl2/MAO systemsas well as other syn-
dioselective catalyst systems that incorporate propylene
in a 1,2 fashion3ssuffers from two kinds of stereoerrors
that decrease the syndiotacticity of the produced poly-
mer: enantiofacial misinsertions, giving rise to a double
m, i.e., a [...rrmmrr...] stereoerror sequence, and site
epimerizations (inversion at the stereogenic metal cen-
ter by polymeryl chain swinging), giving rise to an
isolated m, i.e., a [...rrmrrr...] stereoerror sequence
(Scheme 1).4
After Ewen’s initial discovery, several metallocene-
based systems have been devised that exceed the
original with respect to syndioselectivity, activity, mo-
lecular weight, and polymer melting temperature (Tm)
(Scheme 2).3,4 Following reports of the original singly
bridged metallocenes 1 and 2,1 Razavi et al. developed
bridge-modified analogues 3 and 4.5 Alt, Zenk, et al. and
Ewen et al. have prepared several fluorenyl-substituted
metallocenes, including 5.6,7 We have reported doubly
[SiMe2]-bridged, syndioselective metallocenes; examples
† Current address: Department of Chemistry, Texas A&M Univer-
sity, College Station, TX 77843-3255.
(1) (a) Ewen, J . A.; J ones, R. L.; Razavi, A.; Ferrara, J . D. J . Am.
Chem. Soc. 1988, 110, 6255-6256. (b) Ewen, J . A.; Elder, M. J .; J ones,
R. L.; Haspeslagh, L.; Atwood, J . L.; Bott, S. G.; Robinson, K.
Makromol. Chem., Macromol. Symp. 1991, 48/ 49, 253-295.
(2) (a) Price, F. P. J . Chem. Phys. 1962, 36, 209. (b) Frisch, H. L.;
Mallows, C. L.; Bovey, F. A. J . Chem. Phys. 1966, 45, 1565. (c) Bovey,
F. A. Acct. Chem. Res. 1968, 1, 175.
(3) For reviews on Ziegler-Natta polymerization see: (a) Brintz-
inger, H.-H.; Fischer, D.; Mu¨lhaupt, R.; Rieger, B.; Waymouth, R. M.
Angew. Chem., Int. Ed. Engl. 1995, 34, 1143. (b) Britovsek, G. J . P.;
Gibson, V. C.; Wass, D. F. Angew. Chem., Int. Ed. 1999, 38, 428. (c)
Coates G. W. Chem. Rev. 2000, 100, 1223. (d) Resconi, L.; Cavallo, L.;
Fait, A.; Piemontesi, F. Chem. Rev. 2000, 100, 1253.
(5) (a) Razavi, A.; Atwood, J . L. J . Organomet. Chem. 1993, 459,
117-123. (b) Razavi, A.; Peters, L.; Nafpliotis, L. J . Mol. Catal. A
Chem. 1997, 115, 129-154.
(6) (a) Alt, H. G.; Zenk, R.; Milius, W. J . Organomet. Chem. 1996,
514, 257-270. (b) Alt, H. G.; Zenk, R. J . Organomet. Chem. 1996, 518,
7-15. (c) Alt, H. G.; Zenk, R. J . Organomet. Chem. 1996, 526, 295-
301.
(4) (a) Veghini, D.; Henling, L.; Burkhardt, T.; Bercaw, J . E. J . Am.
Chem. Soc. 1999, 121, 564-573. (b) Alt, H. G.; Ko¨ppl, A. Chem. Rev.
2000, 100, 1205-1221.
10.1021/om030333f CCC: $27.50 © 2004 American Chemical Society
Publication on Web 03/18/2004

1778 Organometallics, Vol. 23, No. 8, 2004
Miller and Bercaw
Sch em e 2
Ta ble 1. MAO-Coca ta lyzed P r op ylen e
fluorenyl ligand of 1-4) of the doubly bridged metal-
locenes 6 and 7 apparently disfavors the analogous
transition state for misinsertion, as these provide
polymers containing fewer enantiofacial misinsertions
([rmmr] typically <1%).4a Thus, the initial goal of this
study was to prepare a fluorenyl-based metallocene with
similar steric properties such that enantiofacial misin-
sertions would be minimized.
P olym er iza tion Resu lts w ith 1-7
metallocene
Tp (°C)
[r] (%)
[rrrr] (%)
Tm (°C)
ref
1
2
3
4
5
6
7
50
50
50
50
60
20
20
96.0
74.0
97.5
90.3
99.1
97.1
99.4
138
118
139
101
141
151
151
5
5
5
5
7
8
8
98.7 [rr]
93.4
97.5
Metallocene 5 bears added steric bulk in the form of
two tert-butyl groups at the 2 and 7 positions of the
fluorenyl ligand. However, only a modest increase in
syndioselectivity is observed over that of the parent
system.6,7 A simple steric analysis suggests that steric
bulk placed at the 3 and 6 positions of the fluorenyl
ligand might better approximate the steric situation
present in the doubly bridged metallocenes. Therefore,
the sterically expansive substituted fluorenes 1,1,4,4,7,7,
10,10-octamethyl-1,2,3,4,7,8,9,10-octahydrodibenzo[b,h]-
fluorene (OctH) and 1,1,4,4-tetramethyl-1,2,3,4-tetrahy-
drobenzo[b]fluorene (TetH) were prepared according to
Scheme 3.10
This improved synthesis, the Friedel-Crafts double
cycloalkylation11 of fluorene, which proceeds in high
isolated yields (up to 95%), allows the large-scale (ca.
100 g) preparation of sterically loaded fluorene OctH.
All starting materials are inexpensive and commercially
available with the exception that 2,5-dichloro-2,5-di-
methylhexane must be prepared in one step from the
corresponding, inexpensive diol (see Experimental Sec-
tion). TetH is prepared by the analogous single cy-
cloalkylation and must be separated from residual
fluorene and OctH, which forms competitively. None-
theless, acceptable yields (43%) of pure TetH can readily
be prepared in multigram quantities (10-20 g).
are given by 6 and 7.8 Table 1 presents representative
data regarding polymers made by MAO-cocatalyzed
polymerizations. The doubly bridged metallocenes pro-
duce the most stereoregular and highest melting syn-
diotactic polypropylene. For a given ligand system, the
hafnium analogues generally provide polymers of lower
syndiotacticity. In this report we describe new singly
[CR2]-bridged zirconocene catalysts having cyclopenta-
dienyl linked to sterically expansive substituted fluo-
renyl ligands, i.e., those with mono- and bis(tetrameth-
yltetrahydrobenzo) substituents on the flourenyl ligand.
These new catalyst systems afford syndiotactic polypro-
pylenes with very high [rrrr] pentad contents and the
highest melt transition temperatures reported to date
for thermally quenched polymers (no annealing).
Resu lts a n d Discu ssion
Syn th esis of Zir con ocen e a n d Ha fn ocen e Di-
ch or id es. According to the current understanding of
propagative stereocontrol,9 enantiofacial misinsertions
occur in parent fluorenyl systems whenever the growing
polymer chain is directed toward the fluorenyl ligand
during the transition state for insertion. The more three-
dimensional nature of the lower cyclopentadienyl ligand
having the isopropyl substituents (vis-a`-vis the flatter
(10) The syntheses of OctH and TetH have been reported: (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. For the first reports of Oct-
containing metallocenes, see ref 19 and: (c) Miller, S. A.; Bercaw, J .
E. U.S. Patent 6,469,188, 2002 (application date: 1/20/00). (d) Miller,
S. A.; Bercaw, J . E. U.S. Patent 6,693,153, 2004. (e) Kawai, K.;
Yamashita, M.; Tohi, Y.; Kawahara, N.; Michiue, K.; Kaneyoshi, H.;
Mori, R. PCT Int. Appl. WO 01/27124, 2001 (application date 10/5/
00). Additionally, two of our Oct-containing metallocenes have been
cited in recent reviews (8 in ref 3c and 12 in ref 3d).
(11) For Friedel-Crafts cycloalkylation reactions employing 2,5-
dichloro-2,5-dimethylhexane, see: (a) Bruson, H. A.; Kroeger, J . W. J .
Am. Chem. Soc. 1940, 62, 36-44. (b) Roberts, R. M.; Khalaf, A. A.
Friedel-Crafts Alkylation Chemistry. A Century of Discovery; Dek-
ker: New York, 1984. (c) Olah, G. A.; Krishnamurti, R.; Prakash, G.
K. S. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I.,
Eds.; Pergamon Press: Oxford, 1991; Vol. 3, pp 292-339.
(7) (a) Ewen, J . A. Macromol. Symp. 1995, 89, 181-196. (b)
Shiomura, T.; Kohno, M.; Inoue, N.; Asanuma, T.; Sugimoto, R.;
Iwatani, T.; Uchida, O.; Kimura, S.; Harima, S.; Zenkoh, H.; Tanaka,
E. Macromol. Symp. 1996, 101, 289.
(8) Herzog, T. A.; Zubris, D. L.; Bercaw, J . E. J . Am. Chem. Soc.
1996, 118, 11988-11989.
(9) (a) Longo, P.; Grassi, A.; Pellecchia, C.; Zambelli, A. Macromol-
ecules 1987, 20, 1015. (b) Cavallo, L.; Guerra, G.; Vacatello, M.;
Corradini, P. Macromolecules 1991, 24, 1784. (c) Corradini, P.; Guerra,
G.; Cavallo, L.; Moscardi, G.; Vacatello, M. In Ziegler Catalysts, Recent
Scientific Innovations and Technological Improvements; Fink, G.,
Mu¨lhaupt, R., Brintzinger, H. H., Eds.; Springer: Berlin, 1995; pp
237-249. (d) Guerra, G.; Corradini, P.; Cavallo, L.; Vacatello, M.
Macromol. Symp. 1995, 89, 307. (e) Gilchrist, J . H.; Bercaw, J . E. J .
Am. Chem. Soc. 1996, 118, 12021-12028. (f) J aniak, C. In Metal-
locenes: Synthesis, Reactivity, Applications; Togni, A., Halterman, R.
L., Eds.; Wiley-VCH: Weinheim, 1998; pp 547-623.

Highly Stereoregular Syndiotactic Polypropylene
Organometallics, Vol. 23, No. 8, 2004 1779
Sch em e 3
Sch em e 4
Zirconocene and hafnocene dichlorides incorporating
the Oct and Tet ligands are prepared according to
Scheme 4. A single crystal of the intermediate diprotio
ligand Ph2C(OctH)(C5H5) was subjected to X-ray dif-
fraction analysis (Figure 1),12 revealing the presence of
four different molecules residing in the asymmetric unit.
These varied because of alternate positioning of the sp3
carbon in the cyclopentadienyl ring and because of
different conformers possible for the 2,5-dimethylhex-
ane-2,5-diyl substituents (Table 2). For example, Figure
1 shows the structure with the proximal (same side of
the fluorene ring as 9-H) methyl groups near the 3 and
6 positions of fluorene occupying axial (left) and equato-
rial (right) positions.
Str u ctu r a l Ch a r a cter iza tion of th e Zir con ocen e
a n d Ha fn ocen e Dich lor id es. Using the procedures
similar to that shown in Scheme 4, zirconocene and
hafnocene dichlorides having the Oct or Tet ligands
were prepared (Scheme 5). Compounds 8, 9, and 12
display time-averaged Cs symmetry in solution (1H
F igu r e 1. X-ray crystal structure of Ph2C(OctH)(C5H5).
Thermal ellipsoids are shown with 50% probability.
NMR, 25 °C), suggesting rapid interconversion of the
two chairlike conformers possible for the tetrameth-
yltetrahydrobenzo substituents. Additionally, com-
pounds 10 and 11sa racemic mixture of enantiomerss
behave as single, C1-symmetric species in solution. As
seen for the parent fluorenyl complexes 3 and 4,5
rotation around the C-phenyl bonds of 8-11 is slow
on the NMR time scale.
(12) The diprotio ligand Ph2C(OctH)(C5H5) (crystallized from etha-
nol) is monoclinic C2/c , a ) 39.813(15) Å, b ) 12.631(6) Å, c )
29.671(15) Å, â ) 98.34(4)°, V ) 14763(12) Å3, Z ) 16, T ) 85 K.
Crystallographic data have been deposited at the CCDC, 12 Union
Road, Cambridge CB2 1EZ, UK, and copies can be obtained on request,
free of charge, by quoting the publication citation and the deposition
number 105607.

1780 Organometallics, Vol. 23, No. 8, 2004
Miller and Bercaw
Ta ble 2. Descr ip tion of F ou r Solid Sta te
Str u ctu r es for P h 2C(OctH)(C5H5)
position of left
position of right
position of Cp proximal methyl proximal methyl
molecule
sp3 carbon
group
group
A (Figure 1)
3
3
4
4
axial
axial
axial
equatorial
equatorial
axial
equatorial
axial
B
C
D
Sch em e 5
F igu r e 2. X-ray crystal structure of 8. Thermal ellipsoids
are shown with 50% probability.
The solid state crystal structures of zirconocene
dichlorides 8 (Figure 2) and 10 (Figure 3) have been
determined by X-ray crystallography.13,14 These were
each crystallized from 1,2-dichloroethane and have
empirical formulas of 8‚(C2H4Cl2)1.5 and 10‚(C2H4Cl2)1.5
.
In contrast to the several conformations found in the
solid state for the diprotio ligand Ph2C(OctH)(C5H5),
metallocenes 8 and 10 present only single conforma-
tions. The proximal methyl groups of 8 are described
as axial/equatorial (Figure 2), and the proximal methyl
group of 10 is described as axial (Figure 3).
F igu r e 3. X-ray crystal structure of 10. Thermal ellipsoids
are shown with 50% probability.
P olym er iza tion Resu lts. The five metallocene pre-
catalysts shown in Schemes 4 and 5 were subjected to
MAO-cocatalyzed polymerizations of propylene. The
polymerization results are summarized in Table 3.
Figure 4 depicts the 13C NMR spectrum of the methyl
region for two polymers made by metallocene 8/MAO
(entry 1 and entry 2), and at the signal-to-noise level of
this spectrum, no pentads other than [rrrr] are ob-
served.15 In addition to these precatalysts, 3, 4, and
those shown in Scheme 6 were also prepared and tested,
primarily for obtaining comparative data. Precatalysts
3, 4,5 and 1416 have been reported, while 13 is the
diphenylmethylidene-bridged variant of the reported
isopropylidene-bridged compound 5.7 Precatalyst 15 is
a Cs-symmetric Oct-containing analogue of 14.
Since the thermal history of a syndiotactic polypro-
pylene sample can often impact its melting behavior,
polymer melting temperature for several polymers was
monitored as a function of thermal history, as described
in Table 4.17 Generally, the first melting temperature
is significantly higher than that obtained during sub-
sequent scans. Those values measured during the
second, third, and fourth scans correspond to the melt-
ing temperature of the thermally quenched polymer and
are quite similar. Annealing processes can often lead
to unusually high melting temperatures,18 as shown for
entry 3 following annealing for 16 h at 145 °C. The
highest melting thermally quenched polymer reported
(16) 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-Verlag: Berlin, 1995; pp 111-147.
(13) Ph2C(C29H36)(C5H4)ZrCl2 (8) (crystallized from 1,2-dichloro-
ethane) is monoclinic P21/n , a ) 13.898(8) Å, b ) 13.698(10) Å, c )
23.275(16) Å, â ) 96.91(6)°, V ) 4399(5) Å3, Z ) 4, T ) 85 K. CCDC
deposition number 112475.
(17) Thermal analyses were performed with a Perkin-Elmer DSC 7
with a scan rate of 10 °C/min. Following a scan from 50 °C to 200 °C,
the sample was brought to 20 °C at a rate of 200 °C/min.
(14) Ph2C(C21H22)(C5H4)ZrCl2 (10) (crystallized from 1,2-dichloro-
ethane) is triclinic P1, a ) 9.1631(10) Å, b ) 12.4550(13) Å, c )
16.7351(18) Å, R ) 70.655(2)°, â ) 87.699(2)°, γ ) 87.820(2)°, V )
1800.0(3) Å3, Z ) 2, T ) 98 K. CCDC deposition number 137697.
(15) The polymer described by entry 15 has been analyzed by L.
Resconi at Montell Polyolefins, and its 13C NMR data are reported in
ref 3d. The pentad distribution (13C NMR, 120 °C, C2D2Cl4): mmmm
) 0.0%; mmmr ) 0.0%; rmmr ) 0.6%; mmrr ) 1.2%; mmrm + rrmr
) 2.7%; mrmr ) 0.1%; rrrr ) 91.7%; rrrm ) 3.8%; mrrm ) 0.0%. The
intrinsic viscosity in tetrahydronaphthalene was determined at 135
°C: 4.21 dL/g. This corresponds to a viscosimetric molecular weight
(Mv) of 730 000. DSC measurements were identical to those reported
in Table 3 (Tm ) 154.3 °C). It was suggested that these polypropylenes
are too syndiotactic to obtain reliable tacticity information. Therefore,
polymer melting temperature is often taken as an alternative measure
of syndiotacticity for the polymers reported herein.
(18) Caution is advised in comparing melting temperatures of
syndiotactic PP, because annealed and thermally quenched polymers
typically afford widely different values. (a) Annealing at 140 °C for 2
h provided a Tm of 170 °C: Grisi, F.; Longo, P.; Zambelli, A.; Ewen, J .
A. J . Mol. Catal. A-Chem. 1999, 140 (3), 225-233. (b) Annealing at
140 °C provided a Tm of 161.2 °C. The thermally quenched polymer
melted at 146.3 °C: Lovinger, A. J .; Lotz, B.; Davis, D. D.; Schumacher,
M. Macromolecules 1994, 27, 6603-6611. (c) A recent report of
syndiotactic polypropylenes prepared via chain-end control with non-
metallocene titanium-based catalysts indicated melting temperatures
up to 156 °C. The reported triad content of [rr] ) 94% suggests that
these polymers were annealed. See: Mitani M.; Furuyama, R.; Mohri,
J .; Saito, J .; Ishii, S.; Terao, H.; Kashiwa, N.; Fujita, T. J . Am. Chem.
Soc. 2002, 124, 7888-7889.

Highly Stereoregular Syndiotactic Polypropylene
Organometallics, Vol. 23, No. 8, 2004 1781
Ta ble 3. MAO-Coca ta lyzed P olym er iza tion Resu lts w ith 8-12, 3, 4, a n d 13-15
b
metallocene
(mg)
MAO
(equiv)
Tp
(°C)
toluene
(mL)
C3H6
(mL)
time
(min)
yield
(g)
Tm
entry
activitya
(°C)
[r] (%)
Mw
Mw/Mn
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15c
16
17
18
19
20
21
22
23
24
25
26
8 (0.5)
8 (0.5)
8 (2.0)
8 (2.0)
8 (2.0)
8 (2.0)
9 (3.0)
9 (3.0)
10 (2.0)
10 (2.0)
11 (3.0)
11 (3.0)
12 (0.5)
12 (0.5)
12 (2.0)
12 (2.0)
3 (1.0)
3 (1.0)
4 (3.0)
4 (3.0)
13 (2.0)
13 (2.0)
14 (0.5)
14 (0.5)
15 (0.5)
15 (0.5)
2000
2000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
2000
2000
1000
1000
1000
1000
1000
1000
1000
1000
2000
2000
2000
2000
0
20
0
20
0
20
0
20
0
20
0
20
0
20
0
20
0
20
0
20
0
20
0
20
0
1.0
1.0
2.0
2.0
30.0
30.0
2.0
2.0
2.0
2.0
2.0
2.0
1.0
1.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.0
1.0
1.0
1.0
30
30
30
30
3
10
10
10
10
20
5
15
10
10
5
60
20
10
10
10
10
10
10
30
15
5
0.48
1.16
0.82
1.44
1.44
1.68
0.25
0.98
0.27
1.60
0.27
1.27
0.26
0.26
1.16
4.79
0.31
1.13
0.16
0.70
1.12
2.24
3.00
5.09
1.09
2.85
5700
14 000
2500
4300
2200
10 000
330
2000
800
9600
90
1300
3100
3100
3500
14 000
1900
6800
110
153
148
149
146
146
140
111
88
140
137
141
120
151
147
154
153
142
136
124
119
144
139
n.o.
n.o.
n.o.
n.o.
>99
>98
961 000
843 000
2.1
1.8
>98
3
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
88.6
89.8
>98
92.2
89.7
>98
97.5
535 000
310 000
2.0
2.0
940
6700
22 000
72 000
120 000
26 000
68 000
3
5
5
5
50.2
50.5
81.3
74.1
20
5
a
b
Reported in g PP/(g metal‚h). n.o. ) not observed. c See ref 15.
F igu r e 4. 13C NMR spectrum of the methyl region for polypropylenes made by 8/MAO (entry 1 and entry 2).
Sch em e 6
here is that produced by metallocene 12/MAO, which
is capable of producing syndiotactic polypropylene with
a melting temperature of 154.3 °C (entry 15). On the
basis of comparing melting points of the polypropylenes
produced by this series of zirconocene dichlorides, the
syndioselectivity of the Oct complex 8 is higher than
that of 10, 3, and 13, which contain Tet, Flu, and 2,7-
di-tert-butylfluorenyl, respectively. For the hafnium
analogues, it is the Tet-containing catalyst system (11/
MAO) that produces the highest melting polymer.

1782 Organometallics, Vol. 23, No. 8, 2004
Ta ble 4. P olym er Meltin g Tem p er a tu r es Dep en d on th e Th er m a l Histor y of th e Sa m p lea
first scan second scan third scan fourth scan fifth scan sixth scan seventh scan eighth scan
Miller and Bercaw
entry
2
Tm (°C)
Hm (J /g)
Tm
Hm
Tm
160.6
83.2
165.1
80.2
157.3
73.6
148.0
38.2
148.9
48.2
146.1
43.1
148.0
36.1
148.9
47.0
146.1
44.3
148.0
37.2
148.9
47.6
146.1
44.9
3
4
anneal at
145 °C
168.6
65.7
148.7
48.2
148.7
47.0
148.7
48.0
Hm
15
16
17
Tm
Hm
Tm
Hm
Tm
Hm
159.9
78.8
156.6
81.1
148.7
61.0
154.3
52.1
153.1
47.2
141.6
37.9
154.3
53.2
153.4
47.9
141.6
37.9
154.3
52.9
153.4
48.1
141.6
37.6
a
Entry numbers correspond to those of Table 3.
F igu r e 5. Space-filling model of 8 and H-Cl interatomic
distances (in Å).
The magnitude of the effects of the (seemingly distal)
tetramethyltetrahydrobenzo substitutents for the Oct
and Tet ligands on polypropylene tacticity might seem
rather unexpected. On the other hand, inspection of
space-filling models derived from the X-ray crystal
structures of 8 and 10 reveals surprisingly close contacts
between the chloride ligands and the “distal” 2,5-
dimethyl-2,5-hexanediyl substituents. Figure 5 depicts
metallocene 8 and provides the interatomic H-Cl
distances. Remarkably, the shortest H-Cl distances
3.113 Åsis that between chlorine and a hydrogen on
the equatorially positioned methyl group of the Oct
ligand. In fact, the only other atom in the molecule
closer to chlorine is zirconium (to which it is covalently
bound with Zr-Cl ) 2.416 Å), despite the fact that the
H-Cl interaction is separated by seven covalent bonds
(H-C-C-C-C-C-Zr-Cl). Figure 6 depicts metal-
locene 10 and provides the interatomic H-Cl distances.
In this molecule, the H-Cl distances for the axial
methyl group are slightly shorter than those for the
axial methyl group of 8 (3.252 vs 3.272 Å). Notwith-
standing the shorter zirconium-R-carbon bond present
in the active polymerization species, it is probable that
there exist significant steric interactions between the
distal substituents and the â- and γ-carbons of the
growing polymer chain.19
F igu r e 6. Space-filling model of 10 and H-Cl interatomic
distances (in Å).
(MAO) counteranion(s) in olefin polymerization cata-
lysts is not established and likely variable,20 and its
ability to coordinate has been shown to impact relative
rates of stereochemically important events.21 On the
other hand, little has been reported on how the size of
the cationic transition metal species can affect ion-
pairing phenomena. Thus the varying sizes of the Flu,
Tet, and Oct ligands may significantly alter the ion-
pairing dynamics and result in different stereoselectivi-
ties.
Ster eoer r or An a lysis. As shown in Scheme 7 an
isolated enantiofacial misinsertion will produce pentads
of the type rrrm, rrmm, and rmmr in a 2:2:1 ratio. A
(19) (a) It is proposed that nonbonded repulsive interactions in the
propagative transition state greatly disfavor enantiofacial misinser-
tions for Oct-containing metallocenes. At the same time, ground state
attractive interactions possible between the growing polymer chain and
the Oct ligand may increase the barrier to site epimerization. These
combined effects result in the highest apparent syndioselectivity
observed for a fluorenyl-based metallocene polymerization catalyst. For
details on this preliminary theoretical model see: (a) Miller, S. A.;
Bercaw, J . E. 217th Am. Chem. Soc. Nat. Meet., Anaheim 1999, INOR
151. (b) Miller, S. A. Ph.D. Thesis, California Institute of Technology,
2000; Chapter 3.
(20) Babushkin, D. E.; Brintzinger, H.-H. J . Am. Chem. Soc. 2002,
124, 12869-12873.
(21) (a) Mohammed, M.; Nele, M.; Al-Humydi, A.; Xin, S.; Stapleton,
R. A.; Collins, S. J . Am. Chem. Soc. 2003, 125, 7930-7941. (b) Chen,
M.-C.; Marks, T. J . J . Am. Chem. Soc. 2001, 123, 11803-11804.
An alternative explanation for enhanced catalyst
stereoselectivity may be found upon consideration of
subtle ion-pairing effects (vide infra). The size of the

Highly Stereoregular Syndiotactic Polypropylene
Organometallics, Vol. 23, No. 8, 2004 1783
Sch em e 7
Ta ble 5. Lea st-Squ a r es Deter m in a tion of th e
Rela tive P r oba bilities of En a n tiofa cia l
Misin ser tion (P e.m .) a n d Site Ep im er iza tion (P s.e.
single, isolated site epimerization will lead to pentads
rrrm and rrmr in a 2:2 ratio.22Assuming that stereoer-
rors are isolated (separated by 4 or more correctly
inserted monomer units),23 one can calculate the relative
likelihood of enantiofacial misinsertion versus site
epimerization. If the relative probability of enantiofacial
misinsertion is defined as Pe.m. and the relative prob-
ability of site epimerization is defined as Ps.e., then the
following relationships can be written, where Ixxxx is the
measured intensity for the xxxx pentad:
a
)
entry
7
8
11
12
15
metallocene dichloride
9
9
11
11
12
Tp (°C)
0
20
0
20
0
pentad (%)
[mmmm]
[mmmr]
[rmmr]
[mmrr]
[mrmm] + [rmrr]
[mrmr]
[rrrr]
0.0
0.0
3.0
7.0
9.9
0.0
69.6
10.6
0.0
0.0
0.0
2.0
4.2
12.2
0.0
66.0
15.7
0.0
0.0
0.0
2.8
5.1
4.9
0.0
76.4
10.9
0.0
0.0
0.0
3.4
7.1
6.7
0.0
69.2
13.6
0.0
0.0
0.0
0.6
1.2
2.7
0.1
91.7
3.8
0.0
Irrrm ) 2Pe.m. + 2Ps.e.
Irmmr ) Pe.m.
[rrrm]
[mrrm]
Pe.m.
0.025 0.020 0.027 0.035 0.006
0.039 0.060 0.026 0.033 0.013
Immrr ) 2Pe.m.
Irmrr ) 2Ps.e.
Ps.e.
P
s.e./Pe.m.
% rms error
1.53
1.83
3.03
0.21
0.95
0.25
0.94
0.09
2.21
0.04
a
Entry numbers correspond to those of Table 3.
A least-squares calculation24 was performed on the
relevant pentad intensities for entries 7, 8, 11, and 12
for polymers made with Oct-containing (9, entries 7 and
8) and Tet-containing (11, entries 11 and 12) hafnocene
catalysts. The same calculation was performed for entry
15, which is a polymer made with an Oct-containing
zirconocene (12). The results are summarized in Table
5. For catalyst systems 9/MAO and 11/MAO, an increase
in polymerization temperature results in an increase in
s.e.. This outcome is anticipated because the rate for
P
the unimolecular site epimerization process is more
temperature dependent (∆Sq is near zero; ∆Hq relatively
large), while that for the bimolecular propagation step
increases more slowly with increasing temperature (∆Sq
is large and negative; ∆Hq thus relatively small). The
unimolecular process is therefore expected to become
increasingly more prevalent at higher temperatures.
Less clear is why the Tet-containing hafnocene catalyst
derived from 11 is less likely than the Oct-containing
analogue 9 to undergo site epimerization. This result
is contrary to that observed for the zirconocene series.
An additional apparent discrepancy is that 9 allows a
considerable fraction of enantiofacial misinsertions,
while the isosteric zirconium analogue 12 practically
forbids them. It is possible that the hafnium catalyst
allows chain epimerizations,25 producing mm stereo-
(22) (a) Farina, M. Top. Stereochem. 1987, 17, 1-111. (b) Price, F.
P. In Markov Chains and Monte Carlo Calculations in Polymer Science;
Lowry, G. G., Ed.: Marcel Dekker: New York, 1970; Chapter 7. (c)
Bovey, F. A. High-Resolution NMR of Macromolecules; Academic
Press: New York, 1972. (d) Randall, J . C. Polymer Sequence Determi-
nation: Carbon-13 NMR Method; Academic Press: New York, 1977.
For leading references, see (e) Randall, J . C. Macromolecules 1997,
30, 803-816.
(23) For example, for entry 7 (Hf catalyst), 11.4% of the dyads are
m dyads (dyads resulting from mistakes). The probability of sequential
m dyads is about (0.114)(0.114) ) 1.3%. We considered this number
to be comparatively small (close to the experimental error of the 13C
NMR analysis), and thus the assumption that stereoerrors are isolated
was retained.

1784 Organometallics, Vol. 23, No. 8, 2004
Miller and Bercaw
Ta ble 6. P olyp r op ylen e P en ta d Distr ibu tion s fr om
En tr ies 23, 24, 25, a n d 26a
UV/Visible Sp ectr a of MAO-Activa ted Dich lo-
r id es vis-a` -vis P r op ylen e P olym er iza tion Activity.
The variety of color changes observed during polymer-
izations with fluorenyl-containing metallocenes prompted
a UV/visible spectroscopic analysis of these new Tet and
Oct derivatives for comparison. While addition of MAO
(500 equiv) to a toluene solution of the “parent” fluor-
enyl-cyclopentadienyl complex, the “Flu” complex 3
(Scheme 6), results in a small red shift of the absorption
band (λmax from 498 to 518 nm; ∆E ) -775 cm-1), an
exceptionally large red shift is observed following ad-
dition of MAO (500 equiv) to a toluene solution of 8 (λmax
from 520 to 612 nm; ∆E ) -2890 cm-1). The intensity
of these bands strongly suggests a ligand to zirconium-
(IV) charge transfer transition. Quite possibly, the large
red shift observed for activated 8 could arise from the
sterically demanding Oct ligand, disfavoring associated
ion pairs; thus the transition involves a more nearly true
cationic zirconium. Dichloride 10, which contains the
Tet ligand, undergoes MAO-induced color transitions
which are very similar to those observed with Flu
compound 3 (purple polymerization reactions), suggest-
ing that extensive substitution on both sides of metal-
locene 8 is requisite for observing drastic shifts in the
LMCT band. Although it is rare for d0 complexes to
emit,28 several fluorenyl-containing complexes studied
gave emission spectra; for example, irradiation of Ph2C-
(Oct)(C5H4)ZrCl2 (8, λmax ) 520) causes emission with
a maximum near 590 nm. Although the lifetime of the
zirconium species is apparently very short, the fluores-
cence lifetime for Ph2C(Oct)(C5H4)HfCl2 (9, λmax ) 492)
was determined (in methylcyclohexane): τ ) 155 ns (λEM
) 610 nm).29
entry
23
24
25
26
metallocene dichloride
14
14
15
15
Tp (°C)
0
20
0
20
pentad
[mmmm]
[mmmr]
[rmmr]
[mmrr]
[mrmm] + [rmrr]
[mrmr]
[rrrr]
(%)
4.8
(%)
5.9
(%)
1.4
2.3
4.9
11.8
4.7
(%)
2.7
3.9
5.9
12.2
9.4
11.2
8.8
12.2
24.3
13.6
6.3
10.5
7.1
13.7
23.9
14.4
6.6
3.7
5.3
49.1
16.0
6.1
41.9
15.2
3.5
[rrrm]
mrrm]
13.3
5.5
11.4
6.5
[r]
[m]
50.1
49.9
50.5
49.5
81.3
18.7
74.1
25.9
a
Entry numbers correspond to those of Table 3.
errors, the same stereochemical defects produced by
isolated enantiofacial misinsertions. Admittedly, the
calculated value for Pe.m. is a combination of unimolecu-
lar chain epimerization and bimolecular enantiofacial
misinsertion. Since the calculated value for Pe.m. is not
significantly altered over the narrow temperature range
studied, the relative contributions of these two processes
cannot presently be disentangled.
Dista l Desym m etr iza tion of a C2v-Sym m etr ic
Meta llocen e Ca ta lyst. In the above catalytic systems
that incorporate the Oct or Tet ligands, the distal ligand
perturbation has been applied to a parent metallocene
of Cs symmetry. An alternative approach is to apply the
steric perturbation to a C2v-symmetric metallocene. To
this end, C2v-symmetric metallocene 1416 was desym-
metrized to the Cs-symmetric metallocene 15 by incor-
poration of one Oct ligand.26 The polymerization results
are reported in Table 3 in entries 23 and 24 (14) and in
entries 25 and 26 (15); the pentad distributions are
reported in Table 6. It is immediately apparent that the
distal ligand perturbations have a pronounced effect on
polymer stereochemistry. Whereas the polymers made
from 14 are atactic with [rrrr] pentad fractions of
approximately 6%, this pentad constitutes well over 40%
of the polymers made from 15. A stereochemical analy-
sis of the pentad distributions22,27 suggests that the
latter polymers are made primarily via syndioselective
enantiomorphic site control (Table 7) and not by chain
end control, although a competing site epimerization
process likely contributes to a nonideal statistical fit.
To the extent that the red shift is proportional to
counterion dissociation, one might expect catalyst activ-
ity to correlate with the magnitude of the red shift. This
phenomenon has been reported with the indenyl-
containing complex rac-C2H4(1-C9H6)2ZrCl2 following
the addition of varying amounts of MAO.30 The same
correlation between propylene polymerization activity
and the observed red shift is also found for the Flu-,
Tet-, and Oct-containing metallocene series. Table 8
shows that Oct-containing metallocenes 8 and 9 are the
most active (1000 equiv of MAO, units ) kg polymer/
(mol M‚h)), despite being the most sterically crowded.
Con clu sion s
For the series of syndioselective polymerization cata-
lysts of the type R2C(Flu′)(C5H4)ZrCl2/MAO, it was
found that the catalyst system with Flu′ ) octamethyl-
octahydrodibenzofluorenyl (Oct) provided polypropylene
of significantly higher melting temperature as compared
to metallocene dichlorides containing other substituted
(24) The least-squares minimization was performed for four pentads
(rmmr, mmrr, rmrr, and rrrm) according to % rms error ) (((∑(Iobs
-
I
calc)2)/4)0.5) × 100.
(25) (a) Busico, V.; Cipullo, R. J . Am. Chem. Soc. 1994, 116, 9329.
(b) Resconi, L.; Fait, A.; Piemontesi, F.; Colonnesi, M.; Rychlicki, H.;
Zeigler, R. Macromolecules 1995, 28, 6667. (c) Reference 9f, p 596. (d)
Yoder, J . C.; Bercaw, J . E. J . Am. Chem. Soc. 2002, 124, 2548.
(26) A similar desymmetrization has been reported to arrive at the
Cs-symmetric 2,7-disubstituted metallocene: Me2Si(C13H8)(2,7-di-tert-
butyl-C13H6)ZrCl2. Patsidis, K.; Alt, H. G.; Milius, W.; Palackal, S. J .
J . Organomet. Chem. 1996, 509, 63-71. For a propylene polymerization
conducted at 70 °C (1000 equiv of MAO), the resulting polymer
contained 32.46% rr triads. The unsubstituted analogue 14 provides
polymers with rr triad contents of 25.1% and 24.5% (Table 8, entries
23 and 24), while the Oct-containing analogue 15 provides polymers
with rr triad contents of 71.2% and 60.6% (Table 8, entries 25 and
26). These results highlight the importance of substitution at the 3
and 6 positions of fluorenyl for effecting a significant change in polymer
stereochemistry.
(28) (a) Heinselman, K. S.; Hopkins, M. D. J . Am. Chem. Soc. 1995,
117, 12340-12341. (b) Williams, D. S.; Thompson, D. W.; Korolev, A.
V. J . Am. Chem. Soc. 1996, 118, 6526-6527.
(29) Rack, J . R.; Gray, H. B.; Miller, S. A.; Bercaw, J . E. Unpublished
results.
(30) (a) Coevoet, D.; Cramail, H.; Deffieux, A. Macromol. Chem.
Phys. 1998, 199, 1451-1457. (b) Coevoet, D.; Cramail, H.; Deffieux,
A. Macromol. Chem. Phys. 1998, 199, 1459-1464. (c) Pieters, P. J . J .;
Vanbeek, J . A. M.; Vantol, M. F. H. Macromol. Rapid Commun. 1995,
16, 463-467. (d) Wieser, U.; Brintzinger, H.-H. In Organometallic
Catalysts and Olefin Polymerization; Blom, R., Follestad, A., Rytter,
E., Tilset, M., Ystenes, M., Eds.; Springer: Berlin, 2001; pp 3-13. (e)
Paczkowski, N.; Gregorius, H.; Kristen, M. O.; Sueling, C.; Suhm, J .;
Brintzinger, H.-H.; Wieser, U. PCT Int. Appl., WO 03/06961, 2003.
(27) The least-squares minimization was performed on the nine
resolvable pentad intensities according to rms error ) (((∑(Iobs - Icalc)2)/
9)0.5).

Highly Stereoregular Syndiotactic Polypropylene
Organometallics, Vol. 23, No. 8, 2004 1785
Ta ble 7. Sta tistica l F its of P olym er s fr om En tr ies 25 a n d 26 for Ca ta lyst 15 to th e Ch a in En d Con tr ol
Mod el a n d th e Syn d ioselective En a n tiom or p h ic Site Con tr ol Mod ela
entry 25 (Tp ) 0 °C)
entry 26 (Tp ) 20 °C)
pentad
[mmmm]
[mmmr]
[rmmr]
[mmrr]
[mrmm] + [rmrr]
[mrmr]
[rrrr]
observed (%)
chain end control (%)
site control (%)
observed (%)
chain end control (%)
site control (%)
1.4
2.3
4.9
11.8
4.7
3.7
49.1
16.0
6.1
81.3
18.7
0.1
0.6
1.7
3.3
19.1
3.3
51.7
18.5
1.7
84.8
15.2
0.152
1.5
3.1
7.8
15.6
6.1
3.1
45.8
15.6
1.5
75.3
24.7
2.7
3.9
5.9
12.2
9.4
5.3
41.9
15.2
3.5
74.1
25.9
0.1
1.1
2.4
4.7
21.3
4.7
43.2
20.2
2.4
81.1
18.9
0.189
1.8
3.7
8.0
16.1
7.4
3.7
41.3
16.1
1.8
72.9
27.1
[rrrm]
[mrrm]
[r]
[m]
σ
R
0.855
2.52
0.838
1.87
rms error
6.03
5.29
a
Entry numbers correspond to those of Table 3.
Ta ble 8. Activity Com p a r ison s for F lu -, Tet-, a n d
Oct-Con ta in in g Meta llocen e Dich lor id e/MAO
Ca ta lyst System sa
atmosphere of argon or nitrogen using standard glovebox,
Schlenk, and high-vacuum line techniques.31 Solvents are dried
according to standard procedures. The following were pur-
chased from Aldrich and used as received: redistilled pyrro-
lidine (99.5+%); fluorene (98%); n-butyllithium (1.6 M in
hexanes); zirconium tetrachloride (99.5%); aluminum chloride
(99.99%); 2,5-dimethyl-2,5-hexanediol (99%); benzophenone
(99%); and nitromethane (96%). Dicyclopentadiene was ob-
tained from Aldrich and cracked following standard procedures
prior to use. Crystalline LiCH2Si(CH3)3 (Aldrich, solution in
pentane) is obtained by condensing the pentane solution. 1,2-
Dibromoethane (Aldrich) was dried over calcium hydride and
isolated by vacuum transfer. Hafnium tetrachloride (99%) was
obtained from Cerac and used as received.
Flu
Tet
Oct
Tp (°C)
Zr (3)
Hf (4)
Zr (10)
Hf (11)
Zr (8)
Hf (9)
0
20
1000
3800
70
600
540
6400
69
960
1700
7800
280
1700
a
Activity measured in units of kg polypropylene/(mol metal‚h).
fluorenyl ligands, including parent fluorenyl, tetrameth-
yltetrahydrobenzofluorenyl (Tet), and 2,7-di-tert-bu-
tylfluorenyl. For the Oct-containing zirconocene dichlo-
rides, melting temperatures as high as 153 °C (8, R )
Ph) or 154 °C (12, R ) Me) were measured for the
thermally quenched polymer. These results underscore
the importance of bulky substituents in the 3 and 6
positions of the fluorenyl ligand.
For the hafnocene dichlorides, it is found that Ph2C-
(Tet)(C5H4)HfCl2/MAO (11/MAO) produces syndiotactic
polymer with the highest melting temperature (141 °C),
surpassing both the fluorenyl and Oct analogues. This
results from the unexpected and diminished propensity
of the Tet-containing catalyst to perform site epimer-
izations. That this is contrary to the results obtained
from the zirconocenes warrants further investigation,
perhaps by molecular modeling studies.
The switch in polymer microstructure on replacement
of a fluorenyl of C2v-symmetric catalyst 14 with an Oct
ligand is clearly the most compelling evidence of the
importance of the distal substituents. While the parent
catalyst C2H4(Flu)2ZrCl2/MAO produces essentially atac-
tic polypropylene ([rrrr] ) 6.3%), the distal substituents
of C2H4(Oct)(Flu)ZrCl2/MAO (15/MAO) yield a moder-
ately syndiotactic polypropylene with an [rrrr] pentad
content of 49.1%.
In str u m en ta tion . NMR spectra were recorded on a J EOL
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 chromato-
graph connected to a Hewlett-Packard 5989A mass spectome-
ter. 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.
Meta llocen e Dich lor id e Syn th eses. 6,6-Dip h en ylfu l-
ven e. (Synthesis modified from that previously reported.32
)
Sodium methoxide (41.00 g, 759.0 mmol), ethanol (500 mL),
and benzophenone (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.
Finally, despite the added steric bulk, the activities
of Oct-containing metallocenes generally exceed those
of less substituted catalyst systems. Evidence that this
might be related to diminished ion pairing is found in
an exceptionally large red shift of the LMCT band upon
exposure of Ph2C(Oct)(C5H4)ZrCl2 (8) to MAO (∆E )
-2890 cm-1).
2,5-Dich lor o-2,5-d im eth ylh exa n e. 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 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
Exp er im en ta l Section
(31) Burger, B. J .; Bercaw, J . E. New Developments in the Synthesis,
Manipulation, and Characterization of Organometallic Compounds;
1987; ACS Symposium Series 357.
Gen er a l Con sid er a tion s. Unless otherwise noted, all
reactions and procedures are carried out under an inert
(32) Thiele, J . Chem. Ber. 1900, 33, 666.

1786 Organometallics, Vol. 23, No. 8, 2004
Miller and Bercaw
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). Analysis Calcd for C8H16Cl2:
C, 52.47; H, 8.81. Found: C, 52.65, 52.35; H, 9.74, 9.39.
Octa m eth ylocta h yd r od iben zoflu or en e. 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-MS):
6.24, 6.28 (m, 3H, Cp-CH1), 7.11-7.18 (m, 10H, phenyl-H),
7.35, 7.35 (s, 4H, Oct-H). 13C NMR (Cl2DCCDCl2, 100 °C): δ
32.00, 32.03, 32.12, 32.39 (CH3), 34.37, 34.50 (CH0), 35.74,
35.85 (CH2), 40.85 (Cp-CH2), 53.16 (9-Oct-CH1), 60.08 (C(Oct)-
(Cp)(Ph)2), 116.14, 125.40, 125.82 (Cp-CH1), 127.04, 130.18
(Oct-CH1), 126.76, 128.84, 130.31, 131.08, 135.93 (phenyl-CH1),
140.04, 142.34, 142.38, 143.59 (Oct-CH0), other CH0 not
determined. Minor isomer (24%): 1H NMR (Cl2DCCDCl2, 100
°C): δ 1.08, 1.25, 1.33, 1.36 (s, 24H, CH3), 1.72 (m, 8H, CH2),
2.94 (s, 2H, Cp-CH2), 5.49 (s, 1H, 9-Oct-H), 6.36, 6.49, 6.58
(m, 3H, Cp-CH1), 7.11-7.18 (m, 10H, phenyl-H), 7.29, 7.29 (s,
4H, Oct-H). Anal. Calcd for C47H52: C, 91.50; H, 8.50. Found:
C, 90.36, 90.47; H, 7.72, 7.76.
P h 2C(C5H4)(C29H36)Zr Cl2 (8). A 250 mL flask was charged
with Ph2C(C5H4)(C29H36)H2 (10.00 g, 16.21 mmol) and LiCH2-
Si(CH3)3 (3.053 g, 32.42 mmol) and equipped with a 180°
needle valve. Diethyl ether (75 mL) was condensed in at -78
°C and the cold bath removed. Upon warming, 25 mL of
tetrahydrofuran was condensed in. After 45 h, solvent was
removed and ZrCl4 (3.78 g, 16.2 mmol) was added. Petroleum
ether (75 mL) was condensed in at -78 °C, and the cold bath
was removed. After 47 h, solvent was removed and the pink
material was extracted in a cellulose extraction thimble with
200 mL of diethyl ether for 2 days. The volume was reduced
to 100 mL, and the precipitate was collected and dried in
vacuo: 5.03 g (39.9%). Two additional crops were obtained for
a total mass of 5.519 g (43.8%). MS (LC-MS): m/z 776.8 (M+).
1H NMR (C6D6): δ 1.01, 1.07, 1.35, 1.51 (s, 24H, Oct-CH3), 1.61
(m, 8H, Oct-CH2), 5.68, 6.21 (s, 4H, Cp-H), 6.42, 8.42 (s, 4H,
Oct-H), 6.97, 7.08, 7.12 (t, 3J HH ) 7.0, 7.3, 8.1 Hz, 6H, phenyl-
1
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.
Calculated for C29H38: C, 90.09; H, 9.91. Found: C, 89.07,
89.16; H, 8.94, 8.85.
Alt er n a t ive P r ep a r a t ion of Oct a m et h yloct a h yd r o-
d iben zoflu or en e. An argon-purged 2 L vessel was charged
with fluorene (45.30 g, 0.2725 mol), 2,5-dichloro-2,5-dimeth-
ylhexane (100.00 g, 0.5461 mol), and nitromethane (800 mL).
The solids were dissolved by gentle heating. A solution of AlCl3
(44.65 g, 0.335 mol) in 60 mL of nitromethane was syringed
in over 6 min. During the addition, much HCl evolved through
an oil bubbler and precipitate was rapidly formed. After
stirring for 18 h, the steel blue reaction was filtered and the
solid collected on filter paper. Water (300 mL) was slowly
added to the filtrate, and the formed precipitate was collected
by suction filtration. The combined precipitates were added
slowly to 400 mL of water. Hexanes (200 mL) were added to
this, and the slurry was stirred overnight to quench the
aluminum chloride. The water layer was removed and the
solvent removed from the remaining slurry by rotary evapora-
tion. The solid was extracted over a period of 3 days with 300
mL of diethyl ether from a cellulose extraction thimble. Diethyl
ether was removed by rotary evaporation and the remaining
solid boiled in 100 mL of hexanes, cooled, filtered, and washed
with 50 mL of hexanes. In vacuo drying afforded 87.75 g of
octamethyloctahydrodibenzofluorene as a white powder (83.3%).
MS (GC-MS): m/z 386.5 (M+). 1H NMR (CDCl3): δ 1.32, 1.38
(s, 24H, CH3), 1.72 (apparent s, 8H, CH2), 3.77 (s, 2H, CH2),
7.43, 7.66 (s, 4H, Flu-H).
P h 2C(C5H4)(C29H36)H2. A 300 mL flask was charged with
octamethyloctahydrodibenzofluorene (12.00 g, 31.04 mmol),
equipped with a 180° needle valve, evacuated, and charged
with diethyl ether (120 mL) by vacuum transfer. At 0 °C,
n-butyllithium (21.0 mL, 33.6 mmol, 1.6 M in hexanes) was
syringed in over 3 min, giving much yellow precipitate. After
21 h, the solvent was removed and 6,6-diphenylfulvene (7.148
g, 31.04 mmol) was added. Diethyl ether (150 mL) was
condensed in and the reaction stirred at room temperature for
5 days before 60 mL of aqueous NH4Cl was added slowly at 0
°C. The organic layer was isolated, and the aqueous layer was
extracted with diethyl ether (4 × 100 mL). The combined
organic layers were dried over MgSO4, filtered, and rotavapped
to provide the crude product in quantitative yield (19.15 g).
The material can be recrystallized from ethanol. MS (GC-
MS): m/z 616.8 (M+). Major isomer (76%): 1H NMR (Cl2-
DCCDCl2, 100 °C): δ 1.02, 1.22, 1.33, 1.36 (s, 24H, CH3), 1.71
(m, 8H, CH2), 3.00 (s, 2H, Cp-CH2), 5.54 (s, 1H, 9-Oct-H), 6.24,
3
H), 7.70, 7.74 (d, J HH ) 7.7, 7.7 Hz, 4H, phenyl-H). 13C NMR
(CD2Cl2): δ 30.97, 31.75, 32.42, 33.50 (CH3), 34.56, 34.80,
34.83, 35.05 (CH0 and CH2), 57.92 (PhCPh), 74.25, 108.78,
119.78, 121.09, 144.90, 146.12, 146.81 (Cp-, phenyl-, and Oct-
CH0), 102.05, 117.83 (Cp-CH1), 121.61, 122.18, 126.71, 127.17,
129.01, 129.10, 129.46 (phenyl- and Oct-CH1). Anal. Calcd for
C
47H50Zr1Cl2: C, 72.65; H, 6.49. Found: C, 69.93; H, 6.10.
P h 2C(C5H4)(C29H36)HfCl2 (9). A 100 mL flask was charged
with Ph2C(C5H4)(C29H36)H2 (3.500 g, 5.637 mmol) and LiCH2-
Si(CH3)3 (1.100 g, 11.68 mmol) and equipped with a 180°
needle valve. Diethyl ether (50 mL) was condensed in at -78
°C and the cold bath removed. Upon warming, 5 mL of
tetrahydrofuran was condensed in. After 45 h, solvent was
removed and HfCl4 (2.00 g, 6.244 mmol) was added. Petroleum
ether (50 mL) was condensed in at -78 °C, and the cold bath
was removed. After 43 h, solvent was removed and the yellow
material was extracted in a cellulose extraction thimble with
150 mL of diethyl ether overnight. The volume was reduced
to 50 mL, and the precipitate was collected and dried in
vacuo: 0.975 g (19.9%). MS (LC-MS): m/z 864.5 (M+). 1H NMR
(C6D6): δ 1.02, 1.08, 1.35, 1.51 (s, 24H, Oct-CH3), 1.61 (m, 8H,
3
Oct-CH2), 5.65, 6.16 (t, J HH ) 2.6, 2.7, 4H, Cp-H), 6.47, 8.40
3
(s, 4H, Oct-H), 6.98, 7.08, 7.13 (t, J HH ) 7.4, 7.6, 9.0 Hz, 6H,
3
phenyl-H), 7.70, 7.77 (d, J HH ) 7.8, 7.9 Hz, 4H, phenyl-H).
13C NMR (CD2Cl2): δ 31.13, 31.81, 32.47, 33.64 (CH3), 34.59,
34.78, 34.78, 35.07 (CH0 and CH2), 57.73 (PhCPh), 74.03,
111.64, 118.78, 119.55, 145.18, 145.55, 146.54 (Cp-, phenyl-,
and Oct-CH0), 99.66, 116.96 (Cp-CH1), 121.23, 121.93, 126.72,
127.15, 128.97, 129.10, 129.49 (phenyl- and Oct-CH1). Anal.
Calcd for C47H50Hf1Cl2: C, 65.31; H, 5.83. Found: C, 63.53;
H, 5.33.
Tetr a h yd r otetr a m eth ylben zoflu or en e. A 2 L argon-
purged vessel was charged with fluorene (200.0 g, 1203 mmol),
2,5-dichloro-2,5-dimethylhexane (25.00 g, 136.5 mmol), and
1000 mL of nitromethane. The solids were dissolved by heating
to 60 °C, and when the vessel had cooled to 50 °C, a solution
of AlCl3 (25.00 g, 187.5 mmol) in nitromethane (40 mL) was
syringed in over 2 min, giving a dark green homogeneous
solution. After 15 min, 60 mL of water was syringed in very
slowly, followed by addition of 100 mL of water. The formed
precipitate (>99% fluorene by GC) was removed by filtration.

Highly Stereoregular Syndiotactic Polypropylene
Organometallics, Vol. 23, No. 8, 2004 1787
The filtrate (77.5:22.5 ratio of FluH:TetH) was extracted with
hexane (7 × 100 mL). Solvent was removed by rotavap from
the combined hexane extracts, giving 64.7 g of FluH and TetH.
This material was subjected to Kugelrohr distillation under
high vacuum. Several fractions were removed at 100 °C until
22.54 g of material remained. This material was dissolved in
300 mL of boiling ethanol, hot filtered, and slowly cooled. The
pure product was obtained by collection of the precipitate and
in vacuo drying: 16.06 g (42.5%). Essentially pure fluorene
was recovered from several of the removed fractions and by
Kugelrohr distillation of the original filtrate: 133.7 g, 75.4%
of the theoretically recoverable amount of 177.3 g. MS (GC-
P h 2C(C5H4)(C21H22)HfCl2 (11). A swivel frit was charged
with Ph2C(C5H4)(C21H22)Li2 (1.40 g, 2.70 mmol) and HfCl4
(1.000 g, 3.12 mmol). Petroleum ether (60 mL) was condensed
in at -78 °C, and the cold bath was removed. The reaction
was allowed to warm slowly, and after 40 h, the solvent was
removed. Diethyl ether (100 mL) was added by vacuum
transfer, and the solution was filtered. The filtrate was
condensed to 25 mL, and the product precipitated very slowly.
The yellow solid was collected and dried in vacuo: 0.518 g
(25.4%). MS (LC-MS): m/z 754.7 (M+). 1H NMR (CD2Cl2): δ
0.84, 0.95, 1.39, 1.45 (s, 12H, CH3), 1.61, 1.68 (m, 4H, CH2),
5.61, 5.67 (q, 3J HH ) 2.6, 2.6 Hz, 2H, Cp-H), 6.25, 6.25 (t, 3J HH
) 2.9, 2.9 Hz, 2H, Cp-H), 6.28, 8.12 (s, 2H, Tet-H), 6.43, 8.13
(d, 3J HH ) 8.8, 8.0 Hz, 2H, Tet-H), 6.93, 7.32 (t, 3J HH ) 8.8, 6.0
Hz, 2H, Tet-H), 7.33, 7.47 (m, 6H, phenyl-H), 7.85, 7.87, 7.95,
1
MS): m/z 276.3 (M+). H NMR (CDCl3): δ 1.38, 1.42 (s, 12H,
CH3), 1.78 (apparent s, 4H, CH2), 3.88 (s, 2H, CH2), 7.29 (t,
3
3J HH ) 7.3 Hz, 1H, Flu-H), 7.38 (t, J HH ) 7.0 Hz, 1H, Flu-H),
3
3
7.53 (d, J HH ) 8.1 Hz, 1H, Flu-H), 7.54 (s, 1H, Flu-H), 7.78
7.96 (d, J HH ) 7.7, 8.8, 6.6, 7.0 Hz, 4H, phenyl-H). 13C NMR
3
(s, 1H, Flu-H), 7.80 (d, J HH ) 7.7 Hz, 1H, Flu-H). 13C NMR
(CD2Cl2): δ 31.07, 31.80, 32.43, 33.62 (CH3), 34.56, 34.78,
34.81, 34.96 (CH0 and CH2), 100.48, 100.70, 116.91, 117.72 (Cp-
CH1), 118.71, 119.83, 120.52, 121.41 (Tet-CH0), 121.25, 121.92,
123.70, 124.49, 124.56, 126.60, 126.78, 127.18, 127.25, 127.68,
128.99, 128.99, 129.07, 129.16, 129.22, 129.51 (Tet-CH1 and
phenyl-CH1), 145.03, 145.28, 146.09, 147.06 (Tet-CH0 and
phenyl-CH0), remaining CH0 not determined. Anal. Calcd for
(CDCl3): δ 32.27, 32.40 (CH3), 34.66, 34.73 (CH0), 35.40, 35.40
(CH2), 36.69 (CH2), 117.67, 119.65, 123.05, 125.03, 126.33,
126.65 (CH1), 139.47, 140.65, 141.98, 143.54, 143.68, 143.90
(CH0). Anal. Calcd for C21H24: C, 91.25; H, 8.75. Found: C,
90.51, 90.17; H, 7.99, 7.65.
P h 2C(C5H4)(C21H22)Li2. A 200 mL flask was charged with
tetramethyltetrahydrobenzofluorene (4.500 g, 16.28 mmol),
equipped with a 180° needle valve, evacuated, and charged
with 100 mL of diethyl ether by vacuum transfer. At room
temperature, an n-butyllithium (11.0 mL, 17.6 mmol, 1.6 M
in hexanes) solution was syringed in over 3 min. After 23 h of
stirring, solvent was removed and 6,6-diphenylfulvene (3.749
g, 16.28 mmol) was added. Diethyl ether (100 mL) was
condensed in, and the homogeneous reaction slowly formed a
white precipitate. After stirring for 23 days, 60 mL of water
was slowly added. The organic layer was isolated, and the
aqueous layer was extracted with diethyl ether (3 × 100 mL).
The combined organic layers were dried over MgSO4 and
filtered, and the volume was reduced to 100 mL. After 48 h at
-78 °C, the liquid was decanted from the formed precipitate,
which was collected and dried in vacuo (7.05 g, 13.9 mmol,
85.5%). The flask was attached to a swivel frit, and 100 mL of
diethyl ether was condensed in before an n-butyllithium
solution (19.0 mL, 30.4 mmol, 1.6 M in hexanes) was syringed
into the white slurry over 3 min at room temperature. The
reaction was stirred for 22 h before the orange solid was
collected by filtration and dried in vacuo: 7.21 g (quantitative
yield).
C
39H36Hf1Cl2: C, 62.12; H, 4.81. Found: C, 61.11; H, 5.02.
6,6-Dim eth ylfu lven e (Synthesis modified from ref 33). A
2 L vessel was charged with methanol (875 mL), cyclopenta-
diene (101.0 g, 1528 mmol), and acetone (230 mL). Pyrrolidine
(13.5 mL, 112 mmol) was syringed in over 10 min. The reaction
was stirred for 18 h before an aqueous acetic acid solution (50
mL of acetic acid/200 mL of water) was added over 2 min.
Diethyl ether (500 mL) and water (1000 mL) were added, and
the organic layer was isolated. The aqueous layer was ex-
tracted with diethyl ether (3 × 100 mL), and the combined
organic layers were extracted with water (5 × 150 mL). The
ether layer was dried over MgSO4, filtered, and rotavapped
at 40 °C for 2 h. This provided ether-free fulvene (156.81 g,
96.7%). This material was further cleaned by pushing the neat
liquid through a short column of alumina: 111.2 g (68.5%).
Me2C(C5H 4)(C29H 36)H 2.
Octamethyloctahydrodibenzo-
fluorene (9.625 g, 24.89 mmol) was massed into a 250 mL
round-bottom Schlenk flask. This was evacuated, backfilled
with argon, and charged with 100 mL of tetrahydrofuran via
syringe. A solution of n-butyllithium in hexanes (16.0 mL, 1.6
M, 25.6 mmol) was syringed in over 10 min, giving initially a
red solution, which later formed some red precipitate. After
100 min, 6,6-dimethylfulvene (3.0 mL, 2.64 g, 24.9 mmol) was
syringed in, yielding a homogeneous solution. After 22 h, 60
mL of aqueous NH4Cl was slowly syringed in 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 give the
product as a yellow crystalline solid, 12.27 g in theoretical
yield.
Me2C(C5H4)(C29H36)Li2. A 250 mL round-bottom flask was
charged with Me2C(C5H4)(C29H36)H2 (12.27 g, 24.89 mmol) and
attached to a swivel frit before 75 mL of diethyl ether was
condensed in. A solution of n-butyllithium in hexanes (32.0
mL, 1.6 M, 51.2 mmol) was syringed in over 3 min at 0 °C.
After stirring for 17 h at room temperature, solvent was
removed and 75 mL of benzene was condensed in. The solution
was frozen and lyophilized to give 11.80 g of the dilithio salt
as an orange powder (93.9%).
P h 2C(C5H4)(C21H22)Zr Cl2 (10). A 100 mL flask was charged
with Ph2C(C5H4)(C21H22)Li2 (3.895 g, 7.51 mmol) and ZrCl4
(1.750 g, 7.51 mmol) and equipped with a 180° needle valve.
Petroleum ether (60 mL) was condensed in at -78 °C, and the
cold bath was removed. The reaction was allowed to warm
slowly, and after 18 h the solvent was removed. The solid was
placed in a cellulose extraction thimble and extracted with 150
mL of methylene chloride overnight. The filtrate was switched
onto a swivel frit, and solvent was removed. Toluene (200 mL)
was condensed in and the solution filtered. The filtrate was
reduced in volume to 40 mL, and the pink precipitate was
collected and dried in vacuo: 2.149 g (42.9%). MS (LC-MS):
1
m/z 666.6 (M+). H NMR (CD2Cl2): δ 0.83, 0.94, 1.40, 1.46 (s,
3
12H, CH3), 1.61, 1.68 (m, 4H, CH2), 5.66, 5.71 (q, J HH ) 2.6,
3
2.9 Hz, 2H, Cp-H), 6.31, 6.31 (t, J HH ) 2.6, 2.6 Hz, 2H, Cp-
3
H), 6.24, 8.15 (s, 2H, Tet-H), 6.39, 8.16 (d, J HH ) 8.8, 8.8 Hz,
3
1H, Tet-H), 6.96, 7.23 (t, J HH ) 7.7, 7.3 Hz, 2H, Tet-H), 7.33,
3
Me2C(C5H4)(C29H36)Zr Cl2 (12). In the glovebox, a swivel
frit apparatus was charged with Me2C(C5H4)(C29H36)Li2 (3.246
g, 6.436 mmol) and zirconium tetrachloride (1.500 g, 6.437
mmol). Petroleum ether (50 mL) was condensed in and the
reaction stirred at room temperature for 51 h before solvent
removal. Dichloromethane (20 mL) was condensed in, stirred,
and removed. Then, 30 mL of diethyl ether was condensed in,
7.49 (m, 6H, phenyl-H). 7.86, 7.88, 7.93, 7.96 (d, J HH ) 8.8,
7.7, 6.6, 6.6, 4H, phenyl-H). 13C NMR (CD2Cl2): δ 30.95, 31.75,
32.41, 33.53 (CH3), 34.53, 34.53, 34.86, 34.95 (CH0 and CH2),
102.90, 103.13, 117.88, 118.67 (Cp-CH1), 121.58, 122.17,
123.98, 124.63, 125.01, 126.60, 126.77, 127.22, 127.29, 127.93,
128.18, 129.07, 129.10, 129.18, 129.21, 129.49 (Tet-CH1 and
phenyl-CH1), 144.76, 145.02, 146.62, 147.35 (Tet-CH0 and
phenyl-CH0), remaining CH0 not determined. Anal. Calcd for
C
39H36Zr1Cl2: C, 70.25; H, 5.44. Found: C, 71.46; H, 5.55.
(33) Stone, K. J .; Little, R. D. J . Org. Chem. 1984, 49, 1849-1853.

1788 Organometallics, Vol. 23, No. 8, 2004
Miller and Bercaw
3
stirred, and removed. In the glovebox, the solid was transferred
to a cellulose extraction thimble, and this was extracted
overnight with 100 mL of diethyl ether. The obtained slurry
was transferred back to the swivel frit and the volume reduced
to 30 mL. The orange precipitate (12) was collected on the frit
and dried in vacuo: 1.649 g (39.2%). MS (LC-MS): m/z 653.7
Hz, 4H, Cp-H), 6.46, 8.16 (d, J HH ) 8.8, 8.4 Hz, 4H, Flu-H),
6.97, 7.31, 7.35 (t, 3J HH ) 7.0, 7.4, 7.4 Hz, 6H, phenyl-H), 7.45,
3
3
7.52 (t, J HH ) 7.7, 7.4 Hz, 4H, Flu-H), 7.86, 7.94 (d, J HH
7.4, 8.1 Hz, 4H, phenyl-H). Anal. Calcd for C31H22Hf1Cl2: C,
57.82; H, 3.44. Found: C, 53.97; H, 3.23.
)
P h 2C(C5H4)(2,7-di-ter t-bu tylflu or en yl)H2. A 250 mL flask
was charged with 2,7-di-tert-butylfluorene34 (10.00 g, 35.91
mmol) and equipped with a 180° needle valve. The flask was
evacuated, and 125 mL of diethyl ether was condensed in
before an n-butyllithium solution (23.0 mL, 36.8 mmol, 1.6 M
in hexanes) was syringed in over 3 min at room temperature.
After 3.5 h, solvent was removed and 6,6-diphenylfulvene
(8.271 g, 35.91 mmol) was added. Diethyl ether (100 mL) was
condensed in, and the reaction was stirred for 26 days. Water
(60 mL) was slowly added, and the white precipitate that
slowly formed was collected by suction filtration. A white
powder (13.11 g, 71.8%) was isolated after drying in vacuo at
80 °C for several hours. MS (GC-MS): m/z 508.6 (M+). Anal.
Calcd for C39H40: C, 92.08; H, 7.92. Found: C, 89.42; H, 7.07.
P h 2C(C5H4)(2,7-d i-ter t-bu tylflu or en yl)Li2. A swivel frit
was charged with Ph2C(C5H4)(2,7-di-tert-butylfluorenyl)H2
(12.00 g, 23.59 mmol), and 150 mL of diethyl ether was added
by vacuum transfer. n-Butyllithium solution (32.0 mL, 51.2
mmol, 1.6 M in hexanes) was added over 8 min at room
temperature. After 19 h, the orange precipitate was collected
by filtration and dried in vacuo: 12.28 g, (quantitative yield).
P h 2C(C5H4)(2,7-d i-ter t-bu tylflu or en yl)Zr Cl2 (13). A 100
mL flask was charged with Ph2C(C5H4)(2,7-di-tert-butylfluor-
enyl)Li2 (3.351 g, 6.437 mmol) and ZrCl4 (1.500 g, 6.437 mmol).
Petroleum ether (60 mL) was condensed in at -78 °C, and the
cold bath was removed. The reaction was allowed to warm
slowly, and after 23 h the solvent was removed. The solid was
placed in a cellulose extraction thimble and extracted with 150
mL of methylene chloride overnight. The filtrate was switched
onto a swivel frit, filtered, and condensed to 10 mL. The orange
precipitate was collected and dried in vacuo: 1.271 g (29.5%).
MS (LC-MS): m/z 668.5 (M+). 1H NMR (CD2Cl2): δ 1.04 (s,
1
(M+). H NMR (CD2Cl2): δ 1.23, 1.38, 1.39, 1.49 (s, 24H, Oct-
CH3), 1.73 (m, 8H, Oct-CH2), 2.32 (s, 6H, (CH3)2C-Oct-Cp), 5.57,
3
6.20 (t, J HH ) 2.6, 2.2 Hz, 4H, Cp-H), 7.63, 8.04 (s, 4H, Oct-
H). 13C NMR (CD2Cl2): δ 28.60, 31.88, 32.22, 32.37, 33.50
(CH3), 34.78, 34.86, 35.03, 35.11 (CH0 and CH2), 40.06 (MeC-
Me), 75.23, 112.89, 121.05, 121.50, 145.68, 147.68 (Cp- and
Oct-CH0), 100.45, 118.44 (Cp-CH1), 120.63, 122.06 (Oct-CH1).
Anal. Calcd for C37H46Zr1Cl2: C, 68.07; H, 7.10. Found: C,
60.14; H, 6.45.
P r ep a r a tion of 3 a n d 4. The syntheses of 3 and 4 were
accomplished as follows by slight modification of the literature
procedure.5a
P h 2C(C5H4)(C13H8)H2. A 500 mL flask was charged with
fluorene (28.87 g, 173.7 mmol), equipped with a 180° needle
valve, evacuated, and backfilled with argon before 180 mL of
diethyl ether was syringed in. n-Butyllithium (110.0 mL, 176
mmol, 1.6 M in hexanes) was syringed in over 8 min at 0 °C.
After 15 h, all solvent was removed, 6,6-diphenylfulvene (40.00
g, 173.7 mmol) was added, and 200 mL of diethyl ether was
condensed in. After stirring for 6 days, the vessel was cooled
to 0 °C and 120 mL of water was very slowly added, followed
by 60 mL of aqueous NH4Cl solution. The slurry was suction
filtered, and the crude product was boiled in 400 mL of ethanol
for 2 h. This was filtered hot, and the collected solid was dried
in vacuo (53.39 g, 77.5%). Three crops were obtained with a
total mass of 64.72 g (94.0%). MS (GC-MS): m/z 396.4 (M+).
Anal. Calcd for C31H24: C, 93.90; H, 6.10. Found: C, 92.30,
92.33; H, 5.27, 5.47.
P h 2C(C5H4)(C13H8)Li2. A 500 mL flask was charged with
Ph2C(C5H4)(C13H8)H2 (21.00 g, 52.96 mmol) and attached to a
swivel frit. Diethyl ether (200 mL) was condensed in before
an n-butyllithium solution (70.0 mL, 112 mmol, 1.6 M in
hexanes) was syringed into the white slurry over 12 min at
room temperature. After 18 h, the orange slurry was heated
at 40 °C for 5 h, when the orange solid was collected on the
frit and dried in vacuo. The product as a diethyl ether adduct
(27.71 g) was obtained (theoretical: 21.63 g for ether-free
product; 25.55 g for mono diethyl ether adduct; 29.48 g for bis
diethyl ether adduct).
3
18H, C(CH3)3), 5.69, 6.34 (t, J HH ) 2.6, 3.7 Hz, 4H, Cp-H),
6.34, (s, 2H, Flu-H), 7.62, 8.05 (d, 3J HH ) 8.8, 8.8 Hz, 4H, Flu-
H), 7.89, 7.98 (d, 3J HH ) 8.1, 8.1 Hz, 4H, phenyl-H), 7.31, 7.38,
7.48 (t, 3J HH ) 7.3, 7.3, 7.7 Hz, 6H, phenyl-H). 13C NMR (CD2-
Cl2): δ 30.36 (C(CH3)3), 34.98 (C(CH3)3), 58.28 (C(Cp)(Flu)-
(Ph)2), 77.82, 110.21, 120.82, 121.63, 144.86, 150.76 (Flu-CH0,
phenyl-CH0, and Cp-CH0), 102.86, 118.39, 119.94, 123.95,
124.66, 126.65, 127.26, 129.07, 129.14, 129.47 (Flu-CH1 and
phenyl-CH1). Anal. Calcd for C39H38Zr1Cl2: C, 70.03; H, 5.73.
Found: C, 67.36; H, 5.13.
P h 2C(C5H4)(C13H8)Zr Cl2 (3). A 100 mL flask was charged
with Ph2C(C5H4)(C13H8)Li2 (4.381 g, 10.73 mmol) and ZrCl4
(2.500 g, 10.73 mmol) and equipped with a 180° needle valve.
Petroleum ether (60 mL) was condensed in at -78 °C and the
cold bath removed. After 24 h, solvent was removed from the
pink slurry, and the solid was extracted in a cellulose thimble
with 150 mL of methylene chloride overnight. The volume of
the filtrate was reduced to 50 mL, and the precipitate was
collected and dried in vacuo: 1.32 g (22.1%). MS (LC-MS): m/z
P r ep a r a tion of 14. Metallocene 14 was synthesized as
reported in the literature.16 MS (LC-MS): m/z 518.5 (M+).
Anal. Calcd for C28H20Zr1Cl2: C, 64.85; H, 3.89. Found: C,
62.92; H, 3.62.
2-(9-F lu or en yl)br om oeth a n e. A 500 mL Schlenk flask
was charged with fluorene (50.00 g, 300.8 mmol), evacuated,
backfilled with argon, and charged with 180 mL of tetrahy-
drofuran via syringe. At 0 °C, an n-butyllithium solution (200.0
mL, 320 mmol, 1.6 M in hexanes) was syringed in over 20 min.
After 1 h this fluorenyllithium solution was cannulated over
60 min into a 1 L flask containing 1,2-dibromoethane (180 mL,
2090 mmol) and 180 mL of petroleum ether. After stirring for
29 h, all volatiles were removed by vacuum transfer. Then,
300 mL of water and 400 mL of methylene chloride were
added. The organic layer was isolated, and the aqueous layer
was extracted with methylene chloride (3 × 50 mL). The
organic layers were dried over MgSO4, filtered, and ro-
tavapped. The resulting orange oil was stirred in 100 mL of
boiling hexanes; this was cooled and filtered. Solvent was
removed from the filtrate, and the resulting oil was subjected
to Kugelrohr distillation under high vacuum at 80 °C to remove
1
3
556.6 (M+). H NMR (CD2Cl2): δ 5.79, 6.36 (t, J HH ) 3.0, 2.6
3
Hz, 4H, Cp-H), 6.42, 8.19 (d, J HH ) 8.8, 8.4 Hz, 4H, Flu-H),
6.99, 7.31, 7.35 (t, 3J HH ) 7.0, 7.3, 7.7 Hz, 6H, phenyl-H), 7.45,
3
3
7.56 (t, J HH ) 7.4, 8.1 Hz, 4H, Flu-H), 7.86, 7.94 (d, J HH
)
7.7, 7.7 Hz, 4H, phenyl-H). Anal. Calcd for C31H22Zr1Cl2: C,
66.89; H, 3.98. Found: C, 61.12; H, 3.65.
P h 2C(C5H4)(C13H8)HfCl2 (4). A 100 mL flask was charged
with Ph2C(C5H4)(C13H8)Li2 (2.55 g, 6.24 mmol) and HfCl4
(2.000 g, 6.24 mmol) and equipped with a 180° needle valve.
Petroleum ether (60 mL) was condensed in at -78 °C and the
cold bath removed. After 21 h, solvent was removed from the
yellow slurry, and the solid was extracted in a cellulose thimble
with 150 mL of methylene chloride for 2 days. The filtrate was
switched onto a swivel frit and gravity filtered. The volume of
the filtrate was reduced to 40 mL, and the precipitate was
collected and dried in vacuo: 1.15 g (28.5%). MS (LC-MS): m/z
(34) Kajigaeshi, S.; Kadowaki, T.; Nishida, A.; Fujisaki, S.; Noguchi,
1
3
644.7 (M+). H NMR (CD2Cl2): δ 5.75, 6.30 (t, J HH ) 3.0, 2.6
M. Synthesis 1984, 335-337.

Highly Stereoregular Syndiotactic Polypropylene
Organometallics, Vol. 23, No. 8, 2004 1789
1 mL of material. The second crop, collected at 110 °C, was
dissolved in 100 mL of hexanes; the solution was pushed
through a column of alumina, rotavapped, and dried in vacuo
to provide 50.55 g (61.5%) of the product as an oil, which
crystallized upon standing. MS (GC-MS): m/z 272.2 (M+). 1H
NMR (CDCl3): δ 2.49, 3.29 (m, 4H, CH2), 4.16 (m, 1H, 9-Flu-
mL Hamilton syringe rated to 200 psi. Temperature mainte-
nance was monitored by an affixed pressure gauge. Polymer-
ization reactions were vented and quenched with a small
volume of methanol/concentrated HCl (12:1), and the polymers
were separated from hydrolyzed aluminoxanes by precipitation
from methanol, followed by filtration. Residual amounts of
toluene and methanol were removed from the obtained poly-
mers by in vacuo drying.
H), 7.32, 7.38, 7.51, 7.75 (m, 8H, Flu-H). Anal. Calcd for C15H13
-
Br1: C, 65.95; H, 4.80. Found: C, 67.35; H, 4.46.
1,2-Eth a n o(C29H36)(C13H8)H2. A 250 mL flask was charged
with octamethyloctahydrodibenzofluorene (10.00 g, 25.87 mmol),
equipped with a 180° needle valve, and evacuated before 80
mL of diethyl ether was condensed in. n-Butyllithium (17.0
mL, 27.2 mmol, 1.6 M in hexanes) was syringed in over 3 min,
providing a yellow slurry. After 2 h, solvent was removed and
2-(9-fluorenyl)bromoethane (7.066 g, 25.87 mmol) was added
in the glovebox. Diethyl ether (80 mL) was condensed in, and
the reaction was stirred for 24 days before cooling to 0 °C and
addition of 100 mL of water and 150 mL of diethyl ether. The
organic layer was isolated, and the aqueous layer was ex-
tracted with diethyl ether (2 × 50 mL). The organic layers were
dried over MgSO4, filtered, and condensed to 50 mL. Cooling
to -78 °C provided six crops of product, which were further
dried in vacuo: 13.36 g (89.2%). MS (GC-MS): m/z 578.7 (M+).
1H NMR (CDCl3): δ 1.30, 1.37, 1.38, 1.42 (s, 24H, CH3), 1.74
(apparent s, 8H, Oct-CH2), 1.64, 1.72 (m, 4H, CH2), 3.71, 3.85
(m, 2H, 9-Flu-H and 9-Oct-H), 7.20, 7.62 (s, 4H, Oct-H), 7.31,
Rep r esen ta tive P olym er iza tion P r oced u r es. En tr y 1.
A 100 mL Lab Crest glass pressure reactor was charged with
MAO (0.075 g, 1.3 × 10-3 mol [Al]). Propylene (30 mL) was
condensed in at 0 °C. A solution of Ph2C(Oct)(C5H4)ZrCl2 (8,
0.0005 g, 6 × 10-7 mol) in toluene (1.0 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.
En tr y 2. A 100 mL Lab Crest glass pressure reactor was
charged with MAO (0.075 g, 1.3 × 10-3 mol [Al]). Propylene
(30 mL) was condensed in at 0 °C. A solution of Ph2C(Oct)-
(C5H4)ZrCl2 (8, 0.0005 g, 6 × 10-7 mol) in toluene (1.0 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.
En tr y 13. A 100 mL Lab Crest glass pressure reactor was
charged with MAO (0.089 g, 1.53 × 10-3 mol [Al]). Propylene
(30 mL) was condensed in at 0 °C. A solution of Me2C(Oct)-
(C5H4)ZrCl2 (12, 0.0005 g, 7.7 × 10-7 mol) in toluene (1.0 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.
3
3
7.38 (t, J HH ) 7.3, 6.8 Hz, 4H, Flu-H), 7.77, 7.77 (d, J HH
)
7.3, 7.3 Hz, 4H, Flu-H). Anal. Calcd for C44H50: C, 91.29; H,
8.71. Found: C, 91.01; H, 8.38.
1,2-Eth a n o(C29H36)(C13H8)Zr Cl2 (15). A 100 mL flask was
charged with 1,2-ethano(C29H36)(C13H8)H2 (3.000 g, 5.182
mmol), evacuated, and charged with 50 mL of diethyl ether.
n-Butyllithium (7.0 mL, 11.2 mmol, 1.6 M in hexanes) was
syringed in over 3 min. After 43 h, solvent was removed from
the yellow slurry and ZrCl4 (1.300 g, 5.58 mmol) was added.
Petroleum ether was condensed in at -78 °C, and the reaction
was allowed to warm slowly. After 29 h, solvent was removed,
50 mL of diethyl ether was condensed in, and this was stirred
an additional 40 h. Solvent was removed and the pink-red solid
was extracted from a cellulose extraction thimble overnight
with 150 mL of diethyl ether. The volume of the filtrate was
reduced to 30 mL, and the precipitate was collected on a frit
and dried in vacuo: 0.779 g (20.3%). MS (LC-MS): m/z 738.7
(M+). 1H NMR (C6D6): δ 1.20, 1.36, 1.37, 1.39 (s, 24H, Oct-
CH3), 1.60 (m, 8H, Oct-CH2), 3.93 (m, 4H, ethano-H), 6.98, 7.09
(t, J HH ) 8.4, 8.4 Hz, 4H, Flu-H), 7.50, 7.54 (d, J HH ) 9.2,
8.4 Hz, 4H, Flu-H), 7.52, 7.89 (s, 4H, Oct-H). 13C NMR (CD2-
Cl2): δ 29.68, 30.00 (CH2-CH2), 31.61, 32.35, 32.57, 33.87
(CH3), 34.87, 34.97, 35.00, 35.07 (CH0 and CH2), 102.10, 105.40,
120.88, 121.28, 126.55, 127.35, 145.74, 146.91 (Flu- and Oct-
CH0), 119.59, 121.97, 122.49, 124.39, 124.49, 127.84 (Flu- and
Oct-CH1). Anal. Calcd for C44H48Zr1Cl2: C, 71.51; H, 6.55.
Found: C, 69.50; H, 6.07.
En tr y 14. A 100 mL Lab Crest glass pressure reactor was
charged with MAO (0.089 g, 1.53 × 10-3 mol [Al]). Propylene
(30 mL) was condensed in at 0 °C. A solution of Me2C(Oct)-
(C5H4)ZrCl2 (12, 0.0005 g, 7.7 × 10-7 mol) in toluene (1.0 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.
P olym er Ch a r a cter iza tion . Polymer melting tempera-
tures 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 distributions were determined by integration
of the nine resolved peaks in the methyl region (19-22 ppm)
of the 13C NMR spectra obtained.35 Spectra were acquired at
124 °C with tetrachloroethane-d2 as solvent. A 90° pulse was
employed with broadband decoupling. A delay time of 3 s and
a minimum of 1000 scans were used.
3
3
Ack n ow led gm en t. This work has been funded by
the USDOE Office of Basic Energy Sciences (Grant No.
DE-FG03-88ER13431) and by a DOD National Defense
Science and Engineering Fellowship awarded to S.M.
The authors thank Dr. Michael Day and Mr. Lawrence
Henling for obtaining the X-ray crystal structures. The
authors also thank Dr. Terry Burkhardt at ExxonMobil
for providing molecular weight data. Dr. Luigi Resconi
at Montell Polyolefins (now Basell Polyolefins) kindly
provided polymer analysis of the sample from entry 15.
P r op ylen e P olym er iza tion P r oced u r es. CAUTION: All
P olym er iza tion P r oced u r es Sh ou ld Be P er for m ed Be-
h in d
a Bla st Sh ield . 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. Polymerizations were
conducted in a 3 oz. Lab Crest glass reaction vessel and were
stirred with a magnetic stir bar. Monomer was condensed into
the vessel over several min at 0 °C. The vessel was then
equilibrated at either 0 or at 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
Su p p or tin g In for m a tion Ava ila ble: X-ray crystal struc-
ture data for Ph2C(OctH)(C5H5), Ph2C(C29H36)(C5H4)ZrCl2 (8),
and Ph2C(C21H22)(C5H4)ZrCl2 (10). This material is available
free of charge via the Internet at http://pubs.acs.org.
OM030333F
(35) (a) Busico, V.; Cipullo, R.; Corradini, P.; Landriani, L.; Vacatello,
M.; Segre, A. L. Macromolecules 1995, 28, 1887-1892. (b) Busico, V.;
Cipullo, R.; Monaco, G.; Vacatello, M. Macromolecules 1997, 30, 6251.