Zr-alkyl isomerization in ansa-zirconocene-catalyzed olefin polymerizations. Contributions to stereoerror formation and chain termination

Article abstract of DOI:10.1021/ja961157n

In polymers made from (E)- or (Z)-[1-D]propene with methyl alumoxane (MAO)-activated ansa-zirconocene catalysts, signals of deuterium-labeled mrrm pentads document that stereoerrors arise mainly from an isomerization of the Zr-bound chain end. Reduced D-atom redistribution in poly( [2-D]propene) indicates a kinetic isotope effect of k(H)/k(D) ~ 3 for the β-H/D transfer associated with the isomerization reaction. In all poly([1- and [2-D]propenes) studied, D-labeled mmmm pentads occur with a probability similar to that of D-labeled mrrm pentads; this observation requires further mechanistic clarification. Olefinic chain ends of polymers obtained with C₂H₄(thind)₂ZrCl₂-MAO from (E)- or (Z)-[1-D]propene deviate in their D-label distributions from the expected stereochemistry. Isomerization of the Zr-bound chain end, probably via a Zr-bound tertiary alkyl intermediate, thus contributes also to chain-growth termination. With the sterically hindered catalyst Me₂Si(2-Me-4-tBu-C₅H₂)₂ZrCl₂-MAO, almost all chain terminations appear to occur via chain-end isomerization, which does not lead to stereoerrors here. With the high-performance catalyst Me₂Si(2-Me-benz[e]indenyl)₂ZrCl₂-MAO, finally, no D-label is found in the mrrm and mmmm pentad signals; here, Zr-alkyl isomerization can apparently not compete with the high rate of olefin insertion.

Full text of DOI:10.1021/ja961157n

9024
J. Am. Chem. Soc. 1996, 118, 9024-9032
Zr-Alkyl Isomerization in ansa-Zirconocene-Catalyzed Olefin
Polymerizations. Contributions to Stereoerror Formation and
Chain Termination1
Margarete K. Leclerc and Hans H. Brintzinger*
Contribution from the Fakulta¨t fu¨r Chemie, UniVersita¨t Konstanz, D-78464 Konstanz, Germany
ReceiVed April 8, 1996X
Abstract: In polymers made from (E)- or (Z)-[1-D]propene with methyl alumoxane (MAO)-activated ansa-zirconocene
catalysts, signals of deuterium-labeled mrrm pentads document that stereoerrors arise mainly from an isomerization
of the Zr-bound chain end. Reduced D-atom redistribution in poly([2-D]propene) indicates a kinetic isotope effect
of kH/kD ≈ 3 for the â-H/D transfer associated with the isomerization reaction. In all poly([1- and [2-D]propenes)
studied, D-labeled mmmm pentads occur with a probability similar to that of D-labeled mrrm pentads; this observation
requires further mechanistic clarification. Olefinic chain ends of polymers obtained with C2H4(thind)2ZrCl2-MAO
from (E)- or (Z)-[1-D]propene deviate in their D-label distributions from the expected stereochemistry. Isomerization
of the Zr-bound chain end, probably Via a Zr-bound tertiary alkyl intermediate, thus contributes also to chain-growth
termination. With the sterically hindered catalyst Me2Si(2-Me-4-tBu-C5H2)2ZrCl2-MAO, almost all chain terminations
appear to occur Via chain-end isomerization, which does not lead to stereoerrors here. With the high-performance
catalyst Me2Si(2-Me-benz[e]indenyl)2ZrCl2-MAO, finally, no D-label is found in the mrrm and mmmm pentad signals;
here, Zr-alkyl isomerization can apparently not compete with the high rate of olefin insertion.
Introduction
R-agostic interactions control the stereochemistry of olefin-
insertions by stabilizing alternative transition states.7 This
hypothesis is supported by kinetic isotope effects associated with
the presence of a metal-bound R-CHD-group.8-11 Stereocontrol
by chain-segment orientation and by R-agostic interaction are
entirely compatible with each other and in fact mutually
supportive: Interaction of one or the other of the two H atoms
of the Zr-bound R-CH2 group with the Zr center gives two
orientations of the C(R)-C(â) segment, of which one is
sterically unencumbered and indeed just that predicted to yield
minimal van der Waals repulsions.4 The alternative ZrrH
interaction (which would give rise to a stereoirregular olefin
insertion) is sterically inaccessible for chiral ansa-zirconocene
complexes, in which the coordination sites are unilaterally
blocked by one â-substituent at each C5-ring ligand. This
explanation for the stereoselectivity of chiral ansa-metallocene
polymerization catalysts is supported by several recent theoreti-
cal studies by ab initio and local-density methods.12
Well-defined homogeneous ansa-zirconocene catalysts for the
stereoregular polymerization of R-olefins2 offer an opportunity
to clarify the mechanisms of polymer chain growth and, in
particular, the origins of its stereoselectivity. NMR studies by
Zambelli and co-workers3 on the stereochemistry of olefin
insertions into metal-bound 13CH3 and 13CH2CH3 groups and
molecular-mechanics studies by Corradini and co-workers4 on
ansa-metallocene catalysts have led to the notion of chain-
segment-mediated catalytic-site control: The enantiotopicity of
each coordination site at the catalyst center5 determines the
orientation of the last-inserted, metal-bound C(R)-C(â) chain
segment; the latter then controls the enantiofacial orientation
of the following olefin insertion, such that the olefin substituent
is placed trans to the C(R)-C(â) segment at the incipient C‚‚‚C
bond.6
This model of chain-segment-mediated catalytic-site stereo-
control has recently been complemented by the notion that
The rigidity of the three-membered ZrrH-C(R) ring and
of the adjoining four-membered cycle of the olefin-insertion
transition state would lead to the expectation that high (if not
essentially complete) selectivities for stereoregular olefin inser-
X Abstract published in AdVance ACS Abstracts, September 1, 1996.
(1) ansa-Metallocene derivatives, part 34; for part 33, see: Barsties, E.;
Schaible, S.; Prosenc, M. H.; Rief, U.; Ro¨ll, W.; Weyand, O.; Dorer, B.;
Brintzinger, H. H. J. Organomet. Chem. 1996, in press.
(2) Reviews: Aulbach, M.; Ku¨ber, F. Chem. i.u. Ztg. 1994, 28, 197.
Mo¨hring, P. C.; Coville, N. J. J. Organomet. 1994, 479, 1. Sinclair, K. B.;
Wilson, R. B. Chem. Ind. 1994, 857. Kaminsky, W. Catalysis Today 1994,
20, 257. Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt, R.; Rieger, B.;
Waymouth, R. M. Angew. Chem. Int. Ed. Engl. 1995, 34, 1143.
(3) Longo, P.; Grassi, A.; Pellecchia, C.; Zambelli, A. Macromolecules
1987, 20, 1015. Zambelli, A.; Pellecchia, C. Makromol. Chem., Macromol.
Symp. 1993, 66, 1.
(4) Corradini, P.; Guerra, G.; Vacatello, M.; Villani, V. Gazz. Chim. Ital.
1988, 118, 173. Corradini, P.; Guerra, G.; Cavallo, L.; Moscardi, G.;
Vacatello M. In Ziegler Catalysts; Fink, G., Mu¨lhaupt, R., Brintzinger, H.
H., Eds., Springer-Verlag: Berlin, 1995; p 237. Guerra, G.; Corradini, P.;
Cavallo, L.; Vacatello M. Makromol. Chem., Macromol. Symp. 1995, 89,
p 77.
(7) Laverty, D. T.; Rooney, J. J. J. Chem. Soc., Faraday Trans. 1, 1983,
79, 869; Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983, 250,
395; Brookhart, M.; Green, M. L. H.; Wong, L. Prog. Inorg. Chem. 1988,
36,1.
(8) Clawson, L.; Soto, J.; Buchwald, S. L.; Steigerwald, M. L.; Grubbs,
R. H. J. Am. Chem. Soc. 1985, 107, 3377. Coates, G. W.; Grubbs, R. H.
Acc. Chem. Res. 1996, 29, 85.
(9) Piers, W. E.; Bercaw, J. E. J. Am. Chem. Soc. 1990, 112, 9406.
Coughlin, E. B.; Bercaw, J. E. J. Am. Chem. Soc. 1992, 114, 7606. Burger,
B. J.; Cotter, W. D. Coughlin, E. B.; Chacon, S. T.; Hajela, S.; Herzog, T.
A.; Ko¨hn, R.; Mitchell, J. T.; Piers, W. E.; Shapiro, P. J.; Bercaw, J. E. In
Ziegler Catalysts; Fink, G., Mu¨lhaupt, R., Brintzinger, H. H., Eds.;
Springer-Verlag: Berlin, 1995; p 317.
(5) Farina, M.; Di Silvestro, G.; Terragni, A. Macromol. Chem. Phys.
1995, 196, 353.
(6) This “chain-segment-mediated stereocontrol” is to be distinguished
from “chain-end stereocontrol”, the latter being derived from alternative
configurations of the last-inserted unit, the former from its spatial orientation.
(10) Krauledat, H.; Brintzinger, H. H. Angew. Chem. Int. Ed. Engl. 1990,
29, 1412.
(11) Leclerc, M. K.; Brintzinger, H. H. J. Am. Chem. Soc 1995, 117,
1651.
S0002-7863(96)01157-2 CCC: $12.00 © 1996 American Chemical Society

Zr-Alkyl Isomerization in Olefin Polymerizations
J. Am. Chem. Soc., Vol. 118, No. 38, 1996 9025
Scheme 1
tion are found for all appropriately substituted ansa-zirconocene
catalysts. Quite in contrast to this, many chiral ansa-zirconocene
catalysts will give polypropene with high isotacticities (i.e. with
[mmmm] values above 90%) only at temperatures around or
below ambient; polymerizations at temperatures of 50 °C or
higher lead to strongly decreased stereoregularities, with
[mmmm] values down to ca. 50%.14-16
Exceptions in this regard are zirconocene catalysts with bulky
â-substituents, such as t-butyl or 1-methylcyclohexyl groups,
which produce polypropene with isotacticities of [mmmm] >
95% even at 80 °C, albeit at the expense of decreased polymer
chain lengths.16 Modern ansa-zirconocene catalysts of practical
utility, such as Me2Si(2-Me-benz[e]indenyl)2ZrCl2-MAO17,18 or
Me2Si(2-Me-4-arylindenyl)2ZrCl2-MAO,17 on the other hand,
combine high stereoselectivities (i.e. values of [mmmm] > 95%)
with high polymer chain lengths and also with very high
productivities.
mere presence or absence of a â-substituent in a particular
quadrant of the ansa-zirconocene complex determines the
stereochemical course of each olefin insertion.19 Important
insights with regard to the origin of stereoselectivity variations
among different chiral ansa-zirconocene catalysts were obtained
by Cipullo and Busico20,21 and by Resconi and co-workers:22
Decreased monomer concentrations were found to lead to
strongly diminished stereoregularities, especially at elevated
reaction temperatures. Catalysts of relatively low activity appear
to be distinctly more sensitive to decreased monomer concentra-
tions than high-activity catalysts such as Me2Si(2-Me-benz[e]-
indenyl)2ZrCl2-MAO.20 This effect of monomer concentrations
on stereoselectivities was ascribed to an epimerization at the
asymmetric methine center of the last-inserted, Zr-bound C3H6
unit. This reaction appears to compete with the insertion of
another olefin whenever the insertion rates are diminished, be
it by reduced monomer concentrations or by low activities of
the catalyst employed.
The strong stereoselectivity variations observed among these
catalysts do not appear to be compatible, prima facie, with such
a “binary” stereoselection model, which would predict that the
In studying the effects of R-deuteration on polymer-chain
growth11 we have obtained independent evidence for the
occurrence of such a chain-end isomerization: In the 13C NMR
spectra of polymers produced with chiral ansa-zirconocene
catalysts from cis- or trans-R-deuterated propenes, we observed
that the mrrm pentad signal, i.e. the signal due to stereoinverted
CH3 groups, is largely replaced by a triplet arising from
stereoinverted CH2D groups. The origin of these mrrm-framed
CH2D groups was tentatively ascribed to the rearrangement
represented in Scheme 1. In the following, we present further
evidence, from deuterium labeling patterns arising in the
polymerization of deuterated monomers with a number of typical
zirconocene catalysts, for the nature and the possible mecha-
nisms of this rearrangement reaction. We also present evidence
that isomerization of the Zr-bound chain end contributes not
only to the formation of stereoerrors but also to chain-growth
termination.
(12) (a) Yoshida, T.; Koga, N.; Morokuma, K.; Organometallics 1995,
14, 746; Yoshida, T.; Koga, N.; Morokuma, K. Organometallics 1996, 15,
766. (b) Hart, J. R.; Rappe´, A. K. J. Am. Chem. Soc. 1993, 115, 6159. (c)
Woo, T. K.; Fan, L.; Ziegler, T. Organometallics 1994, 13, 432. Ibid. 1994,
13, 2252. Ziegler Catalysts; Fink, G., Mu¨lhaupt, R., Brintzinger, H. H.,
Eds.; Springer-Verlag: Berlin, 1995; p 291. Lohrenz, J. C. W.; Woo, T.
K.; Ziegler, T. J. Am. Chem. Soc. 1995, 117, 12793. (d) Weiss, H.; Ehrig,
M.; Ahlrichs, R. J. Am. Chem. Soc. 1994, 116, 4919. Yet to be assessed is
the size of the energy increment by which an R-agostic ZrrH interaction
stabilizes the sterically more favorable transition state.13
(13) An estimate of ca. 20 kJ/mol for the R-agostic stabilization of the
olefininsertion transition state has been derived from an Extended-Hu¨ckel
study: Prosenc, M. H.; Janiak, C.; Brintzinger, H. H. Organometallics 1992,
11, 4036.
(14) Kaminsky, W.; Ku¨lper, K.; Brintzinger, H. H.; Wild, F. R. W. P.
Angew. Chem. Int. Ed. Engl. 1985, 24, 507.
(15) Rieger, B.; Chien, J. C. W. Polym. Bull. 1989, 21, 159. Rieger, B.;
Mu, X., Mallin, D. T., Rausch, M. D.; Chien, J. C. W. Macromolecules
1990, 23, 3559.
(16) Rieger, B.; Reinmuth, A.; Ro¨ll, W.; Brintzinger, H. H. J. Mol. Catal.
1993, 82, 67. Reinmuth, A. Dissertation, Universita¨t Konstanz, 1992.
(17) Spaleck, W.; Ku¨ber, F.; Winter, A.; Rohrmann, J.; Bachmann, B.;
Antberg, M.; Dolle, V.; Paulus, E. F. Organometallics 1994, 13, 954.
Spaleck, W.; Aulbach, M.; Bachmann, B.; Ku¨ber, F.; Winter, A. Makromol.
Chem., Macromol. Symp. 1995, 89, 237.
(19) Pino, P.; Cioni, P.; Wei, J. J. Am. Chem. Soc. 1987, 109, 6189.
(20) Busico, V.; Cipullo, R. J. Am. Chem. Soc. 1994, 116, 9329. J.
Organomet. Chem. 1995, 497, 113. Busico, V. Makromol. Chem Macromol.
Symp. 1995, 89, 277.
(18) (a) Stehling, U.; Diebold, J.; Kirsten, R.; Ro¨ll, W.; Brintzinger, H.
H.; Ju¨ngling, S.; Mu¨lhaupt, R.; Langhauser, F. Organometallics 1994, 13,
964. (b) Ju¨ngling, S.; Mu¨lhaupt, R.; Stehling, U.; Brintzinger, H. H.; Fischer,
D.; Langhauser, F. J. Polym. Sci. 1995, A-33, 1305.
(21) Busico, V.; Caporaso, L.; Cipullo, R.; Landriani, L.; Angelini, G.;
Margonelli,A.; Segre A. L. J. Am. Chem. Soc. 1996, 118, 2105.
(22) Resconi, L.; Fait, A.; Piemontesi, F.; Colonnesi, M.; Rychlicki, H.;
Zeigler, R. Macromolecules 1995, 28, 6667.

9026 J. Am. Chem. Soc., Vol. 118, No. 38, 1996
Leclerc and Brintzinger
Table 1. Degree of Polymerization, PN, and Pentad Distribution
(in %) for Polypropene Obtained from (E)-[1-D]- and
(Z)-[1-D]Propene (1 bar, 0.33 mol/La) with Catalysts 1-4c
catalyst/monomer
(polym, temp °C)
mrrm
D-mrrm
PN mmmm mmrr (rr-CH3) (rr-CH2D)
1/(E)-[1-D] (30)
1/(Z)-[1-D] (30)
1/(E)-[1-D] (53)
1/(Z)-[1-D] (53)
2/(E)-[1-D] (50)
2/(Z)-[1-D] (50)
3/(E)-[1-D] (50)
3/(Z)-[1-D] (50)
4/(E)-[1-D] (54)
4/(Z)-[1-D] (54)
134
101
78
45
59
46
124
94
b
82
79
52
47
87
7.2
8.4
19.2
21.2
5.2
1.4
1.2
2.7
3.5
0.8
2.2
3.0
6.9
7.1
1.8
1.8
<0.5
<0.5
0
86
5.6
1.0
>98
>98
95
<1
<1
2.0
2.0
<0.5
<0.5
1.0
b
95
1.0
0
a Solubility data: ref 18b. b Not determinable by 13C NMR; PN
700 ( 50 by GPC. c Polymerization conditions: [Zr] ) 1.5‚10-5
for catalysts 1-3, 2.5‚10-6 M for catalyst 4; [Al] ) 0.045 M.
≈
M
Table 2. Degree of Polymerization, PN, and Pentad Distribution
(in %) for Poly((E)-[1-D]propene) Obtained with C2H4(thind)2ZrCl2/
MAO (Catalyst 1) at 30 °C at Different Monomer Pressuresa
p(C3H5D), c(C3H5D),
mrrm
D-mrrm
bar
mol/Lc
PN mmmm mmrr (rr-CH3) (rr-CH2D)
0.9b
2c
0.42
0.99
1.58
140
290
440
73
80
84
11
8.0
6.4
1.4
1.2
1.2
3.9
2.8
2.0
3c
a Other conditions are as in Table 1. b Total pressure, including the
vapor pressure of toluene at 30 °C. c Partial pressure of C3H5D, added
to the catalyst solution under 1 bar of Ar. d c(C3H5D) ≈ (p(C3H5D)/
bar - 0.1)0.53 mol/L (from data in refs 18b and 22).
Additional signals are apparent in the 13C-NMR spectra of
poly((E)- and (Z)-[1-D]propene) close to the mmmm pentad
signal. The DEPT spectrum shows three signals of equal
intensity, with an equidistant spacing of 19 Hz and with inverted
sign, centered 0.3 ppm upfield of the mmmm pentad signal
(Figure 1). These characteristics indicate that poly((E)- and (Z)-
[1-D]propene) contain also stereoregular, mm-centered CH2D
units. The intensities of these D-mmmm signals are similar to
those of the D-mrrm signals discussed above. These and other
analytical data of the polymers obtained from (E)- and (Z)-[1-
D]propene (Table 1) will be further discussed below.
Figure 1. 13C NMR spectra (pentad region) of poly((Z)-[1-D]propene)
produced with C2H4(thind)2ZrCl2-MAO at 50 °C, measured at 150 MHz
on a Bruker DRX 600 spectrometer in 1H broad-band decoupled (top)
and DEPT 135 (bottom) NMR modes. Asterisks denote CH2 and CH3
signals associated with n-propyl chain ends and 3,1-misinsertions,
respectively.
Results and Discussion
2. Effects of Monomer Concentration and Catalyst Com-
position. In order to ascertain the origin of the CH2D groups
found in these polymers, we have studied the effects of changing
monomer concentrations on the D-label distribution. Polymers
obtained with the catalyst system C2H4(thind)2ZrCl2-MAO
(catalyst 1) (30 °C) at (E)-[1-D]propene pressures ranging from
0.9 to 3 bar yield the data listed in Table 2.
In accord with earlier observations on undeuterated poly-
propenes,20-22,24 isotacticities and chain lengths are found to
increase with increasing CHDdCHCH3 pressure. The increase
in isotacticity with increasing monomer pressure is clearly due
to a decrease of the fraction of stereoerrors associated with the
rr-CH2D triplet centered at 19.35 ppm. The fraction of
undeuterated rr-triads, on the other hand, is found to be
essentially constant at [mrrm] ≈ 1.2-1.4%, independent of the
propene pressure. This is in accord with the view20-22 that a
relatively low fraction of “intrinsic” stereoerrors is caused by
faulty enantiofacial olefin insertions and that these intrinsic
errors are superimposed by increasing fractions of isomerization-
induced stereoerrors when olefin insertion rates are decreased
by reduced monomer concentrations.
1. D-Label Distribution in the Poly([1-D]propene) Chain.
When (E)- or (Z)-[1-D]propene is polymerized with the catalyst
system C2H4(thind)2ZrCl2-MAO, the isotactic polymer chains
contain, as expected, CHD instead of CH2 groups in their
polymer backbone, recognizable by 13C NMR triplet signals in
the region of 45.4-45.8 ppm. Additional changes are apparent
in the pentad region of the 13C NMR spectra of these
polymers: Instead of the catalytic-site controlled 2:2:1 distribu-
tion of mmmr, mmrr, and mrrm pentad signals observed with
undeuterated propene, one finds a mrrm pentad signal with an
intensity reduced to about one-quarter of the expected size and
slightly upfield from it, a 1:1:1 triplet with J ) 19 Hz, which
signals the presence of a D atom in stereoinverted methyl groups
(Figure 1).
Support for an interchange of CHD and mm-centered CH3
groups for CH2 and rr-centered CH2D groups (cf. Scheme 1),
is derived from a splitting of the mmrr signal into two barely
resolved signals of equal intensity in poly((E)-[1-D]propene)
and by its broadening in poly((Z)-[1-D]propene):23 The rear-
rangement represented in Scheme 1 implies that the two mr-
centered CH3 groups adjacent to a rr-centered CH2D group are
inequivalent, the one closer to the metal center being flanked
by a CH2 and a CHD group rather than by two CHD groups.
Further insights with regard to the competition between olefin
insertion and chain-end isomerization can be obtained from a
(24) (a) Chien, J. C. W.; Wang, B. P. J. Polym. Sci. 1990, A 28, 15. (b)
Fischer, D. Dissertation, Universita¨t Freiburg 1992. (c) Dro¨gemu¨ller, H.;
Niedoba, S.; Kaminsky, W. Polym. React. Eng. 1986, 299.
(23) Different gauche effects caused by the unequal configurations of
the flanking CHD groups are the most likely cause of this difference.

Zr-Alkyl Isomerization in Olefin Polymerizations
J. Am. Chem. Soc., Vol. 118, No. 38, 1996 9027
Table 3. Degree of Polymerization, PN, and Pentad Distribution
(in %) for Undeuterated Polypropene and Poly([2-D]propene)
Obtained with Catalyst 1 at 30 and 50 °Ca
to be entirely due to a diminished rate of isomerization: While
the relative integrals of the undeuterated mrrm pentad signals
due to “intrinsic” stereoerrors in poly([2-D]propene) are, with
1.0-1.4%, quite close to those observed in poly([1-D]propene)
obtained under the same conditions, a substantially reduced
probability of isomerization-induced stereoerrors is apparent
from the reduced fraction of D-mrrm pentads: In poly([2-D]-
propene) prepared with catalyst 1 at 30 °C for instance, the
triplet signal due to stereoinverted CH2D groups is hardly
detectable; accordingly, the isotacticity almost reaches its
“intrinsic” value of [mmmm] ≈ 95% here. For poly([2-D]-
propene) prepared at 50 °C, the rate of chain-end isomerization
appears to be only about one-third of that for undeuterated
polypropene or poly([1-D]propene) obtained with catalyst 1 at
the same temperature. These data lead to the estimate that an
isotope effect of kH/kD ≈ 3 is associated with the breakage of
the â-C-H/D bond in the rate-limiting step of the isomerization
process.25
An isotope effect is also evident with regard to the reactions
which control the chain lengths of these polymers: Both at 30
and 50 °C, the mean degree of polymerization, PN, of poly([2-
D]propene) is higher by a factor of 1.5-1.6 than that of
undeuterated polypropene (Table 4). This isotope effect is
undoubtedly connected with the rate-determing step of the chain
termination reaction (cf. section 4).
4. D-Label Distribution in the Unsaturated Chain End.
The control of chain lengths is an important aspect for the
practical utilization of ansa-zirconocene-based polymerization
catalysts; it would thus appear useful to clarify the details of
the underlying chain-termination reactions. We have thus
extended our present D-label distribution study also to an
investigation of the reaction paths which generate the unsaturated
chain ends of these polymers.
monomer
mrrm
D-mrrm
(polym temp, °C) PN mmmm mmrr (rr-CH3) (rr-CH2D)
D0-propene (30)
2-D-propene (30) 188
D0-propene (50) 73
2-D-propene (50) 110
124
84
95
61
84
6.4
2.0
15.5
6.4
3.2
1.0
8.0
1.0
0
<0.5
0
2.2
a Other conditions are as in Table 1.
comparison of different ansa-zirconocene catalyst (Table 1):
essentially analogous isotactities and D-label distributions as
with catalyst 1 are found with Me2Si(2,4-Me2-C5H2)2ZrCl2-
MAO (catalyst 2). The catalyst system Me2Si(2-Me-4-tBu-
C5H2)2ZrCl2-MAO (catalyst 3), on the other hand, produces
polymers which are so highly isotactic that it is barely possible
to detect any error pentad signals.16 Accordingly, no triplet
signal due to mrrm-CH2D groups is discernible in poly((E)- or
(Z)-[1-D]propene) made with this catalyst.
Catalyst 4, Me2Si(2-Me-benz[e]indenyl)2ZrCl2-MAO, finally,
yields polymers with an isotacticity of [mmmm] ≈ 95% even
at higher polymerization temperatures. In poly((E)- or (Z)-[1-
D]propene) prepared with this catalyst, the mrrm error pentad
signals are clearly discernible, however, and in fact indistin-
guishable with regard to its relative intensity from that in
undeuterated polypropene prepared with catalyst 4; we can thus
exclude that any significant fraction of the rr-centered CH3
groups of these polymers carry a deuterium atom. In accord
with the isotacity data obtained by Cipullo and Busico,20 olefin
insertion appears to occur so fast in this highly active catalyst
system even at a low monomer pressure of 0.5 bar that chain-
end isomerization cannot significantly compete with it.
DEPT NMR spectra of poly((E)- or (Z)-[1-D]propene), made
with either catalyst 3 or catalyst 4, do not show any D-labeled
satellites of the mmmm pentad signal. We regard this as
evidence that the mm-centered CH2D groups of polymers made
with catalyst 1 arise only in connection with the chain-end
isomerization discussed above and not, for example, by an
isomerization of free [1-D]propene to [3-D]propene.
In polypropene prepared with C2H4(thind)2ZrCl2-MAO or
related catalyst systems, 2-propenyl units are the dominant end
groups resulting from chain termination;26 they arise by â-H
transfer either to the metal center or to a coordinated olefin
(Scheme 2).18,27 In either case, the cis-stereochemistry of olefin
insertion and â-H transfer reactions predicts that poly((E)-[1-
D]propene) should have a [1-D]-2-propenyl end group with its
D-atom trans to the polymer chain (A), while poly((Z)-[1-D]-
propene) is expected to have a propenyl end group with cis-
3. Polymers obtained from [2-D]propene. Polymerizations
of â-deuterated propene with C2H4(thind)2ZrCl2-MAO (catalyst
1) have recently been studied by Busico and co-workers.21 In
accord with their observations, we find that poly([2-D]propene)
has a substantially higher isotacity than undeuterated polypro-
(25) A primary KIE of kH/kD ) 2.0 has been observed for the â-H transfer
reaction Cp2ZrCH2CH2D(NCCH3) f Cp2Zr(H,D)(NCCH3) + CH2CH-
(H,D): Alelyunas, Y. W.; Guo, Z.; LaPointe, R. E.; Jordan, R. F.
Organometallics 1993, 12, 544. Related isotope effects: Evans, J.; Schwartz,
J.; Urquhart, P. W., J. Organomet. Chem. 1974, 81, C37. Alibrandi, G.;
Scolaro, L. M.; Minniti, D.; Romeo, R. Inorg. Chem. 1990, 29, 3467.
Negishi. E. I; Nguyen, T.; Maye, J. P. Choueiri, D. Suzuki, N.; Takahashi,
T. Chem. Lett. 1992, 2367.
(26) Not detectable in the polymers considered here are end groups
derived from â-methyl transfer (Hajela, S.; Bercaw, J. E. Organometallics
1994, 13, 1147. Resconi, L.; Jones, R. L.; Rheingold, A. L.; Yap, G. P. A.
Organometallics 1996, 15, 998; Horton, A. D. Organometallics 1996, 15,
2675), from C-H activation of the monomer (Siedle, A. R.; Lamanna, W.
M.; Newmark, R. A.; Stevens, J.; Richardson, D. E.; Ryan, M. Makromol.
Chem. Macromol. Symp. 1993, 66, 215), or from â-H transfer from a
secondary unit (Shiono, T.; Soga, K. Macromolecules 1992, 25, 3356; refs
18b and 22). Isopropyl signals arising from alkyl exchange between Zr
and Al centers (Resconi, L.; Bossi, S.; Abis, L. Macromolecules 1990, 23,
4489; Mogstad, A. L.; Waymouth, R. M. Macromolecules 1992, 25, 2282;
refs 22, 24a) occur with an intensity of ca. 30% of n-propyl end groups in
polymers made with catalyst 1 and about equally frequent as n-propyl end
groups in polymers made with catalyst 3 (cf. ref 16). Occasionally,
2-propenyl end groups are observed to isomerize to 1-isobutenyl groups
(presumably by traces of acid) during 13C NMR measurement in C2D4Cl2
at temperatures above 100 °C.
pene produced under identical conditions (Table 3). In the 13
C
NMR spectra of poly([2-D]propene) one observes, besides the
expected methine C-D signals in the range of 27.8-28.3 ppm,
again the triplet at 19.35 ppm, associated with stereoinverted
CH2D groups. The â-D atom in a ZrCH2CD(CH3)-pol unit is
thus obviously shifted to the newly formed methyl group during
the chain-end isomerization process. As noted by Busico and
co-workers, this can be taken as evidence that the isomerization
reaction proceeds Via a tertiary intermediate of the type ZrCMe2-
pol.21
In addition to these rr-centered CH2D groups, however, we
also observe heres as in poly((E)- and (Z)-[1-D]propene)san
additional negative DEPT triplet of similar intensity 0.3 ppm
upfield of the mmmm pentad. Stereoregular, mm-centerd CH2D
groups are thus clearly present also in these poly([2-D]propene)
samples. As discussed in more detail below, this observation
is not easily explicable in terms of the reaction mechanisms
commonly considered for metal-alkyl isomerizations.
As noted above (Table 4), the isotacticity of poly([2-D]-
propene) is substantially higher than that of undeuterated
polypropene or of poly((E)- or (Z)-[1-D]propene). This appears
(27) Tsutsui, T.; Mizuno, A.; Kashiwa, N. Polymer 1989, 30, 428. Busico,
V.; Cipullo, R.; Corradini, P. Makromol. Chem. Rapid Commun. 1993, 14,
97. Busico, V.; Cipullo, R.; Chadwick, J. C.; Modder, J. F.; Sudmeijer, O.
Macromolecules 1994, 27, 7538.

9028 J. Am. Chem. Soc., Vol. 118, No. 38, 1996
Leclerc and Brintzinger
Table 4. End Group Distribution (in %) in Polymers Obtained from (E)-[1-D]-, (Z)-[1-D]-, and [2-D]Propene with Catalysts 1 and 3a
integral ratios of signals at various ppm
end group ratios
monom pressure,
catalyst/monomer
ar (polym temp, °C)
4.72
4.70
4.64
4.62
A
B
C
1/(E)-[1-D]
1/(E)-[1-D]
1/(Z)-[1-D]
1/2-D
1/2-D
3/(E)-[1-D]
3/(Z)-[1-D]
1 (50)
2 (50)
1 (50)
1 (30)
1 (53)
0.75 (50)
0.75 (50)
0.99
0.98
0.98
1
1
0.99
0.96
0.28
0.31
0.88
<.03
0.1
0.24
0.19
1
1
1
1
1
1
1
0.70
0.80
0.27
<.03
0.1
0.35
0.17
36
38
13
<3
8
22
13
14
15
41
<3
8
15
14
50
47
46
>95
84
63
73
a Conditions as in Table 1.
Scheme 2
deuterated geometry (B). The HDC terminus of both end groups
appears in the 13C NMR spectra of poly((E)- or (Z)-[1-D]-
propene) as a 1:1:1 triplet with J ) 19.6 Hz at 111.0 ppm. In
addition, however, the signal of the undeuterated CH2 terminus
is also apparent at 111.3 ppm. Some D-atom redistribution is
thus obviously involved also in the generation of these unsatur-
ated end groups.
In order to assign possible isotopomers, we have undertaken
an analysis of the 1H NMR signals associated with 2-propenyl
end groups. Two-dimensional HMQC and ROESY spectra of
undeuterated polypropene and of poly((E)- and (Z)-[1-D]-
propene) samples, discussed in detail in the Experimental
Section, lead to the assignment summarized in Figure 2. For
each of the protons at the CH2 terminus, the signal is shifted
upfield by ca. 0.02 ppm, to 4.70 and 4.62 ppm, when a D atom
occupies the opposite terminal position (Figure 3). The ratios
of the different end group isotopomers can thus be estimated
from the relative intensities of these signals: The weights of
end groups A and B are measured by the integrals of the signals
at 4.62 and 4.70, respectively, while that of end group C is
obtained from the averaged integral of the signals at 4.64 and
4.72 ppm.
Figure 2. 1H NMR signal assignments (δ in ppm) for unsaturated
2-propenyl chain ends (cf. Experimental Section).
all three types of unsaturated end groups, i.e. trans- and cis-
HDCdCMe-pol (A and B) as well as H2CdCMe-pol (C). Most
of the D-label content missing in the terminal positions of
H2CdCMe-pol appears to be associated with the methyl group
of this unit, as indicated by a partial replacement of the 13C
NMR signal of the 2-propenyl methyl group at 22.5 ppm by a
triplet with J ≈ 19 Hz, centered at 22.2 ppm. The D-label
content of the ZrCHDCHMe unit thus appears to be mostly
redistributed among its three C atoms during â-H transfer, rather
than being lost from it.28
The 1H NMR spectra of polymers obtained from either (E)-
or (Z)-[1-D]propene (Figure 3) show that both polymers contain
The proportions of the three isomeric end groups A, B, and
C are found to be different for polymers derived from (E)-[1-

Zr-Alkyl Isomerization in Olefin Polymerizations
J. Am. Chem. Soc., Vol. 118, No. 38, 1996 9029
Scheme 3
In the following, we attempt to analyze the experimentally
observed end group distributions in terms of the rate constants
indicated in Scheme 3: In poly((E)-[1-D]propene), end group
A arises from the normally prevalent meso-configurated inter-
mediate {m} (by the selective reaction path) with rate constant
ksel and from the tertiary intermediate {t} (by the nonselective
path) with rate constant knsl. End group B, on the other hand,
arises only from {t}, with a rate constant which must be equal
to knsl as the two H atoms of the CH2D group will be eliminated
with equal probability. We can thus define a “selectivity ratio”
SR ) ([A] - [B])/([A] + [B] + [C]) for the unsaturated end
groups in poly((E)-[1-D]propene); for poly((Z)-[1-D]propene)
this “selectivity ratio” is SR ) ([B] - [A])/([A] + [B] + [C]).
From lines 1-4 in Table 4, we derive an estimate of SR ≈
1:(4-5) for this ratio for catalyst 1 at 50 °C. Of all chain
terminations, only ca. 20% thus appear to occur by direct â-H
transfer. This ratio does not appear to vary substantially with
propene pressure.
Figure 3. 1H NMR spectra (olefin region) of poly((E)-[1-D]propene)
(top) and poly((Z)-[1-D]propene) (bottom). Assignments according to
Scheme 3.
End group C, finally, arises mainly from the tertiary
29
intermediate {t}, with a rate constant knsl′
.
For the latter, one
D]propene and from (Z)-[1-D]propene: As expected from the
reaction sequences outlined in Scheme 2, end group A is more
frequent with poly((E)-[1-D]propene) while B prevails with
poly((Z)-[1-D]propene) (Table 4).
can estimate, from a minor secondary isotope effect of 1.05-
1.1, that the reaction is likely to favor transfer of a â-H atom
from a CH3 over that from a CH2D group30 and, from a statistical
choice of three H atoms, a ratio of knsl′/knsl ≈ 3.2. The close
accord of this estimate with the ratio of [C]/[B] ≈ (3.1-3.5)
found for poly((E)-[1-D]propene) and of [C]/[A] ≈ 3.5 for poly-
((Z)-[1-D]propene) (cf. Table 4) confirms the chain-termination
sequence represented in Scheme 3.
The validity of this reaction sequence is further supported
by end group distributions for poly([2-D]propene) (Table 4):
At 30 °C, where no isomerization is apparent for poly([2-D]-
propene), deuterium is not incorporated to any noticeable degree
into the 2-propenyl end group either. At 50 °C, where deuterium
Two different reaction paths for â-H transfer apparently
compete with each other: One leads stereoselectively to the
expected products while the other involves a scrambling of
D-atoms within the last-inserted unit. At the propene pressures
studied, the stereoselective â-H transfer path appears to account
only for a minor part of the unsatured end groups. The dominant
D-scrambling reaction path, on the other hand, appears to be
related to the isomerization which causes the formation of
stereoerrors: An intermediate ZrC(CH3)(CH2D)-pol, arising by
a migration of Zr to the tertiary C atom, would be expected to
generate end groups A, B, and C with comparable probability
(Scheme 3).
(29) The low fraction of intermediates {d-r} and {d-m}, multiplied by
the relatively small rate constants knsl′ and ksel make any additional
contributions to the formation of C negligeable.
(28) We find no signs that extra D atoms would be incorporated into
position 2 of the n-propyl end group by Zr-D units arising from abstraction
of â-D atoms.
(30) Secondary kinetic isotope effects on â-H transfer and related
reactions: Schro¨der, D.; Schwarz, H. Chimia 1989, 43, 317. Poirier, R. A.;
Wang, Y.; Westaway, K. C. J. Am. Chem. Soc. 1994, 116, 2526.

9030 J. Am. Chem. Soc., Vol. 118, No. 38, 1996
Leclerc and Brintzinger
is transferred into rr-centered CH2D groups, it also appears in
the terminal positions of the 2-propenyl end group. Nonselective
formation of end groups A and B can occur here only from
{t}, with rate constant knsl. Comparison with the data for (E)-
and (Z)-[1-D]propene shows that isomerization of {m} to {t}
is less frequent than in poly([1-D]propene) by a factor of ca.
1.7, undoubtedly due to a retardation of â-D Vs â-H transfer.25
A related kinetic isotope effect is probably also responsible
for the increased PN values of poly([2-D]propene) mentioned
above. This increase in PN could in principle reflect an increase
in kins or a decrease in ksel and/or kiso for the 2-deuterated olefin.
Of these, kins (the rate constant of olefin insertion) is unlikely
to be influenced to any substantial degree by â-deuteration of
the monomer.30 The rate constants which control the rate of
chain termination, ksel and/or kiso, thus appear to be associated
with a kinetic isotope effect of kH/kD ) 1.5-2, as one would
expect if a â-H transfer limits the rates of these reactions.25
The increase of the mean degree of polymerization, PN, with
increasing monomer concentrations, [M], observed for catalyst
1 (Table 2, cf. refs 20, 22, and 24), is explicable by a rate of
polymer growth proportional to [M], VP ) kins[M], and a
monomer-independent rate of chain termination by the pre-
dominant route Via intermediate {t}. This would imply that
isomerization of {m} to {t} occurs by way of a species which
does not contain a coordinated olefin, presumably Via â-H
transfer to the metal (vide infra). Because of the still unclarified
significance of a higher-order dependence of the rate of polymer
growth on [M] (VP ≈ kins[M]n with n ≈ 1.5-2) reported for
this and related catalysts,18,22 the numbers of olefin molecules
participating in each of the competing reactions in Scheme 3
will eventually require further elaboration.
relative to chain growth by a factor which is ca. 20 times larger
than for catalyst 1.
The further fate of intermediate {t} is also different from
that in catalyst 1 where isomerization of {t} to {D-r} or {D-
m} occurs 10-15 times more frequently than generation of
chain ends A, B, or C: Catalyst 3 generates almost no
stereoerrors at all (Table 1) but high proportions of nonstereo-
selective end groups (Table 4). Any intermediate {t} formed
with catalyst 3 thus appears to undergo chain termination
exclusively, rather than further isomerization to a {D-r}- or {D-
m}-configurated Zr-alkyl unit. Again, the bulky ligand sub-
stituents appear to impede chain-end isomerization relative to
competing reaction modes.
In catalyst 4, finally, intermediate {m} appears to escape from
â-H transfer and chain-end isomerization, probably due to an
exceptionally high rate of olefin insertion. It remains to be
understood through which mechanisms different complex
structures affect the relative rates of competing reaction modes
of intermediates {m} and {t}.
Conclusions
As shown above, stereoerror distributions as well as polymer
chain lengths of polyolefins produced by a particular ansa-
zirconocene catalyst will be determined by the rate of chain-
end isomerizations, relative to that of olefin insertion and of
chain-termination by â-H transfer. It would thus be of
substantial interest to clarify the mechanisms and rate-limiting
factors of these isomerization reactions.
The mechanism most frequently considered for an isomer-
ization of metal-alkyl units is â-H elimination to a metal-
hydride-olefin complex, rotation of the olefin ligand around
the metal-olefin bond, and reinsertion to form the rearranged
metal-alkyl unit.31 This mechanism has been proposed to
account, for example, for the rearrangement of 2,1-regioinverted
units to n,1-units, which competes with olefin insertion into the
secondary Zr-alkyl bond,15,20,32 as well as for related migrations
of Ni and Pd centers along metal-bound polymer chains in the
formation of 2,ω-polyolefins and branched ethene polymers.33,34
Busico and co-workers have proposed that such a reaction
sequence is operative also for the chain-end isomerizations
discussed here (Scheme 4).20,21
This reaction sequence would plausibly explain the formation
of stereoinverted D-rr units in poly([1- and [2-D]propene), as
well as the kinetic isotope effect caused by a â-D atom at the
tertiary carbon center. It is not apparent, however, how this
reaction sequence could explain the occurrence of stereoregular,
mm-centered CH2D groups (D-mm units) observed in poly-
((E)- and (Z)-[1-D]propene) as well as in poly([2-D]propene):
for this rearrangement, the Zr-H fragment would have to move
from one enantioface of the olefin ligand H2CdC(CH2D)-pol
to the other (Scheme 4).35 Once lost from the Zr-center,
Phenomenologically, anyhow, we can delineate the fate of
the preponderant intermediate {m}, e.g. at 50 °C, if we take
the reciprocal mean degree of polymerization, (PN)-1, as the
frequency of chain terminations and the“'nonselectivity” ratio
NR ) 1 - SR ) ([C] + 2[B])/([A] + [B] + [C]) ≈ 0.75 as
the fraction of unsaturated end groups arising Via {t}. From
the frequency of stereoselective end groups (SR(PN)-1) on the
one hand and from the sum of the frequencies of nonstereose-
lective end groups (NR(PN)-1) and of D-rr and (equally
frequent) D-mm units on the other (Tables 1 and 4), we estimate
that {m} is channelled toward stereoselective chain termination,
toward isomerization, and toward chain growth in a ratio of
ksel:kiso:{kins[M]} ≈ (0.028 ( 0.007):1:{(16.7 ( 0.5)[M]}.
Isomerization to {t} thus appears to occur 30-40 times more
frequently than chain termination by direct â-H transfer.
As for the further fate of intermediate {t}, alternative chain
terminations Via â-H transfer from the CH3 and CH2D groups
and chain-end isomerization to intermediates {D-r} and {D-
m} compete with each other. From the data in Tables 1 and 4
we can estimate for the rate constants knsl, knsl′, and kiso′swhich
govern the distribution of {t} toward end groups A and B,
toward C, and toward intermediates {D-r} and {D-m},
respectivelysa ratio of knsl:knsl′:kiso′ ≈ 1:(3.5 ( 0.1):(32 ( 8)
for catalyst 1. The tertiary intermediate {t} will thus give rise
to intermediates {D-r} and {D-m} 10-15 times more frequently
than to an unsaturated chain end.
(31) Cramer, R. Acc. Chem. Res. 1968, 1, 186. Hartley, F. R. Chem.
ReV. 1969, 69, 799. Whitesides, G. M., Gaasch, J. F. Stedronsky, E. R. J.
Am. Chem. Soc. 1972, 94, 5258.
(32) Soga, K.; Shiono, T.; Takemura, S.; Kaminsky, W. Makromol.
Chem., Rapid Commun. 1987, 8, 305. Grassi, A.; Ammendola, P.; Longo,
P.; Albizzati, E.; Resconi, L.; Mazzocchi, R. Gazz. Chim. Ital. 1988, 118,
539. Tsutsui, T.; Mizuno, A.; Kashiwa, N. Makromol. Chem. 1989, 190,
1177. Asakura, T.; Nakayama, N.; Demura, M.; Asano, A. Macromolecules
1992, 25, 4876.
(33) Mo¨hring, V. M.; Fink, G. Angew. Chem. Int. Ed. Engl. 1985, 24,
1001.
(34) Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc.
1995, 117, 6414.
For Me2Si(2-Me-4-tBu-C5H2)2ZrCl2-MAO (catalyst 3), the
difference between [A] and [B] is too small for a reliable
determination of ksel. For this catalyst we can assume that chain
termination almost always occurs Via chain-end isomerization;
the data in Table 1 thus lead to the estimate that intermediate
{m} is channelled here to {t} and to chain growth in a ratio of
kiso:(kins[M]) ≈ 1:{(320 ( 50)[M]}. In this sterically encum-
bered catalyst system, isomerization appears to be disfavored
(35) To test for the possibility that isomerization of the monomer,
CHDdCHCH3 f CH2DCHdCH2, gives rise to D-mm units, we have
conducted a polymerization with catalyst 1/MAO (conditions as in Table
1) in which the pressure of (Z)-[1-D]propene was allowed to fall below 1

Zr-Alkyl Isomerization in Olefin Polymerizations
J. Am. Chem. Soc., Vol. 118, No. 38, 1996 9031
Scheme 4
-10 °C. After stirring for 3 h, the mixture was filtered and the filtrate
evaporated to dryness. The resulting white solid (25 g, 0.527 mol)
was suspended in 150 mL of toluene. At 0 °C, 22 g (0.67 mol) of
MeOD was added; the gas evolved was condensed through a cold trap
(-50 °C) into a storage vessel at -196 °C to yield 12 g (0.279 mol,
39%). The 2H-NMR spectrum showed that the product contained 98%
(E)-[1-D]propene and 2% (Z)-[1-D]propene.
however, the disubstituted olefin H2CdC(CH2D)-pol would
hardly be capable of competing with the large excess of propene
present for reinsertion into the Zr-H unit.36
Equally puzzling is the question why isomerization of
intermediate {t}, while producing equal fractions of D-rr and
D-mm units, does not also lead to equal fractions of undeuter-
ated rr (as well as mm) units. As discussed above, undeuterated
rr units are found only in minor amounts and appear to be
derived from faulty olefin insertions rather than from chain-
end isomerizations. Substantial further experimental and theo-
retical studies are obviously required to ascertain through which
reaction sequences these chain-end isomerizations do actually
proceed.
[2-D]Propene. 2-Bromopropene (62 g, 0.52 mol) was added to
freshly cut lithium metal (13 g, 1.87 mol) in 600 mL of ether at 0 °C
under exclusion of light. After stirring for 2 h, the mixture was filtered,
the filtrate evaporated to dryness, and the solid residue suspended in
150 mL of toluene. At 0 °C, 29 g (0.88 mol) of MeOD was added;
the gas evolved was collected through a cold trap at -50 °C in a storage
vessel kept at -196 °C to yield 17 g (0.395 mol, 76%). The 1H-NMR
spectrum showed complete deuteration at C(2) (δ ) 5.00 ppm for cis-
HCHdCDMe, 4.92 for trans-HCHdCDMe, and 1.70 for H2CdCDCH3).
Experimental Section
Polymerizations and Polymer Analysis. For a typical experiment,
2 mL of a stock solution of the ansa-zirconocene dichloride studied (c
) 1.5‚10-4 M in toluene) was added, under an Ar atmosphere, to 17
mL of toluene containing 0.6 mL of a 10% solution of MAO (Witco,
MG ≈ 900) in toluene in a 50-mL Schlenk vessel; the mixture was
thermostated at 30 °C. After evacuating the vessel to the vapor pressure
of toluene, (E)- or (Z)-[1-D]propene or -[2-D]propene was admitted to
the specified total pressure. The pressure was monitored and held
constant by further additions of monomer by way of an electronically
controlled magnetic valve. After 60 min, the polymerization was
quenched by addition of 50 mL of methanol and 0.5 mL of concentrated
hydrochloric acid. The polymer (typical yield ca. 200 mg) was collected
on a filter paper and dried overnight in the air. Polymerization at
elevated pressures were carried out analogously in a 100-mL autoclave,
but without evacuation of the Ar atmosphere prior to the admission of
C3H5D; the partial pressure of the latter is indicated in these cases in
Table 2.
To determine the polymer microstructure, 13C and 1H NMR spectra
were recorded on a JEOL JNM GX400 or Bruker DRX 600 spectrom-
eter. The more soluble polymers with shorter chain lengths, obtained
at 50 °C, were measured in CDCl3 solutions at 303 K to avoid any
extraneous chain-end isomerization, e.g. by traces of acid, which is
occasionally observed during NMR measurements at temperatures
above 100 °C. The less soluble, long-chain polymers were measured
in C2D2Cl4 solution at 373 K. Chemical shifts are referenced to residual
protons in deuterated solvents (tetrachloroethane-d2, δ ) 5.98 ppm;
CDCl3, δ ) 7.24 ppm) for 1H NMR and to the characteristic triplets of
tetrachloroethane-d2 at δ ) 74.1 ppm and of CDCl3 at δ ) 77.0 ppm
for 13C NMR. Pentad signals were assigned according to literature
General. All operations were performed in an inert atmosphere with
rigorous exclusion of oxygen and moisture using Schlenk, vacuum-
line, or glovebox techniques. Solvents were thoroughly dried (ether
and THF over Na/benzophenone, toluene over Na) and distilled prior
to use. Lithium metal contained 1.3% sodium.
(Z)-[1-D]Propene. A 11.3 g (0.342 mol) sample of MeOD was
added to 13.7 g (0.298 mol) of propynyllithium (from propyne and
nBuLi); the [1-D]propyne thus generated was added to a suspension of
dicyclohexylborane37 in diethyl ether at -50 °C. The mixture was
stirred for 4 h while the temperature was allowed to rise to -10 °C.
After complete dissolution of the white solid, the solvent was removed
in vacuum. The resulting dicyclohexyl[1-D]propen-1-ylborane was
dissolved in 150 mL toluene and cooled to -20 °C. (Z)-[1-D]propene
was liberated by adding 22 g (0.37 mol) of acetic acid and condensed
through a cold trap (-50 °C) into a storage vessel at -196 °C to yield
2
7.3 g (0.17 mol, 57%). The H NMR spectrum showed a purity of
94% for the (Z)-isomer (δ ) 4.98 ppm for the (E)-isomer, 5.08 ppm
for the (Z)-isomer, in CHCl3).
(E)-[1-D]Propene. (E)-1-Chloro-1-propene (54 g, 0.71 mol) was
added to freshly cut lithium (15 g, 2.16 mol) in 300 mL of ether at
atm. Residual propene, collected after quenching with CH3OH by condensa-
tion into 1 mL of CHCl3, gave a 2H NMR spectrum indistinguishable from
that of an unreacted sample of (Z)-[1-D]propene; no CH2D signal is apparent
above the detection limit of ca. 2-3%.
(36) That a metal can traverse between alkene enantiofaces without
dissociation has been observed in chiral rhenium olefin complexes (Peng,
T.-S.; Gladysz, J. A. J. Am. Chem. Soc. 1992, 114, 4174), but this
isomerization requires many hours even at elevated temperatures.
(37) Pelter, A.; Smith, K.; Brown, H. C. Borane Reagents; Academic
Press: New York, 1988; p 422.

9032 J. Am. Chem. Soc., Vol. 118, No. 38, 1996
Leclerc and Brintzinger
data,32b,38 pentad distributions estimated from the integral ratios of the
mmmm and mmrr pentad signals (which are not superimposed by end-
group and misinsertion signals), assuming a catalytic-site-controlled
ratio of mmmr:mmrr:mrrm ≈ 2:2:1. The relative proportions of
undeuterated and deuterated mrrm pentad and chain-end signals were
estimated by the cut-and-weigh method and mean degrees of polym-
erization, PN, from the integral ratio of all 13C NMR signals in the
range of 22.5-19.0 ppm39 to that of the n-propyl end group at 14.3
ppm.
cross-relaxation with the diastereotopic proton signals of the CH2 group
at 2.06 and 1.69 ppm, is assigned to the H atom cis to this CH2 group.
The assignment of the two diastereotopic CH2 protons is derived from
1H NMR spectra of polymers made from (E)- and (Z)-[1-D]propene:
In poly((E)-[1-D]propene), the combined integral of the signal group
at 1.67/1.69 ppm is diminished by about 1 H; in poly((Z)-[1-D]propene),
the signal at 2.06 ppm is absent. This implies that the signal at 1.69
is due to the proton in the cis position relative to the last m-CH3 group
in the saw-horse projection shown in Scheme 3, while that at 2.06 is
due to the proton trans to this CH3 group.
1
The H NMR signals associated with 2-propenyl end groups were
assigned as follows: The olefinic protons appear at 4.64 and 4.72 ppm,
the CH3 group next to the double bond at 1.67 ppm, and the adjacent
CH2 group, finally, has diastereotopic protons of which one gives a
signal at 2.06 ppm while the other appears at 1.68 ppm, partly coincident
with the multiplet of the CH3 group at 1.67 ppm.40 This assignment is
based on two-dimensional HMQC spectra which show cross peaks for
the signals at 4.64 and 4.72 ppm with the 13C resonance at 111.3 ppm,
for the signal at 1.69 ppm with the 2-propenyl CH3 13C signal at 22.5
ppm and for the signals at 2.06 and 1.69 ppm with that of the methine
carbon at 45 ppm.41
Acknowledgment. Competent help by Dr. A. Geyer and Ms.
1
M. Cavegn with 13C and H NMR studies and with signal
assignments by HMQC, COSY, ROESY, and DEPT methods
is gratefully acknowledged. For communication of results prior
to publication and for helpful discussions, we wish to thank
Professor J. E. Bercaw, Professor V. Busico, Professor J. A.
Gladysz, Professor R. H. Grubbs, Dr. G. W. Coates, and Dr. L.
Resconi. This work was financially supported by BMBF, by
BASF AG, and by funds of the University of Konstanz. MAO
was donated by Witco GmbH.
Further assignments given in Figure 2 are derived from a ROESY
spectrum of a polypropene sample: The signal at 4.72, which shows
cross-relaxation with the CH3 signal at 1.67, is assigned to the H atom
cis to this methyl group, while the signal at 4.64 ppm, which shows
Supporting Information Available: 1H-, ROESY-, 13C-,
and DEPT-NMR-spectra of deuterated and undeuterated polypro-
pene samples (olefinic and pentad regions) (13 pages). See any
current masthead page for ordering and Internet access
instructions.
(38) Ewen, J. A. J. Am. Chem. Soc. 1984, 106, 6355. Cheng, H. N.;
Ewen, J. A. Macromol. Chem. 1989, 190, 1931.
(39) In this range, each C3H6 unit, including misinsertions and n-propenyl
and 2-propenyl end groups, has just one CH3 signal.
(40) A COSY spectrum also shows cross signals for the protons at 2.06
and 1.69 ppm with those at 4.64 and 4.72 ppm.
(41) These shift values refer to polypropene solutions in CDCl3; in C2D2-
Cl4 solutions, these resonances are shifted to higher fields by ca. 0.08 ppm.
JA961157N