Living polymerization of 1-hexene by cationic zirconium and hafnium complexes that contain a diamido/donor ligand of the type [H3 CC(2-C5H4N)(CH2 NMesityl)2]2-. A comparison of methyl and isobutyl initiators

Article abstract of DOI:10.1021/om030438i

Addition of [Ph₃C]B(C₆F₅)₄] to [MesNpy]Hf(i-Bu)₂ (1a; [MesNpy]^2- = [H₃CC(2-C₅H₄N)-(CH₂N-2,4,6- Me₃C₆H₂)₂]^2-) yields {[MesNpy]Hf(i-Bu)}[B(C₆F₅)₄] (2a) quantitatively. Compound 2a decomposes in a first-order manner at 0°C in C₆D₅Br with k_d = 9.2 × 10^-6 s^-1. The polymerization of 1-hexene by 2a at 0°C in C₆D₅Br was found to follow first-order kinetics with a k_p = 0.10(1) M^-1 s^-1. The polymerization is well-behaved up to 10°C with ΔH^? = 10.82 kcal mol^-1 and ΔS^? = -23.01 cal mol^-1 K^-1 (at [Hf] = 1 M). Up to 600 equiv of 1-hexene were polymerized by 2a in chlorobenzene to give atactic poly[1-hexene] with the expected M_n values and polydispersities between 1.02 and 1.05. Compound 1a could be activated with B(C₆F₅)₃ to yield [MesNpy]Hf(CH₂CHMe₂)} [HB(C₆F₅)₃] (2a′), which will polymerize 1-hexene in C₆D₅Br at 0°C with k_p = 0.048(1) M^-1 s^-1 and in toluene with k_p = 0.0111(3) M^-1 s^-1. Well-behaved inhibitors for 1-hexene polymerization include diisopropyl ether, hexyltrimethylsilyl ether, triethylamine, and tributylamine, but not diphenyl ether or dimethylaniline, both of which undergo CH activation. Activation of [MesNpy]Zr(i-Bu)₂ (1b) with [Ph₃C][B(C₆F₅)₄] in C₆D₅Br gave {[MesNpy]Zr(i-Bu)}[B(C₆F₅)₄] (2b), which decomposed at 0°C by β-hydride elimination in a first-order manner with k_d = 8.3 × 10^-6 s^-1. Polymerization of 100 equiv of 1-hexene by 2b appeared to be well-behaved at temperatures up to 10°C (ΔH^? = 8.08 kcal mol^-1 and ΔS^? = -33.34 cal mol^-1 K^-1 for [Zr] = 1 M), although sensitivity of 1b to light limited reproducibility in general. Activation of [MesNpy]ZrMe₂ (4) with [Ph₃C][B(C₆F₅)₄] resulted in formation of primarily catalytically inactive {[MesNpy]₂Zr₂Me₃}[B(C₆F₅) ₄]. However, ～10% of the theoretical amount of {[MesNpy]ZrMe}[B(C₆F₅)₄] is not trapped by [MesNpy]ZrMe₂ and therefore is available to initiate polymerization of 1-hexene slowly at room temperature. Bulk polymerization results obtained for this system support the assertion that only ～10% of the added metal centers are active for polymerization. Activation of [MesNpy]HfMe₂ with [Ph₃C][B(C₆F₅)₄] also results in formation of inactive {[MesNpy]₂Hf₂Me₃}[B(C₆F₅) ₄].

Full text of DOI:10.1021/om030438i

Organometallics 2003, 22, 4569-4583
4569
Livin g P olym er iza tion of 1-Hexen e by Ca tion ic
Zir con iu m a n d Ha fn iu m Com p lexes th a t Con ta in a
Dia m id o/Don or Liga n d of th e Typ e
[H3CC(2-C5H4N)(CH2NMesityl)2]2-. A Com p a r ison of
Meth yl a n d Isobu tyl In itia tor s
Parisa Mehrkhodavandi, Richard R. Schrock,* and Lara L. Pryor
Department of Chemistry, Massachusetts Institute of Technology, 77 Massachusetts Avenue,
Cambridge, Massachusetts 02139
Received J une 5, 2003
Addition of [Ph3C][B(C6F5)4] to [MesNpy]Hf(i-Bu)2 (1a ; [MesNpy]2- ) [H3CC(2-C5H4N)-
(CH2N-2,4,6-Me3C6H2)2]2-) yields {[MesNpy]Hf(i-Bu)}[B(C6F5)4] (2a ) quantitatively. Com-
pound 2a decomposes in a first-order manner at 0 °C in C6D5Br with kd ) 9.2 × 10-6 s-1
.
The polymerization of 1-hexene by 2a at 0 °C in C6D5Br was found to follow first-order
kinetics with a kp ) 0.10(1) M-1 s-1. The polymerization is well-behaved up to 10 °C with
∆Hq ) 10.82 kcal mol-1 and ∆Sq ) -23.01 cal mol-1 K-1 (at [Hf] ) 1 M). Up to 600 equiv of
1-hexene were polymerized by 2a in chlorobenzene to give atactic poly[1-hexene] with the
expected Mn values and polydispersities between 1.02 and 1.05. Compound 1a could be
activated with B(C6F5)3 to yield {[MesNpy]Hf(CH2CHMe2)}[HB(C6F5)3] (2a ′), which will
polymerize 1-hexene in C6D5Br at 0 °C with kp ) 0.048(1) M-1 s-1 and in toluene with kp )
0.0111(3) M-1 s-1. Well-behaved inhibitors for 1-hexene polymerization include diisopropyl
ether, hexyltrimethylsilyl ether, triethylamine, and tributylamine, but not diphenyl ether
or dimethylaniline, both of which undergo CH activation. Activation of [MesNpy]Zr(i-Bu)2
(1b) with [Ph3C][B(C6F5)4] in C6D5Br gave {[MesNpy]Zr(i-Bu)}[B(C6F5)4] (2b), which
decomposed at 0 °C by â-hydride elimination in a first-order manner with kd ) 8.3 × 10-6
s-1. Polymerization of 100 equiv of 1-hexene by 2b appeared to be well-behaved at
temperatures up to 10 °C (∆Hq ) 8.08 kcal mol-1 and ∆Sq ) -33.34 cal mol-1 K-1 for [Zr]
) 1 M), although sensitivity of 1b to light limited reproducibility in general. Activation of
[MesNpy]ZrMe2 (4) with [Ph3C][B(C6F5)4] resulted in formation of primarily catalytically
inactive {[MesNpy]2Zr2Me3}[B(C6F5)4]. However, ∼10% of the theoretical amount of
{[MesNpy]ZrMe}[B(C6F5)4] is not trapped by [MesNpy]ZrMe2 and therefore is available to
initiate polymerization of 1-hexene slowly at room temperature. Bulk polymerization results
obtained for this system support the assertion that only ∼10% of the added metal centers
are active for polymerization. Activation of [MesNpy]HfMe2 with [Ph3C][B(C6F5)4] also results
in formation of inactive {[MesNpy]2Hf2Me3}[B(C6F5)4].
In tr od u ction
carry out what appear to be living R-olefin polymeriza-
tions for various olefins under certain conditions.8-22
The living Ziegler-Natta polymerization of R-olefins
has been a topic of interest for a number of years.1
However, progress in the area took a leap forward in
1996 with McConville’s report of a diamido titanium
system that carried out living 1-hexene polymerization
(in neat 1-hexene) at room temperature.2,3 (This catalyst
turned out not to be stable at low olefin concentration.4)
In the past several years several groups, including ours,
have reported nonmetallocene group IV catalysts5-7 that
(6) Britovsek, G. J . P.; Gibson, V. C.; Wass, D. F. Angew. Chem.,
Int. Ed. 1999, 38, 428.
(7) Gade, L. H. Chem. Commun. 2000, 173.
(8) Baumann, R.; Davis, W. M.; Schrock, R. R. J . Am. Chem. Soc.
1997, 119, 3830.
(9) Schrock, R. R.; Bonitatebus, P. J ., J r.; Schrodi, Y. Organometal-
lics 2001, 20, 1056.
(10) J ayaratne, K. C.; Sita, L. R. J . Am. Chem. Soc. 2000, 122, 958.
(11) J ayaratne, K. C.; Keaton, R. J .; Henningsen, D. A.; Sita, L. R.
J . Am. Chem. Soc. 2000, 122, 10490.
(12) Keaton, R. J .; J ayaratne, K. C.; Henningsen, D. A.; Koterwas,
L. A.; Sita, L. R. J . Am. Chem. Soc. 2001, 123, 6197.
(13) Tshuva, E. Y.; Goldberg, I.; Kol, M. J . Am. Chem. Soc. 2000,
122, 10706.
(14) Tshuva, E. Y.; Goldberg, I.; Kol, M.; Goldschmidt, Z. Chem.
Commun. 2001, 2120.
(15) Tshuva, E. Y.; Groysman, S.; Goldberg, I.; Kol, M.; Goldschmidt,
Z. Organometallics 2002, 2002, 662.
(16) Tian, J .; Hustad, P. D.; Coates, G. W. J . Am. Chem. Soc. 2001,
123, 5134.
(1) Coates, G. W.; Hustad, P. D.; Reinartz, S. Angew. Chem., Int.
Ed. 2002, 41, 2236.
(2) Scollard, J . D.; McConville, D. H. J . Am. Chem. Soc. 1996, 118,
10008.
(3) Scollard, J . D.; McConville, D. H.; Vittal, J . J .; Payne, N. C. J .
Mol. Catal. A Chem. 1998, 128, 201.
(4) Scollard, J . D.; McConville, D. H.; Rettig, S. J . Organometallics
1997, 16, 1810.
(17) Keaton, R. J .; J ayaratne, K. C.; Fettinger, J . C.; Sita, L. R. J .
Am. Chem. Soc. 2000, 122, 12909.
(5) Kempe, R. Angew. Chem., Int. Ed. 2000, 39, 468.
10.1021/om030438i CCC: $25.00 © 2003 American Chemical Society
Publication on Web 09/25/2003

4570 Organometallics, Vol. 22, No. 22, 2003
Mehrkhodavandi et al.
We have been interested primarily in multidentate
diamido ligands that contain a donor between the two
amido groups, so-called diamido/donor ligands. Most
recently we have focused on zirconium and hafnium
complexes that contain the [H3CC(2-C5H4N)(CH2N-
Mes)2]2- ([MesNpy]2-) ligand.23-25 The trimethylsilyl-
substituted version of this ligand was reported by
Gade,7,26,27 who has explored its coordination chemistry
extensively in recent years. Our belief that trimethyl-
silyl substituents on nitrogen are not likely to be as
stable in a cationic group 4 metal complex as aryl or
tert-butyl substituents28 led us to prepare the mesityl-
substituted versions, H2[MesNpy]. We have reported the
synthesis of zirconium and hafnium dialkyl complexes
that contain [ArNpy]2- ligands (Ar ) mesityl or 2,4,6-
triisopropylphenyl).25 These dialkyl species are stable
when M ) Hf even when R ) i-Pr, n-Bu, or i-Bu. We
also have reported in a preliminary fashion the activa-
tion of zirconium dimethyl and diisobutyl complexes in
bromobenzene-d5 with [Ph3C][B(C6F5)4] to give cationic
species that are active for the polymerization of 1-hex-
ene.23
found to have a polydispersity between 1.02 and 1.05
and a molecular weight equal to that expected for a
living system.
In this paper we report the full details of the activa-
tion of zirconium and hafnium methyl and isobutyl
complexes and the living polymerization of 1-hexene by
the resulting monoalkyl cations. These results will serve
as a base for future investigations of zirconium and
hafnium complexes and their use as catalysts for the
polymerization of 1-hexene and other olefins. It will be
instructive to compare these findings to those obtained
with other relatively well-defined “single-site” polym-
erization catalyst systems based on metallocenes, which
until recently29,30 have not been studied using direct
methods.
Resu lts
1-H exen e P olym er iza t ion w it h {[MesNp y]H f-
(i-Bu )}[B(C6F 5)4]. We will begin with a description of
the most stable and well-defined cationic monoalkyl
complex, {[MesNpy]Hf(i-Bu)}[B(C6F5)4] ([MesNpy]2-
)
[H3CC(2-C5H4N)(CH2NMesityl)2]2-).
Addition of 1 equiv of [Ph3C][B(C6F5)4] to [MesNpy]-
Hf(i-Bu)2 (1a ) or an analogous species in which the
isobutyl group is labeled in the R-position with 13C (1a *)
gave (with respect to Ph3CH) a quantitative yield of
{[MesNpy]Hf(i-Bu)}[B(C6F5)4] (2a or 2a *), along with
isobutene and Ph3CH (eq 1). The isobutene was slowly
The {[MesNpy]Zr(i-Bu)}[B(C6F5)4] catalyst formed by
activation of the diisobutyl species consumed 100 equiv
of 1-hexene between -20 and 10 °C in a smooth and
well-behaved reaction that was first-order in 1-hexene.
On the other hand, activation of [MesNpy]ZrMe2 with
[Ph3C][B(C6F5)4] produced a catalyst that would poly-
merize 1-hexene only very slowly, even at room tem-
perature, as a consequence of formation of largely
inactive {[MesNpy]2Zr2Me3}[B(C6F5)4]. Activation of
[MesNpy]Hf(i-Bu)2 in C6D5Br with 1 equiv of [Ph3C]-
[B(C6F5)4] at -20 °C led to formation of a {[MesNpy]-
Hf(i-Bu)}[B(C6F5)4] species that appeared to be more
stable than the zirconium analogue.24 Consumption of
1-hexene by {[MesNpy]Hf(i-Bu)}[B(C6F5)4] in C6D5Br at
0 °C followed first-order kinetics, and the resulting poly-
(1-hexene) (containing up to 600 equiv of 1-hexene) was
polymerized in the presence of traces of [Ph3C][B(C6F5)4];
isobutene is not polymerized when a different method
of activation is employed that generates the [HB(C6F5)3]-
anion (see later). The 13C{1H} NMR spectrum (0 °C,
C6D5Br) of 2a * showed a singlet at 93.3 ppm corre-
sponding to the Hf-13CH2 carbon atom (cf. the two
R-carbon resonances at 80.9 and 83.0 ppm in 1a *). The
resonance for 13CH2dC(CH3)2 is observed at 111.1 ppm.
(18) Keaton, R. J .; Koterwas, L. A.; Fettinger, J . C.; Sita, L. R. J .
Am. Chem. Soc. 2002, 124, 5932.
(19) J eon, Y.-M.; Park, S. J .; Heo, J .; Kim, K. Organometallics 1998,
17, 3161.
(20) Saito, J .; Mitani, M.; Mohri, J .; Yoshida, Y.; Matsui, S.; Ishii,
S.; Kojoh, S.; Kashiwa, N.; Fujita, T. Angew. Chem., Int. Ed. 2001, 40,
2918.
(21) Mitani, M.; Mohri, J .; Yoshida, Y.; Saito, J .; Ishii, S.; Tsuru,
K.; Matsui, S.; Furuyama, R.; Nakano, T.; Tanaka, H.; Kohoh, S.;
Matsugi, T.; Kashiwa, N.; Fujita, T. J . Am. Chem. Soc. 2002, 124, 3327.
(22) Mitani, M.; Furuyama, R.; Mohri, J .; Saito, J .; Ishii, S.; Terao,
H.; Kashiwa, N.; Fujita, T. J . Am. Chem. Soc. 2002, 124, 7888.
(23) Mehrkhodavandi, P.; Bonitatebus, P. J ., J r.; Schrock, R. R. J .
Am. Chem. Soc. 2000, 122, 7841.
(24) Mehrkhodavandi, P.; Schrock, R. R. J . Am. Chem. Soc. 2001,
123, 10746.
(25) Mehrkhodavandi, P.; Schrock, R. R.; Bonitatebus, P. J ., J r.
Organometallics 2002, 21, 5785.
(26) Friedrich, S.; Schubart, M.; Gade, L. H.; Scowen, I. J .; Edwards,
A. J .; McPartlin, M. Chem. Ber. Recl. 1997, 130, 1751.
(27) Blake, A. J .; Collier, P. E.; Gade, L. H.; McPartlin, M.;
Mountford, P.; Schubart, M.; Scowen, I. J . Chem. Commun. 1997, 1555.
(28) Schrock, R. R.; Liang, L.-C.; Baumann, R.; Davis, W. M. J .
Organomet. Chem. 1999, 591, 163.
1
The H NMR spectrum (-20 °C, C6D5Br) for 2a clearly
shows resonances for the methyl, methylene, and me-
thine protons at 0.55, 0.44, and 1.73 ppm, respectively.
Variable-temperature 13C{1H} NMR studies of 2a * show
that the isobutyl R-carbon resonance at 93.3 ppm
remains sharp as the temperature of the sample is
(29) Liu, Z.; Somsook, E.; Landis, C. R. J . Am. Chem. Soc. 2001,
123, 2915.
(30) Liu, Z.; Somsook, E.; White, C. B.; Rosaaen, K. A.; Landis, C.
R. J . Am. Chem. Soc. 2001, 123, 11193.

Living Polymerization of 1-Hexene
Organometallics, Vol. 22, No. 22, 2003 4571
F igu r e 1. 13C{1H} NMR spectra (0 °C, C6D5Br) after addition of 1-hexene to 2a *: (a) 1-2 equiv of 1-hexene, (b) 45 equiv
of 1-hexene; L ) [MesNpy]2-
.
lowered to -50 °C. Decomposition of 2a at 0 °C could
be followed by observing the decrease in the isobutyl
resonance at 0.55 or 0.44 ppm, or both together, over a
period of 1 to 2 days. First-order rate constants at 0 °C
in bromobenzene was found to be 0.00045, 0.00056,
0.00057, and 0.00060 min-1 for an average of 0.00055-
(6) min-1 or 9.2(2) × 10-6 s-1. Isobutene is generated
as a product of the decomposition of 2a and is polym-
erized irreproducibly in the presence of traces of trityl,
as described above. Although we suspect that the
immediate organometallic product of the decomposition
of 2a is a cationic hydride complex, we have observed
only complex mixtures. Therefore we do not know
whether the initial “hydride” simply decomposes further
rapidly, even in the presence of 1-hexene, or under some
conditions can react with 1-hexene to restart polymer
formation. Decomposition of 2a at 22 °C is complex and
depends on metal concentration. The rate of decomposi-
tion of 2a at 22 °C also appears to depend on isobutene
concentration, which is present as a product of the
initial activation (eq 1) and is also formed upon decom-
position of 2a . The nature of this complex decomposition
phenomenon is not understood at this time, so no details
are reported here.
Addition of 1-2 equiv of 1-hexene to a solution of 2a *
leads to a new 13C{1H} NMR resonance at 49.0 ppm
(* in Figure 1a). This sharp resonance is attributed to
the 13C label in the first 1,2 insertion product,
{[MesNpy]Hf(CH2CH(n-Bu)(13CH2CHMe2)}[B(C6F5)4]. A
resonance for the isobutyl R-carbon atom in unreacted
2a * was also observed at 93.3 ppm. Addition of a further
44 equiv of 1-hexene led to the disappearance of the
resonance at 93.3 and 49.0 ppm, as well as others, and
appearance of a resonance at 44.3 ppm (* in Figure 1b).
Therefore, all of 2a * and the first insertion product (and
the second, etc.) are converted into multiple 1,2 insertion
products whose 13C labels give rise to the resonance at
44.3 ppm. Natural abundant 13C resonances for atactic
poly(1-hexene) were also observed (Figure 1b).31,32 No
F igu r e 2. Plot of ln[1-hexene/Ph2CH2] vs time (min) for
two consecutive additions of 60 equiv of 1-hexene (0.80 M)
to 2a (0.015 M) (0 °C, C6D5Br).
significant amount of 2,1 insertion into the initial Hf-
(isobutyl) bond in 2a * takes place since no other
significant resonance for the 13C label in the multiple
insertion product was found. Since the reaction of the
isobutyl species is a close analogue for the propagation
reaction, we propose that the propagating species reacts
with 1-hexene primarily, if not solely, to give 1,2
insertion products. This is supported further by the
nature of the olefin formed upon â-hydride elimination
from the propagating species (see below).
The disappearance of 1-hexene in the presence of 2a
at 0 °C in C6D5Br was found to follow first-order kinetics
(versus an internal standard, Ph2CH2), as shown in
Figure 2. (The 1-hexene that was employed contained
∼0.1% of a H2CdCRR′ species with vinylidene proton
resonances that are coincident with those in 2-methyl-
1-pentene. Neither the 2-methyl-1-pentene impurity nor
the isobutene generated in the activation step reacted
(31) Asakura, T.; Demura, M.; Nishiyama, Y. Macromolecules 1991,
24, 2334.
(32) Babu, G. N.; Newmark, R. A.; Chien, J . C. Macromolecules 1994,
27, 3383.

4572 Organometallics, Vol. 22, No. 22, 2003
Mehrkhodavandi et al.
Ta ble 1. Su m m a r y of th e P r op a ga tion Ra te
Con sta n ts for P olym er iza tion of Va r iou s r-Olefin s
by 2a in C6D5Br a t 0 °C
olefin
kp (M-1 s-1
)
1-hexene
1-octene
1-decene
0.10(1)
0.13(1)
0.14(1)
1-tetradecene
3-methyl-1-pentene
4,4-dimethyl-1-pentene
0.15(1)
7.2 × 10-4
6.2 × 10-3
with the cationic hafnium alkyl.) Two consecutive
additions of 60 equiv of 1-hexene to a solution of 2a in
C6D5Br at 0 °C resulted in linear plots of ln[1-hexene]
versus time. If d[1-hexene]/dt ) -kp[Hf]o[1-hexene], kp
) 6.0(2) M-1 min-1 (0.10(2) M-1 s-1). The rate of
1-hexene polymerization with 2a was monitored (in 6
runs) over catalyst concentrations in the range 8-23
mM. The values for kobs in these runs were found to
correlate directly with [Hf]o, as expected. Propagation
rate constants obtained over this concentration range
resulted in an average kp value of 6.0(6) M-1 min-1
(0.10(1) M-1 s-1) at 0 °C (Table 1).
F igu r e 3. Analysis of poly(1-hexene) obtained employing
2a (0 °C, C6H5Cl). Polydispersity values are between 1.02
and 1.05 (see Table 1).
Ta ble 2. Ch a r a cter istics of P oly(1-h exen e)
P r ep a r ed Em p loyin g Va r iou s Dia lk yls Activa ted
w ith [P h 3C][B(C6F 5)4] a t 0 °C in Ch lor oben zen e or
C6D5Br a
conc olefin
Mn
Mn
The polymerization of 1-hexene by 2a is well-behaved
at 10 °C or below with kp values (in M-1 s-1 at degrees
K) equal to 0.020 (253 K), 0.049 (263 K), 0.065 (268 K),
0.098 (273 K), and 0.16 (278 K) at [Hf]o concentrations
near 15mM. (The catalyst concentration was determined
in situ by integration versus an internal standard.) An
Eyring plot (R ) 0.997) of ln(k/T) vs 1/T, where k is the
theoretical first-order rate constant and [Hf]o is set at
1 M, yielded ∆Hq ) 10.82 kcal mol-1 and ∆Sq ) -23.01
cal mol-1 K-1. The relatively large and negative ∆Sq is
consistent with a rate-limiting bimolecular propagation
step.
dialkyl complex
(mM) (equiv) theory found fd/th PDI
[MesNpy]Hf(i-Bu)2 10.6
98
101
182
296
398
423
493
607
107
220
63
8248
8500
8012 0.97 1.04
7888 0.93 1.04
9.7
5.3
5.1
5.7
2.7
5.2
15 317 14 380 0.94 1.04
24 911 24 800 1.00 1.03
33 496 31 370 0.94 1.03
35 600 34 220 0.96 1.02
41 491 40 610 0.98 1.03
51 085 49 970 0.98 1.03
4.9
[MesNpy]Zr(i-Bu)2 15
7.3
9005
9092 1.00 1.03a
18 515 19 190 1.04 1.02a
5300 60 000 11.3 1.02
6480 65 000 10.0 1.07
[MesNpy]ZrMe2
1.2
1.0
77
a
Up to 600 equiv of 1-hexene were polymerized by 2a
in chlorobenzene in a series of batchwise experiments
that were carried out for 2 h at 0 °C. (The half-life of
1-hexene at a catalyst concentration of 0.015 M is 7.7
min. Therefore all 1-hexene is polymerized in ∼40 min
at 0 °C.) The molecular weights and polydispersities of
these samples were determined by GPC using a com-
bination of light-scattering and refractometer GPC data.
(The value for dn/dc determined by the manufacturer
of the light-scattering instrument (Wyatt) on atactic
poly(1-hexene) samples prepared in our laboratory using
2a was found to be 0.076; this value was employed in
molecular weight measurements.) The plot of Mn versus
the number of equivalents employed was linear over the
range 100 to ∼600 equiv, and polydispersity values were
between 1.02 and 1.05 (Figure 3; Table 2). In one
experiment, 91 equiv of 1-hexene were added to a
solution of 2a (8.8 mM, C6D5Br). After 2 h at 0 °C
another 91 equiv of 1-hexene were added and polymer-
ization resumed at 0 °C for another 2 h. The resulting
polymer sample had a PDI ) 1.04 and Mn ) 13 650
(theoretical Mn ) 15 320), which is within the error limit
for molecular weight determinations of this type in this
range. These data suggest that little catalyst decom-
poses (e.g., by â-hydride elimination) during this time
period. Unfortunately, we have not been able to find an
NMR resonance that is unique for the living species and,
therefore, have not been able to measure the rate of its
decomposition directly at 0 °C. If we say that the rate
of decomposition of 2a is approximately the same as the
rate of decomposition of the living polymer, then in 2 h
In C6D5Br at 0 °C after NMR studies (see Experimental
Section).
at 0 °C only ∼4% of the propagating species would have
decomposed. As noted above all 1-hexene is polymerized
at 40 min at 0 °C, so decomposition of the living polymer
is likely to be negligible during this period, as is also
suggested by the data in Figure 3.
No resonances at 4.83 and 4.85 ppm or between 5.3
and 5.5 ppm were observed in polymerization runs of
up to 600 equiv of 1-hexene at 0 °C. However, if
polymerization reactions that are completed at 0 °C are
allowed to warm to room temperature, then olefinic
resonances characteristic of â-elimination from a 1,2
insertion product appear at 4.83 and 4.85 ppm.33,34 No
resonances at 5.3-5.5 ppm characteristic of an internal
olefin (the product of â-elimination from a 2,1 insertion
product33,34) were observed. This confirms that the chain
end that is present results from a 1,2 insertion. The â-
(1,2) elimination product was observed to grow in with
time. Unfortunately, a rate for forming this product
could not be obtained since it was polymerized slowly
(along with isobutene) in the presence of traces of trityl.
Addition of 100 equiv of 1-hexene to an initially cold
(-20 °C) solution of 2a followed by warming the sample
to room temperature as the polymerization was com-
(33) Moscardi, G.; Resconi, L.; Cavallo, L. Organometallics 2001,
20, 1918.
(34) Resconi, L.; Piemontesi, F.; Camurati, I.; Sudmeijer, O.; Nifant’ev,
I. E.; Ivchenko, P. V.; Kuz’mina, L. G. J . Am. Chem. Soc. 1998, 120,
2308.

Living Polymerization of 1-Hexene
Organometallics, Vol. 22, No. 22, 2003 4573
pleted led to polymer samples with PDI ) 1.04 and Mn
) 7500 (expected Mn ) 8420). Addition of 100 equiv of
1-hexene to this sample at room temperature led to
polymers that clearly showed bimodal GPC traces (by
both refractive index and light-scattering methods),
consistent with significant chain termination by â-hy-
dride elimination before and during renewed chain
growth at room temperature at the concentrations of
1-hexene employed.
Activation of 1a * with B(C6F5)3 led to clean abstrac-
tion of a â-hydride and formation of {[MesNpy]Hf-
(13CH2CHMe2)}[HB(C6F5)3] (2a ′*) as well as 13CH2d
1
C(CH3)2 (eq 2). The H and 13C{1H} NMR spectra (0 °C,
F igu r e 4. Plots of ln([1-hexene]/Ph2CH2) vs time (min)
after addition of 1-hexene to [(MesNpy)Hf(i-Bu)][A] (∼15
mM) (0 °C, S): (a) A ) [B(C6F5)4]-, S ) C6D5Br; (b) A )
[HB(C6F5)3]-, S ) C6D5Br; (c) A ) [HB(C6F5)3]-, S ) C6D5-
CD3.
monitored over 6 runs at a 2a concentration range of
7-27 mM; the average kp was 9.1(3) M-1 min-1. All data
are gathered in Table 1 in units of M-1 s-1. At this stage
we do not want to make much of the small increase in
kp with increasing chain length.
Polymerizations of both 3-methyl-1-pentene and 4,4-
dimethyl-1-pentene at 0 °C were very much slower than
unbranched terminal olefins, with kp ) 7.2 × 10-4 s-1
for the former and 6.2 × 10-3 s-1 for the latter. (Both
are single runs.) The rate of polymerization of 3-methyl-
1-pentene is more than 100 times slower than 1-hexene.
We ascribe the slower rates to bulk steric effects, with
the effect being greater for substitution closer to the
double bond.
C6D5Br) of 2a ′* show resonances for the Hf-CH2 protons
and R-carbon at 0.4 and 94.0 ppm, respectively, which
are close to those reported for 2a *. The 1H NMR
resonance for the proton attached to boron in [HB-
(C6F5)3]- could not be found. The only byproduct of the
reaction is isobutene, which is not polymerized over a
period of 24 h at room temperature. (In contrast,
activation of diisobutyl complexes with [Ph3C][B(C6F5)4]
results in polymerization of isobutene in the presence
of traces of [Ph3C][B(C6F5)4] at 0 °C.) In contrast to
decomposition of 2a (vide supra) decomposition of 2a ′
Well-Beh a ved In h ibitor s for 1-Hexen e P olym er -
iza tion . It has been known for some time in single-site
cationic group 4 catalyst systems in general that po-
lymerization is quenched in the presence of donors such
as diethyl ether. We find too that polymerization of
1-hexene is quenched after addition of even stoichio-
metric quantities of diethyl ether or THF to the cations.
For example, 4 equiv of diethyl ether were added to a
22 mM solution of 2a in bromobenzene and 72 equiv of
1-hexene were added; no polymer was observed at 0 °C
after 2 h. However, if the catalysts are as well-behaved
as they appear to be, then we should be able to find some
(larger) bases that will bind to the metal with a strength
approximately the same as that of the olefin. If we make
the assumption that an inhibitor binds to the cationic
catalyst to yield an inactive base adduct (cat(B), eq 3),
that the “on” and “off” rates for the base are both fast
with respect to insertion of olefin into the metal-carbon
bond in the base-free catalyst (cat), and that [cat]o )
[cat] + [cat(B)], then kobs ) kp[cat]o/(1 + K[B]), which is
rewritten as eq 4. A plot of [cat]o/kobs versus [B] then
should give a line with slope K/kp and intercept 1/kp,
from which both K and kp can be extracted. The value
of kp (0.10 M-1 s-1) is known in this case by direct
measurement in the absence of inhibitor.
at 22 °C proceeded in a first-order manner with kd
)
0.0044 (2 runs) and 0.0045 min-1 for an average kd )
0.0044(1) min-1 or 7.3(2) × 10-5 s-1 (3 runs total). At 0
°C kd ) 2.8 × 10-6 s-1, which is roughly one-third the
rate of decomposition of 2a .
Polymerization of 1-hexene by 2a ′ in C6D5Br at 0 °C
was well-behaved with kp ) 0.048(1) M-1 s-1 (Figure
4). This should be compared with kp ) 0.10(1) M-1 s-1
for the [B(C6F5)4]- salt. Since compound 2a ′ is sparingly
soluble in toluene (2a is essentially insoluble in toluene),
1-hexene polymerization with 2a ′ also could be carried
out in toluene, in which kp ) 0.0111(3) M-1 s-1 (Figure
4), a further reduction of kp by approximately a factor
of 4 compared to bromobenzene and approximately 1
order of magnitude less than kp for the [B(C6F5)4]- salt
in bromobenzene (Table 2).
P olym er iza tion of Oth er Ter m in a l Olefin s. Po-
lymerizations of several other terminal olefins were
examined. The polymerization of 1-octene was carried
out with 2a concentrations ranging from 7.5 to 28 mM
over 6 runs at 0 °C in bromobenzene. The value for kp
was found to be 7.7(3) M-1 min-1. In seven trials with
2a concentrations ranging from 7 to 27 mM, it was
determined that the average kp for 1-decene was 8.4(4)
M-1 min-1. The polymerization of 1-tetradecene was
cat + Byk1zcat(B)
(3)
(4)
k-1
[cat]o
K
1
)
[B] +
kobs
kp
kp

4574 Organometallics, Vol. 22, No. 22, 2003
Mehrkhodavandi et al.
F igu r e 5. Plot of [Hf]o/kobs (M min) vs [B] for polymeri-
zation of 1-hexene by 2a (∼15 mM) in the presence of
(i-Pr)2O (0 °C, C6D5Br).
F igu r e 6. 1H NMR spectra (0 °C, C6D5Br) for the ortho
pyridyl region after addition of NMe2Ph (18 equiv) and
1-hexene (100 equiv) to {[MesNpy]Hf(i-Bu)}[B(C6F5)4]
(0.013M) (b) ∼10 min after mixing and (a) 2 h later.
Diisopropyl ether was the first well-behaved base to
be discovered. A plot of [Hf]o/kobs vs [B] (Figure 5)
produced a value for kp of 0.11(1) M-1 s-1 and a K )
41(5) M-1. The derived value for kp agrees with the
experimental value of 0.10(1) M-1 s-1, which technically
could also be employed as a point (where [B] ) 0) in
the plot in Figure 5. In one experiment the rate of
polymerization of 52 equiv of 1-hexene by 2a was first
determined, then 14 equiv of (i-Pr)2O were added to the
same solution, and another 52 equiv of 1-hexene were
added. The resulting decrease in the slope of ln[1-
hexene] versus time was that predicted on the basis of
the above analysis. Polymerization of a bulk sample (141
equiv of 1-hexene) in the presence of 13 equiv of (i-Pr)2O
Proton NMR spectra during polymerization revealed
that the sharp pyridyl Ho resonance for 2a at 8.39 ppm
had been replaced by a broad pyridyl Ho resonance at
8.72 ppm characteristic of the propagating species (B
in Figure 6b), as expected, although a new, sharp
resonance began to appear at 9.03 ppm (C in Figure 6)
over a period of 2 h. The 9.03 ppm resonance must be
ascribed to a nonpolymeric species, since it is sharp, as
in 2a . Resonances corresponding to vinylidene protons
were not observed; that is, chain termination via â-hy-
dride elimination did not take place. We propose that
the new Ho resonance is that in a species formed in a
reaction between the propagating species and dimethyl-
aniline (3 in eq 5). The disappearance of the resonance
at 8.72 ppm was found to be first-order with k ) 4.5 ×
10-4 s-1 or a second-order rate constant for the reaction
between the living species and dimethylaniline (kAn) of
1.7 × 10-4 M-1 s-1. The time required to consume 50%
of the initial 1-hexene at 0 °C when [Hf] ) 0.013 M is
∼9 min. During that time ∼20% of the living species
would have reacted with dimethylaniline (if [dimethyl-
aniline]o ) 0.26 M). In 2 h ∼75% of the living species
would be consumed. These data are consistent with the
observations shown in Figure 6. It should be noted that
if the aniline concentration is equal to the catalyst
concentration, then the initial rate of reaction of aniline
with the living polymer is only 1/20 as fast. In fact,
reactions in which the [B(C6F5)4]- salt of dimethyl-
anilinium was employed as an initiator showed only a
few percent of 3. Therefore the [B(C6F5)4]- salt of
dimethylanilinium can be a relatively well-behaved
initiator for 1-hexene polymerization, depending upon
the circumstances, even though dimethylaniline is a
poorly behaved inhibitor at high concentrations. Addi-
tion of 1-hexene to samples containing only 3 led to
essentially no poly(1-hexene) being formed at 0 °C, i.e.,
3 is inactive.
gave a polymer sample with PDI ) 1.05 and Mn
)
11 900 (expected Mn ) 11 870). These data suggest that
the presence of diisopropyl ether is not deleterious in
any way to the polymerization, except in rate; that is,
it is a well-behaved inhibitor. As noted above, polymer-
izations are quenched at 0 °C upon addition of g1 equiv
of diethyl ether.
Polymerization of 1-hexene was examined in the
presence of (TMS)2O (114 equiv relative to catalyst )
14 mM). The value for kp was the same as in the absence
of (TMS)2O. However, polymerization of 1-hexene at 0
°C was inhibited by hexyltrimethylsilyl ether, with a K
) 220 M-1 being determined in runs in the presence of
10, 30, and 100 equiv of hexyltrimethylsilyl ether. This
K is roughly 5 times the K value for diisopropyl ether.
Triethylamine and tributylamine also proved to be
well-behaved inhibitors. Polymerization of 1-hexene
with 2a in the presence of NEt3 proceeded with first-
order kinetics and no observable initiation period. A plot
of [Hf]o/kobs versus [B] was linear and gave a K value of
73(7) M-1. Polymerization of 247 equiv of 1-hexene with
a 10 mM solution of 2a (0 °C, C6H5Cl) with 9 equiv NEt3
resulted in a polymer sample with Mn ) 22 900 (Mn
expected ) 20 790) and PDI ) 1.08. Tributylamine was
actually a slightly more efficient inhibitor than triethyl-
amine. The linear plot of [Hf]o/kobs versus [B] showed a
Complex 2a itself was found to react with dimethyl-
aniline to yield a species with the same pyridyl reso-
nance at 9.03 ppm in the proton NMR spectrum. The
reaction between 2a (0.022 M) and dimethylaniline (0.99
M) was found to be (pseudo) first-order with a first-order
rate constant kd ) 1.11(2) × 10-4 s-1 (t1/2 ) 104 min at
this concentration of dimethylaniline) or a second-order
K value of 107(11) M-1
.
Polymerization of 115 equiv of 1-hexene ([1-hexene]o
) 1.50 M) by a mixture of 2a (0.13 M) and 20 equiv of
dimethylaniline (0.26 M) resulted in a nonlinear plot of
ln([1-hexene]/std) versus time over several half-lives.

Living Polymerization of 1-Hexene
Organometallics, Vol. 22, No. 22, 2003 4575
k ) 1.1 × 10-4 M-1 s-1 at 0 °C, which is slightly smaller
than that for the reaction between the living polymer
and dimethylaniline. (The reaction was followed by
monitoring the disappearance of the ligand backbone
P olym er iza t ion of 1-H exen e w it h Act iva t ed
[MesNp y]Zr (i-Bu )2. Although [MesNpy]Zr(i-Bu)2 (1b)
is thermally unstable and sensitive to light,25 it can be
activated with 1 equiv of [Ph3C][B(C6F5)4] (at -20 °C
in C6D5Br) to give {[MesNpy]Zr(i-Bu)}[B(C6F5)4] (2b) (eq
6) quantitatively. The 1H and 13C{1H} NMR spectra
1
resonance for 2a at 4.4 ppm in the H NMR spectrum.)
When 2a * was employed, the 13C{1H} NMR spectrum
showed resonances for 13CH2dC(CH3)2 (at 111.14 ppm)
from the initial activation process, 2a * (at 94.22 ppm,
cf. at 93.3 ppm in the base-free species), and 13CH3CH-
(CH3)2 (at 24.82 ppm). The formation of isobutane is
consistent with the proposed CH activation process (eq
5).
(-30 °C, C6D5Br) of 2b show only one set of peaks
corresponding to the isobutyl’s R-methylene protons (at
0.62 ppm) and carbon atom (at 83.3 ppm). Compound
2b decomposes at 0 °C to give isobutene and a complex
mixture of products. Decomposition was followed by 1H
NMR spectroscopy (0 °C, C6D5Br) with respect to an
internal standard (Ph2CH2) and was found to be first-
order over several half-lives. However, the rate constant
that was obtained varied significantly from batch to
A compound analogous to 3 was synthesized at 0 °C
over a 24 h period by adding 1 equiv of NMe2Ph to
{[MesNpy]Hf(i-Bu)}[HB(C6F5)3]. It was isolated as a
yellow powder in 72% yield and was characterized by
elemental analysis as well as various NMR spectroscopic
methods. The 1H NMR spectrum (20 °C, C6D5Br) is
essentially identical to that obtained for the decomposi-
tion product during polymerization of 1-hexene in the
presence of dimethylaniline. A gCOSY experiment al-
lowed a full assignment of the spectrum. Three reso-
nances in the aliphatic region were attributed to methyl
groups of the mesityl ring. The fourth resonance at 2.13
ppm is assigned to the N-CH3 groups of a C-H activated
dimethylaniline. Resonances for the four protons of the
N-C6H4 ring were also observed and assigned. The 13C-
{1H} NMR spectrum shows a resonance at 47.9 ppm for
the N-CH3 carbon atom. All NMR data can be found in
the Experimental Section. Unfortunately, no crystals
suitable for an X-ray structural study could be obtained.
Diphenyl ether was also unsuitable as an inhibitor
at high concentrations (∼20× the catalyst concentration
of ∼0.15 M) as a consequence of what we assume to be
a CH activation related to that found for dimethyl-
aniline. In addition to the broad resonance for the
pyridyl proton in the propagating species, two sets of
sharp doublets at 8.70 and 8.74 ppm were observed to
grow in during the polymerization reaction. Therefore
we believe that CH activation in diphenyl ether takes
place to yield two different products, the precise nature
of which is not known. For this reason we did not
explore inhibition by diphenyl ether further.
batch of 1b, with values kd ) 6.7 × 10-6, 9.8 × 10-6
,
1.5 × 10-5, and 3.0 × 10-5 s-1 being obtained, with more
rapid decompositions of 2b being recorded when the
initiator was prepared from progressively older samples
of 1b. (Samples of 1b were stored at -30 °C in the dark.)
The reason for more rapid decompositions of 2b from
“aged” samples of 1b is not known. As decomposition of
2b progresses, the concentration of isobutene in solution
increases, but then decreases as (we believe) it is
polymerized in the presence of traces of [Ph3C][B(C6F5)4].
The decomposition is believed to proceed via â-hydride
elimination to yield isobutene; the zirconium product
of the decomposition could not be determined. If 2b is
prepared at 0 °C or below and kept at that temperature,
then virtually none decomposes before 1-hexene is
added.
Addition of 100 equiv of 1-hexene to 2b (6.8 mM) in
C6D5Br led to a plot of ln[1-hexene] versus time that
was linear over several half-lives at -20, -10, 0, and
10 °C. The respective kp values were 0.027, 0.044, 0.084,
and 0.16 M-1 s-1, respectively. An Eyring plot (for [Zr]
) 1 M) led to ∆Hq ) 8.09 kcal mol-1 and ∆Sq ) -33.34
cal mol-1 K-1. A comparison of kp for {[MesNpy]Zr(i-
Bu)}[B(C6F5)4] (0.084 M-1 s-1) with that for {[MesNpy]-
Hf(i-Bu)}[B(C6F5)4] (0.10 M-1 s-1) reveals that they are
virtually identical. The polymerization at 0 °C was
repeated with a different batch of catalyst by a second
We conclude that well-behaved inhibitors can be
found, although CH activation in phenyl rings that are
directly attached to the coordinating atom in the base
appear to be a complication (at least for 1-hexene under
the conditions employed) that can compromise the living
nature of the polymerization process at relatively high
concentrations of added base.
individual; the result was kp ) 0.097 M-1 s-1
.
As shown in Table 1, the molecular weights of the
polymers obtained with 2b as a catalyst were those
expected for a living system. It should be noted, how-
ever, that higher than expected molecular weights have
been observed if the catalyst is not freshly prepared and
if the reactions are not quenched with methanol im-

4576 Organometallics, Vol. 22, No. 22, 2003
Mehrkhodavandi et al.
equivalent methyl groups, while in 5 the downfield
resonance is that ascribed to the two equivalent methyl
groups.
Observation of 5 and 7 suggests that activation of
[ArNpy]ZrMe2 complexes with 1.0 equiv of [Ph3C]-
[B(C6F5)4] does not result in formation of the expected
{[ArNpy]ZrMe}[B(C6F5)4] species in a readily observable
amount. On the basis of formation of structurally
characterized {[MesN2NMe]2Zr2Me3]+ (where [MesN2-
NMe]2- ) [(MesNCH2CH2)2NMe]2-),35 we propose
that 5 and 7 are dimeric trimethyl monocations, i.e.,
{[ArNpy]2Zr2Me3}[B(C6F5)4]. In {[MesN2NMe]2Zr2Me3]+
one bridging methyl group (J CH ) 133 Hz), in which Zr-
C-Zr ) 180°, and two terminal methyl groups (J CH
)
115 Hz) are present. Since the three methyl groups in
5 and 7 have the same, and relatively low, value for J CH
we propose that 5 and 7 contain three bridging methyl
groups, as shown in eq 7. The three methyl groups must
,
F igu r e 7. Variable-temperature 13C{1H} NMR spectra (1:1
C6D5Br:C6D5CD3) of [MesNpy]Zr13Me2 activated with [Ph3C]-
[B(C6F5)4] (1.0 equiv).
mediately after they are complete. Variable results of
this nature confirm that polymerizations by 2b are
potentially much more problematic than polymeriza-
tions by 2a .
Activa tion of [Ar Np y]Zr Me2 a n d La belin g Stu d -
ies. Activation of [MesNpy]ZrMe2 (4) with 1.0 equiv of
[Ph3C][B(C6F5)4] (20 °C, C6D5Br) results in formation
of Ph3CMe and a zirconium cation (5). An 1H NMR
spectrum of this cation shows a sharp resonance at 2.03
ppm, which corresponds to the three protons of Ph3-
CCH3, and a broader resonance at 0.66 ppm, which
corresponds to approximately nine protons, i.e., three
methyl groups are present in this cation. In the 1H NMR
spectrum of activated [MesNpy]Zr13Me2 (4*, which
yields 5*) the methyl resonance splits into two reso-
nances with a J CH of 108 Hz, while the 13C{1H} NMR
spectrum shows a broad singlet at 38 ppm at room
temperature (Figure 7).
scramble intramolecularly, since species of this type are
slow to lose [ArNpy]ZrMe2 (see below). Only 0.5 equiv
of [Ph3C][B(C6F5)4] is required to form these dimeric
monocations. Therefore the solutions generated by add-
ing 1.0 equiv of activator are dark orange, the color of
unreacted [Ph3C][B(C6F5)4].
Variable-temperature 13C{1H} NMR studies of 5*
were carried out in a 1:1 mixture of C6D5Br and C6D5-
CD3 (Figure 7). These studies show that as the solution
is cooled to -60 °C, the resonance at 38 ppm broadens
and splits into two peaks (at 41.9 and 32.1 ppm) in a
2:1 ratio. A proton-coupled 13C NMR experiment at -60
°C showed the J CH for both resonances to be 108 Hz.
The three methyl groups therefore are likely to be of
the same type. A 1H ROESY experiment (-25 °C, C6D5-
Br) shows NOE cross-peaks between the Zr-CH3 and
the ortho pyridyl proton resonances and between
Zr-CH3 and one set of the mesityl ortho methyl groups.
It can be shown readily that the solution obtained
upon addition of 1.0 equiv of [Ph
3C][B(C6F5)4] contains
0.5 equiv of unreacted [Ph3C][B(C6F5)4]. One equivalent
of 6 was activated with 1.0 equiv of [Ph3C][B(C6F5)4] to
1
yield an orange solution whose H NMR spectrum (20
°C, C6D5Br) showed resonances for {[TripNpy]2Zr2Me3}-
[B(C6F5)4] (7) and Ph3CMe. One equivalent of 13C-
1
labeled 6* was then added. The H NMR spectrum of
this solution showed resonances for Ph3CMe and Ph3-
C13Me in approximately a 1:1 ratio, while the 13C{1H}
NMR spectrum showed resonances for Ph3C13Me and
7*.
If 0.5 equiv of [Ph3C][B(C6F5)4] is added to [ArNpy]-
ZrMe2 complexes, then pale yellow solutions result,
Activation of [TripNpy]ZrMe2 (6) with 1 equiv of
[Ph3C][B(C6F5)4] (Trip ) 2,4,6-triisopropylphenyl) re-
sults in formation of Ph3CMe and a cationic species (7)
since little or no excess trityl is present. [TripNpy]Zr13
-
Me2 (6*) was activated with 0.5 equiv of [Ph3C]-
[B(C6F5)4] to generate {[TripNpy]2Zr213Me3}[B(C6F5)4]
(7*). Variable-temperature 13C{1H} studies of this yel-
low solution in a 1:1 mixture of C6D5Br and C6D5CD3
are essentially identical to those of 6 activated with 1.0
equiv of [Ph3C][B(C6F5)4].
1
that is similar to 5. The H NMR spectrum of 7 (20 °C,
C6D5Br) shows a single broad peak at 0.9 ppm corre-
sponding to the Zr-CH3 group, with a J CH value of 109
Hz. Variable-temperature studies of the 13C-labeled
cation (obtained from activation of 6*) in a 1:1 mixture
of C6D5Br and C6D5CD3 reveal two resonances at -70
°C (at 47.2 and 36.3 ppm) in a 1:2 ratio. In 7 the most
upfield Zr-13Me resonance is that ascribed to the two
(35) Schrock, R. R.; Casado, A. L.; Goodman, J . T.; Liang, L.-C.;
Bonitatebus, P. J ., J r.; Davis, W. M. Organometallics 2000, 19, 5325.

Living Polymerization of 1-Hexene
Organometallics, Vol. 22, No. 22, 2003 4577
NMR spectra (13C{1H} studies in 1:1 C6D5Br:C6D5-
CD3) of a mixture of 7* and 5 showed no evidence for
partially 13C-labeled 5. Treatment of 5 with excess
[Ph3C][B(C6F5)4] at 25 °C for 2 h also led to no further
activation of the cationic dinuclear species. However,
heating the sample to 40 °C for a further 2 h led to
approximately 10% further activation, as evident by the
presence of Ph3C13Me resonances in the 1H and 13C{1H}
NMR spectra (40 °C, C6D5Br). The dinuclear monoca-
tions also do not react readily with excess diethyl ether,
THF, or DME at room temperature to give [ArNpy]-
ZrMe2 and solvent adducts, {[ArNpy]ZrMeSx}[B(C6F5)4].
All of these observations suggest that the dinuclear
monocations are relatively stable toward dissociation to
give {[ArNpy]ZrMe}[B(C6F5)4] and [ArNpy]ZrMe2. Com-
pounds 5, 5*, 7, and 7* were all prepared and isolated
as yellow powders. They were all characterized by
F igu r e 8. Plot of ln[1-hexene/Ph2CH2] vs time (min) for
addition of 160 equiv of 1-hexene to [MesNpy]ZrMe2 (0.007
M) activated with 1 equiv of [Ph3C][B(C6F5)4] (25 °C, C6D5-
Br).
1
elemental analysis, as well as by H and 13C{1H} NMR
spectroscopy, and had spectra identical to their coun-
terparts generated in situ.
zation of 1-hexene, we were surprised to find that
activation of [MesNpy]ZrMe2 with 1 equiv of [Ph3C]-
[B(C6F5)4] in chlorobenzene at room temperature fol-
lowed by addition of 1-hexene does lead to formation of
poly(1-hexene), but slowly at room temperature. In a
series of carefully controlled polymerizations, [MesNpy]-
ZrMe2 was activated with 1 equiv of [Ph3C][B(C6F5)4]
in bromobenzene-d5 at -30 °C, and the rates of con-
sumption of up to 200 equiv 1-hexene at 25 °C were
determined by monitoring the decrease in 1-hexene
concentration with respect to an internal standard using
1H NMR spectroscopy. (A control experiment showed
that 1-hexene was not polymerized by [Ph3C][B(C6F5)4]
in the absence of zirconium, as expected.) An example
is shown in Figure 8. The plot is slightly convex for the
first 100 min, but becomes essentially linear thereafter.
We propose that under the initiation conditions some
{[MesNpy]ZrMe}[B(C6F5)4] escapes being captured by
[MesNpy]ZrMe2 and therefore is available to act as an
initiator for polymerization. The nonlinearity in the
early part of the plot can be attributed to a slow
initiation process that is able to compete with chain
growth in the early stages of polymerization, or perhaps
simply to completion of the initiation process. (The
1-hexene was added at low temperature, where the
reaction between [Ph3C][B(C6F5)4] and [MesNpy]ZrMe2
may not have been complete before the sample was
warmed to 25 °C.) If we assume that the linear portion
of the plot is an approximate measure of the rate
constant for polymerization at 25 °C, then kobs ) 2.7(2)
Treatment of {[MesNpy]2Zr2Me3}[B(C6F5)4] (5) with
2 equiv of [TripNpy]Zr(13CH3)2 (6) led to a slow exchange
of methyl groups between these two species over a
period of 4 h at 25 °C, according to 13C NMR studies.
The most plausible mechanism of methyl exchange is
for 5 to dissociate as shown in eq 8 to yield {[MesNpy]-
ZrMe}[B(C6F5)4] and [MesNpy]ZrMe2. At room temper-
ature this equilibrium lies far toward 5, and virtually
no {[MesNpy]ZrMe}[B(C6F5)4] is available in solution.
However at 40 °C, the rate of forming {[MesNpy]ZrMe}-
[B(C6F5)4] and [MesNpy]ZrMe2 is large enough to result
in methyl group exchange between {[MesNpy]ZrMe}-
[B(C6F5)4] and [TripNpy]ZrMe2 or between [TripNpy]-
ZrMe2 and [MesNpy]ZrMe2,25 or both.
Addition of a large excess of 1-hexene to a solution of
isolated {[MesNpy]2Zr2Me3}[B(C6F5)4] in chlorobenzene
yielded no poly(1-hexene) over a period of 24 h at 25
°C. Since the exchange described immediately above
involves loss of [MesNpy]ZrMe2 from {[MesNpy]2Zr2-
Me3}[B(C6F5)4] and generation of some [MesNpy]ZrMe}-
[B(C6F5)4], we suspect that the rate of reaction between
1-hexene and {[MesNpy]ZrMe}[B(C6F5)4] under the
conditions employed simply does not compete with the
back reaction between [MesNpy]ZrMe2 and {[MesNpy]-
ZrMe}[B(C6F5)4]. In contrast, the rate of the reaction
between {[MesNpy]ZrMe}[B(C6F5)4] and [TripNpy]-
ZrMe2 would be comparable to the rate of the reaction
between {[MesNpy]ZrMe}[B(C6F5)4] and [MesNpy]-
ZrMe2. Therefore slow methyl group exchange is ob-
served.
1
× 10-4 s-1 at 25 °C. H and 13C NMR spectra (25 °C,
C6D5Br) indicate that after all the olefin is consumed,
most, if not all, of the initial {[MesNpy]2Zr2Me3}-
[B(C6F5)4] remains in solution. In another experiment
80 equiv of 1-hexene were added to a sample of catalyst
([Zr]o ) 0.01 M) and the consumption of olefin was
monitored at 25 °C. After all the olefin was consumed,
a further 80 equiv of 1-hexene were added to the
solution and the polymerization process was again
monitored. The observed rate constant (kobs) values were
2.7(2) × 10-4 and 3.0(2) × 10-4 s-1 for the first and
second additions, respectively, and little or no
{[MesNpy]2Zr2Me3}[B(C6F5)4] appeared to be consumed.
No significant initiation period was observed for the
second addition. If we calculate the theoretical rate of
polymerization of 1-hexene at 25 °C using the activation
In view of the above results concerning the formation
of dimeric monocations that are inactive for polymeri-

4578 Organometallics, Vol. 22, No. 22, 2003
Mehrkhodavandi et al.
F igu r e 9. 13C{1H} NMR spectrum (20 °C, C6D5Br, CH3
resonances) of [MesNpy]Zr13Me2 activated with 1 equiv of
[Ph3C][B(C6F5)4] after addition of 2 equiv of 1-hexene.
parameters that we found when {[MesNpy]Zr(isobutyl)}-
[B(C6F5)4] was employed as the initiator, we obtain a
theoretical kp ) 0.337 M-1 s-1 or kobs ) 3.30 × 10-3 s-1
(at [Zr] ) 0.01 M), which is ∼10 times what is observed.
Therefore only ∼10% of the zirconium in the activated
dimethyl system actually is active for polymerization.
It is likely that the amount of {[MesNpy]ZrMe}-
[B(C6F5)4] that ends up in solution and that therefore
is available for initiating polymerization will vary with
the temperature of initiation and other slight procedural
differences between various experiments.
F igu r e 10. Plot of [olefinic byproducts]/C6Me6 vs time
(min) for two consecutive additions of 80 equiv of 1-hexene
to [MesNpy]ZrMe2 (0.01 M) activated with 1 equiv of [Ph3C]-
[B(C6F5)4] (25 °C, C6D5Br).
all of the olefin was consumed even after 24 h at room
temperature, possibly due to catalyst decomposition by
â-hydride elimination under these conditions. Activation
of [MesNpy]ZrMe2 (0.0213 M in C6D5Br) with 0.4 equiv
of [Ph3C][B(C6F5)4] led to a solution that was virtually
inactive as a catalyst for the polymerization of 1-hexene
(34 equiv in 24 h at 22 °C).
It was possible to observe the insertion of 1-hexene
into the Zr-CH3 bond of the monomeric monocation by
treating a solution of [MesNpy]Zr13Me2 (5*, 0.01 M) with
1 equiv of [Ph3C][B(C6F5)4] followed by 2 equiv of
1-hexene. Although no appreciable amount of polymer
was evident after 90 min (because of the low concentra-
tion of 1-hexene in this experiment), after 24 h reso-
Bulk polymerization results obtained for this system
support the assertion that only a small fraction of metal
centers catalyze polymerization. Polymer samples ob-
tained from a solution of [MesNpy]ZrMe2 activated with
1.0 equiv of [Ph3C][B(C6F5)4] at 25 °C have Mn values
that are ∼10 times greater than expected (Table 2).
These data again indicate that only ∼10% of the catalyst
is active for polymerization, in agreement with other
estimates. In view of the possible complications in the
polymerization process at 25 °C (â-hydride elimination
and 2,1 insertion into the initial Zr-Me bond; see
below), it is surprising that the PDI values are so low.
PDI values of this low magnitude in the literature are
often said to be “proof” that a polymerization is “living”.
2,1 In ser tion of 1-Hexen e in to {[MesNp y]Zr Me}-
[B(C6F 5)4]. In the early stages of polymerization in the
activated [MesNpy]ZrMe2 system, two sets of weak
1
nances for poly(1-hexene) were observed in the H NMR
spectrum. The 13C{1H} NMR spectrum also showed a
resonance at 21 ppm as a result of multiple insertion
products,36 as well as peaks corresponding to the un-
reacted cation and the byproduct, Ph3CMe (Figure 9).
The intensity of the resonance at 21 ppm relative to that
ascribed to {[MesNpy]2Zr213Me3}[B(C6F5)4] at 39 ppm
should approximately represent the amount of {[MesNpy]-
Zr13Me}{B(C6F5)4} that has escaped being captured to
give {[MesNpy]2Zr213Me3}[B(C6F5)4] and that therefore
is available to initiate polymerization. Although these
resonances cannot be integrated accurately under the
circumstances, we can at least say that only a fraction
of the 13C originally added to the system turns up in
the polymer in this experiment, consistent with our
earlier assertion. The quality of this spectrum does not
allow us to exclude the possibility of some initial 2,1
insertion into the Zr-Me bond followed by further 1,2
insertions or 2,1 insertion in the propagating species
followed by further 1,2 insertions, i.e., enchainment of
the regioerror. However, we show below that a 2,1
misinsertion into the initial Zr-Me bond probably
results exclusively in â-elimination.
1
resonances appear in the olefinic region of the H NMR
spectra between 5.3 and 5.5 ppm. These resonances are
in the region characteristic of internal olefins, which
would be formed by â-hydride elimination after a 2,1
insertion.33,34 The rate of formation of these byproducts
(monitored versus a C6Me6 internal standard) is great-
est at the early stages of the polymerization period
(Figure 10), and no increase in the â-elimination prod-
ucts was observed after the 1-hexene was fully con-
sumed. A similar pattern (initial increase, then eventual
cessation as 1-hexene was consumed) was observed
upon addition of another 80 equiv of 1-hexene (Figure
10); the amount formed in the second stage is less than
in the first stage. Clearly formation of these products
is hexene dependent.
Activation of [MesNpy]ZrMe2 (0.019 M) with 0.5 equiv
of [Ph3C][B(C6F5)4] followed by addition of 43 equiv of
1-hexene led to formation of poly(1-hexene]) slowly. The
ln[1-hexene] versus time plot is similar to that shown
in Figure 8, but the reaction was clearly much slower
with kobs ) 5.8(2) × 10-5 s-1, which is one-fifth the value
shown in Figure 8. Therefore only ∼2% of the zirconium
added to the system is active in this experiment. Not
The 1H NMR spectra of reaction mixtures that contain
these internal olefin byproducts show a splitting pattern
that is superimposable on the pattern of a 4:1 mixture
(36) Baumann, R.; Schrock, R. R. J . Organomet. Chem. 1998, 557,
69.

Living Polymerization of 1-Hexene
Organometallics, Vol. 22, No. 22, 2003 4579
Sch em e 1. P r op osed Or igin of 2-Hep ten esa
a
L ) [MesNpy]2-, anion ) [B(C6F5)4]-.
of trans-2-heptene and cis-2-heptene, as shown in Figure
11. Therefore it would appear that these olefins form
via 2,1 insertion of 1-hexene into the Zr-CH3 bond in
{[MesNpy]ZrMe}[B(C6F5)4], as shown in Scheme 1,
followed by â-elimination. 3-Heptenes do not form,
perhaps because of the more facile â-hydride elimination
in the ethyl group versus the butyl group in the internal
alkyl. No â-hydride elimination products of this type
were observed when {[MesNpy]Zr(isobutyl)}[B(C6F5)4]
was employed as the initiator at 0 °C. Therefore the 2,1
regioerror in {[MesNpy]ZrMe}[B(C6F5)4] at 25 °C can
be ascribed to differences (largely steric) between inser-
tion into a methyl group (at 25 °C) versus that into an
isobutyl group (at 0 °C).
Integration versus a C6Me6 internal standard reveals
that the amount of olefin present after addition of 160
equiv of 1-hexene is ∼4% of the total metal added to
the system. This integration is relatively inaccurate
because of the small quantities. We also know that the
amount of {[MesNpy]ZrMe}[B(C6F5)4] initiator present
in solution initially is only 10-15% of the total metal
that has been added. Therefore a “misinsertion” into the
initial Zr-Me bond followed by â-hydride elimination
could take place approximately one-third of the time (at
room temperature). We are proposing that no 2,1
misinsertion product is enchained and that each 2,1
insertion error is followed by â-hydride elimination. We
do not know whether the metal-containing product of
â-hydride elimination reacts with 1-hexene or decom-
poses too quickly to do so. Nevertheless, and perhaps
surprisingly, bulk polymerization of 1-hexene at room
temperature with the Zr-Me initiator results in poly-
mers with low PDI values, even though the Zr-Me
initiator is clearly not an ideal living system, to say the
least.
Activation of [MesNpy]HfMe2 an d Labelin g Stu d-
ies. Activation of [MesNpy]HfMe2 with 1.0 equiv of
[Ph3C][B(C6F5)4] consumes only half of the activator and
results in formation of the dinuclear complex {[Mes-
1
Npy]2Hf2Me3}[B(C6F5)4] (eq 9). The H NMR chemical
shifts for this species correspond closely to those for
{[MesNpy]2Zr2Me3}[B(C6F5)4]. Activation of [MesNpy]-
HfMe2 with 0.5 equiv of B(C6F5)3 results in methyl
abstraction and formation of {[MesNpy]2Hf2Me3}[MeB-
(C6F5)3]. This species and its 13CH3 labeled analogue
have been isolated and characterized by proton and
carbon NMR spectroscopy (see Experimental Section)
and by elemental analysis. In view of the (now not
unexpected) fact that addition of 96 equiv of 1-hexene
to a solution of isolated {[MesNpy]2Hf3Me3}[B(C6F5)4]
(0.009 M in C6D5Br) resulted in no polymerization in 2
h at 22 °C the hafnium dimeric monocation was not
studied further. We have no reason to believe that the
hafnium system would differ in any significant way from
the zirconium system.
Discu ssion
It has become clear that the term “living polymeri-
zation” means different things to different people.37
However, the strict interpretation of “living”, as outline
in a recent review article1 by Coates, relies on seven
criteria. With some changes in wording they are the
following: (1) all monomer is consumed, and addition
F igu r e 11. Proton NMR spectra of the olefinic byproducts
of 1-hexene polymerization by activated [MesNpy]ZrMe2
compared to a 4:1 mixture of trans- and cis-2-heptene.
(37) Darling, T. R.; Davis, T. P.; Fryd, M.; Gridnev, A. A.; Haddleton,
D. M.; Ittel, S. D.; Mathieson, R. R., J r.; Moad, G.; Rizzardo, E. J .
Polym. Sci.: Part A: Polym. Chem. 2000, 38, 1706.

4580 Organometallics, Vol. 22, No. 22, 2003
Mehrkhodavandi et al.
of more monomer leads to further consumption at the
same rate; (2) Mn increases linearly as a function of
conversion; (3) the number of active centers remains
constant; (4) Mn can be controlled precisely through
stoichiometry; (5) polydispersities are essentially those
expected in theory from a perfect living polymerization,
i.e., close to 1 (e.g., 1.05) for a polymer of reasonable
length (e.g., a 500-mer or larger); (6) block copolymers
can be prepared by sequential addition of two different
monomers; and (7) the living polymer chain ends can
be end-functionalized quantitatively. Of course it is not
absolutely necessary that all criteria be fulfilled. For
example, the first six might be fulfilled, but an end-
functionalizing reagent might not have been found. That
would not invalidate the living nature of the reaction.
On the basis of what we report here we would suggest
that two more criteria could be added to this list: (8)
the living propagating species should be observable
using some technique (such as NMR methods); and (9)
the kinetics of polymerization should be understood and
explicable. Again, to some extent these are restatements
of other criteria. We believe that at least for the Hf
systems reported here, and we should state that this is
true so far only for 1-hexene, we feel that seven of these
(now) nine criteria have been satisfied.
Dimethyl complexes of group 4 metals have been
widely studied since they are usually much more stable
than dialkyl complexes in which the alkyl contains one
or more â-protons, and they are readily activated in
several ways. However, it has been known in metal-
locene systems for some time that methyl groups can
readily bridge between metals.38-46,47 For example,
activation of Cp*2ZrMe2 with an excess of the B[C6F4-
(2-C6F5)]3 was found to yield only the dinuclear complex
{(Cp*2ZrMe)2Me}{MeB[C6F4(2-C6F5)]3}.48 This complex
has one bridging methyl group with a J CH value of ∼130
Hz, characteristic of sp2 hybridization.48 As stated
previously, {[MesN2NMe]2Zr2Me3}[B(C6F5)4] ([MesN2-
NMe]2- ) [(MesNCH2CH2)2NMe]2-) is formed upon
activation of [MesN2NMe]ZrMe2 with 0.5 equiv of [Ph3C]-
[B(C6F5)4].35 (The (linear) bridging methyl group in this
complex has a J CH value of 133 Hz.) However, {[MesN2-
NMe]2Zr2Me3}[B(C6F5)4] dissociates readily into {[MesN2-
NMe]ZrMe}[B(C6F5)4] and [MesN2NMe]ZrMe2 and there-
fore reacts further with 0.5 equiv of [Ph3C][B(C6F5)4] to
give a high yield of {[MesN2NMe]ZrMe}[B(C6F5)4], the
expected product of complete activation. The bimetallic
trimethyl monocation is not stable enough to be ob-
served upon activation of [t-BuNON]ZrMe2 ([t-BuNON]2-
) [(t-BuN-o-C6H4)2O]2-) by [Ph3C][B(C6F5)4] in bro-
mobenzene,49 presumably because the [t-BuNON]2-
system is too sterically demanding. The relatively “open”
nature of {[MesNpy]ZrMe}+ allows a strong adduct with
[MesNpy]ZrMe2 to form. The presence of three bridging
methyl groups also probably significantly reduces the
possibility of losing the [MesNpy]ZrMe2 “base” compared
to a complex in which a single bridging methyl group is
present.
Reports in the literature have suggested that the alkyl
group in the initiator can have a significant effect on
the polymerization process. For example, Fink et al.
reported that the rate of monomer insertion into a
metal-alkyl bond (alkyl * Me) can be up to 100 times
faster than insertion into a metal-Me bond.50-53 (No
details of the initiation process were elucidated.) Boch-
mann et al. have reported that the use of Al(i-Bu)3 as a
cocatalyst with L2ZrMe2 systems (L2 ) Cp2, rac-Me2Si-
(Ind)2) is beneficial as a consequence of alkyl exchange
with the cocatalyst to yield Zr(i-Bu) species.54 In this
case, the bulkier alkyl group is thought to prevent the
formation of catalytically dormant aluminum adducts
and thereby increase catalyst efficiency. Marks et al.
also have reported significant Zr-alkyl effects on ion pair
formation thermodynamics.55 We have shown here why
it is dangerous to draw conclusions about the rate of
insertion into various metal-alkyl bonds simply on the
basis of the rate of polymerization of a given monomer
after what is usually assumed to be complete activation
of the dialkyl uncomplicated by side reactions. Never-
theless it is sensible to propose that the rate of insertion
into a metal-methyl bond in an initiator is slower than
the rate of insertion into the MR bond in a propagating
species as a consequence of the anion interacting more
strongly with the sterically less demanding [MMe]+
system.
The reaction of a terminal olefin with a primary metal
alkyl often proceeds largely with 1,2 regiochemistry.
Theoretical studies show a clear preference for 1,2
insertion of olefin due to both electronic and steric
factors.56,57 However, regioerrors due to 2,1 insertion of
olefin are commonly observed in metallocene systems.58
In these cases, polymer molecular weights are lowered
as chain termination competes with propagation and the
hydride product of chain termination reacts with 1-hex-
ene to start another chain.59 We find that 2,1 insertion
into a Zr-Me bond at room temperature is surprisingly
facile at 25 °C. Recently, Landis has quantified many
of the features of polymerization of 1-hexene by a Zr
indenyl system and has found that 2,1 insertion is
followed by rapid â-elimination, which gives rise to the
(38) Bochmann, M.; Lancaster, S. J . Angew. Chem., Int. Ed. Engl.
1994, 33, 1634.
(39) Beck, S.; Prosenc, M. H.; Brintzinger, H. H.; Goretzki, R.;
Herfert, N.; Fink, G. J . Mol. Catal. A 1996, 111, 67.
(40) Koehler, K.; Piers, W. E.; J arvis, A. P.; Xin, S.; Feng, Y.;
Bravakis, A. M.; Collins, S.; Clegg, W.; Yap, G. P. A.; Marder, T. B.
Organometallics 1998, 17, 3557.
(41) Bochmann, M.; Lancaster, S. J .; Hursthouse, M. B.; Malik, K.
M. A. Organometallics 1994, 13, 2235.
(49) Goodman, J . T.; Schrock, R. R. Organometallics 2001, 20, 5205.
(50) Fink, G.; Zoller, W. Makromol. Chem. 1981, 182, 3265.
(51) Fink, G.; Schnell, D. Angew. Makromol. Chem. 1982, 105, 31.
(52) Mynott, R.; Fink, G.; W., F. Angew. Makromol. Chem. 1987,
154, 1.
(53) Fink, G.; Fenzl, W.; Mynott, R. Z. Naturforsch., B: Chem. Sci.
1985, 40b, 158.
(42) J ia, L.; Yang, X.; Stern, C. L.; Marks, T. J . Organometallics
1997, 16, 842.
(43) Buchwald, S. L.; Lucas, E. A.; Davis, W. M. J . Am. Chem. Soc.
1989, 111, 397.
(44) Watson, P. L.; Parshall, G. W. Acc. Chem. Res. 1985, 18, 51.
(45) Watson, P. L. J . Am. Chem. Soc. 1983, 105, 6491.
(46) Waymouth, R. M.; Santarsiero, B. D.; Grubbs, R. H. J . Am.
Chem. Soc. 1984, 106, 4050.
(47) Chen, E. Y.-X.; Marks, T. J . Chem. Rev. 2000, 100, 1391.
(48) Chen, E. Y.-X.; Stern, C. L.; Yang, S.; Marks, T. J . J . Am. Chem.
Soc. 1996, 118, 12451.
(54) Zhou, J .; Lancanster, S. J .; Walker, D. A.; Beck, S.; Thornton-
Pett, M.; Bochmann, M. J . Am. Chem. Soc. 2001, 123, 223.
(55) Beswick, C. L.; Marks, T. J . Organometallics 1999, 18, 2410.
(56) Kawamura-Kuribayashi, H.; Koga, N.; Morokuma, K. J . Am.
Chem. Soc. 1992, 114, 8687.
(57) Kawamura-Kuribayashi, H.; Koga, N.; Morokuma, K. J . Am.
Chem. Soc. 1992, 114, 2359.
(58) Resconi, L.; Cavalo, L.; Fait, A.; Piemontesi, F. Chem. Rev. 2000,
100, 1253.
(59) Resconi, L.; Camurati, I.; Sudmeijer, O. Top. Catal. 1999, 7,
145.

Living Polymerization of 1-Hexene
Organometallics, Vol. 22, No. 22, 2003 4581
Ta ble 3. Com p a r ison of Sever a l Ra te Con sta n ts a t
hexene dependence of â-elimination via 2,1 insertion in
that system. Virtually no â-hydride elimination is
observed in the hafnium isobutyl systems at 0 °C upon
synthesis of up to a 600-mer of poly[1-hexene].
0 °C for th e P olym er iza tion of 1-Hexen e
kp
compound
solvent (M-1 s-1
)
Today it is generally believed that all anions interact
to some degree with the cation in a group 4 ionic
polymerization system and that, in general, bulky,
fluorinated borate anions are some of the more readily
accessible weakly coordinating anions.42,60-63 It also is
generally believed that anion effects can be interpreted
most readily in terms of the degree of cation/anion
“separation”; the more separated and solvated the ion
pairs, the more active (but perhaps also the more
unstable) the catalyst.64 We have seen some evidence
for well-behaved anion and solvent effects in this work
that could be interpreted in this manner. However, we
were surprised to find in preliminary experiments that
{[MesNpy]Hf(i-Bu)}[HB(C6F5)3] was half as active as {-
[MesNpy]Hf(i-Bu)}[B(C6F5)4] in bromobenzene. We had
expected that a Hf-H-B interaction would dramatically
decrease the rate of reaction of an early metal “cation”
with an olefin. Apparently that is not the case.
We were surprised to find that hafnium and zirco-
nium complexes polymerize 1-hexene at essentially the
same rate. We had expected that the anion would form
a tighter ion pair with hafnium and therefore slow the
reaction between the ion pair and the olefin. However,
the increased strength of the interaction between the
olefin and hafnium apparently compensates and the
rate overall for hafnium ends up being much the same
as for zirconium.
The idea of inhibiting a polymerization reaction
catalyzed by a cationic early metal species with an
otherwise inert “base” that binds more strongly than
an olefin to the metal and thereby limits access of the
olefin to the metal is not new. However, we could find
no quantitative example of the simple inhibitions that
we have demonstrated here, perhaps largely because
inhibition is easiest to demonstrate quantitatively in a
well-behaved living system. One of the interesting
findings is that dimethylaniline may not be a good
choice as an inhibitor, although it has been assumed to
be innocuous by us9,49 as well as many other workers
in group 4 metal catalyst systems (largely metallocene
systems) in the past. Whether CH activation in dim-
ethylaniline takes place in other catalyst systems will
depend on the nature of the system, the concentration
of catalyst and dimethylaniline, and the monomer being
polymerized. For example, ethylene may be polymerized
before CH activation can take place, while 1-hexene, for
example, may not be. Knowledge of inhibitors in the
systems described here ultimately will allow one to
choose a protecting group for a given functionalized
olefin that would then allow the protected functionalized
olefin to be polymerized smoothly.65
{[MesNpy]ZrR}[B(C6F5)4]
{[t-BuNON]ZrR}[B(C6F5)4]a
{rac-[C2H4(1-indenyl)2]ZrMe}[MeB(C6F5)3] C6H5CH3
{[MesNpy]HfR}[B(C6F5)4]
{[MesNpy]HfR}[HB(C6F5)3]
{[MesNpy]HfR}[HB(C6F5)3]
C6D5Br
C6D5Br
0.084
0.13
2.0
0.10
0.048
0.011
C6D5Br
C6D5Br
C6D5CD3
a
[t-BuNON]2- ) [(t-BuN-o-C6H4)2O]2- (see refs 36 and 49).
3, along with a value for {[t-BuNON]ZrR}[B(C6F5)4]
([t-BuNON]2- ) [(t-BuN-o-C6H4)2O]2- 49
and a recent
)
value obtained by Landis for a nonliving metallocene
system.29,30 The rates for the diamido/donor systems
vary by approximately an order of magnitude. However,
the metallocene system explored by Landis is ∼200
times more active than {[MesNpy]HfR}[HB(C6F5)3] in
toluene and ∼20 times more active than {[MesNpy]-
HfR}[B(C6F5)4] in bromobenzene. We ascribe the lower
activity of the diamido/donor system in general (com-
pared to a metallocene-based catalyst system) to a
delocalization of the metal’s positive charge out onto the
amido nitrogens. It is also worth noting that the anion
may interact more strongly with the cation in the more
“open” [MesNpy]2- systems than in a metallocenium ion
pair and, therefore, may further reduce the electrophi-
licity of the “cation” and also sterically block the olefin’s
access to the metal.
In future studies we hope to explore other [MAryl-
Npy]2- ligand systems (Aryl ) triisopropylphenyl24 or
3,5-dichlorophenyl66 are known), to explore other types
of ligands that appear to give rise to living catalysts, to
explore the possibility of polymerizing the appropriately
protected functionalized olefin (in which the functional-
ity is remote from the double bond) in a living manner,
and to explore the possibility of preparing asymmetric
diamido/donor catalysts that will allow isotactic poly-
mers to be prepared in a living manner, a phenomenon
that has been reported recently by both Sita10,12,17 and
Kol.13-15 We also hope to be able to prepare various alkyl
cations, even internal alkyl (e.g., isopropyl) cations,24
study the rate and mechanism of their decomposition,
and study the rate at which they react with various
olefins. In this manner we hope to address relative
stabilities of various alkyls and their reactivities toward
olefins directly.
Exp er im en ta l Section
Gen er a l P r oced u r es. All manipulations, with the excep-
tion of the synthesis of ligand precursors, were performed
under N2 in a glovebox or using standard Schlenk procedures.
Solvents were dried using conventional procedures.67 All
catalysts were activated in a glovebox free of ether, THF, and
other coordinating solvents. Chlorobenzene (HPLC grade) and
deuterated solvents were degassed and stored over and
distilled from CaH2. Commercial reagents were used without
further purification. NMR spectra were recorded on a Varian
Rate constants for the polymerization of 1-hexene at
0 °C that we have determined here are listed in Table
1
(60) Vanka, K.; Chan, M. S. W.; Pye, C. C.; Ziegler, T. Organo-
metallics 2000, 19, 1841.
(61) Vanka, K.; Ziegler, T. Organometallics 2001, 20, 905.
(62) Beck, W.; Sunkel, K. Chem. Rev. 1988, 88, 1405.
(63) Beck, S.; Lieber, S.; Schaper, F.; Geyer, A.; Brintzinger, H.-H.
J . Am. Chem. Soc. 2001, 123, 1483.
(64) Bochmann, M. Angew. Chem., Int. Ed. Engl. 1992, 31, 1181.
(65) Kesti, M. R.; Coates, G. W.; Waymouth, R. M. J . Am. Chem.
Soc. 1992, 114, 9679.
INOVA 500 spectrometer. H NMR chemical shifts are given
in ppm versus residual protons in the deuterated solvents as
follows: δ 7.16 C6D6, δ 2.09 toluene-d8 (methyl), δ 7.29 C6D5-
(66) Araujo, J . P.; Wicht, D. K.; Bonitatebus, P. J . J .; Schrock, R. R.
Organometallics 2001, 20, 5682.
(67) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Tinners, F. J . Organometallics 1996, 15, 1518.

4582 Organometallics, Vol. 22, No. 22, 2003
Mehrkhodavandi et al.
Br (most downfield resonance). 13C{1H} NMR chemical shifts
are given in ppm versus residual 13C in the solvents as
follows: δ 128.39 C6D6, δ 20.4 toluene-d8 (methyl), δ 122.25
C6D5Br (most upfield resonance). Inhibitors were dried thor-
oughly over CaH2 and distilled. Dialkyl complexes were
prepared as reported in the literature.25
7.83 (m, 1H, py-CH), 7.85 (m, 1H, anil-CH), 9.02 (m, 1H, py-
o-CH). 13C{1H} NMR (125 MHz, C6D5Br, 293 K): δ 18.13 (s,
o-CH3), 19.82 (s, o-CH3), 20.82 (s, p-CH3), 22.72 (s, CH3), 43.17
(s, CR4), 47.89 (s, N-CH3), 64.48 (s, CH2), 99.74 (s, Ar-C), (some
aryl peaks omitted here), 186.06 (s, Ar-C). Anal. Calcd for
C
53H44N4BF15Hf: C, 52.56; H, 3.66; N, 4.63. Found: C, 52.63;
H, 3.61; N, 4.57.
GPC analyses were carried out on a system equipped with
two J ordi-Gel DVB mixed bed columns (250 mm length × 10
mm inner diameter) in series. HPLC grade THF was supplied
at a flow rate of 1.0 mL/min with a Knauer 64 HPLC pump.
A Wyatt Technology mini Dawn light-scattering detector
coupled with a Knauer differential refractometer was em-
ployed. Data analysis was carried out using Astrette 1.2
software (Whatt Technology). Mn and Mw values for poly(1-
hexene) were obtained using dn/dc ) 0.076 mL/gr (Wyatt
Technology), and the auxiliary constant of the apparatus (5.9
In h ibition Stu d ies. P olym er iza tion of 1-Hexen e w ith
{[MesNp y]Hf(i-Bu )}[B(C6F 5)4] in th e P r esen ce of (i-P r )2O.
Solutions of [MesNpy]Hf(i-Bu)2 (0.0129 g, 0.0186 mmol) and
[Ph3C][B(C6F5)4] (0.0178 g, 0.0193 mmol), each in C6D5Br (0.4
mL), were prepared and cooled to -30 °C. The solutions were
mixed while still cold, and (i-Pr)2O (0.015 mL, 0.11 mmol) was
added and the mixture was stirred vigorously. 1-Hexene (0.200
mL, 1.60 mmol) was then added to the mixture. The NMR
samples were obtained directly from this solution and frozen
in liquid nitrogen prior to examination.
× 10-4) was calibrated using a polystyrene standard (Mn
)
2.2 × 105).
Other inhibition studies were carried out similarly.
Obser va tion of {[MesNp y]Hf(i-Bu )}[B(C6F 5)4]. Solutions
of [MesNpy]Hf(i-Bu)2 (0.0091 g, 0.013 mmol) and [Ph3C]-
[B(C6F5)4] (0.0122 g, 0.0132 mmol), each in C6D5Br (0.5 mL),
were prepared and cooled to -30 °C. The solutions were mixed
while still cold, and the resulting yellow solution was trans-
ferred to an NMR tube and frozen in liquid nitrogen within 2
min of the preparation of the sample. 1H NMR (500 MHz, C6D5-
Br, 273 K): δ 0.44 (d, 2H, Hf-CH2CH(CH3)2), 0.55 (d, 3H, Hf-
CH2CH(CH3)2), 1.28 (s, 3H, CH3), 1.57 (br s, 6H, o-CH3), 1.62
(s, 6H, CH2C(CH3)2), 1.73 (m, 1H, Hf-CH2CH(CH3)2), 2.18 (s,
6H, p-CH3), 2.35 (br s, 6H, o-CH3), 2.95 (d, 2H, CH2), 4.22 (d,
2H, CH2), 4.73 (s, 2H, CH2C(CH3)2), 5.45 (s, 1H, Ph3CH), 6.77
(br s, 2H, CH), 6.85 (br s, 2H, CH), 7.23 (m, 1H, py-CH), 7.38
(m, 1H, py-CH), 7.69 (m, 1H, py-CH), 8.54 (m, 1H, py-o-CH).
13C{1H} NMR (125 MHz, C6D5Br, 273 K): δ 18.57 (s, o-CH3),
20.73 (s, p-CH3), 24.20 (s, CH2C(CH3)2), 24.91 (s, CH3), 27.58
(s, Hf-CH2CH(CH3)2), 28.75 (s, Hf-CH2CH(CH3)2), 42.79 (s,
CR4), 56.79 (s, Ph3CH), 64.25 (s, CH2), 93.34 (s, Hf-CH2CH-
(CH3)2), 111.23 (s, CH2C(CH3)2). Some aryl peaks are omitted
from both the proton and the carbon NMR data.
Obser va tion of {[MesNp y]Zr (i-Bu )}[B(C6F 5)4]. Note: all
of the following manipulations were carried out in the absence
of light. Solutions of [MesNpy]Zr(i-Bu)2 (0.022 g, 0.036 mmol)
and [Ph3C][B(C6F5)4] (0.034 g, 0.036 mmol), each in C6D5Br
(0.5 mL), were prepared and cooled to -30 °C. The solutions
were mixed while still cold, and the resulting yellow/orange
solution was transferred to an NMR tube and frozen in liquid
nitrogen within 2 min of the preparation of the sample. 1H
NMR (500 MHz, C6D5Br, 243 K): δ 0.60 (d, 6H, Zr-CH2CH-
(CH3)2), 0.62 (d, 2H, Zr-CH2CH(CH3)2), 0.84 (m, 1H, CH2-
CH(CH3)2), 1.31 (s, 3H, CH3), ∼1.5 (br s, 6H, o-CH3), 1.62 (s,
6H, CH2C(CH3)2), 2.18 (s, 6H, p-CH3), 2.32 (br s, 6H, o-CH3),
2.79 (d, 2H, CH2), 3.98 (d, 2H, CH2), 4.73 (s, 1H, Ph3CH), 5.45
(s, 2H, CH2C(CH3)2), some aryl peaks omitted here, 7.36 (m,
1H, py-CH), 7.65 (m, 1H, py-CH), 8.39 (m, 1H, py-o-CH). 13C-
{1H} NMR (125 MHz, C6D5Br, 243 K): δ 18.65 (s, o-CH3), 20.79
(s, p-CH3), 24.30 (s, CH2C(CH3)2), 24.96 (s, CH3), 26.28 (s, Zr-
CH2CH(CH3)2), 27.533 (s, Zr-CH2CH(CH3)2), 44.76 (s, CR4),
56.64 (s, Ph3CH), 64.77 (s, CH2), 83.33 (s, Zr-CH2CH(CH3)2),
111.23 (s, CH2C(CH3)2). (Some aryl peaks are omitted.) The
reaction is cleanest when carried out with freshly prepared
and recrystallized [MesNpy]Zr(i-Bu)2.
Rea ction s of {[MesNp y]Zr (i-Bu )}[B(C6F 5)4] w ith 1-Hex-
en e: Kin etic Stu d ies. Note: all of the following manipula-
tions were carried out in the absence of light. Solutions of
[MesNpy]Zr(i-Bu)2 (0.0082 g, 0.0136 mmol) and [Ph3C]-
[B(C6F5)4] (0.0136 g, 0.0147 mmol), each in C6D5Br (0.8 mL),
were prepared and cooled to -30 °C. The solutions were mixed
while still cold, and 1-hexene (0.170 mL, 1.36 mmol) was added
to the mixture. The total volume was increased to 2 mL in a
volumetric flask. The NMR samples were obtained directly
from this solution and frozen with liquid nitrogen prior to
experimentation. The reaction is most reproducible when
carried out with freshly prepared and recrystallized [MesNpy]-
Zr(i-Bu)2. In two cases (107 and 220 equiv of 1-hexene with
[Zr] ) 15 and 7.3 mM, respectively) the samples were
quenched with methanol after polymerization was complete,
and the solvent was removed in vacuo. The polymer sample
was then redissolved in pentane, the solution was passed
through silica, and the pentane was removed in vacuo (16 h).
Rea ction s of Activa ted [MesNp y]Zr Me2 w ith 1-Hex-
en e: Kin etic Stu d ies. Solutions of [MesNpy]ZrMe2 (0.0298
g, 0.0572 mmol) and [Ph3C][B(C6F5)4] (0.0535 g, 0.0580 mmol),
each in C6D5Br (2 mL), were prepared and cooled to -30 °C.
The solutions were mixed while still cold, and hexamethyl-
benzene (HMB) (0.0582, 0.0359 mmol) was added to the
resulting orange solution. The total volume was increased to
5 mL in a volumetric flask, and the solution was stored at -30
°C for further use for up to 5 days. 13C labeling studies were
carried out in a similar fashion using [MesNpy]Zr13Me2.
The NMR samples were prepared by adding 1-hexene (0.100
mL, 0.800 mmol) to the above stock solution (0.900 mL, 0.0099
mmol) in a calibrated NMR tube. 1H NMR (500 MHz, C6D5-
Rea ction s of {[MesNp y]Hf(i-Bu )}[B(C6F5)4] with 1-Hex-
en e: Kin etic Stu d ies. Solutions of [MesNpy]Hf(i-Bu)2 (0.0091
g, 0.013 mmol) and [Ph3C][B(C6F5)4] (0.0122 g, 0.013 mmol),
each in C6D5Br (0.5 mL), were prepared and cooled to -30 °C.
The solutions were mixed while still cold, and 1-hexene (0.100
mL, 0.80 mmol) was added to the mixture. The NMR samples
were obtained directly from this solution and frozen in liquid
nitrogen prior to examination.
Polymerizations of the other olefins listed in Table 1 were
carried out similarly.
Rea ction s of {[MesNp y]Hf(i-Bu )}[B(C6F5)4] with 1-Hex-
en e: Bu lk P olym er iza tion . Solutions of [MesNpy]Hf(i-Bu)2
(0.0131 g, 0.0189 mmol) and [Ph3C][B(C6F5)4] (0.0177 g, 0.0192
mmol), each in C6H5Cl (1.5 mL), were prepared and cooled to
-30 °C. The solutions were mixed while still cold, and 1-hexene
(0.700 mL, 5.60 mmol) was added to the mixture. The samples
were stirred at 0 °C for 2 h and quenched with methanol. The
solvent was then removed in vacuo and the residue dissolved
in pentane. The solution was passed through silica and the
pentane removed in vacuo (16 h) at room temperature.
{[MesNp y]Hf(C6H4NMe2)}[HB(C6F 5)3]. Solutions of [Mes-
Npy]Hf(i-Bu)2 (0.324 g, 0.468 mmol) and B(C6F5)3 (0.240 g,
0.468 mmol) in C6H5Cl and toluene, respectively (total volume
) 3.0 mL), were prepared and cooled to -30 °C for 30 min.
The solutions were mixed while still cold, and the yellow
mixture was stirred at 0 °C for 24 h. The solvent was removed
in vacuo, and the residue was triturated with pentane to yield
1
yellow powder; yield 0.410 g (72%). H NMR (500 MHz, C6D5-
Br, 293 K): δ 1.36 (s, 3H, CH3), 1.50 (s, 6H, o-CH3), 2.03 (s,
6H, p-CH3), 2.13 (s, 6H, N-CH3), 2.32 (s, 6H, p-CH3), 2.94 (d,
2H, CH2), 4.42 (d, 2H, CH2), 6.71 (m, 1H, anil-CH), 6.43 (s,
2H, mes-CH), 6.73 (s, 2H, mes-CH), 6.99 (m, 1H, anil-CH), 7.21
(m, 1H, py-CH), 7.43 (m, 1H, py-CH), 7.47 (m, 1H, anil-CH),

Living Polymerization of 1-Hexene
Organometallics, Vol. 22, No. 22, 2003 4583
Br) spectra were obtained by comparison of the most downfield
methylene peak of 1-hexene δ 1.96 (m, 2H, CH2dCHCH2(CH2)2-
CH3) to the hexamethylbenzene peak 2.10 (s, 18H, CH3).
Activa tion of [MesNp y]Zr Me2 w ith [P h 3C][B(C6F 5)4].
Solutions of [MesNpy]ZrMe2 (0.021 g, 0.040 mmol) and [Ph3C]-
[B(C6F5)4] (0.037 g, 0.040 mmol), each in C6D5Br (0.6 mL), were
prepared and cooled to -30 °C. The solutions were mixed while
still cold, and the resulting orange solution was transferred
to an NMR tube. The following NMR data correspond to
activated [MesNpy]Zr(13Me)2, which was obtained in an analo-
and the resulting yellow solution was stirred at room temper-
ature for 5 min. The solvent was removed in vacuo and
remaining residue triturated with pentane for 2 h. The
resulting yellow powder was washed with pentane to remove
Ph3CMe and dried in vacuo; yield 0.320 g (75%). {[Trip-
Npy]2Zr213Me3}[B(C6F5)4] was prepared in a similar fashion
1
from [TripNpy]Zr13Me2. H NMR (500 MHz, C6D6, 295 K): δ
0.64 (d, 6H, CH3), 0.81 (d, 6H, CH3), 0.89 (s, 3H, CH3), 1.02 (s,
4.5H, Zr-CH3), 1.05 (d, 6H, CH3), 1.26 (d, 6H, CH3), 1.33 (d,
12H, CH3), 1.67 (m, 2H, CH), 2.83 (m, 2H, CH), 3.00 (d, 2H,
CH2), 3.56 (m, 2H, CH), 3.73 (d, 2H, CH2), 6.84 (m, 1H, py-
CH), 6.85 (s, 2H, CH), 6.91 (m, 1H, py-CH), 6.93 (s, 2H, CH),
7.39 (m, 1H, py-CH), 8.59 (m, 1H, py-o-CH). Anal. Calcd for
1
gous process. H NMR (500 MHz, C6D5Br, 295 K): δ 0.66 (d,
J CH ) 108 Hz, 3H, Zr-13CH3), 1.21 (s, 3H, CH3), 1.22 (s, 6H,
o-CH3), 2.03 (d, J CH ) 128 Hz, 3H, 13CH3Ph3), 2.15 (s, 6H,
p-CH3), 2.29 (s, 6H, o-CH3), 2.78 (d, 2H, CH2), 3.67 (d, 2H, CH2),
6.70 (s, 2H, CH), 6.80 (s, 2H, CH), some aryl peaks omitted
here, 6.87 (m, 1H, py-CH), 7.51 (m, 1H, py-CH), 7.92 (m, 1H,
py-o-CH). 13C{1H} NMR (125 MHz, C6D5Br, 295 K): δ 17.14
(s, o-CH3), 18.10 (s, o-CH3), 20.80 (s, p-CH3), 23.81 (s, CH3),
30.42 (s, 13CH3Ph3), 38.81 (br s, Zr-13CH3), 48.95 (s, CR4), 67.40
(s, CH2) (Some aryl resonances are omitted.)
C
105H123N6Zr2BF20: C, 61.75; H, 6.07; N, 4.11. Found: C, 61.64;
H, 6.18; N, 3.98.
Activa tion of [MesNp y]HfMe2 w ith [P h 3C][B(C6F 5)4].
Solutions of [MesNpy]HfMe2 (0.0102 g, 0.0168 mmol) and
[Ph3C][B(C6F5)4] (0.0162 g, 0.0176 mmol), each in C6D5Br (0.5
mL), were prepared and cooled to -30 °C. The solutions were
mixed while still cold, and the resulting orange solution was
transferred to an NMR tube.
Isola tion of {[MesNp y]2Zr 2Me3}[B(C6F 5)4]. Solutions of
[MesNpy]ZrMe2 (0.205 g, 0.394 mmol) and [Ph3C][B(C6F5)4]
(0.182 g, 0.197 mmol), each in C6H5Cl (1.5 mL), were prepared
and cooled to -30 °C. The solutions were mixed while still cold,
and the resulting yellow solution was stirred at room temper-
ature for 5 min. The solvent was removed in vacuo and the
remaining residue triturated with pentane for 2 h. The
resulting yellow powder was washed with pentane to remove
Ph3CMe and dried in vacuo; yield 0.233 g (64%). The 13C-
labeled analogue was prepared in a similar fashion from
[MesNpy]Zr13Me2. 1H NMR (500 MHz, C6D6, 295 K): δ 0.74
(s, 4.5 H, Zr-CH3), 0.86 (s, 3H, CH3), 1.22 (s, 6H, o-CH3), 2.14
(s, 6H, p-CH3), 2.27 (s, 6H, o-CH3), 2.65 (d, 2H, CH2), 3.67 (d,
2H, CH2), 6.67 (m, 1H, py-CH), 6.73 (s, 2H, CH), 6.78 (s, 2H,
CH), 6.88 (m, 1H, py-CH), 7.18 (m, 1H, py-CH), 7.99 (m, 1H,
{[MesNp y]2Hf2Me3}[B(C6F 5)4]. Solutions of [MesNpy]H-
fMe2 (0.318 g, 0.523 mmol) in C6H5Cl (1.5 mL) and B(C6F5)3
(0.139 g, 0.271 mmol) in toluene (1.5 mL) were prepared and
cooled to -30 °C. The solutions were mixed while still cold,
and the resulting clear, colorless solution was stirred at room
temperature for 15 min. The solvent was removed in vacuo
and the remaining white residue triturated with pentane for
13
2 h and dried in vacuo; yield 0.628 g (69%). {[TripNpy]2Hf2
-
Me3}[B(C6F5)4] was prepared in a similar fashion from [TripNpy]-
Hf 13Me2. 1H NMR (500 MHz, C6D5Br, 295 K): δ 1.00 (s, 4.5H,
Zr-CH3), 1.17 (s, 3H, B-CH3), 1.18 (s, 3H, CH3), 1.23 (s, 6H,
o-CH3), 2.17 (s, 6H, o-CH3), 2.31 (s, 6H, p-CH3), 2.90 (d, 2H,
CH2), 3.86 (d, 2H, CH2), 6.72 (s, 2H, CH), 6.81 (s, 2H, CH),
6.92 (m, 1H, py-CH), 7.31 (m, 1H, py-CH), 7.52 (m, 1H, py-
CH), 7.89 (m, 1H, py-o-CH). 13C{1H} NMR (125 MHz, C6D5-
Br, 233 K): δ 11.12 (br s, B-13CH3), 32.52 (s, Hf-13CH3), 41.35
(s, Hf-13CH3). Anal. Calcd for C76H78N6Hf2BF15: C, 52.82; H,
4.55; N, 4.86. Found: C, 52.67; H, 4.62; N, 4.73.
1
py-o-CH). H NMR (500 MHz, C6D5Br, 295 K): δ 0.66 (d, J CH
) 108 Hz, H, Zr-13CH3), 1.21 (s, 3H, CH3), 1.22 (s, 6H, o-CH3),
2.15 (s, 6H, p-CH3), 2.29 (s, 6H, o-CH3), 2.78 (d, 2H, CH2), 3.67
(d, 2H, CH2), 6.70 (s, 2H, CH), 6.80 (s, 2H, CH), 6.87 (m, 1H,
py-CH), 7.25 (m, 1H, py-CH), 7.51 (m, 1H, py-CH), 7.92 (m,
1H, py-o-CH). Anal. Calcd for C93H75N6Zr2BF20: C, 60.38; H,
4.09; N, 4.54. Found: C, 60.46; H, 4.19; N, 4.59.
Ack n ow led gm en t. This research was supported in
part by the Department of Energy (DE-FG02-86ER13564)
and the U.S. Army through the Institute for Soldier
Nanotechnologies, under Contract DAAD-19-02-D-0002
with the U.S. Army Research Office. The content does
not necessarily reflect the position of the Government,
and no official endorsement should be inferred. We also
thank NSERC of Canada for a postgraduate scholarship
(PGS B) to P.M.
Activa tion of [Tr ip Np y]Zr Me2 w ith [P h 3C][B(C6F 5)4].
Solutions of [TripNpy]ZrMe2 (0.0102 g, 0.0148 mmol) and
[Ph3C][B(C6F5)4] (0.0139 g, 0.0151 mmol), each in C6D5Br (0.5
mL), were prepared and cooled to -30 °C. The solutions were
mixed while still cold, and the resulting orange solution was
transferred to an NMR tube.
{[Tr ip Np y]2Zr 2Me3}[B(C6F 5)4]. Solutions of [TripNpy]-
ZrMe2 (0.290 g, 0.421 mmol) and [Ph3C][B(C6F5)4] (0.194 g,
0.210 mmol), each in C6H5Cl (1.5 mL), were prepared and
cooled to -30 °C. The solutions were mixed while still cold,
OM030438I