Sterically hindered aluminum alky is: Weakly interacting scavenging agents of use in olefin polymerization

Article abstract of DOI:10.1021/om060474s

Statically hindered aluminum methyl compounds derived from reaction of hindered phenols with AlMe₃ (i.e., MeAl(BHT)₂ and MeAl(BHT*)₂; BHT = 2,6-di-te>t-butyl-4-methylphenoxide; BHT* = 2,4,6-triterf-butylphenoxide) are useful scavenging agents in olefin polymerization using metallocene catalysts. They do not, or only slowly, react with activators such as B(C₆F₅)₃ or [Ph₃C][B(C₆F₅)₄] at 25 °C, nor do they coordinate to or react with metallocenium ion-pairs derived from metallocene dialkyls and these activators. A mixture of AlMe₃ and a large excess of MeAl(BHT)₂ proves advantageous for catalysts that are susceptible to reaction with BHT-H, the hydrolysis product of MeAl(BHT) ₂. Ethylene polymerization experiments establish that the activity of [Cp₂ZrMe][MeB(C₆F₅)₃] is only slightly inhibited by AlMe₃ in the presence of a significant excess of MeAl(BHT)₂. Spectroscopic studies have revealed that AlMe ₃ is in equilibrium with MeAl(BHT)₂, forming Me ₂Al(BHT). At low temperature using ¹³C NMR spectroscopy, a 1:1 mixture of AlMe₃ and MeAl(BHT)₂ is shown to consist of Al₂Me₆, MeAl(BHT)₂, and primarily Me ₂Al(M-BHT)₂AlMe₂. A higher temperature, both intra- and intermolecular exchange of both Al-Me and Al-BHT groups, coupled with the temperature dependence of the various equilibria involved, lead to ¹H and ¹³C NMR spectra that are consistent with monomeric Me₂Al(BHT). ¹H and ¹⁹F NMR spectroscopic studies of mixtures of the ion-pairs [Me₂C(Cp)IndMMe][MeB(C ₆Fs)₃] (M = Zr, Hf) or [Me₂SiCp ₂ZrMe][MeB(C₆F₅)₃] with various quantities of AlMes in the presence of MeAl(BHT)₂ were conducted. The AlMe₃-mediated degradation of ion-pairs that are susceptible to B(CeF₅)₃ dissociation is largely absent in the presence of excess MeAl(BHT)₂, although reversible formation of [Me ₂SiCp₂Zr(μ-Me)₂AlMe₂][MeB(C ₆F₅)₃] and related adducts is observed at low ratios of MeAl(BHT)₂ to AlMe₃.

Full text of DOI:10.1021/om060474s

Organometallics 2006, 25, 5083-5092
5083
Sterically Hindered Aluminum Alkyls: Weakly Interacting
Scavenging Agents of Use in Olefin Polymerization
Russell A. Stapleton, Abdulaziz Al-Humydi, Jianfang Chai, Brandon R. Galan, and
Scott Collins*
Department of Polymer Science, The UniVersity of Akron, Akron, Ohio 44325-3909
ReceiVed May 29, 2006
Sterically hindered aluminum methyl compounds derived from reaction of hindered phenols with AlMe
3
(
i.e., MeAl(BHT)
tert-butylphenoxide) are useful scavenging agents in olefin polymerization using metallocene catalysts.
They do not, or only slowly, react with activators such as B(C or [Ph C][B(C ] at 25 °C, nor do
they coordinate to or react with metallocenium ion-pairs derived from metallocene dialkyls and these
activators. A mixture of AlMe and a large excess of MeAl(BHT) proves advantageous for catalysts
that are susceptible to reaction with BHT-H, the hydrolysis product of MeAl(BHT) . Ethylene
polymerization experiments establish that the activity of [Cp ZrMe][MeB(C ] is only slightly inhibited
by AlMe in the presence of a significant excess of MeAl(BHT) . Spectroscopic studies have revealed
that AlMe is in equilibrium with MeAl(BHT) , forming Me Al(BHT). At low temperature using
NMR spectroscopy, a 1:1 mixture of AlMe and MeAl(BHT) is shown to consist of Al Me , MeAl-
BHT) , and primarily Me Al(µ-BHT) AlMe . A higher temperature, both intra- and intermolecular
exchange of both Al-Me and Al-BHT groups, coupled with the temperature dependence of the various
2 2
and MeAl(BHT*) ; BHT ) 2,6-di-tert-butyl-4-methylphenoxide; BHT* ) 2,4,6-tri-
F
6 5
)
3
3
6 5 4
F )
3
2
2
2
6 5 3
F )
3
2
1
3
3
2
2
C
3
2
2
6
(
2
2
2
2
1
13
equilibria involved, lead to H and C NMR spectra that are consistent with monomeric Me
2
Al(BHT).
C(Cp)IndMMe][MeB(C
] with various quantities of AlMe in the presence of
-mediated degradation of ion-pairs that are susceptible to B(C
dissociation is largely absent in the presence of excess MeAl(BHT)
1
19
H and F NMR spectroscopic studies of mixtures of the ion-pairs [Me
M ) Zr, Hf) or [Me SiCp ZrMe][MeB(C
were conducted. The AlMe
2
6 5 3
F ) ]
(
2
2
6
5
F )
3
3
MeAl(BHT)
2
3
F )
6 5 3
2
, although reversible formation of
[
Me
2
SiCp
2
Zr(µ-Me)
2
AlMe
2
][MeB(C
6 5
F )
3
] and related adducts is observed at low ratios of MeAl(BHT)
2
to AlMe
3
.
3
Introduction
[Ph3C][B(C6F5)4] and metallocene dichlorides, the activation
process is complicated and leads to the formation of different
catalyst precursors as recent mechanistic studies have shown.3
Further, a specific order of catalyst/cocatalyst introduction is
required due to the incompatibility of [Ph3C][B(C6F5)4] with
higher AlR3 (due to facile â-H abstraction; even with AlMe3
slower Me abstraction and activator degradation is observed).3b
In the case of metallocene dialkyls activated by B(C6F5) or
related compounds, the use of preformed ion-pairs is preferable.
In situ generation of these ion-pairs in the presence of AlR3
can be complicated by competing AlR3-mediated degradation
The mechanistic study of metallocene-catalyzed olefin poly-
merization has been considerably simplified through the devel-
opment of discrete activators such as B(C6F5)3, which give rise
a-c
1
to thermally stable and isolable ion-pairs. The ion-pairs can
be characterized in both the solid state and solution, and a wealth
of information is now available on their intrinsic behavior.
Because of the extreme air- and moisture-sensitivity of these
compounds, polymerization studies may require the use of high-
vacuum techniques and/or relatively high concentrations of these
compounds (ca. 0.1 mM). In contrast, MAO-activated metal-
locene catalysts are typically generated using micromolar
concentrations of metallocene complexes in efficiently stirred
autoclave reactors, where a large excess of MAO serves as both
scavenger and activator.2
4
of B(C6F5)3. Even when preformed ion-pairs are employed,
the amount of AlR3 must be controlled to avoid reversible
catalyst inhibition, which results from formation of thermally
stable, but unreactive adducts of the type [Cp2Zr(µ-R)2AlR2]-
MeB(C6F5)3]. Finally, some ion-pairs that are prone to
5
[
Although trialkylaluminum compounds are used as scavengers
and alkylating agents) in polymerization studies involving
6
dissociation of B(C6F5)3 (e.g., hafnocene complexes ) can be
readily degraded in the presence of AlR3 compounds (vide
(
7
*
To whom correspondence should be addressed. E-mail: collins@
infra).
uakron.edu.
(
1) (a) Bochmann, M.; Lancaster, S. J.; Hannant, M. D.; Rodriguez, A.;
(3) (a) Gotz, C.; Rau, A.; Luft, J. J. Mol. Catal. A: Chem. 2002, 184,
95. (b) Bochmann, M.; Lancaster, S. J. Organometallics 1993, 12, 633. (c)
Bochmann, M.; Lancaster, S. J. J. Organomet. Chem. 1992, 434, C1. (d)
Chien, J. C. W.; Tsai, W.-M.; Rausch, M. D J. Am. Chem. Soc. 1991, 113,
8570.
Schormann, M.; Walker, D. A.; Woodman, T. J. Pure Appl. Chem. 2003,
7
1
5, 1183-1195. (b) Chen, E. X.-Y.; Marks, T. J. Chem. ReV. 2000, 100,
391-1434.
(2) (a) Kaminsky, W. J. Polym. Sci., Part A: Polym. Chem. 2004, 42,
3
1
911-3921. (b) Zurek, E.; Ziegler, T. Prog. Polym. Sci. 2004, 29, 107-
(4) (a) Klosin, J.; Roof, G. R.; Chen, E. X.-Y.; Abboud, K. A.
Organometallics 2000, 19, 4684-4686. (b) Bochmann, M.; Sarsfield, M.
J. Organometallics 1998, 17, 5908-5912.
(5) Bochmann, M.; Lancaster, S. J. Angew. Chem. 1994, 106, 1715-
18.
48. (c) Bryliakov, K. P.; Semikolenova, N. V.; Yudaev, D. V.; Zakharov,
V. A.; Brintzinger, H. H.; Ystenes, M.; Rytter, E.; Talsi, E. P. J. Organomet.
Chem. 2003, 683, 92-102. (d) Babushkin, D. E.; Brintzinger, H. H. J. Am.
Chem. Soc. 2002, 124, 12869-12873.
1
0.1021/om060474s CCC: $33.50 © 2006 American Chemical Society
Publication on Web 08/24/2006

5
084 Organometallics, Vol. 25, No. 21, 2006
Stapleton et al.
Several years ago, it occurred to us that if one could use
Scheme 1
8
monomeric alkylaluminums as scavenging agents, both inhibi-
tion of catalysts and degradation of organoborane or even borate
activators might be mitigated, thus simplifying the conduct of
olefin polymerization using discrete, metallocenium ion-pairs
at least in autoclave reactors. As the key step in both processes
involves formation of hetero-dinuclear species with µ-R bridges,
by definition, a monomeric alkylaluminum should not as readily
participate in such processes. Finally, we anticipated that the
use of a discrete and “weakly interacting” alkylaluminum
scavenger of this type would lead to chemically cleaner systems
useful in mechanistic study of metallocene catalysts and catalyst
precursors under conditions that are relevant to an industrial
setting (i.e., efficiently stirred autoclave reactors).
We were attracted to the use of MeAl(BHT)2 for this purpose
due to its crystallinity and ready availability from inexpensive
9
starting materials. Also, since this compound lacks â-H atoms,
it should be more compatible with activators such as [Ph3C]-
[
B(C6F5)4]. This compound proved to be an effective scavenger
hydrolysis product of MeAl(BHT)2 is in fact the corresponding
of use in ethylene or propene polymerization using either
preformed or in situ generated ion-pairs derived from a variety
of metallocene and even some nonmetallocene catalysts. It
13
phenol (BHT-H) rather than CH4. Preliminary work established
that this reaction proceeds through formation of a tetrahedral
adduct (BHT)2AlMe‚OH2, which was structurally character-
1
0
could be used in large excess without inhibiting catalyst activity
14
and is not pyrophoric and thus easily handled.11 Activities in
ized and which decomposed in THF solution to form 1.6 equiv
of BHT-H, 0.2 equiv of MeAl(BHT)2, and only 0.1 equiv of
CH4 (Scheme 1). Material balance required the formation of a
species with the empirical formula Al0.8Me0.7O0.7(OH)0.3, and
though formation of a methylaluminoxane species was evident
in the NMR spectra of these mixtures, it has resisted purification
and characterization.
ethylene polymerization using, for example, [Cp2ZrMe][MeB-
(C6F5)3] were ca. 10 times higher than could be achieved in the
i
presence of conventional AlR3 scavengers (R ) Me, Et, Bu
etc.).
1
0e,f
On the other hand, it became apparent that with less hindered
and generally much less active) catalysts, the use of MeAl-
BHT)2 as scavenger inhibited polymerization for reasons that
(
(
In a preliminary communication we showed that although
BHT-H does react with [Cp2ZrMe][MeB(C6F5)3] to form [Cp2-
ZrBHT][MeB(C6F5)3] and CH4 (Scheme 1), the rate of that
reaction is about 7 orders of magnitude slower than the apparent
were not apparent. NMR spectroscopic studies indicated no
reaction nor even evidence for coordination to the ion-pairs.
It is during these studies that we became aware that the
1
2
14
rate of ethylene insertion at 25 °C. Consequently, the presence
of BHT-H during catalysis using many catalysts is innocuous
and one can even deliberately add significant amounts of (dry)
BHT-H to a polymerizing system without ill effect.14
(
6) (a) Beswick, C. L.; Marks, T. J. J. Am. Chem. Soc. 2000, 122, 10358-
1
1
0370. (b) Deck, P. A.; Beswick, C. L.; Marks, T. J. J. Am. Chem. Soc.
998, 120, 1772-1784.
(7) Collins, S.; Dai, C.; Li, Z.; Mohammed, M.; Tian, J.; Tomazsewski,
However, for other less hindered and less active catalysts,
where the rate of reaction of BHT-H with the ion-pair is
essentially instantaneous on mixing at typical NMR concentra-
R.; Vollermhaus, R. Polym. Mater. Sci. Eng. 2001, 84, 921-22.
8) (a) Roesky, H. W.; Walawalkar, M. G.; Murugavel, R. Acc. Chem.
Res. 2001, 34, 201-211. (b) Wehmschulte, R. J.; Power, P. P. Polyhedron
000, 19, 1649-1661. (c) Healy, M. D.; Power, M. B.; Barron, A. R. Coord.
Chem. ReV. 1994, 130, 63-135.
9) (a) Healy, M. D.; Wierda, D. A.; Barron, A. R. Organometallics 1988,
, 2543-2548. (b) Shreve, A. P.; Mulhaupt, R.; Fultz, W.; Calabrese, J.;
(
2
12
tions (1-10 mM), it was evident that a scavenger that only
forms alkane on initial hydrolysis was required. Motivated by
(
reports in the patent literature on the utility of R2AlBHT (R )
7
i
15
Robbins, W.; Ittel, S. D. Organometallics 1988, 7, 409-416. (c) Skow-
ronska-Ptasinska, M.; Staowieyski, K. B.; Pasynkiewicz, S.; Carewska, M.
J. Organomet. Chem. 1978, 160, 403-409. (d) Skowronska-Ptasinska, M.;
Starowieyski, K. B.; Pasynkiewicz, S. J. Organomet. Chem. 1977, 141,
Bu, etc.) and related compounds, we investigated the utility
16
of mixtures of MeAl(BHT)2 and AlMe3 and have discovered
that this combination is very effective for use in ethylene
1
49-156. (e) Skowronska-Ptasinska, M.; Starowieyski, K. B.; Pasynkiewicz,
S. J. Organomet. Chem. 1975, 90, C43-C44.
10) (a) Al-Humydi, A.; Garrison, J. C.; Youngs, W. J.; Collins, S.
(13) Related chemoselective protonolyses of aluminum phenoxides or
their donor adducts are known. (a) VanPoppel, L. G.; Bott, S. G.; Barron,
A. R. J. Chem. Crystallogr. 2001, 31, 417-420. (b) Taden, I.; Kang, H.
C.; Massa, W.; Spaniol, T. P.; Okuda, J. Eur. J. Inorg. Chem. 2000, 441-
445. (c) McMahon, C. N.; Barron, A. R. J. Chem. Soc., Dalton Trans. 1998,
3703-3704. (d) Healy, M. D.; Leman, J. T.; A, R., Barron J. Am. Chem.
Soc. 1991, 113, 2776-2777. (e) Healy, M. D.; Ziller, J. W.; Barron, A. R.
Organometallics 1991, 10, 597-608. (f) Healy, M. D.; Mason, M. R.;
Gravelle, P. W.; Bott, S. G.; Barron, A. R. J. Chem. Soc., Dalton Trans.
1993, 441-454.
(14) Stapleton, R. A.; Galan, B. R.; Collins, S.; Simons, R. S.; Garrison,
J. C.; Youngs, W. J. J. Am. Chem. Soc. 2003, 125, 9246-9247.
(15) (a) Chen, E. Y.; Kruper, W. J., Jr.; Roof, G. R.; Schwartz, D. J.;
Storer, J. W. PCT Int. Appl. WO 2000,009514, 2000, 32 pp. (b) Chen, E.
Y.; Kruper, W. J., Jr.; Roof, G. R.; Schwartz, D. J.; Storer, J. W. PCT Int.
Appl. WO 2000,009513, 2000, 32 pp. (c) Rosen, R. K.; Vanderlende, D.
D. PCT Int. Appl. WO 9735893, 1997, 27 pp. (d) Rosen, R. K.; Stevens,
J. C.; Tracy, J. C. PCT Int. Appl. WO 9727228, 1997, 23 pp.
(16) For an early report on the use of this mixture to activate Cp2ZrCl2
for ethylene polymerization see: Reddy, S. S.; Shashidhar, G.; Sivaram, S.
Macromolecules 1993, 26, 1180-1182. This mixture was no more effective
than AlMe3 at the same concentrations, etc.
(
Organometallics 2005, 24, 193-196. (b) Mohammed, M.; Nele, M.; Al-
Humydi, A.; Xin, S.; Stapleton, R. A.; Collins, S. J. Am. Chem. Soc. 2003,
1
25, 7930-7941. (c) Metcalfe, R. A.; Kreller, D. I.; Tian, J.; Kim, H.;
Taylor, N. J.; Corrigan, J. F.; Collins, S. Organometallics 2002, 21, 1719-
726. (d) Vollmerhaus, R.; Rahim, M.; Tomaszewski, R.; Xin, S.; Taylor,
1
N. J.; Collins, S. Organometallics 2000, 19, 2161-2169. (e) Williams, V.
C.; Irvine, G. J.; Piers, W. E.; Li, Z.; Collins, S.; Clegg, W.; Elsegood, M.
R. J.; Marder, T. B. Organometallics 2000, 19, 1619-1621. (f) Williams,
V. C.; Dai, C.; Li, Z.; Collins, S.; Piers, W. E.; Clegg, W.; Elsegood, M.
R. J.; Marder, T. B. Angew. Chem., Int. Ed. 1999, 38, 3695-3698.
(11) For other studies involving the use of MeAl(OAr)2 or methylalu-
minoxane modified by hindered phenols see: (a) Busico, V.; Cipullo, R.;
Cutillo, F.; Friedrichs, N.; Ronca, S.; Wang, B. J. Am. Chem. Soc. 2003,
1
7
25, 12402-12403. (b) Kissin, Y. V. Macromolecules 2003, 36, 7413-
421. (c) R o¨ sch, J. Eur. Pat. Appl. EP 0781783, 1997, 14 pp. (d) Tran, N.
H.; Davenport, D. L.; Malpass, D. B.; Rabbit, C. S. Eur. Pat. Appl. EP
561476, 1993, 20 pp. (e) Marks, T. J.; Yang, X.; Mirviss, S. B. PCT Int.
Appl. WO 9410180, 1994, 18 pp.
12) Tomaszewski, R.; Vollmerhaus, R.; Al-Humydi, A.; Wang, Q.;
Taylor, N. J.; Collins, S. Can. J. Chem. 2006, 84, 214-224.
0
(

Sterically Hindered Aluminum Alkyls
Organometallics, Vol. 25, No. 21, 2006 5085
Figure 1. 1H (300 MHz, C
D
, left) and 19F NMR (282 MHz, C
and B(C
D
3
, right) spectra of (a) a 5:1 mixture of MeAl(BHT)
. Signals due to [Cp ZrMe][MeB(C ] are highlighted in gray.
6
6
6
6
2 2
and [Cp ZrMe]-
[
MeB(C ] and (b) a 5:1 mixture of MeAl(BHT)
6
F
5
)
3
2
F
6 5
)
2
6 5 3
F )
polymerization. In this paper we provide details of this work
and the mechanistic reasons underlying the efficacy of these
hindered scavengers in olefin polymerization using metallocene
catalysts.
elevated temperature. Further, competing decomposition of the
ion-pair results in disappearance of labeled Zr-Me at a faster
rate compared to that at which labeled MeAl(BHT)2 appears.
Thus, the observed rate constant and activation free energy
are based on the latter data. An estimate for the pseudo-first-
13
order rate constant for scrambling of the label from Zr- Me
Results and Discussion
-
5
-1
into Al-Me is 4.76 × 10 s . As [MeAl(BHT)2] ) 0.106 M
Compatibility of MeAl(BHT)2 with Metallocenium Ions
in this experiment, this corresponds to a second-order rate
1
and Common Activators. As shown in Figure 1a, the H and
-4
-1 -1
q
-1
constant of 4.49 × 10
M
s
or ∆G ) 23.2 kcal mol at
1
9
F NMR spectra of a mixture of [Cp2ZrMe][MeB(C6F5)3] and
excess MeAl(BHT)2 ([Zr] ) 0.02 M, [Al] ) 0.10 M) shows no
4
0 °C. Thus, formation of a transient hetero-dinuclear adduct
between [Cp2ZrMe][MeB(C6F5)3] and MeAl(BHT)2 must occur
during this exchange process, but it is evident that the equilib-
rium is quite unfavorable compared to simple AlR3, and AlMe3
1
1
reaction between these two species. Further, the 2D H- H
NOESY spectrum of this mixture (see the Supporting Informa-
tion) shows no evidence of exchange correlation between Al-
Me and Zr-Me (or B-Me) groups despite the use of mixing times
as long as 1.0 s. This puts an upper limit on the rate constant
for Me exchange between these two compounds of k ) 1.9 ×
5
in particular.
MeAl(BHT)2 is inert toward prototypical metallocenium ion-
pairs, as illustrated above. It is also unreactive toward Lewis
acidic organoboranes used to activate metallocene dialkyls. The
-
3 -1
1
0
s
at 22 °C based on the S:N ratios of the diagonal peaks
1
19
H and F NMR spectra of excess MeAl(BHT)2 and B(C6F5)3
17
in this 2D spectrum. This corresponds to a lower limit for
∆
20
(Figure 1b), the chelating diborane 1,2-C6F4[B(C6F5)2]2, or the
q
-1
G g 21.0 kcal mol at 22 °C. On the other hand, the
21
strongly Lewis acidic 9,10-(C6F4)2[B(C6F5)]2 are unchanged
compared to those of the pure compounds.
expected cross-peaks between B-Me and Zr-Me resonances due
to competing ion-pair reorganization and borane dissociation
are seen (see the Supporting Information). A rough estimate of
the total exchange rate constant based on the diagonal versus
10f
MeAl(BHT)2 does react slowly with [Ph3C][B(C6F5)4] to form
a mixture of oligomeric methylaluminum compounds of unde-
termined structure. The Al-Me resonances are undiminished in
intensity during degradation, although they become broad, as
do all other resonances (Figure 2a). Additional signals charac-
teristic of aliphatic Me groups appear at δ 0.7-0.9, suggesting
1
8
-1
q
cross-peak intensities is 0.043 s , corresponding to ∆∆G )
-1
1
9.3 kcal mol at 295 K. This value is in reasonable agreement
with data reported in the literature for unbridged ion-pairs of
1
this type.
t
degradation or rearrangement of the Ar- Bu groups during this
It is known from the work of Siedle and co-workers that
degenerate exchange of Me groups occurs between Cp2ZrMe2
process. The only readily identifiable product is Ph C-H (δ 5.31,
Figure 2a), while the corresponding F NMR spectra indicate
no anion degradation.
3
1
9
q
-1
and MeAl(BHT)2 with ∆G ) 22.4 kcal mol at 22 °C based
on kinetic measurements using 13C-labeled materials.
19
In monitoring these reactions, it is apparent that Ph CH is
3
1
3
Surprisingly, the analogous process involving [Cp2Zr Me],
MeB(C6F5)3], and MeAl(BHT)2 appears equally slow at room
formed about 2 times faster than the rate of MeAl(BHT)
2
[13
-4
-1 -1
consumption (kobs ) 5.8 vs 3.0 × 10
M
s ). Further, only
temperature. Under pseudo-first-order conditions using a 5-fold
excess of MeAl(BHT)2 there is limited, if any, scrambling on
mixing at 22 °C at [Zr] ) 0.02 M (see the Supporting
Information). Ion-pair thermal instability limited the range of
temperature and times over which this reaction could be
investigated, but satisfactory data were obtained at 40 °C.
Analysis of these data is also complicated by the fact that
exchange of Zr-Me and B-Me groups is rapid on the chemical
time scale, while the latter resonances are line-broadened at this
catalytic amounts of [Ph C][B(C F ) ] are required for this
3
6 5 4
reaction. Taken together, the results suggest that [Ph C]-
3
[B(C F ) ] initiates degradation of MeAl(BHT) , possibly by
6
5 4
2
initial hydride abstraction, but is otherwise uninvolved in the
subsequent degradation of this compound. Hydride abstraction
by [Ph C][B(C F ) ] would be expected to generate a p-quinone
3
6 5 4
methide (eq 1); normally these species are reasonably stable
when derived from hindered phenols such as BHT-H,22 but the
Lewis acidic aluminum center would make this species highly
(
17) Chen, M.-C.; Roberts, J. A. S.; Marks, T. J. J. Am. Chem. Soc.
(20) Williams, V. C.; Piers, W. E.; Clegg, W.; Elsegood, M. R. J.; Collins,
S.; Marder, T. B. J. Am. Chem. Soc. 1999, 121, 3244-3245.
(21) Metz, M. V.; Schwartz, D. J.; Stern, C. L.; Nickias, P. N.; Marks,
T. J. Angew. Chem., Int. Ed. 2000, 39, 1312-1316.
2
004, 126, 4605-25.
(
18) Perrin, C. L.; Dwyer, T. J. Chem. ReV. 1990, 90, 935-967.
(19) Siedle, A. R.; Newmark, R. A.; Lamanna, W. M.; Schr o¨ pfer, J. N.
Polyhedron 1990, 9, 301-308.
(22) Filar, L. J.; Winstein, S. Tetrahedron Lett. 1960, 9-16.

5
086 Organometallics, Vol. 25, No. 21, 2006
Stapleton et al.
Figure 2. 1H NMR spectra (300 MHz, C
B(C ] and MeAl(BHT*) (right) as a function of time. Note that the latter units are in hours [vs minutes in the case of MeAl(BHT)
C]-
].
6
5 3 6 5 4 2 3
D -Br, 25 °C) of a 1:1 mixture of (a) [Ph C][B(C F ) ] and MeAl(BHT) (left) and (b) [Ph
[
F
6 5
)
4
2
2
electrophilic.23 Since the use of BHT-H in the presence of AlCl3,
as a reagent for the electrophilic tert-butylation of other aromatic
compounds, is known,24 it is possible this or a related process
is involved in the further degradation of MeAl(BHT)2.
from the second-order rate constant for degradation (kobs ) 3.0
-4
-1 -1
× 10
M
s ), and the activator and scavenger concentrations
used, it can be estimated that minimal degradation of either
compound could have occurred (i.e., the extent of [Ph3C]-
[
B(C6F5)4] or MeAl(BHT)2 degradation within 6 min corre-
-
10
sponds to a decrease in concentration of (2-4) × 10
M).
Mixtures of MeAl(BHT)2 with Al2Me6. As indicated in the
Introduction, unhindered or less reactive polymerization catalysts
1
2
can be deactivated by BHT-H, the kinetic hydrolysis product
1
4
of MeAl(BHT)2. To that end, we considered the use of R2-
Al(OAr) (R ) Me, etc.) and related compounds as scavengers
that would be prone to formation of alkane on hydrolysis. These
compounds exist as monomers, dimers, or even higher-order
aggregates depending on both R and OAr; when polynuclear,
they are almost invariably bridged by the phenoxide group.
Structurally characterized examples of mono- and dinuclear
As MeAl(BHT)2 has benzylic C-H bonds, which would be
most susceptible to either hydrogen or hydride abstraction, we
investigated the compatibility of MeAl(BHT*)2 (BHT* ) 2,4,6-
tri-tert-butylphenoxide) with [Ph3C][B(C6F5)4]. Degradation of
this material was much slower (ca. 33× slower) than in the
case of MeAl(BHT)2. Although only trace amounts of Ph3C-H
formed in this reaction, it would appear that similar Al
byproducts were formed (Figure 2b). Here, however, there was
a noticeable loss of Al-Me intensity over time in this experiment.
It is quite possible that the degradation of both of these
aluminum alkyls is redox-initiated by [Ph3C][B(C6F5)4] and in
the case of MeAl(BHT)2 is accompanied by hydrogen abstrac-
25
compounds have been reported, while other compositions have
been characterized by solution and solid-state NMR spectros-
2
5,26
copy.
The compound Me Al(BHT) has been structurally character-
2
ized through formation of adducts with donors such as CdO
2
7
compounds, amines, or phosphines. The equilibrium between
MeAl(BHT) and Me Al, which furnishes Me Al(BHT), was
•
tion by Ph3C to produce Ph3C-H. We do note in this context
2
3
2
that MeAl(BHT)2 is susceptible to oxidative degradation by O2
studied some time ago by Ittel and co-workers by variable-
•
where the formation of BHT and Al-centered radicals has been
invoked.9
c
(25) (a) Firth, A. V.; Stewart, J. C.; Hoskin, A. J.; Stephan, D. W. J.
Organomet. Chem. 1999, 591, 185-193. (b) Aitken, C. L.; Barron, A. R.
J. Chem. Crystallogr. 1996, 26, 293-295. (c) Petrie, M. A.; Olmstead, M.
M.; Power, P. P. J. Am. Chem. Soc. 1991, 113, 8704-8. (d) Benn, R.;
Janssen, E.; Lehmkuhl, H.; Rufinska, A.; Angermund, K.; Betz, P.; Goddard,
R.; Krueger, C. J. Organomet. Chem. 1991, 411, 37-55. (e) Kumar, R.;
Sierra, M. L.; De Mel, V. S. J.; Oliver, J. P. Organometallics 1990, 9,
Under catalytic conditions using in situ catalyst formation
from Cp2ZrMe2 and [Ph3C][B(C6F5)4] in the presence of a large
excess of MeAl(BHT)2 or MeAl(BHT*)2 this degradation
reaction does not interfere with catalysis. Thus, nearly identical
6
polymerization activities of 7.2 and 7.8 × 10 g PE/mol Zr ×
4
84-9. (f) Dzugan, S. T.; Goedken, V. L. Inorg. Chim. Acta 1988, 154,
h were observed when 2.2 µM [Ph3C][B(C6F5)4] was precon-
tacted with either MeAl(BHT)2 or MeAl(BHT*)2 ([Al] ) 1.0
mM) for 6 min prior to the addition of Cp2ZrMe2 ([Zr] ) 2.0
µM) into toluene saturated with ethylene at 30 psig at 30 °C.
An earlier report that concluded that [Ph3C][B(C6F5)4] and
MeAl(BHT)2 are incompatible must be in error1 since an
identical protocol was employed here and the results indepen-
dently confirmed using the two different scavengers. Further,
169-75.
(26) (a) Benn, R.; Janssen, E.; Lehmkuhl, H.; Rufinska, A. J. Organomet.
Chem. 1987, 333, 155-68. (b) Starowieyski, K. B.; Skowronska-Ptasinska,
M.; Muszynski, J. Organomet. Chem. 1978, 157, 379-87. (c) Pasynkiewicz,
S.; Starowieyski, K. B.; Skowronska-Ptasinska, M. J. Organomet. Chem.
1
973, 52, 269-74.
0e
(27) (a) Akakura, M.; Yamamoto, H.; Bott, S. G.; Barron, A. R.
Polyhedron 1997, 16, 4389-4392. (b) Barron, A. R. IUC, Crystallogr. Symp.
1991, 4, 164-70. (c) Healy, M. D.; Ziller, J. W.; Barron, A. R.
Organometallics 1991, 10, 597-608. (d) Healy, M. D.; Leman, J. T.; Barron,
A. R. J. Am. Chem. Soc. 1991, 113, 2776-7. (e) Power, M. B.; Bott, S. G.;
Clark, D. L.; Atwood, J. L.; Barron, A. R. Organometallics 1990, 9, 3086-
97. (f) Healy, M. D.; Ziller, J. W.; Barron, A. R. J. Am. Chem. Soc. 1990,
112, 2949-54.
(
23) Toteva, M. M.; Moran, M.; Amyes, T. L.; Richard, J. P. J. Am.
Chem. Soc. 2003, 125, 8814-8819.
24) Tashiro, M. Synthesis 1979, 12, 921-36.
(

Sterically Hindered Aluminum Alkyls
Organometallics, Vol. 25, No. 21, 2006 5087
1
Figure 3. Variable-temperature H NMR spectra (400 MHz, toluene-d
8
3 2 2 o
) of a 1:1 mixture of AlMe and MeAl(BHT) with [MeAl(BHT) ]
)
0.500 M. Signals due to MeAl(BHT)
2
and Me
2
AlBHT are labeled, and those due to Al
2
Me are highlighted in gray.
6
1
temperature H NMR spectroscopy; the low-temperature spectra
had to be simulated with reference to assumed equilibria, as
the slow exchange limit was not accessible.9
of BHT and Me groups given the broadening of all signals in
the H NMR spectrum.
1
b
We have reinvestigated this process using variable-temper-
1
13
ature H and C NMR spectroscopy. As shown in Figure 3,
1
the 400 MHz H NMR spectrum of a 1:1 mixture of AlMe3
and MeAl(BHT)2 exhibits a set of line-broadened signals
characteristic of Me2Al(BHT) and MeAl(BHT)2 at 25 °C. On
t
cooling, the signals due to Ar-H, Me, and Bu protons sharpen,
while the AlMe signal due to Me2Al(BHT) broadens, and at
-
13 °C, a set of separate and sharp signals due to MeAl(BHT)2
are seen. The ratio of Me2Al(BHT) to MeAl(BHT)2 is 2.13:1
Note that the adduct depicted in eq 3a is similar to that
invoked by Ittel and co-workers to explain their temperature-
dependent NMR spectra. However, there is no evidence in these
spectra for the independent existence of this species.
at -13 °C.
On further cooling, all of the signals due to Me2Al(BHT)
line-broaden, while those due to MeAl(BHT)2 remain sharp
1
3
The variable-temperature C NMR spectra were more
revealing in terms of the nature of the compounds present.
Depicted in Figure 4 are the low (a) and high (b) field regions
(aside from viscosity-induced line-broadening at low temperature
in this concentrated solution). The Al-Me signal of Me2Al(BHT)
shows decoalescence behavior and at, for example, -57 °C;
separate signals due to the terminal and bridging Me groups of
Al2Me6 are present (highlighted in gray). In contrast, the Al-
Me signal due to Me2Al(BHT) (as well as the other signals
characteristic of this compound) remains broadened even at the
lowest temperature investigated. The Al-Me region of the
spectrum at -63 °C can be integrated, and from this the relative
amounts of the three compounds present can be determined to
be Me2Al(BHT):MeAl(BHT)2:Al2Me6 ) 6.05:2.00:1, while Keq
for the reaction depicted in eq 2 is 33.3 M at this temperature.
1
3
of the C NMR spectra of this mixture at various temperature.
The Al-Me region (Figure 4b) mirrors the results just discussed
1
for the H NMR spectra over this same temperature range, and
assignments are as indicated. At the lowest temperature
investigated, it is evident that there is residual exchange between
an intense Al-Me signal at ca. -3 ppm and a very weak signal
at higher field (ca. -7.5 ppm), while the signals due to Al2Me6
and MeAl(BHT)2 are quite sharp at this temperature.
In the low-field region at room temperature, two line-
broadened signals are seen at δ 153.3 and 152.6 due to the ipso
O-CAr carbons of Me2Al(BHT) and MeAl(BHT)2, respectively.
The signal at higher field due to MeAl(BHT)2 sharpens at low
temperature, while the other broadens and shifts to higher field.
Concomitantly, a signal at higher field (138.5 ppm due to an
Al Me + 2 MeAl(BHT) h 4 “Me AlBHT” with K )
2
6
2
2
eq
4
2
[
Me AlBHT] /[MeAlBHT ] [Al Me ] (2)
2 2 2 6
t
ipso Bu-CAr carbon) also line-broadens and shifts to lower field.
Although not readily evident, due to overlap with the solvent
resonances, a signal at δ 126.7 also shows similar behavior.
All of these signals undergo decoalescence, and at low tem-
perature, three signals are resolved at δ 147.4, 140.6, and 132.4
in a ratio of 1:2:1 in these inverse gate-decoupled spectra. In
addition a very weak signal is evident at δ 153.6; we suspect
there are additional signals present, but they are difficult to
reliably detect because of overlap and low sensitivity.
It is evident from these spectra that there are at least three
exchange processes occurring in this system, in addition to the
known exchange between the terminal and bridging groups of
28
Al2Me6. At lowest temperature, Me2Al(BHT) is still fluxional
1
on the H NMR time scale, while at higher temperature this
compound is involved in degenerate Me exchange with Al2-
Me6 (eq 3a) and at still higher temperature, with MeAl(BHT)2
(eq 3b). The latter process may involve degenerate exchange
Our interpretation of these results is the lowest temperature
exchange process involves monomeric and dinuclear forms of
Me2Al(BHT), where the dinuclear compound is exclusively
(
28) Cerny, Z.; Fusek, J.; Kriz, O.; Hermanek, S.; Sole, M.; Casensky,
B. J. Organomet. Chem. 1990, 386, 157-165.

5
088 Organometallics, Vol. 25, No. 21, 2006
Stapleton et al.
Figure 4. Partial VT 13C NMR spectra (100 MHz, toluene-d
high-field region. Signals due to dinuclear and mononuclear Me
) of a 1:1 mixture of Al
Me
and MeAl(BHT)
: (a) low-field region; (b)
8
2
6
2
2
Al(BHT) are highlighted in light and dark gray, respectively.
phenoxide bridged (eq 4). The very weak signal at lowest field
is due to the ipso O-CAr carbons of monomeric Me2Al(BHT)
by analogy with the signal that appears for MeAl(BHT)2 in this
on addition of 0.5 equiv of Al2Me6, while the Zr IP is more
stable. In the former case, the principle degradation products
2
9
are a mixture of (C6F5)nAlMe3-n (and the corresponding IP
3
0
same region. The remaining three signals are in a ratio of 1:2:
formed from these alanes) and (C6F5)2BMe. These same
compounds are also present in smaller quantities in the case of
the Zr analogue, and there is also evidence in both spectra of
t
1
, and they can be assigned to the ipso O-CAr, ipso Bu-CAr,
and ipso Me-CAr, respectively, of the dinuclear complex. The
upfield chemical shift of the ipso O-CAr is in agreement with
other spectroscopic studies of hindered, phenoxide-bridged
complexes of this type.2
3
0
C6F5 transfer to the metal (o-F at δ 114-115). Thus, the
hafnocene IP, which is more prone to B(C6F5)3 dissociation, is
degraded by the Al2Me6 that is present according to the
following general scheme:
5,26
Taken together, the H and 13C NMR spectra indicate that
Me2Al(BHT) is an equilibrating mixture of four different
compounds: monomeric Me2Al(BHT), the symmetrical di-
nuclear complex Me2Al(µ-BHT)2AlMe2, while Al2Me6 and
MeAl(BHT)2 comprise the reactant side of the equilibria
depicted in eq 4, where the relative amounts of all species are
depicted at -75 °C. Evidently, the position of these equilibria
1
Note that such a mixture of compounds can still be competent
for olefin polymerization, but of course the identity of the active
catalyst is now clouded by uncertainty with respect to the nature
of the counteranion(s).
The behavior of the hafnocene IP in the presence of various
quantities of MeAl(BHT)2 and Al2Me6 is depicted in Figure
5
b. It can be appreciated from these spectra that IP degradation
1
are temperature dependent, and from the room-temperature H
is much less problematic when MeAl(BHT)2 is present com-
pared to the situation where it is absent. However, at higher
ratios of Al2Me6:Hf a new IP with a less coordinating MeB-
1
3
and C NMR spectra it seems probable that the mononuclear
form of Me2Al(BHT) is the dominant species present. This
explains, in part, why adducts of this compound can be isolated
by treatment of such mixtures with donors.27
(
C6F5)3 anion (∆δ ) 2.6 ppm) is generated at the expense of
the original ion-pairs. Although not readily evident from Figure
b, a second set of signals with the same ∆δ value is present
Mixtures of MeAl(BHT)2, Al2Me6, and Metallocenium
Ion-Pairs. The experiments just discussed illustrate that sig-
nificant quantities of Al2Me6 are present in mixtures of MeAl-
5
in these spectra in variable amounts with respect to the major
set of signals, depending on [Hf]. We attribute this second set
of signals to formation of variable amounts of a liquid clathrate
phase that forms at these relatively high concentrations (0.01-
(BHT)2 and Al2Me6 at least at low temperature. To further
investigate this in the context of polymerization catalysis, the
behavior of in situ generated ion-pairs, derived from metallocene
dimethyl complexes and B(C6F5)3, in the presence of mixtures
of these two compounds was investigated.
The complexes studied were the ansa-metallocene complexes
Me2Si(Cp)IndMMe2 (M ) Zr, Hf) and the symmetrical Me2-
SiCp2ZrMe2 complex. Typically, a stock solution of MeAl-
3
1
0
.02 M) when the counteranion is rendered less coordinating.
19
Although the F NMR spectra of these mixtures were readily
1
interpreted, the corresponding H NMR spectra were complex
due to the fluxional and unsymmetrical nature of the various
compounds present. For this reason, we studied analogous
(BHT)2 and the relevant ion-pair (IP) was prepared in toluene-
(29) (a) Chen, E. Y.-X.; Kruper, W. J.; Roof, G.; Wilson, D. R. J. Am.
d8, and this stock solution was used to prepare solutions differing
in the amount of Al2Me6. Representative 19F NMR spectra for
the [Me2Si(Cp)IndMMe][MeB(C6F5)3] ion-pairs in the absence
and presence of Al2Me6 are depicted in Figure 5a). It can be
readily seen that the Hf-based IP suffers extensive degradation
Chem. Soc. 2001, 123, 745-746. (b) Bochmann, M.; Sarsfield, M. J.
Organometallics 1998, 17, 5908-5912.
(
30) It is interesting to note that the two isomeric ion-pairs present in
10b
the case of M ) Zr react at different rates with Al2Me6.
31) Beck, S.; Geyer, A.; Brintzinger, H.-H. Chem. Commun. 1999,
2477-2478.
(

Sterically Hindered Aluminum Alkyls
Organometallics, Vol. 25, No. 21, 2006 5089
Figure 5. 19F NMR spectra (276 MHz, toluene-d
) of (a) [Me
. Signals due to (C
BMe are highlighted in light and dark gray, respectively, while the o- p-, and m-F signals due to the isomeric ion-pairs are labeled. (b)
Me SiCp(Ind)HfMe][MeB(C
Me . Signals due to [Me SiCp(Ind)Hf(µ-Me)
SiCp(Ind)MMe][MeB(C
] (IP: M ) Zr, Hf) in the absence (bottom
8
2
6 5 3
F )
spectrum) and presence of 0.5 equiv of Al
2
Me
6
6
5
F )
n
AlMe3-n [or their adducts with Me
2
SiCp(Ind)MMe ] and (C
2
6 5 2
F ) -
[
2
F
6 5
)
3
] ([Hf] ) 0.0125 M) in the presence of various amounts of MeAl(BHT)
2
2
([Al] ) 0.03125 M) and Al -
6
2
2
AlMe
2
6 5 3
][MeB(C F ) ] adducts are highlighted in light gray.
Figure 6. 19F (282 MHz, toluene-d
BHT) , Al Me , and [Me SiCp ZrMe][MeB(C
ZrMe][MeB(C ] (IP) and Al Me . Signals due to the [Me
signal due to TFX are indicated with an asterisk.
, -30 °C) and H NMR spectra (300 MHz, toluene-d
1
, -30 °C) of (a) a 5:2.5:2 mixture of MeAl-
SiCp
] adduct are highlighted in gray, while
8
8
(
2
2
6
2
2
6
F
5
)
3
] (IP) containing TFX as an internal standard and (b) a ca. 4:1 mixture of [Me
SiCp Zr(µ-Me) AlMe ][MeB(C
2
2
-
6
F
5
)
3
2
6
2
2
2
2
6 5 3
F )
reactions of the more symmetrical IP [Me2SiCp2ZrMe][MeB-
remaining signals are due to the unsymmetrical contact IP [Me2-
SiCp2ZrMe][MeB(C6F5)3].
1
19
32
(
C6F5)3], and the H and F NMR spectra of mixtures of this
compound with Al2Me6 and MeAl(BHT)2 are depicted in Figure
. The behavior seen in the 19F NMR spectra was similar to
We thus conclude that the principle ion-pair formed in the
presence of MeAl(BHT)2 and Al2Me6 is [Me2SiCp2Zr(µ-
Me)2AlMe2][MeB(C6F5)3] and that neither MeAl(BHT)2 nor
Me2Al(BHT) coordinates to the metal center, or at least to the
same extent as Al2Me6. The signals are line-broadened presum-
ably due to exchange of AlMe3, coordinated to Zr, with excess
Al2Me6 present in this solution.
6
that for the unsymmetrical ion-pairs discussed above. In
particular, at a MeAl(BHT)2:Al2Me6:Zr ratio of 5:2.5:2 forma-
tion of an IP with a less coordinating counteranion (∆δ ) 2.6)
was essentially complete.
1
The room-temperature H NMR spectrum of a mixture of
MeAl(BHT)2/Al2Me6/Zr (5:2.5:2) was not informative due to
exchange broadening, but on cooling, new signals highlighted
in gray in Figure 6a were observed. Since there are only two
Cp signals and a single SiMe2 resonance, it is evident that this
species is symmetrical, consistent with the formation of [Me2-
SiCp2Zr(µ-Me)2AlMe2][MeB(C6F5)3]. The signals present are
identical (aside from their line-width) to those observed for this
ion-pair in the presence of 0.25 equiv of Al2Me6 at low
temperature (Figure 6b). In this spectrum, separate signals for
the bridging and terminal AlMe signals are resolved, while the
The results presented in this section can be interpreted in the
following manner. In essence, MeAl(BHT)2 and metallocenium
ion-pairs compete for Al2Me6 through formation of Me2Al(BHT)
and hetero-dinuclear Al,M adducts, respectively. At sufficiently
high MeAl(BHT)2:Al2Me6 ratios, essentially all of the Al2Me6
is sequestered in the form of Me2Al(BHT), where the latter
(32) (a) Beck, S.; Prosenc, M.-H.; Brintzinger, H.-H.; Goretzki, R.;
Herfert, N.; Fink, G. J. Mol. Catal. A: Chem. 1996, 111, 67-79. (b) Beck,
S.; Lieber, S.; Schaper, F.; Geyer, A.; Brintzinger, H.-H. J. Am. Chem.
Soc. 2001, 123, 1483-1489.

5
090 Organometallics, Vol. 25, No. 21, 2006
Stapleton et al.
Figure 7. (a) Ethylene consumption vs time for [Cp
solution, 30 °C, 30 psig, [Zr] ) 2.0 µM, [Al Me ] ) 20 µM. (b) Simulated consumption of MeAl(BHT)
Al] ) 600 µM based on kinetic data for hydrolysis of MeAl(BHT)
obtained in THF.14
2
ZrMe][MeB(C
6
F
5
)
3
] for mixtures of MeAl(BHT)
2
and Al
2
Me
6
. Conditions: Toluene
2
6
2 2 o
assuming [H O] ) 60 µM and
[
o
2
compound, or at least its monomeric form, does not strongly
bind to metallocenium ions.
Under these conditions, scavenging of water is effective (and
timely), and as we show here, the effects of any excess Al2Me6
are largely mitigated through the addition of a sufficient quantity
of MeAl(BHT)2 prior to catalyst/cocatalyst introduction.
Thus, from a practical perspective, the deleterious effects of
Al2Me6 on metallocenium ion-catalyzed olefin polymerization
should be largely alleviated through the addition of a sufficient
excess of MeAl(BHT)2. This concept is only partially realized
during actual catalysis of ethylene polymerization by [Cp2ZrMe]-
Conclusions. The use of sterically hindered aluminum alkyls
confers an advantage in the fundamental study of olefin
polymerization using metallocene or other early metal catalysts
and single-component activators. They represent what one might
term weakly interacting scavengers in the sense that they do
not strongly inhibit polymerization through formation of hetero-
dinuclear adducts, as seen with conventional alkylaluminum
compounds, nor do they readily degrade many activators of
interest. Further, as we show here, the deleterious effects of
simple alkylaluminum compounds can be largely eliminated
through the addition of a sufficient quantity of one of these
sterically hindered scavengers; this is important for catalysts
that are susceptible to protonolysis by BHT-H, the principle
2
[MeB(C6F5)3], where initial monomer uptake versus t curves
at various ratios of MeAl(BHT)2:AlMe3 are shown in Figure 7.
Even at a 50:1 ratio, the activity of this catalyst is still not
equivalent to that observed in the absence of AlMe3. However,
we suspect that any inhibition of catalysis under these conditions
does not result from free AlMe3 but perhaps from weak and
reversible interaction of “Me2Al(BHT)” with the metal center
during catalysis.
A final comment concerning practical aspects of polymeri-
zation experiments using these scavengers deserves mention.
This relates to effective and timely removal of protic impurities
in autoclave reactors. MeAl(BHT)2 is only sluggishly reactive
toward water, and from the rates of hydrolysis measured in THF
one can estimate that 4 h of scavenging time would be required
to reduce water to sub-micromolar concentrations (see Figure
product of MeAl(BHT) hydrolysis.
Experimental Section
All materials were obtained from Aldrich Chemical Co. or Strem
Chemical Co. and purified as required. All synthetic procedures
14
7
b). Although we have been unable to measure the rate of
hydrolysis in hydrocarbon solution, it is significantly faster than
in THF, though the product distribution is the same; hydrolysis
is accompanied by formation of ca. 100 µM BHT-H if, for
example. [H2O]o ) 60 µM (Scheme 1).
It is interesting to note that the reaction of BHT-H with a
mixture of Al2Me6 and excess MeAl(BHT)2 is much slower than
independent reaction of BHT-H with Al2Me6. Even when the
latter compound is added to about an equimolar amount of
were conducted under a N atmosphere using Schlenk techniques
or in a glovebox. Isolation of solids from supernatant liquids was
2
performed using a filter-tipped cannula or by filtration. Tetrahy-
drofuran, diethyl ether, toluene, hexane, and dichloromethane were
purified by passage through activated La Roche A-2 alumina and
3
3
Engelhard CU-0226s Q-5 columns.
Routine H, 19F, and 13C NMR spectra were recorded on a Varian
1
Mercury or Gemini 300 MHz instrument. Benzene-d
and bromobenzene-d were distilled from Na or Na/K alloy prior
to use. H NMR spectra were referenced with respect to residual
6 8
, toluene-d ,
BHT-H ([BHT-H] ) 0.0091 M) and excess MeAl(BHT)2 ([Al]
5
1
1
)
0.0682 M) at 25 °C, the initial H NMR spectra reveal the
13
protonated solvent, while C NMR spectra were referenced with
presence of BHT-H, MeAl(BHT)2, and Me2Al(BHT) ([Al] )
.0078 M) with very little CH4 production (see the Supporting
respect to deuterated solvent. 19F NMR spectra were referenced
0
with respect to tetrafluoro-p-xylene (TFX: δ -145.69 vs CFCl
in toluene-d ). Variable-temperature experiments were performed
3
Information).
8
A plot of ln{[BHT-H][Me2Al(BHT)]o/[BHT-H]o[Me2Al-
using Varian Innova 400 MHz or VXR 300 MHz instruments. The
spectrometer thermocouple was calibrated to within 5% of actual
temperature using a sample of MeOH. IR spectra were obtained
on a DigiLab Excalibur FTS 3000 spectrometer and were not
calibrated. Elemental analyses were performed by either Oneida
Research Services or Galbraith Laboratories.
Tris(perfluorophenyl)borane, donated by Nova Chemicals, was
predried in a hexane solution containing 4 Å activated molecular
sieves and recrystallized from hexane at -30 °C. 2,4,6-Tri-tert-
butylphenol was purchased from Aldrich Chemical Co. and was
purified by recrystallization from hexane at 0 °C. The compounds
(
1
BHT)]} vs t was linear over about 2 half-lives with k ) 5.7 ×
-
2
-1 -1
0
M
s . If we assume that [BHT-H]o ) [AlMe3]o ) 100
µM with excess MeAl(BHT)2, as in the simulation depicted in
Figure 7b, one would predict that decline of BHT-H to 1.0 µM
concentration would require in excess of 4800 h for completion!
This is far too slow for practical purposes and dictates that
AlMe3 be added first to an autoclave reactor to scavenge water.
From a practical perspective we use [AlMe3] ) [H2O] even
though this represents a 1.5-fold excess of Al-Me over that
required for complete reaction. After a much shorter period of
time (typically 1 h or less), a 10- to 100-fold excess of MeAl-
(BHT)2 is then added, followed by activator and metallocene
(33) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
catalyst.
Timmers, F. J. Organometallics 1996, 15, 1518-1520.

Sterically Hindered Aluminum Alkyls
],34 Cp ,35 Cp
Organometallics, Vol. 25, No. 21, 2006 5091
mm septum-capped NMR tube, 150 µL of the MeAl(BHT) (5.0
equiv, 65 µmol) and 500 µL (13 µmol, 26 mM) of the ion-pair
solutions were added. To this solution was then syringed 150 µL
13
,36 Me
[
(C
6
H
5
)
3
C][B(C
6
F
5
)
4
2
ZrMe
2
2
Zr Me
2
2
10b
SiCp
were
2
-
2
19
10b
2
ZrMe ,
Me
2
SiCp(Ind)ZrMe and Me SiCp(Ind)HfMe
2
,
2
2
prepared according to literature procedures.
t
Methyl Bis(2,6-di- Bu-4-methylphenoxy)aluminum. MeAl-
BHT) was synthesized using a modification of literature proce-
3
of the AlMe solution (5.0 equiv, 65 µmol) at 25 °C. The NMR
1
(
2
9
tube was then lowered into a precooled probe at -80 °C. H and
19
dures. Trimethyl aluminum (3.45 g, 47.9 mmol) was dissolved in
0 mL of n-hexane. This solution was slowly added over a period
F NMR spectra were acquired at 10 °C increments during slow
2
warming of the probe to 30 °C.
of 5 min via cannula to a rapidly stirred solution of Ar′OH (20.04
g, 90.9 mmol) in 120 mL of n-hexane. Caution: Addition must be
done slowly to aVoid uncontrolled methane formation. The solution
was stirred for 1 h at 25 °C, during which time a white precipitate
separated. The mixture was concentrated to dryness in vacuo at
Reaction of [Me
2
SiCp(Ind)MMe][MeB(C
and MeAl(BHT) . These reactions were
performed in a manner similar to the above experiments, with the
6 5 3
F ) ] (M ) Zr or
10b
Hf)
with AlMe
3
2
10b
[
Me
amounts of AlMe
added at 25 °C to a mixture of the ion-pair and MeAl(BHT)
the following amounts: [Zr] or [Hf] ) [AlMe ] ) 0.025 M with
MeAl(BHT) ] ) 0.0, 0.025, and 0.25 M. The spectra are depicted
in Figure 4b.
Spectroscopic Studies of [Cp
BHT) . (a) NOESY Spectroscopy. A solution of MeAl(BHT)
121 mg, 0.25 mmol) and B(C (28 mg, 0.055 mmol) was
. To this solution was added 0.305
in C (20 mg in 0.47 mL; 0.17
2
SiCp(Ind)MMe][MeB(C
F
6 5
)
3
] ion-pair and using various
. In each case the AlMe was
in
3
and MeAl(BHT)
2
3
0.01 mmHg overnight. The crude product (21.3 g, 98% yield) could
2
be used without further purification. Spectroscopic data were
3
consistent with the literature.
[
2
t
Methyl Bis(2,4,6-tri- Bu-phenoxy)aluminum. MeAl(BHT*)
2
was synthesized using a modification to the literature procedure.9d
Analogous to the synthesis of MeAl(BHT) , a solution of 2,4,6-
tri- Bu-phenol (4.94 g, 17.5 mmol) in hexane was rapidly stirred
with an excess of Al(CH (0.95 g, 13 mmol). The mixture was
2 6 5 3
ZrMe][MeB(C F ) ] and MeAl-
2
(
2
2
t
(
6 5 3
F )
3 3
)
prepared in 0.612 mL of C
6 6
D
concentrated to dryness in vacuo at 0.01 mmHg overnight. The
crude product (4.85 g, 98% yield) could be used without further
purification. Spectroscopic data were consistent with the literature.
mL of a solution of Cp ZrMe
2
2
6 6
D
M) so as to give a final solution where [IP] ) 0.056 M. A series
of NOESY spectra were acquired with different mixing times, the
longest of which was τ ) 1.0 s. This spectrum and F slices through
1
Reaction of [(C
septum-sealed 5 mm NMR tube, 39 mg (42 µmol) of [(C
B(C ] was dissolved in 483 µL of bromobenzene-d . Twenty
milligrams (42 µmol) of solid MeAl(BHT) was then added and
6
H
5
)
3 6
C][B(C F
5
)
4
] with MeAl(BHT)
2
. In a
6
5 3
H ) C]-
the relevant diagonal and cross-peaks at δ 0.24 (Zr-Me), 0.08 (B-
Me), and -0.34 (Al-Me) are depicted in the Supporting Information.
[
F
6 5
)
4
5
2
13
(
b) C-Scrambling Experiment. A solution of MeAl(BHT)
60 mg, 0.125 mmol) and B(C (14 mg, 0.0275 mmol) was
prepared in 0.600 mL of C . To this solution was added 0.350
Zr Me in C (20 mg in 1.0 mL; 0.079
M) so as to give a final solution where [IP] ) 0.029 M. H and
2
1
19
dissolved. The reaction solution turned deep red. H and F NMR
spectra were recorded at room temperature.
(
6 5 3
F )
6
3
D
6
Reaction of [(C
septum-sealed 5 mm NMR tube 38.5 mg (41.7 µmol) of [(C
B(C ] was dissolved in 468 µL of bromobenzene-d . Twenty-
four milligrams (46 µmol) of solid MeAl(BHT*) was then added
6
H
5
)
3
C][B(C
6
F
)
5 4
] with MeAl(BHT*)
2
. In a
1
mL of a solution of Cp
2
2
6 6
D
H ) C]-
6 5 3
1
19
F
[
F
6 5
)
4
5
NMR spectra were acquired about 5 min after mixing at room
temperature and showed trace levels of scrambling between Zr-
Me and Al-Me. A H spectrum acquired 10 min later revealed
minimal change. These spectra were integrated to record the
intensity of (a) the low-field 13C satellite of the Bu protons of MeAl-
2
1
and dissolved. The reaction solution slowly turned deep red. H
and 19F NMR spectra were recorded at room temperature.
1
Mixtures of MeAl(BHT)
mm NMR tube, 132 mg (0.28 mmol) of MeAl(BHT)
and dissolved into a solution of 560 µL of toluene-d
0 mg (0.28 mmol) of Al(CH . The equimolar ratio of MeAl-
BHT) and Al(CH was confirmed by acquiring a H NMR
2
and Al(CH
3
)
3
. In a septum-sealed 5
was weighed
containing
t
2
1
3
(
BHT)
2
as an internal standard, (b) the two signals due to Zr- -
8
CH
3
, and (c) the 13C satellite at highest field due to the Al-Me
2
(
3 3
)
protons, as these peaks were of similar intensity and resolved. After
ejecting the sample and equilibrating the probe at 40 °C, the sample
was reinserted and spectra were recorded over a 2 h period,
corresponding to ca. one half-life (see the Supporting Information
for representative spectra and integral data as a function of time).
At this point there was significant decomposition of the IP as judged
1
2
3 3
)
13
1
spectrum at room temperature. Low-temperature C and H NMR
spectra were recorded by cooling the probe to between -30 and
-
80 °C.
Reaction of [Me
SiCp
2
ZrMe][MeB(C
F )
6 5 3
]32 with AlMe
SiCp ZrMe and 179 mg
were each dissolved into 6.25 mL of
containing 50 mM TFX standard. The Me SiCp ZrMe
solution was added to the B(C solution to generate a 26 mM
ion-pair solution, which turned from yellow to red upon mixing.
In a separate vial, 31 mg (430 µmol) of AlMe was weighed and
dissolved into 1.00 mL of toluene-d (containing 50 mM TFX),
3
. Into
2
separate vials, 100 mg (325 µmol) of Me
350 µmol) of B(C
toluene-d
2
2
2
1
by multiple resonances in the Cp region of the H NMR spectrum.
(
6 5 3
F )
Reaction of BHT-H with a Mixture of MeAl(BHT)
Me . A stock solution of MeAl(BHT) (22 mg, 0.045 mmol)
and BHT-H (1.0 mg, 0.0045 mmol) in 0.692 mL of C containing
TFX was prepared. A 1H NMR spectrum of the mixture was
recorded, and then 32 µL of a solution of Al Me (0.0505 M in
) was added via syringe with shaking and the tube immediately
immersed in the probe. An initial spectrum revealed the presence
of BHT-H and Me Al(BHT) but no methane. Spectra were recorded
2
and
8
2
2
2
Al
2
6
2
6 5 3
F )
6 6
D
3
2
6
8
6 6
C D
making a 430 mM solution. Into a 5 mm septum-capped NMR
tube, 500 µL (13 µmol) of the ion-pair solution was added and
cooled to -80 °C. Thirty microliters of the AlMe (1.0 equiv, 13
3
2
over a 3 h period, during which time both of these materials were
consumed with concomitant production of methane. Selected spectra
along with kinetic data are included as Supporting Information.
µmol) solution was then syringed in and gently shaken at -80 °C.
The NMR tube was then lowered into a precooled probe at -80
1
19
°
C. H and F NMR spectra were acquired at 10 °C increments
during slow warming of the probe to 30 °C.
Reaction of [Me SiCp ZrMe][MeB(C
:1 Solution of MeAl(BHT) and AlMe
µmol) of MeAl(BHT) was dissolved into 1.0 mL of toluene-d
containing 50 mM TFX) to make a 440 mM solution. Into a 5
Polymerization of Ethylene at Various Ratios of Al
MeAl(BHT) . A 1.0 L autoclave reactor was heated to 100 °C under
vacuum and evacuated and refilled 3× with dry N at this
temperature. It was allowed to cool to 30 °C under N , and then
2 6
Me :
2
2
6
F
5
)
3
] with 5 equiv of a
2
1
2
3
. In a vial, 210 mg (440
2
2
8
2
(
448-498 mL of dry toluene (impurity level 40 µM as determined
by titration of an aliquot using a stock solution of benzophenone
ketyl in xylenes/tetraglyme) was added using an overpressure of
(
34) (a) Chien, J. C. W.; Tsai, W. M.; Rausch, M. D. J. Am. Chem. Soc.
991, 113, 8570-1. (b) Doellein, G. US Patent 5,399,781, 1995, 3 pp.
35) Wailes, P. C.; Weigold, H.; Bell, A. P. J. Organomet. Chem. 1972,
2, 155-164.
36) Dahmen, K. H.; Hedden, D.; Burwell, R. L., Jr.; Marks, T. J.
Langmuir 1988, 4, 1212-14..
1
2 2 6
dry N . A solution of Al Me in toluene (1.0 mL of 0.01 M) was
(
added via syringe at 30 °C, and the mixture was then saturated
with ethylene at 30 psig for 1-2 h. After saturation was complete
as detected by calibrated mass flow meter (i.e., zero flow), a solution
3
(

5092 Organometallics, Vol. 25, No. 21, 2006
Stapleton et al.
of MeAl(BHT)
amounts corresponding to 0:1, 10:1 ([MeAl(BHT)
or 50:1 ([MeAl(BHT) ] ) 2.0 mM) equivalents based on [AlMe
In a control experiment no Al Me was added to the reactor, but
an equivalent amount of MeAl(BHT) was added so as to give
MeAl(BHT) ] ) 2.0 mM prior to saturation with ethylene.
A stock solution of [Cp ZrMe][MeB(C ] (0.001 M in toluene)
was prepared, and for each reaction the same fixed quantity of
solution (1.0 mL) was added via syringe after a further 1 h of stirring
2
following the addition of MeAl(BHT) . After a short delay, ethylene
2
in toluene (0.01 M in toluene) was added in
] ) 0.4 mM),
].
Acknowledgment. We thank the University of Akron for
financial support of this work. R.A.S. acknowledges the Eastman
Chemical Co. for a scholarship. J.C. thanks the American
Chemical Society, Petroleum Research Fund, for partial stipend
support.
2
2
3
2
6
2
[
2
2
6 5 3
F )
Supporting Information Available: Additional NMR spectra
for mixtures of MeAl(BHT) with various compounds. This
2
information is available free of charge via the Internet at
http://pubs.acs.org.
uptake was detected and the reactions were stirred at 1000 rpm
and 30 °C for several hours until steady state had been achieved.
Representative ethylene uptake curves are depicted in Figure 6a.
OM060474S