On the first Insertion of α-olefins in hafnium pyridyl-amido polymerization catalysts

Article abstract of DOI:10.1021/om900705v

Reactions of [{N-,N,C_naphthy1}HfMe][MeB(C₆F5) ₃] (2) precatalyst with a series of a-olefins have been investigated in order to intercept the active polymerization species generated by an in situ modification of the precurs

Full text of DOI:10.1021/om900705v

Organometallics 2009, 28, 5445–5458 5445
DOI: 10.1021/om900705v
On the First Insertion of r-Olefins in Hafnium Pyridyl-Amido
Polymerization Catalysts
Cristiano Zuccaccia,† Vincenzo Busico,‡ Roberta Cipullo,‡ Giovanni Talarico,‡
Robert D. J. Froese,^ Paul C. Vosejpka,^ Phillip D. Hustad,§ and Alceo Macchioni*,†
†Department of Chemistry, University of Perugia, Via Elce di Sotto, 8-06123, Perugia, Italy, ‡Department of
Chemistry, University of Napoli, Via Cintia, 80126, Napoli, Italy, §The Dow Chemical Company,
2301 North Brazosport Boulevard, B-3827, Freeport, Texas 7754, and ^The Dow Chemical Company,
1776 Building, Midland, Michigan 48674
Received August 10, 2009
Reactions of [{N-,N,Cnaphthyl-}HfMe][MeB(C6F5)3] (2) precatalyst with a series of R-olefins
have been investigated in order to intercept the active polymerization species generated by an in situ
modification of the precursor by insertion of a single monomer unit into the Hf-CAryl bond. In all cases
the first migratory insertion of monomer occurs into the Hf-CAryl bond rather than the Hf-CAlkyl
bond. A low-temperature polymerization with 170 equiv of 1-hexene activated with tris-
(pentafluorophenyl)borane (FAB) allows for the complete NMR characterization of a Hf-CAlkylaryl
methyl cation. This structure agrees with DFT studies of the kinetically favored diastereomer, although a
number of other structures are more stable thermodynamically. Attempts to trap an inserted catalyst
through the stoichiometric addition of 2-vinylpyridine or 3-ethoxy-1-propylene led to complicated
reactions, but in both cases, experimental and computational data suggest that both processes initiate
through insertion of the substrate into the Hf-CAryl bond. In addition, an activated Hf-dibutyl complex
is studied in an attempt to minimize differences in the rates of initiation and propagation, as a butyl
group is a reasonable mimic for a propagating polymer chain. Tritylborate activation of this complex
cleanly generates 1 equiv of 1-butene in close proximity to the Hf-butyl cation via β-hydride abstraction.
This reaction results in formation of isotactic poly(1-butene) with a fairly high molecular weight rather
than butene oligomers. The observed molecular weight is consistent with a small fraction of active
species, and quenching studies show a similar fraction of butene-modified ligand. These results are
consistent with slow insertion into the Hf-CAryl bond followed by fast polymerization kinetics, with the
latter rate constant being 3 orders of magnitude faster than the former.
Introduction
lead to extremely efficient catalysts for olefin polymeri-
zation. Discovered and optimized through high-through-
put parallel screening techniques,2-4 these species have
been successfully applied in high-temperature solution pro-
cesses for the production of highly isotactic, high mole-
cular weight polypropylene.5 More recently, these species
Pyridyl-amide complexes have recently demonstrated
great promise in the field of olefin polymerization catalysis.
Upon activation, these {N-,N,CAryl-}HfMe2 C1-symmetric
arylcyclometalated hafnium-pyridyl-amido precatalysts1
*Corresponding author. E-mail: alceo@unipg.it.
(1) {N-,N,C-} indicates a tridentate ligand with an anionic Hf-amide
(N-), neutral Hf-pyridyl bond (N), and Hf-CAryl/Alkyl bonds (C-).
(2) (a) Boussie, T. R.; Diamond, G. M.; Goh, C.; Hall, K. A.;
LaPointe, A. M.; Leclerc, M. K.; Lund, C.; Murphy, V. (Symyx
Technologies, Inc.) U.S. Pat. Appl. Publ. No. US 2006/0135722 A1.
(b) Boussie, T. R.; Diamond, G. M.; Goh, C.; Hall, K. A.; LaPointe,
A. M.; Leclerc, M. K.; Lund, C.; Murphy, V. (Symyx Technologies, Inc.)
U.S. Pat. 7,018,949, 2006. (c) Boussie, T. R.; Diamond, G. M.; Goh, C.;
LaPointe, A. M.; Leclerc, M. K.; Lund, C.; Murphy, V. (Symyx
Technologies, Inc.) U.S. Pat. 6,750,345, 2004. (d) Boussie, T. R.; Dia-
mond, G. M.; Goh, C.; Hall, K. A.; LaPointe, A. M.; Leclerc, M. K.; Lund, C.;
Murphy, V. (Symyx Technologies, Inc.) U.S. Pat. 6,713,577, 2004.
(e) Boussie, T. R.; Diamond, G. M.; Goh, C.; Hall, K. A.; LaPointe, A. M.;
Leclerc, M. K.; Lund, C.; Murphy, V. (Symyx Technologies, Inc.) U.S. Pat.
6,706,829, 2004. (f) Boussie, T. R.; Diamond, G. M.; Goh, C.; Hall, K. A.;
LaPointe, A. M.; Leclerc, M. K.; Lund, C.; Murphy, V. (Symyx Technologies,
Inc.) PCT Int. Appl. Publ. No. WO 2002/046249. (g) Boussie, T. R.;
Diamond, G. M.; Goh, C.; Hall, K. A.; LaPointe, A. M.; Leclerc, M. K.;
Lund, C.; Murphy, V. (Symyx Technologies, Inc.) PCT Pat. Appl. Publ. No.
WO 2002/038628.
(3) Boussie, T. R.; Diamond, G. M.; Goh, C.; Hall, K. A.; LaPointe,
A. M.; Leclerc, M. K.; Lund, C.; Murphy, V.; Shoemaker, J. A.W.;
Tracht, U.; Turner, H.; Zhang, J.; Uno, T.; Rosen, R. K.; Stevens, J. C.
J. Am. Chem. Soc. 2003, 125, 4306.
(4) Boussie, T. R.; Diamond, G. M.; Goh, C.; Hall, K. A.; LaPointe,
A. M.; Leclerc, M. K.; Murphy, V.; Shoemaker, J. A. W.; Turner, H.;
Rosen, R. K.; Stevens, J. C.; Alfano, F.; Busico, V.; Cipullo, R.;
Talarico, G. Angew. Chem., Int. Ed. 2006, 45, 3278.
(5) (a) Arriola, D. J.; Bokota, M.; Timmers, F. J. (Dow Chemical
Company) PCT Int. Appl. Publ. No. WO 2004/026925 A1. (b) Frazier,
K. A.; Boone, H.; Vosejpka, P. C.; Stevens, J. C. (Dow Chemical
Company) U.S. Pat. Appl. Publ. No. US 2004/0220050 A1. (c) Tau,
L.-M.; Cheung, Y. W.; Diehl, C. F.; Hazlitt, L. G. (Dow Chemical
Company) U.S. Pat. Appl. Publ. No. US 2004/0087751 A1. (d) Tau,
L.-M.; Cheung, Y. W.; Diehl, C. F.; Hazlitt, L. G. (Dow Chemical
Company) U.S. Pat. Appl. Publ. No. US 2004/0242784 A1. (e) Coalter,
J. N., III; Van Egmond, J. W.; Fouts, L. J., Jr.; Painter, R. B.; Vosejpka,
P. C. (Dow Chemical Company) PCT Int. Appl. Publ. No. WO 2003/
040195 A1. (f) Stevens, J. C.; Vanderlende, D. D. (Dow Chemical
Company) PCT Int. Appl. Publ. No. WO 2003/040201 A1.
r
2009 American Chemical Society
Published on Web 09/01/2009
pubs.acs.org/Organometallics

5446 Organometallics, Vol. 28, No. 18, 2009
Zuccaccia et al.
Scheme 1. Activation with Bronsted Acids
e
the Hf-CAryl bond are starting points for entering the
catalytic cycle. Although these activation studies dis-
close several important aspects of the peculiar chemistry of
{N-,N,CAryl-}HfMe2 compounds, they do not afford any
rationalization of their multisite behavior in catalysis. One
proposal for the generation of multiple sites again turns to
the potential reactivity of the Hf-CAryl bond. Analogous to
protonation during activation with Bronsted acids, the
e
incoming olefin can choose to react with either the Hf-CAlkyl
or Hf-CAryl bonds (Scheme 2).
The former case represents the “classical” migratory in-
sertion reaction, leading to formation of species AI, which
can continue to insert olefins (Scheme 2, pathway A). On the
contrary, a first insertion into the Hf-CAryl bond (Scheme 2,
have also found use in the production of a new class of
olefin block copolymers through chain shuttling polymeri-
zation.6
-
pathway B) causes a modification of the {N-,N,CAryl
}
ligand through formation of BI, which can then successively
undergo “classical” multiple migratory insertion reactions
into the Hf-CAlkyl bond. This possibility was first proposed
by Froese et al. to account for multisite behavior in ethylene/
octene copolymerizations.9 Through theoretical calcula-
tions, they showed that (i) insertion into the Hf-CAryl bond
has a lower barrier than the corresponding insertion into the
Hf-CAlkyl bond and (ii) the monoinserted species BI displays
several characteristics consistent with polymerization data,
including a low barrier to site epimerization.9 This olefin
insertion could result in many different BI diastereomers
depending on the orientation of the naphthyl moiety and
stereochemistry of the insertion in the case of R-olefins.
Consequently, BI is a good candidate to be one of the active
species, but the experimental data do not preclude 2 from
also being another.
One distinguishing structural feature of these species over
most of the other precatalysts for olefin polymerization7,8
is the presence of both Hf-CAlkyl and Hf-CAryl bonds. On
one hand, the Hf-CAryl bond plays a key role in determin-
ing the outstanding performance of these species,3 but it
also adds several layers of complexity on both the activa-
tion and polymerization mechanisms. Particularly, it
offers alternatives for both protic attack, in the case of
activation with Bronsted acids, and olefin insertion. An in-
e
depth analysis of the molecular weight distributions of
polymers generated with the above-mentioned cata-
lysts9 indicates that polymerizations are carried out by
several active sites. It was proposed that these multiple
catalytic sites may be derived from this peculiar structural
feature.
In addition to computations, some experimental evidence
has been found to support pathway B. First, a small fraction
of 4-methyl-1-pentene-appended ligand, arising from olefin
insertion into the Hf-CAryl bond, has been isolated and
characterized from a 4-methyl-1-pentene polymerization in
the presence of 2.9 Second, when 13CH2d13CH2 was polym-
erized, the 13C NMR spectra showed resonances at 69 and
35 ppm that are consistent with ethylene insertion into the
Hf-CAryl bond.9 Related Cs-symmetric achiral precursors
produce poly(R-olefins) with prevalently isotactic micro-
structures and complex multimodal molecular weight
distributions.12-14 Coates and co-workers have also re-
cently synthesized a cationic species having a single chiral
sp3-hybridized carbon bridging the naphthyl moiety and
hafnium that produces isotactic polypropylene in a living
fashion.15
Although these observations are sufficient to establish the
viability of pathway B, they do not necessarily demonstrate
that the first olefin insertion preferentially occurs into the
Hf-CAryl bond nor that the resulting isomeric ion pairs (BI-
type) are the main active species. In this article, we describe
our experimental attempts to intercept and characterize
monoinserted species and possibly evaluate the relative
activity of species 2, AI, and BI. Careful studies of the
With the aim of intercepting and identifying the cata-
lytically active sites, we initiated a systematic study of the
activation of {N-,N,CAryl-}HfMe2 precatalysts and their
subsequent reactions with olefins. The results of the acti-
vation studies were reported in a recent article.10 The activa-
tion of these precatalysts with Lewis acids proceeds by
methyl abstraction,10 in line with other metallocene and
post-metallocene systems.11 On the contrary, Bronsted
e
acids, such as [HNR3][B(C6F5)4], selectively protonate the
metalated aryl arm of the tridentate ligand, forming di-
methyl intermediate I stabilized by an Hf-η2-naphthyl
interaction (Scheme 1).10 This dimethyl species slowly under-
goes naphthyl remetalation assisted by a nucleophile L
(Scheme 1).10
Activation and 1-alkene polymerization studies using
different precatalyst/cocatalyst combinations demonstrate
that intermediate I is likely more distant from the cata-
lytically active species than the remetalated ion pair.10 These
data suggest that the monomethyl ion pairs bearing
(6) Arriola, D. J.; Carnahan, E. M.; Hustad, P. D.; Kuhlman, R. L.;
Wenzel, T. T. Science 2006, 312, 714.
(7) For a case in which olefin inserts in a Ti-CAryl bond see: Gielens,
E. E. C. G.; Dijkstra, T. W.; Berno, P.; Meetsma, A.; Hessen, B.; Teuben,
J. H. J. Organomet. Chem. 1999, 591, 88.
(8) For another precatalyst for olefin polymerization having both
M-CAlkyl and M-CAryl bonds (M = Ti, Zr, and Hf) see: Tam, K.-H.;
Lo, J. C. Y.; Guo, Z.; Chan, M. C. W. J. Organomet. Chem. 2007, 692,
4750.
(12) Domski, G. J.; Lobkovsky, E. B.; Coates, G. W. Macromolecules
2007, 40, 3510.
(13) Niu, A.; Stellbrink, J.; Allgaier, J.; Richter, D.; Hartmann, R.;
Domski, G. J.; Coates, G. W.; Fetters, L. J. Macromolecules 2009, 42,
1083.
(9) Froese, R. D. J.; Hustad, P. D.; Kuhlman, R. L.; Wenzel, T. T. J.
Am. Chem. Soc. 2007, 129, 7831.
(14) Busico, V.; Cipullo, R.; Pellecchia, R.; Rongo, L.; Talarico, G.;
Macchioni, A.; Zuccaccia, C.; Froese, R. D. J.; Hustad, P. D. Macro-
molecules 2009, 42, 4369.
(15) Domski, G. J.; Edson, J. B.; Keresztes, I.; Lobkovsky, E. B.;
Coates, G. W. Chem. Commun. 2008, 6137.
(10) Zuccaccia, C.; Macchioni, A.; Busico, V.; Cipullo, R.; Talarico,
G.; Alfano, F.; Boone, H. W.; Frazier, K. H.; Hustad, P. D.; Stevens, J.
C.; Vosejpka, P. C.; Abboud, K. A. J. Am. Chem. Soc. 2008, 130, 10354.
(11) Chen, E. Y.-X.; Marks, T. J. Chem. Rev. 2000, 100, 1391.

Article
Organometallics, Vol. 28, No. 18, 2009 5447
Scheme 2. First Migratory Insertion of Olefin in the Hf-CAlkyl (pathway A) or Hf-CAryl (pathway B) Bonds (P = polymeryl chain)
Scheme 3. Activation of 1 with B(C6F5)3 in the Presence of an
Excess of 1-Hexenea
activation of complex 1 with B(C6F5)3 in the presence of
1-hexene are described. Attempts to characterize a stoichio-
metric Hf-CAryl insertion by using chelating olefins,
2-vinylpyridine or 3-ethoxy-1-propylene,16 are also detailed.
In addition, 1-butene is generated in close proximity to
each metal center by activation of the {N-,N,CAryl-}-
Hf(n-Bu)2 precursor with [CPh3][B(C6F5)4], which ab-
stracts a β-H17 from one n-Bu group. Where possible, the
assistance of molecular modeling is used to predict and
explain results.
a The stereoisomer with R-configuration at C* is shown, although a
racemic mixture was used.
1
complete assignment of H and 13C NMR resonances of 3
Results and Discussion
is carried out starting from the singlet at 6.16 ppm in the
1H NMR spectrum, which in this family of compounds
corresponds to H17 (Figure 1A),10 following the scalar and
dipolar connectivity in the 1H-COSY, 1H,13C-HMQC,
In Situ Generation and Characterization of the Aryl Mono-
inserted Species. Several activation reactions of the precata-
lyst 1 with B(C6F5)3 in the presence of a small excess of
1-hexene (3-6 equiv) were carried out under different experi-
mental conditions (different temperature, solvent, order of
addition of reagents, etc.) within an NMR tube. In all cases,
resonances typical of poly(1-hexene) appear in the 1H NMR
spectrum without any appreciable variation of the reso-
nances due to the ion pair 2, which forms as a consequence
of methyl abstraction by B(C6F5)3 (Supporting Infor-
mation). These data suggest that the rate of initiation is
much lower than propagation, and consequently, a small
fraction of active sites are responsible for the observed
polymerization.18,19
1
1H,13C-HMBC, and H-NOESY experiments, respectively.
Considering the importance of characterization of mono-
inserted species 3, some details of the NMR assignments are
highlighted here, especially those protons essential for eluci-
dating the three-dimensional structure of 3.
The first key point is the assignment of the three broad
septets corresponding to H24, H31, and H34 due to the CH
moieties of the three isopropyl groups. H31 is distinguished
from H24 and H34 by the lack of an NOE interaction with H17
(Figure 1B). The discrimination between H24 and H34 is
possible considering that the former is scalarly coupled with
H24a (Figure 1D), which is easily assigned due to its NOE
interaction with H14 (Figure 1B). A second key point concerns
discrimination between H2 and H3 aromatic resonances, which
is complicated by the partial overlap with the doublet of H5. H2
is identified by a weak long-range heteronuclear correlation
with a carbon resonance at 48 ppm consequently assigned to Cb
(Figure 1C). The resonances corresponding to Cb and Ca (δC =
78 ppm) are in reasonable agreement with those observed
previously in experiments using 13CH2d13CH2.9 Finally, reso-
nances corresponding to the Hf-Me group (C38) are found at
On the contrary, when precatalyst 1 is activated with
B(C6F5)3 in the presence of a large excess of 1-hexene
(170 equiv) at low temperature (-56 ꢀC), a new set of
1
resonances is observed in the H NMR spectrum (Figures
4S-7S, Supporting Information) in addition to those
of poly(1-hexene) (polymerization was complete in about
50 min). A detailed multinuclear and multidimensional
NMR study indicates the new set of resonances is due to
the aryl monoinserted species 3 shown in Scheme 3. A
δ
H = 0.06 ppm and δC = 60.6 ppm.
(16) Bijpost, E. A.; Zuideveld, M. A.; Meetsma, A.; Teuben, J. H. J.
Organomet. Chem. 1998, 551, 159. For a more recent paper describing the
use of 3-(methylthio)-1-propylene for similar studies see: Richter, B.;
Meetsma, A.; Hessen, B.; Teuben, J. H. Angew. Chem., Int. Ed. 2002, 41,
2166.
(17) (a) Lian, B.; Toupet, L.; Carpentier, J.-F. Chem.;Eur. J. 2004,
10, 4310. (b) Mehrkhodavandi, P.; Schrock, R. R. J. Am. Chem. Soc. 2001,
123, 10746. (c) Bochmann, M.; Sarsfield, M. J. Organometallics 1998, 17,
5908. (d) Jerkunica, J. M.; Traylor, T. G. J. Am. Chem. Soc. 1971, 93, 6278.
(e) Hannon, J. S.; Traylor, T. G. J. Org. Chem. 1981, 46, 3645.
(18) (a) Landis, C. R.; Rosaaen, K. A.; Sillars, D. R. J. Am. Chem.
Soc. 2003, 125, 1710. (b) Klamo, S. B.; Wendt, O. F.; Henling, L. M.; Day, M.
W.; Bercaw, J. E. Organometallics 2007, 26, 3018.
The 13C-JMOD NMR spectrum allows assignment of the
regiochemistry of the migratory insertion. In fact, it demon-
strates that Ca and Cb are secondary and tertiary carbons,
respectively, thus revealing that a 1,2-insertion of 1-hexene
occurs into the Hf-CAryl bond. As far as the stereochemistry
of the migratory insertion is concerned, the observation of a
strong NOE contact between Hb and H2 (Figure 1B) in-
dicates that the re-enantioface is selected by the catalyst
having an R-configuration at the bridging carbon (and
conversely the si-enantioface by the S-one). Consequently,
the butyl chain in the β-position is oriented back toward the
(19) Mehrkhodavandi, P.; Bonitatebus, P. J.; Schrock, R. R. J. Am.
Chem. Soc. 2000, 122, 7841.

5448 Organometallics, Vol. 28, No. 18, 2009
Zuccaccia et al.
Figure 1. NMR characterization of the monoinserted product 3, including (A) a molecular sketch of ion pair 3 with red arrows
indicating relevant NOEs; (B) a section of the 1H-NOESY NMR spectrum of 3; (C) a section of the 1H,13C-HMBC NMR spectrum of 3
showing the long-range correlation between H2 and Cb; (D) a section of the 1H-COSY NMR spectrum of 3.
pyridine ring of the ligand (Figure 1A). The strong interac-
tions observed in the 1H-NOESY spectrum between H38 and
H19, H31, and H32 (Figures 1B and S15) show that the
stereochemistry at the metal remains the same as that of the
starting ion pair 2, with the terminal methyl group oriented
on the opposite side with respect to H17.10 The presence of
the Ca-Cb bridge introduces some flexibility, allowing the
naphthyl ring to release the eclipsing interaction with the
backbone of the pyridine ring present in 2;10 consistently, H8
is found spatially close to H38, as indicated by the observa-
While the chirality of the metal center was lost, the assign-
ment of the remaining two stereocenters agrees with the
assigned structure in Figure 1A.
Finally, the structure of the ion pair 3 is determined. The
chemical shift difference between p-F and m-F resonances
(Δm-F,p-F = 4.64 ppm)20 of the counterion indicates that 3 is
present in solution as an inner-sphere ion pair21 where the
methyl group of the anion steadily points toward the metal
center. In full agreement, the 19F,1H HOESY NMR spectra
1
tion of a strong contact between them in the H NOESY
(20) The value of Δ(m,p-F) (19F NMR) is a good probe of the
spectrum (Figure 1B). A previous work9 had determined the
solid state structure of the ligand residue of a quenched
catalyst during a 4-methyl-1-pentene homopolymerization.
coordination of MeB(C6F5)3 at cationic d0 metals. Horton, A. D.;
-
de With, J. Organometallics 1997, 16, 5424, and references therein.
(21) Macchioni, A. Chem. Rev. 2005, 105, 2039.

Article
Organometallics, Vol. 28, No. 18, 2009 5449
Scheme 4. Possible Initiation and Propagation Starting from the
Aryl-Monoinserted Species BI
iPrPh and inserted olefin. The stereo- and regiochemistry
depend on whether the methyl is located on R1-R4. The
addition of the counterion to the calculations adds another
-
level of complexity, as X1 and X2 can be Me or MeB(C6F5)3
.
If X1=MeB(C6F5)3-, then the anion is opposite or trans to
the inserted olefin, while if X2 = MeB(C6F5)3-, then the
anion is adjacent to the inserted olefin.
Table 1 collects the data on the reactant (C1, C2) species,
the transition states (TSs) (C3-C6), and the products
(C7-C10) of ethylene and propylene insertions.23 The
data are enthalpies relative to the lowest energy uninserted
species, 2. The data without the counterion are also included
in the table as the lower set of entries (note that there are only
half as many structures since X1=Me and X2 is the open
coordination site). There is a large difference in the scales
with and without counterion; with the anion, the barriers
are realistic values for enthalpies of activation, but with-
out a counterion, the reactions are too exothermic and the
barriers too low. However, in both cases, the different TSs
can be compared to predict the selectivity, while with the
anion, the barrier to activation can also be estimated quali-
tatively.
The first thing to observe is that, regardless of whether
anions are included, insertions are always faster for ethylene
(e) than propylene. Since our focus is on the selectivity of R-
olefin insertion, we shall not discuss ethylene insertions any
further. As observed from the data with no counterion,
insertions are favored syn, i.e., on the same face as the iPrPh
group, as the lowest energy TS is C4a. The next lowest energy
TSs are C4b and C6a. These results agree with previous
computational work focused on a comparison of insertions
into the Hf-CAryl and Hf-CMe bonds.9 However, this result
is inconsistent with the species assigned by NMR from the
previous section. The structure C6a corresponds to the
experimentally determined one, despite the fact that it resides
3.5 kcal/mol above C4a.
Following this inconsistency, the calculations were ex-
tended to include the anion. It appears that the counterion
prefers to reside adjacent to the inserting moiety, position X2,
in agreement with its experimentally determined position in
both the reactant10 and product. When the anion is included,
the facial selectivity changes, and C6a is the lowest energy
TS, in agreement with the experimentally identified isomer.
The next lowest energy structures, C5b and C6b, correspond
to 1,2-insertions on the same face, only with opposite
stereochemistry (the anion is in the two different positions).
These TSs are less than 1 kcal/mol above C6a. Even if the
position of the anion is ignored, all four 1,2-insertions on
both faces lie within 2.3 kcal/mol of C6a. The lowest 2,1-
inserted TS (C6c) is only 4.3 kcal/mol higher in energy. While
the computations and experiments agree on the most likely
species, it is clear that a myriad of different diastereomers
Figure 2. Reactant species and transition states/products that
can be formed from insertion of an olefin into the Hf-CNaphthyl
bond.
show interionic contacts between the o-F nuclei and
some resonances of the cation that are much stronger than
those due to m-F and p-F nuclei.22 Specifically, the anion
preferentially interacts with H38, Ha, Ha, and H32 and, more
weakly, with H35, H31, Hb, H2, H27, H28, and H29
(Supporting Information).
It is worth noting that the monoinserted ion pair 3 is
intercepted under catalytic conditions, and its formation
occurs on a time scale comparable with that of olefin
enchainment. Interestingly, 3 has an intact Hf-Me moiety,
indicating that a single olefin insertion occurs exclusively
into the Hf-CAryl bond. An a posteriori inspection shows
that 3 is already present before complete 1-hexene consump-
tion even in the first spectrum recorded approximately 2 min
after the addition of B(C6F5)3 (Figure 2S, Supporting In-
formation). On the other hand, organometallic species bear-
ing a polymeryl chain are never observed under these
conditions. This suggests that either initiation of 3 (a BI-like
species) to give species BII is much slower than subsequent
propagation (Scheme 4) or, alternatively, 3 is an “out of
cycle” species and is irrelevant to catalysis.
Computational Probe of Selectivity of First Olefin Inser-
tion. Computationally, we chose to examine insertions into
the Hf-CAryl bond using density functional theory (DFT).
Previous calculations indicated that these reactions proceed
with 1,2-regioselectivity, stereoselectivity directs the substi-
tuent toward the pyridine, and the facial selectivity occurs
from the same side as the isopropylphenyl (iPrPh).9 While the
regio- and stereochemistries agree with the experimental
assignment of the inserted species, the facial selectivity does
not. We therefore revisited the calculations including the
borane activator {B(C6F5)3} to determine its role in the
Hf-CAryl insertion process.
Figure 2 shows possible insertion modes with the syn and
anti species drawn to indicate the relative positions of the
(22) (a) Zuccaccia, C.; Stahl, N. G.; Macchioni, A.; Chen, M. C.;
Roberts, J. A.; Marks, T. J. J. Am. Chem. Soc. 2004, 126, 1448. (b) Song,
F.; Lancaster, S. J.; Cannon, R. D.; Schormann, M.; Humphrey, S. M.;
Zuccaccia, C.; Macchioni, A.; Bochmann, M. Organometallics 2005, 24,
1315.
(23) Computational structures are labeled with the prefix C.

5450 Organometallics, Vol. 28, No. 18, 2009
Zuccaccia et al.
Table 1. DFT Energies (kcal/mol) Relative to the Reactants (C1, C2) of the TSs (C3-C6) and Products (C7-C10) for Ethylene (e) and
Propylene (a-d) Insertion into the Hf-CAryl Bond with MeB(C6F5)3- Anion and without Aniona
TS syn
TS anti
product syn
C8
product anti
C3
C4
C5
C6
C7
C9
C10
R1 = Me, R2-4 = H (a)
X1 = MeB(C6F5)3-, X2 = Me
X1 = Me, X2 = MeB(C6F5)3
21.2
25.3
-1.2
-8.4
-11.8
-8.6
-
20.4
-6.8
19.0
-3.3
-9.9
-9.6
X1 = Me, X2 = vacancy
R2 = Me, R1,3-4 = H (b)
X1 = MeB(C6F5)3-, X2 = Me
-25.2
-22.7
22.6
30.7
25.6
18.7
19.7
40.0
26.2
22.6
-4.7
-5.3
2.0
-
X1 = Me, X2 = MeB(C6F5)3
21.3
-5.2
20.0
-1.7
-13.9
-29.2
-13.0
-26.8
X1 = Me, X2 = vacancy
R3 = Me, R1-2,4 = H (c)
X1 = MeB(C6F5)3-, X2 = Me
-
X1 = Me, X2 = MeB(C6F5)3
27.2
1.4
23.3
4.9
-11.6
-27.4
-9.7
-20.4
X1 = Me, X2 = vacancy
R4 = Me, R1-3 = H (d)
X1 = MeB(C6F5)3-, X2 = Me
-8.4
-
X1 = Me, X2 = MeB(C6F5)3
X1 = Me, X2 = vacancy
R1-4 = H (e)
27.9
-2.7
24.9
-0.1
-8.0
-26.8
-3.6
-24.3
X1 = MeB(C6F5)3-, X2 = Me
X1 = Me, X2 = MeB(C6F5)3
X1 = Me, X2 = vacancy
-9.5
-17.3
-
17.0
-7.7
15.4
-4.3
-18.4
-33.5
-17.8
-27.9
a See Figure 2 for the structures.
could be formed. In addition, the enthalpic barrier for propylene
insertion (C6a) is 19.0 kcal/mol, suggesting slow activation.
The product energies are also collected in Table 1. A very
interesting result is that the kinetically favored product,
C10a (from TS C6a), is not nearly the most stable thermo-
dynamic species. In fact, even if the position of the anion is
ignored, five of the eight other diastereomers are more stable,
including two different 2,1-inserted species, C8c and C10c.
The favored kinetic product, C10a, is 4.3 kcal/mol less stable
than C8b, which is the most stable product.
Scheme 5. Species Derived from the Reaction of 2 with 3-Ethoxy-
1-propylenea
Trapping the Monoinserted Species by Means of Chelating
Olefins. Although the above reactions directly demonstrate
the ability for R-olefin insertion into the Hf-CAryl bond of 2,
quantitative generation of a monoinsertion product is hin-
dered by polymer formation. In order to explore the scope of
olefin insertion into the Hf-CAryl bond and, at the same
time, avoid polymer formation, we turned our investigation
to stoichiometric reactions of heteroatom-containing olefins
that should prevent multiple insertions due to chelation
effects. This strategy has been successfully applied by Teuben
and co-workers in migratory insertion of 3-ethoxy-1-propy-
lene into the Zr-Me bond of [(η5-C5Me5)2ZrMe][MeB-
(C6F5)3], leading to a single 1,2-insertion with formation of
an oxygen-stabilized cationic zirconacyclopentane.16 Re-
sults of reactions of 2 with both 3-ethoxy-1-propylene and
2-vinylpyridine as chelating olefins are described below.
Computationally, we examined olefin insertions into the
simplified achiral catalyst (no isopropylphenyl group) with
no counterion. The goal was to understand the stabilization
gained by the chelate to see if insertion into the Hf-CAryl
bond is predicted. The relative energies of the six transition
states for olefin insertion into the Hf-CAryl and Hf-CMe
bond are collected in Table 2 for propylene (C11), 4-vinyl-
pyridine (C12), 2-vinylpyridine (C13), and 3-ethoxy-1-pro-
pylene (C14). For the simple R-olefin, propylene, insertions
are favored into the Hf-CAryl bond in a 1,2-fashion (C11a).
However, the substrates 2-vinylpyridine and 3-ethoxy-1-
propylene preferentially insert in a 2,1-fashion, maximizing
interactions with the metal. The calculations suggest
that for these latter two substrates insertions into the
a X is a molecule of 3-ethoxy-1-propylene coordinated at the metal as
a neutral ligand through the oxygen atom. Values of δ (ppm) for selected
13C resonances are shown in red. The stereoisomer with R-configuration
at C* is shown, although a racemic mixture was used.
Hf-CAryl bond (C13d, C14d) are preferred over the those
into the Hf-CMe bond (C13g, C14g), although both produce
substantially more stable products than 1,2-insertions. These
results suggest that the use of a chelating olefinic substrate
could trap an aryl-inserted monomer in a stoichiometric
reaction.
(a) 3-Ethoxy-1-propylene. The reaction of 2 with a stoi-
chiometric amount of 3-ethoxy-1-propylene (EOP) turned
out to be more complicated than expected, producing a mix-
ture of species under all examined experimental conditions.
Despite the complicated product mixture, it is clear that a
fast and complete anion displacement from the first coordi-
nation sphere occurs, since the 19F NMR spectrum shows a
single set of sharp resonances corresponding to a noncoor-
-
dinated MeB(C6F5)3 anion (Δm-F,p-F = 2.42 ppm).20 On
1
the contrary, the H NMR spectrum shows rather broad
resonances with several singlets in the Hf-Me region. Addi-
tion of a small excess of 3-ethoxy-1-propylene (2-3 equiv)
leads to separation of an oil from the solution. NMR
investigation of this oil (Supporting Information) reveals
that a mixture of two products is formed in approximately a
70:30 molar ratio; the structure of the major component of
this oil is assigned to the ion pair 4 (Scheme 5). A sample
enriched (ca. 50-60% of the mixture) in the other species

Article
Organometallics, Vol. 28, No. 18, 2009 5451
Scheme 6. Postulated Pathway for Reaction of 2 with 3-Ethoxy-1-propylene Leading to the Formation of 4 and 5a
a X is the MeB(C6F5)3 anion, and Y is a molecule of 3-ethoxy-1-propylene coordinated at the metal as a neutral ligand through the oxygen atom.
Table 2. DFT Energies of Six TSs Relative to the 1,2-Aryl Insertion Species, a, for Four Different Monomersa
substrate
1,2-aryl, a
1,2-aryl, b
2,1-aryl, c
2,1-aryl, d
1,2-Me, f
2,1-Me, g
propylene C11
4-vinylpyridine C12
2-vinylpyridine C13
3-ethoxy-1-propylene C14
0.0
0.0
0.0
0.0
1.9
2.6
3.3
1.7
6.9
4.5
3.6
2.5
3.7
2.6
-17.0
-16.0
7.0
8.7
11.1
9.1
11.2
7.8
-7.9
-7.9
a Only insertions into the Hf-CMe with the olefinic substituent directed away from the DIP were examined. The structures of the Hf-CAryl insertion
species (a-d) correspond to that in Figure 2.
Scheme 7. Reaction of Ion Pair 2 with 2-Vinylpyridinea
was obtained when 2 was reacted with a smaller excess of
3-ethoxy-1-propylene. Although complete assignment was
strongly hampered by extensive overlapping, NMR spectra
(Supporting Information) suggest the molecular structure 5
shown in Scheme 5.
Evidently, the reactivity of the ion pair 2 with EOP is
much more complex than that of [(η5-C5Me5)2ZrMe][MeB-
(C6F5)3]16 and involves a C-O bond cleavage. Three differ-
a The stereoisomer with R-configuration at C* is shown although a
racemic mixture was used.
ent mechanisms were proposed to explain the formation of 4
and 5, and a thorough computational study was performed
to provide insight into the likely path. Two possibilities begin
24
with a first insertion of EOP into either the Hf-CMe or
Hf-CAryl bond, β-ethoxy elimination, and reinsertion of the
propenyl group into the Hf-CMe bond.
Hf-CAryl bonds, while the third involves a concerted route
to arrive at 5 directly through breaking the Hf-Cnaphthyl
bond.25 Complete computational details and a thorough
discussion of the differences in these three pathways
are given in the Supporting Information. Overall, the evi-
dence overwhelmingly suggests that the reaction proceeds as
shown in Scheme 6 through a 2,1-insertion of EOP into the
(b) 2-Vinylpyridine. In order to avoid complications due
to C-heteroatom activation, we decided to use a substituted
R-olefin in which the donor atom is incorporated into a more
robust aromatic ring. The reaction of 2 with 2-vinylpyridine
leads to the prevalent (80-90%) formation of the ion pair 6
(Scheme 7) through a 2,1-insertion of the olefin into the
Hf-CAryl bond. The structural assignment was ascertained
by the absence of resonances around 200 ppm in the 13C NMR
spectrum, where the naphthyl cyclometalated carbon reso-
(24) Komiya, S.; Srivastava, R. S.; Yamamoto, A.; Yamamoto, T.
Organometallics 1985, 4, 1504, and references therein.
(25) (a) Cheung, M. S.; Chan, H.-S.; Xie, Z. Organometallics 2005, 24,
5217. (b) Trifonov, A. A.; Spaniol, T. P.; Okuda, J. Dalton Trans. 2004, 2245.
(c) van der Boom, M. E.; Liou, S.-Y.; Ben-David, Y.; Shimon, L. J. W.;
Milstein, D. J. Am. Chem. Soc. 1998, 120, 6531. (d) Deelman, B.-J.; Booij,
M.; Meetsma, A.; Teuben, J. H.; Kooijman, H.; Spek, A. L. Organometallics
1995, 14, 2306, and references therein.
1
nance falls,10 and by the H,13C-HMBC NMR experiment
that establishes the connection Ci-C39-C40-C1 (Figure 3).
The regiochemistry of the insertion is again determined by the
13C-JMOD NMR spectrum; C39 and C40 are found to be

5452 Organometallics, Vol. 28, No. 18, 2009
Zuccaccia et al.
8.1 Hz, while the corresponding constant between H39 and
H40eq is below the experimental detectability (<1 Hz). Accord-
ing to the Karplus relationship, these values suggest a quasi-
eclipsed conformation (Figure 5). In agreement with this
steric arrangement, H40ax rather than H40eq shows an intense
NOE interaction with H39. Coordination of the inserted olefin
through the pyridine nitrogen is further supported by the
NOE cross-peak between H38 and Hd (Figure 4A). Again,
the presence of the C39-C40 bridge introduces some flexibi-
lity to release the eclipsing interaction with the backbone
pyridine;10 as a consequence, H8 is found spatially close to
H38 (Figure 4A).
Although complete characterization of the two byproducts
accounting for 10-20% of the product mixture is impossible,
resonances around δH = -1.2 ppm typical of the Hf-CMe
moiety are present (Figure 29S, Supporting Information). NMR
evidence for olefin insertion into the Hf-CMe bond is not
observed in this reaction, suggesting that the first insertion of
2-vinylpyridine proceeds quantitatively into the Hf-CAryl bond.
Following characterization of ion pair 6, the possibility
that this species may itself catalyze R-olefin polymeriza-
tions was investigated. This species shows no reaction with
1-hexene, probably a consequence of N-pyridine coordina-
tion to hafnium. Reactivity is observed upon addition of a
stoichiometric amount of 2-vinylpyridine, leading to com-
plete consumption of 6 with formation of ion pair 729 and
evolution of methane (Scheme 8). In 7, the olefinic carbon
directly bonded to the hafnium resonates at δC = 211 ppm,
in agreement with the literature.26 The coupling constant of
the olefinic protons (ca. 15 Hz) indicates a trans geometry
around the double bond. As in 6, the 19F NMR spectrum
shows a single set of resonances corresponding to an outer-
-
sphere MeB(C6F5)3 anion.20 Considering the trans geo-
metry of the double bond, it seems more plausible that
the N-pyridine atom of the inserted olefin coordinates at
the metal center, pushing the anion to the outer coordina-
tion sphere. As in other cases, NOE studies establish that
the CH moiety directly bonded to hafnium is positioned in
a pseudo-trans position with respect to H17. Similar
C(sp2)-H bond activations of 2-vinylpyridine have
been observed for a number of metal complexes,30 but to
our knowledge, these examples are all cyclometalations
Figure 3. Three sections of the 1H,13C HMQC (A) and 1H,13C
HMBC (B and C) NMR spectra of complexes 6 (C6D5Cl, 298
K), demonstrating the connection Ci-C39-C40-C1.
tertiary and secondary carbons, respectively. These data are all
consistent with a 2,1-insertion, which is typically observed for
aromatic olefins.26,27 The 19F NMR spectrum shows a single set
-
of resonances corresponding to an outer-sphere MeB(C6F5)3
anion,20 strongly suggesting coordination of the nitrogen of 2-
vinylpyridine to the metal center.
(29) A similar alkyl-displacement reaction has been recently reported
by Esteruelas and co-workers for osmium and ruthenium derivatives:
Buil, M. L.; Esteruelas, M. A.; Goni, E.; Olivan, M.; Onate, E.
Organometallics 2006, 25, 3076, and references therein.
The three-dimensional structure of 6 in solution is
determined through 1H-NOESY spectroscopy (Figure 4).
The strong interactions of H38 with H19, H31 and H32
(Figure 4A and C) indicate that the stereochemistry at the metal
is the same as the starting ion pair 2,10 with the terminal methyl
group oriented on the opposite side with respect to H17.
Concerning the olefin insertion stereochemistry, the strong
NOE contacts between H39 and both H34 and H35
(Figure 4D), as well as the absence of any appreciable interac-
tion between H38 and H39, strongly support a selection of the
re-enantioface.28 The close proximity of H40eq and H2
(Figure 4B) also provides a hint for 2,1-regiochemistry. The
3JHH coupling constant between H39 and H40ax amounts to
(30) (a) Foot, R.; Heaton, B. T. J. Chem. Soc., Chem. Commun. 1973,
838. (b) Bruce, M. I.; Goodall, B. L.; Matsuda, I. Aust. J. Chem. 1975, 28,
1259. (c) Foot, R.; Heaton, B. T. J. Chem. Soc., Dalton Trans. 1979, 295. (d)
Burgess, K.; Holden, H. D.; Johnson, B. F. G.; Lewis, J.; Hursthouse, M. B.;
Walker, N. P. C.; Deeming, A. J.; Manning, P. J.; Peters, R. J. Chem. Soc.,
Dalton Trans. 1985, 85. (e) Newkome, G. R.; Theriot, K. J.; Cheskin, B. K.;
Evans, D. W.; Baker, G. R. Organometallics 1990, 9, 1375. (f) Jia, G.; Meek,
D. W.; Gallucci, J. C. Organometallics 1990, 9, 2549. (g) Wong, W.-Y.;
Wong, W.-T. J. Organomet. Chem. 1996, 513, 27. (h) Wong-Foy, A. G.;
Henling, L. M.; Day, M.; Labinger, J. A.; Bercaw, J. E. J. Mol. Catal. A:
Chem. 2002, 189, 3. (i) Lau, J. P.-K.; Wong, W.-T. Inorg. Chem. Commun.
2003, 6, 174. (j) M€uller, J.; Hirsch, C.; Ha, K. Z. Anorg. Allg. Chem. 2003,
629, 2180. (k) Navarro, J.; Sola, E.; Martín, M.; Dobrinovitch, I. T.; Lahoz, F.
J.; Oro, L. A. Organometallics 2004, 23, 1908. (l) Ozerov, O. V.; Pink, M.;
Watson, L. A.; Caulton, K. G. J. Am. Chem. Soc. 2004, 126, 2105. (m)
ꢀ
~
Eguillor, B.; Esteruelas, M. A.; Olivan, M.; Onate, E. Organometallics 2005,
ꢀ
ꢀ
~
24, 1428. (n) Esteruelas, M. A.; Fernandez-Alvarez, F. J.; Olivan, M.; Onate,
E. J. Am. Chem. Soc. 2006, 128, 4596. (o) Buil, M. L.; Esteruelas, M. A.;
(26) Guram, A. S.; Jordan, R. F. Organometallics 1991, 11, 3470.
(27) Correa, A.; Galdi, N.; Izzo, L.; Cavallo, L.; Oliva, L. Organo-
metallics 2008, 27, 1028, and references therein.
(28) The si-enantioface is consequently selected when the isomer with
the S-configuration at the bridging amine-pyridyl carbon is considered.
ꢀ
~
Goni, E.; Olivan, M.; Onate, E. Organometallics 2006, 25, 3076. (p) Azam,
K. A.; Bennett, D. W.; Hassan, M. R.; Haworth, D. T.; Hogarth, G.; Kabir, S.
E.; Lindeman, S. V.; Salassa, L.; Simi, S. R.; Siddiquee, T. A. Organome-
tallics 2008, 27, 5163.

Article
Organometallics, Vol. 28, No. 18, 2009 5453
Figure 4. Four sections of the 1H-NOESY NMR spectrum of complex 6 (C6D5Cl, 298 K).
Scheme 8. Reaction of Ion Pair 6 with 2-Vinylpyridinea
in formation of a BI-type species with potential catalytic
activity. At the same time, these data do not exclude that the
aryl cyclometalated ion pair 2 could also be an active
catalyst.
In order to perform a more stringent competitive experi-
ment between these two species, we decided to investigate
the activation of the dibutyl analogue of 1 with [CPh3]-
[B(C6F5)4]. Dialkyl complexes of group IV metals bearing
alkyl substituents with one or more β-hydrogens are not
typically thermally robust,32 but several examples of stable
group IV dibutyl complexes have been reported.17,33 The
motivation for this experiment centers around the possibility
of generating a single molecule of 1-butene in close proxi-
mity to each [{N-,N,CAryl-}Hf(n-Bu)][B(C6F5)4] ion pair
through β-H abstraction at one n-butyl group (Scheme 9).17
The experiment is designed to favor pathway A; since the
Hf-n-Bu species is a better mimic for the propagating
Hf-polymeryl species, rates of initiation and propagation
a The stereoisomer with R-configuration at C* is shown although a
racemic mixture was used.
(32) (a) Negishi, E.-I.; Takahashi, T. Acc. Chem. Res. 1994, 27 5, 124,
and references therein. (b) Chang, B.-H.; Tung, H.-S.; Brubaker, C. H., Jr.
Inorg. Chim. Acta 1981, 51, 143. (c) McDermott, J. X.; Wilson, M. E.;
Whitesides, G. M. J. Am. Chem. Soc. 1976, 98, 6529. (d) Binger, P.; M€uller,
P.; Benn, R.; Rufínska, A.; Gabor, B.; Kr€uger, C.; Betz, P. Chem. Ber. 1989,
122, 1035. (e) Sinnema, P.-J.; van der Veen, L.; Spek, A. L.; Veldman, N.;
Teuben, J. H. Organometallics 1997, 16, 4245.
(33) For leading examples, see: (a) Paolucci, G.; Pojana, G.; Zanon,
J.; Lucchini, V.; Avtomonov, E. Organometallics 1997, 16, 5312. (b)
Amor, F.; Spaniol, T. P.; Okuda, J. Organometallics 1997, 16, 4765. (c)
Amor, F.; Butt, A.; du Plooy, K. E.; Spaniol, T. P.; Okuda, J. Organometallics
1998, 17, 5836. (d) Mehrkhodavandi, P.; Bonitatebus, P. J., Jr.; Schrock, R. R.
J. Am. Chem. Soc. 2000, 122, 7841. (e) Mehrkhodavandi, P.; Schrock, R. R.;
Bonitatebus, P. J. Organometallics 2002, 21, 5785. (f) Ernst, R. D.; Harvey,
B. G.; Arif, A. M. Z. Kristallogr.-New Cryst. Struct. 2004, 219, 398. (g)
Schrock, R. R.; Adamchuk, J.; Ruhland, K.; Lopez, L. P. H. Organometallics
2005, 24, 857. (h) Boone, H. W.; Coalter, J. N., I. I. I.; Frazier, K. A.; Iverson,
C. N.; Munro, I. M.; Vosejpka, P. C. PCT Int. Appl. Publ. No. WO 2007/
130307.
Figure 5. Newmann projection of complex
C40-C39 bond. Red arrows indicate the observed NOE inter-
actions.
6 along the
facilitated by chelation of the pyridine nitrogen.31 The
trans configuration of the double bond indicates that
this C-H activation process proceeds without chelation
assistance.
Activation of {N-,N,CAryl-}Hf(n-Bu)2 (8) with [CPh3]-
[B(C6F5)4]. The results discussed above clearly suggest that
the first insertion of olefin preferentially occurs into the
Hf-CAryl bond even under catalytic conditions, resulting
(31) Jun, C.-H.; Moon, C.-W.; Lee, D.-Y. Chem.;Eur. J. 2002, 8,
2423. and references therein.

5454 Organometallics, Vol. 28, No. 18, 2009
Scheme 9. Proposed Pathways for Reaction of 8 with [CPh3][B(C6F5)4]
Zuccaccia et al.
should be competitive. According to the literature,34-36 this
should lead to an increased percentage of active sites and
the formation of mainly C8-dimers, if this path is opera-
tive. On the contrary, if pathway B is operative, little
difference should be observed with respect to polymeriza-
tions conducted with 2 under the assumption that the first
insertion into the Hf-CAryl bond is slow (relative to
propagation) and thus limits the formation of the active
species.9 If olefin binding is reversible, this mechanism
should result in formation of higher insertion products
rather than butene dimers with a product distribution
corresponding to the relative rates of insertion into the
Hf-CAryl and Hf-CButyl bonds.
Reaction of 8 with [CPh3][B(C6F5)4] in toluene-d8 at
room temperature within an NMR tube results in a quan-
titative formation of HCPh3 and two diastereoisomeric
organometallic species reasonably assignable to [{N-,N,
CAryl-}Hf(n-Bu)][B(C6F5)4] 9a and 9b ion pairs. n-Bu
cations of this type are unusual for electrophilic group IV
species, as they are susceptible to decomposition via
β-hydride elimination.17b,33d,g However, the stability of
9a/b is not surprising considering the ability of these catalysts
to form very high molecular weight polyolefins at high
reaction temperatures.2-5 Decomposition products derived
from the transfer of the C6F5 moiety(ies) from the counterion
to the cation are also observed.37 Interestingly, no evidence is
Scheme 10. Activation of 8 with [CPh3][B(C6F5)4]a
a The stereoisomer with R-configuration at C* is shown although a
racemic mixture was used.
observed for the presence of free/coordinated 1-butene or
oligomers formed by insertion(s) of 1-butene into the
Hf-CButyl bond. However, the 13C NMR spectrum of the
reaction mixture (acquired during the first 8 h of reaction)
instead shows relatively sharp resonances corresponding to
poly(1-butene). Integration of the 13C resonances of the
polymer with respect to those of HCPh3 indicates that all
the liberated 1-butene is polymerized. Consequently, the
poly(1-butene) is likely formed on a time scale similar to
that of the formation of HCPh3. This is in sharp contrast to
the behavior reported by Mehrkhodavandi and Schrock,
who observed products from single monomer insertions
into Hf-CEthyl and Hf-CisoPropyl bonds under a similar
scheme with diamido pyridine complexes (pathway A in
Scheme 9).17b Encouraged by this result, we decided to carry
out a reaction on a much larger scale in order to isolate and
characterize the resulting poly(1-butene) in this unusual
“stoichiometric polymerization”.
Toward this aim, activation of 2.5 g of 8 with [CPh3]-
[B(C6F5)4] produced a stoichiometric amount of low
molecular weight, highly isotactic poly(1-butene). 13C
NMR of the polymer shows saturated P-(n-butyl) and
CH3CH(C2H5)-P chain ends in approximately a 1:1 molar
ratio. Peaks due to stereo- and regiodefects are below the
level of detection, and olefinic protons are not observed
in the 1H NMR spectrum. The number-average degree
of polymerization estimated by 13C NMR is ca. 25, indicat-
ing activity of approximately 4% of the Hf centers.
(34) Landis, C. R.; Christianson, M. D. Proc. Natl. Acad. Sci. 2006,
103, 15349.
(35) Sillars, D. R. Ph.D. thesis, Univ. of Wisconsin, Madison, WI,
2003.
ꢀ ꢀ
(36) Ducere, J. M.; Cavallo, L. Organometallics 2006, 25, 1431, and
references therein.
(37) For some examples of C6F5 transfer to metal, see: (a) Phomphrai,
K.; Fenwick, A. E.; Sharma, S.; Fanwick, P. E.; Caruthers, J. M.; Delgass,
W. N.; Abu-Omar, M. M.; Rothwell, J. P. Organometallics 2006, 25, 214.
(b) Metcalfe, R. A.; Kreller, D. I.; Tian, J.; Kim, H.; Taylor, N. J.; Corrigan, J. F.;
Collins, S. Organometallics 2002, 21, 1719. (c) Woodman, T. J.; Thornton-Pett,
M.; Hughes, D. L.; Bochmann, M. Organometallics 2001, 20, 4080. (d) Guerin,
F.; Stewart, J. C.; Beddie, C.; Stephan, D. W. Organometallics 2000, 19,
2994. (e) Scollard, J. D.; McConville, D. H.; Rettig, S. J. Organometallics
1997, 16, 1810. (f) Yang, X. M.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc.
1994, 116, 10015. For a recent theoretical work, see: Wondimagegn, T.;
Xu, Z.; Vanka, K.; Ziegler, T. Organometallics 2004, 23, 3847, and references
therein.

Article
Organometallics, Vol. 28, No. 18, 2009 5455
The number-average molecular weight (Mn) value of
950 g/mol, deduced from gel permeation chromatography
(GPC), corresponds to approximately 6% active centers.
The poly(1-butene) possesses a narrow polydispersity index
of 1.2, consistent with “living” catalyst behavior and in line
with the observed end groups. GC/MS analysis of the mother
liquors after quenching shows ca. 10% of a fraction,
consistent with a 1-butene-appended ligand (m/z = 568)
and, more importantly, no evidence for 1-butene oligomers
(C8, C12, C16, etc.).
chelating olefin, 2-vinylpyridine.40 The simultaneous pre-
sence of Hf-Me and Hf-alkylaryl moieties in both of these
examples confirms the relative ease of insertion into the
Hf-CAryl bond. Insertion of 3-ethoxy-1-propylene led to
complicated activation, but from computational studies and
logic, one can deduce that insertion first goes through the
Hf-CAryl bond.
Finally, the generation of precisely 1 equiv of 1-butene
in close proximity to each metal center (i.e., perfect
mixing) resulted in the formation of low molecular weight,
highly isotactic poly(1-butene) rather than a stoichio-
metric insertion product or oligomers. These characteris-
tics indicate the presence of a very low percentage of active
metal centers, consistent with slow initiation relative to
propagation. These data suggest that the active species is
generated upon insertion of monomer into the Hf-CAryl
bond.
These results indicate that initiation is again much slower
than propagation in this peculiar polymerization reaction.
A simple kinetic scheme assuming that 5% of the catalysts
are active and consume the remainder of the monomer leads
to an estimate of the relative rate constants of 1000-fold for
polymerization over activation. Since we have estimated that
insertion into Hf-CPoly is about an order of magnitude
slower than insertion into the Hf-CAryl,
39 these data suggest
that the polymerization rates of catalysts of the AI-type
are 4 orders of magnitude slower than those of the BI-type
(for tetrakis-anions). Under the assumption that the n-butyl
group is a reasonable model for the growing polymer chain,36
the rate of the initial insertion into the Hf-CButyl bond is
expected to be similar to the rate of polymerization. The
simultaneous presence of a large fraction of uninitiated n-
butyl cations 9a/b and poly(1-butene) is not consistent with
pathway A of Scheme 9. These data strongly suggest that the
active sites are BI-type species from pathway B of Scheme 9,
and that AI-like ion pairs are not active or, at least, much less
active for R-olefin polymerization.
Experimental Section
All manipulations were performed in flamed Schlenk-type
glassware interfaced to a high-vacuum line (<10-5 Torr), or in a
nitrogen-filled Vacuum Atmospheres glovebox (<1 ppm O2).
Molecular sieves (MS) were activated for 24 h at 200-230 ꢀC
under dynamic vacuum. All the solvents and liquid reagents
were freeze-pump-thaw degassed on the high-vacuum line,
dried over the appropriate drying agent, vacuum-transferred to
a dry storage tube with a PTFE valve, and stored over activated
MS. Benzene-d6, toluene-d8, pentane, hexane, toluene, and 3-
ethoxy-1-propene were dried over Na/K alloy; 1,2-F2C6H4 was
refluxed for three days over P2O5 and finally dried over CaH2; 2-
vinylpyridine, chlorobenzene-d5, and methylene chloride-d2
were dried over CaH2; 4-methyl-1-pentene and 1-hexene were
dried over CaH2 or LiAlH4. [HNMe2Ph][B(C6F5)4] and
[CPh3][B(C6F5)4] were obtained from Boulder Scientific Com-
pany and were used as received. B(C6F5)3 was obtained from
Boulder Scientific Company and was purified by sublimation
(40-60 ꢀC, 10-5 Torr). HfCl4 (sublimed grade, 99.9þ% Hf,
<1.0% Zr) was purchased from Strem Chemical and used
as received. n-BuLi (2.5 M solution in hexanes) and BuMgCl
(2.0 M solution in diethyl ether) were purchased from
Acros and Aldrich Chemical Co., respectively, and used as
received.
NMR samples were prepared in oven-dried J-Young
NMR tubes. 1H, 13C{1H}, 1H-COSY, 1H-NOESY, 1H-ROESY
1H,13C-HMQC, 1H,13C-HMBC, 19F, and 19F,1H-HOESY
NMR experiments were performed on a Bruker Avance DRX
400 instrument. Referencing by residual solvent signals is rela-
tive to TMS. Typical mixing time for the Overhauser experi-
ments was in the range 120-800 ms. Complexes 1 and 2 were
prepared as described previously.10
Generation of the Monoinserted Species 3. Complex 1 (7.2 mg,
0.01 mmol) and 1-hexene (0.21 mL, 1.7 mmol, 170 equiv) were
dissolved in toluene-d8 (see Figure 1S for the corresponding 1H
NMR spectrum at -56 ꢀC) and cooled at -60 ꢀC using an
ethanol/liquid nitrogen bath. Then 0.20 mL of a toluene-d8
solution containing B(C6F5)3 (8 mg, 0.015 mmol) was injected
into the NMR tube, and while shaking, the tube was quickly
transferred to the NMR spectrometer (precooled to -56 ꢀC).
The polymerization was followed by 1H NMR; 95% of 1-hexene
was consumed after approximately 22 min, while resonances
due to 1-hexene were not detectable after 40-50 min (see
Figure 2S for 1H NMR spectra taken approximately 2 min
(black), 7 min (red), 15 min (blue), and 50 min (green) after the
addition of B(C6F5)3). When the olefin was completely con-
sumed, the temperature inside the NMR probe was raised
to -29 ꢀC in order to obtain sharper resonances, and the
organometallic complex present in solution was identified
Conclusions
Our quest to develop a detailed mechanistic understanding
of these pyridyl-amido complexes has uncovered a wealth of
interesting activation chemistry centered on the unusual
Hf-CAryl bond. Ascertaining the identity of the true active
species remains of great interest considering the utility of
these complexes in the production of specialty polyolefins.
The data presented here provide direct NMR evidence
concerning the preference for initial olefin insertion into
the Hf-CAryl bond over the Hf-CAlkyl bond. In the presence
of a large excess of 1-hexene at low temperature, a BI-type
monoinserted ion pair was directly observed and completely
characterized. Computational studies including the FAB
counterion showed that the kinetic product of insertion
was the same as that observed experimentally. However,
thermodynamically, this diastereomer is only the sixth
most stable species out of eight possible R-olefin insertion
products.
Calculations suggested chelating olefins, such as 2-vinyl-
pyridine or 3-ethoxy-1-propylene, would insert into the Hf-
CAryl bond in a 2,1-fashion to form a stable product. An
analogous BI species was generated and fully characterized
by the reaction of 2 with an equimolar amount of the
(38) The relative fraction of 1-butene-appended ligand is estimated
from uncorrected flame-ionization detection data.
(39) For the most likely pathway, the difference in these transition
states was estimated at 2.2 kcal/mol with insertion into the Hf-CAryl
bond lower than insertion into the Hf-CnButyl bond (ref 9).
(40) Bijpost, E. A.; Zuideveld, M. A.; Meetsma, A.; Teuben, J. H. J.
Organomet. Chem. 1998, 551, 159. For a more recent paper describing the
use of 3-(methylthio)-1-propylene for similar studies see: Richter, B.;
Meetsma, A.; Hessen, B.; Teuben, J. H. Angew. Chem., Int. Ed. 2002, 41,
2166.

5456 Organometallics, Vol. 28, No. 18, 2009
Zuccaccia et al.
as 3 by multinuclear and multidimensional NMR methods
(Figures 4S-25S).
and HB), 3.39 (m, H34, H41, and HC), 3.23 (m, HC), 3.17 (m,
HfOCH2CH3), 3.08 (m, HD), 2.97 (m, HfOCH2CH3), 2.87 (m,
HD), 2.66 (sept, 3JHH = 6.7, H24), 2.46 (sept, 3JHH = 6.7, H31),
2.28 (dd, 2JHH = 13.7, 3JHH = 6.8, H41), 2.10 (m, H39), 1.68 (m,
H39), 1.45 (m, H40),1.41 (d, 3JHH = 6.7, H36), 1.28 (d, 3JHH
=
6.7, H35), 1.19 (br, Me-B), 1.13 (t, JHH = 7.1, H38), 1.10 (d,
3
3JHH = 6.7, H24a), 0.77 (d, 3JHH = 6.7, H32), 0.61 (t, 3JHH
=
=
3
3
7.1, HE), 0.48 (t, JHH = 7.1, HfOCH2CH3), 0.44 (d, JHH
3
6.7, H24b), 0.13 (d, JHH = 6.7, H33). 13C{1H} NMR (C6D6/
C6D5Cl, 298 K): δ 169.9 (s, C15), 155.8 (s, C11), 149.6 (s, C23),
145.27 (s, C26), 145.23 (s, C26), 144.8 (s, C30), 142.7 (s, C1),
141.4 (s, C13), 139.2 (s, C18), 133.7 (s, C4), 132.0 (s, C3), 131.9
(s, C9), 130.7 (s, C10), 130.67 (s, C19), 130.0 (s, C2), 129.8 (s,
C5), 129.1 (s, C7), 128.0 (s, C6), 127.8 (s, CA), 127.0 (s, C12 and
C28), 127.4, 127.1, 127.0 (s, C20, C21, C22, CB), 125.6 (s, C29),
125.4.3 (s, C14), 125.2 (s, C27), 124.9 (s, C8), 77.6 (s, C40), 76.6
(s, CC), 72.5 (s, CD), 72.1 (s, C16), 67.7 (s, HfOCH2CH3), 36.3 (s,
C41), 29.5 (s, C31), 29.3 (s, C24), 29.07 (s, C34), 27.9 (s, C35),
27.1 (s, C39), 26.3 (s, C24a), 26.27 (s, C32), 25.0 (s, C36), 23.7 (s,
C33), 22.3 (s, C24b), 18.1 (s, HfOCH2CH3), 16.3 (s, C38), 13.7
(s, CE). 19F NMR (C6D6/C6D5Cl, 298 K, J values in Hz): δ
-132.16 (brd, 3JFF = 20.7, o-F), -164.7 (t, 3JFF = 21.0, p-F),
-167.2 (m, m-F).
Synthesis of Complex 6. Complex 2 (15 mg, 0.0122 mmol) was
charged into a J-Young NMR tube and dissolved using
a mixture of C7D8 and C6D5Cl (0.7 mL). 2-Vinylpyridine
(66 μL, of a 0.184 M solution in toluene-d8, 0.0122 mmol) was
added in three portions. The mixture was allowed to react for a
few minutes, and then all the volatiles were removed under
vacuum. The residue was then redissolved in C6D5Cl and
analyzed by NMR. Complex 6 can be isolated as a deep orange
oil starting from 70 mg of 2 and performing the reaction in neat
C7D8. Several attempts to grow single crystals of 6 suitable for
X-ray analysis were unsuccessful thus far (1H and 13C NMR
spectra of 6 are reported in the Supporting Information, Figures
29S-38S).
1H NMR (C7D8, 244K, J values in Hz): δ 7.76 (d, 3JHH = 8.3,
H3), 7.53 (d, 3JHH = 8.0, H5), 7.51 (d, 3JHH = 8.3, H2), 7.21 (m,
H6 and H7), 7.14 (buried under solvent resonances, H8), 7.09
(buried under solvent resonances, H19), 6.95 (m, H28 and H29),
6.86 (m, H20, H21, and H22), 6.60 (t, 3JHH = 7.8, H13), 6.51 (d,
3JHH = 7.8, H14), 6.42 (m, H12 and H27), 6.16 (s, H17), 3.83
(brm, Hb), 3.22 (sept, 3JHH = 6.7, H34), 2.89 (sept, 3JHH = 6.7,
H31), 2.71 (sept, 3JHH = 6.7, H24), 2.48 (br, Me-B), 1.38 (buried
under polymer resonances, Ha and Ha0), 1.25 (buried under
polymer resonances, H36), 1.15 (buried under polymer reso-
nances, H35), 1.08 (buried under polymer resonances, H24a),
0.97 (buried under polymer resonances, Hd), 0.79 (buried
under polymer resonances, Hd0), 0.78 (buried under polymer
resonances, Hc), 0.77 (d, 3JHH = 6.7, H32), 0.75 (m, Hg), 0.58
3
(m, Hc0), 0.31 (d, JHH = 6.7, H24b), 0.06 (s, H38), -0.28 (d,
3JHH = 6.7, H33). 13C{1H} NMR (C7D8, 244 K): δ 167.5 (s,
C15), 155.1 (s, C11), 147.1 (s, C23), 146.4 (s, C26), 145.7 (s, C30),
143.3 (s, C1), 140.6 (s, C13), 139.3 (s, C25), 138.4 (s, C18), 133.6
(s, C9), 133.3 (s, C10), 132.6 (s, C3), 129.9 (s, C2), 129.6 (s, C5),
129.3 (s, C19), 129.1 (s, C20, C21 or C22), 128.4 (s, C28), 127.5
(s, C12), 127.5 (s, C20, C21 or C22), 127.3 (s, C4), 126.3 (s, C6
and C7), 126.2 (s, C20, C21 or C22), 125.4 (s, C29), 125.3 (s,
C27), 123.9 (s, C8), 123.1 (s, C14), 78.1 (s, Ca), 72.4 (s, C16), 60.6
(s, C38), 48.0 (s, Cb), 39.6 (s, Me-B), 36.6 (s, Cc), 30.6 (s, Cd),
28.8 (s, C24), 28.6 (s, C34), 28.5 (s, C31), 26.1 (s, C35), 25.4 (s,
C36), 24.7 (s, C33), 24.5 (s, C24a), 24.0 (s, C32), 23.4 (s, Ce), 23.1
(s, C24b), 14.3 (s, Cg). 19F NMR (C7D8, 244 K J values in Hz): δ
-133.44 (brd, 3JFF = 23.2, o-F), -159.87 (t, 3JFF = 21.3, p-F),
-164.51 (m, m-F).
Generation of Complex 4. Complex 2 (30 mg, 0.0244 mmol)
was dissolved in C6D6 (0.7 mL), and 3-ethoxy-1-propene (8 μL,
0.0730 mmol) was added in one portion. After 3 h a deep-yellow-
orange oil separated from the solution. The supernatant was
removed and the oil redissolved in C6D6/C6D5Cl for NMR
analysis.
1H NMR (C6D5Cl, 298 K, J values in Hz): δ 7.71 (d, 3JHH
=
8.3, H3), 7.65 (d, 3JHH = 8.1, H5), 7.63 (d, 3JHH4 = 8.0, Ha), 7.58
3
(d, JHH = 7.9, H13), 7.46 (dt, 3JHH = 8.0, JHH = 1.7, Hb),
7.38 (m, H7), 7.33 (m, H6), 7.28 (d, 3JHH = 8.3, H2), 7.18 (m,
Hd), 7.17-7.10 (m, H12, H14, H20, H22, H28, and H29), 7.03
3
(d, JHH = 8.2, H8), 7.02-6.96 (m, H19, H21, H27), 6.57
(s, H17), 6.38 (m Hc), 4.06 (d, 2JHH = 14.8, H40eq), 3.31 (dd,
2JHH = 14.8, 3JHH = 8.1, H40ax), 3.18 (sept, 3JHH = 6.7, H31),
3.09 (sept, 3JHH = 6.7, H34), 2.89 (sept, 3JHH = 6.7, H24), 2.13
(d, 3JHH = 8.1, H39), 1.42 (d, 3JHH = 6.7, H36), 1.24 (d, 3JHH
=
6.7, H24a), 1.17 (br, Me-B), 0.90 (d, 3JHH = 6.7, H32), 0.74 (d,
3JHH = 6.7, H35), 0.54 (d, 3JHH = 6.7, H24b), 0.13 (d, 3JHH
=
6.7, H33), -1.01 (s, H38). 13C{1H} NMR (C6D5Cl, 298 K): δ
167.8 (s, C15), 163.3 (s, Ci), 154.1 (s, C11), 149.3 (s, C26), 148.7
(s, C30), 147.7 (s, C23), 145.1 (s, Cd), 143.1 (s, Cb), 142.7 (s,
C13), 138.9 (s, C1), 137.2 (s, C18), 134.0 (s, C9), 133.1 (s, C4),
132.2 (s, C3), 131.3 (s, C25), 129.1 (s, C5), 129.0 (s, C19), 128.6
(s, C7), 127.2 (s, C2), 126.7 (s, C6), 124.9 (s, Ca), 123.2 (s, C8),
119.3 (s, Cc), 129.57, 129.54, 128.56, 127.29, 126.77, 126.12,
124.97, 124.51 (s, C12, C14, C20, C21, C22, C27, C28, C29), 70.4
(s, C16), 64.8 (s, C39), 61.3 (s, C38), 31.6 (s, C40), 28.69 (s, C24),
28.66 (s, C32), 28.63 (s, C34), 28.2 (s, C31), 24.9 (s, C24a), 24.7
1H NMR (C6D6/C6D5Cl, 298 K, J values in Hz): δ 7.73 (d,
3JHH = 8.0, H5), 7.70 (d, 3JHH = 8.5, H3), 7.54 (d, 3JHH = 8.4,
H8), 7.40 (m, H6), 7.33 (m, H7), 7.21 (d, 3JHH = 7.7, H13), 7.19
3
(d, JHH = 8.5, H2), 7.08 (m, H29), 7.06-6.90 (m, H20, H21,
H22, and H28), 6.81 (m, H12 and H27), 6.72 (d, JHH = 7.7,
H14), 6.41 (s, H17), 6.37 (m, H19), 4.95 (m, HA), 4.79 (m, HA
3

Article
Organometallics, Vol. 28, No. 18, 2009 5457
(s, C35), 24.5 (s, C36), 23.1 (s, C33), 22.7 (s, C24b), 11.0 (v br,
additional hexane (30 mL). Solvent was removed by vacuum
from the combined hexane solutions to give the desired product
as a yellow solid: 3.2 g; 80%.
Me-B). 19F NMR (C6D5Cl, 298 K, J values in Hz): δ -132.09
3
3
(brd, JFF = 20.7, o-F), -164.8 (t, JFF = 21.0, p-F), -167.2
(m, m-F).
Generation of Ion Pair 7. Inside the glovebox, an NMR tube
containing a solution of 6 in C6D5Cl was treated with
a solution of 2-vinylpyridine in C6D5Cl (1 equiv). NMR
analysis showed complete consumption of 6 and formation
of complex 7. Alternatively, complex 7 may be conveniently
obtained by adding 2 equiv of 2-vinylpyridine to a solution of
ion pair 2.
1H NMR (C7D8, 298 K, J values in Hz): δ 8.71 (d, 3JHH = 7.6,
H2), 8.19 (m, H8), 7.77 (d, 3JHH = 7.6, H3), 7.67 (m, H5), 7.48
(d, 3JHH = 8.0, H12), 7.39 (m, H19), 7.26 (m, H7 and H6), 7.15
(dd, 3JHH = 7.5, 4JHH = 1.9, H29), 7.1-7.0 (m, H20, H21, H22,
H27, and H28), 6.88 (d, 3JHH = 8.0, H13), 6.61 (d, 3JHH = 8.0,
3
H14), 6.58 (s, H17), 3.75 (sept, JHH = 6.7, H31), 3.36 (sept,
3JHH = 6.7, H34), 2.91 (sept, 3JHH = 6.7, H24), 2.09 (m, Hβ),
1.85 (m, Hβ’), 1.49 (m, HR), 1.41 (m, H36 and H32), 1.55 (m, Hγ
1H NMR (C6D6/C6D5Cl, 298 K, J values in Hz): δ 7.72-7.65
(m, H8 and H5), 7.63 (d, 3JHH = 8.2, H3), 7.54 (brd, 3JHH = 8.2,
3
3
and Hγ ), 1.26 (m, HR), 1.20 (d, JHH = 6.7, H24a), 1.18 (d,
0
Ha), 7.45-735 (m, Hb, H6, and H7), 7.32 (t, JHH = 7.9,
3
H13), 7.25 (m, H21), 7.19 (d, JHH = 8.2, H2), 7.19-7.15
3JHH = 6.7, H35), 1.07 (m, HR’), 0.86 (t, 3JHH = 7.3, Hδ3or Hδ’),
3
0
(m, H20, H22, and H28), 7.12 (m, H19), 7.09 (d, JHH=14.5,
0.84 (t, 3JHH = 7.3, Hδ’ or Hδ), 0.82 (m, HR ), 0.67 (d, JHH
=
H41), 7.00-7.03 (m, H12, H14, and Hb0), 7.00 (dd, 3JHH = 7.9,
4JHH = 1.7, H27), 6.93 (dd, 3JHH = 7.7, 4JHH = 1.6, H29), 6.74
(s, H17), 6.68 (m, Hd), 6.64 (m, Hd), 6.60 (d, 3JHH = 14.5, H42),
6.46 (m, Hc), 6.39 (m, Hc0), 6.32 (brd, 3JHH = 7.9, Ha0), 4.39 (d,
2JHH = 13.9, H40), 3.55 (dd, 2JHH = 13.9, 3JHH = 5.9, H40),
6.7, H24b), 0.31 (d, 3JHH = 6.7, H33). 13C{1H} NMR (C7D8,
298 K): δ 205.5 (s, C1), 170.7 (s, C15), 164.4 (s, C11), 147.5 (s,
C26), 146.8 (s, C23), 146.4 (s, C30), 146.3 (s, C25), 144.2 (s, C10)
141.5 (s, C18), 140.6 (s, C13), 135.6 (s, C4), 134.6 (s, C2), 130.8
(s, C9), 130.4 (s, C19), 129.8 (s, C5), 129.7 (s, C3), 129.3, 128.8,
128.0, 126.8, 125.9 (s, C20, C21, C22, C27 and C28), 126.7 (s,
C7), 125.3 (s, C6), 124.9 (s, C29), 124.1 (s, C8), 120.5 (s, C12),
119.4 (s, C14), 88.3 (s, CR0), 83.6 (s, CR), 76.9 (s, C16), 30.6 (s,
Cβ0), 30.22 (s, Cβ), 30.18 (s, Cγ or Cγ0), 29.6 (s, Cγ0 or Cγ), 28.7
(s, C24), 28.6 (s, C34), 28.3 (s, C31), 27.0 (s, C32), 26.2 (s, C36),
25.5 (s, C35), 25.2 (s, C24a), 24.2 (s, C33), 23.1 (s, C24b), 14.2 (s,
Cδ), 14.1 (s, Cδ0).
3
3
3.23 (sept, JHH = 6.7, H34), 3.91 (sept, JHH = 6.7, H31),
2.78 (sept, 3JHH = 6.7, H24), 2.15 (d, 3JHH = 5.9, H39), 1.35 (d,
3JHH = 6.7, H36), 1.32 (br, Me-B), 1.25 (d, 3JHH = 6.7, H24a),
0.81 (d, 3JHH = 6.7, H32), 0.59 (d, 3JHH = 6.7, H24b), 0.16 (d,
3
3JHH = 6.7, H33), 0.02 (d, JHH = 6.7, H35). 13C{1H} NMR
(C6D6/C6D5Cl, 298 K): δ 212.7 (s, C41), 171.1 (s, Ci), 168.5 (s,
C15), 160.6 (s, Ci0), 155.7 (s, C11), 148.9 (s, C23), 147.9 (s, Cd0),
147.1 (s, C26), 146.8 (s, C30), 145.1 (s, Cd), 145.0 (s, C42), 142.5
(s, Cb), 142.4 (s, Cb0), 142.2 (s, C25), 140.18 (s, C18), 140.17 (s,
C13), 139.2 (s, C1), 133.9 (s, C4), 133.6 (s, C9), 132.2 (s, C10),
131.5 (s, C3), 130.4 (s, C19), 129.9 (s, C5), 129.6 (s, C2), 129.0 (s,
C6 or C7), 128.9 (s, C12), 128.5 (s, C28), 127.7 (s, C20 or C21),
127.4 (s, C29), 127.2 (s, C22 and C7 or C6), 126.1 (s, Ca), 125.9
(s, C8), 125.3 (s, C27), 124.9 (s, C14), 124.1 (s, Ca0), 122.9 (s, Cc0),
121.1 (s, Cc), 72.9 (s, C16), 61.0 (s, C39), 35.9 (s, C40), 29.4 (s,
C31), 29.3 (s, C24), 28.3 (s, C34), 26.8 (s, C32), 26.5 (s, C36), 26.3
(s, C24a), 24.12 (s, C33), 24.09 (s, C35), 22.6 (s, C24b), 11.5 (v br,
Me-B). 19F NMR (C6D6/C6D5Cl, 298 K, J values in Hz): δ
Isolation of Poly-1-butene from Scale-up Activation of 8 with
[CPh3][B(C6F5)4]. A 36 mL portion of a 0.0846 mM solution of 6
in methylcyclohexane (2.445 g, 3.05 mmol) was charged in a
250 mL flask equipped with a PTFE valve inside the glovebox.
The flask was connected to the high-vacuum line and the solvent
removed under reduced pressure. Inside the glovebox, 2.80 g of
[CPh3][B(C6F5)4] was added in the flask and approximately
80 mL of dry toluene was added in one portion under magnetic
stirring. After 45 min the flask was taken out from the glovebox
and connected to the high-vacuum line, where the solvent was
removed under reduced pressure. The foamy residue was
scratched with a spatula while the flask was open to the air.
Regular toluene (ca. 25 mL) was added in one portion, and the
mixture was vigorously stirred for few minutes. After a while, a
dark red-brown oily phase separated from the solution. The oil
(fraction 1) was separated from the solution by means of a
Pasteur pipet. The solution was concentrated to approximately
10-15 mL, resulting in the separation of an additional amount
of oil. The last oily phase (fraction 2) was separated from the
residual solution (fraction 3) by centrifugation. Fraction 3 was
poured in a flask and the solvent removed under vacuum.
Workup of fraction 2: approximately 200 mL of acidified
MeOH was added to the oil. The resulting cloudy suspension
was filtered. The mother liquor (fraction 2A) was stored in open
air to allow solvent evaporation. The solid residue on the frit was
redissolved in CH2Cl2 and transferred to a small vial, where the
solvent was removed by N2 flush. Approximately 3 mL of
MeOH was added to the solid residue; the suspension was
vigorously shaken by sonication and then stored at -20 ꢀC
overnight. Finally, the solid (fraction 2B) was collected by
3
3
-132.1 (brd, JFF = 20.7, o-F), -164.6 (t, JFF = 21.0, p-F),
-167.05 (m, m-F).
Synthesis of Complex 8. A glass jar was charged with 5.2 mmol
of ligand dissolved in 30 mL of toluene. To this solution was
added 5.40 mmol of n-BuLi (2.5 M solution in hexanes, Acros)
by syringe. This solution was stirred for 30 min, and the toluene
was removed using a vacuum system attached to the drybox.
Hexane was added, removed by vacuum, and added again, and
the resulting slurry was filtered to give a tan solid. A glass jar was
then charged with the tan solid dissolved in 30 mL of toluene. To
this solution was added 5.0 mmol of solid HfCl4. The jar was
capped with an air-cooled reflux condenser, and the mixture was
heated at reflux for about 4 h. After cooling, 18.1 mmol of
BuMgCl (3.5 equiv, 2.0 M solution in diethyl ether, Aldrich
Chemical Co.) was added by syringe, and the resulting mixture
was stirred overnight at ambient temperature. Solvent (toluene
and diethyl ether) was removed from the reaction mixture by
vacuum. Hexane (30 mL) was added to the residue, the mixture
was filtered, and the residue (magnesium salts) was washed with

5458 Organometallics, Vol. 28, No. 18, 2009
Zuccaccia et al.
centrifugation, washed with 1 mL of MeOH, and dried over-
night under vacuum (12 mg was obtained). Workup of fraction
3: approximately 150 mL of MeOH (to which 3 drops of 37%
aqueous HCl was added) was poured in the flask containing the
solid residue. The resulting suspension was filtered. The mother
liquor (fraction 3A) was stored in open air to allow solvent
evaporation. The solid residue on the frit was redissolved in
CH2Cl2 and transferred to a small vial, where the solvent was
removed by N2 flush. Approximately 3 mL of MeOH was added
to the solid residue; the suspension was vigorously shaken by
sonication and then stored at -20 ꢀC overnight. Finally, the
solid (fraction 3B) was collected by centrifugation, washed with
1 mL of MeOH, and dried overnight under vacuum (60 mg was
obtained). According to subsequent characterization, fractions
2B and 3B turned out to poly-1-butene samples with identical
characteristics. The theoretical yield in polymers (or oligomers)
ranges from 171 (virtually 0% active sites) to 342 mg (virtually
100% active sites). 13C NMR spectrum and GPC data are
reported in the Supporting Information, Figures 39S and 40S,
respectively.
Acknowledgment. We thank Roger Kuhlman, Timothy
Wenzel, James Stevens, Daryoosh Beigzadeh, and
Harold Boone for helpful discussions and the Ministero
ꢁ
dell’Universita e della Ricerca (MUR, Rome, Italy;
PRIN 2007) and The Dow Chemical Company for
support.
Supporting Information Available: Additional NMR data and
computational details. This material is available free of charge
via the Internet at http://pubs.acs.org.