Measurement of the barrier to β-hydride elimination in a β-agostic palladium-ethyl complex: A model for the energetics of chain-walking in (α-diimine)PdR+ olefin polymerization catalysts

Article abstract of DOI:10.1021/om010197j

The cationic Pd(II) ethyl complex [[(2,6-(ⁱPr)₂C₆H₃)N=C(An)C(An)=N(2,6 -(ⁱPr)₂C₆H₃)]Pd (CH₂CH₂-μ-H)]BAr′₄ (8), which exhibits a β-agostic interaction, has been synthesized via protonation of the corresponding diethyl complex with H(OiPr₂)₂BAr′₄ (Ar′ = 3,5-(CF₃)₂C₆H₃). The fluxional behavior of this ethyl cation has been studied quantitatively using ¹H and ¹³C NMR spectroscopy. Two dynamic processes have been observed: (1) β-H elimination and reinsertion with opposite regiochemistry and (2) rotation about the C-C bond of the ethyl group. β-H elimination is the lower energy process, with ΔG? = 7.1 kcal/mol at -108°C. Eyring analysis of the ¹³C NMR data indicates ΔH? = 6.1 ± 0.2 kcal/mol and ΔS? = -5.2 ± 0.9 eu. Rotation of the agostic methyl group occurs with ΔG? ca. 8.4 kcal/mol at -108°C. These data provide insight into the energetics of chain walking, which is responsible for the unusual polyolefin microstructures produced using (α-diimine)PdR⁺ catalysts.

Full text of DOI:10.1021/om010197j

Organometallics 2001, 20, 3975-3982
3975
Mea su r em en t of th e Ba r r ier to â-Hyd r id e Elim in a tion in
a â-Agostic P a lla d iu m -Eth yl Com p lex: A Mod el for th e
En er getics of Ch a in -Wa lk in g in (r-Diim in e)P d R+ Olefin
P olym er iza tion Ca ta lysts
Leigh Huff Shultz and Maurice Brookhart*
Department of Chemistry, University of North Carolina at Chapel Hill, CB3290 Venable Hall,
Chapel Hill, North Carolina 27599-3290
Received March 12, 2001
The cationic Pd(II) ethyl complex [[(2,6-(iPr)2C6H3)NdC(An)C(An)dN(2,6-(iPr)2C6H3)]Pd(CH2-
CH2-µ-H)]BAr′4 (8), which exhibits a â-agostic interaction, has been synthesized via
protonation of the corresponding diethyl complex with H(OiPr2)2BAr′4 (Ar′ ) 3,5-(CF3)2C6H3).
The fluxional behavior of this ethyl cation has been studied quantitatively using 1H and 13C
NMR spectroscopy. Two dynamic processes have been observed: (1) â-H elimination and
reinsertion with opposite regiochemistry and (2) rotation about the C-C bond of the ethyl
group. â-H elimination is the lower energy process, with ∆Gq ) 7.1 kcal/mol at -108 °C.
Eyring analysis of the 13C NMR data indicates ∆Hq ) 6.1 ( 0.2 kcal/mol and ∆Sq ) -5.2 (
0.9 eu. Rotation of the agostic methyl group occurs with ∆Gq ca. 8.4 kcal/mol at -108 °C.
These data provide insight into the energetics of chain walking, which is responsible for the
unusual polyolefin microstructures produced using (R-diimine)PdR+ catalysts.
In tr od u ction
There has been considerable recent interest in the use
of late transition metal complexes for olefin polymeriza-
tion.1-5 In 1995, we described a unique family of
catalysts based on cationic Ni(II) and Pd(II) species
bearing bulky aryl-substituted R-diimine ligands which
are capable of polymerizing ethylene, R-olefins, and
F igu r e 1. Cationic Ni and Pd R-diimine catalysts for olefin
polymerization.
internal and cyclic olefins to high molecular weight
polymers (Figure 1).6
The polyethylenes produced by these systems are
distinguished from those made with early metal d0
catalysts7 by their structure, having branches along the
main polymer backbone and possessing branches on
these branches (a “hyperbranched” structure).8-10
Previous studies of the Pd and Ni catalysts have
resulted in the mechanism for propagation and isomer-
ization shown in Scheme 1.11,12
The catalyst resting state, observed by 1H NMR
spectroscopy, is an alkyl olefin complex (1). A key
feature of this propagation mechanism which results in
formation of branched polymer is rapid migration of the
metal center along the chain in the â-agostic alkyl
complexes (e.g., 2, 4) prior to insertion. Bulky, axially
oriented substituents on the imine aryl rings retard
chain transfer and chain termination reactions. This
chain-walking process13 occurs via a series of â-hydride
elimination/readdition reactions (a common feature of
d8 metal alkyls14) coupled with rotation around the
Pd-CR bond, which interchanges â-agostic hydrogens
on geminal Câ carbons in secondary alkyl complexes.
We have previously reported the synthesis, charac-
terization, and dynamic behavior of the isopropyl agostic
cation [[(2,6-(iPr)2C6H3)NdC(An)C(An)dN(2,6-(iPr)2C6H3)]-
Pd(CH(CH3)CH2-µ-H)]BAr′4 (6) (Ar′ ) 3,5-(CF3)2C6H3),
shown in eq 1, which can be isolated from the insertion
(1) Ittel, S. D.; J ohnson, L. K.; Brookhart, M. Chem. Rev. 2000, 100,
1169-1203.
(2) Britovsek, G. J . P.; Gibson, V. C.; Wass, D. F. Angew. Chem.,
Int. Ed. 1999, 38, 429-447.
(3) Keim, W.; Appel, R.; Storeck, A.; Kruger, C.; Goddard, R. Angew.
Chem., Int. Ed. Engl. 1981, 20, 116-117.
(4) Keim, W. Angew. Chem., Int. Ed. Engl. 1990, 29, 235.
(5) Klabunde, U.; Ittel, S. D. J . Mol. Catal. 1987, 41, 123-134.
(6) J ohnson, L. K.; Killian, C. M.; Brookhart, M. J . Am. Chem. Soc.
1995, 117, 6414-6415.
(7) Brintzinger, H. H.; Fischer, D.; Mulhaupt, R.; Rieger, B.;
Waymouth, R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143-1170.
(8) Cotts, P. M.; Guan, Z.; McCord, E.; McLain, S. Macromolecules
2000, 33, 6945-6952.
(9) Guan, Z. B.; Cotts, P. M.; McCord, E. F.; McLain, S. J . Science
1999, 283, 2059-2062.
(10) McLain, S. J .; McCord, E. F.; J ohnson, L. K.; Ittel, S. D.; Nelson,
L. T. J .; Arthur, S. D.; Halfhill, M. J .; Teasley, M. F.; Tempel, D. J .;
Killian, C.; Brookhart, M. Polym. Prepr. (Am. Chem. Soc., Div. Polym.
Chem.) 1997, 38 (1), 772-773.
(11) Svejda, S. A.; J ohnson, L. K.; Brookhart, M. J . Am. Chem. Soc.
1999, 121, 10634-10635.
(12) Tempel, D. J .; J ohnson, L. K.; Huff, R. L.; White, P. S.;
Brookhart, M. J . Am. Chem. Soc. 2000, 122, 6686-6700.
(13) For an early example of chain-walking in an R-olefin oligomer-
ization process, see: (a) Mo¨hring, V. M.; Fink, G. Angew. Chem., Int.
Ed. Engl. 1985, 24, 1001-1003 (b) Schubbe, R.; Angermund, K.; Fink,
G.; Goddard, R. Makromol. Chem. Phys. 1995, 196, 467-468.
(14) Rix, F. C.; Brookhart, M. J . Am. Chem. Soc. 1995, 117, 1137-
1138.
10.1021/om010197j CCC: $20.00 © 2001 American Chemical Society
Publication on Web 08/18/2001

3976 Organometallics, Vol. 20, No. 19, 2001
Sch em e 1. P r op osed Mech a n ism for P olym er iza tion of Eth ylen e
Shultz and Brookhart
of a single equivalent of ethylene into the Pd-methyl
bond of the catalyst precursor shown in Figure 1.15
Several cationic agostic ethyl species have been de-
scibed in the literature for group 9 and 10 transition
metals. Spencer16,17 and Brookhart18,19 have reported
cations of the formula [(C5R5)Co(CH2CH2-µ-H)L]BF4 (R
) H, Me; L ) C2H4, PMe3, P(OMe)3, P(o-tolyl)3, PMe2-
Ph) (Figure 2).
These species undergo three dynamic processes: CR-
Câ bond rotation, â-H elimination, and inversion at the
chiral Co center. The cation [(C5Me5)Co(CH2CH2-µ-H)-
(PMe3)]BF4 is the only species in the series for which
the barrier to inversion at Co can be measured (13.4
kcal/mol).20 The barrier to C-C bond rotation is lower,
at 10.9 kcal/mol, while â-H elimination requires >15.6
kcal/mol. The other cations in the series have similar
barriers, with C-C bond rotation always being a lower
energy process than â-H elimination. Significantly,
these species are active catalysts for the production of
linear polyethylene, with the catalyst resting state being
the agostic alkyl complex.21
1
In the static H NMR spectrum (-115 °C), the agostic
proton resonance appears as a triplet (2J HH ) 17 Hz) at
-8.00 ppm. Above -10 °C, a broad resonance is ob-
served at -0.84 ppm, accounting for the six methyl
protons. In solution, the isopropyl agostic cation 6 exists
in equilibrium with the n-propyl agostic isomer 7, with
the isopropyl species being favored by 20:1. Measure-
ment of the barrier to interconversion between the two
species, which presumably occurs through a propene
hydride complex, is complicated not only by the lopsided
equilibrium but also by bond rotations in the isopropyl
cation: CR-Câ(agostic) bond rotation occurs with a barrier
of 9.2 kcal/mol, while Pd-CR bond rotation has a barrier
of 9.6 kcal/mol. Because of these complications, we were
only able to estimate an upper limit for the barrier to
â-hydride elimination of 10.7 kcal/mol.12 Significantly,
no inversion at the Pd center is observed between -115
and 0 °C for these species.
The complications described above that prevent ac-
curate measurement of the agostic alkyl isomerization
process involving the olefin hydride can be circumvented
by the synthesis of an agostic ethyl cation; â-H elimina-
tion in this species is a degenerate process occurring
via a symmetrical ethylene hydride complex, and thus
isotopic labeling could be used to quantify the energetics
involved (eq 2).
Spencer has also published a series of agostic ethyl
complexes of Ni, Pd, and Pt having chelating bisphos-
(16) Cracknell, R. B.; Orpen, A. G.; Spencer, J . L. J . Chem. Soc.,
Chem. Commun. 1984, 326.
(17) Cracknell, R. B.; Orpen, A. G.; Spencer, J . L. J . Chem. Soc.,
Chem. Commun. 1986, 1005.
(18) Brookhart, M.; Green, M. L. H.; Pardy, R. B. A. J . Chem. Soc.,
Chem. Commun. 1983, 691.
(19) Schmidt, G. F.; Brookhart, M. J . Am. Chem. Soc. 1985, 107,
1443-1444.
(20) Brookhart, M.; Lincoln, D. M.; Volpe, A. F., J r.; Schmidt, G. F.
Organometallics 1989, 8, 1212-1218.
(15) Tempel, D. J .; Brookhart, M. Organometallics 1998, 17, 2290-
(21) Brookhart, M.; DeSimone, J . M.; Grant, B. E.; Tanner, M. J .
Macromolecules 1995, 28, 5378-5380.
2296.

Chain-Walking in (R-Diimine)PdR+ Olefin Catalysts
Organometallics, Vol. 20, No. 19, 2001 3977
of EtMgCl at -78 °C in diethyl ether affords the diethyl
complexes (10a ,b) in moderate yields (eq 3).
F igu r e 2. Cobalt(III) agostic ethyl cations reported by
Spencer and Brookhart.
F igu r e 3. Bisphosphine group 10 ethyl agostic cations
reported by Spencer.
Removal of the magnesium salts is accomplished by
filtration of the cold reaction mixture through Florisil,
which also serves to remove any decomposition products
from solution. The diethyl complexes are surprisingly
stable for late transition metal alkyl complexes possess-
ing â-hydrogens. They survive in the solid state for days
at ambient temperature and give satisfactory combus-
tion analyses. They are moderately stable (ca. 1 day) in
solution in the absence of oxygen, but eventually
decompose to ethylene, ethane, and unidentified pal-
ladium species.
phine ligands; these species are synthesized by proto-
nation of the corresponding M(II) diethyl species or M(0)
ethylene complexes.22-25 The complexes [(tBu2P(CH2)3Pt-
Bu2)M(CH2CH2-µ-H)]BF4 (M ) Ni, Pd, Pt) (Figure 3)
1
have been characterized by H and 13C NMR spectros-
copy; in the Pt case, the agostic proton resonates at -2.6
1
ppm (-110 °C) with a J CH of 60 Hz.22,23
All five hydrogens and both carbons in the Pt-ethyl
group are equivalent above -30 °C; an overall upper
limit for â-H elimination and C-C bond rotation pro-
cesses of 8 ( 1.5 kcal/mol can be estimated from
coalescence data at -77 °C. The Ni and Pd cations are
The 1H NMR spectrum of the diethyl complexes is
straightforward; the presence of only one type of ligand
methyl group (in the case of 10a ) or one kind of isopropyl
methine hydrogen (in the case of 10b) is indicative of a
symmetric dialkyl complex. The methyl and methylene
groups of the ethyl ligands appear as a multiplet at 0.92
ppm; the Pd center shields the methylene protons,
causing them to be nearly coincident with the methyl
groups. The observed C-H coupling constants for the
ethyl group are 127 Hz (CH2) and 129 Hz (CH3), in the
normal range for sp3 carbon centers.28
F or m a tion of th e â-Agostic Eth yl Com p lex. When
the diethyl complexes 10a and 10b are treated with
H(OEt2)2BAr′4 (Ar′ ) 3,5-(CF3)2C6H3)29 in CD2Cl2 at low
temperature (-78 °C), ethane is formed, and the solu-
tion turns from deep brown to a light, clear orange color,
indicative of protonolysis to form a cationic Pd(II) ethyl
species. 1H NMR spectra of the products at -78 °C show
that they are cationic ethyl ether complexes (11a ,b)
rather than ethyl agostic species (eq 4).
1
reported, although the static H NMR spectra are not
resolved.
We report here the synthesis and characterization of
the ethyl agostic cation [[(2,6-(iPr)2C6H3)NdC(An)C-
(An)dN(2,6-(iPr)2C6H3)]Pd(CH2CH2-µ-H)]BAr′4 (8), which
serves as a model for intermediates involved in the
cationic (R-diimine)Pd(II)-catalyzed polymerization of
ethylene. Observation of the dynamic processes exhib-
ited by this complex yields the barrier to â-H elimina-
tion, one of the elementary steps involved in the
migration of the Pd center along the polymer chain
during the polymerization process.
Resu lts a n d Discu ssion
Syn th esis of P r ecu r sor Dieth yl Com p lexes 10a
a n d 10b. The R-diimine Pd diethyl complexes can be
synthesized in a straightforward manner from the
analogous dichloride complexes, which were known in
the literature but had not been fully characterized.26,27
Treatment of the dichloride complexes 9a ,b with 2 equiv
Treatment of complexes 10a and 10b with H(OiPr2)2-
BAr′4 at -78 °C in CD2Cl2 appears to give similar
results; initially, both solutions become clear orange and
a gas is evolved. However, species in solutions produced
from 10a begin to decompose immediately, precipitating
(22) Conroy-Lewis, F. M.; Mole, L.; Redhouse, A. D.; Litster, S. A.;
Spencer, J . L. J . Chem. Soc., Chem. Commun. 1991, 1601-1603.
(23) Mole, L.; Spencer, J . L.; Carr, N.; Orpen, A. G. Organometallics
1991, 10, 49-52.
(24) Spencer, J . L.; Mhinzi, G. S. J . Chem. Soc., Dalton Trans. 1995,
3819-3824.
(25) Carr, N.; Mole, L.; Orpen, A. G.; Spencer, J . L. J . Chem. Soc.,
Dalton Trans. 1992, 2653-2662.
(26) van Asselt, R.; Elsevier, C. J .; Smeets, W. J . J .; Spek, A. L.;
Benedix, R. Recl. Trav. Chim. Pays-Bas 1994, 113, 88-98.
(27) van Asselt, R.; Elsevier, C. J . Organometallics 1994, 13, 1972-
1980.
1
Pd(0) and giving products that are unidentifiable by H
NMR spectroscopy. Protonation of 10b with H(OiPr2)2-
(28) Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. Spectrometric
Identification of Organic Compounds, 5th ed.; Wiley: New York, 1991;
p 247.
(29) Brookhart, M.; Grant, B.; Volpe J r., A. F. Organometallics 1992,
11, 3920-3922.

3978 Organometallics, Vol. 20, No. 19, 2001
Shultz and Brookhart
agostic proton with the two geminal protons on the
â-carbon while the R-methylene protons remain unaf-
fected.
1
Observation of the H NMR spectroscopic behavior of
the ethyl moiety as the probe temperature is raised from
-130 to 0 °C allows relative rate measurements for
these processes. At -100 °C, the methylene resonance
originally at 2.2 ppm has coalesced with the other
methylene resonance (originally at 1.4 ppm) to a broad
peak at 1.8 ppm; the resonance for the agostic proton
remains distinct. This observation indicates that â-hy-
dride elimination and reinsertion is the faster of the two
fluxional processes. Using coalescence methods, a first-
order rate constant k ) 1448 s-1 is obtained for â-H
elimination at Tc ) -108 °C (∆Gq ) 7.1 ( 0.4 kcal/mol).
From the line broadening of the agostic resonance at
this temperature (-108 °C, ∆ω ) 12.5 Hz), the C-C
bond rotation barrier is estimated to be ca. 8.4 kcal/mol
(k ) 40 s-1). In the fast exchange limit at 0 °C, one broad
resonance is observed at -0.4 ppm, accounting for all
five ethyl hydrogens. This is consistent with the
superposition of both dynamic processes illustrated in
Scheme 2.
BAr′4 gives the desired product, [[(2,6-(iPr)2C6H3)Nd
C(An)C(An)dN(2,6-(iPr)2C6H3)]Pd(CH2CH2-µ-H)]BAr′4
(8) (eq 5).
Both of the dynamic processes of the â-agostic ethyl
cation result in 1H site exchange and are therefore
visible in the variable-temperature 1H NMR spectra.
However, the two processes can more easily be distin-
guished using 13C NMR; â-hydride elimination and
reinsertion with opposite regiochemistry equilibrate CR
and Câ, while C-C bond rotation results in no site
exchange for CR and Câ. Therefore, 13C-labeling was
used to quantify â-hydride elimination and reinsertion
within a range of temperatures and without the com-
plication of C-C bond rotation.
Syn th esis of 13C-La beled â-Agostic Eth yl Com -
p lex 8-13C. The 13C-labeled diethyl complex (10b-13C)
was synthesized in the same manner as the unlabeled
complex, using CH313CH2MgBr. Protonation of 10b-13C
with H(OiPr2)2BAr′4 in CDCl2F at -80 °C results in
quantitative loss of ethane-13C and formation of [[(2,6-
(iPr)2C6H3)NdC(An)C(An)dN(2,6-(iPr)2C6H3)]Pd(13CH2-
CH2-µ-H)]BAr′4 (8-13C). Although only the R-methylene
carbon is initially labeled with 13C, very rapid â-hydride
elimination and reinsertion at -80 °C results in
equal distribution of the 13C label between the R- and
â-carbons by the time an NMR spectrum is acquired
(Scheme 3).
Sta tic Sp ectr u m of th e Eth yl Agostic Com p lex
8. At -130 °C in CDCl2F, the agostic hydrogen appears
as a broad (ca. 60 Hz at half-height) triplet (2J HH ) 16
Hz) at -8.9 ppm, due to coupling to the geminal protons
1
on the â-carbon. The J CH for the agostic proton,
indicative of donation of the C-H σ bond density to the
metal center, was determined to be 67 Hz,30 which is a
clear indication of an agostic Pd- -H- -C three-center,
two-electron interaction.31 The geminal protons on Câ
appear as a broad peak at 1.4 ppm, while the methylene
protons on the R-carbon appear at 2.2 ppm. Because the
complex possesses a plane of symmetry (corresponding
to the square plane of the complex), only two methine
resonances are observed for the diisopropyl ligand, at
3.24 and 3.14 ppm.
Dyn a m ic P r ocesses Obser ved in 8. Two distinct
dynamic processes occur that serve to equilibrate the
five ethyl protons of 8. The first, â-hydride elimination
and reinsertion with opposite regiochemistry (Scheme
2(i)), interchanges the two sets of methylene protons,
keeping the agostic proton (Ha) distinct. The second,
C-C bond rotation (Scheme 2(ii)), interchanges the
The R-carbon appears at 38.5 ppm in the 13C spec-
1
trum, having a J CH of 153 Hz; the â-carbon resonance
1
at 19.3 ppm exhibits a J CH of 155 Hz with the
1
nonagostic methylene protons and a J CH of 67 Hz with
the agostic proton. The large C-H coupling constants
between CR and Câ and the methylene protons are in
the range for sp2 C-H coupling constants, indicating a
significant amount of ethylene hydride character in this
â-agostic ethyl species.28
Va r ia ble-Tem per a tu r e 13C NMR Spectr a . As men-
tioned earlier, â-hydride elimination is the only one of
the two dynamic processes that is observable by 13C
NMR; this process serves to interchange CR and Câ.
Thus, in the slow exchange limit (ca. -120 °C), two
broad resonances are observed at 38.5 ppm (8-13CR) and
19.3 ppm (8-13Câ) in the proton-decoupled 13C NMR
spectrum, corresponding to the two labeled complexes
1
(30) This J CH was measured using the 13C-labeled species 8-13C.
(31) Brookhart, M.; Green, M. L. H.; Wong, L.-L. Prog. Inorg. Chem.
1988, 36, 1-124.

Chain-Walking in (R-Diimine)PdR+ Olefin Catalysts
Organometallics, Vol. 20, No. 19, 2001 3979
Sch em e 2. Dyn a m ic P r ocesses of th e â-Agostic Eth yl Ca tion 8
F igu r e 5. Free energy (∆G) diagram (kcal/mol) for isomer-
ization of ethyl ethylene complexes (-55 °C).
F igu r e 4. Eyring plot for â-hydride elimination in 8-13C.
necessary for â-hydride elimination.32 On the basis of
these results and the observation from previous studies
that the barrier to trapping with ethylene is higher than
the barrier to â-H elimination, the free energy diagram
shown in Figure 5 can be constructed for isomerization
of ethyl ethylene complexes, which involves formation
of the agostic ethyl cation. Using gradient-corrected
DFT, Ziegler et al. have calculated the enthalpic barrier
to â-hydride elimination in the analogous generic (di-
imine ) -NHCHCHNH-) Pd system to be 5.84 kcal/
mol for an agostic butyl system, in good agreement with
our measured barrier (∆Hq) of 6.1 kcal/mol.33 The
agreement between DFT calculations using a generic
diimine system (electronic effects only are considered)
and the experimental barrier suggests that steric factors
do not play a large role in determining the ground state
or transition state energies for these cationic ethyl
complexes.
Sch em e 3. P r oton a tion of 10b-13C a n d Su bsequ en t
Equ ilibr a tion of 8-13C
Su m m a r y a n d Con clu sion s
A (diimine)Pd(II) â-agostic ethyl cation (8) has been
synthesized via protonation of a (diimine)Pd(II) diethyl
complex with H(OiPr2)2BAr′4. The cation exhibits dy-
in equilibrium (Keq ) 1 for degenerate species). As the
probe temperature is raised, the two resonances broaden
and disappear into the baseline, reappearing as a single,
averaged peak (28.4 ppm) at warmer temperatures.
Using the slow exchange approximation at -120 °C, a
first-order rate constant of k ) 590 s-1 for exchange was
determined, corresponding to a barrier (∆Gq) to â-hy-
dride elimination of 6.8 kcal/mol. In the fast exchange
region (-55 °C), a first-order rate constant of k ) 2.9 ×
105 s-1 can be estimated (∆Gq ) 7.2 kcal/mol). An Eyring
analysis of the first-order rate data between -40 and
-125 °C (Figure 4) gives ∆Hq ) 6.1 ( 0.2 kcal/mol and
∆Sq ) -5.2 ( 0.9 eu.
1
namic H and 13C NMR spectra which can be assigned
to two dynamic processes: (1) â-H elimination (∆Gq )
6.9 kcal/mol, -108 °C) and (2) CR-Câ bond rotation (∆Gq
ca. 8.4 kcal/mol, -108 °C). The energy ordering of these
two processes distinguishes this agostic ethyl cation
from those previously described in the literature, in
which bond rotations were observed to be faster than
â-H elimination.
The â-H elimination barrier determined by these
experiments is one of two important barriers in the
(32) Collman, J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry, 2nd
ed.; University Science Books: Mill Valley, CA, 1987.
The modest negative value of the entropy of activation
is most likely due to the ordered, planar transition state
(33) Michalak, A.; Ziegler, T. Organometallics 1999, 18, 3998-4004.

3980 Organometallics, Vol. 20, No. 19, 2001
Shultz and Brookhart
Sch em e 4. Elem en ta r y P r ocesses In volved in Ch a in -Wa lk in g
Hz. Elemental analyses were performed by Atlantic Microlab,
Inc. of Norcross, GA.
chain-walking process, the other being Pd-C bond
rotation. Although Pd-C bond rotation does not produce
site exchange in the ethyl cation, precluding measure-
ment of its energetic cost, studies of the â-agostic
isopropyl cation 6 (see eq 1) do allow measurement of
this bond rotation barrier, ∆Gq ) 9.6 kcal/mol. The
chain-walking process, involving both Pd-C rotation
and â-H elimination, is illustrated in Scheme 4 for a
Pd-C5 alkyl moiety.
Migration of the Pd atom from C3 to C2 involves
initial release of the agostic interaction in 12, rotation
about the Pd-C3 bond, and formation of an agostic bond
with a C-H bond of C2 (13). Then â-H elimination/
readdition can occur, placing the Pd atom on C2, with
C3 in the â-agostic postion (14). Migration to C1 (16)
again involves Pd-C2 bond rotation (14 f 15), followed
by â-H elimination. Thus, the combination of these two
processes, modeled with the ethyl cation (8) and the
isopropyl cation (6), presents a detailed picture of the
energetics of migration of Pd along the chain in the (R-
diimine)Pd cations and suggests that the Pd-CR bond
rotation is the higher barrier, controlling the overall rate
of chain migration.
Rate constants were determined by analysis of NMR line
shapes using standard equations. At coalescence (where two
resonances have merged to a single “flat-topped” peak), k )
π(νa - νx)/21/2, where νa - νx is the difference in resonance
frequencies determined in the slow exchange limit. In the slow
exchange region, k ) π∆ω, where ∆ω is the line-width at half-
height at a given temperature minus the line-width at half-
height in the slow exchange limit. In the fast exchange region,
k ) π(νA - νX)2/2∆ω, where ∆ω is the line-width at half-height
at a given temperature minus the line-width at half-height in
the fast exchange limit. Activation parameters (∆Hq, ∆Sq, and
∆Gq) were calculated from measured rate constants and
temperatures using the Eyring equation. Error analyses for
∆Hq and ∆Sq were based on Girolami’s derivation of σ∆Hq and
σ∆Sq.34 Error analysis of ∆Gq obtained for dynamic processes
was based on Binsch’s derivation of σ∆Gq and incorporated
an estimate of 10% error in k and (1 °C error in temperature
((5 °C error for coalescence data).35
Ma ter ia ls. All solvents used for synthesis were deoxygen-
ated and dried via passage over
a column of activated
alumina.36 Dichlorofluoromethane-d (CDCl2F) was prepared
according to the literature,37 dried over CaCl2, vacuum-
transferred, degassed by repeated freeze-pump-thaw cycles,
and stored over 4 Å molecular sieves at -30 °C under argon.
Methylene chloride-d2 was dried over P2O5, vacuum-trans-
ferred, degassed by repeated freeze-pump-thaw cycles, and
stored over 4 Å molecular sieves under argon. The R-diimine
ligands,38 (PhCN)2PdCl2,39 and H(OEt2)2BAr′429 were prepared
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All manipulations of compounds
were performed using standard high-vacuum or Schlenk
techniques. Argon was purified by passage through columns
of BASF R3-11 catalyst (Chemalog) and 4 Å molecular sieves.
Solid organometallic compounds were transferred in an argon-
filled MBraun drybox. NMR spectra were acquired with
Bruker DRX400 or DRX500 spectrometers. NMR probe tem-
peratures were measured using an Omega model HH21
thermocouple (type T) immersed in anhydrous methanol in a
(34) Morse, P. M.; Spencer, M. D.; Wilson, S. R.; Girolami, G. S.
Organometallics 1994, 13, 1646.
(35) Binsch, G. In Dynamic Nuclear Magnetic Resonance Spectros-
copy; J ackman, L., Cotton, F. A., Eds.; Academic Press: New York,
1975; p 77.
(36) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J . Organometallics 1996, 15, 1518-1520.
(37) Siegel, J . S.; Anet, F. A. L. J . Org. Chem. 1988, 53, 2629-2630.
(38) van Koten, G.; Vrieze, K. Adv. Organomet. Chem. 1982, 21,
151-239.
(39) Kharasch, M. S.; Seyler, R. C.; Mayo, F. R. J . Am. Chem. Soc.
1938, 60, 882-884.
1
5 mm NMR tube. H and 13C chemical shifts are reported in
ppm downfield of TMS and were referenced to residual 1H
NMR signals and to the 13C NMR signals of the deuterated
solvents, respectively. All coupling constants are reported in

Chain-Walking in (R-Diimine)PdR+ Olefin Catalysts
Organometallics, Vol. 20, No. 19, 2001 3981
according to the literature. NaBAr′4 (Ar′ ) 3,5-(CF3)2C6H3) was
purchased from Boulder Scientific and used as received.
EtMgCl was purchased from Aldrich and stored at -30 °C.
CH313CH2Br was purchased from Aldrich, stored at -30 °C,
and used as received. CH313CH2MgBr was synthesized via
standard procedures40 and used as a solution in Et2O; the
molarity of this solution was determined using the method
reported by Love et al.41 The syntheses of the dichloride
complexes (9a and 9b) have been reported without spectral
characterization.42
126.2, 123.3, 18.0 (Ar(CH3)2), 15.8 (PdCH2CH3), 11.5 (PdCH2-
CH3). Anal. Calcd for C32H34N2Pd: C, 69.50; H, 6.20; N, 5.07.
Found: C, 69.65, H, 6.31, N, 5.00.
[(2,6-(i P r )2C6H 3)NdC(An )C(An )dN(2,6-(i P r )2C6H 3)]-
P d (C2H 5)2 (10b ). A procedure identical to that for 2a
was followed, using ((2,6-(iPr)2C6H3)NdC(An)C(An)dN(2,6-(i-
Pr)2C6H3)PdCl2 (9b) (0.326 g, 0.48 mmol) and C2H5MgCl (2.0
M, 0.48 mL, 0.962 mmol). Yield: 0.078 g (24%). 1H NMR (CD2-
Cl2, 400 MHz, 25 °C): δ 8.02 (d, J ) 8.0, 2H, An pH), 7.39 (m,
8H, An mH, ArH), 6.58 (d, J ) 7.0, 2H, An oH), 3.37 (sep, J )
6.8 Hz, 4H, CHMeMe′, C′HMeMe′), 1.38 (d, J ) 6.8, 12H,
CHMeMe′, C′HMeMe′), 0.92 (m, 10H, PdCH2CH3), 0.90 (d, J
) 6.8, 12H, CHMeMe′, C′HMeMe′). 13C NMR (CD2Cl2, 125
MHz, 25 °C): δ 167.8, 143.8, 142.8, 138.8, 131.8, 130.2, 128.9,
128.7, 126.9, 124.3, 124.2, 29.0 (CHMe2), 24.0 (CHMeMe′), 23.4
(CHMeMe′), 16.4 (PdCH2CH3), 12.7 (PdCH2CH3). Anal. Calcd
for C40H50N2Pd: C, 72.22; H, 7.58; N, 4.21. Found: C, 72.44,
H, 7.61, N, 4.18.
[(2,6-(CH 3)2C6H 3)NdC(An )C(An )dN(2,6-(CH 3)2C6H 3)]-
P d Cl2 (9a ). A clean, flame-dried Schlenk flask was charged
with (PhCN)2PdCl2 (2.00 g, 5.21 mmol) and (2,6-(CH3)2C6H3)Nd
C(An)C(An)dN(2,6-(CH3)2C6H3) (2.23 g, 5.74 mmol) in an
argon-filled drybox. The flask was placed under Ar, and 20
mL of CH2Cl2 were added via syringe. The homogeneous
solution was stirred for 20 h. Pentane (30 mL) was added to
cause precipitation of the product, which was isolated by
cannula filtration and washed with pentane (3 × 10 mL). The
orange-red solid was dried under reduced pressure at 25 °C
for 19 h and stored under argon at room temperature. Yield:
[(2,6-(iP r )2C6H 3)NdC(An )C(An )dN(2,6-(iP r )2C6H 3)]P d -
(
13CH2CH3)2 (10b-13C). A procedure identical to that for 10b
was followed, using ((2,6-(iPr)2C6H3)NdC(An)C(An)dN(2,6-
(iPr)2C6H3)PdCl2 (9b) (0.60 g, 0.89 mmol) and CH313CH2MgBr
(0.45 M, 3.9 mL, 1.8 mmol). Yield: 0.253 g (43%). 1H NMR
(CD2Cl2, 400 MHz, 25 °C): δ 8.02 (d, J ) 8.0, 2H, An pH),
7.39 (m, 8H, An mH, ArH), 6.58 (d, J ) 7.0, 2H, An oH), 3.37
1
2.50 g (85%). H NMR (CD2Cl2, 400 MHz, 25 °C): δ 8.19 (d, J
) 8.3, 2H, An pH), 7.53 (dd, J ) 7.4, 8.2, 2H, An mH), 7.39
(dd, J ) 7.0, 8.2, 2H, ArH), 7.30 (m, 4H, ArH), 6.64 (d, J )
7.3, 2H, An oH), 2.43 (s, 12H, Ar(CH3)2). 13C NMR (CD2Cl2,
100 MHz, 25 °C): δ 176.2, 147.9, 143.7, 133.3, 131.8, 130.1,
129.9, 129.1, 128.9, 125.7, 124.8, 18.5. Anal. Calcd for C28H24N2-
PdCl2: C, 59.43; H, 4.28; N, 4.95. Found: C, 59.36, H, 4.69,
N, 4.82.
[(2,6-(i P r )2C6H 3)NdC(An )C(An )dN(2,6-(i P r )2C6H 3)]-
P d Cl2 (9b). A procedure similar to that for 9a was followed,
using (PhCN)2PdCl2 (0.66 g, 1.72 mmol) dissolved in 10 mL of
CH2Cl2 and (2,6-(iPr)2C6H3)NdC(An)C(An)dN(2,6-(iPr)2C6H3)
(0.95 g, 1.89 mmol) dissolved in 90 mL of CH2Cl2. The
homogeneous ligand solution was added via cannula to the Pd
solution. Reaction times, conditions, and workup were identical
to that for 9a . Yield: 0.93 g (80%). 1H NMR (CD2Cl2, 400 MHz,
25 °C): δ 8.16 (d, J ) 8.3, 2H, An pH), 7.53 (m, 4H, An mH,
ArH), 7.40 (m, 4H, ArH), 6.53 (d, J ) 7.3, 2H, An oH), 3.49
(sep, J ) 6.8, 4H, CHMeMe′, C′HMeMe′), 1.51 (d, J ) 6.8, 12H,
CHMeMe′, C′HMeMe′), 0.99 (d, J ) 6.8, 12H, CHMeMe′,
C′HMeMe′). 13C NMR (CD2Cl2, 100 MHz, 25 °C): δ 176.4,
147.4, 141.3, 140.5, 133.1, 131.8, 129.6, 126.4, 125.2, 124.7,
29.9, 23.9, 23.7 (one C missing). Anal. Calcd for C36H40N2-
PdCl2: C, 63.76; H, 5.95; N, 4.13. Found: C, 63.76, H, 6.01,
N, 4.17.
(sep, 4H, J ) 6.8, CHMeMe′, C′HMeMe′), 1.38 (d, 12H, J )
3
6.8, CHMeMe′, C′HMeMe′), 0.93 (dq, 4H, J CH ) 127, J HH
)
7.2, Pd13CH2CH3),0.90 (d, 12H, J ) 6.8, CHMeMe′, C′HMeMe′),
0.87 (t, 6H, J ) 7.2, Pd13CH2CH3).
H(OiP r 2)2BAr ′4 (Ar ′ ) 3,5-(CF 3)2C6H3). iPr2O was distilled
from Na and stored in the absence of light under argon. A
clean, flame-dried Schlenk flask containing activated 4 Å
molecular sieves was charged with NaBAr′4 (2.5 g, 2.8 mmol)
i
in an argon-filled drybox. A 30 mL sample of dry Pr2O was
added, and the solution was allowed to stand overnight. It was
then transferred to a clean, flame-dried Schlenk flask via
cannula and was cooled to 0 °C. HCl(g), generated by the slow
addition of concentrated H2SO4 to NaCl(s), was bubbled
through the solution for 15 min. The mixture was then cooled
to -78 °C, and argon was purged through the solution for 1 h
to remove excess HCl(g). The solution was filtered through
Celite in a jacketed frit kept at -78 °C into a flame-dried
Schlenk flask. The solution volume was then reduced to 10-
15 mL under reduced pressure, and 6 mL of pentane was
added. The flask was sealed and placed in a -30 °C freezer
for 20 h, causing crystallization. The supernatant was removed
via cannula filtration, and the white crystals were washed with
3 mL of dry pentane. The solid was dried in vacuo at 25 °C for
15 min and stored at -30 °C in the drybox. Yield: 1.62 g (54%).
1H NMR (CD2Cl2, 400 MHz, -80 °C): δ 16.39 (quintet, 1H, J
) 2.4, H(OiPr2)2), 7.72 (s, 8H, BAr′4 oH), 7.54 (s, 4H, BAr′4
pH), 4.35 (doublet of septets, 4H, J ) 6.4, 2.4, H(O(CHMe2)2)2),
1.33 (d, 24H, J ) 6.4, H(O(CHMe2)2)2).
Gen er a l P r oced u r e for Va r ia ble-Tem p er a t u r e NMR
Sp ectr oscop ic Exp er im en ts. A tared 5 mm NMR tube was
charged with ca. 0.01 mmol of the desired diethyl complex (10a
or 10b) and 0.01 mmol of either H(OEt2)2BAr′4 or H(OiPr2)2-
BAr′4 in the drybox under Ar. The tube was capped with a
septum and removed from the drybox; the septum was secured
with Teflon tape and Parafilm. The tube was cooled to -78
°C (dry ice/acetone), and CDCl2F was added via a 22-gauge
cannula (∼600-800 µL) or CD2Cl2 was added via a gastight
syringe (700 µL). The tube was shaken and warmed slightly
to dissolve the solids and facilitate the protonation reaction;
gas evolution was observed, and the solution became clear
orange. The tube was then transferred to a precooled (-78 °C)
NMR probe for acquisition of spectra.
[(2,6-(CH3)2C6H3)NdC(An )C(An )dN(2,6-(CH3)2C6H3)]P d-
(C2H5)2 (10a ). A flame-dried Schlenk flask was charged with
((2,6-(CH3)2C6H3)NdC(An)C(An)dN(2,6-(CH3)2C6H3))PdCl2 (9a)
(0.771 g, 1.36 mmol) in the drybox. The flask was then cooled
to -78 °C in a dry ice/isopropyl alcohol bath. The orange solid
was suspended in 20 mL of diethyl ether, and 2 equiv of C2H5-
MgCl (2.0 M, 1.36 mL, 2.73 mmol) was added via syringe. The
mixture became brown within 5 min and was stirred at -78
°C for 2 h. Methanol (0.1 mL) was added via syringe to quench
any unreacted Grignard reagent, and the cold reaction mixture
was filtered through a 2-3 cm column of Florisil under argon
into a receiving flask cooled to 0 °C. The Florisil was extracted
with additional Et2O (3 × 10 mL), and the volume of the red-
brown filtrate was reduced in vacuo, producing a red-brown
solid. The product was dried briefly in vacuo at 25 °C. Yield:
1
0.193 g (26%). H NMR (CD2Cl2, 500 MHz, 25 °C): δ 8.05 (d,
J ) 8.0, 2H, An pH), 7.43 (dd, J ) 7.0, 8.0, 2H, An mH), 7.28
(m, 6H, ArH), 6.70 (d, 7.0, 2H, An oH), 2.28 (s, 12H, Ar(CH3)2),
0.86 (m, 10H, PdCH2CH3). 13C NMR (CD2Cl2, 125 MHz, 25
°C): δ 167.2, 145.7, 142.9, 130.2, 129.3, 128.7, 128.4, 128.2,
(40) Handbook of Grignard Reagents; Silverman, G. S., Ed.; Dek-
ker: New York, 1996.
(41) Love, B. E.; J ones, E. G. J . Org. Chem. 1999, 64, 3755-3756.
(42) van Asselt, R.; Elsevier: C. J .; Amatore, C.; J utand, A.
Organometallics 1997, 16, 317-328.
[[(2,6-(Me)2C6H3)NdC(An )C(An )dN(2,6-(Me)2C6H3)]P d -
(CH2CH 3)(OE t2)]BAr ′4 (11a ). 1H NMR (CD2Cl2, 400 MHz,
-80 °C): δ 8.10 (d, 1H, J ) 8.4, An pH), 8.06 (d, 1H, J ) 8.4,
An pH′), 7.72 (s, 8H, BAr′4 oH), 7.49 (s, 4H, BAr′4 pH), 7.45-

3982 Organometallics, Vol. 20, No. 19, 2001
Shultz and Brookhart
7.25 (m, 8H, An mH, ArH), 6.66 (d, 1H, J ) 7.2, An oH), 6.35
(d, 1H, J ) 7.2, An oH′), 3.70 (q, 4H, J ) 6.8, Pd(O(CH2CH3)2)),
2.22 and 2.18 (2 s, 6H each, ArMe2 and Ar′Me2), 1.46 (q, 2H,
J ) 7.2, PdCH2CH3), 1.45 (t, 6H, J ) 6.8, Pd(O(CH2CH3)2)),
-0.05 (t, 3H, J ) 7.2, PdCH2CH3).
H), 19.3 (1J CH ) 155, 67, PdCH2CH2-µ-H). 13C NMR (CDCl2F,
100 MHz, 0 °C, dynamic, selected peaks): δ 28.4 (PdCH2CH2-
µ-H).
Eyr in g An a lysis via 13C NMR Sp ectr oscop y. The barrier
to â-hydride elimination was determined using the 13C-labeled
ethyl agostic complex 8-13C. Samples were prepared as de-
scribed above and transferred to a precooled (-78 °C) NMR
probe for acquisition of spectra, which were acquired in the
range between -135 and 0 °C. Line widths (ω) were measured
in the 13C{1H} spectra in units of Hz for the two carbons in
the ethyl moiety and were corrected for line widths (ωo) in the
absence of exchange (∆ω ) ω - ωo). There is a 50% probability
of â-hydride elimination being followed by reinsertion with the
same regiochemistry (re-formation of starting materials rather
than formation of products), so the actual rate of â-hydride
elimination is twice the observed rate. To correct for this, all
measured rate constants were doubled. These experiments
were repeated twice with the same results; one of the data
sets is shown in Figure 4.
[[(2,6-(iP r )2C6H3)NdC(An )C(An )dN(2,6-(iP r )2C6H3)]P d -
(CH2CH3)(OE t2)]BAr ′4 (11b ). 1H NMR (CD2Cl2, 500 MHz,
-80 °C): δ 8.05 (d, 1H, J ) 8.5, An pH), 8.01 (d, 1H, J ) 8.5,
An pH′), 7.72 (s, 8H, BAr′4 oH), 7.52 (s, 4H, BAr′4 pH), 7.50-
7.25 (m, 8H, An mH, ArH), 6.51 (d, 1H, J ) 7.0, An oH), 6.26
(d, 1H, J ) 7.0, An oH′), 3.51 (br, 4H, Pd(O(CH2CH3)2)), 3.16
and 3.13 (2 septets, 6H each, J ) 6.5 and 6.5, ArCHMeMe
and Ar′CHMeMe), 1.59 (br, 2H, PdCH2CH3), 1.33 (br, 6H,
Pd(O(CH2CH3)2)), 1.31, 1.28, and 0.82 (3 d, 6H, 6H, and 12H
each, J ) 6.5, 6.5, and 6.5, CHMeMe, CHMeMe, CH′MeMe,
CH′MeMe), 0.19 (br, 3H, PdCH2CH3).
[[(2,6-(iP r )2C6H3)NdC(An )C(An )dN(2,6-(iP r )2C6H3)]P d -
1
(CH2CH2-µ-H)]BAr ′4 (8). H NMR (CDCl2F, 500 MHz, -130
°C): δ 8.10 (d, 1H, J ) 8.5, An pH), 8.08 (d, 1H, J ) 8.5, An
pH′), 7.77 (s, 8H, BAr′4 oH), 7.54 (s, 4H, BAr′4 pH), 7.55-7.40
(m, 8H, An mH, ArH), 6.81 (d, 1H, J ) 7.0, An oH), 6.75 (d,
1H, J ) 6.5, An oH′), 3.24 and 3.14 (m, 1H each, CHMeMe′,
C′HMeMe′), 2.23 (m, 2H, PdCH2CH2-µ-H), 1.42 (m, 2H,
PdCH2CH2-µ-H), 1.40-1.36 and 1.13-1.05 (br, 3H each,
CHMeMe′, CHMeMe′, C′HMeMe′, C′HMeMe′), -8.90 (br t, 1H,
2J HH ) 16, PdCH2CH2-µ-H). 13C NMR (CDCl2F, 100 MHz, -130
°C, static, selected peaks): δ 38.5 (1J CH ) 153, PdCH2CH2-µ-
Ack n ow led gm en t . This work was supported by
the National Science Foundation (CHE-9710380) and
by DuPont. L.H.S. thanks the NSF for a Graduate
Research Fellowship. We thank Dr. Steve Svejda for
synthesis and characterization of H(OiPr2)2BAr′4.
OM010197J