Mechanistic studies of Pd(II)-α-diimine-catalyzed olefin polymerizations

Article abstract of DOI:10.1021/ja000893v

Mechanistic studies of olefin polymerizations catalyzed by aryl-substituted α-diimine-Pd(II) complexes are presented. Syntheses of several cationic catalyst precursors, [(N^{^}N)Pd(CH₃)(OEt₂)]BAr′₄ (N^{^}N

Full text of DOI:10.1021/ja000893v

6686
J. Am. Chem. Soc. 2000, 122, 6686-6700
Mechanistic Studies of Pd(II)-R-Diimine-Catalyzed Olefin
Polymerizations1
Daniel J. Tempel,† Lynda K. Johnson,‡ R. Leigh Huff, Peter S. White, and
Maurice Brookhart*
Department of Chemistry, UniVersity of North Carolina at Chapel Hill, CB#3290,
Chapel Hill, North Carolina 27599-3290
ReceiVed March 13, 2000
Abstract: Mechanistic studies of olefin polymerizations catalyzed by aryl-substituted R-diimine-Pd(II)
complexes are presented. Syntheses of several cationic catalyst precursors, [(N∧N)Pd(CH3)(OEt2)]BAr′4
(N∧N ) aryl-substituted R-diimine, Ar′ ) 3,5-(CF3)2C6H3), are described. X-ray structural analyses of [ArNd
C(H)C(H)dNAr]Pd(CH3)(Cl) and [ArNdC(Me)C(Me)dNAr]Pd(CH3)2 (Ar ) 2,6-(iPr)2C6H3) illustrate that
o-aryl substituents crowd axial sites in these square planar complexes. Low-temperature NMR studies show
that the alkyl olefin complexes, (N∧N)Pd(R)(olefin)+, are the catalyst resting states and that the barriers to
migratory insertions lie in the range 17-19 kcal/mol. Following migratory insertion, the cationic palladium
alkyl complexes (N∧N)Pd(alkyl)+ formed are â-agostic species which exhibit facile metal migration along the
chain (“chain walking”) via â-hydride elimination/readdition reactions. Model studies using palladium-
n-propyl and -isopropyl systems provide mechanistic details of this process, which is responsible for introducing
branching in the polyethylenes made by these systems. Decomposition of the cationic methyl complexes
(ArN∧NAr)Pd(CH3)(OEt2)+ (Ar ) 2,6-(iPr)2C6H3, 2-tBuC6H4) occurs by C-H activation of â-C-H bonds of
the ortho isopropyl and tert-butyl substituents and loss of methane. The rate of associative exchange of free
ethylene with bound ethylene in (N∧N)Pd(CH3)(C2H4)+ is retarded by bulky substituents. The relationship of
these exchange experiments to chain transfer is discussed.
Introduction
We have described a family of cationic palladium(II) and nickel-
(II) complexes bearing bulky aryl-substituted R-diimine ligands
which are capable of polymerizing ethylene, R-olefins, and
internal and cyclic olefins to high molar mass polymers.18
Polymerizations of common, inexpensive monomers using these
catalysts result in the formation of unique materials including
highly branched, amorphous polyethylene, “chain-straightened”
poly(R-olefins), and melt processable polycyclopentene.18-30 In
Prior to reports from these laboratories, a few Ni(II)-based
catalysts were known to polymerize ethylene to linear poly-
ethylene.2-9 Due to rapid chain transfer, dimerization and
oligomerization reactions of ethylene4,10-16 and R-olefins17 were
the most commonly reported features of late metal catalysts.
† Current address: Air Products and Chemicals, Inc. Allentown, PA
18195-1501.
‡ Current address: E. I. DuPont and Co., Central Research and
Development, Wilmington, DE 19880-0328.
(15) Desjardins, S. Y.; Cavell, K. J.; Hoare, J. L.; Skelton, B. W.;
Sobolev, A. N.; White, A. H.; Keim, W. J. Organomet. Chem. 1997, 544,
163-174.
(16) Svedja, S. A.; Brookhart, M. Organometallics 1999, 18, 65-74 and
extensive references therein.
(1) This paper has been adapted, in part, from the following thesis:
Tempel, D. J. Ph.D. Dissertation, University of North Carolina at Chapel
Hill, 1998.
(2) Keim, W.; Appel, R.; Storeck, A.; Kruger, C.; Goddard, R. Angew.
Chem., Int. Ed. Engl. 1981, 20, 116-117.
(3) Klabunde, U.; Ittel, S. D. J. Mol. Catal. 1987, 41, 123-134.
(4) Wilke, G. Angew. Chem., Int. Ed. Engl. 1988, 27, 185-206.
(5) Keim, W.; Kowaldt, F. H.; Goddard, R.; Kruger, C. Angew. Chem.,
Int. Ed. Engl. 1978, 17, 466.
(6) Keim, W. Angew. Chem., Int. Ed. Engl. 1990, 29, 235.
(7) Ostoja Starzewski, K. A.; Witte, J. Angew. Chem. 1985, 97(7), 610-
612.
(8) Ostoja Starzewski, K. A.; Witte, J.; Reichert, K. H.; Vasiliou, G. In
Transition Met. Organomet. Catal. Olefin Polym., [Proc. Int. Symp.].
Kaminsky, W., Sinn, H., Eds.; Springer: Berlin, 1988; pp 349-360.
(9) For additional references, see: (a) Johnson, L. K.; Killian, C. M. In
Metallocene-Based Polyolefins: Preparation, Properties, and Technology;
Scheirs, J., Kaminsky, W., Eds.; Wiley: West Sussex, 2000; p 233. (b)
Ittel, S. D. In Metalorganic Catalysts for Synthesis and Polymerization;
Kaminsky, W., Ed.; Springer-Verlag: Heidelberg, 1999; p 616.
(10) Keim, W.; Behr, A.; Limbacker, B.; Kruger, C. Angew. Chem., Int.
Ed. Engl. 1983, 22, 503.
(11) Peuckert, M.; Keim, W. Organometallics 1983, 2, 594-597.
(12) Cavell, K. J.; Masters, A. F. Aust. J. Chem. 1986, 39, 1129-1134.
(13) Rix, F. C.; Brookhart, M. J. Am. Chem. Soc. 1995, 117, 1137-
1138.
(17) Certain Ni(II) species have been reported which convert R-olefins
to oligomers with DP’s of ∼4-20. A chain-walking mechanism was
proposed to account for the oligomer microstructures observed: (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.
(18) Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc.
1995, 117, 6414-6415.
(19) Killian, C. M. Ph.D. Dissertation, University of North Carolina at
Chapel Hill, 1996.
(20) Brookhart, M. S.; Johnson, L. K.; Killian, C. M.; Arthur, S. D.;
Feldman, J.; McCord, E. F.; McLain, S. J.; Kreutzer, K. A.; Bennett, A.
M. A.; Coughlin, E. B.; Ittel, S. D.; Parthasarathy, A.; Tempel, D. J. WO
9623010, 1996.
(21) McLain, S. J.; Feldman, J.; McCord, E. F.; Gardner, K. H.; Teasley,
M. F.; Coughlin, E. B.; Sweetman, K. J.; Johnson, L. K.; Brookhart, M.
Macromolecules 1998, 31, 6705-6707.
(22) Killian, C. M.; Tempel, D. J.; Johnson, L. K.; Brookhart, M. J.
Am. Chem. Soc. 1996, 118, 11664-11665.
(23) Thompson, D. T. Plat. Metals ReV. 1996, 40, 78-79.
(24) Killian, C. M.; Johnson, L. K.; Brookhart, M. Organometallics 1997,
16, 2005-2007.
(25) Johnson, L. K.; Killian, C. M. In Metallocene-Based Polyolefins:
Preparation, Properties, and Technology; Scheirs, J., Kaminsky, W., Eds.;
Wiley: West Sussex, 2000; p 233.
(14) Desjardins, S. Y.; Cavell, K. J.; Jin, H.; Skelton, B. W.; White, A.
H. J. Organomet. Chem. 1996, 515, 233-243.
10.1021/ja000893v CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/30/2000

Pd(II)-R-Diimine-Catalyzed Olefin Polymerizations
J. Am. Chem. Soc., Vol. 122, No. 28, 2000 6687
Scheme 1. Proposed Mechanism for Ethylene
Polymerization
Figure 1. Palladium-R-diimine complexes used as olefin polymeri-
zation catalysts.
addition, the palladium-R-diimine catalysts are the first transi-
tion metal complexes reported which are capable of copolymer-
izing ethylene and R-olefins with functionalized monomers such
as alkyl acrylates and methyl vinyl ketone.20,31,32 Following our
initial reports of these catalysts, other groups have described
similar results using R-diimine-based systems.33-40 Neutral,
bidentate Ni(II) systems which incorporate bulky aryl-substituted
imine units have also recently been shown to be quite effective
for polymerization of ethylene to high molecular weight
polymers.41-43
In contrast to the extremely active, highly air-sensitive nickel
analogues,18,44 the well-defined cationic palladium-alkyl species
depicted in Figure 1 are more easily synthesized and handled
and exhibit rates of reaction that make them well-suited for study
using variable temperature NMR spectroscopic techniques. On
the basis of our initial NMR observations and analyses of
polymer microstructures, we proposed the polymerization mech-
anism shown in Schemes 1A and 1B(i) (illustrated here for the
case of ethylene polymerization).
metal migration along the polymer chain (“chain running”), and
(3) chain transfer. The first and third processes are common
to all ethylene oligomerization and polymerization catalysts.
Metal migration, on the other hand, is insignificant for most
catalytic systems17,45,46 but is a major reaction pathway for
the Pd-R-diimine catalysts, thus making chain-running the
most distinguishing feature of these systems. For acyclic olefins,
the resting states of the palladium catalysts have been observed
by NMR spectroscopy to be alkyl-olefin complexes. In the
absence of excess olefin or other Lewis bases, the alkyl cations
exist as stable â-agostic species.44,47-49 A â-agostic alkyl
species is observed to be the resting state for cyclopentene
polymerizations.21
Propagation and isomerization (Scheme 1A) are easily
monitored by low-temperature NMR spectroscopy in this
system. Insertion of ethylene from the catalyst resting state 7
produces the linear alkyl agostic species 8, which can be trapped
by free ethylene. Repetition of this cycle (7 f 8 f 7) results
in the growth of a linear polymer chain. Complex 8 can also
â-hydride eliminate and reinsert with opposite regiochemistry
to produce the branched â-agostic species 10. This process may
involve an alkyl-olefin intermediate, 9, though such a species
has never been observed in this system. The branched alkyl
The overall mechanism of polymerization consists of three
main processes beyond initiation: (1) chain propagation, (2)
(26) Ittel, S. D. In Metalorganic Catalysts for Synthesis and Polymeri-
zation; Kaminsky, W., Eds.; Springer-Verlag: Heidelberg, 1999; p 616.
(27) Guan, Z. B.; Cotts, P. M.; McCord, E. F.; McLain, S. J. Science
1999, 283, 2059-2062.
(28) McLain, S. J.; McCord, E. F.; Johnson, 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.
(29) Guan, Z. B.; Cotts, P. M.; McCord, E. F.; McLain, S. J. Polym.
Prepr. (Am. Chem. Soc., DiV. Polym. Chem.) 1998, 39(2), 402-403.
(30) Guan, Z. B. Polym. Mater. Sci. Eng. 1999, 80, 50.
(31) Johnson, L. K.; Mecking, S.; Brookhart, M. J. Am. Chem. Soc. 1996,
118, 267-268.
(32) Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J. Am. Chem.
Soc. 1998, 120, 888-899.
(33) Galland, G. B.; de Souza, R. F.; Mauler, R. S.; Nunes, F. F.
Macromolecules 1999, 32, 1620-1625.
(34) Heinemann, J.; Mu¨lhaupt, R.; Brinkmann, P.; Luinstra, G. Macromol.
Chem. Phys. 1999, 200, 384-389.
(35) Eilerts, N. W. WO 9915569, 1999.
(36) Jurkiewicz, A.; Eilerts, N. W.; Hsieh, E. T. Macromolecules 1999,
32, 5471-5476.
(37) Schleis, T.; Heinemann, J.; Spaniol, T. P.; Mulhaupt, R.; Okuda, J.
Inorg. Chem. Commun. 1998, 1, 431-434.
(45) Exceptions for homopolyethylene (4-12 branches/1000C) in-
clude: (a) Karol, F. J.; Karapinka, G. L.; Wu, C.; Dow, A. W.; Johnson,
R. N.; Carrick, W. L. J. Polym. Sci., A-1 1972, 10, 2621-2637 (b) Karol,
F. J.; Johnson, R. N. J. Polym. Sci., Polym. Chem. Ed. 1975, 13, 1607-
1617.
(38) Schleis, T.; Spaniol, T. P.; Okuda, J.; Heinemann, J.; Mulhaupt, R.
J. Organomet. Chem. 1998, 569, 159-167.
(39) Pappalardo, D.; Mazzeo, M.; Pellechia, C. Macromol. Rapid
Commun. 1997, 18(12), 1017.
(40) For recent reviews of late metal polymerization catalysts, see: (a)
Britovsek, G. J. P.; Gibson, V. C.; Kimberly, B. S.; Maddox, P. J.;
McTravish, S. J.; Solan, G. A.; White, A. J. P.; Williams, D. J. Chem.
Commun. 1998, 849-850 (b) Piccolini, R. J. ChemTech 1999, 29(5), 39-
40 (c) Thompson, D. T. Plat. Metals ReV. 1996, 40, 78-79.
(41) Johnson, L. K.; Bennett, A. M. A.; Ittel, S. D.; Wang, L.;
Parthasarathy, A.; Hauptman, E.; Simpson, R. D.; Feldman, J.; Coughlin,
E. B. WO 9830609, 1997.
(42) Wang, C.; Friedrich, S.; Younkin, T. R.; Li, R. T.; Grubbs, R. H.;
Bansleben, D. A.; Day, M. W. Organometallics 1998, 17, 3149-3151.
(43) Younkin, T. R.; Connor, E. F.; Henderson, J. I.; Friedrich, S. K.;
Grubbs, R. H.; Bansleben, D. A. Science 2000, 287, 460-462.
(44) Svejda, S. A.; Johnson, L. K.; Brookhart, M. J. Am. Chem. Soc.
1999, 121, 10634-10635.
(46) Exceptions for poly(R-olefins) include: (a) Rieger, B.; Mu, X.;
Mallin, D. T.; Rausch, M. D.; Chien, J. C. W. Macromolecules 1990, 23,
3559-3568. (b) Soga, K.; Shiono, R.; Takemura, S.; Kaminsky, W.
Makromol. Chem. Rapid Commun. 1987, 8, 305-310. (c) Spalek, W.;
Antberg, M.; Aulbach, M.; Bachmann, B.; Dolle, V.; Haftka, S.; Kuber,
F.; Rohrmann, J.; Winter, A. In Ziegler Catalysts; Fink, G., Mu¨lhaupt, R.,
Brintzinger, H. H., Eds.; Springer: Berlin, 1995; p 83.
(47) Brookhart, M.; Green, M. L. H.; Wong, L.-L. Prog. Inorg. Chem.
1988, 36, 1-124.
(48) Tempel, D. J.; Brookhart, M. Organometallics 1998, 17, 2290-
2296.
(49) Similar species have been observed previously by Spencer et al.;
see, for example: Conroy-Lewis, F. M.; Mole, L.; Redhouse, A. D.; Litster,
S. A.; Spencer, J. L. J. Chem. Soc., Chem. Commun. 1991, 1601-1603.

6688 J. Am. Chem. Soc., Vol. 122, No. 28, 2000
Tempel et al.
cation 10 can continue to “chain run” by â-hydride elimination
and reinsertion with opposite regiochemistry, producing longer
branches. Alternatively, 10 can be trapped with ethylene, and
insertion produces a polymer chain containing a methyl branch
along the backbone. In the case of nickel, trapping and insertion
are competitive with chain-running, and thus the extent of
branching is dependent on ethylene pressure.18 For palladium
systems, the total number of branches is nearly insensitive to
ethylene pressure, but the number of branches-on-branches and
the overall architecture of the polymer does vary with ethylene
pressure.20,27 In the case of R-olefins, 2,1-insertion followed by
metal migration to the terminal carbon can result in incorporation
of R-olefin in a 1,ω (linear) fashion with the resultant polymer
containing fewer branches than expected for simple 1,2 mono-
mer enchainment, a phenomenon we term “chain-straighten-
ing”.18,20
active catalysts can be approached by several routes (eqs 1-5).
Bulky aryl-substituted R-diimine complexes slow the rate of
chain transfer relative to propagation, resulting in the formation
of high polymer. This rate retardation clearly results from
substituting the ortho positions of the aryl rings with bulky
substituents which then project into the axial sites of the square
planar complexes.50 The precise mode of chain transfer is not
yet completely understood. Initially we proposed that chain
transfer occurred by associative displacement of the unsaturated
chain from an olefin hydride complex (Scheme 1B(i)), a process
slowed by axial bulk. Calculations by Ziegler suggest chain
transfer may occur from the alkyl olefin resting state by direct
â-hydride transfer to monomer (Scheme 1B(ii)).51,52
We describe in this paper mechanistic investigations of olefin
polymerizations catalyzed by the well-defined cationic palla-
dium-alkyl complexes depicted in Figure 1. These investiga-
tions include NMR spectroscopic characterization of catalyst
resting states, measurement of insertion barriers and the effects
of substituents on these barriers, determination of the mode and
energetics of palladium migration, as well as the nature and
relative stabilities of key â-agostic alkyl complexes, and
measurement of associative exchange rates in alkyl-olefin
complexes which relate to the chain transfer process. These
experimental studies are complemented by recent theoretical
calculations from other groups concerning various mechanistic
details of these catalyst systems.51-57
Stable palladium complexes of the general formula (N∧N)-
PdMeCl are formed in diethyl ether or dichloromethane solvent
via displacement of 1,5-cyclooctadiene (COD) from (COD)-
PdMeCl by 1 equiv of R-diimine ligand.61 Subsequent addition
of 0.5 equiv of MgMe2 to the product then gives the desired
dimethyl complex along with 0.5 equiv of MgCl2 (eq 1). This
two-step sequence requires separation of the product from MgCl2
and gives moderate yields.
Single-crystal X-ray structures of (ArNdC(H)C(H)dNAr)-
PdMeCl and of (ArNdC(Me)C(Me)dNAr)PdMe2 (Ar ) 2,6-
C6H3-(i-Pr)2) were obtained. ORTEP diagrams for these two
compounds are included in Figures 2 and 3, respectively, along
with selected bond lengths and angles, which are within the
standard range for these types of complexes.50 Included in Figure
2 is a perspective of the molecular structure perpendicular to
the square plane of the complex along with selected torsion
angles. The two N-aryl rings lie roughly 80° and 67° relative
to the square plane of the molecule. Comparison with similar
molecules in the literature indicates that as the steric bulk of
the R-diimine backbone substituents or of the aryl ortho
substituents increases, the N-aryl rings tend to lie more
perpendicular to the square plane of the molecule, and the ortho
substituents more effectively block the axial sites.
Results and Discussion
The synthesis of (COD)PdMe2, a more reactive precursor,
has been reported; crystalline material can be isolated at low
temperature through addition of methyl cuprate to (COD)PdCl2
followed by workup with aqueous KCN.62 The (COD)PdMe2
complex is unstable but can be stored at temperatures below
ca. -10 °C. It reacts with R-diimines in ether to give nearly
quantitative yields of (N∧N)PdMe2, and product isolation is
straightforward (eq 2). Preparation of (N∧N)PdMe2 complexes
through in situ generation of (COD)PdMe2 has been reported
by Elsevier et al., and we have also found this route to be
effective.63 The desired complexes are obtained upon reaction
of 1 equiv of MeLi with (COD)PdMeCl in ether or THF at low
temperature, followed by addition of R-diimine. The product
must then be extracted from LiCl (eq 3).
Synthesis of Palladium Complexes. Syntheses of the N-aryl-
substituted R-diimine ligands are straightforward,58,59 and the
properties of R-diimines as ligands for transition metal com-
pounds have been reviewed.60 The dimethyl precursors to the
(50) van Asselt, R.; Elsevier, C. J.; Smeets, W. J. J.; Spek, A. L.; Benedix,
R. Recl. TraV. Chim. Pays-Bas 1994, 113, 88-98.
(51) Deng, L.; Margl, P.; Ziegler, T. J. Am. Chem. Soc. 1997, 119, 1094-
1100.
(52) Deng, L.; Woo, T. K.; Cavallo, L.; Margl, P. M.; Ziegler, T. J. Am.
Chem. Soc. 1997, 119, 6177-6186.
(53) Michalak, A.; Ziegler, T. Organometallics 1999, 18, 3998-4004.
(54) Woo, T. K.; Ziegler, T. J. Organomet. Chem. 1999, 591, 204-
213.
(55) Stromberg, S.; Zetterberg, K.; Siegbahn, P. E. M. J. Chem. Soc.,
Dalton Trans. 1997, 4147-4152.
(56) Musaev, D. G.; Svensson, M.; Morokuma, K.; Stromberg, S.;
Zetterberg, K.; Siegbahn, P. E. M. Organometallics 1997, 16, 1933-1945.
(57) Froese, R. D. J.; Musaev, D. G.; Morokuma, K. J. Am. Chem. Soc.
1998, 120, 1581-1587.
Canty reported the preparation of palladium-dimethyl com-
plexes of nitrogen donor ligands via addition of the ligands to
either previously isolated (pyridazine)PdMe2 or in situ generated
(58) Svoboda, M.; tom Dieck, H. J. Organomet. Chem. 1980, 191, 321-
328.
(59) tom Dieck, H.; Svoboda, M.; Greiser, T. Z. Naturforsch 1981, 36b,
823-832.
(60) van Koten, G.; Vrieze, K. AdV. Organomet. Chem. 1982, 21, 151-
239.
(61) Ru¨lke, R. E.; Ernsting, J. M.; Spek, A. L.; Elsevier, C. J.; van
Leeuwen, P. W. N. M.; Vrieze, K. Inorg. Chem. 1993, 32, 5769-5778.
(62) Rudler-Chauvin, M.; Rudler, H. J. Organomet. Chem. 1977, 134,
115-119.
(63) van Asselt, R.; Rijnberg, E.; Elsevier, C. J. Organometallics 1994,
13, 706-720.

Pd(II)-R-Diimine-Catalyzed Olefin Polymerizations
J. Am. Chem. Soc., Vol. 122, No. 28, 2000 6689
Figure 3. ORTEP diagram of (ArNdC(Me)C(Me)dNAr)PdMe2
(Ar ) 2,6-C6H3-(i-Pr)2). Selected bond lengths (Å): Pd(1)-N(1)
2.133(4), Pd(1)-N(4) 2.145(4), Pd(1)-C(7) 2.023(6), Pd(1)-C(8)
2.033(5), N(1)-C(2) 1.317(6), N(1)-C(31) 1.460(6), C(2)-C(3)
1.366(8), C(2)-C(5) 1.524(7), C(3)-N(4) 1.297(6), C(3)-C(6)
1.528(7), N(4)-C(11) 1.459(6). Selected bond angles (deg): N(1)-
Pd(1)-N(4) 74.81(15), N(1)-Pd(1)-C(7) 174.97(19), N(1)-Pd(1)-
C(8) 99.55(20), N(4)-Pd(1)-C(7) 100.17(19), N(4)-Pd(1)-C(8)
174.31(20), C(7)-Pd(1)-C(8) 85.47(23), Pd(1)-N(1)-C(2)
114.2(3), Pd(1)-N(1)-C(31) 123.7(3), C(2)-N(1)-C(31) 122.1(4),
N(1)-C(2)-C(3) 118.2(4), N(1)-C(2)-C(5) 120.0(5), C(3)-C(2)-
C(5) 121.8(4), C(2)-C(3)-N(4) 118.1(4), C(2)-C(3)-C(6) 120.8(4),
N(4)-C(3)-C(6) 121.1(5), Pd(1)-N(4)-C(3) 114.7(3), Pd(1)-N(4)-
C(11) 121.5(3), C(3)-N(4)-C(11) 123.7(4). Selected torsion angles
(deg): N(1)-C(2)-C(3)-N(4) -0.9(3), C(2)-N(1)-C(31)-C(32)
78.6(6), C(3)-N(4)-C(11)-C(16) -79.4(6).
molecule of solvent acts as a Lewis base and occupies the vacant
coordination site. The ether ligand is extremely labile, making
the precursors ideal for low-temperature NMR mechanistic
studies.
Figure 2. (a) ORTEP diagram of (ArNdC(H)C(H)dNAr)PdMeCl
(Ar ) 2,6-C6H3-(i-Pr)2). Selected bond lengths (Å): P(1)-Cl(1)
2.300(4), Pd(1)-C(1) 2.020(11), Pd(1)-N(11) 2.033(9), Pd(1)-N(14)
2.208(9), N(11)-C(12) 1.315(16), N(11)-C(21) 1.448(14), C(12)-
C(13) 1.434(17), C(13)-N(14) 1.262(18), N(14)-C(41) 1.450(16).
Selected bond angles (deg): Cl(1)-Pd(1)-C(1) 89.7(4), Cl(1)-Pd(1)-
N(11) 174.9(3), Cl(1)-Pd(1)-N(14) 99.9(3), C(1)-Pd(1)-N(11)
94.3(4), C(1)-Pd(1)-N(14), N(11)-Pd(1)-N(14) 76.2(4), Pd(1)-
N(11)-C(12) 117.6(8), Pd(1)-N(11)-C(21) 124.1(7), C(12)-N(11)-
C(21) 118.0(9), Pd(1)-N(14)-C(13) 111.3(8), Pd(1)-N(14)-C(41)
126.2(8), C(13)-N(14)-C(41) 122.5(10), N(11)-C(12)-C(13)
115.3(12), C(12)-C(13)-N(14) 119.5(11), N(11)-C(21)-C(22)
120.8(11), N(11)-C(21)-C(26) 116.2(10), N(14)-C(41)-C(42)
119.5(11), N(14)-C(41)-C(46) 116.2(11). Selected torsion angles
(deg): N(11)-C(12)-C(13)-N(14) 1.2(10), C(12)-N(11)-C(21)-
C(26) 100.5(21), C(13)-N(14)-C(41)-C(46) -113.5(25). (b) View
of the molecular structure perpendicular to the square plane of the
complex.
For preparative-scale polymerizations, complexes in which
the vacant coordination site is occupied by a stronger Lewis
base than ether are sufficient. For example, an active catalyst
is formed in situ upon addition of NaBAr′4 to (N∧N)PdMeCl
in the presence of olefin. Acetonitrile complexes can be isolated
under similar conditions in the absence of olefin, but chloride-
bridged, monocationic dimers of the formulas [(N∧N)PdMe-µ-
Cl]2BAr′4 are formed when the same reaction is run in diethyl
ether. Ether is apparently too weakly coordinating to break open
the dimer. Six-membered chelate complexes resulting from
acrylate insertion also display activity for olefin homopolym-
erization and olefin/acrylate copolymerization.31,32
The routes described above were found to be ineffective for
the preparation of a palladium complex from an R-diimine ligand
with an ortho tert-butyl group on the ring, e.g., 2-C6H4(t-Bu)Nd
C(Me)C(Me)dN-2-C6H4(t-Bu), and a different approach was
required. Addition of NaBAr′4 to (COD)PdMeCl at low tem-
perature in a mixture of CH2Cl2 and acetonitrile yields [(COD)-
PdMeNCMe][BAr′4].65 This salt can be used directly or isolated
and stored at -30 °C and is stable for extended periods.
Displacement of COD by the ortho tert-butyl-substituted R-di-
(Me2S)2PdMe2.64 Displacement of pyridazine from the former
also occurs for the N-aryl R-diimines and gives moderate yields
of product in ether solvent (eq 4). R-Diimine displacement of
dimethyl sulfide from the in situ generated precursor in ether
gives somewhat lower yields of product, but a modification of
this route employing Mg(13CH3)2 proved to be effective for
synthesis of a 13C-labeled analogue (eq 5).48
Ether adducts of well-defined cationic Pd(II)-R-diimine salts
-
of BAr′4 (Ar′ ) 3,5-C6H3(CF3)2) are synthesized through
addition of H(OEt2)2BAr′4 to complexes with the general
formula (N∧N)Pd(CH3)2 (eq 6). Through protonolysis, the
oxonium acid cleaves one of the Pd-methyl bonds, opening
up a coordination site on the metal center through loss of
methane. When the reaction is carried out in diethyl ether, one
(65) This compound was developed by C. M. Killian, M. Brookhart,
and J. Feldman. See Brookhart et al. WO 9623010, 1996 for procedure.
(64) Byers, P. K.; Canty, A. J. Organometallics 1990, 9, 210-220.

6690 J. Am. Chem. Soc., Vol. 122, No. 28, 2000
Tempel et al.
Scheme 2. Possible Intermediates for Palladacycle
imine occurs for this highly electrophilic precursor to yield the
Formation via R-Diimine C-H Activation
expected acetonitrile adduct, 6b (eq 7).
Catalyst Stability. At room temperature and in the absence
of olefin, the cationic acetonitrile adducts are stable for weeks
in methylene chloride or chloroform solution. However, under
similar conditions solutions of the cationic ether adducts are
observed to undergo decomposition via C-H activation of the
ortho aryl substituents to give a six-membered palladacycle and
methane (eq 8).66
activation of 5a is more rapid than for 1a and 2a and the product,
5a′, is more stable than both 1a′ and 2a′. Lack of a highly
unfavorable interaction between the ligand backbone and the
ortho hydrogen of the in-plane aryl group likely accounts for
the rapid activation of 5a and the relative stability of 5a′.
Introduction of 20 equiv of ethylene to 5a′ leads to displacement
of ether to give ethylene-ligated palladacycle 5c′. However,
unlike 1c′, palladacycle 5c′ exhibits little or no evidence for
migratory insertion even after 24 h at room temperature.
The rates of C-H activation of the ether adducts 1a, 2a, and
5a decrease substantially upon addition of an excess of ether.
For example, upon addition of 10 equiv of Et2O to a 0.015 M
CD2Cl2 solution of 5a at 0 °C, the C-H activation process is
too slow to be monitored over a reasonable time period, and
at 10 °C C-H activation occurs with a half-life of more than
1.5 h. One equivalent of acetonitrile readily displaces ether from
5a to give 5b, and this complex exhibits only slow deactivation
even at room temperature. A 0.015 M solution of the acetonitrile
complex 4b (R ) Me; aryl ) o-tert-butyl) exhibits a half-life
for C-H activation of several days at room temperature, and
the resulting palladacycle 4b′ has been the most stable complex
of this type encountered.
Inhibition of C-H activation by an increased concentration
of ether or addition of a stronger Lewis base such as acetonitrile
strongly suggests that the C-H activation takes place through
an intermediate that is formed reversibly. More specifically,
intramolecular activation of a C-H bond of one of the ligand
methyl groups likely occurs through initial formation of an
agostic interaction between that C-H bond and the metal center
(Scheme 2). This mechanism requires displacement of L in order
to open up a coordination site for an agostic interaction to form.
Displacement of L may occur through a dissociative mechanism
involving formation of a 14-electron intermediate, but the
possibility also exists that L is displaced associatively by the
incoming C-H bond. It is not known whether oxidative
addition/reductive elimination or concerted σ-bond metathesis
is responsible for completion of the reaction.
This intramolecular C-H activation requires one of the ligand
aryl rings to rotate into the square plane of the complex.
Increasing the steric bulk of the ligand is therefore expected to
make this process less favorable, and conversely, the process
should become more facile upon decreasing the steric bulk of
the ligand. For the reaction indicated in eq 8, changing the
backbone substituents from hydrogens to methyl groups results
in a significant decrease in rate for the overall process. At
25 °C, the reaction occurs with a half-life of less than 5 min
for 1a (R ) H) and a half-life of ca. 30 min for 2a (R ) CH3)
for 0.015 M dichloromethane solutions of these complexes.
Palladacycles 1a′ and 2a′ were not isolated, but complex 1a′
is stable for days as a concentrated dichloromethane solution.
In contrast, for the decomposition of ether adduct 2a, only
methane and nonselective decomposition products are observed.
The proposed palladacycle 2a′, if formed, is only a transient
intermediate. Presumably, unfavorable steric interactions be-
tween the coplanar backbone and the ortho isopropyl substituents
make palladacycle 2a′ very unstable.
Addition of ethylene to 1a′ in solution at low temperature
results in the formation of an observable ethylene adduct (1c′),
and addition of an excess of ethylene results in branched
polyethylene formation with initiation occurring more slowly
than propagation.
The unsymmetrically substituted ether complex 5a (R,R )
An, 0.015 M in CD2Cl2) undergoes deactivation with a half-
Migratory Insertion Rates. In studying the mechanism of
olefin polymerization as catalyzed by the palladium R-diimine
catalysts, rates for the migratory insertion of olefins were
determined using 1H NMR spectroscopy (eq 10). For example,
life of ca. 60 min at -3 °C (eq 9) which corresponds to t1/2
<
2 min at 25 °C. Therefore, under similar conditions C-H
(66) A similar observation was reported for a square planar N-alkyl
R-diimine complex of nickel in which intramolecular C-H activation of a
ligand methyl group occurs to give a five-membered metallacycle: tom
Dieck, H.; Svoboda, M. Chem. Ber. 1976, 109, 1657-1664.
addition of 10-20 equiv of ethylene to the ether adducts 1a,
2a, 3a, 5a, and 6a gave the corresponding methyl ethylene

Pd(II)-R-Diimine-Catalyzed Olefin Polymerizations
Table 1. Kinetic Data for Pd-Catalyzed Insertion of Ethylenea
J. Am. Chem. Soc., Vol. 122, No. 28, 2000 6691
Table 2. Kinetic Data for Pd-Catalyzed Insertion of Propylenea
first insertion
subsequent insertions
first insertion
subsequent insertions
‡
‡
q
q
k1
∆G1
ksub
∆Gsub
k1
∆G1
ksub
∆Gsub
catalyst
(×103 s-1
)
kcal/mol)
(×103 s-1
)
(kcal/mol)
catalyst
(×103 s-1
)
(kcal/mol)
(×103 s-1
)
(kcal/mol)
1c, -30 °C
2c, -30 °C
3c, -26 °C
5c, -14 °C
6c, -20 °C
1.9
1.7
0.83
2.1
0.63
17.2
17.2
17.8
18.3
18.4
0.88
3.4
0.67
0.71
0.42
17.5
16.9
18.0
18.8
18.6
2a, -30 °C
3a, -13 °C
0.54
1.3
17.8
18.6
1.2
1.5
17.4
18.5
a Error in ∆G‡ values is (0.1 kcal/mol.
Scheme 3. Chain-Running Leading to 1,2-Migration of
a Error in ∆Gq measurements is (0.1 kcal/mol.
Palladium
(R ) Me) are most active and catalysts based on structures 1
(R ) H) and 3 (R,R ) An) exhibit fairly similar activities
to one another which are lower than activities for 2.18 Catalyst
5a exhibits very few turnovers and gives only low molecular
weight polyethylene.
Barriers to insertion in both (N∧N)Ni(olefin)R+ and (N∧N)-
Pd(olefin)R+ species have been theoretically addressed by
Ziegler,51,52 Morokuma,56,57 and Siegbahn.55 The calculated
barriers for Ni systems invariably fall below those of the
corresponding Pd systems, in agreement with experimental
observations.44 In addition, moving from model systems of the
type (HNdC(H)C(H)dNH)M(olefin)R+ to systems bearing
bulky ortho-disubstituted aryl groups and substituted backbone
positions (ArNdC(R′)C(R′)dNAr)M(olefin)R+, the calculated
barriers decrease,51,52 which is consistent with the observations
above that bulkier ortho aryl substituents cause a decrease in
the migratory insertion barrier. There is reasonable quantitative
agreement between the barriers calculated for bulky aryl-
palladium diimine systems and the barriers measured in this
work. For example, Morokuma57 has calculated a barrier for
insertion of 14.1 kcal/mol for complex 2c in which the
experimentally measured barrier is 17.2 kcal/mol.
Alkyl Intermediates. The number of alkyl branches observed
for polyethylene made using a particular Pd-R-diimine catalyst
is relatively independent of reaction conditions including
temperature, ligand structure, and ethylene concentration,
although the type of branches or polymer architecture observed
is dependent upon these variables.18,20,21,27,31,32,36 Polyethylene
samples made with the 2,6-diisopropyl-substituted catalysts
(structures 1-3) typically exhibit ca. 100-120 branches for
every 1000 carbons. This is in sharp contrast to MAO-activated
nickel analogues which can be fine-tuned by choice of reaction
conditions to give polyethylene ranging in structure from almost
completely linear to moderately branched.
Figure 4. Eyring plot for subsequent ethylene insertions by 2c.
complexes 1c, 2c, 3c, 5c, and 6c. The ensuing first insertion of
ethylene into the Pd-methyl bond is first order in catalyst, and
the rate for this step is determined by monitoring the decrease
of the integral for the Pd-Me signal. The combined rate for
subsequent insertions is zero order in ethylene and is determined
from the turnover frequency based on the decreasing signal for
free ethylene in solution. Rates for first and subsequent insertions
were used to calculate rate constants and Gibbs free energies
of activation (Table 1). For complex 2c, rate constants for
subsequent insertions were obtained over a 40 °C temperature
range (-40 to 0 °C). The Eyring plot of the data is shown in
Figure 4 and gives ∆Hq ) 14.2 ( 0.1 kcal/mol and ∆Sq )
-11.2 ( 0.8 eu.
The clearest indication of catalyst activity for comparative
analysis is the kinetic data for subsequent insertions and more
specifically values for ∆Gsubq. The data for the diisopropyl aryl-
substituted catalysts 1c-3c indicate that 2c (R ) Me) is the
most active for ethylene polymerization, suggesting that ligand
steric bulk is the predominant factor affecting the overall rate
of polymerization. Increased steric bulk about the metal center
likely raises the ground-state energies of the methyl ethylene
complexes, resulting in lower barriers to migratory insertion.
The o-tert-butyl aryl-substituted complex 5c and the o-methyl
aryl-substituted complex 6c exhibit significantly higher barriers
to insertion than 1c-3c and are, relatively, poorer catalysts for
ethylene polymerization.
As previously described, polyethylene branching arises from
facile â-hydride elimination and reinsertion of Pd-alkyl
intermediates (chain-running). An olefin-hydride intermediate67
resulting from â-hydride elimination of a Pd-alkyl species may
reinsert with opposite regiochemistry to give a new alkyl
complex (Scheme 3). A branch is incorporated into the polymer
every time ethylene insertion occurs from any alkyl intermediate
other than a primary Pd-alkyl olefin species. Characterization
by 13C NMR spectroscopy of polyethylene prepared from
Kinetic data for the migratory insertion of propylene using
catalysts 2a and 3a were collected in a fashion identical to that
described for ethylene. The results for these two catalysts are
presented in Table 2 and indicate the same trend in activities
q
observed for ethylene insertion. In each case, ∆Gsub for
q
propylene propagation increases by 0.5 kcal/mol over ∆Gsub
(67) We describe here a discrete olefin-hydride intermediate; however,
â-H transfer may occur through a concerted process with an olefin-hydride-
like transition state, as suggested by Ziegler et al.: (a) Deng, L.; Margl, P.;
Ziegler, T. J. Am. Chem. Soc. 1997, 119, 1094-1100 (b) Deng, L.; Woo,
T. K.; Cavallo, L.; Margl, P. M.; Ziegler, T. J. Am. Chem. Soc. 1997, 119,
6177-6186.
determined for ethylene propagation. Importantly, the general
trends apparent in Tables 1 and 2 are consistent with what
is observed for preparative-scale polymerizations. Namely, for
both ethylene and propylene, catalysts based on structure 2

6692 J. Am. Chem. Soc., Vol. 122, No. 28, 2000
Tempel et al.
Scheme 4. 2 Formation of Alkyl Agostic Species via
Multiple C2H4 Insertions from 1c-3c
Scheme 5. Single Insertion of Ethylene To Give â-Agostic
Isopropyl Complex i-16
Figure 5. Equilibrium constants for trapped propyl isomers.
expected on the basis of analogy to the normal stability order
(1° > 2° > 3°) of metal alkyl complexes (presumed to be a
steric effect). However, the inductive electron-donating ability
of a secondary alkyl group is greater than that of a primary
alkyl group and should provide more electron density to the
electrophilic Pd metal center. Theoretical calculations have been
reported for â-agostic n-propyl and isopropyl Pd-R-diimine
compounds that predict a â-agostic isopropyl species to be
favored by as little as 0.5 kcal/mol or by as much as 2.0 kcal/
mol.57 Although we have not directly observed n-16 by NMR
spectroscopy, ∆G for the i-16/n-16 equilibrium was determined
to be ca. 1.2-1.3 kcal/mol based on trapping experiments using
dimethyl sulfide (vide infra).
palladium catalysts indicates that the catalysts incorporate
branches that contain branches.27,29,68
Well-defined Pd-alkyl species derived from 1c-3c have
been studied using variable temperature NMR spectroscopy to
specifically probe factors contributing to the formation of
polymer branches. Complexes 1c-3c insert ethylene to yield a
mixture of branched products after several turnovers, and the
displaced ether does not recoordinate upon complete con-
sumption of added olefin. The resulting 1H NMR spectra at ca.
-95 °C are relatively complex and contain a group of broad
shifts at ca. -8.4 ppm that are consistent with the formation of
alkyl agostic species (Scheme 4). Complexes arising from a
single insertion of olefin into the Pd-CH3 bond of 3c exhibit
relatively simple NMR spectra and are therefore ideally suited
as models for polymerization intermediates. In particular, the
acenaphthyl backbone of the ligand contains hydrogens that
exhibit unique, resolved 1H NMR shifts that simplify observation
and characterization of reaction intermediates. These experi-
ments are described below.
Isomerization of Cationic Pd-Alkyl Complexes. Various
trapping ligands have been used to observe Lewis base adducts
of i-16 and n-16, yielding insight into the mechanism of chain-
running.69 Addition of 2 equiv of acetonitrile to an NMR sample
of i-16/n-16 at -78 °C gives almost exclusively the isopropyl
1
acetonitrile complex i-16b by H NMR spectroscopy. Isomer-
ization to the n-propyl acetontrile complex n-16b begins to occur
at temperatures above -70 °C, and an equilibrium value of 0.29
is observed for n-16b/i-16b at -66 °C. Use of ethylene as the
trapping ligand under the same conditions results in initial
trapping as i-16c with isomerization to n-16c again occurring
above -70 °C. At -66 °C the n-propyl ethylene complex is
favored with Keq ) 19 for n-16c/i-16c. In contrast to acetonitrile
1
and ethylene, the H NMR spectra indicate that addition of
â-Agostic Isopropyl Complex. Addition of a single equiva-
lent of ethylene to an NMR tube containing 3a dissolved in
CDCl2F leads to displacement of coordinated ether by ethylene.
Migratory insertion of the resulting methyl ethylene complex
3c takes place at temperatures above -30 °C (∆Gq ) 17.8 kcal/
mol) to give a single species that displays dynamic behavior.
propylene to i-16/n-16 at -78 °C yields predominantly the
n-propyl propylene complex, n-16d, with Keq ) 43 at -65 °C
for n-16d/i-16d. However, addition of 2 equiv of propylene at
-130 °C allowed trapping of i-16 as i-16d, with i-16d readily
isomerizing to n-16d at temperatures above ca. -85 °C.
Consideration of the equilibrium values shown in Figure 5
indicates that formation of the sterically less congested n-propyl
complex becomes more favored with bulkier trapping ligands.
Isomerization of the isopropyl ethylene complex i-16c to the
n-propyl ethylene complex n-16c models the chain-running
process (eq 11). Several possible mechanisms may apply to this
1
The static H and 13C NMR spectra of this dynamic species
can be observed at temperatures below ca. -110 °C. These static
spectra are consistent with a â-agostic isopropyl complex, i-16,
which presumably forms via rapid isomerization of the initial
Pd-n-propyl insertion product, n-16 (Scheme 5). The signal
for the agostic hydrogen appears as a broad triplet at -8.0 ppm
1
in the H NMR spectrum with a geminal coupling constant of
17 Hz. The carbon of the agostic methyl group appears in the
13C NMR spectrum at 19.5 ppm with JCH ) 152 and 65 Hz.
The ether-free salt of i-16 is fairly stable and can be prepared
on a large scale and stored at -30 °C for extended periods
without decomposition.48
isomerization. One possibility is that isomerization occurs
without ethylene loss, for example, through a five-coordinate
The stability of the secondary isopropyl agostic complex
relative to the primary n-propyl agostic complex was not
(69) For previous studies of alkyl isomerization in unhindered Pd systems,
see: (a) Reger, D. L.; Garza, D. G.; Baxter, J. C. Organometallics 1990, 9,
873-874 (b) Reger, D. L.; Garza, D. G.; Lebioda, L. Organometallics 1991,
10, 902-906 (c) Reger, D. L.; Garza, D. G.; Lebioda, L. Organometallics
1992, 11, 4285-4292.
(68) Nelson, L. T. J.; McCord, E. F.; Johnson, L. K.; McLain, S. J.;
Ittel, S. D.; Killian, C. M. Polym. Prepr. (Am. Chem. Soc., DiV. Polym.
Chem.) 1997, 38(1), 133-134.

Pd(II)-R-Diimine-Catalyzed Olefin Polymerizations
J. Am. Chem. Soc., Vol. 122, No. 28, 2000 6693
Figure 6. Free energy diagram for isomerization of i-16c according
to Mechanism I.
Figure 7. Free energy diagram for isomerization of i-16c according
to Mechanism II.
bis-olefin hydride intermediate. Alternatively, isomerization may
occur via ethylene loss, 1,2-migration of palladium in the alkyl
cation, and recapture of ethylene. For the latter process, two
limiting kinetic situations can apply and are illustrated as
Mechanisms I and II in Scheme 6.
Table 3. Rate of Isomerization of i-16c to n-16ca
equiv of C2H4
k1 (×104 s-1
2.9, -66 °C
2.8, -66 °C
2.5, -66 °C
1.8, -67 °C
)
∆G‡ (kcal/mol)
1
5
10
45
15.3
15.3
15.4
15.4
In Mechanism I, a rapid preequilibrium exists between i-16c
and i-16 + C2H4 (i.e., k-1[C2H4] > k2). The transition state for
overall conversion of i-16c to n-16c lies between i-16 and n-16.
Kinetic analysis as shown in Scheme 6 indicates that if this
mechanism pertains, the rate of isomerization of i-16c to n-16c
should be inverse in ethylene and thus retarded by added
ethylene. In Mechanism II the interconversion of alkyl cations
i-16 and n-16 is rapid relative to the trapping reactions (i.e., k2,
k-2 > k-1[C2H4], kT[C2H4]). The rapidly equilibrating cations
i-16 and n-16 can, for kinetic analysis, be treated as a single
species. As shown in Scheme 6, kinetic analysis of Mechanism
II predicts no inhibition of the rate of conversion of i-16c to
n-16c by added ethylene.
a Error in ∆Gq values is (0.1 kcal/mol.
in Mechanism I the trapping barriers are much lower than the
barrier to interconversion of the Pd-alkyl cations i-16 and n-16,
while in Mechanism II the trapping barriers are higher than the
barriers to interconversion.
Experiments were carried out in which the rate of isomer-
ization of i-16c to n-16c was monitored as a function of added
ethylene. Results are summarized in Table 3. Addition of 5 equiv
of ethylene results in no measurable decrease in isomerization
rate. Addition of 10 and 45 equiv of ethylene results in a small
decrease in rate, but clearly the isomerization rate shows little
dependence on the concentration of ethylene. These results
suggest that either isomerization occurs from the four-coordinate
It is instructive to consider free energy diagrams for these
two limiting mechanisms as shown in Figures 6 (Mechanism I)
and 7 (Mechanism II). The key difference in these cases is that
Scheme 6. Possible Mechanisms for Alkyl Isomerization

6694 J. Am. Chem. Soc., Vol. 122, No. 28, 2000
Tempel et al.
Table 4. Ground-State Energy Difference for i-16 and n-16
temp (°C)
Keq (i-16/n-16)
∆G (kcal/mol)
-51.2
-65.2
-84.3
0.056
0.053
0.036
1.3
1.2
1.3
data in Table 3 suggests that the very small rate suppression
observed at high free ethylene concentration is likely due to an
increase in the rate of trapping to the point where trapping is
becoming competitive with i-16/n-16 interconversion. Under
these conditions, the limiting kinetic analysis shown for Mech-
anism II will begin to break down and some rate retardation
will be observed in this “intermediate” case.
Two additional features of these isomerizations should be
noted. First, consistent with a mechanism involving loss of L
for isomerization of [(N∧N)Pd(i-Pr)(L)+] to [(N∧N)Pd(n-Pr)-
(L)+], rates track the binding energies of the ligands70 with the
rate of L ) CH3CHdCH2 > CH2dCH2 > CH3CN > (CH3)2S.
Second, the fact that ethylene exchange in palladium-methyl
ethylene complexes occurs by an associative mechanism (vide
infra) suggests that dissociation of ethylene (or other L) from
[(N∧N)Pd(Me)(L)+] complexes may be highly disfavored.
However, in contrast to the methyl complexes, [(N∧N)Pd-
(propyl)(L)+] complexes contain a â-hydrogen, and thus ligand
dissociation may be greatly accelerated by an intramolecular
displacement of L to form, in a concerted fashion, the â-agostic
palladium-alkyl complexes n-16 and i-16.
Figure 8. Free energy diagram for isomerization of i-16e.
resting state without ethylene loss or that Mechanism II applies
in which the trapping barriers are higher than the interconversion
barriers.
We reasoned that if Mechanism II applies in the case of
ethylene, use of a small, much more nucleophilic trapping ligand
would result in much lower barriers to trapping, and a limiting
case, as described by Mechanism I, might be reached in which
isomerization is now retarded by added trapping ligand. In
contrast, if isomerization occurs without loss of trapping ligand
then the rate should not be affected by additional ligand.
On the basis of previous equilibrium binding studies of
ligands to Pd(II) centers, we chose to examine trapping and
isomerization using dimethyl sulfide.70 Addition of a single
equivalent of Me2S to i-16/n-16 at -78 °C yields predominantly
the isopropyl adduct i-16e, and subsequent isomerization to
the n-propyl complex n-16e occurs only at temperatures above
-50 °C (eq 12). Quantitative rate measurements were carried
Rates of Interconversion of Agostic Species. Experiments
conducted thus far suggest that interconversion of i-16 and n-16
is quite rapid, but no quantitative data are available to estimate
these rates. Dynamic NMR experiments are described here
1
which attempt to address this problem. The static H NMR
spectrum of â-agostic isopropyl complex i-16 contains a
resonance at 2.88 ppm corresponding to the methine hydrogen
of the Pd-isopropyl group. Conversion of i-16 to n-16 and
return could result in exchange of the methine hydrogen with
the methyl hydrogen; thus line shape analysis of this shift could
be useful for investigating interconversion of i-16 and n-16.
However, this shift is obscured by the isopropyl methine signals
of the ligand. In fact, the methine signal could only be assigned
at 2.88 ppm on the basis of a COSY experiment performed at
ca. -110 °C.48 To circumvent this problem, complex 6a was
employed in the synthesis of a â-agostic isopropyl analogue
i-17 bearing methyl rather than isopropyl substituents in the
ortho positions of the ligand aryl rings (eq 13).
out at -43 °C. Addition of 1 equiv of Me2S to i-16/n-16
generates i-16e without free Me2S present. Isomerization occurs
by a clean first-order process with k ) 1.1 × 10-4 s-1 at
-43 °C. Addition of 5 equiv of Me2S to i-16/n-16 yields i-16e
with 4 equiv of free Me2S present. First-order isomerization to
n-16e occurs at -43 °C with k ) 0.25 × 10-4 s-1. Clearly,
free Me2S greatly retards isomerization.
This result establishes that isomerization of i-16e occurs via
Mechanism I and does not take place from i-16e without loss
of Me2S. The free energy diagram for the Me2S case is shown
in Figure 8. Since trapping below -50 °C is very fast relative
to interconversion of i-16 and n-16, the equilibrium ratio of
i-16:n-16 can be assessed by trapping at low temperature and
carefully measuring the ratio of i-16e:n-16e. Table 4 summarizes
Keq values for i-16/n-16 at three different temperatures together
with the corresponding ∆G values, which are ca. 1.2-1.3
kcal/mol.
6a
i-17
The methine hydrogen of the Pd-isopropyl group of i-17
resonates as a septet at 2.9 ppm and begins to display line
broadening at ca. -65 °C. Broadening results from exchange
of the methine hydrogen (H′) with a hydrogen from one of the
isopropyl methyl groups (i.e., i-17 f i-17′, Scheme 7). This
occurs via â-hydride elimination from i-17 and reinsertion to
give n-17, followed by rotation about the CR-Câ bond of n-17
to give n-17′, and finally â-hydride elimination/reinsertion to
give i-17′.
The results using dimethyl sulfide as the trapping ligand
strongly suggest that in the case of ethylene, a weaker trapping
ligand, Mechanism II applies. Reconsideration of the kinetic
Approximate rate constants for this exchange (kH′) were
determined by line-shape analysis (Table 5). If the barrier to
rotation leading to agostic exchange (n-17 f n-17′: ∆Grotq) is
(70) Rix, F. C.; Brookhart, M.; White, P. S. J. Am. Chem. Soc. 1996,
118, 4746-4764.

Pd(II)-R-Diimine-Catalyzed Olefin Polymerizations
J. Am. Chem. Soc., Vol. 122, No. 28, 2000 6695
Scheme 7. Methine (H′) Exchange in â-Agostic Isopropyl
Complex i-17
Figure 10. Modeled interconversion of â-agostic species through a
propylene hydride intermediate.56,57
Scheme 8. Associative Ethylene Exchange
Table 5. Kinetic Data from Exchange of the Methine Hydrogen of
i-17a
q
q
T
(°C)
∆W1/2
kH′
)
∆GH′
kconv
∆Gconv
c
(Hz)b
(s-1
35
115
(kcal/mol)
(s-1
)
(kcal/mol)
-55
-45
11
37
11.1
11.1
70
230
10.7
10.7
a Error in ∆Gq is (0.1 kcal/mol. b Change in width at half-height
relative to the static spectrum at -110 °C. c From slow exchange
approximation, k ) π∆ν.
Table 6. Rate Constants for Ethylene Exchange
complex
rate constant (M-1 s-1
)
1c (R ) H, R′ ) i-Pr)
8200
45
560
2600
2c (R ) CH3, R′ ) i-Pr)
3c (R,R ) An, R′ ) i-Pr)
6c (R,R ) An, R′ ) Me)
mol. Using a more sterically encumbered model, an energy
difference between the â-agostic isopropyl intermediate and the
propylene-hydride complex is reported to be only 0.7 kcal/
mol, indicating that rearrangement should be incredibly facile.
Although the predicted energy barriers to â-hydride elimination
for both model systems are significantly different from one
another, these results suggest that the overall barrier to inter-
conversion between i-17/n-17 above may be substantially less
than the upper limit of 10.7 kcal/mol determined by line-shape
analysis.
Chain Transfer. Slow rates of chain transfer relative to
propagation distinguish these Pd- and Ni-based R-diimine
systems from most other late metal catalysts (Scheme 1B). Chain
transfer may occur by exchange of bound olefinic polymer with
free ethylene from an olefin hydride species (Scheme 1B(i)).
Alternatively, Ziegler et al. have suggested based on DFT
calculations that chain transfer occurs by concerted â-hydride
transfer from the growing polymer chain to bound ethylene.51,52
Subsequent exchange of the unsaturated polymer with ethylene
must then occur from an alkyl olefin intermediate rather than
from an olefin hydride species (Scheme 1B(ii)).
Figure 9. Free energy diagram for interconversion of i-17 and n-17.
assumed to be low relative to â-hydride elimination and
reinsertion (i-17 f n-17), kH′ will be equal to one-half the
rate constant for conversion of i-17 to n-17 (since half the time
n-17/n-17′ could return with no site exchange of H′). A barrier
for conversion (∆Gconvq) of 10.7 kcal/mol can be calculated in
this case by simply doubling kH′ (Table 5). However, the
assumption of a low n-17fn-17′ (and n-17′ f n-17) barrier
relative to i-17 f n-17 conversion may be incorrect. A value
q
of 9.2 kcal/mol can be estimated for ∆Grot based on results
obtained using i-16,48 and ∆G for n-17/i-17 can be estimated
at 1.2-1.3 kcal/mol from results previously described here.
As shown in Figure 9, assuming the barrier to conversion of
i-17 to n-17 to be less than n-17 to n-17′, the predicted barrier
q
for conversion of i-17 to i-17′ is ∆G + ∆Grot ) 1.3 kcal/mol
+ 9.2 kcal/mol ) 10.5 kcal/mol, which is quite close to the
observed barrier of 10.7 kcal/mol. Thus, based on this analysis,
the magnitude of the i-17 to n-17 barrier is not certain. This
barrier cannot be greater than 10.7 kcal/mol but may in fact be
substantially less. If the barrier is significantly less than 10.7
kcal/mol, this implies that a 1,2-migration of Pd between the R
and â carbons in a â-agostic alkyl complex is much more facile
than migration to the carbons adjacent to CR and Câ.
To assess the influence of steric bulk on the associative
displacement of ethylene from the coordination sphere of these
complexes (as in the proposed chain transfer mechanism shown
in Scheme 1B(i)), the rate of exchange of bound ethylene with
free ethylene (Scheme 8) was monitored for ethylene complexes
1
1c-3c and 6c by H NMR line-broadening experiments at
-85 °C (Table 6, rate constants). Line widths (w) were measured
at half-height in units of hertz for complexed ethylene and were
corrected for line widths in the absence of exchange (wo). The
exchange rates were determined from the standard equation for
the slow exchange approximation: π(w - wo)/[C2H4], where
[C2H4] is the concentration of free ethylene.
The process of interconversion between â-agostic n-propyl
and â-agostic isopropyl-palladium species has been modeled
by Morokuma, et al. (Figure 10).56,57 As previously mentioned,
the energy difference between the two agostic intermediates is
predicted to be ca. 0.5 kcal/mol. The barrier to â-hydride
elimination from each agostic intermediate is reported to be only
0.2-0.8 kcal/mol above the energy of the resulting propylene
hydride species.56 â-Hydride elimination from the isopropyl
species of an unsubstituted model is predicted to give an olefin
hydride species that is 7.4 kcal/mol higher in energy, so the
barrier to this process can be estimated to be less than 8.2 kcal/
The exchange rate of bound ethylene is clearly proportional
to the concentration of free ethylene, which indicates that
exchange takes place through an associative mechanism. As the
size of the ligand backbone substituents is increased, the ligand
aryl rings are forced to a greater extent into the plane
perpendicular to the square plane of the complex. This causes

6696 J. Am. Chem. Soc., Vol. 122, No. 28, 2000
Tempel et al.
2the isopropyl groups to block the axial coordination sites of
the metal center more effectively. As coordination of a second
equivalent of ethylene becomes less favored, the rate of
exchange is observed to decrease in the order R ) H > R,R )
An > R ) CH3. This is in agreement with experiment where
the Mn’s of the polyethylenes made using the same three
catalysts are observed to increase in the same order. The rate
of exchange of ethylene also increases as the size of the
substituents on the aryl rings decreases from isopropyl to methyl
(complex 3c to complex 6c in Table 6), indicating the increased
effectiveness of the isopropyl groups relative to methyl at
blocking the axial coordination sites in these systems.
It is also significant to note that ethylene exchange in (phen)-
Pd(CH3)(C2H4)+ is too fast to be measured by NMR techniques,
even at -100 °C. These systems dimerize ethylene, indicating
that the rate of chain transfer is much greater than propagation,
consistent with the complete absence of axial bulk in these
systems.
with chain transfer occurring via associative exchange from a
palladium olefin hydride intermediate as originally proposed,
where the transition state places both incoming and outgoing
olefins in axial positions. It is also consistent with â-hydrogen
transfer to monomer in the resting state, since Ziegler’s
calculations suggest that the transition state for this process
resembles a bis-olefin hydride with both olefins occupying axial
positions.
(7) Decomposition of (ArN∧NAr)Pd(CH3)(L)+ species can
occur by C-H activation of ortho isopropyl and tert-butyl
substituents after loss of L to generate CH4 and metallacycles.
It is not clear whether such decomposition pathways are
responsible for catalyst decay.
Experimental Section
General Considerations. All manipulations of compounds were
performed using standard high-vacuum, Schlenk, or drybox techniques.
Argon was purified by passage through columns of BASF R3-11
catalyst (Chemalog) and 4 Å molecular sieves. 1H and 13C NMR
chemical shifts were referenced to residual 1H NMR signals and to the
13C NMR signals of the deuterated solvents, respectively, unless
otherwise noted. NMR probe temperatures were measured using an
external anhydrous methanol sample. All coupling constants are reported
in hertz. Elemental analyses were performed by Atlantic Microlab Inc.
of Norcross, GA, or by Oneida Research Services of Whitesboro, NY.
Materials. All solvents were deoxygenated and dried via passage
over a column of activated alumina.71 Dichlorofluoromethane-d (CDCl2F)
was prepared according to the literature.72 Dichlorofluoromethane-d and
methylene chloride-d2 were dried over CaH2 or P2O5, degassed by
repeated freeze-pump-thaw cycles, vacuum-transferred, and stored
over 4 Å molecular sieves. Polymer grade ethylene and CP grade
propylene were used without further purification for preparative
scale polymerizations and NMR experiments. The R-diimine ligands
(except for 2-C6H4(t-Bu)NdC(R)C(R)dN-2-C6H4(t-Bu), R ) Me and
R,R ) An) were prepared according to the literature,50,58-60 and (2,6-
C6H4(i-Pr)2-NdC(R)C(R)dN-2,6-C6H4(i-Pr)2)PdMe250 has previously
been prepared and characterized. Our group has earlier reported the
preparation of dimethyl complexes (2,6-C6H4(i-Pr)2NdC(R)C(R)dN-
2,6-C6H4(i-Pr)2)PdMe2 (R ) H and R ) Me), and their ether complexes
1a and 2a.18 In addition, characterization of the ethylene, propylene,
and butene adducts of 1a and 2a and details of the resulting insertion
products were reported in the Supporting Information of this same
reference.18 (COD)PdMeCl (COD ) 1,5-cyclooctadiene),61 (COD)-
PdMe2,62 (pyridazine)PdMe2,64 NaBAr′4,73,74 and H(OEt2)2BAr′4 (Ar′ )
3,5-C6H3(CF3)2)75,76 were prepared as described in the literature.
Synthesis of Ligands. (a) 2-C6H4(t-Bu)NdC(Me)C(Me)dN-2-
C6H4(t-Bu). A flask was charged with 2-tert-butylaniline (5.00 mL,
32.1 mmol) and 2,3-butanedione (1.35 mL, 15.4 mmol). Addition of
10 mL of methanol and ca. 1 mL of formic acid gave formation of a
yellow precipitate within 30 s. A yellow solid was collected by filtration
after stirring for 19 h. This solid was dissolved in diethyl ether, and
the solution was stirred over Na2SO4, filtered, condensed, and placed
into a -30 °C freezer for crystallization. Large, bright yellow crystals
were collected via filtration and dried under vacuum. After isolating a
second crop of crystals, an overall yield of 86% (4.6 g) was obtained.
1H NMR (CDCl3, 300 MHz) δ 7.43 (dd, 2H, J ) 7.85, 1.33 Hz), 7.20
(ddd, 2H, J ) 7.4, 7.2, 1.4 Hz), 7.09 (ddd, 2H, J ) 7.6, 7.6, 1.4 Hz),
These observations are not only consistent with the associative
exchange mechanism 1B(i), but they are also fully consistent
with the Ziegler â-H transfer to monomer mechanism 1B(ii),
since the transition state for this process effectively places two
olefinic moieties in axial sites (see 14q, Scheme 1B(ii)).
Summary
Pd(II) and Ni(II) complexes of aryl-substituted R-diimines
bearing bulky ortho substituents polymerize ethylene to high
molecular weight, highly branched polyethylene and R-olefins
to polymers with unique microstructures bearing fewer branches
than expected. This initial discovery18 was a significant advance
in the development of late metal catalysts for olefin polymer-
ization and has resulted in much increased activity in this area.
The studies described here have exposed numerous mechanistic
details concerning these Pd(II) catalysts. The key findings are
as follows:
(1) The catalyst resting states are (N∧N)Pd(R)(olefin)+
complexes. Turnover frequencies are controlled by the rates of
migratory insertions of these species, and thus the rate of chain
growth is independent of olefin concentration.
(2) Increasing the bulk of the ortho substituents on the aryl
groups (as well as on the backbone carbons) lowers the barrier
to insertions and raises the TOF. Presumably this arises from
greater relief of steric crowding in the transition state for
insertion relative to the ground state for the bulkier substituents.
(3) The intermediates formed upon migratory insertion of
(N∧N)Pd(alkyl)(olefin)+ complexes (catalyst resting states) are
dynamic, â-agostic (N∧N)Pd(R)+ complexes.
(4) Branching in the polyethylene is a result of metal
migration along the chain in the cationic â-agostic palladium-
alkyl complexes via repetitive â-elimination/reinsertion se-
quences.
(5) Chain walking can occur in the palladium alkyl olefin
resting states, but these species do not undergo â-hydride
elimination/reinsertion reactions. Rather, the process involves
(reversible) loss of ethylene to yield cationic â-agostic palla-
dium-alkyl species which undergo facile â-elimination/re-
insertion reactions. Judging from the model systems studied
here, the difference in barriers to 1,2 chain isomerization in the
resting state and migratory insertion are ca. 2.5-3.0 kcal/mol.
Thus, an average of ca. 100 1,2-migrations occur at 25 °C prior
to insertion. This clearly explains why polyethylene produced
from palladium catalysts is so highly branched.
(71) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518-1520.
(72) Siegel, J. S.; Anet, F. A. L. J. Org. Chem. 1988, 53, 2629-2630.
(73) Nishida, H.; Takada, N.; Yoshimura, M.; Sonoda, T.; Kobayashi,
H. Bull. Chem. Soc. Jpn. 1984, 57, 2600-2604.
(74) Caution: Preparation of the Grignard reagent from 3,5-(CF3)2C6H3-
Br can result in explosions. This material is commercially available from
Boulder Scientific.
(75) Brookhart, M.; Grant, B.; Volpe Jr., A. F. Organometallics 1992,
11, 3920-3922.
(76) The chemical shift originally reported for the oxonium proton in
H(OEt2)2BAr4′ (δ 11.1) is incorrect; this peak is caused by the presence
of H2O in the sample. The correct chemical shift is δ 16.7 (9 peaks, J )
1.5 Hz, 1H).
(6) The rate of associative ethylene exchange in (ArN∧NAr)-
Pd(alkyl)(C2H4)+ complexes decreases with increasing bulk of
ortho and backbone substituents. This observation is consistent

Pd(II)-R-Diimine-Catalyzed Olefin Polymerizations
J. Am. Chem. Soc., Vol. 122, No. 28, 2000 6697
6.52 (dd, 2H, J ) 7.7, 1.4 Hz), 2.22 (s, 6H), 1.36 (s, 18H). 13C{1H}
NMR (CDCl3, 75 MHz) δ 167.0, 149.5, 139.6, 126.62, 126.58, 124.2,
119.5, 35.32, 29.7, 16.62. Anal. Calcd for C24H32N2: C, 82.71; H, 9.25;
N, 8.04. Found: C, 82.69; H, 9.21; N, 8.01.
temperature for 1 h. The ether solution was filtered into another Schlenk
flask, and the remaining product was extracted with CH2Cl2, leaving
behind a significant amount of Pd(0). The solvents were removed under
vacuum, and the residue was washed with pentane until the washes
were colorless (5 × 20 mL). Recrystallization from ether layered
with pentane (ca. 2:1), filtration, and drying in vacuo gave 0.376 g
(0.717 mmol) of green microcrystals (two crops, 37.4% yield). 1H NMR
(CDCl3, 300 MHz) δ 8.00 (d, 2H, J ) 8.4 Hz, An: Hp), 7.41 (dd, 2H,
J ) 8.3, 7.2 Hz, An: Hm), 7.23-7.19 (m, 6H, 6 Haryl), 6.71 (d, 2H,
J ) 7.2 Hz, An: Ho), 2.29 (s, 12H, 2,6-C6H3(Me)2), 0.02 (s, 3H, PdMe2).
Anal. Calcd for C30H30N2Pd: C, 68.63; H, 5.76; N, 5.34. Found: C,
68.64; H, 5.82; N, 5.32.
(b) 2-C6H4(t-Bu)NdC(An)C(An)dN-2-C6H4(t-Bu). A flask was
charged with 2-tert-butylaniline (3.00 mL, 19.2 mmol) and acenaph-
thenequinone (1.71 g, 9.39 mmol). The reagents were partially dissolved
in 50 mL of methanol (acenaphthenequinone was not completely
soluble), and formic acid (1-2 mL) was added. An orange solid formed
and was collected via filtration after stirring for 19 h. This solid was
dissolved in CH2Cl2, the solution was stirred over Na2SO4 and filtered,
and the product reprecipitated upon cooling (3.51 g, 84.1%). 1H NMR
(CDCl3, 300 MHz) δ 7.85 (d, 2H, J ) 8.2 Hz, An: Hp), 7.59 (m, 2H,
Ar: Hm), 7.37 (dd, 2H, J ) 7.8, 7.6 Hz, An: Hm), 7.27 (m, 4H,
Ar: Hm and Hp), 7.01 (m, 2H, Ar: Ho), 6.91 (d, 2H, J ) 7.2 Hz,
An: Ho), 1.46 (s, 18H, C(CH3)3). 13C{1H} NMR (CDCl3, 75 MHz) δ
159.9, 150.6, 141.8, 139.3, 131.2, 129.1, 128.8, 127.7, 126.83, 126.75,
124.6, 123.9, 119.0, 35.5, 29.8. Anal. Calcd for C32H32N2: C, 86.45;
H, 7.25; N, 6.30. Found: C, 86.31; H, 7.29; N, 6.22.
Synthesis of Catalysts. (a) Spectral Data for the BAr′4 Counter-
1
ion. The H and 13C NMR resonances of the [B{3,5-C6H3(CF3)2}4]-
anion in CD2Cl2 were essentially invariant for different complexes, and
temperatures and are not repeated in all of the spectroscopic data for
each of the cationic complexes: 1H NMR (CD2Cl2) δ 7.7 (s, 8, Ho),
7.5 (s, 4, Hp); 13C{1H} NMR (CD2Cl2) δ 162.2 (q, 1JCB ) 49.8, Cipso),
2
1
135.2 (Co), 129.3 (q, JCF ) 31.7, Cm), 125.0 (q, JCF ) 272.5, CF3),
117.9 (Cp).
Synthesis of Dimethyl Complexes. (a) (ArNdC(An)C(An)dNAr)-
Pd(CH3)2 (Ar ) 2,6-C6H3(i-Pr)2). A Schlenk flask was charged with
0.2658 g (1.003 mmol) of (COD)PdMeCl and cooled to -30 °C. Ether
(25 mL) was added via syringe followed by MeLi (0.75 mL of a
1.6 M solution in ether, ca. 1.2 equiv), and the mixture was stirred
with warming to -10 °C. At this stage white LiCl had precipitated out
and tert-butyl bromide (0.05 mL, 0.43 mmol) was added to consume
any excess MeLi. The ligand, 2,6-C6H3(i-Pr)2NdC(An)C(An)dN-2,6-
C6H3(i-Pr)2 (0.5051 g, 1.002 mmol), was added as a cooled (-10 °C)
slurry in 25 mL of diethyl ether, and the resulting mixture was stirred
for about 1 h as the flask warmed to 25 °C. The product was transferred
to another flask using ca. 125 mL of Et2O, the solvent was removed
under vacuum, and the residue was washed with pentane until the
washes were colorless (5 × 10 mL). The solid was dried overnight in
vacuo to give 0.289 g (0.454 mmol) of a green powder (45.3% yield).
The 1H and 13C NMR spectroscopic data for this compound have been
reported.60 Anal. Calcd for C38H46N2Pd: C, 71.63; H, 7.28; N, 4.40.
Found: C, 71.60; H, 7.24; N, 4.48.
(b) (ArNdC(An)C(An)dNAr)Pd(CH3)2 (Ar ) 2-C6H4(t-Bu)). A
Schlenk flask was charged with 0.097 g (0.396 mmol) of (COD)PdMe2
in a glovebox equipped with a -30 °C freezer, taking precautions to
keep the precursor and glassware as cold as possible at all times. The
flask was brought out of the glovebox and rapidly cooled to -40 °C
in a dry ice/2-propanol bath. The solid was dissolved in 15 mL of
ether, and the diimine 2-C6H4(t-Bu)NdC(An)C(An)dN-2-C6H4(t-Bu)
(0.178 g, 0.400 mmol) was cannulated onto the stirring solution as a
slurry in 25 mL of ether. The reaction was warmed to 0 °C, and stirring
was continued for approximately 2 h. The reaction flask was stored
at -30 °C overnight, resulting in formation of a green precipitate
which was isolated via filtration and dried in vacuo. A second green
solid resembling the first was collected upon evaporation of solvent
under vacuum from the supernatant. The solids were combined,
washed with hexanes (2 × 10 mL), and dried in vacuo to yield 0.197
g (85.6%) of the desired product (two isomers, presumbly from cis
and trans ligand conformations, in a ratio of ca. 97:3). 1H NMR
(CD2Cl2, 300 MHz) major: δ 8.01 (d, 2H, J ) 8.3 Hz, An: Hp), 7.66
(m, 2H, Haryl), 7.43-7.34 (m, 6H, An: Hm, 2Haryl), 7.05 (m, 2H, Haryl),
6.54 (d, 2H, J ) 7.5 Hz, An: Ho), 1.45 (s, 18H, C(CH3)3), -0.05
(s, 6H, PdMe2); minor: δ 7.90 (d, 2H, J ) 8.3 Hz), 7.56 (m, 2H),
obscured (2Haryl), 7.25 (m, 4H), 6.93 (m, 2H), 6.85 (d, 2H, J ) 7.1
Hz), 1.37 (s, 18H), obscured (6H, PdMe2). 13C{1H} NMR (CD2Cl2,
75 MHz) δ 167.7 (NdC-C-N), 147.2 (Ar: Cipso), 143.7, 131.6 (An:
2 quaternary C), 140.6 (NdC(C)-C(C)dN), 128.24 (Ar: C-C(CH3)3),
130.22, 128.84, 128.43, 127.52, 126.68, 125.23, 122.49, 36.3 (C(CH3)3,
31.5 (C(CH3)3, -6.29 (PdMe, JCH ) 128.1 Hz) (minor isomer not
detected). Anal. Calcd for C34H38N2Pd: C, 70.27; H, 6.59; N, 4.82.
Found: C, 69.98; H, 6.87; N, 4.69.
(b) [(ArNdC(An)C(An)dNAr)Pd(CH3)(OEt2)]BAr′4 (Ar ) 2,6-
C6H3(i-Pr)2), 3a. A Schlenk flask was charged with 0.195 g (0.306
mmol) of (2,6-C6H3(i-Pr)2NdC(An)C(An)dN-2,6-C6H3(i-Pr)2)Pd(CH3)2
and 0.309 g (0.305 mmol) of H(OEt2)2BAr′4. The flask was cooled to
-78 °C, and 10 mL of ether were added via syringe. The stirring
suspension was warmed to 0 °C in an ice bath, resulting in formation
of a dark red solution. The solution was filtered, and the ether was
removed under vacuum to give an orange powder which was further
dried in vacuo, yielding 0.425 g of product (89% yield). 1H NMR
(CD2Cl2, -30 °C, 300 MHz): δ 8.08 (d, 1H, J ) 8.4 Hz, An: Hp),
8.04 (d, 1H, J ) 8.4 Hz, An: Hp′), 7.36-7.34 (m, 8H, An: Hm, Hm′;
6 Haryl), 6.64 (d, 1H, J ) 7.3 Hz, An: Ho), 6.37 (d, 1H, J ) 7.3 Hz,
An: Ho′), 3.36 (q, 4H, J ) 7.0 Hz, O(CH2CH3)2), 3.16 (2 septets, 2H
each, J ) 6.8 Hz, CHMe2, C′HMe2), 1.35 (d, 6H, J ) 6.9 Hz,
CHMeMe′), 1.32 (d, 6H, J ) 6.9 Hz, CHMeMe′), 1.26 (t, 6H, J ) 6.9
Hz, O(CH2CH3)2), 0.86 (2d, 6H each, J ) 6.5 Hz, C′HMeMe′), 0.67
(s, 3H, PdMe). 13C{1H} NMR (CD2Cl2, -30 °C, 75 MHz): δ 174.5
and 168.6 (NdC-C′dN), 140.9 and 139.4 (Ar, Ar′: Cipso), 138.2 and
137.1 (Ar, Ar′: Co), 144.7, 132.5, 131.8, 130.5, 129.2, 128.7, 128.2,
128.0, 127.8, 125.4, 125.2, 124.9, 124.6, and 124.4 (An: 4 quaternary
C; An: Co,Co′, Cp, Cp′, Cm, Cm′; Ar, Ar′: Cm, Cp), 71.7 (O(CH2CH3)2),
28.7 and 28.4 (CHMe2, C′HMe2), 23.86, 23.56, 22.59, and 22.45
(CHMeMe′, C′HMeMe′), 14.7 (O(CH2CH3)2), 9.6 (PdMe). Anal. Calcd
for C73H65N2BF24OPd: C, 56.22; H, 4.20; N, 1.80. Found: C, 56.46;
H, 4.14; N, 1.82.
(c) [(ArNdC(An)C(An)dNAr)Pd(CH3)(OEt2)]BAr′4 (Ar ) 2-C6H4-
(t-Bu)), 5a. This compound was synthesized using a procedure identical
to that described above for 3a. The dimethyl complex, 2-C6H4(t-Bu)Nd
C(An)C(An)dN-2-C6H4(t-Bu))Pd(CH3)2 (0.106 g, 0.182 mmol), and
H(OEt2)2BAr′4 (0.185 g, 0.183 mmol) were combined in a Schlenk
flask. Ether (5 mL) was added, and the reaction was warmed to 0 °C.
Isolation via removal of the solvent under vacuum yielded 0.260 g of
a dark red solid (0.173 mmol, 94.8%). Importantly, only a single isomer
is observed by NMR spectroscopy. Each of several attempts at preparing
this compound was accompanied by 5-15% intramolecular C-H
1
activation to give the six-membered palladacycle. H NMR (CD2Cl2,
-80 °C, 300 MHz) δ 8.09 (d, 1H, J ) 9.0 Hz, An: Hp), 8.06 (d,
1H, J ) 9.3 Hz, An: Hp′), 7.72-7.32 (m, 8H, An: Hm, Hm′; 6 Haryl),
7.04 (dd, 2H, J ) 7.2, 7.2 Hz, 2 Haryl), 6.54 (d, 1H, J ) 7.5 Hz,
An: Ho), 6.24 (d, 1H, J ) 7.5 Hz, An: Ho′), 3.37 (q, 4H, J ) 6.9 Hz,
O(CH2CH3)2), 1.43 (s, 9H, C(CH3)3), 1.38 (s, 9H, C′(CH3)3), 1.09
(t, 6H, J ) 6.9 Hz, O(CH2CH3)2), 0.80 (s, 3H, P,Me).
(d) [(ArNdC(An)C(An)dNAr)Pd(CH3)(OEt2)]BAr′4 (Ar ) 2,6-
C6H3(Me)2), 6a. Following the above procedure for protonation
described for catalysts 3a and 5a in this case yielded a clean mixture
of two products that could not be separated (ca. 4:1 desired/side
product). The side product was not clearly identified but may consist
of a methyl-bridged dimeric monocation. A second attempt was made
under slightly modified conditions. A Schlenk flask was charged with
0.578 g (0.571 mmol) of H(OEt2)2BAr′4 and cooled to -78 °C and
8 mL of ether was added via syringe. The flask was carefully warmed
(c) (ArNdC(An)C(An)dNAr)Pd(CH3)2 (Ar ) 2,6-C6H3(Me)2). A
Schlenk flask was charged with 0.4150 g (1.916 mmol) of [(pyridazine)-
Pd(CH3)2]n and 0.7520 g of ArNdC(An)C(An)dNAr (1.936 mmol).
The flask was cooled to 0 °C, 50 mL of diethyl ether was added via
syringe, and the mixture was stirred at 0 °C for 2 h and at room

6698 J. Am. Chem. Soc., Vol. 122, No. 28, 2000
Tempel et al.
just until the acid was dissolved. A second flask, charged with a
suspension of 0.300 g (0.571 mmol) of (2,6-C6H3(Me)2NdC(An)C-
(An)dN-2,6-C6H3(Me)2)Pd(CH3)2 in 10 mL of ether, was also cooled
to -78 °C. Most of this suspension was transferred via cannula into
the flask containing the dissolved acid. Additional ether (4 mL) was
added to the residual dimethyl complex; this suspension was cooled
and added as well. The reaction mixture became a bright orange/red
solution with a small amount of dark solid after stirring at -78 °C for
ca. 15 min. The solution was rapidly filtered into another cooled Schlenk
flask (-78 °C), and the solvent was removed under reduced pressure
to give 0.635 g (0.439 mmol) of a light orange powder (76.5% yield).
1H NMR (CD2Cl2, -80 °C, 300 MHz) δ 8.09 (d, 1H, J ) 9.0 Hz, An:
Hp), 8.05 (d, 1H, J ) 9.3 Hz, An: Hp′), 7.46-7.17 (m, 8H, An: Hm,
Hm′; 6 Haryl), 6.67 (d, 1H, J ) 7.5 Hz, An: Ho), 6.39 (d, 1H, J ) 7.5
Hz, An: Ho′), 2.23 (q, 4H, J ) 6.3 Hz, O(CH2CH3)2), 2.23 and 2.19
(2 s, 6H each, 2,6-C6H3(Me)2 and 2,6-C6H3(Me′)2), 1.48 (t, 6H, J )
1.5 Hz), O(CH2CH3)2), 0.53 (s, 3H, PdMe). 13C{1H} NMR (CD2Cl2,
-80 °C, 300 MHz) δ 174.4 and 168.5 (NdC-C′dN), 144.9, 143.2
(Ar, Ar′: Cipso), 141.9 (2 C, Ar, Ar′: Co), 132.6, 132.0, 130.2, 129.1,
128.9, 128.1, 127.9, 127.2, 126.7, 126.6, 124.9, 124.7, 124.6, 124.3,
71.8 (O(CH2CH3)2), 17.5 and 17.3 (2,6-C6H3(Me)2, 2,6-C6H3(Me′)2),
16.4 (O(CH2CH3)2), 9.5 (PdMe). Anal. Calcd for C65H49N2BF24OPd:
C, 53.94; H, 3.41; N, 1.94. Found: C, 54.22; H, 3.46; N, 2.05.
(C′HMe2), 23.8, 23.7, 23.6, and 23.0 (CHMeMe′, C′HMeMe′), 21.5
and 20.0 (NdC(Me)-C′(Me)dN), 9.8 (JCH ) 136.0 Hz, PdMe). Anal.
Calcd for (C90H98BClF24N4Pd2): C, 55.41; H, 5.06; N, 2.87. Found:
C, 55.83; H, 5.09; N, 2.63.
(g) [[(ArNdC(H)C(H)dNAr)PdMe]2(µ-Cl)]BAr′4 (Ar ) 2,6-
C6H3(i-Pr)2). The above procedure was followed with one exception.
The removal of CH2Cl2 in vacuo yielded a product that was partially
an oil. Dissolving the compound in Et2O and then removing the
1
Et2O in vacuo yielded a microcrystalline red solid (85.5%). H NMR
(CD2Cl2, 400 MHz) δ 8.20 and 8.09 (s, 2 each, NdC(H)-C′(H)dN),
7.37 (t, 2, J ) 7.73 Hz, Ar: Hp), 7.28 (d, 4, J ) 7.44 Hz, Ar: Hm),
7.24 (t, 2, Ar′: Hp), 7.16 (d, 4, J ) 7.19 Hz, Ar′: Hm), 3.04 (septet, 4,
J ) 6.80 Hz, CHMe2), 2.93 (septet, 4, J ) 6.80 Hz, C′HMe2), 1.26
(d, 12, J ) 6.79 Hz, CHMeMe′), 1.14 (d, 12, J ) 6.83 Hz, CHMeMe′),
1.11 (d, 12, J ) 6.80 Hz, C′HMeMe′), 1.06 (d, 12, J ) 6.79 Hz,
C′HMeMe′), 0.74 (s, 6, PdMe); 13C{1H} NMR (CD2Cl2, 100 MHz) δ
166.0 (JCH ) 180.4 Hz, NdC(H)), 160.8 (JCH ) 179.9 Hz, NdC′(H)),
143.5 and 143.0 (Ar, Ar′: Cipso), 139.8 and 138.9 (Ar, Ar′: Co), 129.3
and 128.5 (Ar, Ar′: Cp), 124.3 and 123.7 (Ar, Ar′: Cm), 29.2 and 28.9
(CHMe2, C′HMe2), 24.5, 24.1, 23.0, and 22.5 (CHMeMe′, C′HMeMe′),
10.3 (PdMe). Anal. Calcd for (C86H90BClF24N4Pd2): C, 54.52; H, 4.97;
N, 2.96. Found: C, 54.97; H, 4.72; N, 2.71.
Intramolecular C-H Activation. (a) [ArNdC(H)C(H)dN-2,6-
C6H3-i-Pr,CHMeCH2Pd(OEt2)]BAr′4 (Ar ) 2,6-C6 H3(i-Pr)2), 1a′.
A 700 µL CD2Cl2 solution of [(ArNdC(H)C(H)dNAr)PdMe(OEt2)]-
BAr′4 (68.4 mg, 0.0477 mmol) was allowed to stand at 25 °C for several
hours and then at -30 °C overnight. Such highly concentrated solutions
(e) [(ArNdC(Me)C(Me)dNAr)Pd(CH3)(NCMe)]BAr′4 (Ar )
2-C6H4(t-Bu)), 4b. A Schlenk flask was charged with 0.180 g (0.679
mmol) of (COD)PdMeCl and 0.602 g (0.679 mmol) of NaBAr′4. After
cooling the flask to -40 °C, 25 mL of CH2Cl2 and 25 mL of NCMe
were added by syringe. The reaction was stirred with warming to -20
°C, resulting in formation of [(COD)Pd(Me)(NCMe)]BAr′4 accompa-
nied by precipitation of NaCl. After allowing the precipitate to settle,
the solution of NCMe adduct was cannula filtered into another Schlenk
flask cooled to 0 °C and containing a suspension of 0.237 g (0.679
mmol) of 2-C6H4(t-Bu)NdC(Me)C(Me)dN-2-C6H4(t-Bu) in 20 mL of
acetonitrile. The reaction was stirred at room temperature overnight,
and the solvents were removed under vacuum to give a yellow oil.
The oil was redissolved in 15 mL of CH2Cl2 and 15 mL of hexanes.
Removal of solvents in vacuo now gave 0.791 g (0.576 mmol) of a
bright yellow powder (84.8% yield). Two isomers are observed in the
1H NMR spectrum in an approximate 9:1 ratio, but the aryl shifts of
the minor isomer are obscured. 1H NMR (CDCl3, 300 MHz) δ 7.58-
7.50 (m, 2H, Haryl), 7.27 and 7.21 (m, 2H each, Haryl), 6.59 (m, 2H,
1
of the resulting metallacycle were stable for hours at 25 °C. H NMR
(CD2Cl2, 400 MHz, 41 °C) δ 8.17 (s, 2, NdC(H)-C′(H)dN), 7.5-
7.0 (m, 6, Haryl), 3.48 (q, 4, J ) 6.88 Hz, O(CH2CH3)2), 3.26 (septet,
1, J ) 6.49 Hz, CHMe2), 3.08 (septet, 1, J ) 6.86 Hz, C′HMe2), 2.94
(septet, 1, J ) 6.65 Hz, C′′HMe2), 2.70 (dd, 1, J ) 6.67 Hz, 0.90,
CHMeCHH′Pd), 2.43 (dd, 1, J ) 7.12 Hz, 4.28, CHMeCHH′Pd), 2.23
(br m, 1, CHMeCH2Pd), 1.54 (d, 3, J ) 6.86 Hz, CHMeCH2Pd), 1.43
(d, 3, J ) 6.79 Hz, C′′HMeMe′), 1.40 (d, 3, J ) 7.12 Hz, CHMeMe′),
1.37 (d, 3, J ) 6.95 Hz, C′HMeMe′), 1.27 (d, 6, J ) 6.79 Hz,
C′HMeMe′, C′′HMeMe′), 1.12 (d, 3, J ) 6.54 Hz, CHMeMe′), 1.23
(br m, 6, O(CH2CH3)2), 0.21 (CH4); 13C{1H} NMR (CD2Cl2, 100 MHz,
41 °C) δ 162.5 (JCH ) 181.5 Hz, NdC(H)), 161.2 (JCH ) 178.4 Hz,
NdC′(H)), 145.8 and 144.5 (Ar, Ar′: Cipso), 141.6, 140.7, 140.3, and
138.8 (Ar, Ar′: Co, Co′), 131.6 and 129.8 (Ar, Ar′: Cp), 128.1, 127.6,
125.2, and 124.5 (Ar, Ar′: Cm, Cm′), 72 (br, O(CH2CH3)2), 43.2
(CHMeCH2Pd), 40.5 (CHMeCH2Pd), 29.5,. 29.1 and 28.8 (CHMe2,
C′HMe2, C′′HMe2), 26.2 (br), 25.3, 25.2, 25.1, 24.5 (br), 23.3 and
22.1 (CHMeMe′, C′HMeMe′, C′′HMeMe′, CHMeCH2Pd), 15.5 (br,
O(CH2CH3)2), -14.8 (CH4).
(b) Ethylene Polymerization Catalyzed by [ArNdC(H)C(H)d
N-2,6-C6H3-i-Pr,CHMeCH2Pd(OEt2)]BAr′4 (Ar ) 2,6-C6H3-(i-Pr)2)
(1a′). Addition of ethylene to a CD2Cl2 solution of this compound
resulted in displacement of Et2O to give the corresponding ethylene
adduct 1c′. Warming of the ethylene adduct in the presence of excess
ethylene resulted in branched polymer formation: 1H NMR (CD2Cl2)
δ 1.3 ppm (CH2)n, 0.9 ppm (CH3). Rates of initiation were significantly
slower than rates of propagation.
Haryl), 2.16 and 2.14 (minor: 2.20 and 2.18) (s, 3H each, NdC(CH3)-
C(CH3′)dN), 1.69 (minor: 1.71) (s, 3H, NCMe), 1.43 and 1.41
(minor: 1.383 and 1.380) (s, 9H each, C(CH3)3), 0.57 (minor: 0.62)
(s, 3H, PdMe). 13C{1H} NMR (CD2Cl2, 300 MHz) (minor product
not reported) δ 179.7 and 172.2 (NdC-C′dN), 144.8 and 144.0 (Ar,
Ar′:
Cipso), 140.3 and 139.7 (Ar, Ar′: C-C(Me)3), 129.7, 128.66,
128.64, 128.0, 127.8, 127.7, 122.5, 121.0, 120.9, 36.4, and 35.9 (C(Me)3
and C′(Me)3), 31.9 and 31.1 (C(Me)3 and C′(Me)3), 22.5 and 20.8
(NdC(Me)-C(Me′)dN), 7.7 (PdMe), 1.8 (NCMe). Anal. Calcd for
C73H65N2BF24OPd: C, 51.57; H, 3.67; N, 3.06. Found: C, 52.11;
H, 3.91; N, 3.06.
(f) [[(ArNdC(Me)C(Me)dNAr)PdMe]2(µ-Cl)]BAr′4 (Ar ) 2,6-
C6H3(i-Pr)2). Et2O (25 mL) was added to a mixture of (ArNdC(Me)C-
(Me)dNAr)PdMeCl (0.81 g, 1.45 mmol) and 0.5 equiv of NaBAr′4
(0.64 g, 0.73 mmol) at 25°C. A golden yellow solution and NaCl
precipitate formed immediately upon mixing. The reaction mixture was
stirred for 19 h and then filtered. After the Et2O was removed in vacuo,
the product was washed with 25 mL of hexane. The yellow powder
was then dissolved in 25 mL of CH2Cl2, and the resulting solution was
filtered in order to removed traces of unreacted NaBAr′4. Removal of
CH2Cl2 in vacuo yielded a golden yellow powder (1.25 g, 88.2%): 1H
NMR (CD2Cl2, 400 MHz) δ 7.33 (t, 2, J ) 7.57 Hz, Ar: Hp), 7.27
(d, 4, J ) 7.69 Hz, Ar: Hm), 7.18 (t, 2, J ) 7.64 Hz, Ar: Hp), 7.10
(d, 4, J ) 7.44 Hz, Ar′: Hm), 2.88 (septet, 4, J ) 6.80 Hz, CHMe2),
2.75 (septet, 4, J ) 6.82 Hz, C′HMe2), 2.05 and 2.00 (s, 6 each, Nd
C(Me)-C′(Me)dN), 1.22, 1.13, 1.08, and 1.01 (d, 12 each, J ) 6.61-
6.99 Hz, CHMeMe′, C′HMeMe′), 0.41 (s, 6, PdMe); 13C{1H} NMR
(CD2Cl2, 100 MHz) δ 177.1 and 171.2 (NdC-C′dN), 141.4 and 141.0
(c)[ArNdC(H)C(H)dN-2,6-C6H3-i-Pr,CHMeCH2Pd(H2CdCH2)]-
1
BAr′4 (Ar ) 2,6-C6H3-(i-Pr)2) (1c′). H NMR (CD2Cl2, 400 MHz,
-61 °C) δ 8.25 and 8.23 (NdC(H)-C′(H)dN), 7.55-7.16 (m, 6, Haryl),
4.67 (m, 2, HH′CdCHH′), 4.40 (m, 2, HH′CdCHH′), 2.95 (septet, 1,
J ) 6.30 Hz, CHMe2), 2.80 (septet, 2, J ) 6.36 Hz, C′HMe2 and
C′′HMe2), 2.53 (br m, 1, CHMeCH2Pd), 2.43 (d, 1, J ) 8.16 Hz,
CHMeCHH′Pd), 1.73 (dd, 1, J ) 8.16, 2.84 Hz, CHMeCHH′Pd), 1.45
and 1.19 (d, 3 each, J ) 6.79-6.40 Hz, CHMeMe′), 1.42 (d, 3, J )
7.05 Hz, CHMeCH2Pd), 1.30, 1.30, 1.19, and 0.99 (d, 3 each, J )
6.40-6.65 Hz, C′HMeMe′ and C′′HMeMe′); 13C{1H} NMR (CD2Cl2,
100 MHz, -61 °C) δ 162.7 (JCH ) 179.7 Hz, NdCH), 162.1 (JCH
)
180.9 Hz, NdC′H), 144.7, 141.7, 141.2, 139.2, 137.5, and 137.1 (Ar,
Ar′: Cipso, Co, C′o), 131.0 and 129.0 (Ar, Ar′: Cp), 124.6 and 124.0
(Ar, Ar′: Cm), 92.3 (JCH ) 162.4 Hz, H2CdCH2), 45.1 (CH2Pd), 41.1
(CHMeCH2Pd), 28.9, 28.5, and 28.2 (CHMe2, C′HMe2, C′′HMe2), 26.1,
25.6, 25.1, 24.9, 24.6, 22.9, and 21.4 (CHMeMe′, C′HMeMe′, C′′HMeMe′,
CHMeCH2Pd).
(Ar, Ar′:
Cipso), 138.8 and 138.1 (Ar, Ar′: Co), 128.6 and 127.8
(Ar, Ar′: Cp), 124.5 and 123.8 (Ar, Ar′: Cm), 29.3 (CHMe2), 29.0

Pd(II)-R-Diimine-Catalyzed Olefin Polymerizations
J. Am. Chem. Soc., Vol. 122, No. 28, 2000 6699
(d) [ArNdC(Me)C(Me)dN-C6H4-2-C(CH3)2CH2Pd(OEt2)]BAr′4
(Ar ) 2-C6H4(t-Bu)), 4b′. A solution of 4b in CDCl3 exhibited slow
deactivation to 4b′ at 50 °C and gave the following spectra: 1H NMR
(CDCl3, 300 MHz) δ 7.54 (m, 1H, Haryl), 7.38 (ddd, 1H, J ) 7.7, 7.6,
1.3 Hz, Haryl), 7.32-7.26 (m, 1H, Haryl), 7.20 (2 m, 1H each, Haryl),
6.88 (dd, 1H, J ) 7.9, 1.3, Haryl), 6.62-6.56 (2 m, 1H each, Haryl),
2.55 and 2.20 (s, 3H each, NdC(CH3)-C(CH3′)dN), 2.26 (2 coincident
s, 1H each, C(Me)2CHH′Pd), 1.72 (s, 3H, NCMe), 1.40 (br s, 3H,
CMeMe′CH2Pd), 1.32 (s, 18H, C(Me)3), 1.29 (br s, 3H, CMeMe′CH2-
Pd). 13C{1H} NMR (CDCl3, 75 MHz) δ 173.4 and 171.8 (NdC-C′d
Hz, CHMeMe′), 0.89 (d, 6H, J ) 6.6 Hz, C′HMeMe′), 0.82 (d, 6H,
J ) 6.6 Hz, C′HMeMe′), 0.46 (s, 3H, PdMe).
(b) [(ArNdC(An)C(An)dNAr)Pd(CH3)(CH2dCH2)]BAr′4 (Ar )
2-C6H4(t-Bu)) (5c). Complex 5a used for the kinetic runs contained
15% deactivated palladacycle 5a′ which did not promote migratory
insertion of ethylene. This was accounted for in the measurement of
1
turnovers. H NMR (CD2Cl2, -80 °C, 300 MHz) δ 8.09 (d, 1H, J )
8.7 Hz, An: Hp), 8.05 (d, 1H, J ) 8.7 Hz, An: Hp′), 7.67-7.45 (m,
8H, An: Hm, Hm′; 6 Haryl), 7.07 (m, 1H, Haryl), 6.90 (m, 1H, Haryl), 6.50
(d, 1H, J ) 7.5 Hz, An: Ho), 6.40 (d, 1H, J ) 7.2 Hz, An: Ho′), 4.58
(dm, 4H, J ) 20.1 Hz (d), C2H4), 1.33 (2 overlapping s, 9H each,
C(CH3)3 and C′(CH3)3), 0.52 (s, 3H, PdMe).
N), 148.0 and 145.0 (Ar, Ar′:
Cipso), 140.7 and 140.1 (Ar, Ar′:
C-C(Me)3), 130.9, 128.5, 127.9, 127.51, 127.49, 126.7, 123.6, 120.7,
119.4, 45.9 (CMe2CH2Pd), 37.7 and 35.8 (C(Me)3 and C6H4-2-CMe2-
CH2Pd), 31.9 and 31.0 (CMeMe′CH2Pd), 31.1 (C(Me)3); 22.6 and 20.7
(NdC(Me)-C(Me′)dN), 1.9 (NCMe).
(c) [(ArNdC(An)C(An)dNAr)Pd(CH3)(CH2dCH2CH3)]BAr′4
1
(Ar ) 2,6-C6H3(i-Pr)2) (3c). H NMR (CD2Cl2, -80 °C, 300 MHz)
δ 8.12 (d, 1H, J ) 8.4 Hz, An: Hp), 8.08 (d, 1H, J ) 8.4 Hz,
An: Hp′), 7.57-7.41 (m, 8H, An: Hm, Hm′; 6 Haryl), 6.53 (d, 1H, J )
7.2 Hz, An: Ho), 6.51 (d, 1H, J ) 7.5 Hz, An: Ho′), 5.24 (m, 1H,
CH2dCHCH3), 4.56 and 4.39 (2 br m, 2H, CH2dCHCH3), 3.12-2.92
(4 septets, 1H each, CHMe2, C′HMe2, C′′HMe2, and C′′′HMe2), 1.78
(s, 3H, J ) 6.0 Hz, CH2dCHCH3), 1.39, 1.34-1.29 (2), 0.92, 0.89,
0.89, 0.85, and 0.83 (8 d, 3H each, J ) 6.3-6.9 Hz, CHMeMe′,
C′HMeMe′, C′′HMeMe′, C′′′HMeMe′), 0.58 (s, 3H, PdMe).
(e) [ArNdC(An)C(An)dN-C6H4-2-C(CH3)2CH2Pd(OEt2)]BAr′4
(Ar ) 2-C6H4(t-Bu)), 5a′. Upon sitting at 0 °C for several hours, a
solution of 5a in CH2Cl2 changed in color from red/orange to deep
red. Removal of solvent under vacuum gave a dark red brittle solid
that was scraped into a powder. The product is consistent with
1
intramolecular C-H activation to give 5a′. H NMR (CD2Cl2, 0 °C,
300 MHz) δ 8.17 (d, 1H, J ) 8.2 Hz: An: Hp), 8.10 (d, 1H, J ) 8.2
Hz: An: Hp′), 7.89 (d, 1H, J ) 7.4 Hz, Haryl), 7.70-7.27 (m, 9H, An:
Hp′, Hm, Hm′, Ho′; 5 Haryl), 7.04 (dd, 1H, J ) 7.5, 1.53 Hz, Haryl), 6.59
(d, 1H, J ) 7.2 Hz, Ar: Ho), 3.51-3.32 (br, 4H, O(CH2CH3)2), 2.53
(d, 1H, J ) 8.1 Hz, C(Me)2CHH′Pd), 2.24 (d, 1H, J ) 7.9 Hz,
C(Me)2CHH′Pd), 1.53 (s, 3H, CMeMe′CH2Pd), 1.44 (s, 3H, CMeMe′CH2-
Pd), 1.38 (s, 18H, C(Me)3), 1.33 (t, 6H, O(CH2CH3)2.
Alkyl Agostic Complexes. [(ArNdC(An)C(An)dNAr)Pd(CH-
(CH2-µ-H)(CH3))]BAr′4 (Ar ) 2,6-C6H3(Me)2) (i-17). Synthesis of
the â-agostic isopropyl complex i-16 has been reported.48 The â-agostic
isopropyl complex i-17 was prepared in a similar fashion: In a drybox
under an argon atmosphere, an NMR tube was charged with ca. 15 mg
of 6a (ca. 1 × 10-5 mol). The tube was capped with a rubber septum
and removed from the drybox. After securing the septum with Teflon
tape and Parafilm, the tube was cooled to -78 °C. CDCl2F was added
to the NMR tube via a 22 gauge cannula (∼600-800 µL), and the
septum was sealed with silicon grease and rewrapped with Parafilm.
The tube was shaken and warmed slightly to facilitate dissolution of
the complex. After acquiring a spectrum at -80 °C, 0.95 equiv of
ethylene was added via gastight syringe to the solution of 6a cooled to
-78 °C, and the NMR tube was briefly shaken to completely dissolve
the olefin. The tube was then transferred to the cooled NMR probe,
and migratory insertion of the methyl ethylene complex 6c was followed
at 0 °C. Formation of the desired â-agostic complex was confirmed by
acquiring a spectrum at ca. -115 °C for i-17, which exhibited a broad
triplet at -8.0 ppm, characteristic of the agostic hydrogen of a methyl
group. The spectra acquired in CDCl2F described below were referenced
to the ortho hydrogens of BAr′4- at 7.71 ppm, as the residual proton
shifts from the solvent were partially or completely obscured by aryl
(f) [ArNdC(An)-C(An)dN-C6H4-2-C(CH3)2CH2Pd(CH2dCH2)]-
BAr′4 (Ar ) 2-C6H4(t-Bu)) (5c′). Addition of 20 equiv of ethylene
to 5a′ resulted in displacement of ether by ethylene. No migratory
1
insertion was observed upon standing at 25 °C after 1 day. H NMR
(CD2Cl2, 0 °C, 300 MHz) δ 8.23 (d, 1H, J ) 8.4 Hz, An: Hp), 8.15 (d,
1H, J ) 8.4 Hz, An: Hp′), 8.07 (d, 1H, J ) 7.2 Hz, Haryl), 7.80 (dd,
1H, J ) 7.8, 1.2 Hz, Haryl), 7.72-7.41 (m, 8H, An: Hm, Hm′; 6 Haryl),
6.91 (dd, 1H,
J ) 7.2, 1.8 Hz, Haryl), 7.60 (d, 1H, J )
7.2 Hz, An: Ho), 4.92 and 4.46 (2 m, 2H each, HH′CdCHH′, HH′Cd
CHH-), 2.41 (d, 1H, J ) 9.0 Hz, C(Me)2CHH′Pd), 1.68 (d, 1H, J )
9.0 Hz, C(Me)2CHH′Pd), 1.53 and 1.52 (s, 3H each, C(MeMe′)CH2Pd
and C(MeMe′)2CH2Pd), 1.34 (s, 9H, C(CH3)3), (5.4 (unbound C2H4,
3.42 (q from Et2O), 1.14 (t from Et2O), 0.2 (CH4)).
General Procedure for Variable Temperature NMR Experi-
ments. In a drybox under an argon atmosphere, an NMR tube was
charged with ca. 0.01 mmol of [(ArNdC(R)C(R)dNAr)Pd(Me)(L)]-
BAr′4 (or [(ArNdC(R)C(R)dNAr)Pd-(CH(CH2-µ-H)(CH3))]BAr′4).
The tube was capped with a rubber septum and removed from the
drybox. After securing the septum with Teflon tape and Parafilm, the
tube was cooled to -78 °C. CD2Cl2 was added to the NMR tube via
gastight syringe (700 µL) or CDCl2F was added via a 22 gauge cannula
(∼600-800 µL), and the septum was rewrapped with Parafilm. The
tube was shaken and warmed slightly to facilitate dissolution of
the complex. After acquiring a spectrum at -80 °C, olefin or other
trapping ligand was added via gastight syringe to the solution cooled
to -78 °C, and the NMR tube was briefly shaken to completely dissolve
the additive. The tube was then transferred to the cooled NMR probe
for acquisition of spectra. The concentrations of the alkyl agostic
species, trapped alkyl species, free olefin, and free trapping ligands
were calculated using the BAr′4 or p-acenaphthyl peaks as an internal
standard.
1
signals. H NMR (CDCl2F, -120 °C (static), 300 MHz): δ 8.25-
7.93 (2 br doublets, 1H each, An: Hp and Hp′), 7.22-7.14 (br m, 8H,
An: Hm, Hm′, 6 Haryl), 6.70 (d, 1H, J ) 8.1 Hz, An: Ho), 6.68 (d, 1H,
J ) 8.1 Hz, An: Ho′), 2.83 (br septet, 1H, PdCH(CH2-µ-H)(CH3)),
2.17-2.12 (4 br singlets, 3H each, 2,6-C6H3(Me,Me′), 2,6-C6H3-
(Me,Me′), 2,6-C6H3′(Me,Me′), and 2,6-C6H3′(Me,Me′)), 1.38 (PdCH-
(CH2-µ-H)(CH3)), 0.27 (br s, 3H, Pd(CH(CH2-µ-H)(CH3)), -7.85 (br
1
t, 1H, J ) 16 Hz, Pd(CH(CH2-µ-H)(CH3)). H NMR (CDCl2F, 0 °C
(dynamic), 300 MHz): δ 8.08 (d, 1H, J ) 8.7 Hz, An: Hp), 8.06 (d,
1H, J ) 8.4 Hz, An: Hp′), 7.51-7.26 (m, 8H, An: Hm, Hm′, 6 Haryl),
6.81 (d, 1H, J ) 7.2 Hz, An: Ho), 6.78 (d, 1H, J ) 7.2 Hz, An: Ho′),
2.26 (s, 6H, 2,6-C6H3(Me2)), 2.23 (s, 6H, 2,6-C6H3′(Me2)), -0.60 (br,
7H, PdC3H7).
Trapping of â-Agostic Species i-16/n-16. â-Agostic i-16 can be
isolated and stored or formed in situ.48 Samples of i-16 were added to
or prepared in an NMR tube in either CD2Cl2 or CDCl2F according to
the procedures described above. The â-agostic isopropyl complex was
trapped with acetonitrile (b), ethylene (c), propylene (d), and dimethyl
sulfide (e) (added via gastight microliter syringe) as isopropyl species
at temperatures low enough to prevent isomerization to the n-propyl
species. A temperature of -78 °C was sufficient for all trapping ligands
except for propylene, which was introduced to an NMR tube cooled to
-130 °C in a pentane/liquid N2 bath. The 1H NMR data for the initially
trapped isopropyl complexes as well as the isomerized n-propyl
complexes are listed below. Equilibrium constants for the trapped
n-propyl/isopropyl species are indicated in Figure 5.
Migratory Insertion Rates. Rates of migratory insertion were
monitored for ethylene and propylene as previously described (eq 10)
and are listed in Tables 1 and 2. 1H NMR spectra for the initially formed
ethylene complexes 3c and 5c and for the propylene complex 3d are
described below (spectra for 1c and 2c have been reported).18
(a) [(ArNdC(An)C(An)dNAr)Pd(CH3)(CH2dCH2)]BAr′4 (Ar )
1
2,6-C6H3(i-Pr)2) (3c). H NMR (CD2Cl2, -80 °C, 300 MHz) δ 8.09
(d, 1H, J ) 8.4 Hz, An: Hp), 8.05 (d, 1H, J ) 8.4 Hz, An: Hp′),
7.55-7.39 (m, 8H, An: Hm, Hm′; 6 Haryl), 6.55 (d, 1H, J ) 7.2 Hz,
An: Ho), 6.53 (d, 1H, J ) 7.5 Hz, An: Ho′), 4.62 (br s, 4H, C2H4),
3.03 (septet, 2H, J ) 6.9 Hz, CHMe2), 2.96 (septet, 2H, J ) 6.9 Hz,
C′HMe2), 1.34 (d, 6H, J ) 6.6 Hz, CHMeMe′), 1.29 (d, 6H, J ) 6.3

6700 J. Am. Chem. Soc., Vol. 122, No. 28, 2000
(a) [(ArNdC(An)C(An)dNAr)Pd(CH(CH3)2)(NCMe)]BAr′4
Tempel et al.
MHz): δ 8.12 (d, 1H, J ) 8.1 Hz, An: Hp), 8.09 (d, 1H, J ) 8.1 Hz,
An: Hp′), 7.56-7.34 (m, 8H, An: Hm, Hm′, 6 Haryl), 6.70 (d, 1H, J )
7.2, Hz, An: Ho), 6.45 (d, 1H, J ) 7.5 Hz, An: Ho′), 3.12 (septet, 2H,
J ) 6.9 Hz, CHMe2), 3.0 (septet, 2H, J ) 6.9 Hz, C′HMe2), 2.05
(s, 6H, SMe2), 1.37, 1.32, 0.93, and 0.84 (4 d, 6H each, J ) 6.9, 6.6,
6.9, and 6.6 Hz, CHMeMe′, CHMeMe′, C′HMeMe′, C′HMeMe′), 0.51
(t, 3H, J ) 7.2 Hz, PdCH2CH2CH3), (PdCH2CH2CH3 obscured).
Isomerization of i-16L to n-16L. Rates were determined for
isomerization of i-16c to n-16c (ethylene-trapped species) under various
concentrations of ethylene by treatment of the process as a first-order
reaction coming to equilibrium and based on Keq ) 19 at -65 °C. The
process was monitored by loss of the acenaphthyl Hortho signal at 6.19
ppm standarized to the two acenaphthyl Hpara signals at ca. 8 ppm.
Rates for isomerization of i-16e to n-16e (dimethyl sulfide-trapped
species) were determined by monitoring the same integrals but in this
case the process could be treated as an irreversible first-order reaction.
Ethylene Associative Exchange Rates. The rate of exchange of
free and bound ethylene was determined by NMR line broadening
experiments at -85 °C for [(ArNdC(R)C(R)dNAr)Pd(Me)(H2Cd
CH2)]BAr′4 (R ) H, An, Me; Ar ) 2,6-C6H3-(i-Pr)2). Samples were
prepared according to the following procedure: The corresponding
palladium ether adducts [(ArNdC(R)C(R)dNAr)Pd(Me)(H2CdCH2)]-
BAr′4 (R ) H, An, Me; Ar ) 2,6-C6H3-(i-Pr)2) were weighed (∼15
mg) into a tared NMR tube in a nitrogen-filled drybox. The tube was
then capped with a septum and Parafilm and cooled to -80 °C. Dry,
degassed CD2Cl2 (700 µL) was then added to the palladium complex
via gastight syringe, and the tube was shaken and warmed briefly to
give a homogeneous solution. After acquiring a -85 °C NMR spectrum,
ethylene was added to the solution via gastight syringe, and a second
NMR spectrum was acquired at -85 °C. The molarity of the BAr′4
counterion was calculated according to the moles of the ether adduct
placed in the NMR tube. The molarity of the ethylene complexes and
free ethylene were calculated using the BAr′4 peaks as an internal
standard. Line widths (w) were measured at half-height in units of
hertz for complexed ethylene and were corrected for line widths (wo)
in the absence of exchange. The exchange rates were determined
from the standard equation for the slow exchange approximation: k )
π(w - wo)/[C2H4], where [C2H4] is the molar concentration of free
ethylene. A pulse delay of 60 s and a 30° pulse width were used. (The
T1 of free ethylene is 15 s.) These experiments were repeated twice,
and the averaged rate constants are reported in Table 6.
(Ar ) 2,6-C6H3(i-Pr)2) (i-16b). 1H NMR (CD2Cl2, -80 °C, 300
MHz): δ 8.11 (d, 1H, J ) 6.6 Hz, An: Hp), 8.08 (d, 1H, J ) 6.9 Hz,
An: Hp′), 7.52-7.37 (m, 8H, An: Hm, Hm′, 6 Haryl), 6.97 (d, 1H, J )
6.9, Hz, An: Ho), 6.30 (d, 1H, J ) 7.2 Hz, An: Ho′), 3.15 (2 over-
lapping septets, 2H each, CHMe2, C′HMe2), 1.73 (s, 3H, NCMe), 1.37,
1.32, 0.98, and 0.83 (4 d, 6H each, J ) 6.0, 5.7, 6.0, and 6.0 Hz,
CHMeMe′, CHMeMe′, C′HMeMe′, C′HMeMe′), 0.67 (d, 6H, J ) 5.7
Hz, PdCH(CH3)2), (PdCH(CH3)2 obscured).
(b) [(ArNdC(An)C(An)dNAr)Pd(CH2CH2CH3)(NCMe)]BAr′4
(Ar ) 2,6-C6H3-(i-Pr)2) (n-16b). 1H NMR (CD2Cl2, -10 °C, 300
MHz): δ 8.15 (d, 1H, J ) 6.0 Hz, An: Hp), 8.13 (d, 1H, J ) 6.0 Hz,
An: Hp′), 7.55-7.32 (m, 8H, An: Hm, Hm′, 6 Haryl), 6.97 (d, 1H, J )
7.2, Hz, An: Ho), 6.47 (d, 1H, J ) 7.5 Hz, An: Ho′), 3.28-3.13 (2
overlapping septets, 2H each, CHMe2, C′HMe2), 1.82 (s, 3H, NCMe),
1.43, 1.38, 1.37, and 1.05 (4 d, 6H each, J ) 6.9, 6.6, 6.6, and 6.9 Hz,
CHMeMe′, CHMeMe′, C′HMeMe′, C′HMeMe′), 0.91 (t, 3H, J ) 6.6
Hz, PdCH2CH2CH3), (PdCH2CH2CH3 obscured).
(c) [(ArNdC(An)C(An)dNAr)Pd(CH(CH3)2)(CH2dCH2)]BAr′4
(Ar ) 2,6-C6H3(i-Pr)2) (i-16c). 1H NMR (CDCl2F, -80 °C, 300
MHz): δ 7.92 (d, 1H, J ) 9.0 Hz, An: Hp), 7.89 (d, 1H, J ) 9.0 Hz,
An: Hp′), 7.71 (s, 8H, BAr′4: Ho), 7.44 (s, 4H, BAr′4: Hp), 7.51-7.34
(m, 8H, An: Hm, Hm′, 6 Haryl), 6.41 (d, 1H, J ) 7.2, Hz, An: Ho), 6.19
(d, 1H, J ) 7.2 Hz, An: Ho′), 4.56 (s, 4H, C2H4), 3.05 (septet, 2H,
J ) 6.0 Hz, CHMe2), 2.89 (septet, 2H, J ) 6.0 Hz, C′HMe2), 2.4 (septet,
1H, J ) 6.0 Hz, PdCH(CH3)2), 1.31, 1.26, and 0.75 (2 coincident)
(4 d, 6H each, J ) 5.7, 5.7, 5.4, and 5.4 Hz, CHMeMe′, CHMeMe′,
C′HMeMe′, C′HMeMe′), 0.63 (d, 6H, J ) 5.7 Hz, PdCH(CH3)2).
(d) [(ArNdC(An)C(An)dNAr)Pd(CH2CH2CH3)(CH2dCH2)]-
1
BAr′4 (Ar ) 2,6-C6H3-(i-Pr)2) (n-16c). H NMR (CDCl2F, -60 °C,
300 MHz): δ 7.99 (d, 1H, J ) 8.4 Hz, An: Hp), 7.95 (d, 1H, J ) 8.4
Hz, An: Hp′), 7.53-7.37 (m, 8H, An: Hm, Hm′, 6 Haryl), 6.51 (d, 1H,
J ) 7.2, Hz, An: Ho), 6.46 (d, 1H, J ) 7.2 Hz, An: Ho′), 5.60 (s, 4H,
C2H4), 3.01 (septet, 2H, J ) 6.9 Hz, CHMe2), 2.92 (septet, 2H, J )
6.6 Hz, C′HMe2), 1.47 (t, 2H, J ) 8.7 Hz, PdCH2CH2CH3), 1.35, 1.30,
0.85, and 0.82 (4 d, 6H each, J ) 6.9, 6.6, 7.2, and 7.2 Hz, CHMeMe′,
CHMeMe′, C′HMeMe′, C′HMeMe′), 0.47 (t, 3H, J ) 7.2 Hz, PdCH2-
CH2CH3), (PdCH2CH2CH3 obscured).
(e) [(ArNdC(An)C(An)dNAr)Pd(CH(CH3)2)(CH2dCHCH3)]-
1
BAr′4 (Ar ) 2,6-C6H3(i-Pr)2) (i-16d). H NMR (CDCl2F, -110 °C,
300 MHz): δ 7.88 (d, 1H, J ) 8.1 Hz, An: Hp), 7.84 (d, 1H, J ) 8.1
Hz, An: Hp′), 7.49-7.28 (m, 8H, An: Hm, Hm′, 6 Haryl), 6.11 (d, 1H,
J ) 7.5, Hz, An: Ho), 6.08 (d, 1H, J ) 6.9 Hz, An: Ho′), 5.08 (m, 1H,
CH2 ) CHCH3), 4.46-4.11 (m, 2H, CH2dCHCH3), 3.14, 3.05, 2.95,
and 2.64 (4 br septets, 1H each, CHMe2, CH′Me2, C′HMe2, C′H′Me2),
2.27 (br septet, 1H, PdCHMe2), 1.77 (br doublet, 3H, CH2dCHCH3),
1.33, 1.19, 0.80, and 0.66 (8 br doublets, 3H each CHMeMe′,
C′HMeMe′, C′′HMeMe′, C′′′HMeMe′), 0.39 (br doublet, 6H, PdCH-
(CH3)2).
X-ray Structure Determinations. Data were collected on a Rigaku
AFC 6/S diffractometer with graphite-monochromated Mo KR radiation
(λ ) 0.71073 Å) using a θ/2θ Å scan; reflections with I > 2.5σ were
considered observed and included in subsequent calculations. The
structures were solved by direct methods. Refinement was by full-
matrix least squares with weights based on counter statistics. Hydrogen
atoms were included in the final refinement using a riding model with
thermal parameters derived from the atom to which they are bonded.
Crystal data and experimental conditions are given in the Supporting
Information. All computations were performed using the NRCVAX
suite of programs.77,78
(f) [(ArNdC(An)C(An)dNAr)Pd(CH2CH2CH3)(CH2dCHCH3)]-
1
BAr′4 (Ar ) 2,6-C6H3-(i-Pr)2) (n-16d). H NMR (CDCl2F, -60 °C,
300 MHz): δ 7.97 (d, 1H, J ) 8.4 Hz, An: Hp), 7.93 (d, 1H, J ) 8.7
Hz, An: Hp′), 7.53-7.23 (m, 8H, An: Hm, Hm′, 6 Haryl), 6.40 (2
coincident doublets, 1H Hz, An: Ho, Ho′), 5.30 (m, 1H, CH2 ) CHCH3),
4.61 and 4.20 (m, 2H, CH2dCHCH3), 3.11-2.88 (4 septets, 1H
each, CHMe2, C′HMe2, C′′HMe2, C′′′HMe2,), 1.80 (d, 3H, J ) 6.0
Hz, CH2dCHCH3), 1.37-1.28 and 0.84-0.74 (8 overlapping doublets,
3H each CHMeMe′, C′HMeMe′, C′′HMeMe′, C′′′HMeMe′), 0.39 (t, 3H,
J ) 6.6 Hz, PdCH2CH2CH3), (PdCH2CH2CH3 obscured).
Acknowledgment. This work was supported by the National
Science Foundation (CHE-9710380) and by DuPont. R.L.H. and
L.K.J. thank the NSF for graduate and postdoctoral research
fellowships, respectively. We also wish to thank Dr. Chi-Duen
Poon for NMR technical assistance.
Supporting Information Available: Graphs for the deter-
mination of rates of migratory insertion for complex 5a and
rates of isomerization for i-16c/n-16c and i-16e/n-16e, tables
of crystal data, structure solution and refinement, atomic
coordinates, bond lengths and angles, and anisotropic thermal
parameters (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.
(g)
[(ArNdC(An)C(An)dNAr)Pd(CH(CH3)2)(SMe2)]BAr′4
(Ar ) 2,6-C6H3(i-Pr)2) (i-16e). 1H NMR (CD2Cl2, -45 °C, 300
MHz): δ 8.09 (d, 1H, J ) 8.4 Hz, An: Hp), 8.06 (d, 1H, J ) 8.4 Hz,
An: Hp′), 7.57-7.38 (m, 8H, An: Hm, Hm′, 6 Haryl), 6.46 (d, 1H, J )
7.5, Hz, An: Ho), 6.18 (d, 1H, J ) 7.5 Hz, An: Ho′), 3.15 (septet, 2H,
J ) 6.6 Hz, CHMe2), 3.05 (septet, 2H, J ) 6.6 Hz, C′HMe2, 2.24
(septet, 1H, J ) 6.0 Hz, PdCHMe2), 2.04 (s, 6H, SMe2), 1.40, 1.32,
0.85, and 0.80 (4 doublets, 6H each, J ) 6.9, 6.9, 6.9, and 6.9 Hz,
CHMeMe′, CHMeMe′, C′HMeMe′, C′HMeMe′), 0.68 (br doublet, 6H,
PdCH(CH3)2).
JA000893V
(77) Gabe, E. J.; LePage, Y.; Charland, J.-P.; Lee, F.; White, P. S. J.
Appl. Crystallogr. 1989, 22, 384-387.
(78) Johnson, C. K. Technical Report No. ORNL-5138; Oak Ridge
National Laboratory: Oak Radge, TN, 1976.
(h) [(ArNdC(An)C(An)dNAr)Pd(CH2CH2CH3)(SMe2)]BAr′4
(Ar ) 2,6-C6H3-(i-Pr)2) (n-16e). 1H NMR (CD2Cl2, -60 °C, 300