Unusual reactivity of N, N,N′,N′-tetramethylethylenediamine- coordinated neutral nickel(II) polymerization catalysts

Article abstract of DOI:10.1021/om900198c

Tmeda-coordinated species [(N,O)NiCH₃(tmeda)] (tmeda = N,N,N′,N′- tetramethylethylenediamine), obtained by reaction of [(tmeda)NiMe₂] with salicylaldimines, (N,O)H, are reactive and versatile intermediates for olefin polymerization c

Full text of DOI:10.1021/om900198c

4048 Organometallics 2009, 28, 4048–4055
DOI: 10.1021/om900198c
Unusual Reactivity of N,N,N0,N0-Tetramethylethylenediamine-
Coordinated Neutral Nickel(II) Polymerization Catalysts
€
Andreas Berkefeld, Heiko M. Moller, and Stefan Mecking*
Chair of Chemical Materials Science, Department of Chemistry, University of Konstanz,
Universitatsstrasse 10, D-78457 Konstanz, Germany
::
Received March 16, 2009
Tmeda-coordinated species [(N,O)NiCH3(tmeda)] (tmeda=N,N,N0,N0-tetramethylethylenediamine),
obtained by reaction of [(tmeda)NiMe2] with salicylaldimines, (N^O)H, are reactive and versatile
intermediates for olefin polymerization catalysis. Solution NMR spectroscopic studies of 1-tmeda
(N,O=2,6-(3,5-(F3C)2C6H3)2C6H3-NdCH-(3,5-I2-2-OC6H2)) revealed two major binding modes of the
tmeda ligand, open κ1- and, unexpectedly, chelating κ2-fashion, which interconvert slowly on the NMR
chemical shift time scale, and form equilibria with solvent complexes 1-L (L=dmso, methanol). Binding
of tmeda is favored by 2-3 orders of magnitude at the temperatures studied (25 to 80 °C) over binding of
solvent. Chelating κ2-coordination of tmeda renders the monoanionic bidentate salicylaldiminato ligand
κ1-coordinate. Exposure of dmso solutions of 1-tmeda to excess ethylene in an NMR tube at 55 °C
resulted in the very minor formation of propylene and an equilibrium mixture of Ni(II)-ethyl complexes
2-dmso and [(κ1-N,O)Ni(RCH2βCH3)(κ2-tmeda)] (2-κ2-tmeda). Ethylene is primarily dimerized to bu-
tenes, which qualitatively parallels the reactivity observed for tmeda-free solutions of 2-dmso, but tmeda-
coordinated Ni(II)-alkyl complexes appeared unreactive, i.e., dormant, toward ethylene. Carrying out
the aforementioned reaction under aqueous conditions revealed that hydrolysis of Ni(II)-Me species to
methane is a relevant deactivation pathway of the catalyst precursor, which clearly contrasts the reactivity
observed in the absence of tmeda. Observed pseudo-first-order rate constants of overall disappearance
of 1-tmeda split into two independent contributions according to kobs,Me,1-tmeda= kins,Me,1-tmeda
+
khydr,1-tmeda[water], khydr,1-tmeda=1.9 ꢀ 10-4 M-1 s-1, and kins,Me,1-tmeda ≈ 2.4 ꢀ 10-4 M-1 s-1 (∼0.1
M [C2H4]). Determination of activation parameters of the bimolecular elimination of ethane from
1-tmeda (ΔHq= 97 ( 7 kJ mol-1 and ΔSq ≈ -5 J mol-1 K-1), a generally relevant deactivation
mechanism of Ni(II)-methyl complexes, points out that tmeda-coordinated Ni(II)-methyl complexes,
despite being inactive toward activation with ethylene, are actively involved in decomposition reactions.
Introduction
lene and 1-olefins can be copolymerized with electron-defi-
cient polar monomers such as acrylates to higher molecular
weight polymers.2,3 Acrylates can be homooligomerized by
an insertion mechanism,4 and polymerizations can be carried
out in aqueous emulsions.5 These studies prompted renewed
interest in neutral nickel(II) ethylene polymerization cata-
lysts.6-9 The perhaps most prominent class of neutral Ni(II)
polymerization catalysts are salicylaldiminato complexes.7
These are versatile catalysts that afford high molecular
Olefin polymerization by complexes of d8 metals (late
transition metals) has been studied intensely in the past
decade.1 A major motivation is the functional group toler-
ance of these catalysts, by comparison to their much more
oxophilic early transition-metal counterparts. Thus, ethy-
*Corresponding author. E-mail: stefan.mecking@uni-konstanz.de.
(1) (a) Ittel, S. D.; Johnson, L. K.; Brookhart, M. Chem. Rev. 2000,
100, 1169–1203. (b) Gibson, V. C.; Spitzmesser, S. K. Chem. Rev. 2003,
103, 283–315. (c) Mecking, S. Angew. Chem., Int. Ed. 2001, 40, 534–540.
(d) Mecking, S. Coord. Chem. Rev. 2000, 203, 325–351. (e) Guan, Z.
Chem.;Eur. J. 2002, 8, 3086–3092. (f) Domski, G. J.; Rose, J. M.;
Coates, G. W.; Bolig, A. D.; Brookhart, M. Prog. Polym. Sci. 2007, 32,
30–92. (g) Berkefeld, A.; Mecking, S. Angew. Chem., Int. Ed. 2008, 47,
2538–2542.
(2) (a) Johnson, L. K.; Mecking, S.; Brookhart, M. J. Am. Chem. Soc.
1996, 118, 267–268. (b) Mecking, S.; Johnson, L. K.; Wang, L.;
Brookhart, M. J. Am. Chem. Soc. 1998, 120, 888–899. (c) Johnson, L.;
Wang, L.; McLain, S.; Bennett, A.; Dobbs, K.; Hauptman, E.; Ionkin,
A.; Ittel, S.; Kunitsky, K.; Marshall, W.; McCord, E.; Radzewich, C.;
Rinehart, A.; Sweetman, K. J.; Wang, Y.; Yin, Z.; Brookhart, M. ACS
Symp. Ser. 2003, 857, 131–142. (d) Li, W.; Zhang, X.; Meetsma, A.;
Hessen, B. J. Am. Chem. Soc. 2004, 126, 12246–12247. (e) Popeney, C.
S.; Camacho, D. H.; Guan, Z. J. Am. Chem. Soc. 2007, 129, 10062–
10063.
(3) Drent, E.; van Dijk, R.; van Ginkel, R.; van Oort, B.; Pugh, R. I.
Chem. Commun. 2002, 744–745.
::
€
(4) (a) Guironnet, D.; Roesle, P.; Runzi, T.; Gottker-Schnetmann, I.;
Mecking, S. J. Am. Chem. Soc. 2009, 131, 422-423. Also cf. (b) Chen, C;
Luo, S.; Jordan, R. F. J. Am. Chem. Soc. 2008, 130, 12892–12893.
(5) (a) Held, A.; Bauers, F. M.; Mecking, S. Chem. Commun. 2000,
301–302. (b) Tomov, A.; Broyer, J.-P.; Spitz, R. Macromol. Symp. 2000,
150, 53–58. (c) Bauers, F. M.; Mecking, S. Angew. Chem., Int. Ed. 2001,
40, 3020–3022. (d) Soula, R.; Novat, C.; Tomov, A.; Spitz, R.; Claverie,
J.; Drujon, X.; Malinge, J.; Saudemont, T. Macromolecules 2001, 34,
€
2022–2026. (e) Gottker-Schnetmann, I.; Korthals, B.; Mecking, S.
J. Am. Chem. Soc. 2006, 128, 7708–7709. (f) Weber, C. H. M.;
Chiche, A.; Krausch, G.; Rosenfeldt, S.; Ballauff, M.; Harnau, L.;
€
Gottker-Schnetmann, I.; Tong, Q.; Mecking, S. Nano Lett. 2007, 7,
2024-2029. Reviews: (g) Claverie, J. P.; Soula, R. Prog. Polym. Sci. 2003,
28, 619–662. (h) Mecking, S. Colloid Polym. Sci. 2007, 285, 605–619.
r
pubs.acs.org/Organometallics
Published on Web 05/29/2009
2009 American Chemical Society

Article
Organometallics, Vol. 28, No. 14, 2009 4049
weight polyethylene, are capable of homooligomerization
and copolymerization of 1-olefins (also with polar functional
groups providing these are separated from the vinyl group by
a spacer), and polymerize in aqueous emulsion to afford
unique nanoparticles. The degree of branching of ethylene
homopolymers, and thus their crystallinity and thermal
properties, can be varied over a wide range by remote
substituents.7
In studies of catalytic polymerizations, neutral or cationic
Ni(II)- and Pd(II)-methyl complexes represent versatile
well-defined single-component precursors for catalysis and
mechanistic studies. Convenient and increasingly utilized
reagents for the preparation of neutral M(II)-Me complexes
in general are [(tmeda)M(CH3)2]10,11 (M=Ni, Pd; tmeda=
N,N,N0,N0-tetramethylethylenediamine). Reaction with a
stoichiometric amount of a monoanionic bidentate ligand
in its protonated form, e.g., of free salicylaldimine, (N,O)H,
with [(tmeda)Ni(CH3)2], results in protonation of one M(II)-
methyl moiety to methane and formation of [(N,O)NiCH3-
(tmeda)].6h,p,q,7c-m,12-14 The tmeda ligand is relatively la-
bile, and in the presence of other coordinating species L, the
corresponding complexes [(N,O)NiCH3(L)] can be formed
(e.g., L= pyridine, phosphines, primary amines).7 Due to
their reactivity and lability, the tmeda species are also
precursors to highly active catalysts.7g,14a However, the
nature and properties of the [(N,O)NiCH3(tmeda)] species
are unclear. We now report the results of comprehensive
NMR studies of this issue.
(6) (a) Braunstein, P.; Pietsch, J.; Chauvin, Y.; Mercier, S.; Saussine,
L.; DeCian, A.; Fischer, J. J. Chem. Soc., Dalton Trans. 1996, 3571–
3574. (b) Braunstein, P.; Pietsch, J.; Chauvin, Y.; DeCian, A.; Fischer, J.
J. Organomet. Chem. 1997, 529, 387–393. (c) Pietsch, J.; Braunstein, P.;
Chauvin, Y. New J. Chem. 1998, 22, 467–472. (d) Rachita, M. J.; Huff,
R. L.; Bennett, J. L.; Brookhart, M. J. Polym. Sci. Part A 2000, 38, 4627–
4640. (e) Hicks, F. A.; Brookhart, M. Organometallics 2001, 20, 3217–
3219. (f) Soula, R.; Broyer, J. P.; Llauro, M. F.; Tomov, A.; Spitz, R.;
Claverie, J.; Drujon, X.; Malinge, J.; Saudemont, T. Macromolecules
2001, 34, 2438–2442. (g) Gibson, V. C.; Tomov, A.; White, A. J. P.;
Results and Discussion
Reactions of [(tmeda)Ni(CH3)2] with salicylaldimines (N,O)H
proceed straightforwardly and intrinsically introduce the weakly
coordinating bidentate ligand N,N,N0,N0-tetramethylethylene-
diamine, tmeda, to the product complexes [(N,O)NiCH3(tme-
da)] (eq 1). The particular salicylaldimine (N,O)H = 2,6-(3,5-
(F3C)2C6H3)2C6H3-NdCH-(3,5-I2-2-OH-C6H2) was chosen for
this study, as Ni(II) complexes of this ligand are long-
lived, robust, and very active precatalysts for ethylene polymer-
ization.5e,5f,7h,13
€
Williams, D. J. Chem. Commun. 2001, 719–720. (h) Schroder, D. L.;
Keim, W.; Zuideveld, M. A.; Mecking, S. Macromolecules 2002, 35,
6071–6073. (i) Hicks, F. A.; Jenkins, J. C.; Brookhart, M. Organometal-
€
lics 2003, 22, 3533–3545. (j) Heinicke, J.; Kohler, M.; Peulecke, N.;
He, M.; Kindermann, M. K.; Keim, W.; Fink, G. Chem.;Eur. J. 2003,
9, 6093–6107. (k) Jenkins, J. C.; Brookhart, M. J. Am. Chem. Soc. 2004,
126, 5827–5842. (l) Speiser, F.; Braunstein, P.; Saussine, L. Acc. Chem.
Res. 2005, 38, 784–793. (m) Zhang, L.; Brookhart, M.; White, P. S.
ꢀ
Organometallics 2006, 25, 1868–1874. (n) Kuhn, P.; Semeril, D.;
Jeunesse, C.; Matt, D.; Neuburger, M.; Mota, A. Chem.;Eur. J.
2006, 12, 5210–5219. (o) Nowack, R. J.; Hearley, A. K.; Rieger, B. Z.
Anorg. Allg. Chem. 2005, 631, 2775–2781. (p) Yu, S.-M.; Berkefeld, A.;
::
€
Gottker-Schnetmann, I.; Muller, G.; Mecking, S. Macromolecules 2007,
40, 421–428. (q) Guironnet, D.; Runzi, T.; Gottker-Schnetmann, I.;
::
€
Mecking, S. Chem. Commun. 2008, 4965–4967. (r) Lavanant, L.;
Rodriguez, A.-S.; Kirillov, E.; Carpentier, J.-F.; Jordan, R. F. Organo-
metallics 2008, 27, 2107–2117. (s) Rodriguez, B. A.; Delferro, M.;
Marks, T. J. Organometallics 2008, 27, 2166–2168. (t) Azoulay, J. D.;
Itigaki, K.; Wu, G.; Bazan, G. C. Organometallics 2008, 27, 2273–2280.
(u) Zhou, X.; Bontemps, S.; Jordan, R. F. Organometallics 2008, 27,
4821–4824.
(7) (a) Johnson, L. K.; Bennett, A. M. A.; Ittel, S. D.; Wang, L.;
Parthasarathy, A.; Hauptman, E.; Simpson, R. D.; Feldman, J.; Cough-
lin, E. B. (DuPont) WO98/30609, 1998. (b) Wang, C.; Friedrich, S.;
Younkin, T. R.; Li, R. T.; Grubbs, R. H.; Bansleben, D. A.; Day, M. W.
Organometallics 1998, 17, 3149–3151. (c) Younkin, T. R.; Connor, E. F.;
Henderson, J. I.; Friedrich, S. K.; Grubbs, R. H.; Bansleben, D. A.
Science 2000, 287, 460–462. (d) Connor, E. F.; Younkin, T. R.; Hen-
derson, J. I.; Hwang, S.; Grubbs, R. H.; Roberts, W. P.; Litzau, J. J. J.
Polym. Sci. Part A 2002, 40, 2842–2854. (e) Connor, E. F.; Younkin, T.
R.; Henderson, J. I.; Waltman, A. W.; Grubbs, R. H. Chem. Commun.
2003, 2272–2273. (f) Darensbourg, D. J.; Ortiz, C. G.; Yarbrough, J. C.
Inorg. Chem. 2003, 42, 6915–6922. (g) Bastero, A.; Kolb, L.; Wehrmann,
Polymerization experiments carried out at precatalyst
loadings of 1 μmol of 1-tmeda in toluene solution at 20 °C
revealed that overall catalytic activities (measured in units
of mol(C2H4) mol(1-tmeda)-1 h-1= TO h-1) are nearly
independent of the ethylene concentration in the pressure
range from 10 bar (3.0 ꢀ 104 TO h-1) to 30 bar (3.4 ꢀ
104 TO h-1). This indicates that under these conditions
tmeda does not compete with monomer binding to the metal
center. A highly linear semicrystalline polyethylene is ob-
tained, with, for example, a molecular weight of Mn=1.5 ꢀ
105 g mol-1 (Mw/Mn=2.5, ca. 50% crystallinity; prepared at
30 bar).
€
P.; Bauers, F. M.; Gottker gen. Schnetmann, I.; Monteil, V.; Thomann,
R.; Chowdhry, M. M.; Mecking, S. Polym. Mat. Sci. Eng. 2004, 90, 740–
1H NMR Spectroscopic Studies of 1-tmeda. Two Ni(II)-
methyl species coexist in a close to 1:1 ratio in dmso-d6
solutions of 1-tmeda at ambient temperatures. Characteristic
€
741. (h) Zuideveld, M. A.; Wehrmann, P.; Rohr, C.; Mecking, S. Angew.
Chem., Int. Ed. 2004, 43, 869–873. (i) Wehrmann, P.; Mecking, S.
Macromolecules 2006, 39, 5963–5964. (j) Wehrmann, P.; Zuideveld,
M. A.; Thomann, R.; Mecking, S. Macromolecules 2006, 39, 5995–
€
6002. (k) Gottker Schnetmann, I.; Wehrmann, P.; Rohr, C.;
Mecking, S. Organometallics 2007, 26, 2348–2362. (l) Bastero, A.;
€
€
(10) Kaschube, W.; Porschke, K. R.; Wilke, G. J. Organomet. Chem.
1988, 355, 525–532.
(11) de Graaf, W.; Boersma, J.; Smeets, W. J. J.; Spek, A. L.;
van Koten, G. Organometallics 1989, 8, 2907–2917.
(12) For [(Me3P)2NiMe(OMe)] as a precursor see: Klein, H.-F.
Inorg. Chim. Acta 1996, 248, 111–114.
€
€
Gottker-Schnetmann, I.; Rohr, C.; Mecking, S. Adv. Synth. Catal.
2007, 349, 2307–2316. (m) Wehrmann, P.; Mecking, S. Organometallics
2008, 27, 1399-1408. Also see: (n) Brasse, M.; Campora, J.; Palma, P.;
Alvarez, E.; Cruz, V.; Ramos, J.; Reyes, M. L. Organometallics 2008, 27,
4711–4723.
(8) For early related work on phosphino enolate catalyst systems see:
(a) Keim, W.; Kowaldt, F. H.; Goddard, R.; Kruger, C. Angew. Chem.,
(13) Berkefeld, A.; Mecking, S. J. Am. Chem. Soc. 2009, 131, 1565–
1574.
::
Int. Ed. Engl. 1978, 17, 466–467. (b) Ostoja Starzewski, K. A.; Witte, J.
Angew. Chem., Int. Ed. Engl. 1985, 24, 599–601. (c) Klabunde, U.; Ittel,
S. D. J. Mol. Catal. 1987, 41, 123–134.
(9) For computational studies see: (a) Chan, M. S. W.; Deng, L.;
Ziegler, T. Organometallics 2000, 19, 2741–2750. (b) Zeller, A.; Strass-
ner, T. J. Organomet. Chem. 2006, 691, 4379–4385.
(14) For a corresponding reaction of [(tmeda)PdMe2] with salicylal-
dimine: ref 7f. With (zwitterionic) phosphinosulfonic acids, (P,O)H:
(a) Skupov, K. M.; Marella, P. R.; Simard, M.; Yap, G. P. A.; Allen, N;
Conner, D.; Goodall, B. L.; Claverie, J. P. Macromol. Rapid Commun.
2007, 28, 2033–2038. (b) Vela, J.; Lief, G. R.; Shen, Z.; Jordan, R. F.
Organometallics 2007, 26, 6624–6635 and ref 4a.

4050 Organometallics, Vol. 28, No. 14, 2009
Scheme 1. Equilibrium between 1-κ2-tmeda and 1-dmso As Observed by NMR Spectroscopy
Berkefeld et al.
low- and high-field 1H NMR resonances of the imine
(-HCdN-, H7, see eq 1) and Ni(II)-CH3 protons, respectively,
were observed at 11.08 and -1.79 ppm for the first and at 8.11
and -1.19 ppm for the second Ni(II)-methyl species (Figure S1,
Supporting Information (SI)). Two-dimensional rotating-frame
nuclear Overhauser effect spectroscopy (2D ROESY) indicates
the chemical exchange of both the imine and the Ni(II)-methyl
protons, respectively, between the two species (Figure S2, SI).
The exchange occurs slowly on the NMR chemical shift time
scale. By contrast to NMR spectra of 1-tmeda, a single defined
Ni(II)-methyl species is observed for the separately prepared13
dmso-coordinated complex [(N,O)NiCH3(dmso)] (1-dmso),
and characteristic 1H NMR resonances of the imine and
Ni(II)-CH3 protons were observed at 8.13 and -1.19 ppm,
respectively. Notably, the addition of 1 equiv of tmeda to a
dmso-d6 solution of neat 1-dmso resulted in a 1H NMR
spectrum entirely identical to that observed for 1-tmeda
in dmso-d6, consistent with the partial ligand exchange of
(deuterated) dmso by tmeda and the formation of a tmeda-
coordinated Ni(II)-methyl complex. Further, the Ni(II)-methyl
species observed at -1.79 ppm became the dominant species
(with a molar fraction x > 0.9) after the addition of 3 equiv of
tmeda to either sample (Figure S3, SI), which shifts the equili-
brium concentrations between dmso- and tmeda-coordinated
Ni(II)-methyl complexes in favor of the latter.
orientation of the Ni(II)-methyl and imine moieties and, as
a consequence, no NOEs between these protons have been
observed for these complexes. The relative orientation of the
Ni(II)-methyl and imine moiety toward each other reveals
further evidence of an unhindered rotation around the N-C8
and C1-C7 bonds in 1-κ2-tmeda. In particular, rotation
around the C1-C7 bond is indicative of a κ1-O-coordination
of the (N,O)-ligand and, third, is assumed to cause the
1
significant low-field shift of the imine H NMR resonance
detected at 11.08 ppm as opposed to 8.13 ppm observed for
1-dmso and 8.08 ppm in the case of the free salicylaldimine
(N,O)H. The complete 1H and 13C NMR spectroscopic
assignment of 1-κ2-tmeda is provided in the Experimental
Section.
The equilibrium constant KMe(T)=[1-dmso][tmeda]free
/
([1-κ2-tmeda][dmso]) was determined in the temperature
range from 27 to 80 °C by 1H NMR spectroscopy. The
formation of 1-κ2-tmeda is thermodynamically favored over
the solvent species 1-dmso; values of KMe(T) increased from
2.9 ꢀ 10-3 to 4.0 ꢀ 10-2 over the temperature range studied
(27 to 80 °C). Linear regression of a plot of ln(KMe(T)) vs
reciprocal temperatures provided a reaction enthalpy ΔH°
of 44 ( 1 kJ mol-1, and a difference of the free enthalpies of
1-κ2-tmeda and 1-dmso, ΔG°=15 kJ mol-1, was calculated
from KMe(T) at room temperature (Figure S4, SI).
1H NMR, and 2D ROESY data in particular, reveal the
coordination of tmeda to the Ni(II)-alkyl moiety to occur
preferentially in a κ2-fashion with the salicylaldiminato
ligand κ1-coordinated (1-κ2-tmeda, Scheme 1) in dmso-d6
solution, surprisingly. This coordination mode contrasts
with the known observation that tmeda can be facily re-
placed from [(N,O)NiMe(tmeda)] species by ligands such as
L=pyridine or phosphine to afford the κ2-salicylaldiminato
complexes [(N,O)NiMe(L)].7c,e,f,h,k-m This unexpected
structure is clearly evident from, first, the κ2-coordination
of tmeda in 1-κ2-tmeda rendering the N-CH3 groups inequi-
Presumably, 1-κ2-tmeda and 1-dmso equilibrate via the
Ni(II)-methyl species 1-κ1-tmeda (Scheme 1). The latter was
not observed in dmso-d6 solution, but studies of 1-tmeda in
methanol-d4 and toluene-d8 solution revealed that complex
1-κ1-tmeda is a relevant species. Values for the equilibrium
constant KMe,2=[1-κ2-tmeda]/[1-κ1-tmeda] were found to
range from 0.66 (ΔG°= 1.0 kJ mol-1) in methanol-d4 to
0.88 (ΔG°=0.3 kJ mol-1) in toluene-d8 solution at 25 °C,
providing a measure for the competition of the formation of
the (five-membered) κ2-tmeda chelate vs the (six-membered)
κ2-salicylaldiminato chelate.15,16
1
valent, and distinct H NMR resonances were observed at
2.36, 2.01, 1.82, and 1.64 ppm (Figure S1, SI). 2D ROESY
data indicate that the N-methyl groups of the κ2-tmeda
ligand are subject to (i) self-exchange and (ii) exchange with
free tmeda from solution. Both processes proceed slowly on
the NMR chemical shift time scale. Distinct dipolar proton-
proton couplings of all four N-methyl to the imine and
Ni(II)-methyl protons as well as to the four ortho protons
of the trifluoromethyl-substituted phenyl rings of the ter-
phenylaniline unit 2,6-(3,5-(F3C)2C6H3)2C6H3 (Figure S2,
SI) are in agreement with unhindered rotation around the
C1-C7 and N-C8 bond (C1-HC7dN-C8, see Scheme 1).
Second, particular cross-peaks were observed by 2D ROESY
consistent with dipolar coupling of Ni(II)-methyl and imine
protons for 1-κ2-tmeda (Figure S2, SI). Notably, κ2-co-
ordination of the (N,O)-ligand in complexes [(N,O)Ni-
CH3(L)] (with, e. g., L=dmso) enforces the opposite spatial
(15) Key 1H NMR resonances observed in methanol-d4 at 25 °C:
10.19 and -1.64 (-HCdN- and Ni(II)-CH3, 1-κ2-tmeda); 7.69, 2.40,
2.32, and -0.72 (-HCdN-, {(H3C)2NCH2}1/2, {(H3C)2NCH2}1/2, and
Ni(II)-CH3, 1-κ1-tmeda); 7.62 and -1.28 (-HCdN- and Ni(II)-CH3,
1-O(D)CD3) ppm. Resonances of free tmeda were observed at 2.01
({(H3C)2NCH2}1/2) and 2.26 ({(H3C)2NCH2}1/2) ppm. Equilibrium
constants according to 1H NMR spectra (see Scheme 1): KMe,1 =5.3 ꢀ
10-3 (ΔG°=1 3 k J m o l -1), KMe=KMe,1/KMe,2=8 ꢀ 10-3 (ΔG°=1 4 k J m o l -1).
Key 1H NMR resonances (imine/Ni(II)-CH3) of 1-κ2-tmeda and
1-κ1-tmeda observed in toluene-d8 at 25 °C: 11.80/-1.45 and 8.88/
-1.39 ppm, respectively. A third Ni(II)-methyl species detected at
-1.29 ppm (Ni(II)-CH3, x=0.21) was tentatively assigned to the tmeda-
bridged bimolecular complex [{(N,O)NiCH3}2(μ2-tmeda)] (1-μ2-tmeda),
consistent with the resonances of half an equivalent of free tmeda being
observed in the toluene-d8 solution.
(16) The stepwise addition of N,N-dimethylbutylamine to solutions of
1-tmeda in dmso-d6 (15 mM; 1.2 equiv amine) or methanol-d4 (6 mM; 19
equiv amine) did not yield observable quantities of the corresponding κ1-
amine complex [(N,O)NiCH3(N(Me)2C4H9)], an observation that under-
lines the particular binding properties of the bidentate tmeda ligand.

Article
Organometallics, Vol. 28, No. 14, 2009 4051
dissociation of the five-ring κ2-tmeda chelate (the rigidity of
which is also evidenced by its NMR spectroscopic properties,
see above), which results in a gain of conformational degrees
of freedom, occurs in a concerted fashion with the rate-
determining step of the deactivation reaction. The potential
bridging coordination of tmeda may also affect the bimole-
cular reaction.
Reactivity of 1-tmeda toward Ethylene and Water in dmso
Solution. Chain growth by ethylene insertion is a key step in
the utilization of salicylaldiminato complexes as polymeriza-
tion catalysts. Exposure of dmso-d6 solutions of 1-tmeda
(15-30 mM) to ethylene (∼0.1-0.2 M) in an NMR tube at
T=55 °C resulted in the partial reaction of 1-dmso and
1-κ2-tmeda to form propylene and Ni(II)-ethyl species (at an
overall conversion of Ni(II)-Me e 10%). The latter are the
resting state of catalytic ethylene dimerization to butenes,
which results in gradual consumption of ethylene. This
qualitatively parallels the behavior reported previously13
for tmeda-free solutions of 1-dmso toward ethylene in
dmso. In the latter case, propylene was formed with a
pseudo-first-order rate kins,Me=(6.8 ( 0.3) ꢀ 10-4 s-1
(at 0.15 M ethylene), and the Ni(II)-ethyl complex [(N,O)Ni-
(RCHβ2CH3)(dmso)] (2-dmso) catalyzed the conversion of
ethylene to butenes. 2-dmso is subject to interconversion of
the RC and βC moieties via an intermediate [(N,O)NiH-
(ethylene)] complex, which occurs slowly on the NMR
chemical shift time scale.13
Figure 1. Eyring plot of second-order rate constants kMe-Me,
1-tmeda(T) for the decomposition of 1-tmeda to C2H6 in dmso-d6
solution (as determined from the decrease of the total Ni(II)-
CH3 integral in the temperature range from T = 55 to 80 °C).
Scheme 2. Proposed Mechanism of Ethane Formation via the
Bimolecular Coupling of Ni(II)-Methyl Moietiesa
1H NMR resonances of the Ni(II)-alkyl species consider-
ably sharpened upon cooling of the sample to room tem-
perature and revealed the presence of a second Ni(II)-ethyl
species in a close to 1:1 equilibrium mixture with 2-dmso
(Figure 2). Characteristic 1H and 13C NMR resonances
(13C chemical shifts were detected indirectly by 2D hetero-
nuclear single-quantum correlation (HSQC) spectroscopy)
of the new Ni(II)-ethyl species were observed at 11.71 and
a L=tmeda may represent both κ1- and κ2-tmeda Ni(II)-methyl
species.
Decomposition of Ni(II)-Methyl Species. A relevant de-
composition pathway specific to Ni(II)-methyl complexes is
the bimolecular homocoupling of Ni(II)-methyl moieties to
ethane. The temperature dependence of second-order rate
constants kMe-Me(T) of ethane formation from neat 1-dmso
determined by variable-temperature 1H NMR spectroscopic
monitoring yielded the activation parameters ΔHq=57 (
1 kJ mol-1 and ΔSq=-(129 ( 2) J mol-1 K-1, correspond-
3
∼161 (-HCdN-), -0.39 (triplet, JH-H=7.2 Hz) and
13.5 (Ni(II)-CHHCH3), and -0.89/-1.02 (as broad multi-
plets each) and -1.5 (Ni(II)-CHHCH3) ppm. By comparison
to the NMR spectroscopic data obtained for 1-κ2-tmeda, the
Ni(II)-ethyl species was assigned as the κ2-tmeda-coordi-
nated complex 2-κ2-tmeda on the basis of the significant
low-field shift of the imine resonance observed at 11.71 ppm
as opposed to 8.21 ppm for 2-dmso, and the inequivalence of
the R-methylene protons [(N,O)Ni(CHRHR’CH3)(κ2-tmeda)]
(Figure S6, SI). The very low degree of conversion of 1-tmeda
into 2-κ2-tmeda (e5%) unfortunately precluded detection
of the corresponding resonances of the tmeda-ligand in both
1H and 2D NMR spectra.
ing to ΔGq(298 K)=95 ( 1 kJ mol-1 13
. In agreement with the
assumed formation of a dinuclear nickel species in the rate-
determining step of ethane formation, the activation entropy
was found to particularly contribute to the free activation
enthalpy. Analogous kinetic studies of the decomposition of
1-tmeda to ethane17 (Figure S5, SI) revealed a similar free
activation enthalpy ΔGq(298 K) ≈ 99 kJ mol-1 but consider-
ably different activation parameters ΔHq = 97 ( 7 kJ mol-1
and ΔSq ≈ -5 J mol-1 K-1 (Figure 1).
The R-methylene protons of 2-κ2-tmeda are subject to
interconversion, as monitored in the temperature regime of
slow exchange by 2D ROESY and nuclear Overhauser effect
spectroscopy (NOESY) methods. Notably, the interconver-
sion of RC and βC moieties of 2-κ2-tmeda appears to proceed
very slowly (if at all), and no indicative exchange cross-peaks
were observed by 2D NOESY or ROESY as opposed to the
dynamics of 2-dmso. However, positive cross-peaks between
the β-methyl proton resonances indicate that a slow ligand
exchange occurs between 2-dmso and 2-κ2-tmeda (Scheme 3,
Figure S7, SI).18 As observed for Ni(II)-methyl species
A tentative explanation for this observation is that
the required breaking of the Ni(II)-N bond(s) to tmeda
accounts for the relatively higher activation enthalpy. This is,
however, partially compensated by an overall activation
entropy, which is quite high (low negative value) given the
bimolecular nature of the overall reaction (Scheme 2). The
specific bidentate property of tmeda by comparison to
dmso apparently affects the rate-determining step. Possibly,
(17) Second-order rate constants kMe-Me,1-tmeda(T) were determin-
ed by 1H NMR spectroscopic monitoring of the time-dependent de-
crease of the total Ni(II)-CH3 integral of 1-tmeda (as the sum of
1-κ2-tmeda and 1-dmso) for temperatures from 55 to 80 °C (Figure S5).
(18) According to the exchange cross-peaks observed, the intercon-
version of RC and βC moieties of 2-dmso and the ligand exchange with
2-κ2-tmeda apparently occur at similar rates on the time scale of the
mixing times τmix applied in the 2D spectra.

4052 Organometallics, Vol. 28, No. 14, 2009
Berkefeld et al.
1
Figure 2. Typical high-field H NMR chemical shift region of Ni(II)-alkyls 2-dmso and 2-κ2-tmeda prepared in an NMR tube in
dmso-d6 at T = 55 °C (spectrum shown acquired at 25 °C).
Scheme 3. Dynamic Processes of 2-dmso and 2-κ2-tmeda
law rCH3D= khydr,1-tmeda[1-tmeda][D2O]. A pseudo-first-
order rate constant of CH3D formation kCH3D=6.3 ꢀ 10-6 s-1
Observed by 2D ROESY at T = 25 °C
was determined from the y-intercept (Figure 4), correspond-
ing to a second-order rate constant khydr,1-tmeda=1.9 ꢀ
10-4 M-1 s-1 (for [1-tmeda]t=0=34 mM).19
That hydrolysis occurs to a significant extent contrasts the
behavior of the analogous tmeda-free system,13 for which
hydrolysis could be detected under appropriate conditions,
but only represents a minor reaction pathway (only trace
amounts of CH3D gradually formed under otherwise
identical reaction conditions). Apparently, tmeda species
are actively involved in hydrolysis reactions. Presumably,
the hydrogen-bonding capability of free amines is relevant in
this context. For example, hydrogen bonding of water to the
noncoordinated amine group in the intermediately occurring
1-κ1-tmeda may promote the coordination of water and
hydrolysis reactions.
(see above), tmeda has a considerably higher binding affinity
toward the Ni(II)-ethyl species than dmso, and a value of
∼3 ꢀ 10-3 was estimated for the equilibrium constant KEt(T)=
Consideration of khydr,1-tmeda in the calculation of pseudo-
first-order insertion rate constants kins,Me,1-tmeda (corrected
for hydrolysis) from the kinetic data determined in the
1
[2-dmso][tmeda]free/([2-κ2-tmeda][dmso]) from the H NMR
presence of water according to kobs,Me,1-tmeda=kins,Me,1-tmeda
+
spectra at 25 °C.
khydr,1-tmeda[D2O] yielded kins,Me,1-tmeda ≈ 2.6 ꢀ 10-4 s-1
(as the average calculated for different concentrations of
D2O in the range from 0.69 to 1.38 and 2.12 M), which is
equivalent to the value determined in the absence of added
water (2.1 ꢀ 10-4 s-1) within experimental accuracy.
Concerning the activation of salicylaldiminato Ni(II)-Me
species for catalysis, by comparison to a tmeda-free system,
the presence of tmeda results in a retardation due to equili-
brium formation of 1-κ2-tmeda, which itself apparently
is dormant for insertion of ethylene.20 At the same time,
however, the κ2-tmeda species is reactive for hydrolysis as
a deactivation reaction (and also bimolecular deactivation,
see above).
Kinetic studies of the overall ethylene insertion into the
metal-carbon bond of the precursor Ni(II)-methyl species
were carried out to survey the reactivity of Ni(II)-alkyls [{κ1-
N,O}NiR(κ2-tmeda)], R g CH3, in the catalytic C-C linkage
of olefins. In this context, the reactivity of 1-tmeda toward
the most ubiquitous polar reagent, water (D2O), was also
addressed. Solutions of 1-tmeda in dmso-d6 (34 mM) were
exposed to ethylene (∼0.1 M) at T=5 5 °C in the NMR tube,
and the time-dependent decrease of the total Ni(II)-methyl
integral was monitored by 1H NMR spectroscopy. Apparent
pseudo-first-order rate constants kobs,Me,1-tmeda were deter-
mined by linear regression of plots of ln([1-tmeda]t/[1-
tmeda]t=0) vs time (Figure 3).
Decomposition of Higher Ni(II)-Alkyls. As outlined,
the bimolecular homocoupling of Ni(II)-methyl moieties
to ethane is a relevant decomposition pathway specific
to precursor Ni(II)-methyl complexes. This raises the ques-
tion for a corresponding reactivity of higher alkyls, Ni(II)-R
Unexpectedly, apparent rate constants kobs,Me,1-tmeda in-
creased in the presence of increasing amounts of added water
(D2O) in the reaction medium. For example, a value of 6.0 ꢀ
10-4 s-1 was determined in the presence of 60 equiv of added
water (data set (+), Figure 3) as opposed to 2.1 ꢀ 10-4 s-1 in
1
the absence of water. According to the H NMR spectra
(19) An averaged value of 1.9 ꢀ 10-4 M-1 s-1 was calculated
independently for khydr,1-tmeda from initial rates of methane formation
rCH3D and initial concentrations of 1-tmeda and D2O according to
khydr,1-tmeda =rCH3D/([1-tmeda]t=0[D2O]t=0).
considerable amounts of CH3D gradually formed from
Ni(II)-methyl bond hydrolysis of 1-tmeda in the presence
of water (D2O) and ethylene under reaction conditions
(catalytic dimerization of ethylene to butenes). Initial rates
of methane formation from 1-tmeda (rCH3D) were deter-
mined by 1H NMR spectroscopic monitoring of the
gradual growth of the characteristic 1:1:1 triplet resonance
of CH3D (observed at 0.16 ppm) as a function of [D2O]
(for kinetic plots see Figure S8, SI). Linear regression of
a plot of log10(rCH3D) vs log10([D2O]) ([D2O] =0.69, 1.38,
and 2.12 M) yielded a reaction order with respect to water
of 1.2 ( 0.1 (from the slope), consistent with the rate
(20) Comparison of the average value of 2.4 ꢀ 10-4 s-1 for kins,Me,1-
tmeda with k*ins,Me,1-dmso ≈ 4.5 ꢀ 10-4 s-1 at [C2H4] ≈ 0.1 M (calculated
from kins,Me,1-dmso=6.8 ꢀ 10-4 s-1 at [C2H4] ≈ 0.15 M; see ref 13)
indicates that activation of 1-tmeda may occur preferentially from
1-dmso instead of 1-κ2-tmeda under the conditions studied. Considering
1-κ2-tmeda as a dormant and 1-dmso as a reactive component toward
ethylene under the conditions studied, independent calculation of a
corrected pseudo-first-order insertion rate constant according-1to
kins,Me,1-tmeda=k*ins,Me,1-dmso (1/(1 + [tmeda]free{KMe(328 K)[dmso-d6]} ))
yields k*ins,Me,1-tmeda=3. 4 ꢀ 10-4 s-1 ([1-tmeda]t=0=34 mM; [dmso- d6]=
14.14 M; KMe(328 K)=1.2 ꢀ 10-2).

Article
Organometallics, Vol. 28, No. 14, 2009 4053
(R > Me), which are the active species in ethylene polymer-
ization. In the absence of tmeda,13 a homocoupling of higher
Ni(II)-alkyl moieties had not been observed (but higher Ni-
(II)-alkyls coupled with Ni(II)-Me present). However, the
formation of alkanes from the bimolecular coupling reaction
of Ni(II)-alkyl with Ni(II)-hydride species, formed by β-H
elimination, was found to be a relevant deactivation pathway
intrinsic to neutral Ni(II) catalysts. In view of the specific
involvement of tmeda coordination in deactivation reactions
of Ni(II)-Me species, namely, both bimolecular elimination
of ethane and hydrolysis, the effect of tmeda on deactivation
of the active species occurring during polymerization was
studied.
The gradual formation of CH4 and C2H6 was observed
from solutions of dmso- and tmeda-coordinated Ni(II)-
methyl and Ni(II)-ethyl species in the presence of ethylene
1
under nonaqueous reaction conditions at T=55 °C (see H
NMR spectrum at the top of Figure 5). The addition of
increasing amounts of water (D2O) resulted in the formation
1
of CH3D (see above) in addition to CH4 and C2H6 (see H
NMR spectrum at the bottom of Figure 5), but no hydrolysis
products of higher Ni(II)-alkyl species, e.g., C2H5D, were
detected.
Summary and Conclusion
The NMR spectroscopic studies reported reveal the
very specific nature and reactivity of tmeda-coordinated
salicylaldiminato Ni(II)-alkyl complexes, which are highly
reactive and versatile intermediates for preparative organo-
metallic chemistry as well as olefin polymerization catalysis.
The NMR spectrocopic studies of the Ni(II)-methyl com-
plex 1-tmeda reported are consistent with two relevant
binding modes of tmeda, an open κ1- and, unexpectedly,
a chelating κ2-coordination fashion (Scheme 1), which are
in equilibrium with each other and the solvent species 1-L
(L=dmso, methanol) and interconvert slowly on the NMR
chemical shift time scale. The position of the equilibria
depends on the solvent properties, σ-donor ability and polar-
ity, and the temperature. Coordination in a chelating κ2-
fashion renders the salicylaldimine κ1-O-coordinate. That
tmeda is κ2-coordinate under these conditions is somewhat
suprising in view of the lability of tmeda toward other ligands
such as pyridine or phosphines, in combination with the
finding that even an excess of the latter (monodentate) ligands
does not result in displacement of the imine moiety.7c-7m
The exposure of dmso solutions of 1-tmeda to ethylene in
an NMR tube resulted in the partial reaction of 1-tmeda (at a
very low degree of conversion of ∼10% of Ni(II)-methyl)
to form propylene and the dmso- and tmeda-coordinated
Ni(II)-ethyl species 2-dmso and 2-κ2-tmeda at 55 °C, which
Figure 3. Pseudo-first-order plots of the time-dependent de-
crease of 1-tmeda in the presence of ethylene and 0 to 60 equiv
of D2O (pseudo-first-order rate constants kobs,Me,1-tmeda deter-
mined by linear regression); 34 mM 1-tmeda, ∼0.1 M ethylene.
Figure 4. Concentration dependence of observed initial rates of
hydrolysis rCH3D on added water as a plot of log10(rCH3D) vs
log10([D2O]) ([1-tmeda]t=0 = 34 mM, [D2O] = 0.69, 1.38, and
2.12 M, T = 55 °C).
Figure 5. Formation of the decomposition products CH4 (CH3D) and C2H6 of Ni(II)-alkyls prepared from the reaction of 1-tmeda
with C2H4 in the absence (presence) of water (20 equiv of D2O; bottom spectrum) at T = 55 °C in the NMR tube.

4054 Organometallics, Vol. 28, No. 14, 2009
Berkefeld et al.
were observed in a close to 1:1 equilibrium mixture after
cooling of the sample to 25 °C, demonstrating that
κ2-coordination of tmeda is a relevant binding mode also
in higher Ni(II)-alkyl species. Ethylene is catalytically dimer-
ized to butenes. The interconversion of RC and βC moieties
of 2-κ2-tmeda apparently proceeds slowly as compared to the
corresponding interconversion observed for 2-dmso and the
ligand exchange of tmeda vs dmso between the Ni(II)-ethyl
species. Studies of the binding affinities of tmeda toward Ni-
(II)-alkyls in dmso-d6 (methanol-d4) solution, quantified as
the equilibrium constants KMe(T) and KEt(T) by variable-
temperature 1H NMR spectroscopy, showed that the coor-
dination of tmeda is favored by ∼2-3 oders of magnitude
over the solvent complexes for temperatures from 25 to
55 °C. Overall, by comparison to the analogous tmeda-free
system,13 the formation of tmeda complexes merely retards
the ethylene insertion reactions, for which the κ2-tmeda
species appear to be unreactive, i.e., dormant. However,
tmeda specise are actively involved in deactivation reactions.
Bimolecular coupling of Ni(II)-methyl moieties to ethane is a
relevant deactivation pathway of 1-tmeda under the reaction
conditions studied. An Eyring analysis of second-order rate
constants of ethane formation, kMe-Me,1-tmeda(T), afforded
2-dmso) gradually decomposed via bimolecular elimination
reactions involving either unreacted Ni(II)-methyl or Ni(II)-
hydride species to form CH4 and C2H6, consistent with observa-
tions of solutions of Ni(II)-alkyls in the absence of tmeda.
In summary, these findings reveal that tmeda, which is
introduced in the synthesis of Ni(II) complexes and also in
in situ catalysts via the reactive and versatile precursor
[(tmeda)Ni(CH3)2], is not an innocent and relatively weakly
coordinating ligand but can displace chelating formally
monoanionic ligands and specifically promotes irreversible
decomposition of Ni(II)-Me species by hydrolysis as well as
bimolecular reductive coupling.
Experimental Section
All manipulations of air- and moisture-sensitive substances
were carried out using standard Schlenk, vacuum, and glovebox
techniques under argon or dinitrogen. Deuterated solvents
(purity and degree of deuteration g99.5%) were purchased
from Eurisotop. All solvents were thoroughly degassed and
saturated with argon prior to use. Dimethyl sulfoxide was
˚
distilled from 4 A molecular sieves or freshly calcined CaO,
toluene-d8 was distilled from a NaK alloy, and methanol-d4 was
used as received. The salicylaldimine (N,O)H, 1-C(H)dNAr-2-
OH-3,5-I2C6H2 (Ar=2,6-(3,5-(F3C)2C6H3)2C6H3), and 1-tme-
da were prepared by modification of reported procedures.7h
[(tmeda)NiMe2] was purchased from MCAT (Konstanz) and
stored at -30 °C in a glovebox. Ethylene of 3.5 grade supplied by
Gerling Holz + Co was used as received. NMR spectra were
recorded on a Varian Inova 400 spectrometer equipped with either
a direct or indirect detection broadband probe, or a Bruker Avance
DRX 600 instrument equipped with an H/C/N-TXI inverse probe.
1H and 13C chemical shifts were referenced to residual proton and
the naturally abundant 13C resonance of the deuterated solvent,
respectively. Assignments of chemical shifts are based on 1H,
13C{1H}, DQF-COSY, 1H,13C-HSQC, 1H,13C-HMBC, ROESY,
NOESY, and TOCSY NMR spectra. In order to avoid TOCSY
artifacts during ROESY, a 180x-180-x spin lock scheme was
employed.21 When necessary, high-intensity signals of the solvent,
tmeda, and butenes were suppressed by presaturation22 at multiple
irridiation frequencies and/or WATERGATE.23 An overall
recycle delay of 7 and 5 s was applied for the collection of
1H and 13C NMR spectroscopic data, respectively. In general,
variable-temperature NMR spectroscopy was carried out with
prior temperature determination using a sample of neat methanol
(T e 40 °C) or ethylene glycol (T g 40 °C). NMR spectra
were processed and analyzed with MestreNOVA (v5.3.0). Gel
permeation chromatography (GPC) was carried out in 1,2,
4-trichlorobenzene/0.0125% BHT at 160 °C at a flow rate of
1 mL min-1 on a Polymer Laboratories 220 instrument equipped
with Olexis columns with differential refractive index, viscosity,
and light-scattering (15° and 90°) detectors. Data reported were
determined via triple detection employing the PL GPC-220 soft-
ware algorithm. The instrument was calibrated with narrow poly-
styrene and polyethylene standards; data given are referenced to
linear polyethylene. Differential scanning calorimetry (DSC) was
measured on a Netzsch DSC 204 F1 instrument with a heating/
cooling rate of 10 K min-1. DSC data reported are from second
heating cycles.
activation parameters ΔHq= 97 ( 7 kJ mol-1 and ΔSq
≈
-5 J mol-1 K-1. These differ substantially from ΔHq=57 (
1 kJ mol-1 and ΔSq=-(129 ( 2) J mol-1 K-1 in the absence of
tmeda. This demonstrates that dissociation of coordinated
tmeda presumably occurs in a concerted fashion with the
rate-determining step of the overall reductive coupling reac-
tion. A tentative explanation for the higher enthalpy is the
need for breaking of Ni-N bonds to tmeda, which is partly
compensated by the degrees of freedom gained by opening of
the rigid five-membered κ2-tmeda chelate.
In the presence of water, hydrolysis of the Ni(II)-Me
bond occurs to a significant extent in dmso solutions of
1-tmeda. This contrasts with the behavior of the analogous
tmeda-free system, for which no substantial hydrolysis oc-
curred under identical conditions. Likely, the hydrogen-
bonding capability of free amine moieties is relevant, e.g.,
interaction with water in the intermediate 1-κ1-tmeda (or in
1-κ2-tmeda) could promote water coordination and hydro-
lysis. Quantitative kinetic data of the time-dependent
formation of methane (rCH3D) from aqueous (D2O) dmso-
d6 solutions of 1-tmeda in the presence of ethylene at T=
55 °C are consistent with a rate law rCH3D=khydr,1-tmeda[1-
tmeda][D2O], khydr,1-tmeda=1.9 ꢀ 10-4 M-1 s-1 (ΔGq=
113 kJ mol-1 for [1-tmeda]t=0=34 mM). Splitting of observed
pseudo-first-order rate constants for the disappearance of the
total Ni(II)-methyl resonance of 1-tmeda (as the sum of 1-dmso
and 1-κ2-tmeda) into the contributions of ethylene insertion and
Ni(II)-methyl bond hydrolysis according to kobs,Me,1-tmeda
=
kins,Me,1-tmeda + khydr,1-tmeda[D2O] confirmed that the insertion
rate constant kins,Me,1-tmeda is invariant irrespective of the ab-
sence or presence of water.
In contrast to the promotion of irreversible decomposition
of Ni(II)-Me complexes in the presence of tmeda by hydro-
lysis as well as bimolecular coupling of tmeda complexes, no
evidence for an analogous enhancement of decomposition of
higher alkyl (R > Me) complexes, which are the intrinsic
intermediates of polymerization catalysis, was observed
under the conditions studied. By contrast to CH3D forma-
tion from 1-tmeda, no hydrolysis products of higher Ni(II)-
alkyl species were observed. Mixtures of Ni(II)-methyl (1-κ2-
tmeda and 1-dmso) and Ni(II)-ethyl species (2-κ2-tmeda and
Characterization of 1-κ2-tmeda. 1H NMR (dmso-d6,
600 MHz, 25 °C): δ 11.08 (s, 1H, H7); 8.24 (br s, 4H, H15, 21,
23, 29); 7.90 (br s, 2H, H18, 26); 7.77 (d, JH-H=7.6 Hz, 2H,
3
(21) Hwang, T.-L.; Shaka, A. J. J. Am. Chem. Soc. 1992, 114, 3157–
3159.
(22) (a) Hoult, D. I. J. Magn. Reson. 1976, 21, 337–347. (b) Schaefer,
J. J. Magn. Reson. 1972, 6, 670–671.
(23) Piotto, M.; Saudek, V.; Sklenar, V. J. Biomol. NMR 1992, 2,
661–665.

Article
Organometallics, Vol. 28, No. 14, 2009 4055
H10, 12); 7.77 (br s, 1H, H6); 7.63 (br s, 1H, H4); 7.48 (br t,
3JH-H=7.6 Hz, 1H, H11); 2.36, 2.01, 1.82, and 1.64 (br s, 3H each,
4ꢀ N-CH3, κ2-tmeda); 2.09 (obscured by N-methylene protons of
free tmeda, N-CH2-, κ2-tmeda); -1.79 (s, 3H, Ni(II)-CH3) ppm.
13C NMR (dmso-d6, 151 MHz, 25 °C): δ 168.9 (C2); 161.2 (C7);
150.4 (C8); 146.9 (C4); 141.3 (C14, 22); 134.4 (C6), 129.7 (C9, 13);
131.0 (C10, 12); 130.3 (C15, 21, 23, 29); 129.9 (q, 2JC-F=33 Hz, C16,
19, 24, 27); 125.1 (C11); 123.0 (q, 1JC-F=273 Hz, C17, 20, 25, 28);
120.0 (C18, 26); 118.9 (C1); 99.5 (C3); 73.6 (C5); 60.4 and 54.6
(2ꢀ N-CH2-, κ2-tmeda); 45.9, 47.5, 45.9, and 48.6 (4ꢀ N-CH3,
κ2-tmeda); -15.0 (Ni(II)-CH3) ppm.
General Procedure for Quantitative Studies of 1-tmeda by
Variable-Temperature 1H NMR Spectroscopy. In typical experi-
ments, a 34 mM dmso-d6 solution of 1-tmeda was prepared in an
NMR tube sealed with a rubber septum in the glovebox. The
tubes were kept at room temperature outside the spectrometer,
and NMR spectra were acquired with the NMR probe pre-
warmed to desired temperatures. Relative concentrations
were determined from integration of the respective resonances
and normalization to an integral value for Ni(II)-CH3=
1 ([1-tmeda]t=0=34 mM; [dmso-d6]=14.14 M). Equilibrium
constants were calculated from 1H NMR spectra, linear regres-
sion of plots of ln(K(T)) vs reciprocal temperatures yielded
reaction enthalpies ΔH°, and free reaction enthalpies ΔG° were
calculated from equilibrium constants at 25 °C. Water ([D2O] in
the range from 0.69 to 1.38 and 2.12 M equal to 20, 40, and
60 equiv per 1-tmeda) and ethylene (3-4 mL) were added via
gastight syringes. The gradual disappearance of the Ni(II)-
CH3 resonances of 1-tmeda was monitored at T= 55 °C
after ethylene addition. Linear regression yielded values of
kobs,Me,1-tmeda (from plots of ln([1-tmeda]t/[1-tmeda]t=0) vs
time) and the reaction order with respect to water and khydr,1-
tmeda (from a plot of log10(rCH3D) vs log10([D2O])).
General Procedure for the Kinetic Analysis of Ethane Forma-
tion from 1-tmeda. NMR samples of 1-tmeda in dmso-d6 (34
mM) were prepared in a glovebox. Tubes were sealed with a
rubber septum and removed from the glovebox. Samples were
inserted into the prewarmed NMR probe and the decrease of the
total Ni(II)-CH3 resonance (as the sum of 1-dmso and 1-κ2-
1
tmeda) was followed with time by H NMR spectroscopy for
temperatures from 55 to 80 °C. Integral values were normalized
to the value at t=0 and converted into concentrations by
multiplication with the starting concentration of 1-tmeda. Sec-
ond-order rate constants kMe-Me,1-tmeda(T) were calculated from
linear regression of plots of 1/[1-tmeda]t - 1/[1-tmeda]t=0
vs time. Linear regression of a plot of ln{kMe-Me,1-tmeda(T)/T}
vs 1/T provided the activation parameters ΔHq and ΔSq.
Acknowledgment. Financial support by the DFG
(Me1388/3-2) is gratefully acknowledged. We thank
Anke Friemel for technical support and Lars Bolk for
GPC and DSC measurements. S.M. is indebted to the
Fonds der Chemischen Industrie and to the Hermann
Schnell-Foundation.
Supporting Information Available: General procedure of
high-pressure polymerization experiments, NMR spectra of
1-κ2-tmeda and 2-κ2-tmeda in dmso-d6 solution, and plots for
the determination of kinetic and thermodynamic data obtained
from NMR spectroscopic studies. This material is available free
of charge via the Internet at http://pubs.acs.org.