Preparation of an active neodymium catalyst for regioselective butadiene cis-polymerization supported by a dianionic modification of the 2,6-diiminopyridine ligand

Article abstract of DOI:10.1021/om0496434

Treatment of the 2,6-diiminopyridine ligand 2,6-{[2,6-(i-Pr) ₂C₆H₃]N=C(CH₃)}₂(C ₅H₃N) with 2 equiv of Me₃SiCH₂Li afforded the corresponding {[2,6-{[2,6-(i-Pr)₂C₆H ₃]N-C=(CH₂)}₂-(C₅H ₃N)]^2- (1) dianion via deprotonation of the two methyl groups attached to the two imine functions. Reaction of 1 with NdCl ₃(THF)₃ yielded {[2,6-{[2,6-(i-Pr)₂C ₆H₃]N-C=(CH₂)}₂-(C₅H ₂N)]Nd(THF)}(μ-Cl)₂[Li(THF)₂]·0. 5(hexane) (2a), whose recrystallization from DME gave the corresponding ionic {[2,6-{ [2,6-(i-Pr)₂C₆H₃]N-C=(CH ₂)}₂(C₅H₂N)]NdCl₂(DME)} {Li(DME)₃]-(2b). With the exception of the reaction with (allyl)MgBr, which proceeded readily with 2b to form the allyl derivative {[2,6-{[2,6-(i-Pr)₂C₆H₃]N-C=(CH ₂)}₂(C₅H₃N)]Nd(η³- C₃H₅)Br}{Li-(DME)₃} (3), compounds 2a,b are not suitable starting substrates for further alkylation reactions. A viable synthetic strategy for the preparation of alkyl derivatives of 2 consisted instead of the pretreatment of NdCl₃(THF)₃ with RLi [R = Me₃SiCH₂, CH₃] at low T followed by treatment with either the diimine ligand or 1. According to this procedure, the terminally bound alkyl derivative [2,6-{[2,6-(i-Pr)₂C₆H ₃]N-C=(CH₂)}₂(C₅H ₃N)]Nd[CH₂Si (CH₃)₃](THF) (4) was prepared and subsequently crystallographically characterized. The same synthetic procedure with MeLi afforded {[2,6-{[2,6-(i-Pr)₂C₆H ₃]N-C=(CH₂)}₂(C₅H ₃N)]Nd}(μ-CH₃)₂[Li-(THF)₂] (5) and {[2,6-{[2,6-(i-Pr)₂C₆H₃]N-C=(CH ₂)}₂(C₅H₃N)]Nd}μ-Cl)(μ-X)[Li (THF)₂] (6) [X = Cl 53%, Me 47%] depending on the MeLi/Nd ratio. Both 5 and 6 as well as 2a are potent catalysts for the cis-polymerization of butadiene.

Full text of DOI:10.1021/om0496434

5054
Organometallics 2004, 23, 5054-5061
P r ep a r a tion of a n Active Neod ym iu m Ca ta lyst for
Regioselective Bu ta d ien e cis-P olym er iza tion Su p p or ted
by a Dia n ion ic Mod ifica tion of th e 2,6-Diim in op yr id in e
Liga n d
Hiroyasu Sugiyama,† Sandro Gambarotta,*,† Glenn P. A. Yap,†
David R. Wilson,*,‡ and Sven K.-H. Thiele§
Department of Chemistry, University of Ottawa, Ottawa, Ontario, K1N 6N5, Canada,
Chemical Sciences Capability, The Dow Chemical Company, Midland, Michigan 48674,
and Dow Olefinverbund GmbH (An Affiliate of The Dow Chemical Company),
Werk Schkopau, Merseburg, Germany
Received May 17, 2004
Treatment of the 2,6-diiminopyridine ligand 2,6-{[2,6-(i-Pr)2C6H3]NdC(CH3)}2(C5H3N) with
2 equiv of Me3SiCH2Li afforded the corresponding {[2,6-{[2,6-(i-Pr)2C6H3]N-Cd(CH2)}2-
(C5H3N)]2- (1) dianion via deprotonation of the two methyl groups attached to the two imine
functions. Reaction of 1 with NdCl3(THF)3 yielded {[2,6-{[2,6-(i-Pr)2C6H3]N-Cd(CH2)}2-
(C5H3N)]Nd(THF)}(µ-Cl)2[Li(THF)2]‚0.5(hexane) (2a ), whose recrystallization from DME gave
the corresponding ionic {[2,6-{[2,6-(i-Pr)2C6H3]N-Cd(CH2)}2(C5H3N)]NdCl2(DME)}{Li(DME)3]-
(2b). With the exception of the reaction with (allyl)MgBr, which proceeded readily with 2b
to form the allyl derivative {[2,6-{[2,6-(i-Pr)2C6H3]N-Cd(CH2)}2(C5H3N)]Nd(η3-C3H5)Br}{Li-
(DME)3} (3), compounds 2a ,b are not suitable starting substrates for further alkylation
reactions. A viable synthetic strategy for the preparation of alkyl derivatives of 2 consisted
instead of the pretreatment of NdCl3(THF)3 with RLi [R ) Me3SiCH2, CH3] at low T followed
by treatment with either the diimine ligand or 1. According to this procedure, the terminally
bound alkyl derivative [2,6-{[2,6-(i-Pr)2C6H3]N-Cd(CH2)}2(C5H3N)]Nd[CH2Si (CH3)3](THF)
(4) was prepared and subsequently crystallographically characterized. The same synthetic
procedure with MeLi afforded {[2,6-{[2,6-(i-Pr)2C6H3]N-Cd(CH2)}2(C5H3N)]Nd}(µ-CH3)2[Li-
(THF)2] (5) and {[2,6-{[2,6-(i-Pr)2C6H3]N-Cd(CH2)}2(C5H3N)]Nd}(µ-Cl)(µ-X)[Li (THF)2] (6)
[X ) Cl 53%, Me 47%] depending on the MeLi/Nd ratio. Both 5 and 6 as well as 2a are
potent catalysts for the cis-polymerization of butadiene.
In tr od u ction
unanticipated ability of the diimine to be involved in
the reactivity of the metal center3,4 and to undergo a
number of diverse transformations.4 Of particular inter-
est to us is the possibility shown by the conjugated
π-system of efficiently accepting considerable amounts
of electronic charge to form radical anions.5 The recent
isolation and characterization of a NdI2 adduct6 indi-
The discovery of the ability of the tridentate diimi-
nopyridine ligand 2,6-{[2,6-(i-Pr)2 C6H3]NdC(CH3)}2-
(C5H3N) to support exceptionally high levels of catalytic
activity with several transition metals of very different
natures1 has marked a milestone in Ziegler-Natta
catalysis and also prompted efforts aimed at under-
standing the reasons for the unique performance of this
ligand.2 The results of these studies have shown an
(3) (a) Bruce, M.; Gibson, V. C.; Redshaw, C.; Solan, G. A.; White,
A. J . P.; Williams, D. J . J . Chem. Soc., Chem. Commun. 1998, 2523.
(b) Korobkov, I.; Gambarotta, S.; Yap, G. P. A.; Budzelaar, P. H. M.
Organometallics 2002, 21, 3088. (c) Clentsmith, G. K. B.; Gibson, V.
C.; Hitchcock, P. B.; Kimberly, B. S.; Rees, C. W. J . Chem. Soc.,
Chem. Commun. 2002, 1498. (d) Kooistra, T. M.; Knijnenburg, Q.;
Smits, J . M. M.; Horton, A. D.; Budzelaar, P. H. M.; Gal, A. W. Angew.
Chem., Int. Ed. 2001, 40, 4719. (e) Gibson, V. C.; Humphries,
M. J .; Tellmann, K. P.; Wass, D. F.; White, A. J . P.; Williams, D. J .
Chem. Commun. 2001, 2252. (f) Gibson, V. C.; Tellmann, K. P.;
Humphries, M. J .; Wass, D. F. Chem. Commun. 2002, 2316. (g)
Britovsek, G. J . P.; Bruce, M.; Gibson, V. C.; Kimberley, B. S.; Maddox,
P. J .; Mastroianni, S.; McTavish, S. J .; Redshaw, C.; Solan, G. A.;
Stroemberg, S.; White, A. J . P.; Williams, D. J . J . Am. Chem. Soc. 1999,
121, 8728.
(4) (a) Reardon, D.; Conan, F.; Gambarotta, S.; Yap, G. P. A.; Wang,
Q. J . Am. Chem. Soc. 1999, 121, 9318. (b) Sugiyama, H.; Aharonian,
G.; Gambarotta, S.; Yap, G. P. A.; Budzelaar, P. H. M. J . Am. Chem.
Soc 2002, 124, 12268. (c) Reardon, D.; Aharonian, G.; Gambarotta, S.;
Yap, G. P. A. Organometallics 2002, 21, 786. (d) De Bruin, B.; Bothe,
E.; Weyhermuller, T.; Wieghardt, K. Inorg. Chem. 2000, 39, 2936.
(5) Enright, D.; Gambarotta, S.; Yap, G. P. A.; Budzelaar, P. H. M.
Angew. Chem., Int. Ed. 2002, 41, 3873.
* To whom correspondence should be addressed. E-mail: sgambaro@
science.uottawa.ca.
† University of Ottawa.
‡ The Dow Chemical Company.
§ Dow Olefinverbund GmbH.
(1) (a) Small, B. L.; Brookhart, M. S.; Bennett, A. M. A. J . Am. Chem.
Soc. 1998, 120, 4049. (b) Britovsek, G. J . P.; Gibson, V. C.; Kimberley,
B. S.; Maddox, P. J .; McTavish, S. J .; Solan, G. A.; White A. J . P.;
Williams, D. J . J . Chem. Soc., Chem. Commun. 1998, 849. (c)
Esteruelas, M. A.; Lopez, A. M.; Mendez, L.; Olivan, M.; Onate, E.
Organometallics 2003, 22, 395. (d) Bryliakov, K. P.; Semikolenova, N.
V.; Zudin, V. N.; Zakharov, V. A.; Talsi, E. P. Catal. Commun. 2004,
5, 45. (e) Britovsek, G. J . P.; Gibson, V. C.; Spitzmesser, S. K.;
Tellmann, K. P.; White, A. J . P.; Williams, D. J . J . Chem. Soc., Dalton
Trans. 2002, 6, 1159.
(2) (a) Budzelaar, P. H. M.; de Bruin, B.; Gal, A. W.; Wieghardt, K.;
van Lenthe, J . H. Inorg. Chem. 2001, 40, 4649. (b) Margl, P.; Deng,
L.; Ziegler, T. Organometallics 1999, 18, 5701. (c) Griffiths, E. A. H.;
Britovsek, G. J . P.; Gibson, V. C.; Gould, I. R. Chem. Commun. 1999,
1333.
10.1021/om0496434 CCC: $27.50 © 2004 American Chemical Society
Publication on Web 09/08/2004

Preparation of an Active Nd Catalyst
Organometallics, Vol. 23, No. 21, 2004 5055
3.20 mmol, 77%). Crystals suitable for X-ray analysis were
grown from a THF-hexane mixture. 1H NMR (C6D6, 500 MHz,
25 °C) δ: 1.27 (m, 20H, OCH2CH2), 1.32 (d, J ) 7.0 Hz, 12H,
(CH3)2CH), 1.57 (d, J ) 6.9 Hz, 12H, (CH3)2CH), 3.18 (m, 20H,
OCH2CH2), 3.41 (br s, 2H, NCdCH2), 3.74 (quint, J ) 6.9 Hz,
4H, (CH3)2CH), 4.05 (br s, 2H, NCdCH2), 7.17 (t, J ) 7.7 Hz,
cates that this characteristic can effectively be used for
the stabilization of highly reactive species.
During these studies, the ability to perform mono- and
double-deprotonations of the methyl groups attached
directly to the imine carbon atoms was serendipitously
discovered.3b,4b This has in turn provided a novel class
of dianionic ligands which, although very similar in
structure to the versatile diketimine ligand, are elec-
tronically very different. Given the remarkable perfor-
mance of the intact ligand in supporting exceptionally
high catalytic activity,1,3 the next most obvious step
was to probe the rational preparation and use of the
mono- and double-deprotonated forms of 2,6-{[2,6-(i-
Pr)2C6H3]NdC(CH3)}2(C5H3N) as supporting ligands for
polymerization catalysts. The dianion in particular is
of interest given its close relationship to a large family
of dianionic chelating ligands based on nitrogen donor
atoms, which have been employed for the preparation
of nonmetallocene catalysts7 capable of performing a
wide range of polymerization reactions including ste-
reospecific polymerization of dienes.8
i
1H, p-C5H3N), 7.24 (t, J ) 7.6 Hz, 2H, p-C6H3 Pr2), 7.39 (d, J
i
) 7.6 Hz, 4H, m-C6H3 Pr2), 7.76 (d, J ) 7.7 Hz, 2H, m-C5H3N).
13C{1H} NMR (C6D6, 123.72 MHz, 25 °C) δ: 25.52 ((CH3)2CH)),
25.54 ((CH3)2CH)), 25.78 (OCH2CH2), 28.56 ((CH3)2CH), 68.54
(OCH2CH2), 71.74 (NCdCH2), 118.06 (m-C5H3N), 121.93 (p-
i
i
C6H3 Pr2), 123.69 (m-C6H3 Pr2), 136.72 (p-C5H3N), 144.18 (o-
i
i
C6H3 Pr2), 153.61 (ipso-C6H3 Pr2), 160.31 (o-C5H3N), 163.36
(NCdCH2). Meaningful analytical data could not obtained
probably due to spontaneous loss of the crystallization solvent.
P r ep a r a t ion of {[2,6-{[2,6-(i-P r )2C6H 3]N-Cd(CH 2)}2-
(C5H3N)]Nd(THF)}(µ-Cl)2[Li(THF)2]‚0.5 (h exan e) (2a). Com-
pound 1 (1.50 g, 1.76 mmol) was added to a solution of
NdCl3(THF)3 (0.81 g, 1.75 mmol) in THF (60 mL), and the
resulting mixture was stirred overnight at room temperature.
After removal of the volatiles, the residue was extracted with
Et2O (80 mL) and centrifuged to remove LiCl. After further
evaporation to dryness, the residue was recrystallized from
THF (10 mL)/hexane (80 mL) to give brown-orange crystals
of the crude product, which were again recrystallized to give
analytically pure 2a (0.93 g, 0.97 mmol, 55%). Anal. Calcd
(found) for C48H72Cl2NdLiN3O3: C 59.98 (58.58), H 7.55 (7.36),
N 4.37 (4.50), Cl 7.38 (7.27), Nd 15.01 (15.10). The NMR
spectra showed only very broad and overlapping features. µeff
) 3.65 µB.
In this first paper we describe the preparation and
characterization of the [2,6-{[2,6-(i-Pr)2C6H3]N-Cd
(CH2)}2(C5H3N)]2- dianion and its use in the prepara-
tion of potent Nd catalysts for stereoselective butadiene
polymerization.
Exp er im en ta l Section
P r ep a r a t ion of {[2,6-{[2,6-(i-P r )2C6H 3]N-Cd(CH 2)}2-
(C5H3N)]Nd Cl2(DME)}{Li(DME)3] (2b). Complex 2a (0.93
g, 0.97 mmol) was dissolved into a minimum amount of DME
(20 mL) with heating. The solution was allowed to stand at
room temperature overnight, upon which crystals of 2b
separated (0.78 g, 0.74 mmol, 76% yield). Anal. Calcd (found)
for C49H81Cl2NdLiN3O8: C 55.40 (55.28), H 7.69 (7.66), N 3.96
(3.89). µeff ) 3.59 µB.
P r ep a r a t ion of {[2,6-{[2,6-(i-P r )2C6H 3]N-Cd(CH 2)}2-
(C5H3N)]Nd (η3-C3H5)Br }{Li(DME)3} (3). Complex 2a (0.19
g, 0.20 mmol) was gradually dissolved into DME (8 mL), from
which it precipitated as 2b. The addition of C3H5MgBr (0.2
M, 2 mL, 0.40 mmol) at 0 °C to the suspension gave a
homogeneous orange solution. An insoluble material was
gradually formed while stirring overnight at ambient temper-
ature. The mixture was concentrated to about 6 mL and
centrifuged. Cyclohexane (2 mL) was added to the solution,
and the resulting suspension was allowed to stand for 1 day
at room temperature. The precipitated solid was eliminated
by centrifugation, and the clear solution was layered with
cyclohexane to afford crystals of 3 (0.14 g, 0.11 mmol, 56%
yield). Anal. Calcd (found) for C48H76N3BrNdLiO6: C 56.40
(56.38), H 7.49 (7.36), N 4.11 (4.02). µeff ) 3.61 µB.
P r ep a r a tion of [2,6-{[2,6-(i-P r )2C6H3]N-Cd(CH2)}2-
(C5H3N)]Nd (CH2Si(CH3)3)(THF ) (4). Meth od A. A THF (20
mL) solution of NdCl3(THF)3 (0.23 g, 0.5 mmol) was treated
at -60 °C with a THF (24 mL) solution of LiCH2Si(CH3)3 (0.09
g, 1.0 mmol.). Stirring was continued at 0 °C for 1 h. A THF
(30 mL) solution of 1 (0.43 g, 0.5 mmol) was added, and the
resulting mixture was stirred overnight. Solvent was evapo-
rated to dryness. The residue was extracted with warm hexane
(50 °C, 200 mL), and the extract was transferred to a Schlenk
tube and concentrated to 40 mL. The mixture was warmed
again with a hot water bath and centrifuged to remove the
colorless precipitate. The resulting green solution was cooled
to -36 °C, upon which yellowish-green crystals of 4 separated
(0.21 g, 0.26 mmol, 52%). Anal. Calcd (found) for C47H74N3-
NdOSi: C 64.93 (64.56) H 8.58 (8.39) N 4.83 (4.71). µeff ) 3.65
µB.
All operations were performed under an inert atmosphere
by using standard Schlenk type techniques. NdCl3(THF)39 and
2,6-{[2,6-(i-Pr)2C6H3]NdC(CH3)}2(C5H3N)1a were prepared ac-
cording to published procedures. Samples for magnetic sus-
ceptibility measurements at room temperature were pre-
weighed inside a drybox equipped with an analytical balance
and flame sealed into calibrated 5 mm o.d. quartz tubes.
Magnetic measurements were carried out using a J ohnson
Matthey magnetic balance. Background corrections for the
sample holder were included in the magnetic calculations.
Standard corrections for the underlying diamagnetism were
applied to the data.10 Elemental analyses were carried out with
a Perkin-Elmer 2400 CHN analyzer. Data for the X-ray crystal
structure determinations were obtained with a Bruker dif-
fractometer equipped with a Smart CCD area detector. NMR
spectra were recorded on a Bruker AMX-500 spectrometer.
P r ep a r a t ion of {[2,6-{[2,6-(i-P r )2C6H 3]N-Cd(CH 2)}2-
(C5H3N)]Li(THF )}{Li(THF )4} (1). Treatment of a solution
of 2,6-{[2,6-(i-Pr)2C6H3]NdC(CH3)}2(C5H3N) (2.0 g, 4.15 mmol)
in THF (30 mL) with LiCH2Si(CH3)3 (0.85 g, 9.03 mmol) in
THF (40 mL), followed by evaporation of the solvent, afforded
a solid residue. The mass was washed with two portions of
hexane (20 mL) and dried under vacuum. Compound 1 was
thus obtained as a very air-sensitive yellow powder (2.73 g,
(6) Sugiyama, H.; Korobkov, I.; Gambarotta, S.; Mo¨ller, A.; Budze-
laar, P. H. M. Inorg. Chem., in press.
(7) (a) Estler, F.; Eickerling, G.; Herdtweck, E.; Anwander, R.
Organometallics 2003, 22, 1212. (b) Ward, B. D.; Dubberley, S. R.;
Maisse-Franc¸ois, A.; Gade, L. H.; Mountford, P. J . Chem. Soc., Dalton
Trans. 2002, 4649. (c) Piers, W. E.; Emslie, D. J . H. Coord. Chem. Rev.
2002, 233-234, 131.
(8) (a) Hou, Z.; Wakatsuki, Y. Coord. Chem. Rev. 2002, 231, 1. (b)
Yasuda, H. Top. Organomet. Chem. 1999, 2, 255.
(9) (a) Anhydrous NdCl3 was prepared following a standard proce-
dure: (a) Freeman, J . H.; Smith, M. L. J . Inorg. Nucl. Chem. 1958, 7,
224, and was transformed into the corresponding tetrahydrofuranate.
(b) Manzer, L. E. Inorg. Synth. 1982, 21, 135. (c) Angelici, R. J . Inorg.
Synth. 1990, 27, 136.
(10) (a) Mabbs, M. B.; Machin, D. Magnetism and Transition Metal
Complexes; Chapman and Hall: London, 1973. (b) Foese, G.; Gorter,
C. J .; Smits, L. J . Constantes Selectionnes, Diamagnetisme, Paramag-
netism, Relaxation Paramagnetique; Masson: Paris, 1957.
Meth od B. A 250 mL Schlenk flask was loaded with solid
NdCl3(THF)3 (0.47 g, 1.0 mmol) and LiCH2Si(CH3)3 (0.38 g,

5056 Organometallics, Vol. 23, No. 21, 2004
Ta ble 1. Cr ysta l Da ta a n d Str u ctu r e An a lysis Resu lts
Sugiyama et al.
1
2b
3
4
6
formula
MW
C
53H81Li2N3O5
C
49H81Cl2LiN3NdO8
C
48H76BrLiN3NdO6
C
47H74N3NdOSi
C44.47H66.41Cl1.53NdLiN3O3
908.48
854.09
1062.25
1022.21
869.42
cryst syst
space group, Z
a (Å)
b (Å)
c (Å)
monoclinic
P2/c, 2
11.8557(11)
13.2992(13)
18.4462(17)
monoclinic
P21/c, 4
9.2962(11)
23.836(3)
25.096(3)
monoclinic
P21/n, 4
15.045(2
18.592(3)
20.340(3)
monoclinic
P21/c, 4
15.267(3)
16.123(3)
20.375(4)
triclinic
P1h
10.896(2)
13.002(3)
18.807(4)
70.579(7)
80.152(4)
68.397(4)
2332.7(9)
0.71073
R (deg)
â (deg)
γ (deg)
V (Å3)
λ(Mo KR)
T (K)
108.480(2)
96.225(2)
111.432(2)
109.613(4)
2758.5(5)
0.71073
298(2)
1.028
0.064
5528.0(12)
0.71073
203(2)
1.276
1.085
2228
0.0731, 0.1789
1.025
5295.9(13)
0.71073
223(2)
1.286
1.282
2124
0.0623, 0.572
1.056
4724.4(15)
0.71073
203(2)
1.222
1.159
1836
0.0617, 0.0742
1.074
203(2)
1.293
1.240
946
0.0408, 0.1009
1.030
Dcalcd (g cm-3
)
µcalcd (cm-1
)
F000
R, Rw
932
2 a
0.0963, 0.3235
1.024
GoF
a
2
R ) ∑o - Fc/∑FoRw ) [(∑(Fo - Fc)2/∑wFo )]1/2
.
4.0 mmol). A mixture of hexane (30 mL) and Et2O (10 mL)
was subsequently added after cooling at -60 °C, and the
resulting suspension was stirred for 30 min. The mixture
became emerald green, indicating the formation of partially
alkylated Nd(III) species. To this mixture, a THF solution (40
mL) of the neutral 2,6-diiminopyridine (0.48 g, 1.0 mmol) was
added. The color immediately turned brown. The mixture was
allowed to warm to room temperature, and stirring was
continued overnight. The resulting dark orange-brown mixture
was evaporated to dryness under reduced pressure. The dried
residue was extracted with hexane (60 mL) and centrifuged
to remove some insoluble residue. The resulting green solution
was concentrated to ca. 30 mL and cooled to -36 °C to afford
yellowish-green crystals of 4 (0.19 g, 0.22 mmol, 22%).
P r ep a r a t ion of {[2,6-{[2,6-(i-P r )2C6H 3]N-Cd(CH 2)}2-
(C5H3N)]Nd (THF )}(µ-CH3)2[Li(THF )2]‚0.5(h exa n e) (5). A
suspension of NdCl3(THF)3 (1.45 g, 3.11 mmol) in THF (60 mL)
was treated at 0 °C with a solution of CH3Li in ether (1.4 M,
8.9 mL, 12.5 mmol). The mixture was allowed to warm to room
temperature, upon which the neutral 2,6-diiminopyridine
ligand (1.50 g, 3.11 mol) was added to the mixture. Stirring
was continued overnight. The solvent was evaporated at
reduced pressure, and the residue was extracted with Et2O
(50 mL) and centrifuged to remove a small amount of insoluble
precipitate. After evaporation of Et2O at reduced pressure, the
residue was redissolved in hexane (80 mL) and cooled to -35
°C, upon which crystalline 5 separated (1.46 g, 1.58 mmol,
51%). Anal. Calcd (found) for C50H78NdLiN3O3: C 65.25 (64.58),
H 8.54 (8.36), N 4.57 (4.50). µeff ) 3.63 µB.
P r ep a r a tion of th e Cocr ysta llite {[2,6-{[2,6-(i-P r )2C6H3]-
N-Cd(CH2)}2(C5H3N)]Nd }(µ-Cl)(µ-X)[Li(THF )2] (6) [X ) Cl
53%, Me 47%]. A suspension of NdCl3(THF)3 (1.45 g, 3.11
mmol) in THF (60 mL) containing the neutral 2,6-diiminopy-
ridine ligand (1.50 g, 3.11 mol) was treated with a solution of
CH3Li in ether (1.4 M, 6.7 mL, 9.4 mmol) at 0 °C. The mixture
was allowed to warm to room temperature, and stirring was
continued overnight. The solvent was evaporated at reduced
pressure, and the residue was extracted with Et2O (50 mL)
and centrifuged to remove a small amount of insoluble solid.
After evaporation of Et2O at reduced pressure, the residue was
redissolved in THF (6 mL) and layered with hexane (80 mL)
to afford crystalline 6 (1.25 g, 1.37 mmol, 44%). µeff ) 3.57 µB.
X-r a y Cr yst a llogr a p h y. Compounds 1, 2b , 3, 4, and 6
consistently yielded crystals that diffracted weakly, and the
results presented are the best of several trials. The crystals
were mounted on thin glass fibers using paraffin oil and cooled
to the data collection temperature. Data were collected on a
Bruker AXS SMART 1k CCD diffractometer using 0.3° ω-scans
at 0°, 120°, and 240° in æ. Initial unit cell parameters were
determined from 45 data frames collected at the different
sections of the Ewald sphere. Semiempirical absorption cor-
rections based on equivalent reflections were applied.11 Sys-
tematic absences in the diffraction data-set and unit-cell
parameters were consistent with monoclinic, P2/c for 1, 2b,
and 4; monoclinic, P21/n for 3; triclinic, P1h for 6. Solutions in
the centrosymmetric space groups yielded chemically reason-
able and computationally stable results of refinement. The
structures were solved by direct methods, completed with
difference Fourier syntheses, and refined with full-matrix
2
least-squares procedures based on F . The compound molecule
was not located on symmetry operators in 2b, 3, 4, and 6. Vice
versa, the molecule of 1 was located on a 2-fold axis. One
symmetry-unique molecule of hexane was found in the lattice
of 4. All non-hydrogen atoms, except those of the hexane
solvent molecule in 4, were refined with anisotropic displace-
ment coefficients. Hexane solvent molecule carbon atoms in 4
were refined isotropically. All hydrogen atoms were treated
as idealized contributions. All scattering factors are contained
in several versions of the SHELXTL program library with the
latest version used being v.6.12 (Sheldrick, G. M., Bruker AXS,
Madison, WI, 2001). Crystallographic data and relevant bond
distances and angles are reported in Tables 1 and 2.
Bu ta d ien e P olym er iza tion . The polymerizations were
performed under exclusion of moisture and oxygen under a
nitrogen atmosphere. The products were characterized by
means of SEC (size exclusion chromatography), elemental
analysis, NMR, and IR. The IR samples were prepared by
using CS2 as a swelling agent and using a 2- or 4-fold
dissolution. Mn and Mw were determined by a universal
calibration of SEC against a polystyrene standard. The ratio
between the 1,4-cis, 1,4-trans, and 1,2-polydiene contents of
the butadiene was determined by IR and 13C NMR spectros-
copy. Polymerization experiments were carried out in a double-
wall 2 L steel reactor, which was purged with nitrogen before
the addition of organic solvent, metal complex, and cocatalyst
at 50 °C. In a typical experiment, a solution of MAO cocatalyst
(30.3 mmol) in either cyclohexane or toluene (500 mL) was
treated with butadiene (54 g) and stirred for 1 h. The addition
of a solution of the appropriate amount of catalyst (0.1 mmol)
started the polymerization. The reaction mixture was quenched
with methanol and stripped of the volatile components in
vacuo, and the resulting polymer was dried in an air-circulat-
ing oven at 70 °C for at least 24 h until constant weight.
Resu lts a n d Discu ssion
The possibility of deprotonating the methyl groups
attached to the two imino functions of 2,6-{[2,6-(i-
(11) Blessing, R. Acta Crystallogr. 1995, A51, 33.

Preparation of an Active Nd Catalyst
Organometallics, Vol. 23, No. 21, 2004 5057

5058 Organometallics, Vol. 23, No. 21, 2004
Sugiyama et al.
The NMR spectra feature the four isopropyl groups
as two sets of methyl groups and only one set of
methyne. This indicates the presence of a substantial
magnetic anisotropy between the methyl groups point-
ing on the side of the dCH2 groups and those pointing
in the opposite direction. There was only one resonance
present in the 7Li NMR spectrum (δ ) 1.63 ppm),
indicating a rapid exchange in solution between the two
nonequivalent lithium atoms of the solid state structure
(one coordinated to the ligand and the other solvated
by four molecules of THF). The two ene-amido hydrogen
atoms gave two slightly broad singlets at 3.41 and 4.05
ppm coupled to the same 13C resonance at 71.74 ppm
1
of the H{13C} HMQC spectrum.
The dianion 1 is closely reminiscent of the diketene-
iminato benzene dianion reported by Burger.12 It reacts
readily at room temperature with NdCl3(THF)3 in THF
to afford {2,6-[(2,6-(i-Pr)2Ph)N-Cd(CH2)]2(C5H3N)}Nd-
(THF)(µ-Cl)2[Li(THF)2]‚0.5(hexane) (2a ), which was iso-
lated as a brown crystalline solid (Scheme 2). While the
NMR spectra were uninformative due to the paramag-
netism, simple recrystallization from DME afforded the
DME-solvated analogue {[2,6-{[2,6-(i-Pr)2C6H3]N-Cd
(CH2)}2(C5H3N)]NdCl2(DME)}{Li(DME)3] (2b), which
produced crystals suitable for X-ray diffraction analysis.
The structure of 2b is formed by one anionic unit
containing the Nd atom and a cationic moiety consisting
of a Li atom surrounded by three molecules of DME.
The heptacoordination around the Nd metal in the
anionic moiety (Figure 2) is defined by two chlorine
atoms [Nd-Cl(1) ) 2.720(2) Å, Nd-Cl(2) ) 2.7372(19)
Å] located trans to each other [Cl(1)-Nd-Cl(2) )
158.69(7)°], the three N atoms of the ligand [N(1)-Nd
) 2.435(5) Å, N(2)-Nd ) 2.556(5) Å, N(3)-Nd ) 2.439-
(5) Å], and the two oxygen atoms of one molecule of
DME [Nd-O(2) ) 2.715(6) Å, Nd-O(1) ) 2.691(6) Å].
The overall geometry around the metal is distorted
pentagonal bipyramidal with the equatorial plane de-
fined by the three N atoms of the ligand and the two
oxygen atoms of DME [N(1)-Nd(1)-N(2) ) 64.48(18)°,
N(3)-Nd(1)-N(2) ) 64.14(17)°, N(1)-Nd(1)-O(1) )
86.5(2)°, N(3)-Nd(1)-O(2) ) 86.05(19)°, O(1)-Nd(1)-
O(2) ) 60.2(2)°]. The two chlorine atoms are located
above and below the molecular plane [N(1)-Nd(1)-Cl-
(2) ) 95.14(14)°, N(3)-Nd(1)-Cl(2) ) 96.46(14)°, O(1)-
Nd(1)-Cl(2) ) 78.12(15)°, N(1)-Nd(1)-Cl(1) ) 92.92-
(14)°, N(3)-Nd(1)-Cl(1) ) 94.27(14)°, O(1)-Nd(1)-Cl(1)
) 82.72(15)°, O(2)-Nd(1)-Cl(1) ) 83.68(15)°]. The two
ene-amido groups display values for the C-C distances
[C(1)-C(2) ) 1.362(10) Å, C(8)-C(9) ) 1.351(11) Å] as
expected for conjugated double bonds and compare well
with those of the other complexes reported in this work.
Accordingly, the C-N bond distances [N(1)-C(2) )
1.380(9) Å, N(3)-C(8) ) 1.395(9) Å] are also shorter
than C-N single bonds, thus indicating the presence
of a delocalized π-system along the CH2-C-N frame-
work. The Nd-N distances are slightly longer than
those of other lanthanide complexes of chelating amide
ligands,13 possibly indicating that the bonding character
is somewhat intermediate between Nd-N(amido) and
Nd-N(imine).
F igu r e 1. ORTEP plot of the anionic unit of 1 showing
the thermal ellipsoids drawn at the 30% probability level.
Sch em e 1
Pr)2C6H3]NdC(CH3)}2(C5H3N) was serendipitously dis-
covered while attempting the alkylation of the diimine
complex with MeLi.3b The reaction is rather complex
since it proceeds via initial alkylation of the pyridine N
atom followed by elimination of methane.3a,b In addition,
the reaction is not suitable for the large-scale prepara-
tion of the corresponding monoanion given the modest
yield and the formation of more than one product in the
reaction mixture. However, a similar reaction carried
out with 1 equiv of LiCH2Si(CH3)3 in THF cleanly
afforded the corresponding dilithium salt [{2,6-[(2,6-(i-
Pr)2Ph]N-Cd(CH2)}2(C5H3N)]Li(THF)][Li(THF)4] (1)
where both methyl groups attached to the imine func-
tions have been deprotonated (Scheme 1). The use of 2
equiv of the alkyllithium reagent gave the expected yield
improvement (up to 72%). The formula and connectivity
of 1 were provided by single-crystal diffraction data
(Figure 1). Unfortunately, attempts to obtain a satisfac-
tory data file for the crystal structure were always
plagued by low-quality data, which resulted in high
convergence values of the anisotropic refinement indi-
ces. Nonetheless, the data were sufficient to provide the
connectivity. The structure shows an ionic arrangement
comprising one tetrahedrally solvated lithium cation
and one symmetry-generated anionic unit. The anionic
unit is formed by one double-deprotonated ligand [C(1)-
C(2) ) 1.358(7) Å] surrounding a second Li cation [Li-
(1)-N(1) ) 2.104(4) Å, Li(1)-N(2) ) 2.020(12) Å]. The
C-N bond distances [C(2)-N(1) ) 1.366(7) Å] are
substantially longer for a C-N double bond but shorter
for a regular single bond. One molecule of THF com-
pletes the saddle-shaped coordination geometry of the
Li cation [N(1)-Li(1)-N(1A) ) 155.9(6)°, N(2)-Li(1)-
O(1) ) 180.00(1)°, N(1)-Li(1)-O(1) ) 102.1(3)°, N(1)-
Li(1)-N(2) ) 77.9(3)°].
Attempts to replace the two chlorine atoms on 2a
were unsuccessful. Its treatment with a variety of
(12) Nuckel, S.; Burger, P. Organometallics 2000, 19, 3305.

Preparation of an Active Nd Catalyst
Organometallics, Vol. 23, No. 21, 2004 5059
Sch em e 2
[Nd-Br ) 2.8358(12) Å] and the second terminal carbon
atom of the allyl group. Similar to the case of some other
Nd-allyl derivatives,14 the allyl group [C(34)-C(35)-
C(36) ) 144(2)°] displays an arrangement with respect
to the Nd, which deviates from the classical η3-bonding
mode in the sense that the metal is nearly coplanar with
the three carbon atoms rather than being placed on the
axis perpendicular to the allyl plane. However, the
central carbon atom shows rather large thermal param-
eters possibly indicative of some disorder, which could
not be modeled. Thus, it is possible that the observed
coplanar arrangement may in reality be an artifact of
the disorder of the allyl groups over two skewed posi-
tions. However, even in the most favorable case, the
deviation of the Nd atom from perpendicularity to the
allyl plane remains substantial. The three allyl carbon
atoms form bonding distances with the metal [Nd-C(34)
) 2.782(10) Å, Nd-C(35) ) 2.718(11) Å, Nd-C(36) )
2.671(10) Å] comparable to those of other Nd-allyl
derivatives14 and which, when combined with the C-C
distances [C(34)-C(35) ) 1.32(2) Å, C(35)-C(36) ) 1.24-
(2) Å], indicate a substantial contribution of a σ,π-η1:
η2-bonding mode for the allyl group.15 Similar to 2b,
the two ene-amido groups form C-C bond distances
[C(1)-C(2) ) 1.365(11) Å, C(8)-C(9) ) 1.332(12) Å] in
good agreement with the presence of a substantial C-C
double-bond character. Overall, the ligand displays bond
distances and angles comparable to those of complex 2b,
although the Nd-N distances are a little shorter. Sim-
ilarly, an octahedrally solvated Li cation surrounded by
three molecules of DME completes the structure of 3.
With the exception of the formation of 3, complexes
2a ,b are generally not suitable starting materials for
further alkylation reactions. Nevertheless, alkyl deriva-
tives can still be prepared by modifying the synthetic
pathway (Scheme 3). Namely, partial alkylation of
NdCl3(THF)3 with RLi [R ) Me3SiCH2, CH3] followed
by reaction with either the diimime or 1 (depending on
the Nd/R ratio) enabled the synthesis of the neutral,
halogen- and alkali cation-free compound [2,6-{[2,6-(i-
P r )2C6H 3]N-Cd(CH 2)}2(C5H 3N)]Nd[CH 2Si(CH 3)3]-
(THF) (4). The coordination geometry around the Nd
atom can be described as a severely distorted trigonal
bipyramid. The equatorial plane is defined by two N
donor atoms from the ligand [N(1)-Nd-N(3) ) 126.6-
F igu r e 2. ORTEP plot of the anionic unit of 2b showing
the thermal ellipsoids drawn at the 30% probability level.
F igu r e 3. ORTEP plot of the anionic unit of 3 showing
the thermal ellipsoids drawn at the 30% probability level.
alkylating agents, including several Grignard and or-
ganolithiums, unfortunately gave only intractable ma-
terials. Similarly disappointing behavior was observed
for 2b, with the only exception being the reaction with
(allyl)MgBr. The resulting crystalline species was for-
mulated as {[2,6-{[2,6-(i-Pr)2C6H3]N-Cd(CH2)}2(C5H3N)]-
Nd(η3-C3H5)Br}{Li(DME)3} (3). The crystal structure
shows that the ionic arrangement was preserved during
the allylation reaction, although a Li rather than the
Mg counterion was retained in the structure. In addi-
tion, one of the two chlorine atoms of 2b was replaced
by the allyl group, while the second was surprisingly
replaced by one bromine atom. The crystal structure
(Figure 3) shows a severely distorted pseudo-octahedral
arrangement around the Nd atom in the anionic moiety
[N(1)-Nd-N(3) ) 124.1(2)°, N(2)-Nd-C(34) ) 169.4-
(3)°, Br-Nd-C(36) ) 132.6(3)°, Br-Nd-N(1) ) 101.75-
(15)°, Br-Nd-N(2) ) 90.86(13)°, Br-Nd-N(3) ) 98.32-
(16)°]. The ligand displays the usual arrangement with
the three N donor atom attachments [Nd-N(1) ) 2.332-
(6) Å, Nd-N(2) ) 2.521(5) Å, Nd-N(3) ) 2.375(6) Å]
nearly coplanar with Nd. The fourth coordination site
is occupied by one of the terminal carbon atoms of the
allyl group. Orthogonally placed are the bromine atom
(13) See for example: (a) Graf, D. D.; Davis, W. M.; Schrock, R. R.;
Organometallics 1998, 17, 5820. (b) Fryzuk, M. D.; Yu, P.; Patrick, B.
O.; Can. J . Chem. 2001, 79, 1194. (c) Lee, L.; Berg, D. J .; Bushnell, G.
W.; Inorg. Chem. 1994, 33, 5302.
(14) See for example: (a) Taube, R.; Maiwald, S.; Sieler, J . J .
Organomet. Chem. 2001, 621, 327. (b) Wu, W.; Chen, M.; Zhou, P.
Organometallics 1991, 10, 98,
(15) There are a few example of this type of bonding for allylic
groups; see for example: (a) Perez, J .; Riera, V.; Rodriguez, A.; Garcia-
Granda, S. Angew. Chem., Int. Ed. 2002, 41, 1427. (b) Barbaro, P.;
Currao, A.; Hermann, J .; Nesper, R.; Pregosin, P. S.; Salzmann, R.
Organometallics 1996, 15, 1879. (c) Abbenhuis, H. C. L.; Burckhardt,
U.; Gramlich, V.; Martelletti, A.; Spencer, J .; Steiner, I.; Togni, A.
Organometallics 1996, 15, 1614.

5060 Organometallics, Vol. 23, No. 21, 2004
Sugiyama et al.
Sch em e 3
displays very good activity as a stereospecific butadiene
polymerization catalyst. Addition of a solution of 2a in
cylohexane to a cyclohexane solution of butadiene
containing excess MMAO at 50 °C started a polymeri-
zation affording cis-polybutadiene with 95-97% cis
configuration in good yield (about 70%). Such high
stereoselectivity of the polymerization reaction, in ad-
dition to the high catalyst activity, the high molecular
weight, the relatively narrow polydispersity, and other
desirable rheological properties, places this catalyst
above other transition metal-based catalysts17 but below
the remarkably performing Cp*2Sm.18 The activity also
is substantially higher with respect to that of NdCl3-
(THF)2,19 thus indicating that the ligand system is
indeed beneficial to the catalytic behavior of the metal.
While it is reasonable to expect that an in situ-generated
Nd-C function is the polymerization initiator, attempts
to form such a group via reaction of 2a or 2b with Me3-
Al (including in large excess) gave only intractable
material. Furthermore, the use of more aggressive
alkylating agents such as MeLi resulted only in com-
plicated mixture. It is possible that, similar to the cases
of V, Cr, and even Al, the ligand pyridine ring engages
in complex ring alkylation and cycloaddition reactions.4
By using the same synthetic protocol described for 4 (i.e.,
treatment of NdCl3(THF)3 with 4 equiv of MeLi followed
by addition of the intact diimine ligand 2,6-{[2,6-(i-
Pr)2C6H3]NdC(CH3)}2(C5H3N) in lieu of treating 2a ,b
with MeLi) it was possible to isolate a new species,
formulated as {2,6-[(2,6-(i-Pr)2Ph)N-Cd(CH2)]2(C5H3N)}-
Nd (THF)(µ-Me)2[Li (THF)2] (5). Complex 5 is also a
potent polymerization catalyst while activated by alkyl
aluminum derivatives. Although the activity was some-
what lower than 2a , the molecular weight of the
polymer produced by this organometallic complex was
higher. The stereoselectivity decreased only slightly,
while the overall yield of the polymer was disappoint-
ingly low. To date, attempts to isolate suitable crystals
of complex 5 have failed. Nevertheless, support for
F igu r e 4. ORTEP plot of 4 showing the thermal ellipsoids
drawn at the 30% probability level.
(2)°] and the C atom of the Me3SiCH2 group [N(1)-Nd-
C(34) ) 102.1(2)°, N(3)-Nd-C(34) ) 104.7(2)°] (Figure
4), whereas the O atom of THF [Nd-O ) 2.463(5) Å]
and the N atom from the pyridine moiety of the ligand
system [Nd-N(2) ) 2.475(5) Å] are located in the two
axial positions [O-Nd-N(2) ) 156.09(19)°]. The plane
of the ligand system is defined by the three N donor
atoms [Nd-N(1) ) 2.316(5) Å, Nd-N(3) ) 2.321(5) Å]
and is folded along the Nd-N(2) vector. The bond
distances and angles of the ligand system compare well
with those of the other compounds reported in this work
[C(1)-C(2) ) 1.347(9) Å, C(8)-C(9) ) 1.337(8) Å, C(2)-
C(3) ) 1.492(9) Å, C(7)-C(8) ) 1.486(9) Å]. The Nd-C
distance [Nd-C(34) ) 2.432(7) Å] also is in the expected
range.
Neodymium alkyl derivatives are well-known to pro-
mote a range of catalytic processes. Allyl derivatives,
in particular, are widely used as stereoselective buta-
diene polymerization catalysts.16 However, complex 3
displayed only marginal activity as a butadiene polym-
erization initiator and similarly disappointing behavior
was also observed for 4. By contrast, complex 2a
(17) (a) Gehrke, K.; Harwart, M. Plast. Kautsch. 1993, 40, 356. (b)
Anzai, S.; Irako, K.; Onishi, A.; Furukawa, J . Kogyo Kagaku Zasshi
1969, 72, 2081. (c) Kaita, S.; Kobayashi, E.; Sakikibara, S.; Aoshima,
S.; Furukawa, J . J . Polym Sci. Polym. Chem. Ed. 1996, 34, 3431. (d)
Oehme, A.; Gebauer, U.; Gehrke, K. Macromol. Rapid Commun. 1995,
16, 563. (e) Zambelli, A.; Proto, A.; Longo, P.; Oliva, P. Macromol.
Chem. Phys. 1994, 195, 2623.
(18) (a) Kaita, S.; Hou, Z.; Wakatsuki, Y. Macromolecules 1999, 32,
9078. (b) Kaita, S.; Hou, Z.; Wakatsuki, Y. Macromolecules 2001, 34,
1539.
(16) (a) Taube, R.; Windisch, H.; Maiwald, S.; Hemling, H.; Schu-
mann, H. J . Organomet. Chem. 1996, 513, 49. (b) Taube, R.; Windisch,
H.; Hemling, H.; Schumann, H. J . Organomet. Chem. 1998, 555, 201.
(c) Taube, R.; Windisch, H.; Go¨rlitz, F. H.; Schumann, H. J . Organomet.
Chem. 1993, 445, 85. (d) Maiwald, S.; Taube, R.; Hemling, H.;
Schumann, H. J . Organomet. Chem. 1998, 552, 195.
(19) Yang, J . H.; Tsutsui, M.; Chen, Z.; Bergbreiter, D. E. Macro-
molecules 1982, 15, 230.

Preparation of an Active Nd Catalyst
Organometallics, Vol. 23, No. 21, 2004 5061
Ta ble 3. P olym er iza tion Resu ltsa
run
complex
activity
yield (%)
cis-1,4
trans-1,4
1,2-PB
Mw
Mn
Mz
Mw/Mn
1
2
3
4
5
6
6
6
6
2a
5
0.74
77.3
67.5
84.8
78.6
17.2
75.0
97.0
95.0
96.1
97.0
95.9
61.8
2.5
4.5
4.5
2.3
3.3
0.5
0.5
3.4
0.7
0.8
1.8
659 000
380 000
385 000
725 000
910 000
96 000
263 000
164 000
164 000
262 000
155 000
44 000
1 220 000
715 000
727 000
1 440 000
2 200 000
235 000
2.5
0.12(5)
0.034(5)
0.71
0.40
0.08
2.31
2.34
2.8
4.9
2.18
4
36.4
a
The activity is calculated from polymer samples taken from the reactor after 15 min run time for runs 1-3 and 5, after 10 min for
run 4, and after 30 min for run 6. Activity is given as kg of polymer per mmol of catalyst/h. Runs 2 and 3 were performed in toluene
rather than cyclohexane. In run 3 the order of addition of MMAO and Nd catalysts was inverted.
of THF. The overall geometry around Nd is distorted
octahedral [N(1)-Nd-N(3) ) 127.54(9)°, N(2)-Nd-Cl-
(1) ) 156.39(7)°, O(1)-Nd-C(46) ) 159.0(5)°]. The
ligand system shows the usual arrangement where the
pyridine ring is nearly coplanar with the two ene-amide
functions, while the planes of the two bulky phenyl rings
are oriented perpendicularly. The Nd-N and Nd-C
bond distances do not show any particularly significant
feature.
Complex 6 is an excellent polymerization catalyst for
the preparation of cis-polybutadiene (Table 3) and has
characteristics very similar to those of 2a . This clearly
indicates that partial alkylation is occurring during the
treatment of 2a with MMAO. Furthermore, the choice
of solvent plays an important role in the polymerization
reaction. Polymerizations carried out in toluene rather
than in cyclohexane afforded substantially lower activity
F igu r e 5. ORTEP plot of 6 showing the thermal ellipsoids
and produced polymers with significantly lower molec-
drawn at the 30% probability level.
ular weights. Such a marked effect of the solvent on the
polymerization reaction, as well as the lack of catalytic
the formulation of 5, as suggested by analytical data,
activity of 5 and 6 in the absence of cocatalyst, can be
was indirectly provided by the isolation and structural
taken as an indication (among other hypotheses) that
crystallographic determination of {[2,6-{[2,6-(i-Pr)2-
one of the primary functions of the Lewis acidic cocata-
C6H3]N-Cd(CH2)}2(C5H3N)]Nd}(µ-Cl)(µ-X)[Li(THF)2] [X
lyst is to dissociate the residual THF from the coordi-
) Cl 53%, Me 47%] (6). This new species was prepared
natively saturated Nd atom. In other words, labile
by following the same protocol adopted for 5 except for
coordination of toluene might be a competitive process
the use of a different stoichiometric ratio of Nd/MeLi.
for the coordination of butadiene at the metal’s empty
The rationale for attempting its preparation was simply
coordination site, which is the preliminary step to
to obtain a species in which one chlorine atom would
trigger the polymerization. Finally, the high stereospec-
be retained without substantial structural modification.
ificity of the polymerization process and the linearity
It was hoped that the presence of the electron-with-
of the polymer strongly indicate that the coordination
drawing halogen could lead to improved catalytic activ-
of butadiene occurs in a η4 mode with a substantial
ity, hopefully without the need of initiators, while still
participation of the Nd atom in a back-bonding interac-
preserving the characteristics of 2 in terms of polymer
tion.
quality.
In conclusion, we have described the preparation of
the dianion modification of the versatile diiminopyridine
ligand and its use in the preparation of novel chloro and
alkyl neodymium complexes, which were identified and
The structure of 6 shows the species being a cocrys-
tallite of 2a and {[2,6-{[2,6-(i-Pr)2C6H3]N-Cd(CH2)}2-
(C5H3N)]Nd}(µ-Cl)(µ-CH3)[Li(THF)2]. The anisotropic
refinement of the thermal parameters clearly shows
characterized by standard techniques. The dilithium
that one of the two bridging atoms between Nd and the
salt 1 and the neodymium complex 2b and organome-
tallic derivatives 3, 4, and 6 were characterized by
single-crystal X-ray crystallography. Complexes 2a , 5,
and 6 were found to be active precatalysts for stereo-
Li(THF)2 unit is a chlorine with 100% occupancy.
Conversely, the second bridging atom could only be
refined by arbitrarily attributing the 47% Me and 53%
Cl occupancy. Figure 5 shows the structure of 6. The
selective cis-butadiene polymerization.
coordination sphere around the metal is defined by the
chlorine atom [Nd-Cl(1) ) 2.8024(10) Å] and the partly
Ack n ow led gm en t. This work was supported by the
occupied Me group [Nd-C(46) ) 2.74(3) Å] located cis
Natural Science and Engineering Council of Canada
(NSERC).
to each other [Cl(1)-Nd-C(46) ) 82.0(5)°], the three N
atoms of the ligand [N(1)-Nd ) 2.385(3) Å, N(2)-Nd
) 2.515(3) Å, N(3)-Nd ) 2.379(3) Å], and the O atom
Su p p or tin g In for m a tion Ava ila ble: Complete crystal-
of one molecule of THF [Nd-O(1) ) 2.584(2) Å]. The
lographic data (CIF) for all the complexes. This material is
chlorine and the methyl carbon atoms are also bridged
available free of charge via the Internet at http://pubs.acs.org.
to the same Li cation [Li-Cl(1) ) 2.371(7) Å, Li-C(46)
) 2.27(3) Å], which in turn is solvated by two molecules
OM0496434