Rare-earth metal bis(tetramethylaluminate) complexes supported by a sterically crowded triazenido ligand

Article abstract of DOI:10.1039/b925837j

Complexes [NNN]Ln(AlMe₄)₂ (Ln = Y, La, Nd, Lu) bearing the sterically demanding aryl-substituted triazenido ligand [(Tph) ₂N₃] (Tph = [2-(2,4,6-iPr₃C₆H ₂)C₆H₄]) can be obtained from homoleptic complexes Ln(AlMe₄)₃ in moderate yields, both via protonolysis with [(Tph)₂N₃]H and a salt metathesis reaction pathway utilizing [(Tph)₂N₃]K. In the solid state the Y and Lu derivatives are isostructural, with both tetramethylaluminate groups coordinated in an η² fashion, while one of the [AlMe ₄] ligands of the Nd derivative features a distorted η² coordination mode. Due to the high affinity of the triazenido ligand toward the more Lewis-acidic and harder aluminium cation compared to the softer rare-earth metal centres, ligand redistribution is observed in solution and formation of byproduct [(Tph)₂N₃]AlMe₂ is prominent. While the monoanionic triazenido ligand coordinates the rare-earth metal centres in an asymmetrical syn/anti fashion, it adopts an almost symmetric syn/syn configuration in the aluminium complex. Attempts were also made to produce putative dimethyl complexes {[(Tph)₂N₃]LnMe ₂} (Ln = Y, Lu) via cleavage of the aluminate moieties with diethyl ether. Furthermore, the intrinsic redistribution reactions are proposed to affect the performance of complexes [(Tph)₂N₃]Ln(AlMe ₄)₂ in isoprene polymerization. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/b925837j

View Online / Journal Homepage / Table of Contents for this issue
This article is published as part of the Dalton Transactions themed issue entitled:
New Horizons in Organo-f-element Chemistry
Guest Editor: Geoff Cloke
University of Sussex, UK
Published in issue 29, 2010 of Dalton Transactions
Image reproduced with the permission of Tobin Marks
Articles in the issue include:
PERSPECTIVES:
Organo-f-element catalysts for efficient and highly selective hydroalkoxylation and hydrothiolation
Charles J. Weiss and Tobin J. Marks, Dalton Trans., 2010, DOI: 10.1039/c003089a
Non-classical divalent lanthanide complexes
François Nief, Dalton Trans., 2010, DOI: 10.1039/c001280g
COMMUNICATIONS:
A bimetallic uranium μ-dicarbide complex: synthesis, X-ray crystal structure, and bonding
Alexander R. Fox, Sidney E. Creutz and Christopher C. Cummins
Dalton Trans., 2010, DOI: 10.1039/c0dt00419g
PAPERS:
Coordination polymerization of renewable butyrolactone-based vinyl monomers by lanthanide and
early metal catalysts
Garret M. Miyake, Stacie E. Newton, Wesley R. Mariott and Eugene Y.-X. Chen,
Dalton Trans., 2010, DOI: 10.1039/c001909g
Visit the Dalton Transactions website for more cutting-edge inorganic and
organometallic chemistry research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
Rare-earth metal bis(tetramethylaluminate) complexes supported by a
sterically crowded triazenido ligand†
Rannveig Litlabø,a Hyui Sul Lee,b Mark Niemeyer,b Karl W. To¨rnroosa and Reiner Anwander*a,c
Received 8th December 2009, Accepted 23rd January 2010
First published as an Advance Article on the web 12th March 2010
DOI: 10.1039/b925837j
Complexes [NNN]Ln(AlMe4)2 (Ln = Y, La, Nd, Lu) bearing the sterically demanding aryl-substituted
triazenido ligand [(Tph)2N3] (Tph = [2-(2,4,6-iPr3C6H2)C6H4]) can be obtained from homoleptic
complexes Ln(AlMe4)3 in moderate yields, both via protonolysis with [(Tph)2N3]H and a salt metathesis
reaction pathway utilizing [(Tph)2N3]K. In the solid state the Y and Lu derivatives are isostructural,
2
with both tetramethylaluminate groups coordinated in an h fashion, while one of the [AlMe4] ligands
2
of the Nd derivative features a distorted h coordination mode. Due to the high affinity of the
triazenido ligand toward the more Lewis-acidic and harder aluminium cation compared to the softer
rare-earth metal centres, ligand redistribution is observed in solution and formation of byproduct
[(Tph)2N3]AlMe2 is prominent. While the monoanionic triazenido ligand coordinates the rare-earth
metal centres in an asymmetrical syn/anti fashion, it adopts an almost symmetric syn/syn
configuration in the aluminium complex. Attempts were also made to produce putative dimethyl
complexes {[(Tph)2N3]LnMe2} (Ln = Y, L u ) via cleavage of the aluminate moieties with diethyl ether.
Furthermore, the intrinsic redistribution reactions are proposed to affect the performance of complexes
[(Tph)2N3]Ln(AlMe4)2 in isoprene polymerization.
Me3C6H2)2C6H3]; Tph = [2-(2,4,6-iPr3C6H2)C6H4]; Ln = Eu,
Yb).11 In addition asymmetric mesityl azido and adamantyl azido
ligands afforded complexes [L(MesN3)]Ln[(MesN3)(CH2SiMe3)]2
(Ln = Lu, Sc; L = (2,6-Me2C6H3)NCH2C6H4P(C6H5)2)12 and
Introduction
N-coordinating (chelating) ligands provide unique scaffolds for the
synthesis and isolation of highly reactive monomeric organometal-
lic reagents.1 Importantly, d-transition and rare-earth metal com-
plexes supported by such ancillary ligands emerged as highly effi-
cient catalysts for the enantioselective fabrication of polymers and
fine chemicals.2 More specifically, rare-earth metal bis(alkyl) com-
plexes containing amidinato,3 b-diketiminato,4 aminopyridinato,5
aza-crown,6 and other N-coordinating ancillary ligands7 have
proved themselves active in the polymerization of 1,3-dienes,
ethylene, and a-olefins upon activation by organoboron and/or
organoaluminium cocatalysts.
2
(C5Me5)2La[h -(N,N¢)-(C5Me5)NN¢N¢¢Ad](N3Ad),13 respectively.
We have recently shown that homoleptic complexes Ln(AlMe4)3
(Ln = Y, La, Nd, Lu) can be used as precursors for the synthesis
of non-cyclopentadienyl complexes with donor-functionalized
diamido [NON]2- and [NNN]2-, imino-amido [NNN]-, as well as
tris(pyrazolyl)borato (Tp) [NNN]- ancillary ligands.5b,14 Herein we
add to this tetramethylaluminate-based postmetallocene library15
the triazenido ligand [(Tph)2N3]-,16 and show how rare-earth
metal bis(tetramethylaluminate)s can be obtained by two com-
plementary reaction protocols, namely alkane elimination and
salt metathesis.14b,17 We also report on our attempts to syn-
thesize dimethyl complexes of the type [(Tph)2N3]LnMe2 via
donor-induced cleavage of the tetramethylaluminate moieties in
[(Tph)2N3]Ln(AlMe4)2. And finally, we describe how the latter
rare-earth metal bis(tetramethylaluminate) complexes perform in
isoprene polymerization.
Although monoanionic triazenido ligands are structurally
related to amidinato ligands, their utilization in organometal-
lic catalysis is scarce.8 Since triazenido ligands are weaker
donors than the isoelectronic amidinato and the related
b-diketiminato ligands, enhanced electrophilicity of the metal
centres can be anticipated, and hence different reactivity.9 In
rare-earth metal chemistry, X-ray structurally authenticated ex-
amples are limited to the solvated phenyl-substituted triazenide
complexes LnIII[(C6H5)2N3]3(NC5H5)2 and Cp2LnIII[(C6H5)2N3](4-
tBuNC5H4) (Ln = Er, Lu),10 and the unsolvated pentafluorophenyl
Results and discussion
derivatives [(Dmp)(Tph)N3]LnII(C6F5) (Dmp
=
[2,6-(2,4,6-
Synthesis and characterization of [(Tph)2N3]Ln(AlMe4)2 (2)
aDepartment of Chemistry, University of Bergen, Alle´gaten 41, 5007, Bergen,
Norway
Bis(tetramethylaluminate) complexes [(Tph)2N3]Ln(AlMe4)2 (2)
(Ln = Y(a), La(b), Nd(c), Lu(d)) were prepared by applying a
protonolysis protocol based on [(Tph)2N3]H and Ln(AlMe4)3 (1),
or via a salt metathesis reaction pathway utilizing the potassium
salt of the triazene proligand instead (Scheme 1). An instantaneous
colour change of the reaction mixtures from light to bright
yellow, or orange brown in the case of neodymium, accompanied
bInstitut fu¨r Anorganische und Analytische Chemie, Johannes Gutenberg-
Universita¨t Mainz, Duesbergweg 10-14, 55128, Mainz, Germany
cInstitut fu¨r Anorganische Chemie, Universita¨t Tu¨bingen, Auf der Mor-
genstelle 18, 72076, Tu¨bingen, Germany. E-mail: reiner.anwander@uni-
tuebingen.de; Fax: +49(0)7071-29-2436; Tel: +49(0)7071-29-72069
† CCDC reference numbers 757139–757142. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/b925837j
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6815–6825 | 6815
©

p-arene interactions like those observed in the potassium complex
[(Tph)2N3]K16 or the divalent europium and ytterbium complexes
[(Dmp)(Tph)N3]LnII(C6F5) (Dmp = [2,6-(2,4,6-Me3C6H2)2C6H3],
Tph = [2-(2,4,6-iPr3C6H2)C6H4], Ln = Eu, Yb).11 The metal atoms
are shifted slightly out of the N3 plane of the ligand (∠N6–N4–
N5–Ln2 8.6◦ (2a); 7.5◦ (2d)). Both [AlMe4] ligands coordinate in
2
a h fashion with one of the ligands closer to planarity than the
other (∠C55–Ln2–C56–Al4 -4.6◦ (2a); -4.3◦ (2d), ∠C51–Ln2–
C52–Al3 9.9◦ (2a); 9.7◦ (2d)). All Ln–C bond lengths are in the
expected range (2a 2.495 A (av); 2d 2.451 A (av)).21,22 Interestingly,
˚
˚
the Y–C bond lengths are slightly shorter than those observed in
the similar amidinate complex [(NCNdipp)Y(AlMe4)2]3a (2.533 A
˚
(av)) therefore reflecting the weaker donor character9,18b,c of the
triazenido compared with the amidinato ligand. Even though the
triazenido ligand coordinates in an asymmetrical syn/anti fashion,
the two Ln–N bond lengths are very similar (Y–N 2.345(3)-
˚
˚
2.365(3) A, Lu–N 2.296(6)-2.332(6) A).
Scheme 1 Synthesis of [(Tph)2N3]Ln(AlMe4)2 (2) via protonolysis and
salt metathesis reaction pathways.
by methane formation and [KAlMe4] precipitation, respectively,
indicated coordination of the monoanionic triazenido ligand
to the rare-earth metal centre. Upon removal of the solvent
and the volatile byproducts (or [KAlMe4]), bright yellow solids
of complexes 2 were obtained, however, contaminated by the
dimethylaluminium byproduct [(Tph)2N3]AlMe2 (3).18a
Both the bis(tetramethylaluminate)s 2, except 2b, and the
byproduct 3 are highly soluble in aromatic and aliphatic solvents,
1
which hampers their separation. Based on the H NMR spectra
of the crude reaction products of the diamagnetic complexes, the
approximate ratio of yttrium complex 2a and undesired byproduct
3 could be estimated as 1 : 1, while in the case of lanthanum the
ratio was slightly higher, and for lutetium a bit lower. This can be
rationalized on the basis of steric effects: due to the high affinity
of the triazenido ligand toward the more Lewis-acidic and harder
aluminium cation the rare-earth metal centre of the sterically
more crowded Lu(AlMe4)3 is less accessible. This tendency of
N-chelating ligands to coordinate to Al(III) centres has been
observed previously.14d,19,20
In an attempt to increase the yield of complexes 2 relative to
byproduct 3 the reactions were performed with two equivalents
of proligand. This led to the complete (La) or almost complete
(Y and Lu) disappearance of the homoleptic precursor 1, left-
overs of which were observed in the raw products from all of the
1 : 1 reactions. Unfortunately, the two-equivalent reactions also
considerably increased the yield of byproduct 3.
Fig. 1 Molecular structure of molecule 2 of 2a. Atoms are represented by
atomic displacement ellipsoids set at the 50% level. Hydrogen atoms and
solvent are omitted for clarity.
2
Also in neodymium complex 2c the h -bonded triazenido
ligand coordinates in an asymmetrical syn/anti fashion with no
additional metal-p-arene interactions (Fig. 2, Table 2). However,
the two [AlMe4] ligands feature distinct coordination motifs
2
in the solid state; one of them shows the routine planar h
Through fractional crystallization we were able to collect single
crystals of 2a, 2c and 2d suitable for X-ray structure analysis. The
yttrium (2a) and lutetium (2d) derivatives are isostructural and
crystallize in the monoclinic space group P21/n, with one molecule
hexane per two crystallographically independent molecules (the D
and L enantiomers) of the product, while the neodymium deriva-
tive (2c) crystallizes without solvent molecules in the monoclinic
space group P21/c. An ORTEP drawing of the yttrium complex
2a is shown in Fig. 1, and selected bond distances and angles
of complexes 2a and 2d are listed in Table 1. The rare-earth
metal centre is six-coordinate and adopts a distorted octahedral
coordination (∠C5–Nd1–C6–Al2 2.5◦), while the other one is
◦
2
bent (∠C1–Nd1–C2–Al1 -37.5 ). Such a bent h coordination
of the aluminate group will accomplish a better steric saturation
of the larger neodymium metal centre. Non-planar tetraalkyl-
aluminate coordination has been observed earlier in neodymium
complexes (C5Me5)Nd(AlMe4)2,23 [C5H3(SiMe3)2]Nd(AlMe4)2,15
[C5H2(CMe3)3]Nd(AlMe4)2,15 (h -PC4Me4)Nd(AlMe4)2,17 and [h -
PC4Me2(SiMe3)2]Nd(AlMe4)2.17 For complex 2c, the Nd ◊ ◊ ◊ C3
contact originating from this aluminate bending is significantly
5
5
˚
shorter than observed earlier (2.973(6) A, compared to 3.088–
3
˚
3.326 A), almost approaching a h coordination as found in
2
coordination geometry. The h -bonded triazenido ligand displays
the pentaneodymium cluster {Cp*5Nd5[(m-Me)3AlMe](m4-Cl)(m3-
Cl)2(m2-Cl)6} (Nd ◊ ◊ ◊ C 2.88(2), 2.88(2), 2.78(2) A).24 As in the case
˚
an asymmetrical syn/anti conformation with no additional metal-
6816 | Dalton Trans., 2010, 39, 6815–6825
This journal is
The Royal Society of Chemistry 2010
©

Table 1 Selected Interatomic Distances, Angles, and Dihedral Angles for [(Tph)2N3]Ln(AlMe4)2, Ln = Y (2a), Lu (2d)
2a (Ln = Y)
2d (Ln = Lu)
Molecule 1
Molecule 2
Molecule 1
Molecule 2
˚
Bond Distances/A
Ln1(2)–N2(5)
Ln1(2)–N3(6)
2.345(3)
2.365(3)
3.044(1)
3.061(1)
2.476(4)
2.506(4)
2.478(4)
2.515(5)
4.66
2.093(5)
2.075(5)
1.940(6)
1.961(6)
2.069(4)
2.076(5)
1.965(5)
1.962(5)
1.315(4)
1.324(4)
2.358(3)
2.358(3)
3.066(1)
3.052(1)
2.478(4)
2.518(4)
2.472(4)
2.512(4)
4.67
2.083(4)
2.077(5)
1.961(4)
1.965(4)
2.090(5)
2.075(5)
1.956(5)
1.959(5)
1.307(4)
1.330(4)
2.296(6)
2.325(6)
2.997(3)
3.021(2)
2.418(8)
2.471(8)
2.429(8)
2.452(9)
4.70
2.093(9)
2.073(8)
1.928(11)
1.956(11)
2.066(8)
2.069(8)
1.970(8)
1.960(9)
1.322(8)
1.326(8)
2.304(6)
2.332(6)
3.015(2)
3.000(2)
2.440(8)
2.471(8)
2.427(8)
2.467(8)
4.70
2.079(8)
2.077(8)
1.960(8)
1.958(8)
2.089(8)
2.071(8)
1.955(10)
1.957(9)
1.306(8)
1.317(8)
Ln1(2) ◊ ◊ ◊ Al1(3)
Ln1(2) ◊ ◊ ◊ Al2(4)
Ln1(2)–C1(51)
Ln1(2)–C2(52)
Ln1(2)–C5(55)
Ln1(2)–C6(56)
Ln ◊ ◊ ◊ C(Phsyn) (av)
Al1(3)–C1(51)
Al1(3)–C2(52)
Al1(3)–C3(53)
Al1(3)–C4(54)
Al2(4)–C5(55)
Al2(4)–C6(56)
Al2(4)–C7(57)
Al2(4)–C8(58)
N1(4)–N2(5)
N1(4)–N3(6)
Bond Angles (◦)
N2(5)–Ln1(2)–N3(6)
Ln1(2)–C1(51)–Al1(3)
Ln1(2)–C2(52)–Al1(3)
Ln1(2)–C5(55)–Al2(4)
Ln1(2)–C6(56)–Al2(4)
C1(51)–Ln1(2)–C2(52)
C1(51)–Al1(3)–C2(52)
C5(55)–Ln1(2)–C6(56)
C5(55)–Al2(4)–C6(56)
N2(5)–N1(4)–N3(6)
Dihedral Angles (◦)
N3(6)–N1(4)–N2(5)–Ln1(2)
N1(4)–N2(5)–C9(59)–C29(79)
N1(4)–N3(6)–C30(80)–C50(100)
54.59(11)
83.09(15)
82.72(15)
84.12(15)
83.05(15)
85.47(15)
108.43(18)
84.04(14)
107.46(18)
109.9(3)
54.58(11)
84.00(15)
83.10(15)
83.53(15)
82.80(15)
83.92(14)
106.86(17)
85.12(15)
108.08(18)
110.2(3)
55.7(2)
82.9(3)
82.0(3)
84.0(3)
83.4(3)
87.0(3)
107.8(3)
85.3(3)
106.2(3)
109.1(5)
55.6(2)
83.3(3)
82.6(3)
82.9(3)
82.3(3)
85.5(3)
106.6(3)
86.7(3)
107.7(3)
111.1(6)
-3.6(3)
-147.2(3)
-43.8(5)
8.6(3)
153.5(3)
38.9(5)
-3.2(5)
-146.7(6)
-43.7(9)
7.5(5)
152.9(6)
38.8(9)
of the yttrium and lutetium derivatives 2a and 2d, the Nd–N bond
lengths are similar (2.454(4) and 2.470(4) A).
Because of the high affinity of the triazenido ligand toward
aluminium, and formation of the byproduct 3, clean NMR spectra
of products 2 were difficult to obtain. Even NMR spectra of
crystallized compounds 2 (confirmed by X-ray structural and
elemental analysis) revealed a product/byproduct mixture, which
implies that formation of byproduct 3 is favourable over time.
Overlapping peaks in the 1H NMR spectra were therefore a
problem for the assignment of some of the peaks and integral
˚
1
values. The H NMR spectra of complexes 2a and 2b in C6D6,
at ambient temperature, are indicative of more symmetric ligand
environments than found in the solid state, which implies that
the aromatic substituents of the coordinated ligand are highly
flexible in solution. The methyl protons of the isopropyl groups
give three doublets, two for the o-positioned isopropyl groups
(2a 1.05, 1.15 ppm; 2b 1.06, 1.15 ppm) and one for those in p-
position (2a 1.30 ppm; 2b 1.32 ppm), each counting twelve protons.
For comparison, the proligand shows a much less symmetric
environment in C6D6 giving six different doublets, one for each
isopropyl group. Correspondingly, the methine protons give two
septets with a ratio of 2 : 1, which can be assigned to those in o-
and p-position, respectively (2a 2.74, 2.88 ppm; 2b 2.78, 2.91 ppm),
while the proligand shows three different methine signals. Also the
[AlMe4] moieties of complex 2a and 2b show high fluxionality in
solution. Only one signal is observed for the aluminium bound
methyl groups at 25 ◦C (2a -0.24 ppm; 2b -0.15 ppm), and
Fig. 2 Molecular structure of 2c. Atoms are represented by atomic
displacement ellipsoids set at the 50% level. Hydrogen atoms and
iPr-substituents are omitted for clarity.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6815–6825 | 6817
©

Table 2 Selected Interatomic Distances, Angles, and Dihedral Angles for
[(Tph)2N3]Nd(AlMe4)2 (2c)
Table 3 Selected Interatomic Distances, Angles, and Dihedral Angles for
[(Tph)2N3]AlMe2 (3)
2c
3
˚
˚
Bond Distances (A)
Bond Distances (A)
Nd1–N2
Nd1–N3
Nd1 ◊ ◊ ◊ Al1
Nd1 ◊ ◊ ◊ Al2
Nd1–C1
Nd1–C2
Nd1 ◊ ◊ ◊ C3
Nd1–C5
Nd1–C6
Nd1 ◊ ◊ ◊ C(Phsyn) (av)
Al1–C1
Al1–C2
Al1–C3
Al1–C4
Al2–C5
2.454(4)
2.470(4)
2.9180(17)
3.1663(16)
2.713(6)
2.708(6)
2.973(6)
2.640(5)
2.562(5)
5.035
2.060(7)
2.048(7)
1.987(7)
1.949(6)
2.055(6)
2.085(6)
1.964(7)
1.976(6)
1.325(6)
1.304(6)
Al1–C1
Al1–C2
Al1–N1
Al1–N2
Al1 ◊ ◊ ◊ C(Ph) (av)
N1–N3
N2–N3
1.933 (2)
1.950 (2)
1.983(1)
1.975(1)
4.193
1.311(2)
1.323(2)
Bond Angles (◦)
C1–Al1–C2
N1–Al1–N2
N1–N3–N2
Dihedral Angles (◦)
N1–N3–N2–Al1
N3–N1–C3–C4
N3–N2–C24–C25
122.68(7)
64.49(5)
106.6(1)
7.2(1)
19.8(2)
-16.9(2)
Al2–C6
Al2–C7
Al2–C8
N1–N2
The presence of byproduct 3 also hampered the interpretation
of the 13C NMR spectra. For yttrium complex 2a the assignment
of all product peaks was possible, even though byproduct 3 gives
the strongest peaks. The [AlMe4] signal at 3.5 ppm revealed a
13C-89Y scalar coupling (1JYC = 4.6 Hz). As opposed to all other
complexes, the lanthanum derivative 2b is less soluble in aliphatic
hydrocarbons than byproduct 3, facilitating its isolation and
purification. Attempts to dissolve these purified samples in C6D6
to measure 13C NMR spectra resulted in the same type of ligand
scrambling and, despite the considerable higher concentration of
complex 2b, only a few peaks could be assigned definitely.
N1–N3
Bond Angles (◦)
N2–Ln1–N3
Nd1–C1–Al1
Nd1–C2–Al1
Nd1–C3–Al1
Nd1–C5–Al2
Nd1–C6–Al2
C1–Ln1–C2
C1–Al1–C2
C1–Ln1–C3
C1–Al1–C3
C2–Ln1–C3
C2–Al1–C3
C5–Ln1–C6
C5–Al2–C6
N2–N1–N3
Dihedral Angles (◦)
N3–N1–N2–Nd1
N1–N2–C9–C29
N1–N3–C30–C50
52.20(13)
74.0(2)
74.25(18)
68.81(18)
83.83(18)
85.24(19)
72.3(2)
102.2(3)
68.2(3)
104.3(4)
69.78(18)
107.6(3)
81.14(18)
109.6(2)
110.9(4)
Synthesis and characterization of [(Tph)2N3]AlMe2 (3)
Byproduct 3 was independently synthesized in quantitative yield
via addition of AlMe3 to a hexane solution of proligand
[(Tph)2N3]H. Instant colour change of the reaction mixture from
light to bright yellow and methane formation indicated the
formation of compound [(Tph)2N3]AlMe2 (3). Crystallization of
complex 3 from a hexane solution produced single crystals suitable
for X-ray structure analysis. 3 crystallizes in the monoclinic space
13.2(4)
144.0(5)
50.2(7)
in the case of the yttrium derivative a characteristic splitting
of the peak is observed due to coupling to the yttrium metal
centre (2JYH = 2.3 Hz). Variable temperature H NMR studies
1
of complex 2a in toluene-d8 did not show any decoalescence
for the signals of these methyl groups at temperatures down to
-80 ◦C (Dn1/2 ª 8 Hz at -80 ◦C), similar to homoleptic precursor
1a.21 This implies rapid exchange between bridging and terminal
methyl groups even at low temperatures. Furthermore, there is
no splitting of the peaks corresponding to the isopropyl groups
of the ancillary ligand at -80 ◦C. The ligand environment in
complex 2d is slightly less symmetric in C6D6, most likely due
to the high steric crowding around the small lutetium metal
centre. However, as mentioned earlier, no additional metal-p-
arene interactions are observed in the solid-state structure. For
complex 2d five different methyl signals (1.31, 1.27, 1.15, 1.12,
1.05 ppm) are observed for the isopropyl groups, assignable to two
p-positioned, two o-positioned, and two equivalent o-positioned
isopropyl groups, respectively. As for the proligand, three methine
signals are observed (2.71, 2.84, 2.87 ppm). The [AlMe4] moieties
of complex 2d show enhanced fluxionality even at temperatures
down to -80 ◦C (Dn1/2 ª 8 Hz at -80 ◦C).
Fig. 3 Molecular structure of 3. Atoms are represented by atomic
displacement ellipsoids set at the 50% level. Hydrogen atoms are omitted
for clarity.
6818 | Dalton Trans., 2010, 39, 6815–6825
This journal is
The Royal Society of Chemistry 2010
©

Scheme 2 Synthesis of [(Tph)2N3]LnMe2 (4) via donor-induced cleavage of [(Tph)2N3]Ln(AlMe4)2.
group P21/n and adopts a distorted tetrahedral geometry (Fig. 3,
of the formation of soluble dimethyl complexes 4. Attempts to
crystallize such dimethyl derivatives have so far only resulted in
the isolation of byproduct 3.
Table 3). In contrast to the rare-earth metal complexes 2, the
2
h -coordinated triazenido ligand features an almost symmetric
syn/syn configuration. Additional p-interaction between the
metal centre and the arene rings of the ligand are not observed
Polymerization of isoprene
˚
(Al–Cph 4.193 A (av)). The aluminium atom is shifted slightly
As an extension of our tetramethylaluminate-based postmet-
allocene library15 rare-earth metal complexes 2 were initially
examined as precatalysts for the polymerization of isoprene. Given
the intrinsic ligand scrambling and redistribution reactions in
solution, this turned out to be a challenging task. In addition
to the rare-earth metal complexes 2 under investigation, we had to
consider additional active/co-influencing components originating
from residual homoleptic precursors 1 and byproduct 3. In order to
investigate the effect of these side-products, pure samples of these
compounds were also tested as precatalysts. Significant amounts
of isolated complexes 2 were only obtained for the derivatives of
the larger metal centres lanthanum and neodymium, by exploiting
distinct solubility behaviour and fractional crystallization, respec-
tively. Accordingly, isolated samples of 2b and 2c were used, in
addition to in situ formed catalyst mixtures. For yttrium derivative
2a, representing the smaller rare-earth metal centres, only the
in situ formed complex was tested as a precatalyst. To reduce
the amount of homoleptic rare-earth metal complexes 1 present in
the in situ formed catalyst mixtures of complexes 2, 1 : 2 ratios of
precursors 1 and the proligand were employed. [Ph3C][B(C6F5)4]
(A), [PhNMe2H][B(C6F5)4] (B), and B(C6F5)3 (C) were used as
activators. The polymerization results are summarized in Table 4.
The aluminium compound 3 did not polymerize isoprene,
neither without nor with addition of borane or borate activators
(not shown in Table 4).32 Homoleptic complexes 1 did, how-
ever, show good to excellent activity in the polymerization of
isoprene upon activation with B(C6F5)3 (C) or [Ph3C][B(C6F5)4]
(A)/[PhNMe2H][B(C6F5)4] (B), respectively (Table 4, runs 1–9).
The stereoselectivities produced by these binary catalyst mixtures
are pretty low, with similar cis-1,4 and trans-1,4 contents for
complexes 1b and 1c (Table 4, runs 4-8), and a slightly higher
cis-1,4 content for complex 1a (Table 4, runs 1–3). Notewor-
thy exception is catalyst system Nd(AlMe4)3 (1c)/B(C6F5)3 (C)
revealing excellent activity with quantitative yield and 87% cis-
1,4 selectivity (Table 4, run 9). Runs 2 and 8 involving cata-
lysts Y(AlMe4)3 (1a)/[PhNMe2H][B(C6F5)4] (B) and Nd(AlMe4)3
(1c)/[PhNMe2H][B(C6F5)4] (B) are in accordance with previously
reported performances of these binary systems, nicely demonstrat-
ing the reproducibility of such polymerization data.33,34
out of the N3 plane of the ligand (∠N1–N3–N2–Al1 7.2◦), and
˚
˚
the Al–C bond lengths (Al1–C1 1.933(2) A, Al1–C2 1.950(2) A)
are similar to those observed in structurally related dimethyl-
aluminium amidinate ([MeC(NSiMe3)2]AlMe2 (1.942 A (av)),
25
˚
26
26
˚
[MeC(NCy)2]AlMe2
˚
(1.958 A (av)), [tBuC(NCy)2]AlMe2
(1.954 A (av)) and b-diketiminate ([(2,6-iPr2C6H3)2(nacnac)]-
27,28
27
˚
A
AlMe2
(1.967
(av)), [(HC(CMe)2)(N-p-tolyl)2]AlMe2
29
˚
˚
(1.958 A (av)), [(HC(CMe)2)(NC6F5)2]AlMe2 (1.952 A (av))
complexes.
As expected, the 1H NMR spectrum of the aluminium complex
3 in C6D6 shows a quite symmetric ligand environment. As for
complexes 2a and 2b, three doublets are observed for the methyl
protons of the isopropyl groups (o-position: 1.06, 1.18 ppm; p-
position: 1.29 ppm) and two septets in a ratio of 2 : 1, accounting
for o- (2.71 ppm) and p-positioned (2.85 ppm) methine protons,
respectively.
Attempted synthesis of {[(Tph)2N3]LnMe2}
Attempts were made to synthesize dimethyl complexes of the
formula [(Tph)2N3]LnMe2 (4) via donor-induced cleavage of
complexes 2a and 2d (Scheme 2), as described earlier for the
corresponding pentamethylcyclopentadienyl complexes.30 Addi-
tion of an excess of Et2O to a hexane solution of the respective
complexes 2 (contaminated with byproduct 3) did not result in any
visual changes. The mixture remained a bright yellow solution.
1
From the H NMR of the product mixture one could, however,
observe the disappearance of the [AlMe4] peak. Unfortunately,
the concentration of putative product 4 was too low for any
peaks to be assigned, due to ligand redistribution and extensive
formation of 3. Upon addition of two equivalents of AlMe3 to
this mixture the [AlMe4] peak reappeared, which might indicate a
reversible formation of dimethyl complexes 4 from 2. Furthermore,
the redistribution of complexes 2 in solution, to give mixtures
of 2 and 3, is always accompanied by formation of some of
the homoleptic starting material Ln(AlMe4)3 (1). Donor-induced
cleavage of complexes 1 with Et2O should produce insoluble
[LnMe3]n derivatives,31 precipitation of which, however, did not
occur in the present cleavage reactions. This would be in favour
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6815–6825 | 6819
©

Table 4 Polymerization of isoprene by precatalysts [(Tph)2N3]Ln(AlMe4)2 (2)35
Structured
Entrya Precatalyst
Cocatalystc
Yield (%)
trans-1,4-
cis-1,4-
3,4-
Mn e(¥ 105)
Mw/Mn
Efficiency f(%)
e
1
2
3
4
5
6
7
8
Y(AlMe4)3 (1a)
A
B
C
A
B
C
A
B
C
—
A
B
C
—
A
B
C
A
B
C
—
A
B
>99
>99
56.2
>99
>99
80.0
>99
>99
>99
—
23.6
21.2
35.1
51.4
46.8
59.5
44.0
44.9
10.1
—
69.1
68.8
62.5
46.3
49.8
39.1
51.0
50.8
86.6
—
7.3
10.1
2.4
2.3
3.4
1.4
5.1
4.3
3.4
—
0.9
1.1
1.2
0.5
0.6
3.3
0.5
0.4
1.6
—
2.09
1.93
2.03
1.29
1.23
1.18
1.89
1.77
1.65
—
75
62
32
138
120
16
147
158
42
—
88
123
—
—
112
125
23
108
95
33
—
Y(AlMe4)3 (1a)
Y(AlMe4)3 (1a)
La(AlMe4)3 (1b)
La(AlMe4)3 (1b)
La(AlMe4)3 (1b)
Nd(AlMe4)3 (1c)
Nd(AlMe4)3 (1c)
9
Nd(AlMe4)3 (1c)
10
11
12
13
14
15
16
17
18b
19b
20b
21
22
23
24
25b
26b
27b
[(Tph)2N3]Y(AlMe4)2 (2a)
[(Tph)2N3]Y(AlMe4)2 (2a)
[(Tph)2N3]Y(AlMe4)2 (2a)
[(Tph)2N3]Y(AlMe4)2 (2a)
[(Tph)2N3]La(AlMe4)2 (2b)
[(Tph)2N3]La(AlMe4)2 (2b)
[(Tph)2N3]La(AlMe4)2 (2b)
[(Tph)2N3]La(AlMe4)2 (2b)
[(Tph)2N3]La(AlMe4)2 (2b)
[(Tph)2N3]La(AlMe4)2 (2b)
[(Tph)2N3]La(AlMe4)2 (2b)
[(Tph)2N3]Nd(AlMe4)2 (2c)
[(Tph)2N3]Nd(AlMe4)2 (2c)
[(Tph)2N3]Nd(AlMe4)2 (2c)
[(Tph)2N3]Nd(AlMe4)2 (2c)
[(Tph)2N3]Nd(AlMe4)2 (2c)
[(Tph)2N3]Nd(AlMe4)2 (2c)
[(Tph)2N3]Nd(AlMe4)2 (2c)
>99
>99
—
41.0
31.4
—
49.6
57.5
—
9.4
11.1
—
0.8
0.6
—
1.41
1.52
—
—
—
—
—
—
—
>99
>99
10.3
>99
>99
40.2
—
>99
>99
84.9
>99
>99
2.2
74.2
82.6
89.4
81.0
91.7
89.2
—
81.2
76.0
61.3
90.0
89.4
28.7
23.1
14.9
8.6
16.1
6.3
8.8
—
16.4
21.7
35.6
8.0
8.6
67.2
2.7
2.5
2.0
3.0
2.1
2.0
—
2.4
2.3
3.2
2.0
2.0
4.1
0.6
0.5
0.3
0.6
0.7
0.9
—
0.5
0.5
1.3
0.6
0.5
1.4
1.35
1.13
2.02
1.28
1.30
2.04
—
1.40
1.42
2.93g
1.51
1.36
3.63g
135
143
26
110
133
<1
C
A
B
C
a Conditions: 0.02 mmol precatalyst formed in situ, [Ln]/[cocat] = 1 : 1, 8 mL toluene, 20 mmol isoprene, 24 h, ambient temperature. b Isolated sample used
as precatalyst. c Cocatalyst: A = [Ph3C][B(C6F5)4], B = [PhNMe2H][B(C6F5)4], C = B(C6F5)3; the catalyst was preformed for 30 min at ambient temperature.
d Determined by 1H and 13C NMR spectroscopy in CDCl3. e Determined by means of size-exclusion chromatography (SEC) against polystyrene standards.
f Initiation efficiency = Mn(calculated)/Mn(measured). g Bimodal distributions with significantly different Mn values were obtained for catalyst systems
2c/C (run 24, major part (72%): Mn = 1.0 ¥ 105, Mw/Mn = 1.07; run 24, minor part (28%): Mn = 7.4 ¥105, Mw/Mn = 1.44; run 27, major part (51%):
Mn = 4.9 ¥ 105, Mw/Mn = 1.66; run 27, minor part (49%): Mn = 0.1 ¥ 105, Mw/Mn = 1.81).
All precatalysts 2 showed extremely high activities upon acti-
vation with [Ph3C][B(C6F5)4] (A) or [PhNMe2H][B(C6F5)4] (B),
independent on the size of the metal cation. The activation
capability of B(C6F5)3 (C) seems to be generally low and depends
considerably on the metal cation size, since the yttrium derivative
2a did not give any polyisoprene upon addition of B(C6F5)3
(Table 4, run 13). For all activations of complexes 2, a colour
change from bright yellow to orange was observed independent
on the boron activator.
Due to the interference of homoleptic complexes 1 in solutions
of precatalysts 2, it is difficult to discuss the effect of the rare-earth
metal on the stereoselectivity of the polymerization reactions in
detail. One clearly sees a different performance of the smaller
yttrium compared to the larger neodymium and lanthanum
derivatives; while the latter precatalysts 2b and 2c display relatively
high trans-1,4 selectivities (Table 4, runs 15–20 and 22–27), yttrium
complex 2a produces slightly higher cis-1,4 contents, and also a
slightly higher 3,4 content (Table 4, runs 11 and 12). This could
be a metal size effect, but at the same time the yttrium system
implies a higher concentration of homoleptic complex 1a, which
will cause a shift toward higher cis-1,4 contents of the polymers
(Table 4, runs 1–3).
All catalyst mixtures produce polyisoprene with high trans-1,4
content, comparable Mn and relatively narrow molecular weight
distributions (Table 4, runs 15-20), with the highest trans-1,4
selectivity of 91.7% obtained with activator [PhNMe2H][B(C6F5)4]
(B) (Table 4, run 19). The performance of the neodymium
derivative 2c seems to be considerably affected by homoleptic
complex 1c present in the system. This is particularly evident for
the B(C6F5)3 (C)-based initiators (Table 4, runs 24 and 27), which
show a relatively high cis-1,4 content and pronounced bimodal
molecular weight distributions. Except for these latter initiators, all
catalyst systems tested revealed relatively low molecular weights,
which could be explained by polymer chain-transfer promoted by
the various organoaluminium species present in solution.36 The
molecular weight distributions produced by the initiators 2/A and
2/B are smaller than 1.5.
Table 5 summarizes important polymerization data of bis(alkyl)
rare-earth metal complexes supported by monoanionic N-
coordinating ancillary ligands including amidinato, pyrrolido,
aminopyridinato and other types of substituted amido ligands
(Chart 1).3a,5a,7a,c-f,37 It is very clear that the polymerization perfor-
mance and in particular the stereoselectivity is crucially affected by
the N-coordinating ancillary ligands. In 2007 Hou et al. reported
on bis(phosphinophenyl)amido (PNP) supported rare-earth metal
bis(alkyls) I that promote the living cis-1,4-polymerization of iso-
prene upon cationization with borate [PhNMe2H][B(C6F5)4] (B).37
Having a closer look at the polymerization reactions performed
with precatalyst 2b, there seems to be negligible interference of
homoleptic complex 1b, independent on which activator is used.
6820 | Dalton Trans., 2010, 39, 6815–6825
This journal is
The Royal Society of Chemistry 2010
©

Table 5 Previously reported results for isoprene polymerization by bis(alkyl) rare-earth metal complexes containing N-coordinating ancillary ligands
Structure
Precatalyst
Cocatalysta Time
T/◦C Yield (%) 3,4- cis-1,4- trans-1,4- Mn (¥ 105) Mw/Mn Ref.
[(2-(Ph2P)C6H4)2N]Y(CH2SiMe3)2(thf) I
B
A
3 h
0
100
100
100
97
0.4
99.5 0.5
1
95
99.6
—
—
<1
—
1.3
1.7
4.0
1.3
1.06
1.4
1.7
37
3a
3a
5a
[(2,6-iPr2C6H3N)2PhC]Y(o-CH2C6H4NMe2)2 II
20 min -20
16 h
20 h
[(2,6-iPr2C6H3N)2PhC]Y(o-CH2C6H4NMe2)2 II A/AlMe3
-20
20
>98
5
{(2,6-iPr2C6H3)[6-(2,4,6-iPr3C6H2)C5H3-2-
A
1.68
N]N}Sc(CH2Ph)2(thf) III
{(2,6-iPr2C6H3)[6-(2,4,6-iPr3C6H2)C5H3-2-
N]N}Sc(CH2Ph)2(thf) III
B/AlMe3
A/AlEt3
20 h
20
20
96
10
90
76.7
—
0.7
1.0
4.51
1.55
5a
7e
=
{2-[(N-2,6-Me2C6H3)N CH]-
1 min
100
9.0
14.3
C4H3N}Sc(CH2SiMe3)2(thf) IV
{[2-(Me2NCH2)-C4H3N]Y(CH2SiMe3)2}2 V
[(N-2,4,6-Me3C6H2)NPPh2N(Ph)]-
Sc(CH2SiMe3)2(thf) VI
A/AlEt3
B/AliBu3
5 h
2 h
20
-40
99
97
14.4 15.2
94.7
70.4
5.3b
0.1
9.9
1.37
1.55
7e
7c
=
{7-[(N-2,6-iPr2C6H3)N CH]-
A/AliBu3
A/AliBu3
A
5 min
6 h
20
96
9.1
87.7
3.2
—
4.9
1.5
0.3
1.56
1.5
7d
7a
7f
C8H5N}Sc(CH2SiMe3)2(thf) VII
[(N-2-EtC6H4)NPPh2N(N-2,6-
-20
25
81
99.4 0.6
70
iPr2C6H3)]Sc(CH2SiMe3)2(thf) VIII
[2,6-iPr2C6H3N(SiMe3)]Lu(CH2SiMe3)2(thf) IX
5 min
100
30b
1.12
a A = [Ph3C][B(C6F5)4], B = [PhNMe2H][B(C6F5)4]. b cis-1,4 : trans-1,4 ratio not specified.
One year later, the same group demonstrated that the initiator
mono(amidinate) bis(aminobenzyl) yttrium (II)/[Ph3C][B(C6F5)4]
(A) exhibits high 3,4-stereoselectivity, a selectivity that is totally
switched to cis-1,4 upon addition of AlMe3 involving the forma-
tion of the corresponding bis(tetramethylaluminate) complex.3a
More recently, Kempe et al. described the same type of stereoselec-
tivity switch when adding different alkylaluminium compounds to
the bis(alkyl) scandium aminopyridinate complex III.5a Cui et al.
also reported on several bis(alkyl) rare-earth metal complexes (IV-
VIII) containing different types of monoanionic N-coordinating
ancillary ligands, and their effect in isoprene polymerization.7a,c–e
It was concluded that 3,4-selectivity is favoured in the case of
sterically oversaturated metal centres, as evidenced for the smallest
metal centre scandium in VIII showing the highest 3,4-selectivities,
even in the presence of alkylaluminium cocatalysts. However, for
initiators with sterically less saturated metal centres the addition of
alkylaluminium cocatalysts results in the same cis-1,4-selectivity
switch as reported by Hou and Kempe, suggesting the formation of
tetraalkylaluminate rare-earth metal complexes as active species in
these polymerizations. The dinuclear pyrrolido supported yttrium
bis(alkyl), [(2-(Me2NCH2)-C4H3N)Y(CH2SiMe3)2]2 (V)7e is a rare
example, which gave a high trans-1,4 selectivity. This was argued
to be a result of the special spatial environment around the metal
centre. The beneficial effect of N-chelating ancillary ligands for
achieving high stereoselectivities is evident from the mediocre
performance of complex IX.7f Importantly, for the triazenide
complexes 2 under study, it is the combination of a highly flexible
[NNN] ligand with large Ln(III) centres which favours trans-1,4 se-
lectivity. This is in agreement with the polymerization performance
of half-sandwich bis(tetramethylaluminate) complexes.15
Conclusions
Homoleptic complexes Ln(AlMe4)3 can be utilized in both
protonolysis and salt metathesis reactions for the synthe-
sis of triazenide bis(tetramethylaluminate) postmetallocene-type
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6815–6825 | 6821
©

complexes. The triazenido ligand [(Tph)2N3]- (Tph = [2-(2,4,6-
iPr3C6H2)C6H4]) proves to be a versatile ancillary ligand giving
access to complexes [(Tph)2N3]Ln(AlMe4)2 of both the smaller
(Y and Lu), and larger rare-earth metals (Nd and La). How-
ever, competition of the more Lewis-acidic harder aluminium
cation for the triazenido ligand gives product mixtures. Com-
plexes [(Tph)2N3]Ln(AlMe4)2 are active in the polymerization
of isoprene, their performance being governed by the following
factors: a) ligand redistribution in solution to give Ln(AlMe4)3
and [NNN]AlMe2 produces additional active sites, interference
being most effective for yttrium or cocatalyst C; b) large
Ln(III) metal centres reveal high trans-1,4 selectivities (91.7% for
[(Tph)2N3]La(AlMe4)2/[PhNMe2H][B(C6F5)4]); c) relatively low
molecular weights suggest polymer chain transfer via organoalu-
minium species. The binary system Nd(AlMe4)3/B(C6F5)3 revealed
also excellent activity with 87% cis-1,4 selectivity.
experiments on the AV500 and AV600 spectrometers at ambient
temperature, using CDCl3 as a solvent.
General procedure for the synthesis of [(Tph)2N3]Ln(AlMe4)2 (2)
via protonolysis
In a glovebox, a stirred suspension of [(Tph)2N3]H in 3 mL of
hexane was added dropwise to a solution of Ln(AlMe4)3 (1)
dissolved in 3 mL of hexane. Instant gas formation was observed,
and the mixture turned from a light yellow suspension into a bright
yellow, clear solution. The reaction mixture was stirred another
0.5–1 h at ambient temperature, and then dried under vacuum to
yield yellow powdery solids of 2 and the byproduct 3.
General procedure for the synthesis of [(Tph)2N3]Ln(AlMe4)2 (2)
via salt metathesis
A more general message from this polymerization study is
that rare-earth metal alkyl complexes carrying N-coordinating
ligands are especially prone to ligand redistribution reactions in
the presence of (small) organoaluminium reagents such as AlMe3
and AlEt3. This will involve the formation of homoleptic Ln(III)
tetraalkylaluminate species, which might drastically affect the
polymerization performance.
In a glovebox, a stirred suspension of [(Tph)2N3]K in 3 mL of
toluene was added dropwise to a solution of Ln(AlMe4)3 (1)
dissolved in 3 mL of toluene. Instantaneously, the mixture turned
from a yellow suspension into a clearer and more orange solution.
Then it turned yellow again and gave a new suspension. The reac-
tion mixture was stirred for another 2 h at ambient temperature.
After centrifugation, the residue was washed several times with
toluene and then the solution fractions combined and dried under
vacuum to yield yellow powdery solids of 2 and the byproduct 3.
The salt metathesis reactions gave approximately the same
yields as the protonolysis reactions, but a slight contamination
of remaining [KAlMe4] made the latter more favourable.
Experimental
General considerations
All operations were performed with rigorous exclusion of air
and water, using standard Schlenk, high-vacuum, and glovebox
techniques (MBraun MBLab; <1 ppm O2, <1 ppm H2O). Hexane,
toluene and diethyl ether were purified by using Grubbs columns
(MBraun SPS, solvent purification system) and stored in a glove-
box. C6D6 and toluene-d8 were obtained from Aldrich, degassed,
dried over Na for 24 h, and filtered. AlMe3 was purchased
from Aldrich and used as received. Homoleptic Ln(AlMe4)3 (1)
(Ln = Y, La, Nd, Lu) were prepared according to literature
methods.21,22 [(Tph)2N3]H (Tph = [2-(2,4,6-iPr3C6H2)C6H4]) was
synthesized as described earlier,16 and [(Tph)2N3]K by reacting
the protonated ligand with K[N(SiMe3)2]. The NMR spectra of
air and moisture sensitive compounds were recorded by using
J. Young valve NMR tubes at 25 ◦C on a Bruker DMX-400
Avance (1H: 400.13 Hz; 13C: 100.61 MHz), a Bruker-BIOSPIN-
[(Tph)2N3]Y(AlMe4)2 (2a)
Following the protonolysis procedure described above, Y(AlMe4)3
(1a, 57 mg, 0.16 mmol) and [(Tph)2N3]H (99 mg, 0.16 mmol)
yielded 2a, contaminated by 3, as a powdery yellow solid (140 mg,
1
45%, based on H NMR). Salt metathesis gave a reaction yield
of 44%. Crystallization from hexane at -30 ◦C afforded yellow
single crystals of 2a suitable for X-ray diffraction analysis. IR
(Nujol, cm-1): 1610 w, 1574 w, 1460 vs (Nujol), 1378 vs (Nujol),
1323 m, 1284 s, 1197 m, 1103 w, 1057 w, 1000 w, 943 w, 881 w,
757 m, 721 m, 705 m, 581 w, 524 w cm-1. H NMR (600 MHz,
1
◦
C6D6, 25 C): d = 7.21 (s, 4 H, m-Trip), 7.13-7.18 (m, 2 H, ar),
7.11 (d, 3JHH = 7.4 Hz, 2 H, ar), 6.95 (t, 3JHH = 7.4 Hz, 2 H, ar),
3
3
6.92 (d, JHH = 8.0 Hz, 2 H, ar), 2.88 (sep, JHH = 6.8 Hz, 2 H,
p-CH(CH3)2), 2.74 (sep, 3JHH = 6.8 Hz, 4 H, o-CH(CH3)2), 1.30 (d,
3JHH = 6.8 Hz, 12 H, p-CH(CH3)2), 1.15 (d, 3JHH = 6.8 Hz, 12 H,
o-CH(CH3)2), 1.05 (d, 3JHH = 6.8 Hz, 12 H, o-CH(CH3)2), -0.24 (s,
2JYH = 2.3 Hz, 24 H, Al(CH3)4) ppm. 13C NMR (151 MHz, C6D6,
25 ◦C): d = 149.0 (p-Trip), 147.1 (o-Trip), 146.9 (ar), 135.4 (i-Trip),
133.4 (ar), 131.7 (ar), 128.5 (ar), 122.8 (ar), 121.9 (m-Trip), 120.9
(ar), 34.7 (p-CH(CH3)2), 30.8 (o-CH(CH3)2), 26.3 (o-CH(CH3)2),
1
AV500 (5 mm BBO, H: 500.13 Hz; 13C: 125.77 MHz), and a
Bruker-BIOSPIN-AV600 (5 mm cryo probe, 1H: 600.13 MHz;
1
13C: 150.91 MHz). H and 13C shifts are referenced to internal
solvent resonances and reported in parts per million relative to
TMS. IR spectra were recorded on a NICOLET Impact 410
FTIR spectrometer as Nujol mulls sandwiched between CsI plates.
Elemental analyses were performed on an Elementar Vario EL III.
The molar masses (MW/Mn) of the polymers were determined by
size-exclusion chromatography (SEC). Sample solutions (1.0 mg
polymer per mL THF) were filtered through a 0.2 mm syringe
filter prior to injection. SEC was performed with a pump supplied
by Viscotek (GPCmax VE 2001), employing ViscoGEL columns.
Signals were detected by means of a triple detection array (TDA
1
24.5 (p-CH(CH3)2), 23.1 (o-CH(CH3)2), 3.5 (s br, JYC = 4.6 Hz,
Al(CH3)4) ppm. Anal. Calcd. for C50H78N3Al2Y: C, 69.50; H, 9.10;
N, 4.86. Found: C, 69.21; H, 9.38; N, 4.80.
[(Tph)2N3]La(AlMe4)2 (2b)
302) and calibrated against polystyrene standards (MW/Mn
<
Following the protonolysis procedure described above,
La(AlMe4)3 (1b, 80 mg, 0.20 mmol) and [(Tph)2N3]H (120 mg,
0.20 mmol) yielded 2b, contaminated by 3, as a powdery yellow
1.15). The flowrate was set to 1.0 mL min-1. The microstructure
of the polyisoprenes was examined by means of 1H and 13C NMR
6822 | Dalton Trans., 2010, 39, 6815–6825
This journal is
The Royal Society of Chemistry 2010
©

1
solid (193 mg, 49%, based on H NMR). Salt metathesis gave a
0.06 mmol, 36% from crystallization). Salt metathesis gave a
reaction yield of 37%. IR (Nujol): 1610 w, 1564 w, 1460 vs (Nujol),
1383 vs (Nujol), 1326 w, 1274 s, 1233 m, 1207 m, 1186 w, 1098
w, 1078 w, 1052 w, 1005 w, 938 w, 876 w, 757 m, 721 m, 710 m,
reaction yield of 51%. A small amount of pure product 2b could
1
be obtained by washing the mixture with cold hexane. H NMR
(400 MHz, C6D6, 25 ◦C): d = 7.21 (s, 4 H, m-Trip), 7.14-7.19
(m, 2 H, ar), 7.05 (d, JHH = 7.6 Hz, 2 H, ar), 6.92 (t, JHH
=
659 w, 591 w, 529 w cm-1. H NMR (600 MHz, C6D6, 25 ◦C):
3
3
1
3
7.6 Hz, 2 H, ar), 6.81 (d, JHH = 8.0 Hz, 2 H, ar), 2.91 (sep,
d = 7.12-7.16 (m, 6 H, ar), 7.10 (d, 3JHH = 7.5 Hz, 2 H, ar), 6.93
(t, 3JHH = 7.5 Hz, 2 H, ar), 6.71 (d, 3JHH = 8.0 Hz, 2 H, ar), 2.87
(sep, 3JHH = 6.8 Hz, 2 H, p-CH(CH3)2), 2.84 (sep, 3JHH = 6.8 Hz,
3JHH = 6.8 Hz, 2 H, p-CH(CH3)2), 2.78 (sep, JHH = 6.8 Hz, 4
3
3
H, o-CH(CH3)2), 1.32 (d, JHH = 6.8 Hz, 12 H, p-CH(CH3)2),
3
3
3
1.15 (d, JHH = 6.8 Hz, 12 H, o-CH(CH3)2), 1.06 (d, JHH
=
2H, o-CH(CH3)2), 2.71 (sep, JHH = 6.8 Hz, 2 H, o-CH(CH3)2),
3
3
6.8 Hz, 12 H, o-CH(CH3)2), -0.15 (s, 24 H, Al(CH3)4) ppm.
13C NMR (151 MHz, C6D6, 25 ◦C): d = 34.6 (p-CH(CH3)2),
31.1 (o-CH(CH3)2), 25.6 (o-CH(CH3)2), 24.3 (p-CH(CH3)2), 23.5
(o-CH(CH3)2), 5.0 (s br, Al(CH3)4) ppm (the intensities of signals
in the aromatic region were too low to be interpreted). Anal.
Calcd. for C50H78N3Al2La: C, 65.70; H, 8.60; N, 4.60. Found: C,
65.16; H, 8.58; N, 4.48.
1.31 (d, JHH = 6.8 Hz, CH(CH3)2), 1.27 (d, JHH = 6.8 Hz,
3
3
CH(CH3)2), 1.15 (d, JHH = 6.8 Hz, CH(CH3)2), 1.12 (d, JHH
=
6.8 Hz, CH(CH3)2), 1.05 (d, 3JHH = 6.8 Hz, CH(CH3)2), -0.02 (s,
24 H, Al(CH3)4) ppm. Anal. Calcd. for C50H78N3Al2Lu: C, 63.21;
H, 8.28; N, 4.42. Found: C, 63.81; H, 6.96; N, 4.22.
Synthesis of [(Tph)2N3]AlMe2 (3)
In a glovebox, [(Tph)2N3]H (201 mg, 0.33 mmol) was suspended in
3 mL hexane and an excess AlMe3 (49 ml, 0.51 mmol) was added via
micropipette while stirring. Instant gas formation was observed,
and the mixture turned from a light yellow suspension into a bright
yellow, clear solution. The reaction mixture was stirred another
0.5 h at ambient temperature, and then dried under vacuum to
yield a yellow powdery solid of 3 in quantitative yield (212 mg,
0.32 mmol, 98%). Crystallization from hexane at -30 ◦C afforded
yellow single crystals of 3 suitable for X-ray diffraction analysis.
IR (Nujol, cm-1): 1610 w, 1569 w, 1460 vs (Nujol), 1378 vs (Nujol),
1290 vs, 1207 m, 1197 m, 1166 w, 1159 w, 1103 w, 1067 w, 1052 w,
1005 m, 938 w, 881 m, 762 s, 721 s, 710 s, 659 m, 591 w, 529 w cm-1.
1H NMR (400 MHz, C6D6, 25 ◦C): d = 7.41 (d, 3JHH = 8.0 Hz, 2
H, ar), 7.17 (s, 4 H, m-Trip), 7.14-7.20 (m, 2 H, ar), 7.12 (d, 3JHH
[(Tph)2N3]Nd(AlMe4)2 (2c)
Following the protonolysis procedure described above,
Nd(AlMe4)3 (1c, 81 mg, 0.20 mmol) and [(Tph)2N3]H (120 mg,
0.20 mmol) yielded 2c, contaminated by 3, as a powdery greenish
solid. Crystallization from hexane at -30 ◦C afforded light green
single crystals of 2c suitable for X-ray diffraction analysis (72 mg,
0.08 mmol, 39% from crystallization). IR (Nujol): 1610 w, 1569 w,
1460 vs (nujol), 1383 vs (Nujol), 1290 m, 1207 w, 1166 w, 1098
w, 1005 w, 943 w, 881 w, 757 m, 721 m, 664 w, 586 w cm-1. Anal.
Calcd. for C50H78N3Al2Nd: C, 65.32; H, 8.55; N, 4.57. Found: C,
65.37; H, 8.18; N, 4.38.
[(Tph)2N3]Lu(AlMe4)2 (2d)
3
= 7.3 Hz, 2 H, ar), 6.98 (t, JHH = 7.3 Hz, 2 H, ar), 2.85 (sep,
Following the protonolysis procedure described above,
Lu(AlMe4)3 (1d, 69 mg, 0.16 mmol) and [(Tph)2N3]H (95 mg,
0.16 mmol) yielded 2d, contaminated by 3, as a powdery yellow
solid. Crystallization from hexane at -30 ◦C afforded yellow
single crystals of 2d suitable for X-ray diffraction analysis (55 mg,
3JHH = 6.8 Hz, 2 H, p-CH(CH3)2), 2.71 (sep, 3JHH = 6.8 Hz, 4 H,
3
o-CH(CH3)2), 1.29 (d, JHH = 6.8 Hz, 12 H, p-CH(CH3)2), 1.18
3
3
(d, JHH = 6.8 Hz, 12 H, o-CH(CH3)2), 1.06 (d, JHH = 6.8 Hz,
12 H, o-CH(CH3)2), -0.99 (s, 6 H, Al(CH3)2) ppm. 13C NMR
(151 MHz, C6D6, 25 ◦C): d = 149.2 (p-Trip), 147.0 (o-Trip), 143.9
Table 6 Crystal Data and Data Collection Parameters of Complexes 2a, 2c, 2d and 3
2a
2c
2d
3
Chemical formula
Formula Mass
Crystal system
C106H170N6Al4Y2
1814.22
Monoclinic
10.0997(2)
51.7483(12)
20.7090(5)
90.00
94.419(1)
90.00
10791.2(4)
100(2)
P21/n
4
19067
0.0614
0.1178
0.0866
0.1263
1.137
C50H78N3Al2Nd
919.35
C106H170N6Al4Lu2
1986.34
Monoclinic
10.0659(3)
51.7349(14)
20.7342(6)
90.00
94.371(1)
90.00
10766.1(5)
123(2)
P21/n
4
18995
0.0669
0.1073
0.0980
0.1150
1.156
C44H60N3Al
657.93
Monoclinic
22.4306(13)
9.5164(5)
25.0736(14)
90.00
107.802(1)
90.00
5095.9(5)
100(2)
P21/c
4
9684
0.0553
Monoclinic
16.9941(7)
13.1084(5)
17.6874(7)
90.00
97.496(1)
90.00
3906.5(3)
100(2)
P21/n
4
8687
0.0473
˚
a/A
˚
b/A
˚
c/A
alpha/◦
beta/◦
gamma/◦
3
˚
Unit cell volume/A
T/K
Space group
No. of formula units per unit cell, Z
No. of independent reflections
Final R1 valuesa
Final wR2(F2) valuesa
0.1313
0.0677
0.1376
1.065
0.0995
0.0639
0.1066
1.041
Final R1 values (all data)a
Final wR2(F2) values (all data)a
Goodness of fit on F2a
a R1 = R(ꢀFo|-|Fcꢀ)/R|Fo|; wR2 = {R[w(Fo2-Fc2)2]/R[w(Fo2)2]}1/2; GOF = {R[w(Fo2-Fc2)2]/(n-p)}
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6815–6825 | 6823
©

(ar), 134.4 (i-Trip), 132.6 (ar), 132.1 (ar), 128.6 (ar), 125.4 (ar),
121.6 (m-Trip), 120.0 (ar), 35.0 (p-CH(CH3)2), 31.0 (o-CH(CH3)2),
25.6 (o-CH(CH3)2), 24.4 (p-CH(CH3)2), 23.3 (o-CH(CH3)2), -11.1
(Al(CH3)2) ppm. Anal. Calcd. for C44H60N3Al: C, 80.32; H, 9.19;
N, 6.39. Found: C, 80.25; H, 10.20; N, 6.10.
Kretschmer, A. Meetsma, B. Hessen, T. Schmalz, S. Qayyum and R.
Kempe, Chem.–Eur. J., 2006, 12, 8969.
6 (a) S. Bambirra, D. van Leusen, C. G. J. Tazelaar, A. Meetsma and
B. Hessen, Organometallics, 2007, 26, 1014; (b) S. Bambirra, D. van
Leusen, A. Meetsma, B. Hessen and J. H. Teuben, Chem. Commun.,
2001, 637.
7 (a) S. Li, D. Cui, D. Li and Z. Hou, Organometallics, 2009, 28, 4814;
(b) D. Wang, S. H. Li, X. L. Liu, W. Gao and D. Cui, Organometallics,
2008, 27, 6531; (c) S. Li, W. Miao, T. Tang, W. Dong, X. Zhang and
D. Cui, Organometallics, 2008, 27, 718; (d) Y. Yang, Q. Wang and D.
Cui, J. Polym. Sci., Part A: Polym. Chem., 2008, 46, 5251; (e) Y. Yang,
B. Liu, K. Lv, W. Gao, D. Cui, X. Chen and X. Jing, Organometallics,
2007, 26, 4575; (f) Y. Luo, M. Nishiura and Z. Hou, J. Organomet.
Chem., 2007, 692, 536.
8 (a) A. G. M. Barrett, M. R. Crimmin, M. S. Hill, P. B. Hitchcock, G.
Kociok-Ko¨hn and P. A. Procopiou, Inorg. Chem., 2008, 47, 7366; (b) S.
Brase, S. Dahmen, F. Lauterwasser, N. E. Leadbeater and E. L. Sharp,
Bioorg. Med. Chem. Lett., 2002, 12, 1849.
General procedure for the polymerization of isoprene
A detailed polymerization procedure (Table 4, run 18) is described
as a typical example. [Ph3C][B(C6F5)4] (18 mg, 0.02 mmol, 1
equiv) was added to a solution of 2b (18 mg, 0.02 mmol) in
toluene (8 mL) and the mixture was aged at ambient temperature
for 30 min. After the addition of isoprene (2.0 mL, 20 mmol),
the polymerization was carried out at ambient temperature for
24 h. The reaction was terminated by pouring the polymerization
mixture into a large quantity of acidified 2-propanol containing
0.1% (w/w) 2,6-di-tert-butyl-4-methylphenol as a stabilizer. The
polymer was washed with 2-propanol and dried under vacuum at
ambient temperature to constant weight. The polymer yield was
determined gravimetrically.
9 S.-O. Hauber, F. Lissner, G. B. Deacon and M. Niemeyer, Angew.
Chem., Int. Ed., 2005, 44, 5871.
10 D. Pfeiffer, I. A. Guzei, L. M. Liable-Sands, M. J. Heeg, A. L. Rheingold
and C. H. Winter, J. Organomet. Chem., 1999, 588, 167.
11 S. O. Hauber and M. Niemeyer, Inorg. Chem., 2005, 44, 8644.
12 B. Liu and D. Cui, Dalton Trans., 2009, 550.
13 W. J. Evans, T. J. Mueller and J. W. Ziller, J. Am. Chem. Soc., 2009, 131,
2678.
14 (a) M. Zimmermann, J. Takats, G. Kiel, K. W. To¨rnroos and R.
Anwander, Chem. Commun., 2008, 612; (b) R. Litlabø, M. Zim-
mermann, K. Saliu, J. Takats, K. W. To¨rnroos and R. Anwander,
Angew. Chem., Int. Ed., 2008, 47, 9560; (c) M. Zimmermann, K. W.
To¨rnroos and R. Anwander, Angew. Chem., Int. Ed., 2007, 46, 3126;
(d) M. Zimmermann, F. Estler, E. Herdtweck, K. W. To¨rnroos
and R. Anwander, Organometallics, 2007, 26, 6029; (e) M. Zimmer-
mann, K. W. To¨rnroos and R. Anwander, Organometallics, 2006, 25,
3593.
15 M. Zimmermann, K. W. To¨rnroos, H. Sitzmann and R. Anwander,
Chem.–Eur. J., 2008, 14, 7266.
16 H. S. Lee and M. Niemeyer, Inorg. Chem., 2006, 45, 6126.
17 E. Le Roux, F. Nief, F. Jaroschik, K. W. To¨rnroos and R. Anwander,
Dalton Trans., 2007, 4866.
18 (a) D. Vindusˇ and M. Niemeyer, unpublished results, and ref. 16
(Supporting Information); (b) H. S. Lee, S.-O. Hauber, D. Vindusˇ and
M. Niemeyer, Inorg. Chem., 2008, 47, 4401; (c) H. S. Lee and M.
Niemeyer, Inorg. Chem., 2010, 49, 730.
19 R. Duchateau, C. T. vanWee, A. Meetsma, P. T. vanDuijnen and J. H.
Teuben, Organometallics, 1996, 15, 2279.
X-ray crystallography and crystal structure determination of 2a,
2c, 2d, and 3
Crystals suitable for diffraction experiments were selected in a
glovebox and mounted in Paratone-N (Hampton Research) inside
a nylon loop. Data collection was done on a Bruker AXS SMART
2 K CCD diffractometer using graphite-monochromated Mo-
˚
Ka radiation (l = 0.71073 A) performing w-scans in four j
positions. Raw data were collected using the SMART software
package,38 and reduced and scaled with the SAINT program.39
Numerical absorption corrections were done using SHELXTL.40
The structures were solved by direct methods and refined with
standard difference Fourier techniques.40 All plots were generated
using the ORTEP-3 program.41 For further experimental details
on refinement and crystallographic data see Table 6.
20 C. Do¨ring and R. Kempe, Eur. J. Inorg. Chem., 2009, 412.
21 W. J. Evans, R. Anwander and J. W. Ziller, Organometallics, 1995, 14,
1107.
Acknowledgements
Financial support from the Norwegian Research Council (Project
No. 182547/I30) and the Deutsche Forschungsgemeinschaft (SPP
1166) is gratefully acknowledged.
˚
22 M. Zimmermann, N. A. Frøystein, A. Fischbach, P. Sirsch, H. M.
Dietrich, K. W. To¨rnroos, E. Herdtweck and R. Anwander, Chem.–
Eur. J., 2007, 13, 8784.
23 H. M. Dietrich, E. Herdtweck, R. Anwander, unpublished results.
24 H. M. Dietrich, O. Schuster, K. W. To¨rnroos and R. Anwander, Angew.
Chem., Int. Ed., 2006, 45, 4858.
25 R. Lechler, H. D. Hausen and J. Weidlein, J. Organomet. Chem., 1989,
359, 1.
26 M. P. Coles, D. C. Swenson, R. F. Jordan and V. G. Young,
Organometallics, 1997, 16, 5183.
27 B. X. Qian, D. L. Ward and M. R. Smith, Organometallics, 1998, 17,
3070.
Notes and references
1 (a) F. T. Edelmann, Chem. Soc. Rev., 2009, 38, 2253; (b) F. T. Edelmann,
D. M. M. Freckmann and H. Schumann, Chem. Rev., 2002, 102, 1851;
(c) W. E. Piers and D. J. H. Emslie, Coord. Chem. Rev., 2002, 233–234,
131; (d) L. H. Gade, Acc. Chem. Res., 2002, 35, 575; (e) R. Kempe,
Angew. Chem., Int. Ed., 2000, 39, 468.
2 (a) P. M. Zeimentz, S. Arndt, B. R. Elvidge and J. Okuda, Chem. Rev.,
2006, 106, 2404; (b) S. J. Park, Y. Y. Han, S. K. Kim, J. S. Lee, H. K.
Kim and Y. K. Do, J. Organomet. Chem., 2004, 689, 4263; (c) V. C.
Gibson and S. K. Spitzmesser, Chem. Rev., 2003, 103, 283.
3 (a) L. Zhang, M. Nishiura, M. Yuki, Y. Luo and Z. Hou, Angew.
Chem., Int. Ed., 2008, 47, 2642; (b) S. Bambirra, M. W. Bouwkamp,
A. Meetsma and B. Hessen, J. Am. Chem. Soc., 2004, 126, 9182; (c) S.
Bambirra, D. van Leusen, A. Meetsma, B. Hessen and J. H. Teuben,
Chem. Commun., 2003, 522.
28 C. E. Radzewich, M. P. Coles and R. F. Jordan, J. Am. Chem. Soc.,
1998, 120, 9384.
29 D. Vidovic, J. N. Jones, J. A. Moore and A. H. Cowley, Z. Anorg. Allg.
Chem., 2005, 631, 2888.
30 H. M. Dietrich, C. Zapilko, E. Herdtweck and R. Anwander,
Organometallics, 2005, 24, 5767.
31 H. M. Dietrich, G. Raudaschl-Sieber and R. Anwander, Angew. Chem.,
Int. Ed., 2005, 44, 5303.
32 The reaction mixture [(Tph)2N3]AlMe2 + [Ph3C][B(C6F5)4] did in fact
yield a precipitate in the attempted isoprene polymerization, which
could be isolated as a white powder. However, by NMR and SEC
investigations we could exclude the possibility of this being any form
of polyisoprene. Attempts to characterize this compound have not yet
been successful.
4 P. G. Hayes, W. E. Piers and R. McDonald, J. Am. Chem. Soc., 2002,
124, 2132.
5 (a) C. Do¨ring, W. P. Kretschmer, T. Bauer and R. Kempe, Eur. J. Inorg.
Chem., 2009, 4255; (b) M. Zimmermann, K. W. To¨rnroos, R. M.
Waymouth and R. Anwander, Organometallics, 2008, 27, 4310; (c) W. P.
6824 | Dalton Trans., 2010, 39, 6815–6825
This journal is
The Royal Society of Chemistry 2010
©

33 S. Arndt, K. Beckerle, P. M. Zeimentz, T. P. Spaniol and J. Okuda,
Angew. Chem., Int. Ed., 2005, 44, 7473.
34 C. Meermann, K. W. To¨rnroos, W. Nerdal and R. Anwander, Angew.
Chem., Int. Ed., 2007, 46, 6508.
37 L. Zhang, T. Suzuki, Y. Luo, M. Nishiura and Z. Hou, Angew. Chem.,
Int. Ed., 2007, 46, 1909.
38 SMART v. 5.054, Data Collection Software for Bruker AXS CCD,
Bruker AXS Inc., Madison, WI, 1999.
35 A. Fischbach, M. G. Klimpel, M. Widenmeyer, E. Herdtweck,
W. Scherer and R. Anwander, Angew. Chem., Int. Ed., 2004, 43,
2234.
36 All systems show bimodal distributions with closely overlapping
peaks.
39 SAINT v. 6.45a, Data Integration Software for Bruker AXS CCD,
Bruker AXS Inc., Madison, WI, 2002.
40 SHELXTL v. 6.14, Structure Determination Software Suite, Bruker
AXS Inc., Madison, WI, 2000.
41 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6815–6825 | 6825
©