Chiral metal architectures in aminopyridinato complexes of zirconium

Article abstract of DOI:10.1039/b413023e

Optically pure 2-alkylaminopyridines (HL) are synthesised readily from bromopyridines and chiral amines [(S)-l,2,3,4-tetrahydro-1-naphthylamine and (S)-(-)-α-methylbenzylamine] using palladium-catalysed amination. Protonolysis reactions of these proligands with ZrX₄ (X = NMe ₂, CH₂Ph, CH₂Bu^t) yield zirconium aminopyridinates, usually of the type [ML₂X₂], some of which have been characterised by X-ray crystallography. Control of absolute configuration at the metal centre is pursued by investigation of the effects of chiral amine substituent, substitution at the pyridine rings and the identity of co-ligands. Surprisingly the conformationally flexible a-melhylbenzyl based aminopyridinato ligands promote much better control of chirality-at-zirconium than do the cyclic tetrahydronaphthyl analogues. One complex of the former class displays complete control of stereochemistry at 193 K; only one diastereomer out of eight possible structures is observed. It is found that there is an excellent correlation between observed selectivities and calculated diastereomer energy differences from DFT. All the complexes studied are in dynamic exchange between diastereomers. The rate of these processes (ΔH^? ca. 40 kJ mol^-1) as studied by Selective Polarisation Transfer-Selective Inversion Recovery experiments (SPT-SIR) and lineshape analyses are significantly faster than those for aminopyridines containing bulkier amido substituents (ΔH^? ca. 70 kJ mol ^-1). This type of dependence on steric effects, and the impact of the trans effect, is consistent with an N-dissociative mechanism, i.e. conversion from six- to five-coordinate structure followed by rapid intramolecular scrambling.

Full text of DOI:10.1039/b413023e

View Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
Chiral metal architectures in aminopyridinato complexes of
zirconium†
Edward J. Crust, Adam J. Clarke, Robert J. Deeth, Colin Morton and Peter Scott*
Department of Chemistry, University of Warwick, Coventry, UK CV4 7AL.
E-mail: peter.scott@warwick.ac.uk; Fax: 024 7652 2710; Tel: 024 7652 3238
Received 24th August 2004, Accepted 12th October 2004
First published as an Advance Article on the web 9th November 2004
Optically pure 2-alkylaminopyridines (HL) are synthesised readily from bromopyridines and chiral amines
[(S)-1,2,3,4-tetrahydro-1-naphthylamine and (S)-(−)-a-methylbenzylamine] using palladium-catalysed amination.
Protonolysis reactions of these proligands with ZrX4 (X = NMe2, CH2Ph, CH2But) yield zirconium
aminopyridinates, usually of the type [ML2X2], some of which have been characterised by X-ray crystallography.
Control of absolute configuration at the metal centre is pursued by investigation of the effects of chiral amine
substituent, substitution at the pyridine rings and the identity of co-ligands. Surprisingly the conformationally
flexible a-methylbenzyl based aminopyridinato ligands promote much better control of chirality-at-zirconium than
do the cyclic tetrahydronaphthyl analogues. One complex of the former class displays complete control of
stereochemistry at 193 K; only one diastereomer out of eight possible structures is observed. It is found that there is
an excellent correlation between observed selectivities and calculated diastereomer energy differences from DFT. All
the complexes studied are in dynamic exchange between diastereomers. The rate of these processes (DH‡ ca.
40 kJ mol−1) as studied by Selective Polarisation Transfer–Selective Inversion Recovery experiments (SPT-SIR) and
lineshape analyses are significantly faster than those for aminopyridines containing bulkier amido substituents (DH‡
ca. 70 kJ mol−1). This type of dependence on steric effects, and the impact of the trans effect, is consistent with an
N-dissociative mechanism, i.e. conversion from six- to five-coordinate structure followed by rapid intramolecular
scrambling.
omers (e.g. Fig. 1), we have recently shown that it is quite possible
Introduction
to design highly diastereoselective systems.6 In this contribution
When a metal coordination sphere includes one or more chiral
we describe our studies toward the control stereochemistry at
non-racemic ligands, the metal center may itself become stere-
zirconium using the system AB = aminopyridinato.7 This work
ogenic as a result of diastereoselective interactions.1,2 While the
is informed by our recent studies on methods by which structure
concept of chirality-at-metal has been of interest for some time,3
can be controlled in achiral aminopyridinate systems.8,9 While
the realisation that such chiral metal centers are present in many
we do not necessarily expect this ligand type to be of great
use in applications to e.g. enantioselective catalysis, it has the
advantage that a wide range of analogues can be made rapidly
and the effects on structure and selectivity studied by standard
enantioselective catalyses4 is promoting more intense study of
methods by which the absolute stereochemistry at the metal
may be predetermined.1 Studies in this area are complicated
by the great structural variety possible in organometallic and
techniques.
coordination compounds and also the range in kinetic stabilities
of coordination environments. Accordingly, most studies on
chiral metal complexes have involved later transition elements
where the rates of topographical exchange are generally slow
enough to be studied on convenient experimental timescales.
For the early transition metals and f elements, where the rates
of exchange are greater, very high levels of thermodynamic
discrimination are needed in order to make systems that are
of practical value.
The real challenge, for systems containing any type of metal, is
to produce complexes in which not only is the metal stereochem-
istry selected efficiently, but also by virtue of the presence of one
or more labile ligand sites is the complex suitable for catalytic
applications. Traditionally, this has been achieved through the
use of multidentate ligands, although the design and synthesis
of such coordination environments presents different synthetic
challenges.5 Bidentate ligands are more readily available and
more diverse in nature than multidentate systems. If the aim
is to produce chiral metal catalysts then, for example Group 4
metals, this points us toward octahedral complexes of the type
[M(AB)2X2], where AB is a chiral bidentate monoanionic ligand.
While this class of complex can in principle form eight diastere-
Fig.
1 Six-coordinate isomers of the compounds [Zr(amino-
pyridinate)2X2] (N = amido ligand atom). The stereochemical descrip-
tors are based on the CIP priority sequence NPy > N > X, which is true
for X = alkyl.
Results and discussion
The aminopyridines HL1–4 were synthesised from the cor-
responding bromopyridine and commercially available chiral
non-racemic amines via palladium catalysed amine arylation
† Electronic supplementary information (ESI) available: Details of
NOESY and SPT-SIR NMR spectra, rotatable structure files for X-ray
and DFT output. See http://www.rsc.org/suppdata/dt/b4/b413023e/
4 0 5 0
D a l t o n T r a n s . , 2 0 0 4 , 4 0 5 0 – 4 0 5 8
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
©

View Online
(Scheme 1).10 HL1 was recrystallised from pentane to give a
pale yellow solid. Proligands HL2–4 were isolated from the crude
reaction mixtures via distillation under reduced pressure. Upon
cooling HL2 gave a colourless crystalline solid whereas HL3,4
gave colourless oils. Our expectation was that the cyclic structure
present in L1 and L2 would provide for more efficient control of
stereochemistry at metal than would L3 and L4, although the
opposite was eventually shown to be the case.
[ZrL3 (NMe2)] as judged by NMR spectroscopy. Accordingly,
treatment of [Zr(NMe2)4] with three equivalents of the same
3
ligand led to the isolation of this compound. A similar benzyl
complex [ZrL3 (CH2Ph)] was isolated, but [ZrL3 (CH2Ph)2] was
3
2
inaccessible. In contrast, the reaction of [Zr(CH2But)4] with two
equivalents of HL3 yielded [ZrL3 (CH2But)2].
2
The molecular structure of this compound, determined
by X-ray crystallography (Fig. 2, Table 1) showed a C2-
symmetric K-trans,cis,cis-isomeric structure corresponding to
structure II (Fig. 1) with the a-methylbenzyl aryl groups
pointing towards the back of the molecule. The structure has
mutually cis neopentyl groups [C(32)–Zr(1)–C(27) = 98.2(3)◦]
with the pyridine nitrogen atoms mutually trans [N(1)–Zr(1)–
N(3) = 172.0(2)◦]. The bond lengths and angles are otherwise
unremarkable.8,11
Scheme 1 Synthesis of aminopyridines [conditions; toluene, 90 ◦C, 24
h]; (i) (S)-1,2,3,4-tetrahydro-1-naphthylamine, cat. binap/[Pd2(dba)3],
30–50%; (ii) (S)-(−)-a-methylbenzylamine, cat. dppp/[Pd2(dba)3], ca.
70%.
Zirconium complexes of L1,2
The proligand HL1 was treated with [Zr(NMe2)4], [Zr(CH2Ph)4]
and [Zr(CH2But)4] in various metal/ligand ratios. Each reaction
gave very soluble yellow solids for which 1H NMR spectra
showed many products. We have previously shown that the
presence of a 2-methyl group at the pyridine ring can have
a profound effect on the chemistry of such ligands,8 and
correspondingly treatment of [Zr(NMe2)4] with two equivalents
1
of HL2 yielded the complex [ZrL2 (NMe2)2]. The H spectrum
2
of this compound in C6D6 at 298 K showed a single set of ligand
resonances. The pyridine methyl group appeared as a singlet at
2.11 ppm and the dimethylamido groups appeared as a broad
singlet at 2.96 ppm; the resonances had a ratio 1 : 2 indicating
the stoichiometry of the complex. The broadness of some of
the resonances indicated a degree of fluxionality in the complex.
The spectrum in d6-benzene at 343 K was significantly sharper,
but on cooling only further broadening occurred down to the
viscosity limit of the solvent.
Fig.
2 Thermal ellipsoid plot of the molecular structure of
K-trans,cis,cis-[ZrL3 (CH2But)2] (H atoms omitted).
2
1
The H NMR spectrum of this compound at 298 K (Fig. 3)
showed a single set of aminopyridinato resonances, for example
the quartet (a) for the a-methylbenzyl methine group at ca. 4.25,
and two well-resolved pyridine C–H resonances (b) and (c) at
ca. 6.0 and 6.4 ppm. The neopentyl CH2 resonances appear
as a pair of AB doublets at ca. 1.60 ppm. At 343 K (not
shown) the CH2 resonances sharpened, but there were no other
significant changes in the spectrum. The spectrum broadened on
lowering the temperature to 223 K then began to sharpen as the
temperature was reduced. By 193 K peaks for major and minor
components were resolved. The multiplets (a)–(c) have each split
into two resonances in the ratio ca. 88 : 12, consistent with the
presence of two diastereomers (both C2 symmetric) in this ratio.
NOESY NMR studies† established that the minor product was
linked in an exchange process with the major product.
HL2 was treated with [Zr(CH2Ph)4] to yield the complex
[ZrL2 (CH2Ph)2]. The benzyl CH2 groups appear as a broad
2
singlet resonance at 2.45 ppm in the 1H NMR spectrum. Since
the system is inherently chiral, notwithstanding the presence
of any exchange processes, we expect the Zr–CH2 groups to
behave as AB spin systems, and indeed upon heating to 343 K
this peak sharpens to give the expected pair of AB doublets.
Upon cooling, the resonances broaden extensively then sharpen
again at 203 K. The spectrum at this temperature is consistent
with the presence of at least one C2-symmetric and one C1-
symmetric species in the ratio ca. 2 : 1, although the appear-
ance of additional broad minor resonances, possibly arising
from other diastereomers, prevents a confident assessment of
diastereoselection. Suffice to say that there is little control of
stereochemistry with respect to the metal coordination sphere.
The treatment of HL2 with one equivalent of [Zr(CH2But)4]
yielded the yellow oil [ZrL2(CHBut)3]. When higher ratios of
ligand to metal were tried the same major product was isolated.
The above studies indicate that while the tetrahydro-
naphthylamine-based ligands HL1,2 may form complexes with
the appropriate stoichiometry, there seems to be little control in
terms of diastereoselection of chirality-at-zirconium.
While this diastereoselection is far from perfect, it does give
us an ideal opportunity to test the predictive power of DFT
◦
3
t
˚
Table 1 Selected bond lengths [A] and angles [ ] for [ZrL (CH2Bu )2]
2
Zr(1)–N(1)
Zr(1)–N(2)
Zr(1)–N(3)
2.334(5)
2.210(6)
2.360(5)
Zr(1)–N(4)
Zr(1)–C(27)
Zr(1)–C(32)
2.189(6)
2.282(8)
2.277(8)
N(1)–Zr(1)–N(3)
N(2)–Zr(1)–N(1)
N(2)–Zr(1)–N(3)
N(2)–Zr(1)–C(27)
N(2)–Zr(1)–C(32)
N(4)–Zr(1)–N(1)
N(4)–Zr(1)–N(2)
N(4)–Zr(1)–N(3)
172.0(2)
58.0(2)
116.1(2)
98.0(3)
135.7(3)
116.0(2)
99.2(2)
58.1(2)
N(4)–Zr(1)–C(27)
N(4)–Zr(1)–C(32)
C(27)–Zr(1)–N(1)
C(27)–Zr(1)–N(3)
C(32)–Zr(1)–N(1)
C(32)–Zr(1)–N(3)
C(32)–Zr(1)–C(27)
135.0(3)
97.8(3)
108.3(3)
77.0(3)
77.8(3)
107.7(3)
98.2(3)
Zirconium complexes of L3
Treatment of [Zr(NMe2)4] with two equivalents of HL3 gave
a mixture of compounds, with the major product being
D a l t o n T r a n s . , 2 0 0 4 , 4 0 5 0 – 4 0 5 8
4 0 5 1

View Online
Fig. 3 Variable temperature NMR spectra of [ZrL3(CH2But)2].
calculations with respect to thermodynamic diastereoselection;
the complexes are six-coordinate isomers of one-another in a
poorly-coordinating low polarity medium, and so the issue of
solvation is unlikely to be of importance. Using the X-ray crystal-
lography data as a starting point, the structure of K-trans,cis,cis-
[ZrL3 (CH2But)2] was calculated using DFT. The converged
2
output structure [Fig. 4(a)] was found to be superimposable
on the X-ray crystallographic structure (Fig. 2). The seven
remaining diastereomers of [ZrL3 (CH2But)2] were investigated
2
also; the only energetically feasible diastereomer D-trans,cis,cis
[Fig. 4(b)] was found to be 4 kJ mol−1 higher in energy than the
crystallographically-observed structure. Assuming that this is a
difference in Gibbs free energy, it corresponds to a ratio of 92 : 8
in good agreement with the experimental value of 88 : 12.
Fig. 5 Chem3D representation (some atoms removed for clarity)
of the two diastereomers of [ZrL3 (CH2But)2] (Fig. 4), showing the
2
a-methylbenzyl group orientations.
responsible for bringing the energy of this structure relatively
close to the crystallographically observed K-trans,cis,cis-isomer.
We wished to explore the exchange process between di-
astereomers, and since standard variable temperature NMR
coalescence experiments are inappropriate for this system, the
method of Selective Polarisation Transfer–Selective Inversion
Recovery (SPT-SIR)12 was used.† Observed rates j for exchange
between diastereomers of 5.36 s−1 and 12.64 s−1 were obtained
for two independent experiments, corresponding to a value of
Fig. 4 Chem3D representations (neopentyl But groups removed) of the
DFT-optimised structures of (a) K-trans,cis,cis-[ZrL3 (CH2But)2] and
2
(b) the diastereomer closest in energy D-trans,cis,cis-[ZrL3 (CH2But)2].
2
It may be apparent from Fig. 4, or indeed from the chirality
descriptors, that the major K-trans,cis,cis diastereomer (a) and
the minor D-trans,cis,cis diastereomer (b) are related by simple
inversion of chirality of the inner coordination spheres (compare
II and I, Fig. 1). We note also that in both structures the a-
methylbenzyl substituents are oriented (via rotation about the
C–N bonds) so that the methyl and phenyl substituents avoid
the pyridine 3-H atom. In the observed diastereomer (a) with
K helicity at zirconium, this orients a phenyl group into a
region of steric space at the back of the complexes as shown
in Fig. 5(a). For the D diastereomer (b) the methyl groups
occupy this position and the phenyl groups are caused to interact
with the pyridine rings of the aminopyridinate ligand opposite
Fig. 5(b). While this is sterically unfavourable, there will be an
attractive contribution from p–p interactions, which is perhaps
DG‡ = 43 3 kJ mol−1. As may be apparent from the spectra
193
in Fig. 3 it was not possible to perform SPT-SIR experiments
over a sufficient range of temperatures in to allow determination
of DH‡.
Zirconium complexes of L4
Treatment of [Zr(NMe2)4] with two equivalents of HL4 in
pentane followed by cooling to −4 ◦C overnight yielded a yellow
crystallinesolid[ZrL4 (NMe2)2]. The1HNMRspectrumshowed
2
a single set of resonances at room temperature. High and low
temperature NMR studies (343–183 K) showed no significant
4 0 5 2
D a l t o n T r a n s . , 2 0 0 4 , 4 0 5 0 – 4 0 5 8

View Online
◦
◦
4
4
˚
˚
Table 3 Selected bond lengths [A] and angles [ ] for [ZrL (CH2Ph)2]
Table 2 Selected bond lengths [A] and angles [ ] for [ZrL (NMe2)2]
2
2
Zr(1)–N(1)
Zr(1)–N(2)
Zr(1)–N(3)
2.405(3)
2.205(3)
2.381(3)
Zr(1)–N(4)
Zr(1)–N(5)
Zr(1)–N(6)
2.216(3)
2.040(4)
2.048(4)
Zr(1)–N(1)
Zr(1)–N(2)
Zr(1)–N(3)
Zr(1)–N(4)
2.3767(17)
2.1767(16)
2.3119(16)
2.1753(18)
Zr(1)-C(29)
Zr(1)–C(36)
Zr(1)–C(37)
2.338(2)
2.280(2)
2.6951(18)
N(2)–Zr(1)–N(1)
N(2)–Zr(1)–N(3)
N(2)–Zr(1)–N(4)
N(3)–Zr(1)–N(1)
N(4)–Zr(1)–N(1)
N(4)–Zr(1)–N(3)
N(5)–Zr(1)–N(1)
N(5)–Zr(1)–N(2)
57.91(13)
92.37(12)
141.00(11)
79.67(11)
89.99(13)
57.73(13)
96.84(16)
103.90(16)
N(5)–Zr(1)–N(3)
N(5)–Zr(1)–N(4)
N(5)–Zr(1)–N(6)
N(6)–Zr(1)–N(1)
N(6)–Zr(1)–N(2)
N(6)–Zr(1)–N(3)
N(6)–Zr(1)–N(4)
158.39(15)
101.21(17)
97.7(2)
157.66(14)
101.99(16)
92.59(14)
103.64(17)
N(2)–Zr(1)–N(1)
N(2)–Zr(1)–N(3)
N(2)–Zr(1)–C(29)
N(2)–Zr(1)–C(36)
N(3)–Zr(1)–N(1)
N(3)–Zr(1)–C(29)
N(4)–Zr(1)–N(1)
N(4)–Zr(1)–N(2)
N(4)–Zr(1)–N(3)
58.37(6)
91.57(6)
146.86(7)
104.63(8)
95.54(6)
81.56(7)
145.98(6)
97.32(6)
58.98(6)
N(4)–Zr(1)–C(29)
N(4)–Zr(1)–C(36)
C(29)–Zr(1)–N(1)
C(36)–Zr(1)–N(1)
C(36)–Zr(1)–N(3)
C(36)–Zr(1)–C(29)
C(30)–C(29)–Zr(1)
C(37)–C(36)–Zr(1)
106.39(7)
87.70(7)
89.89(7)
119.59(6)
144.81(7)
99.22(9)
109.62(14)
88.92(12)
change in the spectra, indicating that the system is probably in
the fast exchange regime within this range of temperatures.
The molecular structure of [ZrL4 (NMe2)2] (Fig. 6, Table 2)
2
corresponds to the type IV (Fig. 1, X = NMe2). The
structure obtained is approximately C2-symmetric with cis
dimethylamido units [N(5)–Zr(1)–N(6)
=
97.7(2)◦]. The
pyridine units are approximately trans to the dimethylamido
groups [N(6)–Zr(1)–N(1) = 157.66(14) and N(5)–Zr(1)–N(3) =
158.39(15)◦]; hence like the two lowest energy isomers of
[ZrL3 (CH2But)2], this structure follows the trans influence
2
series Namide>Calkyl>Npy.8 Unlike the alkyl complex structure(s)
however the pyridine units in [ZrL4 (NMe2)2] are at the ‘back’
2
of the complex. This has a profound effect on the issue of
chirality at zirconium in that the a-methylbenzyl substituents
are placed in an area of relative steric freedom and thus their
chirality is only poorly expressed in the complex architecture.
Fig. 7 Thermal ellipsoid plot of the molecular structure of K-cis,cis,cis-
[ZrL4 (CH2Ph)2] (H atoms omitted).
2
exchange.† This confirms that the complex is one C1-symmetric
compound rather than two species in a 50 : 50 mixture. At 263 K
these resonances coalesce to one AB system and at 298 K they
sharpen somewhat and the AB ‘quartet’ structure is resolved.
Over the same temperature range, resonances for the pyridine
methyl groups (ca. 1.90 ppm) and benzylic CH groups (ca.
1.54 ppm) undergo similar processes. A NMR lineshape analysis
(Win-Dyna software, Bruker) was undertaken in the range 203–
263 K using the well-resolved pyridine methyl group resonances.
This led via a standard Eyring plot to a value of DH‡ = 42.4
4.7 kJ mol−1. We also note here the corresponding DS‡ = −32
20 J K mol−1, but since this is obtained by extrapolating a line
constructed over a relatively narrow temperature range to 1/T =
0 we do not attach any great significance to this value or the
calculated error.13
Fig.
6 Thermal ellipsoid plot of the molecular structure of
2
[ZrL4 (NMe2)2] (H atoms omitted).
Treatment of [Zr(CH2Ph)4] with two equivalents of HL4 in
toluene followed by cooling to 4 ◦C yielded a orange/yellow
crystalline solid [ZrL4 (CH2Ph)2]. The molecular structure
2
(Fig. 7, Table 3) corresponds to the K-cis,cis,cis diastereomer
VI (Fig. 1). Hence unlike the structures described above, the
aminopyridinato ligands are oriented ‘head to tail’ to give a C1-
symmetric structure. This might on first sight seem to disobey
the trans influence series described above, particularly in that
one pyridine unit at N(3) is nominally opposite an alkyl ligand
at C(36), but at 144.81(7)◦, C(36)–Zr(1)–N(3) is rather low
to be considered a trans angle. The benzyl ligand at C(36) is
NMR studies on a closely related (achiral) aminopyridinato
2
complex [Zr(g -PyNAd)2(CH2But)2] (Ad = adamantyl) led to
values DH‡ = 71.7
2.6 kJ mol−1 and DS‡ = 30
9 J
K−1 mol−1 for inversion of chirality at the metal.8 The complex
[ZrL4 (CH2Ph)2], and indeed [ZrL3 (CH2But)2] above thus have
2
2
2
g coordinated, as is frequently observed for electron-deficient
much lower barriers to inversion of chirality at the metal than
zirconium benzyls, with Zr(1)–C(36)–C(37) of 88.92(12)◦ and
does [Zr(g -PyNAd)2(CH2But)2]. This observation is consistent
2
˚
Zr–Cipso distance of 2.6943(18) A.
with an N-dissociative mechanism for topographical exchange;
we would expect pyridine dissociation to become less favourable
as the steric demand of R is increased, simply because the
aminopyridine ligand must pivot about the amido atom as
the Zr–NPy distance increases (Fig. 9). N-Dissociation is the
proposed mechanism in related systems.14–16
The 1H NMR spectrum of [ZrL4 (CH2Ph)2] in the slow
2
exchange regime at 203 K (Fig. 8) is consistent with such a
C1-symmetric structure. The alkyl region shows two slightly
broadened peaks for the a-methylbenzyl methyl at 1.42 and
1.47 ppm, with the 6-methylpyridine singlets at 1.88 and
2.04 ppm. The benzyl CH2 groups, which appear as two pairs of
AB doublets at ca. 2.29 ppm and at 2.76 ppm have been shown
by NOESY experiments at the same temperature to be in mutual
Most importantly, however, the diastereoselection of the sin-
gle chiral-at-metal isomer K-cis,cis,cis-[ZrL4 (CH2Ph)2] appears
2
to be essentially complete (>95 : 5) since no other diastereomer is
D a l t o n T r a n s . , 2 0 0 4 , 4 0 5 0 – 4 0 5 8
4 0 5 3

View Online
Fig. 8 Results of low temperature NMR spectroscopic experiments on [ZrL4 (CH2Ph)2].
2
All the complexes studied are in dynamic exchange between
diastereomers via an N-dissociative mechanism, i.e. conver-
sion from six- to five-coordinate structure followed by rapid
intramolecular scrambling. The rate of this process is highly
dependent on steric effects, but it is also interesting to note
the effect of the geometry on both this exchange rate and the
thermodynamic diastereoselection. We mentioned above that as
a result of the trans influence, the stereogenic substituents in
[ZrL4 (NMe2)2] are oriented such that they have little influence
Fig.
9 Pyridine N-dissociation in zirconium aminopyridinato
2
complexes.
on the complex architecture (Fig. 6). In addition, N-dissociation
is promoted in this type of structure since the pyridine N-atoms
are trans to amide, a strong trans effect ligand.
Hence although it is feasible to control chiral architecture
in early transition metal complexes of the form [M(AB)2X2],
significant challenges remain. As a result of the observations
made here, our current work is focussed on the development of
chiral ligands with strong trans influence donor atoms A and B,
and in which the anionic charge is delocalised more across the
AB unit.
observed. The origins of this excellent diastereoselectivity appear
to be similar to those for [ZrL3 (CH2But)2] above although
2
the emergence of an unsymmetrical diastereomer as the lowest
energy species is perhaps surprising. It was attempted to build
models of all eight diastereomers of [ZrL3 (CH2But)2], but
2
the two with trans alkyl ligands were judged to be sterically
unfeasible. The structures of the remaining six diastereomers
were calculated using DFT. The crystallographically observed
K-cis,cis,cis diastereomer, for which the calculated structure was
essentially superimposable on the crystallographic structure is
the more stable, the others being 24–74 kJ mol−1 higher in
energy.† This confirms that the single C1 symmetric diastereomer
observed by NMR spectroscopy is indeed the same as that in
Experimental
General comments
the solid state, i.e. K-cis,cis,cis-[ZrL4 (CH2Ph)2].
Where necessary, procedures were carried out under an inert
atmosphere of argon by using a dual manifold vacuum/argon
line and standard Schlenk techniques, or in an MBraun dry
box. Solvents were dried by refluxing for three days under dini-
trogen over the appropriate drying agents (sodium for toluene;
potassium for THF and benzene; sodium–potassium alloy for
diethyl ether, petroleum ether, and pentane; calcium hydride for
dichloromethane) and degassed before use. Solvents were stored
in glass ampoules under argon. All glassware and cannulae
were stored in an oven (>373 K) and flame dried immediately
prior to use. Most chemicals and reagents were purchased from
either Aldrich Chemical Company, Acros Chemical Company,
Lancaster or Strem and used without further purification.
Deuterated solvents were freeze–thaw degassed and dried by
heating to their normal boiling points over potassium (or
calcium hydride for d2-dichloromethane) in vacuo for three days
before vacuum distilling (trap-to-trap) to a clean, dry Young’s
tap ampoule and being stored in the dry box. Deuterated
2
Unfortunately we are unable to compare the diastereose-
lectivity induced by L3 and L4 for the same co-ligands since
[ZrL3 (CH2But)2] is inaccessible; treatment of HL4 with one
2
equivalent of [Zr(CH2But)4] in pentane gave the complex
[ZrL4(CH2But)3] as a sticky yellow oil.
Conclusions
Surprisingly, the conformationally flexible a-methylbenzyl based
aminopyridinato ligands promote much better control of
chirality-at-zirconium (diastereoselectivity) in this organometal-
lic system than do the tetrahydronaphthyl analogues. For
example, several diastereomers of the complex [ZrL2 (CH2Ph)2]
2
were detected, while [ZrL3 (CH2But)2] exists as a ca. 90 : 10
2
mixture of two isomers only. The complex [ZrL4 (CH2Ph)2]
2
displays excellent control of stereochemistry (>95 : 5) at 193 K
although perhaps surprisingly the observed diastereomer is
unsymmetrical. As we commented previously,9 the factors that
control structure in these complexes are subtle, but it does
appear that there is an excellent correlation between observed
selectivities and calculated energy differences from DFT.
˚
chloroform was dried in the bottle over molecular sieves (4A).
NMR spectra were recorded on Bruker ACF-250, DPX-
300, DPX-400, AV-400 and DRX-500 spectrometers and the
spectra were referenced internally using residual protio solvent
4 0 5 4
D a l t o n T r a n s . , 2 0 0 4 , 4 0 5 0 – 4 0 5 8

View Online
resonances relative to tetramethylsilane (d = 0 ppm). EI/CI mass
spectra were obtained on a VG Autospec mass spectrometer.
Infrared spectra were obtained either as Nujol mulls using
a Perkin-Elmer Paragon 1000 FTIR spectrometer, or directly
using an Avatar 320 FTIR instrument. Elemental analyses were
performed by Warwick Analytical Services. Carbon analysis for
these compounds were consistently low by ca. 0.7%, despite the
use of high combustion temperatures and combustion aids. This
can be ascribed to carbide formation.17 Flash chromatography
was performed with a FlashMaster Personal chromatography
system and a selection of pre-packed disposable columns. Thin-
layer chromatography was performed using Merck 0.25 mm
silica layer foil-backed plates.
884, 850. Anal. Calcd. for C16H18N2: C, 80.63; H, 7.61; N, 11.75.
Found: C, 80.37; N, 7.56; H, 11.72%. MS (EI) m/z 238 (M+).
(1-Phenylethyl)-pyridin-2-yl-amine HL3. Toluene (50 ml)
was added to a large Schlenk vessel charged with (S)-(−)-a-
methylbenzylamine (2.0 g, 16.5 mmol), 2-bromopyridine (3.0 g,
19.0 mmol), Pd2(dba)3 (200 mg, 0.22 mmol), dppp (200 mg,
0.48 mmol) and NaOBut (2.4 g, 25.0 mmol). The resulting deep
orange mixture was heated at 100 ◦C overnight with stirring.
On cooling to room temperature the resultant red solution was
treated with activated charcoal and passed through silica and
the silica washed with diethyl ether (75 ml). The yellow/orange
solution was dried over MgSO4 and the solvent removed under
reduced pressure. The resulting yellow oil was eluted through
a silica column using pentane : diethyl ether (5 : 1). The first
two compounds eluted were impurities, the column was then
flushed with dichloromethane to give the product as an opaque
Syntheses
Pyridin-2-yl-(1,2,3,4-tetrahydronaphthalen-1-yl)-amine HL1.
Toluene (30 mL) was added to a Schlenk vessel charged with
[Pd2(dba)3] (311 mg, 0.3 mmol), NaOBut (4.57 g, 0.05 mol) and
( )-BINAP (423 mg, 0.7 mmol). Toluene (10 mL) was added
to a vessel charged with (S)-1,2,3,4-tetrahydro-1-naphthylamine
(5.00 g, 0.03 mol) and 2-bromopyridine (5.5 g, 0.03 mol) and
degassed. The solutions were mixed and heated to 90 ◦C for
48 h. On cooling to room temperature the resultant red solution
was treated with activated charcoal and passed through silica
and the silica washed with diethyl ether (75 ml). The pale
yellow solution was then dried over MgSO4 and the solvent
removed under reduced pressure. The solid pale yellow product
1
oil (2.85 g, 75%). H NMR (293 K, d6-benzene) d 1.26 (d, 3H,
CH3, 3JHH = 7 Hz), 4.79 (m, 2H, NH/CH), 6.04 (d, 1H, Py-CH,
3
3JHH = 8 Hz), 6.40 (t, 1H, Py-CH, JHH = 8 Hz), 7.06 (t, 1H,
Py-CH, 3JHH = 8 Hz), 7.20 (m, 5H, Ar-CH), 8.31 (d, 1H, Py-CH,
1
3JHH = 8 Hz). 13C{ H} NMR (293 K d6-benzene) d 24.3 (CH3),
52.1 (CH), 107.4, 113.5 (Py-CH), 126.6, 127.4 (Ar-CH), 129.2,
137.4, 149.2 (Py-CH). IR (neat, cm−1): 3408, 3246, 3024, 2970,
2927, 2868, 1597, 1572, 1482, 1442, 1335, 1290, 1208, 1155, 1095,
985, 769, 700, 632. Anal. Calcd. for C13H14N2: C, 78.75; H, 7.12;
N, 14.13. Found: C, 78.62; H, 7.17; N, 14.03%. MS (EI) m/z 198
(M+), 183 (M+ − CH3).
1
was recrystallised from Et2O (3.2 g, 48%). H NMR (293 K,
(6-Methyl-pyridin-2-yl)-(1-phenyl-ethyl)-amine HL4. Toluene
(50 ml) was added to a large Schlenk vessel charged with (S)-
(−)-a-methylbenzylamine (2.0 g, 16.5 mmol), 2-bromo-6-methyl
pyridine (3.0 g, 17.4 mmol), Pd2(dba)3 (200 mg, 0.22 mmol),
dppp (200 mg, 0.48 mmol) and NaOBut (2.3g, 23.9 mmol). The
d6-benzene) d 1.47–1.68 (m, 2H, cyclohexyl-CH2), 1.78–1.90 (m,
2H, cyclohexyl-CH2), 2.62–2.46 (m, 2H, cyclohexyl-CH2), 4.53
(d, 1H, NH, 3JHH = 8 Hz), 5.30 (qt, 1H, CHN, 3JHH = 8 Hz), 5.98
(d, 1H, Py-CH, 3JHH = 8 Hz), 6.39 (t, 1H, Py-CH, 3JHH = 8 Hz),
6.98 (d, 1H, Ar-CH, 3JHH = 8 Hz), 7.01–7.12 (m, 3H, Ar-CH/
Py-CH), 7.35 (d, 1H, Ar-CH, 3JHH = 8 Hz), 8.21 (d, 1H, Py-CH,
◦
resulting deep orange mixture was heated at 100 C overnight
1
3JHH = 8 Hz). 13C{ H} NMR (293 K d6-benzene) d 20.5, 30.1,
with stirring. On cooling to room temperature the resultant
red solution was treated with activated charcoal and passed
through silica and the silica washed with diethyl ether (75 ml).
The yellow/orange solution was dried over MgSO4 and the
solvent removed under reduced pressure. The resulting yellow
oil was eluted through a silica column using pentane : diethyl
ether (5 : 1). The first two compounds eluted were impurities,
the column was then flushed with dichloromethane to give the
30.3 (ring CH2), 49.6 (CHN), 108.1, 113.1 (Py-CH), 126.8, 127.6,
129.6, 130.0 (Ar-CH), 137.3 (Py-CH) 138.1 (Ar-Cq), 139.1 (Ar-
Cq), 149.2 (Py-CH), 159.0 (Py-Cq). IR (neat, cm−1): 3211, 3017,
2940, 2921, 1596, 1575, 1524, 1488, 1455, 1439, 1340, 1322, 1290,
1270, 1203, 1171, 1153, 1117, 1088, 988, 763, 736. Anal. Calcd.
for C15H16N2: C, 80.32; H, 7.19; N, 12.49. Found: C, 80.25; N,
7.23; H, 12.45%. MS (EI) m/z 224 (M+).
1
product as an opaque oil (2.4 g, 69%). H NMR (293 K, d6-
(6-Methylpyridin-2-yl)-(1,2,3,4-tetrahydronaphthalen-1-yl)-
amine HL2. Toluene (30 mL) was added to a Schlenk
vessel charged with [Pd2(dba)3] (311 mg, 0.3 mmol), NaOBut
(4.57 g, 0.05 mol) and ( )-BINAP (423 mg, 0.7 mmol).
Toluene (10 mL) was added to a vessel charged with (S)-
1,2,3,4-tetrahydro-1-naphthylamine (5.00 g, 0.03 mol) and
2-bromo-6-methylpyridine (5.9 g, 0.03 mol) and degassed.
The solutions were mixed and heated to 90 ◦C for 48 h. On
cooling to room temperature the resultant red solution was
treated with activated charcoal and passed through silica and
the silica washed with diethyl ether (75 ml). The pale yellow
solution was then dried over MgSO4 and the solvent removed
under reduced pressure. The mixture was the distilled using
Kugelrohr apparatus at 120 ◦C. The first fraction was unreacted
bromopyridine and the second was the required product, a clear
oil (2.29 g, 32%). 1H NMR (293 K, d6-benzene) d 1.45–1.69 (m,
2H, cyclohexyl-CH2), 1.73–1.93 (m, 2H, cyclohexyl-CH2), 2.45
(s, 3H, Py-CH3), 2.48–2.60 (m, 2H, cyclohexyl-CH2), 4.28 (br d,
3
benzene) d 1.20 (d, 3H, CH3, JHH), 2.41 (s, 3H, Py-CH3), 4.66
3
(qn, 1H, CH, JHH), 4.76 (s, 1H, NH), 5.88 (d, 1H, Py-CH,
3
3
3JHH), 6.29 (d, 1H, Py-CH, JHH), 6.99 (t, 1H, Py-CH, JHH),
7.15 (m, 5H, Ar-CH). 13C{ H} NMR (293 K d6-benzene) d 24.5
1
(CH3), 24.9 (Py-CH3), 52.2 (CH), 104.2, 112.7 (Py-CH), 126.6,
127.4, 129.2 (Ar-CH), 138.0 (Py-CH), 145.9 (Ar-Cq), 157.6 (Py-
NCqN), 158.5 (Py-Cq). IR (neat, cm−1): 3406, 3027, 2967, 2925,
2868, 1594, 1494, 1464, 1373, 1334, 1215, 1148, 779, 700, 632.
Anal. Calcd. for C14H16N2: C, 79.21; H, 7.60; N, 13.20. Found:
C, 79.29; H, 7.59; N, 13.12%. MS (EI) m/z 212 (M+), 197 (M+ −
CH3).
[ZrL2 (NMe)2]. Pentane (20 ml) was added to a Schlenk
2
vessel charged with HL2 (300 mg, 1.26 mmol) and [Zr(NMe2)4]
(170 mg, 0.64 mmol) at room temperature. The reaction mixture
was stirred overnight at ambient temperature under argon. The
solvent was removed under reduced pressure. The yellow solid
was redissolved in pentane, filtered and the solvent volume
reduced. Upon cooling to −30 ◦C the yellow solid [ZrL2(NMe2)2]
precipitated (263 mg, 64%). 1H NMR (293 K, d6-benzene) d 1.79
(m, 2H), 1.94 (m, 2H), 2.04 (m, 2H, ring CH2), 2.11 (s, 6H, Py-
CH3), 2.17 (m, 2H), 2.72 (m, 2H), 2.86 (m, 2H, ring CH2), 2.96
(br s, 12H, NMe2), 4.88 (br m, 2H, ring CH), 5.79 (d, 2H, Py-
1H, NH, J = 11 Hz), 5.32 (br qt, 1H, cyclochexyl-CH, 3JHH
=
8 Hz), 5.89 (d, 1H, Py-CH, 3JHH = 11 Hz), 6.36 (d, 1H, Py-CH,
3JHH = 11 Hz), 6.95–7.14 (m, 4H, Ar-CH/Py-CH), 7.38 (d,
1
1H, Ar-CH, 3JHH = 12 Hz). 13C{ H} NMR (293 K d6-benzene)
d 20.6 (ring CH2), 25.1 (Py-CH3), 30.1, 30.5 (ring CH2), 49.5
(CH), 104.9, 112.3 (Py-CH), 126.8, 127.5, 129.5, 129.8 (Ar-CH),
137.9 (Py-CH), 138.0, 139.3 (Ar-Cq), 157.7 (Py-NCqN), 158.5
(Py-Cq-CH3). IR (neat, cm−1): 3414, 3019, 2927, 2860, 1593,
1489, 1458, 1386, 1331, 1271, 1222, 1156, 1100, 1034, 987, 949,
3
3
CH, JHH = 7 Hz), 5.85 (br d, 2H, Py-CH, JHH = 8 Hz), 6.84
(t, 2H, Py-CH, 3JHH = 7 Hz), 7.11 (m, 6H, Ar-CH), 7.62 (d, 2H,
Ar-CH, 3JHH = 6Hz). 13C{ H} NMR (293 K d6-benzene) d 22.6
1
(Py-CH3), 23.0, 30.4, 30.8 (ring CH2), 42.8 (NMe2), 56.5 (ring
D a l t o n T r a n s . , 2 0 0 4 , 4 0 5 0 – 4 0 5 8
4 0 5 5

View Online
CH), 102.6, 107.4 (Py-CH), 125.8, 126.2, 128.9, 137.2 (Ar-C),
139.5 (Py-CH), 140.9 (Py-C), 154.1 (NCN). Anal. Calcd. for
C36H46N6Zr: C, 66.11; H, 7.09; N, 12.85. Found: C, 65.36; H,
for C41H45N7Zr C, 67.73; H, 6.24; N, 13.49. Found: C, 67.09; H,
6.18; N, 13.38%. MS (EI) m/z 681 (M+ − NMe2), 528 (M+ − L).
[ZrL3 (CH2Ph)]. Toluene (20 ml) was added to a Schlenk
3
6.98; N, 12.72%. MS (EI) m/z 608 (M+ − NMe2), 564 (M+
−
vessel charged with HL3 (250 mg, 1.27 mmol) and [Zr(CH2Ph)4]
(220 mg, 0.64 mmol) at room temperature. The reaction mixture
was stirred overnight at ambient temperature under argon. The
solvent was removed under reduced pressure. The yellow solid
was redissolved in pentane (10 mL) and filtered and the solvent
volume reduced. Upon cooling to −30 ◦C a yellow powder
precipitated. The product was isolated by filtration via cannula
and residual solvent removed under reduced pressure (223 mg,
68%). 1H NMR (293 K, d6-benzene) d 1.57 (d, 9H, 3JHH = 7 Hz,
2NMe2).
[ZrL2 (CH2Ph)2]. Toluene (20 ml) was added to a Schlenk
2
vessel charged with HL2 (230 mg, 0.97 mmol) and [Zr(CH2Ph)4]
(220 mg, 0.48 mmol) at room temperature. The reaction mixture
was stirred overnight at ambient temperature under argon. The
solvent was removed under reduced pressure. The yellow solid
was redissolved in pentane (10 mL), filtered and the solvent
volume reduced. Upon cooling to −30 ◦C a yellow powder
precipitated. The product was isolated by filtration via cannula
3
3
CH3), 2.87 (d, 1H, JHH = 10 Hz), 3.12 (d, 1H, JHH = 10 Hz,
Bn-CH2), 4.53 (q, 3H, 3JHH = 7 Hz, CH), 5.85 (m, 6H), 6.74 (t,
3H, 3JHH = 7 Hz, Py-CH), 6.92 (t, 3H, 3JHH = 8Hz), 7.03–7.21
1
(222 mg, 62%). H NMR (293 K, d6-benzene) d 1.19–1.35 (m,
2H), 1.58–1.69 (m, 2H), 1.76–1.87 (m, 4H, ring CH2), 1.96 (s,
6H, Py-CH3), 2.45 (br s, 4H, Bn-CH2), 2.57–2.78 (m, 4H, ring
1
(m, 17H, Ar-CH), 7.58 (d, 3H, 3JHH = 7 Hz, Py-CH). 13C{ H}
3
NMR (293 K, d6-benzene) d 22.4 (CH3), 55.7 (CH), 66.5 (Bn-
CH2), 106.0, 107.4 (Py-CH), 118.7 (Bn-Cq), 125.0, 127.4 (Ar-
CH), 138.8, 142.1 (Py-CH), 145.1 (Ar-Cq), 170.2 (Py-NCqN).
Anal. Calculated for C46H46N6Zr C, 71.37; H, 5.99; N, 10.86.
Found: C, 70.84; H, 5.87; N, 10.80%. MS (EI) m/z 773 (M+),
681 (M+ − Bn).
CH2), 4.98 (br s, 2H, ring CH), 5.48 (d, 2H, JHH = 8 Hz, Py-
3
3
CH), 5.89 (d, 2H, JHH = 8 Hz, Py-CH), 6.70 (t, 2H, JHH
=
1
8 Hz, Py-CH), 6.87–7.18 (m, 18H, Ar-CH). H NMR (343 K,
d6-benzene) d 1.23–1.35 (m, 2H), 1.59–1.70 (m, 2H), 1.76–1.88
(m, 4H, ring CH2), 1.99 (s, 6H, Py-CH3), 2.48 (q, 4H, J = 10 Hz,
Bn-CH2), 2.58–2.78 (m, 4H, ring CH2), 5.01 (br t, 2H, J = 8 Hz,
ring CH), 5.52 (d, 2H, J = 8 Hz), 5.90 (d, 2H, J = 8 Hz), 6.74
(t, 2H, J = 8 Hz, Py-CH), 6.85 (t, 2H, J = 7 Hz, Bn-CH), 6.92–
[ZrL3 (CH2But)2]. Toluene (20 ml) was added to a Schlenk
2
vessel charged with [Zr(CH2But)4] (467 mg, 1.25 mmol) and HL3
(493 mg, 2.5 mmol) at room temperature. The reaction mixture
was stirred overnight at ambient temperature under argon. The
solvent was removed under reduced pressure and the residue was
redissolved in pentane (5 ml) and filtered. Cooling overnight
7.14 (m, 16H, Ar-CH). 13C{ H} NMR (293 K d6-benzene) d 23.0
1
(Py-CH3), 23.3, 29.2, 30.4, 58.2 (ring CH), 72.8 (Bn-CH2), 104.8,
110.8 (Py-CH), 122.5 (Bn-CH), 126.6, 126.7, 127.5, 128.6, 128.7
(Ar-CH), 137.7, 139.9 (cyclohexyl-Cq), 140.6 (Py-CH), 155.0
(Py-NCqN), 170.2 (Py-Cq-CH3). Anal. Calcd. for C46H48N4Zr:
C, 73.85; H, 6.47; N, 7.49. Found: C, 73.27; H, 6.35; N, 7.43%.
MS (EI) m/z 655 (M+ − Bn).
◦
1
at −30 C afforded orange crystals (452 mg, 58%). H NMR
(293 K, d6-benzene) d 1.31 (s, 18H, Np-But), 1.56 (bq, 4H, Np-
3
3
CH2), 1.66 (d, 6H, JHH = 7 Hz, CH3), 4.35 (q, 2H, JHH
=
7 Hz, CH), 5.85 (d, 2H, 3JHH = 8 Hz), 5.98 (t, 2H 3JHH = 8 Hz),
[ZrL2(CH2But)3]. Pentane (20 ml) was added to a Schlenk
vessel charged with HL2 (230 mg, 0.97 mmol) and [Zr(CH2But)4]
(363 mg, 0.97 mmol) at room temperature. The reaction mixture
was stirred overnight at ambient temperature under argon. The
solvent was removed under reduced pressure. The yellow solid
was redissolved in pentane (10 mL), filtered and the solvent
removed under reduced pressure. A yellow thick oil was obtained
and determined to be [ZrL2(CH2But)3] by NMR spectroscopy.
1H NMR (293 K, d6-benzene) d 1.24 (s, 27H, But), 1.48 (dd, 6H,
Np-CH, J = 12 Hz), 1.60–1.88 (m, 2H), 1.92–2.15 (m, 2H, ring
CH2), 2.16 (s, 3H, Py-CH3), 2.60–2.89 (m, 2H, ring CH2), 5.12
3
6.85 (t, 2H, JHH = 8 Hz, Py-CH), 7.03 (m, 6H), 7.34 (m, 4H,
Ar-CH), 8.05 (d, 2H, Py-CH, 3JHH = 8 Hz). 1H NMR (343 K,
d6-benzene) d 1.24 (s, 18H, Np-But), 1.52 (q, 4H, 3JHH = 14 Hz,
Np-CH2), 1.65 (d, 6H, 3JHH = 7 Hz, CH3), 4.41 (q, 2H, 3JHH
=
7Hz, CH), 5.92 (d, 2H, 3JHH = 8 Hz), 6.03 (t, 2H, 3JHH = 8 Hz),
3
6.93 (t, 2H, JHH = 8 Hz, Py-CH), 6.99–7.08 (m, 6H), 7.34 (d,
2H, JHH = 8 Hz, Ar-CH), 8.02 (d, 2H, 3JHH = 8 Hz, Py-CH).
3
13C{ H} NMR (293 K, d6-benzene) d 24.6 (Np-But), 34.9 (Cq),
1
35.1 (CH3), 57.4 (Ar-CH), 87.7 (CH2), 106.7, 109.0 (Py-CH),
126.6, 126.7, 128.4 (Ar-CH), 142.2, 142.3 (Py-CH), 145.6, 171.9
(Cq). Anal. Calculated for C36H48N4Zr C, 68.85; H, 7.70; N, 8.92.
(br s, 1H, ring CH), 5.66 (d, 1H, 3JHH = 8 Hz), 5.84 (d, 1H, 3JHH
=
Found: C, 68.21; H, 7.58; N, 8.83%. MS (EI) m/z 556 (M+
−
3
3
8 Hz), 6.75 (t, 1H, JHH = 8 Hz, Py-CH), 7.02 (d, 1H, JHH
=
Np).
7 Hz), 7.08 (t, 1H, 3JHH = 7 Hz), 7.17 (t, 1H, 3JHH = 7 Hz), 7.74
1
(d, 1H, 3JHH = 7 Hz, Ar-CH). 13C{ H} NMR (293 K d6-benzene)
[ZrL4 (NMe2)2]. Pentane (20 mL) was added to a Schlenk
2
d 23.2 (ring CH2), 23.4 (Py-CH3), 30.7 (ring CH2), 35.46 (But),
36.4 (ring CH2), 99.1 (Np-CH2), 101.1, 110.7 (Py-CH), 127.1,
127.1, 129.8 (Ar-CH), 142.3 (Py-CH), 154.9 (Py-NCqN), 168.8
(Py-Cq-CH3).
vessel charged with HL4 (350 mg, 1.65 mmol) and [Zr(NMe2)4]
(220 mg, 0.83 mmol) at room temperature. The colour of the
reaction mixture immediately turned yellow and was stirred at
ambient temperature for 1 h. Cooling overnight to 4 ◦C afforded
yellow crystals (152 mg, 30%). 1H NMR (293 K, d6-benzene) d
[ZrL3 (NMe2)]. Pentane (20 ml) was added to a Schlenk
3
3
1.73 (d, 6H, JHH = 7 Hz), 2.08 (bs, 6H, CH3), 3.19 (bs, 12H,
vessel charged with HL3 (400 mg, 2.03 mmol) and [Zr(NMe2)4]
(181 mg, 0.68 mmol) at room temperature. The reaction mixture
was stirred overnight at ambient temperature under argon. The
solvent was removed under reduced pressure. The yellow solid
was redissolved in pentane (10 mL), filtered and the solvent
volume reduced. Upon cooling to −30 ◦C a yellow powder
precipitated which was then isolated by filtration via cannula.
The residual solvent was removed under reduced pressure and
NMe2), 4.44 (q, 2H, 3JHH = 7 Hz, CH), 5.74 (d, 2H, 3JHH = 8 Hz),
5.87 (d, 2H, 3JHH = 8 Hz), 6.77 (t, 2H, 3JHH = 8 Hz, Py-CH), 7.17
1
(t, 2H, 3JHH = 7 Hz), 7.27 (m, 4H), 7.59 (m, 4H, Ar-CH). 13C{ H}
NMR (293 K d6-benzene) d 22.5 (CH3), 27.0 (Cq), 44.4 (NMe2),
58.6 (CH), 103.4, 108.9 (Py-CH), 126.9, 127.0, 129.2 (Ar-CH),
1
136.9 (Cq), 140.5 (Py-CH), 149.1 (Cq). H NMR (343 K, d6-
benzene) d 1.73 (d, 6H, 3JHH = 7 Hz, CH3), 2.08 (s, 6H, Py-CH3),
3
3.17 (s, 12H, NMe2), 4.51 (q, 2H, JHH = 7 Hz, CH), 5.77 (d,
the resulting yellow solid was determined to be [ZrL3 (NMe2)]
3
2H, 3JHH = 8 Hz), 5.88 (d, 2H, 3JHH = 8 Hz), 6.82 (t, 2H, 3JHH
7 Hz), 7.51 (d, 4H, 3JHH = 7 Hz, Ar-CH). Anal. Calculated for
C32H42N6Zr C, 63.85; H, 7.03; N, 13.96. Found: C, 63.07; H,
6.94; N, 13.86%. MS (EI) m/z 600 (M+).
=
=
1
(348 mg, 71%). H NMR (293 K, d6-benzene) d 1.66 (d, 9H,
3
3
8 Hz, Py-CH), 7.13 (t, 2H, JHH = 7 Hz), 7.27 (t, 4H, JHH
3JHH = 7 Hz, CH3), 3.14 (s, 6H, NMe2), 4.68 (q, 3H, 3JHH = 7 Hz,
CH), 5.87 (t, 3H, 3JHH = 8 Hz), 5.95 (d, 3H, 3JHH = 8 Hz), 6.79
(t, 3H, 3JHH = 8 Hz, Py-CH), 7.08 (m, 9H), 7.27 (d, 6H, 3JHH
=
1
7 Hz, Ar-CH), 7.62 (d, 3H, 3JHH = 8 Hz, Py-CH). 13C{ H} NMR
(293 K, d6-benzene) d 23.2 (CH3), 41.8 (N(CH3)2), 56.6 (CH),
107.1, 107.9 (Py-CH), 126.1, 126.8, 128.5 (Ar-CH), 138.9, 143.4
(Py-CH), 146.8 (Ar-Cq), 170.1 (Py-NCqN). Anal. Calculated
[ZrL4 (CH2Ph)2]. Toluene (20 ml) was added to a Schlenk
2
vessel charged with HL4 (250 mg, 1.18 mmol) and [Zr(CH2Ph)4]
(275 mg, 0.60 mmol) at room temperature. The reaction mixture
4 0 5 6
D a l t o n T r a n s . , 2 0 0 4 , 4 0 5 0 – 4 0 5 8

View Online
Table 4 Experimental data for the X-ray diffraction studies
[ZrL3 (CH2But)2]
[ZrL4 (NMe2)2]
[ZrL4 (CH2Ph)2]
2
2
2
Molecular formula
Formula weight
Crystal system
Space group
C36H48N4Zr
628.00
Orthorhombic
P212121
10.7595(3)
12.7065(2)
25.2894(4)
—
3457.46(12)
180(2)
C32H42N6Zr
601.94
Monoclinic
P21
9.2445(4)
15.8682(7)
11.4990(5)
110.592(4)
1579.06(12)
180(2)
C42H44N4Zr
696.03
Orthorhombic
P212121
9.9881(7)
18.2656(13)
19.6287(14)
—
3581.0(4)
180(2)
˚
a/A
˚
b/A
˚
c/A
b/◦
3
˚
Cell volume/A
T/K
Z
4
2
4
l/mm−1
0.346
0.378
0.342
Total reflections
17938
6081
0.0779, 0.1163
10721
6657
0.0467, 0.0918
23492
8707
0.0331, 0.0673
Independent reflections
R1,a wR2b [I>2r(I)]
2
2
a Conventional R = R||Fo| − |Fc||/R|Fo| for observed reflections having Fo > 2r(Fo2). b wR2 = [Rw(Fo − Fc2)2/Rw(Fo2)2]1/2 for all data.
was stirred overnight at ambient temperature in the absence
of light under argon. The solvent was removed under reduced
pressure. The yellow solid was redissolved in pentane and filtered
and the solvent volume reduced. Upon cooling to a yellow solid
Crystallography
Crystals were coated in an inert oil prior to transfer to a
cold nitrogen gas stream on a Bruker-AXS SMART three
circle CCD area detector diffractometer system equipped with
1
was obtained (242 mg, 30%). H NMR (293 K, d6-benzene) d
˚
Mo-Ka radiation (k = 0.71073 A). Data were collected using
1.54 (d, 6H, 3JHH = 7 Hz, CH3), 1.90 (s, 6H, Py-CH3), 2.60 (d,
narrow (0.3◦ in x) frame exposures. Intensities were corrected
semiempirically for absorption, based on symmetry-equivalent
and repeated reflections (SADABS18). Structures were solved by
direct methods (SHELXS19) or by the location of heavy atom
sites by Patterson interpretation of DF-data with additional light
atoms found by Fourier methods. All non-hydrogen atoms were
refined anisotropically. All H atoms were constrained with a
riding model; U(H) was set at 1.2 (1.5 for methyl groups) times
Ueq for the parent atom. Programs used were Bruker AXS
SMART20 (control), SAINT (integration), and SHELXTL for
structure solution, refinement, and molecular graphics. See
Table 4 for crystallographic data.
3
3
2H, JHH = 11 Hz), 2.81 (d, 2H, JHH = 11 Hz, Bn-CH2), 4.48
3
(br qn, 2H, NCH), 5.67 (d, 2H, JHH = 8 Hz), 5.87 (d, 2H,
3JHH = 8 Hz), 6.73 (t, 2H, JHH = 8 Hz, Py-CH), 6.83 (t, 2H,
3
3JHH = 7 Hz), 7.05–7.12 (m, 18H, Ar-CH). H NMR (193 K,
1
d2-dichloromethane) d 1.43 (br s, 3H), 1.47 (br s, 3H, CH3), 1.88
3
(s, 3H), 2.04 (s, 3H, Py-CH3), 2.18 (d, 1H, JHH = 8 Hz), 2.39
3
3
(d, 1H, JHH = 8 Hz), 2.65 (d, 1H, JHH = 8 Hz), 2.87 (d, 1H,
3JHH = 8 Hz, Py-CH2), 4.12 (br qt, 1H), 4.34 (br qt, 1H, CH),
3
3
5.46 (d, 1H, JHH = 7 Hz), 5.61 (d, 1H, JHH = 7 Hz), 6.10 (d,
1H, 3JHH = 7 Hz), 6.32 (d, 1H, 3JHH = 7 Hz, Py-CH), 6.48–6.64
(m, 5H), 6.84–7.08 (m, 15H, Ar-CH), 7.20 (t, 1H, 3JHH = 7 Hz,
Py-CH). 13C{ H} NMR (293 K d6-benzene) d 23.2 (Py-CH3),
1
CCDC reference numbers 248563–248565.
See http://www.rsc.org/suppdata/dt/b4/b413023e/ for cry-
stallographic data in CIF or other electonicformat.
27.3 (CH3), 58.5 (NCH), 74.2 (Bn-CH2), 104.7, 111.5 (Py-CH),
122.2 (Ar-Cq), 126.9, 127.0 - 129.0 (Ar-CH), 140.7 (Py-CH),
146.7 (Ar-Cq), 159.8 (Py-NCqN), 168.9 (Py-Cq-CH3). 13C{ H}
1
NMR (193 K d2-dichloromethane) d 22.8, 24.9 (Py-CH3), 27.8
(CH3), 57.9, 73.7 (CH), 72.1, 73.7 (Bn-CH2), 103.4, 105.2, 112.0,
113.4 (Py-CH), 120.7, 125.4, 126.3, 126.7, 126.9, 127.2, 129.0,
129.3, 129.6, 132.0, 135.8, 140.7, 141.7, 148.4, 150.5, 154.2, 155.7
(Ar-CH), 169.7, 169.9 (Py-CH). Anal. Calcd. for C42H44N4Zr:
C, 72.47; H, 6.37; N, 8.05. Found: C, 71.64; H, 6.34; N, 8.12%.
MS (CI) m/z 619 (M+ − Ph).
Acknowledgements
P. S. thanks the University of Warwick for a studentship (to
E. J. C.) and EPSRC for support.
References
1 (a) H. Brunner, Eur. J. Inorg. Chem., 2001, 905; (b) H. Brunner,
Angew. Chem., Int. Ed., 1999, 38, 1194; (c) C. Ganter, Chem. Soc.
Rev., 2003, 32, 130; (d) J. W. Faller and A. R. Lavoie, New. J. Chem.,
2003, 27, 899.
2 U. Knof and A. von Zelewsky, Angew. Chem., Int. Ed., 1999, 38, 302;
A. von Zelewsky and O. Mamula, J. Chem. Soc., Dalton Trans., 2000,
38, 219.
3 A. P. Smirnoff, Helv. Chim. Acta, 1920, 3, 177.
4 K. Mun˜iz and C. Bolm, Chem. Eur. J, 2000, 6, 2309; M. T. Ashby
and J. Halpern, J. Am. Chem. Soc., 1991, 113, 589; R. Noyori and S.
Hashiguchi, Acc. Chem. Res., 1997, 30, 97.
[ZrL4(CH2But)3]. Pentane (20 ml) was added to a Schlenk
vessel charged with HL4 (200 mg, 0.95 mmol) and [Zr(CH2But)4]
(356 mg, 0.95 mmol) at room temperature. The reaction mixture
was stirred overnight at ambient temperature in the absence
of light under argon. The solvent was removed under reduced
pressure. The yellow solid was redissolved in pentane (10 mL)
and filtered and the solvent removed. A thick yellow oil was
5 P. D. Knight and P. Scott, Coord. Chem. Rev., 2003, 243, 125.
6 I. J. Munslow, A. J. Clarke, R. J. Deeth, I. Westmoreland and P. Scott,
Chem. Commun., 2002, 1868.
1
obtained. H NMR (293 K, d6-benzene) d 1.24 (s, 27H, But),
1.59 (s, 6H, CH2), 1.76 (d, 3H, 3JHH = 7 Hz, CH3), 2.16 (s, 3H,
7 R. Kempe, Eur. J. Inorg. Chem., 2003, 791.
Py-CH3), 4.99 (q, 1H, 3JHH = 7 Hz, CH), 5.72 (d, 3JHH = 8 Hz),
8 C. Morton, P. O’Shaughnessy and P. Scott, Chem. Commun., 2000,
3
3
2099.
5.84 (d, 1H, JHH = 8 Hz), 6.74 (t, 1H, JHH = 8 Hz, Py-CH),
3
3
9 E. J. Crust, I. J. Munslow, C. Morton and P. Scott, Dalton Trans.,
2004, 2257.
10 J. P. Wolfe, H. Tomori, J. P. Sadihghi, J. Yin and S. L. Buchwald,
J. Org. Chem., 2000, 65, 1158.
11 M. Oberthu¨r, G. Hillebrand, P. Arndt and R. Kempe, Chem.
Ber./Recl., 1997, 130, 789.
7.12 (m, 1H), 7.26 (t, 2H, JHH = 7 Hz), 7.53 (d, 2H, JHH
=
7 Hz, Ar-CH). 13C{ H} NMR (293 K d6-benzene) d 21.2 (Ar-
CH3), 23.1 (CH3), 35.1 (Np-But), 36.1 (Np-Cq), 56.1 (CH), 98.8
(Np-CH2), 103.2, 110.7 (Py-CH), 126.8, 126.9, 128.7 (Ar-CH),
141.9 (Py-CH),144.8 (Ar-Cq), 154.4 (Py-Cq), 168.3 (NCqN).
1
D a l t o n T r a n s . , 2 0 0 4 , 4 0 5 0 – 4 0 5 8
4 0 5 7

View Online
12 (a) F. W. Dahlquist, K. J. Longmur and R. B. Du Vernet, J. Magn.
Reson., 1970, 17, 406; for recent example see: (b) D. A. Leigh, A.
Murphy, J. P. Stuart, M. S. Deleuze and F. Zerbetto, J. Am. Chem.
Soc., 1998, 120, 6458.
13 L. Blank, Statistical Procedures for Engineering, Management and
Science, McGraw-Hill, New York, 1980, 1st edn., ch. 27, pp. 487–515;
L. C. Hamilton, Regression With Graphics - A second course in applied
statistics, Brooks/Cole, California, 1st edn., 1992, ch. 2, pp. 29–64.
14 D. G. Bickley and N. Serpone, Inorg. Chem., 1979, 18, 2200.
15 T. Tsukahara, D. C. Swenson and R. F. Jordan, Organometallics,
1997, 16, 3303.
16 I. Kim, Y. Nishihara, R. F. Jordan, R. D. Rogers, A. L. Rheingold
and G. P. A. Yap, Organometallics, 1997, 16, 3314.
17 D. W. Carpenetti, L. Kloppenburg, J. T. Kupec and J. L. Petersen,
Organometallics, 1996, 15, 1572.
18 G. M. Sheldrick, SADABS, Program for area detector adsorption cor-
rection, Institute for Inorganic Chemistry, University of Go¨ttingen,
Germany, 1996.
19 G. M. Sheldrick, SHELXS-97, Program for solution of crystal
structures, University of Go¨ttingen, Germany, 1997.
20 SMART and SAINT, Bruker Analytical X-ray Systems, Inc., Madi-
son, WI, 1998.
4 0 5 8
D a l t o n T r a n s . , 2 0 0 4 , 4 0 5 0 – 4 0 5 8