Radical and migratory insertion reaction mechanisms in Schiff base zirconium alkyls

Article abstract of DOI:10.1016/j.jorganchem.2005.03.043

Four salicylaldimine derivatives H₂L^4-7 of 2,2′-diamino-6,6′-dimethylbiphenyl, where the CN bond is sterically protected by substituents on the phenol ring, form alkyls of zirconium, cis-α-[ZrL^4-7(CH₂Ph)₂

Full text of DOI:10.1016/j.jorganchem.2005.03.043

Journal of Organometallic Chemistry 690 (2005) 5125–5144
www.elsevier.com/locate/jorganchem
Radical and migratory insertion reaction mechanisms in Schiff
base zirconium alkyls
a
Paul D. Knight , Guy Clarkson , Max L. Hammond ,
a
a
b
Brian S. Kimberley , Peter Scott
a,
*
a
Department of Chemistry, University of Warwick, Gibbett Hill Road, Coventry CV4 7AL, UK
Research & Technology Centre, BP Chemicals snc, Boite Postale No. 6, 13117 Lavera, France
b
Received 9 February 2005; received in revised form 22 March 2005; accepted 23 March 2005
Available online 4 May 2005
Abstract
Four salicylaldimine derivatives H2L4–7 of 2,20-diamino-6,60-dimethylbiphenyl, where the C@N bond is sterically protected by
substituents on the phenol ring, form alkyls of zirconium, cis-a-[ZrL4–7(CH2Ph)2]. Rather than decomposing via the established
pathway of 1,2-migratory insertion of an alkyl group to imine, they undergo a radical mechanism. This is evidenced by the large
number of products observed, kinetic and thermodynamic data (Rice-Herfeld, 3/2 order, positive DSꢀ), response to steric factors,
and the fact that switching to a less stable radical leaving group inhibits the reaction. In contrast, the 1,2-migratory insertion is
a clean, first-order intramolecular process with negative DSꢀ. The steric modification of the ligands nevertheless transforms an inac-
tive precatalyst into a stable system for the polymerisation of ethene. Closely related unbridged salicylaldimine catalysts are known
to be highly active catalysts, but in most cases they appear to suffer from high temperature instability. The first examples of zirco-
nium alkyls of this class are isolated, and it is found that they are inherently much more resistant to decomposition by either path-
way (migratory insertion or radical). Structural studies are used to interpret this variance in behaviour; the biaryl-bridged complexes
are pre-organised for both reactions, while the unbridged systems would have to undergo significant ordering prior to activation.
Correspondingly, the unbridged systems are not noticeably affected by the same steric modification of the ligand, and it is concluded
that the more likely mechanism of catalyst death in the latter is ligand loss (i.e. transfer to aluminium from co-catalyst).
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Mechanism kinetics; Zirconium; Alkene polymerisation
1. Introduction
C@N units with less reactive linkers, and while this
has met with some successes in enantioselective catalysis
[2], the favourable properties of Schiff base systems –
strong ligand–metal bond, ease of synthesis, tunability,
crystallinity, structural rigidity – have encouraged us
to investigate the possibilities for improvement of their
stability in early transition complexes. In this context,
salicylaldimine (SA) complexes of the group 4 metals,
[M(SA)2Cl2] (Fig. 1), which when combined with, e.g.
methylaluminoxane (MAO) yield extremely active or
otherwise useful catalysts for the polymerisation of alk-
enes, are of particular interest [3]. It is assumed that me-
tal alkyls are involved in these catalyses, and evidence
A fundamental concern for those using Schiff base
complexes in catalytic applications is the reactivity of
the C@N bond, specifically where this limits the number
of turnovers. This issue is particularly important in early
transition chemistry, where coordination of the imine
unit renders it highly electrophilic [1]. We have sought
to avoid this issue through the replacement of the
*
Corresponding author. Tel.: +44 24 7652 3238; fax: +44 24 7657
2710.
E-mail address: peter.scott@warwick.ac.uk (P. Scott).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.03.043

5126
P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
R''
R'
R
O
O
M
O
R''
N
Cl
R
O
M = Ti, Zr
Cl
N
R'
R'
=
N
N
N
O
N
O
R
N
N
Zr
O
R
Zr
O
R
R'
R'
Fig. 1. Group 4 salicylaldimine complexes.
R''
has been presented that alkyl cation species
[M(SA)2Me]+ are formed on treatment of [M(SA)2Cl2]
with MAO [4]. Coates has mentioned that analogous
ketimino ligands form stable alkyl complexes on treat-
ment with [Ti(CH2Ph)4] [5]. In contrast with many other
olefin polymerisation systems however an alkyl cation
has not been isloated.
R''
Scheme 1. 1,2-Migratory insertion of a metal bound benzyl group to
an imine carbon [7].
2. Results and discussion
2.1. Ligand designs
We have recently shown that biaryl-bridged salicyl-
aldimine derivatives H2L1–3 (Fig. 2) form, under appro-
Reducing the steric demand in the phenolate 2-posi-
tion (R0 in Scheme 1) reduces the rate of 1,2-migratory
insertion in the metal benzyl complexes [ZrLn(CH2Ph)2]
(i.e. L3 < L1 < L2) [7]. This is however an unsatisfactory
resolution of the problem of complex stability for two
reasons; (i) even when this group is hydrogen the 1,2-
migratory insertion pathway is still accessible and occurs
over a short period of time (<48 h), (ii) it is known that
group 4 iminophenolate complexes require sterically
demanding substituents (e.g. tBu) in this position to fur-
nish highly active catalysts for alkene polymerisation [3].
We thus sought other modifications.
priate conditions, isolable alkyls of zirconium [ZrL1–3R2]
with cis-a geometry (C2-symmetric with cis alkyl
ligands) [7]. Subsequently, however, they decompose
via 1,2-migratory insertion of an alkyl group to imine
(Scheme 1) followed in some instances by a second sim-
ilar reaction. This provides an explanation for their
complete inactivity in olefin polymerisation. Here we re-
port a detailed kinetic investigation of this reaction, an
attempt to prevent the process by ligand modification,
discovery of a new decomposition mechanism, and the
development of a stable polymerisation catalyst system.
We also describe some attempts to apply the lessons
learned to the SA catalyst system. Part of this work
has been briefly communicated [6].
A space-filling model of the molecular structure of
[ZrL1(CH2But)2] [7] is shown in Fig. 3(a), with the elec-
trophilic imine carbon atom indicated . We envisaged
*
that notionally moving the 4-methyl substituent to the
5-position [Fig. 3(b)] would effectively block the ap-
proach of a zirconium bound alkyl group to the imine
carbon atom. The series of ligands L1, L4, L5 was de-
signed to examine the effect of steric demand in this 5-
position. In addition, it was envisioned that the series
L6, L4, L7 would allow investigation of the effect of steric
demand in the 2-position for this unusual salicylaldi-
mine substitution pattern.
R4
R3
R2
Me
N
2
HO
R1
All R = H except:
H2L1 R1 = R3 = Me
H2L2 R1 = iPr
H2L3 R3 = tBu
H2L4 R1 = R4 = Me
H2L5 R1 = Me, R4 = iPr
H2L6 R2 = R4 = Me
H2L7 R1 = iPr, R4 = Me
H2L8 R1 = tBu, R4 = Me
H2L9 R1 = tBu, R3 = Me
Fig. 3. Space-filling models of (a) [ZrL1(CH2CMe3)2] from X-ray
molecular structure and (b) [ZrL4(CH2CMe3)2] based on (a); phenolate
Fig. 2. Numbering scheme for proligands.
4 and 5 positions and imine carbon atom ( ) indicated.
*

P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
5127
2.2. Synthesis of zirconium alkyl complexes
The negative value of entropy of activation for the
decomposition of [ZrL1(CH2Ph)2] is consistent with
the formation of an ordered (four-membered) cyclic
transition state in an intramolecular 1,2-migratory inser-
tion process. As we had proposed, placing a methyl
group in the 5-position as in complex [ZrL4(CH2Ph)2]
inhibits the formation of this cyclic transition state,
but unexpectedly a new mechanistic pathway is opened
up as evidenced by the number of products formed
and by a change in the order of reaction to 3/2. Both
observations are consistent with Rice–Herzfeld radical
propagation kinetics [9,10] (vide infra).
We propose an initiation step involving homolytic fis-
sion of the Zr–CH2Ph bond, leading to formation of two
radical species (Eq. (1), Bn = CH2Ph).2 The benzyl rad-
ical can then attack the ligand (e.g. at an imine position)
on another complex molecule to form a new radical spe-
cies (Eq. (2)). We have previously described a very clo-
sely related reaction at a Nb(IV) Schiff base system
(Scheme 2) leading to oxidation to diamagnetic Nb(V)
[11]. In the case of Zr the radical character is retained
by the ligands (Eq. (2)), and the system may react to
form a closed shell complex and a further benzyl radical
(Eq. (3)). The lack of a higher oxidation state for zirco-
nium thus enables a radical propagation process that
terminates when two radicals combine (Eq. (4)). Assum-
ing steady state conditions, Eq. (5) is eventually ob-
tained [10]. We note that the observed rate constant
kobs (Eq. (6)) and thus the thermodynamic values ob-
tained from the Eyring analysis for the process of ‘‘acti-
vation’’ are actually composite parameters arising from
initiation, propagation and termination. The positive
entropies of activation for the complexes of L4–7 (Table
1) are nevertheless consistent with a radical process.
The synthesis of [ZrL1(CH2Ph)2] was reported previ-
ously [7]. Reaction of H2L4 with zirconium tetrabenzyl
in acetonitrile yielded a precipitate which was found to
be mainly unreacted H2L4. This was probably due to
the low solubility of H2L4 in acetonitrile. The complex
[ZrL4(CH2Ph)2] was successfully synthesised in dichlo-
romethane at ꢀ78 ꢁC. The reaction of the more soluble
proligand H2L5 with [Zr(CH2Ph)4] in acetonitrile pro-
ceeded cleanly giving [ZrL5(CH2Ph)2] in high purity.
The complexes [ZrL6(CH2Ph)2] and [ZrL7(CH2Ph)2]
were prepared similarly. The NMR spectra of freshly
prepared solutions of these complexes were consistent
with cis-a geometry.
2.3. Solution stability of [ZrL4–7(CH2Ph)2]: initial
observations
We were surprised to find that the four complexes
[ZrL4–7(CH2Ph)2] underwent fairly rapid decomposition
in solution, although the spectra of the products were
markedly different from those expected for 1,2-migra-
1
tory insertion processes. After ca. 3 h at 298 K, the H
NMR spectrum of a solution of [ZrL4(CH2Ph)2] in d2-
dichloromethane displayed a very large number of new
peaks in the imine (d 7.5–9.5 ppm) and aliphatic regions.
These features, which were similar for all four complexes
with a 5-substituent, are consistent with radical decom-
position processes. Attempts at isolation of one of the
many decomposition products were unsuccessful.
2.4. Kinetic studies of the decomposition processes
The decomposition of [ZrL1(CH2Ph)2] in d2-dichloro-
1
methane was followed by H NMR spectroscopy be-
Initiation
k1
Å
Å
tween 283 and 303 K. Values for the integration of the
complex imine peak relative to the residual protio sol-
vent resonance were obtained over ca. two half-lives
where possible. First-order plots were satisfactory
over the whole temperature range (Fig. 4). Similar vari-
able temperature kinetic studies on the complexes
[ZrL4–7(CH2Ph)2] showed that they did not decompose
via first-order processes and after much experimentation
it was found that the only satisfactory fit was via 1.5 or-
der plots (e.g. Fig. 5). This unusual order was confirmed
using Vanꢁt Hoff plots of the data.
½ZrLðBnÞ ꢁ !½ZrLðBnÞꢁ þ Bn
Propagation
ð1Þ
2
k2
Å
2
Å
Bn þ ½ZrLðBnÞ ꢁ !½ZrðBn–LÞðBnÞ ꢁ
ð2Þ
ð3Þ
2
k3
Å
Å
½ZrðBn ꢀ LÞðBnÞ ꢁ !½ZrðBn–LÞðBnÞꢁ þ Bn
Termination
2
k4
Å
Å
Bn þ Bn ! Bn2
ð4Þ
ð5Þ
ð6Þ
ꢀ ꢁ
0.5
k1
k4
1.5
Activation parameters (Table 1) were subsequently
obtained via Eyring plots. The uncertainties recorded
were calculated using standard methods [8].1
Rate ¼ k2
½ZrLðBnÞ ꢁ
2
1.5
Rate ¼ kobs½ZrLðBnÞ ꢁ
2
pffiffiffiffiffiffiffiffiffiffi
using the LINEST function in Microsoft Excel.
1
Standard error in the slope, SEm ¼ se= TSSx and standard error in
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ꢀ2
2
the intercept, SEc ¼ se ð1=nÞ þ ðX =TSSxÞ. These were computed
These equations are representative examples of the type of process
described.

5128
P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
1.5
0.5
-0.5
283 K
293 K
303 K
288 K
298 K
-1.5
-2.5
0
10000
20000
30000
40000
50000
60000
Time/s
Fig. 4. First-order plots for decomposition of [ZrL1(CH2Ph)2] at various temperatures.
283 K
293 K
303 K
288 K
298 K
323 K
6.5
5.0
3.5
2.0
0.5
0
10000
20000
30000
Time/s
40000
50000
Fig. 5. Normalised 1.5 order plots for [ZrL4(CH2Ph)2].
Table 1
Activation parameters and relative rates for decomposition of
[ZrLn(CH2Ph)2]
But
But
N
Complex
krel (298 Ka) DHꢀ (kJ molꢀ1
)
DSꢀ (J Kꢀ1 molꢀ1
)
But
But
[ZrL1(CH2Ph)2]
–
+88 ( 2)
–32 ( 7)
N
O
O
N
N
N
N
[ZrL4(CH2Ph)2] 1.0
[ZrL5(CH2Ph)2] 1.8
[ZrL6(CH2Ph)2] 0.3
[ZrL7(CH2Ph)2] 1.1
a
+101 ( 5)
+113 ( 10)
+108 ( 7)
+91 ( 5)
+16 ( 16)
+62 ( 34)
+29 ( 23)
+17 ( 16)
NEt2
NEt2
Nb
O
Nb
O
NEt2
NEt2
But
But
Relative rate constant versus [ZrL4(CH2Ph)2].
But
But
Scheme 2. Addition of diethylamine-2-yl radical to a Schiff base
niobium (IV) complex.
Returning to the initiation step, we propose that the
proximity of a benzylic H atom to the imine carbon
may facilitate the homolytic fission of the metal–benzyl
bond (Fig. 6). We note that the observed rate of decom-
position of the L5 complex is almost twice that of L4 and
this difference arises in the main from the much greater
positive DSꢀ in the former where the proposed H atom
donor unit is Pr rather than Me.
Subject to the caveat mentioned above regarding the
composite nature of the thermodynamic parameters, we
also note that the enthalpies of activation decrease for
the complexes in the order L6 > L4 > L7, with steric bulk
in the phenolate 2-position increasing in the same order
(H < Me < iPr). This is the same trend as observed for
the first order 1,2-migratory insertion process and is
consistent with greater steric compression in the more
substituted complexes. The large increase in krel on
i

P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
5129
2.7. Synthesis of [ZrL8,9Cl2] and polymerisation
catalysis
R
H
R
R
R
Ar
H
N
Ar
N
+ PhCH2
These zirconium chloride complexes were synthesised
and characterised as before. Ethylene polymerisation re-
sults are summarised in Table 2. As can be seen, the
presence of the phenolate 2-tert-butyl group in [ZrL9Cl2]
does not give rise to significant polymerisation activity
on its own. In combination with 5-methyl substituent,
as in [ZrL8Cl2], polymerisation activity was observed
for the first time using this type of ligand, albeit moder-
ate according to Gibsonꢁs classification [14]. The catalyst
is unusually stable, and there is no noticeable loss of
activity at 25 and 50 ꢁC over at least a period of 1 h
(Fig. 7). The lower average productivity and activity at
50 ꢁC is due to the reduced solubility of ethylene in tol-
uene at higher temperatures.
Zr O
Zr O
Ph
Fig. 6. Radical initiation mechanism.
moving from L6 to L4 is not continued on moving to L7
however, principally because the general trend in DSꢀ is
acting against this.
2.5. Reaction of H2L4 with [Zr(CH2CMe3)4]
We explored briefly the effect of the presence of a less
stable radical leaving group on the decomposition pro-
cess. Reaction of H2L4 with zirconium tetrakis(neopen-
tyl) in dichloromethane proceeded smoothly, and a high
purity sample of the complex [ZrL4(CH2CMe3)2] was
obtained by crystallisation from pentane. This complex
only began to show traces of decomposition in solution
after leaving for several days at room temperature.
2.8. Structural implications of phenolate 5-methyl
substitution
We obtained X-ray molecular structures (Table 3) of
[ZrL7Cl2] and [ZrL9Cl2] via crystals obtained from sam-
ples of the pure complexes in d2-dichloromethane. The
molecular structures are shown in Fig. 8 with selected
bond lengths and angles given in Table 4. The view
along the Zr–O axes (Fig. 9) highlights the influence of
2.6. Synthesis and polymerisation activity of [ZrL1ꢀ7Cl2]
Since the benzyl complexes were prone to decomposi-
tion, we attempted to generate the chloride complexes as
these had previously been observed to be far more ro-
bust [12]. Treatment of the ligands with sodium hydride
followed by [ZrCl4(THF)2] proceeded without problem.
The zirconium chloride complexes, [ZrL1–7Cl2], were
then purified by sublimation at ca. 300 ꢁC, 10ꢀ6 mm Hg
in all cases and found to possess C2 symmetry. However,
treatment with MAO in toluene in the presence of ethyl-
ene at ambient temperature and 1.2 bar led to very low
uptake of gas. This lack of activity may be attributed to
the lack of steric bulk in the phenolate 2-position [3,13].
Table 2
Polymerisation activity of complexes [ZrL8,9Cl2]: precatalyst,
1.39 · 10ꢀ2 mmol; toluene (500 ml); ethene pressure, 1.2 bar; MAO:Zr
molar ratio, 1000:1
Precatalyst
Temperature (ꢁC)
Average productivity (1 h)
(kg PE/mol–Zr bar–C2 h)
[ZrL8Cl2]
[ZrL9Cl2]
[ZrL8Cl2]
[ZrL9Cl2]
25
25
50
50
65
–
40
–
60
50
298 K
323 K
40
30
20
10
0
0
0.25
0.5
Time/h
0.75
1
Fig. 7. Plot of ethene uptake vs. time/h for [ZrL8Cl2].

5130
P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
Table 3
Experimental data for the X-ray diffraction studies
[ZrL7Cl2]
[ZrL9Cl2]ÆCH2Cl2
[ZrL11Cl2]
Colour
Habit
Yellow
Block
Yellow
Block
Yellow
Block
Molecular formula
Crystal system
Space group
C
36H38Cl2N2O2Zr
C39H44Cl4N2O2Zr
Monoclinic
C2/c
C36H40Cl2N2O2Zr
Monoclinic
C2/c
Monoclinic
C2/c
˚
a (A)
19.975(4)
10.551(2)
15.521(3)
97.95(3)
3239.7(11)
4
13.349(3)
14.379(3)
20.131(4)
96.74(3)
21.166(4)
7.925(2)
˚
b (A)
˚
c (A)
20.282(4)
100.770(19)
3342.2(14)
4
b (ꢁ)
Cell volume (A )
3
˚
3837.4(13)
4
0.601
Z
l (mmꢀ1
Total reflections
)
0.539
10,364
0.523
13,587
12,481
4650 [0.1109]
0.0768, 0.1916
Independent reflections [R(int)]
R1, wR2 [I > 2r(I)]
3987 [0.0183]
0.0235, 0.0617
4189 [0.0182]
0.0225, 0.0585
Table 4
˚
Selected bond lengths (A) and angles (ꢁ) for molecular structures of
[ZrL7Cl2], [ZrL9Cl2] and cis-a-[Zr(L11)2Cl2] (vide infra)
[ZrL7Cl2]
[ZrL9Cl2]
[Zr(L11)2Cl2]
Zr–O
Zr–Cl
Zr–N
1.9987(11)
2.4246(6)
2.3221(12)
165.30(6)
75.09(6)
1.986(3)
2.4313(17)
2.347(4)
172.2(2)
72.9(2)
1.981(9)
2.432(5)
2.348(11)
157.40(5)
84.31(6)
94.38(3)
O–Zr–O
N–Zr–N
Cl–Zr–Cl
103.82(3)
108.36(10)
imine carbon and a metal bound alkyl group (in place
of Cl). This twisting of the phenolate ring may have
an additional consequence in that the phenolate 2-alkyl
substituent has been directed away from the ‘‘active
sites’’ of the catalyst. This may reduce steric compres-
sion at these sites resulting in reduced propensity for
decomposition as well as opening the sites for increased
catalytic activity.
(a)
2.9. Synthesis and stability of active species in
polymerisation catalysis
The question remained as to whether zirconium alkyl
complexes of L8 and L9 decompose via different path-
ways. All attempts failed at generating the desired com-
plexes by reaction of proligands with zirconium
tetrabenzyl and zirconium tetrakis(neopentyl). 1H
NMR spectra indicated that the reaction was much
slower than for less bulky ligands; resonances corre-
sponding to unreacted and mono-deprotonated ligand
were evident for the neopentyl reaction, and several
other products were evident for the benzyl. Attempts
at alkylation of the zirconium chloride complexes using
Grignard and lithium reagents produced a number of
unidentified products. Reactions of [ZrL8Cl2] and
[ZrL9Cl2] in NMR tubes with previously dried MAO
(10 molar equivalents) in d8-toluene resulted in
(b)
Fig. 8. Thermal ellipsoid plots of the molecular structures of (a)
[ZrL7Cl2] and (b) [ZrL9Cl2].
the 5-methyl substituent; steric compression between
this group and the imine-CH has led to a twist in the
plane of coordination of the phenolate ring, such that
the phenolate–CH3 lies between the imine group and
the adjacent Zr–Cl group. This increases the hindrance
of the pathway for 1,2-migratory insertion between the

P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
5131
Fig. 9. Salicylaldimine–Zr–Cl fragments of (a) [ZrL7Cl2] and (b) [ZrL9Cl2] extracted from X-ray molecular structures, highlighting effects of
substituents ortho to the imine carbon.
consumption of the starting complexes and significant
broadening of the spectra, but no unambiguous indica-
tions of the formation of alkyl or alkyl cation species.
Since we could not generate zirconium alkyl com-
plexes of L8 and L9, we decided to use complexes of
L4 and L1 as models. An NMR tube was charged with
[ZrL1(CH2Ph)2] and B(C6F5)3, and d2-dichloromethane
ture range, however three major new imine peaks are
observed along with the disappearance of the pair of
doublets. No clear evidence for 1,2-migratory insertion
processes was observed in either case.
2.10. Application to other catalyst systems
1
was distilled into it at ꢀ78 ꢁC. The H NMR spectra
We sought to investigate the effect of 5-alkyl substitu-
tion on Fujitaꢁs zirconium iminophenolate catalysts.
Four proligands HL10–13 (Fig. 10) were synthesised via
were then recorded at ꢀ80 ꢁC and then at increments
of +10 ꢁC up to room temperature. A similar reaction
was carried out using [ZrL4(CH2Ph)2]. In both cases,
the 1H NMR spectra indicated formation of
[B(C6F5)3(CH2Ph)]ꢀ [15]. Below ca. ꢀ40 ꢁC, two imine
peaks were observed that do not correspond to the start-
ing complexes, and the presence of six methyl reso-
nances indicate that the species formed were C1
symmetric. For both reactions, new pairs of doublets oc-
curred in the region d 2.0–3.5 ppm, and we tentatively
assign all these features to the desired cationic species
[ZrL(CH2Ph)]+. Upon warming the solution containing
the proposed L4 complex cation, significant decomposi-
tion occurred between ꢀ30 and 0 ꢁC; a large number of
new imine peaks are generated and the pair of doublets
disappeared. The solution containing the L1 complex
also decomposes significantly within the same tempera-
R4
R3
N
R2
HO
R1
HL10 R1 = tBu, R4 = Me
HL11 R1 = tBu, R3 = Me
HL12 R1 = R4 = Me
HL13 R1 = R3 = Me
Fig. 10. Modified salicylaldimines and suitable control ligands.

5132
P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
condensation of the appropriate salicylaldehydes with
aniline in ethanol. Two of these ligands (HL10,12)have
a methyl group in the 5-position and two control ligands
HL11,13 have the traditional 2,4-substitution (HL11 has
previously appeared) [16].
Reactions of HL10 and HL11 with sodium hydride in
THF followed by [ZrCl4(THF)2] resulted in the produc-
tion of yellow/orange solids which were sublimated at
ca. 300 ꢁC, 10ꢀ6 mm Hg to yield yellow solids of stoichi-
ometry [Zr(L10,11)2Cl2], as indicated by mass spectrome-
1
try and CHN analysis. H NMR spectra revealed that
two isomeric complexes were present in both cases.
The major products were C2-symmetric, as indicated
by the presence of single imine, phenolate methyl and
tert-butyl resonances. The minor products (ca. 27% for
1
Fig. 12. Crystal packing between adjacent molecules of [ZrL121Cl2ꢁ.
½ZrL120Cl2ꢁ and ca. 36% for ½ZrL211Cl2ꢁ) had broad H
NMR spectra at room temperature, but cooling
(253 K for ½ZrL10Cl2ꢁ and 203 K for [ZrL11Cl2]) gave rise
Samples of ½ZrL210Cl2ꢁ and ½ZrL211Cl2ꢁ were tested at
BP laboratories under supported gas-phase conditions
with MAO co-catalyst. Both complexes had similarly
high activities and catalytic lifetimes were <10 min at
50–80 ꢁC in both instances. Thus, while substitution of
the ligand with a methyl group ortho to the imine carbon
atom does not significantly alter the intrinsic polymeri-
sation activity of the complexes, it fails to increase lon-
gevity of the catalyst in this instance. In the hope of
shedding further light on the issue of iminophenolate
catalyst stability we undertook a study of some alkyl
derivatives.
2
to sharp resonances. Two imine peaks and two pheno-
late methyl and tert-butyl resonances were observed in
both complexes. We therefore assigned these species as
having the C1-symmetric cis-b topography. The ratio
of cis-a to cis-b remains unchanged over a period of
days. Interestingly, Fujita and coworkers [16] did not
note the presence of cis-b isomers of [ZrL11Cl2],
although their NMR data are possibly consistent with
this. Coates [5] has noted the presence of cis-b isomers
of ketimino titanium complexes and others.
Crystals of cis-a-½ZrL121Cl2ꢁ were grown from toluene
solution. X-ray analysis revealed that the C2-symmetric
complex (Fig. 11) crystallises as a dimer via a face-face
p–p stacking interaction between N-aryl rings (Fig.
12). The distance between H(16A) and the centroid of
2.11. Synthesis and properties of [Zr(L10–13)2(CH2Ph)2]
In NMR tube scale experiments, the reactions
between two equivalents each of HL10–13 with [Zr(CH2-
Ph)4] were shown to give cleanly [Zr(L10–13)2(CH2Ph)2].
One example, [ZrL102(CH2Ph)2], was synthesised on a
preparative scale and was characterised in the usual
way. Attempts to grow single crystals for X-ray analysis
were unsuccessful.
˚
the proximal aryl ring is ca. 3.37 A [17]. The bond dis-
tances and angles about Zr(1) are unremarkable for this
type of complex [16] and are discussed in more detail
through comparison with those of a biaryl-bridged com-
plex in Section 2.12.
The 1H NMR spectrum of ½ZrL10ðCH2PhÞ ꢁ indi-
2
2
cated the adoption of the cis-a structure; in particular
the appearance of the CH2Ph as a pair of AB
doublets indicated that interconversion between the
chiral-at-metal structures (Fig. 13) is slow on this
O
O
N
N
CH2Ph
CH2Ph
CH2Ph
CH2Ph
Zr
Zr
N
N
O
O
Λ
∆
cis- α
cis- α
Fig. 13. Interconversion of complex enantiomers for cis-a-[Zr(L10–13)2-
(CH2Ph)2].
Fig. 11. X-ray molecular structure of cis-a-[Zr(L11)2Cl2].

P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
5133
timescale. The spectrum of ½ZrL121ðCH2PhÞ ꢁ was
All the above complexes were found to be relatively
stable with respect to the decomposition reactions de-
tailed above. After several days in solution at ambient
temperature, samples of the bulky ligand complexes
[Zr(L10,11)2(CH2Ph)2] began to show signs of formation
of 1,2-migratory insertion products, viz. a new single
imine peak and a set of three quartets in the region ca.
2
similar. For [Zr(L12,13)2(CH2Ph)2] these CH2Ph reso-
nances appeared as a broad singlet. This trend in con-
figurational stability is consistent with an N-dissociative
mechanism since bulky groups in the phenolate 2-posi-
tion would cause steric compression on lengthening of
the N–Zr bond, and hence hinder isomerisation [18].
Fig. 14. Comparison of the molecular structures of cis-a-½ZrL9Cl2ꢁ and cis-a-½ZrL11Cl2ꢁ.
2

5134
P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
d
2.5–6.5 ppm [1a]. Solutions of the complexes
[Zr(L12ꢀ13)2(CH2Ph)2] showed very little, if any, decom-
position over a period of several days.
a methyl group ortho to the imine consistently impedes
this reaction, thus leading to the remarkable increase in
catalyst stability detailed above. The unbridged com-
plexes are not pre-organised for this reaction, and as a
result the 1,2-migratory insertion is inherently slower
for these compounds. The observation that phenolate
5-methyl substitution does not affect catalyst longevity
suggests that either; (i) despite the fact that complexes
[Zr(L10–13)2(CH2Ph)2] appear to be rather stable with
respect to 1,2-migratory insertion, the greater flexibility
allows for imine migratory insertion even in the modified
catalyst, or perhaps more likely (ii) that other processes
are responsible for catalyst deactivation (vide infra).
2.12. Stability of ZrL120CH2Ph
þ
We attempted to generate an alkyl cation complex by
reaction of ½ZrL210ðCH2PhÞ ꢁ with B(C6F5)3 in an NMR
2
tube at ꢀ78 ꢁC, using d2-dichloromethane as solvent. 1H
NMR spectra were recorded at ꢀ80 ꢁC and at +10 ꢁC
increments up to room temperature. Resonances for
[B(CH2Ph)(C6F5)3]ꢀ were observed at low temperature
[15], indicating that a reaction had taken place, but
while some resonances assignable to a cationic species
[Zr(L10)2(CH2Ph)]+ were observed, by ꢀ40 ꢁC extensive
decomposition had occurred. This result and the at-
tempts to form [ZrL1,4(CH2Ph)]+ (Section 2.8), do not
bode well for isolation of a stable alkyl cationic species
such as that implicated in olefin polymerisation catalyses
by these complexes. Fujita [4] has also detected NMR
resonances consistent with such a species on treatment
of a dichloride complex with dried MAO.
3. Conclusions
Our attempt here to block sterically the 1,2-migratory
insertion process in our Schiff base group 4 alkyl com-
plexes was successful principally because of the lack of
flexibility of the system. This also results however in
the complexes being pre-organised for decomposition
via a radical process. Kinetic analysis and 1H NMR spec-
troscopy data highlight the differences between the two
pathways. For the 1,2-migratory insertion mechanism a
single product was formed in a highly diastereoselective
intramolecular manner to give an unstable intermediate.
The reaction displayed first order reaction kinetics with a
negative entropy of activation associated with ordered
transition state. The radical process gave many products
in a 1.5 order Rice–Herzfeld reaction with positive entro-
py of activation, but was inhibited through the use of a
less stable radical leaving group (neopentyl) at the metal,
thus finally giving a stable metal alkyl.
2.13. Biaryl-bridged complexes vs. non-bridged complexes
Complexes of our biaryl-bridged ligands above and
the non-bridged salicylaldimine type ligands are similar
in terms of functionality. Nevertheless, considerable dif-
ferences are observed for polymerisation activity. We
can see that in comparison to the constrained structure
of the biaryl complex [ZrL9Cl2] [Fig. 14(a)], the N-aryl
units in the non-bridged ligand complex ½ZrL121Cl2ꢁ (b)
are directed away from one another. The presence of
the biaryl unit also constrains the N–Zr–N0 angle to
ca. 72.9ꢁ compared with 84.3ꢁ for ½ZrL121Cl2ꢁ, (Table
4), and perhaps most importantly reduces the size of
the active site by forcing the phenolate units forwards;
the O–Zr–O0 angles for [ZrL9Cl2] and ½ZrL211Cl2ꢁ are
172.2ꢁ and 157.4ꢁ, respectively. The top view of the com-
plexes shows that the phenolate tert-butyl substituents
are positioned directly above the zirconium chloride
sites in the L9 complex, whereas in the L11 complex these
tert-butyl groups are situated above the zirconium cen-
tre. The resultant steric compression in [ZrL9Cl2] forces
the chlorides farther apart (108.4ꢁ) than in the L11 com-
plex (94.4ꢁ). Given these structural differences, the vari-
ance in intrinsic catalytic activity between [ZrL8Cl2] and
½ZrL120Cl2ꢁ is not surprising.
The effect on polymerisation catalysis using the bia-
ryl-bridged complexes with this ligand modification is
significant as we transformed an inactive system to one
that displays activity (albeit moderate) and also demon-
strated that the catalyst is long lived.
Application of this simple ligand modification to the
unbridged salicylaldimine systems did not lead to an in-
crease in polymerisation catalyst lifetime at higher tem-
peratures. At least two explanations are available which
are consistent with observations to date. If imine reac-
tivity is at the heart of the instability of the unbridged
systems, then the lack of success in inhibiting 1,2-migra-
tory insertion by our method might be traced to the dif-
ferences in precatalyst structure detailed in Section 2.13.
If on the other hand, loss of a salicylaldimine ligand (e.g.
via transfer to aluminium from MAO) causes catalyst
death, then this steric modification would not be ex-
pected to make a significant difference. We note the
growing body of evidence for the latter picture, such
as the relatively poor polymerising capabilities of
mono-salicylaldimine complexes [19] and improved sta-
bility of more electron-rich phenolate systems [20].
The variance in response of the two catalysts systems
to attempted steric blocking of the 1,2-migratory inser-
tion reaction also requires comment. For the biaryls,
the geometry is essentially pre-organised for the approach
of metal-coordinated alkyl towards the carbon atom.
Steric compression from phenolic ortho substituents
(top view, Fig. 14) encourages this further. Nevertheless,
the constrained biaryl ligand geometry also dictates that

P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
5135
It remains however that tetradentate systems have
4.2. Synthesis of salicylaldehydes [25]
been shown here to yield a stable catalyst, and if this
along with high activity and ability to polymerise a-ole-
fins in a stereoselective manner is the goal, then attempts
to balance ligand constraint and flexibility might be
fruitful.
The reaction was performed with rigorous exclusion of
moisture. Acetonitrile and triethylamine were dried over
CaH2, paraformaldehyde was dried over P2O5 and
‘‘anhydrous’’ MgCl2 was dried over P2O5 at 120 ꢁC. A
1 l side arm round bottom flask with stirrer bar was
placed under an argon atmosphere and charged with
the appropriate phenol (100 mmol) and dry acetonitrile
(500 ml). To this was added dry triethylamine (52.2 ml,
375 mmol), anhydrous MgCl2 (14.28 g, 150 mmol) and
the solution was stirred for 15 min. Dry paraformalde-
hyde (20.25 g, 675 mmol) was added and a wide bore con-
denser fitted to the round bottom flask. The solution was
heated at reflux temperature under argon for ca. 2.5 h.
The solution was allowed to cool to room temperature
and added to 5% aqueous HCl (800 ml) followed by stir-
ring for 30 min. This was extracted with diethyl ether
(7 · 100 ml portions) and the ether fractions collected to-
gether and washed with saturated NaCl(aq) (3 · 100 ml
portions). The ether layer was dried over anhydrous
MgSO4 followed by filtration. Volatiles were removed
under reduced pressure to yield the corresponding salicyl-
aldehydes, usually contaminated with the starting phe-
nol. Purification prior to reaction with the biaryl
diamine (vide infra) was found to be generally unneces-
sary since the Schiff base product precipitates from the
reaction solution. Samples of the products were neverthe-
less isolated and fully characterised, as described below.
4. Experimental
4.1. General details
Unless stated otherwise, organic preparations were
carried out in air. Organometallic manipulations were
performed under an atmosphere of dry argon, using con-
ventional Schlenk line techniques and an MBraun glove
box (<1 ppm O2/H2O). For organometallic prepara-
tions, hydrocarbon and ether solvents were pre-dried
over sodium wire. These were then dried under reflux
conditions over sodium (for toluene), potassium (THF
and benzene), sodium–potassium alloy (diethyl ether,
petroleum ether and pentane), then distilled and
degassed before use. Dichloromethane and acetonitrile
were dried under reflux conditions over calcium hydride
then distilled and degassed. Deuterated solvents were de-
gassed by the freeze–thaw method and dried over potas-
sium (toluene, benzene and THF) or calcium hydride
(dichloromethane and acetonitrile) before trap-to-trap
distillation and storage in the glove box. NMR samples
of air sensitive species were prepared in the glovebox in
tubes sealed with Youngꢁs concentric stopcocks.
4.2.1. 2-Hydroxy-3,6-dimethylbenzaldehyde
Unless stated otherwise, commercial chemical re-
agents were used as received. Chiral ligands and com-
plexes were racemic mixtures.
Using the general procedure with 2,5-dimethylphenol
(12.21 g), the reaction mixture was heated at reflux tem-
perature for 3 h during which time the solution turned
green. A yellow solid (14.92 g) was obtained, which
was found to contain the target compound (90% by
1H NMR) along with the starting phenol. A sample of
the mixture (2.00 g) was purified by column chromatog-
raphy using hexane:diethyl ether 4:1 as the eluent. Sim-
ilar fractions (TLC analysis) were collected together and
volatiles were removed under reduced pressure to leave a
pale green crystalline solid (1.71 g).
NMR spectra were recorded at ca. 298 K on Bruker
AC-250, DPX-300, DPX-400 or AC-400 spectrometers
and the spectra referenced internally using residual pro-
tio solvent resonances relative to tetramethylsilane
(d = 0.0 ppm). Proton and carbon NMR assignments
were confirmed routinely by 1H–1H (COSY) or
1H–13C (HMQC) experiments. NMR kinetic data were
obtained using Bruker AC-400 or DPX-500 spectrome-
ter with calibrated temperature probes. Infrared spectra
were carried out on sodium chloride plates in an airtight
holder, and obtained as thin film (dichloromethane as
solvent) or nujol mulls on a Perkin–Elmer FT-IR spec-
trometer. EI and CI mass spectra were obtained on a
VG Autospec mass spectrometer. Elemental analyses
were performed by Warwick Analytical Services. For
some alkyl complexes, analyses of %C were found to
be lower than calculated values despite the use of high
combustion temperatures and the addition of combus-
tions aids; this is frequently the case in such systems [21].
The compounds LiCH2CMe3 [22], [Zr(CH2CMe3)4]
[23] and [Zr(CH2Ph)4] [24] were synthesised by literature
methods.
Yield from 2.00 g or crude product = 1.71 g, 84%.
Anal. found (Calculated for C9H10O2)% C 71.57
(71.99), H 6.73 (6.71).
1H NMR 300 MHz (CDCl3) d ppm 12.17 (s, 1H,
ArOH), 10.29 (s, 1H, HC@O), 7.23 (d, 1H, ArH,
3
3JHH = 7 Hz), 6.61 (d, 1H, ArH, JHH = 7 Hz), 2.56 (s,
3H, Me), 2.20 (s, 3H, Me).
13C{1H} NMR 75 MHz (CDCl3) d ppm 195.4, 161.5,
139.3, 138.0, 124.9, 121.1, 117.8, 17.8, 14.8.
IR (CH2Cl2 thin film) m cmꢀ1 2955, 1645 (s, C@O),
1626, 1513, 1462, 1432, 1408, 1376, 1330, 1280 (m, C–
O), 1252, 1236, 1058, 1030, 974, 822, 775, 740, 696.
MS (EI) m/z 150 [M]+, 149 [M ꢀ H]+, 135
[M ꢀ CH3]+.

5136
P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
1
4.2.2. 2-Hydroxy-6-isopropyl-3-methylbenzaldehyde
Using the general procedure with 5-isopropyl-2-
methylphenol (carvacrol) (15.4 ml), the mixture was
heated at reflux temperature for 3.5 h during which time
the solution turned yellow. A yellow/brown oil (17.24 g)
was obtained and found to contain the target compound
by H NMR). A sample of the oil (1.60 g) was purified
by flash column chromatography (hexane:diethyl ether
8:1). Similar fractions (TLC analysis) were collected to-
gether and volatiles were removed under reduced pres-
sure to leave a pale green oil (1.08 g).
Yield from 1.60 g of crude product = 1.08 g, 66%.
Anal. found (Calculated for C11H14O2)% C 74.09
(74.13), H 7.95 (7.92).
1
(75% by H NMR). A sample of the oil (1.00 g) was
purified by column chromatography (hexane:diethyl
ether 4:1). Similar fractions (TLC analysis) were col-
lected together and the solvent was removed under re-
duced pressure to leave a pale yellow/green oil (0.72 g).
Yield from 1.00 g of crude product = 0.72 g, 70%.
Anal. found (Calculated for C11H14O2)% C 74.10
(74.13), H 7.99 (7.92).
1H NMR 300 MHz (CDCl3) d ppm 12.30 (s, 1H,
ArOH), 10.29 (s, 1H, HC@O), 7.31 (d, 1H, ArH,
3JHH = 8 Hz), 6.66 (d, 1H, ArH, JHH = 8 Hz), 3.32
3
3
(m, 1H, CHMe2, JHH = 7 Hz), 2.56 (s, 3H, Me), 1.21
3
(d, 6H, CHMe2, JHH = 7 Hz).
13C{1H} NMR 75 MHz (CDCl3) d ppm 195.5
(HC@O), 160.7, 139.0, 135.1, 133.8, 121.2, 117.8 (Ar),
25.8 (CHMe2), 22.1 (CHMe2), 17.7 (Me).
1H NMR 300 MHz (CDCl3) d ppm 12.42 (s, 1H,
ArOH), 10.40 (s, 1H, HC@O), 7.31 (d, 1H, ArH,
3JHH = 8 Hz), 6.76 (d, 1H, ArH, JHH = 8 Hz), 3.60
IR (thin film) m cmꢀ1 3041 (w, OH), 2963, 2875, 2579,
2361, 2343, 1896, 1821, 1633 (s, C@O), 1584, 1508, 1427,
1382, 1350, 1328, 1302, 1273 (m, C–O), 1248, 1226,
1178, 1151, 1110, 1097, 1068, 1049, 1035, 1000, 967,
923, 866, 821, 774, 735, 684, 660, 605, 595, 549.
MS (EI) m/z 178 [M]+, 163 [M ꢀ CH3]+, 148
[M ꢀ (CH3)2]+.
3
3
(m, 1H, CHMe2, JHH = 7 Hz), 2.20 (s, 3H, Me), 1.31
3
(d, 6H, CHMe2, JHH = 7 Hz).
13C{1H} NMR 75 MHz (CDCl3) d ppm 195.1
(HC@O), 161.7, 150.2, 138.2, 124.3, 116.3, 115.5 (Ar),
27.2 (CHMe2), 24.2 (CHMe2), 14.9 (Me).
IR (CH2Cl2 thin film) m cmꢀ1 2965, 2926, 2874, 1892,
1632 (s, C@O), 1584, 1457, 1425, 1407, 1314, 1266 (m,
C–O), 1256, 1228, 1176, 1155, 1107, 1063, 1034, 976,
752, 870, 823, 787, 716, 693, 654, 545.
4.2.5. Synthesis of 2-hydroxy-3-tert-butyl-6-methyl-
benzaldehyde
MS (EI) m/z 178 [M]+, 177 [M ꢀ H]+, 163
Using the general procedure with 2-tert-butyl-5-
methylphenol (16.43 g), the mixture was heated at reflux
temperature for 1.5 h during which time the solution
turned green/yellow. A yellow/orange gum was obtained
(15.44 g) which was found to contain the target com-
pound (66% by 1H NMR). A sample of the solid
(2.00 g) was purified by flash column chromatography
(hexane:diethyl ether 4:1), to yield a yellow/green oil.
The gum was also purified by distillation at 110 ꢁC,
2 mm Hg, to yield a yellow/green oil.
[M ꢀ CH3]+.
4.2.3. 2-Hydroxy-4,6-dimethylbenzaldehyde
Using the general procedure with 3,5-dimethylphenol
(12.21 g), the mixture was heated at reflux temperature
for 3 h, during which time the solution turned green.
A yellow solid was obtained.
Yield = 14.75 g, 97%.
Anal. found (Calculated for C9H10O2)% C 71.98
(71.98), H 6.62 (6.71).
1H NMR 300 MHz (CDCl3) d ppm 11.93 (s, 1H,
ArOH), 10.19 (s, 1H, HC@O), 6.59 (s, 1H, ArH), 6.50
(s, 1H, ArH), 2.52 (s, 3H, Me), 2.28 (s, 3H, Me).
13C{1H} NMR 75 MHz (CDCl3) d ppm 194.4
(HC@O), 163.2, 149.1, 141.7, 123.0, 116.4, 115.9 (Ar),
22.0, 17.8.
Yield = 8.21 g, 43% (from distillation).
Anal. found (Calculated for C12H16O2)% C 75.11
(74.97), H 8.42 (8.39).
1H NMR 300 MHz (CDCl3) d ppm 12.72 (s, 1H,
ArOH), 10.29 (s, 1H, HC@O), 7.37 (d, 1H, ArH,
3JHH = 8 Hz), 6.64 (dd, 1H, ArH, JHH = 8 Hz,
3
4JHH = 1 Hz), 2.55 (s, 3H, Me), 1.40 (s, 9H, CMe3).
13C{1H} NMR 75 MHz (CDCl3) d ppm 195.7
(HC@O), 162.7, 139.6, 136.3, 134.4, 120.9, 118.2 (Ar),
34.4 (CMe3), 29.1 (CMe3), 17.7 (Me).
IR (CH2Cl2 thin film) m cmꢀ1 3426 (b, OH), 2928,
2305, 1643 (s, C@O), 1571, 1504, 1450, 1377, 1348,
1310, 1292, 1266 (s, C–O), 1237, 1193, 1153, 1063,
1038, 969, 896, 874, 846, 804, 750, 727.
IR (thin film) m cmꢀ1 3473 (b, OH), 3262, 2959, 2772,
2590, 1898, 1765, 1638 (s, C@O), 1578, 1485, 1452, 1427,
1401, 1380, 1362, 1332, 1306, 1285, 1266 (m, C–O),
1241, 1196, 1133, 1083, 1051, 1028, 969, 948, 930, 818,
794, 735, 643, 607, 595.
MS (EI) m/z 150 [M]+, 149 [M ꢀ H]+, 132
[M ꢀ H2O]+.
4.2.4. 2-Hydroxy-3-isopropyl-6-methylbenzaldehyde
Using the general procedure with 2-isopropyl-5-
methylphenol (thymol) (15.02 g), the mixture was heated
at reflux temperature for 3 h during which time the solu-
tion turned yellow. A yellow oil was obtained (17.32 g),
which was found to contain the target compound (70%
MS (EI) m/z 192 [M]+, 177 [M ꢀ CH3]+.
4.2.6. 2-Hydroxy-3-tert-butyl-5-methylbenzaldehyde
Using the general procedure with 2-tert-butyl-4-
methylphenol (16.43 g), the mixture was heated at reflux

P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
5137
temperature for 3 h during which time the solution
turned orange/yellow. A yellow solid was obtained and
was found to contain the target compound (67% purity
128.5, 128.3, 124.1, 119.8, 116.3, 116.2 (Ar), 19.9 (Me),
18.6 (Me), 15.4 (Me).
IR (CH2Cl2 thin film) m cmꢀ1 3428 (b, OH), 2923,
1605 (s, N@C), 1569, 1454, 1379, 1359, 1283 (m, C–
O), 1253, 1224, 1056, 1034, 985, 938, 804, 759.
MS (EI) m/z 476 [M]+.
1
by H NMR). The solid was dissolved in pentane, con-
centrated and stored at ꢀ30 ꢁC overnight to yield pale
green crystals which were isolated by vacuum filtration.
These were found to be the pure aldehyde. Further crops
were obtained by concentration of the supernatant.
Yield = 9.70 g, 50%.
4.3.2. H2L5
Following the general procedure, 2-hydroxy-3-
methyl-6-isopropylbenzaldehyde (1.10 g, 6.17 mmol),
2,20-diamino-6,60-dimethylbiphenyl (0.655 g, 3.09 mmol)
and ethanol (40 ml) were used. A yellow solid was
obtained.
Anal. found (Calculated for C12H16O2)% C 75.08
(74.97), 8.46 (8.39).
1H NMR 300 MHz (CDCl3) d ppm 11.61 (s, 1H,
ArOH), 9.81 (s, 1H, HC@O), 7.33 (d, 1H, ArH,
4
4JHH = 2 Hz), 7.17 (d, 1H, ArH, JHH = 2 Hz), 2.32 (s,
Yield = 1.37 g, 84%.
Anal. found (Calculated for C36H40N2O2)% C 80.83
(81.17), H 7.53 (7.57), 5.21 (5.26).
3H, Me), 1.41 (s, 9H, CMe3).
13C{1H} NMR 75 MHz (CDCl3) d ppm 196.9
(HC@O), 159.0, 137.8, 135.3, 131.3, 128.0, 120.2 (Ar),
34.6 (CMe3), 29.1 (CMe3), 20.4 (Me).
1H NMR 300 MHz (CDCl3) d ppm 13.47 (s, 2H,
ArOH), 8.89 (s, 2H, N@CH), 7.34 (t, 2H, ArH,
IR (CH2Cl2 thin film) m cmꢀ1 3357 (b, OH), 3002,
2964, 2868, 1649 (s, C@O), 1599, 1514, 1461, 1392,
1356, 1323, 1266 (m, C–O), 1232, 1197, 1161, 1020,
972, 932, 866, 793, 768, 747, 710.
3JHH = 8 Hz), 7.22 (d, 2H, ArH, JHH = 7 Hz), 7.07 (d,
3
2H, ArH, 3JHH = 8 Hz), 7.01 (d, 2H, ArH,
3
3JHH = 7 Hz), 6.61 (d, 2H, ArH, JHH = 8 Hz), 3.28
3
(m, 2H, CHMe2, JHH = 7 Hz), 2.09 (s, 6H, Me), 2.06
MS (EI) m/z 192 [M]+, 177 [M ꢀ CH3]+, 163
(s, 6H, Me), 1.21 (d, 6H, CHMe2, JHH = 7 Hz), 1.12
3
3
[M ꢀ CHO]+.
(d, 6H, CHMe2, JHH = 7 Hz).
13C{1H} NMR 75 MHz (CDCl3) d ppm 160.5, 160.0
(N@CH), 147.8, 147.6, 137.0, 134.0, 132.9, 128.5, 128.3,
123.6, 116.2, 114.9, 114.2 (Ar), 27.9 (CHMe2), 24.2
(CHMe2), 23.7 (CHMe2), 19.9 (Me), 15.5 (Me).
IR (CH2Cl2 thin film) m cmꢀ1 2955, 2867, 2295, 1603
(s, N@C), 1568, 1454, 1428, 1378, 1364, 1299, 1261 (m,
C–O), 1225, 1171, 1107, 1042, 983, 941, 847, 817, 798,
780, 762, 737, 649.
4.3. Synthesis of Schiff base proligands
A 100 ml round bottom flask with stirrer bar was
charged with the appropriate salicylaldehyde and a
stoichiometric amount of the appropriate amine (race-
mic 2,20-diamino-6,60-dimethylbiphenyl for H2L4ꢀ9
and aniline for HL10–13). The reactants were dissolved
in ethanol or methanol (40–100 ml), and heated at re-
flux for ca. 18 h, using a condenser fitted with a dry-
ing tube containing CaCl2. A precipitate generally
formed during the reaction and the solid was isolated
by vacuum filtration and washed with cold ethanol or
methanol. All remaining volatiles were removed in
vacuo.
MS (EI) m/z 532 [M]+, 517 [M ꢀ CH3]+.
4.3.3. H2L6
Following the general procedure, 2-hydroxy-4,6-dim-
ethylbenzaldehyde (5.00 g, 33.33 mmol), 2,20-diamino-
6,60-dimethylbiphenyl (3.53 g, 16.65 mmol) and ethanol
(100 ml) were used. An orange solid was obtained.
Yield = 6.82 g, 86%.
4.3.1. H2L4
Following the general procedure, 2-hydroxy-3,6-di-
Anal. found (Calculated for C32H32N2O2)% C 79.17
(80.64), H 6.87 (6.77), N 5.52 (5.88).
methylbenzaldehyde
(2.00 g,
13.33 mmol),
2,20-
1H NMR 300 MHz (CDCl3) d ppm 13.16 (s, 2H,
ArOH), 8.71 (s, 2H, N@CH), 7.38 (t, 2H, ArH,
diamino-6,60-dimethylbiphenyl (1.41 g, 6.65 mmol) and
ethanol (40 ml) were used. An orange solid was obtained.
Yield = 2.44 g, 77%.
3
3JHH = 8 Hz), 7.25 (d, 2H, ArH, JHH = 7 Hz), 7.08 (d,
3
2H, ArH, JHH = 8 Hz), 6.50 (s, 2H, ArH), 6.41 (s,
Anal. found (Calculated for C32H32N2O2)% C 80.69
(80.64), H 6.80 (6.77), N 6.00 (5.88).
2H, ArH), 2.27 (s, 6H, Me), 2.21 (s, 6H, Me), 2.08 (s,
6H, Me).
13C{1H} NMR 75 MHz (CDCl3) d ppm 162.0, 159.1
(N@CH), 147.2, 143.9, 139.2, 137.0, 133.1, 128.4, 128.2,
121.7, 115.7, 115.3, 114.6 (Ar), 21.6 (Me), 19.7 (Me),
18.6 (Me).
IR (CH2Cl2 thin film) m cmꢀ1 3428 (b, OH), 2360,
1608 (s, N@C), 1558, 1453, 1356, 1302, 1268, 1229 (s,
C–O), 1202, 1152, 1062, 1017, 971, 940, 836, 783, 736.
MS (EI) m/z 476 [M]+.
1H NMR 300 MHz (CDCl3) d ppm 13.24 (s, 2H,
ArOH), 8.68 (s, 2H, N@CH), 7.33 (t, 2H, ArH,
3
3JHH = 8 Hz), 7.22 (d, 2H, ArH, JHH = 7 Hz), 6.99 (d,
2H, ArH), 6.99 (d, 2H, ArH), 6.47 (d, 2H, ArH,
3JHH = 8 Hz), 2.26 (s, 6H, Me), 2.09 (s, 6H, Me), 2.07
(s, 6H, Me).
13C{1H} NMR 75 MHz (CDCl3) d ppm 160.5
(N@CH), 160.4, 147.9, 137.0, 137.0, 133.9, 132.7,

5138
P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
4.3.4. H2L7
diamino-6,60-dimethylbiphenyl (1.08 g, 5.09 mmol) and
ethanol (40 ml) were used. An orange solid was
obtained.
Following the general procedure, 2-hydroxy-3-iso-
propyl-6-methylbenzaldehyde (1.55 g, 8.65 mmol), 2,20-
diamino-6,60-dimethylbiphenyl (0.91 g, 4.29 mmol) and
ethanol (40 ml) were used. An orange solid was
obtained.
Yield = 2.52 g, 88%.
Anal. found (Calculated for C38H44N2O2)% C 81.33
(81.39), H 7.97 (7.91), N 5.04 (5.00).
Yield = 1.62 g, 71%.
Anal. found (Calculated for C36H40N2O2)% C 81.13
(81.17), H 7.68 (7.57), N 4.96 (5.26).
1H NMR 300 MHz (CDCl3) d ppm 12.84 (s, 2H,
ArOH), 8.37 (s, 2H, N@CH), 7.32 (t, 2H, ArH,
3
3JHH = 8 Hz), 7.21 (d, 2H, ArH, JHH = 8 Hz), 7.07 (d,
1H NMR 300 MHz (CDCl3) d ppm 13.40 (s, 2H,
ArOH), 8.69 (s, 2H, N@CH), 7.35 (t, 2H, ArH,
4
2H, ArH, JHH = 2 Hz), 7.02 (2H, ArH, JHH = 8 Hz),
3
4
6.85 (d, 2H, ArH, JHH = 2 Hz), 2.23 (s, 6H, Me), 2.09
3JHH = 8 Hz), 7.24 (d, 2H, ArH, JHH = 7 Hz), 7.09 (d,
(s, 6H, Me), 1.32 (s, 18H, CMe3).
3
2H, ArH, 3JHH = 8 Hz), 7.01 (d, 2H, ArH,
13C{1H} NMR 75 MHz (CDCl3) d ppm 162.1
(N@CH), 158.2, 146.9, 137.0, 136.9, 133.2, 130.8,
130.0, 128.2, 128.1, 126.2, 118.5, 115.1 (Ar), 34.6
(CMe3), 29.1 (CMe3), 20.5 (Me), 19.7 (Me).
IR (CH2Cl2 thin film) m cmꢀ1 3357 (b, OH), 2957,
2916, 2864, 2740, 1619 (s, N@C), 1568, 1462, 1438,
1390, 1384, 1359, 1318, 1266 (w, C–O), 1230, 1210,
1163, 1104, 1024, 980, 948, 927, 861, 813, 786, 772,
746, 688.
3JHH = 8 Hz), 6.55 (d, 2H, ArH, JHH = 8 Hz), 3.22
3
3
(m, 2H, CHMe2, JHH = 7 Hz), 2.25 (s, 6H, Me), 2.11
3
(s, 6H, Me), 1.16 (d, 6H, CHMe2, JHH = 7 Hz), 1.15
3
(d, 6H, CHMe2, JHH = 7 Hz).
13C{1H} NMR 75 MHz (CDCl3) d ppm 160.5
(N@CH), 159.5, 147.9, 136.9, 136.7, 134.4, 132.6,
129.3, 128.4, 128.1, 119.8, 116.3, 116.0 (Ar), 26.0
(CHMe2), 22.3 (CHMe2), 22.1 (CHMe2), 19.8 (Me),
18.4 (Me).
MS (EI) m/z 560 [M]+, 545 [M ꢀ CH3]+.
4.3.7. HL10
Following the general procedure, 2-hydroxy-3-tert-
butyl-6-methylbenzaldehyde (1.50 g, 7.81 mmol), aniline
(0.73 g, 7.85 mmol) and methanol (40 ml) were used.
The solution was kept at ꢀ30 ꢁC before filtration and
an orange solid was obtained.
IR (CH2Cl2 thin film) m cmꢀ1 3476 (b, OH), 2960,
2870, 1602 (s, N@C), 1569, 1438, 1380, 1360, 1300,
1274 (m, C–O), 1247, 1219, 1152, 1108, 1067, 1048,
987, 941, 813, 771, 740.
MS (EI) m/z 532 [M]+.
4.3.5. H2L8
Following the general procedure, 2-hydroxy-3-tert-
butyl-6-methylbenzaldehyde (1.36 g, 7.08 mmol), 2,20-
diamino-6,60-dimethylbiphenyl (0.74 g, 3.49 mmol) and
ethanol (40 ml) were used. An orange solid was obtained.
Yield = 1.56 g, 80%.
Yield = 1.75 g, 84%.
Anal. found (Calculated for C18H21NO)% C 80.81
(80.86), H 7.99 (7.92), N 5.24 (5.24).
1H NMR 300 MHz (CDCl3) d ppm 14.71 (s, 1H,
ArOH), 8.98 (s, 1H, N@CH), 7.44 (t, 2H, ArH,
3JHH = 8 Hz), 7.31–7.27 (m, 4H, ArH), 6.65 (d, 1H,
Anal. found (Calculated for C38H44N2O2)% C 81.38
(81.39), H 7.87 (7.91), N 4.99 (5.00).
3
ArH, JHH = 8 Hz), 2.51 (s, 3H, Me), 1.47 (s, 9H,
1H NMR 300 MHz (CDCl3) d ppm 13.87 (s, 2H,
ArOH), 8.71 (s, 2H, N@CH), 7.34 (t, 2H, ArH,
CMe3).
13C{1H} NMR 75 MHz (CDCl3) d ppm 162.2, 161.0
(N@CH), 148.7, 137.3, 135.8, 130.5, 129.3, 126.5, 121.2,
120.0, 116.8 (Ar), 34.6 (CMe3), 29.3 (CMe3), 18.8 (Me).
IR (CH2Cl2 thin film) m cmꢀ1 3383 (b, OH), 2999, 2941,
2552 (b, OH), 1606 (s, N@C), 1580, 1481, 1457, 1434,
1384, 1360, 1301, 1286 (m, C–O), 1190, 1137, 1078,
1025, 978, 966, 927, 904, 851, 813, 777, 758, 731, 693.
MS (EI) m/z 267 [M]+, 252 [M ꢀ CH3]+, 250
[M ꢀ OH]+.
3
3JHH = 8 Hz), 7.23 (d, 2H, ArH, JHH = 7 Hz), 7.12 (d,
2H, ArH, 3JHH = 8 Hz), 7.04 (d, 2H, ArH,
3
3JHH = 8 Hz), 6.48 (dd, 2H, ArH, JHH = 8 Hz,
4JHH = 1 Hz), 2.23 (s, 6H, Me), 2.11 (s, 6H, Me), 1.30
(s, 18H, CMe3).
13C{1H} NMR 75 MHz (CDCl3) d ppm 161.6, 159.9
(N@CH), 147.3, 137.2, 137.0, 135.5, 133.0, 129.9, 128.3,
128.2, 119.3, 116.7, 115.4 (Ar), 34.4 (CMe3), 29.1
(CMe3), 19.8 (Me), 18.4 (Me).
IR (CH2Cl2 thin film) m cmꢀ1 3426 (b, OH), 2957,
2871, 2361, 2094, 1643, 1601 (s, N@C), 1568, 1484,
1436, 1361, 1288, 1266 (m, C–O), 1236, 1197, 1136,
1051, 1018, 976, 942, 813, 792, 770, 739.
MS (EI) m/z 560 [M]+, 545 [M ꢀ CH3]+.
4.3.6. H2L9
4.3.8. Synthesis of HL11 [16]
Following the general procedure, 3-tert-butyl-2-hy-
droxy-5-methylbenzaldehyde (1.87 g, 9.74 mmol), ani-
line (0.91 g, 9.78 mmol) and methanol (40 ml) were
used. The solution was kept at ꢀ30 ꢁC before filtration
and an orange solid was obtained.
Yield = 2.05 g, 79%.
Anal. found (Calculated for C18H21NO)% C 80.89
(80.86), H 7.94 (7.92), N 5.27 (5.24).
Following the general procedure, 2-hydroxy-3-tert-
butyl-5-methylbenzaldehyde (2.00 g, 10.40 mmol), 2,20-

P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
5139
1H NMR 300 MHz (CDCl3) d ppm 13.69 (s, 1H,
ArOH), 8.56 (s, 1H, N@CH), 7.40 (t, 2H, ArH,
IR (thin film) m cmꢀ1 3030, 2982, 2919, 2771, 1944,
1752, 1621 (s, N@C), 1585, 1490, 1474, 1435, 1380,
1364, 1324, 1280, 1267 (s, C–O), 1205, 1167, 1076, 1049,
1025, 976, 961, 906, 862, 810, 781, 766, 747, 708, 692.
MS (EI) m/z 225 [M]+, 210 [M ꢀ CH3]+.
3
3JHH = 8 Hz), 7.26 (d, 2H, ArH, JHH = 8 Hz), 7.25 (t,
1H, ArH), 7.19 (d, 1H, ArH, 4JHH = 2 Hz), 7.03 (d, 1H,
ArH, 4JHH = 2 Hz), 2.30 (s, 3H, Me), 1.46 (s, 9H, CMe3).
13C{1H} NMR 75 MHz (CDCl3) d ppm 163.2
(N@CH), 158.2, 148.5, 137.3, 131.4, 130.3, 129.2,
126.9, 126.5, 121.0, 118.6 (Ar), 34.7 (CMe3), 29.3
(CMe3), 20.6 (Me).
4.4. Na2Ln ÆTHFx and NaLn ÆTHFx
A Schlenk vessel with stirrer bar was charged with the
appropriate proligand (0.5–1.5 g), and a twofold excess
of NaHÆTHF (40 ml) was added and the solution stirred
overnight under a vented inert atmosphere. The stirring
was stopped and unreacted NaH allowed to settle before
filtering the solution through a cannula. THF was re-
moved in vacuo to yield a solid. The product was ana-
IR (CH2Cl2 thin film) m cmꢀ1 2957, 2913, 2760, 1941,
1764, 1727, 1618 (s, N@C), 1581, 1485, 1440, 1390, 1361,
1320, 1267 (s, C–O), 1236, 1206, 1164, 1076, 1027, 981,
930, 903, 862, 801, 782, 759, 691.
MS (EI) m/z 267 [M]+, 252 [M ꢀ CH3]+, 250
[M ꢀ OH]+.
1
lysed by H NMR spectroscopy in d5-pyridine and the
4.3.9. HL12
THF content thereby estimated.
Following the general procedure, 2-hydroxy-3,6-
dimethylbenzaldehyde (0.53 g, 3.53 mmol), aniline
(0.33 g, 3.55 mmol) and methanol (40 ml) were used.
The solution was kept at ꢀ30 ꢁC before filtration and
an orange solid was obtained.
4.5. Zirconium alkyl complexes
4.5.1. [ZrL4(CH2Ph)2]
A Schlenk vessel with stirrer bar was charged with
H2L4 (247 mg, 0.52 mmol) and [Zr(CH2Ph)2] (250 mg,
0.55 mmol). The vessel was placed in a dry ice/acetone
bath at ꢀ78 ꢁC and dichloromethane (30 ml) was added.
This was stirred for 30 min and then filtered at ꢀ78 ꢁC,
followed by removal of solvent under reduced pressure.
The orange solid obtained was then washed with aceto-
nitrile, filtered and dried in vacuo.
Yield = 0.72 g, 91%.
Anal. found (Calculated for C15H15NO)% C 79.47
(79.97), H 6.73 (6.71), N 6.04 (6.22).
1H NMR 300 MHz (CDCl3) d ppm 14.28 (s, 1H,
ArOH), 8.85 (s, 1H, N@CH), 7.32 (t, 2H, ArH,
3JHH = 8 Hz), 7.18 (t, 1H, ArH), 7.17 (d, 2H, ArH),
3
7.02 (d, 1H, ArH, JHH = 7 Hz), 6.51 (d, 1H, ArHc,
3JHH = 8 Hz), 2.39 (s, 3H, Me), 2.18 (s, 3H, Me).
13C{1H} NMR 75 MHz (CDCl3) d ppm 160.5, 160.5
(N@CH), 148.6, 136.9, 134.2, 129.3, 126.5, 124.3, 121.1,
120.2, 116.2 (Ar), 18.7 (Me), 15.3 (Me).
IR (CH2Cl2 thin film) m cmꢀ1 3033, 3004, 2949, 2924,
2734, 2672, 1940, 1867, 1609 (s, N@C), 1578, 1503, 1484,
1459, 1434, 1380, 1362, 1308, 1283 (s, C–O), 1253, 1201,
1171, 1157, 1076, 1054, 1034, 986, 943, 905, 852, 806,
770, 754, 738, 714, 692, 593, 559.
MS (EI) m/z 225 [M]+, 208 [M ꢀ OH]+.
4.3.10. HL13
Yield = 266 mg, 68%.
Anal. found (Calculated for C46H44N2O2Zr)% C
73.56 (73.86), H 5.60 (5.93), N 3.85 (3.74).
1H NMR 400 MHz (CD2Cl2) d ppm 8.14 (s, 2H,
3
N@CH), 7.24 (d, 2H, ArH, JHH = 8 Hz), 7.06 (t, 2H,
ArH, JHH = 8 Hz), 7.02 (d, 2H, ArH, JHH = 7 Hz),
3
3
3
6.96 (t, 4H, ArH, JHH = 8 Hz), 6.79 (t, 2H, ArH,
3
3JHH = 7 Hz), 6.64 (d, 4H, ArH, JHH = 7 Hz), 6.55 (d,
2H, ArH, 3JHH = 7 Hz), 6.53 (d, 2H, ArH,
3JHH = 7 Hz), 2.34 (s, 6H, Me), 2.25 (s, 6H, Me), 2.09
2
(s, 6H, Me), 1.89 (d, 2H, CH2Ph, JHH = 8 Hz), 1.61
2
(d, 2H, CH2Ph, JHH = 8 Hz).
Following the general procedure, 2-hydroxy-3,5-dim-
ethylbenzaldehyde (0.978 g, 6.52 mmol), aniline (0.61 g,
6.56 mmol) and methanol (40 ml) were used. Evapora-
tion of volatiles gave an orange oil.
Yield = 1.28 g, 87%.
Anal. found (Calculated for C15H15NO)% C 79.88
(79.97), H 6.73 (6.71), N 6.10 (6.22).
1H NMR 300 MHz (CDCl3) d ppm 13.37 (s, 1H,
ArOH), 8.56 (s, 1H, N@CH), 7.43 (t, 2H, ArH,
3JHH = 8 Hz), 7.28 (t, 1H, ArH), 7.28 (d, 2H, ArH),
7.09 (s,1H, ArH), 7.03 (s, 1H, ArH), 2.32 (s, 3H, Me),
2.31 (s, 3H, Me).
13C{1H} NMR 75 MHz (CDCl3) d ppm 162.6
(N@CH), 157.1, 148.5, 135.1, 129.7, 129.2, 127.4,
126.5, 125.7, 121.0, 117.9 (Ar), 20.2 (Me), 15.3 (Me).
13C{1H} NMR 100 MHz (CD2Cl2) d ppm 165.7
(N@CH), 161.9, 150.4, 144.1, 138.8, 136.7, 136.0,
131.3, 128.4, 128.2, 127.9, 127.7, 126.0, 120.9, 120.0,
120.0, 120.0 (Ar), 60.3 (C2Ph), 19.8 (Me), 19.2 (Me),
16.9 (Me).
IR (Nujol) m cmꢀ1 2727, 2284, 1600 (m, N@C), 1562,
1297 (m, C–O), 1249, 1219, 1140, 1062, 1027, 993, 968,
940, 915, 876, 815, 748, 720, 697, 646.
MS (CI) m/z 747 [M ꢀ H]+, 733 [M ꢀ CH3]+, 718
[M ꢀ (CH3)2]+, 657 [M ꢀ CH2Ph]+.
4.5.2. [ZrL5(CH2Ph)2]
A Schlenk vessel with stirrer bar was charged with
H2L5 (449 mg, 0.84 mmol) and [Zr(CH2Ph)4] (402 mg,
0.88 mmol). The vessel was placed in an acetone/dry

5140
P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
ice bath at ꢀ78 ꢁC and acetonitrile (20 ml) added. This
was transferred to an ice bath at 0 ꢁC and allowed to stir
for 20 min. An orange precipitate formed and was iso-
lated by filtration at 0 ꢁC, followed by washing with cold
acetonitrile. The orange solid was dried in vacuo.
Yield = 486 mg, 72%.
4.5.4. [ZrL7(CH2Ph)2]
The complex was synthesised in an analogous manner
to [ZrL5(CH2Ph)2], using H2L7 (366 mg, 0.69 mmol) and
[Zr(CH2Ph)4] (313 mg, 0.69 mmol). The complex was
obtained as an orange solid.
Yield = 396 mg, 71%.
Anal. found (Calculated for C50H52N2O2Zr)% C
72.28 (74.68), H 6.44 (6.52), N 3.34 (3.48).
Anal. found (Calculated for C48H48N2O2Zr)% C
72.74 (74.28), H 6.47 (6.23), N 3.38 (3.61).
1H NMR 500 MHz (CD2Cl2) d ppm 8.06 (s, 2H,
1H NMR 400 MHz (CD2Cl2) d ppm 8.26 (s, 2H,
3
3
N@CH), 7.33 (d, 2H, ArH, JHH = 8 Hz), 7.06 (t, 2H,
ArH, JHH = 8 Hz), 7.01 (d, 2H, ArH, JHH = 7 Hz),
N@CH), 7.26 (d, 2H, ArH, JHH = 8 Hz), 7.03 (t, 2H,
ArH, JHH = 8 Hz), 6.99 (d, 2H, ArH, JHH = 7 Hz),
3
3
3
3
3
6.95 (t, 4H, ArH, JHH = 8 Hz), 6.80 (t, 2H, ArH,
3
6.92 (t, 4H, ArH, JHH = 8 Hz), 6.77 (t, 2H, ArH,
3
3
3JHH = 7 Hz), 6.68 (d, 2H, ArH, JHH = 8 Hz), 6.63 (d,
3JHH = 7 Hz), 6.59 (d, 2H, ArH, JHH = 8 Hz), 6.55 (d,
4H, ArH, 3JHH = 7 Hz), 6.42 (d, 2H, ArH,
2H, ArH, 3JHH = 7 Hz), 6.54 (d, 4H, ArH,
3
3
3JHH = 7 Hz), 3.07 (m, 2H, CHMe2, JHH = 7 Hz), 2.33
(s, 6H, Me), 2.07 (s, 6H, Me), 1.89 (d, 2H, CH2Ph,
3JHH = 8 Hz), 3.57 (m, 2H, CHMe2, JHH = 7 Hz), 2.22
(s, 6H, Me), 2.07 (s, 6H, Me), 1.77 (d, 2H, CH2Ph,
2
2
2JHH = 8 Hz), 1.49 (d, 2H, CH2Ph, JHH = 8 Hz), 1.26
2JHH = 8 Hz), 1.51 (d, 2H, CH2Ph, JHH = 8 Hz), 1.36
3
(d, 2H, CHMe2, JHH = 7 Hz), 1.11 (d, 2H, CHMe2,
3
(d, 6H, CHMe2, JHH = 7 Hz), 1.16 (d, 6H, CHMe2,
3JHH = 7 Hz).
3JHH = 7 Hz).
13C{1H} 75 MHz NMR (CDCl3) d 165.4 (N@CH),
161.9, 150.9, 149.6, 144.5, 136.9, 136.4, 131.3, 128.6,
128.3, 128.1, 128.0, 125.7, 121.2, 120.3, 118.5, 114.9
(Ar), 60.4 (CH2Ph), 28.7 (CHMe2), 24.5 (CHMe2),
23.8 (CHMe2), 20.0 (Me), 17.3 (Me).
IR (Nujol) m cmꢀ1 2725, 2359, 1593 (s, N@C), 1580,
1559, 1405, 1316, 1304, 1278 (m, C–O), 1256, 1216,
1176, 1157, 1102, 1044, 1026, 997, 970, 939, 911, 872,
816, 801, 778, 761, 754, 742, 719, 696, 647.
MS (EI) m/z 622 [M+ ꢀ (CH2Ph)2].
13C{1H} NMR 100 MHz (CD2Cl2) d ppm 165.6
(N@CH), 160.4, 150.5, 144.0, 138.6, 136.8, 136.4,
131.5, 131.5, 128.4, 128.0, 127.8, 127.5, 120.9, 120.4,
120.2, 120.0 (Ar), 60.1 (C2Ph), 26.2 (CHMe2), 23.9
(CHMe2), 21.9 (CHMe2), 19.9 (Me), 19.2 (Me).
IR (Nujol) m cmꢀ1 2726, 2362, 1600 (s, N@C), 1559,
1406, 1365, 1353, 1307, 1282 (m, C–O), 1242, 1210,
1171, 1149, 1100, 1056, 1027, 1000, 964, 940, 901, 850,
810, 780, 742, 697, 648, 611.
MS (EI) m/z 803 [M]+, 621 [M ꢀ (CH2Ph)2]+.
4.5.3. [ZrL6(CH2Ph)2]
4.5.5. [ZrL120ðCH2PhÞ ]
2
The complex was synthesised analogously to
[ZrL5(CH2Ph)2] using H2L6 (279 mg, 0.59 mmol) and
[Zr(CH2Ph)4] (275 mg, 0.60 mmol). A pale orange solid
was obtained.
The complex was synthesised in an analogous manner
to [ZrL4(CH2Ph)2] using HL10 (450 mg, 1.68 mol) and
[Zr(CH2Ph)4] (383 mg, 0.84 mol). An orange/red solid
was obtained.
Yield = 274 mg, 62%.
Yield = 360 mg, 53%.
Anal. found (Calculated for C44H40N2O2Zr)% C
71.30 (73.40), 5.49 (5.60), N 3.39 (3.89).
1H NMR 400 MHz (CD2Cl2) d ppm 8.15 (s, 2H,
Anal. found (Calculated for C50H54N2O2Zr)% C
73.99 (74.49), H 6.52 (6.75), N 3.02 (3.47).
1H NMR 400 MHz (d6-benzene) d ppm 8.33 (s, 2H,
3
3
N@CH), 7.05 (t, 2H, ArH, JHH = 8 Hz), 7.03 (d, 2H,
ArH, JHH = 8 Hz), 7.00 (t, 4H, ArH, JHH = 8 Hz),
N@CH), 7.35 (d, 2H, ArH, JHH = 8 Hz), 7.13 (d, 4H,
ArH, JHH = 8 Hz), 7.05 (t, 4H, ArH, JHH = 7 Hz),
3
3
3
3
3
6.84 (t, 2H, ArH, JHH = 7 Hz), 6.64 (d, 2H, ArH,
3
7.03 (t, 2H, ArH, JHH = 8 Hz), 6.83 (t, 4H, ArH,
3JHH = 7 Hz), 6.52 (s, 2H, ArH), 6.49 (s, 2H, ArH),
3JHH = 7 Hz), 6.78 (t, 2H, ArH, JHH = 7 Hz), 6.74 (d,
3
3
4
6.28 (dd, 2H, ArH, JHH = 7 Hz, JHH = 2 Hz), 2.33 (s,
6H, Me), 2.26 (s, 6H, Me), 2.05 (s, 6H, Me), 1.68 (d,
4H, ArH, 3JHH = 7 Hz), 6.52 (d, 2H, ArH,
2
3JHH = 8 Hz), 2.98 (d, 2H, CH2Ph, JHH = 10 Hz), 2.88
2
2H, CH2Ph, JHH = 8 Hz), 1.27 (d, 2H, CH2Ph,
2
(d, 2H, CH2Ph, JHH = 10 Hz), 1.93 (s, 6H, Me), 1.51
2JHH = 8 Hz).
(s, 18H, CMe3).
13C{1H} NMR 100 MHz (CD2Cl2) d ppm 164.7
(N@CH), 163.7, 150.1, 147.3, 144.0, 141.1, 136.6,
131.1, 128.8, 128.2, 128.0, 127.9, 122.2, 121.1, 120.2,
118.2, 117.8 (Ar), 59.6 (C2Ph), 21.9 (Me), 19.8 (Me),
19.3 (Me).
13C{1H} NMR 100 MHz (d6-benzene) d ppm 167.0
(N@CH), 161.8, 153.3, 147.4, 139.9, 137.8, 133.3,
128.7, 128.2, 127.5, 126.3, 123.2, 123.0, 121.3, 120.7
(Ar), 69.2 (C2Ph), 35.0 (CMe3), 30.0 (CMe3), 19.1
(Me).
IR (Nujol) m cmꢀ1 2360, 2341, 1580 (m, N@C), 1558,
1487, 1404, 1298 (s, C–O), 1266, 1205, 1186, 1136, 1056,
1028, 979, 964, 907, 852, 815, 768, 744, 727, 697, 668,
639.
IR (Nujol) m cmꢀ1 2726, 2361, 1592 (m, N@C), 1533,
1410, 1348, 1308, 1228, 1203 (w, C–O), 1161, 1073, 1028,
976, 938, 837, 782, 722, 696, 647.
MS (EI): did not produce assignable peaks.

P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
5141
MS (EI) m/z 728 [M ꢀ C6H5]+, 714 [M ꢀ CH2Ph]+,
4.6.1. [ZrL1Cl2]
651 [M ꢀ (C6H5)2]+, 623 [M ꢀ (CH2Ph)2]+.
Following the general method, Na2L1 ÆTHF0.48
(700 mg, 1.26 mmol) and [ZrCl4(THF)2] (487 mg,
1.29 mmol) were used.
4.5.6. [ZrL4(CH2But)2]
A Schlenk vessel with stirrer bar was charged with
H2L4 (360 mg, 0.76 mmol) and [Zr(CH2CMe3)4]
(284 mg, 0.76 mmol). This was cooled to ꢀ78 ꢁC and
dichloromethane (30 ml) added, with stirring of the
solution. The solution was allowed to warm to room
temperature and stirring was continued for a further
30 min, during which time the solution turned from yel-
low to orange. Removal of volatiles under reduced pres-
sure yielded an orange/yellow solid. This was dissolved
in pentane and the solution was filtered, concentrated
and kept at ꢀ30 ꢁC overnight to yield a yellow precipi-
tate, which was isolated by filtration and dried in vacuo.
Yield = 226 mg, 42%.
Yield = 722 mg, 90%.
Anal. found (Calculated for C32H30Cl2N2O2Zr)% C
59.99 (60.36), H 4.72 (4.75), N 4.26 (4.40).
1H NMR 300 MHz (CD2Cl2) d ppm 8.21 (s, 2H,
N@CH), 7.30 (s, 2H, ArH), 7.27 (t, 2H, ArH,
3JHH = 8 Hz), 7.20 (d, 2H, ArHj, JHH = 7 Hz), 7.07 (s,
3
3
2H, ArH), 6.98 (d, 2H, ArHl, JHH = 7 Hz), 2.29 (s,
6H, Me), 2.26 (s, 6H, Me), 2.05 (s, 6H, Me).
13C{1H} NMR 75 MHz (CD2Cl2) d ppm 169.7
(N@CH), 158.0, 149.3, 139.4, 137.9, 132.0, 130.7,
129.7, 129.7, 129.2, 128.8, 127.3, 122.1 (Ar), 20.2 (Me),
19.8 (Me), 15.8 (Me).
IR (Nujol) m cmꢀ1 2729, 1614 (s, N@C), 1593, 1556,
1288, 1263 (s, C–O), 1218, 1171, 1160, 1105, 1063,
1016, 980, 961, 942, 902, 860, 851, 836, 817, 791, 777,
756, 736, 627.
Anal. found (Calculated for C42H52N2O2Zr)% C
68.98 (71.24), H 6.99 (7.40), N 4.07 (3.96).
1H NMR 400 MHz (d6-benzene) d ppm 8.18 (s, 2H,
3
N@CH), 7.15 (d, 2H, ArH, JHH = 8 Hz), 7.07 (d, 2H,
ArH, JHH = 7 Hz), 6.77 (t, 2H, ArH, JHH = 8 Hz),
MS (EI) m/z 635 [M]+, 600 [M ꢀ Cl]+, 585 [M ꢀ (Cl
3
3
and Me)]+, 530 [M ꢀ {(Cl)2 and Me}]+.
3
6.49 (d, 2H, ArH, JHH = 8 Hz), 6.32 (d, 2H, ArH,
3JHH = 7 Hz), 2.62 (s, 6H, Me), 1.91 (s, 6H, Me), 1.89
4.6.2. [ZrL2Cl2]
2
(s, 6H, Me), 1.88 (d, 2H, CH2CMe3, JHH = 13 Hz),
Following the general method, Na2L2 ÆTHF0.88
(800 mg, 1.25 mmol) and [ZrCl4(THF)2] (523 mg,
1.38 mmol) were used.
2
1.59 (d, 2H, CH2CMe3, JHH = 13 Hz), 1.37 (s, 18H,
CH2CMe3).
13C{1H} NMR 100 MHz (d6-benzene) d ppm 168.1
(N@CH), 162.7, 151.4, 138.0, 136.8, 136.7, 132.1,
128.1, 127.6, 127.1, 120.2, 120.0, 120.0 (Ar), 87.2
(C2CMe3), 35.6 (CH2CMe3), 35.1 (CH2 CMe3), 20.0
(Me), 18.8 (Me), 17.2 (Me).
Yield = 596 mg, 71%.
Anal. found (Calculated for C34H34Cl2N2O2Zr)% C
61.42 (61.43), H 4.97 (5.16), N 3.86 (4.21).
1H NMR 300 MHz (CD2Cl2) d ppm 8.22 (s, 2H,
N@CH), 7.51 (dd, 2H, ArH, 3JHH = 7 Hz, 4JHH = 2 Hz),
IR (Nujol) m cmꢀ1 2728, 2360, 1602 (s, N@C), 1561,
1400, 1322, 1302 (s, C–O), 1258, 1216, 1143, 1096,
1062, 1041, 993, 975, 938, 891, 803, 767, 747, 720, 669,
646, 600, 554.
7.26 (dd, 2H, ArH), 7.26 (t, 2H, ArH, JHH = 8 Hz),
3
7.19 (d, 2H, ArH, JHH = 7 Hz), 7.03 (d, 2H, ArH,
3
3JHH = 7 Hz), 6.96 (t, 2H, ArH, JHH = 8 Hz), 3.37 (m,
3
3
2H, CHMe2, JHH = 7 Hz), 2.07 (s, 6H, Me), 1.32 (d,
MS (EI) m/z 1414 [M2]+, 1343 [M2 ꢀ CH2CMe3]+,
6H, CHMe2, JHH = 7 Hz), 1.27 (d, 6H, JHH = 7 Hz).
13C{1H} NMR 75 MHz (CD2Cl2) d ppm 170.0
(N@CH), 159.3, 149.1, 138.1, 137.9, 134.1, 132.4,
130.9, 129.2, 128.7, 122.5, 121.8, 120.4 (Ar), 28.1
(CHMe2), 22.1 (CHMe2), 21.9 (CHMe2), 19.8 (Me).
IR (Nujol) m cmꢀ1 1599 (s, N@C), 1553, 1320, 1278
(m, C–O), 1239, 1206, 1150, 1104, 1055, 1017, 984,
943, 891, 857, 806, 770, 751, 738, 710, 653, 579.
MS (EI) m/z 663 [M]+, 628 [M ꢀ Cl]+, 613 [M ꢀ (Cl
and Me)]+, 593 [M ꢀ (Cl)2]+.
3
3
651 [M ꢀ Me2CH2]+, 636 [M ꢀ CH2CMe3]+.
4.6. Synthesis of zirconium chloride complexes
A Schlenk vessel with stirrer bar was charged with the
sodium salt of the appropriate ligand (0.5–1.2 g) and a
slight excess of [ZrCl4(THF)2]. The vessel was cooled
to ꢀ78 ꢁC and THF (40 ml) added. The solution was al-
lowed to warm to room temperature and stirred for
20 h, during which time a precipitate formed and the
solution turned usually turned from yellow to orange.
Stirring was ceased and the precipitate allowed to settle
before filtering the solution through a cannula. The sol-
vent was then removed under reduced pressure to give
an orange/yellow solid. The solid was then dissolved in
dichloromethane and filtered, followed by removal of
the solvent under reduced pressure. The solid obtained
was then sublimated along a horizontal glass tube at
250–350 ꢁC, 10ꢀ6 mm Hg, to yield the product.
4.6.3. [ZrL3Cl2]
Following the general method, Na2L3 ÆTHF0.48
(800 mg, 1.31 mmol) and [ZrCl4(THF)2] (508 mg,
1.35 mmol) were used.
Yield = 634 mg, 70%.
Anal. found (Calculated for C36H38Cl2N2O2Zr)% C
61.72 (62.41), 5.37 (5.53), N 3.88 (4.04).
1H NMR 300 MHz (CD2Cl2) d ppm 8.32 (s, 2H,
N@CH), 7.67 (dd, 2H, ArH, 3JHH = 9 Hz, 4JHH = 3 Hz),

5142
P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
4
7.44 (d, 2H, ArH, JHH = 2 Hz), 7.32 (t, 2H, ArH,
128.7, 125.1, 122.0, 119.6, 116.8 (Ar), 29.2 (CHMe2),
24.4 (CHMe2), 23.1 (CHMe2), 19.8 (Me), 15.8 (Me).
IR (Nujol) m cmꢀ1 2360, 1868, 1583 (s, N@C), 1556,
1404, 1336, 1311, 1258 (s, C–O), 1216, 1156, 1102,
1049, 1019, 996, 939, 897, 875, 836, 817, 793, 778, 761,
754, 739, 727, 686, 657, 625, 571.
3
3JHH = 8 Hz), 7.26 (dd, 2H, ArH, JHH = 8 Hz,
3
4JHH = 1 Hz), 7.00 (dd, 2H, ArH, JHH = 8 Hz,
3
4JHH = 1 Hz), 6.86, (d, 2H, ArH, JHH = 9 Hz), 2.07 (s,
6H, Me), 1.34 (s, 18H, CMe3).
13C{1H} NMR 75 Mhz (CD2Cl2) d ppm 169.8
(N@CH), 159.4, 149.2, 143.7, 138.1, 135.6, 131.2,
130.6, 129.5, 129.0, 122.5, 122.3, 118.1 (Ar), 34.2
(CMe3), 31.1 (CMe3), 19.8 (Me).
MS (EI) m/z 691 [M]+, 656 [M ꢀ Cl]+, 641 [M ꢀ (Cl
and CH3)]+, 618 [M ꢀ (Cl)2]+.
4.6.6. [ZrL6Cl2]
Following the general method, Na2L6 ÆTHF0.59
(780 mg, 1.39 mmol) and [ZrCl4(THF)2] (540 mg,
1.43 mmol) were used.
IR (Nujol) m cmꢀ1 1614 (s, N@C), 1590, 1543, 1362,
1312, 1298, 1260 (s, C–O), 1209, 1184, 1151, 1143,
1098, 1026, 992, 982, 954, 895, 881, 838, 790, 776, 762,
738, 722, 704, 680, 667, 616.
MS (EI) m/z 692 [M]+, 677 [M ꢀ CH3]+, 657
Yield = 497 mg, 56%.
[M ꢀ Cl]+, 622 [M ꢀ (Cl)2]+.
Anal. found (Calculated for C32H30Cl2N2O2Zr)% C
60.11 (60.36), H 4.74 (4.75), N 4.30 (4.40).
1H NMR 300 MHz (CD2Cl2) d ppm 8.43 (s, 2H,
4.6.4. [ZrL4Cl2]
Following the general method, Na2L4 ÆTHF0.79
(800 mg, 1.39 mmol) and [ZrCl4(THF)2] (564 mg,
1.50 mmol) were used.
N@CH), 7.82 (t, 2H, ArH, JHH = 8 Hz), 7.21 (d, 2H,
3
3
3
ArH, JHH = 7 Hz), 7.03 (d, 2H, ArH, JHH = 7 Hz),
6.65 (s, 2H, ArH), 6.59 (s, 2H, ArH), 2.38 (s, 6H, Me),
2.33 (s, 6H, Me), 2.08 (s, 6H, Me).
Yield = 720 mg, 81%.
Anal. found (Calculated for C32H30Cl2N2O2Zr)% C
59.94 (60.36), H 4.75 (4.75), N 4.24 (4.40).
1H NMR 300 MHz (CD2Cl2) d ppm 8.47 (s, 2H,
13C{1H} NMR 75 MHz (CD2Cl2) d ppm 166.1
(N@CH), 161.9, 149.6, 149.2, 141.9, 137.7, 131.0,
129.1, 128.7, 124.3, 122.2, 119.5, 116.9 (Ar), 22.0 (Me),
19.8 (Me), 19.5 (Me).
IR (Nujol) m cmꢀ1 1930, 1606 (N@C), 1585, 1530,
1411, 1391, 1346, 1303, 1256 (w, C–O), 1228, 1209,
1159, 1102, 1072, 1017, 980, 942, 885, 865, 850, 840,
784, 759, 736, 656, 601.
MS (EI) m/z 636 [M]+, 599 [M ꢀ Cl]+.
4.6.7. [ZrL7Cl2]
Following the general method, Na2L7 ÆTHF1.12
(800 mg, 1.22 mmol) and [ZrCl4(THF)2] (476 mg,
1.26 mmol) were used.
3
N@CH), 7.34 (d, 2H, ArH, JHH = 8 Hz), 7.27 (t, 2H,
ArH, JHH = 8 Hz), 7.20 (dd, 2H, ArH, JHH = 8 Hz,
3
3
3
4JHH = 1 Hz), 7.06 (dd, 2H, ArH, JHH = 8 Hz,
3
4JHH = 1 Hz), 6.71 (d, 2H, ArH, JHH = 8 Hz), 2.38 (s,
6H, Me), 2.26 (s, 6H, Me), 2.09 (s, 6H, Me).
13C{1H} NMR 75 MHz (CD2Cl2) d ppm 166.9
(N@CH), 160.3, 149.6, 139.4, 137.6, 137.5, 131.0,
129.1, 128.6, 125.5, 122.1, 122.0, 121.0 (Ar), 19.8 (Me),
19.3 (Me), 15.7 (Me).
IR (Nujol) m cmꢀ1 2350, 2290, 1587 (m, N@C), 1556,
1401, 1320, 1292 (m, C–O), 1248, 1217, 1159, 1098,
1062, 994, 977, 939, 816, 762, 748, 720, 681, 651, 598,
561.
Yield = 602 mg, 71%.
Anal. found (Calculated for C36H38Cl2N2O2Zr)% C
61.48 (62.41), H 5.32 (5.53), N 4.47 (4.04).
1H NMR 300 MHz (CD2Cl2) d ppm 8.43 (s, 2H,
MS (EI) m/z 635 [M]+, 600 [M ꢀ Cl]+, 585 [M ꢀ (Cl
and CH3)+].
3
N@CH), 7.37 (d, 2H, ArH, JHH = 8 Hz), 7.25 (t, 2H,
ArH, JHH = 8 Hz), 7.18 (dd, 2H, ArH, JHH = 7 Hz,
4.6.5. [ZrL5Cl2]
3
3
Following the general method, Na2L5 ÆHF0.62
(800 mg, 1.29 mmol) and [ZrCl4(THF)2] (524 mg,
1.39 mmol) were used.
4JHH = 1 Hz), 7.08 (dd, 2H, ArH, JHH = 7 Hz,
3
3
4JHH = 1 Hz), 6.74 (d, 2H, ArH, JHH = 8 Hz), 3.33
3
(m, 2H, CHMe2, JHH = 7 Hz), 2.37 (s, 6H, Me), 2.09
3
(s, 6H, Me), 1.31 (d, 6H, CHMe2, JHH = 7 Hz), 1.26
Yield = 662 mg, 74%.
3
Anal. found (Calculated for C36H38Cl2N2O2Zr)% C
61.75 (62.41), H 5.48 (5.53), N 4.03 (4.04).
1H NMR 300 MHz (CD2Cl2) d ppm 8.57 (s, 2H,
(d, 6H, CHMe2, JHH = 7 Hz).
13C{1H} NMR 75 MHz (CD2Cl2) d ppm 167.2
(N@CH), 159.8, 149.5, 139.3, 137.7, 135.9, 133.7, 131.2,
129.0, 128.5, 122.2, 121.8, 121.2 (Ar), 28.0 (CHMe2),
22.2 (CHMe2), 21.8 (CHMe2), 19.8 (Me), 19.3 (Me).
IR (Nujol) m cmꢀ1 3051, 1953, 1872, 1804, 1600 (m,
N@C), 1582, 1557, 1462, 1412, 1393, 1361, 1350, 1310,
1286 (m, C–O), 1258, 1240, 1213, 1162, 1144, 1098,
1058, 1015, 979, 938, 906, 812, 779, 764, 744, 652, 618, 570.
MS (EI) m/z 692 [M]+, 657 [M ꢀ Cl]+, 622
[M ꢀ (Cl)2]+.
3
N@CH), 7.42 (d, 2H, ArH, JHH = 8 Hz), 7.27 (t, 2H,
ArH, JHH = 8 Hz), 7.19 (d, 2H, ArH, JHH = 7 Hz),
3
3
7.05 (ArH, 3JHH = 8 Hz), 6.83 (d, 2H, ArH,
3
3JHH = 8 Hz), 3.18 (m, 2H, CHMe2, JHH = 7 Hz), 2.25
(s, 6H, Me), 2.07 (s, 6H, Me), 1.29 (d, 6H, CHMe2,
3
3JHH = 7 Hz), 1.14 (d, 6H, CHMe2, JHH = 7 Hz).
13C{1H} NMR 75 MHz (CD2Cl2) d ppm 166.4
(N@CH), 160.1, 149.8, 149.8, 137.6, 137.5, 130.9, 129.0,

P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
5143
4.6.8. [ZrL8Cl2]
Yield = 461 mg, 77%.
Anal. found (Calculated for C36H40Cl2N2O2Zr)% C
61.85 (62.23), H 5.78 (5.80), N 3.78 (4.03).
Following the general method, Na2L8 ÆTHF1.23
(900 mg, 1.30 mmol) and [ZrCl4(THF)2] (500 mg,
1.33 mmol) were used.
1
cis-a-½ZrL120Cl2ꢁ H NMR 300 MHz (CD2Cl2) d ppm
3
Yield = 704 mg, 77%.
8.49 (bs, 2H, N@CH), 7.33 (d, 2H, ArH, JHH = 8 Hz),
Anal. found (Calculated for C38H42Cl2N2O2Zr)% C
62.91 (63.31), H 5.95 (5.87), N 3.61 (3.89).
1H NMR 300 MHz (CD2Cl2) d ppm 8.38 (s, 2H,
7.10 (br, ArH), 6.65 (d, 2H, ArH, 3JHH = 8 Hz), 2.36 (s,
6H, Me), 1.37 (s, 18H, CMe3).
cis-a-[Zr(L10)2Cl2] 13C{1H} NMR 75 MHz (CD2Cl2)
d ppm 167.2 (N@CH), 160.2, 151.1, 140.2, 137.4,
134.3, 128.8, 126.9, 123.4, 122.2, 121.9 (Ar), 34.7
(CMe3), 29.4 (CMe3), 19.0 (Me).
3
N@CH), 7.46 (d, 2H, ArH, JHH = 8 Hz), 7.23 (t, 2H,
ArH, JHH = 8 Hz), 7.17 (d, 2H, ArH, JHH = 7 Hz),
3
3
3
7.11 (d, 2H, ArH, JHH = 8 Hz), 6.72 (d, 2H, ArH,
3JHH = 8 Hz), 2.35 (s, 6H, Me), 2.11 (s, 6H, Me), 1.47
(s, 18H, CMe3).
IR (Nujol) m cmꢀ1 1601 (m, N@C), 1576, 1555, 1486,
1409, 1362, 1289, 1262 (s, C–O), 1189, 1138, 1059, 1027,
985, 968, 912, 857, 823, 768, 750, 703, 671, 646.
MS (EI) m/z 694 [M]+, 657 [M ꢀ Cl]+, 428
[M ꢀ (L10)]+.
13C{1H} NMR 75 MHz (CD2Cl2) d ppm 167.7
(N@CH), 161.0, 149.4, 140.0, 137.7, 137.3, 134.3,
131.3, 129.0, 128.5, 122.1, 122.0, 121.7 (Ar), 34.8
(CMe3), 29.4 (CMe3), 19.9 (Me), 19.3 (Me).
IR (Nujol) m cmꢀ1 1877, 1598 (N@C), 1580, 1552,
1402, 1384, 1300, 1268 (m, C–O), 1227, 1197, 1138,
1097, 1060, 1030, 978, 937, 904, 833, 816, 803, 780,
763, 743, 648, 618, 564.
4.6.11. [ZrL11Cl2]
Following the general method, NaL11 ÆTHF0.45
(556 mg, 1.73 mmol) and [ZrCl4(THF)2] (326 mg,
0.86 mmol) were used. After sublimation at 300 ꢁC,
10ꢀ6 mm Hg, a yellow solid was obtained.
Yield = 520 mg, 87%.
MS (EI) m/z 720 [M]+, 685 [M ꢀ Cl]+, 651
[M ꢀ (Cl)2]+.
Anal. found (Calculated for C36H40Cl2N2O2Zr)% C
62.10 (62.23), H 5.81 (5.80), N 4.09 (4.03).
4.6.9. [ZrL9Cl2]
Following the general method, Na2L9 ÆTHF2.35
(1.13 g, 1.46 mmol) and [ZrCl4(THF)2] (564 mg,
1.50 mmol) were used. After sublimation at 350 ꢁC,
10ꢀ6 mm Hg, a yellow solid was obtained. Analytical
data were obtained on this material. The product was
also crystallised from dichloromethane/pentane in simi-
lar yield. Single crystals were grown from
dichloromethane.
cis-a-½ZrL121Cl2ꢁ H NMR 300 MHz (CD2Cl2) d ppm
1
8.13 (bs, 2H, N@CH), 7.46 (br, 2H, ArH), 7.29 (br, 4H,
ArH), 7.08 (br, 4H, ArH), 6.95 (br, 2H, ArH), 6.84 (br,
2H, ArH), 2.29 (s, 6H, Me), 1.34 (s, 18H, CMe3).
cis-a-½ZrL121Cl2ꢁ 13C{1H} NMR 75 MHz (CD2Cl2) d
ppm 170.6 (N@CH), 157.6, 151.0, 139.1, 136.0, 133.1,
129.1, 128.9, 126.9, 123.8, 123.2 (Ar), 34.8 (CMe3),
29.4 (CMe3), 20.5 (Me).
Yield = 420 mg, 40% (From sublimation).
IR (Nujol) m cmꢀ1 1610 (m, N@C), 1588, 1555, 1488,
1435, 1405, 1360, 1334, 1292, 1269 (s, C–O), 1243, 1211,
1192, 1168, 1072, 1027, 984, 964, 906, 862, 844, 795, 776,
763, 702, 688, 618, 554.
Anal. found (Calculated for C38H42Cl2N2O2Zr)%
62.88 (63.31), H 5.83 (5.87), N 3.39 (3.89).
1H NMR 300 MHz (CD2Cl2) d ppm 8.10 (s, 2H,
4
N@CH), 7.40 (d, 2H ArH, JHH = 2 Hz), 7.22 (t, 2H,
ArH, JHH = 8 Hz), 7.15 (d, 2H, ArH, JHH = 7 Hz),
MS (EI) m/z 694 [M]+, 657 [M ꢀ Cl]+, 428
3
3
[M ꢀ (L11)]+.
4
7.02 (d, 2H, ArH, JHH = 2 Hz), 7.01 (d, 2H, ArH),
2.30 (s, 6H, Me), 2.06 (s, 6H, Me), 1.45 (s, 18H, CMe3).
13C{1H} NMR 75 MHz (CD2Cl2) d ppm 170.3
(N@CH), 158.7, 149.1, 139.0, 137.9, 136.1, 132.5,
131.1, 129.2, 129.1, 128.6, 123.1, 121.6 (Ar), 34.9
(CMe3), 29.3 (CMe3), 20.5 (Me), 19.8 (Me).
IR (Nujol) m cmꢀ1 1610 (N@C), 1590, 1548, 1431,
1410, 1340, 1295, 1269 (m, C–O), 1245, 1210, 1174,
1097, 1036, 1010, 979, 930, 902, 842, 794, 777, 764,
740, 549.
4.7. Crystallography
Crystals were coated with inert oil and transferred to
the cold (180 K) N2 gas stream on the diffractometer
(Bruker-AXS SMART three-circle with CCD area
detector). Graphite monochromated Mo Ka radiation
˚
k = 0.71073 A was used. Absorption correction was per-
formed by multi-scan (SADAB). The structures were
solved by direct methods using SHELXS [26] with addi-
tional light atoms found by Fourier methods. Crystals
of [ZrL9Cl2] contained a molecule of dichloromethane
disordered over an inversion centre Hydrogen atoms
were added at calculated positions and refined using a
riding model with freely rotating methyl groups. Aniso-
tropic displacement parameters were used for all non-H
atoms; H atoms were given isotropic displacement
MS (EI) m/z 720 [M]+, 685 [M ꢀ Cl]+, 651
[M ꢀ (Cl)2]+.
4.6.10. [ZrL120Cl2]
Following the general method, NaL10 ÆTHF0.54
(567 mg, 1.73 mmol) and [ZrCl4(THF)2] (326 mg,
0.86 mmol) were used.

5144
P.D. Knight et al. / Journal of Organometallic Chemistry 690 (2005) 5125–5144
[8] L. Blank, Statistical Procedures for Engineering, Management
and Science, McGraw-Hill, New York, first ed., pp. 487–515
(Chapter 27);
parameters Uiso(H) = 1.2Ueq(C) or 1.5Ueq(C) for methyl
groups. The structures were refined using SHELXL 96 and
SHELXTL [27].
L.C. Hamilton, Regression With Graphics – A Second Course in
Applied Statistics, Brooks/Cole, Belmont, CA, first ed., pp. 29–64
(Chapter 2).
Crystallographic data for the structural analysis has
been deposited with the Cambridge Crystallographic
Data Centre, CCDC 262887–262889.
[9] F.O. Rice, K.F. Herzfeld, J. Am. Chem. Soc. 56 (1934) 284.
[10] K.J. Laidler, Chemical Kinetics, third ed., Harper & Row, New
York, 1995, Section 8.5.
[11] P.R. Woodman, C.J. Sanders, N.W. Alcock, P.B. Hitchcock, P.
Scott, New J. Chem. 23 (1999) 815.
Acknowledgement
[12] P.R. Woodman, I.J. Munslow, P.B. Hitchcock, P. Scott, J. Chem.
Soc., Dalton Trans. (1999) 4069.
PS thanks EPSRC for support (PDK) and BP for a
CASE award (PDK).
[13] T.R. Younkin, E.F. Conner, J.I. Henderson, S.K. Friedrich,
R.H. Grubbs, D.A. Bansleben, Science 287 (2000) 460.
[14] G.J.P. Britovsek, V.C. Gibson, D.F. Wass, Angew. Chem., Int.
Ed. 38 (1999) 428.
References
[15] P.-J. Sinnema, T.P. Spaniol, J. Okuda, J. Organomet. Chem. 598
(2000) 179.
[1] (a) For example, see: P.R. Woodman, N.W. Alcock, I.J.
Munslow, C.J. Sanders, P. Scott, J. Chem. Soc., Dalton Trans.
(2000) 3340;
[16] S. Matsui, M. Mitani, J. Saito, Y. Tohi, H. Makio, N. Matsuk-
awa, Y. Takagi, K. Tsuru, M. Nitabaru, T. Nakano, H. Tanaka,
N. Kashiwa, T. Fujita, J. Am. Chem. Soc. 123 (2001) 6847.
[17] (a) C.A. Hunter, J.K.M. Sanders, J. Am. Chem. Soc. 112 (1990)
5525;
(b) P.A. Cameron, V.C. Gibson, C. Redshaw, J.A. Segal, A.J.P.
White, D.J. Williams, J. Chem. Soc., Dalton Trans. (2002) 415;
(c) D.G. Black, D.C. Swenson, R.F. Jordan, Organometallics 14
(1995) 3539;
(b) C. Janiak, J. Chem. Soc., Dalton Trans. (2000) 3885;
(c) H. Suezawa, T. Yoshida, Y. Umezawa, S. Tsuboyama, M.
Nishio, Eur. J. Inorg. Chem. (2002) 3148.
(d) H. Tsurugi, Y. Matsuo, T. Yamagata, K. Mashima, Organo-
metallics 23 (2004) 2797.
[18] (a) D.G. Bickley, N. Serpone, Inorg. Chem. 18 (1979) 2200;
(b) I.J. Munslow, A.J. Clarke, R.J. Deeth, I. Westmoreland, P.
Scott, Chem. Commun. (2002) 1868.
[2] (a) P.D. Knight, C. Morton, K.M. Gillespie, P. Scott, Chem.
Commun. (2003) 1770;
(b) P.N. OꢁShaughnessy, P. Scott, Tetrahedron: Asymmetry 14
(2003) 1979;
[19] D.A. Pennington, W. Clegg, S.J. Coles, R.W. Hartington, M.B.
Hursthouse, D.L. Hughes, M.E. Light, M. Schormann, M.
Bochmann, S.J. Lancaster, J. Chem. Soc., Dalton Trans. (2005)
561.
(c) P.D. Knight, I.J. Munslow, P.N. OꢁShaughnessy, P. Scott,
Chem. Commun. (2004) 894;
(d) P.N. OꢁShaughnessy, K.M. Gillespie, P.D. Knight, C. Morton,
P. Scott, J. Chem. Soc., Dalton Trans. (2004) 2251.
[3] (a) S. Matsui, T. Fujita, Catal. Today 66 (2001) 63;
(b) G.W. Coates, J. Chem. Soc., Dalton Trans. (2002) 467;
(c) H. Makio, N. Kashiwa, T. Fujita, Adv. Synth. Catal. 344
(2002) 477;
[20] N. Matsukawa, S. Matsui, M. Mitani, J. Saito, K. Tsuru, N.
Kashiwa, T. Fujita, J. Mol. Catal. A 169 (2001) 99.
[21] D.W. Carpenetti, L. Kloppenburg, J.T. Kupec, J.L. Petersen,
Organometallics 15 (1996) 1572.
[22] (a) R.R. Schrock, J.D. Fellmann, J. Am. Chem. Soc. 100 (1978)
3359;
(d) Y. Suzuki, H. Terao, T. Fujita, Bull. Chem. Soc. Jpn. 76
(2003) 1493;
(b) Available from: <http://www.syntheticpages.org/pages/120>.
[23] (a) P.J. Davidson, M.F. Lappert, R. Pearce, J. Organomet. Chem.
57 (1973) 269;
(e) M. Mitani, J. Saito, S.I. Ishii, Y. Nakayama, H. Makio, N.
Matsukawa, S. Matsui, J.I. Mohri, R. Furuyama, H. Terao, H.
Bando, H. Tanaka, T. Fujita, Chem. Rec. 4 (2004) 137.
[4] H. Makio, T. Fujita, Macromol. Symp. 213 (2004) 221.
[5] S. Reinartz, A.F. Mason, E.B. Lobkovsky, G.W. Coates,
Organometallics 22 (2003) 2542.
(b) Available from: <http://www.syntheticpages.org/pages/135>.
[24] (a) U. Zucchini, E. Albizzati, U. Giannini, J. Organomet. Chem.
26 (1971) 357;
(b) Available from: <http://www.syntheticpages.org/pages/1>.
[25] N.U. Hofsløkken, L. Skattebøl, Acta Chem. Scand. 53 (1999)
258.
[6] P.D. Knight, A.J. Clarke, B.S. Kimberley, R.A. Jackson, P.
Scott, Chem. Commun. (2002) 352.
[26] G.M. Sheldrick, SHELXTL Ver 5.1, Bruker Analytical X-ray
Systems, 1997.
[27] G.M. Sheldrick, Acta Crystallogr., Sect. A 46 (1990) 467.
[7] P.D. Knight, P.N. OꢁShaughnessy, I.J. Munslow, B.S. Kimberley,
P. Scott, J. Organomet. Chem. 683 (2003) 103.