The synthesis, coordination chemistry and ethylene polymerisation activity of ferrocenediyl nitrogen-substituted ligands and their metal complexes

Article abstract of DOI:10.1016/j.jorganchem.2005.05.051

Treatment of aminoferrocene with substituted 2-hydroxybenzaldehydes yields the air- and moisture-stable ligands 1-4, which were then reacted to form the chromium dichloride complexes 5-7 and the nickel bis-chelate species 8 and 9. The metal compounds are very air-sensitive but the chromium compounds act as pre-catalysts for the polymerisation of ethylene. Reaction of 1,1′-bis(amino)ferrocene with similarly substituted 2-hydroxybenzaldehyes or simple benzaldehyde gives the ligands 10-12 and 17, respectively. The X-ray crystal structure of 11 shows the molecule to have non-crystallographic C ₂ symmetry and to be linked by C-H...π interactions between the anthracene rings. Titanium-containing complexes 13-16 can be formed utilising ligands 10-12 and there is a change in geometry within the complexes dependent on the adjacent co-ligands, whilst ligand 17 can be reacted with PdClMe(COD) to form the chelate complex 18. Cyclic voltammetric studies have been carried out on 18 and its oxidised analogue 19, but both complexes are inactive towards ethylene polymerisation.

Full text of DOI:10.1016/j.jorganchem.2005.05.051

Journal of Organometallic Chemistry 690 (2005) 6271–6283
www.elsevier.com/locate/jorganchem
The synthesis, coordination chemistry and ethylene
polymerisation activity of ferrocenediyl nitrogen-substituted
ligands and their metal complexes
Vernon C. Gibson ,a, Charlotte K.A. Gregson , Catherine M. Halliwell ,
a
b
*
a,
a
a
a
Nicholas J. Long *, Philip J. Oxford , Andrew J.P. White , David J. Williams
a
Imperial College London, Exhibition Road, South Kensington, London, SW7 2AZ, United Kingdom
The National Physical Laboratory, Hampton Road, Teddington, Middlesex TW11 0LW, United Kingdom
b
Available online 11 November 2005
Abstract
Treatment of aminoferrocene with substituted 2-hydroxybenzaldehydes yields the air- and moisture-stable ligands 1–4, which were
then reacted to form the chromium dichloride complexes 5–7 and the nickel bis-chelate species 8 and 9. The metal compounds are very
air-sensitive but the chromium compounds act as pre-catalysts for the polymerisation of ethylene. Reaction of 1,10-bis(amino)ferrocene
with similarly substituted 2-hydroxybenzaldehyes or simple benzaldehyde gives the ligands 10–12 and 17, respectively. The X-ray crystal
structure of 11 shows the molecule to have non-crystallographic C2 symmetry and to be linked by C–Hꢀ ꢀ ꢀp interactions between the
anthracene rings. Titanium-containing complexes 13–16 can be formed utilising ligands 10–12 and there is a change in geometry within
the complexes dependent on the adjacent co-ligands, whilst ligand 17 can be reacted with PdClMe(COD) to form the chelate complex 18.
Cyclic voltammetric studies have been carried out on 18 and its oxidised analogue 19, but both complexes are inactive towards ethylene
polymerisation.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Ferrocene; Catalysis; N-ligands; Transition metals; Ethylene polymerisation
1. Introduction
the polymerisation of rac-lactide [10,11] and methyl meth-
acrylate polymerisation [11].
Schiff-base ligands play pivotal roles in homogeneous
catalysis. Indeed, salicylaldiminato complexes of group 4
metals [1–5], nickel [6,7] and chromium [8] have all been
used as active pre-catalysts for ethylene polymerisation.
Similarly group 4 metal complexes based on the family of
salen ligands have also been used for ethylene and propyl-
ene polymerisation catalysis [9]. These latter ligands have
also been successfully utilised in a variety of other catalytic
processes, for instance in conjunction with aluminium for
In other aspects of catalysis and ligand design, ferrocene
ligands have an important role to play: (i) as a backbone or
substituent in ligands providing a specific and unique
geometry and (ii) via their redox activity and the potential
control of the reactivity of a coordinated, catalytically-
active metal centre [12,13]. Although substituted ferrocenyl
amines are known [14], the formation of N-substituted ferr-
ocenyl and ferrocenediyl species has been limited. In recent
years though, efficient routes to ferrocenediyldiamine [15]
have been reported and the titanium and zirconium com-
plexes of these and related ligands [16–18] have been high-
lighted [19] and shown to exhibit moderate ethylene
polymerisation activity.
*
Corresponding author. Tel.: +44 20 75945781; fax: +44 20 75945804.
E-mail addresses: v.gibson@imperial.ac.uk (V.C. Gibson), n.long@
imperial.ac.uk (N.J. Long).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.05.051

6272
V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
Thus, as part of our on-going quest to find new steri-
likely that a mixture of products is formed during these
reactions.
cally-hindered and redox-active ferrocene ligands, this
manuscript features the introduction of a ferrocene moiety
into a variety of salicylaldimine, salen and bis(imine)
ligands, subsequent coordination to a range of transition
metals and an investigation into their potential as catalysts
for ethylene polymerisation.
1.3. Polymerisation tests using chromium catalysts 5–7
In order to test the ethylene polymerisation activity of
the chromium species 5–7 the compounds were redissolved
in toluene and 100 equiv. of MAO were added to the reac-
tion mixture which was then exposed to 1 bar ethylene
pressure for 1 h during which time the precipitation of solid
polyethylene was observed. The polymerisation test was
terminated by the removal of the ethylene supply and the
addition of acidified methanol. The polyethylene was sepa-
rated from the reaction mixture by filtration and dried in a
vacuum oven overnight. The results are summarised in
Table 1. The catalysts showed moderate activity, and the
anthracenyl-substituted complex (5) was found to be com-
parable with the non-ferrocenyl analogues [8]. Analysis of
the polymer formed from 5 showed it to be high molecular
weight with a relatively narrow molecular weight distribu-
tion (Mw – 203000, Mn – 79000, PDI 2.6). This is similar to
the literature values [8] for this type of system. For com-
pounds 6 and 7, insoluble polyethylene was obtained indic-
ative of very high molecular weight (>2M Da.).
1.1. Synthesis of ferrocenyl salicylaldimines
The ferrocenyl salicylaldimines 1–4 were conveniently
synthesised by the Schiff base condensation of ferrocenyl-
amine with a range of salicylaldehydes (Scheme 1). The
reactions were carried out by adding a dichloromethane
solution of the substituted salicylaldehyde to a dichloro-
methane solution of ferrocenylamine and heating the
reaction mixture to reflux for 5 h. During this time a col-
our change of orange to red was observed and after
removal of the solvent in vacuo, the resulting red sticky
solid was easily recrystallised from ethanol to give the
desired products in moderate yields. All analyses were
1
consistent with the named compounds. H NMR spectra
show a characteristic singlet resonance due to the OH
proton between 13 and 14 ppm, the high field chemical
shift being due to intramolecular H-bonding with the
imine nitrogen. A singlet resonance due to the aldimine
proton between 8.1 and 8.8 ppm is also observed. All
compounds exhibit two pseudo triplets and a singlet
resonance in the ferrocenyl region, which is indicative
of mono-substitution of one of the cyclopentadienyl
rings.
1.4. Synthesis and evaluation of nickel catalysts supported by
ferrocenyl salicylaldimines
Grubbs and co-workers [6] have recently shown that a
series of salicylaldiminato complexes of Ni with a bulky
group ortho to the OH group can form highly active ethyl-
ene polymerisation catalysts. The bulky groups retard asso-
ciative displacement reactions in much the same way as for
cationic Ni(a-diimine) systems, whilst the formation of
inactive bis-ligand complexes is also prevented. Thus, a
pronounced dependence of activity on the phenyl substitu-
ent (in this case anthracenyl) is observed. The highest activ-
1.2. Synthesis of chromium catalysts supported by ferrocenyl
salicylaldimines
Work in our laboratories on chromium catalysts for eth-
ylene polymerisation has shown that a series of substituted
salicylaldimine complexes of chromium can lead to active
catalysts [20]. The nature of the phenyl ring substituent(s)
and the nitrogen alkyl/aryl group can lead to a difference
in the catalytic activity and physical properties of the
resulting polymer. Monochelate analogues have also been
found to be accessible if the phenoxide ortho position con-
tains a bulky substituent such as an anthracenyl group [8].
In order to evaluate the activity of the ferrocenyl salicyl-
aldimines, in situ polymerisation studies were carried out
by the addition of a toluene solution of the ferrocenyl sal-
icylaldimines with p-tolylCrCl2 Æ 3THF in a Schlenk vessel
to form compounds 5–7. The reaction proceeds at room
temperature and after 3 h the solvent was removed under
vacuum to eliminate any THF, which may hinder catalysis.
NMR analysis of the chromium compounds 5–7 was prob-
lematic due to their paramagnetic nature – only broad
unassignable peaks were observed. Elemental analysis sug-
gests that the chromium dichloride complexes are formed,
but mass spectrometry indicates the possible formation of a
bis chelate-chromium monochloride compound and it is
ity, of several hundred g mmolꢁ1
h
ꢁ1 barꢁ1 is observed
with complexes containing a weakly coordinating acetoni-
trile ligand [6]. In order to investigate whether the ferroce-
nyl analogues would form active nickel catalysts, reaction
of 1 and 2 with (TMEDA)NiMe2 in the presence of aceto-
nitrile was carried out (Scheme 1). A solution of the salicyl-
aldimine in acetonitrile was added dropwise to a solution
of (TMEDA)NiMe2 also in acetonitrile at ꢁ78 ꢁC and
the reaction mixture stirred for 1 h after which it was
warmed to room temperature. The resulting red coloured
solution was stirred for a further 30 min and the solvent
removed in vacuo to give red-orange solids 8 and 9. In both
1
cases the H NMR spectra of the product showed broad
resonances indicative of a tetrahedral paramagnetic com-
pound plus unreacted (TMEDA)NiMe2. The new com-
pounds could be isolated by washing the crude product
with cold diethyl ether in order to remove the unreacted
starting materials and elemental analysis suggested that
in all cases the bis-chelate compound was formed. Floris
[21] has previously shown that 1 can react with a range

V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
6273
R
1
2
3
4
R = R' = H
R
R = H, R' = 9-Anthracenyl
R = H, R' = Tryptocenyl
R = R' = tBu
NH
N
2
HO
R'
Fe
Fe
+
O
HO
R'
(TMEDA)NiMe2
CrCl2(thf)3
R
- TMEDA
-2CH4
Fc
N
Fc
O
R'
Ni
N
O
R'
O
CrCl
2
N
Fc
R
R'
R
8
9
R = R' = H
R =H, R' = 9-Anthracenyl
5
6
7
R = H, R' = 9-Anthracenyl
R = H, R' = 1-Trypticenyl
R = R' = tBu
NH
2
Fe
NH
2
O
O
HO
R
-2H2O
N
N
N
17
Fe
10 R = H
11 R = 9-Anthracenyl
12 R = tBu
HO
HO
R
R
Fe
N
Ti(OiPr)4
TiCl4.thf2
-2HCl
-2 iPrOH
N
Cl
Pd
18
Fe
Cl
Ti
Cl
O
N
N
R
N
N
Me
O
O
O
O
R
Fe
Fe
Ti
O
N
+
-
(BAF)
13
14 R = H
15 R = 9-Anthracenyl
16 R = tBu
N
N
Cl
19
Pd
Fe
Me
Scheme 1. The formation of compounds 1–19.
of metal salts (Co(II), Ni(II), Zn(II) and Cu(II)) to give bis-
chelate compounds. It was hoped that the bulky substitu-
ents would have prevented the formation of the bis-chelate
(which is catalytically inactive), but it is likely that the com-
paratively small size of the cyclopentadienyl ring as
opposed to a substituted phenyl ring means that bis-chelate

6274
V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
Table 1
Ethylene polymerisation tests with chromium catalysts 5–7
Cat
Cat. conc. (lmol)
MAO (equiv.)
Time
Temperature (ꢁC)
Mass polymer (g)
Activity (g mmolꢁ1 barꢁ1 hꢁ1
)
5
5
5
6
7
a
50
10
10
10
10
100
100
100
100
100
20 mina
1 h
1 h
1 h
20
20
50
20
20
2.15
0.94
0.46
0.22
0.34
129
94
46
22
34
1 h
Test stopped after 20 min due to difficulty in stirring.
formation is possible. The nickel compounds were tested as
ethylene polymerisation catalysts but in all cases were
found to be inactive.
1.5. Synthesis of ferrocenyl bis-salicylaldimine (Salen type)
‘Salfen’ ligands
The Schiff base condensation of 1,10-bis(amino)ferro-
cene [15,22] with a range of substituted salicylaldehydes
proceeds readily in dichloromethane solution to form
ligands 10–12 (Scheme 1). The reactions were carried out
by adding a dichloromethane solution of the substituted
salicylaldehyde to a dichloromethane solution of 1,10-
bis(amino)ferrocene and heating the reaction mixture to
reflux for 1 h. During this time a colour change of orange
to burgundy was observed. The solvent was removed under
vacuum and the resulting burgundy coloured solid was eas-
ily recrystallised from dichloromethane:pentane (25:75)
solution to give the products in good yields (68–85%).
The relationship between this family of ligands and the
family of salen ligands can easily be seen and they have
recently been coined ꢀsalfenꢁ ligands [15]. In a conventional
salen ligand the backbone contains a C2 unit, which has
effectively been replaced by a ferrocene unit in a salfen
ligand. The distance between the two nitrogen atoms is
somewhat larger in a salfen ligand, although it is possible
for the ferrocenediyl unit to twist and deform in order to
coordinate to a metal centre. All analyses are consistent
Fig. 1. The molecular structure of 11. The intramolecular hydrogen bonds
˚
have Oꢀ ꢀ ꢀN, Hꢀ ꢀ ꢀN (A), O–Hꢀ ꢀ ꢀN (ꢁ): (a) 2.61, 1.83, 145; (b) 2.62, 1.79,
154.
1
with the named compounds. The H NMR spectra of the
Table 2
˚
Selected bond lengths (A) and angles (ꢁ) for (11)
ligands show two pseudo-triplets in the ferrocenyl region,
which is consistent with a time-averaged C2v symmetry in
solution.
C(1)–N(11)
N(11)–C(11)
N(32)–C(32)
1.397(4)
1.286(4)
1.287(4)
C(6)–N(32)
C(11)–C(12)
C(32)–C(33)
1.402(4)
1.445(4)
1.439(5)
A single crystal structure analysis of 11 showed the mol-
ecule to have approximate non-crystallographic C2 symme-
try about an axis passing through the iron centre and
bisecting the vector linking the centres of the two phenyl
rings (Fig. 1). The two cyclopentadienyl rings are inclined
by ca. 2ꢁ and staggered by ca. 21ꢁ. The pattern of bonding
in the C(1)–C(12) and C(6)–C(33) chains (Table 2) do not
differ significantly from those observed in the structure of
N-(o-hydroxybenzylidene)-ferroceneamine [13]. Here, in
11 the C(12)- and C(33)-containing phenyl rings are rotated
by ca. 17ꢁ and 15ꢁ, respectively, out of the planes of their
parent cyclopentadienyl rings (cf. ca. 10ꢁ in the literature
complex). In both arms of the molecule there is an intramo-
lecular O–Hꢀ ꢀ ꢀN hydrogen bond between the phenolic oxy-
N(11)–C(1)–C(5)
N(32)–C(6)–C(10)
C(11)–N(11)–C(1)
C(17)–C(12)–C(11)
C(32)–N(32)–C(6)
C(38)–C(33)–C(32)
121.7(4)
N(11)–C(1)–C(2)
N(32)–C(6)–C(7)
N(11)–C(11)–C(12)
C(13)–C(12)–C(11)
N(32)–C(32)–C(33)
C(34)–C(33)–C(32)
129.9(4)
122.5(3)
121.4(3)
119.5(3)
120.3(3)
119.7(3)
130.0(3)
122.2(3)
121.6(3)
122.5(3)
122.0(3)
gen atom and the amino nitrogen. The anthracene and
phenyl rings are oriented approximately orthogonally to
each other, the torsional twists about the C(14)–C(18)
and C(35)–C(39) bonds being ca. 78ꢁ and 89ꢁ, respectively.
The two phenyl rings are p-stacked with a mean interplanar
˚
separation of ca. 3.65 A and a centroidꢀ ꢀ ꢀcentroid distance

V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
6275
X
O
O
X
O
N
O
N
N
N
X
X
O
N
X
N
X
O
planar (trans isomer)
α-cis
β-cis
Fig. 3. Possible coordination geometries of salen ligands.
complicated and shows eight individual ferrocenyl reso-
nances. Arnold [15] has shown that 12 can coordinate to
Ti(Bn)4 and Zr(Bn)4 in a planar fashion and the 1H
NMR spectra of the ferrocene region is reported to have
two pseudo triplet references, suggesting that the salfen-
TiCl2 complex 13 has a similar geometry. In contrast, the
1H NMR spectrum of the salfenTi(OiPR)2 complexes are,
however, much more complicated. The 1H NMR spectrum
of 16 shows two sets of resonances for the imino protons
(at 8.26 and 8.51 ppm), the aryl protons and the tert-butyl
protons. The ferrocenyl region shows eight multiplets –
indicating that each of the ferrocenyl protons is inequiva-
lent. Furthermore, two distinct septets are visible for the
CH(CH3)2 isopropyl protons and four doublets are noted
for isopropyl CH3 groups. This indicates that one of the
ligand O atoms is trans to an isopropyl unit and the other
is cis to an isopropyl unit. The reason for the change in
geometry is probably due to the larger size of the isopropyl
groups compared to a chloride or benzyl group.
Fig. 2. Part of one of the C–Hꢀ ꢀ ꢀp linked chains of molecules in the
˚
structure of 11. The Hꢀ ꢀ ꢀring-centroid distances (A) are: (a) 3.10; (b) 3.18;
(c) 3.16; (d) 2.78; (e) 2.80; (f) 2.76.
Isolation of X-ray quality crystals for all the titanium
species was difficult and only the structures of oxo-bridged
dimers could be obtained. Two are detailed here: A is the
oxo-bridged titanium dichloride complex of ligand 10,
whilst B is the oxo-bridged dimer of 14. The X-ray analysis
of A revealed an oxo-bridged dimer structure, as illustrated
in Fig. 4. The geometries at the two titanium centres are
˚
of 4.00 A; the rings are inclined by ca. 8ꢁ. Centrosymmetri-
cally related pairs of molecules are linked by C–Hꢀ ꢀ ꢀp
interactions between the anthracene rings to form chains
that extend in the crystallographic 110 direction (Fig. 2).
1.6. Reaction of 10–12 with titanium precursors
The reaction of the Salfen type ligands with titanium
precursors TiCl4 Æ (THF)2 and Ti(OiPr)4 was investigated
(Scheme 1). Reactions were carried out by addition of a tol-
uene solution of the ligand to a toluene solution of the tita-
nium precursor and stirring at 40 ꢁC for 3 h. The solvent
was removed in vacuo and the product washed with cold
pentane to give the products. The dichloride product 13
could be successfully recrystallised from dichlorometh-
ane:pentane (25:75) solution and isolated as brown needles
in 66% yield, The diisopropyl products 14–16 were isolated
as red powders in moderate yields (45–77%). The size of
Ti(IV) means that the Salfen ligand canꢁcoordinate to the
titanium as a tetradentate [N,Oꢁ,N0,O0 ] ligand. Octahe-
dral salen metal complexes have been shown to display a
range of coordination modes (Fig. 3) and, depending on
the metal precursor, the salfen ligands can also show differ-
ent coordination geometries. In the case of the dichloride
product 13 the ferrocenyl region of the 1H NMR spectrum
shows two pseudo triplet resonances, whereas the NMR
spectra of the diisopropyl complexes 14–16 is much more
Fig. 4. The molecular structure of A.

6276
V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
distorted octahedral with cis angles at the metal centres
ranging between 82.4(1)ꢁ and 99.8(1)ꢁ at Ti(1), and
82.1(1)ꢁ and 100.8(1)ꢁ at Ti(2); the Cl(1)–Ti(1)–O and
Cl(2)–Ti(2)–O angles are 166.2(1)ꢁ and 168.2(1)ꢁ, respec-
tively. The gross structure is very similar to that observed
in the related oxo-bridged complex dichloro-(l2-oxo)-
bis(N,N0-disalicylidene-(R,R)-1,2-cyclohexane-diamine-
O,O0,N,N0)-di-titanium(iv) [23]. The Ti–O(oxo) distances
˚
[1.838(3) and 1.824(3) A] and the Ti–O(salicylidene) bond
˚
lengths [range 1.846(3)–1.878(3) A] (Table 3) do not differ
significantly from those in the above literature structure,
though the folding at the oxo atom in A is greater at
155.8(2)ꢁ cf. 168ꢁ. The Ti–N distances in A [range
˚
˚
2.250(4)–2.279(4) A] are all ca. 0.1 A longer than those in
the above structure reflecting the change from a cyclohexyl
to a ferrocenyl bridge between the nitrogen centres. The
N(1),O(1) chelate ring is nearly planar with Ti(1) lying only
˚
0.06 A out of the plane of the remaining atoms, which are
coplanar to within 0.02 A. By contrast, the other three N,O
˚
Fig. 5. The molecular structure of B.
chelate rings are all folded significantly out of plane about
their Nꢀ ꢀ ꢀO vectors with fold angles of 27ꢁ, 23ꢁ, and 26ꢁ for
the N(2),O(2), N(3),O(3), and N(4),O(4) containing rings,
respectively; the deviations of their associated titanium cen-
Table 4
˚
Selected bond lengths (A) and angles (ꢁ) for B
˚
tres are 0.69, 0.61 and 0.68 A respectively. The bending at
Ti–O
Ti–O(1)
Ti–N(18)
1.912(3)
1.948(3)
2.365(4)
Ti–O0
Ti–N(7)
Ti–O(24)
1.792(3)
2.247(3)
1.910(3)
the oxo bridge results in the N(3)@C(31) double bond over-
laying the C(1)–C(6) phenyl ring in a p–p stacking arrange-
ment, the distance of the centre of the bond to the ring
O0–Ti–O(24)
O(24)–Ti–O
O(24)–Ti–O(1)
O0–Ti–N(7)
O–Ti–N(7)
O0–Ti–N(18)
O–Ti–N(18)
N(7)–Ti–N(18)
101.07(13)
97.39(13)
92.26(13)
90.85(13)
89.74(12)
167.75(13)
84.52(12)
89.94(13)
O0–Ti–O
O0–Ti–O(1)
O–Ti–O(1)
O(24)–Ti–N(7)
O(1)–Ti–N(7)
O(24)–Ti–N(18)
O(1)–Ti–N(18)
Ti0–O–Ti
83.26(13)
96.07(13)
170.28(13)
166.72(14)
80.57(13)
79.66(13)
96.12(13)
96.74(13)
˚
plane being only 3.25 A. Both ferrocenyl units have stag-
gered geometries, with stagger angles of 21ꢁ and 28ꢁ for
Fe(1) and Fe(2), respectively; the cyclopentadienyl rings
are inclined by ca. 1ꢁ [Fe(1)] and ca. 4ꢁ [Fe(2)]. The struc-
ture contains an MeCN molecule of solvation and there are
no noteworthy intermolecular packing interactions.
A single crystal structure of B showed the complex to be
dimeric and centrosymmetric with the tetradentate ligands
adopting an axial/equatorial coordination (Fig. 5) in con-
trast to the all-equatorial geometry observed in A. The
geometry at titanium is distorted octahedral with cis angles
in the range 79.7(1)–101.1(1)ꢁ and trans angles of between
166.7(1)ꢁ and 170.3(1)ꢁ (Table 4). The gross structure is
very similar to that of the cyclohexyl containing analogue,
bis((l2-oxo)-(N,N0-(cyclohexane-1,2-diyl)-bis(salicylaldimi-
nato))-titanium(iv)) [24]. The Ti–O(salicylidene) bond
Table 3
˚
Selected bond lengths (A) and angles (ꢁ) for A
Ti(1)–O
1.838(3)
1.878(3)
2.250(4)
1.824(3)
1.873(3)
2.252(4)
Ti(1)–O(1)
Ti(1)–N(1)
Ti(1)–Cl(1)
Ti(2)–O(3)
Ti(2)–N(3)
Ti(2)–Cl(2)
1.846(3)
2.279(4)
2.393(2)
1.861(3)
2.268(4)
2.4176(14)
Ti(1)–O(2)
Ti(1)–N(2)
Ti(2)–O
Ti(2)–O(4)
Ti(2)–N(4)
O–Ti(1)–O(1)
97.5(2)
94.44(14)
173.8(2)
84.86(13)
177.29(14)
166.16(11)
95.15(11)
83.45(11)
94.9(2)
O–Ti(1)–O(2)
O–Ti(1)–N(2)
96.83(14)
88.23(14)
82.38(13)
83.21(14)
99.82(13)
88.44(12)
86.51(10)
100.77(14)
95.13(14)
173.57(14)
91.01(14)
173.8(2)
O(1)–Ti(1)–O(2)
O(1)–Ti(1)–N(2)
O–Ti(1)–N(1)
O(2)–Ti(1)–N(1)
O–Ti(1)–Cl(1)
O(2)–Ti(1)–Cl(1)
N(1)–Ti(1)–Cl(1)
O–Ti(2)–O(4)
O(2)–Ti(1)–N(2)
O(1)–Ti(1)–N(1)
N(2)–Ti(1)–N(1)
O(1)–Ti(1)–Cl(1)
N(2)–Ti(1)–Cl(1)
O–Ti(2)–O(3)
O(3)–Ti(2)–O(4)
O(3)–Ti(2)–N(4)
O–Ti(2)–N(3)
O(4)–Ti(2)–N(3)
O–Ti(2)–Cl(2)
O(4)–Ti(2)–Cl(2)
N(3)–Ti(2)–Cl(2)
˚
˚
lengths [1.910(3) A (trans to N) and 1.948(3) A (trans to
oxo)] are very similar to those observed in the related liter-
˚
ature structure, whereas those to nitrogen [2.247(3) A to
˚
˚
N(7) and 2.365(4) A to N(18)] are ca. 0.1 A longer. The
˚
two Ti–O(oxo) bonds in B [1.792(3) and 1.912(3) A] are
markedly asymmetric, though these distances are the same
as those observed in the N–CH2–CH2–N linked analogue
bis((l2-oxo)-(N,N0-disalicylidene-1,2-diaminoethane)-tita-
nium(iv)) [25]. The two N,O chelate rings have more steeply
folded geometries than those observed in A with fold angles
about their Nꢀ ꢀ ꢀO vectors of 40ꢁ [N(7),O(1)] and 31ꢁ
[N(18),O(24)] and associated out-of-plane displacements
O–Ti(2)–N(4)
85.66(13)
84.2(2)
O(4)–Ti(2)–N(4)
O(3)–Ti(2)–N(3)
N(4)–Ti(2)–N(3)
O(3)–Ti(2)–Cl(2)
N(4)–Ti(2)–Cl(2)
Ti(2)–O–Ti(1)
82.09(13)
97.93(13)
89.31(11)
84.30(10)
155.8(2)
168.16(10)
90.29(12)
84.18(10)

V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
6277
˚
of the titanium atom of 1.03 and 0.84 A, respectively. The
ferrocenyl unit has a staggered geometry (29ꢁ) with its
cyclopentadienyl rings being inclined by ca. 9ꢁ. The crystals
contain a CHCl3 molecule of solvation which is disordered,
and the only intermolecular packing interaction of note is a
p–p stacking of the C(1)–C(6) rings of centrosymmetrically
related pairs of molecules. The centroidꢀ ꢀ ꢀcentroid and
˚
mean interplanar separations are 4.10 and 3.40 A,
respectively.
1.7. Synthesis of N,N0-bis(benzylimido)ferrocenediamine
(17) and some palladium complexes (18 and 19)
The Schiff-base condensation of 1,10-bis(amino)ferro-
cene with benzaldehyde proceeds readily in toluene solu-
tion at room temperature. Freshly distilled benzaldehyde
was added to a toluene solution of 1,10-bis(amino)ferrocene
and the reaction mixture stirred for 3 h during which time
the initial yellow colour became deep red. The toluene and
excess benzaldehyde were removed in vacuo to yield a bur-
gundy solid, which was washed with pentane to give the red
coloured, air-stable bis(imine) in 75% yield (Scheme 1) [18].
Although compound 17 has a bis-imino type functionality,
attempts at forming coordination compounds with metal
precursors (dme)NiBr2 and FeCl2 failed; it was thought
that the twist energy of the imine arms and their relative
rigidity precluded coordination. However, it was antici-
pated that a larger metal such as palladium might form
complexes, and indeed this has now been achieved. Reac-
tion of 17 with Pd(COD)MeCl [26] in DCM at 40 ꢁC for
72 h, and removal of the solvent gave an orange terra-
cotta-coloured solid 18, that was crystallised at ꢁ30 ꢁC
from pentane/dichloromethane (Scheme 1). Spectroscopic
analysis of the compound in both 1H and 13C NMR
revealed that the two halves of the ligands were now no
longer equivalent, owing to the different influences exerted
by the methyl and chloride groups on the palladium centre,
hence two imine peaks are observed and the ferrocene gives
eight separate peaks in the 13C-spectrum. Orange needle-
like crystals of 18 suitable for X-ray diffraction studies were
obtained from a 2-layer solution of dichloromethane/hex-
ane. The structure of 18 (Fig. 6) is very similar to that of its
dichloro analogue (1,10-bis(benzylideneamino)ferrocene-
N,N0)-dichloro-palladium(II) [27], the principal difference
being in the inclination of the phenyl rings to their associ-
ated cyclopentadienyl rings. In 18 the C(2)–C(6) ring is
inclined by ca. 74ꢁ to C(13)–C(18) whereas the C(7)–
C(11) ring is inclined by only ca. 7ꢁ to C(20)–C(25). In
the former case the inclination is made up of torsional
twists of ca. 48ꢁ and 25ꢁ in the same sense about the
C(2)–N(12) and C(12)–C(13) bonds, respectively, whereas
in the later the torsional twists about C(7)–N(19) and
C(19)–C(20), each ca. 29ꢁ, are in opposite senses. In the lit-
erature structure, which contains two independent mole-
cules, the cyclopentadienyl and their associated phenyl
rings are inclined by angles ranging between only 10ꢁ and
18ꢁ. The geometry at palladium is distorted square planar
˚
Fig. 6. The molecular structure of 18. The Feꢀ ꢀ ꢀPd separation is 3.60 A.
with cis angles in the range 88.2(5)–92.8(5)ꢁ with trans
angles of 172.8(3) and 177.1(5)ꢁ (Table 5). The Pd–N(12)
and Pd–Cl distances do not differ significantly from those
observed for the dichloro analogue. The Pd–N(19) bond
length is significantly longer, reflecting the difference of N
trans to Me cf. N trans to Cl. The ferrocenyl unit has a
slightly staggered geometry (ca. 6ꢁ) the rings being inclined
by ca. 5ꢁ. The only packing interaction of note is a weak p–
p stacking of the C(13)–C(18) ring of one molecule with the
C(20)–C(25) ring of a symmetry related neighbour, and
˚
vice versa; the mean interplanar separation is 3.66 A.
With a view to potential application as an ethylene poly-
merisation catalyst 18 was reacted with sodium tetra-
kis(3,5-bis-trifluoromethyl-phenyl)borate (BAF) [28] to
form the new compound [{1,10-bis(benzylimino)ferro-
cene}Pd+Me(MeCN)][BAFꢁ] (19). As in the preparation
of 18, reactants for 19 were heated together in dichloro-
methane at 40 ꢁC, this time overnight. Removal of the sol-
vent gave a brown/orange/yellow lustrous solid in 59%
yield. As before the two halves of the compound are still
inequivalent due to one imine being trans to a methyl group
whilst the other is trans to a loosely coordinated acetoni-
trile molecule. Multiple splittings are therefore again seen
in both the 1H and 13C NMR, particularly in the aryl
region, where full assignment was made by comparison
with literature values [28] for BAFꢁ. Further evidence for
the existence of 18 and 19 came from mass spectra and ele-
mental analysis. Attempts are currently underway to
Table 5
˚
Selected bond lengths (A) and angles (ꢁ) for 18
Pd–Cl
Pd–N(12)
N(12)–Pd–C(1)
C(1)–Pd–N(19)
C(1)–Pd–Cl
2.298(4)
2.030(13)
88.5(6)
177.1(5)
88.2(5)
Pd–C(1)
Pd–N(19)
N(12)–Pd–N(19)
N(12)–Pd–Cl
N(19)–Pd–Cl
2.046(10)
2.200(10)
90.2(4)
172.8(3)
92.8(3)

6278
V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
-6
obtain crystals of 19 that are suitable for X-ray diffraction
studies in order to see whether any change in the structure
occurs as a result of the presence of the Pd+ species such as
a palladium-iron bond, analogous to those found for
[FcS2]Pd/Pt and [FcP2]Pd-type species [29].
In order to test the potential ethylene polymerisation
activity of 19, the palladium compound was dissolved in
DCM and exposed to 1 bar ethylene pressure for 1 h during
which time the orange solution turned green, and there was
no visible evidence of any solid polyethylene being formed.
Upon quenching the reaction with HCl the solution turned
lilac and upon standing became a pale yellow colour. An
aliquot was removed for GPC analysis which revealed that
no oligomers had been formed either, meaning that com-
pound 19 unfortunately showed no activity as an ethylene
polymerisation catalyst.
6.0x10
4.0x10
2.0x10
Compound 17
Compound 18
Compound 19
-6
-6
0.0
-6
-2.0x10
-4.0x10
-6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Voltage (V)
Fig. 7. Cyclic voltammogram of 17–19.
1.8. Redox chemistry
Although complex 19 showed no activity for ethylene
polymerisation, it was considered possible that the oxidised
ferrocenium species may exhibit activity. Investigations into
the electrochemistry of 19 and its precursors 17 and 18 were
therefore performed in order to determine what common
chemical oxidants were appropriate to generate the ferroce-
nium-active complex. The experiments were performed on
0.001 M dichloromethane solutions of compounds 17–19,
with supporting electrolyte 0.1 M nBu4NPF6, using Ag/
AgCl as the reference electrode, along with platinum work-
dation of the solvent at these high potentials cannot be
ruled out entirely.
Complexation to the palladium centre shifts the oxida-
tion potential to higher values, in other words the Fe(II)–
Fe(III) is harder to access due to the electron withdrawing
capacity of the palladium fragment. This means that FcPF6
may not be used to chemically oxidise compound 19 as it
has a lower redox potential than 19. Indeed such a large
anodic shift in the case of 19 may indicate some sort of
an interaction between the iron and the palladium centres,
that results in binding of ethylene to the palladium centre,
with a resultant positive charge on the iron. This may
explain the lack of ethylene polymerisation activity, and
the observed green-colour, as Fc+ species are typically
blue/green colour in solution.
ing and auxiliary electrodes, and a scan rate of 100 m V sꢁ1
.
The results of the electrochemical investigations are shown
in Table 6 and Fig. 7. In each case a quasi-reversible
(DEp < 0.1 V), one-electron transfer for the Fe(II)–Fe(III)
redox couple is observed. Upon incorporation of the palla-
dium centre in 18 there is a large shift in the standard redox
potential to a more positive voltage as compared to 17. This
shift allows observation of a small imine to amine reduction
peak at approximately 0.45 V that was previously masked
in 17 by the iron redox couple. A similar phenomenon has
been observed in other Fc-imine ligands where the imine–
amine reduction peak occurs at 0.16–0.19 V [22]. There is
another anodic shift for the positively charged palladium
centre in 19, which means that the iron redox couple occurs
very close to the limits of the solvent window. Extending the
scans to more positive potentials (1.4 V) gave some indica-
tion of another reduction peak (ꢂ1.1 V) that overlapped
with the Fe(III)–Fe(II) peak that may speculatively be
attributed to a Pd(II)–Pd(I) reduction, although the degra-
2. Experimental
General. All preparations were carried out under an
atmosphere of nitrogen using standard Schlenk techniques
or an inert atmosphere (nitrogen) glove box, unless other-
wise stated. All solvents were distilled over standard drying
agents under nitrogen directly before use and deoxygen-
ated. Silica gel (Kieselgel grade 60) was used for chromato-
graphic separations. All NMR spectra were recorded using
a Delta upgrade on a JEOL EX270 MHz spectrometer.
Chemical shifts are reported in d (ppm) using residual pro-
ton impurities in CDCl3 (1H d 7.25 ppm, 13C d 77.0 ppm)
as the reference solvent. Mass spectra were recorded using
positive FAB methods for organometallic compounds, and
EI for organic compounds, on a micromass Autospec spec-
trometer (primary ion beam 35 keV Cs+ and 3-nitrobenzyl
alcohol matrix) by Mr J. Barton, Imperial College. Micro-
analyses were carried out by Mr S Boyer of the London
Metropolitan University.
Table 6
Cyclic voltammetry studies on Fc, FcPF6, and compounds 17–19
Compound
EAp ðVÞ
ECp ðVÞ
E1/2 (V)
DEp (V)
Fc
0.489
0.460
0.516
0.778
0.858
0.403
0.368
0.425
0.692
0.768
0.441
0.414
0.471
0.735
0.813
0.096
0.092
0.091
0.086
0.090
FcPF6
17
18
Polyethylene analysis was performed at BP Chemicals
Limited by J. Boyle (NMR) and G. Audley (GPC). GPC
19

V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
6279
traces were recorded using PL gel 2 · mixed bed – D, 30 cm,
5 micron columns, trichlorobenzene eluent and a flow rate
of 1.0 ml/min at 150 ꢁC using a refractive index detector.
yield crystalline 3 (0.21 g, 76%). Anal. Calc. for
C37H27FeNO: C, 79.71; H, 4.85; N, 2.51. Found: C,
1
79.48; H, 4.31; N, 2.43%. H NMR (CDCl3): d 4.14 (s,
5H, C5H5), 4.26 (tr, 2H, C5H4), 4.54 (tr, 2H, C5H4), 5.41
(s, 1H, bridgehead CH C20H13) 6.96 (m, 8H, Ar–H), 7.16
(m, 1H, Ar–H), 7.22 (m, 11H, Ar–H), 7.41 (m, 4H, Ar–
H), 7.53 (m, 1H, Ar–H) , 8.55 (m, 1H, Ar–H) 8.77 (s,
1H, CH@N), 14.18 (s, 1H, OH); 13C NMR (CDCl3); d
55.30 (bridgehead-C C20H13), 58.5 (bridgehead-C
C20H13), 62.4 (C5H5-Cp), 67.7 (C5H4-Cp), 69.9 (C5H4-
Cp), 101.9 (ipso-C C5H4-Cp), 118.25, 120.54, 123.50,
124.37, 124.53, 124.89, 125.01, 128.24, 128.92, 131.11,
134.44, 145.55, 146.21, 159.72 (Ar–C), 161.06 (CH@N).
FAB+ve; m/z: 557 [M]+.
2.1. Starting materials
Starting materials were prepared according to adapted
literature procedures (referenced where appropriate), and
1
were characterised by H NMR and 13C NMR and mass
spectrometry where appropriate. All other chemicals were
purchased from Aldrich Chemical Company and used
without further purification unless stated. Research grade
ethylene (BOC) was used for all ethylene polymerisation
experiments. Methyl aluminoxane (MAO) was obtained
as a 1.6 M solution in toluene. Ferrocenylamine [30],
3-anthracenyl-2-hydroxybenzaldehyde, and 3-triptycenyl-
2-hydroxybenzaldehyde [8] were formed via modified
literature procedures. (TMEDA)NiMe2 was synthesised
in our laboratories by Dr. D.J. Jones. All other starting
materials were obtained from commercial sources. 1,10-
2.5. Synthesis of N-(2-hydroxy, 3,5-di-tert-butylbenzyli-
dene) ferroceneamine (4)
To a solution of aminoferrocene (1 g, 5 mmol) in tolu-
ene (20 ml) was added a solution of 3,5-ditertbutyl-2-
hydroxybenzaldehyde (1.16 g, 5 mmol) also in toluene.
The mixture was heated to 80 ꢁC for 3 h and the volatiles
removed in vacuo to yield a resulting red coloured solid.
The product 4 was recrystallised from ethanol (1.2 g,
57%). Anal. Calc. for C25H31FeNO: C, 71.94; H, 7.43; N,
3.36. Found: C, 71.89; H 7.58; N, 3.26%. 1H NMR
(CDCl3): d 1.33 (s, 9H, C(CH3)3), 1.46 (s, 9H, C(CH3)3),
4.18 (s, 5H, C5H5), 4.26 (t, 2H, C5H4), 4.59 (t, 2H,
C5H4), 7.14 (d, 1H, C6H2), 7.42 (d, 1H, C6H2), 8.64 (s,
1H, CH@N), 13.76 (s, 1H, OH); 13C NMR (CDCl3); d
29.6 (CH3), 31.6 (CH3), 34.2 (C(CH3)3), 35.1 (C(CH3)3),
62.4 (C5H5-Cp), 67.4 (C5H4-Cp), 69.8 (C5H4-Cp), 102.6
(ipso-C C5H4-Cp) 118.8 (C6H2), 125.8 (C6H2), 127.0
(C6H2), 136.9 (C6H2), 140.4 (C6H2), 157.9 (C6H2), 161.4
(CH@N). FAB+ve; m/z: 417 [M]+.
Bis(amino)ferrocene was formed using
procedure of Nesmeyanov [31].
a
modified
2.2. Synthesis of N-(2-hydroxybenzylidene)ferroceneamine
(1)
The experimental method of Floris [21] was used with
minor modifications as necessary to produce a red coloured
product (55%).
2.3. Synthesis of N-(2-hydroxy,3-anthracenylbenzylidene)
ferroceneamine (2)
To a solution of ferrocenylamine (0.65 g, 3.2 mmol) in
dichloromethane (20 ml) was added a solution of 3-anth-
racenyl-2-hydroxybenzaldehyde (0.96 g, 3.2 mmol) also in
dichloromethane (30 ml). The mixture was heated to reflux
for 5 h and the volatiles removed in vacuo to yield a red
coloured solid. The product was recrystallised from ethanol
to yield crystalline 2 (1.3 g, 84%). Anal. Calc. for
C31H23FeNO: C, 77.33; H, 4.78; N, 2.91. Found: C,
77.49; H 4.89; N, 2.84%. 1H NMR (CDCl3): d 4.15 (s,
5H, C5H5), 4.26 (t, 2H, C5H4), 4.54 (t, 2H, C5H4), 7.17
(m, 2H, C14H9), 7.47 (m, 2H, C14H9), 7.75 (d, 2H,
C14H9), 8.07 (d, 2H, C14H9), 8.55 (s, 1H, ipso-C14H9)
8.78 (s, 1H, CH@N), 13.72 (s, 1H, OH); 13C NMR
(CDCl3); d 62.4 (C5H5-Cp), 67.6 (C5H4-Cp), 69.6 (C5H4-
Cp), 101.9 (ipso-C C5H4-Cp), 118.9 (Ar–H), 119.9, 125.1,
125.5, 126.6, 126.9, 128.6, 130.4, 131.0, 131.5, 135.3,
159.1 (Ar–H), 159.9 (CH@N). FAB+ve; m/z: 481 [M]+.
2.6. Synthesis of N-(2-hydroxy,3-anthracenylbenzylidene)
ferroceneamine chromium dichloride (5)
To a solution of p-tolylCrCl2 Æ 3THF (0.18 g, 0.42 mmol)
in toluene (50 ml) was added a solution of 2 (0.2 g,
0.42 mmol) also in toluene (50 ml). The resulting solution
was stirred over a period of 3 h during which time the colour
had changed from red to brown. The volatiles were
removed in vacuo and the resulting red-brown solid washed
with pentane. The solid was dried under vacuum (0.21 g,
83%). Anal. Calc. for C31H22Cl2CrFeNO: C, 61.70; H,
3.65; N, 2.32. Found: C, 61.83; H 3.75; N, 2.19%. FAB+ve;
m/z: 1047 [(2)2CrCl]+, 1012 [(2)2Cr]+, 603 [(2)CrCl2]+.
2.7. In situ polymerisation using N-(2-hydroxy,3-anthracen-
ylbenzylidene)ferroceneamine chromium dichloride (5)
2.4. Synthesis of N-(2-hydroxy,3-triptycenylbenzylidene)
ferroceneamine (3)
In a Schlenk vessel 2 (0.0048 g, 0.01 mmol) was dis-
solved in toluene (25 ml) and added to p-tolylCrCl2 Æ 3THF
(0.0043 g, 0.010 mmol) also in toluene (25 ml). The reaction
mixture was stirred for 2 h and the volatiles removed in
An analogous procedure to that employed for 2 was uti-
lised using aminoferrocene (0.1 g, 0.50 mmol) and 3-tript-
ocenyl-2-hydroxybenzaldehyde (0.18 g, 0.48 mmol) to

6280
V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
vacuo. The product was redissolved in toluene (100 ml) and
100 equiv. of MAO (1.6 M in toluene) added. The Schlenk
vessel was exposed to 1 bar ethylene for 1 h. The polymer-
isation test was terminated by removal of the ethylene sup-
ply and addition of acidified methanol. The polymer was
filtered and dried in a vacuum oven overnight.
2.12. Synthesis of N,N0bis-(2-hydroxy-3-anthracenyl-
benzaldimine)ferroceneamine (11)
An analogous procedure to that employed for 10 was
utilised using 1,10-bis(amino)ferrocene (0.21 g, 1 mmol)
and
3-anthracenyl-2-hydroxybenzaldehyde
(0.58 g,
2 mmol) to yield crystalline 11 (0.65 g, 85%). Anal. Calc.
for C52H36FeN2O2: C, 80.41; H, 4.64; N, 3.61. Found:
C, 80.50; H, 4.82; N, 3.42%. H NMR (CDCl3): d 4.24
2.8. In situ polymerisation using N-(2-hydroxy,3-triptlcenyl-
benzylidene)ferroceneamine chromium dichloride (6) and
N-(2-hydroxy-3,5-di-tert-butylbenzylidene)ferroceneamine
chromium dichloride (7)
1
(t, 4H, C5H4 , J 1.9 Hz), 4.43 (tr, 4H, C5H4, J 2.0 Hz),
6.94 (m, 2H, Ar–H), 7.17 (m, 2H, Ar–H), 7.38 (m, 8H,
Ar–H), 7.70 (m, 4H, Ar–H), 8.10 (d, 4H, Ar–H), 8.56
(s, 2H Ar–H), 8.60 (s, 2H, C@NH), 13.65 (s, 2H, OH);
13C NMR (CDCl3); d 64.60 (C5H4), 69.40 (C5H4),
102.32 (ipso-C C5H4), 118.89, 125.13, 125.52, 126.58,
126.96, 128.62, 130.41, 131.27, 131.58, 135.42, 152.55,
159.09, 160.49, 162.11 (Ar–C), 166.07 (C@N). FAB+ve;
m/z: 776 [M]+. Crystal data for 11: C52H36N2O2-
An analogous procedure to that employed for 5 was
followed, using 3 (0.0055 g, 0.01 mmol) and p-tolyl-
CrCl2 Æ 3THF (0.0043 g, 0.010 mmol) to form 6, and react-
ing 4 (0.0042 g, 0.01 mmol) and p-tolylCrCl2 Æ 3THF
(0.0043 g, 0.010 mmol) to form 7.
ꢀ
˚
Fe Æ 2CH2Cl2, M = 946.53, triclinic, P1 (no. 2), a =
2.9. Synthesis of bis-N-(2-hydroxybenzylidene)-
ferroceneamine nickel(II) (8)
11.1166(8), b = 13.5027(8), c = 17.1512(9) A, a = 95.435(4),
3
˚
b = 106.702(5), c = 111.536(4)ꢁ, V = 2235.0(2) A , Z = 2,
Dc = 1.406 g cmꢁ3, l(Cu-Ka) = 5.261 mmꢁ1, T = 193 K,
orange/red platy prisms; 6608 independent measured
reflections, F2 refinement, R1 = 0.051, wR2 = 0.118, 5059
independent observed absorption-corrected reflections
[jFoj > 4r(jFoj), 2hmax = 120ꢁ], 631 parameters. CCDC
271797.
To a solution of (TMEDA)NiMe2 (0.1 g, 0.5 mmol) in
acetonitrile (20 ml) was added a solution of 1 (0.15 g,
0.5 mmol) also in acetonitrile (20 ml). The resulting orange
coloured solution was stirred for 2 h during which time a
red precipitate formed. The product was filtered and
washed with two portions of cold pentane (ꢁ30 ꢁC). The
resulting orange solid was dried under vacuum to give 8
(0.07 g, 21%). Characterising data were found to be in
agreement with the literature values [21].
2.13. Synthesis of N,N0 bis-(2-hydroxy-3-tertbutylben-
zaldimine)ferroceneamine (12)
An analogous procedure to that employed for 10 was
utilised using 1,10-bis(amino)ferrocene (0.21 g, 1 mmol)
and 3-tertbutyl-2-hydroxybenzaldehyde (0.46 g, 2 mmol)
to yield crystalline 12 (0.44 g, 68%). Anal. Calc. for
C32H36FeN2O2: C, 71.64; H, 6.72; N, 5.22. Found: C,
2.10. Synthesis of bis-N-(2-hydroxy-3anthracenyl-benzylid-
ene)ferroceneamine nickel(II) (9)
An analogous procedure to that employed for 8 was uti-
lised using (TMEDA)NiMe2 (0.1 g, 0.5 mmol) and 2
(0.24 g, 0.5 mmol). Anal. Calc. for C62H44Fe2N2NiO2: C,
73.01; H, 4.32; N, 2.75. Found: C, 73.49; H 4.89; N,
2.70%. FAB+ve; m/z: 1019 [M]+.
1
71.11; H, 6.72; N, 5.01%. H NMR (CDCl3): d 1.46 (s,
18H, CH3), 4.31 (t, 4H, C5H4), 4.56 (t, 4H, C5H4), 6.69
(m, 2H, C6H3), 6.92 (m, 2H, C6H3), 7.30 (m, 2H, C6H3),
8.46 (s, 2H, CH@N), 13.80 (s, 2H, OH); 13C NMR
(CDCl3); d 28.76 (C(CH3)3), 34.14 (C(CH3)3), 63.69
(C5H4), 68.39 (C5H4), 102.16 (ipso-C C5H4), 117.57,
118.73, 128.74, 129.06, 136.45, 159.28 (C6H3), 160.97
(CH@N). FAB+ve; m/z: 536 [M]+.
2.11. Synthesis of N,N0 bis-(2-hydroxybenzaldimine)
ferroceneamine (10)
To a solution of 1,10-bis(amino)ferrocene (0.21 g,
1 mmol) in dichloromethane (50 ml) was added salicylalde-
hyde (0.61 g, 5 mmol). The solution was refluxed for 1 h
and the volatiles removed in vacuo. The resulting oily solid
was recrystallised from ethanol to yield the burgundy col-
oured 10 (0.36 g, 85%). Anal. Calc. for C24H20FeN2O2:
C, 67.92; H, 4.72; N, 6.60. Found: C, 68.10; H, 4.40; N,
2.14. Synthesis of N,N0 bis-(2-hydroxy-3-tertbutyl-
benzaldimine)ferroceneamine TiCl2 (13)
A solution of 12 (0.49 g, 1 mmol) in toluene (30 ml) was
added dropwise to a solution of TiCl4THF2 (0.33 g
1 mmol) also in toluene (30 ml). The resulting brown col-
oured solution was stirred for 18 h and the solvent removed
under vacuum. The crude product was recrystallised from
dichloromethane:pentane (25:75) to give analytically pure
13 (0.43 g, 66%). Anal. Calc. for C32H34Cl2FeN2O2Ti: C,
58.81; H, 5.21; N, 4.29. Found: C, 58.87; H, 5.01; N,
1
6.40%. H NMR (CDCl3): d 4.32 (t, 4H, C5H4), 4.59 (t,
4H, C5H4), 6.72 (m, 4H, C6H4), 7.01 (m, 2H, C6H4), 7.17
(m, 2H, C6H4), 8.37 (s, 2H, N@CH), 13.05 (s, 2H, OH);
13C NMR (CDCl3); d 64.08 (C5H4), 69.09 (C5H4), 102.42
(ipso-C C5H4), 116.84, 118.85, 119.51, 131.10, 131.86,
160.42 (C6H4), 161.19 (C@N). FAB+ve; m/z: 424 [M]+.
1
3.86%. H NMR (CD2Cl2): d 1.59 (s, 9H, C(CH3)3), 4.31

V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
6281
(t, 4H, C5H4), 4.96 (t, 4H, C5H4), 7.12 (m, 2H, C6H3), 7.46
(m, 2H, C6H3), 7.66 (m, 2H, C6H3), 8.52 (s, 2H, CH@N);
13C NMR (CDCl3); d 30.59 (C(CH3)3), 35.69 (C(CH3)3),
66.81 (C5H4), 69.54 (C5H4), 110.64 (ipso-C C5H4), 122.93,
127.62, 133.28, 133.41, 137.71, 163.66 (C6H4), 168.68
(CH@N). FAB+ve; m/z: 653 [M+].
C5H4), 4.32 (m, H, C5H4), 4.52 (sept, H, CH(CH3)2), 4.54
(m, H, C5H4), 4.97 (sept, H, CH(CH3)2), 5.77 (m, H,
C5H4), 6.56 (t, H, C6H3), 6.79 (t, H, C6H3) 7.30 (m, 4H,
C6H3), 8.26 (s, H, CH@N), 8.51 (s, H, CH@N); 13C
NMR (C6D6); d 24.92, 25.10, 26.33, 26.41 (CH(CH3)2),
29.44, 29.98 (C(CH3)3), 34.57, 35.10 (C(CH3)3), 62.88
(C5H4), 64.38 (CH(CH3)2), 64.63, 66.57, 67.71, 68.24,
68.73 (C5H4), 69.07 (CH(CH3)2), 70.19, 10.47 (C5H4),
79.69, 107.95 (ipso-C C5H4), 111.88, 115.54, 116.95,
123.81, 124.26, 130.68, 131.30, 131.54, 132.74, 138.00,
2.15. Synthesis of N,N0 bis-(2-hydroxybenzaldimine)-
ferroceneamine Ti(iOPr)2 (14)
i
A solution of 10 (0.42 g, 1 mmol) in toluene (30 ml) was
added dropwise to a solution of Ti(iOPr)4 (0.28 g, 1 mmol)
also in toluene (30 ml). The resulting red coloured solution
was stirred for 18 h and the solvent removed under vac-
uum. The crude product was recrystallised from dichloro-
methane:pentane (25:75) to give analytically pure 14
(0.42 g, 71%). Anal. Calc. for C30H32FeN2O4Ti: C, 61.22;
139.72, 164.62 (C6H3). FAB+ve; m/z: 657 [M+ ꢁ Pr].
Crystal data for A: C50H39N5O5Cl2Fe2Ti2, M = 1068.26,
monoclinic, P21/c (no. 14), a = 14.889(3), b = 13.2811(13),
3
˚
˚
c = 23.064(2) A, b = 97.575(12)ꢁ, V = 4520.8(10) A , Z =
4, Dc = 1.570 g cmꢁ3
,
l(Mo-Ka) = 1.143 mmꢁ1
,
T =
203 K, dark brown blocks; 7960 independent measured
reflections, F2 refinement, R1 = 0.050, wR2 = 0.099, 5461
independent observed absorption-corrected reflections
[jFoj > 4r(jFoj), 2hmax = 50ꢁ], 595 parameters. CCDC
271798. Crystal data for B: C48H36N4O6Fe2Ti2 Æ 2CH2Cl2,
M = 1142.16, monoclinic, P21/n (no. 14), a = 10.7063(9),
1
H, 5.44; N, 4.76. Found: C, 61.06; H, 5.50; N, 4.71%. H
NMR (CDCl3): d 0.92 (m, 6H, CH(CH3)2), 1.51 (m, 6H,
CH(CH3)2), 3.62 (m, H, C5H4), 3.79 (m, H, C5H4), 3.85
(m, H, C5H4), 3.92 (m, H, C5H4), 3.96 (m, H, C5H4),
4.12 (m, H, C5H4), 4.65 (m, H, C5H4), 4.80 (sept, H,
CH(CH3)2), 5.18 (sept, H, CH(CH3)2), 5.79 (m, H,
C5H4), 6.98 (m, 4H, C6H4), 8.04 (s, H, CH@N), 8.12 (s,
˚
b = 13.5040(7), c = 16.3743(7) A, b = 95.637(5)ꢁ, V =
3
2355.9(3) A , Z = 2 [Ci symmetry], Dc = 1.610 g cmꢁ3
,
˚
l(Cu-Ka) = 10.170 mmꢁ1, T = 293 K, ruby red rhombs;
3497 independent measured reflections, F2 refinement,
R1 = 0.050, wR2 = 0.120, 2715 independent observed
i
H, CH@N). FAB+ve; m/z: 588 [M+ ꢁ Pr].
2.16. Synthesis of N,N0bis-(2-hydroxy-3-anthracenyl-
benzaldimine)ferroceneamine Ti(iOPr)2 (15)
absorption-corrected reflections [jFoj > 4r(jFoj), 2hmax
=
120ꢁ], 344 parameters. CCDC 271799.
An analogous procedure to that utilised for 14 was
employed using 11 (0.77 g, 1 mmol) and Ti(iOPr)4
(0.28 g,1 mmol) to yield 15 (0.72 g, 77%). Anal. Calc. for
C58H48FeN2O4Ti: C, 73.96; H, 5.10; N, 2.98. Found: C,
2.18. Synthesis of N,N0-bis(benzaldimine)ferroceneamine
(17)
Benzaldehyde (5 ml) was added to a toluene (50 ml)
solution of 1,1-bis(amino)ferrocene (0.25 g, 1.16 mmol).
The solution was stirred for 3 h during which the initial yel-
low colour became deep burgundy. The toluene and excess
benzaldehyde were removed in vacuo to yield a burgundy
solid 17, which was washed with pentane (20 ml). (0.34 g,
75%). Anal. Calc. for C24H20FeN2: C, 73.47; H, 5.10; N,
7.14. Found: C, 73.50; H, 5.10; N, 7.10%. 1H NMR
(CDCl3): d 4.29 (4H, t, C5H4), 4.61 (4H, t, C5H4), 7.57
(6H, m, C6H5), 7.63 (4H, m, C6H5), 8.47 (2H, s, NC@H);
13C NMR (CDCl3) d 64.42 (C5H4), 68.65 (C5H4), 105.55
(ipso-C C5H4), 127.97, 128.55, 130.31, 136.57 (C6H5),
158.31 (C@N). FAB+ve; m/z: 392 [M+].
1
73.94; H, 5.08; N, 2.88%. H NMR (C6D6): d ꢁ 0.65 (d,
3H, CH(CH3)2), ꢁ0.02 (d, 3H, CH(CH3)2), 0.68 (d, 3H,
CH(CH3)2), 3.44 (sept, H, CH(CH3)2), 4.05 (m, H,
C5H4), 4.12 (m, H, C5H4), 4.18 (m, H, C5H4), 4.23 (m,
2H, C5H4), 4.26 (sept, H, CH(CH3)2), 4.41 (m, H, C5H4),
4.19 (m, H, C5H4), 5.68 (m, H, C5H4), 6.18 (m, 2H, Ar–
H), 6.38 (m, 1H, Ar–H), 6.69 (m, H, Ar–H), 6.79 (m, 2H,
Ar–H), 7.34 (overlapping m, 6H, Ar–H), 7.58 (m, H, Ar–
H), 7.60 (m, H, Ar–H), 7.75 (m, H, Ar–H), 7.81 (m, H,
Ar–H), 7.86 (m, H, Ar–H), 7.90 (m, H, Ar–H), 7.98 (m,
H, Ar–H), 8.01 (m, H, Ar–H), 8.09 (m, 2H, Ar–H) 8.35
(s, H, CH@N), 8.46 (s, H, CH@N), 8.72 (m, 2H, Ar–H).
i
FAB+ve; m/z: 897 [M+ ꢁ Pr].
2.19. Synthesis of [1,10-bis(benzylimino)ferrocene]
palladium methyl chloride (18)
2.17. Synthesis of N,N0 bis-(2-hydroxy-3-tertbutyl-
benzaldimine)ferroceneamine Ti(iOPr)2 (16)
Compound 17 (0.35 g, 0.9 mmol) and Pd(COD)MeCl
[26] (0.24 g, 0.9 mmol) were dissolved in dry dichloro-
methane and stirred at 40 ꢁC for 72 h. The solvent was
removed in vacuo to leave a bright orange/red solid that
was washed with cold pentane to remove excess COD.
The solid was then redissolved in dichloromethane/pen-
tane, hot filtered and cooled to ꢁ30 ꢁC to obtain the puri-
fied product 18 (0.23 g, 46%). Anal. Calc. for
An analogous procedure to that utilised for 14 was
employed using 12 (0.056 g, 0.1 mmol) and Ti(iOPr)4
(0.028 g, 0.1 mmol) to yield 16 (0.032 g, 46%). H NMR
(CD2Cl2): d 0.82 (m, 6H, CH(CH3)2), 1.02 (d, 3H,
CH(CH3)2), 1.15 (s, 9H, C(CH3)3), 1.23 (d, 3H,
CH(CH3)2), 1.47 (s, 9H, C(CH3)3), 3.77 (m, H, C5H4),
4.03 (m, H, C5H4), 4.14 (m, H, C5H4), 4.23 (m, H,
1

6282
V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
C25H23ClFeN2Pd: C 54.68, H 4.22, N 5.10. Found: C
54.49, H 4.18, N 4.98%. 1H NMR (CDCl3): d 0.81 (s,
3H, PdMe), 4.23 (br s, 2H, C5H4), 4.46 (br s, 2H, C5H4),
4.53 (br s, 2H, C5H4), 5.82 (br s, 1H, C5H4), 6.18 (br s,
1H, C5H4), 7.19 (m, 2H, Ar–H), 7.45 (m, 4H, Ar–H),
8.25 (s, 1H, HC@N), 8.33 (s, 1H, HC@N), 8.58 (m, 2H,
Ar–H), 8.68 (m, 2H, Ar–H); 13C NMR (CDCl3,): d 7.54
(PdMe), 59.42, 60.68, 67.43, 70.23, 71.63, 72.72 (C5H4),
110.45, 110.92 (ipso- C5H4), 128.45, 128.71 (para-C, Ar),
130.21, 130.51 (meta-C, Ar), 131.36, 131.85 (ortho-C, Ar),
132.64, 134.34 (ipso-C, Ar), 164.76, 164.94 (C@N). FAB
+ve; m/z: 548 (M+), 533 (M ꢁ Me+), 513 (M ꢁ Cl+), 497
(M ꢁ MeCl+), 392 (M ꢁ PdMeCl+). Crystal data for 18:
C25H23N2ClFePd, M = 549.15, orthorhombic, Pbca (no.
the electrolyte in dichloromethane. The sample solution
was approximately 1 mM concentration in this electrolyte.
The working electrode was platinum – circular and 1 mm in
diameter, the reference was an Ag/AgCl electrode and the
counter electrode was a platinum wire. The cyclic voltam-
mograms were measured at rates of 100 m V sꢁ1 and the
potentials calculated using ferrocene as a comparison.
References
[1] S. Matsui, M. Mitani, J. Saito, Y. Tohi, H. Makio, N. Matsuawa, Y.
Takagi, K. Tsuru, M. Nitabaru, T. Nakano, H. Tanaka, N. Kashiwa,
T. Fujita, J. Am. Chem. Soc. 123 (2001) 6847.
[2] S. Matsui, M. Mitani, J. Saito, Y. Tohi, H. Makio, H. Tanaka, T.
Fujita, Chem. Lett. (1999) 1263.
[3] Y. Yoshida, J. Mohri, S. Ishii, M. Mitani, J. Saito, S. Matsui, H.
Makio, T. Nakano, H. Tanaka, M. Onda, Y. Yamamoto, A. Mizuno,
T. Fujita, J. Am. Chem. Soc. 126 (2004) 12023.
[4] S. Matsui, Y. Tohi, M. Mitani, J. Saito, H. Makio, H. Tanaka, M.
Nitabaru, T. Nakano, T. Fujita, Chem. Lett. (1999) 1065.
[5] Y. Inoue, T. Nakano, H. Tanaka, N. Kashiwa, T. Fujita, Chem.
Lett. (2001) 1060.
˚
61), a = 16.138(4), b = 11.550(3), c = 23.677(7) A, V =
3
4413(2) A , Z = 8, Dc = 1.653 g cmꢁ3
,
l(Cu-Ka) =
˚
13.060 mmꢁ1, T = 293 K, orange needles; 3270 indepen-
dent measured reflections, F2 refinement, R1 = 0.066,
wR2 = 0.118, 1520 independent observed absorption-
corrected reflections [jFoj > 4r(jFoj), 2hmax = 120ꢁ], 249
parameters. CCDC 271800.
[6] T.R. Younkin, E.F. Connor, J.I. Henderson, S.K. Friedrich, R.H.
Grubbs, Science 287 (2000) 460.
[7] C.M. Wang, S. Friedrich, T.R. Younkin, R.T. Li, R.H. Grubbs,
D.A. Bansleben, Organometallics 17 (1998) 3149.
[8] D.J. Jones, V.C. Gibson, S.M. Green, P.J. Maddox, Chem. Commun.
(2002) 1038.
2.20. Synthesis of [{1,10-bis(benzylimino)ferrocene}
Pd+Me][BAFꢁ](MeCN) (19)
[Fc(NBenz)2]PdMeCl 18 (0.15 g, 0.27 mmol) and
NaBAF [28] (0.242 g, 0.27 mmol) were weighed out into
a Schlenk tube in the glovebox, then dissolved in DCM
(30 ml) and dry degassed MeCN (2 ml) added. The dark
purple solution was heated overnight at 40 ꢁC, then the sol-
vent removed and the solid washed with pentane to give the
product, a brown/orange/yellow lustrous solid 19 (0.23 g,
59%). Anal. Calc. for C59H38BF24FeN3Pd: C 49.97, H
2.70, N 2.96. Found: C 49.86, H 2.74, N 2.81%. 1H
NMR (CDCl3,): d 0.91 (s, 3H, PdMe), 2.26 (s, 3H, MeCN),
4.28 (m, 2H, C5H4), 4.46 (m, 2H, C5H4), 4.60 (m, 2H,
C5H4), 5.44 (m, 1H, C5H4), 5.48 (m, 1H, C5H4), 7.27 (m,
2H, Ar–H), 7.44 (m, 4H, Ar–H), 7.50 (br m, 4H, para-H
BAF), 7.70 (br m, 8H, ortho-H BAF) 8.08 (m, 2H, Ar–
H), 8.22 (s, 1H, HC@N), 8.32 (m, 2H, Ar–H), 8.41 (s,
1H, HC@N); 13C NMR (CDCl3): d ꢁ 4.40 (PdMe), 3.02
(MeCN), 60.81, 61.94, 68.65, 68.70, 69.98, 70.38, 70.92
(C5H4), 109.64, 109.80 (ipso-C5H4), 117.49 (para-C BAF),
121.32, 123.49, 125.66, 127.82 (CF3 BAF), 128.56, 128.82,
129.07, 129.75, (meta-C BAF), 128.96, 129.27 (para-C,
Ar), 130.14, 130.44 (meta-C, Ar), 132.92, 133.27 (ortho-C,
Ar), 134.02 (ipso-C, Ar), 134.81 (ipso-C, Ar, and ortho-C
BAF), 161.13, 161.53, 161.92, 162.32 (ipso-C BAF),
167.14, 167.76 (C@N). FAB +ve m/z: 513 (M+), 497
(M ꢁ Me+), 392 (M ꢁ PdMe+).
[9] J.P. Corden, W. Errington, P. Moore, M.G.H. Wallbridge, Chem.
Commun. (1999) 323.
[10] A. Bhaw-Luximon, D. Jhurry, N. Spassky, Polymer Bull. 44 (2000) 31;
M. Wisniewski, A. Le Borgne, N. Spassky, Macromol. Chem. Phys.
198 (1997) 1227;
N. Nomura, R. Ishii, M. Akakura, K. Aoi, J. Am. Chem. Soc. 124
(2002) 5938;
N. Spassky, M. Wisniewski, C. Pluta, A. Le Borgne, Macromol. Chem.
Phys. 197 (1996) 2627;
T.M. Ovitt, G.W. Coates, J. Am. Chem. Soc. 121 (1999) 4072;
T.M. Ovitt, G.W. Coates, J. Polym. Sci., Polym. Chem. 38 (2000) 4686;
C.P. Radano, G.L. Baker, M.R. Smith, J. Am. Chem. Soc. 122 (2000)
1552;
Z. Zhong, P.J. Dijkstra, J. Feijen, Angew. Chem., Int. Ed. 41 (2002)
4510;
Z. Tang, X. Chen, X. Pang, Y. Yang, X. Pang, J. Sun, X. Zhang, X.
Jing, J. Polym. Sci. Polym. Chem. 42 (2004) 5974.
[11] P.A. Cameron, D. Jhurry, V.C. Gibson, A.J.P. White, D.J. Williams,
Macromol. Rapid Commun. 20 (1999) 616.
[12] For a detailed literature review see A. Togni, T. Hayashi (Eds.),
Ferrocenes: Homogeneous Catalysis – Organic Synthesis – Materials
Science, VCH, Weinheim, Germany, 1995;
A. Togni, R.L. Halterman (Eds.), Metallocenes, Wiley-VCH, Wein-
heim, Germany, 1998.
[13] For a comprehensive overview of ferrocene and other metallocene
chemistry see N.J. Long, in: Metallocenes: An Introduction to
Sandwich Complexes, Blackwell Science, Oxford, 1998.
[14] K.-P. Stahl, G. Boche, W. Massa, J. Organomet. Chem. 277 (1984)
113;
H. Plenio, D. Burth, Organometallics 15 (1996) 4054;
H. Plenio, D. Burth, Angew. Chem., Int. Ed. Engl. 34 (1995) 800;
B. Neumann, U. Siemeling, H.-G. Stammler, U. Vorfeld, J.G.P.
Delis, P.W.N.M. van Leeuwen, K. Vrieze, J. Fraanje, K. Goubitz, F.
Fabrizi de Biani, P. Zanello, J. Chem. Soc., Dalton Trans. (1997)
4705;
2.21. Cyclic voltammetry
In all cases measurements were performed using an
Autolab PGSTAT12 potentiometer with Gpes software;
with 0.1 M tetrabutylammonium-hexafluorophophate as
M. Herberhold, M. Ellinger, W. Kremnitz, J. Organomet. Chem. 241
(1983) 227;

V.C. Gibson et al. / Journal of Organometallic Chemistry 690 (2005) 6271–6283
6283
J.D. Carr, S.J. Coles, M.B. Hursthouse, M.E. Light, E.L. Munro,
J.H.R. Tucker, J. Westwood, Organometallics 19 (2000) 3312;
J.D. Carr, S.J. Coles, M.B. Hursthouse, J.H.R. Tucker, J. Organo-
met. Chem. 637–639 (2001) 304;
[20] V.C. Gibson, S. Mastroianni, C. Newton, C. Redshaw, G.A. Solan,
A.J.P. White, D.J. Williams, J. Chem. Soc., Dalton. Trans. (2000) 1969.
[21] M. Bracci, C. Ercolani, B. Floris, M. Bassetti, A. Chiesi-Villa, C.
Gaustini, J. Chem. Soc., Dalton. Trans. (1990) 1357.
B. Wrackmeyer, H.E. Maisel, M. Herberhold, J. Organomet. Chem.
637–639 (2001) 727;
U. Siemeling, U. Vorfield, B. Neumann, H.-G. Stammler, P. Zanello,
F. Fabrizi de Biani, Eur. J. Inorg. Chem. 38 (1999) 1;
K. Kavallieratos, S. Hwang, R.H. Crabtree, Inorg. Chem. 38 (1999)
5184;
[22] V.C. Gibson, C.M. Halliwell, N.J. Long, P.J. Oxford, A.M. Smith,
A.J.P. White, D.J. Williams, J. Chem. Soc., Dalton Trans. (2003) 918.
[23] (a) K. Nakajima, C. Sasaki, M. Kojima, T. Aoyama, S. Ohba, Y.
Saito, J. Fujita, Chem. Lett. (1987) 2189;
(b) T. Aoyama, S. Ohba, Y. Saito, C. Sasaki, M. Kojima, J. Fujita,
K. Nakajima, Acta Crystallogr., Sect. C. 44 (1988) 1309.
I.R. Butler, J. Organomet. Chem. 552 (1997) 53;
S. Bradley, P.C. McGowan, K.A. Oughton, M. Thornton-Pett, M.E.
Walsh, Chem. Commun. (1999) 77;
[24] Y.N. Belokon, S. Caveda-Cepas, B. Green, N.S. Ikonnikov, V.N.
Khrustalev, V.S. Larichev, M.A. Moscalenko, M. North, C. Orizu,
V.I. Tararov, M. Tasinazzo, G.I. Timofeeva, L.V. Yashkina, J. Am.
Chem. Soc. 121 (1999) 3968.
[25] M. Tsuchimoto, Bull. Chem. Soc. Jpn. 74 (2001) 2101.
[26] R.E. Rulke, J.M. Ernsting, A.L. Spek, C.J. Elsevier, P.W.N.M. van
¨
Leeuwen, K. Vrieze, Inorg. Chem. 32 (1993) 5769.
[27] Z. Weng, L.-L. Koh, T.S.A. Hor, J. Organomet. Chem. 689 (2004) 18.
[28] M. Brookhart, B. Grant, A.F. Volpe, Organometallics 11 (1992)
3920.
[29] S. Akabori, T. Kumagai, T. Shirahige, S. Sato, K. Kawazoe, C.
Tamura, M. Sato, Organometallics 6 (1987) 2105;
O.V. Gusev, A.M. Kalsin, M.G. Peterleitner, P.V. Petrovski, K.A.
Lyssenko, N.G. Akhmedov, C. Bianchini, A. Meli, W. Oberhauser,
Organometallics 21 (2002) 3637.
S. Bradley, K.D. Camm, X. Liu, P.C. McGowan, R. Mumtaz, K.A.
Oughton, T.J. Podesta, M. Thornton-Pett, Inorg. Chem. 41 (2002) 715.
[15] A. Shafir, M.P. Power, G.D. Whitener, J. Arnold, Organometallics
19 (2000) 3978;
A. Shafir, D. Fiedler, J. Arnold, J. Chem. Soc., Dalton Trans. (2002)
555.
[16] A. Shafir, M.P. Power, G.D. Whitener, J. Arnold, Organometallics
20 (2001) 1365;
A. Shafir, J. Arnold, Organometallics 22 (2003) 567.
[17] U. Siemeling, O. Kuhnert, B. Neumann, A. Stammler, H.-G.
Stammler, B. Bildstein, M. Malaun, P. Zanello, Eur. J. Inorg. Chem.
(2001) 913.
[18] V.C. Gibson, N.J. Long, E.L. Marshall, P.J. Oxford, A.J.P. White,
D.J. Williams, J. Chem. Soc., Dalton Trans. (2001) 1162.
[19] M. Herberhold, Angew. Chem., Int. Ed. Engl. 41 (2002) 956.
[30] D. van Leusen, B. Hessen, Organometallics 20 (2001) 224.
[31] A.N. Nesmeyanov, V.N. Drozd, V.A. Sazonova, Dokl. Akad. Nauk
SSSR 150 (1963) 321.