Cobalt chloride complexes of N3 and N4 donor ligands

Article abstract of DOI:10.1002/ejic.200390090

A number of cobalt(II) chloride complexes of pyridine/amine N₃ and N₄ donors have been prepared and structurally characterised. Even though they are paramagnetic, assignment of the ₁H NMR signals was possible in several cases. With the exception of tpa, which formed a cationic monochloride complex, all new complexes have cis-coordinated chlorides. Nevertheless, only the relatively rigid pyp ligands gave rise to (low) ethene polymerisation activity on activation with MAO. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejic.200390090

FULL PAPER
Cobalt Chloride Complexes of N3 and N4 Donor Ligands
T. Martijn Kooistra,[a] Koen F. W. Hekking,[a] Quinten Knijnenburg,[a] Bas de Bruin,[a]
[a]
[a]
[a]
[a]
´
Peter H. M. Budzelaar, Rene de Gelder, Jan M. M. Smits, and Anton W. Gal*
Keywords: Cobalt / N Ligands / Polymerisation
A number of cobalt(II) chloride complexes of pyridine/amine
N3 and N4 donors have been prepared and structurally char-
Nevertheless, only the relatively rigid pyp ligands gave rise
to (low) ethene polymerisation activity on activation with
acterised. Even though they are paramagnetic, assignment MAO.
of the 1H NMR signals was possible in several cases. With
the exception of tpa, which formed a cationic monochloride
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
complex, all new complexes have cis-coordinated chlorides.
Germany, 2003)
Introduction
the optimal coordination geometries for olefin polymeris-
ation, we decided to study a range of flexible pyridine/
amine complexes and compare these with the more rigid
dip complexes. The ligands are of the picolylamine type and
are multidentate, with amine (Nam) and pyridine (Npy) ni-
trogens (Figure 1).[4]
Cobalt complexes with nitrogen donor ligands have been
used to mimic biological systems such as vitamin B-12 co-
enzyme.[1] They are also of interest in the context of dioxy-
gen activation at transition metals.[2] Recently they have at-
tracted attention as precatalysts for olefin polymerisation.[3]
Typically, the precatalysts are the cobalt() chloride com-
plexes of diiminepyridine (dip) ligands (see Figure 1). Struc-
tural characterisation of these complexes requires X-ray
analysis because the paramagnetic CoII centre complicates
NMR studies. In the hope of obtaining some insight into
For polymerisation the two chlorides must be converted
into an alkyl chain and a vacant coordination site cis to it.
In five-coordinate complexes of dpa-type N3 ligands, the
small bite angles of the two five-membered chelate rings
should disfavour equatorial coordination of all three nitro-
gens in a trigonal-bipyramidal coordination geometry; this
leaves the fac and mer arrangements shown in Figure 2,
both of which have a cis orientation of the remaining coor-
dination sites (for the more rigid dip ligands, only the mer
arrangement is accessible). Of the N4 ligands, pyp is the
least flexible one. In practice, it strongly prefers a cis-octa-
hedral coordination geometry (Figure 2),[5] although other
arrangements have been observed.[6,7] Six-coordinate com-
plexes of tpa will also have a cis-octahedral structure; in
addition, this ligands supports five-coordination, usually
with an apical Nam donor (Figure 2). Finally, bpc-type li-
gands have been observed in three distinct coordination
geometries shown in Figure 2: planar (‘‘N4’’), symmetric
cis-octahedral (‘‘N2/N2’’) and asymmetric cis-octahedral
(‘‘N3/N’’). The different coordination geometries possible
for the N3 and N4 donors also result in different patterns
of Nam and Npy donors trans to the two sites potentially
involved in polymerisation.
Results and Discussion
Synthesis
Figure 1. Pyridine/amine ligands used in this work; the dip ligand
has been included for comparison
The N3 and N4 ligands used have all been reported previ-
ously.[8] Their CoCl2 complexes were easily obtained by re-
action with CoCl2·6H2O in the appropriate solvent.
[a]
Toernooiveld 1, 6525 ED Nijmegen, The Netherlands
E-mail: gal@sci.kun.nl
648
 2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ1948/03/0204Ϫ0648 $ 20.00ϩ.50/0 Eur. J. Inorg. Chem. 2003, 648Ϫ655

Cobalt Chloride Complexes of N3 and N4 Donor Ligands
FULL PAPER
Figure 3. X-ray structure of one of the two independent cations
؉
˚
1 ; selected bond lengths [A] and angles [°]: Co1AϪN1A 2.028(15),
Co1AϪN2A 2.038(14), Co1AϪN3A 2.055(14), Co1AϪN4A
2.240(14), Co1AϪCl1A 2.284(5), N1AϪCo1AϪN4A 76.3(6),
N2AϪCo1AϪN4A 78.8(5), N3AϪCo1AϪN4A 76.7(5); corres-
ponding values for the second cation: Co1BϪN1B 2.072(15),
Co1BϪN2B 2.075(16), Co1BϪN3B 2.055(14), Co1BϪN4B
2.229(15), Co1BϪCl1B 2.280(6), N1BϪCo1BϪN4B 77.9(6),
N2BϪCo1BϪN4B 76.2(6), N3BϪCo1BϪN4B 77.1(5); corres-
ponding values for [(tpa)Co(MeCN)]2ϩ (ref.[9]): CoϪN1 2.041(3),
CoϪN2 2.037(3), CoϪN3 2.053(3), CoϪN4 2.177(3), CoϪN5
2.046(3),
N1ϪCoϪN4
79.1(1),
N2ϪCoϪN4
78.9(1),
N3ϪCoϪN4 78.7(1)
Figure 2. Coordination geometries of N3 and N4 ligands
[(bpc)CoCl2] (5) and [(Me-bpc)CoCl2] (6)
Both bpc complexes adopt a C2-symmetric, cis-octahed-
ral N2/N2 geometry (Figure 4). These two are the first ex-
amples of bis(picolylamino)ethylenediamine CoCl2 com-
plexes to be characterised by an X-ray structure determina-
tion. The structure of [(bpc)CrCl2]ϩ (an analogue of 5) has
been reported.[10] In addition, several metal dichloride com-
Whereas most of the complexes are air stable, [(bpc)CoCl2]
(5) proved very sensitive to air and water. This complex was
therefore synthesised by reaction of the ligand with dry
CoCl2 in anhydrous acetonitrile. The related complex [(Me-
bpc)CoCl2] (6) decomposes only very slowly in the presence
of air and water.
Crystal Structures of N4-Ligand Complexes
[(tpa)CoCl]2CoCl4
We expected to prepare an octahedral complex from tpa
and CoCl2·6H2O. However, X-ray analysis showed that the
five-coordinate cation [(tpa)CoCl]ϩ (1ϩ) had formed (Fig-
2Ϫ
ure 3), with CoCl4 as the counterion. The unit cell con-
tains two crystallographically independent cations 1ϩ. Each
of these has an almost C3-symmetric trigonal-bipyramidal
coordination geometry with Nam and ClϪ at the two axial
positions. The CoII centre is displaced approximately 0.5 A
˚
from the equatorial N3 plane in the direction of the chlor-
ide.
The structure of 1ϩ is very similar to that of the dicat-
ionic complex [(tpa)Co(MeCN)]2ϩ (2) reported by Karlin.[9]
The CoϪNpy distances of 1ϩ and 22ϩ are identical within
experimental error. The CoϪNam distance for 1ϩ is longer
than that for 22؉, as expected on the basis of the larger
trans-influence of ClϪ compared to acetonitrile.
The structures of [(tBu-pyp)CoCl2] (3)[5d] and [(Me-pyp)-
CoCl2] (4)[5a] have been reported before. Both have a cis-
octahedral coordination; the steric bulk of the tBu substitu-
ents in 3 forces the ligand to twist, lowering the symmetry
from C2v (as observed in 4) to C2.
˚
Figure 4. X-ray structures of 5 and 6; selected bond lengths [A] and
angles [°] for 5: Co1ϪN1 2.187(4), Co1ϪN2 2.205(4), Co1ϪCl1
2.4187(14), N1ϪCo1ϪN1a 170.1(2), N2ϪCo1ϪN2a 79.8(2),
Cl1ϪCo1ϪCl1a 98.24(8) (atoms labeled a symmetry-related by Ϫx,
y, 1/2 Ϫ z); corresponding values for 6: CoϪN1 2.147(8), CoϪN2
2.134(9), CoϪN3 2.255(9), CoϪN4 2.236(7), CoϪCl1 2.436(3),
CoϪCl2 2.435(3), N1ϪCoϪN2 173.1(4), N3ϪCoϪN4 78.7(3),
Cl1ϪCoϪCl2 98.24(13)
Eur. J. Inorg. Chem. 2003, 648Ϫ655
649

A. W. Gal et al.
FULL PAPER
plexes of bis(picolylamino)ethylenediamine ligands lacking
the cyclohexane ring have been structurally charac-
terised.[11] A number of bpc and Me-bpc complexes of co-
balt with counterions other than chloride have also been
reported; all of these contain CoIII. Remarkably, all previ-
ously reported bpc cobalt complexes have the N3/N coor-
dination geometry,[12] whereas the three Me-bpc cobalt
complexes reported so far all have the N2/N2 arrange-
ment.[13] Considering the wider class of bis(picolyl)ethylene-
diamine cobalt complexes, there are only a few examples
of the N2/N2 arrangement: these include the more highly
substituted system A[14] and the atypical, oxalato-bridged
dimer B.[15]
˚
Figure 5. X-ray structure of 7; selected bond lengths [A] and angles
[°]: CoϪN1 2.178(2), CoϪN2 2.182(2), CoϪN3 2.124(2), CoϪCl1
2.2999(9),
CoϪCl2
2.2784(9),
N1ϪCoϪN2
157.98(10),
N3ϪCoϪCl1 111.33(6), N3ϪCoϪCl2 114.24(6), Cl1ϪCoϪCl2
134.41(4); corresponding values for (Me-dpa)NiBr2 (ref.[17]):
NiϪN1 2.042(10), NiϪN2 2.050(11), NiϪN3 2.007(14), NiϪBr1
2.473(3), NiϪBr2 2.494(3), N1ϪNiϪN2 163.6(5), N3ϪNiϪBr1
112.1(6), N3ϪNiϪBr2 98.7(5), Br1ϪNiϪBr2 149.2(1)
Judging from these examples, the N3/N arrangement is
preferred over N2/N2 by only a small margin; substituents
at the backbone may cause increased steric hindrance at the
‘‘planar’’ Nam and so promote the more extensive folding
of the N2/N2 geometry. The N2/N2 geometry found in the
oxalato-bridged dimer might be caused by steric hindrance
between the monomer units in the N3/N geometry. Finally,
only a single example of the N4 arrangement has ever been
reported,[16] and in that case (C, above) the geometry is
probably enforced by the rigid ligand backbone.
Returning to our CoCl2 complexes, it seems remarkable
that even bpc adopts the N2/N2 structure here. We believe
that the larger size of CoII (compared to CoIII) disfavours
formation of the N3/N structure and hence is responsible
for shifting the balance towards N2/N2.
In contrast to 3, where the two enantiomeric conforma-
tions of the ligand are only slightly different and will in-
terconvert in solution, the chiral geometry in 5 and 6 is
locked through the coordination of the two pyridyl groups.
On going from 5 to 6, the introduction of the Nam methyl
substituent results in a longer CoϪNam distance, which is
compensated for by a strengthening of the CoϪNpy interac-
tion. In addition, the two pyridyl fragments become more
nearly orthogonal (angle between planes: 121° and 104°,
respectively). The rather similar CoϪCl distances in com-
plexes 3؊6 indicate that the trans-influence of Nam and Npy
on the chlorides is similar.
[(Me-dip)CoCl2] (8)
Complex 8 (Figure 6) has a trigonal-bipyramidal geo-
metry in the solid state, like its Fe analogue[18] and dpa
complex 7. In contrast, the CoCl2 and FeCl2 complexes of
the isopropyl-substituted dip ligand have structures which
are more nearly square pyramidal.[19] Apparently, the bulk-
ier aryl groups promote this kind of distortion; a similar
effect has been observed in a series of β-diiminate com-
plexes.[20] In both dpa and dip complexes, the equatorial
nitrogen is closer to the metal centre than the axial nitro-
gens. This effect is most pronounced for the dip complex,
possibly due to geometric constraints of the ligand back-
bone.
Crystal Structures of N3-Ligand Complexes
˚
Figure 6. X-ray structure of 8; selected bond lengths [A] and angles
[(Me؊Bz؊dpa)CoCl2] (7)
[°]: CoϪN1 2.247(2), CoϪN2 2.265(2), CoϪN3 2.0371(19),
CoϪCl1 2.2578(7), CoϪCl2 2.2730(7), N1ϪCoϪN2 150.42(7),
N3ϪCoϪCl1 119.90(6), N3ϪCoϪCl2 127.14(6), Cl1ϪCoϪCl2
112.92(3); corresponding values for Fe analogue (ref.[18]): FeϪN1
2.271(6), FeϪN2 2.266(5), FeϪN3 2.110(6), FeϪCl1 2.312(2),
FeϪCl2 2.278(2), N1ϪFeϪN2 145.5(2), N3ϪFeϪCl1 118.9(2),
N3ϪFeϪCl2 131.3(2), Cl1ϪFeϪCl2 109.9(1)
Complex 7 (Figure 5) has a pseudo trigonal-bipyramidal
coordination geometry, with the two axial positions occu-
pied by the two Npy atoms. The rather similar crystal struc-
ture of [(Me-dpa)NiBr2] has previously been reported.[17]
650
Eur. J. Inorg. Chem. 2003, 648Ϫ655

Cobalt Chloride Complexes of N3 and N4 Donor Ligands
FULL PAPER
Five versus Six Coordination
Complex 3 reveals a nearly axial spectrum with no re-
solved hyperfine couplings, whereas the spectrum of 7 is
rhombic and reveals hyperfine coupling with the cobalt
nucleus at one of the g-tensors (Co, I ϭ 7/2, 102 G). The
apparent symmetry from the EPR spectra seems to correl-
ate well with the cobalt coordination environments seen in
the X-ray structures. The approximate real g values and
E/D rhombicity factors derived from simulation of the ex-
perimental EPR spectra are g11 ϭ 2.17, g22 ϭ g33 ϭ 2.30,
E/D ϭ 0 for complex 3 and g11 ϭ 2.03, g22 ϭ 2.05, g33 ϭ
2.20, E/D ϭ 0.17 for complex 7. Apart from establishing
the spin-state (S ϭ 3/2) and the apparent coordination en-
vironment symmetry, these very broad EPR spectra are not
very informative.
The present work shows examples of both five- and six-
coordinate structures. It seems that the ligand has a large
influence on the preference for five- or six-coordination.
The N3 donor complexes 7 and 8 are naturally five-coordin-
ate, but even the cobalt chloride complex of tpa prefers an
ionic five-coordinate structure (1ϩ). Five-coordination must
be very favourable over six-coordination here (compared to
pyp and bpc ligands) since even if a large excess of ligand
is used, one-third of the cobalt is found as the counterion
[CoCl4]2Ϫ and free ligand remains. When NH4PF6 is then
added to the reaction mixture, all cobalt is complexed by
the ligand and the (presumably also five-coordinate) com-
plex [1]PF6 can be isolated. The other N4 complexes all
seem to prefer six-coordinate structures, although for com-
plex 6 five-coordination in solution cannot be completely
ruled out (see NMR section below). For 4 the coordination
behaviour in solution has been described in the literature.[5a]
Conductivity studies performed by Krüger showed that one
chloride is exchanged for a coordinated solvent molecule in
methanol, but only partial dissociation of one chloride
takes place in acetonitrile. Although acetonitrile is generally
considered to be a better ligand than methanol, methanol
is very well suited for the solvation of chloride anions. This
may be the driving force in the dissociation of chloride from
4. The NMR spectrum of 3 shows a similar solvent depend-
ence and suggests that this complex can also undergo chlor-
ide substitution in methanol. Clearly, complexes of pyp and
bpc prefer six-coordination, although a chloride can some-
times be displaced by a polar solvent molecule. It is possible
that five-coordination only occurs when the ligand can eas-
ily accommodate a trigonal-bipyramidal arrangement.
NMR Analysis
1
A complete interpretation of the paramagnetic H NMR
spectrum of [(tpa)Co(CH3CN)]2ϩ is presented in the report
of Karlin,[9] and this helped us to assign all the peaks in the
spectrum of [(tpa)CoCl]ϩ (1ϩ) on the basis of peak posi-
tions, linewidths and integrals. The signals from pyridine-
H3, -H4, and -H5 are distinctly sharper than those of pyrid-
ine-H6 and the methylene protons. This seems to be the
case for most pyridine-amine cobalt complexes studied
here. Because the NMR shifts of [1]PF6 and [1]2CoCl4 are
identical, the same trigonal bipyramidal structure is as-
sumed for both. Unfortunately, we did not succeed in grow-
ing X-ray quality crystals of [(dpa)CoCl2] (9). However, its
spectrum shows enough of
a similarity to that of
[(tpa)CoCl]ϩ to allow a complete interpretation. For
(MeϪBzϪdpa)CoCl2 (7), however, the extra signals from
the benzyl substituent complicate the NMR spectrum too
much to allow unambiguous assignments. Also, we ob-
served no signals for the pyridine-H3 and -H5 protons be-
tween δ ϭ 40 and 60 ppm, although this is the most consist-
ent feature of our NMR spectra. Since paramagnetic spec-
tra are very sensitive to the environment at the metal centre,
the large differences between the spectra of 7 and 9 may
mean that these complexes do not have the same geometry.
One possibility would be a switch to fac coordination of the
dpa ligand in 9. This would be inconsistent with the de-
creased tendency towards folding of bpc over Me-bpc dis-
cussed earlier. However, the two dpa complexes of Co that
have been structurally characterised both have fac-coordin-
ated dpa ligands.[21]
EPR Measurements
Six-coordinate [(tBu-pyp)CoCl2] (3) and five-coordinate
[(Me-Bz-dpa)CoCl2] (7) both have very broad X-band EPR
spectra in the range g ϭ 2Ϫ6, characteristic of high-spin
(S ϭ 3/2) cobalt() species with a large zero-field splitting
(D ϾϾ hν). The EPR spectrum of 3 is shown in Figure 7.
1
The H NMR spectrum of [(bpc)CoCl2] (5) could not be
fully assigned, because all signals (apart from the NH) have
approximately the same intensity and no distinct pattern
could be discovered. The number of signals, however, agrees
with the symmetrical structure observed in the solid state.
[(Me-bpc)CoCl2] (6) was never obtained as a single com-
pound. A mixture of two products in a ratio of 10:6 or
10:4 was formed in the reaction of the ligand with CoCl2 in
1
CH2Cl2 and CH3CN, respectively. H NMR spectroscopy
shows that both products produce twelve paramagnetic sig-
nals, the number expected for a symmetrically coordinated
ligand (i.e. N4 or N2/N2, but not N3/N). Slow evaporation
of an NMR sample resulted in crystallisation of one of the
Figure 7. Experimental and simulated X-band EPR spectrum of 3;
experimental spectra in acetone/MeOH (2:3) at 6 K, 9.64 GHz.
(attn. 55 dB, mod. 10 G)
Eur. J. Inorg. Chem. 2003, 648Ϫ655
651

A. W. Gal et al.
FULL PAPER
products. This is probably the major product of the reac- priate N3 or N4 ligand. Most complexes are air-stable for
tion, because both the crystal structure and the NMR spec- at least a couple of days. Different modes of coordination
trum resemble those of 5, while the spectrum of the minor became apparent from the obtained crystal structures.
product shows no obvious similarities to that of 5. Since These were correlated with the 1H NMR spectra of the
the minor product also has a symmetrical ligand, one pos- complexes and helped in their interpretation. Variation of
sibility would be the N4 structure, although this seems un- the solvent in the NMR experiments was informative for
likely given that CoII should have an even larger aversion the behaviour in solution of the pyp complexes. The pyp
to this arrangement than CoIII. Another possibility is that ligands seemed to promote six-coordination, since even
it is a cationic five-coordinate complex like [(tpa)CoCl]ϩ. A when a chloride dissociates in solution, a solvent molecule
distinction between these two possibilities cannot be made occupies the free coordination site. In contrast, tpa appears
at this stage. We were not able to obtain a 13C NMR spec- to favour five-coordination. The bpc complexes 5 and 6 are
trum for any of the paramagnetic complexes.
both six-coordinate in the solid state, but for 6 formation
of a five-coordinate complex in solution cannot be ex-
cluded. Only two of the pyridine/amine complexes prepared
(3 and 4) showed (low) activity in ethene polymerisation.
Ethene Polymerisation
All the complexes except 8 were tested for polymerisation
activity by treating them with 100 equivalents of MAO in
chlorobenzene under an ethene atmosphere (7 and 35 bar).
Most complexes did not exhibit any polymerisation activity
at all, but pyp complexes 3 and 4 produced some polymer.
The tert-butyl-substituted complex was particularly inter-
esting as it remained active for a very long time (35 hours),
resulting in an activity of 70 g·molϪ1·hϪ1·barϪ1. This is a
factor of 7000 less than the reported cobalt-dip system un-
der optimised conditions.[3a,3b] GPC analysis of the polymer
formed at 35 bar showed that, besides a small amount of
high molecular weight polyethene, the bulk of the ‘‘poly-
mer’’ had a molecular weight of approximately 1000. It was
difficult to optimise the pyp system because the complex
seemed to decompose when a fresh batch of MAO was used
for the polymerisation experiments.
It has been suggested that the dip-cobalt and -iron com-
plexes used by Brookhart and Gibson polymerise ethene
to high molecular weight polyethene because the large aryl
substituents prevent associative displacement of the grow-
ing polymer chain. [(Me-Bz-dpa)CoCl2] (7) has nearly the
same cobalt environment as precatalyst 8, but lacks the
steric bulk above and below the active site on cobalt. In-
deed, 7 is not active in ethene polymerisation (nor is 9). If
this were caused by rapid chain transfer, formation of
dimers would be expected. The fact that no ethene con-
sumption was observed indicates that the lack of steric bulk
is not the only obstruction for the reactivity of our N3 com-
plexes. Rather, it seems likely that the electronic properties
of the ligand are also important. For example, [(dip)CoCl2]
is apparently activated by reduction to a low-spin CoI spe-
cies, although the nature of the actual active species is still
unclear.[22] Our pyridine-amine ligands are much weaker π-
acceptors than dip ligands and are unlikely to stabilize the
CoI oxidation state, let alone induce the switch to a low-
spin configuration. The main problem here is could be the
formation of the active species, rather than the activity it
would have once formed.
Experimental Section
General: All procedures were carried out under N2 using standard
Schlenk techniques unless specified otherwise. Acetonitrile, THF,
dichloromethane and diethyl ether were distilled under nitrogen.
Ethanol and methanol were deoxygenated by bubbling a stream of
N2 through the solvent for 15 minutes. For the synthesis of 5 and
6, CoCl2·6H2O was dried for 4 hours at 150 °C before use. Ligands
were prepared according to literature procedures.[8] All other chem-
icals are commercially available and were used without further
purification. NMR measurements were performed on Bruker
DPX200, AMX400, and AMX500 spectrometers. X-band EPR
spectra were recorded on a Bruker ER 220 spectrometer. Samples
were measured in acetone/MeOH (2:3) at 6Ϫ10 K. Simulations
were performed with the program WEPR (F. Neese, University of
Konstanz). Chemical shifts are listed in ppm with TMS as external
reference (δ ϭ 0 ppm). NMR signals for the NH protons of para-
magnetic complexes were never observed. FABϩ- and ESIϩ-MS
spectra were recorded on a Finnigan MAT 900XL instrument.
[(tpa)CoCl]2[CoCl4]
([1]2[CoCl4]):
CoCl2·6H2O
(205 mg,
0.86 mmol) was added to a solution of tpa (248 mg, 0.86 mmol;
1.0 equiv.) in 20 mL of ethanol. A blue solid precipitated almost
immediately. After stirring for 1 hour at room temperature, the
mixture was filtered. The solid was washed with diethyl ether and
dried in vacuo; yield 0.19 g (66%). Crystals suitable for X-ray dif-
fraction were grown by vapour diffusion of THF into a saturated
acetonitrile solution. The blue product is air-stable as a solid and
in solution. C36H36Cl6Co3N8 (970.25): calcd C 44.56, H 3.74, N
11.55; found C 44.41, H 3.79, N 11.47. 1H NMR (200 MHz, 298 K,
ppm) CD3OD: δ ϭ 130.1 (3 H, py-H6), 108.0 (6 H, NCH2), 58.2
(3 H, py-H5), 49.2 (3 H, py-H3), Ϫ1.7 (3 H, py-H4). CD3CN: δ ϭ
134.3 (3 H, py-H6), 107.3 (6 H, NCH2), 60.8 (3 H, py-H5), 46.2 (3
H, py-H3), Ϫ3.5 (3 H, py-H4). FAB-MS: m/z
ϭ 803
[(tpa)2Co2Cl3]ϩ, 384 [M]ϩ, 349 [M Ϫ Cl]ϩ, 291 [M Ϫ CH3Py]ϩ, 256
[M Ϫ Cl, Ϫ CH3Py]ϩ. ESI-MS: m/z ϭ 384 [M]ϩ, 349 [M Ϫ Cl]ϩ.
[(tpa)CoCl]PF6 ([1]PF6): CoCl2·6H2O (330 mg, 1.39 mmol) was ad-
ded to a solution of tpa (424 mg, 1.46 mmol; 1.05 equiv) in 25 mL
of methanol. After addition of NH4PF6 (1.2 g, 7.2 mmol, 5.2
equiv.) a green microcrystalline solid precipitated. This solid was
filtered off, washed three times with diethyl ether and dried under
vacuum; yield 676 mg (92%). The product is air-stable as a solid
and in solution. C18H18ClCoF6N4P (529.72): calcd C 40.81, H 3.43,
N 10.58; found C 40.64, H 3.42, N 10.44. 1H NMR (200 MHz,
Conclusions
Analytically pure paramagnetic cobalt() complexes were
easily synthesised by the reaction of CoIICl2 with the appro- 298 K, ppm): CD3OD δ ϭ 129.8 (3 H, py-H6), 107.6 (6 H, NCH2),
652
Eur. J. Inorg. Chem. 2003, 648Ϫ655

Cobalt Chloride Complexes of N3 and N4 Donor Ligands
FULL PAPER
57.91 (3 H, py-H5), 49.1 (3 H, py-H3), Ϫ1.44 (3 H, py-H4).
CD3CN: δ ϭ 134.4 (3 H, py-H6), 107.1 (6 H, NCH2), 60.8 (3 H,
py-H5), 46.2 (3 H, py-H3), Ϫ3.6 (3 H, py-H4).
calcd C 56.57, H 5.37, N 11.47; found C 56.48, H 5.32, N 11.41.
1H NMR (200 MHz, CD3CN, 298 K, ppm): δ ϭ 152.0 (2 H), 28.0
(2H), 15.6 (2 H), 14.6 (2 H), 14.2 (2 H), 12.1 (2 H), 7.7 (1 H, bz-
p), Ϫ12.0 (2 H), Ϫ61.1 (6 H, py-CH3). FAB-MS: m/z ϭ 446 [M]ϩ,
411 [M Ϫ Cl]ϩ.
[(tBu-pyp)CoCl2] (3):[5d] CoCl2·6H2O (172 mg, 0.72 mmol) was ad-
ded to a warm solution (40 °C) of tBu-pyp (256 mg, 0.72 mmol;
1.00 equiv.) in 35 mL of THF. After refluxing for 30 min a light-
blue solid precipitated. The mixture was cooled to room temper-
[(Me-dip)CoCl2] (8):[3a] Me-dip (564 mg, 1.53 mmol) and
CoCl2·6H2O (363 mg, 1.53 mmol, 1.00 equiv.) were dissolved in
ature and filtered. The residue was washed twice with small por- 40 mL of THF. A colour change could be observed almost immedi-
tions of THF and dried under vacuum; yield 275 mg (79%). ately and a green solid precipitated. After stirring for an additional
C22H32Cl2CoN4 (483.26): calcd C 54.78, H 6.69, N 11.62; found C hour at room temperature, this solid was filtered off, washed with
54.66, H 7.78, N 11.55. 1H NMR (200 MHz, 298 K, ppm) CD2Cl2:
THF (2ϫ) and hexane (2ϫ), and recrystallised from a layered
δ ϭ 81.4 (4 H, py-H3), 26.2 (2 H, py-H4), 3.1 (?), Ϫ9.4 (4 H, methanol/hexane (1:1) solution, giving 266 mg (0.53 mmol, 35%)
NCH2), Ϫ35.8 (4 H, NCH2), Ϫ43.5, Ϫ76.3 (tBu?). CD3OD: δ ϭ of the product as green air-stable crystals. 1H NMR (200 MHz,
73.8 (4 H, py-H3), 28.3 (4 H, NCH2), 18.8 (2 H, py-H4), Ϫ4.7 (4 298 K, ppm) CD2Cl2: δ ϭ 112.0 (2 H, py-H3), 36.1 (1 H, py-H4),
H, NCH2), Ϫ38.6 (tBu). FAB-MS: m/z ϭ 446 [M Ϫ Cl]ϩ, 411 [M
Ϫ 2Cl]ϩ, 353 [M Ϫ 2Cl, Ϫ tBuH]ϩ.
6.1 (4 H, Ar-Hm), Ϫ1.2 (6 H, NϭCMe), Ϫ12.1 (2 H, Ar-Hp),
Ϫ28.2 (12 H, Ar-Me). CD3OD: δ ϭ 111.9 (2 H, py-H3), 36.2 (1
H, py-H4), 6.2 (4 H, Ar-Hm), Ϫ1.1 (6 H, NϭCMe), Ϫ12.0 (2 H,
Ar-Hp), Ϫ28.1 (12 H, Ar-Me). FAB-MS: m/z ϭ 463 [M Ϫ Cl]ϩ.
[(Me-pyp)CoCl2]·H2O (4·H2O): This compound was synthesised
according to a literature procedure.[5a] 1H NMR (200 MHz, 298 K,
ppm) CD2Cl2: δ ϭ 48.0 (4 H, NCH2), 46.4 (4 H, py-H3), 24.0 (6 [(dpa)CoCl2] (9): A solution of CoCl2·6H2O (136 mg, 0.57 mmol)
H, NCH3), 14.2 (4 H, NCH2), 7.2 (2 H, py-H4). CD3OD: δ ϭ in 5 mL of ethanol was added to a solution of dpa (102 mg,
117.7 (6 H, NCH3), 88.8 (4 H, NCH2), 46.9 (4 H, py-H3), 31.5 (4 0.52 mmol; 0.91 equiv.) in 5 mL of ethanol. The colour of the mix-
H, NCH2).
ture changed from light purple to dark-blue after stirring for 5
minutes at room temperature. After stirring for 16 hours at room
temperature, the product was precipitated by adding diethyl ether.
The mixture was filtered and the residue washed with diethyl ether
and dried in vacuo; yield 73 mg (42%). The purple product is air-
stable as a solid, however it decomposed in solution after being
exposed to air for one day. 1H NMR (200 MHz, 298 K, ppm)
CD3CN: δ ϭ 126.1 (2 H, py-H6), 120.8 (2 H, NCH2), 113.5 (2 H,
NCH2), 55.9 (2 H, py-H5), 44.0 (2 H, py-H3), Ϫ1.1 (2 H, py-H4).
CD2Cl2: δ ϭ 127.0 (2 H, py-H6), 113.6 (2 H, NCH2), 110.3 (2 H,
NCH2), 55.7 (2 H, py-H5), 44.6 (2 H, py-H3), Ϫ0.80 (2 H, py-H4).
CDCl3: δ ϭ 126.4 (2 H, py-H6), 112.9 (4 H, NCH2), 55.1 (2 H, py-
H5), 45.2 (2 H, py-H3), Ϫ0.7 (2 H, py-H4). FAB-MS: m/z ϭ 293
[M Ϫ Cl]ϩ, 258 [M Ϫ 2Cl]ϩ.
[(bpc)CoCl2] (5): A solution of dried CoCl2 (106 mg, 0.82 mmol) in
10 mL of CH3CN was added to a solution of bpc (239 mg,
0.81 mmol; 0.99 equiv.) in 7 mL of CH3CN. The mixture immedi-
ately turned bright red. Evaporation of the solvent yielded a bright
pink compound, which was washed twice with THF and dried un-
der vacuum; yield 237 mg (68%). Single crystals suitable for X-ray
diffraction were grown from a saturated CD3CN solution. The
1
product is highly air-sensitive as a solid and in solution. H NMR
(200 MHz, CD3CN, 298 K, ppm): δ ϭ 195.1 (2 H), 88.9 (2 H), 52.4
(2 H), 26.0 (2 H), 22.0 (4 H), 16.2 (2 H), 9.2 (2 H), 5.6 (2 H), 4.2
(2 H), 0.4 (2 H).
[(Me-bpc)CoCl2] (6): Dried CoCl2 (115 mg, 0.88 mmol) was added
to a solution of Me-bpc (296 mg, 0.91 mmol; 1.03 equiv.) in 12 mL
of CH3CN. A purple solid precipitated immediately. After 2 hours
the mixture was filtered and the dark pink residue was dried in
vacuo; yield 244 mg (61%). Additional product precipitated from
the mother liquor on standing at Ϫ20 °C. Crystals suitable for X-
ray diffraction were obtained by slow evaporation of a CD2Cl2 so-
lution. The product is air-stable as a solid and in solution, although
it is very poorly soluble in most organic solvents. C20H28Cl2CoN4
(454.31): calcd C 52.88, H 6.21, N 12.33; found C 51.66, H 6.22,
Polymerisation Experiments: Chlorobenzene was distilled from a
small amount of sodium. MAO was purchased as a solution in
toluene (4.96% Al) and used as such. Polymerisation reactions un-
der 7 bar of ethene were performed in a glass autoclave; reactions
under 35 bar of ethene were done in a steel vessel (Hastelloy C).
A suspension of 33 µmol of the precatalyst in 10 mL of chloroben-
zene was transferred under nitrogen into a 100 mL autoclave. Un-
der nitrogen-flow 2 mL of MAO (100 equiv.) was transferred into
N 12.00. 1H NMR (500 MHz, CD2Cl2, 298 K, ppm): δ ϭ 164.9 a steel tube connected to the top of the pressure vessel. The com-
(2H), 104.3 (2H), 102.6 (1.2 H), 89.4 (1.2 H), 83.0 (3.6 H, NCH3),
61.1 (6 H, NCH3), 51.8 (2 H), 47.5 (1.2 H), 29.7 (1.2 H), 24.7 (1.2 MAO was added to the reaction mixture and the autoclave was
H), 24.4 (1.2 H), 23.9 (2H), 21.8 (0.8 H), 21.1 (2 H), 20.5 (2 H), closed. When ethylene consumption had stopped, or after about 6
18.6 (2 H), 17.4 (1.2 H), 15.9 (1.2 H), 9.3 (1.2 H), 8.1 (2 H), 6.6 (2 hours if no reaction was observed, the autoclave was opened and
plete system was put under ethylene pressure (7 or 35 bar), the
H), 5.8 (2 H), 3.3 (1.2 H), Ϫ16.93 (2 H). FAB-MS: m/z ϭ 418 [M
the reaction mixture was neutralised by adding a mixture of meth-
anol and aqueous HCl (5:1). The mixture was filtered and the poly-
ethylene Ϫ if formed Ϫ was washed several times with methanol
and dried under vacuum.
Ϫ Cl]ϩ, 383 [M Ϫ 2Cl]ϩ.
[(Me-Bz-dpa)CoCl2] (7):
A solution of CoCl2·6H2O (88 mg,
0.37 mmol) in 10 mL of acetonitrile was added to a solution of Me-
Bz-dpa (111 mg, 0.35 mmol; 0.95 equiv.) in 5 mL of acetonitrile. X-ray Structure Determinations: Single crystals were mounted in air
The colour of the mixture immediately changed to blue/purple.
After stirring for 7 hours at room temperature the volume was re-
duced by half under reduced pressure and a purple precipitate
on glass fibres. Intensity data were collected at room temperature
on an EnrafϪNonius CAD4 diffractometer. Unit cell dimensions
were determined from the angular setting of a limited number of
formed. The solid was collected by filtration, washed three times reference reflections. Intensity data were corrected for Lorentz and
with diethyl ether and dried under vacuum; yield 85 mg (54%). The
purple product is air-stable as a solid and in solution. Crystals of
polarisation effects, and for absorption using a semi-empirical ψ-
scan correction.[23] Structures were solved by the program system
7 suitable for X-ray diffraction were obtained by slow evaporation DIRDIF[24] using the program PATTY[25] to locate all non-hydro-
of an acetonitrile solution. 7·CH3CN, C23H26Cl2CoN4 (488.32):
gen atoms and refined with standard methods (refinement against
653
Eur. J. Inorg. Chem. 2003, 648Ϫ655

A. W. Gal et al.
FULL PAPER
F2 of all reflections) using SHELXL-97.[26] All non-hydrogen atoms
anisotropic thermal displacement parameters showed physically
were refined anisotropically; hydrogens were placed at calculated unacceptable values when determined using default least-squares
positions and refined isotropically in a riding mode, unless other-
wise noted. Geometrical calculations (PLATON[27]) revealed nei-
ther unusual geometric features, nor unusual short intermolecular
contacts, no higher symmetry and no (further) solvent-accessible
areas. Details of the data collection and structure determinations
are collected in Table 1.
refinement parameters. The SQUEEZE procedure[27] was used to
account for the residual electron density in a solvent area of 50 A ,
3
˚
resulting in a total electron count for this void of 44 electrons in the
unit cell. This electron density should probably also be attributed to
water molecules. Next the DIFABS procedure[28] was used,
lowering the anisotropic R-value from 8.3% to 7.9%, resulting in
considerably improved anisotropic thermal displacement para-
meters. The DELU and SIMU options still had to be applied to
all atoms, however, and the ISOR option specifically to N2A,
C51A, C55A, C62A, C51B, C65B, C72B, C76B, and O1.
CCDC-186122Ϫ186127 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge at www.ccdc.cam.ac.uk/conts/retrieving.html [or from the
Cambridge Crystallographic Data Centre, 12, Union Road, Cam-
bridge CB2 1EZ, UK; Fax: (internat.) ϩ44Ϫ1223/336-033; E-mail: 5: One molecule of acetonitrile was located and refined. The refine-
deposit@ccdc.cam.ac.uk].
ment was hampered by the severe deterioration of the crystal
during the data collection. A correction of almost 50% for crystal
decay was necessary during the data reduction process.
Notes on Individual Structures
[1]2[CoCl4]: One molecule of water and one molecule of acetonitrile
6: One molecule of water was located and refined; hydrogen atoms
were located and refined (hydrogen atoms for the water molecule
for this water molecule were not included in the refinement. Based
were not included in the refinement). Hydrogen atoms were placed solely on donor-acceptor distances hydrogen bonds could probably
at calculated positions and were subsequently refined freely. Some
be formed between O1 and Cl1 and between O1 and Cl2.
Table 1. Details of crystal structure determinations
Compound
[1]2[CoCl4]
5
6
7
8
Crystal colour
Crystal shape
dark green
transparent
red-brown
very irregular
fragment
0.52ϫ0.48ϫ0.28
C22H30Cl2CoN6
508.35
298(2)
Monoclinic,
C2/c
transparent
purple-red
rather regular
fragment
0.29ϫ0.17ϫ0.10
C20H30Cl2CoN4O
472.31
293(2)
Orthorhombic,
Pna21
purple
green
regular
regular
small, regular
fragment
0.13ϫ0.11ϫ0.08
C26H31Cl2CoN3O
531.37
150(2)
Monoclinic,
P21/n
fragment
0.24ϫ0.18ϫ0.15
C19H20.5Cl3Co1.5N4.5O0.5
514.64
293(2)
Monoclinic,
Cc
fragment
0.42ϫ0.24ϫ0.12
C23H26Cl2CoN4
488.31
293(2)
Monoclinic,
P21/c
Crystal size [mm
Empirical formula
Molecular weight
Temperature [K]
Crystal system,
space group
Unit cell det. #,
θ range
15,
25,
25,
25,
13258,
6.504···11.817
12.095(3)
27.587(9)
14.074(3)
90
93.967(17)
90
4685(2)
8, 1.459
1.432
ω
19.293···22.715
12.9525(6)
12.1804(10)
16.3098(10)
90
103.698(4)
90
2499.9(3)
4, 1.351
0.921
11.961···14.468
16.3169(11)
14.7110(10)
8.8110(6)
90
90
90
2115.0(2)
4, 1.483
1.083
ω/2θ
988
10.488···20.064
8.9564(12)
20.817(5)
13.1776(19)
90
106.227(10)
90
2359.0(7)
4, 1.375
0.971
1.830···25.350
14.1667(12)
14.6541(8)
13.1684(11)
90
109.558(4)
90
2576.0(3)
4, 1.370[a]
0.897[a]
˚
a [A]
˚
b [A]
˚
c [A]
α [°]
β [°]
γ [°]
3
˚
Volume [A ]
Z, dcalc [g·cmϪ3
]
Abs. coeff. [mmϪ1
Scan
ω/2θ
1060
ω/2θ
1012
Area det. θ, ω
F(000)
2092
1108[a]
θ range [°]
2.78···23.25
0 Յ h Յ 13
-30 Յ k Յ 0
Ϫ15 Յ l Յ 15
3546/3546
2.57···27.46
Ϫ16 Յ h Յ 0
0 Յ k Յ 15
Ϫ20 Յ l Յ 21
2985/2863
[0.0320]
2.85···27.46
0 Յ h Յ 21
0 Յ k Յ 19
0 Յ l Յ 11
2576/2576
2.56···25.61
0 Յ h Յ 10
0 Յ k Յ 25
Ϫ16 Յ l Յ 15
4722/4431
[0.0242]
1.83···25.35
Ϫ17 Յ h Յ 15
Ϫ17 Յ k Յ 16
Ϫ15 Յ l Յ 12
13258/4707
[0.0501]
Index ranges
Collected/unique
[Rint
]
Observed [Io Ͼ 2σ(Io)]
Abs struct par
DIFABS used?
1888
0.02(4)
yes
2210
1314
0.00(5)
no
3067
3265
no
no
no
Relat. transm. fact.
Data/restraints/pars
GoodnessϪofϪfit on F2
SHELXL-97 wt pars
R1, wR2 [I Ͼ 2σ(I)]
R1, wR2 (all data)
1.052···0.962
3546/531/515
1.004
1.051···0.916
2863/0/142
1.332
1.021···0.965
2576/1/250
1.053
0.0361, 1.3080
0.0705, 0.1074
0.1700, 0.1331
0.402, Ϫ0.501
1.069···0.961
4431/0/375
1.084
0.0403, 0.6266
0.0405, 0.0846
0.0741, 0.0963
0.278, Ϫ0.224
4707/0/286
0.960
0.0402
0.0400, 0.0802
0.0689, 0.0869
0.309, Ϫ0.371
0.0692
0.2000
0.0733, 0.1365
0.1635, 0.1608
0.629, Ϫ0.537
0.0995, 0.2789
0.1207, 0.3140
2.084, Ϫ1.124
Ϫ3
˚
Diff. peak, hole [e·A
]
[a]
3
˚
Excluding contents of a ‘‘void’’ of 165 A ; see text for details.
654
Eur. J. Inorg. Chem. 2003, 648Ϫ655

Cobalt Chloride Complexes of N3 and N4 Donor Ligands
FULL PAPER
A. Nanthakumar, S. Fox, N. N. Murthy, K. D. Karlin, J. Am.
Chem. Soc. 1997, 119, 3898.
[9]
7: One molecule of acetonitrile was located and refined. Hydrogen
atoms were refined freely.
[10]
[11]
Y. Yamamoto, Y. Hata, Y. Shimura, Chem. Lett. 1981, 1559.
Cr: [11a] Y. Hata, Y. Yamamoto, Y. Shimura, Bull. Chem. Soc. Jpn.
1981, 54, 1255; Fe: [11b] N. Arulsamy, D. J. Hodgson, J. Glerup,
Inorg. Chim. Acta 1993, 209, 61. [11c] Y. Mekmouche, S. Menage,
C. Toia-Duboc, M. Fontecave, J.-B. Galey, C. Lebrun, J. Pecaut,
Angew. Chem. Int. Ed. 2001, 40, 949. [11d] P. Mialane, J.-J. Girerd,
J. Guilhem, L. Tchertanov, Inorg. Chim. Acta 2000, 298, 38. [11e]
J. Simaan, S. Poussereau, G. Blondin, J.-J. Girerd, D. Defaye, C.
Philouze, J. Guilhem, L. Tchertanov, Inorg. Chim. Acta 2000,
299, 221. [11f] B. Rieger, A. S. Abu-Surrah, R. Fawzi, M. Steiman,
J. Organomet. Chem. 1995, 497, 73; Cu: [11g] C. Ng, M. Sabat, C.
L. Fraser, Inorg. Chem. 1999, 38, 5545.
3
8: Calculations[27] showed a void of 165 A , containing 32 electrons,
˚
3
˚
resulting in a large volume of about 35 A per atom. Hexane and
methanol were used in the synthetic route but no reasonable guess
could be made about the nature of the solvent moiety or moieties pre-
sent, nor was it possible to assign any physically meaningful para-
meters to the electron density. Therefore the SQUEEZE procedure
was applied to account for this electron density. As a consequence
this electron density could not be taken into account while calculat-
ing the physical·molecular properties as given in Table 1.
[12] [12a] M. A. Cox, T. J. Goodwin, P. Jones, P. A. Williams, F. S. Ste-
Acknowledgments
We thank Dr. E. J. Reijerse for recording the EPR spectra of 3 and 7.
[12b]
phens, R. S. Vagg, Inorg. Chim. Acta 1987, 127, 49.
E. F.
Birse, M. A. Cox, P. A. Williams, F. S. Stephens, R. S. Vagg, In-
org. Chim. Acta 1988, 148, 45. [12c] E. F. Birse, P. A. Williams, F.
S. Stephens, R. S. Vagg, Inorg. Chim. Acta 1988, 148, 63. [12d] J.
A. Chambers, T. J. Goodwin, P. A. Williams, F. S. Stephens, R.
S. Vagg, J. Coord. Chem. 1988, 17, 277. [12e] M. A. Anderson, E.
[1]
See e.g.: [1a] Y. Murakami, J.-I. Kikuchi, Y. Hisaeda, O. Hayash-
ida, Chem. Rev. 1996, 96, 721. [1b] J. Halpern, in: Chemistry and
Significance of Vitamin B12 Model Systems (Ed.: D. Dolphin);
John Wiley: New York, 1982; vol. 1, p. 501.
F. Birse, M. J. E. Hewlins, J. P. G. Richards, F. S. Stephens, R.
[2]
See e.g.: C. Bianchini, R. W. Zoellner, Adv. Inorg. Chem. 1997,
[12f]
S. Vagg, P. A. Williams, Inorg. Chem. 1991, 30, 3774.
P. D.
44, 263.
Newman, F. S. Stephens, R. S. Vagg, P. A. Williams, Inorg. Chim.
Acta 1993, 204, 257.
[3]
[3a] B. L. Small, M. Brookhart, A. M. A. Bennett, J. Am. Chem.
Soc. 1998, 120, 4049Ϫ4050. [3b] G. J. P. Britovsek, V. C. Gibson,
B. S. Kimberley, P. J. Maddox, S. J. McTavish, G. A. Solan, A. J.
P. White, D. J. Williams, Chem. Commun. 1998, 849Ϫ850. [3c] B.
L. Small, M. Brookhart, J. Am. Chem. Soc. 1998, 120,
7143Ϫ7144. [3d] G. J. P. Britovsek, M. Bruce, V. C. Gibson, B. S.
Kimberley, P. J. Maddox, S. Mastroianni, S. J. McTavish, C. Red-
shaw, G. A. Solan, S. Strömberg, A. J. P. White, D. J. Williams, J.
Am. Chem. Soc. 1999, 121, 8728Ϫ8740. [3e] G. J. P. Britovsek, S.
Mastroianni, G. A. Solan, S. P. D. Baugh, C. Redshaw, V. C. Gib-
son, A. J. P. White, D. J. Williams, M. R. J. Elsegood, Chem. Eur.
[13] [13a] P. Leverett, J. Petherick, P. A. Williams, R. S. Vagg, J. Coord.
[13b]
Chem. 1996, 37, 195.
R. R. Fenton, F. S. Stephens, R. S.
Vagg, P. A. Williams, Inorg. Chim. Acta 1991, 182, 67. [13c] R. R.
Fenton, F. S. Stephens, R. S. Vagg, P. A. Williams, Inorg. Chim.
Acta 1995, 236, 109.
[14]
[15]
[16]
R. R. Fenton, F. S. Stephens, R. S. Vagg, P. A. Williams, Inorg.
Chim. Acta 1992, 197, 233.
J. Glerup, P. A. Goodson, D. J. Hodgson, K. Michelson, Inorg.
Chem. 1995, 34, 6255.
J. Ratilainen, K. Airola, R. Frohlich, M. Nieger, K. Rissanen,
Polyhedron 1999, 18, 2265.
J. 2000, 6, 2221Ϫ2231.
[4]
Me-dip
ϭ 2,6-bis[1-(2,6-dimethylphenylimino)ethyl]pyridine;
[17]
[18]
J. Rogers, R. A. Jacobson, J. Chem. Soc. (A) 1970, 1826.
FeCl2 complex of 2,4,6-trimethylphenyl-substituted dip ligand:
see ref.[3d]
iPr-dip ϭ 2,6-bis[1-(2,6-diisopropylphenylimino)ethyl]pyridine;
dpa ϭ N,N-bis(2-pyridylmethyl)amine; Me-Bz-dpa ϭ N-benzyl-
N,N-bis[(6-methyl)pyrid-2-ylmethyl)]amine; tpaϭ N,N,N-tris(2-
pyridylmethyl)amine; Me-pyp ϭ 2,11-dimethyl-2,11-diaza[3,3]-
[19] [19a] (iPr-dip)CoCl2: ref.[3d] [19b] (iPr-dip)FeCl2: refs.[3a,3b]
[20]
P. H. M. Budzelaar, A. B. Van Oort, A. G. Orpen, Eur. J. Inorg.
Chem. 1998, 1485.
(2,6)pyridinophane; tBu-pyp
diaza[3,3](2,6)pyridinophane; bpc ϭ trans-N,NЈ-bis(2-pyridyl-
methyl)cyclohexane-1,2-diamine; Me-bpc trans-N,NЈ-di-
ϭ
2,11-di-tert-butyl-2,11-
[21] [21a]
T. Ama, K.-I. Okamoto, T. Yonemura, H. Kawaguchi, Y.
ϭ
[21b]
Ogasawara, T. Yasui, Chem. Lett. 1996, 29.
R. M. Harts-
methyl-N,NЈ-bis(2-pyridylmethyl)cyclohexane-1,2-diamine
[5a] H.-J. Krüger, Chem. Ber. 1995, 128, 531. [5b] C.-M. Che, Z.-Y.
Li, K.-Y. Wong, C.-K. Poon, T. C. W. Mak, S.-M. Peng, Polyhed-
ron 1994, 13, 771. [5c] Y. Yamamoto, Y. Hata, Y. Shimura, Chem.
Lett. 1981, 1559Ϫ1560. [5d] S. P. Meneghetti, P. J. Lutz, J. Fischer,
J. Kress, Polyhedron 2001, 20, 2705.
[5]
horn, S. G. Telfer, J. Chem. Soc., Dalton Trans. 2000, 2801.
[22] [22a]
V. C. Gibson, M. Humphries, K. Tellmann, D. Wass, A. J.
P. White, D. J. Williams, Chem. Commun. 2001, 2252. [22b] T. M.
Kooistra, Q. Knijnenburg, J. M. M. Smits, A. D. Horton, P. H.
M. Budzelaar, A. W. Gal, Angew. Chem. Int. Ed. 2001, 40, 4719.
A. C. T. North, D. C. Philips, F. S. Mathews, Acta Crystallogr.,
Sect. A 1968, 24, 351.
[23]
[24]
[6]
[7]
For an example of eight-coordinate Co, see: W. O. Koch, J. T.
Kaiser, H.-J. Krüger, Chem. Commun. 1997, 2237.
For κ3-coordinated pyridinophane ligands, see: [7a] H. Kelm, H.-
J. Krüger, Eur. J. Inorg. Chem. 1998, 1381. [7b] S. P. Meneghetti,
P. J. Lutz, J. Kress, Organometallics 2001, 20, 5050. [7c] T. Scia-
rone, J. Hoogboom, P. P. J. Schlebos, P. H. M. Budzelaar, R. De
Gelder, J. M. M. Smits, A. W. Gal, Eur. J. Inorg. Chem. 2002, 457.
P. T. Beurskens, G. Beurskens, W. P. Bosman, R. de Gelder, S.
´
Garcıa-Granda, R. O. Gould, R. Israe¨l, J. M. M. Smits,
DIRDIF-96. A computer program system for crystal structure de-
termination by Patterson methods and direct methods applied to
difference structure factors, Crystallography Laboratory, Univer-
sity of Nijmegen: The Netherlands, 1996.
P. T. Beurskens, G. Beurskens, M. Strumpel, C. E. Nordman, in:
Patterson and Pattersons (Eds.: J. P. Glusker, B. K. Patterson, M.
Rossi); Clarendon Press: Oxford, 1987; p. 356.
[8] [8a]
Me-dip, iPr-dip: ref.[3d] [8b]dpa: H. Nagao, N. Komeda, M.
[25]
[26]
Mukaida, M. Suzuki, K. Tanaka, Inorg. Chem. 1996, 35, 6809.
[8c] Me-Bz-dpa: B. De Bruin, J. A. Brands, J. J. J. M. Donners, M.
P. J. Donners, R. De Gelder, J. M. M. Smits, A. W. Gal, A. L.
[8d]
G. M. Sheldrick, SHELXL-97. Program for the refinement of
crystal structures; University of Göttingen: Germany, 1997.
Spek, Chem. Eur. J. 1999, 5, 2921.
tpa: F. Højland, H.
Toftlund, S. Yde-Andersen, Acta Chem. Scand., Ser. A 1983, 37,
251. [8e] Me-pyp: F. Bottino, M. de Grazia, P. Finocchario, F. R.
Frinczek, A. Mamo, S. Pappalardo, J. Org. Chem. 1988, 53, 3521.
[27] [27a] A. L. Spek, Acta Crystallogr., Sect. A 1990, 46, C34. [27b] A.
L. Spek, PLATON-93. Program for display and analysis of crystal
and molecular structures; University of Utrecht: The
Netherlands, 1995.
[8f]
tBu-pyp: ref.[5b] [8g] bpc: M. A. Heinrichs, D. J. Hodgson, K.
[8h]
Michelsen, E. Pedersen, Inorg. Chem. 1984, 23, 3174.
Me-
[28]
N. Walker, D. Stuart, Acta Crystallogr., Sect. A 1983, 39, 158.
bpc: J. Glerup, P. A. Goodson, A. Hazall, R. Hazall, D. J. Hodg-
son, C. J. McKanzie, K. Michelsen, U. Rychlewska, H. Toftlund,
Inorg. Chem. 1994, 33, 4105.
Received June 6, 2002
[I02298]
Eur. J. Inorg. Chem. 2003, 648Ϫ655
655