Iron and cobalt complexes of 4-alkyl-2,6-diiminopyridine ligands: Synthesis and ethylene polymerization catalysis

Article abstract of DOI:10.1002/ejic.200701220

A new methodology for selective alkylation of the 2,6-diiminopyridine (pdi) ligands allows the ready synthesis of new derivatives displaying 2-methyl-2-phenylpropyl (neophyl), benzyl, or allyl groups in the 4-position of the pyridine ring. A set of Fe and Co complexes containing the new ligands have been synthesized and fully characterized. The presence of 4-neophyl or -benzyl groups does not perturb the performance of the complexes as ethylene polymerization catalysts, but the introduction of the 4-allyl group leads to the production of higher molecular weight polymers. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejic.200701220

FULL PAPER
DOI: 10.1002/ejic.200701220
Iron and Cobalt Complexes of 4-Alkyl-2,6-diiminopyridine Ligands: Synthesis
and Ethylene Polymerization Catalysis
Juan Cámpora,*[a] A. Marcos Naz,[a] Pilar Palma,[a] Antonio Rodríguez-Delgado,[a]
Eleuterio Álvarez,[a] Incoronata Tritto,[b] and Laura Boggioni[b]
Keywords: Iron / Cobalt / Diiminopyridine ligands / Ethylene Polymerization / Self-immobilization
A new methodology for selective alkylation of the 2,6-di- of 4-neophyl or -benzyl groups does not perturb the perform-
iminopyridine (pdi) ligands allows the ready synthesis of new ance of the complexes as ethylene polymerization catalysts,
derivatives displaying 2-methyl-2-phenylpropyl (neophyl), but the introduction of the 4-allyl group leads to the pro-
benzyl, or allyl groups in the 4-position of the pyridine ring. duction of higher molecular weight polymers.
A set of Fe and Co complexes containing the new ligands (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
have been synthesized and fully characterized. The presence Germany, 2008)
duce ω-alkenyloxy groups in the 4-position of the pyridine
Introduction
ring,[9] but electron-withdrawing substituents at the hetero-
Since their discovery in the late 1990s,[1] iron and cobalt
cyclic core are usually detrimental towards the properties
olefin polymerization catalysts containing 2,6-diiminopyr-
of the catalyst.[9,10] In spite of these problems, the pyridine
idine (pdi) ligands have been the subject of much interest.[2]
3/5- and 4-sites provide good attachment points when
Much effort has been devoted to the modification of the
modification of the ligand is not intended to alter the ac-
ligand structure, in order to gain control of catalyst activity
tivity of the catalyst (e.g. ligand immobilization on sur-
and polymer properties. The ligand has also been modified
faces), since these points remain necessarily far from the
to provide attachment points for catalyst immobilization.[3]
metal center, in contrast to the N-aryl and imine positions.
A large part of that work has concentrated on the system-
For these purposes, alkyl groups are the most obvious
atic variation of the nitrogen aryl substituents, since these
choice, given their chemical stability and electronically ne-
have a deep influence on the catalyst performance. Many
arly neutral character, but they are particularly difficult to
introduce. Perhaps for this reason routes towards pdi li-
N-aryl[4] and N-heteroaryl[5] groups displaying different
substitution patterns have been investigated. Substitution at
the α-imine position is another modification strategy, and
not only the well-known 2,6-bis(formaldimino)pyridine and
2,6-bis(acetaldimino)pyridine ligands but a number of li-
gands displaying alkyl groups attached to the heterocycle
remain essentially unexplored.[11] Recently, we disclosed a
new procedure that allows the selective alkylation of 2,6-
bis{[1-(2,6-diisopropylphenyl)imino]ethyl}pyridine
(
iPrpdi)
gands that display different types of α-alkyl and aryl
groups[6] as well as heteroatom-based functionalities (OR,
SR, etc)[7] have been prepared and studied. These structural
modifications of the pdi framework provide access to a wide
variety of polyethylene materials and are usually based on
relatively straightforward methodologies. This constitutes
one of the most attractive qualities of the pdi-based catalyst
family. In contrast, modification of the ligand core by either
substitution of the pyridine nucleus by different heterocy-
cles[8] or introduction of substituents in the ring[9–11] poses
a more stringent synthetic challenge. For instance, Alt has
reported an elegant but laborious multistep route to intro-
ligands at the pyridine 4-position by a simple one-pot pro-
cedure.[12] In this contribution, we expand our previous re-
sults, describing the synthesis of the corresponding Fe and
Co complexes and their performance as ethylene polymeri-
zation catalysts.
Results and Discussion
We reported that the reaction of MnR2 reagents with
iPrpdi at low temperatures leads to the thermally unstable
dialkylmanganese species 1, which undergoes a spontane-
ous rearrangement involving the migration of one of the
alkyl groups from the metal center to the 4-position of the
heterocyclic ring (Scheme 1).[12] These changes are marked
by a characteristic color change form the dark, nearly black
hue of dialkyl species 1 to the red burgundy color of the
alkyl(amido)manganese(I) complex 2 (Scheme 1). The latter
[a] Instituto de Investigaciones Químicas, Universidad de Sevilla –
CSIC,
c/ Americo Vespucio, 49. 41092, Seville, Spain
Fax: +34-954460565
E-mail: campora@iiq.csic.es
[b] ISMAC-C.N.R.,
Via E. Bassini 15, 20133 Milano, Italy
Eur. J. Inorg. Chem. 2008, 1871–1879
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1871

J. Cámpora et al.
FULL PAPER
Scheme 1.
is prone to loss of hydrogen in solution to afford the 4- were isolated in good yields without any special modifica-
alkylated pdi derivative 3, but this process is quite slow. tion of the general procedure (see Experimental Section).
Quenching the solution with methanol readily releases the
As expected, treatment of ligands L1, L2, or L3 with
modified organic ligands contained in 2 and 3 to afford a suspensions of metal halides (FeCl2·4H2O or CoCl2) in or-
mixture of 4-alkyl-2,6-diimino-1,4-dihydropyridine and the ganic solvents gives rise to the corresponding complexes 4–
4-alkyl-pdi ligand. The dihydropyridine derivative is slowly 9 (Scheme 2). We did not pursue the synthesis of the metal
aromatized on exposition to air, a reaction that is ac- derivatives corresponding to L4 or L5, although very likely
celerated in the presence a catalytic amount of CrO3 on they would also readily be obtained following this standard
K2CO3, which leads to the desired alkylated pdi ligand as procedure. The new complexes precipitate out from the re-
the only significant organic product. Homoleptic or sol- action media (usually thf) and are readily separated by fil-
vated dialkylmangenese(II) compounds are extremely air- tration, followed by washing with hexane. Their solubility
sensitive species, but they can be generated and used in situ, is somewhat higher than that of the parent complexes, and
thus avoiding their manipulation and isolation. In this way, in some cases, the synthesis is advantageously performed in
4-alkylated iPrpdi derivatives L1–L3 are conveniently ob- diethyl ether, rather than thf in order to ensure precipi-
tained in good isolated yields.
The 1,5-alkyl migration step that takes place in the inter- paramagnetic solids with magnetic moments typical of
mediate organomanganese complex is a rather unusual pro- high-spin electronic configurations (Fe complexes, µeff
tation. The complexes are obtained as microcrystalline,
≈
cess, and therefore its generalization to other pdi derivatives 5.3 BM; Co, µeff ≈ 4.8 BM), and their colors are similar to
displaying different substitution patterns at the nitrogen- those of the parent compounds, dark blue for Fe and brown
bound aryl groups is not obvious. In order to expand the for Co. Their UV/Vis spectra (CH2Cl2) are similar to those
scope of the method, we investigated its application to two of the parent compounds[13] and display a single, broad ab-
new pdi derivatives displaying decreasing degrees of steric sorption band in the visible region that is assigned to
hindrance. One of these compounds, iPr,Mepdi, retains one MLCT; the band is more intense for the iron complexes (at
ortho-isopropyl substituent at each of the N-aryl groups, ca. 690 nm, ε ≈ 103) than for the cobalt complexes (700 nm,
while the other is replaced by a methyl group, and in the ε ≈ 102). A group of bands probably associated with internal
second, Mespdi, both ortho and para positions of the N-aryl pdi ligand πǞπ* transitions are found in the ultraviolet
group are substituted by methyl groups. In both cases the range between 250 and 300 nm. The presence of the alkyl
reaction follows the same pathway, and ligands L4 and L5 group has only a very slight effect on the position of the
Scheme 2.
1872
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 1871–1879

Iron and Cobalt Complexes of 4-Alkyl-2,6-diiminopyridine Ligands
UV/Vis absorption bands. The frequency of the MLCT the parent complexes reported by Gibson.[1b,4c] During one
band is expected to be more sensitive to the electronic influ- attempt of crystallizing 4, a crop of brown crystals was ob-
ence of the 4-R group; however, a very small bathochromic tained. Its X-ray structure (Figure 3) showed it to be a
shift of this band relative to that in FeCl2(iPrpdi) was ob- mixed valence binuclear iron compound, probably arising
served only in the Fe complexes (3 nm for 4 and 5 nm for 5 from oxidation by adventitious traces of oxygen. An almost
and 6), while its position remains essentially unchanged for identical result was obtained by Brookhart and Small, who
the Co compounds. This suggests that the presence of the reported the analogous structure of an oxidation product
alkyl substituent does not significantly perturb the elec- isolated on attempted crystallization of a FeCl2(pdi) com-
tronic structure of the complexes.
plex.[4b]
1
The H NMR spectra of complexes 4–9 display typical
paramagnetically shifted signals, which are generally
broader in the Fe than in the Co derivatives. All signals
were assigned on the basis of their relative intensity and
width and by comparison with the spectra of the corre-
sponding MCl2(iPrpdi) complexes. One of the most notice-
able features of these spectra is the lack of the pyridine 4-
H signal, which occurs at 82.4 and 116.5 ppm in the Fe and
Co parent complexes, respectively. Other signals for the pdi
framework are little affected by the presence of the 4-alkyl
substituent. The atoms of the latter groups are farthest from
the metal center, and thus they are expected to be less affec-
ted by the paramagnetism. This is true in general, and these
signals usually appear sharper and occur close to the dia-
magnetic zone. For instance, the signal for the para-H atom
of the benzyl group in 8 is sharp enough to allow resolution
of the vicinal H–H coupling; this signal appears as a triplet
at δ = 12.6 ppm with 3JHH = 6.5 Hz. However, the aromatic
pyridine ring effectively delocalizes the spin density, and
some of the resonances of the pending group may display
large contact shifts, which can vary significantly from one
compound to another. For instance, the ring-bound CH2
group gives rise to a signal at δ = –13.7 ppm in 6, –35.0 ppm
in 7 and –31.1 ppm in 8, while this signal appears within
the +10 to +15 ppm region for the three cobalt derivatives.
Crystals of compounds 4 and 7 bearing a neophyl sub-
stituent in the pyridine ring were grown from CH2Cl2/hex-
ane (9:1) solutions at –20 °C over several days. Their molec-
ular structures are shown in Figure 1 and Figure 2, and se-
lected bond lengths and angles are collected in Table 1,
where they are compared with the analogous parameters in
Figure 2. X-ray crystal structure of complex 7.
Table 1. Selected bond lengths [Å] and angles [°] for complexes 4
and 7 and comparison with the parent compounds MCl2(iPrpdi).
[a]
[a]
Bond/Angle
4
∆
Fe
7
∆
Co
M1–N1
M1–N2
M1–N3
M1–Cl2
M1–Cl1
N1–C5
N1–C1
N2–C8
N3–C6
C5–C8
C4–C5
C3–C4
C2–C3
C1–C2
2.0781(17) –0.0099 2.0529(11) 0.0019
2.2149(16) –0.0231 2.2090(11) –0.002
2.2371(15) –0.0129 2.2186(11) 0.0076
2.2651(6)
2.3139(6)
1.343(2)
1.342(2)
1.289(3)
1.284(2)
1.488(2)
1.385(3)
1.400(3)
1.398(3)
1.388(3)
1.489(3)
117.64(2)
147.86(5)
95.53(4)
–0.0009 2.2554(5)
0.0044
–0.0083
0.0029
0.005
0.005
0.004
0.004
0.005
0.001
0.012
0.027
–0.009
0.008
0.19
2.2847(4)
1.3343(18) –0.0047
1.3377(17) 0.0007
1.2828(17) 0.0058
1.2866(17) 0.0066
1.4931(18) 0.0031
1.3921(18) 0.0091
1.4018(18) 0.0128
1.3955(19) 0.0265
1.3893(18) –0.0077
1.4897(19) –0.0053
116.726(16) 0.2
C1–C6
Cl1–M1–Cl2
N1–M1–Cl2
N1–M1–Cl1
0.00
–0.07
148.60(3)
94.65(3)
2.0
1.75
[a] Bond length or angle differences with the analogous parameters
in corresponding MCl2(iPrpdi) complexes (M = Fe;[1b] Co[3c]).
Compounds 4 and 7 are isomorphous and crystallize in
the same point group (monoclinic Cc), with very close crys-
tal unit parameters and identical molecular arrangements.
The neophyl fragment is extended, with the essentially co-
planar phenyl and pyridine rings disposed in antiperiplanar
position with respect to the C11–C10 bond. The metal cen-
ters are in approximately square-pyramidal environments.
The MCl2(pdi) frameworks show only slight differences
Figure 1. X-ray crystal structure of complex 4.
Eur. J. Inorg. Chem. 2008, 1871–1879
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1873

J. Cámpora et al.
FULL PAPER
other bonds are almost insensitive to the presence of the
alkyl group in the cobalt derivative, but some small differ-
ences can be noticed for the iron complex, notably in the
Fe–N bonds which are 0.01–0.02 Å shorter in 4 than in
FeCl2(iPrpdi). The shortening of the three Fe–N bonds sug-
gests that the alkyl substituent increases the donor capa-
bility of the pdi ligand towards the FeII center, but this is
not supported by the Fe–Cl2 bond length, which would be
expected to be longer because of the stronger trans influ-
ence of the ligand but remains the same length (within the
experimental uncertainty) as in the parent compound.
Like the analogous dinuclear compound reported by
Brookhart and Small, the molecule of compound 10 dis-
plays two iron centers bridged by an oxygen atom. The
angular Fe–O–Fe unit [Fe1–O1–Fe2 146.7(1)°] suggests that
the bridging atom does not belong to an oxido group but
rather to a hydroxido group, and consequently, the com-
pound is probably a mixed valence FeII–FeIII complex.
However, there is no further support for this proposal, as
the IR spectrum does not show any band that might be
unambiguously assigned to the hydroxido group. The con-
formation of the molecule of 10 in its crystal structure dif-
fers from that observed for 4 and 7. The phenyl and pyr-
idine groups are placed in a gauche position with respect to
Figure 3. X ray crystal structure of complex 10. Selected bond
lengths [Å] and angles [°]: Fe1–N1 2.1725(18), Fe1–N2 2.0754(18),
Fe1–N3 2.1765(18), Fe1–O1 1.7624(15), Fe1–Cl1 2.2067(6), Fe2–
Cl(average) 2.2241(7), Fe1–O1–Fe2 146.7(1).
from those of the unsubstituted complexes. Thus, the the C9–C10 bond, and the two rings have approximately
average differences in the bond lengths in the substituted perpendicular orientations. This difference could be due to
and unsubstituted complexes (calculated as a standard devi- crystal packing forces, imposed by the different molecular
ation over the selected bond lengths contained in Table 1) shape.
are 0.012 Å for the Fe complexes and 0.009 Å for the Co
The performance of iron and cobalt complexes 4–9 as
complexes, which indicate a somewhat larger variation in precatalysts for ethylene polymerization has been investi-
the former case. The C2–C3 and C3–C4 bonds, next to the gated, by using methylaluminoxane (MAO) as cocatalyst,
alkyl group attachment point, are elongated to the same with an M/Al ratio 1:500. The polymerization reactions
extent in 4 and 7 (0.012 and 0.027 Å) with respect to the were carried out under near atmospheric pressure of ethyl-
reference complexes, possibly because of steric effects. The ene (1.1 bar) in magnetically stirred reactors with an exter-
Table 2. Data for ethylene polymerization experiments.[a]
[c]
Entry
Catalyst
PE yield
[g]
Activity
Mn
Mw
Mw/Mn
Mv
(ϫ10–3)[b]
(ϫ10–3)[c]
(ϫ10–3)[c]
(ϫ10–3)
1
2
3
4
5
6
7
8
FeCl2(iPrpdi)
1.0
1.1
1.3
1.2
1.1
1.4
0.8
1.1
1.4
1.2
1.1
0.9
0.8
0.7
0.6
0.7
0.6
0.7
0.5
0.8
1.4
1.5
1.8
1.6
1.5
1.9
1.1
1.5
1.9
1.6
1.5
1.2
1.1
1.0
0.8
1.0
0.8
1.0
0.7
1.1
9.2
62.6
6.8
63.0[c]
FeCl2(iPrpdi)
46.1[d]
FeCl2(iPrpdi)
FeCl2(iPrpdi)
4
10.6
7.3
13.4
83.7
37.2
56.1
7.9
5.1
4.2
84.1[c]
37.1[c]
56.0[c]
34.6[d]
175.3[c]
180.0[c]
182.0[d]
29.5[c]
28.4[c]
4
5
5
9
6
7.1
8.4
176.1
179.7
24.8
21.4
10
11
12
13
14
15
16
17
18[e]
19[e]
20[e]
6
6
CoCl2(iPrpdi)
10.9
11.9
29.4
33.3
2.7
2.8
CoCl2(iPrpdi)
7
7
8
8
9
9
9
20.7[d]
28.1[c]
10.4
25.7
2.7
[a] Polymerization conditions: catalyst amount, 4 µmol; solvent, toluene (100 mL); external bath temperature, 30 °C; ethylene pressure,
1.1 bar; cocatalyst, MAO (M/Al = 1:500); reaction time, 10 min. [b] Activity in kg PE(mol catalyst)–1 bar–1 h–1. [c] Determined by GPC.
[d] Determined by viscosimetry. [e] Insoluble polymer.
1874
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 1871–1879

Iron and Cobalt Complexes of 4-Alkyl-2,6-diiminopyridine Ligands
nal bath thermostated at 30 °C. Catalytic activities and of the reactor.[9] Alt recently described the synthesis and
polymer characterization data are collected in Table 2. For catalytic activity of a wide range of FeCl2(pdi) complexes
comparison, Table 2 includes polymerization data corre- similar to those previously reported by the Herrmann and
sponding to the reference complexes FeCl2(iPrpdi) and Co- Jin groups, but, somewhat paradoxically, no conclusive evi-
Cl2(iPrpdi) under the same experimental conditions as those dence of catalyst self-immobilization was found.[18] In order
employed for their functionalized derivatives. In our hands, to obtain some additional evidence of the existence of un-
the behavior of these two catalysts matches that reported in usual effects on the catalytic behavior of the 4-allyl-func-
the literature, although the activity figures are somewhat tionalized complexes 6 and 9, we decided to examine the
lower.[1,3] In general, the iron catalysts produce higher mo- morphology of the polymers generated with these catalysts
lecular weight polymers than their cobalt analogues, but by using scanning electron microscopy (SEM). Self-immo-
with a markedly broader molecular weight distribution. As bilization of catalysts is thought to lead the formation of
expected, the iron catalysts are significantly more active microscopic polymer particles, which mimic the initial im-
than the cobalt catalysts. Previously reported FeCl2(iPrpdi) mobilized catalyst seeds. As an example of this behavior,
catalysts bearing electron-withdrawing groups in the pyr- Jin mentioned the production of micron-sized spheres when
idine 4-position (OR,[9] Cl[10]) show activities that are signif- using either nickel or iron catalysts bearing pending alkenyl
icantly lower than those of their unsubstituted congeners. groups.[16] Figure 4 shows SEM images of polymer samples
In contrast, the 4-alkylated Fe derivatives 4–6 maintain es- prepared with the two 4-allyl-functionalized pyridine cata-
sentially unaltered activities, and a similar conclusion can lysts and those generated with the FeCl2(iPrpdi) and Co-
be drawn for the cobalt analogues 7–9. The polymer sam- Cl2(iPrpdi) reference complexes. The latter complexes give
ples generated by the Fe and Co neophyl and benzyl deriva- rise to rather structured polymer materials in the 1–2 mi-
tives show insignificant variations in the molecular weight cron scale (Figure 4C and D, respectively). The cobalt cata-
and polydispersity that can be attributed to changes in the lyst gives formations that resemble platelet clusters, whilst
internal reactor temperature during the experiments. How- the iron catalyst gives rise to partially spherical, hollow,
ever, both the Fe and Co allyl derivatives 6 and 9 produce “shell-like” particles, which are likely to consist of curved
appreciably less-soluble polyethylene samples, a feature that lamellae. It seems likely that these structures arise from the
suggests they have higher molecular weights. Although this partially crystalline character of the materials, particularly
fact prevented the characterization of the polymers ob- in the case of those produced with the cobalt catalyst, since
tained with 9, GPC and viscosimetric analysis of the prod- this catalyst gives relatively small polyethylene molecules
uct of the iron catalyst 6 (Entries 9 and 10) confirms a sub- with a narrow molecular weight distribution. In contrast,
stantial increase in Mw and Mv, to ca. 180000. The GPC the images of the polymers produced with 6 and 9 appear
traces of these polymers indicate very broad, bimodal mo- to be more homogeneous and lack structural features (Fig-
lecular weight distributions, with peaks at Mp = 68000 ure 4A and B). This might possibly be attributed to the
(major) and 890. At this point, it is difficult to give a rea- higher molecular weight and wider range of molecular sizes
sonable explanation for these findings, but the presence of (at least in the case of the Fe catalyst), which results in an
a polymerizable vinyl unit in the structure of 6 and 9 sug- amorphous material. However, it seems clear that the
gests that these catalysts might become incorporated in the images provide no evidence of polymer particles resulting
growing polymer chain, which leads to a somewhat modi- from catalyst self-immobilization.
fied propagating species. Alt demonstrated this possibility
for metallocene catalysts containing pending alkenyl chains
attached to the Cp ligands. This so-called self-immobiliza-
tion of the catalyst usually results in substantially modified
activities and polymer structures.[14] The concept of self-
immobilization has been extended to other catalytic systems
that include late transition metals.[3] Herrmann[15] and
Jin[16] functionalized FeCl2(pdi) complexes with alkenyl
chains at the α-imino or the N-aryl positions, respectively,
and they obtained some evidence of the self-immobilization
phenomena in spite of the poor capability of iron catalysts
for α-olefin incorporation.[4b,4d,17] For instance, Jin men-
tioned the production of polymers with large polydispersity
indexes (Mw/Mn) of up to 22, which are similar to those
generated by 6. In contrast, Alt reported that catalysts
modified with pending O-alkenyl groups at the 4-position
of the pyridine ring produce polyethylenes with a monomo-
dal molecular weight distribution, in contrast with the bi-
Figure 4. SEM images of polyethylene samples produced with com-
modal polymers obtained with the analogous nonfunction-
plexes
6 (A) and 9 (B) and with the reference complexes
FeCl2(iPrpdi) (C) and CoCl2(iPrpdi) (D). The bar scale represents
alized catalysts. The presence of the 4-alkenoxy group was
also found to be beneficial in that it helps to prevent fouling 5 µm.
Eur. J. Inorg. Chem. 2008, 1871–1879
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1875

J. Cámpora et al.
FULL PAPER
as follows: o-dichlorobenzene + 0.05% 2,6-di-tert-butyl-4-meth-
ylphenol (BHT, antioxidant) as mobile phase, 0.8 mL/min as flow
Summary and Conclusions
Several pdi ligands displaying alkyl substituents at the 4- rate, and a column temperature of 145 °C. The column set was
composed of three GMHXL-HT columns from Toso Haas (Stutt-
gart, Germany). The universal calibration was constructed from 18
narrow MMD polystyrene standards, with the molar mass ranging
from 162 to 5.48ϫ106 g/mol. The MM values reported are those
obtained with the DV detector. Methylaluminoxane (MAO) was
purchased from Aldrich as a 7-% solution in heptane. The pdi li-
gands were synthesized from 2,6-diacetylpyridine according to
standard procedures. The preparation of derivatives iPrpdi-4-R
[where R = CH2CMe2Ph (L1); CH2Ph (L2); CH2CH=CH2 (L3)]
was carried out following the synthetic method reported by us in
position of the central pyridine ring have been prepared by
a simple procedure involving the reaction of in situ gener-
ated MnR2 reagents with readily available nonfunction-
alized ligands. The scope of the method has been expanded,
thus allowing the synthesis of new 4-alkylated pdi ligands
displaying N-aryl substituents with different degrees of ste-
ric hindrance. A set of Fe and Co complexes (4–9) contain-
ing 4-R pdi ligands (R = neophyl, benzyl, and allyl) have
been synthesized and fully characterized. The presence of
the alkyl substituent at the pyridine ring increases the solu- the literature.[10] FeCl2·4H2O and CoCl2 were purchased from Ald-
rich and used without further purification. The synthesis of
FeCl2(iPrpdi) and CoCl2(iPrpdi) was carried out as shown below
for other complexes. UV/Vis data (CH2Cl2): FeCl2(iPrpdi), λmax (ε,
bility of the new complexes relative to the corresponding
parent complexes, but it has very minor effect on their
structures and physical properties. The impact of the alkyl
substituent on the catalytic properties of the 4-neophyl and
4-benzyl derivatives in ethylene polymerization is also negli-
gible. This suggests that our methodology would be advan-
tageous for the introduction of special groups suitable for
immobilization, in comparison with other methods in
which the linker functionality is placed in positions where
it may have a significant influence on the catalyst activity.
Similarly to the neophyl and benzyl groups, introduction
of a 4-allyl group on the pyridine ring does not alter the
productivity of the corresponding Fe and Co catalysts (6

–1 cm–1) = 294 (3800), 365 sh. (740), 689 (1300) nm; CoCl2(iPrpdi),
λmax (ε, –1 cm–1) = 293 (3800), 362 (1100), 445 sh. (460), 701 (110)
nm.
Synthesis of 4-Allyl-2,6-bis[1-(mesitylimino)ethyl]pyridine (L5): A
pink suspension of thf (20 mL) and anhydrous MnCl2 (326 mg,
2.61 mmol) was treated with Mg(CH2CH=CH2)Cl (4.21 mL of a
1.3  solution) in diethyl ether (5.5 mmol, 2.1 equiv.) at –78 °C. The
mixture was stirred for 10 min and it turned pale green. The cooling
bath was removed, and as the mixture warmed up, its color
changed from pale green to brown. The color changed again to
pale yellow while stirring at room temperature for 10–15 min. The
and 9), but the polymers generated by them are appreciably solution, containing the “Mn(CH2CH=CH2)2” reagent, was trans-
ferred to a yellow solution of Mespdi (884 mg, 2.08 mmol) in thf
less soluble. GPC analyses of the polymer produced by cat-
(60 mL) stirred at –78 °C. This resulted in an instantaneous color
alyst 6 shows much larger values of Mw (up to 180000) and
change to dark brown. Vigorous stirring was continued for 10 min
a larger polydispersity index (Mw/Mn = 21). The origin of
this effect is currently unknown, but it is conceivable that a
self-immobilization phenomenon, arising from the copoly-
merization of the pending allyl group into the growing poly-
at –78 °C, and then the mixture was warmed to room temperature.
On continued stirring for 90 min, its color changed to deep purple,
and then it was quenched with excess anhydrous methanol (8–
10 mL) and the solvents evaporated under vacuum. The remaining
ethylene chain, could be involved.
orange oil was extracted with hexane (3ϫ25 mL) and toluene
(2ϫ20 mL), and the extracts were filtered from a brown precipi-
tate. The oily mixture was taken up in thf (100 mL) and stirred in
air with a catalytic amount of CrO3/K2CO3 (10%) for 1 h at room
temperature. After evaporation, the oil was dissolved in hexane and
Experimental Section
All preparations were carried out under oxygen-free nitrogen by
conventional Schlenk techniques or in a nitrogen-filled glove-box
unless otherwise stated. Microanalyses were performed by the Mi-
croanalytical Service of the Instituto de Investigaciones Químicas
(Sevilla, Spain). Infrared spectra were recorded on a Bruker Vector
22, UV/Vis spectra on a Perkin–Elmer Lambda 12 spectrophotom-
eter, and NMR spectra on Bruker DRX 300, 400 and 500 MHz
spectrometers. The 1H and 13C{1H} resonances of the solvent were
used as the internal standard, but the chemical shifts are reported
with respect to TMS. Magnetic susceptibility measurements were
made on a Sherwood Scientific balance model MSB-Auto. HPLC
grade organic solvents were freshly distilled prior to use. Diethyl
ether and thf were distilled from sodium benzophenone. CH2Cl2
was distilled from CaH2, and hexane, toluene, and benzene from
sodium benzophenone ketyl. Upon collection, all of them were
deoxygenated prior to use.
evaporated again to dryness to afford L4 as a yellow solid. Yield:
0.71 g (78%). H NMR (CDCl3, 298 K, 300 MHz): δ = 8.29 (s, 2
H, m-CHpy), 6.79 (s, 4 H, CHar(aniline)) 3.00 (s, 2 H, CH2CMe2Ph),
1
5.93–5.84 (m,
1 H, CH2CH=CH2), 5.13–5.08 (m, 2 H,
CH2CH=CH2), 3.47 (m, 2 H, CH2CH=CH2), 2.37 (s, 6 H,
CH3C=N), 1.91 (s, 6 H, o-CH3) 1.83 (s, 6 H, m-CH3), 1.81 (s, 6 H,
p-CH3) ppm. ESI-MS: m/z = 438.3 [M + 1]+.
Synthesis of 2,6-Bis{[1-(2-isopropyl-6-methylphenyl)imino]ethyl}-4-
neophylpyridine (L4): A pink suspension of thf (20 mL) and anhy-
drous MnCl2 (187 mg, 1.48 mmol) was treated with
Mg(CH2CMe2Ph)Cl (2.43 mL of a 1.3  solution) in diethyl ether
(3.16 mmol, 2.1 equiv.) at –78 °C. The mixture was stirred for
10 min and it turned pale green. The cooling bath was removed,
and as the mixture warmed up, its color changed from pale green
to brown. The color changed again to pale yellow while stirring
at room temperature for 10–15 min. The solution, containing the
“Mn(CH2CMe2Ph)2” reagent, was transferred to a yellow suspen-
sion of iPr,Mepdi (580 mg, 1.19 mmol) in thf (60 mL) stirred at
–78 °C. This resulted in an instantaneous color change to dark
brown. Vigorous stirring was continued for 10 min at –78 °C, and
then the mixture was warmed to room temperature. On continued
stirring for 90 min, its color changed to deep purple, and then it
The molar mass distribution (MMD) and polydispersity measure-
ments were performed on a high temperature dual-detector size
exclusion chromatography (SEC) system. The SEC system was a
GPCV2000 from Waters (Milford, MA) that uses two on-line detec-
tors: a differential viscometer (DV) and a differential refractometer
(DRI) as concentration detector. The experimental conditions were
1876
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 1871–1879

Iron and Cobalt Complexes of 4-Alkyl-2,6-diiminopyridine Ligands
was quenched with excess anhydrous methanol (8–10 mL) and the
solvents evaporated under vacuum. The remaining orange oil was
extracted with hexane (3ϫ25 mL) and toluene (2ϫ20 mL), and
the extracts were filtered from a brown precipitate. The filtrate was
dried, and the resulting oil (0.65 g, 89%) was found by NMR spec-
troscopy to consist of a mixture of L5 and the corresponding dihy-
dropyridine in a relative ratio of 1:2. The oily mixture was taken
up in thf (100 mL) and stirred in air with a catalytic amount of
CrO3/K2CO3 (10%) for 1 h at room temperature. After evapora-
tion, the residue was crystallized from hexane. Yield: 0.52 g (71%).
1H NMR (CDCl3, 298 K, 300 MHz): δ = 7.95 (s, 2 H, m-CHpy),
7.28–6.99 (m, 11 H, CHar(Ph) and CHar(neof)), 3.00 (s, 2 H,
CHpy), 14.29 (∆ν1/2 = 24 Hz, 4 H, m-CHar), 13.79 (∆ν1/2 = 44 Hz,
1 H, CH2=CHCH2), 10.32 (∆ν1/2 = 34 Hz, 1 H, CH2=CHCH2),
7.72 (∆ν1/2 = 24 Hz, 1 H, CH2=CHCH2), –5.44 [∆ν1/2 = 73 Hz, 12
H, CH(CH3)2], –6.59 [∆ν1/2 = 29 Hz, 12 H, CH(CH3)2], –10.49
(∆ν1/2 = 24 Hz, 2 H, p-CHar), –24.56 [∆ν1/2 = 290 Hz, 4 H,
CH(CH3)2], –30.24 (∆ν1/2 = 78 Hz, 6 H, CH3C=N), –31.09 (∆ν1/2
= 68 Hz, 2 H, CH2=CHCH2) ppm. IR (Nujol): ν(C=N) =
1596 cm–1. UV/Vis (CH2Cl2): λmax (ε, –1 cm–1) = 300 (4600), 304
sh. (4500), 369 sh. (1800), 692 (2500) nm. C36H47Cl2FeN3 (647.25):
calcd. C 66.77, H 7.30, N 6.48; found C 65.11, H 7.12, N 6.12.
Synthesis of CoCl2(L1) (7): A thf solution (10 mL) of L1 (0.674 g,
1.1 mmol) was added dropwise to a stirred suspension of CoCl2
(0.128 g, 1 mmol) in thf (10 mL) at room temperature. The re-
sulting mixture turned from blue to brown. The suspension was
stirred for 24 h at room temperature, and the dark brown solid
was filtered out, washed with hexane (2ϫ10 mL), and dried under
vacuum to yield 0.8 g of a brown powdery solid, which was recyr-
stallized from CH2Cl2/hexane (9:1) at –30 °C. Yield: 0.43 g (58%).
1H NMR: δ = 114.83 (∆ν1/2 = 92 Hz, 2 H, 3,5-CHpy), 26.04 (∆ν1/2
= 17 Hz, 2 H, m-CHar(neof)), 20.15 (∆ν1/2 = 21 Hz, 2 H, o-CHar(neof)
), 14.66 (∆ν1/2 = 15 Hz, 6 H, CMe2(neof)), 10.84 (∆ν1/2 = 22 Hz, 2
3
CH2CMe2Ph), 2.72 [sept, JHH = 5.4 Hz, 2 H, CH(CH3)2], 2.24 (s,
6 H, CH3C=N), 1.95 (s, 6 H, o-CH3), 1.17 (s, 6 H, CH2CMe2Ph),
3
3
1.14 (d, JHH = 5.4 Hz, 6 H), 1.10 [d, JHH = 5.4 Hz, 6 H, CH-
(CH3)2] ppm. 13C{1H} NMR: δ = 17.1 (CH3C=N), 18.4 (Mear)
23.1, 23.4 (CHMe2), 28.5 (CH2CMe2Ph), 29.3 (CHMe2), 39.1
(CH2CMe2Ph), 50.8 (CH2CMe2Ph), 123.4 (3,5-Car), 124.2 (4-Car),
125.3 (3,5-Car(py)), 126.1 (4-Car(Ph)), 127.9 (2,6-Car(Ph)), 128.3 (3,5-
Car(Ph)), 136.6 (Car), 147.8 (Car), 147.9 (Car(py)), 148.9 (Car), 154.6
(Car(py)), 167.4 (CH3C=N) ppm. IR (Nujol mull): ν(C=N) = 1643,
1619, 1591, 1555 cm–1.
H, CH2(neof)), 8.70 (∆ν1/2 = 17 Hz, 1 H, p-CHar(neof)), 8.09 (∆ν1/2
=
18 Hz, 4 H, CHar), 2.10 (∆ν1/2 = 34 Hz, 6 H, CH3C=Nar), –9.41
(∆ν1/2 = 27 Hz, 2 H, p-CHar), –18.86 [∆ν1/2 = 29 Hz, 24 H,
CH(CH3)2], –88.32 [s, 2 H, CH(CH3)2] ppm. IR (Nujol mull):
ν(C=N) = 1597 cm–1. UV/Vis (CH2Cl2): λmax (ε, –1 cm–1) = 302
(4500), 356 (1800), 434 sh. (800), 703 (150) nm. µeff (magnetic
susceptibility balance, 293 K) = 4.67 µB. C43H55Cl2CoN3 (742.31):
calcd. C 69.44, H 7.45, N 5.65; found C 69.15, H 7.50, N 5.11.
Synthesis of FeCl2(L1) (4): A thf solution (10 mL) of L1 (0.674 g,
1.1 mmol) was added dropwise to
a stirred suspension of
FeCl2·4H2O (0.198 g, 1 mmol) in thf (10 mL) at room temperature.
The mixture turned from yellow to green. The suspension was
stirred for 24 h at room temperature, and the green solid was re-
moved by filtration and washed with hexane (2ϫ10 mL) and dried
in vacuo to obtain 0.8 g of a green powdery solid. Yield: 693 mg
(85%). 1H NMR (CD2Cl2, 298 K, 300 MHz): δ = 84.60 (∆ν1/2
=
Synthesis of CoCl2(L2) (8): The synthesis of this complex was car-
156 Hz, 2 H, 3,5-CHpy), 14.59 (∆ν1/2 = 37 Hz, 4 H, m-CHar), 11.08
(∆ν1/2 = 37 Hz, 2 H, m-CHar(neof)), 8.65 (∆ν1/2 = 32 Hz, 2 H, o-
ried out in the same manner as that for 7, but with L2. Yield: 0.26 g
1
(45%). H NMR (CD2Cl2, 298 K, 300 MHz): δ = 116.28 (∆ν1/2
=
CHar(neof)), 7.25 (∆ν1/2 = 32 Hz, 1 H, p-CHar(neof)), 5.04 (∆ν1/2
=
71 Hz, 2 H, m-CHpy), 24.10 (∆ν1/2 = 16 Hz, 2 H, m-CHBz), 24.00
(∆ν1/2 = 16 Hz, 2 H, o-CHBz), 14.50 (∆ν1/2 = 16 Hz, 2 H, CH2, Bz),
12.61 (t, 3JHH = 6.7 Hz, 1 H, p-CHBz), 8.91 (∆ν1/2 = 18 Hz, 4 H, m-
CHar), 3.37 (∆ν1/2 = 27 Hz, 6 H, CH3C=N), –9.18 (∆ν1/2 = 27 Hz, 2
H, p-CHar), –18.62 [∆ν1/2 = 14 Hz, 12 H, CH(CH3)2], –19.52
[∆ν1/2 = 66 Hz, 12 H, CH(CH3)2] –88.75[∆ν1/2 = 255 Hz, 4 H,
37 Hz, 6 H, CMe2), –4.83 [∆ν1/2 = 87 Hz, 12 H, CH(CH3)2], –6.13
[∆ν1/2 = 41 Hz, 12 H, CH(CH3)2], –10.33 (∆ν1/2 = 41 Hz, 2 H, p-
CHar) –13.74 (∆ν1/2 = 105 Hz, 2H, CH2(neof)) –23.07 [∆ν1/2
=
344 Hz, 4 H, CH(CH3)2], –27.39 (∆ν1/2 = 170 Hz, 6 H, CH3C=N)
ppm. IR (Nujol mull): ν(C=N) = 1625, 1590 cm–1. UV/Vis
(CH2Cl2): λmax (ε, –1 cm–1) = 298 (3800), 309 sh. (8100), 360 sh.
(4600), 689 (1400) nm. µeff (magnetic susceptibility balance, 293 K)
= 5.39 µB.
CH(CH3)2] ppm. IR (Nujol): ν(C=N)
=
1597 cm–1. UV/Vis
(CH2Cl2): λmax (ε, –1 cm–1) = 300 (4800), 357 (2100), 435 sh. (980),
703 (190) nm. µeff (magnetic susceptibility balance, 293 K) =
4.90 µB. C40H49Cl2CoN3 (700.26): calcd. C 68.47, H 7.04, N 5.99;
found C 67.89, H 6.86, N 5.73.
Attempts to grow crystals of this compound from a CH2Cl2/hexane
(9:1) mixture also afforded a crop of brown crystals of compound
10.
Synthesis of CoCl2(L3) (9): This compound was obtained in the
same manner as that described for the synthesis of 7 and 8, as a
crystalline brown solid. Yield: 0.36 g (55%). 1H NMR (CD2Cl2,
298 K, 300 MHz): δ = 116.26 (∆ν1/2 = 117 Hz, 2 H, m-CHpy), 25.35
(∆ν1/2 = 71 Hz, 1 H, CH2=CHCH2), 23.49 (∆ν1/2 = 47 Hz, 2 H,
CH2=CHCH2), 19.74 (∆ν1/2 = 51 Hz, 1 H, CH2=CHCH2), 15.42
(∆ν1/2 = 47 Hz, 1 H, CH2=CHCH2), 8.9 (∆ν1/2 = 51 Hz, 4 H, m-
CHar), 3.40 (∆ν1/2 = 68 Hz, 6 H, CH3C=N), –9.17 (∆ν1/2 = 53 Hz,
2 H, p-CHar), –18.66 [∆ν1/2 = 47 Hz, 12 H, CH(CH3)2], –19.45
[∆ν1/2 = 108 Hz, 12 H, CH(CH3)2], –89.09 [∆ν1/2 = 438 Hz, 4 H,
Synthesis of FeCl2(L2) (5): The synthesis of this compound was
carried out in the same manner and scale as that of 4, but with L2.
Yield: 0.35 g (45%). 1H NMR (CD2Cl2, 298 K, 300 MHz): δ =
83.40 (∆ν1/2 = 41 Hz, 2 H, 3,5-CHpy), 14.45 (∆ν1/2 = 18 Hz, 4 H,
m-CHar), 13.22 (∆ν1/2 = 18 Hz, 2 H, m-CHBz), 9.68 (∆ν1/2 = 18 Hz,
2 H, o-CHBz), 9.31(∆ν1/2 = 18 Hz, 1 H, o-CHBz), –5.58 [∆ν1/2
=
55 Hz, 12 H, CH(CH3)2], –6.76 [∆ν1/2 = 18 Hz, 12 H, CH(CH3)2],
–10.59 (∆ν1/2 = 23 Hz, 2 H, p-CHar), –25.16 [∆ν1/2 = 234 Hz, 4 H,
CH(CH3)2], –32.74 (∆ν1/2 = 36 Hz, 6 H, CH3C=N), –35.04 (∆ν1/2
= 14 Hz, 2 H, CH2(Bz)) ppm. IR (Nujol mull): ν(C=N) = 1593 cm–1.
UV/Vis (CH2Cl2): λmax (ε, –1 cm–1) = 296 (6700), 305 sh. (6200),
367 sh. (2300), 689 (2200) nm. µeff (magnetic susceptibility balance,
293 K) = 5.24 µB. C40H49Cl2FeN3 (697.25): calcd. C 68.77, H 7.07,
N 6.02; found C 68.55, H 7.35, N 6.08.
CH(CH3)2] ppm. IR (Nujol): ν(C=N)
=
1598 cm–1. UV/Vis
(CH2Cl2): λmax (ε, –1 cm–1) = 299 (6900), 350 (2300), 430 sh. (980),
703 (150) nm. µeff (magnetic susceptibility balance, 293 K) =
4.79 µB. C36H47Cl2CoN3 (650.25): calcd. C 66.36, H 7.27, N 6.45;
found C 65.79, H 7.29, N 6.24.
X-ray Structure Determination
Synthesis of FeCl2(L3) (6): The synthesis of this compound was
carried out in a similar manner to that of 4, but with diethyl ether
as solvent and L3. Yield: 0.41 g of a blue solid (58%). 1H NMR
(CD2Cl2, 298 K, 400 MHz): δ = 83.83 (∆ν1/2 = 83 Hz, 2 H, 3,5-
Crystal Data for 4: C44H57Cl4FeN3 [C43H55Cl2FeN3, CH2Cl2], Mw
=
825.58.
A single crystal of suitable size, green block
(0.19ϫ0.13ϫ0.10 mm) from CH2Cl2/thf, coated with dry perfluo-
Eur. J. Inorg. Chem. 2008, 1871–1879
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1877

J. Cámpora et al.
FULL PAPER
ropolyether, was mounted on a glass fiber and fixed in a cold nitro-
gen stream, 100(2) K, to the goniometer head. Monoclinic, space
group Cc (no. 9), a = 22.5569(19) Å, b = 12.1920(9) Å, c =
Acknowledgments
Financial support from the Ministerio de Educación y Ciencia
(Project PPQ2003-000975) and Junta de Andalucía is gratefully ac-
knowledged. Support for personnel exchange from the European
Union Network of Excellence IDECAT (contract No NMP3-CT-
2005-011730) is gratefully acknowledged. A. R.-D. thanks the Min-
isterio de Educación y Ciencia for a Ramón y Cajal contract and
A. M. N for a predoctoral research grant from CSIC (I3P pro-
gram). We thank Mrs. M. C. Jimenez de Haro for SEM images
of the polyethylene samples and Mr. A. Giacometti Schieroni for
carrying out the GPC analysis.
17.623(3) Å, β = 119.276(2)°, V = 4227.5(9) Å3, Z = 4, ρcalcd.
=
=
1.297 gcm–3, λ(Mo-Kα1
) = 0.71073 Å, F(000) = 1744, µ
0.644 mm–1. 29461 Reflections were collected from a Bruker-Non-
ius X8Apex-II CCD diffractometer in the range 5.72 Ͻ 2θ Ͻ
61.00°, and 12129 independent reflections [R(int) = 0.0482] were
used in the structural analysis. The data were reduced (SAINT)
and corrected for Lorentz polarisation effects and absorption by
the multiscan method applied by SADABS.[19,20] The structure was
solved by direct methods (SIR-2002)[21] and refined against all F2
data by full-matrix least-squares techniques (SHELXL97)[22] con-
verged to final R1 = 0.0396 [I Ͼ 2σ(I)] and wR2 = 0.0808 for all
data, with a goodness-of-fit on F2, S = 0.996 and 469 parameters.
[1] a) B. L. Small, M. Brookhart, A. M. A. Bennett, J. Am. Chem.
Soc. 1998, 120, 4049–4050; b) G. J. P. Britovsek, V. C. Gibson,
B. S. Kimberley, P. J. Maddox, S. J. Mc Tavish, G. A. Solan,
A. J. P. White, D. J. Williams, Chem. Commun. 1998, 849–850.
[2] a) V. C. Gibson, C. Redshaw, G. A. Sola, Chem. Rev. 2007, 107,
1745–1776; b) C. Bianchini, G. Giambastiani, I. Guerrero Rios,
G. Mantovani, A. Meli, A. M. Segarra, Coord. Chem. Rev.
2007, 250, 1391–1418.
Crystal Data for 7: C44H57Cl4CoN3 [C43H55Cl2CoN3, CH2Cl2], Mw
=
828.66.
A
single crystal of suitable size, red prism
(0.48ϫ0.14ϫ0.13 mm) from CH2Cl2, coated with dry perfluo-
ropolyether, was mounted on a glass fiber and fixed in a cold nitro-
gen stream, 100(2) K, to the goniometer head. Monoclinic, space
group Cc (no. 9), a = 22.4998(16) Å, b = 12.1667(9) Å, c =
[3] J. R. Severn, J. C. Chadwick, R. Duchateau, N. Friederichs,
Chem. Rev. 2005, 105, 4073–4147.
17.618(2) Å, β = 119.213(2)°, V = 4209.5(6) Å3, Z = 4, ρcalcd.
=
=
[4] See for exsample: a) V. C. Gibson, S. K. Spitzmesser, Chem.
Rev. 2003, 103, 283–315; b) B. L. Small, M. Brookhart, Macro-
molecules 1999, 32, 2120–2130; c) G. J. P. Britovsek, M. Bruce,
V. C. Gibson, B. S. Kimberley, P. J. Maddox, S. Mastroianni,
S. J. McTavish, C. Redshaw, G. A. Solan, S. Strömberg, A. J. P.
White, D. J. Williams, J. Am. Chem. Soc. 1999, 121, 8728–8740;
d) Y. Chen, C. Qian, J. Sun, Organometallics 2003, 22, 1231–
1321; e) K. P. Tellmann, V. C. Gibson, A. J. P. White, D. J. Wil-
liams, Organometallics 2005, 24, 280–286; f) A. S. Ionkin, W. J.
Marshall, D. J. Adelman, B. Bobik Fones, B. M. Fish, M. F.
Schiffhauer, Organometallics 2006, 25, 2978–2992.
[5] a) C. Amort, M. Malaun, A. Krajete, H. Kopacka, K. Wurst,
M. Christ, D. Lilge, M. O. Kristen, B. Bildstein, Appl. Or-
ganomet. Chem. 2002, 16, 506–516; b) C. Bianchini, G. Giam-
bastiani, G. Mantovani, A. Meli, D. Mimeau, J. Organomet.
Chem. 2004, 689, 1356–1361; c) G. J. P. Britovsek, V. C. Gib-
son, B. S. Kimberley, S. Mastroianni, C. Redshaw, G. A. Solan,
A. J. P. White, D. J. Williams, J. Chem. Soc., Dalton Trans.
2001, 1639–1644; d) G. A. Luinstra, J. Queisser, B. Bildstein,
H.-H. Görtz, C. Amort, M. Malun, A. Krajete, G. Werne,
M. O. Kristen, N. Huber, C. Genert, in Late Transition Metal
Polymerization Catalysts (Eds.: B. Rieger, L. Saunders Baugh,
S. Kacker, S. Strigler), Wiley-VCH, Weinheim, Germany, 2003.
[6] a) R. Schmidt, M. B. Welch, S. J. Palackal, H. G. Alt, J. Mol.
Catal. A 2002, 179, 155–173; b) S. McTavish, G. J. P. Britovsek,
T. M. Smit, V. C. Gibson, A. J. P. White, D. J. Williams, J. Mol.
Catal. A 2007, 261, 293–300; c) A. M. Archer, M. W. Bouwk-
amp, M.-P. Cortez, E. Lobkovsky, P. J. Chirik, Organometallics
2006, 25, 4269–4278.
1.308 gcm–3, λ(Mo-Kα1
) = 0.71073 Å, F(000) = 1748, µ
0.696 mm–1. 53090 Reflections were collected from a Bruker-Non-
ius X8Apex-II CCD diffractometer in the range 5.74 Ͻ 2θ Ͻ
61.14°, and 12060 independent reflections [R(int) = 0.0329] were
used in the structural analysis. Data reduction, solving and refine-
ment was performed in the same manner as that for 4 converged
to final R1 = 0.0270 [I Ͼ 2σ(I)] and to wR2 = 0.0647 for all data,
with a goodness-of-fit on F2, S = 1.047 and 469 parameters.
Crystal Data for 10: C43H55Cl4Fe2N3O, Mw = 883.40. A single
crystal of suitable size, orange needle (0.58ϫ0.15ϫ0.12 mm) from
CH2Cl2/thf/hexane, coated with dry perfluoropolyether, was
mounted on a glass fiber and fixed in a cold nitrogen stream,
100(2) K, to the goniometer head. Monoclinic, space group C2/c
(no. 15), a = 30.4435(14) Å, b = 17.1922(8) Å, c = 23.7690(18) Å,
β = 128.0240(10)°, V = 9800.0(10) Å3, Z = 8, ρcalcd. = 1.197 gcm–3,
λ(Mo-Kα1) = 0.71073 Å, F(000) = 3696, µ = 0.842 mm–1. 58740
Reflections were collected from a Bruker-Nonius X8Apex-II CCD
diffractometer in the range 4.28 Ͻ 2θ Ͻ 52.80°, and 9959 indepen-
dent reflections [R(int) = 0.0531] were used in the structural analy-
sis. Data reduction, solving and refinement was performed in the
same manner as that for 4 converged to final R1 = 0.0353 [I Ͼ
2σ(I)] and to wR2 = 0.0974 for all data, with a goodness-of-fit on
F2, S = 1.054 and 490 parameters.
CCDC-666530 for 4, CCDC-666531 for 7, and CCDC-666532 for
10 contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[7] T. M. Smit, A. K. Tomov, V. C. Gibson, A. J. P. White, D. J.
Williams, Inorg. Chem. 2004, 43, 6511–6512.
[8] G. J. P. Britovsek, V. C. Gibson, O. D. Hoarau, S. K. Spitz-
messer, A. J. P. White, D. J. Williams, Inorg. Chem. 2003, 42,
3454–3465.
Ethylene Polymerization Experiments: Polymerization reactions
were carried out in round-bottom flasks connected to a vacuum
line operating with ethylene. The temperature was maintained with
an external water bath thermostated at 30 °C. After the flasks were
evacuated for 15 min, they were charged with toluene (100 mL) and
flushed with ethylene at the line pressure (1.1 bar). A solution of
the cocatalyst (MAO, 500 equiv.) was added from a pipette, fol-
lowed by a solution of CH2Cl2 (2 mL) and precatalyst (4 µmol).
After stirring for 10 min, the reaction mixture was quenched with
acidified ethanol. The resultant polymer was filtered and dried in
a vacuum oven at 50 °C until a constant weight was obtained.
[9] M. Seitz, W. Milius, H. G. Alt, J. Mol. Catal. A 2007, 261, 246–
253.
[10] F. Pelascini, M. Wesolek, F. Peruch, P. J. Lutz, Eur. J. Inorg.
Chem. 2006, 4309–4316.
[11] a) S. Milione, C. Cavallo, C. Tedesco, A. Grassi, J. Chem. Soc.,
Dalton Trans. 2002, 1839–1846; b) A. M. A. Bennett, World
Patent, WO9827124, 1998; c) P. Barbaro, C. Bianchini, G. Gi-
ambastiani, I. Guerrero Rios, A. Meli, W. Oberhauser, A. M.
Segarra, L. Sorace, A. Toti, Organometallics 2007, 26, 4639–
4651; d) C. Bianchini, G. Giambastiani, I. Guerrero Rios, A.
Meli, W. Oberhauser, L. Sorace, A. Toti, Organometallics 2007,
26, 5066.
1878
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 1871–1879

Iron and Cobalt Complexes of 4-Alkyl-2,6-diiminopyridine Ligands
[12] J. Cámpora, C. M. Pérez, A. Rodríguez-Delgado, A. Mar- [18] C. Görl, H. G. Alt, J. Mol. Catal. A 2007, 273, 118–132.
cos Naz, P. Palma, E. Álvarez, Organometallics 2007, 26, 1104–
[19] Apex 2, version 2.1, Bruker AXS Inc., Madison, Wisconsin,
USA, 2004.
1107.
[13] G. J. P. Britovsek, V. C. Gibson, S. K. Spitzmesser, K. P. [20] SAINT and SADABS, Bruker AXS Inc., Madison, Wisconsin,
Tellmann, A. J. P. White, D. J. Williams, J. Chem. Soc., Dalton
Trans. 2002, 1159–1171.
USA, 2001.
[21] SIR2002: The Program: M. C. Burla, M. Camalli, B. Carroz-
zini, G. L. Cascarano, C. Giacovazzo, G. Polidori, R. Spagna,
J. Appl. Crystallogr. 2003, 36, 1103.
[14] H. G. Alt, Dalton Trans. 2005, 3271–3276.
[15] F. A. R. Kaul, G. T. Puchta, H. Schneider, F. Bielert, D. Mih-
alios, W. A. Herrmann, Organometallics 2002, 21, 74–82.
[16] G. Jin, D. Zhang, J. Polym. Sci. Part A 2004, 42, 1018–1024.
[17] B. L. Small, A. J. Marcucci, Organometallics 2001, 20, 5738–
5744.
[22] G. M. Sheldrick, SHELXL97, University of Göttingen, Ger-
many, 1997.
Received: November 12, 2007
Published Online: February 22, 2008
Eur. J. Inorg. Chem. 2008, 1871–1879
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1879