Synthesis and molecular structures of nickel(II) and cobalt(III) complexes with 2-(arylimino)-3-(hydroxyimino)butane

Article abstract of DOI:10.1002/ejic.200300030

We report new series of Ni^II and Co^III complexes with nitrogen-donor chelating ligands of the (E,E)-2-(arylimino)-3-(hydroxyimino) butane type (Ar-N∩N-OH). These ligands are characterized by a hydrophilic (OH group) and a hydrophobic region (aryl group). Ni^II derivatives were obtained either as trimers of formula [Ni₃(Ar-N∩N-OH)₃Br₄(OH)][Br], with the hydrophobic groups oriented on the same side, or as bis(chelated) derivatives with cis geometry, depending on the steric hindrance of the aryl groups. Co^III complexes were obtained only as bis(chelated) derivatives, with the two ligands coplanar. The "iso-oriented" arrangement of ligands in bis(chelated) Co^III complexes is favored by weak interactions between the two ligands, namely O-H...O hydrogen bond and stacking interactions between the aryl groups. Co^III complexes might find application as catalysts for living or controlled radical polymerization of polar olefins, and preliminary results are reported. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejic.200300030

FULL PAPER
Synthesis and Molecular Structures of Nickel(II) and Cobalt(III) Complexes
with 2-(Arylimino)-3-(hydroxyimino)butane
Ennio Zangrando,*[a] Manuela Trani,[b] Elisa Stabon,[a] Carla Carfagna,[c] Barbara Milani,[a]
and Giovanni Mestroni[a]
Keywords: Cobalt / Crystal structures / N-donor ligands / Nickel / Stacking interactions
We report new series of NiII and CoIII complexes with nitro-
gen-donor chelating ligands of the (E,E)-2-(arylimino)-3-(hy- bis(chelated) CoIII complexes is favored by weak interactions
droxyimino)butane type (Ar−NʝN−OH). These ligands are
between the two ligands, namely O−H···O hydrogen bond
characterized by a hydrophilic (OH group) and a hydro- and stacking interactions between the aryl groups. CoIII com-
gands coplanar. The ‘‘iso-oriented’’ arrangement of ligands in
phobic region (aryl group). NiII derivatives were obtained
either as trimers of formula [Ni3(Ar−NʝN−OH)3Br4(OH)][Br],
with the hydrophobic groups oriented on the same side, or
as bis(chelated) derivatives with cis geometry, depending on
the steric hindrance of the aryl groups. CoIII complexes were
obtained only as bis(chelated) derivatives, with the two li-
plexes might find application as catalysts for living or con-
trolled radical polymerization of polar olefins, and prelimin-
ary results are reported.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Introduction
Recent studies report that the molecular weights of poly-
mers synthesized from olefin monomers with the aid of
transition-metal catalysts or promoters can be tuned by the
steric hindrance of the ancillary ligands.[1Ϫ6] In particular,
the steric properties can be modulated by the introduction
of 2- or 2,6-substituted aryl groups in planar bi-, tri- and
tetradentate nitrogen-donor chelates. Some notable ex-
amples are reported in Figure 1.
When the substituted aryl groups are arranged perpen-
dicular to the plane of the chelate, they create a particular
steric hindrance above and below the plane of the chelate
itself (Figure 1). Bi- and tridentate ligands of variable steric
hindrance are easily accessible thanks to the synthetic pro-
cedure based on the condensation reaction between biacetyl
(or 2,6-diacetylpyridine) and an appropriate aromatic am-
ine.[1,2] On the other hand, convenient modulation of the
steric hindrance is more difficult to achieve for tetradentate
N-donor chelating ligands such as porphyrins or bis(dime-
thylglyoximate) derivatives.
Figure 1. Examples of bi-, tri-, and tetradentate nitrogen-donor
ligands and of the corresponding complexes
[a]
`
Dipartimento di Scienze Chimiche, Universita di Trieste,
Via L. Giorgeri 1, 34127 Trieste, Italy
Fax: (internat.) ϩ 39-040/5583903
E-mail: zangrando@univ.trieste.it
Department of Chemistry, Sanford S. Atwood Chemistry
Center,
[b]
[c]
Such nitrogen-donor chelating ligands are of particular
interest in the field of homogeneous catalysis. Depending
on the nature of the metal, the olefin, and the experimental
conditions, different reactions can be performed: nickel
1515 Pierce Dr, Atlanta, GA 30322, USA
`
Istituto di Scienze Chimiche, Universita di Urbino,
Piazza Rinascimento 6, 61029 Urbino, Italy
Eur. J. Inorg. Chem. 2003, 2683Ϫ2692
DOI: 10.1002/ejic.200300030
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2683

E. Zangrando, M. Trani, E. Stabon, C. Carfagna, B. Milani, G. Mestroni
FULL PAPER
complexes with bidentate chelating ligands or cobalt and
These ligands are characterized by a hydrophilic (OH)
iron complexes with tridentate chelating ligands, in the pres- and a hydrophobic region (Ar). Therefore, with a ligand/
ence of MAO, promote the oligomerization or polymeriz- metal ratio of 2:1, the formation of planar bis(chelated)
ation of ethylene or propylene by
a coordinative species can be predicted, thanks to the possibility of estab-
mechanism.[1Ϫ5] Moreover, cobalt complexes with planar lishing weak interactions between the two molecules of Ar-
tetradentate chelating ligands have successfully been used NʝNϪOH, such as the hydrogen bond between the oxygen
in oligomerization and polymerization of polar olefins by a atoms and/or the stacking interaction between the aromatic
radical mechanism in the presence of AIBN [2,2Ј- rings (Figure 2, b). Unlike that in the planar bis(dimethyl-
azobis(isobutyronitrile)].[6Ϫ10] In particular, organometallic glyoxime) derivatives, the steric bulk in these complexes is
CoIII complexes with planar sterically hindered N-donor directed normally to the equatorial plane and can be modu-
tetradentate chelating ligands polymerize polar olefins by a lated by varying the nature of substituents in the 2- and the
living radical mechanism.[6] In the absence of steric bulk, 6-positions of the aromatic ring.
these complexes promote the oligomerization of the same
Some results on the catalytic activity of the cobalt deriva-
substrates, yielding oligomers characterized by terminal Cϭ tives in the polymerization of methyl methacrylate (MMA)
C double bonds.[7] Cobalt derivatives with dimethylglyox- are also presented.
ime (cobaloximes)[9] and analogous complexes, in which the
bridging H atom is replaced by a BF2 group,[10] turned out
to be very efficient and cheap oligomerization catalysts. The Results and Discussion
presence of two OH···O bridging moieties in cobaloxime
Synthesis and Characterization of 2-(Arylimino)-3-
(hydroxyimino)butane Ligands 1a؊1i
derivatives ensures coplanarity of the two molecules of li-
gand, which is necessary for these complexes to show cata-
lytic activity.[7]
The chelating ligands (E,E)-2-(arylimino)-3-(hydroxyim-
ino)butane are obtained by condensation of biacetyl
monoxime with a series of different substituted aromatic
amines (Scheme 1).
Interestingly, the effect of steric hindrance induced by
chelating ligands is the same regardless of the type of poly-
merization mechanism. In both cases the chain termination
step, a β-H elimination process, proceeding either through
a coordinative[1] or a radical mechanism,[9,11,12] is slowed
down with increasing bulkiness of the aryl moiety.
Very recently we reported that complexes of type cis-
[Co(CH2Ph)2(NϪN)2][PF6] (NϪN ϭ bipy, phen) promote
acrylonitrile polymerization by a controlled radical mecha-
nism,[13] with behavior intermediate between that exhibited
by cobalt complexes with sterically hindered and nonhin-
dered planar tetradentate N-donor chelating ligands.
In this study we report the synthesis and characterization
of nickel and cobalt complexes with a series of bidentate
N-donor chelating ligands: namely, (E,E)-2-(arylimino)-3-
(hydroxyimino)butanes (ArϪNʝNϪOH) (Figure 2, a). The
ligands in which the aryl group is phenyl or o-tolyl, are
known[14,15] and the series has now been extended to other
substituted amines.
Scheme 1. Synthesis of ArϪNʝNϪOH ligands 1aϪ1i
Syntheses were conducted in methanol at reflux for 5 h,
starting from stoichiometric amounts of reagents, and in
the presence of formic acid in catalytic amount. All ligands
were obtained in high yields as white, crystalline solids.
Crystals of 1h suitable for X-ray diffraction analysis were
obtained upon slow cooling of the reaction mixture.
Its structural characterization, with regard to the two
NϭC double bonds, shows a syn-methyl,syn-methyl confor-
mation (Figure 3) that would require a rotation about the
C(2)ϪC(3) bond in order to behave as a chelating ligand.
The planarity of the OϪNϭCϪCϭN fragment, with a tor-
sion angle of 162.7(5)°, is indicative of π-electron delocaliz-
ation in the heterodiene fragment, as confirmed by the
˚
C(2)ϪC(3) bond length of 1.462(6) A. This in fact appears
˚
to be slightly shorter than the accepted value of 1.48 A for
a C(sp2)ϪC(sp2) single bond.[16] It may be of interest to
point out the difference between the CϭNϪC angles at the
N(2) imino and the N(1) oxime nitrogen atoms [120.1(5)
and 111.8(4)°, respectively]. The mesitylene ring is roughly
Figure 2. (a) Generic 2-(arylimino)-3-(hydroxyimino)butane ligand
(ArϪNʝNϪOH);(b) corresponding bis(chelated) planar complex
2684
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2683Ϫ2692

Nickel(II) and Cobalt(III) Complexes with 2-(Arylimino)-3-(hydroxyimino)butane
FULL PAPER
perpendicular to the planar residue of the molecule, making Synthesis and Characterization of the NiII Complexes 2a,
a dihedral angle of 73°, an arrangement induced by steric 2b, 2d, 2e, 2h, 2i; 3h
hindrance between the groups. The crystal packing consists
Synthesis of complexes was carried out in CH2Cl2, at
of chains with a zigzag motif, running along the crystallo-
room temperature, for 3 h, starting from [NiBr2(C4H10O2)]
graphic c axis, where O(1)ϪH ··N(2) hydrogen bonds
(C4H10O2 ϭ ethylene glycol dimethyl ether) and using a li-
˚
[O···N ϭ 2.749(5) A] connect adjacent molecules. The mo-
gand/nickel ratio of 1:1 or 2:1 (Scheme 2, a). The products
were isolated as brown solids upon addition of diethyl ether
or n-pentane to the reaction mixture.
lecular geometry and the packing resemble those found in
the p-tolyl analogue.[15]
Figure 3. Molecular structure of ligand 1h (ORTEP drawing, ther-
˚
mal ellipsoids 40% probability); selected bond lengths [A] and
angles [°]: O(1)ϪN(1) 1.395(4), N(1)ϪC(2) 1.283(5), N(2)ϪC(3)
1.269(5), N(2)ϪC(5) 1.434(6), C(1)ϪC(2) 1.487(5), C(2)ϪC(3)
1.462(6), C(3)ϪC(4) 1.493(5); C(2)ϪN(1)ϪO(1) 111.8(4),
C(3)ϪN(2)ϪC(5) 120.1(5)
1
The ligands have been characterized by H NMR spec-
1
troscopy in CDCl3 (Table 1). H NMR spectra exhibit two
characteristic regions: aromatic (protons of the aryl groups
at δ ഠ 6Ϫ7.5 ppm) and aliphatic (methyl groups of the
biacetyl monoxime backbone and aromatic ring substitu-
Scheme 2. Synthesis of (a) nickel and (b) cobalt complexes
ents at δ ഠ 2Ϫ3 ppm). A broad peak belonging to the OH
The products of the reaction were characterized by el-
group is also visible between δ ഠ 7.80 and 8.40 ppm. The emental analyses (Table 2) and, depending on the nature of
assignment of signals to protons was based on NOE and the ligand, complexes of different stoichiometry were ob-
COSY experiments. The signal of methyl group (A) in all tained. In particular, when ligands of low steric hindrance,
the ligands was further downfield than that of the methyl such as 1a and 1b, were used, the products were bis(chel-
group (B) (Figure 2, a); for the aryl protons, a downfield ated) complexes of formula [NiBr2(ArϪNʝNϪOH)2] (2a,
shift is observed on going from the β to the γ and to the α 2b), regardless of the ligand/nickel ratio used in the syn-
protons. The latter trend was also observed for alkyl sub- thesis. With the more hindered ligands 1d, 1e, and 1i, a
stituents on the aromatic ring. These chelating ligands were monochelated species was obtained when the synthesis was
used for the synthesis of both NiII and CoIII complexes.
carried out with a ligand/nickel ratio of 1:1 (2d, 2e, 2i). No
1
Table 1. H NMR spectroscopic data for ligands 1aϪ1i
Ligand[a] CH3 (A) CH3 (B) OH
Hα
HαЈ
Hβ
HβЈ
Hγ
1a
1b
1c
1d
1e
1f
1g
1h
1i
2.20 (s) 2.01 (s) 8.25 (t) 6.72 (d)
2.23 (s) 2.05 (s) 8.19 (s) 1.94 (s)
6.72 (d)
6.56 (d)
6.62 (d)
6.51 (d)
6.40 (d)
7.03 (d)
6.34 (s)
1.93 (s)
7.30
7.30
7.09 (t)
7.16Ϫ7.20 (m) 7.16Ϫ7.20 (m) 7.00 (t)
2.18 (s) 2.00 (s) n.d.
6.62 (d)
7.12 (d)
7.29 (d)
7.39 (d)
7.03 (d)
2.30 (s)
6.85 (s)
7.12 (d)
7.05Ϫ7.17 (m) 7.05Ϫ7.17 (m)
7.15 (t)
6.91 (t)
2.30 (s)
6.85 (s)
2.33 (s)
2.24 (s) 1.99 (s) 7.85 (s) 1.15 (d), 2.88 (m)
2.26 (s) 2.01 (s) 8.39 (s) 1.30 (s)
2.25 (s) 1.84 (s) 7.90 (s) 1.96 (s)
2.18 (s) 2.00 (s) 7.90 (s) 6.34 (s)
2.25 (s) 1.83 (s) 8.14 (s) 1.93 (s)
7.05 (t)
6.72 (s)
2.22 (s)
2.27 (s) 1.87 (s) 8.34 (s) 1.13 (d), 2.58Ϫ2.65 (m) 1.13 (d), 2.58Ϫ2.65 (m) 7.05Ϫ7.15 (m) 7.05Ϫ7.15 (m)
[a]
Spectra recorded in CDCl3, at room temperature; δ, ppm; s ϭ singlet, d ϭ doublet, t ϭ triplet, m ϭ multiplet; n.d. ϭ not determined;
the signals of the protons of the alkyl substituents on the aryl group are reported in the place of the corresponding protons.
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2683Ϫ2692
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2685

E. Zangrando, M. Trani, E. Stabon, C. Carfagna, B. Milani, G. Mestroni
FULL PAPER
Table 2. Elemental analyses for the nickel complexes 2a, 2b, 2d, 2e,
2h, 2i and 3h and the cobalt complexes 4aϪ4d, 4fϪ4h, 5aϪ5d,
5fϪ5h
Ni complexes (yield) C[a]
H[a]
N[a]
2a (53%)
2b (51%)
2d[b] (57%)
2e[b] (44%)
2h (60%)
2i[b] (53%)
3h (52%)
41.30 (42.10) 4.13 (4.10)
42.90 (44.11) 4.63 (4.71)
35.20 (37.54) 4.39 (4.44)
36.80 (37.29) 4.49 (4.47)
48.80 (47.67) 5.87 (5.53)
40.80 (40.13) 5.26 (5.05)
38.80 (40.10) 4.65 (4.70)
9.66 (9.81)
9.23 (9.35)
6.34 (6.74)
6.01 (6.34)
8.83 (8.55)
5.60 (5.84)
6.87 (7.20)
Co complexes (yield) C[a]
H[a]
N[a]
4a (71%)
5a (83%)
4b (75%)
5b (76%)
4c (80%)
5c (78%)
4d (82%)
5d (80%)
4f (72%)
5f (74%)
4g (74%)
5g (80%)
4h (73%)
5h (73%)
49.20 (49.91) 4.71 (4.81) 11.50 (11.64)
42.10 (42.13) 4.00 (4.06) 9.84 (9.82)
51.10 (51.88) 5.28 (5.34) 11.00 (11.00)
44.20 (44.10) 4.58 (4.71) 9.39 (9.35)
51.70 (51.88) 5.35 (5.34) 11.00 (11.00)
Figure 4. Molecular structure of compound 2a (ORTEP drawing,
44.10 (44.10) 4.51 (4.71)
54.90 (55.23) 6.25 (6.24)
47.72 (47.72) 5.39 (5.39)
9.41 (9.35)
9.81 (9.91)
8.59 (8.56)
˚
thermal ellipsoids 40% probability); selected bond lengths [A] and
angles [°]: NiϪN(1) 2.026(8), NiϪN(2) 2.109(7), NiϪN(3) 2.018(8),
NiϪN(4) 2.094(8), NiϪBr(1) 2.561(1), NiϪBr(2) 2.561(2);
N(1)ϪNiϪN(2) 76.4(3), N(3)ϪNiϪN(4) 75.8(3), N(1)ϪNiϪN(3)
172.5(3), N(2)ϪNiϪBr(1) 166.1(3), N(4)ϪNiϪBr(2) 165.0(2),
Br(1)ϪNiϪBr(2) 93.38(5)
53.30 (53.64) 5.78 (5.81) 10.20 (10.24)
45.50 (45.95) 4.90 (5.14) 8.77 (8.93)
53.60 (53.64) 5.80 (5.81) 10.40 (10.42)
46.30 (45.95) 5.03 (5.14)
55.40 (55.23) 6.29 (6.24)
46.80 (47.72) 5.36 (5.39)
9.01 (8.93)
9.93 (9.91)
8.36 (8.56)
[a]
[b]
Calculated values in parentheses. The yield was calculated by
formulating the complex as [NiBr2(ArϪNʝNϪOH)].
bis(chelated) species was isolated when a ligand/nickel ratio
Ͼ 1:1 was used. For the ligand with the mesityl group, 1h,
the bis(chelated) derivative [NiBr2(1h)2] (2h) was the prod-
uct when the synthesis was performed with an excess of
ligand. On the other hand, a monochelated species was ob-
tained when a ligand/nickel ratio of 1:1 was used.
Single crystals suitable for X-ray analysis were obtained
for the complexes with ligands 1a (Figure 4) and 1h (Fig-
ure 5), by slow precipitation from a dichloromethane solu-
tion with diethyl ether.
The structural determination of complex 2a confirms its
formulation as a bis(chelated) species with the nickel atom
in a distorted octahedral coordination geometry. The two
bromide ions are in a cis arrangement and the two cis chel-
ating ligands 1a are bound to the metal atom in a head-to-
tail configuration with a dihedral angle of 85.8(4)° (Fig-
ure 4). The NiϪN coordination bond lengths involving ox-
ime nitrogen donors trans to bromide are longer than the
NiϪN(1) and NiϪN(3) bonds trans to each other by about
Figure 5. Molecular structure of compound 3h; for sake of clarity
aryl C atoms are indicated as spheres of fixed radius; selected bond
˚
lengths [A] and angles [°]: NiϪNiЈЈ 3.026(2), NiϪBr(1) 2.598(2),
NiϪBr(2Ј) 2.551(2), NiϪBr(2) 2.578(2), NiϪO(2) 1.999(5),
NiϪN(1) 2.024(8), NiϪN(2) 2.051(8); N(1)ϪNiϪN(2) 78.6(4),
O(2)ϪNiϪN(1)
O(2)ϪNiϪBr(1)
O(2)ϪNiϪBr(2Ј)
N(2)ϪNiϪBr(1)
105.9(4),
76.8(3),
82.73(5),
98.8(2),
O(2)ϪNiϪN(2)
O(2)ϪNiϪBr(2)
N(1)ϪNiϪBr(1)
Br(2)ϪNiϪBr(2Ј)
175.5(4),
82.02(5),
175.4(3),
164.67(7),
˚
˚
0.08 A. The NiϪBr bond lengths of 2.561(2) A are between
the values found for the bridging BrϪ in 3h (see below). The
N(2)ϪNiϪBr(1) and N(4)ϪNiϪBr(2) bond angles of about
165.5° deviate significantly from those expected for a reg-
ular octahedron. Nevertheless, each oxime hydrogen atom
NiϪBr(1)ϪNiЈЈ 71.24(6), NiϪBr(2)ϪNiЈЈ 72.32(6), NiϪO(2)ϪNiЈЈ
98.4(3); symmetry codes: (Ј) Ϫx ϩ y ϩ 1, Ϫ x ϩ 1, z; (ЈЈ) Ϫ y ϩ
1, x Ϫ y, z
˚
is pointing, at about 2.5 A, toward the halide ion and the
The X-ray analysis of the complex with ligand 1h reveals
values for the bond angles could give support to an that it is a trinuclear species, corresponding to the formu-
OH···Br interaction. lation [Ni3Br4(OH)(1h)3][Br] (3h).
2686
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2683Ϫ2692

Nickel(II) and Cobalt(III) Complexes with 2-(Arylimino)-3-(hydroxyimino)butane
FULL PAPER
The unique and unexpected structure of complex 3h is binds a close copper unit through the O donor, becoming
shown in Figure 5, together with selected bond lengths and almost coplanar with the metal triangle.
angles. Its C3 symmetry is illustrated by the view along the
It is worth noting that in the current compound the elec-
symmetry axis (Figure 6). The metal ions of three monoch- tronic structural features of ligand 1h give rise to an as-
elated Ni(1h) fragments form an equilateral triangle. On sembly possessing well-defined hydrophobic and hydro-
each edge, the pair of nickel atoms, separated by 3.026(2) philic domains. The former is delimited by the mesitylene
˚
A, is µ2-bridged by a bromide ligand, slightly displaced by groups. The aromatic rings, inclined to form an angle of
˚
0.2 A from the Ni3 plane, with NiϪBr(2) bond lengths of 80.7(3)° with the Ni3 plane, enclose the µ3-bridging Br(1).
˚
2.551(2) and 2.578(2) A. Another bromide ion and a The hydrophilic area is confined at the oxime groups side,
hydroxy oxygen atom µ3-bridge all the nickel ions with co- where a bromide ion is displaced by an hydroxy oxygen µ3-
˚
ordination distances of 2.578(2) and 1.999(5) A, respec- bridge. The bromide ion, Br(3), is settled on the threefold
tively. These capping species lie on the threefold axis, above axis and accounts for the neutral charge of the whole com-
˚
˚
˚
[Br(1) ϩ1.923(2) A] and below [O(2) Ϫ0.97(1) A] the metal pound. The latter anion is located at 3.236(10) A from the
˚
plane. The coordination geometry around each nickel atom oxime oxygen atoms O(1) and at 3.322(7) A from O(2),
thus involves the nitrogen-donor atoms of a chelating li- probably interacting with the hydrogen atoms of the
gand 1h, three bromide ions, and a hydroxy group with hydroxy groups (Figure 5).
bond angles that considerably deviate from the ideal octa-
hedral values.
On the basis of these results, a reaction mechanism is
proposed for the formation of the different synthetic prod-
ucts (Scheme 3).
Scheme 3. Proposed mechanism for the synthesis of nickel com-
plexes
Figure 6. Projection down the threefold axis of the molecular struc-
ture of 3h
The reaction proceeds by initial formation of the mono-
chelated monomeric species, which, depending on the steric
hindrance of the ligand, can then evolve along different
pathways. For ligands of low steric hindrance (1a, 1b), the
reaction moves rapidly towards the formation of bis(chel-
ated) complexes, even at low ligand/Ni ratios. For the mes-
ityl derivative, with ligand/Ni ϭ 1:1, the monochelated de-
rivative is isolated as the trimer 3h, while with an excess of
ligand the bis(chelated) species 2h is the product. With the
sterically more hindered ligands 1d, 1e and 1i, only the for-
mation of monochelated species of unknown nuclearity is
observed.
˚
In the ligand, the C(2)ϪC(3) bond length of 1.522(14) A,
although of low accuracy, agrees with a single bond charac-
ter indicating that, upon coordination, the delocalization
inside the OϪNϭCϪCϭN moiety of the ligand seems to
be removed.
To the best of our knowledge, two structures containing
a strictly analogous Ni3X3 core (X ϭ halide)[17,18] and with
metal atoms capped by an oxygen atom and another halide
ion have been reported; their NiϪNi distances, averaged to
˚
3.05 A, are closely similar to those here. One of these, with
tetramethylethylenediamine as chelating ligand (L) and with
chloride replacing the bromide anions, has the same stoichi-
ometry [Ni3X4(OH)L3][X].[17a] A trimeric copper species
[Cu3(ClO4)(OH)LЈ3]ϩ containing the deprotonated ligand
1a (LЈ) has also been reported.[17b] However, the crystal
structure shows that LЈ behaves as a tridentate species: it
Synthesis and Characterization of CoIII Complexes 4a؊4d,
4f؊4h, 5a؊5d, 5f؊5h
Synthesis was carried out in the presence of air, in n-
chelates a metal ion with the N atoms and simultaneously butanol, at room temperature, for 3 h, starting from
Eur. J. Inorg. Chem. 2003, 2683Ϫ2692
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2687

E. Zangrando, M. Trani, E. Stabon, C. Carfagna, B. Milani, G. Mestroni
FULL PAPER
CoX2·nH2O salts (X ϭ Cl, Br) and using a ligand/cobalt
ratio of 1:1 or 2:1 (Scheme 2, b). In all cases, the formation
of a red solid is observed immediately after addition of the
ligands, followed by its transformation in the final product
into a green microcrystalline solid. A total of 14 complexes
(chloro and bromo derivatives) have been synthesized.
However, the synthesis of 4a and 5a derivatives was re-
ported several years ago.[14]
On the basis of elemental analyses (Table 2), all the com-
plexes were formulated as [CoX2(ArϪNʝNϪO)-
(ArϪNʝNϪOH)] [X ϭ Cl (4), Br (5)]. Therefore, in con-
trast to the nickel derivatives, only octahedral bis(chelated)
species are obtained with all the ligands tested, regardless
of the ligand/cobalt ratio used.
Single crystals suitable for X-ray analysis were obtained
for 5b and 5h by crystallization from CH2Cl2/n-pentane.
Figures 7 and 8 (ORTEP drawings, thermal ellipsoids at
50% probability) give perspective views of the molecular
structures of compounds 5b and 5h, the latter being located
on a crystallographic twofold axis passing through the me-
tal atom and in the middle between the oxime oxygen
atoms. Unlike in the nickel derivatives, the two bromide
ions are in a trans arrangement, so the chelating ligands
define the equatorial plane of the molecule. Moreover, they
are bound to the cobalt atom in a head-to-head configura-
tion, in contrast to the head-to-tail arrangement proposed
for 4a and 5a on the basis of steric considerations.[14] The
CoϪBr and CoϪN coordination bond lengths have com-
parable values in the two complexes. In contrast, the values
of the BrϪCoϪBr bond angles, probably affected by steric
Figure 8. Molecular structure of compound 5h with atom labelling
scheme of the crystallographically independent moiety; selected
˚
bond lengths [A] and angles [°]: CoϪBr 2.386(1), CoϪN(1)
1.912(6),
CoϪN(2)
2.006(6);
80.7(2),
N(1Ј)ϪCoϪN(1)
N(1)ϪCoϪN(2Ј)
95.2(3),
175.9(2),
N(1)ϪCoϪN(2)
N(2)ϪCoϪN(2Ј) 103.5(3), N(1Ј)ϪCoϪBr 85.1(2), N(1)ϪCoϪBr
85.3(2), N(2)ϪCoϪBr 94.4(2), N(2Ј)ϪCoϪBr 94.4(2), BrϪCoϪBrЈ
165.64(7); primed atoms at Ϫ x ϩ 1, Ϫy ϩ 1, z
hindrance between the bromide ions and the proximal methyl groups on the phenyl rings, are 174.1(1) and
165.64(7)° in the o-tolyl and in the mesityl derivative,
respectively. The atoms of the five-membered rings are close
to coplanar and the calculated angles of 1.6 (5b) and 5.4°
(5h) represent a slight tilting between the chelating ligand
moieties.
The head-to-head arrangement of the mixed Schiff-base/
oxime ligands favors the formation of an OH···O bridge.
The mean O···O distance for the two structures of 2.42(1)
˚
A is shorter than those found in the widely studied
[Co(DH)2] and [Co(DO)(DOH)(pn)] complexes[19] [mean
[20]
˚
values of 2.487(2) and 2.45(1) A,
respectively]. It was
suggested that the O···O distance should parallel the ease
of deprotonation, as indicated by the slower exchange of
the proton in (DO)(DOH)pn than in (DH)2 complexes of
similar axial ligands.[21,22] The formation of the H-bridge in
˚
5h causes a remarkable shortening of the NϪO [1.288(7) A]
˚
and of the N(oxime)ϭC [1.292(9) A] bonds with respect to
those detected in the free ligand, and the same trend is also
observed for 5b, although the values found here are less ac-
curate.
Moreover, the aromatic rings, with an orientation similar
to that found in the uncoordinated ligand, ensure intramol-
ecular π-stacking. In fact, the interplanar angle formed by
the facing o-tolyl rings in 5b is 15.0(3)° and the C···C dis-
tances between the ipso-carbon atoms and the carbon
Figure 7. Molecular structure of compound 5b; selected coordi-
˚
nation bond lengths [A] and angles [°]: CoϪBr(1) 2.389(2),
CoϪBr(2) 2.385(2), CoϪN(1) 1.895(11), CoϪN(2) 2.006(11),
CoϪN(3) 1.906(11), CoϪN(4) 1.997(11); N(1)ϪCoϪN(2) 80.6(3),
N(1)ϪCoϪN(3)
N(2)ϪCoϪN(3)
96.2(5),
N(1)ϪCoϪN(4)
N(2)ϪCoϪN(4)
176.3(5),
103.1(5),
˚
atoms in para positions are 3.35 and 3.84 A, respectively. It
is worth noting that the dihedral angle between the aryl
176.7(5),
N(3)ϪCoϪN(4) 80.1(3), Br(1)ϪCoϪBr(2) 174.12(13)
2688
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2683Ϫ2692

Nickel(II) and Cobalt(III) Complexes with 2-(Arylimino)-3-(hydroxyimino)butane
FULL PAPER
groups in 5h is reduced to 5.5(3)°, and that the C···C separ- CH3I, to yield a red, microcrystalline solid. The elemental
ations are correspondingly somewhat shorter (3.18 and 3.76 analysis confirms the ratio Co/1g ϭ 1:2 and is in agreement
˚
A, respectively), indicating a closer approach and hence a with the formation of the organometallic derivative
more efficient π-interaction between the bulky mesitylene [Co(CH3)(I){1g-(H)}(1g)] (6g). The 1H NMR spectrum, re-
groups.
corded in CDCl3 at room temperature, shows three singlets
Therefore, the head-to-head arrangement of the chelating of the same intensity in the region of the aromatic protons,
ligands gives rise to an H-bridge and to an aryl π-stacking and five singlets for the methyl groups, one of which is of
that act as a double clamp providing additional stability to half intensity with respect to the other four peaks. This last
the complexes. The geometrical data for the cobalt com- is therefore attributed to the methyl group bound to the
pounds reported here are quite comparable, but a detailed cobalt atom. The other assignments of signals to protons
analysis shows that the steric effects around the metal were based on comparison with the spectrum of the starting
center can be modulated by changing the nature of the complex 4g. In particular, the groups in the α- and β-posi-
aryl substituents.
tions of the aromatic rings generate two different signals,
These complexes were characterized in solution by re- thus indicating the lack of the equatorial plane of the com-
cording NMR spectra in CDCl3 at room temperature plex as a plane of symmetry. The CoϪCH3 and CoϪI
(Table 3). The signal of the methyl group (A) bound to the bonds are therefore in a trans configuration. Finally, it is
carbon atom closest to the oxime nitrogen atom is the most reasonable to assume that the two NʝN ligands are bound
shifted (almost 0.40 ppm downfield shift) with respect to to the cobalt atom in the head-to-head arrangement, as in
the signals of the free ligand. No signal due to the hydroxy the starting complex. This behavior is analogous to that
group is visible in the spectra of the complexes. It can be typical of the dimethylglyoxime and (DO)(DOH)pn deriva-
seen that the spectrum of complexes with the o-tolyl-substi- tives.[23]
tuted ligand presents two sets of signals of different inten-
Polymerization of Methyl Methacrylate
sity both for the alkyl and aryl protons, thus indicating the
presence of two species in solution. At room temperature
the signals are quite broad. An NOE was observed between
the methyl group of o-tolyl and HαЈ and Hβ. On the other
hand, when the signals of the methyl groups (A and B) of
the skeleton were irradiated in different NOE experiments,
no NOE was seen, other than that expected between the
two signals themselves. These results suggest that the two
species in solution correspond to two rotamers, differing in
the relative orientations of the methyl group of the o-tolyl
substituents: namely syn and anti isomers. The broadness
of the signals suggests that the two rotamers are in equilib-
rium and the rate of this equilibrium is slow on the NMR
timescale.
Some preliminary test polymerizations of methyl meth-
acrylate were performed with complex 5b by carrying out
the reaction in the presence of AIBN as initiator, in THF
as solvent. The result is compared with the activity of the
system based on AIBN itself (Table 4).
Table 4. Polymerization of methyl methacrylate
Promoter[a]
Yield [g]
Mw
Mw/Mn
AIBN
2.34
5.59
420 000
83 300
1.7
1.4
AIBN ϩ 5b[b]
[a] Reaction conditions: nAIBN ϭ 2.9·10Ϫ2 mmol; solvent THF V ϭ
11 mL; MMA V ϭ 12 mL; T ϭ 60 °C; t ϭ 5 h. [AIBN]/[Co] ϭ
The CoIII complex with ligand 1g, [Co(Cl)2{1g-(H)}(1g)],
was treated with NaBH4 in methanol in the presence of 15.
[b]
1
Table 3. H NMR spectroscopic data for cobalt complexes 4aϪ4d, 4fϪ4h, 5aϪ5d, 5fϪ5h
Complexes[a] CH3 (A) CH3 (B) Hα
HαЈ
Hβ
HβЈ
Hγ
4a
5a
4b
5b
4c
5c
4d
5d
4f
2.67 (s)
2.66 (s)
2.67 (s)
2.68 (s)
2.65 (s)
2.65 (s)
2.67 (s)
2.69 (s)
2.70 (s)
2.72 (s)
2.65 (s)
2.64 (s)
2.69 (s)
2.71 (s)
2.05 (s)
2.07 (s)
2.04 (s)
2.06 (s)
2.06 (s)
2.08 (s)
2.10 (s)
2.12 (s)
2.06 (s)
2.08 (s)
2.05 (s)
2.07 (s)
2.07 (s)
2.09 (s)
6.77 (d)
6.85Ϫ6.92 (m)
1.96 (s)
2.04 (s)
6.64Ϫ6.71 (m)
6.70Ϫ6.74 (m)
6.77 (d)
6.91 (t)
6.91 (t)
7.10 (t)
6.85Ϫ6.92 (m) 6.85Ϫ6.92 (m) 6.85Ϫ6.92 (m) 7.08 (t)
6.64 (d)
6.65 (d)
6.64Ϫ6.71 (m) 6.64Ϫ6.71 (m) 6.64Ϫ6.71 (m) 2.30 (s)
6.70Ϫ6.74 (m) 6.70Ϫ6.74 (m) 6.70Ϫ6.74 (m) 2.30 (s)
6.90Ϫ7.00 (m) 7.28 (dd)
6.85Ϫ7.00 (m) 7.48 (d)
6.90Ϫ7.00 (m)
6.85Ϫ7.00 (m)
1.10 (d), 1.28 (d) 2.90 (m) 6.75 (dd)
1.12 (d), 1.32 (d) 2.97 (m) 6.97 (dd)
6.42 (t)
6.41 (t)
6.64 (d)
6.64 (d)
2.16 (s)
2.16 (s)
6.47 (s)
6.48 (s)
7.13 (dd)
7.13 (d)
6.64 (d)
6.64 (d)
2.16 (s)
2.16 (s)
6.47 (s)
6.48 (s)
7.04 (t)
7.02 (t)
6.86 (t)
6.86 (t)
6.71 (s)
6.70 (s)
2.21 (s)
2.23 (s)
2.03 (s)
2.10 (s)
6.44 (s)
6.53 (s)
1.99 (s)
2.06 (s)
2.03 (s)
2.10 (s)
6.44 (s)
6.53 (s)
1.99 (s)
2.06 (s)
5f
4g
5g
4h
5h
[a]
Spectra recorded in CDCl3, at room temperature; δ [ppm]; s ϭ singlet, d ϭ doublet, dd ϭ double doublet, t ϭ triplet, m ϭ multiplet;
n.d. ϭ not determined; the signals of the protons of the alkyl substituents on the aryl group are reported in the place of the correspond-
ing protons.
Eur. J. Inorg. Chem. 2003, 2683Ϫ2692
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2689

E. Zangrando, M. Trani, E. Stabon, C. Carfagna, B. Milani, G. Mestroni
FULL PAPER
[(3,5-dimethylphenyl)imino]-3-(hydroxyimino)butane (1g), (E,E)-2-
(hydroxyimino)-3-[(2,4,6-trimethylphenyl)imino]butane (1h), (E,E)-
The addition of the complex, even if in low amount with
respect to AIBN, resulted in a doubling of the yield of poly-
merization and, at the same time, in a fivefold decrease in
the molecular weight together with a decrease in polydis-
persity. These complexes therefore show a behavior typical
of chain-transfer catalysts (CCTCs).[7]
2-[(2,6-diisopropylphenyl)imino]-3-(hydroxyimino)butane
(1i)].
These ligands were synthesized by heating a solution of biacetyl
monoxime (0.02 mol) and the stoichiometric amount of amine at
reflux in methanol (5 mL) in the presence of formic acid (3 drops)
as catalyst for 4Ϫ5 h. Ligands 1a, 1b, 1c, and 1d were obtained by
cooling the solutions at 4 °C for 2Ϫ3 h after reflux. White crystals
were filtered off, washed with ether, and dried in vacuo. Ligand 1d
was recrystallized from n-pentane. Ligands 1e, 1f, 1g, 1h, and 1i
were obtained by cooling the solutions at room temp after reflux.
Ligand 1i precipitated from the heated solution after 1.5 h reaction
time. Crystalline precipitates were filtered, washed with ether, and
dried in vacuo. All ligands are colorless, with the exception of 1e,
which precipitated as a yellow solid. The values of elemental analy-
ses are reported in Table 5.
Conclusion
The coordination chemistry of the new N-ligands
towards NiII and CoIII has been studied, different behavior
being observed depending both on the ligand and on the
metal. In the case of NiII, when unhindered ligands were
used, octahedral bis(chelated) species were obtained, while
the product with more hindered ligands was a monochel-
ated trimer. In this last compound the three molecules of
chelating ligands are ‘‘iso-oriented’’, producing one hydro-
philic and one hydrophobic hole. In the case of CoIII, octa-
hedral bis(chelated) species are obtained with all the ligands
used. Their main structural feature is the presence of weak
interactions between the two coplanar N-ligands, namely
OϪH···O hydrogen bond and stacking interactions between
the aryl groups.
Table 5. Elemental analyses for ligands 1aϪ1i (calculated values
in parentheses)
Ligands
C
H
N
1a
1b
1c
1d
1e
1f
1g
1h
1i
67.30 (68.15)
68.70 (69.45)
67.90 (69.45)
71.30 (71.52)
71.40 (72.38)
70.60 (70.56)
69.60 (70.56)
71.50 (71.52)
74.60 (74.38)
6.78 (6.86)
7.40 (7.42)
7.43 (7.42)
8.37 (8.31)
8.55 (8.67)
7.92 (7.89)
7.85 (7.89)
8.28 (8.31)
9.45 (8.58)
15.70 (15.89)
14.80 (14.72)
14.90 (14.72)
12.90 (12.83)
12.00 (12.05)
13.80 (13.71)
13.70 (13.71)
12.83 (12.83)
10.90 (10.84)
These bis(chelated) CoIII complexes have several simi-
larities to the well-known cobaloximes[7] and to (DO)-
(DOH)pn derivatives,[24] previously studied by us and
further investigated by Finke.[25,26] As reported in the in-
troduction, these complexes are efficient chain transfer
catalysts, and give oligomers in high yield for methyl meth-
acrylate and styrene. The new complexes reported here
could behave very similarly but, thanks to the variable steric
hindrance, could give polymers instead of oligomers. The
molecular mass of the polymers should increase on increas-
ing the steric bulk of the N-ligands. The study of their cata-
lytic behavior and the characterization of polymers is cur-
rently under investigation.
Synthesis of Bis(chelated) NiII Complexes 2a, 2b, and 2h: Bis(chel-
ated) NiII complexes were synthesized by addition of [NiBr2-
(C4H10O2)] (9.7·10Ϫ4 mol) to a stirred solution of ligand (1.94·10Ϫ3
mol) in CH2Cl2 (7 mL). Solutions were stirred for 3 h. Complexes
2a, 2b, and 2h were isolated upon concentration of the solution and
addition of diethyl ether or n-pentane. The products precipitated as
brown solids, which were filtered and dried under vacuum.
Synthesis of Monochelated NiII Complexes 2d, 2e, 2i, and 3h: Mono-
chelated NiII complexes were prepared by addition of [NiBr2-
(C4H10O2)] (C4H10O2 ϭ ethylene glycol dimethyl ether) (9.7·10Ϫ4
mol) to a stirred solution of ligand (1.56·10Ϫ3 mol) in CH2Cl2
(7 mL). Solutions were stirred for 3 h. For 3h, 2d, and 2e, insoluble
impurities were filtered off. Solutions were concentrated under vac-
uum to 1Ϫ2 mL, and then treated with ether. Light brown powders
were filtered, washed with ether or n-pentane, and dried in vacuo.
Insoluble 2i precipitated as a light brown powder, which was fil-
tered, washed with ether, and dried in vacuo.
Synthesis of Bis(chelated) CoIII Complexes 4aϪ4d, 4fϪ4h, 5aϪ5d,
and 5fϪ5h: Syntheses of bis(chelated) CoIII complexes were per-
formed by addition of the ligand (2.9·10Ϫ3 mol) to a stirred solu-
tion of the CoII salt hexahydrate (CoCl2·6H2O, CoBr2·6H2O)
(1.4·10Ϫ3 mol) in n-butanol (10 mL). Solutions were stirred for 3 h.
Solids were filtered off, washed with n-butanol, and dried in vacuo.
Crystalline precipitates were obtained by recrystallization from
CH2Cl2/pentane or CH2Cl2/ether (4a and 5a only). Complexes 5f,
5h, and 5d are brown, while all other complexes are green.
Experimental Section
Materials and Instrumentation: Chemicals and solvents were ob-
tained from commercial sources (Aldrich, CarloϪErba) and were
used as received. NMR spectra were recorded with a Jeol EX 400
spectrometer operating at 400.0 MHz for 1H. 1H chemical shifts
were measured relative to the residual solvent peak versus TMS
(CDCl3 at δ ϭ 7.26 ppm and CD2Cl2 at δ ϭ 5.33 ppm for 1H).
Two-dimensional correlation spectra (COSY) were obtained with
the automatic program of the instrument. NOE experiments were
1
run with a H pulse of 90° and 12.5 µs. THF was dried with met-
allic sodium. MMA used in the polymerization reactions was puri-
fied by extraction with a water solution of NaOH and distilled.
Synthesis of Ligands: (E,E)-2-(Hydroxyimino)-3-(phenylim-
ino)butane (1a), (E,E)-2-(hydroxyimino)-3-(o-tolylimino)butane Synthesis of Bis(chelated) Organometallic CoIII Complex 6g: Com-
(1b), (E,E)-2-(hydroxyimino)-3-(p-tolylimino)butane (1c), (E,E)-2-
(hydroxyimino)-3-[(2-isopropylphenyl)imino]butane (1d), (E,E)-2-
[(2-tert-butylphenyl)imino]-3-(hydroxyimino)butane (1e), (E,E)-2-
[(2,6-dimethylphenyl)imino]-3-(hydroxyimino)butane (1f), (E,E)-2-
plex 6g was prepared by addition of CH3I (0.5 mL) and an excess
of NaBH4, in that order, to a stirred solution of the chloride deriva-
tive 4g (0.16 g) in methanol (10 mL) under an inert gas. The solu-
tion was stirred at room temperature for 1 h, and the red crystalline
2690
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2683Ϫ2692

Nickel(II) and Cobalt(III) Complexes with 2-(Arylimino)-3-(hydroxyimino)butane
FULL PAPER
Table 6. Crystallographic data for compounds 1h, 2a, 3h, 5b, and 5h
1h
2a·0.5CH2Cl2
3h·H2O
5b
5h
Empirical formula
Formula mass
Crystal system
Space group
C13H18N2O
218.29
C20.5H25Br2ClN4NiO2
613.43
C39H56Br5N6Ni3O5
1264.58
C22H27Br2CoN4O2
598.23
orthorhombic
Pna21 (no. 33)
15.176(7)
8.273(3)
18.738(7)
C26H35Br2CoN4O2
654.33
tetragonal
I41cd (no. 110)
13.623(2)
monoclinic
P21/c (no. 14)
7.379(5)
15.635(6)
11.167(5)
93.74(5)
1285.6(12)
4
1.128
472
0.072
2.24Ϫ25.02
293(2)
monoclinic
C2/c (no. 15)
21.737(5)
15.002(4)
16.382(4)
110.46(3)
5005(2)
trigonal
¯
R3 (no. 148)
˚
a [A]
13.945(4)
˚
b [A]
˚
c [A]
50.665(16)
29.901(8)
β [°]
3
˚
V [A ]
8532.5(44)
6
1.477
3786
4.534
2.33Ϫ26.98
293(2)
17970
2352.6(16)
4
1.689
1200
4.153
2.68Ϫ27.95
293(2)
3992
5549.2(19)
8
1.566
2656
3.528
2.52Ϫ29.95
293(2)
4153
Z
8
ρcalcd. [g cmϪ3
F(000)
]
1.628
2456
4.099
1.69Ϫ24.41
130(2)
µ (Mo-Kα) [mmϪ1
θ range [°]
T [K]
]
Reflns. collected
2394
7621
Reflns. unique/R(int)
Reflns. obsd. [I Ͼ 2σ(I)]
Variables
2271/0.0730
1592
151
0.0692
0.0849
0.901
0.168, Ϫ0.181
3965/0.0509
2781
296
0.0596
0.1276
1.066
0.675, Ϫ0.770
4005/0.1358
1495
182
0.0646
0.1541
1.160
2898/0.0580
1408
179
0.0586
0.0997
1.083
1932/0.0690
1246
164
0.0423
0.0712
1.044
R1[a]
wR2[a]
GOF (F2)[b]
Residuals [e/A ]
3
˚
0.981, Ϫ0.734
0.688, Ϫ0.777
0.470, Ϫ0.354
[a]
2
2 2
2 2
[b]
2
2 2
R1 ϭ Σ||Fo| Ϫ |Fc||/Σ|Fo|; wR2 ϭ {Σw(Fo Ϫ Fc ) /Σw(Fo ) }1/2 for observed reflections. GOF ϭ {Σw(Fo Ϫ Fc ) /Σw (n Ϫ p)}1/2
.
solid was then filtered off and dried in vacuo. C25H34CoIN4O2
Polymerization Reactions: The polymerization reactions were car-
(608.40): calcd. C 49.35, H 5.63, N 9.21; found C, 50.00, H 5.73, ried out by use of Schlenk techniques, under an inert gas. The
1
N 9.26. H NMR (CDCl3) δ ϭ 1.68 (s, 3 H), 1.73 (s, 6 H), 2.12 (s,
AIBN and the cobalt complex were added to the dry THF solution
6 H), 2.15 (s, 6 H), 2.30 (s, 6 H), 5.96 (s, 2 H), 6.60 (s, 2 H), 6.89 containing MMA at 60 °C for 5 h. The amounts are reported in
(s, 2 H) ppm.
Table 4. The polymer was obtained as a white powder after the
addition of ethanol. The powder was filtered and vacuum-dried.
X-ray Crystallographic Study: Crystal data, data collections, and
refinement parameters for the structures reported are summarized
in Table 6. Data sets of 1h, 2a, 3h, 5b, and 5h were collected at
room temperature by the ω/2θ scan technique with a CAD4
Molecular Mass Measurements: The analyses were recorded with a
Knauer HPLC (pump K-501, UV detector K-2501) with a PLgel
5 µm 104 A column and THF as solvent (flow rate 0.6 mL/min).
˚
EnrafϪNonius diffractometer, equipped with graphite mono- Polystyrene standards were used. The polyMMA samples were dis-
˚
chromator using Mo-Kα radiation (λ ϭ 0.71073 A). Intensities of
three standard reflections, monitored during data collections,
showed no noticeable variation in intensity for any of the crystals.
Intensities were corrected for Lorentz polarization effects and data
sets of 3h, 5b, and 5h also for absorption, based on an empirical
ψ-scan method. Diffraction data of 2a were obtained with a Nonius
DIP-1030H system with Mo-Kα radiation (graphite-monochrom-
ated). A total of 30 frames were collected, each with an exposure
time of 20 min, over a half of reciprocal space with a rotation of
6° about the detector being located at 100 mm from the crystal.
Cell refinement and data reduction were carried out with the aid
of the programs Mosflm[27] and Scala.[27] The structures were
solved by Patterson or direct methods and successive Fourier analy-
ses (SHELXS-97)[28] and refined by the full-matrix least-squares
method based on F2, with the aid of SHELXL-97.[28] A difference
Fourier map allowed a disordered molecule of CH2Cl2 to be de-
tected (half occupancy) in 2a, while a residual in 3h was assigned
to a water oxygen atom. All the calculations were performed with
the aid of the WinGX System, version 1.64.03.[29] CCDC-199885,
-199886, -199887, -199888, and -199889 contain the supplementary
crystallographic 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, Cambridge CB2 1EZ, UK; Fax: (internat.) ϩ 44-1223/336-
033; E-mail: deposit@ccdc.cam.ac.uk].
solved as follows: 3 mg of each sample in 120 µL of 1,1,1,3,3,3-
hexafluoro-2-propanol, and THF was then added up to 10 mL.
The statistical calculations were performed by use of the Bruker
Chromstar software program.
Acknowledgments
This work was supported by the Ministero dell’Universita e della
`
Ricerca Scientifica (MURST-Rome) Grant no. 9903153427.
[1]
D. J. Tempel, M. Brookhart, Organometallics 1998, 17,
2290Ϫ2296.
[2] [2a]
L. Johnson, M. Brookhart, C. M. Killian, S. D. Arthur, J.
Feldman, E. F. McCord, S. J. McLain, K. A. Kreutzer, M. A.
Bennett, E. B. Coughlin, S. D. Ittel, A. Parthasarathy, D. J.
Tempel, Wo 96/23010, 1996.
[2b]
S. D. Ittel, L. K. Johnson, M.
Brookhart, Chem. Rev. 2000, 100, 1169Ϫ1203.
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. William,
Chem. Commun. 1998, 849Ϫ850.
T. R. Younkin, E. F. Connor, J. I. Henderson, S. K. Friedrich,
R. H. Grubbs, D. A. Bansleben, Science 2000, 287, 460Ϫ462.
B. L. Small, M. Brookhart, Macromolecules 1999, 32,
2120Ϫ2130.
[3]
[4]
[5]
[6]
[7]
B. B. Wayland, G. Poszmik, S. L. Mukerjee, J. Am. Chem. Soc.
1994, 116, 7943Ϫ7944.
A. A. Gridnev, S. D. Ittel, Chem. Rev. 2001, 101, 3611Ϫ3659.
Eur. J. Inorg. Chem. 2003, 2683Ϫ2692
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2691

E. Zangrando, M. Trani, E. Stabon, C. Carfagna, B. Milani, G. Mestroni
FULL PAPER
[8]
[18]
N. S. Enikopolopyan, B. R. Smirnov, G. V. Ponomarev, I. M.
Belgovskii, J. Poly. Sci. Polym. Chem. Ed. 1981, 19, 879Ϫ889.
A. F. Burczyk, K. F. O’Driscoll, G. L. Rempel, J. Polym. Sci.
1984, 22, 3255Ϫ3262.
D. Kukulj, J. P. A. Heuts, T. P. Davis, Macromolecules 1998,
31, 6034Ϫ6041.
E. Bouwman, P. Evans, H. Kooijman, J. Reedijk, A. L. Spek,
J. Chem. Soc., Chem. Commun. 1993, 1746Ϫ1748.
[9]
[19]
(DH)2 ϭ bis(dimethylglyoximato); tetradentate ligand (DO)-
(DOH)pn
oxime).
ϭ
N2,N2Ј-propanediylbis(2-imino-3-butanone
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[20]
[21]
L. Randaccio, N. Bresciani Pahor, E. Zangrando, L. G. Marz-
illi, Chem. Soc., Rev. 1989, 18, 225Ϫ250.
N. Bresciani Pahor, M. Forcolin, L. G. Marzilli, L. Randaccio,
M. F. Summers, P. J. Toscano, Coord. Chem. Rew 1985, 63,
1Ϫ125.
R. J. Gulshl, T. L. Brown, Inorg. Chem. 1974, 13, 959Ϫ965.
G. Costa, G. Mestroni, Tetrahedron Lett. 1967, 41, 4005Ϫ4007.
B. E. Daikh, R. G. Finke, J. Am. Chem. Soc. 1991, 113,
4160Ϫ4172.
J. Halpern, F. T. T. Ng, G. L. Rempel, J. Am. Chem. Soc. 1979,
101, 7124Ϫ7126.
F. T. T. Ng, G. L. Rempel, J. Halpern, J. Am. Chem. Soc. 1982,
104, 621Ϫ623.
B. Milani, E. Stabon, E. Zangrando, G. Mestroni, Inorg. Chim.
Acta, in press.
[22]
[23]
[24]
E. Uhlig, E. Dinjus, M. Helbig, Z. Anorg. Allg. Chem. 1970,
373, 279Ϫ284.
[25]
[26]
[27]
[28]
[29]
V. Bertolasi, G. Gilli, A. C. Veronese, Acta Crystallogr., Sect.
B 1982, 38, 502Ϫ511.
B. E. Daikh, R. G. Finke, J. Am. Chem. Soc. 1992, 114,
2938Ϫ2943.
F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G.
Orpen, R. Taylor, ‘‘Typical interatomic distances: organic com-
pounds’’, in International Tables for Crystallography, vol. C
(Ed.: A. J. C. Wilson), Kluwer Academic Publishers, Dord-
recht, The Netherlands, 1995.
B. E. Daikh, T. J. R. Weakely, R. G. Finke, Inorg. Chem. 1992,
31, 137Ϫ139.
Collaborative Computational Project, Number 4, Acta Crys-
tallogr., Sect. D 1994, 50, 760Ϫ763.
G. M. Sheldrick, SHELX97, Programs for Crystal Structure
Analysis, release 97-2, University of Göttingen, Germany, 1998.
L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837Ϫ838.
Received January 7, 2003
[17] [17a]
U. Turpeinen, A. Pajunen, Finn. Chem. Lett. 1976, 6Ϫ11.
[17b] R. J. Butchers, C. J. OЈConnor, E. Sinn, Inorg. Chem. 1981,
20, 537Ϫ545.
2692
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2683Ϫ2692