Highly flexible O,O′,N ligands and their Fe, Ni, Cu and Zn complexes

Article abstract of DOI:10.1016/j.ica.2011.02.072

Three new chiral ligands bearing an O,O′,N donor set (O _methoxyO_hydroxyN_pyridine) were synthesised and coordinated to Fe^III, Fe^II, Ni^II, Cu ^II and Zn^II to yield complexes with the general formula [M(OON)Cl_x]_y. While the pyridine N and the hydroxy O atoms coordinate strongly to all applied metal ions, the methoxy donor seems not to be involved in coordination, although some evidence for a weak interaction between O_Me and the Zn^II were found in NMR spectra. In the bidentate O′,N coordination mode the new ligands exhibit several coordination geometries as analysed in the solid compounds by XRD, EXAFS and EPR and in solution by UV-Vis absorption, cyclic voltammetry, EXAFS, EPR or NMR spectroscopy.

Full text of DOI:10.1016/j.ica.2011.02.072

Inorganica Chimica Acta 374 (2011) 350–358
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Highly flexible O,O0,N ligands and their Fe, Ni, Cu and Zn complexes
b
Katharina Butsch a, Axel Klein a, , Matthias Bauer
⇑
a Universität zu Köln, Department für Chemie, Institut für Anorganische Chemie, Greinstrasse 6, D-50939 Köln, Germany
b Karlsruhe Institut of Technology (KIT), Institut für Technische Chemie und Polymerchemie, Engesserstr. 20, D-76131 Karlsruhe, Germany
a r t i c l e i n f o
a b s t r a c t
Three new chiral ligands bearing an O,O0,N donor set (OmethoxyOhydroxyNpyridine) were synthesised and
coordinated to FeIII, FeII, NiII, CuII and ZnII to yield complexes with the general formula [M(OON)Clx]y.
While the pyridine N and the hydroxy O atoms coordinate strongly to all applied metal ions, the methoxy
donor seems not to be involved in coordination, although some evidence for a weak interaction between
OMe and the ZnII were found in NMR spectra. In the bidentate O0,N coordination mode the new ligands
exhibit several coordination geometries as analysed in the solid compounds by XRD, EXAFS and EPR
and in solution by UV–Vis absorption, cyclic voltammetry, EXAFS, EPR or NMR spectroscopy.
Ó 2011 Elsevier B.V. All rights reserved.
Article history:
Available online 3 March 2011
Dedicated to Wolfgang Kaim on the
occasion of his 60th birthday.
Keywords:
Oxido-pincer ligands (O,O,N ligands)
Iron(III)
Nickel(II)
Copper(II)
Zinc(II)
EPR spectroscopy
1. Introduction
exclusively bind to the pyridine nitrogen [10,11]. As a result,
oxido-pincer ligands show a wealth of structural motifs including
Ligands bearing a tridentate O,E,O donor set where E = C, N, P
etc. are well known. The most common, called oxido-pincer li-
gands, contain an aromatic central core and two pendant arms car-
rying various oxygen donor functions. Most important are ligands
with a central pyridine ring, which have found a number of appli-
cations, e.g. in catalysis [1]. Typical oxido donors functions can be
carboxylates [2], carbonyl [3] or alcohol functions [4]. The latter
are most interesting, since they provide the opportunity of step-
wise deprotonation upon coordination which leads to a large
diversity of complexes and structures. An increasing tendency to
deprotonation is expected with increasing hardness (HSAB) of
the metal ion. Consequently, complexes of early transition metals
(groups 4–6) exclusively contain the fully deprotonated ligand
dianions [5]. Intermediate cases (partly deprotonated) were found
in polynuclear manganese complexes [6], while late transition
metals (groups 9–12) usually coordinate to the fully protonated
ligands [4a,7]. However, there are also exceptions from this
square planar (Pt, Pd), trigonal-bipyramidal (Co, Ni, Zn), square
pyramidal (Cu) and octahedral (Co, Ni, Cu, Zn) coordination
[10,12,13].
Most of the studies reported so far focus on symmetric oxido-
pincer ligands with an O,N,O donor set. The modification of the do-
nor atom sequence from O,N,O to O,O,N or O,O0,N (Scheme 1) gives
some interesting possibilities because of the intrinsic asymmetry
and the option to build up hemilabile systems resulting from the
trans-effect of the (presumably) strong N–M interaction which
weakens the peripheral M–O bond.
In this contribution three new chiral O,O0,N pincer ligands were
synthesised (Scheme 2). The new ligands were designed focussing
on four main aspects. (i) Compared to the frequently used oxido-
pincer ligands with a central pyridine core and two pendant
carboxylate or methanol functions, the new ligands provide not
only a change in donor atom sequence O,N,O ? O,O0,N but also
an altered sequence of (expected) binding strength Oweak,Nstrong
O
,
correlation, e.g. as found in [Cu2(g2
[7d,8] and in [Cu(pydimH2)(pydimH)]+ [7d]
,l
-pydimH)2(
g
3-pydimH2)2]2+
weak ? Oweak,O0weak,Nstrong towards soft metal ions as indicated
in Scheme 1. (ii) The ligands provide the possibility of tris-chelate
binding, upon forming two five-membered chelate rings. In con-
trast to ligands containing rigid aromatic cores such as pyridine
In agreement with the HSAB concept [9], weak bonds are ob-
served between the O-donor functions and late transition metal
ions like CoII, NiII, CuII and ZnII, while the metal ions PdII and PtII
which allow only
position is far more flexible and might open the option of a
g
3-meridional binding, here the central binding
g
3-fa-
cial binding. The steric bulk at the central C atom increases within
the series OON1 < OON2 < OON3, while the acidity of the hydroxy
function is expected to decrease along the series OON3 >
OON1 > OON2. (iii) Only one of the two oxido donor functions
⇑
Corresponding author.
E-mail address: axel.klein@uni-koeln.de (A. Klein).
0020-1693/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2011.02.072

K. Butsch et al. / Inorganica Chimica Acta 374 (2011) 350–358
351
2.2. Preparation of the complexes
O
Msoft Ohard Nsoft Msoft Ohard
hard
Nsoft
Ohard
The complexes were obtained by stirring the metal chloride
salts with the ligands in a 1:1 ratio at 298 K in methanol for
16 h, without the use of external base. Strongly coloured products
of the general formulae [MII(O,O0,N)Cl2] (M = Ni, Cu, Zn) or
[FeIII(O,O0,N)Cl3] (except OON2) were obtained in high yields.
While the Ni, Cu and Zn compounds (despite [Cu(OON1)Cl2]n)
show good solubility in polar organic solvents such as acetone,
methanol, ethanol, THF, DMF or MeCN, the Fe compounds are less
soluble. Several attempts to obtain the FeIII complex [Fe(OON2)Cl3]
failed – in all cases virtually insoluble orange material was ob-
tained which gave elemental analyses far from a reasonable stoi-
chiometry (see Section 4). The iron complexes [Fe(OON1)Cl3] and
[Fe(OON3)Cl3] require strongly polar solvents such as MeCN or
DMF to be dissolved.
Scheme 1. Schematic binding situation of O,N,O oxido-pincer ligands compared to
O,O0,N ligands to soft transition metal ions.
H3C
O
H
N
N
N
C
C
C
*
*
*
OH
OH
OH
O
O
H3C
H3C
OON3
OON1
H3C
OON2
Scheme 2. The new chiral O,O0,N pincer-ligands; left: 2-(hydroxy(pyridin-2-yl)-
methyl)methoxyphenol (OON1), middle: 2-(1-hydroxy-1-(pyridin-2-yl)ethyl)phe-
nol (OON2), right: 2-(hydroxy(phenyl)(pyridin-2-yl)methyl)phenol (OON3).
2.3. Crystal and molecular structures from XRD
From the compounds [Fe(OON1)Cl2]2, [Cu(OON1)Cl2]n and
[Cu(OON3)Cl2]2 single crystals were obtained which could be ana-
lysed by X-ray diffraction. Details of the structures are shown in
Figs. 1–3, essential crystal and structural data is collected in the
Supplementary material.
When recrystallising the complex [Fe(OON1)Cl3] from acetone
solution, single crystals of the dimeric complex [Fe(OON1)Cl2]2
were obtained (see Fig. 1). The compound has obviously been
formed through elimination of HCl (Eq. (1))
can be deprotonated upon coordination, since the second is a
methoxy group. The methoxy and hydroxyl functions are expected
to exert approximately the same binding strength but deprotona-
tion of the central hydroxyl group will greatly enhance its coordi-
nation capacity over that of the methoxy group (creating a
O
weak,O0strong,Nstrong binding situation). (iv) The new ligands are
chiral due to four different substituents at the central C atom.
The chiral centre is very close to the metal atom, which might
be important for the transfer of chiral information to a substrate
bound at the central metal ion.
2½FeðOON1HÞCl3ꢀ¡2HCl þ ½FeðOON1ÞCl2ꢀ2
ð1Þ
The crystal structure was solved and refined in the triclinic
ꢀ
space group P1 (Tables 1, and Supplementary material). In the
The three ligands were reacted with precursors containing FeIII,
NiII, CuII and ZnII. These metal ions are expected to increase in
‘‘hardness’’ along the series CuII < NiII < FeIII and also provide differ-
ent preferences for coordination polyhedra. The structures of the
new complexes in the solid were studied by a combination of
XRD on single crystals, and EXAFS and EPR spectroscopy on pow-
ders of selected compounds (focussing on CuII and FeIII). The struc-
tures in solution were investigated using UV–Vis absorption
spectroscopy, and in the case of the diamagnetic ZnII compounds
additionally by 1H NMR. The paramagnetic CuII and FeIII complexes
were investigated by EPR spectroscopy, and all complexes have
been submitted to electrochemical experiments (mainly cyclic
voltammetry).
crystal structure the two iron atoms are bridged by the central
deprotonated oxido function of the OON1 ligand, forming a polyhe-
dron of two edge-sharing (distorted) trigonal bipyramids with a
nearly planar Fe2O2 core (with a dihedral angle Fe1–O1–Fe2–O2
of 9.8(2)° and small O–Fe–O angles of 73.31(2)° and 73.12(2)°).
The trigonal planes around FeIII are formed by the two chlorido li-
gands and one of the bridging O atoms. The axial positions are
occupied by the pyridine N atoms and the second bridging O atom,
2. Results and discussion
2.1. Preparation and structure of the ligands
The ligands were synthesised by Grignard coupling reaction of
2-iodo-anisole with 2-formylpyridine (OON1), 2-acetylpyridine
(OON2), or 2-benzoylpyridine (OON3), respectively. The resulting
ligands were obtained as racemates and characterised by NMR
and elemental analysis. All ligands were used as racemates for
complex synthesis and characterisation. The ligands exhibit a
good solubility in organic solvents and were purified by recrystal-
lisation from acetone, upon which single crystals of OON2 and
OON3 were obtained and analysed by XRD (details in the Supple-
mentary material). In the molecular structures of the free ligands
the O,O0,N binding pocket is not available, however, the coordi-
nating sites (O,O0,N) are arranged around the central chiral C
atom in a rather flexible way and the conformational changes
necessary to allow tridentate facial coordination are probably
very small.
Fig. 1. ORTEP representation (with 50% probability ellipsoids) of the molecular
structure of [Fe(OON1)Cl2]2, hydrogen atoms are omitted for clarity.

352
K. Butsch et al. / Inorganica Chimica Acta 374 (2011) 350–358
Fig. 2. View on the polymeric chain structure of [Cu(OON1)Cl2]n (30% probability level) (left); ORTEP representation of the asymmetric unit in the chain structure with the
disordered anisole ring (right); H atoms are omitted for clarity in both figures.
Table 1
Selected distances and angles of [FeCl2(OON1)]2 and [CuCl2(OON1)]n.
Distances
(Å)
Distances
(Å)
Angles (°)
[Fe(OON1)2Cl2]
Fe1–O1
Fe2–O1
1.951(4) Fe1–Cl2
1.981(3) Fe2–Cl4
2.173(2) Fe2–N2
2.212(2) O1–Fe1–O3 73.3(1)
2.176(2) O1–Fe2–O3 73.2(1)
Fe1–Cl1
2.088(4) Cl1–Fe1–
111.45(9)
Cl2
2.350(1) Cl3–Fe2–
Cl4
Fe2–Cl3
2.223(2) O1ꢁꢁꢁO3
112.79(9)
Fe1–N1
Fe1–O3
Fe2–O3
2.109(5) Fe1ꢁꢁꢁFe2
1.986(3) Fe1ꢁꢁꢁOMe
1.965(4) Fe2ꢁꢁꢁOMe
3.147(1) N1–Fe1–O3 147.0(2)
4.994(2) N2–Fe2–O1 150.1(2)
4.931(2) N1–Fe1–
99.2(1)
Cl1
N2–Fe2–
Cl3
95.2(1)
Fig. 3. ORTEP representation (with 50% probability ellipsoids) of the molecular
structure of [Cu(OON3)Cl2]2, hydrogen atoms are omitted for clarity.
Distances (Å)
[Cu(OON1)Cl2]n
Cu1ꢁꢁꢁOMe
Angles (°)
the angles N1–Fe1–O3 (147.0(2)°) and N2–Fe2–O1 (150.1(2)°)
deviate markedly from the ideal angle of 180°. The methoxy func-
tion remains uncoordinated. Interestingly, the dimer is formed by
two mononuclear units which contain ligands of the same chiral-
ity; one of the molecules in the unit cell possesses ligands of R con-
figuration, this is depicted in Fig. 1, the other molecule possesses
ligands of S chirality, thus in the crystal structure both types of
stereoisomers occur.
4.089(4)
space group P21/c (Tables 2 and Supplementary material). The
structure shows a binuclear -Cl2-bridged complex entity (see
Fig. 3). The crystal structure reveals a short intramolecular H
bridge between O2 and H1 with a distance of 1.94(5) Å. Further-
more, intermolecular
is observed for the anisole rings of neighbouring molecules. The
two rings are coplanar (0.00(3)°) and exhibit a rather short inter-
planar distance (3.53(2) Å). In the binuclear [Cu(OON3)Cl2]2 mole-
cules, the CuII ions (see Fig. 3) are coordinated by the N and O
atoms of the OON3 ligands, one peripheral chlorido ligand and
two bridging chlorido ligands. The resulting coordination polyhe-
l
The CuII compound [Cu(OON1)Cl2]n was crystallised from ace-
tone solution and the structure was solved and refined in the
monoclinic space group P21/c (Tables 1 and Supplementary mate-
rial). The structure (Fig. 2) shows a one-dimensional stack of
p stacking along the crystallographic a axis
square planar coordination units consisting of CuII, with an g2
-
N,O bound OON1 ligand and two Cl ligands, connected along the
crystallographic c axis by rather long axial Cl–Cu bonds (using
one of the Cl ligands). While the Cu–Cl bond in the apical position
is 2.848(2) Å, the Cu–Cl distance opposite is 3.144(4) Å (Fig. 2).
Although this is a rather long Cu–Cl distance, similar [14,15] or
even longer distances (about 3.5 Å) [16] have been found for chlo-
ride bridged dimers. The overall geometry around the Cu atom is
thus square pyramidal (or asymmetrically elongated octahedral).
Similar to the FeIII complex, OON1 coordinates in a bidentate mode
to the CuCl2 fragment and a 50% disorder of the methoxy groups
was found. The OON1 ligands alternate in the chain structure so
that all ligands on one side of the chain possess the same chirality.
The CuII compound [Cu(OON3)Cl2]2 was crystallised from ace-
tone and the structure was solved and refined in the monoclinic
dron can be described as distorted square-pyramidal with the
l-
Cl (Cl1) atom in the axial positions or as distorted trigonal-bipyra-
midal with N1 and Cl10 as axial ligands. The two OON3 ligands
which are part of the binuclear molecule reveal opposite chirality,
in contrast to the iron complex. Again, the methoxy group is non-
coordinating (CuꢁꢁꢁOMe = 4.121(2) Å).
2.4. Structure determination using XAS
Since single crystals were not obtained for most of the new
compounds we embarked on a XAS study of selected compounds.
The edge position and the weak signal at around 8.977 keV in the

K. Butsch et al. / Inorganica Chimica Acta 374 (2011) 350–358
353
Table 2
iterated as a cross check, because the obtained coordination
numbers should agree with the XRD structures if the models used
to fit the EXAFS data are correct. In all cases, the M–C number is in
good agreement with the XRD values.
The Cu/Fe–O/N and Cu/Fe–C distances from the EXAFS measure-
ments are in all cases in good agreement with the XRD structures.
While in the case of [Cu(OON1)Cl2]n the adjustment of one copper-
backscatterer decreased the R-factor slightly, the R-factor was
slightly increased on introducing a Cu shell to the dimeric
[Cu(OON3)Cl2]2. Thus the assignment of [Cu(OON2)Cl2]2 having a
dimeric nature with a short Cu. . .Cu distance of about 3.2 Å seems
to be justified, the fit in this case was also slightly better after
the incorporation of a Cu shell.
Angles found for [Cu(OON3)Cl2]2 compared to angles of ideal coordination polyhedra.
Angles
Found (°)
TBP-ideal (°)
TPy-ideal (°)
O1–Cu1–Cl2
O1–Cu1–Cl1
Cl2–Cu1–Cl1
171.23(1)
85.42(1)
102.07(1)
358.06(2)
358.72(1)
120
120
120
180
90
90
R
R
aa
360
bb
360
a
R
R
a = (N1–Cu1–O1) + (N1–Cu1–Cl2) + (Cl1–Cu1–O1) + (Cl1–Cu1–O1).
b = (O1–Cu1–Cl2) + (O1–Cu1–Cl1) + (Cl2–Cu1–Cl1).
b
copper spectra indicates an oxidation state of CuII. The rather high
pre-peak intensity in the iron spectra is in accordance with the
fivefold coordination.
An FeꢁꢁꢁFe contribution was found for[Fe(OON1)Cl2]2; the FeꢁꢁꢁFe
distance agrees well with the XRD value. On the other hand, in
[Fe(OON1)Cl3] and [Fe(OON3)Cl3] no FeꢁꢁꢁFe interaction was found,
pointing to five-coordinated monomeric complex species. The EX-
AFS spectra are supplied in the Supplementary material.
The XANES spectra of the copper and iron samples are given in
Fig. S4. For both metals, all the samples look similar to each other,
i.e. no significant changes in the nearest neighbour contributions
are found, hence we assume the same O/N and Cl shells as for
the examples where XRD-structures were obtained. Since oxygen
and nitrogen can not be distinguished due to their similar ampli-
tude and phase functions, they are grouped in one backscattering
contribution in the EXAFS analysis (see Table 3 and Supplementary
material). Only the metalꢁꢁꢁcarbon coordination numbers were
2.5. EPR spectroscopy of CuII and FeIII compounds
For the CuII and FeIII compounds EPR spectra were measured at
298 K on powder samples and solutions (data collected in Table 4).
Table 3
EXAFS parameters determined for copper and iron complexes.
Sample
Abs-Bs
N(Abs)
Distance (XRD) (Å)
R(Abs-Bs) (Å)
r
(Å2)
R (%)
[Cu(OON1)Cl2]n
Cu–N/O
Cu–Cl
Cu–C
2
3
1.9863(3)/2.0016(2)
2.2494(3)/2.2549(3)/2.8481(4)
1.99 0.02
2.24 0.02
2.90 0.03
3.83 0.04
0.008 0.001
0.009 0.001
0.022 0.004
0.025 0.005
10.5
2.8 0.6
1
Cu–Cu
3.8146(4)
[Cu(OON2)Cl2]2
[Cu(OON3)Cl2]2
[Fe(OON1)Cl2]2
Cu–N/O
Cu–Cl
Cu–C
2
3
1.97 0.02
2.20 0.02
2.88 0.03
3.32 0.03
0.019 0.002
0.020 0.002
0.012 0.002
0.025 0.005
15.07
14.11
6.67
2.1 0.4
1
Cu–Cu
Cu–N/O
Cu–Cl
Cu–C
2
3
1.9951(7)/2.0024(5)
2.2676(6)/2.647(1)/2.2057(9)
1.99 0.02
2.21 0.02
2.89 0.03
3.31 0.03
0.024 0.002
0.016 0.002
0.014 0.003
0.025 0.005
2.4 0.5
1
Cu–Cu
3.268(1)
Fe–N/O
Fe–Cl
Fe–C
3
2
1.9852(7)/1.9507(7)/2.1072(6)
2.1741(7)/2.213(1)
1.97 0.02
2.20 0.02
2.96 0.03
3.12 0.03
0.013 0.001
0.007 0.001
0.002 0.001
0.011 0.002
1.8 0.4
1
Fe–Fe
3.1467(9)
[Fe(OON1)Cl3]
[Fe(OON3)Cl3]
Fe–N/O
Fe–Cl
Fe–C
2
3
1.98 0.02
2.19 0.02
2.99 0.03
0.025 0.003
0.012 0.002
0.025 0.005
15.99
12.55
3.4 0.7
Fe–N/O
Fe–Cl
Fe–C
3
2
1.99 0.02
2.21 0.02
3.04 0.03
0.014 0.003
0.005 0.001
0.025 0.005
2.9 0.6
Table 4
X-band EPR data of CuII and FeIII complexes.a
Sample
State
gav
g1 or g||
g2 or g\
g3
D
g
Geometryb
[Cu(OON1)Cl2]
[Cu(OON1)Cl2]n
[Cu(OON1)Cl2]
[Cu(OON2)Cl2]
[Cu(OON2)Cl2]
[Cu(OON3)Cl2]
[Cu(OON3)Cl2]
[Fe(OON1)Cl3]
[Fe(OON1)Cl2]2
[Fe(OON3)Cl3]
amorphous
polycryst.
dissolvedc
amorphous
dissolvedc
amorphous
dissolvedc
amorphous
polycryst.
amorphous
2.148
2.148
2.128
2.145
2.128
2.145
2.131
2.024
2.063
2.212
2.267
2.286
2.207
2.247
2.210
2.283
2.209
2.128
2.079
2.128
2.134
2.127
2.076
2.130
2.024
2.052
2.048
2.048
-
2.050
2.053
2.048
-
0.219
0.207
0.157
0.194
0.162
0.207
0.155
OE or Spy
OE or Spy
OE
OE or Spy
OE
OE or Spy
OE
TBP or OE
TBP
2.054
2.723
3.081
1.414
1.508
1.309
1.573
TBP or OE
a
b
c
CuII and FeIII measured at 298 K; averaged g value gav = (g|| + 2g\)/3 or gav = (g1 + g2 + g3)/3; g anisotropy
Symmetry assigned by XRD or assumed from EPR spectroscopy; SPy = square pyramidal, TBP = trigonal bipyramidal, OE = octahedral elongated.
Dissolved in an acetone/THF (1:1) mixture.
D
g = g\ꢂg|| or
D
g = g1ꢂg3.

354
K. Butsch et al. / Inorganica Chimica Acta 374 (2011) 350–358
g1
g1
g
g2
g2
100 G
500 G
g
100 G
g3
g3
Fig. 4. X-band EPR spectra of an amorphous powder of [Cu(OON1)Cl2] (left), polycrystalline powder of [Cu(OON1)Cl2]n (middle) and [Cu(OON1)Cl2] in acetone/THF (1:1)
(right); samples measured at 298 K.
Fig. 4 shows EPR spectra recorded for the amorphous material
obtained from the preparation procedure [Cu(OON1)Cl2] (parent
material), the recrystallised and structurally characterised binu-
clear complex [Cu(OON1)Cl2]n and an acetone/THF solution of the
parent material. While the spectra of the parent material
[Cu(OON1)Cl2] and its solution show rhombic symmetry, the poly-
meric compound exhibits an axial EPR signal with g|| > g\ (see
Table 4).
The latter signal shape is in line with the elongated square pyra-
midal (octahedral) coordination revealed from XRD. The rhombic
symmetry of the other two spectra indicates distorted square-
pyramidal coordination as found in the recently described complex
[(pydimH2)CuCl2] (gav = 2.149, g1 = 2.311, g2 = 2.094, g3 = 2.041,
Table 5
Selected NMR data for free OON ligands and ZnIIcomplexes.a
In CDCl3
d H6py
D
d
d OH
D
d
d OMe
Dd
OON1
[ZnCl2(OON1)]
OON2
[ZnCl2(OON2)]
OON3
[ZnCl2(OON3)]
8.54
8.71
8.49
8.73
8.55
8.74
5.20
7.30
5.12
6.55
5.71
7.72
3.86
3.75
3.56
3.71
3.53
3.81
0.17
0.24
0.19
2.10
1.43
2.01
0.11
0.15
0.28
In [D6]-acetone
OON1
[ZnCl2(OON1)]
OON2
[ZnCl2(OON2)]
OON3
8.50
8.76
8.42
8.70
8.46
8.73
5.33
8.77
5.10
7.94
5.67
8.18
3.83
3.84
3.55
3.49
3.49
3.66
0.26
0.28
0.27
3.44
2.84
2.51
0.01
0.06
0.17
D
g = 0.270) [13]. From the similarity of the average g value and
[ZnCl2(OON3)]
the g anisotropy ( g) of the parent compound and [(pydimH2)-
D
In [D7]-DMF
OON3
[ZnCl2(OON3)]
CuCl2] we conclude that in both compounds the ligand binds in a
tridentate fashion. From the significantly different values from
the solution spectrum it can be concluded that the binding situa-
tion has changed markedly upon dissolving the parent compound.
We assume a square pyramidal coordination sphere with a N,O0
bidentate binding mode and a solvent molecule binding to Cu, in
line with the findings from NMR spectroscopy of the corresponding
ZnII complexes (see later). This is true for all three Cu complexes.
The spectra of the iron complexes [Fe(OON1)Cl3] (parent mate-
rial) and [Fe(OON1)Cl2]2 (recrystallised) are not identical. The par-
ent material exhibits a broad isotropic EPR signal with a g value of
8.56
8.58
6.03
6.16
3.56
3.57
0.02
0.13
0.01
a
Chemical shifts d in ppm vs. TMS. Change of the chemical shift
dination in ppm.
D
d upon coor-
relatively small for the H6py atom and the OMe group (
0.28 ppm) but large for the OH group (1.4–2.1 ppm). In
6]-acetone the d values for the H6py and OH protons point to
coordination, while the d values for the OMe group for the com-
plexes with OON1 and OON2 drop to zero (indicating non-coordi-
nation). Qualitatively the same holds for [Zn(OON3)Cl2] in [ 4]-
Dd = 0.1–
[D
D
D
2.024 (
to this, the binuclear species reveals a rhombic spectrum with a
smaller g value (gav = 2.063) and a g of 1.309. The spectrum of
amorphous [Fe(OON3)Cl3] also has rhombic symmetry but higher
g values and g. The total spectral width of all three spectra is very
D
H ꢃ 3000 G) and a half-field signal at g = 4.305. In contrast
D
methanol, although the OMe proton signal overlaps with the signal
of the solvent protons and thus cannot be studied. Importantly, the
OH proton of the ligands can be unequivocally detected, thus the
ligands remain protonated upon coordination. In the coordinating
D
D
similar and importantly, the low g values of all complexes are in
line with a FeIII low-spin configuration [17]. Dissolving the two
parent amorphous FeIII complexes in an acetone/THF mixture leads
solvents [D7]-DMF and [D5]-pyridine, the Dd values strongly indi-
cate that the OON ligands were replaced by solvent molecules;
the shift observed for the OH proton might also be due to H bridge
formation with solvent molecules.
in both cases to isotropic, narrow (
value of 2.014. So far we could not assign this species.
D
H ꢃ 60 G) EPR spectra with a g
For the NiII complexes the 1H NMR spectra show paramagnetic
broadening, meaning that 2D experiments failed and no assign-
ments of the signals could be made. Tetrahedral or (distorted) octa-
hedral configurations can be assumed from the observed
paramagnetism. This is in contrast to recently reported Ni com-
plexes of oxido-pincer ligands with phenolate groups [2,6-bis(2-
methoxyphenyl)pyridine (LOMe2), 2,6-bis(2-hydroxyphenyl)pyri-
dine (LOH2), 2,6-bis-(2,4-dimethoxyphenyl)pyridine (LOMe4)]
which were diamagnetic, bromido bridged dimers [12].
2.6. NMR and UV–Vis spectroscopy in solution
From the ZnII compounds (d10 diamagnetic) NMR spectra could
be obtained, Table 5 lists selected data (for complete data see Sec-
tion 4). Special attention was paid to the chemical shifts of the OH,
OCH3 and H6py atoms since they are most sensitive to coordination
of the metal to the adjacent donor atoms. Since this coordination
might depend on the solvent polarity and/or the solvent’s ability
to act as a ligand, spectra were recorded in CDCl3 (unpolar, non-
UV–Vis absorption spectra were recorded in the range 200–
1000 nm (figures and data collected in the Supplementary mate-
rial). The absorption spectra of all complexes show intense absorp-
coordinating), [
D
6]-acetone (polar, presumably non-coordinating),
(highly polar, potentially coordinating),
5]-pyridine and [ 7]-DMF (both polar, presumably coordinating).
Upon coordination signals of the protons adjacent to donor
atoms usually exhibit a low-field shift. In CDCl3 these shifts are
[D
4]-methanol
tion bands in the UV range (k1), which can be assigned to p–
p⁄
[D
D
transitions within the ligands. As expected, the Zn complexes do
not show any further absorptions. Interestingly, the three iron
complexes exhibit quite similar absorption spectra indicating that

K. Butsch et al. / Inorganica Chimica Acta 374 (2011) 350–358
355
FeIII is still coordinated, in line with the EPR experiments. In addi-
tion to the
p⁄ absorption bands the iron and the copper com-
both the anisole and the pyridine ring contribute to the singly
occupied molecular orbital. The stabilisation of these radicals
might thus be similar as for triphenyl-methyl radicals and deriva-
tives [19]. Of course, more detailed investigations including quan-
tum chemical calculations, detailed spectroelectrochemical EPR
and UV–Vis experiments and further derivatives are necessary to
provide evidence for this effect.
p
–
plexes show bands in the range 300–400 nm. Based on their
intensities and their negative solvatochromism these bands were
assigned in both cases to halide-to-metal charge transfer (LMCT)
transitions. Within the series of experiments to determine the sol-
vatochromism, ligand de-coordination was observed for the copper
complexes for some solvents. In MeOH and EtOH solution correla-
tion of the LMCT energy with typical solvent parameters as the
Dimroth–Reichardt parameter (ET) [18] fails, probably due to the
protic character of these solvents (for details see Supplementary
material). While the iron complexes do not exhibit further bands,
for the copper and nickel complexes long-wavelength d–d (ligand
field) absorptions were observed in the visible range. For the cop-
per complexes it could be noted that the spectra of DMF solutions
differ markedly from spectra in other solvents indicating a de-
coordination of the OON ligands, in line with the NMR results ob-
tained for the zinc complexes. The long-wavelength bands of the
nickel complexes clearly show d–d bands typical for an octahedral
coordination around NiII giving proof for the binuclear character of
these complexes.
For the copper complexes reversible CuII/CuI redox couples
were found around 0 V indicating high flexibility of the coordina-
tion polyhedra. The three ligands do not influence the redox poten-
tials markedly. This is in line with the EPR experiments, which
already revealed that the CuII ions in the three complexes are mag-
netically (and thus structurally) identical in solution (Table 6). Fur-
ther reductions (presumably ligand-centred) and oxidations occur
irreversibly at potentials around ꢂ2 and 1 V, respectively (for de-
tails see Supplementary material).
The two iron complexes [Fe(OON1)Cl3] and [Fe(OON3)Cl3] show
reduction waves corresponding to the FeIII/FeII couple. To establish
the fully reversible character of these redox couples we synthes-
ised the corresponding FeII complexes (including the OON2 deriv-
ative, see Section 4) and found reversible oxidation waves at the
same potentials. Unfortunately, these three FeII compounds are vir-
tually insoluble, largely hampering further characterisation. For
the binuclear FeIII complex [Fe(OON1)Cl2]2 a reversible reduction
wave was detected at far lower potentials, basically in line with
the increased ligand-strength of the central O donator atom upon
deprotonation (OH ? Oꢂ).
2.7. Electrochemical and spectroelectrochemical investigations
Cyclic voltammetry was measured on the free ligands and the
new Cu, Fe and Ni complexes (Table 6). Surprisingly, the first
reduction of the ligands OON2 and OON3 was found to be fully
reversible, while the corresponding reduction wave for OON1 lies
far more negative and is irreversible. A UV–Vis spectroelectro-
chemical experiment confirmed that even under these conditions
(several minutes) the processes for OON2 and OON3 were revers-
ible (for Details see Supplementary material). The isotropic EPR
spectrum of the reduced species generated from OON3 using K/
cryptand [2.2.2] showed a three line pattern at g = 2.0034 (figure
and simulation in the Supplementary material). The best simula-
tion assumed two 1H nuclei coupling to the unpaired electron with
aH = 2.1 G. Comparison of the redox potentials of the three ligands
shows that variation of the residue (H, CH3, Ph) highly influences
the first reduction potential of the O,O0,N ligands and the stability
of the corresponding radicals. More importantly, for the corre-
sponding FeIII, NiII, CuII and ZnII complexes no such reversible li-
gand-centred reduction can be observed. In these complexes
ligand-centred reductions occur irreversibly and at very negative
potentials (as for OON1). This has lead us to the assumption that
a specific geometry is required to stabilise the corresponding rad-
icals, which is not the case for OON1 and the coordinated ligands
OON2 and OON3. The small coupling constants and the absence
of (or very small) coupling to the 14N nucleus let us assume that
The nickel complexes cannot be oxidised in a range between 0.0
and 2.0 V, and irreversible reductions were observed around
ꢂ1.8 V. They were assigned to NiII/NiI reduction followed either
by splitting of chlorido ligands or disproportionation to NiII and
Ni0. In both cases disintegration of the complexes will result and
indeed only the free ligands could be detected (by NMR) in electro-
lysed solutions.
3. Conclusion
The chiral ligands with O,O0,N donor sets presented here suc-
cessfully coordinate to the metal ions ZnII, CuII, NiII and FeIII. Single
crystals of [Fe(OON1)Cl2]2, [Cu(OON1)Cl2]n and [Cu(OON3)Cl2]2
were obtained and analysed by X-ray diffraction. EXAFS investiga-
tions reveal that these compounds were not the primary products
from the synthesis procedures but result from recrystallisation.
The primary products were mononuclear for CuII [Cu(O,O0,N)Cl2],
and FeIII [Fe(O,O0,N)Cl3]. For the ZnII derivatives [Zn(O,O0,N)Cl2]
the excellent solubility points to a mononuclear character, while
the very low solubility and NMR spectra of the NiII derivatives
points to binuclear complexes [Ni(O,O0,N)Cl2]2 with chlorido-
bridged, octahedrally configured NiII. In the solid structures (XRD
and EXAFS) and in solutions of coordinating solvents (as DMF,
THF or pyridine) the ligands were N,O0 coordinated with the weak-
est donor OMe uncoordinated. In non- or very weakly coordinating
solvents there is strong evidence that the ligands coordinate in a
tridentate fashion. Both CuII and FeIII complexes exhibit highly
reversible redox couples (CuII/CuI and FeIII/FeII) pointing to high
coordination flexibility for the ligands. Both the potentials of these
redox couples and the EPR spectra (of powders, recrystallised sam-
ples and solutions) vary only slightly upon variation of the ligands.
Table 6
Redox potentials of the CuII and FeIII complexes.a
Compound
E
½
(V)
Solvent
OON1
OON2
OON3
[Cu(OON1)Cl2]
[Cu(OON2)Cl2]
[Cu(OON3)Cl2]
[Fe(OON1)Cl3]
[Fe(OON3)Cl3]
[Fe(OON1)Cl2]2
ꢂ3.23 irr.b
ꢂ1.34 (65)
ꢂ1.35 (61)
ꢂ0.17 (81)
ꢂ0.15 (78)
ꢂ0.10 (79)
ꢂ0.39 (83)
ꢂ0.30 (80)
ꢂ1.36 (85)c
MeCN
MeCN
MeCN
THF
THF
THF
MeCN
MeCN
MeCN/DMF
For
the
three
crystallised
complexes
[Fe(OON1)Cl2]2,
[Cu(OON1)Cl2]n and [Cu(OON3)Cl2]2 both enantiomers of the chiral
ligands were found in the structures and very probably the chiral-
ity does not play a marked role in the spectroscopic and electro-
chemical results presented here. Detailed investigations on these
complexes in the future will be devoted to this aspect. Also, the
separation of the enantiomers and the investigation of their use
a
From cyclic voltammetry in solvent/nBu4NPF6 mixtures; potentials in V vs.
FeCp2/FeCp2+, peak-to-peak separation in parenthesis.
b
Irreversible reduction, given is the cationic peak potential in V.
An additional minor reversible reduction wave is observed at ꢂ0.34 V resulting
c
presumably from a mononuclear species from dissociation.

356
K. Butsch et al. / Inorganica Chimica Acta 374 (2011) 350–358
in catalysis will be attempted. Surprisingly, the ligands OON2 and
OON3 exhibit reversible reduction waves at moderately negative
potentials, while the ligand OON1 and all metal complexes exhibit
irreversible reduction behaviour with very negative potentials.
Although the generated reduced species could be studied by UV–
Vis and EPR spectroelectrochemistry, this interesting phenomenon
also requires some more investigations to allow an explanation.
consistently by taking the energy at half the edge jump [23]. To
determine the smooth part of the spectrum, corrected for pre-edge
absorption, a piecewise polynomial was used. It was adjusted in
such a way that the low-R components of the resulting Fourier
transform were minimal. After division of the background–sub-
tracted spectrum by its smooth part, the photon energy was con-
verted to photoelectron wave numbers k. The resulting
v(k)-
function was weighted with k3 and Fourier transformed using a
Hanning window function. Data analysis was performed in k-space
with Fourier filtered data. The filtered range was chosen according
to the range of significant data and is given in table A together with
the results of the fitting procedure. Adjustment of the common
theoretical EXAFS expression
4. Experimental
4.1. Instrumentation
Elemental analyses were carried out using a HEKAtech CHNS
EuroEA 3000 Analyzer. UV–Vis–NIR absorption spectra were mea-
sured on Varian Cary50 Scan or Shimadzu UV-3600 photo spec-
trometers. EPR spectra were recorded in the X-band on a Bruker
System ELEXSYS 500E equipped with a Bruker Variable Tempera-
ture Unit ER 4131VT. g values were calibrated using a dpph sample.
Electrochemical experiments were carried out in 0.1 M nBu4NPF6
solutions using a three-electrode configuration (glassy carbon elec-
trode, Pt counter electrode, Ag/AgCl reference) and an Autolab
PGSTAT30 potentiostat and function generator. The ferrocene/fer-
rocenium (FeCp2/FeCp2+) couple served as internal reference. Spec-
troelectrochemical UV–Vis experiments were performed using a
THF/nBu4NPF6 solution of OON3. The ligand radical anion for EPR
experiments were generated using a THF solution of OON3 and
reducing under Argon with potassium/cryptand [2.2.2].
X
ꢂ2k2r2j
Nj
kr2j
S2ðkÞFjðkÞe
e
ꢂ2r =k sin½2krj þ djðkÞꢀ
ð2Þ
j
vðkÞ ¼
0
j
(Nj: one type of neighbour atoms j in a shell; rj: distance of atoms j
from the X-ray absorbing atom; S2: amplitude reduction factor; Fj:
0
backscattering amplitude;
r
2: Debye–Waller like factor; dj: overall
phase shift) according to the curved wave formalism of the EX-
CURV98 program with XALPHA phase and amplitude functions [24].
The mean free path of the scattered electrons was calculated from
the imaginary part of the potential (VPI set to ꢂ4.00 eV). An inner
potential correction Ef was introduced when fitting experimental
data with theoretical models that accounts for an overall phase shift
between the experimental and calculated spectra.
In the fitting procedure, it was taken into account that the num-
ber of fitted parameters (Npars) did not exceed the degrees of free-
dom (Nind) which are calculated according to Nind = (2DkDR/p)
[25]. The quality of fit is given in terms of the R-factor according
4.2. Single crystal X-ray diffraction
to [26].
Crystal structure determinations were performed at 293(2) K
3
using graphite-monochromatised Mo Ka radiation (k = 0.71073 Å)
on a IPDS I (STOE and Cie.). The structures were solved by direct
k jvexpðkiÞ ꢂ vtheoðkiÞj
X
R ¼
ꢁ 100%
ð3Þ
k3jvexpðkiÞj
i
methods (SHELXS-97) [20] and refined by full-matrix least-squares
techniques against F2
(SHELXL-97) [21]. The numerical absorption
corrections (X-RED V1.22; Stoe & Cie, 2001) were performed after
optimising the crystal shapes using X-SHAPE V1.06 (Stoe & Cie,
1999) [22]. The non-hydrogen atoms were refined with anisotropic
displacement parameters without any constraints. All H atoms of
the free ligands including the OH group were found during the
refinement process.
4.4. Materials
4.4.1. 2-(Methoxy(pyridin-2-yl)methyl)phenol (OON1)
0.41 g (17.02 mmol, 2 eq) Mg was suspended in 20 mL dry
diethyl ether and 1.67 mL (3.0 g, 12.77 mmol, 1.5 eq) 2-iodometh-
oxybenzene in 20 mL dry diethyl ether were added slowly. The
reaction mixture was heated to 50–60 °C until it was brown and
turbid (ꢃ45 min). Then 0.91 g (8.5 mmol, 1 eq) pyridin-2-carboxal-
dehyde was added drop by drop and the mixture became orange.
After stirring overnight, 50 mL of water was slowly added to the
reaction mixture. After phase separation and extraction of the
aqueous phase with diethyl ether the combined organic phases
were dried over Na2SO4. The solvent was then removed under vac-
uum and the remaining solid was washed with a small portion of
acetone to yield 0.95 g (4.4 mmol, 52%) of an off-white powder.
4.3. XAS measurements
XAS measurements were performed at beamline X1 at the
Hamburger Synchrotron Strahlungslabor (HASYLAB) under ambi-
ent conditions. A Si(1 1 1) double crystal monochromator was used
for measurements at the Fe (7.112 keV) and Cu K-edge (8.979 keV).
The second monochromator crystal was tilt for optimal harmonic
rejection. The energy resolution for the Cu K-edge energy is
estimated to 1.0 eV. The spectra were recorded in transmission
mode with ionisation chambers filled with nitrogen. The individual
pressures were adjusted to optimise the signal to noise ratio.
Energy calibration was performed with an iron or copper metal
foil, respectively, which was measured simultaneously with the
samples between the second and third ionisation chamber. To
avoid errors in the XANES region due to small changes in the
energy calibration between two measurements, all spectra were
calibrated to the edge position of the copper foil. The solid samples
were embedded in an oxygen free cellulose matrix and pressed
into pellets.
1H NMR ([
D6]-acetone): 8.50 Hpy6, 1H, d; 7.69 Hpy4, 1H, t; 7.40
H
phen4/6, 2H, t; 7.22 Hpy3/5, 2H, m; 6.98 Hphen3, 1H, d; 6.92 Hphen5,
1H, t; 6.20 Hmethanol, 1H, s; 5.33 OH, 1H, s; 3.83 OMe, 3H, s. 13C
NMR (acetone-d6): 162 Cpy2, 1C, s; 157 Cphen2, 1C, s; 148 Cpy6, 1C,
d; 137 Cpy4, 1C, d; 132 Cphen1, 1C, s; 128 Cpy3, 1C, d; 127 Cphen6
1C, d; 122 Cpy5, 1C, d; 121 Cphen4, 1C, d; 120 Cphen5, 1C, d; 110
,
C
phen3, 1C, d; 69 Cmethanol, 1C d; 55 OMe1; 1C, q. Elemental Anal.
Calc. for C13H13O2N; M = 215.25 g molꢂ1: C, 72.54; H, 6.09; N,
6.51. Found: C, 73.45; H, 6.04; N, 6.46%.
4.4.2. 2-(Methoxy(pyridin-2-yl)ethyl)phenol (OON2)
Data evaluation started with background absorption removal
from the experimental absorption spectrum by subtracting a
Victoreen-type polynomial. Due to several inflection points in the
absorption edge, the threshold energy E0 was determined
2.78 mL 2-iodo methoxybenzene (5 g, 21 mmol, 1 eq) and
0.61 g (25 mmol, 1.2 eq) Mg were reacted in dry THF to give the
corresponding Grignard reagent. To this 2.75 mL (2.54 g, 21 mmol,
1 eq) 2-acetylpyridine were added drop by drop and the reaction

K. Butsch et al. / Inorganica Chimica Acta 374 (2011) 350–358
357
mixture was stirred for 16 h at 298 K. Then 100 mL of water was
4.4.7. [Zn(OON3)Cl2]
added and the resulting phases were separated. The aqueous phase
was extracted with diethyl ether, the combined organic phases
were dried over Na2SO4 and then the solvent was removed under
vacuum to yield a brown oil. After a few hours big colourless crys-
tals were formed and were separated by filtration and washing
100 mg (0.34 mmol, 1 eq) OON3 and 46 mg (0.34 mmol, 1 eq) of
anhydrous ZnCl2 were reacted to yield 104 mg (0.24 mmol, 71%) of
a colourless powder. NMR (300 MHz, [
6Py), 8.07 (t, 1H, H4Py), 7.65 (t, 1H, H5Py), 7.43 (t, 1H, H4Ph), 7.34 (m,
6H, H2,3,4,5,6Bnz), 7.14 (m, 2H, H5,6Ph), 6.91 (d, 1H, H3Ph), 6.72 (s, 1H,
D
6]-acetone): 1H: 8.73 (d, 1H,
H
with cold acetone. Yield: 3.0 g (13 mmol, 64%) 1H NMR ([
D
6]-ace-
tone): 8.42 Hpy6, 1H, d; 7.71 Hpy4, 1H, t; 7.65 Hpy3, 1H, d; 7.53
phen6, 1H, d; 7.23 Hphen4, 1H, t; 7.18 Hpy5, 1H, t; 6.96 Hphen3, 1H,
HOH), 3.66 (s, 3H, HOMe). Elemental Anal. Calc. for C19H17O2NCl2Zn;
M = 427.64 g molꢂ1: N, 3.28; C, 53.36; H, 4.01. Found: N, 3.15; C,
52.09; H, 4.04%.
H
t; 6.90 Hphen5, 1H, d; 5.09 OH 1H, s; 3.55 OMe, 3H, s; 1.86 Me,
3H, s. 13C NMR (acetone-d6): 168 Cphen1, 1C, s; 158 Cpy2, 1C, s;
148 Cphen2, 1C, s; 147 Cpy6, 1C, d; 136 Cpy4, 1C, d; 128 Cphen4, 1C,
4.4.8. [Cu(OON1)Cl2]
220 mg (1 mmol, 1 eq) OON1 were reacted with 138 mg
(1 mmol, 1 eq) of anhydrous CuCl2 to yield 283 mg (0.81 mmol,
81%) of a green solid. Elemental Anal. Calc. for C13H13O2NCl2Cu;
M = 349.70 g molꢂ1: N, 4.01; C, 44.65; H, 3.75. Found: N, 3.88; C,
45.21; H 3.89%.
d; 127 Cpy3, 1C, d; 121 Cpy5, 1C, d; 120 Cphen5/6, 2C, d; 112 Cphen3
,
1C, d; 76 COH, 1C, s; 55 Cphen2, 1C, q; 28 CMe, 1C q. Elemental Anal.
Calc. for C14H15O2N; M = 229.27 g molꢂ1: C, 73.34; H, 6.59; N, 6.11.
Found: C, 73.23; H, 6.54; N, 6.06%.
4.4.3. 2-(Methoxy(pyridin-2-yl)benzyl)phenol (OON3)
4.4.9. [Cu(OON2)Cl2]
2.78 mL 2-iodo methoxybenzene (5 g, 21 mmol, 1 eq) and
612 mg (25 mmol, 1.2 eq) Mg were reacted in 30 mL dry THF to
give the corresponding Grignard reagent. To this 3.84 g (21 mmol,
1 eq) benzoylpyridine were added slowly. The reaction mixture
was stirred overnight at 298 K. Then 50 mL water was slowly
added and the resulting phases were separated. The aqueous phase
was extracted with diethyl ether, the combined organic phases
were dried over Na2SO4 and the solvent was removed under vac-
uum. The crude product, a colourless solid, was recrystallised from
acetone to yield 5.4 g (15 mmol, 72%) of colourless crystals. 1H
100 mg (0.44 mmol, 1 eq) OON2 were reacted with 59 mg (0.
44 mmol, 1 eq) of anhydrous CuCl2 to yield 107 mg (0.29 mmol,
66%) of a green solid. Elemental Anal. Calc. for C14H15O2NCl2Cu;
M = 363.73 g molꢂ1: N, 3.85; C, 46.23; H, 4.16. Found: N, 3.97; C,
47.92; H, 4.26.
4.4.10. [Cu(OON3)Cl2]
100 mg (0.34 mmol, 1 eq) OON3 were reacted with 46 mg
(34 mmol, 1 eq) of anhydrous CuCl2 to yield 103 mg (0.24 mmol,
71%) of a green solid. Elemental Anal. Calc. for C19H16O2NCl2Cu;
M = 424.79 g molꢂ1: N, 3.30; C, 53.72; H, 3.80. Found: N, 3.27; C,
52.99; H, 3.73%.
NMR ([
1H, d; 7.39 Hbenz2/6, 2H, d; 7.27 Hbenz3/4/5/phen4/py5, 5H, m; 6.99
phen3, 1H, d; 6.95 Hphen6, 1H, d; 6.85 Hphen5, 1H, t; 5.67 OH 1H,
s; 3.49 OMe, 3H, s; 13C NMR (acetone-d6): 167 Cpy2, 1C, s; 159
phen1, 1C, s; 147 Cbenz1, 1C, s; 148 Cpy6, 1C, d; 137 Cpy4, 1C, d;
130 Cphen6, 1C, d; 128 Cbenz2/6/phen4, 3C, d; 127 Cbenz3/4/5, 3C; d;
123 Cpy3/5, 2C, d; 121 Cphen5, 1C, d; 121 Cphen1, 1C, s; 113 Cphen3
1C, d; 82 COH, 1C, s; 56 Cphen2, 1C, q. Elemental Anal. Calc. for
19H17O2N; M = 365.57 g molꢂ1: C, 78.33; H, 5.88; N, 4.81. Found:
D6]-acetone): 8.46 Hpy6, 1H, d; 7.76 Hpy4, 1H, t; 7.59 Hpy3,
H
4.4.11. [Ni(OON1)Cl2]
C
100 mg (0.46 mmol, 1 eq) OON1 were reacted with 110 mg
(0.46 mmol, 1 eq) NiCl2ꢁ6H2O to yield 60 mg (0.17 mmol, 37%) of
,
a green solid. NMR (300 MHz, [D
6]-acetone): 1H: 11.46; 7.70;
7.46; 7.20; 6.94; 6.80; 5.38; 4.80; 4.28; 3.83 ppm. Elemental Anal.
Calc. for C13H13O2NCl2Ni; M = 344.85 g molꢂ1: N, 4.06; C, 45.28; H,
3.80. Found: N, 3.88; C, 43.72; H, 3.98%.
C
C, 78.14; H, 6.01; N, 5.01%.
4.4.4. General procedure for the synthesis of the complexes
1 eq of the OON-ligand was dissolved in methanol and 1 eq of
the metal chloride (MCl2 or MCl3, respectively), dissolved in meth-
anol/acetone (1:1) was added. The mixture was stirred at 298 K for
14 h. The solvent was removed under reduced pressure and the
remaining solids were washed with cold acetone.
4.4.12. [Ni(OON2)Cl2]
100 mg (0.43 mmol, 1 eq) OON2 were reacted with 102 mg
(0.43 mmol, 1 eq) NiCl2ꢁ6H2O to yield 74 mg (0.21 mmol, 49%) of
a green solid. NMR (300 MHz, [D
6]-acetone): 1H: 15.49; 13.27;
8.42; 7.51; 6.95; 5.52; 4.49; 2.00 ppm. Elemental Anal. Calc. for
C
14H15O2NCl2Ni; M = 356.98 g molꢂ1: N, 3.90; C, 46.85; H, 4.21.
Found: N, 4.09; C, 44.52; H, 4.05%.
4.4.5. [Zn(OON1)Cl2]
100 mg (0.46 mmol, 1 eq) of OON1 were reacted with 79 mg
(0.46 mmol, 1 eq) of anhydrous ZnCl2 to yield 160 mg (0.45 mmol,
4.4.13. [Ni(OON3)Cl2]
100 mg (0.34 mmol, 1 eq) OON3 were reacted with 82 mg
(0.34 mmol, 1 eq) NiCl2ꢁ6H2O to yield 84 mg (0.20 mmol, 59%) of
97%) of a colourless powder. NMR (300 MHz, [
8.77 (s, 1H, HOH), 8.76 (d, 1H, H6Py), 8.12 (t, 1H, H4Py), 7.71 (t, 1H,
5Py), 7.48 (d, 1H, H6Ph), 7.48 (d, 1H, H3Py), 7.39 (t, 1H, H4Ph), 7.11
(d, 1H, H3Ph), 6.97 (t, 1H, H5Ph), 6.75 (s, 1H, Hmethanol), 3.84 (s, 3H,
D
6]-acetone): 1H:
a green solid. NMR (300 MHz, [D
6]-acetone): 1H: 14.10; 8.12;
H
7.47; 6.93; 4.60; 3.54 ppm. Elemental Anal. Calc. for C19H17O2N-
Cl2Ni; M = 420.94 g molꢂ1: N, 3.33; C, 54.21; H, 4.07. Found: N,
3.39; C, 52.85; H, 4.24%.
H
OMe). Elemental Anal. Calc. for C13H13O2NCl2Zn; M = 351.54
g molꢂ1: N, 3.98; C, 44.42; H, 3.73. Found: N, 3.81; C, 43.20; H,
3.71%.
4.4.14. [Fe(OON1)Cl3]
100 mg (0.46 mmol, 1 eq) OON1 were reacted with 92 mg
(0.46 mmol, 1 eq) FeCl3ꢁ6H2O to yield 157 mg (0.42 mmol, 91%)
of a dark red solid. Elemental Anal. Calc. for C13H13O2NCl3Fe;
M = 377.45 g molꢂ1: N, 3.71; C, 41.37; H, 3.47. Found: N, 3.84; C,
42.67; H, 3.56%.
4.4.6. [Zn(OON2)Cl2]
100 mg (0.44 mmol, 1) eq OON2 and 59 mg (0.44 mmol, 1 eq) of
anhydrous ZnCl2 were reacted to yield 101 mg (0.28 mmol, 64%) of
a colourless powder. NMR (300 MHz, [
6Py), 8.02 (t, 1H, H4Py), 7.94 (s(br), 1H, HOH), 7.76 (d, 1H, H6Ph),
7.63 (t, 1H, H5Py), 7.23 (d, 1H, H3Py), 7.43 (t, 1H, H4Ph), 7.08 (t, 1H,
D
6]-acetone): 1H: 8.70 (d, 1H,
H
4.4.15. [Fe(OON1)Cl2]2
H
5Ph), 6.99 (d, 1H, H3Ph), 3.49 (s, 3H, HOMe), 2.08 (s, 3H, HMe).
50 mg (0.13 mmol, 2 eq) [FeCl3(OON1)] were dissolved in ace-
tone and stirred for 60 h. A dark red precipitate was formed and fil-
tered off. After washing with acetone 37 mg (0.05 mmol, 92%) were
Elemental Anal. Calc. for C14H15O2NCl2Zn; M = 365.57 g molꢂ1: N,
3.83; C, 46.00; H, 4.14. Found: N, 3.90; C, 44.91; H, 4.23%.

358
K. Butsch et al. / Inorganica Chimica Acta 374 (2011) 350–358
obtained. Anal. Calc. for C26H24O4N2Cl6Fe2; M = 681.98 g molꢂ1: N,
3.55; C, 45.79; H, 4.11. Found: N, 4.06; C, 44.68; H, 3.46%.
(c) J.M. Hawkins, K.B. Sharpless, Tetrahedron Lett. 28 (1987) 2825;
(d) M. Albrecht, G. van Koten, Angew. Chem. Int. Ed. 40 (2001) 3750.
[2] (a) J.A. Thich, C.C. Ou, D. Powers, B. Vasiliou, D. Mastropaolo, J.A. Potenza, H.J.
Schugar, J. Am. Chem. Soc. (1976) 1425;
(b) P.A. Brayshaw, J.-C.G. Bünzli, P. Froidevaux, J.M. Harrowfield, Y. Kim, A.N.
Sobolevt, Inorg. Chem. 34 (1995) 2068;
(c) F. Renaud, C. Piguet, G. Bernardinelli, J.-C.G. Bünzli, G. Hopfgartner, J. Am.
Chem. Soc. 121 (1999) 9326.
4.4.16. [Fe(OON2)Clx]y
170 mg (0.74 mmol, 1 eq) OON2 were reacted with 148 mg
(0.74 mmol, 1 eq) FeCl3ꢁ6H2O to yield 253 mg of a bright orange so-
lid. The product is presumably polymeric, contains more than one
FeIII species (EPR spectroscopy) and the exact stoichiometry of this
compound could not be determined. Elemental Anal. found: N,
3.07; C, 49.66; H, 3.75%.
[3] R. Castarlenas, M.A. Esteruelas, E. Onate, Organometallics 26 (2007) 3082.
[4] (a) M. Melnik, C.E. Holloway, J. Coord. Chem. 49 (1999) 69;
(b) R.M. Gauvin, J.A. Osborn, J. Kress, Organometallics 19 (2000) 2944;
(c) E. Gómez, V. Santes, V. de la Luz, N. Farfán, J. Organomet. Chem. 622 (2001)
54.
[5] (a) K.V. Zaitsev, M.V. Bermeshev, S.S. Karlov, Y.F. Oprunenko, A.V. Churakov,
J.A.K. Howard, G.S. Zaitseva, Inorg. Chim. Acta 360 (2007) 2507;
(b) R. Fandos, B. Gallego, A. Otero, A. Rodriguez, M.J. Ruiz, P. Terreros, C. Pastor,
Organometallics 26 (2007) 2896;
(c) R.J. Fites, A.T. Yeager, T.L. Sarvela, W.A. Howard, G. Zhu, K. Pang, Inorg. Chim.
Acta 359 (2006) 248;
(d) J.M. Berg, R.H. Holm, Inorg. Chem. 22 (1983) 1768.
[6] (a) C. Boskovic, W. Wernsdorfer, K. Folting, J.C. Huffman, D.N. Hendrickson, G.
Christou, Inorg. Chem. 41 (2002) 5107;
4.4.17. [Fe(OON3)Cl3]
100 mg (0.34 mmol, 1 eq) OON3 were reacted with 68 mg
(0.34 mmol, 1 eq) FeCl3ꢁ6H2O to yield 124 mg (0.27 mmol, 79%)
of a brown-red solid. Elemental Anal. Calc. for C19H17O2NCl3Fe;
M = 453.55 g molꢂ1: N, 3.09; C, 50.32; H, 3.78. Found: N, 3.07; C,
49.66; H, 3.75%.
(b) J. Yoo, E.K. Brechin, A. Yamaguchi, M. Nakano, J.C. Huffman, A.L. Maniero, L.-
C. Brunel, K. Awaga, H. Ishimoto, G. Christou, D.N. Hendrickson, Inorg. Chem.
39 (2000) 3615;
4.4.18. [Fe(OON1)Cl2]
(c) E.K. Brechin, J. Yoo, M. Nakano, J.C. Huffman, D.N. Hendrickson, G. Christou,
Chem. Commun. (1999) 783.
[7] (a) V.T. Yilmaz, S. Güney, O. Andac, W.T.A. Harrison, J. Coord. Chem. 56 (2003)
21;
100 mg (0.46 mmol, 1 eq) OON1 were reacted with 93 mg
(0.46 mmol, 1 eq) FeCl2ꢁ4 H2O to yield 140 mg (0.41 mmol, 89%)
of an orange-brown solid. Elemental Anal. Calc. for C13H13O2N-
Cl2Fe; M = 342.00 g molꢂ1: C, 45.65; H, 3.83; N, 4.10. Found: C,
47.87; H, 3.56; N, 4.02%.
(b) O. Andac, S. Güney, Y. Topcu, V.T. Yilmaz, W.T.A. Harrison, Acta Crystallogr.,
Sect. C 58 (2002) m17;
(c) S. Winter, W. Seichter, E. Weber, J. Coord. Chem. 57 (2004) 997;
(d) S. Winter, W. Seichter, E. Weber, Z. Anorg. Allg. Chem. 630 (2004)
434;
(e) M. Koman, M. Melnik, Polyhedron 16 (1997) 2721;
(f) M. Melnik, M. Koman, L. Macaskova, T. Glowiak, R. Grobelny, J. Mrozinsky, J.
Coord. Chem. 43 (1998) 159;
4.4.19. [Fe(OON2)Cl2]
94 mg (0.41 mmol, 1 eq) OON2 and 82 mg (0.41 mmol, 1 eq)
FeCl2ꢁ4 H2O were dissolved in 5 mL methanol each. Both solutions
were combined and stirred for 6 h at room temperature. During
this time an intensive coloured, orange precipitate was formed,
which was filtered off and washed with three portions of cold
methanol to yield 139 mg (0.39 mmol, 95%). Elemental Anal. Calc.
for C14H15O2NCl2Fe; M = 356.03 g molꢂ1: C, 47.23; H, 4.25; N,
3.93. Found: C, 48.65; H, 4.20; N, 3.98%.
(g) M. Koman, M. Melnik, J. Moncol, Inorg. Chem. Commun.
262.
3 (2000)
[8] A.K. Gupta, J. Kim, Acta Crystallogr., Sect. C 59 (2003) m262.
[9] (a) J.A. Davies, F.R. Hartley, Chem. Rev. 81 (1981) 79;
(b) R.G. Pearson, Inorg. Chem. 27 (1988) 734;
(c) R.G. Pearson, J. Am. Chem. Soc. 85 (1963) 3533.
[10] A. Klein, S. Elmas, K. Butsch, Eur. J. Inorg. Chem. (2009) 2271.
[11] (a) T. Kawato, H. Koyama, H. Kanatomi, Y. Muramoto, Inorg. Chim. Acta 183
(1991) 107;
(b) F.D. Rochon, A.L. Beauchamp, C. Bensimon, Can. J. Chem. 74 (1996)
2121.
[12] A. Klein, K. Butsch, J. Neudörfl, Inorg. Chim. Acta 363 (2010) 3282.
[13] A. Klein, K. Butsch, S. Elmas, D. Heift, S. Nitsche, I. Schlipf, H. Bertagnolli,
Polyhedron, submitted for publication.
[14] J.C. Wilson, P.D. Verweij, W.L. Driessen, J. Reedijk, Inorg. Chim. Acta 192 (1992)
219.
[15] G.J.A.A. Koolhaas, W.L. Driessen, P.J. Van Koningsbruggen, J. Reedijk, A.L. Spek,
J. Chem. Soc., Dalton Trans. (1993) 3803.
4.4.20. [Fe(OON3)Cl2]
97 mg (0.33 mmol, 1 eq) OON3 and 66 mg (0.33 mmol, 1 eq)
FeCl2ꢁ4 H2O were dissolved in 5 mL methanol each. Both solutions
were combined and stirred for 6 h at room temperature. The sol-
vent of the resulting orange-brown solution was removed under
vacuum to yield 83 mg (0.20 mmol, 61%) of a brown powder. Ele-
mental Anal. Calc. for C19H17O2NCl2Fe; M = 418.09 g molꢂ1: C,
54.58; H, 4.10; N, 3.35. Found: C, 53.88; H, 3.92; N, 3.45.
[16] C.-D. Zhang, S.-X. Liu, C.-Y. Sun, F.-J. Ma, Z.-M. Su, Cryst. Growth Des. 9 (2009)
3655.
[17] R. Viswanathan, M. Palaniandavar, T. Balasubramanian, P.T. Muthiah, J. Chem.
Soc., Dalton Trans. (1996) 2519.
[18] C. Reichardt, Angew. Chem. 1 (1965) 30.
Acknowledgments
[19] (a) N.I. Tzerpos, A.K. Zarkadis, R.P. Kreher, L. Repas, M. Lehnig, J. Chem. Soc.,
Perkin Trans. 2 (1995) 755;
We would like to thank Dr. Ingo Pantenburg and Ingrid Müller
from the Institute for Inorganic Chemistry for the X-ray measure-
ments and the ‘‘Studienstiftung des Deutschen Volkes’’ for financial
support (KB). Dr. Eric J. L. McInnes, School of Chemistry, The Uni-
versity of Manchester, UK is acknowledged for helpful comments..
(b) D. Dünnebacke, W.P. Neumann, A. Penenory, U. Stewen, Ber. (1988)
533;
(c) M. Ando, S. Yoshiike, T. Suzuki, T. Ichimura, T. Okutsu, M. Ueda, H. Horiuchi,
H. Hiratsuka, A. Kawai, K. Shibuya, J. Photochem. Photobiol. A: Chem. 174
(2005) 194.
[20] (a) G.M. Sheldrick, Acta Cryst. A (2008) 112;
[b] G.M. Sheldrick, SHELX-97, Programs for Crystal Structure Analysis, Göttingen,
1997.;
Appendix A. Supplementary material
(c) L.J. Farrugia, J. Appl. Cryst. 32 (1999) 837.
[21] G.M. Sheldrick, SHELXL-97, Program for the Refinement of Crystal Structures,
Universität Göttingen, 1997.
[22] [a] STOE X-RED, Data Reduction Program, Version 1.22/Windows, STOE & Cie,
Darmstadt, 2001;
[b] STOE X-SHAPE, Crystal Optimisation for Numerical Absorption Correction,
Version 1.06/Windows, STOE & Cie, Darmstadt, 1999.
[23] (a) T.S. Ertel, H. Bertagnolli, S. Hückmann, U. Kolb, D. Peter, Appl. Spectrosc. 46
(1992) 690;
(b) M. Newville, P. Livins, Y. Yacoby, J.J. Rehr, E.A. Stern, Phys. Rev. B 47 (1993)
14126.
[24] S.J. Gurman, N. Binsted, I. Ross, J. Phys. C. 17 (1984) 143.
[25] E.A. Stern, Phys. Rev. B 48 (1993) 9825.
CCDC 801045–801049 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Supplementary data associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.ica.2011.02.072.
References
[26] N. Binsted, in: F. Mosselman (Ed.), EXCURV98, CCLRC Daresbury Laboratory
Computer Program, Manual, 1998.
[1] (a) J.T. Singleton, Tetrahedron 59 (2003) 1837;
(b) M.E. van der Boom, D. Milstein, Chem. Rev. 103 (2003) 1759;