Electron transfer studies on Cu(II) complexes bearing phenoxy-pincer ligands

Article abstract of DOI:10.1016/j.ica.2010.06.011

Oxido-pincer ligands with phenolate-groups [2,6-bis(2-methoxyphenyl) pyridine (LOMe₂), 2,6-bis(2-hydroxyphenyl)-pyridine (LOH ₂), 2,6-bis-(2,4-dimethoxyphenyl)-pyridine (LOMe₄)] coordinate to Cu^II forming binucl

Full text of DOI:10.1016/j.ica.2010.06.011

Inorganica Chimica Acta 363 (2010) 3282–3290
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Electron transfer studies on Cu(II) complexes bearing phenoxy-pincer ligands
Axel Klein , Katharina Butsch , Jörg Neudörfl b
a,*
a
a
Universität zu Köln, Department für Chemie, Institut für Anorganische Chemie, Greinstrasse 6, D-50939 Köln, Germany
Universität zu Köln, Department für Chemie, Institut für Organische Chemie, Greinstrasse 4-6, D-50939 Köln, Germany
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Oxido-pincer ligands with phenolate-groups [2,6-bis(2-methoxyphenyl)pyridine (LOMe ), 2,6-bis(2-
hydroxyphenyl)-pyridine (LOH ), 2,6-bis-(2,4-dimethoxyphenyl)-pyridine (LOMe )] coordinate to Cu
2 4
2
Received 3 February 2010
Received in revised form 31 May 2010
Accepted 8 June 2010
II
forming binuclear complexes which can be easily and reliably converted into mononuclear species. Their
physical properties were analysed using EPR, optical spectroscopy and (spectro-)electrochemical meth-
Available online 15 June 2010
II
ods. The results were compared to those of related Ni complexes and discussed in view of Cu-containing
II
I
metalloenzymes. Due to the ligands flexibility the Cu /Cu redox couple exhibits high reversibility, while
the ligand-centred oxidation leads to highly reactive phenoxy radicals. Reduction of the LOH complex
leads to sequential deprotonation. The ligand LOMe and the derived complexes show blue luminescence,
which can be rationalised from its molecular structure (analysed by XRD).
Keywords:
2
Oxido-pincer ligands (O,N,O ligands)
Coordination chemistry
Electrochemistry
Acid–base-properties
EPR spectroscopy
4
Ó 2010 Elsevier B.V. All rights reserved.
1
. Introduction
state [9]. Complex systems designed to model enzyme functions
have to spotlight on the coordination geometry and the resulting
electrochemical properties. Suitable models need a flexible copper
coordination and ligands that exhibit ligand to metal interactions
Copper is an important metal for oxidation catalysis in biologi-
cal systems. Blue copper proteins (e.g., azurin or plastocyanin),
tyrosinase, amine oxidase, -ascorbat oxidase, cytochrom C oxi-
3
I
L
2
[10] supporting the O affinity of Cu [11] which is necessary for
dase, galactose oxidase (GO) and catecholase are just a few exam-
ples for copper containing enzymes. The copper ions in these
enzymes are able to catalyse electron transfer of one electron
biological catalytic oxidation reactions.
Recently we reported on a number of complexes containing
O,N,O-pincer ligands derived from 2,6-bis(hydroxymethyl)pyridine
[12]. These oxido-pincer ligands bind (in their protonatedforms) in a
tridentate mode, forming two five-membered coplanar rings in
(
e.g., blue copper proteins [1]), two electrons (e.g., GO [2] or
catecholase [3]) or four electrons (e.g., ascorbat oxidase [4]).
Ascorbat oxidase and catecholase both transfer more than one
electron due to the number of copper atoms per active centre
2
+
2 2
complexes [(O,N,O)CuCl ] or [(O,N,O) Cu] [12]. This motivated us
to extend our studies to phenoxido-pincer-ligands as depicted in
Scheme 1. These ligands can bind through two six-membered
chelates, which should allow effective binding combined with
(
two or four, respectively), while GO contains one copper atom
Å+
linked to a secondary ‘‘built-in” co-factor, a tyrosyl-radical ([Tyr] )
which accepts the second electron. Therefore, GO is able to catalyse
the oxidation of terminal alcohols to the corresponding aldehydes
along with the formation of hydrogen peroxide [5]. The copper
atoms in these metalloenzymes change their redox state (Cu /Cu )
reversibly and rapidly, which is important for efficient electron
transfer [6]. The Cu /Cu mediated electron transfer reactions
occur at low potentials, e.g., the copper redox couple of GO has a
potential of 0.16 V versus NHE and the oxidation potential of the
tyrosyl-radical is 0.41 V versus NHE [7]. This corresponds to
II
strong ligand distortion. As recently shown for Cu the size of the
0
O,N,O binding site in 2,6-bis-(2 -hydroxyphenyl)pyridine is far to
small to allow a coplanar ligand arrangement of the three aryl rings
[13]. As a result the coordination flexibility in such ligands should be
higher and a fast conformation change associated with the change of
I
II
II
I
II
I
the redox state (Cu to Cu ) should be possible. Therefore, complexes
of such phenolate-pincers ligands are promising candidates for the
modelling ofbio-inspiredoxidationcatalysts. In this paper wereport
on the syntheses of phenoxido-pincer ligands (Scheme 1) and their
II
I
Å+
II
II
ꢀ
0.24 V (Cu /Cu reduction) and 0.01 V (ligand oxidation [L ])
2
complexes with Cu and Ni . LOH is the only ligand in this series,
þ
versus FeCp
2
/FeCp2 [8]. The low redox potentials in natural
which can alternatively bind in its protonated or deprotonated
forms. Furthermore it is reasonable to assume that the deprotona-
tion is facilitatedupon coordination. Generally, acid–base properties
of the redox active centre are important, since oxidation/oxygena-
tion reactions are often associated with proton transfer. To this
enzymes are always linked with unusual coordination geometries
and strongly distorted coordination polyhedra, the so called entatic
*
Corresponding author.
2
end, the comparison between the LOH containing model system
E-mail address: axel.klein@uni-koeln.de (A. Klein).
0
020-1693/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2010.06.011

A. Klein et al. / Inorganica Chimica Acta 363 (2010) 3282–3290
3283
N
N
N
O
O
O
O
O
O
OH HO
CH3
CH3 CH3
CH3
CH3 CH3
LOMe4
LOMe2
LOH2
Scheme 1. O,N,O oxido-pincer ligands used in this study. 2,6-Bis-(2,4-dimethoxyphenyl)-pyridine (LOMe
hydroxyphenyl)pyridine (LOH ).
4 2
), 2,6-bis(2-methoxyphenyl)pyridine (LOMe ) and 2,6-bis(2-
2
with its LOMe
2
derivative is worthwhile studying. The LOMe
4
ligand
ligands and their metal complexes we will focus the description of
the molecular structures to the torsion angles between the ring-
planes (Table 1).
In the ligand LOMe both phenol rings are markedly tilted from
2
the central pyridine ring in a way that the two methoxy groups
point into different directions one lying above the pyridine ring
was introduced to this series for the electron-releasing influence of
the additional methoxy substituents and its higher steric bulk. The
new compounds were fully characterised spectroscopically (NMR,
UV–Vis–NIR absorption, UV–Vis emission, EPR) and electrochemi-
II
cally. The suitability of the Cu complexes to serve as oxidation cat-
alysts was investigated in detail by absorption spectroscopy of
parent, reduced and oxidised states using spectroelectrochemical
techniques and by catalytic test reactions [14].
plane and the other one below. Thus the
p
-overlap between the
both tilt an-
different ring-planes in LOMe is negligible. In LOMe
2
4
gles are slightly smaller and the methoxy groups point in the same
direction opposite of the binding pocket, representing a situation
far from the terdentate coordination mode in metal complexes.
One of the methyl groups in p-position is disordered (1:3). The tilt
2
. Results and discussion
2
angles of the LOH ligand are the smallest found in this series and
2.1. Syntheses and characterisation of the ligands
the two hydroxy groups point into the same direction. In summary
all three ligands exhibit molecular structures which result from
steric interactions and do not represent the terdentate coordina-
tion mode. In order to coordinate in a bis-chelate manner they
have to undergo conformational changes by rotating the phenyl
substituents. The corresponding rotating energies are probably
rather small, while on the other hand the systems might gain en-
2
While the ligands 2,6-bis(2-methoxyphenyl)pyridine (LOMe )
and 2,6-bis(2-hydroxyphenyl)pyridine (LOH
before [13,15,16], the 2,6-bis-(2,4-dimethoxyphenyl)pyridine
LOMe ) ligand is a new compound and was synthesised via a
Kumada-coupling reaction (for details see Section 4). The LOMe li-
gand was first isolated as a MgBr complex as inferred from NMR
spectroscopy and elemental analysis. The free LOMe ligand was
obtained by removing the MgBr from the compound using a
2
) have been described
(
4
4
2
ergy from more efficient
mation. However, from a study on the related ligand 2-(2-
hydroxy-5-methylphenyl)-6-(2-hydroxyphenyl)-pyridine (4-
) and its dipyridine Cu complex [18] it
is evident that one of the phenol substituents is in a coplanar
p-delocalisation in the all-planar confor-
4
2
cryptand, while other methods like acidifying the system, filtration
over silica or treatment with crown ether were unsuccessful. The
three ligands were obtained in sufficient yields and fully character-
II
methyl derivative of LOH
2
ꢁ
ꢁ ꢁ
1
13
arrangement to the central pyridine core (supported by a O–H N
ised by elemental analysis, H and C NMR spectroscopy (see Sec-
tion 4), by absorption spectroscopy and cyclic voltammetry (CV).
II
hydrogen bond), while in the corresponding Cu complex all three
rings are tilted toward each other, due to the small cavity provided
by the O,N,O donor atoms (the mentioned proton fits to the cavity).
From LOMe
tained from acetone solutions. The structure of LOMe
in monoclinic Cc, while LOMe crystallises in the chiral orthorhom-
bic space group P2 . This is in line with the chirality of the li-
gand scaffold in the solid state [17]. Fig. 1 shows the two molecular
2
and LOMe
4
single crystals suitable for XRD were ob-
2
was solved
II
Thus an all-planar configuration cannot be expected for the Cu or
4
II
Ni complexes.
1 1 1
2 2
structures (for details of the LOMe
porting information).
4
crystal measurement see Sup-
Table 1
Selected data of the molecular structures of LOMe
2
and LOMe
LOMe
4
.
The crystal structure of LOMe
clinic space group Ia together with the crystal structure of LOH
16]. Although Ia is a non-standard but completely equivalent set-
ting of Cc, the parameters reported there are not identical to ours.
Assuming that -overlap between the different ring-planes is cru-
cial for the spectroscopic and electrochemical properties of the free
2
has been reported in the mono-
Dihedral angles
LOH
2
2
LOMe
4
2
a
a
a
PlaneO1–planeN
PlaneO2–planeN
PlaneO1–planeO2
ꢂ26
ꢂ30
ꢂ51
44.40(6)
51.75(8)
78.92(8)
35.39(9)
40.31(9)
38.06(9)
[
p
a
Averaged values from three independent molecules from Ref. [16].
Fig. 1. View of the molecular structure of LOMe
2 4
(left) and LOMe (right) (thermal ellipsoids represent 50% probability); the H atoms were omitted for clarity. Note that one of
the methoxy groups of LOMe is disordered with a 30/70 occupancy ratio.
4

3
284
A. Klein et al. / Inorganica Chimica Acta 363 (2010) 3282–3290
2.2. Synthesis of the metal complexes
(higher potentials for the metal-centred redox chemistry). The
nickel compounds were synthesised using ligand exchange reac-
Our synthesis strategy for the copper complexes was mainly
tions starting from [(PPh
pounds [(LOMe )NiBr
3
)
2
NiBr
2
]. The obtained dimeric com-
)NiBr and [(LOMe )NiBr
determined by the initial goal to generate mononuclear com-
pounds. For deprotonated LOH and related ligand types the for-
2
2
]
2
,
[(LOH
2
2
]
2
4
2 2
]
2
were rapidly formed but the isolated yields were low, because
OPPh is formed as a side product and has to be removed by recrys-
tallisation from CH Cl
mation of oligonuclear copper complexes with copper atoms
bridged by the negatively charged oxido ligand functions (pheno-
lates) was reported [15,16]. The oligonuclear products can be
transformed to mononuclear complexes by adding suitable ligands
like pyridine [18], however, this method is insufficiently controlla-
ble, due to the unknown nuclearity of the starting materials (stoi-
3
2
2
.
2.3. NMR spectroscopy
II
For the Cu complexes no NMR spectra could be obtained due to
II
chiometry) and the formation of by-products as [Cu(py)
2
Cl
2
]
.
their paramagnetism. Ni compounds are not necessarily diamag-
Therefore, we sought for more reliable strategies leading directly
to mononuclear complexes. Using ligands with protonated or
netic but if their geometry is square planar or distorted octahedral
they are suitable for NMR-analyses. Indeed, for the dimeric com-
1
methoxylated oxido functions LOH
2
, LOMe
2
, LOMe
4
(Scheme 1)
plexes [(O,N,O)NiBr ] , H NMR spectra were obtained and can be
2
2
1
3
formation of oligonuclear species through oxido-ligand bridging
is prohibited. The reaction of the three oxido-pincer ligands with
compared to spectra of the free ligands. C NMR spectra were re-
corded for the free ligands (see Section 4) while low solubility of
the nickel complexes did not allow such measurements.
CuCl
(LOH
2
in methanol yielded the brown compounds [(LOMe
2 2 2
)CuCl ] ,
[
2
)CuCl and [(LOMe )CuCl which were binuclear as can
2
]
2
4
2
]
2
From Table 2 it is evident that the proton signals of the ligands
show remarkable low field shifts upon coordination. Especially the
shifts of the pyridine ring protons indicate the metal ion coordina-
tion, while the phenol-protons are not extremely low field shifted.
The observed shifts for the methoxy and hydroxy proton signals
indicate that both oxido-donors are bound to the metal in any case.
The spectra were found to be unchanged after days, proving the
stability of the formed complexes in acetone solution.
be inferred from their colour and EPR spectra (see below). In many
reports binuclear complexes or compounds with even higher
nuclearity are described to exhibit brownish colours, while mono-
meric derivatives are green [18]. We assume that in our dimeric
products the copper atoms were octahedrally configured with
two chlorido bridging ligands (Scheme 2). These solids can be dis-
solved in DMF resulting reliably in octahedrally configured mono-
nuclear species [(O,N,O)CuCl
2
(DMF)] (Scheme 2) as indicated by
The LOH nickel complex still shows proton signals for two OH-
2
their absorption spectra and EPR spectroscopy (see below).
Also in acetonitrile (MeCN) mononuclear species were obtained,
however at the same time these species rapidly disproportionate
following Eq. (1) in line with recent reports on related chlorido-
bridged copper complexes [19] and with the previously reported
protons, revealing that the complex formation is not connected to
deprotonation of the coordinated OH-groups. However, the d value
has shifted from 9.88 to 11.67 ppm indicating that the protons are
far more acidic.
0
0
pentacoordinated complexes [(RR pydimH
2
)CuCl
2
]
(RR pydim-
2.4. EPR spectroscopy
H
2
= oxido pincer ligands based on 2,6-bis(hydroxymethyl)pyri-
dine) [12]
EPR spectra of all copper compounds were measured on solid
samples at ambient temperatures and 110 K as well as on glassy
frozen DMF solutions at 110 K (data listed in Table 3). The free li-
gands and nickel complexes were found to be diamagnetic. All ob-
2
þ
ꢃ2ꢀ
2
½ðO; N; OÞCuCl
2
ꢃ¡½ðO; N; OÞ Cuꢃ þ ½CuCl
4
ð1Þ
2
An alternative synthetic strategy was to perform ligand ex-
change reactions in acetonitrile starting from [Cu(MeCN)
4
](TFA)
2
(
TFA = trifluoroacetic acid) as a precursor. In the case of LOMe
2
Table 2
no reaction took place (see Section 4), while for LOMe
4
the com-
Selected 1H NMR data for the free oxido-pincer ligands and their MgII and NiII
a
plex [(LOMe
The LOH
CuCl and the ligand in methanol is completely protonated. It can
be easily deprotonated by adding an excess of BuOK or NEt
solution of the complex in methanol. The deprotonated species
4
)Cu(TFA)
2
] was obtained as a light green solid.
complexes.
2
ligand in the complex [(LOH
2
)CuCl , obtained from
2 2
]
Compound
LOMe
(LOMe
LOH
[(LOH
LOMe
[(LOMe
(LOMe
Py4 Py3,5 Phen6 Phen4 Phen5 Phen3 OMe/OH
2
t
2
7.84 7.84
8.52 8.30
8.00 7.72
8.00 8.00
7.69 7.78
8.63 8.29
8.94 8.55
7.97
8.04
7.69
7.86
8.01
6.87
7.00
7.39
7.59
7.35
7.32
7.15
7.35
7.05
7.00
6.68
6.93
7.04
7.07
7.22
7.00
7.00
6.68
8.01
8.18
3.92
4.10
9.88
11.67
3.91/3.87
4.19/3.98
4.40/4.05
3
to a
[
2
)NiBr
2
]
2
2
[
(LOH)CuCl]
2
precipitates immediately and can be easily isolated
2
)NiBr
2
]
2
by filtration.
4
II
4
)MgBr
)NiBr
2
]
In addition to Cu complexes, which were the main focus of this
[
4
2
]
2
II
study, some Ni derivatives were synthesised for the sake of com-
parison concerning the coordination geometry (detectable by UV–
Vis–NIR-, NMR- or EPR-spectroscopy) and the electrochemistry
a
Chemical shifts d in ppm, as measured in [D
6
]-acetone.
OR
RO
Cl
Cl
Cl
Cl
Cu
DMF
N
N
N
O
N
Cu
OR
Cl
OH
Cl
Cu
Cu
OR
Cl
RO
O
N
DMF
R
N
DMF
N
DMF
(A)
(B)
Scheme 2. Schematical representation of the proposed structures for the binuclear complexes [(O,N,O)Cu(
The monomer obtained for the complex containing the deprotonated ligand LOH is shown in (B).
l-Cl)
2
Cu(O,N,O)] and the monomers obtained in DMF solution (A).
ꢀ

A. Klein et al. / Inorganica Chimica Acta 363 (2010) 3282–3290
3285
Table 3
a
X-band EPR data of the oxido-pincer copper complexes.
Compound
g
av
g
g
A
k
Cu
D
g
Symmetryb
Solvent/T
k
?
[
[
[
[
[
[
[
[
[
(LOH
(LOMe
(LOMe
(LOH)CuCl]
(LOH )CuCl
(LOMe
(LOMe
2
)CuCl
2
]
2
2.178
2.170
2.158
2.181
2.164
2.166
2.163
2.156
2.146
2.327
2.230
2.333
2.346
2.331
2.336
2.296
2.313
2.326
2.104
2.140
2.070
2.099
2.081
2.081
2.097
2.078
2.056
0.223
0.090
0.263
0.351
0.250
0.255
0.199
0.235
0.270
OD
OD
OD
OD
OE or SP
OE or SP
OE or SP
OE or SP
OE or SP
Solid/298 K
Solid/298 K
Solid/298 K
Solid/110 K
DMF/110 K
DMF/110 K
DMF/110 K
DMF/110 K
Solid/110 K
2
)CuCl
)CuCl
2
]
]
2
4
2
2
2
2
2
(DMF)]
139 G
129 G
123 G
170 G
165 G
2
)CuCl
)CuCl
2
(DMF)]
(DMF)]
4
2
(LOH)CuCl(DMF)
(LOMe )Cu(TFA) ]
2
]
4
2
a
g
av = Averaged g value = (g þ 2g /3);
D
g ¼ gk ꢀ g
.
k
?
?
b
Symmetry assignment based on EPR spectroscopy (see text), OD = octahedral dimeric, OE = octahedral elongated, SP = square pyramidal.
II
served g values lie in the range expected for Cu complexes, while a
do not know if this species contains one or two DMF ligands, since
assuming a square planar arrangement of the O,N,O and the Cl
coligand, the presence of one or two DMF coligands in the axial
position will not have marked influence on the spectroscopy. For
close look reveals decent differences in the signal symmetry and
subtle variations in g values and g anisotropy (
D
g). The signal form
of Cu is a direct hint to the complex geometry [20]. The binuclear
products [(O,N,O)CuCl from the preparation in methanol exhibit
II
2
]
2
4 2
the complex [(LOMe )Cu(TFA) ] in the solid we found an axial
ill-resolved axial spectra (no hyperfine splitting) and no indication
of a half-field signal. Such spectra are typical for octahedrally con-
figured chloride-bridged binuclear species (OD) [19,21]. Similar
spectra have been observed also for chlorido-bridged binuclear
complexes with square pyramidal surrounding of the copper ion
spectrum (Fig. 2), very similar to those of the monomeric com-
plexes with chlorido ligands and we assume a similar monomeric
structure for this complex.
2.5. Luminescence properties
[
22,23], indicating a marginal influence of the sixth ligand [24]. A
markedly different g anisotropy ( g) is observed for the LOMe
D
2
To further substantiate the question of tight binding, rigidity or
complex in comparison to the other two complexes (Fig. 2), while
the signal symmetry is the same. In a number of related com-
pounds the g anisotropy reflects subtle distortions of the geometry
around the copper ion imposed by the crystal structure [19,21].
Since we could not obtain crystal structures of our complexes we
can only speculate, that the same mechanisms operate here. The
species observed in glassy frozen DMF solutions all exhibit axial
flexibility of the ligands, we examined the emission properties of
the new compounds. The ligand LOMe and its complexes show
blue luminescence upon irradiation into the long-wavelength
4
3
absorption band (assigned to evolve from a
p–p excited state) in
the solid and in DMF solution. Comparable emission has been re-
II
ported for the Zn complex [(LO)
Neither in the solid, nor in DMF solution LOH
and their complexes exhibit emission at ambient temperature.
We assume that the corresponding emission for the ligands LOH
and LOMe and their complexes is quenched by radiationless de-
cay, and we assume that the higher distortion (from steric strain)
in the free ligand LOMe and its complexes compared to LOMe
4
(Py)
4
Zn
4
] in the solid state [14].
2
or LOMe ligands
2
spectra with coupling constants (ACu) about 140 G for the g com-
k
ponent, which are typical for square-based pyramidal (SP), tetrag-
onally elongated octahedral (OE) or trigonal bipyramidally (TBP)
2
II
2
configured Cu complex [20,25–28] The three cases can be win-
nowed by their g value range. TBP compounds usually have g val-
k
4
2
ues around 2.0 and g around 2.2, while for the other two
?
and LOH
2
accounts for the different behaviour. The emission max-
configurations a smallest g value >2.04 can be expected [20].
Following this classification the complexes [(O,N,O)-
ꢀ1
ima (around 470 nm) and the Stokes shifts (around 5400 cm ) are
4
quite similar for free LOMe and its complexes, while the intensi-
CuCl
Scheme 2), although we cannot rule out, that the contribution of
the DMF ligand is marginal and the coordination is more of a
2
(DMF)] are mononuclear octahedrally configured complexes
ties of the emission and the quantum yields vary markedly (Ta-
ble 4; representative spectra are supplied in the Supporting
information). The quantum yield is higher for the magnesium com-
pound than for the nickel and copper derivative, reflecting the bet-
ter fit of the ion into the narrow binding pocket of the ligand (see
discussion on the molecular structures) thus providing a rigid sys-
tem (better u than the free ligand). For a more detailed picture of
the luminescence properties further experiments have to be car-
ried out in the future (low-temperature measurements and time-
(
square pyramid. The g anisotropy (
complexes. The complex containing the deprotonated LOH ligand
D
g) is quite similar for all three
ꢀ
[
2
(LOH)CuCl] exhibits very similar spectra, thus we conclude that
in the solid also a dimer is present. This would imply that the
deprotonated oxido function takes part in the bridging between
the two copper atoms. In DMF solution clear indication for a mono-
mer complex is provided by the obtained EPR spectra. However, we
2 2
resolved), focussing on the failure of LOH and LOMe and their
complexes to exhibit luminescence under the applied conditions.
2.6. Electrochemical measurements
The electrochemical measurements of the copper dichlorido
complexes were carried out in DMF as solvent (and nBu
electrolyte) in which they were in their mononuclear form
(O,N,O)CuCl (DMF)]. The compound [(LOMe )Cu(MeCN)](TFA)
2
4 6
NPF as
[
2
4
4
00 G
5
00 G
was measured in acetonitrile due to a better solubility. The nickel
complexes, which were measured for the sake of comparison to the
II
Cu derivatives had to be measured in THF solution since they are
Fig. 2. X-band EPR spectra solid samples of (a) [(LOMe
2
)CuCl
]
2 2
(solid line) and
] at 298 K
not completely stable in acetonitrile or DMF. The assignment of re-
[
(
(LOH
2
)CuCl
]
2 2
(dashed line) at 298 K (left) and (b) [(LOMe )Cu(TFA)
4
2
II
II
right).
dox waves is based on the assumption that Ni and Cu complexes

3
286
A. Klein et al. / Inorganica Chimica Acta 363 (2010) 3282–3290
Table 4
Absorption, excitation and emission data of LOMe
4
and the derived metal complexes.
Compound
k absorption maximuma
k excitation
maximum
k emission maximum
exc = 390 nm
Stokes shift
uc
a,b
ꢀ1
k
(cm
)
3
LOMe
4
270 (21260); 298(sh) (16150); 307(sh) (17110); 313
18160); 367 (1110)
316 (1530); 371 (550)
(DMF)] 318 (6600); 372 (3860)
361; 392
466
5137
0.21 ꢄ 10ꢀ
(
ꢀ
3
[
[
(LOMe
(LOMe
4
)MgBr
)NiBr
2
]
344(sh); 402
307(sh); 380(sh);
464
467
5449
5469
3.52 ꢄ 10
ꢀ
3
4
2
0.12 ꢄ 10
395
3
[
(LOMe
4
)CuCl
2
(DMF)] 316 (14820); 371(sh) (2290); 439(sh) (330)
346; 408
475
5901
0.53 ꢄ 10ꢀ
a
b
c
ꢀ1
ꢀ1
Absorption, excitation and emission maxima in nm, intensities (in parentheses) in L mol cm as measured in DMF solution.
Excitation spectra obtained for the maximum emission wavelength.
Quantum yield.
II
I
might exhibit similar ligand-centred oxidation or reductions,
while the metal-based electrochemistry should differ largely
type copper enzymes (e.g., ꢀ0.24 V for Cu /Cu in galactose oxi-
dase). Furthermore, the ligand-centred oxidation occurs presum-
ably in the coordinated phenol or phenolate, which corresponds,
to enzymes containing radical co-factors and therefore transfer-
ring more than one electron (e.g., galactose oxidase). Unfortu-
nately, the ligand oxidation processes occur irreversibly (even at
high scan rates) and occur at rather high potentials, far higher
II
II
I
since Cu might be easily reduced (Cu /Cu couple), whereas for
II
II
III
Ni oxidation (Ni /Ni ) should be observable at comparably
low potentials.
Indeed, the complexes under investigation show reversible
II
reduction waves at around 0 V for the Cu complexes, while corre-
II
Å+
sponding Ni complexes are oxidised at around +0.4 V (irrevers-
than natural co-factors (e.g., 0.01 V for [Tyr] in galactose oxi-
ibly). The irreversible ligand-centred oxidation occurs for both
systems (and the free ligands) at quite similar values (0.7–1 V) (Ta-
dase). Noteworthy, the complex containing the deprotonated
ligand LOH shows a relatively low potential of 0.28 V. The reason
ꢀ
ble 5) except for the chlorido copper complex [(LOH)CuCl(DMF)
2
]
for the very high ligand centred oxidation potentials in our com-
plexes is very probably that the ligand systems presented herein
do not have any radical-stabilising groups. Upon stabilising the
ꢀ
containing the deprotonated LOH ligand, for which the oxidation
occurs at far lower potential. Furthermore, irreversible reduction
waves were observed on cathodic scans, which were assigned to li-
Å+
radical species [L ] (usually by tert-butyl groups in 2 and 4 posi-
gand-centred processes. For the complex [(LOH
found waves on the reversed scan after reduction, which might
be assigned to the deprotonated complex [(LOH)CuCl(DMF) ] as re-
2
)CuCl
2
(DMF)] we
tion of the phenolate) the compounds would not only show a
reversible oxidation process, but would have also lower oxidation
potentials [29].
2
ferred from their potential. However, this would mean that the
complex is deprotonated upon reduction. This is rather unusual,
since normally one would expect a deprotonation after augmenta-
tion of positive charge (=oxidation). Essential redox potentials of
the different ligands and metal complexes are summarised in Ta-
ble 5 (representative CVs are supplied in the Supplementary
information).
2.6.1. Optical spectroscopy and spectroelectrochemical measurement
The complexes [(O,N,O)CuCl
field (d–d) transitions around 1000 nm (
2
(DMF)] show typical weak ligand
ꢀ
1
ꢀ1
e
ꢂ 100 L mol cm
)
indicative for a Jahn–Teller distorted square pyramidal (or octahe-
dral) structure (representative spectra and data are collected in the
Supporting information) [30]. The energy of the ligand field
absorptions for the Cu–OMe containing compounds lie at some-
what lower energy (1097 nm/1033 nm) compared to the Cu–OH
containing derivative (929 nm), reflecting a weaker ligand field im-
posed by the methoxy donor ligands compared to hydroxido. This
points to ligand distortion around the copper ion, since weaker
binding might be the result of a strained coordination geometry.
The ligands exhibit strong bands in the UV range of the spectrum.
The two bands discernible at around 270 and 310 nm are assigned
The essential electrochemical parameters for the question, if
the present complexes might serve as model compounds for
entatic copper enzymes are the reversibility and the potential of
II
I
the Cu /Cu reduction wave. These potentials are markedly lower
II
2+
4
for the Cu chlorido complexes than for [(LOMe )Cu(MeCN)] . At
II
I
the same time, the Cu /Cu potentials are high compared to wild-
Table 5
Electrochemical data of free oxido-pincer ligands and their CuII and NiII complexes.a
to ligand-centred
p–p* transitions and do not shift markedly upon
I
II
coordination (Table 6).
E
pa ox/ligand
E
½
Cu /Cu
Solvent
The first spectroelectrochemical experiments (using an optical
transparent thin-layer electrode (OTTLE) cell) were devoted to
demonstrate that the ligands tolerate the change in oxidation state
of the copper atom from Cu to Cu . When reducing the parent
complexes the blue colour of the ligand field transition vanishes,
while the ligand-centred absorption bands decrease only slightly
in intensity (5–30%) and the absorption maxima are virtually un-
changed. Only the complex with the deprotonated ligand [(LOH)-
Ligands
LOH
2
0.93
0.80
0.98
DMF
DMF
DMF
LOMe
LOMe
2
II
I
4
Cu chlorido complexes
LOH
LOH
2
1.01
0.28
0.90
0.72
ꢀ0.11
ꢀ0.03
ꢀ0.04
ꢀ0.01
DMF
DMF
DMF
DMF
ꢀ
LOMe
LOMe
2
4
2
CuCl(DMF) ] exhibits a slight bathochromic shift (Table 6, for
Cu(MeCN) complexes
LOMe
spectra see Supporting information). We thus conclude that coor-
4
1.54
0.21
MeCN
I
dination is retained upon reduction to Cu and the ligand easily fol-
Ni complexesb
LOH
E
pa Ni /Ni
II
III
lows the associated change of the coordination geometry. We
assume that this can happen through the deformations (mainly
rotation of the phenoxy substituent) discussed above.
2
0.82
>1.04
0.96
0.41
0.56
0.52
THF
THF
THF
LOMe
LOMe
2
4
a
The oxidation of the complexes occurs irreversibly as has been
shown by cyclic voltammetry. Nevertheless, we studied the spec-
troscopic response upon oxidation. We found for all complexes
4 6
From cyclic voltammetry in nBu NPF /solvent solutions; potentials in V vs.
þ
FeCp
2
/FeCp2
.
b
We assume that the nickel complexes are dimeric in THF solution.

A. Klein et al. / Inorganica Chimica Acta 363 (2010) 3282–3290
3287
Table 6
Data of absorption–titration experiments and spectroelectrochemical measurements in DMF, all bands in nm.
Titration
k (nm)
Equivalents
With CuCl
With CuCl
2
2
278, 318, 436a
278, 360
0–5
0–3
3
(nBu N as base)
Spectroelectrochemistry
Oxidation
Parent (CuII)
CuII ? CuI
Reduction (I)b
278, 310, 425
Reduction (II)b
278, 351sh, 408
[
[
[
[
(LOH)CuCl(DMF)
2
]
278, 322, 410sh
278, 316, 439
258, 324, 342sh
264, 321, 371
278, 343
278, 316, 436
258, 304
277, 358
278, 318
258, 306
264, 314
a
(LOH )CuCl (DMF)]
2
2
278, 350sh, 408, 535
(LOMe
(LOMe
2
)CuCl
)CuCl
2
(DMF)]
(DMF)]
4
2
264, 312
a
2ꢀ
4
Assigned to [CuCl ] .
For experimental conditions see text.
b
3
00
350
wavelength / nm
400
450
300
350
400
450
500
550
wavelength / nm
þ
Fig. 3. Spectroelectrochemical measurement of oxidation processes at +1.0 V (vs. FeCp
2
/FeCp2 ) in DMF/nBu
4
NPF
6
left: [(LOH)CuCl(DMF)
2 2 2
]; right: [(LOMe )CuCl (DMF)].
2
ꢀ
2ꢀ
Note, that the band at 436 nm, indicative for [CuCl
4
]
, also vanishes upon oxidation thus [CuCl
4
]
is oxidised under these conditions.
quite similar spectra mainly characterised by a band at around
20 nm (for all complexes) and further long-wavelength bands
varying from 340 to 440 nm, dependent on the ligand (Fig. 3). They
can be assigned to the generation of phenoxy radical species [31],
confirming our assignment of the waves around 1 V to ligand-
based oxidations.
change the spectrum, thus a fully deprotonated complex species,
2ꢀ
3
containing LO is not accessible using nBu
tion of the ligand LOH under the same conditions without base
leads to the formation of [(LOH )CuCl (DMF)], as indicated by an
identical spectrum of the isolated complex (Table 6).
To establish if the complex [(LOH )CuCl (DMF)] is deprotonated
3 a
N (pK = 10.9). Titra-
2
2
2
2
2
The LOH
base properties using absorption spectroscopy of the different
de)protonated species combined with calculations [13]. From this
study we know that a band at 318 nm belongs to the protonated
2
ligand had been already analysed concerning its acid–
upon reduction, as inferred from the CV experiments, we studied
the complex by optical spectroelectrochemistry. Under reducing
conditions (ꢀ2.7 V) the main absorption band of the complex
(
2 2
[(LOH )CuCl (DMF)] at 318 nm vanishes while a new band appears
LOH
2
ligand (
p
–
p
*). An absorption maximum of 349 nm indicates
at 425 nm (Fig. 4, right). Further reduction (ꢀ3.0 V) finally leads to
ꢀ
the singly deprotonated Ligand LOH , while a band at 408 nm de-
a strong band at 408 nm. While the 408 nm band is unequivocally
notes the completely deprotonated LO2 ligand. The absorption
ꢀ
indicative for the doubly deprotonated ligand, the band at 425 nm
ꢀ
spectrum of the [(LOH
the ligand remains protonated upon coordination. However, it is
reasonable to assume, that coordination of the LOH ligand will
strongly facilitate the deprotonation and we thus performed titra-
tion experiments (see Fig. 4, left) to investigate the pH-dependent
behaviour of the metal complexes [32].
2
2
)CuCl (DMF)] complex thus confirms that
is assigned to the species [(LOH)CuCl
2
(DMF)] . When studying the
complex carrying the deprotonated ligand [(LOH)CuCl(DMF)
observed the 408 nm band evolving immediately under reductive
electrolysis. It is important to note, that reduction of [(LOH )-
CuCl (DMF)] and subsequent deprotonation in first instance leads
2
to [(LOH)CuCl (DMF)] not to [(LOH)CuCl(DMF) ], therefore the
2
] we
2
2
2
ꢀ
2
In a first experiment 0.3 mL nBu
3
N (1.3 mmol) were added to
spectra are not identical (but very similar). Anyway our experi-
ments give strong evidence, that the ligand-centred reduction pro-
3
mL of a solution of the LOH ligand (0.6 mmol). Since no spectral
2
changes were observed, we conclude that the free ligand is not
deprotonated under these conditions. To this mixture small por-
2 2
cesses of the complex [(LOH )CuCl (DMF)] are strongly coupled to
deprotonation, which is unusual, as outlined in Section 2.6. This is
further supported by the reductive spectroelectrochemistry of
tions (10
absorption spectra showed a decreasing absorption at 318 nm
corresponding to the ligand LOH ), while a new absorption band
2
lL) of CuCl in DMF were added. The corresponding
2 2 4 2
[(LOMe )CuCl (DMF)] and [(LOMe )CuCl (DMF)] where we did
(
2
not observe any defined absorption bands upon reduction. A gen-
eral overview of the titration experiments and spectroelectro-
chemical measurements is presented in Table 6, further
representative figures are supplied in the Supporting information.
at 360 nm appeared (see Fig. 4, left). From the spectral fingerprint,
we can assign the reaction to a deprotonation of the coordinated
ꢀ
ligand (formation of LOH ). Further addition of base does not

3
288
A. Klein et al. / Inorganica Chimica Acta 363 (2010) 3282–3290
3
00
350
400
450
500
550
wavelength / nm
3
00
400
500
600
300
400
500
600
700
800
900
wavelength / nm
wavelength / nm
Fig. 4. Left: titration of a LOH
FeCp2 ), the inset shows further reduction at ꢀ2.9 V presumably leading to [(LO)CuCl]
2
/nBu
3
solution with CuCl
2
in DMF; right: spectroelectrochemical reduction of [(LOH
2
)CuCl
2
(DMF)] in DMF/nBu
4
NPF
6
at ꢀ2.7 V (vs. FeCp
2
/
þ
2ꢀ
.
3
. Conclusions
and HMBC experiments. All 2D NMR experiments were performed
using standard pulse sequences from the Bruker pulse program li-
brary. Chemical shifts were relative to TMS. UV–Vis–NIR absorp-
tion spectra were measured on Varian Cary50 Scan or Shimadzu
UV-3600 photo spectrometers. UV–Vis emission spectra were re-
corded with a Spex FlouroMax-3. Elemental analyses were carried
out using a HEKAtech CHNS EuroEA 3000 Analyzer. EPR spectra
were recorded in the X band on a Bruker System ELEXSYS 500E,
with a Bruker Variable Temperature Unit ER 4131VT. g values were
calibrated using a dpph sample. Electrochemical experiments were
A number of new binuclear CuII complexes containing bis-phe-
nol-pyridine O,N,O-pincer ligands were synthesised, which can be
reliably converted into mononuclear species [(O,N,O)CuCl (DMF)]
in DMF solution. These mononuclear compounds exhibit electro-
chemical behaviour comparable to natural oxidation catalysts as
2
II
could be established from comparison with corresponding Ni
II
I
complexes. The Cu /Cu transformation occurs fully reversible at
rather low potentials. Spectroelectrochemistry has allowed the
in situ observation of the associated geometrical transformation
of the coordination sphere around the copper ion. Furthermore,
all three ligands form phenoxyl radicals. Unfortunately, these oxi-
dations occur irreversibly at rather high potentials and were not
comparable to natural co-factors in this respect. From our detailed
titration experiments in combination with reductive spectroelect-
4 6
carried out in 0.1 M nBu NPF solutions using a three-electrode
configuration (glassy carbon electrode, Pt counter electrode, and
Ag/AgCl reference) and an Autolab PGSTAT30 potentiostat and
function generator. The ferrocene/ferrocenium couple (FeCp
2
/
þ
FeCp
) served as internal reference. UV–Vis spectroelectrochemi-
2
cal measurements were performed with an optical transparent
thin-layer electrode (OTTLE) cell [36].
rochemistry we have strong evidence that the system [(LOH
2
)-
CuCl (DMF)] exhibits coupled reduction/deprotonation
2
a
chemistry which would allow to generate deprotonated species
electrochemically. The electrochemical and spectroscopic behav-
iour (absorption and emission) is fully in line with the assumption
that the ligands offer a rather strained coordination geometry to
the metal ions, comparable to those found in copper enzymes
4
.3. X-ray crystallographic studies
Crystal structure determinations were performed at 293(2) K
using graphite-monochromatised Mo Ka radiation (k = 0.71073 Å)
on a IPDS II (STOE and Cie). The structures were solved by direct
methods (SHELXS-97) [37] and refined by full-matrix least-squares
techniques against F
corrections (X-RED V1.22; STOE and Cie, 2001) were performed
after optimising the crystal shapes using X-SHAPE V1.06 (STOE
and Cie, 1999) [39]. The non-hydrogen atoms were refined with
anisotropic displacement parameters without any constraints. All
H atoms of the free oxido-pincer ligands including the OH group
were found during the refinement process.
(
entatic state).
. Experimental
.1. General
,6-Dibromopyridine,
2
(SHELXS-97) [38]. The numerical absorption
4
4
2
1,3-dimethoxybenzene,
pyridinium
hydrochloride and urea-hydrogen peroxide (UHP) were purchased
from Aldrich. 2-Iodo-methoxybenzene was purchased from Alfa
2 3 2
Aesar. [Ni(dppe)Cl ] [33], [Ni(PPh )Br ] [34] were synthesised
4
4
1
.4. Synthesis of the oxido-pincer ligands
according to literature and 4-iodo-1,3-dimethoxybenzene in a
modified synthesis [35]. The Grignard reactions and the Kumada-
coupling reactions were carried out under inert gas conditions
and performed by using Schlenk techniques. THF was dried using
a MBRAUN MB SPS-800 solvent purification system.
.4.1. 4-Iodo-1,3-dimethoxybenzene
1
.38 g (10 mmol) 1,3-dimethoxybenzene were mixed with
.27 g (5 mmol) I (finely powdered) and 0.56 g (6 mmol) UHP
2
(
finely powdered). After exposing to ultrasound for 10 h, the
mixture was extracted with 100 mL methyl tbutyl ether (MTBE).
The organic phase was washed with an aqueous solution of Na SO
10%) and then with water. After drying the MTBE-phase using
4.2. Instrumentation
2
3
(
NMR spectra were recorded on a Bruker Avance II 300 MHz
4
MgSO , the solvent was removed under reduced pressure to yield
1
n
1
spectrometer, using a triple resonance H, BB inverse probe head.
2.62 g (99%) of a brown oil. H NMR (300 MHz, CDCl
3
): d = 7.60
1
13
The unambiguous assignment of the H and C resonances was
(d, 1H, 5-Phen), 6.42 (d, 1H, 2-Phen), 6.31 (dd, 1H, 6-Phen), 3.84
1
1
1
13
13
obtained from H NOESY, H COSY, gradient selected H, C HSQC
3
(s, 3H, 3-OMe), 3.78 (s, 3H, 1-OMe) ppm. C NMR (75 MHz, CDCl ):

A. Klein et al. / Inorganica Chimica Acta 363 (2010) 3282–3290
3289
d = 161 (1C, 1-Phen), 159 (1C, 3-Phen), 139 (1C, 5-Phen), 107 (1C,
-Phen), 99 (1C, 2-Phen), 75 (1C, 4-Phen), 56 (1C, 3-OMe), 55
1C, 1-OMe) ppm. C I (264.06): Anal. Calc. C, 36.39; H, 3.44.
Found: C, 36.30; H, 3.45%.
4.4.5. 2,6-Bis-(2,4-dimethoxyphenyl)pyridine
3.0 g (5.2 mmol, 1 eq) of [(LOMe )MgBr
ethyl acetate and an aqueous solution of 1.0 g Kryptofix- (2.2.2)
was added until all starting material has dissolved. Then the phases
were separated and the organic phase was subsequently washed
with two small portions of Kryptofix solution. After final phase
6
(
4
2
] were suspended in
Ò
H
8 9
O
2
4
.4.2. 2,6-Bis(2-methoxyphenyl)pyridine
1.7 g of 2-Iodo-methoxybenzene (50 mmol) were reacted with
.8 g (75 mmol) magnesium in THF to give the Grignard compo-
separation the organic phase was dried over anhydrous Na
2 4
SO .
1
After filtration the solvent was removed under vacuum leaving a
1
1
yellow-orange solid. Yield 1.57 g (87%). H NMR (300 MHz, [D
6
]-
nent. The resulting solution was added dropwise to a stirred solu-
tion of 2,6-dibromopyridine and [(dppe)NiCl ] (10 mol%) the
acetone): d = 8.01 (d, 2H, 6-Phen), 7.79 (d, 2H, 3,5-Py), 7.69 (t,
2
1
6
4
H, 4-Py), 6.68 (m, 4H, 5,6-Phen), 3.91 (s, 6H, 2-OMe), 3.87 (s,
reaction mixture was stirred for 12 h and the brown solution was
quenched with 150 mL half concentrated HCl. After phase separa-
tion, the aqueous phase was extracted with CH Cl . The volume of
2 2
13
H, 4-OMe) ppm. C NMR (75 MHz, [D
6
]-acetone): d = 163 (2C,
-Phen), 160 (2C, 2-Phen), 156 (2C, 2,6-Py), 135 (1C, 4-Py), 132
(
9
2C, 6-Phen), 135 (2C, 3,5-Py), 105 (2C, 5-Phen), 10 (2C, 1-Phen),
8 (2C, 3-Phen), 55 (4C, 2,4-OMe) ppm. C21 (351.41): Anal.
Calc. C, 71.78; H, 6.02; N, 3.99. Found: C, 71.63; H, 6.05; N, 4.00%.
the combined organic solutions was reduced and the product was
precipitated by adding saturated ammonium chloride solution. The
H21NO
4
1
obtained solid was pale yellow. Yield 5.47 g (78%). Mp = 130 °C. H
NMR (300 MHz, CDCl
3
): d = 7.93 (dd, 2H, 6-Phen), 7.76 (m, 3H,
4
4
.5. Synthesis of the oxido-pincer complexes
3
5
,4,5-Py), 7.36 (t, 2H, 4-Phen), 7.14 (d, 2H, 3-Phen), 7.07 (t, 2H,
-Phen), 3.88 (s, 6H, OMe) ppm.
13
3
C NMR (75 MHz, CDCl ):
.5.1. Tetrakis(acetonitrile)copper(II) bis(trifluoroacetate)
d = 158 (2C, 2-Phen), 155 (2C, 2,6-Py), 135 (1C, 4-Py), 132 (2C, 6-
Phen), 130 (2C, 4-Phen), 123 (2C, 3,5-Py), 121 (4C, 1,5-Phen,),
[Cu(MeCN)
4 2
](TFA)
2
.0 g (15 mmol) anhydrous CuCl was dissolved in 100 mL ace-
2
1
12 (2C, 3-Phen), 55 (2C, 2-OMe) ppm. C19
2
H17NO (291.34): Anal.
tonitrile and 20 mL trifluoroacetic acid and refluxed over night. The
remaining solution was evaporated to 50 mL and the product was
precipitated at ꢀ25 °C as a blue fine-crystalline solid. Yield 4.30 g
Calc. C, 78.33; H, 5.88; N, 4.81. Found: C, 78.33; H, 5.85; N, 4.82%.
4.4.3. 2,6-Bis(2-hydroxyphenyl)pyridine
12 6 4 4
(63%). C12H F N O Cu (453.81): Anal. Calc. C, 31.76; H, 2.67; N,
A mixture of 0.50 g (1.7 mmol) 2,6-bis(2-methoxyphenyl)pyri-
12.35. Found: C, 31.74; H, 2.66; N, 12.33%.
dine and 5.18 g (44 mmol) pyridinium hydrochloride was heated
up to 190 °C and was stirred for 1 h. The resulting yellow-green
solution was cooled down to room temperature, where the liquid
solidified. The solid was suspended in water using an ultrasonic
bath. The suspension was extracted several time using a total of
4.5.2. Dichlorido(2,6-bis(2-methoxyphenyl)pyridine)copper(II)
[(LOMe )CuCl ]
2
2 2
200 mg LOMe (0.69 mmol) and 92 mg (0.69 mmol) anhydrous
2
CuCl were separately dissolved in 5 mL methanol and the copper
2
4
00 mL CH
a saturated sodium carbonate solution and dried over anhydrous
Na SO . After filtration the solvent was removed under vacuum
to leave a beige solid. Yield 344 mg (76%). Mp = 139 °C. H NMR
300 MHz, CDCl ): d = 9.88 (s (br), 2H, OH), 8.00 (t, 1H, 4-Py),
.72 (d, 2H, 3,5-Py), 7.69 (d, 2H, 6-Phen), 7.35 (t, 2H, 4-Phen),
.05 (d, 2H, 3-Phen), 7.00 (t, 2H, 5-Phen) ppm. 13C NMR (75 MHz,
): d = 156 (2C, 2-Phen), 151 (2C, 2,6-Py), 140 (1C, 4-Py), 132
2C, 4-Phen), 128 (2C, 6-Phen), 121 (2C, 1-Phen), 120 (4C,
2
Cl
2
. The combined organic phases were washed with
solution was added slowly to the ligand solution. The reaction mix-
ture was stirred at room temperature over night. The solvent was
removed under vacuum and the remaining orange-brown solid
was washed with a small portion cold acetone and dried. Yield
210 mg (71%). C19H17NO CuCl (425.80): Anal. Calc. C, 53.60; H,
2
4
1
(
7
7
CDCl
(
3
2
2
4.02; N, 3.29. Found: C, 53.57; H, 4.00; N, 3.30%.
3
4
[
.5.3. Dichlorido(2,6-bis(2-hydroxyphenyl)pyridine)copper(II)
(LOH )CuCl
2 2
.20 g (0.76 mmol) LOH and 0.10 g anhydrous CuCl were dis-
2
2 2
]
3
2
,5Py,5-Phen), 118 (2C, 3-Phen) ppm. C17H13NO (263.30): Anal.
0
Calc. C, 77.55; H, 4.98; N, 5.32. Found: C, 77.54; H, 4.99; N, 5.31%.
solved separately in 5 mL methanol. Both solutions were combined
and stirred at room temperature for 12 h. After removing the sol-
vent under vacuum a black solid was obtained. Yield: 211 mg
4
[
.4.4. 2,6-Bis-(2,4-dimethoxyphenyl)pyridine magnesium bromide
(LOMe )MgBr
A Grignard reagent was prepared from 12.5 g (47 mmol) 4-
(
72%). C17
2 2
H13NO CuCl (397.75): Anal. Calc. C, 51.34; H, 3.29; N,
4
2
]
3
.52. Found: C, 51.28; H, 3.29; N, 3.51%.
iodo-1,3-dimethoxybenzene and 2.0 g (excess) magnesium in
THF. The Grignard-solution was added slowly to a solution of
4
.5.4. Chlorido(2-(2-hydroxidophenyl)-6-(hydroxyphenyl)pyridine)
copper(II) [(LOH)CuCl]
0 mg (0.13 mmol) of [(LOH
methanol and 0.5 mL NEt were added. A green-brown solid pre-
cipitated immediately and was filtered off. Yield 35 mg (77%).
CuCl (361.29): Anal. Calc. C, 56.52; H, 3.35; N, 3.88.
5
[
.57 g (23.5 mmol) 2,6-dibromo-pyridine and 0.97 g (8 mol%
(dppe)NiCl ] in dry THF at 0 °C. The reaction mixture was stirred
at 0 °C overnight. After 12 h the reaction was stopped by adding
50 mL of HCl/water (1:1) and the reaction product was precipi-
tated by adding 400 mL of CH Cl . The bright yellow solid was fil-
l
)
2
5
2 2
)CuCl ] were dissolved in 7 mL
2
3
1
17 2
C H12NO
Found: C, 56.52; H, 3.38; N, 3.89%.
2
2
tered off and washed with a small portion of cold acetone. The
product was dried at 60 °C and then stored in a brown glass vessel
to prevent the yellow solid from turning dark orange. Yield: 8.03 g
4.5.5. Dichlorido(2,6-bis-(2,4-dimethoxyphenyl)pyridine)copper(II)
1
(
6
98%). H NMR (300 MHz, [D ]-acetone): d = 8.63 (t, 1H, 4-Py), 8.29
[(LOMe
0.20 g (0.58 mmol) LOMe
drous CuCl were dissolved separately in 5 mL methanol. Both
4 2 2
)CuCl ]
(
d, 2H, 3,5-Py), 8.01 (d, 2H, 6-Phen), 6.93 (d, 2H, 5-Phen;), 6.87 (dd,
4
and 78 mg (0.58 mmol, 1 eq) anhy-
1
3
2
H, 3-Phen), 4.19 (s, 6H, 2-OMe), 3.88 (s, 6H, 4-OMe) ppm.
]-acetone): d = 167 (2C, 4-Phen), 161 (2C, 2-
Phen), 151 (2C, 2,6-Py), 146 (1C, 4-Py), 133 (2C, 3-Phen), 123 (2C,
,5-Py), 108 (2C, 6-Phen), 101 (2C, 1-Phen), 100 (2C, 5-Phen), 57
4C, 2,4-OMe) ppm. C21 MgBr (535.52): Anal. Calc. C,
7.10; H, 3.95; N, 2.62%. Found: C, 47.13; H, 4.02; N, 2.61%.
C
2
NMR (75 MHz, [D
6
solutions were combined and stirred at room temperature for
2 days. Removing the solvent from the mixture yielded a brown so-
lid, which was washed with a small amount of cold acetone and
3
(
4
H21NO
4
2
dried on air. Yield (190 mg, 60%). C42
42 2 8 2 4
H N O Cu Cl (971.71): Anal.
Calc. C, 51.91; H, 4.36; N, 2.88. Found: C, 51.89; H, 4.33; N, 2.88%.

3
290
A. Klein et al. / Inorganica Chimica Acta 363 (2010) 3282–3290
4
.5.6. Bis(trifluoracetato)(2,6-bis-(2,4-dimethoxyphenyl)pyridine)
copper(II) [(LOMe )Cu(TFA)
79 mg (0.29 mmol, 1.5 eq) [Cu(MeCN)
0.19 mmol, 1 eq) [(LOMe )MgBr ] were mixed as solids and dis-
solved in 15 mL acetonitrile. The green solution was stirred for
h at room temperature then the acetonitrile was removed under
vacuum leaving a dark green solid. Yield 103 mg (84%). C25
CuF (641.00): Anal. Calc. C, 46.84; H, 3.30; N, 2.19. Found: C, 46.82;
H, 3.33; N, 2.20%.
ments. Additional figures show the
p-stacking in the crystal
4
2
]
structure of LOMe , representative cyclic voltammogramms, emis-
4
1
4
](TFA)
2
and 100 mg
sion spectra and UV–Vis-spectroelectrochemical experiments)
associated with this article can be found, in the online version, at
doi:10.1016/j.ica.2010.06.011.
(
4
2
6
H21NO8-
References
6
[
1] (a) M.R. Redinbo, D. Cascio, M.K. Choukair, D. Rice, S. Merchant, T.O. Yeates,
Biochemistry 32 (1993) 10560;
(
b) R. Huber, Angew. Chem. 101 (1989) 849;
4
[
.5.7. Dibromido(2,6-bis(2-methoxyphenyl)pyridine)nickel(II)
(LOMe )NiBr
5 mg (0.29 mmol, 1 eq) of LOMe
methanol. A methanolic solution of 215 mg (0.29 mmol, 1 eq)
(PPh NiBr ] (7 mL) was added in one portion and the resulting
Angew. Chem., Int. Ed. 28 (1989) 848.
2] N. Ito, S.E.V. Phillips, C. Stevens, Z.B. Ogel, M.J. McPherson, J.N. Keen, K.D.S.
Yadav, P.F. Knowles, Nature 350 (1991) 87.
[
[
2
2 2
]
8
2
were dissolved in 7 mL
3] E.I. Solomon, U.M. Sundaram, T.E. Machonkin, Chem. Rev. 94 (1994) 737.
[4] (a) G.C.M. Steffens, R. Bielwald, G. Buse, Eur. J. Biochem. 164 (1987) 295;
b) G.C.M. Steffens, T. Soulimane, G. Wolff, G. Buse, Eur. J. Biochem. 213 (1993)
149.
(
1
[
3
)
2
2
mixture was stirred for 6 h. The solvent was removed under vac-
uum and the remaining turquoise solid was washed with small
portions of cold acetone and dried, Yield 98 mg (45%). 1H NMR
[
5] (a) F. Himo, L.A. Eriksson, F. Maseras, P.E.M. Siegbahn, J. Am. Chem. Soc. 122
(2000) 8031;
(b) J.W. Whittaker, Chem. Rev. 103 (2003) 2347.
6] (a) A.G. Sykes, Adv. Inorg. Chem. 36 (1991) 377;
[
(
300 MHz, [D
Py), 8.04 (d, 2H, 6-Phen), 7.59 (t, 2H, 4-Phen), 7.35 (t, 2H, 5-Phen),
.22 (d, 2H, 3-Phen), 4.10 (s (br), 6H, OMe) ppm. C19 NiBr
509.86): calcd. C 44.76, H 3.36, N 2.75; found: C 44.75, H 3.35, N
.74.
6
]-acetone): d = 8.52 (t, 1H, 4-Py), 8.30 (d, 2H, 3,5-
(
b) H.E.M. Christensen, L.S. Conrad, K.V. Mikkelsen, M.K. Nielsen, J. Ulstrup,
Inorg. Chem. 29 (1990) 2808.
[7] W. Kaim, Dalton Trans. (2003) 761.
7
(
2
H17NO
2
2
[
[
8] N.G. Connelly, W.E. Geiger, Chem. Rev. 96 (1996) 877.
9] (a) B.L. Vallee, R.J.P. Williams, Proc. Natl. Acad. Sci. USA 59 (1968) 498;
R.J.P. Williams, J. Mol. Catal. (Review issue) (1986) 1.
[
10] (a) K.M. Merz, R. Hoffmann, Inorg. Chem. 27 (1988) 2120;
b) M. Hakansson, S. Jagner, E. Clot, O. Eisenstein, Inorg. Chem. 31 (1992) 5389.
(
4
[
.5.8. Dibromido(2,6-bis(2-hydroxyphenyl)pyridine)nickel(II)
(LOH )NiBr
5 mg (0.30 mmol, 1 eq) of LOH
anol. A methanolic solution of 223 mg (0.30 mmol, 1 eq)
(PPh NiBr ] (7 mL) was added in one portion and the resulting
[
[
11] K.D. Karlin, Y. Gultneh, Prog. Inorg. Chem. 35 (1987) 219.
12] A. Klein, S. Elmas, K. Butsch, Eur. J. Inorg. Chem. (2009) 2271.
2
2 2
]
7
2
were dissolved in 7 mL meth-
[13] S. Steinhauser, U. Heinz, J. Sander, K. Hegetschweiler, Z. Anorg. Allg. Chem. 630
2004) 1829.
14] W. Kaim, A. Klein, Spectroelectrochemistry, RCS Publishing, Cambridge, UK,
008.
[15] H.-Y. Zhang, K.-Q. Ye, J.-Y. Zhang, Y. Liu, Y. Wang, Inorg. Chem. 45 (2006) 1745.
(
[
[
3
)
2
2
2
mixture was stirred for 6 h. The solvent was removed under vac-
uum and the remaining green solid was washed with small por-
[
16] A.M.S. Silva, L.M.P.M. Almeida, J.A.S. Cavaleiro, C. Foces-Foces, A.L. Llamas-Saiz,
C. Fontenas, N. Jagerovic, J. Elguero, Tetrahedron 53 (1997) 11645.
1
tions of cold acetone and then dried. Yield 106 mg (47%).
NMR (300 MHz, [D
3
2
4
H
[
17] Although the goodness of fit is not very good there is no doubt on the applied
space group P2 2 2 (concluded from its typical extinctions). The Flack
6
]-acetone): d = 11.67 (s (br), 2H, OH), 8.00 (m,
H, 3,4,5-Py), 7.86 (dd, 2H, 6-Phen), 7.32 (t, 2H, 4-Phen), 7.00 (m,
H, 3,5-Phen) ppm. NiBr (481.81): Anal. Calc. C,
2.38; H, 2.72; N, 2.91. Found: C, 42.36; H, 2.75; N, 2.88%.
1
1 1
parameter is not relevant in our case; compare also H.D. Flack, G. Bernadinelli,
J. Appl. Cryst. 33 (2000) 1143.
18] E. Ludwig, U. Schilde, E. Uhlemann, H. Hartl, I. Briidgam, Z. Anorg. Allg. Chem.
C
17
H13NO
2
2
[
[
622 (1996) 701.
19] I.A. Koval, M. Sgobba, M. Huisman, M. Lüken, E. Saint-Aman, P. Gamez, B.
Krebs, J. Reedijk, Inorg. Chim. Acta 359 (2006) 4071.
20] B.J. Hathaway, D.E. Billing, Coord. Chem. Rev. 5 (1970) 143.
21] P. Kapoor, A. Pathak, R. Kapoor, P. Venugopalan, M. Corbella, M. Rodríguez, J.
Robles, A. Llobet, Inorg. Chem. 41 (2002) 6153.
22] S. Thakurta, P. Roy, G. Rosair, C.J. Gómez-García, E. Garribba, S. Mitra,
Polyhedron 28 (2009) 695.
23] F. Yraola, F. Albericio, M. Corbella, M. Royo, Inorg. Chim. Acta 361 (2008) 2455.
4
[
.5.9. Dibromido(2,6-bis-(2,4-dimethoxyphenyl)pyridine)nickel(II)
(LOMe )NiBr
00 mg (0.19 mmol, 1 eq) of [(LOMe
mL methanol. A methanolic solution of 215 mg (0.29 mmol,
.5 eq) [(PPh NiBr ] (7 mL) was added in one portion and the
[
[
4
2 2
]
1
4 2
)MgBr ] were dissolved in
[
[
7
1
3
)
2
2
resulting mixture was stirred for 6 h. The solvent was removed un-
der vacuum and the remaining yellow-green solid was washed
[24] A. Bencini, D. Gatteschi, Inorg. Chim. Acta 31 (1978) 11.
[
[
25] D. Reinen, C. Friebel, Inorg. Chem. 23 (1984) 791.
26] (a) N. Wei, N.N. Murthy, K.D. Karlin, Inorg. Chem. 33 (1994) 6093;
G. Kokoszka, K.D. Karlin, F. Padula, J. Baranowski, C. Goldstein, Inorg. Chem.
(1984) 4378.
with small portions of cold acetone and subsequently dried. Yield
1
8
3 mg (79%). H NMR (300 MHz, [D
6
]-acetone): d = 8.94 (t, 1H, 4-
[
[
27] A.W. Addison, H.M.J. Hendriksen, J. Reedijk, L.K. Thompson, Inorg. Chem. 20
Py), 8.55 (d, 2H, 3,5-Py), 8.18 (d, 2H, 6-Phen), 7.04 (d, 2H, 5-Phen),
(
1981) 103.
7
C
.00 (d, 2H, 6-Phen), 4.40 (s, 6H, 2-OMe), 4.05 (s, 6H, 4-OMe) ppm.
NiBr (569.92): Anal. Calc. C, 44.26; H, 3.71; N, 2.46.
28] T. Glowiak, I. Podgorska, Inorg. Chim. Acta 125 (1986) 83.
21
H21NO
4
2
[29] Y. Wang, T.D.P. Stack, J. Am. Chem. Soc. 118 (1996) 13097.
[30] C. Furlani, Coord. Chem. Rev. 3 (1968) 141.
Found: C, 44.26; H, 3.70; N, 2.48%.
[
31] M.R. Ganapathi, R. Hermann, S. Naumov, Ortwin Brede, Phys. Chem. Chem.
Phys. 2 (2000) 4947.
Acknowledgments
[
32] (a) A. Janscó, Z. Paksi, N. Jakab, B. Gyurcsik, A. Rockenbauer, T. Gajda, Dalton
Trans. (2005) 3187;
Dr. I. Pantenburg and I. Müller from the University of Cologne
are thanked for the X-ray measurements. A. Kaiser is acknowl-
edged for the synthesis of the nickel catalyst for the C–C-coupling.
We also would like to thank the ‘‘Studienstiftung des Deutschen
Volkes” for financial support.
(b) Z. Paksi, A. Janscó, F. Pacello, N. Nagy, A. Battistone, T. Gajda, J. Inorg.
Biochem. 102 (2008) 1700;
(
c) A. Pérez-Cadenas, L. Godino-Salido, R. López-Garzón, P. Arranz-Mascarós, D.
Gutiérrez-Valero, R. Cuesta-Martos, Transition Met. Chem. 26 (2001) 581.
[33] K. Matsumoto, N. Kotoku, T. Shizuka, R. Tanaka, S. Okeya, Inorg. Chim. Acta 321
2001) 167.
(
[
[
[
[
34] J. Chatt, B.L. Shaw, J. Chem. Soc. (1960) 1718.
35] J. Pavlinac, M. Zupan, S. Stavber, Org. Biomol. Chem. 5 (2007) 699.
36] M. Krej cˇ ík, M. Da nˇ ek, F. Hartl, J. Electroanal. Chem. 317 (1991) 179.
37] G.M. Sheldrick, SHELXS-97: A Program for Crystal Structure Solving, University
of Göttingen, 1997.
Appendix A. Supplementary material
CCDC 751119 and 751120 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 (crys-
[38] G.M. Sheldrick, SHELXL-97: A Program for the Refinement of Crystal Structures,
University of Göttingen, 1997.
[
39] (a) STOE X-RED, Data Reduction Program, Version 1.22/Windows, STOE and
Cie, Darmstadt, 2001.;
(b) STOE X-SHAPE, Crystal Optimisation for Numerical Absorption Correction,
Version 1.06/Windows, STOE and Cie, Darmstadt, 1999.
2 4
tal data on LOMe and LOMe and a table on absorption measure-