Copper(II) and zinc(ii) complexes with a photoactive bridging ligand based on carbazole: Synthesis, structures, electronic and magnetic properties

Article abstract of DOI:10.1002/ejic.201000970

The synthesis of the new Schiff base ligand N-(4-methoxyphenyl)-3,4- bis(salicylidenimino)carbazole (H₂SalCarbOMe) is described. It readily reacts with copper(II) andzinc(II) ions to form metallacycles of the general formula [M₂(SalCarbOMe)₂] (M = Cu, Zn). Their structures are determined by X-ray crystallography and show significant differences for the coordination at the metal ion with a nearly tetrahedral arrangement for Zn^II, but a considerable distortion towards square-planar geometry for Cu^II. This in turn leads to substantial differences for the dihedral angles observed between the salicylidene and carbazole moieties. As a consequence different arrangements in the unit cell are observed, which becomes particularly obvious when the additional solvent molecules of crystallization are considered. For the Zn^II compound this leads to the incorporation of a chloroform and a methanol molecule in fixed positions, whereas in the structure of the Cu^II compound a higher number of solvent molecules (chloroform, methanol, and water) in highly disordered crystallographic positions is observed. The optical and magnetic properties have been characterized for both complexes. They both show two band fluorescence in the visible region with emission maxima at 531 and 586 nm for the Zn^II compound and 488 and 580 nm in the Cu^II complex. The electronic properties have been further elucidated by quantum chemical calculations. For the two copper(II) ions an antiferromagnetic interaction through the carbazole bridging ligand with an exchange coupling constant of J = -3.1 cm^-1 is observed. ESR spectroscopy reveals a rhombic signal with g values [g_x, g_y, g_z] = [2.057, 2.125, 2.200] consistent with the distorted coordination geometry around the copper(II) ions.

Full text of DOI:10.1002/ejic.201000970

FULL PAPER
DOI: 10.1002/ejic.201000970
Copper(II) and Zinc(II) Complexes with a Photoactive Bridging Ligand Based
on Carbazole: Synthesis, Structures, Electronic and Magnetic Properties
Eike T. Spielberg[a] and Winfried Plass*[a]
Dedicated to Professor Ekkehardt Fluck on the occasion of his 80th birthday
Keywords: Copper / Zinc / EPR spectroscopy / Fluorescence / Magnetic properties / UV/Vis spectroscopy / Density
functional calculations
The synthesis of the new Schiff base ligand N-(4-meth-
oxyphenyl)-3,4-bis(salicylidenimino)carbazole (H2SalCarb-
OMe) is described. It readily reacts with copper(II) and
zinc(II) ions to form metallacycles of the general formula
[M2(SalCarbOMe)2] (M = Cu, Zn). Their structures are deter-
mined by X-ray crystallography and show significant differ-
ences for the coordination at the metal ion with a nearly tet-
rahedral arrangement for ZnII, but a considerable distortion
towards square-planar geometry for CuII. This in turn leads
to substantial differences for the dihedral angles observed
between the salicylidene and carbazole moieties. As a conse-
quence different arrangements in the unit cell are observed,
which becomes particularly obvious when the additional sol-
vent molecules of crystallization are considered. For the ZnII
compound this leads to the incorporation of a chloroform and
a methanol molecule in fixed positions, whereas in the struc-
ture of the CuII compound a higher number of solvent mole-
cules (chloroform, methanol, and water) in highly disordered
crystallographic positions is observed. The optical and mag-
netic properties have been characterized for both complexes.
They both show two band fluorescence in the visible region
with emission maxima at 531 and 586 nm for the ZnII com-
pound and 488 and 580 nm in the CuII complex. The elec-
tronic properties have been further elucidated by quantum
chemical calculations. For the two copper(II) ions an antifer-
romagnetic interaction through the carbazole bridging li-
gand with an exchange coupling constant of J = –3.1 cm–1 is
observed. ESR spectroscopy reveals a rhombic signal with g
values [gx, gy, gz] = [2.057, 2.125, 2.200] consistent with the
distorted coordination geometry around the copper(II) ions.
mophores is known and most have been characterized ex-
tensively as far as the nature of transitions and their other
optical properties are concerned. It is therefore desirable to
combine magnetic ions with photoactive ligands.
Introduction
During recent years a lot of effort has been spent on the
development of new magnetic materials.[1] The combination
of magnetic characteristics with other properties, for exam-
ple luminescence, promises new materials which may open
up new fields of applications.[2] Metal compounds are ideal
candidates for use as magnetic materials as metal ions with
unpaired electrons are good carriers for magnetic moments
because of their rather large momentum assembled in a d-
or f-shell, the stability of the open shell compounds, and
the possibility to tune their properties through the choice
of the metal ion and the design of the coordination
sphere.[3] In addition, optical properties can be tuned in a
very distinct way in organic compounds.[4] The influence
of several substituents on the absorbance and emission of
chromophores is well established. A vast number of chro-
Carbazoles are well known building blocks in organic
photo- and electrochemistry.[5,6] N-aryl-substituted carb-
azoles constitute a subclass of triphenylamines, which have
been extensively studied. The design of organic light-emit-
ting diodes,[7] photovoltaic elements,[8] and other organic
electronic devices[6,9] are only three fields of interest for this
type of molecule. In addition, aryl-substituted carbazoles
have been studied due to their unusual fluorescence behav-
ior, which has been discussed in terms of twisted intramo-
lecular charge transfer[10] and recently by other mecha-
nisms.[11]
We report here the synthesis of an N-aryl-substituted
carbazole bridging ligand containing two Schiff base-de-
rived metal ion binding pockets. This ligand is utilized to
generate a copper(II) complex combining magnetic interac-
tions and the photophysical properties of carbazole deriva-
tives. As zinc(II) ions are known to give very similar com-
plexes in many cases, the corresponding zinc(II) complex
has been synthesized as diamagnetic comparison, which ex-
cludes magnetic interactions.
[a] Institut für Anorganische und Analytische Chemie, Friedrich-
Schiller-Universität Jena,
Humboldtstr. 8, 07743 Jena, Germany
Fax: +49-3641-948132
E-mail: sekr.plass@uni-jena.de
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201000970.
826
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 826–834

CuII and ZnII Complexes with a Photoactive Bridging Ligand
plexes (see Supporting Information) the equilibrium con-
stants of these two reactions can be fitted to the experimen-
tal spectra giving the equilibrium constant Ktot for the total
reaction; see Equation (1).
Results and Discussion
Synthesis and Complex Formation
The amine precursor 3,4-diamino-N-(4-methoxyphenyl)-
carbazole (1) was synthesized by Ullmann-type coupling of
carbazole with p-bromoanisole according a variation of the
synthetic procedure described by Zhang et al.[12] and subse-
quent introduction of the amino groups by nitration and
reduction with elemental hydrogen. The reaction of 1 with
two equivalents of salicylaldehyde in methanol gives the
Schiff base ligand H2SalCarbOMe (2, Scheme 1).
1/2
Ktot = KIKII
(1)
For the titration with copper acetate the best fit was ob-
tained with KI = 1.14ϫ107 m–1 and KII = 9793 m–1, giving
Ktot = 1.13ϫ109 m–1. As only two species are detected in
the spectra, KI and KII are strongly correlated. However,
the value of Ktot remains constant within the expected toler-
ance (see Supporting Information). The large value of Ktot
confirms the high overall stability of the dinuclear complex
3.
Although the spectra for the formation of 4 look very
similar (see Figure S1 in the Supporting Information)
plausible results could not be obtained from the simulation
of the spectroscopic data. This is most probably due to nu-
merical problems within the fitting procedure.
Structural Characterization
The crystal structures of 3 and 4 were determined by X-
ray crystallography. Both complexes crystallize in the space
group P1 with two independent centrosymmetric dinuclear
molecules in the unit cell. As these two independent com-
plex molecules show similar structural features for both
Scheme 1. Synthesis of 2: (a) Cu(NO3)2, AcOH, Ac2O; H2, 55 bar,
Raney nickel, AcOEt, 80 °C, 4 h. (b) Salicylaldehyde, MeOH, re-
flux, 10 min.
¯
Ligand 2 reacts readily with copper(II) acetate and
zinc(II) acetate in a solvent mixture of methanol/chloro-
form (1:1) to form the dinuclear complexes [Cu2(SalCarb-
OMe)2] (3) and [Zn2(SalCarbOMe)2] (4), respectively. Both
complexes can be isolated as crystalline solids that contain
cocrystallized solvent molecules. In 4 thermogravimetric
and elemental analyses indicate the presence of one meth-
anol and one chloroform molecule per dinuclear complex
which are tightly bound. Whereas for 3 the solvent mole-
cules were found to be only loosely bound leading to the
decomposition of the crystals outside the mother liquor due
to the loss of cocrystallized solvent molecules.
The formation of the complexes was further investigated
by UV/Vis titrations of the ligand vs. copper and zinc acet-
ate solutions. To a solution of 2 in methanol/chloroform
(1:1) (copper: c0,L
=
3.65ϫ10–5 m; zinc: c0,L
=
3.08ϫ10–5 m) was added stepwise a solution of copper acet-
ate (c0,Cu = 3.68ϫ10–3 m, 99 mL per step) or zinc acetate
(c0,Zn = 3.68ϫ10–3 m, 85 mL per step). The spectra ob-
tained are depicted in the Supporting Information (Figure
S1). In both cases the two absorption bands of 2 at 358
and 380 nm (vide infra) decrease, while the bands of the
corresponding metal complexes steadily increase giving rise
to isosbestic points at 404 and 315 nm for copper acetate
and 400 and 453 nm for zinc acetate. Their presence indi-
cates a first order reaction. Hence there is direct formation
of the dinuclear complexes without intermediates at a UV/
Vis detectable concentration. The spectra were analyzed
with SpecReg using nonlinear regression methods. Since a
four component reaction is statistically improbable, the pro-
gram only regards two compound reactions. Assuming a
Figure 1. Molecular structure and numbering scheme for one of
the independent complex units of 3. Views perpendicular (top) and
parallel (bottom) to the bridging carbazole. Hydrogen atoms are
omitted for clarity, thermal ellipsoids are drawn at the 30% prob-
two-step mechanism for the formation of the metal com- ability level.
Eur. J. Inorg. Chem. 2011, 826–834
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
827

E. T. Spielberg, W. Plass
FULL PAPER
compounds, only one is depicted for 3 and 4 in Figures 1 Table 2. Selected bond lengths [pm] and angles [°] for 4.
and 2, respectively. Selected bond lengths and angles for 3
i = 1
i = 2
and 4 are given in Tables 1 and 2, respectively.
Zni–O1i
Zni–O2iA
Zni–N1i
Zni–N2iA
O1i–Zni–O2iA
O1i–Zni–N1i
O1i–Zni–N2iA
O2iA–Zni–N1i
O2iA–Zni–N2iA
N1i–Zni–N2iA
190.6(2)
189.1(2)
201.1(3)
200.4(2)
119.43(9)
96.71(9)
122.72(9)
119.73(10)
97.57(9)
100.75(9)
188.2(2)
189.9(2)
201.6(2)
201.6(3)
122.44(10)
97.59(10)
121.20(11)
120.50(10)
97.03(10)
96.69(10)
by the two N–M–N moieties (Mi, N1i, N2iA, MiA, N1iA,
N2i) of the metallacycle only slightly deviate from an or-
thogonal orientation with respect to the planes of the
carbazole bridging ligands, leading to dihedral angles of 83
and 87° for 4 and 75 and 80° for 3. Although the molecular
structures of 3 and 4 are quite similar, this indicates some
differences due to the nature of the coordinated metal ions,
which might be attributed to their characteristic electronic
structures. For a d9 copper system a square planar arrange-
ment should be favored, whereas for the d10 zinc system a
tetrahedral coordination geometry can be expected. This is
consistent with the bond angles observed at the metal cen-
ters of 3 and 4 (Tables 1 and 2).
Moreover, the characteristic differences in coordination
geometry for the metal centers in 3 and 4 are also evident
from the dihedral angles between the mean planes of the
two coordinating salicylidene moieties, which in 3 deviate
considerably from both ideal tetrahedral and square planar
arrangements (Table 3). The related geometrical strain is
also reflected in the large deviation of the copper ions from
the respective mean planes.
Figure 2. Molecular structure and numbering scheme for one of
the independent complex units in 4. Views perpendicular (top) and
parallel (bottom) to the bridging carbazole. Hydrogen atoms are
omitted for clarity, thermal ellipsoids are drawn at the 50% prob-
ability level.
Table 3. Dihedral angles [°] between mean planes defined by the
salicylidene (Sal1i and Sal2i)[a] and the cabazole units (Carbi)[b] as
well as for the tilting between the two benzene moieties (Ar1i and
Ar2i)[c] within the carbazole units and the deviation d(Mi) [pm] for
the metal centers from the salicylidene mean planes for the dinu-
clear complex units in 3 and 4.
Table 1. Selected bond lengths [pm] and angles [°] for 3.
i = 1
i = 2
Cui–O1i
Cui–O2iA
Cui–N1i
Cui–N2iA
O1i–Cui–O2iA
O1i–Cui–N1i
O1i–Cui–N2iA
O2iA–Cui–N1i
O2iA–Cui–N2iA
N1i–Cui–N2iA
188.2(4)
189.5(4)
194.2(5)
193.9(5)
93.0(2)
189.9(5)
189.9(4)
195.0(6)
195.2(5)
97.3(2)
M = Cu
M = Zn
i = 1
i = 2
i = 1
i = 2
ЄSal1i/Sal2i
47
34
59
46
38
6
50
34
51
47
28
6
78
6
13
57
46
2
74
7
17
50
42
1
94.3(2)
94.0(2)
d(Mi) form Sal1i
d(Mi) form Sal2i
ЄSal1i/Carbi
ЄSal2i/Carbi
ЄAr1i/Ar2i
148.2(2)
145.8(2)
93.6(2)
144.9(2)
142.6(2)
94.4(2)
97.5(2)
96.4(2)
[a] Salicylidene mean planes are defined by the following atoms.
Sal1i: O1i, N1i, C1i, C2i, C3i, C4i, C5i, C6i, and C7i; Sal2i: O2i,
N2i, C20i, C21i, C22i, C23i, C24i, C25i, and C26i. [b] Carbazole
mean planes are defined by the following atoms. Carbi: C8i, C9i,
C10i, C11i, C12i, C13i, C14i, C15i, C16i, C17i, C18i, and C19i. [c]
Mean planes of the benzene rings of the Carbazoles are defined by
the following atoms. Ar1i: C8i, C9i, C10i, C11i, C12i, and C13i;
Ar2i: C14i, C15i, C16i, C17i, C18i, and C19i.
In both 3 and 4 the metal ions are tetracoordinated by a
[N2O2] donor set comprising the phenolate oxygen and im-
ine nitrogen atoms of the salicylidene moiety of two bridg-
ing carbazole ligands. This leads to the formation of binu-
clear complex units [M2(SalCarbOMe)2] with coplanar car-
bazole moieties at distances of 313–322 pm. The resulting
intramolecular metal–metal distances within the metalla-
cycles, defined by the carbazole bridging ligand connected
The shape of the coordination polyhedra of the metal
by two N–M–N moieties, are about 1070 pm (Cu1: 1072; centers in 3 and 4 can be analyzed utilizing continuous
Cu2: 1074; Zn1: 1067; Zn2: 1074 pm). The plane defined shape measures (CSM) introduced by Anvir et al.,[13] which
828
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 826–834

CuII and ZnII Complexes with a Photoactive Bridging Ligand
gives a quantitative measure for the distortion of a given chloroform molecules are located near the phenolate oxy-
coordination sphere from tetrahedral and square planar by gen atoms O11 and O22, respectively (Figure S4, Support-
the corresponding shape measures, S(Td) and S(D4h). These ing Information). These crystallographic positions are fully
values give a measure for the difference from ideal geome- occupied and no sign of disorder is observed, which is con-
try, where zero indicates ideal geometry. Data is given in sistent with the solvent content in the isolated bulk material
Table 4. For 3 these values indicate a coordination geometry (vide supra). Data is given in Table 5. One dinuclear com-
about halfway between tetrahedral and square planar, plex molecule is solely hydrogen-bonded to two methanol
whereas for 4 a distorted tetrahedral geometry is observed. molecules, whereas the other is hydrogen-bonded to two
chloroform molecules. Consequently, the crystal structure
Table 4. CSM parameters for the coordination geometries of 3 and
can be described as 2D layered networks of [Zn2(SalCarb-
4 (for notation see text).
OMe)2]·2CHCl3 and [Zn2(SalCarbOMe)2]·2MeOH units
with no solvent molecules located between the layers.
M = Cu
i = 1
M = Zn
i = 1
i = 2
i = 2
Table 5. Summary of bond lengths (pm) and angles (°) for the hy-
Si(Td)
Si(D4h
φi(Td ǞD4h
10.34
8.12
53%
8.71
9.70
47%
1.70
24.06
20%
1.86
23.97
21%
drogen
bonding
interactions
in
[Cu2(SalCarbOMe)2]·
)
1.75CHCl3·1.25MeOH·5.25H2O (3)[a] and [Zn2(SalCarbOMe)2]·
CHCl3·MeOH (4).
[a]
)
[a] φi(Td ǞD4h) gives the angular fraction along the path from a
tetrahedron to a square, for definition see ref.[15]
D–H···A
d(D–H) d(H···A) ЄD–H···A
d(D···A)
.
Compound 3 (see Figure S5, Supporting Information)
Significant differences between 3 and 4 are also found
for the dihedral angles between the salicylidene and the car-
bazole moieties, with the somewhat larger angles observed
in 4 (see Table 3). This is most probably related to the differ-
ences observed for the supramolecular structures of 3 and
4 (vide infra). In general, the overall structures of 3 and 4
are similar to those reported for the unsubstituted carb-
azole derivative.[14] Nevertheless, significant differences are
observed for the supramolecular arrangement in the crystal
structures, as the presence of the imine N–H group of the
carbazole units leads to extensive hydrogen bonding, which
is absent in 3 and 4.
The crystal structures of 3 and 4 are primarily deter-
mined by π-π interactions between the aromatic moieties
(salicylidene, carbazole, and methoxyphenyl substituents),
leading to 2D-layered networks oriented along the crystal-
lographic (100) plane (Figures S2 and S3, Supporting Infor-
mation). The diameter of these sheets in 3 is found to be
1475 pm, whereas in 4 a somewhat smaller value of
1263 pm is observed. The two dinuclear complex units are
found to have different orientations with respect to the 2D
layers. The Cu1 and Zn1 units are arranged diagonally to
the sheets with dihedral angles (metal–metal vector and
normal vector) of 37 and 46°, respectively, whereas the Cu2
and Zn2 units are oriented almost coplanar within the lay-
ers with dihedral angles of 83 and 88°, respectively. The
resulting angle between the metal–metal vectors of the two
dinuclear units are 82 and 86° for 3 and 4, respectively.
As a consequence the intermolecular metal–metal dis-
C1C3–H···O21 100
C1C4–H···O21 100
C1C2–H···O12 100
C1C2–H···O22 100
216
207
246
224
172
152
144
140
316(2)
299(2)
332(2)
307(2)
Compound 4 (see Figure S4, Supporting Information)
O1M–H···O11
C2Cl–H···O22
84
100
197
211
157
152
276.2(4)
302.8(5)
[a] Chloroform molecules in 3 are disordered with site occupation
factors of 0.5 (C1C2, C1C3) and 0.25 (C1C4).
Although the dinuclear complex units [Cu2(SalCarb-
OMe)2] in compound 3 show a similar structural arrange-
ment, significant differences are observed for the location
of the solvent molecules. The chloroform molecules are dis-
ordered over four crystallographic positions, three of which
enable hydrogen bonding to the phenolate oxygen atoms
O12, O21, and O22 (Table 5 and Figure S5, Supporting In-
formation). The remaining solvent molecules are located
between the layers and disordered over several crystallo-
graphic positions. This leads to additional interactions with
the methoxy-substituted phenyl rings of the bridging li-
gands of the dinuclear units containing Cu2, as they are
oriented towards the surfaces of the layers. As a conse-
quence this substituent at the carbazole unit is disordered
over two positions.
Electronic Properties
Absorption spectra for 1 and 2 were recorded in THF
tances are considerably shorter than the intramolecular dis- solution, whereas those of 3 and 4 were measured in dichlo-
tances. Within the layers the shortest distances are found romethane (Figure 3). The nature of the absorption bands
between the independent dinuclear units with values of 827 was investigated utilizing semiempirical ZINDO/S calcula-
(Cu1 and Cu2) and 698 pm (Zn1 and Zn2). The shortest tions on DFT optimized structures. A comparison of simu-
metal–metal distances across the layers are between dinu- lated and measured spectra and a list of the character of
clear units containing Cu1 and Zn1 with values of 907 and orbitals near the HOMO–LUMO region are provided in
817 pm, respectively. This is consistent with the more dense Figure S6 and Table S1 of the Supporting Information.
packing in 4.
Fluorescence spectra were recorded by irradiating in the
Consequently, the major difference for the crystal struc- highest aromatic wavelength and are depicted in Figure 4.
tures arises from the amount and arrangement of incorpo- Excitation spectra were recorded to prove that all bands
rated solvent molecules. In compound 4 the methanol and originate from the desired compounds.
Eur. J. Inorg. Chem. 2011, 826–834
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
829

E. T. Spielberg, W. Plass
FULL PAPER
fluorescence bands are shifted to 530 nm (18870 cm–1) and
589 nm (16980 cm–1) exhibiting Stokes shifts of 6770 and
8660 cm–1, respectively. These large values indicate major
reorganization after excitation. The intensity is strongly re-
duced compared to 1, which is most probably due to new
relaxation pathways along the salicylidene units.
Upon complexation the absorption band of the Schiff
base ligand at 271 nm is only slightly shifted (3: 280 nm/
35710 cm–1 with logε = 4.87; 4: 278 nm/35970 cm–1 with
logε = 4.80), whereas the band at 358 nm is shifted to a
higher wavelength by about 40 nm [3: 412 nm (24270 cm–1)
with logε = 4.70; 4: 397 nm (25190 cm–1) with logε = 4.74].
The observed increase of the extinction coefficients can be
explained by the fact that the complexes contain two ligand
Figure 3. Absorption spectra of 1 in THF (solid line), 2 in THF
(dashed line), 3 in CH2Cl2 (dash-dotted line), and 4 in CH2Cl2
(dash-dot-dotted line). The inset enlarges the absorption band of 3 moieties per molecule.
at around 700 nm.
Compound 3 shows two additional broad absorption
bands of low intensity. The first is at around 480 nm
(20830 cm–1) with logε = 3.71 and can be attributed to π*–
π* transitions from the antibonding C–O bond orbital into
antibonding orbitals at the C–N bond with contributions
from the copper d orbitals. The second is at around 710 nm
(14080 cm–1) with logε = 2.93 and can be attributed to π*–
d(Cu) transitions from the antibonding C–O bond orbital
into the copper d orbitals. Consequently, this absorption
band is assigned as a phenolate to copper charge transfer
transition.
In THF solution 3 shows a fluorescence band at 436 nm
(22940 cm–1) with a Stokes shift of 1330 cm–1, whereas 4
shows a fluorescence maximum at 486 nm (20580 cm–1)
with a Stokes shift of 4610 cm–1. In both cases additional
weak fluorescence bands at around 530 and 590 nm are ob-
served, which are reminiscent of 2. This either indicates a
partial dissociation in solution, which is unlikely based on
the results of the UV/Vis titrations (vide supra), or a dif-
ferent fluorescent state, which is independent of metal coor-
dination.
Figure 4. Fluorescence spectra of 1 (solid line), 2 (dashed line), 3
(dash-dotted line), and 4 (dash-dot-dotted line). The spectra were
recorded in THF and have been normalized to one.
The spectrum of 1 consists mainly of four bands: the first
Square-wave voltammetry of 1 shows one oxidation pro-
with a maximum at around 390 nm (25640 cm–1) with logε cess at 0.69 V (Figure S7, Supporting Information). Al-
= 3.61, the second around 320 nm (31250 cm–1) with logε though this process is reversible, the differences between
= 4.29, the third at 287 nm (34840 cm–1) with logε = 4.23, oxidation and reduction current suggests further decay re-
and the fourth below 250 nm (Ͼ 40000 cm–1) with logε Ͼ actions. Measurements of 2, 3, and 4 exhibited quite unde-
4.6. Fluorescence spectra of 1 recorded in THF show a fined oxidation processes at similar redox potentials. A pos-
strong band at around 428 nm (23360 cm–1) with a shoulder sible explanation is the anodic polymerization described for
at 446 nm (22420 cm–1). The bands are shifted by 2280 and carbazoles.[16]
3220 cm–1 compared to the absorption. The interpretation
of the spectra is complicated by the obvious color depen-
dence on the solvent used for recrystallization. Using eth-
anol results in a weakly colored solid, whereas using toluene
Magnetic Properties
gives better yields of a dark purple solid. However, the
The temperature dependence of the magnetic suscep-
NMR spectra show no differences. The 320 nm transition tibility of 3 was determined in the temperature range be-
is assigned as an n–π* transition from the lone pair of the tween 2 and 300 K. Plots of χM = f(T) and χMT = f(T) are
central nitrogen atom into the aromatic backbone of the depicted in Figure 5. At high temperatures the value of χMT
carbazole unit based on calculations.
is 0.88 cm3 Kmol–1 that corresponds to the spin-only value
Upon Schiff base condensation two new absorption for two independent copper(II) ions (0.75 cm3 Kmol–1 for g
bands at 271 nm (36900 cm–1) with logε = 4.66 and 358 nm = 2). At very low temperatures χMT decreases indicating
(27930 cm–1) with logε = 4.61 arise, which can be attributed antiferromagnetic interactions between the copper centers.
to transitions from the carbazole (backbone and nitrogen The χMT data were fitted utilizing the Bleaney–Blowers
lone pair) into the π* of the Schiff base imine bond. The equation.[17]
830
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 826–834

CuII and ZnII Complexes with a Photoactive Bridging Ligand
ESR spectra for 3 were recorded on powder samples at
room temperature and 77 K as well as on a frozen solution
in chloroform at 77 K (see inset in Figure 5). The spectra
can be simulated using the rhombic g values [gx, gy, gz] =
[2.057, 2.125, 2.200] leading to an isotropic value of giso
=
2.127, which is in good agreement with the measured
susceptibilities and DFT calculations on mononuclear
model complexes (see Table 6). Moreover, the g values are
consistent with the geometric distortion observed for the
coordination environment of the copper centers in 3, which
is almost halfway between tetrahedral and square planar
(cf. Table 4).[21] With increasing distortion, given by the ro-
tation angle between the two chelate units, the normally
observed axial ESR spectrum becomes rhombic. Rotation
angles of around 50°, as observed in 3 (ЄSal1i/Sal2i, see
Table 3), lead to a clear separation of the two perpendicular
components (gy and gx).[22]
Figure 5. Temperature dependence of the magnetic susceptibility
χM (black squares) and the product χMT (open circles) of 3
measured at an applied field of 2000 G. The solid lines represent
the best simulation. Inset: ESR spectra of 3 recorded as a frozen
solution at 77 K (solid line), a powdered sample at 77 K (dashed
line), and at room temperature (dash-dotted line).
Table 6. ESR parameters for 3 from simulation of experimental
data and DFT calculations for mononuclear models of the copper
centers of the two independent molecules (Cu1 and Cu2).
The best fit for Equation (2) is obtained with the param-
eters g = 2.074(1), JCuCu = –3.14(1) cm–1, and χTIP
=
gx
gy
gz
giso
0.000268(4) cm3 Kmol–1 leading to an agreement factor R2
of 0.99972. The observed antiferromagnetic coupling con-
stant is in accordance with the value found for a similar
system with an unsubstituted carbazole bridge.[14] A similar
magnitude but ferromagnetic sign of the coupling constant
is observed for meta-phenylene-bridged copper complexes,
where spin-polarization effects are present.[18] However, a
superexchange mechanism between the coupled copper cen-
ters and meta-phenylene bridges leads to antiferromagnetic
interactions, but with significantly larger magnitude.[19]
exp.
DFT Cu1
DFT Cu2
2.057
2.049
2.053
2.125
2.054
2.058
2.200
2.178
2.187
2.127
2.094
2.099
However, in the spectra two features are observed for the
gz component, a sharp signal and a very broad signal. This
indicates the presence of two different entities, which is in
agreement with the crystal structure of 3, for which signifi-
cant differences are observed for the hydrogen bonding in-
teractions of the two crystallographic independent mole-
cules (Cu1 and Cu2, vide supra). Both the sharp and the
broad signal are also present in the spectrum of the frozen
solution, with the sharp signal being less intense compared
to the powder spectra. In line with the ESR spectra on the
powdered samples it is therefore tempting to attribute this
to the presence of an equilibrium of the complex [Cu2(Sal-
CarbOMe)2] with its hydrogen bonded aggregates in solu-
tion, with the latter giving rise to the broad signal. As the
large excess of solvent favors the hydrogen bonding, this
causes the sharp signal to decrease, whereas the broad sig-
nal gains in intensity (Table 6).
χM T = N μB/(3k)[1 + 1/3exp(–JCuCu/(kT))]–1 + χTIP
T
(2)
DFT calculations for the dinuclear copper complex uti-
lizing the B3LYP functional have been performed to further
investigate the magnetic properties. For these calculations
the structural parameters of the Cu1 unit derived from the
X-ray structure of 3 were used. The coupling constant is
determined by the broken symmetry (BS) approach [J =
E(BS) – E(T), where E(BS) and E(T) are the energies of the
broken symmetry and the triplet state, respectively].[20]
These calculations yield a value of –3.5 cm–1, which is in
excellent agreement with the experimental coupling deter-
mined for 3. Moreover, the alternating sign of the atomic
spin populations derived by Mulliken population analysis
indicates that a spin-polarization mechanism is in operation
Conclusions
We have described the synthesis of an N-(p-meth-
in the case of 3 (Figure S8, Supporting Information). The oxyphenyl) substituted carbazole as a bridging ligand. This
population analysis further shows that the delocalization of ligand contains two Schiff base-derived metal binding pock-
the spin densities at the copper centers into the carbazole ets and has been utilized for the synthesis of metallacyclic
bridging unit is quite small and comparable to that found binuclear copper and zinc complexes [Cu2(SalCarbOMe)2]
for the delocalization into the salicylidene chelates. This is (3) and [Zn2(SalCarbOMe)2] (4). Both complexes show sim-
consistent with rather weak coupling between the copper ilar structural features, but significantly differ in the geome-
centers through the bridging carbazole ligand. Moreover, it try of the metal coordination environment. For the zinc
should be noted that spin density is not found at the carbaz- center slightly distorted tetrahedral geometry is observed,
ole nitrogen atom and consequently also no delocalization whereas the copper centers have a distorted coordination
into the p-methoxyphenyl substituent is observed in the cal- geometry halfway between tetrahedral and square planar.
culations.
This leads to differences in the crystal packing with highly
Eur. J. Inorg. Chem. 2011, 826–834
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
831

E. T. Spielberg, W. Plass
FULL PAPER
(400 MHz, CDCl3, 25 °C): δ = 3.84 (s, 3 H, OCH3), 4.72 (s, 4 H,
NH2), 6.69–6.71 (m, 2 H, H2), 7.00–7.02 (m, 2 H, H3), 7.13–7.15
(m, 2 H, H4,5), 7.15 (s, 2 H, H1), 7.39–7.4 (m, 2 H, H4,5) ppm. 13C
NMR (100 MHz, CDCl3, 25 °C): δ = 55.9, 104.3, 110.0, 115.3,
115.5, 123.6, 127.7, 131.5, 135.5, 142.1, 157.9 ppm. MS-EI: m/z (%)
= 303 (100) [M], 288 (27) [M – CH3], 31 (38) [OCH3].
disordered solvent molecules for 3, which is also reflected
in the ESR spectra. Magnetic measurements show a weak
antiferromagnetic coupling between the two copper centers
in 3. The spectroscopic features show that most contri-
butions to the observed low energy absorption bands arise
from n–π* transitions with the lone pair of the carbazole
nitrogen. However, for 3 significant contributions of the
copper d orbitals are present leading to LMCT characteris-
tic of these absorption bands. Further investigations will be
focused on the interplay between photo activity and mag-
netism of the carbazole bridging system.
N-(4-Methoxyphenyl)-3,4-bis(salicylidenimino)carbazole (2): Salicyl-
aldehyde (240 μL, 0.28 g, 2.3 mmol) was added to a suspension
of 3,4-diamino-N-(4-methoxyphenyl)carbazole (0.35 g, 1.15 mmol)
in methanol (20 mL). The suspension was heated for 10 min under
reflux and allowed to cool to room temperature. The yellow pre-
cipitate was collected by filtration and washed with a small amount
of methanol. After drying under reduced pressure product was ob-
tained (0.48 mg, 0.94 mmol, 82%); m.p. 300 °C (dec.). C33H25N3O3
(511.58): calcd. C 77.48, H 4.93, N 8.21; found C 77.09, H 4.80, N
Experimental Section
8.03. Selected IR data (KBr): ν = 3437 [s, ν(O–H)], 3053 (w), 2930
˜
Materials: The chemicals for synthesis were used as received with-
out further purification. N-(p-Methoxyphenyl)carbazole was syn-
thesized according to a literature procedure.[12] THF for fluores-
(w), 2836 [w, ν(OCH3)], 1618 [s, ν(C=N)], 1514 [s, νarom(C–C)],
1285 (s), 1247 (s), 1149 (m), 1031 (m), 754 (s) cm–1. 1H NMR
(400 MHz, CDCl3, 25 °C): δ = 3.95 (s, 3 H, OCH3), 6.8–7.6 (m, 16
H, Harom), 8.11 (s, 2 H, H7), 8.80 (s, 2 H, N=CH), 13.61 (s, 2 H,
OH) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 55.7, 110.7,
112.5, 115.3, 117.2, 119.0, 119.6, 120.6, 123.7, 128.4, 129.7, 132.0,
132.7, 141.1, 141.4, 159.2, 160.5, 161.1 ppm. MS-ESI positive: m/z
(%) = 512 (100) [M + H]+, 413 (28).
cence measurements was obtained from Roth as RotiSolv®
99.9% UV/IR grade and was used without further purification.
Ն
1
1
1
Instrumentation: H, 13C, H-1H COSY, and H-13C heteronuclear
correlation NMR spectra were recorded with a Bruker Avance
400 MHz spectrometer. IR spectra were measured with a Bruker
IFS55/Equinox spectrometer with a Raman unit FRA 106/S on
samples prepared as KBr pellets. Mass spectra were measured with
a Bruker MAT SSQ 710 spectrometer. Elemental analysis (C,H,N)
were carried out with a Leco CHNS-932 and El Vario III elemental
analyzers. Thermogravimetric analysis for powdered samples was
performed with a NETZSCH STA409PC Luxx apparatus under
constant flow of air ranging from room temperature up to 1000 °C
with a heating rate of 1 °C/min. UV/Vis spectra were recorded with
a VARIAN CARY 5000 UV/Vis spectrometer. Fluorescence spec-
tra were recorded with a Jasco FP-6300 spectrofluorometer. ESR
measurements were performed with a Bruker ESP 300E using X-
Band (9 GHz). Magnetic susceptibilities were obtained from pow-
[Cu2(SalCarbOMe)2] (3): To a solution of 2 (102 mg, 0.2 mmol) in
chloroform/methanol (1:1, 10 mL) was added Cu(AcO)2·H2O
(40 mg, 0.2 mmol) dissolved in chloroform/methanol (1:1, 2 mL).
The resulting red solution was allowed to stand at room tempera-
ture for the solvent to slowly evaporate. Red brown crystals suitable
for X-ray crystallography were obtained within a day. Total
yield: 47 mg (0.04 mmol, 40%); m.p. 300 °C (dec.).
C66H46Cu2N6O6·0.8CHCl3 (1241.72): calcd. C 64.61, H 3.80, N
6.77; found C 64.68, H 3.81, N 6.78. Selected IR data (KBr): ν =
˜
3435 [s, ν(O–H)], 3052 [w, νarom(C–H)], 2988 (w), 2927 (w), 1605
[vs, ν(C=N)], 1530 (s), 1514 [vs, νarom(C–C)], 1461 (s), 1444 (s),
1248 (s), 1148 (s), 757 (m) cm–1. MS-ESI positive: m/z (%) = 1167
(100) [M + Na]+, 1146 (36), M+. ESR (X-band, powder, 77 K): g
= [2.057, 2.125, 2.200].
dered samples in gelatin capsules using
a Quantum-Design
MPMSR-5S SQUID magnetometer equipped with a 5 T magnet
in the range from 300 to 2 K (for details see ref.[23]). The measured
data were corrected for diamagnetism of the capsules and the in-
trinsic diamagnetism of the sample, estimated by measurements on
a similar ligand system.
[Zn2(SalCarbOMe)2] (4): To a solution of 2 (102 mg, 0.2 mmol) in
chloroform/methanol (1:1, 10 mL) was added ZnAc2·2H2O (44 mg,
0.2 mmol) dissolved chloroform/methanol (1:1, 2 mL). The re-
sulting orange-brown solution, which showed green fluorescence,
was allowed to stand at room temperature for the solvent to slowly
evaporate. Crystals suitable for X-ray crystallography were ob-
tained within a day. Total yield: 52 mg (0.045 mmol, 45%); m.p.
400 °C (dec.). C66H46N6O6Zn2·CHCl3·MeOH (1301.31): calcd. C
62.76, H 3.95, N 6.46; found C 62.50, H 3.63, N 6.45. Selected IR
3,4-Diamino-N-(4-methoxyphenyl)carbazole (1): Cu(NO3)2·3H2O
(7.94 g, 33 mmol) was added to a mixture of acetic acid (15 mL)
and acetic acid anhydride (30 mL) and stirred for 10 min at room
temperature. N-(4-Methoxyphenyl)carbazole (7.50 g, 27 mmol) was
added slowly over five minutes. After adding acetic acid (15 mL)
the resulting solution was stirred for an additional 30 min. The
solution was poured into ice water (250 mL), the yellow precipitate
was collected by filtration, and washed three times with water (150
mL). The product was dried for one day under reduced pressure at
70 °C. This solid was suspended in ethyl acetate and several spatu-
las of Raney nickel were added. Gaseous hydrogen was applied
with a pressure of 55 bar. The mixture was stirred at 80 °C for 4 h.
After cooling the nickel was removed by filtration and washed three
times with ethyl acetate. After drying over Na2SO4 the solvent was
removed to give the crude product (7.50 g). Recrystallization from
toluene yielded 3,4-diamino-N-(p-methoxyphenyl)carbazole (5.54
data (KBr): ν = 3435 [m, ν(O–H)], 3044 (w), 3001 (w), 2957 (w),
˜
2933 (w), 2905 (w), 2834 [w, ν(OCH3)], 1605 [vs, ν(C=N)], 1534 (s),
1514 [vs, νarom(C–C)], 1463 (s), 1442 (s), 1248 (s), 1173 (s), 1025
(m), 757 (m) cm–1. MS-DEI: m/z (%) = 1151 (0.5) [L2Zn2]+, 575
(0.1) [LZn]+, 83 (100), 47 (21).
Computational Details: The evaluation of the absorption spectra is
based on structures which were optimized with DFT calculations
utilizing the BP86 functional[24] with the RI approximation and
def2-TZVP basis sets[25–27] as provided by the program package
TurboMole.[28] Absorption spectra were calculated for the opti-
g, 18.3 mmol, 67%); m.p. 320 °C (dec.). C19H17N3O (303.36): calcd. mized structures using the ZINDO/S method[29] as implemented
C 75.23, H 5.65, N 13.85; found C 75.15, H 5.73, N 13.94. Selected
within the ORCA program package[30,31] and plotted with the orca_
IR data (KBr): ν = 3377 [s, ν (NH )], 3292 [s, ν (NH )], 2956 (m), mapspc functionality. To study the magnetic properties of 3 DFT
˜
as
2
s
2
2933 (m), 2838 [m, ν(OCH3)], 1613 [m, νarom(C–C)], 1580 [m, calculations with the program package ORCA[30] were performed
νarom(C–C)], 1514 [s, νarom(C–C)], 1218 (s) cm–1. 1H NMR utilizing the hybrid functional B3LYP[32] in combination with def2-
832
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 826–834

CuII and ZnII Complexes with a Photoactive Bridging Ligand
SVP (C,N,O,H) and def2-TZVP (Cu) basis sets.[25,26] The geometry
Table 7. Summary of crystallographic data for [Cu2(SalCarbOMe)2]·
was taken from the crystallographic structure for the dinuclear unit 1.75CHCl3·1.25MeOH·5.25H2O (3) and [Zn2(SalCarbOMe)2]·
CHCl3·MeOH (4).
Cu1, with the positions of the hydrogen atoms optimized. The data
was analyzed according the phenomenological Heisenberg Hamil-
tonian H = –JS1S2, where the coupling constant J is given by the
energy difference between the relevant triplet and singlet states [J
= E(S) – E(T)]. Therefore, a negative value of J indicates antiferro-
magnetic coupling with a triplet ground state. The BS approach
originally proposed by Noodleman was used to cope with the DFT
inherent problem that the singlet state cannot be expressed by a
single determinant.[33] In this formalism, the energy of the singlet
3
4
Formula
C69H63.25Cl5.25Cu2N6O12.5 C68H51Cl3N6O7Zn2
Formula weight
Crystal system
Space group
a [pm]
b [pm]
c [pm]
1489.70
triclinic
P1
1301.24
triclinic
P1
¯
¯
15.6401(9)
15.9390(8)
16.8759(9)
63.449(3)
71.314(2)
77.162(2)
3549.4(3)
2
14.3107(3)
14.9621(6)
16.9319(5)
112.754(1)
96.694(2)
112.014(2)
2950.92(16)
2
state is computed on the basis of a single determinate of broken α [°]
β [°]
spin and space symmetry built up by localizing spin up and spin
down electrons onto the two copper centers, which is not a spin
eigenstate. Although the original formalism was based on spin pro-
jection to extract the singlet energy, it has been shown that this
energy can be directly estimated from the energy of the BS solution
[E(BS)] without performing any spin projection according to J =
E(BS) – E(T). This procedure is in accordance with the proposal
of Perdew et al., who suggested that the BS determinant is the
γ [°]
V [106 pm3]
Z
T [K]
183(2)
1.394
1533
0.861
1.70–27.47
24816
183(2)
1.464
1336
1.011
2.36–27.47
21441
dcalc [gcm–3]
F(000)
μ [mm–1]
θ range [°]
correct solution of the Kohn–Sham equations for the singlet Measured data
state.[34] The ESR parameters were calculated with the program
package ORCA[30,35] utilizing the one parameter hybrid functional
PBE0[36] in combination with VDZ (C,N,O,H) and CoreProp (Cu)
basis sets.[25,37] The calculations were performed on mononuclear
model systems for the two crystallographic independent molecules
(Cu1 and Cu2), derived from the crystal structure of 3, that contain
the full coordination environment of the copper ions, with the car-
bazole moiety replaced by a phenyl group and the positions of the
hydrogen atoms optimized.
Unique data (Rint
Data [IϾ2σ(I)]
GOF on F2
R1 (observed data) 0.0853
wR2 (all data) 0.2637
)
16007 (0.0505)
8401
1.028
13406 (0.0369)
9104
1.018
0.0491
0.1127
CCDC-774835 (for 3) and -774836 (for 4) contain the supplemen-
tary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Centre
via www.ccdc.cam.ac.uk/data-request/cif.
Electrochemical Measurements: Cyclic square-wave voltammetric
measurements were carried out by a three-electrode technique using
a home-built computer-controlled instrument. The experiments
were performed in acetonitrile solutions containing 0.25 m tetra-n-
butylammonium hexafluorophosphate under a blanket of solvent
saturated with argon. Ag/AgCl was used as the reference electrode
in acetonitrile containing 0.25 m tetra-n-butylammonium chloride.
The working electrode was a hanging mercury drop. The potentials
reported in this paper refer to the ferrocenium/ferrocene couple,
which has been measured at the end of each experiment (for details
see ref.[38]).
Supporting Information (see footnote on the first page of this arti-
cle): UV/Vis titrations with ligand 2 and copper and zinc salts,
graphics showing the crystal packing and hydrogen bonding for 3
and 4, comparison of measured and computed absorption spectra
for compounds 1–4, square-wave voltammograms of compound 1,
Mulliken spin population for the broken symmetry state derived
from DFT calculations on 3, and details on the orbital character
in the HOMO–LUMO region derived from ZINDO/S calculations
on 3.
Acknowledgments
X-ray Crystallography: The crystallographic data were collected
with a Nonius KappaCCD diffractometer using graphite-mono-
chromated Mo-Kα radiation (71.073 pm). A summary of crystallo-
graphic and structure refinement data is given in Table 7. The struc-
ture was solved by direct methods with SHELXS-97 and was full-
matrix least-squares refined against F2 using SHELXL-97.[39] All
non-hydrogen atoms were refined anisotropically. Hydrogen atoms
were calculated and treated as riding atoms with fixed thermal pa-
rameters. For the dinuclear unit containing Cu2 of compound 3
the p-methoxyphenyl substituent on the carbazole was found to be
disordered over two positions in a 3:1 ratio. The chloroform mole-
cules are disordered over four positions with occupancy factors of
0.5 (C1C1, C1C2, and C1C3) and 0.25 (C1C4), whereas the meth-
anol molecules are disordered over three positions with occupancy
factors of 0.5 (C1M1 and C1M2) and 0.25 (C1M3). The water
molecules in the crystal structure are disordered over 16 positions
and hydrogen bonded to the methanol molecules and the chloro-
form molecule with C1C1, which is not hydrogen bonded to one
of the independent dinuclear copper complexes (Cu1 and Cu2).
The latter molecules are all located in the region between the 2D
layers observed for 3 (Figure S2, Supporting Information).
We thank the Deutsche Forschungsgemeinschaft (DFG) (SPP 1137
“Molecular Magnetism”) for financial support. E. T. S. gratefully
acknowledges scholarships from the Carl-Zeiss-Stiftung and the
Jenaer Graduiertenakademie. In addition, we thank Dr. M. Ru-
dolph for the measurement of the square-wave voltammograms and
helpful discussions.
[1] a) J. S. Miller, M. Drillon (Eds.), Magnetism: Molecules to Ma-
terials, vol. I–V, Wiley-VCH, Weinheim, Germany, 2001–2005;
b) O. Kahn, Acc. Chem. Res. 2000, 33, 647–657; c) D. Gattes-
chi, R. Sessoli, J. Villain, Molecular Nanomagnets, Oxford Uni-
versity Press, Oxford, 2006; d) W. Plass, Chem. Unserer Zeit
1998, 32, 323–333; e) A. Cornia, A. F. Costantino, L. Zobbi,
A. Caneschi, D. Gatteschi, M. Mannini, R. Sessoli, Struct.
Bonding (Berlin) 2006, 122, 133–161.
[2] a) Y. Zheng, F. Cardinali, N. Armaroli, G. Accorsi, Eur. J. In-
org. Chem. 2008, 12, 2075–2080; b) M.-J. Li, Z. Chen, V. W.-
W. Yam, Y. Zu, ACS Nano 2008, 2, 905–912; c) Y. Zhang, S.
Pan, X. Teng, Y. Luo, G. Li, J. Phys. Chem. C 2008, 112, 9623–
9626; d) R.-S. Zhou, L. Ye, H. Ding, J.-F. Song, X.-Y. Xu, J.-
Eur. J. Inorg. Chem. 2011, 826–834
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
833

E. T. Spielberg, W. Plass
FULL PAPER
Q. Xu, J. Solid State Chem. 2008, 18, 567–575; e) L.-Q. Yu,
R.-D. Huang, Y.-Q. Xu, T.-F. Liu, W. Chu, C.-W. Hu, Inorg.
Chim. Acta 2008, 361, 2115–2122; f) C. A. Black, J. S. Costa,
W. T. Fu, C. Massera, O. Roubeau, S. J. Teat, G. Aromi, P.
Gamez, J. Reedijk, Inorg. Chem. 2009, 48, 1062–1068.
[3] W. Plass, Coord. Chem. Rev. 2009, 253, 2286–2295.
[4] S. R. Marder, B. Kippelen, A. K.-Y. Jen, N. Peyghambarian,
Nature 1997, 388, 845–851.
[22]
a) S. K. Hoffmann, J. Goslar, J. Solid State Chem. 1982, 44,
343–353; b) A. Bonamartini-Corradi, L. P. Battaglia, J. Rub-
enaker, R. D. Willett, T. E. Grigereit, P. Zhou, J. E. Drumheller,
Inorg. Chem. 1993, 32, 3859–3863.
a) W. Plass, Angew. Chem. Int. Ed. Engl. 1996, 35, 627–631; b)
W. Plass, Inorg. Chem. 1997, 36, 2200–2205.
a) J. P. Perdew, Phys. Rev. B 1986, 33, 8822–8824; b) A. D.
Becke, Phys. Rev. A 1988, 38, 3098–3100.
A. Schäfer, H. Horn, R. Ahlrichs, J. Chem. Phys. 1992, 97,
2571–2577.
[23]
[24]
[25]
[26]
[5] a) J. Morin, M. Leclerc, D. Ades, A. Siove, Macromol. Rapid
Commun. 2005, 26, 761–778; b) S. Grigalevicius, Synth. Met.
2006, 156, 1–12.
F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7,
3297–3305.
[6] O. Paliulis, J. Ostrauskaite, V. Gaidelis, V. Jankauskas, P.
Strohriegl, Macromol. Chem. Phys. 2003, 204, 1706–1712.
[7] a) W.-Y. Wong, C.-L. Ho, Z.-Q. Gao, B.-X. Mi, C.-H. Chen,
K.-W. Cheah, Z. Lin, Angew. Chem. Int. Ed. 2006, 45, 7800–
7803; b) W.-Y. Wong, L. Liu, D. Cui, L. M. Leung, C.-F.
Kwong, T.-H. Lee, H.-F. Ng, Macromolecules 2005, 38, 4970–
4976; c) D. Velasco, S. Castellanos, M. Lopez, F. Lopez-Cala-
horra, E. Brillas, L. Julia, J. Org. Chem. 2007, 72, 7523–7532.
[8] T. Xu, R. Lu, X. Liu, X. Zheng, X. Qiu, Y. Zhao, Org. Lett.
2007, 9, 797–800.
[27]
[28]
F. Weigend, Phys. Chem. Chem. Phys. 2006, 8, 1057–1065.
a) R. Ahlrichs, M. Bär, M. Häser, H. Horn, C. Kölmel, Chem.
Phys. Lett. 1989, 162, 165–169; b) O. Treutler, R. Ahlrichs, J.
Chem. Phys. 1995, 102, 346–354; c) M. von Arnim, R. Ahlrichs,
J. Comput. Chem. 1998, 19, 1746–1757; d) M. von Arnim, R.
Ahlrichs, J. Chem. Phys. 1999, 111, 9183–9190; e) R. Ahlrichs,
Phys. Chem. Chem. Phys. 2004, 6, 5119–5121.
a) J. Ridley, M. C. Zerner, Theor. Chim. Acta 1973, 32, 111–
134; b) A. D. Bacon, M. C. Zerner, Theor. Chim. Acta 1979,
53, 21–54; c) M. C. Zerner, G. H. Loew, R. F. Kirchner, U. T.
Mueller-Westerhoff, J. Am. Chem. Soc. 1980, 102, 589–599; d)
W. P. Anderson, W. D. Edwards, M. C. Zerner, Inorg. Chem.
1986, 25, 2728–2732.
F. Neese, ORCA, ab initio density-functional and semiempir-
ical program package, version 2.7, University of Bonn, Ger-
many, 2008.
T. Petrenko, F. Neese, J. Chem. Phys. 2007, 127, 1664319.
a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) C. Lee,
W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
L. Noodleman, J. Chem. Phys. 1981, 74, 5737–5743.
a) J. P. Perdew, A. Savin, K. Burke, Phys. Rev. A 1995, 51,
4531–4541; b) J. P. Perdew, M. Ernzerhof, K. Burke, A. Savin,
Int. J. Quantum Chem. 1997, 61, 197–205.
F. Neese, J. Chem. Phys. 2001, 115, 11080–11096.
C. Adamo, V. Barone, J. Chem. Phys. 1999, 110, 6158–6170.
a) K. Eichkorn, O. Treutler, H. Öhm, M. Häser, R. Ahlrichs,
Chem. Phys. Lett. 1995, 240, 283–289; b) K. Eichkorn, F. Weig-
end, O. Treutler, R. Ahlrichs, Theor. Chem. Acc. 1997, 97, 119–
124; c) A. J. H. Wachters, J. Chem. Phys. 1970, 52, 1033–1036;
d) TurboMole basis sets: ftp://ftp.chemie.uni-karlsruhe.de/pub/
basen and ftp://ftp.chemie.uni-karlsruhe.de/pub/jbasen; e) R.
Ahlrichs and co-workers, unpublished results.
a) S. Nica, A. M. Rudolph, H. Görls, W. Plass, Inorg. Chim.
Acta 2007, 360, 1743–1752; b) S. Nica, A. Buchholz, M. Ru-
dolph, A. Schweizer, M. Wächtler, H. Breitzke, G. Buntkowsky,
W. Plass, Eur. J. Inorg. Chem. 2008, 2350–2359.
[29]
[30]
[9] Y. Song, C. Di, Z. Wei, T. Zhao, W. Xu, Y. Liu, D. Zhang, D.
Zhu, Chem. Eur. J. 2008, 14, 4731–4740.
[10] a) W. Rettig, M. Zander, Chem. Phys. Lett. 1982, 87, 229–234;
b) F. Evers, J. Giraud-Girard, S. Grimme, J. Manz, C. Monte,
M. Oppel, W. Rettig, P. Saalfrank, P. Zimmermann, J. Phys.
Chem. A 2001, 105, 291–2924; c) S. Kapelle, W. Rettig, R. La-
pouyade, Chem. Phys. Lett. 2001, 348, 416–424; d) T. Xu, R.
Lu, X. Liu, P. Chen, X. Qiu, Y. Zhao, J. Org. Chem. 2008, 73,
1809–1817.
[31]
[32]
[11] A. Samanta, S. Saha, R. W. Fessenden, J. Phys. Chem. A 2001,
105, 5438–5441.
[12] H. Zhang, Q. Cai, D. Ma, J. Org. Chem. 2005, 70, 5164–5173.
[13] S. Keinan, D. Avnir, Inorg. Chem. 2001, 40, 318–323.
[14] A. R. Paital, A.-Q. Wu, Guo, G. Aromi, J. Ribas-Arino, D.
Ray, Inorg. Chem. 2007, 46, 2947–2949.
[15] J. Cirera, E. Ruiz, S. Alvarez, Chem. Eur. J. 2006, 12, 3162–
3167.
[33]
[34]
[35]
[36]
[37]
[16] J. Ambrose, R. Nelson, J. Electrochem. Soc. 1968, 115, 1159–
1164.
[17] O. Kahn, Molecular Magnetism, Wiley-VCH, Weinheim 1993.
[18] a) T. Glaser, H. Theil, M. Heidemeier, C. R. Chim. 2008, 11,
1121–1136; b) E. T. Spielberg, W. Plass, Eur. J. Inorg. Chem.
2010, 3093–3096.
[19] a) D. Plaul, A. Buchholz, H. Görls, W. Plass, Polyhedron 2007,
26, 4581–4590; b) D. Plaul, D. Geibig, H. Görls, W. Plass, Poly-
hedron 2009, 28, 1982–1990.
[20] a) W. Plass, A. Pohlmann, J. Rautengarten, Angew. Chem. Int.
Ed. 2001, 40, 4207–4210; b) W. Plass, Angew. Chem. Int. Ed.
Engl. 1996, 35, 627–631; c) E. T. Spielberg, M. Fittipaldi, D.
Geibig, D. Gatteschi, W. Plass, Inorg. Chim. Acta 2010, 363,
4269–4276.
[38]
[39]
G. Sheldrick, SHELXS-97 and SHELXL-97, University of
Göttingen, Germany, 1997.
[21] S. Keinan, D. Avnir, J. Chem. Soc., Dalton Trans. 2001, 941–
Received: September 10, 2010
Published Online: January 14, 2011
947.
834
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 826–834