Correlated coordination and redox activity of a hemilabile noninnocent ligand in nickel complexes

Article abstract of DOI:10.1002/chem.201304316

The compound [Ni(Q_M)₂], Q_M=4,6-di-tert- butyl-N-(2-methylthiomethylphenyl)-o-iminobenzoquinone, is a singlet diradical species with approximately planar configuration at the tetracoordinate metal atom and without any Ni-S bonding interaction. One-electron oxidation results in additional twofold Ni-S coordination (d_Ni-S≈2.38 A) to produce a complex cation of [Ni(Q_M)₂](PF₆) with hexacoordinate Ni^II and two distinctly different mer-configurated tridentate ligands. The O,O′-trans arrangement in the neutral precursor is changed to an O,O′-cis configuration in the cation. The EPR signal of [Ni(Q_M)₂](PF₆) has a very large g anisotropy and the magnetic measurements indicate an S=3/2 state. The dication was structurally characterized as [Ni(Q_M)₂](ClO ₄)₂ to exhibit a similar NiN₂O ₂S₂ framework as the monocation. However, the two tridentate (O,N,S) ligands are now equivalent according to the formulation [Ni^II(Q_M⁰)₂]²⁺. Cyclic voltammetry reflects the qualitative structure change on the first, but not on the second oxidation of [Ni(Q_M)₂], and spectroelectrochemistry reveals a pronounced dependence of the 800-900 nm absorption on the solvent and counterion. Reduction of the neutral form occurs in an electrochemically reversible step to yield an anion with an intense near-infrared absorption at 1345 nm (ε=10400 M^-1 cm ^-1) and a conventional g factor splitting for a largely metal-based spin (S=1/2), suggesting a [(Q_M^.-)Ni ^II(Q_M^2-)]^- configuration with a tetracoordinate metal atom with antiferromagnetic Ni^II-(Q _M^.-) interactions and symmetry-allowed ligand-to-ligand intervalence charge transfer (LLIVCT). Calculations are used to understand the Ni-S binding activity as induced by remote electron transfer at the iminobenzoquinone redox system. Radical change: Oxidation of the singlet diradical species [Ni^II(Q_M^.-) ₂], which contains the hemilabile redox-active ligand Q_M, results in coordination of the S-donor functions and spin-crossover at the nickel atom, although the electron is removed not from the metal but from the noninnocent chelate ligand (see scheme).

Full text of DOI:10.1002/chem.201304316

DOI: 10.1002/chem.201304316
Full Paper
&
Redox Chemistry
Correlated Coordination and Redox Activity of a Hemilabile
Noninnocent Ligand in Nickel Complexes
ˇ [b]
[a]
[a]
[b]
[a]
Alexa Paretzki, Martina Bubrin, Jan Fiedler, Stanislav Zꢀlis, and Wolfgang Kaim*
Abstract: The compound [Ni(QM)2], QM =4,6-di-tert-butyl-N-
(2-methylthiomethylphenyl)-o-iminobenzoquinone, is a sin-
glet diradical species with approximately planar configura-
according to the formulation [NiII(QM0)2]2+. Cyclic voltamme-
try reflects the qualitative structure change on the first, but
not on the second oxidation of [Ni(QM)2], and spectroelectro-
tion at the tetracoordinate metal atom and without any Niꢀ chemistry reveals a pronounced dependence of the 800–
S bonding interaction. One-electron oxidation results in ad-
ditional twofold NiꢀS coordination (dNiꢀS ꢁ2.38 ꢁ) to produce
a complex cation of [Ni(QM)2](PF6) with hexacoordinate NiII
and two distinctly different mer-configurated tridentate li-
gands. The O,O’-trans arrangement in the neutral precursor
is changed to an O,O’-cis configuration in the cation. The
EPR signal of [Ni(QM)2](PF6) has a very large g anisotropy and
900 nm absorption on the solvent and counterion. Reduc-
tion of the neutral form occurs in an electrochemically rever-
sible step to yield an anion with an intense near-infrared ab-
sorption at 1345 nm (e=10400mꢀ1 cmꢀ1) and a conventional
g factor splitting for a largely metal-based spin (S=1= ), sug-
2
gesting a [(QM )NiII(QM2ꢀ)]ꢀ configuration with a tetracoordi-
Cꢀ
Cꢀ
nate metal atom with antiferromagnetic NiII–(QM ) interac-
the magnetic measurements indicate an S=3= state. The di-
tions and symmetry-allowed ligand-to-ligand intervalence
charge transfer (LLIVCT). Calculations are used to understand
the NiꢀS binding activity as induced by remote electron
transfer at the iminobenzoquinone redox system.
2
cation was structurally characterized as [Ni(QM)2](ClO4)2 to ex-
hibit a similar NiN2O2S2 framework as the monocation. How-
ever, the two tridentate (O,N,S) ligands are now equivalent
Introduction
Hemilabile ligands[1] are increasingly acknowledged as valuable
components of functional coordination compounds.[2] The con-
trol of hemilability beyond equilibria situations is an attractive
goal, for example, in catalysis.[1,2] One parameter to control
hemilability is the redox state of the metal, which may prefer
larger or smaller coordination numbers. The copper(II/I) redox
pair is a prominent case in point.[3] However, the hemilabile
ligand itself can also be redox active, leading to potentially
noninnocent behavior[4] in transition metal complexes. In con-
nection with the H2-activation research described by Ringen-
berg, Rauchfuss et al.,[5] we have thus described a redox-active
iminobenzoquinone ligand with an additional thioether donor
function, Qy =4,6-di-tert-butyl-N-(2-methylthiophenyl)-o-imino-
benzoquinone, where the S-donor atom coordinates to
Scheme 1. Intramolecular oxidative addition involving a hemilabile noninno-
cent ligand Qy.
[Ir(C5Me5)]2+ in the semiquinone state, but not in the fully re-
duced amidophenolate form (Scheme 1).[6]
II
Cꢀ
Cꢀ
On the other hand, the three-spin complex [(Qy )Cu (Qy )]
showed a single weak metal–S coordination accompanied by
twisting of the CuN2O2 core and ligand-based spin[7] in contrast
II
Cꢀ
Cꢀ
to the planar parent system [(Q )Cu (Q )] without S-donor
function (Scheme 2).[8]
[a] Dipl.-Chem. A. Paretzki, Dipl.-Chem. M. Bubrin, Prof. Dr. W. Kaim
Institut fꢀr Anorganische Chemie
Universitꢁt Stuttgart, Pfaffenwaldring 55
70550 Stuttgart (Germany)
E-mail: kaim@iac.uni-stuttgart.de
ˇ
[b] Dr. J. Fiedler, Dr. S. Zꢂlis
´
J. Heyrovsky Institute of Physical Chemistry
v.v.i., Academy of Sciences of the Czech Republic
ˇ
Dolejskova 3, 18223 Prague (Czech Republic)
Supporting information for this article contains crystallographic, spectro-
scopic, and computational data and is available on the WWW under
http://dx.doi.org/10.1002/chem.201304316.
Scheme 2. Twisting and S coordination in copper(II) complexes of o-imino-
benzosemiquinones (ref. [7]).
Chem. Eur. J. 2014, 20, 5414 – 5422
5414
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
One of the most striking examples in the popular[5–11] coordi-
nation chemistry that involves redox-active amidophenolate/
iminobenzoquinone ligands, is the planar configuration at the
nickel atom, analyzed experimentally and theoretically as
Results and Discussion
Synthesis, structure and spectroscopy of 1
II
[8,9]
Cꢀ
Cꢀ
[(Q )Ni (Q )].
Similar to the o-diiminobenzosemiquinonato
The new, hemilabile, noninnocent ligand QM was obtained in
the conventional way.[7,8,11] The reaction with NiCl2 yields com-
plex 1, identified by elemental analysis, mass spectroscopy,
1H NMR spectroscopy (see the Experimental Section) and by
two crystal structures, one of a solvent-free form (Tables 1,2,
and S3 in the Supporting Information, Figure 1) and one of an
acetonitrile solvate (Tables S1,S2 and Figure S1 in the Support-
analogues,[12–14] the diamagnetic compounds were addressed
as “singlet diradicals”[13–15] due to the antiferromagnetic cou-
pling of the structurally identified iminosemiquinone ligands
through the low-spin nickel center.[9] Metric parameters are
usually accepted as valid indicators for the oxidation state of
these Q ligands.[8,9,13,16] Redox reactions have shown that the
cations of the all-N donor systems are accessible,[13a] whereas
apparent two-electron oxidation processes of the iminosemi-
quinonato complexes lead to dications.[8,10b,17,18] The structures
II
Cꢀ
Cꢀ
of [(Qy )Ni (Qy )][17] and related species[18] show a similar
II
Cꢀ
Cꢀ
Cꢀ
planar NiN2O2 configuration as in [(Q )Ni (Q )], Q =4,6-di-
tert-butyl-N-phenyl-o-iminobenzosemiquinone,[8] without NiꢀS
bonding.
The particular electrochemical oxidation behavior during
cyclic voltammetry experiments[17] of [(Qy )Ni (Qy )] has
prompted us to investigate this process in detail, employing
the more flexible, tridentate, new, (O,N,S) ligand QM =4,6-di-
tert-butyl-N-(2-methylthiomethylphenyl)-o-iminobenzoquinone.
Spectroelectrochemical (UV-visible-NIR, EPR) and structural in-
formation could now be collected for the redox series
II
Cꢀ
Cꢀ
Figure 1. Molecular structure of 1 in the crystal.
[Ni(QM)2]0/+/2+, 10/+/2+ whereas the anion was studied with
,
spectroelectrochemistry. The purpose was to establish the ap-
propriate oxidation-state combinations for each state on the
basis of experimental and DFT-calculated spectroscopic and
structural changes.
ing Information). Both structures show an approximately
planar NiO2N2 configuration with the thioether sulfur atoms at
nonbonding distances (5.507(1) and 5.767(1) ꢁ for 1, 5.066(1)
and 3.712(1) ꢁ for 1ꢃ0.3 CH3CN). The configuration at the
metal atom reveals slight deviation from planarity (Table S4 in
the Supporting Information). In spite of this slight distortion,
which has also been observed for the corresponding neutral
[8,9]
Cꢀ
Cꢀ
complexes [(L )Ni(L )],
the potentially O,N,S tridentate li-
gands coordinate only in a bidentate fashion with trans-posi-
tioned O and N donors. The results are reproduced by DFT cal-
culations (Table 2). In agreement with earlier reports[8–14] on
similar species, complex 1 displays a very intense absorption
with a lmax value at 894 nm (e=26000mꢀ1 cmꢀ1).
1
The H NMR spectrum of 1 in [D8]toluene is influenced by
Nickel complexes [NiL2]n with noninnocent, quinonoid, che-
late ligands L have attracted attention for several decades[8–14]
because of the multiple redox activity/processes, the puzzling
structures, and oxidation-state situations. Whereas the elec-
tronic structures have been discussed in detail by Wieghardt,
Neese, and co-workers, the ionic species have only been par-
tially characterized: Anions were obtained as intermediates of
two one-electron steps, whereas the oxidations showed a less
well-defined response, partially appearing as irreversible two-
electron processes.[8,10b,13,14,17,18] Our present contribution will
shed light on the changes that occur during oxidation of com-
pound 1 and thus provide further insight into the bonding po-
tential of hemilabile noninnocent ligands.
dynamic effects, which involve the flexible, saturated, ꢀ
CH2SCH3 side arms,[19] and by marginal paramagnetism (see
below). Temperature-dependent measurements (Figure S2 in
the Supporting Information) show the expected, slightly
broadened signals in the unsymmetrically frozen configuration
at 223 K (see the Experimental Section) and broadening, shift-
ing, and partially coalescing signals at higher temperatures,
until 348 K. Superconducting quantum interference device
(SQUID) susceptibility and EPR measurements of 1 (powder) at
variable temperatures showed only very weak paramagnetic
response, in agreement with the largely planar configuration
around the metal atom.
Chem. Eur. J. 2014, 20, 5414 – 5422
www.chemeurj.org
5415
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
established coordination change. On the other hand,
Table 1. Crystallographic data.
once generated, the cation 1+ exhibits a reversible
1[a]
[1](PF6)ꢃCH2Cl2
[1](ClO4)2 ꢃ0.25CH2Cl2
oxidation to 12+
.
The electrochemical (and spectroelectrochemical)
results can be understood from the molecular struc-
tures (Tables 1,2, and S3 in the Supporting Informa-
tion, Figure 3, Figure 4). Compound 1 could be oxi-
dized chemically to [1](PF6) and [1](ClO4)2 by silver
salts. Both ions 1+ and 12+ show hexacoordination in
the crystal with two normal NiꢀS bond lengths of
about 2.38 ꢁ and with changed configuration, that is,
cis-positioned O donors at the nickel atom in contrast
to trans-positioned O centers in nonoxidized 1. This
configuration, which results from the necessity to
have two mer-coordinating tridentate ligands at the
central metal ion, must involve an intermediate bond
dissociation and rearrangement (Scheme 3). At low
temperature a single two-electron oxidation peak is
observed (Figure 2); this suggests that the prevailing
oxidation mechanism under these conditions in-
cludes two close one-electron transfers, followed by
a rearrangement in the coordination sphere, that is,
an EEC process different from the EC+E mechanism
observed at room temperature.
formula
Mr
crystal system
space group
cell dimensions [ꢁ,8]
C44H58N2NiO2S2 C45H60Cl2F6N2NiO2PS2 C44.25H58.5Cl2.5N2NiO10S2
769.75
monoclinic
P21/n
999.65
triclinic
¯
P1
989.90
monoclinic
C2/c
a=18.277(1) a=11.028(1)
b=10.1072(6) b=15.140(1)
c=23.657(1) c=15.299(1)
a=79.521(4)
b=13.5097(7)
c=20.7353(9)
a=90
a=90
a=100.215(4)
b=105.044(4) b=93.383(4)
b=103.894(2)
g=90
21624(2)
g=90
g=95.550(4)
2494.5(4)
V [ꢁ3]
crystal size [mm]
4220.3(4)
0.14; 0.07;
0.02
0.25; 0.15; 0.04
0.12; 0.07; 0.06
Z
T [K]
l [ꢁ]
q range
index range
4
2
2
100(2)
1.54178
2.74–66.528
100(2)
0.71073
1.37–28.628
100
1.54178
2.29–66.628
ꢀ93ꢂhꢂ87
ꢀ15ꢂkꢂ15
ꢀ17ꢂlꢂ24
119471
ꢀ20ꢂhꢂ21 ꢀ14ꢂhꢂ14
ꢀ11 ꢂkꢂ12
ꢀ27ꢂlꢂ27
24888
ꢀ20ꢂkꢂ20
ꢀ20ꢂlꢂ20
90974
reflns collected
independent reflns
data/restraints/parame-
ters
goodness-of-fit
R(int)
7197
12621
18246
7197/0/460
12621/0/564
18246/0/1163
1.019
1.026
1.035
0.0734
0.0616
0.0990
0.1555
0.1783
0.0442
0.0442
0.0619
0.1046
0.1119
0.0619
0.0572
0.0710
0.1587
Inspection of [1](PF6) and [1](ClO4)2 shows that the
monocation 1+ contains two distinctly different li-
R1[I>2s(I)]
R1 (all data)
wR (F2) [F2 >2s(F2)]
wR (F2) (all data)
0
Cꢀ
gands, assigned as QM and QM on the basis of the
0.1695
well-established[8,9,14,16] metrical parameters CꢀO, Cꢀ
N and CꢀC within the quinonoid ring. The dicationic
D1max; D1min [eꢁꢀ3
]
1.490; ꢀ0.783 3.293; ꢀ1.084
1.735; ꢀ0.606
[a] Crystals from CH2Cl2/CH3OH solution. The structure of an acetonitrile solvate with
slightly better crystal quality is given in Tables S1–S3 and Figure S1 in the Supporting
Information.
12+ contains two very similar ligands, identified as
0
QM
.
The DFT-optimized geometry of the neutral com-
plex 1 reproduces the experimental crystal structures
Electrochemistry and structures of oxidized forms
Compound 1 exhibits two reduction and two oxida-
tion waves in cyclic voltammetry experiments, shown
for ambient and low-temperature situations (see
Figure 2) and Table 3). The first reduction is com-
pletely reversible, whereas the second reduction pro-
cess is not fully reversible, as confirmed by UV-visi-
ble-NIR spectroelectrochemistry. The one-electron ox-
idation wave in the cyclic voltammogram (Figure 2) is
considerably distorted; this reflects the structurally
Scheme 3. Conversion of complexes 1n on electron transfer.
Figure 2. Cyclic voltammograms of 1 in CH2Cl2/Bu4NPF6 (0.1m) at 298 and
233 K (100 mVsꢀ1 scan rate).
Figure 3. Molecular structure of 1+ in the crystal of [1](PF6)ꢃCH2Cl2.
Chem. Eur. J. 2014, 20, 5414 – 5422
www.chemeurj.org
5416
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tronic configurations of the oxi-
dized species were examined in
order to find the lowest energy
state. The DFT calculations show
the 4A state of the octahedral
conformation of 1+ as the most
stable, The 4A state lies 0.05 eV
lower than the 2A state and
about 1 eV lower than the
planar 2A configuration. The 3A
state was found at the lowest
energy level of doubly oxidized
12+. Bonding parameters of the
lowest energy states are listed in
Table 2, indicating how the oxi-
dation and reduction influences
the structure of 1n.
Table 2. Selected bond lengths of complexes in the crystals[a] and DFT calculated selected bond lengths.
Selected distances [ꢁ]
1[b]
[1]+
[1]2+
[1]ꢀ
calcd
exptl
calcd
exptl
calcd
exptl
calcd
Ni(1)ꢀO(1)
Ni(1)ꢀO(2)
Ni(1)ꢀN(1)
Ni(1)ꢀN(2)
Ni(1)-S(1)
Ni(1)ꢀS(2)
O(1)-C(1)
C(1)ꢀC(2)
C(2)-C(3)
C(3)ꢀC(4)
C(4)-C(5)
C(5)ꢀC(6)
C(6)-C(1)
N(1)ꢀC(6)
O(2)ꢀC(30)
C(30)ꢀC(31)
C(31)ꢀC(32)
C(32)ꢀC(33)
C(33)-C(34)
C(34)ꢀC(35)
C(35)ꢀC(30)
N(2)ꢀC(35)
1.832(3)
1.831(3)
1.835(3)
1.839(3)
5.507(1)
5.767(1)
1.315(5)
1.413(5)
1.377(6)
1.430(6)
1.371(6)
1.414(5)
1.431(5)
1.353(5)
1.322(4)
1.417(5)
1.381(5)
1.424(5)
1.368(5)
1.418(5)
1.429(5)
1.351(5)
1.817
1.819
1.827
1.825
5.215
5.413
1.297
1.422
1.381
1.423
1.375
1.411
1.433
1.348
1.297
1.422
1.381
1.424
1.375
1.411
1.433
1.348
2.112 (2)
2.018(2)
2.058(2)
2.013(2)
2.3957(6)
2.3778(6)
1.239(3)
1.459(3)
1.351(3)
1.465(3)
1.344(3)
1.438(3)
1.516(3)
1.299(3)
1.295(2)
1.432(3)
1.372(3)
1.430(3)
1.364(3)
1.422(3)
1.454(3)
1.353(3)
2.089
2.002
2.102
2.014
2.426
2.422
1.232
1.459
1.356
1.458
1.354
1.435
1.511
1.298
1.285
1.433
1.376
1.427
1.369
1.418
1.456
1.345
2.087(2)
2.112(2)
2.056(3)
2.056(3)
2.356(1)
2.365(1)
1.235(4)
1.456(5)
1.350(5)
1.468(5)
1.347(5)
1.433(5)
1.514(5)
1.304(4)
1.240(4)
1.459(5)
1.343(5)
1.471(5)
1.347(5)
1.426(5)
1.512(4)
1.306(4)
2.058
2.058
2.087
2.087
2.405
2.405
1.239
1.452
1.359
1.460
1.356
1.432
1.512
1.303
1.239
1.452
1.359
1.460
1.356
1.432
1.512
1.303
1.829
1.833
1.841
1.837
5.063
5.370
1.314
1.412
1.394
1.405
1.388
1.402
1.421
1.368
1.314
1.412
1.395
1.405
1.389
1.402
1.422
1.366
EPR, magnetism, and UV-visi-
ble-NIR spectroelectrochemis-
try
The odd-electron species 1ꢀ and
1+ were generated for X-band
EPR spectroscopy by in situ elec-
trolysis (1ꢀ) or chemical oxida-
tion (AgClO4: 1+). The fully re-
[a] For crystal details see Table 1. [b] Crystals from CH2Cl2/CH3OH solution. The structure of an acetonitrile sol-
vate with slightly better crystal quality is given in Tables S1–S3 and Figure S1 in the Supporting Information.
Table 3. Electrochemical data[a] from cyclic voltammetry.
versible reduction of 1 produces an unresolved EPR signal at g
= 2.041 at room temperature and an axial g factor component
splitting in the frozen state at 110 K (Figure 5). The data
Process
1 (in CH2Cl2)
[1](PF6) (in CH3CN)
0.03 (E1/2
DE=72
)
0.06 (E1/2
DE=75
)
1+,12+
ꢀ0.28 (Epa1)
ꢀ0.09 (Epa2)
ꢀ0.38 (Epc)
ꢀ0.23 (Epa
)
1,1+
ꢀ0.33 (Epc)
ꢀ1.04 (E1/2
DE=70
)
ꢀ0.88 (E1/2
DE=64
)
1,1ꢀ
ꢀ1.87 (Epc)
ꢀ1.62 (E1/2
DE=101
)
1ꢀ,12ꢀ
ꢀ1.72 (Epa
)
[a] Potentials in V vs. Fc+/0 and DE in mV from cyclic voltammetry in
Bu4NPF6 (0.1m) solutions at 298 K, scan rate 100 mVsꢀ1
.
Figure 5. Experimental and simulated EPR spectrum of electrolytically gener-
ated 1ꢀ in CH2Cl2/Bu4NPF6 (0.1m) at 110 K.
(Table 4) suggest a predominantly organic radical, a semiqui-
none radical anion,[8,20] with non-negligible metal contribution,
which is responsible for the g anisotropy due to the spin-orbit
coupling contributions from nickel. Within a localized descrip-
tion, a [(QM )NiII(QM2ꢀ)]ꢀ situation can thus be postulated, with
Cꢀ
a low-spin NiII center bound by catecholate and with the resid-
ual spin (S=1= ) largely on one of the semiquinone ligands.
2
Figure 4. Molecular structure of 12+ in the crystal of [1](ClO4)2 ꢃ0.25CH2Cl2.
1.5ꢀ
The delocalized alternative with two equivalent ligands QM
is favored by DFT calculations (Table 2), however, it would also
I
Cꢀ
(Tables 2, and S2 and S3 in the Supporting Information). The
bond lengths of 1 are confirmed within 0.02 ꢁ (excluding the
nonbonding Ni–S distances). Several conformations and elec-
involve ligand-centered spin. Yet another formulation, [(Q )Ni-
ꢀ
[7]
Cꢀ
(Q )] , involving three spins, can be constructed; this must,
however, imply one antiferromagnetically coupled semiqui-
Chem. Eur. J. 2014, 20, 5414 – 5422
www.chemeurj.org
5417
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 4. EPR data for complex cation and anion.
[1]+
[1]ꢀ[a]
g1 (110 K)
g2
4.88
3.32
2.00
n.o.
n.o.
2.104
2.014
2.001
1.0
g3
A3(14N)[b]
giso (295 K)
2.041
[a] From in situ electrolysis of 1 in CH2Cl2/Bu4NPF6 (0.1m).[b] In mT.
Figure 7. Temperature dependence of the effective magnetic moment of
[1](ClO4).
none-metal pair in order to produce the observed ligand-cen-
tered spin. Such semiquinone-d9-semiquinone arrangements
have been reported for copper(II) species[7,8] with metal-based
spin for the planar system[8] and ligand-centered spin for
a twisted configuration (Scheme 2).[7]
The mono-oxidized form 1+ is EPR-silent at ambient temper-
ature, but displays a broad EPR signal with high g anisotropy
at 110 K (Figure 6, Table 4). Combined with the structural evi-
not obvious from the crystal structure of the PF6 salt; the situa-
tion may change with a different anion. Intermolecular antifer-
romagnetic exchange may also be the reason for the increase
of the effective magnetic moment with temperatures above
20 K; this should not occur with an S=3= ground state result-
2
ing from intramolecular ferromagnetic coupling, as suggested
by EPR and DFT calculations.
Further support for intramolecular ferromagnetic and inter-
molecular antiferromagnetic coupling in 1(ClO4) is provided by
the low-temperature (1.8 K) magnetization curve (Figure 8).
Figure 6. Experimental and simulated EPR spectrum of oxidatively (AgClO4)
generated 1+ in CH2Cl2 at 110 K.
dence for [(QM )NiII(QM0)]+, this result can be traced to an S=
Cꢀ
3
= situation (DFT: 4A state) as described earlier for related semi-
Figure 8. Field dependence of the magnetization of [1](ClO4).
2
quinonatonickel(II) and phenoxylnickel(II) species;[21] this is typ-
ical for a transition of a Kramers’ doublet with an S=3=
Compound 1(ClO4) has a magnetization value of 2.63 mB at 7 T,
2
ground state and zero-field splitting larger than the X-band mi-
crowave energy (ca. 0.3 cmꢀ1). Due to the odd electron
number, the ground and exited state split into Kramers’ dou-
blets, and owing to the zero-field splitting, there are only tran-
sitions possible between the doublet states. The g-values are
which is much higher than expected for an S=1= system, but
2
still lower than expected for S=3= . This deviation may again
2
be attributed to intermolecular antiferromagnetic exchange.
During reduction of 1, the added electron is accepted in the
ligand-based LUMO with contribution from Ni d orbitals,
whereas the oxidation is accompanied by a structural reorgani-
attributed to an effective spin S’=1= . The high-spin NiII config-
2
uration is favored by the neutral and weak[22] o-iminoquinone
zation. Figure 9 depicts the spin densities for 1ꢀ, 1+, and 12+
.
ligand and by the hexacoordination.
ADF/BP calculated spin densities on Ni are 0.216, 1.680, and
1.677 for reduced, oxidized, and doubly oxidized species, re-
spectively, in agreement with the oxidation and spin-state
combination shown in Scheme 4.
The temperature dependence of the magnetic moment of
1(ClO4) is shown in Figure 7. The value of meff(300 K)=4.55 mB
fits well with meff calculated for a noninteracting S=(1+1= )
2
system with g=2 (meff(s.o.)=4.56 mB; s.o.=spin only). On lower-
ing the temperature, the magnetic moment decreases to
a value of meff(20 K)=3.73 mB, that is, close to the spin-only
value for S=3= (meff(s.o.)=3.88 mB). On further lowering of the
2
temperature, the magnetic moment shows a pronounced de-
crease to meff(1.8 K)=2.59 mB, attributed to zero-field splitting,
which is also responsible for the inter-Kramers’ doublet transi-
tion detectable by EPR. An alternative explanation would be
intermolecular antiferromagnetic coupling, which, however, is
Figure 9. Spin density plots for 1ꢀ, 1+ and 12+ from left to right.
Chem. Eur. J. 2014, 20, 5414 – 5422
www.chemeurj.org
5418
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
for the neutral complex 1. The
essential features in the experi-
mental spectrum are well de-
scribed by the calculations
(Table 6). The assignment of indi-
vidual transitions based on the
frontier orbitals depicted in
Figure 11 indicates that the very
intense transition calculated at
Scheme 4. Conversion and spin arrangements in complexes 1n.
UV-visible-NIR spectroelectrochemistry with an optically
transparent thin-layer electrolysis (OTTLE) cell and diode array
detection (Figure 10) showed chemically reversible formation
of 1+ and then 12+ from 1, whereas the cathodic reduction
gave only 1ꢀ as a 100% regenerable product. The reversible
transition 1/1ꢀ resulted in the diminishing of the original
894 nm band and the emergence of an intense (e=
Table 5. Absorption maxima.[a]
lmax [nm] (10ꢀ3 ꢃe [Mꢀ1 cmꢀ1])
[1]2+ 891 (1.4), 525 (6.3), 431 sh (5.7), 387 (6.8), 295 sh (10.6), 227 (23.2)
[1]+ 893 (4.9), 721 sh (1.7), 490 (4.8), 385 (8.8), 280 sh (12.4), 227 (23.7)
[1]0 894 (26.0), 715 sh (3.2), 595 sh (1.5), 446 sh (1.8), 294 (16.7), 231
(25.7)
[1]ꢀ 1345 (10.4), 897 (3.4), 615 (1.4), 301 (14.6), 231 (25.0)
10400mꢀ1 cmꢀ1
) absorption at lmax =1345 nm (Figure 10,
Table 5). Table 6 lists the TD-DFT-calculated singlet transitions
[a] From spectroelectrochemistry in CH2Cl2/Bu4NPF6 (0.1m).
914 nm has mixed intraligand/ligand-to-metal charge-transfer
(IL/LMCT) character.[8,9,13,14] Calculations correctly reproduce the
shift of this intense transition to longer wavelengths (1345 nm)
after reduction (Figure S3 in the Supporting Information). The
transition calculated for 1ꢀ at 1289 nm is formed by an excita-
tion from the bHOMO to the bLUMO, those MOs correspond-
ing to HOMO and LUMO of the neutral complex (Figure 11).
The calculated excitation energies for the redox series of these
complexes (Figures S3 and S4) qualitatively reproduce the
spectral variations caused by redox changes. Based on TD-DFT
calculations and on the [(QM·ꢀ)NiII(QM2ꢀ)]ꢀ formulation a ligand-
to-ligand intervalence charge-transfer (LLIVCT) transition
p(QM2ꢀ)!p*(QM ) is postulated. The high absorptivity reflects
Cꢀ
optimum orbital overlap in a coplanar situation. The DFT ge-
ometry optimization for 1ꢀ (Table 2) and the Nernstian behav-
ior of the corresponding 1/1ꢀ wave support (Figure 2) the
notion of maintained planarity during that redox transition
with O,N-bidentate chelate ligands. Apparently, the addition of
an electron does not activate the thioether S donors towards
metal coordination.
In contrast, the removal of an electron in the first oxidation
of 1 to 1+ =[(QM )NiII(QM0)]+ does not produce an intense
Cꢀ
low-energy-shifted absorption (Figure 8), in agreement with
the non-coplanar arrangement of the redox-active p systems
of the two now O,N,S tridentate ligands in 1+. The crystal
structure (Figure 3, Table 2) shows a localized situation with
0
Cꢀ
one QM and one QM ligand. The second oxidation to 12+ re-
sults in further diminished NIR absorbance (Figure 10), whereas
typical[7–11] quinone transitions at about 500 nm become more
intense. TD-DFT calculations qualitatively describe the diminish-
ing intensity of the long-wavelength transition and the increas-
ing intensity of transitions around 500–600 nm for complex
ions 1+ and 12+ (Figure S4 in the Supporting Information). Fig-
ure S5 and S6 in the Supporting Information, which depict
MOs predominantly contributing to transitions around 550 nm
indicate IL quinone character of these transitions.
Figure 10. UV-visible-NIR spectroelectrochemistry of the steps 1+!12+
(top), 1!1+ (center), and 1!1ꢀ (bottom), each in CH2Cl2/Bu4NPF6 (0.1m).
Chem. Eur. J. 2014, 20, 5414 – 5422
www.chemeurj.org
5419
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 6. Selected G09/PBE0/PCM calculated lowest allowed TD-DFT singlet transitions for 1 with oscillator strengths larger than 0.005.
State
Main character [%]
Transition energies [eV]
(wavelengths [nm])
calcd
Oscillator strengths
calcd
lmax [nm]
(10ꢀ3ꢃe [Mꢀ1 cmꢀ1])
exptl
a1A
b1A
c1A
d1A
e1A
f1A
99 (HOMO!LUMO)
1.36 (914)
2.22 (557)
2.55 (485)
3.87 (319)
4.21 (294)
4.28 (290)
4.41 (281)
0.519
0.058
0.019
0.039
0.048
0.083
0.040
894 (26.2)
595 sh (1.5)
446 sh (1.8)
66 (HOMO-2!LUMO); 24 (HOMOꢀ4!LUMO)
87 (HOMOꢀ5!LUMO);
40 (HOMO!LUMO+2)
71 (HOMO-14!LUMO); 14 (HOMO!LUMO+5)
71 (HOMO!LUMO+5); 14 (HOMOꢀ14!LUMO)
mixed
294 (16.7)
g1A
a maximum at 800 to 900 nm accompanied by significant in-
tensity increase from 1100mꢀ1 cmꢀ1 (X=ClO4 in CH2Cl2) to
8200mꢀ1 cmꢀ1 (X=PF6 in CH3CN) suggests coordinative lability
in solution despite the seemingly stable configuration of
[1](PF6) in the crystal (Figure 3). We assume that the oxidatively
induced double S coordination in 1+ is not strong enough to
prevent partial dissociation; the dissociation is dependent on
counterions and solvents with different coordinative and die-
lectric properties.
Nevertheless, the results collected in this study can be sum-
marized in Schemes 4 and 5, showing predominantly ligand-
based electron transfer, but changed coordination situations
on the metal as a consequence of
the redox states of the noninno-
cent hemilabile ligands. The de-
scribed example of reversible intra-
molecular double oxidative addi-
tion, based not on a metal redox
process, but on ligand electron
transfer, complements previous re-
activity studies[5,23] where remote
electron transfer at noninnocent li-
gands has activated addition reac-
tions at the metal or at the ancil-
lary ligand. This present example
of nickel complexes thus comple-
Scheme 5. Assignment of oxi-
dation states in the redox
series of 1n.
ments previous related studies of
copper,[3] iridium,[6,24] ruthenium,
and rhodium compounds[24] and
may be extended to complexes of
other metals. Dinickel complexes of redox-active quinonoid
bischelate ligands, which have been well investigated by
Braunstein et al.,[25] might thus be modified accordingly in
order to add another dimension to the systems described
here.
Experimental Section
Figure 11. The representation of FMOs of 1.
General
Commercially available compounds were purchased from Aldrich
and used as received. Solvents for the complexes were dried with
standard Schlenk techniques. 2-(Methylthiomethyl)aniline was pre-
pared according to a reported procedure.[26] Purification was car-
A remarkable electrolyte effect was observed for the long-
wavelength (NIR) band of [1](X) in CH2Cl2 or CH3CN (Figure S7
and Table S5 in the Supporting Information). The shift from
Chem. Eur. J. 2014, 20, 5414 – 5422
www.chemeurj.org
5420
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ried out by column chromatography (silica gel, CH2Cl2/pentane,
1:1).
was recrystallized from hot acetonitrile. Yield: 122 mg (0.16 mmol,
32%). 1H NMR (250 MHz, CD2Cl2, 303 K): d (ppm)=1.03 (s, 18H,
CH3), 1.08 (s, 18H, CH3), 1.98 (s, 6H, SCH3), 3.70–4.38 (brm, 4H,
SCH2), 6.17–7.62 (brm, 12H, Ar-H and SQ-H); 1H NMR (400 MHz,
[D8]toluene, 223 K): d (ppm)=1.08 (s, 10.8H, CH3), 1.11 (s, 7.2H,
CH3), 1.27 (s, 7.2H, CH3), 1.31 (s, 10.8H, CH3), 1.64 (s, 2.4H, SCH3),
1.74 (s, 3.6H, SCH3), 3.87 (d, 1JHꢀH =14.0 Hz, 0.8H, SCH2), 3.93 (d,
1JHꢀH =14.0 Hz, 1.2H, SCH2), 4.46 (d, 1JHꢀH =14.0 Hz, 1.2H, SCH2),
4.54 (d, 1JHꢀH =14.0 Hz, 0.8H, SCH2), 6.64 (s, 1.2H, SQ-H), 6.67 (s,
Instrumentation
1H NMR spectra were obtained with a AV250 spectrometer from
Bruker. UV-visible-NIR spectra were recorded on
a Shimadzu
UV 160 spectrometer. UV-visible-NIR spectroelectrochemical studies
were performed in CH2Cl2/Bu4NPF6 (0.1m) at 298 K by using an op-
tically transparent thin layer electrochemical (OTTLE) cell[27a] in con-
nection with a J&M TIDAS spectrophotometer. The EPR measure-
ments were made with a Bruker EMX spectrometer (9.5 GHz), with
a two-electrode cell.[27b] Magnetic susceptibility measurements
were carried out with a Quantum Design MPMS SQUID magneto-
meter. Temperature-dependent magnetization between 1.8–350 K
was measured in 500 Oe external field. Experimental magnetic sus-
ceptibilities were corrected with the help of Pascal’s tables. Cyclic
voltammetric measurements were carried out with a M273A poten-
tiostat and a function generator M175 from EG&G. Platinum work-
ing and auxiliary electrodes and silver wire as pseudoreference
electrode were used in a three-electrode configuration. The sup-
porting electrolyte was Bu4NPF6 (0.1m) and the solute concentra-
tion was about 10ꢀ3 m. The ferrocenium/ferrocene couple Fc+/0
was used as internal reference.[28] Elemental analysis was carried
out with a Perkin–Elmer Analyzer 240. Electrospray mass spectra
were recorded on a Bruker Daltronics Microtof Q spectrometer. X-
ray diffraction was performed with a Bruker Kappa Apex II Duo
system. The structures were solved and refined by full-matrix least-
squares techniques on F2 by using the SHELX-97 program.[29] The
absorption corrections were done numerically or by the multiscan
technique. All data were corrected for Lorentz and polarization ef-
fects, and the non-hydrogen atoms were refined anisotropically.
Hydrogen atoms were included in the refinement process as per
the riding model.
0.8H, SQ-H), 7.07 (d, JHꢀH =7.74 Hz, 0.8H, Ar-H), 7.15 (d, JHꢀH
7.74 Hz, 1.2H, Ar-H), 7.20–7.26 (m, 4H, Ar-H and SQ-H), 7.35 (d,
HꢀH =7.74 Hz, 1.2H, Ar-H), 7.42 (d, JHꢀH =7.74 Hz, 0.8H, Ar-H), 7.47
=
J
(d, JHꢀH =7.74 Hz, 1.2H, Ar-H), 7.60 (d, JHꢀH =7.74 Hz, 0.8H, Ar-H);
elemental analysis calcd (%) for C44H58N2NiO2S2: C 68.65, H 7.59, N
3.64; found: C 68.66, H 7.65, N 3.67; HRMS (ESI): m/z: 768.3262
[M]+ (calcd m/z: 768.3288). Two different kinds of single crystals
for X ray diffraction were grown either out of a dichloromethane/
methanol mixture at room temperature or from acetonitrile at 08C.
ꢀ
Synthesis of [1]X (Xꢀ: PF6ꢀ, ClO4
)
CAUTION: PERCHLORATE SALTS CAN DECOMPOSE EXPLOSIVELY
AND SHOULD BE HANDLED WITH CARE. Compound 1 (38 mg,
0.05 mmol) and AgX (0.05 mmol) were dissolved in dry CH2Cl2
(5 mL) and stirred for one hour. The solution was filtered, concen-
trated and layered with dry n-hexane (10 mL). After filtration, the
ꢀ
solid was dried in vacuo. Elemental analysis calcd (%) for [1]ClO4
(C44H58ClN2NiO6S2 ꢃ0.6 CH2Cl2): C 58.01, H 6.47, N 3.03; found: C
58.04, H 6.40, N 3.04. Single crystals could be obtained through
slow diffusion of hexane into a dichloromethane solution of the
hexafluorophosphate complex in the presence of a silver wire.
Synthesis of [1](ClO4)2
Compound 1 (77 mg, 0.1 mmol) and AgClO4 (45 mg, 0.2 mmol)
were dissolved in dry CH2Cl2 (8 mL) and stirred for five hours. The
solution was filtered, concentrated, and layered with dry n-hexane
(10 mL). After filtration, the solid was dried in vacuo. Elemental
analysis calcd (%) for C44H58Cl2N2NiO10S2: C 54.56, H 6.04, N 2.89;
found: C 54.42, H 6.10, N 2.95. Cyclic voltammetry showed essen-
tially the same response as that of [1](ClO4), the UV-visible-NIR
spectrum represents that observed by spectroelectrochemistry of
1 (see Figure 8). Susceptibility measurements up to 300 K revealed
values <2.3 mB. Single crystals were obtained from [1](ClO4)
through disproportionation, which takes place during the slow dif-
fusion of hexane into a dichloromethane solution of the monocat-
ion in the absence of a silver wire.
Synthesis of 4,6-di-tert-butyl-2-(2-methylthiomethylphenyl)-
aminophenol
2-(Methylthiomethyl)aniline (3.06 g, 20.0 mmol) and 3,5-di-tert-bu-
tylcatechol (4.45 g, 20.0 mmol) were refluxed for eight hours in dry
n-hexane (60 mL) with NEt3 (0.8 mL) under an argon atmosphere.
The solvent was removed at 708C and the oily residue was dis-
solved in n-hexane (2 mL). After seven days at ꢀ48C the solvent
was removed and the brown oil was subjected to chromatography
on a silica gel column with diethylether/n-pentane (5:95) as eluent.
The desired product was found in the second fraction. Yield: 3.4 g
1
(10 mmol, 50%). H NMR (250 MHz, CDCl3, 303 K): d (ppm)=1.28 (s,
9H, CH3), 1.44 (s, 9H, CH3), 2.06 (s, 3H, SCH3), 3.83 (s, 2H, SCH2),
5.93 (brs, 1H, OH), 6.25 (brs, 1H, NH), 6.46–6.50 (m, 1H, Ar-H), 6.76
ꢀ6.82 (m, 1H, Ar-H), 7.01 (d, 4JHꢀH =2.3 Hz, 1H, Cat-H), 7.06–7.13
Quantum chemical calculations
The electronic structures of 1n =[Ni(QM)2]n (n=ꢀ1, 0, 1, 2) were cal-
culated by DFT methods by using the Gaussian 09[30] and Amster-
dam Density Functional (ADF2013.01)[31,32] program packages. G09
calculations employed the Perdew, Burke, Ernzerhof[33,34] PBE0
hybrid functional (G09/PBE0). The geometry of the open-shell sys-
tems was calculated by the UKS approach. Geometry optimization
was followed by vibrational analysis in order to characterize sta-
tionary states and enumerate free energies DG. For H and C atoms
polarized triple-z basis sets 6–311G(d), together with polarized
triple-z basis sets 6–311G(3df) for N, O, S, and Ni were used.[35] Elec-
tronic excitations were calculated by TD-DFT with polarizable con-
tinuum model (PCM)[36] solvent correction.
4
(m, 2H, Ar-H), 7.22 (d, JHꢀH =2.3 Hz, 1H, Cat-H); HRMS (ESI): m/z:
380.2011 [M+Na]+ (calcd m/z: 380.2019).
Synthesis of bis(4,6-di-tert-butyl-N-(2-methylthiomethylphe-
nyl)iminosemiquinonato)-nickel(II) (1)
4,6-Di-tert-butyl-2-((2-methylthiomethylphenyl)amino)-phenol
(358 mg, 1.00 mmol) and NiCl2 ꢃ6 H2O (119 mg, 0.50 mmol) were
refluxed with NEt3 (0.28 mL) in dry acetonitrile (10 mL) for two
hours under an atmosphere of argon. The light green solid was fil-
tered and suspended in CHCl3. After 10 min, the dark-green solu-
tion was filtered and the solvent was removed under reduced pres-
sure. The residue was dissolved in n-pentane and filtered to
remove HNEt3Cl. After removal of the solvent, the dark-green solid
Within the ADF program, Slater type orbital (STO) basis sets of
triple-z quality with two polarization functions for Ni, C, N, O, and
Chem. Eur. J. 2014, 20, 5414 – 5422
www.chemeurj.org
5421
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[14] a) K. Chlopek, E. Bothe, F. Neese, T. Weyhermꢅller, K. Wieghardt, Inorg.
Chem. 2006, 45, 6298; b) K. Ray, T. Petrenko, K. Wieghardt, F. Neese,
Dalton Trans. 2007, 1552.
[15] M. Abe, Chem. Rev. 2013, 113, 7011.
[16] a) S. Bhattacharya, P. Gupta, F. Basuli, Inorg. Chem. 2002, 41, 5810;
b) S. N. Brown, Inorg. Chem. 2012, 51, 1251.
[17] a) S. Ye, Ph. D. Thesis, University of Stuttgart, 2005; b) R. Hꢅbner, Ph. D.
Thesis, University of Stuttgart, 2011.
[18] A. Mukherjee, R. Mukherjee, Indian J. Chem. 2011, 50A, 484.
S atoms and double-z with one polarization function for H atom
were employed. Inner shells were represented by the frozen core
approximation (1s for C, N, and O; 2p for S and Ni were kept
frozen). Within ADF the functional including Becke’s gradient cor-
rection to the local exchange expression in conjunction with Per-
dew’s gradient correction to local density approximation (LDA)
with VWN parameterization of electron gas data was used (ADF/
BP86).[37,38]
ˇ
[19] A. Knçdler, W. Kaim, V. K. Jain, S. Zꢀlis, J. Organomet. Chem. 2002, 655,
218.
Acknowledgements
[20] W. Kaim, Coord. Chem. Rev. 1987, 76, 187.
[21] a) C. Benelli, A. Dei, D. Gatteschi, L. Pardi, Inorg. Chem. 1988, 27, 2831;
b) K. Sik Min, T. Weyhermꢅller, K. Wieghardt, Dalton Trans. 2003, 1126;
c) O. Rotthaus, F. Thomas, O. Jarjayes, C. Philouze, E. Saint-Aman, J.-L.
Pierre, Chem. Eur. J. 2006, 12, 6953.
We thank Dr. Wolfgang Frey for crystallographic data collection
and Dr. Lapo Bogani and Prof. Dr. Martin Dressel at the Physi-
kalisches Institut I der Universitꢄt Stuttgart for the use of the
SQUID instrument and Dr. Ralph Hꢅbner for valuable discus-
sions. Support from the COST programme CM1202 of the EU
and the Land Baden-Wꢅrttemberg is gratefully acknowledged.
S.Z. thanks the Ministry of Education of the Czech Republic
(grant LD11086) for support.
ˇ
[22] S. Roy, B. Sarkar, D. Bubrin, M. Niemeyer, S. Zalis, G. K. Lahiri, W. Kaim, J.
Am. Chem. Soc. 2008, 130, 15230.
[23] a) J. L. Boyer, T. R. Cundari, N. J. DeYonker, T. B. Rauchfuss, S. R. Wilson,
Inorg. Chem. 2009, 48, 638; b) K. J. Blackmore, J. W. Ziller, A. F. Heyduk,
Inorg. Chem. 2005, 44, 5559.
[24] W. Kaim, M. Bubrin, R. Hꢅbner, in Advances in Organometallic Chemistry
and Catalysis (Ed.: A. J. L. Pombeiro), Wiley, Hoboken, 2014, pp. 667.
[25] a) Q.-Z. Yang, A. Kermagoret, M. Agostinho, O. Siri, P. Braunstein, Orga-
nometallics 2006, 25, 5518; b) O. Siri, J.-P. Taquet, J.-P. Collin, M.-M.
Rohmer, M. Bꢆnard, P. Braunstein, Chem. Eur. J. 2005, 11, 7247.
[26] J. P. Chupp, T. M. Balthazor, M. J. Miller, M. J. Pozzo, J. Org. Chem. 1984,
49, 4711.
Keywords: coordination change · hemilabile ligands · nickel ·
noninnocent ligands · redox chemistry
ˇ
[27] a) M. Krejcik, M. Danek, F. Hartl, J. Electroanal. Chem. Interfacial Electro-
[1] a) J. C. Jeffrey, T. B. Rauchfuss, Inorg. Chem. 1979, 18, 2658; b) P. Braun-
stein, F. Naud, Angew. Chem. 2001, 113, 702; Angew. Chem. Int. Ed. 2001,
40, 680; c) D. Steinborn, Fundamentals of Organometallic Catalysis,
Wiley-VCH, Weinheim, 2012, p. 155.
[2] C. W. Machan, A. M. Lifschitz, C. L. Stern, A. A. Sarjeant, C. A. Mirkin,
Angew. Chem. 2012, 124, 1498; Angew. Chem. Int. Ed. 2012, 51, 1469
and references cited.
chem. 1991, 317, 179; b) W. Kaim, S. Ernst, V. Kasack, J. Am. Chem. Soc.
1990, 112, 173.
[28] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877.
[29] G. M. Sheldrick, Acta Crystallogr. 2008, A64, 112.
[30] Gaussian 09, Revision C.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Ko-
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyen-
gar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cio-
slowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
[31] G. te Velde, F. M. Bickelhaupt, S. J. A. van Gisbergen, C. Fonseca Guerra,
E. J. Baerends, J. G. Snijders, T. Ziegler, J. Comput. Chem. 2001, 22, 931.
[32] ADF2013.01, SCM, Theoretical Chemistry, Vrije Universiteit, Amsterdam,
The Netherlands, http://www.scm.com.
[33] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865.
[34] C. Adamo, V. Barone, J. Chem. Phys. 1999, 110, 6158.
[35] L. A. Curtiss, M. P. McGrath, J.-P. Blaudeau, N. E. Davis, R. C. Binning Jr., L.
Radom, J. Chem. Phys. 1995, 103, 6104.
[3] J. Schnçdt, J. Manzur, A.-M. Garcia, I. Hartenbach, C.-Y. Su, J. Fiedler, W.
Kaim, Eur. J. Inorg. Chem. 2011, 1436.
[4] a) Forum on Redox-Active Ligands, Inorg. Chem. 2011, 50, 9737–9914;
b) Special Issue: Cooperative and Redox Non-Innocent Ligands in Di-
recting Organometallic Reactivity, Eur. J. Inorg. Chem. 2012, 340–580;
c) V. K. K. Praneeth, M. R. Ringenberg, T. R. Ward, Angew. Chem. 2012,
124, 10374; Angew. Chem. Int. Ed. 2012, 51, 10228; d) K. G. Caulton, Eur.
J. Inorg. Chem. 2012, 435.
[5] a) M. R. Ringenberg, S. L. Kokatam, Z. M. Heiden, T. B. Rauchfuss, J. Am.
Chem. Soc. 2008, 130, 788; b) M. R. Ringenberg, M. J. Nilges, T. B. Rauch-
fuss, S. R. Wilson, Organometallics 2010, 29, 1956; c) M. R. Ringenberg,
T. B. Rauchfuss, Eur. J. Inorg. Chem. 2012, 490.
ˇ
[6] R. Hꢅbner, S. Weber, S. Strobel, B. Sarkar, S. Zꢀlis, W. Kaim, Organometal-
lics 2011, 30, 1414.
[7] S. Ye, B. Sarkar, F. Lissner, Th. Schleid, J. van Slageren, J. Fiedler, W. Kaim,
Angew. Chem. 2005, 117, 2140; Angew. Chem. Int. Ed. 2005, 44, 2103.
[8] P. Chaudhuri, C. N. Verdani, E. Bill, E. Bothe, T. Weyhermꢅller, K. Wie-
ghardt, J. Am. Chem. Soc. 2001, 123, 2213.
[9] V. Bachler, G. Olbrich, F. Neese, K. Wieghardt, Inorg. Chem. 2002, 41,
4179.
[10] a) X. Sun, H. Chun, K. Hildenbrand, E. Bothe, T. Weyhermꢅller, F. Neese,
K. Wieghardt, Inorg. Chem. 2002, 41, 4295; b) E. Bill, E. Bothe, P. Chaud-
huri, K. Chlopek, D. Herebian, S. Kokatam, K. Ray, T. Weyhermꢅller, F.
Neese, K. Wieghardt, Chem. Eur. J. 2005, 11, 204.
[36] M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 2003, 24,
669.
[37] A. D. Becke, Phys. Rev. A 1988, 38, 3098.
[38] J. P. Perdew, Phys. Rev. B 1986, 33, 8822.
[11] A. I. Poddel’sky, V. K. Cherkasov, G. A. Abakumov, Coord. Chem. Rev.
2009, 253, 291.
[12] a) A. L. Balch, F. Rçhrscheid, R. H. Holm, J. Am. Chem. Soc. 1965, 87,
2301; b) A. L. Balch, R. H. Holm, J. Am. Chem. Soc. 1966, 88, 5201.
[13] a) D. Herebian, E. Bothe, F. Neese, T. Weyhermꢅller, K. Wieghardt, J. Am.
Chem. Soc. 2003, 125, 9116; b) D. Herebian, K. E. Wieghardt, F. Neese, J.
Am. Chem. Soc. 2003, 125, 10997.
Received: November 4, 2013
Published online on March 24, 2014
Chem. Eur. J. 2014, 20, 5414 – 5422
www.chemeurj.org
5422
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim