Electronic structures of octahedral Ni(II) complexes with click derived triazole ligands: A combined structural, magnetometric, spectroscopic, and theoretical study

Article abstract of DOI:10.1021/ic3026123

The coordination complexes of Ni(II) with the tripodal ligands tpta (tris[(1-phenyl-1H-1,2,3-triazol-4-yl)methyl]amine), tbta ([(1-benzyl-1H-1,2,3- triazol-4-yl)methyl]amine), and tdta (tris[(1-(2,6-diisopropyl-phenyl)-1H-1,2,3- triazol-4-yl)methyl]amine) and the bidentate ligand pyta (1-(2,6- diisopropylphenyl)-4-(2-pyridyl)-1,2,3-triazole), [Ni(tpta)₂](BF ₄)₂ (1), [Ni(tbta)₂](BF₄) ₂ (2), [Ni(tdta)₂](BF₄)₂ (3), and [Ni(pyta)₃](BF₄)₂ (4), were synthesized from Ni(BF₄)₂·6H₂O and the corresponding ligands. Complexes 2 and 4 were also characterized structurally using X-ray diffraction and magnetically via susceptibility measurements. Structural characterization of 2 that contains the potentially tetradentate, tripodal tbta ligand revealed that the Ni(II) center in that complex is in a distorted octahedral environment, being surrounded by two of the tripodal ligands. Each of those ligands coordinate to the Ni(II) center through the central amine nitrogen atom and two of the triazole nitrogen donors; the Ni-N(amine) distances being longer than Ni-N(triazole) distances. In case of 4, three of the bidentate ligands pyta bind to the Ni(II) center with the binding of the triazole nitrogen atoms being stronger than those of the pyridine. Temperature dependent susceptibility measurements on 2 and 4 revealed a room temperature χ_MT value of 1.18 and 1.20 cm³ K mol^-1, respectively, indicative of S = 1 systems. High-frequency and -field EPR (HFEPR) measurements were performed on all the complexes to accurately determine their g-tensors and the all-important zero-field splitting (zfs) parameters D and E. Interpretation of the optical d-d absorption spectra using ligand field theory revealed the B and Dq values for these complexes. Quantum chemical calculations based on the X-ray and DFT optimized geometries and their ligand field analysis have been used to characterize the metal-ligand bonding and its influence on the magnitude and sign of the zfs parameters. This is the first time that such extensive HFEPR, LFT, and advanced computational studies are being reported on a series of mononuclear, distorted octahedral Ni(II) complexes containing different kinds of nitrogen donating ligands in the same complex.

Full text of DOI:10.1021/ic3026123

Article
pubs.acs.org/IC
Electronic Structures of Octahedral Ni(II) Complexes with “Click”
Derived Triazole Ligands: A Combined Structural, Magnetometric,
Spectroscopic, and Theoretical Study
David Schweinfurth, J. Krzystek,*,∫ Igor Schapiro,§ Serhiy Demeshko,∥ Johannes Klein,ζ
Joshua Telser,*,# Andrew Ozarowski,∫ Cheng-Yong Su,¶ Franc Meyer,∥ Mihail Atanasov,*,§,$
Frank Neese,*,§ and Biprajit Sarkar*,
Institut fur Chemie und Biochemie, Freie Universitat Berlin, Fabeckstraße 34-36, D-14195 Berlin, Germany
̈
̈
∫ National High Magnetic Field Laboratory, Florida State University, Tallahassee, Florida 32310, United States
§Max Planck Institute for Chemical Energy Conversion, Stiftstraße 34-36, D-45470 Mulheim an der Ruhr, Germany
̈
∥Institut fur Anorganische Chemie, Georg-August Universitat Gottingen, Tammanstraße 4, D-37077 Gottingen, Germany
̈
̈
̈
̈
ζInstitut fur Anorganische Chemie, Universitat Stuttgart, Pfaffenwaldring 55, D-70569 Stuttgart, Germany
̈
̈
#Department of Biological, Chemical and Physical Sciences, Roosevelt University, Chicago, Illinois 60605, United States
¶Lehn Institute of Functional Materials, School of Chemistry and Chemical Engineering, Sun Yat-Sen University, Guangzhou 510275,
China
$Institute of General and Inorganic Chemistry, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria
S
* Supporting Information
ABSTRACT: The coordination complexes of Ni(II) with the
tripodal ligands tpta (tris[(1-phenyl-1H-1,2,3-triazol-4-yl)-
methyl]amine), tbta ([(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]-
amine), and tdta (tris[(1-(2,6-diisopropyl-phenyl)-1H-1,2,3-
triazol-4-yl)methyl]amine) and the bidentate ligand pyta (1-
(2,6-diisopropylphenyl)-4-(2-pyridyl)-1,2,3-triazole), [Ni-
(tpta)2](BF4)2 (1), [Ni(tbta)2](BF4)2 (2), [Ni(tdta)2](BF4)2
(3), and [Ni(pyta)3](BF4)2 (4), were synthesized from
Ni(BF4)2·6H2O and the corresponding ligands. Complexes 2
and 4 were also characterized structurally using X-ray
diffraction and magnetically via susceptibility measurements.
Structural characterization of 2 that contains the potentially tetradentate, tripodal tbta ligand revealed that the Ni(II) center in
that complex is in a distorted octahedral environment, being surrounded by two of the tripodal ligands. Each of those ligands
coordinate to the Ni(II) center through the central amine nitrogen atom and two of the triazole nitrogen donors; the Ni−
N(amine) distances being longer than Ni−N(triazole) distances. In case of 4, three of the bidentate ligands pyta bind to the
Ni(II) center with the binding of the triazole nitrogen atoms being stronger than those of the pyridine. Temperature dependent
susceptibility measurements on 2 and 4 revealed a room temperature χMT value of 1.18 and 1.20 cm3 K mol−1, respectively,
indicative of S = 1 systems. High-frequency and -field EPR (HFEPR) measurements were performed on all the complexes to
accurately determine their g-tensors and the all-important zero-field splitting (zfs) parameters D and E. Interpretation of the
optical d−d absorption spectra using ligand field theory revealed the B and Dq values for these complexes. Quantum chemical
calculations based on the X-ray and DFT optimized geometries and their ligand field analysis have been used to characterize the
metal−ligand bonding and its influence on the magnitude and sign of the zfs parameters. This is the first time that such extensive
HFEPR, LFT, and advanced computational studies are being reported on a series of mononuclear, distorted octahedral Ni(II)
complexes containing different kinds of nitrogen donating ligands in the same complex.
screening such metal complexes in various fields such as
electron transfer,3 photochemistry,4 magnetism,5 antitumor
agents,6 supramolecular chemistry,7 and catalysis.8 We have
been utilizing metal complexes of “click” derived ligands for
INTRODUCTION
■
The Cu(I) catalyzed cycloaddition reaction between azides and
alkynes, a so-called “click” reaction, has established itself as an
extremely powerful synthetic tool in organic chemistry.1
Coordination chemists have taken advantage of this highly
appealing reaction to design ligands for synthesizing a variety of
metal complexes.2 Several groups have undertaken projects of
Received: November 29, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
understanding their fundamental properties and a variety of
applications.9 One field that is our focus at the moment is the
magnetic properties of complexes of these ligands with 3d
metal centers. We have thus shown that Co(II) complexes of
tripodal triazole ligands can undergo temperature dependent
spin crossover.10a Recently, we have reported on the catalytic
poly- and oligomerization reactions with Fe(II), Co(II), and
Ni(II) complexes of tripodal10b as well as bidentate10c “click”
derived triazole ligands and have investigated their geometric
and electronic structures.
meridional-N3N′3 (in 4). Single crystal X-ray diffraction studies
are presented on representative complexes 2 and 4 to
determine the structural features and bonding anisotropies
around the metal centers. SQUID magnetometric measure-
ments are presented for the same two complexes to gather
information about the bulk magnetic properties of the
substances. HFEPR studies on these open shell S = 1 systems
are reported, and spin Hamiltonian parameters are determined
by that method. Ligand field theory is used to gain information
about B and Dq parameters from the energies of the observed
d−d transitions. Quantum chemical calculations and their
ligand field (angular overlap model) analysis are used to gain
insight into the origin of the metal−ligand bonding and its
effect on the zfs in these complexes.
Synthesizing new kinds of metal complexes and the complete
understanding of their geometric and electronic structures is an
all important goal in coordination chemistry. It is only through
the development of innovative synthetic tools, as well as the
structural understanding of metal complexes, that a rational
basis for their application in various fields of chemistry can be
achieved. High-frequency and -field electron paramagnetic
resonance (HFEPR) spectroscopy has established itself as an
extremely powerful tool for elucidating the electronic structures
of metal complexes that are not amenable to conventional EPR
techniques.11 Traditionally, paramagnetic metal complexes
containing more than one unpaired electron such as
mononuclear tetrahedral or octahedral Ni(II) complexes have
been investigated by magnetometry.12 In these complexes the
operation of zero-field splitting (zfs) often shifts the resonances
outside the window available in standard EPR measurements.13
Thus this type of complexes has been investigated with HFEPR
spectroscopy, including model complexes for Ni-containing
enzymes.14 Octahedral Ni(II) complexes (S = 1) with
RESULTS AND DISCUSSION
■
Synthesis and Characterization. The ligands tbta and
tpta were synthesized by reported methods18 (Figure 1), and
14i
N4O2,14h,k,15d N3O3,14l and N2O4 donor sets have been
Figure 1. Ligands used in this work.
subjected to such investigations. Very recently HFEPR studies
have been reported also on Ni(II) complexes with a O4E2 (E =
S or Se) donor set.14m To the best of our knowledge, reports
on HFEPR spectroscopy on mononuclear Ni(II) complexes
containing a N6 coordination sphere are restricted to
homoleptic complexes (where all six N donors are identical)
and such reports number only a few.14a,c
Ligand field theory (LFT) has been broadly used as a tool to
interpret spectral and magnetic phenomena in extended solids
and molecular complexes of transition metals. Nowadays this
model has been widely supplemented by advanced first
principle density functional theory (DFT) and ab initio
multireference electronic structure theory. In particular the
latter theory was successfully applied to predict and interpret
electronic multiplet structures and zfs tensors of transition
metal 3d-complexes of unprecedented size using a well
documented computational protocol as implemented in the
program ORCA.15 A one-to-one mapping procedure of the
general nonadditive ligand field parametrization using ab initio
wave functions allows to extract ligand field parameters from
state average complete active space self consistent field (SA-
CASSCF)16 wave functions and from N-electron valence
second order perturbation theory (NEVPT2).17
Herein we present the synthesis of four new complexes of
Ni(II) with the tripodal ligands tpta (tris[(1-phenyl-1H-1,2,3-
triazol-4-yl)methyl]amine), tbta ([(1-benzyl-1H-1,2,3-triazol-4-
yl)methyl]amine), and tdta (tris[(1-(2,6-diisopropyl-phenyl)-
1H-1,2,3-triazol-4-yl)methyl]amine) and the bidentate ligand
pyta (1-(2,6-diisopropylphenyl)-4-(2-pyridyl)-1,2,3-triazole)),
[Ni(tpta)2](BF4)2 (1), [Ni(tbta)2](BF4)2 (2), [Ni(tdta)2]-
(BF4)2 (3), and [Ni(pyta)3](BF4)2 (4), from Ni(BF4)2·6H2O
and the corresponding ligands. The donor sets in these
complexes can be described as trans-N4N′2 (in 1−3) and
the new ligands tdta and pyta were synthesized by a modified
Cu(I) catalyzed “click” protocol. The modifications were
undertaken to simplify the purification steps for this ligand
(see Experimental Section).
The metal complexes were synthesized by reaction of
Ni(BF4)2·6H2O with the respective ligands for 1 h under
reflux (see Experimental Section). Cooling of the reaction
mixture to room temperature and filtration delivered good
product yields after recrystallization. The purity of the
complexes and their composition was verified by using
elemental analysis and mass spectrometry (peaks correspond-
ing to m/z values of [M]2+ ions were observed (see
Experimental Section)).
X-ray Diffraction Studies. Single crystals of 2 and 4
suitable for X-ray diffraction studies were grown by slow
diffusion of diethyl ether into acetonitrile and ethanol solutions,
respectively, at ambient temperatures. The complexes 2 and 4
crystallize in the triclinic P1 and monoclinic C2/c space groups,
̅
respectively (Table 1). In both 2 and 4, the Ni(II) center is in a
distorted octahedral environment, coordinated in the case of 2
by four triazole (Nta) and two trans amine (Nam) nitrogen
donors (trans-Nta Nam coordination mode, Figure 2) and in
4
2
the case of 4 by pyridine (Npy) and triazole N-donors in a
meridional arrangement (mer-Nta , Npy3, Figure 3).
3
In the centrosymmetric complex with the tripodal ligand
tbta, 2, the nickel center is coordinated by each tbta ligand
through the nitrogen atoms N11 and N21 of two 1,2,3-triazole
rings and the central amine nitrogen donors (N1), while the
third triazole ring of each tbta ligand remains uncoordinated
(Figure 2). The four nitrogen donors from the triazole rings
thus make up the equatorial coordination at the nickel center
B
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Crystallographic Data for 2 and 4
2
4
formula
Mr
C60H60B2F8N20Ni
1293.61
C57H66B2F8N12Ni
1151.55
Monoclinic
C2/c
crystal system
space group
a (Å)
triclinic
P1
̅
10.171(1)
11.408(3)
13.979(3)
93.79(2)
105.24(1)
94.67(1)
1553.2(6)
1
43.564(5)
15.497(2)
18.910(2)
90
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
Z
105.320(7)
90
12313(2)
8
Dcalc (g cm−3
)
1.383
1.242
T (K)
μ (mm−1
173(2)
0.394
100(2)
)
0.385
Mo Kα (nm)
F(000)
0.71073
670
0.71073
4816
meas/ indep refl
obsvd [I > 2σ(I)] refl
R(int)
6435/6077
4599
29659/10743
5060
0.0399
0.1207
R[F2 > 2σ(F2)]
wR (F2)
0.0390
0.0898
−
0.1032
0.2426
Figure 3. ORTEP plot of 4. The BF4 ions are left out for clarity.
S
0.863
1.091
Ellipsoids are drawn at 50% probability.
Δρmax, Δρmin (e Å−3
)
0.329, −0.529
1.150, −0.522
∠NamNiNta (α, distortion denoted by δα) from 90° (see Table
2). These differences in bond lengths and bond angles thus
introduce bonding strain into the structure and reduce the
symmetry around the Ni(II) from Oh to Ci. As will be shown
this has a strong impact on the magnetic and spectral behavior.
The benzyl substituents of the uncoordinated 1,2,3-triazole
rings of the tbta ligands participate in strong intramolecular C−
H···π interactions with the benzyl substituents of a coordinated
triazole ring (Figure 2). The distance between the C−H proton
of one benzyl ring and the center of the phenyl ring of the other
benzyl group is 2.7271(7) Å. The formation of such strong
noncovalent interactions within the second coordination sphere
of the metal center is a trademark of the tbta ligand in its
bis(ligand)-complexes. We have recently shown that these
noncovalent interactions are responsible for magnetic bistability
in Co(II) complexes of tbta and will show here that these
interactions affect the sign of the zero-field splitting.10a
Despite several attempts, we were unable to produce suitable
−
crystals of complex 1 with BF4 as counteranion. Attempts at
crystallizing a corresponding [Ni(tpta)2](ClO4)2 analog (1′)
were somewhat successful, but the quality of the X-ray
diffraction data is not good enough to provide accurate bond
lengths in that complex. To this end, DFT geometry
optimizations were carried out on the entire 1−4 series (see
Table 2 for a list of relevant Ni−N bond distances and N−Ni−
N bond angles). The coordination geometry of 1 is very close
to that of 2; similar to 2, the nickel center in 1′ is also
coordinated through triazole nitrogen atoms in the equatorial
plane and through amine nitrogen atoms in the axial direction.
One triazole ring from each tpta ligand remains uncoordinated
in 1′, just as is the case of 2 with tbta. Thus the Ni(II) center in
1′ is observed (and also computed) to have the lowest
symmetry (C1). Furthermore, the direct phenyl substituents of
tpta are unlikely to form C−H···π interactions of the kind
discussed above for 2, where the more flexible benzyl groups
make such interactions possible.
−
Figure 2. ORTEP plot of 2. The BF4 ions are left out for clarity.
Ellipsoids are drawn at 50% probability.
with Ni−Nta distances of 2.040(2) and 2.071(2) Å (Table 2).
In contrast, the Ni−Nam distance is much longer with 2.242(2)
Å. As will be shown by LF analysis (vide infra), the amine
nitrogen atoms are poorer donors than the triazole nitrogen
atoms, and this is reflected in the larger Ni−Nam distances
compared to the Ni−Nta distances. DFT geometry optimization
using a polarizable continuum model (COSMO) reproduces
reasonably well the X-ray geometries (see Table 2). In addition
to the bond length anisotropy, there is a significant angular
distortion of the average value of the axial−equatorial angle,
C
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. Selected Bond Lengths (in Å) and Equatorial−Axial Angles α and δα (in deg.) for Complexes 2 and 4 (from X-ray
Data) and from DFT Geometry Optimizations (for 1 to 4)
1
2
3
4
structural parameter
DFT
exp
DFT
2.224
DFT
structural parameter
exp
DFT
2.034
Ni−Nam(ax)
2.204
2.209
2.027
2.091
2.050
2.066
2.058
90.1
2.242(2)
2.242(2)
2.040(2)
2.040(2)
2.071(2)
2.071(2)
2.056
2.237
Ni−Nta(+z)
2.021(5)
2.033(5)
2.068(5)
2.111(5)
2.070(6)
2.101(6)
2.088
Ni−Nta(eq)
2.224
2.019
2.019
2.094
2.094
2.056
2.236
2.072
2.041
2.068
2.076
2.064
Ni−Nta(−z)
2.040
2.059
2.083
2.088
2.115
2.086
Ni−Nta(+x)
Ni−Npy(−x)
Ni−Npy(+y)
Ni−Npy(−y)
Ni−Nta,py(eq) average
Ni−Nta(eq) average
αav
90.0
90.0
90.0
90.1
90.1
δα
−12.0
+13.8
−9.7
+9.4
−10.5
+10.5
−13.1
13.3
−11.1
+7.8
−10.8
+8.3
Similar to complex 1 we could not obtain single crystals of 3
appropriate for X-ray diffraction. DFT optimizations of this
complex and the comparison of bond lengths and angles (Table
2) show that the coordination of Ni(II) in 3 is very similar to 1
and 2. This is also supported by the zfs tensor parameters from
analysis of the HFEPR (vide infra).
The nickel center in complex 4 is coordinated by three of the
bidentate pyridyl-triazole ligands, pyta (Figure 3). The three
triazole N donors take up meridional positions around the
nickel centers, as do the pyridyl N donors.
Unlike 2 and similar to 1, the Ni(II) complex in 4 does not
contain a center of symmetry. The three principal bonding axes
of the octahedron in 4 are also all different. This is a
consequence of the meridional coordination mentioned above,
together with the bis-chelating nature of the pyta ligand. Thus
one of the bonding axis has two triazole nitrogens trans to each
other (N13−Ni1−N17, Figure 3), the second one has two
pyridine nitrogens trans to each other (N18−Ni1−N22), and
the final one has a pyridine and a triazole nitrogen trans to each
other (N20−Ni1−N23). On average the Ni−Nta bond lengths
are shorter than the Ni−Npy bond lengths (Table 2). Similar
observations have been made by us and others for metal
complexes with related pyridyl-triazole ligands.7a,9a,b As
supported by the ligand field analysis of the ab initio results,
this trend reflects the slightly larger donor ability of the triazole
nitrogen atom as compared to the pyridine nitrogen atom.
Thus the three different bonding axes make the coordination
around the nickel center in 4 less anisotropic than that in 1−3.
The distortion around the Ni(II) center in 4 is also apparent
from the angles between the metal center and the donor atoms;
for instance, the N20−Ni1−N22 angle is 78.8(2)°.
Magnetic Properties. The magnetic susceptibility of 2 and
4 was measured as a function of temperature between 2 and
295 K. Complex 2 was chosen as a representative for the series
of complexes with the tripodal ligands, and additionally the
magnetic properties of 4, which contains bidentate ligands,
were also investigated. The χMT versus T plots are shown in
Figure 4.
The room temperature χMT values for 2 and 4 are 1.18 and
1.20 cm3 K mol−1, respectively. The magnetic susceptibility for
both complexes remains constant down to ca. 8 K. Within this
range both complexes follow the Curie−Weiss behavior as can
be expected for an octahedral Ni(II) center isolated from its
immediate environment through bulky ligands. Below 8 K, the
χMT values for both the complexes show a sharp drop, which is
a consequence of zfs. Fitting the spin Hamiltonian parameters
that included terms for axial zero-field and isotropic Zeeman
Figure 4. Plots of χMT versus T for 2 (top) and 4 (bottom) at 0.5 T.
The solid lines represent the calculated curve fits (see text for
appropriate parameters).
splitting (see Experimental Section) delivered |D| values larger
for 2 (2.3 cm−1) compared to 4 (1.5 cm−1).
Additionally, the fits indicated important spin−orbit
contributions to the spin-only value of g = 2.00 (2.16 and
2.15 for 2 and 4, respectively). The data thus reveal the
significant magnetic anisotropy in complexes 2 and 4. However,
the magnitude and sign of D obtained from such powder
measurements are not reliable. This is not only a well-known
problem in magnetometry, as has been recently pointed out in
a work dealing with Ni(II) complexes,14m but also a possible
influence of other effects such as weak intermolecular
interactions or measurement artifacts at low temperatures due
to the orientation of the microcrystals in the magnetic field.
Attempts to include a rhombic zfs parameter E in magnetic data
analysis did not lead to any improvement of the fit quality. The
more reliable data extracted out of detailed HFEPR measure-
ments are discussed in the next section.
HFEPR Spectroscopy. Complexes 1−4 delivered HFEPR
spectra characteristic for an S = 1 spin state. The quality of
those spectra, however, varied from one complex to another. In
general the observed line widths were very large; single-crystal
line widths assumed in simulations were typically of the order
of 200 mT for the allowed (ΔMS = 1) transitions, and 50 −
70 mT for the “half-field” (ΔMS = 2) transition. Figure 5
shows the collection of spectra acquired at 5 K and at
approximately the same frequency (295 − 305 GHz). The
experimental spectra are accompanied by powder-pattern
simulations, using S = 1 spin Hamiltonian parameters obtained
from multifrequency experiments (see below). Only in the case
D
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. Experimental spectra (black traces) of complex 1 at 295.2 GHz and 2−4 at 304.8 GHz, both at 5 K. The red traces are simulations
assuming S = 1 spin Hamiltonian parameters as in Table 3 for all complexes except 3, for which they were as follows: D = +3.08 cm−1, E = +0.50
cm−1, g = [2.20, 2.15, 2.15]. Single-crystal line widths assumed in the simulations were [100, 100, 200 mT] for ΔMS = 1 transitions, 70 mT for ΔMS
= 2 transitions in complex 1; isotropic 200 mT for ΔMS = 1 transition, 50 mT for ΔMS = 2 transitions in complexes 2 and 3; isotropic 250 mT for
ΔMS = 1 transitions, 50 mT for ΔMS = 2 transition in 4. The particular turning points are labeled for complex 1. DQ stands for a double-quantum
transition that is not simulated.
of complex 3 we needed to “tweak” the parameters a bit to
achieve a better agreement between experiment and simulation
for that particular frequency. Single-frequency simulations
proved that in each case except complex 4, the zfs parameters
are positive (simulations for negative parameters are not
shown). In the case of 4, the rhombicity of its zfs tensor makes
the sign of little relevance; yet the simulations using negative
values were slightly better than the positive ones.
Because of well-known problems with torquing in high
magnetic fields and consequent distortions of the ideal powder
patterns, we chose to extract the spin Hamiltonian parameters
from a multifrequency 2-D map of resonances for each complex
rather than from single-frequency spectra, according to the
principle of tunable-frequency EPR.19 Figure 6 shows a
collection of such maps for each of the investigated complexes.
The spin Hamiltonian parameters fitted to the 2-D maps as
in Figure 6 are collected in Table 3. From those, one can see
that complexes 1 and 3 have basically identical parameters with
D ca. (positive) 3 cm−1, while complex 2 has zfs larger by about
50%, and complex 4 is smaller by a factor of 3. These findings
will be analyzed below. The g factors remain in the same range
of 2.12−2.18 for all the compounds, and their anisotropy is
rather small. Note that although the D values determined by
magnetometry for 2 and 4 differ from those determined by
HFEPR by 50%, the isotropic g value determined by
magnetometry for each complex agrees perfectly with the
average of the anisotropic g values determined by HFEPR.
Ligand Field Theory. Complexes 1, 2, and 3 have
sufficiently similar electronic absorption spectra such that
there is no point in attempting individual fits. Rather, we use
the following consensus values for the three spin-allowed
transitions for octahedral Ni(II) (see Figure 7 and Supporting
3
Information) in these complexes taken together: A2g(3F) →
3
3
3T2g at 11000 100 cm−1, A2g(3F) → T1g(3F) at 18250
3
3
100 cm−1, and, A2g(3F) → T1g(3P) at 28900
100 cm−1.
Fitting ligand-field parameters as defined by Ballhausen20 to
these transitions gives B = 875 cm−1, Dq = 1130 cm−1.
Complex 4 similarly yielded B = 850 cm−1, Dq = 1160 cm−1.
These values are in good agreement with those determined for
an extensive series of tetraazamacrocyclic complexes of Ni(II)
with a variety of axial ligands, as reported by Busch and co-
workers.21 For example, Ni([14]aneN4)(NCS)2, a complex
́
with a similar N4N2 donor set, gave B = 860 cm−1, Dq = 1418
cm−1.21b The electronic absorption data do not permit
extraction of parameters characterizing the axial distortion of
the ligand field (Ds, Dt), in contrast to the studies by Busch and
co-workers. For there to be zfs, such distortion must be present
and it is suggested by the different Ni−N bond lengths (2.04,
2.07, and 2.24 Å in 2). Inclusion of small tetragonal distortion
that is consistent with the electronic transitions with ζ = 540
E
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 6. Two-dimensional maps of turning points in the HFEPR spectra of the investigated complexes. Squares are experimental data at 5 K; lines
were drawn using the spin Hamiltonian parameters as in Table 3. Red lines represent B0//x turning points, blue lines represent B0//y, and black lines
represent B0//z resonances.
Table 3. Spin Hamiltonian Parameters of the Investigated Complexes As Determined by HFEPR
complex
D (cm−1
)
E (cm−1
)
E/D
gx
gy
gz
1
+2.92(2)
+4.35(2)
+2.91(3)
1.01(3)
+0.45(1)
+0.738(4)
+0.50(3)
0.23(1)
0.15
0.17
0.17
0.23
2.135(5)
2.143(7)
2.18(1)
2.108(6)
2.178(6)
2.120(15)
2.152(7)
2.138(6)
2.150(6)
2.130(6)
2.152(6)
a
2
3
b
4
2.140(2)
a
Variable temperature magnetic susceptibility for 2 gave |D| = 2.3 cm−1 and giso = 2.16, which agrees exactly with the average g value from HFEPR.
b
Variable temperature magnetic susceptibility for 4 gave |D| = 1.5 cm−1 and giso = 2.15, which agrees exactly with the average g value from HFEPR.
cm−1 (∼80% of the free-ion value)22 leads to D ≈ 0.8 cm−1,
much less than the experimental value for complexes 1−3, but
comparable to that seen for complex 4. This approach does not
include spin−spin coupling (SSC), which is expected to yield
only negligible contributions for S = 1 states. A more detailed
analysis of the electronic absorption spectra was performed by
quantum chemical calculations and their ligand field analysis as
discussed below.
Quantum Chemical Calculations. To gain a better insight
into the origin of zfs in the complexes under investigation and
for better understanding their absorption spectra, elaborate ab
initio calculations at the NEVPT2 level of theory17,15 were
performed.
As was recently shown, ab initio calculations correlate well
Figure 7. Absorption spectrum of 1 in CH3CN together with
calculated (NEVPT2) transitions (red sticks). The bands are identified
based on the complexes’ idealized Oh point group symmetry.
with the traditional LFT approach.15 For this reason, the
electronic transitions in the absorption spectra of 1, 2, and 4
were assigned on the basis of the NEVPT2 method. Correlated
calculations on complex 3 were prohibited by its large size. A
F
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 4. Absorption Spectra of 1 in CH3CN and Their Deconvolution Compared to Calculated Transitions with the NEVPT2
Method Based on the Optimized Geometry of 1
experiment
ε (M−1 cm−1
NEVPT2
)
a
a
transitions (in Oh)
maximum (cm−1
11 200
)
)
deconv (cm−1
)
fosc (× 10−6
)
energy (cm−1
fosc (× 10−9
)
3
3A2g(3F) → T2g
34.0
43.8
76.5
8200
11 000
11 700
17 700
18 700
20 700
28 800
225
229
253
426
72
11 936
12 628
12 856
19 357
20 922
21 731
31 604
33 148
33286
2251
251
8910
3403
56
3
3A2g(3F) → T1g(3F)
18 400
28 700
654
1572
939
3
3A2g(3F) → T1g(3P)
4693
1745
183
a
Listed numbers for the (dimensionless) oscillator strength fosc have to be multiplied by the factors given in parentheses.
Table 5. Contributions to the Zero-Field Splitting Parameters (in cm−1) from SA-CASSCF and NEVPT2 Calculations Using 14
a
Excited States (9 Triplets and 5 Singlets)
SA-CASSCF
NEVPT2
compd
method
2nd PT
state
energy
D
E
energy
D
E
1
3A
3A
3A
1A
1A
1A
8350
8879
25.2
11.7
15.7
−12.8
−1.1
−6.8
6.4
11 559
12 255
12 612
27 104
27 730
28 523
18.2
5.2
1.9
−16.7
−6.8
−5.2
11.8
−4.3
0.2
9467
−31.2
−7.2
−6.6
13.6
26 060
26 480
27 247
−5.2
4.8
0.0
0.1
sum of all
+5.80
+4.80
+1.26
+1.06
+3.36
+2.94
+2.92(2)
17.0
+0.74
+0.66
+0.45(1)
17.8
QDPT
expt
2
2nd PT
3A
3A
3A
1A
1A
1A
8085
8515
24.5
22.3
25.3
−23.7
0.1
11 109
11 672
13 091
26 724
27 198
28 897
14.9
−16.7
0.2
9848
−38.9
−7.1
−6.8
13.4
−27.2
−6.7
−6.3
25 849
26 161
27 521
−7.2
6.9
−6.9
6.6
−0.0
+1.25
+1.20
12.4
−0.0
+0.75
+0.66
+0.738(4)
sum of all
+7.74
+5.95
+4.48
+3.79
+4.35(2)
QDPT
expt
a
Optimized geometry of 1 and X-ray geometry of 2 were used.
full list of the energies of the spin-allowed d−d transitions of
complexes 2 and 4 based on their X-ray and DFT geometry
optimized structures for 1, 2, and 4 are listed in Table S3
(Supporting Information). The energies obtained from
calculations based on optimized structures compare well with
the ones based on the X-ray structure.
distributions in excellent agreement with the experimental
spectrum (SI, Figures S1−S3).
The absorption spectra of 1, 2, and 4 in acetonitrile were
calculated on the SA-CASSCF/NEVPT2 level of theory
including all of the possible 24 excited states (9 triplets and
15 singlets, which with the triplet ground state result in a total
of 45 microstates of d8). In Table 4, the energies dominated by
the triplet states for 1 are shown; energies for 2 and 4 can be
found in the Supporting Information (Table S1). All triplet
states contribute to the absorption spectrum due to spin-
allowed transitions. Spin-forbidden transitions have not been
observed.
3
The reduction of symmetry leads to the splitting of the T2g
and 3T1g states into three nondegenerate sublevels. In contrast,
the experimental absorption spectra show only three main
bands, which are due to overlapping absorption bands which
can be resolved starting with the NEVPT2 excitation energies
and refining their values by a fit to the experimental d−d
absorption band profile. For further analysis the experimental
absorption spectra were subjected to spectral deconvolution,
allowing for a better comparison with the theoretical results.
Figure 7 shows the experimental absorption spectrum of 1 in
CH3CN (black trace) together with calculated transitions on
the NEVPT2 level (red bars). Each of the three main bands
could be fitted with three Gaussian distributions representing
the three times three possible transitions with the sum of these
The theoretical investigation helps to clarify the experimental
results: Even though the reduction of symmetry leads to
splitting of a given state, there is always one out of three
transitions that has much stronger oscillator strength (fosc) than
the others. For compound 1, the three calculated transitions
with the major oscillator strength show up at 12856, 19357, and
31604 cm−1, which compares reasonably well with the
experimental maxima at 11200, 18400, and 28700 cm−1
respectively. Therefore, in the experimental spectrum only
G
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
three main bands are observed, despite considerable axial
elongation. However, these experimental bands are relatively
broad and, in the case of the lowest energy band, somewhat
distorted in line shape. Thus, spectral deconvolution helps to
uncover the underlying weaker transitions, which compare well
with the calculated ones (Figures S1−S3 in the Supporting
Information).
V =
⎡
⎢
⎢
⎢
⎢
⎢
⎢
⎢
⎤
⎥
⎥
⎥
⎥
⎥
⎥
⎥
Vxy,xy
Vyz,xy
Vz2,xy
Vxz,xy
Vyz,xy
Vyz,yz
Vz2,yz
Vxz,yz
Vz2,xy
Vz2,yz
Vz2,z2
Vxz,z2
Vxz,xy
Vxz,yz
Vxz,z2
Vxz,xz
Vx2−y2,xy
Vx2−y2,yz
Vx2−y2,z2
Vx2−y2,xz
After thorough evaluation of the absorption spectra, ab initio
methods were used for examination of the zfs parameters. In a
first attempt, all excited states (9 triplets and 15 singlets) were
included in the calculation (see Supporting Information, Table
S2). In a recent series of papers,23 a computational strategy was
presented wherein not all excited states were included in the
state-averaged CASSCF wave function. This has significantly
improved the description of the orbitals of the zero order wave
function since not all roots contribute to the zfs phenomenon.
Even though the other states do not contribute to the zfs, they
were equally weighted with roots that contribute. Those were
V
Vx2−y2,yz Vx2−y2,z2 Vx2−y2,xz Vx2−y2,x2−y2
⎢
⎣
⎥
⎦
x2−y2,xy
(1)
Equation 1 represents the ligand field matrix in the most
general form independent of any additive ligand field
approximations (see Supporting Information eqs S8 and
S10−S14 for numerical values of the matrix V for complexes
1, 2, and 4). Diagonalization of the matrix in eq 1 yields ligand
field orbital energies for complexes 1, 2, and 4 as depicted in
Figure 8.
3
mainly the first set of excited triplet states, A, and a set of
excited singlet states, 1A.14m This strategy leads to a less biased
description, hence the number of excited states was reduced
from 15 to 5, including only the states important for the zfs.
In Table 5, only the states with significant contributions to
the zfs are shown. Compound 4 is not discussed because of the
rhombicity of the system. The major contribution to the zfs
3
comes from the excited triplet states stemming from the T2g
octahedral term, whereas the excited singlet states contribute
only in a minor way, as revealed by second order perturbation
theory (PT2). Comparing the SA-CASSCF and the NEVPT2
approaches, it can be seen in particular that, due to the dynamic
electron correlation added by the NEVPT2 treatment, the
excited triplet states shift to higher energies. As a result, the zfs
parameters D and E are reduced almost by 50%, as zfs
qualitatively is inversely proportional to excited state energies,
which is an effect taking into account the covalency of the
system. The final D values calculated by quasi degenerate
perturbation theory (QDPT) are +2.94 and +3.79 cm−1 for 1
and 2, respectively, which is in excellent agreement with the
values determined by HFEPR (+2.92(2) and +4.35(2) cm−1,
respectively). Because QDPT considers all contributions to the
zfs, it yields very accurate values, but in comparison to the PT2
approach, it does not allow to discuss the single contribution of
the excited states to the total zfs.
These results clearly point to the necessity of NEVPT2
calculations on top of the CASSCF treatment to add dynamic
correlation and thereby covalency. This multireference
approach is applicable even for large molecules, like the
complexes under investigation, thus showing the broad
applicability of this methodology.
Ni(II)−Ligand Bonding as Revealed by a Ligand Field
Analysis of the Ab Initio Results. A considerable insight
into the nature of the metal ligand bond can be gained by
deriving ligand field parameters from the rich database offered
by CASSCF/NEVPT2 results (ab initio ligand field theory).15a,c
Numerical values of the parameters of the ligand field given by
the matrix elements of the symmetric 5 × 5 ligand field matrix
of eq 1 and in addition, the Racah parameter B of d−d
interelectronic repulsion have been computed from a best fit to
the 10 × 10 matrix of the total of the 10 triplet states of Ni(II)
in the considered complexes (for a detailed outline of the
procedure for complex 4, taken as an example, see Supporting
Information.)
Figure 8. Metal based d-MO energies from ab initio ligand field
analysis (AILFT) of NEVPT2 multireference electronic structure
calculations for 1 (from the DFT optimized complex geometry) and
for 2 and 4 (with geometries from X-ray data). Labeling of orbitals
pertain to a coordinate system choice with the z, y, and x axes
respectively oriented parallel to the trans Nam−Nam, Nta−Nta, and Nta−
Nta bond directions in 1 and 2, and the Nta−Nta, Npy−Npy, and Nta−
Npy bond directions in 4 (see also Table 2).
2
2
2
An energy order of d-MOs dyz ≤ dxz < dxy < dz < dx −y (for 1
2
2
2
and 2) and dxz ≤ dyz < dxy < dx −y < dz (for 4) results from ab
initio LFT analysis. On the basis of the very small splitting of
the dxz, dyz, and dxy type orbitals we infer from this analysis that
π Ni−N bonding is weak in all reported complexes. Correlated
calculations on complex 3 were prohibited by its large size.
However, on the basis of the very similar DFT optimized
coordination geometries for 1, 2, and 3 (see Table 2) we can
safely assume that bonding in 3 is similar to that in 1 and 2, i.e.,
it is not significantly altered by more distant substituents. The
very weak metal ligand π interactions reflected by the parameter
eπ is consistent with the dxz, dyz, and dxy orbitals being doubly
occupied in the ground state configuration t2g6eg2 of Ni(II). This
compensates the energy stabilization due to Ni−N donor
interactions and agrees with the weak Ni−N π-bonding inferred
from analysis of d−d absorption and EPR spectra of Ni(II)
2
complexes with similar N-donor ligands.24 The large dz −dx −y
2
2
MO splitting features the distinctly weaker σ-antibonding
character of the Ni(II)-N amine compared with the Ni−N
triazole ligand. The resulting tetragonally elongated octahedral
geometry is also reflected by the large and positive D values
deduced from the HFEPR results (see Table 3). In contrast to
H
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 6. Ni−N AOM Parameters, the Racah Parameter B, and Zero-Field Splitting Parameters (in cm−1) from a Best Fit of the
AOM to Ab Initio NEVPT2 Results
complex
1
2
4
b
DFT opt
X-ray
DFT-opt
X-ray
DFT-opt
epσy
3706
−131
3859
−276
3824
−200
3827
−113
py
e
πs
etσa
4083
4131
167
4034
ta
e
eaσm
217
199
πs
3256
2802
1200
421
3033
B
1202
1199
1212
282
1204
a
σ(AOM)
393
454
269
ta ta
ta
D(Dexp
E(Eexp
)
+2.30 (+2.92)
+0.87 (+0.45)
+3.23
+0.76
−4.14 (+4.35)
−0.54 (+0.74)
e: DN
m: DN
= −0.98 (12°)
= +0.02 (18°)
= +0.96 (12°)
DN ta = −0.91 (18°); Dexp
=
=
1.00
0.23
N
N
py py
N
ta py
N
ta py
)
DN
= +0.28 (18°); Eexp
= +0.64 (13°)
N
py py
h: DN
DN
N
a
Standard deviation between ligand field matrix elements from NEVPT2 calculations using DFT optimized (1, 2, 3) and X-ray (2, 4) geometries and
b
their respective AOM values calculated using the listed best fit AOM parameters. The letters “e”, “m”, and “h” in the last row imply the directions of
the easy, medium, and hard magnetic axis as identified by the computed signs of the main values of zfs tensor; the entries listed in parenthesis show
the deviations (in deg) between the calculated “e”, “m”, and “h” directions and the corresponding NtaNta, NpyNpy, and NtaNpy bond axes in 4.
the coordination geometries of 1, 2, and 3, the orbital energies
of 4 reflect an almost homogeneous coordination environment.
bonding type. The very nearly octahedral ligand field of
complex 4 is in agreement with its electronic spectrum which
shows no resolved splitting of the transitions of the octahedral
parent geometry (Figure S3). In agreement with this
conclusion, D and E parameters resulting from the analysis of
the HFEPR spectrum of 4 are considerably smaller than those
for 1−3 (Tables 3 and 6).
2
2
2
A much smaller dx −y −dz energy difference reflects the very
close σ-antibonding properties of the two triazole and pyridine
ligands with the triazole ligand acting as a slightly stronger σ-
donor toward Ni(II). Based on the slightly compressed
2
2
2
geometry (dz > dx −y ) a negative sign of D would be expected.
However, the additional angular distortions and the resulting
very low (C1) symmetry renders the computed (and also
observed by HFEPR) zfs tensors almost entirely rhombic (see
Tables 3 and 6). A quantification of the qualitative conclusions
based on the ligand field MO diagram (Figure 8) is possible by
an angular overlap model (AOM)25,26 analysis of the ab initio
LF matrices V (eq 1 and discussion in Supporting Information).
This model is less general, but more appealing for chemists,
although the AOM parametrization is not always unique. In
standard applications of the AOM, up to four independent
AOM parameters (deriving from energy differences due to
complete C1 low-symmetry splitting of the five d-orbitals) can
be computed. However, focusing on complex 4, six such
parameters eσ, eπs and eπc, respectively denoting the σ-bonding,
out-of-plane π-bonding, and in-plane π-bonding parameters for
each of the two ligand types, the N-pyridine (py) and N-
triazole (ta), result. Here we adopt a parametrization in which
the in-plane π parameters, eπc, for both ligand types have been
set to zero, as there are no energetically favorable interactions
of the N ligand orbitals in the plane, as these are involved in
stronger bonds with the ligand framework atoms attached to
them. To discern the effect of the out-of-plane parameter, eπs,
on the AOM matrix elements, we used a dummy atom attached
to each N-ligand defining the (py) and (ta) ligand planes.
Utilizing the exact geometry of complex 4 from the X-ray
structure and from DFT geometry optimizations, AOM
parameters from a least-squares fit to the V matrices were
derived (Table 6, see a detailed outline of the procedure
applied to this complex in Supporting Information). As seen by
the rather small standard deviation (282 cm−1, NEVPT2
results), the consistency of the fit is remarkable. From these
results we can deduce that Ni−N bonding is dominated by Ni−
Nta and Ni−Npy σ-antibonding interactions with comparable
values for the two ligands. In contrast, out-of-plane π-bonding
AOM parameters deduced from NEVPT2 results for
complex 4 have been compared with those resulting from
CASSCF (accounting for static electron correlation, Table 7)
Table 7. AOM Parameters for 4 from LF Analysis of
CASSCF and NEVPT2 Results
parameter
CASSCF (cm−1
)
NEVPT2 (cm−1
)
epσy
3210
−121
3379
−246
3706
−131
3859
−276
py
e
πs
etσa
ta
e
πs
where the effect of the excited charge transfer states on the 3d8
multiplets of Ni(II) have been neglected. From this comparison
we conclude that Ni−N covalence (charge transfer, dynamical
correlation) yields important contributions to AOM parame-
ters.
Complexes 1 and 2 present coordination spheres with two
axial Ni−Nam in trans positions to each other (forming longer
Ni−Nam bonds, 2.242 (in 2) and 2.206 0.003 Å (in 1, average
value) and four shorter Ni−Nta bonds in equatorial positions
(2.056 0.015 Å (2) and 2.058 0.033 Å (1)), respectively.
The rather similar coordination environments of the two
complexes give rise to similar multiplet energies. In Supporting
Information (Table S2) we list their values from CASSCF and
NEVPT2 calculations where electronic energy levels have been
assigned using the Oh parent symmetry. However, the true, low
symmetry is important as it leads to a splitting of the triplet
cubic terms.
The NEVPT2 matrices for the DFT optimized structures of
1 and 2 and for the X-ray structure of 2 were further analyzed
with the AOM adopting the following parametrization: the in-
plane π-bonding parameters, eπc, for the Nta ligand was set to
zero as in complex 4 and only σ-antibonding were taken into
account in the case of Nam. For understanding the out-of-plane
py
ta
reflected by the parameters e = −131 and e = −276 cm−1 is
πs
πs
of very small magnitude and negative, i.e., of the π-back
I
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
parameter eπs, we used as for 4 a dummy atom attached to each
of the Nta ligand planes. Utilizing the exact geometry of 1 from
experimental values. In addition, theory uncovered the spin−
orbit coupling to triplet excited states as the major contribution
to the zfs. Ligand field analysis of the ab initio results gave
insight into chemical bonding. The Nta atoms proved to be
better σ-donors than the Nam atoms, which compares well with
the bond length analysis of the X-ray structures.
the DFT geometry optimization and for 2 from the X-ray data
ta
we have adjusted the three AOM parameters eaσm, eσta, e to the
πs
(over)complete set of elements of the matrices VNEVPT2 (Table
6). This leads us to the following values of these parameters:
eaσm = 3256, eσta = 4083, e = 217 cm−1 (for 1) and eaσm = 2802,
am
πs
etσa = 4131, e = 167 cm−1 (for 2). Reasonably small standard
ta
πs
CONCLUSIONS
■
deviations between calculated AOM matrix elements, using
We have reported here on the synthesis and characterization of
Ni(II) complexes with “click” derived triazole ligands, one of
which is new: the tripodal tdta. For complexes 1−3 an axially
distorted octahedral coordination environment is observed
around the Ni(II) center that reduces their local symmetry
from an ideal Oh to D2h and further to C1 and Ci. In the case of
4, which contains bidentate ligands, the triazole N atoms are
seen to bind more strongly to the Ni(II) center compared to
the pyridine N donors. A comprehensive HFEPR study
together with simulations delivered precise zfs parameters
from which it can be seen that the magnitude of D for the
complexes 1−3 is larger than that for 4, a fact that correlates
with the axial structural distortion of 1−3. LFT and ab initio
calculations at the SA-CASSCF and NEVPT2 levels of theory
allowed us to interpret the absorption spectra and extract ligand
field parameters. Quantum chemical calculations are in line
with the experimental zfs parameters and provided insight into
their origin. Furthermore, ligand field analysis of the ab initio
results allowed distinguishing between the different σ-donating
properties of the different nitrogen donor atoms.
these parameters and their original ab initio NEVPT2 values
(393 cm−1 for 1 and 421 cm−1 for 2, see Table 6) demonstrate
the consistency of the fit and the adequacy of the para-
metrization scheme. From these results we can deduce, as for 4,
that Ni−N bonding in 1 and 2 is dominated by Ni−Nta and
Ni−Nam σ-antibonding interactions. However, σ-donation for
Nta is distinctly more pronounced as compared to Nam. This is
also reflected in the shorter Ni−N bond distances for the latter
ligand compared to the former.
Comments on Combined Experimental and Theoreti-
cal Results. The complexes 1−4 provide a unique opportunity
of comparing six-coordinated Ni(II) complexes with two
different types of substituted 1,2,3-triazole ligands; whereas
the tripodal varieties in 1−3 contain a N4N′2 coordination
around the Ni(II) center, 4 contains a N3N′3 donor set with the
triazole N and pyridyl N taking up a meridional coordination.
Structurally, 2 shows a pronounced axial elongation because of
the weak amine donors of the tripodal ligands. This is also seen
for 1′ (and hence for 1) and is expected for 3, for which we
unfortunately could not generate appropriate single crystals.
Complex 2 has unique, intramolecular weak interactions in its
secondary coordination sphere because of the benzyl
substituents of tbta. These interactions most likely contribute
to the total ligand field stabilization energy, as has been shown
by us recently for the related Co(II) complex.10a In the case of
4, even though each axis of the octahedron has a different
donor set, the metal ligand bond lengths are rather similar, and
hence there is no pronounced axial distortion in this case as
opposed to 1−3 even though the symmetry is much lower than
Oh.
To the best of our knowledge, this is the first time that such
comprehensive structural, magnetic, HFEPR, LFT, and
quantum calculations have been carried out on six-coordinate
Ni(II) complexes with N4N′2 or N3N′3 donor sets. The results
show that structural and hence electronic differences can arise
despite the donating atoms being similar. Our studies also
highlight the continuing versatility of “click” derived ligands in
coordination chemistry and establish the combination of
HFEPR and quantum chemical calculations as an extremely
powerful tool for elucidating the electronic structure of metal
complexes bearing more than one unpaired electron.
The room temperature magnetic moment for both 2 and 4
matches well with values expected for isolated octahedral Ni(II)
centers. HFEPR data together with simulations provided
accurate magnitudes and signs for the D parameter, along
with the rhombic parameter, E. It was seen that the D values are
positive for complexes 1−3 which is expected from the
structural data that clearly show an axial elongation for these
complexes with the tripodal ligands. Within the complexes 1−3,
2 has a higher D value. It is tempting to assign this difference to
the unique C−H···π interactions observed within the secondary
coordination sphere of 2, but this is purely speculation.
An LFT parameter fit was seen to reproduce the
experimental absorption spectra with reasonable accuracy.
However, quantum chemical calculations show that there are
more underlying transitions that are not considered in a simple
LFT fit. The reduction of symmetry from Oh to C1 (in 1 and 4)
and Ci (in 2) leads to the appearance of multiple transitions,
albeit only with small oscillator strengths. A Gaussian
deconvolution, together with quantum chemical calculations
made the assignment of all the nine spin-allowed transitions
possible. High-level ab initio calculations at the SA-CASSCF/
NEVPT2 level of theory gave more insights into the zfs
parameters. In particular, the absolute values of the zfs
calculated by NEVPT2 for 1 and 2 compare well with
EXPERIMENTAL SECTION
■
Materials and General Methods. All chemicals were used as
received unless otherwise mentioned. The ligands tbta and tpta were
synthesized according to reported procedures.18 Elemental analyses
1
were performed with a Perkin-Elmer Analyzer 240. H NMR spectra
were recorded at 250.13 MHz on a Bruker AC250 instrument. Mass
spectrometry experiments were carried out on a Bruker Daltronics
Mictrotof-Q mass spectrometer.
Magnetic Susceptibility Measurements. Temperature-depend-
ent magnetic susceptibility measurements were carried out with a
Quantum-Design MPMS-XL-5 SQUID magnetometer equipped with
a 5 T magnet in the range from 295 to 2.0 K at a magnetic field of 0.5
T. The powdered sample was contained in a gel bucket and fixed in a
nonmagnetic sample holder. Each raw data file for the measured
magnetic moment was corrected for the diamagnetic contribution of
the sample holder and the gel bucket. The molar susceptibility data
were corrected for the diamagnetic contribution. Magnetic properties
were simulated using the julX program (E. Bill, Max Planck Institute
for Chemical Energy Conversion, Mulheim/Ruhr, Germany) using the
̈
following spin Hamiltonian:
⎛
⎞
⎟
2
1
3
̂
̂
⃗
⃗
⎜
H = D Sz − S(S + 1) + gμBBS
⎝
⎠
(2)
J
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Temperature-independent paramagnetism (TIP) was included accord-
ing to χcalc = χ + TIP (50 × 10−6 cm3 mol−1 and 150 × 10−6 cm3 mol−1
for 2 and 4, respectively).
H). 13C NMR (62.5 MHz, CDCl3): δ 24.0; 28.4; 47.1; 123.7; 126.9;
130.7; 133.3; 143.4; 145.9.
pyta. 2-Pyridylacetylene (206 mg, 2.0 mmol), 2,6-di-isopropylphe-
nylazide (406 mg, 2.0 mmol), CuSO4·5H2O (25 mg, 0.1 mmol),
sodium ascorbate (79 mg, 0.4 mmol), and TBTA (11 mg, 0.02 mmol)
were dissolved in CH2Cl2/H2O/tert-BuOH (2.5 mL/2.5 mL/5 mL)
and stirred for 3 d at 50 °C. Then water (50 mL) was added, and the
reaction mixture was extracted with CH2Cl2 (3 × 20 mL). The organic
phase was separated and washed with an EDTA/ammonia solution (1
M) (3 × 10 mL). Finally the solution was dried over Na2SO4, and the
solvent was evaporated. After flash chromatography over silica
(CH2Cl2:CH3OH, 99:1) the product was isolated as a white solid
(460 mg) in 75% yield. Anal. Calcd for C19H22N4: C, 74.48; H, 7.24;
N, 18.29. Found: C, 74.06; H, 7.37; 18.10. HRMS (ESI) m/z: calcd for
HFEPR Spectroscopy. HFEPR spectra were recorded using the
Electron Magnetic Resonance (EMR) Facility at the National High
Magnetic Field Laboratory (NHMFL, Tallahassee, FL) using a single-
pass transmission homodyne spectrometer. The spectrometer employs
a Virginia Diodes (Charlottesville, VA) source operating at a base
frequency of 12−14 GHz and multiplied by a cascade of multipliers in
conjunction with a 15/17 T superconducting magnet. Detection was
provided with an InSb hot-electron bolometer (QMC Ltd., Cardiff,
U.K.). The magnetic field was modulated at 50 kHz. A Stanford
Research Systems SR830 lock-in amplifier converted the modulated
signal to dc voltage. Low temperature was provided by an Oxford
Instruments (Oxford, U.K.) continuous flow cryostat with temperature
controller (see ref 27 for more details). The typical sample amount
was 40−100 mg. All the investigated complexes yielded spectra
recognizable as S = 1 powder patterns. These spectra were analyzed by
simultaneously fitting the parameters of the standard spin Hamil-
tonian:
1
C19H23N4 ([M + H]+), 307.1917; found, 307.1915. H NMR (250
MHz, CDCl3): δ 1.15 (d, 3JH−H = 6.8 Hz, 6H, CH3); 1.17 (d, 3JH−H
=
3
6.8 Hz, 6H, CH3); 2.35 (septet, JH−H = 6.9 Hz, 2H, CH); 7.28 (t,
3
3JH−H = 7.5 Hz, 1H, pyridyl); 7.32 (d, JH−H = 7.9 Hz, 2H, phenyl);
3
3
7.51 (t, JH−H = 7.8 Hz, 1H, phenyl); 7.84 (t, JH−H = 7.7 Hz, 1H,
3
pyridyl); 8.23 (s, 1H, 5-triazole-H); 8.32 (d, JH−H = 7.9 Hz, 1H,
pyridyl); 8.62 (multiplet, 1H, pyridyl). 13C NMR (62.5 MHz, CDCl3):
δ 24.2; 24.4; 28.6; 120.7; 123.2; 124.0; 125.0; 131.0; 133.3; 137.1;
146.3; 148.3; 149.7; 150.5.
2
2
2
̂
̂
̂
̂
/ = βB·g·S + D[Sz − S(S + 1)/3] + E[Sx − Sy ]
(3)
e
[Ni(TPTA)2](BF4)2, 1. Ni(BF4)2·6H2O (70 mg, 0.20 mmol) and
TPTA (200 mg, 0.41 mmol) were dissolved in MeOH (5 mL). The
solution was refluxed for 1 h. After the solution cooled, the light pink
product precipitated. It was filtered, washed with ether, and isolated in
70% yield (170 mg). Anal. Calcd for C54H48B2F8N20Ni: C, 53.63; H,
4.00; N, 23.16. Found: C, 52.94; H, 4.31; N, 23.01. HRMS (ESI) m/z:
calcd for C54H48N20Ni ([M]2+), 517.1857; found, 517.1747. IR cm−1:
613(s), 654(s), 687(vs), 756(vs), 824(s), 865(s), 955(s), 1056(br),
1194(s), 1250(s), 1359(s), 1439(s), 1467(s), 1502(s), 1597(s),
3160(br).
[Ni(TBTA)2](BF4)2, 2. Ni(BF4)2·6H2O (70 mg, 0.20 mmol) and
TBTA (218 mg, 0.41 mmol) were dissolved in acetonitrile (5 mL).
The solution was refluxed for 1 h. After the solution cooled, ether (5
mL) was added and the light pink product was allowed to crystallize.
Then it was filtered, washed with ether, and isolated in 87% yield (225
mg). Single crystals suitable for X-ray diffraction were grown by slow
diffusion of ether into an acetonitrile solution. Anal. Calcd for
C60H60B2F8N20Ni: C, 55.71; H, 4.68; N, 21.66. Found: C, 55.67; H,
4.72; N, 21.78. HRMS (ESI) m/z: calcd for C60H60N20Ni ([M]2+),
559.2326; found, 559.2309. IR cm−1: 579(s), 697(s), 718(vs), 756(s),
823(br), 957(s), 1040(br), 1247(s), 1330(s), 1455(s), 1497(s),
3145(br).
[Ni(TDTA)2](BF4)2, 3. Ni(BF4)2·6H2O (70 mg, 0.20 mmol) and
TDTA 303 mg, 0.41 mmol) were dissolved in EtOH (5 mL). The
solution was refluxed for 1 h. After the solution cooled, the light pink
product precipitated. It was filtered, washed with ether, and isolated in
85% yield (287 mg). Anal. Calcd for C90H120B2F8N20Ni: C, 63.05; H,
7.06; N, 16.34. Found: C, 62.51; H, 7.15; N, 16.24. HRMS (ESI) m/z:
calcd for C90H120N20Ni ([M]2+), 769.4674; found, 769.4697. IR cm−1:
624(br), 676(s), 708(s), 759(vs), 803(vs), 862(s), 935(s), 993(s),
1054(br), 1179(s), 1254(br), 1364(br), 1386(s), 1473(br), 2964(br).
[Ni(pyta)3](BF4)2, 4. Ni(BF4)2·6H2O (74 mg, 0.22 mmol) and pyta
(200 mg, 0.65 mmol) were dissolved in EtOH (7.5 mL). The solution
was refluxed for 1 h. The hot solution was filtered and the filtrate
collected. The filtrate evaporated slowly, and the light pink product
precipitated. It was filtered, washed with ether, and isolated in 60%
yield (151 mg). Single crystals suitable for X-ray diffraction were
grown by slow diffusion of ether into an ethanol solution. Anal. Calcd.
for C57H66B2F8N12Ni: C, 59.45; H, 5.78; N, 14.60. Found: C, 59.20;
H, 5.80; N, 14.58. HRMS (ESI) m/z: calcd for C57H66N12Ni ([M]2+)
,488.2438; found, 488.2338. IR cm−1: 569(s), 643(s), 715(s), 761(s),
785(vs), 806(s), 935(s), 982(s), 1057(br), 1277(s), 1329(s), 1366(s),
1388(s), 1454(vs), 1575(s), 1614(s), 2965(br), 3112(br).
to the complete two-dimensional (field vs frequency) map of turning
points, following the principles of tunable-frequency EPR.19 The sign
of zfs was obtained by simulating single-frequency spectra using
software (program SPIN) available from A. Ozarowski.
Quantum Chemical Calculations. The program package ORCA
2.9.128 was used for all calculations. The coordinates from X-ray
structures of 2 and 4 were used without modification. To generate a
structural model for 1, the X-ray structure of 1′ was subjected to a
geometry optimization with the BP86 functional.29 In all calculations,
triple-ζ valence quality basis sets (def2-TZVP) were used for all
atoms30 along with empirical van der Waals corrections for the
nonbonding interactions.31 Convergence criteria for the geometry
optimization were set to default values and “tight” convergence criteria
were used for the SCF procedure. For single point calculations (DFT
and unrestricted Hartree−Fock), a combination of the resolution of
the identity and the “chain of spheres exchange” algorithms
(RIJCOSX) was employed32,33 with matching auxiliary basis
sets.29,34 The environmental effect were included using the
conductor-like screening model (COSMO).35 The active space of
the CASSCF calculations consisted of the 5 metal-based d-orbitals
leading to an active space with eight electrons in five orbitals
[CAS(8,5)]. The state averaged approach was used where all excited
triplet states (10 roots) and all excited singlet states (15 roots) were
equally weighted. In addition, for the sake of analysis a reduced
number of singlet states (five roots) was used. To recover dynamic
correlation, multireference N-electron valence perturbation theory of
second order (NEVPT2)17 was employed on top of the state-averaged
CASSCF reference wave functions. MOs were visualized via the
program Molekel.36 Spectral deconvolution was achieved with the
ORCA subprogram orca_asa.28
Synthesis. tdta. 2,6-Diisopropylphenylazide (3.5 g, 17.2 mmol),
tripropargylamine (500 mg, 3.8 mmol), CuSO4·5H2O (143 mg, 0.6
mmol), and TBTA (30 mg, 0.06 mmol) were dissolved in tert-
butanol/H2O/DCM (50/25/25 mL). Then sodium ascorbate (454
mg, 2.3 mmmol) was added and the mixture stirred for 3 d at 70 °C.
Then water (200 mL) was added, and the reaction mixture was
extracted with CH2Cl2 (3 × 50 mL). The organic phase was separated
and washed with an EDTA/ammonia solution (1 M) (3 × 20 mL).
Finally, the solution was dried over Na2SO4, and the solvent was
evaporated. After flash chromatography over silica (CH2Cl2:CH3OH,
98:2) the product was isolated as a white solid (2.11 g) in 75% yield.
Anal. Calcd for C45H60N10·CH3OH: C, 71.47; H, 8.34; N, 18.12.
Found: C, 71.20; H, 8.32; 18.41. HRMS (ESI) m/z: calcd for
C45H60N10Na ([M + Na]+), 763.4895; found, 763.4876. 1H NMR
X-ray Crystallography. Suitable single crystals of 2 and 4 were
selected and mounted onto a thin glass fiber. X-ray intensity data were
measured at 150 K on an Oxford Gemini S Ultra diffractometer with
the Enhance X-ray Source of Cu Kα radiation (λ = 1.54178 Å) using
3
(250 MHz, CDCl3): δ 1.13 − 1.16 (m, 36H, CH3); 2.29 (qt, JH−H
=
3
6.8 Hz, 6H, CH); 4.00 (s, 6H, CH2); 7.28 (d, JH−H = 7.8 Hz, 6H,
phenyl); 7.48 (t, 3JH−H = 7.9 Hz, 3H, phenyl); 7.98 (s, 3H, 5-triazole-
K
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
the ω−φ scan technique37 or with a four circle diffractometer P4
(Siemens, Madison WI) at 173 K. Empirical absorption correction was
applied using spherical harmonics implemented in SCALE3
ABSPACK scaling algorithm.38 The structure was solved by direct
methods and refined by full-matrix least-squares against F2 of all data
using the SHELXTL program package.39 Anisotropic thermal factors
were assigned to the non-hydrogen atoms, while the positions of the
hydrogen atoms were generated geometrically, assigned isotropic
thermal parameters, and allowed to ride on their respective parent
atoms before the final cycle of least-squares refinement. CCDC-
910983 and 910984 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from the
Cambridge Crystallographic Data Centre via www.ccdc.cam.uk/data_
request.cif.
Winter, A.; Hager, M. D.; Hoogenboom, R.; Schubert, U. S. Chem.
Asian J. 2009, 4, 154.
(4) (a) Swanick, K. N.; Ladouceur, S.; Zysman-Colman, E.; Ding, Z.
Chem. Commun. 2012, 48, 3179. (b) Brown, D. G.; Sanguantrakun, N.;
Schulze, B.; Schubert, U. S.; Berlinguette, C. P. J. Am. Chem. Soc. 2012,
134, 12354. (c) Manbeck, G. F.; Brennessel, W. W.; Eisenberg, R.
Inorg. Chem. 2011, 50, 3431. (d) Mydlak, M.; Bizzarri, C.; Hartmann,
D.; Sarfert, W.; Schmid, G.; De Cola, L. Adv. Funct. Matter 2010, 20,
1812.
(5) (a) Ostermeier, M.; Berlin, M.-A.; Meudtner, R. M.; Demeshko,
S.; Meyer, F.; Limberg, C.; Hecht, S. Chem.Eur. J. 2010, 16, 10202.
(b) Guha, P. M.; Phan, H.; Kinyon, J. S.; Brotherton, W. S.; Sreenath,
K.; Simmons, J. T.; Wang, Z.; Clark, R. J.; Dalal, N. S.; Shatruk, M.;
Zhu, L. Inorg. Chem. 2012, 51, 3465.
́
(6) (a) Maisonial, A.; Serafin, P.; Traikia, M.; Debiton, E.; Thery, V.;
ASSOCIATED CONTENT
* Supporting Information
Experimental and deconvoluted spectra for 1, 2, and 4. Tables
of calculated triplet energies for 2 and 4; calculated D and E
values for 1, 2, and 4; ligand field matrices; and AOM
parameters extracted from ligand field analysis of the ab initio
results. This material is available free of charge via the Internet
at http://pubs.acs.org.
Aitken, D. J.; Lemoine, P.; Viossat, B.; Gautier, A. Eur. J. Inorg. Chem.
2008, 298. (b) Chevry, A.; Teyssot, M. L.; Maisonial, A.; Lemoine, P.;
Viossat, B.; Traikia, M.; Aitken, D. J.; Alves, G.; Morel, L.; Nauton, L.;
Gautier, A. Eur. J. Inorg. Chem. 2010, 3513.
■
S
(7) (a) Crowley, J. D.; Bandeen, P. H. Dalton Trans. 2010, 39, 612.
(b) Bai, S.-Q.; Leelasubcharoen, S.; Chen, X.; Koh, L. L.; Juo, J.-L.;
Hor, T. S. A. Cryst. Growth Des. 2010, 10, 1715. (c) Steinmetz, V.;
Couty, F.; David, O. R. P. Chem. Commun. 2009, 343. (d) Collason,
B.; Paul, N. L.; Mest, Y. L.; Reinaud, O. J. Am. Chem. Soc. 2010, 132,
4393. (e) Aucagne, V.; Berna, J.; Crowley, J. D.; Goldup, S. M.; Hanni,
K. D.; Leigh, D. A.; Lusby, P. J.; Ronaldson, V. E.; Slawin, A. M. Z.;
Viterisi, A.; Walker, D. B. J. Am. Chem. Soc. 2007, 129, 11950.
(f) Clough, M. C.; Zeits, P. D.; Bhuvanesh, N.; Gladysz, J. A.
Organometallics 2012, 31, 5231. (g) Urankar, D.; Pevec, A.; Kosmrlj, J.
Cryst. Growth Des. 2010, 10, 4920. (h) Younes, A. H.; Clark, R. J.; Zhu,
L. Supramol. Chem. 2012, 24, 696. (i) Crowley, J. D.; Gavey, E. L.
Dalton Trans. 2010, 39, 4035. (j) Scott, S. O.; Gavey, E. L.; Lind, S. L.;
Gordon, K. C.; Crowley, J. D. Dalton Trans. 2011, 40, 12117.
(k) Stevenson, K. A.; Melan, C. F. C.; Fleischel, O.; Wang, R.;
Petitjean, A. Cryst. Growth Des. 2012, 12, 5169. (l) White, N. G.; Beer,
P. D. Supramol. Chem. 2012, 24, 473.
AUTHOR INFORMATION
Corresponding Author
*E-mail: krzystek@magnet.fsu.edu (J.K.), jtelser@roosevelt.edu
(J.T.), mihail.atanasov@cec.mpg.de (M.A.), frank.neese@cec.
mpg.de (F.N.), biprajit.sarkar@fu-berlin.de (B.S.).
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are indebted to the Fonds der Chemischen Industrie (FCI)
for financial support of this project (Chemiefondsstipendium
for D.S.). Denis Bubrin is kindly acknowledged for the crystal
structure determination of 2 and Stephan Hohloch for 1′. We
are thankful to Sebastian Weber for synthetic help. The HFEPR
work was done at the National High Magnetic Field
Laboratory, which is funded by the NSF through the
Cooperative Agreement DMR 1157490, the State of Florida,
and the U.S. Department of Energy.
(8) (a) Gu, S.; Xu, H.; Zhang, N.; Chen, Y. Chem. Asian. J. 2010, 5,
1677. (b) Cambeiro, X. C.; Pericas, M. A. Adv. Synth. Cat. 2011, 353,
113. (c) Lalrempuia, R.; McDaniel, N. D.; Muller-Bunz, H.; Bernhard,
̈
S.; Albrecht, M. Angew. Chem., Int. Ed. 2010, 49, 9765. (d) Kilpin, K. J.;
Paul, U. S. D.; Lee, A.-L.; Crowley, J. D. Chem. Commun. 2011, 47,
328. (e) Saravanakumar, R.; Ramkumar, V.; Sankararaman, S.
Organometallics 2011, 30, 1689. (f) Hua, C.; Vuong, K. Q.;
Bhadbhade, M.; Messerle, B. A. Organometallics 2012, 31, 1790.
(9) (a) Schweinfurth, D.; Pattacini, R.; Strobel, S.; Sarkar, B. Dalton
Trans. 2009, 9291. (b) Schweinfurth, D.; Strobel, S.; Sarkar, B. Inorg.
Chim. Acta 2011, 374, 253. (c) Hohloch, S.; Su, C.-Y.; Sarkar, B. Eur. J.
Inorg. Chem. 2011, 3067.
REFERENCES
■
(10) (a) Schweinfurth, D.; Weisser, F.; Bubrin, D.; Bogani, L.; Sarkar,
B. Inorg. Chem. 2011, 50, 6114. (b) Schweinfurth, D.; Demeshko, S.;
Khusniyarov, M. M.; Dechert, S.; Gurram, V.; Buchmeiser, M. R.;
Meyer, F.; Sarkar, B. Inorg. Chem. 2012, 51, 7592. (c) Schweinfurth,
D.; Su, C.-Y.; Wei, S.-C.; Braunstein, P.; Sarkar, B. Dalton Trans. 2012,
41, 12984.
(11) (a) Krzystek, J.; Ozarowski, A.; Telser, J. Coord. Chem. Rev.
2006, 250, 2308. (b) Telser, J.; Ozarowski, A.; Krzystek, J. Specialist
Periodical Reviews of the Royal Chemical Society: Electron Paramagnetic
Resonance 2013, 23, 209.
(1) (a) Rostovtsev, V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int. Ed. 2002, 41, 2596. (b) Tornoe, C. V.; Christensen,
C.; Meldal, M. J. Org. Chem. 2002, 67, 3057. (c) Lau, Y. H.; Rutledge,
P. J.; Watkinson, M.; Todd, M. H. Chem. Soc. Rev. 2011, 40, 2848.
(d) Hua, Y.; Flood, A. H. Chem. Soc. Rev. 2010, 39, 1262.
(2) (a) Liang, L.; Astruc, D. Coord. Chem. Rev. 2011, 255, 2933.
(b) Struthers, H.; Mindt, T. L.; Schibli, R. Dalton Trans. 2010, 39, 675.
(c) Crowley, J. D.; McMorran, D. In Topics in Heterocyclic Chemistry;
Kosmrlj, J., Ed.; Springer: Berlin/Heidelberg, 2012; Vol. 28, pp 31.
(d) Schweinfurth, D.; Deibel, N.; Weisser, F.; Sarkar, B. Nachr. Chem.
2011, 59, 937. (e) Aromi, G.; Barrios, L. A.; Robeau, O.; Gamez, P.
Coord. Chem. Rev. 2011, 255, 485.
(3) (a) Li, Y.; Huffman, J. C.; Flood, A. H. Chem. Commun. 2007,
2692. (b) Meudtner, R. M.; Ostermeier, M.; Goddard, R.; Limberg, C.;
Hecht, S. Chem.Eur. J. 2007, 13, 9834. (c) Fletcher, J. T.;
Bumgarner, B. J.; Engels, N. D.; Skoglund, D. A. Organometallics
2008, 27, 5430. (d) Yang, W. W.; Wang, L.; Zhong, Y. W.; Yao, Y.
Organometallics 2011, 30, 2236. (e) Richardson, C.; Fitchett, C. M.;
Keene, F. R.; Steel, P. J. Dalton Trans. 2008, 2534. (f) Schulze, B.;
Friebe, C.; Hager, M. D.; Winter, A.; Hoogenboom, R.; Goerls, H.;
Schubert, U. S. Dalton Trans. 2009, 787. (g) Happ, B.; Friebe, C.;
(12) (a) Boca
̌ ̌
, R. Coord. Chem. Rev. 2004, 248, 757. (b) Titis, J.;
Boca, R. Inorg. Chem. 2010, 49, 3971.
̌
(13) Gispert, J. R. Coordination Chemistry: Wiley-VCH: Weinheim,
2008.
(14) (a) van Dam, P. J.; Klaassen, A. A.K.; Reijerse, E. J.; Hagen, W.
R. J. Magn. Reson. 1998, 130, 140. (b) Pardi, L.; Hassan, A. K.;
Hulsbergen, F. B.; Reedjik, J.; Spek, A. L.; Brunel, L. C. Inorg. Chem.
2000, 39, 159. (c) Collison, D.; Helliwell, M.; Jones, V. M.; Mabbs, F.
E.; McInnes, E. J. L.; Riedi, P. C.; Smith, G. M.; Pritchard, R. G.;
Cross, W. I. J. Chem. Soc. Faraday Trans. 1998, 94, 3019. (d) Krzystek,
J.; Park, J. H.; Meisel, M. W.; Hitchman, M. A.; Stratemeier, H.;
L
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Brunel, L. C.; Telser, J. Inorg. Chem. 2002, 41, 4478. (e) Rebilly, J. N.;
Charron, G.; Riviere, E.; Guillot, R.; Barra, A. L.; Serrano, M. D.; van
Slageren, J.; Mallah, T. Chem.Eur. J. 2008, 14, 1169. (f) Mrozinski,
J.; Skorupa, A.; Pochaba, A.; Dromzee, Y.; Verdaguer, M.; Goovaerts,
E.; Varcammen, H.; Korybut-Daszkiewicz, B. J. Mol. Struct. 2001, 559,
107. (g) Yang, E. C.; Kirman, C.; Lawrence, J.; Zakharov, L. N.;
Rheingold, A. L.; Hill, S.; Hendrickson, D. N. Inorg. Chem. 2005, 44,
3827. (h) Rogez, G.; Rebilly, J. N.; Barra, A. L.; Sorace, L.; Blondin, G.;
Kirchner, N.; Duran, M.; van Slageren, J.; Parsons, S.; Ricard, L.;
Marvilliers, A.; Mallah, T. Angew. Chem., Int. Ed. 2005, 44, 1876.
(i) Dobrzynska, D.; Jerzykiewicz, L. B.; Duczmal, M.; Wojciechowska,
A.; Jablonska, K.; Palus, J.; Ozarowski, A. Inorg. Chem. 2006, 45,
10479. (j) Desrochers, P. J.; Telser, J.; Zvyagin, S. A.; Ozarowski, A.;
Krzystek, J.; Vicic, D. A. Inorg. Chem. 2006, 45, 8930. (k) Wojcie-
chowska, A.; Daszkiewicz, M.; Staszak, Z.; Trusz-Zdybek, A.; Bienko,
A.; Ozarowski, A. Inorg. Chem. 2011, 50, 11532. (l) Charron, G.;
Bellot, F.; Cisnetti, F.; Pelosi, G.; Rebilly, J. N.; Riviere, E.; Barra, A. L.;
Mallah, T.; Policar, C. Chem.Eur. J. 2007, 13, 2774. (m) Maganas,
D.; Krzystek, J.; Ferentinos, E.; Whyte, A. M.; Robertson, N.;
Psycharis, V.; Terzis, A.; Neese, F.; Kyritsis, P. Inorg. Chem. 2012, 51,
7218.
(29) Becke, A. D. Phys. Rev. 1988, A38, 3098. Perdew, J. P. Phys. Rev.
1986, B33, 8822.
(30) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297.
(31) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
2010, 132, 154104.
(32) Neese, F.; Wennmohs, F.; Hansen, A.; Becker, U. Chem. Phys.
2009, 356, 98.
(33) Neese, F. J. Comput. Chem. 2003, 24, 1740.
(34) (a) Eichkorn, K.; TGreutler, O.; Ohm, H.; Haser, M.; Ahlrichs,
R. Chem. Phys. Lett. 1995, 242, 652. (b) Eichkorn, K.; Weigend, F.;
Treutler, O.; Ahlreichs, R. Theor. Chem. Acc. 1997, 97, 119.
(35) Klamt, A.; Schuurmann, G. J. Chem. Soc., Perkin Trans. 2 1993,
̈
̈
799.
(36) Portmann, S. Molekel, version 5.4.0.8; CSCS/UNI Geneva:
Geneva, Switzerland, 2009.
(37) CrysAlis CCD, version 1.171.31.7; Oxford Diffraction Ltd.
(38) CrysAlis RED, version 1.171.31.7; Oxford Diffraction Ltd.
(39) SHELXTL, version 5.10; Bruker Analytical X-ray Systems:
Madison, WI, 1998.
(15) (a) Atanasov, M.; Ganyushin, D.; Sivalingam, K.; Neese, F.
Struct. Bonding (Berlin, Ger.) 2012, 143, 149. (b) Atanasov, M.;
Ganyushin, D.; Pantazis, D. A.; Sivalingam, K.; Neese, F. Inorg. Chem.
2011, 50, 7640. (c) Atanasov, M.; Zadrozny, J. M.; Long, J. R.; Neese,
F. Chem. Sci. 2013, 4, 139. (d) Atanasov, M.; Comba, P.; Helmle, S.;
Muller, D.; Neese, F. Inorg. Chem. 2012, 51, 12324. (e) Neese, F.
̈
WIREs Comput. Mol. Sci. 2012, 2, 73.
(16) Malmqvist, P.-Å; Roos, B. O. Chem. Phys. Lett. 1989, 155, 189.
(17) (a) Angeli, C.; Cimiraglia, R.; Malrieu, J.-P. Chem. Phys. Lett.
2001, 350, 297. (b) Angeli, C.; Cimiraglia, R.; Evangelisti, S.;
Leininger, T.; Malrieu, J.-P. J. Chem. Phys. 2001, 114, 10252.
(c) Angeli, C.; Cimiraglia, R.; Malrieu, J.-P. J. Chem. Phys. 2002,
117, 9138. (d) Borini, S.; Cestari, M.; Cimiraglia, R. J. J. Chem. Phys.
2004, 121, 4043. (e) Angeli, C.; Bories, B.; Cavallini, A.; Cimiraglia, R.
J. Chem. Phys. 2006, 124, 54108.
(18) Hein, J. E.; Krasnova, L. B.; Iwasaki, M.; Fokin, V. V. Org. Synth.
2011, 88, 238.
(19) Krzystek, J.; Zvyagin, S. A.; Ozarowski, A.; Trofimenko, S.;
Telser, J. J. Magn. Reson. 2006, 178, 174.
(20) Ballhausen, C. J. In Introduction to Ligand Field Theory;
McGraw-Hill: New York, 1962; pp 99−103.
(21) (a) Lovecchio, F. V.; Gore, E. S.; Busch, D. H. J. Am. Chem. Soc.
1974, 96, 3109. (b) Martin, L. Y.; Sperati, C. R.; Busch, D. H. J. Am.
Chem. Soc. 1977, 99, 2968.
(22) Bendix, J.; Brorson, M.; Schaeffer, C. E. Inorg. Chem. 1993, 32,
2838.
(23) (a) Liakos, D. G.; Ganyushin, D.; Neese, F. Inorg. Chem. 2009,
48, 10572. (b) Duboc, C.; Ganyushin, D.; Sivalingam, K.; Collomb,
M.-N.; Neese, F. J. Phys. Chem. A 2010, 114, 10750. (c) Maganas, D.;
Sottini, S.; Kyritsis, P.; Groenen, E. J. J.; Neese, F. Inorg. Chem. 2011,
50, 8741. (d) Domingo, A.; Carvajal, M. A.; de Graaf, C.; Sivalingam,
K.; Neese, F.; Angeli, C. Theor. Chem. Acc. 2012, 131:1264, 1.
(24) Bencini, A.; Benelli, C.; Gatteschi, D. Coord. Chem. Rev. 1984,
60, 131.
(25) Jørgensen, C. K.; Pappalardo, R.; Schmidtke, H.-H. J. Chem.
Phys. 1963, 39, 1422.
(26) Schaffer, C. E.; Jørgensen, C. K. Mol. Phys. 1965, 9, 401.
̈
(27) Hassan, A. K.; Pardi, L. A.; Krzystek, J.; Sienkiewicz, A.; Goy, P.;
Rohrer, M.; Brunel, L. C. J. Magn. Reson. 2000, 142, 300−312.
(28) Neese, F. with contributions from Becker, U.; Ganyushin, G.;
Hansen, A.; Izsak, R.; Liakos, D. G.; Kollmar, C.; Kossmann, S.;
Pantazis, D. A.; Petrenko, T.; Reimann, C.; Riplinger, C.; Roemelt, M.;
Sandhofer, B.; Schapiro, I.; Sivalingam, K.; Wennmohs, F.; Wezisla, B.
̈
and contributions from our collaborators: Kal
́
lay, M.; Grimme, S.;
Valeev, E. ORCA - An ab initio, DFT and semiempirical SCF-MO
package, Version 2.9.1; Mulheim a.d.R. The binaries of ORCA are
̈
available free of charge for academic users for a variety of platforms.
M
dx.doi.org/10.1021/ic3026123 | Inorg. Chem. XXXX, XXX, XXX−XXX