R. Herges, F. Tuczek et al.
19F NMR (470 MHz, [D8]toluene, 258C, CFCl3): d=ꢀ136.82 (br, Fortho),
ꢀ152.97 (br, Fpara), ꢀ163.21 ppm (br, Fmeta).
sis, and the plot of cm (cm3 molꢀ1) versus Tꢀ1 (Kꢀ1) exhibits a straight line
with a slope of 1.15ꢁ0.03.[13] According to Equation (4):
Titration: The paramagnetism of the Ni2+ in the penta- and hexacoordi-
nated complexes gave rise to a strong downfield shift and a broadening
of the pyrrole proton resonance in porphyrins such as NiTPPF20 (1). Due
to the very rapid coordination/decoordination process, the time-averaged
Nm2B
1
Nm2
3k
1
T
ð4Þ
cm
¼
m2
¼
B g2SðS þ 1Þ
3k eff T
1
in which N is the Avogadro constant (6.022ꢃ1023 molꢀ1), mB the Bohr
magneton, meff the effective magnetic moment, k the Boltzmann constant,
g the Landꢅ factor, S the spin quantum number, and T the temperature
(K), a slope of 1 is expected for two unpaired electrons. The observed
slope of 1.15 corresponds to a magnetic moment of 3.0, in good agree-
ment with the above room-temperature value of cm.
signal of the dia- and paramagnetic species was observed in the H NMR
spectra. The signal of the pyrrole protons was followed as a function of
the axial ligand concentration (25 different ligand concentrations and
four different pyridines: 2a, 2c–e). K1S and K2 were determined from
these data by using the linear analysis outlined below. All titration ex-
periments were performed at four different temperatures (298, 308, 318,
and 328 K) to determine reaction enthalpies and entropies. More than
Single-crystal structure analysis: The investigation was performed with an
imaging plate diffraction system (IPDS-1) with MoKa radiation from
STOE & CIE. The structure solutions were carried out by direct methods
1
400 H NMR spectra were measured with an automatic sampler. A differ-
ent NMR tube was used for each ligand concentration. Each 1H NMR
tube was filled with 200 mL of a porphyrin solution (10.0 mmolLꢀ1) in
2
with SHELXS-97. These refinements were performed against jFj by
[D8]toluene. Then different amounts of
a solution of ligand in
using SHELXL-97.[29] All non-hydrogen atoms except some of the disor-
dered C atoms in compound 2b were refined with anisotropic displace-
ment parameters. All hydrogen atoms were positioned with idealized ge-
ometry and were refined with fixed isotropic displacement parameters
[Ueq(H)=ꢀ1.2ꢃUeq(C)] by using a riding model with dCꢀH =0.93 ꢄ. In
compound 2a one of the two crystallographically independent toluene
molecules is disordered and in compound 2b both molecules are disor-
dered. They were refined by using a split model. Details of the structure
determination are given in the Supporting Information. CCDC 766988
(2a), 766989 (2b) and 766990 (2 f) contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of charge
ac.uk/data_request/cif.
[D8]toluene (10 vol%) were added and the tubes were filled up to
600 mL with [D8]toluene and sealed. NMR titration experiments could
not be performed for the pyridines 2b and 2 f because of solubility prob-
lems.
Susceptibility measurements after Evans: To quantify the paramagnetic
properties of our NiII complexes, we determined their magnetic suscepti-
bility by using the “Evans method”.[11, 24–26] A set of two coaxial NMR
tubes consisting of an ordinary outer NMR tube and an inner tube with a
smaller diameter was used. The inner tube contained the reference sol-
vents [D5]pyridine and 2 vol% tert-butanol. The outer tube was filled
with the same mixture and additionally with paramagnetic complex.
[D5]Pyridine was used as the solvent as well as the axial ligand to make
sure that the hexacoordinated 2:1 complex was formed. A concentration
of m=0.0113 gcmꢀ3 of porphyrin 1 in [D5]pyridine was applied for the
actual Evans experiment. The mass susceptibility cg of the complex was
calculated from the difference of the chemical shifts Df of the tert-butanol
proton signals in the inner and outer tubes. A downfield shift of 60.5 Hz
was measured for the paramagnetic outer solution. The susceptibility was
corrected by the diamagnetic susceptibility of the solvent and the Ni-por-
phyrin [Eq. (1)]:
Derivation of Equation (7): We derived an equation for the determina-
tion of the equilibrium constants K1S and K2 from NMR and UV titration
data. Usually the coordination of one ligand without spin change is de-
scribed by K1, and if there is also a spin crossover during this process the
equilibrium constant is denoted K1S.[10] Equilibrium constant K2 describes
coordination of a second ligand to the five-coordinate, high-spin (HS)
nickel complex. The simultaneous coordination of two ligands to a four-
coordinate Ni center is described by b [Eqs. (a)–(d)]:
3Df
K1
ð1Þ
cg
¼
þ cg;pyr ꢀ cg;Ni
4pfm
Ni þ L
Ni þ L
NiL
ðaÞ
ðbÞ
ðcÞ
ðdÞ
!
LS
K1S
NiL
!
HS
in which f is the frequency of the NMR instrument (Hz), Df the frequen-
cy shift of tert-butanol (Hz), m the concentration of porphyrin 1 (gcmꢀ1),
cg,pyr the mass susceptibility of solvent [D5]pyridine (cm3 gꢀ1), and cg,Ni the
K2
NiLHS þ L
NiL
!
2
diamagnetic correction for porphyrin 1 (cm3 gꢀ1).
b
Ni þ 2L NiL
!
The molar susceptibility of pyridine is known (ꢀ49.2ꢃ10ꢀ4 cm3 molꢀ1
)
[27]
2HS
and the diamagnetic correction of porphyrin 1 was calculated from an in-
crement system.[28] The molar susceptibility cg =3.14ꢃ10ꢀ3 cm3 molꢀ1 was
calculated from the mass susceptibility and the molecular weight M of
the complex [Eq. (2)]:
It is generally accepted that the quadratic pyramidal 1:1 complexes of Ni-
porphyrins as well as the quadratic bipyramidal 2:1 complexes are high
spin. Thus, the complexation of the first axial ligand is accompanied by a
spin crossover from low to high spin. The validity of this approach is
shown by the excellent regression coefficient obtained when fitting exper-
imental data to equations based on these assumptions. La Mar and
Walker derived an equation which describes the ratio between paramag-
netic and diamagnetic species [Eq. (5)]:[11]
cm ¼ cgM
ð2Þ
The magnetic moment m in Bohr magnetons (B.M.), with T=tempera-
ture, is defined as follows [Eq. (3)]:
½paraꢄ
½diaꢄ
d ꢀ d0
½NiL2HSꢄ þ ½NiLHSꢄ
pffiffiffiffiffiffiffiffiffi
¼
¼
ð5Þ
ð3Þ
m ¼ 2:828 cmT
dmax ꢀ d
½NiLSꢄ þ ½NiLLSꢄ
The magnetic moment calculated from our data at 300 K of 2.9 B.M. is in
good agreement with the literature values for two unpaired electrons in
an octahedral ligand field. Depending on the contribution of the orbital
moment, values of 2.8 to 3.5 B.M. were found for the magnetic moment
m.
in which d0 is the chemical shift of the pyrrole protons of the porphyrin
in the absence of an axial ligand, d is the chemical shift of the pyrrole
protons at different ligand concentrations, dmax is the chemical shift of the
pyrrole protons of completely coordinated porphyrin, [NiL2HS] is the con-
centration of the hexacoordinate high-spin complex, [NiLHS] is the con-
centration of the pentacoordinate high-spin complex, [NiLS] is the con-
centration of the low-spin porphyrin, [NiLLS] is the concentration of the
pentacoordinate low-spin complex, and [L] is the concentration of the
ligand.
The magnetic moment m was also measured at 12 different temperatures
(from 293 to 283 K in increments of 10 K) to make sure that the number
of unpaired electrons was not temperature dependent. A plot of the ob-
served shift difference Df as a function of T does not exhibit any hystere-
10082
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10074 – 10083