A New Class of Lanthanide Complexes with Three Ligand Centered Radicals: NMR Evaluation of Ligand Field Energy Splitting and Magnetic Coupling

Article abstract of DOI:10.1002/chem.201901388

Combination of three radical anionic Ph-BIAN ligands (Ph-BIAN=bis-(phenylimino)-acenaphthenequinone) with lanthanoid ions leads to a series of homoleptic, six-coordinate complexes of the type Ln(Ph-BIAN)₃. Magnetic coupling data were measured by paramagnetic solution NMR spectroscopy. Combining ¹H NMR with ²H NMR of partially deuterated compounds allowed a detailed study of the magnetic susceptibility anisotropies over a large temperature range. The observed chemical shifts were separated into ligand- and metal-centered contributions by comparison with the Y analogue (diamagnetic at the metal). The metal-centered contributions of the complexes with the paramagnetic ions could then be separated into pseudocontact and Fermi contact shifts. The latter is large within the Ph-BIAN scaffold, which shows that magnetic coupling is significant between the lanthanide ion and the radical ligand. Pseudocontact shifts were further correlated to structural data obtained from X-ray diffraction experiments. Ligand-field parameters were determined by fitting the temperature dependence of the observed magnetic susceptibility anisotropies. The electronic structure determined by this approach shows, that the Er and Tm analogues are candidates for single molecule magnets (SMM). These results demonstrate the possibilities for the application of NMR spectroscopy in investigations of paramagnetic systems in general and single molecule magnets in particular.

Full text of DOI:10.1002/chem.201901388

A Journal of
Accepted Article
Title: A New Class of Lanthanide Complexes with three Ligand
Centered Radicals: NMR Evaluation of Ligand Field Energy
Splitting and Magnetic Coupling
Authors: Markus Hiller, Thomas Sittel, Hubert Wadepohl, and Markus
Enders
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201901388
Link to VoR: http://dx.doi.org/10.1002/chem.201901388
Supported by

10.1002/chem.201901388
Chemistry - A European Journal
A New Class of Lanthanide Complexes with three
Ligand Centered Radicals: NMR Evaluation of Ligand
Field Energy Splitting and Magnetic Coupling
Markus Hiller1, Thomas Sittel1, Hubert Wadepohl1, and Markus Enders*1
1 Institute of Inorganic Chemistry, Heidelberg University, Im Neuenheimer Feld 270,
69120 Heidelberg, Germany
* To whom correspondence should be addressed. Tel: +49-6221-546247, Fax: +49-6221-541616247,
E-mail: markus.enders@uni-heidelberg.de
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
ABSTRACT
Combination of three radical anionic Ph-BIAN ligands (Ph-BIAN
=
bis-(phenylimino)-
acenaphtenequinone) with Lanthanoid ions leads to a series of homoleptic, six-coordinate complexes
of the type Ln(Ph-BIAN)3. Magnetic coupling data were measured by paramagnetic solution NMR
spectroscopy. Combining 1H NMR with 2H NMR of partially deuterated compounds allowed a detailed
study of the magnetic susceptibility anisotropies over a large temperature range. The observed
chemical shifts were separated into ligand- and metal-centered contributions by comparison with the
Y analogue (diamagnetic at the metal). The metal-centered contributions of the complexes with the
paramagnetic ions could then be separated into pseudocontact and Fermi contact shifts. The latter is
large within the Ph-BIAN scaffold, which shows that magnetic coupling is significant between the
lanthanide ion and the radical ligand. Pseudocontact shifts were further correlated to structural data
obtained from X-ray diffraction experiments. Ligand field parameters were determined by fitting the
temperature dependence of the observed magnetic susceptibility anisotropies. The electronic
structure determined by this approach shows, that the Er and Tm analogues are candidates for single
molecule magnets (SMM). These results demonstrate the possibilities for the application of NMR
spectroscopy in investigations of paramagnetic systems in general and single molecule magnets in
particular.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
INTRODUCTION
Single molecule magnets (SMMs) are a class of compounds capable of retaining molecular
magnetization below a certain blocking temperature TB. In order to achieve this behavior, a bistable
ground state is necessary, combined with a preferably large thermal barrier preventing the flip of the
magnetization. While the effect was first discovered in d metal compound (“Mn12OAc”)[1], in recent
years lanthanide ions have increasingly entered the focus of research and SMM behavior was observed
for compounds featuring only a single paramagnetic ion.[2] In contrast to many d metal compounds the
lanthanide ions feature large orbital contributions to the total angular momentum of the unpaired
electrons, in particular in the second half of the series (Tb3+–Yb3+), where spin-orbit-coupling is
additive. The key factor for achieving bistability in lanthanide based SIMs is the ligand field which
removes the degeneracy of the 2J+1 microstates.[3] Depending on the specific geometry and the type
of ligand, this splitting may favor microstates with highest, intermediate or even lowest possible mJ
values. Furthermore, this preference depends on the ion and thus for a given ligand field environment
only certain ions are suitable candidates for SMMs.[3a] Attempts have been made to categorize the
lanthanide ions into oblate and prolate ones, based on the asphericity of their f electron density.[4]
While this approach is beneficial for the quick selection of promising candidates and has been applied
successfully in the discovery of the best-performing SMM system to date, it neglects that due to ligand
field effects the shape of the free ion ground state may not reflect that of the mJ state of lowest energy
in a coordination compound. Hence, the prolate or oblate character of a particular ion is subject to the
influence of the ligand field. One prominent example for this behavior is Dy3+ (J = 15/2) in [DyCp2]+,
DyPc2 and [Dy(COT)2]− systems (Cp = substituted cyclopentadienide, Pc = Phtalocyaninato, COT =
cyclooctatetraene). While in the first two cases the ground state is considered to be predominantly
|±13/2> (oblate)[5], it is assumed to be |±9/2> (weakly prolate) in the latter.[6] This illustration
demonstrates that insight into the electronic structure of the lanthanide ions is of paramount
importance in the field of SMM research. Consequently, the theoretical treatment of such systems by
means of CASSCF calculations is developing rapidly.[7] On the other hand there are also experimental
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
methods like SQUID magnetometry as well as torque magnetization measurements,[8] which provide
helpful information. It has been demonstrated that NMR techniques can be included as well,[5, 9] yet
examples for their application are sparse. In this report we want to demonstrate how NMR
spectroscopy alone can be used as a tool for the determination of magnetic parameters and the
identification of promising SMM candidates. We have therefore investigated a previously unknown
series of homoleptic lanthanide complexes featuring three monoanionic radical ligands.
The enforcement of desired coordination geometries of lanthanide ions is far more challenging than
in the case of d group metals. Reasons for this are the comparatively large radii and the strong Lewis
acidity. Thus, high coordination numbers are commonly observed, often combined with irregular
coordination polyhedra. While these observations by no means preclude SMM behavior, they are in
contrast to the comparatively highly symmetric ligand fields encountered in many high-performing
lanthanide SIMs.[7b] We have therefore intended to achieve highly symmetric coordination of the
lanthanide ions in combination with a relatively small coordination number of six. In addition to that,
we envisaged to use ligands containing unpaired electrons themselves in order to study the magnetic
coupling phenomena within the ligand scaffold and between ligand and metal centered unpaired
electrons. Such couplings are important for opening magnetic hysteresis loops as has been shown
recently by a number of papers.[10]
RESULTS
A new group of neutral lanthanide complexes was synthesized by combining a Ln3+ ion with three
bidentate, monoanionic bis-(phenylimino)-acenaphtenequinone (Ph-BIAN) ligands. This ligand was
first described in 1967[11] and can be converted to its radical anion by addition of one equivalent of
alkali metal[12] (potassium was used herein). The resulting intensely green metal salt can be isolated or
reacted in situ with anhydrous LnCl3 in tetrahydrofuran at room temperature or at elevated
temperature in toluene. To the best of our knowledge, the reaction of this ligand with lanthanide
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
halogenides has so far not been reported in a 3:1 stoichiometry.[13] Notably, similar vanadium
complexes with a related ligand have been described previously.[14]
Scheme 1: Synthesis of the Ln(Ph-BIAN)3 compounds and isolated yields.
As summarized in scheme 1, the Ln(Ph-BIAN)3 compounds were obtained in high yields. Due to their
neutral charge, they are soluble in common organic solvents and can therefore be isolated by
extraction with toluene as intensely red solids. The stoichiometry of the reaction should be met exactly
to avoid the formation of toluene-soluble byproducts. The compounds are very sensitive to air and
moisture, particularly in solution, and have to be handled accordingly. All compounds were
characterized by NMR spectroscopy and elemental analysis. The reaction sequence was further carried
out using a partially deuterated ligand (prepared from aniline-D5, for details see ESI). This allowed the
independent identification of the NMR signals arising from the acenaphthene backbone and the phenyl
groups. As the NMR relaxation rate in paramagnetic compounds depends on the gyromagnetic ratio
of the investigated nucleus, 2H gives much smaller line widths compared to 1H, which in turn facilitates
2
the detection of the NMR signals.[15] Furthermore, the fact that H has a quadrupole moment opens
the possibility of gaining structural geometric information from residual quadrupolar couplings (see
below).[16]
Single crystals of the toluene solvate Ho(Ph-BIAN)3·2 C7H8 were obtained from a toluene solution and
the solid-state molecular structure was determined by X-ray crystallography (see Figure 1, left). In the
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
crystals, Ho(Ph-BIAN)3 attains C2 point symmetry. However, the molecular structure is close to D3
symmetry as evidenced by the corresponding bond lengths and angles. The length of the
acenaphthene backbone carbon carbon bond C1-C2 [C25-C25', respectively] can be used as a measure
for the redox state of the ligands and is found to be 1.445(2) [1.446(3)] Å, which is in line with reported
values for a mono-anion.[12]
Figure 1: Left: Molecular structure of Ho(Ph-BIAN)3∙2 C7H8. Hydrogen atoms and co-crystallized toluene are
omitted for clarity; displacement ellipsoids are drawn at the 50% probability level. Selected bond lengths and
angles: Ho-N1: 2.3519(14), Ho-N2: 2.3792(14), Ho-N3: 2.3736(15) Å, N1-Ho-N2: 70.60(5), N3-Ho-N3': 71.04(7),
N1-Ho-N3': 93.39(5), N1-Ho-N2': 103.43(5), N2'-Ho-N3': 93.86(5). Right: Overlay of the solid-state structure of
Ho(Ph-BIAN)3 (orange) and a model with enforced D3 symmetry (blue).
The solution structure was explored further by DFT-optimization of the Y analogue. Y3+ was chosen as
it does not contribute to the paramagnetism of the molecule and has an ionic radius close to Ho3+ (Y3+:
115.8 pm, Ho3+: 115.5 pm for eight-fold coordination),[17] which places it in the center of the
investigated series. The comparison of the solid-state structure of Ho(Ph-BIAN)3 and the structure of 1
with enforced D3 symmetry (Figure 1, right) demonstrates that only small structural changes result in
a significantly higher symmetry. Moreover, almost no variation of the atom positions of the first
coordination sphere is required.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
NMR Analysis
From the perspective of NMR spectroscopy the combination of three organic radical ligands with a
paramagnetic metal ion may appear disadvantageous, as organic radicals are usually difficult to study
by NMR due to their very broad resonance lines. Line widths in NMR spectra of paramagnetic
molecules are strongly influenced by the electron relaxation rate.[18] Slow electron relaxation as
commonly observed in organic radicals or transition ions with half-filled shells (e.g. Mn2+, Gd3+), causes
extremely fast relaxation of the NMR nuclei, usually increasing the NMR line widths beyond the
detection limit. However, if the organic radical is combined with a paramagnetic metal ion, the
unpaired electrons at the metal influence the relaxation of the ligand unpaired electron. This increases
the electron relaxation rates and in turn reduces the line broadening of the NMR nuclei. At the same
time, the interaction of the NMR nuclei with unpaired electrons of the metal leads to additional line
broadening. However, if the metal ion exhibits high electron relaxation rates, the overall line
broadening is smaller compared to NMR signals of purely ligand based paramagnetic compounds.
Consequently, the NMR detectability of organic radicals can be improved by coordinating them to
paramagnetic metal ions. The system studied here, presents an example of this case, which is
evidenced by the fact that NMR line widths are significantly larger for the analogous Y complex, with
some signals being undetectable (see below).
The 1H NMR spectra of the series Tb3+ – Tm3+ in toluene-d8 consist of only four relatively sharp signals
at room temperature with equal integrals. However, six or eight signals would be expected depending
on whether the rotation of the phenyl groups is fast or slow on the NMR timescale. The presence of
only four signals is due to the disappearance of the ortho- and meta-protons of the phenyl groups
caused by intermediate fast rotation of the phenyl rings with coalescence near room temperature. At
higher temperatures two new signals appear whereas at lower temperatures these split into four
signals (see Figures S5 and S6). The excellent solubility of the compounds was exploited for a detailed
measurement of the temperature dependence of the NMR shifts covering the range from 235 K to
365 K.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
In the case of the Y3+ analogue 1 four broad signals are observed as well. However, three of them stem
1
from the phenyl ring whereas only one H resonance of the three H atoms from the acenaphthene
moiety can be observed. In this compound, the phenyl ring rotation is fast on the NMR time scale
because the chemical shift difference of the interconverting H atoms (ortho-phenyl and meta-phenyl,
respectively) is small due to the absence of lanthanide-centered pseudocontact shift. This
2
interpretation is supported by the H NMR of the deuterated analogue where three signals from the
phenyl rings are present at room temperature. On the other hand, the unfavorable NMR relaxation
properties of organic radicals lead to significant line broadening of the resonances of the 1H atoms in
the acenaphthene group so that only one (H4) out of three NMR signals is observable (see Fig. 3).
1 provides the possibility to investigate the interaction among the three radical ligands. If their spins
are oriented parallel, a quartet spin state results whereas an anti-parallel orientation gives a spin-
frustrated doublet state. Which of the two possibilities is lower in energy depends on whether the
spin-spin coupling interaction is ferromagnetic or antiferromagnetic in nature.
1
The H NMR spectrum of Y(Ph-BIAN)3 (Figure S1,) changes only slightly with temperature, in analogy
to the description of a related vanadium complex featuring highly similar ligands.[14] This behavior
disagrees with a pure doublet or quartet state, for which a T−1 dependence of the chemical shifts is
expected. However, the observed temperature dependence is in agreement with a doublet ground
state and quartet state close in energy. Upon increase of the temperature, the Boltzmann occupancy
of the quartet state increases, for which higher chemical shifts occur resulting from the larger number
of unpaired electrons. Simultaneously, the T−1 dependence of the chemical shifts of both the ground
and excited states counteracts this variation, leading to only minor net changes of the observed
spectra. The fit of the observed chemical shifts provided a value of 288.0 ± 4.5 cm−1 for the energy
separation in favor of the doublet state. This corresponds to an antiferromagnetic coupling constant
of J = −(96 ± 1.5) cm−1, resulting in a nearly linear increase of the occupancy of the quartet state from
14.7 % to 24.2 % in the investigated temperature range. In addition, the ligand-based Fermi contact
contributions associated with both spin states were obtained from the fitting procedure, which is
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
described in more detail in the ESI. In order to assign the signals, the spin densities for the respective
nuclei were derived from DFT calculations of Y(Ph-BIAN)3. For this purpose, four different functionals
(B3LYP, TPSSh, PBE and BP86) were employed in geometry optimizations of the complex in both spin
states. The largest proton spin densities were in all cases found for the H3 and H5 positions (compare
Figure 2). As a consequence, fast relaxation broadens the NMR signals of these nuclei beyond the
detection limit. For the H4 and H10 nuclei positive spin densities were calculated while the respective
values are negative for H9 and H11, agreeing with the recorded NMR spectrum of 1 (Figure 2).
Comparison with the spectrum of the partially deuterated analogue allowed the distinction of H4 and
H10, while the lower relative integral of H11 allows the differentiation from H9.
Figure 2: Top left: Calculated (DFT, TPSSh) versus experimental chemical shifts of 1 at 296.2 K containing 19.5 %
of the quartet state. Top right: Comparison of the temperature-dependent chemical shifts of 1 and the values
fitted corresponding to a coupling constant of J = −96 cm−1. Values are plotted against the inverse temperature
to emphasize the deviation from Curie behavior. Bottom: Spin density distribution of the doublet (left) and
quartet (right) spin states of 1.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
Valuable information about the magnetic anisotropies of the respective lanthanide ions can be
extracted from the NMR spectra. The observed chemical shifts are influenced by the paramagnetism
of both the radical anions and the lanthanide ions. Both contributions can be separated into a Fermi-
contact and a pseudocontact shift.[19] The complete description is given in Eq. 1.
푖,푗
표푏푠
= 훿표푖푟푏 + 훿푙푖푓푐 + 훿푙푖푝푐 + 훿푚푖,푗푓푐 + 훿푚푖,푗푝푐
(1)
훿
푖,푗
표푏푠
The observed chemical shift 훿 of a nucleus i in the complex of lanthanide j thus consists of the orbital
shift contribution 훿표푖푟푏, the ligand-based Fermi-contact-contribution 훿푙푖푓푐 as well as the ligand-based
pseudocontact contribution 훿푙푖푝푐 and, finally, the metal-based Fermi-contact 훿푖,푗 and pseudocontact
푚푓푐
푖,푗
푚푝푐
shift 훿
. Please note, that the analysis in this way assumes a simple additive behavior of all shift
contributions. In order to extract information about the magnetic properties of the lanthanide ion, in
particular the magnetic anisotropy, the separation of the individual shifts is necessary. The sum of the
first three terms in eq. 1 corresponds to a system without unpaired f-electrons (i.e. the Y3+, La3+ or Lu3+
derivatives). The larger (or smaller, respectively) ionic radii of La3+ and Lu3+ as compared to the series
from Tb3+ to Tm3+ leads to considerable differences in the inter-ligand distances, thus significantly
modifying the size of the magnetic coupling between the ligands. The presence of energetically
accessible f-electrons should not influence the interligand coupling seriously, as f orbitals are shielded
rather effectively by the more outward-lying, larger d orbitals. The size of the Y3+ ion is in the middle
of the investigated series (Tb3+ to Tm3+) and thus we choose complex 1 as reference system. Therefore,
eq. 1 can be rewritten as eq. 2.
푖,푗
표푏푠
푖,푗
푖,푗
푖,푌
표푏푠
(2)
훿
= 훿
+ 훿
+ 훿
푚푝푐
푚푓푐
푖,푗
푚푝푐
The metal-based pseudocontact shift 훿
of a lanthanide ion j is described with Eq. 3 for an axial
system.[20] This contribution depends on the axial anisotropy of the magnetic susceptibility (∆휒푎푗푥) as
well as the metal-nucleus distance ri and the polar angle 휃푖 of the metal-nucleus vector with respect
to the magnetic main axis. This geometric information is often combined in the geometrical factor Gi.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
2
(
)
1
3푐표푠 휃푖 − 1
1
푖,푗
푚푝푐
(3)
훿
=
∙ ∆휒푎푗푥
∙
=
∙ ∆휒푎푗푥 ∙ 퐺푖
푟3
12휋
12휋
푖
For compounds 2 – 7 we assume that the effective magnetic main axis lies along the approximated
molecular C3 axis. This assumption is in line with previous calculations for lanthanide ions in trigonal-
prismatic environments.[21] Tilting of the axis should produce a higher number of NMR signals unless
these are averaged by rotation. This process would influence individual nucleus positions in a different
way due to the angular and radial dependence of the geometric factor. Consequently, if all metal-
푖,푗
푚푝푐
centered pseudocontact shifts 훿
of a specific lanthanide ion can be simultaneously fitted to a set
of geometric factors Gi of a geometric model (calculated with respect to a specific magnetic main axis)
and a single Δχaxj value, the direction of the effective magnetic axis is correct. It is further assumed that
the contraction of the ionic radii along the investigated series is negligibly small, so that all compounds
can be described by the same set of geometrical factors Gi.
푖,푗
푚푓푐
The metal-based Fermi contact shift 훿
is expressed by eq. 4 involving the reduced average spin
〈 〉 [22]
polarization 푆푧 . The shift further depends on the temperature T, the gyromagnetic ratio γi of the
investigated nucleus and the hyperfine coupling constant Ai/ħ for this nucleus (in units of rad·s–1); μB
represents the Bohr magneton and k is Boltzmann’s constant.
휇퐵 퐴푖
푖,푗
푚푓푐
〈 〉푗
∙ 푆푧
(4)
훿
=
∙
3푘푇훾푖
ℏ
푖,푗
푚푓푐
훿
is usually small for most nuclei in Lanthanide compounds, as the unpaired electrons are located
within f-orbitals that do interact only weakly with the ligand atoms. Therefore, this contribution is
often disregarded, however, the presence of the radical ligands may enhance spin delocalization
between f-centered and ligand centered unpaired electrons. Accordingly, we included this
contribution in our analysis and combined eqs. 2, 3 and 4 to eq. 5:
휇퐵 퐴푖
∙
1
푖,푗
표푏푠
∙ ∆휒푎푗푥 ∙ 퐺푖
푖,푌
표푏푠
〈 〉푗
∙ 푆푧
(5)
훿
= 훿
+
+
3푘푇 ℏ
12휋
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
As the susceptibility anisotropies of the different lanthanide ions are of particular importance, a second
2
methodology to obtain their values was employed by investigation of the H NMR spectra of the
analogous complexes bearing partially deuterated ligands. Due to the large magnetic anisotropy a
partial orientation of the molecules in magnetic fields is induced. In the case of nuclei with quadrupole
moments (nuclear spin I > ½) this leads to a splitting of the observable signal into 2·I lines, i.e. a doublet
in the case of 2H.[16, 23] The size of the splitting is described by Eq. 6.
2
푒2푞푄
(
)
퐵0
2
(6)
|( |
)
|Δ휈푄| =
∙
∙ 3푐표푠 휙 − 1 ∙ ∆휒푎푥
20휇0푘푇
ℎ
In this expression B0 represents the applied magnetic flux density and µ0 is the vacuum permeability.
The second factor e²qQ/h is the nuclear quadrupole coupling constant of deuterium, which differs
slightly depending on the specific environment. Suitable values can be found in the literature and
herein a value of 186 ± 6 kHz has been used.[24],[25],[26] Similar to eq. 3 an angular dependence is
encountered. In this case, however, the angle of interest is measured between the magnetic main axis
and the electric field gradient surrounding the investigated nucleus. As each deuterium nucleus is
bound to only one carbon atom, the latter lies along the direction of the C–2H bond. The partial
orientation depends on the susceptibility anisotropy of the entire molecule, including diamagnetic and
paramagnetic contributions from the ligands. However, the ligand contributions to the total magnetic
anisotropy of the series is negligible, as the 2H NMR spectrum of 1 does not show a splitting or in other
words the RQC splitting in 1 is considerably smaller than the line width of the observed 2H resonance
line (ca. 40 Hz at 296 K, Lorentzian line shape).
Signal assignment
In order to obtain useful information, the correct assignment of the observed signals is required. Due
to the different contributions that are not easily estimated, this process is far from trivial but can be
achieved by some general considerations. Firstly, as described above, one of the observed signals
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
belongs to the proton in para-position of the phenyl group (H11, see Fig. 3). The signal of this nucleus
in the different complexes is easily identified by comparison with the 2H NMR spectra. The NMR shifts
of nuclei which are several bonds apart from the lanthanide ions are usually dominated by the
pseudocontact contribution, which is proportional to the susceptibility anisotropies of the individual
lanthanide ions. However, due to the unpaired electron in the BIAN scaffold, a stronger delocalization
of unpaired f-electron density to the 1H atoms in the BIAN backbone has to be considered. As shown
below this contribution may exceed 30 ppm for atoms H3 and H5. Nonetheless, pseudocontact
dominates the shift of H11 due to its large distance (in terms of bonds) from the lanthanide ion and the
non-participation of the phenyl group in ligand unpaired electron delocalization. The geometric factor
G for H11 is positive and consequently large positive NMR shifts for this nucleus correspond to large
positive Δχaxj values and vice versa. Therefore, an arrangement of the spectra in the order of increasing
chemical shift for the H11 proton (Fig. 3) represents the decreasing order of the Δχaxj values. This order
is found to be Tb3+ < Dy3+ < Ho3+ < Er3+ < Tm3+, which is consistent with values predicted by the
literature for this coordination polyhedron.[21] Comparison with the spectra of Y(Ph-BIAN)3 shows, that
the Δχaxj values switch signs between Ho3+ and Er3+, which is in line with the classification of oblate and
prolate ions as well as the values predicted for this coordination geometry.[4b, 21]
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
1
Figure 3: Left: overlay of H NMR spectra (15.1 T, 296 K, toluene-d8) of the Ln(Ph-BIAN)3 compounds (top to
bottom: Ln = Tb3+, Dy3+, Ho3+, Er3+ and Tm3+) with expansions showing the RQC splitting of the 2H NMR signal of
D11 in the partially deuterated analogues 3D30.– 7D30 Colored lines illustrate the shifts of the signals arising from
identical nuclei. Top right: Labeling scheme for the compounds with colored positions corresponding to the
1
2
signals in the NMR spectra. Bottom right: Stacked H and H NMR spectra (15.1 T, 296 K, toluene-d8) of 1 and
1D30.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
From this arrangement it is easily found, that the remaining signal exhibit similar shift patterns.
However, due to the different geometrical factors of these nuclei, the strength and direction of the
shift variations differs. The line widths of the signals support the observed patterns, which is most
clearly seen in the signals of the H3 group as these are the broadest for all lanthanide ions. It must be
noted, that the signal of H3 in Tb(Ph-BIAN)3 does notably deviate from the otherwise regular pattern.
1
This effect is a result of a sizeable metal-based Fermi contact shift. The simultaneous fitting of all H
NMR signals as described below leads to a separation of metal centered Fermi-contact and
pseudocontact shifts. The result for the complex Tb(Ph-BIAN)3 is presented in figure 4 (for charts of the
other complexes see the Supporting Information). Large Fermi-contact shifts induced by the unpaired
f-electrons is distributed onto the BIAN scaffold (mainly on H3 and H5) whereas H11 at the phenyl
substituent carries only very small contact shifts.
Figure 4: Shift contributions for Tb(Ph-BIAN)3 at 296.2 K with 0 = shift of Y(Ph-BIAN)3, fc = metal induced Fermi-
contact shift, pc = metal induced pseudocontact shift, calc = shift resulting from fitting, obs = observed chemical
shift.
1
All observable H NMR shifts were simultaneously fitted to an alternative expression of Eq. 5, as
described previously for substituted lanthanide-cyclooctatetraene systems.[20b, 22b, 27] For this purpose,
the geometric factors Gi are replaced by relative geometric factors G’i (see ESI for details), which is
fixed to unity for one nucleus chosen arbitrarily. Simultaneously, only relative magnetic susceptibility
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
anisotropies result. This treatment allows the analysis of the NMR shifts without presuming structural
information apart from isostructurality. The resulting relative geometric factors can then be compared
for different geometric models and used for exploring the solution averaged structure. Once a suitable
model has been identified, its geometric factors Gi can be used to translate the relative magnetic
j
susceptibility anisotropies to the physically meaningful Δχax values. The relative geometric factors
obtained from the fitting (ESI, Table S.14) were compared to the values obtained from the solid-state
structure of Ho(Ph-BIAN)3 (Fig. 5, right). The ligand situated on the molecular C2 axis has very similar
geometric factors as obtained by the NMR fitting procedure. Thus, the NMR determined solution
molecular structure of these paramagnetic complexes is very similar to the results of the X-ray
j
diffraction. Accordingly, the susceptibility anisotropies Δχax of the individual lanthanide ions were
calculated from the relative values (ESI, Table S.15) and they correspond in an excellent way to the
results of the analysis of the RQC splittings as well as the predictions by Mironov et al.[21] (Table 1, Fig.
5, left).
Table 1: Values of Δχaxj for the individual lanthanide ions in Ln(Ph-BIAN)3 at different temperatures obtained by
the fitting procedure.
Δχax [10-31 m3]
235.8 K
256.8 K
278.8 K
296.2 K
315.6 K
336.7 K
363.6 K
−9.43
± 2.54
−8.33
± 2.23
−7.36
± 1.96
−6.73
± 1.77
−6.09
± 1.56
−5.48
± 1.37
−4.88
± 1.18
Tb3+
Dy3+
Ho3+
Er3+
−8.48
± 2.03
−7.53
± 1.83
−6.67
± 1.63
−6.11
± 1.52
−5.56
± 1.41
−5.03
± 1.33
−4.48
± 1.22
−6.06
± 1.55
−5.20
± 1.33
−4.46
± 1.11
−3.98
± 0.98
−3.53
± 0.85
−3.09
± 0.73
−2.69
± 0.62
3.13
± 0.80
2.85
± 0.66
2.54
± 0.59
2.33
± 0.55
2.03
± 0.51
1.85
± 0.49
1.62
± 0.45
9.38
± 2.39
8.34
± 2.00
7.28
± 1.65
6.63
± 1.45
5.83
± 1.28
5.24
± 1.15
4.61
± 1.00
Tm3+
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
Figure 5: Left: Comparison of Δχaxj values from the fitting procedure (black) and the RQC splittings (blue) with the
values predicted in the literature.[21] Right: Comparison of relative geometric factors obtained from the fitting
procedure with those calculated from one of the ligands in the solid-state structure of Ho(Ph-BIAN)3.
Determination of Ligand field parameters
The low number of observable NMR signals indicates, the solution structure of the Ln(Ph-BIAN)3
compound series to be of higher symmetry than C2, as inferred from the solid-state structure of the
Ho analogue. The comparison with a D3 symmetric model (Figure 1) shows only minor deviations.
Considering only the first coordination sphere (i.e. the nitrogen donor atoms) an almost perfect
agreement of both structural models is found. Accordingly, a distorted trigonal prism is observed as
the coordination polyhedron of the lanthanide ions, which conforms to the point group D3d. The
associated ligand field Hamiltonian[28] is given in the operator equivalent[29] form in Eq. 7. Apart from
the three axial parameters 퐵푘0 (with k = 2,4,6) an additional parameter 퐵66 is encountered, leading to a
mixing of the different mJ states in the ground state wave functions ψi.
퐷
3푑
0
0
0
0
0
0
6
6
̂
̂
̂
̂
̂
(7)
퐻퐿퐹 = 퐵2 푂2 + 퐵4 푂4 + 퐵6 푂6 + 퐵6 푂6
푞
푞
푘
[3a]
〈 〉
The parameters 퐵푘 are composed of the ligand field parameters 퐴푘 푟 and the Stevens parameters
kJk, specific for the lanthanide ions, as described by Eq. 8.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
푞
푞
푘
〈 〉
퐵푘 = 퐴푘 푟 ∙ 푘퐽푘
(8)
For the determination of the ligand field parameters the approach recently applied in substituted
bis-cyclooctatetraenyl lanthanide double decker compounds was adapted.[30] By diagonalization of the
energy matrices, the contributions of the various mJ states to the wave functions ψi and their relative
energies were obtained. Subsequently, the magnetic susceptibility tensor components were calculated
using Eq. 9.[21, 31] Therein, the index p represents the direction in the magnetic axis system, i.e. either
̂
x, y or z, 퐽푝 is the associated total angular momentum operator and Ei is the relative energy of ψi.
̂
̂
〈
〉〈
〉
휓푖|퐽푝|휓푗 휓푗|퐽푝|휓푖
)
푖 [ 휓푖|퐽푝|휓푖 휓푖|퐽푝|휓푖 − 2푘푇 ∙
] ∙ 푒(
−퐸 /푘푇
푖
̂
̂
∑ 〈 ∑
〉〈
〉
푗≠푖
휇0휇퐵2 ꢀ퐽2
2푘푇
퐸푖 − 퐸
(9)
푗
휒푝푐푎푙푐
=
∙
)
푖 푒(
−퐸 /푘푇
푖
∑
푐푎푙푐
푎푥
Finally, calculated values for the axial anisotropies of the magnetic susceptibility (∆휒
obtained according to the definition in Eq. 10 for the different lanthanides j.
) can be
1
푗,푐푎푙푐
푎푥
= 휒푧푗,푐푎푙푐 − (휒푥푗,푐푎푙푐 + 휒푦푗,푐푎푙푐
)
(10)
∆휒
2
Ligand field parameters were obtained by minimization of the root-mean-square deviation of
푗,푐푎푙푐
푎푥
j
calculated (∆휒
) and experimental anisotropies (Δχax ). Following the approach used in the
푞
푘
literature[5, 9] a linear variation of the parameters 퐴푘 푟 with the atomic number was imposed,
reducing the number of parameters for the five investigated ions from 20 to 8. For the numeric
calculations a custom-written script running in Octave 4.2 was used. A more detailed description of
the fitting procedure and the script file can be found in the ESI. The results are presented in Table 2.
〈 〉
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
Table 2: Ligand field parameters for the individual ions obtained by the fitting procedure.
Parameter (cm−1)
Tb3+
Dy3+
Ho3+
Er3+
Tm3+
0〈 〉
−381 ± 7.6
−95 ± 3.8
−52 ± 2.6
850 ± 17
−422 ± 10
−107 ± 5.1
−56 ± 3.6
1008 ± 22
−463 ± 13
−120 ± 6.3
−60 ± 4.5
1165 ± 26
−504 ± 15
−133 ± 7.6
−63 ± 5.4
1323 ± 31
−544 ± 17
−146 ± 8.9
−67 ± 6.4
1480 ± 36
2
퐴2 푟
0〈 〉
4
퐴4 푟
0〈 〉
6
퐴6 푟
6〈 〉
6
퐴6 푟
The obtained parameters result in excellent agreement of the observed and calculated anisotropies as
0〈 〉
2
demonstrated in Fig. 6. Most notably, the primary parameter 퐴2 푟 is determined to be negative for
the Ln(Ph-BIAN)3 compounds series, indicating a predominantly equatorial ligand field. The large
6〈 〉
6
퐴6 푟 parameter results in a pronounced mixing of different mJ states with ΔmJ = 6 in the ground
state wave functions ψi (see Tables S.36–S.40).
Figure 6: Comparison of experimental susceptibility anisotropies (data points) with those calculated from the set
of ligand field parameters summarized in Table 2 (full lines).
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
Figure 7: Relative energies of the wave functions ψi in 3 – 7. Please note, that the lowest two Kramers
doublets for the Er compounds 6 are very close in energy.
Conclusion
A series of neutral, six-coordinate and homoleptic lanthanide complexes with three additional ligand-
centered unpaired electrons was synthesized. The neutral charge of the compounds results in excellent
solubility in solvents of low polarity like toluene, and due to the large temperature range of liquid
toluene, reliable magnetic coupling data could be extracted from paramagnetic solution NMR
spectroscopy. The yttrium derivative allowed the NMR experimental determination of the inter-ligand
coupling behavior of the anti-ferromagnetically coupled and thus spin-frustrated ligand environment
with J = −(96 ± 1.5) cm−1. NMR chemical shifts resulting from the paramagnetic ligand framework of
the Y derivative were then subtracted from the observed chemical shifts of the other lanthanide
complexes. This allowed the determination of the contributions of the paramagnetic lanthanide ions
to the NMR shifts in the crystal field created by the three BIAN ligands. A simultaneous fitting of the
1H NMR shifts of the complex series allowed the subsequent separation into Fermi-contact and
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
pseudocontact shifts induced by the unpaired f-electrons. The coupling between the unpaired
electrons at the Ph-BIAN scaffold and the f-electrons at the Ln3+ centers leads to large Fermi-contact
shifts at H atoms of the ligand. This observation is important for the enhancement of metal ligand
interactions in the design of new magnetic materials. Subsequently, the temperature behavior of the
NMR shifts and of residual quadrupolar couplings was determined as a basis for a ligand field
parameter fitting. The ligand field is of predominantly equatorial type as described by the negative
0〈 〉
2
3+
value of 퐴2 푟 . In such a system, the prolate ions Er and Tm3+ are expected to exhibit easy-axis
magnetic anisotropy., which is in agreement with the NMR data. The constitution of the individual
wave functions indicates a significant mixing of mJ states for the Er derivative, while a non-degenerate
ground state was derived for the Tm3+ ion.
Experimental Section
General
All reactions were carried out under standard Schlenk or Glovebox techniques. Anhydrous solvents
were obtained from a MBRAUN SPS-800 solvent purification system. Deuterated toluene for NMR
spectroscopy was dried by heating over K/benzophenone and distilled. Celite was heated for several
hours to 140 °C under high vacuum to remove traces of moisture. NMR spectra were recorded on a
Bruker AVANCE II NMR spectrometer (9.4 T, 400 MHz for 1H) or a Bruker AVANCE III NMR spectrometer
(15.1 T, 92.1 MHz for 2H). The temperature controlling unit was calibrated using a standard substance
(ethylene glycol in DMSO or methanol). Spectra were calibrated using the residual solvent resonance
reported at 2.08 ppm.[32] Elemental analyses were carried out at the Microanalytical laboratory of the
University Heidelberg using a vario MICRO cube (Elementar). Despite numerous attempts, the
elemental analyses could not be improved beyond the reported values. The fact that the determined
nitrogen content is lower than expected in all cases may indicate a systematic problem of combustion
analyses of these compounds.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
DFT calculations
All geometry optimizations and spin density calculations were carried out with the Gaussian 09
program suite (Rev. D.01)[33] applying the basis sets Stuttgart RSC Segmented/ECP[34] for Y3+ and
6-31g(d,p) for the remaining atoms. The solid-state structure of Ho(Ph-BIAN)3 was chosen as the
starting point of the calculations and Ho was replaced by Y. For the calculation of the spin densities of
1 with the various functionals, no symmetry restrictions were imposed.
Synthesis of Ph-BIAN
The ligand was synthesized according to known literature procedures. Direct reaction of
acenaphthenequinone with aniline[35] produced only the monocondensation product. Hence, it was
found necessary to synthesize the ligand via its ZnCl2 adduct followed by removal of the metal.[36] All
analytical data was identical to reported literature values.
Synthesis of D10-Ph-BIAN
The partially deuterated ligand was synthesized according to the same procedure but employing D5-
aniline, which was prepared by modifications of literature syntheses. A detailed description can be
found in the ESI.
Synthesis of Ln(Ph-BIAN)3 1 - 7
Ph-BIAN (typically 100 mg, 301 µmol) was dissolved in anhydrous thf (5 mL) and K metal (typically
11.8 mg, 301 µmol, 1.0 eq.) was added. After stirring at room temperature overnight, the respective
anhydrous lanthanide trichloride (100 µmol, 0.33 eq.) was added. The reaction was allowed to stir
overnight before all volatile material was removed under reduced pressure. The residue was then
extracted with toluene (5 mL), centrifuged and filtered. The solvent was removed under reduced
pressure to yield a dark red solid, which was washed with a small amount of pentane.
The partially deuterated analogues 1D30 – 7D30 were synthesized in the same way employing the
corresponding partially deuterated ligand in similar yields.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
Y(Ph-BIAN)3 (1)
Yield: 387 mg (356 µmol) from 500 mg Ph-BIAN, 71 %.
Elemental analysis: calcd. for C72H48N6Y: C 79.62, H 4.45, N: 7.74; found: C 79.65, H 5.18, N 7.35.
Gd(Ph-BIAN)3 (2)
Yield: 93 mg (81 µmol) from 100 mg Ph-BIAN, 80 %.
Elemental analysis: calcd. for C72H48N6Gd: C 74.91, H 4.19, N: 7.28; found: C 73.88, H 4.77, N 6.91.
Tb(Ph-BIAN)3 (3)
Yield: 65 mg (56 µmol) from 100 mg Ph-BIAN, 56 %.
Elemental analysis: calcd. for C72H48N6Tb: C 74.80, H 4.18, N: 7.27; found: C 75.04, H 4.57, N 6.37.
Dy(Ph-BIAN)3 (4)
Yield: 109 mg (94 µmol) from 101 mg Ph-BIAN, 93 %.
Elemental analysis: calcd. for C72DyH48N6: C 74.57, H 4.17, N: 7.25; found: C 74.50, H 4.78, N 6.27.
Ho(Ph-BIAN)3 (5)
Yield: 194 mg, (167 µmol) from 250 mg Ph-BIAN, 67 %.
Elemental analysis: calcd. for C72H48HoN6: C 74.41, H 4.16, N: 7.23; found: C 73.82, H 4.75, N 6.36.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
Er(Ph-BIAN)3 (6)
Yield: 113 mg (97 µmol) from 100 mg Ph-BIAN, 97 %.
Elemental analysis: calcd. for C72ErH48N6: C 74.26, H 4.15, N: 7.22; found: C 74.05, H 4.56, N 6.25.
Tm(Ph-BIAN)3 (7)
Yield: 107 mg (92 µmol) from 100 mg Ph-BIAN, 92 %.
Elemental analysis: calcd. for C72H48N6Tm: C 74.16, H 4.15, N: 7.21; found: C 74.05, H 4.49, N 6.92.
ACKNOWLEDGEMENTS
This work has been supported by the Fonds der Chemischen Industry with a Kekulé scholarship. The
authors acknowledge support by the state of Baden-Württemberg through bwHPC and the German
Research Foundation (DFG) through grant No. INST 40/467-1 FUGG.
REFERENCES
[1]
[2]
a) R. Sessoli, H. L. Tsai, A. R. Schake, S. Wang, J. B. Vincent, K. Folting, D. Gatteschi, G. Christou,
D. N. Hendrickson, J. Am. Chem. Soc. 1993, 115, 1804-1816; b) R. Sessoli, D. Gatteschi, A.
Caneschi, M. A. Novak, Nature 1993, 365, 141-143.
a) N. Ishikawa, M. Sugita, T. Ishikawa, S. Koshihara, Y. Kaizu, J. Am. Chem. Soc. 2003, 125, 8694-
8695; b) C. A. P. Goodwin, F. Ortu, D. Reta, N. F. Chilton, D. P. Mills, Nature 2017, 548, 439-
442; c) F. S. Guo, B. M. Day, Y. C. Chen, M. L. Tong, A. Mansikkamaki, R. A. Layfield, Angew.
Chem., Int. Ed. 2017, 56, 11445-11449; d) F.-S. Guo, B. M. Day, Y.-C. Chen, M.-L. Tong, A.
Mansikkamäki, R. A. Layfield, Science 2018, 362, 1400-1403; e) K. R. Meihaus, J. R. Long, J. Am.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
Chem. Soc. 2013, 135, 17952-17957; f) S. K. Gupta, T. Rajeshkumar, G. Rajaraman, R.
Murugavel, Chem. Sci. 2016, 7, 5181-5191.
a) A. Abragam, B. Bleaney, Electron Paramagnetic Resonance of Transition Ions, Oxford
University Press, 1970; b) S.-D. Jiang, B.-W. Wang, S. Gao, in Molecular Nanomagnets and
Related Phenomena, Vol. 164, Ch. 2, (Ed.: S. Gao), 2015, pp. 111-143.
a) J. Sievers, Zeitschrift fur Physik B: Condensed Matter 1982, 45, 289-296; b) J. D. Rinehart, J.
R. Long, Chem. Sci. 2011, 2, 2078-2085.
[3]
[4]
[5]
[6]
N. Ishikawa, M. Sugita, T. Okubo, N. Tanaka, T. Iino, Y. Kaizu, Inorg. Chem. 2003, 42, 2440-2446.
L. Ungur, J. J. Le Roy, I. Korobkov, M. Murugesu, L. F. Chibotaru, Angew. Chem., Int. Ed. 2014,
53, 4413-4417.
[7]
a) L. F. Chibotaru, in Molecular Nanomagnets and Related Phenomena, Vol. 164, Ch. 4, (Ed.: S.
Gao), 2015, pp. 185-231; b) L. Ungur, L. F. Chibotaru, Inorg. Chem. 2016, 55, 10043-10056.
M. Perfetti, E. Lucaccini, L. Sorace, J. P. Costes, R. Sessoli, Inorg. Chem. 2015, 54, 3090-3092.
a) N. Ishikawa, T. Iino, Y. Kaizu, J. Phys. Chem. A 2002, 106, 9543-9550; b) N. Ishikawa, J. Phys.
Chem. A 2003, 107, 5831-5835; c) A. Santria, A. Fuyuhiro, T. Fukuda, N. Ishikawa, Dalton Trans.
2019, 10.1039/C9DT00915A.
[8]
[9]
[10]
a) S. Demir, J. M. Zadrozny, M. Nippe, J. R. Long, J. Am. Chem. Soc. 2012, 134, 18546-18549; b)
S. Demir, M. Nippe, M. I. Gonzalez, J. R. Long, Chem. Sci. 2014, 5, 4701-4711; c) S. Demir, I.-R.
Jeon, J. R. Long, T. D. Harris, Coord. Chem. Rev 2015, 289–290, 149-176; d) S. Demir, M. I.
Gonzalez, L. E. Darago, W. J. Evans, J. R. Long, Nat. Commun. 2017, 8, 2144.
a) I. Matei, T. Lixandru, Buletinul Institutului Politehnic din Iaşi 1967, 13, 245; b) M. H. Normant,
Comptes Rendus Acad. Sci. C 1969, 268, 1811-1813.
[11]
[12]
[13]
I. L. Fedushkin, A. A. Skatova, V. A. Chudakova, G. K. Fukin, Angew. Chem., Int. Ed. 2003, 42,
3294-3298.
a) I. L. Fedushkin, O. V. Maslova, A. G. Morozov, S. Dechert, S. Demeshko, F. Meyer, Angew.
Chem., Int. Ed. 2012, 51, 10584-10587; b) I. L. Fedushkin, O. V. Maslova, A. N. Lukoyanov, G. K.
Fukin, Comptes Rendus Chimie 2010, 13, 584-592; c) I. L. Fedushkin, O. V. Maslova, E. V.
Baranov, A. S. Shavyrin, Inorg. Chem. 2009, 48, 2355-2357; d) I. L. Fedushkin, O. V. Maslova,
M. Hummert, H. Schumann, Inorg. Chem. 2010, 49, 2901-2910; e) I. L. Fedushkin, D. S.
Yambulatov, A. A. Skatova, E. V. Baranov, S. Demeshko, A. S. Bogomyakov, V. I. Ovcharenko, E.
M. Zueva, Inorg. Chem. 2017, 56, 9825-9833.
[14]
[15]
[16]
[17]
[18]
J. Bendix, K. M. Clark, Angew. Chem., Int. Ed. 2016, 55, 2748-2752.
I. Bertini, C. Luchinat, Coord. Chem. Rev 1996, 150, 77-110.
A. A. Bothner-By, P. J. Domaille, C. Gayathri, J. Am. Chem. Soc. 1981, 103, 5602-5603.
R. Shannon, Acta Cryst. A 1976, 32, 751-767.
a) G. N. La Mar, W. D. Horrocks Jr, R. H. Holm, NMR of Paramagnetic Molecules – Principles
and Applications, Academic Press, INC., New York and London, 1973; b) I. Bertini;, C. Luchinat,
in Physical Methods for Chemists (Ed.: R. S. Drago), Saunders College Publishing, 1992; c) J. D.
Satterlee, Concepts in Magnetic Resonance 1990, 2, 119-129.
[19]
a) I. Bertini, C. Luchinat, G. Parigi, Solution NMR of Paramagnetic Molecules. Applications to
Metallobiomolecules and Models, Vol. 2, Elsevier, 2001; b) J. D. Satterlee, Concepts in Magnetic
Resonance 1990, 2, 69-79.
[20]
[21]
[22]
a) I. Bertini, C. Luchinat, G. Parigi, Prog. Nuc. Magn. Res. Spectrosc. 2002, 40, 249-273; b) J. A.
Peters, J. Huskens, D. J. Raber, Prog. Nuc. Magn. Res. Spectrosc. 1996, 28, 283-350.
V. S. Mironov, Y. G. Galyametdinov, A. Ceulemans, C. Gꢀrller-Walrand, K. Binnemans, J. Chem.
Phys. 2002, 116, 4673-4685.
a) R. M. Golding, M. P. Halton, Aust. J. Chem. 1972, 25, 2577-2581; b) C. N. Reilley, B. W. Good,
R. D. Allendoerfer, Anal. Chem. 1976, 48, 1446-1458.
[23]
[24]
P. J. Domaille, J. Am. Chem. Soc. 1980, 102, 5392-5393.
R. Knorr, H. Hauer, A. Weiss, H. Polzer, F. Ruf, P. Löw, P. Dvortsák, P. Böhrer, Inorg. Chem. 2007,
46, 8379-8390.
[25]
E. W. Bastiaan, C. Maclean, P. C. M. Van Zijl, A. A. Bothner, in Annual Reports on NMR
Spectroscopy, Vol. Volume 19 (Ed.: G. A. Webb), Academic Press, 1987, pp. 35-77.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
[26]
Please note, that the value given by Knorr et. al. contains a typing error. The source cited by
the authors gives the correct value.
[27]
[28]
M. Hiller, M. Maier, H. Wadepohl, M. Enders, Organometallics 2016, 35, 1916-1922.
C. Görller-Walrand, K. Binnemans, in Handbook on the Physics and Chemistry of Rare Earths,
Vol. 23, Ch. 155, (Eds.: K. A. Gschneidner, L. Eyring), 1996, pp. 121-283.
[29]
[30]
[31]
K. W. H. Stevens, Proc. Phys. Soc. A 1952, 65, 209-215.
M. Hiller, S. Krieg, N. Ishikawa, M. Enders, Inorg. Chem. 2017, 56, 15285-15294.
a) M. Gerloch, R. F. McMeeking, J. Chem. Soc. Dalton Trans. 1975, 2443-2451; b) J. H. Van
Vleck, The Theory of Electric and Magnetic Susceptibilities, Oxford University Press, 1932.
G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb, A. Nudelman, B. M. Stoltz, J. E.
Bercaw, K. I. Goldberg, Organometallics 2010, 29, 2176-2179.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G.
Scalmani, V. Barone, B. Mennucci, 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. J. A.
Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N.
Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, 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. Cioslowski,
D. J. Fox, Gaussian 09, Revision D.01, Gaussian, Inc., Wallingford CT, 2013.
[32]
[33]
[34]
[35]
D. Andrae, U. Häußermann, M. Dolg, H. Stoll, H. Preuß, Theor. Chim. Acta 1990, 77, 123-141.
C. D. Nunes, P. D. Vaz, V. Felix, L. F. Veiros, T. Moniz, M. Rangel, S. Realista, A. C. Mourato, M.
J. Calhorda, Dalton Trans. 2015, 44, 5125-5138.
[36]
C. S. K. Mak, H. L. Wong, Q. Y. Leung, W. Y. Tam, W. K. Chan, A. B. Djurišić, J. Organomet. Chem.
2009, 694, 2770-2776.
This article is protected by copyright. All rights reserved.

10.1002/chem.201901388
Chemistry - A European Journal
For Graphical Abstract
3 + nf-electrons + Ligand Field: Magnetic anisotropy and J-coupling was measured by NMR
spectroscopy in a new series of homoleptic Lanthanide complexes. Temperature dependent
quadrupolar NMR splitting (RQC) and subsequent mathematical analysis describe the ligand field as
equatorial with energy splitting of the mJ manifold up to 1000 cm-1.
This article is protected by copyright. All rights reserved.