Trinuclear Complexes Derived from R/S Schiff Bases – Chiral Single-Molecule Magnets

Article abstract of DOI:10.1002/ejic.201601138

The employment of enantiomerically pure Schiff bases in manganese chemistry is revealed to be an excellent method to obtain chiral single-molecule magnets and has allowed the characterization of several pairs of enantiomers, for which the magnetic properties were investigated. The reported systems consist of Mn^III–Mn^II–Mn^IIIlinear trimers or Mn^III₃cations in a triangular arrangement including the first example of a μ₃-Cl bridge in an isolated manganese triangle.

Full text of DOI:10.1002/ejic.201601138

DOI: 10.1002/ejic.201601138
Full Paper
Chiral Magnets
Trinuclear Complexes Derived from R/S Schiff Bases – Chiral
Single-Molecule Magnets
Albert Escuer*[a] Julia Mayans,[a] Merce Font-Bardia,[b] Lorenzo Di Bari,[c] and
Marcin Górecki[c]
Abstract: The employment of enantiomerically pure Schiff which the magnetic properties were investigated. The reported
bases in manganese chemistry is revealed to be an excellent systems consist of MnIII–MnII–MnIII linear trimers or MnIII
3
method to obtain chiral single-molecule magnets and has al- cations in a triangular arrangement including the first example
lowed the characterization of several pairs of enantiomers, for of a μ3-Cl bridge in an isolated manganese triangle.
Introduction
Research on coordination clusters of paramagnetic 3d or 4f
cations has been key in the search for single-molecule magnets
(SMMs) or single-ion magnets (SIMs)[1] and more recently for
their catalytic,[2] bioinorganic,[3] and optical[4] properties. The
exploration of molecular systems that crystallize in chiral space
groups is still an emerging field, and the possible chiral organi-
zation of magnetic moments can produce quite exotic proper-
ties such as the predicted skyrmions[5] or chiral solitons.[6] The
Scheme 1. General structural formula for the H2sae ligand and its substituted
derivatives. The variation of the R1 and R2 groups allows the synthesis of
chiral ligands.
spontaneous crystallization of nonchiral components in chiral
groups or the resolution of racemic mixtures are unusual[7] or
experimentally difficult; therefore, syntheses with enantiomeri-
cally pure chiral ligands have become the optimal method to
as neutral mononuclear MnIV complexes,[9] one MnIII 1D sys-
reach systems of this kind.[8]
[11]
tem,[10] and several homometallic MnIIMnIII
,
2
MnIII (ring),[12]
4
The condensation of salicylaldehyde and 1,2-aminoethanol
yields the Schiff base 2-[(2-hydroxyethyl)iminomethyl]phenol
(H2sae, Scheme 1). H2sae is a popular ligand and has been em-
ployed widely in transition-metal and lanthanide coordination
chemistry (around 110 entries in the CCDC) owing to its good
chelating properties and its ability to generate polynuclear sys-
tems through the potentially bridging O-phenolate and O-alk-
oxido donors.
MnIII4 (butterfly),[12c,13] MnIII6, MnII MnIII4, and MnIII8 clusters[13,14]
2
as well as heterometallic[15] CuII MnIII and NiII MnIII clusters.
3
2
2
Among them, the magnetic properties of the tetranuclear cyclic
systems with general formula [MnIII4X4L4] (X = Cl, Br) have been
the most interesting, as they display ferromagnetic interactions
and SMM responses.[12b]
The substitution on the aromatic ring or the C atoms of the
ethyl fragment can increase the number of Schiff bases enor-
mously, and chirality can be induced if the substitution occurs
at the 1- or 2-position of the hydroxyethyl group (Scheme 1).
Following our work in this field, we have explored the react-
ivity in manganese chemistry of the enantiomerically pure li-
gands (R)- and (S)-2-[(2-hydroxy-1-phenylethyliminomethyl)-
phenol] (H2L1) and (R)- and (S)-2-[(3-hydroxy-1-phenylpropyl-
iminomethyl)phenol] (H2L2, Scheme 2).
The reactions of these ligands with different manganese salts
under appropriate reaction conditions allowed the characteriza-
tion of the first polynuclear derivatives of H2L1/H2L2 consisting
of three pairs of trinuclear chiral clusters with formulas
[Mn3(L1)2(PhCOO)4(MeOH)] [(R)-1·MeOH and (S)-1·MeOH],
[Mn3(L2)2(PhCOO)4] [(R)-2·5H2O and (S)-2·2MeCN·MeOH·
1.5H2O], and (Phgly)[Mn3(L1)3(μ3-Cl)(Cl)3] [(R)-3·0.5MeCN·
0.25MeOH·0.5H2O and (S)-3·0.5MeCN·0.5H2O; Phgly = phenyl-
The chemistry of manganese with nonchiral Schiff bases
such as H2sae or their substituted analogs has been partially
explored in the recent past to yield a variety of systems such
[a] Departament de Química Inorgànica i Orgànica,
Secció Inorgànica and Institute of Nanoscience (IN2UB)
and Nanotecnology, Universitat de Barcelona,
Av. Diagonal 645, 08028 Barcelona, Spain
E-mail: albert.escuer@ub.edu
http://www.ub.edu/inorgani/recerca/MagMol/magmol.htm
[b] Departament de Mineralogia, Cristallografia i Dipòsits Minerals and
Unitat de Difracció de R-X, Centre Científic i Tecnològic de la Universitat
de Barcelona (CCiTUB), Universitat de Barcelona,
Solé i Sabarís 1–3 08028 Barcelona, Spain
[c] Dipartimento di Chimica e Chímica Industriale, Università di Pisa,
Via Moruzzi 13, I-56124 Pisa, Italy
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under http://dx.doi.org/10.1002/ejic.201601138.
Eur. J. Inorg. Chem. 0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 2. Structures of H2L1 and H2L2 ligands (top) and their crystallograph-
ically established coordination mode in complexes 1–3 (bottom).
Figure 1. Top: partially labeled plot of 1 [common labels for (R)-1 and (S)-1].
Bottom: 1D arrangement of trinuclear units linked by intermolecular H bonds
(red dashed bonds). H atoms omitted for clarity. Color key: MnII orange, MnIII
dark green, O red, N navy, C grey.
glycinate]. Complexes 1 and 2 have linear {MnIIMnIII2(μ-O)2-
(RCOO)4} cores, whereas 3 has a triangular arrangement with an
unprecedented {MnIII3(RO)3(μ3-Cl)} linkage and SMM response.
ido atom of the neighboring clusters to afford a 1D arrange-
ment of trimers. The O(1w)···O(2 W) and O(1w)···O(11) distances
are 2.764 and 2.984 Å, respectively.
Results and Discussion
Structural Descriptions
Each pair of enantiomers is very similar; therefore, a common [Mn3(L2)2(PhCOO)4(MeOH)]·Solvent [(R)-2·5H2O,
structural description is provided for the corresponding R (S)-2·2MeCN·MeOH·1.5H2O]
enantiomer. The oxidation states of the manganese atoms have
been assigned on the basis of structural considerations and
bond valence sum (BVS) calculations. The bond parameters for
each pair of enantiomers are summarized in Table S2 for (R)-1
and (S)-1, Table S3 for (R)-2 and (S)-2, and Table S4 for (R)-3.
Some significant bond parameters for (S)-3 are also provided in
Table S4.
The trinuclear compounds (R)-2 and (S)-2 show the same gen-
eral formulas and connectivities as (R)-1 and (S)-1 but signifi-
cant differences in the manganese coordination spheres and
the arrangement of the ligands. A view of the complex is shown
in Figure 2, and the main bond parameters are summarized in
Table S3.
[Mn3(L1)2(PhCOO)4(MeOH)]·MeOH [(R)-1·MeOH and
(S)-1·MeOH]
Compounds (R)-1 and (S)-1 can be described as linear trinuclear
MnIII2MnII systems, in which the MnII cation is in the central
position and linked to both MnIII cations by two syn–syn carb-
oxylate bridges and one alkoxido bridge. A view of the complex
is shown in Figure 1, and the main bond parameters are sum-
marized in Table S2. The trinuclear system is almost linear [the
Mn(1)···Mn(2)···Mn(3) angle is close to 178°]. The divalent Mn(2)
cation shows an MnO6 octahedral environment, whereas the
trivalent Mn(1) cation exhibits an octahedral MnO5N coordina-
tion with the elongated Jahn–Teller axis directed toward the
O(5)-carboxylato atom and the coordinated methanol molecule,
and Mn(3) shows a square-pyramidal MnO4N environment with
the apical position occupied by the O(8) donor from one of the
carboxylate groups.
The angle between the main O10–O11–O12–N2 and O1–O2–
O3–N1 planes is 6.2°; thus, the easy axes of the two MnIII
cations, defined by the Mn3–O8 and O5–Mn1–O1w directions,
are approximately parallel.
Figure 2. Partially labeled plot of 2 [common labels for (R)-2 and (S)-2].
For 2, the two moieties of the complex are related by one
C2 axis, which results in a more symmetrical molecule than 1.
In this case, the trinuclear complex has a Mn(1)···Mn(2)···Mn-
(1′) angle of 135°, and the carboxylate ligands are placed on
the same side of the mean trimer plane. Mn(1) and symmetry-
related Mn(1′) are pentacoordinate with MnO4N square-pyrami-
The crystallization methanol molecule establish two H bonds dal environments, whereas the divalent Mn(1) cation is octa-
with the coordinated methanol molecule and the O(11) alkox- hedrally coordinated.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The angle determined by the mean planes that define the Comments on the Syntheses
base of the square pyramids (O1–O2–O3–N1 and the symmetry-
The reactions of H2L1 and H2L2 with manganese carboxylates
related counterpart) is 44.6°; consequently, the angle between
yielded the trinuclear MnIIMnIII systems 1 and 2, which show
2
the easy axes of the MnIII cations is also close to 45°.
the very stable {MnIIMnIII2(μ-O)2(RCOO)4} core. This kind of
complex is well known for most transition-metal cations; for
manganese, they often form from the reactions of manganese
carboxylates with bi- or tridentate ligands. Notably, the cores of
1 and 2 are closely related, and the main difference lies in the
rotation of one of the moieties by ca. 60° (Figure 4).
(Phgly)[Mn3(L1)3(μ3-Cl)(Cl)3]·0.5MeCN·0.5H2O·0.25MeOH
[(R)-3·0.5MeCN·0.5H2O·0.25MeOH]
Complex (R)-3 can be described as a triangular arrangement of
three MnIII ions linked by O-alkoxido donors and one μ3-Cl
bridge. A view of the complex is shown in Figure 3, and the
main bond parameters are summarized in Table S4. The MnIII
cations are octahedrally coordinated with a MnCl2O3N environ-
ment. One L12– Schiff base is coordinated to each manganese
atom through their three donor atoms and acts as an O-alkox-
ido bridge to the neighboring cation. The coordination sphere
is completed by two chlorido ligands in trans positions, one of
which acts as a μ3-Cl bridge. The easy axis corresponding to
the elongated axis of the octahedron follows the Cl–Mn–Cl di-
rection, and the mean O3N planes of the MnIII cations are tilted
by ca. 58° as consequence of this arrangement.
Figure 4. Cores of 1 (left) and 2 (right).
More interesting are the reactions of H2L (H2sae or substi-
tuted derivatives) with MnCl2 (Scheme 3). Tetranuclear rings
with four μ-O and four μ-Cl bridges formed through the reac-
tions of MnCl2 and the Schiff base (1:1) in ethanol followed by
recrystallization in acetonitrile/diethyl ether.[12a,12b]
Scheme 3. Tri-, tetra-, and hexanuclear cores of the clusters reported previ-
ously through the reactions of MnCl2 and H2sae (or substituted derivatives).
All oxygen bridges are provided by the O-alkoxido atoms of the sae2– ligands.
Figure 3. Left: partially labeled plot of (R)-3. The partial resolution of the
structure confirmed the same core for (S)-3. Right: H bonds involving the
trimeric complex and the protonated Phgly+ cation. Color scheme: MnIII dark
green, O red, N navy, Cl violet, C grey, H pink.
In that case, the deprotonation of the ligand was induced by
the trivalent manganese centers because no base was added.
The trinuclear entity is monoanionic, and charge balance is The same core was also formed through the reaction of MnCl2,
provided by one protonated phenylglycinate cation [(R)- or (S)- the Schiff base, and sodium benzoate (in a 4:2:1 ratio) in hot
Phgly+ for (R)-3 or (S)-3, respectively], which is linked to the acetonitrile.[12c] In contrast, a hexanuclear MnII MnIII core was
2
4
three terminal chlorido ligands and the O-alkoxido atoms obtained through the reaction of MnCl2, the Schiff base, and
through six H bonds promoted by the RNH3+ fragment. No rele- triethylamine (in a 1:1:2 ratio) in methanol.[14] In our case, (R)-3
vant intermolecular interactions were found.
and (S)-3 were obtained through the reaction of MnCl2, the
chiral Schiff base, and sodium benzoate (in a 1:1:1 ratio) in
acetonitrile under reflux.
It should be emphasized that complexes 3 are the only ex-
amples of discrete triangular MnIII systems with one μ3-Cl
bridge. This kind of bridge has been reported previously for
The deprotonation of the H2L ligands occurs in neutral and
cubanes with {Mn4O3Cl} cores in which one corner[16] is occu- basic media, and a comparison of the reaction conditions sug-
pied by the chlorido ligand or as a fragment of a larger cluster. gests that it is hard to extract any justification for the different
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
nuclearities. The most probable factor that allows the formation solution for both enantiomers of 2 and 3 are perfect mirror
of triangular systems 3 is the partial breaking of the ligand images of each other (see Supporting Information, Figures S1
during the heating of the reaction mixture under reflux to re- and S2); this confirms the enantiomeric purity and also artefact-
generate the Phgly+ cation, which helps to stabilize the struc- free character of the solid-state spectra. The solid-state ECD
ture. The only conclusion is that these systems are very sensitive spectra again showed some regularities. For (S)-1 and (S)-3,
to small changes in the synthetic procedures, and a rich cluster there is a common sequence of ECD bands: negative in the
chemistry can be advanced by the systematic study of reactions range λ = 650–800 nm, positive at λ ≈ 550 nm, and negative at
with the variation of the strength of the basic medium, the λ ≈ 400 nm [also present for (S)-2]. Owing to the low solubility
H2sae substituents, or the solvent.
of 1, the acetonitrile solution was prepared by heating for sev-
eral minutes, and we noticed that the spectra for the two
enantiomers were not exact mirror images (Figure S3). The sum
of the two ECD spectra should give zero or, for different enan-
tiomeric purity, a spectrum identical to that of one of the two
enantiomers. This is not the case, as shown in Figure S4; there-
fore, some inequivalent decomposition occurs during the heat-
ing process.
Electronic Circular Dichroism Spectra
Electronic circular dichroism (ECD) spectroscopy represents a
crucial tool in structural studies of chiral coordination com-
pounds.[17] Generally, the chirality in such systems may be in-
duced by a chiral ligand, by the intrinsic chirality of metal coor-
dination, or both. The ECD and absorption (UV/Vis) spectra of
the complexes in solution (CH3CN) and in the solid state (KCl
pellets) were measured. The solid-state ECD spectra of both
enantiomers of 1–3 (Figure 5) and those recorded in acetonitrile
All of the absorption spectra of the complexes in acetonitrile
solution are fairly similar to each other (see Figures S1–S3). The
spectra show a very weak band at λ ≈ 470 nm (this band is not
developed for 3), a moderately intense band at λ ≈ 390 nm,
and two intense bands at λ = 270 and λ ≈ 230 nm. Similarly to
the absorption spectra, the ECD spectra display a few common
features, especially in the range λ = 300–800 nm. All of the
complexes with the S ligand configuration, that is, (S)-1, (S)-2,
and (S)-3, show three negative bands: one in the range λ =
500–550 nm, the second at λ ≈ 400 nm, and the third one in
the range λ = 300–350 nm. Furthermore, for (S)-1 and (S)-3, a
weak positive broad band appears in the visible range centered
at λ = 675 and 595 nm, respectively. As expected, this range is
not readily detectable in the absorption spectra (in the concen-
tration range 2.0–2.7 × 10–4
d–d character.
M) as the transitions have mainly
In summary, although the chiral complexes show very differ-
ent structures, the presence of the aforementioned correlations
support the hypothesis that the ECD spectrum is directed
mainly by the chirality of the chelating ligands rather than the
chirality of the manganese coordination sphere by itself. To ob-
tain more evidence, we compared the solution and solid-state
ECD spectra. Such comparisons will give a pronounced view
into the differences between the molecular species in both
states and shed more light on the origin of the transitions in
such complex systems. For clarity, the comparisons were done
for one enantiomer for each pair of compounds.
For (S)-1, the solid-state and solution ECD spectra show sig-
nificant variations (Figure S5); in particular, the signs of two
bands in the visible region at λ ≈ 550 and 675 nm are opposite.
This demonstrates that the solid-state ECD spectra are gov-
erned by intercrystalline interactions and the measured spec-
trum does not reflect the most stable conformer(s) in solution.
Time-dependent DFT (TDDFT) calculations of high-spin coor-
dination compounds are still very demanding and were not
performed because of the complexity of the system as a result
of the complicated relationship between the geometry and the
electronic configuration.[18]
In contrast to the previous case, complexes 2, which are
more symmetric than 1 (as stated above), exhibits rather similar
ECD spectra in both media (Figure 6). The only difference is the
Figure 5. Solid-state ECD spectra for the pairs of enantiomers of 1–3 (red
lines, S enantiomers; blue lines, R enantiomers).
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
relative intensity of the band λ = 300 nm, which suggests that spectively, clearly lower than the expected value of
the dominant conformation in solution is similar to that in the 10.375 cm3 mol–1 K for two MnII and one MnIII isolated cations
solid state. Furthermore, one can notice that the intensity of (g = 2.00). On cooling, the ꢀMT values decreases continuously
the ECD bands (in solution) are the highest among the studied to a plateau at 1.5 cm3 mol–1 K at ca. 6 K. At very low tempera-
set of Mn complexes.
tures, the decrease is more pronounced, probably because of a
mixing of the intermolecular coupling mediated by H bonds,
the D effect induced by the MnIII cations, or the weak interac-
tion between the terminal MnIII ions.
Figure 7. Plots of ꢀMT versus T for (R)-1, (R)-2, (R)-3 (left) and (S)-1, (S)-2,
(S)-3 (right). The solid lines are the fits of the data; see the text for the fit
parameters.
To avoid overparameterization, the fit of the experimental
data was performed in the 300–15 K temperature range with
the conventional isotropic Hamiltonian for a linear arrangement
of three S = 2, 5/2, 2 spins [Equation (1)]:
Figure 6. Comparison between the solid-state and solution ECD spectra for
(R)-2 (top) and (R)-3 (bottom). Note that the solid-state spectrum of (R)-2 is
divided by 1.5 and that of (R)-3 is divided by 4.
H = –2J(S1·S2 + S2·S3)
(1)
It is apparent from Figure 6 that the spectra for (R)-3 in the
solid state and solution are different. This suggests that the
crystal structure is not close to the predominant structure(s) in
acetonitrile solution, as could be expected from the weak H-
bond interactions between the triangular cluster and the phe-
nylglycinate counteraction, which could be broken readily by
interactions with the solvent.
We noticed that the studied complexes in solution and the
solid state are characterized by several well-developed ECD
bands. In the solution spectra, ECD bands are good indicators
of the absolute configuration of the ligand, both for linear (1
and 2) and triangular (3) Mn cores. The geometry of the most
prevalent conformer(s) in solution is in good agreement with
the solid-state structure only for 2. Additionally, significant dif-
ferences between the ECD spectra of 1 and 3 in the two media
suggest a strong overlap of electronic transitions between the
manganese coordination spheres and the arrangement of the
ligands.
The best fit parameters were J = –5.7 cm–1 and g = 2.00 with
R = 1.53 × 10–5 for (R)-1 and J = –6.7 cm–1 and g = 2.01 with
R = 5.17 × 10–6 for (S)-1.
As expected from their bond parameters, complexes (R)-2
and (S)-2 exhibit similar magnetic responses to that of 1, and the
room-temperature ꢀMT values are 8.19 and 8.09 cm3 mol–1 K,
respectively. The fit of the experimental data was performed
with the same Hamiltonian and conditions used for (R)-1 and
(S)-1. The best fit parameters were J = –7.1 cm–1 and g = 2.00
with R = 2.95 × 10–4 for (R)-2 and J = –8.1 cm–1 and g = 2.02
with R = 8.71 × 10–5 for (S)-2. The magnetization under the
maximum field of 5 T reached an unsaturated value close to
2.25 NμB for the four complexes. These values are in agreement
with those reported previously for similar Schiff bases and carb-
oxylate bridges.[11a] The small difference between the values for
each pair of enantiomers must be attributed to the different
solvent molecules in the structures. Complexes (R)-3 and
(S)-3 show room-temperature ꢀMT values of 10.03 and
10.20 cm3 mol–1 K. respectively, higher than the expected value
of 9.00 cm3 mol–1 K for three isolated MnIII cations. As the tem-
Magnetic Properties
The ꢀMT versus T plots for the pairs of enantiomers of 1–3 are perature decreases, the ꢀMT products of both compounds in-
shown in Figure 7. The room-temperature ꢀMT values for the crease to a maximum value of ca. 20 cm3 mol–1 K, which indi-
enantiomers (R)-1 and (S)-1 are 8.62 and 8.41 cm3 mol–1 K re- cates a moderate intramolecular ferromagnetic coupling.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
As the Mn–Cl–Mn and Mn–O–Mn bond angles show minor from the Ea = DS2 relationship is in agreement with the values
differences, the system was modeled as an equilateral triangle. obtained from the magnetization experiments.
A first attempt to fit the experimental data isotropically in the
10–300 K temperature range was made with one J coupling
constant with the Hamiltonian shown in Equation (2):
H = –J(S1·S2 + S1·S3 + S2·S3)
(2)
However, the experimental plots were not reproduced; thus, a
Dion term was also taken into account. An excellent match to
the experimental data was obtained under these conditions
with the best fit parameters J = +1.6 cm–1, Dion = 3.5 cm–1, and
g = 2.03 with R = 9.4 × 10–5 for (R)-3 and J = +1.5 cm–1, Dion
=
3.4 cm–1, and g = 2.03 with R = 1.3 × 10–4 for (S)-3. From these
data, an S = 6 ground state can be proposed.
Magnetization measurements performed at 2 K show a non-
saturated value of 10.6 NμB for (R)-3 and 10.7 electrons for
(R)-3 under the maximum field of 5 T. The fits of the magnetiza-
tions for an isolated S = 6 ground state were satisfactory for
D = –0.42 cm–1 and g = 1.95 for both compounds (Figure S6).
These data confirm the S = 6 state and indicate a significant
ground-state anisotropy.
Figure 9. Left: out-of-phase peaks for (S)-3 under a 0.1 T field at the indicated
frequencies. Right: Arrhenius plot.
Among other factors, a parallel alignment of the easy axis of
the MnIII cations contributes to the enhanced global anisotropy
A more precise determination of the D parameter was per- of the system and the SMM response. Low-temperature studies
formed through reduced magnetization experiments (Figure 8). (0.04–2 K) of ferromagnetic tetranuclear [Mn4(sae)4Cl4] rings,
These measurements nicely confirmed the anisotropic character which show a quasiperpendicular easy axes, revealed a weak
of the ground state, and excellent fits of the experimental data anisotropy and energy barriers of ca. 5 cm–1. The triangular
were obtained for D = –0.39 cm–1 and g = 1.93 for (R)-3 and systems derived from Schiff bases with one μ3-O ligand dis-
D = –0.37 cm–1 and g = 1.92 for (S)-3.
placed from the Mn3 plane are scarce,[19] and an S = 6 ground
state has been reported for only one of them.[19b] As in the
above case, the easy axes of the MnIII cations are roughly per-
pendicular, the ground state was weakly anisotropic, and no
SMM response was observed. In contrast, the easy axes for
(R)-3 and (S)-3 are directed towards the μ3-Cl ligand and form
a mean angle close to 65°, which allows a moderate total
anisotropy and an appreciable Ea barrier.
Conclusions
The employment of enantiomerically pure Schiff bases is an
excellent method to produce chiral clusters and chiral SMMs.
The reported systems are the first polynuclear derivatives of the
employed ligands, and a discrete triangular arrangement of
MnIII cations linked by one μ3-Cl bridge was isolated for the
first time and gives unambiguous evidence of its ferromagnetic
response. Complexes 3 exhibit the largest energy barriers for
this family of clusters. Electronic circular dichroism is a funda-
mental tool for the full characterization of these kind of com-
pounds, and a comparison between the solid and solution
spectra provided information about the stabilities of the com-
plexes in different media.
Figure 8. Reduced magnetization plots for (R)-3 (left) and (S)-3 (right). The
solid lines show the best fits of the data.
In light of the above results, the SMM properties were ex-
plored. Alternating-current (ac) measurements at zero field only
showed the tails of the out-of-phase signals, but measurements
performed under a transverse magnetic field of 0.1 T broke the
tunneling of the magnetization and allowed the observation of
well-defined ac peaks above 2 K. The ac peaks for (R)-3 and
(S)-3 were compared at the arbitrary frequency of 1000 Hz. They
were exactly identical (Figure S7) and, thus, the complete set of
measurements was performed for only one of the enantiomers
[(S)-3, Figure 9]. The Arrhenius fit of the positions of the peak
maxima gives a barrier for the reversal of the magnetization of
Experimental Section
Materials and Methods: The IR spectra (ν = 4000–400 cm–1) were
˜
recorded with a Bruker IFS-125 FTIR spectrometer with samples pre-
Ea = 17.1 cm–1 and τo = 9.3 × 10–9. The D value of 0.48 cm–1 pared as KBr pellets. Variable-temperature magnetic studies were
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
6
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
performed with a Quantum Design MPMS-5 magnetometer operat-
ing at 0.03 T in the 300–2.0 K range. Diamagnetic corrections were
Na(PhCOO) and Mn(NO3)2. Mn(PhCOO)2 precipitated immediately
as a white powder, which was washed with cold water to remove
applied to the observed paramagnetic susceptibilities by using Pas- soluble ions and air-dried. (R)- and (S)-2-phenylglycinol and (R) and
cal's constants. The analysis of the magnetic data was performed (S)-phenylalaninol were purchased from TCI chemicals and used
with the PHI program.[20] The qualities of the fits were parametrized
without further purification. Each pair of enantiomers was prepared
by the same procedure; thus common syntheses will be described.
The yields for 1–3 were ca. 25 %, and well-formed crystals were
obtained and employed for instrumental measurements. The sam-
ples for analysis were dried gently to remove volatile solvents.
by the value of R = (ꢀMTexp – ꢀMTcalcd.)2/(ꢀMTexp)2.
The ECD and UV/Vis spectra were recorded with a Jasco J-715
spectrometer at room temperature with samples in spectroscopy-
grade acetonitrile. Solutions in the concentration range
2.0–2.7 × 10–4 mol dm3 (i.e., 1.33–1.50 mg of sample per 5 mL of
acetonitrile) were measured in three quartz cells with pathlengths
of 2 cm (λ = 850–450 nm), 1 cm (λ = 450–315 nm), and 0.1 cm (λ =
315–200 nm). In each case, a tiny amount of the sample would not
dissolve. All spectra were recorded at a scanning speed of 100 nm
min–1, a step size of 0.1 nm, a bandwidth of 2 nm, a response time
of 0.5 s, and averaged over four accumulations. The baselines of
the spectra were corrected with the solvent (acetonitrile) spectrum
recorded under the same conditions immediately before or after
the sample measurement. The ECD spectra were normalized to the
UV/Vis spectra. The solid-state ECD spectra were obtained by plac-
ing a pellet in a rotating holder as close as possible to the photo-
multiplier tube of a Jasco J-715 spectrometer. Freshly prepared pel-
lets with KCl as a matrix were measured in the range λ = 240–
800 nm. In all cases, it was not possible to obtain a good quality
spectrum below λ = 270 nm. The following measurement parame-
ters were applied: scanning speed 100 nm min–1, step size 0.1 nm,
bandwidth 2 nm, response time 0.5 s, and four scans. The back-
grounds of the resulting spectra were corrected. In each case, sev-
eral spectra were obtained for one pellet, for which the disk was
rotated around the incident axis direction and then flipped. The
spectra were very similar, and no major differences were observed
on the variation of rotation angle; therefore, the absence of spectral
artefacts from linear dichroism and birefringence was confirmed.
H2L1 and H2L2 Schiff Bases: Equimolar amounts of (R)- or (S)-2-
phenylglycinol (for H2L1) or phenylalaninol (for H2L2) and salicyl-
aldehyde were mixed in ethanol, and the mixture was heated under
reflux for 1 h. The ligands were collected as yellowish solids in high
yields after the concentration of the mother solutions.
[Mn3(L1)2(PhCOO)4(MeOH)]·MeOH [(R)-1·MeOH and (S)-1·
MeOH]: Mn(PhCOO)2 (0.5 mmol, 0.167 g) and (R)- or (S)-H2L1
(0.5 mmol, 0.121 g) were mixed in MeOH/MeCN (1:1), and the mix-
ture was stirred for 1 h at room temperature. The resulting dark
brown solution was filtered and left to diffuse slowly with Et2O.
Dark brown crystals suitable for XRD appeared after 24 h.
C58H46Mn3N2O12: calcd. C 61.77, H 4.11, N 2.48; found [(R)-1/(S)-1]
C 60.9/61.3, H 4.2/4.0, N 2.3/2.3. IR: ν = 3423.67 (br), 2925.21 (br),
˜
1598.48 (s), 1565.85 (s), 1538.66 (m), 1445.84 (w), 1395.35 (s),
1338.50 (m), 1287.31 (m), 1201.53 (w), 1147.38 (w), 1023.48 (m),
865.77 (s), 724.00 (m), 704.36 (w), 675.43 (w), 630.21 (w), 584.37 (w),
549.39 (m), 452.23 (w) cm–1
.
[Mn3(L2)2(PhCOO)4]·Solvent [(R)-2·5H2O, (S)-2·2MeCN·MeOH·
1.5H2O]: Mn(PhCOO)2 (0.5 mmol, 0.167 g) and (R)- or (S)-H2L2
(0.5 mmol) were mixed in MeOH/MeCN (1:1), and the mixture was
stirred for 1 h at room temperature. The resulting solution was fil-
tered and left to evaporate slowly. Dark brown crystals appeared
after 24 h. C61H54Mn3N2O13: calcd. C 61.68, H 4.58, N 2.36; found
[(R)-2/(S)-2] C 62.0/61.5, H 4.3/4.4, N 2.3/2.5. IR: ν = 3434.23 (br),
˜
Single-Crystal X-ray Crystallography: Prismlike specimens of the
R and S enantiomers of 1–3 were used for the X-ray crystallographic
analysis. The X-ray intensity data were measured with a Bruker D8-
Venture system equipped with a multilayer monochromator and a
Mo microfocus source (λ = 0.71073 Å). The frames were integrated
with the Bruker SAINT software package with a narrow-frame algo-
rithm. The final cell constants were based upon the refinement of
the xyz centroids of reflections above 20σ(I). The data were cor-
rected for absorption effects by the multiscan method (SADABS).
The structures were solved with the Bruker SHELXTL software pack-
age and refined with SHELXL.[21] Details of the crystal data, collec-
tion, and refinement for the pairs of enantiomers 1–3 are summa-
rized in Table S1. The analyses of the structures and the preparation
of the plots for publication were performed with the Ortep3[22] and
POVRAY programs.
3059.21 (w), 2930.64 (w), 1598.78 (s), 1569.96 (s), 1540.78 (s),
1491.65 (w), 1469.41 (w), 1446.43 (s), 1374.50 (s), 1299.23 (m),
1149.83 (w), 1051.26 (w), 1023.34 (w), 764.13 (w), 754.62 (w), 721.38
(m), 672.72 (w), 577.72 (m), 453.71 (w) cm–1
.
(Phgly)[Mn3(L1)3(μ3-Cl)(Cl)3]·Solvent [(R)-3·0.5MeCN·0.25MeOH·
0.5H2O, (S)-3·0.5MeCN·0.5H2O]: MnCl2·4H2O (0.5 mmol, 0.197 g),
(R)- or (S)-H2L1 (0.5 mmol, 0.121 g), and Na(PhCOO) (0.5 mmol,
0.036 g) were dissolved in MeCN (20 mL), and the mixture was
heated under reflux for 30 min. The solution was filtered and left
to evaporate slowly. Dark brown crystals suitable for XRD appeared
after two weeks.
The same product was obtained from an attempt to introduce an
azide ion into the cluster with Na(PhCOO) replaced with NaN3
(0.5 mmol, 0.033 g). C53H51Cl4Mn3N4O7: calcd. C 54.75, H 4.42, N
4.81; found [(R)-3/(S)-3] C 55.1/55.4, H 4.1/4.5, N 4.9/4.6. IR: ν =
3446.7 (br), 2926.72 (br), 1607.85 (s), 1541093 (m), 1492.22(w),
The quality of the structure for (S)-3 was below the quality standard
for publication; thus, the complete structural data are not included
in the work but unambiguous characterization was provided by the
cell parameters and space group (Table S1), which agree with those
obtained for (R)-3, and the partial data of the cluster confirms an
identical structure.
˜
1441.13 (m), 1384.07 (w), 1293.49 (m), 1149.25 (w), 981.56 (w),
757.49 (m), 703.00 (m), 643.11 (w), 591.05 (m), 553.57 (s) cm–1
.
Supporting Information (see footnote on the first page of this
article): Structural and ECD data.
CCDC 1492960 [for (R)-1], 1492961 [for (S)-1], 1492962 [for (R)-2],
1492963 [for (S)-2], 1492964 [for (R)-3], and 1492965 [for (S)-3] con-
tain the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre.
Acknowledgments
Funding from the Ministerio de Economía y Competitividad,
Project CTQ2015-63614-P, is acknowledged. M. G. thanks the
Polish Ministry of Science and Higher Education (“Mobilnosc
Synthetic Procedures: Mn(PhCOO)2 was synthesized in high yield
(>80 %) by mixing stoichiometric amounts of aqueous solutions of Plus” grant no. 1286/MOB/IV/2015/0) for support.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
7
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[11] a) Y.-G. Li, L. Lecren, W. Wernsdorfer, R. Clerac, Inorg. Chem. Commun.
2004, 7, 1281–1284; b) H.-C. Yao, M.-M. Li, L.-M. Zheng, Z.-J. Li, J. Coord.
Chem. 2008, 61, 2814–2822.
Keywords: Manganese · Schiff bases · Single-molecule
magnets · Chirality · Cluster compounds
[12] a) C. Boskovic, E. Rusanov, H. Stoeckli-Evans, H. U. Gudel, Inorg. Chem.
Commun. 2002, 5, 881–886; b) C. Boskovic, R. Bircher, P. L. W. Tregenna-
Piggott, H. U. Gudel, C. Paulsen, W. Wernsdorfer, A.-L. Barra, E. Khatsko,
A. Neels, H. Stoeckli-Evans, J. Am. Chem. Soc. 2003, 125, 14046–14058; c)
S. Nayak, G. Novitchi, S. Muche, D. Luneau, S. Dehnen, Z. Anorg. Allg.
Chem. 2012, 638, 1127–1133.
[13] Y.-G. Li, Q. Wu, L. Lecren, R. Clerac, J. Mol. Struct. 2008, 890, 339–345.
[14] N. Hoshino, T. Ito, M. Nihei, H. Oshio, Inorg. Chem. Commun. 2003, 6,
377–380.
[1] a) D. N. Woodruff, R. E. P. Winpenny, R. A. Layfield, Chem. Rev. 2013, 113,
5110–5148; b) J. D. Rinehart, J. R. Long, Chem. Sci. 2011, 2, 2078–2085;
c) L. Ungur, S.-Y. Lin, J. Tang, L. F. Chibotaru, Chem. Soc. Rev. 2014, 43,
6894–6905.
[2] a) K. Griffiths, C. W. D. Gallop, A. Abdul-Sada, A. Vargas, O. Navarro, G. E.
Kostakis, Chem. Eur. J. 2015, 21, 6358–6361; b) G. Maayan, G. Christou,
Inorg. Chem. 2011, 50, 7015–7021; c) N. C. Anastasiadis, D. A. Kalofolias,
A. Philippidis, S. Tzani, C. P. Raptopoulou, V. Psycharis, C. J. Milios, A.
Escuer, S. P. Perlepes, Dalton Trans. 2015, 44, 10200–10209.
[3] J.-C. G. Bünzli, S. V. Eliseeva, Chem. Sci. 2013, 4, 1939–1949.
[4] J.-C. G. Bünzli, S. V. Eliseeva, Chem. Soc. Rev. 2010, 39, 189–227.
[5] a) S. Muhlbauer, B. Binz, F. Jonietz, C. Pfleiderer, A. Rosch, A. Neubauer,
R. Georgii, P. Boni, Science 2009, 323, 915–919; b) X. Z. Yu, Y. Onose, N.
Kanazawa, J. H. Park, J. H. Han, N. Nagaosa, Y. Tokura, Nature 2010, 465,
901–904.
[15] a) Q. Wu, Q. Shi, Y.-G. Li, E.-B. Wang, J. Coord. Chem. 2008, 61, 3080–3091;
b) Y. Ding, J. Xu, Z. Pan, H. Zhou, X. Lou, Inorg. Chem. Commun. 2012,
22, 40–43.
[16] a) Q. Li, J. B. Vincent, E. Libby, H.-R. Chang, J. C. Huffman, P. D. W. Boyd,
G. Christou, D. N. Hendrickson, Angew. Chem. Int. Ed. Engl. 1988, 27,
1731–1733; Angew. Chem. 1988, 100, 1799; b) S. Wang, K. Folting, W. E.
Streib, E. A. Schmitt, J. K. McCusker, D. N. Hendrickson, G. Christou,
Angew. Chem. Int. Ed. Engl. 1991, 30, 305–306; Angew. Chem. 1991, 103,
314; c) D. N. Hendrickson, G. Christou, E. A. Schmitt, E. Libby, J. S. Bashkin,
S. Wang, H.-L. Tsai, J. B. Vincent, P. D. W. Boyd, J. C. Huffman, K. Folting,
Q. Li, W. E. Streib, J. Am. Chem. Soc. 1992, 114, 2455–2471; d) M. W.
Wemple, H.-L. Tsai, K. Folting, D. N. Hendrickson, G. Christou, Inorg. Chem.
1993, 32, 2025–2031; e) S. Wang, H.-L. Tsai, E. Libby, K. Folting, W. E.
Streib, D. N. Hendrickson, G. Christou, Inorg. Chem. 1996, 35, 7578–7589.
[17] T. Wu, X.-Z. You, P. Bouř, Coord. Chem. Rev. 2015, 284, 1–18.
[18] a) A. Ipatov, F. Cordova, L. J. Doriol, M. E. Casida, J. Mol. Struct. 2009,
914, 60–73; b) M. Enamullah, M. A. Quddus, M. R. Hasan, G. Pescitelli, R.
Berardozzi, G. Makhloufi, V. Vasylyeva, Ch. Janiak, Dalton Trans. 2016, 45,
667–680.
[6] a) J. Kishine, K. Inoue, Y. Yoshida, Prog. Theor. Phys. Suppl. 2005, 159, 82–
95; b) Y. Togawa, T. Koyama, K. Takayanagi, S. Mori, Y. Kousaka, J. Akim-
itsu, S. Nishihara, K. Inoue, A. S. Ovchinnikov, J. Kishine, Phys. Rev. Lett.
2012, 108, 107202.
[7] a) H. Asada, M. Ozeki, M. Fujiwara, T. Matsushita, Chem. Lett. 1999, 28,
525–526; b) H. Asada, M. Ozeki, M. Fujiwara, T. Matsushita, Polyhedron
2002, 21, 1139–1148; c) M. Dey, C. P. Rao, P. K. Saarenketo, K. Rissanen,
E. Kolehmainen, P. Guionneau, Polyhedron 2003, 22, 3515–3521; d) C. P.
Pradeep, T. Htwe, P. S. Zacharias, S. K. Das, New J. Chem. 2004, 28, 735–
739; e) C. P. Pradeep, P. S. Zacharias, S. K. Das, Polyhedron 2005, 24, 1410–
1416; C. P. Pradeep, P. S. Zacharias, S. K. Das, J. Chem. Sci. 2006, 118, 311–
317; f) S.-Z. Zhan, Y.-J. Zhang, J.-G. Wang, Acta Crystallogr., Sect. E: Struct.
Rep. Online 2006, 62, m2713–m2714; g) C. P. Pradeep, P. S. Zacharias,
S. K. Das, Inorg. Chem. Commun. 2008, 11, 89–93.
[19] a) M. Nihei, N. Hoshino, T. Ito, H. Oshio, Chem. Lett. 2002, 31, 1016–1017;
b) P. Chaudhuri, R. Wagner, T. Weyhermuller, Eur. J. Inorg. Chem. 2010,
1339–1342; c) C.-M. Liu, D.-Q. Zhang, D.-B. Zhu, Dalton Trans. 2010, 39,
1781–1785.
[20] N. F. Chilton, R. P. Anderson, L. D. Turner, A. Soncini, K. S. Murray, J.
Comput. Chem. 2013, 34, 1164–1175.
[8] R. Inglis, F. White, S. Piligkos, W. Wernsdorfer, E. K. Brechin, G. S. Papaef-
stathiou, Chem. Commun. 2011, 47, 3090–3092.
[21] G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr. 2008, 64,
112–122.
[22] Ortep-3 for Windows: L. J. Farrugia, J. Appl. Crystallogr. 1997, 30, 565.
[9] a) Y.-Y. Zhu, C. Cui, Y.-Q. Zhang, J.-H. Jia, X. Guo, C. Gao, K. Qian, S.-D.
Jiang, B.-W. Wang, Z.-M. Wang, S. Gao, Chem. Sci. 2013, 4, 1802–1806; b)
Y.-Y. Zhu, C. Cui, K. Qian, J. Yin, B.-W. Wang, Z.-M. Wang, S. Gao, Dalton
Trans. 2014, 43, 11897–11907.
[10] L.-F. Zhang, Z.-H. Ni, Z.-M. Zong, X.-Y. Wei, C.-H. Ge, H.-Z. Kou, Acta Crystal-
logr., Sect. C: Cryst. Struct. Commun. 2005, 61, m542–m544.
Received: September 15, 2016
Published Online: ■
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
8
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Chiral Magnets
The employment of enantiomerically
pure Schiff bases allows the synthesis
of the R and S pairs of trinuclear chiral
A. Escuer,* J. Mayans,
M. Font-Bardia, L. Di Bari,
M. Górecki ............................................ 1–9
clusters. Two discrete triangular MnIII
3
enantiomers with one μ3-Cl bridge ex-
hibit single-molecule magnet (SMM)
responses.
Trinuclear Complexes Derived from
R/S Schiff Bases – Chiral Single-Mol-
ecule Magnets
DOI: 10.1002/ejic.201601138
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
9
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim