A new family of 1D exchange biased heterometal single-molecule magnets: Observation of pronounced quantum tunneling steps in the hysteresis loops of quasi-linear {Mn2Ni3} clusters

Article abstract of DOI:10.1021/ja108740c

First members of a new family of heterometallic Mn/Ni complexes [Mn ₂Ni₃X₂L₄(LH)₂(H ₂O)₂] (X = Cl: 1; X = Br: 2) with the new ligand 2-{3-(2-hydroxyphenyl)-1H-pyrazol-1-yl}ethanol (H₂L) have been synthesized, and single crystals obtained from CH₂Cl₂ solutions have been characterized crystallographically. The molecular structures feature a quasi-linear Mn^III-Ni^II-Ni^II-Ni ^II-Mn^III core with six-coordinate metal ions, where elongated axes of all the distorted octahedral coordination polyhedra are aligned parallel and are fixed with respect to each other by intramolecular hydrogen bonds. 1 and 2 exhibit quite strong ferromagnetic exchange interactions throughout (J_Mn-Ni ≈ 40 K (1) or 42 K (2); J_Ni-Ni ≈ 22 K (1) or 18 K (2)) that lead to an S_tot = 7 ground state, and a sizable uniaxial magnetoanisotropy with D_mol values -0.55 K (1) and -0.45 K (2). These values are directly derived also from frequency- and temperature-dependent high-field EPR spectra. Slow relaxation of the magnetization at low temperatures and single-molecule magnet (SMM) behavior are evident from frequency-dependent peaks in the out-of-phase ac susceptibilities and magnetization versus dc field measurements, with significant energy barriers to spin reversal U_eff = 27 K (1) and 22 K (2). Pronounced quantum tunnelling steps are observed in the hysteresis loops of the temperature- and scan rate-dependent magnetization data, but with the first relaxation step shifted above (1) or below (2) the zero crossing of the magnetic field, despite the very similar molecular structures. The different behavior of 1 and 2 is interpreted in terms of antiferromagnetic (1) or ferromagnetic (2) intermolecular interactions, which are discussed in view of the subtle differences of intermolecular contacts within the crystal lattice.

Full text of DOI:10.1021/ja108740c

ARTICLE
pubs.acs.org/JACS
A New Family of 1D Exchange Biased Heterometal Single-Molecule
Magnets: Observation of Pronounced Quantum Tunneling Steps
in the Hysteresis Loops of Quasi-Linear {Mn2Ni3} Clusters
Animesh Das,† Klaus Gieb,‡ Yulia Krupskaya,§ Serhiy Demeshko,† Sebastian Dechert,† R€udiger Klingeler,§
Vladislav Kataev,*,§ Bernd B€uchner,§ Paul M€uller,*,‡ and Franc Meyer*,†
†Institut f€ur Inorganic Chemistry, Georg-August-University G€ottingen, Tammannstrasse 4, D-37077 G€ottingen, Germany
‡Department of Physics, Universit€at Erlangen-N€urnberg, Erwin-Rommel-Strasse 1, D-91058 Erlangen, Germany
§Leibniz-Institute for Solid State and Materials Research IFW Dresden, P.O. Box 270116, D-01171 Dresden, Germany
S Supporting Information
b
ABSTRACT: First members of a new family of heterometallic
Mn/Ni complexes [Mn2Ni3X2L4(LH)2(H2O)2] (X = Cl: 1;
X = Br: 2) with the new ligand 2-{3-(2-hydroxyphenyl)-1H-
pyrazol-1-yl}ethanol (H2L) have been synthesized, and single
crystals obtained from CH2Cl2 solutions have been characterized
crystallographically. The molecular structures feature a quasi-
linear MnIII-NiII-NiII-NiII-MnIII core with six-coordinate
metal ions, where elongated axes of all the distorted octahedral
coordination polyhedra are aligned parallel and are fixed with
respect to each other by intramolecular hydrogen bonds. 1 and 2 exhibit quite strong ferromagnetic exchange interactions throughout
(JMn-Ni ≈ 40 K (1) or 42 K (2); JNi-Ni ≈ 22 K (1) or 18 K (2)) that lead to an Stot = 7 ground state, and a sizable uniaxial
magnetoanisotropy with Dmol values -0.55 K (1) and -0.45 K (2). These values are directly derived also from frequency- and
temperature-dependent high-field EPR spectra. Slow relaxation of the magnetization at low temperatures and single-molecule magnet
(SMM) behavior are evident from frequency-dependent peaks in the out-of-phase ac susceptibilities and magnetization versus dc field
measurements, with significant energy barriers to spin reversal Ueff = 27 K (1) and 22 K (2). Pronounced quantum tunnelling steps are
observed in the hysteresis loops of the temperature- and scan rate-dependent magnetization data, but with the first relaxation step shifted
above (1) orbelow(2) the zero crossing of the magnetic field, despite the very similar molecular structures. The different behavior of 1and
2 is interpreted in terms of antiferromagnetic (1) or ferromagnetic (2) intermolecular interactions, which are discussed in view of the
subtle differences of intermolecular contacts within the crystal lattice.
’ INTRODUCTION
devoted to the synthesis and characterization of new polynuclear
clusters that display slow magnetic relaxation below a certain blocking
temperature (TB). There are now many examples of homometal
SMMs, mostly containing MnIII,5 butalsoothermetalionssuchasV,6
Fe,7 Co,8 or Ni.9 Still, the Mn12 family represented the SMMs with the
highest TB (∼3.5 K) and anisotropy barriers Ueff (up to 74 K) until
2007, when a Mn6 complex with enhanced blocking temperature
(∼4.5 K) and an effective energy barrier to magnetization reversal
of 86.4 K was found, the record holder to date.10,11 Among the
essential requirements for SMM behavior are a high-spin ground state
Stot and a strong easy axis-type magnetic anisotropy (negative zero-
field splitting parameter Dmol), which define the energy barrier Ueff ≈
Certain high-spin molecules with an easy axis (Ising-type)
magnetic anisotropy show bistability due to slow magnetic
relaxation of the spin reorientation along the magnetic anisotro-
py axis. These molecules, called single molecule magnets
(SMM), have a double minimum potential with a significant
barrier for the reversal of the magnetic moment, which is why at
very low temperature the spin flips via quantum processes only.
This results in a hysteresis of the magnetization that is of purely
molecular origin.1 SMMs are attracting particular interest in
chemistry, physics, and materials sciences because they show
novel phenomena at the classical/quantum interface, and be-
cause they are considered as promising nanoscale objects for,
inter alia, future high-density memory storage devices and
quantum computing technologies.2
|Dmol|Stot .
2 12 While impressive progress has been made in terms of
increasing Stot, synthetic strategies for manipulating D are a much less
developed.13,14 More recently, it has even been proposed that the
anisotropy barrier does not increase with Stot as Stot2, but the
Since the discovery in 1993 of the first SMM, the dodecanuclear
mixed-valent manganese complex [Mn12O12(O2CMe)16(OH2)4]
with an Stot = 10 ground state,3 various derivatives of this so-called
Mn12 archetype have been investigated,4 and much effort has been
Received: September 28, 2010
Published: February 18, 2011
r
2011 American Chemical Society
3433
dx.doi.org/10.1021/ja108740c J. Am. Chem. Soc. 2011, 133, 3433–3443
|

Journal of the American Chemical Society
ARTICLE
dependence is on the order of unity.15 It has also been questioned
whether large magnetic anisotropy and high spin can coexist at all.16
Given the intrinsic relationship between the Stot and Dmol parameters,
it was concluded from the latter study that perfect alignment of the
Jahn-Teller axes with the largest possible number of paramagnetic
centers should be most advantageous for obtaining systems large
energy barriers, though this poses a particular challenge for controlling
the ferromagnetic exchange. New types of SMMs are thus needed to
extend and substantiate the experimental basis in these directions.
Of particular value are families of related complexes with
relatively minor, yet controllable perturbations to the structures,
since they allow probing the subtle effects of structural or electronic
variations on SMM characteristics and, in the end, lead to strategies
for the deliberate tuning of SMM properties.17 Most of the earlier
studies had focused on individual high-nuclearity molecules that
should, in the ideal case, be well isolated in the crystal lattice for
investigating pristine SMM behavior. However, the absence of
intermolecular interactions is not a prerequisite for observing a
sufficiently large energy barrier to magnetization reversal. In fact, in
more recent years various systems that feature intermolecular
contacts, either pairwise or in extended arrangements in the crystal
lattice, were found to exhibit slow relaxation of the magnetization
because of an energy barrier of molecular origin.18-20 As a new
feature it was observed for these systems that the quantum
tunnelling of magnetization (QTM) has shifted away from zero
field, and hence, they were called “exchange-biased SMMs”. Such
novel SMMs offer promising prospects for studying the effects of
intermolecular interactions on SMM, and it has even been suggested
that intermolecular interactions may provide a means of fine-tuning
the quantum tunnelling of magnetization in SMMs.21,22 Modulation
of the intermolecular interactions certainly is a challenging task and
requires proper adjustment of ligand functionalities in the periphery
of the molecular complexes, which emphasizes the particular value
of families of robust SMMs with closely related oligometallic core
but variable terminal ligand decoration.
Scheme 1. Synthesis of H2L
’ RESULTS AND DISCUSSION
Syntheses and Structural Characterizations. The new li-
gand 2-{3-(2-hydroxyphenyl)-1H-pyrazol-1-yl}ethanol (H2L) has
been synthesized in two steps starting from 2-hydroxyacetophe-
none (Scheme 1). The intermediate 1-(2-hydroxyphenyl)-3-di-
methylaminoprop-2-enone was obtained according to a literature
method34 and was then heated to reflux with excess β-hydroxyethyl
hydrazine in methanol solution for 2 h. Extraction of the crude
product with light petroleum using a Soxhlet apparatus gave H2L as
needle-shaped white crystals in 40-50% yield. Several design
considerations make H2L an attractive scaffold for assembling
oligonuclear complexes: (i) H2L may act as a tridentate ligand with
sufficient flexibility in one arm (the hydroxyethyl subunit); (ii) it
provides, after deprotonation, two distinct potential bridging units
(phenoxy and alkoxy); and (iii) it is easy to synthesize in reasonable
quantities. The molecular structure of H2L has been determined
(see Figure S1 in the Supporting Information). It forms a propeller-
like hydrogen-bonded tetramer via the aliphatic OH groups, while
the phenolic OH groups are involved in intramolecular hydrogen-
bonding to the pyrazole-N.
Reaction of H2L with MnX2 xH2O (X = Cl, Br) and
3
NiCl2 6H2O in the presence of NEt3 in acetonitrile at room
3
temperature gave complexes [MnIII2NiII X2L4(LH)2(H2O)2]
3
(X = Cl: 1; X = Br: 2). Both compounds could be crystallized
from dichloromethane solutions as 1 8CH2Cl2 or 2 8CH2Cl2,
The combination of different metal ions in a single cluster for
achieving large magnetic anisotropy and high-spin ground states
significantly extends the structural landscape for SMMs but has
been explored relatively little until now. Mixed 3d/3d transition
metal SMMs include MnCu,23 Mn2Ni2,24,25 Mn11Cr, Fe4Ni4,26
Mn6Cr,27 and Mn3Ni,28 some of them showing hysteresis loops
with QTM. While targeted synthetic approaches to the con-
struction of SMMs are highly sought after,27,29 the most common
strategy to date for obtaining oligonuclear complexes with SMM
behavior is the (more or less serendipity-driven) self-assembly of
paramagnetic metal ions and flexible bridging ligands.30 Diol-
type ligands have been particularly successful in this regard,31
though the variety of ligands in use has remained rather limited.
Phenol-pyrazole hybrid ligands have recently emerged as a
promising ligand class for high-nuclearity MnIII compounds.32
On the basis of our experience with pyrazole-derived ligand
scaffolds,33 we have now synthesized a new tridentate diol ligand
2-[3-(2-hydroxyphenyl)-1H-pyrazol-1-yl]ethanol (H2L) with a
central pyrazole unit, which gives rise to a new family of
heterometal SMMs. These relatively small pentanuclear systems
3
3
respectively. In both cases the complexes crystallize in the
triclinic space group P1 as quasi-linear centrosymmetric Mn-
Ni-Ni-Ni-Mn molecules, with the central nickel atom situ-
ated on a center of inversion. Overall molecular topologies
determined by X-ray crystallography are very similar, and com-
plex 2 is shown as a representative example in Figure 1 (the
molecular structure of 1 is shown in Figure S2). Relevant atom
distances and bond angles are collected in Table 1.
Both complexes contain a [Mn2Ni3(μ-O)8] core where the
coordination number of all metal atoms is six and the coordina-
tion environment is more or less distorted octahedral. Com-
plexes 1 and 2 differ only by the terminal ligands bound to Mn,
which are either Cl in 1 or Br in 2. Oxidation states of the metal
ions, MnIII and NiII, are confirmed by bond valence sum
calculations35 and by the relatively short bonds for the MnIII
ions (see Table 1). The string of metal ions can be best visualized
as five edge-sharing octahedra (Figure 2, bottom). The three NiII
ions are doubly bridged by phenoxy-O atoms, while both
peripheral MnIII ions are linked to the NiII core via double
3
alkoxy-O bridging. MnIII ions are found in {NO4X} environment
with equatorial position occupied by the {ONO} donor site of a
tridentate ligand [L]2- and an alkoxo-O of another ligand, and
with the terminal halide (Cl or Br) and a water molecule in the
axial positions. As expected, the MnIII ions exhibit a strong Jahn-
Teller elongation along the Cl/Br-Mn-OH2 axis (Figure 2,
II
[Mn2IIINi3 L4(LH)2X2(H2O)2] feature a ferromagnetic Stot = 7
ground state, yet sizable anisotropy, and pronounced QTM steps
in the hysteresis loops. Here we report and discuss details of the
synthesis as well as structural, spectroscopic, and magnetic
characterization of two first examples of this new family of
SMMs, which are amenable to valuable structural modifications
for tuning the SMM behavior.
top) with Mn-O/Cl distances of 2.34/2.61 Å in 1 8CH2Cl2 and
3
3434
dx.doi.org/10.1021/ja108740c |J. Am. Chem. Soc. 2011, 133, 3433–3443

Journal of the American Chemical Society
ARTICLE
Figure 1. Molecular structure of complex 2 in 2 8CH2Cl2 (thermal
3
ellipsoids drawn at the 30% probability level). Most hydrogen atoms and
solvent molecules are omitted for clarity. Symmetry transformation used
to generate equivalent atoms: (0) 1-x, 1-y, 1-z.
Table 1. Selected Distances [Å] and Angles [deg] for
1 8CH2Cl2 and 2 8CH2Cl2.a,b
Figure 2. (Top) Coordination environment of the metal atoms in 1.
The arrows emphasize the Jahn-Teller axes. Symmetry transformation
used to generate equivalent atoms: (0) 1-x, 1-y, 1-z. (Bottom)
Emphasis of the five edge-sharing octahedral of the MnIII-NiII-NiII-
NiII-MnIII core.
3
3
1 8CH2Cl2
2 8CH2Cl2
3
3
Ni1-N1
Ni1-O1
Ni1-O3
Ni2-N3
Ni2-O1
Ni2-O20
Ni2-O3
Ni2-O4
Ni2-O6
Mn1-N5
Mn1-O20
Mn1-O5
Mn1-O6
Mn1-O7
Mn1-X1
2.173(7)
2.030(6)
2.120(6)
2.022(7)
2.014(6)
2.062(6)
2.061(6)
2.121(6)
2.043(6)
1.977(7)
1.889(6)
1.872(6)
1.888(6)
2.344(6)
2.609(3)
3.1089(11)
3.0447(17)
3.17(1)
2.77(1)
180
2.186(4)
2.026(3)
2.116(3)
2.030(4)
2.012(3)
2.065(3)
2.056(3)
2.127(3)
2.047(3)
1.994(4)
1.891(3)
1.866(4)
1.886(3)
2.337(4)
2.7775(9)
3.1106(6)
3.0457(10)
3.238(4)
2.757(5)
180
{NO5} coordination sphere. Interestingly, two of the six pyr-
azole-based ligands in 1 and 2 are only deprotonated once. The
remaining neutral alcohol function of these partially deproto-
nated [HL]- serves as a terminal ligand to the NiII core and
3
exhibits a strong intramolecular H-bond to the neighboring Mn-
bound halide ion [d(O4 Cl1) = 3.17(1) Å in 1 8CH2Cl2;
3 3 3
3
d(O4 Br1) = 3.238(4) Å in 2 8CH2Cl2];36 the Ni-O4 bond
3 3 3
3
is elongated compared to all other Ni2-O and Ni2-N bonds
(2.10-2.13 Å versus 2.01-2.07 Å). Another important intra-
molecular H-bond involves the MnIII-bound water and the
proximate phenolato-O3 atom [d(O7 O3) ≈ 2.76 Å]. These
3 3 3
two H-bonding interactions displayed by the axial ligands at Mn
fix the Cl/Br-Mn-OH2 axis relative to the O3-Ni2-O4 axis;
i.e., the Mn1 and Ni2 coordination octahedra have their elon-
gated axes aligned parallel. With respect to the magnetic proper-
ties it is also important to note that the Jahn-Teller axes of the
MnIII ions at both ends of the MnIII-NiII-NiII-NiII-MnIII
core are roughly parallel to each other.
Ni1 Ni2
3 3 3
Ni2 Mn1
3 3 3
O4 X1
3 3 3
O7 O3
3 3 3
N1-Ni1-N10
O1-Ni1-O10
O3-Ni1-O30
N3-Ni2-O20
O1-Ni2-O6
O3-Ni2-O4
N5-Mn1-O20
O5-Mn1-O6
O7-Mn1-X1
Magnetic and EPR Properties. Magnetic properties of 1 and
2 were investigated using a commercial SQUID MPMS system as
well as a homemade micro-Hall-bar magnetometer.37 The result
of SQUID susceptibility measurements of a powder sample of 1
at magnetic field values of 0.5, 1, and 5 T in a temperature range
between 2 and 300 K are shown in Figure 3.
180
180
180
180
171.7(3)
164.7(2)
177.9(2)
167.2(3)
175.9(3)
175.4(2)
-
170.5(1)
165.0(1)
178.9(1)
169.4(2)
177.6(2)
174.5(1)
175(5)
Above 250 K, the χT value is about 11.5 K cm3 mol-1 and is
nearly temperature independent, in accordance with the pre-
sence of two high-spin MnIII (S = 2) and three NiII (S = 1). At low
temperatures, the sharp increase of the susceptibility to around
25 K cm3 mol-1 is an indication for ferromagnetic intramolecular
coupling. An effective Heisenberg Hamiltonian of the form
O4-H4 X1
3 3 3
O7-H7A O3
3 3 3
-
154(7)
a Symmetry transformation used to generate equivalent atoms: (0) 1-x,
1-y, 1-z. b X = Cl: 1; X = Br: 2
H ¼ - J1ðSMn1SNi1 þ SMn2SNi3Þ - J2ðSNi1SNi2 þ SNi2SNi3
Þ
ð1Þ
Mn-O/Br distances of 2.34/2.78 Å in 2 8CH2Cl2. In case of
3
nickel the octahedral environment is more regular with the
was used to fit the data. From the numerically calculated best fit,
shown as red lines in Figure 3, J1 ≈ 46 K for the Mn-Ni coupling
central NiII featuring a {N2O4} and the other two NiII ions a
3435
dx.doi.org/10.1021/ja108740c |J. Am. Chem. Soc. 2011, 133, 3433–3443

Journal of the American Chemical Society
ARTICLE
Figure 3. Temperature dependence of χT for a microcrystalline sample
of 1. The various fields are indicated in the inset. The red line is a
Heisenberg fit to the data (see text).
Figure 4. Magnetization versus field data for different orientations of a
single crystal of 1 8CH2Cl2 at 1.8 K and the corresponding powder data.
3
The red lines represent best fits. The inset shows a top view of a typical
crystal. o1, o2, o3 are the three different orientations indicated by the
sketch in the inset.
and J2 ≈ 22 K for the Ni-Ni coupling were obtained with an
average g factor gav = 2.06; the latter value is in a good agreement
with the value directly measured by ESR, gav(ESR) = 2.08 (see
below). Obviously the quasi-linear [MnIIINiIINiIINiIIMnIII] is a
purely ferromagnetically coupled system with a ground-state spin
of Stot = 7.
Figure 5 shows an ac susceptibility measurement of a poly-
crystalline sample of 1 8CH2Cl2 with a so-called Cole-Cole
3
(Argand) plot in the inset. The dispersion data presented in this
way reveal an almost perfect semicircle. This is clear evidence of a
relaxation process with a single relaxation time τ039 and suggests
The results of variable field magnetization measurements for a
polycrystalline sample and for a single crystal of 1 8CH2Cl2 with
SMM behavior of 1 8CH2Cl2.
Similar measurements were also performed for 2 8CH2Cl2 (see
Figures S4-S6 for variable-temperature and variable-field SQUID
data as well as ac susceptibility for a polycrystalline sample). For
3
3
different field orientations are shown in Figure 4. While the
magnetization of the powder sample and of the single-crystal
sample at orientations o2 and o3 is not saturated at a field of 5 T,
the crystal oriented along o1 saturates already at about 0.6 T.
From the saturation value MS ≈ 14.2 μB one can again derive a
spin ground state of Stot = 7, assuming gav = 2.03. This gav value,
though slightly smaller, is reasonably consistent with the values
obtained from the powder susceptibility and ESR data. The total
gav factor deviates from the value of 2.0 that is characteristic for
MnIII because of the contribution from NiII, which usually has g
values in the range 2.2-2.3.38
3
2 8CH2Cl2 coupling constants J1 ≈ 42 K (for Mn-Ni) and J2
3
≈ 18 K (for Ni-Ni) were found, again with a spin ground state of
Stot = 7. From the single-crystal measurements it was possible to
estimate a uniaxial anisotropy parameter Dmol ≈ -0.4(5) K, which
is well comparable with the result for complex 1.
High-field electron spin resonance (HF-ESR) measurements
in magnetic fields up to 15 T were performed on an oriented
powder sample of 1 8CH2Cl2. Owing to a strong magnetic
3
Since the magnetization measurements, presented in Figure 4,
have been performed at a temperature much lower than the
estimated strength of intramolecular couplings J1 and J2 (see
anisotropy (see below) the single-crystalline powder particles in
the sample were aligned in the magnetic field along their
crystalline magnetic anisotropy axis. A typical ESR spectrum at
a frequency ν of 332 GHz and T = 20 K is shown in Figure 6. The
spectrum features a relatively simple structure consisting of seven
well-defined and almost equally spaced absorption lines with a
separation of the resonance fields Bres by ∼1 T. The intensity of
the lines continuously decreases with increasing the magnetic
field strength. Measurements at different excitation frequencies
reveal a linear relationship between ν and Bres for each absorption
line. The respective ν vs Bres dependencies (resonance branches)
are plotted together in Figure 6. The slope dν/dB of all resonance
branches is almost identical. This enables a straightforward
determination of g = (h/μB)dν/dB = 2.08, which is in very good
eq 1), one can assume that only the ground-state multiplet Stot
=
7 is populated. The first exited multiplet which is sufficiently well
separated from the ground state due to large values of J1 and J2
does not contribute to the magnetization in this case. Therefore,
for the analysis of these data an effective spin Hamiltonian for the
total ground state spin Stot = 7 in the form
ꢀ
ꢁ
1
H ¼ D S2z - SðS þ 1Þ þ EðSx2 - Sy2Þ þ gμB HB BS ð2Þ
3
3
can be applied. From a numerical fit of this equation to the data,
we were able to obtain a rough value for the uniaxial anisotropy
parameter Dmol ≈ -0.5 K. Within this framework, one can get an
estimation for the effective energy barrier for spin reversal of
accordance with the values for 1 8CH2Cl2 derived from the
3
SQUID measurements. As shown in Figure 6 the extrapolation of
the ν(Bres) dependence of the leftmost line in the ESR spectrum
(line 1) to zero magnetic field yields the magnitude of the
magnetic anisotropy gap Δ, which is 150 GHz (7.2 K). Since
Δ is related to the anisotropy parameter Dmol of the molecular
complex as |Dmol| = Δ/[Stot2 - (Stot - 1)2], with Stot = 7 one
obtains the absolute value |Dmol| = 11 GHz (0.55 K) for
2
Ueff = |Dmol|Stot ≈ 25 K. Unfortunately, it was not possible to
determine a reliable value of the E parameter with this method, as
the sample crystals were rather small (∼20 μg) and very thin.
Furthermore, the softness of the crystal made it almost impos-
sible to align them with the necessary angular precision. How-
ever, from the measurements it can be estimated 10 |E| e |D|,
1 8CH2Cl2.40,33b This value again is in good agreement with
3
3
which gives an upper limit for |E| of 45 mK.
the value derived from magnetization data.
3436
dx.doi.org/10.1021/ja108740c |J. Am. Chem. Soc. 2011, 133, 3433–3443

Journal of the American Chemical Society
ARTICLE
Figure 5. Plots of in-phase (m0, left) and out-of-phase (m00, right) ac susceptibility versus temperature for a microcrystalline sample of 1 8CH2Cl2. The
3
data were collected in an ac field of 2.56 Oe oscillating at the indicated frequencies. The inset in the right graph shows a Cole-Cole plot (m00 versus m0).
Figure 6. Frequency ν versus resonance magnetic field Bres dependence
of the ESR lines (symbols) and a representative ESR spectrum of
1 8CH2Cl2 at T = 20 K. Solid lines are linear fits to the experimental
3
data points. Inset: Comparison of the experimental and modeled ESR
spectra (see text).
Figure 7. Temperature dependence of the ESR spectra of 1 8CH2Cl2
at ν = 166 GHz. (a) Experimentally obtained ESR spectra; (b) calculated
3
ESR spectra and energy levels (see text).
The sign of Dmol can be unambiguously determined from the
temperature dependence of the ESR spectrum which is pre-
sented in Figure 7a. Here, one observes a clear shift of the spectral
weight to lower magnetic fields at low temperatures, whereas the
positions of the absorptions do not change. Such redistribution
of the ESR spectral weight occurs in the case of a spin doublet
main features of the experimentally observed ESR spectrum: the
extent of the fine structure of the spectrum, the number of lines,
and their relative intensities. Discrepancies between the model
and the experiment could be due to a nonperfect alignment of the
powder particles as well as to the simplifications of the model that
neglects intermolecular interaction. Beyond this model descrip-
tion in the experimental ESR spectrum at T = 20 K an additional
wide absorption line emerges at around 5.5 T (see Figure 7a),
which can most probably be attributed to transitions inside the
higher-energy spin multiplets.
ground state |(StotZæ corresponding to the negative sign of Dmol
,
whereas for the case of a singlet state |0æ, i.e. for a positive sign of
D
mol, the spectral weight should shift to higher fields.33b This is
illustrated by the energy level diagram in Figure 7b where, for
Dmol < 0 at low temperatures, only the lowest state |-7æ is
sufficiently populated, yielding the strongest intensity for the
leftmost line in the ESR spectrum. Thus, the HF-ESR results give
clear evidence for a bi-stable ‘easy axis’ ground state of
To confirm further the SMM properties of 1 and 2 a micro-
Hall-bar magnetometer was used for low-temperature single-
crystal magnetization measurements. The crystals were oriented
with the easy axis approximately parallel to the applied magnetic
field. Measurements were performed at several temperatures in
the range of 0.3-2.0 K, and sweep rates in the range of 12-200
mT s-1. The left parts of Figures 8 and 9 show hysteresis loops of
1 8CH2Cl2 and 2 8CH2Cl2 with a constant field sweep rate at
1 8CH2Cl2 with a substantial anisotropy barrier U = Stot2|Dmol| =
3
Z
27 K between two degenerate ground states Stot = þ7 and
StotZ = -7.
With the parameters obtained from the above analysis the ESR
spectrum of 1 8CH2Cl2 was modeled using the EasySpin tool-
3
3
3
box for Matlab41 for the case of the parallel orientation of the
‘easy axis’ of the complex to the direction of the magnetic field, i.e.
the situation that is expected for the oriented powder sample.
This model, according to eq 2, describes only the ground state of
the molecule while neglecting the excited higher-energy spin
states. As can be seen in Figure 7 the simulation captures well the
different temperatures. For both complexes, hysteretic behavior
was observed below a temperature TB of approximately 1.5 K.
The sweep rate dependence is depicted in the right parts of
Figures 8 and 9. Both compounds show sharp QTM steps, which
are particularly pronounced for 1 8CH2Cl2. An exemplary low-
3
temperature hysteresis loop of 1 8CH2Cl2 measured at 300 mK
3
3437
dx.doi.org/10.1021/ja108740c |J. Am. Chem. Soc. 2011, 133, 3433–3443

Journal of the American Chemical Society
ARTICLE
Figure 8. Magnetization (M) versus dc field hysteresis loops for a single crystal of 1 8CH2Cl2 at constant sweep rate and variable temperature (left) and
3
constant temperature and variable sweep rate (right).
Figure 9. Magnetization (M) versus dc field hysteresis loops for a single crystal of 2 8CH2Cl2 at constant sweep rate (left) and at a constant
3
temperature of 300 mK and different field scan rates (right).
with the field sweep rate of 200 mT/s is compared in Figure 10
with the energy scheme of the spin levels of the ground state
multiplet. The scheme was calculated using the simplest model
(eq 2), with the parameters taken from the SQUID and ESR
measurements (Stot = 7; Dmol = -0.55 K; gav = 2.08). The model
qualitatively describes the main features of the magnetization
behavior such as the number and approximate positions of the
relaxation steps which correspond to the level crossing points in
the diagram. The correspondence between the crossing points
and the positions of the steps that indicate the flip of the
magnetization vector due to the quantum tunneling is closest
for the highest sweep rate of the magnetization measurement and
gets worse for smaller sweep rates. Indeed, the rate of the spin
flips in the ESR experiment determined by the excitation
frequency is by orders of magnitude higher. Thus, the level
diagram derived from the ESR data should correspond to a
magnetization measurement performed at a virtually infinite
sweep rate.
Figure 10. Magnetization vs dc field hysteresis loops for a single crystal
of 1 8CH2Cl2 at a temperature of 300 mK and the field sweep rate of
3
However, despite similar model parameters the overall
shape of the hysteresis loops look quite different for
1 8CH2Cl2 and 2 8CH2Cl2. In particular, in the case of
200 mT/s (black line) and the calculated energy level scheme.
3
3
between two neighboring molecules can be described by a
simple effective interaction Hamiltonian
1 8CH2Cl2 the first relaxation step in the hysteresis loop is
3
above the zero crossing of the magnetic field, while in the
case of 2 8CH2Cl2 it is below the zero crossing. It is
X
3
2
reasonable to assume that these features are mainly deter-
mined by relaxation paths caused by intermolecular interac-
tions, with ferromagnetic intermolecular interactions in the
HD
¼
ðDSiz þ gμbμ0SizHzÞ - J12S1S2
i ¼ 1,2
Within this framework, one can estimate the intermolecular
coupling interactions for both systems by the following
case of 1 8CH2Cl2 and predominantly antiferromagnetic
3
interactions in the case of 2 8CH2Cl2.19 The interaction
3
3438
dx.doi.org/10.1021/ja108740c |J. Am. Chem. Soc. 2011, 133, 3433–3443

Journal of the American Chemical Society
ARTICLE
equation:18
exceeds the memory capacity of currently available computers. It
is possibly owing to the relative smallness of the intermolecular
interactions and broadness of the absorption lines that the ESR
spectra do not exhibit any characteristic splitting of the resonance
lines due to the exchange bias.19
gμbμ0HS
Jꢀ -
S2
This leads to a value for J of around 25 mK (ferromagnetic) for
1 and -35 mK (anitferromagnetic) for 2. Within this simplified
dimer model the zero field step and some further steps of the
hysteresis loops at higher field can be described, but not all of
them. To fully describe the hysteresis features one would need to
extend this model to a chain of more than two molecules
interacting with each other. At the moment, however, it is not
possible to accurately simulate this extended intermolecular cou-
pling scheme because the dimension of the resulting Hamiltonian
Intermolecular Interactions. In view of the importance of
weak intermolecular interactions for the QTM properties of
SMMs and the differing characteristics of 1 and 2, we have
taken a closer look at the crystal packing and intermolecular
contacts of the new {Mn2Ni3} compounds. One should note
that 1 8CH2Cl2 and 2 8CH2Cl2 are isostructural, and hence
3
3
there are only minor differences. Classic hydrogen bonds in
1 8CH2Cl2 and 2 8CH2Cl2 are all intramolecular (as de-
3
3
scribed above), but additional intermolecular interaction
might occur via short C-H X (X = Cl, Br, O) contacts,
Table 2. Selected Intermolecular Distances [Å] for
1 8CH2Cl2 and 2 8CH2Cl2
3 3 3
3
3
of which several are observed in the crystal structures. In this
context short contacts are defined as a distance shorter than
the sum of the van der Waals radii of the respective atoms
atom1
atom2 symm. op. 1 symm. op. 2 distance length-vdW
with a C-H
X angle greater than 90ꢀ. Such distances were
1 8CH2Cl2:
3
3 3 3
determined using the program Mercury42 and are listed in
Table 2.
Cl1
O5
O7
Cl4
Cl2
Cl3
Cl2
Cl7
H61A
H62A
H64B
H46
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
1-x, 1-y, 2-z 2.664
1-x, 1-y, 2-z 2.401
-0.286
-0.319
-0.335
-0.015
-0.056
-0.213
-0.078
-0.024
Interestingly, within the given range no direct molecular
x, 1þy, z
1-x, 2-y, 2-z 2.935
x, y, z 2.894
2.385
contacts of two or more [MnIII2NiII Cl2L4(LH)2(H2O)2] units
3
are observed for 1 8CH2Cl2. Instead the CH2Cl2 solvent
molecules containing the atoms Cl2/3 and Cl4/5 provide
3
H61B
H43
-1þx, -1þy, z 2.737
1-x, 1-y, 2-z 2.872
interactions to neighboring molecules 1 via a network of C-
H61B
H51B
H
X (X = Cl, O) contacts (Figure 11). However, H-atoms
3 3 3
from neighboring [MnIII2NiII Cl2L4(LH)2(H2O)2] molecules
-1þx, y, z
2.926
3
are not too far away. The closest contact between the manganese-
bound chlorine atom and the next H-atom of a surround-
ing [MnIII2NiII3Cl2L4(LH)2(H2O)2] moiety is 3.013 Å
2 8CH2Cl2:
3
Br1
Br1
Br1
Br1
O5
O7
Cl2
Cl3
H31A
H41
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
2-x, 1-y, 2-z 3.035
2-x, 2-y, 2-z 2.998
1-x, 1-y, 2-z 2.992
1-x, 1-y, 2-z 3.022
1-x, 1-y, 2-z 2.456
-0.015
-0.052
-0.058
-0.028
-0.264
-0.141
-0.003
-0.101
(Cl1 H41-C41[2-x,2-y,2-z]); the latter belongs to a
3 3 3
H61B
H62B
H62B
H64A
H61B
H43
pyrazole group. In case of 2 8CH2Cl2 two short contacts of
3
the bromine atom of one [MnIII2NiII Br2L4(LH)2(H2O)2] to
3
close-by hydrogen atoms H-C (H31A and H41) result in direct
contacts to the neighboring [MnIII2NiII Br2L4(LH)2(H2O)2]
x, 1þy, z
x, y, z
2.579
2.947
3
molecules. Additional intermolecular contacts via solvent mole-
cules containing Cl2/Cl3 are present as well (Figure 12).
-1þx, -1þy, z 2.849
Figure 11. Emphasis of some intermolecular C-H X contacts in 1 8CH2Cl2. Symmetry transformation used to generate equivalent atoms: (0)
3 3 3
3
1-x, 1-y, 2-z. Other symmetry transformations are given in the figure.
3439
dx.doi.org/10.1021/ja108740c |J. Am. Chem. Soc. 2011, 133, 3433–3443

Journal of the American Chemical Society
ARTICLE
It seems that in both compounds at least one close inter-
similar and show an almost parallel alignment of both rings
(—P1-P2 in Table 3). The distance of both ring-centroids is
approximately 3.8 Å, but they are displaced relative to each other,
which is reflected by the angle between the Cg(1)-Cg(2) vector
and the normal to one of the planes formed by the ring atoms
(—CgCg-P in Table 3). Hence, the rings are offset in such a
way that a ring atom almost lies over the center of the other ring.
This is a rather common situation in complexes with aromatic
nitrogen heterocycles and reveals that such an arrangement also
has a contribution from π-σ attraction.44 It is important for the
present discussion, however, that metric parameters collected in
Table 3 are rather similar for 1 8CH2Cl2 and in 2 8CH2Cl2 and,
molecular contact between individual [MnIII2NiII X2L4(LH)2-
3
(H2O)2] molecules is present that involves the manganese-
bound halide atom and the hydrogen atom of a neighboring
pyrazole group, namely H41. Since a center of inversion is
present within each molecule a one-dimensional (1D) chainlike
arrangement is formed within the crystal (Figure 13). However,
this contact is shorter than the sum of the van der Waals radii only
in the case of 2 8CH2Cl2. One might thus speculate that this
3
contact, because of the larger bromine atom, serves as the
dominant exchange pathway in 2 8CH2Cl2, mediating an anti-
3
ferromagnetic interaction, whereas it is negligible in 1 8CH2Cl2.
3
3
3
In the latter case exchange interactions are thus weaker (see the
lower absolute value: þ25 mK for 1 and -35 mK for 2) and are
mainly mediated indirectly via CH2Cl2 solvent molecules in the
crystal lattice, or dipolar interactions through space, resulting in a
ferromagnetic situation. It should be kept in mind though that
this is a rather reductionistic treatment that should be considered
at the present stage of knowledge, cannot rationalize the different
magnetic properties of the two compounds.
’ CONCLUSIONS
The new diprotic and potentially tridentate ligand H2L
allowed to synthesize and characterize first members of a new
with care, since the intermolecular C-H X contact involving
family of heterometallic MnIII2NiII complexes that exhibit a
3 3 3
3
H41 is only one (albeit the most obvious) of several intermole-
cular interactions which, as a whole, build up the 3D crystal
structure.
robust quasi-linear structure supported by various intramolecular
hydrogen bonds, with variable terminal ligands X (Cl and Br in
the case of 1 and 2). These complexes exhibit relatively strong
intramolecular ferromagnetic coupling via the alkoxo and phe-
noxo O atoms, giving an Stot = 7 ground state, and sizable uniaxial
anisotropy that is reinforced structurally by the intramolecular
hydrogen bonds fixing the elongated Jahn-Teller axes of the
individual coordination polyhedra roughly parallel. Hence these
new complexes show properties typical for SMM behavior, with
an energy barrier to thermal relaxation of the magnetization of 27
K (1) or 22 K (2). Noteworthy are the pronounced QTM steps
in the hysteresis curve of the magnetization, which are, however,
distinctly different for 1 and 2 since QTM between the ground
states does not occur at static zero field but is shifted to positive
(1) or negative (2) fields. This is likely due to intermolecular
contacts between the individual SMM units, causing the quan-
tized energy sublevels of the molecules to interact coherently and
produce superposition states.17,18,22 Such opposed exchange bias
The H-bonding contacts in 1 8CH2Cl2 and in 2 8CH2Cl2
3
3
seem to be further supported by π-π ring interactions. Close
inspection of the rings containing the atoms N5, N6, C41-C43
(pyrazole part of the ligand), and C44-C49 (phenol part of the
ligand, excluding O) reveals the possibility for such intermole-
cular interactions. Relevant parameters were calculated with the
program PLATON43 and are listed in Table 3. Values are rather
for the two complexes, viz. ferromagnetic for 1 8CH2Cl2 but
3
antiferromagnetic for 2 8CH2Cl2, is surprising, given that
3
1 8CH2Cl2 and 2 8CH2Cl2 are isostructural. Close inspection
3
3
of the crystal packing reveals only subtle differences, which are
mainly due to the different size of the peripheral Cl versus Br
ligands at MnIII and the resulting extent of C-H X (X = Cl,
3 3 3
Br) contacts with neighboring molecules. These interactions
through C-H X hydrogen bonds, albeit very weak, might
3 3 3
possibly be responsible for the significant effect on the quantum
properties of these SMMs, which shows that at the present stage
of knowledge and the inherent difficulties associated with crystal
Figure 12. Emphasis of some intermolecular C-H X contacts in
3 3 3
2 8CH2Cl2. Symmetry transformations used to generate equivalent
3
atoms: (0) 2-x, 2-y, 2-z, (00) 1-x, 1-y, 2-z, (000) 2-x, 1-y, 2-z.
Figure 13. Emphasis of intermolecular C-H Cl contacts in 1 8CH2Cl2. Symmetry transformations used to generate equivalent atoms: (0) 1-x, 1-
3 3 3
3
y, 1-z, (00) -1þx, -1þy, -1þz.
3440
dx.doi.org/10.1021/ja108740c |J. Am. Chem. Soc. 2011, 133, 3433–3443

Journal of the American Chemical Society
ARTICLE
Table 3. Parameters for Selected Short Ring Interactions for 1 8CH2Cl2 and 2 8CH2Cl2.a
3
3
d(Cg1-Cg2) [Å]b
—P1-P2 [deg]c
d(Cg1-P2) [Å]/d(Cg2-P1) [Å]d
—CgCg-P2 [deg]/—CgCg-P1 [deg]e
1 8CH2Cl2:
3.860
3.818
4.4
6.8
3.311/3.197
3.336/3.130
34.1/30.9
3
2 8CH2Cl2:
34.9/29.1
3
a 1 = N5, N6, C41-C43 [x, y, z], 2 = C44-C49 [2-x, 2-y, 2-z], Cg = ring-centroid, P = plane. b Distance between Cg. Angle between P.
c
d Perpendicular distance of Cg(x) on ring (y). e Angle Cg(1)-Cg(2) vector and normal to P.
engineering any full control of inter-SMM interactions is still far
beyond the means of the chemist.
and H2L (30 mg, 0.147 mmol, 1 equiv) in acetonitrile (15 mL) at room
temperature. The resultant green solution was then added dropwise to a
brown solution of MnCl2 4H2O (29 mg, 0.147 mmol, 1 equiv), H2L(30mg,
Attractive features of this family of SMMs, which are relevant
to further investigations toward the targeted modulation of SMM
properties, are (i) the robustness of the Mn2Ni3 core that is
reinforced by the intramolecular hydrogen bonds, minimizing
3
0.147 mmol, 1 equiv), and NEt3 (23 mg, 0.22 mmol, 1.5 equiv) in
acetonitrile (15 mL), and the reaction mixture was stirred for 8 h at room
temperature. Evaporation of all volatile material under reduced pressure
gave a brown solid. This crude product was dissolved in dichloro-
structural distortions of the MnIII2NiII SMM building blocks;
3
methane (25 mL), and plate-shaped brown crystals of 1 8CH2Cl2 were
(ii) the relatively strong intramolecular ferromagnetic coupling
that separates excites states from the Stot = 7 ground state; (iii)
the pronounced QTM steps in the hysteresis curves, and (iv) the
possibility of exchanging the terminal ligands X at the outer MnIII
ions. In fact, variation of the peripheral ligands X, including
bridging ligands that establish direct linkages between the
individual complex units, should allow the rational assembly of
1D chains of heterometallic SMMs45 and a more detailed study
of the effect of intermolecular interactions (and hence the
exchange bias) in such correlated systems. Studies in this
direction are ongoing in our laboratories.
3
obtained by slow evaporation of the solution. Due to the loss of solvent
molecules the crystals crumble after separation from the mother liquor.
Yield: 40 mg (48%). IR (KBr): υ~ 3582 (w), 3458 (br), 3131 (w), 3111
(w), 3052 (w), 2928 (m), 2841 (m), 1627 (m), 1598 (s), 1563 (m),
1502 (w), 1506 (s), 1470 (s), 1443 (s), 1412 (m), 1357 (w), 1308 (s),
1256 (w), 1234 (w), 1204 (w), 1132 (s), 1084 (s), 948 (w), 857 (m),
850 (w), 805 (w), 766 (m), 743 (w), 701 (m), 642 (w), 609 (m) cm-1
.
Elemental analysis (%) calculated for C67.5H69N12O14Mn2Ni3Cl5
(1 1.5CH2Cl2; 1735.6): C 46.71, H 4.01, N 9.68; found: C 46.46, H
3
4.18, N 9.67. Most of the lattice solvent molecules are lost upon drying
the crystalline material.
II
Synthesis of [Mn2IIINi3 L4(LH)2Br2(H2O)2] (2). Compound 2 was
obtained in a similar manner as described for 1, but using MnBr2 4H2O
3
’ EXPERIMENTAL SECTION
instead of MnCl2 4H2O. Brown crystals of 2 8CH2Cl2 were obtained
3
3
by slow evaporation of a solution of the crude product in dichlor-
methane. Yield: 50 mg (54%). IR (KBr): υ~ 3588 (w), 3437 (br), 3113
(w), 3052 (w), 2928 (m), 2842 (m), 2363 (w), 1629 (w), 1598 (s), 1561
(m), 1507 (s), 1469 (s), 1443 (s), 1412 (m), 1361 (m), 1325 (w), 1306
(s), 1269 (w), 1256 (w), 1233 (w), 1202 (w), 1156 (w), 1134 (s), 1084
(s), 1050 (w), 948 (w), 856 (m), 841 (w), 769 (w), 756 (m), 742 (w),
700 (m), 664 (w), 635 (w), 638 (w), 610 (m) cm-1. Elemental analysis
(%) calculated for C67.5H69N12O14Mn2Ni3Cl3Br2 (1731.5): C 44.44, H
3.81, N 9.21; found: C 44.29, H 4.16, N 9.15.
General Procedures and Materials. All manipulations were
performed under aerobic conditions using chemicals and solvents as
received unless otherwise stated. 1-(2-Hydroxyphenyl)-3-dimethylami-
noprop-2-enone as a precursor of the ligand H2L was synthesized as
described elsewhere.34
Synthesis of 2-[3-(2-Hydroxyphenyl)-1H-pyrazol-1-yl]ethanol
(H2L). A mixture of 1-(2-hydroxyphenyl)-3-dimethylaminoprop-2-
enone (20 g, 0.104 mol) and 2-hydroxyethyl hydrazine (29.3 g, 0.384 mol)
in methanol (100 mL) was heated to reflux with stirring for 3 h. Then the
methanolic solution was poured on ice, and the resulting brown
precipitate was isolated by filtration and dried in air. The crude product
was purified by Soxhlet extraction with light petroleum over several days
to give H2L as needle-shaped white crystals. Yield: 9.3 g (0.045 mol,
Physical Measurements. Melting points were determined with a
SRS (Stanford Research Systems) Opti Melt apparatus using open
capillaries; the values are uncorrected. Elemental analyses were per-
formed by the analytical laboratory of the Institute of Inorganic
Chemistry at Georg-August-Universit€at G€ottingen using an Elementar
Vario EL III. NMR measurements were performed, unless indicated
otherwise, at 25 ꢀC on Bruker Avance 200, 300, and 500 spectrometers.
1H and 13C chemical shifts were calibrated internally to the solvent
signals (7.24 and 77.0 ppm for CDCl3). EI-MS-spectra were recorded on
a Finnigan MAT 95. Values for m/z are given for the most intense peak
of the isotope envelope. IR spectra were recorded on a Digilab Excalibur
series FTS 3000 spectrometer. Variable-temperature, solid-state direct
current (dc) and alternating current (ac) magnetic susceptibility data
down to 1.8 K were collected on a Quantum Design MPMS-5S SQUID
magnetometer equipped with a 5 T dc magnet (for these measurements
the powdered samples were contained in a gel bucket and fixed in a
nonmagnetic sample holder). Diamagnetic corrections were applied to
the observed paramagnetic susceptibilities using Pascal’s constants.
Magnetic studies below 1.8 K were carried out on single crystals, using
a homemade micro-Hall-bar magnetometer equipped with several 10 ꢁ
10 μm2 Hall bars. Micrometer-sized single crystals of the sample were
placed on top of the active area of the Hall bars. The crystals were
oriented with the easy axis approximately parallel to the applied
magnetic field. High-field/high-frequency electron spin resonance
1
44%). Mp 91 ꢀC. H NMR (500 MHz, DMSO-d6): δ 10.68 (s, Ar-
OH), 7.84 (d, 1H, J = 2.5 Hz), 7.70 (dd, 1H, J = 7.5, 1.5 Hz), 7.15 (ddd,
1H, J = 1.5, 6.75, and 8.5 Hz), 6.89 (dd, 1H, J = 1 Hz, 8 Hz), 6.87 (dd, 1H,
J = 1 Hz, 7.5 Hz), 6.85 (d, 1H, J = 1 Hz), 4.93 (t, 1H, J = 5 Hz), 4.22 (t,
2H, J = 5.5 Hz), 3.77 (q, 2H, J = 5.5 Hz) ppm. 13C{1H} NMR (125.75
MHz, DMSO-d6): δ 154.9, 149.7, 132.1, 128.6, 126.6, 119.2, 117.2,
116.4, 102.5, 59.8, 54.1 ppm. MS (EI): m/z (%) = 205 (20), [M þ H]þ,
204 (100), [M]þ, 173 (80), [M - CH2OH], 160 (40), [M -
CH2CH2O]. IR (KBr): υ~ 33478 (w), 3325 (br), 3015 (w), 2958 (w),
2926 (w), 2871 (w), 1620 (m), 1584 (m), 1508 (m), 1487 (m), 1466
(m), 1430 (w), 1415 (m), 1360 (w), 1350 (w), 1317 (w), 1291 (m),
1248 (s), 1211 (w), 1152 (w), 1080 (m), 1059 (s), 1045 (w), 935 (m),
861 (m), 825 (m), 782 (m), 758 (m), 702 (m), 702 (m), 675 (m) cm-1
.
Elemental analysis (%) calculated for C11H12N2O2 (204.2): C 64.69, H
5.92, N 13.72; found: C 64.71, H 6.05, N 13.73. Single crystals were
obtained upon slow cooling of a saturated solution in light petroleum.
The solid-state structure of H2L was determined by X-ray diffraction.
II
Synthesis of [Mn2IIINi3 L4(LH)2Cl2(H2O)2] (1). A solution of NEt3
(23 mg, 0.22 mmol, 1.5 equiv) in acetonitrile (3 mL) was added
dropwise to a solution of NiCl2 6H2O (26 mg, 0.110 mmol, 0.75 equiv)
3
3441
dx.doi.org/10.1021/ja108740c |J. Am. Chem. Soc. 2011, 133, 3433–3443

Journal of the American Chemical Society
ARTICLE
(HF-ESR) studies in a frequency range 50-400 GHz in magnetic fields
up to 15 T were performed using a homemade spectrometer based on a
Millimeter Wave Vector Network Analyzer (AB Millimeteꢀr).46
Cornia, A.; Fabretti, A. C.; Gatteschi, D.; Mortalo, C.; Olivieri, E.;
Parenti, F.; Rosa, P.; Sessoli, R.; Sorace, L.; Wernsdorfer, W.; Zobbi, L.
J. Am. Chem. Soc. 2006, 128, 4742–4755. (c) Boudalis, A. K.; Sanakis, Y.;
Clemente-Juan, J. M.; Donnadieu, B.; Nastopoulos, V.; Mari, A.; Coppel,
Y.; Tuchagues, J.-P.; Perlepes, S. P. Chem.—Eur. J. 2008, 14, 2514–2526.
(d) Ako, A. M.; Mereacre, V.; Lan, Y.; Wernsdorfer, W.; Clꢀerac, R.;
Anson, C. E.; Powell, A. K. Inorg. Chem. 2010, 49, 1–3.
(8) (a) Moubaraki, B.; Murray, K. S.; Hudson, T. A.; Robson, R. Eur.
J. Inorg. Chem. 2008, 4525–4529. (b) Murray, K. S. Aust. J. Chem. 2009,
62, 1081–1101. (c) Murrie, M. Chem. Soc. Rev. 2010, 39, 1986–1995.
(9) (a) Andres, H.; Basler, R.; Blake, A. J.; Cadiou, C.; Chaboussant,
G.; Grant, C. M.; G€udel, H.-U.; Murrie, M.; Parsons, S.; Paulsen, C.;
Semadini, F.; Villar, V.; Wernsdorfer, W.; Winpenny, R. E. P. Chem.—
Eur. J. 2002, 8, 4867–4876. (b) Ochsenbein, S. T.; Murrie, M.; Rusanow,
E.; Stoeckli-Evans, H.; Sekine, C.; G€udel, H.-U. Inorg. Chem. 2002, 41,
5133–5140. (c) Bell, A.; Aromi, G.; Teat, S. J.; Wernsdorfer, W.;
Winpenny, R. E. P. Chem. Commun. 2005, 2808–2810. (d) Aromi, G.;
Parsons, S.; Wernsdorfer, W.; Brechin, E. K.; McInnes, E. J. L. Chem.
Commun. 2005, 5038–5040.
X-ray Crystallography. Crystal data and details of the data
collections are given in Table S1 (Supporting Material). X-ray data
were collected on a STOE IPDS II diffractometer (graphite mono-
chromated Mo KR radiation, λ = 0.71073 Å) by use of ω scans at -140 ꢀC.
The structures were solved by direct methods and refined on F2 using all
reflections with SHELX-97.47 All non-hydrogen atoms were refined
anisotropically. Most hydrogen atoms were placed in calculated posi-
tions and assigned to an isotropic displacement parameter of 0.08 Å2.
In the case of the oxygen-bound hydrogen atoms H4 and H7A/B in
2 8CH2Cl2 DFIX restraints (dO-H = 0.82 Å) were applied. Those
3
hydrogen atoms could not be located in 1 8CH2Cl2. No restraints or
3
constraints were used for the oxygen-bound hydrogen atoms in H2L.
The correct absolute structure of H2L could not be determined. Face-
indexed absorption corrections for all compounds were performed
numerically with the program X-RED.48
(10) Milios, C. J.; Vinslava, A.; Wernsdorfer, W.; Moggach, S.;
Parsons, S.; Perlepes, S. P.; Christou, G.; Brechin, E. K. J. Am. Chem.
Soc. 2007, 129, 2754–2755.
’ ASSOCIATED CONTENT
(11) (a) Cremades, E.; Cano, J.; Ruiz, E.; Rajaraman, G.; Milios,
C. J.; Brechin, E. K. Inorg. Chem. 2009, 48, 8012–8019. (b) Inglis, R.;
Jones, L. F.; Milios, C. J.; Datta, S.; Collins, A.; Parsons, S.; Wernsdorfer,
W.; Hill, S.; Perlepes, S. P.; Piligkos, S.; Brechin, E. K. Dalton Trans.
2009, 3403–3412.
S
Supporting Information. CIF files; plots of the molecu-
b
lar structures of 1 8CH2Cl2, 2 8CH2Cl2 and H2L with atom
distances and bond angles and crystal data. This material is
available free of charge via the Internet at http://pubs.acs.org.
3
3
(12) Gatteschi, D.; Sessolis, R.; Villain, J. Molecular Nanomagnets;
Oxford University Press: New York, 2006.
’ AUTHOR INFORMATION
(13) Cirera, J.; Ruiz, E.; Alvarez, S.; Neese, F.; Kortus, J. Chem.—Eur.
J. 2009, 15, 4078–4087.
(14) Oshio, H.; Nakano, M. Chem.—Eur. J. 2005, 11, 5178–5185.
(15) Waldmann, O. Inorg. Chem. 2007, 46, 10035–10037.
(16) Ruiz, E.; Cirera, J.; Cano, J.; Alvarez, S.; Loose, C.; Kortus, J.
Chem. Commun. 2008, 52–54.
(17) Hill, S.; Datta, S.; Liu, J.; Inglis, R.; Milios, C. J.; Feng, P. L.;
Henderson, J. J.; del Barco, E.; Brechin, E. K.; Hendrickson, D. N. Dalton
Trans. 2010, 39, 4693–4707.
(18) Wernsdorfer, W.; Aliaga Alcalde, N.; Hendrickson, D. N.;
Christou, G. Nature 2002, 416, 406–409.
(19) (a) Hill, S.; Edwards, R. S.; Aliaga-Alcade, N.; Christou, G.
Science 2003, 302, 1015–1018. (b) Tiron, R.; Wernsdorfer, W.; Foguet-
Albiol, D.; Aliaga-Alcade, N.; Christou, G. Phys. Rev. Lett. 2003, 91
(227203), 1–4.
(20) (a) Yamaguchi, A.; Kusumi, N.; Ishimoto, H.; Mitamura, H.;
Goto, T.; Mori, N.; Nakano, M.; Awaga, K.; Yoo, J.; Hendrickson, D. N.;
Christou, G. J. Phys. Soc. Jpn. 2002, 71, 414–417. (b) Affronte, M.;
Lasjaunias, J. C.; Wernsdorfer, W.; Sessoli, R.; Gatteschi, D.; Heath, S. L.;
Fort, A.; Rettori, A. Phys. Rev. B 2002, 66, 064408/1–064408/7. (c)
Asada, H.; Ozeki, M.; Fujiwara, M.; Matsushita, T. Chem. Lett. 1999,
525–526. (d) Dey, M.; Rao, C. P.; Saarenketo, P. K.; Rissanen, K. Inorg.
Chem. Commun. 2002, 5, 924–928. (e) Wittick, L. M.; Murray, K. S.;
Moubaraki, B.; Batten, S. R.; Spiccia, L.; Berry, K. J. Dalton Trans. 2004,
1003–1011.
(21) Boskovic, C.; Bircher, R.; Tregenna-Piggott, W. L. P.; G€udel,
U. H.; Paulsen, C.; Wernsdorfer, W.; Baara, -L. A.; Khatsko, E.; Neels, A.;
Stoeckli-Evans, H. J. Am. Chem. Soc. 2003, 125, 14046–14058.
(22) Miyasaka, H.; Yamashita, M. Dalton Trans. 2007, 399–406.
(23) Oshio, H.; Nihei, M.; Yoshida, A.; Nojiri, H.; Nakano, M.;
Yamaguchi, A.; Karaki, Y.; Ishimoto, H. Chem.—Eur. J. 2005, 11, 843–848.
(24) Oshio, H.; Nihei, M.; Koizumi, S.; Shiga, T.; Nojiri, H.; Nakano,
M.; Shirakawa, N.; Akatsu, M. J. Am. Chem. Soc. 2005, 127, 4568–4569.
(25) Kachi-Terajima, C.; Miyasaka, H.; Saitoh, A.; Shirakawa, N.;
Yamashita, M.; Clꢀerac, R. Inorg. Chem. 2007, 46, 5861–5872.
(26) Li, D.; Parkin, S.; Wang, G.; Yee, T. G.; Cle’ rac, R.;
Wernsdorfer, W.; Holmes, M. S. J. Am. Chem. Soc. 2006, 128, 4214–4215.
(27) Glaser, T.; Heidemeier, M.; Weyherm€uller, T.; Hoffmann,
R.-D.; Rupp, H.; M€uller, P. Angew. Chem., Int. Ed. 2006, 45, 6033–6037.
Corresponding Author
V.Kataev@ifw-dresden.de; phm@physik.uni-erlangen.de; franc.
meyer@chemie.uni-goettingen.de
’ ACKNOWLEDGMENT
Financial support by the DFG (FM: SFB 602 “Complex
Structures in Condensed Matter”; Y.K., R.K., V.K. and B.B.:
FOR 1154 “Towards Molecular Spintronics”) is gratefully ac-
knowledged.
’ REFERENCES
(1) (a) Christou, G.; Gatteschi, D.; Hendrickson, D. N.; Sessoli, R.
MRS Bull. 2000, 25, 66–71. (b) Chudnovsky, E. M. Science 1996, 274,
938–939. (c) Gatteschi, D.; Sessoli, R. Angew. Chem. 2003, 115, 278–
309.
(2) (a) Affronte, M. J. Mater. Chem. 2009, 19, 1731–1737. (b)
Ardavan, A.; Blundell, S. J. J. Mater. Chem. 2009, 19, 1754–1760. (c)
Lehmann, J.; Gaita-Ari~no, A.; Coronado, E.; Loss, D. J. Mater. Chem.
2009, 19, 1672–1677. (d) Camarero, J.; Coronado, E. J. Mater. Chem.
2009, 19, 1687–1684. (e) Bogani, L.; Wernsdorfer, W. Nat. Mater. 2008,
7, 179–186.
(3) (a) Sessoli, R.; Gatteschi, D.; Caneschi, A.; Novak, M. A. Nature
1993, 365, 141–143. (b) Sessoli, R.; Tsai, H. L.; Schake, A. R.; Wang,
S. Y.; Vincent, J. B.; Folting, K.; Gatteschi, D.; Christou, G.; Hendrickson,
D. N. J. Am. Chem. Soc. 1993, 115, 1804–1816.
(4) Bagai, R.; Christou, G. Chem. Soc. Rev. 2009, 38, 1011–1026.
(5) (a) Miyasaka, H.; Clꢀerac, R.; Wernsdorfer, W.; Lecren, L.;
Bonhomme, C.; Sugiura, K.-I.; Yamashita, M. Angew. Chem. 2004,
116, 2861–2865. (b) Sanudo, E. C.; Wernsdorfer, W.; Abboud, K. A.;
Christou, G. Inorg. Chem. 2004, 43, 4137–4144. (c) Taguchi, T.;
Wernsdorfer, W.; Abboud, K. A.; Christou, G. Inorg. Chem. 2010, 49,
199–208.
(6) Sun, Z.; Grant, C. M.; Castro, S. L.; Hendrickson, D. N.;
Christou, G. Chem. Commun. 1998, 721–722.
(7) (a) Gatteschi, D.; Sessoli, R.; Cornia, A. Chem. Commun. 2000,
725–732. (b) Accorsi, S.; Barra, A.-L.; Caneschi, A.; Chastanet, G.;
3442
dx.doi.org/10.1021/ja108740c |J. Am. Chem. Soc. 2011, 133, 3433–3443

Journal of the American Chemical Society
ARTICLE
(28) (a) Feng, L. P.; Beedle, C. C.; Wernsdorfer, W.; Koo, C.;
Nakano, M.; Hill, S.; Hendrickson, N. D. Inorg. Chem. 2007, 46, 8126–
8128. (b) Feng, L. P.; Beedle, C. C.; Koo, C.; Wernsdorfer, W.; Nakano,
M.; Hill, S.; Hendrickson, N. D. Inorg. Chem. 2008, 47, 3188–3204.
(29) Glaser, T.; Heidemeier, M.; Krickemeyer, E.; B€ogge, H.;
Stammler, A.; Fr€ohlich, R.; Bill, E.; Schnack, J. Inorg. Chem. 2007, 48,
607–620.
(30) (a) Winpenny, R. E. P. Dalton Trans. 2002, 1–10. (b) Aromi,
G.; Brechin, E. K. Struct. Bonding (Berlin) 2006, 122, 1–67.
(31) Tasiopoulos, A. J.; Perlepes, S. P. Dalton Trans. 2008, 5537–
5555.
(32) (a) Tanase, S.; Aromi, G.; Bouwman, E.; Kooijman, H.; Spek,
A. L.; Reedijk, J. Chem. Commun. 2005, 3147. (b) Viciano-Chumillas,
M.; Tanase, S.; Aromi, G.; Smits, J. M. M.; de Gelder, R.; Solans, X.;
Bouwman, E.; Reedijk, J. Eur. J. Inorg. Chem. 2007, 2635–2640. (c)
Viciano-Chumillas, M.; de Ruiter, G.; Tanase, S.; Smits, J. M. M.; de
Gelder, R.; Mutikainen, I.; Turpeinen, U.; de Jongh, L. J.; Reedijk, J.
Dalton Trans. 2010, 39, 4991–4998.
(33) (a) Klingele, J.; Dechert, S.; Meyer, F. Coord. Chem. Rev. 2009,
253, 2698–2741. (b) Chaudhuri, P.; Kataev, V.; B€uchner, B.; Klauss,
H.-H.; Kersting, B.; Meyer, F. Coord. Chem. Rev. 2009, 253, 2261–2285.
(34) Pleier, A.-K; Glas, H; Grosche, M.; Sirsch, P.; Thiel, W. R.
Synthesis 2001, 55–62.
(35) Liu, W.; Thorp, H. H. Inorg. Chem. 1993, 32, 4102–4105.
(36) Although hydrogen atoms could not be located in the crystal-
lographic analysis of 1 8CH2Cl2, an H-bonding pattern similar to the
3
situation in 2 8CH2Cl2 is likely.
3
(37) Geim, A. K.; Dubonos, S. V.; Lok, J. G. S.; Grigorieva, I. V.;
Maan, J. C.; Hansen, L. T.; Lindelof, P. E. Appl. Phys. Lett. 1997, 71,
2379–2381.
(38) Abragam, A.; Bleaney, B. Electron Paramagnetic Resonance of
Transition Ions; Oxford University Press: London, 1970.
(39) Gatteschi, D.; Sessoli, R.; Villain, J. Molecular Nanomagnets;
Oxford University Press: Oxford, 2006.
(40) Bencini, A.; Gatteschi, D. EPR of Exchange Coupled Systems;
Springer: New York, 1990.
(41) Stoll, S.; Schweiger, A. J. Magn. Reson. 2006, 178, 42–55.
(42) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.;
McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de
Streek, J.; Wood, P. A. J. Appl. Crystallogr. 2008, 41, 466–470.
(43) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7–13.
(44) Janiak, C. J. Chem. Soc., Dalton Trans. 2000, 3885–3896.
(45) For 1D assembly of homometallic MnIII SMMs, see for
example: (a) Lecren, L.; Roubeau, O.; Coulon, C.; Li., Y.-G.; Le Goff,
X. F.; Wernsdorfer, W.; Miyasaka, H.; Clꢀerac, R. J. Am. Chem. Soc. 2005,
127, 17353–17363. (b) Jeon, I.-R.; Ababei, R.; Lecren, L.; Li, Y.-G.;
Wernsdorfer, W.; Roubeau, O.; Mathoniꢁere, C.; Clꢀerac, R. Dalton Trans.
2010, 39, 4744–4746. (c) Haryono, M.; Kalisz, M.; Sibille, R.;
Lescou€ezec, R.; Fave, C.; Trippe-Allard, G.; Li, Y.; Seuleiman, M.;
Rousseliꢁere, H.; Balkhy, A. M.; Lacroix, J.-C.; Journaux, Y. Dalton Trans.
2010, 39, 4751–4756.
(46) Golze, C.; Alfonsov, A.; Klingeler, R.; B€uchner, B.; Kataev, V.;
Mennerich, C.; Klauss, H.-H.; Goiran, M.; Broto, J.-M.; Rakoto, H.;
Demeshko, S.; Leibeling, G.; Meyer, F. Phys. Rev. B 2006, 73, 224403/1–
224403/8.
(47) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112–122.
(48) X-RED; Stoe & Cie GmbH: Darmstadt, Germany, 2002.
3443
dx.doi.org/10.1021/ja108740c |J. Am. Chem. Soc. 2011, 133, 3433–3443