Inducing single molecule magnetic behavior in a [Co4O 4] cubane via a pronounced solvatomagnetic effect

Article abstract of DOI:10.1021/ic4027656

The pyrazole-based tridentate diol ligand 2-(1-(2-hydroxyethyl)-1H-pyrazol- 3-yl)phenol (H₂L) forms a cubane-type complex [Co₄L ₄(MeOH)₄] (1) that features a {Co₄O ₄} core and four exogenous MeOH ligands. Electrospray ionization mass spectrometry suggests that the MeOH ligands are easily lost, and thermogravimetric analysis evinces a thermally induced release of those methanol molecules from solid material in the temperature range from 380 to 440 K. Desolvation was found to give rise to a pronounced solvatomagnetic effect that causes a switching of the spin ground state of the {Co₄O₄} core from diamagnetic to magnetic. Furthermore, the desolvated naked [Co₄L₄] cube (1*) shows slow relaxation of the magnetization and butterfly-like magnetic hysteresis at 2 K. A comparatively high relaxation barrier U_eff/k_B = 64.4 K and a characteristic relaxation time τ₀ = 3.8 × 10 ^-9 s for 1* have been derived from an Arrhenius plot. These findings thus demonstrate that the emergence of interesting magnetic properties in molecule-based materials can be triggered via a solvatomagnetic process, even for materials that in their solvated form have a diamagnetic (S_T = 0) ground state.

Full text of DOI:10.1021/ic4027656

Article
pubs.acs.org/IC
Inducing Single Molecule Magnetic Behavior in a [Co4O4] Cubane via
a Pronounced Solvatomagnetic Effect
Felix J. Klinke, Animesh Das, Serhiy Demeshko, Sebastian Dechert, and Franc Meyer*
Institute of Inorganic Chemistry, Georg-August-University, Tammannstrasse 4, D-37077 Gottingen, Germany
̈
S
* Supporting Information
ABSTRACT: The pyrazole-based tridentate diol ligand 2-(1-
(2-hydroxyethyl)-1H-pyrazol-3-yl)phenol (H2L) forms a cu-
bane-type complex [Co4L4(MeOH)4] (1) that features a
{Co4O4} core and four exogenous MeOH ligands. Electrospray
ionization mass spectrometry suggests that the MeOH ligands
are easily lost, and thermogravimetric analysis evinces a thermally
induced release of those methanol molecules from solid material in
the temperature range from 380 to 440 K. Desolvation was found
to give rise to a pronounced solvatomagnetic effect that causes a
switching of the spin ground state of the {Co4O4} core from
diamagnetic to magnetic. Furthermore, the desolvated “naked”
[Co4L4] cube (1*) shows slow relaxation of the magnetization and
butterfly-like magnetic hysteresis at 2 K. A comparatively high relaxation barrier Ueff/kB = 64.4 K and a characteristic relaxation time
τ0 = 3.8 × 10−9 s for 1* have been derived from an Arrhenius plot. These findings thus demonstrate that the emergence of interesting
magnetic properties in molecule-based materials can be triggered via a solvatomagnetic process, even for materials that in their solvated
form have a diamagnetic (ST = 0) ground state.
molecules. While some cubane-type cobalt complexes with SMM
properties have been published,9−11 this appears to be the first
report of a molecular system where SMM behavior is triggered via
the solvatomagnetic effect.
INTRODUCTION
■
Molecular magnetism has been a vibrant research field for the
last two decades, since molecule-based materials allow for
combining useful magnetic responses with other properties
characteristic for molecular compounds to generate phenomena
that cannot be achieved with conventional magnets.1 One such
phenomenon, which is of great interest for sensing applications,
is the so-called solvatomagnetic effect (SME), that is, the
switching of magnetic properties by the uptake, exchange, or
release of solvent molecules in molecule-based materials.2,3
Most compounds showing an SME are either coordination
polymers or metal organic frameworks, and often changes in
their structural dimensionality upon solvent influence are
involved.4 Only a few examples are known of compounds showing
SME while remaining truly molecular systems.5−9 Two of these
examples exhibit changes in already-existent single molecule
magnet (SMM) behavior upon release of crystal or weakly
bonded water molecules.5,9 In a previous report we introduced the
tetranuclear cubane-type complex [Ni4L4(MeOH)4]·H2O that
shows a dramatic SME, namely a switching of the spin ground
state from S = 4 to S = 0 induced by the exchange of coordinated
solvent molecules (MeOH versus H2O).8 In the present work we
now expand this concept to show that (i) our ligand system
2-(1-(2-hydroxyethyl)-1H-pyrazol-3-yl)phenol (H2L) (Scheme 1)
is able to form cubane-type complexes with further transition
metal ions like cobalt(II) and (ii) even more importantly that the
resulting complex [Co4L4(MeOH)4] (1) exhibits switching of the
spin ground state from diamagnetic to magnetic and furthermore
becomes an SMM upon release of the coordinated methanol
RESULTS AND DISCUSSION
■
Synthesis and Structural Characterization of the
Complexes. The ligand H2L was reacted with 1 equiv of
Co(ClO4)2·6H2O and 2 equiv of NEt3 as base in dry methanol
under inert conditions to give complex 1 as a brown precipitate
via a self-assembly process (Scheme 1); complex 1 was then
recrystallized by layering a CH2Cl2 solution of 1 with n-hexane.
Light-brown single crystals suitable for X-ray structure analysis
were obtained in the same way after several days. The
molecular structure of 1, determined by X-ray crystallography,
is shown in Figure 1. Selected interatomic distances and angles
are compiled in the Supporting Material and in Table 1.
Four cobalt atoms and four oxygen atoms constitute the
distorted cubane-type {Co4O4} core, the coordination sphere
of each metal atom being completed by the phenolate O and
pyrazole N from the dianionic ligand, as well as by an
exogenous methanol ligand. Thus all cobalt(II) ions are six-
coordinate in a distorted octahedral coordination environment.
Similar to what was observed in the previously published
[Ni4L4(solv)4] compounds, the four exogenous MeOH ligands
serve as hydrogen bond donors toward the phenolate O.
Received: November 5, 2013
Published: February 27, 2014
© 2014 American Chemical Society
2976
dx.doi.org/10.1021/ic4027656 | Inorg. Chem. 2014, 53, 2976−2982

Inorganic Chemistry
Article
a
Scheme 1. Synthesis and Topology of 1 and 1*
a
For clarity only one ligand L2− and its coordination mode (bold) at the {Co4(μ3-O)4} core are shown.
Figure 2. Emphasis of the cubane-like {Co4O4} fragment and
hydrogen bonding interactions in 1. Blue faces (OF) are not spanned
by hydrogen bonds. Complex 1 (two crystallographically independent
molecules): CoA= Co1/Co11, OA1 = O1/O41, OA2 = O2/O42,
OA3 = O3/O43, CoB = Co2/Co12, OB1 = O11/O51, OB2 = O12/O52,
OB3 = O13/O53, CoC = Co3/Co13, OC1 = O21/O61, OC2 = O22/
O62, OC3 = O23/O63, CoD = Co4/Co14, OD1 = O31/O71,
OD1 = O32/O72, OD3 = O33/O73.
Figure 1. Molecular structure of complex 1 (thermal ellipsoids drawn
at the 30% probability level). Most hydrogen atoms have been
omitted. Only one of the two crystallographically independent
molecules is shown.
Information, Table S3). In particular, bond angles at the
bridging oxygen atoms are significantly more acute along the
faces spanned by O−H···O hydrogen bonds (SF), resulting in
shorter Co···Co distances and deformation of the cubane core
away from ideal cubic symmetry. A detailed discussion of these
structural differences and their relevance for magneto structural
correlations was presented earlier for closely related nickel
cubanes8 and is not part of the present work. However,
structural perturbation is a crucial factor determining the
magnetic exchange interactions through O-atom bridges in
cobalt cubanes as well,12 and this issue will be addressed below
when coupling pathways are discussed.
Thermogravimetric Analysis. In view of the potentially
labile MeOH solvent molecules in complex 1, and since we
were interested in potential solvatomagnetic effects of the
{Co4O4} cubane complex, we investigated the thermal
degradation of solid 1 and accompanying changes in its
composition by thermogravimetric (TGA) and elemental
analysis (EA). TGA revealed that 1 smoothly loses all
coordinated methanol molecules in one step in the temperature
range of 380−440 K. Then a stable plateau occurs that lasts to
700 K (Figure 3). The mass loss of about 10.8% until 440 K
corresponds well with the calculated value for loss of four
MeOH molecules (10.9%). The release of all solvent molecules
from the cobalt cube was also confirmed by EA of powdered
material directly after the thermogravimetric measurements;
Anal. Calcd (%) for C44H40Co4N8O8: C 50.59, H 3.86, N 10.73.
Found: C 50.47, H, 3.91, N 10.65. Interestingly, the
a
Table 1. Co−O−Co Angles (deg) for Complex 1
M−O−M
1
CoA−OA2−CoB
CoA−OC2−CoB
CoA−OC2−CoC
CoA−OD2−CoC
CoA−OA2−CoD
CoA−OD2−CoD
CoB−OB2−CoC
CoB−OC2−CoC
CoB−OA2−CoD
CoB−OB2−CoD
CoC−OB2−CoD
CoC−OD2−CoD
av M−O−M (SF)
av M−O−M (OF)
99.07(13)/98.24(12)
93.90(11)/93.19(12)
97.98(13)/98.11(13)
93.61(11)/93.35(12)
101.06(11)/101.63(14)
101.21(11)/100.82(14)
100.83(13)/100.01(11)
101.94(13)/100.71(12)
93.75(11)/94.15(12)
97.43(13)/99.07(13)
92.49(11)/94.50(12)
98.71(12)/98.14(13)
95.87/96.09
101.26/100.79
a
See Figure 2 for numbering scheme.
Thus four faces of the {Co4O4} cube (denoted as side faces,
SF) are spanned by O−H···O hydrogen bonds, while the two
remaining faces, which are located at opposite sides of the cube
(denoted OF and colored in blue in Figure 2), are not bridged.
This leads to distinct bond lengths and angles for the two dif-
ferent types of cube faces SF and OF (Table 1 and Supporting
2977
dx.doi.org/10.1021/ic4027656 | Inorg. Chem. 2014, 53, 2976−2982

Inorganic Chemistry
Article
Figure 3. Thermogravimetric analysis of 1, scan rate 10 K/min.
temperature range for the loss of all coordinated methanol
molecules in 1 is comparable to the release of guest water
molecules in some recently reported Ni7 and Co7 cluster-based
polymers; however, in the case of molecule-based 1 the process
is more abrupt.13 Differential scanning calorimetry (DSC) measure-
ments performed simultaneously with the TGA revealed a
relatively broad endothermic event beginning right after at the
onset of the mass loss at about 380 K with a maximum at 438 K
and ΔH = 276 kJ mol−1.
The information obtained from TGA was then used for the
preparative synthesis of the “naked” cubane 1* that does not
contain any coordinated solvent molecules by grinding
crystalline material of complex 1 and heating those samples
under vacuum conditions slowly to 480 K (Scheme 1). The
release of all solvent molecules was confirmed by EA, and the
desolvation process was accompanied by a slight color change
from brown (1) to beige (1*). Unfortunately, no structural
information for 1* is available, since desolvation of 1 inevitably
goes along with loss of crystallinity of the material; all attempts
to achieve a single-crystal-to-single-crystal (SCSC) trans-
formation failed. Furthermore it seems unlikely that crystal-
lization of the desolvated complex that lacks any additional
solvent ligands can be achieved via solution methods. However,
desolvated 1* is reproducible prepared via the described method,
namely, by thermal treatment of solid 1, which allowed for
studying its magnetic properties by superconducting quantum
interference device (SQUID) magnetometry (see below).
Figure 4. Positive-ion ESI-MS spectrum of (top) 1 and (bottom) 1*
in MeCN; the upper inset shows a magnification of the experimental
spectrum for 1, and the lower inset shows a magnification of the
experimental spectrum for 1* and simulations of the major peak at
1045 m/z. The protonated complex peak [Co4L4H]+ is overlaid with
the singly oxidized form [Co4L4]+ generated in the mass spectrometer.
ESI Mass Spectrometry. This method was used to gain
further information about the composition and stability of 1
and 1*. If kept in a strictly air- and moisture-free environment,
solid 1* proved to be stable: ESI-MS experiments of MeCN
solutions of 1* showed the intact {Co4O4} cubane core even
after storage for several weeks in the glovebox (Figure 4).
Both the original compound 1 and the desolvated complex
1*, when dissolved in dry MeCN, gave nearly identical ESI-MS
spectra in positive-ion mode. The main peak in the spectra
corresponds to the singly charged [Co4L4H]+ fragment of the
protonated complex without any solvent molecules (m/z =
1045.2 (1 + H), 1045.1 (1* + H)), which is overlaid with the
singly oxidized form [Co4L4]+ generated in the mass spectrometer.
Simulations of the isotopic distribution for both fragments are
included in Figure 4 and show good agreement with the experi-
mental spectrum. These experiments prove not only the stability
and existence of the {Co4L4} core in solution, but also reveal the in
situ formation of 1* from 1 under ESI-MS conditions, namely, the
lability of the exogenous MeOH ligands.
Figure 5. Temperature dependence of χMT for 1 and 1*; solid lines
represent the best simulations according to eq 1, with fit parameters
given in Table 2.
[Co4L4] (1*) in the temperature range from 295 to 2.0 K.
Figure 5 shows the plot of χMT versus temperature and
illustrates the striking differences in the magnetic properties for
cubane-type complexes with and without coordinated solvent
molecules.
For 1 the product χMT decreases only slightly from 295 to
100 K and then drops more steeply to reach a minimum value
close to zero at very low temperatures, indicating an ST = 0
Static Magnetic Properties. Magnetic susceptibility data
were collected for complexes [Co4L4(MeOH)4] (1) and
2978
dx.doi.org/10.1021/ic4027656 | Inorg. Chem. 2014, 53, 2976−2982

Inorganic Chemistry
Article
ground state (Figure 5, blue circles). In contrast, the χMT curve
of 1* increases upon lowering the temperature to reach a
maximum of 17.7 cm3 K mol−1 at 16 K (Figure 5, magenta
circles), evidencing a high-spin ground state for the naked
cube.
According to the structural information available for 1,
experimental magnetic data for both complexes were initially
simulated using a fitting procedure to the appropriate
Heisenberg−Dirac−van Vleck (HDvV) spin Hamiltonian for
two isotropic exchange coupling constants (Figure 6, left) and
reproduced well over the full temperature range when using
two J coupling constants only. We thus assume that upon
release of all MeOH ligands the {Co4O4} core undergoes
structural distortion and reduction in symmetry, meaning that
more complicated coupling schemes are needed to properly
describe the system. Since no detailed solid-state structural
information is available for 1*, an arbitrarily selected model
with three J parameters, based on a reduced Cs symmetric
coupling scheme, and with zero field splitting (Figure 6 right,
eq 2) was applied to achieve a reasonable (though far from
perfect) description over the whole temperature range. The
result of this fit is displayed in Figure 7 and confirms the higher
Figure 6. Magnetic coupling schemes for (left) two and (right) three
independent coupling constants. In the first case, constant J2 marked in
red describes the pathway across faces not spanned by hydrogen bonds
(OF, Figure 2).
Zeeman splitting (eq 1).14 Thereby, J1 represents the interaction
via bridged faces (SF), and J2 represents the interaction via open
faces (OF) (see Synthesis and Structural Characterization of the
Complexes for definitions of SF and OF).
Figure 7. Magnetic susceptibility data for 1* at 0.5 T; the solid line
represents a simulation using best-fit parameters with J1, J2, and J3
according to eq 2; see text and Table 2 for details.
̂
̂
̂
̂
̂
̂
̂
̂
̂
̂
̂
̂
̂
H = −2J (S1S3 + S1S4 + S2S3 + S2S4) − 2J2(S1S2 + S3S4)
1
4
+ gμBB ∑ Szi
degree of distortion in 1*, regardless of the exact type of
distortion.
(1)
i=1
Figure 5 shows the simulations with best-fit parameters
obtained using this model (solid lines); values are given in
Table 2. While the antiferromagnetic coupling constant J2 is
almost the same for both complexes, the ferromagnetic
constant J1 is an order of magnitude higher for 1* than it is
for 1, which translates into the change of the spin ground state
from diamagnetic to magnetic. If classified according to the
shape of the χMT curve, all cobalt(II) cubane complexes can be
grouped into three classes, one of which has the typical
signatures found for 1 with dominant antiferromagnetic
interaction (class B).12 Surprisingly, complexes of this class
are rather rare in literature, and we could find only two
examples15,16 that exhibit the same shape of the χMT curve;
however, no magnetic analysis was presented or magnetic data
were only given for temperatures higher than 100 K. It thus
appears that 1 is the first example of a cobalt(II) type B cubane
that has been magnetically analyzed over a wide temperature
range.
̂
̂
̂
̂
̂
̂
̂
̂
̂
̂
̂
̂
̂
H = −2J (S1S2 + S2S3 + S2S4) − 2J2(S1S3 + S1S4) − 2J3S3S4
1
4
4
2
̂
(Szi − Si(Si + 1)/3)
+ gμBB
∑
Si+D
∑
(2)
i=1
i=4
The fitting parameters illustrate that ferromagnetic as well
as antiferromagnetic couplings are operative in both {Co4O4}
complexes; however, in 1* the ferromagnetic interaction
dominates. Switching of the ground state after release of all
exogenous MeOH ligands can arise (i) from slight perturbation
of the core’s solid-state structure, since the bond angles at the
bridging O atoms have been discussed as a major factor
determining the overall magnetic coupling in the “normal”
cubes12 or (ii) from the lowering of the coordination number
from six to five.17
Dynamic Magnetic Properties. Since 1* has a high-spin
ground state and cobalt(II), having a higher orbital angular
momentum than previously used nickel(II), is a suitable ion to
create SMMs, a more detailed study of the magnetic properties
of 1* was performed. Figure 8a shows the imaginary part
(out-of-phase) of the alternating current (AC) susceptibility
measurement of 1* plotted against temperature, reflecting the
For 1* magnetic data simulation was only accurate in the
temperature range from room temperature to 20 K when using
the model with two J parameters (eq 1). Even after including
zero-field splitting, the experimental data for 1* could not be
Table 2. Best-Fit Parameters of Magnetic Data Analysis
complex
g
J1 (cm−1
)
J2 (cm−1
)
J3 (cm−1
)
|D| (cm−1
)
TIP (10−6 cm3 mol−1
)
1
2.52
2.48
2.48
+0.56
+5.1
−2.84
−3.87
−2.69
0
100
400
400
1*
1*
0
a
+5.70
−1.33
11.76
a
Alternative fit parameters obtained with additional coupling constant J3 and axial zero-field splitting; see Figure 7.
2979
dx.doi.org/10.1021/ic4027656 | Inorg. Chem. 2014, 53, 2976−2982

Inorganic Chemistry
Article
Figure 8. (a) AC susceptibility (HAC = 3 Oe) measurement of 1*, out-of-phase signal without applied DC field. (b) Cole−Cole plot generated from
AC susceptibility measurement of 1* without applied DC field. (c) Arrhenius plot generated from AC susceptibility measurement of 1* without
applied DC field. (d) Magnetic hysteresis of 1* measured at 2 K (right).
frequency-dependent shift of the maximum of χ″ for oscillating
frequencies between 1500 and 10 Hz. The corresponding real
(in-phase) part of the AC susceptibility χ′ is provided in the
Supporting Information (Figure S2).
possible reason for the occurrence of such butterfly-like hysteresis
loops might be the so-called phonon bottleneck effect,21,22
meaning that the hysteresis occurs because of poor thermal
contact of the sample with the heat bath; this has previously been
observed for some spin clusters such as V15,23 Fe12,24 or Fe2.25
However, the maximum of the out-of-phase AC susceptibility χ″
for those complexes is only observed upon application of an
additional direct current (DC) magnetic field of 1 to 10 kOe.22,26
For the present system 1* this is not the case, and the maximum
of the out-of-phase AC susceptibility is evident even in zero DC
field. This likely excludes the possibility of phonon bottleneck
effects being the cause of the observed behavior.
To exclude the possibility that the frequency-dependent AC
susceptibility originates from spin-glass behavior, the shift of
the peak temperature (Tp) of the in-phase part χ′ was analyzed
as Φ = (ΔTp/T)/Δ(log ν),18 where ν is the applied frequency
(in our case between 1500 and 10 Hz). The obtained value for
Φ of 0.15 is 2 orders of magnitude larger than typical values
expected for spin-glass systems (where Φ varies in the range of
0.004−0.018) and lies in the usual range for superparamagnets,
SMMs, or SCMs.19 Additionally, the resulting Cole−Cole plot
in which χ″ is plotted against χ′ for all frequencies can be drawn
at fixed temperatures (Figure 8b). The almost-perfect hemi-
cycles for temperatures between 5 and 6 K indicate that there is
only one active relaxation pathway.
To gain closer insight into the relaxation parameters, the data
were analyzed using an Arrhenuis law for a thermally activated
process: τ = τ0 exp(Ueff/kBT), with the energy barrier Ueff/kB
and the relaxation time τ0.20 Both values can be determined
from a plot of ln(τ) versus (1/T). The parameter τ is calculated
as τ = 1/2πν, the used T values are the peak temperatures
of χ″ at a given frequency. The Arrhenius-type plot for 1*
(Figure 8c) gives a relaxation barrier of Ueff/kB = 64.4 K and
the characteristic relaxation time τ0 = 3.8 × 10−9 s. Few
examples of cobalt cubane complexes with SMM properties are
known to literature. Most of them feature relaxation barriers
between 13 and 30 K,9,10 while some rare examples show
barrier heights of around 60 K.11
CONCLUSIONS
■
In conclusion, thermal treatment of the new [Co4L4(MeOH)4]
cubane complex 1 with ST = 0 triggers switching of the spin
ground state upon release of directly coordinated solvent
molecules, leading to a high-spin ground state for the naked
[Co4L4] (1*). Desolvated 1* shows slow relaxation of the
magnetization with a high relaxation barrier of 64.4 K. These
findings add another important aspect to the chemistry of
compounds with a {M4O4} core, since solvent molecules are
usually bound to such cubane-type complexes. It is now
demonstrated that the SME offers the possibility to trigger the
emergence of interesting magnetic properties via controlled
release of solvent ligands even if the initial ground state of the
cubane complex is diamagnetic.
EXPERIMENTAL SECTION
■
Materials and Methods. Solvents were purified by established
procedures.27 All other chemicals were purchased from commercial
sources and used as received. Microanalyses were performed by the
Analytical Laboratory of the Institute for Inorganic Chemistry at
Since the value for 1* is rather high, it was possible to
observe a butterfly-like hysteresis loop at 2 K even by using a
conventional SQUID magnetometer (Figure 8). The absence of
any coercivity can probably be explained by very fast quantum
tunneling of the magnetization at zero field, as it was also observed
in similar cases for intact cobalt cubane complexes.10b,11a Another
Georg-August-University Gottingen using an Elementar Vario EL III.
̈
1
The H and 13C NMR spectra were recorded with a Bruker Avance
300 MHz spectrometer; chemical shifts for the 1H and 13C spectra are
reported in ppm relative to residual solvent signals of CDCl3 (δ =
2980
dx.doi.org/10.1021/ic4027656 | Inorg. Chem. 2014, 53, 2976−2982

Inorganic Chemistry
Article
7.26 ppm and 77.16 ppm). The IR spectra were recorded using a
Digilab Excalibur Series FTS 3000 spectrometer at room temperature.
Mass spectra were measured using a Bruker HTC Ultra (ESI-MS) or
with a Finnigan MAT 95 (EI-MS). Thermogravimetric measurements
were performed using a Netzsch STA409PC LUXX, scan rate:
10 K/min. Magnetic data were measured with a Quantum-Design
MPMS-XL-5 SQUID magnetometer equipped with a 5 T magnet in
the range from 2 to 295 K. Samples were treated as described for
the individual compounds and fixed in a nonmagnetic sample holder.
Each raw data file for the measured magnetic moment was corrected
for the diamagnetic contribution of the sample holder and the sample.
A Curie-behaved paramagnetic impurity (PI) with spin S = 1 (fixed
to 0.1%) and temperature-independent paramagnetism (TIP) were
included according to χcalc = (1 − PI)χ + PI × χmono + TIP. Before
simulation, the experimental data were corrected for TIP. Full-matrix
diagonalization of exchange coupling and Zeeman splitting was
performed with the julX program.14
Table 3. Crystal Data and Refinement Details for Complex 1
1
empirical formula
formula weight
crystal size [mm3]
crystal system
space group
a [Å]
C48H56Co4N8O12
1172.73
0.27 × 0.23 × 0.10
monoclinic
P21/c
17.5372(7)
34.0208(10)
16.9603(7)
90.00
b [Å]
c [Å]
α [deg]
β [deg]
92.167(3)
90.00
γ [deg]
V [Å3]
10111.8(7)
8
Z
Synthesis of the Ligand 2-(1-(2-Hydroxyethyl)-1H-pyrazol-3-
yl)phenol (H2L). 1-(2-Aryl)-3-dimethylaminoprop-2-enone was syn-
thesized by variation of the known literature procedure28 at ambient
pressure with elongated reaction times (45 min). One equivalent of
the precursor was added to a solution of 2-hydroxyethylhydrazine
(4 equiv) in methanol (100 mL), and the reaction mixture was heated
to reflux for 3 h. The solvent was removed under reduced pressure,
and the residue was transferred into ice water. HCl was added until a
clear solution was obtained, and the solution was then extracted with
dichloromethane. The combined extracts were dried over MgSO4, and
the solvent was removed under reduced pressure. After purification via
column chromatography over silica gel with an ethyl acetate/pentane
(1:1) mixture as eluent the product was obtained as a colorless solid.
A batch of 39 mmol of 1-(2-aryl)-3-dimethylaminoprop-2-enone
ρ [g/cm3]
1.541
F(000)
4832
μ [mm−1
min/Tmax
]
1.359
T
0.6595/0.8188
1.31−25.74
21, −41 to 37, 20
116 377
θ range [deg]
hkl range
measured refl
unique refl [Rint]
observed refl (I > 2σ(I))
data/restraints/param
goodness−of−fit (F2)
R1, wR2 (I > 2σ(I))
R1, wR2 (all data)
resid el dens [e/Å3]
19 035 [0.1003]
11 871
19035/8/1328
1.026
1
0.0604, 0.1016
0.1130, 0.1163
−0.321/0.489
yielded 12.7 mmol, 33%. H NMR (300 MHz, CDCl3, 300 K): δ =
3
3
4.04 (t, JH,H = 5.3 Hz, 2H, CH2), 4.31 (t, JH,H = 5.3 Hz, 2H, CH2),
3
3
6.63 (d, JH,H = 2.5 Hz, 1H, CHPz), 6.91 (dt, JH,H = 7.7, 1.2 Hz, 1H,
3
3
CH), 7.02 (dd, JH,H = 8.3, 1.2 Hz, 1H, CH), 7.21 (dt, JH,H = 8.3,
1.2 Hz, 1H, CH), 7.52 (d, 3JH,H = 2.5 Hz, 1H, CHPz), 7.56 (dd, 3JH,H
=
7.7, 1.2 Hz, 1H, CH). 13C NMR (75 MHz, CDCl3, 300 K): δ = 54.4
(CH2), 61.5 (CH2), 102.4 (CH), 116.8 (Cq), 117.1 (CH), 119.5
(CH), 126.4 (CH), 129.3 (CH), 131.6 (CH), 152.0 (Cq), 156.7 (Cq).
Mp: 91 °C. MS (EI+): m/z (rel. intensity) = 204 (100) [M]+, 173
(85) [M − CH2OH]+.
refined on F2 using all reflections with SHELX-97.29 The non-
hydrogen atoms were refined anisotropically. Most hydrogen atoms
were placed in calculated positions and assigned to an isotropic
displacement parameter of 1.2/1.5 Ueq(C). The positional parameters
of the oxygen-bound hydrogen atoms were refined by using DFIX
restraints (dO−H = 0.82 Å). A fixed isotropic displacement parameter
of 0.08 Å2 was assigned to those hydrogen atoms. Face-indexed
absorption corrections were performed numerically with the program
X-RED.30
Synthesis of [Co4(MeOH)4(L)4] (1). Under inert conditions
Co(ClO4)2·6H2O (180 mg, 0.5 mmol) was added to a solution of
the ligand H2L (100 mg, 0.5 mmol) in dry methanol (20 mL). Then
NEt3 (0.14 mL, 1 mmol) in dry methanol (5 mL) was added dropwise
to the reaction mixture, affording a brown precipitate almost
immediately. The precipitate was collected by filtration, washed with
methanol, and dried in vacuum. The resulting powder was then
dissolved in CH2Cl2 (20 mL) and layered with n-hexane. Brown
crystals of the product [Co4(MeOH)4(L)4] (1) were separated after
several days in the glovebox. MS (ESI+, MeCN): m/z (rel. intensity) =
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental details; additional magnetic data for 1 and 1*;
crystallographic data with bond lengths and angles. This
material is available free of charge via the Internet at http://
pubs.acs.org. CCDC-969653 contains the supplementary
crystallographic data for this Paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
1045.2 (100) [M − 4(MeOH) + H]+. IR (KBr): ν
̃
= 3107 (w), 3063
(w), 2914 (br), 2862 (m), 2790 (br), 2569 (w), 1597 (s), 1555 (m),
1516 (s), 1497 (s), 1466 (s), 1443 (s), 1412 (m), 1370 (w), 1351 (w),
1312 (vs), 1252 (m), 1227 (m), 1201 (m), 1156 (w), 1130 (s), 1086
(m), 1065 (s), 1032 (m), 959 (w), 929 (w), 883 (m), 847 (m), 746
(s), 688 (m), 645 (w), 614 (w), 570 (m), 546 (m), 446 (s) cm−1.
Anal. Calcd (%) for C48H56Co4N8O12: C 49.16, H 4.81, N 9.55.
Found: C 48.82, H, 4.67, N 9.62. After thermal treatment in the TGA
apparatus a sample that was heated to 700 K showed the loss of all
four MeOH ligands (mass loss: exp. 10.8%, calc. 10.9%). Anal. Calcd
(%) for C44H40Co4N8O8: C 50.59, H 3.86, N 10.73. Found: C 50.47,
H, 3.91, N 10.65.
AUTHOR INFORMATION
Corresponding Author
*E-mail: franc.meyer@chemie.uni-goettingen.de. Phone: +49
551 3933012. Fax: +49 551 3933063.
■
X-ray Crystallography. Crystal data and details of the data
collections for 1 are given in Table 3 and in Supporting Information,
Table S1. X-ray data were collected on a STOE IPDS II diffractometer
(graphite monochromated Mo Kα radiation, λ = 0.71073 Å) by use of
ω scans at −20 °C. Measurements at temperatures below −20 °C
proved to be impossible due to disaggregation and loss of crystallinity
of the single crystals. The structure was solved by direct methods and
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Financial support by the Deutsche Forschungsgemeinschaft
(SFB 602, Project A16) is gratefully acknowledged.
■
2981
dx.doi.org/10.1021/ic4027656 | Inorg. Chem. 2014, 53, 2976−2982

Inorganic Chemistry
Article
Seiler, E.; Zeppenfeld, G. Organikum, 19th ed.; Johann Ambrosius
Barth Verlag GmbH: Leipzig, Berlin, Heidelberg, 1993, p 668 ff.
(28) Plier, A.-K.; Glas, H.; Grosche, M.; Sirsch, P.; Thiel, W. R.
Synthesis 2001, 55−62.
(29) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(30) X-RED; STOE & CIE GmbH: Darmstadt, Germany, 2002.
REFERENCES
■
(1) Jeon, I.-R.; Clerac, R. Dalton Trans. 2012, 41, 9565−9686.
(2) Sato, Y.; Ohkoshi, S.; Arai, K.; Tozawa, M.; Hashimoto, K. J. Am.
Chem. Soc. 2003, 125, 14590−14595.
(3) Liu, C.-M.; Zhang, D.-Q.; Zhu, D.-B. Chem. Commun. 2008,
368−370.
(4) Hao, Z.-M.; Zhang, X.-M. Dalton Trans. 2011, 2092−2098 and
references therein.
(5) Song, Y.-M.; Luo, F.; Luo, M.-B.; Liao, Z.-W.; Sun, G.-M.; Tian,
X.-Z.; Zhu, Y.; Yuan, Z.-J.; Liu, S.-J.; Xu, W.-Y.; Feng, X.-F. Chem.
Commun. 2012, 48, 1006−1008.
(6) (a) Motreff, A.; Correa da Costa, R.; Allouchi, H.; Duttine, M.;
Mathonier
5623−5625. (b) Motreff, A.; Correa da Costa, R.; Allouchi, H.;
Duttine, M.; Mathoniere, C.; Duboc, C.; Vincent, J.-M. J. Fluorine
Chem. 2012, 134, 49−55.
̀
e, C.; Duboc, C.; Vincent, J.-M. Inorg. Chem. 2009, 48,
̀
(7) Demeshko, S.; Leibeling, G.; Dechert, S.; Meyer, F. Dalton Trans.
2006, 3458−3465.
(8) Das, A.; Klinke, F. J.; Demeshko, S.; Meyer, S.; Dechert, S.;
Meyer, F. Inorg. Chem. 2012, 51, 8141−8149.
(9) Murrie, M.; Teat, S. J.; Stoeckli-Evans, H.; Gudel, H. U. Angew.
̈
Chem., Int. Ed. 2003, 42, 4653−4656.
(10) (a) Moubaraki, B.; Murray, K. S.; Hudson, T. A.; Robson, R.
Eur. J. Inorg. Chem. 2008, 5425−4529. (b) Galloway, K. W.; Whyte, A.
M.; Wernsdorfer, W.; Sanchez-Benitez, J.; Kamenev, K. V.; Parkin, A.;
Peacock, R. D.; Murrie, M. Inorg. Chem. 2008, 47, 7438−7442.
(c) Guedes, G. P.; Soriano, S.; Comerlato, N. M.; Speziali, N. L.; Lahti,
P. M.; Novak, M. A.; Vaz, M. G. F. Eur. J. Inorg. Chem. 2012, 5642−
5648.
(11) (a) Scheurer, A.; Ako, A. M.; Saalfrank, R. W.; Heinemann, F.
W.; Hampel, F.; Petukov, K.; Gieb, K.; Stocker, M.; Muller, P.
̈
Chem.Eur. J. 2010, 16, 4784−4792. (b) Zhang, S.-H.; Zhang, Y. D.;
Zou, H. H.; Guo, J. J.; Li, H. P.; Song, Y.; Liang, H. Inorg. Chim. Acta
2013, 396, 119−125.
(12) Isele, K.; Gigon, F.; Williams, A. F.; Bernardinelli, G.; Franz, P.;
Decurtins, S. Dalton Trans. 2007, 332−341.
(13) Li, J.; Li, B.; Huang, P.; Shi, H.-Y.; Huang, R.-B.; Zheng, L.-S.;
Tao, J. Inorg. Chem. 2013, 52, 11573−11579.
(14) Bill, E. julX, Magnetic Analysis Program; Max Planck Institute
for Chemical Energy Conversion: Mulheim/Ruhr, Germany, 2008.
̈
(15) Aromí, G.; Batsanov, A. S.; Christian, P.; Helliwell, M.; Roubeau,
O.; Timco, G. A.; Wimpenny, R. E. P. Dalton Trans. 2003, 4466−
4471.
(16) Zhang, K.; Dai, J.; Wang, Y.-H.; Zeng, M.-H.; Kurmoo, M.
Dalton Trans. 2013, 42, 5439−5446.
(17) Volkmer, D.; Hommerich, B.; Griesar, K.; Haase, W.; Krebs, B.
Inorg. Chem. 1996, 35, 3792−3803.
(18) Mydosh, J. A. Spin Glasses: an Experimental Introduction; Taylor
& Francis: London, U.K., 1993.
(19) Kumar, A.; Tandon, R. P.; Awana, V. P. S. J. Appl. Phys. 2011,
110, 043926−43.
(20) Gatteschi, D.; Sessoli, R. Angew. Chem., Int. Ed. 2003, 42, 268−
297.
(21) Van Vleck, J. H. Phys. Rev. 1941, 59, 724−729.
(22) Bindilatti, V.; Vu, T. Q.; Shapira, Y. Solid State Commun. 1991,
77, 423−426.
(23) Chiorescu, I.; Wernsdorfer, W.; Muller, A.; Bogge, H.; Barbara,
̈
̈
B. Phys. Rev. Lett. 2000, 84, 3454−3457.
(24) Inagaki, Y.; Asano, T.; Ajiro, Y.; Narumi, Y.; Kindo, K.; Cornia,
A.; Gatteschi, D. J. Phys. Soc. Jpn. 2003, 72, 1178−1183.
(25) Schenker, R.; Leuenberger, M. N.; Chaboussant, G.; Gudel, H.
̈
U.; Loss, D. Chem. Phys. Lett. 2002, 358, 413−418.
(26) Schenker, R.; Leuenberger, M. N.; Chaboussant, G.; Loss, D.;
Gudel, H. U. Phys. Rev. B: Condens. Matter Mater. Phys. 2005, 72,
̈
184403(10).
(27) Becker, H. G. O.; Berger, W.; Domschke, G.; Fanghanel, E.;
̈
Faust, J.; Fischer, M.; Gentz, F.; Gewald, K.; Gluch, R.; Mayer, R.;
Muller, K.; Pavel, D.; Schmidt, H.; Schollberg, K.; Schwetlick, K.;
̈
2982
dx.doi.org/10.1021/ic4027656 | Inorg. Chem. 2014, 53, 2976−2982