Single-molecule magnets with mesomorphic lamellar ordering

Article abstract of DOI:10.1002/anie.200704460

(Figure Presented) Layering up: The single-molecule magnets (SMMs) [Mn ₁₂O₁₂(O₂CR)₁₆(H₂O) ₄] have been shown to exhibit liquid-crystalline phases. The mesophases can be smectic or cubic depending on the surface functionality imparted by the R group. A local squarelike arrangement of the cluster cores is observed within the smectic layers. This result is a first step towards the self-assembly of SMMs on surfaces by a bottom-up approach.

Full text of DOI:10.1002/anie.200704460

Communications
DOI: 10.1002/anie.200704460
Single-Molecule Magnets
Single-Molecule Magnets with Mesomorphic Lamellar Ordering**
Emmanuel Terazzi, Cyril Bourgogne, Richard Welter, Jean-Louis Gallani, Daniel Guillon,
Guillaume Rogez,* and Bertrand Donnio*
Single-molecule magnets (SMMs) are materials that are able
to retain magnetization at the molecular level below a certain
temperature, known as the blocking temperature.[1,2] These
compounds currently elicit sustained research activity: as
quantum objects they are envisioned as qubits for quantum
computers, and their magnetic properties make them the
ultimate storage bits of a molecular magnetic memory.
Schematically, these hybrid molecules are made from metal
ions bound together by various organic ligands, and numerous
types of compounds have been reported to show SMM
behavior: 4f coordination compounds,[3] polymetallic cages,[4]
and oxometallaclusters.[2,5] Despite its poor stability against
water and temperature, the so-called “Mn12” compound
[Mn12O12(OAc)16(H2O)4] is probably the most studied SMM
for at least two reasons: its synthesis is cheap and easy, and it
long held the record for the highest blocking temperature.[2b]
The use of functional molecules in macroscopic devices
(bottom-up technology) requires that some degree of low-
dimensionality self-organization (1D or 2D) is imparted to
the molecules. It is well known that self-organization can only
take place in a system that has some fluidity during the
process so that positioning errors can be corrected automati-
cally. Liquid crystals are a prime example of molecular
assemblies in which order coexists with fluidity.[6]
SMMs for their eventual incorporation in a functional
nanodevice.
To counterbalance the a priori unfavorable molecular
shape and bulkiness of the Mn12 cluster core and the
important geometrical constraints, we applied the strategies
used to obtain thermotropic mesophases with fullerenes
(fulleromesogens[7]), bulky lanthanidomesogens,[8] octahedral
coordination complexes,[9] or supramolecular objects.[10] The
procedure consists of covalent grafting of either a mesomor-
phic promoter onto the inorganic cluster through a flexible
aliphatic spacer[7,8b,10] or strongly lipophilic ligands[8a,9] to
improve interfaces and areas compatibilities between both
moieties and to enhance microsegregation.
Mesomorphic, dodecanuclear manganese complexes
[Mn12O12(O2CR)16(H2O)4] were thus obtained by the replace-
ment of the 16 LC-inert acetate groups (R = Me) with gallate-
derived moieties (Scheme 1). Depending on the chosen
We describe herein how dedicated design can endow the
Mn12 molecule with liquid-crystalline (LC) properties while
preserving the peculiar magnetic properties of the original
core; a side effect of this added functionality is much
improved thermal stability. The formation of positionally
ordered mesophases is a first step on the route to organizing
Scheme 1. Ligand exchange between [Mn12(OAc)16] and HLi to yield
[Mn12Li16] complexes, and chemical structures of the gallic derivatives
2
HL1, H L, and HL3.
benzoate, 1D (smectic) or 3D (body-centered cubic) organ-
ized mesophases were indeed induced. This is, to the best of
our knowledge, the first example of liquid crystalline metal-
laclusters and as such opens new perspectives in the field of
self-organized and ordered structures of SMMs.[11]
The three dodecanuclear complexes, abbreviated
[Mn12Li16], were synthesized from [Mn12(OAc)16] by ligand
exchange between the acetate groups and the corresponding
gallic acid derivatives with methoxy (HL1), dodecyloxy (HL2),
or cyanobiphenyloxyundecyloxy (HL3) groups (Scheme 1);
[Mn12(OAc)16] was prepared in situ from MnOAc2 and
KMnO4, and the gallic acid derivatives HL2 and HL3 were
prepared by following published procedures.[8,9]
[*] Dr. E. Terazzi, Dr. C. Bourgogne, Dr. J.-L. Gallani, Dr. D. Guillon,
Dr. G. Rogez, Dr. B. Donnio
Institut de Physique et Chimie des MatØriaux de Strasbourg
(IPCMS)
UMR7504 (CNRS-UniversitØ Louis Pasteur Strasbourg)
23 rue du Lœss, BP 43, 67034, Strasbourg cedex 2 (France)
Fax: (+33)388-107-246
E-mail: rogez@ipcms.u-strasbg.fr
bdonnio@ipcms.u-strasbg.fr
Homepage: http://www-ipcms.u-strasbg.fr/
Prof. R. Welter
Laboratoire DECOMET
UMR7177 (CNRS-UniversitØ Louis Pasteur Strasbourg)
4 rue Blaise Pascal, 67070 Strasbourg cedex (France)
The characterization of the final compounds relied mostly
on elemental analysis and IR spectroscopy, as NMR spec-
troscopy was not suitable on account of the three para-
magnetic clusters. The covalent grafting and the complete
substitution of the 16 positions by the benzoates onto the
clusters was confirmed by the disappearance of the stretching
[**] We are grateful to the Swiss National Science Foundation, the
CNRS-ULP, and the National Agency for Research (ANR DENDRI-
MAT) for financial support. We thank N. Beyer for the artwork, D.
Burger for TGA measurements, and Dr. B. Heinrich for discussions.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
490
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 490 –495

Angewandte
Chemie
vibration of the carbonyl groups of the free benzoates HL1 to
HL3; this absence also indicated that no free ligand or acetate
is present (see Figures S1–S3 in the Supporting Information).
Elemental analyses were in agreement with the proposed
structures (see Table S1 in the Supporting Information). Also,
cocrystallised solvent molecules were confirmed by TG
measurements (see Figures S4–S6 in the Supporting
Information).
We were able to obtain the single-crystal X-ray structure
of the model compound [Mn12L116], which allowed us to check
the bonding mode of the benzoate groups on the oxocluster.
¯
The compound crystallizes in the P1 space group with two
formula units in the unit cell,[12] and the structure is similar to
that of previously described Mn12 complexes (see Figure S12
in the Supporting Information).[13] It consists of a roughly
planar disk made up of a central MnIV4O4 cube surrounded by
a ring of eight MnIII centers connected to the cube by eight m3-
O2ꢀ and four k2-m2-carboxylate groups perpendicular to the
plane of the disk (two on each side). Furthermore, the
peripheral MnIII centers are connected to each other by eight
equatorial and four axial k2-m2-carboxylate groups. Two types
of MnIII ions can be distinguished: type I, doubly bridged to
one MnIV center; and type II, singly bridged to two MnIV
centers. Four different isomers of Mn12 have been described in
the literature: 1:1:1:1, 1:2:1:0, 1:1:2:0, and 2:2:0:0 (the
notation indicates the number of water molecules coordi-
nated to type II MnIII ions).[2b] The present [Mn12L116] com-
pound corresponds to the 2:2:0:0 form, whereas the predom-
inant isomer of [Mn12(OAc)16] is the 1:1:1:1 form.[2]
The occurrence of a frequency-dependent out-of-phase ac
susceptibility signal (c’’) is considered a signature for SMM
properties.[2,14] In an ac susceptibility measurement, an out-of-
phase signal (c’’) is expected to appear if the relaxation
frequency of the magnetization of the sample becomes close
to the frequency of the ac field. Moreover, ac magnetic
susceptibility can be used to determine the spin of the ground
state and the effective energy barrier (Ueff) for the relaxation
of magnetization. Figure 1 shows plots of c’T versus T and c’’
versus T at various frequencies. The frequency-dependent
decrease in the c’T versus T plot on lowering the temperature
indicates that the relaxation of the magnetization rate
becomes close to the ac field frequency. Accordingly, a c’’
signal appears at the corresponding temperature, indicating
SMM behavior of the samples.
For the three compounds, the c’T values above 8 K are
equal and constant at all frequencies. Assuming that only the
ground state is populated at this temperature, the value of c’T
can be used to determine the spin of the ground state of the
clusters. For [Mn12L116], the value of c’T at the plateau is
around 51 emuKmolꢀ1, which corresponds to a ground state
of S = 10 and g = 1.93. For [Mn12L216], the value of c’T at the
plateau (ca. 42 emuKmolꢀ1) leads to S = 9 and g = 1.94, and,
for [Mn12L316], to S = 9 and g = 1.91 (c’Tꢁ 41 emuKmolꢀ1).
These values are in accordance with the values determined for
other Mn12 complexes: although most complexes possess a
spin ground state of S = 10, a ground state of S = 9 was
observed with propionate,[13d] benzoate,[13c,15a] and meta-chlor-
obenzoate[15b] ligands. These results, determined from ac
susceptibility measurements, are in accordance with
Figure 1. Temperature dependence of c’T (left) and c’’ (right) for
[Mn12L116] (bottom), [Mn12L216] (middle), and [Mn12L316] (top). c’ and c’’
are the real and imaginary components of the molar magnetic
susceptibility at zero dc field with a field of 3.5 Oe oscillating at 1000
~
!
~
&
*
^
*
( ), 750 (&), 500 ( ), 300 ( ), 100 ( ), 50 ( ), 10 ( ), and 1 Hz ( ).
[Mn12L1
]
presenting
value than [Mn12L216] and [Mn12L316] (see below). Whereas
[Mn12L1
exhibits frequency-dependent out-of-phase
a higher high-field magnetization
16
]
a
16
signal in the region 2–4 K, [Mn12L216] presents two signals,
one in the region 4–7 K, the other in the region 2–4 K, and
[Mn12L316] presents essentially one out-of-phase signal in the
region 4–7 K (a low-temperature peak is actually present but
not clearly visible).
The possible occurrence of two peaks in out-of-phase
susceptibility is not an unusual feature for Mn12 derivatives.
Actually, the parent [Mn12(OAc)16] complex exhibits only one
frequency-dependent out-of-phase signal in the region 4–
7 K,[2c] but it has long been observed that some Mn12
derivatives may present two signals, one in the region 4–
7 K, the other in the region 2–4 K, or only the low-temper-
ature signal.[13d,16,17] The origin of these two relaxation
processes has been attributed to Jahn–Teller isomerism: for
[Mn12(OAc)16], all the Jahn–Teller elongation axes of the
peripheral MnIII centers are oriented perpendicularly to the
plane of the molecule, whereas for systems that present the
low-temperature c’’ peak, it has been shown that one of the
peripheral MnIII ions had an unusual orientation of its Jahn–
Teller elongation axis, almost perpendicular to the elongation
axis of the other MnIII center. This Jahn–Teller isomerism is
responsible for the lowering of the temperature at which the
Angew. Chem. Int. Ed. 2008, 47, 490 –495
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
491

Communications
Table 1: Temperatures and enthalpy and entropy changes of the phase
molecule experiences slow relaxation of the magnetization
3
transitions for MeL3, H L, [Mn12L216], and [Mn12L316].
(blocking temperature), provoking an overall lowering of the
magnetic anisotropy of the molecule (a key parameter for the
slow relaxation process) and favoring the occurrence of
rhombic (transverse) zero-field interactions within the mol-
ecules. This results in an increase of the rate of the reversal of
the magnetization by quantum tunneling.[13d] A closer exami-
nation of the crystal structure of [Mn12L116] shows that one of
the MnIII ions has a Jahn–Teller elongation axis almost
perpendicular to the others (see Figure S12 in the Supporting
Information). In view of the very low disorder of the crystal
structure (R1 = 6.1%), this Jahn–Teller-distorted isomer is the
only one significantly present in the structure, which explains
the occurrence of a unique out-of-phase signal at a lower
temperature than for the parent [Mn12(OAc)16] compound.
For [Mn12L216] and [Mn12L316], for which no structure of the
core is available, the presence of two peaks suggests the
coexistence of different Jahn–Teller isomers within the
sample, as already observed for other Mn12 derivatives.[13d]
For the three complexes, the magnetization relaxation
rate t follows an Arrhenius equation (t = t0 exp(ꢀUeff/kT),
see Figure S13 in the Supporting Information). This behavior
is characteristic for a thermally activated Orbach process in
which Ueff is the effective anisotropy energy barrier, k the
Boltzmann constant, and t0 the preexponential factor. For
[Mn12L216] it was possible to extract the parameters for both
phases, those corresponding to the high-temperature peak
(slow relaxation phase, SR), and those of the low-temperature
peak (fast relaxation phase, FR). The values obtained for the
Compd
MeL3 [b]
Mesophases[a,c]
T [8C]
DH [Jgꢀ1
]
DS [mJgꢀ1 Kꢀ1
]
Cr!SmA
SmA!N
N!I
16.5
85.2
99.6
138.2
141.1
ꢀ11.5
150.0
40.5
2.02
0.24
3.21
–
6.97
0.67
8.61
HL3[b]
Cr!N
–
N!I
78.49[d]
24.06
–
223.21[d]
[Mn12L2
[Mn12L3
]
]
Cr!Cub
Cub!dec
G!Smfmr
Smfmr!dec
91.95
16
–
–
–
–
–
16
150.0
[a] Second heating; [b] Polarized optical microscopy (POM); [c] Cr=
crystal, G=glass, I=isotropic liquid, SmA=smectic A, N=nematic,
Cub=cubic, Smfrm =filled random mesh smectic phase (see text and
reference [26]), dec=decomposition; [d] cumulative values for both
transitions.
ments. At 308C, up to 11 sharp reflections (Figure S9,
Table S2, Supporting Information) were detected in the low-
angle range for which the reciprocal d spacings were in the
ratios
pffiffiffiffiffi pffiffiffiffiffi pffiffiffiffiffi pffiffiffiffiffi pffiffiffiffiffi pffiffiffiffiffi pffiffiffiffiffi pffiffiffiffiffi pffiffiffiffiffi pffiffiffiffiffi pffiffiffiffiffi
10: 12: 14: 16: 20: 22: 38: 40: 44: 48: 54,
pointing to a cubic supramolecular organization [(h2 + k2 +
l2)1/2 = a/dhkl)]; the number of reflections decreased to 4–5
upon increasing temperature. A very intense and broad
diffraction peak centered at around 4.5 Š (hch) was present
over the whole temperature range, corresponding to the
molten alkyl chains and confirming the liquid nature of the
mesophase. Complete space group determination of mobile
three complexes (Table S4 in the Supporting Information) all thermotropic cubic phases is difficult but important. The large
fall within the expected ranges for Mn12 complexes. The shape
of the magnetization versus field curves at 1.8 K for the three
compounds is typical for randomly oriented polycrystalline
Mn12 species (Figure S14 in the Supporting Information).[17b]
The magnetization curves of the three clusters exhibit a
hysteresis loop characteristic of SMM behavior.[2b] The weak
coercive field of 52 mTobserved for [Mn 12L116] is in accord-
ance with the very low TB value (1.6 K), which is slightly
below the temperature of measurement (1.8 K); TB is
arbitrarily defined as the temperature for which t = 100 s.[17a]
For [Mn12L216] (m0HC = 125 mT), only the SR species show
hysteresis behavior as TB (FR) < 1.8 K < TB (SR). The
presence of the FR species leads to narrowing of the
hysteresis loop at zero field measured for the sample.[17b]
Finally, [Mn12L316] shows a rather large coercive field of
320 mT, which is in accordance with its unique and high TB
value of 2.4 K.
initial number of theoretical possibilities (36) can be
decreased by logical analysis of the data, as shown below. In
our standard diffraction experiment, all non-centrosymmetric
groups (groups with Laue classes 23, 432, or ꢀ43m) can be
disregarded on account of Friedelꢀs law,[18] which leaves only
17 cubic space groups to consider. The sequence of the ratios
is not compatible with a face-centered cubic network (F), and,
although it is still compatible with a primitive Bravais lattice
(P), this option is highly improbable owing to the 10 or so
absent authorized reflections and the consequently too large
lattice parameter. The symmetry of the cubic phase is thus
characterized by a body-centered cubic network (I) with a
lattice parameter a = 105.1 Š. The reflections were indexed as
(310), (222), (321), (400), (420), (332), (611/532), (620), 622),
(444), and (633/552/721), and all satisfied the reflection
conditions 0kl:k + l = 2n, hhl:l = 2n, h00h = 2n; only the
¯
¯
symmetry of space groups Im3 and Im3m is theoretically
compatible with this set of reflections. Aggregation into the
highest symmetry is generally admitted, and, accordingly, the
Polarized optical microscopy (POM) and small- and wide-
angle X-ray diffraction (SAXS) experiments show that
[Mn12L116] is solid (crystalline) until it decomposes at
around 2008C, whereas [Mn12L216] and [Mn12L316] are meso-
morphic and thermally stable up to around 1508C, where
decomposition occurs before the clearing point is reached
(Table 1).
[19,20]
¯
Im3m space group is finally retained.
The cubic lattice
parameter is almost temperature-independent (a = 106.1 Š at
T= 908C and a = 108.1 Š at T= 1408C). As deduced from the
estimated molecular volume (Vmol ꢁ 20000–22000 Š3,
1 ꢁ 1.0 g.cmꢀ3), the unit cell of the cubic phase (a3) contains
55–60 molecular clusters.
No birefringent texture could be observed for [Mn12L2
]
16
¯
in the explored temperature range, but only the formation of
large black (optically isotropic) viscous areas on increasing
Thermotropic cubic phases with the Im3m space group are
rare in liquid crystals,[19,20] and their structure not clearly
understood,[21] in contrast to their lyotropic analogues. Con-
sidering first the discontinuous micellar model, as for
¯
temperature above 258C. A cubic mesophase with a Im3m
symmetry was eventually deduced from SAXS measure-
492
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 490 –495

Angewandte
Chemie
supramolecular dendromesogens,[19] two discrete, globular
aggregates (see Figure S15 in the Supporting Information)
with an aggregation number of 27–30 molecules would be
formed and regularly organized in the three spatial directions.
Such an ordered, periodic 3D array is, however, unlikely
owing to weakly cohesive and nondirectional intermolecular
and inter-“aggregate” interactions. Another, better, alterna-
tive for the description of the 3D organization would be to
consider multicontinuous cubic structures made of two or
three interwoven, infinite, interconnected columnar net-
works, each separated by lipophilic films (minimal surfaces)
(see Figure S16 in the Supporting Information).[21] The fact
that (321) and (400) are the dominant reflections in the
reflection is weak relative to that of 002, implying a
modulation of the electronic density within the lamellar
periodicity. This result suggests organization in which the
peripheral mesogenic groups are equally distributed on either
side of the metallic cluster in a compact manner (cylindrical
molecular conformation), resembling the microsegregated
smectic structures formed by mesogenic end-capped den-
drimers.[22] The interlayer periodicity d is almost invariant
with temperature in the range 60–1408C (hdi = 42.15 Š).
From the calculated molecular volume (Vmol ꢁ 34000–
35000 Š3, 1 ꢁ 1.0 gcmꢀ3), the molecular and mesogenic
areas (AM = Vmol/d and am = AM/24, respectively) could be
2
estimated: AM ꢁ 810–830 Š and am ꢁ 33.75–34.6 Š2. T he
2
¯
powder X-ray pattern points preferentially to the Im3m (I)
latter value differs from the cross-sectional area of 22–24 Š
triple interpenetrated network model (as demonstrated in
another study by direct Fourier reconstruction of electron
density[21c]), although, a comprehensive explanation of the
supramolecular organization is not yet possible.
expected for a cyanobiphenyl group arranged normal to the
smectic layer (SmA type). In this case, the sublayers contain-
ing the peripheral groups are either partially interdigitated, as
in the case of the SmAd phase,[23] or the aliphatic spacers and
the cyanobiphenyl groups are tilted with an angle of about 508
but without any correlation in the tilt direction, as in the SmA
phase of de Vries type.[24]
In contrast, [Mn12L316] exhibited a homogeneous and fluid
birefringent optical texture from 408C upwards. The lack of
characteristic defects precluded any visual determination, but
X-ray diffraction scans as a function of temperature allowed
This supramolecular organization is validated by molec-
the identification of the mesophase. Two equidistant sharp ular dynamics simulations. A molecular model of a smectic
reflections were observed in the diffractogram in a ratio of 1:2
bilayer was built with a cell height of 200 Š (larger than the
(42.3 and 21.0 Š, 00l reflections with l = 1, 2), which is
measured value so as not to constrain the thickness artifi-
2
indicative of lamellar order (Figure 2 and Table S3 in the cially) and an area S ꢁ 810 Š containing two superimposed
molecules. As none of the available force fields for the
molecular dynamics studies were able to simulate correctly
the ring shape of the Mn12 cluster, each manganese atom of a
given cluster was restricted to its position relative to the
others, according to the X-ray diffraction structure of com-
pound [Mn12L116]. All non-Mn atoms of the cluster are
allowed to move freely, as were the two molecules in the cell.
An average interlayer periodicity of 41.9 Š was obtained,
which is in excellent agreement with the measured value (see
Figure S19 in the Supporting Information). We also
attempted to determine the nature of the aforementioned
in-plane order with molecular modeling. Positional order in
molecular 2D assemblies is most often hexagonal or square-
like.[25] For hexagonal packing of the Mn12 cores, the average
distance between neighboring magnetic centers would then
1
be 30.5 Š (S = 1= ” aHex ” 3 ), and the intralayer order would
2
=
2
2
be 26.5 Š, which is smaller than the measured d’ value (see
Table S3 in the Supporting Information). However, if a
Figure 2. SAXS profile of the lamellar structures of [Mn12L316].
squarelike pattern is considered, an average core-to-core
pffiffiffi
distance of 28.4 Š ( S) is found, which is in better accordance
with the experimental data (Figure 3). The intrinsic fourfold
molecular cluster symmetry could favor the squarelike
packing. In the fluid phase, both patterns are likely to coexist,
with the square lattice prevailing over the hexagonal one.
Thus, this smectic arrangement can be described as
follows: the two incompatible segments (mesogens and
aliphatic spacers) form alternating layers, with tilt and
interdigitation of the mesogens between successive periods,
and the Mn12 cores are located in the aliphatic sublayers and
arranged into a squarelike planar array. This supramolecular
organization resembles that of the so-called “filled mesh
phases” observed with some facial amphiphiles, in which ionic
clusters (formed by the laterally attached alkali metal
Supporting Information). A broad and diffuse scattering,
corresponding to the short-range order of the molten chains,
was detected at 4.5 Š (hch). Another diffraction signal, at
28.5 Æ 0.5 Š (d’), with a different line shape and not com-
mensurate with the peaks of the lamellar structure was also
present. This peak was assigned to some intralayer order
associated with a specific arrangement of the diffracting
centers, that is, the magnetic cores. The features of the peak
(intensity relative to d002 and full width at half maximum)
imply that this order is short-ranged and that the position of
the diffracting centers is not correlated from layer to layer.
Such an intralayer organization would be sterically induced
by the bulky cyanobiphenyl groups. The intensity of the 001
Angew. Chem. Int. Ed. 2008, 47, 490 –495
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
493

Communications
the axial and equatorial positions) to tailor more precisely the
mesomorphic behavior of these SMMs.[27]
Received: September 27, 2007
Published online: November 28, 2007
Keywords: cluster compounds · liquid crystals ·
.
magnetic properties · metallomesogens · self-assembly
[1] “Single-Molecule Magnets and Related Phenomena”: Struct.
Bonding (Berlin) 2006, 122 (Ed.: R. Winpenny).
[2] a) R. Sessoli, Mol. Cryst. Liq. Cryst. 1995, 274, 145 – 147; b) D.
Gatteschi, R. Sessoli, Angew. Chem. 2003, 115, 278 – 309; Angew.
Chem. Int. Ed. 2003, 42, 268 – 297, and references therein; c) A.
Caneschi, D. Gatteschi, R. Sessoli, J. Am. Chem. Soc. 1991, 113,
5873 – 5874.
[3] a) S. Osa, T. Kido, N. Matsumoto, N. Re, A. Pochaba, J.
Mrozinski, J. Am. Chem. Soc. 2004, 126, 420 – 421; b) C. M.
Zaleski, E. C. Depperman, J. W. Kampf, M. L. Kirk, V. L.
Pecoraro, Angew. Chem. 2004, 116, 4002 – 4004; Angew. Chem.
Int. Ed. 2004, 43, 3912 – 3914.
[4] R. E. P. Winpenny, J. Chem. Soc. Dalton Trans. 2002, 1 – 10.
[5] D. Gatteschi, R. Sessoli, A. Cornia, Chem. Commun. 2000, 725 –
732.
[6] T. Kato, N. Mizoshita, K. Kishimoto, Angew. Chem. 2006, 118,
44 – 74; Angew. Chem. Int. Ed. 2006, 45, 38 – 68.
[7] a) N. Tirelli, F. Cardullo, T. Habicher, U. W. Suter, F. Diederich,
J. Chem. Soc. Perkin Trans. 2 2000, 193 – 198; b) R. Deschenaux,
B. Donnio, D. Guillon, New J. Chem. 2007, 31, 1064 – 1073.
[8] a) E. Terazzi, S. Suarez, S. Torelli, H. Nozary, D. Imbert, O.
Mamula, J.-P. Rivera, E. Guillet, J.-M. BØnech, G. Bernardinelli,
R. Scopelliti, B. Donnio, D. Guillon, J.-C. G. Bünzli, C. Piguet,
Adv. Funct. Mater. 2006, 16, 157 – 168; b) T. Cardinaels, K.
Driesen, T. N. Parac-Vogt, B. Heinrich, C. Bourgogne, D.
Guillon, B. Donnio, K. Binnemans, Chem. Mater. 2005, 17,
6589 – 6598.
Figure 3. Views of the lamellar packing of the cluster [Mn12L316].
Above: Side view of the lamellar packing of the cyanobiphenyl groups
and clusters in the smectic layer. Below: Top view of the in-layer lateral
arrangement of the clusters forming a squarelike lattice.
[9] a) R. W. Date, E. Fernandez Iglesias, K. E. Rowe, J. M. Elliott,
D. W. Bruce, Dalton Trans. 2003, 1914 – 1931; b) F. Camerel, R.
Ziessel, B. Donnio, D. Guillon, New J. Chem. 2006, 30, 135 – 139.
[10] a) I. Aprahamian, T. Yasuda, T. Ikeda, S. Saha, W. R. Dichtel, K.
Isoda, T. Kato, J. F. Stoddart, Angew. Chem. 2007, 119, 4759 –
4763; Angew. Chem. Int. Ed. 2007, 46, 4675 – 4679; b) E. D.
Baranoff, J. Voignier, T. Yasuda, V. Heitz, J.-P. Sauvage, T. Kato,
Angew. Chem. 2007, 119, 4764 – 4767; Angew. Chem. Int. Ed.
2007, 46, 4680 – 4683.
carboxylates) are organized in a hexagonal array within the
smectic phase;[26] the carboxylates are organized in the layers
of the calamitics with or without correlation between the
layers (filled mesh phases with random, 3D rhombohedral
and channelled-layer suborganizations). Here, the Mn12
clusters are squarely packed in the alkyl sublayers, with no
long-range correlation between the layers, and can also
accordingly be described as a filled, random mesh smectic
phase.
Thus, to summarize, we report for the first time the
synthesis and characterization of Mn12 clusters forming fluid
mesophases with 3D (cubic) or 1D (smectic) positional order.
Even though it is short-ranged, the intralayer ordering of the
molecular cores in the smectic phase is nevertheless square-
like. The magnetic properties are preserved upon function-
alization, and the molecules are thermally stable up to 1508C.
Moreover, the main molecular axes of the magnetic cores,
which coincide with the magnetic easy axes below TB (i.e.
2 K), share a common alignment direction in a nematic sense.
It seems reasonable to expect that these qualities might
facilitate the 2D ordering of these clusters on surfaces, which
is the ultimate goal. The self-organization on a surface could
possibly be obtained from a solution process or by thermal
evaporation. We are currently modifying the ligand structure
and the regioselective substitution of two different ligands (at
[11] J. Gómez-Segura, J. Veciana, D. Ruiz-Molina, Chem. Commun.
2007, 3699 – 3707.
[12] Crystal data: C160H176Mn12O96, Mr = 4523.61 gmolꢀ1, triclinic,
¯
space group P1, a = 18.539(2), b = 24.383(5), c = 26.353(5) Š,
a = 86.707(10), b = 85.951(10), g = 71.008(10)8, V= 11228(2) Š3,
Z = 2,
1calcd = 1.338 gcmꢀ3
,
l(MoKa) = 0.71073 Š,
m =
0.714 mmꢀ1, T= 173(2) K, 1.398 < q < 30.058, h,k,l: 0/26, ꢀ31/
34, ꢀ29/26, a total of 52992 reflections collected and 34401
reflections with I > 2s(I), 2565 parameters refined, R1 = 0.0611,
R2 = 0.1028, wR2 = 0.1696, GooF = 1.243. CCDC-651104 con-
tains 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.
[13] a) T. Lis, Acta Crystallogr Sect.
B 1980, 36, 2042 – 2046;
b) S. M. J. Aubin, Z. Sun, I. A. Guzei, A. L. Rheingold, G.
Christou, D. N. Hendrickson, Chem. Commun. 1997, 2239 –
2240; c) R. Sessoli, H.-L. Tsai, A. R. Schake, S. Wang, J. B.
Vincent, K. Folting, D. Gatteschi, G. Christou, D. N. Hendrick-
son, J. Am. Chem. Soc. 1993, 115, 1804 – 1816; d) S. M. J. Aubin,
494
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 490 –495

Angewandte
Chemie
Z. Sun, H. J. Eppley, E. M. Rumberger, I. A. Guzei, K. Folting,
P. K. Gantzel, A. L. Rheingold, G. Christou, D. N. Hendrickson,
Inorg. Chem. 2001, 40, 2127 – 2146.
2001, 7, 4134 – 4141; d) I. Bury, B. Heinrich, C. Bourgogne, D.
Guillon, B. Donnio, Chem. Eur. J. 2006, 12, 8396 – 8413.
[20] a) S. Diele, Curr. Opin. Solid State Mater. Sci. 2002, 6, 333 – 342;
b) S. Kutsumizu, Curr. Opin. Solid State Mater. Sci. 2002, 6, 537 –
543.
[21] a) A. M. Levelut, M. Clerc, Liq. Cryst. 1998, 24, 105 – 115; b) M.
ImpØror-Clerc, Curr. Opin. Colloid Interface Sci. 2005, 9, 370 –
376; c) X. Zeng, G. Ungar, M. ImpØror-Clerc, Nat. Mater. 2005, 4,
562 – 567.
[14] a) M. A. Novak, R. Sessoli, A. Caneschi, D. Gatteschi, J. Magn.
Magn. Mater. 1995, 146, 211 – 213; b) R. Sessoli, D. Gatteschi, A.
Caneschi, M. A. Novak, Nature 1993, 365, 141 – 143.
[15] a) K. Takeda, K. Awaga, T. Inabe,Phys. Rev. B 1998, 57, R11062;
b) J. An, Z.-D. Chen, J. Bian, J.-T. Chen, S.-X. Wang, S. Gao, G.-
X. Xu, Inorg. Chim. Acta 2000, 299, 28 – 34.
[16] D. Ruiz-Molina, P. Gerbier, E. Rumberger, D. B. Amabilino,
I. A. Guzei, K. Folting, J. C. Huffman, A. Rheingold, G.
Christou, J. Veciana, D. N. Hendrickson, J. Mater. Chem. 2002,
12, 1152 – 1161.
[17] a) K. Takeda, K. Awaga, Phys. Rev. B 1997, 56, 14560 – 14565;
b) K. Awaga, Y. Suzuki, H. Hachisuka, K. Takeda, J. Mater.
Chem. 2006, 16, 2516 – 2521.
[18] C. Hammond in The Basics of Crystallography and Diffraction,
2nd ed., IUCr, Oxford Science Publications, Oxford, 2001.
[19] a) C. Tschierske, J. Mater. Chem. 1998, 8, 1485 – 1508; b) D. J. P.
Yeardley, G. Ungar, V. Percec, M. N. Holerca, G. Johansson, J.
Am. Chem. Soc. 2000, 122, 1684 – 1689; c) H. Duan, S. D.
Hudson, G. Ungar, M. N. Holerca, V. Percec, Chem. Eur. J.
[22] B. Donnio, S. Buathong, I. Bury, D. Guillon, Chem. Soc. Rev.
2007, 36, 1495 – 1513.
[23] a) D. Guillon, A. Skoulios, Mol. Cryst. Liq. Cryst. 1983, 91, 341 –
352; b) F. Hardouin, A. M. Levelut, M. F. Achard, G. Sigaud, J.
Chim. Phys. 1983, 80, 53 – 64.
[24] A. De Vries, J. Chem. Phys. 1979, 71, 25 – 31.
[25] C. Tschierske, Chem. Soc. Rev. 2007, DOI: 10.1039/b615517k.
[26] B. Chen, X. B. Zeng, U. Baumeister, S. Diele, G. Ungar, C.
Tschierske, Angew. Chem. 2004, 116, 4721 – 4725; Angew. Chem.
Int. Ed. 2004, 43, 4621 – 4625.
[27] C. Boskovic, M. Pink, J. C. Huffman, D. N. Hendrikson, G.
Christou, J. Am. Chem. Soc. 2001, 123, 9914 – 9915.
Angew. Chem. Int. Ed. 2008, 47, 490 –495
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
495