Crystal packing effects on the magnetic slow relaxation of TB(III)-nitronyl nitroxide radical cyclic dinuclear clusters

Article abstract of DOI:10.1021/ic301394x

Four lanthanide-based nitronyl nitroxide radical cyclic molecular clusters of formula [Ln(hfac)₃(NITPhPO(OEt)₂)]₂ (Ln ^III = Gd (1), Tb (2A and 2B), and Dy (3) and NITPhPO(OEt)₂ = 4'-[2-(1-oxyl-3-4,4,5,5-tetramethylimidazoline)phenyl]diethoxylphosphine oxide) have been synthesized. Their X-ray structures have been solved and highlight two different crystal packings. For the particular case of the Tb ^III derivative, both of them can be obtained. In 2A, the molecules are well-isolated, while 2B shows short contacts between N-O radical groups. Static magnetic studies on the Gd^III derivative (1) demonstrate that lanthanides and radicals are ferromagnetically coupled (J = 3.46 ± 0.04 cm^-1). Dynamic magnetic studies show that both compounds 2A and 2B exhibit single molecule magnet behavior. A comparison of their magnetic behaviors highlights that the crystal packing has a crucial influence on the temperature range in which the SMM behavior is observed. In the case of the well-insulated Tb^III-based derivative (2A), the SMM behavior is observed at higher temperatures and lower frequencies than for the one that presents close-packing between the molecules (2B). Comparisons are then possible only under an applied external magnetic field (0.2 T) with Δ = 27.5(6) and 21.0(5) K and τ₀ = 2.64(25) × 10^-9 and 1.76(20) × 10^-9 s for 2A and 2B, respectively.

Full text of DOI:10.1021/ic301394x

Article
pubs.acs.org/IC
Crystal Packing Effects on the Magnetic Slow Relaxation of Tb(III)-
Nitronyl Nitroxide Radical Cyclic Dinuclear Clusters
Fabrice Pointillart,*,†,‡ Kevin Bernot,†,§,∥ Giordano Poneti,†,⊥ and Roberta Sessoli†
†University of Florence, “Ugo Schiff” Chemistry Department & INSTM Research Unit, Via Della Lastruccia 3-13, 50019 Sesto
Fiorentino, Florence, Italy
‡Organomet
́
alliques: Mater
1, 35042, Rennes Cedex, France
§Universite
Europeenne de Bretagne, Rennes, France
́ ́
iaux et Catalyse, UMR 6226 CNRS-UR1 Institut des Sciences Chimiques de Rennes, Universite de Rennes
́
́
∥INSA, SCR, UMR 6226, F-35708 Rennes, France
⊥Universita
̀
degli Studi Guglielmo Marconi, Via Plinio 44, 0193 Rome, Italy
S
* Supporting Information
ABSTRACT: Four lanthanide-based nitronyl nitroxide radical cyclic molecular
clusters of formula [Ln(hfac)3(NITPhPO(OEt)2)]2 (LnIII = Gd (1), Tb (2A and
2B), and Dy (3) and NITPhPO(OEt)2 = 4′-[2-(1-oxyl-3-4,4,5,5-
tetramethylimidazoline)phenyl]diethoxylphosphine oxide) have been synthesized.
Their X-ray structures have been solved and highlight two different crystal
packings. For the particular case of the TbIII derivative, both of them can be
obtained. In 2A, the molecules are well-isolated, while 2B shows short contacts
between N−O radical groups. Static magnetic studies on the GdIII derivative (1)
demonstrate that lanthanides and radicals are ferromagnetically coupled (J = 3.46
0.04 cm−1). Dynamic magnetic studies show that both compounds 2A and 2B
exhibit single molecule magnet behavior. A comparison of their magnetic
behaviors highlights that the crystal packing has a crucial influence on the
temperature range in which the SMM behavior is observed. In the case of the well-
insulated TbIII-based derivative (2A), the SMM behavior is observed at higher
temperatures and lower frequencies than for the one that presents close-packing between the molecules (2B). Comparisons are
then possible only under an applied external magnetic field (0.2 T) with Δ = 27.5(6) and 21.0(5) K and τ0 = 2.64(25) × 10−9
and 1.76(20) × 10−9 s for 2A and 2B, respectively.
compounds playing with the nature of the metal ion. The
modification of the magnetic properties will then be directly
attributed to the variation of the ion, every other parameter
being unchanged. Such a strategy has allowed some of us to put
in evidence the key role of the anisotropy of the metal ion in a
lanthanide based single-chain magnet family.7b Another way is
to use the ligands to tune some parameters of these molecular
systems.8 Coulon et al.9 reported a family of single chain
magnets incorporating different coordinated ligands on the axial
positions and different counteranions. These structural changes
affect then the magnetic properties. Very recently, some of us
investigated the role of intramolecular ferromagnetic exchange
interactions in trinuclear complexes {[Dy(hfac)3]2[M-
(bpca)2]}(CHCl3) (M = FeII, NiII, bpca− = bis(2-
pyridylcarbonyl)amine anion).10 In these complexes, a
diamagnetic substitution of part of the trinuclear units showed
the importance of the noncollinearity of the Ising ions on the
dynamic properties of a SMM. The same approach has been
INTRODUCTION
■
Research on single-molecule magnets (SMMs) has been a very
active field in the past two decades and has been rapidly
expanding since the first magnetically bistable molecule, the
mixed-valent dodecamanganese cluster (Mn12), was discovered
in the early 1990s.1 SMMs exhibit slow relaxation of the
magnetization thanks to a large ground-state spin quantum
number (S), a significant uniaxial magneto-anisotropy (D < 0),
and a good magnetic isolation of the molecule.
Such phenomena can be enhanced if a ferromagnetic
exchange interaction between anisotropic ions like CoII,2
MnIII,1,3 TbIII,4 or DyIII is present.5 To do so, different bridging
ligands have been used, and in particular organic radicals. They
allow the building of complexes that present slow relaxation of
the magnetization.6 One of the most used organic radicals is
without a doubt the stable nitronyl nitroxide radical, but only
few of the published 2p−4f systems highlight a single molecule
magnet or single chain magnet behavior.7
Nowadays, the challenge is to understand the effect of each
of these parameters on the magnetic properties of SMMs. A
promising strategy is to build families of isostructural
Received: June 28, 2012
Published: November 5, 2012
© 2012 American Chemical Society
12218
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
1
diethoxylphosphine as a white solid. Yield: 0.97 g (62%). H NMR
(CDCl3, 293 K, ppm): δ 7.50−7.20 (m, 4H, Ar); 4.51 (s, 1H, NCH);
2.49 (m, 4H, CH2); 1.42 (t, 6H, Me); 1.11 and 1.06 (2s, 12H, Me).
4′-[2-(1-Oxyl-3-4,4,5,5-tetramethylimidazoline)phenyl]-
diethoxylphosphine Oxide (NITPhPO(OEt)2). A solution of NaIO4
(0.44 g, 2.1 mmol) in H2O (30 mL) was added to a suspension of 4′-
[2-(1,3-dihydroxyl-4,4,5,5-tetramethylimidazoline)phenyl]-
diethoxylphosphine (0.5 g, 1.4 mmol) in CH2Cl2 (60 mL). The
reaction mixture was stirred for 15 min, and a deep blue color
appeared in the organic phase, which was isolated, washed with H2O,
dried over MgSO4, and evaporated to dryness. The residue was
purified by chromatography over silica (eluent CH2Cl2/acetone 4:1).
The mean blue band was collected and the solvent removed in vacuo.
NITPhPO(OEt)2 was obtained as a blue crystalline solid by slow
evaporation of a CH2Cl2/hexane (1:1 in volume) solution. Yield: 274
mg (55%). Anal. Calcd (%) for C17H26N2O5P1: C, 55.28; H, 7.05; N,
7.59. Found: C, 55.40; H, 7.19; N, 7.53.
used to enlighten the effect of a weak intramolecular
ferromagnetic coupling and β-diketonate ancillary ligands on
the modulation of the dynamics of the magnetization in the
complex [Dy(hfac)3(NITpPy)]2 (NITpPy = 2-(4-pyridy)-
4,4,5,5-tetramethyl-4,5-dihydro-1H-imidazolyl-3-oxide)7f and
[Ln(Phtfac)3(NITpPy)]2 (LnIII = Gd, Tb, Dy and HPhtfac
=4,4,4-trifluoro-1-phenylbutane-1,3-dione).7h
In this work, we have associated the nitronyl nitroxide
substituted phosphine oxide NITPhOP(OEt)2 to the GdIII (1),
TbIII (2A and 2B), and DyIII (3) ions. The resulting
compounds present an architecture that is similar to [Dy-
(hfac)3(NITpPy)]2, i.e. square molecules featuring the two
lanthanides diagonally opposed. The TbIII derivative crystallizes
in two packing modes. One crystal packing leads to well
isolated molecules (2A), whereas the other leads to short
intermolecular contacts between the nitronyl nitroxide radicals
of different molecular squares (2B).
The structure of the two Tb-based compounds has been
determined by X-ray diffraction on single crystals. The nature
of the intramolecular interaction between the lanthanide and
the radical ligand has been understood thanks to the magnetic
analysis of the isotropic GdIII derivative (1). Finally, the
confrontation of the magnetic properties of the two TbIII
derivatives (2A and 2B) allows to highlight the effect of the
intermolecular exchange interaction on the dynamics of the
magnetization in lanthanide nitronyl-nitroxide based com-
plexes.
[Gd(hfac)3(NITPhPO(OEt)2)]2 (1). A total of 40.7 mg of Gd-
(hfac)3·2H2O (M = 814.3 g·mol−1, n = 0.05 mmol) was dissolved in 15
mL of boiling n-heptane. The solution was kept at 45 °C, while 10 mL
of a CH2Cl2 solution containing 18.5 mg of NITPhPO(OEt)2 (M =
369 g·mol−1, n = 0.05 mmol) was added. The mixture was stirred for
10 min and left without perturbation for slow evaporation. After one
night, dark purple needles of 1, suitable for X-ray diffraction, were
obtained. Yield: 99.8 mg (87%). Anal. Calcd (%) for
C64H58N4O22F36P2Gd2: C, 33.45; H, 2.53; N, 2.44. Found: C, 33.41;
H, 2.59; N, 2.39. IR (KBr): 1652, (s, CO), 1608, 1525 (m), 1507
(m), 1361 (m, N−O), 1356 (m, N−O), 1256 (s, C−F), 1200, 1194 (s,
PO), 1145 (s), 796 (w), 664 (w), 548 (w) cm−1.
[Tb(hfac)3(NITPhPO(OEt)2)]2 (2A and 2B). These two compounds
have been synthesized with the same method described above. After
one night, dark purple needles of 2A·2(CH2Cl2) (yield: 66.6 mg
(58%)) and prisms of 2B (yield: 28.7 mg (25%)) were obtained via
slow evaporation. Both species of single crystals are suitable for X-ray
diffraction. Anal. Calcd (%) for C64H58N4O22F36P2Tb2: C, 33.41; H,
2.52; N, 2.44. Found: C, 33.34; H, 2.61; N, 2.51. IR (KBr): 1649 (s,
CO), 1609, 1529 (m), 1507 (m), 1360 (m, N−O), 1353 (m, N−
O), 1255 (s, C−F), 1199, 1192 (s, PO), 1147 (s), 797 (w), 664
(w), 543 (w) cm−1.
EXPERIMENTAL SECTION
■
Synthesis. General Procedures and Materials. The whole
synthesis of the 4′-[2-(1-oxyl-3-4,4,5,5-tetramethylimidazoline)-
phenyl]diethoxylphosphine oxide ligand (NITPhPO(OEt)2) was
performed under an inert atmosphere. It was adapted from the
synthesis of the 4′-[2-(1-oxyl-3-4,4,5,5-tetramethylimidazoline)-
phenyl]diphenylphosphine oxide ligand.11 The syntheses of 1, 2A,
2B, and 3 were performed under aerobic conditions. The reagents
were purchased from Aldrich and used as received without
purifications. Starting lanthanides salts, Ln(hfac)3·2H2O (LnIII = Gd,
Tb, and Dy) were synthesized according to a literature method.12
(4-Bromophenyl)diethoxylphosphine (P(OEt)2PhBr).13 A solution
of p-dibromobenzene (2.0 g, 8.5 mmol) in THF (6 mL) was added
dropwise over a period of 1 h to a solution of n-BuLi (1.6 M in hexane,
5.3 mL, 8.5 mmol) kept at −70 °C. Afterward, diethyl chlorophosphite
PCl(OEt)2 (1.23 mL, d = 1.082 g mL−1, 8.5 mmol) was slowly added
and the reaction mixture stirred at room temperature for 8 h. After
hydrolysis with H2O (5 mL), the organic phase was extracted with
Et2O, washed with H2O until the aqueous phase reached a pH = 7, and
dried over MgSO4. The solvent was removed in vacuo, leading to
P(OEt)2PhBr as an oil which was used without further purification for
the next step.
[Dy(hfac)3(NITPhPO(OEt)2)]2 (3). This compound has been
synthesized with the same method described above. After one night,
dark prisms of 3, suitable for X-ray diffraction, were obtained via slow
evaporation. Yield: 103.7 mg (90%). Anal. Calcd for
C64H58N4O22F36P2Dy2: C, 33.30; H, 2.51; N, 2.43. Found: C, 33.37;
H, 2.50; N, 2.40. IR (KBr): 1650 (s, CO), 1610, 1527 (m), 1507
(m), 1362 (m, N−O), 1356 (m, N−O), 1255 (s, C−F), 1201, 1194 (s,
PO), 1145 (s), 796 (w), 662 (w), 550 (w) cm−1.
Physical Measurements. Single crystals of 1−3 were mounted on
an APEXII Bruker-AXS diffractometer (Mo Kα radiation source, λ =
0.71073 Å, T = 150(2) K) for data collection, from the Centre de
́ ́
Diffractometrie (CDIFX), Universite de Rennes 1, France. Structures
were solved with a direct method using the SIR-97 program and
refined with a full matrix least-squares method on F2 using the
SHELXL-97 program.14 The X-ray structures reported in this paper
present some disordered pending (ethoxy) and terminal (perfluo-
rated) groups leading to some alerts in the checkcif files. In order to
improve the quality of the crystal structures, some models of disorder
have been used. Thus, for 1, both ethoxy groups are disordered to two
positions. The occupation factors are 0.67 and 0.33 as well as 0.50 and
0.50, respectively, for C14A−C15A and C14B−C14B, and C16A−
C17A and C16B−C17B. The F7 and F9 fluorine atoms are also split
with occupation factors of 0.61 and 0.39. For 2A, the F31 and F33
fluorine atoms are also split with occupation factors of 0.58 and 0.42.
For 2B and 3, the disorder comes from the two whole hfac− anions
which are disordered on two positions with the global occupation
factors of 0.59 and 0.41 as well as 0.59 and 0.41 for 2A and 3,
respectively.
(4-Formylphenyl)diethoxylphosphine (P(OEt)2PhCHO).13a A sol-
ution of P(OEt)2PhBr (1.0 g, 3.6 mmol) in THF (5 mL) was added
dropwise over a period of 1 h to a solution of n-BuLi (1.6 M in hexane,
2.3 mL, 3.6 mmol) kept at −70 °C, followed by a solution of
dimethylformamide DMF (0.3 mL, 3.9 mmol, in slightly excess) in
THF (5 mL). The reaction mixture was then stirred for 7 h at room
temperature. A total of 5 mL of H2O was cautiously added and the
mixture extracted with CH2Cl2. The organic layer was washed with
H2O, dried over MgSO4, and evaporated to dryness to give
P(OEt)2PhCHO as an oil, which was used without further purification
for the next step.
4′-[2-(1,3-Dihydroxyl-4,4,5,5-tetramethylimidazoline)phenyl]-
diethoxylphosphine. A solution of P(OEt)2PhCHO (1.0 g, 4.4 mmol)
and 2,3-dimethyl-2,3-dihydroxyamino-butane (0.67 g, 4.4 mmol) in
MeOH (20 mL) was refluxed for 12 h in a N2 atmosphere. The
resulting precipitate was filtered and washed with MeOH, yielding 4′-
[2-(1,3-dihydroxyl-4,4,5,5-tetramethylimidazoline)phenyl]-
1H NMR was recorded on a Bruker AC 300P spectrometer.
Chemical shifts are reported in parts per million referenced to TMS
1
for H NMR.
12219
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
Figure 1. ORTEP view of the molecular structure of 2A with thermal ellipsoids at 30% probability. Hydrogen atoms and solvent molecules are
omitted for clarity.
Optical spectra were measured using a KBr disk method on a
Perkin-Elmer 1600 Series Fourier transform infrared spectrometer
(resolution 4 cm−1) for IR.
X-Ray Structure Descriptions. [Tb(hfac)3(NITPhPO-
(OEt)2)]2·2(CH2Cl2) (2A)·2(CH2Cl2). A view of the molecular
structure of 2A·2(CH2Cl2) is shown in Figure 1. Selected bond
lengths and angles are reported in Table 2. The asymmetric
unit of 2A·2(CH2Cl2) is made of a Tb(hfac)3 moiety, a
NITPhPO(OEt)2 ligand, and a dichloromethane molecule of
crystallization. These two are bound through the oxygen atoms
of the PO groups. However, the organic radical binds
another Tb(hfac)3 moiety through one of its two N−O groups.
The inversion center and this double bonding of the
NITPhPO(OEt)2 lead to the formation of the cyclic dimer
[Tb(hfac)3(NITPhPO(OEt)2)]2 2A as shown in Figure 1. 2A is
then a molecular rectangle featuring the long edge formed by
the radical−metal bond through the PO group and the
shorter one by the same radical bound to the other ion through
the N−O group. The distance between the TbIII centers along
the diagonal of the rectangle is 9.442(2) Å. Each TbIII ion is
surrounded by eight oxygen atoms coming from three hfac−
ligands, one phosphine oxide group, and one N−O group. The
Tb−O range distance is 2.296(5)−2.417(6) Å. The oxygen O2
atoms of the PO and N−O groups are coordinated in a cis
orientation on the TbIII ion (the O1−Tb1−O2 angle is
81.4(7)°). The coordination polyhedron of the TbIII can be
described as a slightly distorted dodecahedron. The mean
distortion comes from the shorter Tb1−O2(PO) distance
(2.296(5) Å). All the other Tb−O distances are similar
(2.387(6) Å). One can notice that the shorter M−O(PO)
bonds have been previously observed in the [Mn(hfac)2(p-
PhPONIT)]2 molecular rectangle (where p-PhPONIT = 4′-[2-
(1-oxyl-3-4,4,5,5-tetramethylimidazoline)phenyl]-
diphenylphosphine oxide ligand).16 These bond length differ-
ences influence the shape of the coordination polyhedron of the
lanthanide ion. In fact a regular bicapped square face trigonal
prism is characterized by the following four angles: α1 = 0°, α2
= 21.8°, and α3 = α4 = 48.2°.17 The α1, α2, α3, and α4 angles
are described by the angle between the planes (O2O6O7;
O5O6O7), (O1O4O8; O1O3O4), (O4O5O6; O4O5O8), and
(O1O2O7; O1O2O3), respectively. In 2A, the values of these
angles have been found to be equal to 20.5(3)°, 28.4(3)°,
55.8(2)°, and 57.6(2)°. In other terms, the square face of the
regular bicapped square face trigonal prism is folded by an
angle of 21.3° (Figure 2). As a comparison, the tilting between
the phenyl and the imidazole planes is greater in 2A (38.6(3)°)
All magnetic measurements were performed on pellets in order to
avoid orientation of the very anisotropic materials. For 2A and 2B, the
pellets were realized separating single crystals by hand. The dc-
magnetic susceptibility measurements were performed with a
Cryogenic S600 SQUID magnetometer between 2 and 300 K in an
applied magnetic field of 0.05 T for temperatures in the range 2−250
K and 0.1 T for temperatures between 250 and 300 K. These
measurements were all corrected for the diamagnetic contribution as
calculated with Pascal’s constants. The ac-magnetic susceptibility
measurements were performed using a homemade ac probe operating
in the range 100−25000 Hz.15 For 2A in zero-field, only an onset of
SMM behavior is visible in the investigated T range. To limit
overparametrization, the fitting procedures of χ″(ω) plots of 2A in
zero field have been performed using the corresponding in-field
parameters as input values. Then, a 10% evolution of the parameters is
allowed step by step as the temperature is lowered.
RESULTS AND DISCUSSION
■
Synthesis. The three targeted lanthanides are GdIII, TbIII,
and DyIII. This choice has been made because the first ion is
magnetically isotropic and allows easy estimation of the
magnetic interaction (J), while the two others are the most
efficient lanthanides to observe slow magnetic relaxation. All
the compounds were obtained with a classic method which
consists of increasing the Lewis acidity of the metallic center by
the coordination of 1,1,1,5,5,5-hexafluoroacetyacetonate anions
(hfac−), followed by a dehydratation of the Ln(hfac)3·2H2O in
a boiling n-heptane solution. The nitronyl nitroxide substituted
phosphine oxide ligand NITPhPO(OEt)2 is added to the
resulting anhydrous Ln(hfac)3, and after slow crystallization
from a heptane/CH2Cl2 solution, single crystals of [Ln-
(hfac)3(NITPhPO(OEt)2)]2 were obtained. In the case of
LnIII = Gd and Dy, only one form of single crystals was
observed, while for Ln = TbIII, two forms were observed:
needles called 2A (70%) and prisms called 2B (30%). One has
to note that the 2A species appears before (after 12 h of slow
evaporation) the 2B. It has been impossible to find a synthetic
strategy that affords directly pure derivatives. Consequently, the
mother solution was filtered after 12 h to isolate the two
species. Eventual byproducts can be easily removed by hand as
the crystal shapes between 2A·2(CH2Cl2) (needles) and 2B
(prisms) are very different.
12220
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
groups is an intramolecular contact equal to 8.835(10) Å, while
the shortest intermolecular contact is equal to 8.953(11) Å.
Hence, there are no significant intermolecular contacts between
the magnetic rectangles. The shortest intermolecular Tb−Tb
distance is 9.240(9) Å, and each cyclic dimer is well isolated
from the others due to the perfluorated moieties that are
localized in the periphery of the molecule (Figure 3).
[Tb(hfac)3(NITPhPO(OEt)2)]2 (2B). A view of the molecular
structure of 2B is shown in Figure 4. Though 2B is very similar
to 2A from a molecular point of view, significant differences in
the coordination sphere symmetry have been identified. The
characteristic angles take now the following values 16.2(3)°,
18.2(2)°, 37.2(2)°, 37.9(2)° and 16.3(2)°, 16.6(2)°, 37.5(2)°,
39.5(2)° for Tb1 and Tb2 respectively. The two TbIII
environments can be considered similar in 2B but they are
quite different that the one observed in 2A. The crystal packing
is totally different. In 2B, short intermolecular contacts between
the O−N−C−N-O systems takes place (Figure 5) with
distances of about 4 Å that can be compared with the distance
of 10.246(25) Å observed in 2A (Figure 6). The shortest intra-
and intermolecular Tb−Tb distances are similar than 2A with
9.487(28) Å and 10.317(30) Å respectively.
[Gd(hfac)3(NITPhPO(OEt)2)]2 (1) and [Dy(hfac)3(NITPhPO-
(OEt)2)]2 (3). The X-ray structure of 1 (Tables 1 and 2, Figure
S1) shows that this compound is very similar to 2A. The slight
changes of the crystallographic parameters are due to the
dichloromethane molecules of crystallization in 2A which are
not in 1. In any cases, 1 and 2A can be considered magnetically
isolated. Compound 3 (Tables 1 and 2, Figure S2) is
isostructural to 2B.
Figure 2. TbIII polyhedron in 2A with characteristic angles.
than it is in [Mn(hfac)2(p-PhPONIT)]2 (27°),16 in the N-
coordinated NITpPy compounds [Dy(hfac)3(NITpPy)]2
(33.5°)9 and in [Mn(hfac)2(NITpym)]2 (32°) (NITpym = 2-
(5-pyrimidinyl)-4,4,5,5-tetramethyl-4,5-dihydro-1H-imidazo-
line-1-oxyl-3-oxide).18 This difference is due to the different
nature of the metal ion in the first case and to a diverse radical
in the last one. In the [Mn(hfac)2(NITm-Py)]2 (NITm-Py = 2-
(3-pyridyl)-4,4,5,5-tetramethyl-4,5-dihydro-1H-imidazoline-1-
oxyl-3-oxide), the tilting angle is close to the one found in our
complex (38.7° and 43.7°,18 respectively).
When analyzing a crystal packing involving nitronyl−
nitroxide derivatives, one has to closely look to the NO
groups. In fact, because of the strongly delocalized electron
densities along their O−N−C−N−O system, nitronyl−nitro-
xide ligands are able to transmit intermolecular magnetic
interaction even when the distances between the complex seem
to be very large.7g In 2A, the shortest distance between N−O
Figure 3. Crystal packing of 2A showing the arrangement of the well isolated molecules.
12221
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
Figure 4. ORTEP view of the molecular structure of 2B with thermal ellipsoids at 30% probability. Hydrogen atoms are omitted for clarity. Only the
three hfac− ions with the highest occupation factors coordinated to Tb2 are represented.
Figure 5. Crystal packing of 2B showing the short contacts between the neighboring molecular rectangles.
In summary, the reaction between the anhydrous Ln(hfac)3
moieties and the nitronyl nitroxide NITPhPO(OEt)2 gives only
magnetically well isolated cyclic dimers for Ln = GdIII 1, only
nonmagnetically isolated cyclic dimers for Ln = DyIII 3, and a
mix of isolated 2A (about 70%) and nonisolated 2B (about
30%) cyclic dimers for Ln = TbIII that can be easily separated
and analyzed as pure derivatives.
isolated GdIII (S = 7/2) and two radicals (S = 1/2) (χMT = 16.5
cm3 K mol−1). χMT remains almost constant as T is lowered to
ca. 100 K. Below this temperature χMT increases to reach 19.95
cm3 K mol−1 at 2.5 K. This indicates that clear ferromagnetic
interactions are present in the compound. Compound 1 thus
has an S = 4 ground state and an S = 3 excited state, as seen on
many Gd-NIT-R derivatives.19 Studies on similar rectangular
complexes20,21 suggest that these interactions can occur
through different pathways. A strong ferromagnetic interaction
is expected between the radical and the directly bound GdIII ion
(JGd‑Rad). A smaller one can be transmitted through the whole
ligand to reach the GdIII that is on the shorter edge of the
rectangle (J′Gd‑Rad). For example, on the smaller molecular
rectangle of formula [Gd(hfac)3(NITpPy)]2,20 the strongest
It is expected that the 4fn lanthanides with n ≤ 7 crystallize in
same monoclinic phase as 1, while the 4fn lanthanides with n ≥
9 give the same X-ray structure as 3.
Static Magnetic Properties. [Gd(hfac)3(NITPhPO(OEt)2)]2
(1). The thermal variation of the χMT product is depicted in
Figure 7. At room temperature, χMT takes a value of 16.6 cm3 K
mol−1, which is slightly higher than the expected value for two
12222
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
the NITPhPO(OEt)2 ligand avoids the transmission of any
magnetic exchange interaction (J′Gd‑Rad) between the radical
and the opposite GdIII ion. Hence, from the magnetic point of
view, the molecular rectangle 1 can be described as two isolated
Gd(III)−radical pairs. Using H = −JSGd·Srad as the interaction
Hamiltonian provides a septet−nonet energy gap of 4J and a
theoretical expression for χMT like χMT = (4Nβ2g2/k)((30 + 14
exp(−4J/kT))/(9 + 7 exp(−4J/kT)),22 where N is the
Avogadro constant, β is Bohr’s magneton, k is Boltzmann’s
́
constant, and g is the Lande factor of both radicals and metal
centers. The best fit R = 0.995 is shown in Figure 7, and JGd‑Rad
is found to be equal to 3.46 0.04 cm−1 with gGd = gRad = 2.00.
It can be noted that no reliable fit has been obtained
considering an extra J′Gd‑Rad, confirming that the Gd−radical
pairs are isolated. In 1, JGd‑Rad is stronger than in [Gd-
(hfac)3(NITpPy)]2. This may be due to the different nature of
the radical ligand leading to different Gd−O1−N1 angles
(138.0° for [Gd(hfac)3(NITpPy)]2 versus 131.0° for our
compound). This variation could lead to a change of the
overlap of the magnetic π* orbital of the radical with the empty
6s orbital of the GdIII ion, transferring there some unpaired spin
density and polarizing the spins of the seven electrons in the 4f
orbitals.23 The experimental and calculated magnetization of 1
is depicted in the inset of Figure 7. The calculated
magnetization is obtained from a classical Brillouin function
for two uncorrelated S = 7/2 and S = 1/2. The value of the
magnetization is 15 μB at 6 T and is close to the expected
saturated value of 16 μB for two GdIII (S = 7/2) ions and two
radical ligands (S = 1/2). For any field, the calculated
magnetization is lower than the experimental one. This
confirms the ferromagnetic exchange interaction between the
GdIII ion and the nitronyl nitroxide radical. A second molecular
rectangle of formula [Mn(hfac)2(pPONIT)]2 reported by Kahn
et al.16 can be considered to perform a good comparison on the
J′ interaction. In this compound, a very weak J′ antiferro-
magnetic exchange interaction is reported with a magnetic
Figure 6. Shortest intermolecular distances between two N−O radicals
of neighboring molecules.
interaction is J = 1.78 cm−1 and the smallest (through the
nitrogen atom of the pyridine ring) is J′ = 0.17 cm−1. This leads
to a value of χMT at low temperatures (1.6 K) higher than the
maximum value for two isolated S = 4 systems. On 1, the
maximum value of χMT at 3 K corresponds to the expected
value for two isolated Gd-radical pairs (20 cm3 K mol−1), and
no extra J′ interaction is expected. In fact, the large length of
Table 1. Crystallographic Data for [Gd(hfac)3(NITPhPO(OEt)2)]2 (1), [Tb(hfac)3(NITPhPO(OEt)2)]2 (2A and 2B), and
[Dy(hfac)3(NITPhPO(OEt)2)]2 (3)
compound
formula
M/g mol−1
1
2A·2CH2Cl2
2B
3
C64H58N4O22F36P2Gd2
2295.7
C66H62N4Cl4O22F36P2Tb2
2469.0
C64H58N4O22F36P2Tb2
2299.0
C64H58N4O22F36P2Dy2
2306.2
cryst syst
monoclinic
P21/n (No. 14)
a = 12.1997(10) Å
b = 16.2486(17) Å
c = 22.6444(24) Å
α = 90°
β = 99.3007(51)°
γ = 90°
4429.7(8)
monoclinic
P21/n (No. 14)
a = 12.1688(9) Å
b = 22.7303(17) Å
c = 16.4500(12) Å,
α = 90°
β = 90.6503(31)°
γ = 90°
4549.8(6)
triclinic
triclinic
space group
cell parameters
P1(No. 2)
̅
P1 (No. 2)
̅
a = 12.6405(8) Å
b = 18.8158(12) Å
c = 20.4054(13) Å,
α = 102.9512(23)°
β = 104.6762(22)°
γ = 105.1194(22)°
4308.0(5)
a = 12.6338(3) Å
b = 18.8107(5) Å
c = 20.4124(5) Å,
α = 103.0003(12)°
β = 104.4781(12)°
γ = 105.2316(12)°
4307.1(3)
volume/Å3
cell formula units
T/K
Z = 4
Z = 4
Z = 2
Z = 2
150 (2)
150 (2)
150 (2)
150 (2)
diffraction reflection
ρcalcd, Mg/m3
μ, mm−1
5.92° ≤ 2θ ≤ 54.96°
1.721
3.06° ≤ 2θ ≤ 55.28°
1.802
2.18° ≤ 2θ ≤ 55.20°
1.772
4.50° ≤ 2θ ≤ 54.84°
1.778
1.663
1.836
1.812
1.906
number of reflns
independent reflns
Fo > 4σ(Fo)
36001
33489
66183
66998
10069
10372
19387
19219
6819
6877
14163
14869
number of variables
Rint, R1, wR2
599
620
1236
1250
0.0967, 0.1208, 0.2919
0.0810, 0.0658, 0.1770
0.0409, 0.0546, 0.1351
0.0299, 0.0429, 0.1141
12223
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
Figure 7. Temperature dependence of the χMT product for 1
(triangles) and best fit (full black line) in 2.5−300 K temperature
range measured with 0.1 T dc field. In the inset, experimental
(triangle) and calculated for uncorrelated spin system (full black line)
magnetization of 1, at 2.5 K in 0−6 T field range.
exchange interaction of −0.03 cm−1. The origin of this
interaction could not be unambiguously determined from the
data. It can be either of an intramolecular nature (mediated by
the ligand via the PO bond) or intermolecular. Anyway, it is
well-known that in the case of 4f ions the magnetic 2p−4f
interaction is weaker than with 3d ions; hence, our assumption
of a negligible J′ exchange interaction seems to be pertinent.
[(Tb(hfac)3(NITPhPO(OEt)2)]2 (2A and 2B). The temper-
ature dependence of the molar magnetic susceptibilities χM of
2A and 2B are shown in Figure 8 as a plot of χMT vs T. At 300
Figure 8. Temperature dependence of χMT for 2A (circles) and 2B
(squares) measured with a 0.05 T dc field from 2.5 to 250 K and 0.1 T
above. Inset: experimental magnetization of 2A (circles) and 2B
(squares) at 2.5 K in the 0−6 T field range.
K, the χMT product takes a value of 25.3 cm3 K mol−1 for 2A
and 25.4 cm3 K mol−1 for 2B. These values are slightly higher
than the expected value of 24.4 cm3 K mol−1 for two TbIII ions
and two nitronyl nitroxide radicals (S = 1/2, g = 2.00) taken as
isolated. Both complexes show a decrease of χMT on lowering
the temperature, and at 30 K a slope change is observed. For
2A, in the 30−9 K temperature range, the χMT product takes a
quasi constant value of 22.2 cm3 K mol−1, whereas for 2B, it
decreases slowly from 20.5 cm3 K mol−1 to 19.9 cm3 K mol−1.
Below 9 K, the χMT product of both curves decreases abruptly
to reach 21.0 cm3 K mol−1 and 18.3 cm3 K mol−1 at 2.5 K for
2A and 2B, respectively. The experimental magnetization of 2A
and 2B is shown in the inset of Figure 8. At 6 T, the
magnetization values of 2A and 2B are 11.3 μB and 10.3 μB,
respectively. These are lower than the expected saturated value
of 20 μB for two TbIII ions and two radicals.
The ground state of a TbIII ion (electronic configuration 4f 8)
is 7F6 characterized by gJ = 3/2.19 The first excited state (7F5) is
12224
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
separated by 2000 cm−1 from the ground state.22 When the
temperature decreases, the χMT (T) curve decreases because of
the progressive depopulation of the J = 6 multiplet of the TbIII
even in the absence of any exchange interaction.24 From 300 K
to 30 K, the decrease of the χMT product for both dinuclear
complexes is therefore principally attributed to the depopu-
lation of the MJ states of the two TbIII ions. This depopulation
strongly depends on the coordination sphere symmetry, and so,
we may attribute the difference of decreasing between 150 and
50 K to the difference of geometry around the TbIII ions in 2A
and 2B (values in synthesis and structure description section).
Below 50 K, this phenomenon is still present but can be
combined with others. For example, the slope change below 50
K can be a consequence of a ferromagnetic exchange
interaction between the radical (S = 1/2) and the TbIII ion
as seen on the GdIII derivative (1). In the same way, below 9 K,
the χMT evolution can be due also to intermolecular
antiferromagnetic exchange interactions if present. This must
be the cause of the low temperature discrepancies between χMT
vs T curves of 2A and 2B as their crystal packing is different. As
stated before, the N−O groups are known to transmit through
space magnetic interaction provided they are not too
far.20,21,25,7g A close examination of the distance and angles
between those N−O groups confirms that this interaction
should be antiferromagnetic in nature.25
separation of the ground and excited states and thus to fast
relaxation and (ii) to a stabilization of MJ states, which does not
permit slow relaxation of the magnetization.
[(Tb(hfac)3(NITPhPO(OEt)2)]2 (2B). The dynamic properties
of the magnetization of the most close-packed form of the Tb
derivative, 2B, have been investigated. In zero field, a very weak
frequency dependence of the susceptibility is observed (Figures
9 and S4), and a reliable extraction of dynamic parameters is
This antiferromagnetic interaction in 2B is confirmed by the
M vs H curves (inset of Figure 8) where for any field, the
experimental magnetization of 2B is lower than the one of 2A.
If one takes into account that the intramolecular ferromagnetic
exchange interactions are equal in both compounds 2A and 2B,
the observation of the magnetization behavior confirms the
existence of intermolecular antiferromagnetic exchange inter-
action in 2B.
[(Dy(hfac)3(NITPhPO(OEt)2)]2 (3). The temperature depend-
ence of the molar magnetic susceptibility χM of 3 is shown in
Figure S3 as a plot of χMT vs T. At 300 K, χMT takes a value of
29.6 cm3 K mol−1. This value is slightly higher than the
expected value of 29.1 cm3 K mol−1 for two DyIII ions and two
nitronyl nitroxide radicals (S = 1/2, g = 2.00). It then decreases
continuously to reach 19.2 cm3 K mol−1 at 2.5 K. The
experimental magnetization of 3 is depicted in the inset of
Figure S3. At 6 T, the magnetization is 16.3 μB, lower than the
expected saturated value of 22 μB for two DyIII ions and two
radicals.
Compound 3 is iso-structural to 2B, but contrary to this one,
no change in the slope of χMT vs T is observed. Moreover, in
the previously reported [Dy(hfac)3(NITpPy)]2 rectangle, a
ferromagnetic exchange interaction has been found through the
N−O group bonded to the dysprosium ion.10 Anyway, in 3, this
supposed ferromagnetic exchange interaction is not observed
because of the addition of three possible phenomena: (1) the
crystal field effect on the DyIII ion, (2) the intermolecular
antiferromagnetic interaction since the compound is isostruc-
tural to 2B, and (3) ferromagnetic exchange interactions that
are supposed to be weaker in 3 than in the Tb derivatives (2A
and 2B).
Dynamic Magnetic Properties. [(Dy(hfac)3(NITPhPO-
(OEt)2)]2 (3). Our dynamic magnetic investigation has focused
first on the DyIII derivative, this lanthanide cation featuring a
great ability to show magnetic slow relaxation.5−7 Surprisingly,
no frequency dependence of the dynamic susceptibility has
been observed. This is possibly a consequence of the
environment of the DyIII ion that can lead (i) to a small
Figure 9. Frequency dependence of χM″ for 2B, measured from 1.8 K
(blue) to 6 K (red) in zero (top) and 0.2 T static field (bottom).
Measurements were taken from 1.8 to 2 K, 2.2 to 4 K, and 4.4 to 6 K
with 0.1, 0.2, and 0.4 K spacing respectively. Lines are guides to the
eye.
impossible. This can be a consequence of zero-field fast
tunnelling5c,26 that is reported to be cancellable by the
application of a small external static field.5g,27
The field dependence of χM″ has been thus measured at 2 K
(Figure S5). Frequency dependence is observed and a 0.2 T
field is then chosen for in-field dynamic measurements. Clear
frequency dependence is now visible from 1.7 to 2.6 K, and the
χM″ vs frequency curves have been fitted according to an
extended Debye model (Table S1). The characteristic dynamic
parameters have been plotted considering an Arrhenius law (τ
= τ0 exp(Δ/T) where τo is the characteristic relaxation time and
Δ is the activation energy for the process. For 2B, extracted
values are τo = (1.76 0.20) × 10−9 s and Δ = 21.0 0.5 K (R
= 0.997; Figure 10). These relaxation rates are comparable with
the values of the literature for tetranuclear compounds Cu2Tb2
(Δ = 21 K, τo = 2.7 × 10−8 s,4b Δ = 4.2 K, τo = 1 × 10−5 s,28
and Δ = 4.2 K, τo = 1 × 10−5 s,7h) and for the dinuclear
compound CuTb (Δ = 28.5 K, τo = 3.8 × 10−5 s).29 The
distribution of the relaxation times can be studied by plotting
χM″ vs χM′ in a so-called Argand plot. The resulting curves can
be fitted using the Debye model with
χ − χS
T
χ(ω) = χS +
1 + (iωτ)1−α
in which χT = χ(ω→0) is the isothermal susceptibility and χs =
χ(ω→∞) is the adiabatic susceptibility, ω being the angular
frequency of the ac field and τ being the relaxation time of the
system at the temperature of the measurement.30 The α
parameter is introduced in the Debye law to characterize the
distribution of the relaxation times:30,31 α = 0 corresponds to a
12225
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
small field (0.05 T) is enough to remove the quantum
tunnelling (QT) and promote a slower mechanism. The
relaxation rate is so slow that it shifts out of our investigation
range (100−25000 Hz) at 2 K (Figure S7). To allow
comparison with 2B, we then chose a field of 0.2 T to
investigate the in-field temperature dependence of χM″. The in-
field χM″ vs frequency curves show that the two relaxation
regimes (thermally and nonthermally activated) are present at
each temperature (Figure S8). At high temperatures, the
thermally activated regime prevails but the curves can be fitted
considering that the QT regime slightly contributes to the
relaxation. Reciprocally, some features of the thermally
activated relaxation regime are present at low temperatures,
whereas the QT is prominent. The relaxation rates (τ) have
thus been extracted considering a double relaxation process,
and the τ's that belong to the nonpredominant regime have
been reported as gray-filled circles in the Arrhenius plot (Figure
12, Table S3). The energy barrier for the high temperature
Figure 10. Arrhenius plots of the relaxation times extracted from the
in-field measurements with the best linear fit (see text for values) for
2B.
unique relaxation time and is thus expected for an ideal SMM.32
On the contrary, α →1 corresponds to an infinity of relaxation
times and is characteristic of spin-glasses.33 For 2B distribution
of the relaxation times, α is almost constant at 0.33 along the
whole relaxing temperature range (1.7−2.6 K; Table S1).
[(Tb(hfac)3(NITPhPO(OEt)2)]2 (2A). Also in this case the in-
phase (χM′) and out-of-phase (χM″) components of the
magnetic susceptibility were measured in zero static field.
The χM″ vs frequency curve shows a complex behavior with a
set of temperature dependent peaks from 5 to 3 K and a
temperature independent one at lower temperatures (Figure 11
Figure 12. Arrhenius plots of the relaxation times extracted from the
zero-field (empty black circles) and in-field (full black circles)
measurements with the best linear fits (see text for values) for 2A.
Gray filled circles stand for the relaxation times extracted from the
residual peaks of the χM″ vs frequency curves. Arrhenius plot of the
relaxation times extracted from in-field (squares) measurements and
its best fit (dashed blue line) for 2B is reported for comparison.
regime is comparable to the previous one with 27.50 0.60 K
and a characteristic relaxation time of τ0 = 2.64 0.25 × 10−9 s.
The tunnelling rate is 20 times slower with 400 5 Hz. The
presence of mixed relaxation processes is expected to impact
the distribution of the relaxation times. The Argand plot clearly
evidences this as the χM″ vs χM′ curves can be fitted with a
single α parameter only in narrow temperature ranges, i.e., at
the extremes of the temperature range where the contribution
of the minor relaxation process is negligible (Figure 13). At 1.8
K, α = 0.38, χS = 0.7 and χT = 5.05 cm3 mol−1 (R2 = 0.9997). At
3 K, α = 0.28, χS = 1.48, and χT = 6.78 cm3 mol−1 (R2 =
0.9995). The mixed relaxation processes are highlighted here by
the high χS value of this second regime, where a significant part
of the system can still relax according to the low temperature
regime (i.e., QT). In between these extremes, a double
relaxation process has to be considered for each temperature,
and so two different α values are used (Figure S9, Table S3).
The dynamic relaxation figures of 2A and 2B are gathered in
Table 3 for comparison. All extracted parameters are similar,
with τo ranging from 1.76 to 3.36 × 10−9 s and Δ from 21 to
27.5 K. The 0.2 T external field reveals for both complexes a
similar magnetic behavior that is thus supposed to be intrinsic
of the complex. Compound 2A has a crystal packing that
Figure 11. Frequency dependence of χM″ for 2A, measured from 1.8 K
(blue) to 6 K (red) in a zero (top) and 0.2 T static field (bottom) with
the two different relaxation mechanisms highlighted with dashed
(frequency independent) and full (frequency dependent) square.
Measurements were undertaken from 1.8 to 2 K, 2.2 to 4 K, and 4.4 to
6 K with 0.1, 0.2, and 0.4 K spacing, respectively. Lines are guides to
the eye.
and S6). For T > 3 K the energy barrier is estimated at 24
1
K for a characteristic relaxation time of τ0 = 3.36 0.40 × 10−9
s. A transition is observed below 3 K as the energy barrier
vanishes and the complex relaxes through an under barrier
mechanism with a tunnelling rate estimated at 7500 20 Hz
(Table S2).
In order to better understand these dynamic features, the
field dependence of χM″ has been measured at 2 K. In the
absence of field, 2A presents a very fast relaxation. However, a
12226
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
interaction has been evaluated to J = 3.46 cm−1 0.04 cm−1.
While compound 3 does not show any out-of-phase signal,
both compounds 2A and 2B exhibit SMM behavior.
Compound 2B shows a thermally activated regime with pre-
exponential factors of 1.76 0.20 × 10−9 s and energy barriers
of 21 K under an applied external magnetic field. In the well-
isolated form (2A), the thermally activated regime is observed
in both zero and 0.2 T applied external magnetic field (τ0 =
3.36 × 10−9 and 2.64 × 10−9 s, and Δ = 24.0 and 27.5 K,
respectively). The good magnetic insulation of the complexes
in 2A allows the observation of the SMM behavior of the
molecular rectangles in an “observation window” centered at
higher temperatures and slower frequencies.
Very recently, the influence of the β-diketonate ancillary
ligands on the dynamic magnetic properties of similar TbIII
based molecular rectangles has been evidenced. An optimiza-
tion of the SMM properties is possible by an appropriate choice
of the crystal that is supposed to influence the ligand-field
strength of the lanthanide and so the SMM behavior of the
complexes.7h We show here that a similar optimization can be
done by carefully insulating the molecular rectangles. The
combination of the two approaches can possibly lead to
advances in the design of high temperature relaxing Tb-radical
based SMMs.
It is worth notice that the number of metal-radical-based
systems is small over the large amount of reported SMMs.
Moreover, in this small class of SMMs, Tb derivatives are found
to be the most efficient as well as when they are combined with
S = 1/2 spins (CuII, noninnocent organic ligands). This is in
contradiction to what is currently observed in lanthanide-based
SMMs that involve nonmagnetic ligands and where DyIII
derivatives always offer the richest magnetic behavior. Efforts
in the design of TbIII associated to S = 1/2 molecules could lead
to better understanding of the relaxation mechanism of
lanthanide-based SMMs and provide new and exciting SMMs.
Figure 13. Argand plot for 2A with best fits for the low (1.8 K) and
the high temperature (3 K) regions.
provides a good insulation of the complexes, and thus slow
relaxation is observed with or without external field with QT at
low temperatures in both cases. The close-packed derivative
(2B) presents intermolecular magnetic interactions that affect
its dynamic magnetic properties. In zero field, the QT is so fast
that it is not measurable with our instrument (i.e., > 25 000
Hz). In 2B, the dimers are organized into a 1D structure, and a
small intermolecular magnetic interaction is expected. How-
ever, this interaction is not strong enough to permit the
creation of a correlation length at low T as observed on
SCMs.34 This explains why, and contrary to what was observed
on other 1D SMM organizations,35 the dimers organized into a
1D structure (2B) relax faster than isolated dimers (2A). With
a 0.2 T field, the SMM behavior of 2B shifts toward lower
temperatures, and only part of it is visible in our investigation
range (T > 1.6 K). The applied field seems to affect only the
dimers, and interplay with the intermolecular magnetic
interactions can be ruled out. In fact, isolated dimers (2A)
that are not expected to present any intermolecular magnetic
interactions are also highly field sensitive.
ASSOCIATED CONTENT
■
The two derivatives show the same intrinsic behavior, but the
poor magnetic insulation of 2B shifts the “observation window”
of the SMM behavior toward highest frequencies and lower
temperatures.
S
* Supporting Information
ORTEP views of 1 (Figure S1) and 3 (Figure S2). Thermal
dependence of the χMT product and first magnetization of 3
(Figure S3). Field and temperature dependences of the out-of
phase χM″ for 2A and 2B (Figures S4−S7). Frequency
dependence at 2.4 K (Figure S8) and Argand diagram (Figure
S9) for 2A. Dynamic parameters for 2A and 2B (Tables S1−
S3), as well as a CIF file. This material is available free of charge
via the Internet at http://pubs.acs.org.
CONCLUSIONS
■
The synthesis and structures of four new cyclic dinuclear
compounds involving nitronyl nitroxide as radical ligands are
reported. The crystal packing of the [Ln(hfac)3(NITPhPO-
(OEt)2)]2 (Ln = GdIII (1), TbIII (2A and 2B), and DyIII (3))
compounds depends on the trend of the lanthanide ionic radius
along the series with a coexistence of both crystal packings in
the case of the TbIII derivative. In 2A, the molecules are well-
isolated, whereas in 2B they interact through short contacts
between the N−O groups. Static magnetic studies show that
lanthanides and radicals are ferromagnetically coupled. In the
case of the isotropic GdIII derivative, the magnetic exchange
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: Fabrice.pointillart@univ-rennes1.fr.
Notes
The authors declare no competing financial interest.
Table 3. Dynamic Parameters Extracted for 2A and 2B
compound
H (T)
τo (s)
Δ (K)
2B
0
n.a.
n.a.
0.2
1.76 0.20 × 10−9
21.0 0.5
2A
τo (s) HT
Δ (K) HT
24
27.5 0.6
τo (s) LT
Δ (K) LT
0
3.36 0.40 × 10−9
2.64 0.25 × 10−9
1
2.1 0.1 × 10−5 (7500 Hz)
4.0 0.1 × 10−4 (400 Hz)
0
0.2
0
12227
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
X.-Q.; Shi, W.; Song, Y. Inorg. Chem. 2009, 48, 3493. (o) Ritchie, C.;
ACKNOWLEDGMENTS
■
Speldrich, M.; Gable, R. W.; Sorace, L.; Kogerler, P.; Boskovic, C.
̈
We acknowledge financial support from Italian MURST (FIRB
and PRIN grants), from the EC through the Human Potential
Program RTN-QUELMOLNA (MRTN-CT-2003-504880),
and from the NE-MAG-MANET (NMP3-CT-2005-515767),
and F.P. thanks German DFG (SPP1137) for his postdoctoral
fellowship. T. Roisnel is acknowledged for assistance in
crystallographic diffraction.
Inorg. Chem. 2011, 50, 7004. (p) Yan, P.-F.; Lin, P.-H.; Habib, F.;
Aharen, T.; Murugesu, M.; Deng, Z.-P.; Li, G.-M.; Sun, W.-B. Inorg.
Chem. 2011, 50, 7059. (q) Pointillart, F.; Bernot, K.; Sessoli, R.;
Gatteschi, D. Inorg. Chem. 2010, 49, 4355. (r) Guo, Y.-N; Xu, G.-F.;
Wernsdorfer, W.; Ungur, L.; Tang, J.; Zhang, H.-J.; Chibotaru, L. F.;
Powell, A. K. J. Am. Chem. Soc. 2011, 133, 11948. (s) Tian, H.; Wang,
M.; Zhao, L.; Guo, Y. N.; Guo, Y.; Tang, J.; Liu, Z. Chem.Eur. J.
2012, 18, 442. (t) Hewitt, I. J.; Lan, Y.; Anson, C. E.; Luzon, J.; Sessoli,
R.; Powell, A. K. Chem. Commun. 2009, 6765. (u) Tian, H.; Zhao, L.;
Guo, Y.-N.; Guo, Y.; Tang, J.; Liu, Z. Chem. Commun. 2012, 48, 708.
(6) (a) Tsukuda, T.; Suzuki, T.; Kaizaki, S. J. Chem. Soc., Dalton
REFERENCES
■
(1) (a) Lis, T. Acta Crystallogr., Sect. B 1980, 36, 2042. (b) Sessoli, R.;
Tsai, H. L.; Schake, A. R.; Wang, S.; Vincent, J. B.; Folting, K.;
Gatteschi, D.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc.
1993, 115, 1804. (c) Christou, G.; Gatteschi, D.; Hendrickson, D. N.;
Sessoli, R. MRS Bull. 2000, 25, 66.
(2) (a) Zhang, Y.-Z.; Wernsdorfer, W.; Pan, F.; Wang, Z.-M.; Gao, S.
Chem. Commun. 2006, 3302. (b) Yang, E. C.; Hendrickson, D. N.;
Wernsdorfer, W.; Nakano, M.; Zaharov, L. N.; Sammer, R. D.;
Rheingold, A. L.; Ledezma-Gairaud, M.; Christou, G. J. Appl. Phys.
̀
Trans. 2002, 1721. (b) Lescop, C.; Luneau, D.; Rey, P.; Bussiere, G.;
Reber, C. Inorg. Chem. 2002, 41, 5566. (c) Kaizaki, S.; Shirotani, D.;
Tsukahara, Y.; Nakata, H. Eur. J. Inorg. Chem. 2005, 3503. (d) Bernot,
K.; Luzon, J.; Bogani, L.; Etienne, M.; Sangregorio, C.; Shanmugam,
M.; Caneschi, A.; Sessoli, R.; Gatteschi, D. J. Am. Chem. Soc. 2009, 131,
5573. (e) Xu, J. X.; Ma, Y.; Liao, D. Z.; Xu, G. F.; Tang, J. K.; Wang,
C.; Zhou, N.; Yan, S. P.; Cheng, P.; Li, L. C. Inorg. Chem. 2009, 48,
8890. (f) Tian, H. X.; Liu, R. N.; Wang, X. L.; Yang, P. P.; Li, Z. X.; Li,
L. C.; Liao, D. Z. Eur. J. Inorg. Chem. 2009, 4498. (g) Zhou, N.; Ma, Y.;
Wang, C.; Xu, G. F.; Tang, J. K.; Xu, J. X.; Yan, S. P.; Cheng, P.; Li, L.
C.; Liao, D. Z. Dalton Trans. 2009, 8489. (h) Wang, X.-L.; Li, L.-C.;
2002, 91, 7382. (c) Murrie, M.; Teat, S. J.; Stoeckli-Evans, H.; Gudel,
̈
H. U. Angew. Chem., Int. Ed. 2003, 42, 4653. (d) Langley, S. J.;
Helliwell, M.; Sessoli, R.; Rosa, P.; Wernsdorfer, W.; Winpenny, R. E.
P. Chem. Commun. 2005, 5029. (e) Chibotaru, L. F.; Ungur, L.;
Aronica, C.; Elmoll, H.; Pilet, G.; Luneau, D. J. Am. Chem. Soc. 2008,
130, 12445.
(3) (a) Stamatatos, T. C.; Abboud, K. A.; Wernsdorfer, W.; Christou,
G. Angew. Chem., Int. Ed. 2007, 46, 884. (b) Ge, C.-H.; Cui, A.-L.; Ni,
Z.-H.; Jiang, Y.-B.; Zang, L.-F.; Ribas, J.; Kou, H.-Z. Inorg. Chem. 2006,
́
Liao, D.-Z. Inorg. Chem. 2010, 49, 4735. (i) Coronado, E.; Gimenez-
Saiz, C.; Recuenco, A.; Tarazon, A.; Romero, F. M.; Camon, A.; Luis,
F. Inorg. Chem. 2011, 50, 7370.
(7) (a) Bogani, L.; Sangregorio, C.; Sessoli, R.; Gatteschi, D. Angew.
Chem., Int. Ed. 2005, 44, 5817. (b) Bernot, K.; Bogani, L.; Caneschi,
A.; Gatteschi, D.; Sessoli, R. J. Am. Chem. Soc. 2006, 128, 7947.
(c) Bernot, K.; Bogani, L.; Sessoli, R.; Gatteschi, D. Inorg. Chim. Acta
2007, 360, 3807. (d) Liu, R.; Ma, Y.; Yang, P.; Song, X.; Xu, G.; Tang,
J.; Li, L.; Liao, D.; Yan, S. Dalton Trans. 2010, 39, 3321. (e) Liu, R.; Li,
L.; Wang, X.; Yang, P.; Wang, C.; Liao, D.; Sutter, J.-P. Chem.
Commun. 2010, 46, 2568. (f) Poneti, G.; Bernot, K.; Bogani, L.;
Caneschi, A.; Sessoli, R.; Wernsdorfer, W.; Gatteschi, D. Chem.
Commun. 2007, 1807. (g) Bernot, K.; Pointillart, F.; Rosa, P.; Etienne,
M.; Sessoli, R.; Gatteschi, D. Chem. Commun. 2010, 6458. (h) Mei, X.-
L.; Liu, R.-N.; Wang, C.; Yang, P.-P.; Li, L.-C.; Liao, D.-Z. Dalton
Trans. 2012, 41, 2904.
45, 4883. (c) Li, Y.; Wernsdorfer, W.; Cler
C. E.; Powell, A. K. Inorg. Chem. 2006, 45, 2376. (d) Lecren, L.;
Wernsdorfer, W.; Li, Y.-G.; Roubeau, O.; Miyasaka, H.; Clerac, R. J.
Am. Chem. Soc. 2005, 127, 11311. (e) Kachi-Terajima, C.; Miyasaka,
H.; Sugiura, K.; Clerac, R.; Nojiri, H. Inorg. Chem. 2006, 45, 4381.
́
ac, R.; Hewitt, I. J.; Anson,
́
́
(f) Milios, C. J.; Vinslava, A.; Whittaker, A. G.; Parsons, S.;
Wernsdorfer, W.; Christou, G.; Perlepes, S. P.; Brechin, E. K. Inorg.
Chem. 2006, 45, 5272. (g) Aubin, S. M. J.; Wemple, M. W.; Adams, D.
M.; Tsai, H.-L.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc.
1996, 118, 7746.
(4) (a) Costes, J.-P.; Dahan, F.; Wernsdorfer, W. Inorg. Chem. 2006,
45, 5. (b) Osa, S.; Kido, T.; Matsumoto, N.; Re, N.; Pochaba, A.;
Mrozinski, J. J. Am. Chem. Soc. 2004, 126, 420. (c) Novitchi, G.;
Costes, J.-P.; Tuchagues, J.-P.; Vendier, L.; Wernsdorfer, W. New J.
Chem. 2008, 32, 197. (d) Costes, J.-P.; Shova, S.; Wernsdorfer, W.
Dalton Trans. 2008, 1843. (e) Long, J.; Chamoreau, L.-M.; Marvaud,
V. Dalton Trans. 2010, 39, 2188.
(8) 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.
́
(9) Miyasaka, H.; Clerac, R.; Mizushima, K.; Suguira, K.; Yamashita,
M.; Wernsdorfer, W.; Coulon, C. Inorg. Chem. 2003, 42, 8203.
(10) Pointillart, F.; Bernot, K.; Sessoli, R.; Gatteschi, D. Chem.Eur.
J. 2007, 13, 1602.
(5) (a) Aronica, C.; Pilet, G.; Chastanet, G.; Wernsdorfer, W.;
Jacquot, J.-F.; Luneau, D. Angew. Chem., Int. Ed. 2006, 45, 4659.
(b) Ishikawa, N.; Sugita, M.; Ishikawa, T.; Koshihara, S.; Kaizu, Y. J.
Am. Chem. Soc. 2003, 125, 8694. (c) Ishikawa, N.; Sugita, M.;
Wernsdorfer, W. J. Am. Chem. Soc. 2005, 127, 3650. (d) Mori, F.;
Nyuri, T.; Ishida, T.; Nogami, T.; Choi, K.-Y.; Nojiri, H. J. Am. Chem.
Soc. 2006, 128, 1440. (e) Jiang, S.-D.; Wang, B.-W.; Su, G.; Wang, Z.-
M.; Gao, S. Angew. Chem., Int. Ed. 2010, 49, 7448. (f) Li, D.-P.; Wang,
T.-W.; Li, C.-H.; Liu, D.-S.; Li, Y.-Z.; You, X.-Z. Chem. Commun. 2010,
46, 2929. (g) Layfield, R. A.; McDouall, J. J. W.; Sulway, S. A.; Tuna,
F.; Collison, D.; Winpenny, R. E. P. Chem.Eur. J. 2010, 16, 4442.
(11) Rancurrel, C.; Sutter, J.-P.; Kahn, O.; Guionneau, P.; Bravic, G.;
Chasseau, D. New J. Chem. 1997, 21, 275.
(12) Richardson, M. F.; Wagner, W. F.; Sands, D. E. J. Inorg. Nucl.
Chem. 1968, 30, 1275.
(13) (a) Rancurel, C.; Heise, H.; Kohler, F.-H.; Schatzschneider, U.;
Rentschler, E.; Vidal-Gancedo, J.; Veciana, J.; Sutter, J.-P. J. Phys.
Chem. A 2004, 108, 5903. (b) Rancurrel, C.; Sutter, J.-P.; Le Hoerff,
T.; Ouahab, L.; Kahn, O. New J. Chem. 1998, 22, 1333. (c) Leznoff, D.
B.; Rancurrel, C.; Sutter, J.-P.; Golhen, S.; Ouahab, L.; Rettig, S. J.;
Kahn, O. Mol. Cryst. Liq. Cryst. 1999, 334, 425.
(14) (a) CrysAlis CDD; CrysAlis RED, version p 171.29.2; Oxford
Diffraction: Oxfordshire, Great Britain, 2000. (b) Sheldrick, G. M.
́
(h) Lin, P.-H.; Burchell, T. J.; Clerac, R.; Murugesu, M. Angew. Chem.,
Int. Ed. 2008, 47, 8848. (i) Xu, G.-F.; Wang, Q. L.; Gamez, P.; Ma, Y.;
Clerac, R.; Tang, J.; Yan, S.-P.; Cheng, P.; Liao, D.-Z. Chem. Commun.
SHELXL-97; University of Gottingen: Gottingen, Germany, 1997.
̈
̈
́
(15) Midollini, S.; Orlandini, A.; Rosa, P.; Sorace, L. Inorg. Chem.
2005, 44, 2060.
2010, 46, 1506. (j) Tang, J. K.; Hewitt, I.; Madhu, N. T.; Chastanet,
G.; Wernsdorfer, W.; Anson, C. E.; Benelli, C.; Sessoli, R.; Powell, A.
K. Angew. Chem., Int. Ed. 2006, 45, 1729. (k) Hewitt, I. J.; Tang, J.;
Mahdu, N. Y.; Anson, C. E.; Lan, Y.; Luzon, J.; Etienne, M.; Sessoli, R.;
Powell, A. K. Angew. Chem., Int. Ed. 2010, 49, 6352. (l) Guo, F.-S.; Liu,
J.-L.; Leng, J.-D.; Meng, Z.-S.; Lin, Z.-J.; Tong, M.-L.; Gao, S.; Ungur,
L.; Chibotaru, L. F. Chem.Eur. J. 2011, 17, 2458. (m) Luzon, J.;
Bernot, K.; Hewitt, I. J.; Anson, C. E.; Powell, A. K.; Sessoli, R. Phys.
Rev. Lett. 2008, 100, 247205. (n) Chen, Z.; Zhao, B.; Cheng, P.; Zhao,
(16) Rancurel, C.; Leznoff, D. B.; Sutter, J.-P.; Golhen, S.; Ouahab,
L.; Kliava, J.; Kahn, O. Inorg. Chem. 1999, 38, 4753.
(17) (a) Muetterties, E. L.; Guggenberg, L. J. J. Am. Chem. Soc. 1974,
96, 1748. (b) Drew, M. G. B. Coord. Chem. Rev. 1977, 24, 179.
(18) Okada, K.; Nagao, O.; Mori, H.; Kozaki, M.; Shiomi, D.; Sato,
K.; Takui, T.; Kitagawa, Y.; Yamaguchi, K. Inorg. Chem. 2003, 42,
3221.
12228
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229

Inorganic Chemistry
Article
(19) Benelli, C.; Gatteschi, D. Chem. Rev. 2002, 102, 2369.
(20) Benelli, C.; Caneschi, A.; Gatteschi, D.; Pardi, L. Inorg. Chem.
1992, 31, 741.
(21) (a) Benelli, C.; Caneschi, A.; Gatteschi, D.; Laugier, J.; Rey, P.
Angew. Chem., Int. Ed. Engl. 1987, 26, 913. (b) Benelli, C.; Caneschi,
A.; Gatteschi, D.; Pardi, L.; Rey, P.; Shum, D. P.; Carlin, R. L. Inorg.
Chem. 1989, 28, 272. (c) Benelli, C.; Caneschi, A.; Gatteschi, D.;
Guillou, O.; Pardi, L.; Rey, P. Inorg. Chim. Acta 1989, 160, 1.
(d) Benelli, C.; Caneschi, A.; Fabretti, A. C.; Gatteschi, D.; Pardi, L.
Inorg. Chem. 1990, 29, 4153. (e) Benelli, C.; Caneschi, A.; Gatteschi,
D.; Pardi, L.; Rey, P. Inorg. Chem. 1990, 29, 4223.
(22) Kahn, O. Molecular Magnetism; VCH: New York, 1993.
(23) Benelli, C.; Caneschi, A.; Gatteschi, D.; pardi, L.; Rey, P. Inorg.
Chem. 1989, 28, 3230.
(24) (a) Sutter, J.-P.; Kahn, M. L. Magnetism: Molecules to Materials;
Wiley-VCH: Weinheim, Germany, 2005. (b) Rinehart, J. D.; Long, J.
R. Chem. Sci 2011, 2, 2078.
(25) Caneschi, A.; Ferraro, F.; Gatteschi, D.; Rey, P.; Sessoli, R. Inorg.
Chem. 1990, 29, 1756. Caneschi, A.; Ferraro, F.; Gatteschi, D.; Rey, P.;
Sessoli, R. Inorg. Chem. 1990, 29, 4217. Fettouhi, M.; Khaled, M.;
Waheed, A.; Golhen, S.; Ouahab, L.; Sutter, J.-P.; Kahn, O. Inorg.
Chem. 1993, 38, 3967.
(26) (a) Pasatoiu, T. D.; Etienne, M.; Madalan, A. M.; Andruh, M.;
Sessoli, R. Dalton Trans. 2010, 39, 4802. (b) Ishikawa, N.; Sugita, M.;
Wernsdorfer, W. Angew. Chem., Int. Ed. 2005, 44, 2931.
(27) (a) Gatteschi, D.; Sessoli, R. Angew. Chem., Int. Ed. 2003, 42,
268. (b) Jiang, S.-D.; Wang, B.-W.; Sun, H.-L.; Wang, Z.-M.; Gao, S. J.
Am. Chem. Soc. 2011, 133, 4730. (c) Rebilly, J. N.; Catala, L.; Guillot,
R.; Wernsdorfer, W.; Mallah, T. Inorg. Chem. 2005, 44, 8194.
̀
(d) Rebilly, J. N.; Catala, L.; Riviere, E.; Guillot, R.; Wernsdorfer, W.;
Mallah, T. Chem. Commun. 2006, 735. (e) Martinez-Lillo, J.;
Armentano, D.; De Munno, G.; Wernsdorfer, W.; Julve, M.; Lloret,
F.; Faus, J. J. Am. Chem. Soc. 2006, 128, 14218. (f) Martinez-Lillo, J.;
Armentano, D.; De Munno, G.; Wernsdorfer, W.; Clemente-Juan, J.
M.; Krystek, J.; Lloret, F.; Julve, M.; Faus, J. Inorg. Chem. 2009, 48,
3027. (g) Ferbinteanu, M.; Kajiwara, T.; Choi, K.-Y.; Nojiri, H.;
Nakamoto, A.; Kojima, N.; Cimpoesu, F.; Fujimura, Y.; Takaishi, S.;
Yamashita, M. J. Am. Chem. Soc. 2006, 128, 9008.
(28) Costes, J.-P.; Auchel, M.; Dahan, F.; Peyrou, V.; Shova, S.;
Wernsdorfer, W. Inorg. Chem. 2006, 45, 1924.
(29) Costes, J.-P.; Clemente-Juan, J. M.; Dahan, F.; Milon, J. Inorg.
Chem. 2004, 43, 8200.
(30) Cole, K. S.; Cole, R. H. J. Chem. Phys. 1941, 9, 341.
(31) (a) Dekker, C.; Arts, A. F. M.; Wijn, H. W.; van Duyneveldt, A.
J.; Mydosh, J. A. Phys. Rev. B 1989, 40, 11243. (b) Aubin, S. M. J.; Sun,
Z.; Pardi, L.; Krzystek, J.; Folting, K.; Brunel, L.-C.; Rheingold, A. L.;
Christou, G.; Hendrickson, D. Inorg. Chem. 1999, 38, 5329.
(32) Gatteschi, D.; Sessoli, S.; Villain, J. Molecular Nanomagnets;
Oxford University Press: Oxford, U. K., 2006.
(33) Mydosh, J. A. Spin Glasses: An Experimental Introduction; Taylor
& Francis: London, 1993.
́
(34) Coulon, C.; Miyasaka, H.; Clerac, R. Single-Molecule Magnets and
Related Phenomena; Winpenny, R., Ed.; Springer-Verlag Berlin: Berlin,
2006.
(35) (a) Ferbinteanu, M.; Miyasaka, H.; Wernsdorfer, W.; Nakata, K.;
Sugiura, K.; Yamashita, M.; Coulon, C.; Cler
2005, 127, 3090. (b) Lecren, L.; Roubeau, O.; Coulon, C.; Li, Y.-G.;
Le Goff, X. F.; Wernsdorfer, W.; Miyasaka, H.; Clerac, R. J. Am. Chem.
Soc. 2005, 127, 17353.
́
ac, R. J. Am. Chem. Soc.
́
12229
dx.doi.org/10.1021/ic301394x | Inorg. Chem. 2012, 51, 12218−12229