Terbium(III) and Yttrium(III) complexes with pyridine-substituted nitronyl nitroxide radical and different β-diketonate ligands. Crystal structures and magnetic and luminescence properties

Article abstract of DOI:10.1021/ic500779y

A terbium(III) complex of nitronyl nitroxide free radical 2-(2-pyridyl)-4,4,5,5-tetramethyl-4,5-dihydro1H-imidazolyl-1-oxy-3-oxide (NIT2Py), [Tb(acac)3NIT2Py]·0.5H₂O (3) (acac = acetylacetonate), was synthesized for comparison with the previously reported [Tb(hfac)3NIT2Py]·0.5C₇H₁₆ (1) (hfac = hexafluoroacetylacetonate), together with their yttrium analogues [Y(hfac)3NIT2Py]·0.5C₇H₁₆ (2) and [Y(acac)3NIT2Py]·0.5H₂O (4). The crystal structures show that in all complexes the nitronyl nitroxide radical acts as a chelating ligand. Magnetic studies show that 3 like 1 exhibits slow relaxation of magnetization at low temperature, suggesting single-molecule magnet behavior. The luminescence spectra show resolved vibronic structure with the main interval decreasing from 1600 cm^-1 to 1400 cm^-1 between 80 and 300 K. This effect is analyzed quantitatively using experimental Raman frequencies.

Full text of DOI:10.1021/ic500779y

Article
pubs.acs.org/IC
Terbium(III) and Yttrium(III) Complexes with Pyridine-Substituted
Nitronyl Nitroxide Radical and Different β‑Diketonate Ligands.
Crystal Structures and Magnetic and Luminescence Properties
Anthony Lannes,† Mourad Intissar,‡ Yan Suffren,‡ Christian Reber,*,‡ and Dominique Luneau*,†
†Laboratoire des Multimater
Claude Bernard Lyon 1, Campus de la Doua, 69622,
Villeurbanne Cedex, France
‡Dep
́ ́ ́ ́
artement de chimie, Universite de Montreal, Montreal, Quebec H3C 3J7, Canada
́
iaux et Interfaces (UMR 5616), Universite
́
́
S
* Supporting Information
ABSTRACT: A terbium(III) complex of nitronyl nitroxide free radical
2-(2-pyridyl)-4,4,5,5-tetramethyl-4,5-dihydro1H-imidazolyl-1-oxy-3-
oxide (NIT2Py), [Tb(acac)3NIT2Py]·0.5H2O (3) (acac = acetylacet-
onate), was synthesized for comparison with the previously reported
[Tb(hfac)3NIT2Py]·0.5C7H16 (1) (hfac = hexafluoroacetylacetonate),
together with their yttrium analogues [Y(hfac)3NIT2Py]·0.5C7H16 (2)
and [Y(acac)3NIT2Py]·0.5H2O (4). The crystal structures show that in
all complexes the nitronyl nitroxide radical acts as a chelating ligand.
Magnetic studies show that 3 like 1 exhibits slow relaxation of
magnetization at low temperature, suggesting single-molecule magnet
behavior. The luminescence spectra show resolved vibronic structure with the main interval decreasing from 1600 cm−1 to 1400
cm−1 between 80 and 300 K. This effect is analyzed quantitatively using experimental Raman frequencies.
building blocks of molecular-based magnetic materials.42−46
INTRODUCTION
■
Comparatively they have been almost forgotten in the course to
SMMs.41,47−53 However, recent discoveries of a peculiar
enhancement of the energy barrier for reversal of magnetization
in dinuclear lanthanide complexes with bridging dinitrogen54,55
and bipyrimidyl56 radicals have shown that the metal−radical
may also be promising for SMMs.
Herein we report the comparative study of two Tb-radical
complexes based on the same nitronyl nitroxide radical
(NIT2Py) as shown in Figure 1 but with hexafluoroacetylacet-
Single-molecule magnets (SMMs) have attracted more and
more attention since their discovery.1 At first, the design of
coordination compounds that exhibit slow relaxation of the
magnetization has mainly dealt with increasing the ground state
spin of polynuclear compounds based on 3d elements.2,3 The
discovery that single-ion complexes of lanthanide with
unquenched orbital angular momentum may show SMM
behavior has established magnetic anisotropy as a main
factor.4,5 Following this has opened the way to the synthesis
of numerous compounds made of only lanthanide ions or
combined with 3d metal ions to instill anisotropy in high-spin
clusters.6−10 This has resulted in the discovery of many
compounds exhibiting large energy barriers for the reversal of
the magnetization.11−13 Some actinide complexes have also
been found to show SMM behavior.14−20 However, the
blocking temperatures are still observed at very low temper-
ature, and one’s main focus nowadays is to understand the
magneto-structural correlations to guide the synthesis work.
This mainly involves theoretical calculations8 with comple-
mentary experimental techniques such as EPR,21−24 polarized
neutron diffraction,25 or NMR.26,27 Studies of the optical
properties is also helpful to molecular magnetism, as we have
shown previously,28−34 and have become pervasive in recent
studies of lanthanide-based SMMs.35−3940 Complementary to
these approaches is to compare series of related compounds
that show subtle differences in magnetic behavior.41
Figure 1. Chemical structure of the nitronyl nitroxide radical NIT2Py.
onate (hfac) or acetylacetonate (acac) as ancillary ligand,
respectively: [Tb(hfac)3NIT2Py]0.5C7H16 (1), which was
previously reported by L.-C. Li,57 and [Tb(acac)3NIT2Py]·
0.5H2O (3), which is new, together with their yttrium
analogues [Y(hfac)3NIT2Py]·0.5C7H16 (2) and [Y-
(acac)3NIT2Py]·0.5H2O (4). Magnetic studies show that 3
Nitronyl nitroxide (NIT) radicals are a unique type of stable
and versatile free radicals that have been popular magnetic
Received: April 7, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Crystal Data for Tb(hfac)3NIT2Py·0.5C7H16 (1), Y(hfac)3NIT2Py·0.5C7H16 (2), Tb(acac)3NIT2Py·0.5H2O (3), and
Y(acac)3NIT2Py·0.5H2O (4)
a
1
2
3
4
formula
TbC30.5H27O8N3F18
1064.47
Y2C61H54 O16N6F36
1988.92
Tb2C54H76N6O17
1399.08
Y2C54H76N6O17
1259.03
M (g/mol)
cryst shape, color
cryst size (mm)
cryst syst
space group (no.)
T (K)
chunk, pink red
0.10 × 0.03 × 0.02
monoclinic
chunk, dark red
0.08 × 0.08 × 0.08
triclinic
needle, dark violet
0.132 × 0.256 × 0.983
monoclinic
polyhedron, violet
0.126 × 0.270 × 0.406
monoclinic
P21/n (#14)
150
P1 (#2)
C2/c (#15)
293
C2/c (#15)
293
̅
100
λ(Kα Cu) (Å)
λ(Kα Mo) (Å)
a (Å)
1.541 78
1.541 78
0.710 73
27.5929(7)
14.5138(3)
19.3570(6)
90
0.710 73
27.5452(13)
14.5077(5)
19.3707(10)
90
16.298(1)
12.8383(3)
20.5399(13)
90
14.4468(13)
15.0836(13)
18.8539(3)
102.465(4)
102.943(4)
92.014(5)
3894.6(6)
2
b (Å)
c (Å)
α (Å)
β (Å)
110.724(3)
90
114.947(3)
90
114.902(6)
90
γ (Å)
V (Å3)
4019.7(4)
4
7028.8(4)
4
7021.2(6)
4
Z
D (g/cm3)
1.759
1.696
1.322
1.191
μ (mm−1
)
9.889
3.387
2.057
1.706
F(000)
2088
1984
2832
2624
θ range (deg)
reflns (h k l)
3.0−69.8
−19 ≤ 19
−15 ≤ 15
−24 ≤ 24
79 557
2.5−71.7
−17 ≤ 17
−18 ≤ 18
−23 ≤ 23
102 315
14 567
2.9−29.4
−37 ≤ 32
0 ≤ 18
0 ≤ 26
8397
2.9−29.4
−37 ≤ 32
0 ≤ 19
0 ≤ 24
8395
no. of reflns
no. of indep reflns
no. of params
Rint
7510
5961
4643
693
1135
359
357
0.100
0.065
0.029
0.062
b
R1
0.0523
0.0788
0.0358
0.0410
1.0608
0.0571
0.0674
1.059
c
wR2
0.1437
0.2194
GOF on F2
1.035
1.025
Δρmax/Δρmin (e·Å−3
)
0.463/−1.450
2.182/−1.082
1.08/−0.85
0.51/−0.42
a
b
c
For comparison with ref 51. R1 = ∑||Fo| − |Fc||/∑|Fo|. wR2 = [∑(w(Fo2 − Fc2)2)/∑(w(Fo )2)]1/2 with w = 1/[(σ2Fo ) + (aP)2 + bP] and P =
2
2
2
(max(Fo ) + 2Fc2)/3.
was then extracted with 2 × 50 mL of diethyl ether. The organic phase
was dried with sodium sulfate, then filtered and evaporated at room
temperature to give the title compound in 75% yield.
like 1 exhibits slow relaxation of magnetization at low
temperature, suggesting single-molecule magnet behavior.
Luminescence spectra for all four complexes have been
measured on crystalline samples. They show band shapes
similar to the uncoordinated NIT2Py previously reported by
us29 but with small shifts and vibronic structure with unusual
temperature-dependent variations. This was analyzed using the
time-dependent theory of electronic spectroscopy and exper-
imental Raman frequencies.58
[M(hfac)3NIT2Py·0.5C7H16]51,62 with M = TbIII (1) or YIII (2). A
suspension of 100 mg of [M(hfac)3(H2O)2] was prepared in 10 mL of
heptane and boiled at 90 °C for 1 h until total dissolution of the
complex. The heating was stopped, and a few drops of dichloro-
methane were added to avoid crystallization upon cooling. When this
solution was cool enough (<30 °C), 27 mg (0.12 mmol) or 32 mg
(0.13 mmol) of NIT2Py respectively for the Tb and Y compounds
previously dissolved in a minimum of dichloromethane was added.
After 1 week, slow evaporation of the solvent gave single crystals as
dark violet blocks. Yield: 60%.
EXPERIMENTAL SECTION
■
Materials. All reactions were conducted in aerobic conditions. All
reagents were obtained from commercial sources and used as received.
2,3-Bis(hydroxylamino)-2,3-dimethylbutane59 and nitronyl nitroxide
free radical 2-(2-pyridyl)-4,4,5,5-tetramethyl-4,5-dihydro1H-imidazol-
yl-1-oxy-3-oxide (NIT2Py) have been synthesized as previously
reported.60
[M(acac)3NIT2Py·0.5H2O] with M = TbIII (3) or YIII (4). A 100 mg
portion of of M(acac)3·3H2O (M = Tb, 509.92 g/mol, 0.196 mmol; M
= Y, 439.9 g/mol, 0.227 mmol) was dissolved in a 10 mL mixture of
dichloromethane/heptane (3:7). After dissolution by stirring, 40 mg
(0.17 mmol) or 46 mg (0.19 mmol) of NIT2Py respectively for the Tb
and Y compounds, previously dissolved in a minimum of dichloro-
methane, was added. The resultant solution was then stirred for 1 h,
and three drops of water were slowly added. The solution was left
undisturbed for slow evaporation of the solvent at room temperature.
After 2 days, small single crystals were collected as dark violet sticks.
Yield: 25%.
Synthesis of Complexes. [M(hfac)3(H2O)2]61 with M = TbIII or
YIII. A 5 g amount of hexafluoroacetylacetone stored in a refrigerator
was added to 15 mL of cold water. The mixture was stirred strongly for
30 min, then kept in the refrigerator for 3 days to give [hfac·2H2O] as
a solid, which was isolated by filtration. Then 200 mg (0.81 mmol) of
this solid was dissolved in 100 mL of diethyl ether, and 126 μL of
ammonia (25%) was added under stirring. The resulting solution was
then mixed with a solution of TbCl3·6H2O (100 mg, 0.29 mmol) or
YCl3·6H2O (82 mg, 0.29 mmol) in 10 mL of H2O. The aqueous phase
Crystal Structure Determination. Single crystals of [Tb-
(hfac)3NIT2Py]·0.5C7H16 (1) and [Y(hfac)3NIT2Py]·0.5C7H16 (2)
were mounted on a Bruker SMART 6000 diffractometer with optics,
B
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. Metal−Ligand Bond Lengths (Å) in 1 and 2
atoms
distance
atoms
distance
atoms
distance
atoms
distance
atoms
distance
Tb1−O1
Tb1−O8
Tb1−O5
Tb1−O4
Tb1−O6
Tb1−O7
Tb1−O3
Tb1−N3
2.313(4)
2.323(4)
2.330(4)
2.360(3)
2.375(4)
2.375(4)
2.378(4)
2.587(4)
Y1−O11
Y1−O17
Y1−O16
Y1−O13
Y1−O14
Y1−O15
Y1−O18
Y1−N13
2.258(5)
2.296(5)
2.321(5)
2.325(5)
2.328(5)
2.328(5)
2.365(5)
2.609(6)
Y2−O21
Y2−O26
Y2−O24
Y2−O28
Y2−O27
Y2−O23
Y2−O25
Y2−N23
2.270(5)
2.297(5)
2.309(5)
2.310(5)
2.317(5)
2.332(6)
2.381(6)
2.625(6)
Tb1−O11
Tb1−O8
Tb1−O10
Tb1−O9
Tb1−O7
Tb1−O12
Tb1−O5
Tb1−N4
2.323(3)
2.324(3)
2.327(3)
2.331(3)
2.352(3)
2.364(3)
2.405(3)
2.700(3)
Y1−O8
Y1−O11
Y1−O10
Y1−O9
Y1−O7
Y1−O12
Y1−O5
Y1−N4
2.299(3)
2.302(3)
2.306(3)
2.309(2)
2.329(3)
2.342(3)
2.380(3)
2.689(3)
equipped with a monochromatic copper Cu Kα radiation source (λ =
1.547 18 Å) at 100 and 150 K, respectively. Cell refinement and data
reduction were performed using APEX2.63 Absorption corrections
were applied using SADABS.64 Structures were solved by direct
methods using SHELXS97 and refined on F2 by full-matrix least-
squares using SHELXL97.65 All non-hydrogen atoms were refined
anisotropically. Hydrogen atoms were refined on calculated positions
using a riding model.
Single-crystal X-ray studies of [Tb(acac)3NIT2Py]·0.5H2O (3) and
[Y(acac)3NIT2Py]·0.5H2O (4) were carried out using a Gemini
diffractometer and the related analysis software.66 An absorption
correction based on the crystal faces was applied to the data sets
(analytical).67 The structure was solved by direct methods using the
SIR97 program68 combined with Fourier difference syntheses and
refined against F using reflections with [I/σ(I) > 3] with the
CRYSTALS program.68,69 All atomic displacement parameters for
non-hydrogen atoms have been refined with anisotropic terms. The
hydrogen atoms were theoretically located on the basis of the
conformation of the supporting atom and refined keeping restraints
(riding mode). Complexes 3 and 4 contain accessible solvent voids,
with highly disordered solvent molecules, which have been removed
by the SQEEZE program.70
Complexes 1 and 2 show a significant disorder in their crystal
structure, a disorder higher for 2 than for 1. For Tb1 in 1, 12 of 18
fluoride atoms are defined in two positions; the C18, C22, C23, and
C27 carbon atoms have six fluoride atoms each instead of three for
defining a CF3 molecule. For Y1 and Y2 in 2, nine and 12 of 18
fluoride atoms are defined in two positions, respectively. C117, C123,
C127 for Y1 and C213, C218, C223, C227 for Y2 have six fluorides
each. The disorder on the CF3 molecules is commonly encountered in
the coordination complexes. The heptane molecule close to a
symmetry element is disordered over two positions. The NIT2Py
connected to Y2 also has a disorder on the four methyl groups, which
are defined on two positions. For all these reasons, the R1 and wR2
reliability factor for 2 (R1 = 0.0788 and wR2 = 0.2194) are higher than
for 1 (R1 = 0.0523 and wR2 = 0.1437). The hydrogen atoms are
placed arbitrarily during the refinement.
Magnetic Measurements. Magnetic susceptibility data (2−300
K) were collected on powdered polycrystalline samples on a Quantum
Design MPMS-XL SQUID magnetometer under an applied magnetic
field of 0.1 T. Alternating current measurements were performed in
the 2−10 K range using a 2.7 G ac field oscillation in the 1−1500 Hz
range. Magnetization isotherms were collected at 2 K between 0 and 5
T. All data were corrected for the contribution of the sample holder
and the diamagnetism of the samples estimated from Pascal’s
constants.71
response. The luminescence band maxima Emax are determined using
the least-squares fit of a Gaussian to the top 10−15% of each band.
The Raman band maxima are determined using the least-squares fit
of a Gaussian to the top 20−25% of each band. Maxima are given with
a maximum standard deviation of 0.1 cm−1. The relative intensity of
100% is attributed to the most intense band of the spectra.
RESULTS
■
Crystal Structures. The crystal data and refinement
parameters for compounds 1−4 are summarized in Table 1.
Selected bond lengths and angles are given in Table 2.
[M(hfac)3NIT2Py]·0.5C7H16 (M = Tb(III) or Y(III)). [Tb-
(hfac)3NIT2Py]·0.5C7H16 (1) crystallizes in the P21/n
monoclinic space group as was previously reported by Li et
al.51 While 1 is isostructural with the gadolinium derivative
[Gd(hfac)3NIT2Py]·0.5C7H16 reported by Gatteschi et al.,62
complex 2 was found to crystallize in the P1 triclinic space
̅
group. The asymmetric unit of 2 consists of two [Y-
(hfac)3NIT2Py] molecules and one heptane molecule, in
general positions, corresponding to two independent entities in
contrast to 1. However, the two [Y(hfac)3NIT2Py] molecules
have structural features close to those found for the terbium
derivative (1). The Y(III) ion is coordinated to six oxygen from
the hfac ligands, one oxygen from the NO group, and one
nitrogen atom from the pyridine group (Figure S2). The
coordination sphere is a deformed square antiprism. The Y−O
bond lengths are between 2.258(5) and 2.365(5) Å for Y1 and
between 2.270(5) and 2.381(6) Å for Y2, whereas the longer
Y−N bond length is 2.609(6) and 2.625(5) Å for Y1 and Y2,
respectively. The O−Y−O angles are between 70.4(2)° and
144.4(2)° and between 68.9(2)° and 145.9(2)° for Y1 and Y2,
respectively, while the O−Y−N angles are in the range
70.7(2)−146.0(2)° for Y1 and in the range 70.0(2)−
146.9(2)° for Y2. The two minima O−Y−N values correspond
to the chelating NIT2Py angles to Y1 and Y2. The NIT2Py
radical shows normal bond lengths and angles.72,73 The
pyridine and nitronyl nitroxide (ONCNO) moieties make a
dihedral angle of 20.2(4)° and 22.6(4)° respectively for Y1 and
Y2, which are smaller than the 26.8° reported for 1.51
As the crystal cells contain in both case four [M-
(hfac)3NIT2Py] molecules and two heptane molecules, the
crystal packing of compound 2 is significantly more compact;
the cell volumes are 3995.2(14) and 3894.6(6) Å3 for 1 and 2,
respectively. The crystal packing of 1 and 2 resulted in well-
separated [M(hfac)3NIT2Py] with the heptane solvent
molecules inserted in between (Figures S3 and S4). The
shortest intermolecular Tb···Tb distance was reported as 9.273
Å.51 In 2 the shortest intermolecular Y2···Y2 and Y1···Y1
distances are 8.673(1) and 9.010(1) Å, respectively, and the
shortest intermolecular Y1···Y2 distance is 9.755(1) Å. In both
compounds the shortest intermolecular distances between the
uncoordinated NO groups are greater than 7 Å.
Luminescence and Raman Spectroscopy. Luminescence and
Raman spectroscopy measurements were carried out using a Renishaw
inVia imaging microscope system. Excitation sources were a 488 nm
and a 514 nm argon ion laser for luminescence experiments and a 782
nm diode laser for the Raman experiments. The microscope was used
to focus the light onto a spot of approximately 1 μm in diameter and
to collect the scattered light. The backscattered Raman light was
detected with a Peltier-cooled CCD detector. Sample temperatures
were controlled with a Linkam microscope cryostat between 80 and
290 K. All spectra were unpolarized and corrected for spectrometer
C
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
[M(acac)3NIT2Py]·0.5H2O (M = Tb(III) or Y(III)). The crystal
structures of complexes 3 and 4 are isostructural and crystallize
in the C2/c monoclinic space group (Table 1). The asymmetric
unit consists of one [M(acac)3NIT2Py] molecule and a half
water molecule located on a symmetry plane.
Like for 1 and 2, the metal center (Tb, Y) ion is coordinated
to six oxygen atoms of acac ligands, one oxygen atom from a
NO group, and one nitrogen atom from the pryridyl group.
The coordination sphere is a deformed square antiprism
geometry. In compound 3 (Figure 2) the Tb1−O bond lengths
acetylacetonate ligand and the pyridine moiety of a second
entity (Figures S5 and S6). In 3 the shortest intermolecular
Tb···Tb distance is 8.097 Å. In 4 the shortest intermolecular
Y···Y is 8.066 Å. The shortest intermolecular distances between
an uncoordinated NO group and a Tb ion and between two
uncoordinated NO are greater than 6 Å.
Magnetic Properties. [Tb(hfac)3NIT2Py]·0.5C7H16. The
magnetic behavior of 1 was found to be as reported.51 The
χT value at 300 K is in agreement with uncoupled Tb(III) and
the radical. Upon cooling, χT decreases first, then from 40 K
increases to reach a maximum at 10 K, then decreases on
further cooling. The field dependence of the magnetization for
1 does not reach the saturation value due to magnetic
anisotropy. The ac measurements under a zero direct-current
(dc) field with an alternating current (ac) field of 3 Oe
indicated slow relaxation of the magnetization with an energy
barrier of Δ/kB = 17.1 K and preexponential factor τ0 = 9.56 ×
10−7 s [τ = τ0 exp(Δ/KBT)].
[Tb(acac)3NIT2Py]·0.5H2O (3). At 300 K the χT value is
11.78 emu·K·mol−1, in good agreement with a Tb(III) ion (7F6
and g = 3/2) and a free radical that is not interacting. Upon
cooling, χT slowly and continuously decreases as shown in
Figure 3 to finally reach a value of 6.94 emu·K·mol−1 at 2 K.
Figure 2. Representation of complex 3. Ellipsoids are represented at
the 50% probability level for all atoms. Hydrogen atoms are omitted
for clarity.
Figure 3. Plot of χT vs T under an applied field of 1000 Oe, with
magnetic dependence of the magnetization at 2 K for 3.
are in the range 2.321(3)−2.404(3) Å, whereas the unique
Tb1−N4 bond length is 2.702(3) Å. The O−Tb1−O angles
are between 70.00(10)° and 147.56(10)°, while the O−Tb1−
N4 angles are in the range 66.96(9)−147.88(10)°; the
minimum O−Tb1−N4 value corresponds to the chelating
NIT2Py angle to Tb(III). In compound 4 the Y1−O bond
lengths are in the range 2.302(3)−2.380(3) Å, whereas the
unique Y1−N4 bond length is 2.689(3) Å. The O−Y1−O
angles are between 70.06(9)° and 148.14(9)°, while the O−
Y1−N4 angles are in the range 67.16(9)−147.09(9)°; the
minimum O−Tb1−N4 value corresponds to the chelating
NIT2Py angle to Y(III). Like for 1 and 2, NIT2Py shows
normal bond length and angles.72,73 The pyridine and nitronyl
nitroxide (ONCNO) moieties make a dihedral angle of
27.2(3)° and 26.4(3)° respectively for 3 and 4.
The crystal packing of 3 and 4 is identical, as the differences
in crystal cell volume are meaningless in comparison with
compounds 1 and 2. The crystal packing may be described as a
two-dimensional network developing parallel to the ab plane
through intermolecular hydrogen bonds connecting the
[M(acac)3NIT2Py] entities via the crystallized water molecules,
the oxygen of the acac ligand, and the methyl groups of the
imidazolyl moiety (Figure S7). The [M(acac)3NIT2Py] are also
paired by hydrogen bonds involving an oxygen atom of an
The field dependence for 3 was performed at 2 K in the field
range of 0−50 kOe. Upon increasing the magnetic field, the
magnetization increases to reach a value of 4.91 μB, as shown in
Figure 4.
To examine the spin dynamics of complex 3, ac measure-
ments have been carried out in the range 2−10 K with an ac
field of 3 Oe at various frequencies and under dc magnetic
fields of 0, 1000, and 3500 Oe. The temperature dependence of
the in-phase (χ′) and out-of-phase (χ″) ac susceptibilities is
frequency dependent below 4 K, regardless of the magnetic
field (Figures S9 and S10 and Figure 4). In contrast with
compound 1 there are no maxima visible down to 2 K under a
zero dc magnetic field. Under a dc magnetic field of 1000 Oe
(Figure S9a and b) and 3500 Oe (Figure S10a and b), 3
exhibits maxima values. The frequency dependence of the χ″
peak was fitted by the Arrhenius law τ = τ0 exp[Δ/(kBT)] in
which T is the temperature of the maximum χ″ at different
frequencies and τ = 1/2πν. For measurement under a dc
magnetic field of 1000 Oe this gives the preexponential factor
τ0 = 10−7 s, and the energy barrier for the relaxation of the
magnetization Δ/kB = 21.18 K with an R factor of 0.9722
D
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
NIT2Py radical and [Tb(hfac)3NIT2Py]·0.5C7H16 can be
made (uncoordinated NIT2Py/[Tb(hfac)3NIT2Py]·
0.5C7H16). The bands at 616/621, 995/1011, 1572/1575,
and 1589/1596 cm−1 are assigned as modes of the pyridine
group.74,75 Pyridine shows Raman bands at 603/622, 991/1002,
1029, 1067/1071, 1213/1219, 1482, and 1583 cm−1.29,74 The
bands at 1418/1425, 1469/1473, and 1521/1528 cm−1 and the
band at 1167 cm−1 for 1 correspond to the imidazolyle ring
stretching frequencies.76 The Raman bands at 1521/1528 and
1469/1473 cm−1 can be attributed to the ν(O−NC)
stretching frequency.76,77 The ν(N−O•) stretching frequency
is given as approximately 1370 cm−1 in the literature.78 Two
bands are observed in this range, at 1365 and 1376 cm−1, for
uncoordinated NIT2Py and at 1370 cm−1 for 1. The δ(C−C−
C) bending frequencies of the imidazolyl ring are reported at
204, 271, and 318/319 cm−1.76 The ν(Tb−O) and ν(Tb−N)
stretching frequencies are not easily assigned through
comparisons with the literature.79
Figure 4. Temperature dependence of the in-phase (top) and out-of-
phase (bottom) ac magnetic susceptibility for compound 3 in a 0 Oe
dc field with an oscillating field of 3 Oe. The solid lines are a guide for
the eye.
The Raman spectra of 2 have been measured between 80 and
300 K and are very similar to those of 1. A shift of the Raman
maxima to lower frequencies as the temperature increases is
observed for 2, again similar to 1. The vibrational modes
assigned to 1 can also be assigned for 2.
(Figure S9c) suggests an SMM behavior. For measurement
under a dc magnetic field of 3500 Oe this gives the pre-
exponential factor τ0 = 3 × 10−8 s, and the energy barrier for
the relaxation of the magnetization is Δ/kB = 26.036 K with an
R factor of 0.9980 (Figure S10c).
[Y(acac)3NIT2Py]·0.5H2O (4). At 300 K the χT value for 4 is
0.415 emu·K·mol−1, in good agreement with the expected value
for one radical (Figure S11). Upon cooling, χT decreases slowly
down to 30 K, then decreases abruptly to reach a value of 0.166
emu·K·mol−1 at 2 K. The experimental curve has been well
fitted with a Curie−Weiss law for one radical (s = 1/2) to give
θ = −3.1 K, which confirms the weak radical−radical
antiferromagnetic intermolecular interaction.
Raman Spectroscopy. Raman spectra at variable temper-
ature were recorded for [Tb(hfac)3NIT2Py]·0.5C7H16 (1),
[Y(hfac)3NIT2Py]·0.5C7H16 (2), and the uncoordinated
NIT2Py radical between 80 and 300 K for Raman shifts
between 100 and 2000 cm−1. Raman shifts given in the
following correspond to values obtained at 80 K. No Raman
data on the uncoordinated NIT2Py were found in the literature.
The data have been compared to the NITBzImH radical (2-(2-
benzimidazolyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide)
characterized by Raman spectroscopy, a radical similar to
NIT2Py.29
The values of some vibrational modes previously assigned for
1 can be assigned for 2 as the pyridine vibration modes of the
pyridine group at 621, 1008, 1577, and 1596 cm−1. The
imidazolyle ring stretching frequencies are observed at 1426,
1473, 1524, and 1168 cm−1, the δ(C−C−C) bending frequency
of the imidazolyle ring at 210, 277, and 326 cm−1, and the ρr
rocking in the plane of the CH3 methyl groups at 871 cm−1.
The Raman spectra of 3 have been measured in the
temperature range between 78 and 293 K. A slight shift of the
Raman bands to lower energies as the temperature increases is
observed, similar to 1 and 2. Changing the ancillary ligands
should lead to some differences between the Raman spectra of
1 and 3. The Raman bands measured for 3 are relatively broad,
but spectra of 3 show similarities with 1 and 2. The ν(Tb−O)
and ν(Tb−N) stretching frequencies are again not easily
discernible by comparison with the literature.79
The frequencies of some modes previously assigned for 1 can
be assigned for the pyridine group in 3 at 624, 1004, and 1589
cm−1. The peak at 1577 cm−1 is not discernible. The
imidazolyle ring stretching frequencies at 1419, 1454, 1513,
and 1173 cm−1, the δ(C−C−C) bending frequency of the
imidazolyle ring at 218 and 320 cm−1, and the ρr rocking in the
plane of the CH3 methyl groups at 873 cm−1 are assigned. The
band corresponding to the ν(N−O•) stretching mode is most
likely at 1363 cm−1.
The Raman spectra of 4 have been measured in the
temperature range between 80 and 300 K. A slight shift of the
Raman bands to lower frequencies as the temperature increases
is observed, again similar to 1 and 2, with corresponding
assignments.
Luminescence Spectroscopy. Figure 5a shows the
luminescence spectra of [Tb(hfac)3NIT2Py]·0.5C7H16 (1)
between 80 and 300 K. Two energy ranges are characterized:
the first between 10 000 and 15 400 cm−1, which corresponds
to the luminescence of the NIT2Py radical.28 In this area, a
broad band with three maxima, EmaxA, EmaxB, and EmaxC, at
approximately 14 525, 13 070, and 11 700 cm−1 is observed at
80 K. The second energy range between 15 400 and 20 000
cm−1 corresponds to f−f transitions of Tb(III). The spectra of 1
at room temperature and 270 K show three band maxima at
Figure S12a shows the Raman spectra of 1 between 80 and
300 K. Spectra are normalized on the most intense band to
simplify the comparison at variable temperature.
The Raman spectra between 80 and 300 K show no
significant change aside from higher resolution at lower
temperature for some bands. Complex 1 does not show any
phase transitions based on the Raman spectra. Some Raman
bands shift to slightly lower frequencies as the temperature
increases. The Raman bands between 600 and 1600 cm−1
correspond essentially to vibrational modes of the NIT2Py
radical. This observation can be confirmed by the variable-
temperature Raman spectra of the uncoordinated NIT2Py
radical. The vibrational frequencies of uncoordinated NIT2Py
(616, 991, 1067 cm−1 , ...) are lower by a few cm−1 than for the
coordinated NIT2Py in 1 (621, 1002, 1071 cm−1, ...).
Assignments are limited to comparisons with the literature. A
comparison of Raman frequencies for the uncoordinated
E
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. Luminescence spectra at variable temperature (80−300 K) of (a) 1, (b) 3, (c) 2, and (d) 4. The luminescence spectra are offset along the
y-axis for clarity. The bands were assigned according to the literature for Tb(III).
approximately 18 360, 17 130, and 16 100 cm−1. These three
NIT2Py radical are at somewhat different energy than when
coordinated to Tb(III) ions. The maxima for 1 are shifted to
14 525 and 13 070 cm−1 at 80 K, differences of approximately
435 and 500 cm−1, respectively. Similar shifts of the
luminescence band maxima for uncoordinated and coordinated
ligands have already been reported in the literature for Gd(III)
complexes with three hfac ligands and one NITBzImH radical.6
The shift of the luminescence maxima has been attributed to a
metal−ligand charge transfer (MLCT) effect.6 The similarities
between Gd(III) and Tb(III) complexes with NIT2Py and
NITBzImH radicals illustrate that the luminescence behavior
observed for [Tb(hfac)3NIT2Py]·0.5C7H16 is similar to that of
Gd(hfac)3NITBzImH.6
5
maxima correspond to luminescence transitions from the D4
excited state to the 7F5, 7F4, and 7F3 states of Tb(III).80 Tb(III)
luminescence is not visible at temperatures between 80 and 240
K. At 270 K, the intensity of the f−f transitions is significant
and decreases slightly at 300 K.
The resolution of the uncoordinated NIT2Py luminescence
spectra is not as good as for the spectra of 1. The spectra show
two luminescence maxima at approximately 14 050 and 12 950
cm−1 and a shoulder at lower energy visible at 80 K. Four
maxima are reported at 5 K for the uncoordinated NIT2Py
radical, at 14 090, 13 570, 11 750, and 11 010 cm−1.29 Similar
band maxima have been reported for uncoordinated NIT
radicals in the solid state.40 The maxima for uncoordinated
F
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 3. Variation of the Experimental and Calculated Luminescence Band Maxima EmaxA, EmaxB, ΔEmax, EmaxATh, EmaxBTh, and
ΔEmaxTh in Wavenumber Units (cm−1) from the Luminescence Spectra between 80 and 300 K of 1
temperature (K)
EmaxA (cm−1
)
EmaxB (cm−1
)
ΔEmax (cm−1
)
EmaxATh (cm−1
)
EmaxBTh (cm−1
)
ΔEmaxTh (cm−1
)
80
120
160
180
200
220
240
270
300
14 525
14 475
14 465
14 400
14 385
14 360
14 300
14 270
14 200
13070
13050
13005
13030
13065
13050
13000
13020
12960
1455
1425
1360
1370
1320
1310
1300
1250
1240
14520
14474
14446
14385
14343
14350
14305
14250
14196
13075
13054
13078
13025
13067
13038
13000
13030
12940
1445
1420
1368
1360
1284
1312
1305
1220
1256
Table 4. Variation of the Experimental and Calculated Luminescence Band Maxima EmaxA, EmaxB, ΔEmax, EmaxATh, EmaxBTh, and
ΔEmaxTh in Wavenumber Units (cm−1) from the Luminescence Spectra between 80 and 300 K of 3
temperature (K)
EmaxA (cm−1
)
EmaxB (cm−1
)
ΔEmax (cm−1
)
EmaxATh (cm−1
)
EmaxBTh (cm−1
)
ΔEmaxTh (cm−1
)
78
103
123
143
163
183
203
223
243
273
293
14 448
14 404
14 350
14 288
14 257
14 181
14 176
14 151
14 098
14 041
13 960
13 116
13 095
13 055
13 058
13 078
13 050
13 141
13 161
13 127
13 126
13 137
1332
1309
1295
1230
1179
1131
1035
990
14 443
14 399
14 330
14 299
14 233
14 189
14 160
14 131
14 122
14 090
13 981
13 137
13 132
13 081
13 074
13 074
13 112
13 100
13 090
13 085
13 085
13 075
1306
1267
1249
1225
1159
1077
1060
1041
1037
1005
906
971
915
823
Table 5. Variation of the Experimental and Calculated Luminescence Band Maxima EmaxA, EmaxB, ΔEmax, EmaxATh, EmaxBTh, and
ΔEmaxTh in Wavenumber Units (cm−1) from the Luminescence Spectra between 80 and 300 K of 2
temperature (K)
EmaxA (cm−1
)
EmaxB (cm−1
)
ΔEmax (cm−1
)
EmaxATh (cm−1
)
EmaxBTh (cm−1
)
ΔEmaxTh (cm−1
)
80
120
160
180
200
220
240
270
300
14 675
14 540
14 550
14 470
14 445
14 440
14 345
14 330
14 235
13 220
13 140
13 155
13 095
13 090
13 115
13 060
13 050
12 960
1455
1400
1395
1375
1355
1330
1285
1280
1275
14 666
14 542
14 538
14 470
14 448
14 446
14 347
14 333
14 247
13 220
13 131
13 133
13 090
13 090
13 110
13 060
13 050
12 966
1446
1411
1405
1380
1358
1336
1287
1283
1281
For uncoordinated NIT2Py radical, the energy difference
between the two luminescence bands is approximately 1100
cm−1. The corresponding difference is approximately 1455
cm−1 at 80 K for the radical coordinated to Tb(III). This
energy difference between luminescence maxima EmaxA and
EmaxB (noted ΔEmax) for 1 decreases between 80 and 300 K
from 1455 cm−1 to 1240 cm−1, as shown in Table 3 and Figure
S13a. The two luminescence bands shift to lower energies
between 80 and 300 K, by approximately 325 cm−1 from 14 525
to 14 200 cm−1 for EmaxA and by 110 cm−1 for EmaxB, where the
maximum oscillates between 13 070 and 12 960 cm−1 as the
temperature increases. The intensities of EmaxA and EmaxB are
approximately equal between 80 and 160 K, but the intensity
ratio changes at higher temperature, where EmaxB becomes
slightly more intense than EmaxA, particularly for the spectrum at
270 K. The maximum EmaxC between 11 700 and 11 800 cm−1 is
observed as a shoulder in the temperature range between 300
and 160 K.
contrast to 1, only one energy range is characterized, between
10 000 and 15 400 cm−1, corresponding to the luminescence of
the NIT2Py radical.28 In this area, two luminescence band
maxima, EmaxA and EmaxB, at approximately 14 165 and 12 923
cm−1 are observed at 78 K. A shoulder can be observed at lower
energy, corresponding to the third luminescence band observed
in 1, but the resolution is not good enough to determine EmaxC
.
The energy range between 15 400 and 20 000 cm−1
corresponding to the f−f luminescence of the Tb(III) metal
center does not show any luminescence band at any
temperature, in contrast to 1.
The two band maxima at 78 K show shifts to higher and
lower energies compared to uncoordinated NIT2Py ligand,29
with an approximate variation +50 cm−1 for EmaxA from 14 090
to 14 137 cm−1 and −590 cm−1 for EmaxB from 13 570 to 12 980
cm−1. The change from hfac to acac ligands leads to decreases
of approximately 388 and 90 cm−1 for EmaxA and EmaxB
,
respectively, with approximate variations of 3% for EmaxA and
0.5% for EmaxB. The band maxima EmaxA shift to lower energy by
approximately 177 cm−1 (from 14 137 cm−1 to 13 960 cm−1
Figure 5b shows the luminescence spectra of [Tb-
(acac)3NIT2Py]·0.5H2O (3) between 78 and 293 K. In
G
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 6. Variation of the Experimental and Calculated Luminescence Band Maxima EmaxA, EmaxB, ΔEmax, EmaxATh, EmaxBTh, and
ΔEmaxTh in Wavenumber Units (cm−1) from the Luminescence Spectra between 80 and 300 K of 4
temperature (K)
EmaxA (cm−1
)
EmaxB (cm−1
)
ΔEmax (cm−1
)
EmaxATh (cm−1
)
EmaxBTh (cm−1
)
ΔEmaxTh (cm−1
)
78
120
160
180
200
220
240
270
293
14 340
14 290
14 240
14 208
14 186
14 104
14 087
13 995
13 926
13 043
13 013
13 020
13 024
13 078
13 095
13 098
13 057
13 014
1297
1277
1220
1184
1108
1004
989
14 338
14 295
14 198
14 216
14 195
14 090
14 101
13 934
13 899
13 021
13 000
12 980
13 001
13 057
13 000
13 056
13 033
12 965
1317
1295
1218
1214
1138
1090
1045
984
938
912
935
between 78 and 293 K), corresponding to approximate
variations of 1%, as shown in Table 4 and Figure S13b, while
EmaxB remains constant at approximately 12 980 cm−1. The
variation between the two band maxima is higher for 3 than for
1, 274 cm−1 compared to 215 cm−1. The intensity of EmaxA is
higher by a factor of 2 than EmaxB for temperatures between 78
and 293 K. The intensity of the shoulder corresponding to
EmaxC remains constant over the temperature range studied.
Figure 5c shows the luminescence spectra of [Y-
(hfac)3NIT2Py]·0.5C7H16 (2) between 80 and 300 K. As
expected, no luminescence from the Y(III) metal centers is
observed in the recorded energy range. The luminescence
bands of the radical are between 10 000 and 15 400 cm−1,
similar to complex 1, although the luminescence maxima are
slightly higher in energy, as shown in Table 5 and Figure S13c.
At 80 K, three luminescence maxima are observed at
923 cm−1. A shoulder can be observed at lower energy,
corresponding to the third luminescence band observed in 1
and 3, but the resolution is not good enough to allow a
maximum value to be determined. The two maxima at 78 K
show a shift to higher and lower energies compared to the
uncoordinated NIT2Py ligand,29 with an approximate variation
by 250 cm−1 for EmaxA from 14 090 cm−1 to 14 340 cm−1 and
335 cm−1 for EmaxB from 13 570 cm−1 to 13 043 cm−1. The
changes from hfac to acac ligand from 3 to 4 show decreases by
approximately 388 and 180 cm−1 for EmaxA and EmaxB
,
respectively, with approximate variations of 3% for EmaxA and
1.5% for EmaxB. The changes from Tb(III) to Y(III) ions from 2
to 4 show increases by approximately 203 cm−1 and 63 cm−1,
for EmaxA and EmaxB, respectively, with approximate variations of
1.5% for EmaxA and 0.5% for EmaxB. As shown in Table 6 and
Figure S13d, the band maxima EmaxA undergoes linear shifts to
lower energies between 78 and 293 K by approximately 414
cm−1 from 14 340 cm−1 to 13 926 cm−1, while EmaxB remains
constant in energy about 13 050 cm−1. Complex 4 does not
show a behavior similar to 3, but the same behavior as 2, since
only shifts for EmaxA are observed in 4. The variation between
the two band maxima is higher for 4 than for 3, 385 cm−1
compared to 274 cm−1, and is higher in 4 than in 2, 385 cm−1
compared to 180 cm−1. The intensity of EmaxA is higher by a
factor of 3 than EmaxB between 78 and 220 K and by a factor of
2 between 220 and 293 K. The intensity of the shoulder
corresponding to EmaxC remains constant over the studied
temperature range.
approximately 14 675, 13 220, and 11 870 cm−1 for EmaxA
,
EmaxB, and EmaxC, respectively. EmaxC is not easily discernible at
higher temperature. These three maxima are similar in energy
to the NIT2Py radical. Shifts of EmaxA, EmaxB, and EmaxC are
observed, again similar to 1. The three band maxima at 80 K
show larger shifts to higher energies compared to uncoordi-
nated NIT2Py radical,29 with an approximate variation of 585
cm−1 for EmaxA from 14 090 to 14 675 cm−1, 350 cm−1 for EmaxB
from 13 570 to 13 220 cm−1 and 120 cm−1 for EmaxC from
11750 to 11870 cm−1. The changes from Tb(III) to Y(III) ions
show increases by approximately 150 cm−1, 150 and 170 cm−1,
for EmaxA, EmaxB and EmaxC, respectively. As shown in Table 4,
the band maxima EmaxA and EmaxB shift to lower energies
between 80 and 300 K by approximately 440 cm−1 from 14 675
to 14 235 cm−1 and 260 cm−1 from 13 220 to 12 960 cm−1,
corresponding to approximate variations of 3% and 2%,
respectively. Complex 2 shows a behavior similar to 1, shifts
of 440 and 325 cm−1 for EmaxA and of 260 and 110 cm−1 for
EmaxB, respectively. The overall shift of EmaxB is less than that for
EmaxA, and the energy difference between the maxima decreases
from 1455 cm−1 at 80 K to 1275 cm−1 at 300 K. A behavior
similar to that for [Tb(hfac)3NIT2Py]·0.5C7H16 is observed,
but the variation between the two band maxima is lower for 2
than for 1, 180 cm−1 compared to 215 cm−1. The intensities of
EmaxA and EmaxB are approximately equal between 80 and 220 K,
but the intensity ratio changes at higher temperature, where
EmaxB becomes slightly more intense than EmaxA. The intensity
of EmaxC remains constant over the temperature range studied.
Figure 5d shows the luminescence spectra of [Y-
(acac)3NIT2Py]·0.5H2O (4) between 78 and 293 K. Again
no f−f luminescence is observed in the energy range recorded.
Two luminescence maxima, EmaxA and EmaxB, are given in Table
6. At 80 K, EmaxA and EmaxB are at approximately 14 165 and 12
DISCUSSION
■
Changing the ancillary ligand leads to different crystal packing
and change in the space group, which is not surprising, as steric
hindrance in acac is much lower than in hfac. As a consequence,
3 and 4 are not isomorphous with 1 or 2. The half-molecule of
heptane that cocrystallizes in 1 or 2 is replaced by a half-
molecule of water in 3 and 4, leading to the formation of intra-
and intermolecular hydrogen bonds. Then, while 1 or 2 may be
described as discrete and well-isolated molecules, 3 and 4 can
be described as 2D networks due to intermolecular hydrogen
bonds. This may be an important difference since the crystal
packing influences the magnetic properties.
Differences appear also in the metal−ligand bond lengths. In
case of 1 and 2, the M−Ohfac bond lengths are 2.323(4)−
2.378(4) Å and 2.296(5)−2.381 Å, respectively. The M−ONO
bond lengths are 2.313(4) Å for 1 and 2.258(5) and 2.270(5) Å
for 2. The M−N bond lengths are 2.587(4) Å for 1 and
2.609(6) and 2.625(6) Å for 2. In these complexes, the M−
ONO bond lengths are the shortest M−O bond lengths.
H
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
For 3 and 4, the M−Oacac bond lengths are 2.323(3) to
2.364(3) Å and 2.299(3) to 2.342(3) Å, respectively.
Compared to M−Ohfac bond lengths, there is no significant
variation. The M−ONO bond lengths are 2.405(3) and 2.380(3)
Å for 3 and 4, respectively. Compared to M−ONO bond lengths
in 1 and 2, there is an increase of 0.092 Å of the Tb−ONO bond
lengths, i.e., 3.7% longer, and an increase of 0.122 and 0.110 Å
of the Y−ONO bond lengths, i.e., 5% longer. The M−N bond
lengths for 3 and 4 are 2.700(3) and 2.689(3) Å, respectively.
Compared to the M−N bond lengths in 1 and 2, there is an
increase of 0.113 Å of the Tb−N bond length, i.e., 4% longer,
and an increase of 0.080 and 0.064 Å of the Y−N bond length,
i.e., 3% longer. For 3 and 4, the M−ONO bond lengths are the
longest M−O bond lengths.
Metal-β-diketonate bond lengths have already been reported
to be unaffected by the change from hfac to acac.50 However,
one may say that here the crystal structures for 1 and 2 were
determined at low temperature, 150 and 110 K respectively,
whereas the crystal structures of 3 and 4 were determined at
room temperature. To strengthen our assumption, the crystal
structure of 3 has been further determined at 100 K. The Tb−
Oacac bond lengths are between 2.321(3) and 2.351(3) Å, the
Tb−ONO bond length is 2.405(6) Å, and the Tb−N bond
length is 2.679(4) Å. That is, these bond lengths differ by less
than ∼0.02 Å at 100 K as compared with those found at room
temperature. In contrast, even at low temperature, the metal−
radical bond length is still longer by ∼0.1 Å in compound 3 as
compared to compound 1. This is ascribed to the electron-
withdrawing effect of the hexafluoroacetylacetonate ancillary
ligand, which increases, as expected, the Lewis acidity of the
metal center, leading to shorter bond lengths in complex 1 than
in complex 2.
The recent discoveries that the metal−radical strategy may
contribute to the energy barrier for reversal of magnetization in
lanthanide-based SMMs54−56 give some potential to the design
of single-molecule magnets using nitronyl nitroxide radicals.
However, in our case, nitronyl nitroxide radicals act only as
chelating ligands, leading to discrete molecules. Further work
will look at using nitronyl nitroxide as bridging ligands, as they
may contribute to increase both the total spin and the strength
of magnetic interactions to enhance properties of SMMs.
Compounds 2 and 4 of the diamagnetic Y(III) do not show
significant magnetic interest but were studied to have an
estimation of the intermolecular interaction and to obtain the
response of coordinated nitronyl nitroxide radicals in the
luminescence studies.
From the magnetic behavior of 4, it may be seen, and in
agreement with the small negative θ value found for the Curie−
Weiss fitting, that weak intermolecular antiferromagnetic
interactions are operative in 4 and therefore in 3, as they are
isomorphous. For 1 and 3 the χT values at 300 K, respectively
12.16 and 11.78 emu·K·mol−1, are close and in agreement with
that expected. Then upon cooling, the temperature dependence
of the product of the magnetic susceptibility with temperature
(χT) shows that the Tb(III) and the radical NO moiety are
ferromagnetically coupled in 1,51 whereas they are antiferro-
magnetically coupled in 3 (Figure 3). A ferromagnetic
interaction was also found in the gadolinium analogue of 1
reported by Gatteschi et al.62 The nature of the magnetic
interaction Ln−NIT has first been considered as dependent
only on the lanthanide and not from structural features. A
systematic investigation supported that antiferromagnetic
interactions were expected for lanthanide ions up to the 4f5
electronic configurations but ferromagnetic interactions for
higher configurations 4f6 to 4f10.81 This was ascribed to a spin
polarization mechanism involving the 5d and 6s orbitals of the
lanthanide.62,81−83 The ferromagnetic behavior of 1 is in
agreement with this mechanism. The antiferromagnetic
behavior of 3 may thus be due to the intermolecular
antiferromagnetic interactions, evidenced in the isomorphous
compound 4, overcoming the Tb(III)−nitroxide ferromagnetic
interaction. Moreover, we have seen that the Tb(III)−nitroxide
bond lengths increase significantly in 3 as compared to 1. This
may contribute to decrease the strength of the ferromagnetic
coupling and increase the effect of the intermolecular
antiferromagnetic interactions, or this may even result in
antiferromagnetic coupling. Indeed our previous works with
lanthanide−nitroxide compounds have shown that the
magnetic interaction may be subject to structural feature, as
was evidenced from unprecedented cases of Gd(III)−nitroxide
antiferromagnetic interactions31 and further in some Gd(III)−
Cu(II) complexes.84
The field dependence of the magnetization for 3 like that for
1 does not reach saturation at 50 000 Oe, 6.7 μB (1)57 and 4.9
μB for (3), far from the theoretical value (8.87 μB). This is
ascribed to the strong anisotropy of the Tb(III) ion.
Accordingly, the ac measurements show an out-of-phase
frequency dependence for 3, in agreement with a single-
molecule magnet behavior as found for 1. However, the energy
barrier for 3 is smaller than found for 1.51 The χ″ vs T without
an external field (Figure 4) of 1 exhibits maxima values above 2
K, with an energy barrier of about 18 K, while 3 shows only
frequency-dependence and no maxima values. At higher applied
magnetic field 3 exhibits maxima values like for 1. The energy
barrier of 1 is about 27 K and about 36 K at 1000 and 3500 Oe,
respectively, while the energy barrier of 3 is about 21 K and
about 27 K at 1000 and 3500 Oe, respectively. These
differences may have several origins. First this may be the
difference in crystal packing, which contributes to weak
intermolecular interactions, as well the Tb(III)−nitroxide
bond length, which contributes to weaken the strength of the
interaction. Another origin may be the differences in the
geometry of the square antiprism of lanthanides ions in 1 and 3.
Indeed, studies on bis(phthalocyanine)lanthanide(III) deriva-
tives have shown that the relaxation speed was slower when the
geometry of the metal center was close to the ideal D4d
antiprismatic geometry; that is, the distortion between the
two regular squares was 45°.85 For 1, the two faces
O1N3O7O8 and O3O4O6O5 are distorted squares whose
average torsion angle is approximately 38.5°. For 3, the two
O5N4O7O8 and O9O10O11O12 are distorted squares whose
average torsion angle is approximately 35°. 1 has a geometry
closer to the ideal D4d geometry than 3, consistent with a higher
energy barrier.
Radical lanthanide complexes have interesting luminescence
properties. Several studies concerning visible luminescence of
uncoordinated and coordinated nitroxides were re-
ported32,33,78,86−88 The electronic structure of nitronyl nitro-
xide radicals has been probed through luminescence spectros-
copy, most recently in solution, and discussed in terms of
molecular orbitals calculated with DFT theory.40 In the
following, we focus on the vibronic structure and its
temperature variation observed for the title compounds. The
luminescence spectra at variable temperature for 1, 2, 3, and 4
show an unusual decrease of the energy difference between
luminescence maxima as the temperature increases. This effect
I
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 7. Parameter Values Used to Calculate the Luminescence Spectra for Uncoordinated NIT2Py Radical and for Complexes
1, 2, 3, and 4 as the Temperature Increases
T (K)
E00 (cm−1
)
Γ (cm−1
)
νk1 (cm−1
)
νk2 (cm−1
)
νk3 (cm−1
)
Δk1
Δk2
Δk3
n/a
NIT2Py
80
80
n/a
14 620
14 580
14 630
14 580
14 620
14 600
14 560
14 600
14 570
14 740
14 700
14 675
14 630
14 640
14 640
14 590
14 590
14 510
14 460
14 450
14 400
14 380
14 380
14 350
14 350
14 350
14 340
14 340
14 310
14 505
14 480
14 430
14 410
14 400
14 380
14 350
14 340
14 280
n/a
200
215
245
250
255
260
260
260
300
190
255
260
290
300
290
305
325
315
200
215
225
240
255
270
280
290
300
320
340
225
250
270
280
300
320
325
325
340
1469
615.8
n/a
1.45
1.57
1.56
1.58
1.60
1.60
1.60
1.60
1.55
1.60
1.50
1.70
1.60
1.68
1.65
1.58
1.65
1.68
1.65
1.20
1.15
1.17
1.17
1.17
1.17
1.18
1.18
1.24
1.24
1.24
1.23
1.28
1.28
1.27
1.25
1.32
1.25
1.37
1.37
1.4
1
1472.1
1471.8
1471.1
1471.1
1470.3
1469.2
1469.2
1468.5
1467.5
1473.1
1472.1
1471.0
1469.7
1469.3
1469.2
1468.6
1467.8
1467.3
1361.0
1361.0
1365.0
1365.0
1365.0
1365.0
1365.0
1365.0
1365.0
1365.0
1365.0
1446.6
1447.0
1447.0
1447.0
1447.0
1444.6
1447.0
1447.0
1444.6
621.17
621.03
620.63
620.63
620.24
619.67
619.67
619.47
618.00
620.89
621.66
621.00
621.57
620.60
620.47
620.14
620.35
620.25
613.20
613.20
613.20
613.20
613.20
613.20
613.20
613.20
613.20
613.20
613.20
611.64
611.70
611.70
611.70
611.00
613.49
611.70
611.70
613.49
250.88
250.64
250.62
250.62
250.53
250.60
250.60
250.53
249.31
249.22
250.00
249.86
249.67
249.77
250.19
250.11
250.19
250.64
248.00
248.00
248.00
248.00
248.00
248.00
248.00
248.00
248.00
248.00
248.00
258.36
258.00
258.00
258.00
258.00
258.00
258.00
258.00
258.36
1.00
1.00
1.10
1.10
1.30
1.20
1.20
1.30
1.20
1.00
0.90
0.90
0.90
0.90
1.00
1.15
1.00
1.05
1.00
1.00
1.00
1.00
1.00
1.00
1.05
1.05
1.07
1.07
1.07
1.00
1.00
1.00
1.05
1.03
1.10
1.05
1.15
1.15
0.90
0.80
0.90
0.90
0.80
0.90
0.90
1.00
1.00
0.75
1.00
0.90
0.90
1.00
0.90
0.75
0.75
0.95
0.00
0.40
0.50
0.50
0.50
0.80
0.80
0.80
0.80
0.80
1.00
1.00
1.00
1.00
0.85
0.85
0.90
0.90
1.10
1.00
120
160
180
200
220
240
270
300
80
2
120
160
180
200
220
240
270
300
78
3
103
123
143
163
183
203
223
243
273
293
78
4
120
160
180
200
220
240
270
293
is significant, as the energy difference decreases from
approximately 1500 cm−1 to approximately 1000 cm−1, a
surprising change of 500 cm−1. We analyze this unusual effect
through calculated luminescence spectra based on harmonic
potential energy surfaces, a well-established approach using the
time-dependent theory of spectroscopy.58
All parameter values used for these calculations are
summarized in Table 7. Vibrational frequencies used to
calculate the spectra are determined from Raman spectra and
do not show the variation by 500 cm−1 observed for energy
differences in luminescence spectra measured at low and high
temperatures, as illustrated in detail in Figure 7. The calculated
luminescence spectra are in very good agreement with the
experimental spectra especially at temperatures below 240 K, as
shown in Figure 6.
accurately reproduced for all complexes especially at temper-
atures lower than 240 K. It is easier to reproduce luminescence
spectra for 1 and 2 than for 3 and 4. At higher temperatures, it
is difficult to reproduce experimental luminescence spectra due
to the poor resolution. The calculated E00 electronic origin
parameter shows a decrease for 2, 3, and 4 and stays constant
for 1 in the temperature range studied. The E00 value at 80 K is
approximately 14 600, 14 700, 14 500, and 14 500 cm−1, higher
by 200, 300, 100, and 100 cm−1 for complexes 1, 2, 3, and 4,
respectively, than 14 400 cm−1 for the uncoordinated radical.
Offsets along the normal coordinates are similar for all
complexes with no significant variations as the temperature
changes. Slight variations of less than 10% confirm that there
are no strong structural modifications in the complexes, as
supported by the information obtained from Raman spectra.
Several parameters are adjusted to reproduce the exper-
imental spectra, namely, the offsets Δ along each normal
coordinate and the width Γ of each vibronic line, a value
identified as a damping factor in the context of time-dependent
The comparison between experimental and calculated ΔEmax
values in the temperature range from 80 to 300 K is presented
for 1 and 3 in Figure 7a and for 2 and 4 in Figure 7b. The ΔE
variation shows that luminescence experimental spectra are
J
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 6. Experimental (solid line) and calculated (dotted line) luminescence spectra at variable temperature (80−300 K) of (a) 1 (top) and 3
(bottom) and (b) 2 (top) and 4 (bottom).
The energy differences ΔEmax in Figure 7 do not correspond
to experimental Raman frequencies, a situation reminiscent of
the missing mode effect (MIME) discussed in luminescence
spectra of many transition metal compounds.89−91 The MIME
frequency observed in the luminescence spectra of these
compounds is due to emitting-state distortions along several
normal coordinates with different frequencies and is easily
rationalized in the time-dependent theory by exploring the
products of autocorrelation functions along a single normal
coordinate. A formula has been published to calculate the
MIME frequency from parameters used to calculate a spectrum,
such as the values in Table 7 for the title compounds.92 Using
these values, an increase of the calculated MIME frequency is
obtained with increasing temperature, shown in Figure S15, in
contrast to the decrease illustrated in Figure 7. The effect
observed here is therefore different from the well-established
MIME. It involves the magnitude of the widths of vibronic lines
in the calculated spectra (parameter Γ), which become less
resolved as the temperature increases, ultimately leading to an
unresolved envelope with a single maximum, determined by the
short-time dynamics. This is illustrated in Figure S16 for
spectra calculated only with the highest frequency, 1472 cm−1,
mode. For short progressions involving high-frequency modes,
a situation typical for ligand-centered transitions, this regime
leads to vibronic maxima that move closer together as Γ
increases, as shown in Figure S17. The autocorrelation
functions in Figure S18 show that the vibrational recurrence
●
Figure 7. Variations of the ΔE parameter for the experimental ( )
disappears for the range of Γ values where the maxima shift
most visibly, corresponding to the situation reported here for
the title compounds. This effect is illustrated and analyzed for
the first time for a series of experimental spectra, documenting
the rich variety of physical properties of lanthanide−radical
complexes.
▲
and calculated ( ) luminescence spectra as the temperature increases
for 1 and 3 (a) and 2 and 4 (b).
theory, a parameter determining the resolution of the calculated
spectra.
K
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(11) Hewitt, I. J.; Tang, J.; Madhu, N. T.; Anson, C. E.; Lan, Y.;
Luzon, J.; Etienne, M.; Sessoli, R.; Powell, A. K. Angew. Chem., Int. Ed.
2010, 49, 6352.
(12) Blagg, R. J.; Muryn, C. A.; McInnes, E. J. L.; Tuna, F.;
Winpenny, R. E. P. Angew. Chem., Int. Ed. 2011, 50, 6530.
(13) Takamatsu, S.; Ishikawa, T.; Koshihara, S. Y.; Ishikawa, N. Inorg.
Chem. 2007, 46, 7250.
(14) Coutinho, J. T.; Antunes, M. A.; Pereira, L. C. J.; Bolvin, H.;
Marcalo, J.; Mazzanti, M.; Almeida, M. Dalton Trans. 2012, 41, 13568.
(15) Mazzanti, M. Nat. Chem. 2011, 3, 426.
(16) Antunes, M. A.; Pereira, L. C. J.; Santos, I. C.; Mazzanti, M.;
Marcalo, J.; Almeida, M. Inorg. Chem. 2011, 50, 9915.
(17) Mills, D. P.; Moro, F.; McMaster, J.; van Slageren, J.; Lewis, W.;
Blake, A. J.; Liddle, S. T. Nat. Chem. 2011, 3, 454.
(18) Rinehart, J. D.; Long, J. R. Dalton Trans. 2012, 41, 13572.
(19) Rinehart, J. D.; Long, J. R. J. Am. Chem. Soc. 2009, 131, 12558.
(20) Rinehart, J. D.; Harris, T. D.; Kozimor, S. A.; Bartlett, B. M.;
Long, J. R. Inorg. Chem. 2009, 48, 3382.
CONCLUSION
■
We have prepared four complexes of terbium(III) or
yttrium(III) with a nitronyl nitroxide organic radical and
hexafluoroacetylacetonate or acetylacetonate ligands, and we
have compared the properties and structures obtained with the
two ancillary ligands. Their structural and magnetic properties
as well as their luminescence and Raman spectra are compared.
The terbium(III) complexes both exhibit single-molecule
magnet behavior. For all four complexes the vibronic structure
of the luminescence bands cannot be assigned by single
vibrational frequencies, and a variation of the vibronic interval
occurs with temperature. This effect has to be described with
the quantitative variation of all parameters. To the best of our
knowledge, it is the first time that such a behavior is reported
for lanthanide−radical complexes.
ASSOCIATED CONTENT
* Supporting Information
Additional X-ray crystal structures and magnetic ac and dc
susceptibilities; Raman and calculated luminescence spectra;
Raman shift and peak intensities; crystallographic data in CIF
format. This material is available free of charge via the Internet
at http://pubs.acs.org.
■
(21) Feng, X. W.; Liu, J. J.; Harris, T. D.; Hill, S.; Long, J. R. J. Am.
Chem. Soc. 2012, 134, 7521.
(22) Petit, S.; Neugebauer, P.; Pilet, G.; Chastanet, G.; Barra, A. L.;
Antunes, A. B.; Wernsdorfer, W.; Luneau, D. Inorg. Chem. 2012, 51,
6645.
S
(23) Ruamps, R.; Batchelor, L. J.; Maurice, R.; Gogoi, N.; Jimenez-
Lozano, P.; Guihery, N.; de Graaf, C.; Barra, A. L.; Sutter, J. P.; Mallah,
T. Chem.Eur. J. 2013, 19, 950.
(24) Tancini, E.; Rodriguez-Douton, M. J.; Sorace, L.; Barra, A. L.;
Sessoli, R.; Cornia, A. Chem.Eur. J. 2010, 16, 10482.
(25) Borta, A.; Gillon, B.; Gukasov, A.; Cousson, A.; Luneau, D.;
Jeanneau, E.; Ciumacov, I.; Sakiyama, H.; Tone, K.; Mikuriya, M. Phys.
Rev. B 2011, 83.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: christian.reber@umontreal.ca. Fax: +1 514 343 7586.
Tel: +1 514 343 7332.
*E-mail: luneau@univ-lyon1.fr. Fax: +33 4 72 43 11 60. Tel:
+33 4 72 43 14 18.
■
(26) Damjanovic, M.; Katoh, K.; Yamashita, M.; Enders, M. J. Am.
Chem. Soc. 2013, 135, 14349.
(27) Baumgartel, N.; Flambard, A.; Kohler, F. H.; Lescouezec, R.
Inorg. Chem. 2013, 52, 12634.
Notes
The authors declare no competing financial interest.
(28) Beaulac, R.; Bussier
J. Chem. 2003, 27, 1200.
́
(29) Bussiere, G.; Beaulac, R.; Belisle, H.; Lescop, C.; Luneau, D.;
̀
e, G.; Reber, C.; Lescop, C.; Luneau, D. New
ACKNOWLEDGMENTS
The authors thank Francine Bel
in crystal structure determination. A.L. held a CMIRA
■
̀
́
anger Gariep
́
y for her support
Rey, P.; Reber, C. Top. Curr. Chem. 2004, 241, 97.
(30) Lescop, C.; Belorizky, E.; Luneau, D.; Rey, P. Inorg. Chem. 2002,
41, 3375.
Explora’doc 2012 fellowship from la Reg
The collaborative research was made possible by a grant from
the Conseil Franco-Quebecois de Cooperation Universitaire
(CCFQ, administrated by FQRNT, Province of Quebec), from
la Region Rhone-Alpes through the CMIRA COOPERA 2013
́ ̂
ion Rhone-Alpes.
(31) Lescop, C.; Luneau, D.; Belorizky, E.; Fries, P.; Guillot, M.; Rey,
P. Inorg. Chem. 1999, 38, 5472.
́
́
́
́
(32) Lescop, C.; Luneau, D.; Bussier
Chem. 2000, 39, 3740.
(33) Lescop, C.; Luneau, D.; Rey, P.; Bussier
Chem. 2002, 41, 5566.
e, G.; Triest, M.; Reber, C. Inorg.
̀
́
̂
̀
e, G.; Reber, C. Inorg.
program, and by grants from the Natural Sciences and
Engineering Research Council (Canada).
(34) Petit, S.; Baril-Robert, F.; Pilet, G.; Reber, C.; Luneau, D. Dalton
Trans. 2009, 6809.
REFERENCES
■
(35) Cosquer, G.; Pointillart, F.; Jung, J.; Le Guennic, B.; Golhen, S.;
Cador, O.; Guyot, Y.; Brenier, A.; Maury, O.; Ouahab, L. Eur. J. Inorg.
Chem. 2014, 2014, 69.
(36) Pointillart, F.; Le Guennic, B.; Cauchy, T.; Golhen, S.; Cador,
O.; Maury, O.; Ouahab, L. Inorg. Chem. 2013, 52, 5978.
(37) Pointillart, F.; Le Guennic, B.; Golhen, S.; Cador, O.; Maury,
O.; Ouahab, L. Chem. Commun. 2013, 49, 615.
(38) Boulon, M. E.; Cucinotta, G.; Luzon, J.; Degl’Innocenti, C.;
Perfetti, M.; Bernot, K.; Calvez, G.; Caneschi, A.; Sessoli, R. Angew.
Chem., Int. Ed. 2013, 52, 350.
(39) Cucinotta, G.; Perfetti, M.; Luzon, J.; Etienne, M.; Car, P. E.;
Caneschi, A.; Calvez, G.; Bernot, K.; Sessoli, R. Angew. Chem., Int. Ed.
2012, 51, 1606.
(1) 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.
(2) Aromi, G.; Brechin, E. K. Struct. Bonding 2006, 122, 1.
(3) Christou, G. Polyhedron 2005, 24, 2065.
(4) Ishikawa, N.; Sugita, M.; Ishikawa, T.; Koshihara, S.; Kaizu, Y. J.
Am. Chem. Soc. 2003, 125, 8694.
(5) Ishikawa, N.; Satoshi, O.; Kaizu, Y. Angew. Chem., Int. Ed. 2005,
44, 731.
(6) Aronica, C.; Pilet, G.; Chastanet, G.; Wernsdorfer, W.; Jacquot, J.
F.; Luneau, D. Angew. Chem., Int. Ed. 2006, 45, 4659.
(7) 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.
(8) Borta, A.; Jeanneau, E.; Chumakov, Y.; Luneau, D.; Ungur, L.;
Chibotaru, L. F.; Wernsdorfer, W. New J. Chem. 2011, 35, 1270.
(9) Iasco, O.; Novitchi, G.; Jeanneau, E.; Wernsdorfer, W.; Luneau,
D. Inorg. Chem. 2011, 50, 7373.
(40) Tretyakov, E. V.; Plyusnin, V. F.; Suvorova, A. O.; Larionov, S.
V.; Popov, S. A.; Antonova, O. V.; Zueva, E. M.; Stass, D. V.;
Bogomyakov, A. S.; Romanenko, G. V.; Ovcharenko, V. I. J. Lumin.
2014, 148, 33.
(41) Pointillart, F.; Bernot, K.; Poneti, G.; Sessoli, R. Inorg. Chem.
(10) Sessoli, R.; Powell, A. K. Coord. Chem. Rev. 2009, 253, 2328.
2012, 51, 12218.
L
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(42) Caneschi, A.; Gatteschi, D.; Rey, P. Acc. Chem. Res. 1989, 22,
392.
(43) Caneschi, A.; Gatteschi, D.; Rey, P. Prog. Inorg. Chem. 1991, 39,
(79) Guofa, L.; Tongshun, S.; Yongnian, Z. J. Mol. Struct. 1997, 412,
75.
(80) Yamashita, K.; Takeuchi, N.; Taniguchi, H.; Yuyama, S.; Oe, K.;
Mibuka, N.; Suzuki, A.; Mataki, H. J. Lumin. 2009, 129, 526.
(81) Kahn, M. L.; Sutter, J. P.; Golhen, S.; Guionneau, P.; Ouahab,
L.; Kahn, O.; Chasseau, D. J. Am. Chem. Soc. 2000, 122, 9566.
(82) Zhang, C. H.; Sun, N. N.; Zhao, X. Y.; Zang, Y. Y.; Guo, Y. L.
Inorg. Chem. Commun. 2011, 14, 166.
331.
(44) Sutter, J.-P.; Kahn, M. L.; Golhen, S.; Ouahab, L.; Kahn, O.
Chem.−Eur. J. 1998, 4, 571.
(45) Sutter, J.-P.; Kahn, M. L.; Kahn, O. Adv. Mater. (Weinheim, Ger.)
1999, 11, 863.
(83) Xu, J. X.; Ma, Y.; Liao, D. Z.; Xu, G. F.; Tang, J.; Wang, C.;
Zhou, N.; Yan, S. P.; Cheng, P.; Li, L. C. Inorg. Chem. 2009, 48, 8890.
(84) Costes, J. P.; Dahan, F.; Dupuis, A.; Laurent, J. P. Inorg. Chem.
2000, 39, 169.
(85) Wang, H.; Wang, K.; Tao, J.; Jiang, J. Chem. Commun. 2012, 48,
2973.
(86) Godlewska, P.; Gerasymchuk, Y.; Tomachynskii, L.;
Legendziewicz, J.; Hanuza, J. Struct. Chem. 2010, 21, 461.
(87) Tsukahara, Y.; Nakata, H.; Kaizaki, S. Inorg. Chem. 2004, 43,
4383.
(46) Luneau, D.; Rey, P. Coord. Chem. Rev. 2005, 249, 2591.
(47) Vostrikova, K. E.; Luneau, D.; Wernsdorfer, W.; Rey, P.;
Verdaguer, M. J. Am. Chem. Soc. 2000, 122, 718.
(48) Caneschi, A.; Gatteschi, D.; Lalioti, N.; Sangregorio, C.; Sessoli,
R.; Venturi, G.; Vindigni, A.; Rettori, A.; Pini, M. G.; Novak, M. A.
Europhys. Lett. 2002, 58, 771.
(49) Bernot, K.; Pointillart, F.; Rosa, P.; Etienne, M.; Sessoli, R.;
Gatteschi, D. Chem. Commun. 2010, 46, 6458.
(50) Mei, X.-L.; Ma, Y.; Li, L.-C.; Liao, D.-Z. Dalton Trans. 2012, 41,
505.
(51) Wang, X.-L.; Li, L.-C.; Liao, D.-Z. Inorg. Chem. 2010, 49, 4735.
(52) Liu, R. N.; Li, L. C.; Wang, X. L.; Yang, P. P.; Wang, C.; Liao, D.
Z.; Sutter, J. P. Chem. Commun. 2010, 46, 2566.
(53) 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.
(54) Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R. J. Am. Chem.
Soc. 2011, 133, 14236.
(88) Beaulac, R.; Luneau, D.; Reber, C. Chem. Phys. Lett. 2005, 405,
153.
(89) Tutt, L.; Tannor, D.; Heller, E. J.; Zink, J. I. Inorg. Chem. 1982,
21, 3858.
(90) Tutt, L.; Zink, J. I. J. Am. Chem. Soc. 1986, 108, 5830.
(91) Zink, J. I. Coord. Chem. Rev. 2001, 211, 69.
(92) Tutt, L. W.; Zink, J. I.; Heller, E. J. Inorg. Chem. 1987, 26, 2158.
(55) Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R. Nat. Chem.
2011, 3, 538.
(56) Demir, S.; Zadrozny, J. M.; Nippe, M.; Long, J. R. J. Am. Chem.
Soc. 2012, 134, 18546.
(57) Wang, X.; Bao, X.; Xu, P.; Li, L. Eur. J. Inorg. Chem. 2011, 2011,
3586.
(58) Reber, C.; Zink, J. I. J. Chem. Phys. 1992, 96, 2681.
́
(59) Hirel, C.; Vostrikova, K. E.; Pecaut, J.; Ovcharenko, V. I.; Rey, P.
Chem.Eur. J. 2001, 7, 2007.
(60) Osiecki, J. H.; Ullman, E. F. J. Am. Chem. Soc. 1968, 90, 1078.
(61) Richardson, M. F.; Wagner, W. F.; Sand, D. E. J. Inorg. Nucl.
Chem. 1968, 30, 1275.
(62) Benelli, C.; Caneschi, A.; Gatteschi, D.; Pardi, L. Inorg. Chem.
1992, 31, 741.
(63) APEX2, Bruker Molecular Analysis Research Tool; Bruker AXS
Inc.: Madison, WI, USA, 2006.
(64) Sheldrick, G. M. SADABS, Bruker Area Detector Absorption
Corrections; Bruker AXS Inc.: Madison, WI, USA, 1996 and 2004.
(65) Sheldrick, G. Acta Crystallogr. A 2008, 64, 112.
(66) CrysAlisPro, O. D. L., Version 1.171.33.46; CrysAlis171.NET,
2009.
(67) De Meulenaar, J.; Tompa, H. Acta Crystallogr. A 1965, 19, 1014.
(68) Cascarano, G.; Altomare, A.; Giacovazzo, C.; Guagliardi, A.;
Moliterni, A. G. G.; Siliqi, D.; Burla, M. C.; Polidori, G.; Camalli, M.
Acta Crystallogr. A 1996, 52, C.
(69) Watkin, D. J.; Prout, C. K.; Carruthers, J. R.; Betteridge, P. W.
CRYSTALS Issue 11; Chemical Crystallography Laboratory: Oxford,
U.K., 1999.
(70) Farrugia, L. J. Appl. Crystallogr. 1999, 32, 837.
(71) Pascal, P. Ann. Chim. Phys. 1910, 19, 5.
(72) Depanthou, F. L.; Luneau, D.; Laugier, J.; Rey, P. J. Am. Chem.
Soc. 1993, 115, 9095.
(73) Luneau, D.; Risoan, G.; Rey, P.; Grand, A.; Caneschi, A.;
Gatteschi, D.; Laugier, J. Inorg. Chem. 1993, 32, 5616.
(74) Fang, C.; Wu, G. Spectrochim. Acta, Part A 2010, 77, 948.
(75) Xiao, C.; Feng, K.; Mo, Y.; Meng, Q.; Zhang, M.; Wan, M.;
Zhao, J. Chem. Phys. 1998, 237, 73.
(76) Mo, Y.-J.; Li, Y.-C.; Zhao, J.-G.; Fen, K.-A.; Xiao, C.-Y.; Wan,
M.-X. J. Magn. Magn. Mater. 2000, 213, 278.
(77) Tsukahara, Y.; Iino, A.; Yoshida, T.; Suzuki, T.; Kaizaki, S. J.
Chem. Soc., Dalton Trans. 2002, 181.
(78) Rintoul, L.; Micallef, A. S.; Bottle, S. E. Spectrochim. Acta, Part A
2008, 70, 713.
M
dx.doi.org/10.1021/ic500779y | Inorg. Chem. XXXX, XXX, XXX−XXX