Magnetic and photo-physical investigations into Dy^III and Yb^III complexes involving tetrathiafulvalene ligand

Abstract

Three lanthanide-based complexes involving a tetrathiafulvalene derivative (L) in which the lanthanide ion has a pseudo-D_4d symmetry have been reported. One is a dinuclear compound of formula [Dy(hfac)₃(L)]₂ (1) while the two others are isostructural and described as mononuclear complexes of formulae [Ln(tta)₃(L)]·xCH₂Cl₂ (Ln^III = Dy and x = 1.41 (2); Yb and x = 2 (3)). The nuclearity of the species is driven by the nature of the ancillary ligands. Magnetic properties revealed that 1 and 3 behave as single molecule magnets, while 2 does not. The crystal field splitting of the ground multiplet state has been theoretically determined as well as the orientation of the easy axis of the ground M_J state. The results of ab initio calculations are in agreement with the experimental determinations of the anisotropy axis. Irradiation of the lowest-energy charge transfer bands of 3 led to an intense and resolved Yb-centred emission which can be correlated to the magnetic data. Thus 3 can be described as a redox-active luminescent field-induced single-molecule magnet.

---

Introduction

Single Molecule Magnets (SMMs) are fascinating objects which take a preponderant place in both the chemist and physicist communities owing to their potential applications in quantum computing,¹ high-density memory data storage devices² and spintronics.³ One of the actual challenges is to combine, in the same molecule, the SMM property with one or more subsequent physical properties in order to obtain a multifunctional SMM, such as for instance chiral SMM,⁴ ferroelectric SMM⁵ or luminescent SMM.⁶

In this way, the use of lanthanide ions seems to be very promising since they display specific and unique physical and photophysical properties. On one side, they possess a strong magnetic anisotropy and a high magnetic moment, making them ideal candidates for SMM,⁷ and on the other side they display sharp emission lines ranging from the visible to the near infrared (NIR) region arising from f–f transitions.⁸ In order to exploit this potential, lanthanide ions must be coordinated to organic ligands acting as sensitizing antenna in a pre-organized way, resulting in the induction of a crystal (ligand)-field that monitors both the magnetic and luminescence properties.

For several decades, tetrathiafulvalene (TTF)-based ligands have been used for their electron donating ability and so for their electronic conductivity.⁹ Furthermore, their ability to form S⋯S short contacts and π–π stacking makes them attractive candidates as structural agents to organize the tridimensional arrangement in crystals and control the magnetic properties (magnetic isolation, SMM behaviour). In addition, their strong electronic delocalization is responsible for their interesting optical properties and this push–pull chromophore has been demonstrated to efficiently sensitize the lanthanide luminescence by the antenna effect.¹⁰

Considerable efforts have been made to better understand and ultimately predict the magnetic behaviour of lanthanide-based complexes and hence the crystal field effects. To this end, systematic experimental studies explored the magnetic properties of the original structure with the aim of establishing efficient magneto-structural correlations.^10g,11

In parallel, theoretical investigations, based on electrostatic models or high level ab initio calculations, were undertaken to estimate the lanthanide crystal field and the orientation of the magnetic axis in molecular objects.¹²

Recently, some of us reported dinuclear complexes of lanthanide involving the (4,5-bis(2-pyridyl-N-oxidemethylthio)-4′,5′)-ethylenedithiotetrathiafulvene (L¹) or -methyldithiotetrathiafulvene (L²) ligands which displayed either SMM or luminescence behaviour but not both simultaneously.^10g It was demonstrated that a minor chemical change in the nature of the TTF-based ligand induces drastic magnetic modifications since [Dy(hfac)₃(L²)]₂ behaves as a SMM while [Dy(hfac)₃(L¹)]₂ does not. In the present article, we continue the step-by-step study of this ligand family and report the slightly modified 4,5-ethylenedioxy-4′,5′-bis(2-pyridyl-N-oxidemethylthio)tetrathiafulvalene (L) (Scheme 1) ligand featuring the 4,5-ethylendioxy moiety instead of 4,5-ethylenedithio and 4,5-dimethylthio moieties for L¹ and L² respectively. This minor modification allows us to optimize the magnetic properties and to combine both luminescence and SMM behaviours in a single complex.

Scheme 1 Molecular structure of L.

Experimental section

General procedures and materials

The precursors Dy(hfac)₃·2H₂O ¹³ (hfac⁻ = 1,1,1,5,5,5-hexafluoroacetylacetonate anion), Ln(tta)₃·2H₂O ¹⁴ (Ln = Dy^III and Yb^III; tta⁻ = 2-thenoyltrifluoroacetonate anion) and 4,5-ethylenedioxy-4′,5′-bis(cyanoethylthio)tetrathiafulvalene¹⁵ were synthesized following previously reported methods. All other reagents were purchased from Aldrich Co., Ltd and used without further purification.

Synthesis of the ligand 4,5-ethylendioxy-4′,5′-bis(2-pyridyl-N-oxidemethylthio)tetrathiafulvalene (L)

In a three-necked flask under an argon atmosphere, 0.432 g (1 mmol) of 4,5-ethylenedioxy-4′,5′-bis(cyanoethylthio)tetrathiafulvalene^15c was dissolved in 30 mL distilled MeOH. In the dropping funnel, 0.115 g (5 mmol) of metallic sodium was added to 10 mL distilled MeOH and then added dropwise to a solution of 4,5-ethylenedioxy-4′,5′-bis(cyanoethylthio)tetrathiafulvalene under stirring. After the addition was over, the solution was stirred at room temperature for 4 hours. Then, in the dropping funnel, 0.419 g (3.3 mmol) of 2-(chloromethyl)pyridine-1-oxide¹⁶ was dissolved in 20 mL distilled MeOH and added dropwise to the solution and left under stirring overnight. Then 30 mL distilled water was added and stirred for 30 min, followed by the addition of 250 mL CH₂Cl₂. The resulting solution was washed with saturated NaHCO₃ and distilled water. The organic phase was dried with MgSO₄ and the solvent was removed under vacuum. The resulting powder was purified on an Al₂O₃ column with a 99/1 ratio of CH₂Cl₂/MeOH as the eluent. Yield: 200 mg (37%). ¹H NMR (CDCl₃): δ 4.12 (s, 4H), 4.26 (s, 4H), 7.23 (m, 4H), 7.33 (m, 2H), 8.26 (m, 2H). Single crystals were obtained by slow diffusion of n-hexane in a concentrated CH₂Cl₂ solution of L.

Synthesis of complexes 1–3

[Dy(hfac)₃(L)]₂ (1). 30.8 mg of Dy(hfac)₃·2H₂O (0.037 mmol) were dissolved in 10 mL of CH₂Cl₂ and then added to a solution of 10 mL of CH₂Cl₂ containing 20.0 mg of L (0.037 mmol). After 15 minutes of stirring, 25 mL of n-hexane were layered at room temperature in the dark. Slow diffusion leads to red single crystals which are suitable for X-ray studies. Yield: 77 mg (79%). Anal. Calcd (%) for C₇₀H₃₈Dy₂F₃₆N₄O₂₀S₁₂: C 31.73, H 1.44, N 2.12; found: C 31.99, H 1.46, N 2.11.

[Ln(tta)₃(L)]·xCH₂Cl₂ (Ln^III = Dy and x = 1.41 (2); Yb and x = 2 (3)). 0.037 mmol of Ln(tta)₃·2H₂O (32.4 mg for Dy(tta)₃·2H₂O and 32.8 mg for Yb(tta)₃·2H₂O) were dissolved in 10 mL of CH₂Cl₂ and then added to a solution of 10 mL of CH₂Cl₂ containing 20.0 mg of L (0.037 mmol). After 15 minutes of stirring, 25 mL of n-hexane were layered at room temperature in the dark. Slow diffusion leads to red single crystals which are suitable for X-ray studies. Yield: 34 mg (61%) and 39 mg (68%) respectively for 2 and 3. Anal. Calcd (%) for C_45.41H_30.54Cl_2.82DyF₉N₂O₁₀S₉: C 36.67, H 2.05, N 1.88; found: C 37.09, H 2.16, N 1.91. Calcd (%) for C₄₆H₃₂Cl₄YbF₉N₂O₁₀S₉: C 35.68, H 2.07, N 1.81; found: C 35.79, H 2.16, N 1.94.

Crystallography

Single crystals of L and 1 were mounted on a Nonius four circle diffractometer equipped with a CCD camera and a graphite monochromated MoK_α radiation source (λ = 0.71073 Å, T = 293(2) K) while 2 and 3 were mounted on an APEXII Bruker-AXS diffractometer (MoK_α radiation source, λ = 0.71073 Å, T = 150(2) K) for data collection, from the Centre de Diffractométrie (CDIFX), Université de Rennes 1, France. Structures were solved by direct methods using the SIR-97 program and refined by a full matrix least-squares method on F² using the SHELXL-97 program¹⁷ for all the compounds. Crystallographic data are summarized in Table 1. Complete crystal structure results as a CIF file including bond lengths, angles, and atomic coordinates are provided in the ESI.†

Table 1 X-ray crystallographic data for the ligand L and complexes 1–3

Compounds	(L)	[Dy(hfac)3(L)]2 (1)
Formula	C20H16N2O5S6	C70H38Dy2F36N4O20S12
M/g mol^−1	556.7	2705.8
Crystal system	Monoclinic	Triclinic
Space group	P21/a (no. 14)	P1̄ (no. 2)
Cell parameters	a = 13.2244(3) Å	a = 12.4879(3) Å
	b = 12.1843(4) Å	b = 13.3816(4) Å
	c = 15.3079(4) Å	c = 15.6598(6) Å
	β = 107.908(2)°	α = 100.987(2)°
		β = 90.592(2)°
		γ = 107.459(2)°
Volume/Å^3	2347.06(11)	2444.19(13)
Cell formula units	4	2
T/K	293(2)	293(2)
Diffraction reflection	5.92 ≤ 2θ ≤ 54.20	2.66 ≤ 2θ ≤ 54.18
ρ calc/g cm^−3	1.575	1.799
μ/mm^−1	0.619	1.912
Number of reflections	27 589	18 277
Independent reflections	5151	10 729
F o ^2 > 2σ(Fo)^2	3539	7750
Number of variables	298	689
R int, R1, wR2	0.0625, 0.0509, 0.1327	0.0316, 0.0481, 0.1229
Compounds	[Dy(tta)3(L)]·1.41CH2Cl2 (2)	[Yb(tta)3(L)]·2CH2Cl2 (3)
Formula	C45.41H30.54Cl2.82DyF9N2O10S9	C46H32Cl4YbF9N2O10S9
M/g mol^−1	1486.2	1547.1
Crystal system	Triclinic	Triclinic
Space group	P1̄ (no. 2)	P1̄ (no. 2)
Cell parameters	a = 10.4634(8) Å	a = 10.4473(3) Å
	b = 12.1092(10) Å	b = 12.0946(4) Å
	c = 23.9781(18) Å	c = 23.9969(7) Å
	α = 77.004(2)°	α = 76.5010(10)°
	β = 86.822(2)°	β = 86.0280(10)°
	γ = 74.808(2)°	γ = 74.7470(10)°
Volume/Å^3	2856.8(4)	2844.4(15)
Cell formula units	2	2
T/K	150(2)	150(2)
Diffraction reflection	1.74 ≤ 2θ ≤ 55.08	1.74 ≤ 2θ ≤ 55.00
ρ calc/g cm^−3	1.728	1.806
μ/mm^−1	1.853	2.246
Number of reflections	29 912	39 778
Independent reflections	12 500	12 653
F o ^2 > 2σ(Fo)^2	7773	11 826
Number of variables	711	732
R int, R1, ωR2	0.0798, 0.0882, 0.2008	0.0219, 0.0387, 0.1033

---

Physical measurements

The elementary analyses of the compounds were performed at the Centre Régional de Mesures Physiques de l'Ouest, Rennes. ¹H NMR spectra were recorded on a Bruker Ascend 400 spectrometer. Chemical shifts are reported in parts per million referenced to TMS for ¹H NMR. Cyclic voltammetry was carried out in a CH₂Cl₂ solution, containing 0.1 M N(C₄H₉)₄PF₆ as the supporting electrolyte. Voltammograms were recorded at 100 mV s⁻¹ at a platinum disk electrode. The potentials were measured versus a saturated calomel electrode (SCE) (E_Fc/Fc⁺ = 0.45 V). Absorption spectra were recorded on a Varian Cary 5000 UV-visible-NIR spectrometer equipped with an integration sphere. The luminescence spectra were recorded using a Horiba-Jobin Yvon Fluorolog-3® spectrofluorimeter equipped with a three-slit double grating excitation and emission monochromator with dispersions of 2.1 nm mm⁻¹ (1200 grooves per mm). The steady-state luminescence was excited by unpolarized light from a 450 W xenon CW lamp and detected at an angle of 90° for diluted solution measurements or at 22.5° for solid state measurements (front face detection) by a red-sensitive Hamamatsu R928 photomultiplier tube. Spectra were reference corrected for both the excitation source light intensity variation (lamp and grating) and the emission spectral response (detector and grating). Near infrared spectra were recorded at an angle of 45° using a liquid nitrogen cooled, solid indium/gallium/arsenic detector (850–1600 nm). The dc magnetic susceptibility measurements were performed on a solid polycrystalline sample with a Quantum Design MPMS-XL SQUID magnetometer between 2 and 300 K in an applied magnetic field of 2 kOe in the 2–20 K temperature range and 10 kOe between 20 and 300 K for 3. For 1 and 2, 0.2 kOe is applied between 2 and 20 K, 2 kOe between 20 and 80 K and 10 kOe above 80 K. These measurements were all corrected for the diamagnetic contribution as calculated with Pascal's constants.

Computational details

DFT geometry optimizations and TD-DFT excitation energy calculations of the ligand L were carried out using the Gaussian 09 (revision A.02) package¹⁸ employing the PBE0 hybrid functional.¹⁹ All atoms were described with the SVP basis sets.²⁰ The first 50 monoelectronic excitations were calculated. In all steps, a modeling of bulk solvent effects (solvent = dichloromethane) was included through the Polarizable Continuum Model (PCM),²¹ using a linear-response non-equilibrium approach for the TD-DFT step.²² Molecular orbitals were sketched using the Gabedit graphical interface.²³

Wavefunction-based calculations were carried out on model structures of the Ln^III-based complexes 1–3 (see below) by using the SA-CASSCF/RASSI-SO approach, as implemented in the MOLCAS quantum chemistry package (version 8.0).²⁴ In this approach, the relativistic effects are treated in two steps on the basis of the Douglas–Kroll Hamiltonian. First, the scalar terms were included in the basis-set generation and were used to determine the spin-free wavefunctions and energies in the complete active space self-consistent field (CASSCF) method.²⁵ Next, spin–orbit coupling was added within the restricted-active-space-state-interaction (RASSI-SO) method, which uses the spin-free wavefunctions as basis states.²⁶ The resulting wavefunctions and energies are used to compute the magnetic properties and g-tensors of the lowest states from the energy spectrum by using the pseudo-spin S = 1/2 formalism in the SINGLE-ANISO routine.²⁷ Cholesky decomposition of the bielectronic integrals was employed to save disk space and speed up the calculations.²⁸ The atomic positions were extracted from the X-ray crystal structures. For the three complexes, only the modified TTF ligand L was simplified by suppressing the organic moieties beyond the central C=C bond. Moreover, in the case of the dimer 1, only one half of the molecule was considered. For 1 and 2, the active space of the self-consistent field (CASSCF) method consisted of nine 4f electrons of the Dy^III ion spanning seven 4f orbitals, i.e. CAS(9,7)SCF. State-averaged CASSCF calculations were performed for all of the sextets (21 roots), all of the quadruplets (224 roots), and 300 out of the 490 doublets (due to software limitations) of the Dy^III ion. Twenty-one sextets, 128 quadruplets, and 107 doublets were mixed through spin–orbit coupling in RASSI-SO. All atoms were described by ANO-RCC basis sets.²⁹ The following contractions were used: [8s7p4d3f2g1h] for Dy, [4s3p2d] for the O directly coordinated to Dy, [3s2p] for other O atoms; [3s2p1d] for the N atoms; [4s3p] for the S atoms, [3s2p] for the C and F atoms and [2s] for the H atoms. A similar procedure was then applied to the Yb^III complex 3. The active space consisted of 13 4f electrons in seven 4f orbitals, i.e. CAS(13,7)SCF. State-averaged CASSCF calculations were performed for the seven doublets of the Yb^III ion, all of which were included in the RASSI-SO calculation. The effects of dynamical correlation were treated by means of a CASPT2 treatment on top of the spin-free wavefunctions. Similar basis set contractions were used except for Yb, which was described at the [8s7p4d3f2g] level.

Results and discussion

Synthesis and crystal structures

The ligand L (Scheme 1) was synthesized in order to study the influence of the ethylenedioxy (EDO) functionalization on the structural and magnetic properties. First, L was subjected to coordination with the Dy(hfac)₃·2H₂O precursor and the resulting dinuclear complex of formula [Dy(hfac)₃(L)]₂ (1) was compared to two previously published similar compounds.^10c Then the hfac⁻ ancillary ligands were substituted with the tta⁻ ones leading surprisingly to the formation of new mononuclear complexes of formula [Ln(tta)₃(L)]·xCH₂Cl₂ (Ln^III = Dy and x = 1.41 (2); Yb and x = 2 (3)) and not to analogues of 1.

L crystallized in the P2₁/a monoclinic space (Table 1). The asymmetric unit is composed of one molecule of L and one water molecule (Fig. 1). The central C5–C6 bond length (1.378(4) Å) confirms the neutral form of the ligand. Each 2-pyridine-N-oxide-methylthio arm is oriented to opposite sides with respect to the plane formed by the TTF core. The two N-oxide groups realize two hydrogen bonds (O1–O1w = 2.766(6) Å and O2–O1w = 2.783(6) Å) with the water molecule (Fig. 2). The L donors are stacked along the b axis with short S3⋯S1 contacts (3.577(2) Å).

Fig. 1 ORTEP view of the ligand L. Thermal ellipsoids are drawn at 30% probability. Hydrogen atoms are omitted for clarity.

Fig. 2 Crystal packing of L highlighting the intermolecular interactions.

[Dy(hfac)₃(L)]₂. 1 crystallizes in the P1̄ (no. 2) triclinic space group (Table 1) and the asymmetric unit is composed of one half of the dinuclear unit (Fig. 3).

Fig. 3 ORTEP view of the dinuclear complex [Dy(hfac)₃(L)]₂ (1). Thermal ellipsoids are drawn at 30% probability. Hydrogen atoms are omitted for clarity.

The value of the mean Dy–O_hfac distances is 2.359(4) Å (Table 2) that is comparable to what is observed in the previously reported dinuclear complexes involving the L¹ and L² ligands (2.340 Å).^10g The arrangement of the ligands leads to a distorted square antiprism (D_4d symmetry) coordination polyhedron around the Dy^III ion. This distortion is visualized by continuous shape measures performed with SHAPE 2.1 (Table 3).³⁰ The L donors form dimers through S3⋯S5 short contacts (3.691(3) Å) (Fig. 4). The shortest intra- and intermolecular Dy–Dy distances have been found to be equal to 9.314(5) Å and 9.039(5) Å, respectively.

Fig. 4 Crystal packing of 1 highlighting the intermolecular interaction between L ligands and the formation of the dimers of L.

Table 2 Selected bond lengths Å for complexes 1–3

Compounds	1	2	3
Ln1–O1	2.345(3)	2.334(7)	2.295(3)
Ln1–O2	2.309(4)	2.349(7)	2.308(3)
Ln1–O3	2.380(4)	2.406(7)	2.361(3)
Ln1–O4	2.402(4)	2.376(7)	2.326(3)
Ln1–O5	2.370(4)	2.339(7)	2.299(3)
Ln1–O6	2.323(4)	2.327(7)	2.286(3)
Ln1–O7	2.335(4)	2.375(7)	2.332(3)
Ln1–O8	2.404(4)	2.362(7)	2.316(3)

---

Table 3 SHAPE analysis of the coordination polyhedron around the lanthanide ions in complexes 1–3

	CShMSAPR-8 (square antiprism) D4d	CShMBTPR-8 (biaugmented trigonal prism) C2v	CShMTDD-8 (triangular dodecahedron) D2d
1	0.484	1.564	1.373
2	0.410	2.418	2.134
3	0.363	2.362	2.124

---

[Ln(tta)₃(L)]₂ (Ln = Dy^III (2) and Yb^III (3)). Both compounds are isostructural and the description is given for 2 while the values for 3 are given in brackets. The nature of the β-diketonate of the metal precursor has been changed from hfac⁻ for 1 to tta⁻ for 2 and 3. In several examples, this modification leads to a change in the crystal packing but the molecular structure remains very similar.^31,32 In the present case, the molecule obtained from Ln(tta)₃·2H₂O is totally different than the one obtained from Dy(hfac)₃·2H₂O. 2 and 3 crystallize in the P1̄ (no. 2) triclinic space group (Table 1) and the asymmetric unit is composed of one mononuclear unit (Fig. 5 and S1†) and two dichloromethane molecules of crystallization.

Fig. 5 ORTEP view of the mononuclear complex [Dy(tta)₃(L)] (2). Thermal ellipsoids are drawn at 30% probability. Hydrogen atoms and dichloromethane molecules are omitted for clarity.

Using the Ln(tta)₃·2H₂O metal precursor, the two pyridine-N-oxide arms of one ligand are coordinated to the same lanthanide ion. The value of the mean Dy–O_tta [Yb–O_tta] distances is 2.359(7) [2.315(3)] Å (Table 2). The arrangement of the ligands leads to a distorted square antiprism (D_4d symmetry) coordination polyhedron around the Ln^III ion. This distortion is visualized by continuous shape measures performed with SHAPE 2.1 (Table 3).³⁰ The L donors are “head-to-tail” stacked to form a regular pseudo-one-dimensional network of L. Two kinds of S⋯S contacts are identified between (i) the TTF cores (S4⋯S4 = 4.256(3) Å [4.223(4) Å]) and (ii) the thiophene ring and TTF core (S2⋯S9 = 4.067(3) Å [4.141(4) Å]) to form a mono-dimensional organic network along the a axis (Fig. 6). The shortest intermolecular Ln–Ln distances have been found to be equal to 9.789(7) [9.747(5)] Å.

Fig. 6 Crystal packing of 2 highlighting the formation of the mono-dimensional network of L.

Electrochemical properties

The cyclic voltammetry of L and the related complexes 1–3 is investigated to determine their redox properties. The values of the oxidation potentials are listed in Table 4. The cyclic voltammograms show two mono-electronic oxidations which correspond to the formation of a TTF radical cation and a dication, respectively (Fig. S2†). Upon coordination of the electron attracting Ln(hfac)₃ or Ln(tta)₃ moieties, both E¹_1/2 and E²_1/2 are respectively slightly cathodically and anodically shifted compared with the potentials of the free ligands. The electrochemical properties attest the redox-activity of both L and complexes.

Table 4 Oxidation potentials (V vs. SCE, nBu₄NPF₆, 0.1 M in CH₂Cl₂ at 100 mV s⁻¹) of the ligand L and complexes 1–3

	E ^11/2/V		E ^21/2/V
	^Ox E ^11/2	^red E ^11/2	^Ox E ^21/2	^red E ^21/2
L	0.57	0.36	0.93	0.76
1	0.52	0.40	0.96	0.81
2	0.51	0.36	0.99	0.83
3	0.50	0.33	0.96	0.77

---

Magnetic properties

The static and dynamic magnetic properties of the three compounds have been measured. The χ_MT vs. T plots are shown in Fig. 7 while the magnetization curves at 2 K are shown in Fig. S3† (it must be mentioned that the data for 1 have been divided by two to fit with one Dy^III per chemical unit). At room temperature the χ_MT are close to the expected values for isolated Dy(III) (⁶H_15/2: 14.17 cm³ K mol⁻¹) and Yb(III) (²F_7/2: 2.57 cm³ K mol⁻¹). All the χ_MT values decrease on cooling according to the thermal variation of the crystal field level population. Nevertheless, the two Dy(III) derivatives do not behave exactly in the same way although the coordination polyhedra look similar. The decrease is less steep for 2 than for 1 with however lower low temperature limits for 2 than for 1 (11.2 cm³ K mol⁻¹ for 1 and 10.34 cm³ K mol⁻¹ for 2). Differences are also visible in the magnetization curve at 2 K (Fig. S3†): while magnetization of 1 saturates at a value (4.97Nβ) close to that expected for the Ising component (M_J = ±15/2) of the ⁶H_15/2 multiplet ground state, magnetization of 2 linearly increases with the magnetic field at higher values.

Fig. 7 χ _M T vs. T plots for compounds 1 (circles) (divided by two to fit with one Dy^III per chemical unit), 2 (squares) and 3 (triangles) with simulated curves from ab initio calculations.

At this stage we decided to take advantage of the space group P1̄ with one metallic site in the general position for both 1 and 2 to perform rotating single crystal magnetometry (ESI, Fig. S4–S7†) and to compare the orientations of principal magnetic axes with respect to molecular geometry. The principal values of g-tensors, in the frame of the effective spin 1/2 model, are given in Table 5 while the orientations of the principal axis with largest g value are given in Fig. 8. The most striking difference between 1 and 2 is the axiality: 1 is more Ising than 2 which suggests that 1 should be a better SMM than 2. Furthermore the orientation of the most magnetized axis differs significantly: it is oriented toward one oxygen atom of a pyridine-N-oxide moiety which is also coinciding with one of hfac⁻ ligands in 1 whereas in 2, it is oriented between the two oxygen atoms of the two pyridine-N-oxide moieties. Ab initio calculations (see the Computational details section) confirm the large propensity of 1 to behave as a SMM. Indeed, at the SA-CASSCF/RASSI-SO level, both complexes present a predominant M_J = ±15/2 ground state. But due to the larger energy difference between the ground state and the first excited state in 1 than in 2 (80 cm⁻¹vs. 15 cm⁻¹), slighter mixing is observed for 1 than for 2 (Table S1†). This behaviour is also visible in the calculated principal values of the Zeeman tensors with g_x = 0.04, g_y = 0.06, g_z = 19.46 for 1 whereas g_x = 0.77, g_y = 3.59, g_z = 15.63 for 2 (Table 5). Finally, the orientations of the main magnetic anisotropy axes are well reproduced in both Dy^III complexes.

Fig. 8 Representations of the di- and mono-nuclear complexes 1 (top) and 2 (middle) with experimental (orange) and theoretical (green) orientations of the easy magnetic axes at 2 K. Orientations of the magnetic axes for compound 3 are given at the bottom.

Table 5 Experimental and calculated g-tensors principal values in the frame of the effective spin 1/2 model for compounds 1–3

	1		2		3
	Exp.	Calc.	Exp.	Calc.	Exp.	Calc.
g x	1.19	0.04	7.24	0.77	5.84	5.96
g y	18.47	19.46	5.04	3.59	2.00	0.73
g z	3.14	0.06	15.86	15.63	1.77	0.33

---

For compound 3, χ_MT varies from 1.2 cm³ K⁻¹ at 2 K to 2.38 cm³ K mol⁻¹ at room temperature. At 300 K, χ_MT is close to the expected value for the ground state multiplet ²F_7/2 (2.57 cm³ K mol⁻¹). The orientation of the most magnetic axis determined from single crystal rotating magnetometry (ESI, Fig. S8†) is shown in Fig. 8. As for 1 and 2, ab initio calculations were performed on the Yb^III compound 3. However, as usual for Yb^III complexes,^11h,33 dynamical correlation effects had to be taken into account by means of CASPT2 calculations (see the Computational details section). The χ_MT vs. T experimental curve is well reproduced by the calculations (Fig. 7) as well as the orientation of the easy magnetic axes (Fig. 8). The ground state is well separated from the first excited state (234 cm⁻¹) and is mainly composed of M_J = ±5/2 (0.50) but with non-negligible contributions from M_J = ±7/2 (0.34) and M_J = ±5/2 (0.10). As could have been anticipated from dc magnetic data the dynamic susceptibilities of 1 and 2 differ significantly. 2 does not show any out-of-phase component of the ac susceptibility (χ_M′′) whatever the temperature (down to 1.8 K) and whatever the frequency of the oscillating field (between 1 and 1000 Hz) at zero external dc field. On the contrary, χ_M′′ is not null in zero external field for 1 at low temperature. It shows a frequency dependent signal below 7 K in the range 1–1000 Hz (Fig. 9).

Fig. 9 Temperature variations of the ac susceptibility components, χ_M′ and χ_M′′, of compound 1 measured in zero external dc field at various frequencies. Values are given for one metallic site.

The frequency dependences of the two components at various temperatures have been treated using an extended Debye model (see ESI and Table S2†), which allows plotting the thermal variations of the relaxation time τ. It follows an Arrhenius law at high temperature modulated by a thermally independent regime at low temperature (Fig. 10). The complete set of data can be adjusted with the equation (τ⁻¹ = τ₀⁻¹ exp (−Δ/T) + τ_TI⁻¹) with τ being the relaxation time, τ₀ the intrinsic relaxation time, Δ the energy barrier and finally τ_TI the thermally independent relaxation time. The best agreement is obtained with τ₀ = 9 ± 2 × 10⁻⁶ s, Δ = 16 ± 0.8 K and τ_TI = 1.85 ± 0.02 × 10⁻⁴ s.

Fig. 10 Temperature variations of the relaxation time of the magnetization measured at zero external field (empty circles) and at 1 kOe (full circles) for compound 1. Full red lines are the best fitted curves (see the text).

The application of an external dc field dramatically slows down the relaxation process in the thermally independent regime (Fig. S9†). The application of a moderate and constant external field removes the degeneracy of Kramer's doublet with no more direct relaxation between these two states. The most efficient field which is the one for which the relaxation is slowest is found close to 1 kOe. Then the frequency dependences of χ_M′ and χ_M′′ at various temperature but recorded at 1 kOe have been treated using the extended Debye model. The relaxation time obeys now the Arrhenius law (τ⁻¹ = τ₀⁻¹ exp(−Δ/T)) with τ₀ = 1 ± 0.12 × 10⁻⁵ s, Δ = 21 ± 0.3 K. These values are of the same order of magnitude as in zero external field.

The normalized Cole–Cole plots (χ_M′′/χ_Tvs. χ_M′/χ_T) at selected temperatures between 1.8 and 7.0 K are depicted in Fig. 11. They are flattened semicircles and the α values (Table S3†) indicate that a single relaxation time is mainly involved in the present relaxation process.

Fig. 11 Normalized Cole–Cole plots for 1 at 1 kOe between 1.8 and 7.0 K.

Compound 3 does not show any out-of-phase signal in zero external dc field whatever the temperature and the available oscillating field frequency in agreement with the calculated ground state (mainly M_J = ±5/2). However, at 1 kOe a frequency dependent out-of-phase signal shows up at lower temperatures than 4 K (Fig. 12) that may be due to the presence of a non-negligible contribution M_J = ±7/2 in the ground state wavefunction.

Fig. 12 Frequency variations of the ac susceptibility components, χ_M′ and χ_M′′, of compound 3 measured at 1 kOe at various temperatures.

The extended Debye model (Table S4†) allows to extract the temperature dependence of the relaxation which can then be treated in the frame of a thermally activated process only (τ⁻¹ = τ₀⁻¹ exp(−Δ/T) with τ₀ = 1.9 ± 0.2 × 10⁻⁵ s and Δ = 6 ± 0.3 K (Fig. 13).

Fig. 13 Temperature variations of the relaxation time of the magnetization measured at 1 kOe for compound 3. Full red line is the best fitted curve (see the text).

The normalized Cole–Cole plots (χ_M′′/χ_Tvs. χ_M′/χ_T) at selected temperatures between 1.8 and 3.0 K are flattened semicircles due to α values close to 0.05 (Fig. 14 and Table S4†). This value indicates that a single relaxation time is mainly involved in the present relaxation process independently of the temperature. They almost all collapse into one semicircle which intercepts the x axis near the origin. The fast relaxing part of the magnetization is very small and is not related to the distribution of the relaxation time.

Fig. 14 Normalized Cole–Cole plots for 3 at 1 kOe between 1.8 and 3.0 K.

Photo-physical properties

Absorption properties. The UV-visible absorption properties of L have been studied in a CH₂Cl₂ solution (Fig. 15a). Rationalization by TD-DFT calculations was performed following a computational strategy already used successfully on TTF-based systems.³⁴ The molecular orbital diagram and UV-visible absorption spectra were determined for L (Fig. 15b, 16 and Table 6). The experimental absorption curve of L has been decomposed into five bands (Fig. 15a and Table 6). The calculated UV-visible absorption spectrum for L reproduces well the experimental curve (Fig. 15a and b). The lowest energy band (red Gaussian decomposition) was attributed to π–π* HOMO → LUMO TTF to methyl-2-Py-N-oxide charge transfers (ILCT) (Table 6). The absorption bands centred at 29 400 cm⁻¹ and 36 100 cm⁻¹ were also attributed to ILCT transitions while the intermediate absorption band was attributed to intra-TTF excitations. Finally the highest energy absorption band centred at 41 000 cm⁻¹ (purple decomposition) was attributed to intra-ligand (IL) transitions. The replacement of the ethylenedithio fragment with an ethylenedioxy one does not induce significant changes in the absorption properties. The UV-visible absorption properties of the coordination complex 3 have been first studied in the solid-state (Fig. S10†) and then in CH₂Cl₂ solution (Fig. 15c). The absorption spectra have been decomposed into six and seven bands respectively for the solution and solid-state measurements (Table 7). The lowest-energy additional absorption band (15 500 cm⁻¹) which is observed in the solid-state spectra is due to the intermolecular CT.^10c Additional intense absorption excitations have been observed around 27 000–30 000 cm⁻¹ that correspond to π–π* intra-tta⁻ excitations.^11c Complexations induce a weak red shift of the ligand-centered ILCT transition due to the moderate Lewis acid behavior of the Ln(tta)₃ moieties enforcing the electron withdrawing character of the 2-pyridine-N-oxide fragments even if the electronic communication through the methylthio arms is expected to be very weak. Thus, the absorption bands are red-shifted 1300 cm⁻¹ in coordination complexes compared to those in L. Since the shift value is determined from the data in CH₂Cl₂ solution, this is a first indication of the stability of the mononuclear complexes in such a solvent.

Fig. 15 (a) Experimental UV-visible absorption spectra of L in CH₂Cl₂ solution (C = 4 × 10⁻⁵ mol L⁻¹) (open gray circles). Respective Gaussian decompositions (dashed lines) and best fit (full black line) (R = 0.9988). (b) Theoretical absorption spectra of L (black line). The sticks represent the mean contributions of the absorption spectra for L. (c) Experimental UV-visible absorption spectra in CH₂Cl₂ solution of 3 (C = 4 × 10⁻⁵ mol L⁻¹) (open gray circles). Respective Gaussian decompositions (dashed lines) and best fit (full black line) (R = 0.9992).

Fig. 16 Molecular orbital diagram of L. The energy levels of the centred TTF and 2-methylpyridine-N-oxide orbitals are shown in orange and blue, respectively.

Table 6 TD-DFT calculated absorption energies and main compositions of the low-lying electronic transitions for L. In addition, the charge transfer and the pure intramolecular transitions are reported. ID, IA and H, L represent the intramolecular TTF (donor) or intramolecular 2-pyridyl-N-oxidemethylthio (acceptor), and the HOMO, the LUMO, respectively. Therefore, ILCT stands for intra-ligand charge transfer. The theoretical values were evaluated at the PCM(CH₂Cl₂)-PBE0/SVP level of approximation

	Energy exp. (cm^−1)	Energy theo. (cm^−1)	Osc.	Type	Assignment	Transition
L	23 800	25 307	0.04	ILCT		H → L (86%)
	29 400	30 586	0.12	ILCT		H → L+5 (22%)
						H−1 → L+1/+2 (15/19%)
	31 800	31 852	0.07	ID		H → L +5/+6 (44/32%)
	36 100	32 569 + 32 814	0.24 + 0.07	ILCT		H → L+6 (25%)
						H−1 → L (71%)
	41 000	40 484	0.07	IA		H−2 → L+1/+3 (20/20%)
		40 836	0.10	IA		H−1 → L+6 (43%)
		41 455	0.07	ID		H−3 → L/+2/+3/+4 (14/11/8/9/8%)
						H → L +11 (43%)

---

Table 7 Absorption data for the ligand L and coordination complex 3

	L (solid-state)	3 (CH2Cl2 solution)	3 (solid-state)	Type
Energy exp. (cm^−1)	15 500	—	15 500	Inter-LCT
	18 800	22 500	19 800	ILCT
	23 100	28 000	22 900	ILCT
	—	29 900	26 900	Itta
	27 100	32 000	29 700	ILCT
	31 500	37 900	33 000	ID
	39 200	43 900	38 500	IL

---

Emission properties. Emission properties of 3 were measured in the solid state at 77 K (Fig. 17a). The characteristic luminescence profile of Yb^III corresponding to the ²F_5/2 → ²F_7/2 transition is observed upon irradiation at 22 200 cm⁻¹ (450 nm).

Fig. 17 Luminescence spectrum of 3 in the NIR range for λ_ex = 22 220 cm⁻¹ (450 nm) in the solid state at 77 K (a) and luminescence spectrum obtained by direct irradiation of the f–f transitions (b).

The almost complete absence of any residual ligand-centred emission in the visible region indicates that the energy-transfer process is rather efficient. As already observed in previously published TTF-based complexes of Yb^III, the antenna-effect sensitization process is favored and proceeds through energy transfer from the singlet CT state of the L chromophore.^11c–f Eight emission maxima and shoulders are clearly identified at the following energies: 9624 cm⁻¹, 9722 cm⁻¹, 9756 cm⁻¹, 9860 cm⁻¹, 9900 cm⁻¹, 9940 cm⁻¹, 9980 cm⁻¹ and 10 230 cm⁻¹. This number of contributions is higher than the degeneracy of the ²F_7/2 ground state (Kramer's doublets) (maximum 4 contributions). By analogy with one of our previous studies^{11h,i 33a} and the one by Auzel et al.³⁵ the additional emission contributions could be attributed to transitions coming from the second and/or third M_J states of the ²F_5/2 multiplet state. In order to discriminate between the vibronic contribution and the contribution of the excited multiplet state to explain the origin of these additional bands, the emission spectrum was recorded by direct laser excitation of the f–f bands (Fig. 17b).³⁶ The two spectra are very similar although a partial quenching of the lowest-energy contribution (9624 cm⁻¹) by a direct f–f sensitization can be observed. This decrease may be assigned to vibronic contributions. Assuming that the additional emission bands are due to the two M_J excited states of the ²F_5/2 multiplet, the maximum number of expected transitions can be twelve. The total splitting is determined to be equal to 537 cm⁻¹. The values of this splitting for a Yb^III ion in a distorted and a regular³⁷D₃ symmetry are 455 cm⁻¹ and 372 cm⁻¹ respectively while a splitting of 528 cm⁻¹ is found for a Yb complex in a lower symmetry up to 880 nm for an organometallic derivative.³⁸ The value of 537 cm⁻¹ seems to correspond to a quite low symmetry that is in agreement with the lowest symmetry of the coordination sphere around the Yb^III ions in 3 (distorted square anti-prism). In addition, emission spectroscopy provides a direct probe of the ground state multiplet splitting which can be confronted with magnetic data (Fig. 18). Thus the energy splitting of the ²F_7/2 ground state multiplet extracts from the experimental luminescence (M_J = ±5/2 (0 cm⁻¹), M_J = ±7/2 (234 cm⁻¹), M_J = ±3/2 (300 cm⁻¹) and M_J = ±1/2 (488 cm⁻¹)) is in perfect agreement with the results of MS-CASPT2/RASSI-SO calculations (M_J = ±5/2 (0 cm⁻¹), M_J = ±7/2 (234 cm⁻¹), M_J = ±3/2 (298 cm⁻¹) and M_J = ±1/2 (467 cm⁻¹)).

Fig. 18 The solid state 77 K emission spectra are shown with an appropriate shift of the energy scale. Diagram of energy levels for compound 3 (full black sticks) given by the main contribution of the luminescence spectra. Labels correspond to the M_J values. Diagram of energy levels for compound 3 (red sticks) obtained by MS-CASPT2/RASSI-SO calculations.

Conclusions

Three Dy^III and Yb^III complexes involving the 4,5-ethylenedioxy-4′,5′-bis(2-pyridyl-N-oxidemethylthio)tetrathiafulvalene ligand (L) have been reported. The use of hfac⁻ as ancillary ligands for the lanthanide precursor leads to the dinuclear complex 1 while the use of tta⁻ leads to the mononuclear complexes 2 and 3. In all the compounds, the lanthanide ion is surrounded by a distorted O8 D_4d symmetry environment. The electronic distribution around the Dy^III ion in 1 and the Yb^III ion in 3 favours an Ising system and single-molecule magnet behaviours. The Ising character was confirmed by ab initio calculations. Both the calculated and the experimental anisotropy axis have been determined and are in good agreement with each other. Interestingly, compound 2 does not display the out-of-phase component of the magnetic susceptibility whereas 1 does even if the nature of the neighbouring atoms and the symmetry of the coordination sphere are the same. That shows that a subtle change in the ligands can induce a drastic change in the magnetic behaviour. The ligand L guarantees an efficient sensitization of the Yb^III luminescence by the antenna effect and the luminescence data for 3 can be correlated to the magnetic data. 3 thus enlarges the very scarce series of redox-active luminescent SMMs.

Acknowledgements

This work was supported by the CNRS, Rennes Métropole, Université de Rennes 1, Région Bretagne, FEDER and Agence Nationale de la Recherche (no. ANR-13-BS07-0022-01). BLG thanks the French GENCI center for high-performance computing resources (project x2015080649).

Notes and references

1. (a) A. Dei and D. Gatteschi, Angew. Chem., Int. Ed., 2011, 50, 11852–11858; (b) M. N. Leuenberger and D. Loss, Nature, 2001, 410, 789–793; (c) S. Hill, R. S. Edwards, N. Aliaga-Alcalde and G. Christou, Science, 2003, 302, 1015–1018; (d) M. Hosseini, S. Rebic, B. M. Sparkes, J. Twamley, B. C. Buchler and P. K. Lam, Light: Sci. Appl., 2012, 1, e40.
2. (a) R. Sessoli, H. L. Tsai, A. R. Schake, S. Wang, J. B. Vincent, K. Folting, D. Gatteschi, G. Christou and D. N. Hendrickson, J. Am. Chem. Soc., 2002, 115, 1804–1816; (b) M. Mannini, F. Pineider, P. Sainctavit, C. Danieli, E. Otero, C. Sciancalepore, A. M. Talarico, M.-A. Arrio, A. Cornia, D. Gatteschi and R. Sessoli, Nat. Mater., 2009, 8, 194–197; (c) M. Gu, X. Li and Y. Cao, Light: Sci. Appl., 2014, 3, e177.
3. (a) N. Papasimakis, S. Thongrattanasiri, N. I. Zheludev and F. J. Garcia de Abajo, Light: Sci. Appl., 2013, 2, e78; (b) S. Sanvito, Chem. Soc. Rev., 2011, 40, 3336–3355; (c) L. Bogani and W. Wernsdorfer, Nat. Mater., 2008, 7, 179–186.
4. S.-Y. Lin, C. Wang, L. Zhao, J. Wua and J. Tang, Dalton Trans., 2015, 44, 223.
5. J. Long, J. Rouquette, J.-M. Thibaud, R. A. S. Ferreira, L. D. Carlos, B. Donnadieu, V. Vieru, L. F. Chibotaru, L. Konczewicz, J. Haines, Y. Guari and J. Larionova, Angew. Chem., Int. Ed., 2015, 54, 2236–2240.
6. X.-L. Li, C.-L. Chen, H.-P. Xiao, A.-L. Wang, C.-M. Liu, X. Zheng, L.-J. Gao, X.-G. Yanga and S.-M. Fang, Dalton Trans., 2013, 42, 15317–15325.
7. (a) J. C. G. Bünzli and C. Piguet, Chem. Rev., 2002, 102, 1897; (b) R. Sessoli and A. K. Powell, Coord. Chem. Rev., 2009, 253, 2328–2341; (c) J. Luzon and R. Sessoli, Dalton Trans., 2012, 41, 13556–13567; (d) T. Tang, I. Hewitt, N. T. Madhu, G. Chastanet, W. Wernsdorfer, C. E. Anson, C. Benelli, R. Sessoli and A. K. Powell, Angew. Chem., Int. Ed., 2006, 45, 1729–1733.
8. (a) J.-C. G. Bünzli, Acc. Chem. Res., 2006, 39, 53–61; (b) S. V. Eliseeva and J.-C. G. Bünzli, Chem. Soc. Rev., 2010, 39, 189–227; (c) S. V. Eliseeva and J.-C. G. Bünzli, in Springer series on fluorescence, in Lanthanide spectroscopy, Materials, and Bio-applications, ed. P. Hännen and H. Härmä, Springer Verlag, Berlin, 2010, ch. 1, vol. 7; (d) Luminescence of Lanthanide Ions in Coordination Compounds and Nanomaterials, ed. A. De Bettencourt-Diaz, Wiley, 2014.
9. (a) H. Tanaka, H. Kobayashi, A. Kobayashi and P. Cassoux, Adv. Mater., 2000, 12, 1685–1689; (b) S. Uji, H. Shinagawa, T. Terashima, C. Terakura, T. Yakabe, Y. Terai, M. Tokumoto, A. Kobayashi, H. Tanaka and H. Kobayashi, Nature, 2001, 410, 908–910; (c) A. Kobayashi, E. Fujiwara and H. Kobayashi, Chem. Rev., 2004, 104, 5243–5264; (d) T. Enoki and A. Miyasaki, Chem. Rev., 2004, 104, 5449–5478; (e) E. Coronado and P. Day, Chem. Rev., 2004, 104, 5419–5448; (f) L. Ouahab and T. Enoki, Eur. J. Inorg. Chem., 2004, 933–941; (g) H. Fujiwara, K. Wada, T. Hiraoka, T. Hayashi, T. Sugimoto, H. Nakazumi, K. Yokogawa, M. Teramura, S. Yasuzuka, K. Murata and T. Mori, J. Am. Chem. Soc., 2005, 127, 14166–14167; (h) R. Kato, Bull. Chem. Soc. Jpn., 2000, 73, 515–534; (i) D. Lorcy, N. Bellec, M. Fourmigué and N. Avarvari, Coord. Chem. Rev., 2009, 253, 1398–1438 and references cited therein; (j) F. Pointillart, S. Golhen, O. Cador and L. Ouahab, Dalton Trans., 2013, 42, 1949–1960 and references cited therein.
10. (a) S. Faulkner, B. P. Burton-Pye, T. Khan, L. R. Martin, S. D. Wray and P. J. Skabara, Chem. Commun., 2002, 1668–1669; (b) S. J. A. Pope, B. P. Burton-Pye, R. Berridge, T. Khan, P. Skabara and S. Faulkner, Dalton Trans., 2006, 2907–2912; (c) F. Pointillart, T. Cauchy, O. Maury, Y. Le Gal, S. Golhen, O. Cador and L. Ouahab, Chem. – Eur. J., 2010, 16, 11926–11941; (d) F. Pointillart, A. Bourdolle, T. Cauchy, O. Maury, Y. Le Gal, S. Golhen, O. Cador and L. Ouahab, Inorg. Chem., 2012, 51, 978–984; (e) A. D'Aléo, F. Pointillart, L. Ouahab, C. Andraud and O. Maury, Coord. Chem. Rev., 2012, 256, 1604–1620; (f) F. Pointillart, B. Le Guennic, S. Golhen, O. Cador, O. Maury and L. Ouahab, Inorg. Chem., 2013, 52, 1610–1620; (g) M. Feng, F. Pointillart, B. Le Guennic, B. Lefeuvre, S. Golhen, O. Cador, O. Maury and L. Ouahab, Chem. – Asian. J., 2014, 9, 2814–2825; (h) F. Pointillart, B. Le Guennic, T. Cauchy, S. Golhen, O. Cador, O. Maury and L. Ouahab, Inorg. Chem., 2013, 52, 5978–5990.
11. (a) G. Cucinotta, M. Perfetti, J. Luzon, M. Etienne, P. E. Car, A. Caneschi, G. Calvez, K. Bernot and R. Sessoli, Angew. Chem., Int. Ed., 2012, 51, 1606–1610; (b) M.-E. Boulon, G. Cucinotta, J. Luzon, C. Degl'Innocenti, M. Perfetti, K. Bernot, G. Calvez, A. Caneschi and R. Sessoli, Angew. Chem., Int. Ed., 2013, 52, 350–354; (c) F. Pointillart, B. Le Guennic, S. Golhen, O. Cador, O. Maury and L. Ouahab, Chem. Commun., 2013, 49, 615–617; (d) J. Long, R. Vallat, R. A. S. Ferreira, L. D. Carlos, F. A. A. Paz, Y. Guari and J. Larionova, Chem. Commun., 2012, 48, 9974–9976; (e) K. Yamashita, R. Miyazaki, Y. Kataoka, T. Nakanishi, Y. Hasegawa, M. Nakano, T. Yamamura and T. Kajiwara, Dalton Trans., 2013, 42, 1987–1990; (f) K. Ehama, Y. Ohmichi, S. Sakamoto, T. Fujinami, N. Matsumoto, N. Mochida, T. Ishida, Y. Sunatsuki, M. Tsuchimoto and N. Re, Inorg. Chem., 2013, 52, 12828–12841; (g) M. Ren, S.-S. Bao, R. A. S. Ferreira, L.-M. Zheng and L. D. Carlos, Chem. Commun., 2014, 50, 7621–7624; (h) G. Cosquer, F. Pointillart, J. Jung, B. Le Guennic, S. Golhen, O. Cador, Y. Guyot, A. Brenier, O. Maury and L. Ouahab, Eur. J. Inorg. Chem., 2014, 69–82; (i) X. Yi, K. Bernot, V. Le Corre, G. Calvez, F. Pointillart, O. Cador, B. Le Guennic, J. Jung, O. Maury, V. Placide, Y. Guyot, T. Roisnel, C. Daiguebonne and O. Guillou, Chem. – Eur. J., 2014, 20, 1569–1576.
12. (a) J. J. Baldovi, J. M. Clemente-Juan, E. Coronado, Y. Duan, A. Gaita-Arino and C. Gimenez-Saiz, Inorg. Chem., 2014, 53, 9976–9980; (b) J. J. Baldovi, J. M. Clemente-Juan, E. Coronado and A. Gaita-Arino, Inorg. Chem., 2014, 53, 11323–11327; (c) N. F. Chilton, D. Collison, E. J. L. McInnes, R. E. P. Winpenny and A. Soncini, Nat. Commun., 2013, 4, 2551–2558; (d) C. Y. Chow, H. Bolvin, V. E. Campbell, R. Guillot, J. F. Kampf, W. Werndorfer, F. V. Gendron, J. Autschbach, V. L. Pecoraro and T. Mallah, Chem. Sci., 2015, 6, 4148–4159; (e) L. Ungur, S.-Y. Lin, J. Tang and L. F. Chibotaru, Chem. Soc. Rev., 2014, 43, 6894–6905; (f) Y.-N. Guo, G.-F. Xu, W. Wernsdorfer, L. Ungur, Y. Guo, J. Tang, H.-J. Zhang, L. F. Chibotaru and A. K. Powell, J. Am. Chem. Soc., 2011, 133, 11948–11951.
13. M. F. Richardson, W. F. Wagner and D. E. Sands, J. Inorg. Nucl. Chem., 1968, 30, 1275–1289.
14. A. I. Vooshin, N. M. Shavaleev and V. P. Kazakov, J. Lumin., 2000, 91, 49–58.
15. (a) N. Sevenstrup, K. M. Rasmussen, T. K. Hansen and J. Becher, Synthesis, 1994, 809; (b) M. Xu, Y. Ji, J. L. Zuo and X. Z. You, J. Heterocycl. Chem., 2005, 42, 847; (c) L. Binet, J. M. Fabre, C. Montginoul, K. B. Simonsen and J. Becher, J. Chem. Soc., Perkin Trans. 1, 1996, 783.
16. M. Polasek, M. Sedinova, J. Kotek, L. V. Elst, R. N. Muller, P. Hermann and I. Lukes, Inorg. Chem., 2009, 48, 455.
17. G. M. Sheldrick, SHELX97 – Programs for Crystal Structure Analysis (Release 97-2), Institut für Anorganische Chemie der Universität, Tammanstrasse 4, D-3400 Göttingen, Germany, 1998, SIR97; A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori and R. Spagna, J. Appl. Crystallogr., 1999, 32, 115–119.
18. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, Jr., J. A. Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09 Revision A.02, Gaussian Inc., Wallingford, CT, 2009.
19. (a) J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868; (b) C. Adamo and V. Barone, J. Chem. Phys., 1999, 110, 6158–6170.
20. F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7, 3297–3305.
21. J. Tomasi, B. Mennucci and R. Cammi, Chem. Rev., 2005, 105, 2999–3093.
22. (a) M. Cossi and V. Barone, J. Chem. Phys., 2001, 115, 4708–4717; (b) R. Improta, V. Barone, G. Scalmani and M. J. Frisch, J. Chem. Phys., 2006, 125, 054103–054109.
23. A.-R. Allouche, J. Comput. Chem., 2011, 32, 174–182.
24. F. Aquilante, L. De Vico, N. Ferré, G. Ghigo, P. A. Malmqvist, P. Neogrady, T. Bondo Pedersen, M. Pitonak, M. Reiher, B. O. Roos, L. Serrano-Andrés, M. Urban, V. Veryazov and R. Lindh, J. Comput. Chem., 2010, 31, 224–247.
25. B. O. Roos, P. R. Taylor and P. E. M. Siegbahn, Chem. Phys., 1980, 48, 157–288.
26. (a) P. A. Malmqvist, B. O. Roos and B. Schimmelpfennig, Chem. Phys. Lett., 2002, 357, 230–240; (b) P. A. Malmqvist and B. O. Roos, Chem. Phys. Lett., 1989, 155, 189–194.
27. (a) L. F. Chibotaru and L. Ungur, J. Chem. Phys., 2012, 137, 064112–064122; (b) L. F. Chibotaru, L. Ungur and A. Soncini, Angew. Chem., Int. Ed., 2008, 47, 4126–4129.
28. F. Aquilante, P.-A. Malmqvist, T.-B. Pedersen, A. Ghosh and B. O. Roos, J. Chem. Theory Comput., 2008, 4, 694–702.
29. (a) B. O. Roos, R. Lindh, P.-A. Malmqvist, V. Veryazov and P.-O. Widmark, J. Phys. Chem. A, 2004, 108, 2851–2858; (b) B. O. Roos, R. Lindh, P.-A. Malmqvist, V. Veryazov and P.-O. Widmark, J. Phys. Chem. A, 2005, 109, 6576–6586; (c) B. O. Roos, R. Lindh, P.-A. Malmqvist, V. Veryazov, P.-O. Widmark and A. C. Borin, J. Phys. Chem. A, 2008, 112, 11431–11435.
30. M. Llunell, D. Casanova, J. Cirera, J. M. Bofill, P. Alemany and S. Alvarez, S. SHAPE (version 2.1), Barcelona, 2013.
31. F. Habib, P.-H. Lin, J. Long, I. Korobkov, W. Werndorfer and M. Murugesu, J. Am. Chem. Soc., 2011, 133, 8830.
32. G. Cosquer, F. Pointillart, S. Golhen, O. Cador and L. Ouahab, Chem. – Eur. J., 2013, 19, 7895–7903.
33. (a) J. Jung, T. da Cunha, B. Le Guennic, F. Pointillart, L. M. Pereira, J. Luzon, S. Golhen, O. Cador, O. Maury and L. Ouahab, Eur. J. Inorg. Chem., 2014, 3888–3894; (b) F. Pointillart, J. Jung, R. Berraud-Pache, B. Le Guennic, V. Dorcet, S. Golhen, O. Cador, O. Maury, Y. Guyot, S. Decurtins, S.-X. Liu and L. Ouahab, Inorg. Chem., 2015, 54, 5384–5397.
34. G. Cosquer, F. Pointillart, B. Le Guennic, Y. Le Gal, S. Golhen, O. Cador and L. Ouahab, Inorg. Chem., 2012, 51, 8488–8501.
35. P. Goldner, F. Pell, D. Meichenin and F. Auzel, J. Lumin., 1997, 71, 137–150.
36. G. Zucchi, R. Scopelliti, P.-A. Pittet, J. C. G. Bünzli and R. D. Rogers, J. Chem. Soc., Dalton Trans., 1999, 931–938.
37. F. R. Gonçalves Silva, O. L. Malta, C. Reinhard, H. U. Güdel, C. Piguet, J. Moser and J.-C. G. Bünzli, J. Phys. Chem. A, 2002, 106, 1670–1677.
38. G. Lapadula, A. Bourdolle, F. Allouche, M. Conley, I. del Rosa, L. Maron, W. W. Lukens, Y. Guyot, C. Andraud, S. Brasselet, C. Copéret, O. Maury and R. A. Andersen, Chem. Mater., 2014, 26, 1062–1073.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 1407103–1407106. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5qi00087d

---