From Positive to Negative Zero-Field Splitting in a Series of Strongly

This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article pubs.acs.org/IC

From Positive to Negative Zero-Field Splitting in a Series of Strongly Magnetically Anisotropic Mononuclear Metal Complexes Ghénadie Novitchi,† Shangda Jiang,† Sergiu Shova,‡ Fatima Rida,† Ivo Hlavička,† Milan Orlita,† Wolfgang Wernsdorfer,§,⊥ Rana Hamze,∥ Cyril Martins,∥ Nicolas Suaud,∥ Nathalie Guihéry,*,∥ Anne-Laure Barra,† and Cyrille Train*,† †

Laboratoire National des Champs Magnétiques Intenses, UPR CNRS 3228, Université Grenoble-Alpes, 25 rue des Martyrs, B.P. 166, 38042 Grenoble Cedex 9, France ‡ “Petru Poni” Institute of Macromolecular Chemistry, Aleea Gr. Ghica Voda 41A, 700487 Iasi, Romania § Institut Néel, UPR CNRS 2940, Université Grenoble-Alpes, B.P. 166, 38042 Grenoble Cedex 9, France ⊥ Physikalisches Institut and Institute of Nanotechnology, Karlsruhe Institute of Technology, Wolfgang-Gaede-Strasse 1, 76131 Karlsruhe, Germany ∥ Laboratoire de Chimie et Physique Quantiques, UMR 5626, Université de Toulouse 3, Paul Sabatier, 118 route de Narbonne, 31062 Toulouse, France S Supporting Information *

ABSTRACT: A series of mononuclear [M(hfa)2(pic)2] (Hhfa = 1,1,1,5,5,5-hexafluoro-2,4-pentanedione; pic = 4-methylpyridine; M = FeII, CoII, NiII, ZnII) compounds were obtained and characterized. The structures of the complexes have been resolved by single-crystal X-ray diffraction, indicating that, apart from the zinc derivative, the complexes are in a trans configuration. Moreover, a dramatic lenghthening of the Fe−N distances was observed, whereas the nickel(II) complex is almost perfectly octahedral. The magnetic anisotropy of these complexes was thoroughly studied by directcurrent (dc) magnetic measurements, high-field electron paramagnetic resonance, and infrared (IR) magnetospectroscopy: the iron(II) derivative exhibits an out-of-plane anisotropy (DFe = −7.28 cm−1) with a high rhombicity, whereas the cobalt(II) and nickel(II) complexes show in-plane anisotropy (DCo ∼ 92−95 cm−1; DNi = 4.920 cm−1). Ab initio calculations were performed to rationalize the evolution of the structure and identify the excited states governing the magnetic anisotropy along the series. For the iron(II) complex, an out-of-phase alternating-current (ac) magnetic susceptibility signal was observed using a 0.1 T dc field. For the cobalt(II) derivative, the ac magnetic susceptibility shows the presence of two field-dependent relaxation phenomena: at low field (500 Oe), the relaxation process is beyond single-ion behavior, whereas at high field (2000 Oe), the relaxation of magnetization implies several mechanisms including an Orbach process with Ueff = 25 K and quantum tunneling of magnetization. The observation by μ-SQUID magnetization measurements of hysteresis loops of up to 1 K confirmed the single-ion-magnet behavior of the cobalt(II) derivative.

■

INTRODUCTION The design of species displaying slow magnetic relaxation is an intense research field related to applications in data storage and processing. In single-molecule magnets (SMMs), the magnetic bistability is ensured by the existence of a degenerate ground state associated with the MS = ±S spin components of the ground-state spin S. The stability of these states depends on their energy difference to the MS = 0 or ±1/2 components for integer or half-integer spins, respectively.1−9 The corresponding energy barrier is given by the following expression: U = S2|D|

where D is the axial zero-field-splitting (ZFS) parameter of the ground-state spin S. Obtaining more efficient SMMs requires increasing this barrier. During the last 2 decades, the most widespread strategy to reach this objective focused on increasing the spin value of the ground state and preserving the negative sign of the axial ZFS parameter. An impressive variety of metal clusters, in particular with MnIII paramagnetic centers, exhibiting ferroand antiferromagnetic exchange were thus synthesized.4,10−20 An unexpected conclusion of these studies was that the widely considered “increase S” design rule is not as efficient as suggested by eq 1. Indeed, it was shown theoretically that D is inversely

for integer spin

⎛ 1⎞ U = ⎜S2 − ⎟|D| for half‐integer spin ⎝ 4⎠ © XXXX American Chemical Society

Received: July 27, 2017

(1) A

DOI: 10.1021/acs.inorgchem.7b01861 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry proportional to S2 and therefore U does not or only weakly depends on the ground-state total spin.21 Moreover, the misalignment of the anisotropy axes, together with the projection coefficients of the single-ion anisotropy parameters, can drastically decrease the value of D compared to those of the constituting anisotropic metal ions. Finally, the symmetry lowering upon self-assembly favors rhombic anisotropy and, hence, a lowering of the effective energy barrier as well as a higher probability of quantum tunneling of magnetization (QTM). Accordingly, mononuclear complexes have been reconsidered as promising objects to increase the magnetic relaxation barrier, leading to the emergence of single-ion magnets (SIMs).22−35 These molecules are simpler to synthesize with a good versatility toward metal substitution. They are robust and, hence, easier to integrate in devices. Finally, because they have a single magnetic center, they allow deep examination by theoretical means. It first appeared that exotic coordinations of the anisotropic center can lead to strong negative D values. Accordingly, an important effort of ligand design was undertaken to favor such coordinations and observe SIM properties.24−35 More recently, it was shown that mere octahedral metal complexes may exhibit SIM properties, calling for new studies on the most common type of metal complexes. Herein, starting from the known cobalt(II) derivative,36 we report on a series of bis(1,1,1,5,5,5-hexafluoro-2,4-pentanedionato)bis(4-methylpyridine)metal(II) complexes. The crystal structures of the iron, nickel, and zinc derivatives are determined by single-crystal X-ray diffraction. The magnetic properties are studied by static magnetometry, high-field/high-frequency electron paramagnetic resonance (HF-EPR), and, more originally, infrared (IR) magnetospectroscopy. The structure and anisotropy parameters are analyzed and discussed based on the results of ab initio calculations. Finally, the complex dynamic properties of this series of complexes are presented and analyzed thereafter.

■

4 (Zn). Found: C, 39.57; H, 2.30; N, 4.13. Calcd for C22H16O4N2F12Zn: C, 39.69; H, 2.42; N, 4.21. IR (KBr, cm−1): 491.62, 583.35, 666.44, 791.04, 808.21, 1028.22, 1096.31, 1141.21, 1197.11, 1234.28, 1258.18, 1499.21, 1524.91, 1551.81, 1626.24, 1651.39, 1662.70. X-ray Measurements. Crystallographic measurements for 1, 3, and 4 were carried out with an Oxford Diffraction XCALIBUR E CCD diffractometer equipped with graphite-monochromated Mo Kα radiation. Single crystals were positioned at 40 mm from the detector, and 323, 158 and 50 frames were measured during 5, 25, and 3 s for 1, 3, and 4, respectively, over 1° scan width. The unit cell determination and data integration were carried out using the CrysAlis package of Oxford Diffraction.39 All of the structures were solved by direct methods using Olex2 software40 with the SHELXS structure solution program and refined by full-matrix least squares on F2 with SHELXL-97.41 Atomic displacements for non-hydrogen atoms were refined using an anisotropic model. Hydrogen atoms have been placed in fixed, idealized positions, accounting for the hybridation of the supporting atoms. The molecular plots were obtained using the Olex2 program. Table 1 provides a summary of the crystallographic data together with refinement details for compounds 1, 3, and 4. CCDC 1558350, 1558351, and 1558352 contain the supplementary crystallographic data for this contribution for 1, 3, and 4, respectively. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www. ccdc.cam.ac.uk/data_request/cif. Magnetic Studies. Direct-current (dc) magnetic susceptibility data (2−300 K) were collected on powdered samples using a SQUID magnetometer (Quantum Design MPMS-XL), applying a magnetic field of 0.1 T. All data were corrected for the contribution of the sample holder, and the diamagnetism of the samples was estimated from Pascal’s constants.42,43 Magnetization measurements of up to 5 T were performed on smaller quantities of compounds to avoid saturation of the signal. Alternating-current (ac) magnetic susceptibility was measured between 2 and 10 K with an oscillating field magnitude of 3.0 Oe and a frequency ranging between 1 and 1488 Hz in the presence of a dc field of up to 4000 Oe. The relaxation times were extracted from the simultaneous fit of χ′ac and χ′′ac using the generalized Debye model.44,45 Further magnetization measurements were performed with an array of μ-SQUIDs.46,47 This magnetometer works in the temperature range of 0.04 K to ca. 7 K and in fields of up to 1.4 T with sweeping rates as high as 0.28 T s−1. The time resolution is approximately 1 ms. The magnetic field can be applied in any direction of the μ-SQUID plane with precision much better than 0.1° by separately driving three orthogonal coils. In order to ensure good thermalization, each sample was fixed with Apiezon grease. HF-EPR. HF-EPR spectra were recorded on a multifrequency spectrometer48 operating in a double-pass configuration. Several frequency sources were used: a 110 GHz source (Virginia Diodes Inc.), alone or associated with multipliers generating harmonics up to the sixth order (662 GHz); two sources operating at 95 and 115 GHz, respectively (Radiometer Physics GmbH), alone or associated with multipliers up to the fifth harmonic (475 and 575 GHz, respectively). Detection was performed with a hot electron InSb bolometer (QMC Instruments). The exciting light was propagated with a Quasi-Optical setup (Thomas Keating) outside the cryostat and with the help of a corrugated waveguide inside it. The main magnetic field was supplied by a 16 T superconducting magnet associated with a VTI (Cryogenic). The measurements were done on powdered samples pressed into pellets in order to limit torqueing effects. Calculated spectra were obtained with the SIM program49 from H. Weihe (University of Copenhagen). IR Magnetospectroscopy. To measure the IR magnetoabsorbance spectra, the studied compound was dispersed in eicosane with a ratio of 1:20. The pellet was then placed inside a cryostat (at a temperature of 1.6 or 4.2 K) of a superconducting coil delivering magnetic fields up to 13 T. The spectra were recorded with a commercial Bruker IFS 66v/s Fourier transform infrared spectrometer using a globar and composite silicon bolometer as the source and detector of IR radiation, respectively. Theoretical Calculations. Ab initio calculations were performed using the ORCA4.0.0 code.50 Because magnetic anisotropy is very sensitive to the geometry, X-ray structures were used. The positions of

EXPERIMENTAL SECTION

Chemicals. The starting materials and solvents were purchased from Aldrich. These chemicals were used without further purification. [M(hfa)2(H2O)2] (Hhfa = 1,1,1,5,5,5-hexafluoro-2,4-pentanedione; M = FeII, CoII, NiII, ZnII) were prepared according to slightly modified literature procedures.37,38 All of the syntheses for iron(II) were performed in a dry and inert atmosphere (N2). Synthesis of 1−4. The [M(hfa)2(pic)2] [pic = 4-methylpyridine; M = Fe (1), Co (2), Ni (3), Zn (4)] complexes were obtained by dissolving 0.5 mmol of [M(hfa)2(H2O)2] in 15 mL of chloroform containing 1.5 mmol of 4-methylpyridine at ∼50 °C. The solutions were cooled to room temperature. Within 1 h, the compound was obtained as single crystals suitable for X-ray diffraction. They were filtered off and washed with a minimum amount of cold pentane. The synthesis and structure of 2 (Co) were previously reported.36 IR spectra and powder X-ray diffraction patterns are given in Figures S1−S7. 1 (Fe). Found: C, 39.83; H, 2.32; N, 4.31. Calcd for C22H16O4N2F12Fe: C, 40.27; H, 2.46; N, 4.27. IR (KBr, cm−1): 493.02, 590.51, 666.05, 795.47, 809.04, 1018.38, 1069.55, 1101.83, 1139.12, 1192.60, 1220.65, 1258.68,1342.84, 1422.49, 1490.48, 1523.64, 1550.09, 1561.18, 1592.56, 1619.78. 2 (Co). Found: C, 39.79; H, 2.26; N, 4.14. Calcd for C22H16O4N2F12Co: C, 40.08; H, 2.45; N, 4.25. IR (KBr, cm−1): 494.05, 588.32, 670.05, 794.02, 809.91, 1021.61, 1070.24, 1097.98, 1139.90, 1194.02, 1220.34, 1258.81, 1349.45, 1386.08, 1424.03, 1502.73, 1526.74, 1553.64, 1593.30, 1622.07, 1638.31. 3 (Ni). Found: C, 39.75; H, 2.30; N, 4.19. Calcd for C22H16O4N2F12Ni: C, 40.09; H, 2.45; N, 4.25. IR (KBr, cm−1): 495.85, 539.31, 588.86, 673.32, 743.32, 767.40, 793.46, 810.52, 1025.56, 1070.58, 1098.91, 1139.63, 1193.98, 1221.35, 1257.96, 1349.10, 1386.93, 1424.74, 1504.16, 1526.93, 1552.62, 1564.14, 1592.18, 1623.91, 1641.28. B


Article

Inorganic Chemistry

Table 1. Crystallographic Data, Details of Data Collection, and Structure Refinement Parameters for Compounds 1, 3, and 4 empirical formula fw temperature, K cryst syst space group a, Å b, Å c, Å V, Å3 Z Dcalc, mg mm−3 μ, mm−1 cryst size, mm3 θmin−θmax, deg reflns collected indep reflns data/restraints/param R1a [I > 2σ(I] wR2b(all data) GOFc largest diff peak/hole, e Å−3

1

3

4

C22H16F12FeN2O4 656.22 173 orthorhombic Cmcm 9.6502(3) 16.4467(7) 16.0355(5) 2545.06(16) 4 1.713 0.714 0.40 × 0.40 × 0.30 4.9−52.74 8650 1426 [Rint = 0.0394] 1426/0/113 0.0325 0.0817 1.092 0.23/−0.32

C22H16F12NiN2O4 659.08 173 orthorhombic Cmcm 9.7059(3) 16.2556(6) 15.8725(6) 2504.28(15) 4 1.748 0.898 0.35 × 0.15 × 0.10 4.88−52.74 4289 1403 [Rint = 0.0380] 1403/0/113 0.0342 0.0721 1.063 0.30/−0.35

C22H16F12N2O4Zn 665.74 200 monoclinic C2/c 9.0982(7) 17.4929(15) 17.0536(13) 2700.8(4) 4 1.637 1.023 0.3 × 0.3 × 0.03 4.658−50.05 4992 2379 [Rint = 0.0332] 2379/0/187 0.0648 0.1699 1.038 0.64/−0.56

R1 = ∑||Fo| − |Fc||/∑|F0|. bwR2 = {∑[w (Fo2 − Fc2)2]/∑[w(Fo2)2]}1/2. cGOF = {∑[w(Fo2 − Fc2)2]/(n − p)}1/2, where n is the number of reflections and p is the total number of parameters refined.

a

Figure 1. Molecular structures of 1 (a) and 4 (b). DKH-Def2-QZVPP with 24s18p10d4f2g contracted to 14s10p5d4f2g; for C, N, and O, DKH-Def2-TZVP(-f) with 11s6p2d contracted to 6s3p2d, and for H, DKH-Def2-SVP with 4s1p contracted to 2s1p. The values of the ZFS parameters D and E have been extracted from the ab initio energies and wave functions using the effective Hamiltonian theory,58,59 as was already proposed earlier.60

the hydrogen atoms were optimized using density functional theory (DFT) calculations (UB3LYP/6-311g). In the state-average completeactive space self-consistent-field (SA-CASSCF) calculations, the active space contains all d electrons in the five 3d orbitals and five diffuse d orbitals. The orbitals that are discussed further have been optimized in an average way considering all of the states with spin Smax and Smax − 1 spin. Dynamic correlation effects have been treated at the second order of perturbation using the partially contracted N-electron-valence perturbation theory, NEVPT2, method.51−54 Finally, the spin−orbit matrix between the various MS components of the CASSCF states was diagonalized using either the CASSCF or NEVPT2 energies as diagonal elements. The following basis sets55−57 were used: for Fe, Co, Ni,

■

RESULTS AND DISCUSSION Structures. Single-crystal X-ray diffraction revealed that compounds 1 (Figure 1) and 3 are isotypes and crystallize in the Cmcm space group with very close unit cell parameters (Table 1).

C


Article

Inorganic Chemistry Consequently, 1, 2,36 and 3 are isostructural, and the crystal structure will be detailed for 1. The crystal of 1 is composed of isolated neutral [FeII(hfa)2(pic)2] complexes. The central metal atom is surrounded by four oxygen atoms from two hexafluoroacetylacetonate ligands acting as bidentate chelating ligands and by two nitrogen atoms from 4-methylpyridine monodentate ligands. The four coordinated oxygen atoms define a planar equatorial fragment; the complex thus exhibits a trans(N) configuration. Compound 4 (Figure 1) crystallizes in the C2/c space group (Table 1). Its crystal is composed of isolated neutral [ZnII(hfa)2(pic)2] complexes. As for 1 and 3, the central metal atom is surrounded by four oxygen atoms from two hexafluoroacetylacetonate ligands acting as bidentate chelating ligands and by two nitrogen atoms from 4-methylpyridine monodentate ligands, but in contrast with 1 and 3, the complex exhibits a cis(N) configuration. For 1, the O4N2 octahedron exhibits strong axial elongation. The main evolution upon going from 1 to 3 is that the difference between the M−N and M−O distances dramatically decreases (Table 2), with the octahedron being nearly regular for the

Figure 2. Temperature dependence of the χT product at Hdc = 0.1 T for 1 (circles), 2 (diamonds), and 3 (triangles). The solid lines are the simulated susceptibility using the ZFS parameters deduced from other techniques (Table 4).

Table 2. Selected Bond Lenghts (Å) and Distortion Parameter dstr (Å) for Coordination Polyhedra in 1−3a complex

M−O1

M−N1

M−N2

dstr = Rax − Req

1 236 3

2.057(1) 2.060(1) 2.035(1)

2.190(3) 2.145(3) 2.087(3)

2.224(3) 2.167(3) 2.108(3)

0.15 0.096 0.0625

of isolated high-spin centers with g = 2.00 (3.002 cm3 K mol−1 for S = 2, 1.877 cm3 K mol−1 for S = 3/2, and 1.001 cm3 K mol−1 for S = 1). For isolated mononuclear complexes, these increased values can be ascribed to an orbital contribution to g. They are typical of those encountered in iron(II),63−65 cobalt(II),66−69 and nickel(II)70−73 complexes. With decreasing temperature, the χT product continuously decreases. At 2 K, it is worth 2.065 cm3 K mol−1 for 1, 2.084 cm3 K mol−1 for 2, and 0.597 cm3 K mol−1 for 3. In the case of mononuclear complexes, such a temperature dependence of the magnetic susceptibility is essentially due to magnetic anisotropy. In the case of 2, it is expected to be very strong. Nevertheless, other phenomena, in particular intermolecular interactions, can contribute to the departure of the magnetic susceptibility from a Curie behavior, making the dc magnetic susceptibility a delicate method to determine the ZFS parameters unambiguously. Nonetheless, the parameters found by other techniques (Table 4) were used to reproduce the thermal variations of the dc magnetic susceptibility, offering a complementary validation of these parameters (Figure 2). The first technique that was used to reach the ZFS parameters relies on exploitation of the magnetization curves for applied fields varying between 0 and 5 T measured at different temperatures (Figure 3). These curves are fitted simultaneously using the following Hamiltonian:74

b

a The structural data for 2 are taken from ref 36. bRax = (M−N1 + M−N2)/2 and Req = M−O1 distances.

nickel(II) derivative. It is worth noting that 2 is one of the rare examples of an axially elongated cobalt(II) coordination complex.61 With this decrease related to the concomitant evolution of the M−O and M−N distances, it will be further analyzed by theoretical means. The six-membered metallocycle formed by the metal ion and bidentate hexafluoroacetylacetonate ligand has a boat conformation. An intramolecular CH--π [2.6262(1) Å] interaction appears between the protons in the α position of one of the two 4-methylpyridine ligands and the delocalized π orbital of the six-membered metallocycle, whereas given that the two apical coordinated 4-methylpyridine ligands are strictly perpendicular to one another, the second 4-methylpyridine ligand is not implied in any CH--π interaction. Because of this asymmetric intramolecular interaction, the iron center is deviated from the plane by 0.051(1) Å toward the apical coordinated atom N1. In all of the compounds including 4, the most obvious feature of the crystal packing is the absence of any short intermolecular contacts, indicating the presence of π···π and C−H···π interactions. In the case of 1−3, the crystal packing is essentially governed by numerous C−F···F−C62 and C−H···π intermolecular interactions. These interactions lead to the formation of a three-dimensional (3D) supramolecular network (Figure S8). For 4, the main crystal structural motif essentially results from the parallel packing of the supramolecular layers in (101) planes exclusively built from C−F···F−C interactions (Figure S9). Static Magnetic Studies. Thermal variations of the χT product, with χ being the molar magnetic susceptibility, are shown in Figure 2 for the three compounds. At room temperature, the χT products are worth 3.309 cm3 K mol−1 for 1, 3.177 cm3 K mol−1 for 2, and 1.246 cm3 K mol−1 for 3. These values are slightly (1 and 3) to dramatically (2) higher than those

H = DSz 2 + E(Sx 2 − Sy 2) + gμB SB

(2)

where D, E, and g stand for the axial and rhombic ZFS parameters and the isotropic Landé factor of the S spin state and μB and B for the Bohr magneton and magnetic field, respectively. Equation 2 allows one to obtain excellent fits of the experimental data (Figure 3). The corresponding ZFS terms are gathered in Table 3. For the iron(II) complex 1, the ZFS axial parameter exhibits a strong negative value, whereas it is positive for 2 and 3. For 3, the experimental values of D can be compared with the value calculated using the following phenomenological expression:70 Dcalc = 2[25.8(1 − e−0.014dstr )]

(3)

where dstr is the structural distortion defined in Table 2. Using the dstr value extracted from the structural analysis, the ZFS axial parameter calculated with eq 3 is worth +4.32 cm−1, D


Article

Inorganic Chemistry

Figure 3. Field dependence of magnetization data collected on microcrystalline compounds 1−3 at temperatures ranging between 2 and 5 K. The solid lines correspond to the best fits according to the model indicated in the text with the parameters indicated in Table 3.

which arises from the ground-state MS level. This reveals that the ZFS of complex 1 is rather large because it induces a gap of about 20 cm−1 between the levels implied in this 662 GHz transition. The assignment of the transitions is then only possible after the determination of the ZFS parameters through the simulation of the spectra. Conversely, for 3, transitions coming from the ground-state level are observed at all frequencies. Furthermore, the transitions can be assigned at first sight; for instance, at 220.8 GHz, a forbidden transition is observed at 1.7 T, and the MS = −1 to MS = 0 transitions are obbserved at 3.4, 6.1, and 12.2 T for the principal magnetic axes X, Y, and Z, respectively (Figure 4, right). For 1 and 3, considering the Hamiltonian of eq 2, where the isotropic g factor has been replaced by anisotropic ones, gX, gY, and gZ, it is possible to simulate the spectra (Figure 4) to determine the Landé factors and the axial and rhombic ZFS parameters, D and E, leading to gX = 2.02 ± 0.02, gY= 2.00 ± 0.02, gZ = 2.08 ± 0.02, D = −7.28 ± 0.02 cm−1, and E = 0.98 ± 0.03 cm−1 (|E/D| = 0.13) for 1 and gX = 2.20 ± 0.02, gY = 2.22 ± 0.02, gZ = 2.15 ± 0.02, D = 4.920 ± 0.005 cm−1, and E = 1.413 ± 0.003 cm−1 (|E/D| = 0.29) for 3. For 2, the large D value prevents the observation of signals involving the excited-state doublet. The spectra recorded at all frequencies are thus typical of a system that can be described with

Table 3. Results of Simultaneous Fits of the Field Dependence of Magnetization at 2, 3, 4, and 5 K Using the Hamiltonian Given in eq 2 complex 1 2 3

S

D, cm−1

|E|, cm−1

|E/D|

g

2 /2 1

−10.31 +24.17 +5.80

2.45 6.90 0.01

0.238 0.285 0.002

2.40 2.50 2.23

3

which is consistent with the value extracted from magnetization measurements. It should be noted that, following this analysis, 1 and 2 present high ZFS rhombic parameters E (Table 3). Nonetheless, the fitting of the magnetization is not very sensitive to the value of E. To secure the determination of the anisotropy parameters, it is worthwhile to use a spectroscopic method directly sensitive to the magnetic anisotropy parameters, namely, HF-EPR. HF-EPR Spectroscopy. Powder HF-EPR spectra of compounds 1−3 were measured at several frequencies, in fields up to 16 T and at T = 5 and 15 K (Figures 4 and S10−S12). For 1, the signals gain intensity when the temperature is increased, indicating that they originate from the excited-state levels of the multiplet, except for the low-field signal observed at 662 GHz,

Figure 4. Experimental (bold lines) and simulated (thin lines) powder HF-EPR spectra at 460 GHz and T = 5 and 15 K for 1 (left) and at 220.8 and 331.2 GHz and T = 5 K for 3 (right). E


Article

Inorganic Chemistry an effective spin Seff = 1/2. Accordingly, only effective Landé factors can be extracted directly from the HF-EPR spectra, and it is not possible to estimate the ZFS parameters of the S = 3/2 quadruplet. For 2, the three effective Landé factors are g1,eff = 5.98, g2,eff = 3.92, and g3,eff = 2.25. Because the spectra recorded at the highest frequency still correspond to that of Seff = 1/2, it is possible to give a lower limit to the value of the axial anisotropy parameter, that is, |D| > 22 cm−1. Without further information on D, a wide range of in-plane parameters allows one to obtain satisfying simulations of the spectra. It is thus crucial to use a complementary spectroscopic method that is sensitive to high values of the ZFS parameters.75,76 Hereafter, we turned ourselves toward IR magnetospectroscopy to probe the anisotropy parameters of 2. IR Magnetic Spectroscopy. The IR spectra of 2 dispersed in eicosane were measured up to 13 T at 1.6 and 4.2 K (Figure 5).

magnetic susceptibility must be carefully corrected from diamagnetic effects because the models that are used to fit the data deal with the paramagnetic contribution to the magnetic susceptibility. This is particularly difficult for measurements performed in eicosane often used in the case of anisotropic systems to avoid reorientation of the crystal upon application of high fields. Second, magnetic susceptibility is a thermodynamic quantity that averages the contribution of all of the thermally populated states; the risk of overparametrization of the system is therefore always present. The limitations of magnetization measurements in the determination of the anisotropy parameters are illustrated thereafter. A first set of parameters was determined, allowing all of the parameters to be free (Table 3). Then, for compounds 1 and 3, the values of the D parameter were fixed to the values found by HF-EPR, which are 15% to 40% less than the values found previously. For compound 2, the value was fixed to the value found by IR magnetospectroscopy, which is 4 times the value found from the direct fitting of the magnetization. Upon fitting with these constraints, the quality of the fits remains very satisfactory (Figure S14), indicating that magnetization measurement is a good technique to determine the sign of D but does not guarantee a quantitative determination of this parameter. The results for the E parameter when D is fixed are much more dramatic. For 1, the E/D ratio exactly corresponds to that determined by HF-EPR. On the contrary, for 3, the magnetization measurements indicate that the system is almost perfectly uniaxial, whereas HF-EPR concludes that it is almost perfectly rhombic. Going from thermodynamic to spectroscopic measurements is the best way to secure the determination of the magnetic anisotropy parameters. In the present series, HF-EPR appears as the most straightforward technique to determine the anisotropy parameters of 1 and 3 because one can observe transitions whose field positions are directly governed by the anisotropy parameters (Figures 4 and S10−S11). Nonetheless, if, with the highest possible field and/or highest frequency source available, the system remains in the so-called low-field limit, then HF-EPR only gives access to the effective Landé factors for half-integer spins. This is the case encountered for the cobalt(II) derivative 2 (Figure S12). HF-EPR then only sets a lower limit to the absolute value of the D parameter, that is, 22 cm−1 with our setup. To go beyond this limit, several solutions have been proposed.74,75 Herein, we have used IR magnetospectroscopy. With this technique, the limitation of the value is not an issue. When a high magnetic field is applied, it is possible to distinguish the transitions related to the magnetic properties from those arising from the rotavibrational characteristics of the compound. In the case of 2, only one magnetic transition has been observed. Its energy is given by eq 4. Accordingly, IR magnetospectroscopy must be coupled with another technique, HF-EPR in our case, to further determine the anisotropy parameters. The HF-EPR spectra were indeed simulated following two extreme scenarios (Table 4): (i) The rhombic anisotropy parameter E is set to zero, and the in-plane anisotropy is reproduced by introducing different in-plane Landé factors (scenario A). (ii) The in-plane Landé factors are kept equal, and the in-plane anisotropy is only reproduced by introducing the rhombic anisotropy parameter E. One can observe that, even when E reaches E/D = 0.14, the value of D decreases by only 3% (scenario B). In the case of 2, IR magnetospectroscopy thus appears to be a robust method for determination of the magnitude of |D|. On the contrary, there is no possibility to distinguish between the two

Figure 5. IR magnetoabsorbance of 2 recorded at the temperature of 1.6 K (a 3D plot is given in Figure S13).

Three field-independent peaks appear at 183, 207, and 217 cm−1, whereas the position of the fourth peak varies from 192 cm−1 at 0 T to 217 cm−1 at 13 T. The three first peaks are, hence, related to vibrational modes, whereas the latter one has a magnetic origin. This fourth transition is directly related to the ZFS (2D’) between the two Kramers doublets MS = ±3/2 and ±1/2 and is given by74,75 2D′ = 2 D2 + 3E2

(4)

Because only one field-dependent transition is observed, it is not possible to determine D and E independently from IR magnetospectroscopy. Comments on the Experimental Determination of Magnetic Anisotropy. The complete experimental determination of the magnetic anisotropy parameters is a key issue in the field of single-object magnets.77 Working on a series of complexes with different experimental techniques allows one to point out the good techniques or a set of techniques that should be used to reach reliable values of the ZFS parameters. dc magnetic susceptibility and magnetization measurements are currently commercial techniques with user-friendly interfaces. They are therefore the most accessible and widely used techniques. Nonetheless, the difficulty of extracting relevant anisotropy parameters from the experimental data is underestimated for at least two reasons. First, the experimental F


Article

Inorganic Chemistry Table 4. Comparison of the Anisotropy Parameters Found by Different Techniquesa 1

free parameters with constraints

alone

2

D, cm−1

E/D

g

−10.31 −7.28 (HF-EPR)

0.24 0.13

2.40 2.31

0.13

2.02 2.00 2.08

−7.28

with IR

a

D, cm−1

3 D, cm−1

E/D

g

2.50 2.73 2.73 2.23

5.80 4.920 (HF-EPR)

0.01 0.01

2.23 2.14

5.98 3.92 2.25 (geff) 2.48 2.48 2.37 2.92 1.95 2.23

4.920

0.28

2.20 2.22 2.15

E/D

Magnetization 24.17 92 (IR)

0.28 0.14

HF-EPR >22

92 (IR)

0.14

95 (IR)

0

g

When anisotropic, the Landé factors are ordered as gx, gy, and gz.

Figure 6. Main electronic configuration of the ground state and orbital energy diagram for compounds 1−3. The energy of the lowest orbital has been set to zero. The dashed lines indicate the average energy of the three t2g-like and two eg-like orbitals. The axes are those given on the right side of Figure 8 (the magnetic axes of compound 2).

atoms belonging to different (the same) hfa− ligand(s). With this choice, the dxz orbital solely interacts with N1-picoline and the dyz orbital with N2-picoline (Figures 1a and 7), but because the dxy orbital points toward the oxygen atoms, it is a eg-like orbital, and it is more destabilized than the dx2−y2. The metal complexes are close to the octahedral geometry (Oh). The splitting between the eg-like (dxy and dz2) and t2g-like (dxz, dyz, dx2−y2) orbitals decreases from 8440 cm−1 in 1 to 7400 cm−1 in 2 and 7250 cm−1 in 3 (Figure 6). This evolution follows the spectrochemical series of the metal ions. Let us start by discussing the σ effects of the ligands. In the case of 3, where the coordination sphere is very close to a perfect octahedron, the dz2 orbital is more destabilized than the dxy orbital because of the stronger σ-donating ability of the nitrogen atom of the picoline ligands compared to that of the oxygen atoms of the hfa− ligands. Then the increase of the M−N distances from 3 to 1 leads to a relative stabilization of the dz2 and a crossover of the two orbitals. To rationalize the variation of the M−ligand

extreme possibilities: coupling IR magnetospectroscopy and HF-EPR only gives intervals where the in-plane Landé factors and rhombic ZFS parameter can vary (Figure S12). Because the anisotropy of g and the rhombic parameter have the same origin, e.g., the spin−orbit coupling (SOC), and are both related to the λ tensor,78 the actual parameters must be given by an intermediate scenario and fall somewhere within these intervals. It is therefore interesting to investigate theoretically these compounds to see if this helps to unravel both contributions to the spectroscopic data. Theoretical Insights. Analysis of the Geometrical Changes along the Series. The first objective of the theoretical investigation deals with the rationalization of the geometrical variation around the metal ion along the series. In this respect, the main feature of the crystallographic study is the marked decrease of the M−N distances compared to the very limited variations of the M−O distances on going from 1 to 3 (Table 2). For all compounds, the z axis is chosen along the N−M−N bonds, whereas the x axis (y axis) lies between the two oxygen G


Article

Inorganic Chemistry

Table 5. Calculated ZFS Parameters (cm−1) and g Valuesa CASSCF D

Figure 7. Gmolden view of the t2g-like molecular orbitals dxy, dyz, and dxz from left to right. A small contour value of 0.007 was used.

distances along the series, one has to consider the t2g-like orbitals (Figure 7) to evaluate the π effects of the ligands. The dxz and dyz orbitals exhibit large antibonding interactions with the π system of the hfa− ligands, whereas there are almost no π interactions with the picoline ligands. Thus, the decrease of the M−N distances along the series is straightforward: it is related to the decrease of the ionic radius of the metal ion from 1 to 3, as was already observed in other consistent series of compounds involving the corresponding transition-metal ions.79 Concerning the M−O distances (Table 2), their variation results from two antagonistic effects: the decrease of the ionic radii, which tends to lower the distances, is balanced by the antibonding interactions with the hfa− ligand. Consequently, the M−O distance variation is much smaller than that of the M−N distance. The details of the variation of the M−O distances along the series must be related to the electronic configuration of the ions. Indeed, because in both complexes 2 and 3 the dxz and dyz orbitals are doubly occupied, the small decrease of the M−O distances is governed by the decrease of the ionic radius. On the contrary, with the dyz orbital being singly occupied in 1, there is lesser antibonding effect between iron(II) and the hfa− ligands and the M−O distances do not decrease between 1 and 2 (Table 2). Ab Initio Determination of the Magnetic Anisotropy Parameters. The values of the ZFS parameters D and E as well as the g values calculated by introducing the SOC operators are reported in Table 5, whereas the corresponding magnetic axes are given in Figure 8. As was already reported, the comparison between the results obtained at the CASSCF level and using the NEVPT2 method emphasizes the impact of the electronic correlation on the anisotropy parameters. The comparison with experiment indicates that the magnitude of both anisotropy parameters is retrieved all along the series. In order to rationalize the nature and magnitude of the magnetic anisotropy, the contribution of the few excited states that bring the largest contributions to D will be analyzed using a perturbative approach (at the second order) and the following simplified SOC Hamiltonian: Ĥ =

∑ ζi[lZî sẐ i + (l+̂ is−̂ i + l−̂ is+̂ i)/2] i

E

NEVPT2

E/D

1

9.65

2.72

0.28

2

127.4

6.06

0.05

3

4.25

1.00

0.24

g

D

E

E/D

g

gX = 2.15 gY = 2.21 gZ = 2.00 giso = 2.12 gX = 2.54 gY = 2.75 gZ = 1.71 giso = 2.33 gX = 2.26 gY = 2.28 gZ = 2.24 giso = 2.26

−8.34

2.68

0.32

113.4

3.41

0.03

4.01

1.13

0.28

gX = 2.12 gY = 2.18 gZ = 2.00 giso = 2.10 gX = 2.62 gY = 2.71 gZ = 1.77 giso = 2.37 gX = 2.23 gY = 2.25 gZ = 2.21 giso = 2.23

b

a The g axes are those of the ZFS parameters. bFrom a methodological point of view, it is interesting to note that while the fully contracted NEVPT2 method also provides a positive D value, the use of an uncontracted external space restores the correct sign of D. The rationalization for obtaining a positive value of D at both the CASSCF and fully contracted NEVPT2 levels is given in the Supporting Information.

Figure 8. Magnetic axes of 1 and 3 (left) and 2 (right). X is in green, Y in red, and Z in blue.

respectively. As was already shown in previous works,27,29,81 when the ground and excited states have the same spin, a coupling through the l+̂ ŝ− + l−̂ ŝ+(lẑ ŝz) part of the SOC operator results in a positive (negative) contribution to D, while one gets the opposite sign of the contribution if their spin differs (by 1 because the SOC given in eq 5 is a monoelectronic operator). Let us recall that the real orbitals decompose as functions of the Ylm 1 1 spherical harmonics as d Z2 = Y02 , d X2 − Y 2 = 2 Y22 + 2 Y −22 ,

(5)

where the sum runs over all electrons i, l ̂ and ŝ are the angular and spin momenta operators, and ζi are the spin−orbit constants. Let us note that the spin−orbit constants of the isolated ions are 410 cm−1 for iron(II), 533 cm−1 for cobalt(II), and 649 cm−1 for nickel(II).80 As can be seen in Figure 8, the magnetic axes are different from the crystallographic ones (y and z being interchanged) in compounds 1 and 3. For analysis purposes, the orbitals must be relabeled according to this new axis frame, i.e., dxz becomes dXY, dxy becomes dXZ, and d x 2 − y2 and d z 2 become d X2 − Z2 and d Y 2 ,

dXY = −

dYZ =

3 8

i 2

Y22 −

i Y2 2 1

dY 2 = − H

i 2

+

i 2

Y22 −

Y −22 ,

Y −21, 1 2 Y 2 0

−

dXZ = − d X2 − Z 2 = 3 8

1 8

Y22 −

1 Y2 2 1 3 2 Y 4 0

+ +

1 Y2 , 2 −1 1 Y2 , 8 −2

Y22Y −22 . Consequently, excitations DOI: 10.1021/acs.inorgchem.7b01861 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Table 6. Main Electronic Configuration of the Excited States That Bring the Largest Contributions to the Magnetic Anisotropya

a

Di and Ti stand for ith doublet and triplet excited states, respectively, and Qi (for 1) and Qi (for 2) stand for the ith quintet and quartet excited states, respectively. ΔE is the energy difference between the excited and ground states, and CD is the contribution to the D parameter of each excited state, given in cm−1. The molecular orbitals are ordered as in Figure 6.

even if higher than those of compound 2, and generate quite large contributions. Nevertheless, the overall magnetic anisotropy resulting from these opposite-sign contributions appears to be modest. For the three compounds, the theoretical investigation allows one to retrieve the trends in the sign and magnitude of the ZFS parameters found experimentally. Moreover, it identifies the excited states that strongly contribute to the ZFS parameters. In the case of 2, it highlights the existence of two contributions, one from the Landé factors and one from the rhombic anisotropy term, to the in-plane anisotropy. The theoretical values reproduce correctly the magnetization (Figure S9) curves, whereas they do not lead to satisfactory simulation of the HF-EPR spectra because, in the present case, they underestimate the global in-plane anisotropy. Dynamic Magnetic Properties. The high magnetic anisotropy of compounds 1−3 is an open door for the observation of slow relaxation of magnetization. Accordingly, the dynamic properties of compounds 1−3 were investigated using variable-temperature and variable-frequency ac susceptibility measurements. In zero dc field, no signal was observed for 1. When dc fields of up to 0.5 T are applied, a frequency-dependent out-of-phase signal χ′′ac appears (Figure S15). These results are consistent with (i) a negative value of the axial ZFS parameter D, which creates a barrier between the MS = ±2 states, and (ii) with the important value of the rhombic ZFS term E, which favors QTM when the ±MS states are degenerated, e.g., in zero dc field. The dynamic properties of the cobalt(II) derivative compound 2 are much richer. At zero dc field, no signal in the out-of-phase χ′ac was observed. When dc fields of up to 4000 Oe are applied, signals in χ′′ac can be detected with two well-defined maxima at 500 and 2000 Oe. In order to finely analyze this behavior, the frequency and field dependence of the ac susceptibility was mapped at 2 K (Figure 9a). A sharp maximum in the out-of-phase susceptibility χ′′ac can be observed for a dc field of 500 Oe and a frequency of 58 Hz (Figure 9b). A broader maximum of χ′′ac can be detected at 2000 Oe and 40 Hz (Figure 9d). There is a clearcut difference between these two maxima in both frequency (Figure 9c) and field (Figure 9e), indicating that they originate from two independent relaxation processes. The presence of two relaxation processes in a mononuclear cobalt(II) complex has been reported recently.82−86 In order to clarify the origin of these two relaxation processes, the variable-frequency (1− 1500 Hz) and variable-temperature (2−6 K) ac susceptibilities

involving any orbital of the sets (dXY and dX2−Y2) and (dZ2) lead to a zero SOC between the ground and excited states, excitations involving those of the sets (dXZ and dYZ) and (dXY, dX2−Y2, dZ2, dX2−Z2, and dY2) induce a coupling through the l+̂ ŝ− + l−̂ ŝ+ part of the operator, while other excitations result in a coupling through the lẐ ŝZ part. Finally let us recall that, from perturbative arguments, the contribution to the ZFS parameters is inversely proportional to the energy difference between the ground and excited states. For the three compounds, Table 6 gives the main configuration of the excited states that bring large contributions (CD) to D. For 1, magnetic anisotropy is essentially brought by the three first quintet excited states: Q0−Q1, Q0−Q2, and Q0−Q3. They essentially correspond to dXY → dYZ, dXY → dX2−Z2, and dXY → dY2 electron excitations from the ground state, respectively. Their contributions are relatively small because of the small value of the SOC constant of iron(II). As expected from the nature of the excitations, the contribution of the excited state Q1 is positive, while those of Q2 and Q3 are negative. The orientation of the easy axis then differs from one contribution to another; for instance, Q1 favors an easy axis in the Y direction, while Q2 favors the Z magnetic axis. These competing contributions finally result in a small value of D. The very large magnetic anisotropy of 2 is essentially brought by the two first quadruplet excited states (corresponding essentially to dXZ → dX2−Y2 and dYZ → dX2−Y2 electron excitations from the ground state, respectively). These excited states are almost degenerate with the ground state, in line with the quasi-degeneracy of the orbitals involved in the excitations. As for 1, the sign of the contributions is easily rationalized by looking at the orbitals involved in the excitations. In contrast with 1, both contributions are positive and add up with the contribution of the doublet excited state resulting from the jump of one electron from dXY to dX2−Y2. The large positive D value has therefore two origins: (i) the quasi-degeneracy of the orbitals involved in the excitations and (ii) the positive contributions of all excited states, bringing large contributions. Concerning compound 3, the three lowest triplet excited states (resulting from the lift of degeneracy of the T2g state in Oh) are responsible for the main contributions to the magnetic anisotropy. The two first triplets (corresponding essentially to the dX2−Z2 → dXZ and dYZ → dY2 excitations from the ground state, respectively) bring positive contributions to D that are partially compensated for by the third contribution coming from a dXY → dY2 excitation. All of these excited states are low in energy, I


Article

Inorganic Chemistry

Figure 9. Contour map (a) of the out-of-phase χ′′ac susceptibility versus applied dc fields (0−4000 Oe) and frequency of the ac field (Hac = 3.0 Oe) and respective sectional cuts (b−e) at a temperature of 2 K (see also the 3D plot in Figure S16).

that cobalt(II) is a Kramers ion94 and the last term to the Raman process. For the first fitting procedure, all of the parameters of this equation were left free. The simultaneous fitting of the temperature and field dependences of the relaxation times observed at Hdc = 500 Oe did not lead to relevant values of the variable parameters that appear in eq 6. This is an indication that phenomena beyond single-ion behavior, for example, intermolecular interactions, play a role in this relaxation process. To test this hypothesis, it would be necessary to dilute the cobalt(II) complex in a diamagnetic isostructural matrix. Unfortunately, the zinc(II) analogue of 2, [Zn(hfa)2(pic)2] crystallizes in a different space group, where the complex is in a cis configuration, preventing the use of the so-called dilution strategy. For the relaxation process at Hdc = 2000 Oe, the fitting procedure led to a good agreement with the experimental data (Figure 11) for τ0 = 7.04(7) × 10−7 s, Ueff = 24.7(1) K, A = 4.0(3) × 104 s−1 K−1 T−4, and B1 = 85(8) s−1 with a R value of 0.264. In these fits, the contributions of the HB2 and CTn terms are negligible compared to the other terms. In turn, setting these parameters to zero does not change either the quality of the fit or the values found for the relevant parameters (Table S5). Moreover, the value found for Ueff is in line with the value expected for the energy barrier (2E) to be overcome to reverse the in-plane magnetization as previously shown for cobalt(II)32 and rhenium(IV)95 complexes with S = 3/2 and in-plane magnetic anisotropy (D > 0; E ≠ 0).

at 500, 1125, 1250, and 2000 Oe were measured (Figures 10 and S17). At 500 and 2000 Oe, a generalized Debye model for one relaxation model82−86 was used to extract the relaxation time. At 1250 Oe, where two maxima in the out-of-phase susceptibility were detected (Figure 9c), the experimental data were fitted using an extended Debye model87−89 for two relaxation processes (Figure S17). We retrieve by this method the two relaxation times found at 500 and 2000 Oe, supporting the fact that the two relaxation processes evidenced at 500 and 2000 Oe occur simultaneously for intermediate fields without any interference. The temperature dependence of the relaxation times (τT) extracted from the variable-temperature ac susceptibility at 500 and 2000 Oe as well the relaxation times (τH) obtained from the variable-field ac susceptibility are presented in Figures 11 and S18. The origin of slow relaxation of magnetization in mononuclear S = 3/2 complexes is still a matter of debate.32,34,76,90−93 The most general expression for the relaxation time of a single-ion system is given by the following equation:94 τT/H−1(T , H ) =

⎛U ⎞ B1 + τ0−1 exp⎜ eff ⎟ + AH 4T + CT n ⎝ kT ⎠ 1 + HB2 (6)

where the first term is ascribed to QTM, the second term to a thermally activated Orbach process, the third term to direct spin phonon relaxation with a field exponent fixed to 4 owing to the fact J


Article

Inorganic Chemistry

Figure 10. Variable-temperature in-phase (top) and out-of-phase (bottom) components of the ac magnetic susceptibility data for 2 collected in a 3.0 Oe ac oscillating field (1−1500 Hz) between 2.0 and 5.5 K, applying 500 Oe (left) and 2000 Oe (right) dc fields. Solids lines correspond to the best fits by the generalized Debye model (see the text).

Nonetheless, other proposals were made in the literature91−93,96,97 to interpret the slow relaxation of magnetization of mononuclear complexes. Accordingly, we have examined these possibilities by introducing different constraints in the fitting procedure (Table S5). The robustness of Ueff, even when a complementary Raman process is included, together with its compatibility with the value expected from the determination of the rhombic anisotropy (2E),32,97 is a good indication that this process is at work in the relaxation of magnetization of 2, but it is impossible to affirm that it is the only present process. The magnetic properties are further investigated by μ-SQUID magnetometry. The μ-SQUID measurements on compounds 1−3 are presented in Figures 12 and S20. In the case of 1, a small hysteresis loop can be observed at low temperature. The shape of the hysteresis loops is consistent with the typical behavior of SIMs with negative ZFS and QTM. For 2, the hysteretic behavior is observed from 0.2 up to 1 K. The combination of the ac susceptibility and μ-SQUID measurements allows one to assess the presence of an Orbach relaxation process related to the existence of an easy-plane anisotropy.32,97

Figure 11. (a) Temperature dependence of the relaxation time for 2 at 2000 Oe. Diamonds (triangles) represent relaxation times extracted from the frequency (temperature) sweeping ac experiments. (b) Field dependence of the relaxation time for 2 at 2 K. Both curves were fitted with eq 6: the three main processes are indicated as dotted lines, and the best fit parameters are given in the text.

microwave frequencies of up to 662 GHz, HF-EPR at different temperatures appears as the most reliable method to determine the magnetic anisotropy parameters of the iron(II) (D = −7.28 cm−1; E = 0.946 cm−1) and nickel(II) (D = 4.920 cm−1; E = 1.378 cm−1) derivatives because, with an absolute value of the D parameter below 22 cm−1, the transitions observed in HF-EPR give direct access to the anisotropy parameters. With D above this limit, HF-EPR must be associated with other techniques like IR magnetospectroscopy, magnetometry, and theoretical calculations to estimate the magnetic anisotropy parameters. For the cobalt(II) derivative, this approach led to a good estimation of the sign and magnitude of D (∼95 cm−1), whereas for E, the theoretical value of 3.56 cm−1 falls within the interval determined by HF-EPR. With a negative D parameter, the iron(II) derivative fulfills the main condition for observation of SIM properties. It indeed shows a frequency-dependent out-of-phase signal of the ac magnetic susceptibility under an applied dc field of 0.1 T. For the cobalt(II) derivative, two dc field-dependent relaxation phenomena were observed. At low dc field (500 Oe), the relaxation

■

CONCLUSION Within the series of trans-[M(hfa)2(H2O)2] (Hhfa = 1,1,1,5,5,5hexafluoro-2,4-pentanedione; M = FeII, CoII, NiII) complexes, a dramatic elongation of the Fe−N distance (2.19 Å) compared to the Fe−O main distance (2.06 Å) is observed. This distortion gradually decreases upon going to the cobalt(II) and nickel(II) derivatives. Following the results of the theoretical study, the evolution of the M−N distances is directly related to the variation of the ionic radius of the metal center upon going from iron(II) to nickel(II), whereas for the M−O distances, this effect is balanced by the filling of the t2g-like molecular orbitals that are antibonding with the hfa− ligands. As a key parameter of the SIM properties, the magnetic anisotropy has been studied within this series using a multitechniques approach. With a maximum field of 16 T and K


Article

Inorganic Chemistry

Figure 12. Temperature dependence of the hysteresis loops recorded for 1 (left) and 2 (right).

process origin is beyond single-ion behavior. At high dc field (2000 Oe), the relaxation of magnetization is believed to originate from several processes including an Orbach mechanism and QTM. Because this complex has a positive D value, these relaxation processes are related to the in-plane anisotropy of the complex and the rhombic magnetic anisotropy drives the effective barrier of the Orbach mechanism. Such slow relaxation of magnetization has been reported in comparable slightly distorted octahedral complexes.32,97 The robustness of the magnetic anisotropy upon ligand exchange is an invitation to use these types of complexes in the complex-as-ligand strategy toward original SMM and singlechain magnets (SCM).

■

Recherche for her doctoral grant within Project ANR-13-BS080016-01 (CHIRCURIE).

■

(1) Sessoli, R.; Gatteschi, D.; Caneschi, A.; Novak, M. A. Magnetic bistability in a metal-ion cluster. Nature 1993, 365, 141−143. (2) Gatteschi, D.; Sessoli, R. Quantum tunneling of magnetization and related phenomena in molecular materials. Angew. Chem., Int. Ed. 2003, 42, 268−297. (3) Gatteschi, D.; Sessoli, R.; Villain, J. Molecular Nanomagnets; Scholarship Online: Oxford, U.K., 2006. (4) Aromi, G.; Brechin Heidelberg, E. K. Synthesis of 3d metallic single-molecule magnets. In Single-Molecule Magnets and Related Phenomena; Winpenny, R., Ed.; Springer-Verlag: Berlin Heidelberg, 2006; Vol. 122; pp 1−69. (5) Sessoli, R.; Powell, A. K. Strategies towards single molecule magnets based on lanthanide ions. Coord. Chem. Rev. 2009, 253, 2328− 2341. (6) Glaser, T.; Liratzis, I.; Ako, A. M.; Powell, A. K. 2,6Bis(hydroxymethyl)phenols for the synthesis of high-nuclearity clusters. Coord. Chem. Rev. 2009, 253, 2296−2305. (7) Gatteschi, D. Molecular Magnetism − a basis for new materials. Adv. Mater. 1994, 6, 635−645. (8) Andruh, M.; Costes, J. P.; Diaz, C.; Gao, S. 3d-4f Combined Chemistry: Synthetic Strategies and Magnetic Properties. Inorg. Chem. 2009, 48, 3342−3359. (9) Kortz, U.; Muller, A.; van Slageren, J.; Schnack, J.; Dalal, N. S.; Dressel, M. Polyoxometalates: Fascinating structures, unique magnetic properties. Coord. Chem. Rev. 2009, 253, 2315−2327. (10) Gatteschi, D.; Caneschi, A.; Pardi, L.; Sessoli, R. Large clusters of metal-ion − the transition from molecular to bulk magnets. Science 1994, 265, 1054−1058. (11) Aromi, G.; Knapp, M. J.; Claude, J. P.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. High-spin molecules: Hexanuclear Mn(III) clusters with [Mn6O4X4]6+ (X = Cl−, Br−) face-capped octahedral cores and S = 12 ground states. J. Am. Chem. Soc. 1999, 121, 5489−5499. (12) Murugesu, M.; Wernsdorfer, F.; Abboud, K. A.; Christou, G. New structural motifs in manganese single-molecule magnetism from the use of triethanolamine ligands. Angew. Chem., Int. Ed. 2005, 44, 892−896. (13) Scott, R. T. W.; Parsons, S.; Murugesu, M.; Wernsdorfer, W.; Christou, G.; Brechin, E. K. Linking centered manganese triangles into larger clusters: A {Mn32} truncated cube. Angew. Chem., Int. Ed. 2005, 44, 6540−6543. (14) Stamatatos, T. C.; Abboud, K. A.; Wernsdorfer, W.; Christou, G. High-nuclearity, high-symmetry, high-spin molecules: A mixed-valence Mn10 cage possessing rare T symmetry and an S = 22 ground state. Angew. Chem., Int. Ed. 2006, 45, 4134−4137. (15) Stamatatos, T. C.; Abboud, K. A.; Wernsdorfer, W.; Christou, G. ″Spin tweaking″ of a high-spin molecule: An Mn25 single-molecule magnet with an S = 61/2 ground state. Angew. Chem., Int. Ed. 2007, 46, 884−888. (16) Stamatatos, T. C.; Abboud, K. A.; Wernsdorfer, W.; Christou, G. Covalently linked dimers of clusters: Loop- and dumbbell-shaped Mn24

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b01861. IR spectra of 1−4, crystallographic information for 1−4 (powder X-ray diffraction pattern, tables of bond distances and angles, and crystal packing), complementary EPR spectra, complementary fits of magnetization data, details of the theoretical inspection, complementary ac susceptibility data and fits for 1 and 2, and μ-SQUID loops recorded for 3 (PDF) Accession Codes

CCDC 1558350−1558352 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Ghénadie Novitchi: 0000-0002-6109-6937 Shangda Jiang: 0000-0003-0204-9601 Cyrille Train: 0000-0002-6726-2086 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors warmly thank Prof. Roberta Sessoli for allowing them to do preliminary ac magnetic susceptibility measurements in her laboratory. F.R. acknowledges the Agence Nationale de la L


Article

Inorganic Chemistry

(S = 1/2) core complexes, [FeIV(Et2dtc)3−n(mnt)n](n−1)− and [FeIII(Et2dtc)3−n(mnt)n]n− (n = 0, 1, 2, 3): The origin of the magnetic anisotropy. Coord. Chem. Rev. 2013, 257, 27−41. (36) Riblet, F.; Novitchi, G.; Scopelliti, R.; Helm, L.; Gulea, A.; Merbach, A. E. Isomerization Mechanisms of Stereo labile tris- and bisBidentate Octahedral Cobalt(II) Complexes: X-ray Structure and Variable. Inorg. Chem. 2010, 49, 4194. (37) Izumi, F.; Kurosawa, R.; Kawamoto, H.; Akaiwa, H. Adduct of bis(hexafluoroacetylacetonato)metal(II) with unidentate and bidentate nitrogenous bases. Bull. Chem. Soc. Jpn. 1975, 48, 3188−3192. (38) Dikarev, E. V.; Zhang, H.; Li, B. Heterometallic bismuthtransition metal homoleptic beta-diketonates. J. Am. Chem. Soc. 2005, 127, 6156−6157. (39) CrysAlis RED, version 1.171.34.76; Oxford Diffraction Ltd., 2003. (40) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.; Puschmann, H. OLEX2: a complete structure solution, refinement and analysis program. J. Appl. Crystallogr. 2009, 42, 339−341. (41) Sheldrick, G. A short history of SHELX. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64, 112−122. (42) Pascal, P. Magnochemical studies. Ann. Chim. Phys. 1910, 19, 5− 70. (43) Kahn, O. Molecular Magnetism; VCH Publishers, Inc.: New York, 1993. (44) Dekker, C.; Arts, A. F. M.; de Wijn, H. W.; van Duyneveldt, A. J.; Mydosh, J. A. Activated dynamics in a two-dimensional Ising spin glass: Rb2Cu1−xCoxF4. Phys. Rev. B: Condens. Matter Mater. Phys. 1989, 40, 11243. (45) Cole, K. S.; Cole, R. H. Dispersion and absorption in dielectrics I. Alternating current characteristics. J. Chem. Phys. 1941, 9, 341−351. (46) Wernsdorfer, W. Classical and quantum magnetization reversal studied in nanometer-sized particles and clusters. Adv. Chem. Phys. Vol. 118 2001, 118, 99−190. (47) Wernsdorfer, W. From micro- to nano-SQUIDs: applications to nanomagnetism. Supercond. Sci. Technol. 2009, 22, 064013. (48) Barra, A. L.; Hassan, A. K.; Janoschka, A.; Schmidt, C. L.; Schunemann, V. Broad-band quasi-optical HF-EPR spectroscopy: Application to the study of the ferrous iron center from a rubredoxin mutant. Appl. Magn. Reson. 2006, 30, 385−397. (49) Glerup, J.; Weihe, H. Magnetic susceptibility and EPR spectra of (mu-hydroxo)bis pentaamminechromium(III) chloride monohydrate. Inorg. Chem. 1997, 36, 2816−2819. (50) Neese, F. The ORCA program system. Wiley Interdiscip. Rev. Comput. Mol. Sci. 2012, 2, 73−78. (51) Angeli, C.; Bories, B.; Cavallini, A.; Cimiraglia, R. Third-order multireference perturbation theory: The n-electron valence state perturbation-theory approach. J. Chem. Phys. 2006, 124, 054108. (52) Angeli, C.; Cimiraglia, R.; Evangelisti, S.; Leininger, T.; Malrieu, J. P. Introduction of n-electron valence states for multireference perturbation theory. J. Chem. Phys. 2001, 114, 10252−10264. (53) Angeli, C.; Cimiraglia, R.; Malrieu, J. P. N-electron valence state perturbation theory: a fast implementation of the strongly contracted variant. Chem. Phys. Lett. 2001, 350, 297−305. (54) Angeli, C.; Cimiraglia, R.; Malrieu, J. P. n-electron valence state perturbation theory: A spinless formulation and an efficient implementation of the strongly contracted and of the partially contracted variants. J. Chem. Phys. 2002, 117, 9138−9153. (55) Schafer, A.; Horn, H.; Ahlrichs, R. Fully optimized contracted gaussian-basis sets for atoms Li to Kr. J. Chem. Phys. 1992, 97, 2571− 2577. (56) Schafer, A.; Huber, C.; Ahlrichs, R. Fully optimized contracted gaussian-basis sets of triple zeta valence quality for atoms Li to Kr. J. Chem. Phys. 1994, 100, 5829−5835. (57) Weigend, F.; Ahlrichs, R. Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7, 3297−3305. (58) Bloch, C. Sur la theorie des perturbations des etats lies. Nucl. Phys. 1958, 6, 329−347. (59) Descloizeaux, J. Extension d’une formule de Lagrange a des pronlemes de valeurs propres. Nucl. Phys. 1960, 20, 321−346.

and Mn26 single-molecule magnets. Angew. Chem., Int. Ed. 2008, 47, 6694−6698. (17) Bagai, R.; Christou, G. The Drosophila of single-molecule magnetism: [Mn12O12(O2CR)16(H2O)4]. Chem. Soc. Rev. 2009, 38, 1011−1026. (18) Boskovic, C.; Brechin, E. K.; Streib, W. E.; Folting, K.; Bollinger, J. C.; Hendrickson, D. N.; Christou, G. Single-molecule magnets: A new family of Mn12 clusters of formula [Mn12O8X4(O2CPh)8L6]. J. Am. Chem. Soc. 2002, 124, 3725−3736. (19) Boskovic, C.; Pink, M.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. Single-molecule magnets: Ligand-induced core distortion and multiple Jahn-Teller isomerism in [Mn12O12(O2CMe)8(O2PPh2)8(H2O)4]. J. Am. Chem. Soc. 2001, 123, 9914−9915. (20) Brechin, E. K.; Boskovic, C.; Wernsdorfer, W.; Yoo, J.; Yamaguchi, A.; Sanudo, E. C.; Concolino, T. R.; Rheingold, A. L.; Ishimoto, H.; Hendrickson, D. N.; Christou, G. Quantum tunneling of magnetization in a new [Mn18]2+ single-molecule magnet with S = 13. J. Am. Chem. Soc. 2002, 124, 9710−9711. (21) Waldmann, O. A criterion for the anisotropy barrier in singlemolecule magnets. Inorg. Chem. 2007, 46, 10035−10037. (22) Neese, F.; Pantazis, D. A. What is not required to make a single molecule magnet. Faraday Discuss. 2011, 148, 229−238. (23) Atanasov, M.; Aravena, D.; Suturina, E.; Bill, E.; Maganas, D.; Neese, F. First principles approach to the electronic structure, magnetic anisotropy and spin relaxation in mononuclear 3d-transition metal single molecule magnets. Coord. Chem. Rev. 2015, 289−290, 177−214. (24) Zadrozny, J. M.; Long, J. R. Slow Magnetic Relaxation at Zero Field in the Tetrahedral Complex Co(SPh)42‑. J. Am. Chem. Soc. 2011, 133, 20732−20734. (25) Zadrozny, J. M.; Xiao, D. J.; Atanasov, M.; Long, G. J.; Grandjean, F.; Neese, F.; Long, J. R. Magnetic blocking in a linear iron(I) complex. Nat. Chem. 2013, 5, 577−581. (26) Freedman, D. E.; Harman, W. H.; Harris, T. D.; Long, G. J.; Chang, C. J.; Long, J. R. Slow Magnetic Relaxation in a High-Spin Iron(II) Complex. J. Am. Chem. Soc. 2010, 132, 1224−1225. (27) Ruamps, R.; Batchelor, L. J.; Guillot, R.; Zakhia, G.; Barra, A. L.; Wernsdorfer, W.; Guihery, N.; Mallah, T. Ising-type magnetic anisotropy and single molecule magnet behaviour in mononuclear trigonal bipyramidal Co(II) complexes. Chem. Sci. 2014, 5, 3418−3424. (28) Rebilly, J. N.; Charron, G.; Riviere, E.; Guillot, R.; Barra, A. L.; Serrano, M. D.; van Slageren, J.; Mallah, T. Large magnetic anisotropy in pentacoordinate Ni(II) complexes. Chem. - Eur. J. 2008, 14, 1169−1177. (29) Ruamps, R.; Batchelor, L. J.; Maurice, R.; Gogoi, N.; JimenezLozano, P.; Guihery, N.; de Graaf, C.; Barra, A.-L.; Sutter, J.-P.; Mallah, T. Origin of the Magnetic Anisotropy in Heptacoordinate Ni(II) and Co(II) Complexes. Chem. - Eur. J. 2013, 19, 950−956. (30) Ruamps, R.; Maurice, R.; Batchelor, L.; Boggio-Pasqua, M.; Guillot, R.; Barra, A. L.; Liu, J.; Bendeif, E.-E.; Pillet, S.; Hill, S.; Mallah, T.; Guihery, N. Giant Ising-Type Magnetic Anisotropy in Trigonal Bipyramidal Ni(II) Complexes: Experiment and Theory. J. Am. Chem. Soc. 2013, 135, 3017−3026. (31) Zadrozny, J. M.; Liu, J.; Piro, N. A.; Chang, C. J.; Hill, S.; Long, J. R. Slow magnetic relaxation in a pseudotetrahedral cobalt(II) complex with easy-plane anisotropy. Chem. Commun. 2012, 48, 3927−3929. (32) Vallejo, J.; Castro, I.; Ruiz-García, R.; Cano, J.; Julve, M.; Lloret, F.; De Munno, G.; Wernsdorfer, W.; Pardo, E. Field-Induced Slow Magnetic Relaxation in a Six-Coordinate Mononuclear Cobalt(II) Complex with a Positive Anisotropy. J. Am. Chem. Soc. 2012, 134, 15704−15707. (33) Gomez-Coca, S.; Cremades, E.; Aliaga-Alcalde, N.; Ruiz, E. Mononuclear Single-Molecule Magnets: Tailoring the Magnetic Anisotropy of First-Row Transition-Metal Complexes. J. Am. Chem. Soc. 2013, 135, 7010−7018. (34) Gomez-Coca, S.; Urtizberea, A.; Cremades, E.; Alonso, P. J.; Camon, A.; Ruiz, E.; Luis, F. Origin of slow magnetic relaxation in Kramers ions with non-uniaxial anisotropy. Nat. Commun. 2014, 5, 4300. (35) Atanasov, M.; Surawatanawong, P.; Wieghardt, K.; Neese, F. A theoretical study of zero-field splitting in Fe(IV)S6 (S = 1) and Fe(III)S6 M


Article

Inorganic Chemistry (60) Maurice, R.; Bastardis, R.; de Graaf, C.; Suaud, N.; Mallah, T.; Guihery, N. Universal Theoretical Approach to Extract Anisotropic Spin Hamiltonians. J. Chem. Theory Comput. 2009, 5, 2977−2984. Riblet, F.; Novitchi, G.; Scopelliti, R.; Helm, L.; Gulea, A.; Merbach, A. E. Temperature and Pressure NMR Kinetic Investigations. Inorg. Chem. 2010, 49, 4194−4211. (61) Wells, A. F. Structural inorganic chemistry; Oxford University Press, 2012. (62) Baker, R. J.; Colavita, P. E.; Murphy, D. M.; Platts, J. A.; Wallis, J. D. Fluorine-Fluorine Interactions in the Solid State: An Experimental and Theoretical Study. J. Phys. Chem. A 2012, 116, 1435−1444. (63) Carver, G.; Tregenna-Piggott, P. L. W.; Barra, A. L.; Neels, A.; Stride, J. A. Spectroscopic and structural characterization of the Fe(imidazole)62+ cation. Inorg. Chem. 2003, 42, 5771−5777. (64) Whittaker, J. W.; Solomon, E. I. Spectroscopic studies onferrous non-heme iron active-sites − magnetic circular-dichroism of mononuclear Fe sites in superoxide-dismutase. J. Am. Chem. Soc. 1988, 110, 5329−5339. (65) Ozarowski, A.; Zvyagin, S. A.; Reiff, W. M.; Telser, J.; Brunel, L. C.; Krzystek, J. High-frequency and -field EPR of a pseudo-octahedral complex of high-spin Fe(II): Bis(2,2 ’-bi-2-thiazoline)bis(isothiocyanato)iron(II). J. Am. Chem. Soc. 2004, 126, 6574−6575. (66) Titiš, J.; Boča, R. Magnetostructural D Correlations in Hexacoordinated Cobalt(II) Complexes. Inorg. Chem. 2011, 50, 11838−11845. (67) Titiš, J.; Hudák, J.; Kožíšek, J.; Krutošíková, A.; Moncol’, J.; Tarabová, D.; Boča, R. Structural, spectral and magnetic properties of carboxylato cobalt(II) complexes with heterocyclic N-donor ligands: Reconstruction of magnetic parameters from electronic spectra. Inorg. Chim. Acta 2012, 388, 106−113. (68) Papánková, B.; Boča, R.; Dlháň, L.; Nemec, I.; Titiš, J.; Svoboda, I.; Fuess, H. Magneto-structural relationships for a mononuclear Co(II) complex with large zero-field splitting. Inorg. Chim. Acta 2010, 363, 147−156. (69) Rajnák, C.; Titiš, J.; Boča, R.; Moncol’, J.; Padělková, Z. Selfassembled cobalt(II) Schiff base complex: synthesis, structure, and magnetic properties. Monatsh. Chem. 2011, 142, 789−795. (70) Titiš, J.; Boča, R. Magnetostructural D Correlation in Nickel(II) Complexes: Reinvestigation of the Zero-Field Splitting. Inorg. Chem. 2010, 49, 3971−3973. (71) Boca, R.; Titiš, J. Magnetostructural D-correlations for zero-field splitting in nickel (II) complexes. Coord. Chem. Res. Prog. 2008, 247. (72) Singh, S. K.; Gupta, T.; Badkur, P.; Rajaraman, G. Magnetic Anisotropy of Mononuclear NiII Complexes: On the Importance of Structural Diversity and the Structural Distortions. Chem. - Eur. J. 2014, 20, 10305−10313. (73) Atanasov, M.; Comba, P.; Helmle, S.; Muller, D.; Neese, F. ZeroField Splitting in a Series of Structurally Related Mononuclear Ni-IIBispidine Complexes. Inorg. Chem. 2012, 51, 12324−12335. (74) Shores, M. P.; Sokol, J. J.; Long, J. R. Nickel(II)-molybdenum(III)-cyanide clusters: Synthesis and magnetic behavior of species incorporating [(Me3tacn)Mo(CN)3]. J. Am. Chem. Soc. 2002, 124, 2279−2292. (75) Suturina, E. A.; Nehrkorn, J.; Zadrozny, J. M.; Liu, J.; Atanasov, M.; Weyhermüller, T.; Maganas, D.; Hill, S.; Schnegg, A.; Bill, E.; Long, J. R.; Neese, F. Magneto-Structural Correlations in Pseudotetrahedral Forms of the [Co(SPh)4]2− Complex Probed by Magnetometry, MCD Spectroscopy, Advanced EPR Techniques, and ab Initio Electronic Structure Calculations. Inorg. Chem. 2017, 56, 3102−3118. (76) Rechkemmer, Y.; Breitgoff, F. D.; van der Meer, M.; Atanasov, M.; Hakl, M.; Orlita, M.; Neugebauer, P.; Neese, F.; Sarkar, B.; van Slageren, J. A four-coordinate cobalt(II) single-ion magnet with coercivity and a very high energy barrier. Nat. Commun. 2016, 7, 10467. (77) Barra, A. L.; Caneschi, A.; Cornia, A.; Gatteschi, D.; Gorini, L.; Heiniger, L. P.; Sessoli, R.; Sorace, L. The origin of transverse Anisotropy in axially symmetric single molecule magnets. J. Am. Chem. Soc. 2007, 129, 10754−10762. (78) Theoretical foundations of molecular magnetism; Boča, R., Ed.; Elsevier Science, Lausanne, 1999.

(79) Desrochers, P. J.; Brown, J. R.; Arvin, M. E.; Jones, G. D.; Vicic, D. A. Bis hydrotris(4-chloro-3,5-dimethylpyrazolyl)borato nickel(II). Acta Crystallogr., Sect. E: Struct. Rep. Online 2005, 61, M1455−M1458. (80) Reddy, S. L.; Endo, T.; Reddy, G. S. Electronic (absorption) spectra of 3d transition metal complexes. In Advanced Aspects of Spectroscopy; Farrukh, M. A., Ed.; InTech: Rijeka, 2012; pp 3−48. (81) Cahier, B.; Perfetti, M.; Zakhia, G.; Naoufal, D.; El-Khatib, F.; Guillot, R.; Riviere, E.; Sessoli, R.; Barra, A. L.; Guihery, N.; Mallah, T. Magnetic Anisotropy in Pentacoordinate Ni(II) and Co(II) Complexes: Unraveling Electronic and Geometrical Contributions. Chem. - Eur. J. 2017, 23, 3648−3657. (82) Buvaylo, E. A.; Kokozay, V. N.; Vassilyeva, O. Y.; Skelton, B. W.; Ozarowski, A.; Titiš, J.; Vranovičová, B.; Boča, R. Field-Assisted Slow Magnetic Relaxation in a Six-Coordinate Co(II)−Co(III) Complex with Large Negative Anisotropy. Inorg. Chem. 2017, 56, 6999−7009. (83) Rajnak, C.; Varga, F.; Titis, J.; Moncol, J.; Boca, R. FieldSupported Single-Molecule Magnets of Type Co(bzimpy)X2. Eur. J. Inorg. Chem. 2017, 2017, 1915−1922. (84) Rajnak, C.; Titis, J.; Fuhr, O.; Ruben, M.; Boca, R. SingleMolecule Magnetism in a Pentacoordinate Cobalt(II) Complex Supported by an Antenna Ligand. Inorg. Chem. 2014, 53, 8200−8202. (85) Palii, A. V.; Korchagin, D. V.; Yureva, E. A.; Akimov, A. V.; Misochko, E. Y.; Shilov, G. V.; Talantsev, A. D.; Morgunov, R. B.; Aldoshin, S. M.; Tsukerblat, B. S. Single-Ion Magnet Et4N[CoII(hfac)3] with Nonuniaxial Anisotropy: Synthesis, Experimental Characterization, and Theoretical Modeling. Inorg. Chem. 2016, 55, 9696−9706. (86) Boca, R.; Miklovic, J.; Titis, J. Simple Mononuclear Cobalt(II) Complex: A Single-Molecule Magnet Showing Two Slow Relaxation Processes. Inorg. Chem. 2014, 53, 2367−2369. (87) Guo, Y.-N.; Xu, G.-F.; Guo, Y.; Tang, J. Relaxation dynamics of dysprosium(III) single molecule magnets. Dalton Trans. 2011, 40, 9953−9963. (88) Guo, Y. N.; Xu, G. F.; Wernsdorfer, W.; Ungur, L.; Guo, Y.; Tang, J. K.; Zhang, H. J.; Chibotaru, L. F.; Powell, A. K. Strong Axiality and Ising Exchange Interaction Suppress Zero-Field Tunneling of Magnetization of an Asymmetric Dy2 Single-Molecule Magnet. J. Am. Chem. Soc. 2011, 133, 11948−11951. (89) Guo, Y.-N.; Xu, G.-F.; Gamez, P.; Zhao, L.; Lin, S.-Y.; Deng, R.; Tang, J.; Zhang, H.-J. Two-Step Relaxation in a Linear Tetranuclear Dysprosium(III) Aggregate Showing Single-Molecule Magnet Behavior. J. Am. Chem. Soc. 2010, 132, 8538−8539. (90) Frost, J. M.; Harriman, K. L. M.; Murugesu, M. The rise of 3d single-ion magnets in molecular magnetism: towards materials from molecules? Chem. Sci. 2016, 7, 2470−2491. (91) Zhang, Y.-Z.; Gomez-Coca, S.; Brown, A. J.; Saber, M. R.; Zhang, X.; Dunbar, K. R. Trigonal antiprismatic Co(II) single molecule magnets with large uniaxial anisotropies: importance of Raman and tunneling mechanisms. Chem. Sci. 2016, 7, 6519−6527. (92) Novikov, V. V.; Pavlov, A. A.; Nelyubina, Y. V.; Boulon, M.-E.; Varzatskii, O. A.; Voloshin, Y. Z.; Winpenny, R. E. P. A Trigonal Prismatic Mononuclear Cobalt(II) Complex Showing Single-Molecule Magnet Behavior. J. Am. Chem. Soc. 2015, 137, 9792−9795. (93) Colacio, E.; Ruiz, J.; Ruiz, E.; Cremades, E.; Krzystek, J.; Carretta, S.; Cano, J.; Guidi, T.; Wernsdorfer, W.; Brechin, E. K. Slow Magnetic Relaxation in a CoII−YIII Single-Ion Magnet with Positive Axial ZeroField Splitting. Angew. Chem., Int. Ed. 2013, 52, 9130−9134. (94) Abragam, A.; Bleaney, B. Electron paramagnetic resonance of transition ions; Oxford University Press, 2012. (95) Feng, X.; Liu, J.; Harris, T. D.; Hill, S.; Long, J. R. Slow Magnetic Relaxation Induced by a Large Transverse Zero-Field Splitting in a MnIIReIV(CN)2 Single-Chain Magnet. J. Am. Chem. Soc. 2012, 134, 7521−7529. (96) Zadrozny, J. M.; Atanasov, M.; Bryan, A. M.; Lin, C.-Y.; Rekken, B. D.; Power, P. P.; Neese, F.; Long, J. R. Slow magnetization dynamics in a series of two-coordinate iron(II) complexes. Chem. Sci. 2013, 4, 125−138. (97) Atanasov, M.; Zadrozny, J. M.; Long, J. R.; Neese, F. A theoretical analysis of chemical bonding, vibronic coupling, and magnetic anisotropy in linear iron(II) complexes with single-molecule magnet behavior. Chem. Sci. 2013, 4, 139−156. N


From Positive to Negative Zero-Field Splitting in a Series of Strongly

Recommend Documents