Cobalt(II)-Based Single-Ion Magnets with Distorted Pseudotetrahedral

Apr 5, 2016 - ... 1298 (s), 1210 (m), 1166 (s), 1153 (s), 1140 (s), 1109 (w), 968 (m), 859 (m), 829 (s), 769 (s), 749 (vs), 731 (s), 714 (vs), 454 (s)...
2 downloads 0 Views 2MB Size
Article pubs.acs.org/IC

Cobalt(II)-Based Single-Ion Magnets with Distorted Pseudotetrahedral [N2O2] Coordination: Experimental and Theoretical Investigations Sven Ziegenbalg,† David Hornig,† Helmar Görls, and Winfried Plass* Institut für Anorganische und Analytische Chemie, Friedrich-Schiller-Universität Jena, Humboldtstrasse 8, 07743 Jena, Germany S Supporting Information *

ABSTRACT: The synthesis and magnetic properties of cobalt(II) complexes with sterically demanding Schiff-base ligands are reported. The compounds [Co(LBr)2] (1) and [Co(LPh)2]·CH2Cl2 (2·CH2Cl2) are obtained by the reaction of cobalt(II) acetate with the ligands HLBr and HLPh in a dichloromethane/methanol mixture. 1 and 2 crystallize in the space groups P21212 and P1̅, respectively. X-ray diffraction studies revealed mononuclear constitution of both complexes. For 1, relatively short intermolecular Co−Co distances of 569 pm are observed. In compound 2, a hydrogen-bonded dichloromethane molecule is present, leading to a solvent aggregate with remarkable thermal stability for which desolvation is taking place between 150 and 210 °C. Magnetic measurements were performed to determine the zero-field-splitting (ZFS) parameter D for both complexes. Frequency-dependent susceptibility measurements revealed slow magnetic relaxation behavior with spin-reversal barriers of 36 cm−1 for 1 and 43 cm−1 for 2 at an applied external field of 400 Oe. This observation is related to an increasing distortion of the pseudotetrahedral coordination geometry for complex 2. These distortions can be decomposed in two major contributions. One is the elongation effect described by the parameter ϵT, which is the ratio of the averaged obtuse and acute bond angles. The other effect is related to a twisting distortion of the chelate coordination planes at the cobalt center. A comparison with literature examples reveals that the elongation effect seems to govern the overall magnetic behavior in pseudotetrahedral complexes with two bidentate chelate ligands. Ab initio calculations for complexes 1 and 2 using the CASPT2 method show strong splitting of the excited 4T2 term, which explains the observed strong ZFS. Spin−orbit calculations with the RASSI-SO method confirm the single-molecule-magnet behavior because only small transversal elements are found for the lowest Kramers doublet for both complexes.



Mononuclear SMMs as an alternative strategy were first introduced by Ishikawa et al. on the basis of lanthanide complexes,10 which has led to a series of new compounds.11 It was only recently that the pioneering work of Long et al. on low-coordinate high-spin iron(II) complexes superbly disclosed the general potential of 3d transition metals to generate singleion magnets (SIMs).12 In the past few years, this has led to the discovery of an increasing number of new SIMs based on firstrow transition-metal ions.13 Besides iron,14 manganese,15 and nickel,16 a strong focus was placed on high-spin cobalt(II) complexes because the S = 3/2 half-integer spin state should minimize quantum-tunneling effects and lead to favorable relaxation behavior.6,17 Although different coordination geometries and donor environments have been reported for cobalt(II)-based SIMs, the vast majority contain tetrahedral or pseudotetrahedral coordination.18−31 For higher coordination numbers of 5 and 6, only a few complexes have been reported, which comprise examples with square-pyramidal32,33 and trigonal-bipyramidal

INTRODUCTION The seminal discovery of single-molecule-magnet (SMM) properties for the so-called Mn12 cluster1 was the starting point to an intense search for new metal complexes with exchange-coupled paramagnetic centers.2 These investigations were mainly focused on polynuclear complexes based on manganese and iron.3 The first cobalt(II) SMM was reported in 2002,4 which subsequently led to the discovery of several new cobalt-based oligounuclear complexes with slow magnetic relaxation behavior.5 The main focus of these investigations toward high-nuclearity clusters was placed on increasing the total spin of the molecular ground state because the presence of SMM properties is associated with an energy barrier for the reversal of the molecular magnetic moment, which for integer spins is given as Ueff = |D|S2 (for half-integer spins, Ueff = |D|(S2 − 1/4).6 However, this approach appeared to be limited because ever larger spin ground states did not result in higher barriers.7 Additional impetus to change this strategy came from fundamental theoretical considerations by Waldmann8 and later on by Neese and Pantazis,9 which could show that, in fact, the magnetic anisotropy is the parameter that needs to be maximized for obtaining larger relaxation barriers. © XXXX American Chemical Society

Received: February 13, 2016

A

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry geometries33,34 as well as octahedral35 and trigonal-prismatic coordination.31,36 Moreover, it was only recently that cobalt(II) SIMs with uncommon three-coordinate trigonal-planar37 and eight-coordinate square-antiprismatic38 geometries have been reported. Interestingly, among the complexes with the highest spin-reversal barriers, both pseudotetrahedral (Ueff = 118 cm−1)26 and trigonal-prismatic geometries (Ueff = 102.8 cm−1)31 have been observed, where the latter examples also contain a Schiff-base ligand. In 2012, we reported a distorted pseudotetrahedral cobalt(II) complex with two bidentate 2-(1H-imidazol-2-yl)phenol-based ligands exhibiting slow magnetic relaxation behavior with an energy barrier of Ueff = 62 cm−1.19 This was the first example of a cobalt(II) complex with solely an N,O-donor environment showing SIM behavior. This prompted us to perform a search in the Cambridge Structural Database for four-coordinate metal complexes with chelating ligands containing an N,O-donor set. In fact, Schiff-base ligands derived from 2-phenylaniline and 2hydroxy-1-naphthaldehyde are found to be sterically demanding and able to enforce distorted geometric arrangements around metal ions, in particular such as distorted tetrahedral geometries.39 This inspired us to investigate the magnetic properties of this ligand system with and without a sterically demanding phenyl group in the ortho position of the aniline moiety. Here we report the synthesis, crystal structures, and magnetic properties of two new relevant cobalt(II) complexes. This is complemented by sophisticated theoretical calculations with the ab initio CASSCF/CASPT2/RASSI-SO protocol to obtain a detailed picture of the low-lying states of the complexes.



1-[N-(2-Phenylphenyl)carboximidoyl]naphthalen-2-ol (HLPh). Compounds: 2-hydroxy-1-naphthaldehyde (1.02 g, 5.9 mmol) and 2aminobiphenyl (1.00 g, 5.9 mmol). Yield: 85% (1.63 g, 5.0 mmol). 1H NMR (CDCl3): δ 14.86 (s, 1H, NH), 9.36 (s, 1H, NCHAr), 8.06 (d, J = 8.5 Hz, 1H, arom. H), 7.77 (d, J = 9.2 Hz, 1H, arom. H), 7.72 (d, J = 7.9 Hz, 1H, arom. H), 7.36−7.55 (m, 11H, arom. H), 7.07 (d, J = 9.2 Hz, 1H, arom. H). 13C NMR (CDCl3): δ 168.4 (CO), 156.0 (NCHAr), 136.3, 131.1, 129.6, 129.4, 128.8, 128.5, 128.1, 127.7, 127.4, 126.7, 123.5, 121.7, 119.0, 118.8, 109.1. IR (solid, ATR, cm−1): ν̃ 3058 (vw), 3029 (vw), 1610 (s), 1585 (s), 1538 (s), 1479 (m), 1314 (s), 1298 (s), 1210 (m), 1166 (s), 1153 (s), 1140 (s), 1109 (w), 968 (m), 859 (m), 829 (s), 769 (s), 749 (vs), 731 (s), 714 (vs), 454 (s), 408 (m). Synthesis of Complexes. Co(CH3COO)2·4H2O and the corresponding ligand were dissolved in a 1:1 mixture of dichloromethane and methanol (20 mL). After 5 min of stirring, the solution was filtered off and left undisturbed for crystallization. Slow evaporation of the solvent yields the crystalline complex. [Co(LBr)2] (1). Compounds: HLBr (0.20 g, 0.56 mmol) and Co(CH3COO)2·4H2O (0.08 g, 0.32 mmol). Yield: 61% (0.12 g, 0.17 mmol). Anal. Calcd for C34H22Br2CoN2O2 (M = 709.29 g mol−1): C, 57.57; H, 3.13; N, 3.95; Br, 22.53. Found: C, 57.50; H, 2.85; N, 4.43; Br, 22.64. MS (DESI): m/z 709 (60%, [Co(C17H11BrNO)2]•+), 383 (29%, [Co(C17H11BrNO)]•+), 326 (100%, [C17H11BrNO]•+), 245 (78%, [C17H11NO]+). IR (solid, ATR, cm−1): ν̃ 3046 (vw), 1614 (m), 1597 (m), 1568 (m), 1533(s), 1482 (s), 1454 (s), 1424 (s), 1381 (vs), 1342 (s), 1298 (s), 1251 (m), 1216 (m), 1181 (s), 1164 (s), 1136 (s), 1072 (s), 1042 (w), 1005 (s), 978 (m), 945 (w), 891 (w), 831 (vs), 814 (vs), 745 (vs), 710 (m), 666 (m), 581 (s), 496 (vs), 448 (vs), 423 (m). [Co(LPh)2]·CH2Cl2 (2·CH2Cl2). Compounds: HLPh (0.10 g, 041 mmol) and Co(CH3COO)2·4H2O (0.05 g, 0.20 mmol). Yield: 45% (0.06 g, 0.09 mmol). Anal. Calcd for C47H34Cl2CoN2O2 (M = 788.62 g mol−1): C, 71.58; H, 4.35; N, 3.55. Found: C, 71.74; H, 4.32; N, 3.69. MS (DESI): m/z 703 (12%, [CoC46H32N2O2]•+), 381 (100%, [CoC 23 H 16 NO] + ), 323 (26%, [C 23 H 17 NO] •+ ), 180 (60%, [C13H10N]+), 152 (24%, [C12H8]+), 84 (73%). IR (solid, ATR, cm−1): ν̃ 3053 (vw), 1615 (m), 1600 (m), 1569 (s), 1531 (vs), 1474 (m), 1456 (m), 1424 (s), 1384 (vs), 1364 (vs), 1304 (m), 1250 (w), 1178 (s), 1163 (s), 1141 (m), 1116 (w), 1092 (w), 1041 (w), 980 (w), 861 (vw), 827 (s), 775 (m), 753 (vs), 740 (vs), 730 (vs), 696 (vs), 651 (m), 614 (w), 556 (s), 541 (m), 518 (w), 499 (s), 480 (m), 451 (s), 420 (m). Magnetic Measurements. Magnetic measurements were performed on a Quantum Design MPMS-5 SQUID magnetometer. Susceptibility data were obtained in the temperature range from 2 to 300 K with static fields up to 5 T. The polycrystalline samples were molten with paraffin to prevent crystallites from orienting. The collected data were corrected for the diamagnetism of the sample holder, the capsula, the paraffin, and the diamagnetic contribution of the ligand. The dynamic susceptibility was measured with alternating fields of 1 Oe magnitude and a constant direct-current (dc) field of 0, 400, or 1000 Oe in the frequency range from 10 to 1500 Hz at temperatures from 2 to 7 K. The obtained data were fitted for each temperature to eq 1 using the program OriginPro 8.5. Simulations of susceptibility and magnetization data were performed by employing full-matrix diagonalization, as implemented in the MagProp analysis program of the DAVE package.41 This includes an iterative approach to address mean-field theory (MFT) by the molecular-field parameter λ, which is related to the interexchange coupling between the molecular spins.

EXPERIMENTAL DETAILS

Materials. All starting materials were available from commercial suppliers and were used without further purification. The reactions and all manipulations of the samples were carried out under aerobic conditions. The Schiff-base ligands have been synthesized according to modified published procedures.39,40 Instrumentation. IR spectra were measured with a Bruker Equinox spectrometer equipped with a diamond ATR unit in the range of 4000−400 cm−1. Elemental analyses were carried out on Leco CHNS-932 and El Vario III elemental analyzers. Mass spectrometry (MS) spectra were measured with a Bruker MAT SSQ 710 spectrometer. Measurements for thermogravimetric analysis (TGA) were performed with a Netzsch STA409PC Luxx under a constant flow of air in the range from room temperature to 1000 °C at a heating rate of 5 °C min−1. NMR spectra were recorded with a Bruker AVANCE 400 MHz spectrometer. Synthesis of Schiff-Base Ligands. Equimolar amounts of 2hydroxy-1-naphthaldehyde (6 mmol) and the respective aniline derivative (6 mmol) were dissolved in methanol (25 mL) and refluxed for 2 h. After cooling to room temperature, the resulting solid was filtered off and dried under reduced pressure. 1-[N-(4-Bromophenyl)carboximidoyl]naphthalen-2-ol (HLBr). Compounds: 2-hydroxy-1-naphthaldehyde (1.00 g, 5.8 mmol) and 4bromoaniline (1.00 g, 5.8 mmol). Yield: 90% (0.71 g, 5.2 mmol). 1H NMR (CDCl3): δ 15.19 (s, 1H, NH), 9.30 (s, 1H, NCHAr), 8.07 (d, J = 8.4 Hz, 1H, arom. H), 7.81 (s, 1H, arom. H), 7.71 (d, J = 7.9 Hz, 1H, arom. H), 7.49−7.55 (m, 3H, arom. H), 7.32 (dd, J = 7.4 Hz, 1H, arom. H), 7.20 (d, J = 8.4 Hz, 2H, arom. H), 7.09−7.13 (m, 1H, arom. H). 13C NMR (CDCl3): δ 155.7 (CO), 145.1 (NCHAr), 136.6, 133.1, 132.7, 129.4, 128.2, 127.6, 123.8, 122.2, 121.7, 119.9, 119.0. IR (solid, ATR, cm−1): ν̃ 3027 (vw), 1603 (s), 1539 (s), 1481 (s), 1385 (m), 1311 (s), 1211 (m), 1180 (s), 1160 (s), 1144 (s), 1072, 1035 (s), 1004 (s), 957 (m), 856 (m), 823 (vs), 776 (vw), 746 (vs), 724 (s), 662 (m), 606 (m), 492 (vs), 473 (m), 435 (m), 414 (w).

χ (ω) = χS +

χ0 − χS 1 + (iωτc)1 − α

(1)

Crystal Structure Determination. Single crystals of 1 and 2 were obtained by slow evaporation of the solvent from the reaction mixture. Suitable crystals were selected from the mother liquor under a polarization microscope and mounted on a glass fiber. Crystallographic data were collected on a Nonius Kappa CCD diffractometer using B

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry



graphite-monochromated Mo Kα radiation (λ = 0.71073 Å). A summary of crystallographic and structure refinement data is given in Table 1. The obtained data were corrected for Lorentz and

RESULTS AND DISCUSSION Syntheses. The ligands HLBr and HLPh depicted in Scheme 1 are readily obtained by Schiff-base condensation between 2-

Table 1. Crystallographic Data and Structure Refinement Parameters for 1 and 2 formula fw (g mol−1) cryst syst space group a (pm) b (pm) c (pm) α (deg) β (deg) γ (deg) V (106 pm3) Z T (K) δcalc (g cm−3) F(000) μ(Mo Kα) (mm−1) θ range of data collection (deg) measd reflns unique reflns (Rint) no. of param GOF on F2 Flack parameter R1 [I > 2σ(I)] wR2 (all data)

1

2

C34H22Br2CoN2O2 709.29 orthorhombic P21212 955.54(3) 2546.50(9) 568.61(1) 90 90 90 1383.59(7) 2 133(2) 1.703 706 3.544 2.67−27.46 7949 3059 (0.0271) 186 1.119 0.015(7) 0.0246 0.0534

C47H34Cl2CoN2O2 788.59 triclinic P1̅ 937.95(2) 1033.51(2) 2030.76(4) 100.131(1) 92.952(1) 107.721(1) 1834.31(6) 2 133(2) 1.428 814 0.658 2.15−27.42 14002 8164 (0.0171) 623 1.062

Article

Scheme 1. Tautomeric Equilibrium for the Ligands HLBr and HLPh

hydroxy-1-naphthaldeyde and the desired aniline derivative.39,40 It is interesting to note here that the solution behavior of such Schiff bases is generally characterized by a tautomeric equilibrium between their ketoenamine (NH) and enolimine (OH) form (see Scheme 1), where the position of the tautomeric equilibrium (KT = [NH]/[OH]) strongly depends on the actual substituents present at the aromatic rings.50 However, for the ligands HLBr and HLPh, only the ketoenamine tautomer could be detected by NMR spectroscopy. The reaction of cobalt(II) acetate tetrahydrate with 2 equiv of the corresponding ligands HLBr and HLPh in a mixture of dichloromethane and methanol (1:1) at room temperature leads to the formation of neutral complexes 1 and 2, respectively (Scheme 2). The isolated crystalline material for

0.0334 0.0810

Scheme 2 polarization effects but not for absorption effects. The structures were solved by direct methods (SHELXS-97) and refined by fullmatrix least-squares techniques against F2 (SHELXL-97).42 All nonhydrogen atoms were refined using anisotropic thermal parameters. Hydrogen atoms for 1 were treated as riding atoms, whereas for 2, hydrogen atoms were refined with isotropic thermal parameters. CCDC 1424465 and 1424466 contain the supplementary crystallographic data for the structures 1 and 2, respectively. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre (http://www.ccdc.cam.ac.uk). Computational Details. Ab initio calculations were performed with the Molcas 7.8 program package.43 To obtain suitable fragments, the second ring of the naphthyl moiety was replaced by two hydrogen atoms and the substituent at the aniline moiety was replaced by a hydrogen atom. The hydrogen atoms of the fragments were optimized (RI-BP86 and def2-SVP) utilizing the program package Turbomole 6.444 prior to the multiconfigurational calculations. For the Molcas calculations, relativistic ANO-RCC basis sets for all atoms were used (contractions: Co 6s5p4d2f1g, O and N 4s3p2d1f, C 3s2p, H 2s). Scalar relativistic effects were included by using the Douglas−Kroll− Hess Hamiltonian.45 The Choleski decomposition of two-electron integrals was employed to save computational time and disk space (threshold: 10−6). The CASSCF method with an active space of 7 electrons in 10 orbitals (all 3d and 4d orbitals of the cobalt atom) was used to calculate all states arising from a d7 configuration (10 quartets and 40 doublets). Dynamic correlation was treated with the CASPT2 method46 in its multistate formulation for all quartet states and the 12 lowest doublet states, with the latter limit set to avoid intruder states. The RASSI-SO method was used to obtain the energies of the spin− orbit states.47 The magnetic properties were calculated with the Single_Aniso module.48,49

the latter complex 2 was found to contain one cocrystallized molecule of dichloromethane per complex unit, which was evident from elemental analysis, leading to the formula [Co(LPh)2]·CH2Cl2. This was confirmed by TGA because complex 2 undergoes a weight loss of about 11.3% in the temperature range between 150 and 210 °C [see Figure S1 in the Supporting Information (SI)], which is in good agreement with the calculated value corresponding to the loss of one cocrystallized dichloromethane molecule per formula unit (10.8%). Considering the low boiling point of the cocrystallized dichloromethane, this indicates a remarkable thermal stability of the dichloromethane aggregate in 2.51 This is consistent with strong hydrogen bonding in the crystal structure of 2 (vide infra). In contrast, complex 1 is stable up to 300 °C, indicating that there are no solvent molecules present in the crystalline material (Figure S2 in the SI). Crystal Structures. X-ray diffraction experiments on suitable crystals revealed that compounds 1 and 2 crystallize in the orthorhombic space group P21212 and the triclinic space C

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry group P1̅, respectively. For both compounds, a mononuclear constitution of the neutral complexes with the cobalt(II) ion coordinated by two singly deprotonated ligand molecules is observed. The molecular structures of 1 and 2 are depicted in Figures 1 and 2, respectively. Selected bond lengths and angles are summarized in Table 2.

Both complexes exhibit a distorted pseudotetrahedral [N2O2] coordination polyhedron around the cobalt(II) ion. For complex 1, the two coordinating ligand molecules are crystallographically equivalent and related by a C2 axis, while in 2, the cobalt(II) center is not located on a crystallographic symmetry element, leaving its coordinating ligands crystallographically independent. Despite these differences, the direct coordination environments of the cobalt(II) centers in 1 and 2 are very similar as far as the Co−N/O donor bond lengths and bite angles of the chelate ligands are concerned, with deviations of less than 1 pm and 1°, respectively. The Co−N/O donor bond lengths for 1 and 2 are within the typically observed range for tetrahedral cobalt(II) complexes in the high-spin state (cf. Figure S3 in the SI). Nevertheless, both complexes exhibit considerable distortion from the pseudotetrahedral geometry, as indicated by the dihedral angle δ between the coordination planes of the two chelate rings at the cobalt centers (1, 86°; 2, 79°). Moreover, also the other bond angles suggest that the coordination environment in 1 is closer to resembling a pseudotetrahedral geometry. These geometrical differences are illustrated by the overlay of both complex structures depicted in Figure 3.

Figure 1. Molecular structure of 1. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atoms are omitted for clarity.

Figure 3. Overlay of the molecular structures of 1 (blue) and 2 (red). Figure 2. Molecular structure of 2. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atoms not involved in hydrogen bonding are omitted for clarity. Red dotted lines represent hydrogen bonds with pertinent distances: C···O1, 354.7 pm; C···O2, 299.7 pm.

The geometric distortion of the coordination environment of complexes 1 and 2 was further investigated utilizing continuous shape measures (CShM), as proposed by Avnir et al.,52 which describe a quantitative measure for the distortion of a given coordination sphere from tetrahedral and square-planar geometries by corresponding shape measures, S(Td) and S(D4h), respectively. These values describe the deviation from ideal geometries, starting with zero for a perfect match. The CShM data for both complexes are given in Table 3 and indicate significant distortion from the tetrahedral geometry, which is considerably larger in the case of 2. This is consistent with a larger value for 2 as far as the angular fraction along the minimal distortion pathway between the two considered ideal polyhedra is concerned. Consequently, the observed distortion of 1 and 2 is best described as a rotation of the coordination planes defined by the chelate ligands about the corresponding C2 axis, with a minor contribution from elongation of the tetrahedral coordination polyhedra, as indicated by the path deviation function (Δ) and further evidenced by the acute bite angles of the chelate ligands. The bond lengths observed for the coordinating ligand fragments in both complexes indicate their presence in the deprotonated β-ketiminato form, as depicted in Scheme 2 (see

Table 2. Selected Bond Lengths (pm) and Angles (deg) for 1 and 2a 1 Co1−N1 Co1−N2 Co1−O1 Co1−O2 O1−Co1−N1 O1−Co1−N1A/N2 O2−Co1−N2 O2−Co1−N1 N1−Co1−N1A/N2 O1−Co1−O1A/O2

198.0(2) 190.8(3)

94.36(10) 120.18(10)

112.90(15) 116.85(13)

2 198.81(13) 198.25(13) 190.84(11) 191.70(11) 93.51(5) 131.88(5) 93.55(5) 117.98(5) 115.26(5) 106.36(5)

a

Atoms with the suffix A were generated by the symmetry operation −x, −y − 1, z.

D

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Table 3. Calculated CShM Parameters for 1 and 2 S(Td) S(D4h) φ(Td→D4h)a Δ(Td,D4h)a

1

2

1.82 25.9 13.3 9

2.74 22.2 20.3 7

φ(Td→D4h) is the angular fraction along the minimal distortion pathway from tetrahedral (φ = 0) to square-planar coordination (φ = 100). Values in parentheses represent the path deviation function Δ(Td,D4h), which adopts a zero value for structures that are exactly along the path and 100 for structures which are at the same distance from the interconversion path than the two ideal polyhedra are from each other.53 a

Table S1 in the SI). In fact, the C−N bond lengths (about 131 pm) are somewhat elongated compared to the CN distances typically observed for Schiff bases (127.9 pm), whereas the C− O bonds (about 130 pm) are significantly shortened with respect to comparable C−O bonds of aryl alcohols (136.2 pm).54 This suggests a delocalization of the negative charge over both donor atoms of the chelate ring, as is generally observed for other comparable complex systems.55 The dichloromethane molecule found in the crystal structure of 2 forms hydrogen bonds with the phenolate oxygen atoms of the two chelate ligands. In addition to these hydrogen bonds, intramolecular lone pair (lp)−π 56 (O1) and CH/π 57 interactions are present within the cobalt(II) complexes (see Figure S4 in the SI), the former of which also accounts for the differences in strength between the two hydrogen bonds. The complex molecules are further associated by π−π and CH/π interactions to form chains along the crystallographic b axis (see Figure S5 in the SI). The dichloromethane molecules, located in the interspace between these one-dimensional loose molecular aggregates, show additional weak lp−π interactions of the chlorine atoms with neighboring complex molecules (see Figure S6 in the SI). This leads to a firm association of the dichloromethane molecules in the crystal packing of 2, which can explain the remarkably high thermal stability of the solvent aggregate. The overall crystal packing of 2 leads to wellseparated cobalt(II) ions with intermolecular Co−Co distances larger than 860 pm. On the other hand, in the crystal structure of 1, a rather short Co−Co distance of about 569 pm is observed. This is due to stacking of the complex molecules along the crystallographic c axis, which is governed by strong lp (O)−π interactions56 between the coordinating phenolate oxygen atoms and bromophenyl rings of neighboring complex molecules at an O···ring centroid distance of 334 pm and a decline angle (O−centroid−plane) of 78° (see Figure S7 in the SI). Additional strong lp (Br)−π interactions are found between the bromine atom at the para position of the phenyl ring and the outer ring of the naphthyl group of a complex molecule that belongs to a neighboring stack (Br···centroid:, 348 pm; decline angle, 73°; cf. Figure S8 in the SI). Magnetic Properties. All magnetic measurements were performed on powdered samples prepared from the crystalline compounds 1 and 2, which were molten with paraffin to prevent the crystallites from orienting. Magnetic susceptibility data were collected between 2 and 300 K with an applied magnetic field of 0.1 T, and the temperature dependences of the molar susceptibility χM and its product χMT are depicted in Figure 4.

Figure 4. Thermal dependences of χM (squares) and χMT (circles) for 1 (top) and 2 (bottom). Lines represent simulated values for best-fit parameters (see the text and Table 4).

For both compounds 1 and 2, the χMT values at room temperature are virtually the same at about 2.38 cm3 K mol−1. These values largely exceed the spin-only value of an S = 3/2 ion (1.87 cm3 K mol−1) and suggest a g factor of about 2.25, indicating the presence of strong spin−orbit coupling, as can be expected for tetrahedrally coordinated cobalt(II) ions. Upon cooling, χMT stays approximately constant down to about 50 K. For both complexes, a sharp decrease in χMT is observed for temperatures below 50 K, with final values at 2 K of 1.77 and 0.99 cm3 K mol−1 for 2 and 1, respectively. This decrease can basically be attributed to magnetic anisotropy of the cobalt(II) ion and/or intermolecular exchange interactions. Magnetization plots for 1 and 2 depicted in Figure 5 indicate that saturation is not reached for both compounds at a field of 5 T, with a magnetization of M ≈ 2.0 measured at 2 K. Simulations of the susceptibility and magnetization data were performed by employing full-matrix diagonalization, as implemented in the program package DAVE.41 To avoid overparametrization, only Zeeman interaction and the axial zero-field splitting (ZFS) were included in the spin Hamiltonian given in eq 2. ⎤ ⎡ 2 1 Ĥ = gμB BS ̂ + D⎢Sẑ − S(S + 1)⎥ ⎦ ⎣ 3

(2)

In addition, correction terms for temperature-independent magnetic contributions (χTIP) and intermolecular exchange interactions via MFT with the molecular field parameter λ were included as appropriate. The best-fit parameters obtained for both χMT and M data of 1 and 2 used for the simulations depicted in Figures 4 and 5 are listed in Table 4. For comparison, full lists of parameter sets obtained from leastE

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

mol is a signature for a small antiferromagnetic intermolecular interaction, which, nevertheless, is important to accurately describing the experimental data. The overall best fits for the χMT data reveal very similar |D| parameters for both complexes 1 and 2, with a value of about 27 cm−1. At this point, it should be noted that determination of the axial ZFS parameter D from magnetic susceptibility data derived from measurements on polycrystalline samples is generally difficult.58 This not only applies to its absolute value but even more so to the sign of D, which cannot unambiguously be determined on this basis, because variations of χMT are, in general, not very sensitive in this respect. This particularly holds for S = 3/2 systems.59 The more appropriate approach, because it is the more sensitive probe, is the determination of D values from magnetization measurements at intermediate fields and appropriate low temperatures, which is based on the deviation from the anticipated Brillouin behavior. We therefore performed variable-field/variable-temperature magnetization measurements for complexes 1 and 2 at fields ranging from 0 to 5 T and temperatures between 2 and 5 K (see Figure 5). The data show an increase of the magnetization with the applied field, which becomes almost linear for higher fields but does not reach saturation. The observed absence of saturation in combination with the spreading of the curves in the M versus H/T plots (Figure S13 in the SI) indicates the presence of strong ZFS effects in both complexes. Least-squares fits of the experimental magnetization data to eq 2 resulted in good agreement, assuming an isotropic g tensor for 1 and 2 in combination with easy-axis anisotropy (see Table 4). This leads to D values of −36.7 and −39.8 cm−1 for 1 and 2, respectively. The larger D value observed for 2 is consistent with the fact that this complex also shows a larger distortion from the tetrahedral geometry because this enables a more efficient mixing of the ground state with excited states. Attempts to include g anisotropy did not result in a significant improvement of the fit for both complexes. However, for complex 1, the inclusion of intermolecular interactions proved to be essential to sufficiently describing the experimental magnetization data, as was already the case for simulation of the χMT data. The obtained value of λ = −0.538 cm−3 mol is in agreement with the antiferromagnetic intermolecular exchange already indicated by the χMT fits. Not unexpectedly, the obtained absolute values derived from magnetization data deviate from those derived from the susceptibility data. However, because the weighting in the magnetization-based fit is clearly more pronounced on the relevant low-temperature and high-field data, this fit provides the more reliable values for the D parameters. The dynamic behavior of 1 and 2 was investigated by alternating-current (ac) susceptibility measurements at various temperatures using an oscillating magnetic field of 1 Oe in magnitude and frequencies between 10 and 1143 Hz. The inphase (χM′) and out-of-phase (χM″) signals of the susceptibility were measured in the temperature range from 2 to 7 K in steps of 0.1 K. In the absence of a constant external field (dc magnetic field), no out-of-phase susceptibility (χM″) could be observed for 1 and 2. However, with an applied dc magnetic field, distinct maxima for χM″ were detected for both complexes, which is evident from the data obtained for a dc field of 1000 Oe depicted in Figure 6. For comparison, the data collected at 400 Oe are given in Figure S14 in the SI. The maxima for χM″ versus T generally shift to higher temperatures with increasing frequency of the ac field. For 1, the maximum is shifted from 3.0 K for 10 Hz to 4.2 K for 1143 Hz, whereas for

Figure 5. Variable-field magnetization data at different temperatures for 1 (top) and 2 (bottom). Lines represent simulated values for the best-fit parameters (see the text and Table 4).

Table 4. Parameter Sets from Least-Squares Fits on χMT and Magnetization Data for 1 and 2 1 g∥ g⊥ |D| (cm−1) λ (cm−3 mol) χTIP (cm3 mol−1) g D (cm−1) λ (cm−3 mol)

Fit on χMT 2.61 1.86 27.0 −0.187 6.7 × 10−4 Fit on M 2.55 −36.7 −0.538

2 2.48 2.09 26.4 n/a 9.3 × 10−5 2.67 −39.8 n/a

squares fitting of χMT and M data using different combinations of included effects such as anisotropy of the g value as well as molecular interactions based on MFT are summarized in Tables S2 and S3 in the SI, and the corresponding graphical representations are given in Figures S9−S12 in the SI. The comparison of the individual quality and consistency of different parameter sets clearly indicates that inclusion of the anisotropy for the g values is essential for simulation of the experimental χMT data for both compounds, whereas there is a significant difference when it comes to inclusion of the intermolecular interactions. In fact, for compound 1, only inclusion of both the g anisotropy and MFT approach leads to a meaningful parameter set for simulation of the χMT data. This is consistent with the presence of short intermolecular Co−Co distances in the crystal structure of 1, which are absent in the case of 2 (vide supra). The obtained λ value of −0.187 cm−3 F

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 6. Temperature dependence of the out-of-phase ac signal χM″ for complexes 1 (top) and 2 (bottom) at different frequencies with an applied dc field of 1000 Oe. Lines are solely guides for the eyes.

Figure 7. Cole−Cole plots for complexes 1 (top) and 2 (bottom) at different temperatures in steps of 0.1 K with an applied dc field of 1000 Oe. Solid lines represent the best fit.

2, a shift from 3.5 K for 10 Hz to 5.1 K for 1143 Hz is observed. This behavior is indicative of thermally controlled relaxation processes. Fitting the ac susceptibility data of complexes 1 and 2 for each temperature to eq 1 leads to parameter sets that include the isothermal (χ0: for ω → 0) and adiabatic susceptibility (χS: for ω → ∞) as well as relaxation time τc and the α parameter (see Tables S4 and S5 in the SI), which describes the distribution of the relaxation times (α = 0, single relaxation time; α = 1, distribution of infinite width).60 The ac data together with the derived parameter sets are best represented by Cole−Cole plots, which are depicted in Figure 7 for the data obtained at an applied dc field of 1000 Oe. The simulations with the derived parameter sets are in good agreement with the experimental data, which also holds for the lower dc field of 400 Oe (Figure S15 in the SI). For a classical thermally activated relaxation process, which for systems with easy-axis anisotropy like 1 and 2 takes place by 3 excitation from the ground-state Kramers doublet (KD) ± 2

Figure 8. Arrhenius plots of relaxation times as ln τc versus 1/T for complexes 1 and 2 with applied dc fields of 400 and 1000 Oe. Lines represent linear fits for the high-temperature data.

field, leading to a perfect overlay of the data obtained for both fields (400 and 1000 Oe) for both complexes. However, in the low-temperature range (T < 4 K), for complex 2, a pronounced deviation in the ln(τc) versus 1/T behavior is observed when different dc fields (400 and 1000 Oe) are applied. It should be noted here that for complex 1 the χ(ω) fits at low temperatures were associated with large errors in τc and, hence, the related data points were omitted in the ln(τc) versus 1/T plot. Likewise, this holds in the case of complex 2 for data sets obtained at the higher dc field of 1000 Oe and temperatures below 3 K. Least-squares fits of the temperature-dependent data of τc using the Arrhenius law according to eq 3 led to energy barriers

1

to the excited state ± 2 , an Arrhenius-type behavior should be observed for the temperature dependence of the relaxations times τc.61 Consequently, the activation energy barrier Ueff for this process can be determined from the Arrhenius law given in eq 3. τc = τ0 exp(Ueff /kBT )

(3)

In fact, the ln(τc) versus 1/T plots for 1 and 2 depicted in Figure 8 show a virtually linear behavior for the hightemperature range. This is independent from the applied dc G

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Table 5. Comparison of SMM and Structural Parameters for Pseudotetrahedral Cobalt(II) Complexes entry

donor set

D (cm−1)

Hdc (Oe)

Ueff (cm−1)

τ0 (s) −10

S(Td)

φ(Td→D4h)a

bond angles (deg)b

ref

1 2 3 4

N2O2 N2O2 N2O2 N2O2

−36.7 −39.8 −41

400 400 1000 2000

36 43 62 39.4

5.6 8.4 1.0 1.3

× × × ×

10 10−10 10−10 10−8

1.82 2.74 2.95c 3.41

13.3(9) 20.3(7) 28.0(4)c 30.2(5)

94.4/94.4 93.5/93.6 94.2/94.4c 93.4/94.6

this work this work 19 20

5 6 7 8 9

N2Br2 N2Cl2 N2Br2 N2I2 N2(NCS)2

10.6 10.5 12.5 10.3 −10.1

1000 2000 2000 2000 2000

22.9 29.6 27.5 39.6 14.7

3.7 1.9 1.2 3.2 1.86

× × × × ×

10−10 10−10 10−10 10−13 10−8

2.89 2.73 3.14 3.92 0.26

27.7(22) 26.9(17) 28.9(21) 32.4(26) 8.4(4)

84.5/115.6 81.7/117.4 81.9/117.9 81.9/118.3 109.6/115.7d

21 22 22 22 23

10

N3Cl

12.7

1500

24

1.9 × 10−10

2.71

26.9(24)

95.1/118.8e

24

11 12

N4 N4

1500 0

75 118

10.63 6.29

54.0(37) 41.2(23)

71.5/71.5 80.6/80.7

25 26

13 14 15

O4 S4 Se4

−11.1 −62 −83

1400 0 0

21 21 19

7.0 × 10−10 1.0 × 10−6 3.0 × 10−6

0.51 1.10 1.47

11.6(4) 17.1(6) 19.7(8)

104.4/105.9 97.4/98.3 94.3/97.7

27 27 27

16 17 18 19 20 21

P2Cl2 P2Cl2 P2Cl2 P2Br2 P2I2 As2I2

−14 −14.4f −15.4f −12.5 −36.9 −74.7

1000 1000 1000 1000 1000 1000

25.8 24.3 20.8 25.9 21.3 22.7

10−9 10−10 10−9 10−11 10−10 10−8

0.48 0.34 0.35 0.39 0.17 0.23

11.3(2) 9.4(5) 9.6(2) 10.1(1) 6.7(5) 7.7(6)

104.9/104.9 106.6/102.8 103.5/108.0 105.0/105.0 105.4/109.8 104.2/109.7

28 and 63 28 28 29 30 30

−58 −115

2.64 × 10−8 3.89 × 10−8

1.2 2.1 6.0 9.44 4.65 1.5

× × × × × ×

φ(Td→D4h) is the angular fraction along the minimal distortion pathway from tetrahedral (φ = 0) to square-planar coordination (φ = 100). Values in parentheses represent the path deviation function Δ(Td,D4h), which adopts a zero value for structures that are exactly along the path and 100 for structures that are at the same distance from the interconversion path as the two ideal polyhedra are from each other.53 bBite angles for chelate ligands. For all other cases, the smallest angle within the pseudotetrahedron and the second angle given by the remaining donor atoms are summarized. cAverage values are given for two crystallographically independent molecules. dAngles are given in analogy to entries 5−8 as SCN− Co−NCS and N−Co−N. eAverage values are given for the N−Co−N and N−Co−Cl angles because the complex shows a pseudo-C3v symmetry. f Calculated using the CASSCF/CASPT2/RASSI method. a

Ueff of 36 and 43 cm−1 as well as characteristic relaxation times τ0 of 5.6 × 10−10 and 8.4 × 10−10 s for complexes 1 and 2, respectively. These values are within the typically observed range for pseudotetrahedral cobalt(II) SIMs summarized in Table 5. The lack of slow magnetic relaxation in the absence of an applied dc field can be attributed to quantum tunneling of magnetization within individual KDs associated with the two potential wells given by the relative orientation of the spin with respect to the anisotropy axis. Interestingly, for complexes 1 and 2, tunneling effects can be suppressed very efficiently by the application of only moderate dc fields because this seems to effectively lift the degeneracy of the KDs. Assuming solely a thermally activated Orbach-based relaxation mechanism, the energy barriers observed for complexes 1 and 2 (Ueff = 2|D|, for S = 3/2) would correspond to ZFS parameters |D| of about 27 and 32 cm−1, respectively. However, these values are considerably smaller than those determined from the dc magnetization measurements. Nevertheless, at least a qualitative correlation between the Ueff and |D| values is observed for both complexes. In any case, this deviation is not unexpected if the data are compared with related pseudotetrahedral cobalt(II) complexes (Table 5). Reported pseudotetrahedral cobalt(II) complexes showing slow magnetic relaxation are summarized in Table 5 and can be divided into six basic groups according to their donor sets. One of these groups is currently only represented by a single case (N3Cl). By comparison, it becomes obvious that within groups

of cobalt(II) complexes the observed variation of the ZFS parameter D is in most cases only moderate, consistent with the importance of the first coordination sphere on the magnetic properties. However, striking exceptions are found for heavier element donors like the chalcogens sulfur and selenium (entries 14 and 15, Table 5) or the pnictogen arsenic (entry 21, Table 5). The exceptionally large easy-axis magnetic anisotropy of the latter cases might be ascribed to the π-bonding ability of the donor atoms, as suggested by theoretical studies.62 Moreover, it turns out that for most cases easy-axis magnetic anisotropy is found to be independent from the actual donor set, with exceptions observed for cases with geometrically restricted ligands containing N-donor sets, viz., bidentate phenanthroline (entry 5, Table 5) and 2,2-biquinoline (entries 6−8, Table 5) ligands as well as a tripodal ligand (entry 10, Table 5). Interestingly, the energy barriers Ueff are not generally following the trends given by the ZFS parameter D, with the most striking deviation observed for cases with heavier element donors (entries 14, 15, and 21, Table 5), for which very large ZFS only yields moderate values for the energy barriers. The most prominent group among the pseudotetrahedral cobalt(II) complexes is the one with the N4-donor set. For these cases, large easy-axis anisotropy correlates with large energy barriers for the spin reversal (Ueff). To further elucidate the influence of structural features on the magnetic properties, we have included pertinent data derived from CShM analysis and characteristic bond angles H

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

transition-metal complexes and was therefore utilized in this study.48 The calculations were performed on the complex geometries truncated to reduce computational effort (see the Experimental Details). Because the comparison from Table 5 suggested the importance of the first coordination sphere on the magnetic properties, we decided to use the truncation of the computational model to additionally reduce the complexity of the ligand, so as to solely probe geometric effects. The spinfree energies obtained from the CASSCF and CASPT2 calculations are listed in Tables S7 in the SI. The CASSCF energies clearly show a quartet ground state, as expected for tetrahedral cobalt(II). The lowest doublet state is wellseparated from the lowest quartet state with an energy gap of about 17500 and 16500 cm−1 for complexes 1 and 2, respectively. Thus, the doublet states are unlikely to mix strongly into the ground-state KD. Tetrahedrally coordinated cobalt(II) ions have a d 7 configuration and possess a 4A2 ground-state and two excitedstate terms, namely, 4T2 and 4T1, all arising from the 4F free-ion term. A deviation from the ideal tetrahedral symmetry splits the 4 T2 excited-state term and, hence, leads to different mixing of the resulting 4B2 and 4E terms with the 4A2 ground-state term, introducing first-order angular momentum. The CASPT2 calculations for both complexes show a strong splitting of the 4 T2 excited-state term. The first excited quartet state of complex 1 is found at ≈2400 cm−1 above the ground state, whereas the next higher excited states follow at relative energies higher than 6800 cm−1. The situation is similar for 2, where the first excited state lies at ≈2100 cm−1 and further excited states follow at energies above 6300 cm−1. Hence, mixing between the two lowest quartet states is possible and introduces a significant magnetic anisotropy by spin−orbit coupling in both complexes. The energies of the spin−orbit states were obtained from RASSI-SO calculations using the CASPT2 quartet energies (see Table S8 in the SI) and are summarized together with the main values of the g tensor in Table 6. The calculated energy gap

within the pseudotetrahedral arrangement in the comparison summarized in Table 5. For the shape measure S(Td), generally the changes within individual groups are only moderate, with the widest range observed for the N2O2-donor set. Within the latter group, the ZFS parameter D seems to correlate with the structural deviation from the pseudotetrahedral symmetry, which can best be described as twisting of the chelate coordination planes at the cobalt center. This, however, is not the case as far as the energy barrier Ueff is concerned, as is obvious from entry 4 in Table 5, for which, unfortunately, no ZFS parameter has been reported in the literature. The general trends also reflect in the values for φ(Td→D4h), the angular fraction along the minimal distortion pathway from tetrahedral to square-planar coordination. Because this is a measure for a specific geometric distortion, it needs to be additionally characterized by deviation of the actual structure from this desired path. The cases with large deviations are found to contain an elongation component as far as the distortion of the pseudotetrahedral coordination environment is concerned, which is evident from the bond angles. It is interesting to note that the latter cases, which show a large deviation between the two opposite bond angles, exhibit easy-plane magnetic anisotropy. The only other cases with a large deviation from the minimal distortion pathway between tetrahedral to squareplanar coordination are the examples with the N4-donor set. In these cases, a large distortion [S(Td)] is accompanied by the presence of two acute bite angles consistent with a strong elongation of the tetrahedron. Unfortunately, the shape measure concept does not allow for the specific evaluation of the elongation within a tetrahedral system. As a simple approach, we define the elongation parameter ϵT as the ratio of the averaged obtuse and acute angles in an elongated tetrahedron (see Scheme S2 in the SI). These values describe the stretching of a tetrahedron along one of its C2 axes, starting with unity for the undistorted case and increasing values as the elongation increases. For the pseudotetrahedral cobalt(II) complexes from Table 5 with chelate ligands containing N2O2- and N4-donor sets, the relevant data are summarized in Table S6 in the SI. In all other cases from Table 5, this approach is not feasible because it requires two similar bonding situations along opposite edges of the pseudotetrahedron defining two acute angles, which is best realized by two bidentate chelate ligands. For the cases with monoanionic chelate ligands (entries 1−3 and 11 from Table 5), this parameter gives a surprisingly good match with the ZFS behavior, as the projection of the D value on the elongation parameter ϵT indicates, which leads to values of about 32 cm−1. This clearly shows that for these cases the ZFS at least qualitatively correlates with ϵT, indicating that an increase in elongation leads to stronger ZFS. Not unexpected, for the case with a dianionic chelate ligand (entry 12 from Table 5), the corresponding projected value of 74 cm−1 does not match with the previous series. However, this can certainly be attributed to the additional negative charge at the chelate ligand present in the latter case. Interestingly, the inclusion of a formal charge in the projection approach reduces this value to 37 cm−1, which, in turn, is rather close the former average observed for the singly charged chelate ligands. Ab Initio Calculations. For deeper insight into the slow magnetic relaxation and related differences between 1 and 2, ab initio calculations were performed. Recently, the CASSCF/ CASPT2/RASSI-SO protocol implemented in Molcas 7.8 proved to be well-suited to treat the magnetic properties of

Table 6. Main Values of the g Tensor of the Two Lowest KD and Their Relative Energies EKD (cm−1) for 1 and 2 Calculated with the CASPT2/RASSI-SO Procedure KD 1

2

EKD g1 g2 g3 EKD g1 g2 g3

1

2

0 0.0248 0.0284 7.4644 54 2.5255 4.2626 4.3110

0 0.3398 0.3658 7.5596 61 2.5511 3.9620 4.5655

between the ground-state and first excited-state KD is slightly larger for complex 2 (61 cm−1) than for 1 (54 cm−1). However, both results overestimate the experimental Ueff values. The obtained eigenvalues of the ground-state g tensor represent easy-axis anisotropy for both complexes, which is almost ideal for complex 1. Nevertheless, for 2, the g1 and g2 elements of 0.34 and 0.37, respectively, indicate the presence of a small additional rhombic component. With the pseudospin formalism implemented in Single_Aniso, also the ZFS parameters and g values for the effective spin 3/2 have been calculated (see Table S9 in the SI), which are in agreement with easy-axis anisotropy for both complexes. For 2 a D value of −31 cm−1 and a I

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry transversal component E of 1.7 cm−1 are obtained, which indicates a slight rhombic distortion. On the other hand, 1 shows a somewhat smaller D value of −27 cm−1 with almost zero rhombic ZFS. These theoretical values are in good agreement with the experimental findings mentioned above, which also includes the absence of any indication of a transversal component E from the fits of the experimental magnetization data. The orientation of the magnetization easy axis within the molecular framework is depicted in Figure 9 and shows a large

phenolate oxygen atoms toward an associated dichloromethane molecule. The distorted N2O2 coordination environment leads to a considerable spin−orbit coupling and facilitates the observation of slow magnetic relaxation in the presence of small external dc fields. Overall, N,O-based donor sets (N2O2 and N4) are favorable to observe slow magnetic relaxation with substantial energy barriers Ueff in distorted pseudotetrahedral cobalt(II) complexes. This behavior is governed by distortion of the pseudotetrahedral donor arrangement. The major contribution for this group of complexes can be assigned to the degree of elongation induced by the chelate ligands, whereas the differences within the examples containing a N2O2donor set seem to be qualitatively related to a twisting distortion of the chelate coordination planes at the cobalt center. Ab initio calculations on complexes 1 and 2 confirm the observed experimental parameters and support the overall interpretation of their magnetic behavior. Future efforts will have to focus on the investigation of distorted pseudotetrahedral cobalt(II) complexes in order to extend the understanding of structural and possibly also electronic influences on the observed SIM properties.



ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.6b00373. Comparison of the structural data and crystal packing diagrams, further data related to TGA, magnetic characterization, and ab initio calculations (PDF) Additional structural data in CIF format (CIF) Additional structural data in CIF format (CIF)

Figure 9. Orientation of the main axis of the g tensor for the groundstate KD according to the CASPT2/RASSI-SO calculations within the molecular framework of complexes 1 (left) and 2 (right).



similarity between both complexes. Interestingly, these vectors are aligned along the connection of the midpoints of the pseudotetrahedral edges determined by the coordination pockets of the two bidentate Schiff-base ligands, which can be regarded as the axis of slight elongation within the polyhedra (vide supra). This indicates that the geometric restraint given by the charged chelate ligands around the cobalt(II) center is important to determine the magnetic properties. To further compare the picture provided by the calculated values with the experimental data, the static magnetic properties were simulated on the basis of the calculated parameters by employing the Molcas module Single_Aniso (see Figures S16 and S17 in the SI). As expected, for complex 1, inclusion of an intermolecular interaction parameter was necessary to describe the temperature dependence of χMT as well as the magnetization versus field data correctly. It has to be noted here that the calculated and experimental data are in excellent agreement, which confirms that the calculations provide valuable insight.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +49 3641 948130. Fax: +49 3641 948132. Author Contributions †

Contributed equally to this work.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We are grateful to Michael Böhme and Dr. Axel Buchholz for valuable discussions. We thank Florian Reinhardt for measurement of the magnetic data.





REFERENCES

(1) Sessoli, R.; Gatteschi, D.; Caneschi, A.; Novak, M. A. Nature 1993, 365, 141−143. (2) (a) Christou, G.; Gatteschi, D.; Hendrickson, D. N.; Sessoli, R. MRS Bull. 2000, 25, 66−71. (b) Pedersen, K. S.; Bendix, J.; Clérac, R. Chem. Commun. 2014, 50, 4396−4415. (3) (a) Aromı ́, G.; Aubin, S. M. J.; Bolcar, M. A.; Christou, G.; Eppley, H. J.; Folting, K.; Hendrickson, D. N.; Huffman, J. C.; Squire, R. C.; Tsai, H.-L.; Wang, S.; Wemple, M. W. Polyhedron 1998, 17, 3005−3020. (b) Yang, C.-I.; Zhang, Z.-Z.; Lin, S.-B. Coord. Chem. Rev. 2015, 289−290, 289−314. (c) Gatteschi, D.; Sessoli, R.; Cornia, A. Chem. Commun. 2000, 725−732. (d) Glaser, T. Chem. Commun. 2011, 47, 116−130. (e) Aromí, G.; Brechin, E. K. Struct. Bonding (Berlin) 2006, 122, 1−69.

CONCLUSION Two cobalt(II) complexes with bulky Schiff-base ligands have been synthesized. Both complexes exhibit a distorted pseudotetrahedral coordination environment at the cobalt(II) centers. In the case of 1, this distortion is affected by lp−π interactions involving the coordinating phenolate oxygen atoms, which are governing the crystal structure and lead to an arrangement with rather short intermolecular Co−Co distances. For complex 2, the coordination geometry is influenced by strong hydrogen bonding of the coordinating J

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Freedman, D. E. Inorg. Chem. 2014, 53, 10716−10721. (c) Cucos, P.; Tuna, F.; Sorace, L.; Matei, I.; Maxim, C.; Shova, S.; Gheorghe, R.; Caneschi, A.; Hillebrand, M.; Andruh, M. Inorg. Chem. 2014, 53, 7738−7747. (19) Buchholz, A.; Eseola, A. O.; Plass, W. C. R. Chim. 2012, 15, 929−936. (20) Cao, D.-K.; Feng, J.-Q.; Ren, M.; Gu, Y.-W.; Song, Y.; Ward, M. D. Chem. Commun. 2013, 49, 8863−8865. (21) Huang, W.; Liu, T.; Wu, D.; Cheng, J.; Ouyang, Z. W.; Duan, C. Dalton Trans. 2013, 42, 15326−15331. (22) Smolko, L.; Č ernák, J.; Dušek, M.; Miklovič, J.; Titiš, J.; Boča, R. Dalton Trans. 2015, 44, 17565−17571. (23) Nemec, I.; Herchel, R.; Trávníček, Z. Sci. Rep. 2015, 5, 10761. (24) Zadrozny, J. M.; Liu, J.; Piro, N. A.; Chang, C. J.; Hill, S.; Long, J. R. Chem. Commun. 2012, 48, 3927−3929. (25) Carl, E.; Demeshko, S.; Meyer, F.; Stalke, D. Chem. - Eur. J. 2015, 21, 10109−10115. (26) Rechkemmer, Y.; Breitgoff, F. D.; van der Meer, M.; Atanasov, M.; Hakl, M.; Orlita, M.; Neugebauer, P.; Neese, F.; Sarkar, B.; van Slageren, J. Nat. Commun. 2016, 7, 10467. (27) Zadrozny, J. M.; Telser, J.; Long, J. R. Polyhedron 2013, 64, 209−217. (28) Yang, F.; Zhou, Q.; Zhang, Y.; Zeng, G.; Li, G.; Shi, Z.; Wang, B.; Feng, S. Chem. Commun. 2013, 49, 5289−5291. (29) Boča, R.; Miklovič, J.; Titiš, J. Inorg. Chem. 2014, 53, 2367− 2369. (30) Saber, M. R.; Dunbar, K. R. Chem. Commun. 2014, 50, 12266− 12269. (31) Zhu, Y.-Y.; Zhang, Y.-Q.; Yin, T.-T.; Gao, C.; Wang, B.-W.; Gao, S. Inorg. Chem. 2015, 54, 5475−5486. (32) (a) Jurca, T.; Farghal, A.; Lin, P.-H.; Korobkov, I.; Murugesu, M.; Richeson, D. S. J. Am. Chem. Soc. 2011, 133, 15814−15817. (b) Rajnák, C.; Titiš, J.; Fuhr, O.; Ruben, M.; Boča, R. Inorg. Chem. 2014, 53, 8200−8202. (33) Habib, F.; Luca, O. R.; Vieru, V.; Shiddiq, M.; Korobkov, I.; Gorelsky, S. I.; Takase, M. K.; Chibotaru, L. F.; Hill, S.; Crabtree, R. H.; Murugesu, M. Angew. Chem., Int. Ed. 2013, 52, 11290−11293. (34) Ruamps, R.; Batchelor, L. J.; Guillot, R.; Zakhia, G.; Barra, A.-L.; Wernsdorfer, W.; Guihéry, N.; Mallah, T. Chem. Sci. 2014, 5, 3418− 3424. (35) (a) Chandrasekhar, V.; Dey, A.; Mota, A. J.; Colacio, E. Inorg. Chem. 2013, 52, 4554−4561. (b) Wu, D.; Zhang, X.; Huang, P.; Huang, W.; Ruan, M.; Ouyang, Z. W. Inorg. Chem. 2013, 52, 10976− 10982. (c) Herchel, R.; Váhovská, L.; Potočňaḱ , I.; Trávníček, Z. Inorg. Chem. 2014, 53, 5896−5898. (d) Vallejo, J.; Castro, I.; Ruiz-García, R.; Cano, J.; Julve, M.; Lloret, F.; De Munno, G.; Wernsdorfer, W.; Pardo, E. J. Am. Chem. Soc. 2012, 134, 15704−15707. (e) Colacio, E.; Ruiz, J.; Ruiz, E.; Cremades, E.; Krzystek, J.; Carretta, S.; Cano, J.; Guidi, T.; Wernsdorfer, W.; Brechin, E. K. Angew. Chem., Int. Ed. 2013, 52, 9130−9134. (36) (a) Gomez-Coca, S.; Cremades, E.; Aliaga-Alcalde, N.; Ruiz, E. J. Am. Chem. Soc. 2013, 135, 7010−7018. (b) Zhu, Y.-Y.; Cui, C.; Zhang, Y.-Q.; Jia, J.-H.; Guo, X.; Gao, C.; Qian, K.; Jiang, S.-D.; Wang, B.-W.; Wang, Z.-M.; Gao, S. Chem. Sci. 2013, 4, 1802−1806. (37) Eichhöfer, A.; Lan, Y.; Mereacre, V.; Bodenstein, T.; Weigend, F. Inorg. Chem. 2014, 53, 1962−1974. (38) Chen, L.; Wang, J.; Wei, J.-M.; Wernsdorfer, W.; Chen, X.-T.; Zhang, Y.-Q.; Song, Y.; Xue, Z.-L. J. Am. Chem. Soc. 2014, 136, 12213−12216. (39) Cinčić, D.; Kaitner, B. CrystEngComm 2011, 13, 4351−4357. (40) Senier, A.; Clarke, R. J. Chem. Soc., Trans. 1911, 99, 2081−2084. (41) Azuah, R. T.; Kneller, L. R.; Qiu, Y.; Tregenna-Piggott, P. L. W.; Brown, C. M.; Copley, J. R. D.; Dimeo, R. M. J. Res. Natl. Inst. Stand. Technol. 2009, 114, 341−358. (42) Sheldrick, G. M. SHELXS97 and SHELXL97; University of Göttingen: Göttingen, Germany, 1997. (43) Aquilante, F.; De Vico, L.; Ferré, N.; Ghigo, G.; Malmqvist, P.Å.; Neogrády, P.; Pedersen, T. B.; Pitoňaḱ , M.; Reiher, M.; Roos, B.

(4) Yang, E.-C.; Hendrickson, D. N.; Wernsdorfer, W.; Nakano, M.; Zakharov, L. N.; Sommer, R. D.; Rheingold, A. L.; Ledezma-Gairaud, M.; Christou, G. J. Appl. Phys. 2002, 91, 7382. (5) (a) Murrie, M.; Teat, S. J.; Stoeckli-Evans, H.; Güdel, H. U. Angew. Chem., Int. Ed. 2003, 42, 4653−4656. (b) Langley, S. J.; Helliwell, M.; Sessoli, R.; Rosa, P.; Wernsdorfer, W.; Winpenny, R. E. P. Chem. Commun. 2005, 5029−5031. (c) Zhang, Y.-Z.; Wernsdorfer, W.; Pan, F.; Wang, Z.-M.; Gao, S. Chem. Commun. 2006, 3302−3304. (d) Ferguson, A.; Parkin, A.; Sanchez-Benitez, J.; Kamenev, K.; Wernsdorfer, W.; Murrie, M. Chem. Commun. 2007, 3473−3475. (e) Murrie, M. Chem. Soc. Rev. 2010, 39, 1986−1995. (f) Kostakis, G. E.; Perlepes, S. P.; Blatov, V. A.; Proserpio, D. M.; Powell, A. K. Coord. Chem. Rev. 2012, 256, 1246−1278. (6) Gatteschi, D.; Sessoli, R. Angew. Chem., Int. Ed. 2003, 42, 268− 297. (7) (a) Ako, A. M.; Hewitt, I. J.; Mereacre, V.; Clérac, R.; Wernsdorfer, W.; Anson, C. E.; Powell, A. K. Angew. Chem., Int. Ed. 2006, 45, 4926−4929. (b) Ruiz, E.; Cirera, J.; Cano, J.; Alvarez, S.; Loose, C.; Kortus, J. Chem. Commun. 2008, 52−54. (8) Waldmann, O. Inorg. Chem. 2007, 46, 10035−10037. (9) Neese, F.; Pantazis, D. A. Faraday Discuss. 2011, 148, 229−238. (10) Ishikawa, N.; Sugita, M.; Ishikawa, T.; Koshihara, S.-y.; Kaizu, Y. J. Am. Chem. Soc. 2003, 125, 8694−8695. (11) Feltham, H. L. C.; Brooker, S. Coord. Chem. Rev. 2014, 276, 1− 33. (12) (a) Freedman, D. E.; Harman, W. H.; Harris, T. D.; Long, G. J.; Chang, C. J.; Long, J. R. J. Am. Chem. Soc. 2010, 132, 1224−1225. (b) Harman, W. H.; Harris, T. D.; Freedman, D. E.; Fong, H.; Chang, A.; Rinehart, J. D.; Ozarowski, A.; Sougrati, M. T.; Grandjean, F.; Long, G. J.; Long, J. R.; Chang, C. J. J. Am. Chem. Soc. 2010, 132, 18115−18126. (13) (a) Craig, G. A.; Murrie, M. Chem. Soc. Rev. 2015, 44, 2135− 2147. (b) Gómez-Coca, S.; Aravena, D.; Morales, R.; Ruiz, E. Coord. Chem. Rev. 2015, 289−290, 379−392. (c) Frost, J. M.; Harriman, K. L. M.; Murugesu, M. Chem. Sci. 2016, 7, 2470−2491. (14) (a) Zadrozny, J. M.; Xiao, D. J.; Atanasov, M.; Long, G. J.; Grandjean, F.; Neese, F.; Long, J. R. Nat. Chem. 2013, 5, 577−581. (b) Samuel, P. P.; Mondal, K. C.; Sk, N. A.; Roesky, H. W.; Carl, E.; Neufeld, R.; Stalke, D.; Demeshko, S.; Meyer, F.; Ungur, L.; Chibotaru, L. F.; Christian, J.; Ramachandran, V.; van Tol, J.; Dalal, N. S. J. Am. Chem. Soc. 2014, 136, 11964−11971. (c) Weismann, D.; Sun, Y.; Lan, Y.; Wolmershäuser, G.; Powell, A. K.; Sitzmann, H. Chem. - Eur. J. 2011, 17, 4700−4704. (d) Lin, P.-H.; Smythe, N. C.; Gorelsky, S. I.; Maguire, S.; Henson, N. J.; Korobkov, I.; Scott, B. L.; Gordon, J. C.; Baker, R. T.; Murugesu, M. J. Am. Chem. Soc. 2011, 133, 15806−15809. (e) Zadrozny, J. M.; Atanasov, M.; Bryan, A. M.; Lin, C.-Y.; Rekken, B. D.; Power, P. P.; Neese, F.; Long, J. R. Chem. Sci. 2013, 4, 125−138. (f) Feng, X.; Mathonière, C.; Jeon, I.-R.; Rouzières, M.; Ozarowski, A.; Aubrey, M. L.; Gonzalez, M. I.; Clérac, R.; Long, J. R. J. Am. Chem. Soc. 2013, 135, 15880−15884. (g) Mathonière, C.; Lin, H.-J.; Siretanu, D.; Clérac, R.; Smith, J. M. J. Am. Chem. Soc. 2013, 135, 19083−19086. (h) Mossin, S.; Tran, B. L.; Adhikari, D.; Pink, M.; Heinemann, F. W.; Sutter, J.; Szilagyi, R. K.; Meyer, K.; Mindiola, D. J. J. Am. Chem. Soc. 2012, 134, 13651−13661. (15) (a) Grigoropoulos, A.; Pissas, M.; Papatolis, P.; Psycharis, V.; Kyritsis, P.; Sanakis, Y. Inorg. Chem. 2013, 52, 12869−12871. (b) Ishikawa, R.; Miyamoto, R.; Nojiri, H.; Breedlove, B. K.; Yamashita, M. Inorg. Chem. 2013, 52, 8300−8302. (c) Vallejo, J.; Pascual-Á lvarez, A.; Cano, J.; Castro, I.; Julve, M.; Lloret, F.; Krzystek, J.; De Munno, G.; Armentano, D.; Wernsdorfer, W.; Ruiz-García, R.; Pardo, E. Angew. Chem., Int. Ed. 2013, 52, 14075−14079. (16) Poulten, R. C.; Page, M. J.; Algarra, A. G.; Le Roy, J. J.; López, I.; Carter, E.; Llobet, A.; Macgregor, S. A.; Mahon, M. F.; Murphy, D. M.; Murugesu, M.; Whittlesey, M. K. J. Am. Chem. Soc. 2013, 135, 13640− 13643. (17) Gómez-Coca, S.; Urtizberea, A.; Cremades, E.; Alonso, P. J.; Camón, A.; Ruiz, E.; Luis, F. Nat. Commun. 2014, 5, 4300. (18) (a) Zadrozny, J. M.; Long, J. R. J. Am. Chem. Soc. 2011, 133, 20732−20734. (b) Fataftah, M. S.; Zadrozny, J. M.; Rogers, D. M.; K

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry O.; Serrano-Andrés, L.; Urban, M.; Veryazov, V.; Lindh, R. J. Comput. Chem. 2010, 31, 224−247. (44) Turbomole 6.4, a development of University of Karlsruhe and Forschungszentrum Karlsruhe GmbH, 1989−2007, and Turbomole GmbH, since 2007; available from http://www.turbomole.com. (45) Hess, B. A.; Marian, C. M.; Wahlgren, U.; Gropen, O. Chem. Phys. Lett. 1996, 251, 365−371. (46) Finley, J.; Malmqvist, P.-A.; Roos, B. O.; Serrano-Andrés, L. Chem. Phys. Lett. 1998, 288, 299−306. (47) Malmqvist, P. Å.; Roos, B. O.; Schimmelpfennig, B. Chem. Phys. Lett. 2002, 357, 230−240. (48) Chibotaru, L. F.; Ungur, L.; Aronica, C.; Elmoll, H.; Pilet, G.; Luneau, D. J. Am. Chem. Soc. 2008, 130, 12445−12455. (49) Chibotaru, L. F.; Ungur, L. J. Chem. Phys. 2012, 137, 064112. (50) (a) Antonov, L.; Fabian, W. M. F.; Nedeltcheva, D.; Kamounah, F. S. J. Chem. Soc., Perkin Trans. 2000, 2, 1173−1179. (b) Gawinecki, R.; Kuczek, A.; Kolehmainen, E.; Ośmiałowski, B.; Krygowski, T. M.; Kauppinen, R. J. Org. Chem. 2007, 72, 5598−5607. (c) Flores-Leonar, M.; Esturau-Escofet, N.; Méndez-Stivalet, J. M.; Marín-Becerra, A.; Amador-Bedolla, C. J. Mol. Struct. 2011, 1006, 600−605. (51) (a) Dutta, S.; Biswas, P.; Dutta, S. K.; Nag, K. New J. Chem. 2009, 33, 847−852. (b) Chen, C.; Chen, H.; Yan, P.; Hou, G.; Li, G. Inorg. Chim. Acta 2013, 405, 182−187. (52) (a) Zabrodsky, H.; Peleg, S.; Avnir, D. IEEE Trans. Patt. Anal. Mach. Int. 1995, 17, 1154−1166. (b) Pinsky, M.; Avnir, D. Inorg. Chem. 1998, 37, 5575−5582. (53) (a) Casanova, D.; Cirera, J.; Llunell, M.; Alemany, P.; Avnir, D.; Alvarez, S. J. Am. Chem. Soc. 2004, 126, 1755−1763. (b) Cirera, J.; Ruiz, E.; Alvarez, S. Chem. - Eur. J. 2006, 12, 3162−3167. (54) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen, A. G.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1987, S1−S19. (55) (a) Burkhardt, A.; Spielberg, E. T.; Görls, H.; Plass, W. Inorg. Chem. 2008, 47, 2485−2493. (b) Spielberg, E. T.; Plass, W. Eur. J. Inorg. Chem. 2011, 2011, 826−834. (c) Spielberg, E. T.; Plass, W. Eur. J. Inorg. Chem. 2010, 2010, 3093−3096. (56) (a) Mooibroek, T. J.; Gamez, P.; Reedijk, J. CrystEngComm 2008, 10, 1501−1515. (b) Egli, M.; Sarkhel, S. Acc. Chem. Res. 2007, 40, 197−205. (57) (a) Nishio, M. CrystEngComm 2004, 6, 130−158. (b) Roth, A.; Buchholz, A.; Gärtner, M.; Malassa, A.; Görls, H.; Vaughan, G.; Plass, W. Z. Anorg. Allg. Chem. 2007, 633, 2009−2018. (58) Kahn, O. Molecular Magnetism; VCH Publishers: Weinheim, Germany1993. (59) Boča, R. Coord. Chem. Rev. 2004, 248, 757−815. (60) Cole, K. S.; Cole, R. H. J. Chem. Phys. 1941, 9, 341−351. (61) Luis, F.; Mettes, F.; Evangelisti, M.; Morello, A.; de Jongh, L. J. J. Phys. Chem. Solids 2004, 65, 763−771. (62) (a) Suturina, E. A.; Maganas, D.; Bill, E.; Atanasov, M.; Neese, F. Inorg. Chem. 2015, 54, 9948−9961. (b) Maganas, D.; Sottini, S.; Kyritsis, P.; Groenen, E. J. J.; Neese, F. Inorg. Chem. 2011, 50, 8741− 8754. (63) Krzystek, J.; Zvyagin, S. A.; Ozarowski, A.; Fiedler, A. T.; Brunold, T. C.; Telser, J. J. Am. Chem. Soc. 2004, 126, 2148−2155.

L

DOI: 10.1021/acs.inorgchem.6b00373 Inorg. Chem. XXXX, XXX, XXX−XXX