[Co5Tp*4(Me2bta)6]: A Highly Symmetrical Pentanuclear Kuratowski

Experimental Physics V, Center for Electronic Correlations and Magnetism, Institute of Physics, University of Augsburg, Universitaetsstrasse 1, D-8615...
4 downloads 10 Views 3MB Size
Download PDF
Article pubs.acs.org/IC

[Co5Tp*4(Me2bta)6]: A Highly Symmetrical Pentanuclear Kuratowski Complex Featuring Tris(pyrazolyl)borate and Benzotriazolate Ligands Tamas W. Werner,† Stephan Reschke,‡ Hana Bunzen,† Hans-Albrecht Krug von Nidda,‡ Joachim Deisenhofer,‡ Alois Loidl,‡ and Dirk Volkmer*,† †

Chair of Solid State and Materials Chemistry, Institute of Physics, University of Augsburg, Universitaetsstrasse 1, D-86159 Augsburg, Germany ‡ Experimental Physics V, Center for Electronic Correlations and Magnetism, Institute of Physics, University of Augsburg, Universitaetsstrasse 1, D-86159 Augsburg, Germany S Supporting Information *

ABSTRACT: The pentanuclear Co(II) complex [Co5Tp*4(Me2bta)6] containing N-donor ligands (5,6-dimethyl benzotriazolate; Me2bta6) and N-donor capping ligands (tris(3,5-dimethyl-1-pyrazolyl)borate; Tp*) was prepared by a simple and efficient ligand exchange reaction from [Co5Cl4(Me2bta)6] and tetra-n-butyl ammonium tris(3,5-dimethyl-1p y r a z o l y l ) b o r a t e. C o m p ar e d t o t h e p r e c u r s o r c o m p le x [Co5Cl4(Me2bta)6], which contains one Co(II) ion in octahedral and four Co(II) ions in tetrahedral coordination geometry, the title compound features all five Co(II) ions in an octahedral coordination environment while keeping a high complex symmetry. This results in modified properties including improved solubility and distinct magnetic behavior as compared to the precursor complex. The molecular structure and phase purity of the compound was verified by XRPD, UV−vis, ESI-MS, IR, and NMR measurements. Thermal stability of the compound was determined via TGA. The magnetic properties of here reported novel complex [Co5Tp*4(Me2bta)6] as well as its precursor [Co5Cl4(Me2bta)6] were examined in detail via ESR and SQUID measurements, which indicated weak anti-ferromagnetic exchange interactions between high-spin Co(II) centers at T < 20 and 50 K, respectively.

■

to the series of compounds first documented by Marshall in 197812 and later named “Kuratowski-type” complexes by our group11 for their nonplanar connectivity graph K3,3, which was first theoretically described by the mathematician Cazimierz Kuratowski in 1930.13 However, only few of the hitherto described Kuratowski complexes display the full Td symmetry found in 1 and [ZnoZnt4Cl4(Me2bta)6].14 The reason is that a preparation of such highly symmetrical complexes requires almost exclusively to use monodentate ligands (linear and negatively charged), such as chloride ions, since less symmetrical ligands, that is, bidentate ligands such as nitrate or acetylacetonate, lead to breakage of symmetry represented by the local C3-rotation axis running through the central and peripheral metal ions. Aiming at Kuratowski complexes containing five octahedrally Co(II) ions, we selected a trispyrazolylborate ligand (Tp*). Tp* is a C3 symmetrical tridentate ligand that can bind to the peripheral metal ions of a Kuratowski complex, thus retaining its full Td point group symmetry. The Tp* ligand has been studied extensively as one of the fundamental ligands forming

INTRODUCTION The coordination properties of 1,2,3-triazolates have found widespread use in the synthesis of multinuclear, supramolecular structures such as homo-1 and heteronuclear2 complexes, linear3 and planar4 coordination polymers, and metal−organic frameworks (MOFs).5 All of these feature the unique arrangement of three neighboring N-donor atoms, mostly coordinated to transition metals, although coordination to other metals has also been reported.6 The variance in metals, ligands, and in the overall structure leads to many different potential applications including semiconducting porous materials,7 templates for MOF synthesis,8 and (quantum) molecular sieving.9 Under suitable reaction conditions 5,6-dimethylbenzotriazolate (Me2bta) and CoCl2 form a highly symmetrical complex [CooCot4Cl4(Me2bta)6]10 (1; Scheme 1a), which is analogous to the previously reported complex [ZnoZnt4Cl4(Me2bta)6].11 Both compounds contain pentanuclear building units, in which the central metal ion has an octahedral geometry (Mo) and is coordinated to six triazolate ligands. The other four peripherally placed metal ions feature a tetrahedral coordination environment (Mt), and each of them is bound to a monodentate Cl(−) anion and three N-donor atoms stemming from different benzotriazolate ligands. These complexes belong © XXXX American Chemical Society

Received: September 2, 2015

A

DOI: 10.1021/acs.inorgchem.5b01982 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Scheme 1. Syntheses of 1 (a) and 2 (b)

“scorpionate complexes” of the [MIITp2] and [MIITp*X] types with metal(II) ions.15 For these reasons, we decided to introduce the capping Tp* ligand via a ligand exchange reaction on the complex 1 (Scheme 1b) to eliminate a formation of any unintended byproducts of the Tp* with metal ions. The molecular structure and phase purity of the highly symmetrical complex 2, containing five Co(II) ions all coordinated in an octahedral environment, was investigated by XRPD, UV−vis, ESI-MS, IR, and NMR measurements. Additionally, its thermal stability was studied by a TGA measurement. The magnetic properties of both complexes were investigated via ESR and SQUID providing a more detailed understanding of the couplings in these compounds, which share the same number of Co(II) ions but demonstrate different magnetic properties due to their differing coordination geometry.

■

RESULTS AND DISCUSSION Synthesis and Analysis. In this work tris(3,5-dimethyl-1pyrazolyl)borate (Tp*) was chosen as a capping ligand due to its C3 symmetry. The direct synthesis of complex 2 by mixing a Co(II)-metal salt and the ligands, namely, Me2bta and Tp*, is not possible, since the Tp* ligands are known to form mononuclear complexes of the [MII(Tp*)2] or [MIITp*X] type with metal(II) salts.15 However, we discovered that the desired highly symmetrical complex 2 can be prepared by employing a ligand exchange strategy. For this purpose, the recently reported [Co5Cl4(Me2bta)6] complex (1) was prepared as a precursor complex.10 This complex contains one Co(II) ion in an octahedral coordination, whereas the other four Co(II) ions have a tetrahedral geometry, and each of them binds a single chloride ion. The chloride ligands could be exchanged to Tp* ligands via an efficient and fast reaction as we demonstrate in this work. Efforts to synthesize an analogous complex with the unsubstituted Tp ligand, however, resulted invariably in the formation of the mononuclear sandwich complex [CoTp2]. Molecular mechanics modeling studies employing scorpionates with bulkier substituents, however, indicated strong repulsive interactions between the scorpionates and the Kuratowski core in the envisioned complexes. The title complex [Co5Tp*4(Me2bta)6] (2) was prepared by a reaction between 1 and the Tp* ligand (tris(3,5-dimethyl-1pyrazolyl)borate) at room temperature in chloroform. The introduction of Tp* as capping ligands resulted in a change of coordination of the peripheral Co(II) ions from a tetrahedral to an octahedral geometry (Figure 1), which was directly observed as a sudden color change from blue-green (the color of complex 1) to orange (the color of complex 2). As we have pointed out recently,16 each peripheral metal ion in the Kuratowski unit of 1 is structurally analogous to a scorpionate half-sandwich

Figure 1. Connectivity patterns of [Co5Cl4(Me2bta)6] 1 (a) and [Co5Tp*4(Me2bta)6] 2 (b); red = Tp*, blue = Me2bta. Methyl groups and hydrogen atoms are omitted for clarity.

coordination unit. The addition of Tp* ligands changes this environment to a complete sandwich coordination. By letting chloroform slowly evaporate from the reaction mixture, after the addition of a drop of dimethylformamide (DMF), yellow-orange crystals of 2 suitable for single-crystal Xray analysis were obtained. Their morphology was investigated by environmental scanning electron microscopy and optical microscopy (Figure 2). The images revealed the {100} and {010} faces of a tetragonal prism truncated by {111} faces. The phase purity of 2 was proven by XRPD (Figure S4 in Supporting Information). B

DOI: 10.1021/acs.inorgchem.5b01982 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

directly observed in the significant changes in the recorded 1H NMR spectrum (Figure 4). The NMR signals of methyl groups

Figure 4. NMR spectrum of [Co5Tp*4(Me2bta)6] in toluene-d8 at 330 K.

of the Tp* ligand appear at −51.60 and 32.67 ppm. A similar trend, that is, a shift to very high and low chemical shift values, was observed in the spectrum of the [CoTp*2] complex, which was recorded as a reference to assign signals to the chemically different methyl groups in complex 2 (Figure S3 in Supporting Information). Our findings are also in accordance to earlier reported NMR studies on [CoTp*2].17 ESI-MS measurements in the positive mode revealed the presence of desired compound 2 as [2]+, [2+Na]+, and [2+K]+ cations (Figure 5). Neither complex 1 nor any other metal complex could be

Figure 2. ESEM and optical micrographs of [Co5Tp*4(Me2bta)6] single crystals including indexes of visible crystal faces.

UV−vis−NIR spectra were recorded at room temperature to check for the presence of tetrahedrally coordinated Co(II) in a bulk sample (Figure 3), thus indicating an incomplete chloride

Figure 3. UV−vis reflection spectra of [Co5Cl4(Me2bta)6], [CoTp*2], and [Co5Tp*4(Me2bta)6] after Kubelka−Munk transformation.

versus Tp* ligand exchange. The recorded spectrum shows none of the characteristic bands of tetrahedral Co(II) found in the spectrum of 1 at 1530 and 640 nm. Instead, a band at 1050 nm and a shoulder at ∼500 nm were detected. These bands correspond to 4 T 2g - 4 T 1g and 4 T1g(P)-4T1g′ transitions typical for Co(II) in octahedral coordination, which can also be found at similar wavelength ranges for the sandwich complex [CoTp*2] featuring a very similar coordination environment. The new complex 2 was further characterized by NMR, IR (Figure S2 in Supporting Information), and MS measurements. The presence and magnetic influence of the Co(II) ions can be

Figure 5. ESI-MS of [Co5Tp*4(Me2bta)6]. (inset) Magnification of the (m/z) range from 2357 to 2365 Da.

detected. In addition, the thermal stability of complex 2 was studied by TGA. It was found that the complex is stable up to 340 °C and then decomposes (Figure S4 in Supporting Information). Crystallographic Studies. The structure of complex 2 was determined by single-crystal X-ray analysis (Table S1 with detailed crystallographic data in Supporting Information). It was found that the substitution of chloride with multidentate C

DOI: 10.1021/acs.inorgchem.5b01982 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry ligands, in this case Tp*, causes a reduction of the point group symmetry from Td to T, which in consequence results in a much bigger asymmetric unit (Figure 6) compared to the precursor 1, which has only nine non-hydrogen atoms in its asymmetric unit.

Figure 7. Space-filling model of [Co5Tp*4(Me2bta)6] created via Qutemol.19

improved solubility of the complex 2 in organic solvents (including chloroform, dichloromethane, toluene, and benzene) when compared to the precursor 1. Magnetic Properties. Magnetic susceptibility measurements of polycrystalline samples 1 and 2 were performed in a magnetic field of 1000 Oe for temperatures between 1.8 and 400 K. Figure 8 shows the temperature dependence of the molar susceptibility χm(T), the product Tχm(T), and the inverse molar susceptibilities χ−1 m (T) for both samples versus temperature. Both for 1 and 2, Tχm(T) decreases with decreasing temperature; χ−1 m (T) exhibits a linear behavior above 50 and 100 K, respectively, and tends toward zero at low temperatures.

Figure 6. ORTEP18 plots of the asymmetric unit (a) and the coordination environment of the central (b) and peripheral Co atoms (c). Hydrogen atoms are omitted for clarity.

Analysis of the interatomic dihedral angles reveals a slightly distorted octahedral coordination of the peripheral cobalt atoms, as they deviate from the 90° angles found in the perfectly octahedral symmetry of the central cobalt atom by up to 4.24° (N1a−Co2−N8). The bond lengths between cobalt in oxidation state +II and coordinated nitrogen donor atoms range from 2.131 to 2.202 Å (Table 1), which is close to the values found in the [CoTp*2] complex.17 The space-filling model of 2 (Figure 7) shows that the surface of the complex is completely covered by CH substituents (i.e., methyl groups and aromatic hydrogen atoms of the two different organic ligands), and therefore the cobalt atoms are completely encapsulated by a hydrophobic shell. This was further experimentally observed in an overall Table 1. Selected Interatomic Distances (Å) and Angles (deg) of [Co5Tp*4(Me2bta)6] distances

Å

angles

deg

Co1−N2 Co1−N4 Co2−N1 Co2−N3 Co2−N5 Co2−N6 Co2−N8 Co2−N10 B1−N7 B1−N9 B1−N11

2.201(3) 2.132(4) 2.131(4) 2.135(4) 2.136(4) 2.202(4) 2.169(4) 2.162(4) 1.528(10) 1.546(9) 1.523(9)

N4a−Co1−N4b N2a−Co1−N2c N2a−Co1−N2b N2a−Co1−N4a N1a−Co2−N6 N3b−Co2−N10 N5−Co2−N8 N1a−Co2−N3b N3b−Co2−N6 N1a−Co2−N8 N5−Co2−N6

180.0 179.4(2) 90.002(2) 90.32(11) 179.70(18) 175.84(17) 173.97(15) 90.48(14) 89.64(16) 94.24(16) 89.25(16)

Figure 8. Temperature dependence of the molar susceptibility χm(T) and representation Tχm(T) at 1000 Oe for 1 (upper) and 2 (lower). (insets) Inverse molar susceptibilities 1/χm(T). Open symbols show the experimental data, red solid lines represent fits with (mainframe) the model Hamiltonian given in the text and (inset) a Curie−Weiss law. D

DOI: 10.1021/acs.inorgchem.5b01982 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

and g-factors must be treated as tensors, and their spatial orientation must be taken into account. Moreover zero-field splittings (zfs) due to the crystal electric field acting on the Co(II) ions must be considered. However, as we deal with polycrystalline material we average over all possible orientations. Hence, to reduce the fit parameters, we used the following isotropic Heisenberg−Dirac−VanVleck spin Hamiltonian for the exchange coupling combined with terms for the Zeeman splitting in the external magnetic field H⃗ (eq 1), which corresponds to a minimal model for the description of the magnetic properties.

The high-temperature magnetic moments were analyzed assuming a Curie−Weiss law χm(T) = C/(T − θCW) for the linear dependence of χ−1 m (T) above 150 K. This yields a Curie− Weiss temperature of θCW = −37 K and a Curie constant of C = 10.2 emu mol−1 K for 1, which only deviates slightly from the expected spin-only value (g = 2.0) for five non-interacting highspin Co(II) ions with S = 3/2 (C = 9.38 emu mol−1 K). The corresponding values for 2 are θCW = −84 K and C = 21.1 emu mol−1 K, significantly deviating from the spin-only value. Thus, the average g-factor of 3.0, which is also found in other octahedral Co(II) complexes, indicates an orbital contribution of Co(II) in octahedral crystal field, which can be significant.20 Magnetization curves M(H) for both samples were measured in magnetic fields H up to 50 kOe at temperatures of 2, 5, and 10 K as shown in Figure 9.

⎡ ⎢ 5 1 ⎢ ̂ ⃗ H = −2 J1(S1·∑ Si⃗ ) + J2 · ⎢ 2 i=2 ⎢ ⎣

5

∑ i,j=2 i≠j

⎤ ⎥ 5 ⎥ ⃗ ⃗ Si·Sj + gμB ∑ Si⃗ ·H⃗ ⎥ i=1 ⎥ ⎦ (1)

Although this model is only a minimal model, it is able to provide a reasonable estimate of the operating exchange interactions and allows to reproduce temperature-dependent susceptibility and field-dependent magnetization data simultaneously. Because of the high symmetry of these Kuratowski-type complexes, two different exchange constants J1 and J2 were chosen, as shown in Figure 10. For all five Co(II) ions one

Figure 10. Co(II)−Co(II) exchange couplings J1 (green) and J2 (orange) in [Co5Cl4(Me2bta)6] and [Co5Tp*4(Me2bta)6]. Figure 9. Field-dependent magnetization curves M(H) at 2, 5, and 10 K for 1 (upper) and 2 (lower). Open symbols show the experimental data, red solid lines represent fits with the model Hamiltonian.

single g-factor was used. Allowing for two different g-factors for the octahedrally and tetrahedrally coordinated Co(II) site in 1 did not result in significant changes of the exchange constants, and the g-factors tended to attain approximately the same value. For 2 two different g-factors g1 and g2 for the central and the peripheral Co(II) ions slightly improve the fit but also do not significantly change the exchange constants. Including zfs with an axial zfs parameter D for complex 1, which is dominated by Co(II) ions in tetrahedral ligand field, did also not have a major effect on the exchange constants. In fact simulations showed that the decline of Tχm(T) at low temperatures is dominated by the anti-ferromagnetic exchange interactions between the Co(II) ions possibly with small corrections due to zfs. For complex 2 the use of such a zfs term is not applicable because the strong orbital contribution breaks the spin Hamiltonian. From the analysis of our data in terms of Hamiltonian (1) we obtain the following results: For 1, the simultaneous fit of susceptibility and magnetization yields a g value of 2.02 in agreement with the nearly spin-only behavior derived from the

At 10 K one observes a linear increase of the magnetization with increasing field, while for 5 and 2 K a saturation behavior appears at high fields, which becomes more pronounced at low temperatures. Note that the saturation values anticipated from the extrapolation of the 2 K data (MS(1) ≈ 5 μB/f.u. and (MS(2) ≈ 7 μB/f.u.) turn out to be far below the value of 15 μB for five parallel spin-only Co(II) moments. This indicates the dominance of the anti-ferromagnetic exchange couplings between the spins in accordance with the negative Curie− Weiss temperatures. As the magnetic behavior of the two complexes is determined by different exchange interactions, for a deeper analysis the molar susceptibility and magnetization data were fitted using the program PHI21 with a molecular approach for exchangecoupled Co(II) ions. In the general case exchange interactions E

DOI: 10.1021/acs.inorgchem.5b01982 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry high-temperature Curie−Weiss fit described above. For the exchange constants values of J1/kB = −1.24 K and J2/kB = −2.70 K were obtained, indicating anti-ferromagnetic exchange couplings both via J1 and J2. For 2 from a simultaneous fit of magnetic susceptibility and magnetization curves a g value of g = 2.89 was obtained in line with an average g value close to 3.0 as mentioned above. For the exchange constants in 2 antiferromagnetic values of J1/kB = −8.09 K and J2/kB = −4.11 K were found. Like in 1 all Co(II) ions are coupled antiferromagnetically among each other, but in contrast here J1 mediates a stronger anti-ferromagnetic exchange interaction between central and peripheral Co(II) ions than J2 between the peripheral Co(II) ions. This might be explained by different overlapping of the orbitals due to the change of the symmetry of the peripheral Co(II) ions from tetrahedral to octahedral. But one should be careful in the interpretation of the exchange constants of complex 2, because the strong orbital contribution, which is reflected in the large g value, needs deeper analysis beyond the Hamiltonian described by eq 1. It is instructive to compare the obtained exchange constants with the average exchange estimated from the Curie−Weiss temperature in mean-field approximation using 3kBθ = 2JZS(S + 1). With Z = 4 nearest neighbors and spin S = 3/2 we find J/kB(1) = −3.7 K and J/kB(2) = −8.4 K, respectively. This is in satisfactory agreement with the values derived from our microscopic model. For further investigation of the magnetic properties of the two complexes powder X-band (9.48 GHz) and Q-band (34 GHz) ESR measurements were performed using a Bruker Elexsys II and Elexsys 500 CW spectrometer, respectively. For both frequencies typical ESR spectra are shown in Figure 11. The ESR spectra could be observed for temperatures up to ∼25 K for 1 and 80 K for 2, that is, T < |θ| where the exchange

energy within the Co(II) complex is larger than the thermal energy. Obviously at higher temperature ESR spectra cannot be detected because of strongly increasing spin relaxation on occupation of excited states. At X-band frequency the spectra of both compounds consist of a narrow central line accompanied by a broader absorption revealing a shoulder on the low-field side of the main line. Interestingly the spectra strongly differ at Q-band frequency: While for 1 the whole spectrum just shifts to higher fields as expected due to the higher frequency, we observe for 2 that the spectrum decomposes into many lines with the dominant line at even lower fields than in X-band. Because we deal with polycrystalline material and regarding the simplification of our minimal model we refrain from detailed simulation of the ESR spectra. Nevertheless it is possible to understand the basic features considering the energy level schemes resulting from this minimal model using the PHI package: Inserting the fit parameters obtained from the evaluation of the susceptibility we obtain for both complexes a doublet and a quartet coexisting in an energy interval of less than 1 K. This maybe explains the qualitative similarity of the ESR spectra observed at low microwave energies (X-band ∼0.45 K). For complex 1 a further quartet and doublet follow within an energy distance ΔE/kB of ∼3 and 6 K, respectively. Additional multiplets are found in characteristic energy distances of ∼3 K corresponding to the leading exchange constant J2. In complex 2 a sextet is found at 9 K above the ground state, and further multiplets follow in characteristic spacings of ∼9 K comparable to the leading exchange parameter J1. Already at T = 4 K the lower excited states exhibit significant thermal population and, therefore, contribute to the paramagnetic spectrum. Application of an external magnetic field splits all degenerate states with the given gyromagnetic factor. If only dipolar transitions within each multiplet are considered, one should observe only a single absorption line at the resonance field determined by the gyromagnetic factor. However, transitions between the multiplets are also possible, because they cross each other with increasing field. Thus, a rich ESR spectrum can be expected, especially when taking into account further possible zfs due to the crystal field. This is indeed observed for complex 2 at Q-band frequency. For further discussion it is instructive to consider the limit cases of vanishing J1 or J2 when only one of the exchange parameters is dominant. Comparison of X-band and Q-band spectra of 1 shows that the structure of the spectrum is not affected by the frequency change, whereas for 2 in Q-band further resonance lines can be resolved. A strong J2 coupling is resulting in compensation of the spins of the peripheral Co(II) ions and, thus, in a S = 3/2 quartet ground state arising from the central Co(II) ion, whereas for strong J1 all peripheral Co(II) spins are oriented antiparallel to the central Co(II) spin resulting in a S = 9/2 10-fold degenerate ground state. The first excited state is located at 2J2 and 2J1 above the ground state, respectively. Hence, for temperatures small compared to the exchange coupling only the ground state contributes to the ESR spectrum. For strong J2 the Zeeman effect splits the S = 3/2 ground state into four levels allowing for three magnetic dipolar transitions at the same resonance field. In presence of weak crystal field anisotropy this results in a central line and two satellites smeared out into a broad absorption by the averaging over the polycrystal. This is observed for 1 with leading J2 where the simple structure of the spectrum is preserved even at higher frequencies. In the latter case the S = 9/2 state splits into

Figure 11. ESR spectra (X-band, 9.48 GHz) and Q-band (34 GHz) of 1 (top) and 2 (bottom). The marked sharp resonances at H ≈ 3.4 kOe (X-band) and H ≈ 12 kOe (Q-band) result from the background of the resonator. Note that different amplification factors are used in Xband and Q-band. F

DOI: 10.1021/acs.inorgchem.5b01982 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

ravimetric analysis (TGA) was performed with a TGA Q500 analyzer in the temperature range of 25−800 °C in flowing nitrogen gas at the heating rate of 10 K min−1. X-ray powder diffraction data were collected in the 2θ range of 4−70° with 0.02° step width, with a time of 90 min per step, using a Bruker D8 Advance powder diffractometer (40 kV, 40 mA, Cu Kα (λ = 1.541 78 Å). Elemental analysis was measured with a Vario EL III (Elementar-Analysensysteme GmbH). The molecular graphics images of Figure 1 were produced using the Chimera package from the Computer Graphics Laboratory, University of California, San Francisco (supported by NIH P41 RR-01081).22 The calculated powder pattern of 2 was created using Mercury.23 Synthesis. 5,6-dimethylbenzotriazole (Me2btaH) is commercially available as a monohydrate from Sigma-Aldrich and was purified via sublimation before use. NBu4Tp* was prepared via metathesis from NBu4Cl (1.70 g, 5.8 mmol) and NaTp* (1.87 g, 5.8 mmol). Both reactants were dissolved in dry THF (100 and 25 mL, respectively), and the solutions were subsequently combined. The precipitating NaCl was filtered off, and the solvent was removed under reduced pressure. NaTp* was prepared from NaBH4 and 3,5-dimethylpyrazol in kerosene following a procedure described in literature.24 Synthesis of [Co5Cl4(Me2bta)6] (1). Complex 1 was synthesized according to our previously published procedures. Briefly, a solution of Me2btaH (1.5 g, 10.2 mmol) and 2,6-dimethylpyridine (1.05 mL, 9.04 mmol) in methanol (20 mL) was added in 1 min to a solution of CoCl2·6H2O (2.0 g, 8.75 mmol) in methanol (25 mL). The red precipitate that formed was filtered off and dried under vacuum. The crude product was refluxed in bromobenzene (500 mL) for 3 h. Afterward the solution was evaporated to the volume of 250 mL. Green-blue octahedral crystals of 1 were obtained after 3 d (1.59 g, 71%). Synthesis of [Co5Tp*4(Me2bta)6] (2). Complex 2 was prepared by a ligand exchange reaction on the precursor complex 1 by the following procedure. To a solution of compound 1 (100 mg, 0.08 mmol) in chloroform (20 mL), a solution of NBu4Tp* (165 mg, 0.31 mmol) in chloroform (20 mL) was added dropwise over 5 min. The solution changed its color gradually from blue-green to orange. The reaction mixture was filtered over a short column of basic aluminum oxide. Slow evaporation of the solvent resulted in a crystalline orange solid. Unreacted NBu4Tp* was removed via sublimation at 270 °C for 2 h. Yield: 101 mg (56%). For the formation of single crystals, a drop of DMF was added before solvent removal. 1H NMR (400 MHz, toluene-d8, 330 K): δ = 41.24 [s, 4H, B−H], 39.10 [s, 12H, H-bta], 36.10 [s, 12H, H-pz], 32.67 [s, 36H, 5-CH3-pz], −8.70 [s, 36H, CH3bta], −51.60 [s, 36H, 3-CH3-pz]. IR (cm−1): 3078 (w), 2966 (sh), 2926 (br), 2860 (sh), 2730 (w), 2506 (m), 2360 (w), 2325 (w), 2237 (w), 2114 (w), 1445 (s), 1418 (w), 1381 (s), 1351 (s), 1286 (s), 1261 (s), 1205 (s), 1177 (m), 1096 (m), 1065 (s), 1041 (s), 1000 (s). ESIMS (m/z) for [Co5C108H136N42B4]+: 2359.90 (calc.); 2359.90 (found). Elemental analysis (M = 2359.90 g/mol) reveals: 54.9% C, 24.9% N, 5.8% H (calcd.); C 54.67%, N 24.00% H: 6.56% (found). Crystallographic Studies. The single crystals for the analysis were obtained from the reaction mixture after adding a drop of DMF to the solution before the solvent removal as mentioned above in the Synthesis Section. Single-crystal data were collected on a Bruker D8 Venture single-crystal X-ray diffractometer. Intensity measurements were performed using monochromated (doubly curved silicon crystal) Mo Kα radiation (0.710 73 Å) from a sealed microfocus tube. Generator settings were 50 kV and 1 mA. Data collection temperature was 300 K. APEX2 software was used for preliminary determination of the unit cell. Determination of integrated intensities and unit cell refinement were performed using SAINT. The structures were solved and refined using the Bruker SHELXTL Software package.25 Magnetic Studies. Magnetic susceptibility and magnetization measurements were performed using a superconducting quantum interference device (SQUID) magnetometer (Quantum Design, MPMS5) working in the temperature range of 1.8 ≤ T ≤ 400 K at magnetic fields 0 ≤ μ0H ≤ 5 T. X-band ESR spectra (frequency ν = 9.48 GHz) were measured with a Bruker Elexsys II spectrometer using a Bruker ER 4102ST cavity together with a continuous He gas-flow cryostat (Oxford Instruments, ESR 900) working in a temperature

10 levels allowing for nine magnetic dipolar transitions. It is immediately clear that an additional crystal field will result in a much more complex spectrum, and indeed this is realized in 2 with leading J1, where the higher frequency allows to resolve a rich spectrum. Note, however, that these limiting cases are only very rough approximations of the real level schemes, but at least provide an idea about the different behavior observed. Certainly the application of our minimal model must be taken with great care, and the obtained absolute values of the fit parameters must be critically considered. In case of complex 1 the use of the spin Hamiltonian formalism is justified due to the dominant contribution of all tetrahedral peripheral Co(II) sites with completely quenched orbital contribution as evident from g = 2.02. However, in case of complex 2 the spin Hamiltonian formalism turns out to be questionable, because of nonnegligible orbital contributions indicated by the strongly enhanced g value g = 2.89. In that case a deeper theoretical analysis of the crystal field at the different Co(II) sites would be necessary for a complete description of the magnetic properties. Nevertheless, the simultaneous description of field-dependent magnetization and temperature-dependent susceptibility with only three parameters and the principle ESR properties provide a first insight into the microscopic properties of the two Co(II) complexes under consideration.

■

CONCLUSIONS In conclusion, we have synthesized and structurally characterized the novel pentanuclear cobalt complex [Co5Tp*4(Me2bta)6] (2) featuring a Kuratowski-type coordination unit, in which all peripheral metal ions are capped by scorpionate ligands. Compound 2 shows a substantially improved solubility in most organic solvents in addition to the significant changes in magnetic behavior when compared to its precursor complex [Co5Cl4(Me2bta)6] (1), highlighting the relative ease with which these can be accomplished. The almost instant total conversion at room temperature with stoichiometric amounts of reactant demonstrates the preference of the octahedral coordination geometry of the peripheral Co(II) ions. Despite its distortion, which is due to the relatively small size of the ligand, there is no significant difference in Co−N bond lengths, when compared to the central Co(II) ion and its N-donor ligands. The experiences gathered in this study may help to develop similar Kuratowski-scorpionates containing other metal ions, which might show a host of interesting physical properties. Systematic investigations on these unique pentanuclear coordination units will shed more light on the influence of intramolecular couplings of open-shell metal ions in K3,3connected coordination units hosting triazolate ligands in μ1,2,3bridging mode.

■

EXPERIMENTAL SECTION

General Information. All starting materials were of reagent grade and used as received from commercial suppliers if not declared otherwise. Fourier transform infrared (FT-IR) spectra were recorded with ATR in the range of 4000−400 cm−1 on a Bruker Equinox 55 FTIR spectrometer. Diffuse reflectance UV−vis−NIR spectra were recorded in the range of 2000−250 nm on a PerkinElmer λ 750s spectrometer equipped with a Labsphere 60 mm RSA ASSY integrating sphere. Labsphere Spectralon SRS-99 was used as a reference. Molecular masses were measured with a Q-Tof Ultima mass spectrometer (Micromass) equipped with an ESI source. ESEM micrographs were recorded with a Philips XL 30 scanning electron microscope. 1H NMR spectra were recorded with a Mercury plus 400 high-resolution system (Fa. Variant Deutschland GmbH). ThermogG

DOI: 10.1021/acs.inorgchem.5b01982 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry range of 3.7 ≤ T ≤ 300 K. Q-band ESR spectra (frequency ν = 34 GHz) were measured with a Bruker Elexsys 500 CW spectrometer using a Bruker ER5106QT cavity together with an Oxford CF935 cryostat working in a temperature range of 3.7 ≤ T ≤ 300 K. Because of the lock-in technique with field modulation the field derivative of the absorbed power is recorded. Thus, the intensity of the ESR spectra is determined from double integration of the spectra.

■

Acta Crystallogr., Sect. E: Struct. Rep. Online 2005, 61, m1108−m1110. (d) Rajeswaran, M.; Blanton, T. N.; Giesen, D. J.; Whitcomb, D. R.; Zumbulyadis, N.; Antalek, B. J.; Neumann, M. M.; Misture, S. T. J. Solid State Chem. 2006, 179, 1053−1059. (e) Zhu, K.; Chen, H.; Nishihara, S.; Liu, G.-X.; Ren, X.-M. Inorg. Chim. Acta 2009, 362, 4780−4784. (f) Wu, T.; Li, M.; Li, D.; Huang, X.-C. Cryst. Growth Des. 2008, 8, 568−574. (5) Schmieder, P.; Denysenko, D.; Grzywa, M.; Baumgärtner, B.; Senkovska, I.; Kaskel, S.; Sastre, G.; van Wüllen, L.; Volkmer, D. Dalton Trans. 2013, 42, 10786−10797. (6) (a) Zhang, X.; Xu, N.; Zhang, S.-Y.; Zhao, X.-Q.; Cheng, P. RSC Adv. 2014, 4, 40643−40650. (b) Yuan, G.; Shao, K.-Z.; Wang, X.-L.; Lan, Y.-Q.; Du, D.-Y.; Su, Z.-M. CrystEngComm 2010, 12, 1147−1152. (c) Reedijk, J.; Roelofsen, G.; Siedle, A. R.; Spek, A. L. Inorg. Chem. 1979, 18, 1947−1951. (d) Wang, X.; Wang, Y.; Liu, G.; Tian, A.; Zhang, J.; Lin, H. Dalton Trans. 2011, 40, 9299−9305. (7) Sippel, P.; Denysenko, D.; Loidl, A.; Lunkenheimer, P.; Sastre, G.; Volkmer, D. Adv. Funct. Mater. 2014, 24, 3885−3896. (8) (a) Wang, X.-L.; Qin, C.; Wu, S.-X.; Shao, K.-Z.; Lan, Y.-Q.; Wang, S.; Zhu, D.-X.; Su, Z.-M.; Wang, E.-B. Angew. Chem., Int. Ed. 2009, 48, 5291−5295. (b) Zhong, D.-C.; Deng, J.-H.; Luo, X.-Z.; Liu, H.-J.; Zhong, J.-L.; Wang, K.-J.; Lu, T.-B. Cryst. Growth Des. 2012, 12, 1992−1998. (c) Bai, Y.-L.; Tao, J.; Huang, R.-B.; Zheng, L.-S. Angew. Chem., Int. Ed. 2008, 47, 5344−5347. (9) (a) Sastre, G.; van den Bergh, J.; Kapteijn, F.; Denysenko, D.; Volkmer, D. Dalton Trans. 2014, 43, 9612−9619. (b) Teufel, J.; Oh, H.; Hirscher, M.; Wahiduzzaman, M.; Zhechkov, L.; Kuc, A.; Heine, T.; Denysenko, D.; Volkmer, D. Adv. Mater. 2013, 25, 635−639. (c) Chen, Y.-Q.; Qu, Y.-K.; Li, G.-R.; Zhuang, Z.-Z.; Chang, Z.; Hu, T.-L.; Xu, J.; Bu, X.-H. Inorg. Chem. 2014, 53, 8842−8844. (d) Tan, Y.X.; Zhang, Y.; He, Y.-P.; Zheng, Y.-J. New J. Chem. 2014, 38, 5272− 5275. (10) Denysenko, D.; Werner, T.; Grzywa, M.; Puls, A.; Hagen, V.; Eickerling, G.; Jelic, J.; Reuter, K.; Volkmer, D. Chem. Commun. 2012, 48, 1236−1238. (11) Biswas, S.; Tonigold, M.; Volkmer, D. Z. Anorg. Allg. Chem. 2008, 634, 2532−2538. (12) Marshall, J. H. Inorg. Chem. 1978, 17, 3711−3713. (13) Kuratowski, C. Fundam. Math 1930, 15, 271−283. (14) (a) Lan, Y.-Q.; Li, S.; Jiang, H.-L.; Xu, Q. Chem. - Eur. J. 2012, 18, 8076−8083. (b) Liu, Y.-Y.; Grzywa, M.; Tonigold, M.; Sastre, G.; Schüttrigkeit, T.; Leeson, N. S.; Volkmer, D. Dalton Trans. 2011, 40, 5926−5938. (c) Biswas, S.; Tonigold, M.; Speldrich, M.; Kögerler, P.; Weil, M.; Volkmer, D. Inorg. Chem. 2010, 49, 7424−7434. (d) Xue, X.; Li, G.-T.; Peng, Y.-H.; Wu, L.; Wu, B.-L. J. Coord. Chem. 2011, 64, 1953−1962. (15) Trofimenko, S. Chem. Rev. 1993, 93, 943−980. (16) Denysenko, D.; Grzywa, M.; Jelic, J.; Reuter, K.; Volkmer, D. Angew. Chem., Int. Ed. 2014, 53, 5832−5836. (17) Myers, W. K.; Duesler, E. N.; Tierney, D. L. Inorg. Chem. 2008, 47, 6701−6710. (18) Farrugia, L. J. J. Appl. Crystallogr. 2012, 45, 849−854. (19) Tarini, M.; Cignoni, P.; Montani, C. IEEE Trans. Vis. Comput. Graph. 2006, 12, 1237−1244. (20) Lloret, F.; Julve, M.; Cano, J.; Ruiz-Carcia, R.; Pardo, E. Inorg. Chim. Acta 2008, 361, 3432−3445. (21) Chilton, N. F.; Anderson, R. P.; Turner, L. D.; Soncini, A.; Murray, K. S. J. Comput. Chem. 2013, 34, 1164−1175. (22) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.; Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. J. Comput. Chem. 2004, 25, 1605−1612. (23) Macrae, C. F.; Edgington, P. R.; McCabe, P.; Pidcock, E.; Shields, G. P.; Taylor, R.; Towler, M.; van de Streek, J. J. Appl. Crystallogr. 2006, 39, 453−457. (24) Bucher, U. E.; Currao, A.; Nesper, R.; Rueegger, H.; Venanzi, L. M.; Younger, E. Inorg. Chem. 1995, 34, 66−74. (25) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, A64, 112−122.

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.5b01982. Crystallographic data for 2. (CIF) IR, MS, XRPD, and TGA data of 2 and 1H NMR spectrum of [CoTp*2]. (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Funding

S.R., H.-A.K.v.N., and A.L. acknowledge financial support by the German Research Foundation (DFG) via the Transregional Collaborative Research Center TRR 80 (Augsburg, Munich). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Dr. B. Bredenkötter and Dr. M. Grzywa from the Univ. of Augsburg for recording the NMR spectra and single crystal structural data, respectively.

■

REFERENCES

(1) (a) Shaw, R.; Laye, R. H.; Jones, L. F.; Low, D. M.; TalbotEeckelaers, C.; Wei, Q.; Milios, C. J.; Teat, S.; Helliwell, M.; Raftery, J.; Evangelisti, M.; Affronte, M.; Collison, D.; Brechin, E. K.; McInnes, E. J. L. Inorg. Chem. 2007, 46, 4968−4978. (b) Zou, J.-Y.; Shi, W.; Gao, H.-L.; Cui, J.-Z.; Cheng, P. Chin.J.Inorg.Chem. 2014, 30, 149−156. (c) Deeming, A. J.; Whittaker, C.; Arce, A. J.; De Sanctis, Y. J. Organomet. Chem. 1997, 540, 67−76. (d) Han, S.-D.; Chen, Y.-Q.; Zhao, J.-P.; Liu, S.-J.; Miao, X.-H.; Hu, T.-L.; Bu, X.-H. CrystEngComm 2014, 16, 753−756. (e) Biswas, S.; Tonigold, M.; Speldrich, M.; Kögerler, P.; Volkmer, D. Eur. J. Inorg. Chem. 2009, 2009, 3094−3101. (2) (a) Sulway, S. A.; Collison, D.; McDouall, J. J. W.; Tuna, F.; Layfield, R. A. Inorg. Chem. 2011, 50, 2521−2526. (b) Sulway, S. A.; Layfield, R. A.; Bodensteiner, M.; Scheuermayer, S.; Scheer, M.; Zabel, M. Dalton Trans. 2011, 40, 7559−7563. (3) (a) Balanta-Diaz, J. A.; Moya-Cabrera, M.; Jancik, V.; PinedaCedeno, L. W.; Toscano, R. A.; Cea-Olivares, R. Inorg. Chem. 2009, 48, 2518−2524. (b) Liu, J.-J.; Li, Z.-Y.; Yuan, X.; Wang, Y.; Huang, C.-C. Acta Crystallogr., Sect. C: Struct. Chem. 2014, 70, 599−602. (c) Zhou, W.-L.; Liang, J.; Qin, C.; Shao, K.-Z.; Wang, F.-M.; Su, Z.-M. CrystEngComm 2014, 16, 7410−7418. (d) Wang, Y.; Hu, M.-C.; Zhai, Q.-G.; Li, S.-N.; Jiang, Y.-C.; Ji, W.-J. Inorg. Chem. Commun. 2009, 12, 281−285. (e) Chuang, J.; Ouellette, W.; Zubieta, J. Inorg. Chim. Acta 2008, 361, 2357−2364. (4) (a) Luo, F.; Ning, Y.; Luo, M.; Huang, G. CrystEngComm 2010, 12, 2769−2774. (b) Gao, H.; Lou, X.-H.; Li, Q.-T.; Du, W.-J.; Xu, C. Synth. React. Inorg., Met.-Org., Nano-Met. Chem. 2015, 45, 309−311. (c) Sharma, V. C.; Lal, U.; Singh, T.; Chopra, D.; Guru Row, T. N. H

DOI: 10.1021/acs.inorgchem.5b01982 Inorg. Chem. XXXX, XXX, XXX−XXX