Novel Co5 and Ni4 Metal Complexes and Ferromagnets by the Combination of 2-Pyridyl Oximes with Polycarboxylic Ligands

The use of 2-pyridyl oximes in metal complexes chemistry has been extensively investigated in the last few decades as a fruitful source of species with interesting magnetic properties. In this work, the initial combination of pyridine-2-amidoxime (pyaoxH2) and 2-methyl pyridyl ketoxime (mpkoH) with isonicotinic acid (HINA) and 3,5-pyrazole dicarboxylic acid (H3pdc) has provided access to three new compounds, [Ni4(INA)2(pyaox)2(pyaoxH)2(DMF)2] (1), [Co5(mpko)6(mpkoH)2(OMe)2(H2O)](ClO4)6 (2), and [Co5(OH)(Hpdc)5(H2pdc)] (3). 1 displays a square-planar metal topology, being the first example that bears simultaneously HINA and pyaoxH2 in their neutral or ionic form. The neighbouring Ni4 units in 1 are held together through strong intermolecular hydrogen bonding interactions, forming a three-dimensional supramolecular framework. 2 and 3 are mixed-valent Co4IIICoII and Co2IIICoII3 compounds with a bowtie and trigonal bipyramidal metal topology, accordingly. Direct current and alternate current magnetic susceptibility studies revealed that the exchange interactions between the NiII ions in 1 are ferromagnetic (J = 1.79(4) cm−1), while 2 exhibits weak AC signals in the presence of a magnetic field. The syntheses, crystal structures, and magnetic properties of 1–3 are discussed in detail.


Synthetic Discussion
Our group has been exploring the employment of 2-pyridyl oximes in combination with polycarboxylic acids toward the synthesis of new metal compounds. The latter involved experiments with the use of benzene-1,4-dicarboxylic acid, benzene-1,3,5tricarboxylic acid and benzene-1,2,4,5-tetracarboxylic acid [50][51][52]. The initial promising results prompted us to explore further this reaction system by using different carboxylic ligands, such as isonicotinic acid (HINA) and 3,5-pyrazole dicarboxylic acid (H3pdc). To this end, a wide range of experiments was conducted, aiming at studying the effect of different synthetic parameters (temperature, presence/absence or kind of base, metal ratio of the reactants, metal sources, etc.) on the identity and crystallinity of the isolated product(s) and the yield of the reaction.
The reaction mixture of Ni(ClO4)2.6H2O/pyaoxH2/HINA (2:4:1.5) in DMF at 110 °C gave a dark brown solution from which dark brown crystals of [Ni4(INA)2(pyaox)2(pyaoxH)2(DMF)2] (1) were subsequently isolated in good yield. The stoichiometric equation of the reaction that led to the formation of 1 is represented in Equation (1) The kind of base and/or metal source does not have any impact on the identity of the isolated compound, but it affects its crystallinity. The next step included the replacement of Ni II by Co II/III wishing to shed light on how the type of the metal ion affects the identity of the product. Similar reactions that led to 1 were performed by using a Co II source instead of Ni, which, in their vast majority, led to the isolation of previously reported single-ligand Co II MOFs based on INA − [53,54]. Likewise, the use of mpkoH instead of pyaoxH2 provided access to compounds that contain INA − . This prompted us to perform Scheme 1. Schematic representation of 2-pyridyl oximes (left), isonicotinic acid (centre) and 3,5pyrazole dicarboxylic acid (right).

Synthetic Discussion
Our group has been exploring the employment of 2-pyridyl oximes in combination with polycarboxylic acids toward the synthesis of new metal compounds. The latter involved experiments with the use of benzene-1,4-dicarboxylic acid, benzene-1,3,5tricarboxylic acid and benzene-1,2,4,5-tetracarboxylic acid [50][51][52]. The initial promising results prompted us to explore further this reaction system by using different carboxylic ligands, such as isonicotinic acid (HINA) and 3,5-pyrazole dicarboxylic acid (H 3 pdc). To this end, a wide range of experiments was conducted, aiming at studying the effect of different synthetic parameters (temperature, presence/absence or kind of base, metal ratio of the reactants, metal sources, etc.) on the identity and crystallinity of the isolated product(s) and the yield of the reaction.
The kind of base and/or metal source does not have any impact on the identity of the isolated compound, but it affects its crystallinity. The next step included the replacement of Ni II by Co II/III wishing to shed light on how the type of the metal ion affects the identity of the product. Similar reactions that led to 1 were performed by using a Co II source instead of Ni, which, in their vast majority, led to the isolation of previously reported single-ligand Co II MOFs based on INA − [53,54]. Likewise, the use of mpkoH instead of pyaoxH 2 provided access to compounds that contain INA − . This prompted us to perform reactions with an excess of the oximic ligand, which led to the new mixed-valent cationic pentanuclear compound [Co III 4 Co II (mpko) 6 (mpkoH) 2 (OMe) 2 (H 2 O)](ClO 4 ) 6 (2), while the use of H 3 pdc in the place of HINA led to the isolation of the new neutral pentanuclear compound [Co 5 (OH)(Hpdc) 5 (H 2 pdc)] (3). The stoichiometric equations of the reaction that describe the formation of 2 and 3 are shown in Equations (2) and (3).

Description of Structures
Representations of the molecular structures of 1-3 are shown in Figures 1-3. Selected interatomic distances and angles are listed in Tables S1-S3.
The tetranuclear {Ni4} single-decker core of compound 1 has been found previously as a fundamental building unit in a couple of multiple-decker based polynuclear Ni complexes. Being composed by three and four {Ni4} layers/deckers, respectively, Ni12 and Ni16 complexes can be assumed as the trimer and the tetramer version of 1 [40,41]. Both Ni12 and Ni16 are ferromagnets bearing a high-spin ground state of S = 6 and S = 8, respectively. Ferromagnets are a special category of magnetic materials with their properties being derived by the ferromagnetic exchange coupling among the paramagnetic centres. They are candidates for interesting potential applications, such as spintronics, magnetic coolers, etc.

Magnetism Studies
Direct current magnetic susceptibility measurements (DC) were carried out on samples of 1, 2 and 3 in the 2-300 K temperature range and under a field of 0.03 T (1.28 MHz), and they are plotted as χMT vs. T plot. (Figure 4). The χMT values at room temperature (2.30, 1.95 and 4.31 cm 3 mol −1 K for 1, 2 and 3, respectively) are very close to the theoretical spin-only value of 2.00, 1.875 and 5.625 cm 3 mol −1 K corresponding to two Ni II , one Co II and three Co II non-interacting cations, respectively. 1 crystallizes in the monoclinic space group P2 1 /c. Its structure is based on neutral, centrosymmetric [Ni 4 (INA) 2 (pyaox) 2 (pyaoxH) 2 (DMF) 2 ] units ( Figure 1) that are held together through strong hydrogen bonding interactions forming a 3D network ( Figure S1). The Ni 4 compound in 1 is composed of two distorted octahedral (Ni1 and its symmetry equivalent) and two square planar Ni II atoms (Ni2 and its symmetry equivalent), which are connected through two η 1 :η 1 :η 1 :µ pyaoxH -, and two η 1 :η 1 :η 1 :η 2 :µ 3 pyaox 2ligands (Scheme 2). The coordination sphere in Ni1 and Ni1' is completed by one terminally ligated DMF molecule and one monodentate INA − ion. The four metal centres in 1 are co-planar.

Description of Structures
Representations of the molecular structures of 1-3 are shown in Figures 1-3. Selected interatomic distances and angles are listed in Tables S1-S3. 1 crystallizes in the monoclinic space group P21/c. Its structure is based on neutral, centrosymmetric [Ni4(INA)2(pyaox)2(pyaoxH)2(DMF)2] units ( Figure 1) that are held together through strong hydrogen bonding interactions forming a 3D network ( Figure  S1). The Ni4 compound in 1 is composed of two distorted octahedral (Ni1 and its symmetry equivalent) and two square planar Ni II atoms (Ni2 and its symmetry equivalent), which are connected through two η 1 :η 1 :η 1 :μ pyaoxH -, and two η 1 :η 1 :η 1 :η 2 :μ3 pyaox 2-ligands (Scheme 2). The coordination sphere in Ni1 and Ni1' is completed by one terminally ligated DMF molecule and one monodentate INA − ion. The four metal centres in 1 are co-planar.  (Figure 1, right), which can be described as a pseudo 3D polymer or as a 3D supramolecular framework. It is worth mentioning that 1 displays significant thermal stability ( Figure S3), which is a result of its pseudo-polymeric nature. In particular, there is a mass loss of ca 2% below 100 • C, which can be attributed to adsorbed humidity; this is then followed by a plateau with the step at 350 • C corresponding to the compound breakdown.
The tetranuclear {Ni 4 } single-decker core of compound 1 has been found previously as a fundamental building unit in a couple of multiple-decker based polynuclear Ni complexes. Being composed by three and four {Ni 4 } layers/deckers, respectively, Ni 12 and Ni 16 complexes can be assumed as the trimer and the tetramer version of 1 [40,41]. Both Ni 12 and Ni 16 are ferromagnets bearing a high-spin ground state of S = 6 and S = 8, respectively. Ferromagnets are a special category of magnetic materials with their properties being derived by the ferromagnetic exchange coupling among the paramagnetic centres. They are candidates for interesting potential applications, such as spintronics, magnetic coolers, etc.
Co1 and Co4 are six-coordinate with an octahedral coordination geometry. Co2, Co3 and Co5 are five coordinate displaying trigonal bipyramidal geometry (τ = Co2, 0.83; Co3, 0.76; Co5, 0.77) [86]. The Co 5 molecules in 3 are relatively in close proximity with the shortest metal···metal separation between the adjacent Co 5 units being 7.937 Å (Co4···Co4). 3 is the first example of a Co compound bearing H 3 pdc in its neutral or anionic form, being also one of the highest nuclearity discrete metal complexes with this ligand.

Magnetism Studies
Direct current magnetic susceptibility measurements (DC) were carried out on samples of 1, 2 and 3 in the 2-300 K temperature range and under a field of 0.03 T (1.28 MHz), and they are plotted as χ M T vs. T plot. (Figure 4). The χ M T values at room temperature (2.30, 1.95 and 4.31 cm 3 mol −1 K for 1, 2 and 3, respectively) are very close to the theoretical spin-only value of 2.00, 1.875 and 5.625 cm 3 mol −1 K corresponding to two Ni II , one Co II and three Co II non-interacting cations, respectively. Complex 1 is a tetranuclar Ni II system, in which Ni2 and Ni2' present a square-planar geometry being diamagnetic, hence, magnetically, the molecule behaves as a dinuclear Ni II complex. The χMT vs. T curve for 1 shows a weak ferromagnetic interaction between the Ni II paramagnetic cations, reaching a value of 2.96 cm 3 mol −1 K at 9 K; then, there is an abrupt decay until it reaches to a minimum value of 1.64 cm 3 mol −1 K at 2 K, due to weak intermolecular interactions or anisotropy effects. The curve has been fitted using the Hamiltonian Ĥ = -2J (SNi1·S Ni1') + DSz + ΣμgeffHS yielding in the best fitting values of J = 1.79(4) cm −1 , Dion = 2.48(2) cm −1 and g = 2.20 using PHI software [87]. This is in perfect agreement with previously reported multiple decker Ni complexes based on Ni4 layers that exhibit similar magnetic exchange pathways [40,41].
Complex 2 is a pentanuclear Co compound with only one paramagnetic Co II being present in the molecule; thus, it behaves as a mononuclear compound. The χMT decay when lowering temperature is due to the axial zero field splitting of the Co II cation evaluated as Dion = 47.7(7) cm −1 and g = 2.085 (2). The relatively high value of D parameter agrees with the trigonal bipyramid distorted environment. Alternate-current (AC) measurements were performed in the 1-1488 Hz frequency range, and weak tails were observed (Figure 4, right) in the presence of a magnetic field of 0.2 T, which is indicative of the single-ion magnetism behaviour for 2. However, this AC response is too weak to extract any relaxation parameter.
Complex 3 possesses an uncommon magnetic core: usually, μ3-OR cobalt trimers present a defective-cubane topology, whereas 3 is a μ3-O tricobalt complex where the oxo donor and the three cobalt atoms are co-planar, defining a triangular arrangement of cations with the three Co-O-Co bond angles being very similar (~120°). A CCDC search does not reveal any similar structure for which the magnetic study has been performed, and it makes complex 3 a quite interesting study case for the spin-frustrated Co II triangle. The χMT curve for 3 reveals an antiferromagnetic coupling between the three Co II cations; it continuously decreases with decreasing temperature until it reaches the value of 0.200 cm 3 mol −1 K at 2 K. The used Hamiltonian was Ĥ = −2J (SCo2·SCo3+ SCo3·SCo5+ SCo2·SCo5) + DSz+ΣμgeffHS, yielding in the fitting parameters J = −14.18(1) cm −1 , g = 2.075(3) and D = 10.55(4) cm −1 .
Complex 2 is a pentanuclear Co compound with only one paramagnetic Co II being present in the molecule; thus, it behaves as a mononuclear compound. The χ M T decay when lowering temperature is due to the axial zero field splitting of the Co II cation evaluated as D ion = 47.7(7) cm −1 and g = 2.085 (2). The relatively high value of D parameter agrees with the trigonal bipyramid distorted environment. Alternate-current (AC) measurements were performed in the 1-1488 Hz frequency range, and weak tails were observed (Figure 4, right) in the presence of a magnetic field of 0.2 T, which is indicative of the single-ion magnetism behaviour for 2. However, this AC response is too weak to extract any relaxation parameter.
Complex 3 possesses an uncommon magnetic core: usually, µ 3 -OR cobalt trimers present a defective-cubane topology, whereas 3 is a µ 3 -O tricobalt complex where the oxo donor and the three cobalt atoms are co-planar, defining a triangular arrangement of cations with the three Co-O-Co bond angles being very similar (~120 • ). A CCDC search does not reveal any similar structure for which the magnetic study has been performed, and it makes complex 3 a quite interesting study case for the spin-frustrated Co II triangle. The χ M T curve for 3 reveals an antiferromagnetic coupling between the three Co II cations; it continuously decreases with decreasing temperature until it reaches the value of 0.200 cm 3 mol −1 K at 2 K. The used Hamiltonian wasĤ = −2J (S Co2 ·S Co3 + S Co3 ·S Co5 + S Co2 ·S Co5 ) + DS z +Σµg eff HS, yielding in the fitting parameters J = −14.18(1) cm −1 , g = 2.075(3) and D = 10.55(4) cm −1 .
Elemental analysis (C, H, N) was performed by the in-house facilities of the National University of Ireland Galway, School of Chemistry. Solvothermal synthesis was performed in a Binder oven. IR spectra (4000-400 cm −1 ) were recorded on a Perkin-Elmer Spectrum 400 FT-IR spectrometer. Powder X-ray diffraction was performed with an INEL X-ray diffractometer EQUINOX 0000 using Cu K a radiation (λ = 1.54178Å, 35 kV, 25 mA). Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) were carried out in open aluminium crucibles using an STA625 thermal analyzer (Rheometric Scientific, Piscataway, NJ, USA) with nitrogen being purged in ambient mode. Solid-state, variable-temperature and variable-field magnetic data were collected on single crystals of each sample using an MPMS5 Quantum Design magnetometer operating at 0.03 T in the 300-2.0 K range for the magnetic susceptibility and at 2.0 K in the 0-5 T range for the magnetization measurements. Diamagnetic corrections were applied to the observed susceptibilities using Pascal's constants. Method B: Method A was followed with the only difference being the omission of pyaoxH 2 from the reaction mixture. The sealed vial was placed in the oven at 110 • C, and after one day, crystalline purple needles of 2 were observed. Yield: 80%

Compound Synthesis
Method C: Method A was repeated but using CoCl 2 ·6H 2 O (0.023 g, 0.10 mmol) instead of Co(CH 3 COO) 2 ·4H 2 O. The vial was left in the oven at 110 • C, and after one day, crystalline purple needles of 2 were observed. Yield: 90%. The product in Methods B and C was identified as 2 by IR spectral comparison with material obtained in method A ( Figure S5) and unit cell determination.

Single-Crystal X-ray Crystallography
Single-crystal diffraction data for 1 were collected in an Oxford Diffraction Xcalibur CCD diffractometer using graphite-monochromatic Mo-Kα radiation (λ = 0.71073 Å) at room temperature. Single-crystal diffraction data for 2 and 3 were collected in an Oxford-Diffraction SuperNova diffractometer equipped with a CCD area detector and a graphite monochromator utilising Mo-Kα (for 2) and Cu-Kα radiation (for 3). The structures were solved using SHELXT and [90] embedded in the OSCAIL software [91]. The non-H atoms were treated anisotropically, whereas the hydrogen atoms were placed in calculated, ideal positions and refined as riding on their respective carbon atoms. Molecular graphics were produced with DIAMOND [92]. We note that crystal twining occurs in 2, which affects the quality of the structure. We carried out many experiments in order to grow single crystals of better quality; however, this has not been achieved. We collected three sets of data using two different diffractometers and used the best set to solve the structure.
Unit cell data and structure refinement details are listed in Table 1. The crystal structures have been deposited with the Cambridge Crystallographic Data Centre (CCDC 2180525-2180527), and they can be accessed, free of charge, by filling out the application form at https://www.ccdc.cam.ac.uk/structures/.

Conclusions
The combination of 2-pyridyl oximes (pyridine-2 amidoxime, H 2 pyaox; 2-methyl pyridyl ketoxime, Hmpko) with isonicotinic acid (HINA) and 3,5-pyrazole dicarboxylic acid (H 3 pdc) provided access to three new Ni II and Co II/III metal complexes. Among them, [Ni 4 (INA) 2 (pyaox) 2 (pyaoxH) 2 (DMF) 2 (1) is a planar centrosymmetric tetranuclear compound that can be characterised as a pseudo-polymer due to the strong hydrogen bonding interactions between the neighbouring discrete units. It is the first reported metal compound that bears both pyaoxH 2 and HINA in their neutral or anionic form. 2 is a pentanuclear Co III 4 Co II complex with a bowtie topology, while 3 is a Co III 2 Co II 3 complex with a trigonal bipyramidal metal topology. The three Co II cations in 3 are co-planar with the bridging µ 3 -OH-ion providing a unique spin frustration model for triangular Co II 3 systems. The magnetic properties of 1-3 have been studied and revealed that there are ferromagnetic interactions between the two paramagnetic Ni II ions in 1 (J = 1.79(4) cm −1 ), which is in accordance with previously reported multiple decker Ni 12 and Ni 16 compounds based on Ni 4 repeating units [40,41]. 2 displays weak AC signals in the presence of a magnetic field, while in the case of 3, the dominant exchange interactions between the metal ions are antiferromagnetic. 3 is a unique example of a OH-centred triangular Co II 3 system in which all the Co atoms and the OHare co-planar; hence, it is an interesting case study for spin frustration.
Although the isolation of a compound that contains both a 2-pyridyl oxime and isonicotinic acid (HINA) has been successfully achieved in the case of 1, this has not been the case for 3,5-pyrazole dicarboxylic acid (H 3 pdc). Further studies that include the optimisation of the reaction conditions that will favour the presence of both ligands in the same metal complex, e.g., full deprotonation of ligands, nature of metal ion, etc., are currently in progress and will be reported in the near future.
Author Contributions: F.D. performed the synthesis, crystallisation and preliminary characterisation of compounds 1 and 3. C.G.E. and S.P.P. performed the synthesis, crystallisation and preliminary characterisation of compound 2. C.O. and P.M. collected crystallographic data, solved and refined the crystal structure of 1. A.K., E.M. and A.T. collected crystallographic data, solved and refined the crystal structures of 2 and 3. J.M. and E.C.-V. performed the magnetic measurements, interpreted the results and wrote the relevant part of the paper. C.P. coordinated the research and wrote the paper based on the reports of her collaborators. All authors have read and agreed to the published version of the manuscript.

Conflicts of Interest:
The authors declare no conflict of interest.
Sample Availability: Not available.