Synthesis, Crystal Structure and Magnetic Properties of 1D Chain Complexes Based on Azo Carboxylate Oxime Ligand

: Two carboxylate ‐ bridged one ‐ dimensional chain complexes, {[Mn II (MeOH) 2 ][Fe III (L) 2 ] 2 } n (1) and {[Mn II (DMF) 2 ][Mn III (L) 2 ] 2 ∙ DMF} n (2) [H 2 L = ((2 ‐ carboxyphenyl)azo) ‐ benzaldoxime], containing a low ‐ spin [Fe III (L) 2 ] − or [Mn III (L) 2 ] − unit were synthesized. Magnetic measurements show that the adjacent high ‐ spin Mn II and low ‐ spin M III ions display weak antiferromagnetic coupling via the syn–anti carboxyl bridges, with J = − 0.066(2) cm − 1 for complex 1 and J = − 0.274(2) cm − 1 for complex 2.


Introduction
Low-spin (LS) M III (M = Fe or Mn) ions have been widely used for assembling lowdimensional molecular magnets [1][2][3][4][5][6], owing to the magnetic anisotropy and spin-orbit coupling for LS M III (M = Fe or Mn) ions, as well as zero-field splitting (ZFS) for LS Mn III [6]. Moreover, LS M III -based hetero-bimetallic complexes would exhibit a high-spin ground state due to ferromagnetic coupling between LS M III and M' II (e.g., Ni, Cu) according to the rule of strict orbital orthogonality (t2g vs. eg) [7]. According to the crystal-field theory for coordination compounds, LS M III should be surrounded by strong-field ligands. Nevertheless, stable LS M III -containing building blocks are still limited, most of which are cyanide complexes [2][3][4][5][8][9][10][11][12][13]. The search for suitable strong-field ligands can help to obtain new LS M III -containing complexes.
Carboxylate ligands possessing a variety of bridging modes, such as syn-syn, synanti, and anti-anti, play an important role in magnetic coupling propagation [14,15]. Azo carboxylate oxime ligands have emerged and contributed to the formation of stable LS M III (L)2 building units [16][17][18]. However, the valence of metal ions is also strongly related to the self-assembly processes of the complexes. According to our previous report, under one-pot reaction, Mn IV (L)2 was formed in which aromatic azo oxime ligand H2L acts as an (L•) 3-radical to stabilize Mn IV [19]. In this paper, a complex-as-ligand method is used to obtain carboxylate-bridged M III Mn II complexes. The precursor LS M(III) complexes are Et4N[M III (L)2] (M = Fe or Mn) [16,17] (2). Powder X-ray diffraction (PXRD) patterns of complexes 1 and 2 are in good agreement with those simulated by Mercury software (Figure S1), indicating that the obtained crystals have high phase purity.

Crystal Structures
The crystallographic parameters for 1-2 are listed in Table 1. Complex 1 crystallizes in the tetragonal space group P4(2)/n and complex 2 crystallizes in the triclinic space group P-1. The bond lengths for the [M(L)2] − are displayed in Table 2 (Table 3), all above 2 Å, typical of the +2 oxidation state of HS Mn ions, which is further confirmed by the bond valence sum (BVS) calculation (Supplementary Materials, Table S1). The coordination geometry of Mn II ion in complexes 1 and 2 calculated by the SHAPE [20] software approaches Oh, with the smallest deviation value of 1.459 and 0.085, respectively (Table S2). The presence of crystallographic disorder in the coordinating methanol molecules makes the calculated deviation value for complex 1 large, and therefore the deviation value of 1.459 should be treated with care. The difference between these two complexes is that the two [  (Table S2). The calculation results (Table S2) also indicate that the distortion toward trigonal prism (D3h) has been found for all metal ions with the second smallest deviation values.      Figure S5. No intermolecular π-π stacking is present in complexes 1 and 2.

Magnetic Properties
The temperature-dependent magnetic susceptibilities of complexes 1 and 2 were measured under 1000 Oe external field in the range of 2-300 K. The experimental magnetic susceptibilities were corrected for the diamagnetism of the constituent atoms (Pascal's tables). As shown in Figure 3, complexes 1 and 2 have similar magnetic susceptibility curves. The χmT values of each complex remain constant above 30 K. When the temperature was lower than 30 K, the χmT values decrease rapidly with the decrease of temperature due to the intramolecular antiferromagnetic coupling between adjacent metal ions. The data obey the Curie-Weiss law with the negative Weiss constant of θ = −2.71 K and −4.05 K, respectively, which further proves that the existence of antiferromagnetic interactions in complexes 1 and 2. The room temperature χmT per Mn II M III 2 (M = Fe or Mn) values were 5.317 cm 3 K mol −1 for complex 1 and 6.588 cm 3 K mol −1 for complex 2, are slightly higher than the theoretical value of 5.125 cm 3 K mol −1 [two LS Fe III ions (S = 1/2) and one HS Mn II ion (S = 5/2)] for 1 and 6.375 cm 3 K mol −1 [two LS Mn III ions (S = 1) and one HS Mn II ion (S = 5/2)] for 2. As shown in Figure 4, the field dependence (0-50 kOe) of the magnetization shows that with the increase of the field, the magnetization increases gradually and reaches the maximum values of 6.088 Nβ and 5.749 Nβ at 50 kOe for complexes 1 and 2, respectively. The experimental curves for complexes 1 and 2 lie below the Brillouin curves corresponding to non-interacting LS-SFe/SMn and SMn spins with g = 2.0, indicating the existence of overall antiferromagnetic coupling. The Brillouin function, BS, is / coth coth and the magnetization M equals NgβSBS [21].

Discussion
In complex 1, the low-spin octahedral Fe III ion has an electronic configuration of t2g 5 , and there is an unpaired electron on the degenerate π-orbital of dxy, dxz, and dyz. The highspin octahedral Mn II ion with the t2g 3 eg 2 configuration has three unpaired electrons on the degenerate t2g π-orbitals, as well as two unpaired electrons on the degenerate σ-orbitals dx 2 −y 2 and dz 2 . A similar situation occurs in the t2g 4 -t2g 3 eg 2 between low-spin Mn III ion and high-spin Mn II ion in complex 2. According to the theory of the strict orthogonality of magnetic orbitals, the configuration of the above two sets of magnetic orbitals enables both ferromagnetic and antiferromagnetic coupling in complexes 1 and 2, and usually, the latter contribution is dominant. Thus, together with the syn-anti bridged carboxyl group that tends to transfer antiferromagnetic coupling [22], overall antiferromagnetic interaction was observed in the two complexes.
To study the strength of magnetic coupling between metal ions, it is necessary to fit the temperature-dependence magnetic susceptibility. For the [M III ]2-Mn II chain system in complexes 1 and 2, the magnetic susceptibility data (5-300 K) can be fitted by the Fisher model for uniform 1D chains with ∧ ∑ [23] (Equation (1)). A rough approach similar to that previously used for 2D and quasi-2D complexes [24][25][26] was used on the basis of the crystal data, i.e., the 1D chain can be treated as alternating uniform M2Mn trimers ( Figure 5) with the identical intra-trimeric and intrachain exchange constants (Jt = J) on the basis of the Hamiltonian  The best-fit parameters are g = 2.017(8) and J = −0.066(2) cm −1 for complex 1, and g = 2.053(4), J = −0.274(2) for complex 2. The calculated curves based on the above parameters are well consistent with the experimental data, and the negative J values are in accordance with the prediction that the carboxyl group in syn-anti bridging mode tends to transmit antiferromagnetic coupling [22]. The absolute J values are very small, precluding any possibility of a single-chain magnet for the present two complexes. The measurements on ac magnetic susceptibility show that under zero external dc field, the imaginary part of the magnetic susceptibility of complexes 1 and 2 has no signals and maintains zero.

Physical Measurements
The C, H, and N elemental analyses were performed on a Cario Erballo elemental analyzer. IR spectra were recorded on a WQF-510A Fourier transform infrared spectrometer using KBr pellets. Magnetic susceptibility measurements were measured by a Quantum Design MPMS-XL5 SQUID magnetometer. Cyclic voltammetry measurements were tested on a CHI660E electrochemical workstation, using a platinum plate as the working electrode, platinum wire as the counter electrode, Ag/AgCl electrode (Sat. KCl) as the reference electrode, and n-Bu4NClO4 (0.1 M) as support electrolyte in acetonitrile.

X-Ray Crystallography
The single-crystal X-ray diffraction measurements were tested on a Rigaku R-Axis RAPID IP diffractometer by using Mo Kα radiation (λ = 0.71073 Å). The structures were solved by direct methods using the SHELXTL-97 program package and refined with fullmatrix least squares on F 2 .

Conclusions
We used a 'complex as ligand' method to create two one-dimensional chain complexes 1 and 2 based on low-spin [M III (L)2] − units (M = Mn or Fe). The syn-anti carboxyl bridges transmit weak antiferromagnetic coupling between adjacent Mn II -M III ions. Complex 1 possesses a novel deflecting arrangement of [Fe(L)2] − units and is a rare example of carboxylate-bridged hetero-metallic complexes [18,27]. Further work involves the construction of carboxylate-or oxime-bridged bimetallic low-dimensional magnets based on similar low-spin azo carboxylate oxime ligands.
Supplementary Materials: The following are available online at www.mdpi.com/2312-7481/7/7/105/s1, Figure S1: PXRD patterns for complexes 1 and 2 in the range of 5-50 degrees, Figure  S2: (top) Side view of complex 1 along c axis; (bottom) View of the 1D skeleton of complex 1, Figure  S3: Side view of complex 2 along a axis; (bottom) View of the 1D skeleton of complex 2, Figure S4: Cyclic voltammograms (scan rate 50 mV s-1) of 10-3 M acetonitrile solutions of complexes 1 and 2 at 298 K, Figure S5: The intrachain hydrogen bonding interaction between the disordered methanol oxygen and the carboxylate oxygen atoms of L 2− in complex 1.