Crystal and molecular structures and magnetic properties of tetrameric copper(II) complexes with 3-hydroxy-5-hydroxymethyl-4-(4′-hydroxy-4′-phenyl-2′-azabut-1′-en-1′-yl)-2-methylpyridine (H2L3), [Cu4L34]·9CH3OH and 3-hydroxy-5-hydroxymethyl-4-(4′-hydroxy-3′-methyl-4′-phenyl-2′azabut-1′-en-1′-yl)-2-methylpyridine (H2L1), [Cu4-L14]· 8CH3CH2OH : two complexes with ferromagnetic ground states

Abstract

The crystal and molecular structures of the title compounds [Cu₄L³₄]·9CH₃OH (3) and [Cu₄L¹₄]·8CH₃CH₂OH (4) have been determined from single-crystal X-ray diffraction data and refined to final R values of 0.062 (R′= 0.052) using 2 191 independent reflections for (3) and 0.062 (R′= 0.055) using 2 458 independent reflections for (4). Both tetrameric complexes crystallize in the tetragonal space group P4₂/n with two molecules in an unit cell with dimensions a= 18.193(4) and c= 12.615(4)Å for (3) and a= 17.991(4) and c= 14.296(4)Å for (4). Both molecules exhibit the highest possible point symmetry . While compound (3) is comparable with its 4′-(3″,4″-dichlorophenyl) derivative [Cu₄L²₄]·9CH₃OH (2) which crystallizes in a monoclinic lattice having an infinite network of hydrogen bonds, (4) is isomorphous to [Cu₄L¹₄]·8CH₃OH (1). Moreover these two forms of structures are distinguishable by the considerably larger Cu–O distance between two pseudo-monomeric units within the pseudo-dimeric unit in (1) and (4) compared with (2) and (3). The magnetism of the two complexes can be explained on the basis of the isotropic Heisenberg–Dirac–van Vleck model. Compound (4) is very similar in its magnetic behaviour [g= 2.16(2), J₁₂= 0 (fixed), J₁₃= 17.4(2.0) cm^–1, and θ=–1.8(1.0) K] to (1). In contrast to (2), compound (3) definitely has a ferromagnetic ground state [g= 2.085(20), J₁₂=–7.1(2.0), and J₁₃= 28.5(2.0) cm^–1]. The principal difficulties of fitting theoretical magnetic susceptibility equations to experimental ones in ferromagnetically coupled systems is discussed. The discrepancy of the magnetic exchange constants between (3) and (2) which has really C₁ symmetry is discussed in order to test by which model of higher symmetry the magnetism of this compound can be optimally described. It can be shown that a C_2v model leads to more plausible exchange constants for (2) with regard to the geometrical sizes of the highly distorted molecule.