MAGNETISM IN A NUMBER OF METAL ORGANIC FRAMEWORKS (MOFs) WITH 1D AND 3D CHARACTERISTICS: AN EXPERIMENTAL AND ANALYTICAL STUDY

Abstract

Metal Organic Frameworks (MOFs) exhibit many excellent physical properties including magnetic properties for potential applications in devices. More importantly for the subject of this thesis, MOFs are ideal for the realization of low dimensional magnetism because of the large selection of ligands connecting magnetic centers in making the framework. The materials studied in this thesis include ten magnetic MOFs of the form M(L1)(L2) [M = Cu, Ni, Co, Fe, Mn; L1 = NDC, bpdc, BDC, BODC, N3; L2 = DMF, H2O, TED, bpy]. Polycrystalline powder samples as well as single crystal samples were synthesized. Their crystal structures were determined, and their magnetic and thermodynamic properties were measured and analyzed. Eight of these materials were characterized as 1D magnets and two as 3D magnets. In the 1D case it is found that above Tm [the temperature at which the magnetic susceptibility χ(T) has a peak] the magnetic behavior of MOFs (S ≥ 1) can be well described with the Classical Fisher Model (CFM). Near and below TC the spins take a more definite orientation than allowed for in the CFM and hence the Ising Model (IM) was used for fitting. Both CFM and IM yield fairly consistent intrachain couplings (J) when applied in their appropriate temperature region. To estimate the interchain exchange (J′), the susceptibility for a magnetic chain in the mean field of neighboring chains is used. In all cases, as expected, the ratio of J to J′ was less than 10%. The special case of Cu(N3)2bpy (S = ½) was analyzed with the spin ½ IM. Although the specific heat data (Ctotal) for most of the 1D MOFs showed no clear phase transition, a low temperature fit to the electron-phonon specific heats yielded apparent heavy fermion-like &gamma values on the order of several hundred mJ/mol K2. The lattice specific heat (C lattice) was estimated using a Debye-Einstein hybrid model. Subtracting Clattice from Ctotal, magnetic specific heat (CM) with a broad peak characteristic of low dimensional magnetism was obtained. The peak in CM was at temperature near that expected from χ(T) fits. The J values obtained from the magnetic specific heat fits were in good agreement with those obtained from χ(T) fits. Once the magnetic specific heat was accounted for, γtakes values in the expected range of few mJ/mol K2. For 3D MOFs [Mn(N3)2bpy and Fe(N3)2bpy], the existence of long range canted antiferromagnetic ordering was observed in both magnetic and specific heat measurements with phase transitions at 38 K and 20 K in the case of Mn(N3)2bpy and Fe(N3)2bpy, respectively. These transition temperatures are considered fairly high for molecular based materials. In both Mn(N3)2bpy and Fe(N3)2bpy, the χ(T) data fit well to the Heisenberg model for a diamond-type network. The transition can clearly be seen with an abrupt increase in the magnetization below TC and a shift to a higher temperature in the specific heat when measured under an applied magnetic field. The systematic approach in this work led to the successful estimate of C lattice resulting in meaningful fitting of χ(T) and Cmagnetic to the appropriate theoretical models in magnetism. It also led the discovery of ferrimagnets or canted antiferromagnets M(N3)2bpy with large coercivity and rather high transition temperature. The results of this study have been published in three articles in the Journal of Applied physics, and two manuscripts are under preparation for submission [1-5].