AbstractsChemistry

This thesis presents a theoretical and computational approach to the accurate description of magnetic exchange interactions in a variety of complex systems. These include two main families of compounds. The first family is formed by inorganic coordination complexes, presenting localized magnetic centres and well-defined crystal structures. The second family consists of purely organic, π−conjugated odd alternant neutral polyradicals, which display a much larger structural flexibility and greater delocalization of the unpaired electrons over the π system. The thesis has two main parts. The first one refers to the adopted strategy for the accurate extraction of magnetic exchange interactions. The systems used to investigate this issue are coordination compounds of increasing complexity, including heterobinuclear and homotrinuclear complexes, for which experimental crystal structures and magnetic data are available. The adopted strategy is based on the mapping approach, which relies on a one to one correspondence between the non-relativistic, time-independent exact Hamiltonian and two model spin Hamiltonians, the so-called HDVV and Ising. Ultimately, the mapping approach consists on describing both the energy and the wave function of the pure spin states by means of broken symmetry functions, using a spin projector to establish a univocal relation. In this thesis, a detailed analysis of the mapping approach has enabled establishing an alternative and accurate manner for extracting magnetic interactions in complex systems. By pointing out two main deficiencies that make the standard mapping approach proposed by Noodleman not appropriate to certain systems, and following previous work in our group, we propose an alternative approach. This is based on a direct use of the energy of the broken symmetry solutions which are mapped into the energy expectation values of the corresponding broken symmetry solutions of the HDVV Hamiltonian. This strategy relies on the one-to-one correspondence of the diagonal terms of the HDVV and Ising matrix representations. This proposal has been applied to the three-centre three-electron problem, and is further verified by comparison of the calculated coupling constants with the available experimental data and by means of effective Hamiltonian theory. The exchange coupling constant values obtained with this approach are consistent. Additionally, effective Hamiltonian theory offers the possibility to check whether the system can be described as a Heisenberg system. The second part of the thesis deals with purely organic π−conjugated neutral radicals interacting through-bond, and applies the strategy developed for inorganic molecules to extract the magnetic coupling constants between the unpaired electrons in these compounds. On the basis of chemical stability, different building blocks are investigated. Then, by considering different coupling schemes (strategies to assemble them leading to different dimensionalities) and the role of structural flexibility, the main goal of the investigation…