Development of Ab Initio Methods for Lanthanide Complexes

We are developing ab initio methodologies dedicated to the determination of the electronic structure and magnetic properties of ground and low-lying excited states, i.e., the crystal field levels, in lanthanide(III) complexes, which are of interest in current research on Single-Molecule Magnets. Currently, the most popular and successful ab initio approach is the CASSCF/RASSI--SO method, consisting of the optimization of multiple complete active space self-consistent field (CASSCF) spin eigenfunctions, followed by full diagonalization of the spin--orbit coupling Hamiltonian in the basis of the optimized CASSCF spin states featuring spin-dependent orbitals. Based on two simple observations valid for Ln(III) complexes, namely: (i) CASSCF 4f atomic orbitals are expected to change very little when optimized for different multiconfigurational states belonging to the 4f-electronic configuration, (ii) due to strong spin--orbit coupling the total spin is not a good quantum number, we propose here an efficient ab initio strategy which completely avoids any multiconfigurational calculation, by optimizing a unique set of 4f-dominated open-shell molecular orbitals via a configuration-averaged SCF calculation, followed by a complete active space simultaneous diagonalization of both the electron--electron repulsion and the spin--orbit coupling interactions. Our method provides multiconfigurational ab initio wavefunctions, energies and magnetic properties for the crystal field states in Ln(III) complexes at the cost of a single SCF calculation and a small CI diagonalization. We are implementing the proposed methodology in the pilot code CERES (Computational Emulator of Rare Earths Systems), and we have published preliminary applications of the code to the study of a few Ln complexes displaying single-molecule magnet behaviour. The accuracy of these preliminary results is shown to be comparable to that of the computationally more expensive CASSCF/RASSI--SO method. Check out: PCCP, 2016 and arxiv for more information. Other projects aimed at simplifying the computationally demanding multiconfigurational ab initio description of the electronic structure of Ln(III) complexes are under scrutiny.

Spin Detection and Control in Molecular Nanomagnets at Metal Surfaces

Single-Molecule Magnets (SMMs) are molecules with many unpaired electrons, which at low temperatures undergo a superparamagnetic transition, leading to slow-magnetic relaxation and magnetic hysteresis behaviours. Thus, SMMs below the superparamagnetic transition temperature could be harnessed to store and process information encoded in the spin of a single-molecule. Most of the research on SMMs has focused until recent times on their spin dynamics in the crystal phase. However, in order to embed SMMs in electronic circuits and use them as molecular devices a crucial question concerns their behaviour in a non-crystalline environment. Recent progress in fabrication methods and experimental techniques has allowed researchers to fabricate and probe SMMs self-assembled as monolayers or sub-monolayers on magnetised metal surfaces, via e.g. X-ray Magnetic Circular Dichroism (XMCD), so that significant amounts of experimental data are becoming available. The theoretical and computational simulation of these complex systems, which involve both highly delocalised low-dimensional condensed matter components (the conducting ferromagnetic substrate), as well as well-localised strongly correlated molecular and atomic states (the SMMs), poses new challenges that cannot be overcome via black box traditional quantum chemistry approaches. In our group we are developing theoretical models parameterised by multiconfigurational ab initio and DFT calculations, in order to explore the spin dynamics of the SMM spin impurity interacting with a magnetised surface. In particular, one of the main focuses of our research consists in the study of sub-monolayers of the TbPc2 SMM, self-assembled on a ferromagnetic Ni(111) thin film. These systems are fabricated and experimentally probed at low-temperatures via XMCD and other techniques by our collaborators in Italy, the research group of Prof. Marco Affronte and Dr. Andrea Candini at the CNR-Nano Research Centre (University of Modena and Reggio Emilia). In particular, here at the University of Melbourne we are developing theoretical models to describe the formation of local magnetic moments on the Pc organic ligand units directly physisorbed onto the metal Ni(111) ferromagnetic surface. The formation of such localised magnetic moments appears to be a crucial step in establishing a spin-communication channel between the localised 4f electron magnetism of the Tb ion, and the Ni(111) magnetisation of the conducting magnetised substrate. Check out our recent publication Scientific Reports 6, 21740 (2016).

Theory of Chiral Discrimination in Paramagnetic NMR Spectroscopy

Despite its central role in biological processes and in a wide range of chemical reactions, chirality, the property of most molecules to be distinguishable from their mirror image (or enantiomer) remains a challenging property to detect and quantify. Inspired by recent work by Prof. A. D. Buckingham (J. Chem. Phys. 140, 011103 (2014)), we have recently developed a theory predicting that in a paramagnetic NMR experiment, normally blind to chirality, the two mirror images of a chiral paramagnetic molecule are subject to opposite electric forces (green arrows in the picture) triggering distinguishable rotational motions for the two enantiomers in solution. If the molecules have strong magnetic anisotropy (anisotropy axes are represented as red rods in the picture), they will be partially oriented along the NMR magnetic field (blue arrows in the picture), which makes the predicted chiral electric forces large enough to be detected at room temperature, offering a way to achieve direct chiral discrimination via NMR spectroscopy. Read more about our recent work on chiral discrimination in paramagnetic NMR here: Phys. Rev. Lett. 116, 163001 (2016) or check this: arxiv.

Magnetic Anisotropy in Single-Molecule Magnets: Theory and Simulation

Single-Molecule Magnets (SMMs) are nanometer-sized molecules containing transition metal and rare earth ions, which at low temperatures become superparamagnetic, displaying slow-magnetic relaxation and magnetic hysteresis behaviour in time-dependent magnetic fields. The unique spin dynamics of these systems is potentially useful for the development of molecule-sized electronic devices to store and process information. The ability of such molecules to function as efficient SMMs is determined by (i) the existence of a large (pseudo) spin in the ground state (ii) the existence of strong magnetic anisotropy characterising the low-energy spin states, defining a preferred direction within the molecule where the spin can be easily polarised (easy-axis). Mono and poly-nuclear lanthanide-based complexes offer very efficient ways to achieve these properties within a relatively small volume. However, much remains to be understood about which electronic structure properties of a given molecule can lead to an efficient SMM. In our group we develop and apply theoretical models and computational methods to unravel the relationship between electronic structure and spin dynamics of experimentally characterized SMMs. In the picture it is for instance illustrated a successful theoretical model based on classical electrostatics, which was developed by our group in collaboration with Dr. Nick Chilton and Prof. Richard Winpenny from the Molecular Magnetism group of the University of Manchester (UK). Our model is capable of predicting the direction of the magnetic easy axis in low-symmetry dysprosium complexes, see Nat. Commun. 4, 2551 (2013). The direction of the magnetic axis in complexes with little or no symmetry is a crucial piece of information for the design of efficient SMMs. Before this work, magnetic axes could only be determined by means of time consuming high-level ab initio calculations, providing little insight into structure-properties relationship. We managed to map the intricate ab initio quantum chemistry problem into a back-of-the-envelop classical electrostatic energy minimization procedure, which takes seconds to run, and provides a clear and chemically intuitive link between atomic charges on the ligands as determined by pencil-and-paper resonance Lewis structure arguments, and the direction of the magnetic axis in the complex. These results will prove useful to devise synthetic strategies to control the axis direction, and engineer efficient lanthanide-based single molecule magnets. We have other active collaborations aimed to the simulation and understanding of magnetic anisotropy in Ln-complexes, with the magnetochemistry groups of A/Prof Colette Boskovic at the University of Melbourne, Em/Prof Keith Murray at Monash University, and Dr. Richard Mole, Instrument Scientist at the Braggs Institute (ANSTO). Active projects include the study of the crystal field levels and the magnetic anisotropy of polyoxometallate complexes of lanthanide ions, probed via Inelastic Neutron Scattering experiments. Check out: Chem. Commun. 52, 2091-2094 (2016).