Theoretical Physics Projects

Heisenberg models of single magnetic molecules

A successful program for the investigation of magnetic molecules requires detailed comparison between the results of experiment and those of theoretical modeling. The Heisenberg model of exchange interactions between localized spins provides an excellent theoretical platform for calculating the magnetic properties of many single magnetic molecules. The basic parameters of a Heisenberg model Hamiltonian can be determined by comparing the calculated magnetic properties with the results of thermodynamic measurements (e.g., magnetization versus temperature and magnetic field). More refined characteristics of the magnetic molecule (e.g., the spectrum of magnetic energy levels) that are determined from the theoretical model can be compared with the results of inelastic neutron scattering and magnetic resonance studies. Besides equilibrium properties, it is also possible to address time-dependent features at low temperatures such as dynamical hysteresis and magnetization steps in response to pulsed magnetic fields with relatively high sweep rates. Collectively, by an iterative dialog between experiment and theory it is becoming possible to achieve a comprehensive understanding of many distinct magnetic molecules. These efforts provide valuable input to help guide future chemical synthesis efforts as well as deeper insight into the broader issues of the magnetic properties of general, finite arrays of interacting spins.

A variety of powerful theoretical tools are in use within the Ames Laboratory magnetic molecules collaboration. For many magnetic molecules the theoretical analysis can proceed via diagonalization of the Heisenberg Hamiltonian matrix followed by the application of standard methods of statistical mechanics. However, this approach is usually impractical if the dimensionality, D = (2s + 1)N, of the underlying Hilbert space exceeds about 107, a situation that can occur even for relatively small numbers, N, of identical spins s. For these difficult situations we are finding that thermal equilibrium properties can in many cases be successfully determined by employing the quantum Monte Carlo method. Important progress can also be achieved for magnetic molecules when s = 5/2 or larger, by using computer simulational methods of classical spin dynamics and the classical Monte Carlo method to determine the dependence of physical quantities on external variables, including both thermodynamic functions and equilibrium time correlation functions. The latter are of great importance for the analysis of neutron scattering and magnetic resonance measurements. Moreover, by comparing results obtained by invoking both classical and quantum Monte Carlo methods we are able to establish guidelines for when the classical methods, which are relatively easy to implement, can provide useful predictions for finite quantum Heisenberg spin systems.

People and Collaborators

Ames Laboratory:
Marshall Luban (PI) email: luban@ameslab.gov

Main external collaborators:
Hiroyuki Nojiri, Dept. of Physics, Tohoku University, Sendai, Japan
Heinz-Jürgen Schmidt, Fachbereich Physik, Universität Osnabrück, Germany
Jürgen Schnack, Fakultät für Physik, Universität Bielefeld, Germany
Christian Schröder, Fachhochschule Bielefeld, Germany
Richard Winpenny, Dept. of Chemistry, University of Manchester, UK
Larry Engelhardt, Dept. of Physics, Francis Marion University, Florence, SC


Electronic structure and exchange interactions in single magnetic molecules

The slow dynamical processes, which take place in magnetic molecules at moderate magnetic fields, can be reasonably well understood by ignoring the intra-molecular structure, based on the model of a single rigid spin representing the total spin state of the molecule (for instance, single spin S = 10 for description of {Mn12-acetate}, single spin S = 1/2 for {V15As6}, etc.). Such an approach has been successfully applied, for example, to the description of magnetization tunneling in {Mn12} and {Fe8}.
However, deeper understanding of the properties of magnetic molecules calls for a detailed investigation of the intra-molecular properties of different magnetic molecules. On one hand, such an investigation is needed for fundamental scientific reasons, in order to clarify the connection between the structural, electronic, and magnetic properties of the magnetic molecules under study, and to explain how different properties affect each other (for an example of such a study concerning different polyoxovanadate molecules, see below). On the other hand, a detailed and systematic study of the intra-molecular properties, based on the electronic structure calculation, can reveal important information about how the chemical composition and the atomic structure of the molecule affect the intra-molecular exchange interactions. This theoretical information can be very useful for chemists aiming at synthesizing the molecules with prescribed properties. Below we present such a study for different versions of {Mn12} molecules with different ligands.
Moreover, the recent attempts to use magnetic molecules for nanoelectronics applications require detailed theoretical information about the electronic band structure of the molecules. The computations of the band structure based on the local density approximation (LDA) and generalized gradient approximation (GGA) have clarified many aspects of the electronic and magnetic structure of magnetic molecules (such as magnetic moments of individual atoms), but reliable calculations of the energy gaps and magnetic superexchange interactions in the transition metal-oxide systems (which include such molecules as {Mn12}, {V15As6}, etc.) require an account of many-electron correlations originating due to Coulomb repulsion between the electrons. Without taking these correlations into account, the calculated electronic structure gap is noticeably underestimated, and the exchange interactions are significantly (up to a factor of three) overestimated. Therefore, it is important to study how the novel powerful computational approaches based on advanced many-body techniques, such as dynamic mean-field theory (DMFT), can be adapted for calculations of the electronic structure in such large and complex systems as magnetic molecules, comprising 100-200 atoms per unit cell.
Along with investigating the interactions between the transition metal atoms in the magnetic molecules, it is also important to understand the influence of other atoms present in the molecules, in particular, the role of nuclear spins and their impact on the spin tunneling in magnetic molecules. While much previous studies have been directed at an understanding of decoherence by the nuclear spins, many experimental results have not yet been understood. At the same time, detailed understanding of the impact of decoherence on the spin tunneling in nanomagnets, and in magnetic molecules in particular, is important for the ongoing effort aiming at exploring and exploiting the fundamental quantum phenomena for applications in coherent spintronics and in quantum information processing.

People and Collaborators

Ames Laboratory:
Bruce Harmon (PI) email: harmon@ameslab.gov
Viatcheslav Dobrovitski email:slava@ameslab.gov

Main external collaborators:
Mikhail I. Katsnelson, University of Nijmegen, Nijmegen, the Netherlands
Danil W. Boukhvalov, Institute of Metal Physics, Russian Academy of Sciences Urals Division, Ekaterinburg, Russia
Alexander I. Lichtenstein, Universitat Hamburg, Hamburg, Germany
Ernst Z. Kurmaev, Institute of Metal Physics, Russian Academy of Sciences, Urals Division, Ekaterinburg, Russia
Janice L. Musfeldt, University of Tennessee - Knoxville, Knoxville, TN
Naresh Dalal, Florida State University, Tallahassee, FL


last change: Mar 14, 2007