Previous Page  |  Table of Contents  |  Next Page

Charles E. Campbell, Associate Fellow and E. Dan Dahlberg, Co-Principal Investigator

Simulations of Micromagnetic Structure and Quantum Fluid Systems

This quantum many-body physics research program is concerned with the microscopic basis for the structure and dynamics of a wide range of many-body systems. All of the work deals with many degrees of freedom, and most of it is ab initio in the sense that it deals with microscopically accurate representations of the systems of interest, as opposed to model systems (though many qualitative features deduced from model systems are incorporated). Usually, the research projects, and more generally, most projects in the many-body physics of strongly correlated systems, must make use of large scale computations and simulations in order to be relevant to real physical systems, and thus, the power of present supercomputers is often the chief limiting factor for the research. This program has largely focused on the structure and dynamics of strongly correlated quantum and classical fluids. However, during the past year, most efforts have been on simulations of the magnetic microstructure of magnetic materials such as micron scale permalloy films and nano-scale nickel particles.

99/191

These researchers have developed and successfully applied Metropolis Monte Carlo simulations to produce a thorough account of the hysteretic behavior of the permalloy thin films, including especially the temperature and anisotropy dependence of magnetic structure. Relatedly, dynamical Glauber-Metropolis Monte Carlo simulations have been used to study the approach to metastability of the magnetization of the permalloy thin films from certain important initial configurations. This work is being extended to include experimentally accessible film sizes and different geometries. Studies of the magnetic structure on a smaller size scale are well underway to obtain more detailed information of the structure of domain walls and their evolution during the technologically very important process of magnetic reversal. Of particular importance, both technologically and from a fundamental physics perspective, are the mechanisms for the reversal of macroscopic or mesoscopic magnetic moments, whether spontaneous (bad for archival purposes) or driven (good for writing information).

Micromagnetic simulations of nickel particles ranging in size from 50 to 900 nm, and 100 nm thick Ni films have also been performed, with very exciting results. This is being done in collaboration with Professor E. Dan Dahlberg's nano-monopole magnetic force microscopy (nmMFM) experimental program at the MMC on the same physical systems. The simulations are not only in excellent agreement with the nmMFM measurements, they go even further to reveal the full magnetic structure of the nickel particles, whereas nmMFM gives direct information only about the magnetic fields near the surface of these particles. This joint simulational/experimental effort is an outstanding example of the value of such collaborations when the distance scale of the experimental methods overlaps the distance scale possible in simulations.

The simulations on the Ni particles were done using a micromagnetic code which solves Landau Lifshitz Gilbert equation (LLG) for the magnetic microstructure, which is a damped version the classical magnetic equations for precession. This is apparently a very different approach from the Metropolis Monte Carlo simulation method, and is frequently the preferred approach because it is considered to be more physical in its description of the system's approach to metastable states from a given initial state. However, it does not, in its most convenient and available form, include the effects of finite temperature (i.e., it is a zero temperature approach), which on the other hand is the strength of the Metropolis Monte Carlo simulation. A Langevin adaptation of the LLG equations may be used to include finite temperatures, while a Glauber adaptation of the Metropolis Monte Carlo method may be used to make the diffusive dynamics of the Monte Carlo evolution more realistic.

These researchers are also continuing their other large-scale computing efforts on several projects on quantum fluids. The longest standing project is the simulation of liquid 4He in constrained and reduced geometries and the response of these systems to atomic and neutron scattering; this has recently and is currently being extended to the recently discovered rare-gas Bose Einstein condensed systems.

In collaboration with Eckhard Krotscheck at the University of Linz and Artur Polls at the University of Barcelona, the quantum statistical mechanics of superfluid mixtures of helium three is being researched in liquid helium four (3He in 4He), with particular attention to the influence of the fermion component 3He on the superfluid phase transition. Because of the presence of 3He, there is the traditional and vexing "albebraic sign problem" of fermion systems in "exact" quantum Monte Carlo simulations. While the small number of 3He atoms in a 1,000 particle system (i.e., 60 fermions) makes the sign problem potentially more manageable, the resultant statistics associated with such a small number of particles are inadequate to represent a macroscopic system in the vicinity of a superfluid transition. Thus, this work is focusing on the semi-analytic Feenberg-Jastrow Euler-Lagrange approach in order to produce a good enough importance sampling function-in this case, the quantum many-body density matrix-to enable simulations to be applied to considerably larger systems-hopefully 10,000 or more particles.

The second quantum fluids project requiring large-scale computational efforts is the development of a method to do quantum molecular dynamics at finite temperatures applicable to strongly correlated quantum fluids such as the helium liquids and nuclear matter. In the first phase of this research, a time averaging of a pure state density matrix is being implemented with specified energy expectation value.

This information is available in alternative formats upon request by individuals with disabilities. Please send email to alt-format@msi.umn.edu or call 612-624-0528.

HOME | QUESTIONS | FEEDBACK
Employment | Events | Links | People | Programs | Publications | Support | Welcome