Significant progress in treating the most fundamental
few-body atomic collision problems has been enabled recently
by implementation of large-scale computational approaches
for solving the one- and two-electron, time-dependent,
multidimensional Schrodinger equation utilizing lattice
techniques. Results of one variant of this broad class of
methods will be illustrated through description of specific
examples involving inelastic transitions among bound states
and to the multi-center, electronic continuum in fundamental
ion-atom collisions. Advantages of the lattice approach stem
from its circumvention of various shortcomings of
perturbation theories, close coupling expansions, and other
approximation methods, and from the insight it allows one to
develop regarding the underlying quantum dynamics. The
principal goal of this line of development has been to
augment the basic toolkit available to describe fundamental,
atomic-scale, few-body dynamics utlizing direct numerical
approaches.
[S6.002] Recent developments in quantum Monte Carlo methods for electronic structure of atomic clusters
Lubos Mitas (Department of Physics, North Carlina State University, Raleigh)
Recent developments of quantum Monte Carlo (QMC) for electronic structure calculations of clusters, other nanomaterials and quantum systems will be reviewed. QMC methodology is based on a combination of analytical insights about properties of exact wavefunctions, explicit treatment of electron-electron correlation and robustness of computational stochastic techniques. In the course of QMC development for calculations of real materials, small and medium size clusters proved to be invaluable systems both for testing and for revealing unique insights into electron correlation effects in nanostructured materials. The method shows remarkable accuracy which will be demonstrated on calculations of magnetic states of transition metal atoms encapsulated in silicon cluster cages, optical excitations in quantum nanodots and molecules and on studies of reactions in biomolecular metallic centers. Indeed, in some cases QMC turned out to be the only feasible method to provide the necessary accuracy. I will also discuss current QMC developments in using correlated sampling techniques for efficient evaluation of energy differences, efforts to reach beyond the fixed-node approximation and on incorporating QMC methods into multi-scale simulation approaches.
In collaboration with P. Sen, L.K. Wagner, Z.M. Helms, M.
Bajdich, G. Drobny, and J.C. Grossman. Supported by NSF, ONR
and DARPA.
[S6.003] Simulation of Black Hole binaries: pushing computational methods to extreme regimes
Luis Lehner (Louisiana State University)
Binary black hole systems are fascinating. They involve the strongest gravitational fields, in extremely dynamical situations, and represent an ideal source of gravitational waves which could be observed by earth and space-based detectors.
The study of these systems require the ability of dealing
with the full Einstein equations without simplifying
assumptions. The complexity of the equations make numerical
simulations the only practical way to understand the system
in the most interesting regime. For these simulations to be
successful a number of issues, from analytical to
computational, must be addressed. In this talk, I will give
an overview of these issues and discuss their (partial)
answers and outlook for the future.
[S6.004] Atomic and Molecular Double Photoionization and Electron Impact Ionization from First-Principles Computation
C. William McCurdy (Lawrence Berkeley National Laboratory)
Among the last problems in atomic and molecular physics to
be “reduced to computation” – even in principle – have been
processes that lead to the separation of three charged
particles. Double photoionization and ionization by electron
impact are examples of breakup processes in which two
electrons leave behind an atomic or molecular ion creating a
signature of electron correlation in the cross sections. The
key barrier to solving these problems has been that the
formally correct asymptotic boundary conditions for solving
the Schrödinger equation for the breakup of Coulomb systems
have proved to be completely impractical for numerical
calculations. A reformulation involving “exterior complex
scaling” of the electronic coordinates and the application
of modern parallel computing techniques have finally
provided procedures for solving such problems from first
principles. The basic ideas of these new methods will be
discussed and examples of their application to essentially
exact calculations for several systems involving two
electrons will be demonstrated.
[S6.005] The Fall and Rise of Lattice QCD
G. Peter Lepage (Physics Department, Cornell University)
For almost 30 years precise numerical studies of nonperturbative QCD, formulated on a space-time lattice, have been stymied by our inability to include the effects of realistic quark vacuum polarization. This talk, which is aimed at a general physics audience, describes a series of theoretical breakthroughs that have recently led to the first few-percent accurate nonperturbative calculations in the history of QCD. The implications for particle physics, and for theoretical physics more generally will be discussed.