Overview of Theory and Simulation in the

1
Office of Basic Energy Sciences Office of
Science, U.S. Department of Energy
Overview of Theory and Simulation in the
Division of Materials Sciences and Engineering
Dale D. Koelling Program Manager Theoretical
Condensed Matter Physics
2
Director's Office Staff
Office of Basic Energy Sciences
Robert Astheimer F. Don Freeburn Stanley
Staten Fred Tathwell Margie Marrow Program
Analyst (Vacant)
Patricia Dehmer, Director Mary Jo Martin,
Administrative Specialist
Scientific User Facilities Division
Patricia Dehmer, Director (Acting) Linda
Cerrone, Program Support Specialist
Condensed Matter Phys and Materials
Chemistry X-Ray Neutron Scat.
X-ray Neutron Scattering Facilities
Materials and Engineering Physics
William Oosterhuis Melanie Becker, Prog. Asst.
Robert Gottschall Terry Jones, Prog. Asst.
Pedro MontanoVacant
Structure Composition of Materials
Catalysis and Chemical Transformation
Atomic, Molecular, and Optical Science
Plant Sciences
Experimental Condensed Matter Physics
Spallation Neutron Source (Construction)
James Tavares
Vacant Eric Rohlfing uDavid Ederer, ANL
Raul Miranda uJohn Gordon, LANL
Altaf (Tof) Carim
James Horwitz
Biochemistry and Biophysics
Jeffrey Hoy
Sharlene Weatherwax
Nanoscale Science Research Centers (Construction)
Mechanical Behavior of Materials Rad Effects
Separations and Analysis
Chemical Physics
Theoretical Condensed Matter Physics
Richard Hilderbrandt uFrank Tully, SNL
John Miller
Dale Koelling
Yok Chen
Kristin Bennett Altaf (Tof) Carim
Materials Chemistry Biomolecular Materials
Photochemistry Radiation Research
Physical Behavior of Materials
Linac Coherent Light Source (Construction)
Heavy Element Chemistry
Lester Morss Norman Edelstein, LBNL
Mary Gress
Dick Kelley Aravinda Kini
Harriet Kung
Jeffrey Hoy
Synthesis Processing Science
X-ray NeutronScattering
Chemical Energy and Chemical Engineering
Computational and Theoretical Chemistry
SNS, LCLS, and X-ray Neutron Scattering
Instrument MIEs Kristin Bennett
Jane Zhu uDarryl Sasaki
Paul Maupin
Richard Hilderbrandt
Helen Kerch
Geosciences Research
Experimental Program to Stimulate Competitive
Research (EPSCoR)
Engineering Research
Nicholas Woodward lDavid Lesmes, George
Washington U
l IPA u Detailee Detailee, 1/4 time, not
at HQ
Timothy Fitzsimmons
Matesh Varma
February 2004
3
Materials and Engineering Physics Dispersed
Theory
Structure and Composition of Materials
dynamic behavior of nanostructures greater
ability to treat inhomogeneous materials, esp.
disorder includes the effort on the constrained
local moment model for spin dynamics.
Mechanical Behavior of Materials and Radiation
Effects predict material behavior under exposure
conditions (irradiation, temperature, and
mechanical loading) that represent a significant
extrapolation beyond our existing knowledge
base. Physical Behavior of Materials coupling
of length scales from atomic to macroscopic
organic electronic materials --- charge and
energy transfer, electronic structure
calculation, exciton dynamics and transport, spin
dynamics. Synthesis and Processing Science
fundamental understanding of mechanisms and
processes to aid systematic design.
Engineering Physics multiplicity of scales
managing the explosion of data an INCITE award
occurred here.
4
Dynamics of Spins
The Heisenberg model, long used to describe the
magnetic behavior of materials, is a model
involving experimentally determined parameters.
Recently, a fundamental theory has been
formulated for which the Heisenberg model could
be considered an approximation. Within this
theory, the magnitude of the moments can change
as they are rotated as illustrated in the figure.
In addition to such new effects, the formalism
also enables one to calculate and interpret the
strengths of interaction. This will lead to
further insights how materials arrange themselves
to have magnetic moments and, equally important,
how they avoid them. A major step forward
towards understanding magnetic materials, this
theory is extends Density Functional Theory and
utilizes it in regimes where numerous aspects are
still not well understood. Consequently,
applying this theory will simultaneously enhance
our understanding of magnetic materials and of
the basic theory. It is also a non-trivial
computational effort! ?
Calculated state of 512 atoms from paramagnetic
(high temperature) bcc iron. The variation of
magnetic moment is given by the color scheme.
Implementation of the fundamental approach
requires intensive computing --- the code has won
the top prize for computational efficiency ---
but also the resolution of further subtle and
difficult fundamental theoretical issues.
5
INCITE Program

• The projects were selected under a new
  competitive program, entitled Innovative and
  Novel Computational Impact on Theory and
  Experiment (INCITE), announced last July by
  Energy Secretary Spencer Abraham.
• 52 proposals were submitted.
• Three awards amount to 10 percent of the total
  computing time available this year on NERSC's
  current IBM SP3.
• Fluid Turbulence and Mixing at High Reynolds
  Number," led by P.K. Yeung (Georgia Tech.), was
  awarded 1,200,000 processor hours in the area of
  forced isotropic turbulence. The principal
  investigators have NSF grants but their INCITE
  activities are a part of the Engineering Physics
  activity.

6
Condensed Matter Physics Materials Chemistry

• Theory is primarily concentrated in the
  Theoretical Condensed Matter Physics activity ---
  although it is not exclusively so. Multiple
  length and time scales Complex systems Many
  body effects Predictive Multidisciplinary
  Efforts!!! (CMSN 03-17 Solicitation)
  ???????
• Materials Chemistry treatments of
  nanostructured 2 3 dimensional materials with
  chemical accuracy interactions and transport
  phenomena at interfaces novel multiscale
  approaches for large complex systems that link
  spatial and temporal scales.
• While concepts are emphasized here, we are
  effectively using a lot of computing and can
  really benefit from further development!

7
Computational Materials Science Network

• The mission of the Computational Materials
  Science Network is to advance frontiers in
  computational materials science by assembling
  diverse sets of researchers committed to working
  together to solve relevant materials problems
  that require cooperation across organizational
  and disciplinary boundaries.
• http//www.phys.washington.edu/cmsn

8
Criteria for Cooperative Research Team Proposals

• Proposals should
• focus on critical scientific issue operationally
  modified
• have a clear path to relevance (i.e., real
  materials issues)
• be of the type best pursued through broad
  cooperative efforts, as opposed to those key
  problems best tackled by single investigator
  groups
• build on existing BES funded programs
• define some short-term deliverables combined with
  long-term objectives
• a strong synergism with experimental and
  industrial programs is highly encouraged

And so shall ye be judged!
9
Funded Collaborative Research Team Proposals
In FY04, the active teams are

• Excited State Electronic Structure and Response
  Functions (J. Rehr S. Louie)
• Fundamentals of Dirty Interfaces From Atoms to
  Alloy Microstructures (A. Karma A. Rollett)
• Predictive Capability for Strongly Correlated
  Systems (W. Pickett R. Scalletar)
• Microscructural Effects on the Mechanics of
  Materials (R. LeSar D. Wolf)
• Magnetic Materials Bridging Basic and Applied
  Science (B. N. Harmon G.M. Stocks)

10
Theory, Modeling and Simulation in Nanosciences

• Notice 03-17 (February 6, 2003) yielded 34
  applications of which 4 were funded.
• 6.0 M Joint BES and Office of Advanced
  Scientific Computing Research.
• Solicitation encouraged the formation of teams to
  make significant advances.
• Evaluated jointly --- only after decision were
  they binned into divisions
• Computational Nanophotonics
• ANL, Northwestern, Georgia State, Central
  Michigan, U. of Illinois at Chicago
• Predicting the Electronic Properties of 3D,
  Million-Atom Semiconductor Nanostructure
  Architectures
• NREL, LBNL, ORNL, U. of Tennessee
• Scalable Methods of Electronic Excitations and
  Optical Responses of Nanostructures
• LBNL, UCLA, U. of Minnesota and NYU.
• Integrated Multiscale Modeling of Molecular
  Computing Devices
• Vanderbilt, ORNL, NC State, Princeton, U. of
  Colorado, and U. of Tennessee

11
Computing is Important !

• Last year, over 3.35x106 processor hours were
  used on the IBM SP3 at the National Energy
  Research Supercomputer Center and 1.6x106 IBM
  SP3 SP4 processor hours helping evaluate
  computers at Oak Ridge National Laboratory. (The
  two machines involved will become a part of the
  production system this year.)
• At 45 repositories (account for PI and
  coworkers), MATSCI is the program with the
  largest population of users.
• Requests for time ran roughly 3 times the
  resources available this year BEFORE NEW, LARGE
  SPECIAL REQUESTS.