Course Announcement for Spring 2000
Computer Simulation in Materials Physics
PH482: Computers in Physics-- Theory
to be offered as PH4395 under semester conversion
In recent years there has been an ever increasing effort devoted to modeling and solving
complex physical problems using computation-intensive methods of simulation (Monte Carlo and
molecular dynamics) and ab initio quantum calculations. Several new journals such as Journal of
Computer-Aided Materials Design, Computational Materials Science, and Modeling and Simulation
in Materials Science and Engineering attest to the growing importance of computation in materials
research. Along with the rapid development of simulation methods, there is an increasing number
of "black box" computer codes becoming available to investigators (computational specialists or
not) for application to problems of their own particular interest. Such application must be done
with a great deal of care. Unfortunately, details of simulation methods employed often are only
incompletely described, giving the impression that they are trivial or that they are justified where
the simulations produce the anticipated results. As pointed out by Binder [1986], quite a lot of
poor quality simulation work appears, perhaps because of the lack of awareness of the many
difficulties that can occur.
This course is targeted at advanced undergraduate and graduate students in physics,
metallurgical and materials engineering and sciences, chemistry, geology, or other materials-related
disciplines. The course will provide students with (i) a basic understanding of the advantages,
methodologies, assumptions, approximations, and potential pitfalls of computer simulation, and
(ii) some practical experience in addressing these topics using the context of computer simulation in
the study of the physics of materials.
Lectures, simulation exercises, and team projects will revolve around the following simulation
activities:
- Simple Monte Carlo simulation
- Data analysis (going from microscopic to macroscopic)
- Simple molecular dynamics simulation
- Error analysis (averages, fluctuations, correlations, autocorrelations, confidence)
- Advanced molecular dynamics
- Assessing the quality of published simulation work
Students will gain experience in and an appreciation for:
- what it takes to do careful simulation work
- comparing the use of pre-written codes versus developing and using one's own
simulation codes
- advantages and limitations of atomistic computer simulation methods in materials
physics
- confidence in assessing the quality of published simulation work
- working in a team to solve research problems
Instructor: John A. Jaszczak (jaszczak@mtu.edu), Fisher 124A, 487-2255
Text: Computer Simulation of Liquids by M.P. Allen & D.J. Tildesley.
Oxford, 1989.
Text Supplement: Thermal and Statistical Physics Simulations (CUPS) by Gould, Spornick & Tobochnik. Wiley, 1995.
Prerequisites: PH480 or permission of the instructor.
Familiarity with FORTRAN will be very helpful.
Please contact the instructor if you are interested in taking this course.
Back to Jaszczak's home page
Hits since 4 Nov. 1998:
