what

INTRODUCTION TO MOLECULAR DYNAMICS FOR SOLIDS

This page is partly extracted from the carbon movies made by the Computational Physics Group. Versions of these in mpg format can be downloaded: [carbon-1.mpg][carbon-2.mpg][carbon-3.mpg][carbon-4.mpg][carbon-5.mpg][C-Hdiff.mpg].

There are also some new mpgs you may look at: here , on my group's Publications page and the movie that forms part of the advertising video for the PRACE, the European Supercomputer consortium.

This topic can be a little circular - in modelling solids one must use more than one type of algorithm, and it is a bit hard to separate their descriptions. Lots of topics like boundary conditions, interaction potentials and visualization are common for any atomistic modelling process. Molecular dynamics is common to other systems, liquids, gases, molecules, aeroplanes etc.

A very basic understanding at a macrcoscopic level is here.

Lets continue with general introduction (note that the continuation links within the slide belong to other presentations, please use back button to follow this part: [ Background ] [ ctd ]
[ Computational Condensed Matter ] [ Statistical Mechanics ]
Go for it: [What we do ] [ ctd ] [How we do it ] [ Details ]

Lets summarize a bit: Molecular Dynamics is simply solving Newton's equations of motion (F=ma) for atoms and molecules. This requires:

  • CALCULATIONS OF FORCES ( POTENTIALS) - - - from first principles and/or from experimental data. There are a lot of good sites listed here with more information on this.

    (For carbon modeling in the Computational Physics group we used the potentials of Tersoff [Phys.Rev.Lett. 61 (1988) 2879] and Brenner [Phys.Rev.B 42 (1990) 9458])

  • METHODS FOR INTEGRATING EQUATIONS OF MOTION - - - fast, converging algorithms and enough computer time and memory. These are of the type discussed in lecture 4.

    (The Computational Physics group usually uses the leap-frog, Gear predictor-corrector or verlet algorithms. See Guy's project that compared several algorithms.)

  • TECHNIQUES FOR VISUALIZATION OF RESULTS - - - 3D visualization and animation.

    (The computational Physics group does interactive visualization in OpenGL/mesa and C, using our own AViz package.)

  • EXPERIMENTAL DATA FOR COMPARISON - - -

    for example, the Computational Physics group's carbon modeling was motivated by the desire to explain certain ion implantation results of Rafi Kalish and data on CVD diamond films of Alon Hoffman. We also relate to experimental measurements of heat capacity, thermal conductivity, and phonon spectra of diamond and study amorphous diamond (Anastasia Sorkin). Other experimental/computational projects include melting due to vacancies and impurities (Amit Kanigel, Slava Sorkin with Emil Polturak) and aluminium on alumina (Adham Hasibon and Geri Wagner with Wayne Kaplan).

  • Projects using MD include:
    1. Guy Halouia
    2. Yoel Koenka
    3. Pavel Bavli
    4. Alexander Sanek
    5. Yaniv Gershon
    6. Fouad Atrash
    7. Anastasia Sorkin
    8. Slava Sorkin
    9. Polina Pine
    10. Tali Mutat
    11. Ofer Filiba
  • A good link for good programs for MD is Elad Tadmor book_resources.
  • SIMULATED ANNEALING

    The molecular dynamics approach is complemented by methods to find equilibrium configurations for the same atomic systems. In Tali's project these are compared. These methods have been discussed in two papers aimed at undergraduates, one to help understand the formation of crystalline and amorphous states. (The full article can be viewed and was prepared for a special theme issue of the American Journal of Physics.) The other article discusses interactive visualization for animated versions of simulated annealing routines and the program used there can be downloaded. This program uses PGPLOT, and can be compiled with pgplotcl as usual. Download and try it! What temperature schedule do you need to get a nice crystal. If you are working on aluf or a phclass or phelafelnew that should work. For phelafel only, there is a precompiled version screen3_phelafel that you do chmod +x screen3_phelafel then run with screen3_phelafel . You might need to run the pgplot path commands too. Ask me if you need them.

    Back to week 6 page.