Computational analysis of crystal shapes as modified by surrounding microstructure and effect on crystal size distribution

COMP 218

Roddy V. Amenta, amentarv@jmu.edu, Department of Geology and Environmental Science, James Madison University, Harrisonburg, VA 22807
Recent advances in crystallization theory have linked the crystal size distribution (CSD) to the formative crystallization kinetics opening up new possibilities for studying crystallization of ancient igneous rocks. However, the constraints of microstructure on crystal shapes, sizes and CSD need to be more fully explored with computer simulated crystallization, presently the only method possible. Thus we simulated the nucleation and growth of crystals with shape ratios of 1:3:5 using simple kinetic expressions that predicted the number of unit cells added to each crystal in each time step as space permitted, forming a microstructure consisting of several thousand interlocking crystals. Computer rendering of individual grains revealed shape irregularities caused by the microstructure such as embayments due to impingement of adjacent grains and host growth offshoots into adjacent grain boundaries. CSDs based on longest dimensions of crystals compared poorly with CSDs predicted from the kinetic models indicating that such lengths are poor indicators of crystal size. The direct measurement of 3-D grain sizes, in contrast to widely used 2-D measurements, is important for developing CSD work on real rocks using high resolution X-ray tomography. Work in progress is focused on image processing of computer generated 3-D shapes and determining their best fit ellipsoids as indicators of grain sizes.
 

Poster Session
6:00 PM-8:00 PM, Tuesday, August 18, 2009 Walter E. Washington Convention Center -- Ballroom A, Poster

Division of Computers in Chemistry

The 238th ACS National Meeting, Washington, DC, August 16-20, 2009