Computational Diagnostics for Detecting Phase Transitions During Nanoindentation
- PDF / 3,624,494 Bytes
- 6 Pages / 420.48 x 639 pts Page_size
- 80 Downloads / 174 Views
COMPUTATIONAL DIAGNOSTICS FOR DETECTING PHASE TRANSITIONS DURING NANOINDENTATION
Susanne M. Lee*, Carol G. Hoover*, Jeffrey S. Kallman*, William G. Hoover*t, Anthony J. De Groot*, and Frederick Wootent * Lawrence Livermore National Laboratory, P. 0. Box 808, Livermore, California 94551. t Department of Applied Science, UC Davis/Livermore, Livermore, California 94551.
ABSTRACT We study nanoindentation of silicon using nonequilibrium molecular dynamics simulations with up to a million particles. Both crystalline and amorphous silicon samples are considered. We use computational diffraction patterns as a diagnostic tool for detecting phase transitions resulting from structural changes. Simulations of crystalline samples show a transition to the amorphous phase in a region a few atomic layers thick surrounding the lateral faces of the indentor, as has been suggested by experimental results. Our simulation results provide estimates for the yield strength (nanohardness) of silicon for a range of temperatures.
INTRODUCTION Nanometer-machining techniques, using the diamond turning machine at Livermore, produce highly-polished surfaces which are flat to an accuracy of three atomic layers. The production rate is approximately one square meter per year. This slow rate, however, is not economically feasible for large production runs. The surface quality of the samples produced can be measured with scanning tunneling and atomic force microscopy. Sub-surface damage, such as internal cracks, is not measurable. Our work aims to develop parameter guidelines for nanometer-machining experiments and to study surface and bulk characteristics of ceramics and metals. Experiments show that brittle materials, such as silicon and glass, will flow plastically, rather than crack and fracture, for sufficiently small specimen sizes. Nanoindentation, the simplest experiment for measuring material hardness, is a valuable tool for quantifying ductile and brittle behavior and transitions between solid phases of silicon. The experimental samples are a micron in size and the indentation tool tip diameter is typically about 30 nanometers. In this work we simulate the nanoindentation of amorphous and crystalline silicon with nonequilibrium molecular dynamics. We study phase transitions with computer-simulated x-ray diffraction patterns to detect structural modifications. The simulated diffraction patterns augment the visualization of the sample in real space. Measured yield strengths provide a quantitative comparison with experiments. The speed and storage capacity of massively-parallel computers make possible atomistic simulations of tens of millions of particles in three dimensions. This provides the opportunity to study the ductile or brittle behavior as a function of the size of the silicon sample. The results reported here are for sample sizes up to a million particles.
COMPUTATION Nonequilibrium molecular dynamics simulates systems which exchange both heat and work with their surroundings. We perform the indentation under isothermal condition
Data Loading...
