A Study on the Tip-Based Nanomachining of Silicon Wafer Using Atomic Force Microscopy (AFM)

In this work, the principles of nanoindentation and nanoscratching processes are applied in the computational study of two relevant material behaviors of single crystalline silicon wafer, namely phase transformation and hardness. The large-scale atomic/molecular massively parallel simulator (LAMMPS) is used in this study to perform three-dimensional (3D) molecular dynamics simulations (MDS) of these processes on a silicon substrate and a spherical diamond tool tip with a radius of 10Ǻ is considered. In the 3D MDS simulations, the workpiece substrate is a single crystal silicon <100> wafer that contains 2,207,698 Si atoms and a single diamond indenter is modeled as a rigid body. These Si atoms are initially arranged in a diamond cubic lattice structure with a lattice constant of 5.43 Å at room temperature (293 K). The dimensions of the substrate are 500 Å x 500 Å x 350 Å. The outermost layers of the Si workpiece substrate are fixed with the exception of the top surface. These fixed layers are 10.86 Å thick. Within these fixed layers are a 16.29 Å layer of thermostat atoms which are kept at a constant temperature of 293 K. The diamond nanoparticle is a sphere of 10 Å diameter and consists of carbon atoms arranged in diamond cubic structure. The initial gap between the diamond nanoparticle and the substrate is 10 Å. The force-controlled approach is employed for this study. Essentially, investigation is conducted under the premise of three parameters: interatomic potential for the Si-C interaction (i.e. indenter to substrate interaction), applied force on indenter and operating temperature (room temperature and several higher operating temperatures) of the silicon substrate. The effects of these parameters on the phase transformation, depth of indentation and length of scratch are evaluated. Verlet -Velocity algorithm is used to compute the velocity and positions of the atoms. Since we desired to maintain consistency in volume, energy, and the number of particles, the constant-energy ensemble (NVE), also known as microcanonical ensemble is applied in the simulations. Both the Si-Si and C-C interactions are computed using the Tersoff potential throughout the simulations while the Si-C interactions are computed with three different potentials: Morse, Leonard Jones, and SiC Tersoff. The MDS results are visualized and analyzed using the Open Visualization Tool (OVITO). It is found that these three parameters (interatomic potential for the Si-C interaction, applied force on indenter, and operating temperature of the silicon substrate) have substantial effects on the behavior (phase transformation and hardness) of the silicon substrate.