Reactive Force-Field Molecular Dynamics Study of SiGe Thin Film Growth in Plasma Enhanced Chemical Vapor Deposition Processes

Abstract

SiGe thin films, such as hydrogenated amorphous silicon germanium (a-SiGe:H) and hydrogenated microcrystalline silicon germanium (μc-SiGe:H), are formed by plasma enhanced chemical vapor deposition (PECVD) methods from SiH4, GeH4, and H2. Thin film properties such as surface morphology, hydrogen content, and crystal structure are greatly affected by process parameters such as a substrate temperature, gas pressure, and gas composition. For example, SiH3 and GeH3 are dominant species when the SiGe thin film with low defect is deposited under low power density conditions. However, the structural defects are formed by SiH and SiH2 that are highly reactive species under high power density conditions. Although the density and flux of these highly reactive species are much lower than those of the dominant species, the electrical properties of the device are greatly affected by a few structural defects. Therefore, to understand the influence of the composition of each precursor on the thin film properties is important. The purpose of the present study is to clarify the influence of the composition of multiple precursors such as SiH3, GeH3, and H under different substrate temperatures on the surface morphology, hydrogen content, and crystal structure.We performed reactive force-field molecular dynamics (ReaxFF MD) simulations. An existing parameter set was modified to match the dissociation energies in gaseous species because bond formation and breaking are particularly important in this study. The substrate was Si (100)– (2×1), and the size of the simulation box was 30.72 Å× 26.88 Å× 45.00 Å. A periodic boundary condition was applied in the x and y directions, and a fixed boundary condition was applied in the z direction. The first bottom layer of the substrate was fixed to prevent vibrations of the substrate up and down. The second and third layers were controlled constantly at desired temperature by the Berendsen thermostat. The temperature was not controlled on the fourth and subsequent layers to prevent direct influences of changing the velocity of atom due to the temperature control on the surface reaction. The substrate temperature was set at 800 K, 1000 K, 1300 K, and 1600 K. 2000 SiH3 molecules with the velocity corresponding to 1300 K in the direction towards the substrate were deposited one by one onto the substrate surface every 7.5 ps. The initial position of the gaseous species was set at 10 Å above the top of substrate Si atoms, and the in-plane (x-y) position were set randomly. The Velocity-Verlet algorithm was used for time integrations corresponding to 0.25 fs time step. A pre-thermal annealing for 7.5 ps and a post-thermal annealing for 60 ps were performed before and after the deposition to fully equilibrate the substrate and deposited thin films.Figure shows front views of the final configurations and results of analysis when the substrate temperature is 800 K, 1000 K, 1300 K, and 1600 K. The yellow, blue, and red represent substrate Si atoms, added Si atoms, and added H atoms, respectively. Gray shown in the figure are gas-solid surfaces. The surface construction algorithm was used to calculate a surface area that is closely related to the surface morphology. The hydrogen content was calculated using the select type algorithm. The structure identification algorithm for the crystal structure was used to characterize the structure of simulated films. These analyses were carried out by the Open visualization tool (OVITO). The surface area decreases as the substrate temperature increases, indicating that the formation of smoother surface at high temperature. The diffusivity of a SiH3 molecule that reached to the surface increases as the substrate temperature increases, and the atomic scale valleys are filled. This tendency is in good agreement with relevant studies. The hydrogen content was calculated by dividing the number of added H atoms by the total number of atoms existing in the system. The hydrogen content decreases as the substrate temperature increases because a hydrogen desorption is promoted as the substrate temperature increases. Figure 1