Large artificial anisotropic growth rate in on-lattice simulation of obliquely deposited nanostructures

Abstract

On-lattice particle simulation is one of the most common types of Monte Carlo simulations used in studying the dynamics of film growth. We report the observation of a large artificial anisotropic growth rate variation due to the fixed arrangement of particles in an on-lattice simulation of oblique angle deposition (OAD). This unexpectedly large anisotropy is not reported in previous literatures and substantially affects the simulation outcomes such as column angle and porosity, two of the most essential quantities in obliquely deposited nanostructures. The result of our finding is of interest to all on-lattice simulations in obliquely deposited films/nanostructures. PACS number(s): 68.55.J-, 05.10. Ln, 81.15.Aa

Avoiding Anisotropies in On‐Lattice Simulations of Ballistic Deposition

“…For both azimuthal angles, the simulated tilt angles and densities are smaller than predicted by the off‐lattice algorithm and observed in the deposition experiments. The observed tilt angles are in agreement with simulations by Tanto et al [ 19 ] The fact that the azimuthal angle also influences the simulated film parameters implies that all simulations containing a substrate rotation are affected by the grid effect as well. One may ask, why this effect is usually not seen in the many simulations that are used to model deposition processes (for example, refs.…”

“…Tanto et al already have shown that in on‐lattice simulations tilting the whole setup in the simulation cell along a polar angle leads to a change in the observed tilt angles (see Figure 2c,d), even though the angle between particle flux and substrate is held constant. [ 19 ] This contradicts the physical expectation of absolute spatial and angular anisotropy. A closer examination of the tilt angles and densities, obtained from the on‐lattice simulation, reveals more differences ( Figure 3 ).…”

“…This approach facilitates handling of large amounts of particles and complex substrate geometries, as all particle–particle and particle–substrate interactions become trivial. Even though on‐lattice simulations are used for decades, Tanto et al [ 19 ] recently found that the results of this kind of simulation depend strongly on the orientation of the substrate and the particle flux with respect to the axis of the used simulation grid. This violates the requirement of absolute space isotropy, and therefore all on‐lattice simulations that include oblique deposition, substrate movements (as rotation), or beam divergence are rendered questionable.…”

Time Invariant Surface Roughness Evolution during Atmospheric Pressure Thin Film Depositions

“…Figure 1 provides a visualization of several simulation processes. For further description of the simulation, refer to references 15 18 19 20 21 .…”

Growth of SiO2 and TiO2 thin films deposited by reactive magnetron sputtering and PECVD by the incorporation of non‐directional deposition fluxes

“…Monte Carlo (MC) models are widely used for the simulation of the growth of thin films 27–29, and the one we use in this paper has already been successfully applied in many cases 16, 17, 19. It considers the deposition of effective particles on a cubic three‐dimensional N L × N L × N H grid, each cell representing a spatial volume of Δ 3 , with periodic boundary conditions, whose cells may take the values 0 (empty cell), or 1 (a deposited particle occupies the cell).…”