Enlarged Lattice Kinetic Monte Carlo Simulation for Epitaxial Growth
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for simulating epitaxial growth in semiconductor fabrication, such as Lattice Kinetic Monte Carlo (LKMC) models, are computationally slow and fail to accurately predict growth shapes due to sensitivity to orientation and smoothing of corners, limiting their ability to model complex microelectronic devices effectively.
Innovation Solution
The approach involves representing an enlarged crystal lattice structure with a lattice enlargement factor N>1, assigning one simulated particle to each node, and scaling simulation parameters accordingly, which significantly increases simulation speed by a factor of N^3 without compromising accuracy, allowing for more realistic and comprehensive layouts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional LKMC models are used to simulate epitaxial growth, then accuracy in modeling atomic-scale processes is maintained, but simulation speed is computationally slow
Solution Approach 1:
The simulation domain is segmented into multiple processing zones or regions, allowing different computational methods or parameters to be applied to different areas. This enables parallel processing and reduces the overall computation time while maintaining accuracy in critical regions.
Solution Approach 2:
The patent introduces a scaled lattice representation where the simulation operates in an enlarged coordinate system. By scaling the lattice constant and adjusting particle densities accordingly, the simulation can cover larger physical distances with fewer computational steps, effectively adding a dimensional transformation that accelerates computation.
2Measurement precision
If conventional LKMC models simulate epitaxial growth with high resolution, then surface details are captured accurately, but computational resources and time are excessively consumed
Solution Approach 1:
Different regions of the simulation domain are assigned different levels of resolution or computational detail. Areas of interest or regions with higher activity receive finer resolution, while less critical areas use coarser representation. This local adaptation maintains necessary precision where needed while reducing overall computational burden.
Solution Approach 2:
The patent systematically changes key simulation parameters including lattice constant scaling, particle density adjustment, and event frequency modification. By coordinating these parameter changes, the simulation achieves equivalent physical accuracy with reduced computational complexity, effectively trading parameter space for computational efficiency.
3Productivity
If the lattice constant is enlarged by a factor N>1 to speed up simulation, then simulation speed increases by N^3, but surface resolution is reduced
Solution Approach 1:
The patent compensates for the resolution loss from lattice enlargement by systematically adjusting other simulation parameters. The particle density is scaled, event frequencies are modified, and boundary conditions are adapted to maintain the physical accuracy of the simulation despite the coarser lattice representation.
Solution Approach 2:
By transforming to a scaled lattice coordinate system, the patent effectively changes the dimensional scaling of the simulation. This dimensional transformation allows the simulation to cover larger physical spaces with fewer lattice points, and the parameter adjustments ensure that physical quantities remain accurate in this transformed space.
Data Source
AI summary
A method for simulating an epitaxial process in a body having a crystal lattice structure. Roughly described, an enlarged version of the crystal lattice structure is formed, having a lattice constant increased by a lattice enlargement factor N>1. The subject fabrication process is simulated by a Lattice Kinetic Monte Carlo algorithm in which various factors have been scaled in accordance with N. The simulation speed increases by a factor around N3, without significantly degrading the accuracy of the resulting simulated structure. The simulated epitaxial process can later be performed on a physical crystalline body.