A Study on Calculating Epitaxial Layer Thickness Based on a Multi-Parameter Collaborative Optimization Model

Huiming Chen

doi:10.54097/br8kgr37

Authors

Huiming Chen

DOI:

https://doi.org/10.54097/br8kgr37

Keywords:

Drude Model, Fresnel’s Equation, Monte Carlo Simulation, Differential Evolution.

Abstract

Semiconductor materials have progressively become core components in chip manufacturing owing to their unique physical and chemical properties. Taking silicon carbide as an example, this paper establishes a computational model for parameters such as refractive index and epitaxial layer thickness during beam reflection. This model is based on the principles of reflection and transmission at the epitaxial layer-substrate interface, utilising beam reflection principles alongside the Drude model and Fresnel equations. Under specified conditions, this study performed global fitting on the dataset using the least squares method. The refractive index of the epitaxial layer was calculated employing differential evolution principles. Finally, Monte Carlo simulations were conducted for sensitivity analysis to assess the reliability of the fitting results. This paper comprehensively considers multiple characteristics of light beam propagation, providing an innovative solution for determining epitaxial layer thickness via multi-beam interferometry in subsequent practical applications.

References

[1]Ando Y, Makisako R, Takahashi H, et al. Dependence of electrical characteristics on epitaxial layer structure of AlGaN/GaN HEMTs fabricated on freestanding GaN substrates[J]. IEEE Transactions on Electron Devices, 2021, 69(1): 88-95.

[2]Park J, Cho Y J, Chegal W, et al. A review of thin-film thickness measurements using optical methods[J]. International Journal of Precision Engineering and Manufacturing, 2024, 25(8): 1725-1737.

[3]Sun J, Li Z, Zhang H, et al. An improved method for measuring epi-wafer thickness based on the infrared interference principle: Addressing interference quality and multiple interferences in double-layer structures[J]. Results in Physics, 2023, 54: 107146.

[4]Yu T, Ma T. Solving Strategy for Silicon Carbide Epitaxial Layer Thickness Based on Nonlinear Least Squares Method[J]. Frontiers in Engineering, 2025, 1(1): 45-48.

[5]Mirschel G, Daikos O, Scherzer T. In-line monitoring of the thickness distribution of adhesive layers in black textile laminates by hyperspectral imaging[J]. Computers & Chemical Engineering, 2019, 124: 317-325.