Etching of polysilicon in inductively coupled Cl2 and HBr discharges. IV. Calculation of feature charging in profile evolution

• Published in: Journal of vacuum science & technology. B, Microelectronics and nanometer structures processing, measurement and phenomena Microelectronics and nanometer structures processing, measurement and phenomena, 2002-05, Vol.20 (3), p.1084-1095
• Main Authors: Mahorowala, Arpan P., Sawin, Herbert H.
• Format: Article
• Language: English
• ISSN: 1071-1023; 1520-8567
• DOI: 10.1116/1.1481869

The ability to simulate feature charging was added to the Monte Carlo profile evolution simulator described in the companion article of this series [A. P. Mahorowala and H. H. Sawin, J. Vac. Sci. Technol. B 20, 1077 (2002)]. Two electrical assumptions were compared: (1) a perfect insulating feature and (2) a resistive feature. The potential profiles for the entire feature were generated and the ion and electron fluxes were computed along the feature surface. The resistive feature representation enabled the incorporation of bulk conduction and surface leakage that is not possible when using the insulator approximation. For highly resistive surfaces, the solutions for the two assumptions were shown to converge. With the insulator approximation, the potential profiles were calculated by determining the accumulated charge on the feature surface and solving Poisson’s equation over the entire simulation domain. Calculation of the potential profiles with the resistive feature approximation involved the determination of the steady-state current to the feature surface and solution of the current continuity in a resistive feature. Both solutions required the solution of Laplace’s equation with differing boundary conditions. The resistive feature approximation was used to study the role surface and bulk conductivities have on the potential profiles. It was shown for submicron features that the conductivities of most bulk materials could be approximated as perfectly insulating or conducting; however, surface leakage and passivation films could lead to circumstances where realistic conductivities are needed to obtain the correct solution. The charging of features was shown to scale inversely with the feature size. Therefore, charging is more significant in larger features. The materials properties of most materials lead to the conclusion that for current paths on the order of 0.25 μm, bulk materials can be treated as either perfect insulators of conductors. However, surface leakage or ultraviolet radiation may also make bulk oxide significantly conductive. Therefore, modeling it as a resistive material is required. Finally, the deposition of thin insulating layers on conductive materials could easily lead to feature charging.