Simulations shed light on key process in semiconductor device manufacturing
Prior to co-founding the Intel Corporation, Gordon Moore observed in 1965 that regular improvements in semiconductor manufacturing technology would yield corresponding reductions in transistor dimensions. Indeed, “Moore’s law” heralded the information age, facilitated by the dramatic cost reductions associated with these manufacturing advances. Today, state-of-the-art chips contain tens of billions of transistors.
Among the most important steps in semiconductor device manufacturing is removing material from surfaces with atomic-scale precision, following lithographic pattern transfer. One such procedure, atomic layer etching (ALE), separates different aspects of the etch process to allow better control of each part.
Filling a gap in research around this process, Vella et al. examined it through classical molecular dynamics (MD) simulations, a technique in which the time-dependent trajectory of a system of atoms is calculated.
“A key step before using MD to further understand ALE is to first validate the modeling technique against any available experimental data,” said co-author Joseph Vella. “We do this by using MD to model silicon ALE with exposure to chlorine gas and argon ions, then comparing simulations to [previous] experimental work. Our model matches experimentally reported ALE per cycle for a variety of process conditions.”
With a validated mode, simulations provide insight into mechanisms governing the silicon-chlorine-argon ALE process. Argon ion bombardment of silicon surfaces, for instance, induces a surprisingly robust mixing of atoms to a depth of about 1 nanometer.
“While the work demonstrates the ability of MD simulations to accurately predict experimental ALE data, its implications are not yet fully understood,” said co-author David Graves. “But it provides an important foundation for future studies.”
Source: “Molecular dynamics study of silicon atomic layer etching by chorine gas and argon ions,” by Joseph R. Vella, David Humbird, and David B. Graves, Journal of Vacuum Science & Technology B (2022). The article can be accessed at https://doi.org/10.1116/6.0001681 .
This paper is part of the Plasma Processing for Advanced Microelectronics Collection, learn more here .