Session: 12-06-03: Modeling of Nano- and Micro-Scale Phase Change Processes

Paper Number: 168015

Numerical Modeling of Nanostructure Deformation During the Phase Change Process of Sublimation Drying 

Wet processes such as etching and cleaning can result in the presence of moisture and solvents on the semi-conductors, leading to electrical short circuits, non-uniform coating, corrosion, and contamination. To remove these residues from semi-conductors, drying processes are employed. The drying process of semiconductor devices is associated with nano-scale pattern collapse, particularly as production has been scaling down in recent years. Traditionally, nano-scale collapse has been attributed to capillary forces created by liquid residues within the nanostructure. Sublimation drying has been proposed to bypass this challenge, wherein the liquid material is first frozen and then sublimed to prevent liquid residues from becoming trapped in the nanostructures. However, it has been observed that nano-scale collapse can still occur during the freezing process of sublimation drying, a phenomenon that is not yet fully understood in the literature. The interaction of the liquid to solid phase change of the sublimating agent, thermal expansion, and the nanostructure stability is essential to refine the sublimation drying process. Hence, experimental and numerical studies should be conducted to enlighten our understanding of this phenomenon.

The objective of this research is to model nano-scale collapse during the freezing step of the sublimation drying process. To achieve this goal, COMSOL Multiphysics software is used to numerically model the process. Pillar-like nanostructures are considered, with the sublimating agent assumed to be trapped between the pillars. Different sublimating agents, including water and cyclohexane, are considered. The model focuses on the collapse of one cell, which contains two pillars and the liquid in between. This cell may be located either in the middle of the nanostructure or at the edges.

The freezing process is modeled using stepwise functions for the liquid properties. Specifically, the thermophysical properties are defined such that they adopt solid-phase values when the temperature is below the freezing point and liquid-phase values when the temperature is above the freezing point. This technique reduces computational load while maintaining model accuracy. The COMSOL model incorporates heat transfer, solid mechanics, and thermal expansion physics. The elastic properties of the silicon nanostructure and the sublimating agent are included to capture the impact of thermal expansion on nano-scale deformation. The results present the displacement of the nanostructure for both side and middle cells. Furthermore, the variations of the stress and temperature during the process are studied. Experimental results are used to validate the numerical model. The model facilitates the design and implementation of the sublimation drying process for semiconductor manufacturing.

Presenting Author: Minghan Xu University of Toronto

Presenting Author Biography: Minghan Xu is an Assistant Professor in the Department of Civil and Mineral Engineering at the University of Toronto. His research focuses on advancing our fundamental understanding of phase change heat transfer, particularly on freezing and melting processes. He also works on the development of renewable energy solutions in cold and remote regions, including phase change materials for energy storage, spray freezing for mine heating, and artificial ground freezing for permafrost protection. Minghan holds a B.Eng. (Honours) and Ph.D. from McGill University, followed by a Banting Postdoctoral Fellowship at the University of British Columbia.

Authors:

Mohammaderfan Mohit McGill University
Minghan Xu University of Toronto
Yosuke Hanawa SCREEN Holdings Co. LTD.
Jianliang Zhang SCREEN Holdings Co., Ltd.
Yuta Sasaki SCREEN Holdings Co., Ltd.
Koichi Sawada SCREEN Holdings Co., Ltd.
Junichi Yoshida SCREEN Holdings Co., Ltd.
Atsushi Sakuma Faculty of Fiber Science and Engineering, Kyoto Institute of Technology
Agus P. Sasmito McGill University