Multiscale CFD Modeling of an Area-Selective Atomic Layer Deposition Process: Application to Discrete Feeding Reactor Configuration Design

Matthew Tom¹, Henrik Wang¹, Feiyang Ou¹, Gerassimos Orkoulas³, and Panagiotis D. Christofides^1,2

1. Department of Chemical and Biomolecular Engineering, University of California, Los Angeles, CA
2. Department of Electrical and Computer Engineering, University of California, Los Angeles, CA
3. Department of Chemical Engineering, Widener University, Chester, PA

In response to the downscaling of semiconductor feature sizes and the increasing density of transistors occupying a semiconducting wafer, fabrication methods have become more stringent in design specifications. In particular, wafers must be appropriately fabricated to enable nanopatterning and facilitate the bottom-up procedure through self-alignment for transistor stacking [1]. Various manufacturing procedures employ thin-layer deposition processes, which deposit thin films in a self-limiting manner. However, conventional deposition processes such as atomic layer deposition (ALD) lack selective deposition, which disrupts nanopatterning and increases the need for postprocessing steps including atomic layer etching (ALE). In response to aforementioned obstacle, area-selective atomic layer deposition (ASALD) processes have emerged to promote self-limiting and chemoselective deposition through the integration of surface protection layers that interfere with reagent surface adsorption on non-growth area surfaces [2]. Additionally, ultra, thin-film uniformity has been an emerging challenge for large-scale deposi­tion processes, which prompted the development of discrete feeding reactors that better control the behavior of the fluid dynamics through semi-instantaneous adsorbate and purge gas pulses that are confined in discrete re­gions [3,4]. While laboratory paradigms of the processes have been established, optimization and development of ASALD and discrete feeding reactors have not been studied due to the difficulties of relating the atomistic-mesoscopic surface kinetics behavior of the reactions to the macroscopic fluid dynamics of reagents within the reactor. In silico multiscale modeling serves to better characterize the two phases by conjoining a kinetic Monte Carlo (kMC) simulation that describes the surface kinetics behavior with computational fluid dynamics (CFD) that numerically evaluates the spatiotemporal behavior of the adsorbates and byproducts within the reactor. Addi­tionally, multiscale modeling can generate large data sets without requiring substantial computational power and capital for raw materials and experimental design.

This work will establish an in silico multiscale modeling framework for simulating area-selective atomic layer dep­osition of a SiO₂/Al₂O₃ wafer through a discrete feeding reactor by conjoining a mesoscopic simulation using the kMC algorithm and a macroscopic simulation using CFD. First, a discrete feeding reactor configuration (which in­cludes a wafer substrate) will be constructed through a mesh discretization procedure that is appropriately opti­mized for numerical accuracy and computational efficiency. Next, the CFD simulation will be customized to the ASALD process environment by defining various boundary, operating, and numerical method conditions as well as user-defined functions (UDFs), which will be employed to carry out the kMC simulation through each nodal element on the wafer surface. Lastly, the CFD simulation will be conducted for various operating and reactor design conditions (e.g., reagent flow rates and composition and number of discrete feeding ports) to establish optimal reactor configurations and operating conditions to enable process scale-up.

References:

• Li, M.-Y., Su, S.-K., Wong, H.-S. P., & Li, L.-J., 2019. “How 2D semiconductors could extend Moore’s law,” Nature, 567, 169-170.
• Mackus, A. J. M., Merkx, M. J. M., & Kessels, W. M. M., 2019. “From the bottom-up: toward area-selective atomic layer deposition with high selectivity,” Chemistry of Materials, 31, 2-12.
• Kim, D. H., Park, J. C., Park, J., Cho, D.-Y., Kim, W.-H., Shong, B., Ahn, J.-H., & Park, T. J., 2021. “Wafer-scale growth of a MoS₂ monolayer via one cycle of atomic layer deposition: an adsorbate control method,” Chemistry of Materials, 33, 4099-4105.
• Han, J. W., Jin, H. S., Kim, Y. J., Heo, J. S., Kim, W.-H., Ahn, J.-H., Kim, J. H., & Park, T. J., 2022. “Advanced atomic layer deposition: ultrathin and continuous metal thin film growth and work function control using the discrete feeding method,” Nano Letters, 22, 4589-4595.