(388f) Modeling a Fluidized Bed Reactor for Particle Atomic Layer Deposition with CFD-DEM

Particle atomic layer deposition (ALD) is a vapor-phase coating technique that enables precise surface modification of particulate substrates. ALD coatings are uniquely uniform and controllable, which is desirable for many applications such as catalyst synthesis or cathode coating. Particle ALD is typically performed with a repeated sequence of self-limiting gas-solids reactions in a batch fluidized bed reactor. This configuration is capable of achieving high precursor utilization without agglomerating particles. However, the lack of predictive modeling capabilities for fluidized bed particle ALD reactors presents a challenge for implementing new coating chemistries and limits the commercial application of particle ALD. In this work, we develop a physics-based model for a fluidized bed ALD reactor that aims to predict reactor performance and enable intelligent particle ALD reactor design.

A reactor simulation based on computational fluid dynamics and discrete element method (CFD-DEM) is implemented using open-source multiphase flow modeling software MFiX. A well-characterized glass bead substrate is used to validate model predictions. The packed bed pressure drop and minimum fluidization velocity predicted by the model show good agreement with the experimental fluidization curve. Self-limited ALD reaction kinetics are implemented in the model using a first-order, irreversible Langmuir rate expression. The model can be used to predict the transient gas composition exiting the reactor and inform operating conditions to optimize precursor utilization and coating uniformity. Future work will include measurement of first-order reaction rate parameters for different particle substrates and ALD precursors, and comparison of modeling outputs to an experimental fluidized bed reactor.