Abstract: Atomic layer deposition (ALD) is a thin film technique that has become critical for microelectronics, with the ALD and chemical vapor precursor market exceeding $500M/year and forecasted to grow above $1B/year by 2025. ALD has also shown promise in areas such as energy storage, catalysis, and photovoltaics.  One of the challenges of ALD is how to integrate and scale up processes in energy applications where cost and throughput are critical to determine the economic viability of a technology.

As part of Argonne’s ALD research program we have developed models to predict the scale up of ALD, including key variables such as throughput and precursor utilization, particularly as we move to larger surface areas and volumes. These models have allowed us to explore the scale up behavior in areas such as particle coating, batch processing, and roll to roll, and connect surface kinetics with experimental observables. In this presentation, I will first provide current examples of ALD manufacturing in microelectronics, energy, and high value markets. I will then look at different scale up strategies, such as fluidized bed, batch coating, and roll to roll, and compare their optimal performance under ideal ALD conditions. Finally I will briefly discuss how we can use these models to accelerate process optimization in combination with machine learning.