Artificial Intelligence, or AI, is now being used in almost every sector, and people are very much dependent on machine learning and artificial intelligence as it reduces much of the workload. The chip industry is growing very fast, and its production is also growing very rapidly because many industries are using it on a large scale. Currently, computer chips are made using a special type of technology called atomic layer deposition (ALD), which has the capability to create films as fine as one atom thick. This technology is very much used to develop semiconductor devices, but it also has applications in lithium batteries, solar cells, and other energy-related fields.
Today, manufacturers are increasingly relying on ALD to make new types of films, but it takes time to figure out how to fine-tune the process for each new material. Part of the problem is that researchers primarily use trial and error to determine optimal growing conditions. However, a recently published study, one of the first in this scientific field, suggests that the use of artificial intelligence (AI) may be more efficient. In the ACS Applied Materials and Interfaces study, researchers from the Argonne National Laboratory of the USD Department of Energy (DOE) describe several AI-based approaches for the autonomous optimization of AML processes. Their work describes the relative strengths and weaknesses of each approach, as well as insights that can be used to develop new processes more efficiently and economically. "All of these algorithms provide a much faster way to converge to optimal combinations because you don't waste time putting a sample in the reactor, taking it out, taking measurements, etc. like you normally would today, a real-time loop that connected to the reactor," said Argonne senior materials scientist Angel YanguasGil, a co-author on the study.
