Argonne researchers are using AI to optimize a popular material coating technique in real time

Newswise – To make computer chips, technologists around the world rely on atomic layer deposition (ALD), which can create films up to an atom thick. Companies often use ALD to make semiconductor devices, but it also has applications in solar cells, lithium batteries, and other energy-related areas.

Nowadays, manufacturers increasingly rely on ALD to make new types of foils, but it takes time to figure out how to optimize the process for each new material.

Part of the problem is that researchers primarily use trial and error to identify optimal growing conditions. But a recently published study – one of the first in this scientific field – suggests that the use of artificial intelligence (AI) can be more efficient.

In the ACS Applied Materials & Interfaces study, researchers from the Argonne National Laboratory of the US Department of Energy (DOE) describe several AI-based approaches for the autonomous optimization of ALD processes. Your 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 allow you to converge to optimal combinations much faster because you don’t have to spend time putting a sample in the reactor, taking it out, taking measurements, etc. like you normally would today. Instead, you have a real-time loop connected to the reactor, ”said Angel Yanguas-Gil, Argonne senior materials scientist, a co-author of the study.

Modern, but with challenges

In ALD, two different chemical vapors, so-called precursors, adhere to a surface and add a thin layer of film. All of this happens in a chemical reactor and is sequential: a precursor is added and interacts with the surface, then any excess is removed from it. After that, the second precursor is introduced and later removed, and the process repeats. In microelectronics, the ALD thin film could be used to electrically isolate neighboring components in nanotransistors.

ALD excels at growing precise nanoscale films on complex 3D surfaces, such as the deep and narrow trenches patterned in silicon wafers to make modern computer chips. This has motivated scientists around the world to develop new thin-film ALD materials for future generations of semiconductor devices.

However, developing and optimizing these new ALD processes is demanding and labor-intensive. Researchers need to consider many different factors that can change the process, including:

  • The complex chemistry between the molecular precursors
  • Reactor design, temperature and pressure
  • The timing of each dose of its precursor

To find ways to meet these challenges, the Argonne scientists evaluated three optimization strategies – random, expert system and Bayesian optimization – the latter two using different AI approaches.

Put it and forget it

Researchers evaluated their three strategies by comparing how they optimized the dosage and flush times of the two precursors used in ALD. Dosing time refers to the amount of time a precursor is added to the reactor, while purge time refers to the time it takes to remove excess precursors and gaseous chemical products.

The goal: Find the conditions that will produce high and stable film growth in the shortest possible time. Scientists also rated the strategies on how quickly they approach ideal timing by using simulations depicting the ALD process in a reactor.

By linking their optimization approaches to their simulated system, they can measure film growth in real time after each cycle based on the processing conditions that their optimization algorithms generated.

“All of these algorithms allow you to converge to optimal combinations much faster because you don’t spend time putting a sample in the reactor, taking it out, taking measurements, etc. like you normally would. Instead, you have a real-time loop connected to the reactor, ”said Angel Yanguas-Gil, Principal Materials Scientist at Argonne, co-author of the study.

This structure also automated the process for the two AI approaches by forming a closed system.

“In a closed-loop system, the simulation carries out an experiment, receives the results and feeds them into the AI ​​tool. The AI ​​tool then learns from it or interprets it somehow and then suggests the next experiment. And it’s all done without human intervention, ”said Noah Paulson, Argonne computer scientist and lead author.

Despite some weaknesses, the AI ​​approaches have effectively determined the optimal dose and flush times for various simulated ALD processes. This makes the study one of the first to show that thin-film optimization is possible in real time with AI.

“This is exciting because it opens up the possibility of using such approaches to optimize real ALD processes quickly, a step that could potentially save manufacturers valuable time and money in the development of new applications in the future,” said Jeff Elam, chief chemist in Argonne and co-author.

For partnership opportunities, please contact [email protected]

The scientists used the Argonne Blues Cluster in their Laboratory Computing Resource Center. This research was funded by the Laboratory Directed Research and Development (LDRD) program in Argonne.

Argonne National Laboratory seeks solutions to pressing national problems in science and technology. As the country’s first national laboratory, Argonne conducts leading-edge basic and applied scientific research in virtually all scientific disciplines. Argonne researchers work closely with researchers from hundreds of corporations, universities, and federal, state and local agencies to help them solve their specific problems, advance America’s scientific leadership, and prepare the nation for a brighter future. With employees from more than 60 nations, Argonne is administered by UChicago Argonne, LLC for the US Department of Energy’s Office of Science.

The Department of Energy’s Office of Science is the largest single funder of basic research in the physical sciences in the United States, working to address some of the most pressing challenges of our time. Further information can be found at https://energy.gov/science.

Comments are closed.