Redefining Photolithography Formulation With AI

A Citrine Informatics customer manufactures formulations for semiconductor photolithography. As silicon chips shrink in size, this customer’s products must become increasingly performant.

This article describes how Citrine Informatics's customer used the Citrine Platform to develop cutting edge formulations.

Image Credit: sasirin pamai/Shutterstock.com

The Problem

Semiconductor companies must constantly push the limits of science to produce smaller, faster processors.

Meeting these requirements depends on optimizing photolithography, a process that uses UV light to selectively remove a formulation from a silicon wafer to create patterns.

The pattern must be properly defined, which is determined by both the photosensitive molecules and the formulation's composition, as well as the method of application.

To meet industry requirements, Citrine Informatics' customer needed to create a formulation that met extremely stringent standards that had never been reached before.

The Process

Before commencing the project with Citrine, the customer had generated and tested 100 formulations and opted to do 200 trials using traditional methods and 200 using Citrine to recommend experimental possibilities (an excellent way to validate the use of AI).

In addition to the 100 initial datapoints, the Citrine Platform automatically provided extra chemical data for each molecule in each composition.

A sequential learning (SL) technique was used, where experiments were conducted in batches, and the results of each batch were used to retrain the AI model, which then suggested the next batch.

The researchers tracked what the AI model identified as significant for forecasting in each round of SL using the platform's "Feature Importance Table" to determine if it was actually learning over time.

The Outcome

The traditional experimental procedure did not achieve all of the desired qualities simultaneously. The Citrine Platform proposed previously unknown areas of inquiry, learned from the findings, and met the objectives.

  • Citrine Informatics’ Platform proposed candidate materials that broke through the existing Pareto front and met all targets. The Citrine Platform provided the top nine outcomes out of 400.
  • The customer saw this breakthrough as an opportunity to continue innovating and being competitive.
  • The next stage is to integrate the Citrine Platform into daily workflows, allowing the R&D team to quickly respond to customer requests with a formulation that can be modified in collaboration with the customer.

Image

This information has been sourced, reviewed, and adapted from materials provided by Citrine Informatics.

For more information on this source, please visit Citrine Informatics.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Citrine Informatics. (2026, January 09). Redefining Photolithography Formulation With AI. AZoM. Retrieved on January 11, 2026 from https://www.azom.com/article.aspx?ArticleID=24877.

  • MLA

    Citrine Informatics. "Redefining Photolithography Formulation With AI". AZoM. 11 January 2026. <https://www.azom.com/article.aspx?ArticleID=24877>.

  • Chicago

    Citrine Informatics. "Redefining Photolithography Formulation With AI". AZoM. https://www.azom.com/article.aspx?ArticleID=24877. (accessed January 11, 2026).

  • Harvard

    Citrine Informatics. 2026. Redefining Photolithography Formulation With AI. AZoM, viewed 11 January 2026, https://www.azom.com/article.aspx?ArticleID=24877.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.