Trustworthy Physics-informed Machine Learning Enabled Rapid Continuous Nanoscale 3D Printing of Functional Devices

Interdisciplinary Areas: Data and Engineering Applications, Future Manufacturing, Micro-, Nano-, and Quantum Engineering

Project Description

3D printing is one of the most important technology developments in recent history for applications ranging from prototyping and product visualization, to building functional components and devices. Nanoscale 3D printing based on femtosecond laser two-photon polymerization has been used to fabricate a wide range of nanostructured materials and devices with unprecedented properties and functionalities. However, the slow speed of 3D nanoprinting, which is a point-by-point printing process, is a major obstacle to its adoption for a wider spread of applications. Recently, a rapid, continuous, layer-by-layer femtosecond laser projection 3D nanoprinting technology has been developed at Purdue. This project is to incorporate trustworthy physics-informed machine learning (ML) in 3D nanoprinting to improve the robustness of the 3D printing process and the 3D printing accuracy with predictive confidence interval. Various advanced physics-informed ML algorithms, in combination with physics-based modeling, will be used to develop ML and AI tools for the rapid printing of arbitrary nanoscale 3D geometries. Femtosecond laser processing and photo-polymerization processes will also be investigated. The ultimate goal is to develop physics-informed ML -enabled rapid and robust 3D nanoprinting technology to produce precise 3D parts with a significantly higher printing speed compared with state-of-the-art 3D nanoprinting systems.


Start Date



Postdoc Qualifications

Ph.D. in, Mechanical Engineering, Computer Science, Mathematics, or relevant background in femtosecond laser-based manufacturing and/or machine learning.



Xianfan Xu,, James J. and Carol L. Shuttleworth Professor of Mechanical Engineering,

Guang Lin,, Professor, Department of Mathematics (primary appointment), Professor, Department and Statistics, Professor, School of Mechanical Engineering,


Short Bibliography

• Somers, P., Liang, Z., Johnson, J.E., Boudouris, B.W., Pan, L., and Xu, X., 2021, "Rapid, continuous projection multi-photon 3D printing enabled by spatiotemporal focusing of femtosecond pulses", Light: Science & Applications, 10:199, DOI: 10.1038/s41377-021-00645-z
• Hsu, S.H., Chi, T., Kim, J., Somers, P., Boudouris, B.W., Xu, X., and Pan, P., 2021 "High-Speed One-Photon 3D Nanolithography Using Controlled Initiator Depletion and Inhibitor Transport", Adv. Optical Materials, DOI: 10.1002/adom.202102262
• Somers, P., Liang, Z., Chi, T., Johnson, J.E., Pan, L., Boudouris, B.W., Xu, X., 2023, "Photo-activated polymerization inhibition process in photoinitiator systems for high-throughput 3D nanoprinting", Nanophotonics, DOI: 10.1515/nanoph-2022-0611
• Johnson, J.E., Chen, Y. Xu, X., 2022, Model for polymerization and self-deactivation in two-photon nanolithography, Opt. Exp. Vol. 30, pp. 26824-26840.
• Chen, J., Kang, L., Lin, G., 2020, Gaussian Process assisted Active Learning of Physical Laws, Technometrics, DOI: 10.1080/00401706.2020.1817790.