3D µ-printing is a versatile technology with huge potential for fabricating high-quality microstructures. However, most structures initially deviate from their designed dimensions due to photo resin properties and/or optical aberrations.
We present a deep learning approach to predict and subsequently correct these optical aberrations in high numerical aperture systems, commonly employed in multi-photon lithography. The neural network identifies and calculates corrections for prominent aberrations and allows for easy scaling to arbitrary laser wavelengths. We also demonstrate our first steps of a machine learning approach that allows pre-compensation of microstructures without several (intensive) iterative correction prints.
Three-dimensional microprinting via two-photon absorption is the additive manufacturing technique of choice for complex micro-optical systems. Since post-processing of printed micro-optics is not possible in most cases, deviations between design and printed samples affect the intended function and therefore need to be minimized. This is a difficult task since important material properties such as shrinkage and refractive index depend on the cross-linking density and thus on the process parameters. We present first results towards a detailed prediction of 3D printed structures based on a modeling approach combined with machine learning to adjust the corresponding process parameters.
We show first steps towards a simple, fast, and easy to implement algorithm to predict the finally printed topography in Direct Laser Writing (DLW). These robust predictions can be used prior to the printing process to minimize undesired deviations between the experimental 3D print and its target. Consequently, this approach can eliminate the need for a multitude of structural optimization loops to produce highly conformal and high-quality microstructures in the future.
Moreover, we show first neural networks trained with our prediction algorithm being able to pre-compensate known and foreign structures without the need of performing any iterative correction prints.
To further improve the technology of 3D µ-printing, we show a promising deep learning approach for correcting aberrations of the most prominent point spread functions in (STED-inspired) multi-photon lithography.
Moreover, detailed forecasts of 3D printed structures are of high interest. Therefore, an analytical method predicting deformations due to, e.g. proximity or shrinkage effects is presented. These predictions can be used as pre-compensations to achieve a maximum match between target and actual structures from the beginning.
As third topic, we discuss the recently presented continuous frequency band chirp material measure for calibration utilization with regard to its different evaluation routines.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.