Developing ML-based hologram generation algorithm for advance lithography and 3D volumetric printing

Photolithography is the cornerstone of the whole semiconductor industry. Despite the tremendous and exciting achievements of the last 20 years, fundamental limits of traditional projection photolithography start to emerge such as ultimate resolution (i.e. critical dimensions), exponentially growing hardware and operational costs and stagnant throughput due to higher systems complexity. Maskless advanced lithography is a multidisciplinary research field that aims to overcome these limits by directly shaping a light wavefront without the need of an expensive and complex phase mask and by using novel resist formulation.
The proposed technology combines the used of advanced holograms optimized with a novel multi-color resist capable of achieving super-resolution (i.e., below the diffraction limit) critical dimensions whilst retaining a large manufacture area typical of holograms. Unlike traditional projection photolithography, which relies on direct exposure patterns, holography precisely controls the wavefront of light, enabling volumetric and multi-plane structuring with greater flexibility and accuracy. Furthermore, it can enable a completely new filed of additive manufacturing named 3D lithography by volumetric optimization schemes based on ML models.
Our technology enables the fabrication of complex micro- and nanoscale structures with high yield, unlocking new possibilities in advanced electronics, MEMS, energy harvesting, metasurfaces, microfluidic and 3D volumetric printing among other fields. From next-generation chip design to bio-inspired materials and high-throughput optics production, it has the potential to shaping the future of production of semiconductor devices.

The project focuses on improving and optimizing a machine-learning (ML) algorithm inspired by the paper Time-Multiplexed Neural Holography (Choi & Gopakumar, 2022). A primary objective is incorporating a non-linear response of a photoactive compound into the computational pipeline, optimizing the focal stack for multi-colour chemistry using high-performance GPUs available on the imec cluster.

You will create and evaluate multiplexed holograms for different testing geometries and validate their performances using spatial light modulators (SLM) on a 3D volumetric printing system.
We are seeking passionate scientists and engineers with very good programming skills to drive innovation in lithographic techniques and additive processes. If you are eager to contribute to cutting-edge research and translate scientific discoveries into groundbreaking applications, join us in redefining the limits of semiconductor fabrication. You can expect from us thorough training at the start of your job. We teach you how to work with advanced algorithms and you gain a good inside on the experimental part related to your computational work. Supervision with mentors will be held for progress tracking and troubleshooting. validation.
 

Type of Project: Internship; Combination of internship and thesis

Master's degree: Master of Engineering Technology; Master of Science

Master program: Computer Science; Materials Engineering

Duration: 3-6 months

For more information or application, please contact the supervising scientists Giammarco Nalin (giammarco.nalin@imec.be), David Blinder (david.blinder@imec.be) and John Petersen (john.petersen@imec.be). 

Imec allowance will be provided.