AI-SAXS: Decoding Structural Complexity with Intelligent Scattering Analysis

Purpose and goal:

Materials characterization at several length scales are crucial for understanding and tailoring e.g. mechanical and mass transport properties and thus important in product development. Small angle X-ray scattering (SAXS) is of considerable interest for characterizing materials in e.g. pharmaceuticals and packaging. The purpose of the project is to develop new AI methods and models for SAXS data generation and interpretation combined with experiments and material models to improve and speed up the material development process in industry through novel data-driven insights.

Expected results and effects:

The developed SAXS and AI models will contribute to improved understanding and utilization of SAXS experiments, accelerating materials and product development. A demonstrator will be created to compare the results from the AI-powered prototype to traditional methods by the industrial project partners. In parallel to the proposed project, the industry partners will improve their infrastructure to be able to incorporate the methods developed in the project.

Especially, these activities comprise developing pipelines for data-driven methods and training of AI algorithms. Approach and implementation:

The project will be executed in a consortium consisting of RISE, AstraZeneca, Tetra Pak and Max IV. One post-doc will be employed at RISE and work jointly with senior scientists in materials science, SAXS and AI development. The project will investigate relevant materials systems as selected by the industrial partners and modelled by conventional as well as AI models in the different work packages.

Synthesis of the materials will be performed by the industrial partners and validation of the methods and concept will be performed jointly by all project partners.