(361c) Autonomous Construction of Nanomaterial Phase Maps Using High-Throughput X-Ray Scattering | AIChE

(361c) Autonomous Construction of Nanomaterial Phase Maps Using High-Throughput X-Ray Scattering


Vaddi, K. - Presenter, University of Washington
Li, K., University of Washington
Pozzo, L., University of Washington
Scheiwiller, S., University of Washington
Automation in many of the experimental pipelines both at laboratory and central facilities has shifted the bottleneck to autonomously-driven data analysis and decision-making. Exponential growth in tools available for data-driven modeling resulted in the advent of self-driven laboratories (SDL) that aims to automate and accelerate the entire workflow starting from synthesis to characterization and device integration for emerging technologies and energy needs. Platforms based on solution-processible materials (polymers, colloids, and nanoparticles) are amenable to automation both at the synthesis and characterization levels. Techniques such as scattering and spectroscopy provide faster high-throughput alternatives to capture a signal of the underlying structure allowing us to construct composition-structure phase maps. Although obtaining a structure phase map from high throughput structural characterization data is a decade-old problem, many of the state-of-the-art approaches are not generalizable beyond their initial application in X-ray diffraction. In particular, small-angle scattering (SAS) used to probe the structure of nanoscale material results in a continuous signal where the key information is not just in peak positions but also the overall ‘shape’ of the curve. Thus, self-driven laboratories for soft materials should be equipped with tools that can process the shape of SAS curves into useful information for decision-making and optimization. In this talk, we present a novel approach that uses Riemannian metric geometry to develop a shape-based measure for analyzing SAS curves. We use this novel analysis tool in an SDL to build a phase map of the block copolymer system in an iterative closed-loop fashion. Our approach allows the decision-making to be built on top of multiple Riemannian metric fields each corresponding to an order parameter of the phase map. Using a case study of a novel polymer blend, we show that the learned order parameters are continuous and thus have an interpretation in terms of multi-phase transitions.