Elucidation of Relaxation Dynamics in Complex Fluids Through AI-informed X-ray Photon Correlation Spectroscopy

X-ray photon correlation spectroscopy (XPCS) is a useful technique for characterizing the dynamics of evolving systems and has been used successfully in combination with rheology measurements (rheo-XPCS) to observe the relaxation of complex fluids under shear in situ. However, out-of-equilibrium dynamics can produce a variety of unique and complex two-time correlation patterns which makes quantification of dynamics, or even establishing qualitative relationships between samples, extremely difficult. Meanwhile, machine learning and computer vision provide a wide range of unsupervised techniques for processing and understanding data without requiring input from the users, which can be applied to scientific data.

We have developed an unsupervised deep autoencoder capable of encoding raw XPCS data into a feature-rich latent representation, which can then be analyzed to elucidate microstructural relaxation dynamics. We test this approach on experimental data describing relaxation in a model complex fluid and show that microstructural dynamics can be directly related to macroscopic property measurements without requiring prior physical knowledge. Additionally, we will discuss how unsupervised learning can be applied in to aid high-throughput experimentation, to detect transient dynamics in real time, and to facilitate physical modeling of relaxation dynamics across timescales.