Collaborative Research: Understanding and Controlling Multiscale Structural Ordering in Particle Self-assembly

Self-assembly and structural ordering of particles during the slurry drying process are ubiquitous, intricate, and functionally critical. This process significantly influences important applications such as genotyping, biosensing, 3D printing, and the production of thin films for various purposes. Monitoring, understanding, and predicting the multiscale structural dynamics under different drying conditions poses a major challenge in studying particulate and multiphase processes, which involve fundamental phenomena like wetting, evaporation, surface tension, and multiphase flow. This project aims to develop a comprehensive fundamental understanding of the dynamic structural evolution in slurries and to create a predictive machine learning model for guiding the optimization of the drying process. This knowledge and methodology will offer new insights into the dynamics of particle self-assembly, aiding in the design of drying processes to control the microstructure of particulate systems to achieve desired mechanical and electrical properties. The collaboration between University of Texas at Austin and University of Wisconsin-Madison presents unique opportunities for recruiting under-represented students and for engaging with the science-technology-entrepreneurship training programs. This award aims to develop a comprehensive fundamental understanding of the dynamic drying process of a particle-laden slurry. The mechanistic insights will be integrated into a predictive machine learning model to guide the optimization of the drying process for various composite systems. The following research tasks will be conducted. (i) Establishing the correlation between the drying condition and multi-scale structure ordering. (ii) Imaging and predicting the spatiotemporal evolution of the microstructure. (iii) Model-guided optimization of the mechanical and electrical properties of particulate composites. Specifically, 3D in-situ imaging will be applied to model slurry systems consists of thousands of oxide particles with controlled morphology suspended in a liquid solvent with controlled viscosity. A computing module will be developed to identify, recognize, and track all of them in the 4D imaging data (space and time), which will then serve as inputs for the graph-based machine learning effort. Overall, the project will reveal how the system’s non-equilibrium behaviors would affect its final structural ordering and, thus, its mechanical and electrical properties. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.