2026-09-21T12:54:59Z https://www.ideals.illinois.edu/oai-pmh

oai:www.ideals.illinois.edu:2142/129407 2025-10-20 col_2142_5131 col_2142_11615 com_2142_5130 com_2142_9130 com_2142_8903

application/pdf en text Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo terms The student, Dongyu Fan, accepted the attached license on 2025-04-17 at 10:12. The student, Dongyu Fan, submitted this Dissertation for approval on 2025-04-17 at 10:21. This Dissertation was approved for publication on 2025-04-18 at 16:47. DSpace SAF Submission Ingestion Package generated from Vireo submission #21817 on 2025-10-19 at 18:18:24 3D tracking of solute dynamics in heterogeneous polymer networks Fan, Dongyu 2025-04-18 Landes, Christy F. Sing, Charles E Su, Xiao Kuenstler, Alexa S. single molecule tracking single-molecule microscopy polymer networks transport deep learning eng Understanding molecular transport in structurally complex polymer networks is essential for advancing technologies in separation, drug delivery, and soft materials engineering. While theoretical models and ensemble measurements offer macroscopic insights, they fall short in capturing nanoscale heterogeneity and molecular-level dynamics—factors that are critical when these materials are implemented in practical applications. The development of single-molecule microscopy opens new opportunities to revisit this longstanding problem with greater resolution and precision. Recent advances in phase-engineered point spread functions (PSFs) have further extended these capabilities to three dimensions, enabling the direct visualization of solute motion in heterogeneous environments. In this thesis, I first implement three-dimensional (3D) single-molecule tracking (SMT) to investigate solute dynamics within a charged polyelectrolyte brush (PEB). Through high-throughput trajectory analysis, I uncover spatial heterogeneity and identify two distinct transport subpopulations linked to polymer structure. To address the limitations in analyzing fast-diffusing molecules, I then develop D-Blur, a deep learning model that extracts diffusion coefficients directly from motion-blurred PSFs without trajectory linking. Validated on both simulated and experimental datasets, D-Blur enables high-throughput, localization-free diffusion mapping. Together, these experimental and computational developments establish a framework for probing transport phenomena in complex polymer networks at the single-molecule level, opening pathways toward the rational design of functional soft materials. 2025-05 Thesis https://hdl.handle.net/2142/129407 Copyright 2025 Dongyu Fan Chemical & Biomolecular Engr Chemical Engineering University of Illinois Urbana-Champaign Ph.D. Dissertation