Paper Number
ET29
Session
Advanced Experimental Techniques/Methods in Rheology
Title
Observation of semiflexible filament thermal bending and transport in complex porous media
Presentation Date and Time
October 18, 2018 (Thursday) 9:30
Track / Room
Track 5 / San Felipe Room
Authors
- Tang, Zhao (Rice University, Department of Chemical and Biomolecular Engineering)
- Eichmann, Shannon (Rice University, Department of Chemical and Biomolecular Engineering)
- MacKintosh, Fred C. (Rice University, Department of Chemical and Biomolecular Engineering)
- Pasquali, Matteo (Rice University, Department of Chemical & Biomolecular Engineering)
Author and Affiliation Lines
Zhao Tang, Shannon Eichmann, Fred C. MacKintosh, and Matteo Pasquali
Department of Chemical & Biomolecular Engineering, Rice University, Houston, TX 77005
Speaker / Presenter
Tang, Zhao
Text of Abstract
Extracting high-resolution shape and position information of semiflexible filaments in different media from thousands of noisy microscopy images is the basis for understanding their mechanical and transport properties. Recent automated image processing algorithms yield subpixel resolution of the filament backbone shape in free medium via a piece-wise Gaussian-fit on pixel intensities. These algorithms yield the trajectories, orientation and bending mode amplitudes, which in turn can be used to determine mean square displacement (MSD) and mean square angular displacement (MSAD). These latter quantities relate to translational and rotational diffusivity and persistence length. However, the accuracy of the MSD and MSAD are limited by optical resolution, pixilation, and time-resolution; additional diffraction can occur in complex inhomogeneous media. Here we obtain higher resolution single-walled carbon nanotube (SWCNT) backbone points from near infrared microscopy videos of SWCNTs in porous media by further improving pixel intensity sampling method by nearby pixels interpolation. Moreover, we quantify the static error in MSDs from immobile filaments at different noise levels, and we extend dynamic error estimation from spherical particles to thread-like molecules. We apply this method to study the dynamics of highly confined stiff filaments (~10 um in length) in uniformly sized colloidal packing pores a few hundred nanometers in size. We find several translation and rotation Brownian dynamics regimes due to pore structure heterogeneity. Additionally we find that, in straight pores, filament bending energy decreases due to pore confinement; bending energy increases when a filament has to navigate across pores with different orientation, i.e., diffusion must occur against a gradient of elastic energy as a filament changes orientation across pores. This innovative imaging processing algorithm and its results are crucial to understanding filament mechanical and transport properties in emerging systems.