Date of Award

Spring 1-1-2014

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Chemical & Biochemical Engineering

First Advisor

Daniel K. Schwartz

Second Advisor

Andrew Goodwin

Third Advisor

Charles Musgrave

Fourth Advisor

Will Medlin

Fifth Advisor

Joe Falke

Abstract

Molecular interactions with solid-liquid interfaces have long been studied through macroscopic observations. There were, however, only a limited number of ways to observe true molecular phenomena leading to a wide discrepancy between theoretical models and experimental results. The work presented here uses total internal reflection fluorescence (TIRF) microscopy to image individual molecules at the solid-liquid interface as they undergo the dynamic processes of adsorption, diffusion, and desorption.

Studying these dynamic behaviors at the single-molecule level allowed great insight into the macroscopically observed hydrophobic effect as well as the Hofmeister effect. The hydrophobic effect was probed by looking at the response of individual molecules to surfaces with varying alkyl chain lengths. These experiments showed that surface residence time increased and mobility decreased with increasing alkyl chain length despite all of the surfaces having the same nominal hydrophobicity. Experiments using the salts NaF and NaSCN dissolved in water along with a fatty acid probe molecule were conducted to examine the Hofmeister effect at the molecular level. These experiments showed a dramatic change in adsorption rate of the hydrophobic probe onto a hydrophobic surface, but minimal change in diffusion or desorption rate.

We used the knowledge that molecular probes interact with specific surface chemistries very differently to develop a super-resolution imaging technique called MAPT (mapping using accumulated probe trajectories). MAPT created images of a surface using each molecular behavior (e.g. diffusion) as a contrast mechanism. These images were first used to show variations in hydrophobicity on a photopatterned self-assembled monolayer. MAPT images also allowed us to differentiate between the 2D Brownian motion of a molecule on a surface and intermittent 3D flights through solution.

Finally, we developed a technique for identifying surface chemistry using dynamic molecular interactions using an unsupervised Gaussian mixture modeling algorithm. This algorithm identifies regions on a surface that share similar molecular behaviors which can then be compared to the behaviors observed on surfaces of known chemistry. These identifications allow, for the first time, one to create true maps of surface chemistry.

Share

COinS