Date Thesis Awarded


Document Type

Honors Thesis

Degree Name

Bachelors of Science (BS)




Kristin Wustholz

Committee Member

Christopher Abelt

Committee Member

William McNamara


Dye-sensitized solar cells (DSSCs) comprised of sensitizers within a TiO2 semiconductor matrix are promising photovoltaics that use electron transfer (ET) to convert photons into current. Improvements in DSSC efficiency require fundamental understanding of the distribution of forward and back ET dynamics at the dye-semiconductor interface, which ensemble-averaged experiments are incapable of observing. In this study, the distributions of ET dynamics in dye-sensitized TiO2 systems are probed using single-molecule fluorescence (SMF) microscopy. The time-dependent emission (i.e., blinking dynamics) of rhodamine 6G (R6G), rhodamine B (RB), and 5-carboxy-x-rhodamine (5-ROX) sensitized colloidal anatase TiO2 films are quantified by constructing histograms of emissive (“on”) and non-emissive (“off”) events. Robust statistical analysis reveals that the off-time distributions for molecules on TiO2 are not consistent with the power-law hypothesis and are instead well represented by log-normal distributions. These ET data are consistent with the Albery model of a Gaussian distribution of activation energies, where the power-law and log-normal distribution fit parameters are sensitive to experimental time resolution and cannot be analyzed as absolute values. Relative comparison of the fit parameters of R6G, RB, and 5-ROX ET distributions on colloidal TiO2 provides insight into the effect of sensitizer structure on the photophysical system. Notably, log-normally distributed off times produce a larger scaling parameter μ for 5-ROX than for R6G or RB, suggesting the dye binds more strongly to TiO2 than R6G or RB and produces slower back electron transfer (BET) rates due to increased electron delocalization. The on-time distributions for rhodamine dyes on TiO2 are fit by power laws, but they are only operative for long emissive durations comprising less than 15% of the photophysical data. Changes in fluorescent spot density on false-colored images of R6G on bare glass, colloidal anatase TiO2, and single-crystal rutile TiO2 illustrate the impact of substrate on ET dynamics, but single-molecule studies of rhodamine molecules on single-crystal rutile could not be completed due to extremely efficient ET. Future investigations will apply single-molecule methodologies with picosecond resolution to dye-semiconductor systems.

Available for download on Tuesday, May 07, 2019