Date of Award
2024
Degree Type
Dissertation
Degree Name
Doctor of Philosophy in Chemical Engineering
Department
Chemical Engineering
First Advisor
Daniel Roxbury
Abstract
This thesis presents a multifaceted exploration of the optimization and application of single-walled carbon nanotubes (SWCNTs) in biomedical contexts, with a specific focus on enhancing their stability, functionality, and use as optical biosensors. The research spans several key areas, including the development of innovative techniques that couple machine learning with spectral fingerprinting for advanced diagnostics, the investigation of SWCNT chirality and its implications for cellular interactions, and the optimization of lyophilization processes to improve long-term stability and reproducibility.
The first section of the thesis delves into the fundamental properties of SWCNTs, emphasizing their unique electronic and optical characteristics. These properties, largely dictated by the chiral indices of the nanotubes, make SWCNTs particularly suitable for bio-sensing and imaging applications. The research highlights the critical role that chirality plays in determining the optical behavior of SWCNTs and how this can be exploited to design advanced biosensors capable of multiplexed imaging and precise biomarker detection.
A significant portion of the research is devoted to understanding how chiral pure SWCNT species interact with cells compared to multi-chiral samples. This study was motivated by the widespread use of chiral pure species in biosensor applications, despite a limited understanding of their behavior relative to multi-chiral mixtures. The findings revealed that chiral pure SWCNTs exhibit a higher rate of exocytosis and produce significantly brighter and more stable intracellular fluorescence. These insights underscore the potential for advancing biosensor technologies through the adoption of chiral pure SWCNT-based platforms, which could lead to more sensitive and reliable diagnostic tools.
The thesis also introduces a novel approach that integrates machine learning with near-infrared spectral fingerprinting of SWCNTs to accurately identify different macrophage phenotypes. This method not only achieved highly accurate phenotypical identification but also demonstrated performance comparable to traditional techniques. The success of this technique highlights its potential to revolutionize early disease diagnostics by offering a powerful tool for real-time monitoring, personalized medicine, and more efficient healthcare delivery.
In addressing the challenge of SWCNT stability during storage, this research developed an optimized lyophilization protocol combined with the strategic use of cryoprotectants. This protocol effectively preserved the optical properties and prevented aggregation of DNA-functionalized SWCNTs over extended periods, even at room temperature. The study identified glucose and polyethylene glycol (PEG) as the most effective cryoprotectants, with an 80:20 ratio of glucose to PEG providing the best long-term stability and performance. These advancements are particularly significant for the practical deployment of SWCNT-based sensors in clinical and research settings, where consistency and reliability are crucial.
Recommended Citation
Nadeem, Aceer, "NANOTUBE SPECTRAL FINGERPRINTING AND MACHINE LEARNING FOR OPTIMIZED BIOIMAGING/SENSING APPLICATIONS" (2024). Open Access Dissertations. Paper 1706.
https://digitalcommons.uri.edu/oa_diss/1706
Terms of Use
All rights reserved under copyright.
