Calculation, simulation, and experimental analysis of distorted and enhanced spectra from attenuated total reflection (ATR)
Date Issued
2025-01-10
Author(s)
Supervisors
Referees
Abstract
This doctoral thesis extensively investigates the principles, calculations, experiments, and simulations related to the distortion spectrum and surface-enhanced spectroscopy within the framework of attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR). The research encompasses four main aspects:
1. Based on Maxwell's equations, an in-depth exploration is conducted into the origins of spectral distortion with computations utilizing Snell's law. The combination of the established model with Fresnel’s equation yields simulation outcomes that align with the experimental spectral data. Correction of the distortion spectrum is realized through the application of the Kramers-Kronig (KK) transform and an algorithm resembling the Fourier transform (FT). The subsequent analysis challenges conventional understanding by revealing the blue shift associated with the degree of spectral distortion.
2. The study systematically introduces the transition from Fourier Transform (FT) methodologies to contemporary deep learning algorithms, particularly neural networks. Recognizing the intrinsic complexity of conventional correction methods, the investigation performs the classification and correction of distorted spectra using artificial neural network algorithms. Comparative assessments with traditional methods indicate that long short-term memory (LSTM) and Transformer models exhibit accelerated processing speeds and heightened batch correction capabilities.
3. Theoretical calculations for surface plasmon generation, accounting for the thickness of the thin layer, were conducted. The surface enhancement spectrum of Pd nanoparticles is demonstrated through the integration of theoretical calculations with experiments.
4. The thesis expounds on the principles of two-dimensional Fourier transform (2D FT) and provides an in-depth analysis of the classification and fundamental principles of two-dimensional infrared spectroscopy (2D IR). As a technique rooted in third-order nonlinear optical phenomena, 2D IR spectroscopy exhibits distortion and surface-enhanced spectroscopy characteristics similar to those observed in one-dimensional (1D) spectroscopy, particularly near the critical angle. Furthermore, due to the unique principles of 2D IR, it also demonstrates enhanced specificity at the Brewster angle. This thesis offers a comprehensive discussion and comparison of the similarities and differences between 1D and 2D surface-enhanced spectroscopy.
1. Based on Maxwell's equations, an in-depth exploration is conducted into the origins of spectral distortion with computations utilizing Snell's law. The combination of the established model with Fresnel’s equation yields simulation outcomes that align with the experimental spectral data. Correction of the distortion spectrum is realized through the application of the Kramers-Kronig (KK) transform and an algorithm resembling the Fourier transform (FT). The subsequent analysis challenges conventional understanding by revealing the blue shift associated with the degree of spectral distortion.
2. The study systematically introduces the transition from Fourier Transform (FT) methodologies to contemporary deep learning algorithms, particularly neural networks. Recognizing the intrinsic complexity of conventional correction methods, the investigation performs the classification and correction of distorted spectra using artificial neural network algorithms. Comparative assessments with traditional methods indicate that long short-term memory (LSTM) and Transformer models exhibit accelerated processing speeds and heightened batch correction capabilities.
3. Theoretical calculations for surface plasmon generation, accounting for the thickness of the thin layer, were conducted. The surface enhancement spectrum of Pd nanoparticles is demonstrated through the integration of theoretical calculations with experiments.
4. The thesis expounds on the principles of two-dimensional Fourier transform (2D FT) and provides an in-depth analysis of the classification and fundamental principles of two-dimensional infrared spectroscopy (2D IR). As a technique rooted in third-order nonlinear optical phenomena, 2D IR spectroscopy exhibits distortion and surface-enhanced spectroscopy characteristics similar to those observed in one-dimensional (1D) spectroscopy, particularly near the critical angle. Furthermore, due to the unique principles of 2D IR, it also demonstrates enhanced specificity at the Brewster angle. This thesis offers a comprehensive discussion and comparison of the similarities and differences between 1D and 2D surface-enhanced spectroscopy.
Subjects
ATR-FTIR
;
Distortied spectrum
;
Simulation and Correction
;
Deep learning
;
Surface-Enhanced Spectroscopy
;
2DIR
Institution
Type
Dissertation
Language
English
File(s)![Thumbnail Image]()
Loading...
Name
Calculation, simulation, and experimental analysis of distorted and enhanced spectra from attenuated total reflection (ATR).pdf
Size
12.65 MB
Format
Adobe PDF
Checksum
(MD5):2bcf7200462a051957868610cac6cb6b