Resumen de New spectroscopic techniques and architectures for environmental and biomedical applications
The range of application of spectroscopic instruments is so broad nowadays that encompasses various areas of engineering, industry and scientific research. Consequently, different techniques and spectral analysis methods have been developed for the characterization of the numerous samples regularly targeted by spectroscopic sensors. In this doctoral dissertation, contributions have been made to virtually all of the main components that make up spectroscopic systems. Different methods, architectures and spectral data analysis algorithms have been proposed for environmental and biomedical applications particularly. In this way, novel techniques and architectures for molecular spectroscopy based on the measurement of optical dispersion have been presented (unlike most of the current methods that are based on the measurement of absorption). This approach, whereas maintain a reasonably low level of complexity, overcomes most of the limitations associated to absorption-based methods, providing an improved performance in some areas of the analyzers. Even though two of the proposed architectures for the estimation of gas concentration are based on the use of tunable lasers for the characterization of the spectral profile of the sample in the vicinity of an absorption feature, the best performances have been obtained using a dual-comb source. In fact, the development of new robust architectures for dual-comb spectrometers based on combs synthetized by the modulation of continuous wave lasers has been one of the main lines of work of this thesis. Although having a narrower spectral coverage than traditional combs, these sources provide far lower costs and complexity and the robustness of the generators is far higher. Most of the efforts have been made towards the developments of the new dual-comb architectures that allow to take advantage of the use comb-based systems out of the metrology laboratory. Finally, contributions on the integration of complete spectroscopic instrumentation systems, including spectral analysis techniques based on Blind Signal Separation have been made. For that, a non-invasive biomedical spectroscopic instrument has been developed and used as a benchmark to study the viability of diffuse spectroscopic methods and spectral data classification techniques in the monitoring of the state of angiogenesis of a bioengineered skin substitute.
