Infrared absorption spectra can provide analytically useful information on a large variety of compounds, ranging from small ions to large biological molecules. In fact, all analytes that possess a dipole moment that changes during vibration are infrared-active. The infrared (IR) spectrum can be subdivided into far-, mid- and near- regions. The focus of attention in this thesis is the mid-IR region, in which the fundamental vibrations of most organic compounds are located, thus providing scope for positive structural identification. However, while such near-ubiquitous signals can be very useful for monitoring simple molecules in simple systems, they can be increasingly disadvantageous as the number of analytes and/or the complexity of the sample matrix increases. Thus, hyphenation to a separation system prior to detection is desirable. Paper I appended to this thesis presents (for the first time) the on-line hyphenation between Fourier transform infrared spectroscopy, FTIR, and capillary zone electrophoresis, CZE. CZE is a highly efficient separation technique that separates ionic analytes with respect to their charge-to-size ratio. It is most commonly performed in aqueous buffers in fused silica capillaries. Since these capillaries absorb virtually all infrared light an IR-transparent flow cell had to be developed. In further studies (Paper II) the applicability of CZE is expanded to include neutral analytes by the addition of micelles to the buffer, and micellar electrokinetic chromatography, MEKC, was successfully hyphenated to FTIR for the first time. Paper III describes an application of the on-line CZE-FTIR technique in which non-UV-absorbing analytes in a complex matrix were separated, identified and quantified in one run. Measuring aqueous solutions in the mid-IR region is not straightforward since water absorbs intensely in this region, sometimes completely, leaving no transmitted, detectable light. For this reason, quantum cascade lasers are interesting. These lasers represent a new type of mid-IR semiconducting lasers with high output power due to their ingenious design. The laser action lies within one conduction band (intersubband) and can be tailored to emit light in the entire mid-IR region using the same semiconducting material. To investigate their potential to increase the optical path length in aqueous solutions, these lasers were used with an aqueous flow system (Paper IV), and the experience gained in these experiments enabled hyphenation of such lasers to a CZE system (Paper V).

The work presented in this thesis is concerned with development of spectroscopic detection methods based on absorption spectroscopy using semiconductor lasers, with particular ref- erence to the field of medical diagnostics through breath analysis. The first part of this thesis deals with the design and testing of a prototype analyser for simultaneous monitoring of the exchange gases O₂ , CO₂ and H₂O in breath. The aim of this analyser is to provide information required to monitor respiration, with potential use in intensive care monitoring or during anaesthesia. The relatively high concentrations of these gases in breath and read- ily available diode laser sources make detection in the near-infrared (NIR) ideal. However, the relatively weakly absorbing A-band O₂ transitions at 760 nm require the application of a sensitive spectroscopic method, cavity enhanced absorption spectroscopy (CEAS). In contrast, CO₂ and H₂O are monitored using direct single pass absorption spectroscopy, with transitions arising from the 2ν₁ + ν₃ band at 2 μm and ν₁ + ν₃ band at 1.3 μm, respectively. It has been demonstrated that these gases can be detected simultaneously over a short pathlength (2.74 - 4 cm) in the respiratory flow by combining various spectroscopic methodologies and real-time data analysis. This analyser is shown to offer a viable alter- native for monitoring respiration, exhibiting absolute detection limits of changes of 0.26 % O₂ , 0.02 % CO₂ and 0.003 % H₂O with a 10 ms time resolution, which are comparable to current mass spectrometry based methods, but without their inherent delays. Following this, investigations into the detection of the main gas constituents in breath in the NIR employing noise-reduction modulation based spectroscopic techniques, namely wavelength and frequency modulation (WMS and FMS respectively) are also reported. The described WMS studies on water at 1.37 μm provide a demonstration of conventional WMS detection, as well as a “proof-of-principle” example of a relatively new approach to calibrating the non-absolute information obtained from a WMS absorption signal. Typically WMS spectra are calibrated using mixtures of known gas concentrations or an absolute direct absorption spectrum where possible. In this work however, a self-calibrating method, the phasor decomposition method (PDM), is employed and the returned concentration from this calibration is compared to direct absorption measurement. From this, the calculated concentration using the PDM is found to differ by 9 % from the concentration value obtained by direct absorption, providing an alternative method of calibration for when direct absorption measurements are not possible. The use of FMS in the NIR is also demonstrated as a potential alternative to CEAS for monitoring O₂ at 760 nm. FMS detection is performed on atmospherically broadened O₂ and a time-normalised α_min(t) of 2.45 ×10⁻⁶ cm⁻¹ s^1/2 is obtained, which is two orders of magnitude less sensitive than the value of α_min(t) = 2.35 ×10⁻⁸ cm⁻¹ s^1/2 obtained with CEAS. This combined with the experimental requirements of an FMS system, make its use for detection of O₂ a less practicable option compared to CEAS for real-time breath analysis. The latter work in this thesis involves a change in focus to detection of trace gases in breath in the mid-infrared (MIR). The move of spectroscopic detection to the MIR exploits the larger absorption cross-sections available in this region, and to achieve this, a relatively new form of semiconductor laser, the quantum cascade laser (QCL) is used. The design of a continuous wave QCL spectrometer at 8 μm and its operating characteristics are demon- strated and improvements in its performances are also discussed. This QCL system is then utilised to demonstrate the potential of monitoring species in breath, namely the narrow- band absorber methane and the broadband absorber acetone, taking into consideration the potential interference from other absorbing species in breath and the different spectroscopic characteristics exhibited by these molecules. Finally, the potential to further improve the sensitive detection of trace gases in breath in the MIR is also investigated with studies on the use of CEAS and multipass cells. In this work, the molecule of interest is the biomarker OCS, using transitions of the 2ν₂ band at 1031 cm⁻¹ , that are probed using a 10 μm QCL. The application of CEAS in the MIR is not as well developed as in the NIR, and the experimental consequences of using optical cavities at these wavelengths, where equipment tends to be more limited, are investigated and sensitivities discussed in the context of other literature. The experimental procedure of optimising a cavity for CEAS using the off-axis alignment method is also studied in detail, as well as the addition of WMS to further improve the signal quality. An effective absorption pathlength of ∼ 100 m was achieved in the cavity, with a bandwidth reduced α_min(BW) of 1.7 ×10⁻⁷ cm⁻¹ Hz^−1/2 using WMS CEAS achieved. With the poorer quality optics and limitations in equipment in the MIR for CEAS experiments, the use of a multipass cell, a 238 m Herriott cell, is also investigated as an alternative to the use of an optical cavity at 10 μm. Detection of OCS using direct absorption and WMS is demonstrated in the Herriott cell, achieving α_min(BW) = 2.03×10⁻⁸ cm⁻¹ Hz^−1/2 using WMS. This shows an improvement in sensitivity compared to WMS CEAS, and also shows the potential for future work on biomarker detection, as it approaches the ∼ ppb levels required for breath analysis.

Les applications de détection infrarouge active sont nombreuses dans le domaine de la défense et la sécurité. Cependant ces systèmes sont actuellement peu utilisés en pratique à cause de leur portée limitée. Un moyen d’augmenter cette portée est d’améliorer la sensibilité des détecteurs infrarouges qui possèdent des bruits bien plus élevés que leurs équivalents dans le visible. L’idée principale de cette thèse est d’utiliser l’optique non linéaire pour effectuer une conversion de fréquence du signal infrarouge à détecter vers de plus basses longueurs d’ondes et ainsi bénéficier des performances des détecteurs fonctionnants à ces longueurs d’ondes. Les développements récents en cristaux donnent un intérêt nouveau à ces techniques pour certains cas applicatifs identifiés au cours de cette thèse. L’étude détaillée de la conversion multimode à la fois temporelle et spatiale, a permis de proposer une méthode simple et originale pour dépasser l’état de l’art en terme de nombre d’éléments résolus convertis. Pour quantifier précisément les avantages de ces conversions, une étude théorique et numérique de la conversion multimode a été menée et deux séries d’expériences ont été conduites. La première concerne la détection ponctuelle de signaux moyen-infrarouge pour des applications de spectroscopie par conversion dans un cristal d’OP-GaAs. La deuxième concerne la détection de cibles par imagerie active dans le proche infrarouge par conversion dans un cristal de PPLN vers une caméra CMOS. Les performances en sensibilité obtenues sont dans les deux expériences meilleures d’un ordre de grandeur que les détections directes avec les détecteurs habituellement utilisés. / There is a wide range of applications in active infrared detection technologies in defense and security. However, the limited range available by these systems limits their developments. Increasing sensors sensitivity is a key milestone to improve this range. Indeed, noise in infrared detectors is much higher than for visible detectors due to some physical and technological issues. The key idea of this manuscript is to use nonlinear optical technologies to convert the infrared signal to detect into the visible spectrum and use all the benefits of silicon based sensors. Recent advances in optical crystals and in pump laser regimes bring renewed interest to upconversion detection for some specific application cases identified in this thesis. A novel and easy method to improve the number of converted modes has been proposed after a careful study of multimode conversion both temporally and spatially. In order to give figures on detection improvement using upconversion, we conducted a theoretical and numerical study of the multimode conversion as well as two sets of experiments. The first one, using conversion in an OP-GaAs crystal and a monodetector addresses mid-infrared spectroscopy applications. The second one addresses active imaging applications for target recognition and identification in the near-infrared. By using a PPLN crystal, the near-infrared image is detected on a low noise CMOS camera. A key milestone of this work is the sensitivity improvement of such a detection. Sensitivities obtained in each experiment are one order of magnitude better than with direct detection using common infrared sensors.

This thesis deals with the influence of process parameters on the contamination level of powder materials produced by mechanical alloying (MA) technology. For this purpose austenitic stainless steel 316 L and equiatomic CoCrFeNi high-entropy alloy (HEA) were prepared by high-energy ball milling. Both materials were milled in argon and nitrogen atmospheres from 5 to 30 hours. Spark plasma sintering method (SPS) was then used for consolidation of chosen powder samples. Chemical analysis of contamination within MA was carried out using combustion analysers for determination of carbon, oxygen, and nitrogen contents after different lengths of milling. Also differences in chemical composition of powder and corresponding bulk samples were measured. The microstructure analysis using scanning electron microscopy (SEM) of both powder and bulk materials was executed with focus on oxide and carbide presence and dispersion. Increasing content of carbon with increasing milling time was observed across all measured samples. This contamination is attributed to using milling vial made of tool steel AISI D2 (containing 1,55 wt. % of carbon). Increase of carbon content within consolidation using SPS was also observed. Milling of specimens using N2 as milling atmosphere caused higher contamination level in both AISI 316 L and HEA compared to milling in argon.