Elemental analysis in forensic laboratories can be tedious and many trace evidence items are not analyzed to determine their elemental composition. Presently, scanning electron microscopy-energy dispersive x-ray spectroscopy (SEM-EDS) is the primary analytical tool for determining the elemental composition of trace evidence items. However, due to the time it takes to obtain the required vacuum and the limited number of samples that can be analyzed at any one time, SEM-EDS can be impractical for a high volume of evidence items. An alternative instrument that can be used for this type of analysis is laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). While LA-ICP-MS is a very precise and quantitative analytical method that determines elemental composition based on isotopic mass measurements; however, the instrumentation is relatively expensive and therefore is budgetarily prohibitive for many forensic laboratories. It is the purpose of this research to evaluate an inexpensive instrument that can potentially provide rapid elemental analysis for many forensic laboratories. Laser induced breakdown spectroscopy (LIBS) is an analytical method that meets these requirements and offers information about the elemental composition based on ionic, atomic and diatomic molecular emissions.

Laser-induced breakdown spectroscopy (LIBS) is an advanced data analysis technique for spectral analysis based on the direct measurement of the spectrum of optical emission from a laser-induced plasma. Assignment of different atomic and ionic lines, which are signatures of a particular element, is the basis of a qualitative identification of the species present in plasma. The relative intensities of these atomic and ionic lines can be used for the quantitative determination of the corresponding elements present in different samples. Calibration curve based on absolute intensity is the statistical method of determining concentrations of elements in different samples. Since we need an exact knowledge of the sample composition to build the proper calibration curve, this method has some limitations in the case of samples of unknown composition. The current research is to investigate the usefulness of ANN for the determination of the element concentrations from spectral data. From the study it is shown that neural networks predict elemental concentrations that are at least as good as the results obtained from traditional analysis. Also by automating the analysis process, we have achieved a vast saving in the time required for the data analysis.

The detection of trace amounts of H and D present in Zr-2.5%Nb in the form of ZrH and ZrD, respectively, by LIBS was explored. The intended use case for this experimentation was CANada Deuterium Uranium (CANDU) nuclear reactor pressure tube inspections where hydride buildup can lessen the mechanical strength of these components. As these tubes carry coolant and house the fuel bundles, their integrity is paramount. A LIBS inspection method is of interest in the nuclear industry due to the operational flexibility it would introduce and its ability to reduce the cost, time, and radiation dose associated with inspection campaigns of pressure tubes in CANDU nuclear reactors. Using LIBS, simultaneous detection of H and D was achieved in a low-pressure air environment using a microjoule, picosecond pulsed laser and emission being captured by a high-resolution spectrometer. The emission lines of the two species were blended, however, two peaks can be seen. Experiments using a milijoule, nanosecond pulsed laser in a LIBS setup were also conducted at atmospheric pressures. These experiments failed to show D emission, however. In addition to detecting emission from H and D, a Monte Carlo algorithm was developed for estimating the error associated with a LIBS inspection of a pressure tube segment. ZrH and ZrD form heterogeneous structures in the bulk of the Zr-2.5%Nb pressure tube material, meaning that a single measurement would not be indicative of the entire tube. Using metallographs of artificially hydrided pressure tube samples, the error within a given confidence interval was found as a function of number of measurement sites and ablation diameter. Furthermore, the impacts to Zr-2.5%Nb based on intense laser-matter interactions was investigated by optical microscopy and interferometry, allowing for 3-dimensional reconstructions of ablation craters. The morphology of millijoule, nanosecond pulsed laser-matter interaction and microjoule, picosecond pulsed laser-matter interaction were the subjects of this investigation. The salient difference between the two interactions is the evidence of substantial melting and subsequent re-deposition of material in the case of nanosecond interactions, whereas picosecond ablation yielded little melting. These results support the further development of a LIBS-based inspection method for determining the concentration of H and D in Zr-2.5%Nb. It was found that a vacuum environment allows for the simultaneous detection of H and D emission. Further experimentation should explore using low-pressure buffer gas environments as a method to further distinguish emission between the two species. / Thesis / Master of Applied Science (MASc)

Yes / Ultrasmall SnO2 nanocrystals as anode materials for lithium-ion batteries (LIBs) have been synthesized by bubbling an oxidizing gas into hot surfactant solutions containing Sn-oleate complexes. Annealing of the particles in N2 carbonifies the densely packed surface capping ligands resulting in carbon encapsulated SnO2 nanoparticles (SnO2/C). Carbon encapsulation can effectively buffer the volume changes during the lithiation/delithiation process. The assembled SnO2/C thus deliver extraordinarily high reversible capacity of 908 mA.h.g(-1) at 0.5 C as well as excellent cycling performance in the LIBs. This method demonstrates the great potential of SnO2/C nanoparticles for the design of high power LIBs. / National Natural Science Foundation of China (21103039), Anhui Province Natural Funds for Distinguished Young Scientists, https://bradscholars.brad.ac.uk/browse?order=ASC&rpp=20&sort_by=-1&etal=-1&offset=6150&type=authorResearch Fund for the Doctoral Program of Higher Education of China (20110111120008), Beijing National Laboratory for Molecular Sciences (BNLMS), and Deutsche Forschungsgemeinschaft Grant (DFG): H1113/3-5. C.Y. acknowledges the support from the “Thousand Talents Program” and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

Nos últimos anos, a Espectrometria de Emissão Atômica com Plasma induzido por Laser (LIBS) vem despertando grande interesse e pesquisas voltadas às aplicações analíticas dessa técnica tem aumentado significativamente. No entanto, sua aplicação na análise de amostras líquidas não é trivial, podendo haver comprometimento na sensibilidade e na precisão do método. Além disso, a análise quantitativa é considerada o \"calcanhar de Aquiles\" da técnica, especialmente devido à falta de materiais de referência certificados e à forte interferência de matriz. Nesse contexto, com a intenção de viabilizar a análise de matrizes líquidas por LIBS, esse trabalho propõe o uso de cerâmica como fase sólida para extração, pré-concentração e especiação de Cr em amostras de água. Foi realizado também um estudo de otimização dos parâmetros instrumentais do LIBS para o desenvolvimento de um método de medição de Cr na fase sólida, empregando cerâmica enriquecida com concentrações crescentes desse analito como material calibrante sintético. Os resultados obtidos mostraram que a cerâmica adsorve seletivamente a espécie Cr(III) em pH 7, sendo este um processo rápido, de baixo custo, cuja eficiência é pouco afetada pela presença de outros eletrólitos. Os estudos relacionados à otimização dos parâmetros instrumentais do LIBS indicaram que um diâmetro de focalização de 65 µm, um tempo de atraso de 0,75 µs, a energia por pulso de 20 mJ, uma taxa de repetição de 10 Hz e um número de pulsos igual a 403 podem ser recomendados para a quantificação de Cr no material cerâmico. Nessas condições, obteve-se uma curva de calibração analítica que possibilitou a medição Cr adsorvido em cerâmica. Uma amostra de água de torneira foi então enriquecida com 0,9 mg L-1 de Cr(III) e 0,9 mg L-1 de Cr(VI) para que um teste de recuperação fosse realizado com a finalidade de se verificar a aplicabilidade do método proposto, foi possível constatar a adsorção seletiva da espécie Cr(III), com recuperação de 105±2%. A quantificação da espécie Cr(VI) por LIBS também é possível após a redução dessa espécie a Cr(III) utilizando Fe(II) como agente redutor. Os limites de detecção (LOD) e quantificação (LOQ) estimados para o método foram respectivamente 21 mg kg-1 e 70 mg kg-1, equivalentes a 0,1 mg L-1 e 0,35 mg L-1, considerando-se o coeficiente de pré-concentração igual a 200 (m/m). / In recent years, Laser Induced Breakdown Spectroscopy (LIBS) has deserved attention of scientific community and researches using this technique have increased significantly. However, its application in the analysis of liquid samples is not trivial, because the sensitivity and accuracy of the method can be affected. In addition the quantitative analysis is considered the \"Achilles\' heel\" of the technique, especially due to the lack of certified reference materials and to the strong interference matrix. To enable the analysis of liquid matrices by LIBS, this work proposes the use of ceramic as solid phase in the extraction, preconcentration and speciation of Cr in water samples. LIBS instrumental parameters were also optimized to allow the measurement of Cr in this solid phase, for the development of a method for measuring the analyte in the solid phase, which was also used as synthetic calibrating material by the addition of increasing concentrations of the analyte. Results showed the ceramic selectively adsorbs Cr(III) species at pH 7, which is a low cost and rapid process, whose efficiency is not significantly affected by the presence of other electrolytes. Studies related to the optimization of LIBS instrumental parameters indicated that spot size of 65 µm, delay time of 0,75 µs , energy per pulse of 20 mJ, repetition rate of 10 Hz and a number of pulses equal to 403 can be recommended for the quantification of Cr the ceramic material. Under these conditions, it was obtained an analytical calibration curve which allowed a Cr measurement in ceramics. A tap water sample was enriched with 0.9 mg L-1 of Cr(III) and 0.9 mg L-1 of Cr (VI) and a recovery test was done in order to verify the applicability of the proposed method. It was observed the selective adsorption of the Cr(III) species, with recovery of 105 ± 2 %. The quantification of Cr(VI) species by LIBS was also possible, after reduction of this species to Cr(III) using Fe(II) as reducing agent. The limit of detection (LOD) and the limit of quantification (LOQ) obtained by the proposed method were respectively, 21 mg kg-1 and 70 mg kg-1, equivalent to 0.1 mg L-1 and 0.35 mg L-1, considering the preconcentration coefficient of 200 (on a w/w basis).

O manejo adequado do solo é de suma importância na incorporação da matéria orgânica no mesmo, sequestrando carbono da atmosfera. Neste trabalho avaliaram-se dois manejos do sistema produtivo da pecuária, em pastagens de Brachiaria decumbens: (A3) pastagem em recuperação, onde nitrogênio foi aplicado, e (A4) pastagem degradada, onde nenhum tipo de correção é efetuado. Utilizou-se uma área de Mata Atlântica (MT) próxima ao experimento como referência. O foco principal do trabalho foi o acúmulo de carbono no solo e a emissão de CO2. A partir dos teores de carbono e valores de densidade foram estimados os estoques de carbono (ECs), em superfície (0-30 cm) e em profundidade (0-100 cm), em cada um dos sistemas, aplicando as correções por massa equivalente e por teores de argila. Para a estimativa de emissão de CO2 dos solos utilizou-se um sistema portátil que monitora o fluxo de CO2 no interior da câmara através de espectroscopia de absorção no infravermelho. Os resultados mostraram que o sistema A3 acumulou o maior EC, com 142 Mg ha-1, enquanto aproximadamente 99 Mg ha-1 foi obtido para o sistema A4. Na mata o EC foi de 115 Mg ha-1. Os dados de respiração do solo mostraram situação similar para ambas as áreas de pastagem. Outra proposta desse trabalho foi avaliar a qualidade da matéria orgânica do solo (MOS) utilizando a espectroscopia de fluorescência induzida por laser (LIFS). Observou-se o aumento do grau de humificação da MOS (HFIL) com o aumento da profundidade. Os menores valores de humificação foram obtidos na área em recuperação, indicando certa fragilidade deste estoque de carbono ser perdido caso a área não seja bem manejada. Avaliou-se também o potencial da espectroscopia de emissão óptica com plasma induzido por laser (LIBS) para a quantificação de carbono total em solos. Os dados dos teores de carbono obtidos por análise elementar foram utilizados para a calibração do sistema LIBS de acordo com a textura dos solos. O coeficiente de correlação entre os valores de referência (CHNS) e os valores preditos por LIBS na validação foi de 0,87 (erro ~ 25%) para textura arenosa e 0,92 (erro ~ 16%) para textura argilosa. Este trabalho demonstrou a importância do manejo adequado para recuperação de pastagens no Brasil e o risco de se perder esse carbono para atmosfera devido à alta labilidade do carbono armazenado. As pastagens bem manejadas podem levar este sistema produtivo a patamares ambientalmente sustentáveis e contribuir para mitigação do efeito estufa. As técnicas fotônicas apresentaram potencial para determinação de carbono no solo e avaliação de sua estrutura química. Desta forma, espera-se que futuramente este tipo de instrumentação possa ser aplicada ao campo, fornecendo parâmetros que ajudem na tomada de decisões para um desenvolvimento agrícola sustentável. / Proper soil management is important to organic matter incorporation, sequestering carbon from atmosphere. This study was developed for two livestock managements: (A3) pasture recovering of Brachiaria decumbens, where nitrogen is applied, and (A4) degraded pasture of Brachiaria decumbens, with no correction. The reference system was a native forest area (Brazilian Atlantic Forest) next to the experiment. This study aims to compare the soil carbon accumulations and CO2 emissions for both productive systems. Soil carbon content and soil bulk density values were measured to calculate carbon stocks (CSs) in surface (0-30 cm) and at depth (0-100 cm) in each system, applying equivalent mass and clay content corrections. CO2 emissions were estimated using a portable system, which monitors the CO2 flux within a chamber through infrared absorption spectroscopy. The soil carbon stocks were estimated around 142 Mg ha-1 for A3 and 99 Mg ha-1 for A4, while around 115 Mg ha-1 was found in the forest. The results for CO2 emission showed a similar situation for both pasture areas. Another purpose of this study was to evaluate the soil organic matter (SOM) quality using laser-induced fluorescence spectroscopy (LIFS). An increasing of humification degree of SOM (HLIF) was observed with increasing of depth. The lowest HLIFS values were obtained for the recovering area, indicating a certain fragility of carbon stocks, which can be lost if the area is not well managed. In addition, Laser-induced breakdown spectroscopy (LIBS) was used for total soil carbon quantification. Using the data from elemental analysis, mathematical models were developed to calibrate the LIBS system for each class of soil texture. For carbon quantification, the correlation between reference technique (CHN) and LIBS prediction was 0.87 for sandy soils (error around 21%) and 0.92 for clayed soils (error around 16%). This study demonstrated the importance of pasture recovering in Brazil from the point of view of soil carbon sequestration, and the risk of this stock to be lost due to high lability of stored carbon. Well-managed pastures can lead this production system to environmentally sustainable levels and contribute to mitigate the greenhouse effect. Systems with high potential for portability to determine soil carbon and evaluation of its chemical structure have been successfully tested. Thus, it is expected that in a near future, this kind of instrumentation can be effectively applied in the field and provide parameters that help in the decision making for a sustainable agricultural development.

O manejo adequado do solo é de suma importância na incorporação da matéria orgânica no mesmo, sequestrando carbono da atmosfera. Neste trabalho avaliaram-se dois manejos do sistema produtivo da pecuária, em pastagens de Brachiaria decumbens: (A3) pastagem em recuperação, onde nitrogênio foi aplicado, e (A4) pastagem degradada, onde nenhum tipo de correção é efetuado. Utilizou-se uma área de Mata Atlântica (MT) próxima ao experimento como referência. O foco principal do trabalho foi o acúmulo de carbono no solo e a emissão de CO2. A partir dos teores de carbono e valores de densidade foram estimados os estoques de carbono (ECs), em superfície (0-30 cm) e em profundidade (0-100 cm), em cada um dos sistemas, aplicando as correções por massa equivalente e por teores de argila. Para a estimativa de emissão de CO2 dos solos utilizou-se um sistema portátil que monitora o fluxo de CO2 no interior da câmara através de espectroscopia de absorção no infravermelho. Os resultados mostraram que o sistema A3 acumulou o maior EC, com 142 Mg ha-1, enquanto aproximadamente 99 Mg ha-1 foi obtido para o sistema A4. Na mata o EC foi de 115 Mg ha-1. Os dados de respiração do solo mostraram situação similar para ambas as áreas de pastagem. Outra proposta desse trabalho foi avaliar a qualidade da matéria orgânica do solo (MOS) utilizando a espectroscopia de fluorescência induzida por laser (LIFS). Observou-se o aumento do grau de humificação da MOS (HFIL) com o aumento da profundidade. Os menores valores de humificação foram obtidos na área em recuperação, indicando certa fragilidade deste estoque de carbono ser perdido caso a área não seja bem manejada. Avaliou-se também o potencial da espectroscopia de emissão óptica com plasma induzido por laser (LIBS) para a quantificação de carbono total em solos. Os dados dos teores de carbono obtidos por análise elementar foram utilizados para a calibração do sistema LIBS de acordo com a textura dos solos. O coeficiente de correlação entre os valores de referência (CHNS) e os valores preditos por LIBS na validação foi de 0,87 (erro ~ 25%) para textura arenosa e 0,92 (erro ~ 16%) para textura argilosa. Este trabalho demonstrou a importância do manejo adequado para recuperação de pastagens no Brasil e o risco de se perder esse carbono para atmosfera devido à alta labilidade do carbono armazenado. As pastagens bem manejadas podem levar este sistema produtivo a patamares ambientalmente sustentáveis e contribuir para mitigação do efeito estufa. As técnicas fotônicas apresentaram potencial para determinação de carbono no solo e avaliação de sua estrutura química. Desta forma, espera-se que futuramente este tipo de instrumentação possa ser aplicada ao campo, fornecendo parâmetros que ajudem na tomada de decisões para um desenvolvimento agrícola sustentável. / Proper soil management is important to organic matter incorporation, sequestering carbon from atmosphere. This study was developed for two livestock managements: (A3) pasture recovering of Brachiaria decumbens, where nitrogen is applied, and (A4) degraded pasture of Brachiaria decumbens, with no correction. The reference system was a native forest area (Brazilian Atlantic Forest) next to the experiment. This study aims to compare the soil carbon accumulations and CO2 emissions for both productive systems. Soil carbon content and soil bulk density values were measured to calculate carbon stocks (CSs) in surface (0-30 cm) and at depth (0-100 cm) in each system, applying equivalent mass and clay content corrections. CO2 emissions were estimated using a portable system, which monitors the CO2 flux within a chamber through infrared absorption spectroscopy. The soil carbon stocks were estimated around 142 Mg ha-1 for A3 and 99 Mg ha-1 for A4, while around 115 Mg ha-1 was found in the forest. The results for CO2 emission showed a similar situation for both pasture areas. Another purpose of this study was to evaluate the soil organic matter (SOM) quality using laser-induced fluorescence spectroscopy (LIFS). An increasing of humification degree of SOM (HLIF) was observed with increasing of depth. The lowest HLIFS values were obtained for the recovering area, indicating a certain fragility of carbon stocks, which can be lost if the area is not well managed. In addition, Laser-induced breakdown spectroscopy (LIBS) was used for total soil carbon quantification. Using the data from elemental analysis, mathematical models were developed to calibrate the LIBS system for each class of soil texture. For carbon quantification, the correlation between reference technique (CHN) and LIBS prediction was 0.87 for sandy soils (error around 21%) and 0.92 for clayed soils (error around 16%). This study demonstrated the importance of pasture recovering in Brazil from the point of view of soil carbon sequestration, and the risk of this stock to be lost due to high lability of stored carbon. Well-managed pastures can lead this production system to environmentally sustainable levels and contribute to mitigate the greenhouse effect. Systems with high potential for portability to determine soil carbon and evaluation of its chemical structure have been successfully tested. Thus, it is expected that in a near future, this kind of instrumentation can be effectively applied in the field and provide parameters that help in the decision making for a sustainable agricultural development.

L’objectif de ce travail de thèse a été de développer et de valider une méthode d’analyse quantitative du silicium liquide par la technique LIBS (Laser Induced Breakdown Spectroscopy). Pour cela, le dispositif utilisé consiste en une mesure intrusive sur la surface de bulles gazeuses, générées au cœur du métal fondu par soufflage d’un gaz inerte au travers d’un tube. Dans un premier temps, les conditions d’acquisition du signal ont été optimisées pour permettre la répétabilité et la reproductibilité du signal LIBS en dépit des fluctuations paramétriques induites par le bullage. Dans un second temps, les paramètres expérimentaux affectant les propriétés physiques des plasmas produits ont été évalués pour maximiser la sensibilité de la mesure et réduire les limites de détection. Les deux paramètres clefs qui ont été identifiés sont la densité de puissance apportées par l’impulsion laser et la nature du gaz de bullage : l’intensité du signal et donc le nombre d’émetteurs dans les plasmas sont favorisés par les hautes énergies laser et par une atmosphère d’argon plutôt que d’hélium. Le diagnostic spectroscopique des plasmas générés en fonction de ces paramètres a permis de montrer qu’ils n’ont pas d’effet significatif sur la température des électrons, des ions et des espèces neutres. L’augmentation de masse ablatée par le laser et la modification des conditions de relaxation du plasma par le milieu ambiant sont les mécanismes majoritaires à considérer. Finalement un modèle d’analyse quantitatif a été construit dans les conditions optimisées pour les impuretés B, Al, Fe, Cu et Ti du silicium. Il permet d’atteindre des limites de détection qui vont de quelques ppmw pour les éléments B, Fe et Al jusqu’à moins de 0,5 ppmw pour le Cu et le Ti. Les concentrations prédites par LIBS présentent un très bon accord avec celles déterminées par une méthode ex-situ conventionnelle, l’ICP-OES, avec des écarts relatifs inférieurs en moyenne à 10 %. Ceci démontre l’intérêt de la technique LIBS pour contrôler in-situ et en temps réel un procédé d’élaboration du silicium photovoltaïque. / The aim of this PhD research work was to develop and to validate a quantitative method to measure the composition of molten silicon by Laser Induced Breakdown Spectroscopy (LIBS). The device used consists in an intrusive measurement on the surface of gas bubbles which are produced by blowing an inert gas through a pipe inserted into the melt. First, the signal acquisition conditions were optimized to ensure LIBS signal repeatability and reproducibility to overcome experimental fluctuations due to the bubbling. Second, the experimental parameters affecting plasma physics were evaluated to maximize the measurement sensitivity and to lower the limits of detection. The two key parameters identified are the laser power density and the ambient gas: the signal intensity and then the number of emitters inside the plasma are promoted by higher laser energy and an Ar atmosphere rather than a He one. The plasma spectroscopic diagnosis as a function of these parameters showed that they don’t have significant effect on the temperature of electrons, ions and neutral species. The only mechanisms to be considered are then the increase of the mass ablated by the laser and the modification of the plasma relaxation conditions by the ambient atmosphere. Third, calibration curves were established for B, Al, Fe, Cu and Ti impurities under the optimized conditions. Limits of detections from few ppmw for B, Fe and Al to less than 0,5 ppmw for Ti and Cu have been achieved. The predicted concentrations by LIBS are in very good agreement with the ones measured by the conventional ex-situ method, ICP-OES. The mean relative discrepancy is lower than 10 %. These results demonstrate the LIBS benefits for in-situ and in-line monitoring of photovoltaic silicon production processes.

La LIBS est une technique d’analyse élémentaire dont la rapidité d’exécution et la portabilité en font une technique idéale pour l’analyse des matériaux sur site. Elle a déjà largement fait ses preuves dans le cadre de l’analyse de matériaux inorganiques. Mais les molécules organiques usuelles présentent des compositions atomiques similaires (C, O, H, N), rendant leur identification en LIBS difficile. L’utilisation de traitements de données tels que les rapports de raies ou la chimiométrie est donc inéluctable pour séparer efficacement plusieurs types de molécules organiques en LIBS. Malgré le caractère élémentaire de cette technique, les spectres obtenus lors de l’analyse de matériaux organiques montrent la signature spectrale de molécules diatomiques C2 et CN. Longtemps considéré comme des recombinaisons aléatoires des atomes présents dans le plasma, le signal de ces molécules s’avère d´dépendant de la nature des liaisons moléculaires de la molécule d’origine. Ce travail de thèse s’oriente autour de trois grands axes d’études - L’utilisation de traitements de données tels que la chimiométrie pour établir un protocole de traitements optimal visant à la discrimination des polymères dans le cadre du tri des plastiques. Un protocole optimisé a pu être avancé en UV et en IR. - L’identification des matières organiques utilisées en peinture murale (liants, adhésifs et consolidants) a pu être réalisée sur des échantillons de peinture de référence. Les traitements de données utilisés précédemment ont pu être appliqués sur ces échantillons présentant une matrice majoritairement inorganique. - La compréhension des procédés de formation des molécules diatomiques dans le plasma pour améliorer leur utilisation et ainsi optimiser la discrimination des matériaux organiques sur la base du signal moléculaire en LIBS. Le développement d’un système d’imagerie a permis de suivre le comportement spatio-temporel des espèces atomiques et moléculaires dans le plasma sous régimes UV et IR. Une hypothèse de formation des molécules sous ablation UV et IR a pu être établie. / LIBS is an analytical technique usable on-site because of its speed ant its transportability. It has already been used many times for the characterization of inorganic materials. The problematic of organic materials identification is still a subject of many researches, particularly concerning the plastic sorting and the conservation of artworks. Usual organic molecules indeed present similar atomic compositions (C, O, H, N) that make difficult their discrimination by LIBS. The use of data treatments like line ratios or chemometric tools is clearly primordial to separate effectively several organic molecules.Over the elementary nature of LIBS analysis, spectra obtained with this technique from organic materials present molecular signals belonging to diatomic molecules like C2 or CN. Those molecules have been considered as being the result of random recombinations of atoms present in the plasma. But their signals appear to be dependent of the molecular nature of the analyzed organic materials. This PhD work has been based on three studies ; - The use of data treatments like chemometric tools in order to establish a treatment protocol for the discrimination of polymers. An optimized protocol was validated for UV and IR impulsions. - The identification of organic materials used on mural paintings (binders, adhesives and consolidants) was realized on several reference samples. The data treatments used previously were successfully applied on those samples whose nature is mostly inorganic. - The comprehension of the process of formation of diatomic molecules in order to optimize the discrimination of organic materials based on molecular signals obtained by LIBS. A system was developed for the imagery of laser-induced plasmas to detect the spatio-temporal behavior of atomic and molecular species under UV and IR ablation. A hypothesis about their formation in the plasmas was advanced.

Nos últimos anos, a Espectrometria de Emissão Atômica com Plasma induzido por Laser (LIBS) vem despertando grande interesse e pesquisas voltadas às aplicações analíticas dessa técnica tem aumentado significativamente. No entanto, sua aplicação na análise de amostras líquidas não é trivial, podendo haver comprometimento na sensibilidade e na precisão do método. Além disso, a análise quantitativa é considerada o \"calcanhar de Aquiles\" da técnica, especialmente devido à falta de materiais de referência certificados e à forte interferência de matriz. Nesse contexto, com a intenção de viabilizar a análise de matrizes líquidas por LIBS, esse trabalho propõe o uso de cerâmica como fase sólida para extração, pré-concentração e especiação de Cr em amostras de água. Foi realizado também um estudo de otimização dos parâmetros instrumentais do LIBS para o desenvolvimento de um método de medição de Cr na fase sólida, empregando cerâmica enriquecida com concentrações crescentes desse analito como material calibrante sintético. Os resultados obtidos mostraram que a cerâmica adsorve seletivamente a espécie Cr(III) em pH 7, sendo este um processo rápido, de baixo custo, cuja eficiência é pouco afetada pela presença de outros eletrólitos. Os estudos relacionados à otimização dos parâmetros instrumentais do LIBS indicaram que um diâmetro de focalização de 65 µm, um tempo de atraso de 0,75 µs, a energia por pulso de 20 mJ, uma taxa de repetição de 10 Hz e um número de pulsos igual a 403 podem ser recomendados para a quantificação de Cr no material cerâmico. Nessas condições, obteve-se uma curva de calibração analítica que possibilitou a medição Cr adsorvido em cerâmica. Uma amostra de água de torneira foi então enriquecida com 0,9 mg L-1 de Cr(III) e 0,9 mg L-1 de Cr(VI) para que um teste de recuperação fosse realizado com a finalidade de se verificar a aplicabilidade do método proposto, foi possível constatar a adsorção seletiva da espécie Cr(III), com recuperação de 105±2%. A quantificação da espécie Cr(VI) por LIBS também é possível após a redução dessa espécie a Cr(III) utilizando Fe(II) como agente redutor. Os limites de detecção (LOD) e quantificação (LOQ) estimados para o método foram respectivamente 21 mg kg-1 e 70 mg kg-1, equivalentes a 0,1 mg L-1 e 0,35 mg L-1, considerando-se o coeficiente de pré-concentração igual a 200 (m/m). / In recent years, Laser Induced Breakdown Spectroscopy (LIBS) has deserved attention of scientific community and researches using this technique have increased significantly. However, its application in the analysis of liquid samples is not trivial, because the sensitivity and accuracy of the method can be affected. In addition the quantitative analysis is considered the \"Achilles\' heel\" of the technique, especially due to the lack of certified reference materials and to the strong interference matrix. To enable the analysis of liquid matrices by LIBS, this work proposes the use of ceramic as solid phase in the extraction, preconcentration and speciation of Cr in water samples. LIBS instrumental parameters were also optimized to allow the measurement of Cr in this solid phase, for the development of a method for measuring the analyte in the solid phase, which was also used as synthetic calibrating material by the addition of increasing concentrations of the analyte. Results showed the ceramic selectively adsorbs Cr(III) species at pH 7, which is a low cost and rapid process, whose efficiency is not significantly affected by the presence of other electrolytes. Studies related to the optimization of LIBS instrumental parameters indicated that spot size of 65 µm, delay time of 0,75 µs , energy per pulse of 20 mJ, repetition rate of 10 Hz and a number of pulses equal to 403 can be recommended for the quantification of Cr the ceramic material. Under these conditions, it was obtained an analytical calibration curve which allowed a Cr measurement in ceramics. A tap water sample was enriched with 0.9 mg L-1 of Cr(III) and 0.9 mg L-1 of Cr (VI) and a recovery test was done in order to verify the applicability of the proposed method. It was observed the selective adsorption of the Cr(III) species, with recovery of 105 ± 2 %. The quantification of Cr(VI) species by LIBS was also possible, after reduction of this species to Cr(III) using Fe(II) as reducing agent. The limit of detection (LOD) and the limit of quantification (LOQ) obtained by the proposed method were respectively, 21 mg kg-1 and 70 mg kg-1, equivalent to 0.1 mg L-1 and 0.35 mg L-1, considering the preconcentration coefficient of 200 (on a w/w basis).