Introdução - A perda de estrutura dental devido às lesões cervicais não cariosas (LCNC) é um evento clínico recorrente e amplamente discutido na literatura. Os fatores etiológicos relacionados a esse tipo de lesão são: abrasão, erosão e abfração. Segundo vários autores, a particular disposição das estruturas dentinárias torna-se ainda mais interessante e curiosa quando se observa a dentina nas regiões com LCNC. Objetivos - A proposta deste trabalho foi estudar, comparativamente, em microscópio eletrônico de varredura (M.E.V.), a microestrutura dentinária de LCNC, a partir de dentes extraídos e armazenados sob condições controladas, considerando-se o método de preparação dos espécimes. Também se propôs a comparar o conteúdo intratubular da dentina com LCNC com sua contraparte lingual, sem lesão, e observar se a presença de estruturas intratubulares (EIT) tem relação com a ocorrência de hiperestesia dentinária (HD). Materiais e Métodos - Utilizaram-se as superfícies vestibulares e linguais homólogas de 14 dentes humanos, portadores de LCNC, extraídos por razões terapêuticas, e imediatamente fixados em Karnovsky. Os dentes foram divididos em 3 grupos, cada um deles subdivido em dois subgrupos, de acordo com o tratamento dos espécimes. Amostras da região da lesão e suas contrapartes linguais foram, então, pós-fixadas segundo protocolo apropriado para M.E.V. Tanto a superfície da lesão, quanto as estruturas dentinárias subjacentes foram estudadas obedecendo à seguinte seqüência: Grupo I - amostras da lesão seccionadas com disco, subdivididas em: Grupo IA - limpeza com ultra-som por 15 minutos; Grupo IB - limpeza com ultra-som por 15 minutos, seguida de condicionamento com H3PO4 a 5%, por 15 segundos. Grupos II - amostras fraturadas, subdivididas em: Grupo IIA - limpeza com ultra-som por 30 minutos; Grupo IIB - apenas condicionamento com H3PO4 a 5%, por 15 segundos; e Grupo III - amostras fraturadas e assim subdivididas: Grupo IIIA - controle, sem qualquer tipo de tratamento; Grupo IIIB - imersão dos espécimes em NaOCl a 2%, por 5 minutos. Os espécimes foram desidratados, submetidos a secagem em ponto crítico e metalizados com ouro para análise em M.E.V. Resultados - 1 - De maneira geral, todos os métodos interferem, em maior ou menor proporção, com a natureza dos espécimes. Na superfície da lesão, o método que melhor permitiu a visualização das estruturas dentinárias foi o ultra-som seguido do condicionamento ácido (Grupo IIA). Entretanto, é o procedimento que mais altera a estrutura da dentina. Na dentina abaixo da lesão, o método de observação mais apropriado foi o da simples fratura controle (Grupo IIIA). 2 - Tanto a dentina correspondente à região da lesão, quanto à da região não lesionada lingual, apresentaram conteúdos intratubulares semelhantes. 3 ? Não foram notadas diferenças estruturais nas lesões hipersensíveis em comparação com aquelas não-sensíveis. Conclusões - Os métodos de preparação dos espécimes tendem, em maior ou menor extensão, a alterar a natureza do que está sendo observado. Não se pode afirmar, à luz da metodologia empregada, que a existência de EIT é ocorrência particular da dentina em LCNC. Aparentemente, as EITs ocorrem tanto na dentina que apresenta HD como naquela não sensível. / Introduction - The lost of dental structure because of non carious cervical lesion is one of the most important clinical event discussed in dental literature. The most important etiologic factors related to this type of lesions are: abrasion, erosion and abfraction. According to many authors, this particular lost of dental structures became more interesting and curious when the dentin in regions with NCCL is study microscopically. Objectives - The purpose of this investigation was to study, comparatively, in scanning electronic microscopy (S.E.M.), the dental microstructure of NCCL and lesion-dentin dentin surfaces. The null hypothesis were the following: 1- the methods used for specimens preparations lead to different changes on dentin microstructures; 2- there is no difference between NCCL dentin microstructure and that from lesion-free dentin in the same tooth; 3- ITS can be found both in hypersensitive and non-sensitive dentin. Materials and Method - Dentin samples from buccal and lingual surfaces of 14 extracted human teeth, with NCCL on buccal surface, were used. After extraction the teeth were immediately fixed with Karnovsky. Teeth were than, divided in 2 subgroups, in accordance with specimen\'s treatment. Samples of region of lesion and their lingual counterpart were, then, post fixed according to appropriated protocol for SEM. The following groups were studied: Group I - specimens sectioned with diamond disc and subdivided in: Group IA- specimens cleaned in ultrasonic bath for 15 minutes; Group IB - cleaned in ultrasonic bath 15 minutes, followed by etching with 5%, H3PO4, for 15 seconds. Group II - samples were fractured along the lesions axis and subdivided in: Group IIA - ultrasonic bath for 30 minutes; Group IIB -etching with 5%, H3PO4, for 15 seconds. Group III - samples also fractured and subdivided in: Group IIIA (control) - without any kind of treatment; Group IIIB - immersion of the specimens in NaOCl to 2%, for 5 minutes. The specimens were dehydrated, critical point dried and coated with gold for in SEM analysis. Results - 1- On general, all of methods interfere, in larger or lesser proportion, with the nature of the specimens. On the surface of the lesions, the method that better allowed the visualization and recognition of dentine structures was the ultrasound followed by acid conditioning (Group IB). However, this was the procedure that introduced the greatest changes on dentine structures. In dentine under of the lesion, the best method of specimens preparation was the control group (Group IIIA). 2- Both buccal NCCL and lesion-free lingual surfaces presented similar ITS. 3- No structural differences were found in dentin from hypersensitive and non-sensitive lesions. Conclusions - The methods used for specimens tend to modify the characteristics and relationship of dentin structures. The hypothesis that ITS are specific from NCCL could not be confirmed under the light of method used. Apparently, the ITS can be found in both hypersensitive and non-hypersensitive dentin.

Introdução - A perda de estrutura dental devido às lesões cervicais não cariosas (LCNC) é um evento clínico recorrente e amplamente discutido na literatura. Os fatores etiológicos relacionados a esse tipo de lesão são: abrasão, erosão e abfração. Segundo vários autores, a particular disposição das estruturas dentinárias torna-se ainda mais interessante e curiosa quando se observa a dentina nas regiões com LCNC. Objetivos - A proposta deste trabalho foi estudar, comparativamente, em microscópio eletrônico de varredura (M.E.V.), a microestrutura dentinária de LCNC, a partir de dentes extraídos e armazenados sob condições controladas, considerando-se o método de preparação dos espécimes. Também se propôs a comparar o conteúdo intratubular da dentina com LCNC com sua contraparte lingual, sem lesão, e observar se a presença de estruturas intratubulares (EIT) tem relação com a ocorrência de hiperestesia dentinária (HD). Materiais e Métodos - Utilizaram-se as superfícies vestibulares e linguais homólogas de 14 dentes humanos, portadores de LCNC, extraídos por razões terapêuticas, e imediatamente fixados em Karnovsky. Os dentes foram divididos em 3 grupos, cada um deles subdivido em dois subgrupos, de acordo com o tratamento dos espécimes. Amostras da região da lesão e suas contrapartes linguais foram, então, pós-fixadas segundo protocolo apropriado para M.E.V. Tanto a superfície da lesão, quanto as estruturas dentinárias subjacentes foram estudadas obedecendo à seguinte seqüência: Grupo I - amostras da lesão seccionadas com disco, subdivididas em: Grupo IA - limpeza com ultra-som por 15 minutos; Grupo IB - limpeza com ultra-som por 15 minutos, seguida de condicionamento com H3PO4 a 5%, por 15 segundos. Grupos II - amostras fraturadas, subdivididas em: Grupo IIA - limpeza com ultra-som por 30 minutos; Grupo IIB - apenas condicionamento com H3PO4 a 5%, por 15 segundos; e Grupo III - amostras fraturadas e assim subdivididas: Grupo IIIA - controle, sem qualquer tipo de tratamento; Grupo IIIB - imersão dos espécimes em NaOCl a 2%, por 5 minutos. Os espécimes foram desidratados, submetidos a secagem em ponto crítico e metalizados com ouro para análise em M.E.V. Resultados - 1 - De maneira geral, todos os métodos interferem, em maior ou menor proporção, com a natureza dos espécimes. Na superfície da lesão, o método que melhor permitiu a visualização das estruturas dentinárias foi o ultra-som seguido do condicionamento ácido (Grupo IIA). Entretanto, é o procedimento que mais altera a estrutura da dentina. Na dentina abaixo da lesão, o método de observação mais apropriado foi o da simples fratura controle (Grupo IIIA). 2 - Tanto a dentina correspondente à região da lesão, quanto à da região não lesionada lingual, apresentaram conteúdos intratubulares semelhantes. 3 ? Não foram notadas diferenças estruturais nas lesões hipersensíveis em comparação com aquelas não-sensíveis. Conclusões - Os métodos de preparação dos espécimes tendem, em maior ou menor extensão, a alterar a natureza do que está sendo observado. Não se pode afirmar, à luz da metodologia empregada, que a existência de EIT é ocorrência particular da dentina em LCNC. Aparentemente, as EITs ocorrem tanto na dentina que apresenta HD como naquela não sensível. / Introduction - The lost of dental structure because of non carious cervical lesion is one of the most important clinical event discussed in dental literature. The most important etiologic factors related to this type of lesions are: abrasion, erosion and abfraction. According to many authors, this particular lost of dental structures became more interesting and curious when the dentin in regions with NCCL is study microscopically. Objectives - The purpose of this investigation was to study, comparatively, in scanning electronic microscopy (S.E.M.), the dental microstructure of NCCL and lesion-dentin dentin surfaces. The null hypothesis were the following: 1- the methods used for specimens preparations lead to different changes on dentin microstructures; 2- there is no difference between NCCL dentin microstructure and that from lesion-free dentin in the same tooth; 3- ITS can be found both in hypersensitive and non-sensitive dentin. Materials and Method - Dentin samples from buccal and lingual surfaces of 14 extracted human teeth, with NCCL on buccal surface, were used. After extraction the teeth were immediately fixed with Karnovsky. Teeth were than, divided in 2 subgroups, in accordance with specimen\'s treatment. Samples of region of lesion and their lingual counterpart were, then, post fixed according to appropriated protocol for SEM. The following groups were studied: Group I - specimens sectioned with diamond disc and subdivided in: Group IA- specimens cleaned in ultrasonic bath for 15 minutes; Group IB - cleaned in ultrasonic bath 15 minutes, followed by etching with 5%, H3PO4, for 15 seconds. Group II - samples were fractured along the lesions axis and subdivided in: Group IIA - ultrasonic bath for 30 minutes; Group IIB -etching with 5%, H3PO4, for 15 seconds. Group III - samples also fractured and subdivided in: Group IIIA (control) - without any kind of treatment; Group IIIB - immersion of the specimens in NaOCl to 2%, for 5 minutes. The specimens were dehydrated, critical point dried and coated with gold for in SEM analysis. Results - 1- On general, all of methods interfere, in larger or lesser proportion, with the nature of the specimens. On the surface of the lesions, the method that better allowed the visualization and recognition of dentine structures was the ultrasound followed by acid conditioning (Group IB). However, this was the procedure that introduced the greatest changes on dentine structures. In dentine under of the lesion, the best method of specimens preparation was the control group (Group IIIA). 2- Both buccal NCCL and lesion-free lingual surfaces presented similar ITS. 3- No structural differences were found in dentin from hypersensitive and non-sensitive lesions. Conclusions - The methods used for specimens tend to modify the characteristics and relationship of dentin structures. The hypothesis that ITS are specific from NCCL could not be confirmed under the light of method used. Apparently, the ITS can be found in both hypersensitive and non-hypersensitive dentin.

<p> The crystal structures of Ca3(VO4)2, Ca3(asO4)2, alpha-Zn3(VO4)2, alpha-Zn2V2O7, Mg2V2O7, VPO5, and whitlockite (Room temperature and 1200 degree C) have determined by X-ray diffraction methods. Tests on the existing theories on the prediction of bond lengths have been made on the vanadate structures determined in this work as well as other reported structures. The range of validity of these theories have been brought out. The importance of the difference of the structure of whitlockite from that of BetaCa3(PO4)2, have also been discussed in detail. </p> / Thesis / Doctor of Philosophy (PhD)

La gazéification est aujourd'hui l'une des stratégies les plus prometteuses pour valoriser les déchets en énergie. Cette technologie thermo-chimique permet une réduction de 95 % de la masse des intrants et génère des cendres inertes ainsi que du gaz de synthèse (syngaz). Le syngaz est un combustible gazeux composé principalement de monoxyde de carbone (CO), d'hydrogène (H2) et de dioxyde de carbone (CO2). Le syngaz peut être utilisé pour produire de la chaleur et de l'électricité. Il est également la pierre angulaire d'un grand nombre de produits à haute valeur ajoutée, allant de l'éthanol à l'ammoniac et l'hydrogène pur. Les applications en aval de la production de syngaz sont dictées par son pouvoir calorifique, lui-même dépendant de la teneur du gaz en H2. L’augmentation du contenu du syngaz en H2 est rendu possible par la conversion catalytique à la vapeur d’eau, largement répandu dans le cadre du reformage du méthane pour la production d'hydrogène. Au cours de cette réaction, le CO est converti en H2 et CO2 selon : CO + H2O → CO2 + H2. Ce processus est possible grâce à des catalyseurs métalliques mis en contact avec le CO et de la vapeur. La conversion catalytique à la vapeur d’eau a jusqu'ici été réservé pour de grandes installations industrielles car elle nécessite un capital et des charges d’exploitations très importantes. Par conséquent, les installations de plus petite échelle et traitant des intrants de faible qualité (biomasse, déchets, boues ...), n'ont pas accès à cette technologie. Ainsi, la seule utilisation de leur syngaz à faible pouvoir calorifique, est limitée à la génération de chaleur ou, tout au plus, d'électricité. Afin de permettre à ces installations une gamme d’application plus vaste de leurs syngaz, une alternative économique à base de catalyseur biologique est proposée par l’utilisation de bactéries hyperthermophiles hydrogénogènes. L'objectif de cette thèse est d'utiliser Carboxydothermus hydrogenoformans, une bactérie thermophile carboxydotrophe hydrogénogène comme catalyseur biologique pour la conversion du monoxyde de carbone en hydrogène. Pour cela, l’impact d'un phénomène de biominéralisation sur la production d’H2 a été étudié. Ensuite, la faisabilité et les limites de l’utilisation de la souche dans un bioréacteur ont été évaluées. Tout d'abord, la caractérisation de la phase inorganique prédominante lorsque C. hydrogenoformans est inoculé dans le milieu DSMZ, a révélé une biominéralisation de phosphate de calcium (CaP) cristallin en deux phases. L’analyse par diffraction des rayons X et spectrométrie infrarouge à transformée de Fourier de ce matériau biphasique indique une signature caractéristique de la Mg-whitlockite, alors que les images obtenues par microscopie électronique à transmission ont montré l'existence de nanotiges cristallines s’apparentant à de l’hydroxyapatite. Dans les deux cas, le mode de biominéralisation semble être biologiquement induit plutôt que contrôlé. L'impact du précipité de CaP endogène sur le transfert de masse du CO et la production d’H2 a ensuite été étudié. Les résultats ont été comparés aux valeurs obtenues dans un milieu où aucune précipitation n'est observée. Dans le milieu DSMZ, le KLa apparent (0.22 ± 0.005 min-1) et le rendement de production d’H2 (89.11 ± 6.69 %) étaient plus élevés que ceux obtenus avec le milieu modifié (0.19 ± 0.015 min-1 et 82.60 ± 3.62% respectivement). La présence du précipité n'a eu aucune incidence sur l'activité microbienne. En somme, le précipité de CaP offre une nouvelle stratégie pour améliorer les performances de transfert de masse du CO en utilisant les propriétés hydrophobes de gaz. En second lieu, la conversion du CO en H2 par la souche Carboxydothermus hydrogenoformans fut étudiée et optimisée dans un réacteur gazosiphon de 35 L. Parmi toutes les conditions opérationnelles, le paramètre majeur fut le ratio du débit de recirculation du gaz sur le débit d'alimentation en CO (QR:Qin). Ce ratio impacte à la fois l'activité biologique et le taux de transfert de masse gaz-liquide. En effet, au dessus d’un ratio de 40, les performances de conversion du CO en H2 sont limitées par l’activité biologique alors qu’en dessous, elles sont limitées par le transfert de masse. Cela se concrétise par une efficacité de conversion maximale de 90.4 ± 0.3 % et une activité spécifique de 2.7 ± 0.4 molCO·g–1VSS·d–1. Malgré des résultats prometteurs, les performances du bioréacteur ont été limitées par une faible densité cellulaire, typique de la croissance planctonique de C. hydrogenoformans. Cette limite est le facteur le plus contraignant pour des taux de charge de CO plus élevés. Ces performances ont été comparées à celles obtenues dans un réacteur à fibres creuses (BRFC) inoculé par la souche. En dépit d’une densité cellulaire et d’une activité volumétrique plus élevées, les performances du BRFC à tout le moins cinétiquement limitées quand elles n’étaient pas impactées par le transfert de masse, l'encrassement et le vieillissement de la membrane. Afin de parer à la dégénérescence de C. hydrogenoformans en cas de pénurie de CO, la croissance de la bactérie sur pyruvate en tant que seule source de carbone a été également caractérisée. Fait intéressant, en présence simultanée de pyruvate et de CO, C. hydrogenoformans n’a amorcé la consommation de pyruvate qu’une fois le CO épuisé. Cela a été attribué à un mécanisme d'inhibition du métabolisme du pyruvate par le CO, faisant ainsi du pyruvate le candidat idéal pour un système in situ de secours. / Gasification is today one of the most promising strategies to recover energy from waste. This thermo-chemical technology allows a 95% weight reduction of the input and generates inorganic inert ashes as well as a synthesis gas (syngas). Syngas is a gaseous fuel mainly composed of carbon monoxide (CO), hydrogen (H2) and carbon dioxide (CO2). Syngas can be burned to produce heat and electricity. It is also the building block of many high added- value products ranging from ethanol to ammonia and pure hydrogen. Downstream applications of syngas production will depend on its heating value, which is determined by its content in H2. Upgrading the H2 content in syngas is performed by the water-gas shift (WGS) reaction, widely utilized during methane reforming for hydrogen production. During the WGS reaction CO is converted to H2 and CO2 according to: CO + H2O → CO2 + H2. This process is achieved using a metallic catalyst in a heterogeneous gas-phase reaction with CO and steam. The WGS reaction has so far been reserved for large-scale gasification plants and requires high capital and operational expenditures. Hence, smaller scale plants that process low-grade materials (biomass, waste, sludge...), would not have access to such technology. The only possible outcome with the synthesis gas (syngas) produced and which generally has a poor heating value, is to generate heat or at best, electricity. In order to offer small plants access to the WGS reaction and to a higher range of products from their syngas, an alternative to the expensive and energy-intensive established catalyst-based WGS is here considered, such as extreme-thermophilic microbial processes carried out by hydrogenogens. The goal of this thesis was to use Carboxydothermus hydrogenoformans, a thermophilic carboxydotrophic hydrogenogenic bacterium as a biological catalyst for the WGS reaction. This was done by characterizing the impact of a growth-associated biomineralization phenomenon on H2 production and assessing the feasibility and limitations of using the strain in a bioreactor. First, characterization of the predominant inorganic phase when Carboxydothermus hydrogenoformans was inoculated in the DSMZ medium revealed the biomineralization of two crystalline CaP phases. The X-ray diffractometry peaks and Fourier transform infrared spectroscopy spectrum of this biphasic material consistently showed features characteristic of Mg-whitlockite, whereas transmission electron microscopy analysis showed the existence of hydroxyapatite-like nanorods crystals. In both cases, the mode of biomineralization appears to be biologically induced rather than biologically controlled. The impact of the endogenous CaP precipitate on CO mass transfer and H2 production was thus assessed and compared to a medium where no precipitation was observed. In the DSMZ medium, the apparent KLa (0.22 ±0.005 min-1) and H2 production yield (89.11 ±6.69%) were higher than the ones obtained in the modified medium (0.19 ±0.015 min-1 and 82.60 ±3.62% respectively). The presence of the precipitate had no impact on C. hydrogenoformans CO uptake. Overall, the CaP precipitate offers a novel strategy for gas-liquid mass transfer enhancement using CO hydrophobic properties. Second, the conversion of CO into H2 by C. hydrogenoformans was investigated and optimized in a 35 L gas-lift reactor. Upon all operational conditions, the ratio of gas recirculation over CO feed flow rates (QR:Qin) was the major parameter that impacted both biological activity and volumetric gas-liquid mass transfer. The CO conversion performance of the gas lift reactor was kinetically limited over a QR:Qin ratio of 40, and mass transfer limited below that ratio, resulting in a maximum conversion efficiency of 90.4±0.3% and a biological activity of 2.7±0.4 molCO· g–1VSS· day–1. Despite very promising results, CO conversion performance was limited by a low cell density, typical of C. hydrogenoformans planktonic growth. This limitation was found to be the most restrictive factor for higher CO loading rates. Results were compared to the performance of the strain inoculated in a hollow fiber membrane bioreactor where performance, despite the higher cell density and volumetric activity, was biokinetically limited, when not limited by gas–liquid mass transfer, membrane fouling and aging. To avoid any C. hydrogenoformans decay during potential CO shortages, growth of the bacterium on pyruvate as a sole carbon source was characterized. Interestingly, when grown simultaneously on pyruvate and CO, pyruvate consumption was initiated upon CO depletion. This was attributed to the inhibition of pyruvate oxidation by CO, making pyruvate the ideal candidate for an in-situ back-up system.

La gazéification est aujourd'hui l'une des stratégies les plus prometteuses pour valoriser les déchets en énergie. Cette technologie thermo-chimique permet une réduction de 95 % de la masse des intrants et génère des cendres inertes ainsi que du gaz de synthèse (syngaz). Le syngaz est un combustible gazeux composé principalement de monoxyde de carbone (CO), d'hydrogène (H2) et de dioxyde de carbone (CO2). Le syngaz peut être utilisé pour produire de la chaleur et de l'électricité. Il est également la pierre angulaire d'un grand nombre de produits à haute valeur ajoutée, allant de l'éthanol à l'ammoniac et l'hydrogène pur. Les applications en aval de la production de syngaz sont dictées par son pouvoir calorifique, lui-même dépendant de la teneur du gaz en H2. L’augmentation du contenu du syngaz en H2 est rendu possible par la conversion catalytique à la vapeur d’eau, largement répandu dans le cadre du reformage du méthane pour la production d'hydrogène. Au cours de cette réaction, le CO est converti en H2 et CO2 selon : CO + H2O → CO2 + H2. Ce processus est possible grâce à des catalyseurs métalliques mis en contact avec le CO et de la vapeur. La conversion catalytique à la vapeur d’eau a jusqu'ici été réservé pour de grandes installations industrielles car elle nécessite un capital et des charges d’exploitations très importantes. Par conséquent, les installations de plus petite échelle et traitant des intrants de faible qualité (biomasse, déchets, boues ...), n'ont pas accès à cette technologie. Ainsi, la seule utilisation de leur syngaz à faible pouvoir calorifique, est limitée à la génération de chaleur ou, tout au plus, d'électricité. Afin de permettre à ces installations une gamme d’application plus vaste de leurs syngaz, une alternative économique à base de catalyseur biologique est proposée par l’utilisation de bactéries hyperthermophiles hydrogénogènes. L'objectif de cette thèse est d'utiliser Carboxydothermus hydrogenoformans, une bactérie thermophile carboxydotrophe hydrogénogène comme catalyseur biologique pour la conversion du monoxyde de carbone en hydrogène. Pour cela, l’impact d'un phénomène de biominéralisation sur la production d’H2 a été étudié. Ensuite, la faisabilité et les limites de l’utilisation de la souche dans un bioréacteur ont été évaluées. Tout d'abord, la caractérisation de la phase inorganique prédominante lorsque C. hydrogenoformans est inoculé dans le milieu DSMZ, a révélé une biominéralisation de phosphate de calcium (CaP) cristallin en deux phases. L’analyse par diffraction des rayons X et spectrométrie infrarouge à transformée de Fourier de ce matériau biphasique indique une signature caractéristique de la Mg-whitlockite, alors que les images obtenues par microscopie électronique à transmission ont montré l'existence de nanotiges cristallines s’apparentant à de l’hydroxyapatite. Dans les deux cas, le mode de biominéralisation semble être biologiquement induit plutôt que contrôlé. L'impact du précipité de CaP endogène sur le transfert de masse du CO et la production d’H2 a ensuite été étudié. Les résultats ont été comparés aux valeurs obtenues dans un milieu où aucune précipitation n'est observée. Dans le milieu DSMZ, le KLa apparent (0.22 ± 0.005 min-1) et le rendement de production d’H2 (89.11 ± 6.69 %) étaient plus élevés que ceux obtenus avec le milieu modifié (0.19 ± 0.015 min-1 et 82.60 ± 3.62% respectivement). La présence du précipité n'a eu aucune incidence sur l'activité microbienne. En somme, le précipité de CaP offre une nouvelle stratégie pour améliorer les performances de transfert de masse du CO en utilisant les propriétés hydrophobes de gaz. En second lieu, la conversion du CO en H2 par la souche Carboxydothermus hydrogenoformans fut étudiée et optimisée dans un réacteur gazosiphon de 35 L. Parmi toutes les conditions opérationnelles, le paramètre majeur fut le ratio du débit de recirculation du gaz sur le débit d'alimentation en CO (QR:Qin). Ce ratio impacte à la fois l'activité biologique et le taux de transfert de masse gaz-liquide. En effet, au dessus d’un ratio de 40, les performances de conversion du CO en H2 sont limitées par l’activité biologique alors qu’en dessous, elles sont limitées par le transfert de masse. Cela se concrétise par une efficacité de conversion maximale de 90.4 ± 0.3 % et une activité spécifique de 2.7 ± 0.4 molCO·g–1VSS·d–1. Malgré des résultats prometteurs, les performances du bioréacteur ont été limitées par une faible densité cellulaire, typique de la croissance planctonique de C. hydrogenoformans. Cette limite est le facteur le plus contraignant pour des taux de charge de CO plus élevés. Ces performances ont été comparées à celles obtenues dans un réacteur à fibres creuses (BRFC) inoculé par la souche. En dépit d’une densité cellulaire et d’une activité volumétrique plus élevées, les performances du BRFC à tout le moins cinétiquement limitées quand elles n’étaient pas impactées par le transfert de masse, l'encrassement et le vieillissement de la membrane. Afin de parer à la dégénérescence de C. hydrogenoformans en cas de pénurie de CO, la croissance de la bactérie sur pyruvate en tant que seule source de carbone a été également caractérisée. Fait intéressant, en présence simultanée de pyruvate et de CO, C. hydrogenoformans n’a amorcé la consommation de pyruvate qu’une fois le CO épuisé. Cela a été attribué à un mécanisme d'inhibition du métabolisme du pyruvate par le CO, faisant ainsi du pyruvate le candidat idéal pour un système in situ de secours. / Gasification is today one of the most promising strategies to recover energy from waste. This thermo-chemical technology allows a 95% weight reduction of the input and generates inorganic inert ashes as well as a synthesis gas (syngas). Syngas is a gaseous fuel mainly composed of carbon monoxide (CO), hydrogen (H2) and carbon dioxide (CO2). Syngas can be burned to produce heat and electricity. It is also the building block of many high added- value products ranging from ethanol to ammonia and pure hydrogen. Downstream applications of syngas production will depend on its heating value, which is determined by its content in H2. Upgrading the H2 content in syngas is performed by the water-gas shift (WGS) reaction, widely utilized during methane reforming for hydrogen production. During the WGS reaction CO is converted to H2 and CO2 according to: CO + H2O → CO2 + H2. This process is achieved using a metallic catalyst in a heterogeneous gas-phase reaction with CO and steam. The WGS reaction has so far been reserved for large-scale gasification plants and requires high capital and operational expenditures. Hence, smaller scale plants that process low-grade materials (biomass, waste, sludge...), would not have access to such technology. The only possible outcome with the synthesis gas (syngas) produced and which generally has a poor heating value, is to generate heat or at best, electricity. In order to offer small plants access to the WGS reaction and to a higher range of products from their syngas, an alternative to the expensive and energy-intensive established catalyst-based WGS is here considered, such as extreme-thermophilic microbial processes carried out by hydrogenogens. The goal of this thesis was to use Carboxydothermus hydrogenoformans, a thermophilic carboxydotrophic hydrogenogenic bacterium as a biological catalyst for the WGS reaction. This was done by characterizing the impact of a growth-associated biomineralization phenomenon on H2 production and assessing the feasibility and limitations of using the strain in a bioreactor. First, characterization of the predominant inorganic phase when Carboxydothermus hydrogenoformans was inoculated in the DSMZ medium revealed the biomineralization of two crystalline CaP phases. The X-ray diffractometry peaks and Fourier transform infrared spectroscopy spectrum of this biphasic material consistently showed features characteristic of Mg-whitlockite, whereas transmission electron microscopy analysis showed the existence of hydroxyapatite-like nanorods crystals. In both cases, the mode of biomineralization appears to be biologically induced rather than biologically controlled. The impact of the endogenous CaP precipitate on CO mass transfer and H2 production was thus assessed and compared to a medium where no precipitation was observed. In the DSMZ medium, the apparent KLa (0.22 ±0.005 min-1) and H2 production yield (89.11 ±6.69%) were higher than the ones obtained in the modified medium (0.19 ±0.015 min-1 and 82.60 ±3.62% respectively). The presence of the precipitate had no impact on C. hydrogenoformans CO uptake. Overall, the CaP precipitate offers a novel strategy for gas-liquid mass transfer enhancement using CO hydrophobic properties. Second, the conversion of CO into H2 by C. hydrogenoformans was investigated and optimized in a 35 L gas-lift reactor. Upon all operational conditions, the ratio of gas recirculation over CO feed flow rates (QR:Qin) was the major parameter that impacted both biological activity and volumetric gas-liquid mass transfer. The CO conversion performance of the gas lift reactor was kinetically limited over a QR:Qin ratio of 40, and mass transfer limited below that ratio, resulting in a maximum conversion efficiency of 90.4±0.3% and a biological activity of 2.7±0.4 molCO· g–1VSS· day–1. Despite very promising results, CO conversion performance was limited by a low cell density, typical of C. hydrogenoformans planktonic growth. This limitation was found to be the most restrictive factor for higher CO loading rates. Results were compared to the performance of the strain inoculated in a hollow fiber membrane bioreactor where performance, despite the higher cell density and volumetric activity, was biokinetically limited, when not limited by gas–liquid mass transfer, membrane fouling and aging. To avoid any C. hydrogenoformans decay during potential CO shortages, growth of the bacterium on pyruvate as a sole carbon source was characterized. Interestingly, when grown simultaneously on pyruvate and CO, pyruvate consumption was initiated upon CO depletion. This was attributed to the inhibition of pyruvate oxidation by CO, making pyruvate the ideal candidate for an in-situ back-up system.