Este estudo propôs uma avaliação do papel dos três principais parâmetros clássicos da síntese do nitreto de carbono grafítico: temperatura final, tempo de permanência na temperatura final e taxa de aquecimento. Realizou-se a otimização da síntese, via metodologia de superfície de resposta, usando-se como variável-resposta a degradação fotocatalítica de um poluente-modelo (tartrazina). A significância estatística dos fatores foi confirmada, com 95% de confiança. Em seguida, um modelo de segunda ordem foi ajustado às melhores respostas e, no ponto de máxima degradação, as condições foram: 605oC por 183 min, com taxa de aquecimento de 5oC min-1. A taxa de degradação com o fotocatalisador sintetizado foi aproximadamente três vezes maior que a da fotólise. As amostras da região de melhores respostas foram analisadas em uma série de experimentos de caracterização, sendo eles: difratometria de raios X, espectroscopia na região do infravermelho médio, área superficial específica, microscopias de varredura (MEV e MEV-FEG), potencial zeta e espectroscopia de reflectância difusa na região do ultravioleta-visível. O fotocatalisador com maior atividade apresentou menor energia de band gap e maior área superficial especifica do que as relatadas na literatura (2,59 eV e 29,5 m2 g-1, respectivamente). Foram criadas heteroestruturas entre o fotocatalisador sintetizado e o trióxido de tungstênio. A partir de uma série de caracterizações básicas, confirmou-se a formação da heteroestrutura. Com essa heteroestrutura, a taxa de degradação foi aproximadamente cinco vezes maior que a com o nitreto de carbono grafítico. / This study proposed an assessment of the role of the three major classical parameters for synthesizing graphitic carbon nitride: final temperature, residence time at the final temperature and heating rate. The synthesis was optimized, via response surface methodology, using the photocatalytic degradation of a model pollutant (tatrazine) as the response-variable. The statistical significance of the factors was confirmed, within 95% confidence level. Afterwards, a second-order model was adjusted to the better responses and, at the maximum degradation point, the conditions were: 605oC for 183 min, with heating rate of 5oC min-1. The degradation rate with the synthetized photocatalyst was approximately three times greater than the photolytic one. The samples from the better response region were analyzed in a series of characterization experiments: X ray diffractometry, mid-infrared spectrometry, specific surface area, scanning electron microscopy (SEM and FEG-SEM), zeta potential, and ultraviolet-visible diffuse reflectance spectroscopy. The most active photocatalyst showed smaller band gap energy and greater specific surface area than the ones reported in literature (2.59 eV and 29.5 m2 g-1, respectively). Heterostructures were formed between the synthetized photocatalyst and tungsten trioxide. A series of basic characterization techniques confirmed the heterostructure formation. Using this heterostructure, the degradation rate was approximately five times greater than the one with graphitic carbon nitride.

Este estudo propôs uma avaliação do papel dos três principais parâmetros clássicos da síntese do nitreto de carbono grafítico: temperatura final, tempo de permanência na temperatura final e taxa de aquecimento. Realizou-se a otimização da síntese, via metodologia de superfície de resposta, usando-se como variável-resposta a degradação fotocatalítica de um poluente-modelo (tartrazina). A significância estatística dos fatores foi confirmada, com 95% de confiança. Em seguida, um modelo de segunda ordem foi ajustado às melhores respostas e, no ponto de máxima degradação, as condições foram: 605oC por 183 min, com taxa de aquecimento de 5oC min-1. A taxa de degradação com o fotocatalisador sintetizado foi aproximadamente três vezes maior que a da fotólise. As amostras da região de melhores respostas foram analisadas em uma série de experimentos de caracterização, sendo eles: difratometria de raios X, espectroscopia na região do infravermelho médio, área superficial específica, microscopias de varredura (MEV e MEV-FEG), potencial zeta e espectroscopia de reflectância difusa na região do ultravioleta-visível. O fotocatalisador com maior atividade apresentou menor energia de band gap e maior área superficial especifica do que as relatadas na literatura (2,59 eV e 29,5 m2 g-1, respectivamente). Foram criadas heteroestruturas entre o fotocatalisador sintetizado e o trióxido de tungstênio. A partir de uma série de caracterizações básicas, confirmou-se a formação da heteroestrutura. Com essa heteroestrutura, a taxa de degradação foi aproximadamente cinco vezes maior que a com o nitreto de carbono grafítico. / This study proposed an assessment of the role of the three major classical parameters for synthesizing graphitic carbon nitride: final temperature, residence time at the final temperature and heating rate. The synthesis was optimized, via response surface methodology, using the photocatalytic degradation of a model pollutant (tatrazine) as the response-variable. The statistical significance of the factors was confirmed, within 95% confidence level. Afterwards, a second-order model was adjusted to the better responses and, at the maximum degradation point, the conditions were: 605oC for 183 min, with heating rate of 5oC min-1. The degradation rate with the synthetized photocatalyst was approximately three times greater than the photolytic one. The samples from the better response region were analyzed in a series of characterization experiments: X ray diffractometry, mid-infrared spectrometry, specific surface area, scanning electron microscopy (SEM and FEG-SEM), zeta potential, and ultraviolet-visible diffuse reflectance spectroscopy. The most active photocatalyst showed smaller band gap energy and greater specific surface area than the ones reported in literature (2.59 eV and 29.5 m2 g-1, respectively). Heterostructures were formed between the synthetized photocatalyst and tungsten trioxide. A series of basic characterization techniques confirmed the heterostructure formation. Using this heterostructure, the degradation rate was approximately five times greater than the one with graphitic carbon nitride.

Interstitial carbides and nitrides of early transition metals in Groups IV-VI exhibit platinum-like behavior which makes them a promising candidate to replace noble metals in a wide variety of reactions. Most synthetic methods used to prepare these materials lead to bulk or micron size powder which limits their use in reactions in particular in catalytic applications. Attempts toward the production of transition metal carbide and nitride nanoparticles in a sustainable, simple and cheap manner have been rapidly increasing. In this thesis, a new approach was presented to prepare nano-scale transition metal carbides and nitrides of group IV-VI with a size as small as 3 nm through the reaction of transition metal precursor with mesoporous graphitic carbon nitride (mpg-C3N4) that not only provides confined spaces for nanoparticles formation but also acts as a chemical source of nitrogen and carbon. The produced nanoparticles were characterized by powder X-ray diffraction (XRD), temperature-programmed reaction with mass spectroscopy (MS), CHN elemental analyses, thermogravimetric analyses (TGA), nitrogen sorption, X-ray photoelectron spectroscopy (XPS), and transmission electron microscopy (TEM). The effects of the reaction temperature, the ratio of the transition metal precursor to the reactive template (mpg-C3N4), and the selection of the carrier gas (Ar, N2, and NH3) on the resultant crystal phases and structures were investigated. The results indicated that different tantalum phases with cubic structure, TaN, Ta2CN, and TaC, can be formed under a flow of nitrogen by changing the reaction temperatures. Two forms of tantalum nitride, namely TaN and Ta3N5, were selectively formed under N2 and NH3 flow, respectively. Significantly, the formation of TaC, Ta2CN, and TaN can be controlled by altering the weight ratio of the C3N4 template relative to the Ta precursor at 1573 K under a flow of nitrogen where high C3N4/Ta precursor ratio generally resulted in high carbide content rather than nitride. In addition, the reactivity of the transition metals of group IV-VI with the reactive template was investigated under a flow of N2 at different temperatures in the range of 1023 to 1573 K while keeping the weight ratio constant at 1:1. The results show that Ti, V, Nb, Ta, and Cr reacted with mpg-C3N4 at 1023 K to form nitride phase with face centered cubic structure. The nitride phase destabilized at higher temperature ≥1223 K through the reaction with the remaining carbon residue originated from the decomposition of the template to form carbonitride and carbide phases. Whereas, Mo and W produce a hexagonal structure of carbide irrespective of the applying reaction temperature. The tendency to form transition metal nitrides and carbides at 1023 K was strongly driven by the free energy of formation. The observed trend indicates that the free energy of formation of nitride is relatively lower for group IV and V transition metals, whereas the carbide phase is thermodynamically more favorable for group VI, in particular for Mo and W. The thermal stability of nitride decreases at high temperature due to the evolution of nitrogen gas. The electrocatalytic activities of the produced nanoparticles were tested for hydrogen evolution reaction in acid media and the results demonstrated that molybdenum carbide nanoparticles exhibited the highest HER current with over potential of 100 mV vs. RHE, among the samples prepared in this study. This result is attributed to the sufficiently small particle size (8 nm on average) and accordingly high surface area (308 m2 g-1). Also, the graphitized carbon layer with a thickness of 1 nm on its surface formed by this synthesis provides excellent electron pathway to the catalyst which will improve the rate of electron transfer reaction.

O objetivo do presente trabalho foi a otimização, via metodologia de superfície de resposta, da síntese de compósitos de nitreto de carbono grafítico e óxido de grafeno reduzido, por via úmida, para aplicação em fotodegradação de poluentes orgânicos, empregando-se como poluente-modelo o Bisfenol A (BPA) em meio aquoso e em concentrações ambientais (100 µg L-1). O nitreto de carbono foi obtido por pirólise da melamina, enquanto o óxido de grafeno reduzido foi obtido a partir do grafite por um método de Hummers modificado e posteriormente reduzido com hidrazina em refluxo após a impregnação em g-C3N4. A impregnação se baseou em auto-montagem por interação eletrostática via tratamento sonoquímico esfoliativo em pH 3,0, em que os materiais apresentam cargas opostas. Inicialmente, três fatores foram estudados para a formação dos compósitos: o teor mássico de óxido de grafite utilizado, o tempo de sonicação da mistura das suspensões dos materiais precursores esfoliados por ultrassom de ponteira e a quantidade de hidrazina utilizada para redução do óxido de grafeno. Para tanto realizou-se um planejamento inicial de dois níveis e três fatores com réplicas em todos os pontos. Após o ajuste de um modelo linear e de se determinar região ótima percorrendo-se o caminho de máxima inclinação, um Planejamento Composto Central foi utilizado para se otimizar a reação. A quantidade de hidrazina empregada não se mostrou estatisticamente significativa nos níveis estudados. As condições ótimas de síntese foram: 16% em massa de GrO, 8,5 min de sonicação da mistura. A hidrazina foi fixada em seu nível baixo (1:4 GrO em massa). A remoção do BPA chegou a aproximadamente 65% após 0,5 h de adsorção no escuro e 1 h de irradiação, utilizando-se 0,05 g L-1 do compósito. O material obtido com as condições ótimas foi aproximadamente 2,7 vezes mais ativo que o obtido com as condições encontradas na literatura. O sistema reacional apresentou baixas fotólise (3%) e adsorção no escuro (22%). Em termos texturais, o material obtido apresenta área superficial específica maior (86 m2 g-1) que a da matriz original de g-C3N4 (26 m2 g-1) sem o tratamento sonoquímico. Para o compósito otimizado, a estrutura lamelar do g-C3N4 foi preservada no compósito conforme os perfis dos difratogramas de pó. A superfície dos precursores e do compósito mostraram-se bastante puras. As análises de DRS mostraram que as propriedades de separação de carga do compósito não diferem muito do bandgap do g-C3N4 puro (2,7 eV). / The objective of this work was the optimization of the synthesis of grafitic carbon ni-tride and reduced graphene oxide composites by Response Surface Metodology for the photodegradation of organic pollutants, using bisphenol A as a model pollutant in aque-ous medium and environmental concentrations (100 µg L-1). Carbon nitride was ob-tained by pyrolysis of melamine, while the reduced graphene oxide was obtained from grafite by a modified Hummers method. The graphene oxide formed was later reduced by hydrazine in reflux after the impregnation in g-C3N4. The synthesis was based on a self-assembly via eletrostatic interaction and exfoliative sonochemical treatment at pH 3.0, condition in which the materials presented opposite charges. Initially three factors were studied for the composites fabrication: the graphite oxide mass content, the sonication period of the precursor mixture, which was exfoliated by a sonotrode and the amount of hydrazine used for reducing graphene oxide. An initial experimental design with two levels and three factors and replicas at all points was carried out. After determining a linear model and finding the optimum region along the path of steepest ascent, a Central Composite Design was performed to optmize the reaction. The amount of hydrazine employed was not statistically significant within the levels studied. The optimum conditions of syhthesis were 16% in mass of graphite oxide and 8.5 min of soni-cation of the mixture. Hydrazine was set at its low level (1:4 to GrO in mass). The BPA removal reached approximately 65% after adsorption (30 min in the dark) and 1 h of irradiation with 0.05 g L-1 composite. The obtained material according to the optimal conditions was approximately 2.7 times more active than the one obtained using the conditions found in the literature. The reaction system presented low photolysis (3%) and adsorption (22%). Texturally speaking, the obtained material had greater specific surface area (86 m2 g-1) than the original g-C3N4 matrix (26 m2 g-1). In textural terms, the g-C3N4 lamelar structure was preserved in the composite according to the powder dif-fractograms profiles. The surface of precursors and composite was quite pure. DRS analyses showed that the charge separation properties of the composite do not signifi-cantly differ from the pure g-C3N4 bandgap (2.7 eV).

À ce jour, l’humanité est confrontée simultanément à une crise énergétique et environnementale due principalement à deux facteurs: la croissance démographique et la dépendance aux combustibles fossiles. C'est pourquoi l'urgence d'utiliser des sources d'énergie renouvelables, comme l'énergie solaire est une, solution potentielle. A ce titre, la production d'H2 décarboné par dissociation solaire de l'eau est une voie prometteuse. Néanmoins, pour atteindre l'objectif mentionné, il faut trouver un système photocatalytique (semi-conducteurs, SCs) idéal, qui nécessite quatre caractéristiques majeures: (1) une bonne capacité d'absorption de la lumière visible (2) des positions adéquates de BV et BC des SCs par rapport aux potentiels d’oxydation de l’eau et de réduction du proton (3) une utilisation efficace des photons absorbés et charges générées et (4) une bonne stabilité dans le temps. À cette fin, cette thèse contribue à la conception et à l’optimisation de trois matériaux innovants: les composites Au/g-C3N4, Au/TiO2 (P25)-gC3N4, et Au/TiO2 (NTs)-gC3N4 dont l’activité photocatalytique a été corrélé avec les propriétés physico-chimiques pour comprendre leurs performances photocatalytiques de production d'H2 sous irradiation solaire et visible. De manière annexe, certains de ces matériaux se sont également montrés performants pour les conversions du CO. Pour mettre en évidence l'efficacité des composites préparés, des études comparatives ont été testées en utilisant des références commerciales, pertinentes et les mélanges physiques correspondant. / To date, mankind is facing an energy and environmental crisis simultaneously due to mainly two factors: growth population and the dependency on fossil fuels. For this reason, the urgency of using renewables sources, e.g., solar energy, is a potential solution. For example, non-carbon based H2 production from solar light driven water photodissociation is a promising approach. Nevertheless, to target the mentioned objective, an ideal photocatalytic system (semiconductors, SCs) has to meet four main features: (1) capacity of absorption of visible-light (2) suitable VB and CB positions of SCs to undergo the two half reactions of water splitting (3) efficient use of absorbed photons and generated charges and (4) good stability over time. For this purpose, this thesis contributes to the design and optimization of three innovative materials: Au/g-C3N4, Au/TiO2 (P25)-gC3N4, and Au/TiO2 (NTs)-gC3N4 composites. Their photocatalytic activities were correlated with their physico-chemical properties. In addition some of these composites also exhibited interesting CO conversion yields. To highlight the efficiency on the as-prepared composites, comparative studies were tested using commercial, pertinent references, and physical mixtures homologs.

¨Herstellung und Charakterisierung von Kohlenstoffnitridschichten¨ Technische Universitaet Chemnitz, Institut fuer Physik, Dissertation, 1998 (121 Seiten; 61 Abbildungen; 17 Tabellen; 148 Literaturzitate) Spaeth, Christian Mit einem neuartigen Abscheidungsverfahren, der gefilterten Vakuumbogenverdampfung von Graphit in einer Stickstoffatmosphaere, kombiniert mit einer Kaufman-Ionenquelle, wurden Kohlenstoffnitridschichten abgeschieden. Es wurde untersucht, wie sich die Wahl der Schichtabscheidungsparameter auf die Schichtstruktur und -eigenschaften der erzeugten Kohlenstoffnitridschichten auswirkt. Dies geschah insbesondere im Hinblick auf die Herstellung von beta-C_3N_4. Trotz einer breiten Variation der Beschichtungsparameter konnte der Stickstoffgehalt in den Schichten nicht ueber 35 At.% angehoben werden. Die mit ERDA und Profilometrie ermittelte Dichte der Schichten reduziert sich von 2,9 g/cm^3 bei reinen Kohlenstoffschichten auf 2,0 g/cm^3 bei maximalem Stickstoffgehalt. Dies ist in Uebereinstimmung mit Ergebnissen aus der TEM-EELS-Spektrometrie. Die sigma-Plasmonenenergie verringert sich von 30,5 eV auf 26,0 eV und das pi-Plasmon gewinnt deutlich an Intensitaet hinzu. Das starke Anwachsen des Pi*-Vorpeaks an der C1s-Kante zeigte, dass die Verwendung von Stickstoff im Prozess nicht zu einer Stabilisierung von Kohlenstoff sp^3-Bindungen fuehrt, wie dies fuer die Bildung von beta-C_3N_4 notwendig waere. Aus TEM-Untersuchungen in Hochaufloesung und im Beugungsmodus ergab sich, dass das Material amorph ist. Jedoch liessen Raman-Untersuchungen eine Tendenz zur Ausbildung einer graphitischen Ordnung auf sehr kleiner Skala bei erhoehter Substrattemperatur und intensiviertem N-Ionenbeschuss vermuten. Durch die Messung der mittleren Zahl von naechsten Nachbarn, der mittleren Bindungslaenge und des mittleren Bindungswinkel mittels Neutronenbeugung konnte gezeigt werden, dass die diamantartige Nahordnung bei geringen Stickstoffgehalten in eine graphitartige Nahordnung bei hoeheren Stickstoffgehalten uebergeht. Diese Ergebnisse deuten auf ein durch sp^2-Bindungen dominiertes Netzwerk bei hohen Stickstoffgehalten hin. Es ist daher sehr wahrscheinlich, dass, zumindest mittels der ionenassistierten Vakuumbogenverdampfung, die Herstellung von beta-C_3N_4 nicht moeglich ist. Davon unabhaengig weisen amorphe CN_x -Schichten mit ca. 20 At.% Stickstoff interessante mechanische Eigenschaften auf. Bei einer Haerte vergleichbar zu c-BN sind sie aeusserst bruchzaeh.

¨Herstellung und Charakterisierung von Kohlenstoffnitridschichten¨ Technische Universitaet Chemnitz, Institut fuer Physik, Dissertation, 1998 (121 Seiten; 61 Abbildungen; 17 Tabellen; 148 Literaturzitate) Spaeth, Christian Mit einem neuartigen Abscheidungsverfahren, der gefilterten Vakuumbogenverdampfung von Graphit in einer Stickstoffatmosphaere, kombiniert mit einer Kaufman-Ionenquelle, wurden Kohlenstoffnitridschichten abgeschieden. Es wurde untersucht, wie sich die Wahl der Schichtabscheidungsparameter auf die Schichtstruktur und -eigenschaften der erzeugten Kohlenstoffnitridschichten auswirkt. Dies geschah insbesondere im Hinblick auf die Herstellung von beta-C_3N_4. Trotz einer breiten Variation der Beschichtungsparameter konnte der Stickstoffgehalt in den Schichten nicht ueber 35 At.% angehoben werden. Die mit ERDA und Profilometrie ermittelte Dichte der Schichten reduziert sich von 2,9 g/cm^3 bei reinen Kohlenstoffschichten auf 2,0 g/cm^3 bei maximalem Stickstoffgehalt. Dies ist in Uebereinstimmung mit Ergebnissen aus der TEM-EELS-Spektrometrie. Die sigma-Plasmonenenergie verringert sich von 30,5 eV auf 26,0 eV und das pi-Plasmon gewinnt deutlich an Intensitaet hinzu. Das starke Anwachsen des Pi*-Vorpeaks an der C1s-Kante zeigte, dass die Verwendung von Stickstoff im Prozess nicht zu einer Stabilisierung von Kohlenstoff sp^3-Bindungen fuehrt, wie dies fuer die Bildung von beta-C_3N_4 notwendig waere. Aus TEM-Untersuchungen in Hochaufloesung und im Beugungsmodus ergab sich, dass das Material amorph ist. Jedoch liessen Raman-Untersuchungen eine Tendenz zur Ausbildung einer graphitischen Ordnung auf sehr kleiner Skala bei erhoehter Substrattemperatur und intensiviertem N-Ionenbeschuss vermuten. Durch die Messung der mittleren Zahl von naechsten Nachbarn, der mittleren Bindungslaenge und des mittleren Bindungswinkel mittels Neutronenbeugung konnte gezeigt werden, dass die diamantartige Nahordnung bei geringen Stickstoffgehalten in eine graphitartige Nahordnung bei hoeheren Stickstoffgehalten uebergeht. Diese Ergebnisse deuten auf ein durch sp^2-Bindungen dominiertes Netzwerk bei hohen Stickstoffgehalten hin. Es ist daher sehr wahrscheinlich, dass, zumindest mittels der ionenassistierten Vakuumbogenverdampfung, die Herstellung von beta-C_3N_4 nicht moeglich ist. Davon unabhaengig weisen amorphe CN_x -Schichten mit ca. 20 At.% Stickstoff interessante mechanische Eigenschaften auf. Bei einer Haerte vergleichbar zu c-BN sind sie aeusserst bruchzaeh.

Gegenstand dieser Arbeit sind Versuche zur Synthese neuer binärer Kohlenstoffnitride im Allgemeinen und von C3N4-Vorläuferverbindungen im Speziellen. Hierbei werden v. a. die Herstellung und die Eigenschaften organischer Polyazide beschrieben, die aufgrund ihrer Gefährlichkeit durch zahlreiche Folgereaktionen in weniger brisante Moleküle überführt werden mussten. Als Derivatisierungsreaktionen kamen hierbei beispielsweise die 1,3-dipolare Cycloaddition mit Norbornen und Cyclooctin, die STAUDINGER-Reaktion mit verschiedenen Phosphinen sowie die Aza-WITTIG-Reaktion zum Einsatz. Es konnten dabei u. a. zehn Röntgeneinkristallstrukturen erhalten und als Strukturbeweis aufgeführt werden. Zahlreiche hochaufgelöste Massenspektren sowie Elementaranalysen und NMR-Daten bestätigten außerdem alle neu erhaltenen Strukturen. Einen weiteren Schwerpunkt dieser Arbeit stellen Versuche zur Synthese von monomerem C3N4 dar, dessen Herstellung zwar nicht gelang, für dessen Bildung allerdings neue Möglichkeiten ausgehend von verschiedenen Edukten beschrieben werden. Darüber hinaus wurden bereits bekannte Moleküle auf ihre Eignung als C3N4-Vorläufer untersucht, wobei z. B. durch Azid-Addition an Nitrilgruppen unerwartete neue Produkte erhalten werden konnten.

Gegenstand dieser Arbeit sind Versuche zur Synthese neuer binärer Kohlenstoffnitride im Allgemeinen und von C3N4-Vorläuferverbindungen im Speziellen. Hierbei werden v. a. die Herstellung und die Eigenschaften organischer Polyazide beschrieben, die aufgrund ihrer Gefährlichkeit durch zahlreiche Folgereaktionen in weniger brisante Moleküle überführt werden mussten. Als Derivatisierungsreaktionen kamen hierbei beispielsweise die 1,3-dipolare Cycloaddition mit Norbornen und Cyclooctin, die STAUDINGER-Reaktion mit verschiedenen Phosphinen sowie die Aza-WITTIG-Reaktion zum Einsatz. Es konnten dabei u. a. zehn Röntgeneinkristallstrukturen erhalten und als Strukturbeweis aufgeführt werden. Zahlreiche hochaufgelöste Massenspektren sowie Elementaranalysen und NMR-Daten bestätigten außerdem alle neu erhaltenen Strukturen. Einen weiteren Schwerpunkt dieser Arbeit stellen Versuche zur Synthese von monomerem C3N4 dar, dessen Herstellung zwar nicht gelang, für dessen Bildung allerdings neue Möglichkeiten ausgehend von verschiedenen Edukten beschrieben werden. Darüber hinaus wurden bereits bekannte Moleküle auf ihre Eignung als C3N4-Vorläufer untersucht, wobei z. B. durch Azid-Addition an Nitrilgruppen unerwartete neue Produkte erhalten werden konnten.