Catalizadores de base carbonosa para reacciones de oxidación y deshidratación de alcoholes

UNIVERSIDAD DE MÁLAGA Facultad de Ciencias Departamento de Ingeniería Química Portada-Spanish TESIS DOCTORAL Para optar al Título de: Doctor en Ingeniería Química con Mención Internacional Catalizadores de base carbonosa para reacciones de oxidación y deshidratación de alcoholes Autor: María José Valero Romero Directores: Dr. D. José Rodríguez Mirasol Dr. D. Tomás Cordero Alcántara Málaga, 2015 AUTOR: María José Valero Romero EDITA: Publicaciones y Divulgación Científica. Universidad de Málaga Esta obra está sujeta a una licencia Creative Commons: Reconocimiento - No comercial - SinObraDerivada (cc-by-nc-nd): Http://creativecommons.org/licences/by-nc-nd/3.0/es Cualquier parte de esta obra se puede reproducir sin autorización pero con el reconocimiento y atribución de los autores. No se puede hacer uso comercial de la obra y no se puede alterar, transformar o hacer obras derivadas. Esta Tesis Doctoral está depositada en el Repositorio Institucional de la Universidad de Málaga (RIUMA): riuma.uma.es D. TOMÁS CORDERO ALCÁNTARA, Catedrático de Ingeniería Química de la Universidad de Málaga, D. JOSÉ RODRÍGUEZ MIRASOL, Catedrático de Ingeniería Química de la Universidad de Málaga, CERTIFICAN: Que el trabajo de investigación recogido en la presente Memoria ha sido realizado bajo su dirección en el Departamento de Ingeniería Química de la Universidad de Málaga por la Ingeniera Dña. MARÍA JOSÉ VALERO ROMERO, y reúne, a su juicio, contenido científico suficiente y las condiciones necesarias para ser presentado y defendido ante el Tribunal correspondiente para optar al Grado de Doctor con Mención Internacional. Málaga, Abril de 2015 Fdo.: Tomás Cordero Alcántara Fdo.: José Rodríguez Mirasol A mi familia “No importa el tamaño del problema, importa la determinación al resolverlo” (desconocido) Índice de contenidos OVERVIEW .................................................................................................................... 9 Chapter 1: Introducción ............................................................................................... 11 Abstract ........................................................................................................................ 12 1. El carbón activo .................................................................................................... 13 1.1. Métodos de preparación de carbón activo .................................................. 13 1.2. Activación química con H3PO4 .................................................................. 16 1.3. Resistencia a la oxidación .......................................................................... 19 1.4. Carbón activo en la catálisis heterogénea .................................................. 21 2. Los alcoholes ........................................................................................................ 23 2.1. Descomposición de (bio)etanol y (bio)metanol ......................................... 26 3. El objetivo de la oxidación química selectiva ...................................................... 27 4. Catalizadores VPO ............................................................................................... 29 5. El carbón activo como molde (“nanocasting”) ..................................................... 30 Chapter 2: Role of surface phosphorus complexes on the oxidation of porous carbons ....................................................................................................... 41 Chapter 3: On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments .................................................... 75 Chapter 4: Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modeling ................................................................................................... 105 Chapter 5: Kinetic Study of the Decomposition of ethanol on carbon-based acid catalysts .................................................................................................... 141 Chapter 6: Lignocellulosic-derived mesoporous materials: An answer to manufacturing non-expensive catalyts useful for the biorefinery processes ................................................................................................... 175 Chapter 7: Carbon materials as template for the preparation of mixed oxides with controlled morphology and porous structure ................................... 195 GENERAL CONCLUSIONS AND FUTURE WORK ............................................ 221 RESUMEN .................................................................................................................. 223 AGRADECIMIENTOS .............................................................................................. 241 CURRICULUM ........................................................................................................... 243 Índice de c ontenidos   9 OVERVIEW Activated carbons materials have attracted considerable attention because their interesting applications in many fields, such as catalysis, gas and liquid phase adsorption and gas and energy storage. The carbon materials present an additional advantage since they can be prepared from different carbonaceous precursors, as agricultural wastes or by-products, that are abundant, environmentally friendly disposal and cost decrease associated with activated carbon production. Consequently, there is a great research effort in this direction. However, the use of carbon materials in catalysis is limited since they would gasify to CO2 (or CO) in the presence of oxygen at relatively low temperatures. Nevertheless, it has been shown that it is possible to prepare carbon materials with a relatively large amount of phosphorus on the carbon surface by chemical activation of lignocellulosic materials with phosphoric acid. This activation procedure leads to phosphorus surface complexes, in form of COPO3, CPO3 and C3P groups, which remain very stable on the carbon surface, at relatively high temperatures, and confer to the carbon a high oxidation resistance and high acidity, increasing the catalytic applications in which these catalysts can be used. The main purpose of the present thesis is to provide some insights into the role of the different phosphorus species on the oxidation resistance of activated carbons, prepared by activation of biomass with phosphoric acid. In addition to this, it is shown some of their potential applications in catalytic processes, as catalytic supports or as catalysts by themselves, for reactions that take place at relatively high temperatures and under oxidizing conditions. Two types of reacctions have been studied in gas phase; i) the catalytic decomposition (dehydration/dehydrogenation) of bioalcohols and ii) the selective oxidation of light hydrocarbons. This P. h. D. thesis is divided into 7 chapters. The first one, introduction, and a final summary are writing in Spanish, whereas the rest of the thesis is presented in English. Chapters 2 and 3 present an extensive study of the oxidation state evolution of the activated carbons surface groups, after subjecting them to thermal treatments in oxidizing and inert conditions. In these chapters, we will discuss the role of the phosphorus surface groups on the high oxygen content and oxidation resistance of this Overview Chapter 1  16 Al ser menor la producción de volátiles y, particularmente, de compuestos de carbono, el rendimiento en sólido es mayor y la contracción de la matriz sólida resultante, menos significativa. Los catalizadores utilizados en el presente trabajo fueron preparados mediante activación química con H3PO4.  Figura 1: Etapas en la activación física y química para la producción de carbón activo  1.2. Activación química con H3PO4 El proceso de activación de materiales de origen biomásico, con ácido fosfórico, ha sido estudiado en detalle por Jagtoyen y Derbyshiren [18], los cuales analizaron la evolución de la reacción con la temperatura. Los compuestos biomásicos están formados por tres componentes mayoritarios, celulosa, hemicelulosa y lignina. Las proporciones típicas de estos tres biopolímeros en la biomasa son: 4060% de celulosa, 2040% de hemicelulosa y 1025% de lignina. Introducción  17 La reacción de la biomasa con el ácido fosfórico comienza en el proceso de impregnación, cuando se mezclan los componentes. Ya cuando se calienta la mezcla hasta 50 ºC existen evidencias de cambios físicos y químicos en la biomasa. Se cree que el ácido ataca, inicialmente, a la hemicelulosa y a la lignina, posiblemente, debido a que el acceso a estos polímeros amorfos es más sencillo que a la estructura cristalina de la celulosa. La celulosa parece ser más resistente a la hidrólisis ácida que otros polisacáridos, como demuestra el hecho de que la estructura celular permanece prácticamente inalterada después de la reacción con H3PO4, incluso a elevadas temperaturas. Uno de los efectos iniciales del ataque ácido es la hidrólisis de los enlaces glucosídicos de la celulosa y la hemicelulosa y la ruptura de los enlaces ariloéter de la lignina. Estas reacciones se acompañan, además, de deshidratación, degradación y condensación. Las reacciones primarias conllevan una reducción del peso molecular, principalmente de la hemicelulosa y la lignina. Al aumentar la temperatura (150 < T < 450 ºC) predominan las reacciones de entrecruzamiento sobre las reacciones de despolimerización y de ruptura de enlaces. El elevado rendimiento en carbono obtenido por el tratamiento ácido a temperaturas superiores a 300 ºC se debe, precisamente, a este entrecruzamiento que retiene las especies de peso molecular relativamente bajo en el seno de la fase sólida. Evidentemente, existe una relación directa entre el desarrollo de porosidad y el proceso de dilatación de la estructura. La formación de ésteres fosfato por reacción de la celulosa con el ácido fosfórico se muestra en la Figura 2, que ilustra como el ácido fosfórico se puede insertar entre las cadenas de celulosa, sustituyendo los puentes de hidrógeno y, al mismo tiempo, separando las cadenas y produciendo la dilatación de la estructura por la adición o inserción de grupos fosfato. A 280 ºC, las estructuras son pequeñas unidades poliaromáticas conectadas entre sí por enlaces de tipo fosfato o polifosfato, incluidas conexiones del tipo polimetileno [(CH2)n]. Al aumentar la temperatura se producen reacciones de ciclación y condensación, que aumentan la aromaticidad y el tamaño de las unidades poliaromáticas, como consecuencia de la escisión de los enlaces del tipo COP. Entre 350 y 500 ºC, el carbonizado es estable, pero a partir de 430 ºC, la ruptura continua de entrecruzamientos produce un crecimiento del tamaño de las unidades aromáticas. Chapter 1  18 T<450 ºC: Formación de ésteres fosfato entre cadenas de celulosa y entrecruzamiento Los ésteres pueden provenir de ácido orto   , piro, y metafosfórico T>450 ºC: Eliminación del H3PO4 Figura 2. Mecanismo de formación de ésteres fosfato entre las cadenas de celulosa [18] Así, el ácido fosfórico parece actuar de dos formas, (i) como catalizador ácido promoviendo las reacciones de ruptura de enlaces y la formación de entrecruzamientos vía procesos de ciclación y condensación y (ii) combinándose con las especies orgánicas Ácido    Ácido    Introducción  19 para formar enlaces de tipo fosfato, ésteres polifosfatos, que sirven para conectar y entrecruzar los fragmentos de biopolímeros. A temperaturas superiores a 450 ºC, la estructura comienza a contraerse. El volumen de microporo decrece de forma progresiva a partir de temperaturas intermedias (aproximadamente 350 ºC), no observándose variación en el volumen de mesoporo hasta unos 550 ºC, temperatura a partir de la cual comienza a disminuir rápidamente. Entre 450 y 550 ºC, la contracción estructural es absorbida por el estrechamiento del diámetro de poro sin causar un cambio significativo del tamaño de poro. Esta contracción se asocia a la ruptura de los enlaces de tipo fosfato, térmicamente inestables a temperaturas superiores a unos 450 ºC. La reducción en la cantidad de los entrecruzamientos resulta en una estructura más densamente empaquetada con la consecuente reducción en el desarrollo de la porosidad. 1.3. Resistencia a la oxidación La importancia de mejorar la resistencia a la oxidación de los materiales de carbono mediante la introducción de distintos grupos superficiales, silicio [19-21], boro [22-25] o compuestos de fósforo [25-28], se pone de manifiesto al comprobar la gran cantidad de publicaciones que se encuentran en la bibliografía sobre este tema. Esto se debe a la baja resistencia a la oxidación que presentan los materiales de carbono [29] y que limitan su uso en aplicaciones catalíticas en atmósfera oxidante. La inhibición de la oxidación por formación de complejos químicos estables en los sitios activos, parece ser la alternativa más prometedora para evitar la gasificación del soporte de carbono. En la bibliografía se recoge un número limitado de compuestos inhibidores de la oxidación, además de los ya mencionados, destacar los compuestos órganohalogenados, fosfatos, óxido de boro, compuestos órganoboro y compuestos órganofosforado. En la actualidad la mayoría de las investigaciones en este tema se centran en el dopado con compuestos de fósforo. Se ha observado un efecto de inhibición a la oxidación por impregnación de carbón con compuestos organofosforados [25,30], H3PO4 [26], POCl3 [31], fosfatos ácidos [28] y fosfatos metálicos [32-37]. También, se ha observado un efecto inhibidor Chapter 1  20 de la reacción de oxidación mediante la incorporación de pentóxido de fósforo en carbones [38,39], o por lavado de materiales compuestos C/C con ácido fosfórico, seguido de un tratamiento térmico hasta su temperatura de descomposición [40]. En términos generales, se plantean dos explicaciones para el efecto inhibidor de los compuestos de fósforo. La primera se basa en un mecanismo de bloqueo físico y la segunda se basa en el envenenamiento de los centros activos del carbón. El recubrimiento con fosfatos de la superficie exterior es muy efectivo para la protección de los carbones frente a la oxidación [41]. Este tipo de recubrimiento reduce el contacto carbón/catalizador y carbón/oxígeno aumentando, de esta forma, la resistencia a la oxidación catalítica de compuestos carbonosos [35-36]. Un requisito fundamental para la eficacia de este tipo de recubrimientos es una distribución uniforme de los depósitos de fosfatos [36]. El mecanismo de envenenamiento de los sitios activos fue propuesta por McKee et al [22-23] y, posteriormente apoyado por otros autores [25,26,30]. McKee et al propuso que los compuestos de fósforo forman compuestos menos reactivos sobre los centros activos en la superficie del carbón, tales como grupos fosfato o fosfito. Se ha propuesto que los complejos formados son C-O-POCl2, para el caso de impregnación con POCl3 como precursor y grupos C-O-PO3 y C-PO3, en el caso de emplear compuestos organo-fosforados como precursores. En este sentido, Radovic et al [27] describe el efecto inhibidor a la oxidación de carbones activos dopados con POCl3 y CH3OP(OH)2. Los autores describen la presencia de grupos C-O-PO3 y C-PO3 en la superficie de los carbones. Además, sugieren que el bloqueo de los centros activos del carbón es debido, principalmente, a la presencia de grupos de P enlazados al carbón, mientras que los grupos metafosfato funcionan como una barrera física entre el catalizador y la superficie del carbón. La presencia de los grupos C-O-PO3, en los que el fósforo está enlazado al carbón mediante un átomo de oxígeno, son primordiales para mantener el efecto inhibidor de los depósitos de fósforo. Por lo tanto, la pérdida de ese oxígeno o la rotura del enlace C-O resultarían en la pérdida de la inhibición a la oxidación. En este trabajo, además, los autores calcularon las energías de los enlaces de los grupos C-O-P y C-P-O, y concluyeron que el enlace O-P en los grupos C-O-P es el más débil. Por tanto, a elevadas temperaturas los grupos C-O-PO3 descomponen a CO, mientras que el fosforo se reduce a C-PO3. Rosas el at [42] describe la resistencia a la oxidación de carbones obtenidos mediante activación química de residuos lignocelulósicos con H3PO4. A diferencia de Introducción  21 los trabajos descritos anteriormente, la activación con H3PO4 genera carbones activos con distintos grupos superficiales de fósforo (C-O-PO3 y C-P-O3) durante la etapa de activación, no siendo necesario el dopado a posteriori de los carbones. Además, estos grupos superficiales permanecen estables en la superficie de los carbones ya que no se eliminan tras la etapa de lavado. Los autores observaron que la presencia de estos grupos superficiales de fósforo producen un cambio en el mecanismo de gasificación de los carbones activos cuando se comparan con carbonizados del mismo precursor. La oxidación de los carbonizados tiene lugar en toda la superficie disponible de las partículas de carbón, mientras que la gasificación de los carbones activados con ácido fosfórico se produce, principalmente, en el exterior de las partículas (shrinking unreacted core model). El los capítulos 2 y 3 de la presente tesis se estudia la evolución de la oxidación de este tipo de carbones activos a distintas temperaturas de oxidación y el papel de los distintos grupos superficiales de fósforo en la mejora de la resistencia a la oxidación de los mismos.  1.4. Carbón activo en la catálisis heterogénea El carbón activo presenta numerosas aplicaciones que afectan a industrias tan diversas como la alimentaria, farmacéutica, química, nuclear, petrolífera, de tratamiento de aguas de consumo y depuración de efluentes industriales y domésticos, tratamiento de aire y de gases, etc. La multitud de aplicaciones del carbón activo resultan del hecho de que es el adsorbente muy versátil debido a que posee alta superficie específica, distribución polimodal de tamaños de poro, estructura porosa en forma de rendija y capacidad de presentar diferentes tipos de naturaleza química en su superficie. Las aplicaciones del carbón activo, pueden ser en fase líquida o gas. En catálisis heterogénea el carbón activo se utiliza fundamentalmente como soporte de fases activas para múltiples reacciones, aunque también se emplea directamente como catalizador. Hoy en día, el carbón activo es un soporte de catalizadores plenamente consolidado en el mercado mundial utilizándose por ejemplo en la hidrogenación de benceno y de compuestos nitrados tanto alifáticos como aromáticos [41]. Sin embargo, a pesar de que existe un creciente interés, sigue siendo bajo el número de procesos catalíticos que utilizan el carbón activo como catalizador a Chapter 1  22 nivel industrial. De hecho, se estima que las aplicaciones catalíticas que emplean carbón activo representan menos del 5% del volumen total. Entre estas reacciones, la más importante es la producción de fosgeno o cloruro de carbonilo, COCl2, que se usa en la síntesis de diferentes productos químicos y polímeros. La presencia de grupos superficiales de oxígeno e hidrógeno tiene una gran influencia en las propiedades adsorbentes y, por tanto, en las propiedades catalíticas del carbón activo, por lo que su estudio ha atraído mucha atención en los últimos años. Las principales ventajas de la utilización del carbón activo en reacciones de catálisis heterogénea son: •La estructura del carbón es resistente a medios tanto ácidos como básicos. •La estructura es estable a elevadas temperaturas (incluso por encima de 700 ºC), en ausencia de aire. •La estructura porosa se puede controlar en el proceso de obtención, produciendo catalizadores con una distribución de tamaño de poro adecuada para una determinada reacción. •Los carbones activos se pueden preparar con muy diferentes formas y tamaños (granular, polvo, pellets, fibras, telas, aereogeles…). •Aunque la superficie de los carbones es esencialmente hidrofóbica, la química de la superficie se puede modificar para aumentar su carácter hidrofílico e, incluso, se pueden obtener carbones con propiedades de intercambio iónico. •En los catalizadores soportados sobre carbón, la fase activa se puede recuperar fácilmente de los catalizadores agotados mediante combustión del soporte carbonoso, lo cual es una gran ventaja cuando la fase activa es un metal precioso. •El coste de los carbones activos es generalmente menor que el de otros soportes tradicionales como la alúmina o la sílice. Introducción  23 Una desventaja tradicional de los carbones activos, ha sido la presencia de materia inorgánica procedente bien del precursor, o bien introducida durante el proceso de preparación. Aunque es cierto que muchos carbones activos tienen contenidos altos en cenizas (de hasta 1015%), especialmente si se han obtenido a partir de carbones minerales, actualmente es posible preparar carbones porosos con contenidos muy bajos en materia mineral, mediante la elección de un precursor adecuado, tales como los precursores de origen biomásico. Superado este problema, el principal inconveniente para la utilización del carbón activo en algunas reacciones catalíticas puede considerarse que es su baja resistencia a la oxidación. Como se ha mencionado en apartado 1.3, la activación de residuos lignocelulósicos con H3PO4 genera carbones activos con grupos superficiales de fósforo que presentan una elevada estabilidad térmica y química y les confiere a los carbones una elevada resistencia a la oxidación y acidez superficial. Además, la activación con H3PO4 genera carbones con un elevado desarrollo de su estructura mesoporosa a elevadas relaciones de impregnación (H3PO4/precursor) lo que los hace muy interesantes para ser utilizados como soportes catalíticos, o bien, como catalizadores ácidos por sí mismos en reacciones que se lleven a cabo en atmósfera oxidante a temperaturas relativamente elevadas. No obstante, en la bibliografía no existen muchos trabajos en los que se utilicen estos carbones con elevada resistencia a la oxidación como catalizadores o soportes catalíticos, por lo que su estudio resulta muy interesante.  2. Los alcoholes La descomposición catalítica de diferentes tipos de alcoholes, por deshidrogenación o deshidratación, ha sido ampliamente analizada en la bibliografía [43-47]. La actividad y la selectividad de este tipo de reacciones está gobernada por las características texturales y las propiedades ácidobásicas y electrónicas del catalizador. Estas reacciones se utiliza para obtener productos a escala industrial y como reacción tipo test para caracterizar las propiedades ácidobásicas de las superficies de catalizadores, ya que está ampliamente aceptado que los productos de deshidrogenación, es decir, aldehídos y cetonas, se producen preferentemente sobre catalizadores básicos, mientras que los productos de deshidratación, olefinas y éteres se generan Chapter 1  24 fundamentalmente sobre catalizadores ácidos [48-50]. Los alquenos u olefinas se producen por reacción de deshidratación catalizada por un único protón (deshidratación intramolecular), mientras que la formación de éteres supone una reacción de acoplamiento intermolecular (deshidratación intermolecular). La relación entre la cantidad de productos de deshidrogenación y de deshidratación no depende únicamente de la naturaleza del catalizador, sino que también es función de la temperatura de reacción. En general, la reacción de deshidrogenación se produce a mayores temperaturas de reacción que la de deshidratación como consecuencia de su mayor energía de activación [51]. Entre los catalizadores empleados para la descomposición de alcoholes se encuentran óxidos metálicos simples y compuestos [52-56], silicoaluminatos o alúminas [57-60] y zeolitas [61-63]. Sin embargo son pocos los trabajos en los que se ha empleado como catalizador para la descomposición de alcoholes el carbón activo, ya sea como soporte de fase activa o como catalizador. A continuación se analizan la descomposición catalítica de los alcoholes utilizados en esta tesis, etanol y metanol, sobre materiales de carbono. Etanol Szymaski et al. [64] llevó a cabo la descomposición de etanol sobre distintos carbones obtenidos a partir de polifulfuril alcohol por activación física con CO2, sin oxidar y oxidados con ácido nítrico, y con Ni depositado. El carbón activo inicial mostraba una actividad muy baja en la descomposición de etanol. Se producía deshidrogenación y deshidratación, generándose como productos principales acetaldehído (mayoritario), etileno y dietil éter junto con pequeñas cantidades de dietilacetal. La oxidación con ácido nítrico produce un carbón con una superficie más ácida lo que se traduce en un aumento muy significativo en la actividad del carbón hacia la reacción de deshidratación. La introducción del Ni produce un aumento de la reacción de deshidrogenación, como consecuencia de la presencia de cationes Ni2+ sobre la superficie del carbón. La actividad catalítica aumenta con la temperatura del proceso, pero a temperaturas superiores a 120 ºC, se producía una caída de la actividad debida a la descomposición térmica de los grupos superficiales oxigenados. Grunewald y Drago [65], analizaron la conversión de diferentes alcoholes sobre un tamiz molecular Introducción  25 carbonoso. Cuando el etanol se diluía en N2 y reaccionaba sobre el tamiz molecular carbonoso a 230 ºC, el catalizador mostraba una baja actividad con una alta selectividad hacia etileno. Sin embargo, si el gas se cambiaba de N2 a aire la actividad del catalizador aumentaba, alcanzándose conversiones cercanas al 70%. En este caso, los productos dominantes eran acetato de etilo y acetaldehído. La descomposición de etanol también ha sido estudiada por el grupo de Carrasco-Marín [66]. Ellos analizaron el efecto de la oxidación del carbón activo con (NH4)2S2O8 en la reacción de descomposición de etanol. De acuerdo con sus resultados, el carbón activo si oxidar tan solo era capaz de catalizar la reacción de deshidrogenación, siendo el principal producto el acetaldehído. Por otro lado, los carbones oxidados mostraban una actividad moderada y se obtuvieron principalmente productos de deshidratación, siendo el dietil éter el mayoritario y una pequeña cantidad de productos de deshidrogenación cuando el gas de reacción es He. Los autores confirmaron que la actividad es consecuencia de la introducción de grupos carboxilos durante el tratamiento de oxidación. Metanol No existen muchos trabajos publicados sobre la descomposición de metanol sobre materiales de carbono. Zawadzki et al. mostraron que el carbón sin una etapa previa de oxidación no reaccionaba en la descomposición de metanol [67]. Los carbones eran activos sólo cuando se oxidaban en atmósfera de oxígeno. Estos resultados evidenciaron, además, la participación de intermediarios metóxido y formaldehido en la descomposición de metanol. Moreno-Castilla et al. [68] estudiaron la reacción de descomposición de metanol catalizada por carbones activos oxidados con diferentes agentes oxidantes, H2O2, (NH4)2S2O8 y HNO3 para introducir grupos oxigenados superficiales y aumentar su acidez superficial. Los carbones oxidados con (NH4)2S2O8 desarrollaron los grupos ácidos más fuertes y demostraron ser los más activos en la deshidratación de metanol a dimetil éter. Un aumento en la temperatura de reacción producía una rápida disminución de su actividad en la deshidratación de metanol. La razón es una descomposición gradual de los grupos carboxilos presentes en la superficie del carbón a temperaturas superiores a 180ºC. Por tanto, resulta de gran interés la preparación de carbones activos con grupos superficiales ácidos de elevada estabilidad térmica y química y sin la necesidad de una etapa adicional de oxidación. En este sentido, Jorge Bedia et al [69,70], ha estudiado la Chapter 1  32 Titirici y colaboradores [14,115,122] han sido los pioneros en la utilización de esta técnica basándose en el uso de materiales carbonosos con morfología esférica, obtenidos por carbonización hidrotermal, como moldes para sintetizar esferas huecas de óxidos metálicos. Estos materiales resultan muy interesantes por su baja densidad y presentan numerosas aplicaciones [123]: suministro y retirada controlada de medicamentos y drogas, tintas y colorantes, catalizadores y soportes catalíticos, encapsulación y protección de macromoléculas biológicas como proteínas o encimas, encapsulación y retirada de residuos y relleno para compuestos ligeros o de baja densidad [124]. Sin embargo, ésta técnica no permiten un control adecuado de la porosidad (volumen de poros y área superficial) de los óxidos cristalinos resultantes. Por lo tanto, investigar nuevas rutas de síntesis para la fabricación de esferas huecas de óxidos mixtos, que exhiban medias o altas áreas superficiales resulta muy interesante. Referencias [1] J de D LópezGonzález, F MartínezVilchez, F RodríguezReinoso Preparation and characterization of active carbons from olive stones Carbon 1980, 18, 413418 [2] F RodríguezReinoso, M MolinaSabio Activated carbons from lignocellulosic materials by chemical and/or physical activation: an overview Carbon 1992, 30, 11111118 [3] MT González, M MolinaSabio, F RodríguezReinoso Steam activation of olive stone chars, development of porosity Carbon 1994, 32, 14071413 [4] JC González, MT González, M MolinaSabio, F RodríguezReinoso, A SepúlvedaEscribano Porosity of activated carbons prepared from different lignocellulosic materials Carbon 1995, 33, 1175 1177 [5] A Baçaoui, A Yaacoubi, A Dahbi, C Bennouna, R Phan Tan Luu, FJ MaldonadoHodar, J RiveraUtrilla, C MorenoCastilla Optimization of conditions for the preparation of activated carbons from olivewaste cakes Carbon 2001, 39, 425432 [6] C MorenoCastilla, F CarrascoMarín, MV LópezRamón, MA AlvarezMerino Chemical and physical activation of olivemill waste water to produce activated carbons Carbon 2001, 39, 14151420 [7] R UbagoPérez, F CarrascoMarín, D FairénJiménez, C MorenoCastilla Granular and monolithic activated carbons from KOHactivation of olive stones Microp Mesop Mat 2006, 92, 6470 Introducción  33 [8] J RodríguezMirasol, T Cordero, JJ Rodríguez Hightemperature carbons from kraft lignin Carbon 1996, 34, 4352 [9] E González Serrano, T Cordero, J RodríguezMirasol, JJ Rodríguez Development of porosity upon chemical activation of kraft lignin with ZnCl2 Ind Eng Chem Res 1997, 36, 48324838 [10] J RodríguezMirasol, T Cordero, JJ Rodríguez Activated carbons from carbon dioxide partial gasification of eucalyptus kraft lignin Energy& Fuels 1993, 7, 133139 [11] N Tancredi, T Cordero, J RodríguezMirasol, JJ Rodríguez Activated carbons from Uruguayan eucalyptus wood Fuel 1996, 75, 17011706 [12] J RodríguezMirasol, T Cordero, JJ Rodríguez Preparation and characterization of activated carbons from eucalyptus kraft lignin Carbon 1993, 31, 8795 [13] O Ioannidou, A Zabaniotou Agricultural residues as precursors for activated carbon production—A review Renewable and Sustainable Energy Reviews 2007, 11, 19662005 [14] JM Dias, MCM AlvimFerraz, MF Almeida, J RiveraUtrilla, M SánchezPolo Waste materials for activated carbon preparation and its use in aqueousphase treatment: A review J Environ Manag 2007, 85, 833846 [15] JS Macedo, L Otubo, OP Ferreira, IF Jiménez, IO Mazali, LS Barreto Biomorphic activated porous carbons with complex microstructures from lignocellulosic residues Microp Mesop Mat 2008, 107, 276285 [16] P Álvarez, C Blanco, M Granda The adsorption of chromium (VI) from industrial wastewater by acid and baseactivated lignocellulosic residues J Hazar Mat 2007, 144, 400405 [17] Consejo Oleícola Internacional http://wwwinternationaloliveoilorg/ [18] M Jagtoyen, F Derbyshire Activated carbon from yellow poplar and white oak by H3PO4 activation Carbon 1998, 36, 10851097 [19] Zhu YC, Ohtani S, Sato Y, Iwamoto N The improvement in oxidation resistance of CVD-SiC coated C/C composites by silicon infiltration pretreatment Carbon 1998, 36(7–8), 929–35 [20] JE Atwater, JR Akse, TC Wang, S Kimura, DC Johnson Preparation of silicon-carbide-coated activated carbon using a high-temperature fluidized bed reactor Chem Eng Sci 2001, 56(8), 2685–93 [21] Dhami TL, Bahl OP, Awasthy BR Oxidation-resistant carbon– carbon composites up to 1700 ºC Carbon 1995, 33(4), 479–90 [22] DW McKee Borate treatment of carbon fibers and carbon/carbon composites for improved oxidation resistance Carbon 1986, 24(6), 737–41 [23] DW McKee, CL Spiro, EJ Lamby The inhibition of graphite oxidation by phosphorus additives Carbon 1984, 22, 285290 Chapter 1  34 [24] LE Jones, PA Thrower Influence of boron on carbon fiber microstructure, physical properties, and oxidation behavior Carbon 1991, 29(2), 251–69 [25] X Wu, CGM Pantano, LR Radovic Inhibition of catalytic oxidation by boron and phosphorus Extended abstracts, the international carbon conference Lexington, (Kentucky USA): American Carbon Society; 2001 p 2–33 [26] S Labruquère, R Pailler, R Naslain, B Desbat Oxidation Inhibition of Carbon Fibre Preforms and C/C Composites by H3PO4 J European Ceramic Society 1998, 18(13), 1953-1960 [27] X Wu, LR Radovic Inhibition of catalytic oxidation of carbon/carbon composites by phosphorus Carbon 2006, 44(1), 141-151 [28] W Lu, DDL Chung Oxidation protection of carbon materials by acid phosphate impregnation Carbon 2002, 40(8), 1249–54 [29] JX Zhao, RC Bradt, PL Walker Jr Effect of air oxidation at 873 K on the mechanical properties of a carboncarbon composite Carbon 1985, 23, 913 [30] SG Oh, NM Rodríguez In situ electron microscopy studies of the inhibition of graphite oxidation by phosphorus J Mater Res 1993, 8, 28792888 [31] TL Dhami, OP Bahl, BR Awasthy Oxidationresistant carbon–carbon composites up to 1700 ºC Carbon 1995, 33, 479490 [32] WF Wilson, E Tenn Oxidation retardant for graphite US patent 4439491, 1984 [33] LE McAllister, NE Jannasch Barrier coating and penetrant providing oxidation protection for carbon– carbon materials US patent 4837073, 1989 [34] JC Cavalier, A Nale Process for obtaining an article made of a carboncontaining composite materials protected against oxidation US patent 5102689, 1992 [35] ER Stover, RP Dietz Inhibition of catalyzed oxidation of carbon–carbon composites US patent 5401440, 1995 [36] ER Stover Method of inhibiting catalyzed oxidation of carbon–carbon composites US patent 5759622, 1998 [37] TB Walker, LA Booker Inhibition of catalytic oxidation in carbon–carbon composite friction materials In: The international carbon conference Berlin, Alemania, 2000 pp 235–236 [38] JF Rakszawski, WE Parker The effect of group IIIA–VIA elements and their oxides on graphite oxidation Carbon 1964, 2, 53–63 [39] T Durkic, A Peric, M Lausevic, A Dekanski, O Neskovic, M Veljkovic, Z Lausevic Boron and phosphorus doped glassy carbon: I Surface properties Carbon 1997, 35, 1567–1572 [40] EJ Hippo, N Murdie, W Kowbel The effect of acid treatments on subsequent reactivity of carbon–carbon composites Carbon 1989, 27, 331–336 Introducción  35 [41] SA Suvorov, EV Chaikun, SL Korobkin, GS Sokolova Protective mechanism of phosphorus containing coatings Ceramics Abstracts 1988, 9, 24–27 [42] JM Rosas, R Ruiz-Rosas, J Rodríguez-Mirasol, T Cordero Kinetic study of the oxidation resistance of phosphorus containing activated carbons Carbon 2012, 50, 1523-1537 [43] OV Krylov Catalysis by nonmetals Academic Press, Nueva York, EEUU, 1970, p 115 [44] H Knözinger Specific poisoning and characterization of Cacatalytically active oxide surfaces Adv Catal 1975, 25, 184271 [45] F Pepe, C Angeletti, S de Rossi Catalytic behavior and surface chemistry of the ZnO/Al2O3 system for the decomposition of 2propanol J Catal 1989, 118,19 [46] MM Wu, WW Kaeding Conversion of methanol to hydrocarbons : II Reaction paths for olefin formation over HZSM5 zeolite catalyst J Catal 1984, 88, 478489 [47] A Gervasini, A Auroux Acidity and basicity of metal oxide surfaces II Determination by catalytic decomposition of isopropanol J Catal 1991, 131, 190198 [48] JM Campelo, A Garcia, JF Herencia, D Luna, JM Marinas, AA Romero Conversion of Alcohols ,Methylated Series, on ALPO4 Catalysts J Catal 1995, 151, 307314 [49] J March Advanced organic chemistry Reactions, mechanism and structure John Wiley and Sons, 4th ed, Nueva York, EEUU ,1992, pp 9821010 [50] MA Aramendia, V Borau, C Jiménez, JM Marinas, A Parras, FJ Urbano Magnesium oxides as Basic catálisis for organic processes Study of the dehydrogenationdehydration of 2propanol J Catal 1996, 161, 829838 [51] JB Nagy, JP Lange, A Gourgue, P Bodart, Z Gabelita Catalysis by acid and bases Elsevier, Amsterdam, Holanda , 1985, p 127 [52] MP Rosynek, RJ Koprowski, GN Dellisante The nature of catalytic sites on lanthanum and neodymium oxides for dehydration/dehydrogenation of ethanol J Catal 1990, 122, 8094 [53] G Larsen, E Lotero, LM Petkovic DS Shobe Alcohol dehydration reactions over tungstated zirconia catalysts J Catal 1997, 169, 6775 [54] ME Manríquez, T López, R Gómez, J Navarrete Preparation of TiO2ZrO2 mixed oxides with controlled acidbasic properties J Mol Catal A 2004, 220, 229237 [55] N Satoh, JI Hayashi, H Hattori Kinetic study of hydrogen adsorption on sulfated zirconia supported platinum Appl Catal A 2000, 202 ,207213 [56] JA Wang, X Bokhimi, O Novaro, T López, R Gómez Effects of the surface structure and experimental parameters on the isopropanol decomposition catalyzed with sol–gel MgO J Mol Catal A 1999, 145, 291300 Chapter 1  36 [57] H Knözinguer, H Büll, K Kochloefl The dehydration of alcohols on alumina : XIV Reactivity and mechanism J Catal 1972, 24, 5768 [58] H Knözinguer, A Schengllia Influence of steric and inductive effects on product distributions in the dehydration of secondary alcohols on alumina J Catal 1974, 33, 142144 [59] JL Swecker, AK Datye Alcohol dehydration over model nonporous alumina powder J Catal 1990, 121, 196201 [60] R Takahashi, S Sato, T Sodesawa, K Arai, M Yabuki Effect of diffusion in catalytic dehydration of alcohol over silica–alumina with continuous macropores J Catal 2005, 229, 2429 [61] DE Bryant, WL Kranich Dehydration of alcohols over zeolite catalysts J Catal 1967, 8, 813 [62] E Santacesaria, D Gelosa, E Giorgi, S Carrà Role of basic and acid sites in the bimolecular dehydration of alcohols catalyzed by HY zeolite J Catal 1984, 90, 19 [63] PA Jacobs, M Tielen, JB Uytterhoeven Active sites in zeolites: Part 6 Alcohol dehydration over alkali cationexchanged X and Y zeolites J Catal 1977, 50, 98108 [64] GS Szymanski, G Rychlicki, AP Terzyk Catalytic conversion of ethanol on carbon catalysts Carbon 1994, 32, 265271 [65] GC Grunewald, RS Drago Carbon molecular sieves as catalysts and catalysts supports J Am Chem Soc 1991, 113, 16361639 [66] F CarrascoMarín, A Mueden, C MorenoCastilla Surface treated activated carbons as catalists for the dehydration and dehydrogenation reactions of ethanol J Phys Chem B 1998, 102, 92399244 [67] J Zawadzki, B Azambre, O Heintz, A Krzto, J Weber IR study of the adsorption and decomposition of methanol on carbon surfaces and carbon-supported catalysts Carbon 2000, 38, 509-515 [68] C MorenoCastilla, F CarrascoMarín, C ParejoPérez, MV López Ramón Dehydration of methanol to dimethyl ether catalyzed by oxidized activated carbons with varying surface acidic character Carbon 2001, 39, 869875 [69] J Bedia, JM Rosas, J Márquez, J Rodríguez-Mirasol, T Cordero Preparation and characterization of carbon based acid catalysts for the dehydration of 2-propanol Carbon 2009, 47, 286–294 [70] J Bedia, R Ruiz-Rosas, J Rodríguez-Mirasol, T Cordero Kinetic study of the decomposition of 2butanol on carbon-based acid catalyst AIChE J 2010, 56, 1557-1568 [71] P Mckendry, Bioresour Energy production from biomass (part 1): overview of biomass Technol 2002, 83, 37–46 [72] D Puppan, Period Polytechnol Environmental evaluation of biofuels Ser Soc Man Sci 2002, 10, 95–116 [73] M Stocker Methanol-to-hydrocarbons: catalytic materials and their behavior Microporous Mesoporous Mater 1999, 29, 3–48 Introducción  37 [74] FJ Keil Methanol-to-hydrocarbons: process technology Microporous Mesoporous Mater 1999, 29, 49–66 [75] AG Gayubo, PL Benito, AT Aguayo, M Olazar, J Bilbao Ralationship between surface acidity and activity of catalysts in the transformation of methanol into hydrocarbons J Chem Technol Biotechnol 65 (1996) 186–192 [76] AT Aguayo, AG Gayubo, R Vivanco, M Olazar, J Bilbao Role of acidity and microporous structure in alternative catalysts for transformation of methanol into olefins Appl Catal A 2005, 283, 197– 207 [77] AG Gayubo, AT Aguayo, AE Sánchez del Campo, AM Tarrío, J Bilbao Kinetic modelling of methanol transformation into olefins on a SAPO-34 catalyst Ind Eng Chem Res 39 (2000) 292–300 [78] DM Brown, BL Bhatt, TH Hsiung, JJ Lewnard, FJ Waller Novel Technology for the Synthesis of Dimethyl Ether from Syngas A Catal Today 1991, 8, 279–304 [79] WW Kaeding, SA Butter Production of chemicals from methanol: I Low molecular weight olefins J Catal 1980, 61, 155–164 [80] R Le Van Mao, P Levesque, GP McLaughlin, LH Dao Ethylene from ethanol over zeolite catalysts Appl Catal 1987, 34, 163–179 [81] R Le Van Mao, TM Nguyen, GP McLaughlin The bioethanol-to-ethylene (BETE) process Appl Catal 1989, 48, 265–277 [82] PM Morse Taking a measure of sustainability Chem Eng News 1999, 77, 20–22 [83] J Hemminger (chairman) New Science for a Secure and Sustainable Energy Future Basic Energy Science Advisory Committee, (US Department of Energy) (2008) [84] A T Bell The Impact of Nanoscience on Heterogeneous Catalysis Science 2003, 299, 1688-1691 [85] R K Grasselli Fundamental principles of selective heterogeneous oxidation catalysis Topics in Catalysis 2002, 21, 79-88 [86] J D Burrington, R K Grasselli Aspects of selective oxidation and ammoxidation mechanisms over bismuth molybdate catalysts Journal of Catalysis 1979, 59, 79-99 [87] R K Grasselli, J D Burrington Selective Oxidation and Ammoxidation of Propylene by Heterogeneous Catalysis Advances in Catalysis 1981, 30, 133-163 [88] P Arpentinier, F Cavani, F Trifirò The Technology of Catalytic Oxidations Chemical, Catalytic and Engineering Aspects, vol 1; Ed Technip: Paris, 2001 [89] P Arpentinier, F Cavani, F Trifirò The Technology of Catalytic Oxidations Safety Aspects, vol 2; Ed Technip: Paris, 2001 [90] G Centi, F, Cavani, F Trifiró Selective Oxidation by Heterogeneous Catalysis; Kluwer Academic/Plenum Publishers: New York, 2001 Chapter 1  38 [91] S Albonetti, F Cavani, F Trifiro Key aspects of catalyst design for the selective oxidation of parafins Catalysis Reviews-Science and Engineering 1996, 38, 413-438 [92] J Haber Heterogeneous Hydrocarbon Oxidation BK Warren and S (EDs) ACS Symp Series 638, (1996) p 20 [93] E K Novakova, V C Vedrine Metal Oxides, Chemistry and Applications; Fierro, J L G, Ed; CRC Press: New York, 2006 [94] F Cavani, J H Teles Sustainability in Catalytic Oxidation: An Alternative Approach or a Structural Evolution? Chemsuschem 2009, 2(6), 508-534 [95] J M López Nieto The selective oxidative activation of light alkanes From supported vanadia to multicomponent bulk V-containing catalysts Topics in Catalysis 2006, 41(1), 3-9 [96] A Caldarelli et al The design of a new ZrO2-supported V/P/O catalyst for n-butane oxidation to maleic anhydride: The build-up of the active phase during thermal treatment Catalysis Today 2010, 157, 204-210 [97] I Sádaba-Zubiri Catalizadores para biorrefinería: obtención de furfural y su transformación a productos de condensación aldólica Tesis doctoral (2012) Instituto de Catálisis y Petroleoquímica (CSIC) [98] T Okuhara, M Misono Key Reaction Steps and Active Surface Phase of Vanadyl Pyrophosphate for Selective Oxidation of Butane Catalysis Today 1993, 16, 61-67 [99] C J Kiely, G J Hutchings Adventures with vanadium phosphate catalysts: Reflections on a long standing collaboration with J C Volta Applied Catalysis 2007, 325, 194-197 [100] E A Lombardo et al The Effect of Preparation Methods and Activation Strategies upon the Catalytic Behavior of the Vanadium-Phosphorus Oxides Catalysis Today 1992, 15, 407-418 [101] R M Contractor et al Butane Oxidation to Maleic Anhydride over Vanadium Phosphate Catalysts Catalysis Today 1987, 1, 49-58 [102] J S Buchanan et al Pretreatment and Activation of a Vanadium Phosphate Catalyst for Butane Oxidation to Maleic Anhydride Applied Catalysis 1985, 19, 65-75 [103] BK Hodnett Comparison of Models for n-Butane Oxidation to Maleic Anhydride over Vanadium Phosphorus Oxide Catalysts Catalysis Today 1993, 16, 131-138 [104] N Hiyoshi et al Selective oxidation of n-butane over novel V-P-O/silica-composites prepared through intercalation and exfoliation of layered precursor Studies in Surfaces and Catalysis 2000, 130, 175-1720 [105] MO Guerrero-Pérez, JM Rosas, R López-Medina, MA Bañares, J Rodríguez-Mirasol, T Cordero Lignocellulosic-derived catalysts for the selective oxidation of propane Catalysis Communications 2011, 12, 989-992 [106] MO Guerrero-Pérez, JM Rosas, R López-Medina, MA Bañares, J Rodríguez-Mirasol, T Cordero On the Nature of Surface Vanadium Oxide Species on Carbons Journal of Physical Chemistry 2012, 116, 20396-20403 [107] MO Guerrero-Pérez, MJ Valero-Romero, S Hernández, JM López Nieto, J Rodríguez-Mirasol, T Cordero Lignocellulosic-derived mesoporous materials: An answer to manufacturing non-expensive catalysts useful for the biorefinery processes Catalysis Today 2012, 195, 155-161 Introducción  39 [108] MJ Valero-Romero, A Cabrera-Molina, MO Guerrero-Pérez, J Rodríguez-Mirasol, T Cordero. Carbon materials as template for the preparation of mixed oxides with controlled morphology and porous structure Catalysis Today 2014, 227, 233-241 [109] IE Wachs, K Routray Catalysis Science of Bulk Mixed Oxides ACS Catal 2012, 2, 1235-1246 [110] S Cheng, D Yan, JT Chen, RF Zhuo, JJ Feng, HJ Li, HT Feng, PX Yan, J Soft-Template Synthesis and Characterization of ZnO2 and ZnO Hollow Spheres PhysChem C 113 2009, 113, 1363013635 [111] X-L Li, T-JLou, X-M Sun, Y-D Li Highly sensitive WO3 hollow-sphere gas sensors, Inorg Chem 2004, 43, 5442 [112] D Zhang, L Qi, J Ma, H Cheng Adv Mater Synthesis of Submicrometer-Sized Hollow Silver Spheres in Mixed Polymer–Surfactant Solutions 2002, 14, 1499-1502 [113] Y Chang, JJ Teo, HC Zeng Formation of colloidal CuO nanocrystallites and their spherical aggregation and reductive transformation to hollow Cu2O nanospheres Langmuir 2005, 21, 1074-1079 [114] NS Ramgir, DJ Late, AB Bhise, MA More, IS Mulla, DS Joag, K Vijayamohanan ZnO multipods, submicron wires, and spherical structures and their unique field emission behavio J Phys Chem 2006, 110, 18236-18242 [115] MM Titirici, M Antonietti, A Thomas A generalized synthesis of metal oxide hollow spheres using a hydrothermal approach Chemistry of Materials 2006, 18, 3808-3812 [116] M E Davis Ordered porous materials for emerging applications Nature 2002, 417, 813-821 [117] VV Guliants y JH Lin Colloidal synthesis and characterization of alumina-supported Cu/Ni coreshell nanoparticles in water gas-shift reaction Applied Catalysis A: General, 2012, 445–446, 187-194 [118] D Farrusseng, A Julbe, M Lopez, C Guizard Investigation  sol–gel methods   synthesis  VPO membrane materials adapted   partial oxidation  nbutane Catalysis Today, 2000, 56, 211220 [119] A Dong, N Ren et al General synthesis of mesoporous spheres of metal oxides and phosphates Journal American Chemistry Society 2003, 125, 4976-4977 [120] M Kang, D Kim, SH Yi y JU Han Preparation of stable mesoporous inorganic oxides via nanoreplication technique Catalysis Today 2004, 695, 93-95 [121] AH Lu, A Taguchi et al Hollow Mesoporous Silica Fibers: Tubules by Coils of Tubules Angew Chem Int Ed 2002, 41, 3489 [122] MM Titirici, M Antonietti Chemistry and materials options of sustainable carbon materials made by hydrothermal carbonization Chemical Society Reviews 2010, 39(1), 103-116 [123] M Ohmori y E Matijevic Preparation and properties of uniform coated colloidal particles VII Silica on hematite Journal of Colloid Interface Science 1992, 150, 594-598 [124] DL Wilcox et al Hollow and Solid Spheres and Microspheres: Science and Technology Associated with their Fabrication and Applications Materials Research Society, Pittsburgh 1995, 372   Role of surface phosphorus complexes on the oxidation of porous carbons Chapter 2 Chapter 2  48 The total acidity and acid strength distribution of the catalysts were determined by adsorption and temperature-programmed desorption of ammonia. The NH3-TPD was performed using 100 mg of catalysts saturated with NH3 (20% vol in Helium) at 100ºC. After saturation, the NH3 weakly adsorbed was desorbed in a He flow at the adsorption temperature, until no NH3 was detected in the outlet gas. The TPD was performed by raising the temperature up to 500 ºC at a heating rate of 10 ºC/min. Outlet NH3 concentrations were measured by mass spectroscopy (Pfeiffer Vacuum, OmniStar model). 3. Results and discussion 3.1. Characterization of the original activated carbons Table 1 summarizes the yield and the porous structure parameters of the original activated carbons. The activated carbon obtained by gasification with CO2, ACG800, shows very low yield (13.5 %, after carbonization and partial gasification). However, the activated carbon prepared by chemical activation with H3PO4 at the same temperature, ACP2800, presents a yield value of 38.7%, which is similar to those obtained under similar conditions with other biomass residues [2,29,42]. The activation agent restricts the formation of tars and volatiles during the carbonization process, thus increasing the yield of the remaining solid product [43]. Both activated carbons show higher DR micropore volume values when N2 is used instead of CO2 as the adsorbate gas, which indicates the presence of a wide microporous structure [44]. The phosphoric acid activated carbon, ACP2800, present higher values of BET surface area and mesopore volume, which confirm that phosphoric acid activation presents a significant contribution on mesoporosity. The activated carbon ACG800, obtained by physical activation, presents a similar surface area, but in this case, the volume of mesopores is lower. The elemental composition and characterization of the surface chemistry of the activated carbons are shown in Table 2. The main elements found on the surface of the phosphoric acid activated carbons are carbon and oxygen, with lower amounts of phosphorus. Nitrogen was also detected, but at very low concentrations, probably being a remnant of nitrogen originally present in the carbon precursor. Phosphoric acid Role of surface phosphorus complexes on the oxidation of porous carbons  49 appears to activate olive stone through the formation of phosphate and polyphosphate bridges that crosslink biopolymer fragments, avoiding the contraction of the structure by pyrolysis. Part of the activating agent is removed during the washing step, leading to a carbon matrix in an expanded state with an accessible and broader porous structure [2,12]. However, part of the P is retained in the structure during the activation procedure and seems to be stably bonded to carbon because they do not elute after the washing step. The maximum amount of phosphorus surface groups that are introduced into carbon structure during the activation process is attained at 700-800 ºC, according to previous works [2,3,16,29]. Suárez-García et al. [45] reported that higher activation temperatures results in a decrease of the phosphorus content due to volatilization of phosphorus compounds, most probably in the form of polyphosphoric acid or phosphorus pentoxide. The mass surface concentration of phosphorus determined by XPS (3.5 % wt) is similar to that obtained by ICP (3.7 % wt), indicating that phosphorus complexes are well distributed overall the carbon particles. No other inorganic elements were detected on the surface of ACP2800, as they are removed by the phosphoric acid activation treatment. On the other hand, the ACG800 surface is composed mostly by carbon and oxygen. Other mineral components like sodium, calcium and potasium, which are very effective catalysts of gasification reactions [13] and were originally present in the carbon precursor in very low amount, were removed to a large extent after a washing step, being present in quantity lower than 1%. In fact, the ash content reported in Table 2 and obtained from the weight of the ultimate solid residue, determined after exposure to air-TG at 900 ºC, is very low for ACG800 sample, 0.3%. In the case of phosphoric acid activated carbons, the main inorganic constituent of the activated carbon ashes is phosphorus, in form of P2O5, as revealed by XPS analysis (not shown). Table 1. Yields and porous structural parameters values of the activated carbons. Yield N2 isotherm CO2 isotherm Sample (%) ABET (m2/g) VDR (cm3/g) Vmes (cm3/g) ADR (m2/g) VDR (cm3/g) ACG800 13.5 1355 0.570 0.064 870 0.349 ACP2800 38.7 1380 0.514 0.654 662 0.265 Chapter 2  50 Table 2. Ash values, elemental composition and surface chemistry characterization (XPS) of the activated carbons. Ash Elemental Analysis (% wt.) XPS (% wt.) Sample (%) C O H N S C1s O1s P2p ACG800 0.3 92.2 6.1 1.2 0.5 0.1 91.8 6.9 - ACP2800 6.0 79.8 12.5 1.4 0.3 0.0 87.1 9.2 3.5 The elemental composition of activated carbons is also shown in Table 2. A very interesting feature observed here is that ACP2800 carbon contains a considerable amount of oxygen (12.5 %wt.), despite having been activated in an inert atmosphere (N2). The oxygen content in phosphoric acid activated carbons has also shown to reach a maximum for samples activated at 700-800 ºC and decreases at higher temperatures, in parallelism with the phosphorus content [2,3,16,29], which might suggest that incorporation of oxygen to the carbon surface is related to the P surface content. Figure 1. Non-isothermal oxidation resistance profiles (10ºC/min) of the ACG800 and ACP2800 activated carbons. Figure 1 shows the thermogravimetric analysis profiles in air for the different activated carbons. ACP2800 carbon despite having similar BET surface area and higher 0.0 0.2 0.4 0.6 0.8 1.0 300 400 500 600 700 800 Remaining Weight (W/W0) Temperature (ºC) ACG800 ACP2800 Role of surface phosphorus complexes on the oxidation of porous carbons  51 mesopore volume than ACG800 is resistant to air oxidation up to 550 ºC, whereas ACG800 carbon starts to lose weight at about 350 ºC. Therefore, chemical activation with phosphoric acid seems to affect the oxidation behaviour of the produced carbon (probably by modification of the carbon surface available for oxidation). Further discussion regarding the surface oxidation evolution with temperature of the phosphoric acid activated carbon ACP2800 and its chemical characterization are detailed in section 3.2. Information about the chemical nature of oxygen surface groups (OSG) present in the activated carbons is obtained by analysing XPS and TPD profiles. The P2p spectrum for ACP2800 (Figure 2a) shows a band with a main peak at a binding energy of about 133.7 eV, which is characteristic of pentavalent tetracoordinated phosphorus (PO4) as in phosphates and/or polyphosphates [28,29]. This broad band can be deconvoluted in different doublets with an area ratio of 0.5 and a distance between peaks of 0.84 eV. Wu and Radovic assigned the peak at a binding energy about 134.0 eV to C-O-P type bonds [30], in where the P atom is bonded to four O atoms by one double bond and three single bonds, such as in C-O-PO3 groups. These authors suggested that these groups could be in the form of numerous cross-linked structures attached to the carbon surface. Thus, (C-O)2PO2 and/or (C-O)3PO could also be formed on the surface of the carbon [46]. Wu and Radovic also attribute a binding energy of about 133.6 eV to C-P bonding as in C-PO3 groups [30], where P is bonded to one C and three O atoms (two single and one double bonds). In this case, two of these O atoms (with single bond P-O) may also be bonding to C atoms, like in C-O-P. Lower values of binding energy (132.0 eV) can be associated to C2PO2 and/or C3PO groups [30,46,47]. Finally, a low intensity doublet peak with the main peak centered at around 131.0 eV may be ascribed to C3P groups [48]. The XPS P2p results for ACP2800 seem to indicate the presence of mainly C-PO3 (46.9 %) and C-O-PO3 (31.7 %) type groups on the activated carbon surface. The O1s spectrum of ACP2800 sample (Figure 2b) shows a high contribution for oxygen at a binding energy of 532.6 eV, which is characteristic of single bonded oxygen in C-OH, C-O-C and/or C-O-P linkages. Compared to ACG800, the O1s spectrum for ACP2800 shows a more pronounce shoulder at lower binding energies, characteristic of C=O and/or P=O groups (BE = 530.9 eV). A binding energy of 535.5 eV is related to chemisorbed oxygen and/or water [49]. Chapter 2  52 Figure 2. Representative XPS spectra (P2p region) of ACP2800 showing the deconvolution (a). O1s normalized XPS spectra of ACG800 and ACP2800 showing its deconvolution (b). Carbon-oxygen groups of acidic character (carboxylic, lactonic) evolve as CO2 upon thermal decomposition in a typical TPD analysis, whereas non acidic (carbonyl, ether, quinone and phenol) evolve as CO. Anydride surface groups evolve as both CO and CO2 [8,50]. According to previous studies, phenols evolve as CO between 600-700 ºC [8,51,52], ether groups at around 700 ºC, carbonyls between 800-900 ºC and more stable chromenes or pyrone groups have also been reported to evolve as CO at around 1000 ºC [53,54]. Figure 3. Amount of CO, CO2, H2O and H2 evolved with temperature during the TPD performed to ACP2800. 130132134136 Binding energy (ev) C-O-PO3, (CO)3PO, (CO)2PO 31.7% C-PO3 and/or C2PO2 46.9% C3PO 14.3% C3P 7.15% ACP2800 528530532534536538540 Binding Energy (eV) ACP2800 ACG800 C=O and/or P=O chemisorbed oxygen and/or water C-OH, C-O-C and or C-O-P 0.0 0.2 0.4 0.6 0.8 1.0 0 2 4 6 8 10 100 300 500 700 900 CO2, H2, H2O mol/(g·s) CO mol/(g·s) T (ºC) CO CO2 H2O H2 (a) (b) Role of surface phosphorus complexes on the oxidation of porous carbons  53 Figure 3 shows the evolution with temperature of CO, CO2, H2O and H2 during the TPD for ACP2800. Figures 4 a and b present the DTP profiles for ACG800. The TPD profile of ACP2800 (Figure 3) shows a significant CO evolution at temperatures higher than 750 ºC that presents a maximum at 860 ºC, which has been previously assigned to the decomposition of stable C–O–P bonds of C-O-PO3 surface groups, producing C-PO3 groups [2,30]. On the other hand, the CO2 profile shows very low amount desorbed in comparison to that of CO, indicating a lower presence of carboxyl, lactonic and anhydride groups on this sample. The washing step performed after the chemical activation process could have introduced some of these OSG of low thermal stability. The relatively higher amount of CO2 that evolves at temperatures higher to 750 ºC, associated to the peak of CO at those temperatures, is probably due to decomposition of stable O=C–O–P groups [39], although secondary reactions between the CO released at these temperatures and the OSG, cannot be disregarded [55]. The H2O profile of the TPD for ACP2800 sample shows a peak at 150 ºC that corresponds mainly to physisorbed water, while a second desorption peak at 270 ºC can be associated to dehydration of two neighboring carboxylic groups. Besides, the absence of water desorbed at high temperatures indicates the lack of phenol groups on the surface of this sample, suggesting a high dehydration state reached by the phosphoric acid activation at a temperature of 800 ºC. There is also some H2 release starting at 800 ºC, probably as a consequence of both aromatic condensation and the dehydrogenation of the acid phosphates C-O-PO3 groups (or C-PO3) that contains one or two OH) on the carbon surface. In the last case, generation of C-O-PO2-O-C and/or C-O-PO(OC)2 (or C-PO2-O-C and/or C-PO(OC)2) type surface groups might occur, if dehydrogenation of one or two OH of the phosphate group takes place, respectively, followed by the brakeage of the C-O-P bonds at higher temperatures (from 800 to 950 ºC), forming finally C3PO type surface groups. Similar results have been observed for activated carbons obtained by chemical activation with phosphoric acid of different raw materials [46]. The main peak on the CO profile for ACG800 carbon (Figure 4a) is located at about 850 ºC and may be associated to carbonyl/quinone or esthers groups. Besides, the shoulder at 270 ºC and the large tail of the CO2 profile between 350-650 ºC indicated the presence of anhydrides and lactonic groups (Figure 4b). The small amount of CO2 Chapter 2  54 that evolves at temperatures above 800 ºC is, probably, a result of secondary reactions between the CO and the OSG, as we commented before. The amount of oxygen evolved during the TPDs for both carbons have been calculated from the total amounts of CO and CO2 desorbed. The results are 6.0 and 7.2 wt. % of oxygen for ACG800 and ACP2800, respectively. Comparison of these results with those for the elemental composition, showed in Table 2, indicates that most of the oxygen on sample ACG has been desorbed as CO and CO2 during the TPD up to 930 ºC, whereas ACP2800 still retains a relatively high amount of oxygen groups of higher thermal stability after the TPD experiment. A second (-TT) and a third (-TT2) TPD experiments were performed immediately after the first one on ACG800 and ACP2800 activated carbons, without exposure to ambient air, in order to elucidate the thermal stability of the OSG (Figure 4). The results obtained showed a small but constant presence of CO at high temperatures (860 °C) and almost not evolution of CO2 and H2 (data not shown), which was associated to decomposition of C-O-PO3 type groups. Different hypothesis must be examined in order to explain this behaviour. The option that residual O2 in the N2 flow may re-oxidate the carbon surface could be disregarded given that subsequent TPD from ACG800 carbon showed that most of the oxygen complexes were desorbed as CO and CO2 during the first TPD up to 930 ºC (Figure 4a-b). For the second TPD run, ACG800-TT, the amount of oxygen evolved became negligible (0.5 %wt. of O) and only residual CO-evolving groups were detected at high temperature. These results seem to indicate that there was not a significant re-oxidation of the carbon surface by nitrogen impurities (H2O and O2 contents below 3 ppm) or, even, by CO2 evolved during the first TPD. Vivo-Vilche et al. [56] reported that re-oxidation of the carbon surface by CO2 desorbed only takes place during thermal treatments below 700 ºC, based on experimental evidences. Theoretical studies on this question have also been made using different approaches (density functional theories or computational chemistry calculations) [57,58]. Other possibility is that remaining CO-evolving groups of very high thermal stability may show some degree of mobility on the carbon surface when the temperature is going down during the TPD experiment and move to previously generated free sites producing OSG of lower thermal stability that decompose as CO during the successive Role of surface phosphorus complexes on the oxidation of porous carbons  55 TPDs. Experimental evidences of mobility of CO and CO2-evolving groups has been previously reported [51,56]. The total oxygen content evolved from the sample was calculated from the total amounts of CO and CO2 desorbed during the successive TPDs. The oxygen content for ACP2800 is 10.4% wt., which is slightly lower than the amount obtained by the elemental analysis (Table 2), supporting the fact that mobility of oxygen from COevolving oxygen groups of higher thermal stability may produce the oxidation of C-P to C-O-P type group, producing CO at ca. 860 ºC during successive TPDs. Figure 4. Amount of CO and CO2 evolved with temperature during consecutive TPDs performed to ACG800 (a-b) and ACP2800 carbon (c-d), respectively. 0 2 4 6 8 10 300 500 700 900 CO rate mol/(g·s) T (ºC) ACG800 ACG800-TT 0 0.2 0.4 0.6 0.8 1 100 300 500 700 900 CO2, H2O, H2rate mol/(g·s) T (ºC) 0 2 4 6 8 10 300 500 700 900 CO rate mol/(g·s) T (ºC) ACP2800 ACP2800-TT ACP2800-TT2 0 0.2 0.4 0.6 0.8 1 100 300 500 700 900 CO2rate mol/(g·s) T (ºC) (c) (d) (a) (b) Chapter 2  56 3.2. Oxidation of activated carbons The interest on improving the oxidation resistance of carbon materials by introduction of different kinds of surface groups is clear from the large number of research papers on this subject. Nevertheless, this is a matter under continuous discussion due to their inherent complexity. Phosphorus shows an important oxidation inhibition effect, observed in carbon impregnated with different compounds, such as organo-phophorus compounds [35], POCl3 [59] or H3PO4 [59,60]. We should emphasize here that in our case the surface phosphorus groups are directly generated during the preparation process under specific conditions and seem to be strongly bonded to the carbon surface. The phosphorus complexes could be probably located at the edges of carbon crystallites, stabilizing the carbon active sites, which become less reactive for the oxidation reaction, as occurs when carbon materials are doped with phophorus [30,59,61,62]. Nevertheless, they could also act as a physical barrier, blocking the access of oxygen to the active sites. To investigate the oxidation evolution of the carbon surface, the activated carbons were oxidized in air at different temperatures (between 120-350 ºC) for 2 h and then characterized in situ by TPD. It is interesting to mention that ACP2800 is resistant to gasification in air, with no significant weight loss observed during 2 h at 350 ºC. However, ACG800 produces a carbon burn-off of about 20 %wt. after air oxidation at 350 ºC for 2 h. Figure 5 shows the resulting CO and CO2 evolution profiles during the TPD of the oxidized samples. The TPD results after air oxidation at different temperatures evidence that oxidation of the phosphoric acid-activated carbon (ACP2800) at temperatures lower than 300 ºC increases significantly only the amount of CO evolved at about 860 ºC, accompanied by a small evolution of CO2 at the same temperature, suggesting that oxidation of the carbon surface takes place through the oxidation of C-P surface bond, generating C-O-P bond that are thermally stable up to 800 ºC. At higher oxidation temperatures (325 and 350 ºC), the desorption of CO (CO2) at high temperature (860 ºC) seems to keep constant, which suggests that the surface P groups are completely oxidized and other oxygen surface complexes, thermally less stable, are formed, decomposing as CO and CO2 at lower temperatures (~700 ºC for CO and ~500 and 700 ºC for CO2). Role of surface phosphorus complexes on the oxidation of porous carbons  57 Figure 5. Amount of CO and CO2 evolved with temperature during the TPD experiments of ACP2800 (a and b) and ACG800 (c and d) after air oxidation at different temperatures (120, 180, 240, 300, 325 and 350 ºC) for 2h. Oxidation of ACG800 activated carbon that contains no surface P groups causes a more pronounce increase in the evolution of CO2 (and CO) at low temperatures (400500 ºC), compared to that of ACP2800, indicating that oxidation of this sample takes place preferentially on the carbon surface, generating OSG of lower thermal stability. Two broad peaks are found in the CO2 profiles of ACP2800 and ACG800samples (Figures 5 b and d). The low temperature peak (c.a. 500 ºC for ACP2800 and 450 ºC for ACG800) could be attributed to decomposition of lactone groups and the higher temperature peak (c.a. 700 ºC) can be originated from the more stable anhydride groups, which decompose to CO2 and CO. Oxidation at low temperature seems to preferentially form lactone type surface groups on the surface of ACG800 and anhydride ones on the 0 2 4 6 8 10 100 300 500 700 900 CO rate mol/(g·s) T (ºC) ACP2800 ACP2800-O120 ACP2800-O180 ACP2800-O240 ACP2800-O300 ACP2800-O325 ACP2800-O350 0 0.2 0.4 0.6 0.8 1 100 300 500 700 900 CO2 rate mol/(g·s) T (ºC) 0 2 4 6 8 10 100 300 500 700 900 CO rate mol/(g·s) T (ºC) ACG800 ACG800-O240 ACG800-O300 ACG800-O325 ACG800-O350 0.0 0.2 0.4 0.6 0.8 1.0 100 300 500 700 900 CO2 rate mol/(g·s) T (ºC) (d) (a) (b) (c) Chapter 2  64 Figure 8. P2p normalized XPS spectra for ACP2800 before and after being subjected to thermal treatments at 900 ºC in inert atmosphere or/and in air atmosphere at 350 ºC for 2 h. The observed general trend for oxygen surface content on ACG800 and ACP2800 and their corresponding derived carbons was predictable, being lower after the thermal treatment and higher after the oxidation process. However, certain differences can be appreciated between these two carbons. Similar values of O(XPS) and O(TPD) for the fresh ACG800 activated carbon can be observed indicating an homogeneous distribution of oxygen surface groups on the whole particle surface. On the other hand, the high value of O(XPS) compared to that obtained from TPD, O(TPD), for ACP2800, could indicate that a higher proportion of oxygen groups could be concentrated on the external surface of the carbon particles. However, it has to be 130132134136 Binding Energy (eV) ACP2800 ACP2800-O350-TT ACP2800-TT-350 ACP2800-TT-O350-TT ACP2800 ACP2800-O350 ACP2800-TT C3P C3PO C-O-PO3, (CO)2PO2, (CO)3PO C-PO3and/or C2PO2 (a) (b) Role of surface phosphorus complexes on the oxidation of porous carbons  65 pointed out that (and contrary to ACG800 sample) not all the oxygen surface content of this sample is being taking into account during the first TPD experiment, as was shown in Figures 4 c and d. In this sense, the O(TPD) value observed for ACP2800 after the thermal treatment (ACP2800-TT) of 2.0 %wt. represents about 28 % of that for the fresh carbon and this surface oxygen seems to be attached to surface P, given that the value of 1.9 %wt. of O-P(TPD) for this sample represent about 95 % of the O(TPD) value. Thus, successive TPDs become necessary to quantify the total oxygen content O(TPD) for this carbon that contains OSG of very high thermal stability (C-O-PO3 type). The relatively high values of O-P(XPS) and O-P(TPD) observed for ACP2800 and its derived carbons compared to their respective total XPS and TPD oxygen amount indicate that most of the oxygen present in the carbon particles seems to be bonded to surface phosphorus. It is also very interesting the fact that the values of O(XPS) and OP(XPS) for ACP2800 and all the samples derived from it (after thermal treatment or oxidation process) are very similar, respectively (see Table 3). Surprisingly, the values of O(XPS) and O-P(XPS) for ACP2800-TT and ACP2800-TT-O20 are almost identical and very close to those for ACP2800 and ACP2800-TT-O350. It has to be noted that all the samples have been exposed to ambient air after the different treatments, before the XPS analysis was carried out. Thus, as the TPD results indicated, C-PO3 type surface groups, formed after the thermal treatment in inert atmosphere at 900 ºC, can be oxidized to C-O-PO3 type surface groups in ambient air, before the XPS analysis. In this sense, ACP2800-TT sample after stored to ambient air would be identical to ACP2800TT-O20 and, as mentioned before, ACP2800 would be identical to that stored in a stove at 60 ºC. These results also evidence the redox character of these P surface groups that can be oxidized with molecular oxygen and reduced by thermal treatments in inert atmosphere, restoring the original surface nature of the activated carbon. These results also explain the unexpectedly high oxygen content of different carbons prepared by chemical activation with phosphoric acid after being heat treated to high temperature in inert atmosphere reported in the literature [2,27,29,30]. In those cases, the oxygen content measurement was carried out by elemental analysis, FTIR, XPS, etc., in which the samples were exposed to ambient air before the analysis. In this sense, it is Chapter 2  66 important to point out that the characterization of these types of carbons should be carried out by an “in situ” analysis, without exposing the samples to ambient air. These carbons were further characterized by adsorption/desorption of NH3 in order to study the acid character of the carbon surface before and after the different treatment. The NH3-TPD profiles for ACG800, ACP2800, ACP2800-TT, ACP2800O350 and ACP2800-TT-O350 samples, after adsorption and desorption of NH3 at 100 ºC, are compared in Figure 9. ACG800 sample presents an almost negligible desorption of NH3.On the other hand, the NH3 desorption profile of ACP2800 shows two maxima, at around 150 and 250 ºC. The first one is known to arise from desorption of weakly adsorbed NH3 molecules (probably by hydrogen-bonded), identified on zeolite supports [67], although it can also be related to weak acidic groups of C-O-PO3 type on these carbon materials, since ACP2800-TT-O20, which desorbs a slightly higher amount of NH3 at this temperatures compared to that of ACP2800-TT, presented a higher amount of oxygen-containing P groups of C-O-PO3 type, as was showed in Figure 7a. On the other hand, the shoulder at 250 ºC and the large tail at higher temperatures may originate from NH3 adsorbed on strong Brönsted acid sites [46]. The acidity of this type is, mainly, related to the high amount of phosphorus retained on the carbon surface and probably associated to OH in the phosphates groups. The total surface acidity for ACP2800 increases after air oxidation at 350 ºC, given that a high amount of oxygencontaining P groups of C-O-PO3 type are formed, as was showed in Figures 5 a and b and Table 3. The thermal treated sample, ACP2800-TT, was not exposed to ambient air before the adsorption of NH3. This sample desorbed a small amount of CO (and CO2) at high temperatures in the TPDs experiments (Figures 4 c and d), associated to decomposition of C-O-PO3 groups, which suggest the presence of a relatively high amount of surface groups of C2PO2 and/or C3PO type that may contribute to the low acidity of this sample, as can be observed in Figure 9. Finally, the similar NH3-TPD profiles obtained for ACP2800-O350 and ACP2800-TT-O350 samples also confirm the re-generation (re-oxidation) of the carbon surface after thermal treatment in inert atmosphere and the acid character of these oxygen-containing P groups. The results presented in this work support the conclusions derived from the work of Wu and Radovic [30], in the sense that C-O-PO3 type surface groups and, specifically, the bridge O atoms play a key role in the carbon oxidation inhibition mechanism. Activated carbons prepared by chemical activation of lignocellulosic Role of surface phosphorus complexes on the oxidation of porous carbons  67 materials with phosphoric acid present oxygen-containing P surface groups of type CO-PO3 (that include also CO2PO2 and CO3PO groups). These P surface groups are thermally very stable and decompose only at temperatures higher than 750 ºC, producing CO (and CO2) in the gas phase and C-PO3 type surface groups (that may include C2PO2 and C3PO, depending on the treatment temperature), very stable even at 950 ºC, given that P content on the surface of these carbons remains constant or even slightly increases after being treated at this temperature (see Table 3). Oxidation of the carbon surface with molecular oxygen at temperatures lower than 300 ºC takes place through the C-P bond of the later P surface groups type, producing C-O-P bonds; i.e., regenerating the C-O-PO3 type groups (that may include C2PO-O-C, C-PO(OC)2 and (CO)3PO depending on the intensity of the oxidation treatment; although polyphosphate type groups cannot be disregarded) on the surface of the carbon. Only at temperatures higher than 300 ºC, these C-O-P bonds become saturated (all of them of (CO)3PO type groups) and oxidation of the rest of the carbon surface seems to be possible, with production of oxygen surface groups that are less stable, decomposing at lower temperatures (<700 ºC). Figure 9. NH3-TPD profiles of the activated carbons ACG800, ACP2800, ACP2800O350 ACP2800-TT-O20 and ACP2800-TT-O350 after NH3 adsorption/desorption at 100 ºC. 100 200 300 400 500 NH3 desorbed (a.u.) T (ºC) ACG800 ACP2800 ACP2800-TT ACP2800-O350 ACP2800-TT-O20 ACP2800-TT-O350 Chapter 2  68 As Wu and Radovic [30] suggested, the presence of O bonded to both a carbon site and a P group (C-O-P) is a critical factor for maintaining the inhibition effect of P deposits, because these bonds can be generated at very low temperatures and are not broken at temperatures lower than 750 ºC. Thus, the inhibition effect seems to be, in this case, the result of site blockage. McKee et al [34] indicated that the C-O-PO3 group’s type may occupy a large proportion of the surface of the carbon. In this sense and assuming a cross sectional area of 40 Å2 for these surface groups, a P surface content of 3.5 %wt. would occupy an area of approximately 87 m2/g, which represents a larch part of the external surface area of ACP2800 ( 433 m2/g). On the other hand, TPD and XPS results suggest that P surface groups present a uniform distribution on the carbon surface. On the other hand, the surface of these activated carbons, with oxygencontaining P groups, should be a strong acceptor of spillover oxygen [68], which would explain the formation of other oxygen groups of low thermal stability on the carbon surface once P surface groups are saturated with oxygen at temperature higher than 300 ºC (see Figure 5). This effect would also explain the role of molecular oxygen in the reaction mechanism of ethanol decomposition on this type of acid carbon, oxidizing and removing the carbon deposited on the surface acid sites, avoiding catalyst deactivation [16]. 4. Conclusions Chemical activation of olive stone with phosphoric acid produces activated carbons with a relatively high content of phosphorus surface groups that remain very stable on the carbon surface at high temperatures. TPD and XPS results point out that P surface groups preferentially reacts with oxygen, prior to carbon gasification, through the oxidation of C-P bond, forming C-O-P ones, which are thermally stable at temperatures lower than 750 ºC. Thermal treatments at higher temperature (about 860 ºC) decompose the C-O-PO3 type surface groups to less oxygenated phosphorus groups on the carbon surface (of C-PO3 type) generating CO (and CO2). These C-PO3 groups seem to be very reactive and are re-oxidized upon contact with air, even at room temperature, forming again C-O-PO3 type groups. Thus, the presence of these oxygen- Role of surface phosphorus complexes on the oxidation of porous carbons  69 containing phosphorus surface groups with an interesting redox functionality of high chemical and thermal stability seems to be responsible of the high oxygen content and oxidation resistance of this type of porous carbons. Acknowledgments We gratefully thank Junta de Andalucia (P09-FQM-5156) and Spanish Ministry of Economy and Competitiveness (MINECO) and FEDER (Projects CTQ2012-36408) for financial support. M.J.V.R. gratefully thanks MINECO for a FPI fellowship (BES2010-032213). References [1] Tancredi N, Cordero T, Rodríguez-Mirasol J, Rodríguez JJ. CO2 gasification of eucalyptus wood chars. Fuel 1996;75(13):1505–8. [2] Rosas JM, Bedia J, Rodríguez-Mirasol J, Cordero T. Preparation of hemp-derived activated carbon monoliths.Adsorption of water vapor. Ind Eng Chem Res 2008;47(4):1288–96. [3] Puzy AM, Poddubnaya OI, Martínez-Alonso A, Suárez-García F, Tascón JMD. Synthetic carbons activated with phosphoric acid III. Carbons prepared in air. Carbon 2003;41:1181-1191. [4] Rodríguez-Reinoso F. The role of carbon materials in heterogeneous catalysis. Carbon 1998;36(3):159-75. [5] Moreno-Castilla C, Ferro-García MA, Joly JP, Bautista Toledo I, Carrasco-Marín F, Rivera-Utrilla J. Activated carbon surface modifications by nitric acid, hydrogen peroside, and amonium peroxydisulfate treatments. Langmuir 1995;11(11):4386-92. [6] Moreno-Castilla C, Lópes-Ramón MV, Carrasco-Marín F. Changes in surface chemistry of activated carbons by oxidation. Carbon 2000;38:1995-2001. [7] Szymanski GS, Rychlicki G, Terzyk AP. Catalytic conversion of ethanol on carbon catalysts. Carbon 1994;32:265-271 [8] Figueiredo JL, Pereira MFR, Freitas MMA, Órfao JJM. Modification of the surface chemistry of activated carbons. Carbon 1999;37(9): 1379–1389. [9] Byl O, Liu J, Yates JT. Etching of carbon nanotubes by ozones a surface area study. Langmuir 2005;21:4200–4. [10] Xia W, Jin C, Kundu S, Muhler M. A highly efficient gas-phase route for the oxygen functionalization of carbon nanotubes based on nitric acid vapor. Carbon 2009;47:919–22. Chapter 2  70 [11] Gosselink RW, van den Berg R, Xia W, Muhler M, de Jong KP, BitterGas JH. phase oxidation as a tool to introduce oxygen containing groups on metal-loaded carbon nanofibers. Carbon 2012;50:44244431. [12] Rosas JM, Bedia J, Rodríguez-Mirasol J, Cordero T. HEMP-derived activated carbon fibers by chemical activation with phosphoric acid. Fuel 2009;88:19-26. [13] Rosas JM, Ruiz-Rosas R, Rodríguez-Mirasol J, Cordero T. Kinetic study of the oxidation resistance of phosphorus containing activated carbons. Carbon 2012;50:1523-1537 [14] Bedia J, Rosas JM, Márquez J, Rodríguez-Mirasol J, Cordero T. Preparation and characterization of carbon based acid catalysts for the dehydration of 2-propanol. Carbon 2009;47:286–294. [15] Bedia J, Ruiz-Rosas R, Rodríguez-Mirasol J, Cordero T. Kinetic study of the decomposition of 2butanol on carbon-based acid catalyst. AIChE J. 2010;56:1557-1568. [16] Bedia J, Barrionuevo R, Rodríguez-Mirasol J, Cordero T. Ethanol dehydration to ethylene on acid carbon catalysts. Applied Catalysis B: Environmental 2011;103:302-310. [17] Guerrero-Pérez MO, Rosas JM, López-Medina R, Bañares MA, Rodríguez-Mirasol J, Cordero T. Lignocellulosic-derived catalysts for the selective oxidation of propane. Catalysis Communications 2011;12(11):989-992. [18] Guerrero-Pérez MO, Valero-Romero MJ, Hernández S, López Nieto JM, Rodríguez-Mirasol J, Cordero T. Lignocellulosic-derived mesoporous materials: An answer to manufacturing non-expensive catalysts useful for the biorefinery processes. Catalysis Today 2012;195:155-161. [19] Valero-Romero MJ, Cabrera-Molina A, Guerrero-Pérez MO, Rodríguez-Mirasol J, Cordero T. Carbon materials as template for the preparation of mixed oxides with controlled morphology and porous structure. Catalysis Today 2014;227(15):233-241. [20] Bedia J, Rosas JM, Rodríguez-Mirasol J, Cordero T. Pd supported on mesoporous activated carbons with high oxidation resistance as catalysts for toluene oxidation. Applied Catalysis B: Environmental 2010;94(1-2):8-18. [21] Khalil LB, Girgis BS, Tawfik TAM. Decomposition of H2O2 on activated carbon obtained from olive stones. Journal of Chemical Technology and Biotechnology 2001;76(11):1132-1140. [22] Haw KG, Bakar WA, Ali R, Chong JF, Kadir AAA. Catalytic oxidative desulfurization of diesel utilizing hydrogen peroxide and functionalized-activated carbon in a biphasic diesel-acetonitrile system. Fuel Processing Technology 2010;91(9):1105-1112. [23] Puziy AM, Poddubnaya OI, Martínez-Alonso A, Castro-Muñiz Alberto, Suárez-García F, Tascón JMD. Oxygen and phosphorus enriched carbons from lignocellulosic material. Carbon 2007;45(10):19411950. Role of surface phosphorus complexes on the oxidation of porous carbons  71 [24] Puziy AM, Poddubnaya OI, Martínez-Alonso A, Suárez-García F, Tascón J.M.D. Synthetic carbons activated with phosphoric acid II. Porous structure. Carbon 2002;40:1507–1519. [25] Girgis BS, El-Hendawy A-NA. Porosity development in activated carbons obtained from date pits under chemical activation with phosphoric acid. Microporous and Mesoporous Materials 2002;52(2):105117. [26] Jagtoyen M, Derbyshire F. Some considerations of the origins of porosity in carbons from chemically activated wood. Carbon 1993;31(7):1185-1192. [27] Puziy AM, Poddubnaya OI, Socha RP, Gurgul J, Wisniewski M. XPS and NMR studies of phosphoric acid activated carbons. Carbon 2008;46: 2113-2123. [28] Puziy AM, Poddubnaya OI, Ziatdinov AM. On the chemical structure of phosphorus compounds in phosphoric acid-activated carbon. Applied Surface Science 2006;252:8036–8038 [29] Puziy AM, Poddubnaya OI, Martínez-Alonso A, Suárez-García F, Tascón JMD. Surface chemistry of phosphorus-containing carbons of lignocellulosic origin. Carbon 2005;43:2857–2868. [30] Wu X, Radovic LR. Inhibition of catalytic oxidation of carbon/carbon composites by phosphorus. Carbon 2006;44(1):141-151. [31] González-Serrano E, Cordero T, Rodríguez-Mirasol J, Cotoruelo L, Rodríguez JJ. Removal of water pollutants with activated carbons prepared from H3PO4 activation of lignin from kraft black liquors. Water Res. 2004;38(13):3043-3050. [32] Puziy AM, Poddubnaya OI, Martınez-Alonso A, F Suárez-Garcıa, Tascón JMD. Synthetic carbons activated with phosphoric acid: I. Surface chemistry and ion binding properties. Carbon 2002:40(9);1493–1505. [33] Altenor S, Carene B, Emmanuel E, Lambert J, Ehrhardt J-J, Gaspard S. Adsorption studies of methylene blue and phenol onto vetiver rootsactivated carbon prepared by chemical activation. Journal of Hazardous Materials 2009:165(1-3) 1029-1039. [34] McKee DW, Spiro CL, Lamby EJ. The inhibition of graphite oxidation by phosphorous additives. Carbon 1984;22(3):285–90. [35] Lee Y-J, Radovic LR. Oxidation inhibition effects of phosphorus and boron in different carbon fabrics. Carbon 2003;41(10):1987–97. [36] Oh SG, Rodriguez NM. In situ electron microscopy studies of the inhibition of graphite oxidation by phosphorus. J Mater Res 1993;8(11):2879–88. [37] González-Serrano E, Cordero T, Rodríguez-Mirasol J, Cotoruelo L, Rodríguez JJ. Water Res. 2004;38:3043-3050. Chapter 2  72 [38] Rodríguez-Mirasol J, Cordero T and Rodríguez JJ. CO2-Reactivity of eucaliptus kraft lignin chars. Carbon 1993;31:53-61. [39] Brunauer S, Emmett PH, Teller E. Adsorption of gases in multimolecular layers. J Am Chem Soc 1938;60:309-319. [40] Dubinin MM, Zaverina ED, Radushkevich LV. J Phys Chem 1947;21:1351-1362. [41] Kaneko K, Ishii C. Superhigh Area Determination of Microporous Solids. Colloids Surf 1992;67(9):203-212. [42] Vernersson T, Bonelli PR, Cerrella EG, Cukierman AL. Arundodonax cane as a precursor for activated carbons preparation by phosphoric acid activation. Bioresour Technol 2002;83(2):95–104. [43] Jagtoyen M, Derbyshire F. Activated carbons from Yellow Poplar and White Oak by H3PO4 activation. Carbon 1998;36(7):1085–97. [44-] Rodríguez-Reinoso F, Molina-Sabio M, González MT. The use of steam and CO2 as activating agents in the preparation of activated carbons. Carbon 1995;33(1):15–23. [45] F. Suárez-García, S. Villar-Rodil, C. G. Blanco, A. Martíez-Alonso, and J. M. D. Tascón. Effect of Phosphoric Acid on Chemical Transformations during Nomex Pyrolysis. Chem. Mater. 2004;16:26392647. [46] Bedia J, Ruiz-Rosas R, Rodríguez-Mirasol J, Cordero T. A kinetic study of 2-propanol dehydration on carbon acid catalysts. Journal of catalysis 2010;271:33-42. [47] Moulder JF, Stickle WF, Sobol PE, Bomben KD, in:J. Chastain, R.C. King Jr. (Eds.), Handbook of X-ray Photoelectron Spectroscopy, Physical Electronics Inc. Eden Prairie, MN, 1995, pp. 4872-4875. [48] Bedia J, Rosas JM, Vera D, Rodríguez-Mirasol J, Cordero T. Isopropanol decomposition on carbon based acid and basic catalysts. Catalysis Today 2010:158;89-96. [49] D. Briggs, M.P. Seah (Eds.), Practical Surface Analysis, Auger and X-ray Photoelectron Spectroscopy, vol. 1, Wiley, Chichester, UK, 1990. [50] Zielke U, Huttinger KJ, Hoffman WP. Surface-Oxidized Carbon Fibers: I. Surface Structure and Chemistry. Carbon 1996;34:983-998. [51] Moreno-CastillaC, Carrasco-Marín F, Maldonado-Hódar FJ, Rivera-UtrillaJ. Effects of non-oxidant and oxidant acid treatments on the surface properties of an activated carbon with very low ash content. Carbon 1998; 36(1-2):145-51. [52] Marchon B, Carrazza J, Heinemann H, Somorjai GA. TPD and XPS studies of O2, CO2 and H2O adsorption on clean polycristallite graphite. Carbon 1988;26(4):507-14. Role of surface phosphorus complexes on the oxidation of porous carbons  73 [53] Bandosz TJ. in Carbon Materials for Catalysis (eds P. Serp and J. L. Figueiredo), John Wiley & Sons, Inc., Hoboken, NJ, USA. (2008) pp. 45-92, doi: 10.1002/9780470403709.ch2 [54] Montes-Morán MA, Suárez D, Menéndez JA, Fuente E. On the nature of basic sites on carbon surfaces: an overview. Carbon 2004;42(7):1219–25. [55] Peter J. Hall and Joseph M. Calo. Secondary Interactions upon Thermal Desorption of Surface Oxides from Coal Chars. Energy and Fuels 1989;3:370-376. [56] Vivo-Vilches JF, Bailón-García E,. Pérez-Cadenas AF, Carrasco-Marín F, Maldonado-Hódar FJ. Tailoring the Surface chemistry and porosity of activated carbons: Evidence of reorganization and mobility of oxygenated Surface groups. Carbon 2014;68:520-530. [57] Montoya A, Mondragón F, Truong TN. Formation of CO precursors during char gasification with O2, CO2 and H2O. Fuel Process Technol 2002;7778:125–30. [58] Radovic LR. The mechanism of CO2 chemisorption on zigzag carbon active sites: a computational chemistry study. Carbon 2005;43(5):907–15. [59] Labruquère S, Pailler R, Naslain R, Desbat B. Oxidation Inhibition of Carbon Fibre Preforms and C/C Composites by H3PO4. J. European Ceramic Society 1998;18(13):1953-1960. [60] Lu W, Chung DDL. Oxidation protection of carbon materials by acid phosphate impregnation. Carbon 2002;40(8):1249–54. [61] Wu X, Pantano CGM, Radovic LR. Inhibition of catalytic oxidation by boron and phosphorus. Extended abstracts, the international carbon conference. Lexington, (Kentucky USA): American Carbon Society; 2001. p. 2–33. [62] Lu W, Chung DDL. Oxidation protection of carbon materials by acid phosphate impregnation. Carbon 2002;40(8):1249–54. [63] Otake Y, Jenkins RG. Characterization of oxygen-containing surface complexes created on a microporous carbon by air and nitric acid treatment. Carbon 1993;31(1):109–21. [64] Gómez-Serrano V, Piriz-Almeida F, Durán-Valle CJ, PastorVillegas J. Formation of oxygen structures by air activation. A study by FT–IR spectroscopy. Carbon 1999;37:1517–28. [65] F. Carrasco-Marín, J. Rivera-Utrilla, J.P. Joly and C. Moreno-Castilla. Effects of ageing on the oxygen surface complexes on fa oxidized activated carbon. J. Chem. Soc., Faraday TRasn. 1996;92(15):2779-2782. [66] Papier E, Lacroix R, Donnet JB, Nanse G, Fioux P. XPS study of the halogenation of carbon blackPart 1. Bromination. Carbon 1994;32(7):1341-58. Chapter 3  80 3. Results 3.1. Characterization of activated carbons The evolution of the surface chemistry and the porous structure with the impregnation ratio (R, H3PO4/olive stone) and the activation temperature was studied for olive stone in a previous work [21]. In the present paper, the physicochemical properties of two selected activated carbons prepared at activation temperatures of 500 and 800 ºC and an impregnation ratio of R = 2 are presented. In addition, a detailed characterization study of their surface chemistry evolution by XPS and temperatureprogrammed desorption experiments is carried out. Table 2 shows the physicochemical properties of the activated carbons. Yield values of 48 and 38 % for ACP2500 and ACP2800, respectively, were obtained, similar to those obtained from other biomass natural waste pyrolyzed at comparable temperatures [3,5,22-25] and slightly higher than those reported for olive stone with different activation agents [26-28]. The phosphoric acid limits the formation of tars during the carbonization treatment, increasing the yield of the remaining solid product [29]. Moreover, the yield values decrease slightly with the activation temperature studied, as a consequence of a deeper dehydration of the carbonaceous structure of the precursor. The elemental composition of the activated carbon is also shown in Table 2. Oxygen content was determined by difference. An interesting feature observed is the relatively high oxygen values (ca. 9 % for ACP2500 and 12 % for ACP2800), characteristic in this type of activated carbons, as a result of the incorporation into the carbon matrix formed of the oxygen bound to the phosphorus 15,[30]. The values of the structural parameters, as derived from N2 adsorptiondesorption and CO2 adsorption isotherms are also shown in Table 1. The values of micropore volume obtained from the N2 adsorption isotherm, VDRN2, are higher than those obtained from the CO2 adsorption isotherm, VDRCO2, which is indicative of a wide porous structure [31]. Moreover, the carbons present high values of BET surface area and mesopore volume, which confirms the effect of phosphoric acid activation on the porosity development of lignocellulosic residues. The increase of the activation temperature produces a slightly higher BET surface area and micropore volumes measured by N2 and CO2 adsorption isotherms. On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments  81 Table 1. Porous structure and chemical properties of activated carbons Ultimate analysis (% d.a.f.) ACP2500 ACP2800 C 83.7 79.8 H 3.4 1.4 N 0.1 0.3 O* 9.1 12.5 Surface area and porosity N2 adsorption (-196 ºC) ABET (m2/g) 1326 1380 At (m2/g) 476 433 VDR (cm3/g) 0.443 0.514 Vmes (cm3/g) 0.724 0.654 CO2 adsorption (0 ºC) ADR (m2/g) 537 662 VDR (cm3/g) 0.215 0.265 *: obtained by difference (100% – C –N –H –S –ash) The surface chemistry of the activated carbons has been evaluated by XPS and DTP measurements. Table 2 shows the mass surface concentrations of the most abundant elements as obtained from XPS analyses. The well-known presence of stably bonded phosphorus complexes on the surface of phosphoric acid activated carbons has been previously reported [3,5,30]. This activation method produces the reaction between the acid and some organic species, forming phosphate bridges, which are responsible for connecting and crosslinking the biopolymer fragments [29] and have been proven to remain over the carbon surface after the washing step. The retained phosphorus amount on these samples ranges between 2.3 and 3.6 % wt., which is in line with results obtained for phosphoric acid activated carbons of olive stones and similar lignocellulosic biomass wastes. The amount is slightly higher when the chemical activation proceeds at 800 ºC (ACP2800) what suggests that the increase in the activation temperature results in phosphorus combined in more stable forms with the carbon surface, decreasing the efficiency of the washing step. Furthermore, these surface phosphorus groups are thermally stable, as can be seen from the similar or even higher P amounts detected after the thermal treatment at 900 ºC (ACP2500-TT and ACP2800-TT), confirming the phosphorus groups are strongly bonded to the carbon surface. In addition to phosphorus, the pristine activated carbons have an oxygen rich surface which is mostly due to the presence of phosphorus-containing groups, as we mentioned before. However, the presence of phenolic groups cannot be disregarded, Chapter 3  82 especially for ACP2500, as we will be shown later on. The thermal treatment cleansed partly the surface oxygen groups, producing a decrease in the surface oxygen amount measured by XPS from 8.2 to 4.3 % wt. and from 9.2 to 6.5 % wt. for ACP2500 and ACP2800, respectively. The relatively high oxygen amount of the thermal treated samples is related to the presence of oxygen bonded to phosphorus groups like in phosphine and phosphonate groups that either are stable even at 900 ºC or to newly formed polyphosphate groups caused by the surface re-oxidation [10]. Table 2. XPS mass surface concentration (%) and P2p deconvolution of the activated carbons before and after the thermal treatment at 900 ºC under N2. XPS (% wt.) P 2p deconvolution Sample C1s O1s P2p (C-O)nPO C-PO3 and C2PO2 C3PO C3P ACP2800 87.1 9.2 3.5 31.7 46.9 14.3 7.1 ACP2800-TT 89.6 6.5 3.6 24.9 43.6 22.8 8.6 ACP2500 89.0 8.2 2.3 54.8 35.2 10.0 0 ACP2500-TT 93.0 4.3 2.5 19.0 38.6 31.1 11.2 P2p region for the pristine and thermal treated activated carbons is represented in Figure 1a, whereas Figure 1b shows the deconvolution of the P2p region for ACP2800 carbon. The phosphorus spectrum was deconvoluted in four doublets with an area ratio of 0.5 and a distance between peaks of 0.84 eV. Wu and Radovic assigned the peak at a binding energy about 134.0 eV to C-O-P type bonds [7], in where the P atom is bonded to four O atoms by one double bond and three single bonds, such as in C-O-PO3 groups. These authors suggested that these groups could be in the form of numerous crosslinked structures attached to the carbon surface. Thus, C-O-PO2(OC) and/or C-OPO(OC)2 could also be formed on the surface of the carbon [12]. Scheme 1 summarizes the possible C-O-P and C-P-O bonds formed during the chemical activation and thermal treatment. A binding energy of about 133.2 is characteristic of C-P bonding as in C-PO3 and/or C2PO2 groups [7]. Lower intensity peaks at 132.0 and 131.0 eV is associated to C3PO [12,7] and C3P groups [32], respectively. The main peak position was fitted using On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments  83 a margin of ± 0.2 eV, whilst the full width at medium height (FWMH) was set to 1.4± 0.1 eV. The percentages of each group with respect to total P2p spectrum area are compiled at the last columns of Table 2. It can be seen that the increase of the activation temperature to 800 ºC (ACP2800) results in a shift of the maximum of the P2p peak to lower binding energies, indicating the presence mainly of C-PO3/C2PO2 (47 %) and CO-PO3 (31.7 %) type groups on the activated carbon surface, whereas ACP2500 carbon surface is mostly composed of C-O-PO3 type groups (54.8 %). This behavior has been previously observed for other carbons obtained by chemical activation of biomass waste with phosphoric acid [5,21]. The thermal treatment carried out over ACP2500 leaded to a reduction of C-OPO3 groups, whilst the same treatment produced a lesser amount of reduced groups when conducted over the ACP2800 sample. In fact, ACP2500-TT and ACP2800-TT present similar binding energies of the maximum of the P2p peak to that obtained for ACP2800 sample. In our earlier work [10], it was evidenced that C-P bonds are very easily reoxidized to C-O-P bonds upon contact with air, even at low temperatures. Thus, the surface re-oxidation after the samples were exposed to ambient air, before XPS analysis, would explain this result and the relatively high oxygen surface concentration of the thermal treated samples, as aforementioned. Figure 1. Representative XPS spectra (P2p region) of ACP2800 showing the deconvolution (a) and P2p normalized XPS spectra of the activated carbons and thermal treated samples (b).  Binding Energy (eV) ACP2800 ACP2800-TT ACP2500 ACP2500-TT  Binding energy (ev) C-O-PO3 C-PO3 and/or C2PO2 C3PO C3P ACP2800 (a) (b) Chapter 3  84 Scheme 1. Possible phosphorus surface groups formed during H3PO4 activation, as CO-P and C-P-O bonds. Possible C-O-P bonds Possible C-P-O bonds The O1s spectrum for the pristine activated carbons is represented in Figure 2, showing the deconvolution of the O1s region for ACP2800 sample. The O1s spectra of ACP2500 shows a higher contribution for oxygen at a binding energy of 532.6 eV which is characteristic of single bonded oxygen in C-OH, C-O-C and/or C-O-P linkages, whereas ACP2800 O1s spectrum shifts slightly to lower binding energies characteristic of C=O and/or P=O groups (BE = 530.9 eV). A binding energy of 535.5 eV is characteristic of chemisorbed oxygen and/or water [33].                        On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments  85 Figure 2. O1s normalized XPS spectra of ACP2500 and ACP2800 showing the deconvolution. The joint use of XPS and TPD allow a more precise assessment of the surface composition of the activated carbons. TPD technique is recognized as one of the most adequate for identify chemical nature of oxygenated surface groups (OSG), especially those surface oxygen groups that decompose upon thermal treatment producing the evolution of CO and CO2. According to literature, carbon-oxygen groups of acidic character, such as carboxylic and lactonic, evolve as CO2 upon thermal decomposition, whereas non acidic, such as carbonyl, ether, quinone and phenol, evolve as CO. Anydride surface groups evolve as both CO and CO2 [34,35]. Figure 3a and b compares the results of the CO and CO2 TPD profiles, respectively, for ACP2500 and ACP2800 activated carbons, while the total CO and CO2 released obtained by integrating the areas under the TPD peaks are reported in Table 3. For comparison, the TPD profiles of two activated carbons prepared at 350 and 650 ºC, which contain 1.9 and 3.2 % wt. of surface P content, respectively, are also included. As can be observed, the activation process results in the thermal stabilization of the oxygen surface groups that evolve as CO. For instance, the main band for the evolution of CO for carbon ACP2500 appears at ~790 ºC. However, the most significant CO evolution for the activated carbon ACP2800 takes place at higher temperatures (~870 ºC), and this change in the thermal stability of the oxygen surface groups that evolve as CO seems to be significant for activation temperatures above 350 °C. Based on the work of Wu and Radovic [7] and on our previous results with olive stone-derived activated carbons [10], 528530532534536538540 Binding Energy (eV) ACP2500 ACP2800 C=O and/or P=O  ! "#$% &%!' (& )* &%!  Chapter 3  86 the CO evolution at high temperatures can be related to the formation P-surface groups during the phosphoric acid activation, as relatively weak C–O–PO3 surface groups on the carbon surface at high temperatures. Therefore, for activated carbons from olive stone, most of the CO evolved at high temperatures, and a lower amount desorbed at temperatures below 700 ºC probably in the form of anhydride, phenol and ether groups. According to previous studies, phenols evolve as CO between 600-700 ºC [34,36,37], ether groups at around 700 ºC, carbonyls and quinone between 800-950 ºC and more stable chromenes or pyrone groups have also been reported to evolve as CO at around 1000 ºC [38,39]. Furthermore, an interesting feature observed in Figure 3 is the increasing evolution in the amount of CO (and CO2) desorbed at high temperatures (T > 700 ºC), as the activation temperature used for the preparation of the activated carbons increases. This tendency can be due to the higher amount of surface P complexes that decomposes as CO (and CO2), as indicated by XPS, but also these results may suggest that C-OPO2(OC), C-O-PO(OC)2 and C-PO(OC)2) groups, in which the P is bonded to the carbon surface through more than one oxygen, are favourably formed on the surface of phosphoric acid activated carbons prepared at higher activation temperatures. On the other hand, the CO2 profiles (Figure 3b) show lower desorbed amounts compared to those for CO, indicating a lower presence of carboxyl, lactonic and anhydride groups. Moreover, the relatively high amount of CO2 that evolves at the same temperatures that those observed for the maxima release of CO (~790 ºC for ACP2500 and ~860 ºC for ACP2800) is probably as a consequence of decomposition of stable O=C–O–P groups [27] (see Scheme 1). However, secondary reactions between the CO released at these temperatures and the OSG cannot be disregarded [40]. Figure 4 and 5 shows the evolution with temperature of CO, CO2, H2O and H2 during the TPD for ACP2500 and ACP2800, respectively. In addition, a second TPD experiment was performed immediately after the first one on ACP2500 and ACP2800 activated carbons, without exposure to ambient air, in order to elucidate the thermal stability of the OSG. The CO-TPD profiles during the second TPD run (CO-2nd TPD) are also presented (green line). On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments  87 Figure 3. Amount of CO (a) and CO2 (b) evolved with temperature during the TPD performed to ACP2350, ACP2500, ACP2650 and ACP2800. As can be observed, the H2O profile of the TPD for ACP2500 (Figure 4) sample is significantly higher to that observed for ACP2800 carbon (Figure 5), indicating an incomplete dehydration of the phosphoric-impregnated carbon precursor at low activation temperature. The first peak of water desorption at 150 ºC corresponds mainly to the physisorbed water while the second desorption peak at 260 ºC can be associated to the dehydration of two neighboring carboxylic groups. Besides, the coincidence of CO-desorption with the highest peak in the H2O profile at about 700 ºC indicates a large proportion of phenol groups in this sample. Therefore, it is probably that aqueous solution of phosphoric acid protonates surface C-OH groups through hydrogen bonding during the impregnation step which decomposes to CO at about 700 ºC. Puziy et al. [15] and Fu et al. [41] found that an increase of the activation temperature between 400 and 600 ºC resulted in greater amounts of phenol and ether groups. On the other hand, the absence of water desorbed at high temperatures for ACP2800 sample indicates both, the lack of phenol groups on the surface of this sample and thus the high dehydration state reached by the phosphoric acid activation at a temperature of 800 ºC. There is also an important H2 release for ACP2500 which is significantly lower for ACP2800 and both releases coincide with the maximum peak of CO desorption. The desorbed H2 may be a consequence of both aromatic condensation and the dehydrogenation of the acid phosphates C-O-PO3 groups (or C-PO3) that contains one or two OH on the carbon surface. In the last case, generation of C-O-PO2(OC) and/or C-     100 300 500 700 900 CO mol/(g·s) T (ºC) ACP2350 ACP2500 ACP2650 ACP2800 (a) (b) 0.0 0.2 0.4 0.6 100 300 500 700 900 CO2mol/(g·s) T (ºC) Chapter 3  88 O-PO(OC)2 (or C-PO2(OC) and/or C-PO(OC)2) type surface groups might occur, if dehydrogenation of one or two OH of the phosphate group takes place, respectively, followed by the brakeage of the C-O-P bonds at higher temperatures (from 750 to 950 ºC), forming finally C3PO type surface groups (see Scheme 1). These results are similar to those observed for activated carbons obtained by chemical activation with phosphoric acid of different raw materials [10,27]. Figure 4. Amount of CO, CO2, H2O and H2 evolved with temperature during the TPD performed to ACP2500 and amount of CO evolved with temperature during a consecutive TPD. Figure 5. Amount of CO, CO2, H2O and H2 evolved with temperature during the TPD performed to ACP2800 and amount of CO evolved with temperature during a consecutive TPD.               CO2, H2, H2O mol/(g·s) CO mol/(g·s) T (ºC) CO CO-2nd TPD CO2 H2O H2               CO2, H2, H2O mol/(g·s) CO mol/(g·s) T (ºC)  %!+,  ) ) On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments  89 The results obtained during the second TPD for both activated carbons showed a small but constant presence of CO at high temperatures (>800 °C), which was associated to decomposition of C-O-P type bonds, and almost not evolution of CO2 (>800 °C), H2 and H2O (not shown for brevity). According to the literature, remaining CO-evolving groups of very high thermal stability may show some degree of mobility on the carbon surface [10,36,42,]. Thus, when the temperature is going down during the TPD experiment these CO-evolving groups may migrate to previously generated free sites producing OSG of lower thermal stability that decompose as CO during the successive TPDs [10]. A very interesting feature observed during the second TPD run is that the maximum of the CO (and CO2) TPD profile for ACP2500 sample shifts from 790 to ~860 ºC (see CO-2nd TPD in Figure 4). In contrast, ACP2800 sample during the second TPD run presents the maximum of the CO (and CO2) peak at 860 ºC (Figure 5), similar to that observed for pristine ACP2800 that had been previously activated at an elevated temperatures (800 ºC). Thus, it is suggested that decomposition of remaining and/or newly formed C2-PO(OC) and C-PO(OC)2 groups present on the carbon surface after the first TPD run might result again in desorption of CO (and CO2) at relatively higher temperatures (~860ºC) during the second TPD run for ACP2500 and ACP2800. The presence of these phosphate and polyphosphate linkages of higher thermal stabilities on the selected activated carbon surfaces is quite interesting in order to further explain their surface chemistry evolution under oxidizing conditions. 3.2. Oxidation of activated carbons Non-isothermal oxidation experiments before and after a thermal treatment up to 900 ºC in inert atmosphere (designed as –TT) were carried out and the results are shown in Figure 6. These figures depict the thermogravimetric analysis profiles in air for ACP2500 (a) and ACP2800 (b). In general, both activated carbons in spite of having high BET surface areas and mesopore volumes are resistant to air oxidation at about 500 ºC for ACP2500 and about 550 ºC for ACP2800. Thus, this result seems to point out that, at least at lower conversions, the higher the phosphorus content, the lower the oxidation rate. Nevertheless, an increase in the carbonization temperature may also produce a certain ordering of the carbon structure, thus increasing the oxidation onset Chapter 3  96 As expected, the oxygen content decreases after the thermal treatment and increases after the oxidation process. Furthermore, it has to be pointed out that the O-P values obtained after oxidation at 180 ºC for ACP2500 and at 300 ºC for ACP2800 would correspond, approximately, to the total oxygen that may react with the P complexes since the carbon oxidation start to be significant above these oxidation temperatures and other oxygen surface complexes, thermally less stable, are starting to be formed (Figure 7). Therefore, from the XPS and TPD results the ratio mol of O-P (from TPD) per mol of surface P (from XPS), O-P/P, has been calculated for ACP2500O180 and ACP2800-O300 resulting about 1.5 and 3.3, respectively. These values may indicate that phosphorus groups in which the phosphorus atom is bonded to the carbon surface through one or two atoms of oxygen, such as C-OPO3 and C-O-PO2(OC) groups are mostly present on the surface of ACP2500-O180. On the other hand, C-O-PO(OC)2 groups are most probably on ACP2800-O300 carbon surface. The ratio O-P/P could be slightly overestimated given that, the total oxygen content bonded to phosphorus was compared to the surface phosphorus content obtained from XPS analysis (OP(TPD)/P(XPS)). In addition, the O-P/P ratios for ACP2500-TT-O350 and ACP2800-TT-O350 are the same, indicating that the same phosphorus complexes are formed after thermal treatment up to 900 ºC and re-oxidation to 350 ºC. 4. Discussion Phosphoric acid activation of lignocellulosic materials appears to generate phosphate and polyphosphate bridges that crosslink biopolymer fragments, avoiding the contraction of the structure by pyrolysis and thus resulting in activated carbons with an expanded porous structure (At = 476 m2/g for ACP2500 and At = 433 m2/g for ACP2800) [29]. Moreover, the resultant activated carbons present a relatively high amount of surface phosphorus (2.3 %wt for ACP2500 and 3.5 %wt for ACP2800) from the activation agent which seems to be very stable bonded to carbon, even after a thermal treatment up to 900 ºC (Table 2). The formation of oxygen-containing P surface groups during the phosphoric acid activation of type C-O-PO3 (that include also C-OPO2(OC) and C-O-PO(OC)2 groups) has been reported previously in the literature On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments  97 [7,10]. In this sense and assuming a cross sectional area of 40 Å2 for these surface groups, the P surface content on ACP2500 and ACP2800 would occupy an area of approximately 59 and 87 m2/g, respectively, which represent about 13 and 21% of the external surface area of their respective activated carbons. The activation temperature process results in the thermal stabilization of these P surface groups (C-O-PO3) which are thermally very stable and decompose only at temperatures higher than 750 ºC, producing CO (and CO2) in the gas phase and C-PO3 type surface groups (that may include C2PO2 and C3PO, depending on the treatment temperature). Moreover, the TPD profiles of ACP2500 and ACP2800 (Figure 4 and 5, respectively) revealed an important release of H2 (lower for ACP2800) which suggests that dehydrogenation of the acid phosphates C-O-PO3 groups (or C-PO3) and the generation of C-O-PO2-O-C and/or C-O-PO(OC)2 (or C-PO2-O-C and/or C-PO(OC)2) type surface groups might occur, which are further decomposed to CO (and CO2) at higher temperatures, forming finally the C3PO type surface groups. According to this, it is suggested that the degree of dehydrogenation reactions leading to the formation of different C-O-PO3 groups with further decomposes to C-PO3 ones, on the surface of the activated carbons, strongly depends on the activation temperature. In this sense, C-O-PO3 groups (that may include polyphosphates structures) in which the phosphorus is attached to the carbon surface through one oxygen atom and contains two OH are most probable on the surface of ACP2500, as revealed the high amount of H2 desorbed above 500 ºC from this sample, which are thermally stable up to 700 ºC. Although, the presence of phosphate groups thermally more stable is not disregarded due to this sample slightly decomposes to CO (and CO2) at higher temperatures (~860 ºC) (Figure 8a). On the other hand, ACP2800 (or ACP2500-TT) carbon might generate higher amount of C3PO surface groups after activation at 800 ºC (or after a thermal treatment up to 900 ºC) which are re-oxidized, upon contact with air, producing C-O-P bonds; i.e., regenerating the C-O-PO3 type groups (that may include C2-PO(OC), C-PO(OC)2 and C-O-PO(OC)2 depending on the intensity of the oxidation treatment) on the surface of the carbon. These C-O-P bonds are thermally more stable and decompose only at about 860 ºC, producing CO (and CO2). Moreover, these results further explain the higher value of O-P/P for ACP2800 (2.9) compared to that for ACP2500 (1.3). Scheme 2 presents the possible phosphate structures on the surface of phosphorus-containing activated carbons prepared by Chapter 3  98 chemical activation of lignocellulosic precursors with H3PO4 and at different activation temperatures (500 and 800 ºC). Scheme 2. Possible phosphate structures on the surface of phosphorus-containing activated carbons prepared by chemical activation of lignocellulosic precursors with H3PO4 and different activation temperatures. Finally, oxidation of the activated carbons surface seems to proceed through two mechanisms. The presence of mainly C-O-PO3 structures on the surface of ACP2500 correlates quite well with the work reported by Radovic, in which postulated that the bridge O atoms play a key role in the carbon oxidation inhibition mechanism. The ACP2500 ACP2800 On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments  99 inhibition effect seems to be, in this case, the result of site blockage. On the other hand, the presence of P surface complexes with a redox functionality, as that observed by C3PO groups on the surface of ACP2800 and ACP2500-TT carbons (Figure 7, see ACP2800-TT-O350 and ACP2500-TT-O350), seem to enhance the oxidation resistance of these type of porous carbons through the oxidation of C-P bonds to C-O-P ones. 5. Conclusions Chemical structure and thermal stability of phosphorus groups in two activated carbons prepared by phosphoric acid activation at 500 and 800 ºC and an impregnation ratio (H3PO4/olive stone) of 2 were investigated by X-ray photoelectron spectroscopy (XPS) and temperature-programmed desorption experiments (TPD). It has been shown that the activation process results in the thermal stabilization of P surface groups (C-O-P type bonds), which are thermally stable at temperatures up to 700 ºC for ACP2500 and up to 800 ºC for ACP2800. Thermal treatments at higher temperature increases the less oxygenated phosphorus groups on the carbon surface (probably C3-PO) generating CO (and CO2), which upon contact with air re-oxidized forming C-O-P type groups, such as C2-PO(OC), C-PO(OC)2 and C-O-PO(OC)2 depending on the intensity of the oxidation treatment. These groups have been shown to be thermally more stable and more abundant over the carbons prepared at 800 ºC and those thermal treated up to 900 ºC in N2 atmosphere. Thus, the presence of these oxygen-containing phosphorus surface groups with an interesting redox functionality of high chemical and thermal stability seems to be responsible of the high oxygen content and oxidation resistance of this type of porous carbons. On the other hand, the lower oxidation resistance observed for ACP2500 carbon seems to be the result of site blockage by the presence of P groups bonded to carbon sites through an oxygen atom (C-O-PO3 groups). Acknowledgments We gratefully thank Junta de Andalucia (P09-FQM-5156) and Spanish Ministry of Economy and Competitiveness (MINECO) and FEDER (Project CTQ2012-364086) for financial support. M.J.V.R. gratefully thanks MINECO for a FPI fellowship (BES2010-032213). Chapter 3  100 References [1] Bansal RC, Donnet JB, Stoeckli HF, Active Carbon, Marcel Dekker, New York, 1988. [2] Rodríguez-Reinoso F, in: J.W. de Patrick (Ed.), Porosity in Carbons: Characterization and Applications, Edward Arnold, London, 1995. [3] Rosas JM, Bedia J, Rodríguez-Mirasol J, Cordero T. Preparation of hemp-derived activated carbon monoliths. Adsorption of water vapor. Ind Eng Chem Res 2008;47(4):1288–96. [4] Puzy AM, Poddubnaya OI, Martínez-Alonso A, Suárez-García F, Tascón JMD. Synthetic carbons activated with phosphoric acid III. Carbons prepared in air. Carbon 2003;41:1181-1191. [5] Rosas JM, Bedia J, Rodríguez-Mirasol J, Cordero T. HEMP-derived activated carbon fibers by chemical activation with phosphoric acid. Fuel 2009;88:19-26. [6] Puziy AM, Poddubnaya OI, Martínez-Alonso A, Suárez-García F, Tascón J.M.D. Synthetic carbons activated with phosphoric acid II. Porous structure. Carbon 2002;40:1507–1519. [7] Wu X,. Radovic LR Inhibition of catalytic oxidation of carbon/carbon composites by phosphorus. Carbon 2006;44(1):141-151. [8] McKee DW, Spiro CL, Lamby EJ. The inhibition of graphite oxidation by phosphorous additives. Carbon 1984;22(3):285–90. [9] Rosas JM, Ruiz-Rosas R, Rodríguez-Mirasol J, Cordero T. Kinetic study of the oxidation resistance of phosphorus containing activated carbons. Carbon 2012;50:1523-1537. [10] Valero-Romero MJ, Rodríguez-Mirasol J, Cordero T. Role of surface phosphorus complexes on the oxidation of porous carbons. Chapter 1. [11] Guerrero-Pérez MO, Valero-Romero MJ, Hernández S, López Nieto JM, Rodríguez-Mirasol J, Cordero T. Lignocellulosic-derived mesoporous materials: An answer to manufacturing non-expensive catalysts useful for the biorefinery processes. Catalysis Today 2012;195:155-161. [12] Bedia J, Ruiz-Rosas R, Rodríguez-Mirasol J, Cordero T. A kinetic study of 2-propanol dehydration on carbon acid catalysts. Journal of catalysis 2010;271:33-42. [13] Puziy AM, Poddubnaya OI, Socha RP, Gurgul J, Wisniewski M. XPS and NMR studies of phosphoric acid activated carbons. Carbon 2008;46: 2113-2123. [14] Puziy AM, Poddubnaya OI, Ziatdinov AM. On the chemical structure of phosphorus compounds in phosphoric acid-activated carbon. Applied Surface Science 2006;252:8036–8038 [15] Puziy AM, Poddubnaya OI, Martínez-Alonso A, Suárez-García F, Tascón JMD. Surface chemistry of phosphorus-containing carbons of lignocellulosic origin. Carbon 2005;43:2857–2868. On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments  101 [16] González-Serrano E, Cordero T, Rodríguez-Mirasol J, Cotoruelo L, Rodríguez JJ. Water Res. 2004;38:3043-3050. [17] ASTM. Standard test method for total ash content of activated carbon. D286694; 2004. [18] Brunauer S, Emmett PH, Teller E. Adsorption of gases in multimolecular layers. J Am Chem Soc 1938;60:309-319. [19] Dubinin MM, Zaverina ED, Radushkevich LV. J Phys Chem 1947;21:1351-1362. [20] Kaneko K, Ishii C. Superhigh Area Determination of Microporous Solids. Colloids Surf 1992;67(9):203-212. [21] Bedia J, Barrionuevo R, Rodríguez-Mirasol J, Cordero T. Ethanol dehydration to ethylene on acid carbon catalysts. Applied Catalysis B: Environmental 2011;103:302-310. [22] Puziy AM, Poddubnaya OI, Martínez-Alonso A, Sua´ rez-Garcı´a F, Tascón JMD. Surface chemistry of phosphorouscontaining carbons of lignocellulosic origin. Carbon 2005;43(14):2857–68. [23] Vernersson T, Bonelli PR, Cerrella EG, Cukierman AL. Arundodonax cane as a precursor for activated carbons preparation by phosphoric acid activation. Bioresour Technol 2002;83(2):95–104. [24] Guo Y, Rockstraw DA. Physicochemical properties of carbons prepared from pecan shell by phosphoric acid activation. Bioresour Technol 2007;98:1513–21. [25] Blanco Castro J, Bonelli PR, Cerella EG, Cukierman AL. Phosphoric acidactivation of agricultural residues and bagasse from sugar cane: influence of the experimental conditions on adsorption characteristics of activated carbons. Ind Eng Chem Res 2000;39:4166–72. [26] Rodríguez-Reinoso, F.,Molina-Sabio, M. Activated carbons from lignocellulosic materials by chemical and/or physical activation: an overview. Catal. Today 2010;158:89-96. [27] Bedia J, Rosas JM, Vera D, Rodríguez-Mirasol J, Cordero T. Isopropanol decomposition on carbon based acid and basic catalysts. Catalysis Today 2010;158(1-2):89-96. [28] Moreno-Castilla C, Carrasco-Marín F, Mueden A. The creation of acid carbon surfaces by treatment with (NH4)2S2O8. Carbon 1997;35(10-11):1619-1626. [29] Jagtoyen M, Derbyshire F. Activated carbons from yellow poplar and white oak by H3PO4 activation. Carbon. 1998;36(7–8):1085–97. [30] Puziy AM, Poddubnaya OI, Martınez-Alonso A, F Suárez-Garcıa, Tascón JMD. Synthetic carbons activated with phosphoric acid: I. Surface chemistry and ion binding properties. Carbon 2002:40(9);1493–1505. Chapter 3  102 [31] Rodríguez-Reinoso F, Molina-Sabio M, González MT. The use of steam and CO2 as activating agents in the preparation of activated carbons. Carbon 1995;33(1):15–23. [32] Moulder JF, Stickle WF, Sobol PE, Bomben KD, in:J. Chastain, R.C. King Jr. (Eds.), Handbook of X-ray Photoelectron Spectroscopy, Physical Electronics Inc. Eden Prairie, MN, 1995, pp. 4872-4875. [33] D. Briggs, M.P. Seah (Eds.), Practical Surface Analysis, Auger and X-ray Photoelectron Spectroscopy, vol. 1, Wiley, Chichester, UK, 1990. [34] Figueiredo JL, Pereira MFR, Freitas MMA, Órfao JJM. Modification of the surface chemistry of activated carbons. Carbon 1999;37(9): 1379–1389. [35] Zielke U, Huttinger KJ, Hoffman WP. Surface-Oxidized Carbon Fibers: I. Surface Structure and Chemistry. Carbon 1996;34:983-998. [36] Moreno-CastillaC, Carrasco-Marín F, Maldonado-Hódar FJ, Rivera-UtrillaJ. Effects of non-oxidant and oxidant acid treatments on the surface properties of an activated carbon with very low ash content. Carbon 1998; 36(1-2):145-51. [37] Marchon B, Carrazza J, Heinemann H, Somorjai GA. TPD and XPS studies of O2, CO2 and H2O adsorption on clean polycristallite graphite. Carbon 1988;26(4):507-14. [38] Bandosz TJ. in Carbon Materials for Catalysis (eds P. Serp and J. L. Figueiredo), John Wiley & Sons, Inc., Hoboken, NJ, USA. (2008) pp. 45-92, doi: 10.1002/9780470403709.ch2 [39] Montes-Morán MA, Suárez D, Menéndez JA, Fuente E. On the nature of basic sites on carbon surfaces: an overview. Carbon 2004;42(7):1219–25. [40] Peter J. Hall and Joseph M. Calo. Secondary Interactions upon Thermal Desorption of Surface Oxides from Coal Chars. Energy and Fuels 1989;3:370-376. [41] Ruowen Fu, Ling Liu, Wenqiang Huang and Pingchun Sun Studies on the structure of activated carbon fibers activated by phosphoric acid. Volume 87, Issue 14, pages 2253–2261, 1 April 2003 [42] Vivo-Vilches JF, Bailón-García E,. Pérez-Cadenas AF, Carrasco-Marín F, Maldonado-Hódar FJ. Tailoring the Surface chemistry and porosity of activated carbons: Evidence of reorganization and mobility of oxygenated Surface groups. Carbon 2014;68:520-530. [43] Rodríguez-Reinoso F. The role of carbon materials in heterogeneous catalysis. Carbon 1998;36(3):159-75 [44] Lu W, Chung DDL. Oxidation protection of carbon materials by acid phosphate impregnation. Carbon 2002;40(8):1249–54. On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments  103 [45] Wu X, Pantano CGM, Radovic LR. Inhibition of catalytic oxidation by boron and phosphorus. Extended abstracts, the international carbon conference. Lexington, (Kentucky USA): American Carbon Society; 2001. [46] Otake Y, Jenkins RG. Characterization of oxygen-containing surface complexes created on a microporous carbon by air and nitric acid treatment. Carbon 1993;31(1):109–21. [47] Gómez-Serrano V, Piriz-Almeida F, Durán-Valle CJ, PastorVillegas J. Formation of oxygen structures by air activation. A study by FT–IR spectroscopy. Carbon 1999;37:1517–28.                      Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modeling Chapter 4 Chapter 4  112 The concentration of methanol and products in the outlet gas stream were analyzed by gas chromatography (490 micro-GC equipped with PPQ, 5A molsieve and Wax columns, Agilent). The conversion was defined as the molar ratio of methanol converted to methanol fed to the reactor. The selectivity was defined as the molar ratio of a given product to that of the total products formed. The carbon balance was reached with an error lower than 5 %. 3. Results and discussion 3.1. Characterization of the carbon acid catalyst The physicochemical properties of the fresh carbon catalyst are presented in Table 1. The data presented are referred to structural parameters obtained from the N2 adsorption-desorption and CO2 adsorption isotherms, surface mass concentration obtained by XPS and surface acidity of the carbon obtained by pyridine irreversible adsorption. The carbon catalyst presents a high surface area of 1380 m2/g. The higher value of ABETN2 with respect to that of ADRCO2 suggests the existence of a wide microporous structure [45]. The high values of the mesopore volume and the external surface area confirm that the carbon catalyst presents a significant contribution of mesoporosity. The presence of both a large surface area that provides a high amount of potential active sites and a well-developed pore structure that enhances the mass transfer rate are desirable features for catalytic applications. In relation with the surface chemistry of the sample, the main elements found on the activated carbon surface was carbon (87.1 wt. %) and oxygen (9.2 wt. %), with lower amount of phosphorus (3.5 wt. %) which remained over the carbon surface despite the washing process. The high surface oxygen amount on the activated carbon, despite having been activated in an inert atmosphere (N2) to high temperatures, is associated, in a greatest extent, to the presence of phosphorus-containing groups. These surface phosphorus compounds formed during activation step seem to be homogeneously located in the carbon structure and stably bonded to carbon [28,29]. The presence of the phosphorus is supported by the binding energy of the maxima of the XPS P2P peaks, around 132.5-133.5 eV, which have been reported elsewhere as characteristic of the existence of phosphorus groups on the surface of carbon materials [23,46]. From the deconvoluted XPS P2p spectra of the carbon catalyst [30], the main peaks appear at 134.0 and 133.4 eV, suggesting the Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling  113 presence of mainly C-O-P type groups, 31.7 %, (C-O-PO3, (CO)2-PO2 and (CO)3-PO) and C-PO3/C2-PO2 groups, 46.9 %, on the activated carbon surface. The contribution of reduced phosphorus compound as C3-PO (14.3 %, 132.3 eV) and, in a lesser extent, C3P (7.1 %, 131.0 eV) groups cannot be discarded. These C-PO3 type surface groups have showed to be very reactive and are reoxidized upon contact with air, even at room temperature, forming C-O-PO3 type groups (mainly (CO)2-PO2 and (CO)3-PO)) on the surface of ACP2800 carbon [30]. This explains that relatively high carbon surface oxygen content observed for ACP2800 higher to that obtained for activated carbons prepared at lower activation temperatures [31]. The P-OH groups of these surface phosphates ((CO)2-PO2, C-PO2-O-C and/or C2PO2) act as Brönsted acid sites and as we have reported in previous works they play an important role on the acid character of the carbon catalyst for 2-butanol [35] and isopropanol [23] dehydration reactions. The total acidity surface of 1.09 mmol Pyridine adsorbed per gram of activated carbon was determined by pyridine adsorptiondesorption-TPD. This value is high if compared with other phosphoric acid activated carbons prepared at lower activation temperatures [33]. Table 1. Porous structure and chemical properties of fresh catalyst Porous structural parameters XPS - Mass surface concentration (%) Surface acidity (mmol/g)     (m2/g)     (cm3/g)   (cm3/g)     (m2/g)     (cm3/g) C1s O1s P2p Py-adsorbed 1380 0.514 0.654 660 0.265 87.1 9.2 3.5 1.09 The chemical nature of the surface sites of the catalyst has been also investigated by methanol temperature-programmed surface reaction (CH3OH-TPSR). This is a wellestablish technique to determine the chemical nature of the surface sites based on the gaseous desorption products; forlmaldehyde (CH2O) indicates the presence of redox sites, dimethyl ether (CH3OCH3, DME) acidic sites, and carbon oxides basic sites [4750]. But also, it is capable of giving information on surface intermediates [49,51,52] , as well as, monitoring the various steps of a heterogeneous catalytic process [53]. Chapter 4  114 Several publications exist on MeOH adsorption and its surface reaction on metal oxides surfaces [49,50,52] and acidic zeolites [51,53]. However, less is known about MeOH adsorption on phosphoric acid activated carbon surfaces. Figure 1 displays the intensity on m/z lines corresponding to the main detected compounds in the outlet gas during the TPD of ACP2800 after adsorption-desorption of methanol at 100 ºC. The main products were H2O, CO2, CO and H2 (Figure 1a). A blank TPD, where the activated carbon was heated in absence of methanol, showed evolution of some of these compounds but in negligible amounts, so their presence in CH3OH-TPSR can be ascribed to methanol decomposition. The formation of formaldehyde (Figure 1a) was also observed, indicating the presence of redox sites over the carbon surface. Finally, methanol condensation products, i.e., DME and light olefins, were also detected, Figure 1b, confirming the presence of acid sites. As can be observed in Figure 1a, methanol seems to be mainly adsorbed in form of molecular methanol at temperatures lower than 250 ºC. A very interesting feature is that the methanol desorption profile shows two peaks, one at 125 ºC and a broad asymmetric peak at about 163 ºC which have been previously observed for methanol adsorbed over ZSM-5 catalyst [53], TiO2 [49] and Molybdenum-Heteropolyanions [48]. Previous results on the MTG process using different techniques (TPRS [53], IR[54,55], NMR [56]) attributed the higher methanol desorption energy peak to the methanol adsorbed directly on Brönsted sites, whereas the low temperature desorption peak was associated to methanol molecules adsorbed/interacting with the methanol adsorbed directly on the Brönsted sites, thus forming a protonate cluster. However, the presence of different adsorption acid sites for methanol decomposition cannot be discarded. The products of the dehydration of methanol, DME, m/z 45, and H2O, m/z 18 were first observed at temperatures as low as 125 ºC and they were followed by ethylene (m/e 26) and propylene (41 m/e) desorption, which formation would require interaction between methanol and adsorbed DME molecules. Furthermore, DME shows a broad desorption peak between 200 and 400 ºC. This temperature region coincides with appearance of a second water peak and the onset desorption temperature of the hydrocarbons. Concentration of these compounds decayed during the ongoing TPSR, as methanol adsorbed molecules turned more isolated. The formation of these compounds at low temperature reveals the strong acidic character of the selected activated carbon. Another interesting feature is that DME peak in the CH3OH-TPSR experiment appeared at a Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling  115 temperature higher than that for the water peak, indicative that DME initially remains adsorbed on the acid active site. On the other hand, remaining adsorbed methanol started to decompose in form of CO2 at temperatures around 250 ºC and as carbon monoxide at higher temperatures. A similar TPRS profile for methanol adsorbed on H-ZSM-5 catalyst was reported by M. Jayamurzthy et al [53]. However, they also detected propane, butane and pentane along with trace amount of methane and ethane at temperatures higher to 350 ºC, indicative of the stronger acid character of H-ZSM-5 zeolite catalyst with respect to ACP2800. In fact, one of the main disadvantages for the selective formation of DME from methanol over zeolites is the large amounts of olefins formed [15]. Since the goal for this work is the selective formation of DME, the intermediate acidity of the proposed catalyst detected by MeOH-TPRS can be seen as a positive feature. Figure 1. MS results from the CH3OH-TPSR experiment over the ACP2800 activated carbon 3.2. Methanol catalytic dehydration Figure 2 represents the methanol conversion, XMeOH, as a function of time on stream (TOS) at reaction temperatures of 275, 300, 325 and 350 ºC in the presence of oxygen and at 350 ºC in the absence of oxygen (PMeOH = 0.02, W/FMeOH = 0.01 g·s/mol). The presence of phosphorus on the surface of the carbon obtained by chemical activation with phosphoric acid increases the oxidation resistance of the resulting carbon as reported previously by our research group [30,32]. Thus, this allows 100 200 300 400 500 MS signal (a.u.) T (ºC) H2O MeOH CO2 CO H2 CH2O 100 200 300 400 500 MS signal (a.u.) T (ºC) DME C3H6 C2H4 (a) (b ) Chapter 4  116 working at a reaction temperature as high as 350 ºC, under air atmosphere, without a noticeable gasification of the carbon catalyst. In the absence of oxygen, methanol conversion decreased more than 80 % within the first 15 minutes of reaction, indicating that the catalyst undergoes fast deactivation. When air is used as reaction gas, however, a steady state in which methanol conversion remained constant was reached straightaway at temperatures up to 325 ºC. At 350 ºC, steady state conditions are also achieved, although a decrease of about 20 % from the initial methanol conversion is observed after 60 min of reaction, indicating a partial deactivation of the catalyst. Figure 2. Methanol conversion as a function of TOS at different reaction temperatures in the presence (275, 30, 325 and 350 ºC) and absence (350 ºC) of air (PMeOH=0.02 atm, W/FMeOH = 0.1 g·s/mol). Table 2 summarizes the steady state conversion and selectivity values of detected carbon-based products obtained at different reaction temperatures at an inlet methanol vapor pressure of 0.02 atm and space time of 0.10 g·s/mol. When helium (He) was used as reaction gas at 350 ºC, selectivity was mainly to dimethyl ether (DME) and traces of CH4, C3H6, C2H4 and even of dimetoxymethane (DMM) and methyl formate (MF) were measured in the product stream. Given that no molecular oxygen is present in the reaction gas, the presence of DMM and MF suggest that the catalyst itself is the oxygen source for partially oxidizing methanol. Similarly, the use of 0 0.2 0.4 0.6 0.8 1 0 20 40 60 80 100 120 140 xMeOH TOS (min) Air (275 ºC) Air (300 ºC) Air (325 ºC) Air (350 ºC) He (350 ºC) Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling  117 air atmosphere yields DME as the main dehydration product within all the temperature range studied. As shown in Table 2, selectivity higher to 90 % to DME was obtained at temperatures below 350 ºC, whereas selectivities to typical products of the partial oxidation of methanol (CO2, CO, DMM and MF) are more significant than those obtained in He atmosphere and slightly increases with the reaction temperature. It must also be highlighted that in the absence of methanol and under air atmosphere no significant amounts of CO2 and CO were detected, within the range of reaction temperatures, supporting adscription of these compounds to methanol partial oxidation nor direct methanol decomposition and not to carbon gasification. Formaldehyde, which is a commonly observed product with the methanol oxidation reaction, was not detected. Formaldehyde is considered an intermediate in the formation of DMM, MF and carbon oxides [57]. Therefore, the increasing selectivity to COx with reaction temperature can be due to the oxidation of these intermediate products, and also to the strong acid sites of the activated carbon, which are able to retain the reaction intermediates products for longer time, allowing further oxidation of the intermediates. Table 2. Methanol steady state conversion and selectivities to detected carbon-based products at different reaction temperatures for reaction in air and He (PMeOH=0.02 atm, W/FMeOH=0.1 g·s/mol). T (ºC) - Atmosphere MeOH conver. (%) Selectivity (%) CH3OCH3CO2 CO (CH3O)2CH2HCOOCH3C3H6 C2H4 CH4 250-Air 7.3 94.6 1.7 1.6 1.1 0.6 0.4 - - 300-Air 19.5 93.3 2.9 2.7 0.3 0.5 0.3 - - 325-Air 29.6 91.1 3.8 3.2 1.2 0.5 0.2 - - 350-Air 43.0 87.4 5.3 3.9 2.8 0.5 0.1 - - 350-He 5.7 95.9 0.3 0.4 0.02 0.1 1.1 0.2 1.9 The evolution of methanol conversion and selectivities to different products with time on stream (TOS) at 300 ºC is shown in Figure 3. A very small change in conversion (from 22 to 19 %), and DME selectivity (from 94 to 92 %) is observed for the catalysts after 23 h on stream. This suggests that the catalyst does not suffer significant deactivation at longer reaction times, at temperatures up to 300 ºC and under the experimental conditions used in this work, indicating a high stability of the surface Chapter 4  118 acidity available and not deactivated at shorter reaction times. For higher reaction temperatures (350 ºC), however, a decrease in conversions of 10 % is observed for the catalyst after 17 h on stream (data not shown). Figure 3. Conversion of methanol and selectivity to detected carbon-based products as a function of time on stream (TOS) at 300 ºC in air atmosphere (PMeOH = 0.02 atm, W/FMeOH = 0.1 g·s/mol). The methanol dehydration reaction rate obtained in this work at 300 ºC in air (Figure 2) is 2.9·10-6 mol·g-1·s-1, which is in the range of that obtained by using modified alumina as catalysts at the same reaction temperature, as reported by Yaripour et al. [58,59]. A much higher value for the reaction rate, 140.0·10-6 mol·g-1·s-1, was reported by Mollavali et al. [60] at 300 ºC, but at a much higher pressure, 16 atm. This points out that the carbon catalyst presented in this work shows a relevantly high activity if compared to other catalysts reported in the literature, with the noticeable advantage of being obtained from an inexpensive waste. The effect of the presence of water vapor on methanol conversion and on selectivity is a key factor to analyze the potential of these carbon materials for catalyzing the dehydration of methanol to DME, especially given that water is usually found in company of methanol in the reactor feed during the industrial scale production 0.0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1 0 5 10 15 20 25 Selectivity Conversion TOS (h) MeOH DME CO2 CO DMM MF Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling  119 of DME. Figure 4 shows the methanol steady state conversion and selectivities to different products for different inlet partial pressures of water vapor (PH2O, 0.01-0.06 atm), at constant inlet methanol partial pressure of 0.02 atm and at 300 ºC in air. Both conversion and DME selectivity decrease progressively when water content in the feed is raised. The decrease of the methanol dehydration activity in the presence of water vapor has been previously reported in the literature [60,61]. Water is believed to block the active sites responsible of the methanol dehydration through competitive adsorption with methanol on the catalyst surface [2]. However, the ability of methanol to form hydrogen bonds with water molecules might also inhibit the rate of the ether formation. Taqvi et al. [62] reported the ability of methanol to form hydrogen bonds with water molecules, which results in enhanced water and methanol adsorption on activated carbons. J. Rodríguez-Mirasol et al. [63] also found that methanol adsorption is enhanced in the presence of water and ascribed the increased methanol uptake to the formation of water clusters, which were suggested to appear around the chemisorption sites that act as additional adsorption sites. More recently, J. F. DeWilde et al. [64] suggested that the formation of ethanol-water dimer species on the surface of Al2O3 is the cause of the lower dehydration activity of the catalyst in the presence of water vapor. The formation of these dimers might avoid the reaction between a methanol molecule adsorbed on the active sites and a molecule of methanol in the gas phase, which was proposed as the most probable reaction pathway for DME formation in the sight of CH3OH-TPRS results, thus reducing the dehydration activity (Figure 4). The influence of the oxygen concentration in the gas inlet on the catalytic dehydration of methanol was also studied. Figure 5 shows the conversion of methanol after 15 min of TOS as a function of temperature for different inlet oxygen concentrations. As can be observed, the methanol conversion significantly increases with oxygen concentration from a methanol conversion of 5 % in He atmosphere to 23 % in air atmosphere at 300 ºC. Moreover, in the presence of 10 % oxygen the catalyst was no longer deactivated, similarly to that observed in Figure 3. Therefore, timely supply and an excess of oxygen is necessary to keep the catalyst active. Chapter 4  120 Figure 4. Evolution of the methanol conversion and selectivity to main products with the inlet water vapor partial pressure at 300 ºC (PMeOH = 0.020 atm, W/FMeOH = 0.10 gs/µmol). Figure 5. Evolution of the methanol conversion with temperature and different oxygen partial pressures (PMeOH = 0.020 atm, W/FMeOH = 0.10 gs/µmol). 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 0 0.02 0.04 0.06 Conversion PH2O (atm) MeOH DME CO2 CO Selectivity 0.0 0.2 0.4 0.6 0.8 1.0 240 260 280 300 320 340 360 Conversion T (ºC) 0 atm 0.10 atm 0.21 atm O2 partial pressure Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling  121 All these results evidence that ACP2800 catalytic performance on the gas phase methanol conversion strongly depends on the type of atmosphere in which the reaction proceeds, and suggest that oxygen is most likely to actively intervene on the mechanism of this catalytic process inhibiting the deactivation of the catalyst, either by reacting with certain intermediates species or by interacting with the catalyst surface. In this sense, the wasted catalysts resulting from MeOH decomposition in air and He atmospheres are analyzed in detail in the following section, in order to elucidate the role of oxygen in the decomposition reaction of methanol over phosphoric acid activated carbons. 3.3. The role of oxygen on the catalyst activity 3.3.1.Characterization of used catalysts The surface chemistry of the carbon catalyst were analyzed after methanol decomposition using both He (ACP2800-RH) and air (ACP2800-RA) as reaction gases at a reaction temperature of 350 ºC after 2 h. Table 3 summarizes the mass surface concentration of the used catalysts (XPS (%wt.)). It is observed that the amount of surface phosphorus decreases from 3.5 wt. % for the fresh carbon catalyst to 2.1 and 1.7 wt. % for the catalyst after reaction in both air and He atmospheres, respectively. Moreover, when using He as the carrier, the amount of carbon increases significantly due to carbon species deposition, whereas in air the amount of surface carbon remains constant and the amount of surface oxygen increases, probably as the outcome of the oxidation of the carbon surface. Table 3. XPS mass surface concentration (%) and phosphorous surface groups distribution (%) on ACP2800 before and after reaction in air and in He (350 ºC, 2h, PMeOH = 0.02 atm; W/FMeOH = 0.1 g·s/mol) XPS (% wt.) P 2p deconvolution Sample C1s O1s P2p C-OPO3 C-PO3 and C 2 PO 2 C3PO C3P ACP2800 87.1 9.2 3.5 31.7 46.9 14.3 7.1 ACP2800-RA 85.6 11.8 2.1 33.2 47.7 17.1 2.1 ACP2800-RH 91.3 6.6 1.7 14.6 26.3 52.6 6.5 Chapter 4  128 reaction, when it is compared to ACP2800-O350 (stage 1). This loss of activity at short reaction times can be attributed to the deterioration of a fraction of the strongly acid Brönsted sites by stable coke deposits. These stable coke deposits are not oxidized at 350 ºC and it would be necessary higher temperatures for its complete combustion. Moreover, during the regeneration of the deactivated catalysts, traces of CO2 were detected in the outlet gas, supporting the deposition of a light coke on the moderate acid active sites which is slowly removed in air atmosphere under the reaction conditions used. On this question, Pedro L. Benito et al. [68] proved that deposited coke in the catalyst in the range between 300 and 400 ºC by methanol decomposition is very unstable (constituted by alkylated aromatics and oligomers) and it is possible to be partially eliminated by a degasification step (under vacuum of 10-4 mmHg at 300 ºC). They also pointed out that H-ZSM-5 catalyst under the conditions of the MTG process completely recuperates its activity after a coke combustion step with air at 550 ºC [69]. From the characterization results of the wasted catalysts and from the catalyst regeneration experiments it can be concluded that the methanol decomposition over ACP2800 activated carbon may be seen as an equilibrium between the phosphorus groups being reduced by methanol and being regenerated by oxygen, through the oxidation of C-P bonds to C-O-P ones (explaining the higher CO desorption rate of ACP2800-RA when compared to ACP2800-RH, Figure 8). On the other hand, the oxygen spill-over from oxidized P groups to the carbon surface (demonstrated by the CO desorption peak that appears at intermediate temperature in the TPD of ACP2800RA) provide reactive oxygen atoms that are able to oxidize the CxHy intermediates over the acid sites of moderate strength, forming CO, CO2 and water in the process, but they cannot oxidize the coke deposits over the strong acid sites at the temperatures employed in the reaction. 3.4. Mechanism for methanol consumption The ultimate goal of this paper, therefore, is to develop a mechanism for methanol decomposition on phosphoric acid activated carbons. Moreover, the kinetic expression obtained from the proposed kinetic model in the presence of molecular oxygen was used to obtain the kinetic and thermodynamics parameters by numerical optimization of the experimental data. Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling  129 Two types of active sites have been proposed, one acid site (L) in which the alcohol dehydration takes place, this acid site involves both, C-O-PO3 groups and P-OH acid sites. The second active site (L’) is assumed for the dissociation of molecular oxygen, which is that associated to C-PO3 groups, given that at oxidation temperatures lower than 350 ºC oxygen reacts preferentially with the phosphorus surface groups through the oxidation of C-PO3 groups to C-O-PO3 ones [30]. The reaction steps for the formation of DME, DMM, MF, CO2 and CO are given in the following reaction scheme which accounts steps of adsorption on the active sites, surface reaction and desorption of the reaction products from the active sites. Adsorption  ! " # # # # # # # # # # $ % (1) &'( ! " # # # # # # # # $ '%( (2) Surface reactions Methanol dehydration %)*+ , # # # # # # $ %& (3) &%- " # # # # # $ .&/% (4) Secondary reactions %%)012 , # # # # # # # # # # $ 345%6& (5) 345%6')* , # # # # # # # # $ &&%&(6) 345%6)7 , # # # # # # # $ %& (7) 345%6)0 , # # # # # # # $ %& (8) 345%6%)08 , # # # # # # # $ &%& (9) Desorption %9*+ : " # # # # # # # # $  (10) &&%9* : " # # # # # # # # # $ && (11) Chapter 4  130 %97 : " # # # # # # # $ (12) &%908 : " # # # # # # # # $ &(13) %90 : " # # # # # # # $  (14) Scheme 1 outlines the surface methanol dehydration mechanism proposed. Based on our previous results about alcohol dehydration on phosphoric acid activated carbons [23,40], it is proposed that methanol is first molecularly reversibly adsorbed through hydrogen bonds to the bridging hydroxyl on the Brönsted sites or through hydrogen bonds between the oxygen in C-O-P bonds and the alcohol hydroxyl group (1). DME is formed over the active sites, through a substitution reaction, SN2, between the adsorbed methanol and gas methanol by Eley-Rideal mechanism ((3) and (10)). Based on the slight reduction of the methanol conversion and selectivity to DME with increasing water vapor pressures (see Figure 4), it has been assumed that the water formed in this dehydration reaction may be adsorbed on top of the chemisorbed methanol molecules via formation of hydrogen bonds, resulting in the growth of water clusters around the chemisorption active sites, forming H2O-MeOH-L dimers (4) [63].On the other hand, adsorbed MeOH can also react with O-L' to produce formaldehyde (CH2O) and formic acid (HCOOH) intermediates. These intermediates, CH2O and HCOOH, may react with methanol or O-L’ sites to produce DMM and MF and CO, respectively, or CO2. For simplicity the intermediate species have been denoted as CHxOy in the general reaction scheme (5-9). Since all the experiments have been conducted using the same oxygen pressure, and the oxygen chemisorption and spillover are considered to be faster than the CO and CO2 formation, the amount of spillover L’O oxygen can be considered to remain invariable, and will be englobed in the ;<=1>2 and ;<>8 kinetic constants. Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling  131 Scheme1. Scheme of the proposed surface dehydration mechanism from methanol molecule adsorbed on a phosphate group. For the proposed kinetic model, it is assumed that all the adsorption reaches an equilibrium state. Moreover, desorption of the formed products is assumed to be fast, with the adsorbed amounts being negligible, thus they are not accounted in the kinetic expression. Besides, as the total conversion to DMM and MF are always lower than 1.5 %, until temperatures below 325 ºC (which is the scope of the kinetic model) the equations (6) and (7) are not considered. On the other hand, a quasi-equilibrium state is proposed for the formation and consumption rates of the 45 intermediate species. Taking these considerations into account, the kinetic rate expression for the formation of each product can be drawn from the above reaction scheme. ?@AB C ;@AB DEAF>= DGAF>= C ;@AB DHAF>=IJ DEAF>= &DGK(15) ?<> C ;<> DG<=1>2C)0D)012D !DLDMN )0O)08 (16) ?<>8C ;<>8DG<=1>2DGP>C)08D)012D !DLDMN )0O)08 (17) %?AF>= C 'D?@AB ?<> ?<>8 (18) Chapter 4  132 where the free superficial fraction coverage of Lsites is: GKC 9 9O !DL.9O)-DL-OQ012 Q0RQ08/ (19) Finally, the following suppositions were assumed: the reactor is considered as a plug flow integral reactor, homogeneous distribution of active sites on the catalyst surface, catalyst is assumed to operate at steady-state conditions, diffusional constraints and transport limitations are negligible (theoretically proven from Damköhler and Thiele modulus) and changes in temperature and pressure within the reactor are neglected. All these suppositions have been checked in previous work on alcohol dehydration where the same experimental procedure was utilized [23,40]. The dependence of the kinetic and thermodynamic parameters with the temperature was considered to follow an Arrhenius law for the kinetic constant (Eq. 21) or the Van´t Hoff law for the adsorption constants (Eq. 22) ;SC ;STUVWXIBYZ [\ ] (21) HSC HSTUVWXI^=Z [\ ] (22) Finally, the molar balance to the plug reactor is employed for obtaining the methanol conversion and the product formation at a given space time _`Z _a 7C ?S (23) Combining the molar balances, the kinetic rates equations and stoichiometric relationships, and using Runge-Kutta for the numerical solution of each differential equation, the set of differential equations can be solved, and the conversion and Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling  133 selectivity can be estimated for every given operational condition employed in the kinetic study. Thus, the kinetic parameters that rule out the proposed model were calculated by minimizing the objective function (O.F.) using the Levenberg-Marquart optimization method as implemented in Matlab 2010a software: bcbC d`efgZI`F4hZ8 ZS (24) where ijklm represents the value of the conversion obtained experimentally, and inopm the calculated value. This optimization problem was solved given an O.F. value of 8.38 x 10-4. 3.5. Kinetic study Table 5 summarizes the values of the activation energy or enthalpy of adsorption and the preexponential factors (ki0, Ki0) for all the implied reactions and equilibriums. The activation energy for the formation of DME is 85.4 kJ/mol, within the range of apparent activation energies obtained by Moreno-Castilla et al. [26] (65-110 kJ/mol) with different activated carbons oxidized with H2O2, (NH4)2S2O8 and HNO3. Xu et al. [15] reported an apparent activation energy of around 105 KJ/mol for the dehydration of methanol to DME over a -Al2O3. The obtained thermodynamic parameters fulfill the requirements of negative enthalpies of adsorption, negative standard entropies of adsorption (-42.0 and -24.7 J/mol·K for methanol and water, respectively) and standard entropies of adsorption with absolute values smaller than the respective standard entropies in gas phase (239.9 and 188.0 J/mol·K for methanol and water, respectively). The rate constants have been obtained for a reaction temperature of 325 ºC and the results are reported in the last column of Table 5. It can be observed that the kinetic constants for the formation of CO2 and CO are three orders of magnitude lower than the formation constant of the oxygenated intermediates, kCHxOy, which causes the relative amount of the intermediates being the highest in the surface of the active site. As for the water interaction with methanol, it was observed that at the highest methanol Chapter 4  134 conversion, around 6% of adsorbed methanol was unavailable due to the interaction with the formed water. Table 5. Kinetic and thermodynamic parameters obtained for methanol conversion. q m r T ,  s m r T I t r t  (mol/g·s·atm, atm-1) uo m / ^ v m I t r t  (kJ/mol) q m  w'x º y  ,  s m  w'x º y  (mol/g·s·atm, atm-1) KMeOH-L 6.3·10-03 -28.6 2.0 Kw-MeOH-L 5.1·10-02 -30.4 2.3·101 kDME 1.0·10+07 85.4 3.6 kCHxOy 2.6·10+09 132.6 6.8·10-02 kCO21.3·10+03 107.9 4.9·10-07 kCO 2.2·10+03 110.6 4.8·10-07 Figure 10a represents the steady-state methanol conversion as a function of the space time, from 0.050-0.180 gs/µmol, at an inlet methanol partial pressure of 0.02 atm and at different reaction temperatures. On the other hand, Figure 10b displays the methanol steady state conversion as a function of methanol (0.01-0.004 atm) inlet partial pressures at different temperatures and at space time of 0.10 gs/µmol. All shown data refers to experiments performed in air. The results predicted by the model are shown in solid lines for comparison sake. It is noteworthy the good agreement between experimental and calculated values provided by the kinetic model. A significant increment in methanol conversion with increasing temperature, space time and methanol partial pressure in the inlet stream can be noticed. Selectivity to DME remains very high and practically constant, although when temperature is raised there is a slight increase in the selectivities to CO2 and CO, which are the most thermodynamically favored products (data not shown). Figure 11 represents the simulated conversion, Xcal, versus the conversion obtained experimentally, Xexp for dehydration of methanol on ACP2800 catalyst, as well as the calculated and experimental DME, CO and CO2 yields (expressed as the selectivity multiplied by the methanol conversion). A proper fitting of the model to the experimental Methanol conversion and DME yield is observed in Figure 11a. However, calculated CO2 and CO (Figure 11b) yields show a slight deviation from the experimental data due to difficulties in their determination because the low produced amount, affecting the accuracy of the experimental values. Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling  135 Figure 10. Steady state methanol conversion at different temperatures as a function of the space time (PMeOH = 0.020 atm) (a) and as a function of the methanol partial pressure (W/FMeOH = 0.10 gs/µmol) (b) in air. (Symbols: experimental values, lines: calculated values). Figure 11. Calculated versus experimental conversions for (a) methanol and dimethyl ether and (b) carbon monoxide and carbon dioxide. 0.0 0.1 0.2 0.3 0.4 0.5 0.00 0.05 0.10 0.15 0.20 xMeOH W/F MeOH (g·s/mol) 275 ºC 300 ºC 325 ºC 0 0.1 0.2 0.3 0.4 0.5 0.00 0.10 0.20 0.30 0.40 0.50 Xcal X exp MeOH DME 0 0.01 0.02 0.03 0.04 0.05 0.00 0.01 0.02 0.03 0.04 0.05 Xcal X exp XCO XCO2 0.0 0.1 0.2 0.3 0.4 0.5 0 0.01 0.02 0.03 0.04 xMeOH P MeOH (atm) 275 ºC 300 ºC 315 ºC 325 ºC (a) ( b ) (a) ( b ) Chapter 4  136 4. Conclusions A carbon catalyst obtained by chemical activation of olive stone with phosphoric acid at an impregnation ratio of 2 and an activation temperature of 800 ºC has shown to be effective for the selective methanol dehydration to dimethyl ether. XPS analyses demonstrates the presence of phosphorus in form of C-O-PO3 and C-PO3 groups over the carbon surface, which confers to the carbon surface acid and redox sites as confirmed by MeOH-TPRS. The catalytic results evidence that the catalyst performance strongly depends on the type of atmosphere in which the reaction proceeds, and suggest that oxygen plays a key role on this catalytic process. In the absence of oxygen, the catalyst suffers a progressive deactivation due to both coke deposition on the strong Brönsted acid sites and to the reduction of the phosphorus groups (from C-O-P to C-P ones). However, in the presence of air, the carbon surface chemistry is modified through oxygen spillover (favoured by the presence of surface phosphorus groups) on the catalyst surface, where the availability of labile oxygen inhibit catalyst deactivation and allow methanol steady state conversion to be reached. Furthermore, the presence of oxygen leads to significant enhancement of methanol conversion without any significant change in the selectivity or reaction and even partially regenerates the wasted catalysts obtained from inert experiences. Methanol conversion and DME selectivity are slightly reduced when water vapor is added to the reactor. A kinetic study has been carried out where the methanol decomposition was supposed to proceed through an Eley-Rideal mechanism, in which the surface reaction proceeds through a substitution mechanism, SN2, between an adsorbed methanol molecule and a methanol molecule in gas phase. The rate expressions derived from the optimization of the kinetic parameters of the model properly described the experimental results, being the activation energy obtained for the formation of dimethyl ether around 85 kJ/mol. Acknowledgments We gratefully thank Junta de Andalucia (P09-FQM-5156) and Spanish Ministry of Economy and Competitiveness (MINECO) and FEDER (Project CTQ2012-364086) for financial support. M.J.V.R. gratefully thanks MINECO for a FPI fellowship (BES-2010-032213). E.M.C.M. Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling  137 gratefully thanks MECD for a FPU fellowship (AP-2012-01359). R.R.R also thanks MINECO for economic support through a “Juan de la Cierva” contract (JCI-2012-12664). Notation z{B\ |8= apparent surface area obtained by the BET method, m2g1 z@[ <>8 = apparent area of narrow micropores, m2g1 BET = Brunauer, Emmett, and Teller DME = dimethyl ether DMM = dimethoxymethane DR = Dubinin-Raduskevich Eai = activation energy of the formation of i, kJmol1 ER = EleyRideal FMeOH = initial molar flow of methanol, mol·s-1 H0ad = standard enthalpy of adsorption, J mol-1 Hi-L = enthalpy of adsorption of i on an acid site, KJ·mol-1) Hi-L’ = enthalpy of adsorption of i on C-P site, KJ·mol-1 i – L = component i bonded to L site, MeOH-L; DME-L; DMM-L; MF-L; CO-L; CO2-L i – L = component i bonded to L’ site, O-L’ ki = i formation rate constant (mol g-1 s-1) Ki-L = Adsorption equilibrium constant for i on acid site, atm-1 Ki-L’ = Adsorption equilibrium constant for i on C-P site, atm-1 Kio-L = preexponential factor of the i adsorption on acid site, atm-1 Kio-L’ = preexponential factor of the i adsorption on C-P site, atm-1 L = acid site L’= C-Psite M, MeOH = methanol MF = methyl formate NDIR = nondispersive infrared O.F. = objective function (Residual error) Pi0 = component i initial pressure, atm Pi = component i final pressure, atm ri = rate of i formation (mol s-1 g-1) R = universal gas constant, J·mol-1·K-1 Ri = rate equation for component i Si = selectivity to i product Chapter 5  144 process is carried out in inert atmosphere the catalytic activity decreases and the main reaction product is ethylene. However, they reported that when N2 was change by air, the catalysts recovered the activity and the main products were acetaldehyde and ethyl acetate. In particular, our research group has been studying the oxidation resistance and the catalytic properties of activated carbons prepared by chemical activation with phosphoric acid of various bio-renewable carbon precursors [17-19]. This activation method results in the formation of oxygen-phosphorus surface groups of high thermal stability and high surface acidity, which confers to the carbon a high oxidation resistance. These carbons have been used as catalytic supports [20-22] or as catalysts by themselves for 2-2-butanol and 2-propanol decomposition reactions, yielding mainly dehydration products [23,24]. More recently, we have studied the gas phase methanol [25] and ethanol [26] decomposition reaction over different acid carbon catalysts prepared by chemical activation of olive stone with phosphoric acid. We found that in absence of oxygen the catalyst suffer a progressive deactivation, however, the presence of oxygen produces a significant increase of the methanol and ethanol conversion and avoids deactivation of the catalysts under the operation conditions studied. In this work we study the decomposition (dehydration and dehydrogenation) of ethanol by using the acid activated carbon catalyst with the highest activity. Specifically, we examine the effect of oxygen concentration in the carrier gas, reaction temperature, ethanol inlet partial pressure and space time in the conversion and product selectivity of the ethanol decomposition reaction. A kinetic study of the ethanol decomposition on the acid carbon is also presented in which the presence and absence of water vapor in the reactor feed, in concentrations similar to that of the bio-ethanol, has been also analyzed. The corresponding kinetics and thermodynamics parameters were obtained. 2. Experimental procedure 2.1. Catalyst preparation The carbon catalyst used in this work, denoted as ACP2800, was from olive stones, an agricultural waste predominantly produces in the Mediterranean countries. Olive stones were obtained from local olive manufacturers, cleaned with deionized water, dried at 100 ºC, and ground with a roller mill to obtain samples of 1-2 mm Kinetic study of the decomposition of ethanol on carbon-based acid catalysts  145 particle size. This olive stone waste was impregnated with concentrated commercial H3PO4 (85 wt.%, Sigma Aldrich) at room temperature, using a weight ratio of 2/1 (H3PO4/olive stone), and dried for 24 h at 60 °C. The impregnated sample was activated at 800 ºC, under continuous N2 (purity 99.999%, Air Liquide) flow (150 cm3 STP/min) in a conventional tubular furnace. The activation temperature was reached at a heating rate of 10 °C/min and maintained for 2 h. The activated sample was cooled inside the furnace under the same N2 flow and then washed. In the washing process the carbonized sample was immersed in hot distilled water at 60 ºC for 2 h and afterward rinsed with room temperature distilled water at 60 °C until constant pH and negative phosphate analysis in the eluate [26]. Finally, the resulting activated carbon was dried at 100 °C, grinded and sieved (100-300 µm). 2.2. Catalyst characterization The porous structure of the carbon was characterized by N2 adsorption/desorption at -196 ºC and CO2 adsorption at 0 ºC, carried out in an ASAP 2020 model equipment of Michromeritics Instruments Corporation. Samples were previously outgassed during at least 8 hours at 150 ºC. From the N2 adsorption/desorption isotherm, the apparent surface area (ABET) was determined applying the BET equation [27], the micropore volume (Vt) and the external surface area (At) were calculated using the t-method [28] and the mesopore volume (Vmes) was obtained as the difference between the adsorbed volume at a relative pressure of 0.95 and the micropore volume (Vt) [29]. The narrow micropore surface area (ADR) and volume (VDR) were obtained by the Dubinin-Radushkevich method [30] applied to the CO2 adsorption isotherm. The surface chemistry of the carbon was analyzed by X-ray photoelectron spectroscopy (XPS), temperature-programmed desorption (TPD), adsorption and temperature-programmed desorption of ammonia (NH3-TPD). The XPS analyses were obtained using a 5700C model Physical Electronics apparatus, with MgK radiation (1253.6 eV). For the analysis of the XPS peaks, the C1s peak position was set at 284.5 eV and used as reference to position the other peaks. The fitting of the XPS peaks was done by least squares using Gaussian-Lorentzian peak shapes. TPD profiles were obtained in a custom quartz tubular reactor (i.d. 4 mm) placed inside an electrical furnace. The sample was heated from room temperature up to 900 ºC at a heating rate of Chapter 5  146 10 ºC/min in helium (purity 99.999%, Air Liquide) flow (200 cm3 STP/min).The amounts of CO and CO2 desorbed from the samples were monitored with nondispersive infrared (NDIR) gas analyzers (Siemens ULTRAMAT 22). The total acidity and acid strength distribution of the catalyst were determined by temperature programmed desorption of ammonia. The NH3-TPD was performed using 80 mg of catalyst saturated with NH3 (20% vol. in Helium) for 15 min at 100ºC. After saturation, the NH3 weakly adsorbed was desorbed in a H2 flow, at the adsorption temperature, until no NH3 was detected in the outlet gas. The NH3-TPD was performed by raising the temperature up to 500ºC at a heating rate of 10ºC/min. The NH3 was measured by mass spectrometer (Pfeiffer Omnistar GSD-301). 2.3. Ethanol decomposition The catalytic activity of the obtained activated carbon was measured by the decomposition of ethanol performed at atmospheric pressure, in a quartz fixed bed microreactor (4 mm i.d.), placed inside a vertical furnace with temperature control. In a typical experiment 150 mg of catalysts (100-300 m mesh) was used. Ethanol was fed to the system in a controlled way by using a syringe pump (Cole-Parmer® 74900-00-05 model). The reaction was carried out in different atmospheres (He, 2, 4 and 10 % vol. of O2 in He and Air) in the temperature range 250–375 ºC. To prevent condensation of ethanol and its reaction products, all the connections, from the syringe pump to the chromatograph, were heated by a wired resistance up to 120 ºC. The conditions were ethanol partial pressures from 0.01 to 0.08 atm and space times between 0.052 and 0.35 g·s/µmol (GHSV between 72 and 12 m3gas kg-3catalyst h-1). Water was also fed to the system and the water partial pressures were varied between 0.01 and 0.1 atm. The concentration of ethanol and products in the outlet gas stream were analyzed by gas chromatography (490 micro-GC equipped with PPQ, 5A molsieve and Wax columns, Agilent). In all the cases carbon mass balances were closed with errors lower than 5%. The ethanol conversion is defined as the molar ratio of ethanol converted to ethanol fed to the reactor. The selectivity is defined as: ‡mCym dymm Kinetic study of the decomposition of ethanol on carbon-based acid catalysts  147 where Ci is the molar flow of i product in the outlet stream. The carbon-containing products detected in the decomposition of ethanol (CH3CH2OH, EtOH) catalyzed by phosphoric acid activated carbons were ethylene (CH2CH2, E), diethyl ether (CH3CH2OCH2CH3, DEE) and acetaldehyde (CH3CHO, ACC). 3. Results and discussion 3.1. Catalyst characterization Catalyst selection was done based on the results of a previous work, in which different activated carbons were obtained by chemical activation of olive stone with phosphoric acid at different activation temperatures, in the interval 400-800 ºC, and impregnation ratios, between 0.5 and 2 (R, g H3PO4/g olive stone), and characterized and tested for ethanol decomposition [26]. Among the prepared catalysts, ACP2800 (R = 2 and carbonized at 800 ºC) showed the greatest conversions and lowest deactivation kinetics due to the wider porous structure and larger amount of stable surface phosphorous, both characteristics generated by the high impregnation ratio and activation temperature used for its synthesis. The physicochemical properties of the carbon catalyst are presented in Table 1. The data presented are referred to structural parameters obtained from the N2 adsorption-desorption and CO2 adsorption isotherms, surface mass concentration obtained by XPS and surface acidity of the carbon obtained by adsorption, desorption and TPD of ammonia. The carbon catalyst presents a high apparent surface area (1400 m2/g) and a wide microporous structure indicated by the higher value of ABETN2 with respect to that of ADRCO2 [31]. The carbon catalyst presents also a relatively high value of the external area (470 m2/g) and a significant contribution of mesoporosity. The mass surface concentrations measured by XPS reveal a significant amount of surface phosphorus, 3.4 % wt., as a consequence of the activation procedure, which increases with the activation temperature up to 800 ºC and decreases with the impregnation ratio [26,32-34]. It is well known that biomass activation with H3PO4 the acid is combined with organic species to form phosphate and polyphosphate bridges that connect and crosslink polymer fragments [35]. Combining XPS analyses of the P2p Kinetic study of the decomposition of ethanol on carbon-based acid catalysts  167 reaction, E2, was analyzed for ethylene formation. The rate expressions derived from the model fitted properly the experimental results, being the activation energy obtained for the formation of the main product, ethylene, around 165 kJ/mol. Acknowledgements We gratefully thank Junta de Andalucia (P09-FQM-5156) and Spanish Ministry of Economy and Competitiveness (MINECO) and FEDER (Project CTQ2012-364086) for financial support. M.J.V.R. gratefully thanks MINECO for a FPI fellowship (BES2010-032213). Notation ACC = acetaldehyde ABET = apparent surface area (m2 g-1) ADR = narrow micropore surface area (m2 g-1) At = external surface area (m2 g-1) BET = Brunauer, Emmett, and Teller Da = Damköhler number DEE = diethyl ether dp = particle diameter (cm) E = ethylene Eai = activation energy of the formation of i (KJ mol-1) EtOH = ethanol ER = EleyRideal FEtOH = initial molar flow of ethanol (mol·s-1) H0ad = standard enthalpy of adsorption (J mol-1) Hi-L = enthalpy of adsorption of i on an acid site (KJ mol-1) Hi-L” = enthalpy of adsorption of i on a basic site (KJ mol-1) ki = i formation rate constant (mol g-1 s-1) Ki-L = Adsorption equilibrium constant for i on acid site (atm-1) Ki-L” = Adsorption equilibrium constant for i on basic site (atm-1) Kio-L = preexponential factor of the i adsorption on acid site (atm-1) Chapter 5  168 Kio-L” = preexponential factor of the i adsorption on basic site (atm-1) L = acid site L” = basic site Lb = bed length (cm) n = reaction order Pep = particle Peclet number PoEtOH = ethanol partial pressure at the reactor inlet (atm-1) Pi = i partial pressure (atm-1) P/Po = relative pressure R = universal gas constant (R = 8.31 J mol-1 K-1) Rep = particle Reynolds number ri = rate of i formation (mol s-1 g-1) S0ad = standard entropy of adsorption (J mol-1 K-1) S0g = standard entropy in gas phase (J mol-1 K-1) Si = selectivity to i product STP = standard temperature pressure conditions T = temperature (ºC) TOS = time on stream (min) Vads = N2 adsorbed volume (cm3 STP·g-1) VDR = narrow micropore volume (cm3 g-1) Vmes = mesopore volume (cm3 g-1) Vt = micropore volume (cm3 g-1) w = water W = catalyst weight (g) W/FoEtOH = ethanol space time (g s mol-1) Xexp = experimental conversion Xi = conversion of/to i Xcal = calculated conversion XPS = X-ray photoelectron spectroscopy Greek letters  = interphase internal effectiveness factor ext = interphase external effectiveness factor Kinetic study of the decomposition of ethanol on carbon-based acid catalysts  169 O.F. = objective function (Residual error) i = fractional coverage of specie i on acid sites ”i = fractional coverage of specie i on basic sites References [1] Kito-Borsa T, Pacas DA, Selim S, Cowley SW. 1998. Properties of an ethanol diethyl ether water fuel mixture for cold start assistance of an ethanol-fueled vehicle. Industrial and Engineering Chemistry Research 1998; 37:3366-3374. [2] Pereira CJ. New avenues in ethylene synthesis. Science 1999;285:670-671. [3] Gucbilmez Y, Dogu T, Balci S. Ethylene and acetalehyde production by selective oxidation of ethanol using mesoporous V-MCM-41 catalysts. Industrial and Engineering Chemistry Research 2006;45:34963502. [4] Gayubo AG, Alonso A, Valle B, Aguayo AT and Bilbao J. Kinetic Model for the Transformation of Bioethanol into Olefins over a HZSM-5 Zeolite Treated with Alkali. Ind. Eng. Chem. Res. 2010; 49:10836-10844. [5] Balat M, Balat H, Oz C. Progress in bioethanol processing. Prog. Energy Combust. Sci. 2008;34:551573. [6] Navarro RM, Peña MA, Fierro JLG. Hydrogen production reactions from carbon feedstocks: Fossils fuels and biomass. Chem. ReV. 2007;107:3952-3991. [7] Ni M, Leung DYC, Leung MKH. A review on reforming bio-ethanol for hydrogen production. Int. J. Hydrogen Energy 2007;32:3238-3247. [8] Le Van Mao R, Levesque P, McLaughlin GP, Dao LH. Ethylene from ethanol over zeolite catalysts. Applied Catalysis 1987;34:163–179. [9] Le Van Mao R, Nguyen TM, McLaughlin GP. The bioethanol-to-ethylene (BETE) process. Applied Catalysis 1989;48:265–277. [10] Moser WR, Thompson W, Chiang Ch, Tong H. Silicon-rich H-ZSM-5 catalyzed conversion of aqueous ethanol to ethylene. J. Catal. 1989;117:19–32. [11] Le Van Mao R, Nguyen TM, Yao J. Conversion of ethanol in aqueous solution over ZSM-5 zeolites. Influence of reaction parameters and catalyst acidic properties as studied by ammonia TPD technique. Appl Catal. 1990;61:161–173. [12] Oudejans JC, Van Den Oosterkamp PF, Van Bekkum H, Appl. Catal. 1982;3:109–115. Chapter 5  170 [13] Gayubo AG, Aguayo AT, Olazar M, Vivanco R, Bilbao J. Kinetics of the irreversible deactivation of the HZSM-5 catalyst in the MTO process. Chem. Eng. Sci. 2003;58:5239–5249. [14] Szymaski GS, Rychlicki G, Terzyk AP. Catalytic conversion of ethanol on carbon catalysts. Carbon 1994;32:265-271. [15] Carrasco-Marín F, Mueden A, Moreno-Castilla C. Surface-treated activated carbons as catalysts for the dehydration and dehydrogenation reactions of ethanol. Journal of Physical Chemistry B 1998;102:9239-9244. [16] Grunewald GC, Drago RS. Carbon molecular sieves as catalysts and catalyst supports. Journal of American Chemical Society 1991;113:1636-1639. [17] Rosas JM, Bedia J, Rodríguez-Mirasol J, Cordero T. HEMP-derived activated carbon fibers by chemical activation with phosphoric acid. Fuel 2009;88:19-26. [18] Rosas JM, Ruiz-Rosas R, Rodríguez-Mirasol J, Cordero T. Kinetic study of the oxidation resistance of phosphorus containing activated carbons. Carbon 2012;50:1523-1537 [19] Valero-Romero MJ, Rodríguez-Mirasol J, Cordero T. Role of surface phosphorus complexes on the oxidation of porous carbons. Chapter 2. [20] Bedia J, Rosas JM, Rodríguez-Mirasol J, Cordero T. Pd supported on mesoporous activated carbons with high oxidation resistance as catalysts for toluene oxidation. Applied Catalysis B: Environmental 2010;94(1-2):8-18. [21] Guerrero-Pérez MO, Valero-Romero MJ, Hernández S, López Nieto JM, Rodríguez-Mirasol J, Cordero T. Lignocellulosic-derived mesoporous materials: An answer to manufacturing non-expensive catalysts useful for the biorefinery processes. Catalysis Today 2012;195:155-161. [22] Valero-Romero MJ, Cabrera-Molina A, Guerrero-Pérez MO, Rodríguez-Mirasol J, Cordero T. Carbon materials as template for the preparation of mixed oxides with controlled morphology and porous structure. Catalysis Today 2014;227(15):233-241. [23] Bedia J, Ruiz-Rosas R, Rodríguez-Mirasol J, Cordero T. Kinetic study of the decomposition of 2butanol on carbon-based acid catalyst. AIChE J. 2010;56:1557-1568. [24] Bedia J, Rosas JM, Márquez J, Rodríguez-Mirasol J, Cordero T. Preparation and characterization of carbon based acid catalysts for the dehydration of 2-propanol. Carbon 2009;47:286–294. [25] M. J. Valero-Romero, E.M. Calvo-Muñoz, R. Ruiz-Rosas, J. Rodríguez-Mirasol, T. Cordero. Insights into the catalytic performance of a carbon-based acid catalyst in methanol dehydration: Reaction scheme and kinetic modelling. Chapter 4.. [26] Bedia J, Barrionuevo R, Rodríguez-Mirasol J, Cordero T. Ethanol dehydration to ethylene on acid carbon catalysts. Applied Catalysis B: Environmental 2011;103:302-310. Kinetic study of the decomposition of ethanol on carbon-based acid catalysts  171 [27] S Brunauer, PH Emmett, E Teller. Adsorption of gases in multimolecular layers. J Am Chem Soc 1938, 60, 309–19. [28] BC Lippens, JH de Boer. Studies on pore systems in catalysts V. The t method. J Catal 1965, 4, 319– 23. [29] F Rodríguez-Reinoso, M Molina-Sabio, MT González. The use of steam and CO2 as activating agents in the preparation of activated carbons. Carbon 1995, 33, 15–23. [30] MM Dubinin, ED Zaverina, LV Radushkevich. Sorption and structure of active carbons. I. Adsorption of organic vapors. J Phys Chem [URSS] 1947, 21, 1351–62. [31] Rodríguez-Mirasol J, Cordero T, Rodríguez JJ. Activated carbons from CO2 partial gasification of eucalyptus kraft lignin. Energy&Fuels. 1993, 7, 133-8. [32] Rosas JM, Bedia J, Rodríguez-Mirasol J, Cordero T. Preparation of hemp-derived activated carbon monoliths.Adsorption of water vapor. Ind Eng Chem Res 2008;47(4):1288–96. [33] Puzy AM, Poddubnaya OI, Martínez-Alonso A, Suárez-García F, Tascón JMD. Synthetic carbons activated with phosphoric acid III. Carbons prepared in air. Carbon 2003;41:1181-1191. [34] Puziy AM, Poddubnaya OI, Martínez-Alonso A, Suárez-García F, Tascón JMD. Surface chemistry of phosphorus-containing carbons of lignocellulosic origin. Carbon 2005;43:2857–2868. [35] Jagtoyen M, Derbyshire F. Activated carbons from Yellow Poplar and White Oak by H3PO4 activation. Carbon 1998;36(7):1085-97. [36] Valero-Romero MJ, Rodríguez-Mirasol J, Cordero T. On the chemical nature and thermal stability of surface phosphorus groups on carbons by TPD experiments. Submitted to Carbon. [37] Wu X, Radovic LR. Inhibition of catalytic oxidation of carbon/carbon composites by phosphorus. Carbon 2006;44(1):141-151. [38] Varislia D, Dogua T, Dogub G. Ethylene and diethyl-ether production by dehydration reaction of ethanol over different heteropolyacid catalysts. Chemical entineering Science. 2007;62(18-20):53495352. [39] Chen G, Li S, Jiao F, Yuan Q. Catalytic dehydration of bioethanol to ethylene over TiO2/g-Al2O3 catalysts in microchannel reactors. Catalysis Today 1007;125:111-119. [40] Varisli D, Dogu T, Dogu G. Silicotungstic acid impregnated MCM-41-like mesoporous solid acid catalysts for dehydration of ethanol. Ind. Eng. Chem. Res. 2008:47;4071–4076. [41] Zaki T. Catalytic dehydration of ethanol using transition metal oxide catalysts. J. Colloid Interface Sci. 2005;284:606-613. Chapter 5  172  !"#"$%&'()*'+,--&-560. [43] Knözinger H, Köhne R. The Dehydration of Alcohols over Alumina: I. The reaction scheme. Journal of Catalysis. 1966;5(2):264-270. [44] Matthew A. Christiansen, Giannis Mpourmpakis, and Dionisios G. Vlachos. Density Functional Theory-Computed Mechanisms of Ethylene and Diethyl Ether Formation from Ethanol on  Al2O3(100). ACS Catal. 2013(3)1965-1975. [45] Winter O, Ming-Teck E. Make ethylene from ethanol. Hydrocarbon Processing 1976;125-133 [46] Phillips CB, Datta R. Production of ethylene from hydrous ethanol on H-ZSM-5 under mild conditions. Ind. Eng. Chem. Res. 1997;36:4466-4475. [47] Nguyen TM, Mao RLV. Conversion of ethanol in aqueous solution over ZSM-5 zeolites. Appl. Catal. 1990;58:119-129. [48] Zhang X, Wang R, Yang X, Zhang Fo. Comparison of four catalysts in the catalytic dehydration of ethanol to ethylene. Microporous and Mesoporous Materials 2008;(116):1-3)210-215. [49] Zhou B, Machej T, Ruiz P, Delmon B. Catalytic cooperation between MoO3 and Sb2O4 in N-ethylformamide dehydration. II. Nature of active sites and role of spill-over oxygen. J. Catal. 1991;132:183199. [50] Weng LT, Ruiz P, Delmon B and Duprez D. Evidence of the migration of oxygen species from Sb2O4 to MoO3 in MoO3-Sb2O4. J. Mol. Catal. 1989;52:349-360. [51] Mul G, Neeft JPA, Kapteijn F, Moulijn JA. The formation of carbon surface oxygen complexes by oxygen and ozone. The effect of transition metal oxides Carbon 1998;36:1269-1276. [52] Knozinger H, Khone R. The dehydration of alcohols over alumina: I. The reaction scheme. Journal of Catalysis 1996;5:264-270. [53] Knozinger H, Buhl H, Kochloefl K. The dehydration of alcohols on alumina: XIV. Reactivity and mechanism. Journal of Catalysis 1972;24:57-68. [54] Macht J, Janik MJ, Neurock M, Iglesia E. Catalytic Consequences of Composition in Polyoxometalate Clusters with Keggin Structure. Angewandte Chemie – International Edition 46 (2007) 7864-7868. [55] Macht J, Janik MJ, Neurock M, Iglesia E. Mechanistic consequences of composition in acid catalysis by polyoxometalate Keggin clusters. Journal of the American Chemical Society 2008;130:10369-10379. [56] Janik MJ, Macht J, Iglesia E, Neurock M. Correlating Acid Properties and Catalytic Function: A First-Principles Analysis of Alcohol Dehydration Pathways on Polyoxometalates. Journal of Physical Chemistry C 2009;113(5):1872-1885.