Biofiltración de compuestos orgánicos volátiles utilizando diferentes tipos de lecho basados en mezclas de compost de RSU-poda: monitorización mediante diversas técnicas
Abstract
Composting is one of the most successful biological processes for the treatment of the residues enriched in putrescible materials. The optimization of parameters which have an influence on the stability of the products is necessary in order to maximize recycling and recovery of waste components. The influence of the composting process parameters (aeration, moisture, C/N ratio, and time) on the stability parameters (organic matter, N-losses, chemical oxygen demand, nitrate, biodegradability coefficient) of the compost was studied. The composting experience was carried out using Municipal Solid Waste (MSW) and Legume Trimming Residues (LTR) in 200 L isolated acrylic barrels following a Box-Behnken central composite experimental design. Second-order polynomial models were found for each of the studied compost stability parameter, which accurately described the relationship between the parameters. The differences among the experimental values and those estimated by using the equations never exceeded 10% of the former. Results of the modelling showed that excluding the time, the C/N ratio is the strongest variable influencing almost all the stability parameters studied in this case, with the exception of N-losses which is strongly dependent on moisture. Moreover, an optimized ratio MSW/LTR of 1/1 (w/w), moisture content in the range of 40-55% and moderate to low aeration rate (0.05-0.175 Lair kg-1 min-1) is recommended to maximise degradation and to obtain a stable product during co-composting of MSW and LTR. On the other hand, the concentration of volatile organic compounds (VOCs) during the composting of kitchen waste and pruning residues in a pilot plant, and the abatement of VOCs by different compost biofilters was studied. VOCs removal efficiencies greater than 90% were obtained using composts of municipal solid waste (MSW) or MSW-pruning residue as biofilter material. An electronic nose identified qualitative differences among the biofilter output gases at very low concentrations of VOCs. These differences were related to compost constituents, compost particle size (2-7 or 7-20 mm), and a combination of both factors. The total concentration of VOCs determined by a photoionization analyser and inferred from electronic nose data sets were correlated over an ample range of concentrations of VOCs, showing that these techniques could be specially adapted for the monitoring of these processes. Finally, a biofiltration system was designed using mature composts of municipal solid waste (MSW) or MSW mixed with pruning residues (MSW-P) as packing materials to treat vapours of ¿-pinene (a dominant volatile organic compounds (VOC) emitted during the MSW- P co-composting) and Methyl ethyl ketone (MEK). Monitoring the efficiency of the biofiltration system was carried out using a photoionization analyser, a commercial electronic nose (e-nose) and gas chromatography ¿ mass spectrometry (GC/MS). Results indicated that removal efficiencies for both kinds of biofilters were greater than 90% at different stages of the experiment. The acclimatization periods were 10 and 25 days for the MSW biofilter and MSW-P biofilter, respectively in the experiment with ¿-pinene and 14 days for both biofilters when the MEK was the contaminant present in the air stream to treat. Removal efficiency of the system was strongly dependent upon the moisture content of the packing materials, but the nature of the contaminant (hydrophilic or hydrophobic) plays an important role in the degradation. As moisture content in the biofilters fell to below 66% (dw) for the MSW and 51% (dw) for MSW-P, the removal efficiency decreased to less than 90% when the contaminant was an hydrophobic volatile organic compound (¿-pinene). In the case of MEK, the optimal range of moisture content for the packing materials used in this experiment was between 21.5 and 67.5 % (d.w) in the case of the MSW biofilter and between 25.7 and 91.8 % (d.w) for MSW-P biofilter. E-nose and GC/MS data indicate a complete degradation of the ¿-pinene and MEK by biofiltration, although the e-nose did detect background emissions, characteristic of each type of biofilter. Also, was possible to obtain information about the performance of the biofiltration system in different stages of the experiment. Results suggest that e-nose¿s will become a more powerful tool for monitoring VOC compounds in biofiltration and composting processes in the future.
Full text
BIOFILTRACIÓN DE COMPUESTOS ORGÁNICOS VOLÁTILES UTILIZANDO DIFERENTES TIPOS DE LECHO BASADOS EN MEZCLAS DE COMPOST DE RSU-PODA: MONITORIZACIÓN MEDIANTE DIVERSAS TÉCNICAS Sevilla, Abril de 2013 Memoria que presenta Iván Orlando Cabeza Rojas para optar al título de Doctor por la Universidad de Sevilla
BIOFILTRACIÓN DE COMPUESTOS ORGÁNICOS VOLÁTILES UTILIZANDO DIFERENTES TIPOS DE LECHO BASADOS EN MEZCLAS DE COMPOST DE RSU-PODA: MONITORIZACIÓN MEDIANTE DIVERSAS TÉCNICAS Memoria presentada por IVÁN ORLANDO CABEZA ROJAS Para optar al grado de Doctor por la Universidad de Sevilla Visado en Sevilla, a 10 de Abril de 2013 LOS DIRECTORES EL TUTOR EL DOCTORANDO Iván Orlando Cabeza Rojas Dra. Dña. María Elena Fernández Boy Profesora Titular de La Universidad de Sevilla Dr. D. Manuel Jesús Díaz Blanco Profesor Titular de La Universidad de Huelva Dr. D. Rafael López Núñez Científico Titular IRNAS-CSIC
Dr. D. José Manuel Pardo Prieto, DIRECTOR DEL INSTITUTO DE RECURSOS NATURALES Y AGROBIOLOGÍA DE SEVILLA DEL CONSEJO SUPERIOR DE INVESTIGACIONES CIENTÍFICAS Certifica: Que la presente Memoria de Investigación titulada “Biofiltración de Compuestos Orgánicos Volátiles Utilizando Diferentes Tipos de Lecho Basados en Mezclas de Compost de RSU-Poda: Monitorización Mediante Diversas Técnicas”, presentada por D. Iván Orlando Cabeza Rojas para optar al Título de Doctor por la Universidad de Sevilla, ha sido realizada en el Departamento de Protección del Sistema Suelo, Planta, Agua , bajo la dirección del Dr. D. Rafael López Núñez, reuniendo todas las condiciones exigidas a los trabajos de Tesis Doctorales. En Sevilla, a 10 de Abril de 2013 Fdo: D.José Manuel Pardo Prieto.
ÍNDICE Summary I 1. INTRODUCCIÓN 1 1.1. PROBLEMÁTICA DE LOS RESIDUOS 1 1.1.1. RESIDUOS SÓLIDOS URBANOS (RSU) 4 1.1.1.1. Composición de los residuos sólidos urbanos 6 1.1.1.2. Tratamiento de los residuos orgánicos 8 1.2. EL COMPOSTAJE Y EL COMPOST 12 1.2.1. PROCESO DE COMPOSTAJE 12 1.2.1.1. Parámetros del Proceso de Compostaje 15 1.2.2. EL COMPOST 18 1.3. ACTIVIDADES QUE CAUSAN EMISIÓN DE OLORES 20 1.3.1. OLORES EN PLANTAS DE COMPOSTAJE DE RESIDUOS SÓLIDOS URBANOS 22 1.3.1.1. Efectos de los olores en la salud 26 1.4. MÉTODOS DE CONTROL DE OLORES 27 1.5. BIOFILTRACIÓN 31 1.5.1. BIOFILTRO 33 1.6. MÉTODOS DE CARACTERIZACIÓN DE COVs Y OLORES 41 1.6.1. CROMATOGRAFÍA DE GASES 42 1.6.2. NARIZ ELECTRÓNICA 44 1.6.3. DETECTORES DE IONIZACIÓN 48 2. OBJETIVOS Y PLAN DE TRABAJO 57 2.1. OBJETIVOS 57 2.2. PLAN DE TRABAJO 58 3. MAXIMISING MUNICIPAL SOLID WASTE – LEGUME TRIMMING RESIDUE MIXTURE DEGRADATION IN COMPOSTING BY CONTROL PARAMETERS OPTIMIZATION 65 3.1. INTRODUCTION 65 3.2. MATERIALS AND METHODS 68 3.2.1. COMPOSTING MATERIALS 68 3.2.2. EXPERIMENTAL DESIGN FOR THE COMPOSTING PROCESS OPTIMIZATION 70 3.2.3. STATISTICAL ANALYSIS 70 3.2.4. COMPOSTING REACTORS 73 3.2.5. ANALYTICAL METHODS 74 3.3. RESULTS AND DISCUSSION 76 3.3.1. PARETO CHART ANALYSIS 77 3.3.2. ANALYSIS OF COMPOST STABILITY PARAMETERS 79 3.3.2.1. Organic Matter 79 3.3.2.2. Degradability Coefficient (km) 81 3.3.2.3. Nitrate Content 82 3.3.2.4. N-Losses 84 3.3.2.5. Chemical Oxygen Demand (DQO) 86 3.4. CONCLUSIONS 88 4. BIOFILTRATION OF COMPOSTING GASES USING DIFFERENT MUNICIPAL SOLID WASTE-PRUNING RESIDUE COMPOSTS: MONITORING BY USING AN ELECTRONIC NOSE 93 4.1. INTRODUCTION 93 4.2. MATERIALS AND METHODS 96 4.2.1. COMPOSTING PROCESS 96 4.2.2. BIOFILTRATION UNIT 97 4.2.3. TOTAL – VOCs ANALYSIS 100
4.2.4. ELECTRONIC NOSE MEASUREMENTS 100 4.2.5. AIR SAMPLING 101 4.2.6. STATISTICAL ANALYSIS 101 4.3. RESULTS AND DISCUSSION 102 4.4. CONCLUSIONS 116 5. BIOFILTRATION OF α-PINENE VAPOURS USING MUNICIPAL SOLID WASTE (MSW) – PRUNING RESIDUES (P) COMPOSTS AS PACKING MATERIALS 120 5.1. INTRODUCTION 120 5.2. MATERIALS AND METHODS 123 5.2.1. BIOFILTRATION UNIT 123 5.2.2. VOCs ANALYSER MEASUREMENTS 126 5.2.3. ELECTRONIC NOSE ANALYSIS 127 5.2.4. GC/MS ANALYSIS 128 5.3. RESULTS AND DISCUSSION 130 5.4. CONCLUSIONS 143 6. TREATMENT OF GASES CONTAMINATED WITH METHYL ETHYL KETONE (MEK) BY BIOFILTERS PACKED WITH MUNICIPAL SOLID WASTE – PRUNING RESIDUES COMPOSTS: PERFORMANCE ASSESSMENT 148 6.1. INTRODUCTION 148 6.2. MATERIALS AND METHODS 153 6.2.1. BIOFILTRATION UNIT 153 6.2.2. VOCs ANALYSER MEASUREMENTS 157 6.2.3. ELECTRONIC NOSE ANALYSIS 158 6.2.4. GC/MS ANALYSIS 160 6.3. RESULTS AND DISCUSSION 161 6.4. CONCLUSIONS 176 7. GENERAL CONCLUSIONS 183
SUMMARY I SUMMARY Composting is one of the most successful biological processes for the treatment of the residues enriched in putrescible materials. The optimization of parameters which have an influence on the stability of the products is necessary in order to maximize recycling and recovery of waste components. The influence of the composting process parameters (aeration, moisture, C/N ratio, and time) on the stability parameters (organic matter, N-losses, chemical oxygen demand, nitrate, biodegradability coefficient) of the compost was studied. The composting experience was carried out using Municipal Solid Waste (MSW) and Legume Trimming Residues (LTR) in 200 L isolated acrylic barrels following a Box-Behnken central composite experimental design. Second-order polynomial models were found for each of the studied compost stability parameter, which accurately described the relationship between the parameters. The differences among the experimental values and those estimated by using the equations never exceeded 10% of the former. Results of the modelling showed that excluding the time, the C/N ratio is the strongest variable influencing almost all the stability parameters studied in this case, with the exception of N-losses which is strongly dependent on moisture. Moreover, an optimized ratio MSW/LTR of 1/1 (w/w), moisture content in the range of 40-55% and moderate to low aeration rate (0.05-0.175 Lair kg-1 min-1) is recommended to maximise degradation and to obtain a stable product during co-composting of MSW and LTR. On the other hand, the concentration of volatile organic compounds (VOCs) during the composting of kitchen waste and pruning residues in a pilot plant, and the abatement of VOCs by different compost biofilters was studied. VOCs removal efficiencies greater than 90% were obtained using composts of municipal solid waste (MSW) or MSW- pruning residue as biofilter material. An electronic nose identified qualitative differences among the biofilter output gases at very low concentrations of VOCs. These differences were related to compost constituents, compost particle size (2-7 or 7-20 mm), and a
SUMMARY II combination of both factors. The total concentration of VOCs determined by a photoionization analyser and inferred from electronic nose data sets were correlated over an ample range of concentrations of VOCs, showing that these techniques could be specially adapted for the monitoring of these processes. Finally, a biofiltration system was designed using mature composts of municipal solid waste (MSW) or MSW mixed with pruning residues (MSW-P) as packing materials to treat vapours of α-pinene (a dominant volatile organic compounds (VOC) emitted during the MSW- P co-composting) and Methyl ethyl ketone (MEK). Monitoring the efficiency of the biofiltration system was carried out using a photoionization analyser, a commercial electronic nose (e-nose) and gas chromatography – mass spectrometry (GC/MS). Results indicated that removal efficiencies for both kinds of biofilters were greater than 90% at different stages of the experiment. The acclimatization periods were 10 and 25 days for the MSW biofilter and MSW-P biofilter, respectively in the experiment with α-pinene and 14 days for both biofilters when the MEK was the contaminant present in the air stream to treat. Removal efficiency of the system was strongly dependent upon the moisture content of the packing materials, but the nature of the contaminant (hydrophilic or hydrophobic) plays an important role in the degradation. As moisture content in the biofilters fell to below 66% (dw) for the MSW and 51% (dw) for MSW-P, the removal efficiency decreased to less than 90% when the contaminant was an hydrophobic volatile organic compound (α-pinene). In the case of MEK, the optimal range of moisture content for the packing materials used in this experiment was between 21.5 and 67.5 % (d.w) in the case of the MSW biofilter and between 25.7 and 91.8 % (d.w) for MSW-P biofilter. E-nose and GC/MS data indicate a complete degradation of the α-pinene and MEK by biofiltration, although the e-nose did detect background emissions, characteristic of each type of biofilter. Also, was possible to obtain information about the performance of the biofiltration system in different stages of the experiment. Results suggest that e-nose’s will become a more powerful
INTRODUCCIÓN 6 selectivamente2 , seguidas de las comunidades de Madrid y Andalucía, las cuales reportaron cantidades superiores a 300000 t. (MARM, 2011) 1.1.1.1. Composición de los Residuos Sólidos Urbanos La correcta gestión de los residuos sólidos urbanos depende directamente de la composición de los mismos. Esta composición varía en función de tres factores, el nivel de vida de la población, la actividad desarrollada por esta y la climatología propia de la región. Por lo que es claro que existe una relación entre la composición de los residuos sólidos urbanos y el poder adquisitivo de cada colectividad (Thitame et al., 2010). En el caso particular de la UE, la evolución experimentada por la sociedad ha hecho que los residuos orgánicos, tradicionalmente la fracción mayoritaria, hayan dado paso a nuevos productos procedentes principalmente de los embases y embalajes. En España, la tendencia de los productos que conforman la producción de los residuos sólidos urbanos se observa en la Figura 1.3. 2 Residuos recogidos selectivamente: Son el resultado de la recogida diferenciada de materiales orgánicos fermentables y de materiales reciclables, así como cualquier otro sistema de recogida diferenciada que permita la separación de los materiales valorizables contenidos en los residuos. No incluyen residuos recuperados en las plantas de triaje y clasificación.
INTRODUCCIÓN 7 Figura 1.3. Composición RSU en España 2009 (INE, 2010) A continuación se describen los principales componentes de los residuos sólidos urbanos en España • Materia Orgánica: A pesar de la disminución que conlleva el desarrollo de la sociedad, continúa siendo el principal componente de los residuos. Proviene de los restos de alimentos y de las actividades vinculadas a la jardinería principalmente. • Papel y Cartón: La recogida en origen de este componente ha experimentado un importante incremento. Algunos ejemplos son los periódicos, las cajas, embalajes de cartón, etc. • Plástico: Se encuentra en forma de envases, embalajes, piezas de carrocería de coches y elementos de diversa índole. Debido a su versatilidad, bajo coste, facilidad de producción y resistencia a los factores ambientales es usado en diversos sectores industriales en la fabricación de una amplia gama de productos.
INTRODUCCIÓN 8 • Vidrio: La cantidad de este componente de los residuos urbanos recogida fue de 804000 t durante el 2010, lo que equivale a 17.1 kg/cap (INE, 2010). • Textil: Durante el 2010 se recogieron 4.700 t de este material en España (INE, 2010) • Metales: La hojalata, empleada en el sector alimentario e industrial, es el principal compuesto derivado del hierro que se encuentra en los residuos urbanos. A su vez, el aluminio, utilizado en los botes de bebidas carbonatadas y en los tetra-brik, es por su parte el material no férrico de mayor abundancia. • Maderas: Constituyen partes de mobiliario fundamentalmente. En España durante 2010 se produjeron 192.100 t de este tipo de residuos (INE, 2010). • Otros: Este grupo tiene una composición heterogénea y por la naturaleza de algunos de sus componentes es digno de una atención especial por ser considerados residuos peligrosos. 1.1.1.2. Tratamiento de los Residuos Sólidos Urbanos La eliminación de los residuos sólidos urbanos (RSU) es uno de los problemas más importantes y controvertidos que enfrentan los gobiernos locales a nivel mundial (Assamoi & Lawryshyn, 2012). Como se ha mencionado anteriormente, el aumento de la generación de residuos debido al crecimiento demográfico, los cambios sociales de estilo de vida, el desarrollo y el consumo
INTRODUCCIÓN 9 de productos menos biodegradables, han llevado a que la gestión y tratamiento de los RSU en varias ciudades de todo el mundo sea un verdadero reto (Asase et al., 2009). La gestión y destino final de los residuos puede causar diversos efectos sobre la salud y el medio ambiente. Una buena gestión puede proteger la salud pública y la calidad del medio ambiente, contribuyendo a la conservación de los recursos naturales y la economía. A su vez, el abandono o la gestión inadecuada de los residuos producen impactos notables, generando contaminación en el agua, aire, suelo, contribuyendo al cambio climático y afectando a los ecosistemas y a la salud humana. La gestión de los residuos es considerado un indicador mixto ya que contiene variables de presión (vertido de residuos) y de respuesta (valorización de residuos) (OSE, 2011). Cuando se habla de gestión de residuos sólidos urbanos es necesario diferenciar la etapa de recogida y la etapa de tratamiento posterior y/o eliminación, ya que la segunda etapa está fuertemente condicionada por el grado de selectividad de las operaciones de recogida. Sin embargo, actualmente gran parte de los residuos sólidos urbanos sigue eliminándose sin ningún tipo de aprovechamiento en vertederos (Figura 1.4). Según datos de Eurostat, en España durante el año 2009, se depositaron en vertedero 285 kg/cap frente a los 191 kg/cap de la UE-27.
INTRODUCCIÓN 10 Figura 1.4. Gestión de residuos en España año 2009 (MARM, 2011) Una parte de la fracción orgánica de los residuos sólidos urbanos aunque mezclada con otros residuos, se destina a instalaciones de tratamiento mecánico-biológico (compostaje o biometanización). En España en el año 2009, se trataron 9.108.845 t de residuos sólidos urbanos en instalaciones de triaje y compostaje y 3.393.374 t en instalaciones de triaje, biometanización y compostaje (OSE, 2011). El compostaje es una de las formas de valorización de los residuos más empleada y que requiere especial interés (Iqbal et al., 2010; Saha et al., 2008). En el año 2008 se registraron en España, 66 instalaciones de triaje y compostaje distribuidas mayoritariamente en Andalucía. En conjunto con la Comunitat Valenciana y Comunidad de Madrid fueron las que destinaron una mayor cantidad de residuos urbanos a la producción de compost. Cabe anotar que solo un 25% del material de entrada en estas instalaciones fue recuperado y que se utilizó fracción orgánica de recogida selectiva junto con materia
INTRODUCCIÓN 11 orgánica procedente de la fracción de residuos mezclados, lo que supone la obtención de un compost de mala calidad, con usos restringidos (MARM, 2011). La otra opción de tratamiento mecánico-biológico es la que se lleva a cabo en instalaciones de triaje, biometanización y compostaje. En España se registraron durante el 2008, 15 instalaciones, presentes la mayoría en Castilla y León y Cataluña, las cuales recuperaron el 34.5 % de los residuos de entrada (MARM, 2011). Otro tipo de valorización de los residuos es la incineración con recuperación energética. En la cual los residuos son quemados para producir energía y calor. Sin embargo, presenta muchas desventajas como altos coste de montaje y operación, producción de emisiones contaminantes y formación de cenizas en cuya eliminación siguen participando los vertederos tradicionales (Assamoi & Lawryshyn, 2012; Lam et al., 2010; Yang et al., 2012). En España durante el 2009 se incineraron 48 kg/cap de residuos con una tendencia al alza durante el periodo 1995-2009. Al mismo tiempo, la UE-27 valorizó en este mismo año 102 kg/cap de residuos utilizando la técnica de incineración (OSE, 2011). A su vez, es importante tener en cuenta que si no se suman esfuerzos en la optimización de los diversos procesos de tratamiento, los problemas asociados a la producción y la disposición de los residuos pueden persistir. Por lo que los esfuerzos de la comunidad científica deben centrarse en la minimización de los efectos asociados al tratamiento de los residuos sólidos humanos como (Chen, 2010; Ragazzi et al., 2011; Rigamonti et al., 2010):
INTRODUCCIÓN 12 1. Contaminación de suelos. 2. Contaminación de acuíferos por lixiviados. 3. Contaminación de las aguas superficiales. 4. Emisión de gases de efecto invernadero. 5. Creación de focos infecciosos. Proliferación de plagas de roedores e insectos. 6. Producción de malos olores. Sin embargo, a pesar de todos los esfuerzos realizados en los diversos tratamientos utilizados en la gestión de los residuos sólidos urbanos, múltiples trabajos concluyen que energética y ambientalmente hablando la mejor solución es el reciclaje y la concienciación de la comunidad sobre el problema (Rigamonti et al., 2010). 1.2. EL COMPOSTAJE Y EL COMPOST 1.2.1. PROCESO DE COMPOSTAJE El compostaje es definido como la descomposición biológica y estabilización de sustratos orgánicos, bajo condiciones que permiten el desarrollo de temperaturas termófilas como resultado del calor producido biológicamente, con el fin de producir un producto final que es estable, libre de patógenos y semillas de plantas, el cual puede ser beneficiosamente aplicado al suelo (Haug, 1993). En términos técnicos, se puede decir, que el proceso de
INTRODUCCIÓN 13 compostaje consiste en la degradación de la materia orgánica mediante su oxidación y la acción de diversos microorganismos presentes en los propios residuos (Figura 1.5). Figura 1.5. Ecuación Bioquímica General del proceso de compostaje. Hay que tener en cuenta que el proceso de compostaje (Haug, 1993): • Es biooxidativo y controlado: Exige una determinada condición biológica, que lo hace diferente tanto de otros procesos físicos y químicos, como de todos aquellos que no se realicen de forma aeróbica. • En él intervienen diversos microorganismos: Es un proceso microbiológico influido por la naturaleza de los organismos presentes (bacterias, hongos y actinomicetos). • Implica sustratos orgánicos heterogéneos en estado sólido. • Requiere una etapa termófila: Durante esta etapa, ocurre la eliminación de patógenos y se inhibe la producción inicial de fitotoxinas. • Debe producir un material orgánico estabilizado: Un material estabilizado tendrá un alto valor fertilizante para ser empleado en agricultura. La utilización como fertilizante o enmienda en la agricultura del producto resultante del proceso de compostaje de residuos sólidos urbanos es una práctica común en diversos países (Barral et al., 2009; Castillejo & Castelló, 2010; Montemurro et al., 2005; Ostos et al., 2008; Warman & Margarit, 2012).
INTRODUCCIÓN 14 Es por esta razón, que el proceso de compostaje no debe transcurrir espontáneamente, por el contrario, deben controlarse los parámetros que afectan el proceso con el fin de garantizar la valorización final de los residuos obteniendo un producto de calidad a través de este método de tratamiento biológico (Mokhtari et al., 2011; Petric et al., 2012; Vasarevičius et al., 2011). El proceso de compostaje está dividido en dos fases fundamentales (Soliva, 2001): • Fase de Descomposición: En esta fase, ocurre la degradación de las moléculas complejas a moléculas orgánicas más sencillas. Es un proceso exotérmico debido a la actividad biológica y está compuesta por dos etapas, etapa mesófila con temperaturas hasta los 45 ºC y una etapa termófila con temperaturas que pueden llegar a los 70 ºC. La monitorización durante esta fase del proceso es importante, condicionando las propiedades del producto final. • Fase de Maduración: Se compone de dos etapas, una denominada enfriamiento con temperaturas que van desde los 40 ºC y una llamada estabilización con una baja actividad microbiana y donde se presenta la aparición de organismos superiores. Esta fase a diferencia de la anterior no requiere especial control ya que la actividad biológica es mucho más reducida y los requerimientos de oxígeno muy inferiores.
INTRODUCCIÓN 15 1.2.1.1. Parámetros del Proceso de Compostaje. Como se dijo anteriormente, el proceso de compostaje es un sistema complejo en donde los microorganismos cumplen un papel fundamental, por lo que todos los factores que puedan limitar su desarrollo serán limitantes del propio proceso (Barrena-Gomez, 2006). Aunque los parámetros de control que más influyen en el proceso de compostaje pueden variar dependiendo del tipo de materia prima o sustrato a tratar. Diversos trabajos de investigación han concluido que los más importantes son: La temperatura, humedad y porosidad, pH, aireación y cantidad de nutrientes (Relación C/N) (Delgado-Rodríguez et al., 2010; Madejón et al., 2002; Petric et al., 2012; Yañez et al., 2010). En la tabla 1.1 se describen dichos parámetros. Tabla 1.1. Parámetros de control del proceso de compostaje (Barrena-Gomez, 2006; Haug, 1993; Soliva, 2001) PARÁMETROS INFLUENCIA EN EL PROCESO TEMPERATURA Proporciona información directa del funcionamiento del mismo (indica incremento en la actividad biológica) El mantenimiento de temperaturas elevadas asegura la higienización del material pero puede inhibir la actividad microbiana, por lo que se debe procurar un equilibrio de temperaturas entre la máxima higienización
INTRODUCCIÓN 22 Es claro que las actividades de tratamiento biológico de residuos sólidos urbanos como el compostaje son procesos susceptibles a la generación de olores que afecten a los trabajadores y la comunidad en general (López et al., 2011). 1.3.1. OLORES EN PLANTAS DE COMPOSTAJE DE RESIDUOS SÓLIDOS URBANOS Una de las principales amenazas del éxito operativo de la mayoría de instalaciones de compostaje es el potencial inherente del proceso de crear olores desagradables, así como, la generación de polvo y el favorecer la presencia de insectos, aves y roedores. Aunque todos estos problemas pueden ser mermados con el correcto control de las variables operacionales del proceso como por ejemplo, la humedad (en el caso del polvo), el aseguramiento de la fase termófila (destrucción de los huevos de insecto), el correcto manejo de las materias primas del compostaje especialmente cuando se trabaja con residuos sólidos urbanos y el aseguramiento de las condiciones aerobias del proceso (Evita la formación de algunos compuestos que generan mal olor (Stoffella & Kahn, 2005) ). El control de los olores es indudablemente el problema más difícil de controlar en la práctica del compostaje (Haug, 1993). La emisión de olores se inicia normalmente con la recepción de las materias primas del proceso (residuos) a la planta de tratamiento y sobre todo en las fases iniciales del compostaje. A su vez, en el caso de producirse condiciones anaeróbicas en el proceso, debido al escaso control de la aireación, se producirán compuestos de azufre de olor intenso. De la misma forma, una
INTRODUCCIÓN 23 degradación aeróbica incompleta producirá emisiones de alcoholes, cetonas, ésteres y ácidos orgánicos; Por otra parte, un balance de nutrientes equivocado puede dar lugar a emisiones de compuestos orgánicos volátiles y amoniaco (Bianchi et al., 2010). Existen diversos compuestos generadores de olor emitidos por las plantas de tratamiento de residuos. En la Tabla 1.2 se resumen los principales compuestos detectados en dichas instalaciones. Tabla 1.2. Grupo de compuestos emitidos en plantas de tratamiento de residuos (Bidlingmaier & Müsken, 2007; Delgado-Rodríguez et al., 2011; Haug, 1993; Kumar et al., 2011; Mao et al., 2006; Pierucci et al., 2005; Tsai et al., 2008a) COMPUESTO CARÁCTER UMBRAL (µg m-3) Bajo Alto Compuestos de Azufre Sulfuro de hidrógeno Huevos podridos 0,7 14 Oxisulfuro de carbono Picante Disulfuro de carbono Desagradable 24 23000 Sulfuro de dimetilo Col podrida 2,5 50,8 Disulfuro de dimetilo Sulfuro 0,1 34,6 Trisulfuro de dimetilo Sulfuro 6,2 6,2 Metanotiol Sulfuro picante 0,04 82 Etanotiol Sulfuro, a tierra 0,032 92 Compuestos de Nitrógeno Amoniaco Picante, penetrante 26,6 39600
INTRODUCCIÓN 24 Aminometano Pescado picante 25,2 12000 Dimetil amina Pescado, amina 84,6 84 Trimetilamina Pescado, picante 0.8 1 3.metinildol (escatol) Heces 4 10-5 268 Ácidos Grasos volátiles Metanoico Penetrante, picante 45 37800 Etanoico Vinagre 2500 250000 Propanoico Rancio picante 84 60000 Butanoico Rancio 1 9000 Pentanoico Desagradable 2,6 2,6 3-metil butanocio Rancio 52,8 52,8 Cetonas Propanona Frangante, mentol 47500 1,6 106 Butanona Fragante, acetona 737 147000 2-Pentanona Fragante 28000 45000 Otros Benzotiazol Penetrante 442 2210 Etanol Frangante, hierba 0,2 4400 Fenol Medicinal 178 2240 Umbral bajo: Menor límite de detección para las personas más sensibles Umbral alto: Límite de detección para la mayor parte de las personas Los Compuestos Orgánicos Volátiles (COVs) generados en las plantas de compostaje son la principal causa de los malos olores generados en las mismas. Estos malos olores, producen diversas molestias en los alrededores de las zonas de tratamiento y pueden tener efectos adversos en la salud de los
INTRODUCCIÓN 25 trabajadores de las plantas y en los vecinos (Baldwin et al., 1999). Estos COVs provienen fundamentalmente de la degradación microbiana de los residuos y restos de plantas, las cuales son las principales materias primas del proceso. (Müller et al., 2004). Los Compuestos Orgánicos Volátiles se definen como cualquier compuesto que contiene carbón en su estructura molecular, que tiene una presión de vapor superior a 0.1 milímetros de mercurio (mm Hg) a condiciones normales (20ºC y 760 mm Hg) y que participa en las reacciones fotoquímicas atmosféricas, a excepción del monóxido de carbono, dióxido de carbono, ácido carbónico, carburos metálicos o carbonatos y carbonato amónico. Incluyen una variedad de compuestos químicos que pueden causar efectos a corto y largo plazo en la salud de las personas y son los principales causantes del smog. Por otra parte, los gases inorgánicos son llamados compuestos inorgánicos volátiles (CIVs) e incluyen gases como sulfuro de hidrógeno, dióxido de sulfuro, sulfuro de carbono y amoniaco (Revah & Morgan-Sagastume, 2005). Estos últimos también tienen una influencia importante en la producción de malos olores en las plantas de compostaje. Diversos estudios han identificado al limoneno, α-pineno y otros terpenos como los COVs más importantes en la generación de olores en plantas de compostaje (Bidlingmaier & Müsken, 2007; Breza-Boruta et al., 2010; Louhelainen et al., 2001). Por otra parte, se ha comprobado que el escaso control de la aireación en los procesos de compostaje favorece la aparición de zonas anaeróbicas y la formación de compuestos de azufre como: ácido
INTRODUCCIÓN 26 sulfhídrico, metanotiol, sulfuro de dimetilo y disulfuro de dimetilo, entre otros (Pierucci et al., 2005; Staley et al., 2006; Tsai et al., 2008b). El riesgo de formación de estos compuestos es mayor en las primeras fases del proceso y durante el almacenamiento de los residuos orgánicos debido a las condiciones de humedad de los materiales y a que porciones importantes de sustancias orgánicas fácilmente degradables son descompuestas (Bidlingmaier & Müsken, 2007; Tsai et al., 2008b). Con respecto al grupo de las cetonas, la 2-butanona puede ser el compuesto más importante perteneciente a este grupo funcional y emitido en los procesos de compostaje y puede ser considerado uno de los compuestos que influye más significativamente en la producción de los olores (Delgado-Rodríguez et al., 2011). 1.3.1.1. Efectos de los olores en la salud Los trabajadores de las plantas de compostaje están inevitablemente expuestos a los COVs que generan el olor en dichas instalaciones. Estos compuestos pueden producir trastornos secundarios como náuseas y vómitos, reacciones de hipersensibilidad e incluso alteraciones de tipo respiratorio (Schiffman & Williams, 2005). Los efectos más significativos de los COVs son de tipo tóxico sistémico, entre los que cabe destacar trastornos renales, hematológicos, neurológicos, hepáticos e irritaciones mucosonasales (Mücke & Lemmen, 2012). Es importante resaltar que las consecuencias de la inhalación dependen directamente de la concentración de los COVs que entran al sistema olfativo
INTRODUCCIÓN 27 del afectado. Sin embargo, diversos estudios reflejan que en las fases iniciales del proceso, la concentración de los compuestos orgánicos volátiles detectados supera ampliamente la concentración umbral olfativa, en algunos casos en unas 20000 veces como es el caso del limoneno y en unas 3000 veces como es el caso del pineno (Tolvanen & Hänninen, 2007). 1.4. MÉTODOS DE CONTROL DE LOS OLORES Una de las dificultades más importantes en el caso de la gestión y mitigación del olor es que es un rasgo subjetivo y bastante variable. Cabe anotar que aunque en la actualidad se han alcanzado avances significativos en la cuantificación del olor y de los compuestos químicos que lo ocasionan, siempre existirá una amplia variación entre la percepción de la gente ante el mismo fenómeno oloroso (Stoffella & Kahn, 2005). En España, como se mencionó anteriormente, la legislación en el control de olores es relativamente reciente desarrollándose progresivamente por autonomías. En el caso específico de Andalucía, el Decreto 239 de 2011 de la Consejería de Medio Ambiente por el que se regula la calidad de medio ambiente atmosférico, fomenta el interés por parte de los centros de investigación de la comunidad autónoma en unir los esfuerzos en el control de los contaminantes volátiles que causan olor. Por esta razón, la monitorización de los olores, su tratamiento y evaluación de las técnicas más utilizadas hasta el momento son un tema de gran actualidad.
INTRODUCCIÓN 28 En el diseño y construcción de un sistema de control de los olores, se debe tener en cuenta que la construcción de un sistema de tratamiento de gases, cuya finalidad es solucionar un problema, no debe acarrear un nuevo impacto ambiental, por lo que los subproductos del sistema de mitigación deben ser estudiados con detenimiento. En la actualidad existen diversas técnicas para el control de los olores producidos por los COVs y CIVs en la industria. Para seleccionar un método adecuado, es importante considerar las propiedades físicas, termodinámicas y de reactividad de los contaminantes. Entre las propiedades más influyentes en la caracterización de los COVs y la posterior selección del método de control se tienen: La fase (gas o líquido), tipos de enlaces presentes en los compuestos (covalentes o polares), capacidad de ionización, presión de Vapor, solubilidad en agua y reactividad (Morgan-Sagastume & Noyola, 2006). Las tecnologías de control de compuestos orgánicos volátiles y CIVs se clasifican de acuerdo a la naturaleza del control (Física, Química y Biológica) (Figura 1.6.). Entre las tecnologías más empleadas se encuentran la dilución, condensación, adsorción, combustión, enmascaramiento, scrubbers y la biofiltración.
INTRODUCCIÓN 29 Figura 1.6. Tecnologías de control de olores clasificadas según su naturaleza (Revah & Morgan-Sagastume, 2005). Es claro que no todos los sistemas son aplicables en todo el intervalo de flujos y concentraciones. Como norma general, la tendencia en instalaciones de tratamiento de efluentes industriales es utilizar instalaciones compactas, de bajo coste y que sean modulares, con lo que proporcionan un fácil manejo y seguimiento de la instalación. De esta forma en la elección de un sistema de tratamiento de gases odoríferos se deben tener en cuenta los siguientes factores: • Caudal y tipo de caudal.
INTRODUCCIÓN 30 • Características físico químicas de los gases (composición, partículas, etc,) • Concentración de compuestos a tratar. • Espacio necesario para la planta de depuración • Rendimiento requerido del tratamiento • Costes de mantenimiento • Inversión inicial para obra. En la Figura 1.8. se presenta un esquema de las diversas tecnologías utilizadas en el tratamiento de los compuestos orgánicos volátiles clasificadas según las características de caudal y concentración del efluente gaseoso de interés. Figura 1.8. Tecnologías para el control del olor clasificadas según concentración y caudal de la corriente gaseosa a tratar (Cardenas Gonzalez et al., 2003).
INTRODUCCIÓN 31 De la Figura 1.8. se puede concluir que el mecanismo de biofiltración es entonces una de las técnicas de mitigación de olores que representa gran utilidad para el tratamiento de compuestos orgánicos e inorgánicos volátiles a concentraciones y flujos moderados (Dorado et al., 2010; López et al., 2011; Pagans et al., 2005). Dentro de las principales ventajas de los métodos biológicos se tiene que operan a temperaturas normales (10-40 ºC) y presiones atmosféricas, son menos costosos, simples de operar y ecológicamente limpios comparados con los tratamientos fisicoquímicos. Además, permiten tener eficiencias considerables con concentraciones medias de contaminante y flujos entre 150 y 10000 m3/h (Revah & Morgan-Sagastume, 2005). 1.5. BIOFILTRACIÓN Las reacciones microbianas han sido utilizadas durante todo el siglo veinte en el tratamiento de aguas residuales y diversas clases de residuos sólidos, pero es desde 1950 cuando esta técnica empezó a ser empleada en el tratamiento de gases contaminados (Pomeroy, 1957). Durante los inicios de esta tecnología se empleaban normalmente sistemas abiertos en los cuales se utilizaba suelo poroso como material de soporte. Después de la publicación de los fundamentos de la tecnología de biofiltración, la misma se difunde principalmente en los Estados Unidos de América y la Alemania Occidental. Posteriormente, todos los esfuerzos por parte de la comunidad científica se centraron en conseguir mayores eficiencias con mayores cargas de compuestos volátiles: Es entonces en 1970 cuando en Alemania y en los
INTRODUCCIÓN 38 los compuestos aromáticos. Los tiempos de residencia recomendados están en el rango de 30 s a 1 min (Stoffella & Kahn, 2005). • El volumen de los poros debe ser mayor del 80%. Evitando de esta forma posibles zonas anaeróbicas y compactación del lecho que pueden derivar en obstrucciones del biofiltro (Yang et al., 2009). • La caída de presión a través del lecho del filtro depende de la porosidad, contenido de humedad y compactación del medio. En un biofiltro, la síntesis de biomasa hace que durante el tiempo de operación del mismo se acumule masa microbiana sobre el material filtrante, lo que ocasiona un incremento de la resistencia al flujo que se traduce en una caída de presión (Datta & Grant-Allen, 2005). • La altura típica del volumen activo de filtración debe oscilar en un intervalo entre 0,5 a 2,0 m, siendo 1 m de profundidad el típico de un biofiltro. Esta área es suficiente para garantizar un tiempo de residencia correcto y minimizar los requerimientos de espacio (Datta & Grant-Allen, 2005; Detchanamurthy & Gostomski, 2012). • El contenido de nutrientes de los materiales utilizados como lecho en el biofiltro son un parámetro importante a tener en cuenta en el diseño y monitorización de un biofiltro ya que son los responsables del desarrollo de la población microbiana en el mismo. Las fuentes de carbono y nitrógeno para los microorganismos deben provenir del gas contaminante, mientras que otros nutrientes como fósforo, nitrógeno,
INTRODUCCIÓN 39 minerales y elementos traza deben estar disponibles en el material. Concentraciones muy altas de nitrógeno pueden un efecto inhibitorio en la tasa de remoción del contaminante (Lauderdale et al., 2012) La puesta en marcha de un biofiltro requiere un tiempo de aclimatación de los microorganismos (Van Groenestijn & Liu, 2002). Los microorganismos generalmente presentes incluyen a hongos, bacterias y actinomicetos. Para las sustancias fácilmente degradables, este período de aclimatación es normalmente de alrededor de 10 días (Bagherpour et al., 2005). Este proceso también permite que los microorganismos desarrollen tolerancias o aceptación a compuestos que pueden ser normalmente tóxicos y que se convierten en fuente de carbono en el nuevo sistema Las principales ventajas de los biofiltros son: Costos de instalación y de operación son bajos, larga vida útil, condiciones ambientalmente adecuadas, alta capacidad de degradación (98 a 99% para los compuestos típicos del compostaje y para COV en general, 65 a 99%). Dentro de las principales limitaciones que tiene el proceso de biofiltración se encuentra su operación en abierto, la cual está limitada a factores ambientales. En la Tabla 1.3 se enuncian las principales ventajas y desventajas del proceso de biofiltración (Mudliar et al., 2010).
INTRODUCCIÓN 40 Tabla 1.3. Ventajas y Desventajas del proceso de biofiltración. VENTAJAS DESVENTAJAS Costos energéticos de operación bajos Dificultad para controlar crecimiento de biomasa microbiana Excepto de sustancias peligrosas en su operación Eficiencia limitada para concentraciones altas de contaminantes El contaminante es destruido Información limitada de diseño y variables de operación El CO2 producido asociado a esta tecnología es mucho menor Inestabilidad debido a que es un medio vivo Diseño simple sin recirculación de agua Canalización de flujo Un esquema general de un Biofiltro se observa en la figura 1.11 (Revah & Morgan-Sagastume, 2005).
INTRODUCCIÓN 41 Figura 1.11. Diagrama Esquemático de un Biofiltro Abierto 1.6. MÉTODOS DE CARACTERIZACIÓN DE COMPUESTOS ORGÁNICOS VOLÁTILES (COVs) y OLORES Teniendo en cuenta que el olor puede ser definido como un estímulo de las células olfativas ante la presencia de compuestos específicos incluyendo COVs y CIVs, existen diversos métodos de medida de los compuestos orgánicos e inorgánicos que causan olor. Tradicionalmente los métodos empleados en la monitorización y caracterización de los procesos que involucran generación y mitigación de COVs y olores se dividen en olfatometricos y procesos físicoquímicos (Bidlingmaier & Müsken, 2007).
INTRODUCCIÓN 42 1.6.1. CROMATOGRAFÍA DE GASES. El método más fiable para la medición cuantitativa y cualitativa de dichos compuestos (COVs) es la cromatografía de gases (Muñoz et al., 2010) aunque la presencia de un amplio número de compuestos requiere la combinación de diversas técnicas. Sin embargo, debido a la naturaleza compleja de las muestras, esta técnica presenta diversos problemas tales como (Kim et al., 2006): • Utilización de técnicas auxiliares que involucren sensibilidades diferentes o ejecutar diluciones previas al análisis, debido a las variaciones de concentración de los distintos componentes individuales. • El carácter inestable de los compuestos presentes en las muestras gaseosas hace necesaria en muchas ocasiones un muestreo “in situ” y “on line”, con lo que la cromatografía de gases presentaría serios inconvenientes en este caso. • La posible condensación de la humedad de las muestras gaseosas puede generar errores en la cuantificación final de los compuestos analizados, por lo que se hace necesario una minimización del tiempo entre muestreo y análisis. • La complejidad instrumental de la cromatografía de gases hace necesario realizar un muestreo utilizando bolsas de materiales normalizados como el tedlar o utilizando fibras adsorbentes.
INTRODUCCIÓN 43 Por estas razones en el caso de un sistema de biofiltración piloto o una planta de tratamiento de residuos, la monitorización de los compuestos orgánicos e inorgánicos causantes del olor requiere el empleo de técnicas más prácticas aunque pudieran ser de menor selectividad en lo que a la caracterización de compuestos se refiere. Varias técnicas utilizadas en la monitorización de los olores en la actualidad, vienen siendo empleadas hace varios años en otras áreas. Este es el caso de la Cromatografía de Gases la cual se ha utilizado ampliamente en sistemas en los que es necesario separar mezclas orgánicas complejas, compuestos organometálicos y sistemas bioquímicos y como método para determinar cuantitativamente y cualitativamente los componentes de una muestra. En las industrias la cromatografía de gases se enfoca a evaluar la pureza de los reactantes y a monitorear la secuencia de una reacción determinada (Bidlingmaier & Müsken, 2007). En el caso de la medida de los compuestos volátiles responsables del olor, la cromatografía de gases se ha empleado en conjunto con diferentes variaciones instrumentales. La técnica individual más efectiva empleada en este tipo de problemas en la industria o a nivel de planta piloto es la GC/MS (cromatografía de gases/espectrometría de masas) ya que permite la identificación y cuantificación de los compuestos causantes del olor y sus productos de transformación con gran sensibilidad, principalmente en el caso de análisis “offline” (Davoli et al., 2003; Defoer et al., 2002; López et al., 2011; Muñoz et al., 2010). Se han realizado varios intentos por mezclar la cromatografía de gases
INTRODUCCIÓN 44 con técnicas útiles en la evaluación de los olores pero de menor exactitud, estos esfuerzos incluyen la integración de la GC/MS con el análisis olfatométrico permitiendo la determinación de las características del olor de los compuestos separados cromatográficamente (Chen et al., 2008; Sohn et al., 2010). 1.6.2. NARIZ ELECTRÓNICA Los últimos avances en la tecnología de los olores y el interés por simplificar la monitorización de los mismos, reduciendo el análisis, identificación y cuantificación “off-line”, evitando de esta forma posibles errores durante el muestro como la adsorción y la degradación fotoquímica de los contaminantes presentes en las muestras gaseosas, han llevado a la aparición de las narices electrónicas, las cuales debido a su pequeño tamaño y portabilidad permiten realizar fácilmente medidas “in-situ”. La nariz electrónica se basa en un mecanismo de detección química y de un sistema de procesado capaces de identificar el perfil o huella de olor imitando el sentido del olfato (Sohn et al., 2008). Una nariz electrónica consta esencialmente de los siguientes componentes (WinMuster_1.6.2.14, 2010): • Dispositivo de muestreo: Para introducir la muestra se usan diversos dispositivos que bien calientan la muestra (en el caso de componentes con baja presión de vapor) para aumentar la concentración en el gas oloroso, o bien pre-concentran la muestra sobre diferentes soportes absorbentes
INTRODUCCIÓN 45 (Tenax) para después desorberlos térmicamente. De esta forma se pueden conseguir mejoras en el límite de detección de 10 a 1000 veces. El más utilizado es SHS (static headspace) aunque también pueden utilizarse otras técnicas: P&T (purge and trap), DHS (dynamic headspace), SPME (solidphase microextraction), SBSE (stir bar sorptive extraction), INDEX (inside needle dynamic extraction) o MIMS (membrane introduction mass spectrometry). La introducción directa de la muestra es recomendable para la monitorización de procesos en tiempo real. Algunos equipos incorporan también dispositivos de dilución automática del gas oloroso para evitar la saturación por sobrecarga de los detectores y aumentar su vida útil. En la Fig. 1.12 se muestra el esquema de una nariz con introducción directa de la muestra y dilución automática de la misma. Figura 1.12. Esquema de un sistema de monitorización con nariz electrónica • Sistema de detección: El sistema de detección típico de una nariz electrónica es el de un conjunto de sensores de gas (de 5 a más de 20) con especificidad parcial. La selectividad de los diferentes sensores viene
INTRODUCCIÓN 46 determinada por el tipo de material del sensor, el material dopante, la temperatura de trabajo y su geometría. Aunque existen diferentes tipos de sensores los más adecuados para muestras ambientales son los sensores de tipo transistor efecto de campo de metal-óxido (MOSFET) y los de tipo semiconductor metal-óxido (MOS). Como ejemplo, la nariz electrónica PEN3 de Airsense (Figura 1.12) incorpora 10 sensores metal-óxido situados en una pequeña cámara de medida de 1,8 ml. Las últimas tecnologías emplean espectrómetros de masas o espectrometría de movilidad de iones en combinación con el análisis de datos PCA aplicado a los principales picos cromatográficos. • Análisis de datos: La respuesta de una nariz electrónica consiste en el conjunto de medidas de los diferentes sensores. Este conjunto habitualmente se presenta como una “huella de olor”. Este conjunto de datos es analizado para reconocer el olor por procedimientos que habitualmente hacen uso del Análisis de Componentes Principales (PCA) (PCA) o bien mediante Redes Neurales. Normalmente este análisis es realizado por un software residente en un ordenador independiente del sistema de detección. El concepto de nariz electrónica fue introducido originalmente en 1982 en Reino Unido en un intento por realizar una analogía con el olfato humano, reproduciendo la alta sensibilidad del mismo y la alta discriminación entre los
INTRODUCCIÓN 47 compuestos que causan olor. Posteriormente durante el año 1985 en Japón se hicieron valiosos esfuerzos por difundir los resultados experimentales obtenidos en la identificación y cuantificación de olores utilizando un sistema electrónico compuesto por sensores integrados y un microcomputador (Ikegami & Kaneyasu, 1985). Alrededor de 1990 estos prototipos fueron fabricados para su comercialización (Ameer & Adeloju, 2005). En la actualidad existen diversos estudios relacionados con esta clase de dispositivos en el ámbito de la industria alimentaria, más exactamente en el control de calidad de los productos finales y principalmente en la caracterización de los vinos, café y té (Baldwin et al., 2011; Banerjee et al., 2012; Brudzewski et al., 2012; Ghosh et al., 2012; Prieto et al., 2012). Este desarrollo se debe fundamentalmente a sus ventajas prácticas frente a las técnicas cromatográficas de análisis. Sin embargo, deben sumarse esfuerzos para centrar los trabajos futuros en intentar resolver sus principales desventajas como lo son la incapacidad de determinar compuestos químicos específicos en una muestra gaseosa y su cuantificación en términos de unidades de olor. Por otra parte, existen relativamente pocos trabajos relacionados con el uso de la nariz electrónica en la monitorización de procesos generadores de olores como lo son las plantas de tratamiento de residuos sólidos, plantas de compostaje y de tratamiento de aguas residuales, criaderos de animales, estaciones de gasificación (Delgado-Rodríguez et al., 2012; Nicolas et al., 2006; Romain et al., 2005; Sironi et al., 2007; Sohn et al., 2008; Sohn et al., 2009). A su vez, el uso de esta tecnología como herramienta de evaluación de los parámetros de eficiencia en la biofiltración tiene gran potencial, pero es necesario fomentar su uso en base a resultados coherentes
INTRODUCCIÓN 54 nursery growing media: Effects on growth and nutrition of the native shrub Pistacia lentiscus L. Bioresource Technology, 99(6), 1793-1800. Ottengraf, S.P.P., Meesters, J.J.P., Oever, A.H.C., Rozema, H.R. 1986. Biological elimination of volatile xenobiotic compounds in biofilters. Bioprocess and Biosystems Engineering, 1(2), 61-69. Pagans, E., Font, X., Sánchez, A. 2007. Adsorption, absorption, and biological degradation of ammonia in different biofilter organic media. Biotechnology and Bioengineering, 97(3), 515-525. Pagans, E.L., Font, X., Sánchez, A. 2005. Biofiltration for ammonia removal from composting exhaust gases. Chemical Engineering Journal, 113(2- 3), 105-110. Petric, I., Helić, A., Avdić, E.A. 2012. Evolution of process parameters and determination of kinetics for co-composting of organic fraction of municipal solid waste with poultry manure. Bioresource Technology, 117, 107-116. Pierucci, P., Porazzi, E., Martinez, M.P., Adani, F., Carati, C., Rubino, F.M., Colombi, A., Calcaterra, E., Benfenati, E. 2005. Volatile organic compounds produced during the aerobic biological processing of municipal solid waste in a pilot plant. Chemosphere, 59(3), 423-430. Pomeroy. 1957. De-odorizing of gas streams by the use of microbial growth, Vol. 2,793,096. United States. Prieto, N., Rodriguez-Méndez, M.L., Leardi, R., Oliveri, P., Hernando- Esquisabel, D., Iñiguez-Crespo, M., de Saja, J.A. 2012. Application of multi-way analysis to UV-visible spectroscopy, gas chromatography and electronic nose data for wine ageing evaluation. Analytica Chimica Acta, 719, 43-51. Ragazzi, M., Rada, E.C., Antolini, D. 2011. Material and energy recovery in integrated waste management systems: An innovative approach for the characterization of the gaseous emissions from residual MSW bio-drying. Waste Management, 31(9-10), 2085-2091. Rene, E.R., Špačková, R., Veiga, M.C., Kennes, C. 2010. Biofiltration of mixtures of gas-phase styrene and acetone with the fungus Sporothrix variecibatus. Journal of Hazardous Materials, 184(1-3), 204-214. Revah, S., Morgan-Sagastume, J.M. 2005. Methods of Odor and VOC Control. in: Biotechnology for Odor and Air Pollution Control, (Eds.) Z. Shareefdeen, A. Singh, Springer-Verlag Berlin Heidelberg Germany. Rigamonti, L., Grosso, M., Giugliano, M. 2010. Life cycle assessment of subunits composing a MSW management system. Journal of Cleaner Production, 18(16-17), 1652-1662. Romain, A.C., Godefroid, D., Kuske, M., Nicolas, J. 2005. Monitoring the exhaust air of a compost pile as a process variable with an e-nose. Sensors and Actuators, B: Chemical, 106(1 SPEC. ISS.), 29-35. Roshani, B., Torkian, A., Aslani, H., Dehghanzadeh, R. 2012. Bed mixing and leachate recycling strategies to overcome pressure drop buildup in the biofiltration of hydrogen sulfide. Bioresource Technology, 109, 26-30. Saha, S., Pradhan, K., Sharma, S., Alappat, B.J. 2008. Compost production from Municipal Solid Waste (MSW) employing bioinoculants.
INTRODUCCIÓN 55 International Journal of Environment and Waste Management, 2(6), 572- 583. Schiffman, S.S., Williams, C.M. 2005. Science of odor as a potential health issue. Journal of Environmental Quality, 34(1), 129-138. Sironi, S., Capelli, L., Céntola, P., Del Rosso, R., Il Grande, M. 2007. Continuous monitoring of odours from a composting plant using electronic noses. Waste Management, 27(3), 389-397. Sohn, J.H., Hudson, N., Gallagher, E., Dunlop, M., Zeller, L., Atzeni, M. 2008. Implementation of an electronic nose for continuous odour monitoring in a poultry shed. Sensors and Actuators, B: Chemical, 133(1), 60-69. Sohn, J.H., Parcsi, G., Dunlop, M.W., Gallagher, E.M., Stuetz, R.M. 2010. Air quality monitoring and characterisation at two commercial broiler farms. pp. 141-146. Sohn, J.H., Pioggia, G., Craig, I.P., Stuetz, R.M., Atzeni, M.G. 2009. Identifying major contributing sources to odour annoyance using a non-specific gas sensor array. Biosystems Engineering, 102(3), 305-312. Soliva, M. 2001. Compostatge i gestió de residuos orgànics., (Ed.) D.d. Barcelona. Barcelona. Staley, B.F., Xu, F., Cowie, S.J., Barlaz, M.A., Hater, G.R. 2006. Release of trace organic compounds during the decomposition of municipal solid waste components. Environmental Science and Technology, 40(19), 5984-5991. Stoffella, P.J., Kahn, B.A. 2005. Utilización de Compost en los Sistemas de Cultivo Hortícola. Ediciones Mundi-Prensa, Spain. Streese, J., Schlegelmilch, M., Heining, K., Stegmann, R. 2005. A macrokinetic model for dimensioning of biofilters for VOC and odour treatment. Waste Management, 25(9), 965-974. Thitame, S.N., Pondhe, G.M., Meshram, D.C. 2010. Characterisation and composition of Municipal Solid Waste (MSW) generated in Sangamner City, District Ahmednagar, Maharashtra, India. Environmental Monitoring and Assessment, 170(1-4), 1-5. Tolvanen, O.K., Hänninen, K.I. 2007. Occupational hygiene in two combineddrum-and-tunnel composting plants managing source separated biowaste and sludge. Journal of Environmental Health, 69(9), 64-71. Tsai, C.-J., Chen, M.-L., Ye, A.-D., Chou, M.-S., Shen, S.-H., Mao, I.F. 2008a. The relationship of odor concentration and the critical components emitted from food waste composting plants. Atmospheric Environment, 42(35), 8246-8251. Tsai, C.J., Chen, M.L., Ye, A.D., Chou, M.S., Shen, S.H., Mao, I.F. 2008b. The relationship of odor concentration and the critical components emitted from food waste composting plants. Atmospheric Environment, 42(35), 8246-8251. Tyson, R.V., Simonne, E.H., Treadwell, D.D., White, J.M., Simonne, A. 2008. Reconciling pH for ammonia biofiltration and cucumber yield in a recirculating aquaponic system with perlite biofilters. HortScience, 43(3), 719-724.
INTRODUCCIÓN 56 USEPA. 2007. Method 3815: Screening solid samples for volatile organics, in EPA publication SW-846, Test Methods for Evaluating Solid Waste, Physical/Chemical Methods. Van Groenestijn, J.W., Liu, J.X. 2002. Removal of alpha-pinene from gases using biofilters containing fungi. Atmospheric Environment, 36(35), 5501- 5508. Vasarevičius, S., Baltrenas, P., Baltrenaite, E. 2011. Investigation and evaluation of green waste composting parameters. Polish Journal of Environmental Studies, 20(6), 1603-1609. Vergara-Fernández, A., Salgado-Ísmodes, V., Pino, M., Hernández, S., Revah, S. 2012. Temperature and moisture effect on spore emission in the fungal biofiltration of hydrophobic VOCs. Journal of Environmental Science and Health - Part A Toxic/Hazardous Substances and Environmental Engineering, 47(4), 605-613. Vergara-Fernández, A., Van-Haaren, B., Revah, S. 2005. Biofiltration of hydrophobic vocs with filamentous fungi: Modeling and experimentation. pp. 8929. Vigueras, G., Arriaga, S., Shirai, K., Morales, M., Revah, S. 2009. Hydrophobic response of the fungus Rhinocladiella similis in the biofiltration with volatile organic compounds with different polarity. Biotechnology Letters, 31(8), 1203-1209. Warman, P.R., Margarit, S.X. 2012. The effect of ruminant and MSW composts on asparagus growth and soil fertility, Vol. 950, pp. 73-80. WinMuster_1.6.2.14. 2010. Airsense Analytics GmbH. Hagenover, Schwerin, Germany. Yang, C., Chen, H., Zeng, G., Yu, G., Liu, X., Zhang, X. 2009. Modeling variations of medium porosity in rotating drum biofilter. Chemosphere, 74(2), 245-249. Yang, N., Zhang, H., Chen, M., Shao, L.M., He, P.J. 2012. Greenhouse gas emissions from MSW incineration in China: Impacts of waste characteristics and energy recovery. Waste Management. Yañez, R., Bueno, P., Rivera, A., García-Barneto, A., Díaz, M.J. 2010. Selective organic compounds degradation under controlling composting conditions. Waste Management, 30(5), 755-763.
OBJETIVOS Y PLAN DE TRABAJO 57 2. OBJETIVOS Y PLAN DE TRABAJO 2.1. OBJETIVOS Los principales objetivos planteados en este trabajo de investigación se enuncian a continuación: 2.1.1. OBJETIVO GENERAL • Evaluar el proceso de biofiltración empleando diferentes compost de RSU-Poda como lecho filtrante para el tratamiento de compuestos orgánicos volátiles (COVs) generados en el compostaje de RSU mediante diversas técnicas de monitorización. 2.1.2. OBJETIVOS ESPECÍFICOS • Estudiar la influencia de diferentes parámetros del co-compostaje de RSU-Restos de Poda (aireación, humedad, relación C/N y tiempo) en los parámetros de estabilidad del compost a través de ensayos controlados, con el fin de utilizar los productos finales más idóneos como material filtrante en sistemas de biofiltración. • Monitorear las emisiones gaseosas de compuestos volátiles generadas en las diversas fases de un compostaje semiindustrial de la fracción orgánica de residuos sólidos urbanos (FORSU) y restos de jardinería. • Evaluar el rendimiento de cuatro tipos de compost maduro de RSU-Poda utilizados como lecho en un sistema de biofiltración, con el fin de tratar
OBJETIVOS Y PLAN DE TRABAJO 58 COVs producidos durante el compostaje semiindustrial de FORSU y restos de jardinería. • Determinar el rendimiento de un sistema de biofiltración construido a escala de laboratorio utilizando dos tipos de compost como lecho para tratar corrientes gaseosas contaminadas con Metil Etil Cetona (MEK) y αpineno en experimentos individuales controlados. • Analizar la influencia del contenido de humedad de los materiales utilizados como lecho en un sistema de biofiltración a escala de laboratorio en la eficiencia de remoción de corrientes gaseosas contaminadas con COVs hidrofílicos (MEK) e hidrofóbicos (α-pineno) mediante experiencias separadas. • Estudiar el potencial del uso de la nariz electrónica, el detector de fotoionización de COVs y la Cromatografía de Gases en conjunto con Espectrometría de Masas como técnicas de monitorización de los procesos de biofiltración y de evaluación de emisiones gaseosas contaminantes en general. 2.2. PLAN DE TRABAJO Este trabajo de investigación hace parte del proyecto CTM2007- 62117 financiado por la Comisión Interministerial de Ciencia y Tecnología (CYCIT) cuyo objetivo principal es la optimización de los diferentes parámetros que rigen el proceso de compostaje, evaluando la influencia de los mismos sobre el proceso integrado, la producción de olores y compuestos orgánicos volátiles (COVs), con el fin de proponer alternativas para la mitigación de los efectos
OBJETIVOS Y PLAN DE TRABAJO 59 medioambientales, reducir significativamente los costes de inversión y operación para mejorar sustancialmente la eficacia del proceso a gran escala. Las diferentes actividades y trabajos realizados en esta tesis doctoral se enfocan fundamentalmente en la evaluación del rendimiento de la tecnología de biofiltración para tratar COVs cuyo origen principal es el proceso de compostaje de RSU. En la construcción del sistema se utilizarán como materiales de relleno compost maduros provenientes de ensayos controlados donde se determinará la influencia de los diversos parámetros operacionales del compostaje en los parámetros de estabilidad con el fin de obtener un compost con las características más idóneas para su propósito. A su vez, en la monitorización de los diferentes sistemas de biofiltración se utilizarán diversas tecnologías empleadas en el monitoreo de gases como lo es la nariz electrónica, la cromatografía de gases – espectrometría de masas y el detector de fotoionización de COVs. Todas las técnicas mencionadas se utilizarán paralelamente durante los experimentos con el fin de obtener información lo más selectiva posible del rendimiento operativo del sistema estudiado. Simultáneamente, se realizará una valoración de las diferentes técnicas utilizadas en la monitorización con el fin de establecer las ventajas y desventajas de cada una en este tipo de sistemas, con el fin de tener esto como criterio de selección en futuros trabajos. El presente plan de trabajo se divide en las siguientes etapas: 1. Estudio de la influencia de parámetros del co-compostaje de RSU-Poda en los parámetros de estabilidad del compost mediante ensayos controlados.
OBJETIVOS Y PLAN DE TRABAJO 60 Los Residuos Sólidos Urbanos a emplear en esta etapa se obtendrán en la planta de tratamiento de residuos urbanos de Villarrasa (Huelva). Por su parte, los residuos de poda, se recolectaran en el jardín botánico José Celestino Mutis (Palos de la Frontera, Huelva). A su vez, el experimento de co-compostaje se ejecutará en la planta experimental Jack Rodney Harlan de la Escuela Politécnica Superior “La Rábida” de la Universidad de Huelva. Los experimentos a mediana escala se desarrollarán utilizando reactores de polietileno de alta densidad, con una capacidad de 200 litros y termostatizados mediante espuma de poliuretano, para mejorar las condiciones del proceso. Se utilizará un diseño de composición central (Box–Behnken) con 3 factores y tres niveles para estudiar la influencia de las variables independientes (aireación, humedad, relación C/N y tiempo), normalizadas entre -1 y +1, sobre las variables dependientes (materia orgánica, Pérdidas de Nitrogeno, Demanda Química de Oxígeno, contenido de nitratos y coeficiente de biodegradabilidad). La monitorización del proceso incluirá el seguimiento de la temperatura del mismo mediante termopares de sensores en continuo e independientes asociados a un equipo de adquisición de datos. La homogeneización de los materiales a recibir tratamiento aeróbico se garantizará mediante volteos periódicos de los reactores. De la misma forma el flujo de aire en cada uno de los compostadores se mantendrá mediante un sistema de aireación. Dicho sistema consistirá en un anillo perforado de PVC en la base del reactor unido a un compresor de aire. Para la evaluación del proceso de compostaje se realizarán diferentes análisis fisico-químicos mediante el empleo de métodos específicos según normas
OBJETIVOS Y PLAN DE TRABAJO 61 UNE. Estos análisis se efectuarán tanto a las materias primas como a los diferentes muestreos durante la ejecución del experimento. Entre los análisis más importantes realizados se encuentran: Secado y determinación de la humedad de muestras, separación y determinación de impurezas en muestras secas el cual incluye la determinación de la granulometría de las mismas, Molienda de muestras, determinación del contenido en cenizas (Materia Orgánica), Determinación de N - Kjeldahl, composición química (macro y microelementos) mediante plasma (ICPOES),DQO, pH y conductividad de la fracción orgánica soluble de enmiendas. La evolución del conjunto de estos parámetros informará sobre el desarrollo y eficacia del proceso y particularmente sobre la biodegradabilidad del material de partida y la estabilidad del producto final bajo la influencia de las distintas condiciones de operación. Esta evolución se realizará mediante un análisis minucioso de modelos polinomiales empíricos de segundo orden para cada uno de los parámetros de estabilidad obtenidos en base al diseño experimental propuesto. 2. Evaluación del rendimiento de un sistema de biofiltración basado en compost maduros de RSU-Poda para tratar COVs provenientes de una planta piloto semiindustrial de compostaje FORSU y restos de jardinería. Se realizará un ensayo de compostaje a escala piloto en un compostador rotatorio (BIOCOMP 3, Kollvik Recycling S.L., San Sebastián, España) ubicado en el campus de “La Rábida” de la Universidad de Huelva, España. Las materias primas empleadas en el proceso de compostaje ejecutado en esta planta piloto serán restos orgánicos de comida y cocina (FORSU) procedentes
OBJETIVOS Y PLAN DE TRABAJO 62 del comedor universitario de la Universidad de Huelva, separados selectivamente y restos astillados (< 10 cm) de poda de jardinería urbana de Sevilla. Se llevará un control de la cantidad de material adicionado y una caracterización fisicoquímica en diferentes puntos del proceso. A su vez, se asociará al sistema una batería de 12 biofiltros que recibirá el flujo en paralelo de aire en dos series provenientes del compostador piloto. Los biofiltros se rellenarán con compost maduro de RSU y RSU:Poda (1:1) proveniente de los ensayos controlados de la etapa anterior. Se obtendrán 4 tipos diferentes de materiales filtrantes con 3 triplicados de cada uno, basados en el tamaño de partícula de los materiales (2 a 7 mm y 7 a 20 mm). El diámetro externo de cada biofiltro se fijará en 11 cm y 1 m de altura, a su vez, cada uno presentará un reservorio de agua en la parte inferior de la entrada del flujo (flujo ascendente) con el fin de evitar la pérdida de humedad del gas a tratar. La EBRT se fijará en 44 s. La monitorización del tratamiento biológico de los gases provenientes del proceso de compostaje se realizará con una nariz electrónica comercial (PEN3, Airsense), un detector de Compuestos orgánicos volátiles con PID (ppbRAE3000 y MultiRAE, RAE Systems, San José, CA, EEUU) y GC/MS (HP 6890 -Agilent Technologies). Utilizando las técnicas mencionadas, se evaluará el rendimiento de cada tipo de biofiltro observando si existen diferencias entre los mismos en función de la eficiencia de remoción, problemas de operación y señal emitida. Esta evaluación se llevará a cabo con la ayuda de técnicas estadísticas de análisis multivariante como el ACP (análisis de componentes principales).
OBJETIVOS Y PLAN DE TRABAJO 63 3. Evaluación del rendimiento de un sistema de biofiltración en laboratorio para el tratamiento de gases con diferentes concentraciones de contaminantes orgánicos volátiles. Se construirá un sistema de biofiltración compuesto de un par de biofiltros, utilizando dos tipos diferentes de material filtrante. Los materiales a emplear en esta etapa serán los que presenten menores problemas operacionales y tengan las máximas eficiencias de remoción en la experiencia piloto. Las dimensiones de cada biofiltros serán exactamente iguales a las utilizadas en la anterior etapa. El aire contaminado a tratar se preparará artificialmente utilizando un burbujeador que contendrá el contaminante en cuestión y permitía ajustar la concentración de entrada al sistema. A su vez, el área de pre-tratamiento del gas a tratar, contendrá un humidificador para garantizar las condiciones de humedad relativa de la corriente gaseosa recomendada en previos trabajos. Los contaminantes utilizados serán los COVs con presencia mayoritaria durante el compostaje a escala piloto y su naturaleza estará relacionada con los dos tipos de biofiltros escogidos. Las concentraciones de los contaminantes se aumentarán paulatinamente con el fin de encontrar la capacidad máxima de remoción de los biofiltros propuestos y evaluar su capacidad de adaptación a cargas intermitentes con diferentes concentraciones de contaminante en los gases de entrada al sistema. La EBRT se fijará en torno a 66 s y el contenido de humedad del lecho se escogerá en base a los resultados obtenidos en la anterior etapa.
CAPÍTULO 3 70 The materials were mixed carefully due to their heterogeneity. Each mixture obtained with the characteristics of the experimental design, was transferred to the composting reactor. Each one was half filled (to guarantee air flow and aerobic conditions) with 40 kg of the mixture of MSW: LTR. 3.2.2. EXPERIMENTAL DESIGN FOR THE COMPOSTING PROCESS OPTIMIZATION. In order to have different conditions during the composting, a three level, threefactor factorial experimental design was used (Box-Behnken design) (Akhnazarova and Kafarov, 1982). This design allows a designer to adequately quantify a response with a reasonable number of tests. In this sense, Box- Behnken designs require three levels for each factor thus allowing us to evaluate second order models. In this case, the replicated points (with except to center point) can be eliminated from these designs due the high balance and symmetry. For this reason, central composite designs are larger than Box- Behnken designs and provide more degrees of freedom for error estimation. In order to be able to relate the dependent and independent variables to the minimum possible number of experiments, a orthogonal main effect design consisted of a central one point (central experiment, in the centre of a cube, duplicated) and 16 additional points (additional experiments lying at the cube vertices), was used. All of them were evaluated at three levels, low (denoted as -1) centre point or medium (denoted as 0) and high level (denoted as +1). 3.2.3. STATISTICAL ANALYSIS The experimental design described above enabled the construction of secondorder polynomials in the independent variables and the identification of
CAPÍTULO 3 71 statistical significance in the variables (Akhnazarova and Kafarov, 1982). The polynomial model used was of the following type: j)(i 1;11 0 <++= ∑∑ === njni n ji ijni n ii XXdXbaZ (Equation 3.1.) Where Z and Xni denote dependent and normalized independent variables, respectively, a0 is a constant and bi, ci, dij are the regression coefficients obtained from experimental data. Independent variables were normalized (Xn) by using the following equation: ( ) ( ) ( ) 2 minmax XX XX X mean n − − = (Equation 3.2.) Where X is the absolute value of the independent variable concerned, Xmean is the average value of the variable and Xmax and Xmin are their maximum and minimum values, respectively. The independent variables used in the equations relating to both types of variables were those having a statistically significant coefficient (viz. those not exceeding a significance level of 0.05 in Student's t-test and having a 95% confidence interval excluding zero). To obtain such models, the normalized values of independent variables and properties of the compost obtained using the proposed experimental design (Table 3.2) have been correlated. Each value of these properties is an average of three experimental results. The range of values for each independent variable (C/N, aeration and moisture) used in the proposed experimental design is shown in Table 2. At this point, the C/N ratio (C/N) used assumes that the entire N is biodegradable and it is really based on chemical contents. The properties of composting process selected as dependent variables for each model are Organic Matter (%, OM), Nitrate (mg
CAPÍTULO 3 72 kg-1, NI), Chemical Organic Demand (mgO2 kg-1, COD), Nitrogen-losses (%, NL) and Biodegradability coefficient (km). The properties mentioned above are reported as indicator parameters of compost stability by several authors (Bueno et al., 2009b; Diaz et al., 2002; Haug, 1993; Yañez et al., 2010). Table 3.2. Experimental domain and Box-Behnken design matrix used in the studya. Reactor MSW:LTRa (w:w)ratio C/N Aeration L air kg-1 min-1 Moisture % C/N, A, Mb 1 1:2 77 0.300 70 +1, +1, +1 2 1:0 21 0.300 70 -1, +1, +1 3 1:2 77 0.300 40 +1, +1, - 1 4 1:0 21 0.300 40 -1, +1, - 1 5 1:2 77 0.050 70 +1,- 1, +1 6 1:0 21 0.050 70 -1, -1, +1 7 1:2 77 0.050 40 +1, -1, - 1 8 1:0 21 0.050 40 -1, -1, - 1 9 1:1 60 0.300 55 0, +1, 0 10 1:1 60 0. 050 55 0, -1, 0 11 1:1 60 0.175 70 0, 0, +1 12 1:1 60 0.175 40 0, 0, - 1 13 1:2 77 0.175 55 +1, 0, 0 14 1:0 21 0.175 55 -1, 0, 0 15 1:1 60 0.175 55 0, 0, 0 16 1:1 60 0.175 55 0, 0, 0 aMSW: Municipal solid waste; LTR: Legume Trimming Residues b Normalized values for C/N (C/N), aeration (A) and moisture (M) respectively
CAPÍTULO 3 73 A pareto chart was constructed with the purpose to analyse the independent variables which have the greatest cumulative effect in the stability parameters of compost studied (Fig 3.1). This figure shows a plot of each dependent variable (stability parameters of compost) against each independent one constructed by changing all the independent variables (composting process parameters) between the normalized values from –1 to +1 in the models constructed. Having in mind that the influence of the different independent variables on the dependent variables can vary with each value of the first ones, the average change in the dependent variable will be given by: ( ) ( ) [ ] ( ) ( ) ( ) ( ) [ ] min ni max ni X X ni min ni max ni X X dX X ZX Z max ni min ni − − ∫ (Equation 3.3.) Therefore, the change in the dependent variable with that in the independent variable can be expressed as the difference between [Z(Xni)max]max – Z[(Xni)min]min and the previous expression: ( ) [ ] ( ) [ ] { } ( ) ( ) [ ] ( ) ( ) ( ) ( ) [ ] min ni max ni X X ni min ni max ni min min ni max max ni X X dX X ZX Z X ZX Z max ni min ni − − −−= ∫ i DZ (Equation 3.4.) The figure shows the relative DZi value for each variable. These values allow one to weight the relative influences, as percentages, of each independent variable on the variation of each dependent variable 3.2.4. COMPOSTING REACTORS The composting reactors were 200 L capacity acrylic barrels. 40 kg of the raw material were placed in the cylindrical reactor. This equated to 3/4 of the total volume of the reactor. To minimize the conductive heat loss via the reactor wall,
CAPÍTULO 3 74 they were insulated with polyurethane foam. Compressed air (at different rates according to the experimental design, 0.005-0.3 lairkg-1min-1) was introduced into the bottom of each reactor and evenly distributed to the composting mixture through a perforated plate. To maintain initial moisture content, water loss was compensated, every day, by the addition of water during active composting. Two temperature sensors (K thermocouples, TMC6-HA) were placed at the center and the top of the composting mass. An additional temperature sensor was placed outside the reactors to obtain the environmental temperature (Protimeter-MMS-Plus). Temperatures were recorded every 12 h in each reactor by two data loggers (HOBO, U12-006). The normalized time -1,-0.7,-0.3, 1 corresponds to days 0,10,24,36 of active composting respectively at which compost samples were taken. 3.2.5. ANALYTICAL METHODS The feedstocks used in the experiment (MSW and LTR), were obtained immediately before the start phase of the composting process. The MSW particle size distribution was determined by a sieve shaker (CISA MODRP.09 Sieve Shaker) and the impurities (glass, plastics, metals and stones) were hand separated and weighed. In the case of LTR, particle size determination was not possible due to its fibrous character. Compacted Bulk Density was determined according to the CEN EN 13040 standard (EN-13040, 1999). The total organic matter was determined by the weight loss after dry combustion at 540 ºC in a muffle for 4 hours (Heraeus D-6450 Hanau) and organic carbon was estimated multiplying the factor 0.58 by the organic matter results (Haug, 1993). Nitrogen was determined by steam distillation after Kjeldahl digestion using a Tecator Kjeltec System 1002 distilling unit (Bremnen, 1996). In the case of MSW,
CAPÍTULO 3 75 organic matter and nitrogen determinations were done in the < 5 mm size fraction. All the analyses were carried out in triplicate. Compost samples for the analysis mentioned above were taken at days 0, 10, 24 and 36 from the start of the experiment. Moisture content in the compost samples was quickly determined by an Infrared Moisture Analyzer (COBOS IB110) to correct moisture loss in the reactors. Initial moisture content was determined by oven drying at 105 ºC. The pH and the electric conductivity were determined in 1:5 (weight) compost: water extracts using a pHmeter (Crison BASIC 20+) and a conductivity cell (Crison MicroCM, 2201) according to TMECC method 04.10-A (Thompson et al., 2003). The Chemical Oxygen Demand (COD) was measured by a colorimetric method (Spectrophotometer HACH DR/2000) in 1:100 (weight) compost: water extracts, using commercial digestion vials with a range of 500-1500 ppm (HACH) after digestion for two hours to 150 ºC in a COD reactor (HACH, 45600) (Garcia et al., 1991; Zmora-Nahum et al., 2005). Macro and micro elements were obtained by inductively coupled plasma (ICP) spectrometry following nitric/hydrochloric acid digestion (Barnhisel, 1982). The content of nitrate in the different samples of compost was determined in 1:5 (weight) compost: water extracts using a Bran + Luebbe GmbH AA3 dual channel continuous flow auto analyser (Norderstedt, Germany). The Kjeldahl-N losses during the process (Equation 3.5.) have been calculated from the initial content of N-Kjeldahl, and the evolution of the organic matter, assuming that the mineral amount in each mixture is constant (Bernal et al., 1998; Bueno et al., 2008).
CAPÍTULO 3 76 12 21 100100 NX NX NLoss −= (Equation 3.5.) Where N1 and N2 are the initial and final N-Kjeldahl concentrations and X1 and X2 the initial and final ash content respectively. The degradability coefficient (km) for the samples (Equation 3.6.) was obtained, at the same time, from the definition of the conservation of ash principle, assuming that inerts entering the process should equal inerts leaving the process at steady state (Haug, 1993). In this way, km was calculated using Equation 2. ( ) %)100%( 100%% 21 21 OMOM OMOM k m − − = (Equation 3.6.) Where OM1% is the initial total organic matter content, % of total solids and OM2% is the final total organic matter content, % of total solids. 3.3. RESULTS AND DISCUSSION To gather information on the agronomic value of the raw material, analyses on the nutrient content were performed (Table 3.1). In this sense, results revealed adequate properties for both raw materials to be co-composted. The moderate salinity (EC) of MSW and the high C/N ratio of LTR could be counterbalanced among both residues. The low concentration of heavy metals (data not shown) confirms the safety in term of metals of the original materials. Substituting the values of the independent variables for each dependent variable in Table 3.2 into the polynomial expression used by the models, yielded the equations showed in Table 3.3. The differences between the experimental values and those that were estimated using the previous equations never exceeded 10% of the former. The
CAPÍTULO 3 77 equations obtained have a coefficient of determination (r2) from 0.81 to 0.95 which suggest that the model is reliable in order to represent the system studied and predict the optimum values of control parameters. Table 3.3. Equations yielded for each dependenta variable as a function of the independent variablesa. Compounds r2 b Fb dfb OM= 61.88293 - -1.57455 *T -4.26918 * M + 15.18732 *CN -4.96636 * M2 -4.91715 * CN2 -3.47587 * M CN 0.89 65.0 6.53 NI= 37.7710 + 86.0151 * T -49.6935 * CN + 57.5314 * T 2 - 72.1740 * T * CN 0.95 300.2 4.55 COD= 10507.5 -18159.2 *T -3933.1 * A -6745.5 * CN +25632.1 * T2 +8040.0* CN2 +3057.7* T * A +6628.1* T *CN +3921.9 A * CN 0.91 65.6 8.51 NL= -2.48229 -2.68113 *T + 5.00057 * M -4.78448 * T2 +4.64456 M 2 +4.26726 CN 2 +7.57182 A M 0.81 37.9 6.53 KM= 0.635431 +0.213008 *T +0.043159 *M -0.213974 *CN -0.400391 *T2 -0.130694 * T *CN +0.041150 *M * CN 0.90 77.0 6.53 a r2, F and df denote coefficient of determination, Fisher-Snedecor distribution and degrees of freedom respectively. The differences between the experimental values and those estimated by using the previous equations never exceeded 10-15% of the former. Where: T,A, M, CN denote the normalized value of the operation time, aeration, moisture and C/N ratio, respectively as independent variables. In addition, OM= Organic Matter (%). NI= Nitrate (mg kg-1), COD= Chemical Organic Demand (mgO2 kg-1), NL= Nitrogen-losses (%) and KM= Biodegradability coefficient as dependent variables b Each value is the average of three samples, dry weight basis. Percentages with respect to organic matter content. 3.3.1. PARETO CHART ANALYSIS Moreover, the crossed interactions between two independent variables terms in the equations (Table 3.3) make difficult the identification of the degree of influence of the independent on the dependent variables. Thus, a Pareto chart, also called a Pareto distribution diagram, is used to compare the cumulative effect of each of the independent variables (p<0.05). Fig. 3.1 shows a plot of
CAPÍTULO 3 78 each dependent variable (compost stability parameter) and its Pareto chart of standardized effects (as percentages) based on independent variables. Figure 3.1. Variation of dependent variables as a function of normalized independent variables (Pareto Chart).
CAPÍTULO 3 79 As can be seen, the C/N ratio is the strongest variable influencing OM evolution which indicates that the nature of the materials affects its transformation more than the classical variables like moisture or aeration. Moisture has the strongest effect on N-losses; time was found to be the most influential factor on nitrate, COD and biodegradability. Aeration shows a lower effect on the selected composting evolution parameters, which may indicate that even the lower aeration rate selected could maintain aerobic conditions. Moreover, moisture also shows small effect on these parameters with the exception of N-losses which could indicate that the daily moisture adjustments maintained enough water in the system even in the lower level. 3.3.2. ANALYSIS OF COMPOST STABILITY PARAMETERS. In order to determine the values of the independent variables giving the optimum values of the selected dependent variables, the predicted evolution for each variable was plotted at the three selected levels of the most influential independent variable and for a fixed value of the least influential variable (Figs. 3.2–3.6). 3.3.2.1. Organic Matter The OM (Fig. 3.2) variation indicates the relative decrease of the content of this parameter during the composting process. As expected, high OM content under high C/N ratio is observed (Yañez et al., 2009).The higher variation of the OM content was observed under greater C/N ratio ; which agrees with the results of previous reported studies. Nevertheless, this evolution basically depends on the raw materials used and its pre-treatment (Bernal et al., 1998; Diaz et al., 2002; Jiang et al., 2011; Nada et al., 2012; Rao et al., 1995). As expected, for all the C/N ratios, the OM losses tended to stabilise at the end of the composting
CAPÍTULO 3 86 Figure 3.5. N-Losses variation as a function of time at three moisture levels. 3.3.2.5. Chemical Oxygen Demand (COD) The COD (Fig. 3.6.) is a parameter related to compost stability (Haug, 1993) and may indicate the presence either of microbial stabilization, or it may be due to the presence of toxic organics inhibiting the microbial activities. This study showed that the relative decrease of COD was similar under the three C/N conditions, and tended to approach zero at the end of composting. Decreasing of the COD ratio meant that the compost became non-biodegradable, or stable in terms of no further biodegradation (Fdez.-Güelfo et al., 2011; Haug and Ellsworth, 1991). The absolute decrease of COD values was greater under low C/N ratio; this fact could be explained by the higher availability of dissolved Normalize Time (1-36 days) -1.0 -0.5 0.0 0.5 1.0 N-Losses (%) -8 -6 -4 -2 0 2 4 6 8 10 12 14 Low moisture level (40%) Medium moisture level (55%) High moisture level (70%)
CAPÍTULO 3 87 carbon present in the MSW compared with the materials with high C/N ratio such as LTR, where the carbon content is linked to the lignin due to the composition of the raw materials used to adjust this ratio. As expected, the same as with the degradability coefficient, time is an important variable influencing this parameter, which indicates that time has an influence on the process to obtain stable compost products. The time profile of compost stability could be divided into 2 stages, i.e. stage 1 (compost age between 0 and 15 days), stage 2 (after 15 days) which were classified as active degradation and the stability period respectively. The compost would be stable after 30 days. Figure 3.6. COD variation as a function of time at three C/N ratio levels. Normalize Time (1-36 days) -1.0 -0.5 0.0 0.5 1.0 COD (mg kg-1) 0 20000 40000 60000 80000 Low C/N ratio (21) Medium C/N ratio (60) High C/N ratio (77)
CAPÍTULO 3 88 3.4. CONCLUSIONS Both materials (MSW, LTR) evaluated in this study were adequate feedstocks for co-composting. The C/N ratio is the strongest variable influencing OM evolution. The losses of organic matter were reduced when MSW was co-composted with a minimum of LTR (lower C/N ratio). Nevertheless, the biodegradability and the N-losses increased under these conditions. On the other hand, moisture has the strongest effect on N-losses. The minimum N-losses during the composting process were found under medium and low values of this parameter (40-55 %). The weak influence of aeration on the dependent parameters indicates that an air flow of 0.05 Lair kg-1 min-1 is sufficient to guarantee the aerobic process. Time is an important parameter directly influencing the degradation of the materials during the process and affecting the stability of the products. The results of the modelling suggest that an optimized ratio MSW/LTR of 1/1 (w/w) (equivalent to C/N 60), moisture content in the range of 40-55% and moderate to low aeration rate (0.05-0.175 Lair kg-1 min-1) is recommended to maximise degradation and to obtain a stable product during co-composting of the described raw materials. The predicted evolutions and the relations among independent and dependent parameters obtained from the model used are in concordance with the current knowledge of the composting process, showing this model could be effectively applied to the composting process. .
CAPÍTULO 3 89 REFERENCES Adhikari, B.K., Barrington, S., Martinez, J., King, S., 2009. Effectiveness of three bulking agents for food waste composting. Waste Manage. (Oxford) 29, 197-203. Akhnazarova, S., Kafarov, V., 1982. Experiment Optimization in Chemistry and Chemical Engineering. Mir Publisher, Moscow. Barnhisel, R., Bertsch, P.M., 1982. Digestion with pechloric-nitric acids, in: Page, A.L. (Ed.), Methods of soil analysis, Part 2 Chemical and Microbial Properties. Monograph Agronomy ASA, Madison, WI, USA, pp. 279-280. Benito, M., Masaguer, A., Moliner, A., Arrigo, N., Palma, R.M., 2003. Chemical and microbiological parameters for the characterisation of the stability and maturity of pruning waste compost. Biol. Fert. Soils 37, 184-189. Bernal, M.P., Paredes, C., Sánchez-Monedero, M.A., Cegarra, J., 1998. Maturity and stability parameters of composts prepared with a wide range of organic wastes. Bioresour. Technol. 63, 91-99. Box, G., Behnken, D.W., 1960. Some new three level designs for the study of quantitative variables. Technometrics 2, 455-475. Bremnen, J.M., 1996. Methods of Soil Analysis. Part 3. Chemical Methods American Society of Agronomy-Soil Science Society of America, Madison, WI. United States. Bueno, P., Tapias, R., López, F., Díaz, M.J., 2008. Optimizing composting parameters for nitrogen conservation in composting. Bioresour. Technol. 99, 5069-5077. Bueno, P., Yañez, R., Caparrós, S., Díaz, M.J., 2009a. Evaluating environmental parameters for minimum ammonium losses during composting of trimming residues. J. Air Waste Manage. Assoc. 59, 790- 800. Bueno, P., Yañez, R., Rivera, A., Díaz, M.J., 2009b. Modelling of parameters for optimization of maturity in composting trimming residues. Bioresour. Technol. 100, 5859-5864. Chang, J.I., Chen, Y.J., 2010. Effects of bulking agents on food waste composting. Bioresour. Technol. 101, 5917-5924. Delgado-Rodríguez, M., Ruiz-Montoya, M., Giraldez, I., Cabeza, I.O., López, R., Díaz, M.J., 2010. Effect of control parameters on emitted volatile compounds in municipal solid waste and pine trimmings composting. J. Environ. Sci. Heal. A 45, 855-862. Diaz, M.J., Madejón, E., López, F., López, R., Cabrera, F., 2002. Optimization of the rate vinasse/grape marc for co-composting process. Process Biochem. 37, 1143-1150. Dorado, A.D., Lafuente, F.J., Gabriel, D., Gamisans, X., 2010. A comparative study based on physical characteristics of suitable packing materials in biofiltration. Environ. Technol. 31, 193-204. Doublet, J., Francou, C., Poitrenaud, M., Houot, S., 2011. Influence of bulking agents on organic matter evolution during sewage sludge composting; consequences on compost organic matter stability and N availability. Bioresour. Technol. 102, 1298-1307. Eftoda, G., McCartney, D., 2004. Determining the critical bulking agent requirement for municipal biosolids composting. Compost Sci. Util. 12, 208- 218.
CAPÍTULO 3 90 El Kader, N.A., Robin, P., Paillat, J.M., Leterme, P., 2007. Turning, compacting and the addition of water as factors affecting gaseous emissions in farm manure composting. Bioresour. Technol. 98, 2619-2628. EN-13040, 1999. Soils Improvers and Growing Media-Sample Preparation for Chemical and Physical Tests, Determination of Dry Matter Content, Moisture Content and Laboratory Compacted Bulk Density. European Committee for Standardization. European-Commission, 2010. Commission Staff Working Document. Accompanying the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions on the Thematic Strategy on the Prevention and Recycling of Waste. http://ec.europa.eu/environment/waste/pdf/Commission%20Working%20Doc.pd f. Accessed 10 October 2012Eurostat, 2001. Waste management options and climate change. Office for Official Publications of the European Communities, Luxembourg. http://europa.eu/rapid/pressReleasesAction.do?reference=STAT/11/37&for mat=html. Accessed 10 October 2012 Eurostat, 2011. Eurostat-News release. Environment in the EU27. Eurostat Press Office, Luxembourg. http://ec.europa.eu/environment/waste/studies/pdf/climate_change.pdf. Accessed 10 October 2012 Fdez.-Güelfo, L.A., Álvarez-Gallego, C., Sales, D., Romero, L.I., 2011. The use of thermochemical and biological pretreatments to enhance organic matter hydrolysis and solubilization from organic fraction of municipal solid waste (OFMSW). Chem. Eng. J. 168, 249-254. Garcia, C., Hernandez, T., Costa, F., 1991. Changes in carbon fractions during composting and maturation of organic wastes. Environ. Manage. 15, 433- 439. Gea, T., Barrena, R., Artola, A., Sánchez, A., 2007. Optimal bulking agent particle size and usage for heat retention and disinfection in domestic wastewater sludge composting. Waste Manage. (Oxford) 27, 1108-1116. Gómez, X., Cuetos, M.J., Cara, J., Morán, A., García, A.I., 2006. Anaerobic codigestion of primary sludge and the fruit and vegetable fraction of the municipal solid wastes. Conditions for mixing and evaluation of the organic loading rate. Renew. Energ. 31, 2017-2024. Habart, J., Tlustoš, P., Hanč, A., Švehla, P., Váňa, J., Tluka, P., Jelínek, F., 2010. The role of aeration intensity, temperature regimes and composting mixture on gaseous emission during composting. Compost Sci. Util. 18, 194-200. Haug, R.T., 1993. The Practical Handbook of Compost Engineering. Lewis Publishers, United States of America. Haug, R.T., Ellsworth, W.F., 1991. Measuring compost substrate degradability. Biocycle 32, 56-60, 62. Hwang, S., Hanaki, K., 2000. Effects of oxygen concentration and moisture content of refuse on nitrification, denitrification and nitrous oxide production. Bioresour. Technol. 71, 159-165. Illa, J., Sole-Mauri, F., 2008. Modeling the composting process: coupling physical and biological processes, in: Huerta, O., Lopez, M., Martinez, F.
CAPÍTULO 3 91 (Eds.), Proceso y destino del compost, formación, información e interrelaciones entre los agentes del sector. Red Española de Compostaje, Spain. INE, 2010. Environmental Report .National Statistical Institute of Spain Bulletin. http://www.ine.es/prensa/np616.pdf. Accessed 10 October 2012 (In Spanish) Jiang, T., Schuchardt, F., Li, G., Guo, R., Zhao, Y., 2011. Effect of C/N ratio, aeration rate and moisture content on ammonia and greenhouse gas emission during the composting. J. Environ. Sci. 23, 1754-1760. Körner, I., Braukmeier, J., Herrenklage, J., Leikam, K., Ritzkowski, M., Schlegelmilch, M., Stegmann, R., 2003. Investigation and optimization of composting processes - Test systems and practical examples. Waste Manage. (Oxford) 23, 17-26. López, R., Cabeza, I.O., Giráldez, I., Díaz, M.J., 2011. Biofiltration of composting gases using different municipal solid waste-pruning residue composts: Monitoring by using an electronic nose. Bioresour. Technol. 102, 7984-7993. Madejón, E., Jesús Díaz, M., López, R., Cabrera, F., 2002. New approaches to establish optimum moisture content for compostable materials. Bioresour. Technol. 85, 73-78. Mata-Alvarez, J., Macé, S., Llabrés, P., 2000. Anaerobic digestion of organic solid wastes. An overview of research achievements and perspectives. Bioresour. Technol. 74, 3-16. Morgan-Sagastume, J.M., Noyola, A., 2006. Hydrogen sulfide removal by compost biofiltration: Effect of mixing the filter media on operational factors. Bioresour. Technol. 97, 1546-1553. Murillo, J.M., Cabrera, F., López, R., 1997. Response of clover Trifolinm fragiferum L. cv. 'Salina' to a heavy urban compost application. Compost Sci. Util. 5, 15-25. Nada, W.M., Van Rensburg, L., Claassens, S., Blumenstein, O., Friedrich, A., 2012. Evaluation of organic matter stability in wood compost by chemical and thermogravimetric analysis. Int. J. Environ. Res. 6, 425-434. Ostos, J.C., López-Garrido, R., Murillo, J.M., López, R., 2008. Substitution of peat for municipal solid waste- and sewage sludge-based composts in nursery growing media: Effects on growth and nutrition of the native shrub Pistacia lentiscus L. Bioresour. Technol. 99, 1793-1800. Pahl, O., Firth, A., MacLeod, I., Baird, J., 2008. Anaerobic co-digestion of mechanically biologically treated municipal waste with primary sewage sludge - A feasibility study. Bioresour. Technol. 99, 3354-3364. Raj, D., Antil, R.S., 2011. Evaluation of maturity and stability parameters of composts prepared from agro-industrial wastes. Bioresour. Technol. 102, 2868-2873. Rao, N., Grethlein, H.E., Reddy, C.A., 1995. Effect of C/N ratio and moisture content on the composting of poplar wood. Biotechnol. Lett. 17, 889-892. Schlegelmilch, M., Streese, J., Stegmann, R., 2005. Odour management and treatment technologies: An overview. Waste Manage. (Oxford) 25, 928-939. Sole-Mauri, F., Diaz, M.J., 2007. Modelización Matemática del Proceso de Compostaje, in: Moreno, J., Moral, R. (Eds.), Compostaje. Mundi-prensa, Spain.
CAPÍTULO 3 92 Stoffella, P.J., Kahn, B.A., 2005. Utilización de Compost en los Sistemas de Cultivo Hortícola. Ediciones Mundi-Prensa, Spain. Thompson, W., Leege, P., Millner, P., Watson, M.E., 2003. Test Methods for the Examination of Composts and Composting. US Government Printing Office, USA. Tomati, U., Galli, E., Pasetti, L., Volterra, E., 1995. Bioremediation of olive-mill wastewaters by composting. Waste Manage. Res. 13, 509-518. Tuomela, M., Vikman, M., Hatakka, A., Itävaara, M., 2000. Biodegradation of lignin in a compost environment: A review. Bioresour. Technol. 72, 169-183. Walker, L., Charles, W., Cord-Ruwisch, R., 2009. Comparison of static, invessel composting of MSW with thermophilic anaerobic digestion and combinations of the two processes. Bioresour. Technol. 100, 3799-3807. Yañez, R., Alonso, J.L., Díaz, M.J., 2009. Influence of bulking agent on sewage sludge composting process. Bioresour. Technol. 100, 5827-5833. Yañez, R., Bueno, P., Rivera, A., García-Barneto, A., Díaz, M.J., 2010. Selective organic compounds degradation under controlling composting conditions. Waste Manage. (Oxford) 30, 755-763. Zhang, Y.M., Huang, G.H., He, L., 2011. An inexact reverse logistics model for municipal solid waste management systems. J. Environ. Manage. 92, 522- 530. Zmora-Nahum, S., Markovitch, O., Tarchitzky, J., Chen, Y., 2005. Dissolved organic carbon (DOC) as a parameter of compost maturity. Soil Biol. Biochem. 37, 2109-2116.
CAPÍTULO 4 93 4. BIOFILTRATION OF COMPOSTING GASES USING DIFFERENT MUNICIPAL SOLID WASTE-PRUNING RESIDUE COMPOSTS: MONITORING BY USING AN ELECTRONIC NOSE. 4.1. INTRODUCTION Composting facilities are a great source of unpleasant smells, generating a nuisance to nearby residents and causing rejection of these kinds of facilities (Müller et al., 2004; Schlegelmilch et al., 2005; Smet et al., 1999). These smells are caused by substances such as some inorganic gases (ammonia and hydrogen sulphide) and volatile organic compounds (VOCs) released during the biodegradation of organic residues (Müller et al., 2004; Shareefdeen et al., 2005). Generally, the reached concentrations of VOCs and other odorant gases are not of toxicological relevance, but the people working or living in the vicinity of the composting plants may suffer diverse psycho-hygienic effects (Müller et al., 2004). Consequently, several countries have established regulations to guarantee nuisance-free, breathable air. The legislation copes with this problem in two ways: setting minimum distances from the facilities to housing or limiting the odour emissions from them. The abatement of odour and VOC emissions in the facilities could be achieved through an adequate control of the composting conditions. Schelegelmilch et al. (2005) indicated the dependence of odours on some operations (storage, turning, screening, cleaning) carried out in the composting plants. Recently, Delgado-Rodríguez et al. (2010) studied the influence of process parameters (C/N, moisture and aeration) on VOC emissions. Once air streams in a composting plant could be collected,
CAPÍTULO 4 94 biofiltration is considered as a recommended technology for the final VOCs abatement because of its capacity to treat low concentrations of diverse pollutants, its cost-effectiveness, simple operation and absence of secondary contaminated waste streams (Datta and Allen, 2005; Namkoong et al., 2003; Pagans et al., 2006). Common biofilter media include peat, wood bark, wood chips, soil, compost, coated ceramic particles, synthetic media or a combination of these products (Schelegelmilch et al., 2005; Shareefdeen et al., 2005). Nevertheless, few studies have compared different materials or composts (Álvarez-Hornos et al. 2008). Several analytical methods, usually in conjunction, have been used for the monitoring of gas concentrations in the composting and biofiltration process. Characterization and determination of specific VOCs are performed by a variety of gas chromatography (GC) instruments, mass spectrometry (MS) and flame ionization detector (FID) being probably the most frequent detection techniques (European Commission, 1999). Wang and Austin (2006) reviewed sampling and analytical methods for VOCs in air and they concluded that current inventories of VOC emissions remain subject to considerable uncertainty due to variation in methods. Several authors (Karlik et al., 2002; Ojala et al., 2006) have suggested the possibility of using total-VOCs analysers equipped with PID detector to obtain rapid information on the concentration of VOCs. The USEPA (2007) recommends this instrument as a screening procedure for field use. Each different VOC produces different signal intensity in the PID detector. In the case of a single VOC gas, the reading of the VOCs analyser can be a reliable measure of its concentration as long as the instrument is calibrated with the
CAPÍTULO 4 95 same VOC. In the case of gas mixtures differing in composition, the VOCs reading is considered a semiquantitative approach of the total concentration of VOCs, referred to the calibrating compound. In this sense VOCs measurement is similar to odour determination by olfatometry: the obtained value denotes the sum of single VOC signals, each one proportional to a correction factor or threshold. In contrast to odour concentration, readings of the VOC analyser vary linearly with the concentration of VOCs. Additional advantages of the VOCs analyser are its portability, accuracy and quick response time (< 2 s). Chemical (GC-MS) and olfactory analysis have been used serially a few times (Chen et al. 2008). Gas detector tubes for sampling and quantification have also been used for selected compounds or VOCs families (Tsai et al., 2008).To measure odour emission (or concentration) as a whole, olfactometry is generally the selected method (Mao et al., 2006; Tsai et al., 2008). The European Union adopted the standard olfactometric method EN 13725 (CEN, 2003) based on dilution of an odorous sample to the odour threshold detectable to 50 % of a test panel (Schelegelmilch et al., 2005). Even though standardized olfactometric methods permit enough reproducibility and they are reliable for concentrated emissions, their applicability for low concentrations was doubtful (Littarru, 2007). Olfactometry is a time- and cost-intensive method and the applicability of a human panel especially for field measurements involves practical difficulties (Muller et al., 2004; Figueiredo and Stentiford, 2001). Since the eighties, but particularly during the last decade, electronic noses (e-noses) have attracted interest to overcome the limitations of the human sense and have been used for qualitative and quantitative gas analysis in environmental monitoring (Ameer
CAPÍTULO 4 102 correspond to the first two principal components and samples are distributed in this two dimensional space. The legends for the x- and y-axes contain the value of the variance achieved by the PCA component. Euclid and correlation classification have been used as sample classifiers. Partial least squares regression (PLS) has been used for prediction of total VOCs using VOC analyser readings as training results. Each reading from the VOC analyser was assigned to its corresponding e-nose measurement, and the descriptor was recalculated from PLS regression. These statistical analyses were done by the e-nose built-in software (WinMuster, 2010). 4.3. RESULTS AND DISCUSSION Selected physico-chemical characteristics of the biofilter media are shown in table 4.1. In general terms, the differences in compost properties were related to compost type (MSW or MSWP) showing both size grades of MSW composts higher pH, electrical conductivity, nitrogen content and bulk density than both grades of MSWP composts. All the composts kinds showed adequate pH for microorganism activity (pH 7 to 8) (Datta and Allen 2005), Higher bulk density of MSW composts could favour biofilter compaction.
CAPÍTULO 4 103 Table 4.1.Characteristics of the compost-based biofilter filling media Compost type MSW2 MSW7 MSWP2 MSWP7 pH 7.63 a 7.50 a 7.23 b 7.20 b E.C. a (1:5 extract) dS m- 1 5.28 a 5.56 a 1.92 b 1.90 b Organic Carbon g kg-1 123 a 180 a 383 b 484 b Nitrogen g kg-1 15.3 a 13.1 a 11.9 ab 9.1 b C/N 8.1 13.8 32.3 53.3 Bulk density kg m- 3 485 a 354 b 269 c 233 c a Electrical conductivity Values in the same row followed by the same letter are not statistically different (Tuckey’s HSD test, p<0.05) The total-VOCs evolutions during one day period for the input and output air of biofilter MSWP2 are shown in fig. 4.2. Both sets of measurements were done on two consecutive days, starting after the wastes addition to the composter. These previous waste additions were 136 L day-1 (MSW + P) and 105 L day-1 in the case of input and output streams respectively. Average VOCs (24 h) concentrations for input and output air were 4,699 and 901 ppbv respectively but several peaks and fluctuations were observed during the course of measurements due to composter turning and ventilation. Such fluctuations have also been observed during the compost transfer in the reactors (Ryu et al. 2011). The maximum VOC content for input air was 54,229 ppbv and the maximum for output air was 17,372 ppbv. These values were reached 12 h after waste additions, coinciding in time with a period of rotation of the composter and ventilation, which benefitted concentration of VOCs . Several authors detected maximum VOCs emissions during the first 48 h of composting and Pagans et al. (2006) indicating that such emissions were neither related to the biological activity nor the temperature. If both series of readings were
CAPÍTULO 4 104 compared, the (estimated) biofilter removal efficiency at the 12 h peak was lower than during the rest of the time. The decrease in removal efficiency under high VOCs loadings has been previously reported (Datta and Allen, 2005) and it could be due to a need of microorganism acclimatation (Kleinheinz et al., 1999). In these kind of systems subject to ample fluctuations, the monitoring of the concentration of VOCs by means of GC-MS systems could be difficult. To obtain reliable results, composite samples taken during a period of several hours or an elevated number of single time samples would be needed. In spite of its semi-quantitative character, the VOCs analyser could provide enough information on the overall biofiltration (or composting) process if extended sampling times is used. Fig.4.2. Daily evolution of concentration of total-VOCs in input and output air of MSW + pruning waste, particle size 2-7 mm (MSWP2) biofilter. Sampling time (hours) 0 5 10 15 20 Concentration of VOCs (ppbv) 0 10000 20000 30000 40000 50000 60000 Biofilter input MSWP2 biofilter output
CAPÍTULO 4 105 From total-VOCs readings, VOCs removal efficiencies for the 4 biofilter types were calculated (Table 4.2). These values should be taken with caution because the readings were obtained in a short sampling time. Regardless of compost type, the acclimatation period of the biofilters was about 5 days. Periods of acclimatation between 5 to 20 days are consistent with reported periods for compost biofilters (Hernández et al. 2010; Pandey et al. 2010; Raghuvanshi and Babu 2009). Efficiencies were greater than 90% on days 6 and 12 for which the concentrations of VOCs in input air were lower. Using biofilters with similar filling materials Pagans et al. (2006) found similar efficiencies (80-90%) treating gas from MSW composting. With respect to removal efficiency there is not clear differentiation that could be associated with biofilter nature or particle size. Table 4.2.Concentration of VOCs in input air and VOC removal efficiency (%) for the different biofilter types. Composting Day 1 5 6 12 20 Input air (ppbv) 15595 4821 2725 3223 13650 MSW2 72.9 a 44.5 a 92.7 a 84.7 a 81.4 a MSW7 67.3 ab 62.1 b 93.0 a 90.3 ab 82.0 a MSWP2 51.7 ab 62.2 b 94.3 a 95.1 b 89.0 a MSWP7 47.3 b 61.6 b 97.1 a 94.2 b 83.8 a Values in the same column followed by the same letter are not statistically different (Tuckey’s HSD test, p<0.05) The biofilters performance over longer time periods could be altered by several factors such as clogging, changes in chemical properties (e.g. pH), poisoning due to chemical accumulation (e.g. NO3- or SO42-) and this is not considered in this paper.
CAPÍTULO 4 106 Considering separately e-nose data sets for each sampling day, Principal Component Analysis (PCA) revealed clear differences among biofilter input and output air (figure not shown). Fig. 4.3, shows the two principal components given by PCA applied to e-nose data corresponding to day 12 of composting, considering biofilter particle size (greater or lesser than 7 mm regardless of compost raw material, fig. 4.3A) or compost raw material (MSW or MSWP regardless of particle size, fig. 4.3B) as classes. Sample dilution factor was not included in the data set, taking into account that it was the same for the considered samples. On day 12 it was not possible to differentiate outlet air samples from the biofilters by compost size, but samples were separately grouped if compost raw material was the considered class factor. The MSW and MSWP groups were mainly separated along the y axis (function 2) which accounted for 17% of the variance. The loadings analysis of factors makes it possible to check the influence of each sensor on the distribution of data within the PCA-space. The loadings analysis showed that the sensor detecting methane-aliphatic compounds, the two sensors detecting sulphur organic compounds and the broad range sensor accounted for the grouping difference along the y axis. At least at this stage of the composting process, the origin of the biofilter materials seemed to have a moderate influence over the composition of VOCs in the output air.
CAPÍTULO 4 107 Fig. 4.3. Principal Component Analysis plots considering particle size (A) or compost type (B) as classification factor in samples corresponding to day 12 of composting. Wrapping lines correspond to Hem classes. >7mm <7mm MSWP MSW
CAPÍTULO 4 108 PCA analysis of samples on day 12 of composting, considering for classification the four composts used, is shown in fig. 4.4. Based on the pattern including the data of the 10 e-nose sensors (fig. 4.4 A), the loading analysis showed that sensors that mainly detected sulphur-organic, sulphur-chloride and methanealiphatic compounds respectively were the more discriminating ones. These results agreed with the content of the sulphur-compounds (sum of dimethylsulphide and dimethyldisulphide) determined by GC-MS on day 20 which were 2.3 times higher in MSW biofilters than in MSWP biofilters (López et al., 2010). Both sulphur compounds, dimethylsulphide and dimethyldisulphide, were frequently found in the air emissions of composting facilities (Müller et al., 2004; Smet et al., 1999; Tsai et al., 2008). If only the responses corresponding to these 3 sensors are selected for the PCA, a better resolution among classes was observed (fig. 4.4B). The associated variance to the two principal components in PCA was also improved (84.4 % with 10 sensors, 99.9 % with 3 sensors). From the fig. 4.4B, it can be concluded that the air from small-grained biofilters, MSW2 and MSWP2, appeared quite different between them and in the same way, the air from the coarse-grained biofilters (MSW7 and MSWP7), was more similar among them. Both factors, compost origin and particle size seem to have affected the composition of VOCs of the output air. The readings of total-VOCs corresponding to these samples were in the order of one to seven hundred ppbv and therefore, the electronic nose was able to detect very small differences in air composition.
CAPÍTULO 4 109 Fig. 4.4. Principal Component Analysis plots considering the four compost types as classification factor. Analysis considering 10 sensor data (A) and 3 selected sensor data (B) on day 12. Wrapping lines correspond to Hem classes. MSWP7 MSW2 MSWP2 MSW7 A B MSW7 MSW2 MSWP2 MSWP7
CAPÍTULO 4 110 It was not surprising that the e-nose detected such differences (input air-output air, biofilter type), since if you consider the data collected in only one day, the situation can be described as relatively static. Ljungberg Willing et al. (1998), compared human and electronic responses to paperboard odour and they found that some e-nose sensors can be correlated with a selected group of odour descriptors determined by a panel. Sironi et al. (2007) found coherency between the odour detections with an e-nose and the material turning or moving operations inside a composting plant, at least during a short period of 5 days. Using a similar e-nose Littarru (2007) detected qualitative differences between biofilter emissions related to the age of the composting waste. Besides , there are no previous papers which reported, using an e-nose, such differences comparing similar biofilter materials at these low concentrations of total-VOCs . The complete data set including the samplings corresponding to 1, 5, 6, 12 and 20 days of composting was also studied. In spite of the evolution of composting, which originates differences in the composition of exhaust gases, several patterns could be successfully applied. PCA on fig. 4.5A showed good separation if biofilter input (32 samples) and output air (135 samples) are the considered classes. Only 6 sensor signals and the dilution factor were used in this statistical analysis. The less loading sensors in the distribution were those more nonspecific, detecting a broad range of compounds.
CAPÍTULO 4 111 Fig. 4.5. Principal Component Analysis plots for the whole composting process. A: Classes are biofilter input and output air; data set: 6 sensors and dilution factor. B: Classes are the 4 types of compost in biofilter; data set: 3 sensor data and dilution factor. Wrapping lines correspond to Hem classes. A OUTPUT INPUT B MSWP2 MSW7 MSW2 MSWP7
CAPÍTULO 4 118 Application to a Bench-Scale Biofiltration System for VOC Degradation. Arch. Environ. Contam. Toxicol. 37, 151-157. Ljungberg Willing B-I., Brundin A., Ludström I., 1998. Odour Analysis of Paperboard, the Correlation between Human Senses and Electronic Sensors using Multivariate Analysis. Packag. Technol. Sci. 11, 59-67. Littarru P., 2007. Environmental odours assessment from waste treatment plants: Dynamic olfactometry in combination with sensorial analysers “electronic noses”. Waste Manage. 27, 302-309. López R., Cabeza I.O., Giráldez I., Díaz M.J., 2010. Biofiltración de olores en el compostaje de RSU: Monitorización mediante diversas técnicas. II Jornadas de la Red Española de Compostaje: Compostaje de Residuos Orgánicos y Seguridad Medioambiental. Burgos-Palencia. Mao I-F., Tsai C-J., Shen S-H., Lin T-F., Chen W-K., Chen M-L., 2006. Critical components of odors in evaluating the performance of food waste composting plants. Sci. Total Environ. 370, 323-329. Müller T., Thißen R., Braun S., Dott W., Fischer G., 2004. (M)VOC and Composting Facilities. Part 2: (M)VOC Dispersal in the Environment. Environ. Sci. Pollut. R. 11, 152-157. Namkoong W., Park J-S., VanderGheynst J. S., 2003. Biofiltration of gasoline vapour by compost media. Environ. Pollut. 121, 181-187. Ojala S., Lassi U., Keiski R.L., 2006. Testing VOC emission measurement techniques in wood-coating industrial processes and developing a costeffective measurement methodology. Chemosphere 62, 113-120. Orzi V., Cadena E., D’Imporzano G., Artola A., Davoli E., Crivelli M., Adani F., 2010. Potential odour emission measurement in organic fraction of municipal solid waste during anaerobic digestión: Relationship with process and biological stability parameters. Bioresour. Technol. 101, 7330-7337. Pagans E., Font X., Sánchez A., 2006. Emission of volatile organic compounds from composting of different solid wastes: Abatement by biofiltration. J. Hazard. Mater. B131, 179-186. Pandey R.A., Joshi P.R., Mudliar S.N., Deshmukh S.C., 2010. Biological treatment of waste gas containing mixture of monochlorobenzene (MCB) and benzene in a bench scale biofilter. . Bioresour. Technol. 101, 5168- 5174. Raghuvanshi S., Babu B.V., 2009. Experimental studies and kinetic modelling for removal of methyl ethyl ketone using biofiltration. Bioresour. Technol. 100, 3855-3861. Rajamäki T., Arnold M., Venelampi O., Vikman M., Räsänen J., Itävaara M., 2005. An electronic nose and indicator volatiles for monitoring of the composting process. Water, Air, Soil Pollut. 162, 71-87. Ryu H.W., Cho K-S., Lee T-H., 2011. Reduction of ammonia and volatile organic compounds from food waste-composting facilities using a novel anti-clogging biofilter system. Bioresour. Technol. 102, 4654-4660. Schlegelmilch M., Stresse J., Biedermann W., Herold T., Stegman R., 2005. Odour control at biowaste composting facilities. Waste Manage. 25, 917- 927.
CAPÍTULO 4 119 Shareefdeen Z., Herner B. Sing A., 2005. Biotechnology for Air Pollution Control – an Overwiew, in Shareefdeen Z., Sing A. (Eds.), Biotechnology for Odor and Air Pollution Control. Springer-Verlag, Berlin Heidelberg, pp. 3-15. Sironi S., Capelli L., Céntola P., Del Rosso R., Il Grande M., 2007. Continuous monitoring of odours from a composting plant using electronic noses. Waste Manage. 27, 389-397. Smet E., Van Langenhove H., De BO I., 1999. The emission of volatile compounds during the aerobic and the combined anaerobic/aerobic composting of biowaste. Atmos. Environ. 33, 1295-1303. Tsai C-J., Chen M-L., Ye A-D., Chou M-S., Shen S-H., 2008. The relationship of odor concentration and the critical components emitted from food waste composting plants. Atmos. Environ. 42, 8246-8251. USEPA, 2007. Method 3815: Screening solid samples for volatile organics, in EPA publication SW-846, Test Methods for Evaluating Solid Waste, Physical/Chemical Methods, http://www.epa.gov/SW-846, last accessed February 4, 2011. Wang D. K. W., Austin C. C. 2006. Determination of complex mixtures of volatile organic compounds in ambient air: an overview. Anal. Bioanal. Chem. 386, 1089-1098. WinMuster 1.6.2.14, 2010. Airsense Analytics GmbH. Hagenover, Schwerin, Germany.
CAPÍTULO 5 120 5. BIOFILTRATION OF α-PINENE VAPOURS USING MUNICIPAL SOLID WASTE (MSW) – PRUNING RESIDUES (P) COMPOSTS AS PACKING MATERIALS 5.1. INTRODUCTION One of the main problems associated with composting facilities is the odours generated during the process, including the odours produced during the reception and the handling of materials, aerobic treatment, stock piling, etc. The emission of these by-products can cause community annoyance and public opposition to composting plants, not only due to the odours but also due to the potential health risks to workers and inhabitants in the local area (Pagans et al., 2005; Tsai et al., 2008). The most important group of chemicals responsible for this odour are VOCs and some inorganic gases (ammonia, hydrogen sulfide, etc) produced during the biodegradation of organic residues (Beck-Friis et al., 2001; Delgado-Rodríguez et al., 2010; Müller et al., 2004; Stoffella & Kahn, 2005). The family of terpenes is one of the most representative classes of VOCs emitted during the degradation process of vegetal materials (chips, grass clippings, pruning residues, etc). Amongst them, α-pinene is frequently the predominant compound, representing between 10.2 and 72.7 % of the total emissions (Büyüksönmez & Evans, 2007). Furthermore, α-pinene is also emitted by wood processing industries, in particular, saw mills, composite board mills, and paper industries (Mohseni & Allen, 2000; Van Groenestijn & Liu, 2002). Although, the optimization of the operational parameters of composting processes is an important objective for the scientific community in order to reduce the VOC´s emissions (Delgado-Rodríguez et al., 2010; Delgado-
CAPÍTULO 5 121 Rodríguez et al., 2011); different techniques exist when the minimisation at source is ineffective. Amongst these techniques, the most widely used are the adsorption on activated carbon, scrubbers and bioscrubbers, condensation, thermal or catalytic incineration, and biofiltration. (Namkoong et al., 2003; Revah & Morgan-Sagastume, 2005; Schlegelmilch et al., 2005). Biofiltration is considered to be an advantageous system for deodorisation due to low operating costs and their abilty to treat large volumes of waste gas streams containing low concentrations of odorous compounds (Dorado et al., 2010; Mudliar et al., 2010). Biofilters allow for the conversion of gas-phase chemical compounds to transform into common biological degradation products, such as carbon dioxide, water, and mineral salts. In the bioreactor, contaminated air is passed through a bed of porous and moist medium (packing material), and the contaminants are sorbed to the medium surface where they are degraded by microorganisms (Datta & Grant-Allen, 2005). As the treatment of VOCs in this kind of technology requires the transportation of the compound from the gas phase to the biofilm that forms upon a packing material to be available for the degradation by microorganisms (Miller & Allen, 2005), the solubility of the compound in water is a limiting factor in the process. For this reason, the study and optimization of hydrophobic compounds treatment, using biofiltration technology is a challenge for the scientific community (Mohseni & Allen, 2000; Paca et al., 2010; Vergara-Fernández et al., 2012). α-pinene, one of the major hydrophobic compounds, was selected as the marker VOC in this study. Considering the low solubility of α-pinene in the water phase (2.5 mgl-1 at 23ºC) (Bagherpour et al., 2005; Dhamwichukorn et al., 2001), it was interesting to evaluate the treatment of this contaminant through
CAPÍTULO 5 122 biofiltration technology. Biofilter monitoring is usually carried out using GC/MS and olfactometric techniques when an evaluation of the odours is required (Chen et al., 2008; Delgado-Rodríguez et al., 2011; Dever et al., 2007; Romain et al., 2009). The use and advantages of e-noses have already been widely reported in order to evaluate the presence of VOCs and odours (Delgado-Rodríguez et al., 2012; Littarru, 2007; López et al., 2011; Rajamäki et al., 2005; Stuetz & Nicolas, 2001). Nevertheless, few studies have assessed the application of e-noses in the monitoring of biofiltration. Online instruments such as photoionization detector (PID) have also been proposed to obtain rapid information on the concentration of VOCs in different kinds of processes (Hobbs et al., 1995; Karlik et al., 2002; Muñoz et al., 2010; Smith et al., 2007). The main advantages of the VOCs analyser are its portability, accuracy, quick response time and reliability in the case of a specific VOC gas (less than 2 s) (López et al., 2011). This study focused on the evaluation of the VOCs removal efficiency for biofilters that used compost either MSW and MSW-Pruning residues as the packing material. The biofilters treated an artificial stream of gases containing a hydrophobic VOC (α-pinene). This evaluation was supported by several analytical techniques, such as VOC analyser, e-nose and GC/MS analysis. There was a particular focus on the removal efficiencies achieved in the biofilters with different moisture contents in their packing materials.
CAPÍTULO 5 123 5.2. MATERIALS AND METHODS 5.2.1. BIOFILTRATION UNIT The biofiltration system consisted of two laboratory scale biofilters (Fig 5.1.), each using a different type of mature compost as its packing material. Each biofilter consisted of a PVC cylinder of 11 cm in diameter and 1 m in height filled with the packing material in the upper 95 cm (bed volume 9.0 L). One biofilter was filled with compost from MSW and the other one with a mixture of compost from MSW and Pruning Residues (P) in a volumetric ratio 1:1. Some relevant physico-chemical characteristics of the packing materials are shown in Table 5.1., additional details can be found in Delgado-Rodriguez et al. (Delgado- Rodríguez et al., 2010). Table 5.1. Relevant physic-chemical properties of the packing materials used in the biofiltration system proposed (over dry basis) (average ± standard deviation a). MSWb MSW-Pb pH (1:5 extract) 6.55 ± 0.2 5.60 ± 0.1 EC (1:5 extract) mS m-1 12,45 ± 0.1 11,57 ± 0.1 Organic Matter g kg-1 301 ± 82 842 ± 93 Kjeldahl-N2 g kg-1 14.1 ± 1.4 11 ± 0.4 C/N 12.55 45.03 Bulk density g L-1 705 ± 50 374 ± 54 a Average ± standard deviation, over four samples, (d.w.) b MSW: Municipal solid waste composts; MSW-P: Municipal solid waste – Pruning residues composts The granulometry of both packing materials was from 7 to 20 mm, in order to improve removal efficiency and avoid operational problems (clogging, control of air flux, etc). The packing materials had previously been used to treat a gas stream composed of a VOC mixture from the active composting of MSW for up to two months (López et al., 2011). Due to this fact and that natural materials
CAPÍTULO 5 124 were used as bed packing, no inoculation was performed. The initial moisture content of the packing materials was determined by oven drying at 105 ºC. During the course of the experiment the moisture content of the packing materials was controlled periodically by measuring the difference in weight between the complete biofilter bed and its initial weight. When the addition of water was necessary, it was added to the top part of the biofilter. As the nature of the materials was different, the moisture content was expressed as a percentage of the sample's dry weight (d.w) to be able to compare each medium. Figure 5.1. Schematic diagram of the biofiltration system. (1) Pressurized air; (2) Compressor; (3) Humidifier; (4) α-pinene sparger; (5) Biofilter compost MSW; (6) Biofilter compost MSW-P; (7) input sampling port; (8,9) outlet biofilter gas/output sampling port; (10,11) Water drainage
CAPÍTULO 5 125 The inlet gas stream was supplied from the bottom of the column (ascending flow). Pressurized ambience air was obtained using a compressor of oil-free products; the air was taken from the surroundings with the purpose to operate continuously and to facilitate the following humidification process of the air. Depending on the stage of the experiment, the inlet gas was humidified to 90- 100% relative humidity, by sparging the gas stream through a reservoir filled with water. In order to generate the contaminated airflow, a small fraction of humidified air was fed to a sparger, which contained liquid α-pinene (Merck, >95% purity). Subsequently, this stream was joined to the main distribution line and introduced into the biofiltration system. The inlet concentration of the contaminant was adjusted by the flow rate of the gas stream with α-pinene; guaranteeing the concentration of the contaminant fixed during 6 to 8 hours. The average α-pinene concentration during all the stages of the experiment was 11.6 ppmv . The empty bed residence time (EBRT) was adjusted to 66 s. EBRT values of about a 1 min are frequent in biofiltration experiments in order to obtain effective degradation in the system (Box & Behnken, 1960; Dorado et al., 2008; López et al., 2011; Omri et al., 2011; Ramírez et al., 2011). The experiment was separated into four stages, modifying some operative conditions of the system proposed: (i) Days 1-40 -α-pinene concentrations below 5 ppmv with maximum moisture content of the packing materials; (ii) Days 80-144 - Gradual increase of the α-pinene concentration in the inlet gas stream of up to 1600 ppmv, (iii) Days 154-184 - Decrease of the packing material moisture content with a constant α-pinene concentration in the inlet gas stream of 1600 ppmv and (iv) Days 185-266 - Increase of the packing material
CAPÍTULO 5 126 moisture content with a constant α-pinene concentration in the inlet gas stream of 1600 ppmv. 5.2.2. VOCS ANALYSER MEASUREMENTS Rapid in situ measurements of the α-pinene concentrations were performed in the sampling ports using a portable VOCs analyser fitted with a 10.6 eV lamp for photoionization (MULTIRAE IR, PGM-54, RAE systems, San José, CA, USA). The sampling ports were located immediately before the inlet stream of contaminated air and at the output of each biofilter (Fig.5.1.). The instrument sensitivity was 0.1 ppmv in a concentration range from 0 ppmv to 200 ppmv. The suitability of this instrument for semi-quantitative measurements of VOCs has been reported in a number of studies (Delgado-Rodríguez et al., 2010; Karlik et al., 2002; Ojala et al., 2006). As α-pinene is the only VOC in the inlet gas stream, the response of the PID detector is directly proportional to its concentration. The instrument was calibrated with isobutylene, as recommended by the manufacturer, and therefore all the readings obtained were transformed to α-pinene concentrations using a correction factor of 0.31 (RAE_Systems, 2002). Gas samples were taken from the inlet and outlet sampling ports of both biofilters (MSW and MSW-P) by means of an internal sampling pump with a flow rate of 0.150 Lmin-1. The readings were registered during each sampling when signal stabilisation was observed, usually in a time less than 30 s. Replicate measurements were taken three times in each sampling port on a daily basis.
CAPÍTULO 5 127 This instrument does not detect water, but condensation on its UV lamp could produce biased readings (Ojala et al., 2006). For this reason, a teflon filter (0.2 µm pore size) was used as a water and particulate material trap. The removal efficiency was determined as present in the following equation (Eq.1): ( ) in outin C CC RE 100*− = (Eq. 5.1) Where, RE: removal efficiency (%); Cin : Inlet α-pinene concentration (ppm); Cout: Outlet α-pinene concentration (ppm). 5.2.3. ELECTRONIC NOSE ANALYSIS The e-nose is a commercial combination of chemical sensors and software for the recognition of the pattern signs or odour fingerprint of the gaseous samples. A PEN3 e-nose (Portable Electronic Nose, Airsense Analytics GmbH, Hagenover, Schwerin, Germany) was used to analyse the gases before and after the biofilters (at the sampling ports), similar to the measurements with the VOCs analyser. The e-nose consists of an array of 10 different metal oxide sensors (MOS) positioned inside a flow chamber (1.8 mL). The analytical system has an integrated sampling system, which by automatic control (autoranging) prevents overloading of the sensors, and also leads to a better and faster qualitative and quantitative analysis. Sample measurements was taken in replicates with ambient air being used as a blank for the experiment. The signal stabilisation took approximately 60 s for the total range of α-pinene concentrations; however the data used for the statistical analysis corresponded to the final 10 s of the sampling time. Also, a time of 60 s was used as a purging time between consecutive measurements. Principal component analysis (PCA)
CAPÍTULO 5 134 % (Fig. 5.4.). The biofilter performance however was adversely affected when the moisture content of the packing materials was set below 66 % for Bio1 and 51 % for Bio2 with removal efficiencies observed to be lower than 90 %. Figure 5.4. Biofiltration system performance during the decrease of the moisture content of the packing materials
CAPÍTULO 5 135 These results indicated that the optimum moisture content range for the packing materials and the EBRT used in this experiment was between 66 and 112 % (d.w) in the case of the Bio1 and between 51 and 100 % (d.w) for Bio2, respectively. These results are consistent with those found in previous studies (Morales et al., 2003), where the minimum moisture content suggested for the packing material was reported as 66 % (d.w). Figure 5.5. Assessment of the biofilters adaptation capacity during the replacement stage of the beds moisture content
CAPÍTULO 5 136 The four study stage was carried out between day 185 and day 266 (Fig. 5.5.). In this stage, the moisture content was increased progressively up to the optimal reported values in order to evaluate the adaptation capacity of the packing materials. The average inlet concentration of α-pinene during this stage was 11.6 ppmv. Although the moisture content of both packing materials was adjusted progressively to the optimum range found in the previous stage; only Bio1 was able to reach a removal efficiency of 100 %, whereas Bio2 removal efficiency stagnated at a value around 40 %. It is noticeable that the 100 % of removal efficiency in Bio1 was achieved after 34 days from the start of the stage, once the moisture content had been adjusted to the lower limit of the optimal range. Fig. 5.6. shows the results of the GC/MS data obtained during sampling performed on day 0. In this figure, the different peaks observed for each compound were integrated and normalised considering 100 %, the highest possible signal. These results show that even though the VOC detector signal was 0 ppm during sampling, the biofilters had different background emission that are dependent on their packing material characteristics. Another factor might be a consequence of their previous use of the packing materials in previous biofilter (López et al., 2011). Although almost all VOCs found in this sampling were previously reported as common emissions in composting processes (Delgado-Rodríguez et al., 2011), the beta-pinene emission findings may indicate a natural fingerprint by the components of the packing material used, as this peak corresponds to the biofilter with a vegetal origin. It is also possible that the terpene based compounds could have came from the ambient air. No compound related with anaerobic conditions was found, despite the high
CAPÍTULO 5 137 moisture content at the beginning of the experiment. In addition, it is possible that the acetamide, N, N-dimethyl peak may be an artefact caused by the solid phase microextraction method, as this compound is a solvent commonly used for fibers. Figure 5.6. Majority volatile organic compounds found in the biofilters by GC/MS at day 0 of the experiment To avoid this interference, the next analysis was carried out by injecting 0.5 ml of the gas (under investigation or in question) directly into the GC/MS. The chromatograms in Fig.5.7. show the analysis of the gas samples taken from all the sampling ports during day 266 of the experiment. The associated concentrations of α-pinene determined by the VOC detector from all samples were 17.6 ppmv in the inlet stream, 0.0 ppmv in the Bio1 output and 10.4 ppmv in Bio2 output, respectively. These results clearly indicate that the main and more representative compound was α-pinene. However, there were other VOCs present in the chromatograms, which could be impurities of the reactive or 0 20 40 60 80 100 120 Acetamide, N,N-dimethyl- Beta- Pinene Phenol Undecane Relative Area (%) MSW-P MSW
CAPÍTULO 5 138 natural compounds present in the surrounding air. These compounds were octadecanoic acid, the 1, 2-cyclohexanediol, other organic acids and terpenes (limonene, camphene, careen and beta-pinene). Similar to the sampling of day 0, undecane and phenol were present with maximum peaks in samples corresponding to Bio1, while beta-pinene seemed to be only prominent in Bio2. However, these compounds were also present in the input stream, which suggests that they were present in the ambient air and entered into the biofilter systems. Figure 5.7. Chromatograms obtained by GC/MS of gas samples of the inlet stream and the outputs of the biofilters at day 266 The high removal efficiency reported by the VOC´s detector was consistent with the results from the GC/MS analysis. The α-pinene signal in the chromatograms (Fig. 5.7.) disappears completely if it is compared to the input stream and the
CAPÍTULO 5 139 MSW biofilter samples. Table 5.3 shows the concentration of the VOCs found on day 266 in the sampling ports using GC/MS. In the same way as in the chromatograms, most compounds found in the GC/MS analysis are absent in Bio1 samples. Moreover, the average efficiency of Bio2 at this moment of the experiment is represented in these results. GC/MS analysis did not detect compounds related to a partial degradation of the α-pinene in the biofiltration system. Table 3. Concentration of the main volatile organic compounds found on day 266 in the different sampling ports using GC/MS (average ± standard deviationa) Inlet MSWb MSW-Pb α-Pinene μg l air - 1 1969±184 --- 1205±114 β-Pinene 16±1 --- 20.8±0.51 3-Carene 16.0±0.5 --- 3.04±0.18 D-Limonene 103±3 --- 13.7±2.1 Undecane 13.5±1.8 --- --- a Average ± standard deviation, over four samples, (d.w.) b MSW: Municipal solid waste composts; MSW-P: Municipal solid waste – Pruning residues composts Fig. 5.8. presents the PCA carried out using the response of 5 metal oxide sensors and the dilution factor of the e-nose used in the experiment on day 88 of the experiment. This statistical analysis was performed when the output concentration of α-pinene in the biofilters was the same, aiming at detecting any possible influence of the packing material on the results. In this case, the removal efficiency was determined to be 100% in Bio1 and 99.2 % in Bio2. All the PCAs performed showed good discrimination power values for the input classes chosen. The discrimination power is a measure of the overlapping of two groups of measurement points. The less they overlap, the better they can
CAPÍTULO 5 140 be distinguished. On this particular day the associated variance of the two principal components in PCA was 99.94 %, which means that the system was represented correctly by these two new uncorrelated variables. Bio1 and Bio2 groups were mainly separated along the χ axis (Function 1), which accounted for 97.88 % of the total variance. In all cases, the variance of the axis, which represents the function where the separation occurred, accounted for variances over 70 %. The results indicate that the e-nose detected a fingerprint that allowed differentiation between the outputs signals of the packing materials used in the biofiltration systems. The loading analyses carried out, showed that the differentiation of classes is caused by the signal of the broad-range (W5S) and the aromatic-aliphatic (W5C) sensors. The W5C sensor is reported to be sensitive towards alkanes, aromatic compounds and less polar compounds. Previous studies (López et al., 2011; Sironi et al., 2007) reported similar results using different composts as packing materials and composting processes.
CAPÍTULO 5 141 Figure 5.8. Principal Component analysis plots considering MSW Biofilter and MSW-P Biofilter as the classes in samples corresponding to day 88 of the experiment When the PCA was performed using the sampling ports as the classes, the differentiation between groups was also evident. Fig. 5.9. shows the graphic representation of the multivariate analysis carried out on day 223 of the experiment. The moisture content of the biofilters was 57.5 % (d.w) for Bio1 and 43.5 % (d.w) for Bio2, respectively. The removal efficiency values were similar to those found during the GC/MS analysis (98 % Bio1 and 45 % Bio2). The low removal efficiency found in Bio2 is most likely due to the low moisture content of this material. The inlet concentration of α-pinene was 7.8 ppmv and the output concentrations were 0.2 ppmv and 4.3 ppmv for Bio1 and Bio2, respectively. The PCA was performed using the e-nose signal response to the samples taken at the inlet, outputs of the biofiltration system and the surrounding air. At the start, 10 metal oxide sensors were included in the multivariate analysis. Based on the parallel loading analysis, which made it possible to check the influence of each sensor on the distribution of data within the PCA-space, specific sensors were eliminated when interference was produced in the PCA. As a result the discrimination power was improved. Similarly, special attention was given to the sensors that influenced the signal response of the samples with a greater concentration of α-pinene. In Fig. 5.9. shows this representation were the sensors included in the analysis were (i) hydrogen (W6S), which is selective to hydrogen and breath gases, (ii) the sulfur-organic (W1W) that is reported to be sensitive to sulfur compounds, but also to many terpenes and sulfur organic compounds, and (iii) the methane-aliphatic sensor (W3S) that is reported react
CAPÍTULO 5 142 to high concentrations ≥ 100 ppm and sometimes is very selective to methane and other aliphatic compounds. The PCA shows a clear differentiation of all the studied classes and seems to be a function of the α-pinene concentration in the samples. In this sense, the air samples (0 ppmv) and the input samples (7.8 ppmv) were clearly separated along the χ axis, representing 98.21 % of the total variance whereas Bio2 and the input samples are poorly separated between them. Based on the loading analysis observed the differentiations along the χ axis was mainly due to the W6S and the W1W sensors, this indicates that these sensors, related with breath gases, terpenes and smell gases are the cause of the discrimination power obtained. Furthermore, the differences between the classes along the y axis was caused by the W3S signal response, which could be due to the other compounds found in the GC/MS. Figure 5.9. Principal Component Analysis plots considering all the sampling ports and air as the classes in samples corresponding to day 223 of the experiment
CAPÍTULO 5 143 5.4. CONCLUSIONS Despite the hydrophobic properties of the contaminant (αpinene), its moisture content greatly affected the removal efficiency of a biofiltration system using different compost as packing materials. The acclimatization period necessary to reach high removal efficiency was relatively high (10 days for Bio1 and 25 days for Bio2), especially when considering that the materials had previously been used to treat gases containing α-pinene. The optimal range of moisture content for the packing materials used in this experiment was between 66 and 112 % (d.w) in the case of Bio1 and between 51 and 100 % (d.w) in Bio2. Bio2 showed higher removal efficiency during different phases of the experiment, possibly due to the natural affinity of this material with the contaminant. However, it seemed to be more sensitive to moisture changes and more susceptible to moisture losses. This was proven in the phase of the experiment where moisture content was re-adjusted to the lower limit of the optimal range. After low moisture conditions, the acclimatization period seemed to be longer for both packing materials. The GC/MS data support the results obtained with the VOC detector in terms of α-pinene removal efficiencies and did not detect any compound related to a partial degradation of the α-pinene in the biofiltration system. The e-nose could classify the signal emitted by the biofilters when the α-pinene concentration was the same, which indicates that there is a natural fingerprint in the outlet depending on the characteristics of each packing material, and also indicates that the e-nose is highly sensitive and has a high discrimination power to small odour nuances. This instrument is able to correctly differentiate different kind of gas samples based on the concentration of α-pinene in the streams. These