estudio de la gasificación de lodos de edar en lecho fluidizado:...
Post on 19-Jan-2021
7 Views
Preview:
TRANSCRIPT
2015 24
Noemiacute Gil Lalaguna
Estudio de la gasificacioacuten de lodosde EDAR en lecho fluidizado
efecto de la atmoacutesfera reactivaevaluacioacuten energeacutetica y limpieza
del gas producto
DepartamentoIngenieriacutea Quiacutemica y Tecnologiacuteas del MedioAmbiente
DirectoresSaacutenchez Cebriaacuten Joseacute Luis Murillo Esteban Mariacutea Benita
Directores
Tesis Doctoral
Autor
Repositorio de la Universidad de Zaragoza ndash Zaguan httpzaguanunizares
UNIVERSIDAD DE ZARAGOZA
Departamento
Directores
Noemiacute Gil Lalaguna
ESTUDIO DE LA GASIFICACIOacuteN DE LODOS DEEDAR EN LECHO FLUIDIZADO EFECTO DE LA
ATMOacuteSFERA REACTIVA EVALUACIOacuteNENERGEacuteTICA Y LIMPIEZA DEL GAS PRODUCTO
Directores
Ingenieriacutea Quiacutemica y Tecnologiacuteas del Medio Ambiente
Saacutenchez Cebriaacuten Joseacute LuisMurillo ESteban Mariacutea Benita
Tesis Doctoral
Autor
Repositorio de la Universidad de Zaragoza ndash Zaguan httpzaguanunizares
UNIVERSIDAD DE ZARAGOZA
Departamento
Directores
Directores
Tesis Doctoral
Autor
Repositorio de la Universidad de Zaragoza ndash Zaguan httpzaguanunizares
UNIVERSIDAD DE ZARAGOZA
ESTUDIO DE LA GASIFICACIOacuteN DE LODOS DE EDAR EN LECHO
FLUIDIZADO EFECTO DE LA ATMOacuteSFERA REACTIVA
EVALUACIOacuteN ENERGEacuteTICA Y LIMPIEZA DEL GAS PRODUCTO
TESIS DOCTORAL
presentada por
NOEMIacute GIL LALAGUNA
Noviembre de 2014 Zaragoza
A mi familia especialmente para eacutel
La presente Tesis Doctoral que lleva por tiacutetulo ldquoEstudio de la gasificacioacuten de lodos de EDAR en
lecho fluidizado Efecto de la atmoacutesfera reactiva evaluacioacuten energeacutetica y limpieza del gas
productordquo realizada por Dntildea Noemiacute Gil Lalaguna y dirigida por los doctores D Joseacute Luis
Saacutenchez Cebriaacuten y Dntildea Mariacutea Benita Murillo Esteban se presenta como compendio de las
siguientes publicaciones
i N Gil-Lalaguna JL Saacutenchez MB Murillo E Rodriacuteguez G Gea (2014) Air steam
gasification of sewage sludge in a fluidized bed Influence of some operating conditions
Chemical Engineering Journal 248 373-382
ii N Gil-Lalaguna JL Saacutenchez MB Murillo V Ruiz G Gea (2014) Air steam gasification of
char derived from sewage sludge pyrolysis Comparison with the gasification of sewage
sludge Fuel 129 147-155
iii N Gil-Lalaguna JL Saacutenchez MB Murillo M Atienza-Martiacutenez G Gea (2014) Energetic
assessment of air-steam gasification of sewage sludge and of the integration of sewage
sludge pyrolysis and air-steam gasification of char Energy 76 652-662
iv N Gil-Lalaguna JL Saacutenchez MB Murillo G Gea (2015) Use of sewage sludge
combustion ash and gasification ash for high-temperature desulphurization of different gas
streams Fuel 141 99-108
Lo que se hace constar en cumplimiento del Reglamento sobre Tesis Doctorales de la
Universidad de Zaragoza (artiacuteculos 19 y 20) aprobado seguacuten el Acuerdo de 20122013 del
Consejo de Gobierno de la Universidad de Zaragoza
D Joseacute Luis Saacutenchez Cebriaacuten y Dntildea Mariacutea Benita Murillo Esteban
ambos Profesores Titulares de Universidad en el Departamento de Ingenieriacutea Quiacutemica y Tecnologiacuteas del Medio Ambiente de la Universidad de Zaragoza y miembros del Grupo de Procesos Termoquiacutemicos del Instituto Universitario de Investigacioacuten en Ingenieriacutea de Aragoacuten (I3A)
INFORMAN que
La presente memoria titulada
ldquoEstudio de la gasificacioacuten de lodos de EDAR en lecho fluidizado Efecto de la atmoacutesfera reactiva evaluacioacuten energeacutetica y limpieza del gas productordquo
ha sido realizada por Dntildea Noemiacute Gil Lalaguna bajo nuestra direccioacuten en el Departamento de Ingenieriacutea Quiacutemica y Tecnologiacuteas del Medio Ambiente y
AUTORIZAN su presentacioacuten como compendio de publicaciones
Y para que asiacute conste firmamos este informe en Zaragoza a 24 de Noviembre de 2014
Fdo Prof Dr Joseacute Luis Saacutenchez Cebriaacuten Fdo Prof Dra Mariacutea Benita Murillo Esteban
AGRADECIMIENTOS
El trabajo reflejado en esta Tesis Doctoral ha sido realizado en el Grupo de Procesos
Termoquiacutemicos (GPT) del Instituto de Investigacioacuten en Ingenieriacutea de Aragoacuten (I3A) gracias a una
beca predoctoral que me concedioacute el Ministerio de Educacioacuten entre los antildeos 2010 y 2014
(ayuda de posgrado para la formacioacuten de profesorado universitario) Por lo tanto el primer
agradecimiento debe ser para ambas instituciones por el apoyo teacutecnico y econoacutemico
prestado De forma particular me gustariacutea agradecer tambieacuten al director del GPT Rafael
Bilbao por la oportunidad que me brindoacute en el inicio de esta etapa
Si hoy estoy escribiendo estas liacuteneas de agradecimientos es en buena parte gracias a mis
directores la Dra Mariacutea Benita Murillo y el Dr Joseacute Luis Saacutenchez Gracias Mari Beni por
animarme a unirme al grupo hace ya maacutes de 5 antildeos para hacer el Proyecto Fin de Carrera con
vosotros Alliacute empezoacute todo esto Gracias Joseacute Luis por ldquoanimarmerdquo a terminar con esto Gracias
a los dos por vuestros consejos orientacioacuten y apoyo en muchos momentos duros no soacutelo
relacionados con la investigacioacuten
Gracias a mis compantildeeros del GPT tanto a los que seguiacutes por aquiacute como a los que estaacuteis fuera
Gloria gracias por tus consejos y por estar siempre disponible Isabel Javi Fernando Alberto
Ester y Gorka gracias por ensentildearnos y aconsejarnos a los que llegamos por aquiacute un poco
despueacutes que vosotros Ana Javi Mariacutea Lorena Luciacutea Manu Mariacutea y Violeta gracias por las
risas y los buenos momentos en la nave en el despacho en el cafeacute en la comidahellip iexclQue
hubiese sido de miacute algunos diacuteas sin ellos Olga Guille y Beatriz gracias por vuestra ayuda
siempre que os la he pedido Gracias a Eva Victor Ana y David por vuestro trabajo y buena
disposicioacuten en el laboratorio
De puertas para afuera de la universidad tambieacuten tengo mucho que agradecer En primer lugar
a mi padre porque se lo debo todo A mi madre y a mi hermana por su paciencia y por estar
siempre alliacute en lo bueno y en lo malo A mis amigas siempre con palabras de aacutenimo para todo
Y como no a Oacutescar por ser coacutemo es y por haber aguantado todo lo que al final supone hacer
una Tesis
IacuteNDICE
1 INTRODUCCIOacuteN GENERAL Y OBJETIVOS 1
2 ANTECEDENTES 7
21 Lodos de EDAR Problemaacutetica y viacuteas de gestioacuten 7
22 Aspectos generales de la gasificacioacuten 11
23 Influencia de las condiciones de operacioacuten en el proceso de gasificacioacuten 14
24 Limpieza del gas producto de la gasificacioacuten 19
3 MATERIALES Y MEacuteTODOS 24
31 Materiales 24
311 Materia prima para la gasificacioacuten lodos de EDAR y char de piroacutelisis del lodo 24
312 Cenizas de combustioacuten y gasificacioacuten de lodos de EDAR 25
313 Catalizadores de niacutequel 28
32 Instalaciones y procedimiento experimental 30
321 Sistema experimental para la gasificacioacuten 30
322 Sistema experimental para los ensayos de retencioacuten de H2S 32
323 Sistema experimental para los ensayos de actividad de los catalizadores 35
33 Condiciones de operacioacuten y disentildeo de experimentos 36
331 Experimentos de gasificacioacuten 36
332 Experimentos de desulfuracioacuten 38
333 Ensayos de actividad de los catalizadores de niacutequel 40
4 RESULTADOS Y DISCUSIOacuteN 42
41 Gasificacioacuten de lodo y de char 42
42 Evaluacioacuten energeacutetica 62
43 Eliminacioacuten de H2S de diferentes gases con cenizas de lodo 77
44 Estudio de la actividad de catalizadores de niacutequel en el reformado de alquitraacuten 89
5 CONCLUSIONES Y TRABAJOS FUTUROS 100
6 REFERENCIAS BIBLIOGRAacuteFICAS 105
7 APEacuteNDICE COPIA DE LOS TRABAJOS PUBLICADOS 115
Introduccioacuten general y objetivos 1
1 INTRODUCCIOacuteN GENERAL Y OBJETIVOS
La presente Tesis Doctoral ha sido desarrollada en el Grupo de Procesos Termoquiacutemicos
(GPT) perteneciente al Instituto de Investigacioacuten en Ingenieriacutea de Aragoacuten (I3A) de la
Universidad de Zaragoza Entre las principales liacuteneas de investigacioacuten desarrolladas por el GPT
se puede destacar el tratamiento termoquiacutemico de biomasa y residuos orgaacutenicos mediante
procesos de gasificacioacuten y piroacutelisis la produccioacuten de biodiesel y mejora de sus propiedades la
eliminacioacuten de contaminantes de gases de combustioacuten (NOx y holliacuten) y la produccioacuten de
hidroacutegeno a partir del reformado cataliacutetico de corrientes acuosas
El presente trabajo se engloba dentro del campo de la valorizacioacuten energeacutetica de
residuos y maacutes concretamente el proceso estudiado es la gasificacioacuten de lodos de
estaciones depuradoras de aguas residuales urbanas (EDAR) Los lodos de EDAR son el
subproducto generado en el proceso de depuracioacuten de las aguas residuales provenientes de
zonas urbanas La cantidad generada de este residuo ha aumentado mucho en los uacuteltimos
antildeos como consecuencia de la implementacioacuten de la legislacioacuten europea referente al
tratamiento de aguas residuales urbanas (Directiva 91271CEE) En la Unioacuten Europea se
estaacuten produciendo anualmente maacutes de 10 millones de toneladas de lodo seco (Kelessidis y
Stasinakis 2012) Por esta razoacuten la correcta gestioacuten de los lodos de forma econoacutemica y sin
poner en riesgo la salud puacuteblica y el medio ambiente se ha convertido en un reto importante
en el aacutembito del tratamiento de las aguas residuales
Dado el contenido en materia orgaacutenica de los lodos uno de los procesos que puede
plantearse para su aprovechamiento energeacutetico es la gasificacioacuten proceso que centra el
estudio desarrollado en esta Tesis La gasificacioacuten puede definirse como la conversioacuten teacutermica
de un material carbonoso en una atmoacutesfera netamente reductora generando un gas
combustible y un residuo soacutelido El producto de intereacutes de la gasificacioacuten es el gas compuesto
principalmente de CO CO2 H2 vapor de agua CH4 y otros hidrocarburos ligeros y N2 (en el
caso de gasificar con aire) La proporcioacuten de estos gases variacutea en funcioacuten de la composicioacuten de
la materia prima y las condiciones del proceso El gas producto de la gasificacioacuten ofrece varias
opciones para su aprovechamiento desde su uso como combustible en motores de
combustioacuten interna o en turbinas de gas para la generacioacuten de electricidad en ciclos
combinados hasta su uso como materia prima en la obtencioacuten de productos quiacutemicos como
metanol amoniacuteaco o liacutequidos Fischer-Tropsch (Wender 1996) Ademaacutes del producto gaseoso
y del residuo soacutelido durante la gasificacioacuten se genera tambieacuten una mezcla de vapores
orgaacutenicos faacutecilmente condensables denominada alquitraacuten que abandona el gasificador junto
2 Introduccioacuten general y objetivos
con el gas La formacioacuten de alquitraacuten es una de las principales limitaciones para la implantacioacuten
de los procesos de gasificacioacuten ya que su presencia en el gas conlleva problemas
operacionales debido a su facilidad para condensar formar aerosoles y polimerizar dando
lugar a estructuras maacutes complejas y provocando problemas de ensuciamiento y taponamiento
en tuberiacuteas y equipos para el aprovechamiento del gas como motores y turbinas
Los inicios de la investigacioacuten relacionada con la gasificacioacuten de lodos de EDAR en el
Grupo de Procesos Termoquiacutemicos se remontan a la deacutecada de los antildeos 90 cuando se
realizaron las primeras pruebas a escala de laboratorio en colaboracioacuten con la empresa
Cadagua SA En base a los resultados experimentales obtenidos durante los antildeos 2001-2003
el GPT trabajoacute junto con el departamento de I+D de Cadagua SA en el disentildeo y puesta a punto
de una planta piloto de lecho fluidizado de 100 kgmiddoth-1 para la gasificacioacuten de lodos de EDAR En
el antildeo 2003 se comenzoacute con la experimentacioacuten en dicha planta pero los problemas asociados
con el trabajo a gran escala pronto plantearon la necesidad de trasladar de nuevo el estudio
del proceso a escala de laboratorio para conocer mejor la influencia de las condiciones de
operacioacuten Desde entonces la gasificacioacuten y piroacutelisis de lodos de EDAR ha sido una importante
liacutenea de trabajo en el GPT contando con el apoyo financiero de los Ministerios de Educacioacuten y
Ciencia de Ciencia e Innovacioacuten y de Economiacutea y Competitividad en sucesivas convocatorias
de proyectos (CT2004-05528 CT2007-66885 CT2010-20137 y CT2013-47260)
De forma maacutes concreta el desarrollo de esta Tesis ha contado con el apoyo financiero del
Ministerio de Economiacutea y Competitividad gracias al proyecto ldquoValorizacioacuten de lodos de EDAR
mediante un proceso de piroacutelisis estudio y mejora de la aplicabilidad de sus productos
(CTQ2010-20137)rdquo asiacute como del Ministerio de Educacioacuten a traveacutes de una ayuda de posgrado
para la formacioacuten de profesorado universitario (beca FPU referencia AP2009-3446) concedida
a la doctoranda durante los cuatro uacuteltimos antildeos para la realizacioacuten de la Tesis Doctoral
La mayoriacutea de los estudios que se encuentran en la bibliografiacutea sobre gasificacioacuten de lodos
de EDAR (Adegoroye y cols 2004 Dogru y cols 2002 Midilli y cols 2001 Petersen y
Werther 2005 Tae-Young y cols 2009) incluyendo el trabajo previo desarrollado en el GPT
(Aznar y cols 2007 Aznar y cols 2008 Manyagrave y cols 2005 Manyagrave y cols 2006) utilizan aire
para gasificar lodo seco Sin embargo teniendo en cuenta que el contenido de humedad del
lodo antes del secado teacutermico puede superar el 70 tras su estabilizacioacuten y deshidratacioacuten
mecaacutenica mediante filtros prensa y centrifugacioacuten (Manara y Zabaniotou 2012) la gasificacioacuten
del lodo huacutemedo podriacutea ser una opcioacuten interesante para su aprovechamiento Los estudios
publicados acerca de la gasificacioacuten de lodo huacutemedo (o gasificacioacuten de lodo seco con vapor
para simular el proceso) no son muy numerosos (Domiacutenguez y cols 2006 Nipattummakul y
Introduccioacuten general y objetivos 3
cols 2010 Xie y cols 2010 Zhang y cols 2011) El principal inconveniente de la gasificacioacuten
con vapor de agua es que es un proceso endoteacutermico por lo que requiere un aporte continuo
de energiacutea Esta energiacutea puede obtenerse mediante la adicioacuten de aire u oxiacutegeno al medio de
gasificacioacuten gracias a la combustioacuten parcial de parte de la materia orgaacutenica La escasez de
trabajos publicados acerca de la gasificacioacuten de lodos con mezclas de vapor de agua y aire
como agente gasificante motivoacute el estudio realizado en la primera parte de la Tesis Este
estudio pretende profundizar en el conocimiento del efecto de la atmoacutesfera reactiva en la
distribucioacuten de productos y en la calidad del gas Puesto que la materia prima disponible era
un lodo digerido anaeroacutebicamente y secado teacutermicamente la adicioacuten de vapor de agua al
medio de reaccioacuten permitioacute simular el proceso de gasificacioacuten de un lodo huacutemedo Los
resultados de este estudio se recogen en el primer artiacuteculo del compendio de publicaciones
que componen la Tesis
i N Gil-Lalaguna JL Saacutenchez MB Murillo E Rodriacuteguez G Gea (2014) ldquoAir steam
gasification of sewage sludge in a fluidized bed Influence of some operating conditionsrdquo
Chemical Engineering Journal 248 373-382
Como ya se ha comentado anteriormente la presencia de alquitraacuten en el gas de
gasificacioacuten supone un importante inconveniente para el aprovechamiento del gas En
comparacioacuten con la biomasa original y desde el punto de vista de la formacioacuten de alquitraacuten el
soacutelido resultante del proceso de piroacutelisis de la misma (char) puede ser una materia prima
preferible para la gasificacioacuten ya que gran parte de la materia volaacutetil responsable de la
formacioacuten del alquitraacuten se elimina durante el tratamiento de piroacutelisis La piroacutelisis (o
descomposicioacuten teacutermica en atmoacutesfera inerte) de lodos de EDAR ha sido ampliamente
estuadiada en el GPT durante los uacuteltimos antildeos (Fonts y cols 2008 Fonts y cols 2009 Gil-
Lalaguna y cols 2010) Aunque la piroacutelisis raacutepida estaacute orientada a maximizar la produccioacuten de
liacutequido la fraccioacuten mayoritaria en la piroacutelisis de lodo es el soacutelido (en torno a un 50 en masa)
La fraccioacuten orgaacutenica de este char le confiere cierto valor energeacutetico que puede ser
aprovechado mediante procesos de combustioacuten o gasificacioacuten Esta idea es la base de otro de
los estudios desarrollados en la Tesis consistente en la gasificacioacuten del char obtenido en la
piroacutelisis de lodos de EDAR Los resultados de este estudio y su comparacioacuten con los datos
obtenidos en la gasificacioacuten directa del lodo se detallan en el segundo artiacuteculo del compendio
de publicaciones de la Tesis
4 Introduccioacuten general y objetivos
ii N Gil-Lalaguna JL Saacutenchez MB Murillo V Ruiz G Gea (2014) Air steam gasification
of char derived from sewage sludge pyrolysis Comparison with the gasification of
sewage sludge Fuel 129 147-155
Ademaacutes de la viabilidad operacional de los procesos termoquiacutemicos el estudio de los
mismos desde un punto de vista energeacutetico es un aspecto clave de cara a su posible desarrollo
industrial Por lo tanto dado el caraacutecter endoteacutermico de las reacciones de gasificacioacuten con
vapor de agua se consideroacute interesante realizar una evaluacioacuten energeacutetica para determinar si
la energiacutea disponible en los productos de ambos procesos (gasificacioacuten directa del lodo y
combinacioacuten de la piroacutelisis de lodo y gasificacioacuten del char) es suficiente para cubrir el coste
energeacutetico de dichos procesos asiacute como del secado teacutermico del lodo La etapa de secado
teacutermico permite reducir el volumen de residuo facilitando su manipulacioacuten antes del
tratamiento termoquiacutemico Sin embargo el secado del lodo conlleva un gran consumo
energeacutetico que encarece la gestioacuten del residuo El aporte de esta energiacutea mediante los propios
productos de la gasificacioacuten y piroacutelisis del lodo seriacutea la opcioacuten maacutes econoacutemica Los resultados
de este estudio energeacutetico se han publicado en otro de los artiacuteculos que forma parte del
compendio de publicaciones de la Tesis
iii N Gil-Lalaguna JL Saacutenchez MB Murillo M Atienza-Martiacutenez G Gea (2014)
Energetic assessment of air-steam gasification of sewage sludge and of the integration
of sewage sludge pyrolysis and air-steam gasification of char Energy 76 652-662
Los siguientes estudios desarrollados en la Tesis se centran en la limpieza del gas producto
de la gasificacioacuten de lodo Ademaacutes de alquitraacuten este gas contiene otras impurezas que tienen
su origen en la propia composicioacuten del lodo Es el caso del sulfuro de hidroacutegeno (H2S) formado
durante la gasificacioacuten del lodo como consecuencia de la presencia de compuestos con azufre
La presencia de H2S en el gas de gasificacioacuten conlleva problemas ambientales y operacionales
causando la corrosioacuten de tuberiacuteas motores y turbinas asiacute como el envenenamiento de los
catalizadores maacutes comuacutenmente utilizados para el craqueo de los alquitranes por lo general
basados en niacutequel (Hepola y Simell 1997a) Existen diferentes procesos para la eliminacioacuten de
H2S de corrientes gaseosas tanto a baja como a alta temperatura Los procesos de
desulfuracioacuten a alta temperatura se basan en la reaccioacuten quiacutemica de oacutexidos metaacutelicos con el
H2S para formar sulfuros metaacutelicos que quedan retenidos en forma soacutelida Las cenizas
resultantes del tratamiento termoquiacutemico de la biomasa o de residuos orgaacutenicos estaacuten
compuestas por diversos metales y oacutexidos metaacutelicos por lo que su uso para la desulfuracioacuten
de gases puede ser una opcioacuten interesante debido a su bajo coste Es el caso de las cenizas
Introduccioacuten general y objetivos 5
obtenidas en el tratamiento termoquiacutemico del lodo cuya aplicacioacuten para la eliminacioacuten del H2S
generado en el propio proceso supondriacutea una ventaja desde el punto de vista del
aprovechamiento integral de los subproductos Por lo tanto el siguiente estudio de la Tesis se
centroacute en evaluar la capacidad de retencioacuten de azufre de las cenizas obtenidas tanto en la
gasificacioacuten como en la combustioacuten del lodo de EDAR utilizando para ello diversos gases
sinteacuteticos con el fin de evaluar el efecto de la atmoacutesfera reactiva Este estudio se encuentra
publicado en otro de los artiacuteculos de la Tesis
iv N Gil-Lalaguna JL Saacutenchez MB Murillo G Gea (2015) Use of sewage sludge
combustion ash and gasification ash for high-temperature desulphurization of different
gas streams Fuel 141 99-108
La eliminacioacuten del alquitraacuten presente en el gas de gasificacioacuten sigue centrando muchos
estudios en el campo de la gasificacioacuten de biomasa ya que hasta la fecha este problema no ha
sido resuelto de forma satisfactoria a escala industrial El reformado cataliacutetico de los
alquitranes con catalizadores de niacutequel parece ser una de las viacuteas maacutes eficaces (Anis y Zainal
2011 De Lasa y cols 2011) Sin embargo la operacioacuten de estos catalizadores durante largos
periacuteodos de tiempo conlleva ciertos problemas de peacuterdida de actividad por fenoacutemenos de
sinterizacioacuten formacioacuten de carbono sobre los centros activos o envenenamiento con azufre
Por lo tanto el disentildeo de catalizadores de niacutequel resistentes a estos fenoacutemenos de
desactivacioacuten representa todaviacutea un gran desafiacuteo En este contexto se desarrolloacute el uacuteltimo
estudio de la Tesis llevado a cabo durante una estancia de investigacioacuten en el VTT Technical
Research Centre of Finland Dado que el gas obtenido en la gasificacioacuten de lodos de EDAR
presenta un importante contenido de H2S el objetivo de este estudio fue la evaluacioacuten de la
actividad y estabilidad de diversos catalizadores de niacutequel preparados y modificados con varios
promotores El artiacuteculo correspondiente a este estudio se encuentra en fase de revisioacuten para
su publicacioacuten en la revista Fuel ldquoCatalytic activity of nickel-alumina catalysts modified with
iron manganese calcium and copper for tar reforming under a H2S-containing atmosphererdquo
En resumen los artiacuteculos que componen la presente Tesis Doctoral presentan una clara
unidad temaacutetica el estudio de la gasificacioacuten de lodos de EDAR y de diversos procesos para la
limpieza y mejora de la calidad del gas producto
6 Introduccioacuten general y objetivos
Objetivos
El objetivo principal de la Tesis es profundizar en el estudio de la gasificacioacuten de lodos de
EDAR para mejorar las propiedades del gas producto mediante la optimizacioacuten de las
condiciones de operacioacuten y la aplicacioacuten de distintos tratamientos secundarios de limpieza del
gas La consecucioacuten de este objetivo global ha conllevado la realizacioacuten de diversas tareas
bull Revisioacuten y actualizacioacuten de la bibliografiacutea en el campo del aprovechamiento energeacutetico
de lodos de EDAR la gasificacioacuten de biomasa y la limpieza del gas producto
bull Estudio experimental de la influencia de algunas condiciones de operacioacuten en la
distribucioacuten de productos y calidad del gas obtenido en la gasificacioacuten de lodos de EDAR
con mezclas de aire y vapor de agua
bull Estudio experimental de la gasificacioacuten del char obtenido en la piroacutelisis del lodo como
alternativa para reducir la presencia de alquitraacuten en el gas Comparacioacuten con los
resultados obtenidos en la gasificacioacuten directa del lodo
bull Evaluacioacuten energeacutetica de las etapas de gasificacioacuten de lodo y de char incluyendo
tambieacuten el anaacutelisis del proceso de piroacutelisis en el que se genera el char gasificado y de la
etapa previa de secado teacutermico del lodo
bull Estudio experimental de la eliminacioacuten de H2S de diferentes gases sinteacuteticos utilizando
las propias cenizas obtenidas en la combustioacuten y gasificacioacuten de lodos de EDAR
bull Estudio experimental de la actividad y estabilidad de varios catalizadores de niacutequel
modificados con distintos promotores para el reformado de compuestos modelo de
alquitraacuten en presencia de H2S
Antecedentes 7
2 ANTECEDENTES
21 Lodos de EDAR Problemaacutetica y viacuteas de gestioacuten
Los lodos de Estaciones Depuradoras de Aguas Residuales (EDAR) son el subproducto
derivado del proceso de depuracioacuten de las aguas residuales provenientes de zonas urbanas o
de vertidos industriales de composicioacuten similar a las de eacutestas Estos lodos son el resultado de la
acumulacioacuten tanto de microorganismos derivados del tratamiento bioloacutegico de las aguas como
de la materia orgaacutenica y mineral que se encontraba disuelta o suspendida en el agua y que no
ha sido degradada durante el proceso Los lodos de EDAR son considerados como un residuo
no peligroso en el Cataacutelogo Europeo de Residuos (Decisioacuten 2001118CE)
En la Figura 21 se muestra un diagrama de las etapas que habitualmente forman parte del
proceso de depuracioacuten de aguas residuales
Figura 21 Diagrama del proceso de depuracioacuten de aguas residuales (Kiely 1999)
Entrada de agua residual
Residuos soacutelidos a vertedero o incineracioacuten
Decantador primario
Tratamiento bioloacutegico
Lodo primario
Decantador secundario
Filtracioacuten de arena
Lodo activo en exceso (lodo secundario)
Retorno de lodos
Recirculacioacuten de lodos activos
Tratamiento de lodos espesado acondicionamiento deshidratacioacuten estabilizacioacuten
Evacuacioacuten de lodos
Agua efluente al medio receptor
Retorno de agua de lavado
Pretratamiento desbaste desarenado
desengrasado
8 Antecedentes
Tras las etapas de desbaste desarenado y desengrasado en la decantacioacuten primaria se
consigue una separacioacuten de los soacutelidos maacutes finos suspendidos en el agua como consecuencia
de la diferencia de densidades Despueacutes en el tratamiento secundario o tratamiento bioloacutegico
la materia orgaacutenica biodegradable del agua es metabolizada mediante diversos tipos de
microorganismos Este proceso se suele llevar a cabo en un tanque aireado donde las aguas
residuales y los microorganismos permanecen en contacto (lodo activo) y despueacutes los
coaacutegulos microbianos sedimentan en el tanque de decantacioacuten secundaria
El lodo resultante de las etapas de decantacioacuten primaria y secundaria es un residuo
liacutequido biodegradable cuyo contenido de agua puede alcanzar el 95 ocupando por tanto un
gran volumen Por lo tanto antes de su evacuacioacuten y disposicioacuten final este residuo debe
someterse a ciertos tratamientos para facilitar su manejo y evitar problemas sanitarios y
medioambientales Con el espesado (por gravedad o flotacioacuten) y deshidratacioacuten mecaacutenica del
lodo (centriacutefugas filtros prensa filtros banda) se consigue reducir el volumen de residuo
mediante la eliminacioacuten parcial del agua hasta aproximadamente un contenido de humedad
del 70 (Manara y Zabaniotou 2012) Por otro lado los tratamientos de estabilizacioacuten del
fango permiten reducir la presencia de agentes patoacutegenos asiacute como la capacidad de
putrefaccioacuten del lodo y los desagradables olores asociados a ella La digestioacuten anaerobia es
uno de los meacutetodos de estabilizacioacuten maacutes extendidos en las plantas de tratamiento de gran
capacidad pero existen otras opciones como la digestioacuten aerobia el compostaje o la
estabilizacioacuten quiacutemica mediante la adicioacuten de cal Por uacuteltimo el secado teacutermico del lodo como
etapa final del tratamiento da lugar a un residuo soacutelido con un contenido de humedad inferior
al 10 A pesar del coste adicional que conlleva la implantacioacuten de plantas para el secado
teacutermico de los lodos como etapa previa a la incineracioacuten de los mismos ha cobrado especial
importancia en los uacuteltimos antildeos (Kelessidis y Stasinakis 2012)
La composicioacuten del lodo generado en las estaciones depuradoras estaacute fuertemente
condicionada por la carga contaminante del agua residual y por el tipo de tratamiento aplicado
en el proceso En teacuterminos generales estos lodos estaacuten formados por (i) agua que puede
suponer desde un pequentildeo porcentaje hasta un 95 del lodo (ii) materia orgaacutenica no toacutexica
(aproximadamente un 60 en base seca) en la que se incluyen constituyentes bioloacutegicos
como aacutecidos nucleicos proteiacutenas carbohidratos y liacutepidos y materia orgaacutenica no digerida en el
proceso como celulosa (iii) materia inorgaacutenica silicatos aluminatos compuestos de calcio y
de magnesio etc (iv) elementos nutrientes nitroacutegeno foacutesforo y potasio (v) pequentildeas
concentraciones de elementos contaminantes como metales pesados (zinc cromo plomo
cobre niacutequel mercurio etc) compuestos orgaacutenicos persistentes (pesticidas disolventes
Antecedentes 9
industriales colorantes plastificantes agentes tensoactivos etc) y agentes patoacutegenos que
suponen un riesgo medioambiental y sanitario en el caso de una mala gestioacuten de los lodos
(Manara y Zabaniotou 2012 Rulkens 2008)
En cumplimiento con la legislacioacuten europea los estados miembro de la Unioacuten Europea
estaacuten obligados a recoger y tratar las aguas residuales provenientes de nuacutecleos urbanos con
una poblacioacuten superior a 2000 habitantes equivalentes Asiacute lo especifica la Directiva
91271CEE sobre el tratamiento de las aguas residuales urbanas Como resultado del
cumplimiento de esta normativa la produccioacuten de lodos de EDAR ha aumentado
considerablemente en las dos uacuteltimas deacutecadas debido a la implantacioacuten de nuevas estaciones
depuradoras y al desarrollo de tecnologiacuteas de depuracioacuten maacutes eficaces La produccioacuten anual de
lodos en la Unioacuten Europea praacutecticamente se duplicoacute en el periacuteodo de tiempo de 1992 a 2005
pasando de 65 a 109 millones de toneladas anuales de materia seca Las predicciones
apuntan a que en el antildeo 2020 se superaraacuten las 13 millones de toneladas anuales En el caso
particular de Espantildea los uacuteltimos datos disponibles del Registro Nacional de Lodos del
Ministerio de Agricultura Alimentacioacuten y Medio Ambiente (antildeo 2009) lo situacutean entre los cinco
paiacuteses europeos con mayor produccioacuten de lodos generando alrededor de 12 millones de
toneladas anuales de materia seca Desde el antildeo 2000 la produccioacuten de lodos en Espantildea se ha
incrementado en un 41 siguiendo la misma tendencia que los datos europeos
Aunque el residuo final representa soacutelo un pequentildeo porcentaje del volumen total de agua
tratado en una estacioacuten depuradora su acondicionamiento y tratamiento supone maacutes del 50
de los costes de operacioacuten de la planta (Spinosa y cols 2011) Esto unido al fuerte incremento
de su produccioacuten hace que la gestioacuten y eliminacioacuten de los lodos de forma econoacutemica y segura
para la salud puacuteblica y el medio ambiente sea un reto importante en la actualidad
La poliacutetica en materia de gestioacuten de lodos de EDAR estaacute condicionada en gran medida por
factores geograacuteficos culturales econoacutemicos etc y la flexibilidad entre las distintas viacuteas de
gestioacuten variacutea de un paiacutes a otro Sin embargo de forma general dado que los lodos de EDAR
son considerados como un residuo no peligroso (coacutedigo CER 190805) les es de aplicacioacuten el
principio de jerarquiacutea establecido en la normativa vigente de gestioacuten de residuos (Directiva
200898CE) La prevencioacuten reutilizacioacuten reciclado y otras formas de valorizacioacuten (incluyendo
la recuperacioacuten energeacutetica) son por este orden prioritarios frente a la eliminacioacuten del residuo
en vertedero
Actualmente las opciones maacutes comunes para la gestioacuten de lodos de EDAR en la Unioacuten
Europea incluyen su reutilizacioacuten en la agricultura y en la restauracioacuten de terrenos la
10 Antecedentes
incineracioacuten y el depoacutesito en vertedero (Kelessidis y Stasinakis 2012 Manara y Zabaniotou
2012) Dado su contenido en materia orgaacutenica y nutrientes (nitroacutegeno y foacutesforo) la aplicacioacuten
de los lodos en suelos agriacutecolas ya sea de forma directa o tras una etapa de compostaje es la
opcioacuten maacutes utilizada a nivel europeo (53 del lodo producido seguacuten datos del antildeo 2005)
seguida de la incineracioacuten (19) y el depoacutesito en vertedero (17) El porcentaje restante
incluye meacutetodos como el almacenamiento temporal o la recuperacioacuten de terrenos (Kelessidis y
Stasinakis 2012)
Aunque los lodos de EDAR representan una fuente natural de nutrientes y materia
orgaacutenica para su aplicacioacuten en suelos agriacutecolas la presencia de sustancias nocivas (metales
pesados toxinas y agentes patoacutegenos) ha suscitado cierta controversia sobre la reutilizacioacuten
agriacutecola del lodo debido a los posibles efectos adversos de estos contaminantes en la cadena
alimentaria La Directiva 86278CEE relativa a la proteccioacuten del medio ambiente en la
utilizacioacuten de los lodos con fines agriacutecolas regula esta praacutectica haciendo indispensable el
control de las dosis de aplicacioacuten del lodo en funcioacuten de sus caracteriacutesticas y las del suelo
donde va a ser aplicado y estableciendo valores liacutemite para la concentracioacuten de metales
pesados tanto en el lodo como en el suelo Algunos paiacuteses europeos han adoptado liacutemites
mucho maacutes restrictivos que los establecidos en la citada directiva europea asiacute como valores
liacutemite para lo concentracioacuten de contaminantes orgaacutenicos y agentes patoacutegenos (Kelessidis y
Stasinakis 2012) El depoacutesito en vertedero ha sido otra de las viacuteas habituales para la
eliminacioacuten de los lodos de depuradora Sin embargo debido a la prohibicioacuten del vertido de
desechos liacutequidos orgaacutenicos asiacute como a las restricciones establecidas para el depoacutesito de
residuos soacutelidos biodegradables (Directiva 199931CEE) la eliminacioacuten de los lodos en
vertederos mostroacute un continuo y significativo retroceso entre 1992 y 2005 disminuyendo del
33 al 17 de los lodos producidos en la Unioacuten Europea Por otro lado el porcentaje de lodos
incinerados se duplicoacute en dicho periacuteodo pasando del 11 al 21 (Kelessidis y Stasinakis 2012)
La incineracioacuten de los lodos como la de cualquier residuo tambieacuten se encuentra sometida a la
legislacioacuten europea (Directiva 201075UE)
En Espantildea el Plan Nacional Integrado de Residuos (PNIR 2008-2015) establece las
medidas a tomar de cara a la gestioacuten de los lodos de EDAR Entre estas medidas se incluye la
reduccioacuten de la cantidad de fangos destinada a vertedero (12 como maacuteximo en 2015) y se
promueve la reutilizacioacuten del lodo mediante su aplicacioacuten en suelos agriacutecolas fijando un
objetivo miacutenimo del 67 de los lodos para el antildeo 2015 Seguacuten los datos del Registro Nacional
de Lodos estos objetivos han sido claramente sobrepasados en los uacuteltimos antildeos destinando
Antecedentes 11
en torno a un 83 de los lodos a fines agriacutecolas un 8 a vertedero y un 5 a incineracioacuten
(datos del antildeo 2009)
Dado que el uso agriacutecola del lodo parece estar cada vez maacutes cuestionado su valorizacioacuten
energeacutetica mediante procesos termoquiacutemicos (combustioacuten piroacutelisis y gasificacioacuten) ofrece una
alternativa interesante para su gestioacuten convirtiendo la parte orgaacutenica del residuo en energiacutea
uacutetil yo productos de valor antildeadido y quedando soacutelo la fraccioacuten mineral para su disposicioacuten
final Desde el punto de vista del contenido energeacutetico el poder caloriacutefico del lodo seco (12-20
MJmiddotkg-1) es comparable por ejemplo al del lignito (15-27 MJmiddotkg-1) (Manara y Zabaniotou
2012) La gasificacioacuten de lodos de depuradora es el proceso en el que se centra la presente
Tesis Doctoral
22 Aspectos generales de la gasificacioacuten
La gasificacioacuten es un proceso termoquiacutemico en el que un sustrato carbonoso se
transforma en un gas combustible en presencia de un agente gasificante (aire oxiacutegeno vapor
de agua o dioacutexido de carbono) en una atmoacutesfera de reaccioacuten netamente reductora El
producto de intereacutes de la gasificacioacuten es el gas compuesto principalmente de CO CO2 H2
vapor de agua CH4 y otros hidrocarburos ligeros y N2 (en el caso de gasificar con aire) La
proporcioacuten de estos gases variacutea en funcioacuten de la composicioacuten de la materia prima y las
condiciones del proceso El gas producto de la gasificacioacuten ofrece varias opciones para su
aprovechamiento desde su uso como combustible en motores de combustioacuten interna o en
turbinas de gas para la generacioacuten de electricidad hasta su uso como materia prima en la
obtencioacuten de productos quiacutemicos como metanol amoniacuteaco o liacutequidos Fischer-Tropsch
(Wender 1996) Ademaacutes del gas en el proceso de gasificacioacuten se obtiene tambieacuten un residuo
soacutelido con cierto contenido en carbono debido a la incompleta conversioacuten de la materia
orgaacutenica inicial y una mezcla de vapores orgaacutenicos aromaacuteticos y poliaroacutematicos denominada
alquitraacuten
El proceso de gasificacioacuten transcurre a traveacutes de varias etapas (Antal y cols 1979) En
primer lugar se produce el secado del soacutelido y el desprendimiento de la materia volaacutetil
(piroacutelisis) El soacutelido resultante rico en carbono fijo se gasifica mediante su reaccioacuten con O2
CO2 H2 o H2O A la vez se producen tambieacuten otras reacciones secundarias entre los gases y los
productos volaacutetiles generados dando lugar al producto final Las reacciones de gasificacioacuten del
soacutelido son lentas en comparacioacuten con la liberacioacuten de la materia volaacutetil y las reacciones en fase
gas por lo que estas reacciones son habitualmente la etapa limitante del proceso La secuencia
y duracioacuten de las etapas variacutea con el tipo de reactor utilizado
12 Antecedentes
La quiacutemica del proceso de gasificacioacuten es muy compleja pero en general las principales
reacciones gas-soacutelido y gas-gas que tienen lugar en el gasificador son las siguientes (Mondal y
cols 2011)
Combustioacuten parcial C + frac12 O2 rarr CO ∆H298K = -110 kJmiddotmol-1 (ec 21)
Combustioacuten completa C + O2 rarr CO2 ∆H298K = -393 kJmiddotmol-1 (ec 22)
Water-gas C + H2O harr CO + H2 ∆H298K = 131 kJmiddotmol-1 (ec 23)
Boudouard C + CO2 harr 2CO ∆H298K = 172 kJmiddotmol-1 (ec 24)
Metanacioacuten C + 2H2 harr CH4 ∆H298K = -75 kJmiddotmol-1 (ec 25)
Water-gas shift CO + H2O harr CO2 + H2 ∆H298K = -41 kJmiddotmol-1 (ec 26)
Reformado con vapor CH4 + H2O harr CO + 3H2 ∆H298K = 205 kJmiddotmol-1 (ec 27)
Reformado en seco CH4 + CO2 harr 2CO + 2H2 ∆H298K = 247 kJmiddotmol-1 (ec 28)
La gasificacioacuten se presenta como una de las tecnologiacuteas maacutes prometedoras para la
obtencioacuten de energiacutea El gas producto de la gasificacioacuten puede ser utilizado de varias formas
para la produccioacuten de electricidad o calor Los motores de combustioacuten interna en conexioacuten con
gasificadores de lecho fijo o lecho fluidizado a presioacuten atmosfeacuterica ofrecen una interesante
alternativa para potencias eleacutectricas moderadas (entre 50 kWe y 10 MWe) Para una mayor
generacioacuten eleacutectrica (gt 5 MWe) las turbinas de gas son la mejor tecnologiacutea siendo los
gasificadores de lecho fluidizado el tipo de reactor maacutes adecuado (Spliethoff 2001) En
teacuterminos generales los motores o turbinas de gas permiten alcanzar eficiencias eleacutectricas de
hasta el 30 (sin incluir la recuperacioacuten del calor residual) En el caso de las turbinas de gas de
alta potencia eleacutectrica (gt 25 MWe) la eficiencia eleacutectrica puede superar el 40 en los ciclos
combinados en los que ademaacutes de la turbina de gas se incorpora una turbina de vapor y una
caldera para la recuperacioacuten del calor residual Asiacute la integracioacuten de la gasificacioacuten en ciclos
combinados (GICC) se presenta como una alternativa viable econoacutemicamente de mayor
eficiencia que otras tecnologiacuteas convencionales y de menor impacto ambiental ya que permite
la eliminacioacuten de los contaminantes del gas de gasificacioacuten antes de su combustioacuten Esta etapa
de limpieza del gas resulta fundamental para evitar dificultades teacutecnicas en el
aprovechamiento energeacutetico del mismo (Martiacutenez y cols 2012)
En la actualidad existen alrededor de 117 plantas de gasificacioacuten operando alrededor de
todo el mundo de las que un 39 generan combustible 19 generan electricidad y 42
productos quiacutemicos El 49 de las 117 plantas usan carboacuten y un 36 usan coque de petroacuteleo
La capacidad instalada total de las plantas de gasificacioacuten suma 24000 MWe con un
crecimiento anual de alrededor del 10 (Concha y cols 2009) Algunas de las plantas de GICC
Antecedentes 13
maacutes importantes son la de Puertollano en Espantildea (Elcogas SA 335 MWe) la de Buggenum en
Holanda (Willem-Alexander Power Plant 253 MWe) la de Tampa en Florida EEUU (Tampa
Electrics Polk Power Station 260 MWe) y la de West Terre Haute en Indiana EEUU (Wabash
River Generating Station 262 MWe)
En el caso de la gasificacioacuten de biomasa su explotacioacuten comercial presenta una serie de
desafiacuteos tecnoloacutegicos y logiacutesticos relacionados principalmente con la cadena de suministro y el
pretratamiento de la biomasa (Asadullah 2014) A pesar de ello varias plantas de GICC se han
desarrollado a escala de demostracioacuten y a nivel comercial en todo el mundo como alternativa
al uso de combustibles foacutesiles para la produccioacuten de electricidad (Kinoshita y cols 1997) La
primera y maacutes destacable se construyoacute en Vaumlrnamo (Suecia) utilizando astillas de madera para
producir 6 MW eleacutectricos y 9 MW teacutermicos (Staringhl y Neergaard 1998) Esta instalacioacuten funcionoacute
entre 1993 y 1999 pero tuvo que ser cerrada por motivos econoacutemicos Las casi 3600 h de
funcionamiento como planta integrada demostraron la posibilidad de utilizar el gas de
gasificacioacuten obtenido en una turbina de gas en condiciones estables con un poder caloriacutefico de
tan solo 38 MJmiddotkg-1 Desde la deacutecada de los antildeos 90 otros proyectos de gasificacioacuten de
biomasa a escala de demostracioacuten o semi-comercial se han desarrollado en todo el mundo
utilizando diferentes tecnologiacuteas entre las que se pueden citar Lurgi Technology Termiska
Processor Sweden AB Renugas Process etc (Spliethoff 2001)
Una de las principales limitaciones para una mayor implantacioacuten de la gasificacioacuten a nivel
comercial es la presencia de alquitraacuten en el gas producto En el contexto de la gasificacioacuten la
definicioacuten maacutes ampliamente aceptada es la que define los alquitranes como el grupo de
compuestos orgaacutenicos maacutes pesados que el benceno sin tener en cuenta el soot (holliacuten) ni el
char (residuo soacutelido carbonoso) (Neeft y cols 2002) El tolueno y el naftaleno son algunos de
los compuestos mayoritarios en el alquitraacuten junto con el fenol cuando la temperatura de
operacioacuten es baja (lt 800 ordmC) (Spliethoff 2001) La presencia de alquitraacuten en el gas de
gasificacioacuten conlleva problemas operacionales debido a su facilidad para condensar (lt 450 ordmC)
formar aerosoles y polimerizar dando lugar a estructuras maacutes complejas lo cual provoca
problemas de ensuciamiento y taponamientos en tuberiacuteas y equipos donde vaya a ser utilizado
el gas como motores y turbinas (McKendry 2002b) Ademaacutes los compuestos presentes en el
alquitraacuten constituyen un serio problema medioambiental por su caraacutecter persistente y toacutexico
(Nisbet y Lagoy 1992)
El valor liacutemite de concentracioacuten de alquitraacuten en el gas de gasificacioacuten depende del uso
final del mismo Como valor de referencia suele fijarse una concentracioacuten maacutexima de 100
mgmiddotm-3N para el uso del gas en motores de combustioacuten interna y de 5 mgmiddotm-3N para su uso en
14 Antecedentes
turbinas (Spliethoff 2001) Los gasificadores disponibles hoy en diacutea en el mercado superan con
creces estos valores liacutemite variando entre 05 y 100 gmiddotm-3N en funcioacuten de la materia prima las
condiciones de operacioacuten y fundamentalmente el tipo de reactor (Devi y cols 2003) El
proceso de gasificacioacuten puede tener lugar en diversos tipos de reactores siendo los
gasificadores de lecho fijo (updraft o downdraft) de lecho fluidizado o de flujo arrastrado los
maacutes utilizados (Mondal y cols 2011) Los gasificadores de lecho fijo son adecuados para
plantas de baja capacidad mientras que los gasificadores de lecho fluidizado son maacutes
habituales en instalaciones de mayor tamantildeo (gt 5 MWt) (Spliethoff 2001) Con la
configuracioacuten de lecho fluidizado (que es la tecnologiacutea utilizada en el presente trabajo) se
consigue una mayor tasa de conversioacuten del soacutelido en comparacioacuten con los gasificadores de
lecho fijo debido a que el movimiento del lecho favorece el buen contacto soacutelido-gas y mejora
la transferencia de masa y calor El contenido de partiacuteculas y de alquitraacuten en el gas procedente
de los gasificadores de lecho fluidizado es habitualmente superior al de los reactores de lecho
fijo downdraft (corriente descendente) e inferior al de los gasificadores updraft
(contracorriente) (Han y Kim 2008) Los valores habituales de contenido de alquitraacuten en la
gasificacioacuten de biomasa en lecho fluidizado oscilan entre 8 y 15 gmiddotm-3N (Corella y cols 2006)
Dado que estos valores son muy superiores a los liacutemites establecidos la reduccioacuten del
contenido de alquitraacuten en el gas de gasificacioacuten es un aspecto clave para su aprovechamiento
Como se detalla maacutes adelante las tecnologiacuteas de eliminacioacuten de alquitraacuten se dividen en dos
grandes grupos meacutetodos primarios que incluyen todas las medidas adoptadas dentro del
propio gasificador para producir un gas lo maacutes limpio posible y meacutetodos secundarios o de
limpieza del gas aguas abajo del gasificador
23 Influencia de las condiciones de operacioacuten en el proceso de gasificacioacuten
Tanto la formacioacuten de alquitraacuten como la calidad del gas producto de la gasificacioacuten se ven
fuertemente influenciadas por las condiciones de operacioacuten Por lo tanto una adecuada
seleccioacuten de los paraacutemetros de operacioacuten es por siacute misma un meacutetodo primario para reducir la
formacioacuten de alquitraacuten
La temperatura es uno de los paraacutemetros maacutes influyentes en el proceso de gasificacioacuten de
biomasa afectando tanto a la cineacutetica como a la termodinaacutemica de las reacciones y por tanto
a la composicioacuten del gas y a su concentracioacuten de alquitraacuten Dada la complejidad y
simultaneidad de las reacciones durante el proceso de gasificacioacuten la evolucioacuten de los
distintos compuestos gaseosos (H2 CO CO2 CH4) con la temperatura presenta variaciones
en funcioacuten del intervalo de temperatura y del predominio de unas u otras reacciones En
Antecedentes 15
general bajas temperaturas de gasificacioacuten conllevan un alto contenido de alquitraacuten y un bajo
contenido de CO y H2 en el gas producto (Asadullah 2014) El aumento de la temperatura
conduce a un mayor rendimiento a gas en detrimento de la concentracioacuten de alquitraacuten que
puede reaccionar a traveacutes de diferentes mecanismos (Li y Suzuki 2009a) El reformado con
vapor (ec 29) y el reformado en seco (ec 210) son algunas de las reacciones de eliminacioacuten
de alquitraacuten maacutes importantes
Reformado con vapor CnHm + n H2O harr n CO + (n + m2) H2 ∆H gt 0 (ec 29)
Reformado en seco CnHm + n CO2 harr 2n CO + (m2) H2 ∆H gt 0 (ec 210)
El intervalo de temperatura maacutes habitual para la gasificacioacuten de biomasa es 750-900 ordmC
Por lo general se requieren temperaturas de operacioacuten por encima de 800 ordmC para lograr una
alta conversioacuten del carbono soacutelido y un bajo contenido de alquitraacuten en el gas producto (Devi y
cols 2003) Reducciones en el contenido de alquitraacuten de hasta un 75-95 fueron observadas
por diversos autores al aumentar la temperatura de gasificacioacuten desde 700 hasta 800-820 ordmC (Li
y Suzuki 2009a Narvaacuteez y cols 1996) Sin embargo aunque desde el punto de vista de la
eliminacioacuten del alquitraacuten interesa aumentar la temperatura de operacioacuten otros factores
limitan dicha temperatura como puede ser el riesgo de fusioacuten y aglomeracioacuten de las cenizas o
la necesidad de materiales y especificaciones maacutes exigentes para la construccioacuten y
mantenimiento del gasificador (Asadullah 2014)
Las reacciones quiacutemicas envueltas en el proceso de gasificacioacuten no soacutelo se ven afectadas
por la temperatura sino tambieacuten por la presioacuten parcial de los distintos reactivos en el medio
de reaccioacuten Por lo tanto el tipo de agente gasificante es un factor clave en la composicioacuten y
aplicabilidad del gas producto El aire es el agente gasificante maacutes habitual debido a su bajo
coste pero el nitroacutegeno introducido con el aire diluye la mezcla gaseosa (en torno a un 50
vol de N2) dando como resultado un gas con un bajo poder caloriacutefico (PCI = 4-6 MJmiddotm-3N) Este
contenido energeacutetico puede ser suficiente para el uso del gas en calderas motores o turbinas
pero no para el transporte del gas a traveacutes de tuberiacuteas debido a su baja densidad energeacutetica
(Bridgwater 1995) La gasificacioacuten con oxiacutegeno puro evita la dilucioacuten del gas aumentando su
poder caloriacutefico hasta 10-14 MJmiddotm-3N pero el coste del proceso aumenta considerablemente
debido a la necesidad de una unidad de separacioacuten de aire para la obtencioacuten del oxiacutegeno
Ademaacutes del aire u oxiacutegeno el vapor de agua puede ser tambieacuten utilizado como medio de
gasificacioacuten La presencia de vapor de agua favorece el desplazamiento de las reacciones
water-gas (ec 23) water-gas shift (ec 26) o el reformado de hidrocarburos con vapor (ec
27) hacia la produccioacuten de H2 La obtencioacuten de productos quiacutemicos y combustibles sinteacuteticos a
partir del gas de gasificacioacuten requiere un gas de siacutentesis de alta calidad formado
16 Antecedentes
mayoritariamente por H2 y CO en una proporcioacuten adecuada (H2CO = 2-3 molmiddotmol-1) En la
literatura se pueden encontrar valores de concentracioacuten de H2 de hasta un 60 en el gas
producto de la gasificacioacuten de biomasa con vapor (Herguido y cols 1992) La composicioacuten del
alquitraacuten tambieacuten parece verse afectada por el medio de reaccioacuten El alquitraacuten formado
durante la gasificacioacuten con vapor de agua es maacutes reactivo y faacutecil de destruir con catalizadores
que el formado en la gasificacioacuten con aire (Gil y cols 1999a McKendry 2002b) Asimismo Gil
y cols (1999a) encontraron mayores contenidos de alquitraacuten en la gasificacioacuten con vapor (30-
80 gmiddotm-3N) que en la gasificacioacuten con aire (2-20 gmiddotm-3N) aunque esto podriacutea tener su origen en
la menor temperatura de operacioacuten alcanzada en el primer caso
La relacioacuten entre el flujo de agente gasificante alimentado por unidad de masa de
biomasa es tambieacuten un importante factor de operacioacuten en la gasificacioacuten (Devi y cols 2003)
En el caso de gasificar con aire u oxiacutegeno esta ratio viene dada por la relacioacuten equivalente
(RE) que es el cociente entre la cantidad de oxiacutegeno alimentado por gramo de biomasa y la
cantidad estequiomeacutetrica de oxiacutegeno necesaria para la combustioacuten completa de un gramo de
biomasa En la gasificacioacuten con aire RE habitualmente oscila entre 02 y 04 (Narvaacuteez y cols
1996) Un aumento de RE supone una mayor disponibilidad de oxiacutegeno en el medio de
reaccioacuten lo que facilita la combustioacuten de la materia volaacutetil y la disminucioacuten del contenido de
alquitraacuten en el gas mejorando la produccioacuten de gas y la conversioacuten del soacutelido (Kinosita y cols
1994 Narvaacuteez y cols 1996) Sin embargo la composicioacuten del gas se ve negativamente
afectada por el excesivo aumento de RE debido al incremento de las fracciones de CO2 y N2 (al
gasificar con aire) lo que conlleva una disminucioacuten del poder caloriacutefico del gas La influencia de
RE en la concentracioacuten de H2 CO o CH4 puede mostrar diversas tendencias en funcioacuten de otros
factores como el tipo de biomasa el intervalo de temperatura o la presencia de vapor de agua
en el medio (Kumar y cols 2009) En el caso de la gasificacioacuten con vapor de agua el paraacutemetro
habitualmente utilizado es la relacioacuten maacutesica entre la cantidad de vapor de agua y la cantidad
de biomasa alimentados (SB del ingleacutes steam to biomass ratio) El aumento de la relacioacuten SB
favorece la conversioacuten del soacutelido y disminuye la concentracioacuten de alquitraacuten en el gas lo que
puede atribuirse a una mayor extensioacuten de la reacciones de reformado con vapor (ec 29) Sin
embargo el aumento de la relacioacuten SB por encima de cierto liacutemite puede repercutir de forma
negativa en la distribucioacuten de productos como consecuencia de la disminucioacuten de la
temperatura de reaccioacuten
A diferencia de la gasificacioacuten con aire u oxiacutegeno que conlleva la combustioacuten parcial de la
biomasa en condiciones subestequiomeacutetricas la gasificacioacuten del carbono con vapor de agua es
una reaccioacuten endoteacutermica y requiere un aporte continuo de energiacutea Dado que la transferencia
Antecedentes 17
de calor a elevadas temperaturas presenta serias dificultades la operacioacuten del gasificador en
reacutegimen autoteacutermico es la opcioacuten maacutes interesante Para ello la adicioacuten de cierta cantidad de
oxiacutegeno junto con el vapor de agua puede proporcionar la energiacutea necesaria en el medio de
reaccioacuten gracias a la combustioacuten parcial de la materia prima Los estudios relacionados con el
uso de mezclas de aire (u oxiacutegeno) y vapor de agua como agente gasificante en la gasificacioacuten
de biomasa no son muy numerosos (Campoy y cols 2009 Gil y cols 1997 Lv y cols 2004
Pinto y cols 2003) por lo que se requieren maacutes estudios para profundizar tanto en los
aspectos operacionales como energeacuteticos del proceso
Gasificacioacuten de lodos de EDAR
En el caso concreto de la gasificacioacuten de lodos de EDAR (proceso en el que se centra esta
Tesis) los primeros estudios publicados se remontan a mediados de los antildeos 90 (Bacaicoa y
cols 1995) Desde entonces la gasificacioacuten de lodos de EDAR ha sido estudiada como una
posible alternativa para su conversioacuten en energiacutea uacutetil con el objetivo de reducir a la vez el
volumen de residuo y el impacto medioambiental que puede ocasionar su mala gestioacuten Los
trabajos publicados tanto a escala de laboratorio (Adegoroye y cols 2004 Aznar y cols 2008
Manyagrave y cols 2005 Tae-Young y cols 2009) como en planta piloto (Campoy y cols 2014
Dogru y cols 2002 Midilli y cols 2001 Petersen y Werther 2005 Van der Drift y cols 2001)
confirman la posibilidad de obtener un gas combustible a partir del lodo Por ejemplo Midilli y
cols (2001) realizaron experimentos de gasificacioacuten de lodo con aire en un reactor downdraft
(10 kWe) obteniendo un gas combustible con un PCI de aproximadamente 38 MJmiddotm-3N Hasta
la fecha no se conocen proyectos de gasificacioacuten de lodo a nivel comercial pero siacute que se han
desarrollado algunos proyectos a escala de demostracioacuten directamente en estaciones
depuradoras utilizando el gas producto para la produccioacuten de electricidad en un ciclo
combinado (75 kWe en la planta de Balingen en Alemania) o para el secado del lodo (15 MWt
en la planta de Mannheim en Alemania) (Judex y cols 2012)
La mayoriacutea de los estudios de gasificacioacuten de lodo que se encuentran en la bibliografiacutea se
centran en el uso de aire para gasificar lodo seco siendo la relacioacuten equivalente (RE) y la
temperatura las variables maacutes estudiadas Sin embargo teniendo en cuenta que el contenido
de humedad en el lodo tras la deshidratacioacuten mecaacutenica y antes del secado teacutermico puede
alcanzar el 70 (Manara y Zabaniotou 2012) la gasificacioacuten del residuo huacutemedo puede ser
una alternativa interesante Los estudios de gasificacioacuten de lodo huacutemedo que se encuentran en
la literatura no son muy numerosos (Domiacutenguez y cols 2006 Xie y cols 2010 Zhang y cols
2011) pero todos ellos muestran una mejora de la produccioacuten de H2 como consecuencia de la
gasificacioacuten de la materia orgaacutenica del lodo con su propio contenido de humedad Como
18 Antecedentes
simulacioacuten al proceso de gasificacioacuten de lodo huacutemedo Nipattummakul y cols (2010)
observaron una produccioacuten de H2 de hasta tres veces mayor al usar vapor de agua en lugar de
aire para gasificar lodo seco La escasez de estudios acerca del efecto de la atmoacutesfera reactiva
en la calidad del gas producto de la gasificacioacuten de lodos motivoacute parte del estudio desarrollado
en esta Tesis
Gasificacioacuten de char de piroacutelisis
Ademaacutes de las condiciones de operacioacuten la composicioacuten de la biomasa (contenido en
cenizas materia volaacutetil carbono fijo humedadhellip) condiciona la distribucioacuten de productos en la
gasificacioacuten Desde el punto de vista de la reduccioacuten de alquitraacuten el producto soacutelido resultante
de la piroacutelisis de biomasa (char de piroacutelisis) puede ser una materia prima maacutes adecuada para la
gasificacioacuten que la propia biomasa original Ademaacutes de ser una de las primeras etapas en los
procesos de gasificacioacuten y combustioacuten de la biomasa la piroacutelisis es en siacute misma un proceso
termoquiacutemico que consiste en la descomposicioacuten teacutermica de un sustrato carbonoso en
atmoacutesfera inerte Durante la piroacutelisis de biomasa se produce la liberacioacuten de gran parte de su
contenido volaacutetil que tras condensar da lugar al liacutequido de piroacutelisis o bio-oil La fraccioacuten
liacutequida es el producto de intereacutes de la piroacutelisis raacutepida (Bridgwater y Peacocke 2000) aunque
en ella se genera tambieacuten una importante fraccioacuten de producto soacutelido La estructura y
composicioacuten de este soacutelido son bastante diferentes a las de la biomasa original con una mayor
estructura porosa y un mayor contenido de carbono fijo Dicha estructura porosa ha dado
lugar al uso de estos materiales en la preparacioacuten de carbones activos (Gonzaacutelez y cols 2009)
y su aplicacioacuten como adsorbentes para la eliminacioacuten de contaminantes (metales pesados
colorantes fenoles NOxhellip) (Raveendran y Ganesh 1998) Por otro lado el contenido de
carbono remanente en el char puede ser aprovechado energeacuteticamente mediante procesos de
combustioacuten o gasificacioacuten (Di Blasi 2009) El estudio de la gasificacioacuten de char de diferente
origen lignoceluloacutesico ha cobrado especial intereacutes en los uacuteltimos antildeos pero los estudios
publicados hasta la fecha son todaviacutea escasos Algunos de estos trabajos se centran en el
estudio de la reactividad y modelado cineacutetico del proceso (Haykiri-Acma y cols 2006
Maacuterquez-Montesinos y cols 2002 Nilsson y cols 2014) y otros muestran la posibilidad de
obtener un gas de buena calidad con una fraccioacuten molar de CO+H2 que puede alcanzar el 88
del gas (Chaudhari y cols 2003) un contenido en H2 superior al 50 en la gasificacioacuten con
vapor (Yan y cols 2010) y al 25 en la gasificacioacuten con aire (Salleh y cols 2010) y un poder
caloriacutefico superior a 4 MJmiddotm-3N (He y cols 2012)
En el caso particular de la piroacutelisis raacutepida de lodo de EDAR el char puede llegar a ser el
producto mayoritario a pesar de no ser el producto de intereacutes (rendimiento del 35-55)
Antecedentes 19
(Fonts y cols 2008 Fonts y cols 2012 Inguanzo y cols 2002 Pokorna y cols 2009 Shen y
Zhang 2003) Su reutilizacioacuten como material adsorbente ha sido investigado por algunos
autores aunque los resultados han mostrado una baja superficie especiacutefica (50-150 m2middotg-1) en
comparacioacuten con la de los carbones activos comerciales (gt 500 m2middotg-1) debido a su alto
contenido inorgaacutenico (Smith y cols 2009) El aprovechamiento energeacutetico de este tipo de char
apenas ha sido estudiado y los trabajos que se encuentran en la bibliografiacutea son estudios
cineacuteticos y de reactividad que apenas hacen hincapieacute en las propiedades del gas producto
(Nilsson y cols 2012 Nowicki y cols 2011 Scott y cols 2005) por lo que la gasificacioacuten del
char resultante de la piroacutelisis de lodos de EDAR ha sido objeto de otro de los estudios
desarrollados en la Tesis
24 Limpieza del gas producto de la gasificacioacuten
Como se ha comentado anteriormente las tecnologiacuteas de eliminacioacuten de alquitraacuten
pueden clasificarse en meacutetodos primarios y secundarios Ademaacutes de una apropiada seleccioacuten
de las condiciones de operacioacuten la adicioacuten de catalizadores en el propio gasificador es otro de
los tratamientos primarios que puede ayudar a reducir la formacioacuten de alquitraacuten El uso de
minerales naturales como la dolomita o la olivina o de catalizadores metaacutelicos basados en
hierro o niacutequel ha sido ampliamente estudiado obtenieacutendose reducciones del contenido de
alquitraacuten en el gas superiores al 50 y llegando a niveles de hasta 1-2 gmiddotm-3N en algunos casos
(De Andreacutes y cols 2011 Gil y cols 1999b Miccio y cols 2009 Olivares y cols 1997 Rapagnagrave
y cols 2000) Sin embargo los problemas de desactivacioacuten por deposicioacuten de carbono y de
erosioacuten y arrastre de las partiacuteculas en el caso del uso de materiales naturales en lechos
fluidizados impide la operacioacuten durante largos periacuteodos de tiempo (De Andreacutes y cols 2011)
Los exigentes requisitos de calidad fijados para la mayoriacutea de las aplicaciones del gas
hacen necesaria una limpieza adicional del gas aguas abajo del gasificador Los tratamientos
secundarios de eliminacioacuten de alquitraacuten se clasifican en meacutetodos fiacutesicos craqueo teacutermico o
craqueo cataliacutetico Entre los sistemas de limpieza fiacutesicos se incluye el uso de ciclones
precipitadores electrostaacuteticos filtros (filtros de tela filtros de ceraacutemica) torres lavadoras o
scrubbers y adsorcioacuten en carboacuten activo (Abu El-Rub y cols 2004 Anis y Zainal 2011) La
mayoriacutea de estos meacutetodos requieren del enfriamiento del gas para la separacioacuten del alquitraacuten
condensado en pequentildeas gotas o aerosoles lo que conlleva una disminucioacuten de la eficiencia
energeacutetica del proceso Las partiacuteculas soacutelidas arrastradas por el gas tambieacuten quedan retenidas
en este tipo de dispositivos junto con los aerosoles de alquitraacuten La eficiencia en la eliminacioacuten
de alquitraacuten se situacutea por ejemplo en torno a un 40 con el uso de precipitadores
20 Antecedentes
electrostaacuteticos y en un 70 con filtros de tela mientras que las torres lavadoras de gases de
tipo venturi pueden alcanzar eficiencias de hasta el 90 (Anis y Zainal 2011 Han y Kim 2008)
Ademaacutes del enfriamiento del gas la gestioacuten del residuo generado al separar el alquitraacuten del
gas sin ser destruido es otro de los inconvenientes en este tipo de tratamientos de limpieza
Mediante la conversioacuten del alquitraacuten en moleacuteculas maacutes ligeras como H2 CO y CH4 se
evitan los problemas de gestioacuten a la vez que se incrementa la produccioacuten final de gas Dado el
control cineacutetico en estas reacciones (Abu El-Rub y cols 2004 Anis y Zainal 2011) la
descomposicioacuten del alquitraacuten requiere temperaturas extremadamente altas (craqueo teacutermico)
o bien el uso de catalizadores (craqueo o reformado cataliacutetico) El alquitraacuten derivado de la
biomasa es muy refractario y difiacutecil de craquear soacutelo por efecto teacutermico necesitando
temperaturas por encima de los 1000 ordmC para la ruptura de sus enlaces con los problemas
teacutecnicos y energeacuteticos que esto conlleva (Brandt y Henriksen 2000 Bridgwater 1995) El uso
de catalizadores permite disminuir la temperatura de craqueo pudiendo operar incluso a la
temperatura del gas a su salida del gasificador lo que supone un oacuteptimo dese el punto de vista
energeacutetico Todas estas ventajas han hecho que el craqueo cataliacutetico de los alquitranes haya
centrado muchos estudios desde la deacutecada de los 80
El craqueo o reformado cataliacutetico del alquitraacuten implica la adsorcioacuten disociativa de los
hidrocarburos en los centros activos del catalizador (fase metaacutelica) donde se produce la
deshidrogenacioacuten y posteriormente la reaccioacuten con vapor de agua (ec 29) o con CO2 (ec
210) (Han y Kim 2008) Entre los soacutelidos maacutes estudiados para el reformado del alquitraacuten se
pueden destacar algunos minerales y rocas naturales como la olivina la dolomita o la calcita y
catalizadores preparados a base de metales alcalinos y metales de transicioacuten (Abu El-Rub y
cols 2004 Anis y Zainal 2011 De Lasa y cols 2011 Sutton y cols 2001a) La dolomita
calcinada ha sido uno de los minerales maacutes estudiados mostrando conversiones de alquitraacuten
superiores al 95 en determinadas condiciones (Delgado y cols 1997) Para mayores niveles
de pureza los catalizadores basados en metales de transicioacuten y especialmente los
catalizadores de niacutequel ofrecen una buena alternativa A temperaturas superiores a 740 ordmC los
catalizadores de niacutequel no soacutelo favorecen la eliminacioacuten del alquitraacuten sino tambieacuten el
reformado del metano con vapor (ec 27) y el ajuste de la relacioacuten H2CO a traveacutes de la
reaccioacuten water-gas shift (ec 26) (Sutton y cols 2001a) Conversiones de alquitraacuten del 98-99
han sido obtenidas con catalizadores de niacutequel utilizados comercialmente para otros procesos
de reformado con vapor (Aznar y cols 1998 Zhang y cols 2004) Sin embargo la
desactivacioacuten de este tipo de catalizadores es uno de los principales inconvenientes para su
aplicacioacuten en la gasificacioacuten a gran escala Los fenoacutemenos de desactivacioacuten maacutes habituales en
Antecedentes 21
este tipo de catalizadores son (i) la sinterizacioacuten que se produce en condiciones severas de
temperatura e implica la migracioacuten de pequentildeas partiacuteculas de niacutequel dispersas sobre la
superficie del catalizador para formar partiacuteculas maacutes grandes disminuyendo asiacute la presencia de
centros activos (ii) la deposicioacuten de carbono sobre los centros activos que cobra especial
importancia cuando el contenido de alquitraacuten en el gas es alto algunos estudios apuntan a un
valor liacutemite de 2 gmiddotm-3N para evitar en gran medida este fenoacutemeno (Aznar y cols 1998) y (iii)
el envenenamiento de los centros activos debido a la presencia de impurezas en el gas de
gasificacioacuten como cloro o azufre (Abu El-Rub y cols 2004 Anis y Zainal 2011 De Lasa y cols
2011 Sutton y cols 2001a)
Buena parte de la literatura existente acerca del craqueo cataliacutetico de alquitraacuten se centra
en la preparacioacuten y modificacioacuten de catalizadores de niacutequel con diferentes soportes y
promotores para mejorar su actividad y estabilidad El soporte del catalizador juega un papel
clave en la dispersioacuten de la fase activa Diferentes oacutexidos metaacutelicos (Al2O3 MgO ZrO2 TiO2
SiO2 o CeO2) y materiales naturales (dolomita olivina o carboacuten activo) han sido utilizados como
soporte en la preparacioacuten de catalizadores de niacutequel (Courson y cols 2000 Kimura y cols
2006 Li y cols 2009b Miyazawa y cols 2006 Park y cols 2010 Sato y Fujimoto 2007
Srinakruang y cols 2006 Wang y cols 2005) Algunos estudios apuntan al conjunto niacutequel-
aluacutemina (Al2O3) como uno de los catalizadores maacutes eficaces para la eliminacioacuten de alquitraacuten
(Sutton y cols 2001b) aunque sin estar exento de desactivacioacuten (Swierczynski y cols 2007)
La incorporacioacuten de metales alcalinos (Na K) o de elementos de transicioacuten (Ru Rh Mn Mo W
Zr Mn) en catalizadores de niacutequel ha sido tambieacuten objeto de muchos estudios obteniendo en
algunos casos resultados positivos en cuanto a la actividad reducibilidad regenerabilidad
propiedades mecaacutenicas y resistencia frente a los fenoacutemenos de desactivacioacuten especialmente
la debida a la deposicioacuten de carbono (Bona y cols 2008 Dou y cols 2003 Nishikawa y cols
2008 Richardson y Grey 1997 Seok y cols 2002 Yung y cols 2009 Zhang y cols 2007)
Ademaacutes de alquitraacuten el gas procedente de la gasificacioacuten de biomasa o de residuos
orgaacutenicos puede contener otras impurezas como consecuencia de la propia composicioacuten de la
materia prima Es el caso del H2S que se forma durante la gasificacioacuten de materiales que
contienen azufre (Meng y cols 2010) como es el caso de los lodos de EDAR La presencia de
H2S en el gas de gasificacioacuten conlleva tanto problemas ambientales relacionados con las
emisiones de SO2 en la combustioacuten del gas contribuyendo asiacute a la lluvia aacutecida como
operacionales relacionados con la corrosioacuten de tuberiacuteas motores y turbinas y con la
desactivacioacuten de los catalizadores de niacutequel utilizados para el craqueo del alquitraacuten La
presencia de H2S en los gases de gasificacioacuten ha dado lugar a diversos estudios para
22 Antecedentes
profundizar en el fenoacutemeno de envenenamiento de los catalizadores de niacutequel con azufre
(Engelen y cols 2003 Hepola y Simell 1997a Hepola y Simell 1997b Struis y cols 2009) El
azufre queda adsorbido en los catalizadores de niacutequel en diferentes estados quiacutemicos en
funcioacuten de las condiciones de operacioacuten Aunque parte del azufre queda quiacutemicamente
adsorbido de forma irreversible otra parte puede desorberse a alta temperatura (900 ordmC) lo
que hace que el catalizador pueda recuperar parte de su actividad inicial cuando el H2S es
eliminado de la corriente gaseosa (Hepola y Simell 1997b) El envenenamiento de los
catalizadores de niacutequel con H2S puede evitarse mediante el acondicionamiento previo del gas
de gasificacioacuten pero la incorporacioacuten de promotores que puedan mejorar su estabilidad en
presencia de H2S resulta tambieacuten un factor interesante Ambos aspectos han sido abordados
en esta Tesis
Existen diversos tratamientos para la eliminacioacuten del H2S de corrientes gaseosas tanto a
baja como a alta temperatura El lavado de los gases con disolventes baacutesicos es uno de los
procesos maacutes utilizados en la industria quiacutemica (Yildirim y cols 2012) Tambieacuten el uso de
carbones activos y del char procedente de procesos de piroacutelisis (incluyendo la piroacutelisis de lodos
de EDAR) para la adsorcioacuten fiacutesica de H2S ha sido ampliamente estudiado (Bagreev y Bandosz
2005 Bandosz 2002 Gutieacuterrez-Ortiz y cols 2014 Primavera y cols 1998 Ros y cols 2006
Yuan y Bandosz 2007) Ambos procesos requieren el enfriamiento del gas para la retencioacuten del
H2S lo que resulta desfavorable desde el punto de vista energeacutetico En cambio con los
procesos de desulfuracioacuten a alta temperatura se evita el enfriamiento del gas soacutelo para su
limpieza evitando tambieacuten la condensacioacuten del alquitraacuten en el caso del gas de gasificacioacuten Los
procesos de desulfuracioacuten a alta temperatura se basan en la reaccioacuten quiacutemica del H2S con
determinados oacutexidos metaacutelicos de forma que el azufre queda retenido en forma de sulfuros
metaacutelicos Los oacutexidos de zinc manganeso cobre hierro y calcio son algunos de los oacutexidos con
mayor capacidad para retener H2S (Aacutelvarez-Rodriacuteguez y Clemente-Jul 2008 Cheah y cols
2009 Elseviers y Verelst 1999 Meng y cols 2010 Park y cols 2005 Tamhankar y cols 1981
Westmoreland y Harrison 1976) aunque todos ellos presentan liacutemites de operacioacuten
relacionados principalmente con el tipo de atmoacutesfera reactiva y con la temperatura
Dado el contenido metaacutelico de la fraccioacuten inorgaacutenica de la biomasa el uso de las cenizas
resultantes de los tratamientos termoquiacutemicos para la desulfuracioacuten de gases a alta
temperatura podriacutea ser una opcioacuten interesante debido a su bajo coste En el caso de los lodos
de EDAR en los que su fraccioacuten inorgaacutenica se situacutea en torno al 40 con importantes
contenidos de calcio y hierro (Manara y Zabaniotou 2012) la reutilizacioacuten de sus cenizas para
la eliminacioacuten del H2S generado en el propio proceso supondriacutea una ventaja desde el punto de
Antecedentes 23
vista del aprovechamiento integral de los subproductos En este contexto se desarrolloacute el
uacuteltimo de los estudios que componen esta Tesis en el que se utilizaron cenizas de combustioacuten
y de gasificacioacuten de lodo para la eliminacioacuten de H2S bajo distintas atmoacutesferas gaseosas
continuando asiacute con un trabajo previo desarrollado en el GPT (Garciacutea y cols 2011)
24 Materiales y meacutetodos
3 MATERIALES Y MEacuteTODOS
31 Materiales
311 Materia prima para la gasificacioacuten lodos de EDAR y char de piroacutelisis del lodo
Los residuos utilizados como materia prima para los experimentos de gasificacioacuten fueron
el lodo procedente de la estacioacuten depuradora de aguas residuales de Butarque en Madrid y el
char resultante de su proceso de piroacutelisis En esta depuradora las aguas son sometidas a un
tratamiento de depuracioacuten mediante lodos activos y posteriormente los lodos se estabilizan
mediante digestioacuten anaerobia y son secados teacutermicamente La muestra de lodo recibida en
forma granulomeacutetrica se sometioacute a un proceso de molienda y tamizado hasta alcanzar un
tamantildeo de partiacutecula de 250-500 μm
En la Tabla 31 se presenta una breve caracterizacioacuten de ambos materiales El anaacutelisis
inmediato se realizoacute de acuerdo a las especificaciones de normas estaacutendar (ISO-589-1981 para
la humedad ISO-1171-1976 para las cenizas ISO-5623-1974 para la materia volaacutetil) El anaacutelisis
elemental (C H N S) fue realizado en el Servicio de Anaacutelisis del Instituto de Carboquiacutemica
(Zaragoza) utilizando un analizador elemental Carlo Erba EA1108 El poder caloriacutefico superior
de los soacutelidos (PCS) se midioacute en el laboratorio con un caloriacutemetro IKA C-2000 y su capacidad
caloriacutefica especiacutefica (Cp) se determinoacute por calorimetriacutea diferencial de barrido con un equipo
Netzsch DSC 200 Maia (atmoacutesfera inerte 40 mL N2middotmin-1)
Tabla 31 Caracterizacioacuten del lodo y del char
Lodo Char Anaacutelisis inmediato ( maacutesico base huacutemeda)
Humedad 648 170 Cenizas 3904 7420 Volaacutetiles 5009 1502
Carbono fijo (por diferencia) 439 908
Anaacutelisis elemental ( maacutesico base huacutemeda)
C 2950 1549 H 467 097 N 527 185 S 131 035
PCS (MJmiddotkg-1) 128 52 PCI (MJmiddotkg-1) 118 50 Cp25ordmC (kJmiddotkg-1middotK-1) 115 082
Materiales y meacutetodos 25
El porcentaje maacutesico de carbono fijo en el char (908) se duplicoacute con respecto al valor
inicial del lodo (439) como consecuencia del tratamiento teacutermico de carbonizacioacuten que en
mayor o menor medida tiene lugar durante un proceso de piroacutelisis Soacutelo el 15 del carbono
contenido en el lodo estaacute en forma de carbono fijo mientras que este valor alcanza el 59 en
el char
312 Cenizas de combustioacuten y gasificacioacuten de lodos de EDAR
Las cenizas resultantes de los procesos de gasificacioacuten y combustioacuten del lodo
anteriormente descrito han sido utilizadas para la retencioacuten de H2S bajo diferentes atmoacutesferas
gaseosas
Para la obtencioacuten de las cenizas la combustioacuten del lodo se llevoacute a cabo en una mufla en
atmoacutesfera de aire manteniendo una temperatura de 900 ordmC durante dos horas (20 ordmCmiddotmin-1)
Por otro lado las cenizas de gasificacioacuten utilizadas son el residuo soacutelido obtenido en uno de los
experimentos de gasificacioacuten de lodo (experimento 5 en la Tabla 36) Ambas muestras de
ceniza fueron caracterizadas mediante diversas teacutecnicas Los posibles restos de C H S y N se
midieron con un analizador elemental Leco TruSpec Micro Sus propiedades texturales
(superficie especiacutefica volumen de poro y tamantildeo medio de los poros) se determinaron a partir
de isotermas de adsorcioacuten de N2 utilizando un equipo Micromeritics TriStar II 3000 (meacutetodos
BET y BJH) Las muestras fueron previamente desgasificadas a 200 ordmC durante 8 h bajo un flujo
de N2 y despueacutes las isotermas de adsorcioacuten-desorcioacuten se obtuvieron a -196 ordmC y a temperatura
ambiente respectivamente El contenido de metales en las muestras de ceniza fue
determinado por el Servicio de Anaacutelisis Quiacutemico de la Universidad de Zaragoza mediante
espectroscopiacutea de emisioacuten atoacutemica con plasma de acoplamiento inductivo (ICP-OES)
utilizando un espectroacutemetro Thermo Elemental IRIS Intrepid Las muestras fueron disueltas
mediante digestioacuten aacutecida en un sistema de reaccioacuten de microondas (CEM MARS) La Tabla 32
recoge los resultados de todos estos anaacutelisis
26 Materiales y meacutetodos
Tabla 32 Caracterizacioacuten de las cenizas de combustioacuten y gasificacioacuten de lodos de EDAR
Ceniza de combustioacuten
Ceniza de gasificacioacuten
Anaacutelisis elemental ( maacutesico base huacutemeda)
C 015 314 H nd nd N 028 077 S 046 041
Superficie especiacutefica (m2middotg-1) 65 67 Volumen de poro (cm3middotg-1) 002 002 Tamantildeo medio de poro (nm) 120 109 Contenido en metales (mgmiddotg-1
ceniza) Al 52 61 Ca 65 84 Fe 192 116 K 14 na P 63 51
Mg 17 na Na 4 na Si 122 na Ti 4 na
nd no detectado na no analizado
Como se observa en la Tabla 32 el anaacutelisis elemental de la ceniza de gasificacioacuten muestra
un mayor contenido de carbono como consecuencia de la incompleta conversioacuten del
contenido orgaacutenico del lodo durante el proceso de gasificacioacuten Ambas cenizas presentan
ademaacutes una pequentildea fraccioacuten de azufre antes de su uso como material desulfurante (en torno
a 04) Las propiedades texturales de ambos soacutelidos son muy pobres pero su potencial como
material desulfurante radica en su contenido metaacutelico Entre los metales analizados por ICP-
OES los elementos mayoritarios fueron Fe Si Al y Ca Entre ellos la reaccioacuten de los oacutexidos de
hierro y calcio con H2S estaacute termodinaacutemicamente favorecida en determinadas condiciones de
operacioacuten (Westmoreland y Harrison 1976) Por otro lado las diferencias observadas en el
contenido metaacutelico de ambos soacutelidos ponen de manifiesto que la fraccioacuten inorgaacutenica del lodo
no permanece completamente inerte durante los procesos de combustioacuten y gasificacioacuten
Especialmente notable es la diferencia observada en el contenido de hierro que podriacutea
explicarse teniendo en cuenta que parte del contenido inicial de hierro en el lodo puede estar
en forma de cloruro de hierro (FeCl3) como consecuencia de la utilizacioacuten de este compuesto
como agente coagulante en el tratamiento de las aguas residuales Durante la combustioacuten el
exceso de oxiacutegeno favorece la retencioacuten del hierro en el soacutelido en forma de oacutexidos Sin
embargo durante el proceso de gasificacioacuten la escasa presencia de oxiacutegeno limita la
conversioacuten total del FeCl3 a oacutexidos de hierro Esto puede provocar que dado que el punto de
Materiales y meacutetodos 27
ebullicioacuten del FeCl3 se situacutea en torno a 315 ordmC parte del contenido inicial del hierro en el soacutelido
abandone el reactor junto con el gas
Otro de los anaacutelisis realizados a las cenizas de lodo antes de su uso en las pruebas de
desulfuracioacuten fue la determinacioacuten de su estructura cristalina mediante difraccioacuten de rayos X
(XRD) Este anaacutelisis fue realizado por el Servicio de Difraccioacuten de Rayos X y Anaacutelisis por
Fluorescencia de la Universidad de Zaragoza utilizando un difractoacutemetro Rigaku D-Max
equipado con un aacutenodo de cobre (tensioacuten de 40 kV y corriente de 80 mA) Las mediciones se
realizaron en el intervalo de 5ordm a 95ordm del aacutengulo de Bragg (2Ɵ) con una velocidad de barrido de
003ordmmiddots-1 La deteccioacuten de las fases cristalinas se realizoacute de acuerdo con la base de datos del
Centro Internacional de Datos de Difraccioacuten (JCPDS 2000) En la Figura 31 se muestran los dos
difractogramas XRD obtenidos para ambas muestras de ceniza Especies como el cuarzo la
calcita oacutexidos de hierro y diferentes fosfatos de calcio y hierro han sido detectadas mediante
esta teacutecnica El estado de oxidacioacuten del hierro es una de las principales diferencias en la
estructura cristalina de ambos soacutelidos El hierro aparece en forma de hematita (Fe2O3) en la
ceniza de combustioacuten y en forma de magnetita (Fe3O4) en la ceniza de gasificacioacuten debido a la
menor disponibilidad de oxiacutegeno en este segundo caso Coherentemente el color rojizo
caracteriacutestico de la hematita soacutelo se observoacute en las cenizas de combustioacuten
Hematita (Fe2O3) Cuarzo (SiO2) Calcita (CaCO3) Fe3PO7 Ca3(PO4)2 Magnetita (Fe3O4) Whitlockita (Ca18Mg2H2(PO4)14)
Figura 31 Difractogramas XRD de las cenizas de combustioacuten y gasificacioacuten de lodo
20 30 40 50 60
2θ
Ceniza de gasificacioacuten
Ceniza de combustioacuten
28 Materiales y meacutetodos
313 Catalizadores de niacutequel
Se prepararon diferentes catalizadores de niacutequel soportados sobre aluacutemina (NiAl2O3)
mediante la incorporacioacuten de Fe Ca Mn o Cu con el fin de evaluar el efecto de estos
promotores en la estabilidad y actividad del catalizador para el reformado de alquitraacuten en
presencia de H2S La eleccioacuten de estos metales se basoacute en la capacidad de sus oacutexidos para
reaccionar con H2S (Westmoreland y Harrison 1976) En el caso de que la reaccioacuten del H2S con
el promotor se viera favorecida frente a su reaccioacuten con los centros activos de niacutequel cabriacutea
esperar una disminucioacuten o al menos un retardo en la desactivacioacuten del catalizador
Los catalizadores fueron preparados en el laboratorio por el meacutetodo de impregnacioacuten por
humedad incipiente de γ-Al2O3 (250-315 μm) con las disoluciones acuosas de los
correspondientes nitratos de los metales de intereacutes Ni(NO3)2middot6H2O Ca(NO3)2middot4H2O
Fe(NO3)3middot9H2O Cu(NO3)2middot3H2O y Mn(NO3)2middot4H2O Tanto el nitrato de niacutequel como el nitrato de
cada promotor se disolvieron en una uacutenica disolucioacuten y se impregnaron sobre la Al2O3 en un
solo paso Despueacutes de la impregnacioacuten el soacutelido se secoacute en una estufa a 110 ordmC durante 24 h y
despueacutes se calcinoacute en una mufla en atmoacutesfera de aire de acuerdo con la siguiente rampa de
temperatura 120 ordmC durante 20 min 200 ordmC durante 30 min 320 ordmC durante 90 min y por
uacuteltimo 700 ordmC o 900 ordmC durante 120 minutos La temperatura final de calcinacioacuten se varioacute entre
700 y 900 ordmC para evaluar su influencia sobre la actividad y estabilidad del catalizador La
fraccioacuten maacutesica de cada metal en los catalizadores fue de un 8 (tanto de niacutequel como de cada
promotor) Las propiedades texturales y los difractogramas XRD de los soacutelidos calcinados se
obtuvieron de forma anaacuteloga a la descrita en la seccioacuten 312 para las cenizas de lodo Los
resultados obtenidos se muestran en la Tabla 33 y en la Figura 32 respectivamente
Tabla 33 Propiedades texturales de los catalizadores calcinados
Tordf final de calcinacioacuten NiAl2O3 NiCaAl2O3 NiFeAl2O3 NiCuAl2O3 NiMnAl2O3
Superficie especiacutefica (m2middotg-1)
700 ordmC 1205 617 1081 1051 966 900 ordmC 964 527 726 624 712
Volumen de poro (cm3middotg-1)
700 ordmC 035 024 031 032 028 900 ordmC 033 022 025 025 025
Tamantildeo medio de poro (nm)
700 ordmC 111 151 110 118 113 900 ordmC 132 163 135 157 137
La adicioacuten de los promotores supuso una importante reduccioacuten de la superficie especiacutefica
del catalizador asiacute como una disminucioacuten en el volumen de poro Esto parece indicar un
exceso de carga metaacutelica en los catalizadores modificados La adicioacuten de calcio (NiCaAl2O3)
dio lugar a la mayor peacuterdida de superficie especiacutefica (reduccioacuten del 50) al mayor tamantildeo
Materiales y meacutetodos 29
medio de poro y al menor volumen de poro Las propiedades texturales tambieacuten se vieron
afectadas por la temperatura final de calcinacioacuten La superficie especiacutefica se redujo y el tamantildeo
medio de poro aumentoacute al aumentar la temperatura de 700 a 900 ordmC debido probablemente
a la sinterizacioacuten de las partiacuteculas metaacutelicas
A modo de ejemplo la Figura 32 muestra los difractogramas XRD de dos de los
catalizadores calcinados a 700 y 900 ordmC (NiMnAl2O3 y NiCuAl2O3) Las muestras son poco
cristalinas y apenas se observan diferencias en los difractogramas de los catalizadores
preparados a la misma temperatura final de calcinacioacuten independientemente del metal
antildeadido como promotor La fase mayoritaria en las muestras calcinadas a 700 ordmC fue la γ-Al2O3
La anchura de los picos correspondientes a esta fase dificulta la deteccioacuten de otros oacutexidos
metaacutelicos que cabriacutea esperar encontrar en las muestras como el NiO Como excepcioacuten la
muestra de NiCuAl2O3 calcinada a 700 ordmC mostroacute dos picos de difraccioacuten a 356ordm y 388ordm que
se corresponden con el CuO El aumento de la temperatura de calcinacioacuten hasta 900 ordmC supuso
una mayor cristalinidad de los soacutelidos (picos un poco maacutes estrechos y definidos) El NiAl2O4
aparece como la fase mayoritaria en todas estas muestras La identificacioacuten de otros posibles
aluminatos presentes en las muestras es difiacutecil ya que todos ellos son fases de tipo espinela y
presentan patrones XRD muy similares entre siacute y al de la aluacutemina
Figura 32 Difractogramas XRD de los catalizadores NiMnAl2O3 y NiCuAl2O3 calcinados a 900 y 700 ordmC
20 30 40 50 60 70 80 90
γ-Al2O3 NiAl2O4
NiMnAl2O3 (700 oC)
2θ
NiAl2O4 γ-Al2O3
NiMnAl2O3 (900 oC)
20 30 40 50 60 70 80 90
CuO γ-Al2O3
NiAl2O4
NiCuAl2O3 (700 oC)
2θ
NiAl2O4 γ-Al2O3
NiCuAl2O3 (900 oC)
30 Materiales y meacutetodos
32 Instalaciones y procedimiento experimental
321 Sistema experimental para la gasificacioacuten
Los experimentos de gasificacioacuten de lodo de EDAR y de su char de piroacutelisis se llevaron a
cabo en un reactor de lecho fluidizado a escala de laboratorio (lt 1 kgmiddoth-1) operando a presioacuten
atmosfeacuterica La Figura 33 muestra un esquema de la configuracioacuten experimental utilizada
La piroacutelisis del lodo en la que se obtuvo el char posteriormente gasificado se llevoacute a cabo
en una instalacioacuten similar a la que se muestra en la Figura 33 Se utilizoacute N2 como agente
fluidizante (velocidad de fluidizacioacuten unas 8 veces mayor que la velocidad de miacutenima
fluidizacioacuten) y la temperatura de piroacutelisis fue de 530 ordmC El tiempo medio de residencia del
soacutelido en el reactor fue de unos 8 min y el de los gases y vapores producidos de 1 s
Figura 33 Instalacioacuten experimental de gasificacioacuten
El gasificador es un reactor tubular construido en acero refractario AISI-310 El cuerpo del
reactor mide unos 127 cm y su seccioacuten variacutea a lo largo del mismo dividiendo el reactor en dos
zonas una zona para el lecho (4 cm de diaacutemetro interno) y una zona ldquofreeboardrdquo (7 cm de
diaacutemetro interno) El reactor se calienta mediante un horno eleacutectrico que cuenta con tres
Materiales y meacutetodos 31
zonas de calentamiento independientes (lecho freeboard y cicloacuten) midiendo la temperatura
en cada una de ellas con termopares tipo K y utilizando controladores PID para su control
La materia prima soacutelida se encuentra almacenada en una tolva El giro del tornillo sinfiacuten
de la tolva que se acciona mediante un motor controlado con un variador de frecuencia
permite la alimentacioacuten del soacutelido en continuo La alimentacioacuten procedente de la tolva entra a
la parte inferior del lecho a traveacutes de una tuberiacutea inclinada unos 45ordm Para evitar la
descomposicioacuten del soacutelido antes de su llegada al lecho dicha conduccioacuten se refrigera mediante
la circulacioacuten de aire a traveacutes de una camisa externa al tubo El caudal de alimentacioacuten de
soacutelido se fijoacute en torno a 21 gmiddotmin-1 en todos los experimentos de gasificacioacuten (miacutenimo valor
conseguido con el sistema de alimentacioacuten)
Como lecho inicial para empezar cada experimento se utilizoacute ceniza de lodo obtenida en
experimentos anteriores (alrededor de 120 g) Gracias a una tuberiacutea lateral situada unos 30
cm por encima de la placa distribuidora las cenizas acumuladas en el lecho durante el
experimento pueden abandonar el reactor por efecto rebosadero
El efecto de la atmoacutesfera reactiva en el proceso de gasificacioacuten se evaluoacute utilizando
diferentes mezclas de vapor de agua y aire como agente gasificantefluidizante En los casos
con mayor necesidad de oxiacutegeno la corriente de aire se enriquecioacute con oxiacutegeno puro para
mantener una velocidad de fluidizacioacuten similar en todos los experimentos realizados con el
mismo material Esta velocidad de fluidizacioacuten fue 5-7 veces mayor que la velocidad de miacutenima
fluidizacioacuten durante la gasificacioacuten de lodo y 2-3 veces mayor durante la gasificacioacuten de char El
menor contenido orgaacutenico en el char justifica la menor necesidad de agente gasificante y por
tanto esta diferencia en la velocidad de fluidizacioacuten
Los caudales de gas (aire y oxiacutegeno) se ajustan mediante controladores de flujo maacutesico
mientras que el caudal de agua se regula mediante una bomba HPLC y se evapora antes de su
entrada al reactor La mayor parte del agente gasificante se alimenta a traveacutes de la placa
distribuidora situada en la parte inferior del reactor pero una parte del aire (en torno a un
tercio del caudal requerido) se desviacutea hacia el sistema de alimentacioacuten de soacutelido para facilitar
su movimiento a traveacutes de la tuberiacutea Las dos entradas de aire cuentan con manoacutemetros que
permiten observar posibles aumentos de presioacuten producidos por obstrucciones o
taponamiento en el sistema
El tiempo de residencia de los vapores y gases en el reactor fue de alrededor de 7-8 s
durante la gasificacioacuten de lodo y de 17-18 s durante la gasificacioacuten de char debido al menor
caudal de gas utilizado A su salida del reactor la corriente de gas pasa a traveacutes de un cicloacuten y
32 Materiales y meacutetodos
de un filtro caliente (ambos a 450 ordmC para evitar la condensacioacuten de los alquitranes) en los que
se recogen las partiacuteculas soacutelidas arrastradas por el gas A continuacioacuten los gases y vapores
pasan a traveacutes de dos condensadores enfriados con un bantildeo de hielo donde condensan el
agua y el alquitraacuten Para evitar dantildeos en los siguientes dispositivos un filtro de algodoacuten
situado detraacutes de los condensadores retiene los posibles aerosoles arrastrados por la corriente
de gas Despueacutes el volumen de gas seco y libre de partiacuteculas y alquitranes se mide con un
contador volumeacutetrico (G4 Gallus 2000) y su composicioacuten se analiza en liacutenea utilizando un
cromatoacutegrafo de gases portaacutetil (Agilent 3000-A con una columna tipo Plot U y otra de tamiz
molecular) calibrado para determinar los porcentajes volumeacutetricos de H2 CO CO2 CH4 C2H4
C2H6 C2H2 N2 y H2S Los experimentos tuvieron una duracioacuten de 90 min en el caso de la
gasificacioacuten de lodo y de 60 min en la gasificacioacuten de char (menor disponibilidad de material)
Una vez finalizado cada experimento el rendimiento a los productos soacutelido y liacutequido se
determinoacute por diferencia de pesada de los dispositivos de recogida antes y despueacutes del
experimento Ambos productos fueron caracterizados por diferentes teacutecnicas
La fraccioacuten liacutequida se recuperoacute de los condensadores utilizando metanol como disolvente
para su lavado Su contenido de agua se determinoacute mediante valoracioacuten Karl Fischer (equipo
Mettler Toledo V20) de modo que la cantidad de alquitraacuten presente en la muestra podiacutea
determinarse por diferencia descontando tambieacuten la cantidad de metanol antildeadido para el
lavado de los condensadores Ademaacutes la cantidad de carbono orgaacutenico presente en las
muestras liacutequidas se midioacute con un analizador de carbono orgaacutenico total (analizador TOC-L
CSHCSN Shimadzu) obteniendo asiacute otra idea aproximada del contenido de alquitraacuten Por
uacuteltimo la composicioacuten del alquitraacuten (soacutelo del producido en la gasificacioacuten del lodo) se analizoacute
de forma cualitativa mediante un sistema de cromatografiacutea de gases que combina un
espectroacutemetro de masas y un detector de ionizacioacuten de llama (cromatoacutegrafo Agilent 5975C
GCMSD combinado con Agilent 7890A GC)
Respecto al producto soacutelido su composicioacuten elemental (C H N S) se determinoacute con un
analizador elemental Leco TruSpec Micro y su contenido en ceniza se determinoacute de acuerdo
con una norma estaacutendar (ISO 1171-1976)
322 Sistema experimental para los ensayos de retencioacuten de H2S
Las pruebas de desulfuracioacuten se realizaron en una instalacioacuten experimental maacutes pequentildea
que la anterior utilizando un reactor tubular de cuarzo (40 cm de longitud y 1 cm de diaacutemetro
interno) y operando a presioacuten atmosfeacuterica y en configuracioacuten de lecho fijo La Figura 34
muestra un esquema de la instalacioacuten utilizada
Materiales y meacutetodos 33
Figura 34 Instalacioacuten experimental para las pruebas de desulfuracioacuten
En cada experimento se utilizoacute 1 g de soacutelido desulfurante (ceniza de combustioacuten o de
gasificacioacuten de lodo) El soacutelido se introduce en el reactor apoyado sobre un poco de lana de
vidrio a una distancia de unos 18 cm desde la parte superior del reactor (donde no hay
gradientes importantes de temperatura) El reactor se coloca en el interior de un horno
ciliacutendrico La temperatura del soacutelido se mide con un termopar tipo K (116ldquo de diaacutemetro) cuyo
extremo se situacutea en el interior del lecho y se controla con un sistema de control PID
El gas se alimenta por la parte superior del reactor y sale por la parte inferior de eacuteste El
caudal de gas se ajusta con un controlador de flujo maacutesico Dicho caudal fue de 50 mLNmiddotmin-1
en todos los experimentos Se utilizaron dos gases diferentes con el fin de evaluar el efecto de
la atmoacutesfera reactiva en el proceso de desulfuracioacuten Uno de ellos era una mezcla que
conteniacutea soacutelo H2S y N2 (5000 ppm H2S) lo que permitiacutea estudiar el proceso sin la interferencia
de ninguacuten otro gas La otra mezcla gaseosa utilizada fue un gas sinteacutetico de composicioacuten
similar a la de un gas de gasificacioacuten con 5000 ppm de H2S lo que permitiacutea simular
condiciones maacutes reales para la eliminacioacuten de H2S La composicioacuten de ambos gases se muestra
en la Tabla 34
Tabla 34 Composicioacuten de los gases utilizados en las pruebas de desulfuracioacuten ( vol base seca)
Mezcla H2SN2 Gas sinteacutetico de gasificacioacuten
CO -- 100 CO2 -- 150 H2 -- 100
CH4 -- 40 C2H6 -- 02 C2H4 -- 15 C2H2 -- 02 H2S 05 05 N2 995 586
34 Materiales y meacutetodos
Dado que los gases de gasificacioacuten presentan cierto contenido en humedad en algunos
experimentos se antildeadioacute vapor de agua junto con el gas para analizar su impacto en la
capacidad desulfurante de las cenizas El caudal de agua liacutequida (0-1 gmiddoth-1) se reguloacute con una
bomba HPLC y se evaporoacute antes de su entrada al reactor El tiempo de contacto gas-soacutelido se
eligioacute en base a experimentos anteriores (Garciacutea y cols 2011) y una vez que se hubo
comprobado que no existiacutea control difusional externo en el proceso (se obtuvieron resultados
muy similares para distintos caudales de gas) La velocidad espacial del gas varioacute entre 37 y 47
h-1 (en teacuterminos de volumen) seguacuten el caudal de vapor de agua alimentado
A la salida del reactor se colocaron en serie un pequentildeo condensador y un filtro de
algodoacuten para retener la humedad del gas evitando asiacute que pudiese dantildear el cromatoacutegrafo de
gases (Agilent 3000-A) utilizado para analizar la composicioacuten del gas de salida El anaacutelisis en
modo casi continuo de la concentracioacuten del gas permite obtener las llamadas ldquocurvas de
rupturardquo para el H2S en las que se representa la evolucioacuten de su caudal (o concentracioacuten) con
el tiempo El caudal de H2S que abandona el reactor puede calcularse a partir de los datos de
composicioacuten utilizando el nitroacutegeno de los gases como estaacutendar interno debido a su caraacutecter
inerte en el proceso El tiempo de reaccioacuten establecido inicialmente fue de 120 min pero en
algunos casos el experimento se alargoacute maacutes de 300 min hasta detectar el punto de ruptura de
las curvas (momento en el que la presencia de H2S en el gas comienza a ser significativa) Como
referencia para el punto de ruptura se eligioacute una concentracioacuten de H2S de 100 ppm valor
intermedio entre los liacutemites fijados habitualmente en la literatura para por ejemplo la
aplicacioacuten del gas de gasificacioacuten en turbinas de gas (20-750 ppm) (Meng y cols 2010)
Tras los experimentos la cantidad de azufre retenido en las muestras de ceniza se
determinoacute con un analizador elemental Leco TruSpec Micro Ademaacutes a modo de ejemplo una
de las muestras fue caracterizada morfoloacutegica- y quiacutemicamente mediante otras teacutecnicas (i)
microscopiacutea electroacutenica de barrido combinada con espectroscopiacutea de energiacutea dispersiva de
rayos X (SEMEDX) y (ii) espectroscopiacutea fotoelectroacutenica de rayos X (XPS) Ambos anaacutelisis
fueron realizados por el Laboratorio de Microscopiacutea Avanzada del Instituto de Nanociencia de
Aragoacuten El anaacutelisis SEMEDX fue realizado con un microscopio FEI Inspeccione F50 sin aplicar
revestimiento metaacutelico externo a la muestra soacutelida Para el anaacutelisis EDX se utilizoacute el modo de
imagen de electrones retrodispersados Por otro lado el anaacutelisis XPS se realizoacute con un
espectroacutemetro Kratos AXIS Ultra DLD utilizando una fuente de rayos X monocromaacutetica Al Kα
(14866 eV) y una presioacuten en la caacutemara de medida de 3middot10-8 Pa
Materiales y meacutetodos 35
323 Sistema experimental para los ensayos de actividad de los catalizadores
Esta parte del trabajo experimental fue desarrollada durante una estancia de
investigacioacuten en el VTT-Technical Research Centre of Finland El objetivo del estudio fue la
evaluacioacuten de la actividad y estabilidad de varios catalizadores de niacutequel soportados sobre
aluacutemina y modificados con diferentes promotores para el reformado de compuestos modelo
de alquitraacuten en presencia de H2S Estos ensayos de actividad se realizaron en un reactor de
cuarzo de lecho fijo a escala de laboratorio (1 cm de diaacutemetro interno) operando a presioacuten
atmosfeacuterica y en un intervalo de temperatura de 700-900 ordmC La Figura 35 muestra un
esquema de la instalacioacuten experimental
Figura 35 Instalacioacuten experimental para los ensayos de actividad de los catalizadores para el reformado de compuestos modelo de alquitraacuten
El reactor de cuarzo cuenta con una placa fritada en su interior para soportar el lecho de
soacutelido En cada experimento se utilizaron 2 g de catalizador El reactor se calienta dentro de un
horno eleacutectrico ciliacutendrico y la temperatura del lecho se controla gracias a un termopar tipo K
introducido en su interior Ademaacutes para evitar condensaciones en las liacuteneas de la instalacioacuten
la temperatura de las mismas se manteniacutea siempre a 200 ordmC
Para simular el gas obtenido en el proceso de gasificacioacuten de lodos de EDAR se utilizoacute una
mezcla de gases (CO CO2 H2 N2 CH4 y C2H4) vapor de agua y compuestos modelo de alquitraacuten
habitualmente presentes en los gases de gasificacioacuten (tolueno benceno y naftaleno) con una
concentracioacuten de 15 gmiddotm-3N La composicioacuten de la mezcla utilizada se detalla en la Tabla 35 El
caudal de los distintos gases se ajustoacute mediante controladores de flujo maacutesico mientras que el
caudal de los liacutequidos (agua y mezcla de alquitraacuten) se reguloacute mediante sendas bombas HPLC El
caudal total alimentado fue de 1 LNmiddotmin-1 lo que dio lugar a una elevada velocidad espacial del
36 Materiales y meacutetodos
gas (25000 h-1 en condiciones normales de presioacuten y temperatura) con el objetivo de fomentar
la desactivacioacuten del catalizador para profundizar en su estudio
Tabla 35 Composicioacuten de la mezcla sinteacutetica de gases y vapores utilizada en los ensayos de actividad
vol H2O 300 CO 66 CO2 125 H2 136 CH4 22 C2H4 11 H2S 03 N2 334 Contenido de alquitraacuten en el gas 15 gmiddotm-3N Composicioacuten del alquitraacuten toluenonaftalenobenceno
801010 ( maacutesico)
La corriente de gases y vapores que sale del reactor se analiza en liacutenea con un
cromatoacutegrafo de gases equipado con un detector de ionizacioacuten de llama (Agilent 7980A)
calibrado para la cuantificacioacuten del benceno tolueno naftaleno metano y etileno La duracioacuten
de cada anaacutelisis era de unos 33 min A continuacioacuten los compuestos condensables se retienen
en un sistema de condensacioacuten formado por dos frascos lavadores con isopropanol y agua
respectivamente colocados en serie y refrigerados en un bantildeo de hielo Tras secar el gas se
mide su caudal y su correspondiente temperatura y se dirige a un analizador de gases (Sick
Maihak S710) que mide de forma continua la fraccioacuten volumeacutetrica de CO CO2 e H2
Antes de comenzar cada experimento se determinoacute tambieacuten la composicioacuten exacta del
gas que iba a ser alimentado al reactor para comprobar que era la correcta
Las muestras usadas de catalizador fueron caracterizadas mediante difraccioacuten de rayos X y
anaacutelisis de su composicioacuten elemental utilizando los mismos equipos que para la
caracterizacioacuten de las cenizas de lodo (seccioacuten 312)
33 Condiciones de operacioacuten y disentildeo de experimentos
331 Experimentos de gasificacioacuten
El estudio de la influencia de algunos factores de operacioacuten en la distribucioacuten de
productos y calidad del gas producto obtenido tanto en la gasificacioacuten de lodo como en la
gasificacioacuten de char se realizoacute planteando un disentildeo de experimentos factorial 2k siendo 2 el
nuacutemero de niveles (o valores) de cada factor y k el nuacutemero de factores Este disentildeo
experimental permite evaluar no soacutelo el impacto de los factores de operacioacuten sino tambieacuten la
posible existencia de interacciones entre ellos lo que significa que el efecto de un factor sobre
Materiales y meacutetodos 37
una variable respuesta estaacute condicionado por el valor de otro factor El error experimental se
evaluoacute mediante tres reacuteplicas realizadas en el punto central del disentildeo (valor medio de todos
los factores) lo que tambieacuten permite evaluar la linealidad o la curvatura en la respuesta de las
variables analizadas
Los tres factores de operacioacuten modificados durante los experimentos de gasificacioacuten
fueron en ambos casos los siguientes (i) temperatura de gasificacioacuten (770-850 ordmC) (ii)
relacioacuten gasificante (RG) que es la cantidad de agente gasificante (H2O+O2) alimentado por
unidad de masa de lodo o char en base seca y libre de cenizas (08-11 gmiddotg-1org) y (iii) la
composicioacuten del medio de gasificacioacuten representada por la relacioacuten H2OO2 (1-3 molmiddotmol-1)
Estos tres factores y sus respectivos intervalos de estudio se eligieron en base a otros trabajos
publicados en la bibliografiacutea sobre gasificacioacuten de biomasa con mezclas de aire y vapor de agua
en lecho fluidizado (Campoy y cols 2009 Gil y cols 1997 Lv y cols 2004 Pinto y cols 2003)
y teniendo en cuenta tambieacuten los resultados de estudios previos realizados en el GPT el
reacutegimen de fluidizacioacuten en el que se desea trabajar y los liacutemites operacionales de la planta
experimental Las condiciones de operacioacuten en los experimentos de gasificacioacuten se resumen en
la Tabla 36
Tabla 36 Condiciones de operacioacuten en los experimentos de gasificacioacuten de lodo y de char
Gasificacioacuten de lodo de EDAR Nuacutem exp
(T RG H2OO2) valores codificados
T (oC) RG
(gmiddotg-1org)
H2OO2 RE () SB
(g H2Omiddotg-1org)
vol O2 en el aire aire enriquecido
1 111 850 11 3 17 071 21 2 -111 770 11 3 17 071 21 3 1-11 850 08 3 12 052 21 4 -1-11 770 08 3 12 052 21 5 11-1 850 11 1 32 039 33 6 -11-1 770 11 1 32 039 33 7 1-1-1 850 08 1 23 027 27 8 -1-1-1 770 08 1 23 027 27
91011 000 810 095 2 19 052 23
Gasificacioacuten de char de piroacutelisis de lodo Nuacutem exp
(T RG H2OO2) valores codificados
T (oC) RG
(gmiddotg-1org)
H2OO2 RE () SB
(g H2Omiddotg-1org)
vol O2 en el aire aire enriquecido
12 111 850 11 3 17 071 27 13 -111 770 11 3 17 071 27 14 1-11 850 08 3 12 052 21 15 -1-11 770 08 3 12 052 21 16 11-1 850 11 1 32 039 40 17 -11-1 770 11 1 32 039 40 18 1-1-1 850 08 1 23 027 33 19 -1-1-1 770 08 1 23 027 33
202122 000 810 095 2 19 052 29
RE (relacioacuten equivalente) porcentaje del aire estequiomeacutetrico alimentado realmente SB cantidad de vapor de agua alimentado por unidad de masa de lodo o char seco y libre de cenizas
38 Materiales y meacutetodos
La segunda columna de la Tabla 36 muestra el valor de los factores en teacuterminos
codificados siendo -1 el liacutemite inferior y 1 el liacutemite superior del intervalo de estudio de cada
factor (y 0 el punto central) Esta es la forma habitual de expresar el valor de los factores en el
disentildeo de experimentos factorial 2k ya que permite una raacutepida identificacioacuten del factor con
mayor influencia sobre cada variable respuesta
Los resultados experimentales obtenidos para cada variable respuesta han sido analizados
estadiacutesticamente mediante anaacutelisis de varianza (ANOVA) Este anaacutelisis se basa en la
comparacioacuten de la varianza asociada al error experimental con la varianza ocasionada por la
modificacioacuten de los factores La comparacioacuten se realiza mediante el test F de Fischer y permite
discriminar si el efecto observado es estadiacutesticamente significativo frente al error experimental
con un inivel de confianza predeterminado (95 en este estudio) El anaacutelisis ANOVA se realizoacute
con el software Design-Expert 71
La evolucioacuten de las variables respuesta con los factores de operacioacuten puede modelarse
empiacutericamente teniendo en cuenta los efectos significativos mostrados por el anaacutelisis ANOVA
V = α+β1middotF1+β2middotF2+β3middotF3+β12middotF1middotF2+β13middotF1middotF3+β23middotF2middotF3+β123middotF1middotF2middotF3 (ec 31)
donde V representa cualquier valor de una variable respuesta α es el valor promedio de
todo el conjunto de resultados experimentales obtenidos para dicha variable Fi es el valor
codificado del factor i βi es el coeficiente asociado al factor i βij es el coeficiente asociado
a la interaccioacuten de los factores i y j(efecto sineacutergico o antagoacutenico) y β123 es un coeficiente
que representa la interaccioacuten simultaacutenea entre los tres factores
Cuando el anaacutelisis ANOVA detecta la existencia de curvatura en la respuesta de una
variable el modelo lineal no es el maacutes adecuado para predecir su evolucioacuten ante la variacioacuten
de los factores Sin embargo los coeficientes βi pueden utilizarse para evaluar la influencia
relativa de los factores cuanto mayor es el valor absoluto del coeficiente asociado al factor i
mayor es la influencia que ejerce dicho factor sobre la variable siempre que los factores esteacuten
expresados en teacuterminos codificados
332 Experimentos de desulfuracioacuten
Los factores de operacioacuten en las pruebas de desulfuracioacuten fueron los siguientes (i) tipo de
ceniza de lodo (ceniza de combustioacuten o de gasificacioacuten) (ii) temperatura (600-800 ordmC) (iii) gas
alimentado (mezcla H2SN2 o gas sinteacutetico de gasificacioacuten) y (iv) concentracioacuten de vapor de
agua en el gas alimentado (0-30 vol) lo que se traduce en una relacioacuten maacutesica H2OH2S en el
gas de 0 a 45 gH2Omiddotg-1H2S El error experimental se evaluoacute realizando tres reacuteplicas bajo la
Materiales y meacutetodos 39
atmoacutesfera gaseosa de H2SN2 y como en el estudio anterior en valores intermedios de
temperatura (700 ordmC) y de la relacioacuten maacutesica H2OH2S (225 gmiddotg-1)
Con el fin de detectar cualquier efecto secundario causado por la propia configuracioacuten
experimental como por ejemplo la retencioacuten de H2S por su reaccioacuten con las partes metaacutelicas
calientes a la entrada y a la salida del reactor se realizaron blancos sin utilizar lecho de ceniza
bajo las diferentes condiciones de temperatura y atmoacutesfera de reaccioacuten
La Tabla 37 resume las condiciones de operacioacuten en los experimentos de desulfuracioacuten
Tabla 37 Condiciones de operacioacuten en los experimentos de desulfuracioacuten
Nuacutem de experimento
Origen de la ceniza de lodo
Mezcla sinteacutetica de gas T (ordmC) H2OH2S (gmiddotg-1) Duracioacuten del
experimento (min) 1 Combustioacuten H2SN2 600 0 300 2 Combustioacuten H2SN2 800 0 390 3 Combustioacuten H2SN2 600 45 120 4 Combustioacuten H2SN2 800 45 120
567 Combustioacuten H2SN2 700 225 120 8 Gasificacioacuten H2SN2 600 0 120 9 Gasificacioacuten H2SN2 800 0 390
10 Gasificacioacuten H2SN2 600 45 120 11 Gasificacioacuten H2SN2 800 45 120
121314 Gasificacioacuten H2SN2 700 225 120 15 Sin soacutelido H2SN2 600 0 120 16 Sin soacutelido H2SN2 800 0 120 17 Sin soacutelido H2SN2 600 45 120 18 Sin soacutelido H2SN2 800 45 120 19 Sin soacutelido H2SN2 700 225 120 20 Combustioacuten Gas de gasificacioacuten 600 0 240 21 Combustioacuten Gas de gasificacioacuten 800 0 240 22 Combustioacuten Gas de gasificacioacuten 600 45 120 23 Combustioacuten Gas de gasificacioacuten 800 45 120 24 Gasificacioacuten Gas de gasificacioacuten 600 0 120 25 Gasificacioacuten Gas de gasificacioacuten 800 0 240 26 Gasificacioacuten Gas de gasificacioacuten 600 45 120 27 Gasificacioacuten Gas de gasificacioacuten 800 45 120 28 Sin soacutelido Gas de gasificacioacuten 600 0 120 29 Sin soacutelido Gas de gasificacioacuten 800 0 240 30 Sin soacutelido Gas de gasificacioacuten 600 45 120 31 Sin soacutelido Gas de gasificacioacuten 800 45 120
La influencia de los factores de operacioacuten en la cantidad de H2S eliminada del gas hasta el
punto de ruptura se analizoacute estadiacutesticamente mediante anaacutelisis ANOVA utilizando un intervalo
de confianza del 95 en el test F de Fischer
Las pruebas de desulfuracioacuten realizadas en el laboratorio se simularon tambieacuten de forma
teoacuterica determinando asiacute la maacutexima cantidad de H2S que podiacutea ser eliminada del gas desde un
punto de vista termodinaacutemico Para ello se utilizoacute el software HSC Chemistry 61 Este
40 Materiales y meacutetodos
programa utiliza el meacutetodo de minimizacioacuten de la energiacutea de Gibbs para calcular la cantidad de
cada producto en el equilibrio y en condiciones isoteacutermicas e isobaacutericas El sistema de reaccioacuten
debe ser especificado para los caacutelculos incluyendo la temperatura presioacuten cantidad de
reactivos y posibles especies que cabriacutea esperar como productos finales Como reactivo soacutelido
inicial se consideroacute soacutelo el contenido de hierro y de calcio de la ceniza suponiendo que todo el
contenido metaacutelico se encontraba en forma de Fe2O3 y CaO respectivamente Entre las
posibles especies que podiacutean formar parte de los productos se consideraron las siguientes
H2S SO2 COS S Ca CaO CaS CaCO3 CaSO4 CaSO3 Fe FexOy FexSy Fex(SO4)y y Fex(SO3)y
ademaacutes de los propios compuestos gaseosos alimentados con cada mezcla gaseosa
333 Ensayos de actividad de los catalizadores de niacutequel
Los factores de estudio en los ensayos de actividad de los catalizadores de niacutequel fueron
los siguientes (i) promotor antildeadido (Ca Cu Fe o Mn) (ii) temperatura final de calcinacioacuten de
los catalizadores (700 o 900 ordmC) y (iii) procedimiento de reduccioacuten de los catalizadores ya que
en algunos casos los catalizadores se redujeron bajo una atmoacutesfera de H2 antes de comenzar
los ensayos de actividad para transformar el NiO en Ni (900 ordmC durante 1 h con un caudal de 1
LNmiddotmin-1 de una mezcla H2N2 al 50 vol) mientras que en otros casos no se realizoacute esta
reduccioacuten previa La Tabla 38 resume las condiciones de operacioacuten en los ensayos de actividad
de los catalizadores de niacutequel
Tabla 38 Condiciones de operacioacuten en los ensayos de actividad de los catalizadores
Nuacutem de experimento Catalizador Temperatura final
de calcinacioacuten (ordmC) Reduccioacuten previa del
catalizador con H2 1 NiAl2O3 900 siacute 2 NiCaAl2O3 900 siacute 3 NiCuAl2O3 900 siacute 4 NiFeAl2O3 900 siacute 5 NiMnAl2O3 900 siacute 6 NiAl2O3 900 no 7 NiCaAl2O3 900 no 8 NiCuAl2O3 900 no 9 NiFeAl2O3 900 no
10 NiMnAl2O3 900 no 11 NiAl2O3 700 no 12 NiCaAl2O3 700 no 13 NiCuAl2O3 700 no 14 NiFeAl2O3 700 no 15 NiMnAl2O3 700 no
El posible craqueo teacutermico de los alquitranes tambieacuten se tuvo en cuenta realizando un
blanco en el que se utilizoacute un material inerte como lecho soacutelido (SiC)
Materiales y meacutetodos 41
El efecto de la temperatura sobre la actividad del catalizador se estudioacute en todos los casos
modificando la temperatura de reaccioacuten de acuerdo con la siguiente rampa 900-850-800-900-
750-700-900 ordmC Cada temperatura se mantuvo durante 35 h por lo que la duracioacuten total de
los experimentos fue de 245 h La repeticioacuten de varios tramos a 900 ordmC permite evaluar la tasa
de desactivacioacuten del catalizador comparando el nivel de conversioacuten que se alcanza en todos
ellos
42 Resultados y discusioacuten
4 RESULTADOS Y DISCUSIOacuteN
En esta seccioacuten se muestran y se discuten los principales resultados obtenidos en los
diferentes estudios que componen la presente Tesis Los resultados experimentales de la
gasificacioacuten de lodos de EDAR y de la gasificacioacuten de char se presentan en la seccioacuten 41
comparando los resultados obtenidos en ambos procesos y analizando la influencia de los
factores de operacioacuten En la seccioacuten 42 se realiza una evaluacioacuten energeacutetica de la gasificacioacuten
de ambos materiales analizando tambieacuten la demanda energeacutetica del proceso de piroacutelisis en el
que se produce el char y del secado teacutermico previo de los lodos para tener una idea global del
rendimiento energeacutetico de ambos procesos (i) secado y gasificacioacuten del lodo y (ii) secado
piroacutelisis del lodo y gasificacioacuten del char Las secciones 43 y 44 muestran los resultados de los
tratamientos secundarios de limpieza aplicados a diferentes gases sinteacuteticos que simulan el gas
de gasificacioacuten estudio de la eliminacioacuten de H2S mediante el uso de las cenizas obtenidas en la
combustioacuten y gasificacioacuten del lodo (seccioacuten 43) y ensayos de actividad de diferentes
catalizadores de niacutequel en el reformado de compuestos modelo de alquitraacuten (seccioacuten 44)
41 Gasificacioacuten de lodo y de char
El gas es el producto de intereacutes de la gasificacioacuten por lo que la mayoriacutea de las variables
respuesta analizadas estaacuten relacionadas con este producto rendimiento o produccioacuten total de
gas seco composicioacuten del gas (relaciones H2CO y CO2CO) rendimiento o produccioacuten de cada
compuesto gaseoso contenido de alquitraacuten en el gas poder caloriacutefico del gas y eficiencia
energeacutetica de la gasificacioacuten El rendimiento soacutelido y la distribucioacuten del carbono inicial entre los
diferentes productos (soacutelido gas y alquitraacuten) tambieacuten fueron determinados despueacutes de los
experimentos Los resultados de todas estas variables respuesta obtenidos en la gasificacioacuten
del lodo de EDAR se resumen en la Tabla 41 en funcioacuten de las condiciones de operacioacuten
(temperatura relacioacuten gasificante y relacioacuten H2OO2)
Resultados y discusioacuten 43
Tabla 41 Resultados obtenidos en los experimentos de gasificacioacuten de lodo de EDAR
Temperatura (ordmC) 850 770 850 770 850 770 850 770 810
Relacioacuten gasificante RG (gmiddotg-1
org) 11 11 08 08 11 11 08 08 095
Relacioacuten molar H2OO2 3 3 3 3 1 1 1 1 2
Rendimiento a soacutelido (gmiddotkg-1
lodo) 368 401 401 407 356 392 384 400 382 plusmn 1
Fraccioacuten de C remanente en el soacutelido () 3 17 8 23 2 9 2 12 7 plusmn 1
Fraccioacuten de C convertido en gas () 764 613 651 618 897 748 831 657 731 plusmn 08
Fraccioacuten de C convertido en alquitraacuten () 4 7 5 5 4 7 4 6 5 plusmn 1
Rendimiento a gas seco (m3Nmiddotkg-1
lodo sin N2) 072 051 065 053 072 052 071 049 061 plusmn 001
Rendimiento a gas seco (m3Nmiddotkg-1
org sin N2) 132 094 120 097 132 096 130 089 113 plusmn 001
Contenido de alquitraacuten en el gas (gmiddotm-3N) 19 44 19 44 12 22 11 45 15 plusmn 1
Composicioacuten del gas (base seca)
H2 ( vol) 242 184 251 204 180 110 206 136 193 plusmn 01
CO ( vol) 87 57 102 73 116 70 141 77 94 plusmn 01 CO2 ( vol) 171 186 126 155 207 238 160 198 181 plusmn 02
CH4 ( vol) 31 35 36 41 26 28 29 34 33 plusmn 01
C2Hx ( vol) 17 21 14 22 13 16 14 20 17 plusmn 02 H2S ( vol) 044 038 033 033 044 042 038 031 040 plusmn 002
N2 ( vol) 449 514 468 502 453 534 445 532 478 plusmn 02
Produccioacuten de cada compuesto gaseoso (gmiddotkg-1
org)
H2 518 318 504 365 386 201 431 232 372 plusmn 04 CO 260 137 287 182 351 179 414 183 253 plusmn 2
CO2 806 707 557 608 980 962 736 740 767 plusmn 7 CH4 53 49 59 59 45 41 49 46 51 plusmn 1 C2Hx 50 52 40 55 40 42 40 48 47 plusmn 4 H2S 160 112 114 100 161 131 135 90 130 plusmn 08
Relacioacuten molar H2CO en el gas producto 279 325 246 281 154 157 146 177 206 plusmn 001
Relacioacuten molar COCO2
en el gas producto 051 030 081 047 056 029 088 039 052 plusmn 001
PCI del gas (MJmiddotm-3N) 59 53 62 60 52 41 60 49 56 plusmn 01
Eficiencia energeacutetica de la gasificacioacuten () 658 477 647 553 584 391 647 434 558 plusmn 08
Los datos de la uacuteltima columna representan la media plusmn desviacioacuten estaacutendar de las 3 reacuteplicas del punto central
El rendimiento soacutelido (masa de producto soacutelido obtenida por unidad de masa de lodo
alimentado) varioacute entre 356 y 407 gmiddotkg-1lodo Los valores tiacutepicos para otros tipos de biomasa
como madera o paja se encuentran habitualmente por debajo de 80 gmiddotkg-1 (McKendry 2002a)
El alto contenido de ceniza en el lodo (39 en masa) explica esta diferencia En algunos casos
44 Resultados y discusioacuten
la cantidad de soacutelido recogido despueacutes de los experimentos fue menor que el propio contenido
de ceniza del lodo lo que sugiere la volatilizacioacuten de una pequentildea fraccioacuten inorgaacutenica del lodo
durante la gasificacioacuten
La distribucioacuten del contenido inicial de carbono en el lodo entre los diferentes productos
fue la siguiente (i) entre un 2 y un 23 del carbono se mantuvo en el soacutelido como
consecuencia de su incompleta conversioacuten (calculado a partir de los datos de rendimiento a
soacutelido y anaacutelisis elemental del mismo) (ii) la fraccioacuten de carbono en forma de alquitraacuten varioacute
entre un 4 y un 7 (calculado a partir de los datos de carbono orgaacutenico total presente en el
condensado) y (iii) la fraccioacuten de carbono convertido en gases no condensables fue la
mayoritaria variando entre un 613 y un 897 (calculado a partir del rendimiento a gas y la
composicioacuten del mismo) Con estos datos el balance de masa al carbono cierra al 78-95 La
posible formacioacuten de hidrocarburos ligeros no detectados por el cromatoacutegrafo de gases (C3Hx
C4Hx) o la escasa solubilidad de algunos compuestos del alquitraacuten en las disoluciones acuosas
preparadas para la determinacioacuten del carbono orgaacutenico total puede explicar la falta de
carbono en el cierre de los balances
El rendimiento a gas seco definido como el volumen de gas seco y libre de alquitraacuten
producido por kilogramo de lodo orgaacutenico es decir en base seca y libre de ceniza varioacute entre
089 y 132 m3Nmiddotkg-1org (en base libre de N2) Estos datos de produccioacuten de gas no difieren
demasiado de los obtenidos por otros autores al gasificar otros tipos de biomasa bajo similares
condiciones de operacioacuten (Campoy y cols 2009 Gil y cols 1999a Pinto y cols 2003) Como
es habitual en un proceso de gasificacioacuten los principales compuestos gaseosos producidos
durante la gasificacioacuten de lodos de EDAR fueron H2 CO CO2 e hidrocarburos ligeros siendo el
CH4 el hidrocarburo mayoritario Tambieacuten cabe destacar la formacioacuten de H2S debido a la
presencia de un 13 de azufre en el lodo (Tabla 31) Ademaacutes como consecuencia de la
alimentacioacuten del aire como parte del agente gasificante el N2 representoacute un 45-55 del
volumen final de gas En todos los experimentos se alimentoacute una cantidad muy similar de N2
para evitar diferentes efectos de dilucioacuten del gas producto que pudiesen esconder o modificar
el verdadero efecto de los factores de operacioacuten en determinadas variables respuesta
La composicioacuten del gas ha mostrado importantes diferencias en funcioacuten de las
condiciones de operacioacuten Tanto es asiacute que los porcentajes volumeacutetricos de H2 (110-251)
CO (57-141) CO2 (126-238) y CH4 (26-41) pueden llegar a duplicarse al modificar las
condiciones de operacioacuten Estos datos de composicioacuten dan lugar a unas relaciones molares
H2CO y COCO2 de 146-325 y 029-088 respectivamente La relacioacuten molar H2CO es un
paraacutemetro importante de cara al posible uso del gas como gas de siacutentesis valores de 2-3
Resultados y discusioacuten 45
suelen ser necesarios en procesos como la produccioacuten de metanol o la siacutentesis de liacutequidos
Fischer Tropsch (Wender 1996) Por otro lado la relacioacuten COCO2 da una idea de la
distribucioacuten del carbono entre ambos compuestos Valores altos de la relacioacuten COCO2 son
preferibles desde el punto de vista del contenido energeacutetico del gas
El contenido de alquitraacuten en el gas varioacute desde 11 hasta 45 gmiddotm-3N Los valores maacutes altos
(22-45 gmiddotm-3N) se corresponden con la temperatura de operacioacuten maacutes baja (770 ordmC) mientras
que los valores maacutes bajos (11-12 gmiddotm-3N) se encuentran entre los valores habituales obtenidos
en la gasificacioacuten de biomasa en lecho fluidizado (Han y Kim 2008)
El poder caloriacutefico inferior del gas (PCI) se calculoacute como Σ (ximiddotPCIi) donde xi y PCIi son la
fraccioacuten volumeacutetrica y el poder caloriacutefico inferior de cada compuesto del gas respectivamente
El PCI del gas osciloacute entre 41 y 62 MJmiddotm-3N valor suficiente para su combustioacuten en turbinas o
motores (Bridgwater 1995)
La eficiencia energeacutetica de la gasificacioacuten se define como el cociente entre la energiacutea
contenida en el gas friacuteo (producto del rendimiento a gas por su PCI sin tener en cuenta el calor
sensible del gas) y la energiacutea contenida en el lodo (PCI) La eficiencia energeacutetica asiacute calculada
varioacute en un amplio intervalo de valores desde 391 hasta 658
Los resultados experimentales correspondientes a la gasificacioacuten del char obtenidos bajo
las mismas condiciones de operacioacuten que en la gasificacioacuten del lodo se muestran en la Tabla
42
46 Resultados y discusioacuten
Tabla 42 Resultados obtenidos en los experimentos de gasificacioacuten de char
Temperatura (ordmC) 850 770 850 770 850 770 850 770 810
Relacioacuten gasificante RG (gmiddotg-1
org) 11 11 08 08 11 11 08 08 095
Relacioacuten molar H2OO2 3 3 3 3 1 1 1 1 2
Rendimiento a soacutelido (gmiddotkg-1
char) 757 785 750 785 731 771 752 813 775 plusmn 2
Fraccioacuten de C remanente en el soacutelido () 20 41 25 43 15 26 19 41 34 plusmn 3
Fraccioacuten de C convertido en gas () 71 56 62 48 83 67 72 56 62 plusmn 2
Fraccioacuten de C convertido en alquitraacuten () 13 07 29 33 10 57 32 58 28 plusmn 07
Rendimiento a gas seco (m3Nmiddotkg-1
char sin N2) 036 027 031 024 035 028 032 024 029 plusmn 001
Rendimiento a gas seco (m3Nmiddotkg-1
org sin N2) 147 112 130 099 146 115 131 100 121 plusmn 001
Contenido de alquitraacuten en el gas (gmiddotm-3N) 4 3 9 13 3 20 10 22 9 plusmn 2
Composicioacuten del gas (base seca)
H2 ( vol) 293 263 278 248 215 190 220 202 252 plusmn 06
CO ( vol) 195 120 202 128 227 140 237 152 159 plusmn 02
CO2 ( vol) 189 242 162 208 226 295 185 241 219 plusmn 01
CH4 ( vol) 076 091 077 092 059 070 064 084 088 plusmn 001
C2Hx (ppm 150 190 160 220 180 220 150 200 180 plusmn 10
H2S ( vol) 025 012 014 007 017 008 012 006 010 plusmn 001
N2 ( vol) 313 365 349 407 325 367 350 396 361 plusmn 06 Produccioacuten de cada compuesto gaseoso (gmiddotkg-1
org)
H2 567 421 499 368 416 308 397 300 428 plusmn 03 CO 529 268 506 266 615 318 598 316 379 plusmn 10
CO2 808 853 637 679 960 1055 735 786 821 plusmn 20
CH4 117 117 110 109 91 90 92 99 119 plusmn 01 C2Hx 041 042 041 047 050 050 039 042 042 plusmn 004
H2S 82 32 41 18 57 23 37 15 28 plusmn 01
Relacioacuten molar H2CO en el gas producto 150 220 138 193 095 136 093 133 158 plusmn 004
Relacioacuten molar COCO2 en el gas producto 103 049 125 062 100 047 128 063 073 plusmn 001
PCI del gas (MJmiddotm-3N) 596 471 587 465 544 409 563 443 507 plusmn 007
Eficiencia energeacutetica de la gasificacioacuten () 629 411 572 376 574 362 553 357 470 plusmn 06
Los datos de la uacuteltima columna representan la media plusmn desviacioacuten estaacutendar de las 3 reacuteplicas del punto central
La cantidad de alquitraacuten se ha aproximado a la cantidad de carbono orgaacutenico detectado en el condensado
El rendimiento a soacutelido obtenido en la gasificacioacuten de char (731-813 gmiddotkg-1char)
praacutecticamente se duplicoacute respecto al obtenido en la gasificacioacuten de lodo debido al mayor
Resultados y discusioacuten 47
contenido de materia inorgaacutenica en el char Seguacuten las especificaciones de la norma ISO-1171-
1976 el 93-96 de este producto soacutelido era ceniza
En cuanto a la distribucioacuten del contenido inicial de carbono en el char la fraccioacuten de
carbono remanente en el soacutelido osciloacute entre un 15 y un 43 mientras que el valor maacuteximo
obtenido en la gasificacioacuten del lodo fue de aproximadamente un 23 (Tabla 41) Esta
diferencia radica en la diferente estructura carbonosa del soacutelido Soacutelo el 15 del contenido de
carbono en el lodo se encuentra en forma de carbono fijo mientras que este valor alcanza el
59 en el caso del char (Tabla 31) Las reacciones en las que se ve envuelto el carbono soacutelido
son mucho maacutes lentas que la liberacioacuten de la materia volaacutetil y las reacciones en fase gas lo que
supone una reduccioacuten de la conversioacuten total del carbono y por tanto de la fraccioacuten de
carbono convertido a gas durante la gasificacioacuten de char La fraccioacuten de carbono convertido en
alquitraacuten durante la gasificacioacuten de char tambieacuten se redujo en comparacioacuten con los resultados
obtenidos en la gasificacioacuten de lodo aunque no en la misma proporcioacuten que la reduccioacuten
observada en el contenido de materia volaacutetil de ambos soacutelidos (maacutes de tres veces menor en el
char que en el lodo Tabla 31) Esto demuestra que parte de la materia volaacutetil desprendida del
lodo se descompone y reacciona para formar gases maacutes ligeros
Los datos de conversioacuten de carbono durante la gasificacioacuten de char pueden recalcularse
considerando conjuntamente la etapa previa de piroacutelisis de lodo y la posterior gasificacioacuten del
char Para ello ha de utilizarse como base de caacutelculo la cantidad de carbono alimentado
inicialmente al proceso de piroacutelisis Teniendo en cuenta que el rendimiento a char en la
piroacutelisis de lodo se situoacute en torno a un 52 en masa (Gil-Lalaguna y cols 2010) la fraccioacuten del
carbono inicial que queda en forma de soacutelido despueacutes de la piroacutelisis del lodo y la gasificacioacuten
del char se reduce a un 4-11 lo que supone una mejora respecto a la gasificacioacuten directa del
lodo bajo determinadas condiciones de operacioacuten
El rendimiento a gas seco en la gasificacioacuten de char varioacute entre 024 y 036 m3Nmiddotkg-1char
(base libre de N2) o entre 040 y 052 m3Nmiddotkg-1char si se incluye la cantidad de N2 en el volumen
de gas Estos datos suponen un reduccioacuten a la mitad de la produccioacuten de gas por kilogramo de
materia prima en comparacioacuten con la gasificacioacuten de lodo (051-072 m3Nmiddotkg-1lodo en base libre
de N2) El rendimiento a gas obtenido durante el proceso previo de piroacutelisis de lodo (007
m3STPmiddotkg-1
lodo) no compensa esta diferencia ya que dicho proceso fue optimizado para
maximizar la fraccioacuten de liacutequido Sin embargo si la produccioacuten de gas en las etapas de
gasificacioacuten se calcula en base seca y libre de ceniza para el soacutelido la gasificacioacuten de char (099-
147 m3Nmiddotkg-1org) ofrece mejores resultados que la gasificacioacuten de lodo (089-132 m3Nmiddotkg-1
org)
Esto se debe a una mayor concentracioacuten del carbono en la fraccioacuten orgaacutenica del soacutelido despueacutes
48 Resultados y discusioacuten
de la piroacutelisis (064 g Cmiddotg-1org en el char frente a 054 g Cmiddotg-1
org en el lodo) La produccioacuten de gas
por kilogramo de materia orgaacutenica en el char de lodo de EDAR se encuentra en el mismo orden
de magnitud que cuando se gasifica char de origen lignoceluloacutesico como char de bagazo
(Chaudhari y cols 2003) o de ramio (He y cols 2012)
El contenido de alquitraacuten en el gas derivado de la gasificacioacuten de char se redujo a niveles
de 3-4 gmiddotm-3N bajo ciertas condiciones de operacioacuten mientras que el valor maacutes bajo alcanzado
en la gasificacioacuten de lodos de depuradora fue de 11-12 gmiddotm-3N
En cuanto a la composicioacuten del gas el H2 (190-293 vol) CO (120-237 vol) CO2
(162-295 vol) CH4 (059-092 vol) y N2 (313-407 vol) fueron los compuestos
mayoritarios detectados por el cromatoacutegrafo de gases La produccioacuten de CO (en teacuterminos de
gmiddotkg-1org) se vio claramente incrementada al gasificar char en lugar de lodo (45-85 mayor) lo
que puede estar relacionado con su mayor contenido de carbono fijo y con una mayor
extensioacuten de las reacciones water-gas (ec 23) y de Boudouard (ec 24) La produccioacuten de CH4
en la gasificacioacuten de char se redujo en un 75-82 en comparacioacuten con la gasificacioacuten de lodo
(en teacuterminos de gmiddotkg-1org) mientras que las variaciones en la produccioacuten de H2 y CO2 no fueron
tan significativas La relacioacuten molar H2CO en el gas producto de la gasificacioacuten de char (093-
220) fue menor que la obtenida en la gasificacioacuten de lodo (146-325) mientras que la relacioacuten
COCO2 fue mayor en la gasificacioacuten de char (047-128) Soacutelo en la gasificacioacuten de char y bajo
determinadas condiciones de operacioacuten se consigue favorecer la formacioacuten de CO frente a la
de CO2 (COCO2 gt 1)
Tanto el PCI del gas (409-596 MJmiddotm-3N) como la eficiencia energeacutetica (362-629) de la
gasificacioacuten de char oscilan en el mismo intervalo que los valores obtenidos en la gasificacioacuten
del lodo
La influencia de la temperatura (T) relacioacuten gasificante (RG) y composicioacuten del medio de
gasificacioacuten (relacioacuten H2OO2) en los resultados obtenidos en la gasificacioacuten de lodo y de char
ha sido estadiacutesticamente evaluada mediante anaacutelisis ANOVA Las Tablas 43 y 44 muestran los
coeficientes de regresioacuten lineal obtenidos en el anaacutelisis ANOVA de las variables respuesta de
ambos procesos Estos coeficientes (β) obtenidos para valores codificados de los factores
resultan uacutetiles para modelar la respuesta de las variables en el caso de una evolucioacuten lineal (ec
31) y para determinar la influencia relativa de los factores es decir para establecer cuaacutel es el
factor con mayor impacto en cada variable
Resultados y discusioacuten 49
Tabla 43 Coeficientes de regresioacuten lineal (β) para las variables respuesta en la gasificacioacuten de lodo
Valor medio βT βRG βH2OO2 βT-RG βT-H2OO2 βRG-H2OO2 βT-H2OO2-RG Curvatura
Fraccioacuten de C remanente en el soacutelido () 901 -576 -174 329 -150
Fraccioacuten de C convertido en gas () 7248 633 333 -607
Rendimiento a gas seco (m3Nmiddotkg-1
org sin N2) 112 017 002 002 -002 003
Contenido de alquitraacuten (gmiddotm-3N) 2703 -1191 -278 435 315 262 -289
Produccioacuten de cada compuesto gaseoso (gmiddotkg-1org)
H2 3703 906 -137 568 CO 25010 7879 -1751 -3267 -503 -2195 965
CO2 76342 10157 -9253 2159 -1493 1591 CH4 5023 149 -313 480 C2Hx 4613 -328 335 H2S 1266 173 156 062
Relacioacuten molar H2CO en el gas producto 221 -014 008 062 002 -006 011 -005
Relacioacuten molar COCO2 en el gas producto 052 016 -011 -004 -003
PCI del gas (MJmiddotm-3N) 549 037 -031 040 -017 Eficiencia energeacutetica de la gasificacioacuten () 5512 851 347
teacutermino no significativo curvatura significativa
Tabla 44 Coeficientes de regresioacuten lineal (β) para las variables respuesta en la gasificacioacuten de char
Valor medio βT βRG βH2OO2 βT-RG βT-H2OO2 βRG-H2OO2 βT-H2OO2-RG Curvatura
Fraccioacuten de C remanente en el soacutelido () 3013 -916 -330 359
Fraccioacuten de C convertido en gas () 6429 776 482 -516
Rendimiento a gas seco (m3Nmiddotkg-1
org sin N2) 123 016 008
Contenido de alquitraacuten (gmiddotm-3N) 1052 -397 -315 -340 345
Produccioacuten de cada compuesto gaseoso (gmiddotkg-1org)
H2 4093 603 186 543 033 090 118
CO 42716 13509 -3461 -978 CO2 81596 -2909 10491 -6995 -1875
CH4 1033 101 032 C2Hx 044 -003 H2S 354 162 103
Relacioacuten molar H2CO en el gas producto 148 -026 031
Relacioacuten molar COCO2 en el gas producto 085 029 -010 -003 001
PCI del gas (MJmiddotm-3N) 509 063 020 008 Eficiencia energeacutetica de la gasificacioacuten () 4790 1027 145 177 048 083
teacutermino no significativo curvatura significativa
50 Resultados y discusioacuten
La temperatura es el factor maacutes influyente en la fraccioacuten del carbono inicial que queda en
el subproducto soacutelido despueacutes de los procesos de gasificacioacuten Esta fraccioacuten de carbono se
reduce praacutecticamente a la mitad al aumentar la temperatura de gasificacioacuten desde 770 a 850
ordmC (Figura 41) Por otro lado el coeficiente positivo asociado a la la influencia de la relacioacuten
H2OO2 (βH2OO2 gt 0) sugiere una mayor reactividad del carbono con el oxiacutegeno que con el vapor
de agua Nowicki y cols (2011) mostraron un resultado similar al realizar ensayos de
gasificacioacuten de char de lodo en una termobalanza bajo diferentes atmoacutesferas gaseosas (CO2
H2O y O2) ya que la presencia de oxiacutegeno en el medio de gasificacioacuten mejoraba la conversioacuten
del carbono En el caso de la gasificacioacuten de lodo el efecto de la composicioacuten del medio de
gasificacioacuten se reduce considerablemente a alta temperatura (Figura 41a) Ademaacutes como era
de esperar el aumento de la relacioacuten gasificante (RG) conlleva una reduccioacuten de la fraccioacuten de
carbono que queda como subproducto soacutelido despueacutes de la gasificacioacuten (βRG lt 0)
(a) Gasificacioacuten de lodo (b) Gasificacioacuten de char
Figura 41 Fraccioacuten del carbono inicial que queda como subproducto soacutelido despueacutes de la (a) gasificacioacuten de lodo y (b) gasificacioacuten de char (RG = 095 gmiddotg-1
org) Las barras de error en las figuras representan la miacutenima diferencia significativa
Los resultados experimentales muestran una clara relacioacuten entre la fraccioacuten de carbono
remanente en el soacutelido y la fraccioacuten de carbono convertido a gas cuanto menor es la primera
mayor es la segunda Por lo tanto como se muestra en la Figura 42 la fraccioacuten de carbono
convertido a gas se ve afectada positivamente por la temperatura (βT gt 0) y por la relacioacuten
gasificante (βRG gt 0) y negativamente afectada por la relacioacuten H2OO2 (βH2O O2 lt 0) en ambos
casos
770 810 850 0
10
20
30
40
50
Temperatura (ordmC)
Frac
cioacuten
de
C re
man
ente
en
el soacute
lido
()
770 810 850 0
10
20
30
40
50
Frac
cioacuten
de
C re
man
ente
en
el soacute
lido
()
Temperatura (ordmC)
H2OO2 = 3 H2OO2 = 2 H2OO2 = 1
Resultados y discusioacuten 51
(a) Gasificacioacuten de lodo (b) Gasificacioacuten de char
Figura 42 Fraccioacuten del carbono inicial convertido a gas en la (a) gasificacioacuten de lodo y (b) gasificacioacuten de char (RG = 095 gmiddotg-1
org) Las barras de error en las figuras representan la miacutenima diferencia significativa
Como muestran los coeficientes β de las Tablas 43 y 44 la temperatura es el factor con
mayor impacto en el rendimiento a gas afectaacutendole de forma positiva (βT gt 0) Durante el
proceso de gasificacioacuten los compuestos gaseosos se producen en diferentes etapas que se ven
favorecidas por el aumento de la temperatura como la etapa inicial de piroacutelisis el craqueo y
reformado de los alquitranes y las reacciones de gasificacioacuten del carbono soacutelido que son
endoteacutermicas (Pinto y cols 2003) El aumento de la relacioacuten gasificante tambieacuten resulta
favorable para la produccioacuten de gas (βRG gt 0) especialmente en la gasificacioacuten de char
mientras que la composicioacuten de la atmoacutesfera reactiva no ejerce un efecto significativo en la
produccioacuten de gas en la gasificacioacuten de ninguno de los dos materiales Por lo tanto el efecto
negativo antes mencionado de la relacioacuten H2OO2 sobre la fraccioacuten de carbono convertido a
gas no se traduce en una variacioacuten significativa de la produccioacuten total de gas Esto se explica
teniendo en cuenta tambieacuten la produccioacuten de H2 que como se discute maacutes adelante se ve
favorecida con el aumento de la relacioacuten H2OO2 contrarrestando asiacute la disminucioacuten de la
produccioacuten de gases carbonosos Ademaacutes de estas influencias los factores han mostrado
efectos sineacutergicos y antagoacutenicos estadiacutesticamente significativos pero mucho menos
importantes que el efecto individual de la temperatura de gasificacioacuten
La temperatura es de nuevo el factor maacutes influyente en el contenido de alquitraacuten en el
gas producto de la gasificacioacuten de lodo mientras que los tres factores estudiados ejercen
efectos similares durante la gasificacioacuten de char El aumento de temperatura favorece tanto la
produccioacuten total de gas como la descomposicioacuten de los alquitranes mediante reacciones de
reformado lo que se traduce en una menor concentracioacuten de alquitraacuten en el gas producto El
contenido de alquitraacuten en el gas tambieacuten puede reducirse mediante el aumento de la relacioacuten
770 810 850
50
60
70
80
90
Temperatura (ordmC)
Frac
cioacuten
de
C co
nver
tido
a ga
s (
)
770 810 850
50
60
70
80
90 Fr
acci
oacuten d
e C
conv
ertid
o a
gas (
)
Temperatura (ordmC)
H2OO2 = 1 H2OO2 = 3 H2OO2 = 2
52 Resultados y discusioacuten
gasificante (βRG lt 0) aunque este efecto praacutecticamente desaparece cuando se opera a alta
temperatura o con una alta relacioacuten H2OO2 durante la gasificacioacuten del lodo La relacioacuten
H2OO2 utilizada como agente gasificante ha mostrado efectos opuestos sobre el contenido de
alquitraacuten en los gases producidos durante la gasificacioacuten de lodo y de char mostrando un
efecto positivo en el primer caso (βH2OO2 gt 0) y un efecto negativo en el segundo (βH2OO2 lt 0)
En este uacuteltimo caso tal como se muestra en la Figura 43 existe una clara interaccioacuten entre la
temperatura y la relacioacuten H2OO2 utilizada como agente gasificante ya que el efecto de cada
uno de ellos praacutecticamente desaparece al operar con el maacuteximo valor del otro
Figura 43 Contenido de alquitraacuten (gmiddotm-3N) en el gas producto de la gasificacioacuten de char (RG=11 gmiddotg-1org)
El rendimiento o produccioacuten neta (gmiddotkg-1org) de los diferentes compuestos que forman
parte del gas producto puede calcularse a partir de los datos de volumen total y composicioacuten
del gas Como muestran los coeficientes β mostrados en las Tablas 43 y 44 la temperatura de
gasificacioacuten es de nuevo el factor maacutes influyente en la produccioacuten de H2 y de CO Estos gases
estaacuten involucrados en diversas reacciones tanto en forma de reactivos como de productos
pero el aumento de la temperatura favorece su produccioacuten frente a su consumo (βT gt 0) El
reformado del alquitraacuten es una de las reacciones que contribuye a la formacioacuten de CO y H2 a
alta temperatura (ecs 29 y 210) La evolucioacuten de la produccioacuten de CO con la temperatura ha
mostrado variaciones insignificantes o incluso la tendencia opuesta en algunos estudios de la
bibliografiacutea (Gil y cols 1997 Lv y cols 2004) lo que demuestra la importancia de la
naturaleza de la biomasa y de las demaacutes condiciones de operacioacuten en la evolucioacuten de este
compuesto Aunque en menor medida la composicioacuten de la atmoacutesfera reactiva tambieacuten ejerce
un efecto significativo en la produccioacuten de H2 y de CO El aumento de la relacioacuten H2OO2
conlleva un aumento de la produccioacuten de H2 (βH2OO2 gt 0) y un descenso de la produccioacuten de CO
770 790
810 830
850
10 15
20 25
30
3
7
11
15
19
Con
teni
do d
e al
quitr
aacuten e
n el
gas
T (ordmC) H2OO2
Resultados y discusioacuten 53
(βH2OO2 lt 0) Ambas tendencias son consistentes con la reaccioacuten water-gas shift (ec 26) que
es una de las reacciones maacutes influyentes en los procesos de gasificacioacuten con vapor de agua a
temperaturas no demasiado altas hasta unos 830 ordmC (Franco y cols 2003) y que se ve
favorecida al aumentar la presencia de vapor de agua en el medio de reaccioacuten A temperaturas
maacutes altas (830-900 ordmC) las reacciones gas-soacutelido como la reaccioacuten water-gas (ec 23) o la
reaccioacuten de Boudouard (ec 24) toman maacutes importancia (Franco y cols 2003) lo que
contribuye a explicar el efecto positivo de la temperatura sobre la formacioacuten de CO (Figura
44a) Aunque el aumento de la relacioacuten H2OO2 conlleva ademaacutes de una mayor presencia de
vapor de agua una reduccioacuten de la disponibilidad de oxiacutegeno en el medio de gasificacioacuten y una
atenuacioacuten de las reacciones de combustioacuten completa el efecto negativo de la relacioacuten H2OO2
en la produccioacuten neta de CO sugiere que el vapor de agua juega un papel maacutes importante que
el oxiacutegeno en su consumo Ademaacutes como se muestra en la Figura 44a el efecto positivo de la
temperatura se ve aminorado con el aumento de la presencia de vapor de agua como
consecuencia del desplazamiento de la reaccioacuten water-gas shift hacia el consumo de CO
(a) Produccioacuten de CO en la gasificacioacuten de lodo (b) Produccioacuten de CO2 en la gasificacioacuten de lodo
Figura 44 Produccioacuten neta de (a) CO (gmiddotkg-1org) y (b) CO2 (gmiddotkg-1
org) en la gasificacioacuten de lodo
La relacioacuten gasificante (RG) ejerce la mayor influencia en la produccioacuten de CO2
afectaacutendole de forma positiva (βRG gt 0) El aumento de RG conlleva una mayor disponibilidad
de oxiacutegeno y de vapor de agua en el medio de reaccioacuten por lo que la produccioacuten de CO2 en
reacciones de combustioacuten y en reacciones promovidas por la presencia de vapor de agua
(como por ejemplo la reaccioacuten water-gas shift) se ve favorecida El efecto negativo de la
relacioacuten H2OO2 en la produccioacuten de CO2 (βH2OO2 lt 0) revela que las reacciones de combustioacuten
son la principal fuente para la formacioacuten de este compuesto Como se observa en la Figura
44b la mayor produccioacuten de CO2 se corresponde con la mayor presencia de oxiacutegeno en el
770 790
810 830
850
10
15
20
25
30
150
210
270
330
390
Pro
ducc
ioacuten
de C
O (g
kg
lodo
org
aacutenic
o)
T (ordmC) H2OO2
080 088
095 103
110
10
15
20
25
30
580
680
780
880
980
Pro
ducc
ioacuten
de C
O2
(gk
g lo
do o
rgaacuten
ico)
RG (gg org)
H2OO2
54 Resultados y discusioacuten
medio que se obtiene con el mayor valor de RG y el menor valor de H2OO2 La temperatura
soacutelo aparece como un teacutermino estadiacutesticamente significativo en la produccioacuten de CO2 durante
la gasificacioacuten de char En este caso el aumento de temperatura conlleva una reduccioacuten en la
formacioacuten de CO2 (βT lt 0) y como se discutioacute anteriormente favorece la produccioacuten de CO lo
que sugiere un aumento de la reactividad del char con el CO2 a alta temperatura (reaccioacuten de
Boudouard ec 24)
Los coeficientes β resultantes de los anaacutelisis ANOVA (Tablas 43 y 44) muestran que la
produccioacuten neta de hidrocarburos ligeros (CH4 y C2Hx) se ve principalmente influenciada por la
composicioacuten del medio de gasificacioacuten Cuanto mayor es la relacioacuten H2OO2 mayor es la
produccioacuten de hidrocarburos ligeros (βH2OO2 gt 0) lo que sugiere una mayor reactividad de
estos gases con oxiacutegeno que con vapor de agua Ademaacutes la mayor presencia de H2 en la
atmoacutesfera gaseosa al aumentar la relacioacuten H2OO2 puede contribuir a la formacioacuten de CH4 a
traveacutes de la reaccioacuten de metanizacioacuten (ec 25) En el caso de la gasificacioacuten del lodo el
rendimiento a CH4 se ve negativamente afectado por la relacioacuten gasificante (βRG lt 0) y a
diferencia de los resultados mostrados por otros autores (Kim y cols 2001 Pinto y cols
2003) positivamente afectado por la temperatura (βT gt 0) lo que puede deberse a una mayor
importancia de la reaccioacuten de metanizacioacuten a alta temperatura Por otro lado el aumento de
temperatura reduce la produccioacuten neta de C2Hx en la gasificcioacuten de lodo (βT lt 0) Ninguno de
estos efectos ha resultado significativo en la produccioacuten de C2HX en la gasificacioacuten de char
para la que se obtuvieron resultados muy similares en la mayoriacutea de los experimentos (Tabla
42)
Por uacuteltimo la produccioacuten de H2S en ambos procesos de gasificacioacuten se ve acrecentada a
alta temperatura (βT gt 0) yo alta relacioacuten gasificante (βRG gt 0)
Ademaacutes de la produccioacuten especiacutefica de cada compuesto gaseoso la evolucioacuten de las
relaciones molares H2CO y COCO2 en el gas de salida resulta de especial intereacutes para el
posible uso del gas como materia prima para la produccioacuten de quiacutemicos La relacioacuten molar
H2CO se puede incrementar mediante el aumento de la relacioacuten H2OO2 utilizada como
agente gasificante (βH2OO2 gt 0) yo mediante la reduccioacuten de la temperatura de gasificacioacuten (βT
lt 0) La influencia relativa de ambos factores es bastante similar en el gas producto de la
gasificacioacuten de char (Tabla 44) pero el tipo de agente gasificante juega un papel maacutes
importante en la relacioacuten H2CO del gas producto de la gasificacioacuten de lodo (Tabla 43) Por
otro lado la relacioacuten COCO2 en ambos gases de salida se ve significativamente favorecida al
aumentar la temperatura (βT gt 0) yo al disminuir la relacioacuten gasificante (βRG lt 0) mientras que
la composicioacuten del medio de gasificacioacuten no ejerce una influencia significativa en ella
Resultados y discusioacuten 55
En cuanto al contenido energeacutetico del gas el anaacutelisis ANOVA muestra que la temperatura
es el factor maacutes influyente en el PCI del gas cuando se gasifica char (Tabla 44) mientras que la
influencia relativa de los tres factores estudiados en el PCI del gas es similar cuando se gasifica
lodo (Tabla 43) Como se puede observar en la Figura 45 y a diferencia de los resultados
mostrados por algunos autores (Pinto y cols 2003) la temperatura ejerce un efecto positivo
en el PCI de ambos productos gaseosos (βT gt 0) Dado que el contenido de hidrocarburos en el
gas (en teacuterminos de composicioacuten volumeacutetrica) disminuye al aumentar la temperatura cabriacutea
esperar una reduccioacuten del PCI con la temperatura Sin embargo la concentracioacuten de CO2 en el
gas tambieacuten disminuye con la temperatura y lo hace en una mayor proporcioacuten que el
contenido de hidrocarburos (Tablas 41 y 42) lo que al final se traduce en un aumento del PCI
con la temperatura debido a la menor dilucioacuten del contenido energeacutetico del gas El PCI de los
gases tambieacuten se ve favorecido con el aumento de la relacioacuten H2OO2 (βH2OO2 gt 0) debido al
aumento de la fraccioacuten de hidrocarburos y a la disminucioacuten simultaacutenea del porcentaje de CO2
En el caso de la gasificacioacuten del lodo este efecto se ve claramente potenciado a baja
temperatura (Figura 45a)
(a) Gasificacioacuten de lodo (b) Gasificacioacuten de char
Figura 45 Poder caloriacutefico inferior (MJmiddotm-3N) del gas producto de la (a) gasificacioacuten de lodo y (b) gasificacioacuten de char (RG = 095 gmiddotg-1
org) Las barras de error en las figuras representan la miacutenima diferencia significativa
Igual que ocurre con la produccioacuten de gas y con su PCI la eficiencia energeacutetica de ambos
procesos de gasificacioacuten depende principalmente de la temperatura de gasificacioacuten Esta
eficiencia puede mejorarse hasta casi en 20 puntos porcentuales mediante el aumento de la
temperatura de 770 a 850 ordmC (βT gt 0) (Figura 46) gracias al incremento del rendimiento a gas y
de su PCI La relacioacuten H2OO2 tambieacuten afecta positivamente a la eficiencia de gasificacioacuten
(βH2OO2 gt 0) aunque en menor grado que la temperatura Ademaacutes en el caso de la gasificacioacuten
770 810 850 42 45 48 51 54 57 60 63
PCI d
el g
as (M
Jmiddotm-3
N)
Temperatura (ordmC)
770 810 850 42 45 48 51 54 57 60 63
PCI d
el g
as (M
Jmiddotm-3
N)
Temperatura (ordmC)
H2OO2 = 2 H2OO2 = 3 H2OO2 = 1
56 Resultados y discusioacuten
de char la relacioacuten gasificante es tambieacuten un factor significativo (βRG gt 0) ademaacutes de algunos
efectos sineacutergicos y antagoacutenicos entre los factores pero su influencia relativa es mucho menos
importante que el efecto individual de la temperatura
Figura 46 Eficiencia energeacutetica en la gasificacioacuten de lodo () Las barras de error en la figura representan la miacutenima diferencia significativa
La composicioacuten del alquitraacuten recogido tras los experimentos de gasificacioacuten de lodo se
analizoacute cualitativamente por cromatografiacutea de gases A modo de ejemplo la Figura 47
muestra uno de los cromatogramas obtenidos con el detector de ionizacioacuten de llama El
anaacutelisis de la composicioacuten de los alquitranes suele realizarse clasificando los compuestos en
varias familias en funcioacuten de su peso molecular y composicioacuten atoacutemica (Li y Suzuki 2009a) Las
familias y compuestos considerados en este estudio han sido (i) compuestos aromaacuteticos con
un solo anillo (estireno y benceno) (ii) compuestos poliaromaacuteticos con 2 y 3 anillos (indeno
naftaleno metil-naftaleno bifenilo bifenileno fluoreno antraceno y fenantreno) (iii)
compuestos aromaacuteticos heterociacuteclicos con aacutetomos de N (incluyendo metil-piridina
benzonitrilo metil-benzonitrilo quinolina metil-quinolina indol fenil-piridina
naftalenocarbonitrilo benzoquinolina y 5H-indeno-[12-b]-piridina) (iv) compuestos
aromaacuteticos heterociacuteclicos con aacutetomos de O (fenol y benzofurano) y (v) compuestos orgaacutenicos
con S (2-benzotiofeno y 33-tiobis-propanonitrilo)
770 810 850 35 40 45 50 55 60 65 70 75
Temperatura (ordmC)
H2OO2 = 1 H2OO2 = 3
H2OO2 = 2
Efic
ienc
ia e
nerg
eacutetic
a de
la
gasi
ficac
ioacuten
de lo
do (
)
Resultados y discusioacuten 57
1 n-metil-piridina (96 min) 9 2-benzotiofeno (275 min) 17 naftaleno-n-carbonitrilo (445 min) 2 estireno (119 min) 10 quinolina (306 min) 18 fluoreno (469 min) 3 fenol (175 min) 11 n-metil-naftaleno (310 min) 19 33rsquo-tiobis-propanonitrilo (484 min) 4 benzofurano (176 min) 12 n-metil-quinolina (331 min) 20 antracenofenantreno (576 min) 5 benzonitrilo (192 min) 13 indol (343 min) 21 benzoquinolina (589 min) 6 indeno (200 min) 14 bifenilo (356 min) 22 5H-indeno-[12-b]piridina (619 min) 7 n-metil-benzonitrilo (221 min) 15 bifenileno (409 min) 8 naftaleno (264 min) 16 n-fenil-piridina (414 min)
Figura 47 Cromatograma de iones totales (TIC) de la muestra de alquitraacuten del experimento 3 (850 ordmC RG = 08 y H2OO2 = 3
Las aacutereas de los principales picos identificados en los cromatogramas TIC (especificados en
su mayoriacutea en la Figura 47) se han utilizado para evaluar la influencia de las condiciones de
operacioacuten en la composicioacuten de las muestras de alquitraacuten La Tabla 45 recoge estos
porcentajes de aacuterea cromatograacutefica agrupando los compuestos por familias Cabe destacar
que estos datos porcentuales no representan la composicioacuten real de las muestras ya que el
factor de respuesta aacutereaconcentracioacuten no es el mismo para todos los compuestos pero estos
datos son uacutetiles para la comparacioacuten de las muestras y el anaacutelisis de la influencia de los
factores
Tabla 45 Porcentaje de aacuterea cromatograacutefica correspondiente a cada familia de compuestos del alquitraacuten
Temperatura (ordmC) 850 770 850 770 850 770 850 770 810
Relacioacuten gasificante (gmiddotg-1
org) 11 11 08 08 11 11 08 08 095
Relacioacuten molar H2OO2 3 3 3 3 1 1 1 1 2 Aromaacuteticos con 1 anillo
10 12 4 13 5 10 4 10 7 plusmn 2
Poliaromaacuteticos (2-3 anillos) 9 9 34 5 40 42 45 23 36 plusmn 9
Aromaacuteticos con N 75 68 57 69 46 44 44 61 50 plusmn 14 Aromaacuteticos con O 30 77 03 77 06 21 10 60 26 plusmn 08 Orgaacutenicos con S 4 3 4 5 8 3 6 1 5 plusmn 1
Los datos de la uacuteltima columna representan la media plusmn desviacioacuten estaacutendar de las 3 reacuteplicas del punto central
58 Resultados y discusioacuten
De acuerdo con el anaacutelisis ANOVA de los porcentajes de aacuterea cromatograacutefica (Tabla 46)
la temperatura y la relacioacuten H2OO2 son los uacutenicos factores con un efecto significativo en la
composicioacuten del alquitraacuten Hay que destacar que la repetitividad de los resultados no fue todo
lo buena que podriacutea desearse por lo que soacutelo los efectos con una clara influencia han
resultado significativos en el anaacutelisis ANOVA
Tabla 46 Coeficientes de regresioacuten lineal (β) obtenidos del anaacutelisis ANOVA del aacuterea cromatograacutefica de las familias de compuestos en el alquitraacuten de la gasificacioacuten de lodo
Valor medio βT βRG βH2OO2 βT-RG βT-H2OO2 βRG-H2OO2 βT-H2OO2-RG Curvatura
Aromaacuteticos con 1 anillo 851 -281
Poliaromaacuteticos (2-3 anillos) 2578 -1150
Aromaacuteticos con N 5804 911
Aromaacuteticos con O 355 -232
Orgaacutenicos con S 414 126 -126 -084
teacutermino no significativo
Las familias de compuestos aromaacuteticos maacutes ligeros (1 anillo) y de aromaacuteticos con aacutetomos
de O son las maacutes sensibles a la variacioacuten de temperatura disminuyendo su presencia con el
aumento de la temperatura Los compuestos fenoacutelicos parafinas olefinas y aromaacuteticos
alquilados pueden ser faacutecilmente craqueados a altas temperaturas (Ponzio y cols 2006) El
aumento de la fraccioacuten de compuestos sulfurados con la temperatura puede ser soacutelo una
consecuencia de la disminucioacuten de las otras fracciones de compuestos Por otro lado las
familias de compuestos poliaromaacuteticos y de aromaacuteticos con N parecen las maacutes sensibles a la
relacioacuten H2OO2 utilizada como agente gasificante El aumento de la relacioacuten H2OO2 conduce a
una disminucioacuten en la fraccioacuten de poliaromaacuteticos lo que puede deberse a la interferencia del
vapor de agua en las reacciones de polimerizacioacuten El peso molecular de los compuestos que
forman parte del alquitraacuten depende de la presencia de radicales libres de H relacionada con la
disponibilidad de vapor de agua en el medio de gasificacioacuten (Qin y cols 2010) La variacioacuten de
la fraccioacuten de compuestos poliaromaacuteticos se ha visto acompantildeada de una variacioacuten en sentido
opuesto de la presencia de compuestos aromaacuteticos con N lo cual puede ser soacutelo una
consecuencia del hecho de estar analizando datos porcentuales
Produccioacuten teoacuterica de gases basada en datos de equilibrio
El rendimiento o produccioacuten experimental de los distintos compuestos gaseosos
obtenidos en la gasificacioacuten de lodo y de char se ha comparado con la produccioacuten teoacuterica que
cabriacutea esperar en condiciones de equilibrio Esta produccioacuten de gases en el equilibrio se ha
Resultados y discusioacuten 59
calculado con el software HSC Chemistry 61 que utiliza el meacutetodo de minimizacioacuten de la
energiacutea de Gibbs para calcular la composicioacuten de equilibrio en condiciones isoteacutermicas e
isobaacutericas El sistema de reaccioacuten debe ser especificado para los caacutelculos incluyendo la
temperatura de reaccioacuten presioacuten cantidad de reactivos (agente gasificante y anaacutelisis
elemental del soacutelido) y posibles especies que podriacutean formar parte de los productos Las
condiciones de temperatura presioacuten y alimentacioacuten de agente gasificante fueron las mismas
que las estudiadas en el laboratorio Como posibles especies que podriacutean formar parte de los
productos en el equilibrio se especificaron todos los gases obtenidos en los experimentos (H2
N2 O2 CO CO2 H2O CH4 C2Hx y H2S) ademaacutes de NH3 carbono soacutelido (C) y diversos
compuestos modelo habituales en el alquitraacuten de gasificacioacuten (benceno tolueno naftaleno y
piridina) Los gases mayoritarios obtenidos en las simulaciones de equilibrio fueron N2 H2 CO
CO2 H2O CH4 H2S y NH3 Los demaacutes compuestos especificados apareciacutean en concentraciones
insignificantes ( lt 10-10)
En las Figuras 48 y 49 se comparan los datos teoacutericos y experimentales de la produccioacuten
de H2 CO CO2 y CH4 en la gasificacioacuten de lodo y de char respectivamente Los rendimientos
experimentales a H2 y CO estaacuten claramente por debajo de sus correspondientes datos de
equilibrio mientras que los rendimientos experimentales a CO2 y CH4 son mayores que los
valores teoacutericos de equilibrio Estas diferencias ponen de manifiesto que el equilibrio quiacutemico
no fue alcanzado en los experimentos de gasificacioacuten Esto puede deberse a un insuficiente
tiempo de residencia de los gases y vapores en el reactor yo a que en el proceso existen
reacciones limitantes controladas por la cineacutetica o por la transferencia de masa
60 Resultados y discusioacuten
0
20
40
60
80
100
120
5 6 7 8 PC 1 3 2 4
g H 2
kg
org
Experimento
0
200
400
600
800
1000
1200
6 5 2 1 PC 8 7 4 3
g CO
kg
org
Experimento
0
200
400
600
800
1000
1200
3 4 7 8 PC 1 2 5 6
g CO
2 k
g or
g
Experimento
0
10
20
30
40
50
60
g CH4 kg org
(valores de equilibrio)
g
CH 4
kg
org
(val
ores
exp
erim
enta
les)
Experimento
0
1
2
3
4
5
6
5 1 7 6 3 PC 2 8 4
PC Punto central (media de los experimentos 9 10 y 11)
Figura 48 Produccioacuten teoacuterica ( ) y experimental ( ) de H2 CO CO2 y CH4 en la gasificacioacuten de lodo
0
20
40
60
80
100
120
16 17 19 18 PC 15 14 12 13
g H 2
kg
org
Experimento
0200400600800
100012001400
17 16 13 12 PC 19 15 18 14
g CO
kg
org
Experimento
0
200
400
600
800
1000
1200
14 15 18 19 PC 12 13 16 17
g CO
2 k
g or
g
Experimento
02468
101214
16 12 18 17 PC 14 13 19 15
g CH
4 k
g or
g
Experimento PC Punto central (media de los experimentos 20 21 y 22)
Figura 49 Produccioacuten teoacuterica ( ) y experimental ( ) de H2 CO CO2 y CH4 en la gasificacioacuten de char
Resultados y discusioacuten 61
La influencia de los factores de operacioacuten sobre los rendimientos teoacutericos a H2 CO CO2 y
CH4 en el equilibrio ha sido analizada estadiacutesticamente mediante anaacutelisis ANOVA Los
coeficientes de regresioacuten lineal (β) obtenidos para los valores codificados de los factores se
muestran en la Tabla 47
Tabla 47 Coeficientes de regresioacuten lineal (β) obtenidos del anaacutelisis ANOVA de los datos de produccioacuten de gases en el equilibrio
Valor medio βT βRG βH2OO2 βT-RG βT-H2OO2 βRG-H2OO2 βT-H2OO2-RG Curvatura
Gasificacioacuten de lodo
H2 (gmiddotkg-1org) 9504 -115 -320 1480 237
CO (gmiddotkg-1org) 87307 2465 -8722 3638 1583
CO2 (gmiddotkg-1org) 61016 -3650 13861 -5831 -880 752 -2441
CH4 (gmiddotkg-1org) 107 -081 -056 042 043 -033 -017 014
Gasificacioacuten de char
H2 (gmiddotkg-1org) 7197 1358 -082 181
CO (gmiddotkg-1org) 123658 1995 -11574 3059 1657
CO2 (gmiddotkg-1org) 40457 -2471 18733 -5124 -1069 -2455
CH4 (gmiddotkg-1org) 325 -241 -198 115 145 -079 -054 032
teacutermino no significativo curvatura significativa
Los efectos de los factores de operacioacuten sobre la produccioacuten experimental y de equilibrio
de H2 CO CO2 y CH4 muestran algunas diferencias importantes (i) la temperatura era el factor
maacutes influyente en los rendimientos experimentales a H2 y CO pero no lo es en los datos de
equilibrio la relacioacuten H2OO2 ejerce el efecto maacutes significativo en la produccioacuten de equilibrio
de H2 mientras que la relacioacuten gasificante ejerce la mayor influencia en la produccioacuten teoacuterica
de CO (ii) la produccioacuten de equilibrio de H2 apenas se ve afectada por la temperatura soacutelo se
observa un pequentildeo efecto negativo de la temperatura en el caso de la gasificacioacuten de lodo
(iii) la produccioacuten de equilibrio de CO no se ve afectada negativamente por la relacioacuten H2OO2
como ocurriacutea con los datos experimentales sino que le afecta positivamente (iv) la produccioacuten
de equilibrio de CH4 no se ve afectada positivamente por la temperatura como ocurriacutea con los
datos experimentales en la gasificacioacuten de lodo sino que se ve draacutesticamente reducida con el
aumento de temperatura A diferencia de lo que ocurre en un proceso controlado por la
cineacutetica o por fenoacutemenos de difusioacuten en los que un aumento de temperatura favorece tanto
la velocidad de reaccioacuten como la transferencia de masa en el reacutegimen de control
termodinaacutemico el aumento de temperatura no resulta favorable para el transcurso de las
reacciones exoteacutermicas (principio de Le Chacirctelier) Esto justifica las diferencias observadas y
pone de manifiesto la importancia de discernir entre control cineacutetico o control termodinaacutemico
en un proceso de gasificacioacuten de cara a la optimizacioacuten de las condiciones de operacioacuten
62 Resultados y discusioacuten
42 Evaluacioacuten energeacutetica
En esta seccioacuten se plantea un estudio energeacutetico de las etapas de gasificacioacuten de lodo y de
char cuyos resultados experimentales han sido detallados en el apartado anterior Ademaacutes
esta seccioacuten incluye tambieacuten un balance energeacutetico del proceso de piroacutelisis de lodo en el que
se genera el char como subproducto y un anaacutelisis del consumo energeacutetico necesario para el
secado teacutermico del lodo previo a las etapas de piroacutelisis o gasificacioacuten Los balances energeacuteticos
de las distintas etapas permiten tener una idea global del coste energeacutetico asociado al
tratamiento termoquiacutemico del lodo mediante (i) secado y gasificacioacuten del lodo y (ii) secado y
piroacutelisis del lodo y gasificacioacuten del char (Figura 410)
(a) Secado y gasificacioacuten del lodo
(b) Secado y piroacutelisis del lodo y gasificacioacuten del char
Figura 410 Diagrama de dos posibles tratamientos termoquiacutemicos para la gestioacuten de los lodos de EDAR
La demanda energeacutetica de las etapas de gasificacioacuten y piroacutelisis se ha calculado utilizando
los rendimientos experimentales a los distintos productos Como se detalla maacutes adelante
algunas de las propiedades de los reactantes y productos necesarias para los caacutelculos fueron
medidas experimentalmente mientras que otras se obtuvieron de datos bibliograacuteficos Los
balances energeacuteticos se han realizado considerando las siguientes hipoacutetesis y simplificaciones
- Reactor adiabaacutetico No se consideran peacuterdidas de calor
- La capacidad caloriacutefica especiacutefica Cp (T) de los reactivos y productos soacutelidos (lodo char
de piroacutelisis y cenizas de gasificacioacuten) que es una de las propiedades requeridas para los
caacutelculos energeacuteticos se ha considerado constante con la temperatura Aunque estas
Piroacutelisis
Secado teacutermico
Gas
Char
Gas
Gasificacioacuten
Gasificacioacuten
Lodo de EDAR
huacutemedo Secado teacutermico
Gas
Ceniza
Alquitraacuten Agua
Lodo seco
Aire Vapor de agua
Lodo de EDAR
huacutemedo Lodo seco
Liacutequido de piroacutelisis
Aire Vapor de agua
Ceniza
Alquitraacuten Agua
Resultados y discusioacuten 63
capacidades caloriacuteficas especiacuteficas fueron medidas experimentalmente por calorimetriacutea
diferencial de barrido las limitaciones operacionales del equipo no permitieron obtener la
funcioacuten Cp (T) en el intervalo completo de temperatura requerido en cada caso (hasta 530
ordmC para el char y hasta 850 ordmC para las cenizas de gasificacioacuten) por lo que se optoacute por
trabajar con valores constantes de Cp medidos en valores intermedios de dichos
intervalos de temperatura (a 300 ordmC para el char y a 500 ordmC para a ceniza)
- Como se explica maacutes adelante de forma maacutes detallada la composicioacuten de los
condensados recogidos en ambos procesos (alquitraacuten de gasificacioacuten o liacutequido de piroacutelisis)
se ha simplificado considerando soacutelo algunos de sus compuestos con el fin de obtener
datos de propiedades de la bibliografiacutea
La demanda energeacutetica de los procesos de piroacutelisis y gasificacioacuten (Q) puede obtenerse por
diferencia de las entalpiacuteas de las corrientes que entran (ΔHentra) y salen (ΔHsale) del reactor
Q (MJmiddotkg-1materia prima) = ΔHsale - ΔHentra (ec 41)
De acuerdo con la ec 41 Q lt 0 se corresponde con un proceso exoteacutermico mientras que
Q gt 0 hace referencia a un proceso endoteacutermico La entalpiacutea total de cada corriente de entrada
o salida (∆H) por kilogramo de materia prima alimentado se calcula con la ec 42
ΔH (MJ ∙ kgmateria primaminus1 ) = sum mi ∙ (ΔHfi
o + i int Cpi
TTref
(T) dT) (ec 42)
donde
- mi es el rendimiento maacutesico de cada corriente o en el caso de la corriente gaseosa de
cada compuesto gaseoso ya que se conocen los rendimientos individuales de cada uno La
base de caacutelculo utilizada ha sido de 1 kg de lodo en el proceso de piroacutelisis y 1 kg de lodo o
de char en el proceso de gasificacioacuten (kgmiddotkg-1materia prima)
- Tref es la temperatura de referencia (298 K) y T es la temperatura a la que se encuentra
cada corriente (K) Todas las corrientes de entrada se consideran a temperatura ambiente
(298 K) excepto el vapor de agua utilizado en los experimentos de gasificacioacuten que se
alimentoacute a 150 ordmC (448 K)
- ΔHordmfi es la entalpiacutea estaacutendar de formacioacuten (MJmiddotkg-1) de cada corriente o compuesto a la
temperatura de referencia (298 K) Los datos de ΔHordmf de los compuestos gaseosos se han
obtenido de la literatura (Perry y Green 1999) mientras que los correspondientes a los
materiales soacutelidos y al liacutequido de piroacutelisis se han calculado a partir de los datos
experimentales de anaacutelisis elemental y poder caloriacutefico de acuerdo con la siguiente
ecuacioacuten
64 Resultados y discusioacuten
∆Hfi
o = sum mj j ∙ ∆Hfj
o + PCSi (ec 43)
donde j representa cada producto derivado de la combustioacuten completa del material
(CO2 H2O SO2 y NO) mj es la masa de cada gas de combustioacuten producido por kilogramo
de material ΔHordmfj es la entalpiacutea estaacutendar de formacioacuten de cada gas de combustioacuten y PCSi
es el poder caloriacutefico superior del material Esta forma de calcular la ΔHordmf no incluye la
entalpiacutea correspondiente a la fraccioacuten inorgaacutenica de los soacutelidos pero este dato no es
necesario para realizar los balances energeacuteticos ya que la fraccioacuten inorgaacutenica se considera
inerte en el proceso
- Cpi (T) es la capacidad caloriacutefica especiacutefica de cada corriente o compuesto en funcioacuten de la
temperatura (MJmiddotkg-1middotK-1) Como se ha comentado anteriormente el Cp de los soacutelidos se
midioacute experimentalmente por calorimetiacutea diferencial de barrido mientras que los datos
de Cp (T) de los gases y vapores se obtuvieron de la bibliografiacutea (ChemSpider database
Harrison y Seaton 1988 Perry y Green 1999) Si el intervalo de temperatura en la integral
de la ec 42 implica un cambio de fase la entalpiacutea de vaporizacioacuten (∆Hvap) de los
compuestos condensables y su Cp (T) en fase liacutequida tambieacuten deben incluirse en la
ecuacioacuten Estos datos (para los compuestos condensables considerados en cada caso) se
han obtenido de bases de datos y de estudios de la bibliografiacutea (ChemSpider database
Chueh y Swanson 1973)
A continuacioacuten se detallan los datos y consideraciones maacutes especiacuteficas para el caacutelculo de
la demanda energeacutetica en los procesos de piroacutelisis y gasificacioacuten
Piroacutelisis de lodo
La distribucioacuten de productos resultante de la piroacutelisis de lodo asiacute como las propiedades de
los mismos necesarias para el balance de energiacutea se recogen en la Tabla 48 (Gil-Lalaguna y
cols 2010) El liacutequido recogido despueacutes de la condensacioacuten de los vapores estaacute compuesto de
tres fases fase orgaacutenica ligera (FOL) fase orgaacutenica pesada (FOP) y fase acuosa (FA) Las
propiedades de las tres fases se midieron por separado considerando cada fase liacutequida como
un producto distinto para los caacutelculos
Resultados y discusioacuten 65
Tabla 48 Rendimientos y propiedades de los productos de la piroacutelisis de lodo de EDAR
Rendimiento maacutesico () Composicioacuten PCS
(MJmiddotkg-1) ΔHordmf
(MJmiddotkg-1) Cp (T)
(kJmiddotK-1middotkg-1) ΔHvap
(MJmiddotkg-1)
Char
519 plusmn 07 Anaacutelisis elemental ( en masa) C 1549 H 097 N 185 S 035 52 plusmn 02 -118 121 (300 ordmC) ---
Gases no condensables (sin incluir el N2)
101 plusmn 09
( en masa) CO2 743 plusmn 09 CO 132 plusmn 01 H2 17 plusmn 01
CH4 38 plusmn 01 C2H6 14 plusmn 02 C2H4 14 plusmn 01 H2S 43 plusmn 09
80 plusmn 03 -739 118 (25 ordmC) 156 (530 ordmC) ---
Fase orgaacutenica ligera (FOL)
22 plusmn 02
Anaacutelisis elemental ( en masa) C 8592 H 1183 N 180 S 027
Agua 0 ( en masa)
Orgaacutenicos 100 ( en masa)
4310 plusmn 004 -174 185 (liacutequido) 307 (530 ordmC) 018
Fase orgaacutenica pesada (FOP)
94 plusmn 02
Anaacutelisis elemental ( en masa) C 6954 H 897 N 944 S 124
Agua 64 plusmn 03 ( en masa)
Orgaacutenicos 936 plusmn 03 ( en masa)
32 plusmn 2 -349 213 (liacutequido) 236 (530 ordmC) 055
Fase acuosa (FA)
208 plusmn 02
Anaacutelisis elemental ( en masa) C 1117 H 1045 N 652 S 037
Agua 738 plusmn 04 ( en masa)
Orgaacutenicos 262 plusmn 04 ( en masa)
57 plusmn 03 -1244 359 (liacutequido) 212 (530 ordmC) 177
El error experimental se expresa como la media plusmn desviacioacuten estaacutendar de 2 reacuteplicas
El PCS de los soacutelidos y de las fases liacutequidas se midioacute experimentalmente con un
caloriacutemetro mientras que el PCS del gas se calculoacute a partir de su composicioacuten La ΔHordmf de los
soacutelidos y liacutequidos se puede calcular con la ec 43 mientras que la ΔHordmf de los distintos
compuestos gaseosos puede obtenerse de la bibliografiacutea (Perry y Green 1999) Como se ha
comentado anteriormente se han utilizado valores constantes de Cp para los soacutelidos 115middot10-3
MJmiddotkg-1middotK-1 para el lodo y 121middot10-3 MJmiddotkg-1middotK-1 para el char La composicioacuten de las fases liacutequidas
se ha simplificado considerando soacutelo su contenido de agua (determinado mediante el meacutetodo
Karl Fischer) y un compuesto representativo de toda la fraccioacuten orgaacutenica en cada una de las
fases 4-colesteno para la FOL 3-metil-fenol para la FOP y aacutecido aceacutetico para la FA Estos
compuestos fueron elegidos por ser algunos de los compuestos con mayor aacuterea
cromatograacutefica en el anaacutelisis de las muestras por cromatografiacutea con detector de ionizacioacuten de
66 Resultados y discusioacuten
llama La Cp y la ∆Hvap de las fracciones orgaacutenicas de las fases liacutequidas se han equiparado a las
de estos compuestos elegidos como representativos cuyos datos han sido obtenidos de la
bibliografiacutea (ChemSpider database Chueh y Swanson 1973 Harrison y Seaton 1988 Perry y
Green 1999) Los valores globales de Cp y ∆Hvap para cada fase pueden estimarse como un
promedio ponderado de los datos correspondientes al agua y al compuesto orgaacutenico elegido
en cada fase
Aplicando todos estos datos en la ec 42 la entalpiacutea total de las corrientes que entran al
reactor (ΔHentra) es de -328 MJmiddotkg-1lodo y la entalpiacutea total de los productos (ΔHsale) considerados
a la temperatura de piroacutelisis (T = 803 K) es de -313 MJmiddotkg-1lodo La diferencia de estas entalpiacuteas
(ec 41) resulta en una demanda energeacutetica de 015 MJmiddotkg-1lodo para llevar a cabo la
descomposicioacuten teacutermica del lodo durante el proceso de piroacutelisis Este valor es algo menor que
los datos encontrados en la bibliografiacutea para la piroacutelisis de otros tipos de biomasa Por
ejemplo la piroacutelisis de residuos de cosecha conlleva un coste energeacutetico de 03 MJmiddotkg-1
(Mangaro y cols 2011) El mayor contenido de ceniza en el lodo que no se descompone
durante el proceso puede explicar esta diferencia
Si se considera el enfriamiento de los gases y vapores a la salida del reactor hasta
temperatura ambiente (298 K) para aprovechar su energiacutea teacutermica (calor sensible y latente) el
proceso de piroacutelisis pasa a ser exoteacutermico con un calor neto de -070 MJmiddotkg-1lodo Por lo tanto
bajo las hipoacutetesis consideradas la etapa de piroacutelisis podriacutea ser un proceso autosuficiente desde
el punto de vista energeacutetico
Gasificacioacuten de lodo y de char
La distribucioacuten de productos obtenida experimentalmente en la gasificacioacuten del lodo de
EDAR y de su char de piroacutelisis se mostroacute en las Tablas 41 y 42 Estos datos permiten calcular la
demanda energeacutetica de ambos procesos bajo diferentes condiciones de operacioacuten
La composicioacuten de las fracciones de alquitraacuten obtenidas en los experimentos se ha
simplificado a una mezcla equimolar de benceno naftaleno y piridina (compuestos presentes
en la mayoriacutea de las muestras) Asiacute los datos de ∆Hordmf Cp y ΔHvap del alquitraacuten necesarios para
los balances energeacuteticos se han equiparado a los promedios ponderados de los valores
correspondientes a estos tres compuestos que se pueden encontrar en la bibliografiacutea Por
otro lado la Cp de los subproductos soacutelidos de la gasificacioacuten se ha aproximado en todos los
casos a la de la ceniza del lodo ya que estos soacutelidos estaacuten compuestos principalmente de
ceniza (gt 93 de su masa en la mayoriacutea de los casos) La Cp (T) de la ceniza de lodo se midioacute
experimentalmente por calorimetriacutea diferencial de barrido y el valor constante utilizado ha
Resultados y discusioacuten 67
sido 107middot10-3 MJmiddotK-1middotkg-1 (medido a 500 ordmC) La ∆Hordmf de los soacutelidos se ha calculado con la ec 43
utilizando la foacutermula de Dulong para calcular el PCS a partir de su anaacutelisis elemental [PCS
(kJmiddotkg-1) = 339middotC + 1430middot(H - O8 ) + 105middotS]
Todos estos datos se utilizan para calcular con la ec 42 las entalpiacuteas de las corrientes de
entrada y de salida del reactor (considerando los productos a la temperatura de gasificacioacuten) y
con ellas la demanda energeacutetica para las diferentes condiciones de operacioacuten estudiadas (ec
41) Los resultados obtenidos se representan en la Figura 411 en funcioacuten de la temperatura
de gasificacioacuten (770-850 ordmC) y de la composicioacuten del medio gasificante RE (relacioacuten
equivalente) y SB (relacioacuten maacutesica entre el vapor de agua y la cantidad de materia orgaacutenica en
el lodo o en el char)
RE = 32 RE = 23 RE = 19 RE = 17 RE = 12 SB = 039 SB = 027 SB = 052 SB = 071 SB = 052
Figura 411 Demanda energeacutetica en la (a) gasificacioacuten de lodo y (b) gasificacioacuten de char calculada con rendimientos experimentales y considerando los productos a la temperatura de gasificacioacuten
Bajo las hipoacutetesis realizadas la demanda energeacutetica oscila entre -261 y 129 MJmiddotkg-1lodo
para la gasificacioacuten de lodo de EDAR y entre -023 y 120 MJmiddotkg-1char para la gasificacioacuten de char
Por lo tanto las condiciones de operacioacuten ejercen una mayor influencia en el balance
energeacutetico de la gasificacioacuten del lodo Por otro lado a pesar de que el contenido orgaacutenico del
char es bastante menor que el del lodo (Tabla 31) gasificar 1 kg de char conlleva en general
una mayor demanda de energiacutea externa que gasificar 1 kg de lodo Por ejemplo la demanda
energeacutetica en la gasificacioacuten de lodo cuando se opera con RE = 17 SB = 071 y T = 850ordmC es
de 064 MJmiddotkg-1lodo mientras que alcanza 100 MJmiddotkg-1
char en la gasificacioacuten de char Los cambios
observados en la estructura carbonosa del soacutelido despueacutes del proceso de piroacutelisis permiten
explicar esta diferencia La fraccioacuten de materia volaacutetil en el char es menor que en el lodo
mientras que su contenido en carbono fijo es mayor (Tabla 31) Por lo tanto durante la
gasificacioacuten del lodo las reacciones de combustioacuten de la materia volaacutetil (reacciones raacutepidas
Gasificacioacuten de char
-3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=770ordmC
T=850ordmC
T=850ordmC T=810ordmC T=770ordmC
T=770ordmC
(b)
Demanda energeacutetica (MJmiddotkg-1char)
Gasificacioacuten de lodo
-3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=850ordmC
T=850ordmC
T=770ordmC T=810ordmC
T=770ordmC
T=770ordmC
(a)
Demanda energeacutetica (MJmiddotkg-1lodo)
68 Resultados y discusioacuten
gas-gas) aportan buena parte de la energiacutea necesaria para las reacciones endoteacutermicas
mientras que en el caso de la gasificacioacuten de char la presencia de hidrocarburos en la
atmoacutesfera gaseosa es escasa y las principales reacciones de combustioacuten en fase gas involucran
a otros gases como H2 o CO cuyo poder caloriacutefico es bastante inferior al de los hidrocarburos
Como consecuencia de ello la gasificacioacuten de char con mezclas de aire y vapor de agua
aparece como un proceso endoteacutermico bajo la mayoriacutea de las condiciones estudiadas
mientras que la gasificacioacuten de lodo es un proceso exoteacutermico cuando se trabaja
simultaacuteneamente con RE gt 19 y SB lt 052
Si se considera el enfriamiento de los gases y vapores a la salida del reactor hasta
temperatura ambiente (298 K) para aprovechar su energiacutea teacutermica (calor sensible y latente) el
calor neto de los procesos de gasificacioacuten variacutea entre -580 y -165 MJmiddotkg-1lodo en la gasificacioacuten
de lodo y entre -117 y 026 MJmiddotkg-1char en la gasificacioacuten de char Por lo tanto teniendo en
cuenta las hipoacutetesis realizadas ambos procesos parecen autosuficientes desde el punto de
vista energeacutetico (excepto la gasificacioacuten de char con RE = 12 y T = 850 ordmC)
Ademaacutes de calcular la demanda energeacutetica en los procesos de gasificacioacuten se ha
considerado interesante definir un paraacutemetro (llamado eficiencia energeacutetica de la gasificacioacuten)
para medir queacute parte de la energiacutea contenida inicialmente en la materia prima queda
disponible en el gas producto tras cubrir la demanda energeacutetica del propio proceso de
gasificacioacuten y el consumo energeacutetico en la generacioacuten del vapor de agua utilizado como agente
gasificante (ec 44)
Eiciencia energeacutetica de la gasiicacioacuten () = Energiacutea recuperada en el gas minus Qgasiicacioacuten minus Qvapor
PCImateria prima∙ 100 (ec 44)
donde
- La energiacutea recuperada en el gas incluye la energiacutea asociada a su poder caloriacutefico (producto
del volumen de gas seco generado por su PCI) y el aprovechamiento de su calor sensible y
latente multiplicado por un factor de eficiencia de intercambio de calor del 70
- Qgasificacioacuten es la demanda de energiacutea en la gasificacioacuten de lodo o de char considerando los
productos a la temperatura de gasificacioacuten (Figura 411)
- Qvapor es la energiacutea necesaria para calentar y evaporar el agua desde 25 ordmC hasta 150 ordmC
(236 MJmiddotkg-1H2O)
- PCImateria prima es la energiacutea contenida en el lodo o en el char expresada en forma de su
poder caloriacutefico inferior (Tabla 31)
Resultados y discusioacuten 69
Los datos de eficiencia energeacutetica de los procesos de gasificacioacuten de lodo y gasificacioacuten de
char calculados seguacuten la ec 44 se muestran en la Tabla 49
Tabla 49 Datos de eficiencia energeacutetica de la gasificacioacuten de lodo y de char basada en rendimientos experimentales a los productos
Temperatura 850 770 850 770 850 770 850 770 810 RE () 17 17 12 12 32 32 23 23 19 Relacioacuten maacutesica SB 071 071 052 052 039 039 027 027 052
Gasificacioacuten de lodo
Energiacutea recuperada en el gas (MJmiddotkg-1
lodo) 1023 791 963 841 910 677 956 693 872 plusmn 012
Eficiencia energeacutetica () 74 58 65 61 87 75 82 65 71 plusmn 2
Gasificacioacuten de char
Energiacutea recuperada en el gas (MJmiddotkg-1
char) 391 277 349 244 352 244 334 227 298 plusmn 005
Eficiencia energeacutetica () 51 32 40 23 64 49 54 33 40 plusmn 2
Los resultados de eficiencia energeacutetica variacutean del 58 al 87 en la gasificacioacuten de lodo y
del 23 al 64 en la gasificacioacuten de char Tanto la menor demanda de energiacutea como el mayor
rendimiento a gas obtenido en la gasificacioacuten de lodo en comparacioacuten con la gasificacioacuten de
char contribuyen a su mejor eficiencia energeacutetica Como se observa en la Tabla 49 la
eficiencia energeacutetica de la gasificacioacuten (definida seguacuten la ec 44) mejora al aumentar la
temperatura de gasificacioacuten y en ambos casos los mejores resultados se obtienen al operar
con la mayor RE (32) y una relacioacuten SB moderada (039 gmiddotg-1org)
Estudio teoacuterico de la gasificacioacuten de lodo y de char basado en datos de equilibrio
Los resultados experimentales obtenidos en la gasificacioacuten de lodo y de char muestran
que en ninguno de los procesos se alcanzoacute el equilibrio quiacutemico sino que en ambos existiacutean
reacciones controladas por la cineacutetica o por fenoacutemenos difusionales A pesar de ello se ha
considerado interesante incluir en el estudio energeacutetico la situacioacuten de equilibrio para evaluar
sus liacutemites desde un punto de vista termodinaacutemico El caacutelculo de la demanda energeacutetica para
la gasificacioacuten de lodo y de char en la situacioacuten de equilibrio se lleva a cabo de forma anaacuteloga a
la realizada con los datos experimentales pero utilizando la distribucioacuten de productos
calculada con el software HSC Chemistry 61 (Figuras 48 y 49) en lugar de los rendimientos
experimentales Los resultados obtenidos en las simulaciones de equilibrio realizadas bajo las
mismas condiciones de operacioacuten que las estudiadas en el laboratorio se muestran en la
Figura 412
70 Resultados y discusioacuten
RE = 32 RE = 23 RE = 19 RE = 17 RE = 12 SB = 039 SB = 027 SB = 052 SB = 071 SB = 052
Figura 412 Demanda energeacutetica en la (a) gasificacioacuten de lodo y (b) gasificacioacuten de char calculada con
rendimientos de equilibrio y considerando los productos a la temperatura de gasificacioacuten
La comparacioacuten de las Figuras 411 y 412 muestra que el hecho de alcanzar el equilibrio
quiacutemico en ambos procesos de gasificacioacuten conlleva una demanda adicional de energiacutea La
gasificacioacuten de lodos de EDAR en condiciones de equilibrio soacutelo aparece como un proceso
exoteacutermico (en las condiciones experimentales simuladas) cuando la RE se aumenta hasta un
32 mientras que la gasificacioacuten de char en condiciones de equilibrio es un proceso
endoteacutermico en todas las situaciones simuladas La razoacuten es el predominio de las reacciones de
equilibrio endoteacutermicas como la reaccioacuten water-gas (ec 23) la reaccioacuten de Boudouard (ec
24) el reformado con vapor (ec 27) y el reformado en seco (ec 28) y la escasez de
reacciones de equilibrio exoteacutermicas (reaccioacuten water-gas shift ec 26) Estas reacciones son
llevadas a su maacutexima extensioacuten cuando se alcanza el equilibrio quiacutemico lo que implica un
mayor consumo energeacutetico
La eficiencia energeacutetica de ambos procesos de gasificacioacuten en situacioacuten de equilibrio se ha
calculado de forma anaacuteloga al caso experimental (ec 44) Estos resultados teoacutericos se recogen
en la Tabla 410 A pesar de la energiacutea adicional necesaria para alcanzar el equilibrio los datos
de eficiencia energeacutetica en condiciones de equilibrio son mayores que los datos
experimentales eficiencia del 90-94 en la gasificacioacuten de lodo en equilibrio y del 78-84 en
la gasificacioacuten de char en equilibrio Esto se debe a que en el equilibrio quiacutemico se obtiene un
mayor rendimiento a gas que tiene ademaacutes un mayor PCI lo que permite recuperar maacutes
energiacutea en el gas contrarrestando asiacute la mayor demanda energeacutetica
Gasificacioacuten de char
-3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=770ordmC
T=850ordmC
T=850ordmC T=810ordmC T=770ordmC
T=770ordmC
Demanda energeacutetica (MJmiddotkg-1char)
(b) Gasificacioacuten de lodo
-3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=850ordmC
T=850ordmC
T=770ordmC T=810ordmC T=770ordmC
T=770ordmC
Demanda energeacutetica (MJmiddotkg-1lodo)
(a)
Resultados y discusioacuten 71
Tabla 410 Datos de eficiencia energeacutetica de la gasificacioacuten de lodo y de char en condiciones de equilibrio
Temperatura 850 770 850 770 850 770 850 770 810 RE () 17 17 12 12 32 32 23 23 19 Relacioacuten maacutesica SB 071 071 052 052 039 039 027 027 052
Gasificacioacuten de lodo
Energiacutea recuperada en el gas (MJmiddotkg-1
lodo) 1437 1411 1495 1472 1134 1109 1286 1264 1370
Eficiencia energeacutetica () 90 91 91 92 94 94 94 94 92
Gasificacioacuten de char
Energiacutea recuperada en el gas (MJmiddotkg-1
char) 619 609 653 642 494 485 562 553 595
Eficiencia energeacutetica () 78 80 80 81 82 83 82 84 81
En los caacutelculos realizados hasta ahora la temperatura de gasificacioacuten ha sido una variable
ldquoimpuestardquo lo que podriacutea conseguirse mediante el calentamiento o refrigeracioacuten externa del
gasificador Por otro lado el ajuste de la composicioacuten del medio gasificante es decir de la
disponibilidad de oxiacutegeno y de vapor de agua permite controlar la energiacutea liberada en el
proceso lo que en un sistema adiabaacutetico y autoteacutermico se traduce en el control de la
temperatura La temperatura es en este caso la variable desconocida en el balance de energiacutea
que resulta de igualar las entalpiacuteas de las corrientes de entrada y de salida del reactor (ΔHentra =
ΔHsale) La evolucioacuten de la temperatura en funcioacuten de la composicioacuten del medio gasificante se
ha calculado considerando datos de equilibrio Este caacutelculo se ha realizado siguiendo un
meacutetodo iterativo ya que ΔHsale depende del rendimiento a los distintos productos (ec 42) y
eacuteste a su vez depende de la temperatura de reaccioacuten (la temperatura debe especificarse en el
software HSC Chemistry 61 para el caacutelculo de la cantidad de productos en equilibrio)
La Figura 413 muestra la evolucioacuten de la temperatura de equilibrio en funcioacuten de RE
(relacioacuten equivalente) y SC (relacioacuten molar entre el vapor de agua alimentado y el carbono del
soacutelido) en la gasificacioacuten de lodo y de char con mezclas de aire y vapor de agua
72 Resultados y discusioacuten
SC=0 SC=05 SC=1
Figura 413 Temperatura de equilibrio en la (a) gasificacioacuten de lodo y (b) gasificacioacuten de char en funcioacuten de la composicioacuten del medio de reaccioacuten
Como era de esperar la temperatura de equilibrio aumenta con RE y disminuye con SC
El uso de bajas RE da lugar a temperaturas de equilibrio mucho menores que el intervalo de
operacioacuten habitual en un gasificador (gt 700 ordmC) Esto indica que dichos datos de equilibrio se
alejaraacuten mucho del proceso real ya que por ejemplo en la realidad se obtendriacutea una fraccioacuten
de alquitraacuten muy alta algo que no ocurre en el caacutelculo del equilibrio
Como se puede observar en la Figura 413 la gasificacioacuten de char necesita una mayor RE
que la gasificacioacuten de lodo para mantener la misma temperatura de operacioacuten en ambos
procesos Por ejemplo en la gasificacioacuten de lodo se necesita una RE de 33 para operar de
forma autoteacutermica a 800 ordmC y con SC = 05 mientras que este valor alcanza el 45 en la
gasificacioacuten de char El aumento de RE favorece las reacciones de combustioacuten y por tanto la
produccioacuten de CO2 En este ejemplo el 44 del contenido inicial de carbono en el lodo termina
en forma de CO2 mientras que este valor alcanza el 52 en la gasificacioacuten de char La
presencia de CO2 en el gas de gasificacioacuten es indeseable debido al efecto de dilucioacuten del
contenido energeacutetico del gas (PCI = 43 MJmiddotm-3N en la gasificacioacuten de lodo y 31 MJmiddotm-3N en la
gasificacioacuten de char en el ejemplo planteado) y a la reduccioacuten de la formacioacuten de CO ya que la
formacioacuten de CO2 y el consumo de CO y viceversa estaacuten ligados a traveacutes de varias reacciones
como la reaccioacuten water-gas shift (ec 26) o la reaccioacuten de Boudouard (ec 24)
(b) Gasificacioacuten de char
10 15 20 25 30 35 40 45 50 300
500
700
900
1100
ER ()
Tem
pera
tura
de
equi
librio
(ordmC)
10 15 20 25 30 35 40 300
500
700
900
1100
(a) Gasificacioacuten de lodo
ER ()
Tem
pera
tura
de
equi
librio
(ordmC)
Resultados y discusioacuten 73
Secado teacutermico del lodo
Antes del tratamiento termoquiacutemico de los lodos de EDAR mediante piroacutelisis o
gasificacioacuten el secado teacutermico del mismo permite reducir su contenido de agua disminuyendo
asiacute el volumen del residuo y facilitando la manipulacioacuten del soacutelido resultante La energiacutea
necesaria para el secado teacutermico del lodo por kilogramo de residuo final (MJmiddotkg-1residuo final)
puede calcularse con la ec 45
Qsecado =mmateria seca ∙ Cplodo seco + mH2Olodo ∙ CpH2O(l) ∙ ∆T + mH2Oevap ∙ ∆HvapH2O
kg residuo inalkg lodo huacutemedo
(ec 45)
donde
- mmateria seca es el contenido de materia seca en el lodo huacutemedo (kgmiddotkg-1lodo huacutemedo)
- Cplodo seco es la capacidad caloriacutefica especiacutefica del lodo seco lodo (115middot10-3 MJmiddotkg-1middotK-1) Este
valor se obtuvo experimentalmente por calorimetriacutea diferencial de barrido (a 25 ordmC) y se
ha considerado constante en el intervalo de temperatura del secado
- mH2Olodo es el contenido de agua en el lodo huacutemedo (kgmiddotkg-1lodo huacutemedo) Despueacutes de la
deshidratacioacuten mecaacutenica del lodo mediante filtros prensa o centrifugacioacuten y antes del
secado teacutermico el contenido de humedad en el lodo se encuentra habitualmente en
torno al 70 (Manara y Zabaniotou 2012)
- CpH2O(l) es la capacidad caloriacutefica especiacutefica del agua liacutequida (418middot10-3 MJmiddotkg-1middotK-1) que es
praacutecticamente constante en el intervalo de temperatura del secado (Perry y Green 1999)
- ΔT es el incremento de temperatura durante el secado teacutermico (de 25 a 100 ordmC)
- mH2Oevap es la cantidad de agua evaporada durante el proceso de secado (kgmiddotkg-1lodo huacutemedo)
- ΔHvapH2O es la entalpiacutea de vaporizacioacuten del agua a la temperatura final del proceso (226
MJmiddotkg-1H2O a 100 ordmC) (Perry y Green 1999)
La Figura 414 muestra la evolucioacuten de la energiacutea necesaria para el secado teacutermico del
lodo en funcioacuten de su contenido de humedad inicial y final Por ejemplo se requieren casi 8
MJmiddotkg-1residuo final para reducir la fraccioacuten maacutesica de agua del 77 al 65 (datos que se
corresponden con el lodo utilizado) La energiacutea necesaria para el secado teacutermico del lodo se
reduce praacutecticamente a la mitad si el contenido de humedad inicial disminuye de un 77 a un
65 Esta reduccioacuten podriacutea lograrse mejorando la eficiencia de la etapa de deshidratacioacuten
mecaacutenica del lodo anterior al secado teacutermico
74 Resultados y discusioacuten
5 10 15 20 25 30 35 400
1
2
3
4
5
6
7
8
9
Q seca
do (M
Jmiddotkg-1
resid
uo fin
al)
Contenido final de humedad ( en masa)
Contenido inicial de humedad
Figura 414 Energiacutea necesaria para el secado teacutermico del lodo en funcioacuten de su contenido de humedad inicial y final
Evaluacioacuten energeacutetica global de los procesos termoquiacutemicos de dos y tres etapas
El coste energeacutetico para el tratamiento termoquiacutemico del lodo mediante los procesos de
dos etapas (secado y gasificacioacuten del lodo) y de tres etapas (secado del lodo piroacutelisis del lodo y
gasificacioacuten del char) se ha calculado como la suma de la energiacutea neta requerida o liberada en
las etapas individuales Este balance energeacutetico es soacutelo una primera aproximacioacuten ya que soacutelo
se ha tenido en cuenta la demanda energeacutetica de las etapas termoquiacutemicas propiamente
dichas sin incluir otros consumos energeacuteticos relacionados con el uso de bombas
compresoreshellipni rendimientos o eficiencias de combustioacuten al considerar el aprovechamiento
de los gases
- Secado teacutermico del lodo Se ha considerado una reduccioacuten del contenido de humedad en
el lodo del 65 al 65 lo que conlleva un coste energeacutetico de 44 MJmiddotkg-1lodo final (Figura
414)
- Piroacutelisis del lodo Si como en la gasificacioacuten se plantea la recuperacioacuten del contenido
energeacutetico del gas producto de la piroacutelisis (PCI y calor sensible y latente de los gases y
vapores con una eficiencia de intercambio de calor del 70) el calor neto del proceso de
piroacutelisis es -117 MJmiddotkg-1lodo El aprovechamiento del PCI del producto liacutequido (43 MJmiddotkg-1
FOL
y 32 MJmiddotkg-1FOP) no se ha incluido en el balance de energiacutea ya que algunas de sus
propiedades como su escasa estabilidad o alto contenido en nitroacutegeno deben ser
mejorados de cara a su posible uso como combustible (Fonts y cols 2012)
77 70 65
Resultados y discusioacuten 75
- Gasificacioacuten de lodo gasificacioacuten de char El calor neto de las etapas de gasificacioacuten
coincide con el numerador de la ec 44 Para la comparacioacuten de los procesos de dos y tres
etapas se requiere la misma base de caacutelculo en ambos por ejemplo 1 kg de lodo de EDAR
a la salida del secado teacutermico para ser pirolizado o gasificado Por lo tanto los datos
correspondientes a la gasificacioacuten de char (MJmiddotkg-1char) deben convertirse a MJmiddotkg-1
lodo
teniendo en cuenta el rendimiento a char obtenido en la piroacutelisis (052 kgcharmiddotkg-1lodo)
En la Figura 415 se muestran los datos de demanda total de energiacutea para los procesos de
dos y tres etapas en funcioacuten de las distintas condiciones de operacioacuten estudiadas en las etapas
de gasificacioacuten
RE = 32 RE = 23 RE = 19 RE = 17 RE = 12 SB = 039 SB = 027 SB = 052 SB = 071 SB = 052
Figura 415 Demanda total de energiacutea para el (a) proceso de dos etapas y (b) proceso de tres etapas
La demanda total de energiacutea en el proceso de dos etapas (secado y gasificacioacuten del lodo)
variacutea entre -243 y -581 MJmiddotkg-1lodo es decir la energiacutea contenida en el gas producto de la
gasificacioacuten de lodo es suficiente para cubrir la demanda energeacutetica del secado teacutermico y del
proceso de gasificacioacuten en siacute (proceso globalmente exoteacutermico) El balance de energiacutea es
todaviacutea maacutes favorable si se tiene en cuenta que parte de la humedad del lodo podriacutea utilizarse
como agente para la gasificacioacuten La exigencia del secado teacutermico podriacutea reducirse hasta dejar
un contenido de humedad final en el lodo del 19-32 (lo que equivaldriacutea a la alimentacioacuten de
vapor de agua estudiada SB = 027-071) o visto de otro modo el teacutermino Qvap del
numerador de la ec 44 (036-093 MJmiddotkg-1lodo) podriacutea evitarse si parte del secado del lodo se
realizase en el propio gasificador
Por otro lado el coste energeacutetico del proceso de tres etapas (secado y piroacutelisis de lodo y
gasificacioacuten de char) variacutea entre 157 y 264 MJmiddotkg-1lodo (proceso globalmente endoteacutermico) es
decir el aprovechamiento energeacutetico de los gases producto de la piroacutelisis de lodo y de la
Proceso de tres etapas
-3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=770ordmC
T=850ordmC
T=850ordmC T=810ordmC T=770ordmC
T=770ordmC
(b)
Demanda de energiacutea total (MJmiddotkg-1lodo)
Proceso de dos etapas
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=850ordmC
T=850ordmC
T=770ordmC
T=810ordmC T=770ordmC
T=770ordmC
(a)
Demanda de energiacutea total (MJmiddotkg-1lodo)
76 Resultados y discusioacuten
gasificacioacuten de char no es suficiente para cubrir el secado teacutermico del lodo Si en el balance
energeacutetico se incluye tambieacuten el aprovechamiento del poder caloriacutefico de la fraccioacuten orgaacutenica
del liacutequido de piroacutelisis (-392 MJmiddotkg-1lodo) el proceso de tres etapas se convierte en exoteacutermico
con un excedente de energiacutea que variacutea desde -235 a -128 MJmiddotkg-1lodo siempre teniendo en
mente las hipoacutetesis y simplificaciones realizadas
Los resultados energeacuteticos globales maacutes favorables se obtienen cuando en las etapas de
gasificacioacuten se opera simultaacuteneamente a la temperatura maacutes alta (850 ordmC) con la relacioacuten
equivalente maacutes alta (RE = 32) y con una relacioacuten maacutesica entre el vapor de agua y la materia
orgaacutenica moderada (SB = 039)
Resultados y discusioacuten 77
43 Eliminacioacuten de H2S de diferentes gases con cenizas de lodo
En esta seccioacuten se presentan los resultados de la eliminacioacuten de H2S de diferentes gases
sinteacuteticos a alta temperatura (600-800 ordmC) utilizando cenizas de combustioacuten y de gasificacioacuten
de lodos de EDAR Uno de los gases utilizados conteniacutea soacutelo H2S y N2 mientras que el otro
simulaba la composicioacuten de un gas de gasificacioacuten (H2 CO CO2 CH4 C2H6 C2H4 C2H2 N2 y H2S
Tabla 34) Ambos gases se alimentaron en forma seca y con cierto contenido de humedad
(30 vol en la mayoriacutea de los casos) para analizar el efecto de la humedad
Las curvas de ruptura del H2S obtenidas tras el paso de cada gas por el lecho de ceniza se
representan en la Figura 416 en funcioacuten de la atmoacutesfera de gas la temperatura de reaccioacuten y
el tipo de ceniza
(a) Mezcla H2SN2 seca b) Mezcla H2SN2 huacutemeda (30 vol humedad)
(c) Gas sinteacutetico de gasificacioacuten seco d) Gas sinteacutetico de gasificacioacuten huacutemedo (30 vol humedad)
H2S alimentado H2S a la salida del reactor en los experimentos ldquoblancosrdquo 800 ordmC 600 ordmC H2S a la salida del reactor cuando se utiliza la ceniza de combustioacuten 800 ordmC 600 ordmC H2S a la salida del reactor cuando se utiliza la ceniza de gasificacioacuten 800 ordmC 600 ordmC
Figura 416 Curvas de ruptura para el H2S evolucioacuten del caudal de H2S (mLNmiddotmin-1) a la salida del reactor con el tiempo
0 20 40 60 80 100 120 0
005
010
015
020
025
Tiempo (min) H 2S
a la
salid
a de
l rea
ctor
(mLN
middotmin
-1)
0 50 100 150 200 250 0
005
010
015
020
025
Tiempo (min) H 2S
a la
salid
a de
l rea
ctor
(mLN
middotmin
-1)
0 20 40 60 80 100 120 0
005
010
015
020
025
Tiempo (min) H 2S
a la
salid
a de
l rea
ctor
(mLN
middotmin
-1)
0 50 100 150 200 250 300 350 400 0
005
010
015
020
025
H 2S
a la
salid
a de
l rea
ctor
(mLN
middotmin
-1)
Tiempo (min)
78 Resultados y discusioacuten
La concentracioacuten de H2S en el gas de salida del reactor se mantuvo por debajo de 100
ppm (lo que equivaldriacutea a unos 5middot10-3 mLNmiddotmin-1) durante 300 min y 260 min cuando se
utilizaron las cenizas de combustioacuten y de gasificacioacuten respectivamente para desulfurar la
mezcla H2SN2 seca a 800 ordmC (Figura 416a) Sin embargo cuando se alimentoacute el gas sinteacutetico
de gasificacioacuten seco (Figura 416c) el punto de ruptura de las curvas se redujo a unos pocos
minutos (lt 20 min) excepto cuando se utilizoacute la ceniza de combustioacuten a 600 ordmC con la que el
punto de ruptura (H2S gt 100 ppm) se retrasoacute hasta los 165 min tiempo mucho menor que el
obtenido con la mezcla H2SN2 bajo las mismas condiciones de operacioacuten (245 min) Esto
demuestra el efecto negativo que ejerce alguno de los componentes del gas de gasificacioacuten en
el proceso de desulfuracioacuten La reduccioacuten de los oacutexidos de hierro presentes en las cenizas del
lodo (Figura 31) podriacutea explicar este comportamiento La presencia de H2 y CO en el gas de
gasificacioacuten crea una atmoacutesfera reductora que puede causar la reduccioacuten del Fe3O4 y del Fe2O3
a FeO o incluso a Fe elemental en el intervalo de temperatura de 700-1000 ordmC (Tamhankar y
cols 1981 Tseng y cols 2008 Westmoreland y Harrison 1976) Algunos estudios han
mostrado una menor capacidad del FeO y del Fe para reaccionar con H2S (Tseng et al 2008)
lo que explicariacutea los peores resultados de eliminacioacuten de H2S obtenidos al alimentar el gas
sinteacutetico de gasificacioacuten La eficiencia del proceso de desulfuracioacuten dependeraacute por tanto de la
competencia de las reacciones de reduccioacuten y de sulfuracioacuten (reaccioacuten con H2S) que pueden
verse afectadas en mayor o menor medida por las variaciones de temperatura
La presencia de vapor de agua en la atmoacutesfera de reaccioacuten tambieacuten ha mostrado un
impacto negativo en el proceso de eliminacioacuten de H2S de los gases (Figuras 416b y 416d) De
acuerdo con la reaccioacuten general de los oacutexidos metaacutelicos con H2S (ec 46) la termodinaacutemica
predice un efecto negativo del vapor de agua en el proceso de retencioacuten de H2S debido a la
regeneracioacuten simultaacutenea de los sulfuros metaacutelicos formados
MexOy (s) + y H2S (g) harr MexSy (s) + y H2O (g) ∆H lt0 (ec 46)
Bajo control cineacutetico el mayor o menor efecto del vapor de agua dependeraacute de la
diferencia de velocidad de las reacciones directa (formacioacuten del sulfuro metaacutelico) e inversa
(regeneracioacuten del oacutexido metaacutelico) lo que a su vez depende del material utilizado y de las
condiciones de operacioacuten (Cheah y cols 2009) Por ejemplo en su estudio de retencioacuten de H2S
con un material basado en ZnO Kim y cols (2007) comprobaron que el tiempo de ruptura en
la deteccioacuten de H2S llegaba a reducirse a la mitad en presencia de un 45 vol de vapor de
agua a 360 ordmC El impacto negativo del vapor de agua ha sido todaviacutea mayor en el presente
estudio observando una disminucioacuten del tiempo de ruptura del 85 cuando se utiliza la ceniza
Resultados y discusioacuten 79
de combustioacuten a cualquiera de las dos temperaturas (600 o 800 ordmC) con la mezcla H2SN2 o una
peacuterdida total de la actividad de la ceniza de gasificacioacuten
Como se observa en la Figura 416 la cantidad de H2S detectada a la salida del reactor en
los blancos (experimentos sin lecho de ceniza) tambieacuten se vio afectada por la presencia de
vapor de agua en la atmoacutesfera de gas La configuracioacuten experimental mostroacute cierta retencioacuten
del H2S al alimentar los gases secos y especialmente al operar a 800 ordmC lo que puede ser
debido a la reaccioacuten del H2S con las partes metaacutelicas calientes a la entrada y salida del reactor
En condiciones de humedad el caudal de H2S a la salida del reactor tiende hacia el valor de
entrada lo que significa que la presencia de vapor de agua evita la citada corrosioacuten de las
partes metaacutelicas Esto tambieacuten demuestra que la posible absorcioacuten del H2S en la fraccioacuten
acuosa recogida en el condensador no ocurre de forma importante
La Tabla 411 resume los resultados de tiempo de ruptura (truptura tiempo en el que la
concentracioacuten de H2S en el gas de salida supera las 100 ppm) cantidad de H2S eliminado del
gas hasta dicho tiempo de ruptura y contenido de azufre en las muestras soacutelidas despueacutes de
los experimentos tanto los datos reales obtenidos con el analizador elemental como la
concentracioacuten que cabriacutea esperar si todo el H2S eliminado del gas hubiese quedado retenido
en el soacutelido
La cantidad de H2S eliminado del gas hasta el truptura se ha calculado utilizando los datos de
los experimentos blancos como referencia
H2S eliminado del gas hasta truptura (mLN) = VH2S blanco minus VH2S experimento (ec 47)
donde VH2S experimento es la cantidad de H2S (mLN) que sale del reactor hasta truptura en el
experimento y VH2S blanco es la cantidad de H2S (mLN) que sale del reactor en el blanco realizado
en las mismas condiciones de operacioacuten y durante el mismo tiempo que el experimento El
volumen de H2S que abandona el reactor se puede calcular por integracioacuten de las curvas de
ruptura (Figura 416) y graacuteficamente se corresponde con el aacuterea que queda bajo cada curva
hasta truptura
80 Resultados y discusioacuten
Tabla 411 Resultados experimentales de las pruebas de eliminacioacuten de H2S
Tipo de ceniza
Relacioacuten H2OH2S
(gmiddotg-1) T (ordmC) truptura
(min)
H2S eliminado del gas hasta truptura (mLN)
Contenido de S real
(mg Smiddotg-1ceniza)
Contenido de S esperado
(mg Smiddotg-1ceniza)
Gas alimentado H2SN2
Ceni
za d
e
com
bust
ioacuten
0 600 245 54 58 plusmn 1 92
0 800 300 54 63 plusmn 4 100 45 600 40 7 215 plusmn 08 29 45 800 50 9 14 plusmn 01 32
225 700 70 14 208 plusmn 06 38 225 700 62 12 188 plusmn 02 37 225 700 62 12 166 plusmn 06 36
Ceni
za d
e
gasif
icac
ioacuten
0 600 30 4 232 plusmn 05 27
0 800 260 48 644 plusmn 07 100 45 600 5 2 49 plusmn 02 7 45 800 0 0 11 plusmn 01 8
225 700 0 0 141 plusmn 05 17 225 700 0 0 126 plusmn 05 18 225 700 0 0 118 plusmn 04 17
Gas alimentado gas sinteacutetico de gasificacioacuten
Ceni
za d
e co
mbu
stioacute
n 0 600 165 36 464 plusmn 06 64
0 800 13 1 55 plusmn 4 53 45 600 17 4 268 plusmn 05 32
45 800 5 1 85 plusmn 05 11
Ceni
za d
e ga
sific
acioacute
n 0 600 13 1 20 plusmn 1 19 0 800 0 0 332 plusmn 06 31
45 600 5 1 58 plusmn 02 11
45 800 5 1 45 plusmn 05 8
Valor medio plusmn desviacioacuten estaacutendar de tres medidas
La cantidad de H2S eliminada del gas hasta truptura es la uacutenica variable que ofrece
resultados comparables entre siacute (Tabla 37) Esta variable respuesta ha sido por tanto la uacutenica
utilizada para el anaacutelisis estadiacutestico de la influencia de los factores de operacioacuten (temperatura
H2OH2S y tipo de ceniza) cuando se alimenta la mezcla H2SN2 Los coeficientes de regresioacuten
lineal (β) obtenidos para los valores codificados de los factores se muestran en la Tabla 412 El
teacutermino significativo de la curvatura impide el uso del modelo lineal obtenido pero los
coeficientes β pueden utilizarse para comparar la influencia relativa de los factores
Resultados y discusioacuten 81
Tabla 412 Coeficientes de regresioacuten lineal (β) obtenidos del anaacutelisis ANOVA para la cantidad de H2S (mLN) eliminada de la mezcla H2SN2 hasta truptura
Valor medio βT βH2OH2S
βtipo de
ceniza βT-H2OH2S
βT-tipo de
ceniza βH2OH2S-tipo
de ceniza βT-H2OH2S-tipo de
ceniza Curvatura
2236 546 -1776 -891 -536 494 516 -589 Significativa
Los tres factores analizados (temperatura relacioacuten H2OH2S y tipo de ceniza de lodo) asiacute
como sus interacciones afectan de manera significativa a la cantidad de H2S que puede ser
eliminada del gas hasta alcanzar el tiempo de ruptura La Figura 417 muestra las superficies de
respuesta obtenidas con cada tipo de ceniza seguacuten los coeficientes de la Tabla 412 Estas
graacuteficas no se ajustan totalmente a la respuesta real de la variable debido a la existencia de
curvatura pero permiten ver de forma muy clara el efecto de los factores de operacioacuten
(a) Ceniza de combustioacuten (b) Ceniza de gasificacioacuten
Figura 417 Cantidad de H2S (mLN) eliminada de la mezcla H2SN2 hasta truptura con la (a) ceniza de combustioacuten y (b) ceniza de gasificacioacuten de lodo
Seguacuten los coeficientes β obtenidos del anaacutelisis ANOVA (Tabla 412) la relacioacuten H2OH2S es
el factor maacutes influyente y en teacuterminos generales su aumento afecta negativamente a la
cantidad de H2S que puede ser eliminada del gas hasta alcanzar el punto de ruptura (βH2OH2S = -
1776) Sin embargo la existencia de interacciones significativas entre los factores pone de
manifiesto que el impacto de la presencia de vapor de agua estaacute fuertemente condicionado
por la temperatura y el tipo de ceniza Asiacute la relacioacuten H2OH2S es el uacutenico teacutermino significativo
cuanto se utiliza la ceniza de combustioacuten (Figura 417a) En este caso la cantidad de H2S
eliminada del gas hasta la ruptura de la curva es independiente de la temperatura Esto parece
indicar que la variacioacuten observada en el truptura al modificar la temperatura es soacutelo una
consecuencia de la mayor retencioacuten de H2S por parte de que la configuracioacuten experimental a
800 ordmC que a 600 ordmC que se traduce en una disminucioacuten del caudal de H2S potencialmente
600
650
700
750
800 0
11
23
34
45
7
19
31
43
55
mLN
H2S
elim
inad
os h
asta
el t
iem
po d
e ru
ptur
a
Temperatura (ordmC) H2OH2S (gg)
600 650
700
750
800 0
11
23
34
45
0
7
14
21
29
36
43
50
mLN
H2S
elim
inad
os h
asta
el t
iem
po d
e ru
ptur
a
Temperatura (ordmC) H2OH2S (gg)
82 Resultados y discusioacuten
reactivo con la ceniza a 800 ordmC y por tanto en un mayor tiempo de ruptura Por otro lado la
temperatura y su interaccioacuten con la relacioacuten H2OO2 siacute que aparecen como teacuterminos
significativos cuando se utiliza la ceniza de gasificacioacuten Su capacidad de eliminacioacuten de H2S
desaparece casi por completo al operar tanto a la menor temperatura (600 ordmC) como con la
mayor relacioacuten H2OO2 (45 gmiddotg-1) y presenta un claro maacuteximo al operar a 800 ordmC en atmoacutesfera
seca (Figura 417b) alcanzando un resultado muy similar al de la ceniza de combustioacuten
El tipo de ceniza es por tanto otro factor clave en el proceso de eliminacioacuten de H2S (βtipo
de ceniza = -891) El valor negativo de este coeficiente β representa un mejor comportamiento de
la ceniza de combustioacuten (denotado en teacuterminos codificados como -1) frente a la ceniza de
gasificacioacuten (denotado como +1) La diferencia entre la actividad de ambas cenizas es maacutes
significativa a baja temperatura Por ejemplo la ceniza de combustioacuten a 600 ordmC fue capaz de
eliminar 54 mLN de H2S de la mezcla H2SN2 seca hasta alcanzar el tiempo de ruptura mientras
que la ceniza de gasificacioacuten apenas eliminoacute 4 mLN de H2S (Tabla 411) Puesto que las
propiedades texturales de ambas cenizas son muy similares (Tabla 32) las diferencias
observadas en su comportamiento estaacuten relacionadas con su composicioacuten quiacutemica La ceniza
de gasificacioacuten contiene una pequentildea cantidad de carbono (314 en masa) que podriacutea
contribuir a su peor rendimiento debido al obstaacuteculo que puede suponer para el acceso del
H2S a los centros metaacutelicos reactivos Sin embargo esta cantidad de carbono no parece lo
suficientemente alta como para ser la uacutenica causa de las diferencias observadas Los distintos
contenidos metaacutelicos y especies detectadas en ambas cenizas (seccioacuten 312) como
consecuencia de las diferentes atmoacutesferas reactivas en la combustioacuten y gasificacioacuten del lodo
parecen la causa maacutes razonable El contenido de Fe (mayor en la ceniza de combustioacuten) y su
estado quiacutemico (en forma de Fe2O3 en la ceniza de combustioacuten y como Fe3O4 en la ceniza de
gasificacioacuten) son algunas de las principales diferencias En el anaacutelisis por absorcioacuten de rayos X
en estructura fina (EXAFS) de muestras de Fe2O3 y Fe3O4 tras su reaccioacuten con H2S a 400 ordmC
Yoshimura y cols (1995) observaron una menor intensidad del pico correspondiente a la
coordinacioacuten Fe-S en la muestra sulfurada de Fe3O4 que en la de Fe2O3 indicando asiacute una
menor extensioacuten de la reaccioacuten de H2S con Fe3O4 Este hecho puede explicar la escasa
reactividad de la ceniza de gasificacioacuten con el H2S a la menor temperatura (600 ordmC) La cineacutetica
de la reaccioacuten del H2S con Fe3O4 en condiciones secas parece mejorar sensiblemente a 800 ordmC
alcanzando resultados muy similares a los del Fe2O3
Ademaacutes de la evolucioacuten del caudal de H2S a la salida del reactor el contenido de azufre en
las cenizas despueacutes de los experimentos de desulfuracioacuten se midioacute con un analizador
elemental Los resultados obtenidos estaacuten incluidos en la Tabla 411 (contenido de S real)
Resultados y discusioacuten 83
Todos estos datos no son directamente comparables entre siacute porque el tiempo de los
experimentos (y por tanto el grado de exposicioacuten a H2S) no fue el mismo en todos los casos
por lo que la evolucioacuten de esta variable no ha sido analizada por anaacutelisis ANOVA A pesar de
ello los datos de experimentos con la misma duracioacuten han mostrado un menor contenido de
azufre en la ceniza de gasificacioacuten que en la ceniza de combustioacuten cuando la temperatura de
operacioacuten era 600 o 700 ordmC mientras que el contenido de azufre en ambos soacutelidos fue similar
despueacutes de los experimentos realizados a 800 ordmC El maacuteximo contenido de azufre detectado en
las cenizas fue de 63-64 mgmiddotg-1ceniza (tras 390 min de experimento) y se obtuvo al alimentar la
mezcla H2SN2 seca y al operar con cualquiera de las dos cenizas a 800 ordmC En condiciones de
humedad el contenido de azufre retenido finalmente en el soacutelido se vio favorecido con la
disminucioacuten de la temperatura Ante la situacioacuten maacutes cercana a un proceso de desulfuracioacuten
real (gas sinteacutetico de gasificacioacuten huacutemedo) la ceniza de combustioacuten a 600 ordmC mostroacute la mayor
retencioacuten de azufre (268 mgmiddotg-1ceniza despueacutes de 120 minutos)
Estos contenidos de azufre medidos experimentalmente se han comparado con los
resultados que cabriacutea esperar si toda la cantidad de H2S eliminada del gas hubiese quedado
retenida en las cenizas tras los experimentos (ec 48)
Contenido de S esperado (mg S ∙gcenizaminus1 ) = Sinicial +
VH2S blancominusVH2S experimento
224 ∙ 32 (ec 48)
donde Sinicial es el contenido inicial de azufre en la ceniza (mgmiddotg-1ceniza Tabla 32) VH2S blanco
es la cantidad de H2S (mLN) que abandona el reactor durante el blanco (extrapolando a la
duracioacuten del experimento cuando ambas difieran) VH2S experimento es la cantidad total de H2S
(mLN) que abandona el reactor tras un experimento completo 224 (mLNmiddotmmol-1) es el
volumen de 1 mol de gas ideal en condiciones normales de presioacuten (1 atm) y temperatura (0
ordmC) y 32 (mgmiddotmmol-1) es la masa atoacutemica del azufre
Los resultados de contenido de azufre calculados con la ec 48 se han incluido en la
uacuteltima columna de la Tabla 411 Como se puede observar los datos reales de concentracioacuten
de azufre (medidos con el analizador elemental) son en general bastante maacutes bajos que los
datos calculados Puesto que los blancos mostraron que el H2S no quedaba absorbido en la
fraccioacuten de agua condensada a la salida del reactor la justificacioacuten maacutes probable para esta
falta de azufre es que se encuentre formando parte de otros gases que no han sido detectados
por el cromatoacutegrafo a la salida del reactor y cuya formacioacuten se haya visto potenciada por la
presencia de las cenizas Especialmente llamativa es la diferencia en los datos obtenidos para
el experimento 4 (800 ordmC y alimentacioacuten de la mezcla H2SN2 huacutemeda) ya que el contenido de
84 Resultados y discusioacuten
azufre esperado en el soacutelido era de unos 32 mgmiddotg-1ceniza y el analizador elemental soacutelo detectoacute
14 mgmiddotg-1ceniza
Con el objetivo de explicar estas diferencias desde un punto de vista termodinaacutemico se
realizaron simulaciones de equilibrio del proceso utilizando el software HSC Chemistry 61
Estas simulaciones resultaron bastante complejas debido a la propia configuracioacuten del sistema
(lecho fijo de soacutelido y alimentacioacuten continua de gas) que hace que la composicioacuten del soacutelido
cambie con el tiempo Para hacer frente a esta situacioacuten se realizaron sucesivas simulaciones
del equilibrio para pequentildeos intervalos de tiempo (10 min) hasta cubrir un tiempo de reaccioacuten
de 300 min El reactivo soacutelido considerado en la primera simulacioacuten fue la cantidad de Ca (en
forma de CaO) y de Fe (en forma de Fe2O3) presente en 1 g de ceniza de combustioacuten (Tabla
32) Tras el primer caacutelculo los compuestos soacutelidos resultantes de cada simulacioacuten constituiacutean
el soacutelido reactivo para la siguiente La cantidad de gas considerada como alimentacioacuten en cada
simulacioacuten fue la correspondiente a 10 min de experimento en el laboratorio De acuerdo con
este procedimiento las Figuras 418 y 419 muestran la evolucioacuten de la distribucioacuten del azufre
entre los principales productos sulfurados formados en el equilibrio (FexSy CaS H2S CaSO4 y
SO2) cuando se consideran la mezcla H2SN2 y el gas de gasificacioacuten respectivamente
(a) Mezcla H2SN2 seca 600 ordmC (b) Mezcla H2SN2 seca 800 ordmC
(c) Mezcla H2SN2 huacutemeda (30 vol H2O) 600 ordmC (d) Mezcla H2SN2 huacutemeda (30 vol H2O) 800 ordmC
FexSy CaS CaSO4 H2S SO2
Figura 418 Evolucioacuten de la distribucioacuten del azufre entre los principales productos de equilibrio bajo la atmoacutesfera de H2SN2
0102030405060708090
100
Dist
ribuc
ioacuten
del a
zufr
e (
)
0102030405060708090
100
Dist
ribuc
ioacuten
del a
zufr
e (
)
0102030405060708090
100
Dist
ribuc
ioacuten
del a
zufr
e (
)
0102030405060708090
100
Dist
ribuc
ioacuten
del a
zufr
e (
)
Resultados y discusioacuten 85
Bajo la atmoacutesfera de H2SN2 la formacioacuten de CaS se ve termodinaacutemicamente favorecida
frente a la formacioacuten de FexSy en la primera parte de las simulaciones de equilibrio (Figura
418) A medida que disminuye la cantidad disponible de CaO la formacioacuten de FexSy va
tomando maacutes importancia Ademaacutes de la formacioacuten de ambos sulfuros metaacutelicos la
termodinaacutemica predice la formacioacuten de SO2 y CaSO4 eacuteste uacuteltimo a consecuencia de la reaccioacuten
del CaO con el SO2 formado La formacioacuten de SO2 y CaSO4 asiacute como la fraccioacuten del H2S
alimentado que permanece como tal se ven favorecidas por la presencia de vapor de agua El
aumento de la temperatura desde 600 ordmC hasta 800 ordmC resulta favorable para la presencia de
SO2 La formacioacuten de este compuesto podriacutea explicar la falta de azufre en las cenizas despueacutes
de los experimentos De acuerdo con estas simulaciones de equilibrio los valores teoacutericos de
retencioacuten de azufre en las cenizas de combustioacuten seriacutean de 107 103 42 y 38 mgmiddotg-1ceniza para
los experimentos 1 2 3 y 4 respectivamente Los resultados experimentales medidos con el
analizador elemental fueron un 46 39 49 y 96 maacutes bajos que estos resultados
teoacutericos respectivamente
(a) Gas de gasificacioacuten seco 600 ordmC (b) Gas de gasificacioacuten seco 800 ordmC
(c) Gas de gasificacioacuten huacutemedo (30 vol H2O) (d) Gas de gasificacioacuten huacutemedo (30 vol H2O) 600 ordmC 800 ordmC
FexSy CaS H2S
Figura 419 Evolucioacuten de la distribucioacuten de azufre entre los principales productos de equilibrio bajo la atmoacutesfera del gas de gasificacioacuten
Bajo la atmoacutesfera reductora creada por el gas de gasificacioacuten (Figura 419) ni el SO2 ni el
CaSO4 aparecen como productos de equilibrio H2S CaS y FexSy son las principales especies que
0102030405060708090
100
Dist
ribuc
ioacuten
del a
zufr
e (
)
0102030405060708090
100
Dist
ribuc
ioacuten
del a
zufr
e (
)
0102030405060708090
100
Dist
ribuc
ioacuten
del a
zufr
e (
)
0102030405060708090
100
Dist
ribuc
ioacuten
del a
zufr
e (
)
86 Resultados y discusioacuten
contienen azufre en el equilibrio Tambieacuten se ha detectado la formacioacuten de COS aunque en
una proporcioacuten muy pequentildea La formacioacuten de COS en el ambiente reductor del gas de
gasificacioacuten (H2S + CO2 harr COS + H2O) ha sido observada por otros autores (Hepola y Simell
1997a) lo que podriacutea explicar la falta de azufre en las muestras de ceniza tras los
experimentos con el gas sinteacutetico de gasificacioacuten De acuerdo con las simulaciones de
equilibrio realizadas los valores teoacutericos de retencioacuten de azufre en las cenizas de combustioacuten
seriacutean de 85 84 42 y 41 mgmiddotg-1ceniza para los experimentos 20 21 22 y 23 respectivamente
Los resultados experimentales medidos con el analizador elemental fueron un 46 35 36
y 79 maacutes bajos que estos resultados teoacutericos respectivamente
La comparacioacuten de las Figuras 418 y 419 muestra el importante efecto de la atmoacutesfera
gaseosa en la distribucioacuten del azufre entre CaS y FexSy La presencia de CO2 en el gas de
gasificacioacuten puede explicar esta diferencia ya que este gas es el responsable de la reaccioacuten de
carbonatacioacuten del CaO (CaO + CO2 harr CaCO3) El exceso de CO2 desplaza esta reaccioacuten hacia la
formacioacuten de CaCO3 especialmente a bajas temperaturas lo que limita la formacioacuten de CaS a
partir de CaO
Por lo tanto ademaacutes de la competencia de las reacciones de sulfuracioacuten y reduccioacuten de
los oacutexidos de hierro explicada anteriormente la posible carbonatacioacuten del oacutexido de calcio es
otro factor a tener en cuenta para justificar los diferentes resultados experimentales obtenidos
para la mezcla H2SN2 y para el gas de gasificacioacuten
Una vez comprobada la posibilidad de retener azufre en las cenizas de lodo bajo ciertas
condiciones de operacioacuten la presencia de azufre en una de las cenizas (la del experimento 2)
fue caracterizada mediante otras teacutecnicas La Figura 420 muestra una imagen obtenida por
microscopiacutea electroacutenica de barrido (SEM) mediante electrones retrodispersados Los nuacutemeros
en dicha figura indican los puntos superficiales sobre los que se analizoacute la composicioacuten
elemental por espectroscopiacutea de energiacutea dispersiva de rayos X (EDX) Dichas fracciones
atoacutemicas se muestran en la Tabla 413
Figura 420 Imagen SEM obtenida mediante electrones retrodispersados
Resultados y discusioacuten 87
Tabla 413 Composicioacuten elemental (SEMEDX) en diferentes puntos superficiales de la ceniza resultante del experimento 2
Punto en la Figura 420
Composicioacuten elemental ( atoacutemico)
C O Na Mg Al Si P S Ca Fe Zn 1 296 10 20 17 43 238 16 359 2 08 579 34 37 24 133 11 66 106 03 3 22 459 05 15 57 92 115 86 63 88 4 07 690 03 01 03 276 06 09 5 09 507 03 33 62 93 122 46 70 56 6 16 579 24 14 40 117 04 41 161 04 7 335 08 13 78 35 237 41 253
Los porcentajes de C O Na Mg Al Si P S Ca Fe y Zn detectados en la superficie de la
ceniza reflejan la heterogeneidad del material Cabe destacar que los puntos con la mayor
concentracioacuten de S (238 atoacutemico en el punto 1 y 237 atoacutemico en el punto 7) son tambieacuten
los que presentan una concentracioacuten de Fe maacutes alta (359 y 253 respectivamente) lo que
sugiere la formacioacuten de sulfuros o sulfatos de hierro Por otro lado la presencia de S en otros
puntos de la superficie fue praacutecticamente inexistente como en el punto 4 formado
principalmente por O y Si (probablemente en forma de SiO2) o en el punto 6 en el que a pesar
de la importante presencia de Fe (161 atoacutemico) soacutelo se encontroacute un 04 atoacutemico de S En
este uacuteltimo punto asiacute como en los puntos 2 y 3 la alta presencia de Fe coincide con una alta
concentracioacuten de P que indica la presencia de fosfatos de hierro tambieacuten detectados por XRD
(Figura 31)
Por uacuteltimo la Figura 421 muestra el espectro XPS correspondiente al orbital 2p del S para
la muestra de ceniza resultante del experimento 2 en el que se reflejan los diferentes
entornos quiacutemicos del S El pico localizado entre 160 y 164 eV refleja la presencia de sulfuros
metaacutelicos (Sn-2) y el pico obtenido en torno a 169 eV se corresponde con estados maacutes oxidados
del S (SO4-2) La presencia de ambas especies fue predicha en las simulaciones del equilibrio
Figura 421 Espectro XPS correspondiente a la regioacuten S 2p para la ceniza resultante del experimento 2
88 Resultados y discusioacuten
En resumen los experimentos de desulfuracioacuten de gases sinteacuteticos han demostrado la
capacidad de las cenizas del lodo de EDAR especialmente de las generadas en el proceso de
combustioacuten para eliminar H2S de corrientes gaseosas bajo determinadas condiciones Dado
que el rendimiento de desulfuracioacuten se reduce draacutesticamente en presencia de vapor de agua
el uso de las cenizas para la eliminacioacuten de H2S resultaraacute mucho maacutes eficiente en la limpieza de
un gas seco como podriacutea ser el gas de piroacutelisis de lodos de EDAR despueacutes de la condensacioacuten
de los vapores El uso de las cenizas resultantes de la combustioacuten del char de piroacutelisis para la
desulfuracioacuten del propio gas de piroacutelisis plantea una interesante opcioacuten para reintegrar el
subproducto soacutelido en el proceso
La retencioacuten de azufre con las cenizas de combustioacuten de lodo a 600 ordmC bajo el gas sinteacutetico
de gasificacioacuten seco fue de 46 mgmiddotg-1ceniza antes de su saturacioacuten (Tabla 411) Suponiendo un
comportamiento similar de las cenizas resultantes de la combustioacuten del char de piroacutelisis y
teniendo en cuenta los rendimientos maacutesicos a ceniza (390 gmiddotkg-1lodo Tabla 31) y a azufre
gaseoso en forma de H2S (43 gmiddotkg-1lodo Tabla 48) se puede concluir que la ceniza generada
seriacutea suficiente para retener todo el H2S producido en el proceso
Resultados y discusioacuten 89
44 Estudio de la actividad de catalizadores de niacutequel en el reformado de alquitraacuten
En esta seccioacuten se presentan los resultados de la actividad de los distintos catalizadores
de NiAl2O3 preparados y modificados con diferentes promotores metaacutelicos (Ca Fe Mn y Cu)
en el reformado de compuestos modelo de alquitraacuten e hidrocarburos ligeros La atmoacutesfera
gaseosa de la gasificacioacuten se simuloacute mediante la mezcla de diferentes gases (H2 CO CO2 N2
CH4 C2H4 y H2S) vapor de agua y una mezcla de benceno (C6H6) tolueno (C7H8) y naftaleno
(C10H8) como compuestos modelo de alquitraacuten (Tabla 35)
La Figura 422 muestra la evolucioacuten de las concentraciones de C10H8 C6H6+C7H8 CH4 y C2H4
a la salida del reactor cuando se utiliza el catalizador de NiAl2O3 sin promotores calcinado a
900 ordmC y reducido en una atmoacutesfera de H2N2 antes del experimento En el eje superior de
abscisas se ha especificado la rampa de temperatura seguida durante el experimento (35 h
con cada temperatura)
Concentracioacuten de entrada Concentracioacuten de salida
Figura 422 Evolucioacuten de la concentracioacuten de C10H8 C6H6 + C7H8 CH4 y C2H4 en el gas de salida del reactor con el catalizador de NiAl2O3 calcinado a 900 ordmC y reducido previamente en atmoacutesfera de H2N2
0 2 4 6 8 10 12 14 16 18 20 22 24406080
100120140160180200220
Temperatura (oC)
conc
entra
cioacuten
de C
10H 8 (
ppm
v)
Tiempo (h)
900 850 800 900 750 700 900
0 2 4 6 8 10 12 14 16 18 20 22 24600800
10001200140016001800200022002400
Temperatura (oC)
conc
entra
cioacuten
de C
6H6 +
C7H
8 (pp
mv)
Tiempo (h)
900 850 800 900 750 700 900
0 2 4 6 8 10 12 14 16 18 20 22 24150001600017000180001900020000210002200023000
900 850 800 900 750 700 900Temperatura (oC)
conc
entra
cioacuten
de C
H 4 (pp
mv)
Tiempo (h)0 2 4 6 8 10 12 14 16 18 20 22 24
0100020003000400050006000700080009000
conc
entra
cioacuten
de C
2H4 (
ppm
v)
Tiempo (h)
900 850 800 900 750 700 900Temperatura (oC)
90 Resultados y discusioacuten
En el menor intervalo de temperatura (700-750 ordmC) la concentracioacuten de C2H4 en el gas de
salida fue la uacutenica que mostroacute una disminucioacuten apreciable respecto a su valor de entrada y a la
temperatura maacutes alta (900 ordmC) el CH4 fue el hidrocarburo con menor tasa de conversioacuten
Ademaacutes los sucesivos anaacutelisis realizados en cada intervalo de temperatura han mostrado
cierta tendencia a la alza en la concentracioacuten de algunos de los hidrocarburos lo que sugiere la
progresiva peacuterdida de actividad del catalizador Aunque la desactivacioacuten de los catalizadores
de niacutequel por envenenamiento con H2S puede evitarse en gran medida al operar a altas
temperaturas (Hepola y Simell 1997a) el aumento gradual de la concentracioacuten de CH4 y en
menor medida de la concentracioacuten de C6H6 + C7H8 durante las 35 primeras horas de
experimento a 900 ordmC demuestran que a pesar de la elevada temperatura el catalizador de
NiAl2O3 experimentoacute cierta peacuterdida de actividad en el reformado de estos compuestos El
C10H8 y el C2H4 no mostraron variacioacuten en su concentracioacuten cuando la temperatura era de 900
ordmC pero siacute a menores temperaturas siendo especialmente llamativo el raacutepido aumento de la
concentracioacuten de C2H4 durante las 35 h a 750 ordmC Esto corrobora el mencionado efecto de la
disminucioacuten de la temperatura sobre la desactivacioacuten del catalizador por envenenamiento con
H2S Ademaacutes todos estos resultados ponen de manifiesto la diferente sensibilidad de los
hidrocarburos ante la desactivacioacuten del catalizador de NiAl2O3 debido a la competencia de los
mismos por los centros activos de la superficie del catalizador y a los diferentes mecanismos
de descomposicioacuten
La actividad de los distintos catalizadores (NiAl2O3 NiCaAl2O3 NiFeAl2O3 NiCuAl2O3
y NiMnAl2O3) calcinados a 900 ordmC y reducidos previamente en una atmoacutesfera de H2N2 se
muestra en la Figura 423 Los puntos representados en dicha figura muestran los valores
medios de conversioacuten obtenidos en cada intervalo de temperatura (ec 49)
Conversioacuten i() = nientra minus nisalenientra
∙ 100 (ec 49)
donde nientra y nisale representan el caudal molar del compuesto i que entra y sale del
reactor respectivamente Puesto que la formacioacuten de C6H6 estaacute generalmente ligada a la
descomposicioacuten de C7H8 a traveacutes de diversas reacciones como la desalquilacioacuten con vapor de
agua (C7H8 + H2O harr C6H6 + CO + 2H2) o la hidrodesalquilacioacuten (C7H8 + H2 harr C6H6 + CH4) los
resultados de conversioacuten de ambos compuestos mono-aromaacuteticos se han analizado de forma
conjunta
Resultados y discusioacuten 91
0
20
40
60
80
100 0 35 7 105 14 175 21 245Tiempo (h)
900 850 800 900 750 700 900
conv
ersioacute
n de
l C10
H 8 (
)
Temperatura (oC)0
10
20
30
40
50
60Tiempo (h)
0 35 7 105 14 175 21 245
900 850 800 900 750 700 900conv
ersi oacute
n de
l C6H
6 + C
7H8 (
)
Temperatura (oC)
-10-505
1015202530 0 35 7 105 14 175 21 245
Tiempo (h)
900 850 800 900 750 700 900
conv
ersioacute
n de
l CH 4 (
)
Temperatura (oC)
0
20
40
60
80
100
Tiempo (h)0 35 7 105 14 175 21 245
900 850 800 900 750 700 900
conv
ersi oacute
n de
l C2H
4 (
)
Temperatura (oC)
NiAl2O3 NiCaAl2O3 NiFeAl2O3 NiCuAl2O3 NiMnAl2O3
Figura 423 Conversioacuten de los compuestos modelo de alquitraacuten y de los hidrocarburos ligeros obtenida con los distintos catalizadores calcinados a 900 ordmC y reducidos previamente en atmoacutesfera de H2N2
Los resultados del blanco realizado con un lecho de material inerte (SiC) no se han
representado en la Figura 423 pero evidenciaron una nula o escasa contribucioacuten del craqueo
teacutermico en la descomposicioacuten de los hidrocarburos Soacutelo la concentracioacuten de C2H4 fue
ligeramente modificada con una conversioacuten que varioacute del 43 (700 ordmC) al 77 (900 ordmC)
En presencia de los catalizadores tanto la conversioacuten de los alquitranes como la de los
hidrocarburos ligeros se vio claramente afectada por la temperatura Como era de esperar la
disminucioacuten de la temperatura supuso una notable reduccioacuten de la conversioacuten de todos ellos
debido a la disminucioacuten de la velocidad de las reacciones de reformado con vapor de agua (ec
29) o CO2 (ec 210) La conversioacuten del C10H8 fue inferior al 20 al operar a temperaturas
inferiores a 800 ordmC con todos los catalizadores mientras que la conversioacuten conjunta de C6H6 y
C7H8 soacutelo se mantuvo por encima del 30 al operar a 900 ordmC con algunos de los catalizadores
Soacutelo la adicioacuten de Mn resultoacute ventajosa para la conversioacuten del C10H8 que alcanzoacute un 80 a 900
ordmC mejorando en 10 puntos porcentuales la conversioacuten obtenida con el catalizador baacutesico de
NiAl2O3 La conversioacuten de C2H4 tambieacuten mejoroacute con la presencia de Mn pasando de un 92
92 Resultados y discusioacuten
en la primer etapa a 900 ordmC con el catalizador NiAl2O3 a praacutecticamente un 100 con el
catalizador NiMnAl2O3
Por otro lado la conversioacuten maacutexima de C6H6 + C7H8 se obtuvo con el catalizador de
NiAl2O3 En este caso ni la adicioacuten de Mn ni la de ninguacuten otro promotor consiguioacute mejorar la
conversioacuten inicial del catalizador aunque siacute su estabilidad en algunos casos La conversioacuten
media de C6H6 + C7H8 con NiAl2O3 pasoacute de un 55 en la primera etapa a 900 ordmC a un 49 en la
segunda y a un 40 en la tercera etapa a 900 ordmC mostrando asiacute una importante peacuterdida de
actividad a lo largo del experimento La incorporacioacuten de Mn mantuvo los niveles de
conversioacuten de C6H6 + C7H8 maacutes cercanos entre siacute en las tres etapas a 900 ordmC (35-40)
La importante peacuterdida de actividad del catalizador de NiAl2O3 tambieacuten fue observada en
los datos de conversioacuten del CH4 ya que se pasoacute de un valor promedio del 28 en la primera
etapa a 900 ordmC a un 16 en la tercera etapa a la misma temperatura La incorporacioacuten de Mn
en el catalizador permitioacute estabilizar la conversioacuten del CH4 en torno al 24-25 en las tres
etapas a 900 ordmC En el caso del CH4 tambieacuten llaman la atencioacuten los valores negativos de
conversioacuten obtenidos con los catalizadores NiCuAl2O3 y NiFeAl2O3 a 700-800 ordmC lo que
puede deberse a un aumento de la produccioacuten de este compuesto a traveacutes de las reacciones
de metanizacioacuten (CO + 3H2 harr CH4 + H2O C + 2H2 harr CH4)
En definitiva soacutelo la adicioacuten de Mn resultoacute favorable para la actividad o estabilidad del
catalizador de NiAl2O3 calcinado a 900 ordmC y reducido en atmoacutesfera de H2 antes del
experimento La adicioacuten de los demaacutes metales (Ca Fe y Cu) resultoacute perjudicial para la
conversioacuten de todos los hidrocarburos analizados La reduccioacuten de la superficie especiacutefica del
catalizador al incorporar la segunda fase metaacutelica (Tabla 33) parece la explicacioacuten maacutes sencilla
Dado que esta reduccioacuten tambieacuten se produjo al antildeadir el Mn las mejoras observadas con la
incorporacioacuten de dicho metal pueden estar relacionadas con una menor sinterizacioacuten del
catalizador en la fase previa de reduccioacuten llevada a cabo a 900 ordmC
Los demaacutes compuestos gaseosos que forman parte de la atmoacutesfera reactiva (H2 CO y CO2)
aparecen involucrados en diversas reacciones tanto como en forma de reactivos como de
productos lo que hace que el caudal alimentado de cada uno de ellos pueda verse aumentado
o disminuido al atravesar el lecho de catalizador La Figura 424 muestra el cociente entre los
caudales molares de H2 CO y CO2 medidos a la salida y a la entrada del reactor Los datos
representados reflejan los valores medios obtenidos en cada intervalo de temperatura Los
valores superiores a 1 indican una produccioacuten neta del compuesto y los inferiores a 1 se
corresponden con un consumo neto
Resultados y discusioacuten 93
0910111213141516
900 850 800 900 750 700 900
H 2 sal
ida
H2 e
ntra
da
(rela
cioacuten
mol
ar)
Temperatura (oC)
Tiempo (h)0 35 7 105 14 175 21 245
08091011121314151617
900 850 800 900 750 700 900
CO s
alid
a C
O e
ntra
da
(rela
ci oacuten
mol
ar)
Temperatura (oC)
0 35 7 105 14 175 21 245Tiempo (h)
08
09
10
11
12
900 850 800 900 750 700 900
CO2 s
alid
a C
O2 e
ntra
da
(rela
cioacuten
mol
ar)
Temperatura (oC)
Tiempo (h)0 35 7 105 14 175 21 245
NiAl2O3 NiCaAl2O3 NiFeAl2O3 NiCuAl2O3 NiMnAl2O3
Figura 424 Cociente entre los caudales molares de H2 CO y CO2 a la salida y a la entrada del reactor para los distintos catalizadores calcinados a 900 ordmC y reducidos previamente en atmoacutesfera de H2N2
Como se puede observar en la Figura 424 el flujo de H2 obtenido a la salida del reactor
aumentoacute con respecto al valor de entrada con todos los catalizadores utilizados Este aumento
se vio favorecido al aumentar la temperatura de reaccioacuten alcanzando un incremento cercano
al 50 en algunos casos Al igual que ocurrioacute con los datos de conversioacuten del CH4 y del C6H6 +
C7H8 la produccioacuten de H2 con el catalizador de NiAl2O3 se redujo en la uacuteltima parte del
experimento realizada a 900 ordmC reflejando la inestabilidad del catalizador Por otro lado el
caudal de salida de CO tambieacuten mostroacute un notable aumento respecto a su valor de entrada a
temperaturas por encima de 800 ordmC pero a diferencia del flujo de H2 su caudal se mantuvo
praacutecticamente constante o incluso disminuyoacute con algunos de los catalizadores al operar a 700-
750 ordmC Los resultados de CO2 mostraron la tendencia opuesta ya que los maacuteximos en su
produccioacuten se obtuvieron al reducir la temperatura de reaccioacuten El aumento de la velocidad de
las reacciones de reformado del alquitraacuten con la temperatura (control cineacutetico) permite
explicar estos resultados ya que el H2 y el CO son productos en estas reacciones (ec 29 y
210) mientras que el CO2 es el reactivo en las reacciones de reformado en seco (ec 210) En
cuanto a las demaacutes reacciones en fase gas el hecho de alcanzar o no el equilibrio quiacutemico es
94 Resultados y discusioacuten
un aspecto clave para explicar el efecto de la temperatura debido a la existencia de reacciones
exoteacutermicas como la reaccioacuten water-gas shift (ec 26) que no se ven favorecidas con el
aumento de la temperatura
El equilibrio del proceso se ha calculado con el software HSC Chemistry 61 de forma
anaacuteloga a como se hizo en la seccioacuten 41 Las simulaciones realizadas muestran que la
presencia de alquitraacuten (naftaleno benceno y tolueno) en el gas de equilibrio es praacutecticamente
despreciable en todo el intervalo de temperatura analizado (700-900 ordmC) El H2 CO y CO2 son
los compuestos mayoritarios La Tabla 414 muestra el cociente entre el caudal molar teoacuterico
de H2 CO y CO2 a la salida del reactor (en condiciones de equilibrio) y el caudal molar
alimentado de cada uno de ellos
Tabla 414 Cociente entre el caudal molar teoacuterico de H2 CO y CO2 a la salida del reactor en condiciones de equilibrio y el caudal molar alimentado de cada uno
700 ordmC 750 ordmC 800 ordmC 850 ordmC 900 ordmC
H2 salida H2 entrada 211 206 200 195 190
CO salida CO entrada 172 186 199 210 220
CO2 salida CO2 entrada 112 104 098 092 087
La comparacioacuten de los datos experimentales con los datos teoacutericos muestra que el uso de
los catalizadores de niacutequel no fue suficiente para alcanzar el equilibrio quiacutemico lo que puede
tener su origen en la incompleta conversioacuten del alquitraacuten Los datos de produccioacuten de H2 y CO
en el equilibrio son considerablemente maacutes altos que los valores experimentales mientras que
la produccioacuten de CO2 en el equilibrio es ligeramente inferior o muy similar a los datos
experimentales Otra de las diferencias observadas es que la produccioacuten de H2 en el equilibrio
sin presencia de alquitraacuten se ve favorecida al disminuir la temperatura lo cual pone de
manifiesto el peso de la reaccioacuten water-gas shift (ec 26) en la evolucioacuten de la composicioacuten de
equilibrio del gas de gasificacioacuten
Efecto del procedimiento de reduccioacuten de los catalizadores
La Figura 425 muestra los resultados de conversioacuten de los compuestos modelo de
alquitraacuten y de los hidrocarburos ligeros obtenidos con los catalizadores calcinados a 900 ordmC y
utilizados sin un tratamiento previo de reduccioacuten La comparacioacuten de estos resultados con los
datos de la Figura 423 (obtenidos tras la reduccioacuten previa de los catalizadores en atmoacutesfera de
H2N2) permite analizar el efecto del tratamiento de reduccioacuten
Resultados y discusioacuten 95
0
20
40
60
80
100Tiempo (h)
0 35 7 105 14 175 21 245
900 850 800 900 750 700 900
conv
ersioacute
n de
l C10
H 8 (
)
Temperatura (oC)
0
10
20
30
40
50
60
900 850 800 900 750 700 900conv
ersioacute
n de
l C6H
6 + C
7H8 (
)
Temperatura (oC)
0 35 7 105 14 175 21 245Tiempo (h)
-10-505
1015202530
Tiempo (h)
900 850 800 900 750 700 900
conv
ersioacute
n de
l CH 4 (
)
Temperatura (oC)
0 35 7 105 14 175 21 245
0
20
40
60
80
100Tiempo (h)
900 850 800 900 750 700 900
conv
ersi oacute
n de
l C2H
4 (
)
Temperatura (oC)
0 35 7 105 14 175 21 245
NiAl2O3 NiCaAl2O3 NiFeAl2O3 NiCuAl2O3 NiMnAl2O3
Figura 425 Conversioacuten de los compuestos modelo de alquitraacuten y de los hidrocarburos ligeros obtenida con los distintos catalizadores calcinados a 900 ordmC y utilizados sin pretratamiento de reduccioacuten
La conversioacuten media de C10H8 obtenida a 900 ordmC en el primer tramo de los experimentos
realizados sin reduccioacuten previa de los catalizadores de NiFeAl2O3 NiCuAl2O3 y NiAl2O3
mejoroacute en 25 20 y 5 puntos porcentuales respectivamente con respecto a los resultados
obtenidos tras la reduccioacuten de los catalizadores en atmoacutesfera de H2N2 Una posible
explicacioacuten para esto puede ser la desactivacioacuten de dichos catalizadores durante la etapa
previa de reduccioacuten que se realizoacute a 900 ordmC pudiendo causar la sinterizacioacuten de algunas
partiacuteculas de niacutequel Por otro lado la conversioacuten del C10H8 con el catalizador de NiMnAl2O3 se
mantuvo en valores muy similares en ambos casos Esto parece corroborar la hipoacutetesis antes
mencionada del efecto positivo de la adicioacuten de manganeso en la disminucioacuten de la
sinterizacioacuten de las partiacuteculas de niacutequel
Los mejores resultados de conversioacuten de C10H8 se obtuvieron con los catalizadores de
NiAl2O3 y NiMnAl2O3 que no mostraron grandes diferencias para este compuesto (80 de
conversioacuten a 900 ordmC) Sin embargo igual que en el caso anterior el catalizador baacutesico de
NiAl2O3 fue el maacutes activo en la conversioacuten de C6H6 + C7H8 y en este caso tambieacuten en la
96 Resultados y discusioacuten
conversioacuten de CH4 En cuanto a la conversioacuten de C2H4 el efecto maacutes importante del
procedimiento de reduccioacuten se observoacute en el catalizador de NiFeAl2O3 para el que la
conversioacuten media de C2H4 a 900 ordmC pasoacute de un 70-75 cuando se redujo previamente en
atmoacutesfera de H2 a un 95 cuando se evitoacute dicho pretratamiento igualando en este uacuteltimo
caso los resultados obtenidos con NiAl2O3 y NiMnAl2O3
Efecto de la temperatura de calcinacioacuten de los catalizadores
La Figura 426 muestra los resultados de conversioacuten de los compuestos modelo de
alquitraacuten y de los hidrocarburos ligeros obtenidos con los catalizadores calcinados a 700 ordmC y
utilizados sin un tratamiento previo de reduccioacuten La comparacioacuten de estos resultados con los
datos de la Figura 425 (correspondientes a los catalizadores calcinados a 900 ordmC y utilizados sin
pretratamiento de reduccioacuten) permite analizar la influencia de la temperatura de calcinacioacuten
0
20
40
60
80
100
900 850 800 900 750 700 900
conv
ersioacute
n de
l C10
H 8 (
)
Temperatura (oC)
0 35 7 105 14 175 21 245Tiempo (h)
0
10
20
30
40
50
60
900 850 800 900 750 700 900conv
ersioacute
n de
l C6H
6 + C
7H8 (
)
Temperatura (oC)
0 35 7 105 14 175 21 245Tiempo (h)
-10-505
1015202530
900 850 800 900 750 700 900
conv
ersioacute
n de
l CH 4 (
)
Temperatura (oC)
Tiempo (h)0 35 7 105 14 175 21 245
0
20
40
60
80
100
900 850 800 900 750 700 900
conv
ersioacute
n de
l C2H
4 (
)
Temperatura (oC)
Tiempo (h)0 35 7 105 14 175 21 245
NiAl2O3 NiCaAl2O3 NiFeAl2O3 NiCuAl2O3 NiMnAl2O3
Figura 426 Conversioacuten de los compuestos modelo de alquitraacuten y de los hidrocarburos ligeros obtenida con los distintos catalizadores calcinados a 700 ordmC y utilizados sin pretratamiento de reduccioacuten
Resultados y discusioacuten 97
Cuando los soacutelidos se calcinaron a 700 ordmC el catalizador baacutesico de NiAl2O3 fue el maacutes
activo para la conversioacuten de todos los compuestos seguido muy de cerca por los catalizadores
de NiMnAl2O3 y NiFeAl2O3 en la conversioacuten de C10H8 y C2H4
La disminucioacuten de la temperatura de calcinacioacuten de 900 a 700 ordmC en los catalizadores de
NiAl2O3 NiCaAl2O3 y NiFeAl2O3 supuso una mejora de 10 25 y 15 puntos porcentuales
respectivamente en el valor medio de conversioacuten de C10H8 obtenido a 900 ordmC De hecho si se
comparan los resultados de todos los experimentos que forman este estudio cataliacutetico el
catalizador baacutesico de NiAl2O3 calcinado a 700 ordmC fue el que dio lugar a la maacutexima conversioacuten
de C10H8 (87) La influencia de la temperatura de calcinacioacuten en la conversioacuten de C6H6 + C7H8
no fue en general tan significativa como en el caso del C10H8 Por otro lado la conversioacuten de
CH4 a 900 ordmC mejoroacute en torno a 5 puntos porcentuales con todos los soacutelidos al reducir la
temperatura de calcinacioacuten de 900 a 700 ordmC (salvo en el catalizador baacutesico de NiAl2O3 que se
mantuvo muy similar) Teniendo en cuenta que la maacutexima conversioacuten alcanzada para el CH4
fue de un 27 dicha variacioacuten supone una mejora sustancial Por uacuteltimo los resultados de
conversioacuten de C2H4 obtenidos a 900 ordmC con los catalizadores calcinados a 700 ordmC fueron mucho
maacutes similares entre siacute que en los casos anteriores alcanzando valores de conversioacuten superiores
al 90 en todos los casos
La variacioacuten de la superficie especiacutefica con la temperatura de calcinacioacuten (Tabla 33)
parece la explicacioacuten maacutes sencilla para justificar las diferencias observadas al utilizar los soacutelidos
calcinados a 700 ordmC y 900 ordmC Aunque la mayoriacutea de los soacutelidos calcinados a 700 ordmC mostraron
mejores resultados de conversioacuten hubo algunas excepciones Por eso hay que tener en
cuenta que la diferente temperatura de calcinacioacuten puede ocasionar cambios en la estructura
quiacutemica de los soacutelidos dando lugar a especies maacutes o menos activas que puedan contrarrestar
la disminucioacuten de la superficie especiacutefica Esto se intento comprobar con los anaacutelisis XRD
mostrados a continuacioacuten aunque los resultados obtenidos no fueron muy concluyentes
Caracterizacioacuten de los catalizadores tras los experimentos
Las muestras usadas de los catalizadores fueron caracterizadas despueacutes de los
experimentos mediante anaacutelisis elemental y difraccioacuten de rayos X La Tabla 415 presenta los
contenidos de carbono y azufre detectados en todas las muestras
98 Resultados y discusioacuten
Tabla 415 Contenidos de carbono y azufre en las muestras usadas de los catalizadores ( maacutesico)
Meacutetodo de preparacioacuten NiAl2O3 NiCaAl2O3 NiFeAl2O3 NiCuAl2O3 NiMnAl2O3
Calcinados a 900 ordmC Reduccioacuten previa en
H2N2 a 900 ordmC
C 17 plusmn 04 032 plusmn 001 039 plusmn 005 150 plusmn 02 023 plusmn 003
S 224 plusmn 006 16 plusmn 02 227 plusmn 004 31 plusmn 01 168 plusmn 006
Calcinados a 900 ordmC Sin pretratamiento
de reduccioacuten
C 24 plusmn 01 030 plusmn 006 040 plusmn 008 121 plusmn 02 040 plusmn 004
S 20 plusmn 03 18 plusmn 01 216 plusmn 003 344 plusmn 006 221 plusmn 001
Calcinados a 700 ordmC Sin pretratamiento
de reduccioacuten
C 28 plusmn 04 032 plusmn 007 017 plusmn 001 88 plusmn 03 012 plusmn 003
S 19 plusmn 03 206 plusmn 003 19 plusmn 02 374 plusmn 004 152 plusmn 007
Los datos mostrados en la tabla se corresonden con la media plusmn desviacioacuten estaacutendar de tres medidas
Los mayores contenidos de azufre y carbono se detectaron en las tres muestras usadas de
NiCuAl2O3 independientemente de la temperatura de calcinacioacuten y del procedimiento de
reduccioacuten El contenido de carbono alcanzoacute el 15 (en masa) en la muestra de
NiCuAl2O3calcinada a 900 ordmC y reducida en atmoacutesfera de H2N2 lo que parece indicar que la
deposicioacuten de carbono constituye un importante factor en la peacuterdida de actividad mostrada
por este catalizador El contenido de carbono se redujo a un 12 y a un 9 en las muestras no
reducidas y calcinadas a 900 ordmC y 700 ordmC respectivamente Estos datos de formacioacuten de
carbono junto con la formacioacuten neta de CH4 obtenida con este catalizador bajo ciertas
temperaturas (Figuras 423 425 y 426) ponen de manifiesto el importante papel del
catalizador NiCuAl2O3 en la reaccioacuten de metanizacioacuten heterogeacutenea (C + 2H2 harr CH4) La
incorporacioacuten de los demaacutes promotores al catalizador de NiAl2O3 supuso una disminucioacuten en
la deposicioacuten de carbono aunque esta reduccioacuten puede ser soacutelo una consecuencia de la menor
actividad de los catalizadores de NiFeAl2O3 y NiCaAl2O3 Sin embargo en el caso del
NiMnAl2O3 que mostroacute una actividad similar o incluso superior a la del NiAl2O3 en algunos
casos podriacutea significar una buena capacidad del Mn para reducir la deposicioacuten de carbono
(Koike y cols 2013) En el caso en el que la incorporacioacuten del Mn resultoacute en una mejora maacutes
clara de la actividad y estabilidad del catalizador (calcinacioacuten a 900 ordmC y reduccioacuten previa en
atmoacutesfera de H2N2) el contenido final de carbono se redujo del 17 al 023 al incorporar el
Mn Respecto a la influencia de la temperatura de calcinacioacuten o del procedimiento de
reduccioacuten sobre la deposicioacuten de carbono no se ha encontrado una tendencia uniforme sino
que depende de la segunda fase metaacutelica incorporada al catalizador Lo mismo ocurre con el
contenido final de azufre en los soacutelidos pero igual que ocurrioacute con el contenido de carbono el
catalizador de NiCuAl2O3 fue el que presentoacute una mayor retencioacuten La incorporacioacuten de los
demaacutes metales supuso la reduccioacuten del contenido de azufre en determinadas condiciones
aunque sin seguir un patroacuten general Estos resultados por siacute solos no permiten discernir si el
Resultados y discusioacuten 99
azufre quedoacute quimisorbido preferencialmente en los promotores o en los centros activos de
niacutequel
Para intentar identificar las posibles especies formadas las muestras de los catalizadores
se analizaron por difraccioacuten de rayos X (XRD) La Figura 427 muestra los difractogramas
obtenidos para las muestras usadas de los catalizadores calcinados a 900 ordmC y utilizados sin
tratamiento previo de reduccioacuten Los difractogramas obtenidos para las muestras sometidas al
tratamiento previo de reduccioacuten fueron praacutecticamente ideacutenticos a los que se muestran en esta
figura
Figura 427 Difractogramas XRD de las muestras usadas de los catalizadores calcinados a 900 ordmC y
utilizados sin pretratamiento de reduccioacuten
La comparacioacuten de los difractogramas de las muestras usadas (Figura 427) y sin usar
(Figura 32) muestra claras diferencias Tras los experimentos las muestras parecen maacutes
cristalinas y el NiAl2O4 ya no es la fase principal detectada en todos ellos Los difractogramas
de las muestras usadas son maacutes diferentes entre siacute permitiendo la deteccioacuten de los metales
antildeadidos como promotores principalmente en forma de aluminatos MnAl2O4 CuAlO4
FeAl2O4 y CaAl4O7 Especial atencioacuten debe mostrarse tambieacuten a la presencia de azufre que soacutelo
ha podido ser detectada en el catalizador de NiCuAl2O3 en forma de Ni3S2 y CuS2
20 30 40 50 60 70 80 90
NiCuAl2O3
2θ
CuAlO2 Ni3S2 CuS2 γ-Al2O3
MnAl2O4 NiAl2O4 γminus Al2O3 Ni
NiMnAl2O3
20 30 40 50 60 70 80 90
α-Al2O3 FeAl2O4
NiFeAl2O3
NiCaAl2O3
2θ
CaAl4O7 γ-Al2O3
100 Conclusiones y trabajos futuros
5 CONCLUSIONES Y TRABAJOS FUTUROS
Gasificacioacuten de lodo y de char con mezclas de vapor de agua y aire
Dado que el contenido de humedad de los lodos de EDAR antes de su secado teacutermico
puede alcanzar el 70 la gasificacioacuten de este residuo con la propia atmoacutesfera de vapor creada
a partir de su humedad parece una interesante opcioacuten para su valorizacioacuten energeacutetica Dada la
endotermicidad del proceso la adicioacuten de cierta cantidad de aire al medio de gasificacioacuten
puede proporcionar la energiacutea necesaria para el proceso a traveacutes de la combustioacuten parcial de
la materia prima
La produccioacuten de gas durante la gasificacioacuten de lodos de EDAR con mezclas de aire y vapor
de agua osciloacute entre 049 y 072 m3Nmiddotkg-1lodo (gas seco y libre de N2) El contenido energeacutetico de
este gas es suficiente para su aprovechamiento en calderas motores o turbinas (PCI = 41-62
MJmiddotm-3N) Ademaacutes la relacioacuten molar H2CO en el gas producto tambieacuten alcanzoacute el valor de 2 en
algunos casos valor requerido habitualmente para su uso como materia prima para la
produccioacuten de quiacutemicos (Wender 1996) El menor contenido de alquitraacuten obtenido en el gas
fue de 11 gmiddotm-3N valor que se encuentra entre los resultados habituales para la gasificacioacuten de
biomasa en lecho fluidizado (Corella y cols 2006) y que supera los valores liacutemite
recomendados para el aprovechamiento del gas
El char resultante de la piroacutelisis de lodos de EDAR aparece como una materia prima
preferible para la gasificacioacuten desde el punto de vista de la formacioacuten de alquitraacuten En este
proceso el contenido de alquitraacuten en el gas se redujo hasta niveles de 3 gmiddotm-3N en
determinadas condiciones de operacioacuten La reduccioacuten del contenido de materia volaacutetil en el
soacutelido despueacutes de la piroacutelisis explica esta diferencia Tambieacuten la produccioacuten de gas seco por
kilogramo de soacutelido se redujo praacutecticamente a la mitad al gasificar +char en lugar de lodo
siendo eacutesta la principal desventaja del proceso Sin embargo si el rendimiento a gas se calcula
en base seca y libre de cenizas para el soacutelido la gasificacioacuten de char ofrece un mayor
rendimiento a gas (099-147 m3Nmiddotkg-1char orgaacutenico frente a 089-132 m3Nmiddotkg-1
lodo orgaacutenico) con un
importante aumento de la produccioacuten de CO (45-85 mayor en teacuterminos de gramo por
kilogramo de soacutelido seco y libre de cenizas) y una fuerte disminucioacuten de la produccioacuten de CH4 y
C2Hx (80 menor) A pesar de estas diferencias el PCI de los gases de ambos procesos se
mantuvo en el mismo orden de magnitud y la relacioacuten molar H2CO en el gas de gasificacioacuten
de char tambieacuten alcanzoacute el valor de 2 bajo algunas condiciones de operacioacuten
Conclusiones y trabajos futuros 101
La temperatura fue el factor maacutes influyente para la mayoriacutea de las variables analizadas en
ambos procesos El aumento de la temperatura resultoacute favorable para reducir el contenido de
alquitraacuten y mejorar el rendimiento a gas y la produccioacuten especiacutefica de H2 CO asiacute como la
relacioacuten COCO2 en el gas producto el PCI del gas y la eficiencia energeacutetica de la gasificacioacuten
Por otro lado el incremento de temperatura resultoacute desfavorable desde el punto de vista de la
produccioacuten de H2S que se vio favorecida La composicioacuten del medio de gasificacioacuten (relacioacuten
H2OO2) desempentildeoacute un papel importante en la produccioacuten de H2 y en consecuencia en la
relacioacuten H2CO del gas producto que se vio favorecida con la mayor presencia de H2O La
produccioacuten de CH4 tambieacuten aumentoacute al incrementar la relacioacuten H2OO2 mientras que la
conversioacuten del carbono contenido inicialmente en los soacutelidos se vio perjudicada Por uacuteltimo el
aumento del caudal de agente gasificante alimentado por gramo de soacutelido (RG) mostroacute una
importante influencia en la produccioacuten de CO2 y H2S que se vieron favorecidas La Tabla 51
resume todas estas influencias de forma cualitativa (darr y uarr representan la disminucioacuten y el
aumento de la variable respuesta al aumentar el valor del factor de operacioacuten
respectivamente)
Tabla 51 Impacto cualitativo de los factores de operacioacuten en la gasificacioacuten de lodo y de char
Gasificacioacuten de lodo Gasificacioacuten de char
T RG H2OO2 T RG H2OO2
Fraccioacuten de C remanente en el soacutelido () darrdarrdarr darrdarr uarruarruarr darrdarrdarr darrdarr uarruarr Fraccioacuten de C convertido en gas () uarr uarr darr uarruarr uarr darr
Rendimiento a gas seco (m3Nmiddotkg-1 sin N2) uarruarr uarr --- uarruarr uarr ---
Contenido de alquitraacuten en el gas (gmiddotm-3N) darrdarrdarrdarr darr uarruarr darrdarrdarr darrdarrdarr darrdarrdarr Produccioacuten de cada compuesto gaseoso (gmiddotkg-1
soacutelido)
H2 uarruarruarr darr uarruarr uarruarr uarr uarruarr
CO uarruarruarr darr darrdarr uarruarruarr --- darr
CO2 --- uarruarr darrdarr darr uarruarr darr
CH4 uarr darr uarruarr --- --- uarruarr
C2Hx darr --- uarr --- --- ---
H2S uarruarr uarruarr --- uarruarruarruarr uarruarruarr ---
Relacioacuten molar H2CO en el gas producto darr uarr uarruarruarr darrdarr --- uarruarr
Relacioacuten molar COCO2 en el gas producto uarruarruarr darrdarr --- uarruarruarr darrdarr ---
PCI del gas (MJmiddotm-3STP) uarr darr uarr uarruarr --- uarr
Eficiencia energeacutetica de gasificacioacuten () uarruarr --- uarr uarruarr uarr uarr
102 Conclusiones y trabajos futuros
Aspectos energeacuteticos
Dado el caraacutecter endoteacutermico de muchas de las reacciones envueltas en el proceso de
gasificacioacuten con vapor de agua la composicioacuten de la mezcla de aire y vapor de agua utilizada
como agente gasificante es clave para conseguir un balance energeacutetico favorable Bajo las
mismas condiciones y simplificaciones realizadas en los caacutelculos la gasificacioacuten de char
requiere un mayor aporte de energiacutea externo que la gasificacioacuten de lodo para tener lugar
Entre los experimentos realizados el uso de una relacioacuten equivalente de 19 fue suficiente
para tener un proceso exoteacutermico de gasificacioacuten de lodo mientras que esta relacioacuten tuvo que
aumentarse hasta 32 en la gasificacioacuten de char El hecho de alcanzar el equilibrio quiacutemico en
el proceso conllevariacutea un mayor consumo energeacutetico pero a su vez la recuperacioacuten de energiacutea
en el gas producto podriacutea ser mayor de modo que la eficiencia energeacutetica de la gasificacioacuten
mejorariacutea en condiciones de equilibrio
Considerando la recuperacioacuten total e ideal del poder caloriacutefico inferior y del calor sensible
y latente de los gases el contenido energeacutetico del gas obtenido en la gasificacioacuten de lodo
podriacutea ser suficiente para cubrir la demanda energeacutetica del propio proceso de gasificacioacuten y de
la etapa previa de secado teacutermico No ocurre lo mismo si se consideran de forma conjunta los
productos gaseosos de la piroacutelisis de lodo y de la gasificacioacuten de char La fraccioacuten liacutequida del
proceso de piroacutelisis posee el contenido energeacutetico adicional necesario pero algunas de sus
propiedades como su inestabilidad o alto contenido de nitroacutegeno deben mejorarse antes de
plantear su posible uso como combustible
Eliminacioacuten de H2S de gases calientes con cenizas de lodo
Debido a su contenido en metales especialmente hierro y calcio las cenizas de lodos de
EDAR plantean una interesante y econoacutemica opcioacuten para la eliminacioacuten de H2S de gases a alta
temperatura En los experimentos realizados con diferentes gases sinteacuteticos las cenizas de
combustioacuten de lodo mostraron mejores resultados que las cenizas de gasificacioacuten Las
diferencias en su composicioacuten pueden explicar su diferente comportamiento Se detectoacute un
menor contenido de hierro en la ceniza de gasificacioacuten y en forma de distintas especies Fe2O3
en la ceniza de combustioacuten y Fe3O4 en la ceniza de gasificacioacuten
Los mejores resultados de eliminacioacuten de H2S se obtuvieron al alimentar la mezcla de
H2SN2 seca (5000 ppm) obteniendo un gas de salida praacutecticamente libre de H2S (lt100 ppm)
durante 300 min con la ceniza de combustioacuten a 800 ordmC En este caso el contenido final de
azufre en el soacutelido fue de 63 mgmiddotg-1ceniza La eliminacioacuten de H2S del gas se vio claramente
perjudicada por la presencia de vapor de agua en el medio de reaccioacuten debido a la
Conclusiones y trabajos futuros 103
regeneracioacuten simultaacutenea de los sulfuros metaacutelicos formados Ademaacutes los componentes del
gas de gasificacioacuten tambieacuten provocaron un efecto negativo en el proceso tanto por la
atmoacutesfera reductora creada por el H2 (que provoca la reduccioacuten de los oacutexidos de hierro) como
por la presencia de CO2 (que puede ocasionar la carbonatacioacuten de CaO) En las condiciones de
operacioacuten maacutes cercanas a lo que seriacutea un proceso real es decir con el gas sinteacutetico de
gasificacioacuten huacutemedo la ceniza de combustioacuten a 600 ordmC mostroacute los mejores resultados 50 min
hasta alcanzar 100 ppm de H2S en el gas de salida y un contenido de azufre en el soacutelido de 27
mgmiddotg-1ceniza despueacutes de 120 min de experimento
Los anaacutelisis elementales realizados a las cenizas despueacutes de los experimentos revelaron
que no toda la cantidad de H2S que habiacutea sido eliminada del gas estaba retenida en el soacutelido
Esto sugiere que el H2S no era el uacutenico compuesto con azufre en el gas de salida sino que otros
gases como SO2 o COS parecen formarse durante la eliminacioacuten de H2S de corrientes gaseosas
Reformado de compuestos modelo de alquitraacuten con catalizadores de niacutequel
El estudio cataliacutetico consistioacute en la evaluacioacuten de la estabilidad y actividad de diferentes
catalizadores de niacutequel soportados sobre aluacutemina y modificados con diferentes promotores
(Fe Ca Mn y Cu) para el reformado de alquitraacuten e hidrocarburos ligeros en presencia de H2S
Dada la afinidad de los oacutexidos de estos metales por el H2S con la incorporacioacuten de los
promotores se buscaba un menor envenenamiento de los sitios activos de niacutequel y por tanto
una mejora de la estabilidad del catalizador de NiAl2O3 ante la presencia de H2S En la mayoriacutea
de los casos no se obtuvieron dichos resultados sino que la incorporacioacuten de los promotores
resultoacute perjudicial para la actividad del catalizador debido probablemente a un exceso de
carga metaacutelica y a la reduccioacuten de la superficie especiacutefica (10-50 menor) De forma
excepcional los datos de conversioacuten de naftaleno metano y etileno mostraron una mejora de
la actividad y estabilidad del catalizador de NiAl2O3 al incorporarle Mn y tras haber sometido
a los catalizadores a un pretratamiento de reduccioacuten en atmoacutesfera de H2 El anaacutelisis elemental
de ambos catalizadores mostroacute un menor contenido de carbono depositado en el catalizador
de NiMnAl2O3 Por lo tanto aunque es necesario profundizar en el estudio el Mn parece
aportar cierta estabilidad al catalizador de NiAl2O3 para evitar la deposicioacuten de carbono y
tambieacuten el fenoacutemeno de sinterizacioacuten
TRABAJOS FUTUROS
En base a los resultados obtenidos en este trabajo se proponen las siguientes liacuteneas de
estudio para continuar la investigacioacuten relativa a la gasificacioacuten de lodos de EDAR y a la
limpieza del gas producto
104 Conclusiones y trabajos futuros
- Gasificacioacuten de lodo de EDAR huacutemedo (con un contenido de humedad del 20-30) para
obtener la atmoacutesfera de vapor de agua para su gasificacioacuten a partir del propio residuo Esto
requeriraacute probablemente el disentildeo de otro sistema de alimentacioacuten para el lodo huacutemedo
- Gasificacioacuten de lodo de EDAR no digerido anaeroacutebicamente con el fin de aprovechar toda
la fraccioacuten orgaacutenica acumulada durante el tratamiento de aguas residuales
- Completar el estudio energeacutetico de los procesos de gasificacioacuten y piroacutelisis de lodo
teniendo en cuenta factores exergeacuteticos
- Aplicacioacuten de las cenizas resultantes de la combustioacuten del char de piroacutelisis de lodo para la
eliminacioacuten de H2S del gas de piroacutelisis seco tras la condensacioacuten de los vapores
reutilizando asiacute el principal subproducto del proceso
- Profundizar en el estudio del manganeso como aditivo para mejorar la estabilidad de los
catalizadores de niacutequel Aplicacioacuten del catalizador NiMnAl2O3 en el proceso real de
gasificacioacuten de lodo en un reactor aguas abajo del gasificador
Referencias bibliograacuteficas 105
6 REFERENCIAS BIBLIOGRAacuteFICAS
Abu El-Rub Z Bramer EA Brem G (2004) Review of catalysts for tar elimination in biomass
gasification processes Industrial amp Engineering Chemistry Research 43 6911-6919
Adegoroye A Paterson N Li X Morgan T Herod AA Dugwell DR Kandiyoti R (2004) The
characterisation of tars produced during the gasification of sewage sludge in a spouted
bed reactor Fuel 83 1949-1960
Aacutelvarez-Rodriacuteguez R Clemente-Jul C (2008) Hot gas desulphurisation with dolomite sorbent in
coal gasification Fuel 87 3513-3521
Anis S Zainal ZA (2011) Tar reduction in biomass producer gas via mechanical catalytic and
thermal methods A review Renewable amp Sustainable Energy Reviews 15 2355-2377
Antal MJ Edwards WE Friedman KL Rogers FE (1979) Final Progress Report to US
Environmental Protection Agency Princeton University
Asadullah M (2014) Barriers of commercial power generation using biomass gasification gas A
review Renewable amp Sustainable Energy Reviews 29 201-215
Aznar M Gonzaacutelez AE Manyagrave JJ Saacutenchez JL Murillo MB (2007) Understanding the effect
of the transition period during the air gasification of dried sewage sludge in a fluidized bed
reactor International Journal of Chemical Reactor Engineering 5 A18
Aznar M Manyagrave JJ Garciacutea G Saacutenchez JL Murillo MB (2008) Influence of freeboard
temperature fluidization velocity and particle size on tar production and composition
during the air gasification of sewage sludge Energy amp Fuels 22 2840-2850
Aznar MP Caballero MA Gil J Martiacuten JA Corella J (1998) Commercial steam reforming
catalysts to improve biomass gasification with steam-oxygen mixtures 2 Catalytic tar
removal Industrial amp Engineering Chemical Research 37 2668-2680
Bacaicoa PG Bilbao R Uson C (1995) Sewage sludge gasification first studies Proceedings
of the Second Biomass Conference of the Americas Energy Environment and Agricultural
Industry 685-694
Bagreev A Bandosz TJ (2005) On the mechanism of hydrogen sulfide removal from moist air
on catalytic carbonaceous adsorbents Industrial amp Engineering Chemistry Research 44
530-538
Bandosz TJ (2002) On the adsorptionoxidation of hydrogen sulfide on activated carbons at
ambient temperatures Journal of Colloid and Interface Science 246 1-20
Bona S Guillen P Alcalde JG Garciacutea L Bilbao R (2008) Toluene steam reforming using
coprecipitated NiAl catalysts modified with lanthanum or cobalt Chemical Engineering
Journal 137 587-597
106 Referencias bibliograacuteficas
Brandt P Henriksen U (2000) Decomposition of tar in gas from updraft gasifier by thermal
cracking Proceedings of the first world conference on biomass for energy and industry
Bridgwater AV (1995) The technical and economic feasibility of biomass gasification for power
generation Fuel 74 631-653
Bridgwater AV Peacocke GVC (2000) Fast pyrolysis processes for biomass Renewable amp
Sustainable Energy Reviews 4 1-73
Campoy M Goacutemez-Barea A Vidal FB Ollero P (2009) Air-steam gasification of biomass in a
fluidised bed Process optimisation by enriched air Fuel Processing Technology 90 677-
685
Campoy M Goacutemez-Barea A Ollero P Nilsson S (2014) Gasification of wastes in a pilot
fluidized bed Fuel Processing Technology 121 63-69
Chaudhari ST Dalai AK Bakhshi NN (2003) Production of hydrogen andor syngas (H2+ CO)
via steam gasification of biomass-derived chars Energy amp Fuels 17 1062-1067
Cheah S Carpenter DL Magrini-Bair KA (2009) Review of mid- to high- temperature sulfur
sorbents for desulphurization of biomass- and coal-derived syngas Energy amp Fuels 23
5291-5307
ChemSpider database Royal Society of Chemistry Disponible en wwwchemspidercom
Consultado en enero de 2014
Chueh CF Swanson AC (1973) Estimation of liquid heat capacity The Canadian Journal of
Chemical Engineering 51 596-600
Concha A Andalaft A Fariacuteas O (2009) Gasificacioacuten de carboacuten para generacioacuten de energiacutea
eleacutectrica anaacutelisis con valoracioacuten de opciones reales Ingeniare Revista chilena de
ingenieriacutea 17 347-359
Corella J Toledo JM Molina G (2006) Calculation of the conditions to get less than 2 g
tarNm3 in a fluidized bed biomass gasifier Fuel Processing Technology 87 841-846
Courson C Makaga E Petit C Kiennemann A (2000) Development of Ni catalysts for gas
production from biomass gasification Reactivity in steam- and dry-reforming Catalysis
Today 63 427-437
De Andreacutes JM Narros A Rodriacuteguez ME (2011) Behaviour of dolomite olivine and alumina as
primary catalysts in air-steam gasification ofsewage sludge Fuel 90 521-527
Decisioacuten 2001118CE de la Comisioacuten de 16 de enero de 2001 por la que se modifica la
Decisioacuten 2000532CE en lo que se refiere a la lista de residuos DOCE nuacutem L 471 de 16
de 16 de febrero de 2001
De Lasa H Salaices E Mazumder J Lucky R (2011) Catalytic steam gasification of biomass
catalysts thermodynamics and kinetics Chemical Reviews 111 5404-5433
Referencias bibliograacuteficas 107
Delgado J Aznar MP Corella J (1997) Biomass gasification with steam in fluidized bed
effectiveness of CaO MgO and CaO-MgO for hot raw gas cleaning Industrial amp
Engineering Chemistry Research 36 1535-1543
Devi L Ptasinski KJ Janssen FJJG (2003) A review of the primary measures for tar
elimination in biomass gasification processes Biomass amp Bioenergy 24 125-140
Di Blasi C (2009) Combustion and gasification rates of lignocellulosic chars Progress in Energy
and Combustion Science 35 121-140
Directiva 86278CEE del Consejo de 12 de junio de 1986 relativa a la proteccioacuten del medio
ambiente y en particular de los suelos en la utilizacioacuten de los lodos de depuradora en la
agricultura DOUE nuacutem L 1816 de 4 de julio de 1986
Directiva 91271CEE del Consejo de 21 de mayo de 1991 sobre el tratamiento de aguas
residuales urbanas DOUE nuacutem L 13540 de 30 de mayo de 1991
Directiva 199931CEE del Consejo de 26 de abril de 1999 relativa al vertido de residuos
DOUE nuacutem L 1821 de 16 de julio de 1999
Directiva 200898CE del Parlamento Europeo y del Consejo de 19 de noviembre de 2008
sobre los residuos y por la que se derogan determinadas Directivas DOUE nuacutem L 3123
de 22 de noviembre de 2008
Directiva 201075UE del Parlamento Europeo y del Consejo de 24 de noviembre de 2010
sobre las emisiones industriales (prevencioacuten y control integrados de la contaminacioacuten)
DOUE nuacutem L 33417 de 17 de dieciembre de 2010
Domiacutenguez A Meneacutendez JA Pis JJ (2006) Hydrogen rich fuel gas production from the
pyrolysis of wet sewage sludge at high temperature Journal of Analytical and Applied
Pyrolysis 77 127-132
Dogru M Midilli A Howarth CR (2002) Gasification of sewage sludge using a throated
downdraft gasifier and uncertainty analysis Fuel Processing Technology 75 55-82
Dou B Gao J Sha X Baek SW (2003) Catalytic cracking of tar component from high-
temperature fuel gas Applied Thermal Engineering 23 2229-2239
Elseviers WF Verelst H (1999) Transition metal oxides for hot gas desulphurization Fuel 78
601-612
Engelen K Zhang YH Draelants DJ Baron GV (2003) A novel catalytic filter for tar removal
from biomass gasification gas Improvement of the catalytic activity in presence of H2S
Chemical Engineering Science 58 665-670
Fonts I Juan A Gea G Murillo MB Saacutenchez JL (2008) Sewage sludge pyrolysis in fluidized
bed 1 Influence of operational conditions on the product distribution Industrial amp
Engineering Chemistry Research 47 5376-5385
108 Referencias bibliograacuteficas
Fonts I Azuara M Gea G Murillo MB (2009) Study of the pyrolysis liquids obtained from
different sewage sludge Journal of Analytical and Applied Pyrolysis 85 184-191
Fonts I Gea G Azuara M Aacutebrego J Arauzo J (2012) Sewage sludge pyrolysis for liquid
production A review Renewable amp Sustainable Energy Reviews 16 2781-2805
Franco C Pinto F Gulyurtlu I Cabrita I (2003) The study of reactions influencing the biomass
steam gasification process Fuel 82 835-842
Fytili D Zabaniotou A (2008) Utilization of sewage sludge in EU application of old and new
methods - A review Renewable amp Sustainable Energy Reviews 12 116-140
Garciacutea G Cascarosa E Aacutebrego J Gonzalo A Saacutenchez JL (2011) Use of different residues for
high temperature desulphurization of gasification gas Chemical Engineering Journal 174
644-651
Gil J Aznar MP Caballero MA Franceacutes E Corella J (1997) Biomass gasification in fluidized
bed at pilot scale with steam-oxygen mixtures Product distribution for very different
operating conditions Energy amp Fuels 11 1109-1118
Gil J Corella J Aznar MP Caballero MA (1999a) Biomass gasification in atmospheric and
bubbling fluidized bed Effect of the type of gasifying agent on the product distribution
Biomass amp Bioenergy 17 389-403
Gil J Caballero MA Martiacuten JA Aznar MP Corella J (1999b) Biomass gasification with air in
fluidized bed effect of in-bed use of dolomite under different operation conditions
Industrial amp Engineering Chemistry Research 38 4226-4235
Gil-Lalaguna N Fonts I Gea G Murillo MB Laacutezaro L (2010) Reduction of water content in
sewage sludge pyrolysis liquid by selective on-line condensation of the vapors Energy amp
Fuels 24 6555-6564
Gonzaacutelez JF Romaacuten S Encinar JM Martiacutenez GJ (2009) Pyrolysis of various biomass
residues and char utilization for the production of activated carbons Journal of Analytical
and Applied Pyrolysis 85 134-141
Gutieacuterrez-Ortiz FJ Aguilera PG Ollero P (2014) Biogas desulfurization by adsorption on
thermally treated sewage-sludge Separation and Purification Technology 123 200-213
Han J Kim H (2008) The reduction and control technology of tar during biomass
gasificationpyrolysis An overview Renewable amp Sustainable Energy Reviews 12 397-
406
Harrison BK Seaton WH (1988) Solution to missing group-problem for estimation of ideal-
gas heat-capacities Industrial amp Engineering Chemistry Research 27 1536-1540
Haykiri-Acma H Yaman S Kucukbayrak S (2006) Gasification of biomass chars in steam-
nitrogen mixture Energy Conversion and Management 47 1004-1013
Referencias bibliograacuteficas 109
He P Luo S Cheng G Xiao B Cai L Wang J (2012) Gasification of biomass char with air-
steam in a cyclone furnace Renewable Energy 37 398-402
Hepola J Simell P (1997a) Sulphur poisoning of nickel-based hot gas cleaning catalysts in
synthetic gasification gas - I Effect of different process parameters Applied Catalysis B
Environmental 14 287-303
Hepola J Simell P (1997b) Sulphur poisoning of nickel-based hot gas cleaning catalysts in
synthetic gasification gas - II Chemisorption of hydrogen sulphide Applied Catalysis B
Environmental 14 305-321
Herguido J Corella J Gonzaacutelez-Saiz J (1992) Steam gasification of lignocellulosic residues in a
fIuidised bed at a small pilot scale Effect of the type of feedstock Industrial amp Engineering
Chemistry Research 31 1274-1282
Inguanzo M Domiacutenguez A Meneacutendez JA Blanco CG Pis JJ (2002) On the pyrolysis of
sewage sludge the influence of pyrolysis conditions on solid liquid and gas fractions
Journal of Analytical and Applied Pyrolysis 63 209-222
Judex JW Gaiffi M Burgbacher HC (2012) Gasification of dried sewage sludge Status of the
demonstration and the pilot plant Waste Management 32 719-723
Kelessidis A Stasinakis AS (2012) Comparative study of the methods used for treatment and
final disposal of sewage sludge in European countries Waste Management 32 1186-1195
Kiely G (1999) Ingenieriacutea ambiental Fundamentos entornos tecnologiacuteas y sistemas de
gestioacuten Ed McGraw-Hill Buenos Aires
Kim K Jeon SK Vo C Park CS Norbeck JM (2007) Removal of H2S from steam-
hydrogasifier product gas by zinc oxide sorbent Industrial amp Engineering Chemistry
Research 46 5848-5854
Kim YJ Lee SH Kim SD (2001) Coal gasification characteristics in a downer reactor Fuel 80
1915-1922
Kimura T Miyazawa T Nishikawa J Kado S Okumura K Miyao T Naito S Kunimori K
Tomishige K (2006) Development of Ni catalysts for tar removal by steam gasification of
biomass Applied Catalysis B Environmental 68 160-170
Kinoshita CM Wang Y Zhou J (1994) Tar formation under different biomass gasification
conditions Journal of Analytical and Applied Pyrolysis 29 169-181
Kinoshita CM Turn SQ Overend RP Bain RL (1997) Power generation potential of biomass
gasification systems Journal of Energy Engineering 123 88-99
Koike M Ishikawa C Li D Wang L Nakagawa Y Tomishige K (2013) Catalytic performance
of manganese-promoted nickel catalysts for the steam reforming of tar from biomass
pyrolysis to synthesis gas Fuel 103 122-129
110 Referencias bibliograacuteficas
Kumar A Jones DD Hanna MA (2009) Thermochemical biomass gasification a review of the
current status of the technology Energies 2 556-581
Li C Suzuki K (2009a) Tar property analysis reforming mechanism and model for biomass
gasification - An overview Renewable amp Sustainable Energy Reviews 13 594-604
Li C Hirabayashi D Suzuki K (2009b) Development of new nickel based catalyst for biomass
tar steam reforming producing H2-rich syngas Fuel Processing Technology 90 790-796
Lv PM Xiong ZH Chang J Wu CZ Chen Y Zhu JX (2004) An experimental study on
biomass air-steam gasification in a fluidized bed Bioresource Technology 95 95-101
Manara P Zabaniotou A (2012) Towards sewage sludge based biofuels via thermochemical
conversion - A review Renewable amp Sustainable Energy Reviews 16 2566-2582
Manganaro J Chen B Adeosun J Lakhapatri S Favetta D Lawal A (2011) Conversion of
residual biomass into liquid transportation fuel an energy analysis Energy amp Fuels 25
2711-2720
Manyagrave JJ Saacutenchez JL Gonzalo A Arauzo J (2005) Air gasification of dried sewage sludge in
a fluidized bed Effect of the operating conditions and in-bed use of alumina Energy amp
Fuels 19 629-636
Manyagrave JJ Saacutenchez JL Aacutebrego J Gonzalo A Arauzo J (2006) Influence of gas residence time
and air ratio on the air gasification of dried sewage sludge in a bubbling fluidised bed Fuel
85 2027-2033
Maacuterquez-Montesinos F Cordero T Rodriacuteguez-Mirasol J Rodriacuteguez JJ (2002) CO2 and steam
gasification of a grapefuit skin char Fuel 81 423-429
Martiacutenez JD Mahkamov K Andrade RV Lora EES (2012) Syngas production in downdraft
biomass gasifiers and its application using internal combustion engines Renewable
Energy 38 1-9
McKendry P (2002a) Energy production from biomass (part 1) overview of biomass
Bioresource Technology 83 37-46
McKendry P (2002b) Energy production from biomass (part 3) gasification technologies
Bioresource Technology 83 55-63
Meng X Jong W Pal R Verkooijen AHM (2010) In bed and downstream hot gas
desulphurization during solid fuel gasification a review Fuel Processing Technology 91
964-981
Miccio F Piriou B Ruoppolo G Chirone R (2009) Biomass gasification in a catalytic fluidized
reactor with beds of different materialsChemical Engineering Journal 154 369-374
Midilli A Dogru M Howarth CR Ling MJ Ayhan T (2001) Combustible gas production from
sewage sludge with a downdraft gasifier Energy Conversion and Management 42 157-172
Referencias bibliograacuteficas 111
Miyazawa T Kimura T Nishikawa J Kado S Kunimori K Tomishige K (2006) Catalytic
performance of supported Ni catalysts in partial oxidation and steam reforming of tar
derived from the pyrolysis of wood biomass Catalysis Today 115 254-262
Mondal P Dang GS Garg MO (2011) Syngas production through gasification and cleanup
for downstream applications - Recent developments Fuel Processing Technology 92
1395-1410
Narvaacuteez I Orio A Aznar MP Corella J (1996) Biomass gasification with air in an atmospheric
bubbling fluidized bed Effect of six operational variables on the quality of the produced
raw gas Industrial amp Engineering Chemistry Research 35 2110-2120
Neeft JPA Knoef HAM Zielke U Sjoumlstrom K Hasler P Simell PA Dorrington MA
Thomas L Abatzoglou N Deutch S Greil C Buffinga GJ Brage C Suomalinen M
(2002) Guideline for Sampling and Analysis of Tar and Particles in Biomass Producer Gases
(Tar protocol) Energy project EEN5-1999-00507
Nilsson S Goacutemez-Barea A Fuentes-Cano D (2012) Gasification reactivity of char from dried
sewage sludge in a fluidized bed Fuel 92 346-353
Nilsson S Goacutemez-Barea A Fuentes-Cano D Campoy M (2014) Gasification kinetics of char
from olive tree pruning in fluidized bed Fuel 125 192-199
Nipattummakul N Ahmed I Kerdsuwan S Gupta AK (2010) High temperature steam
gasification of wastewater sludge Applied Energy 87 3729-3734
Nisbet ICT Lagoy PK (1992) Toxic equivalency factors (TEFs) for polycyclic aromatic-
hydrocarbons (PAHs) Regulatory Toxicology and Pharmacology 16 290-300
Nishikawa J Nakamura K Asadullah M Miyazawa T Kunimori K Tomishige K (2008)
Catalytic performance of NiCeO2Al2O3 modified with noble metals in steam gasification
of biomass Catalysis Today 131 146-155
Nowicki L Antecka A Bedyk T Stolarek P Ledakowicz S (2011) The kinetics of gasification of
char derived from sewage sludge Journal of Thermal Analysis and Calorimetry 104 693-
700
Olivares A Aznar MP Caballero MA Gil J Franceacutes E Corella J (1997) Biomass gasification
produced gas upgrading by in-bed use of dolomite Industrial amp Engineering Chemistry
Research 36 5220-5226
Park NK Lee DH Lee JD Chang WC Ryu SO Lee TJ (2005) Effects of reduction of metal
oxide sorbents on reactivity and physical properties during hot gas desulphurization in
IGCC Fuel 84 2158-2164
112 Referencias bibliograacuteficas
Park HJ Park SH Sohn JM Park J Jeon JK Kim SS Park YK (2010) Steam reforming of
biomass gasification tar using benzene as a model compound over various Ni supported
metal oxide catalysts Bioresource Technology 101 S101-S103
Perry RH Green DW (1999) Perryrsquos Chemical Engineerrsquos Handbook 7th ed New York
McGraw-Hill
Petersen I Werther J (2005) Experimental investigation and modelling of gasification of
sewage sludge in the circulating fluidized bed Chemical Engineering and Processing
Process Intensification 44 717-736
Pinto F Franco C Andreacute RN Tavares C Dias M Gulyurtlu I Cabrita I (2003) Effect of
experimental conditions on co-gasification of coal biomass and plastics wastes with
airsteam mixtures in a fluidized bed system Fuel 82 1967-1976
Pokorna E Postelmans N Jenicek P Schreurs S Carleer R Yperman J (2009) Study of bio-
oils and solids from flash pyrolysis of sewage sludges Fuel 88 1344-1350
Ponzio A Kalisz S Blasiak W (2006) Effect of operating conditions on tar and gas composition
in high temperature airsteam gasification (HTAG) of plastic containing waste Fuel
Processing Technology 87 223-233
Primavera A Trovarelli A Andreussi P Dolcetti G (1998) The effect of water in the low-
temperature catalytic oxidation of hydrogen sulfide to sulfur over activated carbon
Applied Catalysis A General 173 185-192
Qin YH Feng J Li WY (2010) Formation of tar and its characterization during air-steam
gasification of sawdust in a fluidized bed reactor Fuel 89 1344-1347
Rapagnagrave S Jand N Kiennemann A Foscolo PU (2000) Steam-gasification of biomass in a
fIuidized-bed of olivine particles Biomass amp Bioenergy 19 187-197
Raveendran K Ganesh A (1998) Adsorption characteristics and pore-development of biomass-
pyrolysis char Fuel 77 769-781
Richardson SM Gray MR (1997) Enhancement of residue hydroprocessing catalysts by
doping with alkali metals Energy amp Fuels 11 1119-1126
Ros A Montes-Moraacuten M Fuente E Nevskaia DM Martin MJ (2006) Dried sludges and
sludge-based chars for H2S removal at low temperature influence of sewage sludge
characteristics Environmental Science amp Technology 40 302-209
Rulkens W (2008) Sewage sludge as a biomass resource for the production of energy
Overview and assessment of the various options Energy amp Fuels 22 9-15
Salleh MAM Kisiki NH Yusuf HM Ghani WAK (2010) Gasification of biochar from empty
fruit bunch in a fluidized bed reactor Energies 3 1344-1352
Referencias bibliograacuteficas 113
Sato K Fujimoto K (2007) Development of new nickel based catalyst for tar reforming with
superior resistance to sulfur poisoning and coking in biomass gasification Catalysis
Communications 8 1697-1701
Scott SA Davidson JF Dennis JS Fennell PS Hayhurst AN (2005) The rate of gasification
by CO2 of chars from waste Proceedings of the Combustion Institute 30 2151-2159
Seok S Choi S Park E Han S Lee J (2002) Mn-promoted NiAl2O3 catalysts for stable carbon
dioxide reforming of methane Journal of Catalysis 209 6-15
Shen L Zhang DK (2003) An experimental study of oil recovery from sewage sludge by low-
temperature pyrolysis in a fluidised-bed Fuel 82 465-472
Smith KM Fowler GD Pullket S Graham NJD (2009) Sewage sludge-based adsorbents a
review of their production properties and use in water treatment applications Water
Research 43 2569-2594
Spinosa L Ayol A Baudez JC Canziani R Jenicek P Leonard A Rulkens W Xu G van Dijk
L (2011) Sustainable and innovative solutions for sewage sludge management Water 3
702-717
Spliethoof H (2001) Status of biomass gasification for power production IFRF Combustion
Journal article number 200109 Delft University of Tecnology The Netherlands
Srinakruang J Sato K Vitidsant T Fujimoto K (2006) Highly efficient sulfur and coking
resistance catalysts for tar gasification with steam Fuel 85 2419-2426
Staringhl K Neergaard M (1998) IGCC power plant for biomass utilisation Vaumlrnamo Sweden
Biomass and Bioenergy 15 205-211
Struis RPWJ Schildhauer TJ Czekaj I Janousch M Biollaz SMA Ludwig C (2009) Sulphur
poisoning of Ni catalysts in the SNG production from biomass a TPOXPSXAS study
Applied Catalysis A General 362 121-128
Sutton D Kelleher B Ross JRH (2001a) Review of literature on catalysts for biomass
gasification Fuel Processing Technology 73 155-173
Sutton D Kelleher B Doyle A Ross JRH (2001b) Investigation of nickel supported catalysts
for the upgrading of brown peat derived gasification products Bioresource Technology 80
111-116
Swierczynski D Libs S Courson C Kiennemann A (2007) Steam reforming of tar from a
biomass gasification process over Niolivine catalyst using toluene as a model compound
Applied Catalysis B Environmental 74 211-222
Tae-Young M Bo-Sung K Joo-Sik K (2009) Production of a producer gas with high heating
values and less tar from dried sewage sludge through air gasification using a two-stage
gasifier and activated carbon Energy amp Fuels 23 3268-3276
114 Referencias bibliograacuteficas
Tamhankar SS Hasatani M Wen CY (1981) Kinetic studies on the reactions involved in the
hot gas desulfurization using a regenerable iron oxide sorbent - I Reduction and
sulfidation of iron oxide Chemical Engineering Science 36 1181-1191
Tseng TK Chang HC Chu H Chen HT (2008) Hydrogen sulfide removal from coal gas by the
metal-ferrite sorbents made from the heavy metal wastewater sludge Journal of
Hazardous Materials 160 482-488
Van der Drift A van Doorn J Vermeulen JW (2001) Ten residual biomass fuels for circulating
fluidized-bed gasification Biomass amp Bioenergy 20 45-56
Wang T Chang J Lv P Zhu J (2005) Novel catalyst for cracking of biomass tar Energy amp Fuels
19 22-27
Wender I (1996) Reactions of synthesis gas Fuel Processing Technology 48 189-297
Westmoreland PR Harrison DP (1976) Evaluation of candidate solids for high-temperature
desulfurization of low-btu gases Environmental Science amp Technology 10 659-661
Xie L Li T Gao J Fei X Wu X Jiang Y (2010) Effect of moisture content in sewage sludge on
air gasification Journal of Fuel Chemistry and Technology 38 615-620
Yan F Luo S Hu Z Xiao B Cheng G (2010) Hydrogen-rich gas production by steam
gasification of char from biomass fast pyrolysis in a fixed bed reactor Influence of
temperature and steam on hydrogen yield and syngas composition Bioresource
Technology 101 5633-5637
Yildirim O Kiss AA Huumlser N Leszligmann K Kenig EY (2012) Reactive absorption in chemical
process industry a review on current activities Chemical Engineering Journal 213 371-
391
Yoshimura Y Yasuda H Sato T Shimada H (1995) Utilization of thermodynamic database in
the systems using molybdate and iron based catalysis Coal Science and Technology 24
1275-1278
Yuan W Bandosz TJ (2007) Removal of hydrogen sulfide from biogas on sludge-derived
adsorbents Fuel 86 2736-2746
Yung MM Jablonski WS Magrini-Bair KA (2009) Review of catalytic conditioning of
biomass-derived syngas Energy amp Fuels 23 1874-1887
Zhang R Brown RC Suby A Cummer K (2004) Catalytic destruction of tar in biomass derived
producer gas Energy Conversion and Management 45 995-1014
Zhang R Wang Y Brown RC (2007) Steam reforming of tar compounds over Niolivine
catalysts doped with CeO2 Energy Conversion and Management 48 68-77
Zhang B Xiong S Xiao B Yu D Jia X (2011) Mechanism of a wet sewage sludge pyrolysis in a
tubular furnace International Journal of Hydrogen Energy 36 355-363
APEacuteNDICE COPIA DE LOS ARTIacuteCULOS PUBLICADOS
Artiacuteculo i helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip i1 - i10
Air steam gasification of sewage sludge in a fluidized bed Influence
of some operating conditions
Artiacuteculo ii helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip ii1 - ii9
Air steam gasification of char derived from sewage sludge pyrolysis
Comparison with the gasification of sewage sludge
Artiacuteculo iii helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip iii1 - iii11
Energetic assessment of air-steam gasification of sewage sludge and
of the integration of sewage sludge pyrolysis and air-steam
gasification of char
Artiacuteculo iv helliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphelliphellip iv1 - iv10
Use of sewage sludge combustion ash and gasification ash for high-
temperature desulphurization of different gas streams
Apeacutendice 115
7 APEacuteNDICE COPIA DE LOS TRABAJOS PUBLICADOS
Las publicaciones que recogen los resultados de los estudios realizados en la presente
Tesis Doctoral y que se adjuntan en sucesivas paacuteginas son los siguientes
i N Gil-Lalaguna JL Saacutenchez MB Murillo E Rodriacuteguez G Gea (2014) Air steam
gasification of sewage sludge in a fluidized bed Influence of some operating conditions
Chemical Engineering Journal 248 373-382
Factor de impacto JCR de la revista Chemical Engineering Journal en 2013 4058 ISI
ranking 10 de 133 en el aacuterea de Ingenieriacutea Quiacutemica
ii N Gil-Lalaguna JL Saacutenchez MB Murillo V Ruiz G Gea (2014) Air steam gasification
of char derived from sewage sludge pyrolysis Comparison with the gasification of sewage
sludge Fuel 129 147-155
Factor de impacto JCR de la revista Fuel en 2013 3406 ISI ranking 11 de 133 en el aacuterea
de Ingenieriacutea Quiacutemica
iii N Gil-Lalaguna JL Saacutenchez MB Murillo M Atienza-Martiacutenez G Gea (2014)
Energetic assessment of air-steam gasification of sewage sludge and of the integration
of sewage sludge pyrolysis and air-steam gasification of char Energy 76 652-662
Factor de impacto JCR de la revista Energy en 2013 4159 ISI ranking 14 de 81 en el aacuterea
de Energiacutea amp Combustibles
iv N Gil-Lalaguna JL Saacutenchez MB Murillo G Gea (2015) Use of sewage sludge
combustion ash and gasification ash for high-temperature desulphurization of different
gas streams Fuel 141 99-108
Factor de impacto JCR de la revista Fuel en 2013 3406 ISI ranking 11 de 133 en el aacuterea
de Ingenieriacutea Quiacutemica
La doctoranda ha sido la responsable de la experimentacioacuten anaacutelisis de datos y redaccioacuten
de todos los artiacuteculos
116 Apeacutendice
Ademaacutes durante el desarrollo de la Tesis la doctoranda ha tenido la oportunidad de
presentar su trabajo en varios congresos internacionales participando en algunos de ellos con
ponencias orales
N Gil-Lalaguna JL Saacutenchez MB Murillo G Gea ldquoGas desulfurization with solid by-products
from thermo-chemical conversion of sewage sludgerdquo 22nd European Biomass Conference amp
Exhibition Hamburgo (Alemania) Junio 2014 ISBN 978-88-89407-52-3
N Gil-Lalaguna V Ruiz JL Saacutenchez MB Murillo G Gea ldquoValorization of char from sewage
sludge pyrolysis by means of gasificationrdquo 21st European Biomass Conference amp Exhibition
Copenhague (Dinamarca) Junio 2013 ISBN 978-88-89407-53-0
N Gil-Lalaguna M Atienza I Fonts JL Saacutenchez MB Murillo ldquoThermochemical valorization
of sewage sludge by combination of fast pyrolysis and gasificationrdquo ANQUE International
Congress of Chemical Engineering Innovating for the future Sevilla (Espantildea) Junio 2012
N Gil-Lalaguna A Moreno JL Saacutenchez MB Murillo G Gea ldquoCatalytic steam reforming of
model tar compounds over CaO-Al2O3 catalystsrdquo 20th European Biomass Conference amp
Exhibition Milaacuten (Italia) Junio 2012 ISBN 978-88-89407-54-7
N Gil-Lalaguna JL Saacutenchez MB Murillo M Azuara G Gea ldquoSewage sludge gasification
Effect of the type of gasifying agentrdquo 19th European Biomass Conference amp Exhibition Berliacuten
(Alemania) Junio 2011 ISBN 978-88-89407-55-7
N Gil-Lalaguna JL Saacutenchez MB Murillo ldquoPreliminary study of gasification with
steamenriched air mixtures of sewage sludge and of the char obtained in its pyrolysis
processrdquo BioEnergy III Present and New Perspectives on Biorefineries Lanzarote (Espantildea)
Mayo 2011
Chemical Engineering Journal 248 (2014) 373ndash382
Reprinted with permission from Elsevier
Contents lists available at ScienceDirect
Chemical Engineering Journal
journal homepage wwwelsevier comlocate cej
Airndashsteam gasification of sewage sludge in a fluidized bedInfluence of some operating conditions
httpdxdoiorg101016jcej2014030551385-8947 2014 Elsevier BV All rights reserved
uArr Corresponding author Tel +34 976762224E-mail address noemigilunizares (N Gil-Lalaguna)
N Gil-Lalaguna uArr JL Saacutenchez MB Murillo E Rodriacuteguez G GeaThermo-chemical Processes Group Aragoacuten Institute of Engineering Research (I3A) Universidad de Zaragoza cMariano Esquillor sn 50018 Zaragoza Spain
h i g h l i g h t s
Experimental work on airndashsteam gasification of sewage sludge in a fluidized bed Temperature is the most influential factor for most of the variables analyzed Steam presence favors the gas heating value and the H2CO in the product gas Differences between kinetic and thermodynamic control in a gasification process
a r t i c l e i n f o
Article historyReceived 24 September 2013Received in revised form 13 March 2014Accepted 16 March 2014Available online 24 March 2014
KeywordsAirndashsteam gasificationSewage sludgeFluidized bed
a b s t r a c t
An experimental work was carried out to investigate the viability of energy recovery from the airndashsteamgasification of sewage sludge The relative influence of different factors as well as the effect of their pos-sible interactions has been determined by means of analysis of variance Temperature was found to bethe most influential factor for most of the variables analyzed Solid yield (35ndash41 wt) and tar content(11ndash45 gm3
STP) were largely reduced with temperature whereas gas production (089 132 m3STPkg
sewage sludge dry and ash free) carbon yield to gas phase (62ndash90 wt) gasification efficiency(39ndash66) and H2 and CO yields (20ndash52 and 137ndash414 gkg sewage sludge dry and ash free respectively)were improved at high temperature Other important parameters for the end-use of the gas such as itsheating value (412ndash620 MJm3
STP) and its H2CO molar ratio (146ndash325) were greatly influenced bythe composition of the gasification medium since the increase in the steam to oxygen ratio was favorablefor both The comparison of experimental and theoretical results highlights that equilibrium was notreached during the experimental runs
2014 Elsevier BV All rights reserved
1 Introduction
Biomass is one of the most important primary renewableenergy sources The conversion of biomass to energy encompassesa wide range of materials conversion technologies and end-useapplications of the products such as powerheat generation trans-portation fuels and chemical feedstocks Sewage sludge which isthe waste produced by wastewater treatment processes can beconsidered an important renewable biomass energy source [1]
As a result of the application of the Urban Wastewater Treat-ment Directive (UWWTD) 91271EEC [2] new municipal waste-water treatment strategies have been developed during the lasttwo decades in order to improve the quality of effluents Existingtreatment plants have been upgraded and new and more effectivetreatment plants have been designed and implemented In parallel
to the improvement of the effluent quality environmental aware-ness about sewage sludge management has gained strength Themain commercial means of sewage sludge disposal include itsuse as fertilizer land filling or incineration [34] However becauseof increasing legal limitations on sewage sludge land filling andagricultural reuse energy recovery from sewage sludge remainsan attractive and sustainable way of management Thermal pro-cesses such as pyrolysis gasification or combustion of sewagesludge have thus attracted considerable scientific interest This pa-per presents an experimental work on sewage sludge gasification
Gasification is the conversion of a carbonaceous material into agas fuel by heating it in a gasification medium such as air oxygenor steam Gas from gasification consists of a mixture of carbonmonoxide carbon dioxide hydrogen methane and other lighthydrocarbons nitrogen (if air is used as gasifying agent) and steamThis gas can be used to power gas engines and gas turbines or usedas a chemical feedstock to produce liquid fuels [5] During gasifica-tion a mixture of heavy and condensable hydrocarbons (tars) is
i1
Table 1Proximate and ultimate analyses and heating value of sewage sludge
Proximate analysis (wt wet basis)Moisture 648Ash 3904Volatiles 5009Fixed carbon 439
Ultimate analysis (wt wet basis)C 295H 467N 527S 131HHV (MJkg) 128LHV (MJkg) 118
374 N Gil-Lalaguna et al Chemical Engineering Journal 248 (2014) 373ndash382
also produced The presence of tar in the gas causes problems asso-ciated with condensation formation of aerosols and polymeriza-tion leading to more complex structures which limit thesubsequent utilization of the gas
Operating conditions during gasification (such as the nature ofthe biomass pressure temperature residence time or gasificationmedium) play an important role in both tar formation and gasquality The higher the temperature the lower the tar content inthe product gas [6] but other factors such as the risk of ash sinter-ing limit the operating temperature The use of different gasifyingagents such as air steam steamndashoxygen mixtures or carbon diox-ide has been reported in the literature Both gas composition andgas heating value are noticeably affected by the gasification med-ium because of the variation of selectivity in the gasification reac-tions [7] Generally steam gasification enhances H2 productioncompared to air gasification and also leads to a higher gas heatingvalue because the dilution of the gas with nitrogen is avoided [8]However the steam gasification reactions are endothermic and re-quire a continuous supply of energy Given this background bio-mass gasification with mixtures of air and steam appears to be apotential solution from the economic point of view since the par-tial combustion of biomass inside the gasifier can supply the re-quired energy for the process turning it into an autothermalprocess The improvement in gas quality by feeding a flow of steamtogether with the air stream during biomass gasification has beenreported in several experimental studies [9ndash12]
In the particular case of sewage sludge experimental studiesbased on air gasification [13ndash17] and steam gasification [18] havebeen reported in the literature In general the gas composition andthe gas heating value from sewage sludge gasification are close totypical values obtained from other kinds of biomass which dem-onstrates the potential of sewage sludge as a raw material forthe gasification process However tar formation and other addi-tional problems such as the formation of other pollutants (H2SHCl or NH3) hinder the development of sewage sludge gasificationso new efforts are required in order to optimize the process
In this work an experimental study (based on a 2k factorial de-sign) on sewage sludge gasification in a fluidized bed with mix-tures of air and steam has been developed in order to find outthe influence of several operating conditions (temperature compo-sition of the gasification medium and gasifying agent to biomassratio) on the gasification performance Furthermore experimentalresults have been compared with theoretical data which weredetermined considering equilibrium conditions
2 Materials and methods
21 Sewage sludge
Anaerobically digested and thermally dried sewage sludge (SS)was supplied by a Spanish urban wastewater treatment plantFeedstock analyses were performed at the Instituto de Carboquiacutemi-ca (ICB-CSIC) in Zaragoza (Spain) according to standard methodsmoisture according to ISO-589-1981 ash according to ISO-1171-1976 volatiles according to ISO-5623-1974 ultimate analysis(CHNS) using a Carlo Erba 1108 and heating value according toISO-1928-89 (Table 1) More details about the sewage sludge char-acterization such as FTIR and X-ray diffraction analyses can befound elsewhere [19] Sewage sludge was smashed and sieved toobtain a feed sample in the size range of 250ndash500 lm
22 Experimental setup
Sewage sludge gasification runs have been carried out in alaboratory-scale fluidized bed reactor operating at atmospheric
i2
pressure with continuous feed of solid and continuous removalof ash The gasifier was a tubular reactor made of refractory steel(AISI 310) divided into two parts a bed zone with an inner diam-eter of 40 mm and a freeboard zone with an inner diameter of63 mm Sewage sludge was continuously fed to the reactor by afeeding system composed of a screw-feeder and a variable speedmotor The solid feed rate in each experiment was around 21 gmin Ash from previous sewage sludge gasification tests consti-tuted the solid bed by itself from the beginning of the runs Whenthe amount of bed material inside the reactor exceeded the heightof the bed zone it left the reactor by overflow through a lateralpipe and was collected in a separate vessel The reactor was heatedby an electrical furnace with three different heating zones (bedfree-board and cyclone) which could be controlled independentlyThe bed temperature was one of the factors under study rangingbetween 770 and 850 C (the same as in the free-board) whilethe cyclone temperature was set at 450 C A schematic diagramof the experimental setup can be found elsewhere [20]
The gasifyingfluidizing agent used in the process consisted ofdifferent mixtures of steam and enriched air (air + oxygen) Fur-thermore an additional flow of nitrogen was necessary in two ofthe experiments (those with the lowest air requirement) in orderto avoid differences in the dilution effect of the gas with nitrogenand in the fluidization rate (which was around 5ndash7 times greaterthan the minimum fluidization rate) The feed rate of these gases(air oxygen and nitrogen) was adjusted by using mass flow con-trollers The water was fed through a HPLC pump and vaporizedbefore mixing into the gas stream The composition and theamount of gasifying agent were the other factors under study inthis work The mixture of oxygen steam and approximately 23of the total air required was fed into the fluidized bed reactorthrough its distribution plate while the remaining air was fed withthe solid to facilitate its movement through the feeding pipewhich was externally refrigerated to prevent reactions taking placeoutside the bed
The vapors and gases produced during gasification remained in-side the reactor between 7 and 8 s and then passed through a cy-clone and a hot filter both at 450 C in which the solid particlesswept by the gas were collected Next the gases and vapors passedthrough two ice-cooled condensers where water and condensableorganic compounds (tar) were collected A cotton filter was situ-ated after the condensers in order to remove small particulatesand aerosols swept by the gas The volume of particle- and tar-freegas was measured by a volumetric meter and its composition wasanalyzed on line using a micro gas chromatograph (Agilent 3000-A) which determined the volume percentages of H2 O2 CO CO2CH4 C2H4 C2H6 C2H2 and H2S Water content in the condensedfraction was analyzed off line by Karl Fischer titration (so theamount of tar was determined by difference) and the tar composi-tion was analyzed by gas chromatography with mass spectroscopyand flame ionisation detectors (MSFID GC) The experiments were
N Gil-Lalaguna et al Chemical Engineering Journal 248 (2014) 373ndash382 375
carried out during 90 min to ensure that the stationary state wasreached [21]
23 Experimental design and data analysis
The influence of three operating factors (temperature gasifyingagent to biomass ratio and composition of the gasification med-ium) on sewage sludge gasification performance has been studiedexperimentally by means of a 2k factorial design where k indicatesthe number of factors studied (in this case 3) and 2k represents thenumber of runs (in this case 8) Furthermore three replicates at thecenter point (CP) were carried out in order to evaluate both theexperimental error and the curvature shown by the evolution ofeach response variable that is to say whether this evolution is lin-ear or not within the experimental range studied This experimen-tal design is suitable not only for studying the influence ofoperating conditions but also the influence of their possible inter-actions An interaction occurs when a factor influences a responsevariable in a different way depending on the value of anotherfactor
The three analyzed factors were (i) bed reactor temperature(which ranges between 770 and 850 C) (ii) gasifying ratio (GR)between the mass flow of gasifying agent (oxygen plus steam)and the mass flow of dry and ash-free basis (daf) sewage sludge(which ranges between 08 and 11 gg SS daf) (iii) nature of thegasification medium represented by the H2OO2 molar ratio(which ranges between 1 and 3) The overall flow rate of gasifyingagent was kept constant when the H2OO2 molar ratio was modi-fied These three factors together with their respective ranges ofstudy were chosen on the basis of works of other authors concern-ing gasification of different kinds of biomass in fluidized bed reac-tors [9ndash12]
As can be seen in Table 2 the experimental design consists of 8runs plus 3 replicates at the center point (810 C 095 gg SS daf2 mol H2Omol O2) As usually occurs when an experimental designis planned the lower and upper limits of the factors are coded as1 (in this case T = 770 C GR = 08 and H2OO2 = 1) and 1 (in thiscase T = 850 C GR = 11 and H2OO2 = 3) respectively The use ofcoded levels enables an easy identification of the term with thegreatest influence on the response variable the higher the coeffi-cient the more influential the factor
The response variables analyzed were (i) distribution of prod-ucts (yields to the different gasification products solid gas andtar) (ii) gas composition determined on line using a micro gaschromatograph (iii) production of each gaseous component (iv)lower heating value of the product gas (LHVgas) (v) cold gasifica-tion efficiency (vi) carbon yield to gas phase and (vii) tarcomposition
Statistical analyses of the results have been carried out by anal-ysis of variance (ANOVA) using the Design-Expert 7 software(from Stat-Ease Inc) ANOVA analysis evaluates whether the effectof the factors the interactions between them and the curvaturehave a significant influence or not on the response variables A con-fidence level of 95 for the F-distribution was selected to deter-mine the significant effects
Table 2Operating conditions of gasification tests
Experiment number 1 2 3 4
Coded values 111 111 111 Temperature (C) 850 770 850 7g gasifying agentg sewage sludge daf 11 11 08 0H2OO2 molar ratio in the gasifying agent 3 3 3 3Equivalence ratio (ER) 017 017 012 0Steam to biomass daf mass ratio (SB) 071 071 052 0
3 Results and discussion
31 Distribution of products
Experimental results for the distribution of products are pre-sented in Table 3
Furthermore as a result of the ANOVA analysis Table 4 showsthe relative influence of each factor on the product distributionIn this table the average data represent the average of the wholeset of results obtained for each response variable the coefficientsassociated to the different factors (T GR and H2OO2) show howthe response variables evolve when varying each factor (consider-ing the coded values for the factors within the studied range) andthe coefficients associated to the interactions show whether a fac-tor influences a response variable in a different way depending onthe value of another factor
311 Solid yield and carbon content in the solidThe solid yield is defined as the mass (g) of solid product col-
lected per 100 g of sewage sludge fed Because of the high ash con-tent in the sewage sludge (39 wt) the solid residue is animportant by-product in its gasification process and its yield variedbetween 35 and 41 wt whilst typical values for other kinds ofbiomass such as wood or straw are below 8 wt [22]
Carbon content in the solid product was analyzed using a LecoTruSpec Micro Elemental Analyzer (Table 3) According to the AN-OVA results (Table 4) carbon content in the solid product is re-duced by increasing both the gasification temperature (higherreaction rate) and the gasifying ratio and by decreasing the H2OO2 ratio (Fig 1) which seems to indicate that carbon combustionis faster than its steam gasification Although temperature is themost influential factor for the carbon reaction its effect dependson other operating conditions since its interaction with the H2OO2 ratio is a significant term (Fig 1a) This fact shows that carbonreactions with oxygen are more sensitive to temperature changesthat the reactions with steam
The results of the solid yield together with those of the carboncontent in the solid (Table 3) suggest that inorganic ash com-pounds could have been released to the gas phase during the gas-ification process since some data of solid yield are even below theoriginal ash content of the sewage sludge (39 wt) Both the trans-formation and the release to gas phase of ash compounds duringthermo-chemical processes have been shown in other studies[2324] although this was usually found to take place at highertemperatures
312 Gas yieldThe gas yield is defined as the volume of gas produced (m3
STP
N2-free basis where STP means standard conditions of tempera-ture and pressure at 0 C and 1 atm) per kilogram of SS daf fedThe gas yield data from sewage sludge gasification varied between089 and 132 m3
STPkg SS daf so these values are close to the typicalones found in the literature for similar operating conditions anddifferent kinds of biomass [7912]
5 6 7 8 9 10 11
111 111 111 111 111 00070 850 770 850 770 8108 11 11 08 08 095
1 1 1 1 212 032 032 023 023 01952 039 039 027 027 052
i3
Table 3Experimental results product distribution and gas composition
111 111 111 111 111 111 111 111 000a
Product distributionSolid yield (g solid100 g SS) 368 401 401 407 356 392 384 400 382 plusmn 01Carbon content in the solid product (wt) 456 761 566 1020 051 620 100 709 589 plusmn 033Gas yield (m3
STPkg SS) 072 051 065 053 072 052 071 049 061 plusmn 001
Gas yield (m3STPkg SS daf) 132 094 120 097 132 096 130 089 113 plusmn 001
Tar content (gm3STP) 188 436 186 445 121 224 109 453 148 plusmn 14
Gas composition (vol dry basis)H2 242 184 251 204 180 110 206 136 193 plusmn 01CO 87 57 102 73 116 70 141 77 94 plusmn 01CO2 171 186 126 155 207 238 160 198 181 plusmn 02CH4 31 35 36 41 26 28 29 34 33 plusmn 01C2Hx 17 21 14 22 13 16 14 20 17 plusmn 02H2S 044 038 033 033 044 042 038 031 040 plusmn 002N2 449 514 468 502 453 534 445 532 478 plusmn 02H2CO molar ratio 279 325 246 281 154 157 146 177 206 plusmn 001COCO2 molar ratio 051 030 081 047 056 029 088 039 052 plusmn 001
a Mean value plusmn standard deviation
Table 4Relative influence of the significant factors on the carbon content in the solid product gas yield tar content in the gas and H2CO and COCO2 molar ratios in the product gas
Carbon contentin the solid (wt)
Gas yield(m3
STPkg SS daf)Tar content in the gas(gm3
STP)H2CO molar ratioin the gas
COCO2 molar ratioin the gas
Average 550 112 2703 221 052T 242 017 1191 014 016GR 063 0019 278 0081 011H2OO2 165 435 062
T-GR 0017 315 0021 0044T-(H2OO2) 053 0024 0059 0026GR-(H2OO2) 262 011
T-(H2OO2)-GR 0028 289 0049
Curvature
Non-significant term Curvature is significant
770 810 8500
2
4
6
8
10
H2OO2=3H2OO2=1 H2OO2=2(a)
Temperature (ordmC)
Car
bon
cont
ent i
n th
e so
lid (w
t)
080 095 1100
2
4
6
8
10
(b)
GR (g gasifying agentg daf)
Car
bon
cont
ent i
n th
e so
lid (w
t)
Fig 1 Carbon content in the solid product (wt) (a) Interaction between temperature and H2OO2 molar ratio (GR = 095) (b) Effect of the gasifying ratio (T = 810 C H2OO2 = 2)
376 N Gil-Lalaguna et al Chemical Engineering Journal 248 (2014) 373ndash382
The ANOVA analysis (Table 4) shows that the gas yield does notfollow a linear response within the studied range of the factorssince the curvature is a significant term Temperature is clearlythe most influential factor for the production of gas The significantincrease of the gas yield with temperature may be due to differentprocesses that are favored by higher temperatures greater produc-tion of gas in the initial stage of pyrolysis cracking and steam
i4
reforming of tars and endothermic reactions of char gasification[12] The increase of GR also favors the production of gas althoughits effect is less significant than that corresponding to temperatureSignificant interactions of temperature with both the GR and theH2OO2 ratio have been found the effect of temperature on thegas yield is intensified at the highest value of the GR (Fig 2a)and at the lowest H2OO2 ratio (Fig 2b)
080 095 11009
10
11
12
13
T=770 T=850T=810
770 810 85009
10
11
12
13
H2OO2=3H2OO2=1 H2OO2=2(b)
Temperature (ordmC)
Gas
yie
ld (m
3 STP
kg
SS d
af)
GR (g g daf)
Gas
yie
ld (m
3 STP
kg
SS d
af)
(a)
Fig 2 Gas yield (a) Interaction between temperature and gasifying ratio (H2OO2 = 2) (b) Interaction between temperature and H2OO2 molar ratio (GR = 095)
N Gil-Lalaguna et al Chemical Engineering Journal 248 (2014) 373ndash382 377
313 Tar content in the product gasThe tar content is defined as the mass (g) of condensable organ-
ic compounds collected in each experiment per m3STP of dried gas
measured after condensing the vapors The lowest values of tarcontent obtained in this work are close to the typical values foundfor fluidized bed biomass gasifiers which according to Corella et al[25] usually range between 8 and 15 gm3
STPAs occurred with the gas yield the tar content in the gas does
not follow a linear response within the studied range of the factorsas the curvature is a significant term Temperature is also the mostinfluential factor for tar content (Table 4) The rise in the gasifica-tion temperature from 770 to 850 C causes a clear reduction in tarformation (Fig 3a) because of the enhancement of tar cracking andreforming reactions [6] The tar content is also reduced by decreas-ing the H2OO2 ratio in the gasification medium suggesting that tarcombustion reactions are faster than tar steam reforming and byincreasing the GR although the effect of the latter factor is less sig-nificant The influence of the GR on the tar content disappearswhen working at the highest temperature (Fig 3a) or at the highestH2OO2 ratio (Fig 3b)
32 Gas composition
The gas composition from a gasification process is the result ofmany complex and competing reactions The most representativeof these reactions are given below
770 810 8500
10
20
30
40
50
60(a) GR=08 GR=11GR=095
Temperature (ordmC)
Tar c
onte
nt (g
m
3 STP
)
Fig 3 Tar content in the gas (a) Interaction between temperature and gasifying rat
Oxidation Cthorn O2 $ CO2 DH lt 0 eth1THORN
Partial oxidation Cthorn 1=2O2 $ CO DH lt 0 eth2THORN
Boudouard Cthorn CO2 $ 2CO DH gt 0 eth3THORN
Watermdashgas primary CthornH2O$ COthornH2 DH gt 0 eth4THORN
Watermdashgas secondary Cthorn 2H2O$ CO2 thorn 2H2 DH gt 0 eth5THORN
Watermdashgas shift ethWGSTHORN COthornH2O$ CO2 thornH2 DH lt 0 eth6THORN
Methanation Cthorn 2H2 $ CH4 DH lt 0 eth7THORN
Steam reforming CnHx thorn nH2O$ nCOthorn ethx=2thorn nTHORNH2 DH
gt 0 eth8THORN
Dry reforming CnHx thorn nCO2 $ 2nCOthorn ethx=2THORNH2 DH gt 0 eth9THORN
Cracking CnHx $ Cthorn ethx=2THORNH2 DH gt 0 eth10THORN
As usual in a biomass gasification process the main gases pro-duced during sewage sludge gasification are H2 CO CO2 and lighthydrocarbons CH4 being the most abundant of them In additionH2S is also released during the process due to the presence of sul-fur-compounds in the sewage sludge (Table 1) Statistical analyses
1 2 30
10
20
30
40
50
60(b) GR=08 GR=11GR=095
H2 OO2 (molmol)
Tar c
onte
nt (g
m
3 STP
)
io (H2OO2 = 2) (b) Interaction between H2OO2 and gasifying ratio (T = 810 C)
i5
378 N Gil-Lalaguna et al Chemical Engineering Journal 248 (2014) 373ndash382
of gas composition have not been included in this work because itwas considered preferable to analyze the production or the specificyield of each gaseous compound (gkg SS daf) as detailed in thenext section
The average gas composition (dry basis) obtained in each exper-iment is reported in Table 3 Considerable differences in the frac-tions of the gaseous compounds have been found For exampleH2 (110ndash251 vol) CO (57ndash141 vol) CO2 (126ndash238 vol) orCH4 (26ndash41 vol) can double or halve their percentages depend-ing on the operating conditions These volume percentages lead toH2CO and COCO2 molar ratios in the exit gas ranging from 146ndash325 and 029ndash088 respectively The H2CO molar ratio is animportant parameter in view of possible end uses of the gas andvalues close to 2 are usually required in processes such as metha-nol or Fischer Tropsch synthesis [26] According to the ANOVA re-sults (Table 4) the composition of the gasification medium isclearly the most influential factor for this ratio The higher theH2OO2 ratio used as gasifying agent the higher the H2CO molarratio obtained in the gas product Working at lower temperaturesalso leads to an increase in the H2CO molar ratio
The COCO2 ratio shows how the carbon initially contained inthe sewage sludge is distributed among both compounds Thehigher the gasification temperature the higher the COCO2 ratioobtained in the product gas Furthermore the GR exerts a negativeinfluence on the COCO2 ratio although its effect is less significantthan that of the temperature
33 Production of each gaseous compound
The production or yield of each analyzed gas (H2 CO CO2 CH4C2Hx and H2S) is defined as the mass (g) of each gas produced perkilogram of SS daf fed
Both experimental and theoretical yields of gases are analyzedin this section The theoretical production of each gas during sew-age sludge gasification at equilibrium conditions has been deter-mined using HSC Chemistry 61 software simulating the sameoperating conditions that had been previously tested in the labora-tory that is following the same 2k factorial design According tothe theoretical results obtained the gas product from sewagesludge gasification at equilibrium conditions should only containH2 CO CO2 CH4 H2S and NH3
The experimental and the theoretical yields of gases are com-pared in Fig 4 The points in the same vertical line represent the
Fig 4 Theoretical (d) and experimental (
i6
results obtained under the same operating conditions As can beseen experimental and theoretical data appreciably differ onefrom the other which means that equilibrium was not reachedduring the experimental runs maybe due to insufficient residencetime of the gases in the reactor Experimental yields of H2 and COare clearly below their corresponding theoretical data (up to fourand five times lower in the most unfavorable conditions respec-tively) The lower the gasification temperature the greater is thedifference between the experimental and the theoretical data Incontrast experimental yields of CO2 and CH4 are above their corre-sponding theoretical values CH4 is mainly produced during thepyrolysis step and is hardly reformed during the subsequentprocess
The experimental and theoretical yields of gases have been ana-lyzed statistically by means of ANOVA In the case of the theoreti-cal results most of the yields revealed a curvature so the designwas augmented with central composite points in order to deter-mine the evolution of the response variables in the studied rangesand to find out which factor(s) is (are) causing the curvatureTable 5 presents the ANOVA results for both the experimentaland the theoretical results
As can be seen in Table 5 temperature is the most influentialfactor for the experimental yield of H2 Although this gas is in-volved in many reactions both as reactant and as product the tem-perature rise leads to a global increase in its experimental yieldThe same trend for H2 production has usually been reported inthe literature [111227] Although to a lesser extent the H2 exper-imental yield is also enhanced by increasing the H2OO2 ratio Onthe one hand the increase in the steam presence favors H2 forma-tion (4 5 6 8) and on the other hand H2 combustion is mitigatedby reducing the proportion of oxygen in the gasification mediumThe GR affects the experimental production of H2 in a negativeway H2 consumption outweighs H2 formation when both the ERand the SB ratio are increased In contrast to the experimental re-sults the H2OO2 ratio is the most influential factor for the theoret-ical yield of H2 (Table 5) and it is also the factor responsible for thecurvature observed The influence of the temperature is much lesssignificant in this case and unlike the experimental results thisfactor adversely affects the theoretical production of H2 TheWGS reaction (6) may explain this observed trend at equilibriumconditions due to its exothermic nature
As occurred with the H2 experimental yield temperature is themost influential factor for the experimental production of CO
) production of H2 CO CO2 and CH4
Table 5Relative influence of the significant factors on the yield of each gaseous compound lower heating value of the gas cold gasification efficiency and carbon yield to gas phase(experimental and theoretical results)
Yield of gaseous compounds (gkg SS daf) LHVgas (MJm3STP) Cold gasification efficiency () Carbon yield to gas phase ()
H2 CO CO2 CH4 C2Hx H2S
Coefficients obtained for the experimental resultsAverage 3703 25010 76342 5023 4613 1266 549 5512 7248T 906 7879 149 328 173 037 851 633GR 137 1751 10157 313 156 031 333H2OO2 568 3267 9253 480 335 040 347 607T-GR 503 2159
T-(H2OO2) 2195 017
GR-(H2OO2) 1493
T-(H2OO2)-GR 965 1591 062
Curvature
Coefficients obtained for the theoretical resultsAverage 9953 88323 59521 060 ndash 2554 656 9847 100T 119 2454 3644 076 ndash
GR 309 8609 13658 046 ndash 035 564
H2OO2 1485 3677 5869 032 ndash 034 994
T-GR 496 885 038 ndash
T-(H2OO2) 434 757 027 ndash
GR-(H2OO2) 234 1566 2430 ndash 012 200
T-(H2OO2)-GR ndash
(H2OO2)2 444 1112 1684 ndash 239
T2 042 ndash
GR2 ndash
Non-significant term
N Gil-Lalaguna et al Chemical Engineering Journal 248 (2014) 373ndash382 379
(Table 5) Higher temperatures favor the production of CO throughreactions such as steam and dry reforming (8 9) the Boudouardreaction (3) or the waterndashgas primary reaction (4) However neg-ligible variations or even the opposite trend in CO production arefound in the literature [1011] which reveals the importance ofthe nature of the biomass and the operating conditions in the evo-lution of CO production Both the GR and the H2OO2 ratio affectthe experimental yield of CO in a negative way When the GR is in-creased the higher amount of oxygen fed to the gasifier promotesthe oxidation of CO to CO2 and in addition the higher presence ofsteam favors CO consumption through the WGS reaction More-over the negative effect of the H2OO2 ratio might indicate thatthe consumption of CO in the WGS reaction outweighs its combus-tion process Both negative effects are significantly intensified athigher temperatures In contrast to the experimental results theGR is the most influential factor for the theoretical production ofCO (Table 5) and the H2OO2 ratio shows a positive effect The threeinteractions between the factors are significant terms in the theo-retical production of CO (i) the negative effect of the GR is slightlyreduced when working at high temperatures maybe due to theendothermic nature of the Boudouard reaction (in which CO is pro-duced) (ii) the positive effect of the temperature is slightly re-duced when working at high H2OO2 ratios since increasing thesteam presence shifts the WGS equilibrium towards CO consump-tion (iii) the negative effect of the GR is intensified when the high-est H2OO2 ratio is used as gasification medium
Regarding the production of CO2 the GR is the most influentialfactor for both the experimental and the theoretical yields(Table 5) When the GR is increased more oxygen and steam arefed to the gasifier thus the increased production of CO2 can beattributed to a higher extent of combustion reactions as well asto other reactions promoted by the presence of steam such asthe WGS reaction (6) or the secondary waterndashgas reaction (5) inwhich CO2 is produced The positive effect of the GR on the exper-imental yield of CO2 is intensified at higher temperatures and low-er H2OO2 ratios Although to a lesser extent the increase in theH2OO2 ratio negatively affects the production of CO2 This trendsuggests that combustion reactions are the main source of CO2
In contrast to the theoretical results the experimental yield ofCO2 is not significantly influenced by the temperature Theoreticalresults show that CO2 and CO yields are influenced by the samesignificant factors and interactions but all of them show oppositeeffects since CO production is normally linked with CO2 consump-tion and vice versa (3 6 9)
Regarding the experimental production of light hydrocarbons(CH4 and C2Hx) the H2OO2 ratio is the most influential factor forit (Table 5) Increasing the H2OO2 ratio in the gasification mediumenhances the production of both CH4 and C2Hx thus suggestingthat the steam reforming of light hydrocarbons occurs more slowlythan its combustion process The formation of CH4 via the metha-nation reaction (7) may also be promoted by increasing the H2OO2
ratio due to an increased presence of H2 in the gasification med-ium Although to a lesser extent CH4 production is negatively af-fected by the increase in the GR as its combustion and steamreforming reactions are promoted by increasing the ER and the SB ratio respectively This expected effect is not observed for theC2Hx experimental yield probably because of its large experimentalvariability Unlike the results shown by other authors [1227] theexperimental yield of CH4 is found to increase slightly with thetemperature maybe as a result of the thermal cracking of heavierhydrocarbons while the experimental yield of C2Hx follows theopposite trend with temperature In relation to the theoretical re-sults the presence of C2Hx in the equilibrium gas is practically neg-ligible CH4 is produced at equilibrium conditions but itstheoretical yield is much lower than its experimental yield Tem-perature is the most influential factor for the theoretical yield ofCH4 (Table 5) It has a negative effect due to the enhancement ofthe endothermic reactions in which CH4 is consumed such assteam and dry reforming (8 9) and the restriction of the methana-tion reaction (7) due to its exothermic nature The negative effectof the temperature on the theoretical yield of CH4 is intensifiedby increasing the H2OO2 ratio andor decreasing the GR The tem-perature also seems to be the factor responsible for the curvatureshown by the theoretical yield of CH4
Lastly according to the ANOVA results the experimental pro-duction of H2S is favored by increasing both the temperature and
i7
380 N Gil-Lalaguna et al Chemical Engineering Journal 248 (2014) 373ndash382
the GR although the effect of the latter is slightly smaller than thatof the temperature (Table 5) In contrast to the experimental re-sults non-significant influences of the studied factors on the theo-retical production of H2S have been found within the studiedintervals H2S is the only sulfured-compound considered in theequilibrium gas thus a constant yield of H2S has been obtainedfor all the simulated conditions (2554 gkg SS daf)
34 Lower heating value of the product gas
The lower heating value of the gas (LHVgas) is calculated asP
(xi LHVi) where xi and LHVi are the volumetric fraction andthe lower heating value (MJm3
STP) of each gaseous componentrespectively The LHV of the product gas obtained from the sewagesludge gasification ranged between 412 and 620 MJm3
STP thusthis gas can be considered as a low heating value gas Similar val-ues of LHVgas are usually reported in the literature for air gasifica-tion or airndashsteam gasification of other kinds of biomass [5]
As a result of the ANOVA analysis Table 5 presents the codedcoefficients that explain the influence of the factors on the theoret-ical and experimental gas heating values As can be seen the the-oretical gas heating values are higher than those obtainedexperimentally under the same operating conditions The lowerproduction of CO2 obtained at equilibrium conditions comparedto its experimental production leads to a lower dilution effect ofthe gas from the energy point of view which outweighs the lowerproduction of light hydrocarbons (gas components with the high-est heating value) at equilibrium conditions
The composition of the gasification medium is the most influen-tial factor for the experimental LHVgas When the H2OO2 ratio isincreased the hydrocarbon content increases and the CO2 contentdecreases so both effects contribute to improve the LHVgas Theinfluence of temperature on the experimental LHVgas is almost asimportant as that of the composition of the gasification mediumAlthough the experimental production of CO2 (in terms of gkgSS daf) is not affected by the temperature this result is not thesame when considering the concentration data since a clear reduc-tion in the CO2 fraction with temperature is observed (Table 3) Theeffect of this reduced fraction of CO2 on the gas calorific value ismore significant than that of the reduced fraction of light hydrocar-bons so a global positive effect of temperature on the LHVgas hasbeen found in this study In contrast to this results in the literatureusually show a negative effect of the temperature on the LHVgas
[12] thus showing that the evolution of the gas composition de-pends on the raw material and the operating conditions Althoughits effect is slightly smaller the GR negatively affects the experi-mental LHVgas since both the production of CO2 and the consump-tion of light hydrocarbons are favored at higher GR
The theoretical results show that the GR and the H2OO2 ratiohave almost the same relative influence on the LHVgas whereasthe gasification temperature does not affect it significantly(Table 5)
35 Cold gasification efficiency
The cold gasification efficiency is defined as the ratio betweenthe energy contained in the gas product (m3
STP gas LHVgas) andthe energy contained in the mass of sewage sludge fed(kgSS LHVSS) Cold gasification efficiency assumes a temperatureof 25 C of the product gases so the sensible heat of the gas isnot taken into account
The experimental values of cold gasification efficiency variedbetween 39 and 66 and according to the ANOVA results(Table 5) the temperature and the H2OO2 ratio are the only factorsthat affect it significantly Temperature is the most influentialfactor and its variation from 770 to 850 C improves the cold
i8
gasification efficiency by 17 This enhancement is based on the in-crease of both LHVgas and gas production with temperatureAlthough to a lesser extent the H2OO2 ratio also affects the exper-imental gasification efficiency in a positive way since the LHVgas
increases with the H2OO2 ratio and the production of gas is not af-fected by it
The theoretical cold gasification efficiencies are much higherthan the experimental data Unlike the experimental results theH2OO2 ratio is the most influential factor for the theoretical coldgasification efficiency as well as being the factor responsible forthe curvature exhibited by the results As occurred with the theo-retical LHVgas increasing the GR negatively affects the theoreticalcold gasification efficiency whereas the gasification temperaturedoes not affect it significantly
36 Carbon yield to gas phase
The carbon yield to gas phase is defined as the ratio between themass of carbon contained in the product gas and the mass of car-bon contained in the sewage sludge fed The conversion of solidcarbon during the sewage sludge gasification reached 76ndash98 wt However not all the solid carbon leads to the formationof gaseous compounds as tar is also produced Therefore theexperimental results of carbon yield to gas phase are slightly lowerthan the aforementioned range (62ndash90 wt) whereas a carbonyield to gas phase of 100 is expected at equilibrium conditions
According to the ANOVA results (Table 5) carbon yield to gasphase shows a linear response with the factors within the studiedintervals Temperature is the most influential factor and its varia-tion from 770 to 850 C improves the carbon yield to gas phase by13 wt The rise in temperature not only favors the heterogeneousreactions between the carbon contained in the sewage sludge andthe gas compounds (3ndash5) but also enhances the tar cracking andreforming reactions so a greater amount of carbon leaves the gas-ifier as part of the product gas The effect of the gasification med-ium is slightly lower than that of the temperature Carbon yield togas phase is increased at higher fractions of oxygen and lower frac-tions of steam which suggests that carbon oxidation reactions (12) take place faster than the heterogeneous waterndashgas reactions(4 5) To a lesser extent carbon yield to gas phase is also favoredby the GR since a greater amount of gasifying agent is availableto react with the carbon contained in the sewage sludge
37 Tar composition
The tar composition was analyzed by gas chromatography (MSFID GC) Fig 5 shows a representative chromatogram of the com-ponents detected in most of the tar samples Some researchershave divided tar components into several groups based on theirmolecular weight [28] A similar classification of tar compoundshas been considered in this work in order to analyze the effect ofthe operating conditions on the fractions of the following familiesof compounds heterocyclic aromatics containing N (includingn-methyl-pyridine benzonitrile n-methyl-benzonitrile quinolinen-methyl-quinoline indole n-phenyl-pyridine n-naphthalenecar-bonitrile benzoquinoline and 5H-indeno[12-b]pyridine) hetero-cyclic aromatics containing O (phenol and benzofuran)compounds containing S (2-benzothiophene and propanenitrile330-thiobis-) light aromatics with 1 ring (styrene) and light PAHcompounds with 2 or 3 rings (indene naphthalene n-methyl-naphthalene biphenyl biphenylene fluorene anthracene andphenantrene)
The areas of the main peaks shown by the GC-FID have beenused to compare the composition of the different samples There-fore the results presented in this work do not represent actualcompositions of the tar samples but they are useful for analyzing
Fig 5 Total ion chromatogram (TIC) of a tar sample obtained at 850 C GR = 08 and H2OO2 molar ratio = 3
Table 6Tar composition (percentage of area in the GC-FID signal of each family of tar compounds)
Experiment 111 111 111 111 111 111 111 111 000a
N-aromatics 746 682 571 687 465 440 447 605 500 plusmn 144O-aromatics 30 77 034 77 06 21 10 60 26 plusmn 08S-compounds 36 29 44 51 79 25 57 10 49 plusmn 01Light aromatics (1 ring) 97 122 44 134 51 99 36 98 65 plusmn 18Light PAH compounds (2ndash3 rings) 91 91 338 514 399 415 450 227 360 plusmn 88
a Mean value plusmn standard deviation
N Gil-Lalaguna et al Chemical Engineering Journal 248 (2014) 373ndash382 381
how the factors influence the fraction of each family of compoundsThe percentages of the GC-FID-areas obtained for each sample areshown in Table 6 According to the ANOVA results the temperatureand the H2OO2 ratio are the only factors affecting tar compositionLight aromatics and O-aromatics are the most sensitive families totemperature Their fractions are found to decrease with tempera-ture Similar results have been reported by other researchers[29] showing that phenolic compounds paraffines olefins andalkylated aromatics are easily cracked at high temperatures TheS-compounds fraction has been found to increase with tempera-ture probably as a result of the aforementioned decrease in thefractions of other compounds
On the other hand N-aromatics and light PAH fractions are themost sensitive families to the H2OO2 ratio The increase in this ra-tio leads to a decrease in the fraction of light PAHs thus the pres-ence of steam seems to prevent the polymerization reactionsAccording to Corella et al [30] tars generated in gasification withsteam are easier to eliminate than tars generated in gasificationwith air Tar molecular weight depends on the presence of H freeradicals which is related to the steam added during gasification[31] A simultaneous increase in the fraction of N-aromatics wasfound but this may only be a consequence of the aforementioneddecrease in the light PAH fraction
4 Conclusions
Temperature was found to be the most influential factor formost of the response variables analyzed during sewage sludge
gasification Higher temperatures are favorable for reducing thetar content and improving the gas yield the gasification efficiencyand the carbon yield to gas phase On the other hand the gas heat-ing value and the H2CO molar ratio in the product gas are clearlyfavored by increasing the steam presence and reducing the oxygenpresence in the gasification medium The significant differencesbetween the theoretical and the experimental yields of gases aswell as the differences in the effects of the factors show howimportant it is to distinguish between kinetic and thermodynamiccontrol in a gasification process
Acknowledgements
The financial support received from the Spanish Ministry of Sci-ence and Technology (Research Project CTQ2010-20137) and fromthe Spanish Ministry of Education (pre-doctoral grant awarded toN Gil-Lalaguna AP2009-3446) is gratefully appreciated
References
[1] D Fytili A Zabaniotou Utilization of sewage sludge in EU application of oldand new methods ndash a review Renew Sustain Energy Rev 12 (2008) 116ndash140
[2] EC European Commission Council Directive 91271EEC of May 21 1991 onthe treatment of urban waste water
[3] W Rulkens Sewage sludge as a biomass resource for the production of energyoverview and assessment of the various options Energy Fuel 22 (2008) 9ndash15
[4] J Werther T Ogada Sewage sludge combustion Prog Energy Combust Sci 25(1999) 55ndash116
[5] P McKendry Energy production from biomass (part 3) gasificationtechnologies Bioresour Technol 83 (2002b) 55ndash63
i9
382 N Gil-Lalaguna et al Chemical Engineering Journal 248 (2014) 373ndash382
[6] L Devi KJ Ptasinski FJJG Janssen A review of the primary measures for tarelimination in biomass gasification processes Biomass Bioenergy 24 (2003)125ndash140
[7] J Gil J Corella MP Aznar MA Caballero Biomass gasification in atmosphericand bubbling fluidized bed effect of the type of gasifying agent on the productdistribution Biomass Bioenergy 17 (1999) 389ndash403
[8] G Schuster G Loumlffler K Weigl H Hofbauer Biomass steam gasification ndash anextensive parametric modeling study Bioresour Technol 77 (2001) 71ndash79
[9] M Campoy A Goacutemez-Barea FB Vidal P Ollero Airndashsteam gasification ofbiomass in a fluidised bed process optimisation by enriched air Fuel ProcessTechnol 90 (2009) 677ndash685
[10] J Gil MP Aznar MA Caballero E Franceacutes J Corella Biomass gasification influidized bed at pilot scale with steamndashoxygen mixtures Product distributionfor very different operating conditions Energy Fuel 11 (1997) 1109ndash1118
[11] PM Lv ZH Xiong J Chang CZ Wu Y Chen JX Zhu An experimental studyon biomass airndashsteam gasification in a fluidized bed Bioresour Technol 95(2004) 95ndash101
[12] F Pinto C Franco RN Andreacute C Tavares M Dias I Gulyurtlu I Cabrita Effectof experimental conditions on co-gasification of coal biomass and plasticswastes with airsteam mixtures in a fluidized bed system Fuel 82 (2003)1967ndash1976
[13] M Dogru A Midilli CR Howarth Gasification of sewage sludge using athroated downdraft gasifier and uncertainty analysis Fuel Process Technol 75(2002) 55ndash82
[14] B Gross C Eder P Grziwa J Horst K Kimmerle Energy recovery from sewagesludge by means of fluidised bed gasification Waste Manage 28 (2008) 1819ndash1826
[15] JJ Manyaacute JL Saacutenchez J Aacutebrego A Gonzalo J Arauzo Influence of gasresidence time and air ratio on the air gasification of dried sewage sludge in abubbling fluidised bed Fuel 85 (2006) 2027ndash2033
[16] A Midilli M Dogru CR Howarth MJ Ling T Ayhan Combustible gasproduction from sewage sludge with a downdraft gasifier Energy ConversManage 42 (2001) 157ndash172
[17] I Petersen J Werther Experimental investigation and modelling ofgasification of sewage sludge in the circulating fluidized bed Chem EngProcess 44 (2005) 717ndash736
[18] N Nipattummakul I Ahmed S Kerdsuwan AK Gupta High temperaturesteam gasification of wastewater sludge Appl Energy 87 (2010) 3729ndash3734
i10
[19] N Gil-Lalaguna I Fonts G Gea MB Murillo L Laacutezaro Reduction of watercontent in sewage sludge pyrolysis liquid by selective on-line condensation ofthe vapors Energy Fuel 24 (2010) 6555ndash6564
[20] G Garciacutea E Cascarosa J Aacutebrego A Gonzalo JL Saacutenchez Use of differentresidues for high temperature desulphurization of gasification gas Chem EngJ 174 (2011) 644ndash651
[21] M Aznar AE Gonzaacutelez JJ Manyagrave JL Saacutenchez MB Murillo Understandingthe effect of the transition period during the air gasification of dried sewagesludge in a fluidized bed reactor Int J Chem React Eng 5 (2007) A18
[22] P McKendry Energy production from biomass (part 1) overview of biomassBioresour Technol 83 (2002a) 37ndash46
[23] M Blaumlsing M Muumlller Release of alkali metal sulphur and chlorine speciesfrom high temperature gasification of high- and low-rank coals Fuel ProcessTechnol 106 (2013) 289ndash294
[24] PA Jensen FJ Frandsen K Dam-Johansen B Sander Experimentalinvestigation of the transformation and release to gas phase of potassiumand chlorine during straw pyrolysis Energy Fuel 14 (2000) 1280ndash1285
[25] J Corella JM Toledo G Molina Calculation of the conditions to get less than2 g tarNm3 in a fluidized bed biomass gasifier Fuel Process Technol 87(2006) 841ndash846
[26] I Wender Reactions of synthesis gas Fuel Process Technol 48 (1996) 189ndash297
[27] YJ Kim SH Lee SD Kim Coal gasification characteristics in a downerreactor Fuel 80 (2001) 1915ndash1922
[28] C Li K Suzuki Tar property analysis reforming mechanism and model forbiomass gasification ndash an overview Renew Sustain Energy Rev 13 (2009)594ndash604
[29] A Ponzio S Kalisz W Blasiak Effect of operating conditions on tar and gascomposition in high temperature airsteam gasification (HTAG) of plasticcontaining waste Fuel Process Technol 87 (2006) 223ndash233
[30] J Corella A Orio JM Toledo Biomass gasification with air in a fluidized bedexhaustive tar elimination with commercial steam reforming catalysts EnergyFuel 13 (1999) 702ndash709
[31] YH Qin J Feng WY Li Formation of tar and its characterization during airndashsteam gasification of sawdust in a fluidized bed reactor Fuel 89 (2010) 1344ndash1347
Fuel 129 (2014) 147ndash155
Reprinted with permission from Elsevier
Contents lists available at ScienceDirect
Fuel
journal homepage wwwelsevier comlocate fuel
Air-steam gasification of char derived from sewage sludge pyrolysisComparison with the gasification of sewage sludge
httpdxdoiorg101016jfuel2014030590016-2361 2014 Elsevier Ltd All rights reserved
uArr Corresponding author Tel +34 976762224 fax +34 976762043E-mail address noemigilunizares (N Gil-Lalaguna)
N Gil-Lalaguna uArr JL Saacutenchez MB Murillo V Ruiz G GeaThermo-chemical Processes Group Aragoacuten Institute of Engineering Research (I3A) Universidad de Zaragoza cMariano Esquillor sn 50018 Zaragoza Spain
h i g h l i g h t s
Increased content of fixed carbon in the solid after sewage sludge pyrolysis Higher gas yield from dried and ash-free (daf) char than from sewage sludge (daf) Average tar yield decreased by 45 when gasifying char instead of sewage sludge Average CO yield was 70 higher when gasifying char (daf basis for solids) Temperature was the most influential factor for most of the studied variables
a r t i c l e i n f o
Article historyReceived 16 January 2014Received in revised form 24 March 2014Accepted 25 March 2014Available online 13 April 2014
KeywordsAir-steam gasificationSewage sludgeCharFast pyrolysisFluidized bed
a b s t r a c t
Air-steam gasification of char derived from fast pyrolysis of sewage sludge has been experimentallyevaluated in a fluidized bed as a route towards a full recovery of energy from sewage sludge The resultshave been compared with those obtained from the direct gasification of sewage sludge in order toevaluate how the previous pyrolysis stage affects the subsequent gasification process The fixed carboncontent in the solid increased after the pyrolysis stage so that heterogeneous reactions of carbon withsteam or CO2 assumed greater importance during char gasification than during sewage sludgegasification Furthermore char gasification led to an improvement in the gas yield -calculated on a dryand ash-free basis (daf)- due to the increased concentration of carbon in the organic fraction of the solidafter the pyrolysis step with an increase in the average CO yield of about 70 -in terms of gkg soliddaf- The reduction in the fraction of carbon which forms tar is another advantage of char gasification overthe direct gasification of sewage sludge with an average decrease of about 45 Regarding the influenceof the operating conditions the response variables were mainly controlled by the same factors in bothprocesses
2014 Elsevier Ltd All rights reserved
1 Introduction
Sewage sludge is the waste generated during successive treat-ment stages of urban wastewaters In recent years the productionof sewage sludge in the EU has considerably increased due to theexpansion in the amount and capacity of wastewater treatmentplants [12] For instance the production of sewage sludge in Spainincreased by 41 in the period 2000ndash2009 [3] For this reason theeconomical and environmentally-friendly treatment of sewagesludge has become an important issue The traditional methodsof treatment or disposal of sewage sludge include its use as fertil-izer on croplands incineration and landfilling [124] However as aresult of the environmental and health problems caused by the
application of these techniques energy recovery from sewagesludge by thermo-chemical treatments such as pyrolysis orgasification technologies could be an interesting alternative [2]
A large number of lab-scale studies on sewage sludge pyrolysisfor liquid production (fast pyrolysis) can be found in the literature[5ndash11] The liquid yield and its physicochemical properties dependon the operational conditions (mainly on the temperature) and onthe composition of the sewage sludge [6] Char is the mainby-product of sewage sludge fast pyrolysis Common solid yieldsof around 35ndash55 wt are found in the literature [8ndash11] but itshould be noted that the ash content in these solids is much higherthan those of lignocellulosic origin The use of this solid by-productas adsorbent material has been investigated by some authors Theresults show that char obtained from sewage sludge pyrolysis isnot a very porous material (its surface area ranges 50ndash150 m2g)because of its high inorganic content [12] Despite this some
ii1
148 N Gil-Lalaguna et al Fuel 129 (2014) 147ndash155
authors have reported a certain capacity of this kind of material toremove contaminants such as H2S NOx metals dyes and phenols[12ndash16] Physical activation of this kind of char was proposed aspart of a three-stage thermo-chemical treatment of sewage sludgein a previous work in our group [17]
On the other hand the remaining organic fraction in char givesit a moderate calorific value which could be further exploitedthrough thermo-chemical processes In fact the gasification of charresulting from fast pyrolysis of different types of biomass is beinginvestigated by some authors as a route towards an integralvalorization of biomass [18ndash22] Furthermore as part of volatilematter is removed from biomass during pyrolysis the gasificationof char obtained from pyrolysis instead of the direct gasification ofbiomass should lead to a reduction in the formation of tar duringthe process which is one of the main hurdles for the developmentof gasification technology
The present work is focused on the gasification of char obtainedfrom sewage sludge fast pyrolysis An experimental study has beencarried out in a lab-scale fluidized bed reactor in order to evaluatethe feasibility of gasifying this kind of char The influence of severaloperating conditions (temperature composition of the gasificationmedium and gasifying agent to biomass ratio) on the gasificationperformance has been analyzed statistically in order to determinethe relative influence of each factor Moreover results from chargasification have been compared with those obtained from thedirect gasification of sewage sludge under the same operating con-ditions [23] in order to evaluate how the previous pyrolysis stageaffects the subsequent gasification process
2 Materials and methods
21 Char obtained from sewage sludge pyrolysis
Char obtained from the fast pyrolysis of anaerobically digestedand thermally dried sewage sludge is the feedstock for the gasifica-tion experiments performed in this work Table 1 presents theresults of the proximate and ultimate analyses and heating valueof the char as well as the results obtained for the original sewagesludge The fixed carbon content in this kind of char is considerablylower than in other types of biomass chars [18ndash22] as the compo-sition of sewage sludge and lignocellulosic materials are quitedifferent
22 Experimental setup
Char was produced during sewage sludge fast pyrolysis in a lab-scale fluidized bed reactor operating at a temperature of 530 CThe pyrolysis plant and the operating conditions are described indetail elsewhere [24]
Table 1Proximate and ultimate analyses and lower heating value of both the char derivedfrom sewage sludge pyrolysis and the sewage sludge itself (SS)
Char SS
Proximate analysis (wt wet basis)Moisture ISO-589-1981 170 648Ash ISO-1171-1976 7420 3904Volatiles ISO-5623-1974 1502 5009Fixed carbon By difference 908 439
Ultimate analysis (wt wet basis Carlo Erba 1108 elemental analyzer)C 1549 2950H 097 467N 185 527S 035 131LHV (MJkg) IKA C-2000 calorimeter 50 118
ii2
Char gasification experiments have also been carried out in alab-scale fluidized bed reactor operating at atmospheric pressurewith continuous feed of solid (around 21 gmin of char) and con-tinuous removal of ash Ash from previous gasification tests consti-tuted the solid bed by itself from the beginning of the runs Thegasifyingfluidizing agent used in the process consisted of differentmixtures of steam and enriched air (air + oxygen) Air flow waskept constant in all the experiments and different flows of pureoxygen were fed together with the air thus enriching the air at dif-ferent percentages
The vapors and gases produced during the gasification processremained inside the reactor around 17ndash18 s and then passedthrough a cyclone and a hot filter (both at 450 C) in which thesolid particles swept by the gas were collected Water and con-densable organic compounds (tar) were collected in two ice-cooledcondensers The volume of particle- and tar-free gas was measuredby a volumetric meter and its composition was analyzed on-lineusing a micro gas chromatograph (Agilent 3000-A) The experi-ments were carried out during 60 min Fig 1 shows a diagram ofthe laboratory installation A more detailed description of the plantcan be found elsewhere [23]
Ash content in the solid by-product was determined accordingto ISO-1171-1976 and its carbon content was analyzed using aLeco TruSpec Micro Elemental Analyzer Water content in the con-densed fraction was analyzed off-line by Karl Fischer titration inorder to determine the amount of tar by difference However tarproduction was almost negligible and all the results from the KarlFischer titration were about 100 wt of water so non-significantdifferences in tar production were found by this way Thereforein order to evaluate the effect of the factors tar production fromchar gasification was approximated to the amount of organic car-bon present in the condensate (g Ccondensate) measured by meansof a total organic carbon analyzer (TOC-L CSHCSN Shimadzuanalyzer)
23 Experimental design and data analysis
A 2k factorial experimental design was planned in order todetermine the influence of some operating factors on the char gas-ification performance This kind of experimental design allows theexistence of interactions between the factors to be identified Inother words it can be seen whether a factor influences a responsevariable in a different way depending on the value of anotherfactor
Three factors have been studied in this work (i) gasificationtemperature measured inside the bed (ranging between 770 and850 C) (ii) gasifying ratio (GR) between the mass flow of gasifyingagent (oxygen plus steam) and the mass flow of dry and ash-free(daf) basis char (ranging between 08 and 11 gg char daf) and(iii) composition of the gasification medium represented by theH2OO2 molar ratio (ranging between 1 and 3) The three studiedfactors together with their respective ranges of study were chosenbased on our previous work on sewage sludge gasification [23] inorder to compare the performance of both processes and evaluatehow a previous pyrolysis stage affects the subsequent gasificationprocess The temperature and the ratio between the flow of oxygenor steam and the feed of biomass are among the most studied fac-tors in the air-steam gasification of biomass [2225]
As seen in Table 2 the experimental design consisted of 8 runs (2k
runs where k is the number of factors in this case 3) Furthermorethree replicates at the center point (CP) were added to the experi-mental design in order to evaluate the experimental variability aswell as to determine if the response of each variable was linear ornot within the studied range Coded values of the factors were usedto identify the term with the greatest influence on each responsevariable that is 1 for the lower limits (T = 770 C GR = 08
Fig 1 Laboratory-scale gasification setup
Table 2Operating conditions in the char gasification experiments
Experiment number 1 2 3 4 5 6 7 8 CP (9 10 11)
Coded values (T GR H2OO2) 111 111 111 111 111 111 111 111 000Temperature (C) 850 770 850 770 850 770 850 770 810g gasifying agentg char daf (GR) 11 11 08 08 11 11 08 08 095H2OO2 molar ratio in the gasifying agent 3 3 3 3 1 1 1 1 2Equivalence ratio (ER) 017 017 012 012 032 032 023 023 019Steam to char daf mass ratio (SB) 071 071 052 052 039 039 027 027 052
N Gil-Lalaguna et al Fuel 129 (2014) 147ndash155 149
and H2OO2 = 1) and +1 for the upper ones (T = 850 C GR = 11 andH2OO2 = 3)
The response variables analyzed were the following (i) distri-bution of products (solid gas and tar) (ii) gas composition deter-mined on-line using a micro-gas chromatograph (iii) productionof each gaseous component based on the amount of char daf fed(iv) lower heating value of the product gas (LHVgas) (v) coldgasification efficiency and (vi) carbon yield to gas phase
The experimental results have been analyzed statistically bymeans of analysis of variance (ANOVA) using a confidence levelof 95 for the F-distribution to identify the terms that significantlyaffect each response variable Design-Expert 7 software (fromStat-Ease Inc) was used for the analyses
3 Results and discussion
Experimental results obtained from the char gasification testsare shown in Table 3 Furthermore as a result of the ANOVA anal-yses Table 4 presents the relative influence of each factor on theresponse variables Average data represent the average of thewhole set of results obtained whereas the coefficients associatedto the different factors (T GR and H2OO2) show the effect thatthe change of each factor has on the studied responses (in termsof coded values for the factors) the existence of significant interac-tions between the factors is also denoted by means of coefficientsIn order to compare these results with those corresponding to thedirect gasification of sewage sludge Table 5 presents a summary of
the ANOVA results obtained when sewage sludge was the feed-stock for the gasification process [23] This comparative study isbased on a single type of sewage sludge and char Thereforealthough the expected trends for other kind of materials will besimilar extrapolation of the results should be done carefully
31 Product distribution
311 Solid yield and carbon fraction remaining as solidThe solid fraction was the most abundant by-product during
char gasification because of the high ash content in the char Thesolid yield varied between 73 and 82 wt (based on the amountof char fed) though this solid was mainly composed of ash (93ndash96 wt) Its carbon content ranged between 38 and 62 wt(Table 3) The fraction of carbon remaining as solid after char gas-ification can be calculated from the above data as follows
Carbon fraction as solid ethwtTHORNfrac14 gCsolid by-product=gCchar fed 100 eth1THORN
It should be noted that the amount of solid introduced in thereactor as initial bed (ash from previous gasification tests) was alsopart of the solid collected after the experiments and contained asmall amount of carbon (between 3 and 5 wt) This amount ofcarbon is not included in g Csolid by-product
The fraction of carbon remaining as solid after char gasificationranged between 15 and 43 wt (Table 3) whereas the maximum
ii3
Table 3Experimental results from char gasification
111 111 111 111 111 111 111 111 000a
Solid yield (g solid100 g char) 757 785 750 785 731 771 752 813 775 plusmn 17Carbon content in the solid product (wt) 39 62 45 61 39 45 39 58 56 plusmn 06Carbon fraction remaining as solid (wt) 195 413 254 431 147 262 188 409 338 plusmn 26Gas yield (m3
STPkg char) 036 027 031 024 035 028 032 024 029 plusmn 001Gas yield (m3
STPkg char daf) 147 112 130 099 146 115 131 100 121 plusmn 001Carbon fraction forming tar (wt) 13 07 29 33 10 57 32 58 28 plusmn 07
Gas composition (dry basis)H2 (vol) 293 263 278 248 215 190 220 202 252 plusmn 06CO (vol) 195 120 202 128 227 140 237 152 159 plusmn 02CO2 (vol) 189 242 162 208 226 295 185 241 219 plusmn 01CH4 (vol) 076 091 077 092 059 070 064 084 088 plusmn 001C2Hx (ppmv) 150 190 160 220 180 220 150 200 180 plusmn 10H2S (vol) 025 012 014 007 017 008 012 006 010 plusmn 001N2 (vol) 313 365 349 407 325 367 350 396 361 plusmn 06
H2CO molar ratio 150 220 138 193 095 136 093 133 158 plusmn 004COCO2 molar ratio 103 049 125 062 100 047 128 063 073 plusmn 001LHVgas (MJm3
STP) 596 471 587 465 544 409 563 443 507 plusmn 007Cold gasification efficiency () 629 411 572 376 574 362 553 357 470 plusmn 06Carbon yield to gas phase (wt) 710 555 621 479 829 671 722 557 615 plusmn 15
a Mean value plusmn standard deviation
Table 4Relative influence of the studied factors on the response variables for char gasification
Average T GR H2OO2 TndashGR TndashH2OO2 GR-H2OO2 TndashH2OO2ndashGR Curvature
Carbon fraction remaining as solid (wt) 3013 916 330 359 a a a a a
Gas yield (m3STPkg char daf) 123 016 008 a a a a a b
Carbon fraction forming tar (wt ) 294 090 081 093 095 a a
H2CO molar ratio in the product gas 148 026 a 031 a a a a a
COCO2 molar ratio in the product gas 085 029 010 a 003 a 001 a b
Yield of gaseous compounds (gkg char daf)H2 4093 603 186 543 033 090 118 a b
CO 42716 13509 a 3461 a 978 a a b
CO2 81596 2909 10491 6995 a a 1875 a a
CH4 1033 a a 101 a a 032 a b
C2H4 028 0023 a a a a a a a
H2S 354 162 103 a a a a a a
LHVgas (MJm3STP) 509 063 a 020 a a 008 a a
Cold gasification efficiency () 4790 1027 145 177 048 a 083 a b
Carbon yield to gas phase (wt) 6429 776 482 516 a a a a b
a Non-significant termb Significant curvature
Table 5Relative influence of the studied factors on the response variables for sewage sludge gasification
Average T GR H2OO2 TndashGR TndashH2OO2 GRndashH2OO2 TndashH2OO2ndashGR Curvature
Carbon fraction remaining as solid (wt) 901 576 174 329 a 150 a a a
Gas yield (m3STPkg SS daf) 112 017 002 a 002 002 a 003 b
Carbon fraction forming tar (wt) 528 111 a a 056 a a a a
H2CO molar ratio in the product gas 221 014 008 062 002 006 011 005 b
COCO2 molar ratio in the product gas 052 016 011 a 004 003 a a a
Yield of gaseous compounds (gkg SS daf)H2 3703 906 137 568 a a a a a
CO 25010 7879 1751 3267 503 2195 a 965 a
CO2 76342 a 10157 9253 2159 a 1493 1591 a
CH4 5023 149 313 480 a a a a a
C2Hx 4613 328 a 335 a a a a a
H2S 1266 173 156 a a a a 062 a
LHVgas (MJm3STP) 549 037 031 040 a 017 a a a
Cold gasification efficiency () 5512 851 a 347 a a a a a
Carbon yield to gas phase (wt) 7248 633 333 607 a a a a a
a Non-significant termb Significant curvature
150 N Gil-Lalaguna et al Fuel 129 (2014) 147ndash155
value for sewage sludge gasification was about 24 wt This differ-ence may be explained by the different structure of the carbona-ceous matter in the solids Most of the carbon in sewage sludge
ii4
is in the form of volatile matter (85 wt of the carbon content)which can be easily released during the gasification stage How-ever the volatile matter in sewage sludge was considerably
Char gasification
770 810 8500
10
20
30
40
50
H2OO2=3 H2OO2=1H2OO2=2
(a)
Temperature (ordmC)
Car
bon
fract
ion
rem
aini
ng a
s so
lid (w
t
)
SS gasification
770 810 8500
10
20
30
40
50
H2OO2=3 H2OO2=1H2OO2=2
(b)
Temperature (ordmC)
Car
bon
fract
ion
rem
aini
ng a
s so
lid (w
t
)
Fig 2 Carbon fraction remaining as solid after (a) char gasification and (b) sewage sludge gasification (gasifying ratio = 095 gg solid daf)
770 810 85009
10
11
12
13
14
15
GR=08GR=11 GR=095
(a)
Temperature (ordmC)
m3 ST
Pkg
cha
r daf
770 810 85001234567
H2OO2=3 H2OO2=1H2OO2=2
(b)
Temperature (ordmC)
Car
bon
fract
ion
form
ing
tar (
wt
)
Fig 3 (a) Gas yield during char gasification (H2OO2 = 2) (b) Carbon fraction forming tar during char gasification (gasifying ratio = 095 gg char daf)
N Gil-Lalaguna et al Fuel 129 (2014) 147ndash155 151
reduced during the pyrolysis stage and about 59 wt of the carbonin char is in the form of fixed carbon which is more difficult to gas-ify than the volatile matter
According to the ANOVA results (Table 4) temperature is themost influential factor on the carbon fraction remaining as solidHigher reaction temperatures favor carbon gasification [26] sothat the carbon fraction remaining as solid was reduced by asmuch as half when the temperature increased from 770 to 850 C(Fig 2a) Carbon conversion is also enhanced by increasing the gas-ifying ratio (GR) andor decreasing the H2OO2 ratio thus indicat-ing that carbon reactivity with oxygen is greater than itsreactivity with steam The same trends were observed in the directgasification of sewage sludge (Table 5) although the carbon frac-tion remaining as solid was even more sensitive to the variationof the factors in that case Furthermore the interaction betweenthe temperature and the H2OO2 ratio was denoted as a significantterm with negligible influence of the gasification medium compo-sition at the higher temperature (Fig 2b) The error bars shown inthe figures of results (Figs 2 3 and 5) correspond to the least sig-nificant difference (LSD)
As mentioned above carbon conversion was higher for sewagesludge gasification than for char gasification However results forchar gasification can be recalculated considering both stages (pyro-lysis + gasification) as a whole and taking the initial amount of car-bon in sewage sludge as a reference for calculating the carbonconversion In this way the fraction of carbon remaining as solidafter char gasification is reduced to 4ndash11 wt thus improvingthe carbon conversion obtained in the direct gasification of sewagesludge
312 Gas productionThe gas yield from char gasification varied between 024 and
036 m3STPkg char (N2-free basis) or between 040 and 052 m3
STPkgchar if N2 is included (where STP means standard conditions of tem-perature and pressure at 0 C and 1 atm) Comparing these data withthose corresponding to sewage sludge gasification (049ndash072 m3
STP
N2-freekg SS) [23] it can be observed that the production of gashas been reduced by half mainly due to the higher ash content inchar The production of gas during the pyrolysis stage (around006ndash007 m3
STP N2-freekg SS) is not high enough to offset the differ-ence in the production of gas from the gasification of both materials
On the other hand if the gas yield (N2-free basis) is calculated tak-ing into account only the organic content in the raw material it ran-ged between 099 and 147 m3
STPkg char daf for char gasification and089ndash132 m3
STPkg SS daf for sewage sludge gasification [23] thus indi-cating that the previous pyrolysis stage leads to structural changes inthe organic fraction of the solid that improve the production of gas Gasyield results obtained from the gasification of sewage sludge-derivedchar (expressed on a N2-free and daf basis) are in the same range asthose obtained from char derived from lignocellulosic materials suchas bagasse char [18] or char derived from ramie residues [22]
As with the gasification of sewage sludge temperature is themost influential factor on the production of gas during char gasifi-cation (Table 4) An average gas yield improvement of about 30was obtained when the temperature varied from 770 to 850 C inthe gasification of char (Fig 3a) Although to a lesser extent theincrease in the gasifying ratio (GR) is also favorable for the produc-tion of gas whereas the nature of the gasification medium does notexert a significant influence on the gas yield obtained from the
ii5
0
5
10
15
20
25
30
35
H2
(vol
)
5 6 8 7 CP 4 3 1 2
Number of experiment
05
1015202530354045
CO (v
ol
)
2 6 1 5 CP 4 3 8 7
Number of experiment
0
5
10
15
20
25
30
35
CO2
(vol
)
3 4 7 8 CP 1 2 5 6
Number of experiment
00
02
04
06
08
10
CH4
(vol
)
5 1 7 6 CP 3 2 8 4
Number of experiment
Fig 4 Equilibrium (d) and experimental ( ) fractions of H2 CO CO2 and CH4 in the product gas from char gasification
Char gasification
770 810 850100
200
300
400
500
600
H2OO2=3 H2OO2=1H2OO2=2
(a)
Temperature (ordmC)
g C
O
kg c
har d
af
SS gasification
770 810 850100
200
300
400
500
600
H2OO2=3 H2OO2=1H2OO2=2
(b)
Temperature (ordmC)
g C
O
kg S
S da
f
Fig 5 CO production during (a) char gasification and (b) sewage sludge gasification (gasifying ratio = 095 gg solid daf)
152 N Gil-Lalaguna et al Fuel 129 (2014) 147ndash155
gasification of either char or sewage sludge Therefore as can benoted the negative effect of H2OO2 on the gasified carbon fraction(discussed in the previous section) does not result in a significantgas yield decrease The production of H2 may be the major reasonfor this disagreement because as discussed below H2 formation ispromoted by increasing the H2OO2 ratio (mainly through thewaterndashgas shift reaction) thus counteracting the decrease in theproduction of gaseous carbon-compounds
Gas yield does not follow a linear trend with all factors as cur-vature appears as a significant term in the ANOVA analysis Thismeans that at least one of the three factors has a quadratic effecton the evolution of gas production Some studies reported in theliterature show that excess steam is not favorable for the produc-tion of gas during steam gasification and suggest optimal valuesfor steam to carbon ratios in order to maximize it [20222527]The observed curvature may therefore be associated with the pres-ence of steam in the gasification medium
313 Tar productionThe fraction of carbon which forms tar can be calculated as
follows
ii6
Carbon fraction as tar ethwtTHORNfrac14 gCcondensate=gCchar fed 100 eth2THORN
The fraction of carbon which formed tar during chargasification ranged between 07 and 58 wt (Table 3) andaccording to the ANOVA results (Table 4) it can be reduced byincreasing any of the studied factors though the effect oftemperature disappears at the higher H2OO2 ratio and the effectof H2OO2 is negligible at the higher temperature (Fig 3b) Onthe other hand the average carbon fraction forming tar duringsewage sludge gasification was about 18 times higher than dur-ing char gasification and only the temperature and its interactionwith the gasifying ratio were found to be significant terms(Table 5)
The production rates of tar and gas allow the tar content in theproduct gas to be calculated (g tarm3
STP) The tar content in the gasfrom char gasification (by approximation of the amount of tar to theamount of organic carbon present in the condensate) rangedbetween 2 and 13 g tarm3
STP under most operating conditions whilethe results for sewage sludge gasification ranged between 11 and45 g tarm3
STP [23]
N Gil-Lalaguna et al Fuel 129 (2014) 147ndash155 153
32 Gas composition
The composition of the product gas from a gasification processis the result of many complex and competing reactions The mostrepresentative reactions include the waterndashgas shift reaction(WGS) oxidation reactions waterndashgas reactions (reactions of car-bon with steam) steam and dry reforming of hydrocarbons themethanation reaction and the Boudouard reaction [25ndash27]
As can be seen in Table 3 H2 (190ndash293 vol) CO (120ndash237 vol) CO2 (162ndash295 vol) CH4 (059ndash092 vol) and N2
(313ndash407 vol) were the main gases detected by the micro GCduring char gasification Other minor compounds were alsodetected in the gas such as C2Hx hydrocarbons (mainly C2H4) orH2S which is released during the process due to the presence ofsulfur-compounds in the char (Table 1) Statistical results fromthe analysis of gas composition have not been included in thiswork because it was preferred to analyze the yield of each gaseouscompound in terms of gkg char daf (Section 33) However thevariation of the product gas composition has been evaluatedthrough two molar ratios H2CO and COCO2 On the one handthe H2CO ratio in the product gas is increased by reducing thetemperature andor increasing H2OO2 in the gasification mediumthis last factor being the most influential (Table 4) These trendsare consistent with those obtained for sewage sludge gasification(Table 5) although in that case the gasifying ratio also played a sig-nificant role in the evolution of the H2CO ratio (positive effect) aswell as the interactions between the factors and the curvatureterm Higher values of H2CO were obtained from sewage sludgegasification (146ndash325) than from char gasification (093ndash220)On the other hand temperature is the most influential factor onthe COCO2 ratio in the exit gas in both processes The COCO2 ratiocan be improved by increasing the temperature andor reducingthe amount of gasifying agent fed to the reactor (Table 4) The cur-vature has been denoted as a significant term in the evolution ofCOCO2 as well as some interactions between the factors Thesame trends were observed for sewage sludge gasificationalthough in that case the COCO2 ratio followed a linear responseHigher values of COCO2 have been obtained from char gasification(049ndash128) than from sewage sludge gasification (030ndash088)
The theoretical composition of the gas at equilibrium conditionswas also calculated in order to determine if the gasification processwas kinetically or thermodynamically controlled The HSC Chemis-try61 software was used to obtain the theoretical composition ofthe gas under the same operating conditions tested in the labora-tory The results from the theoretical simulations varied withinthe following ranges H2 (243ndash339 vol) CO (313ndash411 vol)CO2 (30ndash138 vol) N2 (248ndash287 vol) CH4 (87ndash5801 ppmv)H2S (035ndash040 vol) and NH3 (17ndash42 ppmv) The H2CO and COCO2 ratios at equilibrium conditions ranged 062ndash109 and 233ndash1290 respectively Experimental fractions of H2 and CO werelower than their corresponding theoretical values whereas exper-imental fractions of CO2 and CH4 were above their correspondingtheoretical values (Fig 4) The significant differences observed inthe concentration ranges reveal that chemical equilibrium wasnot reached during the experimental tests
33 Production of each gaseous compound
The production or yield of each gas (H2 CO CO2 CH4 C2H4 andH2S) is expressed in terms of mass of gas (g) produced per kilogramof char daf fed
According to the ANOVA results (Table 4) gasification temper-ature is the most influential factor on the production of both H2
and CO during char gasification These gases are involved in manyreactions both as reactants and as products but the temperaturerise seems to enhance their formation rather than their consuming
reactions Although to a lesser extent the production of H2 is alsoimproved by increasing the H2OO2 ratio unlike the CO yield whichdecreases with H2OO2 Both trends are consistent with the WGSreaction (CO + H2O M H2 + CO2) which is one of the most represen-tative reactions for a steam gasification process Besides increasingthe steam presence the oxygen presence is reduced with theincrease in H2OO2 so combustion reactions should be mitigatedThe negative effect of H2OO2 on the production of CO suggeststhat the WGS reaction outweighs the combustion reactions in theevolution of the CO yield The same trend was found when directlygasifying the sewage sludge (Table 5) Similarly Franco et al [27]found that the WGS reaction appeared to be the most dominantreaction in the steam gasification of biomass for the temperaturerange of 730ndash830 C For higher temperatures (830ndash900 C) steamreforming of carbon (waterndashgas reactions) prevailed althoughthese reactions also appeared to contribute significantly at temper-atures lower than 830 C for some types of biomass In the presentstudy an upward trend in CO production with increasing temper-ature was found As the process is controlled by kinetics thisbehavior cannot be explained through the WGS reaction alonebut through the steam reforming of carbon (C + H2O M CO + H2)the Boudouard reaction (C + CO2 M 2 CO) and the steam and dryreforming of hydrocarbons in which CO is formed which seem togain importance at higher temperatures As shown in Fig 5 thepositive effect of temperature on the CO yield slightly diminisheswith increased steam presence due to the enhancement of theWGS reaction
The gasifying ratio (GR) does not significantly affect the CO yieldand only slightly influences the H2 yield in a positive way duringchar gasification (Table 4) However the amount of gasifying agentis the most influential factor on the production of CO2 the higherthe gasifying ratio the greater the amount of CO2 producedIncreasing the gasifying ratio means more oxygen and more steamfed to the gasifier so combustion reactions as well as CO2 forma-tion through other reactions promoted by the presence of steam(such as the WGS reaction) take place to a greater extent The gas-ification temperature and the composition of the gasification med-ium also exert a significant influence on the production of CO2 Thetemperature rise reduces the formation of CO2 and as discussedabove favors the production of CO thus suggesting once againthe importance of the Boudouard reaction at high temperaturesThe negative effect of H2OO2 on the CO2 yield reveals that com-bustion is the main source of CO2 in the process
Regarding the production of light hydrocarbons (CH4 and C2H4)during char gasification the experimental variability was consider-able (15 for C2H4) so only those factors with a very clear effectwere denoted as significant terms in the ANOVA analysis The com-position of the gasification medium was found to be the only factoraffecting the production of CH4 increasing H2OO2 involves greaterCH4 production thus suggesting that its consumption throughcombustion reactions outweighs its steam reforming processMethane formation via the methanation reaction (C + 2H2 M CH4)may also be promoted by increasing the H2OO2 ratio due to anincreased presence of H2 in the gasification medium On the otherhand temperature is the only factor affecting the C2H4 yield(Table 4) and as expected a downward trend with increasing tem-perature was found since higher temperatures provide morefavorable conditions for thermal cracking and reforming reactions[25]
Lastly according to the ANOVA results (Table 4) the productionof H2S during the gasification of char is significantly affected by thegasification temperature and the gasifying ratio although theeffect of the latter factor is less significant The production of H2Sis promoted by the temperature rise (process controlled by kinet-ics) Moreover the production of H2S is favored by the steam pres-ence (COS + H2O M H2S + CO2) [28]
ii7
154 N Gil-Lalaguna et al Fuel 129 (2014) 147ndash155
Some other conclusions can be drawn by comparing the resultsderived from char gasification and sewage sludge gasification
ndash Average yield to H2 was very similar for both feedstocks (41 gkg char daf and 37 gkg SS daf) whereas average yields to COand CO2 (gkg daf) were 70 and 6 higher in the gasificationof char respectively The production of light hydrocarbonsand H2S was much lower when char was gasified due to the pre-vious release of these compounds in the pyrolysis stage (about4ndash5 mg H2Sg sewage sludge released during the pyrolysisstep) However it should be noted that if gas yields are calcu-lated with respect to the whole feedstock and not only consid-ering the dry and ash-free material the production of all the gascomponents is clearly greater during the gasification of sewagesludge
ndash The production of each gas is mainly controlled by the same fac-tor in both processes The gasification temperature is the mostinfluential factor on the production of H2 CO and H2S the gas-ifying ratio is the most significant factor on the CO2 yield andthe composition of the gasification medium exerts the greatestinfluence on the CH4 yield However some differences relatedto minor influences of the factors have also been found Forexample temperature did not affect the production of CO2 inthe gasification of sewage sludge while it had a negative effectduring char gasification An increased reactivity of char withCO2 (Boudouard reaction) may explain this difference Further-more the gasifying ratio did not affect the production of CO inthe gasification of char but it had a negative effect during thegasification of sewage sludge This implies that the consump-tion of CO through combustion or through the WGS reactionduring char gasification is offset by an increased production ofCO from heterogeneous reactions between carbon and steam(waterndashgas reactions) or carbon and CO2 (Boudouard reaction)since the fixed carbon content is higher in char (908 wt) thanin sewage sludge (439 wt)
ndash The production of each gas during the gasification of sewagesludge follows a linear response with the factors whereas cur-vature appears as a significant term in the production of somegases during char gasification
34 Lower heating value of the product gas
The lower heating value of the gas is calculated as follows
LHVgas frac14 Rethxi LHViTHORN eth3THORN
where xi and LHVi are the volumetric fraction and the lower heatingvalue (MJm3
STP) of each gaseous component respectively Thereforethe variation in the gas heating value only depends on the gas com-position evolution
The lower heating value of the product gas from char gasifica-tion ranged between 409 and 596 MJm3
STP (Table 3) thus definingthis gas as a low heating value gas [26] According to the ANOVAresults (Table 4) the gas heating value follows a linear trend withthe gasification temperature and the H2OO2 ratio the temperaturebeing the most influential factor As remarked above the tempera-ture rise leads to a decrease in the production of CO2 and a simulta-neous increase in the yields of H2 and CO These variations outweighthe decrease in the content of light hydrocarbons thus resulting in apositive effect of the temperature on the gas heating value The com-position of the gasification medium also exerts a significant influ-ence on the gas heating value when H2OO2 is increased thecontent of CH4 increases and the content of CO2 decreases so botheffects contribute to improve the LHVgas The effect of the gasifica-tion medium is intensified when more gasifying agent is fed to thereactor (significant interaction between the gasifying ratio andH2OO2)
ii8
Despite feeding different flows of nitrogen to the reactor gaslower heating values from char gasification (409ndash596 MJm3
STP)are in the same range as those obtained from sewage sludge gasifi-cation (412ndash620 MJm3
STP) [23] Temperature plays the most impor-tant role in the evolution of the gas heating value when char isgasified (Table 4) while the three studied factors exerted similar rel-ative influences on the gas heating value from sewage sludge gasifi-cation (Table 5)
35 Cold gasification efficiency
The cold gasification efficiency without taking into account thesensible heat of the gases is defined as follows
Gasification efficiency ethTHORNfrac14 ethGasvolume LHVgasTHORN=ethCharmass LHVcharTHORN 100
eth4THORN
where Gasvolume is the total production of gas (m3STP including the
amount of N2) Charmass is the amount of char fed during each exper-iment (kg) and LHVgas and LHVchar are the lower heating values ofthe product gas and of the char expressed on MJm3
STP and MJkgrespectively
The cold efficiency for char gasification ranged between 36and 63 (Table 3) These values are quite similar to those obtainedfor the gasification of sewage sludge (39ndash66) [23] According tothe ANOVA results (Table 4) the response of char gasification effi-ciency does not follow a linear trend with all the factors since thecurvature was denoted as a significant term Temperature is clearlythe most influential factor on the gasification efficiency and itsvariation from 770 to 850 C improved the char gasification effi-ciency by about 20 As remarked above both the gas heatingvalue and the gas yield were enhanced at high temperatures Theother factors (H2OO2 and gasifying ratio) also have a positiveeffect on the char gasification efficiency but play a less importantrole in its variation Moreover some interactions between the fac-tors appear as significant terms in the evolution of the char gasifi-cation efficiency the positive effects of temperature and H2OO2
are intensified by increasing the gasifying ratioTemperature and H2OO2 also have a positive effect on the sew-
age sludge gasification efficiency (Table 5) the temperature beingthe most influential factor However the gasifying ratio did notexert a significant influence in this case because its positive effecton the production of gas was counteracted by its negative effect onthe gas heating value The response of the sewage sludge gasifica-tion efficiency was linear with its two significant factors
36 Carbon yield to gas phase
The carbon yield to gas phase is defined as follows
Carbon yield to gas phase ethTHORN frac14 gCproduct gas=gCchar fed 100 eth5THORN
Although the conversion of solid carbon during char gasificationreached 57ndash85 wt the carbon yield to gas phase was slightlylower (between 48 and 83 wt) since not all the converted carbonproduced gaseous compounds However both variables are linkedsince a decreased carbon fraction remaining as solid led to anincreased production of carbon-containing gases This link isshown by the ANOVA results as the same factors that affectedthe carbon fraction remaining as solid also affect the carbon frac-tion which forms gas but in opposite directions The same trendswere observed when gasifying sewage sludge (Table 5) thoughthe difference between carbon conversion (76ndash98 wt) and car-bon yield to gas phase (62ndash90 wt) was more significant becauseof the greater formation of tar
Although carbon yield to gas phase achieved in sewage sludgegasification was higher than that for char gasification gas
N Gil-Lalaguna et al Fuel 129 (2014) 147ndash155 155
production calculated on a daf basis was better for char gasifica-tion This may be explained by the increased concentration ofcarbon in the dried and ash-free fraction of the solid after thepyrolysis step (064 g Cg char daf vs 054 g Cg SS daf)
4 Conclusions
Gasification of char obtained from fast pyrolysis of sewagesludge has been experimentally studied in this work The resultshave been compared with those obtained from the direct gasifica-tion of sewage sludge in order to evaluate how the previous pyro-lysis stage affects the subsequent gasification process Most of thecarbon in the sewage sludge was in the form of volatile matter(85 wt) while almost 60 wt of the carbon in char was in theform of fixed carbon thus causing differences in the gasificationperformances of both materials The carbon fraction remaining assolid after char gasification was higher than that for sewage sludgegasification Despite this gas production (expressed on a dry andash-free basis daf) was improved when gasifying char due to theincreased concentration of carbon in the dried and ash-free frac-tion of the solid after the pyrolysis step (064 g Cg char daf vs054 g Cg SS daf)
The comparison of theoretical and experimental results showedthat equilibrium conditions were not reached during the gasifica-tion experiments of either char or sewage sludge so both processeswere controlled by kinetics The average yield to H2 (expressed asgkg solid daf) was very similar for both feedstocks whereas aver-age yields to CO and CO2 (gkg solid daf) were 70 and 6 higher inthe gasification of char respectively On the other hand the pro-duction of light hydrocarbons and tar was significantly reducedduring char gasification due to the reduction in the volatile matterof the solid after the pyrolysis step The gasification efficiency andthe gas heating value varied in similar ranges in both processes
All the studied variables were mainly controlled by the sameoperating factor (temperature composition of the gasificationmedium or gasifying agent to biomass ratio) in both char gasifica-tion and sewage sludge gasification Temperature was the mostinfluential factor on the carbon conversion gasification efficiencygas yield production of H2 CO and H2S and COCO2 ratio in theproduct gas from both processes affecting all of them positivelyThe gasifying ratio was the most significant factor on the produc-tion of CO2 (positive effect) whereas the composition of the gasifi-cation medium exerted the greatest influence on the CH4 yield andH2CO ratio in the product gas (enhanced by the presence ofsteam) Temperature also played the most important role in theevolution of the gas heating value when char was gasified whilethe three studied factors exerted similar relative influences onthe gas heating value from sewage sludge gasification
In summary results show how the increased content of fixedcarbon in the solid after the pyrolysis step leads to a greater impor-tance of heterogeneous reactions at high temperatures such as thesteam reforming of carbon or the Boudouard reaction
Acknowledgements
The authors gratefully acknowledge the financial support pro-vided by the Spanish Ministry of Science and Technology (researchproject CTQ2010-20137) and the Spanish Ministry of Education(pre-doctoral grant awarded to N Gil-Lalaguna AP2009-3446)
References
[1] Fytili D Zabaniotou A Utilization of sewage sludge in EU application of old andnew methods ndash a review Renew Sust Energy Rev 200812116ndash40
[2] Rulkens W Sewage sludge as a biomass resource for the production of energyoverview and assessment of the various options Energy Fuel 2008229ndash15
[3] Statistical Office of the European Communities (Eurostat) Water statisticstotal sewage sludge production from urban wastewater (2009) Retrieved fromeppeurostateceuropaeu
[4] Werther J Ogada T Sewage sludge combustion Prog Energy Combust Sci19992555ndash116
[5] Fonts I Gea G Azuara M Aacutebrego J Arauzo J Sewage sludge pyrolysis for liquidproduction a review Renew Sustain Energy Rev 2012162781ndash805
[6] Fonts I Azuara M Gea G Murillo MB Study of the pyrolysis liquids obtainedfrom different sewage sludge J Anal Appl Pyrol 200985184ndash91
[7] Kim Y Parker W A technical and economic evaluation of the pyrolysis ofsewage sludge for the production of bio-oil Bioresoure Technol2008991409ndash16
[8] Shen L Zhang DK An experimental study of oil recovery from sewage sludgeby low-temperature pyrolysis in a fluidised-bed Fuel 200382465ndash72
[9] Fonts I Juan A Gea G Murillo MB Saacutenchez JL Sewage sludge pyrolysis influidized bed 1 influence of operational conditions on the productdistribution Ind Eng Chem Res 2008475376ndash85
[10] Inguanzo M Domiacutenguez A Meneacutendez JA Blanco CG Pis JJ On the pyrolysis ofsewage sludge the influence of pyrolysis conditions on solid liquid and gasfractions J Anal Appl Pyrol 200263209ndash22
[11] Pokorna E Postelmans N Jenicek P Schreurs S Carleer R Yperman J Study ofbio-oils and solids from flash pyrolysis of sewage sludges Fuel2009881344ndash50
[12] Smith KM Fowler GD Pullket S Graham NJD Sewage sludge-basedadsorbents a review of their production properties and use in watertreatment applications Water Res 2009432569ndash94
[13] Lu GQ Lau DD Characterisation of sewage sludge-derived adsorbents for H2Sremoval 2 Surface and pore structural evolution in chemical activation GasSep Purif 199610103ndash11
[14] Jindarom C Meeyoo V Kitiyanan B Rirksomboon T Rangsunvigit P Surfacecharacterization and dye adsorptive capacities of char obtained frompyrolysisgasification of sewage sludge Chem Eng J 2007133239ndash46
[15] Pietrzak R Bandosz TJ Reactive adsorption of NO2 at dry conditions on sewagesludge-derived materials Environ Sci Technol 2007417516ndash22
[16] Garciacutea G Cascarosa E Aacutebrego J Gonzalo A Saacutenchez JL Use of different residuesfor high temperature desulphurisation of gasification gas Chem Eng J2011174644ndash51
[17] Aacutebrego J Saacutenchez JL Arauzo J Fonts I Gil-Lalaguna N Atienza-Martiacutenez MTechnical and energetic assessment of a three-stage thermochemicaltreatment for sewage sludge Energy Fuel 2013271026ndash34
[18] Chaudhari ST Dalai AK Bakhshi NN Production of hydrogen andor syngas(H2+CO) via steam gasification of biomass-derived chars Energy Fuel2003171062ndash7
[19] Haykiri-Acma H Yaman S Kucukbayrak S Gasification of biomass chars insteam-nitrogen mixture Energy Convers Manage 2006471004ndash13
[20] Yan F Luo S Hu Z Xiao B Cheng G Hydrogen-rich gas production by steamgasification of char from biomass fast pyrolysis in a fixed bed reactorinfluence of temperature and steam on hydrogen yield and syngascomposition Bioresoure Technol 20101015633ndash7
[21] Mohd Salleh MA Kisiki NH Yusuf HM Ghani WAK Gasification of biocharfrom empty fruit bunch in a fluidized bed reactor Energies 201031344ndash52
[22] He P Luo S Cheng G Xiao B Cai L Wang J Gasification of biomass char withair-steam in a cyclone furnace Renew Energy 201237398ndash402
[23] Gil-Lalaguna N Saacutenchez JL Murillo MB Rodriacuteguez E Gea G Air-steamgasification of sewage sludge in a fluidized bed Influence of some operatingconditions Chem Eng J 2014 In press doi 101016jcej201403055
[24] Gil-Lalaguna N Fonts I Gea G Murillo MB Laacutezaro L Reduction of watercontent in sewage sludge pyrolysis liquid by selective on-line condensation ofthe vapors Energy Fuels 2010246555ndash64
[25] Lv PM Xiong ZH Chang J Wu CZ Chen Y Zhu JX An experimental study onbiomass air-steam gasification in a fluidized bed Bioresoure Technol20049595ndash101
[26] McKendry P Energy production from biomass (part 3) gasificationtechnologies Bioresoure Technol 20028355ndash63
[27] Franco C Pinto F Gulyurtlu I Cabrita I The study of reactions influencing thebiomass steam gasification process Fuel 200382835ndash42
[28] Hepola J Simell P Kurkela E Stahlberg P Sulphur poisoning of nickelcatalyst in catalytic hot gas cleaning conditions of biomass gasification InProceedings of the 6th International Symposium Catalyst deactivation 1994p 499ndash506
ii9
lable at ScienceDirect
Energy 76 (2014) 652e662
Reprinted with permission from Elsevier
Contents lists avai
Energy
journal homepage wwwelsevier comlocateenergy
Energetic assessment of air-steam gasification of sewage sludge and ofthe integration of sewage sludge pyrolysis and air-steam gasificationof char
N Gil-Lalaguna JL Sanchez MB Murillo M Atienza-Martiacutenez G GeaThermo-chemical Processes Group Aragon Institute of Engineering Research (I3A) Universidad de Zaragoza cMariano Esquillor sn 50018 Zaragoza Spain
a r t i c l e i n f o
Article historyReceived 23 March 2014Received in revised form25 July 2014Accepted 17 August 2014Available online 15 September 2014
KeywordsSewage sludgeCharAir-steam gasificationPyrolysisThermal drying
Corresponding author Tel thorn34 976762224E-mail address noemigilunizares (N Gil-Lalagu
httpdxdoiorg101016jenergy2014080610360-5442copy 2014 Elsevier Ltd All rights reserved
a b s t r a c t
Thermo-chemical treatment of sewage sludge is an interesting option for recovering energy andorvaluable products from this waste This work presents an energetic assessment of pyrolysis and gasifi-cation of sewage sludge also considering the prior sewage sludge thermal drying and the gasification ofthe char derived from the pyrolysis stage Experimental data obtained from pyrolysis of sewage sludgegasification of sewage sludge and gasification of char (all of these performed in a lab-scale fluidizedreactor) were used for the energetic calculations The results show that the energy contained in theproduct gases from pyrolysis and char gasification is not enough to cover the high energy consumptionfor thermal drying of sewage sludge Additional energy could be obtained from the calorific value of thepyrolysis liquid but some of its properties must be improved facing towards its use as fuel On the otherhand the energy contained in the product gas of sewage sludge gasification is enough to cover theenergy demand for both the sewage sludge thermal drying and the gasification process itself Further-more a theoretical study included in this work shows that the gasification efficiency is improved whenthe chemical equilibrium is reached in the process
copy 2014 Elsevier Ltd All rights reserved
1 Introduction
Sewage sludge is the major by-product of wastewater treat-ment The sludge stemming from thewastewater treatment usuallyappears in the form of a dilute suspension which typically containsfrom 025 to 12 wt of dry solid matter depending on the opera-tion and process used [1] The generated amount of this waste hasincreased in recent years due to the stricter European legislationconcerning urban wastewater treatment [2] which has led to anincrease in the number of wastewater treatment plants As aconsequence sewage sludge management has become an impor-tant issue [3]
Sewage sludge has been traditionally used as fertilizer due to itsorganic matter and nutrient content However the presence ofvarious contaminant elements in the sludge such as heavy metalsorganic contaminants and pathogenic bacteria limits this practicewhich is regulated by European environmental legislation [4]Landfill disposal and incineration are other common ways of
na)
sewage sludge management but they are not exempt from draw-backs [1] Thus different energy valorization technologies arecurrently being developed Among them thermo-chemical pro-cesses such us gasification and pyrolysis represent interesting op-tions since they could provide energy andor valuable productsfrom sewage sludge [56] A large number of lab-scale studies onsewage sludge pyrolysis for liquid production (fast pyrolysis) can befound in the literature [7e9] In addition to the liquid fraction a gasstream and a carbonaceous solid by-product (char) are also ob-tained in the process The remaining organic fraction in char gives itamoderate calorific valuewhich could be further exploited throughthermo-chemical processes such as combustion or gasificationthus providing a route towards the complete energetic valorizationof the biomass [10e12] In addition to the pyrolysis works sewagesludge gasification has been studied since mid-1990s [13] Sincethen numerous studies have been performed at laboratory plants[14e17] and the process has even been tried at demonstration andpilot scale [18e20] Most of these studies used air as a gasificationmedium but steam gasification or supercritical water gasificationof sewage sludge have also been performed in order to enhance H2
production and improve gas quality [2122] However the additionof steam into a gasification process accelerates a series of
iii1
Abbreviations
DH total enthalpyDHcond enthalpy of condensationDH
f standard enthalpy of formation at 298 KDHvap enthalpy of vaporizationAP aqueous phaseCp specific heat capacitydaf dry and ash-freeDSC differential scanning calorimetryER equivalence ratioHHV higher heating valueHOP heavy organic phase
LHV lower heating valueLOP light organic phasem mass flow ratehgas dry gas yieldQ heat of reactionQdrying heat for thermal dryingSB steam to dry and ash-free biomass mass ratioSC steam to carbon molar ratioSS sewage sludgeSTP standard conditions of temperature and pressure (273
K 1 atm)Tb boiling pointTref reference temperature (298 K)
N Gil-Lalaguna et al Energy 76 (2014) 652e662 653
endothermic reactions that result in a temperature decreasemaking it more difficult to achieve an autothermal process [2324]Therefore not only technical and operational aspects should betaken into account for the development of a gasification processEnergetic assessment is also a key issue especially when steam isused as a gasifying agent
The energy needed for steam gasification can be achieved by theaddition of oxygen (or air since the use of pure oxygen raises theprocess cost) together with the steam into the gasificationmediumwhich causes the combustion of part of the organic matter and therelease of energy The gasification temperature is controlled by theoxygen supply itself in an autothermal gasifier while the transfer ofexternal heat is required in an allothermal gasifier to maintain asuitable temperature during the process
Several works reported in the literature describe energetic as-pects related to the gasification and pyrolysis of different types ofbiomass [25e29] but not specifically refer to the use of sewagesludge Given this background this paper presents an energeticassessment of two potential treatments for sewage sludge (i) two-stage process sewage thermal drying thorn air-steam gasification ofsewage sludge and (ii) three-stage process sewage sludge thermaldrying thorn pyrolysis of sewage sludge thorn air-steam gasification of thechar derived from the pyrolysis stage Fig 1 shows a schematicoverview of both treatments The objective of this study is todetermine the overall energy demand of these thermo-chemical
Fig 1 Schematic overview of the thermo-chemical p
iii2
processes by considering the individual energy requirement ofeach stage (drying pyrolysis and gasification) Experimental dataresulting from the pyrolysis and gasification stages were used in theenergy balances Furthermore theoretical simulations of the gasi-fication stages (performed with the HSC Chemistryreg 61 softwareand based on the Gibbs energy minimization method) were alsoconducted in order to evaluate the thermodynamic restrictions ofthe process under different scenarios
2 Materials and methods
21 Sewage sludge and char
Table 1 provides a brief characterization of both the anaerobi-cally digested and thermally dried SS (sewage sludge) onwhich thestudy is based and the char obtained experimentally from thepyrolysis of this sewage sludge Proximate analyses were per-formed according to standard methods (ISO-589-1981 for mois-ture ISO-1171-1976 for ash and ISO-5623-1974 for volatiles) whilethe ultimate analyses were determined with a Carlo Erba EA1108elemental analyzer The heating values of the solids were measuredwith an IKA C-2000 calorimeter and their specific heat capacitieswere determined by differential scanning calorimetry using aNetzsch DSC 200 Maia Thermobalance (inert atmosphere40 mL min1 of nitrogen)
rocesses proposed for sewage sludge treatment
Table 1Characterization of sewage sludge (SS) and char derived from sewage sludgepyrolysis
SS Char
Proximate analysis (wt wet basis)Moisture 648 170Ash 3904 7420Volatiles 5009 1502Fixed carbon 439 908Ultimate analysis (wt wet basis)C 2950 1549H 467 097N 527 185S 131 035LHV (MJ kg1) 118 50Cp25C (kJ kg1 K1) 115 082
N Gil-Lalaguna et al Energy 76 (2014) 652e662654
22 Experimental setup
Sewage sludge fast pyrolysis was performed in a lab-scale flu-idized bed reactor operating at 530 C and using 45 m3
STP min1 ofnitrogen (where STP means standard conditions of temperatureand pressure at 0 C and 1 atm) as fluidizing agent Two ice-cooledcondensers and an electrostatic precipitator were used tocondensate the produced vapors The composition of the dryproduct gas was analyzed on-line with a micro gas chromatograph(Agilent 3000) The pyrolysis plant and the operating conditions aredescribed in detail elsewhere [30] The liquid collected aftercondensation of the vapors was separated into three phases LOP(light organic phase) HOP (heavy organic phase) and AP (aqueousphase) The water content of each phase was analyzed by KarlFischer titration (Mettler Toledo titrator) while its compositionwasqualitatively determined by gas chromatography-mass spectrom-etry (Hewlett Packard HP 5890 A) Ultimate analysis and higherheating value of each liquid phase were determined with a CarloErba EA1108 elemental analyzer and an IKA C-2000 analyticalcalorimeter respectively As commented above the solid by-product resulting from the pyrolysis process was also character-ized and used as a raw material for the gasification process
The experiments of sewage sludge gasification and char gasifi-cation were carried out in a lab-scale fluidized bed reactor oper-ating at atmospheric pressure and in a temperature range of770e850 C More details about the gasification setup can be foundelsewhere [3132] Different mixtures of steam and air (or enrichedair in order to ensure similar fluidization rates) were used asgasifyingfluidizing agent The equivalence ratio (ER ratio of theactual fuel-to-oxygen ratio to the stoichiometric fuel-to-oxygenratio) varied from 12 to 32 while the steam to daf (dry andash-free) biomass ratio (SB) varied from 027 to 071 kg kg1 inboth cases The produced mixture of steam and tar was condensedin two ice-cooled condensers Thewater content and the qualitativecomposition of the mixture were determined using the afore-mentioned equipment The composition of the dry product gas wasanalyzed on-line with a micro gas chromatograph (Agilent 3000)The ultimate analyses of the solid by-products resulting from thegasification processes were determined with a Leco TruSpec Microelemental analyzer and their higher heating values were calculatedaccording to Dulong formula [HHV (kJ kg1) frac14 339 C thorn 1430$(He O8) thorn 105 S]
Fig 2 Heat demand for the thermal drying of sewage sludge (SS) as a function of theinitial and final moisture contents
3 Results and discussion
This section includes the results of the energetic assessment ofthe individual stages forming part of the processes shown in Fig 1sewage sludge drying sewage sludge gasification sewage sludgepyrolysis and char gasification An overview of the total energy
requirement for the two-stage and three-stage processes is alsoincluded at the end of the section
31 Sewage sludge thermal drying
Prior to the thermo-chemical treatment of sewage sludge bymeans of pyrolysis or gasification sewage sludge thermal dryingallows reduction of water content in the waste Thermal drying ofsewage sludge is not a waste elimination method but waste vol-ume is considerably reduced and handling of the dry biosolids iseasier
The heat needed for the sewage sludge thermal drying can becalculated as follows
Qdrying frac14mdried SS$CpSS thornmH2OSS$CpH2OethlTHORN
$DT
thornmH2Oevap$DHvapH2O (1)
where
- Qdrying is the heat needed for sewage sludge thermal drying(MJ kg1dried SS)
- mdried SS is the mass of dried sewage sludge (1 kg as calculationbasis)
- mH2OSS is the mass of water present in the sewage sludge beforethe thermal drying (kg kg1dried SS)
- DT is the difference between the temperature of the sewagesludge at the beginning and at the end of the drying process(from 25 to 100 C)
- CpSS is the specific heat capacity of the dried sewage sludge Thisvalue was experimentally obtained at 25 C(115$103 MJ kg1 K1) and has been considered constant withtemperature for the calculations The variation of CpSS withtemperature could not be obtained in the upper range of tem-perature because of the sewage sludge thermal decompositionobserved during the measurement
- CpH2O(l) is the commonly used specific heat capacity for theliquid water (418$103 MJ kg1 K1) CpH2O(l) is virtually con-stant in the temperature range considered (25e100 C) onlyvarying from 418$103 to 422$103 MJ kg1 K1 [33]
- mH2Oevap is the mass of water evaporated during the sewagesludge thermal drying (kg kg1dried SS)
- DHvapH2O is the enthalpy of vaporization of water at the exittemperature (226 MJ kg1H2O at 100 C) [33]
Fig 2 shows the evolution of the heat needed for sewage sludgedrying as a function of the initial and final moisture contents basedon calculations performed with equation (1) For instance almost8 MJ kg1dried SS are required for reducing the water content from77 wt to 65 wt which represent the actual data of the waste-water treatment plant in which the used sewage sludge was
iii3
N Gil-Lalaguna et al Energy 76 (2014) 652e662 655
generated However the heat required for the sewage sludgethermal drying could be reduced by half if the initial moisturecontent is reduced from 77 to 65wt by improving the efficiency ofthe prior mechanical dewatering of sewage sludge
32 Sewage sludge pyrolysis
If negligible heat losses are considered in the reactor (adiabaticreactor) the heat of reaction for the pyrolysis of sewage sludge canbe calculated from the enthalpies of the streams entering andexiting the reactor as follows
Q frac14 DHout DHin (2)
where Q is the heat of pyrolysis reaction (MJ kg1SS ) and DHin andDHout represent the enthalpies of the streams entering and exitingthe reactor respectively According to equation (2) Q lt 0 corre-sponds to an exothermic process while Q gt 0 refers to an endo-thermic process
The total enthalpy of each stream (DH) can be calculated fromequation (3)
DH frac14Xi
mi$ethDHof i thorn
ZTTref
CpiethTTHORN$dTTHORN (3)
where
- mi is the mass flow rate of each compound (kg kg1SS ) 1 kg ofsewage sludgewas used as calculation basis Themass flow ratesof the products have been calculated according to the experi-mental yields obtained in the pyrolysis process (Table 2)
- Tref is the reference temperature (298 K) and T (K) is the tem-perature of each stream The inlet streams (sewage sludge andnitrogen) were at ambient temperature (298 K) the same as theoutlet stream of gases and vapors which was cooled down toambient temperature in order to take advantage of their
Table 2Yields and properties of the products of the sewage sludge fast pyrolysis
Yield (wt) Composition HHV (MJ kg1)
Char519 plusmn 07 e 52 plusmn 02
Non-condensable gas (N2-free)101 plusmn 09 ( mass fraction)
CO2 743 plusmn 09CO 132 plusmn 01H2 17 plusmn 01CH4 38 plusmn 01C2H6 14 plusmn 02C2H4 14 plusmn 01H2S 43 plusmn 09
80 plusmn 03
Light organic phase (LOP)22 plusmn 02 Elemental analysis (wt wet basis)
C859 H118 N18 S02100 wt of organic compounds
4310 plusmn 004
Heavy organic phase (HOP)94 plusmn 02 Elemental analysis (wt wet basis)
C695 H90 N94 S12Water 64 plusmn 03 wtOrganics 936 plusmn 03 wt
32 plusmn 2
Aqueous phase (AP)208 plusmn 02 Elemental analysis (wt wet basis)
C112 H105 N65 S04Water 738 plusmn 04 wtOrganics 262 plusmn 04 wt
57 plusmn 03
Experimental uncertainty is expressed as mean plusmn deviation standard (two replicates we
iii4
sensible and latent heats The solid product (char) was supposedto leave the reactor at the pyrolysis temperature (803 K)
- DH
f i is the standard enthalpy of formation (MJ kg1) of eachcompound at the reference temperature (298 K) TheDH
f data ofthe gases involved in the process can be easily found in theliterature [33] The DH
f data corresponding to the solid mate-rials (sewage sludge and char) and to the liquid phases (LOPHOP and AP) have been calculated from their ultimate analysesand heating values according to the following equation
X
DHo
f i frac14j
mj $DHof j thorn HHV (4)
where lsquojrsquo represents each product derived from the completecombustion of the material (CO2 H2O SO2 and NO) mj is themass of each combustion gas produced per kilogram of materialDH
f j is the standard enthalpy of formation of each combustiongas and HHV is the higher heating value of the solid material orliquid phase (Table 2) This way of calculating the DH
f does notinclude the DH
f corresponding to the ash content in the solidsbut this is not necessary for the calculations because ash isconsidered an inert material during the process and thecontribution of its DH
f is simplified in the energy balance TheDH
f of the sewage sludge was found to be 328 MJ kg1SS (notethat this value does not include the DH
f of the ash content) TheDH
f of the pyrolysis products are summarized in Table 2- Cpi (T) is the specific heat capacity of each compound as afunction of the temperature (MJ kg1 K1) Only the tempera-ture of the solid product (803 K) is different from the referencevalue (298 K) thus only the Cp (T) of char contributes to theenergy balance However the Cp (T) function could not be ob-tained up to the pyrolysis temperature (803 K) because ofoperational limitations of the thermobalance used Thereforethe Cp of char has been considered constant with temperaturefor the calculations using an experimental value obtained at anintermediate temperature (121$103 MJ kg1 K1 at 573 K)
According to this procedure the heat of pyrolysis reaction(including the cooling and condensation of the vapors) was found
DH
f (MJ kg1) Cp (T) (kJ K1 kg1) DHvap (MJ kg1)
118 082 (25 C)121 (300 C)
e
739 118 (25 C)156 (530 C)
e
174 185 (liquid)307 (530 C)
018
349 213 (liquid)236 (530 C)
055
1244 359 (liquid)212 (530 C)
177
re performed)
N Gil-Lalaguna et al Energy 76 (2014) 652e662656
to be around 070 MJ kg1SS This indicates that in the absence ofsignificant heat losses and if the heat released from the cooling andcondensation of the gases and vapors could be efficiently usedsewage sludge pyrolysis could be an autothermal process
The energy demand corresponding only to the thermal decom-position of sewage sludge without including the energy recoveryfromgasesandvapors has alsobeenapproximatelycalculated In thiscase gases and vapors were supposed to leave the reactor at thepyrolysis temperature that is in gas phase The following assump-tions have been considered for performing the energy balance
- The composition of each liquid phasewas simplified consideringonly its water content and one representative organic com-pound cholest-4-ene for the LOP 3-methyl-phenol for the HOPand acetic acid for the AP These were some of the compoundsdetected by GCeMS with the largest chromatographic area Themass of the representative organic compound in each phasewasequated to the whole organic fraction in the phase Thisassumption slightly affects the actual values of Cp and DHvap ofthe liquid phases but this should not result in misleading con-clusions since the properties of the main organic compoundspresent in each phase are similar to each other
- Although the aforementioned organic compounds were in gasphase in the outlet stream Cp of these compounds in liquidphase are also required as well as their enthalpies of vapor-ization since the temperature range in the integral equation (3)involves a phase change for the produced vapors (Tref frac14 298 KT frac14 803 K)
- The equation of Harrison and Seaton [34] was used for calcu-lating the Cp of the representative organic compounds in liquidphase (considered constant with temperature) while the Cpdata of the compounds in gas phase were found in the literatureas a function of temperature The global Cp of each phase (bothin liquid and gas phases) can be estimated as aweighted averageof the specific heat capacities of water (or steam) and of therepresentative organic compound The results are presented inTable 2
- In the same way the enthalpy of vaporization of the liquidphases (DHvap) was estimated as a weighted average of the en-thalpies of vaporization of the water and of the representativeorganic compound of each phase at their boiling temperaturesThese results are also presented in Table 2
According to this procedure the energy demand for the thermaldecomposition of sewage sludge was around 015 MJ kg1SS Thisvalue is lower than the decomposition heats found in the literaturefor other types of biomass For example a decomposition heat of03 MJ kg1 has been reported for pyrolysis of crop residues [29]The higher ash content in sewage sludge which is not decomposedduring the process can explain this difference The variation in thewater content of sewage sludge will also affect the energy demandfor its thermal decomposition
33 Air-steam gasification of both sewage sludge and char
In addition to the energetic assessment based on experimentaldata from sewage sludge gasification and char gasification atheoretical study based on equilibrium data is also presented in thissection in order to further study the gasification stages and findtheir thermodynamic restrictions Experimental and equilibriumdata are compared
331 Energetic assessment according to experimental resultsGasification experiments were performed under allothermal
conditions since the gasifier required external heat to maintain the
gasification temperature If negligible heat losses are considered inthe gasifier the heat of reaction for the air-steam gasification ofboth raw materials sewage sludge and char can be calculated ac-cording to equation (2) The total DH of each streamwas calculatedas shown in equation (3) considering the following data
- 1 kg of raw material (sewage sludge or char) has been used ascalculation basis in both gasification processes The amount ofgasifying agent varied depending on the ER and SB defined ineach case Tables 3 and 4 present the mass flow rates of theproducts obtained from the experiments of sewage sludgegasification and char gasification under different operatingconditions respectively [3132]
- The raw material (sewage sludge or char) and the air streamwere at ambient temperature (298 K) at the gasifier inlet whilesteam was generated and fed at 448 K All the products (gassolid steam and tar) left the gasifier at the gasification tem-perature (1043e1123 K)
- The collected amount of tar was simplified to an equimolarmixture of benzene naphthalene and pyridine since these weresome of the main compounds detected in the tar mixtures byGCeMS [31] As tar yield was much lower than the yields ofother products its contribution to the energy balance is also lessimportant and this simplification should not result inmisleading calculations
- The DH
f and Cp (T) of the gases and vapors involved in theprocess (N2 O2 H2 CO CO2 CH4 C2H2 C2H4 C2H6 H2S steambenzene naphthalene and pyridine) were taken from the liter-ature [33] The DH
f of the solid by-products were calculatedaccording to equation (4)
- The Cp (T) of the solid by-products were approximated to that ofsewage sludge combustion ash since these solids were mainlycomposed of ash (gt93 wt in most cases) [3132] The Cp (T) ofthe sewage sludge ash was experimentally measured by DSCbut because of operational limitations of the thermobalanceused the variation of Cp with temperature could not be ob-tained up to 1043e1123 K which is the upper limit in the in-tegral equation (3) Thus the Cp of the solid by-products wereconsidered constant with temperature and an experimentalvalue measured at an intermediate temperature(107$103 MJ kg1 K1 at 773 K) was used for calculations
The experimental heats of reaction for the gasification ofsewage sludge and char under different operating conditions aredepicted in Fig 3a and b Despite the lower organic content in thechar than in the sewage sludge (Table 1) the external energydemand for gasifying 1 kg of char was higher than that for gasi-fying 1 kg of sewage sludge For instance the heat of reaction forsewage sludge gasification with ER frac14 17 and SB frac14 071 was064 MJ kg1 at 850 C and 017 MJ kg1 at 770 C while it reached100 MJ kg1 and 078 MJ kg1 for char gasification at 850 and770 C respectively This behavior could be related to theobserved changes in the organic structure of sewage sludge aftercarrying out the pyrolysis process The fraction of volatile matterin the sewage sludge was higher than in the char while thefraction of fixed carbon was higher in the char (Table 1) Thismeans that combustion reactions in gas phase which usuallyshow less diffusional resistance than the solidegas reactionsinvolve vaporized hydrocarbons during the sewage sludge gasifi-cation However in the case of char gasification the main com-bustion reactions in gas phase involve gases such as H2 or CO(produced from the fixed carbon) whose calorific value is lowerthan that of hydrocarbons As a consequence char gasificationwasan endothermic process under most of the experimental condi-tions used (Fig 3b) while sewage sludge gasification was an
iii5
Table 3Experimental results from the gasification of sewage sludge [31]
Temperature 850 770 850 770 850 770 850 770 810a
ER () 17 17 12 12 32 32 23 23 19a
O2 in enriched-air (vol) 21 21 21 21 33 33 27 27 23a
SB (mass ratio) 071 071 052 052 039 039 027 027 052a
Solid product (g kg1SS ) 368 401 401 407 356 392 384 400 382 plusmn 1Tar (g kg1SS ) 25 46 23 49 16 25 14 47 17 plusmn 2H2O (g kg1SS ) 451 515 356 391 352 451 270 336 414 plusmn 8CO2 (g kg1SS ) 439 385 304 332 534 524 401 403 418 plusmn 4CO (g kg1SS ) 142 75 156 99 191 97 226 100 138 plusmn 1H2 (g kg1SS ) 282 173 275 199 211 109 235 126 203 plusmn 02CH4 (g kg1SS ) 291 265 319 321 246 221 267 251 277 plusmn 08C2H6 (g kg1SS ) 40 49 32 42 21 19 24 43 29 plusmn 04C2H4 (g kg1SS ) 22 23 18 25 19 20 19 21 22 plusmn 2C2H2 (g kg1SS ) 09 09 04 09 11 11 10 10 10 plusmn 01H2S (g kg1SS ) 87 61 62 55 88 71 74 49 71 plusmn 04N2 (g kg1SS ) 734 677 533 502 742 747 711 690 703 plusmn 6hgas (m3
STP kg1SS ) 131 106 123 109 131 112 128 104 118 plusmn 001LHVgas (MJ m3STP) 59 53 71 69 52 41 60 49 56 plusmn 01
a Three replicates were performed at the center point of the experimental design Mean plusmn standard deviation of these replicates is shown
Table 4Experimental results from the gasification of char [32]
Temperature 850 770 850 770 850 770 850 770 810a
ER () 17 17 12 12 32 32 23 23 19a
O2 in enriched-air (vol) 27 27 21 21 40 40 33 33 29a
SB (mass ratio) 071 071 052 052 039 039 027 027 052a
Solid product (g kg1char) 757 785 750 785 731 771 752 813 775 plusmn 2Tar (g kg1char) 2 12 2 5 2 9 5 7 5 plusmn 1H2O (g kg1char) 119 160 78 99 88 99 59 69 112 plusmn 11CO2 (g kg1char) 195 206 154 164 232 254 177 190 198 plusmn 5CO (g kg1char) 128 65 122 64 148 77 144 76 91 plusmn 3H2 (g kg1char) 137 101 120 89 100 74 96 72 103 plusmn 01CH4 (g kg1char) 28 28 26 26 22 22 22 24 29 plusmn 01C2H4 (g kg1char) 007 007 006 008 006 008 005 007 007 plusmn 001H2S (g kg1char) 197 076 099 044 138 056 090 036 068 plusmn 003N2 (g kg1char) 205 197 211 204 212 201 213 198 207 plusmn 8hgas (m3
STP kg1char) 052 043 048 040 052 044 049 040 046 plusmn 001LHVgas (MJ m3STP) 60 47 59 47 54 41 56 44 51 plusmn 01
a Three replicates were performed at the center point of the experimental design Mean plusmn standard deviation of these replicates is shown
N Gil-Lalaguna et al Energy 76 (2014) 652e662 657
exothermic process when simultaneously working with ER gt 19and SB lt 052 (Fig 3a) The heat of reaction (based on experi-mental data) ranged from 261 to thorn129 MJ kg1SS for sewagesludge gasification and from 023 to thorn120 MJ kg1char for chargasification Therefore the temperature and the gasification me-dium play a larger role in the energy balance of sewage sludgegasification
The energy demand for carrying out an endothermic gasifi-cation process may be obtained from the product gas eitherfrom its thermal energy (for example by using the product gas topreheat the inlet air stream in a heat exchanger) or from thecombustion of part of the gas The gasification efficiency can bedefined as the fraction of the energy initially contained in theraw material that could be recovered from the product gas aftercarrying out the gasification process (5)
Efficiency ethTHORN frac14Energy in gas Qgasification Qsteam
LHVraw material$100 (5)
where
- ldquoEnergy in gasrdquo is the energy that could be recovered from thethermal and calorific values of the gasification product gas(MJ kg1raw material)(equation 6) A heat exchange efficiency of 70has been considered when taking advantage of the sensible andlatent heats of the gas stream [35]
iii6
Energy in gasfrac14hgas$LHVgasthorn07$64Xmigas$
ZTCpigas ethTTHORN$dT
75
2
i Tref
3
thorn07$mH2O$
264CpH2OethlTHORN$
Tb H2O Tref
thornDHcond H2OthornZT
TbH2O
CpH2OethvTHORNethTTHORN$dT
375
thorn07$Xitar
mitar$
264CpitarethlTHORN$
Tb i tar Tref
thorn $HconditarthornZT
Tb i tar
CpitarethvTHORNethTTHORN$dT
375
(6)
being hgas the dry gas yield (m3STP kg1raw material Tables 3 and 4)
LHVgas the lower heating value of the dry gas (MJ m3STP Tables 3and 4) migas the mass flow rate of the non-condensable gases(kg kg1raw material Tables 3 and 4) Tref frac14 298 K T the gasificationtemperature (K) Cpigas (T) the specific heat capacity of the non-condensable gases as a function of temperature (MJ kg1 K1)
Fig 3 Heats of reaction for sewage sludge (SS) gasification (a c) and char gasification (b d) based on both experimental and equilibrium data
N Gil-Lalaguna et al Energy 76 (2014) 652e662658
[33] mH2O the mass flow rate of produced water(kg kg1raw material Tables 3 and 4) CpH2O(l) the specific heat ca-pacity of liquid water (418$103 MJ kg1 K1) TbH2O the boilingpoint of water (373 K) DHcondH2O the enthalpy of condensationof water at its boiling point (226 MJ kg1) CpH2O(v)(T) thespecific heat capacity of steam as a function of temperature [33]mitar the mass flow rates of tar compounds (kg kg1raw material)Cpitar(l) the specific heat capacity of tar compounds in liquidphase (calculated according to [34]) Tbitar the boiling point oftar compounds DHconditar the enthalpy of condensation of tarcompounds at their boiling points (039 034 and 044 MJ kg1
for benzene naphthalene and pyridine respectively) and Cpi-
tar(v)(T) the specific heat capacity of tar compounds in gas phaseas a function of temperature The obtained results of ldquoEnergy ingasrdquo are shown in Table 5
- Qgasification is the heat of reaction for the gasification process(MJ kg1raw material) previously calculated according to equation(2) under different experimental conditions (Fig 3a and b)
- Qsteam (MJ kg1raw material) is the energy demand for heating andevaporating the inlet flow of water from 25 C to 150 C(236 MJ kg1H2O)
- LHVraw material (MJ kg1raw material) is the energy initially containedin the raw material expressed as its lower heating value(Table 1)
The efficiency results for sewage sludge gasification and chargasification are presented in Table 5 Experimental efficiency datavaried from 58 to 87 for sewage sludge gasification and from 23to 64 for char gasification Better efficiency results were obtainedfor the sewage sludge gasification as a consequence of its lowerheat of reaction and higher gas yield (Tables 3 and 4) The gasifi-cation efficiency (based on experimental data) improved at highertemperatures higher ER and lower SB
332 Energetic assessment according to equilibrium dataThe heat of reaction for the air-steam gasification of sewage
sludge and char when reaching the chemical equilibrium is deter-mined in this section The calculation has been carried out analo-gously to Section 331 but in this case the product mass flows werenot experimental data but equilibrium data HSC Chemistryreg 61software was used to determine themass flow rates of the productsat equilibrium conditions This software uses the Gibbs energyminimization method to calculate the amounts of products atequilibrium in isothermal and isobaric conditions Therefore thereaction system (temperature pressure feed of gasifying agentamounts of C H O S and N that form part of the raw material andspecies expected to be part of the products) must be specified forthe calculations
The main compounds found in the product gas at equilibriumconditions were H2 CO CO2 CH4 H2S NH3 N2 and steam Neithertar nor light hydrocarbons except CH4 were formed at equilibriumconditions In addition to gas production a small fraction of theinitial carbon contained in the rawmaterial (sewage sludge or char)remained in the solid by-product under some of the simulatedconditions
The heats of reaction at chemical equilibrium under the sameoperating conditions used in the laboratory are depicted in Fig 3cand d as a comparison to the experimental data (Fig 3a and b) Ascan be observed reaching the chemical equilibrium in both gasi-fication processes entails additional energy consumption Thereason may be the predominance of endothermic reactions duringthe gasification equilibrium such as steam reforming (C thorn H2O 4
CO thorn H2 DH298K frac14 1314 kJ mol1) dry reforming (CH4 thorn CO2 4
2CO thorn 2H2 DH298K frac14 2468 kJ mol1) or the Boudouard reaction(C thorn CO2 4 2CO DH298K frac14 1723 kJ mol1) against the exothermicequilibrium reactions such as the water-gas shift reaction(CO thorn H2O4 CO2 thorn H2 DH298K frac14 409 kJ mol1) These reactions
iii7
Table 5Energy recovery from the product gas and efficiency of sewage sludge gasification and char gasification according to experimental and equilibrium data
Temperature 850 770 850 770 850 770 850 770 810ER () 17 17 12 12 32 32 23 23 19SB (mass ratio) 071 071 052 052 039 039 027 027 052Sewage sludge gasification (experimental results)Energy in gas (MJ kg1SS ) 1023 791 963 841 910 677 956 693 872 plusmn 012Gasification efficiency () 74 58 65 61 87 75 82 65 71 plusmn 2Sewage sludge gasification (equilibrium results)Energy in gas (MJ kg1SS ) 1437 1411 1495 1472 1134 1109 1286 1264 1370Gasification efficiency () 90 91 91 92 94 94 94 94 92Char gasification (experimental results)Energy in gas (MJ kg1char) 391 277 349 244 352 244 334 227 298 plusmn 005Gasification efficiency () 51 32 40 23 64 49 54 33 40 plusmn 2Char gasification (equilibrium results)Energy in gas (MJ kg1char) 619 609 653 642 494 485 562 553 595Gasification efficiency () 78 80 80 81 82 83 82 84 81
N Gil-Lalaguna et al Energy 76 (2014) 652e662 659
occur to a greater extent at equilibrium which shows the ther-modynamic limit of the process
As can be seen in Fig 3c and d the gasification of sewage sludgeat equilibrium conditions only resulted in an exothermic processwhen ER was increased to 32 while char gasification at equilib-rium conditions was an endothermic process in all the simulatedcases However the gas heating value and the gas yield calculatedfor equilibrium conditions were higher than those obtainedexperimentally [3132] so more energy could be recovered fromthe equilibrium product gas (Table 5) This latter difference out-weighs the difference observed in the experimental and equilib-rium data of heat of reaction (Fig 3) so the gasification efficiency isimproved at equilibrium conditions 90e94 for sewage sludgegasification and 78e84 for char gasification (Table 5)
As an extension of the theoretical study Fig 4 shows the evo-lution of the heat of reaction for air-steam gasification of bothsewage sludge and char as a function of the feed of oxygen andsteam which are represented by ER and SC (steam to carbonmolarratio) respectively Two different gasification temperatures (800and 850 C) have been used for the calculations Operating tem-peratures above 800 C are usually preferred in gasification pro-cesses in order to achieve high carbon conversion and low tarcontent in the product gas [36] The heat of reaction for bothgasification processes at equilibrium conditions decreases with theavailability of oxygen (higher ER enhances the combustion re-actions) andor with the reduced presence of steam (lower SCrestricts the endothermic steam reforming reactions) For examplefor ER frac14 20 and SC frac14 0e1 an external heat transfer of201e243 MJ kg1SS and 136e186 MJ kg1char would be required inorder tomaintain a temperature of 850 C during the gasification ofsewage sludge and char at equilibrium conditions respectively Inthe lower range of ER (up to 25) the energy demand for gasifying1 kg of char is lower than that required for gasifying 1 kg of sewagesludge but this trend is reversed in the upper range of ER since theenergy released from the in situ combustion of sewage sludge be-comes more important
If the gasifier operates at autothermal conditions instead ofbeing heated by external heat transfer the gasification temperatureis the output variable from balancing out the enthalpies of thestreams entering and exiting the gasifier (DHin frac14 DHout assumingnegligible heat losses) The equilibrium temperature has beencalculated under different gasification mediums following an iter-ativemethodDHout depends on themass flow rates of the products(equation 3) and these in turn depend on the gasification tem-perature (temperature has to be specified in the HSC Chemistrysoftware to calculate the amounts of products at equilibrium) Fig 5shows the evolution of the equilibrium temperature as a function ofER and SC for the air-steam gasification of both sewage sludge and
iii8
char Obviously the equilibrium temperature is increased with ERand decreased with SC Furthermore the required ER tomaintain aspecific reaction temperature is higher in char gasification than insewage sludge gasification For instance an ER of 33 would berequired for the autothermal operation of sewage sludge gasifica-tion at 800 C and SC frac14 05 under equilibrium conditions whilethis value reaches 45 in the case of char gasification The higherthe ER the greater the production of CO2 through combustion re-actions The presence of CO2 in the gasification gas is undesirablesince it implies both a dilution effect of the gas heating value and areduction in the formation of CO (production and consumption ofCO and CO2 are connected by reactions such as the water-gas shiftor the Boudouard reaction) In addition to the gas calorific valuethe H2CO ratio in the product gas is an important parameter forusing this gas as a feedstock in the synthesis of chemicals such asmethanol or Fischer Tropsch fuels Values of this ratio close to 2 areusually required in these processes [37] For the aforementionedexample (ER of 33 for sewage sludge gasification and 45 for chargasification to maintain 800 C with SC frac14 05) 44 of the initialcarbon contained in the sewage sludge produces CO2 while thisvalue reaches 52 in the case of char gasification Both the heatingvalue and the H2CO ratio in the product gas of sewage sludgegasification (LHVgas frac14 427 MJ m3STP H2CO frac14 147) are higher thanthose obtained from char gasification (LHVgas frac14 305 MJ m3STP H2CO frac14 089)
34 Energetic assessment of the whole processes
This last section presents an overall energetic assessment ofthe two thermo-chemical processes proposed in Fig 1 for sewagesludge treatment (i) sewage sludge drying thorn sewage sludgegasification (two-stage process) and (ii) sewage sludgedrying thorn sewage sludge pyrolysis thorn char gasification (three-stageprocess) The total energy demand for the whole processes is thesum of the net heats required or released in the involved stages(positive term for endothermic processes and negative value forexothermic processes) Experimental data resulting from thepyrolysis and gasification stages have been used for thecalculations
- Sewage sludge drying The water content in sewage sludge isassumed to be reduced from 65 wt (typical moisture contentbefore thermal drying) to 65 wt during the thermal dryingFor this case Qdrying is around 4 MJ kg1dried SS (Fig 2)
- Sewage sludge pyrolysis The energy contained in the producedgases and vapors could be recovered to be used in the thermaldecomposition of sewage sludge itself and in the prior thermaldrying This energy was calculated analogously to the
Fig 4 Heats of reaction for sewage sludge (SS) gasification (a b) and char gasification (c d) at 800 C and 850 C according to equilibrium data
N Gil-Lalaguna et al Energy 76 (2014) 652e662660
gasification gas energy according to equation (6) An energyrecovery of 132 MJ kg1SS was obtained which turnsinto 117 MJ kg1SS if the thermal decomposition heat is sub-tracted (thorn015 MJ kg1SS ) The use of the calorific value of theorganic liquid product (43 MJ kg1LOP and 32 MJ kg1HOP) has notbeen included in the energy balance as some importantproperties such as its poor stability or its high nitrogen contentmust be improved facing toward its use as fuel [7]
- Sewage sludge gasificationchar gasification The net heats ofthe gasification stages correspond to the numerator of theequation (5) As the same calculation basis is required for thecomparison of the two-stage and three-stage processes (1 kg ofdried sewage sludge) data corresponding to the gasification ofchar (MJ kg1char) must be turned into MJ kg1SS by means of thechar yield obtained during the pyrolysis of sewage sludge(0519 kgchar kg
1SS )
Fig 5 Equilibrium temperature as a function of the equivalence ratio (ER) and the steam t
Fig 6 shows the total energy requirement for the whole pro-cesses considering the different experimental conditions used inthe gasification stages The total energy demand rangedbetween 283 and 621 MJ kg1SS for the two-stage process(exothermic process) and betweenthorn117 and thorn224 MJ kg1SS for thethree-stage process (endothermic process) Thus if the energycontained in the product gas of sewage sludge gasification could beefficiently used it would be enough to cover the energy demand forboth the sewage sludge thermal drying and the gasification processitself It should be noted that the flow of water required for the air-steam gasification of sewage sludge could directly come from theown moisture content in the sludge thus the energy required forthermal drying would be reduced and the total energy balancewould be evenmore favorable (between370 and669MJ kg1SS )On the other hand the energy balance shows that the three-stagetreatment is globally an endothermic process (note that the use
o carbon molar ratio (SC) during sewage sludge (SS) gasification and char gasification
iii9
Fig 6 Total energy demand for the two-stage and three-stage processes for sewage sludge (SS) treatment (experimental data)
N Gil-Lalaguna et al Energy 76 (2014) 652e662 661
of the calorific value of the pyrolysis liquid is not considered) sothat an additional energy input would be needed to carry out thistreatment However assuming a direct and efficient use of thecalorific value of the organic pyrolysis liquid (392 MJ kg1SS ) afavorable energetic assessment of the three-stage process is alsoobtained with a total energy demand ranging from 275to 168 MJ kg1SS (exothermic process) Therefore the use of thecalorific value of the produced pyrolysis liquid is a key issue forreaching an autothermal three-stage process that does not requireexternal heat to take place
Regarding the influence of the gasification operating conditionson the total energy demand of the whole processes the energybalance was more favorable at the highest gasification temperature(850 C) the highest ER (32) and a moderate SB (039)
4 Conclusions
This paper presents an energetic assessment of two potentialthermo-chemical treatments for sewage sludge (i) sewage sludgethermal drying thorn air-steam gasification of sewage sludge (two-stage process) and (ii) sewage sludge thermal drying thorn pyrolysis ofsewage sludgethorn air-steam gasification of the char derived from thepyrolysis (three-stage process) The sewage sludge thermal dryingcan drastically reduce the waste volume which facilitates handlingof the biosolids but it involves high energy consumption Forexample 4 MJ kg1dried SS are required for reducing water contentfrom 65 wt to 65 wt Regarding the pyrolysis stage energeticcalculations based on experimental yields showed that the energyneeded for thermal decomposition of sewage sludge(thorn015MJ kg1SS at 530 C) could be covered by the energy containedin the product stream of gases and vapors An energy outputof117MJ kg1SS could be recovered from the calorific value and thethermal energy of the product gas (heat exchange efficiency of 70)after covering the energy demand in the pyrolysis reaction Despitethe lower organic content in the char (241 wt) than in the sewagesludge (545 wt) higher external energy demand was found forgasifying 1 kg of char than for gasifying 1 kg of sewage sludge(based on experimental yields) This means that less energy isreleased from the in-situ combustion reactions during char gasifi-cation Depending on the operating conditions sewage sludgegasificationwas an exothermic or endothermic process and its heatof reaction varied from 261 MJ kg1SS (T frac14 770 C ER frac14 32 SB frac14 039) to thorn129 MJ kg1SS (T frac14 850 C ER frac14 12 SB frac14 052) Chargasification was an endothermic process in most of the experi-mental conditions and its heat of reaction varied
iii1o
from 023 MJ kg1char (T frac14 770 C ER frac14 32 SB frac14 039)to thorn120 MJ kg1char (T frac14 850 C ER frac14 12 SB frac14 052) A theoreticalstudy performed with equilibrium data (according to Gibbs energyminimization method) showed that both gasification processesrequire more energy to take place at equilibrium conditionsHowever the equilibrium gasification efficiency was higher thanthe experimental results because more energy could be recoveredfrom the equilibrium product gas
In summary the energy balances showed that the energy con-tained in the product gas of sewage sludge gasification is enough tocover the energy demand for both the sewage sludge thermaldrying and the gasification process itself However an additionalenergy input is required to carry out the three-stage process Thisenergy demand could be provided by the calorific value of the bio-oil produced in the pyrolysis stage but some important propertiessuch as its poor stability or its high nitrogen content must beimproved facing toward its use as fuel
Acknowledgments
The authors gratefully acknowledge the financial support pro-vided by the Spanish Ministry of Science and Technology(CTQ2010-20137) and the Spanish Ministry of Education (AP2009-3446)
References
[1] Werther J Ogada T Sewage sludge combustion Prog Energy Combust Sci19992555e116
[2] Council Directive 91271EEC of 21 May 1991 concerning urban waste watertreatment
[3] Fytili D Zabaniotou A Utilization of sewage sludge in EU application of oldand new methods e a review Renew Sustain Energy Rev 200812116e40
[4] Council Directive 86278EEC of 12 June 1986 on the protection of the envi-ronment and in particular of the soil when sewage sludge is used inagriculture
[5] Furness DT Hoggett LA Judd SJ Thermochemical treatment of sewage sludgeJ Chart Inst Water Environ Manage 20001457e65
[6] Rulkens W Sewage sludge as a biomass resource for the production of energyoverview and assessment of the various options Energy Fuels 2008229e15
[7] Fonts I Gea G Azuara M Abrego J Arauzo J Sewage sludge pyrolysis for liquidproduction a review Renew Sustain Energy Rev 2012162781e805
[8] Kim Y Parker W A technical and economic evaluation of the pyrolysis ofsewage sludge for the production of bio-oil Bioresour Technol 2008991409e16
[9] Pokorna E Postelmans N Jenicek P Schreurs S Carleer R Yperman J Study ofbio-oils and solids from flash pyrolysis of sewage sludges Fuel 2009881344e50
[10] Chaudhari ST Dalai AK Bakhshi NN Production of hydrogen andor syngas(H2thornCO) via steam gasification of biomass-derived chars Energy Fuels2003171062e7
N Gil-Lalaguna et al Energy 76 (2014) 652e662662
[11] Haykiri-Acma H Yaman S Kucukbayrak S Gasification of biomass chars insteam-nitrogen mixture Energy Convers Manage 2006471004e13
[12] He P Luo S Cheng G Xiao B Cai L Wang J Gasification of biomass char withair-steam in a cyclone furnace Renew Energy 201237398e402
[13] Bacaicoa PG Bilbao R Uson C Sewage sludge gasification first studies InProceedings of the Second Biomass Conference of the Americas EnergyEnvironment and Agricultural Industry 1995 p 685e94
[14] Adegoroye A Paterson N Li X Morgan T Herod AA Dugwell DR et al Thecharacterisation of tars produced during the gasification of sewage sludge in aspouted bed reactor Fuel 2004831949e60
[15] Manya JJ Sanchez JL Gonzalo A Arauzo J Air gasification of dried sewagesludge in a fluidized bed effect of the operating conditions and in-bed use ofalumina Energy Fuels 200519629e36
[16] Aznar M Manya JJ Garciacutea G Influence of freeboard temperature fluidizationvelocity and particle size on tar production and composition during the airgasification of sewage sludge Energy Fuels 2008222840e50
[17] Tae-Young M Bo-Sung K Joo-Sik K Production of a producer gas with highheating values and less tar from dried sewage sludge through air gasificationusing a two-stage gasifier and activated carbon Energy Fuels 2009233268e76
[18] Dogru M Midilli A Howarth CR Gasification of sewage sludge using athroated downdraft gasifier and uncertainty analysis Fuel Process Technol20027555e82
[19] Petersen I Werther J Experimental investigation and modelling of gasifica-tion of sewage sludge in the circulating fluidized bed Chem Eng Process200544717e36
[20] Judex JW Gaiffi M Burgbacher HC Gasification of dried sewage sludge statusof the demonstration and the pilot plant Waste Manage 201232719e23
[21] Nipattummakul N Ahmed II Kerdsuwan S Gupta AK High temperaturesteam gasification of wastewater sludge Appl Energy 2010873729e34
[22] Xu ZR Zhu W Li M Influence of moisture content on the direct gasification ofdewatered sludge via supercritical water Int J Hydrogen Energy 2012376527e35
[23] McKendry P Energy production from biomass (part 3) gasification technol-ogies Bioresour Technol 20028355e63
[24] Franco C Pinto F Gulyurtlu I Cabrita I The study of reactions influencing thebiomass steam gasification process Fuel 200382835e42
[25] Hosseini M Dincer I Rosen MA Steam and air fed biomass gasificationcomparisons based on energy and exergy Int J Hydrogen Energy 20123716446e52
[26] Karamarkovic R Karamarkovic V Energy and exergy analysis of biomassgasification at different temperatures Energy 201035537e49
[27] Ptasinski KJ Prins MJ Pierik A Exergetic evaluation of biomass gasificationEnergy 200732568e74
[28] Zhang Y Li B Li H Zhang B Exergy analysis of biomass utilization via steamgasification and partial oxidation Thermochim Acta 201253821e8
[29] Manganaro J Chen B Adeosun J Lakhapatri S Favetta D Lawal A Conversionof residual biomass into liquid transportation fuel an energy analysis EnergyFuels 2011252711e20
[30] Gil-Lalaguna N Fonts I Gea G Murillo MB Lazaro L Reduction of watercontent in sewage sludge pyrolysis liquid by selective on-line condensation ofthe vapors Energy Fuels 2010246555e64
[31] Gil-Lalaguna N Sanchez JL Murillo MB Rodriacuteguez E Gea G Air-steam gasi-fication of sewage sludge in a fluidized bed Influence of some operatingconditions Chem Eng J 2014248373e82
[32] Gil-Lalaguna N Sanchez JL Murillo MB Ruiz V Gea G Air-steam gasificationof char derived from sewage sludge pyrolysis Comparison with the gasifi-cation of sewage sludge Fuel 2014129147e55
[33] Perry RH Green DW Perrys chemical engineers handbook 7th ed NewYork McGraw-Hill 1999
[34] Harrison BK Seaton WH Solution to missing group-problem for estimation ofideal-gas heat-capacities Ind Eng Chem Res 1988271536e40
[35] Ccedilengel YA Heat and mass transfer a practical approach 3rd ed MexicoMcGraw-Hill ScienceEngineeringMath 2005
[36] Devi L Ptasinski KJ Janssen FJJG A review of the primary measures for tarelimination in biomass gasification processes Biomass Bioenergy 200324125e40
[37] Wender I Reactions of synthesis gas Fuel Process Technol 199648189e297
iii11
Fuel 141 (2015) 99ndash108
Reprinted with permission from Elsevier
Contents lists available at ScienceDirect
Fuel
journal homepage wwwelsevier comlocate fuel
Use of sewage sludge combustion ash and gasification ashfor high-temperature desulphurization of different gas streams
httpdxdoiorg101016jfuel2014100360016-2361 2014 Elsevier Ltd All rights reserved
uArr Corresponding author Tel +34 976762224E-mail address noemigilunizares (N Gil-Lalaguna)
N Gil-Lalaguna uArr JL Saacutenchez MB Murillo G GeaThermo-chemical Processes Group Aragoacuten Institute of Engineering Research (I3A) Universidad de Zaragoza cMariano Esquillor sn 50018 Zaragoza Spain
h i g h l i g h t s
Hot gas desulphurization (600ndash800 C) by using different sewage sludge ash Better results obtained with the combustion ash than with the gasification ash Different metal content and crystalline phases detected in both solids Negative impact of steam and gasification gas components on H2S removal The total amount of H2S removed from gas was only partially captured in the ash
a r t i c l e i n f o
Article historyReceived 27 May 2014Received in revised form 29 August 2014Accepted 15 October 2014Available online 29 October 2014
KeywordsH2S removalHot gas cleaningSewage sludgeCombustion ashGasification ash
a b s t r a c t
Due to its metal content sewage sludge ash appears as a potential sorbent material for H2S removal athigh temperature The desulphurization ability of the solid by-products of combustion and gasificationof sewage sludge has been evaluated in this work Ash characterization results revealed that metal frac-tion in sewage sludge did not remained completely inert during combustion and gasification processesIron content was lower in the gasification ash and X-ray patterns showed different crystalline phases inthe solids Fe2O3 in the combustion ash and Fe3O4 in the gasification ash These differences resulted in alower sulphur capture capacity of the gasification ash Desulphurization tests were carried out in alab-scale fixed bed reactor operating at 600ndash800 C Different gases containing 5000 ppmv H2S (H2SN2
mixture and synthetic gasification gas) were used The H2S breakthrough curves were negatively affectedby the reducing atmosphere created by the gasification gas and by the presence of steam in the reactionmedium However H2S breakthrough curves alone do not provide enough information to evaluate thesulphur capture capacity of the sorbent materials Ultimate analyses of the spent solid samples showedthat the total amount of H2S removed from the gas was only partially captured in the ash Thermody-namic data pointed to a significant fraction of sulphur forming part of other gases such as SO2 In the bestoperating conditions an outlet gas with less than 100 ppmv H2S was obtained during 300 min thusresulting in a sulphur loading of 63 mg S gash
1 This experimental sulphur content was 39 lower thanthe maximum value predicted by equilibrium simulations
2014 Elsevier Ltd All rights reserved
1 Introduction
Sewage sludge has become an increasingly important residueneeding effective management Due to the organic nature of sew-age sludge thermal processes such as pyrolysis gasification orcombustion have attracted considerable scientific interest as apotential route to its energy valorization [1ndash3] However as occursin most thermo-chemical treatments of wastes and solid fuels var-ious impurities are found in the products of interest deriving from
these processes which significantly limits their final uses One ofthese impurities is H2S present in the gaseous products of boththe pyrolysis and the gasification of sewage sludge due to its initialcontent of sulphur compounds [4] As is well known H2S emissionsto the atmosphere (or SO2 emissions in the case of burning a H2S-containing gas) entail environmental problems related to acid rainIn addition the presence of H2S in the gas leads to operationalproblems such as corrosion in pipes engines or turbines as wellas the deactivation of the common catalysts used for tar crackingand gas reforming after gasification processes [5]
Several low and high temperature processes for H2S removalfrom products or off-gases have been described and developed at
iv1
100 N Gil-Lalaguna et al Fuel 141 (2015) 99ndash108
various stages Wet scrubbing with selected solvents has been awidely used low-temperature process in the chemical processindustry [6] The use of activated carbons for H2S removal at lowtemperature has also been extensively studied [7ndash9] Given thecombination of their unique surface features (high specific surfaceand pore volume) and surface chemistry improved by the additionof functional groups activated carbon-based materials have beenproved to work efficiently as adsorbents of sulphur-containinggases such as H2S SO2 or methyl mercaptans As a result of surfacereactions H2S can be oxidized to either sulphur or SO2 [8] Someactivated carbons of different characteristics have been preparedfrom sewage sludge pyrolysis or carbonization [10ndash13] obtainingin some cases pollutant removal efficiencies comparable to thosecorresponding to commercial activated carbons [13]
On the other hand high temperature desulphurization pro-cesses are advantageous from an energy standpoint as a result ofthe elimination of gas cooling and the associated heat exchangers[14] Different studies concerning the use of metal oxide based sor-bents for hot gas desulphurization can be found in the literature[14ndash20] Zinc manganese copper iron rare earth and calcium sor-bents are among the most promising and most extensively studied[1415] Typically metal oxides are converted to sulphides during asulphur loading stage under reducing hot gas conditions After sul-phidation the spent metal sulphides can be regenerated back tometal oxides by using oxygen steam SO2 or a combination of these[14]
Ash residues derived from thermo-chemical treatment of bio-mass or wastes are known to contain metal oxides in different pro-portions so its use as sorbent materials for H2S removal from hotgases could be an attractive alternative due to its low cost Thisis the case of sewage sludge ash in which this work focuses Anintegrated process could be proposed to remove the H2S producedduring thermo-chemical treatment of sewage sludge by using theown ash resulting from these processes in a downstream cleaningstage The desulphurization capacity of the solid by-productsderived from the combustion and gasification of sewage sludgehas been evaluated in this work Desulphurization tests were car-ried out in a lab-scale fixed bed reactor and the effects of temper-ature type of sewage sludge ash presence of steam and type ofH2S-containing gas were studied thus extending an earlier workperformed by our group [20]
2 Materials and methods
21 Sewage sludge ash
The raw material used to obtain the ash samples was anaerobi-cally digested and thermally dried sewage sludge Sewage sludge
Fig 1 Experimental setup fo
iv2
combustion was performed under air atmosphere in a heatingmuffle furnace at 900 C (heating rate of 20 C min1) during twohours The sewage sludge gasification ash was obtained in a previ-ous gasification study performed in a lab-scale fluidized bed reac-tor at 850 C using a mixture of steam and air as gasifying agent(H2OO2 molar ratio = 1) [4]
Both ash samples were characterized before the desulphuriza-tion tests by various techniques Ultimate analyses were per-formed using a Leco TruSpec Micro elemental analyzer Texturalproperties such as the BET surface area and the average pore sizeand volume (BJH method) were calculated from N2 physisorptionisotherms (BET volumetric method) using a Micromeritics TriStarII 3000 analyzer The N2 adsorptionndashdesorption isotherms wereobtained at 196 C and room temperature respectively overthe whole range of relative pressures The samples were previouslydegasified at 200 C during 8 h in a N2 flow Powder X-ray diffrac-tion (XRD) patterns of the fresh samples were acquired with aD-Max Rigaku diffractometer equipped with a copper anode(voltage of 40 kV and current of 80 mA) The measurements werecompleted in the Braggrsquos angle (2h) range from 5 to 95 using ascanning rate of 003 s1 Phases present in the solid samples weredefined according to the JCPDS-International Centre for DiffractionData 2000 database Lastly metal content in the ash samples wasanalyzed by inductively coupled plasma combined with opticalemission spectroscopy (ICP-OES) using a Thermo Elemental IRISIntrepid ICPndashOES spectrometer The samples were dissolved bymicrowave-assisted acid digestion in a CEM MARS microwavereaction system
22 Experimental setup and operating conditions
Desulphurization tests were performed at atmospheric pressurein a fixed-bed quartz tubular reactor of 12 cm inner diameter and40 cm length A diagram of the installation used for the desulphu-rization tests is shown in Fig 1 The reactor was packed with 1 g ofthe solid material (combustion ash or gasification ash) which wassupported on a fibreglass fleece located 185 cm from the top of thereactor The reactor was electrically heated and a K-type thermo-couple located in the middle of the solid bed was used to measureand control the temperature The flow rate of the H2S-containinggas was adjusted to 50 mLSTP min1 (STP standard conditions oftemperature and pressure at 0 C and 1 atm) by means of a previ-ously calibrated mass flow controller Two different inlet gasescontaining 5000 ppmv H2S were used to evaluate the effect ofthe gas components on the desulphurization process (i) a gas mix-ture only containing H2S and N2 and (ii) a synthetic gasificationgas similar to the dry product gas of sewage sludge gasification[4] Table 1 shows the composition of both gas mixtures On the
r desulphurization tests
Table 1Composition of the gas mixtures (dry basis)
Gas component H2SN2 mixture (vol) Synthetic gasification gas (vol)
CO ndash 100CO2 ndash 150H2 ndash 100CH4 ndash 40C2H6 ndash 02C2H4 ndash 15C2H2 ndash 02H2S 05 05N2 995 586
N Gil-Lalaguna et al Fuel 141 (2015) 99ndash108 101
one hand the H2SN2 mixture allows the assessment of the desul-phurization process without involving any interference from othergases On the other hand the synthetic gasification gas simulatesmore real conditions for H2S removal Steam was also added insome tests in order to evaluate its effect on the desulphurizationability of the solids The required flow of liquid water (0ndash1 g h1)was accurately adjusted by a HPLC pump and was evaporatedbefore entering the reactor Steam concentration in the gas enter-ing the reactor varied from 0 to 30 vol which led to a H2OH2Smass ratio varying from 0 to 45 g H2Og H2S The weight hourlyspace velocity during the experiments ranged between 37 h1
and 47 h1 depending on the steam feed rate The gasndashsolid con-tact time was chosen after preliminary experiments and basedon earlier studies [20]
The reactor containing the ash sample was flushed with a N2
flow while heating The experiment started when the desired tem-perature was reached in the solid bed Just then the N2 flow wasreplaced by the synthetic gasification gas or by the H2SN2 mixtureThe composition of the outlet gas was continuously analyzed dur-ing the experiments by means of an Agilent Micro-GC 3000 (sam-ple injection every four minutes) The H2S outlet flow(mLSTP min1) was calculated from the gas composition data taking
Table 2Operating conditions in the desulphurization tests
Experiment Origin of sewage sludge ash Synthetic gas mixture
1 Combustion H2SN2
2 Combustion H2SN2
3 Combustion H2SN2
4 Combustion H2SN2
5 6 7 Combustion H2SN2
8 Gasification H2SN2
9 Gasification H2SN2
10 Gasification H2SN2
11 Gasification H2SN2
12 13 14 Gasification H2SN2
15 Blank run H2SN2
16 Blank run H2SN2
17 Blank run H2SN2
18 Blank run H2SN2
19 Blank run H2SN2
20 Combustion Gasification gas21 Combustion Gasification gas22 Combustion Gasification gas23 Combustion Gasification gas
24 Gasification Gasification gas25 Gasification Gasification gas26 Gasification Gasification gas27 Gasification Gasification gas
28 Blank run Gasification gas29 Blank run Gasification gas30 Blank run Gasification gas31 Blank run Gasification gas
N2 of the gas mixture as an internal standard The evolution of theH2S outlet flow with time is depicted in the H2S breakthroughcurves Initially the established time for the experiments wastwo hours but reaction time had to be extended in some casesto 240 or 390 min (Table 2) in order to reach the H2S breakthroughtime
After the desulphurization tests sulphur content in the ashsamples was determined with a Leco TruSpec Micro elemental ana-lyzer Furthermore one of the ash samples was morphologicallyand chemically characterized by scanning electron microscopycombined with energy dispersive X-ray spectroscopy (SEMEDX)and by X-ray photoelectron spectroscopy (XPS) A FEI Inspect F50microscope was used for the SEMEDX analysis External metalcoating was not applied to the solid sample The back-scatteredelectron imaging mode was used in the EDX analysis (acquisitiontime of 1 min) The XPS measurements were perfomed with a Kra-tos AXIS Ultra DLD spectrometer by using monochromatic Al Ka(14866 eV) X-ray source and a chamber pressure of around3108 Pa The quantification of the XPS spectra was carried outwith the help of the CasaXPS software and spectra were deconvo-luted by applying Gaussian-Lorentzian line-shapes with Shirley-type background
Table 2 summarizes the operating conditions for the desulphu-rization tests The operating factors studied were the followingbed temperature (600ndash800 C) H2OH2S mass ratio in the inletgas (0ndash45 gg) type of sewage sludge ash (combustion ash or gas-ification ash) and type of H2S-containing gas (H2SN2 mixture orsynthetic gasification gas) Blank tests (with no bed material) wereperformed at the different temperatures and gas atmospheres inorder to assess any side effect of the experimental setup causedfor example by the reaction of H2S with the hot metal parts (madeof steel) at the reactor inlet and outlet Experimental uncertaintywas evaluated through three replicates performed at intermediatevalues of temperature (700 C) and H2OH2S mass ratio (225 gg)when feeding the H2SN2 mixture The impact of the factors was
Temperature (C) H2OH2S mass ratio Test duration (min)
600 0 300800 0 390600 45 120800 45 120700 225 120
600 0 120800 0 390600 45 120800 45 120700 225 120
600 0 120800 0 120600 45 120800 45 120700 225 120
600 0 240800 0 240600 45 120800 45 120
600 0 120800 0 240600 45 120800 45 120
600 0 120800 0 240600 45 120800 45 120
iv3
Table 3Characterization of the ash samples obtained in the combustion and gasification ofsewage sludge
Sewage sludgecombustion ash
Sewage sludgegasification ash
Ultimate analysisC (wt) 015 314H (wt) nda nda
N (wt) 028 077S (wt) 046 041
BET surface(m2 g1)
65 67
Pore volume(cm3 g1)
002 002
Average pore size(nm)
120 109
Metal content (mg gash1 )
Al 52 61Ca 65 84Fe 192 116K 14 nab
P 63 51Mg 17 nab
Na 4 nab
Si 122 nab
Ti 4 nab
a Not detectedb Not analyzed
102 N Gil-Lalaguna et al Fuel 141 (2015) 99ndash108
statistically analyzed by means of analysis of variance ANOVA(confidence level of 95 for the F-distribution) Design-Expert 7software (from Stat-Ease Inc) was used for this purpose The useof coded levels for the factors (1 for the lower limits and +1 forthe upper limits) in the ANOVA analysis enables an easy identifica-tion of the term with the greatest influence the higher the abso-lute value of the coefficient the more influential the factorAlthough the type of sewage sludge ash is not a numerical factora coded value of 1 was assigned to the combustion ash and +1to the gasification ash in order to obtain comparable coefficientsfrom the ANOVA analysis
The desulphurization tests performed at the laboratory werealso theoretically simulated to determine the maximum amountof H2S that could be removed from the gas from a thermodynamicpoint of view HSC Chemistry 61 software was used for this pur-pose This software uses the Gibbs energy minimization method tocalculate the amounts of products at equilibrium in isothermal andisobaric conditions The reaction system must be specified for thecalculations
3 Results and discussion
The characterization results of the fresh ash samples are pre-sented in Section 31 After that desulphurization performanceresults are shown and discussed in two different sections accord-ing to the gas mixture used (Sections 32 and 33) The H2S break-through curves and the sulphur content in the solid samples(expected and measured data) are the main results evaluated
31 Characterization of the fresh ash samples
Table 3 summarizes some characterization results of the ashsamples resulting from combustion and gasification of sewagesludge Both solids contained a small fraction of sulphur and someamount of carbon was also present in the gasification ash (around3 wt) The surface properties of sewage sludge ash are not goodenough for its use as an adsorbent material but its desulphuriza-tion potential is based on its metallic content The main metalsdetected by ICPndashOES were Fe Si Ca and Al According to the
iv4
thermodynamic study reported by Westmoreland and Harrison[21] concerning the desulphurization potential of different metaloxides Ca and Fe oxides are able to react with H2S to form metalsulphides so both ash samples are potential sulphur sorbentsSome differences were found in the metal content of the solidswhich might indicate that the ash fraction of sewage sludge didnot remained completely inert during the combustion and gasifica-tion processes Heterogeneity of sewage sludge may also explainthese differences Particularly striking is the case of Fe contentwhich was much lower in the gasification ash Part of the Fe con-tent in the raw sewage sludge could be in the form of iron chloride(FeCl3) as a result of the use of this compound as a coagulant agentduring the wastewater treatment During combustion excess oxy-gen appears to favour the retention of Fe in the ash in the oxideform However during the gasification process the reduced pres-ence of oxygen prevents the total conversion of FeCl3 to iron oxi-des so some amount of Fe could leave the reactor in the gasphase as FeCl3 evaporates at 315 C [22]
Fig 2 shows the XRD patterns of the combustion ash and gasi-fication ash before the desulphurization tests Species such asquartz calcite iron oxides and different calcium and iron phos-phates were detected in the ash samples The oxidation state ofFe is one of the main differences in the XRD patterns Fe appearsin the form of hematite (Fe2O3) in the combustion ash and in theform of magnetite (Fe3O4) in the gasification ash Concordantlythe reddish colour characteristic of hematite was only observedon the combustion ash
32 Desulphurization performance H2SN2 mixture as inlet gas
Different tolerable sulphur levels can be found in the literaturedepending on the gas application For instance the H2S concentra-tion limit ranges between 20 and 750 ppmv for gas turbine appli-cation [14] An intermediate value of 100 ppmv has been used inthis work to define the H2S breakthrough time
The H2S breakthrough curves obtained at the reactor outletwhen feeding the dry and moist H2SN2 mixtures are depicted inFig 3 as a function of temperature and type of ash An exit gasessentially free of H2S was obtained during 300 min and 260 minby using the combustion ash and the gasification ash respectivelyat 800 C and under dry conditions (Fig 3a) However when steamwas present in the reaction medium the H2S breakthrough timecorresponding to the combustion ash was reduced to 50 min whilethe gasification ash showed a complete loss of its capacity toremove H2S (Fig 3b) According to the general reaction of metaloxides with H2S (1) thermodynamics predicts a negative effect ofsteam on the equilibrium between H2S and metal oxide sorbentsbecause of the simultaneous regeneration of the spent metalsulphides
MexOy ethsTHORN thorn y H2S ethgTHORN $MexSy ethsTHORN thorn y H2O ethgTHORN DH lt 0 eth1THORN
Experimental results reported in the literature show differentsteam impact levels on the sulphidation rate depending on thesorbent material and the operating conditions [15] For instanceKim et al studied the effect of steam on H2S removal by a ZnOsorbent and found that the presence of 45 steam reduced theH2S breakthrough time by almost half at 363 C [23] In generalthe effect of steam on the sulphur sorbent performance is expectedto be more severe at higher temperatures but there are not manystudies concerning this effect [15] In the present work the H2Sbreakthrough time was reduced by 85 in the presence of 30steam when testing the combustion ash at 600ndash800 C while thegasification ash showed even higher loss in its sulphur capturecapacity as a consequence of their different metallic phases Thepresence of carbon in the gasification ash also seemed to slightlyaffect its desulphurization performance The downward trend
20 30 40 50 60
2θ
Gasification ash
Combustion ash
Fig 2 XRD patterns of the fresh samples of sewage sludge combustion ash andsewage sludge gasification ash
(a) Dry H2SN2 mixture
(b) Moist H2SN2 mixture (30 vol H2O)
000
005
010
015
020
025
H2S
outle
t flo
w ra
te (m
L STP
min
)
Time (min)0 50 100 150 200 250 300 350 400
0 20 40 60 80 100 120000
005
010
015
020
025
H2S
outle
t flo
w ra
te (m
L STP
min
)
Time (min)
Fig 3 H2S breakthrough curves evolution of the H2S flow rate (mLSTP min1)leaving the reactor when feeding the H2SN2 mixture
N Gil-Lalaguna et al Fuel 141 (2015) 99ndash108 103
observed in the H2S outlet flow at 800 C (Fig 3b) could be relatedto the steam gasification of the carbon present in the gasificationash Ultimate analyses of the gasification ash before and after theexperiment confirm this as the carbon content was reduced from314 to 025 wt
The presence of steam in the gas atmosphere also affected theblank run results As is well known susceptible alloys especiallysteels react with H2S forming metal sulphides as corrosion by-products This could explain the reduced H2S outlet flow rate inthe blank runs with respect to the inlet gas particularly at 800 Cand under dry operating conditions (Fig 3a) The hot metal partsin the experimental setup did not appear to react with H2S underwet conditions (Fig 3b) since as discussed above the formationof metal sulphides from metal oxides is restricted by the presenceof steam
The amount of H2S (mLSTP) removed from the gas up to thebreakthrough time was calculated using data from the blank runsas a reference
VH2S removed up to breakthrough time frac14 VH2S blank VH2S experiment eth2THORN
where VH2S experiment is the amount of H2S (mLSTP) leaving the reactorup to the H2S breakthrough time during each experiment andVH2S blank is the amount of H2S (mLSTP) leaving the reactor in theblank run during the same experimental time These H2S outlet vol-umes were calculated by integration of the area under the break-through curves (Fig 3) considering only the flow rate data up tothe breakthrough time The amounts of H2S removed from the gasup to the breakthrough time are included in Table 4 These datahave been statistically evaluated by analysis of variance (Table 5)Although the presence of a significant curvature prevents the useof the linear regression model obtained from the experimentaldesign used the relative influence of the factors can be assessedby the coefficients shown in Table 5 The existence of curvatureappears to be due to the sharp reduction of the H2S breakthroughtime when steam was present in the reaction medium The threestudied factors (temperature H2OH2S and ash type) as well astheir interactions significantly affect the amount of H2S removedfrom the gas up to the breakthrough time (p-value lt 00001) TheH2OH2S ratio is the most influential factor ethbH2O=H2S frac14 1776THORNThe higher the presence of steam the smaller the amount of H2Sremoved from gas The origin of the sewage sludge ash is also akey factor in the H2S removal process (bAsh type = 891) Largeramounts of H2S can be removed from the gas with the combustionash than with the gasification ash especially at the lowest temper-ature For instance 54 mLSTP of H2S were removed from the dry gasup to the breakthrough time using the combustion ash at 600 Cwhile the gasification ash only removed 4 mLSTP of H2S beforereaching the breakthrough time (Table 4) Non-significant differ-ences in the surface features of both solids were found (Table 3)so that the difference in their desulphurization performances mustbe related to ash composition On the one hand gasification ashcontains some amount of carbon that could hinder the access ofH2S to the metallic sites However this amount of carbon(314 wt) does not appear large enough to be the only cause forthe observed differences As discussed in Section 31 metal contentand metallic species detected in both types of sewage sludge ashwere not exactly the same as a consequence of the different reactiveatmospheres in the combustion and gasification processes As dis-cussed above one of the main differences in the composition ofthe ash samples was related to Fe content which was detected inthe form of Fe2O3 in the combustion ash and as Fe3O4 in the gasifi-cation ash Yoshimura et al [24] analyzed Fe2O3 and Fe3O4 samplesafter sulphidation with H2S at 400 C and the patterns obtained byExtended X-ray Absorption Fine Structure (EXAFS) showed that theintensity of the peak corresponding to FendashS coordination after sul-phidation of Fe3O4 was lower than that in the sulphided sample
iv5
Table 4Desulphurization performance results H2S removed from the gas up to the breakthrough time and sulphur content in the solid samples after the desulphurization tests(measured and expected data)
Gas mixture Ash type H2OH2Smass ratio
Temperature(C)
VH2 S removed to breakthrough time
(mLSTP)Sulphur content by ultimateanalysis (mg S gash
1 ) aExpected sulphurcontent (mg S gash
1 )
H2SN2 Sewage sludgecombustion ash
0 600 54 58 plusmn 1 920 800 54 63 plusmn 4 10045 600 7 215 plusmn 08 2945 800 9 14 plusmn 01 32225 700 14 208 plusmn 06 38225 700 12 188 plusmn 02 37225 700 12 166 plusmn 06 36
H2SN2 Sewage sludgegasification ash
0 600 4 232 plusmn 05 270 800 48 644 plusmn 07 10045 600 2 49 plusmn 02 745 800 0 11 plusmn 01 8225 700 0 141 plusmn 05 17225 700 0 126 plusmn 05 18225 700 0 118 plusmn 04 17
Syntheticgasificationgas
Sewage sludgecombustion ash
0 600 36 464 plusmn 06 640 800 1 55 plusmn 4 5345 600 4 268 plusmn 05 3245 800 1 85 plusmn 05 11
Syntheticgasificationgas
Sewage sludgegasification ash
0 600 1 20 plusmn 1 190 800 0 332 plusmn 06 3145 600 1 58 plusmn 02 1145 800 1 45 plusmn 05 8
a Mean value plusmn standard deviation
Table 5ANOVA results and linear regression coefficients for the amount of H2S removed from the gas up to the breakthrough time under the H2SN2 atmosphere
Sum of squares (SS) Degrees of freedom p-Value Coefficient (b)a
Model 454422 7 lt00001 ndashIntercept ndash 1 ndash 2236 plusmn 120T 23871 1 lt00001 546 plusmn 120H2OH2S 252405 1 lt00001 1776 plusmn 120Ash type 63546 1 lt00001 891 plusmn 120T-H2OH2S 23005 1 lt00001 536 plusmn 120T-ash type 19503 1 lt00001 494 plusmn 120H2OH2S-ash type 21321 1 lt00001 516 plusmn 120T-H2OH2S-ash type 27730 1 lt00001 589 plusmn 120Curvature 88320 2 lt00001 ndashPure error 267 4 ndash ndashCorrected total 543009 13 ndash ndashR2 = 084 (=SSmodelSScorrected total)
a 95 Confidence interval for the regression coefficients
104 N Gil-Lalaguna et al Fuel 141 (2015) 99ndash108
of Fe2O3 thus indicating more difficulty in sulphiding Fe3O4 at lowtemperatures This fact as well as the lower Fe content detected inthe gasification ash may explain the rapid saturation of the gasifi-cation ash The difference in the desulphurization performance ofboth solids was reduced with increasing temperature probablydue to a significant increase in the Fe3O4 sulphidation reaction rate
As the desulphurization process is based on gasndashsolid reactionsthe sulphidation rate is expected to be controlled by chemical reac-tion kinetics or by mass transfer The effect of the reaction temper-ature on the amount of H2S removed from the gas up to thebreakthrough time is clearly dependent on the ash type and pres-ence of moisture in the gas (bT = 546 bT-Ash type = 494bT-H2O=H2S frac14 536) Thus temperature hardly affected the removedamount of H2S under wet operating conditions andor when usingthe combustion ash whereas a great positive impact was observedwhen the gasification ash was used at dry conditions
In addition to the evaluation of the H2S breakthrough curvessulphur content in the solid samples was measured after the desul-phurization tests by means of an elemental analyzer These resultsare included in Table 4 Only a few of these data are directly com-parable with each other because they refer to the total amount ofsulphur removed in each complete experiment and the duration
iv6
of the experiments was not the same in all cases Comparable datashow lower sulphur content in the gasification ash than in thecombustion ash after the experiments performed at 600 and700 C while similar sulphur contents were detected in both solidsafter the experiments performed at 800 C pointing to similar sul-phidation reaction rates of Fe2O3 and Fe3O4 at higher temperaturesThe highest sulphur content detected in both types of ash wasaround 63ndash64 mg S gash
1 after operating at 800 C and dry condi-tions during 390 min
In order to check the sulphur mass balance the expected sul-phur content in the used ash samples has been calculated assum-ing that all the amount of H2S removed from the gas remained inthe solid after the desulphurization test
Expected sulphur content ethmg S g1ashTHORN
frac14 VH2S blank VH2S experiment
224 32 eth3THORN
where VH2S experiment is the amount of H2S (mLSTP) leaving the reactorduring each complete experiment VH2S blank is the amount of H2S(mLSTP) leaving the reactor during the blank run (extrapolating tothe duration of the experiment where this differs) 224 is the
(a) Dry H2SN2 2SN2 mixture 800 ordmC
(c) Moist H2SN2
mixture 600 ordmC (b) Dry H
mixture 600 ordmC (d) Moist H2SN2 mixture 800 ordmC
0102030405060708090
100
Sulp
hur d
istri
butio
n (
)
0102030405060708090
100
Sulp
hur d
istri
butio
n (
)
0102030405060708090
100
Sulp
hur d
istri
butio
n (
)
0102030405060708090
100
Sulp
hur d
istri
butio
n (
)Fig 4 Evolution of the sulphur distribution into the main sulphur-containing species at equilibrium conditions (H2SN2 mixture as inlet gas)
Fig 5 Back-scattered electron image of the ash resulting from experiment 2
N Gil-Lalaguna et al Fuel 141 (2015) 99ndash108 105
volume of one ideal gas mole at STP (mLSTP mmol1) and 32 is thesulphur atomic mass (mg mmol1) The expected sulphur contentsare included in Table 4 These calculated data are higher than theexperimental results obtained from the elemental analyzer Thissuggests that other sulphur-containing gases not detected by themicro GC could be formed during the desulphurization tests Espe-cially striking is the difference between the results obtained inexperiment 4 (800 C and moist H2SN2 mixture) for which theexpected sulphur content in the solid was around 32 mg S gash
1 while the elemental analyzer only detected 14 mg S gash
1 This lattervalue was significantly lower than that detected at 600 C(215 mg S gash
1 ) so the final sulphur content of the solid underwet conditions was favoured at low temperatures
Equilibrium simulations were conducted in order to try toexplain these observed differences from a thermodynamic pointof view Fig 4 shows the evolution of the theoretical distributionof sulphur into the main sulphur-containing species resulting fromthe equilibrium simulation FexSy CaS H2S CaSO4 and SO2 As thecomposition of the solid evolves over time (discontinuous bed ofsolid and continuous feed of gas) successive simulations for smalltime intervals (10 min) were performed in order to obtain anapproximation to the real process during 300 min The amount ofCa (in the form of CaO) and Fe (in the form of Fe2O3) present in 1g of sewage sludge ash was the initial solid for the first simulationThe gas input for each equilibrium calculation was the amount ofgas fed during 10 min of experiment while the solid input wasthe solid resulting from the previous simulation Successive gasndashsolid contact intervals of 10 min are represented in Fig 4 As canbe noted the formation of CaS is thermodynamically favoured overthe formation of FexSy in the early stage of the desulphurizationprocess The available amount of Ca decreases and FexSy formationbecomes more significant while the desulphurization process pro-gresses The remaining fraction of H2S in the gas stream increasesat high temperature (exothermic nature of the sulphidation reac-tion) and in the presence of steam Besides the formation of metalsulphides thermodynamics predicts the formation of SO2 andCaSO4 (as a result of the reaction of the formed SO2 with CaO)The formation of both SO2 and CaSO4 is favoured at the presenceof steam The temperature increase shifts the reaction to SO2 for-mation Thus SO2 formation may be the reason for the observeddifferences in the expected and measured sulphur contents in
the ash samples Maximum sulphur loadings of 107 103 42 and38 mg S gash
1 were predicted for experiments 1 2 3 and 4 respec-tively Experimental results obtained with the elemental analyzerwere 46 39 49 and 96 lower than the theoretical resultsrespectively (Table 4)
Fig 5 shows a back-scattered electron image of the ash result-ing from experiment 2 Numbers in Fig 5 indicate the points wherethe elemental composition was analyzed The atomic fractionsobtained by EDX in the different superficial points are shown inTable 6 As can be noted ash presents a heterogeneous surfaceC O Na Mg Al Si P S Ca Fe and Zn are detected along the surfacein different fractions It should be noted that the points with thehighest atomic percentage of S (238 in point 1 and 237 in point7) are also those with the highest Fe content (359 and 253respectively) thus suggesting the formation of either iron sulp-hides or iron sulphates On the other hand S was hardly detectedin other points such as in point 4 (mainly formed by O and Si inthe form of SiO2) or in point 6 in which despite of the high fractionof Fe (161) only 04 atomic S was found In this case as well asin points 2 and 3 the high presence of Fe is linked with a high pres-ence of P which indicates the presence of iron phosphates
iv7
Table 6Elemental composition (SEMEDX) on different superficial points of the ash resulting from experiment 2
Point in Fig 5 Elemental composition (atomic percentage)
C O Na Mg Al Si P S Ca Fe Zn
1 296 10 20 17 43 238 16 3592 08 579 34 37 24 133 11 66 106 033 22 459 05 15 57 92 115 86 63 884 07 690 03 01 03 276 06 095 09 507 03 33 62 93 122 46 70 566 16 579 24 14 40 117 04 41 161 047 335 08 13 78 35 237 41 253
005
010
015
020
025
utle
t flo
w ra
te (m
L STP
min
)
106 N Gil-Lalaguna et al Fuel 141 (2015) 99ndash108
Fig 6 shows the XPS spectra in S 2p region of the ash resultingfrom experiment 2 Several doublets can be fitted to the experi-mental signal indicating different chemical states of sulphur inthe ash surface The peak located between 160 and 164 eV is indic-ative of Sn
2 (metal sulphides) but other oxidized sulphur formshave also been detected (169 eV SO4
2)
33 Desulphurization performance synthetic gasification gas as inletgas
The H2S breakthrough curves obtained when feeding the dryand moist (30 vol H2O) synthetic gasification gases are depictedin Fig 7 as a function of temperature and type of ash The H2S con-centration in the outlet gasification gas remained below 100 ppmvfor less time than in the previous case (H2SN2 mixture) Unlikeoccurring with the H2SN2 mixture a significant amount of H2Swas detected in the outlet gasification gas from almost the begin-ning of all the experiments (Fig 7) In this case the H2S break-through time was longer than 20 min only by using thecombustion ash at 600 C and dry conditions (165 min) Thisbreakthrough time was in turn significantly lower than thatobtained for the H2SN2 mixture under the same operating condi-tions (260 min) Reduction of iron oxides present in sewage sludgeash may explain this observed behaviour As discussed in theliterature [172125] the presence of CO and H2 in the gasificationgas creates a reducing atmosphere that causes the conversion ofFe3O4 and Fe2O3 to FeO or even to Fe in the temperature range of700ndash1000 C FeO and Fe show less favourable sulphidation equi-librium which led to a reduction in the sulphur capture capacity[25] The temperature rise from 600 to 800 C was found to bedetrimental for the H2S removal capacity of the combustion ash
Fig 6 XPS spectra in the S 2p region corresponding to the ash resulting fromexperiment 2
iv8
probably as a consequence of the higher reduction rate of Fe2O3
to FeO at 800 C (Fig 7a) However the H2S removal capacity ofthe gasification ash improved at higher temperature Thus theimpact of temperature on H2S removal from the gasification gasis the result of the competition of reduction and sulphidation reac-tion rates The influence of the other factors (presence of steam andtype of sludge ash) on H2S removal from the synthetic gasificationgas was similar to the previous case
Table 4 shows the results of the sulphur content measured in theash samples after the experiments performed with the syntheticgasification gas as well as data calculated according to the sulphurmass balance (Eq 3) Even though the longest H2S breakthroughtime was obtained for the combustion ash at 600 C the highest sul-phur loading was detected in the ash used at 800 C (46 vs55 mg S gash
1 after 240 min) The different shape of the H2S break-through curves explains this result After the desulphurization ofthe moist gasification gas the highest sulphur content was foundin the combustion ash operating at 600 C (268 mg S gash
1 after120 min) This value was slightly higher than that obtained afterthe desulphurization of the moist H2SN2 mixture
000
H2S
o
Time (min)0 50 100 150 200 250
0 20 40 60 80 100 120000
005
010
015
020
025
H2S
out
let f
low
rate
(mL ST
P m
in)
Time (min)
Fig 7 H2S breakthrough curves evolution of the H2S flow rate (mLSTP tmin1)leaving the reactor when feeding the synthetic gasification gas
(a) Dry synthetic gasification gas 600 ordmC (b) Dry synthetic gasification gas 800 ordmC
(c) Moist synthetic gasification gas 600 ordmC (d) Moist synthetic gasification gas 800 ordmC
FexSy CaS H2S
0102030405060708090
100
Sulp
hur d
istri
butio
n (
)
0102030405060708090
100
Sulp
hur d
istri
butio
n (
)
0102030405060708090
100
Sulp
hur d
istri
butio
n (
)
0102030405060708090
100
Sulp
hur d
istri
butio
n (
)Fig 8 Evolution of the sulphur distribution into the main sulphur-containing species at equilibrium conditions (synthetic gasification gas as inlet gas)
20 40 60 80 100 1209
10
11
12
13
14
H2 (v
ol
)
Time (min)
7
8
9
10
11
CO
(vol
)
Time (min)
14
15
16
17
CO
2 (vo
l
)
Time (min)
00
05
10
15
20
C2H
x (vol
)
Time (min)
20 40 60 80 100 120
20 40 60 80 100 12020 40 60 80 100 120
Fig 9 Evolution of the outlet fractions of H2 CO CO2 and C2Hx (dry basis) when feeding the moist synthetic gasification gas
N Gil-Lalaguna et al Fuel 141 (2015) 99ndash108 107
Thermodynamic data obtained for the synthetic gasificationgas feed are shown in Fig 8 In this case neither SO2 nor CaSO4
are present at equilibrium conditions since the reducing atmo-sphere created by the gasification gas prevents its formationH2S CaS and FexSy are the main sulphur-containing species andCOS is also formed in a very low proportion Some authors haveexperimentally detected the formation of COS in the reducingenvironment of the gasification gas (H2S + CO2 M COS + H2O)[26] which could explain the observed differences in the expectedand measured sulphur contents in the ash samples Maximumsulphur loadings of 85 84 42 and 41 mg S g1
ash are theoretically
predicted for experiments 20 21 22 and 23 respectively Experi-mental results measured with the elemental analyzer were 4635 36 and 79 lower than the theoretical results respectively(Table 4) Comparison of Figs 4 and 8 shows a great impact ofthe gas atmosphere on the distribution of sulphur between CaSand FexSy The presence of CO2 in the gasification gas can explainthis difference since this is the responsible gas for the carbonationreaction of CaO (CaO + CO2 M CaCO3) Excess CO2 shifts the reactionto the formation of CaCO3 especially at low temperature thus lim-iting the formation of CaS from CaO Thus besides the reduction ofthe iron oxides carbonation of CaO may also contribute to the
iv9
108 N Gil-Lalaguna et al Fuel 141 (2015) 99ndash108
different results obtained for the H2SN2 mixture and the syntheticgasification gas
In addition to H2S removal the composition of the inlet gasifi-cation gas was modified during some desulphurization tests thussuggesting some catalytic activity of sewage sludge ash The evolu-tion of the outlet concentrations of H2 CO CO2 and C2Hx whenfeeding the moist synthetic gasification gas is depicted in Fig 9As can be noted the use of both types of sewage sludge ash at800 C led to a slight increase in the fractions of H2 and CO2 anda decrease in the percentage of CO with respect to the inlet valuesThis could be attributed to an enhancement of the waterndashgas shiftreaction (CO + H2O M CO2 + H2) Fe content in sewage sludge ashmay explain this catalytic activity [27] Furthermore the fractionof C2Hx was slightly reduced suggesting an enhancement of steamreforming reactions The fraction of CH4 (not shown in Fig 9) wasnot modified in any case (40 vol) As shown in Fig 9 the compo-sition of the synthetic gasification gas was hardly modified duringthe blank runs and during the experiments performed at 600 C
4 Conclusions
The use of the solid by-products of combustion and gasificationof sewage sludge for high-temperature desulphurization (600ndash800 C) of different gases containing 5000 ppmv H2S has been eval-uated in this work In general the gasification ash showed worsedesulphurization ability than the combustion ash Some differ-ences in the composition of both solids may explain their differentbehaviour Lower iron content was detected in the gasification ash(116 mg Fe gash
1 ) than in the combustion ash (192 mg Fe gash1 )
Furthermore iron was forming part of different crystalline speciesFe2O3 in the combustion ash and Fe3O4 in the gasification ash
In the absence of interferences from other gases (gas atmo-sphere only composed of H2S and N2) the H2S breakthrough time(lt100 ppmv H2S) was around 300 min by using the combustionash at 800 C A final sulphur content of around 63 mg S g1
ash wasdetected in this spent ash The H2S breakthrough time was drasti-cally reduced to a few minutes in the presence of 30 steam as aconsequence of the simultaneous regeneration of the spent metalsulphides Thus the use of sewage sludge ash for H2S removal at600ndash800 C is only suitable for dry gas cleaning The desulphuriza-tion process was also negatively affected by the reducing atmo-sphere created by the gasification gas due to the simultaneousreduction of Fe2O3 and Fe3O4 to FeO whose sulphur capture capacityhas been proved to be lower and by the presence of CO2 whichcauses the carbonation of CaO Combustion ash at 600 C led to thebest results during desulphurization of the moist synthetic gasifica-tion gas (most realistic conditions) showing a H2S breakthroughtime of 50 min and a final sulphur content of 27 mg S g1
ash after120 min of experiment
H2S was the only sulphur-containing gas analyzed during theexperiments However thermodynamic data pointed to the possi-ble formation of other sulphur-containing gases during the desul-phurization process such as SO2 or COS (this latter in a very lowproportion) Therefore the analysis of the H2S breakthrough curvesalone can lead to misleading conclusions about the sulphur capturecapacity of a sorbent material In most cases sulphur contentdetected by the elemental analyzer in the spent combustion ashwas 35ndash50 lower than the maximum thermodynamic data calcu-lated assuming that the total content of calcium and iron in thesolid was in the form of CaO and Fe2O3 respectively
Acknowledgements
The authors gratefully acknowledge the financial support pro-vided by the Spanish Ministry of Science and Technology (research
iv10
project CTQ2010-20137) and the Spanish Ministry of Education(pre-doctoral grant awarded to N Gil-Lalaguna AP2009-3446)Authors also wish to thank the Servicio General de Apoyo a laInvestigacioacuten at Universidad de Zaragoza and Laboratorio deMicroscopiacuteas Avanzadas at Instituto de Nanociencia de Aragoacutenfor the XRD ICP-OES XPS and SEMEDX analyses
References
[1] Rulkens W Sewage sludge as a biomass resource for the production ofenergy overview and assessment of the various options Energy Fuel 2008229ndash15
[2] Fytili D Zabaniotou A Utilization of sewage sludge in EU application of old andnew methods ndash a review Renew Sust Energy Rev 200812116ndash40
[3] Manara P Zabaniotou A Towards sewage sludge based biofuels viathermochemical conversion ndash a review Renew Sust Energy Rev2012162566ndash82
[4] Gil-Lalaguna N Saacutenchez JL Murillo MB Rodriacuteguez E Gea G Air-steamgasification of sewage sludge in a fluidized bed Influence of some operatingconditions Chem Eng J 2014248373ndash82
[5] Abu El-Rub Z Bramer EA Brem G Review of catalysts for tar elimination inbiomass gasification processes Ind Eng Chem Res 2004436911ndash9
[6] Yildirim O Kiss AA Huumlser N Leszligmann K Kenig EY Reactive absorption inchemical process industry a review on current activities Chem Eng J2012213371ndash91
[7] Bagreev A Bandosz TJ On the mechanism of hydrogen sulfide removal frommoist air on catalytic carbonaceous adsorbents Ind Eng Chem Res200544530ndash8
[8] Bandosz TJ On the adsorptionoxidation of hydrogen sulfide on activatedcarbons at ambient temperatures J Colloid Interface Sci 20022461ndash20
[9] Primavera A Trovarelli A Andreussi P Dolcetti G The effect of water in thelow-temperature catalytic oxidation of hydrogen sulfide to sulfur overactivated carbon Appl Catal A-Gen 1998173185ndash92
[10] Gutieacuterrez-Ortiz FJ Aguilera PG Ollero P Biogas desulfurization by adsorptionon thermally treated sewage-sludge Sep Purif Technol 2014123200ndash13
[11] Ros A Montes-Moraacuten M Fuente E Nevskaia DM Martin MJ Dried sludges andsludge-based chars for H2S removal at low temperature influence of sewagesludge characteristics Environ Sci Technol 200640302ndash9
[12] Ansari A Bagreev A Bandosz TJ Effect of adsorbent composition on H2Sremoval on sewage sludge-based materials enriched with carbonaceous phaseCarbon 2005431039ndash48
[13] Yuan W Bandosz TJ Removal of hydrogen sulfide from biogas on sludge-derived adsorbents Fuel 2007862736ndash46
[14] Meng X Jong W Pal R Verkooijen AHM In bed and downstream hot gasdesulphurization during solid fuel gasification a review Fuel Process Technol201091964ndash81
[15] Cheah S Carpenter DL Magrini-Bair KA Review of mid- to high- temperaturesulfur sorbents for desulphurization of biomass- and coal-derived syngasEnergy Fuel 2009235291ndash307
[16] Park NK Lee DH Lee JD Chang WC Ryu SO Lee TJ Effects of reduction of metaloxide sorbents on reactivity and physical properties during hot gasdesulphurization in IGCC Fuel 2005842158ndash64
[17] Tamhankar SS Hasatani M Wen CY Kinetic studies on the reactions involvedin the hot gas desulfurization using a regenerable iron oxide sorbent ndash IReduction and sulfidation of iron oxide Chem Eng Sci 1981361181ndash91
[18] Aacutelvarez-Rodriacuteguez R Clemente-Jul C Hot gas desulphurisation with dolomitesorbent in coal gasification Fuel 2008873513ndash21
[19] Elseviers WF Verelst H Transition metal oxides for hot gas desulphurizationFuel 199978601ndash12
[20] Garciacutea G Cascarosa E Aacutebrego J Gonzalo A Saacutenchez JL Use of different residuesfor high temperature desulphurisation of gasification gas Chem Eng J2011174644ndash51
[21] Westmoreland PR Harrison DP Evaluation of candidate solids for high-temperature desulfurization of low-btu gases Environ Sci Technol197610659ndash61
[22] Perry RH Green DW Perryrsquos chemical engineerrsquos handbook 7th ed NewYork McGraw-Hill 1999
[23] Kim K Jeon SK Vo C Park CS Norbeck JM Removal of H2S from steam-hydrogasifier product gas by zinc oxide sorbent Ind Eng Chem Res2007465848ndash54
[24] Yoshimura Y Yasuda H Sato T Shimada H Utilization of thermodynamicdatabase in the systems using molybdate and iron based catalysis Coal SciTechnol 1995241275ndash8
[25] Tseng TK Chang HC Chu H Chen HT Hydrogen sulfide removal from coal gasby the metalndashferrite sorbents made from the heavy metal wastewater sludge JHazard Mater 2008160482ndash8
[26] Hepola J Simell P Sulphur poisoning of nickel-based hot gas cleaning catalystsin synthetic gasification gas I Effect of different process parameters ApplCatal B-Environ 199714287ndash303
[27] Liu QS Zhang QC Ma WP He RX Kou LJ Mou ZJ Progress in waterndashgas-shiftcatalyst Prog Chem 200517389ndash98
ELSEVIER LICENSETERMS AND CONDITIONS
Nov 18 2014
This is a License Agreement between Noemiacute Gil (You) and Elsevier (Elsevier) providedby Copyright Clearance Center (CCC) The license consists of your order details the termsand conditions provided by Elsevier and the payment terms and conditions
All payments must be made in full to CCC For payment instructions please seeinformation listed at the bottom of this form
Supplier Elsevier LimitedThe BoulevardLangford LaneKidlingtonOxfordOX5 1GBUK
Registered CompanyNumber
1982084
Customer name Noemiacute Gil
Customer address CMariano Esquillor sn
Zaragoza 50018
License number 3511870187669
License date Nov 18 2014
Licensed content publisher Elsevier
Licensed content publication Chemical Engineering Journal
Licensed content title Airndashsteam gasification of sewage sludge in a fluidized bed Influenceof some operating conditions
Licensed content author N Gil-LalagunaJL SaacutenchezMB MurilloE RodriacuteguezG Gea
Licensed content date 15 July 2014
Licensed content volumenumber
248
Licensed content issuenumber
na
Number of pages 10
Start Page 373
End Page 382
Type of Use reuse in a thesisdissertation
Portion full article
Format both print and electronic
Are you the author of thisElsevier article
Yes
Will you be translating Yes
Number of languages 1
Languages Spanish
Title of yourthesisdissertation
Estudio de la gasificacioacuten de lodos de EDAR en lecho fluidizadoEfecto de la atmoacutesfera reactiva evaluacioacuten energeacutetica y limpieza delgas producto
Expected completion date Jan 2015
Estimated size (number ofpages)
150
Rightslink Printable License
ELSEVIER LICENSETERMS AND CONDITIONS
Nov 18 2014
This is a License Agreement between Noemiacute Gil (You) and Elsevier (Elsevier) providedby Copyright Clearance Center (CCC) The license consists of your order details the termsand conditions provided by Elsevier and the payment terms and conditions
All payments must be made in full to CCC For payment instructions please seeinformation listed at the bottom of this form
Supplier Elsevier LimitedThe BoulevardLangford LaneKidlingtonOxfordOX5 1GBUK
Registered CompanyNumber
1982084
Customer name Noemiacute Gil
Customer address CMariano Esquillor sn
Zaragoza 50018
License number 3511870438485
License date Nov 18 2014
Licensed content publisher Elsevier
Licensed content publication Fuel
Licensed content title Air-steam gasification of char derived from sewage sludge pyrolysisComparison with the gasification of sewage sludge
Licensed content author N Gil-LalagunaJL SaacutenchezMB MurilloV RuizG Gea
Licensed content date 1 August 2014
Licensed content volumenumber
129
Licensed content issuenumber
na
Number of pages 9
Start Page 147
End Page 155
Type of Use reuse in a thesisdissertation
Intended publisher of newwork
other
Portion full article
Format both print and electronic
Are you the author of thisElsevier article
Yes
Will you be translating Yes
Number of languages 1
Languages Spanish
Title of yourthesisdissertation
Estudio de la gasificacioacuten de lodos de EDAR en lecho fluidizadoEfecto de la atmoacutesfera reactiva evaluacioacuten energeacutetica y limpieza delgas producto
Expected completion date Jan 2015
Rightslink Printable License
ELSEVIER LICENSETERMS AND CONDITIONS
Nov 18 2014
This is a License Agreement between Noemiacute Gil (You) and Elsevier (Elsevier) providedby Copyright Clearance Center (CCC) The license consists of your order details the termsand conditions provided by Elsevier and the payment terms and conditions
All payments must be made in full to CCC For payment instructions please seeinformation listed at the bottom of this form
Supplier Elsevier LimitedThe BoulevardLangford LaneKidlingtonOxfordOX5 1GBUK
Registered CompanyNumber
1982084
Customer name Noemiacute Gil
Customer address CMariano Esquillor sn
Zaragoza 50018
License number 3511870675551
License date Nov 18 2014
Licensed content publisher Elsevier
Licensed content publication Energy
Licensed content title Energetic assessment of air-steam gasification of sewage sludge andof the integration of sewage sludge pyrolysis and air-steamgasification of char
Licensed content author N Gil-LalagunaJL SaacutenchezMB MurilloM Atienza-MartiacutenezGGea
Licensed content date 1 November 2014
Licensed content volumenumber
76
Licensed content issuenumber
na
Number of pages 11
Start Page 652
End Page 662
Type of Use reuse in a thesisdissertation
Intended publisher of newwork
other
Portion full article
Format both print and electronic
Are you the author of thisElsevier article
Yes
Will you be translating Yes
Number of languages 1
Languages Spanish
Title of yourthesisdissertation
Estudio de la gasificacioacuten de lodos de EDAR en lecho fluidizadoEfecto de la atmoacutesfera reactiva evaluacioacuten energeacutetica y limpieza delgas producto
Rightslink Printable License
ELSEVIER LICENSETERMS AND CONDITIONS
Nov 18 2014
This is a License Agreement between Noemiacute Gil (You) and Elsevier (Elsevier) providedby Copyright Clearance Center (CCC) The license consists of your order details the termsand conditions provided by Elsevier and the payment terms and conditions
All payments must be made in full to CCC For payment instructions please seeinformation listed at the bottom of this form
Supplier Elsevier LimitedThe BoulevardLangford LaneKidlingtonOxfordOX5 1GBUK
Registered CompanyNumber
1982084
Customer name Noemiacute Gil
Customer address CMariano Esquillor sn
Zaragoza 50018
License number 3511870874837
License date Nov 18 2014
Licensed content publisher Elsevier
Licensed content publication Fuel
Licensed content title Use of sewage sludge combustion ash and gasification ash forhigh-temperature desulphurization of different gas streams
Licensed content author N Gil-LalagunaJL SaacutenchezMB MurilloG Gea
Licensed content date 1 February 2015
Licensed content volumenumber
141
Licensed content issuenumber
na
Number of pages 10
Start Page 99
End Page 108
Type of Use reuse in a thesisdissertation
Intended publisher of newwork
other
Portion full article
Format both print and electronic
Are you the author of thisElsevier article
Yes
Will you be translating Yes
Number of languages 1
Languages Spanish
Title of yourthesisdissertation
Estudio de la gasificacioacuten de lodos de EDAR en lecho fluidizadoEfecto de la atmoacutesfera reactiva evaluacioacuten energeacutetica y limpieza delgas producto
Expected completion date Jan 2015
Rightslink Printable License
GASIFICATION OF SEWAGE SLUDGE IN A FLUIDIZED BED
REACTOR EFFECT OF THE REACTIVE ATMOSPHERE ENERGETIC
ASSESSMENT AND CLEAN-UP OF THE PRODUCT GAS
PhD Thesis
by
NOEMIacute GIL LALAGUNA
November 2014 Zaragoza
Supervisors Dr Joseacute Luis Saacutenchez Cebriaacuten Dr Mariacutea Benita Murillo Esteban
Dr Saacutenchez Cebriaacuten and Dr Murillo Esteban
Associate Professors at the University of Zaragoza in the Department of Chemical Engineering and Environmental Technologies and research team members of the Thermo-chemical Processes Group of the Aragoacuten Institute of Engineering Research (I3A)
INFORM that
This PhD report entitled
ldquoGasification of sewage sludge in a fluidized bed reactor Effect of the reactive atmosphere energetic assessment and clean-up of the product gasrdquo
has been developed by Ms Gil Lalaguna under our supervision in the Department of Chemical Engineering and Environmental Technologies
We authorize the presentation of this report
In witness whereof we sign this certificate in Zaragoza on November 24 2014
Dr Joseacute Luis Saacutenchez Cebriaacuten Dr Mariacutea Benita Murillo Esteban
OUTLINE
1 PREFACE 1
2 INTRODUCTION 5
3 MATERIALS AND METHODS 15
31 Materials 15
311 Raw materials for gasification sewage sludge and sewage sludge char 15
312 Sewage sludge ash for H2S removal 16
313 Nickel-based catalysts 18
32 Experimental facilities and procedures 20
321 Gasification setup 20
322 Desulphurization setup 23
323 Catalyst testing setup 24
33 Operating conditions and experimental design 26
331 Gasification experiments 26
332 Desulphurization tests 28
333 Catalyst testing 29
4 DISCUSSION OF THE MAIN RESULTS 31
41 Sewage sludge and char gasification 31
42 Energetic assessment 48
43 Desulphurization of different gas streams 61
44 Catalyst activity tests 71
5 CONCLUSIONS AND FUTURE WORK 81
6 REFERENCES 87
Preface 1
1 PREFACE
This PhD thesis has been developed in the Thermochemical Processes Group belonging to
the Aragoacuten Institute of Engineering Research (I3A) This group focuses on various research
areas including thermo-chemical treatment of biomass and organic waste by pyrolysis and
gasification biodiesel production and improvement of its properties pollutants removal from
combustion gases such as NOx or soot and hydrogen production from catalytic reforming of
aqueous streams
The present work falls within the energetic valorization of organic waste and in particular
is focused on sewage sludge gasification Sewage sludge is the waste produced by wastewater
treatment processes The generated amount of this waste has increased in recent years due to
a stricter European legislation concerning urban wastewater treatment (Directive 91271CEE)
As a consequence environmentally and healthy safe management of sewage sludge before
final disposal has become a significant challenge in municipal wastewater treatment Thermo-
chemical processes such as gasification or pyrolysis are potential routes for the energetic
valorization of this organic waste Gasification involves the thermal conversion of a
carbonaceous material into combustible gas and ash in a reducing atmosphere The product
gas from gasification consists of a mixture of CO CO2 H2 CH4 and other light hydrocarbons
steam and N2 (if air is used as gasification medium) which could be used for power generation
in gas engines or turbines or as a chemical feedstock to produce chemicals (Wender 1996)
The Thermochemical Processes Group started to study the gasification of sewage sludge
in the mid-1990s when the first tests were conducted at the laboratory in cooperation with
the company Cadagua SA Based on the earlier experimental results a fluidized bed pilot plant
of 100 kgmiddoth-1 was designed and developed for sewage sludge gasification during the years
2001-2003 Some experiments were conducted in the pilot plant during 2003 but problems
arising from the large scale work raised the need to further study the process at lab scale in
order to evaluate the influence of the operating conditions Since then both sewage sludge
gasification and sewage sludge pyrolysis at lab scale have attracted special attention in the
Thermochemical Processes Group Financial support has been provided by the Spanish
Ministry during the last decade for this research area Specifically this PhD thesis has been
developed with financial support from the Spanish Ministry of Science and Technology through
the research project ldquoSewage sludge valorization by means of pyrolysis study and
improvement of the products use (CTQ2010-20137)rdquo as well as from the Spanish Ministry of
2 Preface
Education through the pre-doctoral grant awarded to the PhD student during the last four
years (AP2009-3446)
Most of the published studies concerning sewage sludge gasification (Adegoroye et al
2004 Campoy et al 2014 Dogru et al 2002 Midilli et al 2001 Petersen and Werther 2005
Tae-Young et al 2009 van der Drift et al 2001) as well as the previous work carried out in
the Thermochemical Processes Group (Aznar et al 2007 Aznar et al 2008 Manyagrave et al
2005 Manyagrave et al 2006) used air to gasify the dried waste However after the mechanical
dewatering by filter pressing or centrifugation the moisture content of the waste still exceeds
70 (Manara and Zabaniotou 2012) so steam gasification could make sense in such waste
Thus part of the experimental work carried out in this PhD thesis aims to extend the
knowledge about the effect of the reactive atmosphere on the gasification of sewage sludge
using different mixtures of air and steam as gasification medium Steam gasification is an
endothermic process so the addition of air into the gasifying medium can provide the
necessary energy through the partial combustion of the feedstock
The pyrolysis (thermal decomposition under inert atmosphere) of sewage sludge has also
been widely studied in the Thermochemical Processes Group during the last years (Fonts et al
2008 Fonts et al 2009 Gil-Lalaguna et al 2010) Fast pyrolysis of sewage sludge is focused
on maximizing the bio-oil production but a significant fraction of char (around 50 wt ) is also
obtained The remaining organic fraction in char gives it a moderate calorific value which could
be further exploited through combustion or gasification processes thus leading to an integral
valorization of the raw material Compared to the raw biomass char resulting from biomass
pyrolysis appears as a preferable feedstock for gasification from the point of view of tar
formation since most of the volatile matter responsible for tar formation is eliminated during
pyrolysis Air-steam gasification of the char obtained in the pyrolysis of sewage sludge has
been experimentally studied in this PhD thesis comparing the results with those obtained in
the direct gasification of sewage sludge
In addition to the operational and technical feasibility of thermo-chemical processes the
energetic assessment of such processes is a key factor to be considered before their industrial
development An energetic study of both the air-steam gasification of sewage sludge and the
integration of sewage sludge pyrolysis and air-steam gasification of char has been performed
Prior to the thermo-chemical treatment of sewage sludge by means of pyrolysis or gasification
sewage sludge thermal drying allows the reduction of water content in the waste thus
reducing the waste volume and facilitating handling of the biosolids As the thermal drying of
sewage sludge involves high energy consumption it is interesting to evaluate if the energy that
Preface 3
could be recovered from the products of sewage sludge gasification and pyrolysis (together
with char gasification) is enough to cover such energy demand
One of the major issues in biomass gasification is to deal with the tar formed during the
process Tar is a complex mixture of condensable organic compounds including single- to
multiple-ring aromatic compounds The presence of tar in the gasification gas leads to
operational problems associated with condensation formation of aerosols and polymerization
such as the blocking of downstream pipelines and fouling of engines and turbines To date tar
removal from gasification gas has not been solved satisfactorily so many studies in the
gasification research area are still focused on tar removal especially by means of catalytic
cracking (Anis and Zainal 2011 De Lasa et al 2011) In the particular case of sewage sludge
gasification an additional trouble is found as a consequence of the significant sulphur content
of this waste This leads to the formation of sulphur-containing gases during the gasification
process being hydrogen sulphide (H2S) the most abundant The presence of H2S in the
gasification gas entails both environmental and operational problems On the one hand
combustion of a H2S-containing gas leads to SO2 emissions partially responsible for acid rain
On the other hand the presence of H2S causes corrosion in pipes engines or turbines as well
as the poisoning of the common catalysts used for tar cracking and gas reforming after
gasification processes which are usually based on Ni These difficulties motivated the
subsequent studies developed in this PhD thesis focused on the gas clean-up
High-temperature desulphurization of various synthetic H2S-containing gases was studied
in order to evaluate the effect of the operating conditions (temperature and reactive
atmosphere) on the sulphur capture process This process is based on the chemical reaction of
H2S with metal oxides to form metal sulphides Due to its metallic content sewage sludge ash
derived from the combustion and gasification processes of this waste were chosen as sorbent
materials in the desulphurization tests The use of ash residues for H2S removal from hot gases
could be an attractive alternative due to its low cost Furthermore an integrated process could
be proposed to remove the H2S produced during the thermo-chemical treatment of sewage
sludge (gasification or pyrolysis) by using the own ash resulting from these processes in a
downstream cleaning stage
The last part of the PhD thesis was developed during a research stay at the VTT-Technical
Research Centre of Finland This work includes a study of catalytic reforming of tar model
compounds present in a H2S-containing synthetic gasification gas by using modified Ni-based
catalysts Different promoters (Ca Fe Mn and Cu) were added to a Ni-Al2O3 catalyst to
evaluate their impact on the tar reforming activity and sulphur poisoning resistance of the
4 Preface
catalysts From a thermodynamic standpoint these chosen promoters would be able to react
with H2S to form metal sulphides (Westmoreland and Harrison 1976) Thus it could be
expected that sulphur chemisorption on nickel active sites and the consequent activity loss
would be hindered by sulphur chemisorption on the promoter sites
In summary the main objective of this work is to improve the quality of the gas product
obtained from sewage sludge gasification through the optimization of the operating conditions
in the gasifier and the application of downstream gas cleaning treatments Various tasks have
been developed for that purpose
bull Study of the impact of some operating conditions on the sewage sludge gasification
performance using mixtures of air and steam as gasification medium
bull Study of the use of the char obtained in the pyrolysis of sewage sludge as a raw material for
the gasification process in order to reduce tar formation
bull Energetic assessment of the air-steam gasification of sewage sludge and of the combination
of sewage sludge pyrolysis and char gasification
bull Study of H2S removal from gases at high temperature by using sewage sludge ash obtained
in thermo-chemical treatments of this waste
bull Study of the activity and stability of nickel-alumina catalysts modified with different
promoters for the reforming of tar model compounds in presence of H2S
Much of the work developed in this PhD thesis has already been published or accepted for
publication in journals with high impact factor
- N Gil-Lalaguna JL Saacutenchez MB Murillo E Rodriacuteguez G Gea (2014) ldquoAir steam
gasification of sewage sludge in a fluidized bed Influence of some operating conditionsrdquo
Chemical Engineering Journal 248 373-382
- N Gil-Lalaguna JL Saacutenchez MB Murillo V Ruiz G Gea (2014) Air steam gasification of
char derived from sewage sludge pyrolysis Comparison with the gasification of sewage
sludge Fuel 129 147-155
- N Gil-Lalaguna JL Saacutenchez MB Murillo M Atienza-Martiacutenez G Gea (2014) Energetic
assessment of air-steam gasification of sewage sludge and of the integration of sewage
sludge pyrolysis and air-steam gasification of char Energy 76 652-662
- N Gil-Lalaguna JL Saacutenchez MB Murillo G Gea (2015) Use of sewage sludge
combustion ash and gasification ash for high-temperature desulphurization of different gas
streams Fuel 141 99-108
Introduction 5
2 INTRODUCTION
Biomass is one of the most important primary renewable energy sources and its thermo-
chemical conversion encompasses a wide range of materials conversion technologies and end-
uses of the products such as powerheat generation transportation fuels and chemical
feedstock Sewage sludge produced by wastewater treatment processes can be considered a
special case of biomass due to its substantial organic fraction and high enough calorific value
(Fytili and Zabaniotou 2008 Rulkens 2008) This sludge is enlisted as a non-hazardous waste
in the European Waste Catalogue (Decision 2001118EC)
The sludge stemming from wastewater treatment usually is in the form of a dilute
suspension and its composition is strongly influenced by the original pollution load and the
purification treatment applied in the process As a rough guide this composition includes (i)
water varying from a small percentage to more than 95 (ii) a substantial fraction of non-
toxic organic material (about 60 dry basis) including biological constituents such as nucleic
acids proteins carbohydrates and lipids as well as organic material undigested in the process
such as cellulose (iii) a fraction of inorganic material that comprises silicates aluminates
calcium- and magnesium-containing compounds and nutrients such as nitrogen phosphorus
and potassium (iv) toxic pollutants as a result of the industrial activity including persistent
organic compounds (pesticides industrial solvents dyes plasticizers surfactants etc) and
heavy metals (Zn Cr Pb Cu Cd Ni As and Hg) whose concentration can vary from less than 1
ppm to more than 1000 ppm (v) pathogens and other microbiological pollutants (Fytili and
Zabaniotou 2008 Rulkens 2008 Manara and Zabaniotou 2012) These toxic pollutants and
pathogens entail environmental and human health risks in the case of poor sludge
management
The production of sewage sludge has considerably increased during the last two decades
as a consequence of the stricter European legislation concerning urban wastewater treatment
(Directive 91271EEC) This Directive requires all cities above 2000 population equivalent to
implement secondary treatment of wastewater Thus new municipal wastewater treatment
plants and renewed technologies have been developed in order to improve the quality of the
effluents As a result the annual production of sewage sludge in the European Union almost
doubled in the period 1992-2005 increasing from 65 to 109 million tonnes of dry matter per
year (Kelessidis y Stasinakis 2012) Spain is among the five European countries with highest
annual sludge production with about 12 million tons of dry matter per year (Eurostat)
6 Introduction
In addition to the production of large quantities of sewage sludge the costs of its
treatment often represent more than 50 of the total wastewater treatment costs (Rulkens
2008) so environmentally and healthy safe management of sewage sludge before final
disposal has become a significant challenge in municipal wastewater treatment
Policy and legislation regarding sludge application and management are heavily dependent
upon local national and regional conditions However sewage sludge is catalogued as non-
hazardous waste and according to European waste policy (Directive 200898CE) the
following hierarchy shall always apply as a priority order in waste management (i) prevention
(ii) reuse (iii) recycling (iv) other recovery forms such as energy recovery and (v) disposal
Currently the most common options for sewage sludge management in the European Union
include its reuse in agriculture land reclamation and restoration incineration and landfill
disposal (Fytili and Zabaniotou 2008 Rulkens 2008 Manara and Zabaniotou 2012) On the
one hand sewage sludge contains components of agricultural value such as organic matter
and nutrients (nitrogen phosphorus and potassium) thus its use as a fertilizer appears as an
interesting option Sewage sludge reuse by direct agricultural application or composting is the
predominant choice for sludge management in the European Union (53 of produced sludge)
(Kelessidis and Stasinakis 2012) However the presence of harmful substances in the sludge
such as heavy metals toxins and pathogens has sparked some controversy over its agricultural
reuse because of the possible adverse effects of these toxic pollutants and pathogens on the
food chain This practice is regulated by European legislation (Directive 86278EEC) but more
stringent legislation has been adopted for sludge disposal in soil by several European
countries setting lower limit values for heavy metals as well as limit values for pathogens and
organic micro-pollutants The degree of flexibility varies from country to country (Kelessidis
and Stasinakis 2012) Landfilling has been another conventional route for sewage sludge
disposal but this practice is on decline in the European Union because of the prohibition of
landfilling of both liquid and untreated wastes as well as restrictions for bio-degradable solid
wastes (Directive 9931EC) Sewage sludge landfilling showed a significant and continuing
decline between 1992 and 2005 from 33 to 17 of the produced sludge while incineration
has been almost doubled (from 11 to 21) (Kelessidis and Stasinakis 2012)
As agricultural use is increasingly regarded as an insecure handling route thermo-chemical
valorization of sewage sludge seems an interesting alternative Thermo-chemical processes
(combustion pyrolysis and gasification) are considered one of the most promising ways to
produce energy and valuable products from waste by removing the organic part and leaving
only the mineral component for final disposal In the particular case of dry sewage sludge its
Introduction 7
calorific value (12-20 MJmiddotkg-1) is comparable to that of coal (146-267 MJmiddotkg-1) (Manara and
Zabaniotou 2012) This work focuses on sewage sludge gasification
Gasification involves the thermal conversion of a carbonaceous material into combustible
gas and ash in a reducing atmosphere The product gas from gasification consists of a mixture
of CO CO2 H2 CH4 and other light hydrocarbons steam and N2 (if air is used as gasification
medium) whose proportions depend on the composition of the raw material and the
operating conditions The gas produced can be used in more versatile ways than the original
biomass for example for power generation in gas engines or turbines or as a chemical
feedstock to produce chemicals (methanol Fischer-Tropchs liquidshellip) (Wender 1996)
Gasification technology appears to offer attractive options for medium to large scale
applications for example by integration of gasification in combined heat and power units
(IGCC) to efficiently convert the energy carried in the fuel to electricity (Manara and
Zabaniotou 2012) Furthermore compared to combustion gasification is a more friendly way
of using biomass for energy purposes since pollutant emissions are reduced in the presence of
non-oxidizing conditions (Franco et al 2003)
The gasification medium is an important parameter influencing the gaseous product
quality Air is the most widely used gasifying agent due to its low cost but N2 introduced with
the air dilutes the product gas (N2 content around 50 vol ) giving a poor-quality gas in terms
of calorific value (lower heating value of 4-6 MJmiddotm-3STP being STP standard conditions of
temperature and pressure at 25 ordmC and 1 atm) This gas is suitable for boiler engine and
turbine operation but not for pipeline transportation due to its low energy density
(Bridgwater 1995) Gasification with pure O2 prevents the dilution of the gas increasing its
lower heating value to 10-14 MJmiddotm-3STP but the cost of the process increases because of the
required air separation unit for the production of pure O2 Instead of airO2 steam can also be
used as gasification medium thus improving H2 production for the use of the gas in fuel cells
or as feedstock for the synthesis of chemicals Synthetic fuel production requires a high quality
syngas with high concentration of CO + H2 and sufficient high H2CO ratio H2 concentration as
high as 60 vol has been reported from steam gasification of biomass (Herguido et al 1992)
However unlike the reaction with O2 the reaction of carbon with steam is endothermic
requiring heat to be transferred at high temperatures which is difficult to achieve The
addition of some amount of O2 to the gasifying medium can provide the necessary energy for
steam gasification through the partial combustion of the feedstock Hence the joint use of
steam and O2 (or air) as gasification medium seems an attractive option Some studies on air-
steam gasification of biomass are reported in the literature (Campoy et al 2009 Gil et al
8 Introduction
1997 Lv et al 2004 Pinto et al 2003) but more efforts are required to assess both the
operational and energetic aspects of the process
In the particular case of sewage sludge gasification the first published studies date back to
mid-1990s (Bacaicoa et al 1995) Since then several studies have been performed at
laboratory facilities (Adegoroye et al 2004 Aznar et al 2007 Aznar et al 2008 Manyagrave et al
2005 Manyagrave et al 2006 Tae-Young et al 2009) and the process has even been tried at
demonstration and pilot scale (Campoy et al 2014 Dogru et al 2002 Judex et al 2012
Midilli et al 2001 Petersen and Werther 2005 van der Drift et al 2001) showing the
feasibility of obtaining a fuel gas from such waste Most of the published studies use air to
gasify the dried waste Steam has been scarcely used as gasification medium (Domiacutenguez et al
2006 Nipattummakul et al 2010 Xie et al 2010 Zhang et al 2011) Nipattummakul et al
(2010) found that H2 production during steam gasification of sewage sludge was three times
higher than that obtained in the air gasification of dried sludge
After stabilization and mechanical dewatering of sewage sludge by filter pressing or
centrifugation water content can be further reduced by means of thermal drying in order to
reduce the waste volume and facilitate its handling However this stage consumes a large
amount of energy and raises the cost of sewage sludge disposal Before thermal drying the
moisture content of sludge still exceeds 70 (Manara and Zabaniotou 2012) so steam
gasification seems to make sense in such waste Thus wet sewage sludge can be assumed as
an interesting feedstock to manufacture H2 (Fytili and Zabaniotou 2008) The high moisture
content of sewage sludge generates at high temperatures a steam-rich atmosphere leading
consequently to an in situ steam reforming of the volatile compounds and to a partial
gasification of the solid char which contributes to the production of H2-rich fuel gas
(Domiacutenguez et al 2006) Hence drying pyrolysis and partial gasification of the raw sludge
take place in the same process
It is generally referred by different authors that the process of biomass gasification occurs
through various steps (Antal et al 1979) initial drying of the raw material subsequent
pyrolysis or thermal decomposition that produces volatile matter and a carbonaceous solid
residue (char) followed by secondary reactions involving the volatile products and finally
gasification reactions of the remaining carbonaceous residue with steam and CO2 These latter
reactions are slow compared to devolatilization and gas phase reactions The sequence and
duration of the stages vary with the type of reactor used
Introduction 9
The main gas-solid reactions taking place in the gasifier can be summarized as follows
(Mondal et al 2011)
Partial oxidation C + frac12 O2 rarr CO ∆H298K = -110 kJmiddotmol-1 (eq 21)
Complete oxidation C + O2 rarr CO2 ∆H298K = -393 kJmiddotmol-1 (eq 22)
Water gas reaction C + H2O harr CO + H2 ∆H298K = 131 kJmiddotmol-1 (eq 23)
Boudouard reaction C + CO2 harr 2CO ∆H298K = 172 kJmiddotmol-1 (eq 24)
Methanation C + 2H2 harr CH4 ∆H298K = -75 kJmiddotmol-1 (eq 25)
As commented above the produced gases can undergo further reactions as follows
Water-gas shift reaction CO + H2O harr CO2 + H2 ∆H298K = -41 kJmiddotmol-1 (eq 26)
Steam reforming CH4 + H2O harr CO + 3H2 ∆H298K = 205 kJmiddotmol-1 (eq 27)
Dry reforming CH4 + CO2 harr 2CO + 2H2 ∆H298K = 247 kJmiddotmol-1 (eq 28)
Most of these reactions are equilibria and can proceed in either direction depending on
the temperature pressure and concentration of the reacting species
One of the major issues in biomass gasification is to deal with the tar formed during the
process Tar is a complex mixture of condensable hydrocarbons produced during the thermal
decomposition of biomass which includes single- to multiple-ring aromatic compounds along
with other oxygen-containing hydrocarbons and complex polycyclic aromatic hydrocarbons
(PAHs) In the context of gasification tar is usually defined as all the organic contaminants with
a molecular weight larger than benzene except soot and char (Neeft et al 2002) The
presence of tar in the gasification gas leads to operational problems associated with
condensation (lt 450 ordmC) formation of aerosols and polymerization of tar compounds to form
more complex structures that block downstream pipelines and foul engines and turbines thus
restricting the direct use of the gas (McKendry 2002b) Moreover tar compounds entail
serious environmental problems due to their persistent and toxic nature (Nisbet and Lagoy
1992) The minimum allowable limit for tar is highly dependent on the kind of process and the
gas end-use Several researchers state that internal combustion gas engines are more tolerant
of contaminants than gas turbines In particular it is possible to have tar content up to 50-100
mgmiddotm-3STP for internal combustion engines but less than 5 mgmiddotm-3
STP is preferable for gas
turbines (Anis and Zainal 2011) However if the gas is fed at the turbine at high temperature
tar content can be not a problem at all (Staringhl and Neergaard 1998)
The tar content in the product gas from biomass gasification usually greatly exceeds the
allowable limits ranging between 05 and 100 gmiddotm-3STP depending on the feedstock the
10 Introduction
operating conditions and the gasifier type (Devi et al 2003) Various gasification reactors
types exist among which the fixed bed (downdraft or updraft) fluidized bed and entrained
flow gasifiers are the most widely used (Mondal et al 2011) Fluidized bed configuration
(which is the gasification technology used in the present work) presents higher gasification
efficiency compared to fixed bed gasifiers since mass and heat transfer phenomena are
enhanced However the product gas usually has increased content in solid particulates and tar
with respect to downdraft fixed bed reactors (Han and Kim 2008) The typical range of tar
content in fluidized bed gasifiers is around 8-15 gmiddotm-3STP (Corella et al 2006) Hence tar
removal becomes one of the most necessary and urgent problems during biomass gasification
Tar removal technologies can be broadly divided into two approaches treatments inside
the gasifier (primary methods) and gas cleaning downstream the gasifier (secondary methods)
Primary methods can be defined as all the measures taken in the gasification step itself to
prevent or convert tar formed in the gasifier The presence of an active material in the gasifier
can largely improve the product gas distribution Tar contents as low as 1-2 gmiddotm-3STP have been
obtained by adding bed materials such as dolomite and olivine in fluidized beds (Gil et al
1999a Olivares et al 1997 Rapagnagrave et al 2000) On the other hand a proper selection of the
operating conditions (temperature pressure gasification medium amount of gasifying agent
residence timehellip) is critical to reduce tar formation (Devi et al 2003)
It is generally referred by different authors that an increasing temperature promotes the
formation of gaseous products at the expense of total tar Different tar decomposition
mechanisms are proposed in the literature suggesting kinetically controlled processes that are
enhanced with increasing temperature (Li and Suzuki 2009a) An operating temperature
above 800 ordmC is usually preferred to achieve high carbon conversion and low tar content in the
resultant product gas However there are other factors that limit the operating temperature
such as are the reduced gas heating value and the risk of ash sintering (Devi et al 2003) Tar
composition is also affected by temperature The increase in temperature drastically reduces
the amount of oxygen-containing components and substituted 1-ring and 2-ring aromatics
while formation of 3- and 4-ring aromatics increases rapidly resulting in an increase in the tar
dew point In summary higher temperatures favor the formation of fewer aromatic tar species
without substituent groups such as benzene naphthalene and phenanthrene (Kinoshita et al
1994)
Selectivity of the gasification reactions varies with the gasification medium thus affecting
both gas composition and tar formation Air gasification produces slow-reacting tar while
steam gasification produces tar with a lower molecular weight (McKendry 2002b) Likewise
Introduction 11
Gil et al (1999b) found that tar generated during steam gasification was easier to destroy
with Ni-based catalysts or with dolomites than tar generated during air gasification They
found higher tar contents in steam gasification (30-80 gmiddotm-3STP) than in air gasification (2-20
gmiddotm-3STP) but the reduced operating temperature set in the first case may influence this result
More studies are necessary to increase knowledge of this effect The ratio between the
gasifying agent flow rate and the biomass feed rate also influences tar content in the gas In
the case of airO2 gasification the equivalence ratio (ER ratio between the actual fuel-to-O2
ratio and the stoichiometric fuel-to-O2 ratio) usually varies from 020 to 040 (Narvaacuteez et al
1996) An ER increase means more availability of oxygen to react with the volatile matter thus
reducing the tar content in the gas (Kinosita et al 1994 Narvaacuteez et al 1996) Tar content of
about 2-4 gmiddotm-3STP was obtained by Narvaacuteez et al (1996) when the ER was increased up to 045
in the gasification of pine sawdust at 800 ordmC However gas composition is also affected by ER
and an increase in the fractions of CO2 and N2 (if gasifying with air) is expected at higher ER
which means a reduced gas heating value
Compared to the raw biomass char resulting from biomass pyrolysis appears as a
preferable feedstock for gasification from the point of view of tar formation The product gas
from the direct gasification of raw biomass is usually rich in tar because of the high volatile
matter content of the solid However in the case of char gasification a product gas with lower
tar content can be obtained since most of volatile matter is eliminated during the pyrolysis
process Fast pyrolysis of biomass focuses on the production of bio-oil being char the main by-
product of the process This remaining solid char shows different properties than the raw
biomass as a consequence of the thermal treatment The most remarkable differences are
related to surface area pore structure and chemical composition (ultimate and proximate
analyses) As a route towards an integral valorization of biomass the solid by-product of
pyrolysis can be reused as a material for activated carbon preparation (Gonzaacutelez et al 2009)
or be further exploited through thermo-chemical processes such as combustion (Di Blasi 2009)
or gasification (Chaudhari et al 2003 Haykiri-Acma et al 2006 He et al 2012 Nilsson et al
2014 Salleh et al 2010 Yan et al 2010)
In the particular case of sewage sludge pyrolysis common solid yields are around 35-55
wt (Fonts et al 2008 Fonts et al 2012 Inguanzo et al 2002 Pokorna et al 2009 Shen
and Zhang 2003) The use of sewage sludge char as an adsorbent material has been
investigated by some authors and the results showed a reduced porosity (surface area of 50-
150 m2middotg-1) compared to that of commercial active carbons (gt 500 m2middotg-1) because of the high
inorganic content in the sewage sludge char (Smith et al 2009) Studies on gasification of
12 Introduction
sewage sludge char are scarce and they are mainly focused on kinetic aspects (Nilsson et al
2012 Nowicki et al 2011 Scott et al 2005)
In addition to the primary methods for tar removal downstream cleaning treatments are
usually required in the gasification processes to further improve gas quality The different
secondary methods for tar removal are classified into mechanicalphysical methods thermal
cracking and catalytic cracking Mechanical cleanup systems include the use of cyclones
electrostatic precipitators filters (fabric filters ceramic filtershellip) activated carbon adsorbents
and scrubbers (Abu El-Rub et al 2004 Anis and Zainal 2011) These methods are considerably
efficient in removing solid particles accompanied by tar liquid droplets Tar removal cannot be
separated from solid particles removal at temperatures below its dew point while tar vapors
are hardly removed at higher temperatures Tar removal efficiencies of 40-70 and 0-50
have been reported by using electrostatic precipitators and fabric filters respectively while
higher removal efficiencies can be achieved by using venturi scrubbers (60-90) Tar removal
in the range of 70 can be expected with tar adsorbents based on activated carbon (Anis and
Zainal 2011 Han and Kim 2008) Tar management is an important drawback in such cleaning
treatments that remove tar from the gas without destroying it
Tar conversion into lighter gases such as H2 CO and CH4 appears as the most effective
method for tar removal These reactions are known to be kinetically limited (Abu El-Rub et al
2004 Anis and Zainal 2011) so either extremely high temperatures or the use of catalysts are
required for tar cracking reactions Biomass-derived tar is very refractory and hard to crack by
thermal treatment alone (Bridgwater 1995) Brandt and Henriksen (2000) found that a
temperature as high as 1250 ordmC and a residence time of 05 s were necessary to achieve a high
tar cracking efficiency The use of catalysts allows a lower operating temperature thus
reducing the energy demand for tar cracking Due to the advantages of converting tar into
useful gases and adjusting the composition of the product gas catalytic cracking has attracted
scientific interest since the mid-1980s Hydrocarbons may be reformed on the catalyst surface
with either steam (eq 29) or CO2 (eq 210) to produce additional CO and H2 The reaction
mechanism involves the dissociative adsorption of the hydrocarbon onto a metal site where
dehydrogenation occurs (Han and Kim 2008)
Steam reforming CnHm + n H2O harr n CO + (n + m2) H2 ∆H gt 0 (eq 29)
Dry reforming CnHm + n CO2 harr 2n CO + (m2) H2 ∆H gt 0 (eq 210)
Various types of catalysts such as minerals and calcined rocks (dolomite olivinehellip) alkali
metal catalysts and transition metal catalysts have been widely tested for tar removal during
Introduction 13
biomass gasification (Abu El-Rub et al 2004 Anis and Zainal 2011 De Lasa et al 2011 Sutton
et al 2001a) Among these Ni-based catalysts appear to be the most suitable choice for both
technical and economic reasons An increase in H2 and CO content of the exit gas as well as
the elimination or reduction of the hydrocarbon and CH4 content is usually observed when
using these catalysts at temperatures above 740 ordmC (Sutton et al 2001a) Heavy tar
destruction efficiencies of 98-99 have been reported with commercial steam reforming Ni-
catalysts (Aznar et al 1998 Zhang et al 2004) Not only the active metal but also the catalyst
support plays an important role in the catalyst activity since it affects the dispersion of the
active phase The use of different metal oxides such as Al2O3 MgO ZrO2 TiO2 CeO2 or SiO2
and natural materials such as dolomite olivine or activated charcoal as supports for Ni-
catalysts has been extensively reviewed in the literature (Courson et al 2000 Kimura et al
2006 Li et al 2009b Miyazawa et al 2006 Park et al 2010 Sato and Fujimoto 2007
Srinakruang et al 2006 Sutton et al 2001b Swierczynski et al 2007 Wang et al 2005) Of
these alumina (Al2O3) is the most commonly used support (Anis and Zainal 2011) Some
studies point to NiAl2O3 as one of the most efficient catalysts for tar removal (Sutton et al
2001b) but this is not stable and eventually deactivates (Swierczynski et al 2007)
Several deactivation mechanisms have been reported for Ni-based catalysts including
poisoning by sulphur chlorine and alkali metals sintering of Ni particles and coke formation
(Abu El-Rub et al 2004 Anis and Zainal 2011 De Lasa et al 2011 Sutton et al 2001a) The
modification of Ni catalysts with promoters can positively affect catalyst activity reducibility
regenerability and coke resistance as well as mechanical strength and attrition resistance
(Yung et al 2009) The addition of a wide variety of metals (Na K Ru Rh Mn Mo W Zr Mn
La Cehellip) to Ni-catalysts has been studied by many researchers (Bona et al 2008 Dou et al
2003 Nishikawa et al 2008 Richardson and Grey 1997 Seok et al 2002 Zhang et al 2007)
showing in some cases good anti-coking ability and improved durability and activity of the
catalysts In addition to coke deposition conventional Ni-based catalysts are very sensitive to
poisoning by sulphur compounds (Yung et al 2009) Strong metal-sulphur chemisorption
occurs during sulphur poisoning thus leading to saturation of the metal active sites This
poisoning effect has been studied by many researchers (Engelen et al 2003 Hepola and
Simell 1997 Struis et al 2009) using low concentrations of sulphur in the gas However
during the gasification of sulphur-containing raw materials such as coal or sewage sludge
sulphur concentration in the gas can easily exceed 1000 ppmv especially in the form of H2S
The presence of H2S in the gasification gas entails both environmental and operational
problems On the one hand combustion of a H2S-containing gas leads to SO2 emissions
14 Introduction
partially responsible for acid rain On the other hand the presence of H2S in gas causes
corrosion in pipes engines or turbines as well as deactivation of the common catalysts used
for tar cracking and gas reforming after gasification processes Several low and high
temperature processes for H2S removal from products and off-gases have been described and
developed at various stages Wet scrubbing with selected solvents has been a widely used low-
temperature process in the chemical process industry (Yildirim et al 2012) The use of
activated carbons for H2S removal at low temperature has also been extensively studied
(Bagreev and Bandosz 2005 Bandosz 2002 Primavera et al 1998) Given the combination of
their unique surface features (high specific surface and pore volume) and surface chemistry
improved by the addition of functional groups activated carbon-based materials have been
proved to work efficiently as adsorbents of sulphur-containing species such as H2S SO2 or
methyl mercaptans As a result of surface reactions H2S can be oxidized to either sulphur or
SO2 (Bandosz 2002) Some adsorbents materials have been prepared from the solid by-
products of sewage sludge pyrolysis or carbonization (Gutieacuterrez-Ortiz et al 2014 Ros et al
2006 Yuan and Bandosz 2007) obtaining in some cases pollutant removal efficiencies
comparable to those corresponding to commercial catalytic activated carbons
On the other hand high temperature desulphurization processes are advantageous from
an energy standpoint as a result of the elimination of gas cooling and the associated heat
exchangers (Meng et al 2010) Furthermore tar condensation after a gasification process can
be prevented by desulphurization at high temperature The high temperature desulphurization
process is based on the chemical reaction of H2S with metal oxides Typically metal oxides are
converted to sulphides during a sulphur loading stage under reducing hot gas conditions After
sulphidation the spent metal sulphides can be regenerated back to metal oxides by using
oxygen steam SO2 or a combination of these (Meng et al 2010) Different studies on the use
of metal oxide based sorbents for hot gas desulphurization can be found in the literature
(Aacutelvarez-Rodriacuteguez and Clemente-Jul 2008 Cheah et al 2009 Elseviers and Verelst 1999
Garciacutea et al 2011 Meng et al 2010 Park et al 2005 Tamhankar et al 1981 Westmoreland
and Harrison 1976) Several metal oxides can be used for high temperature sulphur capture
downstream the gasifier but each solid has its own advantages and limitations at the same
time Sorbents based on zinc manganese copper iron or calcium are some of the most
studied and promising materials (Cheah et al 2009 Meng et al 2010 Westmoreland and
Harrison 1976) Ash residues derived from thermo-chemical treatment of biomass or wastes
are known to contain metal oxides in different proportions so its use as sorbent materials for
H2S removal from hot gases could be an attractive alternative due to its low cost
Materials and methods 15
3 MATERIALS AND METHODS
31 Materials
311 Raw materials for gasification sewage sludge and sewage sludge char
The raw material for gasification was anaerobically digested and thermally dried sewage
sludge (SS) as well as the char resulting from its pyrolysis process The sewage sludge was
supplied by a Spanish urban wastewater treatment plant located in Madrid Table 31 presents
a brief characterization of both materials The proximate analyses were performed according
to standard methods (ISO-589-1981 for moisture ISO-1171-1976 for ash and ISO-5623-1974
for volatiles) the ultimate analyses (CHNS) were determined with a Carlo Erba EA1108
elemental analyzer (Analysis Service of Instituto de Carboquiacutemica) the higher heating values
(HHV) of the solids were measured using an IKA C-2000 calorimeter and their specific heat
capacities (Cp) were determined by differential scanning calorimetry with a Netzsch DSC 200
Maia (inert atmosphere 40 mL N2middotmin-1) Sewage sludge was ground and sieved before the
experiments to obtain a feed sample in the size range of 250-500 μm
Table 31 Sewage sludge and char characterization
Sewage sludge Char Proximate analysis (wt wet basis)
Moisture 648 170 Ash 3904 7420 Volatiles 5009 1502 Fixed carbon 439 908
Ultimate analysis (wt wet basis) C 2950 1549 H 467 097 N 527 185 S 131 035
HHV (MJmiddotkg-1) 128 52 LHV (MJmiddotkg-1) 118 50 Cp25ordmC (kJmiddotkg-1middotK-1) 115 082
Only 15 wt of the carbon content of sewage sludge is in the form of fixed carbon while
it reaches 59 wt in the sewage sludge char because of the structural changes occurring
during the pyrolysis process In terms of mass the amount of fixed carbon is doubled after
pyrolysis (439 vs 908 wt )
16 Materials and methods
312 Sewage sludge ash for H2S removal
The ash obtained from the gasification and combustion of sewage sludge was used as
sorbent material in the desulphurization tests Sewage sludge combustion was performed
under air atmosphere in a heating muffle furnace at 900 ordmC (20 ordmCmiddotmin-1) during two hours
Gasification ash was the solid by-product obtained in one of the sewage sludge gasification
experiments (experiment 5 in Table 36) Both ash samples were characterized before the
experiments by various techniques (Table 32) Ultimate analyses were performed using a Leco
TruSpec Micro elemental analyzer Textural properties such as BET surface area and average
pore size and volume (BJH method) were calculated from N2 physisorption isotherms using a
Micromeritics TriStar II 3000 analyzer The samples were previously degasified at 200 ordmC during
8 h in a N2 flow Then N2 adsorption-desorption isotherms were obtained at -196 ordmC and room
temperature respectively over the whole range of relative pressures Metal content in the ash
samples was analyzed by inductively coupled plasma combined with optical emission
spectroscopy (ICP-OES) using a Thermo Elemental IRIS Intrepid ICP-OES spectrometer
(Chemical Analysis Service of the Universidad de Zaragoza) The samples were dissolved by
microwave-assisted acid digestion in a CEM MARS microwave reaction system
Table 32 Characterization results of sewage sludge combustion ash and sewage sludge gasification ash
Combustion ash Gasification ash
Ultimate analysis (wt wet basis) C 015 314 H nd nd N 028 077 S 046 041
BET surface (m2middotg-1) 65 67 Pore volume (cm3middotg-1) 002 002 Average pore size (nm) 120 109 Metal content (mgmiddotg-1
ash) Al 52 61 Ca 65 84 Fe 192 116 K 14 na P 63 51
Mg 17 na Na 4 na Si 122 na Ti 4 na
nd not detected na not analyzed
Both solids initially contained a small fraction of sulphur Some amount of carbon was also
present in the gasification ash (around 3 wt ) The surface properties of sewage sludge ash
Materials and methods 17
are not good enough for its use as an adsorbent material but its desulphurization potential is
based on its metallic content The main metallic elements detected by ICP-OES were Fe Si Ca
and Al According to the thermodynamic study reported by Westmoreland and Harrison (1976)
concerning the desulphurization potential of different metal oxides Ca and Fe oxides are able
to react with H2S to form metal sulphides so both ash samples are potential sulphur sorbents
Some differences have been found in the metal content of the solids which might indicate
that the initial ash fraction does not remain completely inert during the combustion and
gasification processes Heterogeneity of sewage sludge may also explain these differences
Particularly striking is the case of Fe content which was much lower in the gasification ash Part
of the Fe content in the raw sewage sludge could be in the form of iron chloride (FeCl3) as a
consequence of the use of this compound as a coagulant agent during the wastewater
treatment During combustion excess oxygen appears to favor the retention of Fe in the ash in
the oxide form However during the gasification process the reduced presence of oxygen
prevents the total conversion of FeCl3 to iron oxides so some amount of Fe could leave the
reactor in the gas phase as FeCl3 evaporates at 315 ordmC
Powder X-ray diffraction (XRD) patterns of the fresh samples were acquired with a D-Max
Rigaku diffractometer (Service of X-Ray Diffraction and Fluorescence Analysis of the
Universidad de Zaragoza) equipped with a copper anode (voltage of 40 kV and current of 80
mA) The measurements were completed in the Braggrsquos angle (2θ) range from 5ordm to 95ordm using a
scanning rate of 003ordmmiddots-1 Phases present in the solid samples were defined according to the
JCPDS-International Centre for Diffraction Data 2000 database The obtained XRD patterns are
shown in Fig 31 Species such as quartz calcite iron oxides and different calcium and iron
phosphates have been detected in the ash samples The oxidation state of Fe is one of the
main differences in the XRD patterns Fe appears in the form of hematite (Fe2O3) in the
combustion ash and in the form of magnetite (Fe3O4) in the gasification ash Concordantly the
reddish color characteristic of hematite was only observed on the combustion ash
18 Materials and methods
Figure 31 XRD patterns of the sewage sludge ash samples Hematite (Fe2O3) Quartz (SiO2) Calcite (CaCO3) Fe3PO7 Ca3(PO4)2 Magnetite (Fe3O4) Whitlockite (Ca18Mg2H2(PO4)14)
313 Nickel-based catalysts
Nickel-based catalysts were prepared and modified by adding different metals (Fe Ca Mn
and Cu) in order to evaluate the effect of these promoters on the catalyst stability and activity
for tar reforming From a thermodynamic standpoint these chosen promoters are able to
react with H2S to form metal sulphides (Westmoreland and Harrison 1976) Thus it could be
expected that sulphur chemisorption on nickel active sites and the consequent activity loss
would be hindered by sulphur chemisorption on the promoter sites
The catalysts were prepared by incipient wetness impregnation of γ-alumina (250-315 μm
size) with aqueous solutions of nitrates of the metal of interest Ni(NO3)2middot6H2O Ca(NO3)2middot4H2O
Fe(NO3)3middot9H2O Cu(NO3)2middot3H2O and Mn(NO3)2middot4H2O Both the nickel nitrate and the promoter
nitrate were solved and impregnated in one step After impregnation the catalysts were dried
at 110 ordmC for 24 h followed by calcination under air atmosphere according to the following
temperature ramp 120 ordmC for 20 min 200 ordmC for 30 min 320 ordmC for 90 min and final calcination
temperature (700 ordmC or 900 ordmC) for 120 min Two different calcination temperatures were used
in order to evaluate their influence on the catalyst activity and stability Metal loading in the
calcined samples was around 8 wt of Ni and 8 wt of promoter (Ca Fe Cu or Mn)
Textural properties (Table 33) and XRD patterns (Fig 32) of the calcined powder catalysts
20 30 40 50 60
2θ
Gasification ash
Combustion ash
Materials and methods 19
were obtained analogously to that described in section 312 for the sewage sludge ash A
significant decrease in the BET surface area (reduced by 10-50) as well as a decrease in the
pore volume was found when promoters were incorporated into the catalysts This could be
attributed to plugging of part of the micropores in the support due to an excess of metal
loading Adding Ca resulted in the greatest loss of surface area as well as the greatest average
pore size and the smallest pore volume The surface area was also reduced when the
calcination temperature was increased from 700 to 900 ordmC Pore volume did not show evident
changes in any case but the pore diameter increased with the calcination temperature
probably as a result of sintering of metal particulates
Table 33 Textural properties of the fresh calcined catalysts
Final calcination temperature NiAl2O3 NiCaAl2O3 NiFeAl2O3 NiCuAl2O3 NiMnAl2O3
Surface area (m2middotg-1)
700 ordmC 1205 617 1081 1051 966 900 ordmC 964 527 726 624 712
Pore volume (cm3middotg-1)
700 ordmC 035 024 031 032 028 900 ordmC 033 022 025 025 025
Average pore size nm)
700 ordmC 111 151 110 118 113 900 ordmC 132 163 135 157 137
As an example Fig 32 shows the XRD patterns of two of the fresh catalysts calcined at
700 and 900 ordmC (NiMnAl2O3 and NiCuAl2O3) The XRD patterns show quite low crystallinity
of the samples Non-significant differences are found in the patterns of the catalysts prepared
at the same calcination temperature regardless of the metal added as promoter All the
samples calcined at 700 ordmC showed wide and asymmetric peaks mainly corresponding to the γ-
Al2O3 phase As an exception the NiCuAl2O3 sample calcined at 700 ordmC showed two
diffraction peaks at 356ordm and 388ordm corresponding to CuO Peaks corresponding to other
expected metal oxides such as NiO were not clearly shown by this technique because of the
wide peaks forming the pattern When the calcination temperature was increased to 900 ordmC
the width of the peaks decreased pointing to more crystalline phases NiAl2O4 was the main
phase detected in all the samples calcined at 900 ordmC Other aluminates might also be present in
the catalysts but both the aluminates and the γ- Al2O3 are spinel-type phases and their X-ray
patterns are very similar so their presence cannot be confirmed by the results of this
technique alone
20 Materials and methods
Figure 32 XRD patterns of the fresh samples of NiMnAl2O3 and NiCuAl2O3 calcined at 700 and 900ordmC
32 Experimental facilities and procedures
321 Gasification setup
Sewage sludge and char gasification experiments were carried out in a lab-scale fluidized
bed reactor operating at atmospheric pressure Fig 33 shows a diagram of the gasification
experimental setup
Sewage sludge fast pyrolysis in which the used char was obtained as solid by-product was
conducted in a similar experimental setup to that shown in Fig 33 N2 was used as inert
atmosphere and fluidizing agent providing a fluidizing velocity around 8 times greater than the
minimum fluidization rate The pyrolysis temperature was 530 ordmC The average residence time
of the solid-byproduct and the produced gases and vapors in the reactor was around 8 min and
1 s respectively
20 30 40 50 60 70 80 90
γ-Al2O3 NiAl2O4
NiMnAl2O3 (700 oC)
2θ
NiAl2O4 γ-Al2O3
NiMnAl2O3 (900 oC)
20 30 40 50 60 70 80 90
CuO γ-Al2O3
NiAl2O4
NiCuAl2O3 (700 oC)
2θ
NiAl2O4 γ-Al2O3
NiCuAl2O3 (900 oC)
Materials and methods 21
Figure 33 Lab-scale gasification experimental setup
The gasifier was a tubular reactor (1270 mm height) made of refractory steel (AISI 310)
divided into two parts a bed zone (40 mm inner diameter) and a freeboard zone (70 mm inner
diameter) The reactor was heated by an electrical furnace with three different heating zones
which can be controlled independently (bed free-board and cyclone) Temperature in each
zone was measured by a K-type thermocouple A feeding system composed of a hopper and a
screw-feeder controlled by a variable frequency drive allowed continuous feed of the solid raw
material The solid feed rate was around 21 gmiddotmin-1 in all the gasification experiments
(minimum value that could be obtained with the variable frequency drive) The feed pipe was
cooled down by an air flow through an outer jacket to prevent reactions occurring out of solid
bed The solid bed at the beginning of the experiment was composed of ash from previous
gasification tests (around 120 g) When the amount of accumulated solid in the reactor
exceeded the height of the bed zone (310 mm) excess solid left the reactor by overflow
through a lateral pipe and was collected in a separate vessel
Different mixtures of steam and air (or enriched air) were used as gasifyingfluidizing
agent The air stream was enriched with pure oxygen in the experiments with the highest
demand of oxygen to avoid an abrupt change in the fluidization rate Fluidizing velocity was 5-7
22 Materials and methods
times greater than the minimum fluidization rate during sewage sludge gasification and 2-3
times greater during char gasification The lower organic content in the char is the reason for
this difference
The flow rate of gases (air and oxygen) was adjusted by means of mass flow controllers
while the required flow of liquid water was accurately adjusted by a HPLC pump Water was
evaporated before entering the reactor Most of the gasifying agent was fed through the
distribution plate located at the bottom of the reactor but part of the required air (around one
third) was diverted to the screw feeder to improve the movement of the solid through the
feeding pipe The residence time of the produced vapors and gases was around 7-8 s in the
case of sewage sludge gasification and a bit higher during char gasification (17-18 s) because
of the lower flow rate of gasifying agent used
The product gas passed through a cyclone and a hot filter both at 450 ordmC in which the
solid particles swept by the gas were collected Next gases and vapors passed through two ice-
cooled condensers where water and condensable organic compounds (tar) were collected A
cotton filter was situated after the condensers in order to remove small particulates and
aerosols still present in the gas The volume of particle-free and tar-free gas was measured
with a volumetric meter (G4 Gallus 2000 gas meter) and its composition was analyzed on-line
using a micro gas chromatograph (Agilent 3000-A) which determined the volume percentages
of H2 N2 CO CO2 CH4 C2H4 C2H6 C2H2 and H2S The experiments were carried out during 60-
90 min to ensure the stationary state (Aznar et al 2007)
The produced amount of solid and liquid products was determined after the experiments
by gravimetry that is by weight difference of the collecting devices before and after the
experiment Methanol was used to wash the condensers when collecting the liquid Water
content in the condensed fraction was analyzed off line by Karl Fischer titration (Mettler
Toledo V20 Karl Fischer titrator) so that tar fraction could be determined by difference The
amount of organic carbon in condensates can also give an idea of the tar content This carbon
content was determined with an analyzer of total organic carbon (TOC-L CSHCSN analyzer
Shimadzu) which measures the amount of CO2 produced during the catalytic combustion of
the liquid sample Tar composition from sewage sludge gasification was qualitatively
determined by gas chromatography combined with mass spectroscopy and flame ionization
detection (Agilent 5975C inert GCMSD complemented by Agilent 7890A GC system) Ultimate
analyses of the solid by-products were performed with a Leco TruSpec Micro elemental
analyzer and their ash contents were determined according to a standard method (ISO 1171-
1976)
Materials and methods 23
322 Desulphurization setup
Desulphurization tests were performed at atmospheric pressure in a fixed-bed quartz
tubular reactor of 1 cm inner diameter and 40 cm length A schematic diagram of the
experimental setup is shown in Fig 34
Figure 34 Experimental setup for desulphurization tests
The reactor was packed with 1 g of the sorbent material (sewage sludge combustion ash
or gasification ash) which was supported on a fibreglass fleece located 185 cm from the top
of the reactor The system was electrically heated to the desired temperature A K-type
thermocouple was used to measure and control the temperature in the middle of the solid
bed Two different inlet gases containing 5000 ppmv of H2S were used to evaluate the effect of
the gas components on the desulphurization process (Table 34) The gas containing only H2S
and N2 allows the assessment of the desulphurization process without involving any
interference from other gases while the synthetic gasification gas simulates the composition
of the gas from sewage sludge gasification thus representing more real conditions for H2S
removal
Table 34 Composition of the gas mixtures used in the desulphurization tests (dry basis)
Gas component H2SN2 mixture (vol )
Synthetic gasification gas (vol )
CO -- 100 CO2 -- 150 H2 -- 100
CH4 -- 40 C2H6 -- 02 C2H4 -- 15 C2H2 -- 02 H2S 05 05 N2 995 586
24 Materials and methods
The flow rate of the H2S-containing gases was adjusted to 50 mLSTPmiddotmin-1 by means of a
previously calibrated mass flow controller Steam was also added in some tests in order to
evaluate its effect on the desulphurization ability of the solids The required flow of liquid
water was accurately adjusted by a HPLC pump and was evaporated before entering the
reactor The weight hourly space velocity during the experiments ranged between 37 h-1 and
47 h-1 depending on the steam feed rate The gas-solid contact time was chosen after
preliminary experiments and based on earlier studies (Garciacutea et al 2011)
The composition of the outlet gas was continuously analyzed during the experiments by
means of a gas micro chromatograph (Agilent 3000) The H2S outlet flow was calculated from
the gas composition data taking N2 of the gas mixture as an internal standard Initially the
established experimental time was two hours but reaction time had to be extended in some
cases to 240 or 390 min in order to reach the H2S breakthrough time
Different tolerable levels of sulphur in gases can be found in the literature depending on
the gas application For instance the H2S concentration limit ranges between 20 and 750 ppmv
for gas turbine applications (Meng et al 2010) An intermediate value of 100 ppmv has been
used in this work to define the H2S breakthrough time
After the desulphurization tests sulphur content in the ash samples was determined with
a Leco TruSpec Micro elemental analyzer One of the ash samples was morphologically and
chemically characterized by scanning electron microscopy combined with energy dispersive X-
ray spectroscopy (SEMEDX) and by X-ray photoelectron spectroscopy (XPS) A FEI Inspect F50
microscope (Advanced Microscopy Laboratory of the Instituto de Nanociencia de Aragoacuten) was
used for the SEMEDX analysis External metal coating was not applied to the solid sample The
back-scattered electron imaging mode was used in the EDX analysis (acquisition time of 1 min)
The XPS measurements were performed with a Kratos AXIS ultra DLD spectrometer
(Laboratory of Microstructural Characterization and Spectroscopy of the Instituto de
Nanociencia de Aragoacuten) by using monochromatic Al Kα (14866 eV) X-ray source and a
chamber pressure of around 3middot10-8 Pa The quantification of the XPS spectra was carried out
with the help of the CasaXPS software and spectra were deconvoluted by applying Gaussian-
Lorentzian line-shapes with Shirley-type background
323 Catalyst testing setup
This part of the work was developed during a research stay at the VTT-Technical Research
Centre of Finland Tar reforming activity and stability of several Ni-based catalysts prepared
with different promoters were evaluated The activity tests were performed in a lab-scale
Materials and methods 25
fixed-bed quartz reactor (1 cm inner diameter with 04 cm thermocouple pocket) operating at
atmospheric pressure and in a temperature range of 700-900 ordmC The diagram of the
experimental setup is shown in Fig 35 The reactor was packed with 2 g of the solid powder
catalyst (supported on a quartz frit) and placed in an electrical furnace The reaction
temperature was monitored with a K-type thermocouple located in the middle of the catalyst
bed
Figure 35 Experimental setup for catalyst activity tests
A mixture of various gases (CO CO2 H2 N2 C2H4 C2H4 and steam) was used to simulate the
composition of a typical gas resulting from the air-steam gasification of sewage sludge
Benzene toluene and naphthalene were used as model compounds to simulate the presence
of tar in the gasification gas with a concentration of 15 gmiddotm-3STP The composition of the
resulting gas is shown in Table 35 The gas feed rate was adjusted to 1 LSTPmiddotmin-1 by means of
independent mass flow controllers The liquid reactants (water and the model tar mixture)
were fed independently through HPLC pumps and vaporized before entering the reactor All
the gas lines were heated to 200 ordmC to avoid condensation of the vapors The gas space velocity
(STP) was around 25000 h-1
Table 35 Composition of the synthetic gas used in catalysts testing (wet basis)
Gas component vol H2O 30 CO 66 CO2 125 H2 136 CH4 22 C2H4 11 H2S 03 N2 334 Tar content 15 gmiddotm-3
STP Tar composition toluenenaphthalenebenzene 801010 wt
26 Materials and methods
The product gas from the reactor was analyzed on-line with a gas chromatograph (Agilent
7980A) equipped with a flame ionization detector (FID) calibrated for benzene toluene
naphthalene methane and ethylene (analysis every 33 min) The condensable compounds
were then removed by a cold trap consisting of isopropanol and water in series in an ice bath
The flow rate and temperature of the dry gas thus obtained were measured and the gas was
directed to an on-line gas analyzer (Sick Maihak S710) which continuously measured the
volumetric composition of the permanent gas components (CO CO2 and H2) The accurate
composition of the inlet gas was also determined before each test using the same equipment
Possible thermal cracking of tar compounds was taken into account by performing a blank run
with a SiC bed
The used samples of catalyst were characterized by X-ray diffraction and ultimate analysis
using the same equipment that was mentioned in section 312 for characterization of sewage
sludge ash
33 Operating conditions and experimental design
331 Gasification experiments
The influence of some operating factors on the distribution of products and quality of the
gas obtained from the gasification of sewage sludge and char has been evaluated and
compared using a two-level factorial experimental design (2k design where k indicates the
number of factors studied and 2k represents the number of runs) This experimental design
allows the assessment of the effect of the operating conditions as well as of the existence of
interactions between the factors which occur when a factor influences a response variable in a
different way depending on the value of another factor
The analyzed factors were in both cases the following (i) gasification temperature (770-
850 oC) (ii) gasifying ratio (GR) which is the ratio between the mass flow rate of gasifying
agent (H2O+O2) and the feed rate of biomass on a dry and ash-free basis (08-11 gmiddotg-1daf) (iii)
nature of the gasification medium represented by the H2OO2 molar ratio (1-3) The total flow
rate of gasifying agent was kept constant when the H2OO2 molar ratio was modified These
three factors together with their respective ranges of study were chosen according to other
published studies concerning air-steam gasification of biomass in a fluidized bed reactor
(Campoy et al 2009 Gil et al 1997 Lv et al 2004 Pinto et al 2003)
The operating conditions in the gasification tests are shown in Table 36 Three replicates
at the average points in the studied range of the factors were carried out in order to evaluate
Materials and methods 27
both the experimental error and the linearitycurvature in the response of the analyzed
variables (experiments 9 10 and 11 in sewage sludge gasification and 20 21 and 22 in char
gasification) The two levels of the factors in the 2k experimental designs are usually
represented as -1 (for the first or the lowest level) and +1 (for the second or the highest level)
using 0 for the center points (average values)
Table 36 Operating conditions in the gasification tests
SEWAGE SLUDGE GASIFICATION
Exp Coded values (T GR H2OO2)
T (oC) GR (gmiddotg-1daf) H2OO2 ER ()
SB (g H2Omiddotg-1
daf) vol O2 in
airenriched air 1 111 850 11 3 17 071 21 2 -111 770 11 3 17 071 21 3 1-11 850 08 3 12 052 21 4 -1-11 770 08 3 12 052 21 5 11-1 850 11 1 32 039 33 6 -11-1 770 11 1 32 039 33 7 1-1-1 850 08 1 23 027 27 8 -1-1-1 770 08 1 23 027 27
91011 000 810 095 2 19 052 23
CHAR GASIFICATION
Exp Coded values (T GR H2OO2)
T (oC) GR (gmiddotg-1daf) H2OO2 ER ()
SB (g H2Omiddotg-1
daf) vol O2 in
airenriched air 12 111 850 11 3 17 071 27 13 -111 770 11 3 17 071 27 14 1-11 850 08 3 12 052 21 15 -1-11 770 08 3 12 052 21 16 11-1 850 11 1 32 039 40 17 -11-1 770 11 1 32 039 40 18 1-1-1 850 08 1 23 027 33 19 -1-1-1 770 08 1 23 027 33
202122 000 810 095 2 19 052 29 ER () = fraction of the stoichiometric oxygen that has been actually fed SB = ratio between the mass flow rate of steam and the feed rate of biomass on a dry and ash-free basis
Experimental results obtained for each response variable have been statistically analyzed
by means of analysis of variance (ANOVA) which compares the experimental variance
associated with error with the variance caused by the modification of the factors This
comparison is performed using the F-test (Fischerrsquos test) and allows to discriminate whether
the observed effect is statistically significant compared to error with a predetermined
confidence level (95 in this case) The Design-Expertreg 7 software (from Stat-Ease Inc) was
used for the analyses The evolution of each response variable with the variation of the factors
28 Materials and methods
can be empirically modeled using only the significant effects shown by the ANOVA analysis
according to the following equation
RV = α+β1middotF1+β2middotF2+β3middotF3+β12middotF1middotF2+β13middotF1middotF3+β23middotF2middotF3+β123middotF1middotF2middotF3 (eq 31)
where RV represent an experimental value of the response variable α is the average value
of the whole set of experimental results obtained for this response variable Fi is the coded
value of factor ldquoirdquo βi is the linear coefficient associated to factor ldquoirdquo βij is the coefficient
associated to the interaction of factors ldquoirdquo and ldquojrdquo (synergic or antagonistic effect) and β123 is
the coefficient that represents a simultaneous interaction between the three factors
Coefficients of the non-significant effects do not appear in this equation Although the
existence of a significant curvature prevents the use of the linear regression model the
relative influence of the factors can be still assessed by comparison of coefficients βi
Expressing the equation in terms of coded values of the factors (from -1 to +1) enables an
easier identification of the term with the greatest influence on the response variable the
higher the absolute value of the coefficient the more influential the factor
332 Desulphurization tests
Table 37 summarizes the operating conditions for the desulphurization tests The
operating factors studied were the following (i) temperature (600-800 ordmC) (ii) type of sewage
sludge ash (combustion ash or gasification ash) (iii) type of H2S-containing gas (H2SN2 mixture
or synthetic gasification gas) (iv) steam concentration in the inlet gas (0-30 vol ) which
resulted in a H2OH2S mass ratio ranging between 0 and 45 gH2Omiddotg-1H2S Experimental
uncertainty was evaluated through three replicates performed when feeding the H2SN2
mixture and operating at the average values of temperature (700 ordmC) and H2OH2S mass ratio
(225 gmiddotg-1) The impact of the factors on the H2S breakthrough curves was statistically analyzed
by ANOVA analysis (confidence level of 95 for the F-distribution)
Blank tests (with no bed material) were also performed at the different temperatures and
gas atmospheres in order to assess any side effect of the experimental setup caused for
example by the reaction of H2S with the hot metal parts (made of steel) at the reactor inlet
and outlet
Materials and methods 29
Table 37 Operating conditions in the desulphurization tests
Experiment Origin of sewage sludge ash
Synthetic gas mixture
Temperature (ordmC)
H2OH2S mass ratio
Test duration (min)
1 Combustion H2SN2 600 0 300 2 Combustion H2SN2 800 0 390 3 Combustion H2SN2 600 45 120 4 Combustion H2SN2 800 45 120
567 Combustion H2SN2 700 225 120 8 Gasification H2SN2 600 0 120 9 Gasification H2SN2 800 0 390
10 Gasification H2SN2 600 45 120 11 Gasification H2SN2 800 45 120
121314 Gasification H2SN2 700 225 120 15 Blank run H2SN2 600 0 120 16 Blank run H2SN2 800 0 120 17 Blank run H2SN2 600 45 120 18 Blank run H2SN2 800 45 120 19 Blank run H2SN2 700 225 120 20 Combustion Gasification gas 600 0 240 21 Combustion Gasification gas 800 0 240 22 Combustion Gasification gas 600 45 120 23 Combustion Gasification gas 800 45 120 24 Gasification Gasification gas 600 0 120 25 Gasification Gasification gas 800 0 240 26 Gasification Gasification gas 600 45 120 27 Gasification Gasification gas 800 45 120 28 Blank run Gasification gas 600 0 120 29 Blank run Gasification gas 800 0 240 30 Blank run Gasification gas 600 45 120 31 Blank run Gasification gas 800 45 120
The desulphurization tests performed at the laboratory were theoretically simulated to
determine the maximum amount of H2S that could be removed from the gas from a
thermodynamic point of view HSC Chemistryreg 61 software was used for this purpose This
software uses the Gibbs energy minimization method to calculate the amounts of products at
equilibrium in isothermal and isobaric conditions The reaction system (temperature pressure
reactants and expected species to be part of the products) must be specified for the
calculations The expected species to be part of the products were the following fed gases
sulphur-containing gases (H2S SO2 COS) elemental sulphur (S) calcium-containing species
(Ca CaO CaS CaCO3 CaSO4 and CaSO3) and iron-containing species (Fe FexOy FexSy Fex(SO4)y
and Fex(SO3)y)
333 Catalyst testing
Table 38 summarizes the operating conditions in the activity tests of the Ni-based
catalysts The effects of the added promoter (Ca Cu Fe or Mn) calcination temperature (700
30 Materials and methods
or 900 ordmC) and reduction procedure were evaluated In order to obtain the metallic nickel
active sites the catalysts were reduced in some cases before starting the activity tests in a
H2N2 atmosphere at 900 ordmC for 1 h (1 LSTPmiddotmin-1 5050 vol ) In other cases the reduction
was performed during the activity test itself in the reducing atmosphere created by the
synthetic gasification gas
Table 38 Operating conditions in the catalyst activity tests
Experiment Catalyst Calcination temperature (ordmC)
Reduction pretreatment in H2N2 atmosphere
1 NiAl2O3 900 yes 2 NiCaAl2O3 900 yes 3 NiCuAl2O3 900 yes 4 NiFeAl2O3 900 yes 5 NiMnAl2O3 900 yes 6 NiAl2O3 900 no 7 NiCaAl2O3 900 no 8 NiCuAl2O3 900 no 9 NiFeAl2O3 900 no
10 NiMnAl2O3 900 no 11 NiAl2O3 700 no 12 NiCaAl2O3 700 no 13 NiCuAl2O3 700 no 14 NiFeAl2O3 700 no 15 NiMnAl2O3 700 no
The effect of the reaction temperature on the catalyst activity was also studied by
modifying the temperature during the tests The temperature was programmed according to
the following ramp 900-850-800-900-750-700-900 ordmC Each temperature was maintained for
35 h so the duration of the complete runs was 245 h The loss of catalyst activity may be
evaluated by comparing the conversion results obtained in the successive steps at 900 ordmC
Discussion of the main results 31
4 DISCUSSION OF THE MAIN RESULTS
The main results obtained in the different studies of this PhD thesis are discussed in this
section Experimental results from sewage sludge gasification and char gasification as well as
the impact analysis of the operating factors are presented in section 41 An energetic
assessment of both gasification stages is included in section 42 also considering the pyrolysis
process in which the char is produced and the prior thermal drying of the sewage sludge
Section 43 shows the results concerning desulphurization of various synthetic gases by using
ash derived from the combustion and gasification of sewage sludge Finally the results
obtained in the activity tests of nickel-catalysts prepared with different promoters for tar
reforming under a H2S-containing atmosphere are detailed in section 44
41 Sewage sludge and char gasification
Gas is the product of interest from gasification so most of the analyzed response variables
are related to this product gas production (dry gas yield) and specific yield of each non-
condensable gas compound (gmiddotkg-1solid daf) gas composition (H2CO and COCO2 ratios) tar
content in the gas lower heating value of the gas (LHVgas) and gasification efficiency The solid
yield and the carbon distribution into the different products (solid gas and tar) was also
determined after the gasification runs All these experimental results corresponding to sewage
sludge gasification are summarized in Table 41 as a function of the operating conditions
The solid yield defined as the mass (g) of solid product obtained per kilogram of sewage
sludge fed varied between 356 and 407 gmiddotkg-1SS Typical values for other types of biomass such
as wood or straw are usually below 80 gmiddotkg-1 (McKendry 2002a) The high ash content in the
sewage sludge (39 wt ) explains this difference The solid yield was found to be lower than
the initial ash content of sewage sludge in some cases suggesting some transformations and
release of ash compounds to gas phase
The distribution of initial carbon content of sewage sludge into the different products was
determined from different analysis of the products the fraction of carbon remaining as solid
(2-23) was calculated from the ultimate analyses of the solid samples the fraction of carbon
forming tar (4-7) was calculated from the total organic carbon present in the liquid fractions
and the fraction of carbon converted into non-condensable gases (613-897) was calculated
from the gas composition experimental results Carbon mass balances closed to 78-95 The
possible formation of light hydrocarbons not detected by the micro-GC (C3Hx C4Hx) or the
32 Discussion of the main results
poor solubility of tar compounds in the aqueous solutions for determination of the total
organic carbon may explain the observed defect carbon
Table 41 Experimental results from sewage sludge gasification
Temperature (ordmC) 850 770 850 770 850 770 850 770 810
Gasifying ratio (gmiddotg-1daf) 11 11 08 08 11 11 08 08 095
H2OO2 molar ratio 3 3 3 3 1 1 1 1 2
Solid yield (gmiddotkg-1SS) 368 401 401 407 356 392 384 400 382 plusmn 1
Carbon fraction remaining as solid () 3 17 8 23 2 9 2 12 7 plusmn 1
Carbon fraction converted into gas () 764 613 651 618 897 748 831 657 731 plusmn 08
Carbon fraction forming tar () 4 7 5 5 4 7 4 6 5 plusmn 1
Dry gas yield (N2-free m3
STPmiddotkg-1SS)
072 051 065 053 072 052 071 049 061 plusmn 001
Dry gas yield (N2-free m3
STPmiddotkg-1SS daf)
132 094 120 097 132 096 130 089 113 plusmn 001
Tar content (gmiddotm-3STP) 19 44 19 44 12 22 11 45 15 plusmn 1
Gas composition (dry basis) H2 (vol ) 242 184 251 204 180 110 206 136 193 plusmn 01
CO (vol ) 87 57 102 73 116 70 141 77 94 plusmn 01
CO2 (vol ) 171 186 126 155 207 238 160 198 181 plusmn 02 CH4 (vol ) 31 35 36 41 26 28 29 34 33 plusmn 01
C2Hx (vol ) 17 21 14 22 13 16 14 20 17 plusmn 02
H2S (vol ) 044 038 033 033 044 042 038 031 040 plusmn 002 N2 (vol ) 449 514 468 502 453 534 445 532 478 plusmn 02
Mass yield of each gas compound (gmiddotkg-1
SS daf)
H2 518 318 504 365 386 201 431 232 372 plusmn 04 CO 260 137 287 182 351 179 414 183 253 plusmn 2
CO2 806 707 557 608 980 962 736 740 767 plusmn 7 CH4 53 49 59 59 45 41 49 46 51 plusmn 1 C2Hx 50 52 40 55 40 42 40 48 47 plusmn 4 H2S 160 112 114 100 161 131 135 90 130 plusmn 08
H2CO molar ratio in the product gas 279 325 246 281 154 157 146 177 206 plusmn 001
COCO2 molar ratio in the product gas 051 030 081 047 056 029 088 039 052 plusmn 001
LHV gas (MJmiddotm-3STP) 59 53 62 60 52 41 60 49 56 plusmn 01
Cold gasification efficiency () 658 477 647 553 584 391 647 434 558 plusmn 08
Data in the last column show the ldquomean plusmn standard deviationrdquo obtained from the 3 replicates at the center point
The dry gas yield from sewage sludge varied between 089 and 132 m3STPmiddotkg-1
SS daf
expressing the volume of gas on N2-free basis and the mass of sewage sludge on a dry and ash-
free basis (daf) These values are in the same range as those obtained from the gasification of
Discussion of the main results 33
other types of biomass under similar operating conditions (Campoy et al 2009 Gil et al
1999b Pinto et al 2003) As usual in a biomass gasification process the main gases produced
during sewage sludge gasification were H2 CO CO2 and light hydrocarbons CH4 being the most
abundant hydrocarbon H2S was also released during the process due to the presence of
sulphur-compounds in the sewage sludge (Table 31) As a consequence of the use of air the
presence of N2 accounted for 45-55 of the total gas volume The same amount of N2 was fed
in all tests to avoid different dilution effects
Considerable differences in the gas composition have been found The concentration of H2
(110-251 vol ) CO (57-141 vol ) CO2 (126-238 vol ) or CH4 (26-41 vol ) was
doubled or halved depending on the operating conditions These volume percentages led to a
H2CO molar ratio in the exit gas ranging between 146 and 325 and a COCO2 molar ratio
ranging between 029 and 088 The H2CO molar ratio is an important parameter in view of
possible end uses of the gas and values close to 2 are usually required in processes such as
methanol production or Fischer Tropsch synthesis (Wender 1996) On the other hand the
COCO2 ratio shows how the carbon content of sewage sludge is distributed among both
compounds Higher COCO2 are preferred from an energy standpoint
The tar content in the product gas varied from 11 to 45 gmiddotm-3STP The highest values (22-45
gmiddotm-3STP) were obtained at the lowest operating temperature (770 ordmC) while the lowest values
(lt 19 gmiddotm-3STP) are among the reported values for bubbling fluidized bed gasifiers (Han and Kim
2008)
The lower heating value of the gas (LHVgas) was calculated as Σ (xi middot LHVi) where xi and
LHVi are the volumetric fraction and the lower heating value (MJmiddotm-3STP) of each gaseous
compound respectively The LHV of the product gas of sewage sludge gasification ranged
between 41 and 62 MJmiddotm-3STP The prevailing dilution effect of N2 causes the gas calorific value
from the air-steam gasification to be in the same range as that reported for biomass
gasification when only air is used as gasifying agent (McKendry 2002b)
The cold gasification efficiency is defined as the ratio between the energy contained in the
gas product (gas volume middot LHVgas) not taking into account its sensible heat and the energy
contained in the mass of sewage sludge fed (mass of SS middot LHVSS) A wide range of experimental
values was obtained ranging from 391 and 658
Experimental results corresponding to char gasification obtained at the same operating
conditions that those of sewage sludge gasification are summarized in Table 42
34 Discussion of the main results
Table 42 Experimental results from char gasification
Temperature (ordmC) 850 770 850 770 850 770 850 770 810
Gasifying ratio (gmiddotg-1daf) 11 11 08 08 11 11 08 08 095
H2OO2 molar ratio 3 3 3 3 1 1 1 1 2 Solid yield (gmiddotkg-1
char) 757 785 750 785 731 771 752 813 775 plusmn 2
Carbon fraction remaining as solid () 20 41 25 43 15 26 19 41 34 plusmn 3
Carbon fraction converted into gas () 71 56 62 48 83 67 72 56 62 plusmn 2
Carbon fraction forming tar () 13 07 29 33 10 57 32 58 28 plusmn 07
Dry gas yield (N2-free m3
STPmiddotkg-1char)
036 027 031 024 035 028 032 024 029 plusmn 001
Dry gas yield (N2-free m3
STPmiddotkg-1char daf)
147 112 130 099 146 115 131 100 121 plusmn 001
Tar content (gmiddotm-3STP) 4 3 9 13 3 20 10 22 9 plusmn 2
Gas composition (dry basis)
H2 (vol ) 293 263 278 248 215 190 220 202 252 plusmn 06
CO (vol ) 195 120 202 128 227 140 237 152 159 plusmn 02
CO2 (vol ) 189 242 162 208 226 295 185 241 219 plusmn 01
CH4 (vol ) 076 091 077 092 059 070 064 084 088 plusmn 001
C2Hx (ppmv) 150 190 160 220 180 220 150 200 180 plusmn 10
H2S (vol ) 025 012 014 007 017 008 012 006 010 plusmn 001
N2 (vol ) 313 365 349 407 325 367 350 396 361 plusmn 06 Mass yield of each gas compound (gmiddotkg-1
char daf)
H2 567 421 499 368 416 308 397 300 428 plusmn 03
CO 529 268 506 266 615 318 598 316 379 plusmn 10 CO2 808 853 637 679 960 1055 735 786 821 plusmn 20
CH4 117 117 110 109 91 90 92 99 119 plusmn 01
C2Hx 041 042 041 047 050 050 039 042 042 plusmn 004 H2S 82 32 41 18 57 23 37 15 28 plusmn 01
H2CO molar ratio in the product gas 150 220 138 193 095 136 093 133 158 plusmn 004
COCO2 molar ratio in the product gas 103 049 125 062 100 047 128 063 073 plusmn 001
LHV gas (MJmiddotm-3STP) 596 471 587 465 544 409 563 443 507 plusmn 007
Cold gasification efficiency () 629 411 572 376 574 362 553 357 470 plusmn 06
Data in the last column show the ldquomean plusmn standard deviationrdquo obtained from the 3 replicates at the center point The amount of tar was approximated to the amount of organic carbon detected in the condensed liquid fraction
The solid yield from char gasification varied from 731 to 813 gmiddotkg-1char though this solid
was mainly composed of ash (93-96 wt according to ISO-1171-1976 analysis)
Regarding the distribution of initial carbon content into the different products the
fraction of initial carbon remaining as solid after char gasification ranged between 15 and 43
Discussion of the main results 35
while the maximum value for sewage sludge gasification was about 23 (Table 41) This
difference may be explained by the different structure of the carbonaceous matter in the solid
Only 15 of the carbon content was in the form of fixed carbon in the sewage sludge while
this value reached 59 in the char (Table 31) Gasification reactions of solid carbon are slower
than devolatilization and gas phase reactions which results in both an increased fraction of
initial carbon remaining as solid and a reduced fraction of carbon forming non-condensable
gases during char gasification (48-83) However results of carbon conversion during char
gasification can be recalculated considering both the previous pyrolysis of sewage sludge and
the subsequent gasification of char as a whole The initial carbon content of sewage sludge
(295 wt ) has to be used as a reference for recalculating the carbon conversion in this case
Given a char yield of around 52 wt during sewage sludge pyrolysis (Gil-Lalaguna et al
2010) the fraction of initial carbon remaining as solid after sewage sludge pyrolysis and char
gasification is reduced to 4-11 thus improving in some cases the conversion results obtained
from the direct gasification of sewage sludge
Tar formation during char gasification declined compared to the results obtained in
sewage sludge gasification but in general this reduction was not as high as the reduction in
the initial volatile content of both raw materials (Table 31) The tar content in the gas from
char gasification was as low as 3-4 gmiddotm-3STP under some operating conditions while the lowest
value achieved in sewage sludge gasification was about 11-12 gmiddotm-3STP
The dry gas yield from char gasification varied between 024 and 036 m3STPmiddotkg-1
char (N2-free
basis) or between 040 and 052 m3STPmiddotkg-1
char if N2 is included in the gas volume The
production of gas per kilogram of raw material during char gasification was halved compared
to that from sewage sludge gasification (051-072 m3STPmiddotkg-1
SS N2-free basis) due to the higher
ash content of char The production of gas during the previous fast pyrolysis process of sewage
sludge (around 007 m3STPmiddotkg-1
SS) is not high enough to offset this difference since that process
was focused on maximizing the liquid fraction On the other hand the gas yield expressed on a
dry and ash-free basis for the solid was improved by gasifying the char (099-147 m3STPmiddotkg-1
char)
instead of the sewage sludge (089-132 m3STPmiddotkg-1
SS) since carbon content was more
concentrated in the dry and ash-free fraction of the solid after pyrolysis (064 g Cmiddotg-1char daf vs
054 g Cmiddotg-1SS daf) The dry gas yield (N2-free and daf basis) from the gasification of char derived
from sewage sludge is in the same range as those obtained from lignocellulosic chars such as
bagasse char (Chaudhari et al 2003) or char derived from ramie residues (He at al 2012)
Regarding gas composition H2 (190-293 vol ) CO (120-237 vol ) CO2 (162-295 vol
) CH4 (059-092 vol ) and N2 (313-407 vol ) were the main gases detected by the micro
36 Discussion of the main results
GC during char gasification The CO mass yield (in terms of gmiddotkg-1solid daf) was clearly improved by
gasifying char instead of sewage sludge (45-85 higher in most cases) while CH4 production
was reduced by 75-82 Variations in the production of H2 and CO2 (in terms of gmiddotkg-1solid daf)
were not so significant However it should be noted that gas yields calculated with respect to
the whole feedstock and not on a dry and ash-free basis are considerably higher for sewage
sludge gasification The H2CO molar ratio obtained from char gasification (093-220) was
lower than that obtained from sewage sludge gasification (146-325) while the COCO2 molar
ratio was higher for char gasification (047-128) than for sewage sludge gasification (029-
088) Thus CO formation was only favored rather than CO2 formation (COCO2 gt 1) under
some operating conditions in char gasification The lower heating value of the gas (409-596
MJmiddotm-3STP) and the cold gasification efficiency (362-629) of char gasification were in the
same range as those obtained for sewage sludge gasification
The influence of temperature (T) gasifying ratio (GR) and composition of the gasification
medium (H2OO2 molar ratio) on the aforementioned response variables has been statistically
evaluated Tables 43 and 44 show the linear regression coefficients obtained from the ANOVA
analysis of the experimental results of sewage sludge gasification and char gasification
respectively These coefficients are expressed in terms of coded values for the factors
Table 43 Linear regression coefficients (β) from ANOVA analysis of sewage sludge gasification results
Average βT βGR βH2OO2 βT-GR βT-H2OO2 βGR-H2OO2 βT-H2OO2-GR Curvature
Carbon fraction remaining as solid () 901 -576 -174 329 -150
Carbon fraction converted into gas () 7248 633 333 -607
Dry gas yield (N2-free m3
STPmiddotkg-1SS daf)
112 017 002 002 -002 003
Tar content (gmiddotm-3STP) 2703 -1191 -278 435 315 262 -289
Yield of gaseous compounds (gmiddot kg-1SS daf)
H2 3703 906 -137 568
CO 25010 7879 -1751 -3267 -503 -2195 965 CO2 76342 10157 -9253 2159 -1493 1591
CH4 5023 149 -313 480 C2Hx 4613 -328 335 H2S 1266 173 156 062
H2CO molar ratio in the product gas 221 -014 008 062 002 -006 011 -005
COCO2 molar ratio in the product gas 052 016 -011 -004 -003
LHV gas (MJmiddotm-3STP) 549 037 -031 040 -017
Cold gasification efficiency () 5512 851 347
non-significant term significant curvature
Discussion of the main results 37
Table 44 Linear regression coefficients (β) from ANOVA analysis of char gasification results
Average βT βGR βH2OO2 βT-GR βT-H2OO2 βGR-H2OO2 βT-H2OO2-GR Curvature
Carbon fraction remaining as solid () 3013 -916 -330 359
Carbon fraction converted into gas () 6429 776 482 -516
Dry gas yield (N2-free m3
STPmiddotkg-1char daf)
123 016 008
Tar content (gmiddotm-3STP) 1052 -397 -315 -340 345
Yield of gaseous compounds (gmiddotkg-1char daf)
H2 4093 603 186 543 033 090 118 CO 42716 13509 -3461 -978 CO2 81596 -2909 10491 -6995 -1875 CH4 1033 101 032 C2Hx 044 -003 H2S 354 162 103
H2CO molar ratio in the product gas 148 -026 031
COCO2 molar ratio in the product gas 085 029 -010 -003 001
LHV gas (MJmiddotm-3STP) 509 063 020 008
Cold gasification efficiency () 4790 1027 145 177 048 083
non-significant term significant curvature
Temperature is the most influential factor on the fraction of initial carbon remaining as
solid after the gasification processes This carbon fraction can be practically halved by
increasing the gasification temperature from 770 to 850 ordmC (Fig 41) Furthermore the positive
coefficient associated to the H2OO2 ratio (βH2OO2 gt 0) suggests a higher reactivity of carbon
with O2 compared to its reactivity with steam The similar conclusion was drawn by Nowicki et
al (2011) by conducting some gasification tests in a thermobalance with different gas
atmospheres (CO2 H2O and O2) The gaseous mixtures containing O2 were found to be the
most efficient gasifying agents In the case of sewage sludge gasification the effect of the
gasification medium seems to be mitigated at high temperatures (Fig 41a) As expected the
increase in the gasifying ratio also leads to a reduction in the fraction of initial carbon that
remains in the solid by-product after gasification (βGR lt 0)
38 Discussion of the main results
(a) Sewage sludge gasification (b) Char gasification
Figure 41 Fraction of initial carbon remaining in the solid-byproduct after (a) sewage sludge gasification and (b) char gasification (GR = 095 gmiddotg-1 solid daf) Error bars in the figures represent the least significant
difference (LSD)
Experimental results show a clear connection between the fraction of carbon remaining as
solid and the fraction of carbon converted into gases the lower the carbon fraction remaining
as solid the higher the carbon fraction converted into non-condensable gases Hence the
formation of non-condensable carbonaceous gases is positively affected by the temperature
(βT gt 0) and by the gasifying ratio (βGR gt 0) and negatively affected by the H2OO2 ratio (βH2OO2
lt 0) (Fig 42)
(a) Sewage sludge gasification (b) Char gasification
Figure 42 Fraction of initial carbon converted into gas during (a) sewage sludge gasification and (b) char gasification (GR = 095 gmiddotg-1 solid daf) Error bars in the figures represent the least significant difference
Dry gas yield is strongly dependent on the reaction temperature (βT gt 0) During the
gasification process gas is produced at different stages that are favored at higher
770 810 850
50
60
70
80
90
Carb
on fr
actio
n co
nver
ted
in
to g
as (
)
Temperature (ordmC)
770 810 850
50
60
70
80
90
Carb
on fr
actio
n co
nver
ted
in
to g
as (
)
Temperature (ordmC)
H2OO2 = 1 H2OO2 = 3 H2OO2 = 2
770 810 850 0
10
20
30
40
50
Temperature (ordmC)
Carb
on fr
actio
n re
mai
ning
as
solid
()
770 810 850 0
10
20
30
40
50 Ca
rbon
frac
tion
rem
aini
ng
as so
lid (
)
Temperature (ordmC)
H2OO2 = 3 H2OO2 = 2 H2OO2 = 1
Discussion of the main results 39
temperatures such as the initial stage of pyrolysis the cracking and steam reforming of tars
and the endothermic gasification reactions of the carbonaceous solid (Pinto et al 2003) The
increase of the gasifying ratio (GR) is also favorable for gas production (βGR gt 0) especially
during char gasification whereas the composition of the gasification medium does not exert a
significant influence on the dry gas yield obtained from the gasification of either char or
sewage sludge Hence the above mentioned negative effect of H2OO2 on the gasified carbon
fraction does not result in a significant variation of the dry gas yield Different production rates
of H2 may explain this disagreement since as discussed below H2 formation is promoted by
increasing the H2OO2 ratio thus counteracting the observed decrease in the production of
carbon-containing gases at high H2OO2 ratios Some synergic and antagonistic effects
between the factors are statistically significant on the evolution of the gas yield from sewage
sludge gasification but these are much less important than the individual effect of gasification
temperature
Temperature is also the most influential factor on the tar content of sewage sludge
gasification gas while the three studied factors exert similar effects during char gasification
The rise in temperature not only favors the gasification reactions of solid carbon but also the
tar reforming reactions rate The tar content is also reduced by increasing the gasifying ratio
(βGR lt 0) but this effect practically disappears when operating at high temperatures or high
H2OO2 ratios during sewage sludge gasification The composition of the gasification medium
has shown opposite effects on the tar content of the gases produced during sewage sludge
gasification and char gasification the decrease of the H2OO2 ratio is favorable for reducing the
tar content during sewage sludge gasification (βH2OO2 gt 0) while this reduction involves an
increase of the tar content during char gasification (βH2OO2 lt 0) In this latter case the effect of
the H2OO2 ratio is significantly mitigated at high temperatures Thus the surface response of
tar content cannot be modelled by a horizontal plane due to the importance of the interaction
between the temperature and the H2OO2 ratio (Fig 43)
40 Discussion of the main results
Figure 43 Tar content (gmiddotm-3STP) in the product gas from char gasification (GR = 095 gmiddotg-1
daf)
The production or mass yield (gmiddotkg-1 solid daf) of each non-condensable gaseous compound
was calculated from gas volume and gas composition data As can be noted from β coefficients
shown in Tables 43 and 44 gasification temperature is again the most influential factor on
the production of both H2 and CO These gases are involved in many reactions both as
reactants and as products but the temperature rise seems to enhance their formation rather
than their consuming reactions (βT gt 0) Tar reforming also contributes to the increase of H2
and CO production at high temperatures (eq 29 and 210) Other studies in the literature have
shown negligible variations or even the opposite trend on the CO production when varying the
temperature (Gil et al 1997 Lv et al 2004) which reveals the importance of the biomass
nature and the operating conditions on the evolution of this compound
Although to a lesser extent the production of H2 is also improved by increasing the H2OO2
ratio (βH2O2 gt 0) unlike the CO yield which decreases with H2OO2 (βH2O2 lt 0) Both observed
trends are consistent with the water-gas shift reaction (eq 26) which seems one of the most
representative reactions for a steam gasification process for temperatures up to 830 ordmC (Franco
et al 2003) Franco et al (2003) found that at higher temperatures (830-900 ordmC) solid-gas
reactions such as the water gas reaction (eq 23) and the Boudouard reaction (eq 24)
appeared to prevail which can explain the positive effect of temperature on CO formation
This positive effect is mitigated by increasing the presence of steam (Fig 44a) probably due
to the shift of the water-gas shift reaction to CO consumption Besides increasing the presence
of steam the increase of the H2OO2 ratio entails a reduction of the availability of oxygen in
770
790
810
830
850 1
15
2
25
3
6
10
14
18
22
Tar
con
tent
T (ordmC)
H2OO2
Discussion of the main results 41
the gasification medium which should mitigate the combustion reactions However the
negative effect of H2OO2 on the production of CO suggests that steam plays a more significant
role than O2 in CO consumption
(a) CO yield from sewage sludge gasification (b) CO2 yield from sewage sludge gasification
Figure 44 (a) CO yield (gmiddotkg-1SS daf) and (b) CO2 yield (gmiddotkg-1
SS daf) during sewage sludge gasification
As observed in Fig 44b the CO2 yield is mainly affected by the gasifying ratio (βGR gt 0)
The increase of the gasifying ratio (GR) involves larger amounts of both oxygen and steam in
the gasification medium so combustion reactions as well as other reactions promoted by the
presence of steam in which CO2 is produced (such as the water-gas shift reaction) take place
to a greater extent The negative effect of the H2OO2 ratio (βH2OO2 lt 0) reveals that
combustion reactions are the main source of CO2 On the other hand temperature only
appears as a statistically significant term on CO2 production in the case of char gasification In
this case the temperature rise leads to a reduction in the formation of CO2 (βT lt 0) and as
discussed above favors the production of CO thus suggesting an increased reactivity of char
with CO2 at high temperatures (Boudouard reaction eq 24) The surface response of neither
CO production nor CO2 production can be modelled as horizontal planes due to the significant
synergic and antagonistic effects between the factors (Fig 44)
As can be noted from β coefficients resulting from the ANOVA analyses (Tables 43 and
44) the production of light hydrocarbons (CH4 and C2Hx) is mainly influenced by the
composition of the gasification medium The higher the H2OO2 ratio the higher the light
hydrocarbons yield (βH2OO2 gt 0) This suggests a higher reactivity of these gases with O2
compared to their reactivity with steam Furthermore the formation of CH4 via the
methanation reaction (eq 25) may be partly responsible of the positive effect of the H2OO2
ratio since the presence of H2 in the gasification medium is increased at high H2OO2 ratios In
770 790
810 830
850
10 15
20 25
30
150
210
270
330
390
CO
yie
ld (g
kg
SS
daf
)
T (ordmC) H2OO2
080
088
095
103
110
10 15
20 25
30
580
680
780
880
980
CO
2 yi
eld
(g
kg S
S d
af)
GR (gg daf )
H2OO2
42 Discussion of the main results
the case of sewage sludge gasification the CH4 yield is negatively affected by the gasifying
ratio (βGR lt 0) and unlike the results shown by other authors (Kim et al 2001 Pinto et al
2003) positively influenced by the temperature (βT gt 0) maybe because of the increase of the
methanation reaction rate On the other hand the temperature rise leads to a reduction in the
C2Hx formation during the sewage sludge gasification (βT lt 0) due to the increased reforming
reactions rate Lastly the production of H2S is favored in both gasification processes at high
temperature (βT gt 0) and high gasifying ratio (βGR gt 0)
In addition to the production of each gaseous compound the evolution of the H2CO and
COCO2 molar ratios in the exit gas has been analyzed due to their importance for the gas use
as a feedstock for chemical processes The H2CO molar ratio in the product gas can be
enhanced by increasing the H2OO2 ratio used as gasification medium (βH2OO2 gt 0) andor
reducing the gasification temperature (βT lt 0) The relative influence of both factors is quite
similar in the gasification of char (Table 44) but the gasifying agent plays a more important
role in the gasification of sewage sludge (Table 43) On the other hand the higher the
gasification temperature the higher the COCO2 molar ratio in the exit gas (βT gt 0) The
COCO2 ratio can also be improved by reducing the gasifying ratio (βGR lt 0) while the
composition of the gasification medium does not exert a significant influence on this ratio
ANOVA analyses have revealed that temperature plays the most important role in the
evolution of the gas heating value when char is gasified while the three studied factors exert
similar effects on the gas heating value from sewage sludge gasification (see β coefficients in
Tables 43 and 44) As shown in Fig 45 and in contrast to some results shown in the
literature (Pinto et al 2003) temperature shows a positive effect on the gas lower heating
value from both gasification processes (βT gt 0) The hydrocarbons content of the gas decreases
with increasing temperature so a reduction in the gas heating value would be expected
However CO2 concentration in the gas decreases with temperature in a greater proportion
than the hydrocarbons content (see gas composition data in Tables 41 and 42) thus reducing
the dilution effect of the gas and resulting in an increase of the gas heating value The LHVgas
can also be improved by increasing the H2OO2 ratio used as gasifying agent (βH2OO2 gt 0) since
both an increase in the hydrocarbon content and a decrease in the CO2 fraction are found
when the presence of steam in the gasification medium is increased and the availability of
oxygen is reduced simultaneously In the case of sewage sludge gasification the gasifying ratio
(GR) exerts a significant negative effect on the LHVgas (βGR lt 0) and as shown in Fig 45a the
effect of the gasification medium is significantly mitigated at high temperatures
Discussion of the main results 43
(a) Sewage sludge gasification (b) Char gasification
Figure 45 Lower heating value of the gas (MJmiddotm-3STP) from (a) sewage sludge gasification and (b) char
gasification (GR = 095 gmiddotg-1solid daf)
As occurring with the dry gas yield and the gas heating value the cold gasification
efficiency is mainly dependent on gasification temperature This efficiency can be improved in
20 percentage points by increasing the gasification temperature from 770 to 850 ordmC in both
processes (βT gt 0) This improvement is based on the increase of both the LHVgas and the gas
yield The H2OO2 ratio positively affects the cold gasification efficiency (βH2OO2 gt 0) to a lesser
extent than the temperature In the case of char gasification the gasifying ratio (βGR gt 0) and
some synergic and antagonistic effects between the factors are statistically significant on the
gasification efficiency evolution but these are much less important than the individual effect
of temperature
The composition of the tar formed during sewage sludge gasification was analyzed by gas
chromatography (MSFID GC) Fig 46 shows a representative chromatogram with the main
compounds detected in most of the tar samples Some researchers have divided tar
components into different families based on their molecular weight and atomic composition
(Li and Suzuki 2009a) A similar classification of tar compounds has been considered in this
work (i) light aromatics with 1 ring (styrene) (ii) PAH compounds with 2 or 3 rings (indene
naphthalene n-methyl-naphthalene biphenyl biphenylene fluorene anthracene and
phenantrene) (iii) heterocyclic aromatics containing N (including n-methyl-pyridine
benzonitrile n-methyl-benzonitrile quinoline n-methyl-quinoline indole n-phenyl-pyridine
n-naphthalenecarbonitrile benzoquinoline and 5H-indeno[12-b]pyridine) (iv) heterocyclic
aromatics containing O (phenol and benzofuran) and (v) compounds containing S (2-
benzothiophene and propanenitrile 33-thiobis-)
770 810 850 42 45 48 51 54 57 60 63
LHV
gas (
MJ
m3 ST
P)
Temperature (ordmC)
770 810 850 42 45 48 51 54 57 60 63
LHV
gas (
MJ
m3 ST
P)
Temperature (ordmC)
H2OO2 = 2 H2OO2 = 3 H2OO2 = 1
44 Discussion of the main results
Figure 46 Total ion chromatogram (TIC) of a tar sample obtained at 850 ordmC GR = 08 and H2OO2 = 3
The areas of the main peaks detected by the GC-FID have been used to compare the
composition of the different samples Hence results presented in this work do not represent
the actual composition of the tar samples but the peak areas are useful for comparing how
the factors influence the formation of each family of compounds The percentages of the GC-
FID-areas are shown in Table 45
Table 45 Percentage of area in the GC-FID signal of each family of tar compounds
Temperature (ordmC) 850 770 850 770 850 770 850 770 810 Gasifying ratio (gmiddotg-1
daf) 11 11 08 08 11 11 08 08 095 H2OO2 molar ratio 3 3 3 3 1 1 1 1 2 Light aromatics (1 ring) 97 122 44 134 51 99 36 98 65 plusmn 18 PAH compounds (2-3 rings) 91 91 338 514 399 415 450 227 360 plusmn 88
N-aromatics 746 682 571 687 465 440 447 605 500 plusmn 144 O-aromatics 30 77 034 77 06 21 10 60 26 plusmn 08
S-compounds 36 29 44 51 79 25 57 10 49 plusmn 01
Data in the last column show the ldquomean plusmn standard deviationrdquo obtained from the 3 replicates at the center point
Temperature and H2OO2 ratio are the only significant factors affecting tar composition
Light aromatics and O-aromatics are the most sensitive families to temperature and their
fractions are found to decrease with temperature Similar results have been reported by other
researchers (Ponzio et al 2006) showing that phenolic compounds paraffines olefins and
alkylated aromatics are easily cracked at high temperatures The S-compounds fraction has
Discussion of the main results 45
been found to increase with temperature maybe as a result of the aforementioned decrease
in the fractions of other compounds On the other hand N-aromatics and PAH fractions are
the most sensitive families to the H2OO2 ratio The increase in this ratio leads to a decrease in
the fraction of PAHs so the presence of steam seems to prevent polymerization reactions Tar
molecular weight depends on the presence of H free radicals which is related to the steam
added during gasification (Qin et al 2010) According to Corella et al (1999) tars generated in
gasification with steam are easier to crack than tars generated in gasification with air A
simultaneous increase in the fraction of N-aromatics has been observed but this may be only a
consequence of the aforementioned decrease in the PAH fraction
Theoretical production of gases based on equilibrium data
The experimental production of gaseous compounds has been compared to the expected
production of gases at equilibrium conditions which has been determined with the HSC
Chemistry 61 software This software uses the Gibbs energy minimization method to calculate
the amounts of products at equilibrium in isothermal and isobaric conditions Therefore the
reaction system (temperature pressure feed rate of gasifying agent amount of C H O S and
N that form part of the raw material and expected species to be part of the products) must be
specified for the calculations The same operating conditions studied at the laboratory were
defined for the equilibrium simulations H2 CO CO2 CH4 H2S and NH3 are the main
compounds in the equilibrium gases Neither light hydrocarbons (except CH4) nor tar should be
present in the equilibrium product gas
Fig 47 (sewage sludge gasification) and Fig 48 (char gasification) compare equilibrium
and experimental yields of H2 CO CO2 and CH4 Experimental yields of H2 and CO are clearly
below their corresponding equilibrium data while the experimental yields of CO2 and CH4 are
above the equilibrium values These differences reveal that chemical equilibrium was not
reached during the gasification runs Insufficient residence time of gases and vapors in the
reactor andor mass transfer control or kinetic control in gas-solid reactions could explain why
the chemical equilibrium was not reached in the gasification experiments
46 Discussion of the main results
0
20
40
60
80
100
120
5 6 7 8 CP 1 3 2 4
g H 2
kg
SS d
af
Experiment number
0
200
400
600
800
1000
1200
6 5 2 1 CP 8 7 4 3
g CO
kg
SS d
af
Experiment number
0
200
400
600
800
1000
1200
3 4 7 8 CP 1 2 5 6
g CO
2 k
g SS
daf
Experiment number
0
10
20
30
40
50
60
g CH4 kg SS daf
(equilibrium values)
g CH
4 k
g SS
daf
(e
xper
imen
tal v
alue
s)
Experiment number
0
1
2
3
4
5
6
5 1 7 6 3 CP 2 8 4
CP Center point (mean value from experiments 9 10 11)
Figure 47 Theoretical ( ) and experimental ( ) production of H2 CO CO2 and CH4 during sewage sludge gasification
0
20
40
60
80
100
120
16 17 19 18 CP 15 14 12 13
g H 2
kg
char
daf
Experiment number
0200400600800
100012001400
17 16 13 12 CP 19 15 18 14
g CO
kg
char
daf
Experiment number
0
200
400
600
800
1000
1200
14 15 18 19 CP 12 13 16 17
g CO
2 k
g ch
ar d
af
Experiment number02468
101214
16 12 18 17 CP 14 13 19 15
g CH
4 k
g ch
ar d
af
Experiment number CP Center point (mean value from experiments 20 21 22)
Figure 48 Theoretical ( ) and experimental ( ) production of H2 CO CO2 and CH4 during char gasification
Discussion of the main results 47
The influence of the studied operating factors on the evolution of the equilibrium gas
yields has been statistically analyzed by ANOVA analysis The linear regression coefficients (β)
in terms of coded values of the factors are shown in Table 46
Table 46 Linear regression coefficients (β) from ANOVA analysis of equilibrium gas yields
Average βT βGR βH2OO2 βT-GR βT-H2OO2 βGR-H2OO2 βT-H2OO2-GR Curvature
Sewage sludge gasification
H2 (gmiddotkg-1SS daf) 9504 -115 -320 1480 237
CO (gmiddotkg-1SS daf) 87307 2465 -8722 3638 1583
CO2 (gmiddotkg-1SS daf) 61016 -3650 13861 -5831 -880 752 -2441
CH4 (gmiddotkg-1SS daf) 107 -081 -056 042 043 -033 -017 014
Char gasification
H2 (gmiddotkg-1char daf) 7197 1358 -082 181
CO (gmiddotkg-1char daf) 123658 1995 -11574 3059 1657
CO2 (gmiddotkg-1char daf) 40457 -2471 18733 -5124 -1069 -2455
CH4 (gmiddotkg-1char daf) 325 -241 -198 115 145 -079 -054 032
Some important differences with respect to the experimentally observed effects of the
factors can be highlighted (i) Temperature is not further the most influential factor on the
equilibrium production of H2 and CO the H2OO2 ratio exerts the most significant effect on the
equilibrium production of H2 while the gasifying ratio is the most influential factor on the
equilibrium CO production (ii) Equilibrium H2 production is not positively affected by the
temperature but slightly reduced (sewage sludge gasification) or not affected (char
gasification) (iii) Equilibrium CO production is not negatively influenced by the H2OO2 ratio
but positively affected (iv) Equilibrium CH4 production is not positively affected by the
temperature but strongly decreased with temperature Unlike occurring in a kinetically
controlled process in which the temperature rise favors both mass transfer and chemical
reaction rates the endothermic or exothermic nature of reactions has to be considered to
explain the evolution of the gases at equilibrium conditions
48 Discussion of the main results
42 Energetic assessment
This section presents the energetic assessment of two potential thermo-chemical
treatments for sewage sludge management which include the studied stages of gasification
One of these processes involves the direct gasification of sewage sludge while the other
includes the pyrolysis of sewage sludge and the gasification of the char produced in it (Fig 49)
(a) Air-steam gasification of sewage sludge
(b) Pyrolysis of sewage sludge + air-steam gasification of char
Figure 49 Schematic overview of two thermo-chemical processes for sewage sludge management
The individual energy requirement of each stage (drying pyrolysis and gasification) has
been calculated to determine the overall energy demand of each process
Experimental data concerning mass yields of solid liquid and gas products and some of
their properties (higher heating value and ultimate analysis of liquid and solid streams specific
heat capacity of solids and gases composition) have been used for calculating the enthalpy of
reaction of the gasification and pyrolysis processes As detailed below the other required
properties of gases and liquid compounds were taken from literature Some general
assumptions and simplifications have been considered
- Adiabatic reactor
- The specific heat capacity (Cp) of the solid materials (char and ash) was considered
constant with temperature Although these specific heat capacities were experimentally
measured by differential scanning calorimetry (DSC) the function of Cp with temperature
could not be obtained in the whole range of temperature required in each case because
Dried sewage sludge
Pyrolysis
Wet sewage sludge
Thermal drying
Gas
Char
Bio oil Water
Air-steam Gas
Ash
Tar Water
Gasification
Dried sewage sludge
Gasification
Wet sewage sludge
Thermal drying
Gas
Ash
Tar Water
Air-steam
Discussion of the main results 49
of operational limitations of the equipment used Thus constant Cp values corresponding
to intermediate values of temperature were used for calculations
- As detailed below the composition of the liquid fractions obtained from both processes
(bio-oil from pyrolysis and tar from gasification) was simplified in order to obtain some of
their required properties (specific heat capacity enthalpy of vaporization and standard
enthalpy of formation in the case of tar) from literature
If negligible heat losses are considered in the reactor (adiabatic reactor) the enthalpy of
reaction for the pyrolysis or the gasification processes (Q) can be calculated from the
enthalpies of the streams entering (ΔHin) and exiting (ΔHout) the reactor as follows
Q (MJ middot kg-1) = ΔHout - ΔHin (eq 41)
According to eq 41 Q lt 0 corresponds to an exothermic process while Q gt 0 refers to an
endothermic process The total enthalpy of each stream (ΔH) can be calculated as follows
ΔH (MJ middot kgminus1) = sum mi ∙ (ΔHfio +
i int CpiTTref
(T) dT) (eq 42)
where
- mi is the mass yield of each gas compound liquid stream or solid stream (kgmiddotkg-1solid) 1 kg of
raw material (dried sewage sludge or char) has been used as calculation basis
- Tref is the reference temperature (298 K) and T is the temperature of each stream (K) All
the inlet streams were at ambient temperature (298 K) except the steam used in the
gasification experiments which was generated and fed at 448 K
- ΔHordmfi is the standard enthalpy of formation (MJmiddotkg-1) of each compound at the reference
temperature (298 K) The ΔHordmf data of the gases involved in the process can be easily found
in the literature (Perry and Green 1999) The ΔHordmf data corresponding to the solid materials
(sewage sludge pyrolysis char and solid by-products from gasification) and to the pyrolysis
liquid fractions have been calculated from their ultimate analyses and heating values
according to the following equation
∆Hfi
o = sum mj j ∙ ∆Hfj
o + HHVi (eq 43)
where lsquojrsquo represents each product derived from the complete combustion of the material
(CO2 H2O SO2 and NO) mj is the mass of each combustion gas produced per kilogram of
material ΔHordmfj is the standard enthalpy of formation of each combustion gas and HHVi is
the higher heating value of the solid or liquid phases
50 Discussion of the main results
- Cpi (T) is the specific heat capacity as a function of temperature (MJmiddotkg-1middotK-1) Cp of the solid
materials was experimentally measured by DSC while Cp (T) of the gases and vapors
involved in the processes were taken from literature (ChemSpider database Harrison and
Seaton 1988 Perry and Green 1999) If the range of temperature in the integral of eq 42
involves a phase change both the enthalpy of vaporization and the Cp of the condensable
compounds in the liquid phase have to be included to calculate the total enthalpy These
data can be also found in the literature (ChemSpider database Chueh and Swanson 1973)
Specific assumptions and energetic results for the pyrolysis of sewage sludge and the
gasification of sewage sludge and char are detailed below
Enthalpy of reaction of sewage sludge pyrolysis
The experimental product yields obtained in the pyrolysis of sewage sludge (Gil-Lalaguna
et al 2010) as well as the properties of the products required for the energy balance are
included in Table 47 The liquid collected after condensation of the vapors was separated into
three phases light organic phase (LOP) heavy organic phase (HOP) and aqueous phase (AP)
The higher heating values (HHV) of char and liquid phases were experimentally measured
with a calorimeter while the gas heating value was calculated from its composition ΔHordmf of
solid and liquid products was calculated from eq 43 while ΔHordmf of gases was taken from
literature (Perry and Green 1999)
As mentioned above constant values of Cp were used for the solids 115middot10-3 MJmiddotkg-1middotK-1
for sewage sludge and 121middot10-3 MJmiddotkg-1middotK-1 for char (measured at an average temperature 300
ordmC) The composition of the liquid phases was simplified to determine their Cp by considering
only its water content and one representative organic compound cholest-4-ene for the LOP 3-
methyl-phenol for the HOP and acetic acid for the AP These were some of the main
compounds with the largest chromatographic area detected by GC-MS The mass of the
representative species in each phase was equated to the whole organic fraction in the phase
The global Cp of each phase (both in liquid and gas phases) can be estimated as a weighted
average of the specific heat capacities of water (or steam) and of the representative organic
compound of each phase
In the same way the enthalpy of vaporization of the liquid phases (ΔHvap) was estimated as
a weighted average of the enthalpies of vaporization of water and of the representative
organic compound of each phase at their boiling temperatures
Discussion of the main results 51
Table 47 Yields and properties of the products of sewage sludge fast pyrolysis
Mass yield (wt ) Composition HHV
(MJmiddotkg-1) ΔHordmf
(MJmiddotkg-1) Cp (T)
(kJmiddotK-1middotkg-1) ΔHvap
(MJmiddotkg-1)
Char
519 plusmn 07 Elemental analysis (wt wet basis) C1549 H097 N185 S035 52 plusmn 02 -118 082 (25 ordmC)
121 (300 ordmC) ---
Non-condensable gas (N2-free)
101 plusmn 09
( mass fraction) CO2 743 plusmn 09 CO 132 plusmn 01 H2 17 plusmn 01
CH4 38 plusmn 01 C2H6 14 plusmn 02 C2H4 14 plusmn 01 H2S 43 plusmn 09
80 plusmn 03 -739 118 (25 ordmC) 156 (530 ordmC) ---
Light organic phase (LOP)
22 plusmn 02
Elemental analysis (wt wet basis) C8592 H1183 N180 S027
Water 0 wt
Organics 100 wt
4310 plusmn 004 -174 185 (liquid) 307 (530 ordmC) 018
Heavy organic phase (HOP)
94 plusmn 02
Elemental analysis (wt wet basis) C6954 H897 N944 S124
Water 64 plusmn 03 wt
Organics 936 plusmn 03 wt
32 plusmn 2 -349 213 (liquid) 236 (530 ordmC) 055
Aqueous phase (AP)
208 plusmn 02
Elemental analysis (wt wet basis) C1117 H1045 N652 S037
Water 738 plusmn 04 wt
Organics 262 plusmn 04 wt
57 plusmn 03 -1244 359 (liquid) 212 (530 ordmC) 177
Experimental uncertainty is expressed as mean plusmn standard deviation (two replicates were performed)
Thus considering all the product streams at the pyrolysis temperature at the reactor exit
(T = 803 K) the enthalpies of the streams entering (ΔHin) and exiting (ΔHout ) the reactor have
been found to be -328 MJmiddotkg-1SS and -313 MJmiddotkg-1
SS respectively (eq 42) The difference of
these enthalpies (eq 41) results in an energy demand of 015 MJmiddotkg-1SS for the thermal
decomposition of sewage sludge during the pyrolysis process This value is significantly lower
than decomposition heats found in the literature for other types of biomass For example 03
MJmiddotkg-1 has been reported for the pyrolysis of crop residues (Mangaro et al 2011) The higher
ash content in sewage sludge which is hardly decomposed during the process can explain this
difference
If the outlet stream of gases and vapors is cooled down to ambient temperature (298 K) in
order to take advantage of their sensible and latent heats the net heat of the process is
52 Discussion of the main results
reduced to -070 MJmiddotkg-1SS This indicates that in the absence of significant heat losses and
keeping in mind the considered assumptions the pyrolysis of sewage sludge could be an
autothermal process if the cooling and condensation of the gases and vapors could be
efficiently used
Enthalpy of reaction of sewage sludge gasification and char gasification
The mass yields of the products obtained in the gasification of sewage sludge and char
under the different operating conditions were presented in Tables 41 and 42 respectively
The collected amount of tar has been simplified to an equimolar mixture of benzene
naphthalene and pyridine (typical compounds found in the tar mixtures from sewage sludge
gasification) for determination of their Cp and ΔHvap On the other hand Cp (T) of all the solid
gasification by-products has been approximated to that of sewage sludge ash because these
solids were mainly composed of ash (gt 93 wt in most cases) The constant value used for
sewage sludge ash was 107middot10-3 MJmiddotkg-1middotK-1 (experimentally measured at an average
temperature 500 ordmC)
The enthalpy of reaction of the gasification of sewage sludge and char under the tested
operating conditions can be calculated according to eq 41 The results are depicted in Fig
410 considering the products at the gasification temperature (770-850 ordmC) and as a function
of the gasification medium ER (equivalence ratio) and SB (steam to biomass daf mass ratio)
ER = 32 ER = 23 ER = 19 ER = 17 ER = 12 SB = 039 SB = 027 SB = 052 SB = 071 SB = 052
Figure 410 Enthalpy of reaction of (a) sewage sludge gasification and (b) char gasification based on experimental data (products at the gasification temperature)
As shown in Fig 410 the enthalpy of reaction ranges between -261 and 129 MJmiddotkg-1SS for
sewage sludge gasification and between -023 and 120 MJmiddotkg-1char for char gasification under
the tested conditions Thus enthalpy of reaction is more affected by the operating conditions
Char gasification
-3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=770ordmC
T=850ordmC
T=850ordmC T=810ordmC T=770ordmC
T=770ordmC
(b)
Enthalpy of reaction (MJkg char)
Sewage sludge gasification
-3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=850ordmC
T=850ordmC
T=770ordmC T=810ordmC
T=770ordmC
T=770ordmC
(a)
Enthalpy of reaction (MJkg SS)
Discussion of the main results 53
in the case of sewage sludge gasification Despite the lower organic content in the char than in
the sewage sludge the external energy demand for gasifying 1 kg of char is higher than that
for gasifying 1 kg of sewage sludge For instance the enthalpy of reaction of sewage sludge
gasification at ER = 17 and SB = 071 is 064 MJmiddotkg-1 at 850 ordmC and 017 MJmiddotkg-1 at 770 ordmC
while it reaches 100 MJmiddotkg-1 and 078 MJmiddotkg-1 for char gasification at the same temperature
respectively This behavior could be related to the observed changes in the carbonaceous
structure of sewage sludge after carrying out the pyrolysis process The fraction of volatile
matter in the sewage sludge was higher than in the char while the fraction of fixed carbon was
higher in the char (Table 31) Thus combustion reactions in gas phase which usually show less
diffusional resistance than the solid-gas reactions involve vaporized hydrocarbons during the
sewage sludge gasification while the main combustion reactions in gas phase during char
gasification involve gases such as H2 or CO (produced from fixed carbon reactions) whose
calorific value is lower than those of hydrocarbons As a consequence air-steam gasification of
char was an endothermic process under most of the experimental conditions used while air-
steam gasification of sewage sludge was an exothermic process when simultaneously working
with ER gt 19 and SB lt 052
If the outlet stream of gases and vapors is cooled down to ambient temperature (298 K) in
order to take advantage of their sensible and latent heats the net heat of the gasification
processes varies from -580 to -165 MJmiddotkg-1SS in the case of sewage sludge gasification and
from -117 to 026 MJmiddotkg-1char in the case of char gasification This shows that the required
energy for carrying out the endothermic gasification processes may be obtained from the
product gas either from its thermal energy (for example by using the product gas to preheat
the inlet air stream in a heat exchanger) or from the combustion of part of the gas In this
context a parameter called ldquogasification efficiencyrdquo has been defined as the fraction of the
energy initially contained in the raw material that could be still recovered from the product gas
after carrying out the gasification reaction
Gasiication eficency () =Energy recovery from gas minus Qgasiication minus Qsteam
LHVraw material∙ 100 (eq 44)
where
- ldquoEnergy recovery from gasrdquo includes the gas calorific value (Dry gas yield middot LHVgas) and the
sensible and latent heats of gases and vapors ( sum mi ∙i ∆Hcondi + int Cpi (T) where ldquoirdquo
represents each gas or vapor compound) A heat exchange efficiency of 70 has been
considered for the thermal energy recovery from gases
54 Discussion of the main results
- Qgasification is the enthalpy of reaction of the gasification stages considering the products at
the gasification temperature (Fig 410)
- Qsteam is the energy demand for heating and evaporating the inlet flow of water from 25 to
150 ordmC (236 MJmiddotkg-1H2O)
- LHVraw material is the energy contained in the raw material (sewage sludge or char) expressed
as its lower heating value (Table 31)
Gasification efficiency data obtained for sewage sludge gasification and char gasification
calculated according to eq 44 are presented in Table 48 Efficiency results vary from 58 to
87 for sewage sludge gasification and from 23 to 64 for char gasification Better efficiency
results have been obtained for sewage sludge gasification as a consequence of its lower
enthalpy of reaction and higher gas yield The gasification efficiency improves at higher
temperatures higher ER and lower SB
Table 48 Experimental data of sewage sludge gasification efficiency and char gasification efficiency
Temperature 850 770 850 770 850 770 850 770 810 ER () 17 17 12 12 32 32 23 23 19 SB (mass ratio daf basis) 071 071 052 052 039 039 027 027 052
Sewage sludge gasification
Energy recovery from gas (MJmiddotkg-1
SS) 1023 791 963 841 910 677 956 693 872 plusmn 012
Gasification efficiency () 74 58 65 61 87 75 82 65 71 plusmn 2
Char gasification
Energy recovery from gas (MJmiddotkg-1
char) 391 277 349 244 352 244 334 227 298 plusmn 005
Gasification efficiency () 51 32 40 23 64 49 54 33 40 plusmn 2
Theoretical enthalpy of reaction of the gasification of sewage sludge and char at
equilibrium conditions
The theoretical enthalpy of reaction for the air-steam gasification of sewage sludge and
char at equilibrium conditions has also been determined in order to evaluate the
thermodynamic restrictions of the processes under different scenarios HSC Chemistry 61
software was used to determine the mass flow rates of the products at equilibrium conditions
These theoretical enthalpies of reaction are depicted in Fig 411 simulating the same
operating conditions that were tested at the laboratory
Discussion of the main results 55
ER = 32 ER = 23 ER = 19 ER = 17 ER = 12 SB = 039 SB = 027 SB = 052 SB = 071 SB = 052
Figure 411 Enthalpy of reaction of (a) sewage sludge gasification and (b) char gasification based on
equilibrium data (products at the gasification temperature)
As can be seen in Fig 411 gasification of sewage sludge at equilibrium conditions only
results in an exothermic process when ER is increased up to 32 Char gasification is an
endothermic process in all the simulated cases
Comparison of Figs 410 and 411 shows that reaching the chemical equilibrium in both
gasification processes entails additional energy consumption The reason may be the
predominance of endothermic equilibrium reactions during the gasification such as the water
gas reaction (eq 23) the Boudouard reaction (eq 24) steam reforming (eq 27) and dry
reforming (eq 28) against the exothermic equilibrium reactions such as the water-gas shift
reaction (eq 26) These reactions occur to a greater extent at equilibrium which shows the
thermodynamic limit of the process However the gas heating value and the gas yield
calculated at equilibrium conditions are higher than those obtained experimentally so more
energy could be recovered from the equilibrium product gas (Table 49) This latter difference
outweighs the difference observed in the experimental and equilibrium data of enthalpy of
reaction so the gasification efficiency is improved at equilibrium conditions 90-94 for
sewage sludge gasification and 78-84 for char gasification (Table 49)
Char gasification
-3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=770ordmC
T=850ordmC
T=850ordmC T=810ordmC T=770ordmC
T=770ordmC
Enthalpy of reaction (MJkg char)
(b) Sewage sludge gasification
-3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=850ordmC
T=850ordmC
T=770ordmC T=810ordmC T=770ordmC
T=770ordmC
Enthalpy of reaction (MJkg SS)
(a)
56 Discussion of the main results
Table 49 Theoretical data of sewage sludge gasification efficiency and char gasification efficiency at equilibrium conditions
Temperature 850 770 850 770 850 770 850 770 810 ER () 17 17 12 12 32 32 23 23 19 SB (mass ratio daf basis) 071 071 052 052 039 039 027 027 052
Sewage sludge gasification
Energy recovery from gas (MJmiddotkg-1
SS) 1437 1411 1495 1472 1134 1109 1286 1264 1370
Gasification efficiency () 90 91 91 92 94 94 94 94 92
Char gasification
Energy recovery from gas (MJmiddotkg-1
char) 619 609 653 642 494 485 562 553 595
Gasification efficiency () 78 80 80 81 82 83 82 84 81
If the gasifier operates at autothermal conditions instead of being heated by external heat
transfer the gasification temperature is the output variable from balancing out the enthalpies
of the streams entering and exiting the gasifier (ΔHin = ΔHout assuming negligible heat losses)
The equilibrium temperature has been calculated under different gasification mediums
following an iterative method ΔHout depends on the mass of products (eq 42) and these in
turn depend on the gasification temperature (temperature has to be specified in the HSC
Chemistry software to calculate the amounts of products at equilibrium) Fig 412 shows the
evolution of the equilibrium temperature as a function of ER (equivalence ratio) and SC
(steam to carbon molar ratio) for the air-steam gasification of both sewage sludge and char
SC=0 SC=05 SC=1
Figure 412 Equilibrium temperature as a function of the equivalence ratio (ER) and the steam to carbon molar ratio (SC) during (a) sewage sludge gasification and (b) char gasification
The equilibrium temperature obviously increases with ER and decreases with SC It should
be noted that equilibrium temperature obtained at low ER is too low for a gasification process
(b) Char gasification
10 15 20 25 30 35 40 45 50 300
500
700
900
1100
ER ()
Equi
libriu
m te
mpe
ratu
re (ordm
C)
10 15 20 25 30 35 40 300
500
700
900
1100
(a) Sewage sludge gasification
ER ()
Equi
libriu
m te
mpe
ratu
re (ordm
C)
Discussion of the main results 57
so it does not make sense to think on a tar-free gas product under these conditions which has
been supposed to occur at equilibrium conditions
The required ER to maintain a specific reaction temperature during char gasification is
higher than that in sewage sludge gasification For instance an ER of 33 would be required
for autothermal operation of sewage sludge gasification at 800 ordmC and SC = 05 under
equilibrium conditions while this value reaches 45 in the case of char gasification The higher
the ER the greater the production of CO2 through combustion reactions The presence of CO2
in the gasification gas is undesirable since it implies both a dilution effect of the gas heating
value and a reduction in the formation of CO (production and consumption of CO and CO2 are
connected by reactions such as the water-gas shift or the Boudouard reaction) In addition to
the gas calorific value the H2CO ratio in the product gas is an important parameter for using
this gas as a feedstock in the synthesis of chemicals such as methanol or Fischer Tropsch fuels
Values of this ratio close to 2 are usually required in these processes (Wender 1996) For the
aforementioned example (ER of 33 for sewage sludge gasification and 45 for char
gasification to maintain 800 ordmC when SC = 05) 44 of the initial carbon contained in the
sewage sludge produces CO2 while this value reaches 52 in the case of char gasification
Both the heating value and the H2CO ratio in the product gas of sewage sludge gasification
(LHVgas = 427 MJmiddotm-3STP H2CO = 147) are higher than those obtained for char gasification
(LHVgas = 305 MJmiddotm-3STP H2CO = 089)
Sewage sludge drying
Prior to the thermo-chemical treatment of sewage sludge by means of pyrolysis or
gasification sewage sludge thermal drying allows reduction of water content in the waste
thus reducing the waste volume and facilitating handling of the biosolids The heat needed for
the sewage sludge thermal drying (Q drying) can be calculated as follows
Qdrying (MJ ∙ kg inal SSminus1 ) =
mdry SS ∙ CpSS + mH2OSS ∙ CpH2O(l) ∙ ∆T + mH2Oevap ∙ ∆HvapH2O
kg inal SSkg wet SS
(eq 45)
where
- mdry SS is the dry matter content of the wet sludge (kgmiddotkg-1wet SS)
- mH2OSS is the water content of the wet sludge (kgmiddotkg-1wet SS) After the mechanical
dewatering of sewage sludge by filter pressing or centrifugation (just before thermal
drying) moisture content of sewage sludge is around 70 (Manara and Zabaniotou 2012)
- ΔT is the temperature difference between the beginning and the end of the drying
process (from 25 to 100 ordmC)
58 Discussion of the main results
- CpSS is the specific heat capacity of the dried sewage sludge (115middot10-3 MJmiddotkg-1middotK-1) This
value was experimentally obtained at 25 ordmC by differential scanning calorimetry and has
been considered constant with temperature for calculations
- CpH2O(l) is the specific heat capacity of liquid water (418middot10-3 MJmiddotkg-1middotK-1) which is virtually
constant in the considered temperature range (Perry and Green 1999)
- mH2Oevap is the mass of evaporated water per kilogram of wet sludge (kgmiddotkg-1wet SS)
- ΔHvapH2O is the enthalpy of vaporization of water at the exit temperature (226 MJmiddotkg-1H2O
at 100 ordmC) (Perry and Green 1999)
Fig 413 shows the evolution of the heat needed for sewage sludge drying as a function of
the initial and final moisture contents based on calculations performed with eq 45 For
instance an energy input of 8 MJmiddotkg-1final SS is required for reducing the water content from 77
wt to 65 wt which represent the actual data of the wastewater treatment plant in
which the used sewage sludge was generated However the heat required for the sewage
sludge thermal drying is reduced by half if the initial moisture content is reduced from 77 to 65
wt This reduction could be achieved by improving the efficiency of the prior mechanical
dewatering of sewage sludge
5 10 15 20 25 30 35 400
2
4
6
8
10
Qdr
ying (
MJ
kg
final
SS)
Final moisture content ()
Figure 413 Heat demand for the thermal drying of sewage sludge as a function of the initial and final moisture contents
Energetic assessment of the two-stage and three-stages processes
The energy demand for the two-stage (sewage sludge drying + sewage sludge gasification)
and three-stage processes (sewage sludge drying + sewage sludge pyrolysis + char gasification)
is the sum of the net energy required or released in the individual stages (positive terms for
endothermic processes and negative values for exothermic processes) Note that energy
Initial moisture content 77 wt 70 wt 65 wt
Discussion of the main results 59
consumptions related to the use of pumps compressors motors etc have not been included
in this study Only energetic aspects related to the thermo-chemical stages have been
considered
- Sewage sludge drying Water content in the sewage sludge is assumed to be reduced from
65 wt (typical content of moisture in sewage sludge before its thermal drying) to 65 wt
during the thermal drying Thus Qdrying is around 44 MJmiddotkg-1dried SS (Fig 413)
- Sewage sludge pyrolysis As occurs in the gasification processes the energy contained in
the produced gases and vapors could be recovered to be used in the thermal
decomposition of sewage sludge itself and in the prior thermal drying Taking advantage of
the gas lower heating value and the sensible and latent heats of gases and vapors (given
70 of exchange efficiency for thermal energy recovery) a net heat of -117 MJmiddotkg-1SS is
obtained for the pyrolysis process The use of the calorific value of the liquid product (43
MJmiddotkg-1LOP and 32 MJmiddotkg-1
HOP) is not included in the energy balance as some important
properties such as its poor stability or its high nitrogen content must be improved facing
toward its use as fuel (Fonts et al 2012)
- Sewage sludge gasificationchar gasification The net heats of the gasification stages (based
on experimental data) can be calculated according to the numerator in eq 44 As the same
calculation basis is required for the comparison of the two-stage and three-stage processes
(1 kg of dried sewage sludge fed initially) data corresponding to the gasification of char
(MJmiddotkg-1char) must be turned into MJmiddotkg-1
SS by means of the char yield obtained in the
pyrolysis of sewage sludge (0519 kgcharmiddotkg-1SS)
Fig 414 shows the total energy demand for the two-stage and three-stage processes
considering the different experimental conditions used in the gasification stages
60 Discussion of the main results
ER = 32 ER = 23 ER = 19 ER = 17 ER = 12 SB = 039 SB = 027 SB = 052 SB = 071 SB = 052
Figure 414 Total energy demand for the (a) two-stage and (b) three-stage processes
The energy demand ranges between -243 and -581 MJmiddotkg-1SS for the two-stage process
(exothermic process) and between 157 and 264 MJmiddotkg-1SS for the three-stage process
(endothermic process) Thus energy contained in the product gas of sewage sludge
gasification would be enough to cover the energy demand for both the sewage sludge thermal
drying and the gasification process itself if this energy would be efficiently used Moreover it
should be noted that the flow of water required for the air-steam gasification of sewage sludge
could be directly obtained from the own moisture content of the sludge so the final moisture
of the sewage sludge after the drying stage could be increased up to 19-32 (depending on
the SB ratio defined in the gasification stage) Hence the energy demand during the thermal
drying stage could be reduced around 036-093 MJmiddotkg-1dried SS resulting in a more favorable
energy balance On the other hand the three-stage treatment is globally an endothermic
process (note that the use of the pyrolysis liquid calorific value is not considered) so that an
additional energy input would be needed for this treatment However assuming a direct and
efficient use of calorific value of organic liquid phases obtained from pyrolysis (392 MJmiddotkg-1SS)
a favorable energetic assessment is also obtained for the three-stage process with a total
energy release ranging from -235 to -128 MJmiddotkg-1SS (exothermic process) Thus the use of the
calorific value of the produced pyrolysis liquid is a key issue for reaching an autothermal three-
stage process
Lastly regarding the influence of the gasification conditions on the total energy demand of
the processes the best energetic results are obtained when operating simultaneously at the
highest gasification temperature (850 ordmC) the highest equivalence ratio (32) and a moderate
steam to biomass daf ratio (039)
Three-stage process
-3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=770ordmC
T=850ordmC
T=850ordmC T=810ordmC
T=770ordmC
T=770ordmC
(b)
Total energy demand (MJkg SS)
Two-stage process
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 T=850ordmC
T=850ordmC
T=770ordmC
T=850ordmC
T=850ordmC
T=770ordmC
T=810ordmC T=770ordmC
T=770ordmC
(a)
Total energy demand (MJkg SS)
Discussion of the main results 61
43 Desulphurization of different gas streams
This section presents the results of the desulphurization tests at high temperature (600-
800 ordmC) H2S was removed from different synthetic gases by using sewage sludge combustion
ash and gasification ash as sorbent materials One of these gases only contained H2S and N2
while the other was a synthetic gasification gas containing H2 CO CO2 CH4 C2H6 C2H4 C2H2
N2 and H2S (Table 34) Steam was added in some experiments together with gas in order to
evaluate its impact on the desulphurization performance
The H2S breakthrough curves obtained at the ash bed exit (H2S outlet flow as a function of
experiment time) are depicted in Fig 415 as a function of the type of gas atmosphere
temperature and type of ash
(a) Dry H2SN2 mixture b) Moist H2SN2 mixture (30 vol H2O)
(c) Dry synthetic gasification gas d) Moist synthetic gasification gas (30 vol H2O)
Gas feed Blank run 800 ordmC 600 ordmC
Sewage sludge combustion ash 800 ordmC 600 ordmC Sewage sludge gasification ash 800 ordmC 600 ordmC
Figure 415 H2S breakthrough curves evolution of the H2S flow rate leaving the reactor (mLSTPmiddotmin-1)
An exit gas essentially free of H2S was obtained during 300 min and 260 min by using the
combustion ash and the gasification ash at 800 ordmC under the dry H2SN2 gas atmosphere
0 50 100 150 200 250 300 350 400000
005
010
015
020
025
H 2S o
utle
t flo
w ra
te (m
L STP
min
)
Time (min)0 20 40 60 80 100 120
000
005
010
015
020
025
H 2S o
utle
t flo
w ra
te (m
L STP
min
)
Time (min)
0 50 100 150 200 250000
005
010
015
020
025
H 2S o
utle
t flo
w ra
te (m
L STP
min
)
Time (min)0 20 40 60 80 100 120
000
005
010
015
020
025
H 2S o
utle
t flo
w ra
te (m
L STP
min
)
Time (min)
62 Discussion of the main results
respectively (Fig 415a) On the other hand unlike occurring with the H2SN2 mixture a
significant H2S flow was detected in the exit synthetic gasification gas from almost the
beginning of all the experiments In this case the H2S breakthrough time was longer than 20
min only when operating with the combustion ash at 600 ordmC under dry conditions (165 min
Fig 415c) This breakthrough time was in turn significantly lower than that obtained for the
H2SN2 mixture under the same operating conditions (245 min Fig 415a) The reduction of
the iron oxides present in the sewage sludge ash may explain this observed behavior As
discussed in the literature (Tamhankar et al 1981 Tseng et al 2008 Westmoreland and
Harrison 1976) the presence of CO and H2 in the gasification gas creates a reducing
atmosphere that may cause the conversion of Fe3O4 and Fe2O3 to FeO or even to elemental Fe
in the temperature range of 700-1000 ordmC FeO and Fe show less favorable sulphidation
equilibrium which led to a reduction in the sulphur capture capacity of the ash (Tseng et al
2008) Thus the impact of temperature on the H2S removal from the gasification gas is the
result of the competition of reduction and sulphidation reaction rates
The presence of 30 vol of steam in the reaction medium also affected negatively the H2S
breakthrough time According to the general reaction of metal oxides with H2S (eq 46)
thermodynamics predicts a negative effect of steam on the equilibrium between the H2S and
the metal oxide sorbents because of the simultaneous regeneration of the spent metal
sulphides
MexOy (s) + y H2S (g) harr MexSy (s) + y H2O (g) ∆H lt0 (eq 46)
Experimental results reported in the literature show different steam impact levels on the
sulphidation rate depending on the sorbent material and the operating conditions (Cheah at
al 2009) For instance Kim et al (2007) studied the effect of steam on H2S removal by a ZnO
sorbent and found that the presence of 45 steam halved the H2S breakthrough time at 363
ordmC In general the effect of steam on the sulphur sorbent performance is expected to be more
severe at higher temperatures but there are not many studies concerning this effect (Cheah at
al 2009) In the present work the H2S breakthrough time was reduced by 85 in the presence
of 30 steam when testing the combustion ash at 600-800 ordmC (H2SN2 mixture) while the
gasification ash showed a complete loss of its capacity to remove H2S
The presence of steam in the gas atmosphere also affected the blank run results As it is
well known susceptible alloys especially steels react with H2S to form metal sulphides as
corrosion by-products This could explain the reduced H2S outlet flow rate in the blank runs
with respect to the inlet gas particularly at 800 ordmC and under dry operating conditions (Figs
Discussion of the main results 63
415a and 415c) The hot metal parts in the experimental setup do not appear to react with
H2S under wet conditions (Fig 415b and 415d) since as discussed above the formation of
metal sulphides from metal oxides is restricted by the presence of steam
Table 410 summarizes the H2S breakthrough times (time in which the H2S concentration
exceeds 100 ppmv) as well as other experimental results such as the amount of H2S removed
up to the breakthrough time or the sulphur content in the solid samples which are discussed
below
Table 410 Experimental results from the desulphurization tests
H2OH2S mass ratio T (ordmC)
H2S breakthrough
time (min)
H2S removed up to breakthrough
time (mLSTP)
S content (ultimate analysis
mg Smiddotg-1ash) a
Expected S content
(mg Smiddotg-1ash)
H2SN2 mixture feed
Sewage sludge
combustion ash
0 600 245 54 58 plusmn 1 92
0 800 300 54 63 plusmn 4 100
45 600 40 7 215 plusmn 08 29
45 800 50 9 14 plusmn 01 32 225 700 70 14 208 plusmn 06 38
225 700 62 12 188 plusmn 02 37
225 700 62 12 166 plusmn 06 36
Sewage sludge
gasification ash
0 600 30 4 232 plusmn 05 27
0 800 260 48 644 plusmn 07 100
45 600 5 2 49 plusmn 02 7 45 800 0 0 11 plusmn 01 8
225 700 0 0 141 plusmn 05 17
225 700 0 0 126 plusmn 05 18 225 700 0 0 118 plusmn 04 17
Synthetic gasification gas feed
Sewage sludge
combustion ash
0 600 165 36 464 plusmn 06 64
0 800 13 1 55 plusmn 4 53 45 600 17 4 268 plusmn 05 32 45 800 5 1 85 plusmn 05 11
Sewage sludge
gasification ash
0 600 13 1 20 plusmn 1 19
0 800 0 0 332 plusmn 06 31 45 600 5 1 58 plusmn 02 11
45 800 5 1 45 plusmn 05 8
a mean value plusmn standard deviation
The amount of H2S removed from gas up to the breakthrough time was calculated using
the blank runs data as a reference
H2S removed up to breakthrough time (mLSTP) = VH2S blank minus VH2S experiment (eq 47)
64 Discussion of the main results
where VH2S experiment is the amount of H2S (mLSTP) leaving the reactor up to the H2S
breakthrough time during each experiment and VH2S blank is the amount of H2S (mLSTP) leaving
the reactor in the blank run during the same experimental time These H2S outlet volumes
were calculated by integration of the area under the breakthrough curves (Fig 415)
considering only the flow rate data up to the breakthrough time These results have been
statistically compared by analysis of variance The regression linear coefficients (β) in terms of
coded values of the factors are shown in Table 411 Although the significant curvature
prevents the use of the linear regression model obtained the relative influence of the factors
can be easily evaluated by comparison of the absolute values of coefficients β
Table 411 Linear regression coefficients (β) from ANOVA analysis of the amount of H2S removed from the H2SN2 mixture up to the breakthrough time (mLSTP)
Average βT βH2OH2S βash type βT-H2OH2S βT-ash type βH2OH2S-ash type βT-H2OH2S-ash type Curvature
2236 546 -1776 -891 -536 494 516 -589 Significant
The three studied factors (temperature H2OH2S and ash type) as well as their
interactions significantly affect the amount of H2S removed from the gas up to the
breakthrough time Fig 416 shows the surface response obtained from the ANOVA analysis of
the amount of H2S removed from the H2SN2 mixture by each solid
(a) Sewage sludge combustion ash (b) Sewage sludge gasification ash
Figure 416 H2S removed from the H2SN2 mixture up to the breakthrough time (mLSTP) by using (a) sewage sludge combustion ash and (b) sewage sludge gasification ash
The H2OH2S ratio is the most influential factor (βH2OH2S = -1776) The higher the presence
of steam the smaller the amount of H2S removed from gas The origin of the sewage sludge
ash is also a key factor in the H2S removal process (βash type = -891) Larger amounts of H2S can
600
650
700
750
800 0
11
23
34
45
7
19
31
43
55
H2S
rem
oved
up
to b
reak
thro
ugh
time
(mL
STP
)
Temperature (ordmC) H2OH2S (gg)
600 650
700
750
800 0
11
23
34
45
0
7
14
21
29
36
43
50
H2S
rem
oved
up
to b
reak
thro
ugh
time
(mL
STP
)
Temperature (ordmC) H2OH2S (gg)
Discussion of the main results 65
be removed from the gas with the combustion ash than with the gasification ash especially at
the lowest temperature For instance 54 mLSTP of H2S were removed from the dry H2SN2
mixture up to the breakthrough time by using the combustion ash at 600 ordmC while the
gasification ash only removed 4 mLSTP of H2S before reaching the breakthrough time (Table
410) Non-significant differences in the surface features of both solids were found (Table 32)
so that the difference in their desulphurization performances must be related to ash
composition On the one hand gasification ash contains some amount of carbon that could
hinder the access of H2S to the metallic sites However this amount of carbon (314 wt )
does not appear large enough to be the only cause for the observed differences As discussed
in section 312 metal content and metallic species detected in both types of sewage sludge
ash were not exactly the same as a consequence of the different reactive atmospheres in the
combustion and gasification processes One of the main differences in the composition of the
ash samples was related to Fe content which was detected in the form of Fe2O3 in the
combustion ash and as Fe3O4 in the gasification ash Yoshimura et al (1995) analyzed Fe2O3
and Fe3O4 samples after sulphidation with H2S at 400 ordmC and the patterns obtained by
Extended X-ray Absorption Fine Structure (EXAFS) showed that the intensity of the peak
corresponding to Fe-S coordination after sulphidation of Fe3O4 was lower than that in the
sulphided sample of Fe2O3 thus indicating more difficulty in sulphiding Fe3O4 at low
temperature This fact as well as the lower Fe content detected in the gasification ash may
explain the rapid saturation of the gasification ash The difference in the desulphurization
performance of both solids was reduced with increasing temperature probably due to a rapid
increase in the Fe3O4 sulphidation reaction rate
As the desulphurization process is based on gas-solid reactions the sulphidation rate is
expected to be controlled by chemical reaction kinetics or by mass transfer The effect of the
reaction temperature on the amount of H2S removed from the H2SN2 mixture up to the
breakthrough time is clearly dependent on the ash type and presence of moisture in the gas
(βT = 546 βT-ash type = 494 βT-H2OH2S = -536) Thus temperature hardly affected the removed
amount of H2S up to the breakthrough time when using the combustion ash whereas a great
positive temperature impact was observed when the gasification ash was used at dry
conditions (Fig 416)
In addition to the evaluation of the H2S breakthrough curves sulphur content in the solid
samples was measured after the desulphurization tests by means of an elemental analyzer
The obtained results are included in Table 410 (column denoted as S content ultimate
analysis) Only a few of these data are directly comparable with each other because they refer
66 Discussion of the main results
to the total amount of sulphur removed in each complete experiment and the duration of the
experiments was not the same in all cases Comparable data show lower sulphur content in
the gasification ash than in the combustion ash after the experiments performed at 600 and
700 ordmC while similar sulphur contents were detected in both solids after the experiments
performed at 800 ordmC with the H2SN2 mixture pointing to similar sulphidation reaction rates of
Fe2O3 and Fe3O4 at high temperature These latter values were around 63-64 mg Smiddotg-1ash after
390 min of test being the highest sulphur loading detected in the solids However in the
presence of steam sulphur capture was improved at low temperature Combustion ash
operating at 600 ordmC showed the best sulphur capture ability (268 mg Smiddotg-1ash after 120 min)
under the most realistic operating conditions (H2S removal from the moist synthetic
gasification gas)
In order to check the sulphur mass balance the expected sulphur content in the spent ash
samples has been calculated assuming that all the amount of H2S removed from the gas
remained in the solid after the desulphurization test
Expected sulphur content (mg S middot g-1ash) = VH2S blankminusVH2S experiment
224 middot 32 (eq 48)
where VH2S experiment is the amount of H2S (mLSTP) leaving the reactor during each complete
experiment VH2S blank is the amount of H2S (mLSTP) leaving the reactor during the blank run
(extrapolating to the duration of the experiment where this differs) 224 is the volume of one
ideal gas mole at STP (mLSTPmiddotmmol-1) and 32 is the sulphur atomic mass (mgmiddotmmol-1) The
expected sulphur contents are included in the last column of Table 410 As can be observed
these calculated data are higher than the experimental results obtained from the elemental
analyzer This suggests that other sulphur-containing gases not detected by the micro GC could
be formed during the desulphurization tests Especially striking is the difference between the
results obtained in experiment 4 (800 ordmC and moist H2SN2 mixture) for which the expected
sulphur content in the solid was around 32 mg Smiddotg-1ash while the elemental analyzer only
detected 14 mg Smiddotg-1ash
Equilibrium simulations were conducted in order to try to explain these observed
differences from a thermodynamic point of view As the composition of the solid evolves over
time (discontinuous bed of solid and continuous feed of gas) successive simulations for small
time intervals (10 min) were performed in order to obtain an approximation to the real
process during 300 min The amount of Ca (in the form of CaO) and Fe (in the form of Fe2O3)
present in 1 g of sewage sludge ash was the initial solid for the first simulation The gas input
Discussion of the main results 67
for each equilibrium simulation was the amount of gas fed during 10 min of experiment while
the solid input was the solid resulting from the previous simulation
Figs 417 and 418 show the evolution of the theoretical distribution of sulphur into the
main sulphur-containing species resulting from the equilibrium simulations FexSy CaS H2S
CaSO4 and SO2 Successive gas-solid contact intervals of 10 min are represented
(a) Dry H2SN2 mixture 600 ordmC (b) Dry H2SN2 mixture 800 ordmC
(c) Moist H2SN2 mixture 600 ordmC (d) Moist H2SN2 mixture 800 ordmC
FexSy CaS CaSO4 H2S SO2
Figure 417 Evolution of the sulphur distribution into the main sulphur-containing species at equilibrium conditions when considering the H2SN2 mixture as inlet gas
The formation of CaS is thermodynamically favored over the formation of FexSy in the early
stage of the desulphurization process when feeding the H2SN2 mixture (Fig 417) The
available amount of Ca decreases and FexSy formation becomes more significant while the
desulphurization process progresses The remaining fraction of H2S in gas increases at the
presence of steam (Figs 417b and 417d) Besides the formation of metal sulphides
thermodynamics predicts the formation of SO2 and CaSO4 as a result of the reaction of the
formed SO2 with CaO The formation of both SO2 and CaSO4 is favored at the presence of
steam The temperature increase shifts the reaction to SO2 formation Thus SO2 formation
may be the reason for the observed differences in the expected and measured sulphur
contents in the ash samples Maximum sulphur loadings of 107 103 42 and 38 mg Smiddotg-1ash are
theoretically predicted for experiments 1 2 3 and 4 respectively Experimental results
0102030405060708090
100
Sulp
hur
dist
ribut
ion
()
0102030405060708090
100
Sulp
hur d
istrib
utio
n (
)
0102030405060708090
100
Sulp
hur
dist
ribut
ion
()
0102030405060708090
100
Sulp
hur
dist
ribut
ion
()
68 Discussion of the main results
measured with the elemental analyzer (Table 410) were 46 39 49 and 96 lower than
the theoretical results respectively
Thermodynamic data obtained for the synthetic gasification gas feed are shown in Fig
418
(a) Dry synthetic gasification gas 600 ordmC (b) Dry synthetic gasification gas 800 ordmC
(c) Moist synthetic gasification gas 600 ordmC (d) Moist synthetic gasification gas 800 ordmC
FexSy CaS H2S
Figure 418 Evolution of the sulphur distribution into the main sulphur-containing species at equilibrium conditions when considering the synthetic gasification gas as inlet gas
In this case neither SO2 nor CaSO4 are present at equilibrium conditions since the reducing
atmosphere created by the gasification gas prevents its formation H2S CaS and FexSy are the
main sulphur-containing species at equilibrium conditions COS is also formed in a very low
proportion Some authors have detected the formation of COS in the reducing environment of
the gasification gas (H2S + CO2 harr COS + H2O) (Hepola and Simell 1997) which could explain
the observed differences in the expected and measured sulphur contents in the ash samples
Maximum sulphur loadings of 85 84 42 and 41 mg Smiddotg-1ash are theoretically predicted for
experiments 20 21 22 and 23 respectively Experimental results measured with the
elemental analyzer (Table 410) were 46 35 36 and 79 lower than the theoretical
results respectively
0102030405060708090
100
Sulp
hur
dist
ribut
ion
()
0102030405060708090
100
Sulp
hur
dist
ribut
ion
()
0102030405060708090
100
Sulp
hur
dist
ribut
ion
()
0102030405060708090
100
Sulp
hur
dist
ribut
ion
()
Discussion of the main results 69
Comparison of Figs 417 and 418 shows a great impact of the gas atmosphere on the
distribution of sulphur between CaS and FexSy The presence of CO2 in the gasification gas can
explain this difference since this is the responsible gas for the carbonation reaction of CaO
(CaO + CO2 harr CaCO3) Excess CO2 shifts the reaction to the formation of CaCO3 especially at
low temperatures thus limiting the formation of CaS from CaO Thus besides the reduction of
the iron oxides carbonation of CaO may also contribute to the different results obtained for
the H2SN2 mixture and the synthetic gasification gas
In addition to the elemental analysis of solid samples other characterization techniques
were applied Fig 419 shows a back-scattered electron image of the combustion ash resulting
from experiment 2 Numbers in Fig 419 indicate the points where the elemental composition
was analyzed The atomic fractions obtained by EDX on the different superficial points are
shown in Table 412
Figure 419 Back-scattered electron image of the ash resulting from experiment 2
Table 412 Elemental composition (SEMEDX) on different superficial points of the ash resulting from experiment 2
Point in Fig 419
Elemental composition (atomic percentage)
C O Na Mg Al Si P S Ca Fe Zn 1 296 10 20 17 43 238 16 359 2 08 579 34 37 24 133 11 66 106 03 3 22 459 05 15 57 92 115 86 63 88 4 07 690 03 01 03 276 06 09 5 09 507 03 33 62 93 122 46 70 56 6 16 579 24 14 40 117 04 41 161 04 7 335 08 13 78 35 237 41 253
70 Discussion of the main results
As can be noted from data in Table 412 the composition of the sewage sludge ash surface
is quite heterogeneous C O Na Mg Al Si P S Ca Fe and Zn are detected along the surface
in different fractions It should be noted that the points with the highest atomic percentage of
S (238 in point 1 and 237 in point 7) are also those with the highest Fe content (359 and
253 respectively) thus suggesting the formation of either iron sulphides or iron sulphates
On the other hand S was hardly detected in other points such as in point 4 (mainly formed by
O and Si in the form of SiO2) or in point 6 in which despite of the high fraction of Fe (161)
only 04 atomic S was found In this case as well as in points 2 and 3 the high presence of Fe
is linked with a high presence of P which indicates the presence of iron phosphates
Fig 420 shows the XPS spectra in S 2p region of the ash resulting from experiment 2
Several doublets can be fitted to the experimental signal indicating different chemical states
of sulphur in the ash surface The peak located between 160 and 164 eV is indicative of Sn-2
(metal sulphides) but other oxidized sulphur forms have also been detected (169 eV SO4-2)
Figure 420 XPS spectra in the S 2p region corresponding to the ash resulting from experiment 2
In summary results from desulphurization tests have shown some sulphur capture
capacity of sewage sludge ash especially in the case of the solid by-product derived from the
combustion process The desulphurization performance is dramatically reduced in the
presence of steam so the use of this solid by-product as sorbent material for H2S could be
further exploited for dry gas cleaning such as the sewage sludge pyrolysis gas after
condensation of water and vapors
Discussion of the main results 71
44 Catalyst activity tests
This section presents the results obtained in the Ni-based catalysts activity tests for tar
reforming The Ni-Al2O3 catalysts were modified by adding different metallic promoters Ca Fe
Mn and Cu A synthetic gasification gas containing H2 CO CO2 N2 CH4 C2H4 H2O H2S (3000
ppmv wet basis) and a mixture of benzene toluene and naphthalene as tar model compounds
(Table 35) was used as reactive atmosphere in the experiments
Fig 421 shows the evolution of the outlet concentration of naphthalene (C10H8) mono-
aromatic tar (C6H6+C7H8) methane (CH4) and ethylene (C2H4) when using the non-modified
catalyst (Ni-Al2O3) calcined at 900 ordmC and pre-reduced in a H2N2 atmosphere The temperature
ramp followed during the experiment is shown in the upper x-axis Ethylene was the only
hydrocarbon significantly converted below 800 ordmC In the upper temperature range (800-900
ordmC) naphthalene conversion rate was higher than those of methane and mono-aromatic tar
Inlet concentration Outlet concentration
Figure 421 Evolution of the concentration of naphthalene mono-aromatic tar (benzene + toluene) methane and ethylene with the pre-reduced sample of NiAl2O3 calcined at 900 ordmC
0 2 4 6 8 10 12 14 16 18 20 22 24406080
100120140160180200220
Temperature (oC)
C 10H 8 c
once
ntra
tion
(ppm
v)
Time (h)
900 850 800 900 750 700 900
0 2 4 6 8 10 12 14 16 18 20 22 24600800
10001200140016001800200022002400
Temperature (oC)
C 6H6 +
C7H
8 con
cent
ratio
n (p
pmv)
Time (h)
900 850 800 900 750 700 900
0 2 4 6 8 10 12 14 16 18 20 22 24150001600017000180001900020000210002200023000
900 850 800 900 750 700 900Temperature (oC)
CH4 c
once
ntra
tion
(ppm
v)
Time (h)0 2 4 6 8 10 12 14 16 18 20 22 24
0100020003000400050006000700080009000
C 2H4 c
once
ntra
tion
(ppm
v)
Time (h)
900 850 800 900 750 700 900Temperature (oC)
72 Discussion of the main results
The concentration of tar and hydrocarbons in the outlet gas was not totally stable in the
temperature range in which the catalyst remained active but a slight upward trend was
observed in most of them as a consequence of a progressive activity loss Hepola and Simell
(1997) found that catalyst deactivation by H2S was reduced by increasing the operating
temperature and decreasing the system pressure However despite the high temperature
used in the first 35 h of experiment (900 ordmC) the observed gradual increase in the
concentration of methane and mono-aromatic suggests a quick reforming-activity loss of the
catalyst in the first stage of experiment Naphthalene and ethylene concentrations obtained at
900 ordmC were not affected by catalyst deactivation as much as methane and mono-aromatic tar
conversions The aforementioned effect of temperature on catalyst deactivation by H2S
poisoning can be clearly observed in the ethylene concentration which shows the most abrupt
increase in the stage at 750 ordmC The different behavior of mono-aromatic tar naphthalene
methane and ethylene conversions can be attributed to the different decomposition
mechanisms and to the competition of reactants for active reaction sites on the catalyst
surface
Performance of the promoted catalysts calcined at 900 ordmC and pre-reduced in a H2N2
atmosphere is shown in Fig 422 The depicted points represent the conversion average data
obtained at each temperature The conversion of the tar model compounds methane and
ethylene can be calculated from the reactor mass balance as follows
Conversion i() = niin minus nioutniin
∙ 100 (eq 49)
where niin and niout are the molar flows of compound ldquoirdquo at the reactor inlet and outlet
respectively The formation of benzene is usually linked to the decomposition of toluene
through reactions such as the steam dealkylation (C7H8 + H2O harr C6H6 + CO + 2H2) or
hydrodealkylation (C7H8 + H2 harr C6H6 + CH4) so a global conversion for both compounds has
been calculated and named ldquomono-aromatic tar conversionrdquo
The blank run results are not depicted in Fig 422 but they evidenced that thermal
cracking hardly contributed to the decomposition of tar compounds and light hydrocarbons
under the tested operating conditions Conversion of methane naphthalene and mono-
aromatic tar was found to be virtually zero in the blank run while ethylene conversion varied
from 43 (700 ordmC) to 77 (900 ordmC)
Discussion of the main results 73
0
20
40
60
80
100 0 35 7 105 14 175 21 245Experiment time (h)
900 850 800 900 750 700 900
C 10H 8 c
onve
rsio
n (
)
Temperature (oC)0
10
20
30
40
50
60Experiment time (h)
0 35 7 105 14 175 21 245
900 850 800 900 750 700 900
C 6H6 +
C7H
8 co
nver
sion
()
Temperature (oC)
-10-505
1015202530 0 35 7 105 14 175 21 245
Experiment time (h)
900 850 800 900 750 700 900
CH4 c
onve
rsio
n (
)
Temperature (oC)
0
20
40
60
80
100Experiment time (h)
0 35 7 105 14 175 21 245
900 850 800 900 750 700 900
C 2H4 c
onve
rsio
n (
)
Temperature (oC)
NiAl2O3 NiCaAl2O3 NiFeAl2O3 NiCuAl2O3 NiMnAl2O3
Figure 422 Conversion of tar model compounds and light hydrocarbons as a function of the catalyst temperature and added promoter in the solid samples calcined at 900 ordmC and pre-reduced in a H2N2
atmosphere
As observed in Fig 422 tar conversion was significantly reduced as the catalyst
temperature decreased This is obvious as both the reaction rate of tar decomposition and the
mass transfer rate are favored at higher temperatures Naphthalene conversion was lower
than 20 for temperatures below 800 ordmC with all the catalysts At 900 ordmC the presence of Mn
in the catalyst led to a naphthalene conversion of 80 thus improving the conversion
obtained with NiAl2O3 by 10 percentage points Adding the other metals (Ca Fe and Cu) was
found to be detrimental to naphthalene conversion as well as to mono-aromatic tar methane
and ethylene conversions The reduced surface area obtained when the promoters were
added (Table 33) may explain the worse performance of the promoted catalysts
The maximum mono-aromatic tar conversion was obtained with NiAl2O3 obtaining an
average value of 55 in the first step of 900 ordmC However this average conversion decreased to
49 and 40 in the following temperature steps at 900 ordmC being the most significant
reduction among the tested catalysts The activity loss of NiAl2O3 was also clearly observed in
methane conversion data as the average value was reduced from 28 in the first step at 900
74 Discussion of the main results
ordmC to 16 in the third step at the same temperature Mn addition allowed a stable methane
conversion in the three temperature steps at 900 ordmC (around 24) Negative values of methane
conversion were even found with NiCuAl2O3 and NiFeAl2O3 at temperatures between 700
and 800 ordmC indicating an increased production of this compound through methanation
reactions (CO + 3H2 harr CH4 + H2O C + 2H2 harr CH4) Mn addition also improved ethylene
conversion as this was above 90 with NiAl2O3 and close to 100 with NiMnAl2O3 in the
successive steps at 900 ordmC
Regarding the evolution of the other permanent gases Fig 423 shows the ratio between
the outlet and the inlet molar flows of H2 CO and CO2 as a function of the catalyst temperature
and added promoter The depicted points represent the average ratio obtained at each
temperature Values above 1 indicate that the outlet flow exceeds the inlet flow of the gas
component
0910111213141516
900 850 800 900 750 700 900
H 2 out
let
H2 i
nlet
Temperature (oC)
Experiment time (h)0 35 7 105 14 175 21 245
08091011121314151617
900 850 800 900 750 700 900
CO o
utle
t C
O in
let
Temperature (oC)
0 35 7 105 14 175 21 245Experiment time (h)
08
09
10
11
12
900 850 800 900 750 700 900
CO2 o
utle
t C
O2 i
nlet
Temperature (oC)
Experiment time (h)0 35 7 105 14 175 21 245
NiAl2O3 NiCaAl2O3 NiFeAl2O3 NiCuAl2O3 NiMnAl2O3
Figure 423 Molar ratio between the outlet and inlet molar flows of H2 CO and CO2 as a function of the catalyst temperature and added promoter in the samples calcined at 900 ordmC and pre-reduced in a H2N2
atmosphere