SISTEMAS CATALÍTICOS BASADOS EN DISTINTOS

Departamento de Ingeniería Química y Química Física

TESIS DOCTORAL

SISTEMAS CATALÍTICOS BASADOS EN DISTINTOS

SEMICONDUCTORES PARA LA ELIMINACIÓN

DE CONTAMINANTES EN AGUA MEDIANTE

OZONIZACIÓN FOTOCATALÍTICA SOLAR

ESTEFANÍA MENA RUBIO

Conformidad de los Directores de Tesis:

Fdo.: Ana Rey Barroso Fdo.: Eva M. Rodríguez Franco

Badajoz, 2017

A mis padres

La realización de esta Tesis Doctoral ha sido posible gracias a la concesión de

una beca de Formación de Personal Investigador (Ref. PD12059) financiada por

la Consejería de Empleo, Empresa e Innovación (Junta de Extremadura) y el

Fondo Social Europeo, así como al apoyo económico prestado a través de los

proyectos de investigación de la Junta de Extremadura (GRU10012 y GR15033) y

a través de los proyectos del Programa Nacional de Investigación

CTQ2009/13459/C05/05, titulado “Integración de procesos de fotocatálisis solar en la

depuración biológica de aguas residuales para la eliminación de contaminantes

emergentes”; CTQ2012/35789/C02/01, titulado “Preparación de catalizadores y su

aplicación en la eliminación de contaminantes refractarios de aguas residuales mediante

ozonización fotocatalítica”; y CTQ2015‐64944‐R titulado “Led y fotocatalizadores

polifuncionales basados en grafeno y estructuras metal‐orgánicas para el tratamiento de

aguas por ozonización fotocatalítica” del Ministerio de Economía y Competitividad

(España) y del Fondo Europeo de Desarrollo Regional (FEDER).

Llegado este momento, no puedo dar por finalizado mi trabajo sin antes

agradecer a todas las personas que me han acompañado en esta etapa.

Gran parte de este trabajo es vuestro. A todos, muchas gracias.

ÍNDICE

RESUMEN ………………………………………………………………………...... 1

CAPÍTULO 1: INTRODUCCIÓN Y OBJETIVOS …………………………….. 9

1.1. ANTECEDENTES Y JUSTIFICACIÓN DE LA TESIS ………………... 11

1.1.1. Contextualización del trabajo ...…………........................................ 11

1.1.2. El agua en el mundo actual …..…………........................................ 11

1.1.2.1. Marco legal del agua …..………………...................................... 16

1.1.2.2. Conservación de los recursos hídricos …..………………............ 21

1.1.3. Procesos avanzados de oxidación …..………................................. 25

1.1.3.1. Fotocatálisis heterogénea ………………...................................... 26

1.1.3.2. Combinación de ozono y fotocatálisis heterogénea:

ozonización fotocatalítica …..………………............................................ 30

1.1.4. Optimización de los procesos fotocatalíticos: empleo de

radiación solar y fotocatalizadores apropiados …................................... 38

1.2. OBJETIVOS Y ALCANCE DEL TRABAJO …………………..………... 44

1.3. ORGANIZACIÓN DE LA MEMORIA ………...……………..………... 46

BIBLIOGRAFÍA ………...……………..……………………………….……... 47

Índice

II

CAPÍTULO 2: MATERIALES Y MÉTODOS EXPERIMENTALES ………..... 61

2.1. INSTALACIONES Y PROCEDIMIENTOS EXPERIMENTALES ..….. 63

2.1.1. Tratamientos de eliminación de contaminantes en agua ...…….. 63

2.1.1.1. Ensayos realizados empleando lámparas de luz UVA

(luz negra) como fuente de radiación …..…………….............................. 63

2.1.1.2. Ensayos realizados empleando lámparas de luz solar artificial

como fuente de radiación …..……………................................................. 66

2.1.2. Preparación de los catalizadores empleados …..………............... 68

2.1.2.1. Agitador ………………............................................................... 68

2.1.2.2. Autoclave ………………............................................................. 69

2.1.2.3. Centrífuga …….………............................................................... 69

2.1.2.4. Estufa ……..……………............................................................. 69

2.1.2.5. Horno mufla ……..……………................................................... 69

2.2. EQUIPOS Y MÉTODOS DE ANÁLISIS ……………….……..………... 70

2.2.1. Seguimiento de la eficacia de los tratamientos de eliminación

de contaminantes en agua …..………........................................................ 70

2.2.1.1. Determinación de la concentración de contaminantes modelo

en agua ……………….............................................................................. 72

2.2.1.2. Identificación de productos intermedios de degradación .............. 73

2.2.1.3. Determinación de la concentración de ácidos carboxílicos de

bajo peso molecular y aniones inorgánicos en disolución ......................... 74

2.2.1.4. Determinación de la concentración de carbono orgánico total

en disolución (COT) .................................................................................. 75

2.2.1.5. Determinación de la concentración de ozono en disolución

acuosa …………………………………………………………………… 76

2.2.1.6. Determinación de la concentración de ozono en fase gas ………. 77

2.2.1.7. Determinación de la concentración de peróxido de hidrógeno

en disolución ............................................................................................. 77

2.2.1.8. Determinación de la concentración de hierro total en

disolución .................................................................................................. 78

2.2.1.9. Determinación de la concentración de hierro (II) en disolución .. 79

2.2.1.10. Determinación de la intensidad de radiación ………………..... 80

2.2.1.11. Determinación de la concentración de formaldehído en

disolución ………………………………………………………………. 82

Índice

III

2.2.1.12. Determinación de la demanda química de oxígeno (DQO) ....... 83

2.2.1.13. Determinación de la biodegradabilidad: demanda biológica

de oxígeno (DBO) ..................................................................................... 83

2.2.1.14. Determinación de la absorbancia a 254 nm: Aromaticidad ....... 84

2.2.1.15. Determinación de la concentración de fosfatos en disolución .... 85

2.2.1.16. Determinación de la turbidez ………………………………..... 85

2.2.1.17. Determinación de pH, conductividad y temperatura ……........ 86

2.2.2. Caracterización de los catalizadores empleados …..………......... 86

2.2.2.1. Espectroscopía de emisión atómica de plasma por acoplamiento

inductivo (ICP‐OES) ………………........................................................ 88

2.2.2.2. Difracción de Rayos X (XRD) ..................................................... 88

2.2.2.3. Espectroscopía Raman ………..................................................... 90

2.2.2.4. Análisis termogravimétrico y térmico diferencial (TGA‐DTA) .. 91

2.2.2.5. Microscopía electrónica de transmisión (TEM) ………..………. 92

2.2.2.6. Microscopía electrónica de barrido (SEM) ………….....………. 92

2.2.2.7. Isoterma de adsorción‐desorción de nitrógeno …………………. 93

2.2.2.8. Determinación del pH de potencial de carga cero (pHPZC) por

valoración másica ……………….............................................................. 95

2.2.2.9. Espectroscopía fotoelectrónica de Rayos X …………..………..... 95

2.2.2.10. Espectroscopía ultravioleta‐visible de reflectancia difusa

(DR‐UV‐Vis) ………………..................................................................... 97

2.3. REACTIVOS EMPLEADOS ………...……………..………………..…... 98

BIBLIOGRAFÍA ………...……………..……………………………….……... 103

CAPÍTULO 3 (CHAPTER 3). PAPER 1: ON OZONE‐PHOTOCATALYSIS

SYNERGISM IN BLACK‐LIGHT INDUCED REACTIONS: OXIDIZING

SPECIES PRODUCTION IN PHOTOCATALYTIC OZONATION

VERSUS HETEROGENEOUS PHOTOCATALYSIS ........................................ 107

3.1. INTRODUCTION …………………………………………………….….. 109

3.2. EXPERIMENTAL SECTION ……………..………………………….….. 113

3.2.1. Experimental set‐up and oxidation/ozonation procedure ..…..... 113

3.2.2. Photon fluxes determination ………………………………...…..... 114

3.3. RESULTS AND DISCUSSION ….………..………………………….….. 115

3.3.1. Absorbed photon flux ..…………………………………………..... 115

Índice

IV

3.3.2. Photocatalytic oxidation versus photocatalytic ozonation …...... 121

3.3.2.1. Influence of ozone concentration ………...................................... 125

3.3.2.2. Influence of pH …………………………………………………. 126

3.3.2.3. Synergistic effect and quantum yield of photocatalytic induced

reactions ………………………………………........................................ 129

3.3.2.4. Simplified economic considerations ……………………...…..… 132

3.4. CONCLUSIONS ………………....………..………………………….….. 135

AKNOWLEDGEMENTS ………...………...………………………….……... 136

REFERENCES ……...…...……………..……………………………….……... 136

CAPÍTULO 4 (CHAPTER 4). PAPER 2: INFLUENCE OF STRUCTURAL

PROPERTIES ON THE ACTIVITY OF WO3 CATALYSTS FOR VISIBLE

LIGHT PHOTOCATALYTIC OZONATION …................................................. 141

4.1. INTRODUCTION …………………………………………………….….. 143


4.2.1. Catalysts preparation ………………………………………...…..... 144

4.2.2. Characterization ………………….…………………………...…..... 145

4.2.3. Catalytic activity measurements ………………..…………...…..... 146


4.3.1. Characterization of the photocatalysts ..………………………..... 148

4.3.2. Comparison of processes ………………………………………...... 154

4.3.2.1. Simplified mechanistic approach of IBP photocatalytic

ozonation ……………...………………………........................................ 160

4.3.3. Catalysts screening for photocatalytic ozonation under

visible‐light radiation …………………………………………………...... 164

4.3.4. Simulated solar light photocatalytic ozonation of ECs in

MWW ………………………...…………………………………………...... 168

4.4. CONCLUSIONS ………………....………..………………………….….. 172


REFERENCES ……………..……...………...………………………….……... 173

SUPPLEMENTARY INFORMATION OF CHAPTER 4 ………..….……... 177

Índice

V

CAPÍTULO 5 (CHAPTER 5). PAPER 3: VISIBLE LIGHT PHOTOCATALYTIC

OZONATION OF DEET IN THE PRESENCE OF DIFFERENT FORMS OF

WO3 ………................................................................................................................. 183

5.1. INTRODUCTION …………………………………………………….….. 185






5.3.1. Photocatalysts characterization ………..………………………..... 189

5.3.2. Catalytic activity ……………..…………………………………...... 195

5.4. CONCLUSIONS ………………....………..………………………….….. 202


REFERENCES ……………..……...………...………………………….……... 202

SUPPLEMENTARY INFORMATION OF CHAPTER 5 ………..….……... 205

CAPÍTULO 6 (CHAPTER 6). PAPER 4: NANOSTRUCTURED CeO2 AS

CATALYSTS FOR DIFFERENT AOPs BASED IN THE APPLICATION OF

OZONE AND SIMULATED SOLAR RADIATION.......................................... 213

6.1. INTRODUCTION …………………………………………………….….. 215






6.3.1. Photocatalysts characterization ………..………………………..... 219

6.3.2. Catalytic activity ……………..…………………………………...... 223

6.3.3. Considerations on the reaction mechanism of DEET

photocatalytic ozonation with CeO2 catalysts and solar radiation …... 227

6.4. CONCLUSIONS ………………....………..………………………….….. 230


REFERENCES ………………………………………………..……..….……... 231

Índice

VI

CAPÍTULO 7 (CHAPTER 7). PAPER 5: WO3‐TiO2 BASED CATALYSTS FOR

THE SIMULATED SOLAR RADIATION ASSISTED PHOTOCATALYTIC

OZONATION OF EMERGING CONTAMINANTS IN A MUNICIPAL

WASTEWATER TREATMENT PLANT EFFLUENT ….................................... 235

7.1. INTRODUCTION …………………………………………………….….. 237






7.3.1. Characterization of the photocatalysts ………..………………..... 243

7.3.2. Visible light response of the photocatalysts …………………...... 251

7.3.3. Photocatalytic degradation of ECs in MWW ………………......... 253

7.4. CONCLUSIONS ………………....………..………………………….….. 263


REFERENCES ………………………………………………..……..….……... 264

CAPÍTULO 8 (CHAPTER 8). PAPER 6: REACTION MECHANISM AND

KINETICS OF DEET VISIBLE LIGHT ASSISTED PHOTOCATALYTIC

OZONATION WITH WO3 CATALYST ……….................................................. 269

8.1. INTRODUCTION …………………………………………………….….. 271


8.2.1. Experiments …………………………………………………...…..... 272

8.2.2. Analytical methods ………...…….…………………………...…..... 274


8.3.1. Comparison of processes ………..……………………...………..... 276

8.3.2. Determination of the main species responsible for DEET

degradation and mineralization ………………………..……………...... 283

8.3.3. Identification of TPs ……………………………………………....... 287

8.3.4. Proposed reaction mechanism ……………………..…………....... 294

8.3.5. Kinetic study ……………………………………………………....... 299

8.4. CONCLUSIONS ………………....………..………………………….….. 309


Índice

VII

REFERENCES ………………………………………………..……..….……... 310

CAPÍTULO 9: CONCLUSIONES …………………………………………….…. 315

NOMENCLATURA …………………………………………...……………….….. 321

NOMENCLATURE …………………………………………………………….….. 325

RESUMEN

La Tesis Doctoral que se presenta en esta memoria se encuadra dentro del

trabajo desarrollado por el grupo de investigación TRATAGUAS, perteneciente

al Área de Ingeniería Química de la Universidad de Extremadura. En los últimos

años, la actividad del grupo se ha centrado en la eliminación de contaminantes

emergentes presentes en las aguas mediante procesos avanzados de oxidación y,

más recientemente, en la preparación de catalizadores para su aplicación en

dichos procesos. En concreto, en esta Tesis se han abordado fundamentalmente

algunos de los objetivos del proyecto de investigación titulado “Preparación de

catalizadores y su aplicación en la eliminación de contaminantes refractarios de aguas

residuales mediante ozonización fotocatalítica” (referencia CTQ2012/35789/C02/01),

financiado por el Ministerio de Economía, Industria y Competitividad y el Fondo

Europeo de Desarrollo Regional (FEDER).

En el escenario actual donde la disponibilidad de agua de calidad se está

viendo disminuida, la aparición de los denominados “contaminantes

emergentes” y sus posibles riesgos tanto para la vida acuática como la humana

ha despertado un especial interés. A este grupo de contaminantes pertenecen

Resumen

2

compuestos como fármacos, productos de higiene y cuidado personal, drogas de

abuso, pesticidas, surfactantes o aditivos industriales, entre otros. Todos ellos se

producen a escala mundial, tienen multitud de aplicaciones y han llegado a ser

imprescindibles para la sociedad moderna. Así, su introducción en el medio es

continua, principalmente a través de la descarga de las estaciones depuradoras

de aguas residuales (EDAR), dado que muchos de estos contaminantes son

refractarios a los procesos convencionales existentes. Ante esta situación, se hace

necesaria la aplicación de un tratamiento terciario adicional que permita

degradar dichos compuestos antes del vertido del efluente a cauce público o de

su reutilización.

En este trabajo se han seleccionado varios contaminantes emergentes,

detectados con frecuencia en efluentes secundarios de EDAR, para estudiar la

aplicabilidad de procesos avanzados de oxidación (PAO) como tratamiento

terciario. Estos procesos se basan en la generación de radicales libres,

especialmente radicales hidroxilo ( •HO ), que son especies transitorias de vida

media muy corta y alto poder oxidante, capaces de reaccionar de forma no

selectiva con un amplio grupo de contaminantes resistentes a los tratamientos

de agua convencionales. Dentro de los PAO, la presente investigación se centra

en la aplicación de la fotocatálisis heterogénea combinada con ozono

(ozonización fotocatalítica) empleando distintos catalizadores y fuentes de

radiación. Según algunos estudios previos, se espera que en este tratamiento

combinado aumente el rendimiento de formación de especies oxidantes respecto

de los procesos individuales de fotocatálisis y ozonización, siendo por tanto más

eficaz en la eliminación de contaminantes en agua.

El punto de partida de este trabajo de Tesis ha sido justificar la elección del

proceso de ozonización fotocatalítica como tratamiento a aplicar. Para ello, se ha

seleccionado el metanol como compuesto modelo, empleando radiación

ultravioleta (UVA) y TiO2 comercial (P25) como catalizador. La baja reactividad

del metanol frente al ozono molecular y su alta afinidad por las especies

oxidantes generadas en los procesos fotocatalíticos permiten determinar la

Resumen

3

existencia o no de sinergia al combinar los procesos individuales de fotocatálisis

y ozonización, a partir del cálculo del rendimiento cuántico de los tratamientos

fotocatalíticos.

Para determinar dicha sinergia se siguió la evolución de la concentración de

formaldehído, principal producto de oxidación del metanol, en ensayos de

fotocatálisis y ozonización simple y fotocatalítica, evaluando el efecto de la

concentración de ozono y del pH. Para calcular el rendimiento cuántico de las

reacciones fotocatalíticas se determinó la intensidad de radiación absorbida por

el catalizador y se tuvieron en cuenta las distintas reacciones que pueden tener

lugar dependiendo del proceso y las condiciones del medio: reacción directa

ozono‐metanol, reacciones indirectas por descomposición de ozono en especies

radicales y reacciones fotocatalíticas. Esta última contribución es la que permite

calcular el rendimiento cuántico de foto‐generación de especies oxidantes. A pH

= 3 (pH al cual se inhiben las reacciones indirectas del ozono) la presencia de

ozono ejerció un efecto positivo en la velocidad de formación de especies

oxidantes debido a las reacciones inducidas por la luz, incrementándose el valor

del rendimiento cuántico con respecto al proceso de fotocatálisis. Este parámetro

aumentó aún más a pH = 7 al verse favorecidas las reacciones indirectas del

ozono. El efecto positivo de la presencia de ozono en las reacciones

fotocatalíticas se ha atribuido a su papel como aceptor de electrones, dando

lugar a una mayor concentración de radicales hidroxilo en el medio y

reduciendo en cierta medida el proceso de recombinación en la superficie del

catalizador.

Comprobada la mayor eficiencia del proceso fotocatalítico cuando se

combina con ozono, el resto del trabajo de esta Tesis Doctoral se ha centrado en

el empleo de fotocatalizadores que permitan un mayor aprovechamiento de la

radiación solar con respecto al fotocatalizador utilizado por excelencia (TiO2), y

la posterior aplicación de estos materiales en la eliminación de distintos

contaminantes emergentes mediante ozonización fotocatalítica solar. De este

modo, se intenta paliar uno de los principales inconvenientes de esta tecnología

Resumen

4

derivado del uso de fuentes artificiales de radiación. Todos los ensayos se han

realizado a escala de laboratorio, empleando luz solar simulada y distintos

materiales previamente sintetizados y/o comerciales.

Se han sintetizado catalizadores de WO3 con diferentes propiedades

estructurales, estudiando su actividad en el proceso de ozonización fotocatalítica

solar aplicado a la eliminación de: i) ibuprofeno (IBP) en agua ultrapura

empleando radiación solar restringida a la zona del visible; y ii) una mezcla de

10 contaminantes emergentes añadidos al efluente de una EDAR urbana,

empleando todo el espectro de la radiación solar. Estos catalizadores se

sintetizaron mediante descomposición térmica de un precursor de wolframio

(tungstita, H2O∙WO3) a 300 y 450 °C y diferentes tiempos de calcinación. El

catalizador sometido a descomposición térmica a 450 °C durante 5 minutos

presentó la estructura monoclínica del WO3 y una concentración mayor de

vacantes de oxígeno que el resto de catalizadores, lo que condujo a una mayor

actividad en el proceso de ozonización fotocatalítica. Bajo luz solar visible, la

desaparición de ibuprofeno (CIBP,0 = 10 mg L‐1) fue completa en 20 minutos y la

mineralización superior al 85 % tras dos horas de reacción en las condiciones de

operación empleadas (pH0 = 6,5, T = 20 ‐ 40 °C, V = 0,5 L, CWO3 = 0,25 g L‐1,

CO3,g = 10 mg L‐1, Qg = 20 L h‐1). Bajo radiación solar global, en las condiciones

anteriores este catalizador permitió la degradación completa de la mezcla de

contaminantes (C0 = 0,5 mg L‐1 cada uno) en el efluente de EDAR urbana

(CCOT,0 = 17 mg L‐1) en 60 minutos y una mineralización del 40 % en 2 horas.

Por otro lado, se ha evaluado la influencia de la morfología y de nuevo la

estructura del WO3 en el proceso de ozonización fotocatalítica bajo luz solar

visible, empleando N,N‐dietil‐meta‐toluamida (DEET) como compuesto modelo

en agua ultrapura. Los catalizadores de WO3 se obtuvieron mediante síntesis

sol‐gel e hidrotermal, procesos que favorecieron el desarrollo de diferentes

formas y estructuras cristalinas.

Los resultados indican que el desarrollo de las estructuras cristalinas

Resumen

5

monoclínica y ortorrómbica del WO3 incrementa la actividad fotocatalítica del

material en mayor medida que un área superficial más desarrollada. Asimismo,

en los procesos que emplean ozono la presencia de estados reducidos de W

condujo a una mayor actividad fotocatalítica. En las condiciones de operación

estudiadas (CDEET,0 = 5 mg L‐1, pH0 = 6, T = 20 ‐ 40 °C, V = 0,5 L, CWO3 = 0,25 g L‐1,

CO3,g = 10 mg L‐1, Qg = 15 L h‐1), los mejores catalizadores condujeron a la

degradación total de DEET en menos de 20 minutos y una mineralización

superior al 70 % tras 2 horas de tratamiento.

Otra alternativa al TiO2 fueron los catalizadores de CeO2. Se ha estudiado la

influencia de la morfología del material en su actividad evaluando la misma a

través de ensayos de degradación de DEET en agua ultrapura mediante

ozonización fotocatalítica bajo luz solar visible y radiación solar global. Se

prepararon dos catalizadores nanoestructurados de CeO2 (nanocubos y

nanovarillas) mediante síntesis hidrotermal. En las condiciones de operación

ensayadas (CDEET,0 = 5 mg L‐1, pH0 = 6, T = 35 ‐ 40 °C, V = 0,5 L, CWO3 = 0,25 g L‐1,

CO3,g = 10 mg L‐1, Qg = 15 L h‐1), ambos catalizadores resultaron activos en la

eliminación de DEET mediante ozonización fotocatalítica bajo radiación tanto

visible como solar. Por su menor energía de salto de banda fueron las

nanovarillas las que presentaron una mayor actividad bajo luz visible, logrando

la eliminación total de DEET en 45 minutos y una mineralización del 68 % tras 2

horas de tratamiento. Por el contrario, bajo radiación solar global los nanocubos

tuvieron un mejor rendimiento (degradación total de DEET en 30 minutos y 80

% de mineralización tras 2 horas de reacción), presentando esta morfología,

además, una mayor velocidad específica de mineralización debido a la mayor

actividad de las caras cristalinas expuestas.

Finalmente, se han preparado catalizadores compuestos de WO3‐TiO2 con un

contenido en WO3 de aproximadamente un 4 % en peso empleando dos soportes

de TiO2 diferentes: TiO2 comercial (P25) y nanotubos de TiO2 sintetizados

mediante tratamiento hidrotermal del catalizador comercial. Una vez

Resumen

6

preparados, los materiales han sido empleados en la ozonización fotocatalítica

del efluente de una EDAR urbana dopado con una mezcla de contaminantes

bajo radiación solar global.

El catalizador basado en nanotubos presentó una porosidad más

desarrollada y una mayor eficiencia en el proceso de ozonización fotocatalítica.

Así, en las condiciones de operación estudiadas (C0 = 2 mg L‐1 de cada

contaminante emergente, CCOT,0 = 35 mg L‐1, CCI,0 = 42 mg L‐1, pH0 = 8.3, T = 20 ‐ 40

°C, V = 0,5 L, CNT‐WO3 = 0,5 g L‐1, CO3,g = 10 mg L‐1, Qg = 20 L h‐1), este material

condujo a la total desaparición de los contaminantes emergentes en menos de 40

minutos y a un grado de mineralización superior al 64 % tras 2 horas. La mayor

actividad fotocatalítica del material compuesto en comparación con el

catalizador de TiO2 puro se debe a varios efectos combinados tales como: el

mayor aprovechamiento de la radiación visible; una mayor capacidad de

adsorción de compuestos orgánicos; y una eficiente descomposición del ozono

gracias a la presencia de WO3, el cual resulta activo en la descomposición del

ozono en la oscuridad (ozonización catalítica).

Tras la síntesis, caracterización y aplicación de los distintos materiales, en

base a los resultados obtenidos bajo luz visible se ha seleccionado, por su mayor

actividad, el WO3 con estructura monoclínica y morfología de microesferas, y se

ha aplicado en la ozonización fotocatalítica con luz visible solar de DEET en

disolución acuosa, al objeto de proponer un modelo cinético que describa el

comportamiento de este sistema con vistas a su posible desarrollo y aplicación a

mayor escala.

Mediante el empleo de inhibidores (ter‐butanol y ácido oxálico), se

analizaron las principales especies involucradas en el proceso resultando ser los

radicales •HO los principales responsables de la degradación y mineralización

del DEET. Se identificaron diferentes productos intermedios de degradación de

DEET y se estableció la evolución de sus abundancias relativas a lo largo del

tiempo de reacción. La eficacia del tratamiento de ozonización fotocatalítica se

Resumen

7

manifiesta no solo en la velocidad de eliminación del DEET, sino también en la

de desaparición de los intermedios de degradación y de mineralización. En las

condiciones de operación ensayadas (CDEET,0 = 15 mg L‐1, pH0 = 6, T = 20 ‐ 40 °C,

V = 0,5 L, CWO3 = 0,25 g L‐1, CO3,g = 10 mg L‐1, Qg = 15 L h‐1), los intermedios

formados al inicio, de estructura similar a la del compuesto de partida, fueron

eliminados por completo tras 60 minutos de operación, permaneciendo en

disolución únicamente ácidos orgánicos de cadena corta de muy baja toxicidad y

a concentraciones en concordancia con el grado de mineralización alcanzado (60

% en 2 horas). En base a los resultados obtenidos se ha propuesto para la

ozonización fotocatalítica bajo luz visible un mecanismo de reacción que incluye

distintas etapas basadas en el ataque del radical •HO , desarrollando un modelo

cinético capaz de simular la evolución de la concentración de DEET, la de sus

principales intermedios de degradación y la de los ácidos orgánicos de cadena

corta en términos de carbono orgánico total, lo que proporciona un enfoque

simplificado del proceso.

CAPÍTULO 1 Introducción y objetivos

El crecimiento demográfico e industrial en las últimas décadas ha provocado un incremento exponencial de la contaminación de los recursos hídricos, lo que motiva el desarrollo de procesos y tecnologías que permitan devolver los parámetros físicos, químicos y biológicos de las aguas residuales a su estado natural. En este capítulo se aborda brevemente la naturaleza de la contaminación hídrica, con especial atención a los denominados contaminantes emergentes. Seguidamente, se exponen algunos conceptos teóricos sobre procesos avanzados de oxidación que se proponen para eliminar dichos contaminantes del agua, haciendo referencia a líneas y resultados actuales de investigación. Por último, se indican los objetivos que se persiguen en este trabajo de Tesis, así como la estructura de la memoria que se presenta.

Introducción y objetivos

11

1.1. ANTECEDENTES Y JUSTIFICACIÓN DE LA TESIS

1.1.1. Contextualización del trabajo

El presente trabajo de Tesis Doctoral se ha desarrollado dentro del grupo de

investigación TRATAGUAS del Departamento de Ingeniería Química y Química

Física de la Universidad de Extremadura [1], gracias a la beca de Formación de

Personal Investigador (Ref. PD12059) concedida por la Consejería de Empleo,

Empresa e Innovación (Junta de Extremadura) y el Fondo Social Europeo.

En los últimos años, la actividad del grupo se ha centrado en la eliminación

de contaminantes emergentes presentes en las aguas mediante procesos

avanzados de oxidación y, más recientemente, en la preparación de

catalizadores para su aplicación en dichos procesos. El desarrollo de las líneas

de investigación anteriores ha sido posible gracias a la financiación recibida del

Ministerio de Economía, Industria y Competitividad y del Fondo Europeo de Desarrollo

Regional (FEDER) a través de los proyectos del Programa Nacional de

Investigación “Integración de procesos de fotocatálisis solar en la depuración biológica

de aguas residuales para la eliminación de contaminantes emergentes”

(CTQ2009/13459/C05/05); “Preparación de catalizadores y su aplicación en la

eliminación de contaminantes refractarios de aguas residuales mediante ozonización

fotocatalítica” (CTQ2012/35789/C02/01); y “Led y fotocatalizadores polifuncionales

basados en grafeno y estructuras metal‐orgánicas para el tratamiento de aguas por

ozonización fotocatalítica” (CTQ2015‐64944‐R). De entre los anteriores, el trabajo

de Tesis Doctoral presentado en esta memoria se encuadra fundamentalmente

dentro del proyecto CTQ2012/35789/C02/01, si bien se han abordado algunos

objetivos de los otros dos proyectos citados.

1.1.2. El agua en el mundo actual

El agua es primordial para la salud y la supervivencia de los seres vivos.

Estos necesitan el agua para realizar sus funciones vitales y sin ella no habría

vida. El agua representa entre el 50 y el 90 % de la masa de los seres vivos,

CAPÍTULO 1

12

siendo un 75 % en el caso de los seres humanos. Además, el agua promueve el

crecimiento económico y el desarrollo social de una región, de manera que la

relativa abundancia y seguridad del suministro hídrico en los países

industrializados ha sido, en gran medida, el factor que ha facilitado su

desarrollo económico [2]. A su vez, el aumento en el nivel de vida lleva

aparejado un mayor consumo de agua, por lo que ha dejado de ser un recurso

abundante y disponible en cantidad y calidad suficiente para sus diversos usos.

De toda el agua existente en la Tierra solo una pequeña parte está disponible

para su empleo y consumo, siendo, por tanto, un recurso limitado. La gran

mayoría del agua en el planeta, casi el 97,5 %, se encuentra en los mares y los

océanos y, dada su alta salinidad, no puede aprovecharse de forma directa como

recurso para la población. Del total de agua dulce (2,5 %), más de dos tercios se

encuentra acumulada en los polos. Así pues, solo una pequeña fracción, inferior

al 1 % del total del agua en la Tierra, se encuentra accesible en los ríos o en los

acuíferos, tal y como muestra la Figura 1.1 [3].

Figura 1.1. Distribución de los recursos hídricos en la Tierra [3].


13

La distribución de los recursos hídricos se basa en el ciclo del agua, en el que

el océano es el origen de la mayor parte de las precipitaciones del planeta. La

lluvia sobre la tierra satisface casi todas las necesidades de agua dulce de la

población, junto a una pequeña cantidad, continuamente en aumento, de agua

proveniente de la desalación [4,5]. No obstante, la disponibilidad de agua de

calidad no es la misma en todas partes; su distribución no es homogénea así

como tampoco lo es el sistema de lluvias en la Tierra. De hecho, existen países

húmedos y con escasa población que disponen de mayor cantidad de agua per

cápita que otros áridos y más populosos, algunos de los cuales sufren graves

problemas de escasez del recurso [6] (ver Figura 1.2). Esto se ve agravado hoy en

día por efecto del calentamiento global, el cual se prevé que dará lugar a graves

inundaciones y sequías en amplias zonas del planeta [7].

Figura 1.2. Porcentaje de agua e índice de población de cada continente sobre el total

mundial según datos de la UNESCO y el Departamento de Asuntos Económicos y

Sociales de las Naciones Unidas [6].

A la distribución heterogénea de los recursos hídricos y las consecuencias del

cambio climático se suma el aumento de la población mundial a un ritmo de 80

millones de personas al año, lo que ha triplicado las extracciones de agua en los

CAPÍTULO 1

14

últimos 50 años según cifras del Programa Mundial de Evaluación de los

Recursos Hídricos de las Naciones Unidas [8]. Si las tendencias actuales de

demografía y consumo continúan, se prevé que en 2025 dos de cada tres

personas en el mundo vivirán en zonas con estrés hídrico [8,9]. Además, los

mayores crecimientos de población estarán localizados en países en desarrollo,

tratándose en muchos casos de regiones sin acceso a agua potable ni a un

saneamiento adecuado.

De forma paralela, los diversos usos que se han dado al agua, muy

influenciados por el aumento de la densidad poblacional y los modelos de

consumo actuales, han provocado una disminución en la calidad del agua dulce

disponible. Hasta hace relativamente pocos años, gracias a su conocida

capacidad de auto‐depuración, los ríos han sido con frecuencia los receptores de

las aguas residuales generadas por el hombre. En muchos casos, los vertidos han

sobrepasado con creces la capacidad auto‐depuradora del agua, provocando una

contaminación que altera sus características y hace inservible este recurso

natural para determinados usos. El principal problema lo constituyen los

vertidos residuales procedentes de la industria, la agricultura y la ganadería,

aunque los desechos domésticos también juegan un papel destacado en la

contaminación del medio ambiente acuático [10].

En cifras reales, el 70 % del consumo total de agua en el mundo se destina a

uso agrícola y ganadero, el 22 % a uso industrial y un 8 % al gasto doméstico,

según se constata en la edición del año 2016 del Informe de las Naciones Unidas

sobre el desarrollo de los recursos hídricos en el mundo [10]. El sector productor

no sólo es el que más consume, también es el que más contamina. La necesidad

de alimentos ha llevado a la expansión del riego y a una utilización cada vez

mayor de fertilizantes y plaguicidas con el fin de lograr y mantener

rendimientos superiores. Estas sustancias son arrastradas por el agua de lluvia y

de riego, movilizando consigo desde compuestos de nitrógeno, fósforo y azufre

hasta trazas de otros compuestos tóxicos, como por ejemplo los organoclorados.

El gran contenido en nutrientes de estas escorrentías causa la eutrofización en


15

aguas superficiales y la contaminación de aguas subterráneas por infiltración en

el subsuelo [11].

Por su parte, el sector industrial produce un impacto ambiental aún mayor

debido a la introducción de una gran variedad de sustancias que pueden

contaminar el medio hídrico aportando materia orgánica, metales pesados,

radiactividad, aceites, grasas, ácidos, bases, temperatura extrema, etc. Entre las

industrias más contaminantes destacan las petroquímicas, energéticas,

papeleras, metalúrgicas, alimentarias, textiles y mineras [12].

En cuanto a la contaminación de origen doméstico, las aguas residuales que

se generan constituyen una fuente de patógenos microbianos, sólidos en

suspensión, nutrientes como el nitrógeno y el fósforo y materia orgánica, tanto

coloidal como disuelta, que consumirá oxígeno [13]. Además, cada vez es más

frecuente la aparición en estas aguas de productos químicos presentes en la

composición de fármacos, cosméticos o productos de limpieza [14].

Aparte de los anteriores, existen otros sectores fuertemente dependientes del

agua y, por tanto, fuentes de contaminación de la misma, entre los que se

encuentran la silvicultura, la pesca y la acuicultura continental, la eliminación de

desechos y la minería y la extracción de recursos [10].

En conclusión, el desarrollo demográfico e industrial de las últimas décadas

ha incrementado el uso indiscriminado de los recursos hídricos hasta niveles

insostenibles. En este escenario, surge la necesidad de tomar medidas

legislativas para proteger las aguas tanto en términos cualitativos como

cuantitativos y garantizar así su sostenibilidad. Esta legislación, cada vez más

estricta, obliga a asumir una idea clave para un nuevo enfoque de la gestión

hídrica: considerar el problema de forma global, esto es, el “ciclo integral del

agua”. De esta forma, la captación ya no es algo separado de la potabilización y

la depuración, de manera que si el agua residual se trata y se purifica

suficientemente se convierte en agua regenerada, que puede y debe ser de nuevo

utilizada regresando de este modo al punto inicial del ciclo. No se trata ya de

CAPÍTULO 1

16

buscar nuevas fuentes de agua, sino de recuperar la que se malgasta, regenerar

la que se contamina y reutilizarla [15].

1.1.2.1. Marco legal del agua

La Directiva Marco Europea del Agua (DMA) nace como respuesta a la

necesidad de unificar las actuaciones en materia de gestión de agua en la Unión

Europea. La DMA es el fruto de un proceso extenso de discusión, debate y

puesta en común de ideas entre un amplio abanico de expertos, usuarios del

agua, medioambientalistas y políticos que, por consenso, sentaron los principios

fundamentales de la gestión moderna de los recursos hídricos y que constituyen

hoy por hoy los cimientos de esta directiva. Tras un largo periodo de gestación

de más de cinco años entró en vigor el 22 de diciembre de 2000 (2000/60/CE)

[16], siendo posteriormente modificada por la Directiva 2008/105/CE [17] y, más

recientemente, por la Directiva 2013/39/UE [18].

Esta legislación europea persigue como objetivo último conseguir restablecer

las características naturales de las masas de agua, eliminando de ellas, entre

otras, las sustancias consideradas prioritarias y peligrosas. Para ello, se tienen en

cuenta aspectos como el principio de acción preventiva o el control de la

contaminación en su origen, mediante la fijación de valores límite de emisión y

de normas de calidad ambiental (NCA), entre otros. Las sustancias prioritarias

están definidas por la Directiva 2000/60/CE [16]. A través del artículo 16 de la

misma, se exige el establecimiento de un listado de estas sustancias en base a su

capacidad para representar un riesgo significativo para o a través del medio

acuático. De esta forma, la Decisión 2455/2001/CE [19] estableció la primera lista

de sustancias prioritarias, que debía ser sometida a revisión cada 4 años y

actualizada en función de las nuevas sustancias que pudieran ser identificadas

como prioritarias. Así, el anexo II de la Directiva 2008/105/CE presenta una lista

de 33 sustancias [17] entre las que figuran los metales cadmio, plomo, mercurio

y níquel y sus compuestos, el benceno, los hidrocarburos aromáticos policíclicos

y varios pesticidas [20]. De estas sustancias, 11 fueron identificadas como

peligrosas y, por tanto, sujetas a cese o eliminación progresiva de sus emisiones,


17

descargas y pérdidas en un plazo de tiempo no superior a 20 años. Además, en

el Anexo I de esta directiva también se establecen las NCA relativas a valores

máximos de concentración media y de concentración máxima admisibles de

cada sustancia, debiendo velar los Estados miembros porque se cumplan estos

estándares de calidad [17].

Con la publicación de la Directiva modificativa 2013/39/UE, a la lista inicial se

sumaron 12 sustancias más, contando por tanto con un total de 45 sustancias

prioritarias de las cuales 21 son identificadas como peligrosas. Además, en base

a los últimos conocimientos técnicos y científicos sobre sus propiedades, la

directiva actualiza las NCA de 7 de las 33 sustancias prioritarias originales [18].

Desde el 22 de diciembre de 2015 los planes hidrológicos deben tener en cuenta

estas NCA revisadas con el objeto de alcanzar un buen estado químico de las

aguas superficiales en relación con estas sustancias, a más tardar el 22 de

diciembre de 2021. Asimismo, las 12 sustancias prioritarias más recientes y sus

NCA deberán tenerse en cuenta en la elaboración de los programas de

seguimiento suplementarios y en programas preliminares de medidas a

presentar antes de que finalice 2018, con el fin de lograr un buen estado químico

de las aguas superficiales en relación con estas nuevas sustancias a más tardar el

22 de diciembre de 2027.

Para cada demarcación hidrográfica, los Estados miembros deben elaborar

un inventario de emisiones, vertidos y pérdidas de las sustancias identificadas

por esta directiva. En España, esto se gestiona a través de programas de control

de Sustancias Peligrosas, dependientes del Ministerio de Agricultura y Pesca,

Alimentación y Medioambiente [21].

Por otra parte, la Directiva 2013/39/UE introduce una estipulación según la

cual debe establecerse una lista de observación de sustancias sobre las que han

de recabarse datos de seguimiento a nivel de la Unión Europea, para que sirvan

de base a futuros ejercicios de asignación de prioridad de conformidad [18]. En

este sentido, la Decisión 495/2015/UE [22] estableció una primera lista de

CAPÍTULO 1

18

observación que contenía 10 sustancias y que deberá actualizarse cada dos años.

Se tienen en cuenta, en particular, aquellas sustancias que estuvieron a punto de

ser consideradas prioritarias en la última revisión pero para las que los datos de

seguimiento siguen siendo insuficientes como para confirmar un riesgo

significativo. También se examinaron otras sustancias nuevas sobre las que no se

disponía de datos de seguimiento recientes o estos eran insuficientes. El riesgo

que supone cada una de esas sustancias se determina a partir de la información

disponible sobre su peligrosidad intrínseca y la exposición del medioambiente a

ellas. Por su parte, dicha exposición se estima a partir de los datos sobre el

alcance de la producción y utilización, teniendo en cuenta datos de seguimiento

reales.

Cabe destacar la incorporación en la primera lista de observación de los

fármacos diclofenac, 17‐‐estradiol y 17‐‐etinilestradiol. Los fármacos se

encuentran entre las sustancias conocidas como “contaminantes emergentes o

contaminantes de preocupación emergente”, junto a otros compuestos

empleados en la industria y en la vida diaria tales como productos de higiene y

cuidado personal, drogas de abuso, los metabolitos y/o productos de

degradación de las sustancias anteriores, surfactantes, aditivos industriales,

pesticidas, retardantes de llama, esteroides y hormonas [23].

La aparición de contaminantes emergentes en las aguas se justifica en base al

cambio en los hábitos de consumo de la sociedad actual (mayor densidad y

envejecimiento de la población, mayor cuidado de la imagen exterior, aumento

de enfermedades psicológicas o de cáncer, etc.), y su presencia se ha hecho

evidente gracias al avance de las técnicas analíticas [24]. Se producen a escala

mundial, tienen multitud de aplicaciones y han llegado a ser imprescindibles

para la sociedad moderna. En la mayoría de los casos, estos compuestos son

introducidos en el medio hídrico a través de las aguas residuales urbanas,

aunque también pueden proceder de las aguas empleadas en actividades

industriales o agrícolas y ganaderas que, como puede observarse en la Figura

1.3, se descargan en las estaciones depuradoras de aguas residuales (EDAR) o


19

directamente a los ríos, receptores a su vez de los efluentes de las EDAR.

También pueden introducirse en las aguas subterráneas a través de la lixiviación

de los vertidos o de las filtraciones de las aguas superficiales. Así, se ha

identificado la presencia de un amplio número de contaminantes emergentes en

aguas superficiales, en efluentes de estaciones depuradoras de aguas residuales

[25‐28], e incluso en aguas potables de consumo humano [29‐31], claro indicador

de que los tratamientos convencionales resultan ineficaces en la eliminación de

muchos de estos compuestos [14,32‐34]. Por este motivo, si el agua regenerada

contiene estos compuestos y es utilizada en la irrigación, algunos de los

contaminantes pueden adsorberse en el suelo e incluso llegar a contaminar

aguas subterráneas [35].

Figura 1.3. Introducción de los contaminantes emergentes en el medio hídrico (adaptada

de [36]).

Si bien la concentración de estos compuestos en las aguas es baja (del orden

de ng L‐1), la exposición crónica debida a su introducción continuada puede

representar un nuevo problema medioambiental [37]. Aunque la presencia de la

mayoría de ellos en el medioambiente aún no se analiza de forma rutinaria, su

toxicidad frente a distintos organismos y microorganismos sí ha sido objeto de

CAPÍTULO 1

20

estudio en los últimos años, concluyéndose que la exposición crónica puede

causar alteraciones en la vida acuática, afectando al comportamiento y

reproducción de las especies, a la composición de las comunidades biológicas,

etc. [31,38]. Más aún, algunos trabajos revelan efectos dañinos en el crecimiento,

proliferación o material genético de células humanas [39,40]. El efecto de estos

contaminantes depende no solo de su concentración en el medio hídrico, sino

también de factores tales como su solubilidad en grasas, su persistencia,

bioacumulación, tiempo de exposición o mecanismos de biotransformación y

eliminación [20]. Así, aunque se encuentren en concentraciones muy bajas, la

exposición crónica y la acción sinérgica de estos compuestos o sus metabolitos

con otros contaminantes, unido en muchos casos a su baja biodegradabilidad y a

su posible bioacumulación en la cadena trófica, podría constituir a largo plazo

una amenaza potencial para los ecosistemas acuáticos y terrestres [14,29,41].

Cada vez con más asiduidad, los medios de comunicación se hacen eco de los

resultados de los trabajos realizados por la comunidad científica sobre la

presencia de este tipo de contaminantes en las aguas y sus posibles efectos. En

2005 la prensa española sacó a la luz pública un estudio en el que se ponían de

manifiesto fenómenos de feminización en peces en la cuenca del río LLobregat

como resultado de la presencia de compuestos con actividad estrogénica,

capaces de actuar como alteradores endocrinos [42]. En 2008 los medios

informaban sobre la presencia de medicamentos en el agua potable de un gran

número de ciudades de Estados Unidos [43]. En base a los estudios realizados

por investigadores de distintas instituciones españolas, en 2009 varios medios

alertaban sobre la presencia de drogas y sus metabolitos en el Ebro [44]. En 2012,

la prensa española recogía los resultados obtenidos por investigadores de la

Facultad de Ciencias de la Salud de la Universidad Rey Juan Carlos sobre la

presencia de fármacos en el agua embotellada, según los cuales 5 de las 10

marcas comerciales evaluadas presentaron nicotina en concentraciones de 7 a 15

ng L‐1 [45]. Más recientemente, también en nuestro país los medios

comunicación se han hecho eco del estudio realizado por expertos del Instituto

de Diagnóstico Ambiental y Estudios del Agua (IDAEA), sobre la acumulación


21

de retardantes de llama en delfines que habitan en el Golfo de Cádiz y el

estrecho de Gibraltar, productos capaces de alterar la función de las hormonas

de los cetáceos [46].

La toma de conciencia del riesgo que ocasiona la presencia de este tipo de

contaminantes en el medio ambiente es relativamente reciente. Teniendo en

cuenta que según las tendencias de consumo actuales, ni la producción y uso de

estos compuestos ni la consecuente continua introducción de los mismos en el

medio ambiente van a reducirse, se requiere una investigación más urgente. Es

necesario obtener datos suficientes para valorar de forma apropiada el impacto

de los contaminantes emergentes, determinar si pueden considerarse como

sustancias peligrosas y, en ese caso, establecer niveles máximos para su

concentración en las aguas. Actualmente, este estudio se encuentra entre las

líneas de investigación prioritarias de los principales organismos dedicados a la

protección de la salud pública y medioambiental, tales como la Organización

Mundial de la Salud (OMS), la Agencia de los Estados Unidos para la Protección

del Medioambiente (US‐EPA), o la Comisión Europea.

1.1.2.2. Conservación de los recursos hídricos

Continuamente aparecen nuevos contaminantes potencialmente peligrosos y

la demanda de agua no deja de ir en aumento. Al tratarse de una problemática

directamente relacionada con el desarrollo y la salud pública, se hace necesario

establecer una serie de pautas y acciones en materia de depuración de las aguas

residuales que permita, no solo cumplir con la normativa, sino también

implantar un ciclo de reutilización de las aguas tratadas.

La Directiva 91/271/CEE de 21 de Mayo de 1991 [47], por la que se regula el

tratamiento de las aguas residuales urbanas antes de su vertido, establecía que a

partir del año 2005 todos los núcleos urbanos de más de 2000 habitantes debían

disponer de instalaciones para la recogida y el tratamiento de las aguas

residuales. Posteriormente, en el año 2000, la Directiva Marco del Agua vino a

exigir no solo la depuración de las aguas residuales, sino conseguir que el grado

CAPÍTULO 1

22

de depuración cumpliera los parámetros establecidos para asegurar el buen

estado ecológico de todas las masas de agua, fijando finales del año 2015 como

fecha límite de cumplimiento [16]. Esto implicaba que ya no era suficiente que

una población contara con depuradora, sino que esta tenía que funcionar

adecuadamente.

En España, en respuesta a la Directiva 91/271/CEE sobre el tratamiento de las

aguas residuales, se elaboró el Plan Nacional de Calidad de las Aguas:

Saneamiento y Depuración (PNCA) [48], con un escenario temporal

comprendido entre 2007 y 2015, haciendo coincidir la fecha límite con la de la

Directiva Marco del Agua para así cumplir con la política europea en materia de

medioambiente hídrico. Entre las principales líneas de actuación, el Plan

contemplaba, además de la construcción, explotación y mantenimiento de las

depuradoras, el acondicionamiento de los sistemas de depuración a una

reducción eficaz de nutrientes mediante tratamiento terciario en las zonas

declaradas sensibles. El Plan lleva retraso en su ejecución debido en parte a la

fuerte caída de la inversión pública en infraestructuras del agua. A finales de

2015 operaban en España un total de 2940 plantas EDAR, las cuales depuraron

durante el ejercicio un volumen de 5160 hm3. Es necesaria la renovación y

adecuación tecnológica del parque de EDAR existente, así como la ampliación

futura de la red, de cara a incrementar el grado de conformidad de la carga

contaminante tratada, de acuerdo con la normativa europea [49]. La meta fijada

por el Ministerio para alcanzar los objetivos ambientales es conseguir que en

2021 el 74 % de las masas de agua alcance el buen estado y el 93 % antes de

finalizar 2027 [50].

Por lo general, en la depuración de aguas residuales urbanas suelen aplicarse

únicamente tratamientos primarios y secundarios, con los que se consigue

reducir en gran medida la contaminación de estos efluentes, pero sin lograr a

menudo cumplir la normativa vigente, cada vez más estricta. Además, tal como

se ha comentado anteriormente, muchos de los contaminantes emergentes son

refractarios a los tratamientos convencionales de las EDAR. La tendencia actual


23

es la utilización de tratamientos terciarios, entre los que destacan una o varias

etapas de oxidación química, tanto para cumplir la normativa, como para lograr

reutilizar el agua, que es el gran objetivo del futuro [51].

Con la Directiva Marco del Agua, la reutilización se plantea como una

posible medida complementaria de protección de los recursos hídricos [16]. En

este sentido, en nuestro país el Real Decreto 1620/2007 [52], por el que se

establece el Régimen Jurídico de la Reutilización de las Aguas Depuradas, ha

facilitado su desarrollo de manera que, actualmente, España se sitúa en el

primer puesto de la Unión Europea en reutilización de efluentes de depuradora

[50]. En el citado decreto se aclaran varios conceptos relativos a la reutilización

del agua hasta entonces empleados indistintamente, se decretan los usos

permitidos (urbanos, agrícolas, recreativos, industriales y ambientales) y los

prohibidos (entre los que destaca el consumo humano), fijando para cada uno de

ellos los parámetros de calidad y valores máximos permitidos y determinando el

régimen de control y responsabilidades en relación al mantenimiento de la

calidad, normalizándose los procedimientos administrativos para la obtención

del derecho al uso. Entre los parámetros para evaluar la calidad se encuentran la

carga microbiológica (nematodos intestinales, Escherichia coli, Legionella spp, etc.),

los sólidos en suspensión, la turbidez y el contenido en nitrógeno y fósforo.

En España se comenzó a tramitar el Plan Nacional de Reutilización de Aguas

(PNRA, 2012) [53] como herramienta de planificación y gestión para el fomento

de la reutilización con dos horizontes: 2015 y largo plazo. La capacidad de

regeneración de aguas depuradas prevista para 2015 se estimaba en 1000 hm3 al

año. Actualmente, los planes hidrológicos sitúan la cifra de agua regenerada en

unos 400 hm3 anuales [50], aún muy lejos de la cantidad proyectada por la

Administración. El precio de la regeneración es la principal barrera para el uso

del agua regenerada en España (solo se reutiliza el 11 % de las aguas residuales)

[50]. El PNRA se planteó sin compromisos de financiación y parece haber

sucumbido a los cambios de gobierno. Sin embargo, continúan desarrollándose

multitud de iniciativas de reutilización a escala regional y local. Gracias a estas

CAPÍTULO 1

24

iniciativas, tres cuartas partes del agua regenerada (75 %) se destina a la

agricultura, el 12 % al riego de campos de golf y jardines, el 6 % a servicios

urbanos, el 4 % a recarga de acuíferos y en torno al 3 % a un uso industrial,

según los últimos datos disponibles [50].

Los beneficios del empleo de aguas regeneradas en la agricultura son

notables, no solo por el ahorro de agua dulce que supone sino porque, además,

el agua regenerada puede contener nutrientes provechosos para la tierra de

cultivo dado el valor fertilizante de los mismos. No obstante, la reutilización no

está exenta de riesgos. Puede existir una falta de idoneidad agronómica para

riego si las aguas depuradas presentan una alta salinidad y elevadas

concentraciones de iones fitotóxicos [54]. Independientemente del uso que se le

vaya dar al agua regenerada, existen también riesgos sanitarios y

medioambientales debidos a la posible presencia de microorganismos patógenos

[55] y de compuestos orgánicos capaces de producir efectos perjudiciales a largo

plazo. Cabe destacar nuevamente la presencia de contaminantes emergentes en

las aguas regeneradas en las depuradoras, aspecto que puede poner en peligro el

futuro de la reutilización al constituir una entrada potencial de estos compuestos

en el medioambiente. Así, algunos estudios han demostrado que los

contaminantes emergentes presentes en las aguas reutilizadas son adsorbidos

por el suelo, pudiendo a continuación acumularse en las plantas cultivadas y

llegar a través de la cadena alimentaria al ser humano [29,56,57].

Por los motivos expuestos, al objeto de posibilitar la reutilización de forma

segura de los efluentes depurados y de evitar la llegada de contaminantes

emergentes a los cuerpos de agua, se hace recomendable la aplicación de un

tratamiento terciario adicional a los tratamientos convencionales de las EDAR.

El desarrollo de procesos alternativos y avanzados ha dado lugar a un amplio

campo de estudio en las últimas décadas [58].


25

1.1.3. Procesos avanzados de oxidación

Los procesos avanzados de oxidación (PAO), definidos por Glaze et al. en

1987 [59], son capaces de oxidar de manera no selectiva un amplio grupo de

contaminantes resistentes a los procesos convencionales de tratamiento de aguas

[14,60]. Los PAO se basan en la producción “in situ” de especies transitorias

energéticas, principalmente radicales hidroxilo (•

HO ), que tienen un alto poder

oxidante. Así, bajo condiciones favorables, es posible lograr la mineralización

completa (transformación hasta CO2, H2O y sales inorgánicas inocuas) de los

contaminantes presentes en el agua [61].

Los distintos PAO se diferencian entre sí por el mecanismo de producción

del radical hidroxilo. Así, se puede diferenciar entre los métodos fotoquímicos,

donde estas especies activas se generan inducidas por algún tipo de radiación; y

los no fotoquímicos, en los cuales el radical hidroxilo se produce sin necesidad

de una fuente de radiación. Esta clasificación se recoge en la Tabla 1.1.

Tabla 1.1. Principales procesos avanzados de oxidación [62].

Procesos no fotoquímicos Procesos fotoquímicos

Ozonización en medio alcalino (O3/HO•) Radiación ultravioleta con peróxido de

hidrógeno (UV/H2O2)

Ozonización en presencia de peróxido de

hidrógeno (O3/H2O2) Radiación ultravioleta y ozono (UV/O3)

Ozonización catalítica (O3/catalizador) Radiación ultravioleta con ozono y

peróxido de hidrógeno (UV/O3/ H2O2)

Procesos Fenton (Fe(II)/H2O2) y

relacionados Fotocatálisis heterogénea (UV/catalizador)

Oxidación electroquímica o

electrocatalítica

Foto‐Fenton (Fe(II)/H2O2/UV) y

relacionados

Descarga electrohidráulica‐ultrasonido UV/Fe(III) y relacionados

(UV/ferricomplejos)

Oxidación húmeda y supercrítica Fotólisis en el ultravioleta de vacío (UVV)

Radiólisis γ y tratamiento con haces de

electrones

CAPÍTULO 1

26

Cabe destacar que la combinación de dos o más PAO ha mostrado

frecuentemente efectos sinérgicos en relación con la velocidad de degradación

de los contaminantes [63,64].

El principal inconveniente de los PAO radica en el alto coste de los reactivos

necesarios tales como el ozono, el peróxido de hidrógeno o las fuentes

artificiales de radiación, por lo que su implementación se restringe a situaciones

en las que los procesos biológicos no son eficaces en la eliminación de

determinados contaminantes orgánicos [65]. No obstante, en el caso de los PAO

fotoquímicos podría emplearse la radiación natural proveniente del sol, lo que

supondría un ahorro energético considerable. Muchos de los procesos

fotoquímicos se pueden denominar procesos fotocatalíticos ya que emplean

algún tipo de catalizador. Entre estos se distinguen procesos de fotocatálisis

homogénea (e.g., sistema foto‐Fenton) y procesos de fotocatálisis heterogénea

(e.g., fotocatálisis con TiO2). Puesto que en la presente investigación se ha

aplicado la fotocatálisis heterogénea mediante el uso de distintos catalizadores y

fuentes de radiación, así como su combinación con ozono, a continuación se

describen detalladamente ambos procesos.

1.1.3.1. Fotocatálisis heterogénea

Desde la primera referencia a los procesos fotocatalíticos en 1910 por J.

Plotnikow [66], la fotocatálisis heterogénea ha sido ampliamente estudiada en el

ámbito de degradación de contaminantes, particularmente en fase acuosa.

La base de la fotocatálisis heterogénea es la foto‐excitación de un

semiconductor sólido como consecuencia de la absorción de radiación

electromagnética, procedente de una fuente lumínica con una longitud de onda

determinada (UV e incluso visible). Los semiconductores poseen una estructura

electrónica caracterizada por tener una banda de valencia (BV) llena y una

banda de conducción (BC) vacía, lo que les permite actuar en los procesos redox

inducidos por la luz. Cuando un material semiconductor adecuado es excitado

por fotones que poseen energía de magnitud suficiente para superar la energía


27

correspondiente al salto de banda (band gap), se produce la transición de

electrones desde la banda de valencia hacia la banda de conducción

generándose pares portadores de carga electrón‐hueco positivo [67]. Puesto que

los huecos de la banda de valencia son oxidantes poderosos y los electrones de

la banda de conducción son buenos reductores, estos portadores de carga son

capaces de inducir reacciones de oxidación o reducción, respectivamente [68].

La excitación del semiconductor puede tener lugar de dos formas: por

excitación directa del semiconductor (el caso más habitual y al que generalmente

hace referencia el término de fotocatálisis heterogénea), de manera que es este el

que absorbe los fotones que llegan a su superficie; o por excitación de moléculas

adsorbidas en la superficie del catalizador que, a su vez, ceden electrones al

semiconductor. En estas últimas reacciones se suele hablar de foto‐

sensibilizadores para referirse a los compuestos orgánicos adsorbidos capaces de

ceder electrones al semiconductor.

En la Figura 1.4 se muestra un esquema del mecanismo global del proceso de

fotocatálisis heterogénea que tiene lugar en una partícula de un semiconductor

de banda ancha, en disolución acuosa aireada en presencia de materia orgánica

[69]. Como se observa en el esquema, cuando un semiconductor sólido es

irradiado con luz de energía suficiente se generan pares electrón‐hueco que, en

cuestión de nanosegundos, pueden migrar a la superficie y reaccionar con las

especies adsorbidas (procesos c y d). Los que no lo hacen, se recombinan entre sí

y la energía se disipa, pudiendo tener lugar la recombinación en la superficie o

en el seno del catalizador (procesos a y b, respectivamente). Los huecos y los

electrones generados dan lugar, a través un mecanismo de reacción complejo, a

especies altamente oxidantes que desempeñan un papel clave en la degradación

de contaminantes orgánicos en disolución acuosa. Obviamente, la

recombinación del par electrón‐hueco es perjudicial para la eficacia del proceso

fotocatalítico al reducir la posibilidad de generación de especies activas.

CAPÍTULO 1

28

Figura 1.4. Esquema del proceso fotocatalítico de oxidación de materia orgánica en

disolución acuosa [69].

Profundizando algo más en el proceso, una vez generado el par

electrón/hueco en las bandas de conducción y de valencia, respectivamente

( BC BVe / h , reacción (1.1)), estas especies móviles pueden migrar a la superficie

del catalizador ( s se / h ) y/o ser fácilmente atrapadas formando estados de

menor movilidad ( T Te / h ), de acuerdo a las reacciones (1.2) y (1.3). A su vez,

estas especies pueden dar lugar a la reacción de recombinación (1.4), en la que

e y h representan todas las formas de electrones y huecos (BC, BV,

superficiales o atrapados). La recombinación supone un problema serio en el

desarrollo de las tecnologías fotocatalíticas al limitar severamente los

rendimientos cuánticos que pueden alcanzarse [69].

BC BVSemiconductor h e h (1.1)

BC s Te e e (1.2)

BV s Th h h (1.3)


29

e h Semiconductor calor (1.4)

Existe una gran controversia sobre la naturaleza de los huecos atrapados

[70,71]. En general, se ha asumido que el agua adsorbida sobre la superficie del

catalizador puede ser foto‐reducida por los sh dando lugar a radicales hidroxilo

superficiales, de acuerdo con la reacción (1.5). Sin embargo, otros autores han

demostrado que esta reacción está cinética y termodinámicamente impedida

[70], proponiéndose la reacción (1.6) para explicar la formación de huecos menos

móviles [72]. En ella, los sh son atrapados por iones oxígeno terminales de la

red del catalizador ( 2sO ) formando radicales terminales protonados o

desprotonados, dependiendo del pH.

s s 2 s sh HO (H O ) HO (1.5)

2s s s sh O HO O (1.6)

Por otro lado, en caso de existir oxígeno en la superficie de las partículas de

catalizador, este actuará como aceptor de electrones a través de la reacción (1.7),

en la que e representa los electrones en todas sus formas. Los radicales anión

superóxido ( 2O ) formados podrían recombinarse dando lugar a la aparición de

peróxido de hidrógeno (reacción (1.8)), capaz de reaccionar con electrones y/o

con radicales superóxido adicionales y generar radicales libres hidroxilo a través

de la reacción (1.9).

2 2e O O (1.7)

2 2 2 2 2O O 2H H O O (1.8)

‐2 2 2 2e O H O HO HO O (1.9)

En este mecanismo de reacción complejo (pero simplificado), los radicales

libres hidroxilo, •

HO , y/o los huecos atrapados ( T s sh HO / O ), pueden ser

CAPÍTULO 1

30

responsables de la oxidación no selectiva de la materia orgánica presente en el

medio acuoso y su posterior mineralización (reacción (1.10)). La fotocatálisis

heterogénea es uno de los PAO para el tratamiento de aguas más estudiados,

demostrando ser una tecnología muy eficiente en el tratamiento de sustancias

difícilmente biodegradables y altamente refractarias a los tratamientos

convencionales de las EDAR [69,73,74].

T T 2 2HO h C C ʹ HO h . . . CO H O (1.10)

Entre las variables que pueden afectar al proceso fotocatalítico, aparte de

parámetros propios tales como la intensidad y longitud de onda de la radiación

empleada, la geometría del reactor y la concentración de contaminantes y de

catalizador, algunos investigadores postulan que, si bien el H2O2 generado en la

reacción (1.8) podría actuar como captador de electrones a través de (1.9), es

crítico el aporte continuo de oxígeno al sistema para el mantenimiento de la

reacción (1.7), evitando así la recombinación del par e / h . Por otra parte, existe

un claro acuerdo en el efecto dominante del pH del medio dada su influencia

sobre el estado de la superficie del semiconductor, la banda de potencial y el

grado de disociación de los contaminantes.

1.1.3.2. Combinación de ozono y fotocatálisis heterogénea: ozonización fotocatalítica

La eficacia del proceso fotocatalítico puede mejorarse, tanto en términos

cinéticos (mayor velocidad de degradación de contaminantes), como

económicos (menores costes energéticos), con la presencia de otras especies

oxidantes o mediante su combinación, secuencial o simultánea, con otras

operaciones físicas o químicas de tratamiento de aguas.

En este trabajo de Tesis se ha estudiado la combinación simultánea del

proceso de fotocatálisis heterogénea empleando distintos semiconductores y

ozono (ozonización fotocatalítica). Esta combinación ha demostrado ser un PAO

muy eficaz al aumentar el rendimiento de formación de especies oxidantes en

comparación con los procesos individuales [61,75], y ha sido aplicada con éxito a


31

la degradación de diversos contaminantes emergentes en agua [76‐80].

El ozono (O3) es un agente muy reactivo de elevado poder oxidante que

puede actuar frente a compuestos que poseen determinados grupos funcionales.

En algunos casos, estas reacciones dan lugar a la formación de radicales libres

que se propagan por el medio generando, entre otros, radicales hidroxilo que,

como ya se ha comentado, son extremadamente reactivos frente a la materia

orgánica y algunos de los compuestos inorgánicos presentes en el agua [81]. Por

esta razón, las reacciones del ozono en el agua se pueden clasificar como

reacciones directas e indirectas. Las reacciones directas, también llamadas

reacciones de ozono molecular, son aquellas en las que el ozono ataca de forma

altamente selectiva determinadas moléculas orgánicas (compuestos aromáticos y

aromáticos sustituidos, moléculas con insaturaciones ‐C=C‐, ‐C≡C‐, ‐C=N, ‐C=O,

etc.) e incluso puede reaccionar con iones simples [63]. Por otro lado, las

reacciones indirectas son las reacciones en las que participan los radicales

hidroxilo generados a partir de la descomposición del ozono o de su reacción

directa con algunas especies químicas [82]. A diferencia de las reacciones

directas, las reacciones indirectas vía radicales hidroxilo son muy poco

selectivas, puesto que estas especies atacan a cualquier tipo de compuesto

oxidable presente en el agua [83]. La contribución de cada vía (reacción directa

con el ozono o a través de los radicales •

HO generados en su descomposición),

dependerá de factores tales como la naturaleza de los contaminantes, el pH del

medio y la dosis de ozono empleada, principalmente.

En el tratamiento de aguas el ozono puede emplearse como desinfectante de

agentes patógenos, como oxidante clásico para eliminar iones Fe(II) y Mn(II),

algas, olores, sabores, contaminantes orgánicos como pesticidas, detergentes,

fenoles y aminas, entre otros, y como tratamiento previo para mejorar la eficacia

de otras unidades de operación (coagulación, floculación, sedimentación,

oxidación biológica, etc.) [84]. Independientemente de la finalidad que se

persiga, en la aplicación del ozono en el tratamiento de aguas se distinguen

cuatro etapas claramente diferenciadas: 1) generación de ozono “in situ”; 2)

CAPÍTULO 1

32

transferencia del ozono desde la corriente de gas a la fase acuosa; 3) reacción

química; y 4) eliminación del ozono residual del gas de salida dada su toxicidad.

Tanto en fase gas como en disolución acuosa el ozono es relativamente

inestable. En agua destilada, a pH 7 y 20 °C los valores de vida media del ozono

varían entre 20 ‐ 30 minutos y 160 minutos, aumentando su inestabilidad en

medio básico. Esto hace que no pueda almacenarse y distribuirse para su uso,

por lo que debe ser generado en el momento de su empleo a partir de oxígeno

molecular, según se indica en la reacción (1.11) [85].

‐12 33O 2O H 284,5 kJmol

(1.11)

Se trata de una reacción endotérmica y no espontánea (∆G° = 161,3 kJ mol‐1)

[85], por lo que el ozono no solo no puede ser generado por activación térmica

del oxígeno, sino que además descompone fácilmente por calentamiento. El

método de formación de ozono de mayor aplicación es mediante descarga

eléctrica directa, generando radicales de oxígeno atómico que reaccionan con el

oxígeno molecular y forman una molécula de ozono. También puede generarse

mediante otros métodos electrolíticos o fotoquímicos.

Por otro lado, las reacciones del ozono pueden desarrollarse en fase

homogénea, encontrándose tanto el ozono como los compuestos

orgánicos/inorgánicos disueltos en agua, o en fase heterogénea, con aporte de

ozono en fase gas, en cuyo caso hay que considerar la etapa de transferencia de

materia gas‐líquido. La velocidad a la que se transfiera el ozono a la fase acuosa

dependerá de las propiedades que influyen en la absorción del ozono tales como

el pH, concentración de contaminante, fuerza iónica y temperatura, por lo que

también deben considerarse en el diseño del sistema. Por su parte, los

parámetros característicos de la transferencia de materia en un sistema de

ozonización son los coeficientes individual y volumétrico de transferencia de

materia (kL y kLa, respectivamente) y la constante de Henry del sistema, He, que

expresa la relación de equilibrio entre la concentración de ozono en el seno del

gas y su solubilidad en agua, es decir, la concentración de este en la interfase


33

gas‐líquido.

Una vez transferido el ozono a la fase acuosa o de forma simultánea a esta

etapa tienen lugar las distintas reacciones, entre ellas las reacciones indirectas

por la acción de radicales libres, formados en las etapas de iniciación o

propagación de las reacciones del ozono con otros compuestos tales como el

peróxido de hidrógeno o radiación UV. Así, se han propuesto varios

mecanismos de descomposición de ozono en agua, siendo uno de los más

aceptados el desarrollado por Staehelin, Hoigné y Buhler (SHB) [86] (reacciones

de (1.13) a (1.23)). El mecanismo se muestra a continuación incluyendo, además,

las reacciones directas del ozono y del radical hidroxilo con un contaminante

externo C (reacciones (1.12) y (1.24)).

Reacción directa:

3 oxC zO C (1.12)

Reacciones indirectas:

‐1 1k 70M s

3 2 2O HO HO O (1.13)

6 ‐1 1k 2,2x0 M s ‐

3 2 3 2O HO O HO (1.14)

pK 4,8

2 2HO H O

(1.15)

9 ‐1 1k 1,6x10 M s

3 2 3 2O O O O (1.16)

10 1 1

2 1

k 5x10 M s

3 3k 3,3x10 s

O H HO

(1.17)

5 1k 1,1x10 s

3 2HO HO O (1.18)

5 1 1

7 1 1k 8,3x10 M s

o bien k 9,7x10 M s

2 2 2 2 2 2 2HO HO obienO O H H O O

(1.19)

9 ‐1 1k 2x10 M s

3 2 2O HO HO O (1.20)

CAPÍTULO 1

34

pK 11,3

2 2 2H O HO H (1.21)

7 ‐1 1k 2,7x10 M s

2 2 2 2H O HO HO H O (1.22)

9 ‐1 1k 7 ,5x10 M s

2 2 2HO HO O H O (1.23)

ox 2 2C HO C . . . CO H O (1.24)

En estas reacciones pueden participar, a su vez, una serie de especies

llamadas iniciadoras, promotoras e inhibidoras, que conducen a la

descomposición o estabilización del ozono en agua [87]. Las sustancias

iniciadoras, que como su propio nombre indica inician el mecanismo de

descomposición en cadena, son aquellas que reaccionan directamente con el

ozono, como por ejemplo el ion hidroperóxido ( 2HO) que, a través de la

reacción (1.14), da lugar a la formación del radical ozónido ( •

3O ) y el radical

hidroperóxido (•

2HO ) el cual, de acuerdo con (1.15), se encuentra en equilibrio

con el radical superóxido (•

2O ). Este radical reacciona rápidamente con el ozono

generando más radicales ozónido a través de (1.16). El •

3O formado a través de

(1.14) y (1.16) promueve la aparición del radical •

3HO según el equilibrio (1.17) y,

con ello, la formación de radicales hidroxilo a través de (1.18). Por su parte, las

sustancias promotoras son compuestos orgánicos o inorgánicos que reaccionan

con el radical hidroxilo dando lugar al radical superóxido o hidroperóxido

(dependiendo del pH del medio). Dichos radicales •

2O /

•

2HO se incorporan al

mecanismo de reacción en cadena de descomposición del ozono,

promoviéndolo. Algunos promotores son el ion fosfato, el metanol o el ácido

fórmico. Finalmente, las sustancias inhibidoras son aquellas que reaccionan con

el radical hidroxilo y terminan con las reacciones en cadena, es decir, no dan

lugar al radical hidroperóxido o superóxido, como por ejemplo el ter‐butanol,

determinados iones como carbonatos y bicarbonatos o ciertas sustancias

húmicas.


35

En este tipo de sistemas hay que tener en cuenta que no todo el ozono en el

gas de alimentación se transfiere al agua, existiendo un remanente en el gas de

salida que, por sus propiedades tóxicas, no puede liberarse directamente a la

atmósfera y debe ser eliminado. Para ello, las plantas de tratamiento que

emplean ozono disponen de unos reactores catalíticos de lecho fijo a través de

los cuales se hace circular el gas de salida, destruyendo el ozono e impidiendo

así su emisión al exterior.

Las reacciones directas entre los contaminantes orgánicos y el ozono suelen

ser de segundo orden (primer orden respecto a cada reactante). Aunque existen

compuestos refractarios al ozono, como es el caso del tricloroacetato (kO3‐C =

3x10‐5 M‐1 s‐1), el ter‐butanol (kO3‐C = 3x10‐3 M‐1 s‐1), el etanol (kO3‐C = 0,37 M‐1 s‐1) o

el benceno (kO3‐C = 2 M‐1 s‐1), entre otros [88], los compuestos que en su estructura

presentan insaturaciones o anillos aromáticos (y en especial si en estos últimos

existen sustituyentes activantes como ocurre, por ejemplo, con los fenoles),

reaccionan de forma rápida con el ozono con valores de kO3‐C del orden de 102 ‐

107 M‐1 s‐1. Por su parte, las reacciones entre los contaminantes orgánicos y el

radical hidroxilo, también de segundo orden, tienen lugar a velocidades muy

superiores, con valores de kHO‐C comprendidos entre 107 ‐ 1010 M‐1 s‐1 [82].

Volviendo al conjunto de reacciones (1.12) a (1.24), es importante destacar el

papel del peróxido de hidrógeno en el mecanismo de descomposición del ozono,

tanto por reacción directa entre su forma iónica y el ozono molecular a través de

(1.14), actuando como iniciador, como en la captura de radicales hidroxilo

promoviendo la descomposición mediante (1.22) y (1.23). Por tanto, la

combinación de peróxido de hidrógeno y ozono constituye por sí sola un PAO,

al igual que la combinación de ion hidróxido (HO‾) y ozono (ozonización en

medio básico), según indica la reacción (1.13).

Por otro lado, si el ozono es irradiado con luz UV suficientemente energética

(longitudes de onda inferiores a 310 nm), puede generarse oxígeno atómico en

estado excitado (O(1D) (reacción (1.25)), especie capaz de generar peróxido de

CAPÍTULO 1

36

hidrógeno al reaccionar con el agua a través de (1.26) [89]. A su vez, la fotólisis

del peróxido de hidrógeno podría dar lugar a la formación del radical hidroxilo

a través de la reacción (1.27). Sin embargo, aunque el rendimiento cuántico de

esta reacción en el intervalo de longitudes de onda de 200 ‐ 400 nm es elevado (1

mol einstein‐1, [90]), la absortividad molar del H2O2 para valores de λ superiores

a 310 nm es muy débil, inferior a 0,51 M‐1 cm‐1 [91]. En consecuencia, el

desarrollo de estas reacciones dependerá de la longitud de onda de la radiación

empleada, de manera que a longitudes de onda superiores a 310 nm la reacción

(1.27) perderá importancia aunque la generación de peróxido de hidrógeno

seguirá favoreciendo la descomposición del ozono igualmente.

h 310nm 1

3 2O O O D

(1.25)

12 2 2O D H O H O (1.26)

h 200 ‐ 280nm

2 2H O 2HO

(1.27)

Por su parte, el oxígeno atómico formado a través de (1.25) no solo es capaz

de reaccionar con el agua (reacción (1.26), k = 1,8x1010 M‐1 s‐1), sino que presenta

valores de k del orden de 109 ‐ 1010 M‐1 s‐1 con prácticamente todos los sustratos

[92].

Además de la formación de oxígeno atómico, hay evidencias de la aparición

en fase gas de otros estados del oxígeno a través de 5 posibles vías, en función

de la longitud de onda:

309nm 1 13 2 gO O D O

(1.28)

3411nm 13 2 gO O D O

(1.29)

31180nm 33 2 gO O P O

(1.30)


37

1463nm 33 2 gO O P O

(1.31)

612nm 3 13 2 gO O P O

(1.32)

En disolución acuosa estas reacciones también podrían ocurrir. A longitudes

de onda inferiores a 300 nm los productos principales son el estado excitado

(O(1D)) (con un rendimiento cuántico próximo a 0,9 mol einstein‐1), el oxígeno

singlete (O2(1∆g)) y el oxígeno en su estado fundamental (O2(3∑g‾)) (reacciones

(1.28) y (1.29)) [93‐96]. En comparación, las reacciones (1.30) y (1.31) son menos

importantes ya que el rendimiento cuántico de la fotólisis del ozono disminuye

hasta valores cercanos a 0,1 mol einstein‐1 para longitudes de onda superiores a

320 nm.

En cuanto al oxígeno singlete, este puede reaccionar con el ozono y generar

más especies atómicas:

1 32 g 3 2O O 2O O P (1.33)

El O(3P) formado puede, a su vez, desaparecer por reacción con el O2(3∑g‾)

(reacción (1.34)) [97] o con el ozono (reacción (1.35), k ≈ 10‐3 M‐1 s‐1), aunque se ha

comprobado que en el agua esta última reacción no tiene lugar [98].

9 1 13 k 4x10 M s32 g 3O P O O

(1.34)

33 2O P O 2O (1.35)

Finalmente, cuando el ozono está presente en un proceso de fotocatálisis

heterogénea tendrán lugar, además de los mecanismos de reacción

correspondientes a cada sistema por separado, otros mecanismos de generación

de radicales por interacción entre los distintos sistemas. En este sentido, destaca

el papel del ozono como captador de los e generados en la fotoexcitación del

semiconductor con formación del anión radical ozónido (reacción (1.36)),

potenciando así las reacciones (1.17) y (1.18) y, con ello, la aparición de radicales

CAPÍTULO 1

38

•HO [99,100].

3 3e O O (1.36)

La captura de los e por parte del ozono tiene lugar con velocidades

superiores a las que presenta el O2, de manera que mientras que la constante

cinética de la reacción (1.36) a pH comprendido entre 4 y 6 es de 3,6x1010 M‐1 s‐1,

en el caso del O2 (reacción (1.7)) esta resulta ser de 1,76x1010 M‐1 s‐1 [67]. En

consecuencia, en presencia de ozono el grado de recombinación de electrones y

huecos positivos puede verse disminuido, por una parte gracias a la

participación de (1.36) y, por otra, a una mayor contribución de la reacción (1.9)

dada la mayor concentración esperada de H2O2 en el medio procedente tanto de

las reacciones directas con ozono como de la reacción (1.19) [101‐103]. Además

de lo anterior, en presencia de oxígeno y ozono el ion radical superóxido,

generado tanto en la fotocatálisis como en la descomposición de ozono

(reacciones (1.7), (1.14) y (1.15)), puede reaccionar con ozono y generar el ion

radical ozónido a través de la reacción (1.16), desembocando en la formación

adicional de radicales hidroxilo.

En conclusión, atendiendo a los mecanismos de reacción indicados cabe

esperar la existencia de sinergia entre el ozono y la fotocatálisis heterogénea,

dada la gran cantidad de especies oxidantes formadas y la minimización del

proceso de recombinación de electrones y huecos.

1.1.4. Optimización de los procesos fotocatalíticos: empleo de radiación solar y

fotocatalizadores apropiados

Uno de los principales inconvenientes a los que se enfrentan los procesos

fotocatalíticos es el alto coste derivado del uso de lámparas UV. Por este motivo,

el empleo de radiación natural proveniente del sol en reactores con un diseño

adecuado supondría un ahorro energético considerable, minimizando los costes

de operación asociados a este tipo de tratamientos [69]. No obstante, hay que


39

tener en cuenta que, a diferencia de la luz artificial, la intensidad de la radiación

solar que llega a la superficie terrestre no es constante, sino que se ve afectada

por factores meteorológicos, varía diaria y estacionalmente, depende de la

localización geográfica, etc. La interacción entre la radiación solar y los distintos

componentes atmosféricos (O3, O2, CO2, H2O, aerosoles, etc.) provoca que la

radiación que llega a la superficie de la tierra quede restringida a longitudes de

onda comprendidas entre 300 y 3000 nm [104]. De toda esta radiación, un 5 %

corresponde a la región ultravioleta (95% UVA y 5% UVB), un 46 % a la visible y

un 49 % a la del infrarrojo cercano [105].

Los resultados de las investigaciones sobre el uso de luz solar en los procesos

fotocatalíticos de detoxificación de aguas, iniciada en los años 90 en la

Plataforma Solar de Almería [106], reflejan la eficacia de estos procesos en la

degradación de compuestos en agua, siendo el proceso Fenton y la fotocatálisis

heterogénea los tratamientos más estudiados. Ambos sistemas se han ensayado

con éxito para la eliminación de multitud de contaminantes, muchos de ellos no

biodegradables y/o que por su toxicidad pueden impedir el correcto

funcionamiento de los tratamientos biológicos [69,107,108]. En los últimos años,

la problemática asociada a los contaminantes emergentes ha dado lugar a un

aumento significativo en el número de estudios sobre su eliminación mediante

detoxificación solar [37,109‐112]. En el caso concreto del proceso de ozonización

fotocatalítica solar, estudios recientes informan del éxito de su aplicación en la

degradación de diversos contaminantes emergentes en agua [78,113‐118].

En cuanto a los catalizadores, existen diversos materiales semiconductores

tales como TiO2, ZnO, CdS, FexOy, WO3, CeO2, etc. Se trata de materiales

económicamente asequibles que, en su mayoría, pueden excitarse con radiación

de no muy alta energía (UVA e incluso visible), lo que incrementa el interés en

su empleo en procesos de detoxificación solar. En la Figura 1.5 se muestran las

posiciones de la banda de conducción y de valencia de varios semiconductores,

a partir de las cuales es posible deducir su energía del salto de banda (Eg).

CAPÍTULO 1

40

5

4

3

2

1

0

-1

-2g-C3N4

1,80

-0,90 -0,40

CdS 2,00

-0,75

2,65

0,79

-Fe2O3 2,99

3,47

WO3 0,77

3,75

CeO2

3,13

-0,07

ZnO 3,41

0,21 -0,16

︵rutilo ︶

TiO2

3,04

0,04

pH=0

V vs. ENH

H2O/HO·Eº = +2,73 V

Eº = +0,08 V O2/·O2-

︵anatasa ︶

TiO2

3,04

SnO2

0,25

Figura 1.5. Posiciones de banda (arriba BV y abajo BC) de varios semiconductores

(adaptado de [128]).

En los últimos 30 años, el dióxido de titanio (TiO2) en forma de polvo

policristalino ha sido el semiconductor empleado por excelencia en la

investigación de la descontaminación de aguas mediante procesos fotocatalíticos

debido a su actividad, bajo coste, estabilidad y no toxicidad [69,113,119‐121]. El

producto comercial más conocido es el dióxido de titanio P25 de Degussa

(actualmente Evonik Degussa), constituido por un 20 % de rutilo y un 80 % de

anatasa, siendo esta última la fase cristalina más activa en fotocatálisis. Posee un

área superficial específica de 50 ± 15 m2 g‐1 y un tamaño medio de partícula

cristalina de 20 a 40 nm [122]. Este material ha resultado ser uno de los más

activos en la degradación fotocatalítica de contaminantes del agua empleando

diferentes fuentes de radiación ultravioleta [123‐125]. En lo que respecta a su

empleo en la degradación de compuestos mediante ozonización fotocatalítica,

distintos trabajos ponen de manifiesto su efectividad [78,79,117,118,126,127].

Sin embargo, el TiO2 presenta varios inconvenientes para su aplicación a gran

escala empleando luz solar. El principal de ellos es que al corresponder su

energía del salto de banda (3,0 ‐ 3,2 eV) con longitudes de onda de 387 ‐ 410 nm,

la energía del espectro solar que puede aprovecharse se ve reducida

aproximadamente a un 5 % de la radiación que llega a la Tierra, concretamente a


41

la fracción UV [69,129]. A esto se suma la dificultad de reutilización del

catalizador en polvo debido a la difícil separación del líquido y, además, una

alta capacidad de recombinación. Como se ha comentado en párrafos anteriores,

esto último podría minimizarse mediante el empleo de ozono como oxidante

[69,79,80].

Actualmente se están estudiando distintas estrategias que permitan solventar

o reducir los problemas anteriores. En cuanto al aprovechamiento de la

radiación solar, existen otros semiconductores alternativos al TiO2 que, por su

salto de banda, son aplicables bajo luz solar. Entre ellos se encuentran el óxido

de zinc (ZnO) y el sulfuro de cadmio (CdS) [129], si bien en disolución acuosa

estos catalizadores han presentado algunos problemas de lixiviación del catión

metálico al medio por efectos de fotocorrosión [129].

El óxido de wolframio (WO3), presenta también un salto de banda menor que

el TiO2 (2,4 ‐ 2,8 eV) que permite la absorción de radiación visible. Además, su

fácil preparación, toxicidad nula y su estabilidad en procesos de oxidación

fotocatalíticos hacen de este semiconductor un buen candidato frente al

fotocatalizador tradicional [130‐133]. Sin embargo, dado que el nivel energético

de su banda de conducción es más positivo que el nivel de reducción del

oxígeno (ver Fig. 1.5), este semiconductor por sí solo no resulta eficiente en

fotocatálisis empleando oxígeno, al carecer este último del potencial suficiente

para capturar de forma eficaz los electrones y evitar la recombinación electrón‐

hueco [130,131]. Sin embargo el ozono, que posee un mayor poder oxidante, sí es

capaz de reaccionar rápidamente con los electrones de la banda de conducción

del WO3 y, por tanto, este material sí resulta eficiente en el proceso de

ozonización fotocatalítica como ya han demostrado los estudios de Nishimoto et

al. [134,135].

Por otra parte, el óxido de cerio (CeO2) es un semiconductor que puede

absorber radiación en el ultravioleta cercano e incluso ligeramente en la región

del visible ya que, en función de la morfología y el tamaño de partícula del

CAPÍTULO 1

42

semiconductor, presenta un salto de banda que varía entre 2,9 y 3,2 eV [136‐138].

Además, los pares electrón‐hueco fotogenerados poseen una vida útil lo

suficientemente larga como para desencadenar reacciones fotocatalíticas tanto

en fase líquida como en fase gas [139].

El CeO2 se ha utilizado como fotocatalizador en la degradación numerosos

contaminantes orgánicos en agua [137,140‐143]. Los trabajos realizados ponen de

manifiesto, como se ha indicado anteriormente, la importancia de la morfología

y del tamaño de partícula en su comportamiento como fotocatalizador. Por otra

parte, el CeO2 ha resultado activo en la ozonización catalítica de compuestos

orgánicos de diferente naturaleza [144‐146]. Dada la eficiencia de los procesos

individuales de fotocatálisis y ozonización catalítica empleando CeO2, es de

esperar un efecto sinérgico en el proceso combinado de ozonización

fotocatalítica, si bien no existe hasta la fecha ningún estudio al respecto.

Además del empleo de semiconductores distintos del TiO2, otra posible

estrategia para mejorar la absorción en el visible es sintetizar materiales

compuestos de dos semiconductores, por lo general acoplando al TiO2 un

semiconductor con una energía del salto de banda menor que la suya [69]. Así,

por ejemplo, al ser la banda de conducción del WO3 menos negativa que la del

TiO2 los electrones pueden transferirse desde la banda del TiO2 a la del WO3,

produciéndose una separación de cargas electrón/hueco más efectiva y evitando

así en cierta medida el proceso de recombinación [147,148]. En la Figura 1.6 se

muestra un esquema del proceso fotocatalítico que tendría lugar en el material

compuesto TiO2‐WO3.

Este tipo de materiales se ha aplicado en la eliminación fotocatalítica de

distintos compuestos en agua, demostrando tener una actividad superior al TiO2

bajo radiación visible [149,150]. Sin embargo, no existen estudios sobre su

aplicación en el proceso de ozonización fotocatalítica.


43

Figura 1.6. Proceso de excitación en el sistema fotocatalítico TiO2‐WO3 (adaptada de [69]).

Por otra parte, en los últimos años existe un interés creciente por controlar la

morfología de los fotocatalizadores a niveles nanométricos. Si se les da otra

forma y otro tamaño, es decir, si se modifican sus dimensiones, es de esperar

que varíen sus propiedades fotocatalíticas. Se trata de intentar obtener un

semiconductor con una mayor relación área superficial/volumen para aumentar

con ello la eficiencia fotocatalítica. A modo de ejemplo, el TiO2 en forma de

nanotubos ha demostrado poseer una alta superficie específica, una estrecha

distribución de tamaño de poro, unas propiedades electrónicas que mejoran el

transporte de carga y una alta capacidad de intercambio iónico [151]. Todas

estas propiedades pueden favorecer de una forma u otra su actividad

fotocatalítica. Los nanotubos de titanio pueden sintetizarse mediante diversos

métodos, llegándose a conseguir áreas específicas de hasta 500 m2 g‐1 [151‐153].

En cuanto a la estructura cristalina de los nanotubos, en función de las

condiciones de síntesis se ha encontrado la forma tetragonal anatasa [153] y

también otras estructuras monoclínicas u ortorrómbicas [154,155]. Aunque los

resultados obtenidos empleando nanotubos de TiO2 como fotocatalizadores en

el tratamiento de aguas (utilizando colorantes como compuesto a degradar

CAPÍTULO 1

44

[151]), no han sido muy prometedores, sí son interesantes como soporte de

distintas nanopartículas metálicas u otros semiconductores [156]. En lo que

respecta a su aplicación en ozonización fotocatalítica hasta la fecha se ha

publicado un único trabajo [157].

Por último, es posible favorecer la recuperación y reutilización de los

catalizadores mediante la síntesis de materiales soportados, si bien el

fotocatalizador pierde eficiencia al disminuir la superficie activa [69]. Entre los

soportes comúnmente utilizados se encuentran la sílice, vidrios de distinta

morfología, algunos polímeros y membranas poliméricas, carbón activado, etc.

[158]. Una estrategia muy atractiva que se está investigando en los últimos años

es soportar el catalizador (TiO2 generalmente) sobre partículas magnéticas, ya

sean nanopartículas de magnetita [159,160] o carbón activado magnetizado

[78,161,162], lo que permite separar el catalizador mediante un imán externo.

Los materiales carbonosos introducen, además, las propiedades adicionales del

carbón tales como una alta superficie específica y gran volumen de poros,

proporcionando una elevada capacidad de adsorción de compuestos orgánicos

[162,163]. Asimismo, se ha demostrado que algunos materiales carbonosos

podrían actuar donando electrones al semiconductor, inhibiendo los procesos de

recombinación [164].

Aunque todas estas alternativas se han desarrollado y aplicado en la

oxidación fotocatalítica de distintos contaminantes, la bibliografía sobre su

aplicación en la ozonización fotocatalítica es escasa. Por ello, el trabajo que

recoge esta memoria de Tesis persigue el desarrollo de materiales activos y

estables, con vistas a su aplicación en la eliminación de compuestos emergentes

en las aguas mediante ozonización fotocatalítica empleando luz solar.

1.2. OBJETIVOS Y ALCANCE DEL TRABAJO

Dada la preocupación mundial por la presencia de contaminantes

emergentes en los ecosistemas acuáticos, esta investigación tiene como objetivo

principal contribuir al desarrollo de estrategias más eficientes para su


45

degradación. Se pretende, además, que los posibles procesos a aplicar resulten

atractivos desde el punto de vista económico, empleando para ello la luz solar

como fuente de radiación y sintetizando fotocatalizadores que permitan un

mayor aprovechamiento de la misma.

El trabajo de Tesis se ha centrado en el proceso de ozonización fotocatalítica,

si bien ha sido necesaria la aplicación de otros sistemas con fines comparativos

(fotólisis, ozonización, ozonización fotolítica, adsorción, oxidación fotocatalítica

y ozonización catalítica). Todos los ensayos se han realizado a escala de

laboratorio, empleando luz solar simulada y fotocatalizadores apropiados

previamente sintetizados y/o comerciales.

Para lograr el objetivo general del presente trabajo, se han planteado los

siguientes objetivos específicos:

OBJETIVO 1: Evaluar y comparar la producción de especies oxidantes

fotogeneradas durante los procesos de fotocatálisis y de ozonización

fotocatalítica, determinando el rendimiento cuántico de ambos procesos con

vistas a establecer la sinergia entre el ozono y el proceso de fotocatálisis.

OBJETIVO 2: Preparar y caracterizar fotocatalizadores que puedan ser

fotoexcitados bajo radiación solar, así como mejorar el aprovechamiento de esta

radiación por parte del fotocatalizador más comúnmente utilizado, TiO2,

sintetizando materiales compuestos, evaluando el efecto de las distintas

variables de síntesis a fin de encontrar las formulaciones óptimas para su

empleo en procesos de ozonización fotocatalítica solar.

OBJETIVO 3: Estudiar la degradación de distintos contaminantes emergentes en

agua mediante ozonización fotocatalítica con luz solar, empleando los

catalizadores preparados.

OBJETIVO 4: Desarrollar un modelo cinético para el proceso de ozonización

fotocatalítica solar en base a la identificación de los intermedios de degradación

CAPÍTULO 1

46

de un compuesto modelo y de las principales especies involucradas en el

proceso.

1.3. ORGANIZACIÓN DE LA MEMORIA

Este trabajo de Tesis se estructura en 9 capítulos. En este primero se ha

introducido el trabajo y se han definido los objetivos del mismo. El segundo

capítulo está destinado a las instalaciones y técnicas experimentales empleadas

durante la investigación. En los siguientes capítulos se presentan, como una

colección de artículos de investigación ya publicados en revistas científicas

internacionales, los resultados de los estudios realizados para la consecución de

los objetivos propuestos. Dichos capítulos son los siguientes:

‐ Capítulo 3: On ozone‐photocatalysis synergism in black‐light induced

reactions: Oxidizing species production in photocatalytic ozonation versus

heterogeneous photocatalysis.

‐ Capítulo 4: Influence of structural properties on the activity of WO3

catalysts for visible light photocatalytic ozonation.

‐ Capítulo 5: Visible light photocatalytic ozonation of DEET in the presence

of different forms of WO3.

‐ Capítulo 6: Nanostructured CeO2 as catalysts for different AOPs based in

the application of ozone and simulated solar radiation.

‐ Capítulo 7: WO3‐TiO2 based catalysts for the simulated solar radiation

assisted photocatalytic ozonation of emerging contaminants in a municipal

wastewater treatment plant effluent.

‐ Capítulo 8: Reaction mechanism and kinetics of DEET visible light assisted

photocatalytic ozonation with WO3 catalyst.

Finalmente, en el capítulo 9 se indican las conclusiones más relevantes del

estudio realizado.


47

BIBLIOGRAFÍA

[1] www.unex.es/investigacion/grupos/trataguas (consultada en febrero de

2017).

[2] Radovic, L.R.; Moreno‐Castilla; C.; Rivera‐Utrilla; J. “Carbon materials as

adsorbents in aqueous solutions” en: “Chemistry and Physics of Carbon”.

Radovic L.R.; Dekker M. (eds.), New York (2000), 27, 227‐403.

[3] Sánchez‐Vila, X. en “Aguas continentales: Gestión de recursos hídricos,

tratamiento y calidad del agua”. Informes CSIC, D. Barceló (coord.), CSIC,

Madrid (2008), Cap. 2.

[4] Li, C.; Goswami, Y.; Stefanakos, E. “Solar assisted sea water desalination: A

review”. Renew. Sust. Energy Rev. 19 (2013) 136‐163.

[5] Xiao, C.; Wang, X.; Ni, M.; Wang, F.; Zhu, W.; Luo, Z.; Cen, K. “A review on

solar stills for brine desalination”. Appl. Energy 103 (2013) 642‐652.

[6] www.aysa.com.ar/index.php?id_contenido=323&id_seccion=0 (consultada en

febrero de 2017).

[7] Carrera, J. en “Aguas continentales: Gestión de recursos hídricos, tratamiento

y calidad del agua”. Informes CSIC, D. Barceló (coord.), CSIC, Madrid (2008),

Cap. 1.

[8] www.unesco.org/new/es/natural‐sciences/environment/water/wwap/facts‐

and‐figures/all‐facts‐wwdr3/fact1‐demographics‐consumption/ (consultada en

febrero de 2017).

[9] Blanco, J.; Malato, S.; Fernández‐Ibáñez, P.; Alarcón, D.; Gernjak, W.;

Maldonado, M.I. “Review of feasible solar energy applications to water

processes”. Renew. Sust. Energy Rev. 13 (2009) 1437‐1445.

[10] Informe de las Naciones Unidas sobre el Desarrollo de los Recursos

Hídricos en el Mundo: Agua y Empleo (2016).

unesdoc.unesco.org/images/0024/002441/244103s.pdf (consultada en febrero de

2017).

[11] Barros, R.; Isidoro, D.; Aragüés, R. “Irrigation management, nitrogen

fertilization and nitrogen losses in the return flows of La Violada irrigation

district (Spain)”. Agr. Ecosyst. Environ. 155 (2012) 161‐171.

[12] Wasi, S.; Tabrez, S.; Ahmad, M. “Toxicological effects of major

environmental pollutants: an overview”. Environ. Monit. Assess. 185 (2013)

2585‐2593.

[13] Osorio, F.; Torres, J.C.; Sánchez, M. en “Tratamiento de aguas para la

CAPÍTULO 1

48

eliminación de microorganismos y agentes contaminantes”. Ediciones Díaz de

Santos, Madrid (2010).

[14] Klavarioti, M.; Mantzavinos, D.; Kassinos, D. “Removal of residual

pharmaceuticals from aqueous systems by advanced oxidation processes”.

Environ. Int. 35 (2009) 402‐417.

[15] Bayo, I.F.; Angulo, E. “Gota a gota”. Química e Industria 582 (2009) 16‐24.

[16] Directiva 2000/60/CE del Parlamento Europeo y del Consejo, de 23 de

octubre de 2000 por la que se establece un marco comunitario de actuación en

el ámbito de la política de aguas.

[17] Directiva 2008/105/CE del Parlamento Europeo y del Consejo, de 16 de

diciembre de 2008, relativa a las normas de calidad ambiental en el ámbito de la

política de aguas, por la que se modifica la Directiva 2000/60/CE.

[18] Directiva 2013/39/UE del Parlamento Europeo y del Consejo, de 12 de

agosto de 2013 , por la que se modifican las Directivas 2000/60/CE y 2008/105/CE

en cuanto a las sustancias prioritarias en el ámbito de la política de aguas.

[19] Decisión nº 2455/2001/CE del Parlamento Europeo y del Consejo de 20 de

noviembre de 2001, por la que se aprueba la lista de sustancias prioritarias en el

ámbito de la política de aguas, y por la que se modifica la Directiva 2000/60/CE.

[20] Esplugas, S.; Bila, D.M.; Krause, L.G.T.; Dezotti, M. “Ozonation and

advanced oxidation technologies to remove endocrine disrupting chemicals

(EDCs) and pharmaceuticals and personal care products (PPCPs) in water

effluents”. J. Hazard. Mater. 149 (2007) 631‐642.

[21] www.mapama.gob.es/es/cartografia‐y‐sig/ide/descargas/agua/red‐

programa‐control.aspx (consultada en febrero de 2017).

[22] Decisión de Ejecución (UE) 2015/495 de la Comisión de 20 de marzo de 2015

por la que se establece una lista de observación de sustancias a efectos de

seguimiento a nivel de la Unión en el ámbito de la política de aguas, de

conformidad con la Directiva 2008/105/CE del Parlamento Europeo y del

Consejo.

[23] Díaz‐Cruz, M.S.; García‐Galán, M.J.; Guerra, P.; Jelic, A.; Postigo, C.;

Eljarrat, E.; Farré, M.; López de Alda, M.J.; Petrovic, M.; Barceló, D. “Analysis of

selected emerging contaminants in sewage sludge”. Trends Anal. Chem. 28

(2009) 1263‐1275.

[24] Rodríguez‐Mozaz, S.; López de Alda, M.J.; Barceló, D. “Advantages and

limitations of on‐line solid phase extraction coupled to liquid chromatography‐

mass spectrometry technologies versus biosensors for monitoring of emerging


49

contaminants in wáter”. J. Chromatogr. A 1152 (2007) 97‐115.

[25] Karthikeyan, K.G; Meyer, M.T. “Occurrence of antibiotics in wastewater

treatment facilities in Wisconsin, USA”. Sci. Total Environ. 361 (2006) 196‐207.

[26] Lishman, L.; Smyth, S. A.; Sarafin, K.; Kleywegt, S.; Toito, J.; Peart, J.; Lee, B.;

Servos, M.; Beland, M.; Seto, P. “Occurrence and reductions of pharmaceuticals

and personal care products and estrogens by municipal wastewater treatment

plants in Ontario, Canada”. Sci. Total Environ. 367 (2006) 544‐558.

[27] Rosal, R.; Rodríguez, A.; Perdigón‐Melón, J.A.; Mezcua, M.; Hernando,

M.D.; Letón, P.; García‐Calvo, E.; Agüera, A.; Fernández‐Alba, A.R. “Removal

of pharmaceuticals and kinetics of mineralization by O3/H2O2 in a biotreated

municipal wastewater.” Water Res. 42 (2008) 3719‐3728.

[28] Rosal, R.; Rodríguez, A.; Perdigón‐Melón, J.A.; Petrea, A.; García‐Calvo, E.;

Gómez, M.J.; Agüera, A.; Fernández‐Alba, A.R. “Occurrence of emerging

pollutants in urban wastewater and their removal through biological treatment

followed by ozonation”. Water Res. 44 (2010) 578‐588.

[29] Rodríguez‐Gil, J.L.; Catalá, M.; González Alonso, S.; Romo Maroto, R.;

Valcárcel, Y.; Segura, Y.; Molina, R.; Melero, J.A.; Martínez, F. “Heterogeneous

photo‐Fenton treatment for the reduction of pharmaceutical contamination in

Madrid rivers and ecotoxicological evaluation by a miniaturized fern spores

bioassay”. Chemosphere 80 (2010) 381‐388.

[30] Kasprzyk‐Hordern, B.; Dinsdale, R.M.; Guwy, A.J. “The removal of

pharmaceuticals, personal care products, endocrine disruptors and illicit drugs

during wastewater treatment and its impact on the quality of receiving

waters”. Water Res., 43 (2009) 363‐380.

[31] Hazelton, P.D.; Cope, W.G.; Mosher, S.; Pandolfo, T.J.; Belden, J.B.; Barnhart,

M.C.; Bringolf, R.B. “Fluoxetine alters adult freshwater mussel behavior and

larval metamorphosis”. Sci. Total Environ. 445‐446 (2013) 94‐10.

[32] Ikehata, K.; Jodeiri Naghashkar, N.; Gamal El‐Din, M. “Degradation of

aqueous pharmaceuticals by ozonation and advanced oxidation processes: A

review”. Ozone Sci. Eng. 6 (2006) 353‐414.

[33] Oulton, R.L.; Kohn, T.; Cwiertny, D.M. “Pharmaceuticals and personal care

products in effluent matrices: A survey of transformation and removal during

wastewater treatment and implications for wastewater management”. J.

Environ. Monit. 12 (2010) 1956‐1978.

[34] Verlichi, P.; Al Aukidy, M.; Zambello, E. “Ocurrence of pharmaceutical

compounds in urban wastewater: removal, mass load and environmental risk

CAPÍTULO 1

50

after a secondary treatment ‐ A Review”. Sci. Total Environ. 429 (2012) 123‐155.

[35] Xu, J.; Wu, L.; Chang, A.C. “Degradation and adsorption of

selectedpharmaceuticals and personal care products (PPCPs) un agricultural

soils”. Chemosphere 77 (2009) 1299‐1305.

[36] Petrovic, M.; González, S.; Barceló, D. “Analysis and removal of emerging

contaminants in wastewater and drinking water”. Trends Anal. Chem. 22 (2003)

685‐696.

[37] Bernabeu, A.; Vercher, R.F.; Santos‐Juanes, L.; Simón, P.J.; Lardín, C.;

Martínez, M.A.; Vicente, J.A.; González, R.; Llosá, C.; Arques, A.; Amat, A.M.

“Solar photocatalysis as a tertiary treatment to remove emerging pollutants from

wastewater treatment plant effluents”. Cat. Today 161 (2010) 235‐240.

[38] Muñoz, I.; López‐Doval, J.C.; Ricart, M.; Villagrasa, M.; Brix, R.; Geiszinger,

A.; Ginebreda, A.; Guasch, H.; López De Alda, M.J.; Romani, A.M.; Sabater, S.;

Barceló, D. “Bridging levels of pharmaceuticals in River water with biological

community structure in the Llobregat River basin (Northeast Spain)”. Environ.

Toxicol. Chem. 28 (2009) 1706‐2714.

[39] Pomati, F.; Castigloni, S.; Zuccato, E.; Fanelli, R.; Vigetti, D.; Rossetti, C.;

Calamari, D. “Effects of a complex mixture of therapeutic drugs at

environmental levels on human embryonic cells”. Environ. Sci. Technol. 40

(2006) 2442‐2447.

[40] Zegura, B.; Heath, E.; Cernosa, A.; Filipic, M. “Combination of in vitro

bioassays for the determination of cytotoxic and genotoxic potential of

wastewater, Surface water and drinking water samples”. Chemosphere 75 (2009)

1453‐1460.

[41] Muñoz, I.; Rodríguez, A.; Rosal, R.; Fernández‐Alba, A.R. “Life Cycle

Assessment of urban wastewater reuse with ozonation as tertiary treatment. A

focus on toxicity‐related impacts”. Sci. Total Environ. 407 (2009) 1245‐125.

[42] www.abc.es/hemeroteca/historico‐26‐10‐2005/abc/Sociedad/el‐rio‐llobregat‐

tiene‐contaminantes‐que‐da%C3%B1an‐el‐sistema‐reproductor‐de‐los‐

peces_611832583748.html (consultada en febrero de 2017)‐

[43] usatoday30.usatoday.com/news/nation/2008‐03‐10‐drugs‐tap‐water_N.htm

(consultada en febrero de 2017).

[44] elpais.com/diario/2009/10/28/sociedad/1256684408_850215.html (consultada

en febrero de 2017).

[45] www.abc.es/sociedad/20121126/abci‐trazas‐nicotina‐agua‐embotellada‐

201211261705.html (consultada en febrero de 2017).


51

[46] http://www.lavanguardia.com/vida/20150417/54429985933/delfines‐del‐

estrecho‐de‐gibraltar‐acumulan‐contaminantes‐nocivos‐para‐salud.html

(consultada en febrero de 2017).

[47] Directiva 91/271/CEE del Consejo, de 21 de mayo de 1991, sobre el

tratamiento de las aguas residuales urbanas.

[48] www.mapama.gob.es/es/agua/planes‐y‐estrategias/

PlanNacionalCalidadAguas_tcm7‐29339.pdf (consultada en febrero de 2017).

[49] www.iagua.es/noticias/espana/dbk/16/05/18/negocio‐depuracion‐aguas‐

alcanza‐1230‐millones‐2015 (consultada en febrero de 2017).

[50]www.observatoriosostenibilidad.com/documentos/SOS16_v23_PDF_final.pd

f (consultada en febrero de 2017).

[51] Prat, N.; Rieradevall, M.; Barata, C.; Munné, A. “The combined use of

metrics of biological quality and biomarkers to detect the effects of reclaimed

water on macroinvertebrate assemblages in the lower part of a polluted

Mediterranean river (Llobregat River, NE Spain)”. Ecol. Indic. 24 (2013) 167‐176.

[52] Real Decreto 1620/2007, de 7 de diciembre, por el que se establece el

régimen jurídico de la reutilización de las aguas depuradas.

[53] http://www.mapama.gob.es/es/calidad‐y‐evaluacion‐ambiental/

participacion‐publica/pp_2009p006.aspx (consultada en febrero de 2017).

[54] www.i‐ambiente.es/?q=blogs/aspectos‐considerar‐en‐la‐reutilizacion‐de‐

aguas‐depuradas‐para‐riego (consultada en febrero de 2017).

[55] Scott, T.M.; MccLaughlin, M.R.; Harwood, V.J.; Chivukula, V.; Levine, A.;

Gennaccaro, A.; Lukasick, J.; Farrah, S.R.; Rose, J.B. “Reduction of pathogens,

indicator bacteria, and alternative indicators by wastewater treatment and

reclamation processes”. Water Sci. Technol. Water Supply 3 (2003) 247‐252.

[56] Fatta‐kassinos, D.; Kalavrouziotis, I.K.; Koukoulakis, P.H.; Vasquez, M.I.

“The risks associated with wastewater reuse and xenobiotics in the

agroecological environment”. Sci. Total Environ. 409 (2011) 3555‐3563.

[57] Wu, C.; Spongberg, A.L.; Witter, J.D.; Fang, M.; Czajkowski, K.P. “Uptake

of pharmaceutical and personal care products by soybean plants from soils

applied with biosolids and irrigated with contaminated”. Water Environ. Sci.

Technol. 44 (2010) 6157‐6161.

[58] Wang, J.; Wang, S. “Removal of pharmaceuticals and personal care products

(PPCPs) from wastewater: A review”. J. Environ. Manage. 182 (2016) 620‐640.

[59] Glaze, W. H.; Kang, J.W.; Chapin, D.H. “The chemistry of water treatment

CAPÍTULO 1

52

processes involving ozone, hydrogen peroxide and ultraviolet radiation”. Ozone

Sci. Eng. 9 (1987) 335‐352.

[60] Andreozzi, R.; Caprio, V.; Insola, A.; Marotta, R. “Advanced oxidation

processes (AOP) for water purification and recovery”. Catal. Today 53 (1999) 51‐

59.

[61] Augugliaro, A.; Litter, M.; Palmisano, L: Soria, J. “The combination of

heterogeneous photocatalysis with chemical and physical operations: A tool for

improving the photoprocess performance”. J. Photochem. Photobiol. C

Photochem. Rev. 7 (2006) 127‐144.

[62] Domènech, X.; Jardim, W.F.; Litter, M.I. “Procesos avanzados de oxidación

para la eliminación de contaminantes”. CNEA‐CAC‐UAQ #95‐Q‐03‐05 (2005).

[63] Gogate P.R.; Pandit A.B. “A review of imperative technologies for

wastewater treatment I: oxidation technologies at ambient conditions”. Adv.

Env. Res. 8 (2004) 504‐551.

[64] Beltrán, F.J.; Aguinaco, A.; García‐Araya, J.F; Oropesa, A. “Ozone and

photocatalytic processes to remove the antibiotic sulfamethoxazole from water”.

Water Res. 42 (2008) 3799‐3808.

[65] Esplugas S.; Giménez J.; Contreras S.; Pascual E.; Rodríguez M.

“Comparison of different advanced oxidation processes for phenol

degradation”. Water Res. 36 (2002) 1034‐1042.

[66] J. Plotnikow. Zeitschrift fur Phisikalische chemie‐Stochiometrie und

verwandtschaftslehre 75 (1910) 337‐356.

[67] Bahnemann, D.; Hart, E.J. “Rate constants of the reaction of the

hydratedelectron and hydroxyl radical with ozone in aqueous solution”. J. Phys.

Chem. 86 (1982) 252‐255.

[68] Grätzel, M. “Heterogeneous Photochemical Electron Transfer”. CRC Press,

Boca Raton, FL. (1989).

[69] Malato, S.; Fernández‐Ibáñez, P.; Maldonado, M.I.; Blanco, J.; Gernjak, W.

“Decontamination and disinfection of water by solar photocatalysis: Recent

overview and trends”. Catal. Today 147 (2009) 1‐59.

[70] Salvador, P. “On the nature of photogenerated radical species active in the

oxidative degradation of dissolved pollutants with TiO2 aqueous suspensions: A

revision in the light of the electronic structure of adsorbed water”. J. Phys.

Chem. C 111 (2007) 17038‐17043.

[71] Henderson, M.A. “A surface science perspective on TiO2 photocatalysis”.

Surf. Sci. Rep. 66 (2011) 185‐297.


53

[72] Goldstein, S.; Behar, D.; Rabani, J. “Mechanism of visible light

photocatalytic oxidation of methanol in aerated aqueous suspensions of carbon‐

doped TiO2”. J. Phys. Chem. C. 112 (2008) 15134‐15139.

[73] Choina, J.; Fischer, G.U.; Flechsig, H.; Kosslick, V,A.; Tuan, N.D.; Tuyen,

N.A.; Tuyen, A.S. “Photocatalytic properties of Zr‐doped titania in the

degradation of the pharmaceutical ibuprofen”. J. Photochem. Photobiol. A 274

(2014) 108‐116

[74] Zhang, H.; Zhang, P.; Ji, Y.; Tian, J.; Du, Z. “Photocatalytic degradation of

four non‐steroidal anti‐inflammatory drugs in water under visible light by P25‐

TiO2/tetraethyl orthosilicate film and determination via ultra performance liquid

chromatography electrospray tandem mass spectrometry”. Chem. Eng. J. 262

(2015) 1108‐1115.

[75] Agustina, T.E.; Ang, H.M.; Vareek, V.K. “A review of synergistic effect of

photocatalysis and ozonation on wastewater treatment”. J. Photochem.

Photobiol. C Photochem. Rev. 6 (2005) 264‐273.

[76] Rivas, F.J.; Beltrán, F.J.; Encinas, A. “Removal of emergent contaminants:

Integration of ozone and photocatalysis”. J. Environ. Manag. 100 (2012) 10‐15.

[77] Rodríguez, E.M.; Fernández, G.; Álvarez, P.M.; Beltrán, F.J. TiO2 and Fe (III)

photocatalytic ozonation processes of a mixture of emergent contaminants of

water. Water Res. 46 (2012) 152‐166.

[78] Quiñones, D.H.; Rey, A.; Álvarez, P.M.; Beltrán, F.J.; Plucinski, P.K.

“Enhanced activity and reusability of TiO2 loaded magnetic activated carbon for

solar photocatalytic ozonation”. Appl. Catal. B Environ. 144 (2014) 96‐106.

[79] Xiao, J.; Xie, Y.; Cao, H. “Organic pollutants removal in wastewater by

heterogeneous photocatalytic ozonation (review)”. Chemosphere 121 (2015) 1‐

17.

[80] Mehrjouei, M.; Müller, S.; Möller, D. “A review on photocatalytic ozonation

used for the treatment of water and wastewater”. Chem. Eng. J. 263 (2015) 209‐

219.

[81] Buxton G.V.; Greenstock C.L.; Helman W.P.; Ross A.B. “Critical review of

rate constant for reactions or hydrated electrons, hydrogen atoms and hydroxyl

radicals (∙OH/∙O‐) in aqueous solutions”. J. Phys. Chem. Ref. Data 17 (1988) 513‐

886.

[82] Beltrán, F.J. “Ozone reaction kinetics for water and wastewater systems”.

Lewis Publishers, CRC Press. Boca Ratón, Florida (YS), 2004.

[83] Hoigné, J.; Bader, H. “The role of hydroxyl radical reactions in ozonation

CAPÍTULO 1

54

processes in aqueous solutions”. Water Res. 10 (1976) 377‐386.

[84] Degremont, s.a. “Water Treatment Handbook”. Sixth edition. 1991.

[85] Portolés, J.J.S. “El ozono atmosférico, ¿benefactor o malhechor?”. Caderno

Brasileiro de Ensino de Física 18 (2001) 360‐363.

[86] Staehelin, J., Buhler, R.E., and Hoigné, J., “Ozone decomposition in water

studied by pulse radiolysis. 2. Hydroxyl and hydrogen tetroxide (HO4) as chain

intermediates”. J. Phys. Chem. 88 (1984) 5999‐6004.

[87] Staehelin, J.; Hoigné, J. “Decomposition of ozone in water in the presence of

organic as promoters and inhibitors of radical chain reactions”. Environ. Sci.

Technol. 19 (1985) 1206‐1213.

[88] Hoigné, J.; Bader, H. “Rate constants of reactions of ozone with organic and

inorganic compounds in water‐ I: Non‐dissociating organic compounds”. Water

Res. 17 (1983) 173‐183.

[89] Peyton, G.R.; Glaze, W.H. “Destruction of pollutants in water with ozone in

combination with ultraviolet radiation. Photolysis of aqueous ozone”. Environ.

Sci. Technol. 22 (1988) 761‐767.

[90] Goldstein, S.; Aschengrau, D.; Diamant, Y.; Rabani J. “Photolysis of aqueous

H2O2: Quantum yield and applications for polychromatic UV actinometry in

photoreactors”. Environ. Sci. Technol. 41 (2007) 7486‐7490.

[91] Chu L.; Anastasio C. “Formation of hydroxyl radical from the photolysis of

hydrogen peroxide in ice”. J. Phys. Chem. A 109 (2005) 6264‐6271.

[92] Biedenkapp, D.; Hartshorn, L.G.; Bair E. “The O(1D) + H2O reaction” J.

Chem. Phys. Lett. 5 (1970) 379‐380.

[93] Wayne, R.P. “The photochemistry of ozone” Atmos. Environ. 21 (1987)

1683‐1694.

[94] Taniguchi, N.; Takahashi, K.; Matsumi, Y.J. “Photodissociation of O3 around

309 nm”. J. Phys. Chem. A 104 (2000) 8936‐8944.

[95] Hancock G.; Tyley, P.L. “The near‐UV photolysis of ozone: quantum yields

of O(1D) between 305 and 329 nm at temperatures from 227‐298 K, and the room

temperature quantum yield of O(3P2) between 303 and 310nm, measured by

resonance enhanced multiphoton ionization.” Phys. Chem. Chem. Phys. 3 (2001)

4984‐4990.

[96] Wine, P.H.; Ravishankara, A.R. “O3 photolysis at 248 nm and O(1D2)

quenching by H2O, CH4, H2, and N2O ‐ O(3Pj) yields” Chem. Phys. 69 (1982) 365‐

373.


55

[97] Kläning, U.K.; Sehested K.; Wolff T.J. “Ozone formation in laser flash‐

photolysis of oxoacids and oxoanions of chlorine and bromine” Chem. Soc.,

Faraday Trans. 1 80 (1984) 2969‐2979.

[98] Kondratiev, V.N. “Rate Constants of Gas‐Phase Reactions”. National Bureau

of Standards: Washington DC (1972).

[99] Hernández‐Alonso, M.D.; Coronado, J.M.; Maira, A.J.; Soria, J.; Loddo, V.;

Augugliaro, V. “Ozone enhanced activity of aqueous titanium dioxide

suspensions for photocatalytic oxidation of free cyanide ions”. Appl. Catal. B

Environ. 39 (2002) 257‐267.

[100] Marci, G.; García‐López, E.; Palmisano, L. “Mechanistic aspects of oxalic

acid oxidation by photocatalysis and ozonation”. J. Appl. Electrochem. 38 (2008)

1029‐1033.

[101] Rakowski, S.; Cherneva, D. “Kinetics and mechanism of the reaction of

ozone with aliphatic alcohols”. Int. J. Chem. Kinetics 22 (1990) 321‐329.

[102] Mvula, M.; von Sonntag, C. “Ozonolysis of phenols in aqueous solution”.

Org. Biomol. Chem. 1 (2003) 1749‐1756.

[103] Leitzke, A.; von Sonntag, C. “Ozonolysis of unsaturated acids in aqueous

solution: acrylic, methacrylic, maleic, fumaric and muconic acids”. Ozone Sci.

Eng. 31 (2009) 301‐308.

[104] Blanco J.; Malato S. “Solar detoxification”. Unesco Publishing, Renewable

energies series (2003).

[105] http://solarcellcentral.com/solar_page.html (consultada en febrero de

2017).

[106] www.psa.es (consultada en febrero de 2017).

[107] Robert, D.; Malato, S. “Solar photocatalysis: a clean process for water

detoxification”. Sci. Total. Environ. 291 (2002) 85‐97.

[108] Oller, I.; Malato, S.; Sánchez‐Pérez, J.A. “Combination of advanced

oxidation processes and biological treatments for wastewater decontamination‐

A review”. Sci. Total. Environ. 409 (2011) 4141‐4166.

[109] Prieto‐Rodríguez, L.; Miralles‐Cuevas, S.; Oller, I.; Agüera, A.; Puma, G.L.;

Malato, S. “Treatment of emerging contaminants in wastewater treatment plants

(WWTP) effluents by solar photocatalysis using TiO2 concentrations”. J. Hazard.

Mater. 211‐212 (2012) 131‐137.

[110] Velegraki, T.; Hapeshi, E.; Fatta‐Kassinos, D.; Poulios, I. “Solar‐induced

heterogeneous photocatalytic degradation of methyl‐paraben”. Appl. Catal. B

CAPÍTULO 1

56

Environ 178 (2015) 2‐11.

[111] Cordero‐García, A.; Turnes Palomino, G.; Hinojosa‐Reyes, L.; Guzmán‐

Mar, J.L.; Maya‐Teviño, L.; Hernández‐Ramírez, A. “Photocatalytic behaviour of

WO3/TiO2‐N for diclofenac degradation using simulated solar radiation as an

activation source”. Environ. Sci. Pollut. R. DOI: 10.1007/s11356‐016‐8157‐0.

[112] Rimoldi, L.; Meroni, D.; Falleta, E.; Pifferi, V.; Falciola, L.; Cappelletti, G.;

Ardizzone, S. “Emerging pollutant mixture mineralization by

TiO2 photocatalysts. The role of the water medium”. Photochem. Photobiol. Sci.

16 (2017) 60‐66.

[113] Araña, J.; Herrera, J.A.; Doña, J.M.; González, O.; Viera, A.; Pérez, J.;

Marrero, P.M.; Espino, V. “TiO2‐photocatalysis as a tertiary treatment of

naturally treated wastewater”. Catal. Today 76 (2002) 279‐289.

[114] Bayarri, B.; González, O.; Maldonado, M.I.; Giménez, J.; Esplugas,

S. “Comparative study of 2,4‐dichlorophenol by using different photocatalytic

techniques”. J. Sol. Energy Eng. 129 (2007) 60‐67.

[115] Oyama, T.; Yanigisawa, I.; Takeuchi, M; Koike, T.; Serpone, N.; Hidaka, H.

“Remediation of simulated aquatic sites contaminated with recalcitrant

substrates by TiO2/ozonation under natural sunlight”. Appl. Catal. B Environ. 91

(2009) 242‐246.

[116] Oyama, T.; Otsu, T.; Hidano, Y.; Koike, T.; Serpone, N.; Hidaka, H.

“Enhanced remediation of simulated wastewaters contaminated with 2‐

chlorophenol and other aquatic pollutants by TiO2‐photoassisted ozonation in a

sunlight driven pilot‐plant scale photoreactor”. Sol. Energy 85 (2011) 938‐944.

[117] Márquez, G.; Rodríguez, E.M.; Beltrán, F.J.; Álvarez, P.M. “Solar

photocatalytic ozonation of a mixture of pharmaceutical compounds in water”.

Chemosphere 113 (2014) 71‐78.

[118] Quiñones, D.H.; Álvarez, P.M.; Rey, A.; Contreras, S.; Beltrán, F.J.

“Application of solar photocatalytic ozonation for the degradation of emerging

contaminants in water in a pilor plant”. Chem. Eng. J. 260 (2015) 399‐410.

[119] Kabra, K.; Chaudhary, R.; Sawhney, R.L. “Treatment of hazardous organic

and inorganic compounds through aqueous‐phase photocatalysis: A review”.

Ind. Eng. Chem. Res. 43 (2004) 7683‐7696.

[120] Chong, M. N.; B. Jin, C. W. K. Chow, C. Saint. “Recent developments in

photocatalytic water treatment technology: A review”. Water Res. 44 (2010)

2997‐3027.

[121] Chen, J.; Qiu, F.; Xu, W.; Cao, S.; Zhu, H. “Recent progress in enhancing


57

photocatalytic efficiency of TiO2‐based materials”. Appl. Catal. A Gen. 495

(2015) 131‐140.

[122] Degussa. “Highly dispersed metallic oxides produced by AEROSIL®

process”. Technical Bulletin Pigments, 1990.

[123] Reyes, C.; Fernández, J.; Freer, J.; Mondaca, M.A.; Zaror, C.; Malato, S.;

Mansilla, H.D. “Degradation and inactivation of tetracycline by TiO2

photocatalysis”. J. Photochem. Photobiol. A Chem. 184 (2006) 141‐146.

[124] Xekoukoulotakis, N.P.; Drosou, C.; Brebou, C.; Chatzisymeon, E.; Hapeshi,

E.; Fatta‐Kassinos, D.; Mantzavinos, D. “Kinetics of UV‐A/TiO2 photocatalytic

degradation and mineralization of the antibiotic sulfamethoxazole in aqueous

matrices”. Catal.Today 161 (2011) 163‐168.

[125] Xiong, P.; Hu, J. “Decomposition of acetaminophen (Ace) using

TiO2/UVA/LED system”. Catal. Today 282 (2017) 48‐56.

[126] Beltrán, F.J.; Aguinaco, A.; Rey, A.; García‐Araya, J.F. “Kinetic studies on

black light photocatalytic ozonation of diclofenac and sulfamethoxazole in

water”. Ind. Eng. Chem. Res. 51 (2012) 4533‐4544.

[127] Rodríguez, E.M.; Márquez, G.; León, E.A.; Álvarez, P.M.; Amat, A.M.;

Beltrán, F.J. “Mechanism considerations for photocatalytic oxidation, ozonation

and photocatalytic ozonation of some pharmaceutical compounds in water”. J.

Environ. Manage. 127 (2013) 114‐124.

[128] Ren, H.; Koshy, P.; Chen, W.; Qi, S.; Sorrell, C.C. “Photocatalytic materials

and technologies for air purification”. J. Hazard. Mater. 325 (2017) 340‐366.

[129] Hernández‐Alonso M.D.; Fresno F.; Suarez S.; Coronado J.M.

“Development of alternative photocatalysts to TiO2: Challenges and

Opportunities”. Energ. Environ. Sci. 2 (2009) 1231‐1257.

[130] Zhao, Z.G; Miyauchi, M. “Nanoporous‐walled tungsten oxide nanotubes

as highly active visible‐light‐driven photocatalysts”. Angew. Chem. Int. Ed. 47

(2008) 7051‐7055.

[131] Kim, J.; Lee, C.W.; Choi, W. “Platinized WO3 as an environmental

photocatalyst that generates OH radicals under visible light” Environ. Sci.

Technol. 44 (2010) 6849‐6854.

[132] Sun, S.M.; Wang, W.Z.; Zeng, S.Z.; Shang, M.; Zhang, L.; “Preparation of

ordered mesoporous Ag/WO3 and its highly efficient degradation of

acetaldehyde under visible‐light irradiation” J. Hazard. Mater. 178 (2010) 427‐

433.

[133] Morales, W.; Cason, M.; Aina, O.; Tacconi, N.R.D.; Rajeshwar, K.

CAPÍTULO 1

58

“Combustion synthesis and characterization of nanocrystalline WO3” J. Am.

Chem. Soc. 130 (2008) 6318‐6319.

[134] Nishimoto S.; Mano T.; Kameshima Y.; Miyake M. “Photocatalytic water

treatment over WO3 under visible light irradiation combined with ozonation”

Chem. Phys. Lett. 500 (2010) 86‐89.

[135] Mano, T.; Nishimoto, S.; Kameshima, Y.; Miyake, M. “Investigation of

photocatalytic ozonation treatment of water over WO3 under visible light

irradiation”. J. Ceram. Soc. Jpn. 119 (211) 822‐827.

[136] Huang, W.; Gao, Y. “Morphology‐dependent surface chemistry and

catalysis of CeO2 nanocrystals”. Catal. Sci. Technol. 4 (2014) 3772‐3784.

[137] Aslam, M.; Qamar, M.T.; Tahir‐Soomro, M.; Ismail, I.M.I.; Salah, N.;

Almeelbi, T.; Gondal, M.A.; Hameed, A. “The effect of sunlight induced surface

defects on the photocatalytic activity of nanosized CeO2 for the degradation of

phenol and its derivatives”. Appl. Catal. B Environ. 180 (2016) 391‐402.

[138] Huang, X.; Beck, M.J. “Size‐dependent appearance of intrinsic

Oxq “Activated Oxygen” molecules on ceria nanoparticles”. Chem. Mater. 27

(2015) 5840‐5844.

[139] Hernández‐Alonso, M.D.; Hungría, A.B.; Martínez‐Arias, A.; Fernández‐

García, M.; Coronado, J.M.; Conesa, J.C.; Soria, J. “EPR study of the

photoassisted formation of radicals on CeO2 nanoparticles employed for toluene

photooxidation”. Appl. Catal. B Environ. 50 (2004) 167‐175.

[140] Deng, W.; Chenn, D.; Chenn, L. “Synthesis of monodisperse CeO2 hollow

spheres with enhanced photocatalytic activity”. Ceram. Int. 41 (2015) 11570‐

11575.

[141] Muduli, S.K.; Wang, S.; Chen, S.; Fan‐Ng, C.; Huan, C.H.A.; Sum, T.C.;

Soo, H.S. “Mesoporous cerium oxide nanospheres for the visible‐light driven

photocatalytic degradation of dyes”. Beilstein J. Nanotechnol. 5 (2014) 517‐523.

[142] Yu, Y.; Zhu, Y.; Meng, M. “Preparation, formation mechanism and

photocatalysis of ultrathin mesoporous single‐crystal‐like CeO2 nanosheets”.

Dalton Trans. 42 (2013) 12087‐12092.

[143] Karunakaran, C.; Dhanalakshmi, R. “Semiconductor‐catalyzed degradation

of phenols with sunlight”. Sol. Energ. Mat. Sol. C. 92 (2008) 1315‐1321.

[144] Da Silva, M.F.P.; Soeira, L.S.; Daghastanli, K.R.P.; Martins, T.S.; Cuccovia,

I.M.; Freire, R.S.; Isolani, P.C. “CeO2‐catalyzed ozonation of phenol: The role of

cerium citrate as precursor of CeO2”. J. Therm. Anal. Calorim. 102 (2010) 907‐

913.


59

[145] Faria, P.C.C.; Orfao, J.J.M.; Pereira, M.F.R. “A novel ceria‐activated carbon

composite for the catalytic ozonation of carboxylic acids”. Catal. Comm, 9 (2008)

2121‐2126.

[146] Gonçalves, A.G.; Orfao, J.J.M.; Pereira, M.F.R. “Ozonation of bezafibrate

over ceria and ceria supported on carbon materials”. Environ. Technol. 36 (2015)

776‐785.

[147] Keller, V.; Garin, F. “Photocatalytic behavior of a new composite ternary

system: WO3/SiC‐TiO2. Effect of the coupling of semiconductors and oxides in

photocatalytic oxidation of methylethylketone in the gas phase”. Catal.

Commun. 4 (2003) 377‐383.

[148] Dan, S.; Jingyu, W.; Yupan, T. “Constructing WO3/TiO2 composite

structure towards sufficient use of solar energy”. Chem. Commun. 47 (2011)

4231‐4233.

[149] Ismail, M.; Bousselmi, L.; Zahraa, O. “Photocatalytic behavior of WO3‐

loaded TiO2 systems in the oxidation of salicylic acid” J. Photochem. Photobiol.

A Chem. 222 (2011) 314‐322.

[150] Lana‐Villarreal T.; Monllor‐Satoca D.; Rodes A.; Gómez R. “Photocatalytic

behavior of suspended and supported semiconductor particles in aqueous

media: Fundamental aspects using catechol as model molecule” Catal. Today

129 (2007) 86‐95.

[151] Hernández‐Alonso, M.D.; García‐Rodríguez, S.; Sánchez, B.; Coronado,

J.M. “Revisiting the hydrothermal synthesis of titanate nanotubes: new insights

on the key factors affecting the morphology”. Nanoscale 3 (2011) 2233‐2240.

[152] Jung, J.H.; Lovayashi, H.; Van Bommel, K.J.C.; Shinkai, S.; Shimizu, T.

“Creation of novel helical ribbon and double‐layered nanotube TiO2 structures

using an organogel template”. Chem. Mater. 14 (2002) 1445‐1447.

[153] Kasuga, T.; Hiramatsu, M.; Hoson, A.; Sekino, T.: Niihara, K. “Formation of

titanium oxide nanotube”. Langmuir 14 (1998) 3160‐3163.

[154] Chen, Q.; Du, G.; Zhang, S.; Peng, L.M. ”The structure of trititanate

nanotubes”. Acta Crystallogr. Sect. B 58 (2002) 587‐593.

[155] Wang, W.; Varghese O.K.; Paulose M.; Grimes C.A. ”A study on the

growth and structure of titania nanotubes”, J. Mater. Res. 19 (2004) 417‐422.

[156] Capula, S.I. “Síntesis, caracterización y evaluación de la actividad catalítica

de nanopartículas Pt‐Ir sobre nanotubos de titania”. Tesis Doctoral, Escuela

Superior de Ingeniería Química e Industrias Extractivas, México D.F, México

(2007).

CAPÍTULO 1

60

[157] Pan, L.; Ji, M.; Lu, B.; Wang, X.; Zhao, L. ”Degradation of humic acid by

TiO2 nanotubes/UV/O3”. International conference on environmental science and

information application technology, Vol II (2009) Proceedings, 654‐657.

[158] Shan, A.Y.; Ghati, T.I.M.; Rashid, S.A. “Immobilisation of titanium dioxide

onto supporting materials in heterogeneous photocatalysis: A review”. Appl.

Catal. A Gen. 389 (2010) 1‐8.

[159] Belessi, V.; Lambropoulou, D.; Konstantinou, I.; Zboril, R.; Tucek, J.; Jancik,

D.; Albanis, T.; Petridis, D. “Structure and photocatalytic performance of

magnetically separable titania photocatalysts for the degradation of propachlor”.

Appl. Catal. B Environ. 87 (2009) 181‐189.

[160] Álvarez, P.M.; Jaramillo, J.; López‐Piñeiro, F.; Plucinski, P.K. “Preparation

and characterization of magnetic TiO2 nanoparticles and their utilization for the

degradation of emerging pollutants in water”. Appl. Catal. B Environ. 100 (2010)

338‐345.

[161] Wang, S.; Zhou, S. “Titania deposited on soft magnetic activated carbon as

a magnetically separable photocatalyst with enhanced activity”. Appl. Surf. Sci.

256 (2010) 6191‐6198.

[162] Ao, Y.H.; Xu, J.J.; Fu, D.G.; Shen, X.W.; Yuan, C.W. “Low temperature

preparation of anatase TiO2‐coated activated carbon”. Coll. Surf. A

Physicochem. Eng. Asp. 312 (2008) 125‐130.

[163] Li Puma, G.; Bono, A.; Krishnaiah, D.; Collin, J.C. “Preparation of titanium

dioxide photocatalyst loaded onto activated carbon support using chemical

vapor deposition: A review paper”. J. Hazard. Mater. 157 (2008) 209‐219.

[164] Wang, W.D.; Serp, P.; Kalck, P.; Faria, J.L. “Photocatalytic degradation of

phenol on MWNT and titania composite catalysts prepared by a modified sol‐

gel method”. Appl. Catal. B Environ. 56 (2005) 305‐312.

CAPÍTULO 2 Materiales y métodos experimentales

En este capítulo se describen las instalaciones experimentales empleadas y el procedimiento general de uso de las mismas. Asimismo, se detallan los métodos y equipos de análisis utilizados tanto para el seguimiento de la eficacia de los tratamientos de eliminación de contaminantes en agua como para la caracterización de los catalizadores sintetizados. Finalmente, se indican los productos químicos empleados a lo largo de esta investigación, clasificados en función de su aplicación.

Materiales y métodos experimentales

63

2.1. INSTALACIONES Y PROCEDIMIENTOS EXPERIMENTALES

En este apartado se describen de forma detallada las instalaciones

experimentales empleadas y el procedimiento general de uso de las mismas,

agrupadas en función de la finalidad que persiguen dentro de la presente

investigación.

2.1.1. Tratamientos de eliminación de contaminantes en agua

2.1.1.1. Ensayos realizados empleando lámparas de luz UVA (luz negra) como fuente de

radiación

En la Figura 2.1 se muestra un esquema de la instalación experimental en la

que se ha llevado a cabo el estudio de la degradación de compuestos orgánicos

en agua mediante diferentes procesos (fotólisis, oxidación fotocatalítica u

ozonización fotocatalítica), utilizando como fuente de radiación lámparas de luz

UVA cercana al visible (luz negra).

Figura 2.1. Instalación experimental en ensayos que emplean luz UVA como fuente de

radiación. 1: Botella de oxígeno; 2: Generador de ozono; 3: Rotámetro; 4: Entrada de gas;

5: Barra magnética; 6: Toma de muestra; 7: Termómetro; 8: Lámpara de luz negra y

soporte; 9: Agitador magnético; 10: Salida de gas; 11: Analizador de ozono; 12: Reactor

fotoquímico; 13: Caja negra.

1

2

3

4

5

6

7

8

9

10

11

12

13

CAPÍTULO 2

64

La instalación estaba formada por una caja de madera pintada de color negro

y de dimensiones 50 x 30 x 30 cm en cuyo interior se situaba el reactor. Se trataba

de un reactor cilíndrico de vidrio borosilicato 3.3, de 1 L de capacidad (18 cm de

altura y 10 cm de diámetro), situado sobre un agitador magnético (Agimatic‐E,

P. Selecta®), que inducía el movimiento de una barra magnética recubierta de

teflón situada en el interior del recipiente, garantizando una buena mezcla del

medio de reacción. En dos esquinas opuestas de la caja negra y a una distancia

de 5 cm del reactor, se situaban sendas lámparas de luz negra (LAMP15TBL

HQPOWERTM de 15 W de potencia nominal cada una de ellas, de dimensiones

45 cm de longitud y 2,5 cm de diámetro, fabricadas por Velleman®). El espectro

de emisión de este tipo de lámparas abarca, tal como se muestra en la Figura 2.2,

el intervalo de longitudes de onda comprendido entre 350 ‐ 400 nm, con máximo

de emisión a 365 nm. Este espectro se determinó con ayuda de un

espectrorradiómetro UV‐Vis compacto de fibra óptica que medía en el rango de

longitudes de onda 190 ‐ 850 nm (Black‐Comet modelo C de StellarNet).

200 300 400 500 600 700 800

(nm)

INT

EN

SID

AD

DE

LA

SE

ÑA

L (u

.a.)

Figura 2.2. Espectro de emisión de las lámparas de luz negra empleadas.

La instalación experimental disponía, además, de un sistema de alimentación

de oxígeno que permitía el burbujeo directo de la disolución a través de un


65

difusor, fijando el caudal de gas con ayuda de un rotámetro (Gilmont). La tapa

del reactor contaba con diferentes bocas adaptadas para la entrada‐salida de

gases, la medida de la temperatura y la toma de muestras. En el caso de los

ensayos en los que se empleó ozono, a la instalación experimental descrita se le

acoplaba un sistema de generación de ozono y un equipo de análisis y

destrucción de ozono en la corriente de gas de salida. El ozono era generado a

partir de oxígeno puro en un ozonizador Sander Labor‐Ozonisator (modelo

301.7), fijando la concentración requerida, según el ensayo. La concentración de

ozono en el gas de salida del ozonizador (antes del inicio de la reacción), y en el

gas de salida del reactor (una vez abierto el paso de gases al reactor), se

analizaba en continuo mediante un analizador Anseros Ozomat GM6000 Pro

dotado de un sistema de destrucción catalítica del ozono.

El procedimiento experimental se iniciaba encendiendo las lámparas y

esperando un tiempo de 30 minutos (tiempo suficiente para que las lámparas

alcanzaran el régimen estacionario de emisión), antes de introducir el reactor en

la instalación. Mientras tanto, al reactor se adicionaba la disolución acuosa de

compuesto a degradar, se ajustaba o tamponaba el medio al valor requerido de

pH mediante la adición de ácidos o bases y se ponía en marcha el sistema de

agitación. En los ensayos en los que se empleaba catalizador se tomaba una

muestra inicial del reactor antes de la adición del mismo, se añadía la cantidad

requerida de catalizador y se mantenía la mezcla en agitación y en la oscuridad

durante 30 minutos, con la finalidad de establecer el equilibrio de adsorción del

compuesto a degradar sobre el catalizador. Pasado ese tiempo, se tomaba otra

muestra (que era considerada como la muestra a tiempo cero), y se filtraba

empleando filtros de jeringa de 0,45 μm (Macherey‐Nagel PET). Finalmente, una

vez alcanzado el régimen estacionario de emisión de las lámparas se introducía

el reactor en la instalación y se iniciaba el burbujeo de gas a través de un difusor

sumergido en la disolución (ver Fig. 2.1). A partir de ese instante, se ponía en

marcha el cronómetro y a distintos intervalos de tiempo se tomaban muestras, se

filtraban y se recogían en distintos viales para su análisis posterior. Si se

empleaba ozono en el ensayo, la forma de proceder era la misma pero poniendo

CAPÍTULO 2

66

especial cuidado en evitar fugas de gas, para lo cual debían sellarse con silicona

las uniones esmeriladas de la tapa del reactor. Además, en este tipo de ensayos

las muestras que se extraían del reactor se dividían en dos fracciones, una

dedicada al análisis de la concentración de ozono disuelto y otra al análisis del

resto de especies involucradas en el proceso. Esta última fracción era burbujeada

inmediatamente con aire para eliminar, por arrastre, el posible ozono disuelto

que hubiera en la muestra, evitando con ello que el ozono siguiera reaccionando.

La duración de los experimentos fue de 60 min.

2.1.1.2. Ensayos realizados empleando luz solar artificial como fuente de radiación.

El estudio a escala de laboratorio de la degradación de contaminantes en

agua mediante distintos procesos empleando luz solar artificial, se llevó a cabo

en una instalación experimental como la que se muestra de forma esquemática

en la Figura 2.3.

Figura 2.3. Instalación experimental en ensayos que emplean luz solar simulada como

fuente de radiación. 1: Botella de oxígeno; 2: Generador de ozono; 3: Rotámetro; 4:

Entrada de gas; 5: Barra magnética; 6: Toma de muestra; 7: Lámpara de Xe; 8: Agitador

magnético; 9: Salida de gas; 10: Analizador de ozono; 11: Reactor fotoquímico; 12:

Simulador solar.

La instalación constaba de un simulador solar Suntest CPS+ (Atlas) de

dimensiones 90 x 35 x 35 cm que disponía de una cámara dotada de una

lámpara xenón refrigerada por aire de 1500 W de potencia, siendo el área total

de exposición dentro de la cámara de 560 cm2. La presencia de filtros de cuarzo y

vidrio restringen el espectro de emisión a longitudes de onda superiores a 300

nm. Dicho espectro, medido con ayuda del espectrorradiómetro UV‐Vis


67

indicado en el apartado anterior, se muestra en la Figura 2.4 junto con el

espectro solar. En todos los casos, la intensidad de radiación se fijaba en 550

W m‐2 (valor promedio de la radiación solar global en un día soleado), oscilando

la temperatura en el interior de la cámara entre 20 y 40 °C a lo largo de los

experimentos. En la cámara se introducía el reactor, consistente en un reactor

esférico de vidrio borosilicato 3.3 de 0,5 L de capacidad (10 cm de diámetro). La

distancia de la lámpara a la superficie del reactor era de 15 cm

aproximadamente.

200 300 400 500 600 700 800

Simulador Solar RadiaciÓn Solar

(nm)

INT

EN

SID

AD

DE

LA

SE

ÑA

L (u

.a.

)

Figura 2.4. Espectro de emisión característico del sistema lámpara de Xe + filtro de cuarzo

+ filtro de vidrio del simulador solar Suntest CPS+, junto con el espectro solar.

La instalación disponía de un sistema de alimentación de oxígeno con el que

se burbujeaba directamente la disolución acuosa a tratar a través de un difusor,

fijándose el caudal de gas con ayuda de un rotámetro (Gilmont). El reactor

contaba con diversas bocas para la entrada‐salida de gases y la toma de

muestras. En caso de emplear ozono se procedía de la forma descrita en el

apartado anterior, esto es, generándolo a partir de oxígeno puro en un

ozonizador Sander Labor‐Ozonisator (modelo 301.7) y analizando en continuo

su concentración, tanto en el gas de entrada, con ayuda de un analizador GM‐

CAPÍTULO 2

68

600‐OEM de Anseros, como en el gas de salida empleando un equipo Anseros

Ozomat GM6000 Pro dotado de un sistema de destrucción catalítica del ozono.

El procedimiento seguido fue similar al descrito para los ensayos con luz

UVA: el reactor, cargado con la disolución acuosa de compuesto/s a degradar, se

colocaba en el interior de la cámara del simulador solar soportado sobre un

agitador magnético. En caso de emplear catalizador, se tomaba una muestra

inicial de la disolución y a continuación se añadía la cantidad requerida de

catalizador, dejando la mezcla en agitación y en la oscuridad durante 30 minutos

al objeto de establecer el equilibrio de adsorción. Posteriormente, se tomaba una

nueva muestra (considerada tiempo cero), se filtraba y se recogía en un vial para

su análisis posterior. Se encendía la lámpara del simulador solar (se comprobó

previamente que no era necesario esperar un tiempo para alcanzar el régimen

estacionario de emisión), y se abría el paso del gas (oxígeno o mezcla

oxígeno/ozono), al reactor. En ese instante se ponía en marcha el cronómetro y a

distintos tiempos se tomaban muestras y se procedía de la forma descrita en el

apartado anterior. La duración de los experimentos varió en función del objetivo

perseguido con cada tipo de tratamiento.

2.1.2. Preparación de los catalizadores empleados

A continuación se describen los equipos utilizados en la preparación de

distintos catalizadores indicándose, de forma general, la función de cada uno de

ellos en el procedimiento de síntesis.

2.1.2.1. Agitador

En varias de las etapas seguidas para llevar a cabo la síntesis de catalizadores

(hidrólisis sol‐gel, homogeneización, lavado de los materiales sintetizados, etc.),

se empleó un agitador magnético Agimatic‐E de Selecta®. Este inducía el

movimiento de una barra magnética recubierta de teflón situada en el interior

del vaso de precipitados de vidrio que contenía la mezcla de la síntesis.


69

2.1.2.2. Autoclave

En el caso de los catalizadores sintetizados por vía hidrotermal, el

tratamiento se realizó en una autoclave de acero inoxidable de Parr Instrument

Company, que contenía un vaso de teflón de 125 mL. En el vaso se introducía la

mezcla de reacción evitando que el volumen de la misma superara el 75 % del

volumen total del vaso. Las autoclaves constituyen sistemas sellados en los que

el recinto de reacción se encuentra sometido a condiciones de presión y

temperatura elevadas, como las características de los procesos hidrotermales.

2.1.2.3. Centrífuga

Tras cada etapa de lavado de los materiales sintetizados, las partículas de

catalizador fueron separadas de los disolventes utilizados (ácido clorhídrico,

agua y etanol), con ayuda de una centrífuga Alresa de 200 W de potencia y una

frecuencia de rotación de 50 Hz.

2.1.2.4. Estufa

El secado final del material sintetizado se llevó a cabo en una estufa Conterm

150 L de Selecta®, dotada de un limitador fijo de sobrecalentamiento y un

termostato de seguridad regulable. La estufa permite trabajar con temperaturas

comprendidas entre 45 °C y 250 °C, distribuyendo el calor en su interior por

convección natural. A 100 °C la homogeneidad de esta estufa es de ± 1 °C y la

estabilidad térmica de ± 0,5 °C.

2.1.2.5. Horno mufla

Las partículas de catalizador sintetizadas, así como los propios materiales

precursores en el caso de obtención de catalizadores por descomposición

térmica, se introducían en cápsulas de porcelana y se calcinaban en un horno

mufla Raypa modelo HM, en el que la calefacción se produce mediante placas

termo‐cerámicas con resistencias eléctricas de níquel y cromo. La temperatura se

regula mediante un microprocesador, pudiendo alcanzar hasta 1150 °C, con una

CAPÍTULO 2

70

estabilidad térmica de ± 1 °C y una homogeneidad térmica de ± 5 °C. Asimismo,

los tratamientos hidrotermales de obtención de catalizadores también se

llevaron a cabo en el horno mufla, introduciendo la autoclave en su interior a la

temperatura requerida.

2.2. EQUIPOS Y MÉTODOS DE ANÁLISIS

Dentro del desarrollo experimental de la presente investigación, se pusieron

a punto una serie de métodos de análisis y se emplearon diversos equipos. Estos

se detallan a continuación, clasificados según el objetivo que persiguen.

2.2.1. Seguimiento de la eficacia de los tratamientos de eliminación de

contaminantes en agua

En la Tabla 2.1 se resumen los métodos analíticos y los equipos utilizados

para llevar a cabo la determinación de la eficacia de los distintos tratamientos

ensayados.

Tabla 2.1. Métodos de análisis y equipos utilizados para el seguimiento de la eficacia de

los tratamientos de eliminación de compuestos orgánicos en agua.

Parámetro Método de análisis Equipo

Concentración de

contaminantes modelo

en agua

Cromatografía líquida de alta

resolución (HPLC) ‐ Detector de haz de

diodos (DAD)

Cromatógrafo HPLC‐

DAD Elite LaChrom

Identificación de

productos intermedios

de degradación

Cromatografía líquida de alta

resolución (HPLC) ‐ espectrómetro de

masas (MS) con analizador Q‐TOF

Cromatógrafo HPLC‐

MS‐QTOF Agilent

Concentración de

ácidos carboxílicos e

iones

Cromatografía iónica

Cromatógrafo iónico

Metrohm 881 Compact

Pro

Concentración de

carbono orgánico total

(COT)

Oxidación + espectroscopía infrarroja

Analizador de COT

Shimadzu, TOC‐V

CSH/CSN


71

Tabla 2.1 (Continuación). Métodos de análisis y equipos utilizados para el seguimiento

de la eficacia de los tratamientos de eliminación de compuestos orgánicos en agua.


Concentración de

ozono en disolución

acuosa

Oxidación + Espectrofotometría [1]

Espectrofotómetro

ThermoSpectronic,

Heλios‐α. Cubeta de

cuarzo de 1 cm.

Concentración de

ozono en fase gas Espectrofotometría

Analizador de ozono

Anseros Ozomat

GM6000 Pro

Concentración de

peróxido de hidrógeno

Oxidación + Complejación +

Espectrofotometría [2]

Espectrofotómetro

ThermoSpectronic,


cuarzo de 1 cm.

Concentración de

hierro total

Reducción + Complejación +

Espectrofotometría [3]

Espectrofotómetro

ThermoSpectronic,


cuarzo de 1 cm.

Concentración de

hierro (II) Complejación + Espectrofotometría [4]

Espectrofotómetro

ThermoSpectronic,


cuarzo de 1 cm.

Intensidad de

radiación

Actinometría con ferrioxalato

(Actinómetro de Parker) [5]

Espectrofotómetro

ThermoSpectronic,


cuarzo de 1 cm.

Concentración de

fomaldehído Reacción + Espectrofotometría [6]

Espectrofotómetro

ThermoSpectronic,


cuarzo de 1 cm.

Demanda química de

oxígeno (DQO) Oxidación + Espectrofotometría [7]

Digestor LT200.

Espectrofotómetro

DR2800. Cubetas test

LCK 414 (Hach Lange)

Biodegradabilidad.

Demanda biológica de

oxígeno (DBO)

Oxidación + absorción de oxígeno +

medida manométrica Biómetros Oxitop®

Aromaticidad.

Absorción de

radiación UV de 254

nm

Espectrofotometría

Espectrofotómetro

ThermoSpectronic,


cuarzo de 1 cm.

CAPÍTULO 2

72

Tabla 2.1 (Continuación). Métodos de análisis y equipos utilizados para el seguimiento

de la eficacia de los tratamientos de eliminación de compuestos orgánicos en agua.


Fosfatos

Complejación + Reducción +

Espectrofotometría (Merck

Spectroquant kit)

Espectrofotómetro

ThermoSpectronic,


cuarzo de 1 cm.

Turbidez Nefelometría Turbidímetro Hanna

HI 93414

pH Potenciometría

pH‐metro Crison GLP

21+ con electrodo

combinado

Conductividad Conductimetría Conductímetro Crison

524

Temperatura Termometría Termómetro de

contacto

2.2.1.1. Determinación de la concentración de contaminantes modelo en agua

La concentración de contaminantes modelo en agua se determinó mediante

cromatografía líquida de alta resolución (HPLC), empleando un equipo Elite

LaChrom dotado de desgasificador, bomba cuaternaria Hitachi L‐2130, sistema

de inyección automático Hitachi L‐2200 y detector de haz de diodos Hitachi

L‐2455. El equipo consta, además, de un software (EZ Chrom Elite) para el

registro y tratamiento de datos.

La fase estacionaria utilizada fue una columna de fase reversa Phenomenex

Gemini C18 de tamaño de partícula 5 m y tamaño de poro 100 Å, de 15 x 0,3

cm. Los compuestos eran separados por elución con agua acidificada al 0,1 %

v/v con ácido fosfórico o ácido fórmico (disolvente A), y acetonitrilo (disolvente

B). En la Tabla 2.2 se resumen las condiciones de cada método empleado para el

análisis de la concentración de los contaminantes. Previamente, a partir del

análisis cromatográfico de muestras patrón de distinta concentración, se obtuvo

la recta de calibrado correspondiente a cada compuesto. Los límites de detección


73

se muestran también en la Tabla 2.2.

Tabla 2.2. Condiciones de los métodos de HPLC empleados.

Método Condiciones Compuestos tR

(min) (nm)

LD

(g L‐1)

1

Elución isocrática a 0,7 mL min‐1,

40 % disolvente A (ácido

fosfórico). VINY = 50 L. IBP 6,1 220 161,2

2

Elución con gradiente a 0,2

mL min‐1, 90 ‐ 0 % disolvente A

(ácido fórmico) en 40 min y 20

min de re‐equilibrado. VINY = 50

L.

AAP

MTP

CAF

HCT

ANT

SFM

CAR

KET

DCF

10,5

14,8

15,28

16,9

18,2

23,1

27,1

29,4

36,8

244

225

273

271

240

260

285

320

275

7,1

5,4

33,6

11,5

7,0

4,4

10,3

6,6

5,3

3

Elución con gradiente a 0,5

mL min‐1, 95 ‐ 40 % disolvente A

(ácido fosfórico) en 25 min y 10

min de re‐equilibrado. VINY = 99

L.

CAF

MTP

IBP

13,5

15,1

27,5

220

220

220

12

8,4

11,4

4

Elución isocrática a 0,6 mL min‐1,

70 % disolvente A (ácido

fórmico). VINY = 50 L.

DEET 10,4 220 176,1

tR: tiempo de retención; : longitud de onda de detección; LD: límite de detección; VINY:

volumen de inyección

2.2.1.2. Identificación de productos intermedios de degradación

La identificación de productos intermedios de degradación de alguno de los

contaminantes seleccionados en agua fue realizada con ayuda del Servicio de

Análisis Elemental y Molecular (SAEM) de los Servicios de Apoyo a la

Investigación de la Universidad de Extremadura (SAIUEx), el cual dispone de

un equipo de cromatografía líquida con detector de masas tipo Q‐TOF de

Agilent que permite detectar y cuantificar multitud de compuestos orgánicos,

incluso en concentraciones del orden ng L‐1, sin necesidad de preconcentrar la

muestra. El equipo consta de desgasificador (1260‐‐Degasser), bomba binaria

CAPÍTULO 2

74

(1260 bin Pump), automuestreador (HiP‐ALS SL+), termostatizador (TCC SL) y

detector de masas tipo Q‐TOF (6520 Accurate‐Mass Q‐TOF LC/MS).

En la separación mediante cromatografía líquida la fase estacionaria fue la

columna Zorbax SB C18 de Agilent (3,5 m, 4,6 x 150 mm), empleándose como

fase móvil una mezcla de acetonitrilo (disolvente A) y agua acidificada con

ácido fórmico 25 mM (disolvente B). Para la separación de los intermedios se

empleó un gradiente de 90 % de B en el minuto 0, a 100 % de A en el minuto 40,

con un caudal de 0,2 mL min‐1. El volumen de inyección fue de 10 L.

Por su parte, el detector Q‐TOF operaba en modo positivo bajo los siguientes

parámetros: voltaje del capilar, 3000 V; presión del nebulizador, 30 psig; caudal

de gas, 10 L min‐1; temperatura del gas, 350 °C; voltaje del fragmentador, 175 V;

voltaje de skimmer, 65 V; Octopolo RF, 750 V. El rango de adquisición de masas

de los compuestos fue 30‐1000 m/z. Los datos fueron adquiridos y procesados

con el programa Agilent MassHunter WorkStation (versión B.04.00). La

adquisición precisa de las masas se realizaba con una fuente de doble

calibración, introduciendo junto al flujo del cromatógrafo una solución estándar

de iones de calibración de purina (C5H4N4, m/z 121,0509) y HP‐921

(C18H18O6N3P3F24, m/z 922,0098).

2.2.1.3. Determinación de la concentración de ácidos carboxílicos de bajo peso molecular

y aniones inorgánicos en disolución

La determinación de la concentración de distintos ácidos orgánicos de cadena

corta (en forma de sus aniones correspondientes) y de algunos aniones

inorgánicos se llevó a cabo mediante cromatografía iónica. Para ello, se utilizó

un cromatógrafo iónico con supresor químico, modelo 881 compact IC Pro

(Metrohm), equipado con una columna de aniones modelo Metrosep A Supp 7

de 4 x 150 mm como fase estacionaria (5 m de tamaño de partícula,

termostatizada a 45 °C). Como fase móvil se empleó una disolución de


75

carbonato sódico, aplicándose un gradiente en el que la concentración variaba

de 0,6 a 14,6 mM con un caudal constante de 0,7 mL min‐1. La supresión química

se realizó mediante una disolución 250 mM de ácido sulfúrico. La detección de

los aniones se efectuó mediante medidas de conductividad, determinándose la

concentración de los mismos con ayuda del software MagIC Net 2.4 S, a partir

del calibrado previo realizado con disoluciones patrón que contenían entre 0 y

10 mg L‐1 de cada anión. Los tiempos de retención y los límites de detección de

los aniones fueron los siguientes: acetato, tR = 7,3 min y LD = 46 g L‐1; formiato,

tR = 9,6 min y LD = 83 g L‐1; piruvato, tR = 10,3 min y LD = 275 g L‐1; cloruro, tR

= 13,1 min y LD = 143 g L‐1; nitrato, tR = 25,8 min y LD = 272 g L‐1; fosfato, tR =

34,5 min y LD = 350 g L‐1; sulfato, tR = 35,7 min y LD = 268 g L‐1; y oxalato, tR =

39,0 min y LD = 75 g L‐1.

2.2.1.4. Determinación de la concentración de carbono orgánico total en disolución

(COT)

El análisis del contenido en COT se realizó en un analizador TOC‐V

CSH/CSN de Shimadzu, dotado de un sistema de muestreo e inyección

automático. El método analítico se basa en la formación de dióxido de carbono

por combustión catalítica de la muestra a 680 °C, empleando un catalizador de

Pt/Al2O3. El dióxido de carbono generado se analiza posteriormente mediante

un detector de infrarrojos, determinándose así el carbono total (CT) de la

muestra. Por otro lado, para la determinación de carbono inorgánico (CI), se

inyecta una nueva alícuota en un depósito donde es acidificada con ácido

ortofosfórico al 25 %. En estas condiciones, todos los carbonatos/bicarbonatos

disueltos se transforman en dióxido de carbono, cuya concentración es de nuevo

cuantificada en el detector de infrarrojo. El contenido en COT se obtiene por

diferencia (COT = CT ‐ CI).

Previamente, se prepararon disoluciones patrón de ftalato ácido de potasio

con un contenido en carbono comprendido entre 0 y 100 mg L‐1, y disoluciones

de carbonato/bicarbonato de sodio de 0 a 40 mg L‐1 de carbono, obteniéndose a

CAPÍTULO 2

76

partir del análisis de las mismas las rectas de calibrado del CT y el CI,

respectivamente. Los límites de detección fueron 114 g L‐1 para CT y 247 g L‐1

para CI.

2.2.1.5. Determinación de la concentración de ozono en disolución acuosa

Para la determinación de la concentración de ozono disuelto en las muestras

de reacción se utilizó el método colorimétrico propuesto por Bader y Hoigné en

1981 [1], basado en la decoloración que sufre el 5,5’,7‐indigotrisulfonato de

potasio (índigo carmín) cuando reacciona con el ozono.

Para ello, a un vial que contenía 4 mL de disolución 10‐4 M de índigo carmín

de pH 2 se añadían 2 mL de la muestra analizar (fracción de muestra no

burbujeada con aire, tal como se indicó en el apartado 2.2.1.1). Se agitaba la

mezcla, se filtraba en el caso de los ensayos con catalizador, se trasvasaba a una

cubeta espectrofotométrica de 1 cm de camino óptico y se determinaba la

absorbancia de la mezcla a 600 nm. El blanco se preparaba en idénticas

condiciones pero sustituyendo la muestra por 2 mL de agua ultrapura.

De acuerdo con la Ley de Beer, en ausencia de cualquier otra especie que

absorba radiación a 600 nm se tendrá que:

3dis

0 m TO

600nm m

Abs Abs VC

c V

(2.1)

donde CO3dis es la concentración (M) de ozono disuelto en la muestra analizada;

Abs0 la absorbancia del blanco; Absm la absorbancia de la muestra; 600nm la

absortividad molar del índigo carmín a 600 nm (2x104 M‐1 cm‐1); c el camino

óptico (1 cm); VT el volumen total de mezcla (6 mL) y Vm el volumen de muestra

(2 mL). El límite de detección del método es de 10‐6 M de O3.

La disolución de índigo carmín de concentración 10‐3 M se preparaba por

pesada disolviéndolo en agua ultrapura y en medio ácido fosfórico 2x10‐3 M. A

partir de esta disolución madre se preparaba por dilución 1:20 en agua ultrapura


77

otra de concentración 10‐4 M en medio ácido fosfórico/fosfato monobásico de

potasio de pH 2 (28/35 m/m). Esta disolución se conservaba en la oscuridad y a

una temperatura comprendida entre 2 y 8 °C, siendo estable durante

aproximadamente 4 meses.

2.2.1.6. Determinación de la concentración de ozono en fase gas

Tal como se expuso en el apartado 2.1.1 al describir las instalaciones

experimentales, para determinar la concentración de ozono en fase gas en la

corriente oxígeno/ozono a la entrada y salida del reactor se emplearon los

equipos GM‐600‐OEM y GM‐6000‐PRO de Anseros, respectivamente. La

medida, realizada en continuo durante todo el tiempo de duración del ensayo,

se basa en el análisis espectrofotométrico a 254 nm, longitud de onda a la cual el

oxígeno no absorbe y el ozono presenta un máximo de absorción. Con ayuda de

una lámpara de mercurio de baja presión situada en el interior de una celda

óptica, una fracción de la corriente gaseosa es irradiada a 254 nm. Por aplicación

de la Ley de Beer, la medida de la atenuación de la señal permite determinar la

concentración de ozono en el gas. Los equipos empleados están diseñados para

trabajar a presiones superiores a 2 bar en el rango de concentraciones de O3

comprendido entre 0 y 50 mg L‐1.

2.2.1.7. Determinación de la concentración de peróxido de hidrógeno en disolución

El método empleado ha sido el propuesto por Masschelein et al. en 1977 [2],

basado en la oxidación de Co(II) a Co(III) por parte del peróxido de hidrógeno y

la posterior determinación espectrofotométrica del complejo de bicarbonato‐Co

(III) formado a 260 nm, longitud de onda a la cual el complejo presenta una

elevada absortividad molar (2,664x104 M‐1 cm‐1). Este método es válido para

concentraciones de peróxido de hidrógeno comprendidas entre 3x10‐7 y 5x10‐5

M.

Para llevar a cabo el análisis, en un vial de vidrio que contenía 0,5 mL de

disolución de 4 g L‐1 de Co(II) (preparada disolviendo 1,61 g de cloruro de

CAPÍTULO 2

78

cobalto (II) hexahidrato en 100 mL de agua ultrapura), 10 mL de disolución

saturada de NaHCO3 y 0,5 mL de disolución de 10 g L‐1 de calgón (preparada

disolviendo 1 g de hexametafosfato de sodio en 100 mL de agua ultrapura), se

añadía 1 mL de la muestra analizar. Al objeto de tener en cuenta la posible

contribución de la matriz a la absorbancia a 260 nm, se preparaba otra serie de

viales en los que se sustituían los 0,5 mL de disolución de Co(II) por agua

ultrapura. Los blancos, uno para cada tipo de vial (con y sin cobalto), se

preparaban siguiendo el mismo procedimiento pero reemplazando la muestra

por agua ultrapura. Finalmente, se agitaban las mezclas y se medía la

absorbancia a 260 nm pasados 10‐15 min.

Teniendo en cuenta la Ley de Beer, la concentración molar de peróxido de

hidrógeno en la muestra, CH2O2, puede determinarse mediante la siguiente

expresión:

2 2

2 2

Co(II) H O Co(II) H Om 0 T

H O260nm m

Abs Abs Abs Abs VC

c V

(2.2)

donde (AbsCo(II))0 y (AbsH2O)0 son las absorbancias del blanco en presencia y

ausencia de Co(II), respectivamente; (AbsCo(II))m y (AbsH2O)m las absorbancias de

la muestra en presencia y ausencia de Co(II), respectivamente; 260nm la

absortividad molar del complejo formado a 260 nm; c el camino óptico (1 cm);

VT el volumen total (12 mL) y Vm el volumen de muestra (1 mL).

2.2.1.8. Determinación de la concentración de hierro total en disolución

Para determinar la concentración de hierro total se utilizó un kit comercial

(Spectroquant, Merck). El método se basa en la reducción del Fe(III) existente a

Fe(II) y posterior formación de un complejo estable de color violeta entre el

Fe(II) y la ferrozina presente en el reactivo comercial [3]. Dicho complejo se

analiza mediante espectrofotometría de absorción a 565 nm.

Para el análisis, 3 gotas de reactivo comercial se añadían a 5 mL de muestra o


79

de muestra diluida, según fuera necesario en función de la concentración de

hierro total en la muestra. Pasados quince minutos se analizaba la absorbancia

de las mezclas a 565 nm.

De acuerdo con la Ley de Beer, la concentración molar de especies de hierro

en disolución en la mezcla, CFeT, viene dada por la ecuación (2.3):

T

m 0 TFe

565nm m

Abs Abs VC

c V

(2.3)

donde Abs0 la absorbancia del blanco (5 mL de agua y 3 gotas de reactivo); Absm

la absorbancia de la muestra; 565nm la absortividad molar del complejo a 565 nm

(27044 M‐1 cm‐1, [8]); c el camino óptico (1 cm); VT el volumen total de la muestra

final (5 mL); y Vm el volumen de muestra no diluida.

2.2.1.9. Determinación de la concentración de hierro (II) en disolución

La concentración de Fe(II) en disolución se determinó mediante el método

propuesto por Zuo en 1995 [4], basado en la medida de la absorbancia a 510 nm

del complejo coloreado formado entre el ion ferroso y la o‐fenantrolina en medio

ácido acético/acetato de pH comprendido entre 3 y 4. A 510 nm el coeficiente de

extinción molar de este complejo es 11023 M‐1 cm‐1 [9].

Para llevar a cabo el análisis, en primer lugar se preparaban los reactivos

necesarios: tampón de ácido acético/acetato 0,1 M de pH entre 3 y 4 (preparado

tomando 572 L de ácido acético glacial y enrasando a 100 mL con agua

ultrapura); disolución de 1,10‐fenantrolina al 0,2 % en peso (preparada

disolviendo 0,2 g de la sal en 100 mL de agua ultrapura, manteniéndose la

disolución refrigerada y preservada de la luz); y disolución de fluoruro de

amonio 2 M (preparada disolviendo 7,4 g de la sal en 100 mL de agua ultrapura).

Una vez preparados, en un vial que contenía 1,5 mL de tampón acético/acetato y

1 mL de o‐fenantrolina, se añadían 5 mL de la muestra a analizar (diluida si fue

necesario). Se agitaba la muestra y se adicionaba 1 mL de fluoruro amónico.

Transcurridos al menos 20 minutos se medía la absorbancia de la mezcla a 510

CAPÍTULO 2

80

nm. El blanco se preparaba en idénticas condiciones pero sustituyendo la

muestra por 5 mL de agua ultrapura.

De acuerdo con la Ley de Beer, en ausencia de cualquier otra especie que

absorba radiación a 510 nm la concentración molar de Fe(II) en la muestra

analizada, CFe(II) , vendrá dada por:

m 0 T

Fe(II)510nm m

Abs Abs VC

c V

(2.4)

donde Abs0 es la absorbancia del blanco; Absm la absorbancia de la muestra;

510nm la absortividad molar del complejo o‐fenantrolina‐Fe(II) a 510 nm; c el

camino óptico (1 cm); VT el volumen total (8,5 mL) y Vm el volumen de muestra

no diluída (5 mL).

2.2.1.10. Determinación de la intensidad de radiación

En la práctica, la actinometría se considera la herramienta que permite

determinar la cantidad total de la luz absorbida en un proceso fotoquímico y,

con ello, la determinación de la intensidad de la luz que las lámparas empleadas

ofrecen y la calibración de los equipos de irradiación. Así, la intensidad de

radiación del sistema dotado con lámparas UVA mostrado en la Fig. 2.1 se

determinó con ayuda del actinómetro de Parker [5], basado en la reducción

fotoquímica del Fe(III) presente en el complejo ferrioxalato [Fe(C2O4)3]3‐ a Fe(II).

La fotorreducción tiene lugar con un rendimiento cuántico de 1 ‐ 1,2

mol einstein‐1 en el intervalo de longitudes de onda comprendidas entre 250 ‐

450 nm [10]. El rendimiento cuántico de un determinado proceso fotoquímico es

la relación entre la cantidad de reactante consumido (o de producto formado) en

la fotorreacción, y la cantidad total de luz absorbida por el sistema a la longitud

de onda utilizada en dicho proceso.

Para llevar a cabo la actinometría, en primer lugar se preparaba 1 L de

disolución de ácido perclórico/perclorato de fuerza iónica 0,03 M y pH = 2

(ajuste con hidróxido sódico). Por otra parte, en sendos matraces aforados de 100


81

mL de capacidad se preparaba una disolución de concentración 1,5 M en ácido

oxálico y otra de concentración 5x10‐2 M en Fe(III), ambas en medio ácido

perclórico/perclorato. El reactor se cargaba con 800 mL de la disolución inicial de

ácido perclórico/perclorato y se añadían los 100 mL de la disolución de ácido

oxálico. A continuación, se ponía en marcha el sistema de agitación y se

eliminaba el posible oxígeno disuelto mediante el burbujeo de una corriente de

nitrógeno. Pasados unos 20 minutos se tomaba una muestra inicial y se añadían

los 100 mL de la disolución de Fe(III), obteniéndose así una disolución final de

actinómetro Na3[Fe(C2O4)3] de concentración 5x10‐3 M. Alcanzado el régimen

estacionario de emisión de las lámparas, se introducía el reactor en la instalación

y a distintos intervalos de tiempo se extraían alícuotas y se procedía al análisis

del Fe(II) generado en la fotorreducción del Na3[Fe(C2O4)3].

En ausencia de cualquier otra especie susceptible de fotorreaccionar, la

velocidad con la que fotorreduce el ferrioxalato o, lo que es lo mismo, la

velocidad con la que aparece el Fe(II), viene dada por la expresión:

Fe(III) Fe(II)0 act

dC dCI ∙ ∙ 1 exp 2,303∙L∙ ∙C

dt dt (2.5)

donde I0 es la intensidad de la radiación incidente; φ el rendimiento cuántico de

la reacción fotoquímica a la longitud de onda de la radiación; L el paso efectivo

de radiación a través del reactor; ԑ el coeficiente de extinción molar del

ferrioxalato (500 M‐1 cm‐1 a 365 nm; [11]); y Cact la concentración de actinómetro.

Para valores del término 2,303∙L∙ԑ∙Cact superiores a 2, como ocurre en las

condiciones de trabajo elegidas, la ecuación (2.5) se simplifica en la (2.6):

Fe(III) Fe(II)

0

dC dCI ∙

dt dt (2.6)

de manera que la representación de la evolución de la concentración de Fe(II)

con el tiempo debe dar lugar a una línea recta de pendiente I0∙φ. Dado que el

valor de φ es conocido, puede obtenerse finalmente el valor de la intensidad o

CAPÍTULO 2

82

caudal de radiación incidente, I0.

De esta forma se obtuvo para la instalación de luz negra de la Fig. 2.1 un

valor de I0 = 4,43x10‐7 einstein s‐1.

2.2.1.11. Determinación de la concentración de formaldehído en disolución

El método empleado para la determinación de la concentración de

formaldehído fue el propuesto por Nash en 1953 [6], basado en el análisis

espectrofotométrico a 412 nm de la 3,5‐diaceti‐1,4‐dihidrolutidina, producto de

la reacción entre el formaldehído y la acetilacetona en presencia de iones de

amonio, conocida como reacción de Hantzsch.

En un vial de vidrio se introducían 2 mL de una mezcla de acetilacetona y

acetato de amonio (preparada por adición de 0,2 mL de acetilacetona, 3 mL de

ácido acético y 25 g de acetato de amonio a un matraz aforado de 100 mL de

capacidad, enrasando con agua ultrapura), y se añadían entre 1 y 5 mL de

muestra a analizar, completándose con agua ultrapura hasta los 5 mL si fuera

necesario. Las mezclas obtenidas eran calentadas a 50 °C durante 30 minutos en

la oscuridad, atemperadas y analizadas espectrofotométricamente. El blanco se

preparaba de la misma forma añadiendo 5 mL de la muestra a tiempo cero

(muestra que no contenía formaldehído).

Previamente, mediante el análisis de disoluciones patrón de formaldehído

obtenidas a partir del reactivo comercial, se determinó para la

diacetilhidrolutidina una absortividad molar de 7890 M‐1 cm‐1 a 412 nm.

Asimismo, se calculó un valor del límite de detección de este método de

6,57x10‐7 M.

En base a la Ley de Beer, la concentración molar de formaldehído en la

muestra, CCH2O, puede determinarse mediante la siguiente expresión:


83

2

m 0 TCH O

412nm m

Abs Abs VC

c V

(2.7)

donde Abs0 es la absorbancia del blanco y Absm la de la muestra analizada;

412nm la absortividad molar de la diacetilhidrolutidina a 412 nm; c el camino

óptico (1 cm); VT el volumen total (7 mL) y Vm el volumen de muestra (1 ‐ 5 mL).

2.2.1.12. Determinación de la demanda química de oxígeno (DQO)

La demanda química de oxígeno de una muestra de agua (DQO) representa

la cantidad de oxígeno necesaria para oxidar la materia orgánica e inorgánica

oxidable presente en la misma. Para su análisis se ha empleado un método

colorimétrico de reflujo cerrado [7], en el que la muestra se oxida con dicromato

de potasio en medio ácido sulfúrico y en caliente (148 °C), utilizando sulfato de

plata como catalizador. La interferencia debida a la posible presencia de iones

cloruro se neutraliza adicionando sulfato de mercurio al medio de reacción, que

precipita como cloruro de mercurio. La reducción de Cr(VI) a Cr(III) que tiene

lugar se evalúa fotométricamente por la disminución del color amarillo de la

mezcla.

El análisis de DQO se llevó a cabo empleando cubetas test LCK 414 de Hach

Lange (válidas para el rango de DQO de 5 a 60 mg L‐1 de O2), añadiendo 2 mL

de muestra según el procedimiento indicado por el fabricante. A continuación

las cubetas se introducían en un termorreactor Hach Lange modelo LT200

precalentado a 148 °C y se mantenía la digestión durante 2 horas a dicha

temperatura. Transcurrido ese tiempo, se dejaba que los viales alcanzaran

temperatura ambiente y se analizaba el valor de DQO, proporcionado

directamente por un espectrofotómetro DR2800 de Hach Lange.

2.2.1.13. Determinación de la biodegradabilidad: demanda biológica de oxígeno (DBO)

La demanda biológica de oxígeno de una muestra de agua (DBO5),

representa la cantidad de oxígeno consumido por microorganismos aerobios en

la degradación de la materia orgánica biodegradable presente en la misma

CAPÍTULO 2

84

durante un tiempo de incubación de 5 días. Es, por tanto, una medida de la

biodegradabilidad de un agua en condiciones aerobias.

El análisis de DBO5 del efluente de la EDAR de Badajoz antes/después de su

tratamiento se realizó mediante el método manométrico, basado en la pérdida

de presión en el interior de un recipiente debida al consumo de oxígeno por

parte de las bacterias aerobias involucradas en el proceso. A la vez que degradan

la materia orgánica, dichos microorganismos consumen el oxígeno disuelto en el

agua y liberan dióxido de carbono que se retira del sistema por absorción sobre

hidróxido sódico. El consumo del oxígeno disuelto provoca el desplazamiento

de aire desde el espacio de cabeza al agua para restablecer el equilibrio,

creándose así una depresión tanto mayor cuanto mayor es el oxígeno

consumido. Para determinar la demanda biológica de la materia orgánica

carbonácea el período de incubación es de cinco días.

Para llevar a cabo el análisis, 432 mL de la muestra previamente

acondicionada (filtrada en caso de ensayos con catalizador y con un pH

comprendido entre 6,5 y 7,5) se introducía en un biómetro Oxitop®, dotado de

tapón con dispositivo de medida de la presión del sistema, junto a unas gotas de

disolución inhibidora de nitrificación (disolución 5g L‐1 de aliltiourea, 20 gotas

por litro), y 1 mL de inóculo, procedente de la línea de recirculación de fangos

activos de la depuradora de aguas residuales urbanas de Badajoz. A

continuación, en el interior de la cápsula de goma que forma parte del tapón del

biómetro, se introducían 2 o 3 lentejas de hidróxido sódico para absorber el

dióxido de carbono liberado. El biómetro, cerrado herméticamente, se colocaba

en una cámara de incubación termostatizada a 20 °C y dotada de agitación

individual para cada botella durante cinco días, periodo a lo largo del cual cada

24 horas se almacenaba en la memoria el valor de DBO.

2.2.1.14. Determinación de la absorbancia a 254 nm: Aromaticidad

Los compuestos orgánicos aromáticos se caracterizan por absorber

fuertemente la radiación UV, por lo que la medida de la absorbancia de una


85

muestra de agua en esa zona del espectro es con frecuencia un indicador de la

concentración de este tipo de compuestos. Esta medida suele realizarse a 254

nm, longitud de onda que minimiza las posibles interferencias de otros

compuestos y, a su vez, proporciona las mayores bandas de absorción de los

compuestos de interés. En este trabajo, las medidas espectrofotométricas se

llevaron a cabo empleando cubetas de cuarzo de 1 cm en las que se introducían

las muestras de agua a analizar previamente filtradas.

2.2.1.15. Determinación de la concentración de fosfatos en disolución

La determinación de la concentración de fósforo en forma de fosfato (P‐PO43‐)

en el efluente de la EDAR de Badajoz antes/después de su tratamiento, se llevó a

cabo empleando el Kit LCK 349 de Lange (rango de medida comprendido entre

0,15 y 4,5 mg L‐1).

Los iones fosfato reaccionan en disolución ácida con iones molibdato y

antimonio formando un complejo antimonilfosfomolibdato, que se reduce en

presencia de ácido ascórbico a azul de fosfomolibdeno, compuesto coloreado

que puede analizarse espectrofotométricamente. La digestión previa de la

muestra a 100 °C durante 60 min (Digestor modelo LT200 de Hach Lange),

conduce a la transformación de todo el fósforo de la muestra en fosfato, lo que

permite determinar no solo el contenido en P‐PO43‐, sino también el contenido en

fósforo total (PT).

Para llevar a cabo el análisis se siguió el procedimiento indicado por el

fabricante, realizando la medición fotométrica con ayuda de un

espectrofotómetro DR2800 de Hach Lange.

2.2.1.16. Determinación de la turbidez

La turbidez es una medida del grado de pérdida de transparencia del agua

debido a la presencia de partículas en suspensión. Se mide en unidades

CAPÍTULO 2

86

nefelométricas de turbidez (NTU) y es una propiedad óptica que hace que la luz

sea dispersada y absorbida, en lugar de ser transmitida.

Para llevar a cabo el análisis se empleó un turbidímetro Hanna HI 93414, el

cual se calibraba con la ayuda de suspensiones patrón de distinta turbidez

suministradas por Hanna Instruments. Este instrumento está dotado de un

sistema óptico que incluye una lámpara con filamento de wolframio para

irradiar la muestra de agua contenida en una cubeta de vidrio, un detector de

luz dispersada y un detector de luz transmitida. Para el rango de medida (0 a

1000 NTU), el microprocesador del instrumento calcula, a partir de las señales

que llegan a los dos detectores, el valor NTU mediante un algoritmo efectivo.

2.2.1.17. Determinación de pH, conductividad y temperatura

La medida de pH se efectuó por potenciometría empleando un pH‐metro

Crison GLP 21+ con electrodo combinado, para un rango de pH comprendido

entre 0 y 14 y un intervalo de temperatura de 0 a 80 °C. El pH‐metro se calibraba

diariamente con disoluciones amortiguadoras de pH 4, 7 y 9, siguiendo el

manual del instrumento.

La conductividad de las muestras de agua se midió también por

potenciometría con un conductímetro Crison 524, el cual se calibraba con una

disolución patrón de cloruro de potasio de concentración 0,01 M (conductividad

1413 S cm‐1 a 25 °C). El microprocesador del instrumento mostraba

directamente la conductividad de la muestra en pantalla.

La medida de temperatura se realizó con un termómetro de contacto de

laboratorio que permitía lecturas en el rango de 0 a 60 °C.

2.2.2. Caracterización de los catalizadores empleados

En este apartado se describen las técnicas analíticas y los equipos que se

utilizaron para llevar a cabo la caracterización de los catalizadores ensayados

durante esta investigación, con el fin de poder correlacionar sus características


87

químico‐físicas con su actividad catalítica y/o fotocatalítica. En la Tabla 2.3 se

resumen y clasifican todas las técnicas de caracterización empleadas en función

el tipo de información que proporcionan, detallándose cada una de ellas en los

apartados siguientes.

Tabla 2.3. Clasificación de las técnicas y equipos de caracterización empleados.

Determinación Técnica de análisis Equipo

Análisis elemental:

composición química

Espectroscopía de emisión atómica

de plasma por acoplamiento

inductivo (ICP‐OES)

Espectrofotómetro de

emisión atómica ICP‐

OES Optima 3300DV

(Perkin‐Elmer)

Análisis estructural:

fases cristalinas,

tamaño de fases

cristalinas, proporción

de material amorfo, etc

Difracción de Rayos X (XRD) Difractómetro de Rayos

X Bruker D8 Advance

Espectroscopía Raman

Espectrómetro micro‐

Raman Nicolet Almega

XR Dispersive (Thermo

Scientific)

Análisis termogravimétrico y

térmico diferencial (TGA‐DTA)

Termobalanza SETSYS

Evolution‐16 (Setaram)

Microscopía electrónica de

transmisión (TEM)

Microscopio electrónico

de transmisión JEOL

JEM‐2100F y

microscopio electrónico

de transmisión Tecnai

G20 Twin (FEI

Company)

Análisis morfológico y

textural: topografía

superficial, área

superficial, volumen y

distribución de poros.

Microscopía electrónica de barrido

(SEM)

Microscopio electrónico

de barrido Hitachi

S‐4800

Isotermas de adsorción‐desorción

de nitrógeno

Equipo Autosorb‐1

(Quantachrome)

Análisis superficial:

acidez, grupos

superficiales, estado de

oxidación y porcentajes

atómicos en superficie

Valoración másica para determinar

el pH del potencial de carga cero

(pHPZC)

pH‐metro Crison GLP

21+ con electrodo

combinado

Espectroscopía fotoelectrónica de

Rayos X (XPS)

Espectrómetro de

fotoelectrones con

fuente de Rayos X K‐ (Thermo Scientific)

CAPÍTULO 2

88

Tabla 2.3 (Continuación). Clasificación de las técnicas y equipos de caracterización

empleados.

Determinación Técnica de análisis Equipo

Análisis de la

propiedades

electrónicas: ancho de

la banda de energía

prohibida

Espectroscopía ultravioleta‐visible

de reflectancia difusa (DR‐UV‐Vis)

Espectrofotómetro

UV‐Vis‐NIR Cary 5000

(Varian‐Agilent)

2.2.2.1. Espectroscopía de emisión atómica de plasma por acoplamiento inductivo (ICP‐

OES)

La espectroscopía de emisión atómica es una técnica de análisis elemental

capaz de determinar y cuantificar la mayoría de los elementos de la tabla

periódica en concentraciones desde % en peso hasta partes por millón [12]. Esta

técnica está basada en la excitación de un electrón desde su estado fundamental

hasta un nivel energético superior, debida a la absorción de radiación

electromagnética. El átomo así excitado vuelve nuevamente a su estado

fundamental, emitiendo una radiación cuya energía es característica de cada

elemento en particular y cuya intensidad es proporcional a la cantidad de dicho

elemento en la muestra analizada [12].

Esta técnica se ha empleado para llevar a cabo el análisis del contenido total

en W en algunos catalizadores, mediante un analizador ICP‐OES Perkin Elmer,

modelo Optima 3300 DV, equipado con un detector UV. La disgregación de las

muestras se llevó a cabo mediante digestión ácida, vía microondas.

El análisis fue realizado por la Unidad de Apoyo a la Investigación del

Instituto de Catálisis y Petroleoquímica del CSIC (ICP‐CSIC).

2.2.2.2. Difracción de Rayos X (XRD)

La difracción de rayos X es una técnica muy utilizada para la identificación y

caracterización de las fases cristalinas presentes en un sólido. Aprovechando


89

que las estructuras cristalinas poseen planos (producidos por ordenamientos

repetitivos de átomos), capaces de difractar rayos X, la medida de los ángulos en

los que un haz de rayos X de longitud de onda determinada es difractado por la

muestra permite obtener los diagramas de difracción. El espaciado entre dos

planos hkl (d), que es la base de la caracterización de fases y estructura de un

material, está relacionado con el ángulo de difracción 2θ mediante la ley de

Bragg [13]:

n

d2 sen

(2.8)

donde d es la distancia interplanar, n un número entero que representa el orden

de difracción y λ la longitud de onda de la fuente de rayos X.

Esta es una técnica muy usada por su carácter no destructivo, rapidez y

mínima cantidad de muestra necesaria para identificar las fases cristalinas,

cuantificarlas, determinar el grado de cristalinidad y el tamaño del cristal.

Para la determinación de las dimensiones del cristal se puede emplear la

ecuación de Scherrer [13]:

K

Lcos

(2.9)

siendo L el tamaño de cristal, K una constante relativa a un factor de forma del

cristal (en ausencia de información detallada sobre la forma del cristal K = 0,9 es

una buena aproximación para cristales esféricos [14]) y la anchura a la mitad

de la intensidad máxima de un pico seleccionado.

La señal de difracción emitida por un sólido cristalino es una huella de su

estructura y la intensidad de las líneas de difracción es función de la

concentración de las diferentes fases cristalinas. Por comparación de las

distancias interplanares se puede determinar la fase existente, tomando como

referencia las correspondientes a los compuestos puros [13]. Así, cada

compuesto puro tiene un difractograma específico y unívoco recogido en las

fichas estandarizadas ASTM en la base de datos de la ICDD (International

CAPÍTULO 2

90

Centre for Diffraction Data) [15].

Las medidas de Difracción de Rayos X de distintos catalizadores en polvo

fueron realizadas por el Servicio de Análisis y Caracterización de Sólidos y

Superficies (SACSS) perteneciente a los SAIUEx, empleando un difractómetro

Bruker D8 Advance, que utiliza la radiación Kα del cobre (= 0,1541 nm). Los

difractogramas se registraron para valores de 2θ comprendidos entre 20° y 80°

con paso de barrido de 0,02 s‐1 y tiempo de acumulación de 1 s por punto.

A partir de la intensidad de las líneas de difracción es posible determinar de

forma semicuantitativa el porcentaje de las diferentes fases cristalinas presentes.

Para ello, el equipo dispone del software EVA v.14 (Bruker‐AXS) que, mediante

el método RIR (Reference Intensity Ratio), determina dichos porcentajes

basándose en la relación entre las intensidades de los picos de cada fase

cristalina y la intensidad del corindón como patrón de referencia (Ip/Ipcor).

2.2.2.3. Espectroscopía Raman

La espectroscopía Raman se basa en el efecto Raman, que se produce al

irradiar una sustancia con luz de determinada longitud de onda, teniendo una

parte de la luz dispersada una longitud de onda distinta [16]. Al irradiar una

sustancia con luz monocromática, la mayoría de la dispersión de luz se produce

de forma elástica, es decir, no se produce un intercambio energético neto entre la

radiación y la muestra y, por tanto, la luz dispersada tiene la misma longitud de

onda que la luz incidente, fenómeno al que se denomina dispersión Rayleigh.

Sin embargo, una pequeña parte de la luz incidente produce un cambio en el

estado vibracional del sistema y, en este caso, la luz dispersada tiene una

longitud de onda distinta. Dentro de esta dispersión, conocida como dispersión

Raman, existen dos tipos: la dispersión Stokes, en la que la longitud de onda

final es mayor que la del haz incidente; y la anti‐Stokes, que da lugar a una luz

dispersa con longitud de onda menor que la del haz incidente [17].

De esta manera, la espectroscopía Raman mide frecuencias vibracionales


91

como diferencias energéticas entre la luz incidente y la dispersada. Estas

diferencias son independientes de la frecuencia del haz incidente. El espectro

Raman se representa como intensidad frente a desplazamiento Raman, siendo

este la diferencia entre el número de onda de la luz dispersa y la de excitación.

La espectroscopía Raman es muy útil en la caracterización de sólidos, pues

además de aportar información sobre la simetría del sistema e identificar las

fases existentes por medio del número, intensidad y posición de las bandas

presentes en el espectro, permite realizar un análisis sobre efectos relacionados

con el tamaño de partícula y la presencia de defectos. Así, por ejemplo, un

incremento de la anchura de las bandas puede deberse a la no estequiometría de

la muestra, ocasionada por la presencia de vacantes de oxígeno o bien inducida

por la presencia de fases parcialmente reducidas minoritarias.

Los espectros Raman de algunos de los catalizadores sintetizados fueron

obtenidos por el SAEM‐SAIUEx. Dichos espectros se registraron en un equipo

Nicolet Almega XR Dispersive micro‐Raman (Thermo Scientific) con una

resolución espectral de 2 cm‐1. Como fuente de excitación se utilizó un láser de

λ = 633 nm, con la potencia del mismo al 100 %.

2.2.2.4. Análisis termogravimétrico y térmico diferencial (TGA‐DTA)

El análisis termogravimétrico (TGA) se basa en el registro de los cambios de

peso que se producen en un material al someterlo a calentamiento o

enfriamiento a una velocidad conocida y en una atmósfera controlada. Por otro

lado, en el análisis térmico diferencial (DTA) se registran los cambios de

temperatura que tienen lugar durante el tratamiento térmico con respecto a una

sustancia de referencia térmicamente inerte. Cuando se realizan conjuntamente,

estas técnicas permiten determinar pérdidas de humedad, descomposiciones y

transformaciones de fase, información a partir de la cual pueden establecerse las

condiciones óptimas para llevar a cabo el tratamiento térmico de las muestras.

Dado que una serie de los catalizadores empleados en esta investigación se

CAPÍTULO 2

92

prepararon por descomposición térmica de un precursor comercial, se llevó a

cabo el análisis termogravimétrico y térmico diferencial de dicho precursor. El

análisis, realizado por el SACSS‐SAIUEx, se efectuó en una termobalanza

SETSYS Evolution‐16 (Setaram), con una balanza que detecta cambios de peso

con una resolución de 0,04 g. La muestra, normalmente entre 10 y 20 mg, se

introducía en un crisol de alúmina, empleando un caudal de 50 mL min‐1 de aire

e incrementando la temperatura desde temperatura ambiente hasta 600 °C a una

velocidad de calentamiento de 5 °C min‐1.

2.2.2.5. Microscopía electrónica de transmisión (TEM)

La microscopía electrónica de transmisión (TEM) permite obtener

información estructural y morfológica a nivel nanométrico. Mediante el empleo

de esta técnica se pueden identificar las distintas fases existentes en las

partículas del catalizador, su cristalinidad y la dispersión en la que se encuentra

la fase activa. En esta técnica un haz de electrones se focaliza mediante dos

lentes condensadoras sobre una muestra delgada y transparente a los electrones.

Después de atravesar la muestra, los electrones son recogidos y focalizados por

la lente objetivo dentro de una imagen intermedia ampliada que, además, es

aumentada con las lentes proyectoras y, finalmente, proyectada sobre una

pantalla fluorescente o una película fotográfica [18].

Las muestras en polvo se dispersaron en etanol y se depositaron en una

rejilla de cobre recubierta con un polímero orgánico. Una vez evaporado el

alcohol se procedió al análisis en un microscopio electrónico de transmisión

JEM‐2100F (JEOL) y en otro Tecnai G20 Twin (FEI Company). Ambos equipos,

pertenecientes a ICP‐CSIC y SACSS‐SAIUEx, respectivamente, operaban con

una tensión de aceleración de 200 kV.

2.2.2.6. Microscopía electrónica de barrido (SEM)

La microscopía electrónica de barrido (SEM) es una técnica que permite

obtener información estructural y morfológica a nivel micrométrico sobre el


93

tamaño y la forma de las partículas que constituyen un material sólido.

En un microscopio electrónico de barrido se hace incidir sobre una muestra

gruesa y opaca a los electrones un haz delgado de electrones acelerados, con

energías desde cientos de eV hasta decenas de keV. El haz es desplazado sobre

la superficie de la muestra gracias a un juego de bobinas deflectoras,

rastreándola. Cuando el haz primario entra en contacto con la superficie de la

muestra, parte de los electrones es reflejada y otra parte penetra unas pocas

capas atómicas, siguiendo una trayectoria complicada antes de volver a emerger

a la superficie. La intensidad de estas dos emisiones varía en función del ángulo

que forma el haz incidente con la superficie del material, es decir, depende de la

topografía de la muestra. Las señales emitidas se recogen mediante un detector

y se amplifican. Como resultado se obtiene una imagen topográfica de la

muestra muy ampliada [18].

El equipo utilizado para la adquisición de micrografías de SEM de los

catalizadores, perteneciente al SACSS‐SAIUEx, fue un microscopio electrónico

de barrido de la marca Hitachi S‐4800. Se trabajó con un voltaje de aceleración

de 20‐30 kV y 500 ‐ 2000 de magnificación. Las muestras se dispersaron en un

portamuestras recubierto de carbono y un adhesivo para mantener fijas las

partículas de sólido.

2.2.2.7. Isotermas de adsorción‐desorción de nitrógeno

La isoterma de adsorción de un gas se define como la cantidad de gas

adsorbido en un sólido a distintas presiones relativas de gas, manteniendo

constante la temperatura. Cada punto de la isoterma obtenido representa un

punto de equilibrio entre el volumen del gas adsorbido y la presión relativa del

gas (P/P0). Una vez alcanzadas presiones relativas próximas a la unidad, es

posible determinar la cantidad de adsorbato que permanece retenido para

valores decrecientes de P/P0, conociéndose la curva resultante como isoterma de

desorción. Las curvas de adsorción y desorción no tienen por qué coincidir en

un determinado intervalo de presiones relativas, denominándose a la diferencia

CAPÍTULO 2

94

entre ambas curvas ciclo de histéresis. Para un tipo de sólido, la forma de la

isoterma y del ciclo de histéresis están relacionados con las diferencias en la

energía de interacción entre adsorbato y adsorbente y con la estructura porosa

(micro, meso o macroporos) del sólido. La mayoría de las isotermas que se

pueden encontrar en la bibliografía científica pertenecen a uno de los cinco tipos,

denominados I a V, de la clasificación original de Brunauer, Deming, Deming y

Teller [19] o del tipo VI añadido por la IUPAC [20]. Asimismo se reconocen

cuatro tipos de bucles de histéresis [21].

Las medidas experimentales de adsorción‐desorción de nitrógeno se llevaron

a cabo a ‐196° C en un equipo Autosorb 1 (Quantachrome) perteneciente al

SACSS‐SAIUEx. Las muestras fueron desgasificadas durante 24 horas a 250 °C

con una presión de alto vacío (10‐4 Pa) para asegurar que la superficie estuviera

libre de especies adsorbidas.

Para la determinación de la superficie específica a partir de esta técnica, la

IUPAC recomienda la metodología desarrollada por Brunauer, Emmett y Teller

[22], que desarrollaron la ecuación que hoy se conoce como BET (método BET).

La ecuación BET relaciona el volumen de gas adsorbido a una determinada

presión relativa con el volumen adsorbido en una monocapa de adsorbato sobre

el sólido.

En esta investigación, se calculó el valor de la superficie específica de los

catalizadores a partir de los datos correspondientes a presiones relativas entre

0,02 y 0,15, procesando los mismos con ayuda del software AS1winTM 2.01. El

software también permite determinar el volumen de microporos a partir de la

rama de adsorción de la isoterma mediante el método “t”, aplicando la ecuación

de Halsey [23,24]. En esta determinación, de la representación del volumen

adsorbido frente al parámetro t se obtiene una recta cuya pendiente se relaciona

con el área externa no asociada a los microporos. De esta forma, el área

microporosa se obtuvo por diferencia entre el área BET y el área externa.

Finalmente, el volumen de mesoporos se determinó a partir de la cantidad de


95

nitrógeno adsorbido a la presión relativa de 0,96 en la rama de desorción de la

isoterma, zona equivalente al llenado de todos los poros de diámetro hasta 50

nm, restando al valor obtenido el volumen de microporos calculado por el

método “t”.

2.2.2.8. Determinación del pH del potencial de carga cero (pHPZC) por valoración másica

Para las reacciones en medio acuoso, una medida que da una buena

indicación acerca del carácter ácido o básico superficial de un sólido es la

determinación del pH de la suspensión acuosa del mismo (en terminología

anglosajona “pH slurry”), el cual se relaciona con el pH del potencial de carga

cero (pHPZC) dando una buena indicación acerca de la carga electrónica

superficial del material [25].

Para la estimación del pHPZC de distintas muestras de catalizador se empleó

el método de valoración másica propuesto por Subramanian et al. en 1988 [26].

Para ello, se determinó el pH de la disolución acuosa resultante de mantener

una suspensión de sólido al 5 % en peso en agua ultrapura, con agitación

continua y en un recipiente cerrado. Transcurridas 24 horas se filtró la muestra y

se midió el pH del agua resultante con un pH‐metro Crison GLP 21+. Este

procedimiento se repitió dejando la muestra en agitación hasta que el pH

medido alcanzó un valor constante.

2.2.2.9. Espectroscopía fotoelectrónica de Rayos X (XPS)

La espectroscopía fotoelectrónica de rayos X (XPS) es una de las técnicas de

análisis químico englobadas bajo el nombre de ESCA (Electron Spectroscopy for

Chemical Analysis), utilizada para determinar el estado químico y la

composición superficial de materiales sólidos. Está basada en el efecto

fotoeléctrico: al irradiar una muestra con fotones de energía superior a la energía

de enlace de los electrones de sus átomos, dichos electrones salen del sólido con

una energía cinética igual a la diferencia de energía entre el fotón y la energía de

enlace del electrón. En este sentido, aunque los rayos X utilizados pueden

CAPÍTULO 2

96

penetrar unas pocas micras en una muestra sólida, solo los electrones generados

a unos pocos nanómetros de la superficie pueden salir del sólido. Esto se debe a

que los fotoelectrones producidos en las capas más internas sufren colisiones

inelásticas que provocan una pérdida de energía tal que no pueden abandonar la

superficie del material. Por tanto, esta técnica es especialmente sensible para

análisis químico superficial, proporcionando información química de las 5 ‐ 10

primeras capas atómicas del sólido.

El espectro de XPS es la representación del número de electrones detectados

en un intervalo de energías frente a su energía cinética o, más comúnmente,

frente a su energía de enlace. Las variaciones de energía de enlace de un

elemento respecto a su estado no combinado se deben a las diferencias en el

potencial químico y en la polaridad de los compuestos, por lo que pueden

usarse para identificar el estado químico de los elementos analizados dado que

cada elemento tiene un conjunto de energías de enlace características [27,28].

El análisis de muestras de catalizadores por espectroscopía fotoelectrónica de

rayos X fue realizado por el SACSS‐SAIUEx. Los espectros XPS se obtuvieron

utilizando un espectrómetro de fotoelectrones K‐ de Thermo Scientific

equipado con fuente de rayos X de Al Kα (hν = 1253,6 eV), que operaba a un

voltaje de 12 kV bajo vacío (2x10‐7 mbar). Se realizaron espectros de detalle de

los elementos más importantes en las muestras y, para eliminar el efecto de

carga, las energías de enlace se corrigieron utilizando como patrón el pico C1s

(Energía de enlace 284,6 eV).

En todos los casos el espectro XPS se resolvió en varias componentes,

ajustándose cada componente, mediante el programa de tratamiento de

espectros XPSPEAK 4.0, a la curva experimental con una combinación lineal de

curvas lorentzianas y gaussianas en proporciones variables y escogiendo el

mejor ajuste mediante minimización de residuos χ2. Las relaciones atómicas de

los elementos en superficie se obtuvieron a partir de la siguiente ecuación [29]:


97

i i i

j j j

n I / S

n I / S (2.10)

donde ni es el número de átomos por cm3 del elemento i, Ii es la intensidad del

pico (generalmente expresada como área) y Si es su factor de sensibilidad

atómica.

2.2.2.10. Espectroscopía ultravioleta‐visible de reflectancia difusa (DR‐UV‐Vis)

La espectroscopía UV‐Vis se fundamenta en la absorción electrónica de la

radiación electromagnética cuando interacciona con la materia en el entorno de

longitudes de onda entre 190 nm y 800 nm. En el caso de los catalizadores

estudiados en este trabajo (semiconductores), se observa fundamentalmente la

transición de electrones desde la banda de valencia a la banda de conducción,

cuya energía corresponde a esta región del espectro electromagnético.

La medida de la reflectancia difusa se define como la fracción de radiación

incidente que es reflejada por la muestra en todas direcciones. Para ello, se

emplea un dispositivo llamado esfera integradora, consistente en una esfera

hueca recubierta en su interior de un material altamente reflectante, que envía la

luz reflejada por la muestra al detector [30]. El espectro resultante se obtiene

como tanto por ciento de reflectancia frente a la longitud de onda, fijando como

100 % de reflectancia la obtenida para una muestra de referencia que no absorba

luz en el rango de longitudes de onda utilizado (generalmente BaSO4).

Cerca del borde de absorción se ha establecido que:

ngh A h E (2.11)

siendo h la constante de Planck, equivalente a 4,136x10‐15 eV s; A una constante;

Eg la energía del salto de banda en eV; la frecuencia de la radiación en s‐1; el

exponente n que toma el valor de ½ para transiciones electrónicas directas; y el

coeficiente de absorción [31]. Este coeficiente se determina como:

‐ln R (2.12)

CAPÍTULO 2

98

donde R es el valor de reflectancia medida respecto a la unidad.

De esta manera, reordenando la ecuación (2.11) se tiene que:

2gh A h E (2.13)

Si se representa el primer término de la ecuación (h)2 frente a h se

determina el valor del salto de banda, Eg, como el corte en el eje h cuando α es

igual a cero.

Los análisis de espectroscopía UV‐visible de reflectancia difusa fueron

realizados por el personal de la Unidad de Apoyo a la Investigación del ICP‐

CSIC, en un equipo UV‐Vis‐NIR Cary 500 (Varian‐Agilent Technologies)

equipado con esfera integradora, registrando espectros de absorbancia y

porcentajes de reflectancia para valores de longitudes de onda comprendidos

entre 200 y 900 nm a intervalos de 1 nm.

2.3. REACTIVOS EMPLEADOS

En la Tabla 2.4 se muestran, clasificados en función de su aplicación, los

productos químicos empleados a lo largo de esta investigación, indicándose la

pureza de los mismos y la casa comercial que los suministró.

En todos los casos en los que fue necesaria la preparación de disoluciones

acuosas, se empleó agua ultrapura (grado de pureza milliQ‐Millipore).


99

Tabla 2.4. Reactivos empleados.

Reactivo Pureza Fabricante Aplicación

Acetaminofeno (C8H9NO2) 99 % Sigma‐Aldrich

Compuestos

modelo y matriz

para estudios de

procesos de

degradación en

agua

Agua residual urbana EDAR Badajoz

Antipirina (C11H12N2O) 99 % Sigma‐Aldrich

Cafeína (C8H10N4O2) 99 % Sigma‐Aldrich

Carbamazepina (C15H12N2O) > 98 % Sigma‐Aldrich

Diclofenaco de sodio

(C14H10Cl2NaO2) 99 % Sigma‐Aldrich

Hidroclorotiazida (C7H8ClN3O4S2) 99 % Sigma‐Aldrich

Ibuprofeno sódico (C13H17NaO2) > 98 % Sigma‐Aldrich

Ketorolaco sal tris

(C15H13NO3∙C4H11NO3) > 99 % Sigma‐Aldrich

Metanol (CH3OH) PAI‐ACS

(grado HPLC)

Panreac

Química

Metoprolol tartrato

(C15H25NO3)2∙C4H6O6) > 99 % Sigma‐Aldrich

N,N‐Dietil‐meta‐toluamida

(C12H17NO) 97 % Sigma‐Aldrich

Sulfametoxazol (C10H11N3O3S) > 98 % Fluka

Biochemika

Ácido clorhídrico (HCl) 36,7 % Fisher

Chemical

Ajuste de pH

Ácido ortofosfórico (H3PO4) 85 % Panreac

Química

Ácido perclórico (HClO4) 70 % Panreac

Química

Hidróxido sódico (NaOH) 98 % Panreac

Química

CAPÍTULO 2

100

Tabla 2.4 (Continuación). Reactivos empleados.


Ácido cítrico (C6H8O7) 99 % Panreac

Química

Síntesis de

catalizadores


Chemical

Ácido nítrico (HNO3) 65 % Panreac

Química

Ácido oxálico dihidratado

(C2H2O4∙2H2O) 99,5 % Sigma‐Aldrich

Ácido wolfrámico (H2WO4) > 99 % Sigma‐Aldrich

Cloruro cálcico (CaCl2) 99 % Panreac

Química

Dióxido de titanio P25 (TiO2) > 99 % Evonik‐

Degussa AG

Disolución de amoníaco (NH4OH) 35 % Fisher

Chemical

Etanol (C2H6O) 99,5 % Panreac

Química


Química

Nitrato de cerio (III)

hexahidratado (Ce(NO3)3∙6H2O) 99 % Sigma‐Aldrich

Trióxido de wolframio (WO3) > 99 % Sigma‐Aldrich

Wolframato cálcico (CaWO4) 99,9 % Sigma‐Aldrich

Wolframato sódico (Na2WO4) > 99 % Sigma‐Aldrich

Oxígeno comprimido (O2) > 99,999 % Abelló Linde Agente oxidante y

obtención de ozono

Acetonitrilo (C2H3N) PAI‐ACS

(grado HPLC)

Panreac

Química

Determinación de

la concentración de

contaminantes

modelo en agua e

identificación de

intermedios de

degradación

Ácido fórmico (CH2O2) 98 % Panreac

Química


Química


101




Química

Determinación de

la concentración

de ozono en

disolución acuosa

Fosfato diácido de potasio

(KH2PO4) > 98 %

Panreac

Química

Índigo carmín (5,5’,7‐

indigotrisulfonato de potasio, C16H8N2Na2O8S2)

98 % Sigma‐Aldrich

Bicarbonato de sodio (NaHCO3) 99,7 % Panreac

Química Determinación de

la concentración

de peróxido de

hidrógeno en

disolución

Calgón (Polifosfato de sodio,

((NaPO3)n) 65 ‐ 70 %

Panreac

Química

Cloruro de cobalto (II)

hexahidratado (CoCl2∙6H2O) 99 %

Panreac

Química

Acetato de amonio (CH3CO2NH4) 98 % Sigma‐Aldrich

Determinación de

la concentración

de formaldehído

en disolución

Acetilacetona

(CH3COCH2COCH3) 99 % Sigma‐Aldrich

Ácido acético glacial (CH3COOH) 100 % Merck

Formaldehído (CH2O) 37 % Sigma‐Aldrich


Determinación de

la intensidad de

radiación y de las

concentraciones de

Fe(II) y Fe total en

disolución



Ácido perclórico (HClO4) 70 % Panreac

Química

Fluoruro de amonio (NH4F) 98 % Sigma‐Aldrich

Nitrógeno comprimido (N2) > 99,999 % Abelló Linde

o‐Fenantrolina (C12H8N2∙H2O) 99 % Fluka

Biochemika

Perclorato de hierro(III) hidratado

(Cl3FeO12∙xH2O) PA‐ISO Sigma‐Aldrich

Reactivo para análisis fotométrico

de hierro, Spectroquant ‐ Merck

CAPÍTULO 2

102




Chemical

Determinación de

la concentración

de carbono

orgánico total en

disolución


Química

Aire sintético comprimido (O2/N2) 21 ± 0,5 % O2 Abelló Linde

Bicarbonato de sodio (NaHCO3) PA‐ISO Nacalai Tesque

Carbonato de sodio (Na2CO3) PA‐ISO Nacalai Tesque

Ftalato ácido de potasio

(KHC8H4O4) PA‐ISO Nacalai Tesque


Determinación de

la concentración

de algunos ácidos

orgánicos e iones

inorgánicos en

disolución

Ácido fórmico (CH2O2) 98 % Panreac

Química



Ácido sulfúrico (H2SO4) 95 ‐ 98 % Panreac

Química

Carbonato de sodio (Na2CO3) > 99,8 % Sigma‐Aldrich

Cloruro de sodio (NaCl) 99 % Panreac

Química

Fosfato de potasio dibásico

(K2HPO4) > 98 %

Panreac

Química

Nitrato de sodio (NaNO3) 99 % Panreac

Química

Piruvato de sodio (C3H3NaO3) > 99 % Sigma‐Aldrich

Sulfato de sodio (Na2SO4) 99 % Panreac

Química



Determinación de

parámetros

cinéticos Ácido p‐clorobenzoico (C7H5ClO2) > 98 % Merck

ter‐butanol(C4H10O) 99 % Panreac

Química


103



Cubetas test LCK 349 ‐ Hach‐Lange

Determinación de

la concentración

de fosfatos en

disolución

Cubetas test LCK 414 ‐ Hach‐Lange

Determinación de

la demanda

química de

oxígeno

Aliltiourea (C4H8N2O) > 98 % Panreac

Química Determinación de

la demanda

biológica de

oxígeno


Química

Microorganismos aerobios ‐ EDAR Badajoz

BIBLIOGRAFÍA

[1] Bader, H.; Hoigné, J. “Determination of ozone in water by the indigo

method”. Water Res. 15 (1981) 449‐456.

[2] Masschelein, W.; Denis, M.; Ledent, R. “Spectrophotometric determination of

residual hydrogen peroxide”. Water & Sewage Works (1977) 69‐72.

[3] Stookey, L.L. “Ferrozine‐ A new spectrophotometric reagent for iron”. Anal.

Chem. 42 (1970) 779‐781.

[4] Zuo, Y. “Kinetics of photochemical/chemical cycling of iron coupled with

organic substances in cloud and fog droplets”. Geochim. Cosmochim. Act. 59

(1995) 3123‐3130.

[5] Hatchard, C.G.; Parker, C.A. “A new sensitive chemical actinometer. II.

Potassium ferrioxalate as a standard actinometer”. Proceedings of the Royal

Society, London Service A 235 (1956) 518‐536.

[6] Nash, T. “The colorimetric estimation of formaldehyde by means of the

Hantzsch reaction”. Biochemistry 55 (1953) 416‐421.

[7] Moore, W.A.; Krorner, R.C.; Ruchchoff, C.C. “Dichromate reflux method for

determination of oxygen consumed”. Anal. Chem. 21 (1948) 953‐957.

[8] Rodríguez, E.; Mimbrero, M.; Masa, F.J.; Beltrán, F.J. “Homogeneous iron‐

catalyzed photochemical degradation of muconic acid in water”. Water Res. 41

CAPÍTULO 2

104

(2007) 1325‐1333.

[9] Sandell, E.B. “Colorimetric determination of traces of metals”. Interscience

Pubs., New York 1959.

[10] Goldstein, S.; Rabani, J. “The ferrioxalate and iodide‐iodate actinometers in

the UV region”. J. Photochem. Photobiol. A Chem. 193 (2008) 50‐55.

[11] Safarzadeh‐Amiri, A.; Bolton, J.; Cater, S. “Ferrioxalate‐mediated

photodegradation of organic pollutants in contaminated water”. Water Res. 31

(1997) 787‐798.

[12] Faraldos, M. en “Técnicas de Análisis y Caracterización de Materiales”.

Faraldos, M.; Goberna, C. (eds.), CSIC, Madrid (2003), cap. 9.

[13] Murcia‐Mascarós, S. en “Técnicas de Análisis y Caracterización de

Materiales”. Faraldos, M.; Goberna, C. (eds.), CSIC, Madrid (2003), cap. 10.

[14] Klug, H.P.; Alexander, L.E. “X‐ray diffraction procedures for polycrystalline

and amorphous materials”. 2nd. Edition, John Wiley & Sons (1974).

[15] www.icdd.com (consultada en enero de 2017).

[16] Raman, C.V.; Krishnan, K.S. “A new type of secondary radiation”, Nature

121 (1928) 501‐502.

[17] Bañares, M.A.; Valenzuela, R.X. en “Técnicas de Análisis y Caracterización

de Materiales”. Faraldos, M.; Goberna, C. (eds.), CSIC, Madrid (2003), cap. 5.

[18] Díaz, I.; Landa, A.R.; Otero, L.C. en “Técnicas de Análisis y Caracterización

de Materiales”. Faraldos, M.; Goberna, C. (eds.), CSIC, Madrid (2003), cap. 11.

[19] Brunauer, S.; Deming, L.S.; Deming, W.S.; Teller, E. “On a theory of the van

der Waals adsorption of gases”. J. Am. Chem. Soc. 62 (1940) 1723‐1732.

[20] Haber, J. “Manual on catalyst characterization”. Pure Appl. Chem. 63 (1991)

1227‐1246.

[21] Sing, K.S.W.; Everett, D.H.; Haul, R.A.W.; Moscou, L.; Pierotti, R.A.;

Rouquerol, J.; Siemieniewska, T. “Reporting physisorption data for gas/solid

systems with special reference to the determination of surface area and

porosity”. Pure Appl. Chem. 57 (1985) 603‐619.

[22] Brunauer, S.; Emmett, P.; Teller, E. “Adsorption of gases in multimolecular

Layers”. J. Am. Chem. Soc. 60 (1938) 309‐319.

[23] Halsey, C.P. “Physical adsorption on non‐uniform surfaces”. J. Chem. Phys.

16 (1948) 931‐937.

[24] Lippens, B.C.; Linsen, B.G.; de Boer, J.H. “Studies on pore systems in

catalysts I. The adsorption of nitrogen; apparatus and calculation”. J. Catal. 3


105

(1964) 32‐37.

[25] Menéndez, J.A.; Illán‐Gómez, M.J.; León y León, C.A.; Radovic, L.R. “On the

difference between the isoelectric point and the point of zero charge of carbons”.

Carbon 33 (1995) 1655‐1657.

[26] Subramanian, S., Noh, J.S., Schwarz, J.A. “Determination of the point of

zero‐charge of composite oxides”. J. Catal. 114 (1988) 433‐439.

[27] Moulder, J.F.; Stickle, W.F.; Sobol, P.E.; Bomben, K.D. Handbook of X‐Ray

Photoelectron Spectroscopy: A Reference Book of Standard Spectra for

Identification and Interpretation of XPS Data (Hardcover), Physical Electronics,

Reissue ed. (1995).

[28] Campos, J.M. en “Técnicas de Análisis y Caracterización de Materiales

Faraldos, M.; Goberna, C. (eds.), CSIC, Madrid (2003), cap. 12.

[29] Wagner, C.D.; Davis, L.E.; Zeller, M.V.; Taylor, J.A.; Raymond, R.H.; Gale,

L.H. “Empirical atomic sensitivity factors for quantitative analysis by electron

spectroscopy for chemical analysis”. Surf. Interf. Anal. 3 (1981) 211‐225.

[30] Springsteen, A. en “Applied Spectroscopy”, Workman, J.; Springsteen, A.

(eds.), Academic Press, San Diego (1998), cap.6.

[31] Serpone, N.; Lawless, D.; Khairutdinovt R. “Size effects on the

photophysical properties of colloidal anatase TiO2 particles: size quantization or

direct transitions in this indirect semiconductor?” J. Phys. Chem. 99 (1995)

16646‐16654.

CAPÍTULO 3 (CHAPTER 3) PAPER 1: On ozone-photocatalysis synergism in black-light induced reactions: Oxidizing species production in photocatalytic ozonation versus heterogeneous photocatalysis

E. Mena, A. Rey, B. Acedo, F.J. Beltrán, S. Malato

Chemical Engineering Journal 204-206 (2012) 131-140

ABSTRACT. The synergism produced between ozone and TiO2 black light photocatalytic oxidation of methanol has been studied following the rate of formaldehyde formation during photocatalytic oxidation, ozonation and photocatalytic ozonation experiments. Methanol was selected as a model compound due to its low reaction rate with molecular ozone and its scavenging character for both, free hydroxyl radicals and trapped holes. TiO2-P25 was used as photocatalyst and black light blue lamps (emitting with a maximum at 365 nm) as radiation source. The effect of ozone concentration and pH was evaluated. Absorbed light intensity by the photocatalyst was also determined to calculate the quantum yields of photocatalytic reactions. Three main processes need to be considered during photocatalytic ozonation: direct ozone-methanol reaction, indirect ozone reactions and photocatalytic reactions, which allow calculating the quantum yield of photo-generated oxidizing species. The presence of ozone exerts a positive effect in the reaction rate of oxidizing species formation due to light induced reactions also enhancing the quantum yield from 0.34 to 0.80 mol einstein-1 at pH = 3 (where indirect ozone reactions are negligible). This parameter increased from 0.29 to 3.27 mol einstein-1 at pH = 7 likely due to indirect ozone reactions that cannot be disregarded. The positive effect of ozone in the photocatalytic induced reactions has been attributed to the reaction of dissolved ozone and hydrogen peroxide (formed upon methanol direct ozonation) as electron acceptors, thus reducing the recombination process on the catalyst surface to some extent. A simplified economic study is also presented.

Keywords: Photocatalytic ozonation, ozone, TiO2, synergism, methanol, formaldehyde.

PAPER 1: On ozone‐photocatalysis synergism in black‐light induced reactions:

Oxidizing species production in photocatalytic ozonation versus heterogeneous photocatalysis

109

3.1. INTRODUCTION

Wastewater is one of the most important problems in industrialized regions.

Many of these effluents contain high concentrations of organic pollutants that

must be removed in order to fulfill the required limits of the increasingly

restrictive legislation. In this context, photocatalytic detoxification treatments

based on TiO2 semiconductor have been the focus of numerous investigations in

the last 30 years for the destruction of undesirable contaminants in water [1‐3].

In heterogeneous photocatalysis, photoinduced holes and electrons in

semiconductor particles, through a complex reaction mechanism, give place to

highly oxidizing species which play a key role in degradation of organic

pollutants. The most commonly used semiconductor has been polycrystalline

powders of titanium dioxide due to unique properties such as chemical stability,

safety and low cost [2]. However, its photohole‐electron recombination is a

serious problem for the development of photocatalytically based technologies

since it severely limits the quantum yields achievable [2‐4]. Several strategies

have been proposed to minimize this problem and increase the process

efficiency (ion doping, different semiconductors coupling, using chemical

oxidants or combining photocatalysis with other advanced oxidation processes

(AOPs) [2]). Among them, the combination of ozone and heterogeneous

photocatalysis with TiO2 (photocatalytic ozonation) has demonstrated to be an

efficient treatment enhancing the formation of oxidizing species compared to the

single ozonation or photocatalytic processes [5,6].

When a semiconductor (e.g. TiO2) is irradiated with a photon of energy

greater than its band gap energy an electron/hole pair is formed in the

conduction band (CB) and valence band (VB), respectively. These mobile species

can migrate to the TiO2 surface and/or can be readily trapped forming less

mobile states (to simplify e‐/h+ stand for all the forms of holes and electrons).

2TiO h e h (3.1)

In addition, these species can give place to the recombination reaction (3.2):

CAPÍTULO 3 (CHAPTER 3)

110

2e h TiO (3.2)

The nature of trapped holes has been controversially discussed [7,8]. It has

been generally assumed that adsorbed water could be photo‐oxidized giving

place to surface bounded hydroxyl radicals ( sHO ). However, it has been

reported that this reaction was kinetic and thermodynamically hindered [7] and

the trapping phenomena by terminal oxygen ions of the TiO2 lattice (O ‐22TiO ) has

been proposed, forming terminal protonated or deprotonated radicals

(depending on pH) [9,10]. Regardless of the reaction considered, the formation

of hydroxyl radicals in the TiO2 surface can be described as:

2‐

2 TiO2H O/O

‐s sh HO (O ) (3.3)

In addition, oxygen when present on the particle surface, acts as an electron

acceptor according to reaction (3.4):

•

2 2e O O ‐ (3.4)

Superoxide ion radicals ( •‐2O ) may give place to hydrogen peroxide that can

react with TiO2 electrons and/or additional •‐2O , forming free hydroxyl radicals

through reactions (3.5) and (3.6):

2 2 22 2O O 2H H O O‐ ‐ (3.5)

2 2 2 2e O H O HO HO O (3.6)

In this complex (but simplified) reaction mechanism, free hydroxyl radicals,

•HO , and/or trapped holes ( s sh HO / O ) may be responsible of the non‐

selectively organic matter oxidation and mineralization (reaction (3.7)), being

this process among the most studied AOPs.

2 2HO (h ) R R ʹ HO (h ) ... CO H O (3.7)

When ozone is present, it can react directly and selectively with some organic

compounds (i.e. aromatic and substituted aromatic compounds, molecules with



111

unsaturated bonds e.g. ‐C=C‐, ‐C≡C‐, ‐C=N, ‐C=O, etc.) through different

reaction mechanisms (mainly cycloaddition reactions and/or electrophilic

substitution) included in ‘‘direct ozone reactions’’ terminology (reaction (3.8))

[11].

3 2 2zO R R ʹ H O (3.8)

On the other hand, the presence of radical species (mainly •HO ) coming from

the decomposition of ozone in water gives place to the non‐selectively oxidation

of the organic compounds in water (indirect ozone reactions), process that is

favored in alkaline media [11]:

3 2 2O HO HO O (3.9)

•• ‐3 2 2 3O HO HO O (3.10)

••pKa 4.8 ‐

2 2HO O H (3.11)

• •‐ ‐3 22 3O O O O (3.12)

The generated ozonide radical ( •‐3O ) rapidly reacts with H+ in the solution to

give •

3HO radical, which evolves to give O2 and •HO (reactions (3.13) and (3.14)).

• •‐3 3O H HO (3.13)

•

23HO HO O (3.14)

The benefits of using the combined process, photocatalytic ozonation, are not

only related to the sum of individual processes but also to the fact that the

dissolved ozone can readily react with electrons at the TiO2 surface according to

reaction (3.15) giving place to the ozonide ion radical:

•‐3 3O e O (3.15)

As a consequence, the recombination of electrons and positive holes may be

reduced by this reaction (3.15) and also by reaction (3.6) due to the presence of


112

higher H2O2 concentration in the reaction medium, eventually formed during

direct ozonation reactions (reaction (3.8)) [12‐14]. With this reaction scheme a

synergistic effect between ozone and semiconductor photocatalysis is expected

due to the larger amount of oxidizing species formed (hydroxyl radicals,

bounded hydroxyl radicals and/or positive holes). This has been observed for

the photocatalytic ozonation of several organic compounds in water [2,5,6,15‐

19].

Some of the target compounds subjected to photocatalytic ozonation were

complex molecules that present high direct ozone‐organic rate constants and/or

also complicated reaction pathways involving several steps with different

intermediate species [5,16,19]. This makes difficult to analyze the reaction

mechanism, the species involved in the oxidation steps and/or the synergistic

effect between ozone and photocatalysis. In that sense, it would be interesting to

test the photocatalytic ozonation process using small and refractory to direct

ozone reactions organic compounds.

In this work methanol has been selected due to its known scavenging

character of hydroxyl radicals (kHO = 9.7x108 M‐1 s‐1, [20]) and because it has been

commonly used to test the photocatalytic efficiency of several TiO2 materials

since it also reacts with photogenerated holes [9,10,21‐25]. In addition,

methanol‐ozone direct reaction takes place at slow reaction rate (kO3 = 0.024 M‐1

s‐1) [26]. Therefore, it is expected that under appropriate experimental

conditions, the formaldehyde evolution (formaldehyde is the first product of

methanol oxidation) gives information about the production rate of oxidizing

species (different from O3). The aim of this work was then to evaluate the

production of photo‐generated oxidizing species (hydroxyl radicals and positive

holes) using methanol as target compound comparing both, photocatalytic

oxidation and photocatalytic ozonation, to determine the true quantum yield of

photocatalytic reactions, and to state the synergy degree between ozone and

irradiated semiconductor during photocatalytic ozonation. To our knowledge



113

the determination of the quantum yield of photo‐generated oxidizing species

production in photocatalytic ozonation has not been studied before with any

target compound.

3.2. EXPERIMENTAL SECTION

3.2.1. Experimental set‐up and oxidation/ozonation procedure

Ozonation, photocatalytic oxidation and photocatalytic ozonation

experiments were carried out in a 1 L slurry cylindrical reactor equipped with

magnetic stirring and inlets for measuring temperature, feeding the gas (oxygen

or ozone‐oxygen) through a porous plate situated at the reactor bottom,

sampling port and outlet for the non‐absorbed gas. In photocatalytic

experiments, the reactor was illuminated with 2 black light blue (UVA radiation)

fluorescent lamps (15 W each, from HQPower) placed inside a black box.

The reactor was charged with an aqueous solution containing methanol (2 M,

CH3OH HPLC grade from Panreac) and TiO2 (0.5 g L‐1, Degussa P25, in catalytic

experiments). Initial pH was set to 3 with HClO4, pH = 7 with NaOH or buffered

at pH = 7 with H3PO4 (30 mM) and NaOH (from Panreac). In photocatalytic

experiments, the reactor was then exposed to the radiation (lamps were turned

on 30 min before to stabilize). In ozonation experiments a mixture of ozone‐

oxygen gas (30 L h‐1, 10 ‐ 30 mg L‐1) was also continuously fed to the reactor.

Ozone was generated from pure oxygen in a Sander laboratory ozonator.

Temperature was maintained at 25 °C during the reaction time. Reaction

samples were withdrawn from the reactor at regular intervals for 60 min

reaction time, and then filtered through syringe PET membrane filters

(Chromafil Xtra, 0.20 m). The evolution of the reaction was followed through

the determination of formaldehyde (primary product of methanol oxidation),

hydrogen peroxide, dissolved ozone concentration, ozone concentration in the

outlet gas and pH.

Formaldehyde was determined by the Nash method [27], based on the


114

Hantzsch reaction. In this assay, 2 mL of reagent (0.2 mL of acetylacetone

(Sigma‐Aldrich), 3 mL of acetic acid (Panreac) and 25 g of ammonium acetate

(Fluka) in 100 mL of water) are mixed with 5 mL of the sample and heated for 30

min at 50 °C in the dark [28]. Spectrophotometric measurements were carried

out at 412 nm (= 7890 M‐1 cm‐1) using a Helios‐ Thermo Spectronic

spectrophotometer. Hydrogen peroxide concentration was determined through

the cobalt/bicarbonate method [29], at 260 nm (= 26645 M‐1 cm‐1) using an

Helios‐ spectrophotometer. Dissolved ozone concentration was measured by

following the method proposed by Bader and Hoigné [30] based on the

discoloration of a 5,5,7 indigotrisulphonate solution (= 600 nm, Helios‐

spectrophotometer, = 20000 M‐1 cm‐1). Ozone in the gas phase was monitored

by means of an Anseros Ozomat ozone analyzer, based on the absorbance at 254

nm.

3.2.2. Photon fluxes determination

Ferrioxalate actinometry [31] was used to determine the incident photon flux,

I0, in the photoreactor, that was found to be 2.66x10‐5 einstein min‐1. In these

experiments the Fe(II) concentration was followed by the ‐fenantroline method

[32] using an Helios‐ spectrophotometer at 510 nm (= 11023 M‐1 cm‐1). The

photon flux absorbed by the catalyst, Ia, was estimated through the

determination of the quantum yield of formaldehyde generation in the

photoreactor. This was obtained by applying the protocol of Serpone and

Salinaro [33] for the photocatalytic oxidation of methanol, analyzing the

formaldehyde evolution under the following operating conditions: methanol

concentration from 0.5 to 2 M, TiO2 concentration from 0.025 to 3 g L‐1, pH0 = 7,

pH = 7 (buffered) and pH0 = 3 (adjusted with NaOH, phosphate buffered or

adjusted with HClO4, respectively) and oxygen saturated solution with 30 L h‐1

gas flow rate. The evolution of the reaction was followed as explained in the

previous section. For comparative purposes, additional photocatalytic oxidation

experiments were carried out using formic acid (0.01 M CH2O2) as target



115

compound (0.5 g L‐1 TiO2, pH0 = 3). Formic acid was determined following the

evolution of the CO2 formed by measuring the total organic carbon (TOC)

remaining in the solution using a TOC‐VSCH analyzer from Shimadzu.

3.3. RESULTS AND DISCUSSION

3.3.1. Absorbed photon flux

The determination of the absorbed photon flux by the catalyst is a key issue

in heterogeneous photocatalytic reactions that serves to calculate the quantum

yield of the reaction according to the following equation:

ii

a

rreacted molecules mol i

absorbed photons I einstein

(3.16)

The radiant energy used in a photocatalytic reaction (absorbed, Ia) is

generally lower than that impinging on the reacting system (I0) due to the light

scattered or reflected by the suspended catalyst in the dispersion. The absorbed

light intensity has been usually calculated by applying the radiation transfer

equation (RTE) to the reaction system [34‐38] or experimentally determined by

means of the measurement of the light transmission through a photocatalyst

suspension [39].

In this work, we have indirectly calculated the absorbed photon flux by using

the protocol of Serpone and Salinaro [33] to determine the quantum yield of the

methanol oxidation reaction (formaldehyde formation) under our experimental

conditions.

Methanol photocatalytic oxidation gives place to formaldehyde as primary

oxidation product according to reactions (3.17) and (3.18):

3 2 2CH OH h HO CH OH H H O (3.17)

• •

2 2 2 2 2 22CH OH O O CH OH CH O H O (HO ) (3.18)


116

where either holes (h+) and free hydroxyl radicals ( •HO ) may participate,

forming also superoxide ion radical ( •‐2O ) or hydroperoxide radical ( •

2HO , acidic

pH). Figure 3.1 shows the time‐evolution of CH2O formation under different

experimental conditions. It was nearly linear throughout the range of conditions

used. The rate of CH2O formation reached a plateau around 2 M methanol, as

can be seen in Figure 3.2. This concentration was then used for the experiments

with different TiO2 concentration. At this CH3OH concentration, formaldehyde

is not expected to be significantly oxidized since oxidizing species will mainly

react with the former. In fact, taking into account the kinetic constants of

methanol and formaldehyde with hydroxyl radicals (9.7x108 and 1x109 M‐1 s‐1

[20], respectively) the formaldehyde formation rate with 2 M methanol was 1940

times higher than formaldehyde oxidation rate (using a generic concentration

about 10‐3 M similar to CH2O concentrations observed in this work) proving that

formaldehyde oxidation could be negligible in the overall reaction rate. The

calculated reaction rates of formaldehyde formation at different TiO2 loading

defines a plateau from 0.5 g L‐1 as depicted in Figure 3.3(A). This behavior can be

well described through the following equation:

2 2

2

lim TiO CH O

lim TiO 0

a C r

a C I

(3.19)

where photonic efficiencies are calculated employing the relationship

= rCH2O/I0. The quantum yield of formaldehyde formation (CH2O) was

determined from the limiting photonic efficiency (lim) at high TiO2 loadings. At

these conditions it has been reported that lim = [33]. Figure 3.3(B) shows fitting

results of linearized Eq. (3.19). The quantum yield calculated for formaldehyde

formation at pH0 = 7 (non‐buffered) was CH2O‐pH7 = 0.48 mol einstein‐1. The pH

value did not change significantly (pHf = 6.7). The corresponding absorbed light

intensity at different TiO2 loadings are summarized in Table 3.1 together with

the integrated absorption fraction over the wavelength used here, FS = Ia/I0. From



117

0.5 g L‐1 TiO2 onward, the P25 catalyst presented an absorption fraction higher

than 90 %. On the other hand, in experiments carried out at pH = 7 (phosphate

buffered solution), the reaction rate of formaldehyde formation was lower than

in the previous case at the same operating conditions (catalyst loading, methanol

concentration) as can also be observed in Fig. 3.1 and Table 3.1. This can be due

to phosphate adsorption onto the titania surface [40,41], that can compete for

adsorption sites with methanol and/or react with oxidizing species formed at the

catalyst surface (e.g. with hydroxyl radicals, kHO‐PO43‐ = 1x107 M‐1 s‐1; kHO‐HPO42‐ =

1.5x105 M‐1 s‐1; kHO‐H2PO4‐ = 2x104 M‐1 s‐1; kHO‐H3PO4 = 2.7x106 M‐1 s‐1 [20]), although

this second point could be disregarded due to the great excess of methanol (2 M

CH3OH against 30 mM H3PO4). Since phosphates do not absorb the black light

used, it is expected that the absorbed light intensity, Ia, do not change

significantly compared to the non‐buffered experiments at pH = 7.

0 10 20 30 40 50 600.0

2.0x10-4

4.0x10-4

6.0x10-4

8.0x10-4 2 M 0.025 g L-1 7

2 M 0.5 g L-1 7

0.1 M 0.5 g L-1 7

2 M 0.5 g L-1 7 buffered

2 M 0.5 g L-1 3

CC

H2O

(m

ol L

-1)

TIME (min)

CH3OH TiO2 pH

Figure 3.1. Time‐evolution of formaldehyde formation during methanol photocatalytic

oxidation experiments. Conditions: T = 25 °C, Qg = 30 L h‐1 (O2).


118

0 1 2 3 40.0

3.0x10-6

6.0x10-6

9.0x10-6

1.2x10-5

1.5x10-5

r CH

2O (

mol

L-1

min

-1)

CCH3OH (mol L-1)

Figure 3.2. Rate of formaldehyde formation vs. methanol concentration in photocatalytic

oxidation experiments. Conditions: pH = 7, T = 25 °C, CTiO2 = 2 g L‐1, Qg = 30 L h‐1 (O2 or

O3/O2).

Figure 3.3. (A): Rate of formaldehyde formation vs. TiO2 loading in methanol

photocatalytic oxidation experiments. (B): Fitting results of linearized Eq. (3.19).

Conditions: pH = 7 and 3, T = 25 °C, Qg = 30 L h‐1 (O2 or O3/O2).

0.0 0.5 1.0 1.5 2.0 2.5 3.00.0

3.0x10-6

6.0x10-6

9.0x10-6

1.2x10-5

pH=7 pH=3

r CH

2O

(m

ol L

-1 m

in-1

)

CTiO2 (g L-1)

0 10 20 30 402

4

6

8

10

1/

(ei

nste

in m

ol-1)

1/CTiO2 (L g-1)

(A)

(B)



119

Table 3.1. Rate and photonic efficiency of CH2O formation and absorbed and fraction

light intensity.

CTiO2

(g L‐1) pH0

r CH2O /rCO2

(M min‐1)

ξ

(mol einstein‐1)

Ia

(einstein L‐1 min‐1) Fs

0.025 7 6.05x10‐6 0.23 1.27x10‐5 0.47

0.05 7 8.29x10‐6 0.31 1.73x10‐5 0.65

0.1 7 1.07x10‐5 0.40 2.24x10‐5 0.84

0.3 7 1.16x10‐5 0.44 2.43x10‐5 0.91

0.5 7 1.22x10‐5 0.46 2.55x10‐5 0.96

1 7 1.23x10‐5 0.46 2.57x10‐5 0.97

2 7 1.34x10‐5 0.50 2.80x10‐5 ~1.0

3 7 1.17x10‐5 0.44 2.45x10‐5 0.92

0.5 7a 7.51x10‐6 0.28 2.55x10‐5 0.96

0.025 3 2.65x10‐6 0.10 7.81x10‐6 0.29

0.05 3 4.01x10‐6 0.15 1.18x10‐5 0.44

0.1 3 4.62x10‐6 0.20 1.56x10‐5 0.59

0.5 3 8.77x10‐6 0.33 2.60x10‐5 0.97

1 3 8.82x10‐6 0.33 2.61x10‐5 0.98

0.5 3b 6.76x10‐6 0.26 2.60x10‐5 0.97

aBuffered solution (H3PO4 30 mM), bExperiment with CH2O2 (10 mM)

On the other hand, it is known that the size of the TiO2 aggregates in aqueous

solution depends on the pH value and therefore, could affect the intensity of

light absorbed by the photocatalyst [2]. Thus, additional photocatalytic

oxidation experiments with 0.025 ‐ 1 g L‐1 TiO2 were carried out at pH0 = 3 to

calculate both, the quantum yield (CH2O‐pH3) and the absorbed photon flux at this

pH value. Firstly, the pH value did not significantly change during the reaction

time. It was observed that the rate of formaldehyde formation at pH0 = 3 was

quite lower than at pH0 = 7 (non‐buffered) at the same catalyst loading (see Fig.

3.1). These results have also been plotted in Fig. 3.3(A) whereas Fig. 3.3(B) shows

fitting results of linearized Eq. (3.19) for experiments at pH0 = 3. The quantum


120

yield calculated was CH2O‐pH3 = 0.34 mol einstein‐1. The absorbed light intensity

calculated was somewhat higher than at pH = 7 in experiments at the same

catalyst loading (Table 3.1), according to the lower aggregates size expected at

lower pH. The lower reaction rates and quantum yield observed could be

attributed to the stabilization of the 2 2O CH OH radical at acidic pH [42] and/or

to the presence of ClO4‐ ions (HClO4 used to set pH = 3) that can be absorbed

onto the catalyst surface [43]. Previously, Du and Rabani [42] showed that

quantum yields of formaldehyde formation during photocatalytic oxidation of

methanol with a TiO2 catalyst fell down from 0.2 to 0.05 mol einstein‐1 at

pH = 7 and pH = 3, respectively, whereas when comparing the quantum yield of

CO2 formation during photocatalytic oxidation of formic acid at pH = 3, the

value was quite similar to that found for the methanol system at pH = 7. This

was explained on the basis of the stabilization of the 2 2O CH OH radical at acidic

pH when using methanol. However, our experiments carried out at pH = 3 with

formic acid 0.01 M (high enough to reach a constant reaction rate independent of

formic acid concentration [42]) led to the results summarized in Table 3.1, where

the photonic efficiency calculated for CO2 formation was 0.26 mol einstein‐1,

quite similar (even lower) than the value obtained in the methanol system.

These results seem to be pointed out that the stabilization of the 2 2O CH OH

radical at pH = 3 is not the main responsible of the decrease observed in the

quantum yield at this pH value under the operating conditions used here and,

therefore, we have used also methanol at pH = 3 as model compound.

Finally, under the same experimental conditions, Ia is not expected to

significantly change in presence of ozone during photocatalytic ozonation

experiments since O3 does not absorb black light radiation (= 365 nm). The

calculated Ia values were used to determine the quantum yields for

photocatalytic ozonation processes for comparative purposes.



121

3.3.2. Photocatalytic oxidation versus photocatalytic ozonation

To compare photocatalytic oxidation behavior versus photocatalytic

ozonation, similar experiments were carried out at the same operating

conditions of methanol concentration, gas flow rate, radiation and pH, but

feeding ozone to the system. At 2 M methanol concentration used here, also in

ozonation and photocatalytic ozonation experiments, formaldehyde ozonation is

not expected. According to the kinetic constants of methanol and formaldehyde

with O3 (0.024 and 0.10 M‐1 s‐1 [26], respectively) and the concentration of both

compounds, the formaldehyde formation rate with 2 M methanol was almost

500 times higher than the formaldehyde ozonation (using a generic

concentration about 10‐3 M similar to CH2O concentrations observed in this

work) proving that formaldehyde ozonation could be neglected in the overall

reaction rate.

The comparison of formaldehyde formation during photocatalytic oxidation

(TiO2/O2/UVA) and photocatalytic ozonation (TiO2/O3/UVA) of methanol is

presented in Figure 3.4(A). Also, for comparative purposes single ozonation (O3)

and photolytic ozonation results (i.e. irradiated O3 without TiO2, O3/UVA) have

been plotted. These experiments were carried out at pH = 3 to minimize indirect

ozone reactions. This pH value did not significantly change throughout the

reaction time. As expected, ozonation and photolytic ozonation gave place to

similar formaldehyde evolution since ozone does not absorb black light

radiation. On the other hand, the results observed for photocatalytic oxidation

are also quite similar to the ozone process. The highest rate of formaldehyde

formation was found during photocatalytic ozonation. Reaction rates were

calculated through the slope of the CH2O concentration‐time evolution and are

displayed in Table 3.2. A synergistic effect between ozone and the irradiated

TiO2 can be observed during TiO2/O3/UVA treatment where the reaction rate is

almost twice higher than the sum of the reaction rates of individual process

(2.91x10‐5 M min‐1 versus 1.89x10‐5 M min‐1, respectively).


122

Figure 3.4. Time‐evolution of formaldehyde (A) and hydrogen peroxide (B) concentration

during ozonation and photocatalytic oxidation/ozonation experiments. Conditions: pH =

3, T = 25 °C, CTiO2 = 0.5 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 30 L h‐1 (O2 or O3/O2).

The accumulation of important concentrations of hydrogen peroxide was

also observed during ozonation experiments. Results are plotted in Figure

3.4(B). As can be seen, during photocatalytic oxidation of methanol small

amounts of hydrogen peroxide could be detected whereas during single

ozonation and photolytic ozonation, the evolution of hydrogen peroxide is

comparable to the formaldehyde formation. This is explained on the basis of

ozone‐methanol direct reaction which gives place to hydrogen peroxide [44]

(B)

(A)

0 10 20 30 40 50 600.0

1.0x10-4

2.0x10-4

3.0x10-4

4.0x10-4

5.0x10-4

TIME (min)

CH

2O

2 (

mol

L-1

)

Photocatalytic oxidation Ozonation Photolytic ozonation Photocatalytic ozonation

0.0

4.0x10-4

8.0x10-4

1.2x10-3

1.6x10-3

CC

H2O

(mo

l L-1

)



123

according to:

3 3 2 2 2O CH OH CH O H O (3.20)

It is noticeable that hydrogen peroxide concentration was much lower in the

photocatalytic ozonation experiment likely due to the consumption of H2O2

through reaction (3.6) (H2O2‐eaq‐ reaction: k = 1.1x1010 M‐1 s‐1 [20]). This reaction

would also enhance the generation of hydroxyl radicals improving methanol

oxidation.

In addition, dissolved ozone was measured during all the ozonation

experiments. The concentration was nearly constant at 7.5x10‐6 M and 4.5x10‐6 M

during ozonation and photocatalytic ozonation, respectively. Also the ozone

concentration in the gas phase in the reactor outlet was lower during

photocatalytic ozonation (5.0 mg L‐1) than during ozonation (6.5 mg L‐1). The

lower values found during photocatalytic ozonation suggests that O3 is also

being consumed through reaction (3.15) (O3‐eaq‐ reaction: k = 3.6x1010 M‐1 s‐1,

which is also higher than the one of O2‐eaq‐ reaction: k = 1.9x1010 M‐1 s‐1 [20]),

improving methanol oxidation rate due to the generation of additional hydroxyl

radicals.

Therefore, taking into account the slow rate of direct methanol‐ozone

reaction and the negligible contribution of ozone decomposition at the pH value

used here, the synergism observed may be related to the reaction of dissolved

ozone and hydrogen peroxide as electron acceptors to produce HO radicals,

thus avoiding to some extent the recombination reactions on the TiO2 catalyst.


124

Tabl

e 3.

2. E

xper

imen

tal c

ondi

tions

, rea

ctio

n ra

te c

ontr

ibut

ions

and

qua

ntum

yie

ld o

f the

pho

toca

taly

tic re

actio

ns.

Proc

ess

Irra

diat

ion

CTi

O2

(g L

-1)

CO

3,g

inle

t (m

g L-1

) pH

rC

H2O

(M m

in-1

) rO

3b

(M m

in-1

) rH

O-O

3c

(M m

in-1

) rh

νd

(M m

in-1

) Eh

v

(%)

h ν

(mol

ein

stei

n-1)

Phot

ocat

alyt

ic

oxid

atio

n (T

iO2/O

2/UV

A)

on

0.5

0 3

8.77

x10-6

---

---

8.

77x1

0-6

0 0.

34

on

0.5

0 7

1.22

x10-5

---

---

1.

22x1

0-5

0 0.

48

on

0.5

0 7a

7.51

x10-6

---

---

7.

51x1

0-6

0 0.

29

Ozo

natio

n (O

3)

off

0 10

3

1.02

x10-5

1.

02x1

0-5

0 ---

---

---

off

0 20

3

1.55

x10-5

1.

55x1

0-5

0 ---

---

---

off

0 30

3

1.97

x10-5

1.

97x1

0-5

0 ---

---

---

off

0 30

7a

1.55

x10-4

1.

97x1

0-5

1.35

x10-4

---

---

---

Phot

ocat

alyt

ic

ozon

atio

n (T

iO2/O

3/UV

A)

on

0.5

10

3 2.

91x1

0-5

1.02

x10-5

0

1.89

x10-5

53

0.

73

on

0.5

20

3 3.

63x1

0-5

1.55

x10-5

0

2.08

x10-5

57

0.

80

on

0.5

30

3 3.

91x1

0-5

1.97

x10-5

0

1.94

x10-5

55

0.

75

on

0.5

30

7a 2.

38x1

0-4

1.97

x10-5

1.

35x1

0-4

8.36

x10-5

91

3.

27

a Buf

fere

d so

lutio

n (H

3PO

4 30

mM

), b I

t coi

ncid

es w

ith rC

H2O

for s

ingl

e oz

onat

ion

expe

rim

ents

at p

H =

3, c C

alcu

late

d fr

om o

zona

tion

expe

rim

ents

by

mea

ns o

f the

diff

eren

ce b

etw

een

rCH

2O-r

O3,

d Cal

cula

ted

by m

eans

of t

he d

iffer

ence

bet

wee

n rC

H2O

- rO

3-rH

O-O

3.



125

3.3.2.1. Influence of ozone concentration

The influence of ozone concentration in the feeding gas during photocatalytic

ozonation of methanol was studied at pH = 3. Results of formaldehyde and

hydrogen peroxide formation with time are depicted in Figure 3.5(A) and Figure

3.5(B), respectively. Also, for the sake of comparison, single ozonation results

varying the ozone gas concentration have been plotted. It can be observed that

ozone concentration exerts a positive effect on methanol oxidation rate

(formaldehyde formation), compared to the ozone free experiments (see also

Fig. 3.4(A)) both in single ozonation and photocatalytic ozonation, although the

effect is less important when increasing the ozone gas concentration. Reaction

rates of formaldehyde formation have been calculated and are shown in Table

3.2. The increase of formaldehyde formation rate is proportional to the

increasing ozone concentration in single ozonation experiments whereas it is

more important from 10 to 20 mg L‐1 of ozone than from 20 to 30 mg L‐1 in

photocatalytic ozonation. This behavior has been observed before during the

photocatalytic ozonation process [19] and has been attributed to complex

Langmuir kinetics for substances that absorb and react on the semiconductor

surface.

Regarding the evolution of H2O2 during the experiments at different O3

concentration (Fig. 3.5(B)), it can be observed that H2O2, formed mainly from

direct ozonation of methanol, is further accumulated during ozonation

treatment while is not in the photocatalytic ozonation process and regardless of

the ozone concentration. The latter is likely due to the consumption of H2O2 in

the photocatalytic process through reaction (3.6). Despite this, a positive effect of

ozone concentration is observed on the evolution of H2O2 in both treatments,

more pronounced during ozonation.


126


during methanol ozonation and photocatalytic ozonation experiments. Conditions: pH =

3, T = 25 °C, CTiO2 = 0.5 g L‐1, CO3,g inlet = 10, 20, 30 mg L‐1, Qg = 30 L h‐1 (O3/O2).

3.3.2.2. Influence of pH

Previous experiments have been carried out at pH = 3 to minimize ozone

decomposition reaction and, thus, avoiding indirect methanol‐ozone oxidation

reaction. However, pH plays a crucial role during ozonation processes. Results

of photocatalytic oxidation, ozonation and photocatalytic ozonation of CH3OH

at pH = 3 and pH = 7 (buffered solution) are compared in Figure 3.6. Regarding

0 10 20 30 40 50 600.0

2.0x10-4

4.0x10-4

6.0x10-4

8.0x10-4

1.0x10-3

1.2x10-3

TIME (min)

CH

2O2

(mo

l L-1

)

0.0

4.0x10-4

8.0x10-4

1.2x10-3

1.6x10-3

2.0x10-3

s-1

s-1

s-1

10 mgL-1

20 mgL-1

30 mgL-1

CC

H2O

(m

ol L

-1)

Ozonation Photocatalytic ozonation

(A)

(B)



127

CH2O formation, slight differences were observed in Fig. 3.6(A) for

photocatalytic oxidation reaction at pH = 3 and pH = 7 as commented in the

previous section. Nevertheless, when ozone is present, a large increase in the

rate of CH2O formation is observed at pH = 7 respect to pH = 3. During single

ozonation experiments the formaldehyde formation rate increased around 8

times (from 1.97x10‐5 to 1.55x10‐4 M min‐1, see Table 3.2). Direct ozone‐methanol

reaction is not affected by pH modifications since CH3OH does not dissociate at

the studied pH range [26]; as a consequence, this behavior is related to ozone

decomposition reactions (3.9) ‐ (3.14) that are produced at neutral pH to

generate HO radicals in the reaction medium. In this line, lower dissolved

ozone concentration has been detected at pH = 7 (8x10‐7 M and 1x10‐6 M in

ozonation and photocatalytic ozonation, respectively) compared to pH = 3

(8.2x10‐6 M and 6.1x10‐6 M in ozonation and photocatalytic ozonation,

respectively). Also, ozone concentration in the gas phase at the reactor outlet has

been observed to decrease with pH increase from 3 to 7 (from 15 at pH 3 to 10

mg L‐1 at pH 7 in ozonation, and from 11 at pH 3 to 6 mg L‐1 at pH 7 in

photocatalytic ozonation runs). This corroborates a higher consumption of

ozone at neutral pH likely due to its faster decomposition.

In addition, hydrogen peroxide formed upon direct ozone‐methanol reaction

can be dissociated (reaction (3.21), pKa = 11.3), promoting reaction (3.10) to some

extent:

pKa 11.3

2 2 2H O HO H (3.21)

In fact, hydrogen peroxide concentrations detected at pH = 7 were fairly

lower than at pH = 3 (see Fig. 3.6(B)), indicating that H2O2 is consumed during

the single ozonation process. Another fact that can affect the ozonation

mechanism due to the pH value is that the hydroperoxide radical formed upon

methanol ‐ HO oxidation ( •

2HO ) will be on its dissociated form ( •‐2O ) at neutral

pH (reaction (3.11), pKa = 4.8), then it also participates in the reaction

mechanism of ozone decomposition.


128

0 10 20 30 40 50 600.0

2.0x10-4

4.0x10-4

6.0x10-4

8.0x10-4

1.0x10-3

1.2x10-3

TIME (min)

CH

2O2

(mo

l L-1

)

0.0

3.0x10-3

6.0x10-3

9.0x10-3

1.2x10-2

1.5x10-2

Photocatalytic oxidation Ozonation Photocatalytic ozonation

CC

H2O

(m

ol L

-1)

pH=7 pH=3


during methanol ozonation and photocatalytic oxidation/ozonation experiments.

Conditions: pH = 3, 7, T = 25 °C, CTiO2 = 0.5 g L‐1, CO3,g inlet = 30 mg L‐1, Qg = 30 L h‐1 (O2 or

O3/O2).

The increase of formaldehyde formation rate observed was more significant

in photocatalytic ozonation due to the formation of higher concentration of

oxidizing species at the photocatalyst surface according to the reaction scheme

described above. At pH = 7 and 30 mg L‐1 of ozone inlet concentration, a

synergistic effect is clearly observed in the reaction rate (see Table 3.2), being the

sum of ozone and photocatalytic oxidation rates 1.62x10‐4 M min‐1 fairly lower

than that of photocatalytic ozonation (2.38x10‐4 M min‐1). Therefore, an

(A)

(B)



129

increment in the pH value exerts a positive effect on the photocatalytic

ozonation reaction due to the formation of species that favor the ozone

decomposition in water.

3.3.2.3. Synergistic effect and quantum yield of photocatalytic induced reactions

From the previous experimental results, different contributions to the rate of

formaldehyde formation have been calculated. On one hand, when single

ozonation experiments are carried out at pH = 3, the contribution of O3 indirect

reactions (rHO‐O3) can be neglected and therefore, the observed reaction rate at

this pH would only be due to the direct ozone‐methanol reaction (rO3). This was

extrapolated to photocatalytic ozonation considering the same contribution of

ozone‐methanol direct reaction than in the absence of light. Additionally, for

photocatalytic treatments, another contribution needs to be considered due to

the photogenerated oxidizing species (i.e. h+ and HO from reactions (3.1) ‐ (3.6),

and (3.15)) apart from the direct or indirect ozonation mechanism. This

contribution, named rhv, is the observed rate of CH2O formation in

photocatalytic oxidation, and was calculated as the difference between the

observed reaction rate rCH2O and the direct‐indirect ozone reactions in

photocatalytic ozonation. On the other hand, when reactions were carried out at

pH = 7, the contribution of indirect ozone reactions become very important. In

this case, this contribution was calculated by subtracting the ozone‐direct

reaction rate (calculated at pH = 3 at the same operating conditions) to the

observed reaction rate of formaldehyde formation during single ozonation

experiments. Also, these results were extrapolated to photocatalytic ozonation at

pH = 7. The different contributions to the global reaction rate of formaldehyde

formation during methanol oxidation have been summarized in Table 3.2.

Regarding the photocatalytic contribution, it is noticeable the increase

underwent in the light induced reaction rate when ozone was present compared

to the one of photocatalytic oxidation at the same operating conditions. The

degree of enhancement produced as a consequence of O3 was calculated as

follows:


130

h 2 3 h 2 2

h 2 2

r TiO / O / UVA r TiO / O / UVAE ∙100 %

r TiO / O / UVA

(3.22)

Table 3.2 shows the values obtained for all the photocatalytic ozonation

experiments. The enhancement observed in the photocatalytic contribution

increased from 53 % to 57 % when increasing ozone concentration in the feeding

effluent at pH = 3, although slight differences were observed when using 20 or

30 mg L‐1 O3 as commented in the previous section. The highest enhancement

was observed at pH = 7 with 30 mg L‐1 of ozone. This parameter can be

considered as a real evidence of the synergism produced between ozone and

irradiated TiO2.

At the experimental conditions used (high excess of CH3OH), the reaction

rate calculated for the photocatalytic contribution (rhv) can be well considered as

the reaction rate of oxidizing species formation due to light induced reactions.

Therefore, the quantum yield of these reactions (light induced ones) can be

calculated taking into account the absorbed light intensity at the operating

conditions used (see Table 3.1). The calculated quantum yields (hv) for

photocatalytic oxidation and photocatalytic ozonation processes are presented in

Table 3.2. The quantum yield of photogenerated species formation in

photocatalytic ozonation at pH = 3 was found to be increased twice and near

triple times the value observed for photocatalytic oxidation as ozone

concentration increased. Quantum yield reached a plateau at high ozone

concentrations, as can also be observed in Figure 3.7 where the evolution of the

quantum yield with ozone concentration has been depicted. This trend suggests

that a limiting reaction rate has been reached in the system due to the reaction of

hydroxyl radicals with O3. Sun and Bolton [21] also observed this behavior while

studying the photocatalytic oxidation of methanol with the addition of

hydrogen peroxide as an electron acceptor. On the other hand, the highest

improvement of the quantum yield has been observed in the experiment of

photocatalytic ozonation at pH = 7 reaching a value near 3 photogenerated

oxidizing species per absorbed photon.



131

0 5 10 15 20 25 300.0

0.2

0.4

0.6

0.8

1.0

hv

(m

ol e

inst

ein

-1)

CO3,g inlet (mg L-1)

Figure 3.7. Evolution of quantum yield of oxidizing species formation through light

induced reactions with ozone inlet concentration. Conditions: pH = 3, T = 25 °C, CTiO2 =

0.5 g L‐1, Qg = 30 L h‐1 (O2 or O3/O2).

The maximum quantum yield of formaldehyde formation expected during

photocatalytic oxidation is CH2O,max = hv,max = 1.5 mol einstein‐1, according to the

reaction mechanism proposed where 2 photons are needed to produce 3

oxidizing species (i.e. holes and/or hydroxyl radicals) taking into account also

the radical •‐2O / •

2HO formed from CH3OH oxidation [6]. This maximum

becomes even greater (hv;max‐O3 = 2 ‐ 3) when ozone is present depending on pH,

due to reactions (3.13) ‐ (3.15) where ozone needs 1 e‐ per HO formed, and the

generation of H2O2 from direct methanol ozonation (reaction (3.20)), that acts as

an electron acceptor (reaction (3.6)) needs also 1 e‐ per HO formed. The

maximum quantum yield at basic‐neutral pH when O3 is present has not been

accurately calculated since ozone‐indirect reactions are closely related to the

reactions involving photogenerated species. Thus, the calculated values seem to

be reasonable although they should be taken with caution, especially that

calculated at pH = 7 where the contribution of the indirect‐ozone reaction has

been taken as the same than in single ozonation and may be underestimated.


132

Further work will be carried out to quantitatively calculate the different

contributions through a detailed reaction mechanism for the AOPs studied

(photocatalytic oxidation, ozonation and photocatalytic ozonation). Despite this

it is clear that ozone exerts a positive effect on the photocatalytic induced

reactions (not only O3 direct and indirect reactions) due to two main factors: (i)

ozone acts as an electron acceptor, and (ii) hydrogen peroxide usually formed

during direct ozonation reactions also acts as an electron acceptor, both leading

to higher amount of hydroxyl radicals per photon absorbed and also reducing

the recombination process to some extent.

3.3.2.4. Simplified economic considerations

In addition to the benefits observed with the combined treatment, i.e.

photocatalytic ozonation (TiO2/O3/UVA), in the reaction rate of formaldehyde

formation, the synergistic effect between ozone and irradiated TiO2 and the

highest quantum yield of photo‐generated oxidizing species production, one

important issue in the application of combined treatments is the economy of the

process. Accordingly, to compare the different systems, costs associated with the

normal operation have been evaluated by taking into account the oxygen and

electrical energy consumed to generate the oxidants and/or radiation and the

amount of formaldehyde formed. It has to be highlighted that this simplified

economic study aims only the comparison of the systems (i.e. investments, loss

of TiO2 activity and separation, and other factors are not considered) not

providing an actual costs estimation.

According to Rivas et al. [45] for the same ozone generator, the dependence

between O3 production and current consumption can be estimated as:

‐13O % of max. flow rate in gL 186 ∙current consumption in A (3.23)

with maximum mass flow rate from oxygen 12 g h‐1. From the experimental

conditions for 10 and 30 mg L‐1 O3 at 30 L h‐1 those are 2.5 % and 7.5 % of the

maximum amount produced, respectively. In photocatalytic experiments it is



133

also necessary to take into account the lamps energy consumption (2 lamps with

an input power of 15 W). The cost associated with electricity has been

considered 0.14 € (kWh)‐1 according to the local supplier. In addition, in all the

experiments oxygen from cylinders was used which has a cost of 0.262 € h‐1 [45]

at a flow rate of 30 L h‐1.

With the above data, Table 3.3 shows the results obtained to produce 0.001

mol of formaldehyde from methanol oxidation taking into account the time

needed. It can be observed that the main contribution (of those considered) is

the oxygen consumption. It has to be pointed out that this contribution would be

far different if air from atmosphere was used both in photocatalytic oxidation

and ozonation runs (in the latter two using an air compressor) since there are

ozone generators able to work with air. The operation costs (as calculated here)

are always higher in photocatalytic experiments or ozonation alone than in the

combined process, especially when working at pH = 7, where the reaction time is

highly reduced. Therefore, photocatalytic ozonation treatment is not only

attractive from the point of view of their performance in terms of reaction rates

but also in terms of economic considerations, where the synergism between

ozone and TiO2 photocatalysis is also clear. However, it has to be emphasized

that this simplified economic study could give different results in every

particular case taking into account the necessities of effluent mineralization,

discharge limits required, investment and replacement costs, etc. as reported

before [45] where ozone treatments involved higher costs at the same

mineralization degree obtained.


134

Tabl

e 3.

3. E

stim

ated

cos

ts fo

r the

tran

sfor

mat

ion

of 1

0-3 m

ol C

H3O

H (C

H2O

form

ed).

Proc

ess

pH

CO

3g in

let

(mg

L-1)

Tim

e us

ed

(h)

EO3

(kW

h)

E UV

A

(kW

h)

Elec

tric

ity

(€)

O2

(€)

Cos

t/mm

ol

conv

erte

d (€

mm

ol-1

) Ph

otoc

atal

ytic

ox

idat

ion

(TiO

2/O2 /U

VA

)

3 0

1.90

0

5.70

x10-2

7.

98x1

0-3

4.97

x10-1

5.

05x1

0-1

7 0

2.22

0

6.66

x10-2

9.

32x1

0-3

5.81

x10-1

5.

90x1

0-1

Ozo

natio

n (O

3 )

3 10

1.

63

4.83

x10-3

0

6.76

x10-4

4.

27x1

0-1

4.28

x10-1

7 30

0.

11

9.54

x10-4

0

1.34

x10-4

2.

80x1

0-2

2.80

x10-2

Phot

ocat

alyt

ic

ozon

atio

n (T

iO2/O

3 /UV

A)

3 10

0.

57

1.69

x10-3

1.

72x1

0-2

2.64

x10-3

1.

50x1

0-1

1.52

x10-1

7 30

0.

07

6.21

x10-4

2.

10x1

0-3

3.81

x10-4

1.

80x1

0-2

1.90

x10-2



135

3.4. CONCLUSIONS

Major conclusions reached in this study are:

Absorbed light intensity during photocatalytic oxidation of methanol was

indirectly calculated at different pH values and was always higher than 90 %

of that impinging the reactor for the catalyst loading used (0.5 g L‐1 of TiO2

P25).

The presence of ozone during photocatalytic ozonation at pH = 3 exerts a

positive effect on the reaction rate of formaldehyde formation respect to

photocatalytic oxidation. This was not only due to the direct ozone‐methanol

reaction, but also due to the reaction of dissolved ozone and hydrogen

peroxide (formed upon CH3OH ozonation), electron acceptors to produce

higher concentration of hydroxyl radicals.

An increase in ozone concentration exerts a positive but low effect during

photocatalytic ozonation likely due to self‐scavenging reactions between

HO and ozone or hydrogen peroxide.

The quantum yield calculated for photo‐generated oxidizing species during

photocatalytic oxidation at pH = 3 was hv = 0.34 mol einstein‐1. This

parameter was increased to around hv = 0.80 mol einstein‐1 when combining

TiO2/O3/UVA. This enhancement was even higher at pH = 7 from hv = 0.29 to

hv = 3.27 mol einstein‐1. Therefore, a clear synergism between ozone and

black light photocatalytic oxidation with TiO2 is pointed out due to the

decrease of the recombination process. The reaction of dissolved ozone and

hydrogen peroxide with photo‐generated electrons is likely the reason of the

reaction rate enhancement.

Simplified economic evaluation taking into account only normal operation

has shown that photocatalytic ozonation could be a cost‐effective treatment

depending on the actual necessities of the effluent to be treated.


136

AKNOWLEDGEMENTS

This work has been supported by the Spanish Ministerio de Ciencia e

Innovación (MICINN) and European Feder Funds through the Project CTQ2009‐

13459‐C05‐05/PPQ. A. Rey thanks the University of Extremadura for a post‐

doctoral research contract.

REFERENCES

[1] Kabra, K.; Chaudhary, R.; Sawhney, R.L. “Treatment of hazardous organic

and inorganic compounds through aqueous‐phase photocatalysis: A review”.

Ind. Eng. Chem. Res. 43 (2004) 7683‐7696.




[3] Chong, M.N.; Jin, B.; Chow, C.W.K.; Saint, C. “Recent developments in

photocatalytic water treatment technology: A review”. Water Res. 44 (2010)

2997‐3027.

[4] Adán, C.; Bahamonde, A.; Fernández‐García, M.; Martínez‐Arias, A.

“Structure and activity of nanosized iron‐doped anatase TiO2 catalysts for

phenol photocatalytic degradation”. Appl. Catal. B Environ. 72 (2007) 11‐17.




[6] Augugliaro, V.; Litter, M.; Palmisano, L.; Soria, J. “The combination of

heterogeneous photocatalysis with chemical and physical operations: a tool for


Photochem. Rev. 7 (2006) 127‐144.

[7] Salvador, P. “On the nature of photogenerated radical species active in the

oxidative degradation of dissolved pollutants with TiO2 aqueous suspensions: a

revision in the light of the electronic structure of adsorbed water”. J. Phys.

Chem. C 111 (2007) 17038‐17043.

[8] Henderson, M.A. “A surface perspective on TiO2 photocatalysis”. Surface Sci.

Rep. 66 (2011) 185‐297.



137

[9] Goldstein, S.; Behar, D.; Rabani, J. “Mechanism of visible light photocatalytic

oxidation of methanol in aerated aqueous suspensions of carbon‐doped TiO2”. J.

Phys. Chem. C. 112 (2008) 15134‐15139.

[10] Goldstein, S.; Behar, D.; Rabani, J. “Nature of the oxidizing species formed

upon UV photolysis of C‐TiO2 aqueous suspensions”. J. Phys. Chem. C

113 (2009) 12489‐12494.

[11] Beltrán, F.J. “Ozone Reaction Kinetics for Water and Wastewater Systems”.

CRC Press, Boca Raton, Florida, USA, 2004.

[12] Mvula, M.; von Sonntag, C. “Ozonolysis of phenols in aqueous solution”.

Org. Biomol. Chem. 1 (2003) 1749‐1756.



Eng. 31 (2009) 301‐308.



[15] Rajeswari, R.; Kanmani, S. “Degradation of pesticide by photocatalytic

ozonation process and study of synergistic effect by comparison with

photocatalysis and UV/ozonation processes”. J. Adv. Oxid. Technol. 12 (2009)

208‐214.

[16] García‐Araya, J.F.; Beltrán, F.J.; Aguinaco, A. “Diclofenac removal from

water by ozone and photolytic TiO2 catalysed processes”. J. Chem. Technol.

Biotechnol. 85 (2010) 798‐804.

[17] Zou, L.D.; Zhu, B. “The synergistic effect of ozonation and photocatalysis on

color removal from reused water”. J. Photochem. Photobiol. A Chem. 196 (2008)

24‐32.

[18] Beltrán, F.J.; Rivas, F.J.; Gimeno, O.; Carbajo, M. “Photocatalytic enhanced

oxidation of fluorene in water with ozone. Comparison with other chemical

oxidation methods”. Ind. Eng. Chem. Res. 44 (2005) 3419‐3425.

[19] Beltrán, F.J.; Aguinaco, A.; García‐Araya, J.F. “Mechanism and kinetics of

sulfamethoxazole photocatalytic ozonation in water”. Water Res. 43 (2009) 1359‐

1369.

[20] Buxton, G.V.; Greenstock, C.L.; Helman, W.P.; Ross, A.B. “Critical review of

rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl

radicals (∙OH/O‐) in aqueous solution”. J. Phys. Chem. Ref. Data 17 (1988) 513‐

886.


138

[21] Sun, L.; Bolton, J.R. “Determination of the quantum yield for the

photochemical generation of hydroxyl radicals in TiO2 suspensions”. J. Phys.

Chem. 100 (1996) 4127‐4134.

[22] Tachikawa, T.; Tojo, S.; Kawai, K.; Endo, M.; Fujitsuka, M.; Ohno, T.;

Nishijima, K.; Miyamoto, Z.; Majima, T. “Photocatalytic oxidation reactivity of

holes in the sulfur and carbon‐doped TiO2 powders studied by time‐resolved

diffuse reflectance spectroscopy”. J. Phys. Chem B 108 (2004) 19299‐19306.

[23] Wang, C.; Rabani, J.; Bahnemann, D.W.; Dohrmann, J.K. “Photonic

efficiency and quantum yield of formaldehyde formation from methanol in the

presence of various TiO2 photocatalysts”. J. Photochem. Photobiol. A Chem. 148

(2002) 169‐176.

[24] Gao, R.; Stark, J.; Bahnemann, D.W.; Rabani, J. “Quantum yields of hydroxyl

radicals in illuminated TiO2 nanocrystallite layers”. J. Photochem. Photobiol. A

Chem. 148 (2002) 387‐391.

[25] Marugán, J.; Hufschmidt, D.; López‐Muñoz, M.J.; Selzer, V.; Bahnemann, D.

“Photonic efficiency for methanol photooxidation and hydroxyl radical

generation on silica‐supported TiO2 photocatalysts”. Appl. Catal. B Environ. 62

(2006) 201‐207.

[26] Hoigné, J.; Bader, H. “Rate constant of reactions of ozone with organic and

inorganic compounds in water I. Non‐dissociating organic compounds”. Water

Res. 17 (1983) 173‐183.

[27] Nash, T. “The colorimetric estimation of formaldehyde by means of the

Hantzsch reaction”. Biochemistry 55 (1953) 416‐421.

[28] Flyunt, R.; Leitzke, A.; Mark, G.; Mvula, E.; Reisz, E.; Schick, R.; von

Sonntag, C. “Determination of ∙OH, O2‐, and hydroperoxide yields in ozone

reactions in aqueous solution”. J. Phys. Chem. B 107 (2003) 7242‐7253.

[29] Masschelein, W.; Denis, M.; Ledent, R. “Spectrophotometric determination

of residual hydrogen peroxide”. Water Manag. Water Sewage Works (1977) 69‐

72.


method”. Water Res. 15 (1981) 449‐456.

[31] Hatchard, C.G.; Parker, C.A. “A new sensitive chemical actinometer. 2.

Potassium ferrioxalate as a standard chemical actinometer”. Proc. Roy. Soc.

Lond. Ser. Math. Phys. Sci. 235 (1956) 518‐536.



139

[32] Sandell, E.B. “Colorimetric Determination of Traces of Metals”. Interscience

Pubs., New York, 1959.

[33] Serpone, N.; Salinaro, A. “Terminology, relative photonic efficiencies and

quantum yields in heterogeneous photocatalysis. Part I: suggested protocol”.

Pure Appl. Chem. 71 (1999) 303‐320.

[34] Alfano, O.M.; Cabrera, M.I.; Cassano, A.E. “Photocatalytic reactions

involving hydroxyl radical attack‐I. Reaction kinetics formulation with explicit

photon absorption effects”. J. Catal. 172 (1997) 370‐379.

[35] Cabrera, M.I.; Negro, A.C.; Alfano, O.M.; Cassano, A.E. “Photocatalytic

reactions involving hydroxyl radical attack‐II. Kinetics of the decomposition of

trichloroethylene using titanium dioxide”. J. Catal. 172 (1997) 380‐390.

[36] Brandi, R.J.; Alfano, O.M.; Cassano, A.E. “Rigorous model and experimental

verification of the radiation field in a flat‐plate solar collector simulator

employed for photocatalytic reactions”. Chem. Eng. Sci. 54 (1999) 2817‐2827.

[37] Brucato, A.; Cassano, A.E.; Grisafi, F.; Montante, G.; Rizzuti, L.; Vella, G.

“Estimating radiant fields in flat heterogeneous photoreactors by the six‐flux

model”. AIChE J. 52 (2006) 3882‐3890.

[38] Toepfer, B.; Gora, A.; Li Puma, G. “Photocatalytic oxidation of

multicomponent solutions of herbicides: Reaction kinetics analysis with explicit

photon absorption effects”. Appl. Catal. B Environ. 68 (2006) 171‐180.

[39] Loddo, V.; Addamo, M.; Augugliaro, V.; Palmisano, L.; Schiavello, M.;

Garrone, E. “Optical properties and quantum yield determination in

photocatalytic suspensions”. AIChE J. 52 (2006) 2565‐2574.

[40] Chen, H.Y.; Zahraa, O.; Bouchy, M. “Inhibition of the adsorption and

photocatalytic degradation of an organic contaminant in an aqueous suspension

of TiO2 by inorganic ions”. J. Photochem. Photobiol. A 108 (1997) 37‐44.

[41] Connor, P.A.; McQuillan, A.J. “Phosphate adsorption onto TiO2 from

aqueous solutions: an in situ internal reflection infrared spectroscopic study”.

Langmuir 15 (1999) 2916‐2921.

[42] Du, Y.; Rabani, J. “The measure of TiO2 photocatalytic efficiency and the

comparison of different photocatalytic titania”. J. Phys. Chem. B 107 (2003)

11970‐11978.

[43] Lefevre, G. “In situ Fourier‐transform infrared spectroscopy studies of

inorganic ions adsorption on metal oxides and hydroxides”. Adv. Colloid Interf.

Sci. 107 (2004) 109‐123.


140

[44] Rakovski, S.; Cherneva, D. “Kinetics and mechanism of the reaction of


[45] Rivas, F.J.; Encinas, A.; Acedo, B.; Beltrán, F.J. “Mineralization of bisphenol

A by advanced oxidation processes”. J. Chem. Technol. Biotechnol. 84 (2009)

589‐594.

CAPÍTULO 4 (CHAPTER 4) PAPER 2: Influence of structural properties on the activity of WO3 catalysts for visible light photocatalytic ozonation

A. Rey, E. Mena, A.M. Chávez, F.J. Beltrán, S. Malato

Chemical Engineering Science 126 (2015) 80-90

ABSTRACT. This work is focused on the use of WO3 catalysts with different structural

properties for photocatalytic ozonation using visible light as radiation source. Different

WO3 catalysts were prepared by thermodecomposition of tungstite precursor

(H2O·WO3) at 300 and 450 °C and different time of calcination. Photocatalysts were

characterized by means of TGA-DTA, XRD, N2 adsorption-desorption isotherms, pH of

the point of zero charge (pHPZC), XPS and DR-UV-Vis spectroscopy. Photocatalytic

ozonation of ibuprofen under visible light radiation with the catalyst prepared at 450 °C and 5 min of heat-treatment gave place to complete removal of ibuprofen in less than 20 min with TOC removal up to 87 % at 120 min. In addition, this catalyst gave place to a complete removal of a mixture of ten emerging contaminants in a municipal wastewater in less than 60 min with a mineralization up to 40 % at 120 min. The highest catalytic activity of this material in the reaction under study is due to the formation of monoclinic

structure of WO3 together with the presence of higher concentration of oxygen vacancies.

Keywords: Photocatalytic ozonation, tungsten trioxide, crystalline structure, ozone, visible light.

PAPER 2: Influence of structural properties on the

activity of WO3 catalysts for visible light ozonation

143

4.1. INTRODUCTION

The development of efficient water treatment technologies to remove organic

pollutants such as solvents, dyes, pesticides, phenolic compounds or

pharmaceuticals and personal care products, among others, is an important

issue worldwide. In general, these pollutants are persistent, toxic and/or

recalcitrant to conventional wastewater treatments and have shown to be a

potential risk to human health and to the environment. Therefore, they must be

removed in order to prevent their discharge and, of course, to fulfill the required

limits of an increasingly stringent legislation.

Among the alternative technologies, the combination of ozone and

heterogeneous photocatalysis with a semiconductor, i.e. photocatalytic

ozonation, has demonstrated to efficiently remove organic compounds that are

degraded little or slowly by photocatalysis or ozonation alone. The

improvement of the combined process is mainly related to the formation of

higher concentration of oxidizing species such as hydroxyl radicals compared to

the single treatments [1,2]. Usually, this treatment has been applied by using

TiO2 as photocatalyst and UV irradiation. However, from an environmental and

economic point of view, the use of solar light is a need for the practical

deployment of photocatalytic technologies. In this sense, TiO2 presents some

limitations since it only uses around 5 % of solar radiation (UV range) in spite of

being the archetypical photocatalyst due to its relatively high efficiency, low cost

and availability [3].

Different research efforts have been carried out to overcome this limitation

mainly focusing on the development of doped or modified TiO2 or the use of

different semiconductors such as CdS, WO3, SnO2 or ZnO [3,4]. Among them,

bare tungsten trioxide WO3 has been considered unsuitable for efficient

photocatalytic oxidation due to its conduction band level respect to the

reduction band level of O2, whereas it has demonstrated an efficient behavior in

the presence of O3 [5]. WO3 is a wide‐band‐gap semiconductor (2.6 ‐ 3.0 eV) that


144

can be excited with radiation corresponding to the visible region of the

electromagnetic spectrum [6,7]. This fact together with its relatively low cost, no

toxicity and availability make WO3 an attractive alternative to TiO2 for

photocatalytic ozonation processes under visible or solar light radiation.

The use of WO3 in the photocatalytic ozonation process has been limited to

commercial monoclinic WO3 and phenol as target compound [5,8]. Thus, this

work focuses on the study of WO3 photocatalysts with different structural

properties for the photocatalytic ozonation of a model compound using visible

light radiation. Ibuprofen (IBP), a nonsteroidal anti‐inflammatory drug, has been

selected as target contaminant which is included in the family of emerging

contaminants (ECs) frequently detected in wastewater and different aquatic

environments [9,10]. In addition, a more realistic application of photocatalytic

ozonation with WO3 catalysts has been studied in a mixture of 10 ECs spiked

municipal wastewater treatment plant (MWWTP) effluent using the entire

simulated solar radiation spectrum.


4.2.1. Catalysts preparation

Photocatalysts were prepared by thermodecomposition of commercial

H2WO4 (99 %, Sigma‐Aldrich) yellow powder precursor. The sample was

calcined in air atmosphere in a muffle furnace at different temperatures (300 and

450 °C) for 1 to 30 min in order to obtain different H2WO4/WO3 crystalline

structures according to the method reported by Cao et al. [11]. The

thermodecomposition reaction proceeds through water loss of the tungstite

structure of H2WO4 (H2O∙WO3) and subsequent crystallization as cubic WO3.

Then, at higher temperature cubic WO3 can be transformed into the monoclinic

WO3 crystalline phase [12]. This process is summarized in Eq. (4.1). Table 4.1

shows the nomenclature of the catalysts and heat‐treatment conditions together

with some characterization results.



145

T,t

2 4 2 3 3,cubic 2tungstite

T,t3,cubic 3,monoclinic

H WO H O∙WO WO H O

WO WO

(4.1)

Table 4.1. Nomenclature, calcination conditions and some properties of the

photocatalysts.

CATALYST Calcination

conditions

Tungstite

(%)

Cubic

(%)

Monoclinic

(%)

SBET

(m2 g‐1) pHPZC

Eg

(eV)

W‐0 ‐‐‐ ~100 0 0 27.8 4.71 2.42

W300‐1 300 °C,

1 min 82.0 18.0 0 28.7 4.39 2.43

W300‐3 300 °C,

3 min 0 100 0 40.3 4.13 2.64

W300‐5 300 °C,

5 min 0 85.7 14.3 39.0 4.18 2.65

W300‐60 300 °C,

60 min 0 80.3 19.7 n.m n.m n.m

W450‐5 450 °C,

5 min 0 23.9 76.1 30.2 5.25 2.65

W450‐30 450 °C,

30 min 0 9.5 90.5 21.1 4.98 2.67

n.m.: not measured

4.2.2. Characterization

Thermal gravimetry and differential temperature analysis (TGA‐DTA) was

performed with a SETSYS Evolution‐16 apparatus (Setaram) using the following

conditions: sample loading 20 mg, air flow 50 cm3 min‐1 at a heating rate of

5 °C min‐1 from room temperature to 600 °C.

The crystalline phases present in the photocatalysts were inferred from their

X‐ray diffraction (XRD) patterns recorded using a powder Bruker D8 Advance

XRD diffractometer with a Cu Kα radiation (λ = 0.1541 nm). The data were

collected from 2θ = 10° to 70° at a scan rate of 0.02 s−1 and 1 s per point.

Crystalline composition semi‐quantitative analysis of the samples was

performed on multi‐phase patterns by the Reference Intensity Ratio (RIR)


146

method using reference diffraction patterns by means of EVA v.14 software

(Bruker‐AXS). Due to the overlapping of the main characteristic peaks of the

cubic and monoclinic WO3 patterns, the analyses for the catalysts with

monoclinic contribution were performed using a calculated value of RIR for the

2θ = 28.6° diffraction peak corresponding to the (112) reflection plane.

BET surface area and pore volume of the photocatalysts were determined

from their nitrogen adsorption‐desorption isotherms obtained at ‐196 °C using

an Autosorb 1 apparatus (Quantachrome). Prior to analysis the samples were

outgassed at 150 °C for 24 h under high vacuum (< 10−4 Pa).

The point of zero charge (pHPZC) of each sample was estimated using the

mass titration method proposed by Subramanian et al. [13]. Suspensions of the

solid in deionized water at 5 % (w/w) were prepared and the pH measured after

24 h of stirring.

X‐ray photoelectron spectra (XPS) were obtained with a Kα Thermo Scientific

apparatus with an Al Kα (h = 1486.68 eV) X‐ray source using a voltage of 12 kV

under vacuum (2x10−7 mbar). Due to the lack of C1s signal, binding energies

were calibrated relative to the O1s peak of stoichiometric WO3 at 530.2 eV

[14,15].

Diffuse reflectance UV‐Vis spectroscopy (DR‐UV‐Vis) measurements, useful

in the determination of the semiconductor band gap, were performed with a

UV‐Vis‐NIR Cary 5000 spectrophotometer (Varian‐Agilent Technologies)

equipped with an integrating sphere device.

4.2.3. Catalytic activity measurements

Ibuprofen sodium salt (IBP), was used as target compound to test the

catalytic activity of the synthesized materials. Photocatalytic experiments were

carried out in semi‐batch mode in a laboratory‐scale system consisting of a 0.5 L

glass‐made spherical reactor, provided with a gas inlet, a gas outlet and a liquid



147

sampling port. The reactor was placed in the chamber of a commercial solar

simulator (Suntest CPS, Atlas) provided with a 1500 W air‐cooled Xe arc lamp

with emission restricted to visible light of wavelengths over 390 nm using

quartz, glass and polyester cut‐off filters. The irradiation intensity was kept at

550 W m−2 and the temperature of the system was maintained between 20 and 40

°C throughout the experiments. If required, a laboratory ozone generator

(Anseros Ozomat Com AD‐02) was used to produce a gaseous ozone‐oxygen

stream that was fed to the reactor. In that case, the ozone concentration was

monitored by an Anseros Ozomat GM‐6000Pro gas analyzer.

In a typical photocatalytic ozonation experiment, the reactor was first loaded

with 0.5 L of an aqueous solution containing 10 mg L‐1 of IBP at pH0 = 6.5 (not

buffered). Then, 0.125 g of the catalyst were added and the suspension was

stirred in the darkness for 30 min. After this dark stage, the lamp was switched

on and, simultaneously, a mixture of ozone‐oxygen (10 mg L‐1 ozone

concentration) was fed to the reactor at a flow rate of 20 L h‐1. The irradiation

time for each experiment was 2 h. Samples were withdrawn from the reactor at

intervals and filtered through a 0.2 μm PET membrane to remove the

photocatalyst particles except for dissolved ozone analysis.

Experiments of adsorption (i.e., absence of radiation and ozone), photolysis

(i.e., radiation in absence of catalysts and ozone), photocatalytic oxidation (i.e.,

radiation and catalyst in absence of ozone), ozonation (i.e., absence of radiation

and catalyst), and catalytic ozonation (i.e., absence of radiation) were also

carried out for comparative analysis.

In addition, the effectiveness of the best photocatalytic system was tested in a

more realistic application under complete simulated solar light radiation

(wavelengths from 300 nm). Ten emerging contaminants: acetaminophen (AAP),

metoprolol (MTP), caffeine (CAF), hydrochlorothiazide (HCT), antipyrine

(ANT), sulfamethoxazole (SFM), carbamazepine (CAR), ketorolac (KET),

diclofenac (DCF) and ibuprofen (IBP), frequently found in municipal


148

wastewater effluents (MWW), were selected. They were added at initial

concentration of 0.5 mg L‐1 each to a MWW taken from Badajoz MWWTP

(Badajoz, Spain) designed for 225,000 equivalent inhabitants with an average

inlet flow of 37,500 m3 day‐1. Effluents were collected downstream of the

MWWTP secondary biological treatment and filtered. The carbonate‐bicarbonate

content was eliminated by air stripping after acidifying the MWW. Then pH was

restored and MWW was stored at ‐20 °C until use.

IBP concentration was analyzed by HPLC‐DAD (Hitachi, Elite LaChrom)

using a Phenomenex C‐18 column (5 μm, 150 mm long, 3 mm diameter) as

stationary phase and 0.7 mL min−1 of acetonitrile‐acidified water (0.1 % H3PO4)

as mobile phase (60 ‐ 40 v/v, isocratic). Identification and quantification was

carried out at 220 nm. ECs concentration in MWW matrix were analyzed by the

same HPLC system with 0.2 mL min‐1 of acetonitrile‐acidified water (0.1 %

formic acid) as mobile phase with a gradient from 10 % to 100 % in acetonitrile

in 40 min and 20 min re‐equilibration time. Identification and quantification was

carried out at the maximum wavelength of each compound. Total organic

carbon (TOC) was measured using a Shimadzu TOC‐VSCH analyzer. Aqueous

ozone was measured by following the indigo method using a UV‐Vis

spectrophotometer (Evolution 201, Thermospectronic) set at 600 nm [16]. Ozone

in the gas phase was continuously monitored by means of an Anseros Ozomat

GM‐6000Pro analyzer. Hydrogen peroxide concentration was determined

photometrically by the cobalt/bicarbonate method at 260 nm using a UV‐Vis

spectrophotometer (Evolution 201, Thermospectronic) [17].


4.3.1. Characterization of the photocatalysts

Tungstite is a hydrous tungsten oxide, H2O∙WO3 that crystallizes in the

orthorhombic system and can be thermally decomposed into different WO3

structures. WO3 consists of a three‐dimensional array of corner sharing metal‐



149

oxygen octahedral in a chess‐board type of array. The ideal structure is a ReO3‐

type cubic form but this is easily distorted away since the relatively small

tungsten atoms tend to be displaced from the center of the octahedral. These

displacements are temperature dependent and also influenced by the presence

of impurities which make WO3 to adopt a variety of symmetry types:

monoclinic, triclinic, orthorhombic and tetragonal. In addition, another two

metastable structures of WO3 have been obtained at low temperatures:

hexagonal and pyrochlore [18].

Thus, knowing the thermal behavior of the commercial tungstite precursor is

mandatory to understand the crystallization of the different WO3 structures.

Figure 4.1 shows the TGA‐DTA profiles for the H2WO4 precursor. The evolution

can be divided in 4 different regions. First of all, from ambient temperature to

150 °C, a small weight loss around 1 % is observed accompanied by an

endothermic contribution which can be assigned to the evaporation of adsorbed

H2O. In region II, from 150 to 325 °C, a noticeable weight loss is obtained

together with an endothermic peak at 225 °C which corresponds with the loss of

coordinated H2O from the tungstite structure to form cubic WO3 [11,12]. The

cubic WO3 structure is metastable and it can be stabilized by the presence of Na+

impurities [19]. However, at this temperature, formation of different structures

such as monoclinic WO3 cannot be disregarded as it is expected from the

absence of impurities in the commercial precursor used here. On the other hand,

in the range 325 ‐ 500 °C (region III), it can be observed a slight endothermic

contribution around 371 °C followed by an exothermic peak with a maximum

centered at 447 °C together with a slight weight loss (around 0.2 %). In this

region, the complete transformation of cubic WO3 into monoclinic WO3 occurs

[11,12]. No significant changes are observed in region IV in the range 500 ‐ 600

°C. From these results, the heat‐treatment temperature of the precursor was set

at 300 and 450 °C to obtain different crystalline phases composition of the

catalysts. The time of the heat‐treatment was relatively short (1 ‐ 30 min)

following the work of Cao et al. [11]. One sample was also heat‐treated during 60


150

min for comparative purposes.

0 100 200 300 400 500 6000.93

0.94

0.95

0.96

0.97

0.98

0.99

1.00

III

M/M

0

TEMPERATURE (oC)

TGA

I

IVII-10

-8

-6

-4

-2

0

2

4

6

8

10

DT

A DTA

Figure 4.1. TGA‐DTA profiles of H2WO4 precursor.

The structural characterization of the catalysts was performed by XRD as

shown in Figure 4.2 where XRD patterns of the tungstite H2O∙WO3

orthorhombic phase (JCPDS 018‐1418), cubic WO3 (JCPDS 41‐0905) and

monoclinic WO3 (JCPDS 043‐1035) have been plotted for comparison. It can be

noticed that the main peaks in the diffraction pattern of the precursor, W‐0, are

attributable to the orthorhombic phase of tungstite, although a small shoulder

around 24.2° can also indicate a low contribution of the cubic WO3 phase. When

the precursor is heat‐treated at 300 °C the cubic structure of WO3 is

progressively formed as the time of calcination increases. Thus, for 1 min at 300

°C (W300‐1) it can be observed the coexistence of tungstite orthorhombic phase

together with cubic WO3, as the main peaks of tungstite at 16.5° and 25.6°

decreased in favor of the increasing cubic diffraction peaks. The transformation

of tungstite into cubic WO3 has been completed in the W300‐3 catalyst, where the

main tungstite diffraction peaks have completely disappeared and all the

diffraction peaks are attributable to cubic crystalline structure. After that, an



151

increase in the time of calcination at 300 °C provokes the formation of

monoclinic WO3 to some extent. This can be noticed by the presence of the

contributions centered at 26.5° and 28.6°. These are characteristics of the

monoclinic crystalline structure, together with the widening of the peaks at 23.9

and 34.0° that are affected by the overlapped cubic‐monoclinic contributions,

with the formation of the triple peaks in monoclinic phase at these 2 values

corresponding to reflections from (001), (020) and (200) planes [14]. It is also

noticeable that an increase in the calcination time from 5 to 60 min did not

produce any significant change in the XRD patterns (W300‐5 and W300‐60) which

is indicative of the stabilization of the cubic structure at this temperature.

Finally, the temperature increase from 300 to 450 °C provoked the

transformation of the cubic phase of WO3 into monoclinic to a greater extent

according to previous TGA‐DTA results. Table 4.1 shows the semiquantitative

analysis of crystalline phases in the catalysts.

Table 4.1 also summarizes the BET surface area values of the catalysts from

N2 adsorption‐desorption isotherms. In general, small surface areas are obtained

in this type of materials unless different strategies are used for this purpose. As

can be observed, the surface areas undergo a slight increase (from c.a. 30 to 40

m2 g‐1) when the dehydration of tungstite occurs and cubic crystalline structure

is formed. On the other hand, at higher calcination temperature (450 °C), a

decrease of the surface area is observed in the catalyst, being higher as the time

of the heat treatment increases.


152

10 20 30 40 50 60 70

W300

-60

W450

-30

W450

-5

W300

-5

W300

-3

W300

-1

2 (º)

INT

EN

SIT

Y (

a.u.

)

W-0

H2WO

4, tungstite (orthorhombic)

018-1418

WO3, cubic

41-0905

WO3, monoclinic

043-1035

Figure 4.2. XRD patterns of the catalysts.



153

The pH of the point of zero charge was also determined for the catalysts

studied and the results are summarized in Table 4.1. The acidity of the surface

can play a significant role in the behavior of WO3 based materials in wastewater

treatments related to adsorption processes of organic compounds and species

with unpaired electrons [20]. It can be noticed a slight decrease in the value of

pHPZC together with the transformation of tungstite into WO3 cubic structure so

that the catalysts with this majority crystalline phase have a pHPZC value around

4. On the other hand, an increase in the calcination temperature up to 450 °C led

to pHPZC values around 5 for W450‐5 and W450‐30 catalysts. The decrease of the

surface acidity with increasing the calcination temperature has been previously

reported in monoclinic WO3 [21] and WO3‐ZrO2 catalysts [22], and has been

attributed to the creation of vacancies in the WO3 structure from the extraction

of lattice oxygen, which leads to a decrease in the number of Lewis acid sites

[21]. In addition, a higher calcination temperature could also produce the

dehydration/dehydroxylation of the WO3 surface to some extent lowering the

number of Brönsted acid sites [21].

Surface characterization of the catalysts was performed by XPS to analyze the

stoichiometry and chemical binding states of W. XPS spectra of W4f region are

depicted in Figure 4.S1 of the supplementary information. In all cases, the

spectra obtained showed a doublet with binding energies around 35.5 and 37.6

eV for W 4f7/2 and W 4f5/2, respectively, with the W 4f7/2‐W4f5/2 binding energy

separation being 2.1 eV. The energy position of this doublet corresponds to the

W6+ oxidation state in WO3 [14,15,23]. On the other hand, XPS spectra of O1s

spectral region for all the catalysts are shown in Figure 4.S2. The main

contribution at 530.2 eV has been assigned to oxygen in the WO3 lattice

[14,15,24]. Moreover, an additional contribution centered at 532.5 eV has been

observed for W‐0 and W300‐1 catalysts, which has been assigned to oxygen in

water molecules bound in the tungstite structure (H2O∙WO3) or absorbed in the

catalyst surface [14]. This is consistent with TGA‐DTA and XRD results having


154

these two catalysts more than 80 % of tungstite in their structure.

The diffuse reflectance UV‐Vis spectra of the photocatalysts (Figure 4.3)

showed a higher optical absorbance in the visible region up to ca. 530 nm for

tungstite composed catalysts (W‐0 and W300‐1), whereas decreased to ca. 480 nm

for cubic and monoclinic WO3 composed catalysts (W300‐3, W300‐5, W450‐5, W450‐

30). The optical energy band gap (Eg) was calculated by means of Tauc’s

expression in Figure 4.S3 (supplementary information), and results are

summarized in Table 4.1. These values are approximate due to the need of

extrapolation of the resulting curve as can be observed in Fig. 4.S3 for all the

catalysts. The Eg value for tungstite H2WO4 (major component of W‐0 and W300‐

1) was found to be 2.42 eV which is increased up to 2.64 when the precursor is

completely transformed into cubic WO3 structure (in W300‐3 catalyst). Then, the

Eg value progressively increased up to 2.67 eV as monoclinic WO3 was formed.

These results are quite similar to previously reported values [7,11].

200 300 400 500 600 700 800 9000.00

0.25

0.50

0.75

1.00

1.25

1.50

AB

SO

RB

AN

CE

(u

.a.)

WAVELENGTH (nm)

W-0 W

300-1

W300

-3

W300

-5

W450

-5

W450

-30

Figure 4.3. DR‐UV‐Vis spectra of the catalysts.

4.3.2. Comparison of processes

The effectiveness of the photocatalysts in the use of visible light radiation



155

combined with ozone was tested using IBP as target compound by cutting off all

the wavelengths lower than 390 nm in the solar simulator. First, the performance

of the combined process (photocatalytic ozonation) respect to the single

treatments has been checked with all the catalysts. Results obtained with W450‐5

catalyst are depicted in Figure 4.4 and Figure 4.5, being this catalyst

representative of the behavior of the most active ones. Fig. 4.4(A) shows the time

evolution of IBP dimensionless concentration upon the different treatments

applied. First of all, the absence of IBP depletion due to direct photolysis with

visible light radiation was confirmed which is consistent with the UV‐Vis

absorption spectrum of this compound (supplementary Figure 4.S4). On the

other hand, IBP adsorption capacity of the W450‐5 catalyst at the conditions

studied here was somewhat noticeable, reaching around 10 % of IBP removal

after 2 h contact time. The depletion rate of IBP is increased in photocatalytic

oxidation reaching 25 % of IBP removal after 2 h of treatment although it can be

noticed the low efficiency of the process as a consequence of the high

recombination rate in WO3 materials used in the presence of oxygen [5]. When

O3 is present, all the treatments led to complete IBP removal in less than 1 h of

reaction time (with conversion higher than 99 %) regardless of the

presence/absence of catalyst and/or radiation in contrast to O3‐free treatments.

Although the evolution of IBP is quite similar for all the O3‐treatments, some

slight differences can be observed in the depletion rate. Single ozonation and

photolytic ozonation presented quite similar IBP depletion rate. On the other

hand, the presence of the W450‐5 catalysts and ozone in dark conditions (catalytic

ozonation) did not improve the degradation of IBP thus indicating that this

catalyst does not exert any catalytic effect in ozone decomposition into reactive

species such as hydroxyl radicals or produces an inefficient decomposition for

IBP degradation. Photocatalytic ozonation gave place to the highest IBP

depletion rate reaching complete IBP removal in 20 min compared to 40 ‐ 50 min

necessary for the other O3 treatments.


156

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Dark

Adsorption Photolysis Photocatalytic oxidation Ozonation Catalytic ozonation Photolytic ozonation Photocatalytic ozonation

CIB

P/C

IBP

,0

TIME (min)

Figure 4.4. (A) IBP and (B) TOC dimensionless concentration evolution during all the

treatments applied with W450‐5 catalyst (lines show trends). Conditions: pH0 = 6.5, T = 20 ‐

‐ 40 °C, CIBP,0 = 10 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 20 L h‐1 (O2 or O3/O2).

On the other hand, it is known that the oxidation of IBP proceeds through

different steps giving place to different intermediate compounds which are

eventually transformed into CO2 and H2O (complete mineralization). Thus, the

(A)

(B)

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Dark

TO

C/T

OC

0

TIME (min)

0

0

CT

OC

/CT

OC

,0



157

mineralization of IBP has been followed and presented in Fig. 4.4(B) as

normalized TOC concentration. As expected from previous IBP results, neither

adsorption nor direct photolysis led to significant mineralization percentages.

Photocatalytic oxidation also gave place to negligible mineralization according

to the IBP conversion reached, lower than 25 %. Regarding the O3‐treatments,

noticeable differences have been found depending on the presence/absence of

radiation and/or photocatalyst. Single ozonation led to 25 % of mineralization at

60 min and then stopped as a consequence of the formation of some

intermediate compounds that are refractory to ozone direct reaction (mainly

short‐chain organic acids) [25], and/or due to the absence of IBP in the reaction

medium which eventually produces hydroxyl radicals through IBP‐O3 direct

reaction as discussed below. Fairly similar results were observed during

photolytic ozonation under visible light radiation. On the other hand, the

presence of the catalyst combined with ozone in dark conditions (catalytic

ozonation) did not improve the previous results. This process showed the

poorest behavior among the O3‐treatments in terms of mineralization, with

negligible TOC conversion during the reaction time, confirming the absence of

any positive catalytic effect in the reaction. In fact, the low mineralization

reached compared to ozone alone treatment is likely indicative of an inefficient

consumption of O3 in the catalyst surface in the darkness. These results are

contrary to those found for the mineralization of phenol with a commercial WO3

(monoclinic) catalyst [5], which points out that the target compound nature can

play a key role in the interactions compound‐catalyst surface during catalytic

ozonation. On the contrary, photocatalytic ozonation showed the highest

mineralization rate, leading to 87 % mineralization at 120 min reaction time.

These results also point out the synergism produced between ozone and visible

light irradiated WO3 since final TOC removal of the combined process is much

higher than the sum of these values for the individual processes (87 % in

O3/WO3/Vis vs. 5 % in O2/WO3/Vis and 25 % in O3), confirming also the ability of

O3 to capture the electrons on the WO3 surface compared to O2 [5].


158

0 20 40 60 80 100 1200

1x10-5

2x10-5

3x10-5

4x10-5

5x10-5

6x10-5

CO

3 (M

)

TIME (min)

0 20 40 60 80 100 1200

1x10-5

2x10-5

3x10-5

4x10-5

5x10-5 Photolysis Photocatalytic oxidation Ozonation Catalytic ozonation Photolytic ozonation Photocatalytic ozonation

CH

2O2 (M

)

TIME (min)

Figure 4.5. (A) Dissolved O3 concentration and (B) H2O2 concentration during all the

treatments applied with W450‐5 catalyst (lines show trends). Conditions: pH0 = 6.5, T = 20 ‐

‐ 40 °C, CIBP,0 = 10 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 20 L h‐1 (O2 or O3/O2).

In addition, time profiles of dissolved ozone and hydrogen peroxide

concentration during the treatments have been plotted in Figure 4.5. Regarding

(A)

(B)



159

dissolved ozone, in Fig. 4.5(A) it is seen that ozone accumulated in solution

during the single ozonation experiment reaching a maximum concentration at

about 60 min reaction time. Different profiles were observed in the other O3‐

treatments in which ozone concentration increased in the first minutes of the

experiment and then remained almost constant. The ozone concentration was

always lower in catalytic ozonation than during single ozonation. This

observation together with the low mineralization reached, is an additional

evidence of an inefficient decomposition of O3 on the catalyst surface in the

darkness. On the other hand, in the photocatalytic ozonation process, the

maximum concentration of dissolved ozone was noticeably lower, leading to a

nearly steady concentration of about 1x10‐5 M in contrast to 5x10‐5 M during

single ozonation. The lowest values of dissolved ozone together with the highest

mineralization rate observed in this treatment suggest that O3 has been

consumed on the WO3 catalyst surface through its reaction with electrons

generated in the photocatalytic process, thus enhancing the production of

oxidizing species such as hydroxyl radicals, as with TiO2 catalysts [1,2,26].

On the other hand, hydrogen peroxide is commonly formed through direct

ozone reactions [27‐29]. This has been experimentally confirmed in this work for

IBP, as shown in Fig. 4.5(B). Thus, during single ozonation an increase in H2O2

concentration up to 45 min, when IBP was completely removed, can be

observed. Then, H2O2 concentration remained almost constant until the end of

the experiment. The concentration of H2O2 formed during catalytic and

photolytic ozonation was somewhat lower. On the other hand, during

photocatalytic ozonation the formation of H2O2 takes place at higher rate

reaching a maximum concentration at about 20 ‐ 30 min when IBP was

completely removed. Then the consumption of H2O2 was faster and its

concentration negligible at the end of the treatment. These results point out that

H2O2 is being consumed through photocatalytic reactions probably acting as an

electron acceptor in the WO3 surface thus increasing the production of hydroxyl


160

radicals and avoiding recombination to some extent, as also occurs with TiO2

[26]. Finally, the formation of H2O2 during photocatalytic oxidation was much

lower than in ozone processes confirming the production of H2O2 mainly

through direct IBP‐O3 reaction.

Therefore, compared to individual processes, photocatalytic ozonation

treatment with WO3 and visible‐light radiation led to a faster IBP depletion with

the highest mineralization rate due to the synergistic effect between O3 and

irradiated WO3.

4.3.2.1. Simplified mechanistic approach of IBP photocatalytic ozonation

Direct ozone reactions in water follow second order kinetics and the kinetic

regime can be established through the determination of the Hatta number

according to [25]:

3 3IBP O O IBP

L

k D CHa

k

(4.2)

where DO3 = 1.3x10‐9 m2 s‐1 is the ozone diffusivity in water [30], and kL = 5x10‐5

m s‐1 is the individual liquid phase mass transfer coefficient for a bubbled stirred

tank [31], being CIBP the molar IBP concentration and kIBP‐O3 = 9.1 M‐1 s‐1 the direct

reaction rate constant at pH = 7 [32]. The highest calculated Ha value was 0.023

for the initial conditions thus confirming slow kinetic regime of ozone

absorption. Thus, the depletion rate of IBP through direct O3 reaction is

described as follows:

3 3

3

IBPIBP O O IBP

O

dC 1k C C

dt z

(4.3)

where z = 1 is the stoichiometric coefficient of the reaction in mol O3/mol IBP

and CO3 is the dissolved ozone concentration. The differential equation (4.3) was

solved using the value of the kinetic constant and the experimental values of CO3



161

by means of 4th order Runge‐Kutta. The results of simulated dimensionless CIBP

together with the experimental values during single ozonation are depicted in

Figure 4.6. It can be clearly noticed that the experimental rate of IBP depletion is

higher than the expected results (simulated), thus indicating that additional

reactions may develop and contribute to IBP removal. Indirect reactions due to

O3 decomposition are initiated through the following steps:

3 2 2O HO HO O (4.4)

pKa 11.3

2 2 2H O HO H (4.5)

‐2 3 23HO O O HO (4.6)

pKa 4.8 ‐

2 2HO O H

(4.7)

‐ ‐32 3O O O (4.8)

‐3O H ... HO

(4.9)

These reactions lead to the formation of hydroxyl radicals ( HO) capable of

oxidizing IBP in water. However, the pH of the reaction varied from 6.5 to 4.5

and therefore, reaction (4.4) between 3O / HO seems not to play an important

role on the IBP disappearance. In addition, an accumulation of H2O2 was

observed during the single ozonation experiment (Fig. 4.5(B)), thus indicating

that reaction (4.6) between 3 2O / HO can be also disregarded. These evidences

together with the evolution of mineralization whereas IBP is still present in the

reaction medium lead to hypothesize the formation of ozonide radical through

IBP‐O3 direct reaction, as it has been previously reported for phenol or

diclofenac [28,33,34]:

‐33 OINTOIBP (4.10)

which immediately results in the formation of HOthrough reaction (4.9), being


162

0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

CIB

P/C

IBP

,0

TIME (min)

Experimental Simulated O

3

Simulated O3+HO·

INT any intermediate organic compound.

Figure 4.6. Experimental and simulated IBP dimensionless concentration evolution

during single ozonation. Conditions: pH0 = 6.5, T = 20 ‐ 40 °C, CIBP,0 = 10 mg L‐1, CO3,g inlet

= 10 mg L‐1, Qg = 20 L h‐1 (O3/O2).

With these considerations, the depletion rate of IBP through direct O3

reaction taking into account the formation of HO during this reaction is

described as follows:

3 3

3

IBPIBP O O IBP IBPIBP HO HO

O ‐HO

dCk C C k C C

dt

(4.11)

3 3

3

HOIBP O O IBP IBPIBP HO HO

IBP‐O

dCk C C k C C

dt

(4.12)

where kIBP‐HO∙ is the rate constant between hydroxyl radical and IBP and CHO∙ is

the hydroxyl radical concentration. The differential equations (4.11) and (4.12)

were simultaneously solved using the value of the kinetic constants and the

experimental values of CO3 by means of 4th order Runge‐Kutta (kIBP‐HO∙ = 7.4x109



163

M‐1 s‐1 [32]). These results are also depicted in Figure 4.6, where it can be

observed that the experimental IBP depletion rate is still higher than the

simulated values. This could be related to the accuracy of the rate constants used

for the simulation and/or the development of additional IBP degradation routes.

The confirmation of the different degradation ways needs extra experiments out

of the scope of this work.

According to this and the previous findings, it can be postulated that

photocatalytic ozonation of IBP with WO3 catalyst takes place through reactions

(4.13) ‐ (4.20), where both O3 and HO are responsible for IBP disappearance and

photogenerated oxidizing species ( HO and/or positive holes) are the main

responsible for mineralization.

3 2 2IBP O ... INT H O HO (4.13)

h3 3 3WO WO e WO h

(4.14)

3 3WO e O ... HO (4.15)

3 2 2WO e H O ... HO (4.16)

3WO h IBP INT (4.17)

HO IBP INT (4.18)

3 2 2WO h INT ... CO H O (4.19)

2 2HO INT ... CO H O (4.20)

Reaction (4.13) summarizes the direct IBP‐O3 steps that would lead to some

large organic intermediates together with the H2O2 detected and accumulated

during single ozonation experiment, and eventually with the formation of HO .

On the other hand, visible light radiation provokes the creation of electron‐hole

pairs (reaction (4.14)) and electrons are proposed to react with dissolved O3 and


164

H2O2 through reactions (4.15) and (4.16), according to the profiles of these

species during photocatalytic ozonation versus single ozonation. Finally,

photogenerated holes and hydroxyl radicals can oxidize both IBP and any INT

to achieve the complete conversion into CO2 and H2O according to reactions

(4.17) ‐ (4.20).

The relative contribution of each step to the overall process would depend on

the experimental conditions (e.g. pH, ozone dose or catalyst loading), and the

confirmation of the intermediate species formed requires additional

experimental work that will be the subject of a future work.

4.3.3. Catalysts screening for photocatalytic ozonation under visible‐light

radiation

All the catalysts prepared were tested in the photocatalytic ozonation of IBP

under visible‐light radiation. The results are depicted in Figure 4.7 and Figure

4.8 together with single ozonation results for the sake of comparison. Regarding

the evolution of IBP (Fig. 4.7(A)) results are quite similar although some trends

can be extracted. First of all, slight differences were observed in the IBP

adsorption capacity of the catalysts during the dark stage that is generally

consistent with the values of pHPZC obtained where a higher surface acid

character led to a higher amount of initial IBP adsorbed. However, these

differences seem not to play a crucial role in the behavior of the catalysts. On the

other hand, it can be noticed that W‐0 and W300‐1 catalysts gave place to similar

IBP depletion rate than ozone alone whereas the rest of the catalysts gave place

to a slight higher depletion rate. In general, the catalytic activity is increased as

the heat‐treatment in time and/or temperature was increased. Main differences

between ozone and O3‐photocatalytic processes were found in terms of

mineralization (see Fig. 4.7(B)). Ozone alone gave place to a 25 % mineralization

at about 60 min reaction time. On the other hand, the tungstite precursor, W‐0,

and the subsequent catalysts W300‐1 and W300‐3 showed a very different

normalized TOC profile. It can be noticed a first step in which mineralization is



165

slow, with a TOC removal rate fairly lower than during ozone alone reaction up

to 60 min. After this time, the mineralization rate increased achieving around 35

‐ 40 % of mineralization at 120 min with the three catalysts. It can be noticed that

in addition to the differences in the properties of these three catalysts (Eg, SBET,

pHPZC, etc.), they have in common that no monoclinic WO3 crystallization has

occurred during the heat‐treatment applied. A significant increase in the

catalytic activity is observed with the W300‐5 catalyst in terms of mineralization,

reaching around 70 % TOC removal at 120 min. The main difference between

this and the previous catalysts is the presence of monoclinic crystalline phase to

some extent in its structure according to XRD results. The best results were

obtained with the 450 °C heat‐treated catalysts reaching around 85 ‐ 87 %

mineralization at 120 min reaction time. In these catalysts, a contribution of

monoclinic WO3 higher than 75 % was observed together with a higher pHPZC

which could indicate the presence of oxygen vacancies to some extent, although

this could not be confirmed by XPS. The differences found in the mineralization

profile of W450‐5 and W450‐30 catalysts, the first one leading to the highest

mineralization rate, could be related to the higher SBET found in this catalyst.

In addition, the evolution of dissolved ozone and hydrogen peroxide

concentration formed during the reaction has been plotted in Figure 4.8.

Regarding dissolved ozone (Fig. 4.8(A)), it is noticeable the lowest values of the

steady concentration with the most active catalysts, W450‐5 and W450‐30,

compared to the single ozonation reaction. These results suggest an efficient O3

consumption on the catalysts surface through its reaction with the electrons

generated in the photocatalytic process. Fig. 4.8(B) shows the hydrogen peroxide

concentration evolution for all the catalysts where a faster decomposition rate is

also observed for the most active ones (W450‐5 and W450‐30), thus being likely

indicative of a higher consumption of H2O2 in surface reactions as electron

acceptor.

Therefore, at similar textural and optical properties, the monoclinic structure


166

-20 0 20 40 600.0

0.2

0.4

0.6

0.8

1.0

Dark

W-0 W300-1

W300-3

W300-5

W450-5

W450-30

O3

CIB

P/C

IBP

,0

TIME (min)

in WO3 catalysts seems to be beneficial for the photocatalytic ozonation process

under visible light.

Figure 4.7. (A) IBP and (B) TOC dimensionless concentration evolution during

photocatalytic ozonation with all the catalysts studied (lines show trends). Conditions:

pH0 = 6.5, T = 20 ‐ 40 °C, CIBP,0 = 10 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 20 L

h‐1 (O3/O2).

(A)

(B)

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Dark

TO

C/T

OC

0

TIME (min)

0

0

CT

OC

/CT

OC

,0



167

0 20 40 60 80 100 1200

1x10-5

2x10-5

3x10-5

4x10-5

5x10-5

6x10-5

CO

3 (M

)

TIME (min)

0 20 40 60 80 100 1200

1x10-5

2x10-5

3x10-5

4x10-5

5x10-5

6x10-5

W-0 W300-1

W300-3

W300-5

W450-5

W450-30

O3

CH

2O2 (M

)

TIME (min)

Figure 4.8. (A) Dissolved O3 concentration and (B) H2O2 concentration during

photocatalytic ozonation with all the catalysts studied (lines show trends). Conditions:

pH0 = 6.5, T = 20 ‐ 40 °C, CIBP,0 = 10 mg L‐1, CWO3 = 0.25 g L‐1, CO3 g inlet = 10 mg L‐1, Qg = 20 L

h‐1 (O3/O2).

(A)

(B)


168

4.3.4. Simulated solar light photocatalytic ozonation of ECs in MWW

The selected catalyst W450‐5 which presented the highest catalytic activity in

the photocatalytic ozonation of IBP with visible light was tested in the

photocatalytic ozonation of a mixture of ten ECs (AAP, MTP, CAF, HCT, ANT,

SFM, CAR, KET, DCF and IBP) in a real municipal wastewater effluent. Main

parameters of the MWW effluent are summarized in Table 4.2. The

photocatalytic ozonation treatment was carried out using the entire simulated

solar spectrum (λ > 300 nm). For comparative purposes also single ozonation

treatment was carried out. The evolution of the ECs is depicted in Figure 4.9. It

can be noticed that many of these compounds (AAP, ANT, SFM, CAR, KET and

DCF) presented similar depletion rate both in single ozonation and

photocatalytic ozonation treatments as a consequence of their high reactivity

towards O3. Thus, O3 itself is highly effective to remove these compounds.

Direct rate constants of ECs with O3 are summarized in Table 4.S1. On the other

hand, the ECs whose rate constant is low such as MTP, CAF, HCT and IBP, were

faster eliminated through photocatalytic ozonation (in less than 60 min), being

the differences found between ozone alone and the combined process higher the

lower the direct rate constant of O3‐EC reaction was. Again in the MWW matrix

main differences between ozone and photocatalytic ozonation were found in

terms of TOC removal as shown in Figure 4.10. As it can be seen, ozone alone

gave place to a lower degree of mineralization reaching less than 10 % TOC

removal after 120 min in contrast to more than 40 % for photocatalytic

ozonation. In addition, Table 4.2 also shows the characterization of the

wastewater after the treatments being noticeable the decrease in COD and

aromaticity when the combined treatment was applied. Finally, Figure 4.11

shows the evolution of dissolved O3 concentration together with the formation

of H2O2 during the treatments. The obtained profiles are consistent with the

previous results thus indicating the consumption of both O3 and H2O2 (formed

through O3‐ECs direct reaction) at the WO3 surface leading to the formation of

additional oxidizing species such as hydroxyl radicals, which improve the



169

mineralization reached. Thus, intermediate compounds formed from ECs

oxidation are expected to be more easily eliminated through photocatalytic

ozonation compared to single ozonation.

Therefore, the use of WO3 as catalyst for photocatalytic ozonation of ECs

containing MWW is an attractive alternative that improves the use of solar

radiation leading to a fast elimination of organic contaminants (mainly those

refractory to ozone reactions) and achieving an important mineralization degree.

Table 4.2. Characterization of MWW effluent.

PARAMETER Before treatment Ozonation Photocatalytic ozonation

TOC (mg C L‐1) 14.2 13.1 8.2

IC (mg C L‐1) 0.19 0.15 0.20

Turbidity (NTU) 3.0 n.m. n.m.

pH 6.5 5.9 5.7

Absorbance 254 nm 0.268 0.102 0.015

COD (mg O2 L‐1) 44 33 13

BOD5 (mg O2 L‐1) 28 n.m. n.m.

Phosphate (mg L‐1) 5 5 5

n.m.: not measured


170

-30 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

C/C

0

TIME (min)


AAP

-30 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0MTP

C/C

0

TIME (min)

-30 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0CAF

C/C

0

TIME (min)-30 0 20 40 60 80 100 120

0.0

0.2

0.4

0.6

0.8

1.0

1.2

HCT

C/C

0

TIME (min)

-30 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0ANT

C/C

0

TIME (min)-30 0 20 40 60 80 100 120

0.0

0.2

0.4

0.6

0.8

1.0SFM

C/C

0

TIME (min)

-30 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0CAR

C/C

0

TIME (min)-30 0 20 40 60 80 100 120

0.0

0.2

0.4

0.6

0.8

1.0KET

C/C

0

TIME (min)

-30 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0DCF

C/C

0

TIME (min)-30 0 20 40 60 80 100 120

0.0

0.2

0.4

0.6

0.8

1.0IBP

C/C

0

TIME (min)

Figure 4.9. ECs dimensionless concentration during ozonation and photocatalytic

ozonation with W450‐5 catalyst in a MWW effluent (lines show trends). Conditions: pH0 =

6.5, T = 20 ‐ 40 °C, CECs,0 = 0.5 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 20 L h‐1

(O3/O2).



171

0 20 40 60 80 100 1200.0

4.0x10-6

8.0x10-6

1.2x10-5

1.6x10-5

2.0x10-5

2.4x10-5

CO

3 (M

)

TIME (min)

0 20 40 60 80 100 1200

1x10-5

2x10-5

3x10-5

4x10-5

5x10-5

6x10-5

7x10-5


CH

2O2(

M)

TIME (min)

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

CT

OC

/CT

OC

,0

TIME (min)


Dark

Figure 4.10. TOC dimensionless concentration evolution during ozonation and

photocatalytic ozonation with W450‐5 catalyst in a MWW effluent (lines show trends).

Conditions: pH0 = 6.5, T = 20 ‐ 40 °C, CECs,0 = 0.5 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg

L‐1, Qg = 20 L h‐1 (O3/O2).

Figure 4.11. (A) Dissolved O3 concentration and (B) H2O2 concentration during ozonation

and photocatalytic ozonation with W450‐5 catalyst in MWW effluent (lines show trends).

Conditions: pH0 = 6.5, T = 20 ‐ 40 °C, CECs,0 = 0.5 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet =

10 mg L‐1, Qg = 20 L h‐1 (O3/O2).

(A) (B)


172

4.4. CONCLUSIONS

WO3 catalysts with different structural and surface properties where

prepared by thermodecomposition of commercial tungstite (H2O∙WO3) at

different temperatures, 300 and 450 °C, and short times, from 1 to 30 min. This

heat‐treatment allows obtaining tungstite, cubic and monoclinic crystalline

phases of tungsten trioxide. In general, as time or temperature increased,

monoclinic structure is formed which is also accompanied by a reduction of

specific surface area and an increase in the presence of oxygen vacancies. These

properties, monoclinic structure together with the presence of oxygen vacancies

played a key role in the behavior of the WO3 catalysts in the photocatalytic

ozonation of IBP under visible light radiation, favoring the electron transport in

the catalyst surface to some extent, thus improving the efficiency of the process.

Among the treatments tested with the best catalytic system, photocatalytic

ozonation showed a synergistic effect between ozone and irradiated WO3

leading to a complete IBP depletion in less than 20 min with a degree of

mineralization of about 87 % at 120 min. Finally, the use of WO3 as catalyst for

photocatalytic ozonation of ECs in a MWW effluent with simulated solar

radiation led to a fast elimination of the contaminants in less than 60 min and a

mineralization degree higher than 40 % at 120 min. The mechanism and kinetics

of photocatalytic ozonation with WO3 catalysts will be the subject of further

work.

AKNOWLEDGEMENTS

This work has been supported by the Spanish Ministerio de Economía,

Industria y Competitividad (MINECO) and European Feder Funds through the

project CTQ2012‐35789‐C02‐01. Authors acknowledge the SACSS‐SAIUEX and

UAI‐ICP for the characterization analyses. E. Mena thanks the Consejería de

Empleo, Empresa e Innovación (Gobierno de Extremadura) and European Social

Fund for providing her a predoctoral FPI grant (Ref. PD12059).



173

REFERENCES



Photobiol. C: Photochem. Rev. 6 (2005) 264‐273.

[2] Augugliaro, V.; Litter, M.; Palmisano, L.; Soria, J. “The combination of

heterogeneous photocatalysis with chemical and physical operations: A tool for


Photochem. Rev. 7 (2006) 127‐144.

[3] Hernández‐Alonso, M.D.; Fresno, F.; Suárez, S.; Coronado, J.M.


opportunities”. Energ. Environ. Sci. 2 (2009) 1231‐1257.




[5] Nishimoto, S.; Mano, T.; Kameshima, Y.; Miyake, M. “Photocatalytic water

treatment over WO3 under visible light irradiation combined with ozonation”.

Chem. Phys. Letters 500 (2010) 86‐89.

[6] Abe, R.; Takami, H.; Murakami, N.; Ohtani, B. “Pristine simple oxides as

visible light driven photocatalysts: Highly efficient decomposition of organic

compounds over platinum‐loaded tungsten oxideʺ. J. Am. Chem. Soc. 130 (2008)

7780‐7781.

[7] González‐Borrero, P.P.; Sato, F.; Medina, A.N.; Baesso, M.L.; Bento, A.C.;

Baldissera, G.; Persson, C.; Niklasson, G.A.; Granqvist, C.G.; Ferreira da Silva, A.

“Optical band‐gap determination of nanostructured WO3 film”. Appl. Phys.

Letters 96 (2010) 061909‐1‐061909‐3.

[8] Mano, T.; Nishimoto, S.; Kameshima, Y.; Miyake, M. “Investigation of

photocatalytic ozonation treatment of water over WO3 under visible light

irradiation”. J. Ceramic. Soc. Japan 119 (2011) 822‐827.

[9] Barceló, D.; Petrovic, M. (Eds.). “Emerging contaminants from industrial and

municipal wastes. Occurrence, analysis and effects”. The Handbook of

Environmental Chemistry 5‐S1. Springer, Berlin (Germany), 2008.

[10] Santos, J.L.; Aparicio, I.; Callejón, M.; Alonso, E. “Occurrence of

pharmaceutically active compounds during 1‐year period in wastewaters from

four wastewater treatment plants in Seville (Spain)”. J. Hazard. Mater. 164 (2009)


174

1509‐1516.

[11] Cao, J.; Luo, B. Lin, H.; Xu, B.; Chen, S. “Thermodecomposition synthesis of

WO3/H2WO4 heterostructures with enhanced visible light photocatalytic

properties”. Appl. Catal. B Environ. 111‐112 (2012) 288‐296.

[12] Guery, C.; Choquet, C.; Dujeancourt, F.; Tarascon, J.M.; Lassegues, J.C.

“Infrared and X‐ray studies of hydrogen intercalation in different tungsten

trioxides and tungsten trioxide hydrates”. J. Solid State Electrochem. 1 (1997)

199‐207.

[13] Subramanian, S.; Noh, J.S.; Schwarz, J.A. “Determination of the point of

zero‐charge of composite oxides”. J. Catal. 114 (1988) 433‐439.

[14] Senthil, K.; Yong, K. “Growth and characterization of stoichiometric

tungsten oxide nanorods by thermal evaporation and subsequent annealing”.

Nanotechnology 18 (2007) 395604‐1‐395604‐7.

[15] Pang, H.F.; Li, Z.J.; Xiang, X.; Fu, Y.Q.; Placido, F.; Zu, X.T. “Hierarchical

structured tungsten oxide nanocrystals via hydrothermal route: microstructure,

formation mechanism and humidity sensing”. Appl. Phys. A 112 (2013) 1033‐

1042.


method”. Water Res. 15 (1981) 449‐456.


of residual hydrogen peroxide”. Water Management. Water Sewage Works,

August (1977) 69‐72.

[18] Tilley, R.J.D. “The crystal chemistry of the higher tungsten oxides”. Int. J.

Refractory Metals Hard Mater. 13 (1995) 93‐109.

[19] Szilágyi, I.M.; Fórizs, B.; Rosseler, O.; Szegedi, A.; Németh, P.; Király, P.;

Tárkányi, G.; Vajna, B.; Varga‐Josepovits, K.; László, K.; Tóth, A.L.; Baranyai, P.;

Leskelä, M. “WO3 photocatalysts: Influence of structure and composition”. J.

Catal. 294 (2012) 119‐127.

[20] Tomova, D.; Iliev, V.; Rakovsky, S.; Anachkov, M.; Eliyas, A.; Li Puma, G.

“Photocatalytic oxidation of 2,4,6‐trinitrotoluene in the presence of ozone under

irradiation with UV and visible light”. J. Photochem. Photobiol. A Chem. 231

(2012) 1‐8.

[21] Kanan, S.M.; Lu, Z.; Cox, J.K.; Bernhardt, G.; Tripp, C.P. “Identification of

surface sites on monoclinic WO3 powders by infrared spectroscopy”. Langmuir



175

18 (2002) 1707‐1712.

[22] Sun, W.D.; Zhao, Z.B.; Guo, C.; Ye, X.K.; Wu, Y. ʺStudy of the alkylation of

isobutene with n‐butene over WO3/ZrO2 srong solid acid. 1. Effect of the

preparation method, WO3 loading, and calcination temperature”. Ind. Eng.

Chem. Res. 39 (2000) 3717‐3725.

[23] Lee, J.S.; Jang, I.H.; Park, N.G. “Effects of oxidation state and crystallinity of

tungsten oxide interlayer on photovoltaic property in bulk hetero‐junction solar

cell”. J. Phys. Chem. C 116 (2012) 13480‐13487.

[24] Azimirad, R.; Naseri, N.; Akhavan, O.; Moshfegh, A.Z. “Hydrophilicity

variation of WO3 thin films with annealing temperature”. J. Phys. D Appl. Phys.

40 (2007) 1134‐1137.

[25] Beltrán, F.J. “Ozone reaction kinetics for water and wastewater systemsʺ.

Boca Raton, CRC Press, Florida (USA), 2004.

[26] Mena, E.; Rey, A.; Acedo, B.; Beltrán, F.J.; Malato, S. “On ozone‐

photocatalysis synergism in black‐light induced reactions: Oxidizing species

production in photocatalytic ozonation versus heterogeneous photocatalysis”.

Chem. Eng. J. 204‐206 (2012) 131‐140.



[28] Mvula, M.; Von Sonntag, C. “Ozonolysis of phenols in aqueous solution”.

Org. Biomol. Chem. 1 (2003) 1749‐1756.

[29] Leitzke, A.; Von Sonntag, C. “Ozonolysis of unsaturated acids in aqueous


Eng. 31 (2009) 301‐308.

[30] Johnson, P.N.; Davis, R.A. “Diffusivity of ozone in water”. J. Chem. Eng.

Data 41 (1996) 1485‐1487.

[31] Froment, G.F.; Bishoff, K.B. Chemical Reactor Analysis and Design. John

Willey & Sons. New York (USA), 1979.

[32] Huber, M.M.; Gobel, A.; Joss, A.; Hermann, N.; Loffler, D.; McArdell, C.;

Ried, A.; Siegrist, H.; Ternes, T.A.; Von Gunten, U. “Oxidation of

pharmaceuticals during ozonation of municipal wastewaters effluents: A pilot

study”. Environ. Sci. Technol. 39 (2005) 4290‐4299.

[33] Buffle, M.; von Gunten, U. “Phenols and amine induced HO∙ generation

during the initial phase of natural water ozonation”. Environ. Sci. Technol. 40


176

(2006) 3057‐3063.

[34] Sein, M.M.; Zedda, M.; Tuerk, J.; Schmidt, T.C.; Golloch, A., Von Sonntag, C.

“Oxidation of diclofenac with ozone in aqueous solution”. Environ. Sci. Technol.

42 (2008) 6656‐6662.



177

SUPPLEMENTARY INFORMATION OF CHAPTER 4

Figure 4.S1. High resolution XPS spectra of W4f spectral region for all the catalysts.

40 39 38 37 36 35 34 33 32

W 4f5/2

37.6

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W450

-30 W 4f7/2

35.5

2.1 eV

40 39 38 37 36 35 34 33 32

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W450

-5

W 4f5/2

37.6

W 4f7/2

35.5

2.1 eV

40 39 38 37 36 35 34 33 32

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W300

-5

W 4f5/2

37.5

W 4f7/2

35.4

2.1 eV

40 39 38 37 36 35 34 33 32

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W300

-3

W 4f5/2

37.6

W 4f7/2

35.5

2.1 eV

40 39 38 37 36 35 34 33 32

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W300

-1

W 4f5/2

37.4

W 4f7/2

35.3

2.1 eV

40 39 38 37 36 35 34 33 32

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W-0

W 4f5/2

37.4

W 4f7/2

35.2

2.2 eV


178

Figure 4.S2. High resolution XPS spectra of O1s spectral region for all the catalysts.

536 534 532 530 528 526

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W450

-30 O 1s530.2

536 534 532 530 528 526

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W450

-5 O 1s530.2

536 534 532 530 528 526

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W300

-5 O 1s530.2

536 534 532 530 528 526

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W300

-3 O 1s530.2

536 534 532 530 528 526

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W300

-1

H2O

O 1s530.2

536 534 532 530 528 526

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W-0

H2O

O 1s530.2



179

Figure 4.S3. Optical band gap energy determination according to Tauc’s expression.

2.0 2.2 2.4 2.6 2.8 3.00

1

2

3

4

5

(h)

2 (

eV

)2

h(eV)

2.67 eV

W450

-30

2.0 2.2 2.4 2.6 2.8 3.00

1

2

3

4

5

(h)

2 (

eV

)2

h (eV)

2.65 eV

W450

-5

2.0 2.2 2.4 2.6 2.8 3.00

1

2

3

4

5

(h)

2 (

eV)2

h (eV)

2.65 eV

W300

-5

2.0 2.2 2.4 2.6 2.8 3.00

1

2

3

4

5

(h)

2 (

eV)2

h (eV)

2.64 eV

W300

-3

2.0 2.2 2.4 2.6 2.8 3.00

1

2

3

4

5

(h)

2 (eV

)2

h (eV)

2.43 eV

W300

-1

2.0 2.2 2.4 2.6 2.8 3.00

1

2

3

4

5

(h)

2 (eV

)2

h (eV)

2.42 eV

W-0


180

200 300 400 500 600

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7S

IGN

AL

(u.a

.)

W AVELENGTH (nm )

Figure 4.S4. UV‐Vis spectrum of IBP.

Table 4.S1. Rate constants of direct O3‐ECs reactions (pH = 7).

EMERGING

CONTAMINANT ABBREVIATION

kO3

(M‐1 s‐1) Reference

Acetaminophen AAP 4.1x106 [1S]

Metoprolol MTP 1.4x103 [2S]

Caffeine CAF 6.5x102 [3S]

Hydrochlorothiazide HCT 5.1x103 [4S]

Antipyrine ANT 2.5x104 [5S]

Sulfamethoxazole SFM 4.2x105 [6S]

Carbamazepine CAR 3.0x105 [7S]

Ketorolac KET 3.4x105 [5S]

Diclofenac DCF 6.8x105 [8S]

Ibuprofen IBP 9.1 [9S]



181

References of Supplementary Information

[1S] Andreozzi, R.; Caprio, V.; Marotta, R.; Vogna, D.

“Paracetamol oxidation from aqueous solutions by means of ozonation and

H2O2/UV system”. Water Res. 37 (2003) 993‐1004.

[2S] Benitez, F.J.; Acero, J.L.; Real, F.J.; Roldán, G. “Ozonation of pharmaceutical

compounds: Rate constants and elimination in various water matrices”.

Chemosphere 77 (2009) 53‐59.

[3S] Broséus, R.; Vincent, S.; Aboulfadl, K.; Daneshvar, A.; Sauvé, S.; Barbeau, B.;

Prévost, M. “Ozone oxidation of pharmaceuticals, endocrine disruptors and

pesticides during drinking water treatment”. Water Res. 43 (2009) 4707‐4717.

[4S] Real, F.J.; Acero, J.L.; Benítez, F.J.; Roldán, G.; Fernández, L.C. “Oxidation of

hydrochlorothiazide by UV radiation, hydroxyl radicals and ozone: Kinetics and

elimination from water systems”. Chem. Eng. J. 160 (2010) 72‐78.

[5S] Rivas, F.J.; Sagasti, J.; Encinas, A.; Gimeno, O. “Contaminants abatement by

ozone in secondary effluents. Evaluation of second‐order rate constants”. J.

Chem. Technol. Biotechnol. 86 (2011) 1058‐1066.

[6S] Beltrán, F.J.; Aguinaco, A.; García‐Araya, J.F. “Mechanism and kinetics of


1369.

[7S] Huber, M.M.; Canonica, S.; Park, G.Y.; Von Gunten, U. “Oxidation of

pharmaceuticals during ozonation and advanced oxidation processes”. Environ.

Sci. Technol. 37 (2003) 1016‐1024.

[8S] Sein, M.M.; Zedda, M.; Tuerk, J.; Schmidt, T.C.; Golloch, A.; von Sonntag, C.

“Oxidation of diclofenac with ozone in aqueous solution”. Environ. Sci. Technol.

42 (2008) 6656‐6662.

[9S] Huber, M.M.; Gobel, A.; Joss, A.; Hermann, N.; Loffler, D.; McArdell, C.;


pharmaceuticals during ozonation of municipal wastewater effluents: A pilot


CAPÍTULO 5 (CHAPTER 5) PAPER 3: Visible light photocatalytic ozonation of DEET in the presence of different forms of WO3

E. Mena, A. Rey, S. Contreras, F.J. Beltrán

Catalysis Today 252 (2015) 100-106

ABSTRACT. This work deals with the use of WO3 materials for the removal of DEET, an

emerging contaminant, through photocatalytic ozonation using visible light as radiation

source. A commercial WO3 and own-made WO3 catalysts with different forms and

crystalline structure were synthetized by sol-gel and hydrothermal methods. After catalyst characterization with different techniques, photocatalytic ozonation under visible light radiation was found an efficient process to remove DEET. Monoclinic and/or

orthorhombic structure of WO3 seems to favor the catalytic activity more than a

developed surface area together with the presence of W reduced states in the presence of ozone. The best catalysts lead to a complete removal of DEET in less than 20 min with mineralization up to 70 % in 2 h.

Keywords: Photocatalytic ozonation, tungsten trioxide, DEET, ozone, visible light.

PAPER 3: Visible light photocatalytic ozonation

of DEET in the presence of different forms of WO3

185

5.1. INTRODUCTION

N,N‐diethyl‐meta‐toluamide (DEET) has been widely used as the active

compound in insect repellents for protection against insect bites at 4 ‐ 100 %

concentration in several formulations (lotions, gels, aerosols and sticks) [1].

Thus, DEET contamination has been detected in groundwater, rivers, seawater,

wastewater treatment plants effluents and even in drinking water treated by

conventional systems. This highlights its persistence and degree of recalcitrance

to conventional treatments [2,3]. Toxic effects of DEET to aquatic organisms are

relatively low (e.g. LC50 is in the range 75 ‐ 230 mg L‐1 for Daphnia magna and

Gambusia affinis) [3]. However, it has been reported to have potential

carcinogenic properties in human nasal mucosal cells [4], and additional

research is necessary to examine its fate, transport and ecotoxicity.

Different technologies have been applied for DEET removal such as

ozonation [4]; chlorination and bromination [5]; or AOPs (advanced oxidation

processes) such as heterogeneous photocatalysis with TiO2 [1,3,6]; Fenton and

photo‐Fenton [6,7]; UV/H2O2, O3/H2O2 and UV/O3/H2O2 [7]. The best

performance was obtained using photocatalytic treatments or O3/H2O2 with and

without UVC radiation.

Photocatalytic ozonation, i.e. the combination of ozone and heterogeneous

photocatalysis, has proved to be an efficient AOP improving the efficiency of the

single processes [8,9], and has been also successfully applied for DEET removal

using TiO2 as photocatalyst and UVC as radiation source [6]. However, the use

of UV lamps makes the process expensive for commercial applications so as the

use of solar light is a need for the practical deployment of photocatalytic

technologies. Also, TiO2 presents some limitations since it only uses around 5 %

of solar radiation (UV range) in spite of being the archetypical photocatalyst due

to its relatively high efficiency, low cost and availability [10].

As an attractive alternative to TiO2 for the photocatalytic ozonation process


186

under solar radiation, WO3 is a visible‐light‐responsive semiconductor of

relatively low cost, with no toxicity and availability, which has demonstrated its

catalytic activity in the degradation and mineralization of phenol through this

process [11]. Up to now, only monoclinic WO3 has been used as catalyst in

photocatalytic ozonation. Thus, this work is focused on the use of different

forms of WO3 as photocatalysts for the photocatalytic ozonation of DEET using

visible light radiation.



Ten different catalysts were tested, a WO3 commercial catalyst and nine own‐

made WO3 catalysts prepared by sol‐gel and hydrothermal methods. Sol‐gel

procedure was used to prepare WO3 microspheres as described elsewhere [12].

The method is based in the preparation of a CaWO4 precursor with CaCl2 and

Na2WO4 in the presence of citric acid at pH = 12. Once the precursor was

obtained, it was collected by centrifugation, washed with ultrapure water and

ethanol, and dried in air. Then, this precursor was soaked in HNO3 for 24 h at

room temperature followed by calcination at 500, 600 or 700 °C for 2 h. WO3

flocky‐microspheres were prepared by hydrothermal synthesis following a

modified procedure described elsewhere [13]. Typically, Na2WO4 was dissolved

in ultrapure water and acidified (pH = 1 ‐ 1.2) with HCl solution. After stirring

for 240 min, the system was mixed with ethanol 40 % (v/v) and then transferred

to a 125 mL autoclave heated at 120 °C for 48 h. Finally, the precipitates were

separated by centrifugation, washed with water and ethanol, and dried at 100 °C

for 2 h. Then this sample was calcined at 500, 600 or 700 °C for 2 h. Finally, all

fractions were treated at the reaction conditions (0.25 g L‐1 catalysts, 0.5 L

ultrapure water, 10 mg L‐1 O3 with 15 L h‐1 flow rate under visible radiation).

Catalysts nomenclature and synthesis conditions are summarized in Table 5.1.

PAPER 3: Visible light photocatalytic ozonation of DEET in the presence of different forms of WO3

187

Tabl

e 5.

1. P

rope

rtie

s of

WO

3 cat

alys

ts.

CA

TALY

ST

Synt

hesi

s W

O3

crys

talli

ne

phas

e L

(nm

) (I

808/I

716)R

AM

AN

SB

ET

(m2 g

-1)

Na/

WXP

S

(at/a

t) C

a/W

XPS

(at/a

t) C

/WXP

S

(at/a

t)

W-c

C

omm

erci

al

Mon

oclin

ic

28.2

7 1.

97

9.8

---

---

0.42

W1-

500

Sol-g

el

T =

500

°C

Ort

horh

ombi

c 21

.99

2.19

9.

0 0.

16

0.39

0.

36

W1-

600

Sol-g

el

T =

600

°C

Mon

oclin

ic

20.8

2 1.

98

8.1

0.10

0.

42

0.27

W1-

700

Sol-g

el

T =

700

°C

Mon

oclin

ic

32.9

8 1.

96

5.3

0.08

0.

35

0.35

W2

Hyd

roth

erm

al

Hex

agon

al

13.1

6 ---

82

.0

0.38

---

0.

12

W2-

t H

ydro

ther

mal

Tr

eate

d (O

3+hν

) H

exag

onal

10

.99

---

93.7

0.

22

---

0.06

W2-

500-

t H

ydro

ther

mal

T

= 50

0 °C

Tr

eate

d (O

3+hν

) M

onoc

linic

28

.27

2.01

8.

2 0.

17

---

0.10

W2-

600

Hyd

roth

erm

al

T =

600

°C

Mon

oclin

ic

35.9

6 1.

99

5.8

0.34

---

0.

15

W2-

600-

t H

ydro

ther

mal

T

= 60

0 °C

Tr

eate

d (O

3+hν

) M

onoc

linic

35

.96

1.98

9.

6 0.

13

---

0.09

W2-

700-

t H

ydro

ther

mal

T

= 70

0 °C

Tr

eate

d (O

3+hν

) M

onoc

linic

32

.97

1.95

9.

4 0.

21

---

0.06


188


X‐ray diffraction (XRD) patterns were recorded using a powder Bruker D8

Advance XRD diffractometer with a Cu Kα radiation (λ = 0.1541 nm). The data

were collected from 2θ = 10° ‐ 70° at a scan rate of 0.02 s−1. Raman spectra were

obtained using a Nicolet Almega XR Dispersive micro‐Raman (Thermo

Scientific) with a spectral resolution of 2 cm‐1. The samples were excited with a

633 nm laser. BET surface area was determined from nitrogen adsorption‐

desorption isotherms obtained at ‐196 °C using an Autosorb 1 apparatus

(Quantachrome). The samples were outgassed at 250 °C for 12 h under high

vacuum (< 10−4 Pa). A Hitachi S‐48000 scanning electron microscope with 20 ‐ 30

kV accelerating voltage and 500 ‐ 2000 magnification was used to study the

morphology of the samples. X‐ray photoelectron spectra (XPS) were obtained

with a Kα Thermo Scientific apparatus with an Al Kα (h = 1486.68 eV) X‐ray

source using a voltage of 12 kV under vacuum (2x10−7 mbar). Binding energies

were calibrated relative to the C1s peak at 284.6 eV.


DEET was used as target compound to test the catalytic activity of the

photocatalysts. The chemical structure and UV‐Vis spectrum of DEET are shown

in Figure 5.S1 of the supplementary information. Photocatalytic experiments

were carried out in semi‐batch mode in a 0.5 L glass‐made spherical reactor,

provided with a gas inlet, a gas outlet and a liquid sampling port. The reactor

was placed in a solar simulator (Suntest CPS, Atlas) provided with a 1500 W Xe

lamp with emission restricted to visible light (λ > 390 nm) using different cut‐off

filters. The irradiation intensity was 550 W m−2, incident photon flux in the

reactor was 3.51x10‐4 einstein min‐1 (actinometrical measurements), and the

temperature maintained between 20 ‐ 40 °C. If required, an ozone generator

(Anseros Ozomat Com AD‐02) was used to produce a gaseous ozone‐oxygen

stream that was fed to the reactor.



189

In a typical photocatalytic ozonation experiment, the reactor was loaded with

0.5 L of an aqueous solution containing 5 mg L‐1 of DEET (pH0 = 6). Then, 0.125 g

of catalyst were added and the suspension was stirred in the darkness for 30

min. After that, the lamp was switched on and a mixture of ozone‐oxygen (10

mg L‐1 ozone concentration) was fed to the reactor at 15 L h−1 flow rate. The

irradiation time for each experiment was 2 h. Samples were withdrawn from the

reactor and filtered through a 0.2 μm PET membrane except for dissolved ozone

analysis. Experiments of adsorption, photolysis, photocatalytic oxidation,

ozonation and catalytic ozonation were also carried out for comparative reasons.

DEET concentration was analyzed by HPLC‐DAD (Hitachi, Elite LaChrom)

using a Phenomenex C‐18 column (5 μm, 150 mm long, 3 mm diameter) and 0.6

mL min−1 of acetonitrile‐acidified water (0.1 % formic acid) as mobile phase (30 ‐

70 v/v, isocratic). Identification and quantification was carried out at 220 nm.

Total organic carbon (TOC) was measured using a Shimadzu TOC‐VSCH

analyzer. Aqueous ozone was photometrically determined by the indigo method

at 600 nm [14]. Hydrogen peroxide concentration was analyzed photometrically

by the cobalt/bicarbonate method, at 260 nm [15]. Photometric analyses were

performed in a UV‐Vis spectrophotometer Evolution 201 (Thermospectronic).

Ozone in the gas phase was continuously monitored by means of an Anseros

Ozomat GM‐6000Pro analyzer. Short‐chain organic acids were analyzed by ion

chromatography with chemical suppression (Metrohm 881 Compact Pro) with a

conductivity detector using a MetroSep A sup 5 column (250 mm long, 4 mm

diameter) at 45 °C and 0.7 mL min‐1 of Na2CO3 from 0.6 ‐ 14.6 mM in 50 min (10

min post‐time for equilibration) as mobile phase.


5.3.1. Photocatalysts characterization

Figure 5.1(A) shows the XRD patterns of the WO3 photocatalysts. All the

diffraction peaks of commercial WO3, W‐c, are indexed to monoclinic phase.


190

W1‐500 catalyst presented mainly WO3 orthorhombic structure but this phase

was transformed to monoclinic at calcination temperatures of 600 and 700 °C

(W1‐600 and W1‐700). Monoclinic and orthorhombic phases differ only in the

extent of the W atoms displacement and can transform reversibly into each other

[16]. Also, the two peaks located at 18.5° and 28.7° are indicative of the

tetragonal phase of CaWO4 [12]. On the other hand, W2 synthetized by

hydrothermal treatment was crystalized as hexagonal WO3 structure. After

being washed at the reaction conditions, no modifications were observed in the

XRD patterns of non‐calcined samples (Figure 5.S2). However, calcination

produced the transformation into monoclinic WO3 as can be observed in Figure

5.1 for treated samples. In addition, Na phases as Na2W4O13 are detected in these

calcined catalysts with main diffraction peaks at 10.8°, 21.8° and 32.9° [17],

indicating an unsatisfactory washing procedure. After being treated at reaction

conditions, the relative intensity of these peaks significantly decreased as can be

noticed in Fig. 5.S2 for W2‐600. The crystal size of the WO3 phases was

determined by the Scherrer’s equation (Table 5.1). In general, a growing in the

crystal size is observed as the calcination temperature increased for catalysts

with the same structure. In addition, no significant changes were observed due

to the washing procedure in W2 catalysts.

Raman spectra depicted in Fig. 5.1(B) and Figure 5.S3 (untreated W2

samples) confirmed the XRD results. The bands located at 808 and 716 cm‐1

correspond to O‐W‐O stretching vibrations whereas the peaks at 328 and 270

cm‐1 were identified as O‐W‐O deformation vibrations [16,18]. Since the main

bands of monoclinic and orthorhombic structures are similar, the intensity ratio

of the main Raman peaks allows to distinguish them being around 2.2 for

orthorhombic WO3 and 1.97 for monoclinic [18]. The intensity ratio of the peaks

located at 808 and 716 cm‐1 is presented in Table 5.1 confirming the

orthorhombic structure of W1‐500 catalyst and its transformation into

monoclinic WO3 at higher calcination temperatures. W2 and W2‐t catalysts



191

presented a wide multiple band with a maximum at 799 cm‐1 which can be

composed by the main bands of hexagonal O‐W‐O stretching vibrations at 789,

693 and 651 cm‐1 [16]. The calcination of this catalyst gave place to the

transformation of hexagonal WO3 into monoclinic phase according to the main

characteristic bands and their intensity ratio shown in Table 5.1 (W2 calcined

samples). Finally, the washing procedure of the treated W2 samples did not

produce any significant modification in the WO3 structure (Fig. 5.S3).

Figure 5.1. Structural characterization of WO3 catalysts. (A) XRD patterns (M: Monoclinic

WO3, O: Orthorhombic WO3, H: Hexagonal WO3, C: CaWO4, N: Na2W4O13). (B) Raman

spectra.

(A) (B)

200 300 400 500 600 700 800 900

W2-700-t

INT

EN

SIT

Y (

a.u.

)

RAMAN SHIFT (cm-1)

W2-t

W2-500-t

W2-600-t

710

799

803

805

808

717273328

W-c

W1-500

W1-600

W1-700

807

716

712

330275

269 324

10 20 30 40 50 60 70

W2-700-t

W2-500-t

W1-700

W1-600

W2-600-t

W2-t

W1-500

INT

EN

SIT

Y (

a.u.

)

2

W-c

M

M M

M

M M M

C C O

H H H

H H H

N N N


192

The morphology of the samples was evaluated by SEM (Figure 5.2).

Commercial WO3 presented an irregular morphology whereas the formation of

spherical particles was confirmed in sol‐gel synthetized catalyst W1‐500 with a

heterogeneous distribution in the range 1 ‐ 5 μm. An increase in the calcination

temperature led to the progressive distortion‐breakup of the microspheres as

confirmed for W1‐600 and W1‐700 in Fig. 5.2(C) and Fig. 5.2(D), respectively. On

the other hand, hydrothermal synthetized catalyst presented mainly irregular

morphology with some isolated spherical particles (Fig. 5.2(E)). An image

magnification in Fig. 5.2(F) revealed a dense packing of small WO3 needles

formed on the surface of WO3 particles [13]. Calcination of this sample gave

place to a more irregular morphology with an enlargement of the needles

transforming them into small WO3 particles (Fig. 5.2(G) and Fig. 5.2(H)). Finally,

the washing treatment did not apparently change the morphology of the W2

type samples.

Specific surface areas (SBET) were calculated from N2 adsorption‐desorption

isotherms and are summarized in Table 5.1. The W1 series presented low SBET

values that decreased with the increasing calcination temperature. For similar

crystalline structures this is in agreement with the growing of the crystal size.

These values are quite similar to that of commercial W‐c. On the other hand, the

hydrothermal treatment gave place to catalysts with higher SBET (82.0 m2 g‐1) due

to the morphology and the lower crystal size of W2 samples. However,

calcination provoked not only a crystalline phase transformation but also a very

sharp decrease of the SBET. In addition, it can be noticed that the washing

procedure in W2 samples led to an increase of the SBET to some extent. Taking

into account that the crystal size is not modified, this effect can be related to the

elimination of some impurities blocking the surface of the samples.



193

Figure 5.2. SEM images of WO3 catalysts.

5 um

W-c W1-500

W1-600 W1-700

W2-600 W2-600-t

W2

(A)

(H)(G)

(F)(E)

(D)

(C)

(B)

W2


194

Surface chemical composition and oxidation state of W in the catalysts were

analyzed by XPS. The full spectra of the catalysts (not shown) presented the

main contributions of W4f and O1s. The presence of Na1s was detected in all the

synthetized catalysts and Ca2p in W1 series. In addition C1s contribution was

detected in all samples. The atomic surface ratios of Na, Ca and C to W are

summarized in Table 5.1. For the W1 series, similar values of Ca/W and C/W are

obtained whereas Na/W was lower for the W1‐700 catalyst. These contributions

are due to the precursors used during the synthesis. The most noticeable results

of W2 series are related to the amount of Na in the catalyst surface. This ratio

highly decreased when the catalyst was submitted to the washing procedure

thus indicating that some Na species were released to water. These results are in

agreement to XRD patterns of treated samples. It can be also observed a

decrease on the surface carbon content after the treatment in all the W2 samples.

On the other hand, W4f high resolution spectral region for W‐c, W1 and W2

series are depicted in Figure 5.3. For commercial W‐c and W2 series, two

different peaks corresponding to W4f7/2 and W4f5/2 appear at 35.9 and 38.0 eV,

respectively, which are characteristic of W6+ [16,19]. No significant differences

were found in W2 and W2 treated catalysts (Figure 5.S4). However, the

broadening of these peaks together with their shift to lower binding energies are

indicative of the presence of reduced W5+ to some extent in W1 catalysts [16,19].

This effect is more pronounced in the W1‐700 catalyst as observed in Fig. 5.3.

The partial reduced state of W1 catalysts could be related to the presence of

citric acid from the synthesis procedure that was not completely eliminated

during the washing step, thus acting as a reducing agent. The presence of W5+ in

W1 catalysts could introduce some differences in the electron transport and

charge recombination during photocatalytic processes.



195

41 40 39 38 37 36 35 34 33 32

W2-700-t

W2-600-t

W2-500-t

W2-t

W1-700

W1-600

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W-c

W1-500

38.0 35.9

37.635.4

37.2 35.3

W 4f

Figure 5.3. High resolution XPS spectra of W4f spectral region of WO3 catalysts.

5.3.2. Catalytic activity

The effectiveness of the catalysts in the photocatalytic ozonation process was

tested using DEET under visible light radiation. First, the performance of this

combined process respect to the single treatments has been checked with all the

catalysts. Results for W‐c catalyst are shown in Figure 5.4. In general, the same

behavior was observed for all the synthetized catalysts. Results for W1‐600 and

W2‐600‐t as an example can be found in the supplementary information (Figures


196

5.S5‐5.S8). Fig. 5.4(A) shows the time evolution of DEET dimensionless

concentration upon the different treatments applied. Adsorption and photolysis

showed negligible effect on DEET concentration (lower than 5 % removal). The

absence of DEET depletion due to direct photolysis is consistent with the UV‐Vis

absorption spectrum of this compound (Fig. 5.S1). Photocatalytic oxidation gave

place to a slow degradation of DEET, reaching 22 % depletion after 2 h.

Although WO3 is a visible light responsive semiconductor, in photocatalytic

oxidation it presents a low efficiency as a consequence of the high recombination

rate in the presence of oxygen, an unsuitable oxidant for WO3. This drawback is

overcome in the presence of O3 which can easily react with photoexcited

electrons in the conduction band of WO3 [11]. All the ozone treatments showed

the highest degradation rate of DEET. On the one hand, single ozonation and

catalytic ozonation gave place to similar DEET evolution thus indicating that in

dark conditions W‐c did not present any catalytic effect in ozone decomposition

into more reactive species. On the other hand, DEET degradation rate was

highly increased during photocatalytic ozonation reaching more than 99 %

removal in less than 20 min compared to 60 min necessary for the other O3

treatments. In general, pH0 = 6 evolved to pH ~ 4 during the reaction time in all

the treatments (except for direct photolysis), so that in ozone systems, it is

expected a negligible contribution of indirect reactions due to O3 decomposition

at these pH values [20].

On the other hand, it is known that the oxidation of DEET proceeds through

different steps giving place to intermediate compounds which are eventually

transformed into CO2 and H2O (complete mineralization) [4]. DEET

mineralization has been followed as normalized TOC concentration in Fig.

5.4(B). Adsorption, direct photolysis and photocatalytic oxidation led to

negligible TOC removal according to previous DEET evolution. Regarding the

O3‐treatments, single ozonation led to only 19 % mineralization at 120 min, as a

consequence of the formation of some intermediate compounds that are

refractory to ozone direct reaction (mainly short‐chain organic acids) [20].



197

Oxalic, formic and acetic acids were detected at concentrations of 2.1, 0.6 and 2.7

mg L‐1, respectively, at the end of the reaction. The presence of the catalyst

combined with ozone in dark conditions did not improve the previous results

confirming the absence of any positive catalytic effect in the reaction. On the

contrary, photocatalytic ozonation showed the highest mineralization rate,

leading to 60 % mineralization in 120 min reaction time. These results also point

out the synergism produced between ozone and visible light irradiated WO3

since final TOC removal of the combined process is much higher than the sum

of these values for the individual processes (60 % in O3/WO3/Vis vs. 4 % in

O2/WO3/Vis and 19 % in O3), also confirming the ability of O3 to capture the

electrons on the WO3 surface compared to O2 [11], and the formation of higher

concentration of oxidizing species such as •HO radicals as occurred in TiO2 [8,9].

Dissolved ozone (Fig. 5.4(C)) and hydrogen peroxide concentration (Fig.

5.4(D)) during the treatments confirmed these results. H2O2 is commonly formed

through direct ozone reactions [21], and was experimentally confirmed in this

work for DEET. The profiles obtained for dissolved O3 and H2O2 during the

photocatalytic ozonation showed a faster decomposition of both compounds

with respect to ozonation and catalytic ozonation. This is indicative of their

consumption on the irradiated WO3 surface through photocatalytic reactions,

probably acting as electron acceptors and thus enhancing the production of

oxidizing species as also occurs in TiO2 [8]. The mechanism and kinetics of the

reaction will be the subject of further work.


198

Figure 5.4. (A

) DEET dim

ensionless concentration, (B) TOC dim

ensionless concentration, (C) dissolved

O3 concentration and (D) H

2O2 concentration evolution during all the treatm

ents applied w

ith W‐c

catalyst. E

xperim

ental conditions: pH

0 = 6, T = 20 ‐ 40 °C, C

DEET,0 = 5 m

g L‐1, CWO3 = 0.25 g L‐1, C

O3ginlet = 10

mg L‐1, Q

g = 15 L h‐1 (O

2 or O

3/O

2).

(A)

(B)

(C)

(D)

-20

020

40

6080

100

120

0.0

1.0x

10-5

2.0x

10-5

3.0x

10-5

CO3 (M)

Dar

k

TIM

E (

min

)

-20

02

040

6080

100

120

CH2O2 (M)

Dar

k

TIM

E (

min

)

0.0

1.0x

10-5

2.0x

10-5

3.0x

10-5

4.0x

10-5

TOC/TOC0

Dar

k

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

Dar

k

Ad

sorp

tion

Pho

toly

sis

Pho

toca

taly

tic o

xid

atio

n O

zon

atio

n C

ata

lytic

ozo

nat

ion

Pho

toca

taly

tic o

zona

tion

CDEET/CDEET,0

CTOC/CTOC,0



199

The comparison of the activity and efficiency of the different catalysts during

photocatalytic ozonation is depicted in Figure 5.5. For the W2 series, only

treated catalysts have been selected for this study since the catalysts tested

before washing presented a poor activity as can be observed in Figure 5.S9 of the

supplementary material. This could be due to a negative effect of the precursors

that still remained in the catalysts surface and were reduced after the treatment

(Na and C according to XRD and XPS results).

Regarding the evolution of DEET in Fig. 5.5(A) all the catalysts gave place to

higher DEET depletion rate compared with single ozonation. The highest DEET

degradation rate was observed for W1‐600 and W1‐700 catalysts, reaching a

complete removal in 10 min.

Main differences in the activity of the catalysts were found in terms of

mineralization as shown in Fig. 5.5(B). All the catalysts gave place to higher

mineralization degree than ozone alone. Commercial W‐c catalyst led to a 60 %

mineralization at 120 min reaction time similarly to W1‐500, showing an

intermediate activity compared to the synthetized catalysts. Both catalysts

presented similar textural properties but different crystalline structure although

orthorhombic WO3 is a distortion of the monoclinic structure, being both very

similar [16,22]. On the other hand, W1‐600 and W1‐700 catalysts, with

monoclinic structure, were the most active ones but TOC profiles were very

different. W1‐600 catalyst led to a progressive mineralization reaching a 70 % in

2 h. On the contrary, with W1‐700 catalyst, it can be noticed a first step in which

mineralization is slow up to 45 min and, after this time, the mineralization rate

increased also achieving around 70 % at 120 min. This behavior can be due to

the most reduced state of W1‐700 catalyst. The highest catalytic activity could be

related to the presence of reduced W5+ states according to XPS results that can

favor the electron transport in the WO3 structure [19]. These results are contrary

to those previously found for photocatalytic oxidation where a more oxidized

material had a higher catalytic activity in the presence of oxygen and reported

that W5+/W4+ states could act as recombination centers [16]. However, the


200

presence of O3 could take the advantage of avoiding the recombination to some

extent and thus benefits from the higher electron mobility. For W2 series,

mineralization rate of W2‐t catalyst was noticeably slower regardless of its high

SBET, indicating that the hexagonal structure of this catalyst could not be

appropriated for the process. This is evidenced by the performance of calcined

catalysts (with monoclinic structure and very low SBET) similar or even better

than that of hexagonal W2‐t. It can be noticed the highest catalytic activity of

W2‐600‐t catalyst in terms of mineralization compared to W2‐500‐t and W2‐700‐t

catalysts whose main difference is the higher crystal size of monoclinic phase in

W2‐600‐t.

Finally, the best performance of the W1 series can be observed in the

evolution of short‐chain organic acids formed during the process. Fig. 5.5(C)

represents the sum of the concentrations of main organic acids detected (oxalic,

acetic and formic) in terms of mg C L‐1. The degradation of these intermediates

is only observed for W1 series catalyst.



201

-20 0 20 40 600.0

0.2

0.4

0.6

0.8

1.0 W-c W1-500 W1-600 W1-700 W2-t W2-500-t W2-600-t W2-700-tDark

CD

EE

T/C

DE

ET

,0

TIME (min)

Figure 5.5. (A) DEET dimesionless concentration, (B) TOC dimensionless concentration

and (C) sum of the TOC corresponding to short‐chain organic acids (ΣCTOC,a) evolution

during photocatalytic ozonation with the best WO3 catalysts (dotted lines show single

ozonation results). Experimental conditions: pH0 = 6, T = 20 ‐ 40 °C, CDEET,0 = 5 mg L‐1,

CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 15 L h‐1 (O3/O2).

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Dark

CT

OC

/CT

OC

,0

TIME (min)

(A)

(B)

0 20 40 60 80 100 1200.0

0.4

0.8

1.2

1.6

2.0

CT

OC

,a (

mg

L-1)

TIME (min)

(C)


202

5.4. CONCLUSIONS

WO3 catalysts were synthetized with different morphology (spherical and

irregular) and structure (orthorhombic, monoclinic and hexagonal) through sol‐

gel and hydrothermal procedures. An increase in the calcination temperature

produced the transformation of orthorhombic or hexagonal into monoclinic

structure leading to higher crystal sizes and a less developed surface area. The

presence of citric acid in the sol‐gel synthetized catalysts led to the presence of

W reduced states to some extent. The catalysts were active in the photocatalytic

ozonation of DEET with visible light radiation. A synergistic effect between

ozone and visible light irradiated WO3 was produced leading to a higher

efficiency in the combined process. Monoclinic and/or orthorhombic structure of

WO3 favored the catalytic activity with respect to hexagonal structure, more

than the developed surface area. Also the presence of W reduced states seems to

be beneficial for the process in the presence of ozone. The best catalysts gave

place to a complete removal of DEET in less than 20 min with mineralization up

to 70 % in 2 h.

AKNOWLEDGEMENTS

Authors thank the Spanish MINECO and European Feder Funds (CTQ2012‐

35789‐C02‐01) for economic support. E. Mena thanks the Consejería de Empleo,

Empresa e Innovación (Gobierno de Extremadura) and European Social Fund

for her predoctoral FPI grant (Ref. PD12059).

REFERENCES

[1] Antonopoulou, M.; Giannakas, A.; Deligiannakis, Y.; Konstantinou, I.

“Kinetic and mechanistic investigation of photocatalytic degradation of the N,N‐

diethyl‐m‐toluamide”. Chem. Eng. J. 231 (2013) 314‐325.

[2] Costanzo, S.D.; Watkinson, A.J.; Murby, E.J.; Kolpin, D.W.; Sandstrom, M.W.

“Is there a risk associated with the insect repellent DEET (N,N‐diethyl‐m‐

toluamide) commonly found in aquatic environments?”. Sci. Total Environ. 384

(2007) 214‐220.



203

[3] Adams, W.A.; Impellitteri, C.A. “The photocatalysis of N,N‐diethyl‐m‐

toluamide (DEET) using dispersions of Degussa P‐25 TiO2 particles”. J.

Photochem. Photobiol. A Chem. 202 (2009) 28‐32.

[4] Benítez, F.J.; Acero, J.L.; García‐Reyes, J.F.; Real, F.J.; Roldán, G.; Rodríguez,

E.; Molina‐Díaz, A. “Determination of the reaction rate constants and

decomposition mechanisms of ozone with two model emerging contaminants;

DEET and Nortriptyline”. Ind. Eng. Chem. Res. 52 (2013) 17054‐17073.

[5] Acero, J.L.; Benítez, F.J.; Real, F.J.; Roldán, G.; Rodríguez, E. “Chlorination

and bromination kinetics of emerging contaminants in aqueous systems”. Chem.

Eng. J. 219 (2013) 43‐50.

[6] Benítez, F.J.; Acero, J.L.; Real, F.J.; Roldán, G.; Rodríguez, E. “The

effectiveness of single oxidants and AOPs in the degradation of emerging

contaminants in waters: A comparison study”. Ozone Sci. Eng. 35 (2013), 263‐

272.

[7] Li, W.; Nanaboina, V.; Zhou, Q.; Korshin, G.V. “Effects of Fenton treatment

on the properties of effluent organic matter and their relationships with the

degradation of pharmaceuticals and personal care products”. Water Res. 46

(2012) 403‐412.




Chem. Eng. J. 204‐206 (2012) 131‐140.




[10] Hernández‐Alonso, M.D., Fresno, F., Suárez, S., Coronado, J.M.



[11] Nishimoto, S., Mano, T., Kameshima, Y., Miyake, M. “Photocatalytic water



[12] Zhang, L.; Tang, X.; Lu, Z.; Wang, Z.; Li, L.; Xiao, Y. “Facile synthesis and

photocatalytic activity of hierarchical WO3 core‐shell microspheres”. Appl. Surf.

Sci. 258 (2011) 1719‐1724.

[13] Shen, Y.; Ding, D.; Deng, Y. “Fabrication and characterization of WO3 flocky

microspheres induced by ethanol”. Powder Technol. 211 (2011) 114‐119.


204


method”. Water Res. 15 (1981) 449‐456.


of residual hydrogen peroxide”. Water & Sewage Works (1977) 69‐72.

[16] Szilágyi, I.M.; Fórizs, B.; Rosseler, O.; Szegedi, A.; Németh, P.; Király, P.;

Tárkányi, G.; Vajna, B.; Varga‐Josepovits, K.; László, K.; Tóth, A.L.; Baranyai, P.;

Leskelä, M. “WO3 photocatalysts: Influence of structure and composition”. J.

Catal. 294 (2012) 119‐127.

[17] Sheng, T.; Chavvakula, P.; Cao, B.; Yue, N.; Zhang, Y.; Zhang, H. “Growth

of ultra‐long sodium tungsten oxide and tungsten oxide nanowires: Effects of

impurity and residue deposition”. J. Crystal Growth 395 (2014) 61‐67.

[18] Kustova, G.N.; Chesalov, Y.A.; Plyasova, L.M.; Molina, I.Y.; Nizovskii, A.I.

“Vibrational spectra of WO3∙nH2O and WO3 polymorphs”. Vibrational Spect. 55

(2011) 235‐240.

[19] Lee, J.S.; Jang, I.H.; Park, N.G. “Effects of oxidation state and crystallinity of

tungsten oxide interlayer on photovoltaic property in bulk hetero‐junction solar

cell”. J. Phys. Chem. C 116 (2012) 13480‐13487.

[20] F.J. Beltrán. Ozone reaction kinetics for water and wastewater systems. Boca

Raton, CRC Press, 2004, Florida (USA).

[21] Leitzke, A.; Von Sonntag, C. “Ozonolysis of unsaturated acids in aqueous


Eng. 31 (2009) 301‐308.

[22] Pecquenard, B.; Lecacheux, H.; Livage J.; Julien, C. “Orthorhombic WO3

Formed via a Ti‐Stabilized WO3∙13H2O Phase”. J. Solid State Chem. 135 (1998)

159‐168.



205

200 300 400 500 600

0.0

0.5

1.0

1.5

2.0

2.5

3.0

SIG

NA

L (

u.a

.)

WAVELENGTH (nm)

10 20 30 40 50 60 70

W2-600-t

W2-600

W2-t

W2

INT

EN

SIT

Y (

a.u.

)

2

SUPPLEMENTARY INFORMATION OF CHAPTER 5

Figure 5.S1. UV‐Vis spectrum and chemical structure of DEET.

Figure 5.S2. XRD patterns of non‐treated and treated WO3 catalysts.


206

.

200 300 400 500 600 700 800 900

INT

EN

SIT

Y (

a.u.

)

RAMAN SHIFT (cm-1)

W2

W2-t

W2-600

W2-600-t

805

273

328

716

799

Figure 5.S3. Raman spectra of non‐treated and treated WO3 catalysts.



207

41 40 39 38 37 36 35 34 33 32

W2-600-t

W2

W2-t

INT

EN

SIT

Y (

a.u.

)

BINDING ENERGY (eV)

W2-60038.0 35.9

W 4f

Figure 5.S4. High resolution XPS spectra of W4f spectral region of non‐treated and

treated WO3 catalysts.


208

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Adsorption Photolysis Photocatalytic oxidation Ozonation Catalytic ozonation Photocatalytic ozonation

Dark

CD

EE

T/C

DE

ET

,0

TIME (min)

Figure 5.S5. (A) DEET and (B) TOC dimensionless concentration evolution during all the

treatments applied with W1‐600 catalyst. Experimental conditions: pH0 = 6, T = 20 ‐ 40 °C,

CDEET,0 = 5 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 15 L h‐1 (O2 or O3/O2).

(A)

(B)

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Dark

TO

C/T

OC

0

TIME (min)

0

0

CT

OC

/CT

OC

,0



209

0 20 40 60 80 100 1200.0

1.0x10-5

2.0x10-5

3.0x10-5

CO

3 (M

)

TIME (min)

0 20 40 60 80 100 1200.0

1.0x10-5

2.0x10-5

3.0x10-5

4.0x10-5

5.0x10-5

6.0x10-5 Photocatalytic oxidation Ozonation Catalytic ozonation Photocatalytic ozonation

CH

2O2

(M)

TIME (min)

Figure 5.S6. (A) Dissolved O3 concentration and (B) H2O2 concentration during all the

treatments applied with W1‐600 catalyst. Experimental conditions: pH0 = 6, T = 20 ‐ 40 °C,

CDEET,0 = 5 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 15 L h‐1 (O2 or O3/O2).

(A)

(B)


210

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Adsorption Photolysis Photocatalytic oxidation Ozonation Catalytic ozonation Photocatalytic ozonationDark

CD

EE

T/C

DE

ET

,0

TIME (min)

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Dark

CT

OC

/CT

OC

,0

TIME (min)

Figure 5.S7. (A) DEET and (B) TOC dimensionless concentration evolution during all the

treatments applied with W2‐600‐t catalyst. Experimental conditions: pH0 = 6, T = 20 ‐ 40

°C, CDEET,0 = 5 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 15 L h‐1 (O2 or O3/O2).

(A)

(B)



211

0 20 40 60 80 100 1200.0

1.0x10-5

2.0x10-5

3.0x10-5

CO

3 (M

)

TIME (min)

0 20 40 60 80 100 1200.0

1.0x10-5

2.0x10-5

3.0x10-5

4.0x10-5

5.0x10-5

6.0x10-5 Photocatalytic oxidation Ozonation Catalytic ozonation Photocatalytic ozonation

CH

2O2

(M)

TIME (min)

Figure 5.S8. (A) Dissolved O3 concentration and (B) H2O2 concentration during all the

treatments applied with W2‐600‐t catalyst. Experimental conditions: pH0 = 6, T = 20 ‐ 40

°C, CDEET,0 = 5 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 15 L h‐1 (O2 or O3/O2).

(A)

(B)


212

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Dark

W2 W2-t W2-600 W2-600t

CT

OC

/CT

OC

,0

TIME (min)

Figure 5.S9. TOC dimensionless concentration evolution during photocatalytic ozonation

with untreated and treated W2 catalysts. Experimental conditions: pH0 = 6, T = 20 ‐ 40 °C,

CDEET,0 = 5 mg L‐1, CWO3 = 0.25 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 15 L h‐1 (O3/O2).

CAPÍTULO 6 (CHAPTER 6) PAPER 4: Nanostructured CeO2 as catalysts for different AOPs based in the application of ozone and simulated solar radiation

E. Mena, A. Rey, E.M. Rodríguez, F.J. Beltrán

Catalysis Today 280 (2017) 74-79

ABSTRACT. Two CeO2 catalysts with different morphology (nanocubes and nanorods)

were synthetized by hydrothermal treatment, characterized by TEM, XRD, N2

adsorption-desorption isotherms, XPS and DR-UV-Vis spectroscopy and examined for different AOPs (photocatalytic oxidation, catalytic and photocatalytic ozonation) using simulated solar radiation (λ > 300 nm) or visible radiation (λ > 390 nm) from a Xe lamp, and DEET as target compound. Neither significant DEET removal nor mineralization was achieved by solar photocatalysis and the catalysts did not show an important beneficial effect during catalytic ozonation. The best results were obtained in photocatalytic ozonation process being nanorods more active under visible radiation according to its lower band gap energy whereas nanocubes presented the highest intrinsic photocatalytic

activity under visible and solar radiation. Photocatalytic ozonation of DEET with CeO2

catalysts proceeds through a complex reaction mechanism that involves at least ozone-direct reaction that triggers a radical chain mechanism, indirect ozone reactions, ozone photolytic and photocatalytic decomposition and hydrogen peroxide formation-decomposition.

Keywords: Photocatalytic ozonation, tungsten trioxide, DEET, ozone, visible light.

PAPER 4: Nanostructured CeO2 as catalysts for different

AOPs based in the application of ozone and simulated solar radiation

215

6.1. INTRODUCTION

Water pollution due to specific organic contaminants (phenolic compounds,

aromatic hydrocarbons, pesticides, pharmaceuticals, etc.) is a growing problem

that requires the implementation of advanced technologies capable of removing

not only the initial compounds but also their metabolites. Among the available

technologies, advanced oxidation processes (AOPs) have demonstrated to

efficiently remove a wide variety of organic pollutants due to the generation of

oxidizing species, mainly hydroxyl radicals, which non‐selectively attack the

organic matter [1]. Photocatalytic treatments are promising alternatives due to

the possibility of using natural solar light as a radiation source to minimize

treatment costs. Besides, the combination of photocatalysis with ozone

(photocatalytic ozonation) increases the generation of oxidizing species leading

to higher degrees of mineralization than individual treatments of ozonation or

photocatalysis [2‐4].

For the photocatalytic processes, much of the research effort is focused on the

development of photocatalysts capable of absorbing visible light effectively in

order to increase the fraction of usable solar radiation compared to the

commonly used TiO2 [5,6]. In this sense, cerium oxide CeO2 is a wide band‐gap

semiconducting material usually in the range of 2.9 and 3.2 eV which absorbs

light in the near UV and slightly in the visible region depending on the

morphology and particle size, being its properties as catalyst and catalytic

support highly shape‐size dependent [7‐9]. In addition, CeO2 possesses a

sufficiently long lifetime of photo‐generated electron‐hole pairs to trigger

photocatalytic reactions in liquid and gas phase [10]. Bare CeO2 with different

morphologies/properties (microspheres, nanospheres, nanosheets, etc.) has been

used as photocatalyst in the degradation of organic pollutants in water, most of

them belonging to the dyes family [11‐15] or phenol and phenol derivatives

[8,16]. These works pointed out that the morphology and crystal size of CeO2 are

highly important in the amount and nature of surface defects, oxygen vacancies,

CeO2‐light interaction and, definitely, in the behavior of CeO2 as photocatalyst.


216

On the other hand, CeO2 has demonstrated to be catalytically active during

ozonation of organic pollutants of different nature such as phenol [17], oxalic

and oxamic acids [18] or bezafibrate [19]. Main reaction pathways proposed are

related to the adsorption and subsequent decomposition of ozone into hydroxyl

radicals, being the presence of Ce(III)/oxygen vacancies in the surface the main

responsible of the catalytic activity [17,18].

As far as we know, the combination of photocatalytic oxidation with ozone

using CeO2 as photocatalyst has not been previously reported but it is expected

to take benefits of both processes (photocatalysis and catalytic ozonation)

together with the synergism observed during photocatalytic ozonation

treatment [2]. Therefore, the aim of this work focuses on the application of two

nanostructured CeO2 catalysts with different morphology (nanocubes and

nanorods) in ozone and photocatalysis based AOPs (photocatalysis, catalytic

and photocatalytic ozonation) using simulated solar radiation (300 ‐ 800 nm) and

visible radiation (390 ‐ 800 nm) from a Xe lamp, and N,N‐diethyl‐meta‐

toluamide (DEET) as target compound. DEET is a common insect repellent

found in different aquatic environments [20] that presents a low reactivity with

ozone [21], thus facilitating the comparison between the different processes.



Two CeO2 catalysts were synthetized by hydrothermal treatment according

to a previous work [22]. Briefly, Ce(NO3)3∙6H2O was dissolved in 6 M NaOH

solution and then transferred to a 125 mL autoclave (filled at 75 %) and heated

during 24 h at 100 °C for nanorods (NR) or 180 °C for nanocubes (NC). After the

hydrothermal treatment, the autoclave was cooling down to room temperature

and then, the precipitates were separated by centrifugation, washed sequentially

with water and ethanol and dried at 100 °C overnight.



217


CeO2 catalysts were characterized by different techniques. X‐ray diffraction

(XRD) patterns were recorded using a powder Bruker D8 Advance XRD

diffractometer with a Cu Kα radiation (λ = 0.1541 nm). The data were collected

from 2θ = 20° ‐ 80° at a scan rate of 0.02 s−1. A Tecnai G2 20 (FEI Company)

transmission electron microscope at 200 kV was used to study the crystallinity

and morphology of the samples. Textural properties were analyzed by nitrogen

adsorption‐desorption isotherms at ‐196 °C using an Autosorb‐1 apparatus

(Quantachrome). The samples were outgassed at 250 °C for 12 h under vacuum

(< 10−2 mbar). Surface characterization was performed by X‐ray photoelectron

spectroscopy. XPS spectra were obtained with a Kα Thermo Scientific apparatus

with Al Kα (h = 1486.68 eV) X‐ray source using a voltage of 12 kV under

vacuum (2x10−7 mbar). Binding energies were calibrated relative to the C1s peak

at 284.6 eV and spectra deconvolution was accomplished using XPS‐Peak 4.1

software. Diffuse reflectance UV‐Vis spectroscopy (DR‐UV‐Vis) measurements

were performed with a UV‐Vis‐NIR Cary‐5000 spectrophotometer (Varian‐

Agilent Technologies) equipped with an integrating sphere device. Band gap

values of the samples were obtained using Tauc’s equation, assuming an

indirect band gap semiconductor behavior [23].


Photocatalytic experiments were carried out in a solar simulator (Suntest

CPS, Atlas) provided with a 1500 W Xe lamp operated at 550 W m−2. In some

cases light transmission was restricted to λ > 390 nm by means of a modified

polyester cut‐off filter (Edmund Optics) for visible experiments, whereas a

window glass filter was used for simulating the spectrum of solar radiation

reaching the Earth’s surface (λ > 300 nm). The spectral irradiance of Xe lamp

using the window glass filter, and the transmittance of the polyester filter are

represented in Figure 6.1. The experiments were carried out in semi‐batch mode

(batch with respect to the liquid phase and continuous with respect to the gas


218

phase) using a borosilicate glass‐made round flask provided with gas inlet, gas

outlet and a liquid sampling port. In a typical photocatalytic ozonation

experiment, the reactor was loaded with 500 mL of 5 mg L‐1 of DEET aqueous

solution (pH0 = 6). The catalyst was then added (0.25 g L‐1) and the suspension

was magnetically stirred in the dark for 30 min before switching on the lamp

and feeding ozone to the reactor (15 L h−1 gas flow rate, 10 mg L−1 O3, with an

ozone generator Labor‐Ozonisator from Sander fed with pure oxygen from a

cylinder). The temperature in all the experiments was maintained at 35 ‐ 40 °C.

Besides photocatalytic ozonation (CeO2/O3/h), blank experiments of photolysis

(h), single ozonation (O3), photolytic ozonation (O3/h), adsorption (CeO2),

photocatalysis (CeO2/h) and catalytic ozonation (CeO2/O3) were carried out for

comparative reasons. Most of the experiments were triplicated. Samples were

withdrawn from the reactor at different interval times, filtered with 0.2 μm PET

membranes and analyzed.

300 400 500 600 700 8000

1

2

3

4

5

SP

EC

TR

AL

IRR

AD

IAN

CE

(W

m-2

nm

-1)

AB

SO

RB

AN

CE

(u

.a.)

WAVELENGTH (nm)

NC

NR

0

20

40

60

80

100F

ILT

ER

TR

AN

SM

ITT

AN

CE

(%

)

Figure 6.1. UV‐Vis characteristics of the photocatalytic system applied (spectral

irradiance of the Xe lamp in the solar simulator, transmittance of the UV filter and UV‐

Vis absorbance of the CeO2 catalysts).




219

using a Phenomenex C‐18 column (5 μm, 150 mm long, 3 mm diameter) and 0.6

mL min−1 of acetonitrile‐acidified water (0.1 % formic acid) as mobile phase (30 ‐

70 v/v, isocratic). Identification and quantification was carried out at 220 nm.

Total organic carbon (TOC) was measured using a Shimadzu TOC‐VSCH

analyzer. Aqueous ozone was photometrically determined by the indigo method

at 600 nm [24], in a UV‐Vis spectrophotometer Evolution 201

(Thermospectronic) and ozone in the gas phase was continuously monitored by

an online analyzer (Anseros Ozomat GM‐6000Pro). Hydrogen peroxide

concentration was photometrically determined by the cobalt/bicarbonate

method at 260 nm using the same spectrophotometer [25]. Short‐chain organic

acids were analyzed by ion chromatography with chemical suppression

(Metrohm 881 Compact Pro) and a conductivity detector, using a MetroSep A

sup 5 column (250 mm long, 4 mm diameter) at 45 °C and 0.7 mL min‐1 of

Na2CO3 from 0.6 ‐ 14.6 mM in 50 min (10 min post‐time for equilibration) as

mobile phase.


6.3.1. Photocatalysts characterization

The formation of nanostructured CeO2 with NC or NR morphology was

confirmed by TEM as can be observed in Figure 6.2(A). From the NC images,

also some rectangular prism particles have been detected (a minority in the

images analyzed). The size of NC and NR, determined from several TEM

representative images (not shown), is given in Table 6.1. A size distribution

between 25 and 100 nm was obtained for NC whereas the NR presented a

thickness of ~ 7.2 nm and lengths between 40 and 200 nm. NC and NR showed

{100} and {110} + {100} exposed facets, respectively, which is in accordance with

previous reports [22]. On the other hand, XRD patterns shown in Fig. 6.2(B)

confirmed the formation of pure CeO2 cubic phase (fluorite structure, JCPDS 34‐

0394) in both NC and NR samples [22].


220

Figure 6.2. Structural characterization of the CeO2 nanostructured catalysts: (A) TEM

images and (B) XRD patterns.

The porosity of the catalysts was evaluated by means of N2 adsorption‐

desorption isotherms. BET surface area and total pore volumes are summarized

in Table 6.1. Previous reports have shown that these materials are free of a

developed pore structure except general random stacking of nanoparticles

[22,26]. Due to its morphology and lower crystal size, NR presented higher SBET

and pore volume values than NC. With these characteristics, it is reasonable to

assume that the powder dispersion of these materials (NR and NC) can be

illuminated without internal shading from pores. Thus, the surface area exposed

to light (SBET in this case) should be taken into account since it can affect the

photocatalytic behavior to some extent.

NC

NR

(A) (B)

20 30 40 50 60 70 80

INT

EN

SIT

Y (

a.u.

)

2 (º)

NR-CeO2

NC-CeO2

PAPER 4: Nanostructured CeO2 as catalysts for different AOPs based in the application of ozone and simulated solar radiation

221

Tabl

e 6.

1. P

rope

rtie

s of

the

CeO

2 nan

ostr

uctu

red

cata

lyst

s.

CA

TALY

ST

Synt

hesi

s a S

ize

(nm

) Ex

pose

d fa

cets

CeO

2 cr

ysta

lline

ph

ase

SBET

(m

2 g-1

) V

p

(cm

3 g-1

) b C

e(II

I)

(%, a

t.)

b Ce(

IV)

(%, a

t.)

Eg

(eV

)

NC

H

ydro

ther

mal

NaO

H

6 M

180

°C, 2

4h

25 -

100

{100

} C

ubic

14

0.

185

12.6

87

.4

3.32

NR

Hyd

roth

erm

al N

aOH

6

M 1

00 °C

, 24h

(7

.2 ±

1.1

) x

(40

- 200

) {1

10} +

{100

} C

ubic

10

9 0.

522

23.6

76

.4

3.07

a S

ize o

f the

mai

n di

men

sions

by

TEM

ana

lyse

s; b C

alcu

late

d fro

m C

e 3d

XPS

dec

onvo

lutio

n


222

One of the main characteristics of CeO2 catalysts is their oxygen and electron

mobility due to the presence of surface defects, reduced states of Ce(III) or

oxygen vacancies in the CeO2 structure. High‐resolution XPS spectra

corresponding to Ce3d spectral region of NC and NR samples have been plotted

in Figure 6.3. From this figure, a slight shift of Ce3d peaks to lower binding

energies in NR with respect to NC sample is noticed. This has been attributed to

an increase of the electron density around Ce nucleus, suggesting the existence

of more Ce(III) ions on the surface of NR sample [26]. In addition, this has been

confirmed by the highest intensity of the peaks corresponding to Ce(III) species

(v0, v’, u0, u’) in NR catalyst [18,26,27], leading to a higher percentage of surface

Ce(III) in NR compared to NC sample (see Table 6.1).

920 910 900 890 880 870

NR 23.6% 76.4%NC 12.6% 87.4%

CP

S (

a.u.

)

BINDING ENERGY (eV)

NR

NC

v0

v'

u0u'

Ce(III) Ce(IV)

Figure 6.3. High‐resolution XPS spectra and deconvolution of the Ce3d spectral region of

the CeO2 nanostructured catalysts.



223

Finally, optical properties of the catalysts have been studied by DR‐UV‐Vis

spectroscopy. Fig. 6.1 shows the UV‐Vis absorption spectra of NC and NR where

a noticeable shift to higher wavelength for NR sample is observed, which

suggests this catalyst can use a higher fraction of visible radiation to trigger

photocatalytic reactions. The optical energy band gap values (Eg) calculated

assuming indirect band gap behavior for both catalysts are also summarized in

Table 6.1. A value of 3.32 eV was obtained for NC, somewhat higher than that

reported for bulk CeO2 (around 3.19 eV) due to smaller particle sizes [28]. On the

contrary, a noticeable lower Eg value was calculated for NR indicating a clearly

shift to higher absorption wavelengths as deduced from Fig. 6.1. This effect may

be attributed to the lattice defects and oxygen vacancies present in the ceria NR

[28], in agreement with the results obtained by XPS analyses.

6.3.2. Catalytic activity

The activity of the CeO2 catalysts was evaluated in DEET degradation

through AOPs of photocatalytic oxidation, catalytic ozonation and

photocatalytic ozonation in aqueous solution using visible and simulated solar

radiation. The results obtained for DEET degradation are shown in Figure 6.4.

Some blank experiments are also included for comparative purposes. As

reported in a previous work, at the conditions applied no DEET degradation by

direct photolysis was observed [29]. The adsorption capacity of both catalysts

was also negligible being DEET removal lower than 10 %. In addition, in spite of

the optical properties of the catalysts, no significant DEET removal was achieved

by photocatalysis regardless of the radiation used (visible or solar). These results

differ from those previously reported using different CeO2 catalysts, although in

those cases photosensitizing dyes were used as target compounds [11‐15]. On

the other hand, the highest DEET degradation rate was obtained for ozone‐

based systems. Thus, single ozonation led to a complete removal of DEET in less

than 60 min. A similar DEET evolution was observed during catalytic ozonation

with respect to ozonation regardless of the catalyst used. This points out that

these materials do not show any beneficial effect on DEET removal when


224

combined with ozone in the dark, at the conditions used in this work.

-30 0 30 60 90 1200.0

0.2

0.4

0.6

0.8

1.0

NR NR/h NR/O

3

NR/O3/h

NC NC/h NC/O

3

NC/O3/h

h O

3

O3/h

Dark

CD

EE

T/C

DE

ET

,0

TIME (min)

-30 0 30 60 90 1200.0

0.2

0.4

0.6

0.8

1.0

NR NR/h NR/O

3

NR/O3/h

NC NC/h NC/O

3

NC/O3/h

h O

3

O3/h

Dark

CD

EE

T/C

DE

ET

,0

TIME (min)

Figure 6.4. Evolution of normalized DEET concentration during different treatments

applied under visible (A) and solar (B) simulated radiation (maximum standard

deviation of the DEET concentrations in 3 repeated runs 0.03, not shown). Experimental

conditions: pH0 = 6, T = 35 ‐ 40 °C, CDEET,0 = 5 mg L‐1, CCeO2 = 0.25 g L‐1, CO3,g inlet = 10

mg L‐1, Qg = 15 L h‐1 (O2 or O3/O2).

However, the impact of the radiation (visible and solar) on the efficacy of

ozone‐based processes was significant and, in general, higher degradation rates

(A) VISIBLE

(B) SOLAR



225

were observed when solar light was used (Fig. 6.4(B)). In this sense, DEET was

completely removed after 45 and 30 min when ozone was combined with visible

and solar radiation, respectively, being these periods clearly lower than that

required by single ozonation (60 min). It has been previously demonstrated that

ozone can undergo direct photolysis under solar radiation promoting the

formation of some reactive oxygen species capable of oxidizing the organic

matter in aqueous solution [30,31]. Nevertheless, the interaction between visible

radiation and ozone is unclear and will be investigated in future works. Finally,

according to Fig. 6.4(A), the addition of NC and NR catalysts to the ozone +

radiation system had almost no effect on DEET degradation rate. Only a slight

beneficial effect was observed when visible light was used, regardless of the

type of catalyst.

Main differences between ozone‐based treatments were observed in terms of

DEET mineralization (i.e. transformation into CO2, H2O and inorganic

compounds). Figure 6.5 shows the percentage of mineralization during ozone‐

based treatments at 120 min reaction time. As observed, ozone alone gave rise to

only a 19 % mineralization due to the formation of intermediate compounds

refractory to ozone direct reaction (oxalic, formic and acetic acids were detected

in the reaction medium) [32]. The presence of CeO2 catalysts slightly increased

(by 5 %) the final mineralization, which could be attributable to a positive effect

of cerium oxide on the elimination of some intermediate compounds formed

during DEET ozonation. The combination of ozone and radiation gave rise to

29 % and 47 % of TOC removal under visible and solar light, respectively.

Therefore, compared to visible light the effect of the entire solar spectrum

(which also includes wavelengths between 300 ‐ 390 nm) is much higher.

However, the best results were obtained when photocatalytic ozonation was

applied, being NR the most active under visible radiation (68 % TOC removal),

likely due to its lower band gap energy as a consequence of higher Ce(III)

surface ratio, whereas NC presented the highest photocatalytic activity under

solar radiation (80 % TOC removal). This is in agreement with the highest

energy of {100} facets (thermodynamically more unstable) predominant in NC


226

with respect to NR, allowing higher photocatalytic activity for this material

because once excited the photoinduced holes exhibit superior oxidizing ability,

according to previously reported results [26].

0

20

40

60

80

100

O3/NRO3/NCO3

TO

C R

EM

OV

AL

(%)

Dark Visible Solar

Figure 6.5. TOC removal with the CeO2 catalysts and O3 processes. Experimental


mg L‐1, Qg = 15 L h‐1 (O3/O2).

These results have also been confirmed by calculating the pseudo‐first order

apparent rate constant of TOC removal, kTOC, during photocatalytic ozonation

treatment, according to Eq. (6.1):

TOCTOC TOC

dC‐ k C

dt (6.1)

where CTOC is expressed in mg L‐1, t in min and kTOC in min‐1. Results are

summarized in Table 6.2 together with the correlation coefficient. Taking into

account the different irradiated surface area of the two catalysts (see Table 6.1),

values of normalized to SBET apparent rate constants (kTOC/SBET, min‐1 m‐2 g1) have

been calculated. The values obtained (Table 6.2) confirm the highest

photocatalytic activity of ceria NC. It is also noticeable that solar UV radiation

did not improve the mineralization rate when NR is used probably due to an



227

increase of the recombination rate in the semiconductor.

Table 6.2. Pseudo‐first order apparent rate constant for TOC elimination during

photocatalytic ozonation runs.

CATALYST RADIATION kTOCx103

(min‐1)

kTOC/SBETx104

(min‐1 m‐2 g) R2

NC

Visible 7.2 5.1 0.98

Solar 11.8 8.4 0.96

NR

Visible 11.2 1.1 0.95

Solar 11.5 1.1 0.94

6.3.3. Considerations on the reaction mechanism of DEET photocatalytic

ozonation with CeO2 catalysts and solar radiation

According to the above results, photocatalytic ozonation of DEET using the

CeO2 catalysts studied here, may proceed through a complex mechanism with

the expected participation of ozone (direct reaction), and/or •HO radicals formed

from ozone decomposition (indirect reactions). Ozone can decompose into •HO

by several pathways: (1) in the dark, (2) by interaction with light, (3) by

interaction with the catalyst, and (4) by interaction with both light and the

catalyst. To discuss the possible contributions of each pathway Figure 6.6 shows

the evolution of DEET eliminated, dissolved ozone and hydrogen peroxide

formed during ozone‐based treatments using NC CeO2 and simulated solar

radiation.

Regarding to ozone‐direct reactions, a low reactivity of DEET towards ozone

has been previously reported (kO3‐DEET = 0.123 M‐1 s‐1 [21]). Taking into account

this value, at the conditions used in this work, the reaction between ozone and


228

DEET progress into the slow kinetic regime [32,33], being DEET degradation

rate by direct ozone reaction given by Eq. (6.2):

3 3

DEETO ‐DEET DEET O

dC‐ k C C

dt (6.2)

where CDEET is DEET molar concentration and CO3 the dissolved ozone molar

concentration. The expected evolution of DEET eliminated during ozonation in

the dark considering only its reaction with ozone is shown in Fig. 6.6. As it is

observed, the experimental conversion rate was much higher than that predicted

by Eq. (6.2), indicating that species different from O3 are the main responsible

for the degradation of the target compound. However, it is important to notice

that kO3‐DEET value was reported at 20 °C whereas the temperature during the

experiments was maintained at 35 ‐ 40 °C. Also, due to the low value of kO3‐DEET

some uncertainty on its determination should not be disregarded.

During all ozone‐based treatment pH evolved from 6 to ~ 4. At these pH

values, at the beginning of the experiment some ozone decomposition can be

promoted by hydroxide anion although it likely becomes negligible when pH

decreases [32]. Therefore, during DEET degradation by single ozonation in the

dark it can be hypothesized that direct interaction between DEET‐O3 acts as a

trigger of a chain mechanism that leads to the formation of ozonide and/or •HO

radicals, as has been previously reported for different organic compounds [33].

The key role of •HO during DEET single ozonation has been previously reported

[34].

The evolution of dissolved O3, also depicted in Fig. 6.6, shows a higher

accumulation during ozonation compared to treatments where a catalyst and/or

radiation were also present. This is indicative of the decomposition of ozone

over the catalyst surface (although almost inefficient in the dark) and/or through

its photolysis. According to the literature [30,35], both mechanisms could

promote the formation of hydroxyl radicals, thus increasing the mineralization



229

rate of DEET.

Figure 6.6. Evolution of DEET converted, H2O2 formed and O3 dissolved during the O3

treatments applied using CeO2‐NC catalyst and simulated solar light. Experimental


mg L‐1, Qg = 15 L h‐1 (O3/O2).

In addition, it is well known that hydrogen peroxide is commonly formed

through direct ozone and hydroxyl radical reactions. Accordingly, H2O2 was

detected during DEET ozonation. As observed in Fig. 6.6, during single

ozonation about the same concentration of H2O2 is formed respect to DEET

depleted. When light and/or NC catalyst were also present the ratio between

H2O2 in solution and DEET eliminated was much lower, and more markedly in

the presence of NC thus indicating the decomposition of H2O2 is accelerated by

CeO2 catalyst surface. It is important to notice that H2O2 decomposition seems to

be inefficient towards DEET degradation and mineralization according to the

evolution of DEET and TOC during catalytic ozonation (dark). However, the

0 20 40 60 80 100 120

TIME (min)

NC/O3/h

0 20 40 60 80 100 1200.0

1.0x10-5

2.0x10-5

3.0x10-5O

3/h

C (

M)

TIME (min)

NC/O3

0.0

1.0x10-5

2.0x10-5

3.0x10-5

O3

C (

M)

DEET (converted) H

2O

2 (formed)

DEET (simulated) O

3 (dissolved)


230

role of H2O2 as electron acceptor during solar light photocatalytic ozonation

using CeO2 cannot be disregarded [35]. The role of hydrogen peroxide as

electron acceptor as well as the contribution of the different reaction pathways

to DEET degradation and mineralization by photocatalytic ozonation using solar

light and CeO2 NC as catalyst will be the subject of future work.

6.4. CONCLUSIONS

Nanostructured CeO2 catalysts with different morphology (nanorods and

nanocubes) were active in photocatalytic ozonation treatment of DEET under

visible and simulated solar radiation. CeO2 with nanorod‐like structure

presented higher irradiated surface area due to its morphology and also higher

amount of surface defects and oxygen vacancies, leading to smaller band gap

energy and thus being apparently more active than CeO2 nanocubes under

visible light radiation. However, the presence of the exposed facets {100} in CeO2

nanocubes provokes an increased intrinsic photocatalytic activity in this

material with respect to nanorods under both visible and solar radiation.

Photocatalytic ozonation of DEET with CeO2 catalysts proceeds through a

complex reaction mechanism that involves at least an ozone‐direct reaction that

triggers a radical chain mechanism, indirect ozone reactions, ozone photolytic

and photocatalytic decomposition and hydrogen peroxide formation‐

decomposition. The importance of the different processes and the effect of

visible radiation will be the subject of a future work.

AKNOWLEDGEMENTS

Authors thank the Spanish MINECO/European Feder Funds (CTQ2012‐

35789‐C02‐01) and Junta de Extremadura (Ayuda Exp. GR15‐033) for economic

support. E. Mena thanks the Consejería de Empleo, Empresa e Innovación (Junta

de Extremadura) and European Social Fund for her FPI grant (Ref. PD12059).



231

REFERENCES

[1] Oturan, M.A.; Aaron, J.J. “Advanced oxidation processes in

water/wastewater treatment: Principles and applications. A review”. Crit. Rev.

Environ. Sci. Technol. 44 (2014) 2577‐2641.




[3] Mehrjouei, M.; Müller, S.; Möller, D. “A review on photocatalytic ozonation

used for the treatment of water and wastewater”. Chem. Eng. J. 263 (2015) 209‐

219.

[4] Xiao, J.; Xie, Y.; Cao, H. “Organic pollutants removal in wastewater by

heterogeneous photocatalytic ozonation”. Chemosphere 121 (2015) 1‐17.




[6] Dong, S.Y.; Feng, J.L.; Fan, M.H.; Pi, Y.Q.; Hu, L.M.; Han, X.; Liu, M.L.; Sun,

J.Y.; Sun, J.H. “Recent developments in heterogeneous photocatalytic water

treatment using visible light‐responsive photocatalysts: a review”. RSC Adv. 5

(2015) 14610‐14630.

[7] Huang, W.; Gao, Y. “Morphology‐dependent surface chemistry and catalysis

of CeO2 nanocrystals”. Catal. Sci. Technol. 4 (2014) 3772‐3784.

[8] Aslam, M.; Qamar, M.T.; Tahir‐Soomro, M.; Ismail, I.M.I.; Salah, N.;

Almeelbi, T.; Gondal, M.A.; Hameed, A. “The effect of sunlight induced surface

defects on the photocatalytic activity of nanosized CeO2 for the degradation of

phenol and its derivatives”. Appl. Catal. B Environ. 180 (2016) 391‐402.

[9] Huang, X.; Beck, M.J. “Size‐dependent appearance of intrinsic qxO “activated

oxygen” molecules on ceria nanoparticles”. Chem. Mater. 27 (2015) 5840‐5844.

[10] Hernández‐Alonso, M.D.; Hungría, A.B.; Martínez‐Arias, A.; Fernández‐

García, M.; Coronado, J.M.; Conesa, J.C.; Soria, J. “EPR study of the

photoassisted formation of radicals on CeO2 nanoparticles employed for toluene

photooxidation”. Appl. Catal. B Environ. 50 (2004) 167‐175.

[11] Deng, W.; Chenn, D.; Chenn, L. “Synthesis of monodisperse CeO2 hollow

spheres with enhanced photocatalytic activity”. Ceram. Int. 41 (2015) 11570‐

11575.


232

[12] Muduli, S.K.; Wang, S.; Chen, S.; Fan‐Ng, C.; Huan, C.H.A.; Sum, T.C.; Soo,

H.S. “Mesoporous cerium oxide nanospheres for the visible‐light driven

photocatalytic degradation of dyes”. Beilstein J. Nanotechnol. 5 (2014) 517‐523.

[13] Yu, Y.; Zhu, Y.; Meng, M. “Preparation, formation mechanism and

photocatalysis of ultrathin mesoporous single‐crystal‐like CeO2 nanosheets”.

Dalton Trans. 42 (2013) 12087‐12092.

[14] Feng, T.; Wang, X.; Feng, G. “Synthesis of novel CeO2 microspheres with

enhanced solar light photocatalyic properties”. Mater. Lett. 100 (2013) 36‐39.

[15] Sifontes, A.B.; Rosales, M.; Méndez, F.J.; Oviedo, O.; Zoltan, T. “Effect of

calcination temperature on structural properties and photocatalytic activity of

ceria nanoparticles synthesized employing chitosan as template”. J. Nanomater.

Article ID 265797 (2013) 1‐9.

[16] Karunakaran, C.; Dhanalakshmi, R. “Semiconductor‐catalyzed degradation

of phenols with sunlight”. Sol. Energy Mater. Sol. Cells 92 (2008) 1315‐1321.

[17] da Silva, M.F.P.; Soeira, L.S.; Daghastanli, K.R.P.: Martins, T.S.; Cuccovia,

I.M.; Freire, R.S.; Isolani, P.C. “CeO2‐catalyzed ozonation of phenol: The role of

cerium citrate as precursor of CeO2”. J. Therm. Anal. Calorim. 102 (2010) 907‐

913.

[18] Faria, P.C.C.; Orfao, J.J.M.; Pereira, M.F.R. “A novel ceria‐activated carbon

composite for the catalytic ozonation of carboxylic acids”. Catal. Comm, 9 (2008)

2121‐2126.

[19] Gonçalves, A.G.; Orfao, J.J.M.; Pereira, M.F.R. “Ozonation of bezafibrate

over ceria and ceria supported on carbon materials”. Environ. Technol. 36 (2015)

776‐785.






decomposition mechanisms of ozone with two model emerging contaminants:


[22] Mai, H.X.; Sun, L.D.; Zhang, Y.W.; Si, R.; Feng, W.; Zhang, H.P.; Liu, H.C.;

Yan, C.H. “Shape‐selective synthesis and oxygen storage behavior of ceria

nanopolyhedra, nanorods, and nanocubes”. J. Phys. Chem. B 109 (2005) 24380‐

24385.

[23] Tauc, J. “Absorption edge and internal electric field in amorphous



233

semiconductors”. Mater. Res. Bull. 5 (1970) 721‐729.


method”. Water Res. 15 (1981) 449‐456.



[26] Jiang, D.; Wang, W.; Zhang, L.; Zheng, Y.; Wang, Z. “Insights into the

surface‐defect dependence of photoreactivity over CeO2 nanocrystals with well‐

defined crystal facets”. ACS Catal. 5 (2015) 4851‐4858.

[27] Muñoz‐Batista, M.J.; Gómez‐Cerezo, M.N.; Kubacka, A.; Tudela, D.;

Fernández‐García, M. “Role of interface contact in CeO2‐TiO2 photocatalytic

composite materials”. ACS Catal. 4 (2014) 63‐72.

[28] Zdravkovic, J.; Simovic, B.; Golubovic, A.; Poleti, D.; Veljkovic, I.;

Scepanovic, M.; Brankovic, G. “Study of CeO2 nanopowders obtained by

hydrothermal method from various precursors”. Ceramics Int., 141 (2015) 1970‐

1979.

[29] Mena, E.; Rey, A.; Contreras, S.; Beltrán, F.J. “Visible light photocatalytic

ozonation of DEET in the presence of different forms of WO3”. Catal. Today 252

(2015) 100‐106.

[30] Sánchez, L.; Domènech, X.; Casado, J.; Peral, J. “Solar activated ozonation of

phenol and malic acid”. Chemosphere 50 (2003) 1085‐1093.




[32] Beltran, F.J. “Ozone reaction kinetics for water and wastewater systems”.

Boca Raton, CRC Press, 2004, Florida (USA).

[33] Rey, A.; Mena, E.; Chávez, A.M.; Beltrán, F.J.; Medina, F. “Influence of

structural properties on the activity of WO3 catalysts for visible light

photocatalytic ozonation”. Chem. Eng. Sci. 126 (2015) 80‐90.

[34] Acero, J.L.; Benítez, F.J.; Real, F.J.; Rodríguez, E. “Elimination of selected

emerging contaminants by the combination of membrane filtration and chemical

oxidation processes”. Water Air Soil Pollut. 226 (2015) 139‐1‐139‐14.




Chem. Eng. J. 204‐206 (2012) 131‐140.

CAPÍTULO 7 (CHAPTER 7) PAPER 5: WO3-TiO2 based catalysts for the simulated solar radiation assisted photocatalytic ozonation of emerging contaminants in a municipal wastewater treatment plant effluent

A. Rey, P. García-Muñoz, M.D. Hernández-Alonso, E. Mena, S. García-Rodríguez, F.J. Beltrán

Applied Catalysis B: Environmental 154-155 (2014) 274-284

ABSTRACT. This work is focused on the use of TiO2-WO3 photocatalysts for the

removal of a mixture of emerging contaminants through photocatalytic ozonation using

simulated solar light as radiation source. Own-made TiO2-WO3 catalysts with WO3

content around 4 wt. % were synthetized using P25 and titanate nanotubes as TiO2

starting materials. Photocatalysts were thoroughly characterized by means of ICP-OES,

N2 adsorption-desorption isotherms, XRD, TEM, Raman, XPS and DR-UV-Vis

spectroscopy. Photocatalytic ozonation using WO3 based titanate nanotubes composite

gave place to complete removal of emerging contaminants (caffeine, metoprolol and ibuprofen in a Municipal Wastewater Treatment Plant effluent aqueous matrix) in less than 40 min with TOC removal up to 64 % after 2 h. The highest catalytic activity of this

material in the reaction under study compared to bare TiO2 is due to several effects such

as higher activity under visible light radiation joined to an increase in adsorption capacity

of organic compounds and catalytic ozonation ability caused by the presence of WO3.

Keywords: Photocatalytic ozonation, tungsten oxide-titanium dioxide, titanate nanotubes, ozone, emerging contaminants.

PAPER 5: WO3‐TiO2 based catalysts for the simulated solar radiation assisted photocatalytic

ozonation of emerging contaminants in a municipal wastewater treatment plant effluent

237

7.1. INTRODUCTION

Emerging contaminants (ECs), such as pharmaceuticals and personal care

products, which usually present endocrine disrupting activity, are frequently

detected in wastewater and aquatic environments [1,2]. These contaminants can

cause severe adverse effects in human and wildlife and their removal is of a

great concern on environmental and health risk management [2,3]. However

these compounds are hardly biodegradable so they are not removed by

conventional treatment at municipal wastewater treatment plants (MWWTPs),

and therefore advanced treatment technologies are required [1].

Advanced oxidation processes (AOPs) which involve the formation of highly

reactive species such as hydroxyl radical ( HO ), have demonstrated their

efficiency to degrade many ECs transforming them into harmless products [4].

Among these AOPs, photocatalytic ozonation, i.e. the combination of

photocatalytic oxidation and ozone processes, can greatly enhance the rates of

ECs degradation and, specially, the mineralization achieved by the single

processes [5‐7]. In fact, a synergistic effect has been observed between ozone and

irradiated photocatalyst due to both the strong electron trapping effect of ozone,

that avoids, to some extent, the electron‐hole recombination process, and to the

reaction of ozone with the superoxide ion radical ( 2O ), intermediate species in

photocatalytic reactions. In these two steps the ozonide ion radical ( 3O ) is

generated and further transformed intoHO , thus increasing the concentration

of this highly reactive species in the aqueous medium [5,8,9].

One of the most important aspects in the application of photocatalytic

wastewater treatment technologies is the cost derived from the radiation source

operation and maintenance. The use of solar energy as radiation source avoids

this drawback leading to more economic technologies [10]. However, most of

the photocatalytic ozonation studies conducted to date make use of TiO2 as

photocatalyst that only uses around 5 % of solar radiation (UV range). Different

methods to solve this limitation, mainly applied to ozone‐free photocatalytic


238

oxidation processes, have been developed, such as ion doping, modification of

TiO2 surface with noble metals, metal ion implantation or coupling

semiconductors [10,11]. In this line, composite materials of coupled WO3 and

TiO2 semiconductors can extend the absorption of radiation to the visible region

as a result of the lower band gap energy of WO3 (around 2.7 eV), and also allow

a wide electron‐hole separation avoiding recombination to some extent [11].

TiO2‐WO3 coupled semiconductors have been successfully applied to the

degradation of different organic contaminants in water through photocatalytic

oxidation under visible light radiation [12‐16]. In addition, these materials

combined with gold nanoparticles have also been tested for photocatalytic

ozonation of model compounds such as oxalic acid and 2,4,6‐trinitrotoluene

(TNT) giving place to promising results [17,18].

This work has been focused on the use of different TiO2‐WO3 composite

catalysts to extend the radiation absorption to visible light and make better use

of the solar emission spectrum in photocatalytic ozonation process applied to

wastewater detoxification. Two different TiO2 supports, commercial TiO2 P25

and own‐made titanate nanotubes, which present very different textural and

structural properties, have been used. The synthetized materials were tested in

photocatalytic ozonation of ibuprofen (IBP), metoprolol (MTP) and caffeine

(CAF) spiked in a municipal wastewater effluent from a secondary treatment

and using simulated solar light as radiation. These contaminants, IBP, a

nonsteroidal anti‐inflammatory drug for pain relief and fever reduction; MTP, a

β‐blocker used for several cardiovascular diseases; and CAF, a stimulant drug

mainly from coffee consumption, have been selected because of their frequent

presence in municipal wastewater [2].



Aeroxide TiO2 P25 (anatase/rutile 80/20, 21 nm crystal size) was used as the



239

starting photocatalytic material. The procedure to obtain titanate nanotubes

catalyst (NT) has been reported in previous works [19,20]. Briefly, 1 g of the

precursor P25 was hydrothermally treated at 130 °C in 70 mL of 10 M NaOH in a

Teflon‐lined autoclave, during 48 h. The mixture was stirred for 30 min before

and after the thermal treatment. In a second stage, the obtained powders were

thoroughly washed using diluted HCl. The powders were recovered from the

solution by centrifugation, dried at 100 °C overnight and then calcined in air

atmosphere for 3 h at 350 °C.

P25 and NT photocatalysts were coated with nanosized WO3 particles (P25‐

WO3 and NT‐WO3 samples) as described elsewhere [18]. First 0.216 g H2WO4

were added to 100 mL of ultrapure water and then aqueous ammonia solution

was added drop wise until the tungstic acid was completely dissolved. Further 5

g of P25 or NT were added under continuous stirring and the obtained mixture

was stirred for 30 min and then acidified to pH = 4 with 0.5 M HCl. Then 10 mL

of an aqueous solution of oxalic acid 0.1 M was added to the mixture and stirred

for 1 h at 40 °C to prevent the aggregation of WOX particles in the precipitate.

Finally, the solid was recovered by filtration, dried at 110 °C for 2 h and then

calcined in air atmosphere for 2 h at 420 °C. The TiO2‐WO3 thus prepared had a

theoretical amount of WO3 of 3.8 wt. %. For comparative purposes, NT was also

calcined in air atmosphere for 2 h at 420 °C (sample NT‐T).


Total tungsten content of the catalysts was analyzed by inductively coupled

plasma with an ICP‐OES Optima 3300DV (Perkin‐Elmer) after acidic microwave

digestion of the samples.

BET surface area and pore volume of the photocatalysts were determined

from their nitrogen adsorption–desorption isotherms obtained at ‐196 °C using

an Autosorb 1 apparatus (Quantachrome). Prior to analysis the samples were

outgassed at 250 °C for 12 h under high vacuum (< 10−4 Pa).


240

The crystalline phases present in the photocatalysts were inferred from their

X‐ray diffraction (XRD) patterns recorded using a powder Bruker D8 Advance

XRD diffractometer with a Cu Kα radiation (λ = 0.1541 nm). The data were

collected from 2θ = 20° to 80° at a scan rate of 0.02 s−1 and 1 s per point.

A JEM‐2100F 200 kV transmission electron microscope (JEOL Ltd.) and a

TEM Tecnai G20 Twin 200 kV transmission electron microscope (FEI Company)

were used to study the crystallinity and morphology of the samples.

Raman spectra were obtained using a Nicolet Almega XR Dispersive micro‐

Raman (Thermo Scientific) with a spectral resolution of 2 cm‐1. The samples

were excited at 633 nm with the laser beam power at 100 % and 200 scans

accumulation.

X‐ray photoelectron spectra (XPS) were obtained with a Kα Thermo Scientific

apparatus with an Al Kα (h = 1486.68 eV) X‐ray source using a voltage of 12 kV

under vacuum (2x10−7 mbar). Binding energies were calibrated relative to the

C1s peak at 284.6 eV. The resulting XPS peaks were curve‐fitted to a

combination of Gaussian and Lorentzian functions using a Shirley type

background for peak analysis.

Diffuse reflectance UV‐Vis spectroscopy (DR‐UV‐Vis) measurements, useful

for the determination of the semiconductor band gap, were performed with a

UV‐Vis‐NIR Cary 5000 spectrophotometer (Varian‐Agilent Technologies)

equipped with an integrating sphere device.


Ibuprofen sodium salt (IBP), metoprolol tartrate (MTP) and caffeine (CAF)

were used as target contaminants to test the catalytic activity of the synthesized

materials. They were added to a real municipal wastewater effluent (MWW)

taken from Badajoz MWWTP (Badajoz, Spain) designed for 225,000 inhabitants

(population equivalent) with an average inlet flow of 37,500 m3 day‐1. Effluents



241

were collected downstream of the MWWTP secondary biological treatment,

filtered and stored at ‐20 °C until use.

Photocatalytic experiments were carried out in semi‐batch mode in a

laboratory‐scale system consisting of a 0.5 L glass‐made spherical reactor,

provided with a gas inlet, a gas outlet and a liquid sampling port. The reactor

was placed in the chamber of a commercial solar simulator (Suntest CPS, Atlas)

provided with a 1500 W air‐cooled Xe arc lamp with emission restricted to

wavelengths over 320 nm using quartz, glass and polyester cut‐off filters. The

irradiation intensity was kept at 550 W m−2 and the temperature of the system

was maintained between 20 and 40 °C throughout the experiments. If required,

a laboratory ozone generator (Anseros Ozomat Com AD‐02) was used to

produce a gaseous ozone‐oxygen stream that was fed to the reactor. In that case,

the ozone concentration was monitored by an Anseros Ozomat GM‐6000Pro gas

analyzer. A scheme of the experimental set‐up is depicted in Figure 7.1.

Figure 7.1. Scheme of the experimental set‐up.

In a typical photocatalytic ozonation experiment, the reactor was first loaded

with 0.5 L of the MWW effluent containing 2 mg L‐1 of each EC. Then, 0.25 g of

the catalyst were added and the suspension was stirred in the dark for 30 min

while bubbling air into the system. After this dark stage, the lamp was switched

on and, simultaneously, a mixture of ozone‐oxygen (10 mg L‐1 ozone

concentration) was fed to the reactor at a flow rate of 20 L h−1. The irradiation

time for each experiment was 2 h. Samples were withdrawn from the reactor at


242

intervals and filtered through a 0.2 μm PET membrane to remove the

photocatalyst particles.

Experiments of adsorption (i.e., absence of radiation and ozone), photolysis

(i.e., radiation in absence of catalysts and ozone), photocatalytic oxidation (i.e.,

radiation and catalyst in absence of ozone), ozonation (i.e., absence of radiation

and catalyst), and catalytic ozonation (i.e., absence of radiation) were also

carried out for comparative analysis. In addition, some photocatalytic oxidation

experiments of IBP were carried out under visible radiation using a UV cut‐off (λ

< 390 nm) polyester filter.

ECs concentrations were analyzed by HPLC‐DAD (Hitachi, Elite LaChrom)

using a Phenomenex C‐18 column (5 μm, 150 mm long, 3 mm diameter) as

stationary phase and 0.5 mL min−1 of acetonitrile‐acidified water (0.1 % H3PO4)

as mobile phase (from 5 % to 60 % in acetonitrile during 25 min and 10 min re‐

equilibration time). Identification and quantification was carried out at 220 nm.

Total organic carbon (TOC) and inorganic carbon (IC) were measured using a

Shimadzu TOC‐VSCH analyzer. Aqueous ozone was measured following the

indigo method using a UV‐Vis spectrophotometer (Evolution 201,

Thermospectronic) set at 600 nm [21]. Ozone in the gas phase was continuously

monitored by means of an Anseros Ozomat GM‐6000Pro analyzer. Finally,

chemical oxygen demand (COD) was determined by means of the dichromate

method using Dr. Lange cuvette test, biological oxygen demand (BOD5) was

analyzed by the respirometric method using an Oxitop® WTW system,

aromaticity as the absorbance of the sample at 254 nm and phosphates by means

of colorimetric methods using a Merck Spectroquant kit (UV‐Vis

spectrophotometry, Evolution 201 from Thermospectronic).



243


7.3.1. Characterization of the photocatalysts

Table 7.1 summarizes tungsten content, some textural parameters, band gap

energy and absorption edge wavelength of the photocatalysts. It can be seen that

the W mass composition (expressed as WO3) of the TiO2‐WO3 catalysts was close

but somewhat higher to the nominal value expected (3.8 wt. %).

Table 7.1. Properties of the catalysts.

CATALYST WO3

(wt. %)

SBET

(m2 g‐1)

VP

(cm3 g‐1)

WXPS

(at %)

Eg

(eV)

λ

(nm)

P25 ‐‐‐ 52 0.25 ‐‐‐ 3.19 389

P25‐WO3 4.1 49 0.28 1.6 3.05 407

NT ‐‐‐ 320 1.39 ‐‐‐ 3.18 390

NT‐WO3 4.5 195 1.17 0.9 2.98 416

NT‐T ‐‐‐ 208 1.19 ‐‐‐ ‐‐‐ ‐‐‐

Figure 7.2 shows N2 adsorption‐desorption isotherms where it can be noticed

that P25 based catalysts are non‐porous materials with type II isotherms

whereas NT catalysts presented type IV isotherms with H3 hysteresis loops

characteristics of mesoporous solids, according to IUPAC classifications [22]. On

the other hand, the incorporation of W did not change the isotherm aspect of

these photocatalysts respect to their supports. Main differences are observed for

NT catalysts where the procedure of W incorporation has changed the slope and

shifted the hysteresis loop in NT‐WO3 respect to NT, indicating a lower

porosity. This is mainly related to the heat‐treatment applied, as can be deduced

from the similarity of the isotherms of NT‐T and NT‐WO3. Calcination at 420 °C

triggers structural changes as discussed below [20]. From these analyses, specific

surface area (SBET) and pore volume (VP) were calculated (see Table 7.1). As can

be observed P25 based catalysts show SBET values according to that indicated by


244

the supplier for P25 (around 50 m2 g‐1), and similar pore volume around 0.25 m3

g‐1. Thus, the incorporation of W and calcination of the composite material does

not significantly affect its textural properties. On the other hand, NT presented

higher SBET and pore volume, according to its mesoporous structure as it was

previously reported for similar materials [20]. In this case, the incorporation of

W has favored a decrease in the surface area and pore volume of the NT‐WO3

catalyst mainly due to the calcination step, although some pore blockage after W

deposition cannot be disregarded according to the lower values of textural

parameters of this catalyst respect to the NT‐T sample.

0.0 0.2 0.4 0.6 0.8 1.00

200

400

600

800

1000

1200

N2

AD

SO

RB

ED

VO

LU

ME

(cm

3 g-1

)

P/P0

P25 P25-WO

3

NT NT-WO

3

NT-T

Figure 7.2. N2 adsorption‐desorption isotherms of the catalysts.

Structural characterization of the photocatalysts was accomplished by means

of XRD, TEM and Raman analyses and the results are depicted in Figure 7.3 to

Figure 7.5, respectively. P25 based catalysts gave place to similar XRD patterns

with main diffraction peaks attributable to anatase and rutile TiO2 phases. The

absence of peaks attributable to W species could be due to the low content in the

catalysts and/or to the high dispersion of W. This was also confirmed, as Figure



245

7.4 shows, by TEM micrographs where W particles were undistinguishable. In

addition, heat‐treatment applied to P25‐WO3 photocatalyst does not produce

any significant change in the TiO2 structure. Regarding NT photocatalyst, a

shifted diffraction peak position at 24.9° respect to 25.4° for anatase is observed,

indicating a titanate structure [20]. TEM micrograph confirmed nanotubular

morphology for NT sample (Fig. 7.4). On the other hand, the incorporation of W

(NT‐WO3 photocatalyst) led to a transformation of the titanate structure into

anatase as evidenced by the shift of the corresponding diffraction peaks (e.g.

most intense from 24.9° to 25.3°) together with an increase in their intensity. This

transformation is mainly related to the calcination step at 420 °C as the XRD

pattern and TEM micrograph of the NT‐T sample (subjected to the same heat‐

treatment) confirmed, and also according to the results reported elsewhere [20].

TEM micrographs corroborated the transformation of titanate nanotubular

structures to elongated anatase particles (Fig. 7.4). Fig. 7.5 shows Raman spectra

of TiO2‐WO3 photocatalysts. In Fig. 7.5(A) vibration peaks at 395, 517 and 638

cm‐1 were observed in both catalysts (P25‐WO3 and NT‐WO3) which are

unambiguously attributed to anatase phase of TiO2 [23,24]. On the other hand,

only for P25 based material, rutile phase is observed as a broad peak at 448 cm‐1

according to the P25 structural composition [23,24]. These results also confirm

the transformation of titanate structure into anatase after the calcination step

[20]. The lowest intensity observed in the Raman spectra for the anatase

vibrations can be related to the lowest crystallinity of NT‐WO3 photocatalyst

according to XRD results [23]. On the other hand, Fig. 7.5(B) shows main W

contributions to the Raman spectra. The band located around 792 cm‐1

corresponds to weak second‐order feature of anatase assigned to the first over‐

tone band at 395 cm‐1 [25]. At around 954 cm‐1 appears another contribution

assigned to the stretching mode of terminal W=O bond which is characteristic of

two‐dimensional tungsten oxide surface species WOX [25,26]. In addition, the

weak band located at around 1042 cm‐1 could be related to the presence of WOX

species in tetrahedral coordination [27]. The observed contributions at 954 and


246

1042 cm‐1 together with the absence of any signal close to 800 cm‐1 suggest that

added W does not contribute to the formation of crystalline WO3 [27]. These

results are in a good agreement with XRD and TEM observations. Again it is

noteworthy the lowest intensity of the NT‐WO3 spectrum as commented above.

20 30 40 50 60 70 80

NT-T

R

AA

RR

AA

A

INT

EN

SIT

Y (

a.u

.)

2 (o)

P25

P25-WO3

NT

NT-WO3

A

R

Figure 7.3. XRD patterns of the catalysts. Crystalline phases detected: Anatase (A); Rutile

(R).



247

Figure 7.4. TEM images of the catalysts.


248

Figure 7.5. Raman spectra of P25‐WO3 and NT‐WO3 photocatalysts.

Surface chemical composition of the catalysts was analyzed by XPS. Full

spectra depicted in Figure 7.6(A‐D) confirm the presence of O, Ti and C in all the

samples (O1s, Ti2p and C1s peaks). In addition, the peaks of W4d and W4f

spectral regions appear in the full spectra of the TiO2‐WO3 photocatalysts

confirming the presence of W in their surfaces. No N signal was detected thus

confirming the absence of N‐doping effect due to the NH3 used for the

incorporation of W in the composite catalysts. The surface content of W was

calculated from peak areas and Wagner atomic sensitive factors [28]. Results are

summarized in Table 7.1 where it can be noticed a lower W surface content for

the NT‐WO3 catalyst. This could be related to the higher porosity of the NT

sample which can favor the incorporation of W in inner regions of the porous

structure. On the other hand, Fig. 7.6(E) shows the high‐resolution Ti2p spectral

region of P25‐WO3 catalyst as an example. The binding energy of the Ti2p1/2 and

Ti2p3/2 core levels at 464.7 eV and 459.0 eV, respectively, together with their

separation of 5.7 eV confirm the valence state of Ti as Ti4+ in TiO2 [29,30]. Fig.

7.6(F) displays the high‐resolution and peak‐fitting results of W4f and Ti3p XPS

spectra of P25‐WO3 catalyst as an example. Analysis of the W4f region is

750 875 1000 1125 1250

1042

954

P25-WO3

NT-WO3

INT

EN

SIT

Y (

a.u.

)RAMAN SHIFT (cm-1)

792

300 400 500 600 700

448

638

517

INT

EN

SIT

Y (

a.u.

)

RAMAN SHIFT (cm-1)

P25-WO3

NT-WO3

395

(A)

(B)



249

complicated due to the interference from the Ti3p level of the TiO2. However,

the position of the Ti3p peak can be fixed and its area calculated from the Ti2p

peaks areas [28]. This makes possible to distinguish between the two signals

(Ti3p and W4f) by deconvolution procedure, and to determine the valence of W

from the position of the W4f level. The binding energy of the peaks located at

37.8 eV and 35.7 eV corresponds to W4f5/2 and W4f7/2 components, respectively,

and their area ratio of 3:4 confirms the presence of W as W6+ in WO3 [18,30].

Similar results were obtained for NT‐WO3 catalyst (not shown).

The diffuse reflectance UV‐Vis spectra of the photocatalysts (Figure 7.7)

showed a higher optical absorbance in the visible region (above ca. 400 nm) for

TiO2‐WO3 catalysts due to the WO3 loading. The optical energy band gap (Eg)

was calculated by means of Tauc’s expression and these results are summarized

in Table 7.1 together with the wavelength of absorption edge. These values are

approximate due to the need of extrapolation of the resulting curve (not shown).

However a value of 3.19 eV was calculated for P25 catalyst similar to 3.2 eV

previously reported for bare TiO2 [11]. It is noticeable a decrease of band gap

energy by coupling the TiO2 with WO3 from ca. 3.2 eV to 3.05 eV and 2.98 eV

calculated for P25‐WO3 and NT‐WO3 catalysts, respectively, which is in

agreement to previously reported values for similar materials [15]. These results

indicate that both P25‐WO3 and NT‐WO3 can be promising photocatalysts to be

used under visible light.


250

Figure 7.6. XPS full spectra of the catalysts (A, B, C and D). High‐resolution XPS spectra

of Ti2p (E) and Ti3p/W4f (F) spectral regions of P25‐WO3 catalyst.

600 500 400 300 200 100 0

NT-WO3

Ti3pW4f

W4dC1s

Ti2p

O1s

BINDING ENERGY (eV)

600 500 400 300 200 100 0

NT

Ti3pC1s

Ti2p

O1s

CP

S (

a.u.

)

BINDING ENERGY (eV)

600 500 400 300 200 100 0

Ti3pW4fC1s

Ti2p

O1s P25-WO3

BINDING ENERGY (eV)

W4d

600 500 400 300 200 100 0

CP

S (

a.u.

)

BINDING ENERGY (eV)

P25O1s

Ti2p

C1sTi3p

468 464 460 456

Ti2p3/2

459.0

CP

S (

a.u.

)

BINDING ENERGY (eV)

Ti2p1/2

464.7

5.7 eV

42 40 38 36 34 32

W4f7/2

35.7

Ti3p37.3

W4f5/2

37.8

BINDING ENERGY (eV)

(A) (B)

(C) (D)

(E) (F)



251

300 350 400 450 500 550 6000.0

0.5

1.0

1.5

AB

SO

RB

AN

CE

(u.

a.)

WAVELENGTH (nm)

P25

P25-WO3

NT

NT-WO3

Figure 7.7. DR UV–Vis spectra of the catalysts.

7.3.2. Visible light response of the photocatalysts

The effectiveness of the catalysts in the use of visible light was tested using

IBP as target compound by cutting off the wavelengths lower than 390 nm of the

Xe lamp in the solar simulator. Results of photocatalytic depletion of IBP and its

mineralization with all the catalysts synthetized are presented in Figure 7.8.

Regarding IBP depletion (Fig. 7.8(A)), once confirmed the absence of IBP

degradation through direct photolysis as expected according to its UV‐Vis

absorption spectrum, it can be noticed the positive effect of WO3 loading on the

photocatalytic activity of TiO2‐WO3 photocatalysts under visible light irradiation

compared to bare TiO2. In addition, it is noticeable the lowest IBP degradation

rate with the sample NT‐T, thus indicating that the heat‐treatment and different

structural properties of calcined NT are not responsible for the high catalytic

activity of the NT‐WO3 photocatalyst. Also, once the N‐doping effect has been

ruled out, the enhancement observed in the photocatalytic activity of the WO3

coupled materials can be related to the lowest band gap energy observed (Table

7.1) due to WO3 presence, which makes them easily excited by visible light [15].


252

In addition, the presence of WO3 can promote the charge transfer between

photogenerated electrons from the conduction band of TiO2 to the WO3

conduction band, accompanied by holes transfer from the valence band of WO3

to the TiO2 valence band. The charge separation mechanism has been previously

reported [15,17,18,27], and provokes an increase in the lifetime of the

photogenerated electron/hole pair avoiding its recombination to some extent. As

a consequence of both higher visible light absorption and lower recombination

rate, the photonic efficiency (i.e. reacted molecules/incident photons) of the

photocatalytic process is increased. The improvement of composite materials

respect their corresponding TiO2 was higher for NT‐WO3 photocatalyst than for

P25‐WO3. Taking into account that NT‐T sample led to a lower degradation rate

than NT, the formation of elongated anatase particles from the titanate

structures seems not to be the reason of this behavior. Thus, it seems plausible

that NT material presented a better distribution and higher dispersion of WO3

particles since also offered a developed porous structure and higher surface area

than P25 together with a large ion‐exchange capacity [19,20]. This hypothetic

higher dispersion together with the slightly higher amount of WO3 in NT‐WO3

(Table 7.1) could be the reasons of the highest visible light absorption capacity

and hence the improvement in NT‐WO3 photocatalytic activity compared to

P25‐WO3, although additional characterization analysis confirming this

hypothesis would be needed.

Regarding TOC evolution (Fig. 7.8(B)) it is noticeable the mineralization rate

observed with the NT‐WO3 photocatalyst reaching 40 % TOC removal in 2 h

whereas P25‐WO3 did not show a significant mineralization degree. These and

the previous results point out the highest photocatalytic activity of NT‐WO3

photocatalyst under visible light radiation.



253

Figure 7.8. Time evolution of IBP dimensionless concentration (A) and dimensionless

TOC (B) during photocatalytic oxidation under visible light radiation (λ = 390 ‐ 800 nm).

Conditions: pH = 6.5 T = 20 ‐ 40 °C, CIBP,0 = 5 mg L‐1, CCAT = 0.5 g L‐1, Qg = 20 L h‐1 (O2).

7.3.3. Photocatalytic degradation of ECs in MWW

The effectiveness of the photocatalysts in a more realistic application was

studied using a MWW effluent as aqueous matrix spiked with IBP, MTP and

CAF. The average values of the main MWW secondary effluent parameters are

summarized in Table 7.2. It can be highlighted the amount of

carbonate/bicarbonate as inorganic carbon (IC) and their buffer role maintaining

the pH of the reaction medium around pH = 8.3.

Regarding the process applied, Table 7.3 summarizes the main results of ECs

removal at 120 min upon the treatments tested, together with Figure 7.9 and

Figure 7.10 that show the time‐evolution of ECs and TOC respectively, only for

NT‐WO3 catalyst as an example. It can be noticed that direct photolysis exerts no

effect on the case of IBP removal and only 5 and 7 % removal was observed for

CAF and MTP, respectively. The time‐evolution of ECs concentration during

photolysis is also shown in Fig. 7.9. Although these compounds do not absorb

radiation in the wavelength range used in this work, MWW content could

provoke indirect photolysis reactions [31,32]. The insignificant mineralization

during photolysis can be observed in Fig. 7.10 and the low value of final TOC

-30 0 30 60 90 1200.0

0.2

0.4

0.6

0.8

1.0

CIB

P/C

IBP

,0

TIME (min)

Dark

-30 0 30 60 90 1200.0

0.2

0.4

0.6

0.8

1.0

Photolysis P25 NT P25-WO3

NT-WO3

NT-T

TO

C/T

OC

0

TIME (min)

Dark

0

0

CT

OC

/CT

OC

,0


254

removal is also presented in Table 7.4. Taking into account that the contribution

of the initial concentration of ECs to the initial TOC content is around 10 % (3.7

mg L‐1), the mineralization observed would be mainly associated to the MWW

TOC content.

Table 7.2. Characterization of MWW effluent before and after some treatments with

NT‐WO3 catalyst (t = 120 min).

PARAMETER Before

treatment*

Photocatalytic

oxidation Ozonation

Catalytic

ozonation

Photocatalytic

ozonation

TOC

(mg C L‐1) 35.3 22.8 26.9 20.2 12.8

IC

(mg C L‐1) 42.0 40.1 38.3 41.5 36.4

pH 8.31 8.34 8.29 8.28 8.33

Absorbance

254 nm 0.253 0.164 0.116 0.058 0.017

COD

(mg O2 L‐1) 51 46 41 33 17

BOD5

(mg O2 L‐1) 32 n.m n.m n.m n.m

Phosphate

(mg L‐1) 4 n.m n.m n.m n.m

*Average values between all batches

n.m. not measured

Regarding the adsorption capacity of the catalysts, none of them gave place

to ECs removal higher than 10 % (Table 7.3). It is noticeable that no significant

changes are observed in terms of ECs adsorption either due to W incorporation

or textural properties modification in NT series, although the modification of the

adsorption capacity of these materials has been reported elsewhere [18].

However, a different behavior is observed for the organic matter content of the

MWW since the adsorption capacity of the catalysts increased around 15 % in

both series, P25 and NT, after W incorporation (see values in Table 7.4). It is also

noticeable a somewhat higher adsorption capacity in NT series respect to P25

due to its more developed porous structure. The fact that NT series, in which a

significant decrease in textural parameters was observed after calcination,

presented the same increase in adsorption capacity than P25 series points out



255

that this phenomenon is mainly related to the W species in the catalysts surface.

In fact, it has been reported that the presence of a monolayer of WOX species on

TiO2 can significantly increase the surface acidity leading to the fact that

composite materials TiO2‐WO3 can adsorb more organic reactants [18,33].

Table 7.3. ECs removal (%) after 120 min reaction upon the different treatments applied.

TREATMENT EC No catalyst

Photolysis

IBP 0

CAF 5

MTP 7

TREATMENT EC P25 P25‐WO3 NT NT‐WO3

Adsorption

IBP 7 8 10 7

CAF 8 7 4 6

MTP 6 8 8 6

Photocatalysis

IBP 73 76 28 70

CAF 77 79 31 82

MTP 92 90 61 91

TREATMENT EC No catalysts/All catalysts

Ozonation

Catalytic ozonation

Photocatalytic ozonation

IBP

> 99.9 CAF

MTP


256

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

CIB

P/C

IBP

,0

TIME (min)

Dark

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

CC

AF/C

CA

F,0

TIME (min)

Dark Adsorption Photolysis Photocatalytic oxidation Ozonation Catalytic ozonation Photocatalytic ozonation

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

CM

TP/C

MT

P,0

TIME (min)

Dark

Figure 7.9. Time evolution of ECs dimensionless concentration during all the treatments

applied to MWW effluent under simulated solar light radiation (λ = 320 ‐ 800 nm) and

NT‐WO3 catalyst. Conditions: pH = 8.3, T = 20 ‐ 40 °C, CEC,0 = 2 mg L‐1, CCAT = 0.5 g L‐1,

CO3,g inlet = 10 mg L‐1, Qg = 20 L h‐1 (O2 or O3/O2).



257

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

CT

OC

/CT

OC

,0

TIME (min)

Dark

Adsorption Photolysis Photocatalytic oxidation Ozonation Catalytic ozonation Photocatalytic ozonation

Figure 7.10. Time evolution of TOC dimensionless concentration during all the

treatments applied to MWW effluent under simulated solar light radiation (λ = 320 ‐ 800

nm) and NT‐WO3 catalyst. Conditions: pH = 8.3, T = 20 ‐ 40 °C, CEC,0 = 2 mg L‐1, CCAT = 0.5

g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 20 L h‐1 (O2 or O3/O2).

Table 7.4. TOC removal (%) after 120 min reaction for the different treatments applied.

TREATMENT No catalysts

Photolysis 4

Ozonation 31

TREATMENT P25 P25‐WO3 NT NT‐WO3

Adsorption 3 17 9 23

Photocatalysis 37 42 31 37

Catalytic

ozonation 23 48 39 47

Photocatalytic

ozonation 47 55 42 64

For the photocatalytic oxidation process with P25 series, as shown in Table


258

7.3 the incorporation of W did not give place to higher ECs removal, attaining

around 75, 80 and 90 % for IBP, CAF and MTP, respectively, with both P25 and

P25‐WO3 catalysts. In spite of the improvement observed by introducing W

when irradiated with visible light, bare P25 presents a higher catalytic activity

under UVA radiation and thus, if the entire simulated solar light spectrum (320 ‐

800 nm) is used, the beneficial effect of W is not noticeable. Similar results were

observed in terms of mineralization (evolution not shown, final value given in

Table 7.4). On the contrary, NT photocatalyst showed the poorest behavior

during the reaction but a significant increase in the photocatalytic activity of NT‐

WO3 respect to NT precursor was noticed. Thus, in this case ECs conversions

around 70, 80 and 90 % for IBP, CAF and MTP, respectively, were achieved,

similar to those reached in P25 series. In addition, regardless of the catalyst used

the order of reactivity during photocatalytic oxidation was MTP > CAF > IBP,

different to that of the rate constants of the direct reaction between these ECs

and hydroxyl radicals, HO , (kHO∙‐IBP = 7.4x109 M‐1 s‐1 [34], kHO∙‐CAF = 5.9x109 M‐1 s‐1

[35], kHO∙‐MTP = 2.1x109 M‐1 s‐1 [36]). This may be an indicator that HO is not the

only responsible of ECs removal and other different ways such as direct h+

oxidation can be taking place.

All the ozone treatments led to complete ECs depletion in less than 45 min of

reaction time (conversion higher than 99.9 %) regardless of the presence/absence

of catalysts and/or radiation in contrast to photocatalytic oxidation. Both direct

and indirect ozone reactions may take place according to the pH of the reaction

medium (pH = 8.3) that was kept constant throughout the reaction time due to

the buffering capacity of the MWW. On the other hand, slight differences can be

noticed among ozone processes with a different behavior mainly depending on

the rate constant of the direct ozone‐EC reaction (kO3‐IBP = 9.1 M‐1 s‐1 [37], kO3‐CAF =

6.5x102 M‐1 s‐1 [35], kO3‐MTP = 6.2x103 M‐1 s‐1 (pH = 8) [38]). Although photocatalytic

ozonation gave place to a faster ECs depletion rate, it is noticeable in Fig. 7.9 that

the differences between ozone‐based processes are only significant for IBP

whose direct ozone reaction presents the lowest rate constant, being practically



259

negligible in the case of CAF and MTP. Main differences between ozone and O3‐

catalytic processes were obtained in terms of TOC removal as shown in Table 7.4

and Fig. 7.10. Ozone alone led to 31 % mineralization mainly during the first

minutes of reaction and then stopped likely due to the formation of compounds

refractory to ozone attack. Production of HO , main responsible species for

mineralization, seems to be insufficient at pH = 8.3 in the ozonation process. The

degree of mineralization decreased in the presence of P25. This material does

not exert any positive effect in catalytic ozonation leading to even lower

mineralization than ozone alone probably due to an inefficient consumption of

O3 onto the catalyst surface. In contrast, NT slightly increased TOC removal up

to 39 % although it can be associated to the higher adsorption capacity of this

material since the mineralization rate is similar to that of the ozonation process

(not shown). On the other hand, W containing photocatalysts, P25‐WO3 and NT‐

WO3, led to higher mineralization (around 47 %), with also higher TOC removal

rate than ozone alone (see Fig. 7.10 for NT‐WO3), thus indicating a catalytic

effect to some extent. This improvement can be related to (1) the catalytic effect

of WO3 and/or (2) the higher adsorption capacity of TiO2‐WO3 catalysts. The

catalytic activity of WO3 in aqueous ozone decomposition, i.e., catalytic

ozonation, has not been extensively reported, but some evidences have been

found. Nishimoto et al. observed a higher TOC removal during phenol

ozonation with WO3 suggesting its role as an ozonation catalyst in a similar

manner to MnO2 or TiO2 [39]. In addition, WO3 has been widely used as the

active component of O3 gas sensors, in which O3 undergoes dissociative

adsorption onto the WO3 surface [40,41]. On the other hand, another plausible

explanation for the catalytic activity of TiO2‐WO3 catalysts during ozonation

compared to bare TiO2 is the enhanced adsorption capacity due to the presence

of WO3 [18,33]. Thus, TiO2 has been widely used as catalyst for ozonation

processes and the proposed mechanisms involve surface reactions between

adsorbed O3 and organic compounds [42]. The fact that bare TiO2 samples

studied here do not show any catalytic activity in the process could be related to


260

an inefficient decomposition of O3 due to the lack of organic matter near the

catalyst surface. However, the enhanced adsorption capacity of TiO2‐WO3

catalysts can lead to an increase in the local concentration of organic matter in

the vicinity of their surface [18], thus allowing surface reactions between

adsorbed O3‐organic compounds. Both hypotheses should be considered but

some additional work to clarify the prevailing mechanism is needed. Finally, the

highest mineralization degree was obtained with the photocatalytic ozonation

process regardless of the catalyst used. This is in agreement with the expected

synergistic effect between ozone and irradiated semiconductors. This synergistic

effect is due to the reaction of O3 with the conduction band electrons of TiO2 and

WO3, or with the superoxide ion radical generated ( ‐2O ), in both cases leading to

the formation of higher concentrations of hydroxyl radicals ( HO ), main

responsible for mineralization [5,8,39].

On the other hand, MWW parameters after the most representative

treatments with NT‐WO3 catalyst are summarized in Table 7.2. As commented

above, photocatalytic ozonation gave place to higher TOC depletion than

individual treatments. Similar trends are observed both for COD and

aromaticity, reaching a 66 % COD reduction in the combined treatment

compared to 10 % in photocatalytic oxidation and 19 % in single ozonation.

Also, the effectiveness of catalytic ozonation can be observed in terms of TOC,

COD and aromaticity. In addition, it can be noticed that the pH remained

unalterable after the treatments applied as a consequence of the buffer effect of

the IC content (from carbonate/bicarbonate).

The comparison of the catalysts in the photocatalytic ozonation process is

shown with more detail in Figure 7.11 and Figure 7.12 which depict ECs

concentration and TOC with time, respectively.



261

-30 0 10 20 30 40 50 600.0

0.2

0.4

0.6

0.8

1.0

Ozonation P25 P25-WO

3

NT NT-WO

3

CIB

P/C

IBP

,0

TIME (min)

Dark

-30 0 10 20 30 40 50 600.0

0.2

0.4

0.6

0.8

1.0

CM

TP/C

MT

P,0

TIME (min)

Dark

-30 0 10 20 30 40 50 600.0

0.2

0.4

0.6

0.8

1.0

CC

AF/C

CA

F,0

TIME (min)

Dark

Figure 7.11. Time evolution of ECs dimensionless concentration during ozonation and

photocatalytic ozonation of MWW effluent under simulated solar light radiation (λ = 320

‐ 800 nm) for all the catalysts studied. Conditions: pH = 8.3, T = 20 ‐ 40 °C, CEC,0 = 2 mg L‐1,

CCAT = 0.5 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 20 L h‐1 (O3/O2).


262

-20 0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Ozonation P25 P25-WO

3

NT NT-WO

3

CT

OC

/CT

OC

,0

TIME (min)

Dark

Figure 7.12. Time evolution of TOC dimensionless concentration during ozonation and

photocatalytic ozonation of MWW effluent under simulated solar light radiation (λ = 320

‐ 800 nm) for all the catalysts studied. Conditions: pH = 8.3, T = 20 ‐ 40 °C, CEC,0 = 2 mg L‐1,

CCAT = 0.5 g L‐1, CO3,g inlet = 10 mg L‐1, Qg = 20 L h‐1 (O3/O2).

Again, it can be noticed that ozone alone was able to completely remove the

ECs in a short time. Differences between ozonation and photocatalytic ozonation

were only observed for IBP and CAF depletion according to the lower rate

constant of their direct ozone reactions. In these cases, indirect and

photocatalytic contributions are more important than for MTP. It can be also

noticed the highest catalytic activity of WO3 composite materials compared to

TiO2 precursors although the main differences were observed in TOC evolution

(Fig. 7.12). In addition to their higher adsorption capacity, the highest

mineralization rate was observed for the TiO2‐WO3 catalysts. These materials

take the advantage of using a greater fraction of solar light radiation in the

visible region due to the presence of WO3 as demonstrated in section 7.3.2, but

also a significant contribution of dark catalytic ozone reactions. The best

performance was attained with NT‐WO3 photocatalyst which led to 64 % TOC

removal. The improvement observed in the NT series after introducing the WO3



263

is remarkably higher than in the P25 series, in good agreement with the results

obtained under visible light, probably due to a better dispersion of WO3. To

elucidate the reaction mechanism of photocatalytic ozonation using NT‐WO3

catalyst it is necessary to consider the different processes involved and will be

the subject of future work.

7.4. CONCLUSIONS

TiO2‐WO3 composite catalysts have been synthetized with around 4 wt. % of

WO3 and visible light response, and successfully applied for the removal of

emerging contaminants in urban wastewater and mineralization of the effluent

through photocatalytic ozonation. Ozone alone is able to completely remove the

emerging contaminants in the wastewater matrix but the mineralization degree

reached is relatively low. Photocatalytic ozonation gives place to the highest

mineralization rate regardless of the catalyst used. Titanate nanotubes structure

is not efficient in the photocatalytic oxidation and photocatalytic ozonation

processes, but it is a good precursor of composite catalysts due to its textural

and surface properties. Titanate nanotubes gave place to a composite catalyst

NT‐WO3 with elongated anatase particles, a well‐developed porous structure

and high dispersion of WOX species showing visible light absorption capacity.

The best performance of the NT‐WO3 catalyst compared to bare TiO2 is related

to several mixed contributions such as the use of a greater fraction of solar light

radiation, higher organic compounds adsorption capacity and catalytic activity

in ozone reactions. The contribution of the different phenomena to elucidate the

reaction mechanism with this catalyst will be the subject of future work.

AKNOWLEDGEMENTS

This work has been supported by the Spanish Ministerio de Economía,

Industria y Competitividad (MINECO) and European Feder Funds through the

project CTQ2012‐35789‐C02‐01. Authors acknowledge the SACSS‐SAIUEX and

UAI‐ICP for the characterization analyses. A. Rey thanks the University of


264

Extremadura for a postdoctoral research contract. M.D. Hernández‐Alonso

thanks MINECO for the award of her postpoctoral contract from the “Ramón y

Cajal” program. E. Mena thanks the Consejería de Empleo, Empresa e

Innovación (Gobierno de Extremadura) and European Social Fund for providing

her a predoctoral FPI grant (Ref. PD12059).

REFERENCES

[1] Santos, J.L.; Aparicio, I.; Callejón, M.; Alonso, E. “Occurrence of

pharmaceutically active compounds during 1‐year period in wastewaters from

four wastewater treatment plants in Seville (Spain)”. J. Hazard. Mater. 164 (2009)

1509‐1516.

[2] Barceló, D.; Petrovic, M. (Eds.). “Emerging contaminants from industrial and

municipal wastes. Occurrence, analysis and effects”. The Handbook of

Environmental Chemistry 5‐S1. Springer, Berlin (Germany), 2008.

[3] Halling‐Sorensen, B.; Nors‐Nielsen, S.; Lanzky, P.F.; Ingerslev, F.; Holten

Lützhoft, H.C.; Jorgensen, S.E. “Occurrence, fate and effects of pharmaceutical

substances in the environment‐‐a review”. Chemosphere 36 (1998) 357‐393.

[4] Esplugas, S.; Bila, D.M.; Krause, L.G.T.; Dezotti, M. “Ozonation and

advanced oxidation technologies to remove endocrine disrupting chemicals

(EDCs) and pharmaceuticals and personal care products (PPCPs) in water

effluents”. J. Hazard. Mater. 149 (2007) 631‐642.




[6] Rodríguez, E.M.; Fernández, G.; Alvarez, P.M.; Beltrán, F.J. “TiO2 and Fe (III)

photocatalytic ozonation processes of a mixture of emergent contaminants of

water”. Water Res. 46 (2012) 152‐166.

[7] Rivas, F.J.; Beltrán, F.J.; Encinas, A. “Removal of emergent contaminants:

Integration of ozone and photocatalysis”. J. Environ. Manage. 100 (2012) 10‐15.




Chem. Eng. J. 204‐206 (2012) 131‐140.

[9] Hernández‐Alonso, M.D.; Coronado, J.M.; Maira, A.J.; Soria, J.; Loddo, V.;

Augugliaro, V. “Ozone enhanced activity of aqueous titanium dioxide



265

suspensions for photocatalytic oxidation of free cyanide ions”. Appl. Catal. B

Environ. 39 (2002) 257‐267.




[11] Hernández‐Alonso, M.D.; Fresno, F.; Suárez, S.; Coronado, J.M.



[12] Shifu, C.; Lei, C.; Shen, G.; Gengyu, C. “The preparation of coupled

WO3/TiO2 photocatalyst by ball milling”. Powder Technol. 160 (2005) 198‐202.

[13] Kwon, Y.T.; Song, K.I.; Lee, W.I.; Choi, G.J.; Do, Y.R. “Photocatalytic

behavior of WO3‐loaded TiO2 in an oxidation reaction”. J. Catal. 191 (2000) 192‐

199.

[14] Tryba, B.; Piszcz, M.; Morawski, A.W. “Photocatalytic activity of TiO2‐WO3

composites”. Int. J. Photoenergy (2009) (Article ID 297319) 1‐7.

[15] Ismail, M.; Bousselmi, L.; Zahraa, O. “Photocatalytic behavior of WO3‐

loaded TiO2 systems in the oxidation of salicylic acid”. J. Photochem. Photobiol.

A Chem. 222 (2011) 314‐322.

[16] Xiao, M.; Wang, L.; Huang, X.; Wu, Y.; Dang, Z. “Synthesis and

characterization of WO3/titanate nanotubes nanocomposite with enhanced

photocatalytic properties”. J. Alloys Compd. 470 (2009) 486‐491.

[17] Iliev, V.; Tomova, D.; Rakovsky, S.; Eliyas, A.; Li Puma, G. “Enhancement of

photocatalytic oxidation of oxalic acid by gold modified

WO3/TiO2 photocatalysts under UV and visible light irradiation”. J. Mol. Catal.

A Chem. 327 (2010) 51‐57.

[18] Tomova, D.; Iliev, V.; Rakovsky, S.; Anachkov, M.; Eliyas, A.; Li Puma, G.

“Photocatalytic oxidation of 2,4,6‐trinitrotoluene in the presence of ozone under

irradiation with UV and visible light”. J. Photochem. Photobiol. A Chem. 231

(2012) 1‐8.

[19] Hernández‐Alonso, M.D.; García‐Rodríguez, S.; Sánchez, B.; Coronado, J.M.

“Revisiting the hydrothermal synthesis of titanate nanotubes: new insights on

the key factors affecting the morphology”. Nanoscale 3 (2011) 2233‐2240.

[20] Hernández‐Alonso, M.D.; García‐Rodríguez, S.; Suárez, S.; Portela, R.;

Sánchez, B.; Coronado, J.M. “Highly selective one‐dimensional TiO2‐based

nanostructures for air treatment applications”. Appl. Catal. B Environ. 110

(2011) 251‐259.


266


method”. Water Res. 15 (1981) 449‐456.

[22] Sing, K.S.W.; Everett, D.H.; Haul, R.A.W.; Moscou, L.; Pierotti, R.A.;

Rouquerol, J.; Siemieniewska, T. “Reporting physisorption data for gas/solid

systems with special reference to the determination of surface area and

porosity”. Pure Appl. Chem. 57 (1985) 603‐619.

[23] Yan, J.; Wu, G.; Guan, N.; Li, L.; Li, Z.; Cao, X. “Understanding the effect of

surface/bulk defects on the photocatalytic activity of TiO2: anatase versus rutile”.

Phys. Chem. Chem. Phys. 15 (2013) 10978‐10988.

[24] Arabatzis, I.M.; Antonaraki, S.; Stergiopoulos, T.; Hiskia, A.;

Papaconstantinou, E.; Bernard, M.C.; Falaras, P. “Preparation, characterization

and photocatalytic activity of nanocrystalline thin film TiO2 catalysts towards

3,5‐dichlorophenol degradation”. J. Photochem. Photobiol. A Chem. 149 (2002)

237‐245.

[25] Vuurman, M.A.; Wachs, I.E.; Hirt, A.M. “Structural determination of

supported vanadium pentoxide‐tungsten trioxide‐titania catalysts by in situ

Raman spectroscopy and x‐ray photoelectron spectroscopy”. J. Phys. Chem. 95

(1991) 9928‐9937.

[26] Santato, C.; Odziemkowski, M.; Ulmann, M.; Augustynski, J.

“Crystallographically oriented mesoporous WO3 films: Synthesis,

characterization, and applications”. J. Am. Chem. Soc. 123 (2001) 10639‐10649.

[27] Akurati, K.K.; Vital, A.; Dellemann, J.P.; Michalow, K.; Graule, T.; Ferri, D.;

Baiker, A. “Flame‐made WO3/TiO2 nanoparticles: Relation between surface

acidity, structure and photocatalytic activity”. Appl. Catal. B Environ. 79 (2008)

53‐62.

[28] Wagner, C.D.; Davis, L.E.; Zeller, M.V.; Taylor, J.A.; Raymond, R.H.; Gale,

L.H. “Empirical atomic sensitivity factors for quantitative analysis by electron

spectroscopy for chemical analysis”. Surf. Interface. Anal. 3 (1981) 211‐225.

[29] Xiao, M.W.; Wang, L.; Huang, X.J.; Wu, Y.D.; Dang, Z. “Synthesis and

characterization of WO3/titanate nanotubes nanocomposite with enhanced

photocatalytic properties”. J. Alloys Compd. 470 (2009) 486‐491.

[30] Li, Y.; Chen, L.; Guo, Y.; Sun, X.; Wei, Y. “Preparation and characterization

of WO3/TiO2 hollow microsphere composites with catalytic activity in dark”.

Chem. Eng. J. 181‐182 (2012) 734‐739.

[31] Ryan, C.C.; Tan, D.T.; Arnold, W.A. “Direct and indirect photolysis of

sulfamethoxazole and trimethoprim in wastewater treatment plant effluent”.



267

Water Res. 45 (2011) 1280‐1286.

[32] Jacobs, L.E.; Weavers, L.K.; Houtz, E.F.; Chin, Y.P. “Photosensitized

degradation of caffeine: Role of fulvic acids and nitrate”. Chemosphere 86 (2012)

124‐129.

[33] Kwon, Y.T.; Song, K.Y.; Lee, W.I.; Choi, G.J.; Do, Y.R. “Photocatalytic

behavior of WO3‐loaded TiO2 in an oxidation reaction”. J. Catal. 191 (2000) 192‐

199.

[34] Huber, M.M.; Canonica, S.; Park, G.Y.; Von Gunten, U. “Oxidation of

pharmaceuticals during ozonation and advanced oxidation processes”. Environ.

Sci. Technol. 37 (2003) 1016‐1024.

[35] Broséus, R.; Vincent, S.; Aboulfadl, K.; Daneshvar, A.; Sauvé, S.; Barbeau, B.;

Prévost, M. “Ozone oxidation of pharmaceuticals, endocrine disruptors and

pesticides during drinking water treatment”. Water Res. 43 (2009) 4707‐4717.

[36] Sires, I.; Oturan, N.; Oturan, M.A. “Electrochemical degradation of β‐

blockers. Studies on single and multicomponent synthetic aqueous solutions”.

Water Res. 44 (2010) 3109‐3120.

[37] Huber, M.M.; Gobel, A.; Joss, A.; Hermann, N.; Loffler, D.; McArdell, C.;


pharmaceuticals during ozonation of municipal wastewater effluents: A pilot


[38] Benítez, F.J.; Acero, J.L.; Real, F.J.; Roldán, G. “Ozonation of pharmaceutical

compounds: Rate constants and elimination in various water matrices”.

Chemosphere 77 (2009) 53‐59.




[40] Guerin, J.; Aguir, K.; Bendahan, M. “Modeling of the conduction in a WO3

thin film as ozone sensor”. Sensor. Actuat. B Chem. 119 (2006) 327‐334.

[41] Oison, V.; Saadi, L.; Lambert‐Mauriat, C.; Hayn, R. “Mechanism of CO and

O3 sensing on WO3 surfaces: First principle study”. Sensor. Actuat. B Chem. 160

(2011) 505‐510.

[42] Kasprzyk‐Hordern, B.; Ziólek, M.; Nawrocki, J. “Catalytic ozonation and

methods of enhancing molecular ozone reactions in water treatment”. Appl.

Catal. B Environ. 46 (2003) 639‐669.

CAPÍTULO 8 (CHAPTER 8) PAPER 6: Reaction mechanism and kinetics of DEET visible light assisted photocatalytic ozonation with WO3 catalyst

E. Mena, A. Rey, E.M. Rodríguez, F.J. Beltrán

Applied Catalysis B: Environmental 202 (2017) 460-472

ABSTRACT. This work is focused on the mechanistic investigation of N,N-diethyl-meta-

toluamide (DEET) degradation by photocatalytic ozonation, using WO3 in suspension

and visible radiation (wavelength ≥ 390 nm). This combined process proved to be an efficient treatment to completely remove the contaminant, HO• radicals being the main species involved. Different oxidation products were identified by liquid chromatography time-of-flight mass spectrometry and ion chromatography analyses, and the evolution of their relative abundances with reaction time was established. The efficiency of photocatalytic ozonation treatment was pointed out not only in the DEET depletion rate but also in the evolution of the main intermediate species and mineralization. All the large intermediates initially formed were completely removed within 60 min reaction time and only short-chain organic acids with very low toxicity remained in solution at concentrations in agreement with the mineralization degree achieved (up to 60 % in 2 h). A reaction mechanism of photocatalytic ozonation of DEET involving different chemical reaction steps, with the final formation of short-chain organic acids and mineralization to

CO2, has been proposed. A lumped kinetic model based on TOC and hydroxyl radical

reaction was developed to simulate DEET, intermediates and short-chain organic acids evolution in terms of TOC that provides a simplified approach for this process.

Keywords: Photocatalytic ozonation, DEET, WO3, mechanism, kinetic.

PAPER 6: Reaction mechanism and kinetics of DEET

visible light assisted photocatalytic ozonation with WO3 catalyst

271

8.1. INTRODUCTION

Pharmaceutical and personal care products (PPCPs) are emerging

contaminants of increasing environmental concern due to their continuous

discharge into the aquatic environment, persistence and toxic effect in human

and wildlife health. Among them, N,N‐diethyl‐meta‐toluamide (DEET) has been

widely used as the active compound in insect repellents for protection against

mosquito [1]. Due to its extensive use, in the last few years DEET has been

detected in different aquatic systems, wastewater treatment plant effluents and

even in drinking water treated, thus indicating its recalcitrance to conventional

treatments [2,3]. Although DEET is considered to be slightly toxic to aquatic

invertebrates and freshwater fish [2], it has been reported to have potential

carcinogenic properties in human nasal mucosal cells [4]. In addition, ingestion

of low doses of DEET in children has been reported to result in coma and

seizures [5]. Therefore, it is critical to develop a fundamental understanding of

the fate and degradation of DEET during water treatment processes.

DEET degradation through different ozone‐based processes like ozonation

[4], ozone combined with hydrogen peroxide and/or UV radiation [6], and

photocatalytic treatments has been studied [1,3,6‐8], the last being the most

efficient whereas single ozonation resulted to be ineffective for being so slow.

However, the combination of ozone and heterogeneous photocatalysis

(photocatalytic ozonation), has demonstrated to lead to higher mineralization

rates compared to the individual treatments, due to the generation of higher

concentration of oxidizing species, mainly hydroxyl radicals, and also the

inhibition of e‐/h+ pair recombination to some extent [9,10]. Photocatalytic

ozonation of DEET has been previously investigated using TiO2 as photocatalyst

and UV (λ = 254 nm) as radiation source [6]. For the practical deployment of

photocatalytic technologies, the use of natural solar radiation would be an

advantage. In this line, to improve the efficiency of the process, in a previous

work photocatalytic ozonation of DEET was studied using different forms of

WO3, a visible‐light‐responsive semiconductor. According to this work, a


272

monoclinic WO3 catalyst with W reduced states gave rise to complete DEET

removal and high mineralization degree in less than 2 h [11].

However, it is known that the degradation of DEET during ozonation or

photocatalytic oxidation proceeds through complex redox reactions involving

different steps which can lead to the formation of different intermediates (from

now on transformation products, TPs) if complete mineralization is not achieved

[1,4]. Some of these TPs can be toxic and, in some cases, more persistent than the

parent compound [1]. Thus, the identification of these TPs and their reactivity is

a key step in optimizing the processes conditions in order to increase their

efficiency. To the best of our knowledge, although there are studies dealing with

the identification of TPs generated from DEET during ozonation [4,12] and

heterogeneous photocatalysis with TiO2 [1,8], there are no works focused on the

identification of DEET photocatalytic ozonation TPs. Thus, the purpose of this

work is to study the visible light assisted photocatalytic ozonation of DEET

using WO3 as catalyst with special interest in: 1) identification of DEET TPs and

other intermediates; 2) determination of main species involved in DEET

degradation with the aid of different scavengers; and 3) development of a

kinetic model for the process based on the previous results.


8.2.1. Experiments

All the experiments were carried out in semi‐batch mode in a 0.5 L effective

volume glass‐made spherical reactor, provided with a gas inlet, a gas outlet and

a liquid sampling port. The reactor was placed in the chamber of a solar

simulator (Suntest CPS, Atlas) provided with a 1500 W Xe lamp with emission

restricted to wavelengths above 390 nm using a polyester cut‐off filter (Edmund

Optics). The spectral irradiance of incident radiation shown in Figure 8.1(A),

was measured with a spectral‐radiometer Black Comet C (StellarNet), provided

with an optic fiber for the wavelength range between 190 ‐ 850 nm. The



273

irradiation intensity was set at 550 W m−2 and the temperature increased from 20

‐ 40 °C throughout the experiments. A typical temperature vs time profile is

shown in Figure 8.1(B). An ozone generator (Anseros Ozomat Com AD‐02) was

used to produce a gaseous ozone‐oxygen stream that was fed to the reactor.

200 300 400 500 600 700 8000.0

0.2

0.4

0.6

0.8

1.0

NO

RM

ALI

ZE

D L

IGH

T IN

TE

NS

ITY

AB

SO

RB

AN

CE

(u

.a.)

WAVELENGTH (nm)

0.0

0.2

0.4

0.6

0.8

1.0

0 20 40 60 80 100 12015

20

25

30

35

40

45

T (

º C)

TIME (min)

Figure 8.1. (A) Spectral irradiance of Xe light after passing through the polyester filter

(line) and absorption spectra of DEET (dotted line). (B) Evolution of the temperature of

the reaction media with time at the experimental conditions applied in this work.


274

In a typical ozonation experiment, the reactor was loaded with 0.5 L of an

aqueous solution containing 15 mg L‐1 of DEET (pH0 = 6) and covered with

aluminum foil when needed (experiments in the dark). Then, the Xe lamp was

switched on and at the same time an ozone‐oxygen mixture (10 mg L‐1 ozone

concentration) was fed to the reactor at a flow rate of 15 L h−1. Photolysis and

photocatalytic oxidation of DEET were also performed in absence of ozone. For

catalytic and photocatalytic experiments the same procedure was followed but

0.125 g of the catalyst were added and the suspension was stirred for 30 min

before switching on the lamp and/or opening the gas flow (O2 or O2/O3), in order

to attain the adsorption equilibrium. The catalyst used was a monoclinic WO3‐

microspheres sample synthetized by sol‐gel procedure and calcined at 600 °C as

reported in a previous work [11]. Adsorption of DEET onto the catalyst was also

tested in absence of ozone in the dark.

On the other hand, in some experiments, appropriate amounts of scavengers

(tert‐butyl alcohol (t‐BuOH) and oxalate) were added to the DEET aqueous

solution in order to determine the nature of the main species involved in DEET

degradation. An additional experiment of photocatalytic ozonation of p‐

chlorobenzoic acid (pCBA, 5 mg L‐1) in the presence of oxalic acid (0.01 M) was

carried out at similar operating conditions for kinetics considerations. In all

cases the time for each oxidation experiment was 2 h. At different intervals

samples were withdrawn from the reactor and filtered through a 0.2 μm PET

membrane.

8.2.2. Analytical methods


using a Phenomenex C‐18 column (5 μm, 150 mm long and 3 mm diameter) as

stationary phase and 0.6 mL min‐1 of acetonitrile‐acidified water (0.1 % formic

acid) as mobile phase (30 ‐ 70 v/v, isocratic). Identification and quantification

was carried out at 220 nm. The concentration of pCBA was analyzed in the same

system with the same column and mobile phase at 0.7 mL min‐1, and its



275

identification and quantification was carried out at 238 nm. Total organic carbon

(TOC) was measured using a Shimadzu TOC‐VSCH analyzer. Aqueous ozone

concentration was measured by the indigo method using a UV‐Vis

spectrophotometer (Evolution 201, Thermospectronic) set at 600 nm [13]. In this

assay, 3 mL of sample were mixed with 3 mL of indigo solution and then filtered

to remove catalyst particles. Spectrophotometric measurements were

immediately carried out to avoid possible interferences [14]. Concentration of

ozone in the gas phase was continuously monitored by means of an Anseros

Ozomat GM‐6000 Pro analyzer. Hydrogen peroxide concentration was

determined photometrically by the cobalt/bicarbonate method at 260 nm using

the same UV‐Vis spectrophotometer [15].

DEET organic TPs were identified by HPLC‐qTOF using an Agilent 6530

accurate mass quadrupole time‐of‐flight mass spectrometer bearing with

electrospray ionization (ESI) source coupled with an Agilent 1260 series LC

system. A ZORBAX SB‐C18 column (3.5 μm, 150 mm long, and 4.6 mm

diameter) was used as stationary phase. The column was kept at constant

temperature of 30 °C during each analysis. As mobile phase, 0.2 mL min−1 of

acetonitrile‐acidified water (25 mM formic acid) was used from 10 to 100 % of

acetonitrile in 40 min with 15 min of equilibration. The injection volume was 10

μL. The qTOF instrument was operated in the 4 GHz high‐resolution mode. Ions

are generated using an electrospray ion source Dual ESI. Electrospray conditions

were the following: capillary, 3500 V; nebulizer, 30 psi; drying gas, 10 L min‐1;

gas temperature, 350 °C; skimmer voltage, 65 V; octapoleRFPeak, 750 V;

fragmentor, 175 V. The mass axis was calibrated using the mixture provided by

the manufacturer over the m/z 70 ‐ 3200 range. A sprayer with a reference

solution was used as continuous calibration in positive ion using the following

reference masses: m/z 121.0509 and 922.0098. Data were processed using Agilent

Mass Hunter Workstation Software (version B.04.00).

According to previous works, the generation of formaldehyde during DEET

degradation is expected [1,8]. Thus, concentration of formaldehyde in solution


276

was measured by Hantzsch reaction, measuring the absorbance at 412 nm of the

diacetyldihydrolutidine formed [16]. Finally, short‐chain organic acids and

inorganic ions were analyzed by an ion chromatograph with chemical

suppression (Metrohm 881 Compact Pro) provided with a conductivity detector

using a MetroSep A sup 5 column (250 mm long, 4 mm diameter) at 45 °C and

0.7 mL min‐1 of Na2CO3 from 0.6 ‐ 14.6 mM in 50 min (10 min post‐time for

equilibration) as mobile phase.


8.3.1. Comparison of processes

In a first series of experiments, the effectiveness of adsorption, photolysis

under λ 390 nm (Ph), ozonation (O3), photolytic ozonation (Ph‐O3),

photocatalytic oxidation (PhC‐O2), catalytic ozonation (C‐O3) and photocatalytic

ozonation (PhC‐O3) on DEET removal was studied. On the one hand, from the

obtained results (not shown) it is deduced that contribution of both adsorption

onto the catalyst and direct photolysis to DEET removal is negligible, the latter

as expected since there is not overlapping between DEET absorption and lamp

emission spectra, as deduced from Fig. 8.1(A). On the other hand, time‐

evolution of normalized DEET and TOC concentration for the different

processes studied is shown in Figures 8.2(A) and 8.2(B), respectively. According

to these figures, degradation rate of DEET during photocatalytic oxidation (PhC‐

O2) was very low. Thus, less than 10 % of DEET conversion was attained after

120 min and no mineralization was observed. Taking into account that WO3 is a

visible light responsive semiconductor, these poor results could be attributable

to the fact that oxygen is not able to react with photoexcited electrons in the

conduction band of WO3 [17], leading to a high recombination rate of e‐/h+ pairs

and, as a consequence, presenting a low efficiency in the photocatalytic

oxidation process.



277

Figure 8.2. Evolution of DEET (A) and TOC (B) normalized concentration with time

during the different treatments applied: PhC‐O2; O3; Ph‐O3; C‐O3; PhC‐O3.

Experimental conditions: CDEET,0 = 15 mg L‐1; pH0 = 6; V = 0.5 L; Qg = 15 L h‐1;

*CO3,g = 10 mg L‐1; *I = 550 W m−2; *CWO3 = 0.25 g L‐1 (*if applied).

Regarding single ozonation (O3), at the conditions tested the conversion of

DEET after 2 h was 100 % as shown in Fig. 8.2(A). Given the low reactivity of

ozone towards DEET (kO3 = 0.123 M‐1 s‐1 at 20 °C; [4]), the results suggest the

participation of ozone indirect reactions, namely hydroxyl radical generation

from O3 decomposition, on the elimination of the contaminant. Besides, the

0 30 60 90 1200.0

0.2

0.4

0.6

0.8

1.0

CD

EE

T/C

DE

ET

,0

TIME (min)

(A)

0 30 60 90 1200.0

0.2

0.4

0.6

0.8

1.0

CT

OC/C

TO

C,0

TIME (min)

(B)


278

relatively low TOC removal achieved (~ 20 % after 2 h, see Fig. 8.2(B)) could be

related to the formation of intermediates refractory to ozone direct reaction [18].

All these aspects are discussed in depth in subsequent sections.

From Fig. 8.2 it can also be seen that addition of WO3 catalyst (C‐O3 system)

had no effect on the effectiveness of the ozonation in terms of DEET and TOC

removal, which indicates the inability of the catalyst to promote an efficient O3

decomposition in the dark.

On the other hand, when light (λ 390 nm) and ozone were combined (Ph‐

O3), DEET degradation rate was higher than the observed for the O3 system in

the dark. In this sense, from Fig. 8.2(A) it is deduced that the time needed to

reach a given DEET conversion by Ph‐O3 was almost half the time needed in the

dark. Besides, in line with the above the mineralization was also higher (Fig.

8.2(B)), reaching a TOC removal of 18 % and 25 % in absence and presence of

light, respectively, after 2 h of treatment. Since the direct photolysis of the

pollutant does not occur, the increased efficiency of the Ph‐O3 system compared

to O3 should be related to the interaction between ozone and light leading to the

formation of reactive species as has been previously reported [19,20].

Undoubtedly, addition of WO3 to the Ph‐O3 system, that is, photocatalytic

ozonation (PhC‐O3) led to the best results among the processes tested. As

observed in Fig. 8.2, after 15 min DEET was completely removed and the

mineralization degree was higher than 60 % after 2 h. These results suggest that,

unlike oxygen, recombination of e‐/h+ generated from WO3 photoexcitation is

avoided thanks to ozone electron trapping [17]. On its role as an electron

acceptor ozone would decompose into reactive oxygen species (mainly •HO )

thus increasing the degradation and mineralization rate of the pollutant. These

results are consistent with the evolution of dissolved ozone shown in Figure 8.3.

As observed, in the presence of light the concentration of ozone in solution

diminishes and more markedly when WO3 is also present. Therefore, it seems



279

that both light (λ 390 nm) and WO3 + light promote ozone decomposition

which, in turn, leads to the formation of highly reactive species that would be

responsible for the faster elimination rate of DEET and TOC observed. Another

aspect to highlight from Fig. 8.3 is the time profile of dissolved ozone (O3d)

concentration during single ozonation. As observed, it reaches a maximum at

the beginning of the experiment and then decreases gradually until reaching a

stationary value. This behavior differs from that usually observed during

ozonation of water solutions where O3d concentration is low at the beginning

and gradually increases reaching a stationary value. The explanation to this fact

lies in the temperature profile inside the reactor shown in Fig. 8.1(B). During the

first 60 min, as the temperature increases from ~ 20 °C to 40 °C ozone solubility

decreases. From that moment, since the temperature remains almost constant a

stationary value of ozone in solution is also maintained. In addition to ozone

solubility, temperature can also affect the kinetics of ozone reactions whereas its

effect on the reactivity of transient species is much lower. With this in mind, the

rate constant of the direct reaction between DEET and ozone at 15 °C, 25 °C and

40 °C has been determined by a direct method [18]. Taking into account the low

kO3 values expected according to the literature (0.123 M‐1 s‐1 at 20 °C, [4]), and the

fact that DEET does not dissociate in water, the study was performed at pH 2

(pH adjustment with perchloric acid) in a thermostated bubble column (reaction

volume 250 mL) operating under semi‐batch mode by continuous injection of 15

L h‐1 of a gaseous mixture containing 10 mg L‐1 of O3. In these experiments t‐

BuOH was used as scavenger of •HO radicals that could be generated from O3

decomposition. According to the second‐order rate constants values reported in

the literature (see Table 8.1) for the reaction of both compounds with O3 and

•HO [4,21‐23], the initial concentration of DEET and t‐BuOH was set at 4x10‐5 M

and 4x10‐3 M, respectively. At these conditions, according to Table 8.1, reaction

of DEET with •HO will be avoided whereas t‐BuOH will not be able to compete

with DEET for ozone. Also, the experimental conditions aimed to allow DEET

and ozone to react in the liquid bulk (i.e., slow kinetic regime of ozone


280

absorption was established) so that the DEET mass balance in the

semicontinuous perfectly mixed reactor used would be given by Eq. (8.1):

3

3d

O ‐DEETDEETO DEET

DEET

kdC‐ C C

dt z (8.1)

where CO3d and CDEET are the concentration of ozone and DEET in the liquid,

respectively; and zDEET the stoichiometry of the reaction between O3 and DEET,

which was considered to be 1 mol of O3 consumed per mol of DEET according to

its low reactivity and the structure of DEET molecule. Integration of Eq. (8.1)

leads to Eq. (8.2) which, in case CO3d remains constant, simplifies to Eq. (8.3):

0

3 3d

0

tDEET

O ‐DEET ODEET

t

CLn k C dt

C (8.2)

0

3 3d

DEETO ‐DEET O

DEET

CLn k C t

C (8.3)

According to Eq. (8.3), a plot of the left term against (CO3dt) should lead to a

straight line that intercepts the origin and whose slope is the value of kO3‐DEET.

As an example, in Figure 8.4 the results obtained at 25 °C are shown. As

observed in Fig. 8.4(A) during the run CO3d was practically constant from t = 5

min (CO3d = 3.03 (± 0.18) x 10‐5 M). On the other hand, from experimental data

fitting to Eq. (8.3) (shown in Fig. 8.4(B)) a kO3‐DEET value of 4.24 ± 0.17 M‐1 s‐1 (R2 =

0.985) was obtained at 25 °C. Following a similar procedure, kO3‐DEET values of

2.46 ± 0.09 M‐1 s‐1 and 7.05 ± 0.33 M‐1 s‐1 were obtained at 15 °C and 40 °C,

respectively. To validate these rate constant values, the slow kinetic regime of

ozone absorption (Hatta number, Ha < 0.3, [18]) was verified taking into account

Eq. (8.4) [4]:

O ‐DEET DEET O3 3

L

k C DHa

k

(8.4)

where DO3 is the diffusivity of ozone in water (1.76x10‐9 m2 s‐1; [24]); and kL the



281

individual liquid‐side mass transfer coefficient calculated to be 3.74x10‐4 m s‐1 by

applying the Calderbank’s Equation [25]. At the conditions applied in this work

Ha was always much lower than 0.3, allowing us to validate the kO3‐DEET values

summarized in Table 8.1. As deduced from Table 8.1, although in all cases the

low reactivity of ozone towards DEET is highlighted, kO3‐DEET values obtained in

this work are considerable greater than that reported by Benítez et al. [4]. This

could be attributable to the low CDEET0/Ct‐BuOH0 ratio applied by these authors at

which the alcohol acts as both •HO and ozone scavenger. Finally, the kO3‐DEET

values obtained at different temperatures were fitted to the Arrhenius equation.

The pre‐exponential factor and the activation energy are also indicated in Table

8.1.

0 15 30 45 60 75 90 105 1200

1x10-5

2x10-5

3x10-5

4x10-5

5x10-5

CO

3d (

M)

TIME (min)

Figure 8.3. Evolution of dissolved ozone concentration with time during the application

of different processes. Symbols: O3; Ph‐O3; PhC‐O3. Experimental conditions:

CDEET,0 = 15 mg L‐1; pH0 = 6; V = 0.5 L; Qg = 15 L h‐1; CO3,g = 10 mg L‐1; *I = 550 W m−2;

*CWO3 = 0.25 g L‐1 (*if applied).


282

0 10 20 30 40 50 600.0

1.0x10-5

2.0x10-5

3.0x10-5

4.0x10-5

5.0x10-5

C (

M)

TIME (min)

(A)

0.00 0.02 0.04 0.06 0.080.0

0.1

0.2

0.3

0.4

ln(C

DE

ET

,0/C

DE

ET)

CO3d·t (M·s)

Slope = 4.24 +/- 0.17

R2 = 0.985

(B)

Figure 8.4. Determination of kO3‐DEET at 25 °C. (A) Evolution of DEET () and O3d ()

concentration with time. (B) Fitting of data to Eq. (3). Experimental conditions:

CDEET,0 = 10 mg L‐1; Ct‐BuOH = 0.004 M; pH0 = 2; V = 0.25 L; Qg = 15 L h‐1; CO3,g = 10 mg L‐1.



283

Table 8.1. Second‐order rate constants for the reaction of ozone and hydroxyl radicals

with DEET, t‐BuOH, oxalate and p‐CBA.

Compound kO3 (M‐1 s‐1) Ref. kHO∙ (M‐1 s‐1) Ref.

DEET

20 °C: 0.121 ± 0.005

Conditions: pH = 2 ‐ 9;

CDEET,0 = 10‐6 M; Ct‐BuOH,0 = 0.1 M

[4]

4.95x109

[21]

15 °C: 2.46 ± 0.09 (R2 = 0.991)

25 °C: 4.24 ± 0.17 (R2 = 0.985)

40 °C: 7.05 ± 0.33 (R2 = 0.987)

kO3 = Ae‐Ea/RT; A = 1.2x106;

Ea = 31.3 kJ mol‐1 (R2 = 0.988)

Conditions: pH = 2, CDEET,0 = 4x10‐5 M;

Ct‐BuOH,0 = 4x10‐3 M

This

work

t‐BuOH 0.0011 [22] 6.2x108 [23]

Oxalate < 0.04 [30]

5x107

(pH = 2.8)

7.7x106 (pH = 6)

[31]

p‐CBA 0.015 [39] 5x109 [39]

8.3.2. Determination of the main species responsible for DEET degradation

and mineralization

In order to determine the nature of the species involved on the elimination of

DEET through the most effective systems, namely ozonation, photolytic

ozonation and photocatalytic ozonation, a new series of experiments was carried

out by adding substances capable of acting as scavengers of some of the species

whose generation and participation is expected. In this sense, t‐BuOH has been

selected as scavenger of •HO in the liquid bulk, and oxalate as scavenger of •

HO

both in the liquid bulk and also at the catalyst surface where, in addition,

adsorbed oxalate can be oxidized by the photogenerated positive holes (h+) [26‐


284

29]. As commented in the previous section, to select the appropriate substances

to be used as scavengers their reactivity towards the different species and

reagents involved must be firstly considered. Thus, in Table 8.1 the rate constant

of the reactions oxalate‐ozone and oxalate‐ •HO are also indicated [30,31]. Taking

into account the reactivity of DEET and the scavengers towards the different

species involved, this study was carried out at the following conditions: pH0 = 6;

20 ‐ 40 °C (see Fig. 8.1(B)); CDEET,0 = 8x10‐5 M; COxal,0 = 10‐3 M and/or Ct‐BuOH,0 = 0.03

M. Figure 8.5 shows the influence of the presence/absence of these scavengers in

DEET degradation rate by the different processes applied: O3 (Fig. 8.5(A)), Ph‐O3

(Fig. 8.5(B)) y PhC‐O3 (Fig. 8.5(C)). The theoretical DEET evolution with time has

also been calculated from Eq. (8.1) considering that DEET was only degraded by

direct ozone attack. For this purpose CDEET,0 value, the evolution of both CO3d

(see Fig. 8.3) and temperature (Fig. 8.1(B)) with time as well as the influence of T

on kO3‐DEET according to Arrhenius equation (see Table 8.1) have been taken into

account. These results have also been added to Fig. 8.5 for comparative reasons.

Fig. 8.5(A) shows the effect of the presence of 0.03 M t‐BuOH in 8x10‐5 M

DEET elimination rate by single ozonation. At these conditions, t‐BuOH will

react with all •HO whereas its reaction with O3 will be negligible. As observed in

Fig. 8.5(A), the presence of t‐BuOH negatively affect the elimination rate of the

contaminant, thus indicating not only the decomposition of ozone into •HO

radicals at the conditions tested but also the high contribution of •HO to DEET

degradation by single ozonation. Moreover, the good agreement between CDEET

evolution when t‐BuOH was present and that calculated by Eq. (8.1) indicates

that O3 and •HO are the only species involved on DEET degradation by single

ozonation.



285

0 15 30 45 60 75 90 105 1200

2x10-5

4x10-5

6x10-5

8x10-5

1x10-4

CD

EE

T (

M)

TIME (min)

(A)

0 15 30 45 60 75 90 105 1200

2x10-5

4x10-5

6x10-5

8x10-5

1x10-4

(B)

CD

EE

T (

M)

TIME (min)

0 15 30 45 60 75 90 105 1200

2x10-5

4x10-5

6x10-5

8x10-5

1x10-4

(C)

CD

EE

T (

M)

TIME (min) Figure 8.5. Influence of the presence of different scavengers on the elimination of DEET

by O3 (A), Ph‐O3 (B) and PhC‐O3 (C). Symbols: no scavenger; t‐BuOH 0.03 M;

t‐BuOH 0.03M and Oxalate 10‐3 M. Line: Calculated from Eq. (1) assuming only direct

ozone‐DEET reaction. Experimental conditions: CDEET,0 = 15 mg L‐1; pH0 = 6; V = 0.5 L;

Qg = 15 L h‐1; CO3,g = 10 mg L‐1; *I = 550 W m−2; *CWO3 = 0.25 g L‐1 (*if applied).


286

Similarly, the influence of 0.03 M t‐BuOH in the degradation of 8x10‐5 M

DEET by photolytic ozonation is depicted in Fig. 8.5(B). Again, the presence of

t‐BuOH negatively affects the elimination rate of the contaminant. However, the

elimination rate of DEET calculated from Eq. (8.1) (straight line) is clearly lower

than that experimentally determined in the presence of t‐BuOH. Since it is

expected that at the conditions used most of the •HO was trapped by t‐BuOH,

there must be other species and/or mechanism different than •HO and O3 also

contributing to DEET degradation. This effect is more marked during the first 45

min and then, the rate of DEET depletion becomes similar (see Fig. 8.5(B), Eq.

(8.1) and t‐BuOH experiment). The fact that the positive effect of the presence of

light on TOC removal is much lower than on DEET (Fig. 8.2), that is, the effect is

more noticeable at short reaction times, could indicate the formation of an

intermediate capable of acting as a photosensitizer giving rise to the formation

of reactive species. Once the intermediate is degraded the beneficial effect of

light would disappear.

Finally, the effect of the presence of oxalate and/or t‐BuOH on the rate of

DEET removal by photocatalytic ozonation is shown in Fig. 8.5(C). It can be

observed that the degradation rate of DEET in the presence of t‐BuOH or

t‐BuOH + oxalate was very similar. Considering that t‐BuOH only reacts with

•HO in the bulk, these results are pointing out that •

HO and h+ photogenerated

at the catalyst surface do not contribute to DEET degradation. Moreover, it can

be observed that 100 % DEET removal was attained in less than 15 min by PhC‐

O3 when no scavengers are used, but the evolution of DEET during the same

period when t‐BuOH was present was virtually coincident with that calculated

when only its direct reaction with ozone is considered (Eq. (8.1)). Thus, •HO in

the bulk seems to be the only species involved in DEET elimination by PhC‐O3.

Therefore, as deduced from results in Fig. 8.5, at the conditions used in this

work, during photocatalytic ozonation DEET is mainly removed through its



287

reaction with •HO radicals in the bulk whose formation when visible radiation,

WO3 catalyst and O3 are combined (PhC‐O3) is highly enhanced. The highest

•HO concentration would also be the responsible for the highest efficiency of this

combination in terms of mineralization (Fig. 8.2(B)). In other words, the above

results confirm that not only DEET but also final products oxidation (mainly

carboxylic acids) proceeds through •HO attack during PhC‐O3 process.

Moreover, first TPs degradation is also likely due to •HO reaction in the bulk as

discussed in the following section.

8.3.3. Identification of TPs

The identification of TPs during ozonation and photocatalytic ozonation was

carried out by means of LC‐qTOF and ion chromatography. LC‐qTOF approach

provided accurate mass measurements of ions (m/z values) of the different

compounds formed, together with several peaks corresponding to equal m/z

values. In total, 22 TPs were identified, which are indicated in Table 8.2 together

with the parent compound DEET. They were designated by a number

corresponding to the m/z value and for isomers the number is followed by a

letter in alphabetical order with increasing retention time. The low experimental

relative mass errors obtained indicate the high grade of confidence in the

assignment of the elemental composition. In this sense, relative mass errors

below 5 ppm are generally accepted for the verification of the elemental

composition [32]. Table 8.2 also includes the tentative structures proposed for

the different TPs according to the specific literature [1,4,8,12,21,33,34]. These TPs

identified, similar to those found in previous works for different systems, are

consistent with the attack of •HO generated throughout the processes. Although

neither the tentative structures nor the concentration of the TPs could be

confirmed due to the lack of standards, reaction pathways for DEET

degradation have been proposed and are discussed in the following section.

In general, main first intermediates detected were large molecules, such as


288

C12H15NO2 (m/z = 206), C12H17NO2 (m/z = 208), C10H15NO2 (m/z = 182),

C10H15NO3 (m/z = 198) or C10H13NO (m/z = 164), in which practically the parent

compound structure (C12H17NO, m/z = 192) is still present with slight

modifications. Other smaller intermediates like C5H11NO2 (m/z = 118) or C4H11N

(m/z = 74) that have been identified are formed after the cleavage of

characteristic bonds.

Table 8.2. TPs identified by LC‐qTOF during DEET O3 and PhC‐O3 degradation.

Compound tR

(min)

Molecular

Formula

[M]

Experimental

m/z

[M‐H+]

Error

(ppm) Tentative structure

192 (DEET) 24.56 C12H17NO 192.1385 ‐2.65

240‐A 16.05 240.1234 ‐10.41

240‐B 16.92 C12H17NO4 240.1239 ‐8.32

240‐C 17.4 240.1245 ‐5.83

226‐A 15.38

C11H15NO4

226.1083 ‐4.05

226‐B 15.83 226.1085 ‐4.93

222‐A 18.88

C12H15NO3

222.1128 ‐3.29

222‐B 20.13 222.1129 ‐3.85

214 15.96 C10H15NO4 214.1072 ‐4.74

208‐A 17.54

C12H17NO2

208.1337 ‐2.38

208‐B 19.24 208.1339 ‐3.34

208‐C 20.64 208.1326 2.91

208‐D 21.69 208.1338 ‐2.86



289

Table 8.2 (Continuation). TPs identified by LC‐qTOF during DEET O3 and PhC‐O3

degradation.

Compound tR

(min)

Molecular

Formula

[M]

Experimental

m/z

[M‐H+]

Error

(ppm) Tentative structure

206 21.03 C12H15NO2 206.1177 ‐4.58

198‐A 13.69 198.1129 ‐2.17

198‐B 15.78 C10H15NO3 198.1129 ‐2.17

198‐C 16.83 198.1130 ‐2.68

182 22.38 C10H15NO2 182.1177 ‐5.19

178 22.70 C11H15NO 178.1229 ‐1.46

164 21.48 C10H13NO 164.1069 ‐2.50

146 9.23 C6H11NO3 146.0812 ‐2.26

118 6.55 C5H11NO2 118.0863 ‐2.42

74 6.62 C4H11N 74.0967 ‐6.74

The evolution with time of the intensity of the signal of DEET and those TPs

detected by LC‐qTOF during O3 and PhC‐O3 processes is shown in Figure 8.6.

According to Fig. 8.6 TPs formed were exactly the same regardless of the process

applied although their formation rate was noticeably much faster during the

photocatalytic ozonation treatment. Thus, during PhC‐O3 total TPs

disappearance was observed at 60 min whereas for single ozonation some of

N

O

C N

O

HO

O


290

these TPs still remained after 120 min.

0 15 30 45 60 75 90 105 1200

1x106

2x106

3x106

4x106

5x106

6x106

7x106P

EA

K A

RE

A

TIME (min)

0

1x108

2x108

3x108

4x108

5x108

6x108

(A)

0 15 30 45 60 75 90 105 1200

1x106

2x106

3x106

4x106

5x106

6x106

7x106

PE

AK

AR

EA

TIME (min)

(B)

0

1x108

2x108

3x108

4x108

5x108

6x108

Figure 8.6. Time profiles of the intensity of the main peaks detected during (A) O3 and (B)

PhC‐O3 of DEET. Experimental conditions: CDEET,0 = 15 mg L‐1; pH0 = 6; V = 0.5 L;

Qg = 15 L h‐1; CO3,g = 10 mg L‐1; *I = 550 W m−2; *CWO3 = 0.25 g L‐1 (*if applied).

As deduced in the previous section, the fact that •HO radicals in the bulk are

the main species involved in DEET degradation by O3 and PhC‐O3 could

explain, on the one hand, that TPs formed are the same in both processes and,

206 164 182 198-A 198-B 198-C 214 74 146 178

192 (DEET)



291

on the other hand, that TPs formation rate in the photocatalytic treatment

presents the highest values as a result of the largest concentration of •HO in the

reaction medium. Moreover, once the maximum concentration of each TP is

reached, the removal rate during the PhC‐O3 process is also faster than the

observed during O3 alone process. In summary, during single ozonation direct

O3 reactions are favored as a consequence of the higher concentration of

dissolved molecular ozone and, according to Fig. 8.6, a higher removal rate of

TPs is clearly obtained by PhC‐O3 system. Then, it can be pointed out that, as for

DEET, in both treatments (O3 and PhC‐O3) the •HO radical is the main

responsible species for the degradation of intermediate TPs.

Further oxidation of these TPs usually gives place to the formation of short‐

chain saturated carboxylic acids. Thus, the presence of oxalic, acetic, pyruvic and

formic acids (see molecular formula and structures in Table 8.3) was detected at

long reaction times for both O3 and PhC‐O3 processes. In Figure 8.7, the

evolution with time of CTOC present as carboxylic acids (ΣCTOC‐acids) is shown. As

expected, a higher generation rate and concentration in solution of these acids

for PhC‐O3 is observed being 18 % and 64 % of residual TOC in the form of these

organic acids after 2 h of ozonation and photocatalytic ozonation, respectively.

By comparing Fig. 8.6 and Fig. 8.7, it is deduced that formation of carboxylic

acids needs the previous oxidation of TPs, which is faster for photocatalytic

ozonation as commented before. Hence, the low TOC conversion into carboxylic

acids after 2 h of ozonation indicates that at these conditions intermediate TPs,

that could even be more harmful than the parent compound, are still present.

During photocatalytic ozonation, only short‐chain organic acids were detected

at the end of the treatment, with the complete elimination of the large TPs. Thus,

according to the literature [1,4], a significant decrease in the toxicity is expected

if this process is applied until reaching a high mineralization degree. However,

the information currently available about ecotoxicity is not sufficient and further

work would be necessary [2].


292

Table 8.3. Short‐chain saturated organic acids identified during DEET O3 and PhC‐O3

degradation.

Compound Molecular

Formula

Structure

Oxalic acid C2H2O4

Acetic acid C2H4O2

Pyruvic acid C3H4O3

Formic acid CH2O2

0 15 30 45 60 75 90 105 1200

1

2

3

4

CT

OC

-aci

ds(m

g L-1

)

TIME (min)

Figure 8.7. Evolution of TOC corresponding to short‐chain saturated organic acids

concentration with time during different processes. Symbols: O3; PhC‐O3.

Experimental conditions: CDEET,0 = 15 mg L‐1; pH0 = 6; V = 0.5 L; Qg = 15 L h‐1;

CO3,g = 10 mg L‐1; *I = 550 W m−2; *CWO3 = 0.25 g L‐1 (*if applied).



293

0 15 30 45 60 75 90 105 1200

2x10-5

4x10-5

6x10-5

8x10-5

C (

M)

TIME (min)

Figure 8.8. Evolution of DEET eliminated and H2O2 and CH2O generated with time

during DEET degradation by O3 (circles) and PhC‐O3 (triangles) systems. Symbols: ,

DEET eliminated; , H2O2 formed; , CH2O generated. Experimental conditions:

CDEET,0 = 15 mg L‐1; pH0 = 6; V = 0.5 L ; Qg = 15 L h‐1; CO3,g = 10 mg L‐1; *I = 550 W m−2;

*CWO3 = 0.25 g L‐1 (*if applied).

According to the literature, ozonation and photocatalytic oxidation of DEET

led to the generation of formaldehyde (CH2O) [1,8]; and also, hydrogen peroxide

(H2O2) is commonly formed through direct ozone and hydroxyl radical reactions

[35,36]. Thus, the evolution of the concentration of both compounds with time is

depicted in Figure 8.8 together with the evolution of DEET eliminated during O3

and PhC‐O3. It can be clearly seen for both treatments that the disappearance of

1 mol of DEET leads to the formation and accumulation of ~1 mol of H2O2 in the

reaction medium. Particularly, during PhC‐O3 an increase of H2O2 concentration

is produced up to 45 min coinciding with the time when TPs were almost

completely removed (see Fig. 8.6(B)). From this moment, H2O2 is gradually

consumed likely in photocatalytic reactions as discussed in a previous work [11].

However, in case H2O2 reactions led to the formation of reactive oxidizing

species, they do not seem to participate in TPs removal. In fact, the analysis of

H2O2 evolution during DEET degradation by the other processes applied in this


294

work, together with the results during photocatalytic oxidation of DEET adding

H2O2 (PhC‐O2 + H2O2, not shown) led us to conclude that: 1) H2O2

decomposition during the processes applied mainly occurs due to its interaction

with the irradiated catalyst surface; and 2) H2O2 decomposition does not lead to

the generation of reactive species capable of oxidizing DEET and/or its TPs.

Regarding formaldehyde, a direct relation between DEET degradation and

formaldehyde formation is also observed although in this case the ratio is ~ 0.25

mol of CH2O per mol of DEET eliminated by O3 and PhC‐O3.

8.3.4. Proposed reaction mechanism

Bearing in mind the identified TPs and the fact that hydroxyl radical seems to

be the main species involved on DEET degradation by PhC‐O3 when WO3 and

light of wavelength higher than 390 nm are used, a tentative mechanism is

presented in Schemes 8.1 ‐ 8.3 and discussed below. Both the aliphatic chain and

the aromatic ring of DEET molecule are found to react with hydroxyl radicals

and pathways will be separately discussed. For the reactions at the aliphatic

chain of DEET, the identified TPs and the proposed mechanisms are presented

in Scheme 8.1. In this mechanism, the •HO radical attack on the aliphatic chain

could occur by abstraction of hydrogen to form an organic radical in which

several TPs are based on [12]. For example, the TP named 206 in Table 8.2 with

m/z 206.1177 was generated by a reaction between this organic radical and

oxygen or ozone. On the other hand, the rearrangement of the radical formed

and an additional •HO attack at the aliphatic chain lead to its de‐ethylation to

form another radical, which could further react with water to form the TP

named 164 (m/z 164.1069). Other authors have also reported the existence of

these TPs through the above two different pathways during ozonation [4], and

photocatalytic oxidation of DEET [1,8]. In addition, the rearrangement of the

organic radical could lead to the cleavage of characteristic bonds [12] and

subsequent formation of carbanions which could explain the formation of the



295

smaller TPs detected in the present work. On the one hand, the rearrangement

of the radical produced a new radical in addition to the carbanion of

diethylamine. From the •HO radical attack on this carbanion and reaction with

water, diethylamine can be formed (TP named 74 in Table 8.2 with m/z 74.0967).

On the other hand, the rearrangement of the radical can also form the benzyl

radical (involved in further reactions) and a carbanion which could react with

•HO radicals to produce hydroxydiethylamide (TP 118 in Table 8.2 with m/z

118.0863). Finally, in Scheme 8.1 is also considered the initial •HO attack and

hydrogen abstraction on the methyl group that is directly attached to the

aromatic ring of DEET. The TP 178 with m/z 178.1229 is derived by detachment

of this methyl group from DEET [1,8], through loss of formaldehyde after a new

•HO radical attack.

Regarding the reactions on the aromatic ring of DEET (see Scheme 8.2), the

initial attack of the hydroxyl radical could lead to the generation of

hydroxylated DEET derivatives. Thus, in the present study four isomers have

been identified for monohydroxy DEET (208‐A,B,C,D in Table 8.2 with m/z

208.1326 ‐ 208.1339) corresponding to the different positions that can be

occupied by the •HO moiety in the aromatic ring, as it has been reported for the

degradation of DEET by anodic Fenton, photolytic processes or photocatalytic

treatments [8,21,33,34]. Monohydroxy DEET further reacts with •HO to form

dihydroxylated and trihydroxylated DEET. In this sense, three different isomers

of trihydroxy DEET have been detected (240‐A,B,C in Table 8.2 with m/z

240.1234 ‐ 240.1245). In addition, hydroxylated derivatives of DEET degradation

products were detected. For example, the isomers 222‐A,B in Table 8.2 with m/z

222.1128 ‐ 222.1129 could be formed either by hydroxylation of the 206 TP, or

from isomers 208‐A,B,C,D through oxidation of the aliphatic chain. In the same

way, the isomers 226‐A,B (m/z 226.1083 ‐ 226.1085) can be generated by

hydroxylation of the 178 TP and also from isomers 240‐A,B,C by cleavage of the

C‐C bond and subsequent detachment of the methyl group from the aromatic


296

ring. These were also identified in previous works [1,4,8].

Besides, in Scheme 8.2, some of the identified TPs resulted from the opening

of the aromatic ring due to its breakdown through the •HO radical consecutive

attack on the same carbon atom [8]. Thus, the TP named 182 with m/z 182.1177

was assigned to mono‐oxygenated ring opening TP. After the double attack of

•HO on DEET and subsequent ring opening, the decarboxylation involving the

loss of CO2 and H2O would produce TP 182 [37] whose mono‐ and

dihydroxylation could lead to the formation of TPs 198‐A,B,C and 214 (with m/z

198.1129 ‐ 198.1130 and 214.1072), respectively. On the other hand, ozonolysis of

TP 182 could lead to the formation of the TP 146 (m/z 146.0812) through loss of

hydrogen peroxide and further •HO radical attack [18].

Finally, some of the TPs generated could react with ozone and/or hydroxyl

radicals to produce short‐chain saturated carboxylic acids (oxalic, acetic, pyruvic

and formic acids have been detected) which, given their low reactivity towards

ozone, eventually would evolve to CO2 and H2O by reaction with •HO (see

Scheme 8.3).



297

Scheme 8.1. Proposed DEET degradation pathway through •

HO attack on the aliphatic

chain.


298

Scheme 8.2. Proposed DEET degradation pathway through the •

HO attack on the

aromatic ring.



299

Scheme 8.3. Formation of saturated carboxylic acids and mineralization of DEET.

8.3.5. Kinetic study

According to the above results, during photocatalytic ozonation DEET

degradation and mineralization is mainly due to •HO radical reactions without a

significant direct participation of neither ozone nor positive holes generated on

the irradiated photocatalyst surface. Thus, the mechanism of generation of

hydroxyl radicals can be described by the following reactions and mass transfer

steps:

Ozone mass transfer from the gas flow to the liquid phase:

‐3 ‐1

Lk a 5.1x10 s3g 3O O

(8.5)

where kLa is the volumetric mass‐transfer coefficient that was experimentally

determined through O3 absorption experiments for the reaction system used in

this work according to Beltrán in ref. [18].

Ozone dark decomposition:

‐1 ‐1

d1k 70 M s‐ ‐

3 2 2O HO HO O

(8.6)

6 ‐1 ‐1

d2k 2.2x10 M s ‐‐

2 3 2 3HO O HO O HO

(8.7)

‐pKa 11.4

2 2 2H O H HO (8.8)

O3/ HO● HO●


300

••pKa 4.8 ‐

2 2HO O H (8.9)

9 ‐1 ‐1

d3k 1.6x10 M s‐ ‐

3 32O O O HO

(8.10)

Photocatalytic reactions:

WO3h , ‐

3WO e h (8.11)

10 1 ‐1

d4k 3.6x10 M s‐ ‐

3 3e O O HO

(8.12)

rk‐3e h WO (8.13)

In this mechanism, ozonide radical ( 3O ) readily evolves to form hydroxyl

radical in reactions (8.7), (8.10) and (8.12); oxygen is not able to react with

photogenerated electrons according to [17], so this reaction has not been taken

into account. In addition, although H2O2 is decomposed onto irradiated WO3

surface, the results obtained in terms of DEET degradation and mineralization in

experiments conducted by combining PhC‐O2 + H2O2 (not shown) indicate that

no reactive species are formed during H2O2 photocatalytic decomposition.

Hydroxyl radicals may react with dissolved ozone and hydrogen peroxide

generated according to reactions (8.14) and (8.15):

9 ‐1 ‐1

d5k 3.0x10 M s

3 2 2O HO HO O

(8.14)

7 ‐1 ‐1

d6k 2.7x10 M s

2 2 2 2H O HO HO H O

(8.15)

All the known rate constants have been previously summarized in [18]. The

rate constant of ozone‐electrons reaction has been taken for hydrated electrons

from Buxton et al. [31].

In the presence of any organic contaminant at sufficient concentration,

hydroxyl radicals will mainly react with it. In terms of TOC, the reaction

mechanism of DEET and TOC removal can be described according to the

following lumped reactions:



301

HO ‐TOC1

k

1 2TOC HO TOC (8.16)

HO ‐TOC 2k

2 3 2TOC HO TOC 1 ‐ CO (8.17)

HO ‐TOC 3

k

3 2TOC HO CO

(8.18)

where TOC1 represents DEET expressed in carbon units; TOC3 is the total

organic carbon corresponding to the sum of short‐chain saturated organic acids

and formaldehyde in solution detected during photocatalytic ozonation; and

TOC2 is the total organic carbon corresponding to the sum of the rest of DEET

intermediates detected by LC‐qTOF. In reaction (8.17), the stoichiometric

parameter α represents the TOC2 fraction that is converted to TOC3 whereas

(1‐α) is mineralized to CO2.

TOC2 can be calculated from the TOC balance:

t 1 2 3TOC TOC TOC TOC (8.19)

where TOCt is the analyzed total organic carbon in solution at a given time, t.

Mass balances of each species can be established as follows:

Ozone in the gas phase:

3g 3g

3gi 3g 3d

O O

g O g O L O

C R T dCC ‐ C ‐ V k a ‐C 1 ‐ V

He dt

(8.20)

where g is the gas flow rate; CO3gi and CO3g are the molar ozone concentrations

in the gas phase at the reactor inlet and outlet, respectively; CO3d is the molar

concentration of dissolved ozone; β is the liquid holdup; V the reaction volume;

R the ideal gas constant; He is the Henry law constant for the ozone‐water

system at the temperature T.


302

Ozone in the liquid phase:

3g3d

‐3d 4 3d 5 3d

L OOL O d O d Oe HO

k a C R TdC‐ k a C ‐ k C C ‐ k C C

dt He

(8.21)

where Ce‐ is the concentration of photogenerated electrons; CHO∙ the

concentration of hydroxyl radicals; and kd4 and kd5 the rate constants of

reactions (8.12) and (8.14), respectively. In this balance, the contributions of

reactions (8.6), (8.7) and (8.10) have been neglected according to the pH of the

reaction medium (from 6 to 4).

Photogenerated electrons:

‐

3 4 3d

neWO a d O re h e

dCI k C C k C C

dt (8.22)

where WO3 is the apparent quantum yield of WO3 at the wavelength range used

in this work; Ia is the absorbed radiation flux by the catalyst WO3; the exponent

n is the order of the reaction with respect to the Ia and depends on the efficiency

of e‐/h+ formation and recombination at the catalyst surface taking a value

between 0.5 and 1 when the reaction is kinetically controlled [38]; Ch+ is the

concentration of photogenerated holes and kr is the rate constant of e‐/h+

recombination reaction (8.13). If ozone concentration is high enough, the

recombination process is minimized so it can be considered that all the

photogenerated electrons are trapped by dissolved ozone. Then Eq. (8.22) can be

simplified as:

‐

3 4 3d

neWO a d O e

dCI k C C

dt (8.23)

Hydroxyl radicals:

4 3d 5 3d 6 2 2

1 21 2

33

HOd O d O d H Oe HO HO

TOC TOCHO TOC HO HO TOC HO

TOCHO TOC HO

dCk C C k C C k C C

dtk C C k C C

k C C

(8.24)



303

where CH2O2 is the concentration of hydrogen peroxide in the liquid phase;

CTOC1, CTOC2 and CTOC3 are the concentrations corresponding to TOC1, TOC2 and

TOC3, respectively; and kd6, kHO∙‐TOC1, kHO∙‐TOC2, kHO∙‐TOC3 are the rate constants of

reactions (8.15), (8.16), (8.17) and (8.18), respectively.

Eq. (8.24) can be rewritten as follows:

4 3d

HOd O i ie HO i

dCk C C C k C

dt

(8.25)

where the subscript i represents any species that reacts with hydroxyl radicals.

For Eq. (8.23) and (8.25), net reaction rates were assumed to be zero

according to the hypothesis of the stationary state for transient species. Thus,

hydroxyl radical concentration can be expressed as follows:

nWO a3 i

HOsi ii

I rC

kk C

(8.26)

In Eq. (8.26) the numerator ri represents the reaction rate of initiation for the

formation of hydroxyl radicals, whereas the denominator kS represents the

scavenging factor, taking into account that the species involved in this term are

only the inhibitors of the chain mechanism of ozone decomposition into •HO .

Total organic carbon mass balances:

1

1 11 1

TOC iTOC TOCHO TOC HO HO TOC

s

dC rk C C k C

dt k (8.27)

2

1 21 2

1 21 2

TOC


iTOC TOCHO TOC HO TOC

s

dCk C C k C C

dt

rk C k C

k

(8.28)


304

3

2 32 3

2 32 3

TOC


iTOC TOCHO TOC HO TOC

s

dCk C C k C C

dt

rk C k C

k

(8.29)

where kHO∙‐TOC1 is the rate constant of DEET‐ •HO reaction; kHO∙‐TOC2 represents

the apparent rate constant of the reaction between TOC2 (DEET intermediates

detected by LC‐qTOF) and •HO radicals; and kHO∙‐TOC3 represent an apparent

rate constant of the reaction between compounds of TOC3 (formaldehyde and

oxalic, acetic, pyruvic and formic acids) and •HO .

Firstly, the hypothesis of working in excess of ozone was proved by means of

two different experiments with different ozone concentration at the reactor inlet

(10 and 20 mg L‐1). Results of DEET degradation and mineralization are

represented in Figure 8.9(A) where it can be noticed the similar evolution of

DEET and TOC regardless of the ozone concentration used.

Thus, to solve the proposed model for TOC evolution, the initiation rate ri

was determined in an experiment of photocatalytic ozonation of pCBA in the

presence of oxalic acid. In this system, pCBA is a •HO probe which has very low

reactivity with ozone but readily reacts with •HO (see rate constants in Table 8.1)

[39]. Oxalic acid was used in this experiment at high concentration as •HO

scavenger. Thus, in the presence of both compounds, the evolution of pCBA can

be expressed as:

pCBA ipCBA pCBAHO pCBA HO HO pCBA

s

ipCBAHO pCBA

OxalHO Oxal

dC rk C C k C

dt k

rk C

k ∙C

(8.30)

where CpCBA and COxal are the concentrations of pCBA and oxalic acid,

respectively; kHO∙‐pCBA and kHO∙‐Oxal are the rate constants in Table 8.1 (kHO∙‐Oxal =

4.6x106 M‐1 s‐1, at pH = 4). In this equation, oxalic acid is considered as an



305

inhibitor of the chain mechanism of ozone decomposition into •HO radicals [40].

Integration of Eq. (8.30) leads to Eq. (8.31) which simplifies to Eq. (8.32) since

oxalic acid remains virtually constant while pCBA is depleted and ri will be

constant at the conditions used:

0

0

tpCBA HO ‐pCBA i

pCBA OxalHO ‐Oxal t

kC rLn dt

C k C

(8.31)

0pCBA HO ‐pCBAi

pCBA OxalHO ‐Oxal

kCLn r t

C k C

(8.32)

The representation of Eq. (8.32) is depicted in Fig. 8.9(B), and the calculated

value for ri is indicated in Table 8.4.

Table 8.4. Kinetic parameters and correlation coefficient of the model proposed for

PhC‐O3 DEET mineralization

Parameter Value

ri (M s‐1) 2.07x10‐8

kHO∙‐TOC1 (M‐1 s‐1) 5x109

kHO∙‐TOC2 (M‐1 s‐1) 1.3x109

kHO∙‐TOC3 (M‐1 s‐1) 4x108

α 0.62

R2 0.993


306

0 1x107 2x107 3x107 4x1070.0

0.2

0.4

0.6

0.8

1.0

ln(C

pCB

A ,

0/C

pCB

A)

kHO· - pCBA·t/(kHO· - Oxal·COxal) (M-1s)

Slope = 2.07x10-8 +/- 3.69x10-10

R2 = 0.996

(B)

0 15 30 45 60 75 90 105 1200.0

0.2

0.4

0.6

0.8

1.0

C/C

0

TIME (min)

(A)

Figure 8.9. Determination of kinetic parameters. (A) Time evolution of DEET (triangles)

and TOC (circles) normalized concentrations during PhC‐O3 at different O3 inlet gas

concentration. Symbols: , 10 mg L‐1; , 20 mg L‐1. (B) ri determination from pCBA‐

oxalic acid experiment through data fitting to Eq. (32). (C) ks determination from TOC

evolution. Symbols: TOC1 (line shows fitting of Eq. (33)); TOCt (line shows fitting to

apparent 1st order kinetics); ks. Experimental conditions: CDEET,0 = 15 mg L‐1; pH0 = 6;

V = 0.5 L ; Qg = 15 L h‐1; CO3,g = 10 mg L‐1; *I = 550 W m−2; *CWO3 = 0.25 g L‐1;

*CpCBA,0 = 5 mg L‐1; *COxal,0 = 0.01 M (*if applied).

0 1000 2000 3000 4000 5000 6000 70000.0

0.5

1.0

1.5

2.0

ks=3.67x104·exp(1.84x10-4·t)

Slope (TOCt) = 1.84x10-4+/- 6.35x10-6

R2 = 0.992

ln(C

TO

C,0

/CT

OC

)

Time (s)

Slope (TOC1) = 2.82x10-3 +/- 6.24x10-5

R2 = 0.998

(C)

2.0x104

4.0x104

6.0x104

8.0x104

1.0x105

1.2x105

1.4x105

k s (

s-1)



307

During photocatalytic ozonation of DEET, the scavenging factor, kS could not

be considered constant. This scavenging factor will involve all the species

reacting with hydroxyl radicals that inhibit the ozone decomposition

mechanism. During the first moments, when DEET concentration is high and no

mineralization is observed, kS can be determined from integrated Eq. (8.27):

10

11

TOCi

HO ‐TOCTOC s

C rLn k ∙t

C k (8.33)

As observed in Fig. 8.9(C), kS was found to be 3.67x104 s‐1 (for the initial

moments of reaction) from plotting and fitting experimental data to Eq. (8.33).

However, since the composition of the reaction medium changes with time, this

kS value is not valid for the entire reaction period. In general, as reaction

progresses, the concentration of organic intermediates with inhibiting character

such as oxalic or acetic acid increase. Thus, taking into account that the main

organics involved in the scavenging factor are constituents of TOCt, an

approximation was done taking into account that kS undergoes an exponential

variation with time inversely to TOCt, at t = 0 being kS,0 = 3.67x104 s‐1, as

calculated for the beginning of the experiment. The evolution of ln(CTOC,0/CTOC)

and kS with time is depicted in Fig. 8.9(C), the selected fitting equation for kS

being:

4 4sk 3.67x10 exp 1.84x10 t) (8.34)

With all these calculated data, differential Eqs. (8.27), (8.28) and (8.29) can be

simultaneously solved. For that purpose, kHO∙‐TOC3 apparent rate constant of the

reaction between compounds of TOC3 (formaldehyde and oxalic, acetic, pyruvic

and formic acids) and •HO radicals, has been calculated as an average value

taking into account their individual rate constants [31]. Values of kHO∙‐TOC1 and

kHO∙‐TOC3 are summarized in Table 8.4. The other values, kHO∙‐TOC2 and α, were

used as fitting parameters to minimize the difference between experimental data


308

and those calculated with the proposed model. The resolution of the differential

equations was performed by a fourth‐order Runge‐Kutta method with

Micromath Scientist 3.0 software. Figure 8.10 shows the experimental and

calculated TOC evolution. As observed, the proposed kinetic model fits fairly

well the experimental results (also confirmed by the R2 value in Table 8.4). In

addition, as expected, for the reaction between TOC2 and •HO radicals the value

of the rate constant obtained (kHO∙‐TOC2) is lower than that of DEET‐ •HO reaction

but higher than that of TOC3‐ •HO , which is in accordance with the general trend

of subsequent lower reactivity of the TPs formed in these type of reactions [31].

On the other hand, 62 % of the TOC2 is transformed in TOC3 whereas 38 % can

be mineralized according to the apparent stoichiometric coefficient α.

0 15 30 45 60 75 90 105 1200

2

4

6

8

10

12

CT

OC

(mg

L-1

)

TIME (min)

Figure 8.10. Evolution of TOC with time during DEET degradation by PhC‐O3. Symbols:

TOCt; TOC1 (DEET); □ TOC2 (OPs); TOC3 (Carboxylic acids and formaldehyde);

lines are fitting results from the TOC depletion model. Experimental conditions:

CDEET,0 = 15 mg L‐1; pH0 = 6; V = 0.5 L ; Qg = 15 L h‐1; CO3,g = 10 mg L‐1; I = 550 W m−2;

CWO3 = 0.25 g L‐1.

The proposed kinetic model provides a simplified approach that can be

useful for design purposes, although the nature of the organic pollutant and its



309

reactivity with the different oxidizing species generated during photocatalytic

ozonation, together with the operating conditions, will play an important role in

order to extend the model from this particular case to more general

photocatalytic ozonation studies.

8.4. CONCLUSIONS

A synergistic effect between ozone and visible light irradiated WO3 has been

proved. The highest efficiency of photocatalytic ozonation process compared to

photocatalysis and single ozonation is reflected in both, DEET depletion and

mineralization rates, hydroxyl radicals in the bulk being the main species

responsible according to scavenging experiments. At the conditions used in this

work, photocatalytic ozonation process led to complete removal of 15 mg L‐1

DEET in 15 min with mineralization up to 60 % in 2 h. The detailed study of the

evolution with time of 22 TPs detected during ozonation and photocatalytic

ozonation allowed the proposal of a general reaction mechanism through

different pathways mainly based on the hydroxyl radical attack. The reaction

mechanism involves steps of mono‐ and poly‐hydroxylation and/or oxidation,

de‐alkylation and finally opening of the aromatic ring which evolves through

further oxidation, to the formation of short‐chain saturated organic acids and

mineralization to CO2. The evolution of DEET, intermediates and short‐chain

saturated organic acids was fitted to a lumped kinetic model based on TOC‐

hydroxyl radical reactions and provides a simplified approach for this process

that can be useful for design purposes. However, the nature of the organic

pollutant and its reactivity towards the different reactive species generated

during photocatalytic ozonation (ozone, hydroxyl radicals, positive holes, etc.);

as well as the operating conditions (pH, temperature, radiation intensity and

wavelength, etc.) would play an important role in order to extend the model

from this particular case and will be considered for a future work.


310

AKNOWLEDGEMENTS

Authors thank the Spanish MINECO and European Feder Funds

(CTQ2015/64944‐R) and Junta de Extremadura (Ayuda a Grupos Exp. GR15‐033)

for economic support; and SAEM‐SAIUEX for the LC‐qTOF analyses. E. Mena

thanks the Consejería de Empleo, Empresa e Innovación (Junta de Extremadura)

and European Social Fund for her FPI grant (Ref. PD12059).

REFERENCES

[1] Antonopoulou, M.; Giannakas, A.; Deligiannakis, Y.; Konstantinou, I.

“Kinetic and mechanistic investigation of photocatalytic degradation of the N,N‐

diethyl‐m‐toluamide”. Chem. Eng. J. 231 (2013) 314‐325.

[2] Costanzo, S.D.; Watkinson, A.J.; Murby, E.J.; Kolpin, D.W.; Sandstrom, M.W.

“Is there a risk associated with the insect repellent DEET (N,N‐diethyl‐m‐

toluamide) commonly found in aquatic environments?”. Sci. Total Environ. 384

(2007) 214‐220.






decomposition mechanisms of ozone with two model emerging contaminants;


[5] Petrucci, N.; Sardini. S. “Severe toxic reactions and death following the

ingestion of diethyltoluamide‐containing insect repellents”. Pediatr. Emerg.

Care 16 (2000) 341‐342.

[6] Benítez, F.J.; Acero, J.L.; Real, F.J.; Roldán, G.; Rodríguez, E. “The

effectiveness of single oxidants and AOPs in the degradation of emerging

contaminants in waters: A comparison study”. Ozone Sci. Eng. 35 (2013), 263‐

272.

[7] Li, W.; Nanaboina, V.; Zhou, Q.; Korshin, G.V. “Effects of Fenton treatment

on the properties of effluent organic matter and their relationships with the

degradation of pharmaceuticals and personal care products”. Water Res. 46

(2012) 403‐412.

[8] Medana, C.; Calza, P.; Dal Bello, F.; Raso, E.; Minero, C.; Baiocchi, C.



311

“Multiple unknown degradants generated from the insect repellent DEET by

photoinduced processes on TiO2”. J. Mass Spectrom. 46 (2011) 24‐40.




Chem. Eng. J. 204‐206 (2012) 131‐140.




[11] Mena, E.; Rey, A.; Beltrán, F.J.; Contreras, S. “Visible light photocatalytic

ozonation of DEET in the presence of different forms of WO3”. Catal. Today 252

(2015) 100‐106.

[12] Tay, K.S.; Abd. Rahman, N.; Bin Abas, M.R. “Degradation of DEET by

ozonation in aqueous solution”. Chemosphere 76 (2009) 1296‐1302.


method”. Water Res. 15 (1981) 449‐456.

[14] Ilisz, I.; Bokros, A.; Dombi, A. “TiO2‐based heterogeneous photocatalytic

water treatment combined with ozonation”. Ozone Sci. Eng. 26 (2004) 585‐594.



[16] Flyunt, R.; Leitzke, A.; Mark, G.; Mvula, E.; Reisz, E.; Schick, R.; von

Sonntag, C. “Determination of •OH, O2•‐, and hydroperoxide yields in ozone

reactions in aqueous solution”. J. Phys. Chem. B 107 (2003) 7242‐7253.



Chem. Phys. Lett. 500 (2010) 86‐89.

[18] Beltrán, F.J. “Ozone reaction kinetics for water and wastewater systems”.

Boca Raton, CRC Press, 2004, Florida (USA).

[19] Sánchez, L.; Domènech, X.; Casado, J.; Peral, J. “Solar activated ozonation of

phenol and malic acid”. Chemosphere 50 (2003) 1085‐1093.




[21] Song, W.; Cooper, W. J.; Peake, B. M.; Mezyk, S. P.; Nickelsen, M. G.;

O’Shea, K. E. “Free‐radical‐induced oxidative and reductive degradation of N,


312

N′‐diethyl‐m‐toluamide (DEET): Kinetic studies and degradation pathway”.

Water Res. 43 (3) (2009) 635−642.

[22] Yao, C.C.D.; Haag, W.R. “Rate constants for direct reactions of ozone with

several drinking water contaminants”. Water Res. 25 (1991) 761‐773.

[23] Alam, M.S.; Rao, B.S.M.; Janata, E. “OH reactions with aliphatic alcohols:

evaluation of kinetics by direct optical absorption measurement. A pulse

radiolysis study”. Radiat. Phys. Chem. 67 (2003) 723‐728.

[24] Johnson, P.N.; Davis, R.A. “Diffusivity of ozone in water”. J. Chem. Eng.

Data 41 (1996) 1485‐1487.

[25] Froment, G.F.; Bischoff, K.B. “Chemical reactors analysis and design”. John

Wiley & Sons, 1979, New York (USA).

[26] Staehelin, S.; Hoigné, J. “Decomposition of ozone in water: rate of initiation

by hydroxide ions and hydrogen peroxide”. Environ. Sci. Technol. 16 (1982) 666‐

681.

[27] Beltrán, F.J.; Aguinaco, A.; García‐Araya, J.F. “Mechanism and kinetics of


1369.

[28] Zhang, L.S.; Wong, K.H.; Zhang, D.Q.; Hu, C.; J.C. Yu, C.Y. Chan, Wong,

P.K. “Zn:In(OH)ySz solid solution nanoplates: Synthesis, characterization, and

photocatalytic mechanism”. Environ. Sci. Technol. 43 (2009) 7883‐7888.

[29] Rodríguez, E.M.; Márquez, G.; Tena, M.; Álvarez, P.M.; Beltrán, F.J.

“Determination of main species involved in the first steps of TiO2 photocatalytic

degradation of organics with the use of scavengers: The case of ofloxacin”. Appl.

Catal. B Environ. 178 (2015) 44‐53.

[30] Hoigné, J.; Bader, H. “Rate constants of reactions of ozone with organic and

inorganic compounds in water—II: Dissociating organic compounds”. Water

Res. 17 (1983) 185‐194.

[31] Buxton, G.V.; Greenstock, C.L.; Helman, W.P. “Critical review of rate

constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl

radicals (.OH/.O−) in aqueous solution”. J. Phys. Chem. Ref. Data 17 (1988) 513‐

886.

[32] Coelho, A. D.; Sans, C.; Agüera, A.; Gomez, M.J.; Esplugas, S.; Dezotti, M.

“Effects of ozone pre‐treatment on diclofenac: Intermediates, biodegradability

and toxicity assessment”. Sci. Total Environ. 407 (2009) 3572−3578.

[33] Zhang, H.; Lemley, A. T. “Reaction mechanism and kinetic modeling of

DEET degradation by flow‐through anodic Fenton treatment (FAFT)”. Environ.



313

Sci. Technol. 40 (2006) 4488−4494.

[34] Calza, P.; Medana, C.; Raso, E.; Giancotti, V.; Minero, C. “N,N‐diethyl‐m‐

toluamide transformation in river water”. Sci. Total Environ. 409 (2011)

3894−3901.


solution: Acrylic, methacrylic, maleic, fumaric and muconic acids”. Ozone Sci.

Eng. 31 (2009) 301‐308.

[36] Aguinaco, A.; Beltrán, F.J.; Sagasti, J.J.P.; Gimeno, O. “In situ generation of

hydrogen peroxide from pharmaceuticals single ozonation: A comparative

study of its application on Fenton like systems”. Chem. Eng. J. 235 (2014) 46‐ 51.

[37] Beltrán, F.J.; Gimeno, O.; Rivas, F.J.; Carbajo, M. “Photocatalytic ozonation

of gallic acid in water”. J. Chem. Technol. Biotechnol. 81 (2006) 1787‐1796.




[39] Elovitz, M.S.; von Gunten, U. “Hydroxyl radical/ozone ratios during

ozonation processes. I. The Rct concept”. Ozone Sci. Eng. 21 (1999) 239‐260.

[40] Leitner, N.K.V.; Doré, M. “Mecanisme dʹaction des radicaux OH sur les

acides glycolique, glyoxylique, acetique et oxalique en solution aqueuse:

Incidence sur la consammation de peroxyde dʹhydrogene dans les

systemes H2O2‐UV et O3‐H2O2”. Water Res. 31 (1997) 1383‐1397.

CAPÍTULO 9 Conclusiones

En este capítulo se presentan las conclusiones más relevantes que se han ido extrayendo a partir de los resultados obtenidos en el desarrollo del trabajo de investigación que ha dado lugar a la presente Tesis Doctoral.

Conclusiones

317

La realización de esta Tesis Doctoral ha tenido como objetivo principal

contribuir al desarrollo de nuevas estrategias para el tratamiento de

contaminantes emergentes en aguas, centrándose en el proceso de ozonización

fotocatalítica como alternativa eficiente para su eliminación aprovechando la luz

solar como fuente de radiación y sintetizando fotocatalizadores que permitan un

mayor aprovechamiento de la misma. Para lograr este objetivo general se

plantearon una serie de objetivos específicos, tal como se detalla en el Capítulo 1

de este trabajo. A continuación se presentan las conclusiones generales

obtenidas en función del objetivo perseguido.

I) En primer lugar, del estudio realizado con el fin de demostrar y

justificar la mayor eficacia del tratamiento de ozonización fotocatalítica frente a

los procesos individuales (fotocatálisis heterogénea y ozonización simple), se

concluye que:

‐ Los resultados obtenidos sobre la producción de especies oxidantes en los

distintos procesos ha permitido establecer la importancia de las reacciones

directas e indirectas del ozono y la sinergia entre sistemas. A pH ácido,

cuando las reacciones indirectas de ozono son despreciables, el ozono ejerce

un efecto positivo en la velocidad de formación de especies oxidantes foto‐

generadas, aumentando el rendimiento cuántico de 0,34 mol einstein‐1 en

fotocatálisis a 0,80 mol einstein‐1 en ozonización fotocatalítica. A pH = 7 el

incremento fue mucho mayor (de 0,29 a 3,27 mol einstein‐1), si bien en este

caso no puede descartarse la contribución del ozono por vía indirecta. El

aumento del rendimiento cuántico en presencia de ozono, responsable de la

sinergia observada en el proceso combinado, se debe fundamentalmente al

desempeño del ozono como aceptor de electrones del semiconductor

irradiado, minimizando así el proceso de recombinación.

II) En segundo lugar, del estudio realizado sobre la síntesis de nuevos

fotocatalizadores con vistas a incrementar el aprovechamiento de la radiación

solar en la degradación mediante ozonización fotocatalítica de contaminantes

emergentes en agua pura y agua residual, puede concluirse que:

CAPÍTULO 9

318

‐ El óxido de wolframio WO3 ha demostrado ser un material activo y estable

en la degradación de contaminantes en agua en presencia de ozono

empleando tanto luz visible como radiación solar. A diferencia del oxígeno,

el ozono es capaz de reaccionar con los electrones fotogenerados en la

superficie irradiada del semiconductor. Las formas monoclínica y

ortorrómbica del WO3 son las más activas, viéndose la actividad favorecida

por la presencia de estados reducidos o vacantes de oxígeno en la estructura

del WO3, lo que favorece el transporte de electrones en la superficie del

catalizador. Los resultados indican que son los parámetros estructurales y

superficiales y no el desarrollo de superficie específica los que juegan un

papel más importante en la actividad del material. Bajo luz visible, con el

mejor material de esta serie se obtuvo, en función de las condiciones de

ensayo, un consumo específico de ozono de 15 ‐ 49 mg O3/mg COT eliminado

para conseguir una mineralización del 50 % mediante ozonización

fotocatalítica, empleando DEET como compuesto de prueba.

‐ Los resultados sobre síntesis y aplicación de óxidos de cerio CeO2 con

distinta morfología (nanovarillas y nanocubos), indican que las nanovarillas

de CeO2, por su mayor área superficial irradiada y la mayor cantidad de

defectos superficiales y vacantes de oxígeno (lo que condujo a una menor

energía de salto de banda), presentan una mayor actividad con respecto a los

nanocubos en la ozonización fotocatalítica bajo luz visible. Por el contrario,

bajo radiación solar la presencia en los nanocubos de caras {100} expuestas les

confiere una mayor actividad fotocatalítica intrínseca. Con el mejor material

de entre los preparados en esta serie se obtuvo, en las condiciones de ensayo

y empleando tanto luz visible como radiación solar, un consumo específico

de ozono de 70 mg O3/mg COT eliminado para conseguir una mineralización

del 50 % mediante ozonización fotocatalítica empleando DEET como

compuesto de prueba.

‐ De la síntesis y aplicación de catalizadores compuestos de WO3‐TiO2 con un

4 % en peso de WO3 sobre dos soportes de TiO2 con distintas propiedades

Conclusiones

319

estructurales y texturales (TiO2 comercial P25 y nanotubos de TiO2, NT), se

deduce que bajo luz solar ambos materiales resultaron más eficientes que sus

precursores en la degradación de contaminantes emergentes en agua residual

mediante ozonización fotocatalítica, siendo el material NT‐WO3 el que

presentó un mejor comportamiento. Este catalizador estaba compuesto de

partículas alargadas de anatasa, una estructura porosa muy desarrollada y

una alta dispersión de especies WOx. Estas propiedades provocan un mejor

aprovechamiento de la radiación solar visible y, además, una mayor

capacidad de adsorción de compuestos orgánicos en comparación con el TiO2

P25. Con este material, en las condiciones de ensayo y bajo radiación solar se

obtuvo un consumo específico de ozono de 19 mg O3/mg COT eliminado

para conseguir una mineralización del 50 % de un efluente de EDAR dopado

con contaminantes emergentes mediante ozonización fotocatalítica.

‐ Con independencia de las condiciones de ensayo, de la naturaleza de la

matriz acuosa y del tipo de radiación (luz visible o radiación solar), todos los

catalizadores sintetizados favorecieron la sinergia entre el ozono y el

semiconductor. Atendiendo al comportamiento de los distintos materiales

bajo luz visible, para el siguiente estudio se seleccionó por su mejor consumo

específico de ozono, un catalizador de WO3 con estructura monoclínica.

III) Para finalizar, al objeto de profundizar en el comportamiento de los

catalizadores de WO3 en la degradación de contaminantes en agua mediante

ozonización fotocatalítica bajo luz visible y, con ello, poder proponer para este

sistema un modelo cinético, se identificaron las principales especies oxidantes

involucradas y los intermedios de reacción generados en la degradación del

DEET, compuesto modelo elegido debido a su baja reactividad frente al ozono

molecular. Los resultados obtenidos indican que:

‐ Bajo luz visible los radicales hidroxilo en el seno del líquido son las

principales especies involucradas en la degradación y mineralización de

DEET mediante ozonización fotocatalítica. De forma general, en el caso de

contaminantes que presenten una alta reactividad frente al ozono molecular

CAPÍTULO 9

320

este jugará un papel en la eliminación de los mismos, mientras que los

radicales hidroxilo serán fundamentales en la mineralización de los

intermedios más refractarios al ozono.

‐ Los intermedios de degradación de DEET mediante ozonización simple y

fotocatalítica fueron prácticamente los mismos, lo que ha permitido

establecer un mecanismo de reacción basado en el ataque del radical

hidroxilo y que transcurre a través de la mono‐ y poli‐hidroxilación y

oxidación, de‐alquilación y, finalmente, apertura del anillo aromático,

conduciendo a la formación de ácidos orgánicos de cadena corta y la

mineralización a CO2.

‐ En base al mecanismo propuesto y al papel mayoritario del radical hidroxilo

en el mismo, se desarrolló un modelo cinético en términos de COT que

permite simular la concentración de DEET, la de sus principales intermedios

de degradación y la de los ácidos orgánicos de cadena corta, proporcionando

un enfoque simplificado del proceso de ozonización fotocatalítica. No

obstante, la naturaleza del contaminante orgánico, su reactividad frente a las

distintas especies reactivas en el proceso (ozono, radical hidroxilo, huecos

positivos, etc.), así como las condiciones de operación (pH, temperatura, tipo

de radiación empleada, intensidad, etc.), jugarán un papel muy importante

para poder extender el modelo más allá de este caso en particular.

NOMENCLATURA

A Constante en medidas de espectroscopía ultravioleta‐visible de

reflectancia difusa

Absm Absorbancia de una muestra

Abs0 Absorbancia del blanco

act Actinómetro

BC Banda de conducción de un semiconductor

BV Banda de valencia de un semiconductor

c Camino óptico en la Ley de Beer

C Contaminante orgánico genérico

Ci Concentración de un compuesto i en disolución acuosa

Ci,0 Concentración inicial de un compuesto i en disolución acuosa

Ci,g Concentración de un compuesto i en fase gas

CI Carbono inorgánico

COT Carbono orgánico total

CT Carbono total

d Distancia interplanar en medidas de difracción de rayos X

DBO Demanda biológica de oxígeno

Nomenclatura

322

DBO5 Demanda biológica de oxígeno tras 5 días de incubación

DEET N,N‐dietil‐meta‐toluamida

DMA Directiva Marco Europea del Agua

DQO Demanda química de oxígeno

e Electrón

BCe Electrón generado en la banda de conducción de un

semiconductor

se Electrón móvil que migra a la superficie del semiconductor

Te Electrón atrapado formando un estado de menor movilidad

EDAR Estación depuradora de aguas residuales

Eg Energía de salto de banda de un semiconductor

h Constante de Planck

h Hueco positivo

BVh Hueco positivo generado en la banda de valencia de un

semiconductor

sh Hueco positivo móvil que migra a la superficie del semiconductor

Th Hueco positivo atrapado formando un estado de menor

movilidad

He Constante de Henry

HO Ion hidróxido

•HO Radical hidroxilo

sHO Radical hidroxilo superficial

2HO Ion hidroperóxido

2HO Radical hidroperóxido

h Radiación

IBP Ibuprofeno

Ii Intensidad de pico en medidas de espectroscopía fotoelectrónica

de rayos X

Nomenclatura

323

I0 Intensidad de radiación incidente

K Constante relativa al factor de forma de un cristal en medidas de

difracción de rayos X en la ecuación de Scherrer

kHO‐C Constante de reacción entre los radicales hidroxilo y un

contaminante orgánico C

kL Coeficiente individual de transferencia de materia

kLa Coeficiente volumétrico de transferencia de materia

kO3‐C Constante de reacción directa entre el ozono y un contaminante

orgánico C

L Paso efectivo de radiación a través del reactor en medidas

actinométricas

o tamaño de cristal en medidas de difracción de rayos X

LD Límite de detección

n Exponente que toma el valor de ½ para transiciones electrónicas

directas en medidas de espectroscopía ultravioleta‐visible de

reflectancia difusa

u orden de difracción en medidas de difracción de rayos X

NCA Normas de calidad ambiental

ni Número de átomos por cm3 del elemento i en medidas de

espectroscopía fotoelectrónica de rayos X

NT Nanotubos

2sO Ion oxígeno terminal de la red de un semiconductor

2O Radical superóxido

•

3O Radical ozónido

O3dis Ozono disuelto

OMS Organización Mundial de la Salud

PAO Proceso avanzado de oxidación

pHPZC pH del potencial de carga cero

PNCA Plan Nacional de Calidad de las Aguas

PNRA Plan Nacional de Reutilización de Aguas

PT Contenido en fósforo total

P25 Catalizador de TiO2 comercial P25 (Aeroxide®)

Nomenclatura

324

Qg Caudal de gas

R Valor de reflectancia medida respecto a la unidad en medidas de

espectroscopía ultravioleta‐visible de reflectancia difusa

Si Factor de sensibilidad atómica en medidas de espectroscopía

fotoelectrónica de rayos X

T Temperatura

tR Tiempo de retención

u.a. Unidades arbitrarias

US‐EPA Agencia de los Estados Unidos para la protección del

Medioambiente

UV Radiación ultravioleta

UVA Radiación ultravioleta A

UVB Radiación ultravioleta B

UVV Radiación ultravioleta de vacío

V Volumen de reacción

VINY Volumen de inyección

Vm Volumen de muestra

VT Volumen total

Coeficiente de absorción en medidas de espectroscopía

ultravioleta‐visible de reflectancia difusa

Anchura a la mitad de la intensidad máxima de un pico

seleccionado en medidas de difracción de rayos X

H Variación de entalpía

G Variación de entropía

Coeficiente de extinción molar en la Ley de Beer

Rendimiento cuántico de una reacción fotoquímica

hv Rendimiento cuántico de foto‐generación de especies oxidantes

Longitud de onda de la radiación

Frecuencia de la radiación en medidas de espectroscopía

ultravioleta‐visible de reflectancia difusa

2 Ángulo de difracción en medidas de difracción de rayos X

NOMENCLATURE

A Pre‐exponential constant in the Arrhenius equation

AAP Acetaminophen

ANT Antipyrine

AOP Advanced oxidation process

a.u. Arbitrary units

BOD Biological oxygen demand

BOD5 Biological oxygen demand during 5 days of incubation

CAF Caffeine

CAR Carbamazepine

CAT Catalyst

CB Conduction band in a semiconductor

Ci Concentration of a compound i in the liquid phase

Ci,0 Initial concentration of a compound i in the liquid phase

Ci,g Concentration of a compound i in the gas phase

Ci,g inlet Inlet concentration of a compound i in the gas phase

COD Chemical oxygen demand

Nomenclature

326

DCF Diclofenac

DEET N,N‐diethyl‐meta‐toluamide

DO3 Ozone diffusivity in water

DR‐UV‐Vis Diffuse reflectance UV‐Vis spectroscopy

E Degree of reaction rate enhancement

e Electron

aqe Electron in the aqueous phase

Ea Activation energy

EC Emerging contaminant

Eg Optical energy band gap of a semiconductor

FS Integrated absorption fraction of radiation

h Positive hole

Ha Hatta number

HCT Hydrochlorothiazide

He Henry’s constant

•HO Hydroxyl radical

sHO Surface bounded hydroxyl radical

2HO Hydroperoxide radical

HPLC‐DAD High‐performance liquid chromatography – Diode Array detector

HPLC‐qTOF High‐performance liquid chromatography – quadrupole time of

flight

h Radiation

I Photon flux

Ia Photon flux absorbed by the catalyst

IBP Ibuprofen

IC Inorganic carbon

ICP‐OES Inductively coupled plasma – optical emission spectrometry

Ii Intensity of a Raman peak at i cm‐1.

INT Intermediate organic compound

Nomenclature

327

I0 Incident photon flux

k Kinetic constant

KET Ketorolac

kHO‐i Rate constant for the reaction between i and hydroxyl radical

kL Individual liquid phase mass transfer coefficient

kLa Volumetric mass transfer coefficient

kO3‐i Rate constant for the reaction between i and ozone

kS Scavenging factor

kTOC Pseudo‐first order apparent rate constant of TOC removal

L Crystal size

LC50 Lethal concentration of a substance expected to kill 50 % of

organisms in a given population under a defined set of conditions

M Mass

max Maximum

MTP Metoprolol

MWW Municipal wastewater

MWWTP Municipal wastewater treatment plant

n Reaction order

NC Nanocubes

NT Nanotubes

NR Nanorods

2

2

TiOO Terminal oxygen ion of the TiO2 lattice

2O Superoxide radical

•

3O Ozonide radical

O3d Dissolved ozone

O3g Ozone in the gas phase at the reactor outlet

O3gi Ozone in the gas phase at the reactor inlet

Oxal Oxalic acid

pCBA p‐chlorobenzoic acid

Nomenclature

328

PhC‐O2 Photocatalytic oxidation

PhC‐O3 Photocatalytic ozonation

pHPZC pH at the point of zero charge

PPCP Pharmaceutical and Personal Care Product

P25 Commercial TiO2 P25 (Aeroxide®)

Qg Gas flow rate

R Universal gas constant

rHO‐O3 Rate of ozone indirect reactions

rhv Rate of photocatalytic reactions

ri Rate of reaction or formation of i

rO3 Rate of ozone direct reactions

RTE Radiation transfer equation

SBET BET surface area

SEM Scanning electron microscopy

SFM Sulfamethoxazole

T Temperature

t Time

TEM Transmission electron microscopy

TGA‐DTA Thermal gravimetric and differential temperature analysis

TP Transformation product

TOC Total organic carbon

TOC,a TOC corresponding to short‐chain organic acids

u.a. Units of absorbance

UV Ultraviolet radiation

UVA Ultraviolet A radiation

UVC Ultraviolet C radiation

V Reaction volume

VB Valence band in a semiconductor

Vis Visible

Vp Pore volume

Nomenclature

329

v/v Volume fraction

wt Weight

wt/wt Weight fraction

XPS X‐ray photoelectron spectroscopy

XRD X‐ray diffraction

z Stoichiometric coefficient of a reaction

Stoichiometric parameter that represents the fraction of TOC

converted

Liquid holdup

Increment

Molar extinction coefficient

i Quantum yield of the photocatalytic reaction of i

hv Quantum yield of photo‐generated oxidizing species

Photonic efficiency

lim Limiting photonic efficiency

Radiation wavelength

g Gas flow rate

2 Diffraction angle in XRD analysis

SISTEMAS CATALÍTICOS BASADOS EN DISTINTOS

Documents

Transcript of SISTEMAS CATALÍTICOS BASADOS EN DISTINTOS