Funcionalización química de materiales carbonosos para almacenamiento … · 2019-06-04 · La...

Funcionalización química de materiales

carbonosos para almacenamiento y generación

de energía

María José Mostazo López

https://www.ua.es/

http://www.eltallerdigital.com/es/index.html

Departamento de Química Inorgánica

Departamento de Química Física

Instituto Universitario de Materiales

Facultad de Ciencias

FUNCIONALIZACIÓN QUÍMICA DE MATERIALES CARBONOSOS

PARA ALMACENAMIENTO Y GENERACIÓN DE ENERGÍA

María José Mostazo López

Tesis presentada para aspirar al grado de

DOCTORA POR LA UNIVERSIDAD DE ALICANTE

MENCIÓN DE DOCTORA INTERNACIONAL

Doctorado en Ciencia de Materiales

Dirigida por:

Diego Cazorla Amorós Emilia Morallón Núñez

Catedrático de Química Inorgánica Catedrática de Química Física

Alicante, marzo de 2019

Este trabajo ha sido financiado por la Generalitat Valenciana a través de un contrato de

formación de personal investigador de carácter predoctoral (Programa VALi+d,

ACIF/2015/374).

Diego Cazorla Amorós Emilia Morallón Núñez

Catedrático de

Química Inorgánica

Catedrática de

Química Física

Diego Cazorla Amorós, Catedrático de Química Inorgánica, y Emilia

Morallón Núñez, Catedrática de Química Física, ambos de la

Universidad de Alicante,

CERTIFICAN QUE:

Dña. María José Mostazo López, Licenciada en Química, ha realizado

en el Departamento de Química Inorgánica y en el Instituto

Universitario de Materiales de la Universidad de Alicante, bajo nuestra

dirección, el trabajo que lleva por título: Funcionalización química de

materiales carbonosos para almacenamiento y generación de

energía, que constituye su memoria para aspirar al grado de Doctora,

reuniendo, a nuestro juicio, las condiciones necesarias para ser

presentada y defendida ante el tribunal correspondiente.

Y para que conste a efectos oportunos, en cumplimiento de la

legislación vigente, firmamos el presente certificado en Alicante, a 26

de febrero del 2019.

Capítulo 0

Agradecimientos

En primer lugar, quisiera agradecer a mis directores de Tesis, los

profesores Diego Cazorla Amorós y Emilia Morallón Núñez, por la

oportunidad de realizar este trabajo en sus grupos de investigación, por su

paciencia y su dedicación.

A Ramiro Ruiz Rosas, por todas sus enseñanzas, por estar siempre

dispuesto a ayudar y por los consejos que me ha dado durante estos años.

A la Generalitat Valenciana, por la concesión de un contrato de

formación predoctoral para financiar esta Tesis (ACIF/2015/374), así

como las becas para la realización de estancias en otros centros

(BEFPI/2016/006 y BEFPI/2017/036).

I wish to thank Shiraishi Sensei for the warm welcome I received in

his laboratory at Gunma University. To all my labmates in Shiraishi’s lab,

for their kind help and support. I also want to thank all the friends I met

in Kiryu for the nice moments we shared.

I would like to thank Prof. Andrea Balducci for his kind attention

during my stay in Jena. To my labmates in Jena, for their nice welcome

and kindness.

A mis compañeros del GEPE y del MCMA, a los que estuvieron y a

los que han llegado a la largo de estos años, por los almuerzos, los cafés,

las discusiones científicas y todos los buenos momentos que hemos

pasado juntos. Gracias a todos, porque ha sido un placer compartir con

vosotros esta etapa.

A mis amigos, tanto a los de dentro como a los de fuera de la

universidad, por ayudarme a desconectar y por acompañarme durante

estos años.

A mi familia, y en especial, a mis padres; por creer en mí y por

apoyarme siempre. Todo lo que he conseguido os lo debo a vosotros.

A Juan, por acompañarme desde el principio de esta andadura, por

ayudarme a sacar siempre lo mejor de mí misma, por los kilómetros

recorridos para visitarme y por su apoyo incondicional.

A mis padres

A Juan

Índice

ÍNDICE

Objetivos y estructura general de la Tesis Doctoral

1. Introducción …………………………………………………….. 3

2. Objetivos……………………………………………………........ 3

3. Estructura de la Tesis Doctoral…………………………….......... 7

Capítulo 1. Introducción general.

1.1. Materiales carbonosos ……………………………………..... 15

1.1.1. Carbón activado …………………………………….. 16

1.1.2. Materiales carbonosos nanomoldeados……………... 19

1.2. Química superficial de los materiales carbonosos ………….. 21

1.2.1. Oxígeno ……………………………………………… 21

1.2.2. Nitrógeno …………………………………………….. 22

1.3. Síntesis de materiales carbonosos dopados con nitrógeno …. 24

1.3.1. Síntesis directa a partir de un precursor nitrogenado ... 25

1.3.2 Post-tratamiento de materiales carbonosos con

reactivos nitrogenados ……………………………………… 26

1.3.2.1 Funcionalización con nitrógeno mediante

tratamientos térmicos ………………………………….. 26

1.3.2.2 Funcionalización química basada en reacciones

orgánicas a baja temperatura …………………………... 29

1.3.2.3. Funcionalización electroquímica ……………... 34

1.4. Aplicaciones en almacenamiento y generación de energía … 35

1.4.1. Condensadores electroquímicos ……………………... 35

4

1.4.1.1. Modelos de doble capa eléctrica ………………. 37

1.4.1.2. Características de los condensadores

electroquímicos ………………………………………… 41

1.4.1.3 Configuraciones de condensador electroquímico 45

1.4.1.4. Tipos de electrolitos …………………………… 48

1.4.1.5. Materiales carbonosos como electrodos de

supercondensadores ……………………………………. 52

1.4.2. Pilas de combustible …………………………………. 58

1.4.2.1. Reacción de reducción de oxígeno ……………. 60

1.4.2.2. Electrocatalizadores para la ORR ……………... 62

1.4.2.3. Materiales carbonosos funcionalizados con

nitrógeno como electrocatalizadores de la ORR ………. 63

1.5. Referencias …………………………………………………. 65

Capítulo 2. Materiales, métodos y técnicas experimentales.

2.1. Introducción ………………………………………………... 85

2.2. Materiales ………………………………………………….. 85

2.2.1 Carbones activados …………………………………... 85

2.2.2. Materiales carbonosos nanomoldeados ……………... 86

2.3. Modificación de la química superficial de los materiales

carbonosos ……………………………………………………… 86

2.3.1. Funcionalización química mediante métodos basados

en reacciones orgánicas…………………………………….. 87

2.3.1.1 Oxidación química ……………………………. 88

2.3.1.2 Introducción de grupos funcionales amida…….. 88

Índice

2.3.1.3 Introducción de grupos funcionales amina…….. 89

2.3.2. Polimerización de anilina……………………………. 90

2.3.3. Tratamientos térmicos……………………………….. 90

2.4. Técnicas de caracterización………………………………… 91

2.4.1. Adsorción física de gases……………………………. 91

2.4.1.1 Teoría de Brunauer-Emmet-Teller…………….. 93

2.4.1.2. Ecuación de Dubinin-Radushkevich………….. 95

2.4.1.3 Teoría del funcional de densidad no localizada

bidimensional………………………………………….. 96

2.4.2. Espectroscopia fotoelectrónica de rayos X………….. 98

2.4.3. Desorción a temperatura programada……………….. 100

2.4.4. Caracterización electroquímica……………………… 102

2.4.4.1. Configuraciones de celda electroquímica…….. 103

2.4.4.2. Técnicas electroquímicas……………………... 105

2.5. Referencias…………………………………………………. 122

Chapter 3. N-functionalization of activated carbon.

3.1. Introduction………………………………………………… 129

3.2. Materials and methods……………………………………… 132

3.2.1 Activated carbon……………………………………… 132

3.2.1.1. Chemical oxidation with HNO3………………. 132

3.2.1.2. Generation of amide functionalities…………... 133

3.2.1.3. Generation of amine functionalities…………… 134

3.2.2. Electrochemical characterization……………………. 134

6

3.2.3. Porous texture and surface chemistry characterization 135

3.3. Results and discusión………………………………………. 136

3.3.1. Porous texture characterization……………………… 139

3.3.2. Surface chemistry characterization………………….. 141

3.3.2.1. XPS…………………………………………… 139

3.3.2.2. Temperature Programmed Desorption………... 147


3.3.3.1. Cyclic Voltammetry…………………………... 154

3.3.3.2. Galvanostatic charge-discharge cycles………... 158

3.4. Conclusions………………………………………………… 163

164 3.5.

References…………………………………………………..

Annex to Chapter 3. Nitrogen functionalization of zeolite

templated carbon by electrochemical and chemical methods…... 173

Chapter 4. Electrochemical performance of N-doped activated

carbons in aqueous electrolyte

4.1. Introduction………………………………………………… 183


4.2.1 Activated carbon……………………………………… 185

4.2.2. Chemical functionalization of activaded carbon…….. 186

4.2.2.1. Synthesis of KUA-CONH2……………………. 186

4.2.2.2. Synthesis of KUA-N………………………….. 186

4.2.3. Porous texture and surface chemistry characterization 187


4.2.4.1. Three electrode cell configuration…………….. 188

Índice

4.2.4.2. Two electrode cell configuration……………… 189


4.2.1 Surface chemistry and porous texture characterization. 191

4.3.2 Electrochemical characterization…………………….. 195

4.3.2.1 Characterization of carbon materials…………... 195

4.3.2.2 Characterization of symmetric supercapacitors. 199

4.3.2.3 Characterization of asymmetric supercapacitors 204

4.4. Conclusions………………………………………………… 210

4.5. References………………………………………………….. 212

Chapter 5. Electrochemical performance of N-doped

zeolite templated carbon in aqueous electrolyte

5.1. Introduction………………………………………………… 217


5.2.1 Zeolite templated carbons……………………………. 219

5.2.2. Physicochemical characterization…………………… 220

5.2.3. Electrochemical characterization…………………… 221

5.2.3.1 Three electrode cell configuration……………... 221

5.2.3.2 Two electrode cell configuration………………. 222


5.3.1. Physicochemical characterization…………………… 223


5.3.2.1. Electrochemical characterization before electro-

oxidation……………………………………………….. 225

8

5.3.2.2. Electrochemical characterization after electro-

oxidation………………………………………………... 228

5.3.2.3. ZTC and N-ZTC supercapacitors using acid

electrolyte………………………………………………. 235

5.4. Conclusions………………………………………………… 241

5.5. References………………………………………………….. 243

5.6. Annex to Chapter 5..………………………………………... 250


carbons in organic electrolyte

6.1. Introduction………………………………………………… 255


6.2.1 Synthesis of carbon materials………………………… 261

6.2.1.1. Pristine carbon material……………………...... 257

6.2.1.2 N-functionalization of carbon materials at mild

conditions………………………………………………. 257

6.2.1.3 Heat treatments………………………………… 258

6.2.2. Physicochemical characterization……………………. 258

6.2.3. Electrochemical characterization…………………….. 259


6.3.1. Physicochemical characterization of carbon materials 260

6.3.1.1 Porous texture………………………………….. 260

6.3.1.2 Surface chemistry characterization…………….. 263

6.3.2. Electrochemical characterization…………………….. 271

Índice

6.3.2.1. Effect of surface chemistry modification at

mild conditions………………………………………… 275

6.3.2.2. Effect of surface chemistry modification by

heat treatments…………………………………………. 278

6.4. Conclusions………………………………………………… 281

282 6.5.

References…………………………………………………..

Annex to Chapter 6. Chemical functionalization of a

commercial activated carbon with nitrogen groups……………...

293


carbons in non-conventional electrolytes

7.1. Introduction…………………………………………………... 297

7.2. Materials and methods……………………………………….. 300

7.2.1 Activated carbon……………………………………….. 300

7.2.2. Chemical functionalization of activated carbon………. 300

7.2.3. Porous texture and surface chemistry characterization.. 301

7.2.4. Electrolyte preparation………………………………… 305

7.2.5 Electrochemical characterization………………………. 302

7.2.5.1 Three electrode cell configuration………………. 302

7.2.5.2 EDLC investigation……………………………… 303

7.3. Results and discussion……………………………………….. 304

7.3.1. Surface chemistry and porous texture characterization.. 304

7.3.2. Electrochemical characterization……………………… 307

7.3.2.1 Three-electrode cell configuration………………. 307

10

7.3.2.2. EDLC investigation……………………………... 313

7.4. Conclusions ………………………………………………….. 319

7.5. References……………………………………………………. 323


carbons as electrocatalysts for the ORR

8.1. Introduction………………………………………………… 327


8.2.1 Synthesis of activated carbons……………………….. 330

8.2.1.1 Pristine activated carbon………………………. 330

8.2.1.2 Chemical functionalization of activated carbon

at mild conditions……………………………………… 330

8.2.1.3 Preparation of polyaniline/activated carbon

composite………………………………………………

331

8.2.1.4 Post-thermal treatments……………………….. 331

8.2.2 Porous texture and surface chemistry characterization 331

8.2.3. Electrochemical activity towards ORR……………… 332

8.3. Results and Discussion……………………………………... 333

8.3.1 Surface chemistry characterization…………………... 333

8.3.2. Porous texture characterization……………………… 340

8.3.3 Electroactivity towards ORR…………………………. 342

8.3.3.1 Pristine activated carbon………………………. 343

8.3.3.2 N-doped activated carbons at mild conditions… 343

8.3.3.3 Polyaniline/activated carbon composite……….. 344

Índice

8.3.2.4 Heat-treated activated carbons………………… 346

8.3.2.5 Selectivity to water formation…………………. 347

8.4. Conclusions ………………………………………………... 349

8.5. References …………………………………………………. 351

Annex to Chapter 8……………………………………………… 357

Chapter 9. General conclusions………………………………. 361

Summary/Resumen…………………………………………….. 377

Objetivos y estructura general

de la Tesis Doctoral


3

1. Introducción

La investigación en el área de almacenamiento y producción de

energía eléctrica es un tema de interés creciente debido a la demanda

energética de la sociedad actual. Uno de los retos principales de la

comunidad científica reside en el desarrollo de sistemas más eficientes y

sostenibles medioambientalmente. En este contexto, el empleo de

sistemas electroquímicos, como los supercondensadores y las pilas de

combustible, constituye una alternativa muy prometedora para avanzar

hacia una producción energética sostenible.

En este ámbito, los materiales carbonosos son ampliamente utilizados

debido a sus propiedades únicas, que les confiere una gran versatilidad

para ser utilizados como componentes de diversos dispositivos. No

obstante, es necesario continuar investigando para diseñar materiales que

permitan mejorar las prestaciones de estos sistemas de almacenamiento y

generación de energía. Por estos motivos, la presente Tesis Doctoral se

enmarca en el estudio y desarrollo de materiales carbonosos para su

aplicación en dispositivos electroquímicos de almacenamiento y

producción de energía.

2. Objetivos

La presente Tesis Doctoral tiene como objetivo general el desarrollo

de materiales carbonosos por medio de nuevas metodologías de

funcionalización con nitrógeno para su aplicación como electrodos en

condensadores electroquímicos y electrocatalizadores en pilas de

4

combustible. A continuación, se describen los objetivos específicos de

este trabajo:

(i) Funcionalización química de materiales carbonosos de

porosidad elevada con grupos funcionales nitrogenados por

medio de estrategias basadas en reacciones orgánicas en

condiciones suaves.

(ii) Caracterización fisicoquímica y electroquímica de los

materiales carbonosos funcionalizados con nitrógeno, así

como materiales carbonosos obtenidos por nanomoldeo.

(iii) Evaluación del comportamiento electroquímico de los

materiales como electrodos de supercondensadores en

electrolitos acuosos, orgánicos y líquidos iónicos.

(iv) Estudio del efecto de grupos funcionales nitrogenados en el

comportamiento electroquímico de carbones activados de

elevada porosidad como electrocatalizadores de la reacción de

reducción de oxígeno.

3. Estructura de la Tesis Doctoral

La presente Tesis Doctoral ha sido realizada en el grupo Materiales

Carbonosos y Medio Ambiente (MCMA), perteneciente al Departamento

de Química Inorgánica, y en el Grupo de Electrocatálisis y Electroquímica

de Polímeros (GEPE) del Departamento de Química Física. Ambos

grupos de investigación pertenecen a su vez al Instituto Universitario de

Materiales de la Universidad de Alicante. El desarrollo del trabajo ha sido

complementado por medio de dos estancias de investigación en centros


5

extranjeros. La primera estancia se realizó en el Graduate School of

Science and Technology, de la Universidad de Gunma (Japón), bajo la

supervisión del Profesor Soshi Shiraishi. La segunda estancia se llevó a

cabo bajo la supervisión del Profesor Andrea Balducci, en el Center for

Energy and Environmental Chemistry, perteneciente a la Universidad

Friedrich Schiller de Jena (Alemania).

Dado que la memoria se presenta para aspirar al grado de Doctora por

la Universidad de Alicante con mención de Doctora Internacional, los

capítulos correspondientes a los resultados obtenidos y las conclusiones

generales se han redactado en inglés para cumplir con la normativa

vigente.

La Tesis Doctoral se encuentra dividida en nueve capítulos. A

continuación, se describe el contenido de cada uno de ellos.

Capítulo 1. Introducción general

En este capítulo, se describen las propiedades fisicoquímicas de los

materiales carbonosos estudiados en la Tesis Doctoral, así como diversas

metodologías para sintetizar materiales carbonosos de elevada porosidad

dopados con nitrógeno, con especial hincapié en los métodos basados en

post-tratamientos por medio de reacciones orgánicas. Además, se detallan

los fundamentos teóricos de las aplicaciones energéticas en las que se

utilizan estos materiales: condensadores electroquímicos y pilas de

combustible, prestando especial atención al efecto de la textura porosa y

la química superficial de los materiales en estos sistemas.

6

Capítulo 2. Materiales, métodos de síntesis y técnicas

experimentales

Este capítulo resume las metodologías de síntesis y funcionalización

química de materiales carbonosos utilizados en la presente Tesis Doctoral.

Además, se describen los fundamentos téoricos y experimentales de las

técnicas utilizadas para caracterizar los materiales y evaluar su

comportamiento electroquímico.

Chapter 3. N-functionalization of activated carbon at mild

conditions

En este capítulo se realiza la funcionalización química de un carbón

activado de microporosidad elevada con grupos funcionales nitrogenados

por medio de reacciones orgánicas a temperatura baja. El proceso de

funcionalización engloba reacciones de amidación (en tres etapas) y

aminación (por medio de un reordenamiento de Hoffman). El efecto de la

modificación química en la química superficial y la textura porosa se

describe de forma detallada por medio de distintas técnicas de

caracterización fisicoquímica. Además, se evalua el efecto de los grupos

funcionales nitrogenados en el comportamiento electroquímico de los

materiales en medio acuoso. Por último, se describe en el anexo del

capítulo la aplicación de la metodología de funcionalización química a

materiales nanomoldeados obtenidos empleando zeolitas como plantilla.

Los resultados derivados de este estudio han dado lugar a la

publicación de un artículo en la revista Carbon:


7

Mostazo-López MJ, Ruiz-Rosas R, Morallón E, Cazorla-Amorós D.

Generation of nitrogen functionalities on activated carbons by amidation

reactions and Hofmann rearrangement: Chemical and electrochemical

characterization. Carbon 2015; 91: 252-265


carbons in aqueous electrolyte.

En este capítulo, se describen y comparan dos metodologías de

funcionalización química de carbones activados: el protocolo de

amidación (descrito en el capítulo 3) y una reacción de funcionalización

directa (en una única etapa). Posteriormente, se evalua la influencia de

ambas estrategias de funcionalización en las propiedades fisicoquímicas

y electroquímicas de los materiales carbonosos. Estos materiales se

emplean como electrodos de condensadores simétricos y asimétricos (en

masa). Los dispositivos se caracterizan empleando distintas técnicas

electroquímicas para determinar su capacidad, energía y potencia. La

combinación de las técnicas de caracterización permite deducir el efecto

de la funcionalización química en la estabilidad electroquímica de los

materiales como electrodos de supercondensadores en medio acuoso.

Los resultados obtenidos en este capítulo se han publicado en la

revista International Journal of Hydrogen Energy:

Mostazo-López MJ, Ruiz-Rosas R, Morallón E, Cazorla-Amorós D.

Nitrogen doped activated carbon prepared by a mild method.

Enhancement of the supercapacitor performance. International Journal of

Hydrogen Energy 2016; 41: 19691-19701

8

Chapter 5. Electrochemical performance of N-ZTC in aqueous

electrolyte

En este capítulo, se emplean dos materiales carbonosos

nanomoldeados para evaluar el papel de los grupos funcionales

nitrogenados en su comportamiento como electrodos de

supercondensadores en medio acuoso. Para esto, se utilizan dos materiales

carbonosos nanomoldeados: uno dopado con nitrógeno y el otro sin dopar.

La caracterización fisicoquímica y electroquímica de los materiales

permite esclarecer qué grupos funcionales nitrogenados tienen una mayor

influencia en la aplicación de los materiales carbonosos como electrodos

de supercondensadores en medio ácido.

Los resultados de este trabajo han permitido la elaboración de un

artículo científico, que ha sido publicado en la revista Carbon:

Mostazo-López MJ, Ruiz-Rosas R, Castro-Muñiz A, Nishihara H,

Kyotani T, Morallón E, Cazorla-Amorós D. Ultraporous nitrogen-doped

zeolite-templated carbon for high power density aqueous-based

supercapacitors. Carbon 2018; 129: 510-519


carbons in organic electrolyte

En este capítulo, se describe el efecto de los grupos funcionales

nitrogenados en el comportamiento electroquímico de carbones activados

como electrodos de condensadores en medio orgánico. Para esto, se

emplean distintas estrategias para sintetizar una serie de carbones


9

activados de elevada porosidad con distintos grupos funcionales

nitrogenados. Para esto, se combinan las metodologías de

funcionalización química en condiciones suaves y post-tratamientos

térmicos para sintetizar una serie de carbones activados de elevada

porosidad con distintos grupos funcionales nitrogenados. La evaluación

de condensadores simétricos basados en estos materiales permite deducir

el efecto del nitrógeno en las características de los condensadores

(capacidad, energía, etc.) en medio orgánico. Para profundizar en el efecto

de la funcionalización en la estabilidad de los materiales carbonosos, los

condensadores se evaluan por medio de un test de durabilidad a voltaje

constante y temperatura elevada para acelerar la degradación de los

mismos. Finalmente, en el anexo del capítulo se describe el empleo de la

funcionalización química en condiciones suaves en un carbón activado

comercial para evaluar el efecto en su comportamiento como electrodo de

condensadores en medio orgánico.


carbons in non-conventional electrolytes

En este capítulo, se presenta el estudio del comportamiento

electroquímico de carbones activados obtenidos por funcionalización

química en condiciones suaves en electrolitos no convencionales: líquidos

iónicos y sales innovadoras en electrolito orgánico. Dado que estos

electrolitos permiten trabajar a voltajes de operación muy elevados, se

realizó un estudio de las ventanas de potencial de todos los materiales en

ambos electrolitos con el fin de elucidar el papel de los grupos funcionales

nitrogenados en estos electrolitos. Los materiales son utilizados para

10

preparar condensadores asimétricos en líquido iónico, para evaluar las

mejoras producidas por los electrodos funcionalizados con nitrógeno,

especialmente en lo que concierne a la durabilidad del dispositivo.

Los resultados obtenidos en este capítulo han permitido la

elaboración de un artículo cientifico, que se encuentra actualmente en fase

de revisión:

Mostazo-López MJ, Krummacher J, Balducci A, Morallón E, Cazorla-

Amorós D. Electrochemical performance of N-doped superporous

activated carbons in non-conventional electrolytes. Energy Storage

Materials 2019; en revisión.


carbons as electrocatalysts for the ORR

En este capítulo, se emplean distintas estrategias para sintetizar una

serie de carbones activados de elevada porosidad con distintos grupos

funcionales nitrogenados. Para esto, se combinan las metodologías de

funcionalización química en condiciones suaves, polimerización de

anilina y post-tratamientos térmicos. Estos materiales se emplean para

evaluar el efecto de los grupos funcionales nitrogenados en el

comportamiento electroquímico de materiales microporosos como

electrocatalizadores de la reacción de reducción de oxígeno, prestando

atención tanto a la actividad catalítica como a la selectividad a la

formación de agua.


11

Chapter 9. General conclusions.

En este capítulo, se presentan las conclusiones generales de la Tesis

Doctoral.

Capítulo 1

Introdución general

Introducción general

15

1.1. Materiales carbonosos

Los materiales carbonosos son utilizados en una amplia variedad de

aplicaciones debido a sus propiedades únicas. Los carbones activados y

fibras de carbón activadas son los materiales más empleados como

adsorbentes y catalizadores. No obstante, se han desarrollado una serie de

materiales con propiedades muy prometedoras en distintos campos, como

la industria electrónica, electroquímica, metalurgia, adsorción, catálisis,

etc. Algunos ejemplos de estos materiales son: fulerenos, nanotubos de

carbono, materiales carbonosos nanomoldeados, etc. Todos estos

materiales presentan láminas grafénicas en los que la coordinación y

disposición de los átomos de carbono definen el tipo de nanoestructura

[1]. En las aplicaciones energéticas, el empleo de distintos materiales

carbonosos es muy habitual. No obstante, para alcanzar la elevada

demanda energética y extender el uso de los distintos dispositivos a

aplicaciones nuevas, es necesario desarrollar nuevos materiales con

propiedades óptimas. En ese sentido, la química superficial tiene gran

relevancia, ya que permite adaptar las propiedades fisicoquímicas del

material a la aplicación deseada. En catálisis, la presencia de grupos

funcionales condiciona la interacción del material con adsorbatos o

reactivos, ya sea a través de la modificación de las propiedades texturales,

impidiendo el acceso de estas moléculas a la porosidad o por las

interacciones electrostáticas o químicas que puedan producirse entre la

superficie y la molécula [2]. Del mismo modo y en referencia a las

aplicaciones electroquímicas, la química superficial puede afectar a la

capacidad de almacenamiento de energía [3].

Capítulo 1

16

Por estos motivos, en la actualidad hay un gran interés en la

modificación de la química superficial y en la preparación de materiales

carbonosos dopados con heteroátomos [4,5]. Esta Tesis Doctoral se centra

en el efecto de los grupos funcionales nitrogenados en materiales

carbonosos de elevada porosidad, como son los carbones activados y los

materiales carbonosos nanomoldeados. A continuación, se presenta una

descripción detallada de las características principales de estos materiales.

1.1.1. Carbón activado

Los carbones activados son materiales utilizados en una amplia

variedad de aplicaciones, debido principalmente a su elevada superficie

específica, su estructura porosa y a la presencia de distintos grupos

funcionales en su superficie [6]. Estos materiales están constituidos por

láminas irregulares de carbono, de manera que los espacios interlaminares

conforman la porosidad de los mismos [7].

Estos materiales se utilizan en la industria como adsorbentes,

permitiendo separar compuestos en fase gas y líquida y purificar efluentes

contaminados. También tienen aplicación como catalizadores o soporte

de catalizadores [2]. En el campo de la electroquímica, se utilizan

principalmente como electrodos en dispositivos de almacenamiento y

generación de energía [8,9].

Los carbones activados se sintetizan mediante un proceso

conocido como activación, utilizando como precursor un material rico en

carbono. Los precursores utilizados son muy variados: breas, alquitranes,

carbones minerales, madera, etc. La química superficial y las propiedades


17

texturales del material dependen del precursor, del agente activante y de

la temperatura del tratamiento. La Figura 1.1 resume las etapas principales

de los dos métodos de activación: física y química.

El proceso de activación física se lleva a cabo habitualmente en dos

etapas: (i) proceso de carbonización del precursor (normalmente, entre

400 y 850 ºC); (ii) tratamiento térmico del material carbonizado a 800-

1000 ºC en atmósfera oxidante (vapor de agua, CO2 o mezcla de ambos)

[7,10]. Este proceso de activación permite obtener materiales con un

grado elevado de microporosidad. No obstante, también se forman poros

de mayor tamaño (mesoporos), de manera que la distribución de tamaños

de poros del material resultante es ancha [11].

En el proceso de activación química, se realiza la mezcla del

precursor carbonoso con el agente activante (KOH, H3PO4, ZnCl2, etc.) y

se trata térmicamente a elevada temperatura (entre 400 y 900 ºC) en

atmósfera inerte [7,10,12]. Posteriormente, es necesario realizar una etapa

de lavado para eliminar los productos de reacción y el agente activante

que no haya reaccionado [12]. En comparación con la activación física, la

activación química permite obtener materiales con un mayor desarrollo de

microporosidad. Además, los rendimientos suelen ser superiores [11–14].

En lo que concierne a la textura porosa, el parámetro fundamental (además

de la temperatura) para generar materiales de una porosidad determinada

es la relación entre la cantidad de precursor y el agente activante; sin

embargo, es difícil controlar la química superficial durante el proceso de

activación.

Capítulo 1

18

Figura 1.1. Esquema de los procesos de activación física y activación química [7].

El uso de distintos agentes activantes químicos condiciona la

microporosidad del carbón activado sintetizado. De este modo, la

activación con H3PO4 y ZnCl2 permite obtener materiales microporosos,

pero con una distribución de tamaños de microporos ancha [10,13,15,16].

En consecuencia, no es posible obtener carbones activados con elevada

superficie específica utilizando estos agentes activantes. Sin embargo, el

uso de hidróxidos metálicos como agentes activantes permite obtener

carbones activados esencialmente microporosos con una superficie

específica muy elevada [12,14,17]. Concretamente, la activación química

con KOH proporciona desarrollos mayores de microporosidad estrecha.

La optimización de distintas variables (tiempo de activación, temperatura

de activación, velocidad de calentamiento, etc.) permite obtener carbones

activados con áreas superficiales aparentes muy elevadas y una

distribución de tamaños de poros estrecha, alcanzando valores superiores

a 3000 m2/g con un tamaño medio de poro de 1.4 nm [17].

Precursor

Material

carbonizado

Carbón

activado

Carbón

activado

Activación

física

Activación

química

Carbonización Gasificación

CO2, H2O, etc.

Impregnación/Carbonización/Lavado

KOH, NaOH, H3PO4, etc.


19

1.1.2 Materiales carbonosos nanomoldeados.

Los materiales carbonosos nanomoldeados son una variante de los

materiales carbonosos nanoestructurados cuya característica principal es

su porosidad ordenada. Estos materiales están constituidos por láminas de

átomos de carbono (hibridación sp2) separadas entre sí, de manera que el

espacio resultante entre dichas láminas puede ser considerado como un

poro. A diferencia de los carbones activados, estos materiales presentan

una estructura porosa muy ordenada en el espacio.

Figura 1.2. Estructura de ZTC [18].

Estos materiales se sintetizan utilizando métodos de nanomoldeo, en

los que se emplea una plantilla, de manera que la estructura del material

obtenido depende de la plantilla utilizada. Las plantillas se pueden

clasificar como blandas (“soft-templates”), cuando emplean surfactantes

como moldes, o bien duras (hard-templates), como sólidos inorgánicos.

Capítulo 1

20

Algunos ejemplos de materiales carbonosos obtenidos mediante estas

técnicas son: carbones mesoporosos ordenados (OMC, del inglés ordered

mesoporous carbon), materiales de porosidad jerárquica (HPC,

hierarchical porous carbon) y materiales nanomoldeados obtenidos

utilizando zeolitas como plantilla (ZTC, zeolite templated carbon) [18].

Concretamente, los ZTCs se caracterizan por presentar una porosidad

muy elevada. Están formadas por láminas de grafeno curvadas, cuya

estructura es una réplica inversa de la zeolita empleada [19–21]. En

consecuencia, estos materiales presentan un tamaño de poro uniforme y

definido, que depende del sólido plantilla utilizado en la síntesis. En

particular, los ZTCs obtenidos a partir de zeolitas Y presentan propiedades

especialmente interesantes, debido a la elevada interconectividad de su

microporosidad (tamaño de poro: 1.2 nm), que le confiere una superficie

específica aparente muy elevada (de incluso 4000 m2/g). La Figura 1.2

presenta un modelo de estructura de este tipo de materiales. Además, su

red porosa presenta baja tortuosidad, de manera que facilita la difusión de

moléculas e iones a través de la misma [18]. Debido a la singularidad de

su estructura, estos materiales presentan una concentración muy elevada

de sitios borde accesibles (diez veces superior a los carbones activados

convencionales). Además, estos sitios presentan una reactividad superior

a la habitual [22]. Como consecuencia, estos materiales presentan un

elevado contenido en oxígeno cuando se exponen al aire. Además, su

química superficial se puede modificar por medio de diversas estrategias

[23–25]. Estas propiedades hacen que sea un material de interés en una


21

gran variedad de aplicaciones: supercondensadores, adsorción de gases,

pilas de combustible, baterías, soporte de catalizadores, etc. [18,20].

1.2. Química superficial de los materiales carbonosos

Como se ha indicado, los materiales carbonosos presentan una

química superficial muy heterogénea. Esto se debe a la presencia de sitios

activos en la superficie. Los más comunes son los átomos de carbono

insaturados en los extremos de las láminas grafénicas (sitios “borde”),

que, del mismo modo que las imperfecciones en los planos basales,

presentan una elevada densidad electrónica y, en consecuencia, pueden

quimisorber heteroátomos como el hidrógeno, oxígeno, nitrógeno, azufre,

etc.

1.2.1. Oxígeno

El oxígeno es el heteroátomo más abundante en los materiales de

carbón, debido a que reaccionan espontáneamente con el aire a bajas

temperaturas. Se encuentra en la superficie de los materiales carbonosos

en forma de grupos funcionales identificables por la química orgánica.

Las principales formas químicas del oxígeno en un material carbonoso se

resumen en la Figura 1.3 [6,26].

Se pueden generar grupos oxigenados mediante tratamientos de

oxidación en fase líquida o gaseosa [6]. Algunos de los reactivos más

utilizados son el O2 en fase gas y el HNO3, el (NH4)2S2O8 y H2O2 en fase

líquida. El tipo y cantidad de grupos generados depende del tratamiento

utilizado. Por ejemplo, tanto la oxidación mediante HNO3 como con aire,

genera grupos oxigenados en toda la superficie; sin embargo, el HNO3

Capítulo 1

22

produce una mayor cantidad grupos carboxílicos, mientras que el aire da

lugar a fenoles y carbonilos en mayor medida.

Figura 1.3. Grupos funcionales oxigenados presentes en la superficie de los materiales

carbonosos: (a) ácido carboxílico, (b) lactona, (c) hidroxilo, (d) carbonilo, (e) quinona,

(f) pirona, (g) éter, (h) anhídrido carboxílico, (i) cromeno y (j) lactol.

En general, estas especies se pueden clasificar atendiendo a su

carácter ácido-base. Se consideran grupos ácidos los ácidos carboxílicos,

anhídridos, hidroxilos, lactonas y lactoles. Sin embargo, existe una mayor

controversia en lo que concierne a los grupos funcionales básicos.

Algunos trabajos identifican como básicos a los grupos cromenos, pironas

y quinonas [26], aunque los electrones de los planes basales de las

láminas de grafeno también han sido identificados como sitios básicos

[27].

1.2.2. Nitrógeno

Los carbones activados no presentan nitrógeno de forma espontánea,

como sucede con el oxígeno. Los grupos funcionales nitrogenados


23

presentes en los carbones activados se pueden generar, de forma general,

mediante la carbonización/activación de un precursor que contenga

nitrógeno, o bien, mediante un post-tratamiento del material con un

reactivo nitrogenado, como NH3 o HCN [28].

La Figura 1.4 resume los grupos nitrogenados principales que se

pueden encontrar en la superficie de un material de carbón [6,26,28].

Figura 1.4. Grupos funcionales nitrogenados presentes en la superficie de los materiales

carbonosos: (a) piridina, (b) piridona, (c) piridina N-oxidada, (d) nitro, (e) nitroso, (f)

amida, (g) lactama, (h) nitrilo, (i) amina primaria, (j) amina secundaria, (k) amina

terciaria, (l) imina, (m) nitrógeno cuaternario en el borde de la lámina, (n) nitrógeno

cuaternario en el interior de la lámina y (ñ) pirrol.

Los grupos funcionales nitrogenados presentan, en general,

propiedades básicas que pueden facilitar la interacción entre moléculas

ácidas y la superficie del material carbonoso, mediante enlaces covalentes,

enlaces de hidrógeno y/o interacciones dipolo-dipolo. Esta característica

permite que los carbones activados nitrogenados se puedan utilizar en

distintas aplicaciones, tanto en fase gaseosa como líquida. En lo que

Capítulo 1

24

concierne a interacciones sólido-gas, estos materiales se pueden utilizar

para la eliminación de H2S [29,30], control de SO2 [31] y captura de CO2

[32–34]. En fase líquida, presentan una elevada capacidad de adsorción

de aniones y metales pesados. Además, también se ha investigado su uso

en la reacción de reducción de oxígeno en pilas de combustible y en

supercondensadores [9].

1.3. Síntesis de materiales carbonosos dopados con nitrógeno

Como se ha descrito, la química superficial es un parámetro

fundamental que condiciona gran parte de las aplicaciones de los

materiales carbonosos. En la actualidad, existen numerosos estudios

centrados en la introducción de distintos heteroátomos en este tipo de

materiales [28,35–37].

Se pueden introducir grupos funcionales nitrogenados en un material

carbonoso principalmente mediante dos métodos: (I) empleando en el

proceso de síntesis un precursor carbonoso que contenga nitrógeno; y (II)

el post-tratamiento de un material carbonoso con un reactivo nitrogenado

[10]. Ambas metodologías suelen emplear elevadas temperaturas en el

proceso de síntesis o modificación. La química superficial del material

resultante depende del tratamiento utilizado, que incluye el

precursor/reactivo nitrogenado y la temperatura del tratamiento. Esto se

debe a que no todos los compuestos nitrogenados son estables

térmicamente; los tratamientos a elevada temperatura (superior a 600 ºC)

generan especies aromáticas, como nitrógeno cuaternario, piridina y

pirrol; mientras que los tratamientos a baja temperatura (inferior a 600

ºC), producen especies como aminas, amidas y lactamas, que por


25

tratamiento térmico a temperaturas superiores pueden transformarse en las

especies aromáticas citadas [6].

1.3.1. Síntesis directa a partir de un precursor nitrogenado

Los métodos más comunes para obtener materiales carbonosos de

elevada porosidad funcionalizados con nitrógeno se pueden resumir del

siguiente modo [28,36]:

(i) Carbonización y activación física (con vapor de H2O o

CO2) [38] o química (mediante KOH, H3PO4, etc.) [39] de

un precursor nitrogenado, como poliacrilonitrilo,

polianilina o melamina.

(ii) Depósito químico en fase vapor de un precursor que

contenga nitrógeno (acetonitrilo) empleando distintas

plantillas (zeolitas, sílices, etc.) seguido de un tratamiento

térmico [24,25,40].

(iii) Carbonización hidrotermal de precursores biomásicos que

contengan nitrógeno (glucosamina, ácido cianúrico, etc.).

De forma general, estos materiales presentan mayor contenido y una

distribución homogénea del heteroátomo en la matriz del material

carbonoso (no exclusivamente en la superficie). Además, como la

temperatura habitual en estos procesos de síntesis es elevada (> 600 ºC),

estos materiales presentan una química superficial compuesta

principalmente por especies aromáticas, como pirrol y piridina. También

es habitual la presencia de nitrógeno cuaternario distribuido en las láminas

Capítulo 1

26

grafénicas de los materiales. Un ejemplo es el ZTC dopado con nitrógeno,

que presenta un elevado contenido de nitrógeno cuaternario [25].

1.3.2. Post-tratamiento de materiales carbonosos con reactivos

nitrogenados

Una de las metodologías más utilizadas para incorporar nitrógeno

en los materiales carbonosos se basa en post-tratamientos aplicados a un

material que ha sido sintetizado previamente. Los métodos más habituales

se basan en la reacción de un material carbonoso con un reactivo que

contenga nitrógeno por medio de un tratamiento térmico. No obstante,

también se han utilizado ampliamente estrategias basadas en reacciones

orgánicas para incorporar grupos funcionales en materiales

nanoestructurados, o inmovilizar metales en carbones activados. Por

último, las técnicas electroquímicas son de gran interés debido a su

elevada selectividad. A continuación, se detallan las características

principales de estos métodos de funcionalización.

1.3.2.1. Funcionalización con nitrógeno mediante tratamientos térmicos

Los materiales carbonosos pueden tratarse con reactivos que

contengan nitrógeno, en fase líquida o gaseosa, para introducir grupos

nitrogenados en la superficie. Se han utilizado reactivos muy variados,

como NH3 o HCN en fase gaseosa, o N,N-dimetilformamida, urea o

melamina en fase líquida [31,41]. El reactivo más empleado es el NH3,

que puede utilizarse solo (aminación) o en presencia de aire

(amoxidación). La química superficial y las propiedades texturales del

material dependen del material carbonoso de partida, del reactivo

nitrogenado y de la temperatura del tratamiento [42].


27

De este modo, la química superficial de los carbones nitrogenados

obtenidos mediante post-tratamiento puede ser más heterogénea que la de

los obtenidos a partir de un precursor carbonoso, o proporcionar grupos

funcionales distintos (como amidas, aminas, etc.), ya que pueden

presentar especies de baja estabilidad térmica si la temperatura del

tratamiento no es elevada. Además, el nitrógeno suele anclarse a la

superficie mediante una reacción química con los sitios reactivos de la

superficie del material carbonoso. En consecuencia, el contenido en

nitrógeno suele ser menor que en aquellos materiales obtenidos a partir de

un precursor carbonoso, donde el heteroátomo se encuentra inicialmente

en la estructura del precursor; sin embargo, este método puede ser más

selectivo hacia determinados grupos funcionales. Puesto que esta Tesis

Doctoral, se centra en la introducción de grupos funcionales nitrogenados

por este método, se va a explicar con más detalle la bibliografía

relacionada.

En este tipo de tratamientos, la textura porosa está fuertemente

influenciada por la temperatura. En el caso de tratamientos con NH3,

normalmente tiene lugar un aumento de la microporosidad y de la

superficie específica [43], y sólo en algunos casos disminuye [44] o se

mantiene constante [42]. Cuando se realizan tratamientos de amoxidación,

la estructura porosa y la superficie específica pueden no variar o bien

aumentar o decrecer. En el caso de emplear otros reactivos, como aminas

o HNO3, la superficie específica y la porosidad suele disminuir [45].

En los procesos de post-tratamiento, es frecuente realizar una etapa

previa de preoxidación en fase líquida, puesto que la presencia de grupos

Capítulo 1

28

oxigenados en los materiales carbonosos facilita la introducción de otros

heteroátomos mediante reacciones de sustitución, condensación y

similares, funcionando el grupo oxigenado como punto de unión con la

superficie a funcionalizar. Un ejemplo es la impregnación de un carbón

oxidado en HNO3 con urea, seguida de un tratamiento térmico, en el que

se observó que la adsorción de urea dependía de la cantidad y tipo de

especies oxigenadas presentes [46]. Este tipo de funcionalización se basa

en la interacción ácido-base entre una amina y un ácido carboxílico, en la

que este último es el compuesto que se encuentra sobre la superficie del

material. El ión amonio generado queda anclado a la superficie por un

enlace iónico, y mediante un tratamiento térmico da lugar a la formación

de amidas, imidas y nitrilos.

Otra estrategia para obtener materiales carbonosos de elevada

porosidad con grupos funcionales nitrogenados consiste en realizar

tratamientos térmicos en atmósfera inerte de materiales compuestos

carbón/polímero. Un ejemplo son los carbones activados y fibras de

carbón activadas obtenidas a partir de materiales compuestos

carbón/polianilina y carbón/polipirrol [47–49]. Los materiales

compuestos se pueden preparar por polimerización química o

electroquímica [47]. En estos materiales, el depósito de polímero sobre el

material carbonoso se produce mediante enlaces no covalentes, aunque es

posible la interacción covalente entre grupos funcionales del material

carbonoso y el polímero [47]. Esta metodología permite la generación de

una película de polímero conductor sobre la microporosidad del material

carbonosos [50], de manera que, cuando son tratados térmicamente,


29

conducen a la generación de materiales carbonosos avanzados en los que

se generan residuos del material carbonizado en la microporosidad del

material [48]. Los grupos funcionales generados dependen de la

temperatura a la que se realice el tratamiento térmico. En el caso de

materiales obtenidos a partir de polianilina, a elevada temperatura (800

ºC), se da la formación de heterociclos nitrogenados, con una mayor

selectividad de pirroles/piridonas. Esta estrategia permite obtener

materiales con propiedades electroquímicas interesantes para su

aplicación en condensadores electroquímicos [47,48] y pilas de

combustible [49].

1.3.2.2. Funcionalización química basada en reacciones orgánicas a baja

temperatura

Este procedimiento se ha empleado en diversos materiales

carbonosos para distintas aplicaciones. En carbones activados, se han

realizado estudios para introducir aminas, mediante un enlace amida, para

la adsorción de plomo [51] y para la captura de CO2 [34]; también para

anclar complejos de metales de aplicación catalítica, mediante reacciones

de amidación [52–54]. Las estrategias más habituales emplean un proceso

de oxidación previo para aumentar la concentración de sitios reactivos en

superficie. La generación de ácidos carboxílicos permite que reaccionen

con las aminas sin necesidad de utilizar un tratamiento térmico, que pueda

disminuir la porosidad. El método más utilizado para funcionalizar

materiales carbonosos es el tratamiento con SOCl2 en un disolvente

orgánico. Este tratamiento produce cloruros de ácido en la superficie, los

derivados de ácido son más reactivos, de manera que la sustitución

Capítulo 1

30

nucleofílica en este grupo funcional por aminas, alcoholes, etc., queda

facilitada. Por tanto, se trata de una ruta sintética adecuada para la

funcionalización de materiales carbonosos y, en concreto, para la

introducción de grupos nitrogenados (aminas y amidas) en la superficie.

Sin embargo, estas modificaciones pueden cambiar la textura porosa

del material. Tamai y col. [55] observaron que la microporosidad de los

carbones activados disminuye cuando se introduce etilendiamina y

hexametilendiamina, mientras que la mesoporosidad no resulta afectada.

Esto supone un inconveniente para la aplicación de estos materiales. En

lo que concierne a la captura de CO2, los grupos amina son deseables

porque pueden experimentar quimisorción de CO2, pero si la cantidad de

microporos, muy necesarios para que el material posea una elevada

capacidad de adsorción del mismo, se ve reducida, toda mejora causada

por la introducción de aminas puede verse disminuida y hasta anulada por

completo.

Por otro lado, los grupos nitrogenados pueden mejorar el

comportamiento de los electrodos de supercondensadores, ya que pueden

aumentar la capacidad del material mediante reacciones faradaicas

superficiales (pseudocapacidad), la conductividad eléctrica y la

mojabilidad, entre otros aspectos [9]. La capacidad del material es

directamente proporcional a la superficie del mismo, por lo que, como en

el caso de la captura de CO2, la disminución de su superficie como

consecuencia de la introducción de grupos funcionales es desfavorable.


31

Esta pérdida de superficie específica está relacionada con el tamaño

del grupo funcional anclado a la superficie, ya que los grupos muy

voluminosos pueden bloquear el acceso a los microporos. En este sentido,

existen estudios previos en los que la formación selectiva de aminas

superficiales (directamente enlazadas al anillo aromático del carbón

activado) mediante una ruta distinta de reacciones orgánicas (reducción

de grupos nitro generados por tratamiento en HNO3 concentrado), no

disminuye la porosidad de la muestra de carbón activado [56]. Se debe

tener en cuenta que la diferencia fundamental entre ambos métodos

[55,56] es que el primero introduce las aminas generando un grupo amida,

de tal manera que la superficie resultante presentará, por cada sitio

reactivo, un grupo amida y la amina enlazada al mismo, además de la

cadena alifática correspondiente; mientras que, en el segundo método, las

aminas están directamente ancladas a la superficie del material, por lo que

el volumen de la especie generada es mucho menor.

Por tanto, la introducción de grupos nitrogenados mediante

reacciones orgánicas permite obtener grupos funcionales de forma más

selectiva, sin necesidad de emplear un tratamiento térmico que pueda

influir en las propiedades texturales del material y que modifique de forma

no deseada su química superficial. Las propiedades texturales del material

resultante están relacionadas con el tamaño de los grupos funcionales

generados en la superficie que, si no es elevado, no debe disminuir

significativamente la superficie específica.

Algunos investigadores han realizado funcionalizaciones similares a

las descritas anteriormente con el fin de introducir amidas primarias [57]

Capítulo 1

32

y aminas aromáticas [58] en la superficie de nanotubos de carbono. En

ambos casos, los nanotubos de carbono se tratan previamente para generar

ácidos carboxílicos en la superficie. Posteriormente, reaccionan con

SOCl2 para generar cloruros de ácido y, a continuación, ambas muestras

se tratan con reactivos nitrogenados (disolución acuosa de NH3 y

(NH4)2CO3/Piridina/DMF, respectivamente) para generar amidas

primarias.

En el segundo caso [58], los investigadores realizan una reacción más

para generar aminas primarias a partir de amidas primarias. Para esto,

utilizan la transposición de Hofmann, una reacción ampliamente

estudiada en química orgánica que consiste en tratar una amida primaria

con bromo en medio básico. En estas condiciones, el bromo reacciona con

el nitrógeno de la amida y genera un intermedio de reacción (bromoamida)

que, en medio básico, se desprotona dando lugar a la transposición del

sustituyente de la amida y la pérdida del anión bromuro, de tal manera que

se forma un isocianato. Esta especie reacciona con la base presente en el

medio. En el caso de NaOH, se forma un ácido carbámico, que en medio

básico se desprotona y disocia en forma de CO2, dando lugar a una amina

primaria. Si la base presente es NaOCH3, la especie que se genera es un

carbamato estable. En este caso, la hidrólisis de esta especie genera la

amina primaria [59].

Esta última ruta de modificación es de especial interés si el material

en el que se aplica es un sólido microporoso, ya que, como se ha indicado

anteriormente, permite introducir grupos funcionales nitrogenados en

condiciones suaves que no alteran en gran medida la porosidad, que


33

resulta fundamental en aquellas aplicaciones en las que los carbones

nitrogenados son interesantes. Además, permite determinar la influencia

de grupos superficiales de naturaleza muy diferente en un material

carbonoso concreto y, por tanto, estimar cómo modifica la química

superficial las propiedades de estos materiales y su efecto en distintas

aplicaciones, como captura de CO2, supercondensadores o reducción de

oxígeno.

Por estos motivos, en esta Tesis Doctoral se plantea la modificación

de la química superficial de un carbón activado mediante reacciones

orgánicas y estudiar el efecto de la química superficial en su

comportamiento como electrodos de supercondensadores. La ruta de

modificación consiste en una oxidación química con HNO3 para generar

preferentemente ácidos carboxílicos, la posterior activación de estos

grupos mediante SOCl2, que facilite la creación de grupos funcionales

amida y, consecutivamente, la obtención de aminas a partir de estos. La

ruta de modificación se ilustra en la Figura 1.5.

Figura 1.5. Modificación del carbón activado KUA mediante: (a) oxidación química

con HNO3 (KUA-COOH), (b) tratamiento con SOCl2 (KUA-COCl), (c) amidación

(KUA-CONH2) y (d) aminación (KUA-NH2).

Capítulo 1

34

1.3.2.3. Funcionalización electroquímica

El uso de técnicas electroquímicas ha suscitado gran interés para

modificar la química superficial de materiales carbonosos. Esto se debe a

que estas técnicas son sencillas de aplicar y controlar, se dan a temperatura

ambiente y presión atmosférica, y proporcionan elevada reproducibilidad,

sensibilidad y selectividad [60].

Las técnicas electroquímicas se han utilizado ampliamente para

introducir grupos funcionales nitrogenados en los materiales carbonosos.

El proceso de funcionalización se basa en la aplicación de una corriente

para anclar una molécula disuelta, un ión del elecrolito o incluso una

molécula del disolvente en la superficie del material. La técnica más

conocida es el uso de sales de diazonio con distintos heteroátomos para

modificar el material con grupos amina o grupos nitro, aunque también es

posible introducir otros heteroátomos (Cl, Br, etc.) [61–63]. Otros

ejemplos se basan en la oxidación de aminas o ácidos aminobenzoicos

para anclar diversos grupos funcionales a la superficie [64][65]. Estas

estrategias se han empleado con materiales carbonosos de elevada

porosidad, permitiendo conservar las propiedades derivadas de la textura

porosa del material sin funcionalizar [23].


35

1.4. Aplicaciones en almacenamiento y generación de energía

eléctrica.

1.4.1. Condensadores electroquímicos

Los condensadores electroquímicos o supercondensadores son

sistemas de almacenamiento de energía de elevada potencia compuestos

por dos electrodos en contacto con un electrolito y separados por una

membrana. Estos dispositivos presentan un amplio abanico de

aplicaciones: vehículos eléctricos, dispositivos digitales, etc. En

comparación con los condensadores convencionales, estos dispositivos

proporcionan densidades de energía superiores, debido al uso de

electrodos que permiten un almacenamiento superior de carga. La Figura

1.6 presenta el diagrama de Ragone con los principales dispositivos de

almacenamiento y generación de energía. En comparación con las baterías

y las pilas de combustible, los supercondensadores presentan una potencia

muy superior, pero una densidad de energía limitada. Por estos motivos,

es necesario continuar investigando para poder alcanzar densidades de

energía que permitan extender su uso a aplicaciones propias de las baterías

[9,66–68].

Capítulo 1

36

Figura 1.6. Diagrama de Ragone [69].

Dependiendo del mecanismo de almacenamiento de carga, se pueden

clasificar como condensadores de doble capa eléctrica (EDLC, electric

double-layer capacitor) o pseudocondensadores [8,9]. Los EDLC están

basados en interacciones electrostáticas debidas a la formación de una

doble capa eléctrica en la interfase electrodo-disolución. Este mecanismo

es el predominante cuando se utilizan materiales carbonosos de elevada

porosidad como electrodos. En el caso de los pseudocondensadores, el

almacenamiento de energía viene dado por reacciones farádicas

reversibles que se dan en la superficie del electrodo. Este proceso es el

que rige el mecanismo de almacenamiento de carga cuando se utilizan

óxidos metálicos o polímeros conductores como electrodos del

condensador electroquímico. En esta Tesis Doctoral, los condensadores

electroquímicos empleados son de tipo doble capa eléctrica. La figura 1.7

muestra los componentes de un condensador electroquímico basado en

materiales carbonosos porosos.

Energía específica (Wh/kg)

Po

ten

cia

es

pe

cífic

a (W

/kg

)


37

Figura 1.7. Esquema de un condensador electroquímico o supercondensador basado en

carbones activados.

Debido a la naturaleza electrostática del fenómeno de energía, los

EDLC se pueden cargar y descargar en segundos, lo que les confiere una

potencia muy elevada. Además, el mecanismo es idealmente reversible,

por lo que presentan una elevada eficiencia coulómbica y tiempos de vida

media muy superiores a los de las baterías. Concretamente, los EDLC

pueden cargarse y descargarse durante miles de ciclos (>500000 ciclos),

sin perder la capacidad original del dispositivo.

1.4.1.1. Modelos de doble capa eléctrica

Para comprender en mayor medida los procesos de almacenamiento

de energía que suceden en los EDLC es necesario describir las

características de la doble capa eléctrica que se forma en la interfase

electrodo/disolución. La doble capa eléctrica es una consecuencia de la

separación de cargas que se produce entre dos fases debido a la diferencia

de potencial causada por su distinta composición química.

---------

Carbón activadoMembrana porosaColector de corriente

+++++++++

Capítulo 1

38

Se han propuesto distintos modelos para describir el proceso de

formación de la doble capa eléctrica [70,71]. La Figura 1.8 muestra el

esquema propuesto por Helmholtz (1874) para describir la doble capa

eléctrica. Este modelo se basa en la formación de dos láminas paralelas

cargadas de signo opuesto. Una de ellas se encuentra localizada en la

superficie del electrodo y la otra en la disolución (capa rígida). La

separación entre ambas es del orden del Angstroms. La carga

correspondiente a los iones en disolución neutraliza la carga superficial

del electrodo. Según este modelo, el potencial eléctrico disminuye

linealmente con el aumento de la distancia al electrodo, hasta hacerse nulo

en el plano de la capa rígida.

Figura 1.8. (a) Esquema de la doble capa eléctrica y (b) variación del potencial con la

distancia al electrodo según el modelo de Helmholtz.

Gouy y Chapman propusieron en 1913 un modelo de capa difusa de

la doble capa eléctrica. En este modelo, no tiene lugar la formación de una

capa rígida de los iones en la disolución, sino una capa difusa como

consecuencia de la agitación térmica que neutraliza globalmente la carga

del electrodo (Figura 1.9a). En este modelo, el potencial eléctrico varía de

forma exponencial con la distancia al electrodo (Figura 1.9b).

(a) (b)

-

+++++++++

-

-

-

-

-

-

-

-

-

-

-

- -

+

++

++

+

+

-

-

--

-

-

-

-

-

-

++

++

+

+ +

+

+

+

+

+

--

-

-


39


distancia al electrodo según el modelo de Gouy-Chapman.

Posteriormente, Stern propuso en 1924 un modelo que combina los

modelos propuestos por Helmholtz y Gouy-Chapman, con la formación

de una capa rígida cerca del electrodo y una difusa. Este modelo tiene en

cuenta el movimiento hidrodinámico de los iones en la capa difusa y la

acumulación de los mismos en la superficie del electrodo. De esta forma,

la carga del electrodo se compensa tanto por los iones que se encuentran

cerca del mismo como por los que se encuentran en la capa difusa (Figura

1.10a). En este modelo, el potencial eléctrico varía de forma lineal cerca

del electrodo y disminuye de forma exponencial cuando aumenta la

distancia al mismo (Figura 1.10b) y la capacidad total de la interfase se

puede considerar como dos condensadores en serie.

(a) (b)+++++++++

-

--

- -

---

-

- -

--

-

--

-

- -+

++

+ +

+

+

-

-

-

-

-

-

-

-

-

--

Capítulo 1

40


distancia al electrodo según el modelo de Stern.

Finalmente, Grahame propuso en 1947 un nuevo modelo teniendo en

cuenta la solvatación de los iones en disolución y el carácter dipolar del

agua. Teniendo en cuenta estas variables, se propone la existencia de

distintos planos de acercamiento a la superficie del electrodo,

dependiendo de si el ión se encuentra solvatado o interacciona de forma

directa. Este modelo propone la existencia de diversas capas (Figura

1.11): (i) la capa real en la superficie del electrodo; (ii) una segunda capa

correspondiente al plano interno de Helmholtz en la disolución (IHP), en

la que los iones se encuentran adsorbidos en la superficie del electrodo y

han perdido su esfera de solvatación; (iii) una tercera capa, denominada

plano externo de Helmholtz (OHP), donde se encuentran los iones

solvatados. Las cargas se distribuyen en la zona compacta, de espesor 0.5

nm aprox. (debido a las moléculas de disolvente e iones), y una capa

difusa, de 1 a 100 nm, debido a los iones distribuidos térmicamente.

(a) (b)+++++++++

-

-

---

-

-

-

--

-

--

-

-

-

-

-

-

+

+

+

+

+

+

+

+

++

-

--

-

-

-

-

-


41

Figura 1.11. Esquema de la doble capa eléctrica según el modelo de Grahame.

1.4.1.2. Características de los condensadores electroquímicos

1.4.1.2.1. Energía y potencia

Los dos parámetros fundamentales para caracterizar un sistema de

almacenamiento son la energía y la potencia.

El trabajo realizado cuando se aplica un voltaje en un

supercondensador para trasladar una carga a la interfase

electrodo/electrolito, viene dado por la ecuación (1.2).

𝑑𝑤 = 𝑉 𝑑𝑄 (1.1)

Donde dW es el diferencial de trabajo, V es el voltaje y dQ el

diferencial de carga.

El trabajo realizado es equivalente a la energía almacenada

(considerando despreciable las pérdidas por disipación de calor), y el

voltaje está relacionado con la carga y la capacidad, por lo que se puede

integrar la ecuación (1.2) como se indica a continuación:

+++++++++

+

+

+

+

+

+

+

-

-

-

--

Catión

solvatado-

+

Anión adsorbido

específicamente

IHP

OHP

Capítulo 1

42

𝑊 = 𝐸 = ∫ 𝑉 𝑑𝑄𝑄

0= ∫

𝑄

𝐶

𝑄

0 𝑑𝑄 =

1

2 𝐶𝑉2 (1.2)

Donde W es el trabajo realizado, E es la energía y C es la capacidad.

Por tanto, la energía de un condensador depende fundamentalmente de la

capacidad y el voltaje aplicado. Ambas variables vienen dadas por los

componentes utilizados en su fabricación (electrodos y electrolitos).

La potencia máxima suministrable del condensador electroquímico

viene dada por la ecuación (1.3).

𝑃 = 𝑉2

4 𝐸𝑆𝑅 (1.3)

Donde P es la potencia máxima y ESR es la resistencia equivalente

en serie del dispositivo.

De las expresiones de energía y potencia características de un

condensador electroquímico, se puede deducir que las características

principales de estos dispositivos son: resistencia equivalente en serie,

voltaje y capacidad. A continuación, se detallan los factores principales

que afectan estas variables.

1.4.1.2.2. Resistencia equivalente en serie

La resistencia equivalente en serie de un condensador electroquímico

depende de los componentes de la celda electroquímica: (i) la resistencia

del electrodo (que incluye la resistencia interpartícula e intrapartícula); (ii)

la resistencia entre el electrodo y el colector de corriente; (iii) la resistencia

del electrolito; (iv) la resistencia debida a la difusión de los iones en la


43

estructura porosa del material; (v) resistencia del separador; (vi)

resistencias de contacto externas [72].

La pérdida de potencia generada por el aumento de ESR produce un

aumento significativo del calor del sistema que puede limitar su

comportamiento. Cuando el dispositivo se degrada (debido a su operación

a elevada potencia, elevada temperatura o a la generación de gases),

aumenta la ESR y conduce a un empeoramiento del comportamiento del

dispositivo y reducción del tiempo de vida [72].

Por estos motivos, se debe prestar especial atención a minimizar la

ESR de los electrodos carbonosos. Tradicionalmente, los electrodos de

carbones activados se preparan utilizando un aglomerante que permite

conformar el material [73]. Algunos ejemplos de los aglomerantes son:

teflón (PTFE, del inglés politetrafluoroetilene), fluoruro de polivinilideno

(PVDF, del inglés polyvinylidene) y carboximetilcelulosa (CMC). Este

componente aumenta el contacto interpartícula; sin embargo, los

aglomerantes habituales disminuyen la conductividad global del electrodo

(al tener una conductividad menor que los carbones activados). Por este

motivo, la elaboración de los electrodos incluye el uso de promotores de

conductividad. El más empleado es el negro de carbón, aunque también

se ha propuesto el uso de nanotubos de carbono [74].

Algunas estrategias novedosas para minimizar la resistencia de los

electrodos son: (i) empleo de materiales avanzados de elevada

conductividad como fase activa, como los nanotubos de carbono [74]; (ii)

reducir la tortuosidad de la red porosa de los electrodos, mediante el

Capítulo 1

44

empleo de materiales carbonosos nanoestructurados de porosidad

ordenada [74,75]; (iii) sintetizando el material carbonoso directamente

sobre el colector de corriente [76,77] ; (iv) preparación de monolitos de

materiales carbonosos, con una estructura continua que mejore el contacto

interpartícula en comparación con el carbón activado en polvo [78].

La ESR se puede determinar a partir del diagrama de Nyquist (por

medio de EIS) o bien a partir de la caída óhmica cuando se descarga el

condensador a corriente constante. Estas técnicas se discuten en mayor

medida en el Capítulo 2.

1.4.1.2.3. Voltaje

En las aplicaciones prácticas de los condensadores electroquímicos,

es necesario utilizar ventanas de potencial (o voltajes) elevados para

maximizar el almacenamiento de carga. Sin embargo, el voltaje depende

del electrodo y del electrolito. En el caso del electrolito, la estabilidad

depende intrínsecamente del disolvente (acuoso, orgánico y líquido

iónico) y de la sal conductora utilizada. En medio acuoso, el potencial

termodinámico de descomposición del agua limita la operación de los

supercondensadores a voltajes en torno a 1V. En el caso de electrolitos

orgánicos, se pueden alcanzar valores de 2.5 – 3V. Finalmente, el uso de

líquidos iónicos como electrolitos (libres de disolvente) permiten alcanzar

voltajes de incluso 5V.

Los materiales carbonosos presentan una superficie casi idealmente

polarizable en disoluciones electrolíticas. No obstante, se deben emplear

materiales con elevada estabilidad electroquímica (elevado sobrepotencial


45

para las reacciones de descomposición del electrolito) para obtener una

durabilidad superior.

1.4.1.2.4. Capacidad

Los mecanismos de almacenamiento de energía en un condensador

electroquímico son: (i) formación de la doble capa eléctrica debido a la

adsorción de los iones en la superficie del material carbonoso (proceso

capacitivo) y (ii) reacciones redox rápidas y reversibles, que se dan en la

superficie de los electrodos (proceso pseudocapacitivo). En los

condensadores de doble capa eléctrica (basados en materiales

carbonosos), el mecanismo principal es la electroadsorción de iones en la

porosidad del material. No obstante, la presencia de grupos funcionales en

la superficie puede dar lugar a reacciones redox rápidas que contribuyen

al almacenamiento de energía mediante pseudocapacidad.

Los métodos de determinación de la capacidad de un condensador y

de un electrodo se detallan en el Capítulo 2.

1.4.1.3. Configuraciones de condensador electroquímico

1.4.1.3.1. Condensador simétrico

La configuración simétrica es la más sencilla y común de los

condensadores electroquímicos. Consiste en utilizar dos electrodos del

mismo material con la misma masa. Se trata de la configuración más

frecuente y más sencilla de utilizar en la industria. Sin embargo, en esta

configuración no es posible utilizar totalmente las prestaciones de los

electrodos y electrolitos para maximizar el almacenamiento de carga.

Capítulo 1

46

1.4.1.3.2. Condensador asimétrico

Los materiales utilizados como electrodos en condensadores pueden

no mostrar la misma capacidad en el intervalo de potenciales en los que

cada electrodo trabaja. El caso más destacable es cuando se emplean

materiales esencialmente pseudocapacitivos (polímeros conductores y

óxidos metálicos), en los que las reacciones redox se producen a

potenciales específicos, y por tanto no se da una variación constante de la

capacidad con el voltaje. En estos casos, la capacidad depende del

potencial aplicado. En el caso de los condensadores basados en materiales

carbonosos, este efecto se puede dar cuando se utilizan electrolitos con

tamaños de catión y anión diferentes, y el tamaño de poro efectivo para la

electroadsorción es similar al de uno de los iones [79]. En estos casos, la

adsorción de uno de los iones será superior a la del otro, y presentará

mayor capacidad en el intervalo de potencial en el que su adsorción sea

favorable. En estos casos, se puede optimizar la masa de los electrodos

para maximizar la energía almacenada en el condensador. La relación

óptima de masas fue propuesta por Snook y col. [80]:

𝑚+

𝑚−= √

𝐶−

𝐶+ (1.4)

Donde 𝑚+ y 𝑚− son las masas de los electrodos positivo y negativo,

respectivamente, y 𝐶+ y 𝐶− son las capacidades de los electrodos positivo

y negativo, respectivamente.

Otro de los motivos de desaprovechamiento de la energía máxima de

un condensador es que los materiales sean estables en ventanas de


47

potenciales significativamente diferentes. En un condensador en

funcionamiento, el electrodo positivo se polariza desde el potencial a

circuito abierto hasta potenciales positivos, que están limitados por la

descomposición del electrolito. El electrodo negativo funciona de forma

análoga, pero a potenciales negativos. Si el electrodo positivo trabaja en

una menor ventana de potenciales como consecuencia de que el potencial

a circuito abierto es alto y se alcanza pronto la descomposición del

electrolito, el electrodo negativo no podrá emplear toda su ventana de

potenciales de estabilidad completa; por tanto, no se podrá utilizar el

voltaje teórico completo que proporcionan ambos electrodos en dicho

electrolito. En este caso, se puede determinar la relación óptima de masas

conociendo la ventana de estabilidad y la capacidad de los electrodos en

dichas ventanas, estableciendo una relación igual de cargas entre el

electrodo positivo y negativo [81]:

𝑄 = 𝑄+ = 𝑄− (1.5)

𝑚+ · 𝐶+ · 𝛥𝐸+ = 𝑚− · 𝐶− · 𝛥𝐸− (1.6)

𝑚+

𝑚−=

𝐶−·𝛥𝐸−

𝐶+· 𝛥𝐸+ (1.7)

Donde Q es la carga del condensador electroquímico, 𝑄+ es la carga

del electrodo positivo, 𝑄− es la carga del electrodo negativo, 𝛥𝐸+ y 𝛥𝐸−

son las ventanas de potenciales de estabilidad de los electrodos positivo y

negativo, respectivamente.

Capítulo 1

48

1.4.1.3.3. Condensador híbrido

Los condensadores híbridos son aquellos que emplean distintos

materiales como electrodo positivo y negativo. Normalmente, uno de los

electrodos presenta comportamiento propio de condensador y el otro

presenta un comportamiento característico de batería [8]. En estos

sistemas, se pueden alcanzar voltajes más elevados.

1.4.1.4. Tipos de electrolitos

Los electrolitos utilizados en los condensadores electroquímicos

pueden ser líquidos, sólidos o semi-sólidos o “activos-redox” [82]. En esta

Tesis Doctoral, los electrolitos empleados son líquidos.

Los electrolitos líquidos se clasifican como acuosos, orgánicos o

líquidos iónicos. Las propiedades principales a tener en cuenta para

seleccionar un electrolito son su ventana de potenciales de estabilidad y

su conductividad iónica. La ventana de potenciales de estabilidad es un

parámetro clave en lo que concierne a la energía específica, mientras que

la conductividad iónica tiene una influencia muy notable en la potencia

del dispositivo [83]. A continuación, se describen las propiedades

principales de los distintos tipos de electrolitos.

1.4.1.4.1. Electrolitos acuosos

Los electrolitos acuosos presentan diversas ventajas para su uso en

supercondensadores. En comparación con los electrolitos orgánicos y

líquidos iónicos, tienen mayor conductividad (~ 1 S/cm) y proporcionan

valores superiores de capacidad específica (de hasta 300 F/g por medio de


49

carbones activados de elevada microporosidad [3,84]) Además, presentan

menor coste, son menos perjudiciales para el medio ambiente y facilitan

la construcción del dispositivo, dado que no requieren el uso de cámaras

de atmósfera inerte, como sucede con los electrolitos orgánicos y líquidos

iónicos. Sin embargo, el potencial termodinámico de descomposición del

agua (1.23 V) limita los voltajes de estos supercondensadores a

aproximadamente 1V. Por tanto, los condensadores electroquímicos

basados en electrolitos acuosos proporcionan valores de energía inferiores

a los característicos en medio orgánico o líquido iónico [9,82,83].

Los electrolitos acuosos se pueden clasificar como: ácidos, básicos y

neutros. Los más representativos de cada uno de estos grupos son H2SO4,

KOH y Na2SO4. En medio ácido, el voltaje suele estar limitado a 1-1.3 V

[82]. Sin embargo, dado que los materiales carbonosos pueden presentar

distintos sobrepotenciales para la descomposición del agua, se ha

conseguido ampliar el voltaje a 1.6 V utilizando materiales distintos como

electrodos positivo y negativos y balanceando la masa de los electrodos

[85]. En el caso de los electrolitos neutros, la capacidad específica y la

conductividad iónica es inferior a la de H2SO4 y KOH, por lo que los

dispositivos presentan mayor ESR. Sin embargo, estos electrolitos son

menos corrosivos y presentan mayor ventana de estabilidad, por lo que

permiten alcanzar voltajes superiores. Se han determinado valores de 1.9-

2.2V empleando configuraciones simétricas en Li2SO4 [86,87]. En

Na2SO4, se han obtenidos voltajes de incluso 2.2 V por medio de

configuraciones asimétricas [77].

Capítulo 1

50

1.4.1.4.2. Electrolitos orgánicos

Los electrolitos orgánicos más comunes están basados en sales de

amonio cuaternarias disueltas en un disolvente orgánico, como el

carbonato de propileno (PC) o el acetonitrilo (ACN). Las sales más

empleadas son tetrafluoroborato de tetraetilamonio (TEABF4) y

tetrafluoroborato de tetraetilmetilamonio (TEMABF4). Estos electrolitos

tienen menor conductividad (~ 0.02 S/cm) y mayor coste que los acuosos,

y proporcionan capacidades específicas inferiores (entre 150 y 200 F/g)

[9,88]. Sin embargo, permiten alcanzar valores mayores de energía que

los electrolitos acuosos, debido a la elevada estabilidad electroquímica,

que permite trabajar a voltajes de 2.5 – 2.8 V en electrolitos orgánicos

convencionales. Por estos motivos, son los electrolitos más empleados en

dispositivos comerciales [82,83]. No obstante, los condensadores basados

en electrolitos orgánicos se degradan fácilmente en condiciones de

operación severas (elevada temperatura y elevado voltaje), que son

necesarias en diversas aplicaciones de estos dispositivos. Además, la

presencia de trazas de agua puede disminuir notablemente el voltaje del

dispositivo [9]. La degradación se debe principalmente a la disminución

de la estabilidad electroquímica a elevada temperatura, así como a la

oxidación del material y descomposición del electrolito en condiciones de

elevado voltaje [89,90]. Por estos motivos, se está realizando un gran

esfuerzo investigador para desarrollar electrolitos orgánicos novedosos,

basados en sales conductoras y disolventes orgánicos nuevos, que

permitan mejorar las propiedades fisicoquímicas y aumentar la ventana de

estabilidad de los electrolitos orgánicos, así como optimizar su interacción


51

con los materiales carbonosos. Concretamente, el uso de sales basadas en

el catión pirrolidinio (Pyr14BF4) ha permitido alcanzar voltajes superiores

a 3V en PC. También se ha encontrado que el uso de disolventes orgánicos

como adiponitrilo (ADN) y 2,3 carbonato de butileno (2,3 BC), en

combinación con sales conductoras convencionales, permite trabajar a

voltajes superiores al PC empleando sales conductoras convencionales a

temperatura ambiente [91–93].

1.4.1.4.3. Líquidos iónicos

Los líquidos iónicos se definen como sales que se encuentran en

estado líquido a temperaturas inferiores a los 100 ºC [94,95]. Están

compuestos por un catión orgánico asimétrico y una anión orgánico o

inorgánico [82]. Debido a sus propiedades únicas, esta clase de

compuestos está alcanzando un gran impacto en diversas aplicaciones en

el campo de la energía (supercondensadores, baterías, pilas de

combustible, etc.) debido a que presentan una serie de propiedades que se

pueden optimizar según la aplicación requerida [96].

Las propiedades de los líquidos iónicos varían en función de la

naturaleza química de los cationes y aniones que los componen [96]. En

general, presentan bajas presiones de vapor y baja inflamabilidad, por lo

que se consideran más seguros que los disolventes orgánicos [97].

Además, presentan elevada estabilidad química y electroquímica, y dada

su conductividad intrínseca, se pueden emplear con electrolitos sin

necesidad de añadir un disolvente. Una ventaja adicional es su elevada

Capítulo 1

52

estabilidad térmica, que permite su aplicación en un intervalo amplio de

temperaturas (entre -30 y +60 ºC) [68].

Los condensadores basados en líquidos iónicos proporcionan

capacidades específicas similares a las proporcionadas en electrolitos

orgánicos [98]. No obstante, permiten alcanzar voltajes superiores a 3V,

por lo que pueden alcanzar valores de energía mayores que los basados en

otros electrolitos [94,97]. Sin embargo, los líquidos iónicos presentan

elevada viscosidad y menor conductividad que los electrolitos orgánicos

(< 0.01 S/cm), por lo que limitan la ESR del condensador y, por tanto, su

potencia [92,97]. Para superar está limitación, se ha investigado el uso de

electrolitos basados en mezclas de líquidos iónicos y disolventes

orgánicos. Estos electrolitos permiten aumentar la conductividad y

disminuir la viscosidad, pero mantienen en gran medida el voltaje de

operación, ya que pueden alcanzar valores de incluso 3.5 V [92,99].

1.4.1.5. Materiales carbonosos como electrodos de supercondensadores

Como se ha descrito en las secciones anteriores, los materiales

carbonosos presentan una combinación de propiedades fisicoquímicas

únicas para su aplicación en condensadores electroquímicos. Estos

materiales presentan una conductividad eléctrica relativamente elevada y

una superficie específica elevada, parámetros indispensables para

proporcionar potencia y capacidad elevadas, respectivamente. Además, su

relativamente bajo coste es un parámetro clave para aplicaciones

industriales. Se pueden encontrar en una variedad elevada de morfologías

(polvo, fibras, telas, monolitos, etc.), que permite adecuarlos a las

características necesarias en distintas aplicaciones (dispositivos flexibles,


53

móviles o sistemas estacionarios). Sin embargo, la propiedad más

interesante es la gran diversidad que presentan en términos de textura

porosa y química superficial. Ambas propiedades se pueden modificar por

medio de distintas estrategias de síntesis o post-tratamientos para adecuar

su uso a aplicaciones muy diversas. En condensadores electroquímicos,

se ha demostrado que estos parámetros alteran las propiedades

electroquímicas de los materiales carbonosos, por lo que es posible

adaptar su uso a distintos electrolitos y sistemas [83].

A continuación, se describe la influencia de las características de los

materiales carbonosos en su aplicación como electrodos de

supercondensadores.

1.4.1.5.1. Influencia de la textura porosa

La textura porosa de los materiales carbonosos (definida por

superficie específica, volumen de poros y distribución de tamaños de

poro) es un parámetro clave en su comportamiento como electrodos de

supercondensadores [9,84].

En general, la capacidad específica de los materiales carbonosos es

directamente proporcional a la superficie específica a valores bajos, pero

alcanza un valor constante a superficies específica superiores a 1200-1500

m2/g [100]. Una explicación plausible para este efecto es el decrecimiento

del espesor medio de las paredes de los poros en carbones muy activados,

que produce que el campo eléctrico (y la densidad de carga

correspondiente) no decaiga a cero entre las paredes del poro. Por otro

lado, se ha observado que el tamaño medio de poro aumenta con la

Capítulo 1

54

superficie específica cuando el grado de activación aumenta. Esto sugiere

que la interacción de los iones con las paredes de los poros es más débil

en poros de mayor tamaño y el efecto del desarrollo de la porosidad es

contrarrestado por el aumento de la distancia ión-pared [101], de manera

que la capacidad normalizada (por la superficie específica) aumenta

conforme disminuye el tamaño medio de los microporos. Sin embargo, el

mecanismo de almacenamiento de carga no se puede relacionar

exclusivamente con efectos superficiales. Los tamaños de los iones, de los

poros y la conexión entre estos y con la superficie externa de las partículas

del electrodo tienen una gran influencia en los valores de capacidad,

especialmente en condiciones de elevada potencia [9].

Se ha demostrado que la formación de la doble capa eléctrica está

determinada por el tamaño de los iones y de los poros. La presencia de

microporosidad permite alcanzar valores elevados de capacidad.

Raymundo-Piñero y col. [101] estudiaron el efecto del tamaño de poro en

electrolito orgánico y acuoso. En ambos electrolitos, la capacidad

normalizada aumentaba conforme disminuye el tamaño medio de los

microporos. Dado que el tamaño de los iones es mayor en los electrolitos

orgánicos, el tamaño medio de poro óptimo para la formación de la doble

capa es superior en este medio. Teniendo en cuenta que el tamaño del

catión (no solvatado) es más cercano al tamaño medio de poro que el

catión solvatado, es posible concluir que los iones penetran los poros sin

su esfera de solvatación. No obstante, los materiales carbonosos suelen

presentar distribuciones de tamaños de poros suficientemente anchas que

sugieren una contribución de los poros de tamaño superior al medio. De


55

hecho, la presencia microporos de mayor tamaño (> 0.7 nm) y mesoporos

es beneficiosa en materiales de porosidad jerárquica para promover una

difusión rápida de los iones hacia la entrada de lo microporos estrechos

[102]. Sin embargo, un exceso de mesoporosidad puede ser perjudicial

para la aplicación del material, dado que disminuye la capacidad

volumétrica del dispositivo.

Además, se pueden dar efectos de tamiz molecular, en los que no se

da electroadsorción de cationes comunes (Mg2+, Li+) cuando el tamaño de

poro es menor que el de los cationes hidratados, por lo que no contribuyen

a la formación de la doble capa eléctrica [79,103,104]. Este efecto se

observó también empleando líquidos iónicos como electrolitos, en los que

se puede descartar el efecto de la solvatación del disolvente [105].

Finalmente, la conectividad de los poros con la superficie de la

partícula también juega un papel fundamental en la formación de la doble

capa eléctrica. Bleda-Martínez y col. [84] demostraron la importancia de

la disposición de la microporosidad en la superficie externa del material.

En las fibras de carbón, la microporosidad se encuentra dispuesta

principalmente de forma perpendicular al eje de la fibra, por lo que

presenta baja tortuosidad. En el caso de los carbones activados, se

caracterizan por una red porosa muy desordenada, que le confiere un

grado elevado de tortuosidad. En consecuencia, el carbón activado mostró

una pérdida de capacidad completa en condiciones de medida de elevada

frecuencia (elevada potencia) en 0.5 M Na2SO4, mientras que la fibra de

carbón retuvo el 25% de capacidad. Este se debe a la mayor tortuosidad

en el camino difusional del electrolito en el carbón activado.

Capítulo 1

56

Figura 1.12. Esquema de los problemas difusionales experimentados por los iones para

acceder a la superficie interna de carbones activados y fibras de carbón activadas.

Por tanto, los materiales con una porosidad accesible son deseables

en aplicaciones que requieran elevada potencia. Por estos motivos, los

materiales carbonosos de porosidad jerárquica, donde los macroporos,

mesoporos y microporosos se encuentran ordenados, de manera que

disminuye el camino difusional del electrolito, son de especial interés para

estas aplicaciones [106]. Del mismo modo, los carbones nanomoldeados

con porosidad ordenada y distribución de tamaños de poros homogénea

[18] son de gran utilidad para ser empleados en dispositivos de elevada

potencia.

1.4.1.5.2. Influencia de la química superficial

Como se ha descrito a lo largo de la introducción, la química

superficial es un parámetro característico de los materiales carbonosos que

tiene una gran influencia en diversas aplicaciones. Además, se puede

modificar para mejorar propiedades dependiendo de la aplicación

deseada. Por estos motivos, existen numerosas publicaciones centradas en

el estudio de materiales carbonosos dopados con heteroátomos como

Fibra de carbón activada

+

+

+

+

++

+

+

+

+

+

+

++

+ +

++

Carbón activado


57

electrodos de supercondensadores. La química superficial de los

materiales carbonosos afecta a diversas propiedades fisicoquímicas y

electroquímicas de los materiales, como son: mojabilidad, conductividad

eléctrica, estabilidad electroquímica y contribución a la capacidad por

medio de procesos pseudocapacitivos. No obstante, el efecto de los grupos

funcionales se debe analizar de forma detallada.

En el caso de los grupos funcionales oxigenados, se ha encontrado

una correlación positiva entre la capacidad y la cantidad de grupos que

desorben como CO [3]. Esto se debe principalmente a la contribución a la

pseudocapacidad de los grupos quinona [3,22,107,108] y al carácter

hidrofílico de los hidroxilos, que aumentan la mojabilidad del electrodo.

El efecto de la pseudocapacidad depende del pH del electrolito, debido a

que implica el intercambio de protones. En medio básico, la contribución

farádica de los grupos CO es significativamente menor, aunque se ha

detectado cierta contribución [109]. Además, también se han detectado

procesos pseudocapacitivos en medio orgánico [110].

Los grupos que desorben como CO2, en cambio, son perjudiciales

para los materiales carbonosos, debido a su carácter electrón-aceptor, que

produce una menor deslocalización de la carga y, por tanto, una

disminución de la conductividad eléctrica [107,111]. Por último, se ha

observado que los grupos funcionales oxigenados disminuyen la

estabilidad electroquímica de los materiales carbonosos en electrolitos

acuosos [112] y orgánicos [89].

Capítulo 1

58

En cambio, la naturaleza de la influencia de los grupos nitrogenados

requiere un mayor esfuerzo investigador, debido principalmente a su

coexistencia inevitable con grupos oxigenados. Se ha observado que los

grupos funcionales nitrogenados pueden favorecer distintas propiedades

fisicoquímicas que mejoran su comportamiento como electrodos en

supercondensadores. Concretamente, se han detectado contribuciones

pseudocapacitivas por medio de grupos piridina [113], piridona [114] y

aminas (en medio orgánico) [48]. Además, el nitrógeno cuaternario

permite aumentar la mojabilidad del electrodo [113,115,116]. Finalmente,

se ha observado que el nitrógeno cuaternario aumenta la estabilidad

electroquímica de los materiales cuando se utilizan como electrodos de

supercondensadores en medio orgánico, mientras que los grupos amina

tienen un efecto perjudicial [48].

1.4.2. Pilas de combustible

Las pilas de combustible son dispositivos de generación de energía

eléctrica [117]. A diferencia de las baterías, en estos dispositivos los

reactivos se alimentan de forma continua durante el funcionamiento de la

pila. Estos dispositivos electroquímicos están compuestos por dos semi-

celdas (cátodo y ánodo) separadas por una membrana intercambio iónico.

Estos dispositivos permiten generar elevadas densidades energéticas

(ver diagrama de Ragone en Figura 1.6) de forma eficiente, lo que permite

su empleo en aplicaciones como: sistemas de emergencia, sistemas

estacionarios, dispositivos portátiles y locomoción. En comparación con


59

tecnologías energéticas convencionales, estos dispositivos proporcionan

mayor eficiencia y menores emisiones de productos contaminantes.

Entre todos los combustibles disponibles para estos dispositivos, el

hidrógeno es el más deseado debido a que el único producto de reacción

es agua. No obstante, se han estudiado una amplia variedad de

combustibles, como diversos alcoholes e hidrocarburos, entre los que

destaca el metanol y el etanol. En estos dispositivos, el combustible

(hidrógeno, metanol, etc) se alimenta al ánodo para su oxidación, mientras

que en el cátodo tiene lugar la reacción de reducción de oxigeno (ORR,

del inglés oxygen reduction reaction).

Existen diversos tipos de pilas de combustible. Se clasifican

principalmente atendiendo al electrolito empleado, que determina el tipo

de combustible, temperatura de operación y catalizador utilizado. La

Figura 1.6 resume el funcionamiento de los distintos tipos de pilas de

combustible.

En el desarrollo de esta tecnología, los materiales carbonosos juegan

un papel esencial. Estos se encuentran presentes en distintos componentes

de las pilas de combustible de baja temperatura, debido fundamentalmente

a su elevada conductividad eléctrica, estabilidad en condiciones de

operación de la pila y bajo coste. Concretamente, se utilizan como

material en placas bipolares, componente de los difusores de gases y

soporte del catalizador presente en el cátodo. En los últimos años, además,

han despertado un gran interés como alternativa a los catalizadores

basados en platino u otros metales empleados en la ORR, ya que presentan

Capítulo 1

60

actividad catalítica en ausencia de platino u otros metales nobles en su

composición.

Figura 1.13. Esquema de los tipos de pilas de combustible [118].

1.4.2.1. Reacción de reducción de oxígeno

La reacción de reducción del oxígeno se produce en el cátodo de la

pila de combustible. El potencial de equilibrio teórico para la ORR es

1.229 V (vs NHE) en condiciones estándar y en medio ácido. Sin

embargo, la reacción sucede a sobrepotenciales elevados, disminuyendo

su eficiencia para su aplicación práctica. Según el medio, la reacción

puede darse por medio de dos mecanismos [119]:

ELECTROLITO

CÁTODOÁNODOAire, O2Combustible

Carga

Gases sin reaccionar

(O2, N2)

Gases sin reaccionar

(H2, CO, etc.)

PEMFC (60-90 ºC)

DMFC (60-90 ºC)

PAFC (180-220 ºC)

MCFC (550-650 ºC)

SOFC (800-1000 ºC)

AFC (60-90 ºC)


61

Medio ácido Medio básico

𝑂2 + 4𝐻+ + 4𝑒− → 2𝐻2𝑂 𝑂2 + 2𝐻2𝑂 + 4𝑒− → 4𝑂𝐻− Ec. I

𝑂2 + 2𝐻+ + 2𝑒− → 𝐻2𝑂2 𝑂2 + 𝐻2𝑂 + 2𝑒− → 𝐻𝑂2− + 𝑂𝐻− Ec. II

𝐻2𝑂2 + 2𝐻+ + 2𝑒− → 2𝐻2𝑂 𝐻𝑂2− + 𝐻2𝑂 + 2𝑒− → 3𝑂𝐻− Ec. III

2𝐻2𝑂2 → 2𝐻2𝑂 + 𝑂2 2𝐻𝑂2− → 2𝑂𝐻− + 𝑂2 Ec. IV

En cada electrolito, la reacción se puede producir por un

mecanismo de 2 (ecuaciones II y III) o 4 electrones (ecuación I). La

reacción por medio de 4 electrones es preferible en pilas de combustible

debido a que permite una producción mayor de energía y, además, genera

exclusivamente agua como producto de reacción. En cambio, la reacción

vía de 2 electrones produce la mitad de energía y da lugar a la formación

de productos nocivos. El H2O2 producido por este mecanismo puede

reducirse de nuevo a H2O, o bien, puede producir agua (ecuación III) y

oxígeno debido a una reacción secundaria de desproporción (ecuación

IV). En este último caso, se regeneraría parte del O2 utilizado; parte del

mismo puede ser recirculado de nuevo en el sistema; sin embargo, es

inevitable que se pierda parte del oxígeno producido. Estas reacciones

suceden en paralelo dependiendo de las propiedades del catalizador

utilizado [119,120].

Capítulo 1

62

1.4.2.2. Electrocatalizadores para la ORR

El electrocatalizador más empleado en la ORR es el platino soportado

en materiales carbonosos [121–123]. Este material presenta un

sobrepotencial bajo para esta reacción y una elevada selectividad a la

formación de agua. No obstante, presenta una serie de inconvenientes,

como son: baja estabilidad en las condiciones de operación de la pila de

combustible, sensibilidad al envenenamiento por CO (por migración del

ánodo al cátodo) y elevado coste [121]. En la bibliografía, se pueden

encontrar diferentes estudios para aumentar la eficacia de los catalizadores

basados en platino. Las principales estrategias se centran en: producción

de monocapas de platino sobre otros metales de menor coste (paladio,

hierro, etc.), aleaciones de platino con otros metales (rutenio, paladio,

iridio, plata, etc.), desarrollo de catalizadores basados en nanopartículas

de platino con morfología y tamaño controlados [121,124].

Sin embargo, se ha planteado también el reemplazo del platino como

electrocatalizador de la ORR. En el estudio de catalizadores alternativos

para la ORR, los materiales carbonosos juegan un papel fundamental tanto

como soporte [125–129] como catalizadores en sí mismos (sin presencia

de metal) [130]. Como soporte de catalizadores, se han empleado diversos

materiales (negro de carbón, nanotubos de carbono, nanofibras de

carbono, grafeno, etc.) para soportar tanto platino como otros metales (Ag,

Au, Co, Fe, etc.) [131–134]. La presencia de heteroátomos ha resultado

ser beneficiosa para aumentar la actividad catalítica [125]. Como

electrocatalizadores, numerosos estudios han señalado una elevada

actividad catalítica asociada a la presencia de heteroátomos en superficie


63

(azufre, boro, nitrógeno, etc.) [37]. Concretamente, los materiales

carbonosos dopados con nitrógeno han mostrado excelentes propiedades

catalíticas en la ORR [130]. Además, la estructura de los materiales

carbonosos influye de forma determinante [130,135,136]. Se ha señalado

que la microporosidad puede producir un aumento del potencial de inicio

de reacción [130,137,138].

Dado que esta Tesis Doctoral se centra en el estudio de los grupos

funcionales nitrogenados en las propiedades electroquímicas de los

materiales carbonosos, en la siguiente sección se describirán las

propiedades de los materiales carbonosos funcionalizados con nitrógeno

como electrocatalizadores de la ORR.

1.4.2.3. Materiales carbonosos funcionalizados con nitrógeno como

electrocatalizadores de la ORR

Los materiales carbonosos dopados con nitrógeno son muy

prometedores para sustituir al platino como catalizador de la ORR [139].

La presencia de grupos funcionales nitrogenados se ha relacionado con

una mayor electroactividad. No obstante, la influencia de los distintos

grupos funcionales nitrogenados no ha sido esclarecida.

Algunos autores señalan los sitios N-C-N como activos para la ORR.

Indican, además, que la presencia de nitrógeno produce cambios

inherentes en la geometría que se traducen en un aumento de

electroactividad [140]. Otros estudios proponen que el carácter electrón-

dador de los grupos nitrogenados producen una redistribución de la

densidad electrónica alrededor de los átomos de nitrógeno conduciendo a

Capítulo 1

64

un aumento de la actividad catalítica [139,141,142]. De esta forma, el

aumento de densidad de carga positiva en los átomos de carbono

adyacentes al nitrógeno favorece la quimisorción del oxígeno y debilita el

enlace O-O [139]. Por último, se ha señalado que, además del papel de los

grupos nitrogenados como sitios activos, la presencia de nitrógeno

produce cambios estructurales que contribuyen de forma adicional a un

aumento de la electroactividad.

Dada la dificultad de producir e identificar grupos nitrogenados

específicos, resulta difícil esclarecer qué grupos funcionales son

responsables de la mejora de la respuesta catalítica [130]. Este aspecto es

clave para avanzar hacia el diseño de electrocatalizadores óptimos. Por

estos motivos, se está realizando un gran esfuerzo investigador,

empleando estrategias tanto experimentales como teóricas, para elucidar

los grupos funcionales responsables de la mejora de la actividad

electrocatalítica. Diversos estudios señalan que los materiales carbonosos

con elevada concentración de piridinas presentan mayor actividad

catalítica [143–145] Sin embargo, otros estudios señalan que la mejora de

la actividad catalítica viene dada por la presencia de varios grupos

funcionales: pirroles, nitrógeno cuaternario [146] y oxidados

[147].Recientemente, se han señalado los sitios N-C-O y los grupos N-Q

en posición zigzag como electroactivos para la ORR [148,149]. Además,

el aumento de la actividad electrocatalítica debido a los grupos N-Q

también ha sido señalado mediante estudios teóricos por Ikeda y col [150].

Estos autores también descartaron la contribución de los grupos

piridínicos a la actividad catalítica.


65

En cuando a la selectividad de la reacción, algunos estudios indican

que las piridinas presentan una elevada selectividad a la producción de

agua, mientras que el nitrógeno cuaternario es más selectivo a la

formación de peróxido de hidrógeno. Sin embargo, un estudio reciente

señala un aumento de la selectividad a la formación de agua asociado a la

presencia de nitrógeno cuaternario en posición zigzag en el borde de las

láminas de grafeno [149]. También se ha indicado un aumento de

selectividad asociado a la presencia sitios N-C-O (piridonas) [148].

1.5. Referencias

[1] T.J. Bandosz, Surface chemistry of carbon materials, in: Carbon

Mater. Catal., John Wiley & Sons, Inc, 2009: pp. 45–92.

[2] P. Serp, J.L. Figueiredo, Carbon Materials for Catalysis, John

Wiley & Sons, Inc., Hoboken, New Jersey, USA, 2008.

[3] M.J. Bleda-Martínez, J.A. Maciá-Agulló, D. Lozano-Castelló, E.

Morallón, D. Cazorla-Amorós, A. Linares-Solano, Role of surface

chemistry on electric double layer capacitance of carbon materials,

Carbon 43 (2005) 2677–2684.

[4] C. Yin, M. Aroua, W. Daud, Review of modifications of activated

carbon for enhancing contaminant uptakes from aqueous solutions,

Sep. Purif. Technol. 52 (2007) 403–415.

[5] M.S. Shafeeyan, W.M.A.W. Daud, A. Houshmand, A. Shamiri, A

review on surface modification of activated carbon for carbon

dioxide adsorption, J. Anal. Appl. Pyrolysis. 89 (2010) 143–151.

[6] T.J. Bandosz, C.O. Ania, Surface chemistry of activated carbons

and its characterization, in: T.J. Bandosz (Ed.), Act. Carbon

Surfaces Environ. Remediat., 1st ed., Elsevier, 2006: pp. 159–229.

[7] F. Rodriguez-Reinoso, Activated carbon: structure,

characterization, preparation and applications, in: H. Marsh, E.A.

Capítulo 1

66

Heinz, F. Rodriguez-Reinoso (Eds.), Introd. to Carbon Technol.,

Universidad de Alicante, 1997: pp. 35–102.

[8] B.E. Conway, Electrochemical supercapacitors: Scientific

Fundamentals and Technological Applications, Springer, New

York, 1999.

[9] F. Béguin, E. Frackowiak, eds., Carbons for Electrochemical

Energy Storage and Conversion Systems, 1st ed., CRC Press, 2009.

[10] O. Ioannidou, A. Zabaniotou, Agricultural residues as precursors

for activated carbon production-A review, Renew. Sustain. Energy

Rev. 11 (2007) 1966–2005.

[11] J.A. Maciá-Agulló, B.C. Moore, D. Cazorla-Amorós, A. Linares-

Solano, Activation of coal tar pitch carbon fibres: Physical

activation vs. chemical activation, Carbon 42 (2004) 1367–1370.

[12] A. Linares-Solano, D. Lozano-Castelló, M.A. Lillo-Ródenas, D.

Cazorla-Amorós, Carbon activation by alkaline hydroxides

preparation and reactions, porosity and performance, Chem. Phys.

Carbon 30 (2008) 1–62.

[13] A. Ahmadpour, D.D. Do, The preparation of active carbons from

coal by chemical and physical activation, Carbon 34 (1996) 471–

479.

[14] D. Lozano-Castelló, M.A. Lillo-Ródenas, D. Cazorla-Amorós, A.

Linares-Solano, Preparation of activated carbons from Spanish

anthracite: I. Activation by KOH, Carbon 39 (2001) 741–749.

[15] F. Caturla, M. Molina-Sabio, F. Rodríguez-Reinoso, Preparation of

activated carbon by chemical activation with ZnCl2, Carbon 29

(1991) 999–1007.

[16] M. Molina-Sabio, F. RodRíguez-Reinoso, F. Caturla, M.J. Sellés,

Porosity in granular carbons activated with phosphoric acid,

Carbon 33 (1995) 1105–1113.


67

[17] M. Lillo-Ródenas, D. Lozano-Castelló, D. Cazorla-Amorós, A.


anthracite: II. Activation by NaOH, Carbon. 39 (2001) 751–759.

[18] H. Nishihara, T. Kyotani, Templated nanocarbons for energy

storage, Adv. Mater. 24 (2012) 4473–4498.

[19] H. Itoi, H. Nishihara, T. Kogure, T. Kyotani, Three-Dimensionally

Arrayed and Mutually Connected 1.2-nm Nanopores for High-

Performance Electric Double Layer Capacitor, J. Am. Chem. Soc.

20 (2011) 1165–1167.

[20] H. Nishihara, T. Kyotani, Zeolite-templated carbons - three-

dimensional microporous graphene frameworks, Chem. Commun.

54 (2018) 5648–5673.

[21] H. Nishihara, Q. Yang, P. Hou, M. Unno, S. Yamauchi, R. Saito,

J.I. Paredes, A. Martínez-Alonso, J.M.D. Tascón, Y. Sato, M.

Terauchi, T. Kyotani, A possible buckybowl-like structure of

zeolite templated carbon, Carbon 47 (2009) 1220–1230.

[22] H. Itoi, H. Nishihara, T. Ishii, K. Nueangnoraj, R. Berenguer-

Betrián, T. Kyotani, Large Pseudocapacitance in Quinone-

Functionalized Zeolite-Templated Carbon, Bull. Chem. Soc. Jpn.

87 (2014) 250–257.

[23] C. González-Gaitán, R. Ruiz-Rosas, H. Nishihara, T. Kyotani, E.

Morallón, D. Cazorla-Amorós, Successful functionalization of

superporous zeolite templated carbon using aminobenzene acids

and electrochemical methods, Carbon 99 (2016) 157–166.

[24] H. Nishihara, P. Hou, L. Li, M. Ito, M. Uchiyama, T. Kaburagi, A.

Ikura, J. Katamura, T. Kawarada, K. Mizuuchi, T. Kyotani, High-

Pressure Hydrogen Storage in Zeolite-Templated Carbon, J. Phys.

Chem. C 113 (2009) 3189–3196.

[25] H. Itoi, H. Nishihara, T. Kyotani, Effect of Heteroatoms in Ordered

Microporous Carbons on Their Electrochemical Capacitance,

Langmuir 32 (2016) 11997–12004.

Capítulo 1

68

[26] M.A. Montes-Morán, D. Suárez, J.A. Menéndez, E. Fuente, The

Basicity of Carbons, in: Nov. Carbon Adsorbents, Elsevier Ltd,

2012: pp. 173–203.

[27] C.A. Leon y Leon, J.M. Solar, V. Calemma, L.R. Radovic,

Evidence for the protonation of basal plane sites on carbon, Carbon

30 (1992) 797–811.

[28] W. Shen, W. Fan, Nitrogen-containing porous carbons: synthesis

and application, J. Mater. Chem. A 1 (2013) 999–1013.

[29] M. Seredych, T.J. Bandosz, Role of Microporosity and Nitrogen

Functionality on the Surface of Activated Carbon in the Process of

Desulfurization of Digester Gas, J. Phys. Chem. C 112 (2008)

4704–4711.

[30] A. Bagreev, J. Angel Menendez, I. Dukhno, Y. Tarasenko, T.J.

Bandosz, Bituminous coal-based activated carbons modified with

nitrogen as adsorbents of hydrogen sulfide, Carbon 42 (2004) 469–

476.

[31] E. Raymundo-Piñero, D. Cazorla-Amorós, A. Linares-Solano, The

role of different nitrogen functional groups on the removal of SO2

from flue gases by N-doped activated carbon powders and fibres,

Carbon 41 (2003) 1925–1932.

[32] J. Yu, M. Guo, F. Muhammad, A. Wang, F. Zhang, Q. Li, G. Zhu,

One-pot synthesis of highly ordered nitrogen-containing

mesoporous carbon with resorcinol–urea–formaldehyde resin for

CO2 capture, Carbon 69 (2014) 502–514.

[33] A. Houshmand, M.S. Shafeeyan, A. Arami-Niya, W.M.A.W.

Daud, Anchoring a halogenated amine on the surface of a

microporous activated carbon for carbon dioxide capture, J. Taiwan

Inst. Chem. Eng. 44 (2013) 774–779.

[34] A. Houshmand, W.M.A.W. Daud, M.-G. Lee, M.S. Shafeeyan,

Carbon Dioxide Capture with Amine-Grafted Activated Carbon,

Water, Air, Soil Pollut. 223 (2011) 827–835.


69

[35] M.R. Benzigar, S.N. Talapaneni, S. Joseph, K. Ramadass, G.

Singh, J. Scaranto, U. Ravon, K. Al-Bahily, A. Vinu, Recent

advances in functionalized micro and mesoporous carbon

materials: Synthesis and applications, Chem. Soc. Rev. 47 (2018)

2680–2721.

[36] Y. Deng, Y. Xie, K. Zou, X. Ji, Review on recent advances in

nitrogen-doped carbons: Preparations and applications in

supercapacitors, J. Mater. Chem. A 4 (2015) 1144–1173.

[37] J.P. Paraknowitsch, A. Thomas, Doping carbons beyond nitrogen:

an overview of advanced heteroatom doped carbons with boron,

sulphur and phosphorus for energy applications, Energy Environ.

Sci. 6 (2013) 2839–2855.

[38] D. Mang, H.P. Boehm, K. Stanczyk, H. Marsh, Inhibiting effect of

incorporated nitrogen on the oxidation of microcrystalline carbons,

Carbon 30 (1992) 391–398.

[39] M. Zhou, F. Pu, Z. Wang, S. Guan, Nitrogen-doped porous carbons

through KOH activation with superior performance in

supercapacitors, Carbon 68 (2014) 185–194.

[40] C.O. Ania, V. Khomenko, E. Raymundo-Piñero, J.B. Parra, F.

Béguin, The Large Electrochemical Capacitance of Microporous

Doped Carbon Obtained by Using a Zeolite Template, Adv. Funct.

Mater. 17 (2007) 1828–1836.

[41] E. Raymundo-Piñero, D. Cazorla-Amorós, A. Linares-Solano, J.

Find, U. Wild, R. Schlögl, Structural characterization of N-

containing activated carbon fibers prepared from a low softening

point petroleum pitch and a melamine resin, Carbon 40 (2002) 597–

608.

[42] R.J.J. Jansen, H. van Bekkum, Amination and ammoxidation of

activated carbons, Carbon 32 (1994) 1507–1516.

[43] S. Biniak, G. Szymański, J. Siedlewski, A. ŚwiaTkowski, The

characterization of activated carbons with oxygen and nitrogen

surface groups, Carbon 35 (1997) 1799–1810.

Capítulo 1

70

[44] F. Xie, J. Phillips, I.F. Silva, M.C. Palma, J.A. Menéndez,

Microcalorimetric study of acid sites on ammonia- and acid-

pretreated activated carbon, Carbon 38 (2000) 691–700.

[45] M. Seredych, D. Hulicova-Jurcakova, G.Q. Lu, T.J. Bandosz,

Surface functional groups of carbons and the effects of their

chemical character, density and accessibility to ions on

electrochemical performance, Carbon 46 (2008) 1475–1488.

[46] Y. El-Sayed, T.J. Bandosz, Role of surface oxygen groups in

incorporation of nitrogen to activated carbons via

ethylmethylamine adsorption., Langmuir 21 (2005) 1282–9.

[47] A. Gabe, M.J. Mostazo-López, D. Salinas-Torres, E. Morallón, D.

Cazorla-Amorós, Synthesis of conducting polymer/carbon material

composites and their application in electrical energy storage, in:

Hybrid Polym. Compos. Mater. Process., 2017: pp. 173–209.

[48] D. Salinas-Torres, S. Shiraishi, E. Morallón, D. Cazorla-Amorós,

Improvement of carbon materials performance by nitrogen

functional groups in electrochemical capacitors in organic

electrolyte at severe conditions, Carbon 82 (2015) 205–213.

[49] Ana Cristina Ramírez-Pérez, Javier Quílez-Bermejo, Juan Manuel

Sieben, Emilia Morallón, Diego Cazorla-Amorós, Effect of

Nitrogen-Functional Groups on the ORR Activity of Activated

Carbon Fiber-Polypyrrole-Based Electrodes, Electrocatalysis 9

(2018) 697–705.

[50] D. Salinas-Torres, J.M. Sieben, D. Lozano-Castello, E. Morallón,

M. Burghammer, C. Riekel, D. Cazorla-Amorós, Characterization

of activated carbon fiber/polyaniline materials by position-resolved

microbeam small-angle X-ray scattering, Carbon 50 (2012) 1051–

1056.

[51] J. Zhu, J. Yang, B. Deng, Ethylenediamine-modified activated

carbon for aqueous lead adsorption, Environ. Chem. Lett. 8 (2009)

277–

[52] A.R. Silva, M. Martins, M.M. Freitas, A. Valente, C. Freire, B. de

Castro, J.L. Figueiredo, Immobilisation of amine-functionalised


71

nickel(II) Schiff base complexes onto activated carbon treated with

thionyl chloride, Microporous Mesoporous Mater 55 (2002) 275–

284.

[53] A.R. Silva, V. Budarin, J.H. Clark, C. Freire, B. de Castro, Organo-

functionalized activated carbons as supports for the covalent

attachment of a chiral manganese(III) salen complex, Carbon 45

(2007) 1951–1964.

[54] J.A.C. Alves, C. Freire, B. de Castro, J.L. Figueiredo, Anchoring

of organic molecules onto activated carbon, Colloids Surfaces A

Physicochem. Eng. Asp. 189 (2001) 75–84.

[55] H. Tamai, K. Shiraki, T. Shiono, H. Yasuda, Surface

functionalization of mesoporous and microporous activated

carbons by immobilization of diamine., J. Colloid Interface Sci.

295 (2006) 299–302.

[56] M. Abe, K. Kawashima, K. Kozawa, H. Sakai, K. Kaneko,

Amination of Activated Carbon and Adsorption Characteristics of

Its Aminated Surface, Langmuir 16 (2000) 5059–5063.

[57] C.-T. Hsieh, H. Teng, W.-Y. Chen, Y.-S. Cheng, Synthesis,

characterization, and electrochemical capacitance of amino-

functionalized carbon nanotube/carbon paper electrodes, Carbon

48 (2010) 4219–4229.

[58] A. Gromov, S. Dittmer, J. Svensson, O.A. Nerushev, S.A. Perez-

García, L. Licea-Jiménez, R. Rychwalski, E.E.B. Campbell,

Covalent amino-functionalisation of single-wall carbon nanotubes,

J. Mater. Chem. 15 (2005) 3334–3339.

[59] M.B. Smith, J. March, Rearrangements, in: March’s Adv. Org.

Chem. React. Mech. Estructure, 6th ed., John Wiley & Sons, Inc.,

Hoboken, New Jersey, USA, 2007: pp. 1607–1608.

[60] C. González-Gaitán, R. Ruiz-Rosas, E. Morallón, D. Cazorla-

Amorós, Electrochemical Methods to Functionalize Carbon

Materials, in: Chem. Funct. Carbon Nanomater., CRC Press, 2015:

pp. 230–261.

Capítulo 1

72

[61] D. Bélanger, J. Pinson, Electrografting: a powerful method for

surface modification, Chem. Soc. Rev. 40 (2011) 3995.

[62] M. Toupin, D. Bélanger, Spontaneous functionalization of carbon

black by reaction with 4-nitrophenyldiazonium cations, Langmuir

24 (2008) 1910–1917.

[63] T. Breton, D. Bélanger, Modification of carbon electrode with aryl

groups having an aliphatic amine by electrochemical reduction of

in situ generated diazonium cations, Langmuir 24 (2008) 8711–

8718.

[64] R.S. Deinhammer, M. Ho, J.W. Anderegg, M.D. Porter,

Electrochemical Oxidation of Amine-Containing Compounds: A

Route to the Surface Modification of Glassy Carbon Electrodes,

Langmuir 10 (1994) 1306–1313.

[65] X. Li, Y. Wan, C. Sun, Covalent modification of a glassy carbon

surface by electrochemical oxidation of r-aminobenzene sulfonic

acid in aqueous solution, J. Electroanal. Chem. 569 (2004) 79–87.

[66] E. Frackowiak, Q. Abbas, F. Béguin, Carbon/carbon

supercapacitors, J. Energy Chem. 22 (2013) 226–240.

[67] B.E. Conway, Transition from “supercapacitor” to “battery”

behavior in electrochemical energy storage, J. Electrochem. Soc.

138 (1991) 1539–1548.

[68] P. Simon, Y. Gogotsi, Materials for electrochemical capacitors.,

Nat. Mater. 7 (2008) 845–54.

[69] R. Kötz, M. Carlen, Principles and applications of electrochemical

capacitors, Electrochim. Acta 45 (2000) 2483–2498.

[70] E. Frackowiak, F. Béguin, Carbon materials for the electrochemical

storage of energy in capacitors, Carbon 39 (2001) 937–950.

[71] A.J. Bard, L.R. Faulkner, Electrochemical Methods. Fundamentals

and Applications, 2nd ed., John Wiley & Sons, Inc, New York,

2001.


73

[72] Aiping Yu, V. Chabot, J. Zhang, Electrochemical supercapacitors

for energy storage and delivery. Fundamentals and applications,

CRC Press, Taylor & Francis Group, Boca Raton, 2013.

[73] S. Shiraishi, Electrochemical Performance, in: M. Inagaki (Ed.),

Mater. Sci. Eng. Carbon Charact., Tsinghua University Press

Limited. Elsevier Inc, 2016: pp. 205–226.

[74] Q.-L. Chen, K.-H. Xue, W. Shen, F.-F. Tao, S.-Y. Yin, W. Xu,

Fabrication and electrochemical properties of carbon nanotube

array electrode for supercapacitors, Electrochim. Acta 49 (2004)

4157–4161.

[75] H. Jiang, P.S. Lee, C. Li, 3D carbon based nanostructures for

advanced supercapacitors, Energy Environ. Sci. 6 (2013) 41–53.

[76] S. Leyva-García, D. Lozano-Castelló, E. Morallón, D. Cazorla-

Amorós, Silica-templated ordered mesoporous carbon thin films as

electrodes for micro-capacitors, J. Mater. Chem. A 4 (2016) 4570–

4579.

[77] R. Berenguer, F.J. García-Mateos, R. Ruiz-Rosas, D. Cazorla-

Amorós, E. Morallón, J. Rodríguez-Mirasol, T. Cordero, Biomass-

derived binderless fibrous carbon electrodes for ultrafast energy

storage, Green Chem. 18 (2016).

[78] S. Shiraishi, Highly-durable carbon electrode for electrochemical

capacitors, Bol. Grupo Español Carbón 28 (2013) 18–24.

[79] G. Salitra, A. Soffer, L. Eliad, Y. Cohen, D. Aurbach, Carbon

Electrodes for Double-Layer Capacitors I. Relations Between Ion

and Pore Dimensions, J. Electrochem. Soc. 147 (2000) 2486.

[80] G.A. Snook, G.J. Wilson, A.G. Pandolfo, Mathematical functions

for optimisation of conducting polymer/activated carbon

asymmetric supercapacitors, J. Power Sources 186 (2009) 216–

223.

[81] C. Peng, S. Zhang, X. Zhou, G.Z. Chen, Unequalisation of

electrode capacitances for enhanced energy capacity in

asymmetrical supercapacitors, Energy Environ. Sci. 3 (2010) 1499.

Capítulo 1

74

[82] C. Zhong, Y. Deng, W. Hu, J. Qiao, L. Zhang, J. Zhang, A review

of electrolyte materials and compositions for electrochemical

supercapacitors, Chem. Soc. Rev. 44 (2015) 7484–7539.

[83] F. Béguin, V. Presser, A. Balducci, E. Frackowiak, Carbons and

electrolytes for advanced supercapacitors, Adv. Mater. 26 (2014).

[84] M.J. Bleda-Martinez, D. Lozano-Castelló, D. Cazorla-Amorós, E.

Morallón, Kinetics of Double-Layer Formation: Influence of

Porous Structure and Pore Size Distribution, Energy & Fuels 24

(2010) 3378–3384.

[85] V. Khomenko, E. Raymundo-Piñero, E. Frackowiak, F. Béguin,

High-voltage asymmetric supercapacitors operating in aqueous

electrolyte, Appl. Phys. A Mater. Sci. Process. 82 (2006) 567–573.

[86] K. Fic, G. Lota, M. Meller, E. Frackowiak, Novel insight into

neutral medium as electrolyte for high-voltage supercapacitors,

Energy Environ. Sci. 5 (2012) 5842–5850.

[87] Q. Gao, L. Demarconnay, E. Raymundo-Piñero, F. Béguin,

Exploring the large voltage range of carbon/carbon supercapacitors

in aqueous lithium sulfate electrolyte, Energy Environ. Sci. 5

(2012) 9611–9617.

[88] D. Lozano-Castelló, D. Cazorla-Amorós, A. Linares-Solano, S.

Shiraishi, H. Kurihara, A. Oya, Influence of pore structure and

surface chemistry on electric double layer capacitance in non-

aqueous electrolyte, Carbon 41 (2003) 1765–1775.

[89] D. Cazorla-Amorós, D. Lozano-Castelló, E. Morallón, M.J. Bleda-

Martínez, A. Linares-Solano, S. Shiraishi, Measuring cycle

efficiency and capacitance of chemically activated carbons in

propylene carbonate, Carbon 48 (2010) 1451–1456.

[90] P.W. Ruch, D. Cericola, A. Foelske-Schmitz, R. Kötz, A. Wokaun,

Aging of electrochemical double layer capacitors with acetonitrile-

based electrolyte at elevated voltages, Electrochim. Acta 55 (2010)

4412–4420.


75

[91] A. Brandt, P. Isken, A. Lex-Balducci, A. Balducci, Adiponitrile-

based electrochemical double layer capacitor, J. Power Sources 204

(2012) 213–219.

[92] A. Balducci, Electrolytes for high voltage electrochemical double

layer capacitors: A perspective article, J. Power Sources 326 (2016)

534–540.

[93] J. Krummacher, C. Schütter, L.H. Hess, A. Balducci, Non-aqueous

electrolytes for electrochemical capacitors, Curr. Opin.

Electrochem. 9 (2018) 64–69.

[94] M. Galiński, A. Lewandowski, I. Stepniak, Ionic liquids as

electrolytes, Electrochim. Acta 51 (2006) 5567–5580.

[95] R.D. Rogers, G.A. Voth, Ionic Liquids, Acc. Chem. Res. 40 (2007)

1077–1078.

[96] D.R. Macfarlane, N. Tachikawa, M. Forsyth, J.M. Pringle, P.C.

Howlett, G.D. Elliott, J.H. Davis, M. Watanabe, P. Simon, C.A.

Angell, Energy applications of ionic liquids, Energy Environ. Sci.

7 (2014) 232–250.

[97] A. Brandt, S. Pohlmann, A. Varzi, A. Balducci, S. Passerini, Ionic

liquids in supercapacitors, MRS Bull. 38 (2013) 554–559.

[98] S. Pohlmann, B. Lobato, T.A. Centeno, A. Balducci, The influence

of pore size and surface area of activated carbons on the

performance of ionic liquid based supercapacitors, Phys. Chem.

Chem. Phys. 15 (2013) 17287–17294.

[99] A. Krause, A. Balducci, High voltage electrochemical double layer

capacitor containing mixtures of ionic liquids and organic

carbonate as electrolytes, Electrochem. Commun. 13 (2011) 814–

817. doi:10.1016/j.elecom.2011.05.010.

[100] O. Barbieri, M. Hahn, A. Herzog, R. Kötz, Capacitance limits of

high surface area activated carbons for double layer capacitors,

Carbon 43 (2005) 1303–1310.

Capítulo 1

76

[101] E. Raymundo-Piñero, K. Kierzek, J. Machnikowski, F. Béguin,

Relationship between the nanoporous texture of activated carbons

and their capacitance properties in different electrolytes, Carbon 44

(2006) 2498–2507.

[102] C. Vix-guterl, E. Frackowiak, K. Jurewicz, M. Friebe, J.

Parmentier, F. Béguin, Electrochemical energy storage in ordered

porous carbon materials, Carbon 43 (2005) 1293–1302.

[103] L. Eliad, G. Salitra, A. Soffer, D. Aurbach, Ion Sieving Effects in

the Electrical Double Layer of Porous Carbon Electrodes:

Estimating Effective Ion Size in Electrolytic Solutions, J. Phys.

Chem. B 105 (2001) 6880–6887.

[104] D. Eliad, L., Pollak, E., Levy, N., Salitra, G., Soffer, A., Aurbach,

Assessing optimal pore-to-ion size relations in the design of porous

poly (vinylidene chloride) carbons for EDL capacitors, Appl. Phys.

A Mater. Sci. Process. 82 (2006) 607–613.

[105] C.O. Ania, J. Pernak, F. Stefaniak, F. Be, Solvent-free ionic liquids

as in situ probes for assessing the effect of ion size on the

performance of electrical double layer capacitors, Carbon 44

(2006) 3126–3130

[106] Y. Li, Z.-Y. Fu, B.-L. Su, Hierarchically Structured Porous

Materials for Energy Conversion and Storage, Adv. Funct. Mater.

22 (2012) 4634–4667.

[107] M.J. Bleda-Martínez, D. Lozano-Castelló, E. Morallón, D.

Cazorla-Amorós, A. Linares-Solano, Chemical and

electrochemical characterization of porous carbon materials,

Carbon 44 (2006) 2642–2651.

[108] B.E. Conway, V. Birss, J. Wojtowicz, The role and utilization of

pseudocapacitance for energy storage by supercapacitors, J. Power

Sources 66 (1997) 1–14.

[109] H.A. Andreas, B.E. Conway, Examination of the double-layer

capacitance of an high specific-area C-cloth electrode as titrated

from acidic to alkaline pHs, Electrochim. Acta 51 (2006) 6510–

6520.


77

[110] K. Nueangnoraj, H. Nishihara, T. Ishii, N. Yamamoto, H. Itoi, R.

Berenguer, R. Ruiz-Rosas, D. Cazorla-Amorós, E. Morallón, M.

Ito, T. Kyotani, Pseudocapacitance of zeolite-templated carbon in

organic electrolytes, Energy Storage Mater. 1 (2015) 35–41

[111] C.A. Leon y Leon, L.R. Radovic, Interfacial chemistry and

electrochemistry of carbon surfaces, in: P.A. Thrower (Ed.), Chem.

Phys. Carbon, Vol. 24, Marcel Dekker, Inc New York, 1994: pp.

213–310.

[112] S. Leyva-García, K. Nueangnoraj, D. Lozano-Castelló, H.

Nishihara, T. Kyotani, E. Morallón, D. Cazorla-Amorós,

Characterization of a zeolite-templated carbon by electrochemical

quartz crystal microbalance and in situ Raman spectroscopy,

Carbon 89 (2015) 63–73.

[113] O. Ornelas, J.M. Sieben, R. Ruiz-Rosas, E. Morallón, D. Cazorla-

Amorós, J. Geng, N. Soin, E. Siores, B.F.G. Johnson, On the origin

of the high capacitance of nitrogen-containing carbon nanotubes in

acidic and alkaline electrolytes, Chem. Commun. 50 (2014)

11343–11346.

[114] D. Hulicova-Jurcakova, M. Kodama, S. Shiraishi, H. Hatori, Z.H.

Zhu, G.Q. Lu, Nitrogen-Enriched Nonporous Carbon Electrodes

with Extraordinary Supercapacitance, Adv. Funct. Mater. 19

(2009) 1800–1809.

[115] S.L. Candelaria, B.B. Garcia, D. Liu, G. Cao, Nitrogen

modification of highly porous carbon for improved supercapacitor

performance, J. Mater. Chem. 22 (2012) 9884–9889.

[116] M. Kawaguchi, T. Yamanaka, Y. Hayashi, H. Oda, Preparation and

capacitive properties of a carbonaceous material containing

nitrogen, J. Electrochem. Soc. 157 (2010) A35–A40.

[117] O.Z. Sharaf, M.F. Orhan, An overview of fuel cell technology:

Fundamentals and applications, Renew. Sustain. Energy Rev. 32

(2014) 810–853.

[118] M. Winter, R.J. Brodd, What Are Batteries, Fuel Cells, and

Supercapacitors? Chem. Rev. 104 (2004) 4245–4270.

Capítulo 1

78

[119] J. Masa, W. Xia, M. Muhler, W. Schuhmann, On the Role of Metals

in Nitrogen-Doped Carbon Electrocatalysts for Oxygen Reduction,

Angew. Chemie - Int. Ed. 54 (2015) 10102–10120.

[120] H.-H. Yang, R.L. McCreery, Elucidation of the Mechanism of

Dioxygen Reduction on Metal-Free Carbon Electrodes, J.

Electrochem. Soc. 147 (2000) 3420.

[121] A. Morozan, B. Jousselme, S. Palacin, Low-platinum and

platinum-free catalysts for the oxygen reduction reaction at fuel cell

cathodes, Energy Environ. Sci. 4 (2011) 1238–1254.

[122] V.R. Stamenkovic, B.S. Mun, M. Arenz, K.J.J. Mayrhofer, C.A.

Lucas, G. Wang, P.N. Ross, N.M. Markovic, Trends in

electrocatalysis on extended and nanoscale Pt-bimetallic alloy

surfaces, Nat. Mater. 6 (2007) 241–247.

[123] G. Wu, P. Zelenay, Nanostructured nonprecious metal catalysts for

oxygen reduction reaction, Acc. Chem. Res. 46 (2013) 1878–1889.

[124] C. Wang, H. Daimon, T. Onodera, T. Koda, S. Sun, A general

approach to the size- and shape-controlled synthesis of platinum

nanoparticles and their catalytic reduction of oxygen, Angew.

Chemie - Int. Ed. 47 (2008) 3588–3591.

[125] C.W.B. Bezerra, L. Zhang, K. Lee, H. Liu, A.L.B. Marques, E.P.

Marques, H. Wang, J. Zhang, A review of Fe-N/C and Co-N/C

catalysts for the oxygen reduction reaction, Electrochim. Acta 53

(2008) 4937–4951.

[126] C.W.B. Bezerra, L. Zhang, H. Liu, K. Lee, A.L.B. Marques, E.P.

Marques, H. Wang, J. Zhang, A review of heat-treatment effects on

activity and stability of PEM fuel cell catalysts for oxygen

reduction reaction, J. Power Sources 173 (2007) 891–908.

[127] P. Trogadas, T.F. Fuller, P. Strasser, Carbon as catalyst and support

for electrochemical energy conversion, Carbon 75 (2014) 5–42.

[128] S. Shahgaldi, J. Hamelin, Improved carbon nanostructures as a

novel catalyst support in the cathode side of PEMFC: A critical

review, Carbon 94 (2015) 705–728.


79

[129] F. Zaragoza-Martín, D. Sopeña-Escario, E. Morallón, C.S.-M. de

Lecea, Pt/carbon nanofibers electrocatalysts for fuel cells, J. Power

Sources 171 (2007) 302–309.

[130] K.H. Wu, D.W. Wang, D.S. Su, I.R. Gentle, A Discussion on the

Activity Origin in Metal-Free Nitrogen-Doped Carbons for

Oxygen Reduction Reaction and their Mechanisms,

ChemSusChem. 8 (2015) 2772–2788

[131] L. Kuai, B. Geng, S. Wang, Y. Zhao, Y. Luo, H. Jiang, Silver and

Gold Icosahedra: One-Pot Water-Based Synthesis and Their

Superior Performance in the Electrocatalysis for Oxygen

Reduction Reactions in Alkaline Media, Chem. – A Eur. J. 17

(2011) 3482–3489

[132] H.W. Liang, W. Wei, Z.S. Wu, X. Feng, K. Müllen, Mesoporous

metal-nitrogen-doped carbon electrocatalysts for highly efficient

oxygen reduction reaction, J. Am. Chem. Soc. 135 (2013) 16002–

16005.

[133] A. Gabe, J. García-Aguilar, Á. Berenguer-Murcia, E. Morallón, D.

Cazorla-Amorós, Key factors improving oxygen reduction reaction

activity in cobalt nanoparticles modified carbon nanotubes, Appl.

Catal. B Environ. 217 (2017) 303–312

[134] Y. Liang, H. Wang, P. Diao, W. Chang, G. Hong, Y. Li, M. Gong,

L. Xie, J. Zhou, J. Wang, T.Z. Regier, F. Wei, H. Dai, Oxygen

Reduction Electrocatalyst Based on Strongly Coupled Cobalt

Oxide Nanocrystals and Carbon Nanotubes, J. Am. Chem. Soc. 134

(2012) 15849–15857

[135] J. Park, Y. Nabae, T. Hayakawa, M.A. Kakimoto, Highly selective

two-electron oxygen reduction catalyzed by mesoporous nitrogen-

doped carbon, ACS Catal. 4 (2014) 3749–3754.

[136] Z. Chen, D. Higgins, A. Yu, L. Zhang, J. Zhang, A review on non-

precious metal electrocatalysts for PEM fuel cells, Energy Environ.

Sci. 4 (2011) 3167–3192.

[137] A. Gabe, R. Ruiz-Rosas, C. González-Gaitán, E. Morallón, D.

Cazorla-Amorós, Modeling of oxygen reduction reaction in porous

Capítulo 1

80

carbon materials in alkaline medium. Effect of microporosity, J.

Power Sources (2019) 451–464

[138] J.Y. Choi, R.S. Hsu, Z. Chen, Highly active porous carbon-

supported nonprecious metal-N electrocatalyst for oxygen

reduction reaction in PEM fuel cells, J. Phys. Chem. C 114 (2010)

8048–8053.

[139] S. Maldonado, K.J. Stevenson, Influence of nitrogen doping on

oxygen reduction electrocatalysis at carbon nanofiber electrodes, J.

Phys. Chem. B 109 (2005) 4707–4716.

[140] K. Gong, F. Du, Z. Xia, M. Durstock, L. Dai, Nitrogen-Doped

Carbon Nanotube Arrays with High Electrocatalytic Activity for

Oxygen Reduction, Science 323 (2009) 760–764.

[141] V. Strelko, V. Kuts, P. Thrower, On the mechanism of possible

influence of heteroatoms of nitrogen, boron and phosphorus in a

carbon matrix on the catalytic activity of carbons in electron

transfer reactions, Carbon 38 (2000) 1499–1503

[142] Q. Wei, X. Tong, G. Zhang, J. Qiao, Q. Gong, S. Sun, Nitrogen-

Doped Carbon Nanotube and Graphene Materials for Oxygen

Reduction Reactions, Catalysts 5 (2015) 1574–1602.

[143] N.P. Subramanian, X. Li, V. Nallathambi, S.P. Kumaraguru, H.

Colon-Mercado, G. Wu, J.W. Lee, B.N. Popov, Nitrogen-modified

carbon-based catalysts for oxygen reduction reaction in polymer

electrolyte membrane fuel cells, J. Power Sources 188 (2009) 38–

44.

[144] J.D. Wiggins-Camacho, K.J. Stevenson, Mechanistic discussion of

the oxygen reduction reaction at nitrogen-doped carbon nanotubes,

J. Phys. Chem. C 115 (2011) 20002–20010.

[145] P. Zhang, F. Sun, Z. Xiang, Z. Shen, J. Yun, D. Cao, ZIF-derived

in situ nitrogen-doped porous carbons as efficient metal-free

electrocatalysts for oxygen reduction reaction, Energy Environ.

Sci. 7 (2014) 442–450


81

[146] W. Xia, J. Masa, M. Bron, W. Schuhmann, M. Muhler, Highly

active metal-free nitrogen-containing carbon catalysts for oxygen

reduction synthesized by thermal treatment of polypyridine-carbon

black mixtures, Electrochem. Commun. 13 (2011) 593–596.


Amorós, Functionalization of carbon nanotubes using

aminobenzene acids and electrochemical methods. Electroactivity

for the oxygen reduction reaction, Int. J. Hydrogen Energy 40

(2015) 11242–11253.

[148] J. Quilez-Bermejo, C. González-Gaitán, E. Morallón, D. Cazorla-

Amorós, Effect of carbonization conditions of polyaniline on its

catalytic activity towards ORR. Some insights about the nature of

the active sites, Carbon 119 (2017) 62–71.

[149] J. Quílez-Bermejo, E. Morallón, D. Cazorla-Amorós, Oxygen-

reduction catalysis of N-doped carbons prepared: Via heat

treatment of polyaniline at over 1100 °C, Chem. Commun. 54

(2018) 4441–4444.

[150] T. Ikeda, M. Boero, S.F. Huang, K. Terakura, M. Oshima, J.I.

Ozaki, S.F. Hang, K. Terakura, M. Oshima, J.I. Ozaki, Carbon

Alloy Catalysts: Active Sites for Oxygen Reduction Reaction, J.

Phys. Chem. C 112 (2008) 14706–14709.

Capítulo 2

Materiales, métodos y técnicas

experimentales

Materiales, métodos y técnicas experimentales

85

2.1. Introducción

En este capítulo, se presentan los materiales, métodos de síntesis y

técnicas experimentales utilizados en esta Tesis Doctoral. Además, se

describen de forma general los métodos de funcionalización de la

superficie utilizados para la preparación de nuevos materiales. No

obstante, las condiciones experimentales específicas se encuentran

detalladas en cada uno de los capítulos.

2.2 Materiales

2.2.1 Carbones activados.

En esta Tesis Doctoral, se ha empleado principalmente un carbón

activado sintetizado en los laboratorios del grupo de investigación

Materiales Carbonosos y Medio Ambiente (conocido como KUA) y un

carbón activado comercial (YP50F).

El carbón activado KUA se obtuvo mediante activación química de

una antracita española, cuyas condiciones de síntesis se estudiaron

previamente en el grupo de investigación Materiales Carbonosos y

Medio Ambiente [1]. De forma resumida, este método consiste en

mezclar el precursor carbonoso con hidróxido potásico para

posteriormente realizar una etapa de pirólisis a una temperatura

determinada. Finalmente, se realiza un lavado en medio ácido para

eliminar los restos de agente activante y productos inorgánicos

resultantes de la reacción. De este modo, se obtiene un material de

elevada microporosidad cuyas propiedades fisicoquímicas resultan de

Capítulo 2

86

especial interés en el campo del almacenamiento electroquímico de

energía.

Adicionalmente, se han realizado ensayos de funcionalización

química empleando un carbón activado comercial con aplicación en

supercondensadores (YP50F, Kuraray Chemical, Japón). Las

propiedades de este material se detallan en el anexo del capítulo 6.

2.2.2 Materiales carbonosos nanomoldeados.

Se han empleado dos materiales carbonosos nanomoldeados (ZTC y

N-ZTC), obtenidos a partir de distintos precursores [2-4]. La síntesis de

estos materiales se realizó en el laboratorio del profesor Kyotani en la

Universidad de Tohoku (Japón), como resultado de una colaboración.

Estos materiales se sintetizaron mediante depósito químico en fase vapor

(CVD, del inglés chemical vapor deposition) utilizando una zeolita Y

como plantilla. Se utilizó propileno como precursor del ZTC y

acetonitrilo en el caso del N-ZTC. La información detallada acerca de la

síntesis y caracterización de estos materiales se encuentra en el capítulo

5.

2.3. Modificación de la química superficial de los materiales

carbonosos.

Los materiales KUA, YP50F y ZTC se han utilizado como

materiales de partida para introducir grupos funcionales (principalmente,

nitrogenados) en su superficie. A continuación, se describen de forma

detallada los métodos empleados.


87

2.3.1 Funcionalización química mediante métodos basados en

reacciones orgánicas.

Se han realizado distintos tratamientos de funcionalización química

para introducir grupos funcionales nitrogenados en materiales

carbonosos. Principalmente, se han utilizado dos protocolos basados en

reacciones orgánicas.

El primero consiste en la introducción de grupos funcionales amida

por medio de tres etapas de reacción: (i) oxidación (química o

electroquímica); (ii) tratamiento con SOCl2 para producir cloruros de

ácido y (iii) reacción con un reactivo nitrogenado. Este tratamiento se

ha utilizado para introducir grupos nitrogenados sobre los materiales

KUA y ZTC. Además, este tratamiento también se ha realizado en los

carbones activados KUA e YP50F sin llevar a cabo los procesos previos

de oxidación química y tratamiento con SOCl2.

El segundo tratamiento de funcionalización permite convertir las

amidas generadas por medio del primer protocolo en grupos funcionales

aminas. Este tratamiento se ha realizado sobre el carbón activado KUA.

A continuación, se describe de forma detallada el procedimiento

experimental utilizado para modificar la química superficial de los

materiales.

Capítulo 2

88

2.3.1.1. Oxidación química

La oxidación química se ha realizado utilizando HNO3 como agente

oxidante. Este tratamiento ha sido adaptado a partir de estudios previos

realizados en nuestro grupo de investigación [5].

En un vaso de precipitados se pone en contacto 1g de material con

40 mL de disolución de HNO3 65 % durante 3 horas con agitación

magnética a temperatura ambiente. Posteriormente, el carbón activado se

filtra y se lava con agua destilada hasta que el pH del agua de lavado sea

constante. Por último, la muestra se seca a 100 ºC.

2.3.1.2. Introducción de grupos funcionales amida

Este proceso requiere el uso de dos etapas consecutivas: en primer

lugar, tratamiento para formar cloruros de ácido a partir de los ácidos

carboxílicos y, en segundo lugar, reacción con un reactivo nitrogenado

para anclar el nitrógeno por sustitución nucleofílica sobre el cloruro de

ácido.

Formación de cloruros de ácido

Este procedimiento experimental se ha adaptado a partir de la

referencia [6]. Las reacciones se llevaron a cabo dentro de un sistema

Schlenk. Se introduce 1g de material carbonoso (resultante tras la

oxidación) en 50 mL de tolueno en un matraz de fondo redondo. A

continuación, se añaden 5 mL de SOCl2. La mezcla se mantiene a reflujo

a 120 ºC durante 5 horas en atmósfera de Ar. Posteriormente, se filtra y


89

el polvo obtenido se lava con tolueno y se seca a vacío a 120 ºC durante

14 horas.

Reacción con un reactivo nitrogenado

Las reacciones de amidación y aminación se han adaptado de la

funcionalización propuesta por Gromov y colaboradores para nanotubos

de carbono [7].

1g de KUA-COCl se añade a una disolución 2M de NH4NO3 en

DMF (300 mL) en un matraz de fondo redondo. La mezcla se agita para

obtener una dispersión homogénea. A continuación, se añade lentamente

y con agitación continua 300 mL de piridina. La mezcla resultante se

agita a 70ºC durante 65 horas en atmósfera de Ar. La muestra obtenida

se filtra y se lava con agua y etanol.

Este tratamiento se ha realizado utilizando como material de partida

un material carbonoso prístino y ese mismo material modificado

mediante oxidación química y tratamiento con SOCl2.

2.3.1.3. Introducción de grupos funcionales amina

Se añaden 10 mL de Br2 a 650 mL de disolución 3% de NaOCH3

en CH3OH en un matraz de fondo redondo. A continuación, se añaden

500 mg de material carbonoso a la disolución y la dispersión resultante

se agita a 70ºC durante 5 horas. Posteriormente, se añaden 4 mL de Br2

adicionales y se mantiene en agitación durante 20 horas a 70ºC. El

producto se filtra y se lava con Na2CO3 saturado, agua y etanol. Para

continuar, el material carbonoso se somete a hidrólisis en 500 mL de

Capítulo 2

90

NaOH 0.1M con agitación magnética durante 24 horas. Finalmente, la

muestra se filtra, se lava con agua y se seca a 100ºC en una estufa.

2.3.2. Polimerización de anilina

Este método consiste en la generación de un polímero conductor

(polianilina) en la superficie del material carbonoso, de manera que se

obtiene un material compuesto carbón activado/polianilina. Para esto, se

realiza la síntesis química de polianilina a partir de anilina previamente

adsorbida en el material carbonoso (en este caso, el carbón activado

KUA) [8]. Este método consiste en la oxidación de anilina en medio

ácido en presencia de un oxidante [9]. El proceso se lleva a cabo en un

reactor (que contiene una disolución 1M de HCl a 0 ºC) en el que se

adiciona el material carbonoso en el que se ha realizado la adsorción de

anilina. Posteriormente, se añade el agente oxidante (NH4)2S2O8 disuelto

en 1M HCl (a 0ºC) con una relación monómero:oxidante 1:1. Una vez

obtenido el material compuesto (KUA/PANI), se lava con 1M HCl y se

trata con NH4OH con el fin de obtener la polianilina en su forma

desdopada. Previamente al proceso de polimerización, se estudió la

adsorción del monómero de anilina en el material carbonoso, con el fin

de determinar la cantidad de anilina adsorbida y, de este modo, conocer

la cantidad de oxidante que es necesario adicionar al medio de reacción,

así como obtener diferentes cantidades de anilina adsorbida.


91

2.3.3 Tratamientos térmicos

Los materiales carbonosos obtenidos en las secciones 2.3.1 y 2.3.2,

se han utilizado como materiales de partida para realizar tratamientos

térmicos que modifiquen los grupos funcionales en condiciones que

permitan preservar la textura porosa del material original. Estos

tratamientos se han llevado a cabo en un horno tubular en atmósfera

inerte, utilizando una velocidad de calentamiento lenta (5ºC/min), hasta

alcanzar la temperatura final (500, 600 y 800 ºC), que se mantiene

durante una hora. En estas condiciones, el tratamiento afecta

principalmente a la química superficial, de manera que produce la

descomposición de los grupos funcionales y su conversión en otras

especies de elevada estabilidad térmica. Los detalles específicos de cada

tratamiento se pueden consultar en los capítulos 6 y 8.

2.4 Técnicas de caracterización

2.4.1 Adsorción física de gases

La adsorción física de gases es una técnica muy utilizada para

caracterizar propiedades texturales de los sólidos porosos [10]. El

proceso de adsorción consiste en el fenómeno que experimenta un sólido

(adsorbente) cuando entra en contacto con un gas (adsortivo) en un

recipiente a volumen constante y a una temperatura dada. En estas

condiciones, las fuerzas de interacción (de tipo Van der Waals) entre el

adsorbente y el adsortivo producen la acumulación de las moléculas de

gas en la interfase, que da lugar a una variación de la presión del

Capítulo 2

92

recipiente. Una vez alcanzado el equilibrio, la fisisorción se completa y,

por tanto, la presión alcanza un valor constante.

La cantidad de gas adsorbido se puede calcular de forma

volumétrica (a partir de la variación de presión que experimenta el

recipiente) o de forma gravimétrica (por medio de la determinación del

aumento de masa del sólido). Este cálculo se lleva a cabo a distintas

presiones relativas (P/P0), de manera que la representación de la cantidad

de gas adsorbido en función de las distintas presiones relativas se conoce

como isoterma de adsorción. A partir de estos experimentos, se pueden

determinar parámetros de la textura porosa de los materiales, como el

volumen de microporos, el área superficial, distribución de tamaños de

poros, etc.

Se pueden utilizar distintos adsorbatos (N2, He, Ar, etc.) para

obtener isotermas de adsorción. El adsortivo más común es el N2 a 77K.

A esta temperatura, el gas se adsorbe en un intervalo amplio de

presiones relativas (de 10-8 a 1), lo que permite la adsorción en todo el

intervalo de porosidad. No obstante, este adsorbato no permite

caracterizar aquellos sólidos que presenten una microporosidad muy

estrecha (<0.7 nm) debido a problemas cinéticos derivados de problemas

difusionales del N2 a baja temperatura y el momento cuadrupolar de la

molécula. Por tanto, es necesario un tiempo de equilibrio

extremadamente largo. En estos casos, se utiliza la adsorción de CO2 a

273K, que permite la caracterización de la microporosidad más estrecha,

para completar el estudio de las propiedades texturales de este tipo de

materiales [10–12].


93

En este trabajo, las isotermas de adsorción se han realizado en un

equipo volumétrico (Autosorb-6-Quantachrome). El equipo consta,

además, de una unidad de desgasificación independiente. Las muestras

utilizadas en esta Tesis Doctoral se han desgasificado a 200 ºC durante 4

horas. El intervalo de presiones relativas utilizado para las isotermas de

N2 fue de 0 a 1 y, en el caso de las isotermas de CO2, de 0 a 0.03.

Figura 2.1. Equipos de adsorción física de gases Autosorb-6-Quantachrome.

El análisis de las isotermas de adsorción requiere la aplicación de

varias teorías. Las más comunes son la teoría Brunauer, Emmet y Teller

y el modelo de Dubinin-Radushkevich.

2.4.1.1 Teoría de Brunauer-Emmet-Teller

La teoría de Brunauer-Emmet-Teller (BET) es uno de los modelos

de fisisorción más empleados para la determinación de la superficie

específica de materiales porosos [13]. La teoría BET se basa en el

modelo cinético de adsorción de Langmuir. Esta teoría parte del

concepto de equilibrio de adsorción, en el que la velocidad de adsorción

Capítulo 2

94

y desorción del adsorbato son iguales. Además, la teoría considera los

siguientes supuestos: (i) sólo se puede adsorber una molécula en cada

sitio libre de la superficie del adsorbente; (ii) la superficie del sólido está

compuesta por sitios homogéneos en términos de energía; (iii) la

saturación se alcanza por completo en una monocapa; (iv) las fuerzas (de

repulsión o atracción) entre moléculas adsorbidas son despreciables en

comparación con las interacciones adsorbente-adsortivo.

La teoría de BET consiste en una extensión de la teoría Langmuir a

adsorción en multicapa y establece algunos supuestos adicionales: (i) la

primera capa adsorbida es similar a la propuesta en el modelo de

Langmuir; (ii) el calor de adsorción es igual al calor molar de

licuefacción en todas las capas, excepto en la primera; (iii) las

condiciones de evaporación-condensación son idénticas, excepto en

todas las capas; (iv) cuando P/P0 = 1, el adsortivo condensa en la

superficie del adsorbente (es decir, el número de capas se considera

infinito).

La isoterma de adsorción según la teoría BET se puede expresar

mediante la ecuación (2.1).

𝑃

𝑃𝑜

𝑛(1−𝑃

𝑃𝑜)

=1

𝑛𝑚𝐶+

𝐶−1

𝑛𝑚𝐶

𝑃

𝑃𝑜 (2.1)

donde P y Po son la presión y la presión de saturación, respectivamente

(en Pa o mm Hg); n y 𝑛𝑚 son los moles adsorbidos por gramo de

adsorbente a una determinada presión (P) y en la monocapa superficial,


95

respectivamente; y C es un parámetro relacionado con el calor de

adsorción.

La representación gráfica 𝑃 𝑃𝑜⁄ /(𝑛𝑚(1 − 𝑃 𝑃𝑜⁄ )) en función de

𝑃 𝑃𝑜⁄ da lugar a una recta, de cuya pendiente y ordenada en el origen se

puede obtener 𝑛𝑚 y C. El valor de 𝑛𝑚 permite determinar la superficie

específica utilizando la ecuación 2.2.

𝑆 = 𝑛𝑚 ∙ 𝑎𝑚 ∙ 𝑁𝐴 ∙ 10−18 (2.2)

donde S es la superficie específica aparente del sólido (m2/g), NA es

el número de Avogadro (6.023·1023 moléculas/mol) y am es el área que

ocupa una molécula de adsorbato (en el caso del N2 a -196 ºC, el valor es

0.162 nm2).

La ecuación BET se debe aplicar en un intervalo de presiones

relativas determinado: 0.05P/P00.3. A presiones relativas inferiores a

0.05, la superficie no es energéticamente homogénea, debido a la

presencia de microporos estrechos que producen el aumento del

potencial de adsorción. A presiones relativas superiores a 0.3, se pueden

dar fenómenos de condensación capilar.

2.4.1.2. Ecuación de Dubinin-Radushkevich

La ecuación de Dubinin-Radushkevich (DR) se utiliza para analizar

la textura porosa de materiales microporosos [14]. Esta ecuación está

basada en la teoría del potencial de Polanyi, que se basa en el supuesto

de que las moléculas de gas se condensan en la superficie del sólido

debido a fuerzas de atracción existentes entre la superficie y las

Capítulo 2

96

moléculas. La fuerza de atracción a un punto dado de la superficie se

mide mediante un potencial de adsorción. La teoría considera que el

espacio de adsorción está compuesto por una serie de superficies

equipotenciales.

Esta teoría permite determinar el volumen de poros de material a

partir de las isotermas de adsorción. Esta ecuación supone la

condensación del adsorbato en los microporos en capas de potenciales

iguales. La ecuación es la siguiente (2.3):

𝑉

𝑉𝑜= exp (−

1

(𝐸0𝛽)2 (𝑅𝑇𝐿𝑛𝑃

𝑃0)

2

) (2.3)

donde V es el volumen adsorbido a una presión P, V0 es el volumen de

microporos del sólido, E0 es la energía característica dependiente de la

estructura porosa, es el coeficiente de afinidad característico del

adsorbato y P0 es la presión de saturación del adsorbato a la temperatura

de trabajo. La presentación de LnV en función de (Ln(P0/P))2 permite

obtener el valor de V0 a partir de la ordenada en el origen.

La aplicación de esta ecuación a isotermas de N2 y CO2 permite

determinar el volumen de microporos total (< 2 nm) y el volumen de

microporos más estrechos (< 0.7 nm), respectivamente.

2.4.1.3. Teoría del funcional de densidad no localizada bidimensional

La teoría del funcional de densidad no localizada (NLDFT, del

inglés non local density functional theory) es muy utilizada para

determinar las propiedades texturales de los sólidos microporosos

(superficie específica y distribución de tamaños de poros) [10]. En esta


97

Tesis Doctoral, este modelo se ha utilizado para la determinación de la

distribución de tamaños de poros. Este parámetro gobierna muchas

propiedades de los materiales carbonosos en diversas aplicaciones, como

el almacenamiento de hidrógeno [15] o en condensadores

electroquímicos [16].

Los cálculos NLDFT se basan en la obtención de isotermas de

adsorción en función de distintos tamaños de poros y geometrías [17-

19]. La distribución de tamaños de poros se puede obtener resolviendo la

ecuación integral de adsorción (ecuación 2.4).

𝑁(𝑝) = ∫ 𝑓(𝑤)𝐾(𝑃. 𝑤)𝑑𝑤𝑏

𝑎 (2.4)

Donde N(p) es isoterma de adsorción experimental (en función

de la presión P, w es el tamaño de poro, (a,b) es el intervalo de tamaños

de poros existente, f(w) representa la distribución de tamaños de poros a

determinar y K es el conjunto de isotermas teóricas expresadas en

función de las densidades de los fluidos en los poros (en función de P y

w). La función K se calcula basándose en un modelo de poro y se conoce

como la función “kernel” de la ecuación integral. El cálculo del “kernel”

que representa un modelo de adsorción dado involucra dos procesos: (i)

el modelado del espacio del poro y la evaluación del potencial sólido-

fluido en los poros; y (ii) cálculo de las densidades de equilibrio de un

fluido confinado en los poros (de acuerdo con el modelo del mismo).

Este modelo considera los poros como hendiduras

unidimensionales en las que el espacio del poro está confinado entre

paredes grafíticas semi-infinitas y uniformes energéticamente. En

Capítulo 2

98

consecuencia, las isotermas calculadas mediante NLDFT no coinciden

de forma exacta con las medidas experimentales obtenidas para carbones

activados. El modelo bidimensional (2D-NLDFT) propuesto por Jagiello

[17] proporciona una mejor aproximación a las propiedades texturales de

los sólidos microporosos. Este modelo considera los carbones activados

como una red estructural basada en grafeno e introduce el efecto de las

heterogeneidades energéticas y las curvaturas superficiales en las

paredes de los poros.

2.4.2. Espectroscopia fotoelectrónica de rayos X

La espectroscopía fotoelectrónica de rayos X (XPS, del inglés X-ray

photoelectron spectroscopy) es una técnica espectroscópica muy

utilizada en la caracterización superficial de sólidos [20], debido a que

permite obtener información de las capas más superficiales (de 1 a 10

nm). Consiste en determinar la energía cinética de los electrones que

emite el material cuando este se irradia con un haz de rayos X. La

energía cinética de los electrones fotoemitidos viene dada por la

ecuación (2.5).

𝐸𝐶 = 𝐸ℎ − 𝐸𝑒𝑒 − (2.5)

donde Ec es la energía cinética del electrón emitido, Eh es la energía de

la fuente de rayos X utilizada, Eee es la energía de ligadura del electrón

excitado relativo al nivel de Fermi y es la función de trabajo

combinación del espectrómetro y la muestra.


99

Para un determinado átomo se pueden determinar distintas energías

para los electrones procedentes de las capas internas o de la capa de

valencia. Por tanto, las energías de ligadura dependen tanto del átomo de

procedencia como de su carga, por lo que mediante esta técnica se puede

obtener información de la naturaleza de los átomos y del estado químico

de los mismos en la superficie. En general, la energía de ligadura

aumenta con el estado de oxidación del átomo. Además, cambia por el

entorno químico del átomo en cuestión, por lo que cuanto mayor sea el

carácter electronegativo de los átomos adyacentes, mayor será la energía

de ligadura de este.

El sistema experimental consta de una fuente de rayos X (ánodo de

Al o Mg), un detector de electrones y un analizador de energía de

electrones. Además, presenta una cámara para trabajar en condiciones de

ultra alto vacío (5·10-7 Pa).

El equipo utilizado en este trabajo es un espectrómetro VG-

Microtech Multilab 3000, perteneciente a los servicios técnicos de

investigación de la Universidad de Alicante. El equipo presenta un

analizador de electrones semiesférico (con energía de paso de 2-200 eV)

y una fuente de radiación de rayos X con ánodos de Mg y Al.

En este trabajo, se ha empleado la técnica XPS para determinar la

composición superficial de los materiales carbonosos. Para este

propósito, se han detectado las transiciones electrónicas C(1s), O(1s) y

N(1s). El análisis de estas transiciones permite cuantificar e identificar

los grupos funcionales superficiales de los materiales estudiados. Las

Capítulo 2

100

condiciones utilizadas para deconvolucionar los espectros se detallan en

cada capítulo. En todos los casos, se tomó como referencia la transición

C1s (284.7 eV). Los valores de energía de ligadura tienen una precisión

de ± 0.2 eV.

2.4.3. Desorción a temperatura programada

La técnica de desorción a temperatura programada (TPD, del inglés

temperature programmed desorption) tiene gran importancia en el

estudio de materiales carbonosos, debido a que permite caracterizar su

química superficial. La técnica consiste en someter a la muestra a un

programa de temperaturas controlado (rampa lineal) en atmósfera

controlada (gas inerte). Los gases procedentes de la descomposición de

los grupos superficiales se analizan en un detector, que se encuentra

acoplado a la salida de la termobalanza. El detector más utilizado es el

espectrómetro de masas. El análisis de los gases resultantes de la

desorción permite obtener información acerca de la composición

química y la estabilidad de los grupos funcionales superficiales.

Los gases desorbidos por la muestra son, principalmente, CO, CO2

(Figura 2.2) y H2O [21–23], y pueden proceder de la descomposición de

los grupos funcionales oxigenados superficiales o descomposición de

compuestos inorgánicos presentes y de la desorción del agua adsorbida

en el material. Estos gases desorben a distintas temperaturas,

relacionadas con la distinta energía de descomposición de cada grupo

oxigenado. Los grupos carboxílicos, lactonas y anhídridos descomponen

como CO2 a distintas temperaturas según la estabilidad de los mismos,


101

aunque suelen ser relativamente bajas (300-500ºC), mientras que los

grupos carbonilos, quinona, fenoles o éter son más estables y

descomponen a elevadas temperaturas como CO (600-1200ºC).

Figura 2.2. Grupos funcionales oxigenados presentes en materiales carbonosos y gases

emitidos por su descomposición.

La cantidad total de oxígeno superficial en las muestras se

determina a partir de las cantidades de CO y CO2 desorbidos al someter

las muestras hasta una temperatura de 950 ºC, atendiendo a la ecuación

(2.6).

[O] = 2[CO2] + [CO] (2.6)

En este trabajo, se ha utilizado un equipo DSC-TGA (TA

Instruments, SDT 2960 Simultenaous) acoplado a un espectrómetro de

masas (Thermostar, Balzers, GSD 300 T3). En los experimentos

Capítulo 2

102

realizados, se utilizaron 10 mg de muestra y el análisis se realizó usando

flujo de Helio como gas de arrastre (100 ml/min) y desde temperatura

ambiente hasta 950ºC. La velocidad de calentamiento fue de 20ºC/min.

2.4.4. Caracterización electroquímica

La caracterización electroquímica de los materiales se ha llevado a

cabo utilizado diversas técnicas y dispositivos electroquímicos. En los

siguientes apartados, se describen las configuraciones de celda y los

fundamentos de las técnicas electroquímicas utilizadas en este trabajo.

2.4.4.1. Configuraciones de celda electroquímica

En esta Tesis Doctoral, se han utilizado celdas de dos y tres

electrodos. Las celdas de dos electrodos se han empleado para

caracterizar condensadores electroquímicos, mientras que las celdas de

tres electrodos permiten realizar un estudio fundamental de las

propiedades electroquímicas de los materiales.

Las celdas electroquímicas de dos electrodos son las unidades

elementales a partir de las que se construyen las baterías y los

supercondensadores [24]. Están compuestas por dos materiales activos

(electrodos positivo y negativo), dos colectores de corriente, un

separador y un electrolito. Los colectores de corriente son conductores

eléctricos que garantizan la transferencia de electrones hacia o desde los

materiales. El separador debe ser un conductor iónico (que permita

mantener el flujo de corriente en la célula) y aislante eléctrico para evitar

el cortocircuito del dispositivo. El separador y los materiales activos


103

deben encontrarse en un medio conductor, como un electrolito líquido,

sólido o polimérico.

Figura 2.3. Configuraciones de celda de (a) dos y (b, c) tres electrodos.

La Figura 2.3a muestra el esquema de los componentes de una celda

de dos electrodos utilizada para caracterizar un supercondensador. En

este sistema, la corriente o el voltaje entre los electrodos positivo y

negativo se registran o controlan para evaluar el comportamiento del

condensador [25]. Los métodos galvanostáticos (cronopotenciometría),

+

+

+

+

+

+

+

+

-

-

-

-

-

-

-

-

+

+

+

+

+

+

+

+

e-

e-

e-

e-

e-

e-

e-

e-

(+) (-)

Electrodo

positivoElectrodo

negativo

SeparadorElectrolito Electrodo de

trabajo

Electrolito

Electrodo de

referencia Contraelectrodo

+

+

+

+

+

+

+

+

-

-

-

-

-

-

-

-

+

+

+

+

+

+

+

+

e-

e-

e-

e-

e-

e-

e-

e-Electrodo de

trabajoContraelectrodo

Electrolito

Electrodo de

referencia(c)

(a) (b)

Capítulo 2

104

potenciostáticos (cronoamperometría) o potenciodinámicos (voltametría

cíclica) son los más utilizados para caracterizar estos dispositivos. Sin

embargo, en una celda de dos electrodos, no se pueden caracterizar las

propiedades electroquímicas de los materiales de forma individual. Para

obtener información fundamental acerca de las propiedades de cada

electrodo, es necesario utilizar una configuración de tres electrodos

mediante el uso de un potenciostato.

La configuración de tres electrodos consta de: (i) un electrodo de

trabajo; (ii) electrodo de referencia y (iii) contraelectrodo. Los tres

componentes se encuentran inmersos en el electrolito de trabajo. Las

Figuras 2.3b y c muestran distintos sistemas experimentales utilizados

para construir una celda de tres electrodos. El sistema experimental

propuesto en la Figura 2.3b es el más empleado para estudiar las

propiedades electroquímicas de los materiales. El electrodo de trabajo se

compone del material a caracterizar, que se deposita o aglomera sobre un

colector de corriente. El electrodo de referencia (Ag/AgCl, electrodo

reversible de hidrógeno (ERH), etc.) y contraelectrodo (Pt, Au, etc.)

utilizados depende del medio electrolítico [26]. Este sistema

experimental se ha utilizado en esta Tesis Doctoral para realizar el

estudio electrocatalítico de los materiales carbonosos. Para esto, se ha

utilizado un electrodo rotatorio de disco-anillo (apartado 2.4.4.2.6),

sobre el que se deposita el material carbonoso a caracterizar.

Para caracterizar materiales para su aplicación posterior en

condensadores electroquímicos, se ha utilizado el sistema experimental

propuesto en la Figura 2.3c. En este caso, la celda de tres electrodos se


105

construye simplemente al introducir un electrodo de referencia en la

celda de dos electrodos. Por medio de esta configuración, se pueden

estudiar las propiedades electroquímicas de uno de los dos electrodos,

que constituye el electrodo de trabajo. El otro material se utiliza como

contraelectrodo, cuya masa se sobredimensiona para evitar interferencias

con el electrodo de trabajo debido a que se produzca un aumento de la

densidad de corriente en ese electrodo. Esta configuración permite

caracterizar los electrodos empleando unas condiciones cercanas a las

del dispositivo final y ha sido la empleada en esta Tesis Doctoral. Dado

que en este trabajo se han utilizado distintos tipos de celdas, en cada

capítulo se especifican los detalles referentes a: (i) configuración de

celda; (ii) construcción del electrodo de trabajo; (iii) tipo de electrodo de

referencia y contraelectrodo.

2.4.4.2 Técnicas electroquímicas

2.4.4.2.1. Voltametría cíclica

La voltametría cíclica es una de las técnicas más versátiles para

estudiar procesos electroquímicos [27]. En esta técnica, se produce la

variación del potencial con el tiempo a una velocidad de barrido dada (v)

desde un potencial inicial hasta alcanzar un potencial final (Figura 2.4).

Una vez alcanzado el potencial final, se lleva a cabo el barrido inverso

(de Eb a Ea). La figura 2.4 representa la variación del potencial con el

tiempo que se debe aplicar para obtener un voltagrama cíclico. Durante

el barrio de potencial se registra la corriente que circula entre el

electrodo de trabajo y el contraelectrodo en configuración de 3

electrodos, o la corriente que circula entre ambos electrodos en la

Capítulo 2

106

configuración de 2 electrodos. La representación de la corriente

resultante en función del potencial proporciona un voltagrama cíclico.

Esta técnica permite evaluar el comportamiento de sistemas muy

diferentes: procesos capacitivos, reacciones en disolución, reacciones en

superficie, etc [27]. En esta Tesis Doctoral, se ha empleado

fundamentalmente para evaluar las propiedades electroquímicas de los

materiales carbonosos y la del supercondensador: capacidad, ventana de

estabilidad electroquímica, etc.

Figura 2.4. Variación del potencial (E) con el tiempo (t) durante un voltagrama

cíclico.

La voltamperometría cíclica permite determinar la capacidad de un

electrodo a partir de la ecuación (2.6.) [28].

𝐶 = 𝑞

𝐸=

∫ 𝐼𝑑𝑡

𝐸 (2.6)

Donde I es la corriente obtenida durante el voltamperograma cíclico y E

es el potencial aplicado.

t

E

Ea

Eb

E0


107

Figura 2.5. Voltagrama cíclico característico de (a) un material que presenta un

comportamiento puramente capacitivo y (b) un material con grupos funcionales

electroactivos en su superficie.

La Figura 2.5 muestra los voltagramas característicos de los

procesos electroquímicos más comunes en los materiales carbonosos

estudiados en este trabajo. En el caso de un material carbonoso en el que

la respuesta electroquímica fundamental se deba a procesos de doble

capa eléctrica (es decir, que no presente grupos electroactivos en su

superficie), el voltagrama cíclico correspondiente presenta una forma

idealmente rectangular (Figura 2.5a) [28]. Si el voltagrama se registra en

un intervalo de potenciales en el que el material no es estable, se

observan corrientes farádicas correspondientes a la oxidación del

material y del electrolito.

En el segundo caso (Figura 2.5b), se presenta el comportamiento de

un material carbonoso que, además de la contribución de la doble capa

eléctrica, presenta un comportamiento con corrientes farádicas como

consecuencia de la presencia de grupos funcionales electroactivos, en las

condiciones de estudio, en su superficie (por ejemplo, quinonas).

(a) (b)

Capítulo 2

108

2.4.4.2.2. Voltametría de barrido lineal

La voltametría de barrido lineal es muy similar a la

voltamperometría cíclica. Del mismo modo, se realiza un barrido de

potencial a una velocidad de barrido dada desde un potencial inicial a un

potencial final [29]. En este caso, no se lleva a cabo el barrido inverso y

el experimento finaliza cuando se alcanza Eb. Ambas técnicas permiten

determinar el potencial al que tiene lugar una reacción electroquímica,

por lo que son muy útiles para estudiar procesos electrocatalíticos. No

obstante, la voltametría de barrido lineal es de especial interés en

procesos irreversibles, en los que el barrido inverso de potencial no

proporciona información adicional.

2.4.4.2.3. Cronopotenciometría

La cronopotenciometría es una técnica galvanostática que se basa en

la medida del potencial en función del tiempo cuando se realiza un salto

de corriente desde un valor inicial a un valor final y se mantiene

constante con el tiempo [27]. El uso de esta técnica permite registrar

ciclos de carga-descarga galvanostáticos, puesto que tras alcanzar un

potencial se invierte el signo de la corriente hasta alcanzar potencial

cero. A partir de los mismos, se pueden determinar parámetros

característicos de los supercondensadores, como: la capacidad

específica, reversibilidad, durabilidad, caída óhmica, energía, potencia,

etc [24].

La Figura 2.6a muestra el salto de la corriente con el tiempo durante

un ciclo de carga-descarga galvanostático. Esta técnica consta de dos


109

etapas: (i) un proceso de carga, en el que se aplica una corriente

constante hasta alcanzar un potencial límite; (ii) un proceso de descarga,

en el que se aplica una intensidad de corriente de polaridad opuesta hasta

alcanzar el potencial inicial. En el caso de un proceso capacitivo, los

procesos de carga y descarga se ajustan a una recta [28].

Figura 2.6. (a) Variación de la corriente con el tiempo durante un ciclo de carga-

descarga galvanostático. (b) Ciclo de carga-descarga galvanostático.

Se puede determinar la capacidad específica de un electrodo o un

condensador por medio de la ecuación (2.7) [28].

𝐶 =𝐼𝛥𝑡

𝑚𝛥𝐸 (2.7)

Donde I es la intensidad de corriente eléctrica aplicada (A), Δt es el

tiempo de descarga (s), ΔE es la ventana de potencial cuando se

caracteriza el material o el voltaje de celda cuando se caracteriza al

condensador (V) y m es la masa del electrodo o la de ambos electrodos

cuando se trata del supercondensador.

Los ciclos de carga-descarga galvanostáticos permiten determinar la

caída óhmica, que está relacionada con la resistencia del material y del

t

I

Δt

0 ΔE

t

IR

E

Δt

(b) (a)

Capítulo 2

110

dispositivo. En numerosas publicaciones, se resta el valor de la caída

óhmica al voltaje o potencial aplicado cuando se determina la capacidad.

En estos casos, los valores de capacidad facilitados son mayores que los

obtenidos cuando se incluye la caída óhmica en el cálculo. Ambos

procedimientos son válidos, pero se debe especificar de forma precisa el

método de cálculo [25]. En este trabajo, todos los cálculos de capacidad

específica se han realizado incluyendo la caída óhmica en el valor de la

diferencia de potencial aplicado o el voltaje para determinar la capacidad

del electrodo o condensador, respectivamente.

Se pueden determinar las características principales de los

condensadores electroquímicos empleando ciclos de carga-descarga en

distintas condiciones. Los estudios principales que se utilizan para

caracterizar estos dispositivos son: (i) obtención de ciclos a distintas

densidades de corriente (A/g), para determinar la energía en función de

la potencia (diagrama de Ragone); (ii) obtención de numerosos ciclos de

carga-descarga a una densidad de corriente y voltaje determinados, para

evaluar la durabilidad del dispositivo [28].

2.4.4.2.4. Cronoamperometría.

En esta técnica, se mide la corriente eléctrica resultante en función

del tiempo al aplicar un potencial constante. Se trata de una técnica muy

empleada para evaluar la durabilidad de los condensadores

electroquímicos, y en este caso se realiza un salto de voltaje. Como se ha

indicado en el apartado 2.4.4.2.3, la forma más habitual de caracterizar

los ciclos de vida de un dispositivo de almacenamiento de energía es la


111

aplicación de un número elevado de ciclos; sin embargo, los

condensadores presentan mayor durabilidad que las baterías, por lo que

es necesario aplicar numerosos ciclos para observar una degradación

significativa. Una alternativa para evaluar la durabilidad en

supercondensadores es la aplicación de condiciones de temperatura

elevada y elevado voltaje de carga para acelerar el proceso de

degradación. Utilizando esta técnica, se puede determinar el tiempo de

vida de un condensador electroquímico de forma más rápida que por

medio de ciclos de carga-descarga, lo que supone una ventaja desde el

punto de vista industrial [25, 30].

Estos estudios se llevan a cabo en celdas de dos electrodos en

condiciones de voltaje constante, que se encuentra habitualmente por

encima del límite de estabilidad [30]. Las características del condensador

(capacidad, resistencia, etc.) se determinan antes y después de la etapa

de cronoamperometría. La Figura 2.7 muestra la evolución de la

corriente (conocida como corriente de fuga) durante el pulso de voltaje.

Una vez el condensador está completamente cargado, la corriente de

fuga decrece rápidamente y alcanza un valor constante conforme

aumenta el tiempo. Esta corriente presenta componentes relacionados

con la doble capa eléctrica y con procesos farádicos (debido a procesos

de degradación) [30, 31]. La integral de la curva I-t proporciona la carga

que fluye por la interfase electrodo/electrolito como consecuencia de la

aplicación del pulso de voltaje [28].

Capítulo 2

112

Figura 2.7. Variación de la corriente de fuga durante la aplicación de un voltaje

constante.

2.4.4.2.5. Espectroscopia de impedancia electroquímica.

La espectroscopia de impedancia electroquímica es una técnica muy

utilizada en diversas aplicaciones electroquímicas, como pilas de

combustible, corrosión, caracterización de capas finas y recubrimientos,

sensores, procesos biológicos, etc [32].

Su característica principal es que puede emplearse para estudiar

procesos con escalas de tiempo de distintos órdenes de magnitud. Esta

técnica se basa principalmente en la perturbación del potencial del

electrodo o voltaje de la celda electroquímica con una señal alterna de

amplitud pequeña, que permite realizar medidas en equilibrio o estado

estacionario [33].

Las perturbaciones que se suele emplear en sistemas

electroquímicos son sinusoidales. En este caso, se han utilizado

perturbaciones sinusoidales del potencial o voltaje aplicado:

𝑉 (𝑡) = 𝑉0𝑠𝑒𝑛2𝜋𝑓𝑡 (2.8)

V

t

IVoltaje constante

Corriente de fuga


113

Donde V(t) es el potencial o voltaje aplicado al tiempo t, V0 es la

amplitud del potencial o el voltaje y f es la frecuencia en Hz.

La corriente de respuesta correspondiente es una función sinusoidal

a la misma frecuencia pero con un desplazamiento de fase:

𝐼 (𝑡) = 𝐼0𝑠𝑒𝑛(2𝜋𝑓𝑡 + Ф) (2.9)

Donde I(t) es la corriente al tiempo t, I0 es la amplitud de la corriente, y

Ф es el desplazamiento de fase.

De forma análoga a la ley de Ohm, se define la impedancia como la

relación entre el voltaje y la corriente:

𝑍 = 𝑉 (𝑡)

𝐼 (𝑡) (2.10)

La impedancia presenta una magnitud (𝑍0 = 𝑉0 𝐼0⁄ ) y una fase (Ф) y

por tanto se puede expresar como un vector con la siguiente notación

compleja:

𝑍 = 𝑍0(𝑐𝑜𝑠Ф + 𝑗𝑠𝑒𝑛Ф) = 𝑍′ + 𝑗𝑍′′ (2.11)

Donde j = √(−1), Z’ es la parte real y Z’’ la parte imaginaria de la

impedancia.

Las representaciones más comunes de la impedancia son: el

diagrama de Nyquist (o plano complejo) y el diagrama de Bode [33]. El

diagrama de Nyquist se obtiene al representar la parte imaginaria en

función de la parte real de la impedancia. Estas representaciones son

muy útiles para obtener parámetros de impedancia a partir de los

Capítulo 2

114

espectros. No obstante, no proporcionan una información completa, ya

que no aparecen los valores de frecuencia a los que se llevó a cabo el

experimento. Para representar esta información, se utiliza el diagrama de

Bode, en el que se representa el ángulo de fase (Ф) y el logaritmo de la

magnitud (logZ0) en función del logaritmo de la frecuencia (log f) y, por

tanto, proporcionan toda la información obtenida en los experimentos de

impedancia.

Para modelar los procesos electroquímicos que tienen lugar durante

un experimento de impedancia, se utilizan combinaciones de elementos

de circuitos eléctricos (condensadores, resistencias, etc.), que se conocen

como circuitos equivalentes. La Figura 2.8a muestra un circuito

equivalente básico para representar un condensador electroquímico y su

diagrama de Nyquist correspondiente, que representa un condensador

ideal en serie con una resistencia [24]. La doble capa eléctrica se

representa por medio de la capacidad (C) y la resistencia eléctrica en

serie (ESR). Esta resistencia en serie representa el comportamiento no

ideal del sistema, y es consecuencia de caídas óhmicas diversas en la

celda electroquímica: la resistencia del electrolito (contribución iónica),

la resistencia de contacto (entre las partículas de carbón y en la interfase

del colector y el material carbonoso) y la resistencia intrínseca de los

componentes (colectores de corriente y material carbonoso). El diagrama

de Nyquist correspondiente a este sistema es una recta vertical en

paralelo al eje imaginario. Sin embargo, los condensadores

electroquímicos presentan un diagrama de Nyquist distinto al

correspondiente a un condensador RC simple (figuras 2.8b y 2.9),


115

debido fundamentalmente a la porosidad de los electrodos de materiales

carbonosos. El comportamiento electroquímico de los electrodos porosos

fue propuesto por De Levie [34,35].

Figura 2.8. (a) Condensador C con una resistencia eléctrica en serie R. (b) Diagrama

de Nyquist para el circuito equivalente presentado en (a).

Este modelo sostiene que el circuito equivalente debe presentar una

serie de componentes RC en serie o en paralelo, que comienzan en la

superficie externa del material y engloban la distribución interna de

poros y superficie porosa. Esta serie de componentes RC presentan

distintas constantes de tiempo RC que se pueden observar como la

respuesta eléctrica de la doble capa cargándose en la superficie interna

del electrodo. La Figura 2.9 muestra el diagrama de Nyquist

correspondiente a un condensador electroquímico compuesto de

electrodos porosos. Estos se pueden dividir en tres partes: (i) a

frecuencias muy altas, solo la parte externa del electrodo es accesible a

los iones y, por tanto, el valor de la impedancia es el correspondiente a la

resistencia eléctrica en serie (ESR); (ii) a frecuencias menores, se

observa un aumento de la parte real e imaginaria de la impedancia con

Z

R C

-Z

R

f 0

f ∞

(a) (b)

Capítulo 2

116

un ángulo de 45º (región de Warburg), como consecuencia de la

distribución de resistencias/capacidades en la porosidad del material, y

que se corresponde con la EDR (del inglés, equivalent distributed

resistance); (iii) a las frecuencias más bajas, se alcanza la capacidad

máxima del material.

Figura 2.9. Diagrama de Nyquist de un electrodo poroso compuesto de poros

uniformes. Adaptado de [28].

El modelo de De Levie ha sido actualizado posteriormente para

tener en cuenta otras características como, por ejemplo, la forma de poro

[36] y la distribución de tamaños de poros [37]. Recientemente, Fletcher

y col. [38] propusieron un modelo de circuito equivalente universal que

consiste en una red de RC paralelas que se encuentran en serie con un

componente RC en paralelo (Figura 2.10), que describe los electrodos

f 0

f ∞

ESR + EDRESR

Z

-Z


117

porosos no ramificados. Este modelo permite explicar algunos de los

defectos de los dispositivos actuales, como: pérdida de capacidad a

elevada frecuencia, disminución del voltaje a circuito abierto, etc.

Figura 2.10. Circuito equivalente universal propuesto por Fletcher y col. [38] para un

condensador simétrico basado materiales carbonosos porosos.

2.4.4.2.6. Electrodo rotatorio de disco-anillo

Para estudiar la cinética y el mecanismo de reacción sobre un

electrodo, y concretamente en el caso de la reacción de reducción de

oxígeno (ORR), es necesario utilizar herramientas que permitan

controlar y determinar el transporte del reactivo a la superficie del

electrodo y su efecto en la cinética de transferencia electrónica. El

método más empleado para este propósito es el electrodo rotatorio de

disco (RDE, del inglés rotating disk electrode). Este consiste en un disco

de un material conductor (carbón vítreo, oro, etc.) incorporado en un

material aislante (teflón, etc.).

Capítulo 2

118

En una disolución electrolítica que contenga un exceso de electrolito

soporte, se puede despreciar el efecto de la migración, de manera que

solo existen dos procesos mayoritarios relacionados con el transporte de

masa: la difusión y la convección. En el caso de una disolución sin

convección, el espesor de la capa de difusión en la superficie del

electrodo aumenta progresivamente conforme aumenta el tiempo de

reacción, dando lugar a una densidad de corriente no estacionaria. Sin

embargo, en una disolución con convección (como agitación o rotación

del electrodo), el espesor de la capa de difusión (y, por tanto, la densidad

de corriente) permanece constante. Por lo tanto, la convección controla

el espesor de la capa de difusión y la difusión controla el transporte del

reactivo a través de esta. El uso de un equipo de RDE que controle de

forma precisa la velocidad de rotación, permite analizar

cuantitativamente la cinética de la reacción que se está produciendo en el

electrodo.

La teoría más utilizada para analizar los datos obtenidos por medio

de RDE para el estudio de la catálisis de ORR se conoce como la teoría

de Koutechy-Levich [39], que proporciona la relación existente entre el

número de electrones involucrado en la reacción, el coeficiente de

difusión de oxígeno, la concentración de oxígeno, viscosidad del

electrolito y velocidad de rotación del electrodo. Mediante el estudio de

estos parámetros, se puede determinar la cinética de la reacción y su

mecanismo, a partir de los cuales se pueden seleccionar y diseñar

catalizadores para su uso en aplicaciones que involucren la ORR [40].

La ecuación de Koutecky-Levich expresa la corriente global del disco


119

(iD), que engloba la contribución cinética relacionada con la velocidad de

la reacción de transferencia electrónica (iK) y la velocidad de difusión del

reactivo (iL):

1

𝑖𝐷=

1

𝑖𝐾+

1

𝑖𝐿 (2.12)

En esta ecuación, iD es la corriente medida e iL es la corriente límite

de difusión. El término iL viene dado por la ecuación de Levich:

𝑖𝐿 = 0.62𝑛𝐹𝐴𝐶0(𝐷0)2 3⁄ 𝜈−1 6⁄ 𝜔1 2⁄ (2.13)

Donde n es el número de electrones involucrados en la reacción, F es la

constante de Faraday (96485 C/mol), C0 es la concentración de la

especie electroactiva en el seno de la disolución, D0 es la constante de

difusión de la especie electroactiva, 𝜈 es la viscosidad cinemática del

electrolito, A es el área del electrodo (cm) y 𝜔 es la velocidad angular de

rotación del RDE.

De esta forma, el uso del RDE permite obtener información acerca

de la cinética de la reacción. Sin embargo, para profundizar en el

mecanismo de reacción es necesario utilizar un electrodo rotatorio disco-

anillo (RRDE, del inglés rotating ring-disk electrode), ya que permite

estudiar la formación de intermedios y productos de reacción.

El RRDE consiste en un electrodo de disco concéntrico con un

electrodo en forma de anillo. Ambos electrodos se encuentran

incorporados y separados en un material aislante (Figura 2.11). Dado

que ambos se encuentran aislados eléctricamente, ambos electrodos

Capítulo 2

120

funcionan como dos electrodos hidrodinámicos separados que, durante

la medida, experimentan rotación a la misma velocidad [40]. En este

sistema, la presencia del anillo no afecta a la corriente y el potencial del

electrodo de disco [39].

Figura 2.11. (a) RRDE utilizado en este trabajo (Pine Research Instrumentations,

EE.UU.) [41]. (b) Componentes del RRDE.

El RRDE se utiliza para realizar experimentos muy diversos que

permiten deducir la presencia de productos intermedios de reacción y el

número de electrones involucrados en la reacción.

Un ejemplo típico del uso del RRDE es el estudio de la ORR. Esta

reacción tiene lugar en el electrodo de disco, de manera que puede

producir dos productos de reacción: agua y peróxido de hidrógeno. Una

vez tiene lugar la reacción en el electrodo de disco, los productos de

reacción se alejan de la superficie del mismo y llegan al electrodo anillo

debido a la convección. En el electrodo anillo se pueden oxidar si el

potencial se fija para oxidar selectivamente uno de los productos: el

peróxido de hidrógeno. De este modo, la medida de la corriente del

anillo permite cuantificar la cantidad de peróxido de hidrógeno que se ha

producido durante la ORR.

Electrodo de disco

(carbón vítreo)

Material aislante

(teflón)

Electrodo anillo

(platino)

(a) (b)


121

Dado que el uso de RRDE requiere examinar dos potenciales (el del

anillo y el del disco), es necesario utilizar un bipotenciostato (que

permite controlar el potencial de los dos electrodos por separado) o bien,

un potenciostato y una fuente externa que permita mantener un potencial

constante en el anillo [39].

EL RRDE presenta una eficiencia de colección característica, N, que

depende exclusivamente de las dimensiones geométricas de los

componentes y es independiente de la velocidad de rotación, coeficiente

de difusión, etc [39]. Este parámetro establece la cantidad de productos

generados en el disco que posteriormente alcanzan el anillo. La

eficiencia de colección viene dada por la ecuación (2.14).

𝑁 = −𝑖𝑅

𝑖𝐷 (2.14)

Este parámetro para un electrodo dado se puede determinar a partir

de las dimensiones geométricas del mismo, o bien, experimentalmente

midiendo la corriente en el disco y en el anillo en un sistema en el que el

producto de reacción sea estable [39][40]. En esta tesis, el RRDE

utilizado (E7R9 Thin Gap RRDE, Pine Research Instrumentation, EE.

UU.), presenta una eficiencia de colección de 0.37.

El uso del RRDE permite determinar el número de electrones

involucrado en la ORR y la cantidad de peróxido de hidrógeno generado

durante la reacción. Estos valores vienen dados por las ecuaciones (2.15)

y (2.16) [40].

𝑛 = 4𝑖𝐷

𝑖𝐷+ 𝑖𝑅 𝑁⁄ (2.15)

Capítulo 2

122

% 𝐻2𝑂2 = 1002𝐼𝑅

𝐼𝐷𝑁+ 𝐼𝑅= 100

4−𝑛

2 (2.16)

Por tanto, una vez conocido N, la determinación de iD e iR permite

calcular el número de electrones involucrados en la reacción y el

porcentaje de peróxido de hidrógeno producido y, de esta manera,

evaluar la actividad catalítica del material estudiado para su aplicación

en la ORR.

2.5 Referencias







87 (2014) 250–257.

[3] H. Nishihara, P. Hou, L. Li, M. Ito, M. Uchiyama, T. Kaburagi,

A. Ikura, J. Katamura, T. Kawarada, K. Mizuuchi, T. Kyotani,

High-Pressure Hydrogen Storage in Zeolite-Templated Carbon, J.

Phys. Chem. C 113 (2009) 3189–3196.

[4] R. Berenguer, H. Nishihara, H. Itoi, T. Ishii, E. Morallón, D.

Cazorla-Amorós, T. Kyotani, Electrochemical generation of

oxygen-containing groups in an ordered microporous zeolite-

templated carbon, Carbon 54 (2013) 94–104.



chemistry on electric double layer capacitance of carbon

materials, Carbon 43 (2005) 2677–2684.

[6] C. Willocq, S. Hermans, M. Devillers, Active carbon

functionalized with chelating phosphine groups for the grafting of


123

model Ru and Pd coordination compounds, J. Phys. Chem. C

(2008) 5533–5541.


García, L. Licea-Jiménez, R. Rychwalski, E.E.B. Campbell,

Covalent amino-functionalisation of single-wall carbon

nanotubes, J. Mater. Chem. 15 (2005) 3334–3339.



of activated carbon fiber/polyaniline materials by position-

resolved microbeam small-angle X-ray scattering, Carbon 50

(2012) 1051–1056.

[9] X. Zhang, S.K. Manohar, Polyaniline nanofibers: chemical

synthesis using surfactants, Chem. Commun. (2004) 2360–2361.

[10] D. Lozano-Castelló, F. Suárez-García, D. Cazorla-Amorós, Á.

Linares-Solano, Porous Texture of Carbons, in: F. Béguin, E.

Frackowiak (Eds.), Carbons Electrochem. Energy Storage

Convers. Syst., 1st ed., CRC Press, 2010: pp. 115–162.

[11] D. Cazorla-Amorós, J. Alcañiz-Monge, M.A. De La Casa-Lillo,

A. Linares-Solano, CO2 as an adsorptive to characterize carbon

molecular sieves and activated carbons, Langmuir 14 (1998)

4589–4596.

[12] D. Lozano-Castelló, D. Cazorla-Amorós, A. Linares-Solano,

Usefulness of CO2 adsorption at 273 K for the characterization of

porous carbons, Carbon 42 (2004) 1231–1236.

[13] S. Brunauer, P.H. Emmett, E. Teller, Adsorption of gases in

multimolecular layers, J. Am. Chem. Soc. 60 (1938) 309–319.

[14] M.M. Dubinin, The potential theory of adsorption of gases and

vapors for adsorbents with energetically nonuniform surfaces,

Chem. Rev. 60 (1960) 235–241.

[15] M. Jordá-Beneyto, F. Suárez-García, D. Lozano-Castelló, D.

Cazorla-Amorós, A. Linares-Solano, Hydrogen storage on

Capítulo 2

124

chemically activated carbons and carbon nanomaterials at high

pressures, Carbon 45 (2007) 293–303.

[16] Y. Li, Z.-Y. Fu, B.-L. Su, Hierarchically Structured Porous

Materials for Energy Conversion and Storage, Adv. Funct. Mater.

22 (2012) 4634–4667.

[17] J. Jagiello, J.P. Olivier, 2D-NLDFT adsorption models for carbon

slit-shaped pores with surface energetical heterogeneity and

geometrical corrugation, Carbon 55 (2013) 70–80.

[18] P.I. Ravikovitch, G.L. Haller, A. V Neimark, Density functional

theory model for calculating pore size distributions: pore structure

of nanoporous catalysts, Adv. Colloid. Interface Sci. 76–77

(1998) 203–226.

[19] P.I. Ravikovitch, A. Vishnyakov, R. Russo, A. V. Neimark,

Unified Approach to Pore Size Characterization of Microporous

Carbonaceous Materials from N2, Ar, and CO2 Adsorption

Isotherms, Langmuir 16 (2000) 2311–2320.

[20] S. Tougaard, Surface Analysis: (b) X-Ray Photoelectronic

Spectroscopy, in: P. Worsfold, A. Townshend, C. Pool (Eds.),

Encycl. Anal. Sci., 2nd ed., Elsevier, Oxford, 2005: pp. 445–456.

[21] M.C. Román-Martínez, D. Cazorla-Amorós, A. Linares-Solano,

C.S.M. de Lecea, TPD and TPR characterization of carbonaceous

supports and Pt/C catalysts, Carbon 31 (1993) 895–902.

[22] H. Boehm, Surface oxides on carbon and their analysis: a critical

assessment, Carbon 40 (2002) 145–149.

[23] J. Figueiredo, M.F. Pereira, M.M. Freitas, J.J. Órfão, Modification

of the surface chemistry of activated carbons, Carbon 37 (1999)

1379–1389.

[24] J.-F. Fauvarque, P. Simon, Principles of Electrochemistry and

Electrochemical Methods, in: F. Béguin, E. Frackowiak (Eds.),

Carbons Electrochem. Energy Storage Convers. Syst., 1st ed.,

CRC Press, Taylor & Francis Group, Boca Raton, 2009: pp. 1–36.


125




[26] M. Inagaki, H. Konno, O. Tanaike, Carbon materials for

electrochemical capacitors, J. Power Sources 195 (2010) 7880–

7903.

[27] A.J. Bard, L.R. Faulkner, Electrochemical Methods.

Fundamentals and Applications, 2nd ed., John Wiley & Sons, Inc,

New York, 2001.



York, 1999.

[29] V.D. Parker, Linear Sweep and Cyclic Voltammetry, Compr.

Chem. Kinet. 26 (1986) 145–202.





[31] P. Ratajczak, K. Jurewicz, F. Béguin, Factors contributing to

ageing of high voltage carbon/carbon supercapacitors in salt

aqueous electrolyte, J. Appl. Electrochem. 44 (2014) 475–480.

[32] N. Bonanos, B.C.H. Steele, E.P. Butler, Applications of

Impedance Spectroscopy, in: E. Barsoukov, J.R. Macdonald

(Eds.), Impedance Spectrosc. Theory, Exp. Appl., 2nd ed., 2005:

pp. 205–537.

[33] S. Krause, Impedance Methods, Encycl. Electrochem. (2007)

196–229.

[34] R. de Levie, On porous electrodes in electrolyte solutions: I.

Capacitance effects, Electrochim. Acta 8 (1963) 751–780.

[35] R. de Levie, On porous electrodes in electrolyte solutions—IV,

Electrochim. Acta 9 (1964) 1231–1245.

Capítulo 2

126

[36] H. Keiser, K.D. Beccu, M.A. Gutjahr, Abschätzung der

porenstruktur poröser elektroden aus impedanzmessungen,


[37] H.-K. Song, Y.-H. Jung, K.-H. Lee, L.H. Dao, Electrochemical

impedance spectroscopy of porous electrodes: the effect of pore

size distribution, Electrochim. Acta 44 (1999) 3513–3519.

[38] S. Fletcher, V.J. Black, I. Kirkpatrick, A universal equivalent

circuit for carbon-based supercapacitors, J. Solid State

Electrochem. 18 (2014) 1377–1387. 2328-4.

[39] A.J. Bard, L.R. Faulkner, Methods involving forced convection-

hydrodynamic methods, in: Electrochem. Methods. Fundam.

Appl., 2nd ed., John Wiley & Sons, Inc, 2001: pp. 331–367.

[40] Z. Jia, G. Yin, J. Zhang, Rotating Ring-Disk Electrode Method,

Elsevier B.V., 2014.

[41] https://www.pineresearch.com.

Chapter 3

N-functionalization of

activated carbon at mild

conditions

N-functionalization of activated carbon

129

3.1. Introduction

Surface chemistry plays a major role in determining the

physicochemical properties of carbon materials. It highly relies on the

presence of functionalities from different heteroatoms (i.e. oxygen and

nitrogen, and some others in lesser amounts), either linked to the carbon

surface or introduced inside the carbon atom framework. Along with

their structural and textural properties, surface chemistry dictates the

potential use of carbon materials in a wide variety of applications in the

fields of catalysis, energy storage, environmental protection and

biomedicine [1-3]. More particularly, nitrogen functionalities in porous

carbon materials modifies their electronic properties and the

electrochemical reactivity and confer basic character to the carbon

surface, enhancing the interaction with acid molecules [4].

Consequently, the production of nitrogen-containing porous carbon

materials have raised a great interest due to their promising performance

as electrocatalysts for the oxygen reduction reaction in fuel cells [5], as

electrodes for supercapacitors [6-8] and as adsorbents for the capture of

acid gases [9,10].

Nitrogen-containing porous carbons are usually synthesized by two

methodologies: reaction of the material with a nitrogen-containing

reagent (NH3, HCN, etc) or carbonization/activation of nitrogen-rich

carbon precursors (urea, melamine, polyaniline, etc.) [4,6,7]. The surface

chemistry of the corresponding material depends on the treatment

conditions, which includes the nature of the selected precursor/reagent

and the treatment temperature. Hence, direct synthesis requires the use

Chapter 3

130

of high temperatures (over 600º C) which does not allow to control the

type of functional groups that are formed on the carbon material.

Similarly, post-treatments with gaseous reagents are usually conducted

at high temperatures, though other approaches at low temperatures

(under 600º C) are found to be interesting for attaining different

functional groups, such as amides [11] or imides [6].

Post-treatment through organic reactions can be a quite interesting

approach because they permit the use of soft treatment conditions that

may preserve the structural characteristics of the carbon materials, and

provide different reaction pathways for finely tuning the surface

chemistry with a wide variety of surface groups [12-19]. Nitrogen

functionalization is usually achieved through the oxidation of the carbon

material with HNO3 in order to generate carboxylic acids on its surface,

which act as the anchorage points for the subsequent reactions [6, 18].

They can directly react with nitrogen-containing reagents to attach this

heteroatom onto the surface, but when this material is previously treated

with SOCl2 the amount of nitrogen introduced in the carbon material is

higher [16]. When this reaction is conducted using amines as the

nitrogen-containing reagent, it produces an amide group substituted with

an aliphatic chain, which can block the microporosity depending on the

size of the hydrocarbon chain of the amine used as reagent [16].

However, when the amine is directly attached to the surface [19] or a

primary amide is produced instead of a secondary one, the blocking of

porosity might be lower. Also, primary amides can be converted to


131

amines directly anchored to the carbon surface by Hofmann

rearrangement that has been proven to work with SWCNTs [20].

Figure 3.1. Chemical route used for the modification of the activated carbon KUA:

chemical oxidation with HNO3 (KUA-COOH), SOCl2 treatment and amidation (KUA-

CONH2) and amination (KUA-NH2).

Interestingly, the selectivity of the synthetic route towards the

desired amides or amines is not usually very high due to the complex

surface chemistry of the carbon materials. For these reasons, in this

chapter, the modification of the surface chemistry of a superporous

activated carbon with the introduction of N functional groups through

organic reactions schematized in Figure 3.1, is studied. The modification

approach consists in: (i) chemical oxidation with HNO3 for the

generation of carboxylic acids (KUA-COOH); (ii) amidation treatment

by introduction of acyl chloride functionalities and conversion of these

groups into amide groups (KUA-CONH2); (iii) transformation of amides

into amine groups by Hofmann rearrangement (KUA-NH2). The porous

texture, surface chemistry and electrochemical properties in aqueous

electrolyte of all the obtained materials will be analyzed, trying to

understand the reactions occurring that may explain the structural

changes of the functional groups introduced on the carbon material

Chapter 3

132

surface. This methodology was also applied to functionalize a

microporous zeolite templated carbon (Annex to Chapter 3).

3.2. Materials and methods

3.2.1. Activated carbon

A highly microporous activated carbon prepared in our laboratory

has been used as the starting material for nitrogen incorporation via

organic chemical modification. The pristine material, henceforth named

KUA, has been obtained by chemical activation of an anthracite with

KOH using an impregnation ratio of activating agent to raw material of

4:1 and an activation temperature of 750º C under inert atmosphere,

which was held for 1 hour. More details about the preparation process

are available elsewhere [21].

3.2.1.1. Chemical oxidation with HNO3

The chemical oxidation of the activated carbon KUA was done

using HNO3 as oxidant. Nitric acid is known as strong oxidizing agent

that yields an important amount of carboxylic acids when used for

treating carbon materials, and it has been widely used in literature for

oxygen surface functionalization [22-24].

The oxidation treatment has been adopted from previous studies of

our research group [24]. Briefly, 1g of activated carbon (KUA) was

contacted with 40 mL of HNO3 65 wt% under stirring during 3 hours at

room temperature. After that, KUA-COOH was filtrated and washed


133

with Milli-Q water until the pH of elution was neutral. Finally, the

sample (named KUA-COOH) was dried at 100º C.

3.2.1.2. Generation of amide functionalities

The amidation process consists in two sequential steps: first, acyl

chlorides are generated from carboxylic functionalities and, secondly, a

nitrogen-containing compound is anchored to the carbonyl via

nucleophilic substitution over the acyl chloride. These reactions were

carried out inside a Schlenk system. Although the use of less toxic

compounds is desirable and is considered in the literature [25], this

approach contributes interestingly to the development of mild conditions

for the modification of carbon materials. The acyl chloride formation is

accomplished as follows [26]: Dry KUA-COOH (1 g) was introduced in

a round bottom flask with 50 mL of toluene and 5 mL of SOCl2 was

added to the flask. The mixture was refluxed at 120ºC for 5 hours under

Ar atmosphere. After that time, the sample was washed with toluene and

dried under vacuum for 14h.

The reaction for the generation of the amide group has been adapted

from that proposed by Gromov and coworkers for CNTs [20]. KUA-

COCl was added into a 2 M NH4NO3/DMF solution (activated carbon to

solution ratio of 1g/300mL) in a round bottom flask. Then, 300 mL of

pyridine were added slowly to the round bottom flask under continuous

stirring at room temperature. The mixture was stirred at 70 ºC for 65

hours under Ar atmosphere. The obtained amidated sample (KUA-

Chapter 3

134

CONH2) was washed with abundant water and ethanol, filtered and dried

at 100º C overnight.

3.2.1.3. Generation of amine functionalities

Amine functionalization was carried out using a Hofmann

rearrangement in the previously attached amides [20]: 10 mL of Br2 was

added into a solution of 3% wt. NaOCH3 in CH3OH. KUA-CONH2 (0.5

g) were added to that solution, and the mixture was stirred at 70º C for 4

hours. Then, additional 4 mL Br2 were added, and the mixture was

stirred for 20 hours at 70º C. The product was isolated by filtration,

washed with saturated Na2CO3, water and ethanol. The aminated

activated carbon was obtained after hydrolyzing the recovered sample in

500 mL of 0.1M NaOH for 24 hours, washing with water, and finally

filtering and drying the sample at 100º C overnight. This activated

carbon is named KUA-NH2.

3.2.2. Electrochemical characterization

Carbon electrodes for electrochemical characterization were

prepared by mixing the activated carbon with acetylene black and

polytetrafluoroethylene (PTFE) as binder in a ratio of 90:5:5 (w/w). The

total weight of the electrode was 9 mg (dry basis). For shaping the

electrodes, a sample sheet was cut into a circular shape with an area of

1.2 cm2 and pressed for 5 min at 2 tons to guarantee a homogeneous

thickness. After that, the electrode was placed on a gold disk used as a

current collector. The electrodes were impregnated for 2 days into 1M

H2SO4 previously to electrochemical measurements.


135

The electrochemical characterization was performed by using an

Autolab PGSTAT302 for cyclic voltammetry (CV) and Arbin SCTS for

galvanostatic charge-discharge cycles. The electrochemical

characterization of the different materials synthesized was performed by

using a three-electrode configuration. As counter electrode, more than

20 mg of a KUA electrode was used. Both electrodes were placed

against each other and separated by a nylon membrane (pore size: 320

nm). Ag/AgCl/KCl (3M) was used as reference electrode in all cases.

1M H2SO4 was used as aqueous electrolyte. The electrochemical

performance of all samples was tested by CV at sweep rates between 1

and 50 mV/s and galvanostatic charge-discharge (GCD) cycles at current

densities of 50-50000 mA/g. The potential range used for all the

measurements was 0.2-0.8 V vs Ag/AgCl/KCl (3M). The capacitance in

the CV was calculated from the area of the voltammogram, while in the

GCD experiments, the capacitance was obtained using the next equation:

𝐶𝑔 =𝑗·𝑡

∆𝑉 (3.1)

Where j is the current density (A/g), t is the discharge time and ∆𝑉

is the difference of potential (0.6 V in all experiments). The results are

expressed in F/g, taking into account the weight of the active material of

the electrode.

3.2.3. Porous texture and surface chemistry characterization

The porous texture characterization was carried out by N2

adsorption-desorption isotherms at -196º C and by CO2 adsorption at 0º

C by using an Autosorb-6-Quantachrome apparatus. The samples were

Chapter 3

136

outgassed at 200º C for 4 hours before the experiments. The apparent

surface area was obtained from N2 adsorption-desorption isotherms by

using the BET equation. The micropore volume was determined by

Dubinin-Radushkevich method applied to the CO2 and N2 adsorption

isotherms.

The surface chemistry of the samples was analyzed by TPD and

XPS. TPD experiments were performed by heating the samples (10

mg) to 950º C (at a heating rate of 20º C/min) under a helium flow rate

of 100 mL/min. The analyses were carried out by using a TGA-DSC

instrument (TA Instruments, SDT Q600 Simultaneous) coupled to a

mass spectrometer (Thermostar, Balzers, BSC 200). X-ray Photoelectron

Spectroscopy (XPS) analyses were carried out using a VG-Microtech

Multilab 3000 spectrometer, equipped with an Al anode. The

deconvolution of N1s spectra was carried out by using Gaussian

functions with 20% of Lorentzian component. FWHM of the peaks was

kept between 1.3 and 1.5 eV and a Shirley line was used for estimating

the background signal.

3.3. Results and discussion

3.3.1. Porous texture characterization

Figure 3.2 shows the N2 adsorption-desorption isotherms for the

original activated carbon and the activated carbon at different stages of

the proposed surface chemistry modification protocol. It can be seen that

nitrogen uptake in KUA mainly occurs at low relative pressures,

showing type I isotherm, characteristic of a microporous material. The


137

wide knee observed in the isotherm at low relative pressures indicates

that the material has a wide micropore size distribution. Both

characteristics of the nitrogen isotherm are also observed for all the

modified activated carbons, although the nitrogen uptake seems to be

affected differently by each chemical treatment.

Figure 3.2. N2 adsorption-desorption isotherms of the activated carbons KUA, KUA-

COOH, KUA-CONH2 and KUA-NH2.

Table 3.1 summarizes the porous texture for all the activated carbon

samples. The pristine activated carbon (KUA) presents a large apparent

surface area. The micropore volume determined by CO2 adsorption at 0º

C (VDRCO2) corresponds to the volume of the narrowest micropores (<0.7

nm), while the obtained by N2 adsorption at -196º C (VDRN2) corresponds

to the whole microporosity (<2nm) [27]. The higher micropore volume

measured by nitrogen adsorption indicates that the activated carbon

presents a wide micropore size distribution.

0

200

400

600

800

1000

1200

0 0.2 0.4 0.6 0.8 1

Ad

sorb

ed V

olu

me

(cm

3/g

ST

P)

P/Po

KUA

KUA-COOH

KUA-CONH₂

KUA-NH₂

Chapter 3

138

Table 3.1. Textural properties and XPS elemental surface composition for the pristine

and modified activated carbons.

Sample SBET

(m2/g)

VDRN2

(cm3/g)

VDRCO2

(cm3/g)

CXPS

(at.%)

OXPS

(at.%)

NXPS

(at.%)

KUA 3140 1.10 0.56 90.9 8.8 0.3

KUA-COOH 2680 0.96 0.45 81.8 18.2 -

KUA-CONH2 2080 0.74 0.44 84.2 12.0 3.8

KUA-NH2 2450 0.91 0.42 79.1 17.9 3.0

KUA-

CONH2_T950

- - - 94.5 4.4 1.1

KUA-

NH2_T950

- - - 89.9 8.8 1.3

The sample KUA-COOH showed a decrease in the nitrogen uptake,

in the apparent surface area and in the pore volume (Figure 3.2 and

Table 3.1). This effect is related to the generation of surface groups that

can occupy the entrance and some volume of the microporosity [23, 24].

In the case of KUA-CONH2, the amidation treatment produced a further

decrease of VDRN2, whereas VDR

CO2 remains invariable. This decrease

suggests that the modification of the surface chemistry involves the

largest micropores, while the narrowest microporosity is probably not

accessible to the reagents.

The amination treatment leaded to an increase of the specific surface

area and VDRN2 of KUA-NH2 when compared to KUA-CONH2 sample

(Table 3.1 and Figure 3.2). The narrowest microporosity remains again

practically invariable. If we assume that the desired reaction has been

carried out, the size of the functional groups that occupy or block the

microporosity in KUA-NH2 should be smaller than those in KUA-

CONH2, because of the loss of a CO molecule in the production of


139

amines from amides. This means that the change in the microporosity is

in agreement with the intended functionalization of the activated carbon

sample.

3.3.2. Surface chemistry characterization

3.3.2.1. XPS

Table 3.1 compiles the composition of the surface chemistry of the

samples obtained by XPS. The pristine activated carbon (KUA) is

composed mainly by carbon and a relatively high amount of oxygen.

XPS analysis of KUA-COOH confirms that the oxidation of this

material has been produced during the treatment with HNO3. The same

technique also corroborates that nitrogen is attached onto the surface of

the activated carbon KUA-CONH2, remaining on it after completing the

treatment for amination (KUA-NH2). It is important to highlight that

more than 3 at. % nitrogen content has been successfully loaded onto the

carbon surface while producing only a minor change of the porous

texture of the activated carbon (less than 5% when BET surface area of

KUA-COOH and KUA-NH2 are compared, Table 3.1). This is a

remarkable result since maintaining the porosity of the material is of

huge importance for all the surface-dependent applications of nitrogen-

doped porous carbons. In comparison, other amidation reactions over

activated carbon produced a loss of specific surface area of 42% and

90%, depending on the amine used as reagent in the amidation reaction

[16]. The results indicate that the amidation procedure proposed in this

PhD Thesis is less detrimental to the porous texture than other

Chapter 3

140

approaches involving larger amines. To prove the viability of the method

in other highly microporous carbon materials, the amidation reaction

was successfully applied over a zeolite-templated carbon (ZTC),

producing an attachment of nitrogen of 1.4 at. %. More details about the

N-funtionalization of ZTC are available in Annex 3.

The oxygen amount measured in each step (Table 3.1) can also give

some information about the process responsible for the nitrogen

functionalization. After the amidation step, the introduction of nitrogen

is accompanied by a consumption of oxygen, which evidences that the

attachment of nitrogen to the surface has been produced mainly through

oxygen functional groups. On the other hand, the sample KUA-NH2

presents an oxygen content similar to KUA-COOH (18%,

approximately), and it has also been confirmed by TPD measurements,

see section 3.3.2.2. This result is initially unexpected, because this step

might lead to the loss of CO of the amide group through a rearrangement

and decarboxylation, in order to obtain amine functionality, so that some

loss of oxygen should happen. This increase is probably related to a

collateral oxidation during the amination treatment. This will be

explained later in more detail. The amount of residual bromine

containing species has been determined to be 0.1% at., and therefore are

not considered to be relevant in the electrochemical performance of

KUA-NH2.


141

Figure 3.3. N1s XPS deconvolution of (a) KUA-CONH2, (b) KUA-NH2.

Figures 3.3.a and 3.3.b show the N1s XPS spectra of the samples

KUA-CONH2 and KUA-NH2. They have been deconvoluted according

to the literature [9,14,28-32]. Table 3.2 summarizes the results of the

XPS analysis. The spectra of both samples revealed the existence of at

least three species with different binding energies. In the case of KUA-

CONH2, the spectrum can be separated into three peaks located at 400.7,

399.8 and 398.8 eV. The peak at 399.8 eV is assigned to amides, amines,

lactams and imides on the surface [30]. The formation of lactams and

imides during amidation is plausible from already formed amides

395396397398399400401402403404

I (a

.u.)

Binding Energy (eV)

(a)

395396397398399400401402403404

I (a

.u.)

Binding Energy (eV)

(b)

Chapter 3

142

groups, since they can condense with adjacent hydroxyls and carboxylic

acids that were already over the carbon surface or that can be formed

during the initial stages of the wet acid oxidation of the surface. In this

sense, the generation of these species could take place from anhydrides

[33] and lactones [34], as is showed in Figures 3.4a and 3.4b. Similar

paths have been proposed for the treatment of non-porous carbon

materials with NH3 gas [6]. At this stage of the modification protocol,

the presence of amines should be discarded because their formation

could not be explained from the expected reactions; however, it is not

possible to discard its presence with the information obtained by XPS.

Table 3.2. Assignment of N1s deconvoluted curves to nitrogen functional groups.

Sample Binding

Energy (eV)

Functional

Group

N

(at. %)

Relative

%

KUA-CONH2 400.7 ± 0.2 Pyrrole,

pyridone

0.72 19

399.8 ± 0.2 Amide, Lactam,

Amine, Imide

1.89 50

398.8 ± 0.2 Pyridine, Imine 1.17 31

KUA-NH2 400.5 ± 0.2 Pyrrole,

Pyridone

0.72 27


Amine, Imide

1.25 48

398.5 ± 0.2 Pyridine, Imine 0.65 25

KUA-CONH2

after TPD

398.4 ± 0.2 Pyridine 0.55 49

400.6 ± 0.2 Pyrrole,

Pyridone

0.57 51

KUA-NH2

after TPD

398.3 ± 0.2 Pyridine 0.42 31

400.2 ± 0.2 Pyrrole,

Pyridone

0.92 69


143

The peak at 400.7 eV in the N1s spectra of KUA-CONH2 (400.7

eV) is assigned to pyrrole and pyridone functional groups, while that

found at 398.8 eV is associated to the presence of pyridines and imines

[9,28,30,32]. These signals are not related to the presence of adsorbed

pyridine or DMF, since their desorption has not been detected by

following the 52 and 42 m/e- lines during the TPD experiments,

respectively. The formation of nitrogen heterocycles at such low

temperatures is an unexpected outcome of the proposed surface

modification protocol. This striking result can be explained taking into

account the heterogeneity of the activated carbon surface and the

availability of different surface oxygen groups. Pyridone can be formed

from amide groups when they condense with adjacent hydroxyls in order

to produce lactams, which are tautomers of pyridone (Figure 4c) [31,35].

As for the pyiridines/imines, the imines can be generated by reaction

between a carbonyl group (aldehyde or ketone) and a nitrogen reactant,

such as secondary/tertiary amines and ammonia according to the

literature [36,37]. The imines obtained from the last one are more stable

when they are linked to two carbon atoms forming an aromatic ring (i.e

pyrroles) [37] rather than in form of terminal imines; nevertheless both

of them could coexist in the carbon surface. Possible reaction pathways

for the formation of imines and pyrroles are shown in Figures 3.5a and

3.5b.

Chapter 3

144

Figure 3.4. (a) Formation of lactams from lactones; (b) formation of imides from

anhydrides; (c) tautomerization equilibrium of the functional groups lactam (left) and

pyridone (right); (d) formation of pyridones from pyridines.

When the imines are formed, condensation reactions with aldehydes

and ketones can take place, and nitrogen aromatic heterocycles would be

produced [38], which structure will depend on the distance between the

carbonyls. Hence, two carbonyls separated by two carbon atoms allow

the formation of cycles of five members (pyrrole), Figure 3.5b, whilst a

distance of three carbon atoms, Figure 3.5c, entails the production of six

member rings (pyridine). The condensation required to form a

heterocycle implies the loss of two H2O molecules. It should be pointed

out that the formation of aromatic rings requires the formation of an

intermediate (enamines), as illustrated in Figure 3.5. The atomic


145

composition measured by XPS seems to validate this hypothesis. The

oxygen amount diminishes in 6.2 at. % after the amidation treatment,

whereas the oxygen consumption expected from the amount of freshly

formed amides and heterocycles is 5.7 at.%, according to the reactions

proposed in Figures 3.4 and 3.5.

The analysis of the N1s spectra of KUA-NH2 shows significant

changes in comparison with KUA-CONH2. Both samples present peaks

with similar binding energies, as can be confirmed in Table 3.2, but their

distribution is different. The peak at 399.6 eV should be associated to the

presence of amines on the surface [30,32]. Nevertheless, this assignment

is not enough to confirm the formation of amines from amides, since

both functional groups appear in the same range of binding energy,

according to the literature. On the other hand, the peaks at 400.4 and

398.5 eV can be assigned to the functional groups formed following the

reactions showed in Figure 3.5, as in the case of KUA-CONH2.

However, in the sample KUA-NH2 there exists a larger relative amount

of species of high binding energy (pyrroles and/or pyridones), Table 3.2.

Chapter 3

146

Figure 3.5. Formation of (a) imines, (b) pyrroles and (c) pyridones from aldehydes and

ketones; (d) thermal decomposition of pyridone to form pyrrole through loss of a CO

group.

The increase in the oxygen amount detected by XPS and TPD after

amination indicates that other reactions have taken place in addition to

the conversion of amides to amines. This is confirmed when the step c in

Figure 3.1 is directly applied to KUA-COOH. These results proved that

the attachment of methoxide and/or hydroxyl groups occurs onto the

surface of KUA-NH2 and that it takes place through substitution

reactions without any further oxidation. This can also be related to the

larger relative amount of N-functional groups at 400.4 0.2 eV in this

sample in comparison with KUA-CONH2. A possible explanation is the

nucleophilic substitution during the reaction in the adjacent position of

the nitrogen on the pyridine rings (Figure 3.4.d). Under this hypothesis,

the methoxide present in the media during the first step of the reaction,


147

or the hydroxide ions present in the second step, could be attached to the

adjacent carbon of the nitrogen group due to the influence of the latter,

which confers -deficient character to the ring [35]. This process could

also happen sequentially: first, attachment of methoxide into the pyridine

ring and secondly, the substitution of methoxide by hydroxide ions, as is

shown in Figure 3.4.d. Hence, the substitution in 2 position of pyridine

would generate pyridones, so that the amount of groups at the binding

energy of 400.4 eV would increase as well as the oxygen content.

Finally, the decrease of nitrogen content in KUA-NH2 (Table 3.1)

can be associated to the hydrolysis of imines to carbonyls (explaining

the decrease of the peak at 398.5 eV, Table 3.2), which is favored in acid

media [37], but that can also happen easily in water [39]. As for the

decrease detected at 399.5 eV, it can be related to the loss of part of the

amides and their derivatives of the KUA-CONH2, since amines are

stable in the preparation conditions.

3.3.2.2. Temperature Programmed Desorption

Table 3.3 reports the amount of CO and CO2 desorbed, as well as

the total oxygen amount (calculated as CO + 2CO2) during the TPD

experiments for all the studied samples. It can be appreciated that the

tendencies in the amount of total oxygen for each preparation step

coincide with those obtained by XPS (Table 3.1). Differences in the

obtained amounts are due to the techniques being not quantitatively

comparable. XPS gives information about the most external surface of

the sample; while the oxygen amount obtained from TPD analyses is

Chapter 3

148

usually underestimated, since some oxygen functional groups can

thermally decompose forming water, which has not been quantified, or

can decompose at temperatures higher than those reached at the end of

the TPD experiment.

Table 3.3. Amount of evolved CO, CO2 and total O obtained from the TPD

experiments.

Sample CO

(μmol/g)

OCO

(wt. %)

CO2

(μmol/g)

OCO2

(wt. %)

O

(μmol/g)

O

(wt. %)

KUA 2100 3.4 500 1.6 3100 5.0

KUA-COOH 4030 6.5 1680 5.4 7390 11.8

KUA-CONH2 2490 4.0 1030 3.3 4550 7.3

KUA-NH2 2800 4.5 1480 4.7 5760 9.2

Figure 3.6 shows the CO and CO2 TPD profiles for the pristine and

surface-modified activated carbons obtained after each synthesis step.

The pristine sample (KUA) presents a remarkable variety of oxygen

groups on its surface, all of them generated during the activation

process. Its aimed surface functionalization with carboxylic acids seems

to be accomplished after the nitric acid treatment, as pointed out by the

large amount of CO2 desorption observed between 200 and 400ºC as

consequence of the decomposition of carboxylic acids [22, 40-42].

Collaterally, the amount of desorbed CO, which in pristine KUA sample

results from the decomposition of carbonyls and quinones (CO at

temperatures over 700ºC) and as phenols (related to CO evolution at

600-700ºC) in a lower extent [40], have also increased after the nitric

acid oxidation. Some CO desorption at temperatures lower than 600ºC is

observed because of the presence of anhydride groups, being their

amount also higher for the KUA-COOH sample. These groups thermally


149

decompose as CO and CO2, thus the CO desorption is accompanied by

CO2 evolution in similar amounts at temperatures between 400 and

600ºC. The results from the nitric acid treatment are in line with

previous studies from our research group [24,43] and the abundant

literature regarding wet oxidation of porous carbons [23, 44].

Figure 3.6. Comparison between (a) CO2 and (b) CO TPD profiles of KUA-COOH,

KUA-CONH2 and KUA-NH2.

Furthermore, the comparison between the profiles from the oxidized

and the amidated and aminated samples in Figure 3.6 allows us to clarify

both the changes in the surface oxygen groups produced during the

0

0.5

1

1.5

2

0 200 400 600 800 1000

CO

2(μ

mol/

g)

Temperature (ºC)

KUA

KUA-COOH

KUA-CONH₂

KUA-NH₂

(a)

0

0.5

1

1.5

2

2.5

3

3.5

0 200 400 600 800 1000

CO

(μ

mo

l/g

s)

Temperature (ºC)

KUA

KUA-COOH

KUA-CONH₂

KUA-NH₂

(b)

Chapter 3

150

different steps of the functionalization and the most probable mechanism

for the formation of the nitrogen groups. The functionalization with

nitrogen groups during amidation is produced through substitution and

condensation reactions onto surface oxygen groups, as pointed out by

the lower CO2 and CO (at high temperatures) evolutions measured for

the KUA-CONH2 sample if compared to that of KUA-COOH. More

concretely, the comparison of the CO2 profiles evidences consumption

of carboxylic acids (300ºC) and lactones (350-600ºC), as can be

observed in Figure 3.6a (total consumed amount being 650 μmol/g).

This decrease is expected according to the desired reaction, since the

formation of amide groups and its derivatives requires the consumption

of CO2-desorbing groups (see Figure 3.1). On the other hand, the KUA-

CONH2 sample shows a decrease in the evolution of CO groups at high

temperatures, Figure 3.6.b, that is quantified to be 1500 μmol/g. This

consumption of CO groups is related to the formation of another type of

nitrogen functional groups (Figures 3.5a, 3.5b and 3.5c) that, taking into

account the results obtained by XPS, might be imines, pyrroles,

pyridones and pyridines.

Further evidences of the attained nitrogen functionalization have

been obtained during the TPD experiments. Indeed, Figure 3.7.a, which

presents the m/z=14 signal (arbitrary units) as a function of temperature

for the TPD of KUA-CONH2, shows a peak at 380º C that is related to

the evolution of a nitrogen-containing molecule. It also coincides with a

CO desorption shoulder (Figure 3.6.b), providing enough evidences of

the decomposition of the amide group. Furthermore, the continuous N


151

desorption along the whole temperature range is probably due to lactams

decomposition, which can either desorb at higher temperatures or react

forming pyridines, and the degradation of aromatic rings (pyrroles,

pyridones and pyridines) at even higher temperatures [30]. This is

confirmed by the XPS experiments obtained for the samples thermally

treated until 950 ºC, Table 3.2, where an important decrease in the N

content is detected and only contributions of pyrrole/pyridone and

pyridines are observed.

Additional comparison can be drawn between the TPD profiles

of the KUA-CONH2 and KUA-NH2 samples, Figure 3.6. As was

observed by XPS, the amination reaction increases the amount of

oxygen on the surface in the form of CO and CO2-evolving groups. The

CO2 profile suggests the formation of carboxylic acids during the

amination step, although direct comparison between the profiles of

KUA-COOH and KUA-NH2 (Figure 3.6.a) confirms the overall

consumption of carboxylic acids for the whole synthesis route. A

possible explanation for the increase of CO2-evolving groups in KUA-

NH2 is the esterification of lactams (that have been formed in the

amidation step, Figure 3.4.a), which can react with the methoxide

present in the media and form esters that would subsequently be

hydrolyzed to form again lactone moieties (reverted reaction of Figure

3.4.a), evidencing that the Hofmann rearrangement to produce amines

does not take place over lactams. This is supported by the loss of

nitrogen detected by XPS after amination step (Table 3.2). There have

also been important changes in the CO evolution profile after the

Chapter 3

152

amination of KUA-CONH2 sample, Figure 3.6.b. If the low temperature

region of both desorption profiles are compared, a decrease in the CO

desorption related to amide decomposition (region between 200 and

400ºC) is observed for KUA-NH2. This decrease is probably the

consequence of a consumption of amides and lactams, which evidences

the conversion of these groups into amines and lactones, respectively.

This seems to be supported by the differences in the desorbed amount of

nitrogen species in this region of temperatures in KUA-NH2 (Figure

3.7.b). Moreover, the sample KUA-NH2 presents higher amount of

phenols (CO desorption at 600-700ºC) and a higher CO desorption at the

end of TPD run (temperatures higher than 900ºC). The last one is

produced simultaneously to the increase of the N-related signal recorded

at high temperatures for KUA-NH2, as shown in Figure 3.7.b. This

increase could be a consequence of a favored rearrangement or

formation of N-containing aromatic rings (Figures 3.5.b and 3.5.c)

during amination [30]. As for the CO increase, a favored insertion of

hydroxyl groups during the amination treatment is possible due to the -

deficient behavior of pyridines, which facilitates nucleophilic

substitution reactions on the adjacent carbon (Figure 3.4.d). The

resulting pyridones can be transformed into pyrroles by thermolysis,

which produces the loss of a CO molecule, as is shown in Figure 3.5d

[45]. This is supported by the XPS results obtained for the samples

KUA-CONH2 and KUA-NH2 after the TPD treatment up to 950º C

(Table 3.2), where the higher amount of pyrroles measured after TPD of

KUA-NH2 seems to be connected to the presence of a higher amount of

pyridones in the original KUA-NH2 sample.


153

Figure 3.7. I14 line during TPD experiment for (a) KUA-CONH2 and (b) KUA-NH2

samples.

0 200 400 600 800 1000

I 14

(a.u

.)

Temperature (ºC)

(a)

0 200 400 600 800 1000

I 14

(a.u

.)

Temperature (ºC)

(b)

Chapter 3

154

3.3.3. Electrochemical characterization.

3.3.3.1. Cyclic Voltammetry

The pristine and modified materials have been characterized by

cyclic voltammetry in the 0.2-0.8V potential range in 1M H2SO4 using

different scan rates. Figure 3.8 shows the voltammograms for all

activated carbon electrodes at low scan rate (1 mV/s). Gravimetric and

surface capacitances (gravimetric capacitance divided by the BET

surface area) are reported in Table 3.4. The formation of the electrical

double layer is manifested in all of them, evidenced by the rectangular

shape of their voltammograms between 0.2V and 0.65V. At high

positive potentials an oxidation current is observed for all the samples

and the oxidation current depends on the electrode material. The low

scan rate used for the CV measurements provides enough time for ion

diffusion into the pore network, so all the surface wetted by the

electrolyte will be available for the formation of the double layer. The

calculated capacitance is different for each material due to the

differences in surface area and surface chemistry. However, it can be

seen that the order in the values of the gravimetric capacitance (Table

3.4) does not follow in all the cases the trends observed in the BET

surface area in Table 3.1. Thus, the activated carbon with the highest

capacitance is the pristine carbon (KUA), but the highly oxidized KUA-

COOH shows the lowest capacitance in spite of having higher surface

area than KUA-CONH2 and KUA-NH2.


155

Figure 3.8. 2nd cyclic voltammograms in the potential range 0.2-0.8 V for KUA, KUA-

COOH, KUA-CONH2 and KUA-NH2 electrodes.1M H2SO4. v=1 mV/s.

When capacitance is divided by BET surface area (i.e. surface

capacitance, second column in Table 3.4), values concordant with those

found in the literature for porous carbons are obtained [46], except for

the one obtained for KUA-COOH. These results are an example of the

influence of the surface chemistry in the electrochemical behavior of

carbon materials. The effect of the surface oxygen groups in capacitance

has been studied in previous works of our research group [24]. In

general, oxygen functionalities improve the behavior of the electrodes

mainly due to two effects: increase of wettability of the surface, which

facilitates interactions with the electrolyte, and introduction of functional

groups capable of experimenting redox reactions, generating

pseudocapacitance [24,47]. These advantages are related to CO-

desorbing groups, while the presence of CO2-desorbing groups is

unfavorable because of their electron-withdrawing properties that

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

j (A

/g)

E (V vs Ag/AgCl/KCl)

KUA

KUA-COOH

KUA-CONH₂

KUA-NH₂

Chapter 3

156

diminish the delocalization of the charge (and thereby the electrical

conductivity). This effect explains the lower specific capacitance in

KUA-COOH. However, the specific capacitance for KUA-CONH2 is

similar or even higher than for the other materials. In this sample, a

significant part of the CO2-desorbing groups are subsequently

transformed into amides and derivatives, thus partially deactivating their

negative influence, while keeping the positive influence coming for the

presence of CO-desorbing [24] and other N-containing [48] groups.

Table 3.4. Gravimetric (Cg) and surface capacitances (Cg/SBET) and Ohmic drop

(IRdrop) for all the samples measured at different electrochemical conditions. 1 M

H2SO4.

Sample Cg

[1mV/s]

(F/g)

Cg/SBET

[1mV/s]

(mF/m2)

Cg

[0.5 A/g]

(F/g)

Cg

[20 A/g]

(F/g)

Cg

[50 A/g]

(F/g)

IRdrop

[20 A/g]

(mV)

KUA 299 95 268 82 33 106

KUA-COOH 171 64 138 6 3 380

KUA-CONH2 217 105 206 125 82 66

KUA-NH2 249 102 209 17 2 272

The impact of the changes in the surface chemistry in the

capacitance retention was analyzed by recording CV at different scan

rates, Figure 3.9. The pristine activated carbon is able to keep the CV

characteristics up to 20 mV/s, while KUA-COOH and KUA-NH2

yielded tilted voltammograms at scan rates of 10 mV/s and higher. In

terms of capacitance retention, the three samples seem to behave

similarly. Interestingly, the best results are obtained for the amidated

sample (KUA-CONH2) that is able to retain the CV shape up to 50

mV/s. Since the micropore size distribution seems to be similar for all

the samples, as already discussed in section 3.3.1, these differences in


157

behavior must be related to the changes in the surface chemistry of the

samples.

Figure 3.9. Gravimetric capacitance at different scan rates for KUA, KUA-COOH,

KUA-CONH2 and KUA-NH2 electrodes.1M H2SO4. v= 1, 10, 20 and 50 mV/s.

The negative effect of surface oxygen groups and especially CO2-

desorbing groups in electrical conductivity can explain the low

capacitance retention in KUA-COOH. KUA-NH2 also has similar

capacitance retention, but higher specific capacitance at all scan rates. In

the case of KUA-CONH2, the capacitance experiences the smaller

decrease with scan rate, as proved in Figure 3.9. Consequently, this

electrode has the highest specific capacitance at the highest tested scan

rate. The capacitance retention capability is undoubtedly related to their

surface chemistry, since it presents basically the same pore size

distribution than the parent sample and the lowest specific surface area

of the all studied materials. This effect has been studied in more detail

using galvanostatic charge-discharge analyses.

0

50

100

150

200

250

300

350

0 20 40 60

Cap

acit

an

ce (

F/g

)

Scan rate (mV/s)

KUAKUA-COOH

KUA-CONH₂KUA-NH₂

Chapter 3

158

3.3.3.2. Galvanostatic charge-discharge cycles

The rate performance of the original and the modified samples has

been characterized by chronopotentiometry at different specific currents

(from 0.05 to 50 A/g). Figure 3.10 shows the 4th galvanostatic charge-

discharge cycle at 0.5 A/g, 20 A/g and 50 A/g, and the obtained

capacitance and ohmic drop at 20 A/g are compiled in Table 3.4. The

discharge time, which is directly related to the capacitance reported in

Table 3.4, follows the order KUA>KUA-NH2~KUA-CONH2>KUA-

COOH at 0.5 A/g. However, this tendency changes at higher specific

currents (>10 A/g), being then KUA-CONH2>>KUA>>KUA-

NH2>KUA-COOH. This result is in agreement with the capacitance

values and capacitance retention observed in the cyclic voltammetry

study, section 3.3.3.1. An outstanding capacitance value of 82 F/g was

registered for the KUA-CONH2 at 50 A/g, being high considering that it

is achieved with a surface loading of 8 mg/cm2, a value in the range of

those used in the commercial formulation of supercapacitors [49].

The ohmic drop, reflected by the sudden drop in potential when

moving from charge to discharge at 0.8V in Figures 3.10.b and 3.10.c, is

also affected by the surface chemistry of the materials. This ohmic drop

is associated to the electrical series resistance of the electrode, which

includes the inherent resistance of the materials used in the different

components of the cell (current collector, membrane, etc.), the

interparticle electrical resistance (being connected to the electrode

preparation), the intraparticle electrical resistance (which is related to the

inherent conductivity of the sample) and the ion diffusional resistance


159

(which depends on the pore size distribution and connectivity). Since the

cell and the electrode preparation is identical in all the cases, and ion

diffusional resistance must be similar for the analyzed materials, the

important differences in the ohmic drop values must be related to the

surface chemistry. As discussed before, the higher amount of CO2-

evolving groups results in a worse rate performance in KUA-COOH,

while the presence of N functionalities in KUA-NH2, which improve the

wettability and electrical conductivity [6,8,48], seems to be responsible

of its slightly better performance, although the behavior as

supercapacitor electrode is still worse than the starting material, KUA.

This is probably due to the lower amount of functional groups in KUA

that facilitate the charge delocalization on its surface with respect to that

of KUA-NH2. Very interestingly, the capacitance retention is greatly

improved for the KUA-CONH2 electrode, being much higher than that

of the oxidized carbon KUA-COOH and even higher than that of the

pristine activated carbon. Thus, the gravimetric capacitance of KUA-

CONH2 is higher than that found for KUA at specific currents higher

than 10 A/g, thanks to the lower ohmic drop found for this material

(Table 3.4). This material seems to behave similar in terms of

capacitance retention than other N-containing porous carbon materials

prepared from N-containing precursors reported in literature [4,6,50-54].

Other N-doping techniques, such as ammoxidation or urea treatment

[55,56], do not seem to produce such a good capacitance retention as the

amidation route proposed in this work.

Chapter 3

160

Figure 10. Galvanostatic charge-discharge curves at current densities of (a) 0.5 A/g,

(b) 20 A/g and (c) 50 A/g. Potential range: 0.2-0.8V.

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 200 400 600

E (

V v

s A

g/A

gC

l/K

Cl)

Time (s)

KUA

KUA-COOH

KUA-CONH₂

KUA-NH₂

(a)

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 2 4 6 8

E (

V v

s A

g/A

gC

l/K

Cl)

Time (s)

KUA

KUA-COOH

KUA-CONH₂

KUA-NH₂

(b)

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 0.4 0.8 1.2 1.6 2

E (

V v

s A

g/A

gC

l/K

Cl)

Time (s)

KUA

KUA-CONH₂(c)


161

The mechanism behind the capacitance and the electrical

conductivity improvement in N-containing carbon materials is still under

debate in the literature [6-8,11,57-59]. Theoretical studies [59] have

proven that the nitrogen of pyrrole groups improves the electron

mobility of the carbon matrix through introduction of electron-donor

properties and increases the catalytic activity of the carbon in electron

transfer reactions. Also, the nitrogen atoms present in a six member ring

in the edge of a graphene layer can be considered as pyrrole-like

functionalities due to the conjugation of its two p electrons with the

system of the graphene layer. Hulicova-Hurkacova and coworkers

indicated that lactams and imides can be considered as pyrrole-like

groups, since they present both characteristics [6]; also, they proposed

the participation of nitrogen and oxygen of imide, lactam and pyridone

groups in pseudocapacitance processes. This explains the high

capacitance retention of the electrode KUA-CONH2, which presents a

higher amount of pyrrole-like groups on its surface. On the other hand,

they rejected the participation of amine and amide groups in the

improvement of electro-donating properties. However, we consider that

the presence of amides can also improve the behavior of the sample

KUA-CONH2, because part of the carboxylic acids that were present on

the surface of KUA-COOH has been replaced and, although amides are

also electron-withdrawing groups, when they are attached to the

aromatic ring through the carbonyl group, this character is lower than

that of carboxylic acids. Also, Hsieh and coworkers [11] indicated the

possible contribution of amide groups in pseudocapacitive processes in

supercapacitors of CNTs modified by a synthetic path quite similar to

Chapter 3

162

our approach. Finally, the benefits of pyridines in carbon material with

respect to electrical conductivity is unclear, whereas its positive

influence in capacitance seems to be associated with enhanced

wettability [57], enhanced interaction with ions [58] and the occurrence

of pseudofaradaic processes [57], which is reported to take place through

the following reaction [60]:

HC=N + H2O ↔ CHN=O + 2e- + 2H+

In a different work about the effect of nitrogen functionalities in the

electrochemical behavior of carbon materials [8], the contribution of

pyridines in the surface of nitrogen-containing CNTs to the enhancement

of capacitance was found to be higher in basic electrolyte, so its

presence is not expected to be critical in the behavior of the materials

studied in this work at acidic conditions.

Hence, the improvements in the capacitance retention found in

KUA-CONH2 can be explained by the lower amount of carboxylic acids

in the surface when compared to KUA-NH2 (Figure 3.6) and by the

nitrogen functional groups on its surface. Since the main differences in

the chemical composition in terms of nitrogen surface groups between

KUA-CONH2 and KUA-NH2 is the presence of amide groups and

derivatives in the former that have been replaced by amines and

carboxylic acids in the latter, and considering that amines have a higher

electron-donating character than amides, the enhanced performance of

the KUA-CONH2 is attributable to the cyclic amides (lactams and

imides) and their pyrrol-like effect on the electrochemical performance

of carbon electrodes [6].


163

3.4. Conclusions

Functionalization of activated carbon using an organic chemistry

protocol has been carried out by amidation treatment and Hofmann

rearrangement. Nitrogen content of about 3 at% is achieved through this

process under mild conditions. As consequence of the heterogeneous

surface chemistry of activated carbons, a wide range of functional

groups have been produced onto the surface of the modified samples,

showing different nitrogen and oxygen functional groups on KUA-

CONH2 and KUA-NH2. The amidation treatment produces amides (and

cyclic derivatives, such as imides and lactams) on KUA-CONH2, but the

presence of CO-desorbing groups on KUA-COOH yields the generation

of nitrogen aromatic heterocycles (pyridine, pyridone and pyrrole).

Consequently, KUA-CONH2 presents a complex surface chemistry with

different nitrogen functional groups, some of them also stable at high

temperatures (pyrrole and pyridine). The post-treatment by Hofmann

rearrangement produces the conversion of amides into amines, obtaining

an activated carbon (KUA-NH2) with different nitrogen functionalities

(amines, pyridines and pyrroles) and a high amount of surface oxygen

groups. Different reaction pathways are proposed to explain the

observed changes.

The different surface chemistry of all activated carbons allows us to

study its effect on their electrochemical performance. At low current

densities, the capacitance is mainly governed by the apparent surface

area while the specific capacitance is greater for the N-containing

samples, evidencing the influence of nitrogen functional groups.

Chapter 3

164

Interestingly, KUA-CONH2 showed the highest capacitance retention,

keeping an outstanding value of 83 F/g at 50 A/g, a value especially high

considering the electrode thickness. This capacitance retention was not

registered for any of the other electrodes and, therefore, it might be due

to the lower amount of electron-withdrawing groups, as carboxylic

acids, and to the nitrogen functional groups existing on KUA-CONH2,

such as pyridines, cyclic amides (lactams and imides) and pyrroles,

which improve the charge delocalization and thereby increase the

electrical conductivity.

3.5. References

[1] Bandosz TJ, Ania CO. Surface chemistry of activated carbons and

its characterization. In: Bandosz TJ, editor. Activated Carbon

Surfaces in Environmental Remediation, New York:1st ed,

Elsevier; 2006, p. 159–229.

[2] Bandosz TJ. Surface Chemistry of Carbon Materials. In: Serp P,

Figueiredo JL, editors. Carbon Materials for Catalysis, John

Wiley & Sons, Inc; 2009, p. 45–92.

[3] Radovic LR. Surface Chemical and Electrochemical Properties of

Carbons. In: Beguin F, Frackowiak E, editors. Carbons for

Electrochemical Energy Storage and Conversion Systems, Boca

Raton, FL: Taylor & Francis (CRC Press); 2010, p. 163–219.

[4] Shen W, Fan W. Nitrogen-containing porous carbons: synthesis

and application. J Mater Chem A 2013;1(4): 999-1013.


165

[5] Gong KP, Du F, Xia ZH, Durstock M, Dai LM. Nitrogen-doped

carbon nanotube arrays with high electrocatalytic activity for

oxygen reduction. Science 2009;323:760–4.

[6] Hulicova-Jurcakova D, Kodama M, Shiraishi S, Hatori H, Zhu

ZH, Lu GQ. Nitrogen-enriched nonporous carbon electrodes with

extraordinary supercapacitance. Adv Funct Mater

2009;19(11):1800–9.

[7] Salinas-Torres D, Shiraishi S, Morallón E, Cazorla-Amorós D.



electrolyte at severe conditions. Carbon 2015;82:205-13.

[8] Ornelas O, Sieben JM, Ruiz-Rosas R , Morallón E, Cazorla-

Amorós D, Geng J, et al. On the origin of the high capacitance of

nitrogen-containing carbon nanotubes in acidic and alkaline

electrolytes. Chem Commun 2014;50:11343-6.

[9] Raymundo-Piñero E, Cazorla-Amorós D, Linares-Solano A. The


from flue gases by N-doped activated carbon powders and fibres.

Carbon 2003;41(10):1925–32.

[10] Sevilla M, Valle-Vigón P, Fuertes AB. N-Doped polypyrrole-

based porous carbons for CO2 capture. Adv Func Mater

2011;21(14):2781–7.

[11] Hsieh C-T, Teng H, Chen W-Y, Cheng Y-S. Synthesis,


functionalized carbon nanotube/carbon paper electrodes. Carbon

2010;48(15):4219–29.

Chapter 3

166

[12] Silva AR, Martins M, Freitas MM, Valente A, Freire C, de Castro

B, et al. Immobilisation of amine-functionalised nickel(II) Schiff

base complexes onto activated carbon treated with thionyl

chloride. Microporous Mesoporous Mater 2002;55(3):275–84.

[13] Silva AR, Martins M, Freitas MMA, Figueiredo JL, Freire C,

de Castro B. Anchoring of copper(II) acetylacetonate onto an

activated carbon functionalised with a triamine. Eur J Inorg Chem

2004;10:2027–35.

[14] Silva AR, Budarin V, Clark JH, Freire C, de Castro B. Organo-

functionalized activated carbons as supports for the covalent

attachment of a chiral manganese(III) salen complex. Carbon

2007;45(10):1951–64.

[15] Alves JAC, Freire C, de Castro B, Figueiredo JL. Anchoring of

organic molecules onto activated carbon. Colloids Surf A

2001;189(1-3):75–84.

[16] Tamai H, Shiraki K, Shiono T, Yasuda H. Surface


carbons by immobilization of diamine. J Colloid Interface Sci

2006;295(1):299–

[17] Peñas-Sanjuán A, López-Garzón R, Domingo-García M, López-

Garzón FJ, Melguizo M, Pérez-Mendoza M. An efficient

procedure to bond nanostructured nitrogen functionalities to

carbon surfaces. Carbon 2012;50(11):3977–86.

[18] El-Sayed Y, Bandosz TJ. Role of surface oxygen groups in

incorporation of nitrogen to activated carbons via

ethylmethylamine adsorption. Langmuir 2005;21(4):1282–9.


167

[19] Abe M, Kawashima K, Kozawa K, Sakai H, Kaneko K.

Amination of activated carbon and adsorption characteristics of its

aminated surface. Langmuir 2000;16(11):5059–63.

[20] Gromov A, Dittmer S, Svensson J, Nerushev OA, Perez-García

SA, Licea-Jiménez L, et al. Covalent amino-functionalisation of

single-wall carbon nanotubes. J Mater Chem 2005;15:3334–9.

[21] Lozano-Castelló D, Lillo-Ródenas MA, Cazorla-Amorós D,

Linares-Solano A. Preparation of activated carbons from Spanish

anthracite. I Activation by KOH. Carbon 2001;39(5):741–9.

[22] Boehm HP. Surface oxides on carbon and their analysis: a critical

assessment. Carbon 2002;40(2):145–9.

[23] Moreno-Castilla C, López-Ramón MV, Carrasco-Marın F.

Changes in surface chemistry of activated carbons by wet

oxidation. Carbon 2000;38(14):1995–2001.

[24] Bleda-Martínez MJ, Lozano-Castelló D, Morallón E, Cazorla-

Amorós D, Linares-Solano A. Chemical and electrochemical

characterization of porous carbon materials. Carbon

2006;44(13):2642–51.

[25] Jeong YN, Choi MY, Choi HC. Preparation of Pt- and Pd-

decorated CNTs by DCC-activated amidation and investigation of

their electrocatalytic activities. Electrochim Acta 2012; 60: 78-84.

[26] Willocq C, Hermans S, Devillers M. Active carbon functionalized

with chelating phosphine groups for the grafting of model Ru and

Pd coordination compounds. J Phys Chem C 2008;112(14):5533–

41.

Chapter 3

168

[27] Cazorla-Amorós D, Alcañiz-Monge J, de la Casa-Lillo MA,

Linares-Solano A. CO2 as an adsorptive to characterize carbon

molecular sieves and activated carbons. Langmuir

1998;14(16):4589–96.

[28] Raymundo-Piñero E, Cazorla-Amorós D, Linares-Solano A, Find

J, Wild U, Schlögl R. Structural characterization of N-containing

activated carbon fibers prepared from a low softening point

petroleum pitch and a melamine resin. Carbon 2002;40(4):597–

608.

[29] Biniak S, Szymański G, Siedlewski J, ŚwiaTkowski A. The

characterization of activated carbons with oxygen and nitrogen

surface groups. Carbon 1997;35(12):1799–810.

[30] Jansen RJJ, van Bekkum H. XPS of nitrogen-containing

functional groups on activated carbon. Carbon 1995;33(8):1021–

7.

[31] Kapteijn F, Moulijn JA., Matzner S, Boehm HP. The development

of nitrogen functionality in model chars during gasification in

CO2 and O2. Carbon 1999;37(7):1143–50.

[32] Yamada Y, Kim J, Matsuo S, Sato S. Nitrogen-containing

graphene analyzed by X-ray photoelectron spectroscopy. Carbon

2014;70:59–74.

[33] Smith MB, March J. March’s Advanced Organic Chemistry:

Reaction, Mechanisms and Estructure. 6th ed. Hoboken, New

Jersey, USA: John Wiley & Sons, Inc. 2007: 1429–30.


169



Jersey, USA: John Wiley & Sons, Inc. 2007: 1436.

[35] Clayden J, Greeves N, Warren S, Wothers P. Organic Chemistry.

1st ed. Oxford University Press. 2001: 1150–1.









Reaction, Mechanisms and Estructure. 6th ed., Hoboken, New


[40] Román-Martínez MC, Cazorla-Amorós D, Linares-Solano A,

Salinas-Martínez de Lecea C. Tpd and TPR characterization of

carbonaceous supports and Pt/C catalysts. Carbon

1993;31(6):895–902.

[41] Lopez-Ramon MV, Stoeckli F, Moreno-Castilla C, Carrasco-

Marin F. On the characterization of acidic and basic surface sites

on carbons by various techniques. Carbon 1999;37(8):1215–21.

[42] Figueiredo JL, Pereira MFR, Freitas MMA, Órfão JJ.

Modification of the surface chemistry of activated carbons.

Carbon 1999;37(9):1379–89.

Chapter 3

170

[43] Bleda-Martínez MJ, Maciá-Agulló JA, Lozano-Castelló D,

Morallón E, Cazorla-Amorós D, Linares-Solano A. Role of

surface chemistry on electric double layer capacitance of carbon

materials. Carbon 2005;43(13):2677–84.

[44] Moreno-Castilla C, Ferro-García MA, Joly JP, Bautista-Toledo I,

Carrasco-Marín F, Rivera-Utrilla J. Activated carbon surface

modifications by nitric acid, hydrogen peroxide, and ammonium

peroxydisulfate treatments. Langmuir 1995;11(11):4386–92.

[45] Schmiers H, Friebel J, Streubel P, Hesse R, Kopsel R. Change of

chemical bonding of nitrogen of polymeric N - heterocyclic

compounds during pyrolysis. Carbon 1999;37(12):1965–78.

[46] Qu D. Studies of the activated carbons used in double-layer

supercapacitors. J Power Sources 2002;109(2):403–11.

[47] Conway BE, Birss V, Wojtowicz J. The role and utilization of

pseudocapacitance for energy storage by supercapacitors. J Power

Sources 1997;66(1-2):1–14.

[48] Lota G, Grzyb B, Machnikowska H, Machnikowski J, Frackowiak

E. Effect of nitrogen in carbon electrode on the supercapacitor

performance. Chem Phys Lett 2005;404(1-3):53–8.

[49] Stoller MD, Ruoff RS. Best practice methods for determining an

electrode material's performance for ultracapacitors. Energy

Environ Sci 2010;3:1294-301.

[50] Alabadi A, Yang X, Dong Z, Tan B. Nitrogen-doped activated

carbons derived from a co-polymer for high supercapacitor

performance. J Mater Chem A 2014;2:11697–705.


171

[51] Paraknowitsch JP, Zhang J, Su D, Thomas A, Antonietti M. Ionic

liquids as precursors for nitrogen-doped graphitic carbon. Adv

Mater 2010;22(1):87–92.

[52] Kim K-S, Park S-J. Synthesis and high electrochemical

capacitance of N-doped microporous carbon/carbon nanotubes for

supercapacitor. J Electroanal Chem 2012;673:58–64.

[53] Wu C, Wang X, Ju B, Jiang L, Wu H, Zhao Q, et al.

Supercapacitive performance of nitrogen-enriched carbons from

carbonization of polyaniline/activated mesocarbon microbeads. J

Power Sources 2013;227:1–7.

[54] Wickramaratne NP, Xu J, Wang M, Zhu L, Dai L, Jaroniec M.

Nitrogen enriched porous carbon spheres: attractive materials for

supercapacitor electrodes and CO2 adsorption. Chem Mater

2014;26(9):2820–8.

[55] Pietrzak R, Jurewicz K, Nowicki P, Babeł K, Wachowska H.

Microporous activated carbons from ammoxidised anthracite and

their capacitance behaviours. Fuel 2007;86(7-8):1086–92.

[56] Jurewicz K, Pietrzak R, Nowicki P, Wachowska H. Capacitance

behaviour of brown coal based active carbon modified through

chemical reaction with urea. Electrochim Acta 2008;53(16):5469–

75.

[57] Seredych M, Hulicova-Jurcakova D, Lu GQ, Bandosz TJ. Surface

functional groups of carbons and the effects of their chemical

character, density and accessibility to ions on electrochemical

performance. Carbon 2008;46(11):1475–88.

Chapter 3

172

[58] Jeong HM, Lee JW, Shin WH, Choi YJ, Shin HJ, Kang JK, et al.

Nitrogen-doped graphene for high-performance ultracapacitors

and the importance of nitrogen-doped sites at basal planes. Nano

Lett 2011;11(2):2472–7.

[59] Strelko VV, Kuts VS, Thrower PA. On the mechanism of possible



transfer reactions. Carbon 2000;38(10):1499–503.

[60] Frackowiak E, Lota G, Machnikowski J, Vix-Guterl C, Béguin F.

Optimisation of supercapacitors using carbons with controlled

nanotexture and nitrogen content. Electrochim Acta

2006;51(11):2209–14.


173

ANNEX TO CHAPTER 3.

Nitrogen functionalization of zeolite templated carbon by

electrochemical and chemical methods.

In Chapter 3, the applicability of nitrogen doping through organic

chemistry pathways to an activated carbon was demonstrated. However,

it is interesting to expand the use of this methodology to other carbon

forms. As seen in Chapter 1, zeolite templated carbons (ZTCs) are

materials obtained as a replica of a zeolite that have unique structure and

properties [1,2]. Nevertheless, as consequence of their high reactivity,

their structure is very fragile and hinders their potential use in a wide

range of applications. For this reason, nitrogen doping appears as a

promising strategy for improving the properties of these materials [3].

More information related to this topic is provided in Chapter 5.

Here we propose the nitrogen doping of ZTC through the amidation

protocol developed in Chapter 3. The scheme of the modification

protocol is shown in Figure A3.1. This nitrogen functionalization

method was slightly modified in order to achieve a successful doping of

ZTC. First, all surface modification methods (oxidation and amidation)

were applied over the carbon-zeolite composite prior to the removal of

the template by HF washing (named as cZTC). In this way, the

preservation of the structure of ZTC is guaranteed. The second important

modification of the applied protocol is that the oxidation treatment was

carried out via electrochemical methods, to avoid the damage of the

carbon material.

Chapter 3

174

Figure A3.1. Chemical route used for the modification of the activated carbon KUA:

(a) electrochemical oxidation in 0.5M H2SO4 electrolyte (cZTC-COOH), (b) SOCl2

treatment and amidation (cZTC-CONH2) and (c) removal of the zeolite (ZTC-CONH2).

Table A3.1 shows the textural parameters obtained for cZTC and

ZTC. The results evidence that N2 is not accesible to the microporosity

of the composite. However, the narrow microporosity was found to be

0.11 cm3/g for cZTC. If this value is expressed per gram of carbon

material in cZTC ( 20 % of the composite), the narrow micropore

volume estimated is 0.55 cm3/g of ZTC, what corresponds to a BET

surface area of around 1500 m2/g [4]. Interestingly, this value

corresponds to around 42 % of the BET surface area obtained for ZTC,

evidencing that almost one side of the curved graphene layers of ZTC is

exposed to the CO2 adsorbate in cZTC.

Table A3.1. Textural properties obtained for cZTC and ZTC.

Sample SBET

(m2/g)

VDRN2

(cm3/g)

aVDRCO2

(cm3/g)

bVDRCO

2

(cm3/g)

cZTC 11 0.004 0.11 0.55

ZTC 3600 1.55 - - a m (g) cZTC; b m (g) ZTC

The electro-oxidation of cZTC was carried out by CV in 0.5 M

H2SO4 using the experimental procedure developed by Nueangnoraj and

coworkers for ZTC [5]. Figure A3.2 shows the CVs obtained for ZTC


175

and cZTC upon positive polarization. Under these conditions, ZTC

experiences oxidation reactions with the electrolyte that lead to an

increase of oxygen functionalities [1]. The presence of the zeolite in

cZTC should avoid the electro-oxidation process; however, the materials

show electrochemical response as evidenced by the increase of oxidation

current at 0.8 V. Also, the electrical double layer of cZTC at 0.7 V

(Figure A3.2a, red line) is 60% of that found for the ZTC (Figure

A3.2b, black line), suggesting that only one side of the graphene layer is

exposed to the electrolyte in the carbon-zeolite composite, what is in

agreement with the CO2 adsorption data. The increase of oxygen

functional groups after the electrochemical treatment was confirmed by

XPS (samples cZTC and cZTC_ox, Table A3.2). Nevertheless, the

characteristic redox pair of the quinone-hydroquinone groups is not

clearly observed in cZTC.

Figure A3.2. Cyclic voltammograms for the electro-oxidation of (a) cZTC and (b)

ZTC. v = 1 mV/s. 0.5M H2SO4.

The electro-oxidized carbon-zeolite composite (cZTC_ox sample)

was further functionalized to introduce nitrogen groups on the carbon

surface. Table A3.2 summarizes the chemical composition of the sample

-1000

0

1000

2000

3000

-0.3-0.10.1 0.3 0.5 0.7 0.9 1.1

C (

F/g

)


(b)

-1000

0

1000

2000

3000

-0.3-0.10.1 0.3 0.5 0.7 0.9 1.1

C (

F/g

)


(a)

Chapter 3

176

obtained after N-doping treatment (cZTC-CONH2). The amidation

reaction produces an attachment of 1.4 % at. N that is accompanied by a

large decrease of oxygen functional groups. The amount of nitrogen is

retained on the carbon material (ZTC-CONH2) after the removal of the

zeolite by HF washing as confirmed by XPS (1.4 at. %, Table A3.2).

Figure A3.3 shows the XPS spectra obtained for cZTC-CONH2 and

ZTC-CONH2, evidencing that nitrogen is anchored mainly in form of

amides, pyridines and pyrroles [6–8].

Table A3.2. Chemical composition of the samples obtained by XPS.

Sample CXPS

(% at.)

OXPS

(% at.)

NXPS

(% at.)

SiXPS

(% at.)

AlXPS

(% at.)

cZTC 60.5 16.7 - 15.9 6.9

cZTC_ox 59.5 31.5 - 8.6 1.3

cZTC-CONH2 70.0 13.3 1.3 12.2 3.6

ZTC-CONH2 91.4 6.9 1.4 0.3 -

Figure A3.3. N1s XPS spectra obtained for cZTC-CONH2 and ZTC-CONH2 samples.

395396397398399400401402403404405

I (a

.u.)

Binding Energy (eV)

cZTC-CONH₂

ZTC-CONH₂


177

The N-doped ZTC was characterized by CV in 1M H2SO4. Figure

A3.4a shows the CVs obtained for ZTC and ZTC-CONH2 in a potential

window where the materials are stable. The N-doped ZTC evidences a

similar curve than the pristine carbon material, although the oxidation

current is much lower, what is a characteristic of the N-doped carbons.

When these carbons are exposed to positive polarization (at 1 V), the

electro-oxidation process takes place and the characteristic redox pair of

quinone-hydroquinone is observed for both carbons (Figure A3.3b). This

electrochemical process is strongly dependent on the structure of ZTC

[9] and, consequently, confirms that the porous structure of the pristine

ZTC is retained after the N functionalization treatment.

Figure A3.4. Cyclic voltammograms obtained for ZTC (black curve) and ZTC-CONH2

(red curve) samples (a) before and (b) after electro-oxidation at 1.0 V vs

Ag/AgCl/KCl.

In conclusion, a zeolite-templated carbon was successfully

functionalized by electrochemical and chemical methods based in

electro-oxidation treatments and amidation reactions. To preserve the

porous structure of the pristine ZTC, the doping methods were applied

-500

-250

0

250

500

-0.3 0 0.3 0.6

C (

F/g

)


(a)

-800

-400

0

400

800

-0.3 0 0.3 0.6

C (

F/g

)


(b)

Chapter 3

178

keeping the zeolite template in the composite material. Oxygen

functional groups were attached to the composite and converted

afterwards into nitrogen functionalities, that are retained in the porous

carbon after the removal of the zeolite. The electrochemical

characterization of the N-doped ZTC evidenced that the unique

properties of ZTC remained after the functionalization treatments.

References




87 (2014) 250–257.


Cazorla-Amorós, T. Kyotani, Electrochemical generation of

oxygen-containing groups in an ordered microporous zeolite-

templated carbon, Carbon 54 (2013) 94–104.





[4] T.A. Centeno, F. Stoeckli, The assessment of surface areas in

porous carbons by two model-independent techniques, the DR

equation and DFT, Carbon 48 (2010) 2478–2486.

[5] K. Nueangnoraj, R. Ruiz-Rosas, H. Nishihara, S. Shiraishi, E.

Morallón, D. Cazorla-Amorós, T. Kyotani, Carbon–carbon

asymmetric aqueous capacitor by pseudocapacitive positive and

stable negative electrodes, Carbon 67 (2014) 792–794.

[6] E. Raymundo-Piñero, D. Cazorla-Amorós, A. Linares-Solano,

The role of different nitrogen functional groups on the removal of

SO2 from flue gases by N-doped activated carbon powders and

fibres, Carbon 41 (2003) 1925–1932.


179

[7] R.J.J. Jansen, H. van Bekkum, XPS of nitrogen-containing

functional groups on activated carbon, Carbon 33 (1995) 1021–

1027.

[8] Y. Yamada, J. Kim, S. Matsuo, S. Sato, Nitrogen-containing

graphene analyzed by X-ray photoelectron spectroscopy, Carbon

70 (2014) 59–74.



Characterization of a zeolite-templated carbon by electrochemical

quartz crystal microbalance and in situ Raman spectroscopy,

Carbon 89 (2015) 63–73.

Chapter 4

Electrochemical performance of

N-doped activated carbons in

aqueous electrolyte

Electrochemical performance of N-doped activated carbons in aqueous electrolyte

183

4.1. Introduction

Supercapacitors have attracted considerable interest as energy

storage devices thanks to their high power density, a key missing feature

of fuel cells and electrical batteries. They are based on the formation of

an electrical double layer on the extensive surface of porous carbon

materials, which is profited to store energy that can be delivered in few

seconds. This energy storage mechanism is often complemented with

pseudocapacitive processes produced by fast redox reactions occurring

in electroactive surface functional groups. The development of electrode

materials with large surface area, electroactive functional groups and

high electrochemical stability is crucial for the improvement of the

energy density and durability of these devices [1,2].

Activated carbons are the most employed materials as electrodes for

supercapacitors, mainly due to their large apparent surface area, proper

electrical conductivity, high electrochemical stability and competitive

production cost [2]. Also, their surface chemistry can be conveniently

modified to introduce surface functionalities that could improve the

performance of these materials by an increase of electrochemical

stability, conductivity, wettability or pseudocapacitance [3]. For

instance, oxygen groups and in aqueous electrolytes can increase the

wettability of the surface, improving the electrolyte-electrode interaction

and rendering a larger amount of the surface accessible for the formation

of the electrical double layer, and can also participate in reversible

faradaic reactions that contribute to the energy storage through

pseudocapacitance [4]. Likewise, the presence of nitrogen on the surface

Chapter 4

184

of activated carbons improves the performance of activated carbon as

electrodes of supercapacitors by increasing the wettability of the surface,

the electrochemical stability, the conductivity or the contribution of

pseudocapacitive processes [5-8]. Inversely to the ubiquity of surface

oxygen groups, nitrogen functionalities are not as easily introduced in

the structure of carbon materials, and the development of new

procedures for obtaining nitrogen-containing porous carbons has

therefore attracted a great interest.

Nitrogen-doped porous carbon materials can mainly be obtained by

two pathways: reaction of the carbon material with a nitrogen-containing

reagent (NH3, HCN, etc) or carbonization/activation of nitrogen-rich

carbon precursors (urea, polyaniline, etc.) [6,9]. However, these

approaches are usually carried out at high temperatures and, in the case

of post-treatments, these conditions strongly modify the porosity of the

pristine carbon material [10,11]. Also, the control of the nitrogen

functionalities that are formed during post-treatments is a challenge.

Therefore, post-modification treatments at low temperature are highly

desirable for obtaining different nitrogen functional groups while

preserving the characteristic porous texture of the original carbon

material.

In Chapter 3, the modification of the surface chemistry of a highly

porous activated carbon using organic chemistry reactions was proposed

[11]. Briefly, this treatment consisted in the oxidation of the carbon

material to introduce oxygen functionalities that, afterwards, were

converted into nitrogen functionalities through an amidation treatment.


185

In a last step, the post-conversion of the formed amides into amine

functional groups was achieved by a Hofmann rearrangement [12]. The

modification protocol allowed the incorporation of different nitrogen

functionalities. Interestingly, this method also leads to the formation of

pyridines, pyridones and pyrroles, which have a positive impact in the

electrochemical behavior of carbon materials [6,8], but are only

introduced using high temperature treatments. However, the oxidation

treatment carried out during the first step of the functionalization

protocol produced a remarkable decrease of the microporosity along

with the generation of non-desirable functional groups (CO2-evolving

groups) from the point of view of the capacitor performance. In this

chapter, a modification of this approach that allows the introduction of

nitrogen on carbon materials but preventing the previous incorporation

of oxygen functional groups is proposed. The electrochemical

performance of the pristine and the nitrogen-functionalized activated

carbons has been assessed as electrodes for supercapacitors.

4.2 Materials and methods


A highly microporous activated carbon prepared in our laboratory

has been used as the starting material for nitrogen incorporation via

organic chemical modification. The pristine material, henceforth named

KUA, has been obtained by chemical activation of a Spanish anthracite

(11 wt% of ash content) with KOH using an impregnation ratio of

activating agent to raw material of 4:1 and an activation temperature of

Chapter 4

186

750º C under inert atmosphere, which was held for 1 hour. More details

about the preparation process are available elsewhere [13].

4.2.2. Chemical functionalization of activated carbon

4.2.2.1. Synthesis of KUA-CONH2

In Chapter 3, an approach for the incorporation of amides over

activated carbon was satisfactorily developed [11]. Also, other nitrogen

functional groups were introduced on the surface of the carbon material

via secondary reactions. The amidation treatment was carried out as

follows:

(i) Chemical oxidation with HNO3 in order to introduce oxygen

functional groups on the surface of KUA [14].

(ii) Generation of acyl chloride functionalities by reaction of the

activated carbon obtained on step (i) with SOCl2 in toluene

under Ar atmosphere.

(iii) Reaction of the activated carbon obtained in step (ii) with 2M

NH4NO3/DMF solution and pyridine.

The obtained activated carbon was named KUA-CONH2. More

details about the preparation process are available in Chapter 3 [11].

4.2.2.2. Synthesis of KUA-N

In this case, step (iii) is directly carried out over activated carbon

KUA, in order to obtain nitrogen functional groups by using a single

treatment via reaction with the existing CO2 and CO-evolving groups in


187

the pristine material. In this step (iii), 400mg of KUA were added into

140 mL of 2M NH4NO3/DMF solution (activated carbon to solution

ratio of 1g/300mL) in a round bottom flask. Then, 140 mL of pyridine

were added slowly to the round bottom flask under continuous stirring at

room temperature. The mixture was stirred at 70 ºC for 65 hours. The

obtained sample (KUA-N) was washed with abundant water and ethanol,

filtered and dried at 100º C overnight.






surface area was obtained from N2 adsorption isotherms by using the

BET equation in the 0.05-0.20 range of relative pressures. The total

micropore volume was determined by Dubinin-Radushkevich (DR)

method applied to N2 (relative pressures from 0.01 to 0.05) adsorption

isotherms. The volume of the narrow microporosity (i.e., pore sizes

below 0.7 nm) was calculated from the DR method applied to the CO2

adsorption isotherms (relative pressures from 0.0001 to 0.25) [15]. The

pore size distribution for both samples has been calculated from the N2

adsorption isotherms using the 2D-NLDFT Heterogeneous surface

model [16] and by applying the Solution of Adsorption Integral Equation

Using Splines (SAIEUS, available online at http://www.nldft.com/)

Software.

Chapter 4

188

The surface chemistry of the samples was analyzed by X Ray

Photoelectron Spectroscopy (XPS) and Temperature Programmed

Desorption (TPD). XPS measurements were performed by using a VG-

Microtech Multilab 3000 spectrometer, equipped with an Al anode. The

deconvolution of N1s spectrum was carried out by using Gaussian

functions with 20% of Lorentzian component. FWHM of the peaks was

kept between 1.4 and 1.7 eV and a Shirley line was used for estimating

the background signal. TPD experiments were performed by heating the

samples (10 mg) to 950º C (at a heating rate of 20º C/min) under a

helium flow rate of 100 mL/min. The analyses were carried out by using

a TGA-DSC instrument (TA Instruments, SDT Q600 Simultaneous)

coupled to a mass spectrometer (Thermostar, Balzers, BSC 200).


4.2.4.1. Three electrode cell configuration


prepared by mixing the activated carbon with acetylene black and





thickness. After that, the electrode was attached to a gold disk used as a

current collector by using a conducting adhesive (colloidal graphite

suspension, Hitasol GA-715, Hitachi Chemical Co., Ltd.). The


189

electrodes were impregnated for 2 days into 1M H2SO4 previously to

electrochemical measurements.

The electrochemical characterization of the electrodes was

performed by cyclic voltammetry in a Biologic VSP multichannel

potentiostat and using a T-type Swagelock cell in a three-electrode

configuration. 1M H2SO4 was used as aqueous electrolyte. As counter

electrode, more than 20 mg of KUA was used. Both electrodes were

tightly pressed against each other and separated by a nylon membrane

(Teknokroma membrane filters, pore size: 320 nm). Ag/AgCl/KCl (3M)

was used as reference electrode in all cases. The electrochemical

characterization of all samples was tested by CV at sweep rates of 1 and

2 mV/s. CV capacitance was calculated from the area of the

voltammogram. The results are expressed in F/g, taking into account the

weight of the active material of the working electrode.

4.2.4.1. Two electrode cell configuration

Symmetric and asymmetric in mass capacitors were assembled for

all carbon materials. For symmetric capacitors, two electrodes (surface

area: 0.196 cm2) were prepared with a weight of ~1.3 mg (active phase)

each. The thickness of the electrodes was close to 0.2 mm in all cases. In

case of asymmetric in mass configuration, the mass of the electrodes was

determined following the procedures detailed by Peng et al [17], which

allows to charge the supercapacitor cells in expanded voltage with

respect to the symmetric configuration, taking full profit of the stability

potentials of each electrode. The electrodes were attached to a stainless

Chapter 4

190

steel collector by using a conducting adhesive. Supercapacitors were

constructed by pressing both electrodes against each other and

separating them by a nylon membrane filter (pore size: 320 nm). These

devices were characterized by CV at different scan rates, galvanostatic

charge-discharge (GCD) cycles at current densities from 0.25 to 32 A/g

and Electrochemical Impedance Spectroscopy (EIS) in 1M H2SO4

solution. Impedance spectra were measured at 0.05 V in the frequency

range of 10 mHz to 100 kHz with an amplitude voltage of 10 mV.

Autolab PGSTAT302 potentiostat was employed for EIS and CV

measurements and Arbin SCTS potentiostat for galvanostatic charge-

discharge cycles. A durability test for symmetric and asymmetric

capacitors was performed by 5000 galvanostatic charge-discharge cycles

at a current density of 1 A/g and a voltage of 1.2 V in case of symmetric

configuration and 1.4V for asymmetric cells. Current density and

specific capacitance is defined based on the total active weight of the

carbon material included in both electrodes.

The energy density and power density of symmetric and asymmetric

supercapacitors was calculated in order to obtain all relevant information

about their performance. Energy density was obtained during the

discharge cycle by the following equation (4.1):

E = ∫ 𝑉𝑑𝑄𝑄

0 (4.1)

Where V1 is the cell voltage of charge (1.2V for symmetric

configuration and 1.4V for asymmetric configuration) and V2 the voltage

of discharge (0 V in both cases).


191

Power density was calculated according to equation (4.2):

𝑃𝑚𝑎𝑥 =𝑉𝑚𝑎𝑥

2

4 𝐸𝑆𝑅 𝑚 (4.2)

Where ESR is the equivalent series resistance (determined from

ohmic drop in the charge–discharge cycles), m is the total active mass of

the electrodes and Vmax is the operating voltage.


4.2.1. Surface chemistry and porous texture characterization.

Table 4.1 shows the surface properties of the parent activated carbon

and those obtained after the functionalization treatments. The results of

the XPS analyses evidence that nitrogen atoms have been successfully

anchored through the proposed protocols on the surface of KUA-CONH2

and KUA-N. It should be noted that both functionalization treatments

have leaded to a similar nitrogen content (3.8 and 4.1 at.% for KUA-

CONH2 and KUA-N, respectively), while oxygen content increases

significantly for KUA-CONH2 sample. The changes in surface chemistry

on the sample KUA-CONH2 have been previously studied [11], and we

found that this increase in oxygen was related to the chemical oxidation

treatment carried out prior to amidation step. In case of KUA-N, TPD

measurements show a relevant decrease in the CO-evolving groups, with

a net total diminution (490 µmol /g) that is especially relevant in the

region of thermal decomposition of phenol and quinone-like groups.

However, XPS results evidence an increase of oxygen in both

functionalized activated carbons. It should be considered that this

Chapter 4

192

technique only measures the atomic composition of the most external

region of carbon particles, whereas TPD reflects the CO and CO2

evolving from the thermal decomposition of surface oxygen groups in

both the inner and the outer regions of carbon particles. Hence, TPD

provides more precise information about the overall changes of oxygen

functional groups during the functionalization treatment. The decrease in

oxygen surface groups determined by TPD indicates that part of the

introduced nitrogen groups has been attached to the surface by

consumption of these kind functional groups. These reactions may lead

to the formation of imines, pyridines and pyrroles [11]. The formation of

these functional groups is confirmed by N1s XPS deconvolution

obtained for KUA-N (Figure 4.1), which reveals the presence of

pyrroles/pyridones (400.4 ± 0.2 eV), imines and pyridines (398.8 ± 0.2

eV) [18-21]. Also, other functionalities, such as quaternary nitrogen

(401.8 ± 0.2 eV) [22], amines, amides or cyclic amides (399.6 ± 0.2 eV)

[20], are formed on the surface of this carbon material. The absence of

chemisorbed pyridine after both functionalization treatments was

confirmed by immersing the pristine carbon in pyridine, and following

this step by analogous washing and drying processes carried out for the

synthesis of KUA-N and KUA-CONH2. XPS analysis revealed nitrogen

content of 0.4 at. %, that is negligible in comparison with the nitrogen

content of the parent activated carbon.


193

Table 4.1. Textural properties and elemental surface composition (XPS and TPD) for

the pristine and modified activated carbons.

Sample NXPS

(at.%)

OXPS

(at.%)

CO2 TPD

(µmol/g)

COTPD

(µmol/g)

SBET

(m2/g)

VDRN2

(cm3/g)

VDRCO2

(cm3/g)

KUA 0.3 8.8 500 2100 3460 1.19 0.56

KUA-N 4.1 11.1 700 1610 3450 1.19 0.55

KUA-CONH2 3.8 12.0 1030 2490 2160 0.74 0.44

Figure 4.1. N1s XPS deconvolution of KUA-N.

Figure 4.2 shows the nitrogen adsorption-desorption isotherms of

the pristine and the N-containing activated carbons and the calculated

PSD. It can be seen that the generation of functional groups on KUA-

CONH2 produces a large decrease of the nitrogen uptake, which means

that a noticeable part of the microporosity is blocked (Figure 2b),

producing a large drop of ca. 35% of the surface area and micropore

volume, Table 4.1. The blockage of the porosity has been related to the

damaging effect produced by the oxidation treatment on the

microporosity [11]. However, when the proposed functionalization

396397398399400401402403404405

I (a

.u.)

Binding Energy (eV)

Chapter 4

194

treatment is directly applied over the pristine material (obtaining KUA-

N), the nitrogen adsorption isotherm superimposes with that from the

pristine sample, showing that the porosity of the KUA sample does not

change after the amidation treatment (KUA-N sample) due to the mild

conditions used (Figure 4.2a).

Figure 4.2. (a) N2 adsorption-desorption isotherms of the activated carbons KUA,

KUA-N and KUA-CONH2. (b) Pore-size distributions of KUA, KUA-N and KUA-

CONH2 by DFT calculations.

0

200

400

600

800

1000

1200

0.0 0.2 0.4 0.6 0.8 1.0

Ad

sorb

ed

vo

lum

e (

cm

3/g

)

P/P0

KUA

KUA-N

KUA-CONH₂

(a)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 1 2 3 4 5

Po

re v

olu

me (

cm

3/g

nm

)

Pore width (nm)

KUA

KUA-N

KUA-CONH₂

(b)


195

The minor differences found in the PSD (Figure 4.2b) are due to

small discrepancies within the NL-DFT fitting of both isotherms. Thus,

this pathway seems to be appropriate for functionalization of carbon

materials with a high microporosity development, since it allows the

incorporation of a high amount of nitrogen functionalities (4.1 at.%

XPS) along with the preservation of porous texture.

4.3.2 Electrochemical characterization

4.3.2.1 Characterization of carbon materials

The electrochemical performance of all carbon materials has been

studied in a three electrode cell configuration by cyclic voltammetry.

Figure 4.3 shows the second cyclic voltammograms for all electrodes at

a potential range where they are stable (in the absence of degradation

reactions). The gravimetric capacitance for all the carbon materials has

been determined from CV measurements and is shown in Table 2. It can

be observed that the gravimetric capacitance is mainly related to the

specific area of the sample, with KUA and KUA-N showing the largest

capacitance due to its largest apparent surface area, while KUA-CONH2

displays the lowest capacitance due to the decrease in porosity after the

functionalization treatment.

Although sample KUA-N has a capacitance quite close to sample

KUA, there are interesting differences in the shape of the

voltammogram. Taking into account that this functionalization treatment

does not change the apparent surface area of the pristine activated

carbon, these differences are related to the changes on the surface

Chapter 4

196

chemistry produced by the modification protocol. Thus, KUA shows two

redox processes at around 0.2V and 0.4V, that can be related to the

presence of electroactive surface oxygen groups [14], but KUA-N does

not present these redox processes, and the electrochemical behavior

shown by this electrode seems to be exclusively related to the formation

of electrical double layer on the inner surface of this carbon material

(Figure 4.3). This is in agreement with the characterization of the

materials, since both have similar porous texture (Table 4.1 and Figure

4.1), but sample KUA-N has a lower content of CO-groups (Table 4.1)

which are the electroactive oxygen functional groups in acid medium

[4].

Figure 4.3. 2nd cyclic voltammograms in the potential range between 0V and 0.6V for

KUA, KUA-CONH2 and KUA-N electrodes. 1M H2SO4. v=1 mV/s.

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

j (A

/g)


KUA KUA-N KUA-CONH₂


197

Table 4.2. Gravimetric capacitance (Cg) determined for KUA, KUA-CONH2 and

KUA-N electrodes in the potential range between 0V and 0.6V by cyclic voltammetry.

1M H2SO4. v=1 mV/s.

Sample Cg (F/g)

KUA 256

KUA-N 229

KUA-CONH2 200

These changes related to the absence of electroactive oxygen

functional groups in acid medium may affect the stability of the

electrode. Oxygen functional groups can often be beneficial for the

electrochemical performance of carbon materials since they can provide

pseudocapacitance [4]; however, when these materials reach positive

potentials (>0.8V), some prejudicial functionalities, as the CO2-type

groups, which are the starting point in degradation mechanism of carbon

electrodes [23], can be formed. For these reasons, the potential range of

stability of KUA and KUA-N have been determined in order to assess

the effect of electroactive oxygen groups on the electrochemical stability

of these carbon materials. Figure 4.4a shows the cyclic voltammograms

for KUA and KUA-N in potential range between -0.5V and 0.6V. It is

clearly seen that, beyond the differences already found in Figure 4.3,

both electrodes present a quite similar electrochemical behavior down to

-0.2 V. However, KUA presents a broad peak at -0.35V in the negative

scan that can be related to the reduction of oxygen functionalities. This

peak does not appear in the case of KUA-N, evidencing the conversion

of electroactive oxygen groups into different species. Figure 4.4b shows

the third cyclic voltammograms for KUA and KUA-N in the potential

Chapter 4

198

range between 0V to 1.0V. Again, both electrodes reveal similar

performance at positive potentials evidenced by the oxidation current

happening at 0.9V. However, the effect of the electroxidation on the

materials is different; in case of KUA, the redox processes at around

0.4V resulting from electrooxidation are more noticeable, mainly in the

reduction one, that indicates an increase of surface oxygen groups during

the positive scan, being this broad reduction peak lower in the case

KUA-N activated carbon. Given that the electrochemical degradation of

porous carbon materials in aqueous electrolytes seems to be related to

the formation of oxidized species over the surface of the electrodes when

submitted to positive potentials [24], this is a promising result in terms

of a possible increase in the durability of porous carbon electrodes to

positive potentials.

Figure 4.4. 3rd cyclic voltammograms for KUA and KUA-N electrodes in the potential

range (a) -0.5-0.6V and (b) 0-1.0V. 1M H2SO4. v=2 mV/s.

-3

-2

-1

0

1

2

-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6

j (A

/g)


KUA

KUA-N

(a)

-1.5

-0.5

0.5

1.5

2.5

-0.1 0.1 0.3 0.5 0.7 0.9 1.1

j (A

/g)


KUA

KUA-N

(b)


199

4.3.2.2. Characterization of symmetric supercapacitors

In order to study in detail the effect of surface chemistry on the

performance of these materials from an application point of view,

symmetric supercapacitors for the samples KUA, KUA-N and KUA-

CONH2 have been studied.

Figure 4.5a shows the Nyquist plot obtained for all symmetric

capacitors. At very high frequencies (points closer to the Y-axis), the

capacitors behave like a resistance, while at low frequencies, the

imaginary part of the resistance sharply increases and a curve close to a

vertical line is seen, indicative of a pure capacitive behavior [25]. In the

middle frequency domain, the influence of the electrode porosity and

conductivity can be seen. The resistance of the cell can also be obtained

from the X-axis intercept of the straight line observed at low frequencies

of the diagram (the region where capacitive behavior of the capacitors is

observed [26]) for all cases. The cell resistance is contributed by the

resistance of charge transfer through the carbon particles and grain-

boundaries at the electrode-electrolyte surface (represented by a larger

diameter of the semicircle with increasing the resistance [27]) and by

diffusive problems of ions within the pore system (indicated by the

appearance of a 45º impedance line, also known as the Warburg region,

after the end of the semicircle and prior to the formation of the vertical

line characteristic of capacitive behavior [25]). As can be seen, KUA-

CONH2 evidences the largest resistance (~3.8 Ω). This can be explained

by the presence of a large amount of carboxylic acids and their cyclic

species (Table 4.1), produced prior to the generation of nitrogen

Chapter 4

200

functional groups, that have an electron-withdrawing character and

produce an increase of the resistance of the carbon material. The clear

Warburg region observed for this sample is a consequence of the

changes in porosity and the increased concentration in surface oxygen

groups that hinder the mobility of ions within this pore network. KUA-N

shows an even lower internal resistance than KUA. Since both samples

have similar porous texture and close surface oxygen concentration

(Table 4.1) this difference cannot be explained by diffusional problems

along the porosity. Then, the improvement found for KUA-N should be

related to the generation of electron-donor nitrogen functionalities [8] or

to the decrease of detrimental electron-withdrawing oxygen groups

improving the conductivity of the material. This positive effect of the

conversion of oxygen groups into nitrogen functionalities is not

observed in KUA-CONH2 due to the previous generation of a large

amount of carboxylic acids during the oxidation treatment with nitric

acid.

Table 4.3. Gravimetric capacitance (Cg), energy density (E), maximum power density

(P) and energy efficiency determined for KUA, KUA-CONH2 and KUA-N symmetric

supercapacitors at the voltage 1.2V by galvanostatic charge-discharge cycles. 1M

H2SO4. j= 1A/g.

Sample Cg

(F/g)

E

(Wh/kg)

Pmax

(kW/kg)

Energy

efficiency (%)

KUA 63 10.3 43.3 74

KUA-N 59 10.5 72.4 79

KUA-CONH2 53 8.7 20.0 75


201

Figure 4.5. (a) Nyquist plot for the symmetric supercapacitors KUA (black), KUA-N

(blue) and KUA-CONH2 (red). (b) Ragone plot at 1.2V for all symmetric

supercapacitors. Galvanostatic charge-discharge cycles for the samples at (c) 1A/g and

(d) 16 A/g. 1 M H2SO4 solution.

These differences on the resistance of the carbon material play a

major role in their behavior as electrodes for supercapacitors, since the

performance of supercapacitors at high power density depends on the

rate of charge and discharge, which is faster when the overall resistance

is lower. This explains the best performance observed for KUA-N in the

Ragone plot (measured using GCD experiments at 1.2V under different

specific currents) shown in Figure 4.5b. This supercapacitor keeps larger

energy density at high power density due to its lower resistance, and

provides a maximum power density of 72.4 kW/kg (Table 4.3). As

0

1

2

3

4

5

0 1 2 3 4 5

-Z''

(Ω

)

Z' (Ω)

(a)

1

10

0.1 1 10 100

E (

Wh

/Kg

)

P (kW/Kg)

(b)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 50 100 150

Volt

age (

V)

Time (s)

(c)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 2 4 6

Volt

age (

V)

Time (s)

(d)

Chapter 4

202

expected, KUA shows similar energy density and capacitance at low

power density (10.3 Wh/kg for KUA and 10.5 Wh/kg for KUA-N at 0.5

kW/kg, and gravimetric capacitances of 63 F/g vs 59 F/g, respectively)

but is not able to keep these values when power demand is high due to

its larger internal resistance. This is confirmed in Figure 4.5c and Figure

4.5d, where galvanostatic charge-discharge profiles are shown at

different current densities. As can be observed in Figure 4.5c, at 1 A/g,

KUA and KUA-N based symmetric capacitors show similar capacitance

(Table 4.3) (which is derived from the discharge time of the profiles)

due to its similar apparent surface area as was pointed out in section

4.3.2.1. Although KUA shows a slightly higher capacitance, KUA-N

shows better performance as supercapacitor due to the lower presence of

electroactive species, which make more efficient the charge-discharge

process. This can be noticed in the shape of the profile, which is closer

to the ideal triangular shape in the case of KUA-N, and also in the

energy efficiency (i.e. the ratio between the energy employed in

charging the cell and the energy recovered upon discharge of the cell) of

the KUA-N capacitor (Table 4.3). Also, at high current density (16 A/g),

Figure 4.5d, the best performance of KUA-N is pointed out by the lower

ohmic drop produced during the transition of charge to discharge in the

GCD profile, which provides a larger capacitance retention at high

current. KUA-CONH2 based capacitor evidences lower capacitance and

energy density than the other supercapacitors as consequence of its

lower porosity. Also, its maximum power density is the lowest (Table

4.3) due to the detrimental effect of CO2-evolving groups on the


203

resistance of the carbon material (Figure 4.1a) that produces a larger

ohmic drop and, consequently, a loss of power density.

The effect of nitrogen functionalities upon the durability of these

capacitors has been assessed by carrying out 5000 cycles of

galvanostatic charge-discharge at 1 A/g and 1.2V of cut-off voltage.

Figure 4.6 shows the evolution of gravimetric capacitance along the

whole experiment. All the capacitor cells behaved adequately upon the

test, showing a capacitance retention higher than 92%. The better

electrochemical stability of a nitrogen-containing activated carbon

obtained by carbonization of PANI-functionalized KUA in organic

media has been recently demonstrated [6], and seems to be also found in

aqueous electrolyte. Given the possibility of achieving a larger stability

in these materials, asymmetric in mass capacitors have been constructed

from these carbon materials.

Figure 4.6. Cyclability test for KUA, KUA-N and KUA-CONH2 supercapacitors at

1.2V. 1A/g. 5000 cycles. 1 M H2SO4 solution.

0

10

20

30

40

50

60

70

80

90

100

0 1000 2000 3000 4000 5000

Cg (

F/g

)

Nº Cycles

KUA

KUA-N

KUA-CONH₂

Chapter 4

204

4.3.2.3. Characterization of asymmetric supercapacitors

Asymmetric in mass supercapacitors, where the cells are set up with

the same carbon material as positive and negative electrode, but with

different weights, have been obtained using the method proposed by

Peng et al. [17]. Briefly, this method consists in (i) determining the

working potential windows (∆𝑽+ and ∆𝑽−) of electrochemical stability

for the positive and negative electrodes from the open circuit potential

(EOCP), (ii) determine the capacitances (Cg+ and Cg-) in the established

potential ranges, (iii) determine the optimum weight ratio (𝒘+ 𝒘−⁄ ) of

the electrodes by equaling the charge stored in each electrode while

fixing the values of the potential windows to those determined in (i). The

expression used in the last step reads as follows:

𝒘+ 𝒘−⁄ = 𝑪𝒈−· |∆𝑽−| (𝑪𝒈+

· ∆𝑽+)⁄ (4.3)

Then, by using the capacitance of each electrode in Eqn. (4.3), the

mass ratio for the asymmetric supercapacitors was calculated to be 1.31,

1.12 and 1.02 for KUA, KUA-N and KUA-CONH2, respectively. Table

4.4 shows the capacitances and potential window values, the weight of

the electrodes employed for the construction of each capacitor cell and

the open circuit potential of each carbon material. It is important to note

that the sum of the potential windows of each electrode corresponds to

the working voltage of the capacitor, and that the three cells will be able

to operate up to 1.4 V. Such a high voltage is expected to increase the

energy density of the capacitors, but will also expose the capacitor to


205

harsh conditions, where the possible influence of nitrogen functionalities

could be observed.

Table 4.4. Parameters employed for the design of asymmetric capacitors, and open

circuit potential for each carbon material. 1M H2SO4.

Supercapacitor 𝑪𝒈+

(F/g)

𝑪𝒈−

(F/g)

∆𝑽+

(V)

∆𝑽−

(V)

𝒘+

(mg)

𝒘−

(mg)

EOCP

(V)

KUA 263 298 0.65 0.75 1.30 1.05 0.25

KUA-N 252 283 0.70 0.70 1.22 1.08 0.20

KUA-CONH2 213 199 0.67 0.73 1.18 1.14 0.23

Figure 4.7a shows galvanostatic charge-discharge cycles obtained

for the asymmetric supercapacitors at 1 A/g. In this case, the capacitance

obtained for all asymmetric supercapacitors follows the same order as

for the symmetric capacitors: KUA > KUA-N>> KUA-CONH2.

However, KUA-N based capacitor provides the lower ohmic drop due to

the lower electrical resistance of this carbon material, which provides the

highest power density. Again, a quasi-triangular shape profile is

observed in the case of KUA-N based capacitor. Figure 4.7.b shows the

Ragone plot obtained for all asymmetric supercapacitors working at

1.4V. As in the case of symmetric devices (section 3.2.2), KUA and

KUA-N based capacitors evidence largest energy density than KUA-

CONH2 (Table 4.5) due to their larger apparent surface area. The huge

potential of the asymmetric in mass design for enhancing energy storage

can be observed when the energy densities achieved in this configuration

and those obtained in the symmetric design are compared (Table 4.3). It

is important to note that KUA-N based capacitor shows a slightly higher

energy density than KUA in spite of having lower capacitance, what

Chapter 4

206

indicates again the better performance of this material as electrode for

supercapacitor. In addition, KUA-N based asymmetric capacitor is able

to maintain the superior power density that has already shown in

symmetric configuration.

Figure 4.7. (a) Galvanostatic charge-discharge cycles for KUA, KUA-N and KUA-

CONH2 asymmetric supercapacitors at 1.4V and 1 A/g. (b) Ragone plot at 1.4V for all

asymmetric supercapacitors.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 50 100 150 200 250

Vo

lta

ge (

V)

Time (s)

KUA

KUA-N

KUA-CONH₂

(a)

1

10

100

0.1 1 10

E (

Wh

/Kg

)

P (KW/Kg)

KUA

KUA-N

KUA-CONH₂

(b)


207

Table 4.5. Gravimetric capacitance (Cg), energy density (E), power density (P),

coulombic efficiency and energy efficiency determined for KUA, KUA-CONH2 and

KUA-N asymmetric supercapacitors at the voltage 1.4V by galvanostatic charge-

discharge cycles. 1M H2SO4. j=1A/g.

Sample Cg

(F/g)

E

(Wh/Kg)

Pmax

(kW/kg)

Coulombic

efficiency

(%)

Energy

efficiency

(%)

KUA 72 14.1 52.5 90 55

KUA-N 67 14.5 61.2 99 66

KUA-CONH2 41 8.3 31.6 98 62

The most remarkable result is the improvement in coulombic and

energy efficiency observed in the case of those supercapacitors

constructed with nitrogen-functionalized activated carbon as electrodes

(KUA-N and KUA-CONH2), Table 4.5. Coulombic efficiency is under

the allowable limit in the case of KUA, and points out that either

massive electrolyte decomposition or carbon degradation is occurring in

this capacitor. On the other hand, the coulombic efficiency is close to

100% in the case of the N-containing carbon capacitors. This fact clearly

evidences that the presence on the surface of nitrogen functional groups

can avoid the irreversible faradaic reactions that occur at the surface of

carbon materials in the positive electrode and that are connected to the

malfunctioning of aqueous capacitor cells. More striking results can be

seen on the durability test conducted over these capacitors, Figure 4.8a.

The profile of capacitance for KUA-N and KUA-CONH2 based

capacitors is similar during the GCD test at 1 A/g for 5000 cycles, and

when compared to that of the corresponding symmetric capacitors

(Figure 4.6), a slightly higher capacitance drop can be seen in this case,

providing a capacitance retention of 83 and 85%, respectively. This is an

Chapter 4

208

expected result that can be explained by the higher cut-off voltage of

these capacitors, which forces both electrodes to work closer to its

electrochemical stability limits. On the other hand, the performance of

the cell assembled using KUA shows some oscillations during the

duration of the test, rendering unreliable the KUA asymmetric capacitor.

This behavior is probably connected to the evolution of gases in this cell

during the test. The absence of such problems in the capacitors

constructed using N-containing carbons undoubtedly speaks about the

impact of nitrogen groups upon the electrochemical stability of activated

carbons in aqueous media. However, after around 3000 cycles, the

capacitance stabilizes.

Ragone plot of asymmetric capacitors was measured again after

durability test, Figure 4.8b. These results evidence that KUA-N and

KUA-CONH2 based asymmetric supercapacitors keep most of their

original performance, since they provide high energy densities at high

power density. However, KUA based asymmetric capacitor shows an

overall decrease of energy density, which is especially large at high

power density. It should be noted that, during the last 2000 cycles of

durability test, the gravimetric capacitance determined for KUA is

slightly larger than that of KUA-N. However, KUA-N provides higher

energy density (Figure 4.8b) evidencing again the importance of the

energy efficiency achieved during the charge-discharge processes (Table

4.6). The generation of detrimental functional groups on KUA during

cycles along with irreversible reactions occurring on the electrodes in

this supercapacitor decreases coulombic and energy efficiencies,


209

explaining why KUA has a lower energy density than KUA-N even

when its gravimetric capacitance is higher. Also, a decrease of maximum

power of 62% is attained for KUA capacitor, which is a much severe

drop than those obtained for KUA-N and KUA-CONH2 based

asymmetric capacitors (37 and 14%, respectively, Tables 4.5 and 4.6).

This loss of power density confirms the improved stability of nitrogen-

containing activated carbons as electrodes for asymmetric

supercapacitors in aqueous electrolyte.

Figure 4.8. (a) Durability test for KUA, KUA-N and KUA-CONH2 based asymmetric

capacitors at 1.4V, specific current of 1A/g and 5000 cycles. (b) Ragone plot obtained

after durability test for KUA, KUA-N and KUA-CONH2 based supercapacitors.

0

20

40

60

80

100

0 1000 2000 3000 4000 5000

Cg

(F

/g)

Nº Cycles

KUA

KUA-N

KUA-CONH₂

(a)

1

10

100

0.1 1 10

E (

Wh

/kg

)

P (kW/kg)

KUAKUA-N

KUA-CONH₂

(b)

Chapter 4

210

Table 4.6. Energy density (E), power density (P), and energy efficiency determined for

KUA, KUA-CONH2 and KUA-N asymmetric supercapacitors after durability test at

the voltage 1.4V by galvanostatic charge-discharge cycles. 1M H2SO4. j=4A/g.

Sample Cg

(F/g)

E

(Wh/k

g)

Pmax

(kW/kg

)

Coulombic

efficiency

(%)

Energy

efficiency

(%)

KUA 32.2 5.0 19.7 98 45

KUA-N 44.3 8.4 38.9 99 60

KUA-CONH2 23.9 3.9 27.1 98 49

4.4. Conclusions

Chemical functionalization of an activated carbon with a high

surface area was carried out by following two different pathways: (i)

amidation of an activated carbon previously oxidized (KUA-CONH2)

and (ii) direct treatment of the pristine activated carbon with the same

nitrogen reagents (KUA-N). Both treatments lead to a similar content of

nitrogen species (~4 at. % XPS). The amidation treatment for generation

of KUA-CONH2 produces a decrease of 30% of microporosity due to

the detrimental effect of the oxidation step. In case of KUA-N, this step

is not carried out and consequently the obtained carbon material

preserves 100% of the porous texture of the pristine carbon material,

making this single-step treatment more appropriate for functionalization

of highly microporous carbon materials, which is applied at mild

conditions.

The surface chemistry and porous texture of these carbon materials

clearly influence the electrochemical performance of these carbon

materials. KUA-CONH2 showed the lower capacitance due to its

decrease in porosity. In addition, KUA-N evidences a slight decrease of


211

capacitance due to the removal of electroactive oxygen groups. On the

other hand, impedance spectroscopy showed that the presence of

nitrogen functionalities in KUA-N promoted the conductivity of the

electrode, facilitate ion diffusion and electrolyte polarization. The effect

of these changes on the electrochemical performance of these carbon

materials was thoroughly analyzed by studying their performance as

electrodes of symmetric and asymmetric supercapacitors. KUA-CONH2

based capacitors showed the lower energy density along with a lower

power density as consequence of the presence of a larger amount of

CO2-evolving groups, which strongly affect the conductivity of the

carbon material and the whole device. However, KUA-N based

capacitors showed similar energy density, better capacitance retention

and higher power density than the pristine activated carbon as

consequence of the beneficial effect of the generation of electron-donor

nitrogen groups and the removal of detrimental electroactive oxygen

functionalities, producing a supercapacitor with higher energy and

coulombic efficiency. Also, both nitrogen-functionalized activated

carbons revealed higher stability than the pristine activated carbon,

which is not able to safely operate at 1.4V voltage in aqueous medium.

Since no changes have been made in the textural properties of the parent

carbon, the improvements found for KUA-N based capacitors are

undoubtedly related to the presence of stable nitrogen functionalities in

the activated carbon.

Chapter 4

212

4.5. References

[1] P. Simon, Y. Gogotsi. Materials for electrochemical capacitors,

Nat. Mater. 7 (2008) 845–854.

[2] F. Béguin, E. Frackowiak, editors. Carbons for Electrochemical

Energy Storage and Conversion Systems. 1st ed. CRC Press;

2009.

[3] T.J. Bandosz, C.O. Ania. Surface chemistry of activated carbons

and its characterization. In: T.J. Bandosz, editor. Act. carbon

surfaces Environ. Remediat. 1st ed., Elsevier; 2006, p. 159–229.


Morallón, D. Cazorla-Amorós, A. Linares-Solano. Role of surface




Zhu, G.Q. Lu. Nitrogen-Enriched Nonporous Carbon Electrodes


(2009) 1800–9.

[6] D. Salinas-Torres, S. Shiraish, E. Morallón, D. Cazorla-Amorós.




[7] C.-T. Hsieh, H. Teng, W.-Y. Chen, Y.-S. Cheng. Synthesis,


functionalized carbon nanotube/carbon paper electrodes. Carbon

48 (2010) 4219–4229.


Amorós, J. Geng, et al. On the origin of the high capacitance of

nitrogen-containing carbon nanotubes in acidic and alkaline

electrolytes, Chem. Commun. 50 (2014) 11343–11346.

[9] W. Shen, W. Fan. Nitrogen-containing porous carbons: synthesis

and application, J Mater. Chem. A 1 (2013) 999-1013.


213

[10] H. Tamai, K. Shiraki, T. Shiono, H. Yasuda. Surface


carbons by immobilization of diamine. J Colloid Interface Sci.

292 (2006) 299–302.

[11] M.J. Mostazo-López, R. Ruiz-Rosas, E. Morallón, D. Cazorla-

Amorós. Generation of nitrogen functionalities on activated

carbons by amidation reactions and Hofmann rearrangement:

Chemical and electrochemical characterization. Carbon 91 (2015)

252–65.


García, L. Licea-Jiménez, et al. Covalent amino-functionalisation

of single-wall carbon nanotubes. J Mater. Chem. 15 (2005) 3334–

3339.


Linares-Solano. Preparation of activated carbons from Spanish

anthracite, Carbon 39 (2001) 741-749.


Cazorla-Amorós, A. Linares-Solano. Chemical and


Carbon 44 (2006) 2642–2651.

[15] D. Cazorla-Amorós, J. Alcañiz-Monge, M.A. De La Casa-Lillo

A.Linares-Solano. CO2 as an adsorptive to characterize carbon

molecular sieves and activated carbons. Langmuir 14 (1998)

4589–4596.

[16] J. Jagiello, J.P. Olivier. 2D-NLDFT adsorption models for carbon

slit-shaped pores with surface energetical heterogeneity and

geometrical corrugation. Carbon 55 (2013) 70–80.

[17] C. Peng, S. Zhang, X. Zhou, G.Z. Chen. Unequalisation of


asymmetrical supercapacitors. Energy Environ. Sci. 3 (2010)

1499-1502.

[18] E. Raymundo-Piñero, D. Cazorla-Amorós, A. Linares-Solano.


Chapter 4

214


fibres. Carbon 41 (2003) 1925–1932.


Find, U. Wild, R. Schlögl. Structural characterization of N-


point petroleum pitch and a melamine resin, Carbon 40 (2002)

597–608.

[20] R.J.J. Jansen, H. van Bekkum. XPS of nitrogen-containing


1027.

[21] Y. Yamada, J. Kim, S. Matsuo, S. Sato. Nitrogen-containing


70 (2014) 59–74.

[22] F. Kapteijn, J.A. Moulijn, S. Matzner, H.-P. Boehm. The

development of nitrogen functionality in model chars during

gasification in CO2 and O2, Carbon 37 (1999) 1143–1150.

[23] R. Berenguer, R. Ruiz-Rosas, A. Gallardo, D. Cazorla-Amorós, E.

Morallón, H. Nishihara, et al. Enhanced electro-oxidation

resistance of carbon electrodes induced by phosphorus surface

groups, Carbon 95 (2015) 681–689.

[24] P. Su-Il, L. Eung-Jo, K. Tae-Young, L. Seo-Jae, R. Young-

Gyoon, K. Chang-Soo. Role of surface oxides in corrosion of

carbon black in phosphoric acid solution at elevated temperature,

Carbon 32 (1994) 155–9.

[25] R. Kötz, M. Carlen. Principles and applications of electrochemical

capacitors. Electrochim. Acta 45 (2000) 2483–98.

[26] B.E. Conway. Electrochemical supercapacitors: Scientific

Fundamentals and Technological Applications. New York:

Springer; 1999.

[27] S. Fletcher, V.J. Black, I. Kirkpatrick. A universal equivalent

circuit for carbon-based supercapacitors. J. Solid State

Electrochem 18 (2014) 1377–1387.

Chapter 5

Electrochemical performance

of N-doped zeolite templated

carbon in aqueous electrolyte

Electrochemical performance of N-ZTC in aqueous electrolyte

217

5.1. Introduction

Supercapacitors are one of the most relevant energy storage devices

due to their extraordinary power density, which is much higher than that

provided by other systems, such as batteries and fuel cells. These

devices can be composed by different electrode materials (carbon

materials, conducting polymers and metal oxides) and electrolytes

(aqueous, organic and ionic liquids). However, the development of the

greener, safer and cheaper aqueous-based supercapacitors is severely

limited by their low operation voltage. In consequence, their energy

density is lower than that showed by those supercapacitors based on

organic or ionic liquid based electrolytes [1-3].

Most of the strategies for overcoming this limitation are focused on

the development of electrode materials with high electrochemical

stability and capacitance. Among all the possibilities, the development

of ultraporous carbon materials with well-developed microporosity and

ordered structure is highly desirable. In this sense, zeolite-templated

carbons (ZTC) evidence a unique 3D-ordered microporous structure that

shows extremely profitable properties for this application [4,5]. First,

they have an outstanding apparent surface area and an ordered structure

with uniform pore size distribution (1.2 nm) that maximizes the

formation of the electrical double layer and improves the mobility of

ions within the porosity, making them especially interesting as electrode

materials for supercapacitors. Moreover, a large amount of highly

reactive edge sites are present in their structure, which eases the

incorporation of electroactive oxygen functionalities that produce an

Chapter 5

218

extraordinary pseudocapacitance boost in acid media [6]. However,

these materials can be easily overoxidized, leading to the degradation of

the structure and the loss of their unique properties.

The electrochemical performance of carbon materials can be tuned

by the introduction of heteroatoms (such as O, N, P or B) [7-16].

Concretely, nitrogen functional groups have been found to modify the

electrochemical stability, electrical conductivity, the wettability and

pseudocapacitance of carbon materials [8, 17-20]. Nevertheless, the role

of the different N functionalities still is not fully understood. Nitrogen-

doped carbons can be synthesized following two different strategies: (i)

by using a nitrogen-containing precursor as source or (ii) through post-

treatments of a carbon material with nitrogen reagents [21,22]. However,

these approaches usually require the use of high temperatures that would

damage the fragile porous network of ZTCs. Hence, a plausible strategy

for obtaining N-doped ZTC while preserving its unique structure would

rely on the use of a nitrogen-containing gas as a raw precursor.

Following this premise, Kyotani et al [7, 23, 24] synthesized N-ZTC by

using acetonitrile as CVD precursor. By careful combination of the CVD

conditions and the template, they were able to obtain a nitrogen-rich

microporous ZTC with a similar structure of non-doped ZTC, while

showing a larger microporosity than that found for other N-doped ZTCs

in the literature [25,26]. The advantageous ordered structure of the

resulting material was profited for achieving a better understanding of

the effect of N-doping in the electrochemical performance of carbon

materials in organic electrolyte [27], allowing to use them as model


219

materials for understanding the effect of surface chemistry in

electrochemical properties and supercapacitor performance. However,

the effect of nitrogen doping on the electrochemical performance of

highly microporous N-doped ZTC in aqueous electrolyte and in

supercapacitors application, has not been assessed before in the

literature.

In this chapter, the differences on the performance of N-doped and

non-doped zeolite templated carbons through physicochemical and

electrochemical characterization are shown. The effect of nitrogen

functional groups on the electrochemical behavior of these carbons in

two different electrolytes (1 M H2SO4 and 0.5 M KOH) is thoroughly

analyzed by different techniques. Also, their performance as electrodes

for supercapacitors is studied and related to the functional groups formed

on each zeolite template carbon.


5.2.1. Zeolite templated carbons

Zeolite templated carbons were synthesized by chemical vapor

deposition of different precursors and using zeolite Y as a template (Na-

form, SiO2/Al2O3 = 5.6, obtained from Tosoh Co. Ltd.) by following the

method reported elsewhere [6, 28, 29]. As a result, two different ZTCs

were obtained: non-doped zeolite templated carbon (ZTC) and N-doped

zeolite templated carbon (N-ZTC) [23].

Chapter 5

220

5.2.2 Physicochemical characterization

The structure of the samples was analyzed by X-ray diffraction

(XRD) recorded on Shimadzu XRD-6100 instrument with Cu-K

radiation. The porous texture was characterized by N2 physisorption

technique carried out at -196 °C, by using a volumetric sorption analyzer

(BEL Japan, BELSORP-max). The apparent specific surface area was

calculated by the Brunauer–Emmett–Teller method (SBET) using the N2

adsorption data in the relative pressure (P/P0) range of 0.01–0.05. The

total micropore volume was calculated from the Dubinin–Radushkevich

equation (VDRN2).



Desorption (TPD). XPS analyses were performed by using a VG-

Microtech Multilab 3000 spectrometer with an Al anode. N1s spectra

were deconvoluted by using Gaussian functions with 20% of Lorentzian

component. A Shirley line was used as background and the FWHM of

the peaks was kept between 1.4 and 1.7 eV. TPD experiments were

carried out by using a TGA-DSC instrument (TA Instruments, SDT

Q600 Simultaneous) coupled to a mass spectrometer (Thermostar,

Balzers, BSC 200), by heating the samples (10 mg) up to 950 ºC

(heating rate: 20 ºC/min) under helium atmosphere (flow rate: 100

mL/min).


221


5.2.3.1 Three electrode cell configuration


prepared by mixing the carbon material with acetylene black and





thickness. After that, the electrode was attached to a gold current

collector by means of conducting adhesive (colloidal graphite

suspension, Hitasol GA-715, Hitachi Chemical Co., Ltd.). The

electrodes were soaked for 2 days into the electrolyte (1M H2SO4 or

0.5M KOH) previously to electrochemical measurements.

The electrochemical characterization of the electrodes was

performed by cyclic voltammetry in a Biologic VSP multichannel

potentiostat and using a T-type Swagelock cell in a three-electrode

configuration. 1M H2SO4 was used as aqueous electrolyte. A capacitive

electrode with more than twice the capacitance (electrode weight per

gravimetric capacitance) of the working electrode was prepared from a

commercial activated carbon and used as counter electrode. Both

electrodes were tightly pressed against each other and separated by a

nylon membrane (Teknokroma membrane filters, pore size: 450 nm).

Ag/AgCl/KCl (3M) was used as reference electrode in all cases. The

electrochemical characterization of all samples was studied by cyclic

Chapter 5

222

voltammetry (CV) at sweep rate of 2 mV/s. CV capacitance was

calculated from the area of the voltammogram. The results are expressed

in F/g, taking into account the weight of the active material of the

working electrode. The materials were also characterized by

Electrochemical Impedance Spectroscopy (EIS). Impedance spectra

were measured at 0.3 V vs Ag/AgCl/KCl in the frequency range of 10

mHz to 100 kHz with an amplitude voltage of 10 mV.

5.2.3.2 Two electrode cell configuration

Symmetric capacitors (in mass) were assembled for both carbon

materials. The electrodes were prepared by using the method described

above (section 5.2.3.1) with a weight of ~1.3 mg (active phase) per

electrode and a geometrical area of 0.196 cm2 (thickness: 0.2 mm). The

electrodes were attached to a stainless-steel collector following the

previously mentioned protocol and using the same separator as described

in section 5.2.3.1. These devices were characterized by CV at different

scan rates, galvanostatic charge-discharge (GCD) cycles at current

densities from 1 to 20 A/g and EIS in 1M H2SO4 solution. Impedance

spectra were measured at 0.05 V in the frequency range of 10 mHz to

100 kHz with an amplitude voltage of 10 mV. The measurements were

repeated in freshly prepared cells for their verification. Autolab

PGSTAT302 potentiostat was employed for EIS and CV measurements

and Arbin SCTS potentiostat for galvanostatic charge-discharge cycles.

A durability test was performed by 50000 GCD cycles at a current

density of 5 A/g and a voltage of 1.2 V. Current density and specific

capacitance is defined based on the total active weight of the carbon


223

material in the cell (two electrodes). The energy density and power

density were calculated as described elsewhere [17].

5.3 Results and discussion

5.3.1 Physicochemical characterization

Table 5.1 summarizes the porous texture and chemical composition

of the samples. Although both samples are obtained by following the

same procedure and the replica of the zeolite is successfully obtained

providing an ordered microporous structure (see XRD patterns, Figure

A5.1, and N2 adsorption-desorption isotherms, Figure A5.2, in annex of

Chapter 5), N-ZTC shows ca. 70 % of the apparent surface area and

micropore volume of ZTC.

The surface chemistry of both materials was characterized by XPS

and TPD. The amount of nitrogen and oxygen detected using XPS and

the evolved CO and CO2 quantities during TPD are compiled in Table

5.1. N-ZTC shows nitrogen content of 3.7 at. % detected by XPS.

According to the N1s XPS spectrum (Figure 5.1.b), this nitrogen is

attached to the surface as the following functional groups: quaternary

nitrogen (401.2 ± 0.2), pyrrole/pyridone (400.5 ± 0.2) and pyridines

(398.5 ± 0.2). The percentage of each kind of functional group is 63 %,

17% and 20%, respectively. Hence, quaternary nitrogen is the most

abundant nitrogen functionalities.

TPD experiments are used to study in detail the oxygen

functionalities that exist on the surface of both carbons [30-33]. Both

materials have a large amount of oxygen functional groups, ca. 3.5

Chapter 5

224

mmol/g or 5.5% wt of oxygen. More specifically, N-ZTC presents a

larger amount of CO2-evolving groups in the 200-400 ºC temperature

range (Figure A5.3 in annex of Chapter 5), which points out the presence

of more carboxylic moieties in the nitrogen-doped sample. Interestingly,

both samples show a similar amount of CO-evolving groups, but the

TPD profile for the evolution of CO, Figure 1a, evidences that the

amount of phenolic and ether groups (related to the CO evolution at 600-

700 ºC) is predominant in ZTC, whereas evolution of CO at higher

temperatures is greater in N-ZTC. Since the CO evolution is still high at

the end of the maximum temperature used in the TPD run, it can be

considered that the total amount of oxygen functionalities is larger in

this sample. Hence, surface oxygen groups of high thermal stability

(such as carbonyls and/or even pyrenes) are found on its surface. In

accordance to these findings, the surface oxygen functionalities present

in ZTC has been previously studied by TPD and Fourier transform

infrared spectroscopy and they have been assigned as: acid anhydride

(15%), ether (66%), hydroxyl (15%), and carbonyl (4%), whereas N-

ZTC presents a lower amount of ethers and hydroxyl functional groups

along with a larger amount of carbonyls [7, 34].

Table 5.1. Textural properties and surface chemical composition (XPS and TPD)

obtained for zeolite templated carbons.

Sample SBET

(m2/g)

VDRN2

(cm3/g)

NXPS

(at.%)

CO2 TPD

(µmol/g)

CO TPD

(µmol/g)

O TPD

(µmol/g)

ZTC 3600 1.55 - 470 2330 3280

N-ZTC 2760 1.05 3.7 510 2730 3750


225

Figure 5.1. (a) Comparison between CO TPD profiles of ZTC (black line) and N-ZTC

(blue line) samples. (b) N1s XPS deconvolution of N-ZTC sample.

5.3.2 Electrochemical characterization.

5.3.2.1. Electrochemical characterization before electro-oxidation

The carbon materials were characterized in a three-electrode cell in

acid and alkaline media in order to assess their electrochemical behavior.

It should be noted that ZTC can be easily oxidized at low potentials in

1M H2SO4, producing a very large amount of quinone functional groups

that are responsible of a large boost in pseudocapacitance [6,35]. In

order to avoid such feature, Figure 5.2.a shows the CVs obtained for

ZTC and N-ZTC in a potential range where no important oxidation

occurs and the performance of both pristine carbon materials can be

analyzed. The CVs obtained for both carbons have a broad oxidation

peak during the positive sweep that reaches a maximum at 0.26 V for N-

ZTC and at 0.32 V for ZTC. At the negative sweep, the corresponding

reduction peaks at potentials close to 0.16 and 0.23V are observed

respectively for N-ZTC and ZTC. The redox processes observed in the

ZTC electrode are undoubtedly related to electroactive functional groups

0

0.5

1

1.5

2

2.5

3

0 200 400 600 800

CO

(µ

mo

l/g

s)

Temperature (ºC)

(a)

396398400402404

I (a

.u.)

Binding Energy (eV)

(b)

Chapter 5

226

that, according to previous studies, correspond with the pseudocapacitive

contribution of the quinone/hydroquinone redox pair [6]. However, N-

ZTC shows larger pseudocapacitance in a wider potential range,

delivering a capacitance 38 F/g higher than that of ZTC (see Cg values

on Table 5.2). Since gravimetric capacitance is to some extent

proportional to the surface area of porous carbon materials [36,37], this

result must be a consequence of the presence of a higher amount of

electroactive surface oxygen functionalities contained in this material

together with nitrogen groups on N-ZTC.

Figure 5.2. 3rd cyclic voltammograms for ZTC (black curve) and N-ZTC (blue curve)

electrodes in (a) 1M H2SO4 and (b) 0.5 M KOH. v = 2 mV/s.

Table 5.2. Gravimetric and normalized capacitances of the carbon materials.

Sample Cg

[H2SO4] (F/g)

Cg

[KOH] (F/g)

Cg/SBET

[H2SO4] (µF/cm2)

Cg/SBET

[KOH] (µF/cm2)

ZTC 235 232 6.0 5.9

N-ZTC 273 224 9.9 8.1

In order to achieve a better understanding of the origin of the higher

capacitance of N-ZTC, an analogous CV was recorded in basic medium

(Figure 5.2.b) for both materials. In this case, the current of the redox

-1

-0.5

0

0.5

1

-0.3 -0.1 0.1 0.3 0.5 0.7

j (A

/g)


(a)

-1

-0.5

0

0.5

1

-1.1 -0.9 -0.7 -0.5 -0.3 -0.1

j (A

/g)


(b)


227

peaks previously seen in acid medium is notoriously decreased. Hence,

the pseudocapacitance observed in acid medium is a consequence of

functionalities that are not electroactive in basic medium. In this sense, it

has been reported that pyridine groups can contribute to largest

capacitance in basic media than in acid media via 1 electron process that

is impeded in acid electrolyte due to its protonation in this medium [8,

38]. This is the opposite trend observed for N-ZTC. However, it must be

considered that, since the presence of pyridinic nitrogen in N-ZTC is

low, it should not be expected that they have a high contribution to

capacitance in alkaline electrolyte.

Thus, the highest pseudocapacitive response of pristine N-ZTC in

acid electrolyte should be a consequence of its oxygen functionalities.

As seen in section 5.3.1, N-ZTC has larger amount of carbonyl

functional groups along with other CO-evolving oxygen functionalities

of higher thermal stability (Figure 5.1.a). These functional groups could

take part in redox reactions involving protons [11], explaining why the

pristine N-ZTC shows both a higher capacitance than ZTC in acid

medium and a capacitance decrease in alkaline electrolyte.

Also, it should be noted that the surface capacitance (Cg/SBET in

Table 5.2) is larger for N-ZTC in both media. Since the

pseudocapacitance contribution is mostly neglected in alkaline media,

the difference in surface capacitance in 0.5 M KOH can be related to an

improvement of wettability as consequence of the different surface

chemistry. This improvement can be a consequence of oxygen or

nitrogen functional groups. It is well-known that CO-evolving groups

Chapter 5

228

increase the wettability of carbon materials [11]. Also, the contribution

of N-Q to the wettability of carbons was previously reported [8, 39, 40].

Since both moieties exist on larger amounts on N-ZTC, the improvement

in surface capacitance can be related to an enhanced wettability of this

carbon material. Moreover, the distribution of the functional groups is

expected to be different for N-ZTC and ZTC, with N-Q being found

inside the pore network, increasing the wettability of the whole surface

of N-ZTC, whereas the CO-evolving groups of ZTC are expected to be

preferentially located at edge sites at the entrance of pores, delivering a

lower wetting of the inner porosity.

5.3.2.2 Electrochemical characterization after electro-oxidation

This section reports the performance of zeolite templated carbons

after exposing them to electro-oxidation conditions in acid and alkaline

media. Both carbons were electro-oxidized in 1M H2SO4 by performing

three CVs between -0.2 and 1.0 V at 2 mV/s. Figure 5.3 shows these

CVs for both carbon materials. On the first anodic sweep, both

electrodes show a large irreversible oxidation current when the potential

is shifted to values more positive than 0.6 V. It is well stablished that

this oxidation treatment produces electro-active oxygen functional

groups in zeolite templated carbons [6, 28, 41], and also some

degradation of the structure and electro-gasification [32]. The oxidation

current is larger for the ZTC electrode (see the higher current of the

black CV at 1.0 V in Figure 5.3.a compared to that of Figure 5.3.b),

evidencing that more intensive oxidation occurs in this sample. The net

increase of the currents of the reversible peaks at ca. 0.3V, which are


229

related to electroactivity of the quinone/hydroquinone functionalities,

observed after the electro-oxidation treatments corroborates this

phenomenon (see the differences between red -initial- and purple -final-

CVs in Figure 5.3.a).

Figure 5.3. Cyclic voltammograms for the electro-oxidation of ZTC (a) and N-ZTC

(b). v = 2 mV/s. 1M H2SO4.

Figure 5.4. 3rd cyclic voltammograms for ZTC and N-ZTC after electro-oxidation. v =

2 mV/s. 1M H2SO4.

-2

0

2

4

6

-0.3-0.1 0.1 0.3 0.5 0.7 0.9 1.1

j (A

/g)


ZTC1st cycle3rd cycleZTC ox

(a)

-2

0

2

4

6

-0.3-0.1 0.1 0.3 0.5 0.7 0.9 1.1

j (A

/g)


N-ZTC1st cycle3rd cycleN-ZTC ox

(b)

-2

-1

0

1

2

-0.3 -0.1 0.1 0.3 0.5 0.7 0.9

j (A

/g)


ZTC ox

N-ZTC ox

Chapter 5

230

Table 5.3. Gravimetric and equivalent series resistance of the oxidized carbon

materials.

Sample Cgox

[H2SO4] (F/g)

Cgox

[KOH] (F/g)

ESR+EDR

[H2SO4] (Ω)

ZTC 445 246 1.8

N-ZTC 384 230 0.3

N-ZTC also shows the same features (Figure 5.3.b). However, the

observed increase of these redox processes is lower. Consequently, the

capacitance recorded for ZTC after electro-oxidation (Cgox, Table 5.3) is

larger than that found for N-ZTC. It is interesting to note that the

voltammograms after the oxidation show different redox processes for

both carbon materials. Then, in ZTC an increase in the voltammetric

charge is observed in a wide potential window, from -0.1 up to 0.6 V,

while in the case of N-ZTC, the increase in voltammetric charge is only

observed from 0.2 V to 0.7 V. More significant differences upon the

electro-oxidation between these materials are highlighted by overlapping

their CVs (Figure 5.4). Although both CVs show similar features, ZTC

does show a larger pseudocapacitance from 0.05 to 0.4 V. Since the

structure of both materials is essentially identical [23] (see XRD

patterns, Figure S2 in annex of chapter 5), their different performance

under electro-oxidative conditions should be related to their different

surface chemistry. The electro-oxidized electrodes were analyzed by

TPD and it was found that approximately the same amount of oxygen

functionalities (6690 µmol/g for ZTC and 6740 µmol/g for N-ZTC) are

generated in the surface of both materials after the treatment. Itoi et al.


231

[6] proposed that the incorporation of oxygen groups on ZTC upon the

electro-oxidation happens because of the large amount of reactive edge

sites that exist on these carbons. Since N-ZTC incorporates the same

amount of oxygen functional groups, it can be inferred that larger

irreversible oxidation currents experimented by ZTC may be connected

to gasification reactions, where certain oxygen functionalities may

evolve as CO and CO2 (and therefore, their presence is not detected by

TPD over the oxidized electrode) during the CV sweep. Moreover, the

electroactivity of certain functional groups showed by electro-oxidized

ZTC seem to be hindered or even impeded in N-ZTC, since both carbons

have developed a similar amount and type of oxygen functionalities.

Figure 5.5. Electrochemical impedance spectra for ZTC and N-ZTC electrodes after

electro-oxidation in 1M H2SO4. E = 0.3V vs Ag/AgCl.

0

1

2

3

4

5

0 1 2 3 4 5

-Z''

(Ω

)

Z' (Ω)

ZTC oxN-ZTC ox

Chapter 5

232

EIS measurements were carried out for the electro-oxidized samples

in order to deepen into the changes produced by the electrochemical

oxidation. This technique allows one to distinguish the different

resistance contributions affecting the electrodes (diffusive problems,

electrode-electrolyte interface, etc.) [42]. Figure 5.5 shows the Nyquist

plots obtained in 1 M H2SO4 for both samples after the electro-oxidation

in the acid electrolyte. The main differences between both profiles are

observed at low frequencies, where an almost vertical line, related to

capacitive behavior, is observed. The onset frequency for reaching

capacitive behavior is higher in the case of N-ZTC. The equivalent series

resistance (ESR) plus the equivalent distributed resistance (EDR) of the

cells was calculated from the x-intercept of this line. The values obtained

for both electrodes are summarized in Table 5.3. It can be observed that

ZTC electrode shows a higher resistance than N-ZTC and a remarkably

Warburg region, related to diffusion processes through the pore network

[43]. In this work, the electrodes and the electrochemical cells were

prepared by following the same procedure and, consequently, the

differences observed are related to the electrode materials. As discussed

before (section 5.3.1), the zeolite templated carbons in this study initially

have a similar ordered structure and the same pore size of 1.2 nm. The

main differences of both carbons are their surface chemistry and the

larger apparent surface area of non-doped ZTC. In this sense, N-ZTC

has an important content in N-functionalities that improve both electrical

conductivity and wettability [20, 44-46], which can explain a decrease in

the electrolyte diffusion resistance.


233

The differences in electrochemical stability when exposed to high

oxidation potentials were further studied in alkaline electrolyte. Figure

5.6 shows the cyclic voltammograms obtained for N-ZTC and ZTC in

basic medium under the oxidative conditions. Again, the current density

increases when the potential is shifted to more positive values and an

irreversible faradaic current related to oxidation processes is observed

(black curve, Figure 5.6.a and 5.6.b). As happened in acid medium, the

current density obtained for ZTC is higher than for N-ZTC electrode,

pointing out the higher electrochemical stability of the N-doped carbon.

Figure 5.6. Cyclic voltammograms for the electro-oxidation of ZTC (a) and N-ZTC

(b). v = 2 mV/s. 0.5M KOH.

Furthermore, the CV obtained for N-ZTC remains practically

invariable after positive polarization, while, in case of ZTC, there is a

small increase of capacitance (Table 5.2 and 5.3). This increase of

capacitance can be explained by different phenomena: (i) increase of

wettability and (ii) appearance of pseudocapacitance. In general, carbon

materials have an increase of wettability when oxygen functional groups

(mainly CO-evolving groups) are attached to their surface [47].

Consequently, the improvement of capacitance of ZTC is mainly

-1

-0.5

0

0.5

1

1.5

2

-1.1-0.9-0.7-0.5-0.3-0.1 0.1 0.3

j (A

/g)



(a)

-1

-0.5

0

0.5

1

1.5

2

-1.1-0.9-0.7-0.5-0.3-0.1 0.1 0.3

j (A

/g)



(b)

Chapter 5

234

attributed to the generation of oxygen functionalities on this carbon after

electro-oxidation. As expected, this effect is not observed in N-ZTC

electrode owing to the contribution of N-Q functionalities to the

wettability of this material. Regarding pseudocapacitance, quinones are

reported to be electroactive at lower pHs. However, some small

pseudocapacitive behavior is observed in alkaline electrolyte for both N-

ZTC and ZTC electrodes (Figure 5.7). Previous research reported the

evidence of pseudocapacitance in alkaline medium due to the

contribution of oxygen functionalities in carbon cloths when the pH is

above 11 due the previous activation in acid medium of pyrone

derivatives and another unidentified species [48]. Thus, different CO-

evolving functional groups could be responsible of the electroactivity

showed in KOH electrolyte by both carbon materials, although this is an

issue that needs further research.

Figure 5.7. 3rd cyclic voltammograms for ZTC (a) and N-ZTC (b) after electro-

oxidation. v = 2 mV/s. 0.5M KOH.

-1

-0.5

0

0.5

1

-1.1 -0.9 -0.7 -0.5 -0.3 -0.1 0.1

j (A

/g)


ZTC ox

N-ZTC ox


235

5.3.2.3. ZTC and N-ZTC supercapacitors using acid electrolyte

The performance of the materials as electrodes for supercapacitors

was assessed by using a symmetric configuration (in mass) in 1M

H2SO4. Prior to assembling the two-electrode cell of ZTC, both positive

and negative electrodes were electro-oxidized (as described in section

5.3.2.2) in order to stabilize their response and avoid any problem

related with the large irreversible current shown by ZTC during the first

loading cycles [49]. Since the electro-oxidation current shown by N-

ZTC in the CV studies was lower, the stability of the negative electrode

is not affected (i.e. the negative electrode does not reach potentials

where hydrogen evolution occurs) and, consequently, this pre-treatment

was not needed for N-ZTC electrodes.

Figure 5.8. Steady state cyclic voltammograms for ZTC and N-ZTC based

supercapacitors. v = 10 mV/s. V = 1.2 V. 1M H2SO4.

-2

-1.5

-1

-0.5

0

0.5

1

1.5

-0.1 0.2 0.5 0.8 1.1

j (A

/g)

Voltage (V)

ZTCN-ZTC

Chapter 5

236

Figure 5.8 shows cyclic voltammograms obtained for ZTC and N-

ZTC based capacitors operating at 1.2V and Table 5.4 compiles the

gravimetric capacitance obtained from these CVs. The main contribution

to the gravimetric capacitance is given at low voltage, since the

electrodes in the capacitors are working around the open circuit potential

(OCP, 0.34 and 0.41 V for ZTC and N-ZTC electrodes, respectively),

where the pseudocapacitance contribution is maximized. As the voltage

increases, the current decreases sharply, until reaching approximately a

constant value when the voltage is beyond 0.7 V. At these voltages, the

potentials of positive and negative electrodes are expected to be shifted

at least 0.35 V from the OCP (since the voltage of the cell results from

the difference of potential between both electrodes). Thus, they have

reached potentials where these materials mainly have the contribution of

electrical double layer (Figures 5.3 and 5.4), and therefore a much lower

capacitance is obtained from each electrode and, accordingly, in the

supercapacitor.

In the case of N-ZTC supercapacitor, the charge is lower than for

ZTC at voltages close to zero due to the lower pseudocapacitance

contribution on this material (Figure 5.4), while the charge drop of the

supercapacitor at medium and high voltages in N-ZTC is mitigated due

to the higher charge of this material as the potential of the electrodes is

shifted from OCP.


237

Table 5.4. Parameters obtained for ZTC and N-ZTC based capacitors in 1M H2SO4.

Cell CCV (F/g)

CGCDa

(F/g)

CGCDb

(F/g)

R (Ω)

E

(Wh/kg)

Pmax

(kW/kg)

Cf/Co

(%)

ZTC 60 61 19 3.50 5.9 23 88

N-ZTC 56 57 46 0.85 7.5 98 91 a Capacitance at 1 A/g. b Capacitance at 20 A/g.

Figure 5.9. GCD cycles for ZTC and N-ZTC based supercapacitors at (a) 1 A/g and (b)

20 A/g. V = 1.2 V. (c) Ragone plot obtained for ZTC and N-ZTC based supercapacitors

at 1, 2.5, 5, 10 and 20 A/g. V = 1.2 V. (d) Electrochemical impedance spectra for ZTC

and N-ZTC based supercapacitors. V = 0.05 V. 1M H2SO4.

Figure 5.9.a shows the galvanostatic charge-discharge (GCD) cycles

obtained for ZTC and N-ZTC based capacitors at 1 A/g. The shape of

the cycles differs from the ideal triangle characteristic of carbon

0

0.5

1

1.5

0 50 100 150

Vo

lta

ge (

V)

Time (s)

ZTCN-ZTC

(a)

0

0.5

1

1.5

0 1 2 3 4 5 6

Vo

lta

ge (

V)

Time (s)

ZTC

N-ZTC

(b)

0.1

1

10

0.1 10

E (

Wh

/kg)

P (kW/kg)

ZTCN-ZTC

(c)

0

1

2

3

4

5

0 1 2 3 4 5

-Z''

(Ω

)

Z' (Ω)

ZTCN-ZTC

(d)

Chapter 5

238

materials with pure contribution of electrical double layer formation due

to the strong influence of pseudocapacitive processes and other

phenomena previously discussed (see section 5.3.2.1). Under these

conditions, both cells evidence similar gravimetric capacitance (Table

5.4). However, the increase of current density up to 20 A/g produces a

decrease in capacitance more accused in the case of ZTC, due to a higher

ohmic drop (Figure 5.9b). This plays an important role on the retention

of energy at high power density, as can be seen in the Ragone plot

(Figure 5.9c). Thus, when energy is considered, N-ZTC based

supercapacitor shows only a small improvement compared to ZTC one

at 1 A/g. However, the difference in energy becomes larger at high

current density, since N-ZTC based capacitor only experiences a

decrease of 28% of the initial energy at 20 A/g, while in case of ZTC

cell, a loss of 78% is observed. These differences on the energy retention

at high power density are usually related to certain properties of the

electrodes, such as electrical conductivity and ion mobility within the

porosity. These properties can be better understood by EIS.

Figure 5.9d shows the Nyquist plot obtained for ZTC and N-ZTC

based capacitors at V = 0.05 V recorded after finishing the determination

of the Ragone plot. The profiles shown in Figure 9.d are characteristic of

carbon materials with a mainly capacitive behavior [43]. At high

frequencies, a semicircle is observed in both supercapacitors. This

semicircle is produced by the parallel combination of the bulk

capacitance of the electrolyte and the electrical resistances to charge

propagation in the electrode [50], and it includes contact resistance


239

between particles. At medium frequencies, the Warburg region

(evidenced by a 45º line in the Nyquist plot) is observed, while at low

frequencies, the shape of the curve becomes close to a vertical line,

characteristic of the capacitive behavior [42]. Table 5.4 summarizes the

cell resistance obtained for both capacitors. As happened in three

electrode cell configuration (section 5.3.2.2.), N-ZTC capacitor shows

much lower resistance. These cell resistances are governed by: (i) the

diffusive problems of the ions through the porosity (Warburg region

[43]) and (ii) the electrical resistances to charge propagation in the

electrode (indicated by the diameter of the semicircle) [50]. Both the

Warburg and the semicircle resistances are lower in N-ZTC, although

the latter resistance is much lower. For understanding the difference in

this resistance, it should be noted that both ZTC and N-ZTC

supercapacitors have been assembled using the same electrolyte, the

same electrode composition and by following the same procedure.

Therefore, we propose that the main differences between the cell

resistances of these supercapacitors are mainly related to the inherent

electrical conductivity of the electrode materials that decrease the

contact resistance between particles. As discussed before (section

5.3.2.2), the main differences of both carbons are related to the presence

of nitrogen heteroatoms in N-ZTC. The resistance value obtained for this

capacitor is four times lower than that calculated for ZTC cell (Table

5.4), and it evidences the better electrical conductivity and charge

propagation of N-ZTC electrodes. In accordance to this finding, the

maximum power obtained for N-ZTC based capacitor (calculated from

the resistance measured from the ohmic drop of GCD cycles, which is in

Chapter 5

240

good agreement to that obtained from the EIS analyses) is four times

larger than the value obtained for ZTC cell. This value outperforms

those found in the literature for other carbon electrodes in

supercapacitors [5,51,52], such as activated carbons (61.2 kW/kg in 1M

H2SO4) [17], activated carbon nanofibers (20 kW/kg in 6M KOH) [53],

hierarchical porous carbons (52.7 kW/kg in 1M H2SO4) [54], carbon

nanotubes (43.3 kW/kg in 1M Et4NBF4/propylene carbonate) [55] and

other templated carbons (28kW/kg in 1M H2SO4) [56]. This outstanding

improvement of N-ZTC based capacitor is probably a consequence of its

connected nanopore structure, which facilitates the accessibility of the

electrolyte [4], and the large quantity of N-Q functionalities and pyrrole

on the surface of this material, that are able to increase both wettability

and electrical conductivity [18, 20, 44, 45].

Figure 5.10. Durability test for ZTC and N-ZTC based supercapacitors at 1.2V. 5 A/g.

50000 cycles. 1 M H2SO4 electrolyte.

0

20

40

60

80

100

120

0 10000 20000 30000 40000 50000

C/C

o (%

)

Nº Cycles

ZTCN-ZTC


241

The presence of nitrogen functionalities can also modify the stability

of porous carbon materials [17,19,22]. In order to explore this option,

the durability of both cells was evaluated by 50000 GCD cycles at 1.2 V.

Figure 5.10 shows the evolution of the capacitance retention (C/Co)

during the test. Both capacitors evidence a similar capacitance retention

after the durability test (Table 5.4), although it is somewhat higher in

case of N-ZTC cell. However, it can be seen in Figure 5.10 that ZTC

based capacitor shows larger changes of capacitance along cycles,

evidencing a less stable performance. The improvement of stability in

this capacitor is probably a consequence of the N functionalities formed

on N-ZTC electrodes, since they are able to increase the stability by

preventing the incorporation of detrimental oxygen functionalities [17].

5.4 Conclusions

In this chapter, the electrochemical behavior of non-doped and N-

doped zeolite templated carbons were studied as electrodes for

supercapacitors in different aqueous electrolytes. The materials were

synthesized by CVD of different precursors and they evidenced a

practically identical structure but different surface chemistry. The role of

the different N-functionalities in their electrochemical performance was

elucidated by different techniques. The study carried out by cyclic

voltammetry evidenced a higher resistance to electro-oxidation and

degradation in case of N-ZTC in acid and alkaline media. Also, the

different electroactivity in 1M H2SO4 of the functional groups generated

at positive potentials was demonstrated. The CV study suggested that N-

ZTC has better wettability than ZTC as consequence of the large amount

Chapter 5

242

of N-Q functionalities at its surface. Also, it was shown that N-Q

functionalities provide better conductivity to this carbon.

The effect of the different performance of these carbons as

electrodes for supercapacitors was assessed in acid electrolyte using a

two-electrode cell configuration. The results showed that both capacitors

provide similar capacitance, but larger energy in case of N-ZTC. This

result is the outcome of the large dependence of capacitance on potential

in the highly oxidized ZTC. This behavior produces a large capacitance

at low voltages, but a much smaller capacitance at high voltages, where

most energy would be stored. In consequence, the more capacitive

behavior of N-ZTC renders an improved energy density.

More interestingly, N-ZTC based supercapacitor provided a

maximum power that is four times larger than that showed by ZTC

based supercapacitor (98 and 23 kW/kg), as consequence of the

improvement of electrical conductivity produced by N-Q functionalities

in N-ZTC electrodes. The durability of the capacitors was evaluated by

50000 GCD cycles at 5 A/g and 1.2 V. Both capacitors evidenced high

capacitance retention after the durability test, but N-ZTC based capacitor

showed a most stable performance along cycles, pointing out the

stabilizing effect of N functional groups. These results provide clear

evidences of the advantages of doping advanced porous carbon materials

with nitrogen functionalities for the improvement of the performance of

aqueous based supercapacitors.


243

5.5 References

[1] F. Béguin, E. Frackowiak, Carbons for Electrochemical Energy

Storageand Conversion Systems,1st ed., Boca Raton, FL: Taylor

& Francis (CRC Press) 2010.

[2] A.G. Pandolfo, A.F. Hollenkamp, Carbon properties and their role

in supercapacitors, Journal of Power Sources 157 (2006) 11–27.


Electrolytes for Advanced Supercapacitors, Adv. Mater. 26

(2014) 2219–2251.

[4] H. Itoi, H. Nishihara, T. Kogure, T. Kyotani, Three-

Dimensionally Arrayed and Mutually Connected 1.2 nm

Nanopores for High-Performance Electric Double Layer

Capacitor, J. Am. Chem. Soc. 133 (2011) 1165–1167.

[5] H. Nishihara, T. Kyotani, Templated Nanocarbons for Energy

Storage, Advanced Materials. 24 (2012) 4473–4498.



Functionalized Zeolite-Templated Carbon, Bull Chem Soc Jpn. 87

(2014) 250–257.

[7] H. Itoi, H. Nishihara, T. Kyotani, Effect of Heteroatoms in

Ordered Microporous Carbons on Their Electrochemical

Capacitance, Langmuir 32 (2016) 11997–12004.


Amorós, J. Geng, N. Soin, E. Siores, B.F.G. Johnson, On the

origin of the high capacitance of nitrogen-containing carbon

nanotubes in acidic and alkaline electrolytes, Chem. Commun. 50

(2014) 11343–11346.

[9] M. Seredych, D. Hulicova-Jurcakova, G.Q. Lu, T.J. Bandosz,

Surface functional groups of carbons and the effects of their

chemical character, density and accessibility to ions on

electrochemical performance, Carbon 46 (2008) 1475–1488.

Chapter 5

244

[10] D. Hulicova-Jurcakova, M. Seredych, G.Q. Lu, T.J. Bandosz,

Combined Effect of Nitrogen- and Oxygen-Containing Functional

Groups of Microporous Activated Carbon on its Electrochemical

Performance in Supercapacitors, Adv. Funct. Mater. 19 (2009)

438–447.




Carbon 44 (2006) 2642–2651.


Morallón, H. Nishihara, T. Kyotani, J. Rodríguez-Mirasol, T.

Cordero, Enhanced electro-oxidation resistance of carbon

electrodes induced by phosphorus surface groups, Carbon 95

(2015) 681–689.

[13] D. Hulicova-jurcakova, A.M. Puziy, O.I. Poddubnaya, F. Suárez-

García, J. M. D. Tascón, G. Quing Lu, Highly Stable Performance

of Supercapacitors from Phosphorus-Enriched Carbons, J Am

Chem Soc 2009; 131(4):5026–7.

[14] D.-W. Wang, F. Li, Z.-G. Chen, G.Q. Lu, H.-M. Cheng, Synthesis

and Electrochemical Property of Boron-Doped Mesoporous

Carbon in Supercapacitor, Chem. Mater. 20 (2008) 7195–7200.




Sci. 6 (2013) 2839–2855.

[16] M. Enterría, M.F.R. Pereira, J.I. Martins, J.L. Figueiredo,

Hydrothermal functionalization of ordered mesoporous carbons:

The effect of boron on supercapacitor performance, Carbon 95

(2015) 72–83.


Amorós, Nitrogen doped superporous carbon prepared by a mild

method. Enhancement of supercapacitor performance, Int. J.

Hydrogen Energy 41 (2016) 19691–19701.


245


Amorós, Generation of nitrogen functionalities on activated


Chemical and electrochemical characterization, Carbon 91 (2015)

252–265.





[20] T. Kwon, H. Nishihara, H. Itoi, Q.-H. Yang, T. Kyotani,

Enhancement Mechanism of Electrochemical Capacitance in

Nitrogen-/Boron-Doped Carbons with Uniform Straight

Nanochannels, Langmuir 25 (2009) 11961–11968.

[21] González-Gaitán C, Ruiz-Rosas R, Morallón E, Cazorla-Amorós

D. Electrochemical Methods to Functionalize Carbon Materials.

In: V.K. Thakur, M.K. Thakur, editors. Chemical

functionalization of carbon nanomaterials, Boca Raton, FL:

Taylor & Francis (CRC Press); 2015: p. 230–261.





[23] H. Nishihara, P. Hou, L. Li, M. Ito, M. Uchiyama, T. Kaburagi, et

al., High-Pressure Hydrogen Storage in Zeolite-Templated

Carbon, J. Phys. Chem. C 113 (2009) 3189–3196.

[24] P.-X. Hou, H. Orikasa, T. Yamazaki, K. Matsuoka, A. Tomita, N.

Setoyama, et al., Synthesis of nitrogen-nontaining microporous

carbon with a highly ordered structure and effect of nitrogen

doping on H2O adsorption, Chem. Mater. 17 (2005) 5187–5193.

[25] Y. Kwon, K. Kim, R. Ryoo, N-doped zeolite-templated carbon as

a metal-free electrocatalyst for oxygen reduction, RSC Adv. 6

(2016) 43091–43097.

Chapter 5

246

[26] C.O. Ania, V. Khomenko, E. Raymundo-Piñero, J.B. Parra, F.

Béguin, The large electrochemical capacitance of microporous

doped carbon obtained by using a zeolite template, Adv. Funct.

Mater. 17 (2007) 1828–1836.

[27] K. Nueangnoraj, H. Nishihara, T. Ishii, N. Yamamoto, H. Itoi, R.

Berenguer, et al., Pseudocapacitance of zeolite-templated carbon

in organic electrolytes, Energy Storage Mater. 1 (2015) 35–41.


Cazorla-Amorós, et al., Electrochemical generation of oxygen-

containing groups in an ordered microporous zeolite-templated

carbon, Carbon 54 (2013) 94–104.

[29] Z. Ma, T. Kyotani, Z. Liu, O. Terasaki, A. Tomita, Very high

surface area microporous carbon with a three-dimensional nano-

array structure: Synthesis and its molecular structure, Chem.

Mater. 13 (2001) 4413–4415.

[30] H. Boehm, Some aspects of the surface chemistry of carbon

blacks and other carbons, Carbon 32 (1994) 759–769.

[31] Y. Otake, R.G. Jenkins, Characterization of oxygen-containing

surface complexes created on a microporous carbon by air and

nitric acid treatment, Carbon 31 (1993) 109-121.

[32] M.C. Román-Martínez, D. Cazorla-Amorós, A. Linares-Solano,

C.S.M. de Lecea, TPD and TPR characterization of carbonaceous

supports and Pt/C catalysts, Carbon 31 (1993) 895–902.

[33] P. Burg, D. Cagniant, Characterization of carbon surface

chemistry, in: L.R. Radovic, editor. Chemistry and Physics of

carbons, vol. 30, Boca Raton, FL: Taylor & Francis (CRC Press);

2008: p. 129-175.

[34] H. Nishihara, Q. Yang, P. Hou, M. Unno, S. Yamauchi, R. Saito,

et al., A possible buckybowl-like structure of zeolite templated

carbon, Carbon 47 (2009) 1220–1230.


Nishihara, T. Kyotani, E. Morallón, et al., Characterization of a


247

zeolite-templated carbon by electrochemical quartz crystal

microbalance and in situ Raman spectroscopy, Carbon 89 (2015)

63–73.

[36] D. Cazorla-Amorós, D. Lozano-Castelló, E. Morallón, M.J.

Bleda-Martínez, A. Linares-Solano, S. Shiraishi, Measuring cycle



[37] O. Barbieri, M. Hahn, A. Herzog, R. Kötz, Capacitance limits of

high surface area activated carbons for double layer capacitors,

Carbon 43 (2005) 1303–1310.


Zhu, G.Q. Lu, Nitrogen-enriched nonporous carbon electrodes

with extraordinary supercapacitance, Adv. Funct. Mater. 19

(2009) 1800–1809.

[39] S.L. Candelaria, B.B. Garcia, D. Liu, G. Cao, Nitrogen

modification of highly porous carbon for improved supercapacitor

performance, J. Mater. Chem. 22 (2012) 9884-9889.

[40] M. Kawaguchi, T. Yamanaka, Y. Hayashi, H. Oda, Preparation

and capacitive properties of a carbonaceous material containing

nitrogen, J. Electrochem. Soc. 157 (2010) A35-A40.

[41] R. Ruiz-Rosas, M.J. Valero-Romero, D. Salinas-Torres, J.

Rodríguez-Mirasol, T. Cordero, E. Morallón, et al.,

Electrochemical performance of hierarchical porous carbon

materials obtained from the infiltration of lignin into zeolite

templates., ChemSusChem 7 (2014) 1458–1467.


Fundamentals and Technological Applications. New York:

Springer; 1999.

[43] R. Kötz, M. Carlen, Principles and applications of electrochemical

capacitors, Electrochim. Acta 45 (2000) 2483–2498.

Chapter 5

248

[44] G. Lota, K. Lota, E. Frackowiak, Nanotubes based composites

rich in nitrogen for supercapacitor application, Electrochem.

Commun. 9 (2007) 1828–1832.

[45] Y. Zhou, X. Xu, B. Shan, Y. Wen, T. Jiang, J. Lu, et al., Tuning

and understanding the supercapacitance of heteroatom-doped

graphene, Energy Storage Mater. 1 (2015) 103–111.

[46] V. Strelko, V. Kuts, P. Thrower, On the mechanism of possible



transfer reactions, Carbon 38 (2000) 1499–1503.





[48] H.A. Andreas, B.E. Conway, Examination of the double-layer

capacitance of an high specific-area C-cloth electrode as titrated

from acidic to alkaline pHs, Electrochim. Acta 51 (2006) 6510–

6520.

[49] K. Nueangnoraj, R. Ruiz-Rosas, H. Nishihara, S. Shiraishi, E.

Morallón, D. Cazorla-Amorós, et al., Carbon–carbon asymmetric

aqueous capacitor by pseudocapacitive positive and stable

negative electrodes, Carbon 67 (2014) 792–794.

[50] S. Fletcher, V.J. Black, I. Kirkpatrick, A universal equivalent

circuit for carbon-based supercapacitors, J. Solid State

Electrochem. 18 (2014) 1377–1387.

[51] S.L. Candelaria, Y. Shao, W. Zhou, X. Li, J. Xiao, J. Zhang, et al.,

Nanostructured carbon for energy storage and conversion, Nano

Energy 1 (2012) 195–220.

[52] Y. Zhai, Y. Dou, D. Zhao, P.F. Fulvio, R.T. Mayes, S. Dai,

Carbon materials for chemical capacitive energy storage, Adv.

Mater. 23 (2011) 4828–4850.


249

[53] T. Le, Y. Yang, Z. Huang, F. Kang, Preparation of microporous

carbon nanofibers from polyimide by using polyvinyl pyrrolidone

as template and their capacitive performance, J. Power Sources

278 (2015) 683–692.

[54] D. Salinas-Torres, R. Ruiz-Rosas, M.J. Valero-Romero, J.

Rodríguez-Mirasol, T. Cordero, E. Morallón, et al., Asymmetric

capacitors using lignin-based hierarchical porous carbons, J.

Power Sources 326 (2016) 641–651.

[55] D.N. Futaba, K. Hata, T. Yamada, T. Hiraoka, Y. Hayamizu, Y.

Kakudate, et al., Shape-engineerable and highly densely packed

single-walled carbon nanotubes and their application as super-

capacitor electrodes, Nat. Mater. 5 (2006) 987–994.

[56] G.A. Ferrero, A.B. Fuertes, M. Sevilla, N-doped porous carbon

capsules with tunable porosity for high-performance


Chapter 5

250

ANNEX TO CHAPTER 5

Figure A5.1. XRD patterns obtained for ZTC and N-ZTC carbon materials.

Figure A5.2. N2 adsorption desorption isotherms measured at -196 ºC obtained for

ZTC and N-ZTC samples.

0 10 20 30 40 50

I (a

.u.)

2Ɵ (degrees, Cu-Kα)

0

200

400

600

800

1000

1200

1400

0 0.2 0.4 0.6 0.8 1

Ad

sorb

ed

Volu

me (

Cm

3/g

ST

P)

P/P0


251

Figure A5.3. Comparison between CO2 TPD profiles of ZTC and N-ZTC samples.

0

0.2

0.4

0.6

0.8

1

100 200 300 400 500 600 700 800 900

CO

2(µ

mo

l/g

s)

Temperature (ºC)

ZTC

N-ZTC

Chapter 6



organic electrolyte

Electrochemical performance of N-doped activated carbons in organic electrolyte

255

6.1. Introduction

Electrochemical capacitors are energy storage devices of great

interest mainly due to their high power density, durability and broad

operation temperature range [1,2]. However, it is necessary to increase

their energy density for enabling their use in a wider range of

applications. This requires the use of electrode materials and electrolytes

able to increase both the capacitance and the operation voltage of the

electrochemical capacitor. Regarding the electrode materials, carbon

materials are the most commonly used since they have a well-developed

porosity, electrochemical stability and tunable surface chemistry [3]. In

case of the electrolyte, the use of non-aqueous electrolytes (organic

solvents and ionic liquids) has attracted great interest as consequence of

their possible use at high operation voltage [4,5]. Nonetheless, the use of

high operation voltage dramatically affects the performance of the

electrodes, since carbon materials undergo reactions with the electrolyte

that decrease their electrochemical stability and, consequently, diminish

the cycle life of the device [6-8]. Hence, the development of carbon

materials with improved electrochemical properties is highly needed for

their use in electrochemical capacitors.

The performance of carbon materials can be enhanced by following

different strategies [9]. One of the main approaches consists in

synthesizing materials doped with different heteroatoms [10,11], such as

sulfur, phosphorus or nitrogen, since they modify the electrochemical

behavior of the material [6, 12-14]. In case of nitrogen doping, it has

been proven that can increase the wettability, conductivity and

Chapter 6

256

electrochemical stability [6, 7, 15]. However, the role of the different

nitrogen groups is not still well understood.

Nitrogen doped carbon materials can be prepared by several

methods [16-20], that can be classified as follows: (i) by using a nitrogen

compound as nitrogen and carbon source in different synthesis methods,

such as carbonization (followed by activation), templating methods or

hydrothermal carbonization; (ii) through post-treatments of a material

previously synthesized with a nitrogen-containing reactant in gas or

liquid phase. The second method usually involves the use of high

temperatures to favor the incorporation of nitrogen, wether in gas or

liquid pase. However, this can affect the porosity of the pristine carbon

material and makes difficult to control the type of functionalities that are

formed on the surface. In this sense, we have developed nitrogen

functionalization methods at mild conditions that allow the incorporation

of a wide range of nitrogen functionalities and allows the preservation of

the structure of the pristine carbon material, even when using carbon

materials with extraordinary microporosity (see chapters 3 and 4 and

references [8, 21]).

The aim of this work is to elucidate the role of different nitrogen

functionalities on the performance of activated carbons as electrodes for

electrochemical capacitors in organic electrolyte, mainly focusing on the

effect on the electrochemical stability and durability of the device. We

propose the synthesis of nitrogen-doped activated carbons with high

apparent surface area by combining functionalization methods based on

wet methods at low temperatures and post-thermal treatments at


257

different temperatures, whose combination allows a selective

modification of the surface chemistry. The effect of different nitrogen

functionalities and doping methods on the performance of these

materials as electrodes for electrochemical capacitors is tested in organic

electrolyte and severe conditions of temperature and operation voltage.


6.2.1. Synthesis of carbon materials

6.2.1.1. Pristine carbon material

An activated carbon with high microporosity prepared in our

laboratory has been used as the starting material for nitrogen

incorporation via post-modification treatments. The pristine material

(named as KUA) has been synthethized by chemical activation of a

Spanish anthracite with KOH using an impregnation ratio of activating

agent to raw material of 4:1 and an activation temperature of 750º C

under inert atmosphere, which was held for 1 hour. More details about

the preparation process are available elsewhere [22] (chapter 2).

6.2.1.2. N-functionalization of carbon materials at mild conditions

KUA was further functionalized with nitrogen functional groups by

two different approaches based on the organic chemistry reactions

described in chapters 3 and 4. Briefly, the first approach consisted in a

three-step protocol: (i) chemical oxidation with HNO3, (ii) treatment

with SOCl2 and (iii) amidation reaction with NH4NO3/DMF and

pyridine. The obtained sample was named as KUA-CONH2. In the

Chapter 6

258

second approach, the third step of the first method is directly applied

over pristine sample (KUA). This sample was named as KUA-N. More

details about the preparation methods are found in chapters 3 and 4.

6.2.1.3. Heat treatments

The carbon materials (KUA, KUA-CONH2 and KUA-N samples)

were heat treated at two different temperatures for 1 hour under N2

atmosphere using a flow of 200mL/min and a heating rate of 5 Cº/min.

The obtained samples are named as KUA_T, KUA-CONH2_T and

KUA-N_T, where T is the final heating temperature (500 and 800 ºC).

6.2.2. Physicochemical characterization



Desorption (TPD). XPS analyses were performed using a VG-Microtech

Multilab 3000 spectrometer with an Al anode. N1s spectra were

deconvoluted using Gaussian functions with 20% of Lorentzian

component. A Shirley line was used as background and the FWHM of

the peaks was kept between 1.4 and 1.7 eV. TPD experiments were

carried out using a TGA-DSC instrument (TA Instruments, SDT Q600

Simultaneous) coupled to a mass spectrometer (Thermostar, Balzers,

BSC 200), by heating the samples up to 950 ºC (heating rate: 20 ºC/min)

under helium atmosphere (flow rate: 100 mL/min).




259

C using an Autosorb-6-Quantachrome apparatus. The samples were








adsorption isotherms (relative pressures from 0.0001 to 0.25) [23].


The carbon materials were used for the preparation of two-electrode

cells (Al/Mg-body cell, Honsen Corporation, Japan). The electrodes

were prepared by mixing the carbon material with acetylene black and

PTFE in a ratio of 85:10:5 wt %. The electrodes were pressed to shape

disks with a diameter of 13 mm and a thickness of ~0.5 mm (net weight:

35 mg). Aluminium paper was used as current collector and 1M

triethylmethylamonium tetrafluoroborate diluted in propylene carbonate

(TEMA-BF4/PC) as electrolyte. The cells were assembled using an

Argon glovebox.

The cells were tested by galvanostatic charge-discharge (GCD)

cycles at different current densities (40, 100, 200, 400, 600, 1000 mA/g)

using an operation voltage of 2.5 V and 40ºC. The capacitance was

calculated from the discharge curve of the GCD cycles as reported

previously in the literature [24]. The gravimetric capacitance is referred

Chapter 6

260

to the active mass of the carbon in the full cell. Afterwards, a durability

test (based on GCD cycles and floating test) was performed by applying

the following procedure: (i) 5 GCD cycles at 2.5 V and 40 ºC; (ii) 5

GCD cycles at 2.5 V and 70 ºC; (iii) 5 GCD cycles at 3.2 V and 70 ºC;

(iv) voltage is kept at 3.2 V for 100 hours (floating test). After finishing

the floating test (iv), steps (iii), (ii) and (i) were subsequently repeated.

The capacitance, resistance and integrated leakage current were

calculated. The retention of capacitance is given at 2.5 V and 40 ºC. The

current density was kept as 40 mA/g during the experiment.


6.3.1. Physicochemical characterization of carbon materials

6.3.1.1 Porous texture

Figure 6.1 illustrates the N2 adsorption-desorption isotherms

obtained for all activated carbons. The profiles evidence that all carbon

materials show an adsorption-desorption curve characteristic of

microporous materials. Table 6.1 summarizes the porous texture

parameters obtained from N2 and CO2 isotherms for all the activated

carbons and the pristine KUA, the treated at 800ºC (KUA_800) and the

oxidized in nitric acid (KUA_COOH) has been included for comparison

pourpouse. The changes in the porous texture of the samples obtained by

oxidation (KUA-COOH) and chemical functionalization (KUA-CONH2

and KUA-N) were previously discussed in Chapters 3 and 4. The most

significant change is observed for KUA-COOH and KUA-CONH2


261

samples, as consequence of the generation of oxygen functionalities in

KUA-COOH and their conversion into nitrogen functional groups that

occupy or block part of the microporosity [21] (Table 6.1, Figure 6.2b).

However, when the chemical method is directly applied over the pristine

sample (sample KUA-N), the microporosity of the material is fully

preserved (as evidenced in Figure 6.1a., Table 6.1).

Regarding the heat-treated samples, some loss of the microporosity

is detected when the heat treatment is carried out over KUA and KUA-N

samples (Figure 6.1a and c, Table 6.1). This change could be produced

by some structural rearrangement in the materials [25]. However, the

heat treatments over KUA-CONH2 produce an increase of the micropore

volume (see samples KUA-CONH2_500 and KUA-CONH2_800). This

increase of microporosity is related to the removal of functional groups

that occupies or block part of the microporosity in the pristine sample.

Interestingly, the thermal treatments at different temperatures (500 ºC

and 800 ºC) produce identical modifications on the porous texture of the

pristine sample (KUA-N and KUA-CONH2). As consequence, the N2

adsorption isotherms obtained for samples KUA-N_500 and KUA-

N_800 overlaps (Figure 6.1a.). The same is observed for KUA-

CONH2_500 and KUA-CONH2_800 samples (Figure 6.1b).

Chapter 6

262

0

200

400

600

800

1000

0 0.2 0.4 0.6 0.8 1

Ad

sorb

ed

Vo

lum

e (

cm

3/g

)

P/P0

KUAKUA-NKUA-N_500KUA-N_800

(a)

0

200

400

600

800

1000

0 0.2 0.4 0.6 0.8 1

Ad

sorb

ed

Vo

lum

e (

cm

3/g

)

P/P0

KUAKUA-CONH₂KUA-CONH₂_500KUA-CONH₂_800

(b)

0

200

400

600

800

1000

0 0.2 0.4 0.6 0.8 1

Ad

sorb

ed

Vo

lum

e (

cm

3/g

)

P/P0

KUA

KUA_800

(c)

Figure 6.1. N2 adsorption-desorption isotherms obtained for all activated carbons: (a)

KUA, KUA-N and related heat-treated samples, (b) KUA, KUA-CONH2and related

heat-treated samples, (c) KUA and related heat-treated sample.


263

Table 6.1. Porous texture of the activated carbons.

Sample SBET

(m2/g)

VDRN2

(cm3/g)

VDRCO2

(cm3/g)

KUA 3080 1.19 0.57

KUA_800 2720 1.05 0.49

KUA-N 2960 1.18 0.52

KUA-N_500 2800 1.11 0.49

KUA-N_800 2770 1.09 0.48

KUA-COOH 2770 1.06 0.49

KUA-CONH2 2390 0.97 0.45

KUA-CONH2_500 2630 1.02 0.41

KUA-CONH2_800 2630 1.0 0.43

The CO2 micropore volumes (collected in Table 6.1) provide

information about the narrow microporosity of the samples (< 0.7 nm)

[23]. It is observed that the chemical functionalization and the heat

treatments produce a small decrease of the narrow microporosity. The

largest micropore volume was found for the pristine sample KUA,

followed by the KUA-N activated carbon and its related heat-treated

samples (KUA-N_500 and KUA-N_800).

6.3.1.2 Surface chemistry characterization

The surface chemistry of the activated carbons was characterized by

XPS and TPD. Table 6.2 collects the data related to the chemical

composition of the samples. Figures 6.2 and 6.3 collect the TPD profiles

obtained for all activated carbons. The pristine carbon material KUA

possesses a large amount of oxygen functionalities that decompose as

Chapter 6

264

CO and CO2 (Figure 6.2). As discussed in chapters 3 and 4, the chemical

modification methods produce the attachment of nitrogen to the surface

of the pristine activated carbon by consumption of oxygen functional

groups, as confirmed by XPS (Table 6.2). Figure 6.4 shows the

deconvoluted N1s spectra obtained for all N-doped activated carbons

and Table 6.3 summarizes the relative amount of the detected N

functional groups. In case of KUA-N sample, the attachment of nitrogen

occurs directly over the pristine carbon material via reaction mainly with

CO-evolving groups. Thus, this treatment produces nitrogen groups

derived from these oxygen functionalities, such as imines, amines and N

heterocycles (pyrroles, pyridones, and pyridines) [18,26-28] (Figure

6.2a). In case of KUA-CONH2 sample, an oxidation treatment is carried

out, previously to the incroporation of nitrogen, to increase the amount

of oxygen functional groups that can react during the amidation

treatment. The oxidation treatment with HNO3 produces an increase of

both CO and CO2 evolving-groups, and consequently the nitrogen

moieties produced after the nitrogen doping treatment are mainly in form

of amides (and cyclic amides), pyridines and pyrroles/pyridones [18,26-

28], as was described in chapter 4. Hence, both N-doped carbons

obtained at mild conditions evidence the presence of different N

functional groups, being KUA-CONH2 richer in those derived from

CO2-evolving groups (amides, lactams and imides).


265

Table 6.2. Surface composition of the activated carbons obtained by XPS and TPD.

Sample CO2

(µmol/g)

CO

(µmol/g)

O

(µmol/g)

O

(at. %)

N

(at. %)

KUA 450 1970 2870 8.8 0.3

KUA_800 170 620 960 2.3 -

KUA-N 450 1750 2640 7.5 3.7

KUA-N_500 250 1620 2120 5.1 2.3

KUA-N_800 130 505 720 3.0 1.7

KUA-COOH 1790 3770 7360 15.8 -

KUA-CONH2 1140 2370 4650 10.2 4.2

KUA-CONH2_500 250 1650 2140 8.0 3.4

KUA-CONH2_800 140 570 830 4.5 2.1

The surface chemistry of the carbon materials was subsequently

modified by heat treatments under inert atmosphere to produce activated

carbons with different composition while keeping the porous texture of

the pristine materials. These treatments produce a decrease of oxygen

and nitrogen groups (Table 6.2). TPD profiles allow to get information

about the changes produced in the oxygen functionalities (Figures 6.2

and 6.3). When the treatment at 500 ºC is carried out, some oxygen

functionalities are removed from their surface; the main loss comes from

CO2-evolving groups, since most part of these functionalities are not

thermally stable at temperatures higher than 600 ºC [29]. Thus, no

significant changes are produced on KUA-N at this temperature, as a

result of the small amount of CO2-evolving groups in this sample.

Consequently, KUA-N_500 shows a similar composition than that of

KUA-N, with slightly lower oxygen content. More important differences

are found for KUA-CONH2_500 due to the larger amount of oxygen

(specially, CO2-evolving groups) of the parent carbon (KUA-CONH2).

Chapter 6

266

After the treatment, most of the CO2 groups are removed from this

sample (80 % decrease, Table 6.2). Moreover, an important decrease of

the of CO-evolving groups is also observed (30 % decrease, Table 6.2).

Interestingly, both carbons heat-treated at 500 ºC show almost identical

content of oxygen groups (Table 6.2, Figures 6.2 and 6.3).

0

0.2

0.4

0.6

0.8

1

0 200 400 600 800 1000

CO

2 (µ

mo

l/g

s)

Temperature (ºC)

KUA

KUA_800

KUA-N

KUA-N_500

KUA-N_800

(a)

0

0.5

1

1.5

2

0 200 400 600 800 1000

CO

(µ

mo

l/g

s)

Temperature (ºC)

KUA

KUA_800

KUA-N

KUA-N_500

KUA-N_800

(b)

Figure 6.2. Comparison between (a) CO2 and (b) CO TPD profiles of KUA,

KUA_800, KUA-N, KUA-N_500 and KUA-N_800.


267

0

0.2

0.4

0.6

0.8

1

0 200 400 600 800 1000

CO

2(µ

mo

l/gs)

Temperature (ºC)

KUA

KUA-CONH₂

KUA-CONH₂_500

KUA-CONH₂_800

(a)

0

0.4

0.8

1.2

1.6

2

0 200 400 600 800 1000

CO

(µ

mo

l/g

s)

Temperature (ºC)

KUAKUA-CONH₂KUA-CONH₂_500KUA-CONH₂_800

(b)

Figure 6.3. Comparison between (a) CO2 and (b) CO TPD profiles of KUA, KUA-

CONH2, KUA-CONH2_500 and KUA-CONH2_800.

The heat treatment at 800 ºC removes most of oxygen. Around 70-

80 % of oxygen functionalities are removed from KUA, KUA-N and

KUA-CONH2 when they are heat treated at 800 ºC. N functionalities

also decrease upon heat treatment and around 50% of N remains after

the treatment. Hence, the obtained samples show similar oxygen content

(mainly in form of carbonyles [29]) and almost identical porous texture

(section 6.3.2.1).

Chapter 6

268

396398400402404

I (a

.u.)

Binding Energy (eV)

KUA-N_500

KUA-N_800

KUA-N(a)

396398400402404

I (a

.u.)

Binding Energy (eV)

KUA-CONH2_500

KUA-CONH2_800

KUA-CONH2(b)

Figure 6.4. N1s XPS spectra deconvoluted for the N-doped activated carbons.

The deconvolution of N1s XPS spectra obtained for all N-doped

activated carbons allows to assess the modifications produced on the

nitrogen moieties during the thermal treatments (Figure 6.4). Table 6.3

collects the percentage of each nitrogen group obtained from the N1s

XPS spectra. As previously described, the N-doped carbons at mild

conditions (KUA-N and KUA-CONH2) have similar nitrogen contents

but, according to the modification pathway, some of the generated

groups are different. The peak at 399.8 ± 0.2 eV can be assigned to

amides (and cyclic amides) and amines, being more likely the formation


269

of amides in KUA-CONH2 (due to the previous oxidation process) and

amines in KUA-N [30].


Sample Binding

Energy

(eV)

Functional

Group

N

(at. %)

Percentage of

N species

KUA-N 401.9 ± 0.2 Quaternary 0.4 10

400.7 ± 0.2 Pyrrole,

Pyridone

0.9 25

399.8 ± 0.2 Amide,

Lactam,

Amine, Imide

1.3 35

398.7 ± 0.2 Pyridine, Imine 1.1 30

KUA-N_500 401.5 ± 0.2 Quaternary 0.4

17

400.2 ± 0.2 Pyrrole,

pyridone

1.0 42

399.0 ± 0.2 Pyridine, Imine 0.9 41

KUA-N_800

402.7 ± 0.2 Oxidized N 0.2 14

400.8 ± 0.2 Pyrrole,

Pyridone

0.9 51

398.7 ± 0.2 Pyridine 0.6 35


pyrdidone

0.72 19

399.8 ± 0.2 Amide, lactam,

amine, imide

1.89 50

398.8 ± 0.2 Pyridine, imine 1.17 31

KUA-

CONH2_500

400.9 ± 0.2 Pyrrole,

pyridone

1.16 34


amine, imide

1.28 38

398.5 ± 0.2 Pyridine, imine 0.95 28

KUA-

CONH2_800

402.5 ± 0.2 Oxidized N 0.19 11

400.8 ± 0.2 Pyrrole,

pyridone

0.88 52

398.7 ± 0.2 Pyridine, imine 0.63 37

Chapter 6

270

When the samples are heat treated at 500 ºC, the nitrogen moieties

of lower thermal stability are removed, i.e., those with a single chemical

bond, such as amines and amides [31]. Indeed, the heat treatment of

KUA-N produces a loss of ⁓ 1.4 at. % N (Table 6.2), that is in agreement

with the amount of amines detected in KUA-N, and that does not appear

in the spectrum of KUA-N_500 (Figure 6.4a), evidencing the removal of

mainly this functional group during the treatment. In case of KUA-

CONH2, a lower loss of nitrogen is detected (0.8 at % XPS, Table 6.2).

Likewise, the main decrease is associated to the functional groups at

399.8 ± 0.2 eV, i.e., amide-like functionalities (Table 6.3, Figure 6.4b),

but they are still detected on KUA-CONH2_500. Moreover, this sample

evidences an increase of pyrrole/pyridones contribution (400.7 eV) that

can be related to the rearrangements of amides to generate pyrroles upon

heating [27], explaining the higher retention of nitrogen in this sample.

The heat treatments at 800 ºC strongly decrease the nitrogen groups

of the N-doped samples (Table 6.2, Figure 6.4). The samples KUA-

N_800 and KUA-CONH2_800 show similar nitrogen content (1.7 at. %

and 2.1 at. %, respectively, Table 6.2) and distribution of functionalities,

with a main contribution of pyrroles/pyridones and a lower presence of

pyridines and quaternary nitrogen (Table 6.3). These modifications are

undoubtedly related to the removal of nitrogen groups of lower thermal

stability as well as their conversion into more stable groups, such as

nitrogen heterocycles [27].

Thus, the combination of nitrogen doping at mild conditions with

post-thermal treatments allow the production of activated carbon with


271

similar microporosity and different surface functionalities. Hence, their

electrochemical characterization in organic electrolyte can be useful to

get information about the role of the different N functional groups in the

electrochemical performance of the materials as electrodes for

supercapacitors.


The electrochemical performance of the samples was evaluated as

electrodes in symmetric electrochemical capacitors in organic electrolyte

(1M TEMABF4/PC) by using a two-electrode cell configuration. The

cells were tested by GCD cycles at different current densities. Table 6.4

collects the capacitance and energy obtained from the galvanostatic test

and Figure 6.5 shows, as an example, some of the chronopotentiograms

obtained for the capacitors at 40 mA/g. The curves evidence an ideal

capacitive behaviour, characterized by the triangular shape, for all the

activated carbon-based capacitors. All of them provide large capacitance

(36-41 F/g) and energy values (32-37 Wh/kg) as consequence of their

well-developed microporosity. The small differences found between the

samples are related to their slightly different apparent surface area, as

confirmed by the similar surface capacitance values obtained for these

materials (13-15 µF/cm2). Interestingly, the lowest values were obtained

for the non-doped samples with higher oxygen content (KUA and KUA-

COOH), evindencing a detrimental effect derived from these functional

groups [32].

Chapter 6

272

0

0.5

1

1.5

2

2.5

3

0 2000 4000 6000

Volt

age (

V)

Time (s)

KUA

KUA-N

KUA-CONH₂

(a)

0

0.5

1

1.5

2

2.5

3

0 2000 4000 6000

Vo

lta

ge (

V)

Time (s)

KUA-N_800

KUA-CONH₂_500

KUA-CONH₂_800

(b)(b)

Figure 6.5. GCD cycles obtained for activated carbons-based symmetric capacitors.

1M TEMABF4/PC. j = 40 mA/g. V = 2.5 V.

Figure 6.6 shows the evolution of the capacitance when increasing

the current density during the galvanostatic cycling test. As observed in

aqueous electrolyte (Chapter 3), the oxidation process produces a

detrimental effect on the rate performance of KUA-COOH when used as

electrode for supercapacitors, and again, when the amidation treatment is

carried out, the replacement of oxygen groups by nitrogen functionalities

with higher conductivity (cyclic amides, pyrroles, etc.) [30, 33]


273

improves the behaviour, and consequently, a higher capacitance is

retained when the current density is increased (Figure 6.6a). The heat

treatments of the activated carbons (Figure 6.6b) produce an

improvement of the rate performance, but no significant benefits are

observed for the N-doped samples, evidencing that the increase of the

charge-discharge rate is mainly a consequence of an increase of

conductivity due to the removal of electron-withdrawing oxygen

functionalities.

Table 6.4. Gravimetric capacitance, surface capacitance, energy density, retention of

capacitance at 1000 mA/g (C1/C0), retention of capacitance after floating test (C0/Cf),

increase of resistance and integrated leakage current obtained for all activated carbon-

based capacitors. V = 2.5 V. j = 40 mA/g.

Capacitor C0

(F/g)

C0/SBET

(mF/cm2)

E

(Wh/kg)

C1/C0

(%)

Cf/C0

(%)

ΔR

(Ω)

IL

(mA

h)

KUA 41 13 37 45 29 33 37

KUA_800 38 14 33 48 43 30 29

KUA-N 41 14 36 42 39 20 32

KUA-N_500 41 15 35 45 39 32 36

KUA-N_800 40 14 35 49 44 28 31

KUA-COOH 36 13 31 30 20 50 26

KUA-CONH2 37 15 32 49 55 13 22

KUA-CONH2_500 39 15 33 37 22 86 33

KUA-CONH2_800 38 15 32 72 33 34 41

Chapter 6

274

0

10

20

30

40

50

0 200 400 600 800 1000

Cap

aci

tan

ce (F

/g)

j(mA/g)

KUA

KUA_COOH

KUA-CONH₂

(a)

0

10

20

30

40

50

0 200 400 600 800 1000

Cap

acit

an

ce (

F/g

)

j (mA/g)

KUA_800

KUA-N_800

KUA-CONH₂_800

(b)

Figure 6.6. Gravimetric capacitance at different current densities for the symmetric

capacitors. 1M TEMABF4/PC. j = 40, 100, 200, 500, 750 and 1000 mA/g. V = 2.5 V.

The activated carbon-based capacitors were submitted to a

durability test to assess the effect of surface chemistry in the

electrochemical stability of the carbon materials. The most common

method for the assessment of the durability of energy storage devices

consists in applying thousands of GCD cycles at the operative voltage of

the devices [9]. Since capacitors are based in electrostatic processes and

consequently provide a much larger cycle life than batteries, this

methodology is considerably more time demanding for the former than


275

for the latter [1]. An alternative to this procedure is the use of a floating

test, consisting in keeping the cell charged during long time in order to

accelerate the degradation of the device [34-38]. Moreover, it is well-

known that electrochemical capacitors experience degradation at high

temperature [5,32,39,40]. Hence, the combination of high voltage

holding with the use of high temperature constitutes an adequate

procedure to accelerate the degradation of electrochemical capacitors.

Figure 6.7 shows, as an example, the evolution of capacitance and

coulombic efficiency during the steps of the durability test (described in

section 6.2.3) for some of the capacitors. Table 6.4 compiles the

parameters (retention of capacitance (Cf/C0), increase of resistance (ΔR)

and integrated leakage current (IL)) related to this test determined for all

capacitors. The analysis of these data allows to throroughly discuss the

effect of surface chemistry modification on the performance of carbon

materials as electrodes for supercapacitors in organic electrolyte.

6.3.2.1. Effect of surface chemistry modification at mild conditions

Figure 6.7a shows the performance of the pristine carbon material

during the durability test. Before the floating step, galvanostatic cycling

is performed at different temperatures and voltage conditions, being

these parameters subsequently increased with the evolution of cycles.

The increase of temperature and voltage does not affect the gravimetric

capacitance, but decreases the coulombic efficiency evidencing the

ocurrence of faradaic processes under these conditions. The harmful

effect of the floating conditions is clearly observed in the capacitance

Chapter 6

276

values determined after the durability test. For the pristine sample, only

a 29% of capacitance is retained after the ageing test.

0

20

40

60

80

100

120

0

20

40

60

80

100

0 5 10 15 20 25 30 35

Eff

icie

ncy (

%)

Cap

acit

an

ce (

F/g

)

Nº Cycles

KUA_800

KUA-N_800

KUA-CONH₂_800

40ºC0-2.5V

70ºC0-3.2V

70ºC0-3.2V

70ºC3.2V

100h

70ºC0-3.2V

70ºC0-2.5V

40ºC0-2.5V

(b)

Figure 6.7. Evolution of capacitance and coulombic efficiency during the durability

test. 1M TEMABF4/PC. j = 40 mA/g.

In general, the modified samples show a similar trend than the

pristine samples. Thus, a remarkable loss of capacitance (50-70 %) is


277

detected after the floating test for all supercapacitors (Table 6.4). The

differences in the performance are consequence of their different surface

chemistry. First, it is confirmed that the oxidation process strongly

affects the behavior of the capacitor, since a retention of capacitance of

20 % and an increase of resistance of 50 Ω is detected for KUA-COOH

based capacitor. Thus, the ageing protocol produces an enormous

degradation of the electrodes due to reaction between the electrolyte and

the oxygen groups present on the surface of the electrodes [32].

Otherwise, the N-doping treatments at mild conditions improve the

performance of the carbon materials. Specially, KUA-CONH2 based

capacitor shows the best performance of the all tested cells (Table 6.4).

The retention of capacitance obtained is 55%, which means an increase

of 26 % in comparison with the pristine carbon based capacitor (KUA).

Moreover, KUA-CONH2 based capacitor shows the lowest integrated

leakage current, that is associated to the ocurrence of faradaic processes

during the charging of the device (oxidation, gasification, etc.) [6, 38].

As discussed in section 6.3.1.2, KUA-CONH2 shows predominancy of

amide-like functional groups as well as a lower amount of nitrogen

heterocycles (pyridines and pyrroles/pyridones). Also, the sample has a

larger content of oxygen functional groups than the pristine sample as

consequence of the oxidation process carried out prior to the amidation.

However, the increase in stability cannot be related to the oxygen

functionalities since they damage the electrochemical stability, as

evidenced by KUA-COOH based capacitor. Actually, the increase of

retention of capacitance detected for KUA-CONH2 is even higher when

compared to the former oxidized sample KUA-COOH (36%). Thus, the

Chapter 6

278

improvement on the performance is undoubtely related to the generation

of stable nitrogen surface functional groups.

Furthermore, KUA-N based capacitor also has better performance

than the pristine carbon based capacitor (Figure 6.7a, Table 6.4),

evidencing an increase of capacitance retention of 11%. Since this

sample is synthesized by a direct replacement of oxygen groups by

nitrogen functionalities of KUA, the improvement might be a result of

the combined effect caused by the removal of detrimental oxygen groups

and generation of nitrogen moities with higher electrochemical stability

(pyrroles, pyridines, etc.). Moreover, the comparison of the N-doped

samples (KUA-N and KUA-CONH2) allows to distinguish the effect of

the different nitrogen functional groups and evidence a remarkable

stabilizing effect related to amides and cycles amides, since these

functional groups only exist on the surface of KUA-CONH2 and the

corresponding capacitor shows a higher retention of capacitance and a

lower increase of resistance than KUA-N based capacitor.

In order to deepen into the role of the different functionalities, the

effect of heat treatments on the performance of activated carbons as

electrodes for supercapacitors is throroughly discussed.

6.3.2.2. Effect of surface chemistry modification by heat treatments

The durability of the capacitors built with the heat treated samples

was also analyzed and the related parameters are collected in Table 6.4.

Some meaningful differences are found when comparing the capacitor

performance of the samples heat treated at 500 ºC. KUA-N_500 based


279

capacitor has better performance than the pristine carbon material KUA,

evidenced by an increase of retention of capacitance of 10 %. This

sample has a similar surface chemistry than the parent KUA-N, with

slighly lower content of oxygen and nitrogen groups (section 6.2.1.2).

Thus, the improvement is related, again, to the removal of detrimental

oxygen functional groups as well as the generation of more stable

nitrogen groups. Nevertheless, the analogous KUA-CONH2_500 based

capacitor shows lower stability through the durability test (22 % of

capacitance retention, Table 6.4) than the related carbon based

capacitors. Interestingly, the heat treated N-doped carbons (KUA-

CONH2_500 and KUA-N_500) show different behavior upon durability

even though their surface chemistry is almost identical in terms of

amount of oxygen and nitrogen heteroatoms (section 6.2.1.2). The main

difference arises from their parent carbons (KUA-CONH2 and KUA-N).

KUA-CONH2 has a larger amount of surface functionalities, that upon

heating may decompose and generate a larger amount of reactive sites.

Consequenly, the capacitor based on the activated carbon prepared by

the heat treatment of KUA-CONH2 might experience more interactions

with the electrolyte, leading to a higher degradation process.

The heat treatment of the samples at 800 ºC also affect the durability

of the devices. It is clearly observed that the treatment under these

conditions improve the performance as consequence of the high decrease

of oxygen functionalities. Thus, KUA_800 and KUA-N_800 evidence a

similar behaviour upon the floating test, with an increase of retention of

capacitance of 14 % for both capacitors (in comparison with KUA-

Chapter 6

280

based capacitor). Hence, no further improvement results from the

nitrogen groups present on KUA-N_800. In case of KUA-CONH2_800

based capacitor, the values of retention of capacitance are higher than

the heat treated at 500 ºC due to the generation of surface functionalities

with higher electrochemical stability at this temperature

(pyrroles/pyridones). However, the performance is still poorer than that

found for the capacitors based on heat treated carbons at 800 ºC

(KUA_800 and KUA-N_800), since the surface reactivity of this carbon

material is higher than the found in the related materials.

These results evidence that heat treatments improve the performance

of carbon electrodes upon durability mostly due to the removal of

detrimental oxygen functionalities. However, the performance is further

improved when the doping strategies are carried out by wet methods at

low temperature. KUA-N and KUA-CONH2 show better performance

than the related heat treated samples not only in terms of retention of

capacitance but also in the increase of resistance. Indeed, both N-doped

carbons based capacitors evidence the lowest increase of resistance of

the whole series. This is a consequence of combining the beneficial

effect of the decrease in oxygen content with the production of nitrogen

functional groups with remarkable stabilizing effect, specially in the case

of amide-like functionalities. Also, the generation of reactive sites

during the thermal treatment is avoided under these conditions. Thus, the

nitrogen doping method at mild conditions is a promising strategy to

improve the performance of any carbon material as electrode for

supercapacitors.


281

As a proof of concept, a commercial activated carbon used for

supercapacitor application was succesfully functionalized with nitrogen

and tested upon durability in organic electrolyte (Annex 6.1). After the

durability test, an increase of retention of capacitance of 7 % was

determined, evidencing the viability of this methodology to improve the

electrochemical stability of carbon materials.

6.4. Conclusions

Activated carbons with similar porosity but different surface

chemistry have been prepared by combining chemical functionalization

methods at mild conditions and post-thermal treatments. The doping

methods at low temperature produce the incroporation of 4% at. of N in

form of different functionalities by consumption of oxygen moieties.

The heat treatments diminish the content of surface functionalities and

produce rearrangements of the nitrogen groups. The surface composition

and porosity of the heat-treated samples is almost identical for the whole

series of carbons.

The surface chemistry of these carbon materials clearly influences

the electrochemical performance when tested as electrodes for

electrochemical capacitors in organic electrolyte. They show large

capacitance values with no significant differences as consequence of

their similar porous texture. The presence of oxygen functional groups

affects the rate performance of the capacitor due to the decrease of

conductivity, while nitrogen functional groups (cyclic amides, pyridines

and pyrroles) slightly increase the conductivity of the carbon material.

Chapter 6

282

The effect of surface functionalities upon durability was thoroughly

studied. The oxygen functionalities strongly damage the performance of

the activated carbons. Thus, the heat treatments of the samples produce

an improvement of the electrochemical stability due to the decrease of

detrimental oxygen groups. However, the performance is further

increased by nitrogen doping at mild conditions, since the treatment

combines the positive effect of removing oxygen groups with their

replacement by nitrogen groups with high electrochemical stability,

which is specialy beneficial in case of the generation of amide-like

functional groups.

6.5 References



York, 1999.

[2] W. Raza, F. Ali, N. Raza, Y. Luo, K.-H. Kim, J. Yang, S. Kumar,

A. Mehmood, E.E. Kwon, Recent advancements in supercapacitor

technology, Nano Energy 52 (2018) 441–473.

[3] F. Béguin, E. Frackowiak, eds., Carbons for Electrochemical

Energy Storage and Conversion Systems, 1st ed., Boca Raton, Fl:

Taylor & Francis (CRC Press), 2010.


layer capacitors: A perspective article, J. Power Sources 326

(2016) 534–540.






283



[7] S. Shiraishi, Heat-Treatment and Nitrogen-Doping of Activated

Carbons for High Voltage Operation of Electric Double Layer

Capacitor, Key Eng. Mater. 497 (2011) 80–86.

[8] M.J. Mostazo-López, R. Ruiz-rosas, E. Morallón, D. Cazorla-



Hydrogen Energy 41 (2016) I9691–I9701.

[9] E. Frackowiak, F. Béguin, Carbon materials for the

electrochemical storage of energy in capacitors, Carbon 39 (2001)

937–950.




Sci. 6 (2013) 2839–2855.

[11] Z. Yang, H. Nie, X. Chen, X. Chen, S. Huang, Recent progress in

doped carbon nanomaterials as effective cathode catalysts for fuel

cell oxygen reduction reaction, J. Power Sources 236 (2013) 238–

249.


Morallón, H. Nishihara, T. Kyotani, J. Rodríguez-Mirasol, T.

Cordero, Enhanced electro-oxidation resistance of carbon

electrodes induced by phosphorus surface groups, Carbon 95

(2015) 681–689.






Amorós, J. Geng, N. Soin, E. Siores, B.F.G. Johnson, On the

origin of the high capacitance of nitrogen-containing carbon

nanotubes in acidic and alkaline electrolytes, Chem. Commun. 50

Chapter 6

284

(2014) 11343–11346.

[15] M.J. Mostazo-López, R. Ruiz-Rosas, A. Castro-Muñiz, H.


Ultraporous nitrogen-doped zeolite-templated carbon for high

power density aqueous-based supercapacitors, Carbon 129 (2018)

510–519.

[16] W. Shen, W. Fan, Nitrogen-containing porous carbons: synthesis

and application, J. Mater. Chem. A 1 (2013) 999–1013.







fibres, Carbon 41 (2003) 1925–1932.

[19] H. Itoi, H. Nishihara, T. Kyotani, Effect of Heteroatoms in

Ordered Microporous Carbons on Their Electrochemical

Capacitance, Langmuir 32 (2016) 11997–12004.

[20] H. Nishihara, P. Hou, L. Li, M. Ito, M. Uchiyama, T. Kaburagi,

A. Ikura, J. Katamura, T. Kawarada, K. Mizuuchi, T. Kyotani,

High-Pressure Hydrogen Storage in Zeolite-Templated Carbon, J.

Phys. Chem. C 113 (2009) 3189–3196.





252–265.







285


4589–4596.




aqueous electrolyte, Carbon 41 (2003) 1765–1775.




Carbon 44 (2006) 2642–2651.





597–608.



1027.


graphene analyzed by X-ray photoelectron spectroscopy, Carbon.

70 (2014) 59–74.

[29] J.. Figueiredo, M.F.. Pereira, M.M.. Freitas, J.J.. Órfão,

Modification of the surface chemistry of activated carbons,

Carbon 37 (1999) 1379–1389.


Zhu, G.Q. Lu, Nitrogen-Enriched Nonporous Carbon Electrodes


(2009) 1800–1809.

[31] R. Arrigo, M. Hävecker, S. Wrabetz, R. Blume, M. Lerch, J.

McGregor, E.P.J. Parrott, J.A. Zeitler, L.F. Gladden, A. Knop-

Gericke, R. Schlögl, D.S. Su, Tuning the acid/base properties of

nanocarbons by functionalization via amination, J. Am. Chem.

Soc. 132 (2010) 9616–9630.

Chapter 6

286





[33] V.. Strelko, V.. Kuts, P.. Thrower, On the mechanism of possible







[35] D. Weingarth, H. Noh, a Wokaun, R. Kötz, A reliable

determination method of stability limits for electrochemical

double layer capacitors, Electrochim. Acta 103 (2013) 119–124.

[36] D. Weingarth, A. Foelske-Schmitz, R. Kötz, Cycle versus voltage

hold - Which is the better stability test for electrochemical double

layer capacitors?, J. Power Sources 225 (2013) 84–88.

[37] P.W. Ruch, D. Cericola, A. Foelske-Schmitz, R. Kötz, A.

Wokaun, Aging of electrochemical double layer capacitors with

acetonitrile-based electrolyte at elevated voltages, Electrochim.

Acta 55 (2010) 4412–4420.

[38] P. Ratajczak, K. Jurewicz, F. Béguin, Factors contributing to

ageing of high voltage carbon/carbon supercapacitors in salt

aqueous electrolyte, J. Appl. Electrochem. 44 (2014) 475–480.

[39] R. Kötz, M. Hahn, R. Gallay, Temperature behavior and

impedance fundamentals of supercapacitors, J. Power Sources154

(2006) 550–555.

[40] O. Bohlen, J. Kowal, D.U. Sauer, Ageing behaviour of

electrochemical double layer capacitors: Part I. Experimental

study and ageing model, J. Power Sources 172 (2007) 468–475.


287

ANNEX TO CHAPTER 6.

Chemical functionalization of a commercial activated

carbon with nitrogen groups

In Chapter 6, the effect of nitrogen doping on the performance of

activated carbons as electrodes for supercapacitors has been studied. It

was found that nitrogen doping at mild conditions can improve the

electrochemical stability and consequently increase the durability of the

devices. This is mainly explained by the replacement of oxygen

functional groups (that degrade the carbon materials at extreme

potentials) by nitrogen groups with high electrochemical stability. Thus,

it appears as a promising method for improving the performance of any

carbon electrode, independently of the synthesis method.

In this annex, the nitrogen doping strategy developed in Chapter 4 is

used to functionalize a commercial activated carbon (YP50F, Kuraray

Chemical, Japan) that is used as electrode in supercapacitors. The

nitrogen doping method described in section 6.2.2.2 was applied to

YP50F and the obtained sample was named as YP50F-N.

Table A6.1. Textural properties and elemental surface composition (XPS and TPD) for

YP50F and YP50F-N.

Sample NXPS

(at.%)

OXPS

(at.%)

CO2 TPD

(µmol/g)

COTPD

(µmol/g)

SBET

(m2/g)

VDRN2

(cm3/g)

YP50F - 7.0 160 520 1790 0.71

YP50F-N 2.3 5.8 100 560 1740 0.69

Chapter 6

288

Table A6.1 summarizes the textural properties and surface

composition of the pristine and modified carbons. The nitrogen

functionalization protocol (avoiding the oxidation process) over YP50F

produces the attachment of nitrogen functional groups (2.3 at. % XPS,

Table A6.1) as well as a decrease of oxygen of 1.2 at. % XPS,

evidencing that nitrogen is anchored to the surface by substitution of

oxygen moieties. Figure A6.1 shows the N1s XPS spectrum recorded for

YP50F-N, that evidences that nitrogen is attached to the surface in form

of pyridines, amines/amides and pyrroles/pyridones [1–4].

396398400402404

I (a

.u.)

Binding Energy (eV)

Figure A6.1. N1s XPS spectrum obtained for YP50F-N activated carbon.


289

0

200

400

600

800

1000

0 0.2 0.4 0.6 0.8 1

Ad

sorb

ed

vo

lum

e (

cm

3/g

)

P/Po

YP50F YP50F-N

Figure A6.2. N2 adsorption-desorption isotherms of the activated carbons YP50F and

YP50F-N.

The activated carbon YP50F provides large apparent surface area

with well-developed microporosity (Figure A6.2). As happened when

using the activated carbon KUA (Chapters 4 and 6), the chemical

functionalization does not reduce or block the microporosity of the

pristine carbon material and allows the preservation of the whole

apparent surface area. Thus, these carbons have identical porous texture

but different surface chemistry.

The carbon materials were electrochemically characterized in a

three-electrode cell in aqueous electrolyte. Figure A6.3 shows the CVs

obtained in a potential window where the materials are stable. Both

evidence a rectangular shape characteristic of porous carbons and

provide large capacitance values (156 and and 131 F/g for non-doped

and N-doped activated carbons, respectively). However, the CVs display

Chapter 6

290

some differences in the electrochemical performance of the materials.

YP50F exhibits the characteristic redox pair associated with the

quinone-hydroquinone redox process [5], while this response is not

observed in the case of the N-functionalized carbon. As seen in Chapter

4, these electroactive oxygen groups can be responsible of degradation

processes under oxidation conditions, and consequently, the lower

occurrence of this process may lead to an improvement of the durability

of supercapacitors based on this N-doped carbon. To deepen into the

performance of this material, the durability of the corresponding

capacitors was evaluated in organic medium under severe conditions of

temperature and voltage, as described in Chapter 6. Figure A6.4 shows

the evolution of capacitance and coulombic efficiency during the

durability test performed for YP50F and YP50F-N. As happened with

N-doped KUA, the nitrogen doping does not affect the capacitance in

organic electrolyte due to preservation of the microporosity. Also, the

nitrogen functionalities do not increase the capacitance. However, there

is an improvement of retention of capacitance upon durability of 7 %

that is related to the generation of nitrogen groups with high

electrochemical stability produced by the nitrogen doping at mild

conditions.


291

-250

-200

-150

-100

-50

0

50

100

150

200

-0.2 0 0.2 0.4 0.6 0.8

C (

F/g

)


Figure A6.3. Cyclic voltammograms in the potential range between 0V and 0.6V for

YP50F and YP50F-N electrodes. 1M H2SO4. V=1 mV/s.

0

20

40

60

80

100

120

0

10

20

30

40

50

60

70

0 5 10 15 20 25 30 35

Eff

icie

ncy (

%)

Cap

acit

an

ce (

F/g

)

Nº Cycles

YP50F

YP50F-N

70ºC0-2.5V

70ºC0-3.2V

70ºC3.2V

100h

70ºC0-3.2V

70ºC0-2.5V

40ºC0-2.5V

40ºC0-2.5V

Figure 6.4. Evolution of capacitance and coulombic efficiency during the durability

test for YP50F and YP50F-N based capacitors. 1M TEMABF4. j = 40 mA/g.

Chapter 6

292

In conclusion, a microporous commercial activated carbon used for

supercapacitor application has been functionalized with nitrogen groups

by following organic chemistry protocols. The chemical method allowed

the incorporation of 2 N at. % in form of different groups, by

consumption of oxygen functionalities. The microporosity of the pristine

carbon is fully retained after the chemical treatment. The nitrogen

doping affects the electrochemical response of the carbon in acid

medium, by removing the electroactivity associated to the quinone-

hydroquinone redox pair. Consequently, this leads to an improvement of

the electrochemical stability of the N-doped carbon which is

demonstrated when tested as electrode for electrochemical capacitor in

organic electrolyte.

References.




fibres, Carbon 41 (2003) 1925–1932.





597–608.



1027.



70 (2014) 59–74.


293




Carbon 44 (2006) 2642–2651.

Chapter 7



non-conventionalelectrolytes

Electrochemical performance of N-doped activated carbons in non-conventional electrolytes

297

7.1. Introduction

Electrochemical double-layer capacitors (EDLC), or

supercapacitors, are currently the most suitable energy storage devices

for a wide range of high-power applications. These devices are based on

the physical electrosorption of ions at the interface of an electrode and

an electrolyte [1–3]. Consequently, EDLC can be charged and

discharged in seconds, providing high specific power (10 kW/kg) and

long-cycle life (> 500000 cycles). Nevertheless, the energy density

needs to be increased to introduce this technology into new applications,

such as electric vehicles. This parameter depends on the capacitance and

the cut-off voltage employed during the operation of the device, which

are mostly governed by the two main components of supercapacitors: the

electrolyte and the electrode material. The current commercially

available supercapacitors are based on activated carbons, whose large

surface area produces high capacitance values, and organic electrolytes,

based on solutions of an ammonium salt in propylene carbonate (PC) or

acetonitrile, which can operate at 2.5-2.8 V [1,4]. One of the main

strategies for improving their performance is the use of innovative

electrolytes, such as those based on ionic liquids (ILs) or new

conducting salts (in organic solvents), since they can expand the

operation voltage to values larger than 3V [5,6].

The development of new electrolytes based on ILs has attracted

consiberable attention due to their wide electrochemical stability

window (>3V), high thermal stability, low volatility and non-

flammability [5,7–10]. However, their high viscosity and low

Chapter 7

298

conductivity (in comparison with conventional organic electrolytes)

hinders the capacitance and power provided by EDLC based in solvent-

free ILs. Recently, the use of mixtures of ILs with organic solvents (PC

or acetonitrile), in which IL acts as a conducting salt, has appeared as a

promising alternative because it allows the production of electrolytes

with low viscosity and higher conductivity. However, the operative

voltage oof EDLCs containing these electrolytes is often limited in

comparison with that of EDLC based on solvent-free ILs, but still

overcomes the values of conventional organic electrolytes.

An alternative to the use of ILs is the investigation of new

conducting salts, dissolved in an organic solvent (PC or acetonitrile), for

the preparation of advanced electrolytes [5]. The use of these salts

allows the preparation of electrolytes with high conductivity and low

viscosity. Furthermore, the EDLCs based on these electrolytes perform

at voltages larger than 3V. Several studies have been dedicated to

different cations (phosphonium and quaternary ammonium-based

cations) combined mainly with tetrafluoroborate (BF4-) as anion.

Recently, pyrrolidinium-based cations have appeared as promising

candidates for the generation of advanced electrolytes [1,4]. Since the

ions of the conducting salts are involved on the formation of electrical

double layer, the selection of a conducting salt with adequate ion sizes,

chemical and electrochemical stability is crucial for the development of

advanced electrolytes [5].

Moreover, the appropriate selection of ions has a strong influence on

the carbon-electrolyte interface [11–13]. To take full benefit from the


299

advantages of innovative electrolytes, the physicochemical properties of

carbon materials must be also carefully considered. In this sense, carbon

materials with well-developed microporosity and adequate pore size

distribution (with optimized ion size/pore size ratios) are highly

desirable since they provide high specific capacitance [11–14] However,

their electrochemical stability can be compromised when using high

voltage operation conditions. We have proven previously that nitrogen-

doped carbon materials can improve several properties, such as electrical

conductivity or electrochemical stability (in aqueous and organic

electrolytes) [15–17], avoiding detrimental reactions with the electrolyte

and increasing the durability of the device. However, the role of surface

chemistry, and specifically nitrogen functionalities, has not been

assessed yet in non-conventional electrolytes. As discussed above, the

performance of EDLC strongly depends on the interation at the

electrode-electrolyte interface. Thus, surface chemistry appears as a key

parameter to be investigated for the development of EDLCs based in

non-conventional electrolytes.

In this chapter, the effect of nitrogen functionalization treatments on

the electrochemical properties of the activated carbons with high surface

area was analyzed using non-conventional electrolytes. For this, nitrogen

functionalization is carried out throught chemical post-treatments at mild

conditions over a superporous activated carbon with tailored porous

structure. The effect of nitrogen functional groups on the

electrochemical performance of the carbon materials is assessed in two

PC-based electrolytes: (i) a non-conventional conducting salt (1-butyl-1-

Chapter 7

300

methylpyrrolidium tetrafluoroborate (Pyr14BF4)) in PC and (ii) a mixture

of ionic liquid (1-butyl-1-methylpyrrolidium bis-(tri-

fluoromethylsulfonyl)imide (Pyr14TFSI)) with PC.

7.2. Materials and Methods


A superporous activated carbon prepared in our laboratory has been

used as the starting material for nitrogen incorporation via post-

modification treatments based in organic chemistry reactions. The

pristine material (named as KUA), has been obtained by chemical

activation of a Spanish anthracite with KOH using an impregnation ratio

of activating agent to raw material of 4:1 and an activation temperature

of 750º C under inert atmosphere, which was held for 1 hour. More

details about the preparation process are available elsewhere [18].

7.2.2. Chemical functionalization of activated carbon


two different strategies based on the organic chemistry protocols. This

methodology was accurately described in chapters 3 and 4. Briefly, the

first approach consisted in a three-step protocol: (i) chemical oxidation

with HNO3, (ii) treatment with SOCl2 and (iii) amidation reaction with

NH4NO3/DMF and pyridine. The obtained sample was named as KUA-

CONH2. In the second method, the third step of the first protocol is

directly applied over pristine sample (KUA). This sample was named as

KUA-N.


301












adsorption isotherms (relative pressures from 0.0001 to 0.25) [19].




Microtech Multilab 3000 spectrometer, equipped with an Al anode. TPD

experiments were performed by heating the samples (10 mg) to 950º C

(at a heating rate of 20º C/min) under a helium flow rate of 100 mL/min.

The analyses were carried out by using a TGA-DSC instrument (TA

Instruments, SDT Q600 Simultaneous) coupled to a mass spectrometer

(Thermostar, Balzers, BSC 200).

7.2.4. Electrolyte preparation

Two different electrolytes were employed in this chapter: 1M

Pyr14TFSI/PC and 1M Pyr14BF4/PC. The IL Pyr14TFSI (IoLiTec, 99.5%)

Chapter 7

302

was dried by using a molecular sieve until the water content was lower

than 20 ppm. The conducting salt Pyr14BF4 was purified as described

elsewhere [20]. PC (Sigma Aldrich, 99.7 %) was used as organic

solvent. The water content of the electrolytes was determined to be

lower than 30 ppm, as measured by Karl-Fisher technique.


7.2.5.1. Three electrode cell configuration

The electrochemical characterization of the carbon materials was

performed by using a Swagelock-type cell in a three-electrode

configuration. The working electrodes were prepared by mixing the

activated carbon with acetylene black and polytetrafluoroethylene

(PTFE) as binder in a ratio of 85:10:5 (w/w). The total weight of the

electrode was 4-5 mg (dry basis). For shaping the electrodes, a sample

sheet was cut into a circular shape with an area of 0.95 cm2. The counter

electrodes were prepared using a commercial activated carbon as active

material (Norit DLC Super30, SBET = 1618 m2/g) with a mass loading

larger than 30 mg/cm2, as described elsewhere [14]. The electrodes were

dried overnight at 120 ºC under vacuum prior to assembling the cell.

The cells were prepared inside an Ar-filled glovebox with water and

oxygen contents < 1 ppm. The working and counter electrodes were

tightly pressed against each other and separated by a glass microfiber

membrane (Whatman GF/D, thickness: 675 µm). Prior to assembling the

cell, the separator and the electrodes were soaked with 140 µL of


303

electrolyte and kept under vaccum for 5 minutes. Ag was used as

reference electrode in all cases.

The electrochemical characterization of all samples was tested by

CV at sweep rate of 1 mV/s. The capacitance was calculated from the

electric charge of the CV. The results are expressed in F/g, considering

the weight of the active material of the working electrode.

7.2.5.2. EDLC investigation

Asymmetric in mass capacitors were assembled for all carbon

materials in an Argon glovebox in 1M PYR14TFSI/PC electrolyte. For

this, two electrodes (surface area: 0.283 cm2) were prepared with a

weight of ~1.1 mg (active phase) each. Supercapacitors were constructed

by pressing both electrodes against each other and separating them by a

filter glass microfiber membrane (Whatman GF/D, thickness: 675 µm).

These devices were characterized by CV at different scan rates and

galvanostatic charge-discharge (GCD) cycles at current densities from 1

to 20 A/g in 1M PYR14TFSI/PC. CV measurements were performed in a

Biologic VSP 300 and the galvanostatic tests in a Arbin SCTS

galvanostat. A durability test for asymmetric capacitors was performed

by 2000 GCD cycles at a current density of 1 A/g and a voltage of 3.0 V.

Current density and specific capacitance is defined based on the total

active weight of the carbon material included in both electrodes.

The energy density and power density of asymmetric

supercapacitors was calculated in order to obtain all relevant information

Chapter 7

304

about their performance. Energy density was obtained during the

discharge cycle by the following equation (1):

E = (7.1)

Where V1 is the cell voltage of charge (3V) and V2 the voltage of

discharge (0 V).

Power density was calculated according to equation (2):

(7.2)

Where td is the time of discharge in the GCD cycle.

7.3 Results and Discussion

7.3.1 Surface chemistry and porous texture characterization

Table 7.1 summarizes the main parameters related to the

physicochemical properties of the pristine and N-functionalized

activated carbons. These properties and the effect of nitrogen doping on

the surface chemistry and porous texture of the samples has been

previously discussed (in chapters 3, 4 and 6) and are summarized below.

As evidenced by XPS analyses, nitrogen doping was satisfactorily

achieved for KUA-N and KUA-CONH2 samples through consumption

of oxygen functionalities. Both N-doped carbons show similar nitrogen

spectra and content (⁓ 4 at. % XPS), but the functionalities generated on

their surfaces are different. This is a consequence of the different oxygen

functional groups involved in the reaction. For the production of KUA-N

sample, the nitrogen doping is carried out directly over the pristine


305

carbon material; as a result, the nitrogen functionalities generated on this

carbon are derived mainly from CO-evolving groups. Thus, the main

functional groups found over this carbon material are amines,

pyridines/imines and pyrroles/pyridones (chapters 4 and 6) [16].

Nevertheless, the nitrogen functional groups produced on KUA-CONH2

involve the consumption of CO2 and CO evolving-groups, since the

pristine carbon was previously oxidized to increase the oxygen content

prior to the nitrogen doping step. Hence, there is a contribution of groups

derived from CO2 (amide-like functionalities) and CO groups (pyridines,

pyrroles, etc.), as previously discussed in chapter 3 [21]. Consequently,

the N-doped carbons show similar nitrogen content, but different surface

functionalities. Also, KUA-CONH2 has a larger amount of oxygen

functional groups than the other carbons.

Table 7.1. Surface composition determined for all activated carbons by XPS and TPD.

Sample CXPS

(at. %)

NXPS

(at.%)

OXPS

(at.%)

CO2 TPD

(µmol/g)

CO TPD

(µmol/g)

O TPD

(µmol/g)

KUA 90.9 0.3 8.8 500 2100 3100

KUA-N 87.6 3.6 8.8 450 1750 2640

KUA-CONH2 88.8 4.1 12.0 1140 2370 4650

The porous texture of the samples was assessed by N2 and CO2

adsorption isotherms. The pristine carbon material is a microporous

material with high apparent surface area (chapter 4) [16]. Moreover,

most of the microporosity of the pristine material remains in the N-

doped samples due to the mild conditions of the treatment. When the

nitrogen doping is carried out directly over the pristine carbon (KUA-N

Chapter 7

306

sample), the microporosity is fully preserved since only chemical

reactions involving CO groups take place to attach nitrogen moieties.

However, KUA-CONH2 evidences some loss of the microporosity (30

%) mainly due to the oxidation pre-treatment done prior to the nitrogen

functionalization treatment [21].

Table 7.2. Porous texture parameters of the activated carbons.

Sample SBET

(m2/g)

SNLDFT

(m2/g)

VDRN2

(cm3/g)

VDRCO2

(cm3/g)

KUA 3080 2230 1.19 0.57

KUA-N 2960 2270 1.18 0.52

KUA-CONH2 2390 1800 0.97 0.45

Figure 7.1 illustrates the pore size distributions obtained for the

pristine and modified activated carbons. The sample KUA shows a wide

pore size distribution, with an average pore size of 1.4 nm. Moreover,

the profile shows a contribution of micropores with size lower than 1

nm. The nitrogen doping does not significantly modify the pore size of

the samples. The small differences in the profile are a consequence of

discrepancies in the fitting performed by NL-DFT. However, the

nitrogen functionalization treatment seems to decrease the size of the

micropores in the whole range of porosity. Thus, the nitrogen doping at

mild conditions allows the generation of carbon materials with similar

porous texture but different functionalities.


307

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4

Pore

volu

me (

cm

3/g

nm

)

Pore width (nm)

KUA

KUA-N

KUA-CONH₂

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 1 2 3

Pore

volu

me (

cm

3/g

)

Pore width (nm)

KUAKUA-N

KUA-CONH₂

(b)

Figure 7.1. (a) Differential and (b) cumulative pore size distributions of KUA, KUA-N

and KUA-CONH2 by DFT calculations.


7.3.2.1 Three-electrode cell configuration

The electrochemical performance of the samples was tested in three-

electrode cells in 1M Pyr14TFSI/PC and 1M Pyr14BF4/PC. Figure 7.2

shows the CVs recorded for the carbon materials from the open circuit

potential (EOCP) upon negative (Figure 7.2a) and positive polarization

(Figure 7.2b) in 1M Pyr14BF4/PC. The pristine activated carbon (KUA)

Chapter 7

308

displays faradaic processes under conditions of negative polarization. A

cathodic current is recorded starting at -1.70 V, with the corresponding

reverse anodic peak at -0.30 V. These faradaic charge transfer reactions

can be associated to oxygen functionalities and decomposition of PC

[14,22–24]. At positive polarization, a redox processes are observed at

1.0 V, and an anodic current starts at 1.2 V, evidencing the occurrence of

oxidation processes.

The CVs clearly evidence than nitrogen doping affects the

electrochemical performance of the activated carbons. First, the N-doped

carbons (KUA-N and KUA-CONH2) display lower EOCP than the parent

non-doped carbon, indicating that the generation of nitrogen

functionalities modifies the carbon-ion interaction during non-charging

regime [25]. Moreover, the nitrogen doping affects the electrochemical

stability of the carbon materials. At negative polarization from the open

circuit potential, both samples present a faradaic process at -1.70 V.

However, the current values reached during the reverse scan are much

lower than those of the non-doped carbon, highlighting the lower

ocurrence of degradation processes in the N-doped carbons.

Furthermore, the N-doped carbons show a CV with a square shape

among the whole potential range, evidencing an ideal EDLC behavior

even at potentials close to the limit value of stability.

Similar features are observed at the positive potential range (Figure

7.2a). The lower EOCP detected for the N-doped carbons provides a wider

potential window upon positive polarization. Thus, the symmetric

configuration in supercapacitor cells using these materials might provide


309

higher durability, since a better balance of the charge is expected during

the operation of the device. Moreover, at high positive potentials (> 1.2

V), the faradaic current detected for the N-doped carbon materials is

considerably lower than the one exhibited by KUA, demostrating the

improvement of electrochemical stability under oxidative conditions due

to the formation of nitrogen groups.

-400

-200

0

200

400

600

-0.2 0.2 0.6 1 1.4 1.8

Ca

pa

cit

an

ce (

F/g

)

E vs Ag (V)

(b)

-400

-200

0

200

400

600

-2 -1.4 -0.8 -0.2 0.4

Ca

pa

cit

an

ce (

F/g

)

E vs Ag (V)

(a)

Figure 7.2. Cyclic voltammograms obtained for KUA (black), KUA-N (blue) and

KUA-CONH2 (red) electrodes in (a) negative and (b) positive potential ranges. 1M

Pyr14BF4/PC. v = 1 mV/s.

-400

-200

0

200

400

600

-0.2 0.3 0.8 1.3 1.8

Ca

pa

cit

an

ce (

F/g

)

E vs Ag (V)

(b)

-400

-200

0

200

400

600

-2.5 -1.9 -1.3 -0.7 -0.1

Ca

pa

cit

an

ce (

F/g

)

E vs Ag (V)

(a)

Figure 7.3. Cyclic voltammograms obtained for KUA (black), KUA-N (blue) and

KUA-CONH2 (red) electrodes in (a) negative and (b) positive potential ranges. 1M

Pyr14TFSI/PC. v = 1 mV/s.

Chapter 7

310

The potential window of the carbon materials was also analyzed by

CV in 1M Pyr14TFSI/PC. As detected in 1M Pyr14BF4/PC, the carbon

electrodes exhibit a cathodic current related to decomposition processes.

However, in this case the faradaic current is detected at more negative

potentials (- 2.0 V), evidencing a larger electrochemical stability

window in case of the IL mixture. Under conditions of positive

polarization, KUA exhibits larger faradaic currents when increasing the

potential, evindencing lower electrochemical stability also in this

electrolyte. Furthermore, KUA-CONH2 electrode displays an increase of

faradaic cathodic current at 1.5 V, that is not observed in case of KUA-N

electrode, evidencing lower electrochemical stability for the former than

for the latter.

Even though surface chemistry influences the electrochemical

stability of the carbon materials in the electrolytes, all of them display

larger electrochemical stability window in IL-based electrolyte than in

the other electrolyte. This result is in agreement with the performance of

other carbons in the literature [11,13,26,27].

The gravimetric capacitance was determined using a potential

window in absence of faradaic contributions. Table 7.3 collects the

values obtained under positive (C+) and negative (C-) polarization in

both electrolytes. However, it should be remarked that C+ and C- cannot

be exclusively associated to adsorption of anions and cations,

respectively. Several studies have demonstrated that the charge storage

in supercapacitors is not produced by single electrosorption of ions on

the surface of the charged electrode. The formation of the EDL can be


311

produced by different mechanisms depending on the electrolyte and the

electrode, as well as the electrode polarization [28,29]. Indepently of

mechanism governed in the formation of the EDL, the relative pore/ion

size is a key parameter to be considered [13,14,30,31], since the ions

might diffuse across the pore network involved in the charging process.

The pristine activated carbon displays extraordinary capacitance

values in both electrolytes because its porosity has been tailored to be

used in non-conventional electrolytes [14]. Even though 1M

Pyr14TFSI/PC shows higher viscosity than 1M Pyr14BF4/PC (Table 7.4),

both electrolytes provide large capacitance values when using KUA as

electrode material. The main differences in terms of capacitance are a

consequence of the different sizes of the ions: 0.46 nm (BF4-), 0.79 nm

(TFSI-) and 1.1 nm (Pyr14+) [11,32]. On the other hand, KUA has a large

micropore volume and an average pore size of 1.4 nm. Hence, KUA

provides an adequate pore size distribution to allow the diffusion of the

electrolyte-containing ions to its inner surface.

The N-doped activated carbons show slightly lower capacitance

values in both electrolytes. This is related to the small loss of

microporosity produced during the functionalization treatment. KUA-N

evidenced an almost insignificant modification of porous texture.

However, the ions of the electrolytes have large sizes, and consequently

small changes on the pore width might affect the formation of the EDL.

Thus, the differences in capacitance might arise from the small

modification of the pore sizes in these samples, as well as the slightly

lower available surface area. More interesting features are observed for

Chapter 7

312

KUA-CONH2 activated carbon. This sample displays similar

capacitance values in 1M Pyr14TFSI/PC than KUA-N, but slightly lower

values in 1M Pyr14BF4/PC. Since KUA-CONH2 shows lower micropore

volume than KUA-N, the formation of EDL in the IL-based electrolyte

might not involve the whole microporosity of both carbons, but only a

range of micropores that contain both carbon materials. Since BF4- anion

has lower size than TFSI-, it can access smaller micropores. Hence, the

significantly higher capacitance values detected in 1M Pyr14BF4/PC for

KUA-N, mainly in the positive potential window, can be a consequence

of the significant increase of micropores lower than 1nm observed after

functionalization in this sample.

Table 7.3. Gravimetric capacitance calculated for the carbon electrodes in the

electrolytes in different potential ranges.

Sample 1 M Pyr14BF4/PC 1M Pyr14TFSI/PC

C+ (F/g) C-(F/g) C+ (F/g) C-(F/g)

KUA 187 185 168 193

KUA-N 165 161 153 163

KUA-CONH2 147 153 157 165

Table 7.4. Viscosity and conductivity of the electrolytes at 20 ºC [11].

Electrolyte Viscosity

(mPa s)

Conductivity

(mS/cm)

1M Pyr14BF4/PC 4.2 9.8

1M Pyr14TFSI/PC 5.1 7.3


313

7.3.2.2 EDLC investigation

Asymmetric capacitors (in mass) were constructed to evaluate the

performance of the carbon electrodes at high operation voltage (3 V) in

1M Pyr14TFSI/PC. The mass loading of the carbon materials was

balanced by following the methodology proposed by Peng et al. [33]. By

following this strategy, the performance of the EDLC can be enhanced

and the premature ageing of cells should be avoided. For this, the

capacitance values were determined in the electrochemical stability

windows determined in section 7.3.2.1. The mass ratio was calculated by

using the equation (3):

(7.3)

Where w+/w- is the mass ratio of the electrodes, C+ and C- are the

capacitance values under positive and negative polarization, and ΔV+

and ΔV- are the positive and negative windows, respectively.

Figure 7.4a shows the CV obtained for the three activated carbon-

based capacitors at 20 mV/s. The capacitors display an ideal rectangular

shape, characteristic of electrical double layer, and provide large

capacitance values (37-40 F/g). The lowest value was determined for

KUA-CONH2 as consequence of its lower porosity. Figure 7.4b

illustrates the evolution of capacitance when increasing the scan rate.

The capacitors provide high retention of capacitance at large scan rates,

demostrating values of ⁓ 20 F/g at 500 mV/s. The capacitors based on

KUA-N and KUA-CONH2 supply slighly higher retention of

capacitance, as happened in aqueous (chapters 3 and 4) [16,21] and

Chapter 7

314

organic electrolytes (chapter 6). The improvement of capacitance

retention in all media might be undoubtely related to an enhancement of

electrical conductivity due to the substitution of electron-withdrawing

oxygen functional groups with electron-donor nitrogen functionalities

[34].

-60

-30

0

30

60

0 0.5 1 1.5 2 2.5 3

Cap

acit

an

ce (

F/g

)

Voltage (V)

(a)

0

10

20

30

40

50

0 200 400 600

Cap

acit

an

ce (

F/g

)

Scan rate (mV/s)

(b)

Figure 7.4. (a) Cyclic voltammograms obtained for KUA (black), KUA-N (blue) and

KUA-CONH2 (red) based capacitors. v = 20 mV/s. (b) Evolution of capacitance with

the scan rate (20, 50, 100, 200 and 500 mV/s). 1M Pyr14TFSI/PC.

The performance of the capacitors was analyzed using a cycling test at

different current densities. Figure 7.5 illustrates the GCD cycles and the


315

Ragone plot obtained for the capacitors. Table 7.5 collects the

parameters of the devices obtained from the cycling test. The

gravimetric capacitance values are among the largest reported in the

literature in IL-based electrolytes [35–39]. Furthermore, the devices

show outstanding performance in terms of volumetric capacitance since

the carbon electrodes contain almost negligible mesoporosity, which is

commonly used to enhance the performance in ILs [35,36,38,39] (not

reported in these references). Hence, the combination of highly

microporous activated carbons with IL-based electrolytes allows the

production of devices with excellent energy densities in gravimetric and

volumetric basis (44-48 Wh/kg and 16-17 Wh/L, respectively, Table

7.5).

Table 7.5. Gravimetric capacitance (Cg), energy density (E), coulombic efficiency and

energy efficiency determined for KUA, KUA-CONH2 and KUA-N asymmetric

supercapacitors at the voltage 3V by GCD cycles. 1M Pyr14TFSI/PC. j=1 A/g.

Sample Cg

(F/

g)

Cv

(F/cm3)

Eg

(Wh/Kg)

Ev

(Wh/L)

Coulombic

efficiency

(%)

Energy

efficiency

(%)

KUA 40 14 46 16 99 84

KUA-N 40 14 48 17 100 89

KUA-

CONH2

37 14 44 16 100 87

Moreover, the N-doped carbon-based capacitors evidence larger

coulombic and energy efficiency than the pristine KUA-based capacitor,

providing better reversibility of the N-doped samples upon cycling. This

must be related to the lower occurrence of faradaic processes, as

discussed in section 7.3.2.1. Thus, the generation of stable nitrogen

Chapter 7

316

functionalities, as well as the decrease of detrimental oxygen groups,

enhance the electrochemical performance of carbon electrodes, as

happened in organic and aqueous electrolytes.

0

0.5

1

1.5

2

2.5

3

3.5

0 50 100 150 200 250

Volt

age (

V)

Time (s)

KUA

KUA-N

KUA-CONH₂

(a)

1

10

100

1 10

E (

Wh

/Kg)

P (kW/Kg)

KUA

KUA-N

KUA-CONH₂

(b)

Figure 7.5. (a) Galvanostatic charge-discharge cycles for KUA, KUA-N and KUA-

CONH2 asymmetric supercapacitors at 3V and 1A/g. (b) Ragone plot at 3V for all

asymmetric supercapacitors. j = 1-10 A/g. 1M Pyr14TFSI/PC.

Figure 7.5 illustrates the Ragone plot obtained for all activated

carbon-based capacitors in 1M Pyr14TFSI/PC. At low power density (1.5

kW/kg), the cells provide similar energy density values (44-48 Wh/kg,

Table 7.5). However, the N-doped carbon-based capacitors can keep


317

higher values when increasing the power density up to 9 kW/kg

(14Wh/kg for KUA, and 15-18 Wh/kg for the N-doped activated carbon-

based capacitors). This is related to the improvement of conductivity

produced by the nitrogen functional groups [15,16,21,34].

Preliminary investigation about the durability of the capacitors have

been made carrying out GCD cycles at 3 V. Figure 7.6 shows the

evolution of capacitance along cycles during the cyclability test for

capacitors built with KUA and KUA-N since both have similar

porosities but different electrochemical properties. The non-doped

carbon-based capacitor experiences a remarkable loss of capacitance

during the first thousand cycles, and afterwards it retains around 60 % of

the initial capacitance. Interestingly, the capacitor based on KUA-N

carbon exhibits a stable performance along cycles, providing a

capacitance retention of 90 % after the durability test. This improvement

might be undoubtedly related to the lower ocurrence of faradaic

reactions during the charge-discharge process, that lead to the

degradation of the electrodes and electrolytes [22,24]. Since the

materials employed in the construction of the cells have very similar

porous texture, the differences on the performance might be related to

the modifications on the surface chemistry produced by the nitrogen

doping treatment. The reduction of detrimental oxygen functional groups

[22] as well as the generation of nitrogen groups with higher

electrochemical stability [16,17] are responsible of the improvement of

the performance of N-doped activated carbon-based supercapacitors.

Chapter 7

318

0

0.2

0.4

0.6

0.8

1

1.2

0 500 1000 1500 2000

C/C

0

Nº Cycles

KUA KUA-N

Figure 7.6. Cyclability test for KUA, KUA-N and KUA-CONH2 based supercapacitors

at 3V. j = 1A/g. 2000 cycles. 1M Pyr14TFSI/PC.

7.4. Conclusions

In this chapter, the effect of nitrogen functionalization at mild

conditions on the electrochemical performance of activated carbons in

non-conventional electrolytes was studied. The generation of nitrogen

functionalities at low temperature allows the preservation of most of the

tailored porous texture of the pristine carbon material, which is crucial

for the development of EDLC based in innovative electrolytes.

The electrochemical characterization of the samples in two different

electrolytes, with the same cation but different anion, allows to elucidate

the different parameters affecting the electrode-electrolyte interface. It

was found that surface chemistry does not significantly modify the

capacitance, but has a strong effect on the degradation of both electrodes

and electrolytes. The nitrogen doping decreases the degradation

processes occurring in the carbon electrodes under positive and negative


319

polarization, when approaching the limits of the electrochemical stability

window.

Supercapacitors based on these activated carbons in IL-based

electrolyte displayed very high capacitance and energy densities, with

improved rate performance as consequence of the generation of nitrogen

functional groups. The effect of nitrogen doping on the durability of the

devices was analysed by a cycling test, evidencing an enhancement of

the performance of N-doped activated carbons due to the beneficial

effect of nitrogen doping at mild conditions.

7.5 References



[2] W. Raza, F. Ali, N. Raza, Y. Luo, K.H. Kim, J. Yang, S. Kumar,

A. Mehmood, E.E. Kwon, Recent advancements in supercapacitor

technology, Nano Energy 52 (2018) 441–473..

[3] P. Simon, Y. Gogotsi, Materials for electrochemical capacitors.,

Nat. Mater. 7 (2008) 845–54.

[4] C. Zhong, Y. Deng, W. Hu, J. Qiao, L. Zhang, J. Zhang, A review

of electrolyte materials and compositions for electrochemical

supercapacitors, Chem. Soc. Rev. 44 (2015) 7484–7539.


layer capacitors : A perspective article, J. Power Sources 326

(2016) 534–540.

[6] J. Krummacher, C. Schütter, L.H. Hess, A. Balducci, Non-

aqueous electrolytes for electrochemical capacitors, Curr. Opin.

Electrochem. 9 (2018) 64–69.

[7] A. Brandt, S. Pohlmann, A. Varzi, A. Balducci, S. Passerini, Ionic

liquids in supercapacitors, MRS Bull. 38 (2013) 554–559.

Chapter 7

320

[8] D.R. Macfarlane, N. Tachikawa, M. Forsyth, J.M. Pringle, P.C.

Howlett, G.D. Elliott, J.H. Davis, M. Watanabe, P. Simon, C.A.

Angell, Energy applications of ionic liquids, Energy Environ. Sci.

7 (2014) 232–250.

[9] W.Y. Tsai, R. Lin, S. Murali, L. Li Zhang, J.K. McDonough, R.S.

Ruoff, P.L. Taberna, Y. Gogotsi, P. Simon, Outstanding

performance of activated graphene based supercapacitors in ionic

liquid electrolyte from -50 to 80°C, Nano Energy 2 (2013) 403–

411.

[10] V.L. Martins, R.M. Torresi, Ionic liquids in electrochemical

energy storage, Curr. Opin. Electrochem. 9 (2018) 26–32.

[11] S. Pohlmann, R.S. Kühnel, T.A. Centeno, A. Balducci, The

Influence of Anion-Cation Combinations on the Physicochemical

Properties of Advanced Electrolytes for Supercapacitors and the

Capacitance of Activated Carbons, ChemElectroChem 1 (2014)

1301–1311.

[12] S. Pohlmann, C. Ramirez-Castro, A. Balducci, The Influence of

Conductive Salt Ion Selection on EDLC Electrolyte

Characteristics and Carbon-Electrolyte Interaction, J.

Electrochem. Soc. 162 (2015) A5020–A5030.

[13] S. Pohlmann, B. Lobato, T.A. Centeno, A. Balducci, The

influence of pore size and surface area of activated carbons on the

performance of ionic liquid based supercapacitors, Phys. Chem.

Chem. Phys. 15 (2013) 17287–17294.

[14] S. Leyva-García, D. Lozano-Castelló, E. Morallón, T. Vogl, C.

Schütter, S. Passerini, A. Balducci, D. Cazorla-Amorós,

Electrochemical performance of a superporous activated carbon in

ionic liquid-based electrolytes, J. Power Sources 336 (2016) 419–

426.

[15] M.J. Mostazo-López, R. Ruiz-Rosas, A. Castro-Muñiz, H.


Ultraporous nitrogen-doped zeolite-templated carbon for high

power density aqueous-based supercapacitors, Carbon 129 (2018)


321

510–519.















4589–4596.

[20] C. Schütter, C. Ramirez-Castro, M. Oljaca, S. Passerini, M.

Winter, A. Balducci, Activated carbon, carbon blacks and

graphene based nanoplatelets as active materials for

electrochemical double layer capacitors: A comparative study, J.

Electrochem. Soc. 162 (2015).





252–265.





[23] M. Hahn, A. Würsig, R. Gallay, P. Novák, R. Kötz, Gas evolution

in activated carbon/propylene carbonate based double-layer

capacitors, Electrochem. Commun. 7 (2005) 925–930.

Chapter 7

322

[24] P.W. Ruch, D. Cericola, A. Foelske-Schmitz, R. Kötz, A.

Wokaun, Aging of electrochemical double layer capacitors with

acetonitrile-based electrolyte at elevated voltages, Electrochim.

Acta 55 (2010) 4412–4420.

[25] C.A. Leon y Leon, L.R. Radovic, Interfacial chemistry and

electrochemistry of carbon surfaces, in: P.A. Thrower (Ed.),

Chem. Phys. Carbon, Vol. 24, Marcel Dekker, Inc New York,

1994: pp. 213–310.

[26] A. Brandt, A. Balducci, Theoretical and practical energy

limitations of organic and ionic liquid-based electrolytes for high

voltage electrochemical double layer capacitors, J. Power Sources

250 (2014) 343–351.

[27] G. Moreno-Fernández, C. Schütter, J.M. Rojo, S. Passerini, A.

Balducci, T.A. Centeno, On the interaction of carbon electrodes

and non conventional electrolytes in high-voltage electrochemical

capacitors, J. Solid State Electrochem. 22 (2018) 717–725.

[28] A.C. Forse, C. Merlet, J.M. Griffin, C.P. Grey, New perspectives

on the charging mechanisms of supercapacitors, J. Am. Chem.

Soc. 138 (2016) 5731–5744.

[29] M. Salanne, B. Rotenberg, K. Naoi, K. Kaneko, P.-L. Taberna,

C.P. Grey, B. Dunn, P. Simon, Efficient storage mechanisms for

building better supercapacitors, Nat. Energy 1 (2016) 16070.

[30] C. Largeot, C. Portet, J. Chmiola, P.-L. Taberna, Y. Gogotsi, P.

Simon, Relation between the ion size and pore size for an electric

double-layer capacitor, J. Am. Chem. Soc. 130 (2008) 2730–

2731.




aqueous electrolyte, Carbon. 41 (2003) 1765–1775.

[32] A. Balducci, R. Dugas, P.L. Taberna, P. Simon, D. Plée, M.

Mastragostino, S. Passerini, High temperature carbon – carbon

supercapacitor using ionic liquid as electrolyte, J. Power Sources


323

165 (2007) 922–927.

[33] C. Peng, S. Zhang, X. Zhou, G.Z. Chen, Unequalisation of


asymmetrical supercapacitors, Energy Environ. Sci. 3 (2010)

1499.

[34] V. Strelko, V.. Kuts, P.. Thrower, On the mechanism of possible




[35] M. Sevilla, G.A. Ferrero, N. Diez, A.B. Fuertes, One-step

synthesis of ultra-high surface area nanoporous carbons and their

application for electrochemical energy storage, Carbon 131

(2018) 193–200.

[36] A.B. Fuertes, M. Sevilla, High-surface area carbons from

renewable sources with a bimodal micro-mesoporosity for high-

performance ionic liquid-based supercapacitors, Carbon 94 (2015)

41–52.

[37] E. Redondo, W.-Y. Tsai, B. Daffos, P.-L. Taberna, P. Simon, E.

Goikolea, R. Mysyk, Outstanding room-temperature capacitance

of biomass-derived microporous carbons in ionic liquid

electrolyte, Electrochem. Commun. 79 (2017) 5–8.

[38] S.W. Xu, Y.Q. Zhao, Y.X. Xu, Q.H. Chen, G.Q. Zhang, Q.Q. Xu,

D.D. Zhao, X. Zhang, C.L. Xu, Heteroatom doped porous carbon

sheets derived from protein-rich wheat gluten for supercapacitors:

The synergistic effect of pore properties and heteroatom on the

electrochemical performance in different electrolytes, J. Power

Sources 401 (2018) 375–385.

[39] Y. Zhu, M. Chen, Y. zhang, W. Zhao, C. Wang, A biomass-

derived nitrogen-doped porous carbon for high-energy

supercapacitor, Carbon 140 (2018) 404–412.

Chapter 8

Electrochemicalperformance of

N-doped activated carbons as

electrocatalysts for the ORR

N-doped activated carbons as electrocatalysts for the ORR

327

8.1. Introduction

Fuel cells have attracted considerably attention as energy conversion

devices for stationary and mobile applications mainly due to their high

efficiency and low emisions. However, the development of new catalysts

for the oxygen reduction reaction that takes place in the cathode is still a

key challenge to expand this technology. The most common materials are

based on platinum and other noble metals supported on carbon materials

[1–3]. These catalysts provide the highest electroactivities. However, they

experience several drawbacks that hinders their large-scale application,

such as their high cost, low availability and poor durability [4]. Hence, an

extensive research is been carried out to develop new catalysts capable to

increase the use of fuel cells in industry.

One of the most promising alternatives to replace electrocatalysts

based in noble metals are metal-free carbon materials doped with different

heteroatoms (nitrogen, phosphorous, sulfur, etc.) [5,6]. These materials

have low cost, high surface area, good mechanical and electrical

properties and high stability. In particular, nitrogen-doped carbon

materials have demonstrated outstanding electrocatalytic properties [7].

The presence of nitrogen can modify the electron-donor properties of the

carbon material and provide a redistribution of electronic density,

producing an increase of the electrocatalytic activity for the ORR.

However, the role of the different nitrogen functional groups is not still

well understood [8]. Moreover, several works have pointed out the

enhancement of electrocatalytic performance when increasing the surface

area of the carbon material, either working as catalyst support or catalyst

Chapter 8

328

itself [9]. Thus, the synthesis of new carbon materials with large surface

area and selective nitrogen functionalities appears as a promising

approach for the development of new electrocatalysts for ORR.

Nitrogen-containing porous carbon materials can be synthesized by

following different methods [10]. They are summarized as follows:

(i) Carbonization of a nitrogen-containing precursor (such as

pyridine, melamine, polyaniline, etc), which can be followed

by an activation process (either physical or chemical).

(ii) Hydrothermal carbonization of nitrogen containing-

compounds (glucosamine, cyanuric acid, etc.).

(iii) Templating approaches using a nitrogen-containing precursor

followed by a thermal treatment.

(iv) Post-thermal treatments of a material previously synthesized

with a nitrogen-containing reactant in gas or liquid phase.

The first method has been extensively employed to produce N-doped

carbon materials derived from polyaniline with large nitrogen content

[11]. These carbons display outstanding electroactivity as catalyst

supports [12–14] and good response when working themselves as

electrocatalysts towards ORR [15–21]. The well-defined structure of

polyaniline leads to the formation of carbon materials with predominancy

of certain types of functionalities depending on the treatment temperature

[16]. Moreover, polyaniline has been extensively used to prepare hybrid

material composites by chemical or electrochemical polymerization of

aniline over a carbon support [11]. These composites can be converted


329

into advanced N-doped carbons by thermal post-treatments [11,22]. The

main advantage of this strategy is the generation of porous carbons with

surface functionalities characteristic of polyaniline-derived carbon

materials while remaining most part of the porous structure of the pristine

carbon support.

Another promising methodology to generate N-doped carbon

materials with high surface area is based on chemical post-treatments

through organic chemistry reactions [23,24]. In this procedure, nitrogen

doping is carried out at mild conditions and only depends on the reactivity

between the carbon surface (i.e. oxygen functional groups) and the

nitrogen-containing reaction medium. This approach leads to the

anchoring of a wide range of nitrogen functionalities. Due to the mild

conditions of the treatment, nitrogen groups with low thermal stability (i.e.

amides, amines) are generated on these carbon materials. Thus, the surface

chemistry of these carbons can also be easily tuned by selective post-

thermal treatments under inert atmosphere.

In this chapter, we report the performance of N-doped superporous

activated carbons as electrocatalysts for the ORR. An activated carbon

with very well developed microporosity (SBET > 3000 m2/g) was used as

starting material for further functionalization. The samples are obtained

by different N-doping pathways: (i) chemical polymerization of aniline;

(ii) nitrogen functionalization by organic reactions; (iii) thermal post-

treatments of the samples obtained with the pathways (i) and (ii). The

combination of these strategies allows to preserve most part of the porous

texture of the pristine carbon material while the surface chemistry is

Chapter 8

330

severely modified, as described in chapter 6 and ref. [22]. Thus, these

activated carbons can be used to assess the role of different nitrogen

functionalities in the electroactivity for the ORR.


8.2.1. Synthesis of activated carbons

8.2.1.1 Pristine activated carbon

A superporous activated carbon synthesized in our laboratory has

been used as the starting material for nitrogen functionalization by

different methods. The pristine material (named as KUA), has been

obtained by chemical activation of a Spanish anthracite with KOH using

an impregnation ratio of activating agent to raw material of 4:1 and an

activation temperature of 750º C under inert atmosphere, which was held

for 1 hour. More details about the preparation process are available

elsewhere [25].

8.2.1.2. Chemical functionalization of activated carbon at mild conditions


two different strategies based on the organic chemistry protocols. This

methodology was accurately described in chapters 3 and 4. Briefly, the

first approach consisted in a three-step protocol: (i) chemical oxidation

with HNO3, (ii) treatment with SOCl2 and (iii) amidation reaction with

NH4NO3/DMF and pyridine. The obtained sample was named as KUA-

CONH2. In the second method, the third step of the first protocol is


331

directly applied over pristine sample (KUA). This sample was named as

KUA-N.

8.2.1.3. Preparation of polyaniline/activated carbon composite

PANI/KUA composite was prepared by chemical polymerisation of

aniline as described elsewhere [26]. The aniline (concentration: 70mM)

was adsorbed on the carbon material during 24 h at 30 ºC. Afterwards, the

sample was treated with ammonium persulfate solution in 1M HCl for one

hour in a reactor at 0 ºC. The aniline monomer: ammonium persulfate

molar ratio was 1:1. The composite was washed with 1M HCl, 1M

NH4OH and distilled water, and finally dried under vacuum for 24 h.

8.2.1.4. Post-thermal treatments

The carbon materials (KUA, KUA-CONH2 and KUA-N samples) and

PANI/KUA composite were heat treated at different temperatures under

N2 atmosphere using (200mL/min, 1h) and a heating rate of 5 Cº/min. The

obtained samples are named as S_T, where S es is the name of the pristine

sample (KUA, KUA-N, KUA-CONH2 and KUA/PANI) and T is the final

heating temperature (600 and 800 ºC).


The porous texture characterization was carried out by N2 adsorption-

desorption isotherms at -196º C and by CO2 adsorption at 0º C by using

an Autosorb-6-Quantachrome apparatus. The samples were outgassed at

200º C for 4 hours before the experiments. The apparent surface area was

obtained from N2 adsorption isotherms by using the BET equation in the

Chapter 8

332

0.05-0.20 range of relative pressures. The total micropore volume was

determined by Dubinin-Radushkevich (DR) method applied to N2

(relative pressures from 0.01 to 0.05) adsorption isotherms. The volume

of the narrow microporosity (i.e., pore sizes below 0.7 nm) was calculated

from the DR method applied to the CO2 adsorption isotherms (relative

pressures from 0.0001 to 0.25) [27].




Microtech Multilab 3000 spectrometer, equipped with an Al anode. TPD

experiments were performed by heating the samples (10 mg) to 950º C

(at a heating rate of 20º C/min) under a helium flow rate of 100 mL/min.

The analyses were carried out by using a TGA-DSC instrument (TA

Instruments, SDT Q600 Simultaneous) coupled to a mass spectrometer

(Thermostar, Balzers, BSC 200).

8.2.3. Electrochemical activity towards ORR

The evaluation of electroactivity towards ORR was carried out in a

three-electrode cell using an Autolab PGSTAT302 bipotentiostat

(Metrohm, Netherlands). A rotating ring-disk electrode (RRDE, Pine

Research Instruments, USA) equipped with a glassy carbon disk (5.61 mm

diameter) and a platinum ring was used as the working electrode. A

platinum wire was used as the counter electrode and RHE electrode as

reference electrode. The carbon samples were deposited onto the glassy

carbon disk. The amount of catalyst was optimized to reach the highest


333

limiting current. The optimum value was determined to be 240 µg. The

catalysts were deposited by droping 120 µL of a 2 mg/mL dispersion made

of each sample in ethanol (1.5 % Nafion® perfluorinated resin solution,

Sigma Aldrich), obtaining a catalyst charge of 0.96 mg/cm2.

The electrochemical tests were carried out by Cyclic voltammetry

(CV) and linear sweep voltammetry (LSV) in the potential range 0.0 - 1.0

V vs. RHE. CVs were performed in a N2-saturated and O2-saturated

atmosphere at 5 mV/s. LSV experiments were conducted in an O2-

saturated atmosphere at different rotation rates (between 400 and 2025

rpm) at 5 mV/s on the GC rotating disk electrode. The temperature was

kept at 25 ºC during the experiment. The potential of the Pt ring electrode

was held at 1.5 V vs. RHE during LSV experiments. The electron transfer

number was determined from the RRDE measurements:

𝑛 = 4 𝐼𝑑

𝐼𝑑+ 𝐼𝑟 𝑁⁄ (8.1)

Where Ir and Id are the intensities measured at the ring and the disk,

respectively, and N is the collection efficiency of the ring (0.37).

8.3. Results and Discussion

8.3.1. Surface chemistry characterization

The effect of the different nitrogen functionalization treatments was

analysed by XPS and TPD. Table 8.1 summarizes the surface composition

of the carbon materials. The effect of nitrogen doping at mild conditions

(for the generation of KUA-N and KUA-CONH2) was already discussed

in chapters 3 and 4, and refs. [23] and [28]. For comparison purposes, the

Chapter 8

334

data related to these samples (and their derived heat-treated samples,

KUA-CONH2_800 and KUA-N_800) are included in Table 8.1. Briefly,

the post-functionalization treatment produces the decrease of oxygen

functionalities of the pristine carbon material to attach different nitrogen

moieties. Thus, KUA-N sample has lower oxygen content than the pristine

carbon material and it contains nitrogen functionalities derived from CO-

evolving groups, such as amines and nitrogen heterocycles (chapter 4)

[28]. The synthesis of KUA-CONH2 involves a pre-oxidation treatment to

increase the amount of oxygen functional groups (CO2 and CO-evolving

groups), that afterwards react to produce nitrogen functionalities.

Consequently, this sample has contributions of nitrogen groups derived

from CO2 evolving groups (amide-like functionalities) and from CO

evolving groups (amines, imines, nitrogen heterocycles) (chapter 3) [23].

Moreover, KUA-CONH2 has higher oxygen content than the pristine

carbon material.

Table 8.1. Surface composition of the activated carbons obtained by XPS and TPD.

Sample CO2

(µmol/g)

CO

(µmol/g)

O

(µmol/g)

O

(at. %)

N

(at. %)

KUA 450 1970 2870 8.8 0.3

KUA_800 170 620 960 2.3 -

KUA/PANI 1320 3560 6200 11.4 6.0

KUA/PANI_600 380 1860 2620 4.5 4.7

KUA/PANI_800 210 740 1050 6.7 2.0

KUA-CONH2 1140 2370 4650 10.2 4.2

KUA-CONH2_800 140 570 830 4.5 2.1

KUA-N 450 1750 2640 7.5 3.7

KUA-N_800 130 505 720 3.0 1.7


335

The polymerization of aniline over KUA produces the generation of

a polyaniline film inside the microporosity of the carbon material [26, 29].

The attachment of nitrogen was confirmed by XPS (6.0 at. % N, Table

8.1). KUA/PANI composite also evidences an increase of oxygen content,

as confirmed by XPS and TPD. Figure 8.1 illustrates the TPD profiles

obtained for KUA, KUA/PANI composite and the derived heat treated-

samples (KUA/PANI_600 and KUA/PANI_800). KUA/PANI composite

displays larger amount of CO2 (carboxylic acids, lactones and anhydrides)

and CO evolving groups (phenols, quinones, etc) [30]. This is related to

the use of an oxidizing agent on the polymerization of aniline in presence

of the carbon material that may react with the carbon surface to produce

oxygen functional groups.

The effect of heat treatments over the pristine and N-doped

functionalized activated carbons was previously discussed in chapter 6.

The heat treatment diminishes most of the oxygen and nitrongen

functional groups in KUA-CONH2 and KUA-N materials. Similarly, the

treatments over the composite strongly decrease the amount of oxygen

functional groups. At 600 ºC, ⁓71 % of CO2 evolving groups are removed

from KUA/PANI composite. The amount of CO evolving groups is

severely decreased. These modifications lead to the generation of a N-

doped activated carbon with similar oxygen content than the pristine

sample (2620 and 2870 µmol/g for KUA/PANI_800 and KUA,

respectively, Table 8.1). Nevertheless, the type of functionalities is

different, being remarkably lower the concentration of groups that

thermally desorb at temperatures lower than around 600 ºC (i.e., phenols)

Chapter 8

336

[30]. It is also worth noting that the concentration of carbonyl type groups

(above 600 ºC) is larger in KUA/PANI_600 than in the composite,

indicating that the heat treatment produces the conversion of phenols into

carbonyl groups upon heating at 800 ºC.

Figure 8.1. Comparison between (a) CO2 and (b) CO TPD profiles of KUA, KUA/PANI,

KUA/PANI_600 and KUA/PANI_800.

The heat treatments at 800 ºC diminishes the oxygen content in all

samples. The largest oxygen content was detected for KUA/PANI_800

(1050 µmol/g, Table 8.1). This can be explained by the presence of higher

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 200 400 600 800 1000

CO

2(µ

mo

l/g

s)

Temperature (ºC)

KUA

KUA/PANI

KUA/PANI_600

KUA/PANI_800

(a)

0

0.6

1.2

1.8

2.4

3

3.6

0 200 400 600 800 1000

CO

(μ

mo

l/g

s)

Temperature (ºC)

KUA

KUA/PANI

KUA/PANI_600

KUA/PANI_800

(b)


337

concentration of moieties in the parent composite (KUA/PANI sample),

that may lead to higher reactivity between the carbon surface and the

PANI coating upon heating.

The nitrogen functionalities are also modified as consequence of the

heat treatments. Figure 8.3 shows the devonvoluted N1s XPS spectra

obtained for KUA/PANI composite and the related heat-treated samples.

The assignement of the different nitrogen functionalities is collected in

Table 8.3. The values obtained for the other N-doped samples are included

for comparison purposes. KUA/PANI composite exhibits a main peak

associated to the presence of amines at 399.7 ± 0.2 eV [31–34]. Moreover,

there is also a contribution of pyrroles/pyridones (400.8 ± 0.2 eV) that can

be explained by reactions with oxygen functionalities during the

polymerization of aniline [29]. The carbonization at 600 ºC produces the

decrease of nitrogen groups as well as their conversion into different

surface functionalities. Thus, KUA/PANI after treatment at 600ºC shows

two different peaks related to pyrroles/pyridones (400.6 ± 0.2 eV) and

pyridines/imines (398.6 ± 0.2 eV) [31–34]. After heat treatment at 800 ºC,

the main contribution of N-functional groups is detected at 400.8 ± 0.2

eV, related to pyrroles/pyridones (70 % N, Table 8.3), and a lower

presence of pyridines/imines (398.3 ± 0.2 eV, 30 % N, Table 8.3). These

results are in agreement with the mechanism proposed for carbonization

of PANI [35,36]. During the heat treatment the polymer chain experiences

bond breaking that, upon heating at higher temperatures, react again to

form N-containing heterocycles (pyrroles/pyridones and pyridines)

through cross-linking reactions. Thus, the amines present on KUA/PANI

Chapter 8

338

composite may experience degradation upon heating and react to generate

nitrogen groups with higher thermal stability, such as nitrogen

heterocycles.

Figure 8.2. N1s XPS spectra deconvoluted for the KUA/PANI composite and the N-

doped activated carbons.

As discussed in chapter 6, KUA-N and KUA-CONH2 also experience

loss and conversion of their nitrogen functionalities into groups with

higher thermal stability [37]. At 800 ºC, there are two main contributions

of pyridine (398.7 ± 0.2 eV, Table 8.2) and pyrrole/pyridone (400.7 ± 0.2

eV, Table 8.2). However, the concentration of pyrrole/pyridone groups is

lower than that found for KUA/PANI_800. Hence, the use of different

396398400402404

I (a

.u.)

Binding Energy (eV)

KUA/PANI_600

KUA/PANI_800

KUA/PANI


339

surface modification methods and post-thermal treatments allows to

obtain activated carbons with very different surface chemistry.


Sample Binding

Energy

(eV)

Functional

Group

N

(at. %)

Percentage

of N species

KUA/PANI 400.8 ± 0.2 Pyrrole,

pyridone

2.2 36

399.7 ± 0.2 Amines 3.8 64

KUA/PANI_600 400.6 ± 0.2 Pyrrole,

pyridone

2.9

61

398.6 ± 0.2 Pyridine, Imine 1.8 39

KUA/PANI_800

400.8 ± 0.2 Pyrrole,

pyridone

1.4 70

398.3 ± 0.2 Pyridine, imine 0.5 30

KUA-N 401.9 ± 0.2 Quaternary 0.4 10

400.7 ± 0.2 Pyrrole,

pyridone

0.9 25


Amine, Imide

1.3 35

398.7 ± 0.2 Pyridine, Imine 1.1 30

KUA-N_800 402.7 ± 0.2 Oxidized N 0.2 14

400.8 ± 0.2 Pyrrole,

pyridone

0.9 51

398.7 ± 0.2 Pyridine, imine 0.6 35


pyridone

0.7 19


amine, imide

1.9 50

398.8 ± 0.2 Pyridine, imine 1.2 31

KUA-CONH2_800 402.5 ± 0.2 Oxidized N 0.2 11

400.8 ± 0.2 Pyrrole,

pyridone

0.9 52

398.7 ± 0.2 Pyridine, imine 0.6 37

Chapter 8

340

8.3.2. Porous texture characterization

The porous texture of the samples was analysed by N2 and CO2

adsorption isotherms. Figure 8.3 illustrates the N2 adsorption isotherms

obtained for the pristine activated carbon and the PANI-derived carbon

materials. For comparison purposes, the isotherm of KUA-CONH2 was

also included. All samples evidence a type I isotherm characteristic of

microporous materials.

Figure 8.3. N2 adsorption isotherms obtained for KUA (pristine sample), KUA/PANI

composite and N-doped activated carbons.

Table 8.3 compiles data of the porous texture of the samples. The

effects of nitrogen doping at mild conditions were thoroughly discussed

in chapters 3, 4 and 6. The functionalization at mild conditions produces

some minor modifications on the porous texture of the samples. When

using a single step modification protocol (for the production of KUA-N),

the microporosity of the pristine carbon materials is fully retained [24].

0

200

400

600

800

1000

0 0.2 0.4 0.6 0.8 1

Ad

sorb

ed

vo

lum

e (

cm

3/g

)

P/Po

KUA

KUA/PANI

KUA/PANI_600

KUA/PANI_800

KUA-CONH₂


341

However, the combination of oxidation process and amidation reactions

leads to some loss of the apparent surface area, due to the generation of

surface functionalities that occupies or block some part of the

microporosity (see chapter 3 and ref. [23]).

Regarding KUA/PANI composite, the formation of polyaniline layer

over the pristine carbon material KUA increases the amount of of nitrogen

and oxygen functionalities (see section 8.3.1) but produces the most

important decrease of the apparent surface area and micropore volume

(Table 8.3). After the heat treatments, the obtained activated carbons

(KUA/PANI_600 and KUA/PANI_800) recover part of the porous

texture, due to the decomposition of PANI layer. Hence, all heat-treated

samples display similar apparent surface area (2400-2800 m2/g, Table

8.3). However, PANI-derived samples have around ⁓25 % lower

micropore volume than N-doped heat-treated carbons (KUA-N and KUA-

CONH2). This is a consequence of the carbon layer formed in the heat-

treated substrate from PANI decomposition.

Chapter 8

342

Table 8.3. Porous texture of the samples.

Sample SBET

(m2/g)

VDRN2

(cm3/g)

VDRCO2

(cm3/g)

KUA 3080 1.19 0.57

KUA_800 2680 1.05 0.49

KUA/PANI 1590 0.54 0.37

KUA/PANI_600 2420 0.81 0.49

KUA/PANI_800 2470 0.84 0.56

KUA-COOH 2770 1.06 0.49

KUA-CONH2 2390 0.97 0.45

KUA-CONH2_500 2630 1.02 0.41

KUA-CONH2_800 2630 1.0 0.43

KUA-N 2960 1.18 0.52

KUA-N_500 2800 1.11 0.49

KUA-N_800 2770 1.09 0.48

8.3.3. Electroactivity towards ORR

The electroactivity of the carbons towards the ORR was analysed in

alkaline electrolyte (0.1 M KOH) by using a RRDE. LSV curves were

measured at different rotating rates for all samples. The contribution of

electrical double layer was substracted by recording CV in N2-saturated

0.1 M KOH. Figure 8.4 shows the LSV curves obtained for some samples

at 1600 rpm. The LSV for commercial 20% Pt/Vulcan was included for

comparison purposes. Table 8.4 collects the main properties related to the

performance of the samples as electrocatalysts for the ORR derived from

the LSV curves.


343

8.3.3.1. Pristine activated carbon

The pristine carbon material evidences a two-wave curve

characteristic of microporous carbon materials [9]. The high onset

potential (0.81 V, Table 8.4) demonstrated by this electrocatalyst can be

explained by the high concentration of edge sites provided by its well-

developed microporosity [9], which have been proposed as

electrocatalytic sites [38–43].

8.3.3.2. N-doped activated carbons at mild conditions

The effect of nitrogen functionalities was thoroughly analysed. The

catalysts obtained by chemical methods (KUA-N and KUA-CONH2,

Table 8.1) have not improved electrocatalytic properties in comparison

with the pristine carbon material (KUA). The onset potential was slightly

decreased in both samples. Moreover, the limiting current was severely

diminished for KUA-CONH2 sample (Figure A1). This is mainly related

to the surface chemistry of the carbon material. This sample has large

amount of nitrogen species in form of different functionalities (see section

8.3.1). It has been previously pointed out in the literature that only certain

nitrogen functional groups work as active sites for the ORR [44]. This

sample has N-heterocycles (pyridines, pyrroles/pyridones), which have

been proposed as electrocatalytic sites towards ORR [15,19–21,45,46].

Thus, an improvement of the electrocatalytic response would be expected

on this sample. The decrease on the performance cannot be associated

with a decrease of conductivity since we have previously shown that it is

enhanced in different electrolytes (see chapters 3, 6 and 7). Hence, the

Chapter 8

344

poor electrocatalytic behaviour might be a consequence of the surface

modification protocol used to functionalize these samples. For instance,

the oxidation process involves the generation of oxygen functionalities,

which are attached to the carbon edge sites. Thus, if the functionalization

method does not produce exclusively nitrogen groups with higher

electroactivity (i.e., pyridones, edge type N-Q), the N-doped sample might

result as a worse electrocatalyst, since a noticeable amount of edge sites

is lost during the surface modification reaction. Hence, the sample

experiences a deactivation during the functionalization treatment.

A similar effect is expected for KUA-N sample, even though the

oxidation process is avoided. It is expected that nitrogen functional groups

are anchored to the carbon surface by consumption of oxygen

functionalities. However, the mechanism might affect or involve the

reactivity of the adjacent carbons. This effect was also proposed by Tuci

et al [47]. The electrografting of pyridine functional groups to N-doped

carbon nanotubes resulted in a decrease of inherent electrochemical

properties and “switched them off” towards ORR. Thus, functionalization

at mild conditions might produce also a deactivation of the carbons as

electrocatalysts.

8.3.3.3. Polyaniline/activated carbon composite

The electroactivity towards ORR of KUA/PANI composite was also

evaluated. Table 8.4 compiles the onset potential recorded for this sample.

This sample displays lower onset potential and limiting current than

pristine activated carbon (see Figure A1). This poor performance is


345

explained by its lower microporosity (Table 8.3, section 8.3.2) as well as

its different surface chemistry (section 8.3.1). This sample has a large

concentration of amine functionalities, which are not electroactive

towards ORR [46]. Moreover, the formation of the polyaniline film over

the microporosity of the carbon surface might also decrease the

concentration of accessible edge sites, resulting in an overall decrease of

the catalytic electroactivity of the composite.

Figure 8.4. (a) LSV curves for the catalysts in O2-saturated 0.1 M KOH at 1600 rpm. (b)

LSV curves at different rotating rates for KUA/PANI_800. v = 5 mV/s.

-6

-5

-4

-3

-2

-1

0

0 0.2 0.4 0.6 0.8 1

j (m

A/c

m2)

E (V vs RHE)

Pt/Vulcan

KUA

KUA-N_800

KUA_800

KUA-CONH₂_800KUA/PANI_800

(a)

-6

-5

-4

-3

-2

-1

0

0 0.2 0.4 0.6 0.8 1

j (m

A c

m2)

E (V vs RHE)

400 rpm

2025 rpm

(b)

Chapter 8

346

8.3.2.4. Heat-treated activated carbons

The electrocatalytic performance of the carbon materials is improved

for the N-doped samples after heat treatment at 800 ºC (KUA-N_800,

KUA-CONH2_800 and KUA/PANI_800). All the samples display higher

onset potential, being the largest value for KUA/PANI_800 (0.88 V vs

RHE). Moreover, these carbons provide higher limiting currents than the

other samples, including the pristine KUA sample the heat treated

KUA_800 (Figure 8.4).

Table 8.4. Electrochemical parameters of the electrocatalysts calculated from the RRDE

experiments in O2-saturated 0.1M KOH at 5 mV/s and 1600 rpm.

Sample Eonset

(V vs RHE)

n

(at 0.5 V)

KUA 0.82 2.5

KUA_800 0.82 2.7

KUA/PANI 0.80 2.4

KUA/PANI_600 0.82 2.7

KUA/PANI_800 0.88 3.4

KUA-CONH2 0.79 2.6

KUA-CONH2_800 0.85 3.4

KUA-N 0.81 2.8

KUA-N_800 0.84 3.1

Pt/Vulcan 0.91 3.9

Since the parent non-doped carbon (KUA_800) does not show any

enhancement of the catalytic response (Figure 8.4), the increase of the

performance observed for the N-containing samples after treatment at

800ºC is not only related to the effect of heat treatment. This enhancement

of the electrocatalytic response for the ORR is undoubtedly related to the

existence of nitrogen functional groups. The N-doped carbons have


347

similar content of pyridines (0.5-0.6 at. % XPS, Table 8.1) and

pyrroles/pyridones (0.9 – 1.4 at. % XPS, Table 8.1). The main difference

on the nitrogen functional groups arises for KUA/PANI_800, whose

content of pyrroles/pyridones is larger (1.4 at. % XPS). Moreover, this

carbon has higher content of CO-evolving groups (Table 8.1), being more

likely the presence of pyridones. The improvement of the ORR through

pyridones (N-C-O sites) was already proposed in the literature [15,17].

The enhancement of ORR by pyridine-containing carbon catalysts

has been also proposed [19–21,45], but there is controversy about its

positive effect [44]. Nevertheless, the three samples display similar

pyridine content, and consequently this functional group cannot be

responsible of the large improvement observed in KUA/PANI_800. Its

enhanced electrocatalytic properties might be related to the pyridone

functional groups.

It is also worth noting that in spite of the fact that KUA/PANI_600

has higher amount of N species and similar porous texture in comparison

with KUA/PANI_800, its performance towards ORR is significantly

poorer (Figure A1). This is related to the lower conductivity of the sample

when heat-treated at this temperature (600 ºC) [16].

8.3.2.5. Selectivity to water formation

The electron transfer number involved in the ORR was determined

by monitoring the current registered in the Pt ring electrode during the

experiments. Table 8.4 collects the values at 0.5 V registered for all

samples. Figure 8.5 illustrates the evolution of the number of electrons

Chapter 8

348

with the potential for a selection of samples. The pristine carbon material

(KUA) catalyzes the ORR through a mechanism in two steps [9]. The first

one occurs at high potentials and involves the reduction of oxygen to

hydrogen peroxide. Afterwards, the hydrogen peroxide that remains in the

microporosity of the activated carbon is subsequently reduced to

hydroxile ions. Both steps happen through a 2e- process. Hence, the

increase of number of electrons observed for this sample when shifting

the potential to more negative values is explained by the sum of the

electrons involved in both reaction steps (2 + 2 e- pathway) [9].

Figure 8.5. Electron transfer number calculated from RRDE experiments of activated

carbon electrocatalysts in O2-saturated 0.1 M KOH at 5 mV/s and 1600 rpm.

The N-doped carbons heat-treated at 800ºC involve a larger number

of electrons for the ORR. Specifically, the number measured at 0.5 V is

3.4, 3.4 and 3.1 for KUA/PANI_800, KUA-CONH2_800 and KUA-

N_800, respectively. This points out a higher selectivity for the formation

of water than for the non-doped carbons. For comparison purposes, the

1.5

2

2.5

3

3.5

4

0 0.2 0.4 0.6 0.8

n

E (V vs RHE)

KUA

KUA_800

KUA-N_800

KUA-CONH₂_800KUA/PANI_800


349

measurement of KUA_800 (non-doped and heat-treated pristine sample)

was also included in Figure 8.5. This sample shows an enhancement of

the number of electrons along the whole range of potentials in comparison

with the pristine sample KUA. However, its selectivity to water formation

is lower than that demonstrated by the N-doped ones heat treated at 800ºC.

Hence, the enhancement manifested by these samples is related to the

formation of N-functional groups with higher electrocatalytic activity for

the ORR. These samples have large content of pyridones (N-C-O sites),

which have been proposed to enhance the selectivity to water formation

[15,17]. Moreover, the N-doped carbons does not show the important

increase of number of electrons (to 2 +2 e- pathway) displayed by the non-

doped samples at low potentials. The result is especially interesting in case

of KUA/PANI_800, since the number of transfered electrons decreases

with decreasing the potential and the values at the highest potentials are

closer to 4 (i.e., at around 0.7V). Thus, this sample shows the highest

selectivity to water formation compared to the other N-doped samples.

8.4. Conclusions

In this chapter, several N-doped superporous activated carbons (1-4

at. % XPS) were prepared by following different post-functionalization

treatments, including methods involving organic chemistry pathways,

polymerization of aniline and post-thermal treatments. The combination

of these strategies leads to the production of activated carbons with large

apparent surface area and different surface chemistry. The obtained

samples have different functional groups depending on the post-

modification treatment, such as moieties with low thermal stability

Chapter 8

350

(amines, amides) and groups with high thermal stability (pyrroles,

pyridines, etc.). Thus, they have been used to monitor the effect of

nitrogen functional groups on the performance of highly microporous

activated carbons as electrocatalysts for ORR.

The pristine sample displays remarkable electroactivity towards ORR

due to its well-developed microporosity. The N-doped activated carbons

evidence different electrocatalytic performance depending on the

functionalization strategy. The doping methods at low temperature

produce a decrease of the catalytic response due to the generation of non-

catalytic active sites (amines, amides) and the decrease of catalytic edge

sites. The performance is improved for N-doped activated carbons after

heat treatment at 800ºC as consequence of the generation of

electrocatalytic nitrogen-containing functional groups. Polyaniline-

derived activated carbon (heat-treated at 800 ºC) provided the highest

electroactivity (onset potential of 0.88 V) and improved selectivity to

water formation. This enhanced behaviour is explained by its highest

concentration of N-C-O sites.


351

8.5. References.

[1] C.W.B. Bezerra, L. Zhang, H. Liu, K. Lee, A.L.B. Marques, E.P.

Marques, H.Wang, J. Zhang, A review of heat-treatment effects on

activity and stability of PEM fuel cell catalysts for oxygen

reduction reaction, J. Power Sources 173 (2007) 891–908.

[2] C.W.B. Bezerra, L. Zhang, K. Lee, H. Liu, A.L.B. Marques, E.P.

Marques, H. Wang, J. Zhang, A review of Fe-N/C and Co-N/C

catalysts for the oxygen reduction reaction, Electrochim. Acta 53

(2008) 4937–4951.

[3] X. Yu, S. Ye, Recent advances in activity and durability

enhancement of Pt/C catalytic cathode in PEMFC. Part II:

Degradation mechanism and durability enhancement of carbon

supported platinum catalyst, J. Power Sources 172 (2007) 145–154.

[4] R. Borup, J. Meyers, B. Pivovar, Y.S. Kim, R. Mukundan, N.

Garland, D. Myers, M. Wilson, F. Garzon, D. Wood, P. Zelenay,

K. More, K. Stroh, T. Zawodzinski, J. Boncella, J.E. McGrath, M.

Inaba, K. Miyatake, M. Hori, K. Ota, Z. Ogumi, S. Miyata, A.

Nishikata, Z. Siroma, Y. Uchimoto, K. Yasuda, K.I. Kimijima, N.

Iwashita, Scientific aspects of polymer electrolyte fuel cell

durability and degradation, Chem. Rev. 107 (2007) 3904–3951.

[5] X. Yan, Y. Jia, X. Yao, Defects on carbons for electrocatalytic

oxygen reduction, Chem. Soc. Rev. 47 (2018) 7628–7658.

[6] A. Morozan, B. Jousselme, S. Palacin, Low-platinum and

platinum-free catalysts for the oxygen reduction reaction at fuel cell

cathodes, Energy Environ. Sci. 4 (2011) 1238–1254.

[7] G. Wu, A. Santandreu, W. Kellogg, S. Gupta, O. Ogoke, H. Zhang,

H.L. Wang, L. Dai, Carbon nanocomposite catalysts for oxygen

reduction and evolution reactions: From nitrogen doping to

transition-metal addition, Nano Energy 29 (2016) 83–110.

[8] K.H. Wu, D.W. Wang, D.S. Su, I.R. Gentle, A Discussion on the

Activity Origin in Metal-Free Nitrogen-Doped Carbons for

Oxygen Reduction Reaction and their Mechanisms,

Chapter 8

352

ChemSusChem 8 (2015) 2772–2788.

[9] A. Gabe, R. Ruiz-Rosas, C. González-Gaitán, E. Morallón, D.

Cazorla-Amorós, Modeling of oxygen reduction reaction in porous

carbon materials in alkaline medium. Effect of microporosity, J.

Power Sources (2019) 451–464.




[11] A. Gabe, M.J. Mostazo-López, D. Salinas-Torres, E. Morallón, D.

Cazorla-Amorós, Synthesis of conducting polymer/carbon material

composites and their application in electrical energy storage, in:

Hybrid Polym. Compos. Mater. Process., 2017: pp. 173–209.

[12] S.O. Material, S. Web, N. York, A. Nw, High-Performance

Electrocatalysts for Oxygen Reduction Derived from Polyaniline,

Iron, and Cobalt, Science (80) 332 (2012) 443–447.

[13] H.W. Liang, W. Wei, Z.S. Wu, X. Feng, K. Müllen, Mesoporous

metal-nitrogen-doped carbon electrocatalysts for highly efficient

oxygen reduction reaction, J. Am. Chem. Soc. 135 (2013) 16002–

16005.

[14] N. Gavrilov, I.A. Pašti, M. Mitrić, J. Travas-Sejdić, G. Ćirić-

Marjanović, S. V. Mentus, Electrocatalysis of oxygen reduction

reaction on polyaniline-derived nitrogen-doped carbon

nanoparticle surfaces in alkaline media, J. Power Sources 220

(2012) 306–316.

[15] J. Quilez-Bermejo, C. González-Gaitán, E. Morallón, D. Cazorla-

Amorós, Effect of carbonization conditions of polyaniline on its

catalytic activity towards ORR. Some insights about the nature of

the active sites, Carbon 119 (2017) 62–71..

[16] J. Quílez-Bermejo, E. Morallón, D. Cazorla-Amorós, Oxygen-

reduction catalysis of N-doped carbons prepared: Via heat

treatment of polyaniline at over 1100 °c, Chem. Commun. 54

(2018) 4441–4444.


353

[17] R. Silva, D. Voiry, M. Chhowalla, T. Asefa, Efficient metal-free

electrocatalysts for oxygen reduction: Polyaniline-derived N- and

O-doped mesoporous carbons, J. Am. Chem. Soc. 135 (2013)

7823–7826.

[18] J. Zhang, Z. Zhao, Z. Xia, L. Dai, A metal-free bifunctional

electrocatalyst for oxygen reduction and oxygen evolution

reactions, Nat. Nanotechnol. 10 (2015) 444.

[19] F. Zhou, G. Wang, F. Huang, Y. Zhang, M. Pan, Polyaniline

derived N- and O-enriched high surface area hierarchical porous

carbons as an efficient metal-free electrocatalyst for oxygen

reduction, Electrochim. Acta 257 (2017) 73–81.

[20] A. Zhao, J. Masa, M. Muhler, W. Schuhmann, W. Xia, N-doped

carbon synthesized from N-containing polymers as metal-free

catalysts for the oxygen reduction under alkaline conditions,


[21] L. Lai, J.R. Potts, D. Zhan, L. Wang, C.K. Poh, C. Tang, H. Gong,

Z. Shen, J. Lin, R.S. Ruoff, Exploration of the active center

structure of nitrogen-doped graphene-based catalysts for oxygen

reduction reaction, Energy Environ. Sci. 5 (2012) 7936–7942.




electrolyte at severe conditions, Carbon 82 (2015) 205-213.





252–265.

[24] M.J. Mostazo-López, R. Ruiz-Rosas, S. Shiraishi, E. Morallón, D.

Cazorla-Amorós, Nitrogen doped activated carbons prepared at

mild conditions as electrodes for supercapacitors in organic

electrolyte, in: 7th Int. Conf. Carbon Energy Storage Environ.

Prot., 2017: p. 23.

Chapter 8

354






of activated carbon fiber/polyaniline materials by position-resolved

microbeam small-angle X-ray scattering, Carbon 50 (2012) 1051–

1056.

[27] D. Cazorla-Amorós, J. Alcañiz-Monge, M.A. De La Casa-Lillo, A.

Linares-Solano, CO2 as an adsorptive to characterize carbon

molecular sieves and activated carbons, Langmuir 14 (1998) 4589–

4596.





[29] M.J. Bleda-Martínez, E. Morallón, D. Cazorla-Amorós,

Polyaniline/porous carbon electrodes by chemical polymerisation:

Effect of carbon surface chemistry, Electrochim. Acta 52 (2007)

4962–4968.

[30] J.. Figueiredo, M.F.. Pereira, M.M.. Freitas, J.J.. Órfão,

Modification of the surface chemistry of activated carbons, Carbon

37 (1999) 1379–1389.




point petroleum pitch and a melamine resin, Carbon 40 (2002) 597–

608.

[32] E. Raymundo-Piñero, D. Cazorla-Amorós, A. Linares-Solano, The


from flue gases by N-doped activated carbon powders and fibres,

Carbon 41 (2003) 1925–1932.



355


1027.



70 (2014) 59–74.

[35] S. Kuroki, Y. Hosaka, C. Yamauchi, A solid-state NMR study of

the carbonization of polyaniline, Carbon 55 (2013) 160–167.

[36] Z. Rozlívková, M. Trchová, M. Exnerová, J. Stejskal, The

carbonization of granular polyaniline to produce nitrogen-

containing carbon, Synth. Met. 161 (2011) 1122–1129.

[37] K.G. Latham, W.M. Dose, J.A. Allen, S.W. Donne, Nitrogen doped

heat treated and activated hydrothermal carbon: NEXAFS

examination of the carbon surface at different temperatures,

Carbon 128 (2018) 179–190.

[38] M. Gara, R.G. Compton, Activity of carbon electrodes towards

oxygen reduction in acid: A comparative study, New J. Chem. 35

(2011) 2647–2652.

[39] K. Waki, R.A. Wong, H.S. Oktaviano, T. Fujio, T. Nagai, K.

Kimoto, K. Yamada, Non-nitrogen doped and non-metal oxygen

reduction electrocatalysts based on carbon nanotubes: Mechanism

and origin of ORR activity, Energy Environ. Sci. 7 (2014) 1950–

1958.

[40] N.P. Subramanian, X. Li, V. Nallathambi, S.P. Kumaraguru, H.

Colon-Mercado, G. Wu, J.W. Lee, B.N. Popov, Nitrogen-modified

carbon-based catalysts for oxygen reduction reaction in polymer

electrolyte membrane fuel cells, J. Power Sources 188 (2009) 38–

44.

[41] P.H. Matter, U.S. Ozkan, Non-metal catalysts for dioxygen

reduction in an acidic electrolyte, Catal. Letters. 109 (2006) 115–

123.

[42] P.H. Matter, L. Zhang, U.S. Ozkan, The role of nanostructure in

nitrogen-containing carbon catalysts for the oxygen reduction

Chapter 8

356

reaction, J. Catal. 239 (2006) 83–96.

[43] X. Chu, K. Kinoshita, Surface modification of carbons for

enhanced electrochemical activity, Mater. Sci. Eng. B. 49 (1997)

53–60.

[44] T. Ikeda, M. Boero, S.F. Huang, K. Terakura, M. Oshima, J.I.

Ozaki, S.F. Hang, K. Terakura, M. Oshima, J.I. Ozaki, Carbon

Alloy Catalysts: Active Sites for Oxygen Reduction Reaction, J.

Phys. Chem. C. 112 (2008) 14706–14709.

[45] C.V. Rao, C.R. Cabrera, Y. Ishikawa, In search of the active site in

nitrogen-doped carbon nanotube electrodes for the oxygen

reduction reaction, J. Phys. Chem. Lett. 1 (2010) 2622–2627.


Amorós, Functionalization of carbon nanotubes using

aminobenzene acids and electrochemical methods. Electroactivity

for the oxygen reduction reaction, Int. J. Hydrogen Energy 40

(2015) 11242–11253.

[47] G. Tuci, C. Zafferoni, A. Rossin, L. Luconi, A. Milella, M.

Ceppatelli, M. Innocenti, Y. Liu, C. Pham-Huu, G. Giambastiani,

Chemical functionalization of N-doped carbon nanotubes: A

powerful approach to cast light on the electrochemical role of

specific N-functionalities in the oxygen reduction reaction, Catal.

Sci. Technol. 6 (2016) 6226–6236.


357

ANNEX TO CHAPTER 8

Figure A8.1. LSV curves for the catalysts in O2-saturated 0.1 M KOH at 1600 rpm. v =

5 mV/s.

-5

-4

-3

-2

-1

0

-0.1 0.1 0.3 0.5 0.7 0.9 1.1

j (m

A/c

m2)

E (V vs RHE)

KUA

KUA/PANI

KUA/PANI_600

KUA-CONH₂KUA-N

Chapter 9

General conclusions

General conclusions

361

In this PhD Thesis, several nitrogen functionalization methodologies

were employed to produce carbon materials with similar porosity but

different surface chemistry. The effect of nitrogen functionalization on

their porous texture and surface chemistry was thoroughly analyzed.

Moreover, the effect of surface chemistry in their electrochemical

performance was extensively studied in different electrolytes, mainly

dealing with two applications: as electrodes for supercapacitors and

electrocatalysts for the oxygen reduction reaction. Herein, the main

conclusions derived from this work are summarized.

Nitrogen functionalization of activated carbons at mild conditions

• Nitrogen-functionalization of activated carbon using an organic

chemistry protocol at mild conditions was achieved by three

different approaches, which are summarized as follows: (i)

amidation treatment, through a combination of chemical

oxidation, acyl chloride formation and reaction with nitrogen

reagents; (ii) amination treatment by Hofmann rearrangement;

and (iii) direct reaction with nitrogen reagents. These methods

allowed the attachment of ⁓ 4 at. % N.

• The modification pathways leaded to the formation of a wide

range of surface functionalities. The amidation treatment

produced amides (and cyclic derivatives, such as imides and

lactams) derived from CO2-desorbing groups and nitrogen

aromatic heterocycles (pyridine, pyridone and pyrrole) due to

the consumption of CO-evolving groups. The post-treatment by

Hofmann rearrangement produced the conversion of amides into

Chapter 9

362

amines, obtaining an activated carbon with different nitrogen

functionalities (amines, pyridines and pyrroles) and a high

amount of surface oxygen groups. The direct reaction with

nitrogen reactant produced mainly amines and nitrogen

heterocycles.

• The functionalization treatments slightly decreased de

microporosity of the pristine activated carbon. The direct

incorporation of nitrogen via reaction with nitrogen reagents

fully preserved the porous texture of the pristine carbon

material, making this single-step treatment more appropriate for

functionalization of highly microporous carbon materials at

mild conditions.

Electrochemical performance of N-doped activated carbons in aqueous

electrolyte

• The electrochemical characterization of N-doped activated

carbons in aqueous electrolytes showed that, at low current

densities, the capacitance is mainly governed by the apparent

surface area while the surface capacitance is greater for the N-

containing samples due to the influence of nitrogen functional

groups.

• The N-doped activated carbons obtained by amidation and direct

functionalization displayed higher surface capacitance and

improved conductivity. They showed higher capacitance

retention due to the removal of detrimental oxygen

General conclusions

363

functionalities and generation of electron-donating nitrogen

functional groups, such as cyclic amides, pyridines and pyrroles.

• The capacitor based on the N-doped activated carbon obtained

by direct functionalization showed similar energy density, better

capacitance retention and higher power density than the pristine

activated carbon due to the presence of beneficial N groups and

the removal of detrimental electroactive oxygen functionalities.

Also, the nitrogen-functionalized activated carbons used as

electrodes for supercapacitors revealed higher stability than the

pristine activated carbon when working at high voltage

conditions in aqueous medium.

Electrochemical performance of N-doped zeolite-templated carbons in

aqueous electrolyte

• The physicochemical characterization of non-doped and N-

doped zeolite templated carbons evidenced a practically

identical structure but different surface chemistry.

• The N-doped ZTC displayed higher resistance to electro-

oxidation and degradation in acid and alkaline media. Nitrogen-

doping provided higher wettability and improved conductivity

due to presence of quaternary nitrogen.

• Symmetric supercapacitors based on these carbons provided

similar capacitance, but larger energy in case of N-ZTC due to

its higher capacitive behaviour. N-ZTC based supercapacitor

displayed an outstanding maximum power, that is four times

Chapter 9

364

larger than that showed by non-doped ZTC based supercapacitor

(98 and 23 kW/kg), due to the improvement of electrical

conductivity produced by N-Q functionalities in N-ZTC

electrodes. N-ZTC based capacitor evidenced better

performance of ZTC upon cycling due to the stabilizing effect

of N functional groups.

Electrochemical performance of N-doped activated carbons in organic

electrolytes.

• Activated carbons with similar porosity but different surface

chemistry were prepared by combining chemical

functionalization methods at mild conditions and post-thermal

treatments. The heat treatments diminish the content of surface

functionalities and produce rearrangements of the nitrogen

groups. The porosity of the heat-treated samples is almost

identical for the whole series of carbons.

• The surface chemistry of these carbon materials did not

significantly affect the capacitance due to their similar porous

texture.

• The presence of oxygen functional groups affected the rate

performance of the capacitor due to the decrease of conductivity,

while nitrogen functional groups (cyclic amides, pyridines and

pyrroles) slightly increased the conductivity of the carbon

material and improved the performance.

General conclusions

365

• The effect of surface functionalities upon durability was

thoroughly studied. The oxygen functionalities strongly

damaged the performance of the activated carbons. The heat

treatments of the samples produce an improvement of the

electrochemical stability due to the decrease of detrimental

oxygen groups. The performance was further increased by

nitrogen doping at mild conditions, since the treatment

combines the positive effect of removing oxygen groups with

their replacement by nitrogen groups with high electrochemical

stability, which is specialy beneficial in case of the generation

of amide-like functional groups.

Electrochemical performance of N-doped activated carbons in non-

conventional electrolytes.

• The electrochemical characterization of N-doped activated

carbons obtained at mild conditions in two different non-

conventional electrolytes, based in ionic liquids and new

conducting salts (in organic medium), evidenced that surface

chemistry does not significantly modify the capacitance, but

strongly decreases the degradation processes occurring in the

carbon electrodes under positive and negative polarization,

when approaching the limits of the electrochemical stability

window.

• Supercapacitors based on these activated carbons in ionic liquid-

based electrolyte displayed very high capacitance and energy

densities, with improved rate performance due to the presence

Chapter 9

366

of nitrogen functional groups. The durability of the devices

evidenced an enhancement of the performance of N-doped

activated carbons due to the beneficial effect of nitrogen doping

at mild conditions.

Electrochemical performance of N-doped activated carbons as

electrocatalysts for the oxygen reduction reaction

• The effect of nitrogen functional groups on the performance of

highly microporous activated carbons as electrocatalysts for

ORR was assessed by using samples obtained by different post-

functionalization treatments, including methods involving

organic chemistry pathways, polymerization of aniline and post-

thermal treatments.

• These samples have different functional groups according to the

post-modification treatment, such as moieties with low thermal

stability (amines, amides) and groups with high thermal stability

(pyrroles, pyridines, etc.).

• The pristine sample displays remarkable electroactivity towards

ORR due to its well-developed microporosity. The N-doped

activated carbons evidence different electrocatalytic

performance depending on the functionalization strategy.

Polyaniline-derived activated carbon (heat-treated at 800 ºC)

provided the highest electroactivity (onset potential of 0.88 V)

and improved selectivity to water formation. This enhanced

behaviour is explained by its highest concentration of N-C-O

sites.

Conclusiones generales

367

En esta Tesis Doctoral, se han utilizado diversas metodologías de

funcionalización con nitrógeno para producir materiales carbonosos con

textura porosa similar pero distinta química superficial. Se analizó

detalladamente el efecto de la funcionalización con nitrógeno en la textura

porosa y la química superficial. Además, se estudió el efecto de la química

superficial en el comportamiento electroquímico de los materiales en

distintos electrolitos, con el fin de utilizarlos en dos aplicaciones

principales: como electrodos de supercondensadores y como

electrocatalizadores de la reacción de reducción de oxígeno. A

continuación, se presentan las conclusiones principales derivadas de este

trabajo.

Funcionalización con nitrógeno de carbones activados en condiciones

suaves.

• Se ha realizado la funcionalización química con grupos

funcionales nitrogenados de un carbón activado por medio de un

protocolo basado en reacciones orgánicas en condiciones suaves,

que se puede resumir del siguiente modo: (i) tratamiento de

amidación, mediante reacciones de oxidación química, formación

de cloruro de ácido y reacción con reactivos nitrogenados; (ii)

tratamiento de aminación mediante reordenamiento de Hofmann,

y (iii) reacción directa con reactivos nitrogenados. Estos métodos

permiten la incorporación de ⁓ 4 at. % N.

• Estas rutas de modificación química produjeron la formación de

una amplia variedad de grupos funcionales superficiales. El

tratamiento de amidación produjo amidas (y amidas cíclicas, como

Capítulo 9

368

imidas y lactamas), derivados de grupos funcionales que desorben

como CO2, y heterociclos aromáticos (piridinas, piridonas y

pirroles), debido al consumo de grupos que desorben como CO.

Los post-tratamientos mediante el reordenamiento de Hofmann

produjeron la conversión de amidas en aminas, dando lugar a un

carbón activado con distintos grupos funcionales nitrogenados

(aminas, piridinas y pirroles) y una cantidad elevada de grupos

funcionales oxigenados. La reacción directa con los reactivos

nitrogenados produjo principalmente la formación de aminas y

heterociclos nitrogenados.

• Los tratamientos de funcionalización decrecieron levemente la

microporosidad del carbón activado prístino. La incorporación

directa de nitrógeno mediante reacción con reactivos

nitrogenados, permitió conservar la porosidad del material

original, de manera que este tratamiento en una etapa única resultó

el más apropiado para la funcionalización de materiales

carbonosos de elevada microporosidad en condiciones suaves.

Comportamiento electroquímico de carbones activados dopados con

nitrógeno en electrolito acuoso.

• La caracterización electroquímica de carbones activados dopados

con nitrógeno en condiciones suaves mostró que, a bajas

densidades de corriente, la capacidad está gobernada

principalmente por la superficie específica aparente, mientras que

la capacidad superficial es mayor para las muestras que contienen


369

nitrógeno debido a la influencia de los grupos funcionales

nitrogenados.

• Los carbones activados obtenidos por amidación y

funcionalización directa mostraron capacidades superficiales y

conductividades superiores. Estas muestras presentan mayor

retención de capacidad debido a la eliminación de grupos

funcionales oxigenados perjudiciales y a la generación de grupos

nitrogenados electrón-dadores, como amidas cíclicas, piridinas y

pirroles.

• El condensador basado en el carbón activado dopado con

nitrógeno obtenido mediante funcionalización directa mostró, en

comparación con el condensador basado en el material prístino,

energía similar, retención de capacidad mayor y potencia superior,

debido a la presencia de grupos funcionales nitrogenados

beneficiosos y a la eliminación de grupos oxigenados

perjudiciales. Los carbones activados dopados con nitrógeno

utilizados como electrodos de condensadores mostraron mayor

estabilidad que el material original cuando se emplearon en

condiciones de voltaje elevado en medio acuoso.

Comportamiento electroquímico de materiales carbonosos

nanomoldeados dopados con nitrógeno en electrolito acuoso.

• La caracterización fisicoquímica de los materiales estudiados

(materiales carbonosos nanomoldeados dopados con nitrógeno, N-

ZTC, y sin dopar, ZTC) mostró que estos materiales tienen una

estructura prácticamente idéntica y distinta química superficial.

Capítulo 9

370

• El material N-ZTC mostró mayor resistencia a la electro-

oxidación y a la degradación en medio ácido y alcalino. El dopado

con nitrógeno proporcionó mayor mojabilidad y un aumento de la

conductividad debido a la presencia de nitrógeno cuaternario.

• Los supercondensadores simétricos basados en estos materiales

produjeron valores de capacidad similares, pero mayor energía en

el caso de N-ZTC debido a su mayor carácter capacitivo. El

supercondensador basado en N-ZTC mostró una potencia máxima

extraordinaria, que es cuatro veces superior a la proporcionada por

el condensador basado en ZTC no dopado (98 y 23 Wh/kg,

respectivamente), debido a la mejora de la conductividad eléctrica

producida por el nitrógeno cuaternario presente en los electrodos

de N-ZTC. El condensador basado en N-ZTC mostró mejor

comportamiento durante el ciclado debido al efecto estabilizante

de los grupos funcionales nitrogenados.


nitrógeno en electrolitos orgánicos.

• Se han preparado carbones activados con porosidad similar pero

química superficial distinta mediante la combinación métodos

basados en funcionalización química en condiciones suaves y

post-tratamientos térmicos. Los tratamientos térmicos produjeron

la disminución del contenido de grupos funcionales superficiales

y generaron un reordenamiento de grupos funcionales

nitrogenados. La porosidad de las muestras tratadas térmicamente

es casi idéntica para todos los materiales.


371

• La química superficial de estos materiales no afectó

significativamente a la capacidad debido a la textura porosa

similar de los mismos.

• La presencia de grupos funcionales oxigenados afectó a la

retención de capacidad a elevadas densidades de corriente

debido a la pérdida de conductividad, mientras que los grupos

funcionales nitrogenados (amidas cíclicas, piridinas y pirroles)

producen un aumento leve de la conductividad de los materiales

y mejoran el comportamiento a estas densidades de corriente.

• El efecto de los grupos funcionales nitrogenados en la

durabilidad del condensador se estudió en detalle. Los grupos

oxigenados empeoraron significativamente el comportamiento

de los carbones activados. Los tratamientos térmicos condujeron

a un aumento de la estabilidad electroquímica debido a la

eliminación de grupos funcionales oxigenados. La mejora fue

superior en el caso de la funcionalización con nitrógeno en

condiciones suaves, ya que el tratamiento combina el efecto

positivo de la eliminación de grupos oxigenados con su

sustitución por grupos nitrogenados con elevada estabilidad

electroquímica, especialmente en el caso de la generación de

grupos tipo amida.


nitrógeno en electrolitos no convencionales.

• La caracterización electroquímica de los carbones activados

dopados con nitrógeno en condiciones suaves en dos electrolitos

Capítulo 9

372

no convencionales distintos, basados en líquidos iónicos y sales

conductoras nuevas (en medio orgánico), mostró que la química

superficial no modifica significativamente la capacidad, pero

decrece severamente los procesos de degradación que suceden en

los electrodos en condiciones de polarización a potenciales

negativos y positivos, cuando se aproximan a los límites de

estabilidad electroquímica.

• Los supercondensadores basados en estos carbones activados y

electrolitos basados en líquido iónico mostraron capacidades y

energías muy elevadas, además de una mejora de la retención de

capacidad a elevadas densidades de corriente debido a la presencia

de grupos funcionales nitrogenados. La durabilidad de los

dispositivos mostró una mejora del comportamiento de los

condensadores basados en carbones activados dopados con

nitrógeno debido al efecto beneficioso de la funcionalización con

nitrógeno en condiciones suaves.


nitrógeno como electrocatalizadores de la reacción de reducción de

oxígeno

• Se estudió el efecto de los grupos funcionales nitrogenados en

el comportamiento de carbones activados de elevada

microporosidad como electrocatalizadores de la reacción de

reducción de oxígeno, utilizando distintas muestras sintetizadas

por medio de diversas estrategias de funcionalización, que


373

incluyen métodos basadas en reacciones orgánicas,

polimerización de anilina y post-tratamientos térmicos.

• Estas muestras presentan grupos funcionales diferentes, que

dependen de la temperatura de post-modificación, como grupos

con estabilidad térmica baja (aminas, amidas) y grupos de

elevada estabilidad térmica (pirroles, piridinas, etc.).

• La muestra original presenta una electroactividad destacable en

la reacción de reducción de oxígeno debido a su elevada

microporosidad. Los carbones activados dopados con nitrógeno

mostraron electroactividades distintas dependiendo de la

estrategia de funcionalización. Los carbones activados

derivados de polianilina (tratados térmicamente a 800 ºC)

proporcionaron la mayor electroactividad, con un potencial de

inicio de 0.88 V y mayor selectividad a la formación de agua.

Esta mejora del comportamiento se ha relacionado con la mayor

concentración de sitios N-C-O.

Summary

Summary

377

This PhD Thesis focuses on the functionalization of carbon materials

with high microporosity content with nitrogen functional groups and on

their use as electrodes for supercapacitors and electrocatalysts for the

oxygen reduction reaction. Hence, this work describes the incorporation

of nitrogen functionalities by different methodologies on highly

microporous carbon materials (activated carbons and zeolite templated

carbons), their chemical and electrochemical characterization in different

electrolytes and their performance in the proposed applications.

Nitrogen functionalization of a highly microporous activated carbon

(BET surface area higher than 3000 m2/g) has been achieved using several

post-modification treatments based on organic pathways at mild

conditions. These reaction pathways produced the attachment of 4 at. %

in form of different nitrogen functional groups (amides, amines, pyrroles,

etc.) while preserving most of the microporosity of the pristine activated

carbon. The controlled step-by-step modification of the surface chemistry

allowed to assess the influence of the different nitrogen surface groups in

the electrochemical performance of the carbon materials in different

electrolytes: aqueous (1M H2SO4 and 0.1M KOH), organic (1M

TEMAB4/propylene carbonate (PC) and 1M Pyr14BF4/PC) and ionic

liquid-based electrolytes (1M Pyr14TFSI/PC).

The electrochemical performance of N-doped activated carbons

obtained at mild conditions as electrodes for supercapacitors was assessed

in 1M H2SO4. The N-doped activated carbons showed improved rate

performance, larger stability and energy efficiency than the pristine

carbon material when working at high voltages in aqueous electrolyte.

378

These improvements are related to the presence of surface nitrogen

functionalities that provide higher electrochemical stability, avoiding the

formation of detrimental oxygen groups during the operation of the

supercapacitor.

The effect of nitrogen groups in the performance of carbon materials

as electrodes for supercapacitors is also assessed using two zeolite

templated carbons (ZTC) with comparable structure and different surface

chemistry. These materials were synthesized by chemical vapor

deposition of different precursors, producing a non-doped and a N-doped

carbon material (4 at. % XPS) in which most of the functionalities are

quaternary N. N-doped ZTC evidenced larger surface capacitance in acid

electrolyte, increased electrical conductivity and larger resistance to

electro-oxidation, providing higher electrochemical stability than ZTC.

ZTC and N-ZTC capacitors were constructed using 1M H2SO4. N-ZTC

based capacitor evidenced higher energy density and a power density four

times higher than ZTC-based capacitor, producing an outstanding

maximum power of 98 kW/kg due to the enhancement of electrical

conductivity produced by quaternary nitrogen groups. These results

provide clear evidences of the advantages of doping advanced porous

carbon materials with nitrogen functionalities.

The effect of surface chemistry on the performance N-doped and non-

doped activated carbons as electrodes for symmetric supercapacitors was

analyzed in organic electrolyte (1M TEMABF4/propylene carbonate). For

this purpose, several nitrogen-doped activated carbons were synthesized

by different post-modification methods based on organic chemistry

Summary

379

protocols and selective thermal post-treatments under inert atmosphere.

The combination of both methods allowed the production of carbon

materials with very similar surface area (2400-3000 m2/g) and different

surface chemistry. The capacitors based on these carbons showed high

specific capacitance (37-40 F/g) and energy density (31-37 Wh/kg). The

electrochemical stability of the supercapacitors was evaluated by a

floating test under severe conditions of voltage and temperature. The

results evidence an improvement of the durability of nitrogen-doped

activated carbons modified by chemical treatments at mild conditions due

to the decrease of detrimental oxygen functionalities and the generation

of nitrogen groups with higher electrochemical stability.

The electrochemical performance of nitrogen-doped and non-doped

superporous activated carbons as electrodes for supercapacitors was

assessed in non-conventional electrolytes. The devices provide large

capacitance values (up to 150-180 F/g) in both 1M Pyr14TFSI/PC and 1M

Pyr14BF4/PC due to their tailored porous texture (well-developed

microporosity and low mesopore volume). The nitrogen-doped activated

carbons evidence higher electrochemical stability when exposed to

degradation potentials at positive and negative polarization. The activated

carbon-based capacitors displayed outstanding capacitance (37-40 F/g

and 14 F/cm3) and energy values (44-48 Wh/kg and 16-17 Wh/L) with

promising durability due to the stabilizing effect of nitrogen doping at

mild conditions.

The performance of N-doped activated carbons as electrocatalysts for

the oxygen reduction reaction has been analysed in alkaline electrolyte.

380

The activated carbons were prepared via post-functionalization using

polimerization of aniline, organic chemistry reactions and post-thermal

treatments under inert atmosphere. These treatments leaded to the

formation of different nitrogen functionalities with different content. The

evaluation of the electroactivity of the carbon materials towards ORR

evidenced that nitrogen groups generated at high temperatures were

highly selective towards water formation. Among the investigated

samples, polyaniline-derived activated carbon carbonized at 800 ºC

displayed the best performance (onset potential of 0.88V and an electron

transfer number of 3.4), which was attributed to the highest concentration

of N-C-O sites.

Resumen

381

Esta Tesis Doctoral trata sobre la funcionalización de materiales

carbonosos con elevado contenido de microporosidad con grupos

funcionales nitrogenados y su uso como electrodos de

supercondensadores y electrocatalizadores de la reacción de reducción de

oxígeno. De este modo, este trabajo describe la incorporación de grupos

funcionales nitrogenados mediante distintas metodologías en materiales

carbonosos de elevada microporosidad (carbones activados y materiales

carbonosos nanomoldeados), su caracterización química y electroquímica

en distintos electrolitos y su comportamiento en las aplicaciones

propuestas.

Se ha llevado a cabo la funcionalización con nitrógeno de un carbón

activado de elevada microporosidad (área superficial BET mayor que

3000 m2/g) por medio de distintos métodos de post-modificación basados

en reacciones orgánicas en condiciones suaves. Estas reaccions

produjeron el anclaje de 4 at. % de nitrógeno en forma de distintos grupos

funcionales (amidas, aminas, pirroles, etc.) conservando la mayor parte de

la microporosidad del material original. Esta estrategia de modificación

controlada de la química superficial permitió evaluar la influencia de los

distintos grupos funcionales nitrogenados en el comportamiento

electroquímico de los materiales carbonosos en electrolitos diferentes:

acuoso (1M H2SO4 y 0.1M KOH), orgánico (1M TEMAB4/carbonato de

propileno (PC) y 1M Pyr14BF4/PC) y basados en líquidos iónicos (1M

Pyr14TFSI/PC).

Los carbones activados dopados con nitrógeno en condiciones suaves

se emplearon como electrodos de supercondensadores con el fin de

382

evaluar su comportamiento electroquímico en 1M H2SO4. Los carbones

activados dopados con nitrógeno mostraron mayor retención de capacidad

a elevada densidad de corriente, mayor estabilidad electroquímica y

mayor eficiencia energética que el condensador basado en el material

original a voltajes de operación elevados en electrolito acuoso. Esta

mejora está relacionada con la presencia grupos funcionales nitrogenados

que proporcionan mayor estabilidad electroquímica, ya que evitan la

formación de grupos funcionales oxigenados perjudiciales durante el

funcionamiento del supercondensador.

El efecto de los grupos funcionales nitrogenados en el

comportamiento de los materiales carbonosos como electrodos de

supercondensadores se estudió también empleando dos materiales

carbonosos nanomoldeados (ZTC) con estructura comparable y química

superficial diferente. Estos materiales se sintetizaron mediante depósito

químico en fase vapor utilizando distintos precursores, que producen un

material carbonoso dopado con nitrógeno (N-ZTC) y uno no dopado

(ZTC). N-ZTC presenta un 4 at. de nitrógeno, fundamentalmente en forma

de nitrógeno cuaternario. N-ZTC mostró mayor capacidad superficial en

electrolito ácido, mayor conductividad eléctrica y mayor resistencia a la

electro-oxidación, por lo que proporciona una mayor estabilidad

electroquímica que el material no dopado. Los dos materiales se utilizaron

para construir supercondensadores en 1M H2SO4. El condensador basado

en N-ZTC mostró mayor densidad de energía y una densidad de potencia

cuatro veces superior a la proporcionada por el condensador basado en

ZTC, con una potencia máxima extraordinaria de 98 kW/kg debido al

Resumen

383

efecto beneficioso de la mejora de conductividad eléctrica producida por

el nitrógeno cuaternario. Estos resultados proporcionan evidencias claras

acerca de las ventajas de dopar materiales carbonosos porosos avanzados

con grupos funcionales nitrogenados.

El efecto de la química superficial en el comportamiento

electroquímico de carbones activados dopados con nitrógeno como

electrodos de supercondensadores se estudió también en electrolito

orgánico (1M TEMABF4/carbonato de propileno) empleando una

configuración de celda simétrica. Para este propósito, se prepararon

materiales carbonosos dopados con nitrógeno mediante distintas rutas de

post-modificación basadas en reacciones orgánicas y post-tratamientos

térmicos selectivos en atmósfera inerte. La combinación de ambos

métodos premitió la producción de materiales carbonosos con área

superficial aparente muy similar (2400-3000 m2/g) y química superficial

distinta. Los condensadores basados en estos materiales mostraron

elevada capacidad (37-40 F/g) y densidad de energía (31-37 Wh/kg). La

estabilidad electroquímica de los condensadores se evaluó por medio de

un test de durabilidad en condiciones severas de voltaje y temperatura.

Los resultados evidencian una mejora de la durabilidad de los carbones

activados dopados con nitrógeno mediante tratamientos químicos en

condiciones suaves debido a la eliminación de grupos funcionales

oxigenados y a la generación de grupos funcionales nitrogenados con

mayor estabilidad electroquímica.

El comportamiento electroquímico de carbones activados dopados

con nitrógeno y sin dopar como electrodos de supercondensadores se

384

evaluó también en electrolitos no convencionales. Los dispositivos

presentan elevada capacidad (de 150-180 F/g) en 1M Pyr14TFSI/PC y 1M

Pyr14BF4/PC debido a su adecuada textura porosa (elevada

microporosidad y bajo volumen de microporos). Los carbones activados

dopados con nitrógeno presentan elevada estabilidad electroquímica

cuando se exponen a procesos de degradación a potenciales de

polarización positivos y negativos. Los condensadores basados en estos

carbones activados presentan extraordinarias capacidades (37-40 F/g y 14

F/cm3) y densidades de energía (44-48 Wh/kg y 16-17 Wh/L) con una

durabilidad prometedora debido al efecto estabilizador del dopado con

nitrógeno en condiciones suaves.

El comportamiento de los carbones activados dopados con nitrógeno

como electrocatalizadores de la reacción de reducción de oxígeno ha sido

analizado en electrolito alcalino. Los materiales se prepararon por medio

de tratamientos de post-funcionalización empleando polimerización de

anilina, protocolos basados en reacciones orgánicas y post-tratamientos

térmicos en atmósfera inerte. Estos tratamientos conducen a la formación

de grupos funcionales nitrogenados con distinto contenido. El estudio de

la electroactividad de los materiales carbonosos en la reacción de

reducción de oxígeno mostró que los grupos nitrogenados a elevada

temperatura son muy selectivos a la formación de agua. Entre todas las

muestras investigadas, los carbones activados derivados de polianilina

carbonizados a 800ºC presentan el mejor comportamiento (potencial de

inicio de 0.88 V y número de transferencia de electrones de 3.4), que ha

sido atribuido a la mayor concentración de sitios N-C-O.

Funcionalización química de materiales carbonosos para almacenamiento … · 2019-06-04 · La...

Documents

Transcript of Funcionalización química de materiales carbonosos para almacenamiento … · 2019-06-04 · La...