Nanostructure Science and Technology
Series Editor: David J. Lockwood

Kenneth I. Ozoemena
Shaowei Chen Editors

Nanomaterials
for Fuel Cell
Catalysis


Nanostructure Science and Technology
Series editor
David J. Lockwood, FRSC
National Research Council of Canada
Ottawa, Ontario, Canada

More information about this series at />
www.pdfgrip.com


www.pdfgrip.com


Kenneth I. Ozoemena • Shaowei Chen
Editors

Nanomaterials for Fuel Cell
Catalysis

www.pdfgrip.com

Editors
Kenneth I. Ozoemena
Council for Scientific and
Industrial Research
Pretoria, South Africa

Shaowei Chen
Department of Chemistry and Biochemistry
University of California
Santa Cruz, CA, USA

ISSN 1571-5744
ISSN 2197-7976 (electronic)
Nanostructure Science and Technology
ISBN 978-3-319-26249-9
ISBN 978-3-319-29930-3 (eBook)
DOI 10.1007/978-3-319-29930-3
Library of Congress Control Number: 2016942554
© Springer International Publishing Switzerland 2016
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this

book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained
herein or for any errors or omissions that may have been made.
Printed on acid-free paper
This Springer imprint is published by Springer Nature
The registered company is Springer International Publishing AG Switzerland

www.pdfgrip.com


Preface

Ready availability of sufficient energy resources is critical in virtually every aspect
of our life. Whereas fossil fuels have remained our primary energy sources,
extensive efforts have been devoted to fuel cell research in the past few decades,
which represents a unique technology that will make substantial contributions to
our energy needs by converting the chemical energy stored in small (organic)
molecule fuels into electricity and, more importantly, exert minimal negative
impacts on the environment. In fuel cell electrochemistry, the reactions typically
involve the oxidation of fuel molecules at the anode and reduction of oxygen at the
cathode. Both reactions require appropriate catalysts such that a sufficiently high
current density can be generated for practical applications. Precious metals, in
particular, the platinum group metals, have been used extensively as the catalysts
of choice. Yet their high prices and limited reserves have severely hampered the
widespread commercialization of fuel cell technologies. Therefore, a significant
part of recent research efforts has been focused on the development of effective
electrocatalysts with reduced or even zero amounts (and hence costs) of precious
metals used that exhibit competitive or even improved electrocatalytic performance
as compared to state-of-the-art platinum-based catalysts. Toward this end, it is
imperative to understand the fundamental mechanisms involved such that an

unambiguous structure–activity correlation may be established, from which the
activity may then be further enhanced or even optimized. It should be recognized
that the reaction mechanisms in fuel cell electrochemistry are rather complicated
and not fully understood. Yet advances on the theoretical and experimental fronts
have yielded significant insights which offer important guidelines in the design and
engineering of fuel cell catalysts.
It is within this context that this book volume is conceived and developed. The
chapters are written by some of the leading experts in the field. The key goal is to
highlight recent progress in electrocatalysis at both fuel cell anode and cathode,
with a focus on the impacts of the design and engineering of electrode catalysts on
the catalytic performance. This is a critical first step toward the establishment of a
structure–activity correlation. A significant portion is devoted to oxygen reduction
v

www.pdfgrip.com


vi

Preface

reaction, as this has been recognized as a major bottleneck that largely determines
the overall fuel cell performance due to its sluggish electron-transfer kinetics and
complicated reaction pathways. Relevant research is typically centered around
three aspects. The first involves alloying and surface ligand engineering of
platinum-based electrocatalysts (the conventional catalysts) through the so-called
electronic and geometrical contributions such that the costs may be reduced and
concurrently the performance improved. The second entails the development of
non-platinum precious metal nanoparticle catalysts. The third is focused on cheap
transition-metal oxides or totally metal-free catalysts (e.g., doped carbons). In the

last two, while the costs of the catalysts may be substantially reduced as compared
to those of the platinum-based counterparts, their performances have mostly
remained subpar. Thus, how to further improve their activity is a leading challenge
in the field.
Similar issues have been found with the anode reactions. Conventionally, small
(organic) molecules, such as hydrogen, methanol, and formic acid, have been used
as potential fuels. More recently, C2 molecules such as ethanol and ethylene glycol
have also been attracting extensive interest because of their ready availability, low
toxicity, and high energy density. Yet the oxidation mechanisms involved are far
more complex than those of the C1 counterparts (namely, methanol and formic
acid), leading to reduced efficiency in the reaction. Therefore, a major thrust of
current research is to unravel the reaction mechanisms involved such that the
catalytic performance may be further improved and ultimately optimized.
Whereas these issues represent daunting challenges, one may also choose to
accept them as unique opportunities where breakthroughs will help advance fuel
cell technologies toward commercial applications. This is no doubt a
multidisciplinary endeavor, including materials science, (electro)chemistry, interfacial engineering, and so on. It is our hope that this book will offer a unique
glimpse of the state of the art of fuel cell electrocatalysis and therefore may serve as
a technical reference for researchers at all levels in the areas of nanoparticle
materials and fuel cell technologies.
Santa Cruz, CA, USA
Pretoria, South Africa

Shaowei Chen
Kenneth I. Ozoemena

www.pdfgrip.com


Contents


1

2

3

Electrochemistry Fundamentals: Nanomaterials Evaluation
and Fuel Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Neil V. Rees

1

Recent Advances in the Use of Shape-Controlled Metal
Nanoparticles in Electrocatalysis . . . . . . . . . . . . . . . . . . . . . . . . . .
Francisco J. Vidal-Iglesias, Jose´ Solla-Gullo´n, and Juan M. Feliu

31

Pt-Containing Heterogeneous Nanomaterials for Methanol
Oxidation and Oxygen Reduction Reactions . . . . . . . . . . . . . . . . . .
Hui Liu, Feng Ye, and Jun Yang

93

4

Synthesis and Electrocatalysis of Pt-Pd Bimetallic
Nanocrystals for Fuel Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Ruizhong Zhang and Wei Chen


5

Integrated Studies of Au@Pt and Ru@Pt Core-Shell
Nanoparticles by In Situ Electrochemical NMR,
ATR-SEIRAS, and SERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Dejun Chen, Dianne O. Atienza, and YuYe J. Tong

6

Recent Development of Platinum-Based Nanocatalysts
for Oxygen Reduction Electrocatalysis . . . . . . . . . . . . . . . . . . . . . . 253
David Raciti, Zhen Liu, Miaofang Chi, and Chao Wang

7

Enhanced Electrocatalytic Activity of Nanoparticle Catalysts
in Oxygen Reduction by Interfacial Engineering . . . . . . . . . . . . . . 281
Christopher P. Deming, Peiguang Hu, Ke Liu, and Shaowei Chen

vii

www.pdfgrip.com


viii

Contents

8


Primary Oxide Latent Storage and Spillover
for Reversible Electrocatalysis in Oxygen
and Hydrogen Electrode Reactions . . . . . . . . . . . . . . . . . . . . . . . . . 309
Milan M. Jaksic, Angeliki Siokou, Georgios D. Papakonstantinou,
and Jelena M. Jaksic

9

Metal-Organic Frameworks as Materials for Fuel Cell
Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
Henrietta W. Langmi, Jianwei Ren, and Nicholas M. Musyoka

10

Sonoelectrochemical Production of Fuel Cell Nanomaterials . . . . . 409
Bruno G. Pollet and Petros M. Sakkas

11

Direct Ethanol Fuel Cell on Carbon Supported Pt Based
Nanocatalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
T.S. Almeida, N.E. Sahin, P. Olivi, T.W. Napporn,
G. Tremiliosi-Filho, A.R. de Andrade, and K.B. Kokoh

12

Direct Alcohol Fuel Cells: Nanostructured Materials
for the Electrooxidation of Alcohols in Alkaline Media . . . . . . . . . . 477
Hamish Andrew Miller, Francesco Vizza,

and Alessandro Lavacchi

13

Effects of Catalyst-Support Materials on the Performance
of Fuel Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 517
Paul M. Ejikeme, Katlego Makgopa, and Kenneth I. Ozoemena

14

Applications of Nanomaterials in Microbial Fuel Cells . . . . . . . . . . 551
R. Fogel and J.L. Limson

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577

www.pdfgrip.com


Chapter 1

Electrochemistry Fundamentals:
Nanomaterials Evaluation and Fuel Cells
Neil V. Rees

1.1

Introduction

The field of electrochemistry has enjoyed a renaissance in the last two decades as its
relevance to the ever-expanding realm of nanoscience has been appreciated. As new

nanomaterials have been discovered and designed by physical science, many of
their properties of interest have been associated with electron transfer, as solid state
devices or catalysts, most of which can be probed and investigated via electrochemical techniques. It is in the latter case that we focus this chapter, in particular to
fuel cell catalysis. Clearly the aim of a fuel cell is to achieve full oxidation of its fuel
to extract the maximum thermodynamic output, and we therefore require efficient
catalysts in order to achieve this at low to intermediate temperatures (roughly
333–413 K). Electrochemistry plays two roles in the case of fuel cells therefore:
first in characterising and investigating candidate catalysts, and second in understanding and optimising the reduction and oxidation (“redox”) processes occurring
in the fuel cell itself. This chapter will not attempt to cover the whole of this field:
there are many excellent texts on physical electrochemistry [1–4] and a growing
number of similarly authoritative works on the fundamental science of fuel cells
[5–8], but will provide a brief survey of the key concepts and illustrations from
recent literature that those of us working in the field of fuel cell science should be
aware.

N.V. Rees (*)
School of Chemical Engineering, University of Birmingham, Birmingham B15 2TT, UK
e-mail:
© Springer International Publishing Switzerland 2016
K.I. Ozoemena, S. Chen (eds.), Nanomaterials for Fuel Cell Catalysis,
Nanostructure Science and Technology, DOI 10.1007/978-3-319-29930-3_1

www.pdfgrip.com

1


2

1.2


N.V. Rees

Part I: Ex-Situ Electrochemistry

The aim of ex-situ testing of candidate materials for fuel cells is typically to
simplify conditions such that variables can be carefully controlled. In the majority
of cases the testing takes the form of solution-phase cyclic voltammetry of candidate electrocatalysts and so the degree of catalytic activity needs to be ascertained.
In order to understand how this can be performed correctly, avoiding many common pitfalls, we shall first review the underlying physical electrochemistry of the
cyclic voltammetric response.

1.2.1

The Electrode Interface

In general, any interface comprising of two different phases will develop a potential
difference due to different electrical potentials of the two phases. This is true also
for an electrode placed in a test solution. The existence of a potential difference
therefore leads to an electric field gradient within the solution close to the electrode;
where there are ions in the solution they migrate under the influence of this field
such that an electrical double layer is rapidly established, with the Gouy-ChapmanStern model most commonly used to describe it (see Fig. 1.1). Clearly as the
concentration of ions in solution increases, so the diffuse layer becomes more
compressed.
Changes to the electrode interface due to adsorption, etc, can be detected via the
flow of non-Faradaic currents. These are current flows required to maintain
electroneutrality, and are distinct from Faradaic currents which are associated
with electron transfer to/from electroactive species to effect reduction/oxidation.

1.2.2


Mass Transport

Transport of materials through a fluid can occur via diffusion along a concentration
gradient, migration along an electric field gradient, natural convection due to
thermal gradients, or forced convection (due to stirring, flow, etc. deliberately
imposed on the system).
Diffusion is typically described via Ficks Laws
j ẳ D

c
x

1:1ị

2

c
c
ẳD 2
t
x

www.pdfgrip.com

1:2ị


1 Electrochemistry Fundamentals: Nanomaterials Evaluation and Fuel Cells

3


Fig. 1.1 The GouyChapman-Stern model of
the electrical double layer,
with adsorbed counterions
forming the outer
Helmholtz plane (OHP), a
surfeit of counterions in the
diffuse layer and then bulk
solution. Accordingly, the
potential drops from the
electrode potential (ϕm) to
the solution potential (ϕs)

Diffusion to an electrode is also described in terms of the degree of development of
the diffusion field around the electrode. As the electrode reaches a potential where
electron transfer occurs and so a concentration gradient is established, diffusion is
initially linear (or one-dimensional) and the diffusion field gradually develops such
that eventually it becomes convergent (see Fig. 1.2).
The time taken for the establishment of convergent diffusion, tconv, can be
estimated by the expression [4]
tconv %

r2
D

ð1:3Þ

Migration effects are often undesired in ex-situ experiments, and so an excess of
inert (or supporting) electrolyte is commonly added to eliminate them. However, in
the absence of excess electrolyte (i.e. less than full support), then migration will

become significant and needs to be accounted for via the Nernst-Planck and Poisson
equations [9]


Á
∂C
zF
2
2
ẳD Cỵ
C ỵ C
t
RT

www.pdfgrip.com

1:4ị


4

N.V. Rees

Fig. 1.2 (a) Linear and (b) convergent diffusion fields to an electrode

2 ẳ


s 0


1:5ị

where z is the species charge, ϕ is the potential, εs is the dielectric constant of the
solvent medium, ε0 is the permittivity of free space, and ρ is the local charge
density, found from summing all local charges present
X
1:6ị
ẳ F i zi Ci
Natural convection effects are notoriously unpredictable and so are generally
eliminated by thermostatitng of the electrochemical cell and conducting the experiment over as short a time as practicable, where a timescale of 20–30 s is usually
held to be the maximum desired [10].
Forced convection is imposed deliberately on the system in order to increase
mass transport. As such the formed of convection is usually chosen such that the
hydrodynamics of the system are well-defined [11].

1.2.3

Electrode Kinetics

The simplest case is that of fully reversible behaviour (i.e. where the electron
transfer kinetics are effectively infinitely fast on the timescale of the mass transport), where the Nernst equation will hold at all times at the electrode surface [1].
For quasi-reversible and irreversible systems, account must be made of the
non-Nernstian condition at the electrode surface and this is most commonly
achieved via the Butler-Volmer equations
F

kf ẳ k0 e RT
kb ẳ k0 e

ỵF

RT

1:7ị
1:8ị

where + β ¼ 1, and η ¼ E À E0.
These are routinely available in commercial simulation packages and are phenomenological insofar as there is limited molecular insight gained from their

www.pdfgrip.com


1 Electrochemistry Fundamentals: Nanomaterials Evaluation and Fuel Cells

5

results. More theoretically informative would be the use of Marcus-Hush theory
[12–18], which can be used in a predictive sense as well as modelling experimental
data. Here the reductive and oxidative rates constants (kf and kb respectively) can be
expressed as
n
o
Zỵ1
exp G6ẳ


ị=k
T
B
red
kf ẳ

d
Ared ị
1 ỵ expfe Eị=kB T g

1:9ị

1

n
o
Zỵ1
exp G6ẳ


ị=k
T
B
red
kb ẳ
d
Aox ị
1 ỵ expfỵe EÞ=kB T g

ð1:10Þ

À1

where e is the electronic charge, kB the Boltzmann constant, Ared/ox(ε) is the
pre-exponential factor that includes the influence of the electronic states on metallic
electrode and electroactive species and the density of states of the electrode. If A is

assumed to be independent of energy, ε, then the general form for the reductive and
oxidative rate constants can be written as
(

À
Á)
I red η* ; λ*
À
Á
kf ¼ k0
I red 0; λ*
( À
Á)
I ox * ; *


kb ẳ k0
I ox 0; *

1:11ị
1:12ị

where *,
 λ* aredimensionless overpotential and reorganisation energy given by
*
F
Fλ
, respectively. I(η*, λ*) is an integral of the form
η ¼ RT E E0f and * ẳ RT



I red ;


*

ẳ

n
o
Zỵ1 exp G6ẳ ị
red
1



I ox ;

*



Zỵ1
ẳ
1

1 ỵ expf g

d


ẩ
ẫ
exp G6ẳ
ox ị
d
1 ỵ expfỵ g

1:13ị

1:14ị

F
Eị.
where ẳ RT
It is most commonly encountered in its symmetric representation [19, 20], where


2
*
* ỵ
1ỵ
4
*

2
*
* ỵ
1




ị
ẳ
G6ẳ
sym, ox
4
*

G6ẳ
sym, red ị ẳ

www.pdfgrip.com

1:15ị
1:16ị


6

N.V. Rees

However, this can lead to poor fitting to experimental data in certain circumstances,
for example where the electron transfer is extremely irreversible, and so the
asymmetric representation is potentially more powerful [19, 20].
2
 *
(
 *
2 )
*

*


ỵ


ỵ


ỵ

1ỵ
G6ẳ
ỵ *
1
asym, red ị ẳ
4
*
4*
*
2

*
ỵ
16*

2
 *
(
 *

2 )
*
* ỵ
ỵ
6ẳ
* ỵ
1
Gasym, ox ị ẳ
ỵ
1
4
*
4*
*

1:17ị

2

ỵ

1.2.4

*
16*

1:18ị

The Cyclic Voltammogram


The cyclic voltammogram, commonly abbreviated to CV, shows the current
response to a triangular voltage ramp (see Fig. 1.3) and owes its shape to the
complex interplay of both electrode kinetics and mass transport effects. As such
it can take different forms depending on the exact details of these two factors, often
described as transient or steady-state.

1.2.4.1

Transient Cyclic Voltammetry

Figure 1.3a shows the transient form of the CV, which is obtained for linear
diffusion systems, most commonly “macro”electrodes (of characteristic dimension
>100 μm) or smaller electrodes at very short timescales (i.e. high voltage scan
rates) [21]. Here the characteristic shape is due to the interaction of kinetics with
mass transport: the current maximum is due to the competing factors of (i) an
exponentially increasing rate of electron transfer (see Butler-Volmer equations),
and (ii) rapid depletion of the electroactive material in the double layer caused by
relatively slow diffusion of fresh material from bulk solution. In a sufficiently large
volume of solution, these factors equilibrate at the diffusion limiting current (Ilim).
The transient CV is usually characterised in terms of the peak currents (anodic
and cathodic, Ip,a and Ip,c) and the separation of the anodic and cathodic peak
potentials (ΔEpp ¼ Ep,a À Ep,c). The reversibility of the redox couple is defined in
terms of the peak separation [1] (Fig. 1.4, Table 1.1):

www.pdfgrip.com


1 Electrochemistry Fundamentals: Nanomaterials Evaluation and Fuel Cells

I/A


(b)

I/A

(a)

7

E/V

E/V

Fig. 1.3 Typical cyclic voltammograms for (a) transient and (b) steady-state responses, due to
linear and convergent diffusion respectively

I/A

(Epa, Ipa)

(Epc, Ipc)

E/V
Fig. 1.4 Characteristics of a transient cyclic voltammogram

www.pdfgrip.com


8


N.V. Rees

Table 1.1 Cyclic
voltammetric definitions

ΔEpp ¼ Epa À Epc (mV)
60/n

Definition
Reversible

60/n < Epp < 120/n
>120/n

Quasi-reversible
Irreversible

Other relations
E0 ẳ


Epa ỵEpc
2

E0 ẳ

Epa ỵEpc
2

It should be noted here that the reversibility of a redox system is clearly a

function of the timescale of the experiment, with a redox couple typically
displaying reversible behaviour at slow voltage scan rates, and quasi-reversible or
even irreversible behaviour at higher voltage scan rates.
Analytical results for the peak currents are available for the reversible and
irreversible cases, due to Randles and Se`vcˇ´ık [1]:
À
Á
I p, rev ¼ 2:69 Â 105 nACbulk D1=2 ν1=2
À
Á
3=2
ACbulk D1=2 ν1=2
I p, irrev ẳ 2:99 105 n3=2


1:19ị
1:20ị

where, n is the total number of electrons transferred, nα is the number of electrons
transferred in the rate limiting step, A the geometric electrode area, Cbulk the bulk
concentration of the electroactive species, D the diffusion coefficient, α the transfer
coefficient, and ν the voltage scan rate.

1.2.4.2

Steady-State Voltammetry

In Fig. 1.3b, a steady-state CV is shown: this is most commonly observed for
microelectrode voltammetry (for electrodes with characteristic dimension
<20 μm), where the diffusion field is fully convergent. The sigmoidal shape is

due to the rate of mass transport being sufficiently fast to ‘keep up’ with the
accelerating rate of electron transfer at least until the limiting current (Ilim) is
reached.
Since the reverse scan retraces the forward scan in the absence of hysteresis
effects, the steady-state voltammogram is often recorded as a linear sweep rather
than a triangular voltage ramp. The limiting current is given by [1]
I lim ẳ 4nFCbulk Dr

1:21ị

where n, Cbulk, and D have their usual meaning, F is the Faraday constant and r the
disk radius.

1.2.4.3

The Transition Between Transient and Steady-State

The position of the voltammetric “wave” can be seen to shift to higher
overpotentials as the voltammogram shifts from fully transient to fully steadystate, that is, as the mass transport accelerates from linear to convergent

www.pdfgrip.com


1 Electrochemistry Fundamentals: Nanomaterials Evaluation and Fuel Cells

9

Fig. 1.5 Simulation cyclic voltammograms showing transition from transient to steady state and
the subsequent potential shift in the voltammogram. All parameters are the same, except from
electrode radius which varies: 1 mm (black), 10 μm (magenta), 1 μm (red), 100 nm (green), and

10 nm (blue)

(Fig. 1.5). There is no change to the electrode kinetics in this case, the effect is
purely one of mass transport. A simple way to understand this is that the higher rates
of diffusion sweeps material to and from the electrode more rapidly, hence the
electroactive species spends less time in the near vicinity of the electrode and a
higher overpotential is required to drive the electron transfer at a fast enough rate
for the species to react in that shorter time period.

1.2.4.4

Adsorbed and Thin-Layer Voltammetry

The voltammetry described thus far has been implicitly concerned with a large
volume of solution containing a bulk concentration of electroactive species which is
not significantly depleted by the electrolysis occurring at the electrode.
Figure 1.6 shows a schematic of both adsorbed and thin-layer systems with their
characteristic voltammetric responses. In both cases the current response decays to
zero as the concentration of electroactive species is exhausted. Note that in the
adsorbed case, the anodic and cathodic peaks are ideally symmetrical about the
potential axis due to the absence of mass transport. For thin-layer voltammetry,
there is an asymmetry due to the limited mass transport (usually diffusion) occurring in the thin-layer volume.

www.pdfgrip.com


10

N.V. Rees


Fig. 1.6 Schematic voltammograms for (a) adsorbed species, and (b) thin-layer (finite volume)
solution

1.2.5

Hydrodynamic Voltammetry

There are a wide range of different hydrodynamic electrodes, where forced convection is applied to achieve a well-defined fluid behaviour [11]. These commonly
differ in:
(i)
(ii)
(iii)
(iv)
(v)
(vi)

size of electrode—macro vs micro
geometry—disk, band, tube, ring, ring-disk
method of convection—flow systems, rotating systems, impinging jet systems
symmetry—axisymmetric vs non axisymmetric
uniformity of accessibility
flow regime—laminar vs fully turbulent

Alternative detailed reviews consider this as a topic [11, 22], but here we shall
focus on the rotating disk electrode (RDE) and rotating ring-disk electrode (RRDE)
as they are commonly used within the fuel cell community (Fig. 1.7).
The RDE (and RRDE) have been in use for many years, having been developed
prior to the advent of microelectrodes as a means to shorten the experimental
timescale and measure more rapid processes than could be achieved via
diffusion-only studies with macroelectrodes [23, 24].

The electrode is mounted axisymmetrically on an insulated rotating shaft, and
rotated at sufficient frequency to set up laminar motion in the solution.
In this case the high rates of convective mass transport ensure a steady-state
voltammetric response where the limiting current is given by the Levich equation
[23]:

www.pdfgrip.com


1 Electrochemistry Fundamentals: Nanomaterials Evaluation and Fuel Cells

11

Fig. 1.7 Schematic
diagrams of (a) rotating
disk, and (b) rotating ringdisk electrodes

I lim ¼ 0:62nFACbulk D2=3 υÀ1=6 ω1=2

ð1:22Þ

where υ is the kinematic viscosity of the solution and ω is the angular speed (rad sÀ1).
Kinetic parameters can be extracted via Koutecky-Levich analysis, which
deconvolutes the total current into kinetic and limiting contributions [25, 26]:
À1
I À1 ¼ I 1
lim ỵ I k

1:23ị


Plotting 1/I vs 1/2 at a range of potentials yields intercepts of value 1/Ik. Since Ik is
a function of E, given by
I k ẳ nFACbulk kEị

1:24ị

where k(E) is the potential-dependent electrochemical rate constant given by the
Butler-Volmer expression


F
kEị ẳ k0 exp
RT

1:25ị

Then a further (Tafel) plot of ln Ik vs η provides α and k0. The Koutecky-Levich
method is widely used for multi-electron systems, most notably oxygen reduction
(ORR). However, recent work by Masa et al. suggests that for accurate results it is
necessary to take account of the surface roughness of the catalyst deposit, since the

www.pdfgrip.com


12

N.V. Rees

ratio of actual to geometric surface area will determine how the apparent rate
constant, k0app, differs from the true rate constant, k0 [27].

The RRDE is usually used in a generator-collector type experiment which can be
illustrated by the (simplified) ORR reaction as follows.
k1

O2 gị ỵ 4H ỵ aqị ỵ 4e ! 2H2 Olị
k3

k2

O2 gị ỵ 2H ỵ aqị ỵ 2e ! H 2 O2 ðadsÞ ! H 2 O2 ðaqÞ
k4

H 2 O2 adsị ỵ 2H ỵ aqị ỵ 2e ! 2H2 OðlÞ
The disk is subjected to a normal voltage scan, in this case to reduce oxygen, whilst
the ring potential (Ering) is held at a potential sufficient to oxidise the intermediate
(here, hydrogen peroxide). The disk current therefore provides information on the
rate of reduction of parent species, whilst the ring current provides information on
the amount of intermediate produced.

1.2.6

The Voltammetry of Nanoparticles

The vast majority of catalysts are nanomaterials (assumed spherical) and hence it is
important to identify those characteristics that define their voltammetry.
An isolated nanoparticle on a planar substrate behaves as a spherical electrode
with a hindered convergent diffusion field (see Fig. 1.8).
For which Bobbert et al. have derived the limiting current expression to be [28]:
I lim ẳ 4 ln2ịnFCbulk Dr np
where rnp is the radius of the nanoparticle.

Fig. 1.8 Schematic
diagram showing
convergent diffusion to a
(nano)sphere on a plane

www.pdfgrip.com

ð1:26Þ


1 Electrochemistry Fundamentals: Nanomaterials Evaluation and Fuel Cells

13

The voltammetry for such as system would appear to be relatively straightforward: the nanoscale size of the ‘electrode’ (i.e. particle) ensures that all diffusion is
convergent on all practical timescales (see Eq. 1.3), and the exceptionally high rates
of mass transport associated with this would be expected to cause a shift of the wave
to higher overpotentials.
However, it is not common to measure single particle electrochemistry (see
later), and instead nanocatalysts are usually studied and used in very large numbers.
The catalysts can be deposited directly onto a substrate electrode (usually glassy
carbon, GC) or be deposited upon a carbon support material during their fabrication, and the catalyst/support then deposited onto the substrate. In this case the
carbon support is usually carbon black (often the commercial Vulcan XC-70), but
studies are increasingly using nanotubes, graphenes, etc. There are subtle differences to these two cases, so we consider them in turn.

1.2.6.1

Nanoparticle-Substrate Systems

In this case the nanoparticles are deposited upon the substrate, often via spraying or

drop-casting and subsequent evaporation of solvent. The particles are essentially
randomly distributed across the surface, and typically are in such numbers that they
do not form a monolayer across the entire substrate electrode surface (i.e. coverage
<100 %), shown schematically in Fig. 1.9.
By considering the diffusion to each individual particle, it should be clear that
there are two extreme cases (and several intermediate ones) which may occur as a
result of the relative spacing of the particles from each other [29].
First, where the particles are very far apart (i.e. extremely low coverage) such
that each individual particle develops a convergent diffusion field independent of
every other particle. Each particle is diffusionally isolated, and so the voltammetry
of the ensemble (or N particles) will appear to be a sigmoidal (steady-state)
voltammogram, with limiting current given by
I lim ¼ 4Nπ ðln2ÞnFCbulk Drnp

ð1:27Þ

Second, where the particles are very close together (i.e. high coverages) such that
each individual particle’s diffusion field interferes or overlaps with its neighbour’s,
then the overall effect is for the ensemble to behave as if the whole area covered is
Fig. 1.9 Randomly
deposited particles on a
substrate electrode

www.pdfgrip.com


14

N.V. Rees


Fig. 1.10 The four cases of diffusion to spheres on a plane (particles on a substrate electrode)

subject to linear diffusion. The voltammetry will therefore appear to be of a
transient voltammogram corresponding to an active area equal to that of the
geometric area over which the particles are deposited (i.e. the area of the substrate
electrode). The usual equations relating to macroelectrode voltammetry
(i.e. Randles-Se`vcˇ´ık) will apply.
These two extremes are often termed Case 1 and Case 4 behaviour respectively,
and as they names suggest, two intermediate cases have been identified
corresponding to increasing degrees of overlap of diffusion fields associated with
neighbouring nanoparticles [19] (Fig. 1.10).
When considering the likely Case in operation for a given system, it is worthwhile considering the diffusion length on the experimental timescale, t [30]:
hxi %

pffiffiffiffiffiffiffiffi
2Dt

ð1:28Þ

For the reason that even if the particles are small and relatively far apart, if the
timescale of the experiment is sufficiently long that ‹x› approaches half of the interparticle separation, then diffusion field interference will occur and Cases 2–4 will
apply.

1.2.6.2

Nanoparticle@Support-Substrate Systems

There are three main ways for the presence of carbon support particles to alter the
voltammetry. First, by increasing the resistance between the catalyst particle and
substrate: this is usually not a significant effect in the cases of highly conductive


www.pdfgrip.com


1 Electrochemistry Fundamentals: Nanomaterials Evaluation and Fuel Cells

15

carbons (carbon black, nanotubes, etc.), but can become noticeable for less conductive supports such as graphene oxide.
Second, the partial or full occlusion of the catalyst particle. This can occur as
part of the catalyst fabrication, as this commonly involves a chemical reduction of
the dissolved catalyst precursor salt in the presence of the carbon support in
suspension. Any degree of aggregation of the support particles will necessarily
occlude catalyst particles. In some cases, it has been reported that during fabrication
the catalyst particles may intercalate graphitic particles of carbon support. Alternatively, occlusion may occur during the deposition of the catalyst@support particles onto the substrate through multilayer formation. In these scenarios, the
voltammetric changes will be slight, with Cases I–III affected to a decreasing
degree and Case IV unaffected. In Case I, the voltammetry will remain steadystate, except the limiting current will reflect the lower apparent radius of the
nanoparticles.
Third, the deposition of higher coverages, or strongly aggregated
catalyst@support particles will lead to the existence of voids within the
catalyst@support deposit. There will be extremely hindered diffusion to these
voids from bulk solution and a mixed diffusion regime will be established, where
either convergent or linear diffusion occurs to catalyst particles on the surface of the
deposit and thin-layer diffusion occurs to catalyst particles within voids in the body
of the layer [31–33] (Fig. 1.11).
The resulting voltammetry of such a mixed diffusion regime is therefore some
algebraic sum of linear (or convergent is applicable) diffusion and thin-layer
diffusion signals. Since the latter, by definition, occurs at a lower overpotential
than the former, there is a necessary shift in overpotential to lower values than a
fully linear diffusion response (Fig. 1.12).


1.2.7

Electrocatalysis

An electrocatalyst increases the rate of an electrochemical reaction by providing a
lower energy pathway across the reaction coordinate. Due to the potential dependence of the electrochemical rate constant on overpotential (see Eqs. 1.7 and 1.8)

Fig. 1.11 A multi-layer deposit with illustrative voids

www.pdfgrip.com


16

N.V. Rees

250

Semi-infinite diffusion
Thin layer diffusion

200
150

i / mA

100
50
0

-50
-100
-150
-200
-0.8

-0.6

-0.4

-0.2

-0.0

0.2

0.4

0.6

0.8

E/V
Fig. 1.12 Illustration of how mixed diffusion regime can cause a lowering of overpotential.
Reproduced with permission from [34], copyright 2008 Elsevier

this necessarily means that the overpotential required for the electrochemical
reaction is decreased and the voltammetric wave is shifted to lower overpotentials.
However, the converse is certainly not necessarily true.
There is a widespread naăvety that a candidate catalyst that causes a shift of the

voltammetric wave to lower overpotentials must therefore be catalytic. The preceding discussion should make it clear that this conclusion can only be reached if
the mass transport to the system has been fully characterised and understood. We
have seen that the size of the particle (i.e. electrode) can cause a shift in
overpotential through the increasing rates of mass transport via diffusion as particle
size decreases. Further, the arrangement of the deposited particles will affect the
voltammetric shape (and therefore position on the potential axis) depending on
whether Case I–IV behaviour is followed. Finally, and most subtle, the existence of
a mixed (linear-thin layer) diffusion regime is itself sufficient to cause a lowering of
the observed overpotential.
In any evaluation of nanoparticulate materials as possible catalysts, it is therefore imperative to ensure that the role of coverage, support particles, etc. are taken
into account, through careful control experiments and simulation. Ideally, any study
of candidate catalysts should measure the kinetic parameters for the reaction, as this
is the only truly unambiguous measure of catalytic activity.

www.pdfgrip.com