The Pennsylvania State University

The Graduate School

Department of Energy and Mineral Engineering

CONDUCTIVITY OF ION EXCHANGE MATERIALS

A Thesis in

Energy and Mineral Engineering

by

Tomoki Naya

 2010 Tomoki Naya

Submitted in Partial Fulfillment

of the Requirements
for the Degree of

Master of Science

May 2010
 ii

The thesis of Tomoki Naya has been reviewed and approved* by the following:

Serguei N. Lvov
Professor of Energy and Mineral Engineering &
Material Science and Engineering
Thesis Co-Advisor

Sridhar Komarneni
Distinguished Professor of Clay Mineralogy
Thesis Co-Advisor

Derek Elsworth
Professor of Energy and Mineral Engineering

Jeffrey R.S. Brownson

Assistant Professor of Energy and Mineral Engineering

Yaw D. Yeboah
Professor of Energy and Mineral Engineering
Head of the Department of Energy and Mineral Engineering

* Signatures are on file in the Graduate School

 iii

ABSTRACT

Proton exchange membrane fuel cells (PEMFCs) and alkaline fuel cells (AFCs) have

received much attention as two of the promising energy conversion systems which can be

applied in many areas. In order to improve performance of the cells, the ionic conductivity of the

ion exchange membranes (IEMs) have been studied by many research groups and a considerable

number of IEMs and materials have been synthesized to show high conductivities in wide ranges

of temperature and humidity.

The measurement of ionic conductivity is crucially important in order to evaluate the

performance of a newly synthesized material. Conventionally, the in-plane gas phase

measurement has been used for thin-film organic membranes and the through-plane gas phase

measurement has been used for inorganic-based membranes. However, the in-plane method

measures the conductivity in a different direction from the operating condition of PEMFCs and

AFCs. In order to overcome this problem, we applied the through-plane measurement for the

thin-film membrane and compared the results with those of the conventional in-plane

measurement. Similar approach was implemented for the liquid phase measurements.

As a result, we found that the ionic conductivities in the membrane measured in the

through-plane direction were significantly lower than those measured in the in-plane direction.

Thus, we concluded that the through-plane measurement has some challenges to be solved and

the through-plane measurement should be carried out more carefully than the in-plane

measurement.
 iv

In addition to these studies, we have investigated newly synthesized OH- conductive

inorganic materials using x-ray diffraction (XRD), scanning electron microscopy (SEM), and the

ionic conductivity measurement in the through-plane direction. The materials showed high

conductivities comparable to other reported inorganic materials in the low temperature and low

humidity ranges. However, their conductivities are much lower than the conventionally used in

potassium chloride-based electrolyzers. Therefore, we need further improve the ionic

conductivity of the OH- conductive inorganic materials for the development of an efficient AFC.
 v

TABLE OF CONTENTS

LIST OF FIGURES ..................................................................................................................... viii

LIST OF TABLES ......................................................................................................................... xi

NOMENCLATURE ..................................................................................................................... xii

ACKNOWLEDGEMENT ............................................................................................................ xv

Chapter 1 Introduction .................................................................................................................... 1

1.1 Fuel cells ............................................................................................................................. 1

1.1.1 PEM fuel cells ............................................................................................................... 2

1.1.2. Alkaline fuel cells ......................................................................................................... 4

1.2 Ion exchange membranes .................................................................................................... 5

1.2.1 Basis of ion exchange membranes ............................................................................... 5

1.2.2 Development of IEMs for PEMFCs and AFCs ........................................................... 5

1.3 Problem statement and methodology ................................................................................ 10

Chapter 2 Theory and experimental technique ............................................................................ 11

2.1 Ion conducting mechanisms .............................................................................................. 11

2.1.1 Ion exchange mechanism by thermodynamics .......................................................... 11

2.1.2 Ion exchange mechanism in Nafion membrane ......................................................... 12

2.2 Conductivity measurements .............................................................................................. 14

2.2.1 Conductivity cells ...................................................................................................... 14

2.2.2 Calculation for each measurement ............................................................................. 18

 vi

2.3 Electrochemical impedance spectroscopy ........................................................................ 19

2.3.1 Modeling for equivalent circuit ................................................................................. 19

2.3.2 Basic theory of electrochemical impedance spectroscopy......................................... 20

Chapter 3 Measurement for PEMs ............................................................................................... 23

3.1 In-plane measurement in gas phase .................................................................................. 23

3.1.1 Experimental setup..................................................................................................... 23

3.1.2 Results ........................................................................................................................ 25

3.2 Through-plane measurement in gas phase ........................................................................ 28

3.2.1 Experimental setup..................................................................................................... 28

3.2.2 Results ........................................................................................................................ 28

3.3 Through-plane measurement in liquid phase .................................................................... 33

3.3.1 Experimental setup..................................................................................................... 33

3.3.2 Results ........................................................................................................................ 34

3.3.4 PEM measurement summary ..................................................................................... 35

Chapter 4 Measurement for OH- conductive materials................................................................ 36

4.1 Synthesis and characterizations of inorganic materials .................................................... 36

4.1.1 Background on layered double hydroxides and hydroxyapatite ................................ 36

4.1.2 Synthesis of sample materials .................................................................................... 39

4.1.3 Characterizations of sample materials ....................................................................... 41

4.2 Conductivity measurement ............................................................................................... 45

4.2.1 Sample preparations and experimental setup ............................................................. 45

 vii

4.2.2 Results ........................................................................................................................ 46

Chapter 5 Conclusion.................................................................................................................. 50

BIBLIOGRAPHY ......................................................................................................................... 51
 viii

LIST OF FIGURES

Figure 1-1: The concept of fuel cells……………………………………………………………...1

Figure 1-2: Typical polarization curve for a PEMFC…………………………………………….3

Figure 1-3: The structure of Nafion membrane…………………………………………………...6

Figure 2-1: The interface between the solution and the membrane……………………………...11

Figure 2-2: The structure model of Nafion membrane…………………………………………..13

Figure 2-3: The conductivity cell in liquid phase………………………………………………..14

Figure 2-4: The cross section of the cell…………………………………………………………15

Figure 2-5: The conductivity cell in gas phase…………………………………………………..15

Figure 2-6: The cross section and the width of the membrane…………………………………..16

Figure 2-7: The sample and electrodes of in-plane method……………………………………..16

Figure 2-8: The interface concept between the electrodes and the electrolyte…………………..19

Figure 2-9: The equivalent circuit………………………………………………………………..20

Figure 2-10: Typical impedance plots…………………………………………………………...21

Figure 3-1: Gas phase measurement setup………………………………………………………23

 ix

Figure 3-2: Ohmic responses in the in-plane gas phase measurement…………………………..25

Figure 3-3: Conductivity in the in-plane gas phase measurement……………………………….26

Figure 3-4: The Nyquist plot in the through-plane gas phase measurement…………………….29

Figure 3-5: The equivalent circuit model for the measurement…………………………………30

Figure 3-6: Conductivity in the through-plane gas phase measurement………………………...30

Figure 3-7: Nyquist plot in liquid phase measurement of solution and membrane……………...34

Figure 4-1: Schematic representation of the layered double hydroxide structure……………….37

Figure 4-2: Hydroxyapatite structure projected on x, y plane…………………………………...38

Figure 4-3: Scanning electron micrograph showing the particle morphology and size of Chloride

containing Mg-Al LDH (JK-3)…………………………………………………………………..42

Figure 4-4: Scanning electron micrographs showing the particle morphology and size of Chloride

containing Ca-Al LDH (JK-4) at two different magnifications………………………………….43

Figure 4-5: X-ray diffraction patterns of (a) chloride containing Mg-Al LDH (JK-3) and (b)

chloride containing Ca-Al LDH (JK-4)………………………………………………………….44

Figure 4-6: X-ray diffraction pattern of hydroxide containing Mg:Al LDH (YN 252)…………44

Figure 4-7: X-ray diffraction pattern of hydroxyapatite (YN 148)……………………………...45

Figure 4-8: Nyquist plots of inorganic materials at 95 %RH……………………………………46

 x

Figure 4-9: OH- Conductivities in inorganic materials…………………………………………48

 xi

LIST OF TABLES

Table 1-1: The reported organic membrane examples and their ionic conductivities

for PEMFCs….……………………………………………………………………………………8

Table 1-2: The reported inorganic material examples and their ionic conductivities

for PEMFC………………………………………………………………………………………..8

Table 1-3: The reported membrane examples and their ionic conductivities for AFCs………….9

Table 3-1: The experimental procedure………………………………………………………….24

Table 3-2: Conductivity data in the in-plane gas phase measurement…………………………...27

Table 3-3: Conductivity data in the through-plane gas phase measurement....………………….32

 xii

NOMENCLATURE

A Surface area of a membrane (cm2)

a Activity

a Activity in membranes

B Tafel slope (V)

C Capacitance (F)

Cdl Double layer capacitance (F)

d Diameter of a disk (cm)

E Potential difference (V)

Ediff Diffusion potential (V)

EDon Donnan potential (V)

f Frequency (Hz)

H Height of a membrane in the in-plane direction (cm)

k Reaction rate

L Length of a membrane in the in-plane direction (cm)

l Length of a membrane in the through-plane direction (cm)

I Current (A)

i Current density (A cm-2)

j Imaginary unit

k Boltzmann constant (1.3806504×10-23 J K-1)

M Molarity (mol L-1)

n Degree of phase difference

Q Total electrical charge passed in a circuit (C)

Qo Modified capacitance (F)

 xiii

R Molar gas constant (8.314 J K-1 mol-1)

Rct Charge transfer impedance (Ω)

Rmem Resistance of a membrane (Ω)

Rsol Resistance of a solution (Ω)

Rtot Resistance through the all components (Ω)

T Thermodynamic temperature (K)

t Time (s)

V Applied voltage (V)

V0 Amplitude of AC voltage (V)

W Width of a membrane in the in-plane direction (cm)

Z Impedance (Ω)

Greek

σ Ionic conductivity (mS cm-1)

σ0 Pre-exponential factor

Abbreviations

AFC Alkaline Fuel Cell

PEMFC Proton Exchange Membrane Fuel Cell

PAFC Phosphoric Acid Fuel Cell

MCFC Molten Carbonate Fuel Cell

SOFC Solid Oxide Fuel Cell

FCV Fuel Cell Vehicle

IEM Ion Exchange Membrane

 xiv

IEC Ion Exchange Capacity

S-PPQ Sulfonated Polyphenylquinoxaline

HPA Heteroplyacid

GPTS Glycidoxypropyltri-methoxy

SPS Sulfonated PhenyltriethoxySilane

BMImCl Butyl-Methyl-Imidazolium Chloride

PWA Phosphotungstic Acid

SPEEK Sulfonated Polyether

PAMPS AcrylAmido-Methil-Propane Sulfonic acid

VP VinylPyradine

DABCO Diazabicyclo-Octane

SEM Scanning Electron Microscopy

XRD X-ray Diffraction

RH Relative Humidity

PID Proportional Integral Derivative

V.P. Vapor Pressure

EIS Electrochemical Impedance Spectroscopy

CPE Constant Phase Element

DOE Department of Energy

LDH Layered Double Hydroxides

 xv

ACKNOWLEDGEMENT

I finished this thesis with countless technical and intellectual supports. I would like to

show my gratitude and respect to the individuals who contributed to my work.

Thanks to my thesis advisor, Dr. Serguei N. Lvov for letting me join his laboratory and

supervise my lab works and paper writing. Even though my background in electrochemistry was

not enough and my work was very slow, he patiently gave me great advises that helped me learn

new field on my own. This experience will definitely help me with my future work in both

scientific and military work.

I am grateful to my thesis co-advisor, Dr. Sridhar Komarneni for giving me many

instructive advises and telling me how to prepare the samples, grammatical details in my papers,

and how to find a good reference in the material science field.

Thanks to my committee members, Dr. Derek Elsworth and Dr. Jeffrey R.S. Brownson.

They received my offer to be my committee members with good grace, and gave inspiring

instructions and questions on my thesis work. In tackling these questions, my skill as a

researcher was improved greatly.

Special thanks to Dr. Mark V. Fedkin and Dr. Elena Chalkova. They directly supported

my experiments in the lab, including the preparation for the samples and conductivity

measurements.

Thanks to my colleagues, Chunmei Wang, Justin R. Beck, Mark S. LaBarbera, Richard S.

Schatz, Matthew A. Brown. Delightful conversation with them always encouraged me to conduct
 xvi

my experiment and improved my English skills. Special thanks are to Chunmei Wang who

supported my experimental activity and give me multiple suggestions on my work. Justin Beck

also helped me do proof-reading and many clues for my study and communication skills.

I also wish to thank all my friends who cheered my life in State College, PA. Although it

was sometimes too tough for me to manage to my significant amount of assignments, they

facilitated those works as catalysts.

Finally, thanks to my supervisors in Japan, Col. Inoue and Maj. Ueno. They got in touch

with me closely and dealt with my student life financially and mentally. Financial support for my

student life was provided by the Japanese Ministry of Defense.

 1

Chapter 1 Introduction

1.1 Fuel cells

A fuel cell is an energy conversion system that consumes hydrogen and oxygen in order

to produce electricity. Other fuels can also be consumed for some kinds of fuel cells. The cell has

the half-cell chemical reactions occur on the electrode surfaces. The reduction reaction occurs at

the cathode and the oxidation reaction proceeds at the anode. A simple schematic of the concept

for different kinds of fuel cells is shown in Figure 1-1.

Figure 1-1: The concept of fuel cells[1,2].

AFC, PEMFC, PAFC, MCFC, and SOFC stand for alkaline fuel cell, proton exchange membrane

(polymer electrolyte membrane) fuel cell, phosphoric acid fuel cell, molten carbonate fuel cell,

and solid oxide fuel cell, respectively. The key factors that decide the reaction on the electrode
 2

surfaces are electrodes, electrolytes, and the operating conditions. The half-cell chemical

reactions vary depending on these factors. Since the fuel cell does not work as a heat engine, it

avoids the efficiency limit of the Carnot cycle and shows higher efficiency [3].

PEMFCs and AFCs operate at relatively lower temperature and can potentially use a

solid state ion exchange membrane as the electrolyte, which would help these fuel cells to

increase their compactness, mechanical strength, and the range of applications.

1.1.1 PEM fuel cells

PEMFCs have a wide range of applications, such as the power source for fuel cell

vehicles (FCVs), stationary power plants, and energy storage [1]. The half-cell chemical

reactions in hydrogen PEMFCs are

1
Cathode: O + 2H + + 2e− → H2 O
2 2

Anode: H2 → 2H + + 2e−

Considerable studies have been carried out in order to improve the performance of PEMFCs.

One of the effective ways to examine the performance of a fuel cell is to observe the potential

difference and the current. The plot of this potential difference and current is called “polarization

curve” [4]. The curve typically shows a shape such as in Figure 1-2 below.
 3

Figure 1-2: Typical polarization curve for a PEMFC [4]

This curve has three well defined regions, and these are called, “activation polarization losses”,

“ohmic losses”, and “concentration polarization losses,” respectively. The activation polarization

losses are related to the rate of reactions on the electrode surfaces. According to Tafel equation:

𝑉𝑉 = 𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵|𝑖𝑖| [1-1]

where V, B, and i are the potential difference, Tafel slope, and current density, respectively. This

equation shows the behavior when we assume that only the activation polarization losses exists.

The reaction rate at the electrode is determined by the chemical reaction, and the reaction rate at

cathode is much lower than that at anode in a PEMFC. In order to increase the reaction rate at

cathode and decrease the loss, considerable studies on the electrode catalysts have been

conducted. Although platinum and platinum-ruthenium alloy have been found to be the most

effective catalysts, studies on new materials including tungsten, tin, and molybdenum have been

reported for more reasonable production costs [5]. The concentration polarization losses are

caused by the lack of reactants on the electrode surfaces. For these losses modeling and new flow
 4

channel patterns to prevent the porous electrodes from flooding with water have been reported

[4,6]. The ohmic losses are dominantly caused by the resistance of the PEM. In order to reduce

its resistance, many studies for exploring new materials with high ionic conductivity have been

made.

1.1.2. Alkaline fuel cells

Alkaline fuel cells (AFCs) were developed for the space shuttles because they were

lighter than any other practical power generation systems that could be mounted in the limited

space in the shuttles. These AFCs produced water and electricity from pure oxygen and

hydrogen from the shuttle tanks [7]. The half-cell chemical reactions are as follows:

1
Cathode: O + H2 O + 2e− → OH −
2 2

Anode: H2 + 2OH − → 2H2 O + 2e−

The reaction rate of the reduction reaction at the cathode is much faster than that in PEMFC, and

the cathode does not need a noble metal electrocatalyst. However, in usage on the earth they

showed poor performance and durability. One of the dominant reasons is that the CO2 in the air

reacts with the electrolyte of AFCs, potassium hydroxide (KOH), resulting in the deposition of

potassium carbonate (K2CO3), which causes a decrease of the cell performance [7]. The OH-

conductive solid electrolyte has gathered much attention in order to overcome this problem and

use AFC with air rather that with pure oxygen.

 5

1.2 Ion exchange membranes

1.2.1 Basis of ion exchange membranes

An ion exchange membrane (IEM) is a membrane that allows the permeation of specific

ions through various mechanisms and prevents permeation of other ions and molecules. The

permeable ions are called “counter-ions” because the membrane itself is charged oppositely by

ions fixed inside the membrane. The ions in the electrolyte charged oppositely to counter-ions

are called “co-ions”. For example, when we use a Cl- exchange membrane in a NaCl aqueous

solution the counter-ion is Cl- and the co-ion is Na+. There are many IEMs with varying

properties, such as conductivity, permselectivity between like-charged ions, transport number,

ion exchange capacity (IEC), water uptake, mechanical strength, and so on. Taking these

parameters into consideration, the membrane should be chosen properly to suit the purpose of the

systems. For hydrogen PEMFCs, the IEM is used to provide fuel, in the form of protons, from

anode to cathode, and conductivity is important in order to avoid ohmic losses. In the case of an

electrolyzer that has many ionic species the membrane has to have a good permselectivity in

order to maintain its performance by keeping undesirable ions from permeating through the

membrane [8].

1.2.2 Development of IEMs for PEMFCs and AFCs

Many studies have been carried out in order to develop highly conductive ionic

membranes under various operating conditions. Nafion® 117 has a high conductivity of 120 mS
 6

cm-1 at 120 oC in the low pressured gas phase and 90 mS cm-1 at 25 oC in deionized water [9,10].

It also shows tough mechanical strength that enables it to be used in commercial PEMFCs.

Nafion is made by E.I. du Pont de Nemours & Co. (Inc) and has been applied for a wide range of

electrochemical systems. It has ionomers with hydrophobic C-F chains and hydrophilic sulfonic

acid units (see Figure 1-3) [3].

(CF2－CF2)x (CF －CF2)y

O
CF2
FCーCF
O
CF2
CF2
SO3H

Figure 1-3: The structure of Nafion membrane.

Because of these properties, Nafion has been used in lower temperature conditions desirable for

PEMFCs. However, it cannot be used over 100 oC because heat treatment causes shrinkage of

the membrane and decrease of the water content [11]. Moreover, the morphology of the

membrane can be changed at high temperature [12]. In order to overcome this dilemma,

membranes with various unconventional materials are being developed, and the conductivities

are also being improved as shown in Table 1-1. In parallel with the improvement of organic

materials some inorganic materials and combinations of organic monomers and inorganic
 7

particles, which are called “composite membranes,” have been developed for PEMFCs as shown

in Table 1-2. We found that the reported conductivities at both low temperature and high

temperature have been improved year by year. These kinds of efforts are also being conducted

for AFCs as shown in Table 1-3.

 8

Table 1-1: The reported organic membrane examples and their ionic conductivities for PEMFCs.

Reported Materials Conductivity Temperature Relative Measurement Source

o
Year mS/cm C humidity (Frequency range)
1998 Nafion 28 25 81 % Gas in-plane [13]
Perfluorosulfonic acid 55 100 % (0.1 - 20 kHz)
1998 bis[(perfluo-roalkyl)sulfonyl]imide ionome 8.9 25 31 % Gas in-plane [13]
100 100 % (0.1 - 20 kHz)
2000 Aromatic based 9.8 20 80 % Not mentioned [14]
Sulfonated polyphenylquinoxaline (S-PPQ) clearly
2003 Heteropolyacid (HPA)/sulfonated poly(arylene 80 25 100 % Gas in-plane [15]
ether sulfone) 150 130 (10 Hz - 1 MHz)
2005 Sulfonated block copolyimides 350 80 100 % Gas in-plane [16]
(50 Hz - 50 kHz)

Table 1-2: The reported inorganic material examples and their ionic conductivities for PEMFC
Reported Materials Conductivity Temperature Relative Measurement Source
o
Year mS/cm C humidity (Frequency range)
2003 3-glycidoxypropyltri-methoxysilane(GPTS), 36 120 15 % Gas through-plane [17]
sulfonated phenyltriethoxysilane (SPS), (10m-5MHz)
tetraethoxysilane(TEOS)
2004 Poly (styrene-co-methacrylate)—silica covalent 0.8 25 30 Gas through-plane [18]
by copolymerization of monomers (10-10MHz)
(styrene and 2-hydroxyethyl methacrylate)
2005 Nafion doped with solid acidic inorganic material 1.5/14 90 10/40 % Gas through-plane [19]
(ZrO2, SO4/ZrO2, ZrOH, ZrO2 sol-gel) 1.8/18.5 120 (10m-1MHz)
2006 1-butyl-3-methyl-imidazolium chloride (BMImCl) 2 30 96 % Gas through-plane [20]
and 12-phosphotungstic acid (PWA)
2006 Sulfonated polyether ether ketone(SPEEK) 7.5 30 100 % Gas through-plane [21]
/WO3·2H2O 19 50 (5-5MHz)
2009 Nafion–mesoporous zirconium phosphate 11 30 100 % Gas through-plane [22]
(10-1MHz)
2009 2-acrylamido-2-methyl-1-propane sulfonic acid ~2 30 80 % Gas through-plane [23]
(PAMPS) in diphenylsiloxane-silica 10 80 (1-10MHz)
(Ph2SiO–SiO2)
 9

Table 1-3: The reported membrane examples and their ionic conductivities for AFCs

Reported Materials Conductivity Temperature Relative Measurement Sour

o
Year mS/cm C humidity (Frequency range) ce
2006 4-vinylpyradine (4-VP) and styrene 8 25 Liquid Liquid [24]
(Initiator: 2,2’-azobisisobutyronitrile) (Methanol) through-plane
2007 1,4-diazabicyclo-[2.2.2]-octane (DABCO) and 1-azabicyclo- 2.5 20 98 % Gas through-plane [25]
[2.2.2]-octane (Quinuclidine) on poly(epichlorydrin-allyl 13 60 (5 Hz–13 MHz)
glycidyl ether) copolymer
2010 Random copolymer of poly(methyl methacrylate-co-butyl 8.2 80 60 % Gas in-plane [26]
acrylate-co-vinylbenzyl chloride)
2010 Silica/poly(2,6-dimethyl-1,4-phenylene oxide) 12 30 100 % Gas in-plane [27]
35 90
(2006) Potassium chloride solution (KOH) 200 25 - (Conductivity meter [28]
with 50 kHz)
 10

1.3 Problem statement and methodology

We found that the in-plane measurements with 4 electrodes are mostly used for the ionic

conductivity measurements of organic materials shown in Table 1-1. We can measure the ionic

conductivity for the in-plane direction with this method. However, the protons are conducted

from the anode to the cathode in the operating conditions of PEMFCs. This direction is through-

plane. In order to evaluate the ionic conductivities of PEMs more accurately, we need to conduct

a through-plane measurement.

Furthermore, in tables 1-1, 1-2, and 1-3 we found that the conductivity measurement

methods used in these reports vary depending on the materials. We need to evaluate these

measurement methods in order to compare these results. Thus, we measured the Nafion sample

with the two measurement methods, the in-plane and the through-plane in the gas phase. As for

the liquid phase measurements the conductivity was measured with a through-plane method.

Then the result was compared to the reference data measured with the in-plane method.

In addition to these studies, highly OH- conductive materials are in demand in order to

increase the utility and applications of AFCs. The materials reported so far have shown relatively

lower conductivities than those for PEMFCs. In addition, these organic materials are not stable in

higher temperature over 80 oC. Therefore, we have synthesized new inorganic materials. These

materials were characterized with scanning electron microscopy (SEM) and X-ray diffraction

(XRD), and their conductivities were examined with the through-plane conductivity

measurements as well.
 11

Chapter 2 Theory and experimental technique

In this chapter, we will examine the membrane and ion exchange mechanism first. Then

we will describe the conductivity cells that are used in each of the three ionic conductivity

measurements. Finally, the impedance measurement technique will be explained.

2.1 Ion conducting mechanisms

2.1.1 Ion exchange mechanism by thermodynamics

We first look at the interface example between a cation exchange membrane and

electrolyte solution as shown in Figure 2-1.

a1 a1 a 2 a2

- +
+
+ + + -
-
- - +
+
+ -
-
Electrolyte solution Membrane Electrolyte solution
(Activity: a1) (Activity: a2)

Figure 2-1: The interface between the solution and the membrane.
 12

a1, a1, a2, and a2 are the activities of permeable cations in the electrolyte1, in the membrane at

the electrolyte 1 side, in the electrolyte2, and in the membrane at the electrolyte 2 side,

respectively. The membrane has anions fixed inside of it. This makes its electrochemical

potential different from the bulk solution. This potential difference is termed the Donnan

potential [29]. It can be written as

𝑅𝑅𝑅𝑅 ���1�
𝑎𝑎
𝐸𝐸𝐷𝐷𝐷𝐷𝐷𝐷 = 𝑙𝑙𝑙𝑙 [2-1]
𝐹𝐹 𝑎𝑎 1

On the other side of the membrane the Donnan potential is

𝑅𝑅𝑅𝑅 𝑎𝑎
𝐸𝐸𝐷𝐷𝐷𝐷𝐷𝐷 ′ = 𝑙𝑙𝑙𝑙 �2 [2-2]
𝐹𝐹 𝑎𝑎 2

Also, there occurs the diffusion of ions inside the membrane. The diffusion potential is

𝑅𝑅𝑅𝑅 2𝑎𝑎� 𝑖𝑖 𝑡𝑡 ̅
𝐸𝐸𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 =
𝐹𝐹
∫𝑎𝑎� ∑ 𝑧𝑧 𝑑𝑑 𝑙𝑙𝑙𝑙𝑎𝑎�𝑖𝑖 [2-3]
1 𝑖𝑖

Thus, its total membrane potential is

𝐸𝐸 = 𝐸𝐸𝐷𝐷𝐷𝐷𝐷𝐷 + 𝐸𝐸𝐷𝐷𝐷𝐷𝑛𝑛 ′ + 𝐸𝐸𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 [2-4]

𝑅𝑅𝑅𝑅 ���
𝑎𝑎 1� 𝑅𝑅𝑅𝑅 𝑎𝑎 𝑅𝑅𝑅𝑅 𝑎𝑎�
2 𝑡𝑡 ̅
𝐸𝐸 = 𝑙𝑙𝑙𝑙 + 𝑙𝑙𝑙𝑙 �2 + 𝑖𝑖
∫𝑎𝑎� ∑ 𝑧𝑧 𝑑𝑑 𝑙𝑙𝑙𝑙𝑎𝑎�𝑖𝑖 [2-5]
𝐹𝐹 𝑎𝑎 1 𝐹𝐹 𝑎𝑎 2 𝐹𝐹 1 𝑖𝑖

2.1.2 Ion exchange mechanism in Nafion membrane

As we saw in the Section 1.2.2 and Figure 1-3, Nafion has ionomers inside it. These

ionomers are supported Teflon that has high mechanical strength and chemical stability. Its

structure model is shown in Figure 2-2.

 13

Perfluorosulfonic chains

Teflon

Figure 2-2: The structure model of Nafion membrane. [30]

These hydrophilic units retain moisture inside the membrane and pass protons around. Thus, we

can assume the proton conductivity of Nafion will increase when the humidity around the

membrane increases in the gas phase. The temperature also affects its conductivity [31]. When

the temperature increases, the conductivity of protons increases exponentially following the

equation

−𝐸𝐸𝑎𝑎
𝜎𝜎𝜎𝜎 = 𝜎𝜎0 𝑒𝑒𝑒𝑒𝑒𝑒 � � [2-6]
𝑘𝑘𝑘𝑘

where σ, σ0, and k are the ionic conductivity and the pre-exponential factor, and the Boltzmann

constant, respectively. We see that the ionic conductivity increases as the temperature increases

exponentially.
 14

2.2 Conductivity measurements

2.2.1 Conductivity cells

From the discussion above, we now recognize that we need to control temperature,

pressure, and humidity to evaluate the ionic conductivity of an IEM in the gas phase. For the

measurement in liquid phase the glass cell is appropriate in order to prevent its corrosion and

leakage of the electrolyte solutions. In our experiment, we used a glass conductivity cell as

shown in Figure 3-3.

Figure 2-3: The conductivity cell in liquid phase.

The sample membrane is held between two glass chambers and fixed with a clip. Then the

chambers are filled with sample solution. The membrane is in contact with solution by the cross

section area A, and its through-plane length is l as shown in Figure 2-4.

 15

Figure 2-4: The cross section of the cell.

For the gas phase measurement we used a Bekktech conductivity cell (BT-112, Figure 2-

5) made of metal. It can flow inert nitrogen gas inside the cell so that we can avoid any

undesirable reactions. It is equipped with two platinum disk electrodes that sandwich the sample

membrane.

Figure 2-5: The conductivity cell in gas phase.

 16

The membrane sample was cut into a circle and the Teflon sample holder fixes the sample and

electrodes. The length of the sample is l and the cross section is A as shown in Figure 2-6 below.

Figure 2-6: The cross section and the width of the membrane.

For the in-plane gas phase measurement, we changed this 2-electrode set into the 4-electrode one

as shown in Figure 2-7 below.

Figure 2-7: The sample and electrodes of in-plane method.

 17

L, H, W are the distance between two reference electrodes, the height, and the width of the

sample, respectively. The sample should be cut into a rectangle. The outside working and

counter electrode apply DC voltage and a potential gradient occurs across the sample. At the

same time, the reference electrodes measure the potential difference of the sample, sensing the

flow of protons across the surface of the sample.

For the gas phase measurement the cell is connected to a BT-104 saturator by Bekktech and

a BT-301 pressurized deionized water system. With the saturator, we can control the relative

humidity (RH) which is expressed in a percentage. RH is a simple indicator of humidity, the ratio

of humidifier’s saturated vapor pressure to that in the conductivity cell.

(Saturated vapor pressure of the humidifier )

RH% = × 100 [2-7]
(Saturated vapor pressure of the conductivity cell )

For example, when the temperature of the conductivity cell is 80 Co (vapor pressure, V.P. =

0.0467 MPa) and that of the humidifier is 60 Co (V.P.=0.020), RH = (0.020/0.0467)×100 = 42 %

[11]. The cell also has thermocouples to detect and control the temperature inside. PID

(proportional-integral-derivative) controllers attached to the thermocouples are used to keep the

temperature constant in each component, the conductivity cell, the humidifier, and the inlet. The

temperature of the inlet connecting the humidifier to the cell has to be higher than others set

point temperatures. Otherwise, there occurs vapor condensation inside the inlet, resulting in an

incorrect RH value.
 18

2.2.2 Calculation for each measurement

In the liquid phase through-plane measurement the membrane’s conductivity can be

calculated by the equation,

𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 = 𝑅𝑅𝑡𝑡𝑡𝑡𝑡𝑡 − 𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠 [2-8]

where Rmem, Rtot,and Rsol represent the resistance of the membrane, total resistance, and the

resistance of solution, respectively. Then we calculate its conductivity by the equation,

1 𝑙𝑙
𝜎𝜎 = ∙ [2-9]
𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 𝐴𝐴

where σ, l, and A are the conductivity of the membrane, the length, and the area of the

membrane. This equation is also applicable to the gas phase through-plane measurement because

there is no liquid between electrodes and the membrane. For these through-plane measurements,

the resistance of the cell and membrane are measured by electrochemical impedance

spectroscopy (EIS). In the next section, the concept of EIS and its process will be described.

In gas phase in-plane measurement the length and area of the membrane is different from

those of through-plane as shown in Fig.2-7. In equation 2-9, l changes to L, and A is W・T.

Now the equation for conductivity is,

1 𝐿𝐿
𝜎𝜎 = ∙ [2-10]
𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 𝑊𝑊×𝑇𝑇

For the in-plane measurement, it is not necessary to apply EIS with AC voltage. When we apply

DC voltage to the cell, it returns linear current, or an ohmic response. Its resistance can be

obtained from the slope of the i-V plot.

 19

2.3 Electrochemical impedance spectroscopy

2.3.1 Modeling for equivalent circuit

Every electrochemical system has electrical double layers because there are always phase

boundaries between interfaces of electrodes and electrolytes (see Figure 2-8). This double layer

has two opposite electric charges between the boundaries which part two phases. This double

layer can be regarded as a capacitor with a certain amount of capacitance when we apply low

(lower than 10 mV) AC voltage, so we can assume that the system is linear.

Interface Interface

+ - Anions Cations
+ -
+ - + -
+ - + -
+ - + -
+ - + -

Anode Electrolyte Cathode

Figure 2-8: The interface concept between the electrodes and the electrolyte.

Under this consideration, we can assume an electrical circuit to be equivalent to an

electrochemical system. For example, the boundary of Figure 2-8 is equivalent to the elements in

Figure 2-9 (a), or more simply expressed as (b).

 20

Rct Rct Rct

Rsol Rsol

Cdl
= Cdl
Cdl

(a) (b)

(a) Two interfaces (b) Simplified interface

Figure 2-9: The equivalent circuit.

Rsol, Rct, and Cdl are the resistance of the solution, the charge transfer resistance, and the double

layer capacitance. The charge transfer resistance stands for difficulty of charge transfer on the

surface of electrode which is unique to electrode materials and electrolytes. The double layer

capacitance is the capacitance at the electrode surface as shown in Fig 2-8. When we have this

equivalent circuit, we can easily evaluate it with electrochemical impedance spectroscopy (EIS),

identifying what kind of factors (capacitors, resistance, reactance) are in the system. Then we can

predict the behavior of some materials.

2.3.2 Basic theory of electrochemical impedance spectroscopy

Assuming that the capacitance of the double layer is Cdl, when we apply an AC voltage,

V, to the capacitor,

V = V0 exp(j2πft) [2-11]

where j, f, and t are imaginary unit, frequency of the voltage, and time by second. The current, i,

can be written as

𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
𝑖𝑖 = = 𝐶𝐶𝑑𝑑𝑑𝑑 = (𝑗𝑗2𝜋𝜋𝑓𝑓𝐶𝐶𝑑𝑑𝑑𝑑 )𝐸𝐸0 𝑒𝑒𝑒𝑒𝑒𝑒(𝑗𝑗2𝜋𝜋𝑓𝑓𝑡𝑡) = (𝑗𝑗2𝜋𝜋𝑓𝑓𝐶𝐶𝑑𝑑𝑑𝑑 )𝑉𝑉 [2-12]
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
 21

where Q stands for the total electrical charge passed. Now the impedance of the double layer, Z,

is,

𝑉𝑉 1
𝑍𝑍 = = [2-13]
𝐼𝐼 𝑗𝑗 (2𝜋𝜋𝑓𝑓𝐶𝐶𝑑𝑑𝑑𝑑 )

Thus, we can see that the impedance phases of an electrochemical system with a double layer

will change when we change the frequency. For a simple circuit as in Figure 2-9 the impedance

is

𝑅𝑅𝑐𝑐𝑐𝑐 𝑅𝑅𝑐𝑐𝑐𝑐 2𝜋𝜋𝜋𝜋 𝑅𝑅 2 𝐶𝐶𝑑𝑑𝑑𝑑

𝑐𝑐𝑐𝑐
𝑍𝑍 = 𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠 + = �𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠 + ) 2 � − 𝑗𝑗 2
[2-14]
1+𝑗𝑗 (2𝜋𝜋𝑓𝑓𝑅𝑅𝑐𝑐𝑐𝑐 𝐶𝐶𝑑𝑑𝑑𝑑 ) 1+(2𝜋𝜋𝜋𝜋 𝑅𝑅𝑐𝑐𝑐𝑐 𝐶𝐶𝑑𝑑𝑑𝑑 1+(2𝜋𝜋𝜋𝜋 𝑅𝑅 𝑐𝑐𝑐𝑐 𝐶𝐶𝑑𝑑𝑑𝑑 )

From this equation, we see the impedance phase (the imaginary part) changes as the frequency

changes. Especially, when the frequency increases toward infinite (f → ∞), the impedance

approximates the solution resistance (Rsol). When the frequency increases toward zero (f → 0),

the impedance approximates the sum of the solution resistant and the charge transfer resistance

(Z → Rsol + Rct). Also, the imaginary part reaches its maximum when

1
2𝜋𝜋𝜋𝜋 = [2-15]
𝑅𝑅𝑐𝑐𝑐𝑐 𝐶𝐶𝑑𝑑𝑑𝑑

From these results, we can write its impedance plot as shown in Figure 2-10.

(a) The Nyquist plot (b) Bode plot

Figure 2-10: Typical impedance plots.

 22

The impedance plot as (a) is called the Nyquist plot, and (b) is called the Bode plot. When we

obtain the Nyquist plot and the Bode plot of an electrochemical system we can compare these

plots with those modeled with an equivalent circuit. Then we can identify each element in the

actual electrochemical system and even predict the conducting mechanism of the measured

sample. The frequency range should be adjusted depending on the behavior of the material. The

applied voltage should be low (10 mV – 100 mV) in order not to have any undesirable chemical

reactions.

In the actual model fitting, we can use the constant phase element (CPE) that makes the

modeled Nyquist plot circle twisted so that the modeling fits the results of experiments. This

element corresponds with the surface roughness of the electrodes, a distribution of reaction rates,

varying thickness or composition of the membrane, and non-uniform current distribution [31-36].

The impedance of the CPE can be written as

1
𝑍𝑍𝐶𝐶𝐶𝐶𝐶𝐶 =
𝑄𝑄 𝑜𝑜 (𝑗𝑗2𝜋𝜋𝑓𝑓)𝑛𝑛

where Qo and n represents the modified capacitance and the degree of phase difference. When Qo

is C and n is unit, ZCPE is equal to the impedance of a capacitor with its capacitance of C.
 23

Chapter 3 Measurement for PEMs

3.1 In-plane measurement in gas phase

3.1.1 Experimental setup

We used an EIS system produced by Gamry Instruments in this experiment that has a

power supply and a frequency response analyzer. With this system, we can apply DC and AC

voltage and obtain the Nyquist plot under an arbitrary frequency range within 1.0 mHz – 300

kHz. The system has an inert nitrogen gas cylinder, humidity and PID temperature controls in

addition to the conductivity cell and EIS Gamry system used in the first measurement method in

the liquid phase (See Figure.3-1). Nitrogen gas is first injected into the humidifier and then to the

cell in order to prevent undesirable electrochemical reactions. The flow and pressure of nitrogen

gas is controlled to 18.3 psi and around 350 cm3 min-1 by a flow controller, flow meter, and two

gate valves.

Pressure gauge
BT-301 Pressurized
Deionized water system Thermocouples

V-3

Flow Valve
meter Water cooler
Water trap
Valve

Conductivity cell

Nitrogen cylinder BT-104 Bekktech Beaker

saturator

Figure 3-1: Gas phase measurement setup.

 24

The measurement was conducted in a procedure standardized by DOE and the Bekktech

Company to get conductivity under certain conditions of RH and temperature [35]. We used

25oC, 30 oC, 80 oC, and 120 oC as the temperature of the cell and 20 %RH -100 %RH as the

humidity. The process is as follows in Table 3-1.

Table 3-1: The experimental procedure

Time (min.) Relative humidity Saturator temperature for Saturator temperature

the 120 oC experiment for the 80oC experiment
o o
C C
0* 70 % 108.9 71.4
15 60 % 104.4 68.0
30 50 % 99.2 63.8
45 40 % 93.0 59.0
60 30 % 85.4 53.0
75 25 % 81.1 49.2
90 20 % 75.2 44.8
105 25 % 81.1 49.2
120 30 % 85.4 53.0
135 40 % 93.0 59.0
150 50 % 99.2 63.8
165 60 % 104.4 68.0
180 70 % 108.9 71.4
195 80 % 113.0 74.5
210 90 % 116.7 77.5
225 95 % 118.4 78.7
240 100 % 120.0 80.0

* The sample was calibrated at 70 %RH for 2 hours before measurements

As for the measurement in low temperatures (30 oC), the RH was kept at 100 % because

humidity controlling is difficult and may cause significant errors in the results.
 25

The cell is equipped with four linear shaped electrodes, and the sample was cut to a

rectangle of 2.0 × 0.2 cm. The sample was put on electrodes and fixed with the holder. The cell

was connected to the Gamry system, the saturator, and the gas flow line. Then its ionic

conductivity was measured with the procedure in the previous section except for the usage of DC

voltage.

3.1.2 Results

The voltage-current plot (V-i plot) is shown in Figure 3-2. From the slope of this plot we

obtained the Rmem value and calculate the conductivity with equation 2-10. Ohmic responses and

conductivity data are shown in Figure 3-3, and Table 3-2 below.

30% Voltage V 30% Voltage V

50% 0.15 50% 0.15
80% 80%
90% 0.1 90% 0.1
100% 100%
0.05 0.05
Current A Current A
0 0
-0.0002 -3E-18 0.0002 -0.0002 -1E-18 0.0002
-0.05 -0.05

-0.1 -0.1

-0.15 -0.15

(a) Ohmic response at 80 oC (b) Ohmic response at 120 oC.

Figure 3-2: Ohmic responses in the in-plane gas phase measurement.

 26

80 oC
1.0E+02
Conductivity σ mS/cm

1.0E+01

Measured
Reference

1.0E+00
0 20 40 60 80 100
Relative humidity %

(b) Conductivity at 80 oC.

120 oC
Conductivity σ mS/cm

1.0E+02

1.0E+01

Measured
Reference
1.0E+00
0 20 40 60 80 100
Relative humidity %
(b) Conductivity at 120 oC.

Figure 3-3: Conductivity in the in-plane gas phase measurement.

 27

Table 3-2: Conductivity data in the in-plane gas phase measurement [37].

Relative Conductivity Reference data Conductivity Reference data

humidity at 80 oC mS cm-1 at 80 oC mS cm-1 at 120 oC mS cm-1 at 120 oC mS cm-1
cm-1
70 % 46.13 52 62.06 66
60 % 42.45 49 41.46 48
50 % 39.23 29 27.96 31
40 % 20.98 19 18.84 21
30 % 15.79 12 9.439 12
25 % 9.435 7.5 7.080 7.1
20 % 4.542 4.9 4.333 4.1
25 % 6.145 7.2 6.206 6.2
30 % 8.226 8.5 8.454 8.9
40 % 10.81 15 17.89 18
50 % 19.63 23 26.65 29
60 % 22.83 32 41.95 47
70 % 42.66 45 61.33 65
80 % 50.71 62 79.74 89
90 % 58.37 89 121.7 140
95 % 72.07 110 186.6 175
100 % 105.1 130 214.8 200
Conductivity Reference data
at 30 oC mS cm-1 at 30 oC mS cm-1
1
100 % 81.73 82

The obtained conductivities are very close to the reference data provided by Bekktech. This

means that the system worked properly and has no problem with the cell and the sample. The

data at 80 oC deviate slightly from the reference data. The conductivities are increasing as the

RH decreases, and vice versa. The quality of data could be improved by equilibrating for more

time at each temperature so that the sample absorbs water molecules.

 28

3.2 Through-plane measurement in gas phase

3.2.1 Experimental setup

Through-plane measurement uses the same system as Fig.3-1 except for the conductivity

cell and its electrodes. As for the sample, Ag paste was put on the membrane and dried out for 24

hours in order to ensure the electric connection between the sample and the electrodes. Then the

membrane was cut into a disk (d = 1.6 cm) and sandwiched between the electrode plates and held

in a Teflon bracket and the exterior of the conductivity cell. The cell was connected to the

nitrogen cylinder, and its leakage was checked with an observation of the flow meter and the

pressure gauge. Then the temperatures of the cell, the inlet, and the humidifier were increased up

to experimental value gradually. The ionic conductivities were measured, controlling the RH

inside the cell following the procedure in Table 3-1.

3.2.2 Results

Each impedance measurement was taken three times at each temperature to get the

average value, which is shown in Figure 3-4. From this plot, we read the Rmem value by fitting the

curve with the model as shown in Figure 3-5 and calculate conductivity with equation 2-9. The

obtained conductivity data are shown in Figure 3-6 and Table 3-3 below.
 29

3000

2500
30%

2000
40%
Im{Z} Ω

1500
50%

1000
60%

500 70%

0 80%
0 1000 2000 3000 4000 5000 6000

Re {Z} Ω

(a) Nyquist plot at 80 oC.

140

30%
120

100 40%
Im{Z} Ω

80
50%
60
60%
40

70%
20

0 80%
0 50 100 150
Re {Z} Ω

(b) Nyquist plot at 120 oC.

Figure 3-4: The Nyquist plot in the through-plane gas phase measurement.
 30

Rct
Rsol

Cdl CPE

Figure 3-5: The equivalent circuit model for the measurement.

1.0E+03
80 oC
1.0E+02
Conductivity σ mS/cm

1.0E+01

1.0E+00

1.0E-01 Through-plane
In-plane
1.0E-02
0 20 40 60 80 100 120
Relative humidity %

(a) Conductivity at 80 oC.

 31

1.0E+03
Through-plane 120 oC

Conductivity σ mS/cm
In-plane
1.0E+02

1.0E+01

1.0E+00
0 20 40 60 80 100 120

1.0E-01
Relative humidity %

(b) Conductivity at 120 oC.

Figure 3-6: Conductivity in the through-plane gas phase measurement.

 32

Table 3-3: Conductivity data in the through-plane gas phase measurement.

Relative humidity Conductivity at 80 oC Conductivity at 120 oC

mS cm-1 mS cm-1
70 % 0.5019 3.606
60 % 0.4232 3.180
50 % 0.3270 2.973
40 % 0.2556 2.074
30 % 0.1880 1.541
25 % 0.1528 1.160
20 % 0.1362 0.9378
25 % 0.1461 1.003
30 % 0.1602 1.001
40 % 0.2001 1.086
50 % 0.1929 1.424
60 % 0.2345 1.952
70 % 0.2909 2.228
80 % 0.3186 3.346
90 % 0.3952 8.449
95 % 0.5387 18.04
100 % 0.8911 25.61

The results both at 80 oC and 120 oC are extremely low compared with the in-plane data,

although the slopes are almost the same, especially at 120 oC. However, the isotropic property of

Nafion has reported by Silva. et al [38]. I theory if the methods are equivalent, the measurement

results should be similar. Therefore, we can assume that the measurement procedure have some

problems. There are three clear challenges in the measurement. First, the electrical contact at the

interface between the electrodes and the membrane was not good. The silver paste between the

membrane and the electrodes might be lost somewhat because the paste was not absorbed in the

membrane. This occurred the interface resistance, and it made the membrane resistance increased.

Secondly, the membrane might not get reached equilibrium with the RH in the cell. For the in-
 33

plane measurement, it is easier for the membrane to get equilibrated because the membrane

expose its surface. However, for the through-plane measurement, the surface of the membrane is

covered with the electrodes. The membrane needs more time to have water molecules inside it.

Finally, the extrapolating the membrane resistance from the Nyquist plot for through-plane

measurement could have larger inaccuracy than reading the ohmic slope for in-plane

measurement.

3.3 Through-plane measurement in liquid phase

3.3.1 Experimental setup

The setup consists of conductivity cells for liquid solutions and a measurement EIS

system. The measurement was done under atmospheric pressure and ambient temperature, 25

Co, with a frequency of 100 Hz - 100 kHz. The Nafion 117 sample was cut into circle (its

diameter, d, was about 15 mm) so that its diameter was larger than that of the conductivity cell.

Then it was immersed in sample solutions for 24 hours before the measurement. As for the

solutions, 2 mol L-1 (M) HCl was prepared. The conductivity cell was filled with the HCl

solution, and then titanium electrodes were put into the cell. The conductivity can be calculated

from its resistance with equation 2-9.

 34

3.3.2 Results

The Nyquist plot of HCl solution with the membrane and HCl solution were shown in

Figure 3-7. The total resistance and the resistance of the solution was obtained by the curve

fitting with the same model as the one in the gas phase (Figure 3-5). The membrane resistance

was calculated with equation 2-8, and then the conductivity was calculated with equation 2-9.

The measured values of l and A were 0.0178 cm and 0.865 cm2, respectively.

8.00E+00

2M HCl solution
6.00E+00 HCl + membrane
Im{Z} Ω

4.00E+00

2.00E+00

0.00E+00
0.00E+00 1.00E+00 2.00E+00 3.00E+00 4.00E+00
Re {Z} Ω

Figure 3-7: Nyquist plot in liquid phase measurement.

As a result of the calculation, the ionic conductivity in the 2 M HCl aqueous solution was 84 mS

cm-1 in the through-plane direction. The conductivity of nafion 117 was reported to be 76 mS

cm-1 with the in-plane measurement in the liquid phase in 1 M H2SO4 aqueous solution [39,40].

These results shows that the direction of the conductivity measurement does not affect the ionic

conductivity significantly.
 35

3.3.4 PEM measurement summary

First, we obtained the ionic conductivities with the in-plane gas phase measurement. The

obtained results were very close to the reference data. Thus, we confirmed that our system

worked properly. However, the results with through-plane gas phase measurement were

significantly lower than the result of the first in-plane measurement even though the

measurement in the liquid phase shows very close conductivities to the reference data. We

assume that our measurement method has some challenges to be systematically addressed . In

order to conduct the ionic conductivity measurement for thin-film membranes, we need to make

sure if the measurement system works properly.

 36

Chapter 4 Measurement for OH- conductive materials

4.1 Synthesis and characterizations of inorganic materials

4.1.1 Background on layered double hydroxides and hydroxyapatite

Two layered double hydroxides (LDHs) with Cl- anion in the interlayers were

investigated for chloride ion conductivity. LDHs consist of positively charged layers with two

different valences of metal cations and exchangeable hydrated anions such as Cl- in the

interlayers as shown in Figure 4-1. The LDHs are similar to the mineral hydrotalcite, [Mg6 Al2

(OH)16]2+CO32- • 4H2O [41], and hence they are also called hydrotalcite-like compounds. When

some of the divalent ions such as Mg2+ or Ca2+ are substituted by trivalent cations such as Al3+,

positive charge will be created on the layers. This positive charge on the layers needs to be

satisfied by anions with water molecules in the interlayers. LDHs can be represented by the

formula [42]:

[Mz+1-xM 3+ x (OH)2][Xn-x/n•yH2O].

Most commonly, z = 2, and M2+ is a divalent cation, Ca2+, Mg2+, etc., and M3+ is a trivalent

cation, Al3+, Cr3+, etc, and a variety of anions, X such as Cl-, NO3-, CO32- etc. and y is the

number of moles of coordinated water molecules per formula weight.

 37

One commercially available Mg-Al LDH with carbonate anions in the interlayers was

procured from Aluminum Company of America (Alcoa) located in Pittsburgh, PA. This material

has a Mg:Al ratio of 3:1.

Figure 4-1: Schematic representation of the layered double hydroxide structure [43].
 38

Hydroxyapatite, Ca5(PO4)3OH

In the apatite structure, Ca5(PO4)3X, the X-- ions (X = OH, F, or C1) form one-

dimensional chains (shown in Figure 4-2) parallel to the c-axis [44,45]. The X-X distance is

typically 3.44Å for OH- and F- in apatite. Calcium can be substituted by many ions such as Sr,

Ni, Zn, Pb etc. while phosphate could also be replaced by carbonate. Thus, the apatite structure

is highly amenable for OH conductivity.

Figure 4-2: Hydroxyapatite structure projected on x, y plane [44].

 39

4.1.2 Synthesis of sample materials

For the synthesis of 2Mg:Al and 2Ca:Al LDHs, magnesium chloride, calcium chloride,

aluminum chloride and sodium hydroxide were acquired from Aldrich chemical company.

Chloride containing Mg-Al LDH (JK-3)

The chloride form of LDH with magnesium and aluminum (Mg-Al LDH) was prepared

by coprecipitation of mixed Mg and Al chloride solution with a solution of NaOH. First, aqueous

solutions of magnesium chloride, MgCl2⋅6H2O (1.2M) and aluminum chloride, AlCl3⋅6H2O

(0.6M) were prepared to obtain a Mg/Al molar ratio of 2 using 50 ml of deionized water. An

aqueous solution of sodium hydroxide, NaOH (6M) was prepared separately in 50 ml of

deionized water. The mixed Mg-Al solution was then added drop by drop to the NaOH solution

while mixing at room temperature. This mixing led to precipitation of Mg-Al LDH phase. The

resulting precipitates were washed several times with deionized water and dried at 60 oC prior to

characterization by different techniques.

Chloride containing Ca-Al LDH (JK-4)

The chloride form of Ca-Al LDH (mineral name, hydrocalumite) was synthesized with

calcium and aluminum chlorides in a similar way to that of Mg-Al LDH. The chloride form of

Ca-Al LDH was made with calcium and aluminum chlorides after mixing them together in

sodium hydroxide solution. First, an aqueous solution of NaOH (2.5M) was prepared by

dissolving NaOH crystals in 50 ml of deionized water. Then, a 50 ml aqueous solution of Ca-Al

was made by dissolving CaCl2⋅2H2O (0.9M) and AlCl3⋅6H2O (0.45M) in a Ca/Al molar ratio of 2.

This 50 ml of mixed Ca-Al solution was slowly added drop by drop to 50 ml of sodium
 40

hydroxide solution under vigorous stirring at room temperature. This mixing led to precipitation

of Ca-Al LDH phase. The resulting precipitates were washed several times with deionized water

and dried at 60 oC prior to characterization by different techniques.

Hydroxide containing Mg:Al LDH (YN 252)

The commercial LDH sample that was obtained from Alcoa was calcined at 500oC for 4h

in a tubular furnace to decompose carbonate to carbon dioxide and amorphize the LDH material.

This amorphized LDH was then equilibrated with NH4OH at room temperature to reconstitute

the LDH structure with hydroxide ions in the interlayers as charge balancing anions. Thus, the

LDH containing OH anions was synthesized by the above new approach. The resulting LDH

material was washed with deionized water and dried at 60 oC prior to characterization by

different techniques.

Synthesis of hydroxyapatite (YN 148)

For the synthesis of Ca-hydroxyapatite, (NH4)2HPO4 and Ca(NO3)2 containing15.7 %

H2O (Alfa Aesar) were used as precursors. The molar ratios of Ca: P for Ca- hydroxyapatite

synthesis was 5:3. In a typical synthesis, appropriate amounts of Ca(NO3)2 was dissolved in 20

ml of deionized water and then 40 ml of 0.375 M (NH4)2HPO4 solution was added to the Ca

solution. After mixing, the pH was adjusted to 10 using NH4OH. The mixture was treated in an

autoclave at 50oC for 4.5 h. The resulting precipitates were washed several times with deionized

water and dried at 60 oC.

 41

4.1.3 Characterizations of sample materials

X-ray diffraction

Powder X-ray diffraction (XRD) patterns of synthesized LDHs and hydroxyapatite were

recorded with a Scintag diffractometer operated at 35 kV voltage and 30 mA current using CuKα

radiation. The X-ray diffraction patterns of each sample was recorded using a scanning rate of 5o

(2θ)/min and in the range of 5-45 degrees two theta. The X-ray peaks can be used to identify the

sample comparing it to a standard. XRD can tell us whether the samples are pure phases or

mixed with impurities. Powder XRD clearly showed that pure phases of all the LDHs and

hydroxyapatite were obtained.

Scanning Electron Microscopy

Some of the samples were characterized by scanning electron microscope (SEM) to

determine the particle size and morphology. Scanning electron microscopy was done using a

field emission scanning electron microscope (JSM-6700F, JEOL, Tokyo, Japan) on samples

coated with very thin carbon using a carbon coater. An accelerating voltage of 5 KV was used

for observation with the microscope. The materials were characterized with SEM and XRD. The

results are shown below in Figures 4-3 to 4-7.

 42

Figure 4-3: Scanning electron micrograph showing the particle morphology and size of Chloride containing

Mg-Al LDH (JK-3).

 43

Figure 4-4: Scanning electron micrographs showing the particle morphology and size of Chloride containing

Ca-Al LDH (JK-4) at two different magnifications.

 44

Figure 4-5: X-ray diffraction patterns of (a) chloride containing Mg-Al LDH (JK-3) and

(b) chloride containing Ca-Al LDH (JK-4).

Figure 4-6: X-ray diffraction pattern of hydroxide containing Mg:Al LDH (YN 252).
 45

2.81Å

Intensity
3.43Å 2.72Å

2.62Å
2.26Å
3.08Å

5.29Å 3.87Å

5 10 15 20 25 30 35 40
Degrees 2θ(Cukα)

Figure 4-7: X-ray diffraction pattern of hydroxyapatite (YN 148).

4.2 Conductivity measurement

4.2.1 Sample preparations and experimental setup

300 mg of inorganic particles were pressed into a 13 mm disk pellet with a pressure of

about 5 t cm-2. Then the pellet was annealed at 200 oC for 6 hours in a furnace heated and cooled

with the ramp rate of 1.5 oC min-1. The sample was then painted with Ag paste and dried for 24

hours in order to make sure that an electrical contact was achieved with electrodes of the

conductivity cells. The prepared samples were sandwiched and fixed with two platinum disk

electrodes of the through-plane conductivity cell in the same way of that of the measurements for

the PEMs. Then the cell was connected to the nitrogen cylinder as shown in Figure 3-1, and we
 46

flowed the gas into the cell and controlled the pressure and the temperature. With regard to the

relative humidity inside the cell, we kept the RH constant until we obtained stable data.

4.2.2 Results

We obtained the Nyquist plots of each material as shown in Figure 4-8 and their bulk

resistances were estimated with their inflexion point or intersection point with real axis which

can be thought as the bulk resistance. Then the conductivities were calculated with the equation

[2-9]. The conductivities for each material are shown in Figure 4-9.

2.00E+01
95 %RH JK-3
1.80E+01
1.60E+01 30 C
1.40E+01 50 C
1.20E+01
Im{Z} Ω

80 C
1.00E+01 100 C
8.00E+00 120 C
6.00E+00
4.00E+00
2.00E+00
0.00E+00
0.00E+00 1.00E+02 2.00E+02 3.00E+02 4.00E+02

Re {Z} Ω

(a) Nyquist plot of Mg-Al LDH (JK-3)

 47

4.0E+01
95 %RH JK-4
30 C
3.0E+01
50 C
Im{Z} Ω

80 C
2.0E+01 100 C
120 C
1.0E+01

0.0E+00
0.0E+00 5.0E+01 1.0E+02

Re {Z} Ω

(b) Nyquist plot of Ca-Al LDH (JK-4)

1.0E+02
95 %RH
YN 252
8.0E+01 30 C
50 C
6.0E+01
Im{Z} Ω

80 C
100 C
4.0E+01 120 C

2.0E+01

0.0E+00
0.0E+00 1.0E+02 2.0E+02

Re {Z} Ω

(c) Nyquist plot of Mg:Al LDH (YN 252)

 48

5.0E+02
95 %RH YN148
4.0E+02 30 C

50 C
3.0E+02
Im{Z} Ω

80 C
2.0E+02 120 C

1.0E+02

0.0E+00
0.0E+00 2.0E+02 4.0E+02 6.0E+02 8.0E+02

Re {Z} Ω

(d) Nyquist plot of hydroxyapatite (YN 148)

Figure 4-8: Nyquist plots of inorganic materials at 95 %RH.

1.0E+01
120 oC
1.0E+00
JK3
Conductivity σ S/cm

JK4
1.0E-01
YN252
YN148
1.0E-02

1.0E-03

1.0E-04
0 20 40 60 80 100
RH %

(a) OH- Conductivity at 120 oC.

 49

1.0E+01
95 %RH

Conductivity σ S/cm JK3

JK4
YN252
1.0E+00
YN148

1.0E-01
0 20 40 60 80 100 120
Temperature oC

(b) OH- Conductivity at 95 %RH.

Figure 4-9: OH- Conductivities in inorganic materials.

The Nyquist plot shows the temperature and humidity dependence of the conductivity of

materials. Mg:Al LDH (YN 252) keeps the conductivity larger than 0.01 mS cm-1 in the wide

range of the RHs. Ca-Al LDH (JK-4) also keeps the conductivity in low RHs. It also shows high

conductivity in low temperature ranges.

Comparing these conductivities with those of reported materials in Table 1-3, our

materials do not have comparable conductivities in the high temperature and high RH ranges.

However, they have the possibility to be used in AFCs that are operated at low temperature

(below 100 oC). Also, what we measured in this experiment were the conductivities of particles

instead of synthesized membranes with inorganic particles. In order to make more precise

evaluations, we need to examine the correlations between the conductivity of particles and that

of synthesized membranes.
 50

Chapter 5 Conclusion

We have applied our through-plane gas phase conductivity cell for the organic PEM,

making use of the conductivity measurement for the inorganic materials. As a result, the

membrane in the gas phase showed significantly lower conductivities although the measurement

in the liquid phase of the same membrane showed a proper result. Thus, we conclude that our

measurement has some problems in the procesure and the measurement in the through-plane

direction should further be developed. At the same time, we have examined newly synthesized

inorganic OH- conductive materials. They were characterized with SEM and XRD and their OH-

ionic conductivities were measured. Although the materials showed relatively lower

conductivities than other materials recently reported, their usages in AFCs are worth studying

because they showed stable conductivities in the low temperature and humidity ranges.
 51

BIBLIOGRAPHY

1. Barbir, F. PEM Fuel Cells: Theory and Practice; Academic Press, 2005.

2. Hydrogen and Our Energy Future, U.S. Department of Energy,

http://www1.eere.energy.gov/hydrogenandfuelcells/pdfs/hydrogenenergyfuture_web.pdf

3. Peter Atkins, J. d. P. Physical Chemistry; 8th ed.; Oxford University Press: New York, 2006.

4. Baschuk, J. J.; Li, X. H., Modelling of polymer electrolyte membrane fuel cells with variable

degrees of water flooding. Journal of Power Sources 2000, 86, (1-2), 181-196.

5. Gotz, M.; Wendt, H., Binary and ternary anode catalyst formulations including the elements

W, Sn and Mo for PEMFCs operated on methanol or reformate gas. Electrochimica Acta

1998, 43, (24), 3637-3644.

6. Su, A.; Chiu, Y. C.; Weng, F. B., The impact of flow field pattern on concentration and

performance in PEMFC. International Journal of Energy Research 2005, 29, (5), 409-425.

7. McLean, G. F.; Niet, T.; Prince-Richard, S.; Djilali, N., An assessment of alkaline fuel cell

technology. International Journal of Hydrogen Energy 2002, 27, (5), 507-526.

8. Sata, T.; Sata, T.; Yang, W. K., Studies on cation-exchange membranes having

permselectivity between cations in electrodialysis. Journal of Membrane Science 2002, 206,

(1-2), 31-60.

9. Casciola, M.; Alberti, G.; Sganappa, M.; Narducci, R., On the decay of Nafion proton

conductivity at high temperature and relative humidity. Journal of Power Sources 2006, 162,

(1), 141-145.
 52

10. Halseid, R.; Vie, P. J. S.; Tunold, R., Influence of ammonium on conductivity and water

content of Nafion 117 membranes. Journal of the Electrochemical Society 2004, 151, (3),

A381-A388.

11. Sone, Y.; Ekdunge, P.; Simonsson, D., Proton conductivity of Nafion 117 as measured by a

four-electrode AC impedance method. Journal of the Electrochemical Society 1996, 143, (4),

1254-1259.

12. Kim, Y. S.; Wang, F.; Hickner, M.; McCartney, S.; Hong, Y. T.; Harrison, W.; Zawodzinski,

T. A.; McGrath, J. E., Effect of acidification treatment and morphological stability of

sulfonated poly(arylene ether sulfone) copolymer proton-exchange membranes for fuel-cell

use above 100 degrees C. Journal of Polymer Science Part B-Polymer Physics 2003, 41, (22),

2816-2828.

13. Sumner, J. J.; Creager, S. E.; Ma, J. J.; DesMarteau, D. D., Proton conductivity in Nafion (R)

117 and in a novel bis[(perfluoroalkyl)sulfonyl]imide ionomer membrane. Journal of the

Electrochemical Society 1998, 145, (1), 107-110.

14. Kopitzke, R. W.; Linkous, C. A.; Anderson, H. R.; Nelson, G. L., Conductivity and water

uptake of aromatic-based proton exchange membrane electrolytes. Journal of the

Electrochemical Society 2000, 147, (5), 1677-1681.

15. Kim, Y. S.; Wang, F.; Hickner, M.; Zawodzinski, T. A.; McGrath, J. E., Fabrication and

characterization of heteropolyacid (H3PW12O40)/directly polymerized sulfonated

poly(arylene ether sulfone) copolymer composite membranes for higher temperature fuel cell

applications. Journal of Membrane Science 2003, 212, (1-2), 263-282.

 53

16. Nakano, T.; Nagaoka, S.; Kawakami, H., Preparation of novel sulfonated block copolyimides

for proton conductivity membranes. Polymers for Advanced Technologies 2005, 16, (10),

753-757.

17. Li, S. W.; Liu, M. L., Synthesis and conductivity of proton-electrolyte membranes based on

hybrid inorganic-organic copolymers. Electrochimica Acta 2003, 48, (28), 4271-4276.

18. Aparicio, M.; Duran, A., Hybrid organic/inorganic sol-gel materials for proton conducting

membranes. Journal of Sol-Gel Science and Technology 2004, 31, (1-3), 103-107.

19. Thampan, T. M.; Jalani, N. H.; Choi, P.; Datta, R., Systematic approach to design higher

temperature composite PEMs. Journal of the Electrochemical Society 2005, 152, (2), A316-

A325.

20. Li, Z. Y.; Zhang, Q.; Liu, H. T.; He, P.; Xu, X. D.; Li, J. H., Organic-inorganic composites

based on room temperature ionic liquid and 12-phosphotungstic acid salt with high assistant

catalysis and proton conductivity. Journal of Power Sources 2006, 158, (1), 103-109.

21. Mecheri, B.; D'Epifanio, A.; Di Vona, M. L.; Traversa, E.; Licoccia, S.; Miyayama, M.,

Sulfonated polyether ether ketone-based composite membranes doped with a tungsten-based

inorganic proton conductor for fuel cell applications. Journal of the Electrochemical Society

2006, 153, (3), A463-A467.

22. Sahu, A. K.; Pitchumani, S.; Sridhar, P.; Shukla, A. K., Co-assembly of a Nafion-

Mesoporous Zirconium Phosphate Composite Membrane for PEM Fuel Cells. Fuel Cells

2009, 9, (2), 139-147.

 54

23. Daiko, Y.; Sakakibara, S.; Sakamoto, H.; Katagiri, K.; Muto, H.; Sakai, M.; Matsuda, A.,

Formation of a High Conductivity Fuel Cell Electrolyte by Pressing Diphenylsiloxane-Based

Inorganic-Organic Hybrid Particles. Journal of the American Ceramic Society 2009, 92, (1),

S185-S188.

24. Huang, A. B.; Xia, C. Y.; Xiao, C. B.; Zhuang, L., Composite anion exchange membrane for

alkaline direct methanol fuel cell: Structural and electrochemical characterization. Journal of

Applied Polymer Science 2006, 100, (3), 2248-2251.

25. Stoica, D.; Ogier, L.; Akrour, L.; Alloin, F.; Fauvarque, J. F., Anionic membrane based on

polyepichlorhydrin matrix for alkaline fuel cell: Synthesis, physical and electrochemical

properties. Electrochimica Acta 2007, 53, (4), 1596-1603.

26. Luo, Y. T.; Guo, J. C.; Wang, C. S.; Chu, D., Quaternized poly(methyl methacrylate-co-butyl

acrylate-co-vinylbenzyl chloride) membrane for alkaline fuel cells. Journal of Power Sources

2010, 195, (12), 3765-3771.

27. Wu, Y. H.; Wu, C. M.; Varcoe, J. R.; Poynton, S. D.; Xu, T. W.; Fu, Y. X., Novel

silica/poly(2,6-dimethyl-1,4-phenylene oxide) hybrid anion-exchange membranes for

alkaline fuel cells: Effect of silica content and the single cell performance. Journal of Power

Sources 2010, 195, (10), 3069-3076.

28. Gilliam, R. J.; Graydon, J. W.; Kirk, D. W.; Thorpe, S. J., A review of specific conductivities

of potassium hydroxide solutions for various concentrations and temperatures. International

Journal of Hydrogen Energy 2007, 32, (3), 359-364.

 55

29. Carl H. Hamann, A. H., Wolf Vielstich Electrochemistry; Second ed.; Wiley-vch: Weinheim,

2007.

30. Sata, T. Ion Exchange Membranes; The Royal Society of Chemistry, 2004.

31. Spiegel, C. PEM Fuel Cell Modeling and Simulation Using Matlab; Academic Press, 2008.

32. Mulder, W. H.; Sluyters, J. H.; Pajkossy, T.; Nyikos, L., TAFEL CURRENT AT FRACTAL

ELECTRODES - CONNECTION WITH ADMITTANCE SPECTRA. Journal of

Electroanalytical Chemistry 1990, 285 (1-2), 103-115.

33. Kim, C. H.; Pyun, S.; Kim, J. H., An investigation of the capacitance dispersion on the fractal

carbon electrode with edge and basal orientations. Electrochimica Acta 2003, 48 (23), 3455-

3463.

34. Schiller, C. A.; Strunz, W., The evaluation of experimental dielectric data of barrier coatings

by means of different models. Electrochimica Acta 2001, 46 (24-25), 3619-3625.

35. Jorcin, J. B.; Orazem, M. E.; Pebere, N.; Tribollet, B., CPE analysis by local electrochemical

impedance spectroscopy. Electrochimica Acta 2006, 51 (8-9), 1473-1479.

36. Oldham, K. B., The RC time "constant" at a disk electrode. Electrochemistry

Communications 2004, 6 (2), 210-214.

37. LLC, B. In-Plane Conductivity Testing To Characterize High Temperature Polymer

Electrolyte Membranes, 2006.

38. Silva, R. F.; De Francesco, A.; Pozio, A., Tangential and normal conductivities of

Nafion((R)) membranes used in polymer electrolyte fuel cells. Journal of Power Sources

2004, 134 (1), 18-26.

 56

39. Yoshida, N.; Ishisaki, T.; Watakabe, A.; Yoshitake, M., Characterization of Flemion (R)

membranes for PEFC. Electrochimica Acta 1998, 43 (24), 3749-3754.

40. Slade, S.; Campbell, S. A.; Ralph, T. R.; Walsh, F. C., Ionic conductivity of an extruded

Nafion 1100 EW series of membranes. Journal of the Electrochemical Society 2002, 149

(12), A1556-A1564.

41. Reichle, W. T., Synthesis of Anionic Clay-Minerals (Mixed Metal-Hydroxides, Hydrotalcite).

Solid State Ionics 1986, 22, (1), 135-141.

42. Miyata, S.; Okada, A., Synthesis of Hydrotalcite-Like Compounds and Their

Physicochemical Properties - Systems Mg2+-Al3+-So42- and Mg2+-Al3+-Cro42-. Clays

and Clay Minerals 1977, 25, (1), 14-18.

43. Goh, K. H.; Lim, T. T.; Dong, Z., Application of layered double hydroxides for removal of

oxyanions: A review. Water Research 2008, 42, (6-7), 1343-1368.

44. Kay, M. I.; Young, R. A.; Posner, A. S., Crystal Structure of Hydroxyapatite. Nature 1964,

204, (496), 1050-&.

45. Eysel, W.; Roy, D. M., Hydrothermal Flux Growth of Hydroxyapatites by Temperature

Oscillation. Journal of Crystal Growth 1973, 20, (3), 245-250.