Improved performance of metal hydride electrode of Ni-MH battery with carbon
nanotubes
by
Humara Sultana
Submitted to the School of Materials Science and Engineering
in partial fulfillment of the requirements for the degree of
Master of Science
at
The University of New South Wales, Australia
2006
Abstract
Abstract
In the global search for renewable sources of energy, hydrogen is a promising candidate in transportation and electronic applications. Carbon nanotubes (CNTs) have the largest hydrogen storage capacity among the hydrogen storage materials known at present. The Ni-
MH battery can be used to store and then discharge large amounts of hydrogen reversibly by using hydrogen storage materials as negative electrode. The electrochemical hydrogen storage performances of metal hydride electrodes with different levels of multi wall carbon nanotubes (20%, 15%, 10%, 5% and 2% of Ni-MH battery’s active materials) has been investigated under similar charge-discharge conditions. Electrochemical test cell consisted of a single hydrogen storage negative electrode sandwiched between two NiOOH/Ni(OH)2 positive electrodes. A 6M aqueous KOH solution was used as electrolyte. Electrochemical properties such as specific discharge capacity, high rate charge-discharge capability and cycle life stability have been investigated. The morphology and structure of negative electrode material were examined by scanning electron microscopy, transmission electron microscopy and X-ray diffraction analysis. Chemical analysis of the hydrogen storage alloy was performed using electron probe microanalysis, electron diffraction spectroscopy and induced coupled plasma spectroscopy analysis. Hydrogen absorption-desorption properties were measured in terms of pressure-composition-isotherm curves.
It has been found in this study that the presences of CNTs significantly enhanced the overall electrochemical properties of the Ni-MH battery. Maximum specific discharge capacity was observed for 5% CNTs electrode reaching 243 mAh/g, whereas 0% CNTs could only reach 229 mAh/g. High rate charge and discharge capabilities of 5% CNTs I Abstract
electrodes were ~ 241% and 250% higher than the corresponding values for 0% CNTs electrode. Furthermore, the differences in electrochemical hydrogen storage of CNTs with different diameters of 10-20 nm, 20-40 nm, 40-60 nm, and 60-100 nm were investigated.
Electrochemical results demonstrated that CNTs with different diameters showed a large variation in the electrochemical hydrogen storage capability under the similar experimental condition.
A comparison between electrodes with different CNTs studies was carried out in order to optimize nanotubes choices for Ni-MH battery. It was found that smaller tube diameters,
20-40 nm and 5% CNTs negative electrode showed the best electrochemical properties of
Ni-MH battery system.
II Acknowledgements
This thesis is the result of the inspiring, thoughtful and patient guidance of my supervisor Senior lecturer Dr. Sammy Lap Ip Chan over the last two years. My sincere gratitude goes to my mentor and friend Dr. Rita Khanna for her valuable suggestions, helpful research support, ideas, and inspirations. Without her kind and sincere help this thesis would not be possible. My special thanks go to Professor Oleg Ostrovski, head of the School of Materials Science and Engineering, UNSW, for his guidance and support towards the completion of this work.
I am ever grateful to my friends: Tawfique Hasan, Susan Lee, Jerry Luo, Leida
Queddeng, Farjana Sarwar, Mrs Emon Rahman, MD. Faruqe for making my life in the university and in Sydney a vibrant and truly enjoyable one. Friendly and hard-working faces of fellow postgraduate students Muhammed Mahfuzur Rahman, Mahburur
Rahman, Gavin Parry surely kept me awake during numerous sleepy afternoons.
I am very grateful to the University of New South Wales for providing me with scholarship and opportunity to doing this research. I would like to extend my appreciation to the Professors, staffs and students of School of Materials Science and
Engineering for their support and encouragement.
And finally, I would like to thank my parents and in laws, brothers, sisters and my beloved husband Faisal Sarwar for everyday sacrifices and support in this adventure.
Humara Sultana
iii Dedication
To my parents and in laws.
iv Table of contents
Chapter 1: Introduction 1
1.1 Background 1
1.2 Problem facing Ni-MH batteries 2
1.3 Project objectives 3
1.4 Thesis overview 5
References 7
Chapter 2: Literature review 9
2.1 Introduction 9
2.2 Hydrogen as a clean energy source 9
2.2.1 What is carbon nanotube? 10
2.2.2 CNTs as a mean of hydrogen storage 12
2.2.3 Metal hydrides as a mean of hydrogen storage 15
2.3 Ni-MH battery 18
2.3.1 Comparison of rechargeable battery system 19
2.3.2 Reaction mechanism of Ni-MH battery 21
2.4 Electrochemical performance of Ni-MH battery 23
2.4.1 Discharge capacity of negative electrode 24
2.4.2 Activation profile and cycle life test 26
2.4.3 Self-discharge 31
2.4.4 High rate charge-discharge capabilities 32
2.5 PCI measurements 35
References 37 V Chapter 3: Experimental details 43
3.1 Introduction 43
3.2 Sample details 43
3.2.1 Materials 43
3.2.2 Experimental electrodes 46
3.3 Sample characterization techniques 47
3.4. Preparation of electrodes 49
3.5 Current collector 53
3.6 Electrochemical test cell setup 52
3.7 Electrolyte soaking of electrodes 54
3.8 Apparatus for electrochemical measurements 54
3.9 Electrochemical testing parameters for negative electrode 55
3.10 Experimental procedure 56
References 65
Chapter 4: Results and discussion 67
4.1 Introduction 67
4.2 Surface morphology 67
4.2.1 SEM, EDS and EPMA analysis 67
4.2.2 TEM analysis 74
4.3 Structure analysis 75
4.3.1 Crystal structure analysis 75
4.3.2 Structure analysis of CNTs 79
4.4 Electrochemical measurements 80
4.4.1 Activation and maximum capacity 81
VI 4.4.2 Discharge potential analysis 86
4.4.3 Cycle life characteristics and cycle stability 88
4.4.4 SOC, DOD and self-discharge test 98
4.4.5 High rate discharge characteristics 103
4.4.6 High rate charge characteristics 108
4.5 PCI measurements 111
4.6 Limitations of this study 113
References 115
Chapter 5: Conclusions 121
5.1 Conclusions 121
5.2 A brief summary of main results 121
5.3. Future work directions 125
5.4 Potential applications in electric vehicles 127
References 128
Appendix: Electron probe micro analysis spectrum 129
VII List of Figures
Fig. No. Figure caption Page
Fig. 2.1 Schematic diagram of carbon nanotubes. 11
Fig. 2.2 The power density versus energy density plot distinguishes various 14
electricity storage systems.
Fig. 2.3 Schematic diagram of the electrochemical reaction process of a Ni- 22
MH battery.
Fig. 2.4 The discharge curves of LaNi5 electrodes doped with different ratios 25
of CNTs of 15%, 10%, 7% and 5% in cycle number 40.
Fig. 2.5 Variation of discharge capacity of the two samples with number of 25
charge-discharge cycles at 25° C.
Fig. 2.6 Activation profile of La0.85MgxNi4.5Co0.35Al0.15 hydride electrodes. 27
Fig. 2.7 The hydrogen storage curves of CNTs-Ni electrode with different 27
diameter.
Fig. 2.8 Discharge curve of CNTs-LaNi5 electrode of 20-40 MWNTs treated 30
at different temperatures in nitrogen.
Fig. 2.9 Discharge curve of CNTs-LaNi5 electrode with different diameter 30
CNTs treated with 500° C in Vacuum.
Fig. 2.10 Storage characteristics of Ni-MH battery at different temperature. 31
Fig. 2.11 A solar vehicle used for Ni-MH study, Japan. 33
Fig. 2.12 Charge efficiency of the MH alloy sample. 34
Fig. 2.13 High rate discharge capability of the different electrodes. 35
VIII Fig. 3.1 Schematic diagram of experimental negative electrode. 49
Fig. 3.2 Photograph of the prepared negative electrode. 49
Fig. 3.3 Flow chart of the preparation of pasted type negative electrode. 50
Fig. 3.4 Photograph of as received positive electrode. 51
Fig. 3.5 The schematic diagram of the improved original positive electrode. 51
Fig. 3.6 Schematic diagram of cell assembly. 53
Fig. 3.7 Photograph of charge-discharge REPOWER machine. 54
Fig. 3.8 Flow chart of the experimental test of the negative electrodes. 56
Fig. 3.9 Experimental procedure of the electrochemical measurements. 56
Fig. 3.10 Sketch map of PCI system suitable for hydrogen storage materials 63
used in this experiment.
Fig. 4.1 SEM analysis of 0% CNTs electrode (upper picture) and 5% CNTs 68
electrode (lower picture).
Fig. 4.2 SEM image of ground 10% CNTs, Ni and AB5 alloy. 69
Fig. 4.3 The SEM image of as received AB5 alloy. 69
Fig. 4.4 EDS analysis of as received AB5 alloy. 70
Fig. 4.5 SEM image of as received CNTs diameter ranging from 20-40 nm. 73
Fig. 4.6 SEM image of as received polypropylene separator. 73
Fig. 4.7 TEM image of as received CNTs, diameter ranging from 20-40 nm. 75
Fig. 4.8 Schematic diagram of the crystal cell of AB5 alloy. 76
Fig. 4.9 XRD pattern of as-received AB5 alloy. 76
Fig. 4.10 XRD diffraction of as received CNTs, diameter ranging from 20-40 77
IX nm and ball milled CNTs (1hour) (upper graph) and 10-20 and 60-
100 nm CNTs (lower graph).
Fig. 4.11 XRD diffraction of as received Ni powder. 78
Fig. 4.12 XRD diffraction of prepared CNTs electrodes. 79
Fig. 4.13 Raman spectrum of as received CNTs, diameter ranging from 20-40 80
nm.
Fig. 4.14 Activation profile of different kind of CNTs electrode. 81
Fig. 4.15 Effect of diameter and ball milled CNTs on activation profile. 84
Fig. 4.16 Discharge curves of MH electrodes with various CNTs composition. 87
Fig. 4.17 Variation of the electrochemical discharge capacity of different 89
composition of CNTs electrode as a function of cycle number.
Fig. 4.18 (a) Photograph of the 2% CNTs electrode after cycle life test. 92
(b) Photograph of the deterioration of positive electrode after cycle 92
life test.
Fig. 4.19 XRD pattern of the 2% CNTs, Ni and MH powder and their 93
corresponding XRD after cycle life test.
Fig. 4.20 Effect of diameters and ball milled CNTs in the cycle life 95
performance of 2% CNTs electrode.
Fig. 4.21 Illustrated diagram of the aggregated pores of the CNTs. 97
Fig. 4.22 Specific discharge capacity at different state of charge. 99
Fig. 4.23 Specific discharge capacity at different depth of discharge. 99
Fig. 4.24 Self-discharge behavior of different storage time of the electrodes. 100
X Fig. 4.25 Voltage decay on open circuit at room temperature. 101
Fig. 4.26 High rate discharge capabilities of different types of CNTs 104
electrodes.
Fig. 4.27 Effect of diameter and ball milled CNTs on high rate discharge 106
capabilities.
Fig. 4.28 High rate charge capabilities of the CNTs electrodes. 108
Fig. 4.29 High rate discharge capabilities of the CNTs electrodes. 109
Fig. 4.30 Effect of diameter and ball milled CNTs on high rate charge 110
capabilities.
Fig. 4.31 Hydrogen absorption-desorption of as received MH alloy. 112
Fig. 4.32 Hydrogen absorption-desorption of ball milled CNTs. 113
XI List of Tables
Table No. Table caption Page
Table 2.1 Experimental findings of hydrogen storage capacities in CNTs. 13
Table 2.2 The advantage and disadvantage of AB2 and AB5 alloys. 17
Table 2.3 Cell voltage, specific energy and cycle lifetime of rechargeable 19
batteries.
Table 3.1 Materials used in the electrochemical experiments. 49
Table 3.2 Parameters and procedure for activation and cycle life test. 58
Table 3.3 Experimental procedure and parameter for state of charge test. 60
Table 3.4 Experimental procedure and parameter for depth of discharge 60
test.
Table 3.5 Experimental procedure and parameter for self-discharge test. 61
Table 3.6 Experimental procedure and parameter for high rate discharge 62
test.
Table 3.7 Experimental procedure and parameter for high rate charge test. 62
Table 4.1 Chemical composition of AB5 Powder. 70
Table 4.2 Composition of AB5 alloy obtained by EPMA analysis. 71
Table 4.3 Composition of MH alloy by ICP analysis (%). 72
Table 4.4 Activation and maximum capacity of the electrodes. 82
Table 4.5 Activation and maximum capacity of different diameter and ball 84
milled electrodes.
XII Table 4.6 Discharge mid point voltage obtained from different electrodes. 86
Table 4.7 Experimental findings obtained by cycle life test of the 91
electrodes.
Table 4.8 Effect of diameter and ball milled in 5% CNTs electrodes. 96
Table 4.9 Available specific discharge capacity after self-discharge. 101
Table 4.10 High rate discharge performance of the electrodes at room 104
temperature.
Table 4.11 High rate discharge performance of 5% electrodes at room 107
temperature.
Table 4.12 Specific charge capacity at different high rate charge current. 109
Table 4.13 Specific discharge capacity at different high rate charge current. 109
Table 4.14 Effect of diameter and ball milled CNTs on specific charge 111
capacity at different high rate charge current.
XIII List of commonly used abbreviations
Nickel metal hydride battery Ni-MH
Nickel cadmium battery Ni-Cd
Lithium ion battery Li-ion
Carbon nanotubes CNTs
Single-wall nanotubes SWNTs
Multi-wall nanotubes MWNTs
Nickel Ni
Metal hydride alloy MH
Hydrogen storage alloy AB5
Scanning electron microscopy SEM
Electron diffraction spectroscopy EDS
Electron probe micro analysis EPMA
X-ray diffraction XRD
Transmission electron microscopy TEM
Induced coupled plasma spectroscopy ICP
Polytetrafluorethylene PTFE
State of charge SOC
Depth of discharge DOD
XIV Introduction
Chapter 1
Introduction
1.1 Background
The rapid growth of communications, computers, consumer electronics, and electric and hybrid vehicles have created an urgent need for high energy density storage batteries.
Although conventional Nickel cadmium (Ni-Cd) and lead-acid batteries have been improved in recent years, they still need a better performance and higher power density.
The innate toxicity of cadmium and lead also has become a severe problem.
Applications of rechargeable Nickel metal hydride (Ni-MH) battery began with hydrogen storage alloy as negative electrode in 1990 [1-2]. Ni-MH batteries have received much attention because of their higher energy density, superior charge-discharge characteristics, non memory effect and non polluting nature. Ni-MH batteries have higher gravimetric and volumetric energy densities as compared to Ni-Cd batteries of the same size by approximately 30–40% [3-4]. On the other hand, rechargeable Lithium ion (Li-ion) batteries present as strong competitors to the Ni-MH batteries, with their high specific power density and light weight. For safety reasons, however, they can not be operated without an electronic control. Li-ion batteries are quite expensive as compared to other rechargeable batteries.
1 Introduction
A smart rechargeable battery must have the following performance characteristics: fast initial activation, high energy and power density, longer and stable cycle life, low self- discharge rate, and it should be environmentally friendly, safe to use and economical. In
1973, Philips Research Laboratory in Netherlands introduced negative electrodes made with AB5 hydrogen storage materials [5]. Thereafter many researchers have focused their attention to these AB5 type rare earth based hydrogen storage alloys which are light in weight and are inexpensive. The first AB5 alloy developed was LaNi5, which, through continuous improvement via alloying could reversibly absorb and release hydrogen at room temperature and could easily be activated as well as produce high reacting rate for absorbing and releasing hydrogen [6].
Discovery of carbon nanotubes (CNTs) in 1991 by Sumio Iijima [7], gave the CNTs a ray of hope towards huge potential applications in hydrogen storage fields, owning to their large hydrogen storage capacity inside the nanotubes. It is believed that the potential high hydrogen storage capability is due to the large surface area and large internal empty spaces of the nanotubes. CNTs possess unique properties due to their diameter and structure. They are also chemically stable and have low mass density. Addition of CNTs to alloy particles has been shown to be beneficial to electrode performances in the Ni-MH battery [8].
1.2 Problems facing Ni-MH batteries
Although there are good overall reaction properties of the Ni-MH battery, there are some factors affecting the long term performance of Ni-MH batteries such as capacity decay
2 Introduction
during cycle life, and self-discharge rate. In some applications where high power output is essential, such as electric and hybrid vehicles, Ni-MH batteries are inferior to Ni-Cd batteries without high rate discharge capacity [9]. In this prospect, it is important to have a deep understanding of reactions and deterioration mechanisms of metal hydride electrodes during charge-discharge cycles. During repeated charge-discharge cycles, hydrogen storage alloys (such as AB2, AB5) loose hydrogen storage capacity as a result of oxide formation and surface degradation of the metal hydride alloy pulverization. This phenomenon is not only responsible for their reduced reactivity but also affects the performances of MH electrodes in charge-discharge cycles [3, 10].
Ni-MH battery characteristics such as cycle life depend on the conductivity and electrocatalytic activity. Hydrogen storage alloys generally have poor conductivity and low active sites due to oxidation and decomposition with repeated charge-discharge cycles. For high power applications, high rate capabilities need to be improved for achieving excellent performances.
1.3 Project Objectives
Main issues on the development of Ni-MH battery are the capacity and high power requirements. These properties are determined by kinetics processes occurring during charge-discharge cycles and diffusion rates for hydrogen within the bulk of the alloy. Due to high surface area and abundant pore volume CNTs are considered to absorb a large
3 Introduction
amount of hydrogen [11]. The electrochemical characteristics of the Ni-MH battery can be improved with the hydrogen storage materials CNTs to achieve better performances.
A mixing of active material metal hydride alloy with CNTs in various proportions has been proposed in an attempt to improve the negative electrode characteristics of the Ni-MH battery and to address the problems discussed above.
The primary objectives of this study are
x To investigate the operating characteristics of Ni-MH batteries with AB5 alloy in
the negative electrode.
x To investigate the Ni-MH battery characteristics mixed with 20-40 nm diameter
multi-walled CNTs.
x To observe the influence of different diameters and ball milling of CNTs on the
electrochemical performances of electrodes containing these CNTs.
x To study the hydrogen absorption-desorption behavior of AB5 and CNTs.
Techniques used in this project
Experiments were performed using the following techniques.
x X-ray diffraction for phase identifications. Scanning electron microscope and
Electron diffraction analysis, Transmission electron microscope and Electron probe
micro analysis to characterize the hydrogen storage alloy and CNTs.
x ICP analysis for chemical composition analysis of the metal hydride alloy.
x Raman spectroscopy for structure analysis of the CNTs.
x Galvanostatic charge-discharge to determine the electrochemical characteristics of
the negative electrodes of the Ni-MH battery.
4 Introduction
x Conventional Sievert’s type apparatus for the hydrogen absorption and desorption
rate determination of the hydrogen storage alloy and CNTs.
1.4 Thesis overview
In this thesis, the improvement of negative electrode with carbon nanotube is presented along with a review of different characteristics of metal hydride alloy and CNTs. Optimum percentages of CNTs in the negative electrode are determined through extensive experimentation. Experimental results are presented along with discussion on material characteristics and their influence on the performance of negative electrode.
A brief introduction of the background of this work is presented in Chapter 1. This chapter also outlines the thesis contents.
A detailed discussion on the Ni-MH batteries along with metal hydride alloy negative electrode and CNTs, their electrochemical characteristics, reaction mechanisms and various electrode issues/problems are presented in Chapter 2 with an extensive literature survey on previous research in this field.
Chapter 3 is focused on the basic design methodology of the proposed negative electrodes with metal hydride alloy and CNTs. In this chapter, different negative electrodes are designed for a range of sizes and amounts of CNTs. This chapter also provides general
5 Introduction
guideline and design for the preparation of the electrodes along with experimental methods to study the electrode performance.
In Chapter 4, metal hydride alloys, CNTs and prepared electrodes are characterized using physical and chemical analyses. The electrochemical performances of the negative electrodes are characterized using electrochemical techniques with different CNTs at ambient pressure and temperature. Hydrogen absorption-desoprtion behavior of materials is also investigated using pressure composition isotherm curves. Experimental results are discussed along with the limitations of this investigation.
This thesis is concluded in Chapter 5 with conclusions and summary of this investigation and directions for future developments in this field.
6 Introduction
References
[1] M. Okada, T. Kurriwe, “Role of intermatallics in hydrogen storage materials”,
Materials Science and Engineering A 329-331 (2002) 305-312.
[2] N. Cui, J.L. Luo, “Nickel metal hydride (Ni-MH) battery using Mg2Ni type
hydrogen storage alloy”, Journal of Alloys and Compounds 302 (2000) 218-226.
[3] F. Haschka, W. Warthmann, “Alkaline batteries for hybrid and electric vehicles”,
Journal of Power Sources 72 (1998) 32-36.
[4] F. Fenga, M. Genga, D.O. Northwood, “Electrochemical behaviour of intermetallic-
based metal hydrides used in Ni/metal hydride (MH) batteries: a review”,
International Journal of Hydrogen Energy 26 (2001) 725–734.
[5] Wenlin Zhang “Cell components with emphasis on hydride electrodes for
nickel/hydrogen batteries”, PhD thesis, Texas A & M University, August 1995.
[6] Ping Li , Xin-Lin Wang , “Research of low-Co AB5 type rare-earth-based hydrogen
storage alloy electrodes ”, Journal of Alloys and Compounds 354 (2003) 310–314.
[7] Sumio Iijima “Carbon nanotubes: past, present, and future”, Physica B 323 (2002)
1-5.
[8] Xiaojuan Fu, Haiyan Zhang, “Effect of carbon nanotubes on the electrochemical
hydrogen storage performance of LaNi5 rare earth alloy”, Physica E 25 (2005) 414-
420.
[9] Chan-Yeol Seoa , Seung-Jun Choib, “Effect of vanadium content on
electrochemical properties of la-based AB5-type metal hydride electrodes”,
International Journal of Hydrogen Energy 28 (2003) 967 – 975.
7 Introduction
[10] Junmin Nan, Yong Yang, “Raman spectroscopic study on the surface oxide layer of
AB5-type metal hydride electrodes”, Electrochimica Acta 46 (2001) 1767–1772.
[11] Hui-Ming Cheng, Quan-Hong Yang, “Hydrogen storage in carbon nanotubes”,
Carbon 39 (2001) 1447-1454.
8 Literature review
Chapter 2
Literature review
2.1 Introduction
Electrical energy plays an important role in our everyday life. Electrical energy can be easily converted into light, heat or mechanical energy, but it is very difficult to store electrical energy. The storage of electrical energy requires the conversion into another form of energy. An electrochemical cell is a chemical device used for producing electricity. A basic battery consists of a positive electrode and a negative electrode, separated by a separator. The separator allows the passage of ions between two electrodes, but itself is an electronic insulator. When two or more cells are connected together electrically, either in series or parallel or in series of parallel array, this resulting assembly is known as a battery.
A brief discussion on hydrogen storage in nickel metal hydride (Ni-MH) batteries is presented at the beginning of this chapter. This is followed by a review on mechanisms, performances of Ni-MH battery, as well as improvement of Ni-MH battery with CNTs.
2.2 Hydrogen as a clean energy source
Hydrogen is the simplest element known to man, with an atom being composed of just one proton and one electron. Hydrogen is the most abundant element as well, accounting for more than 90% of atoms observed in the universe [1]. But hydrogen does not exist on earth 9 Literature review
as a gas in nature. It is combined with oxygen to form water. Hydrogen is also present in different compounds such as methane, coal and petroleum after combining with carbon.
Currently most of our energy comes from fossil fuels. Only 6% comes from renewable energy sources. However it is not efficient to get energy from renewable energy sources like solar and wind at all times. On the other hand, hydrogen can store energy until it is needed and can be transferred to preferred location. Hydrogen is an ideal fuel and versatile energy carrier because it is easy to produce, has high efficiency, easy to transport and above all it is green [2].
It may be assumed that hydrogen will lead to a possible solution to solve the problem of the urban pollution, with zero-emission vehicles which are currently propelled by electric motors supplied by hydrogen fuel cells. Among many choices, hydrogen is one of the best options. The storage of hydrogen gas on board automobiles is a problem not solved yet; therefore, it is the focus point of many applied and fundamental researches.
2.2.1 What is carbon nanotube?
The name “Carbon nanotubes” (CNTs) is derived from their size; nanotubes are only a few nanometers wide, of the order of one ten-thousandth the width of a human hair, and their length can be million times greater than their width (almost two millimeters long!).
CNTs were discovered in 1991 by S. Iijima [3-5]. CNTs can be considered as a sheet of graphite (a hexagonal lattice of carbon) rolled into a cylinder, Figure 2.1, with an inner diameter starting from 0.4 nm up to several nm and a length of 10-100 µm [4].
10 Literature review
Figure 2.1: Schematic diagram of CNTs [6]
CNTs have excellent mechanical properties with high elastic modulus of 1.8 TPa and bending strength of 14.2 GPa and better electrical conductivity. They have lower density
(less than Al) and excellent thermal stability (1400˚C in a vacuum). The carbon arrangement becomes very strong when coiled. Because of their nanometer size and many principal properties, such as, high accessible effective surface area, chemical stability, and high electrical conductivity and mechanical strength indicate that CNTs can be used in the negative electrode of Ni-MH battery. The properties of cylindrical carbon molecules in
CNTs make them potentially useful in extremely small scale electronic and mechanical applications [7].
11 Literature review
There are two main types of nanotubes: single-walled nanotubes (SWNT), which are formed with only one single graphite layer; and multi-walled nanotubes (MWNT), which consist of multiple concentric graphite layers. The diameter of SWNTs varies from 0.4 to 3 nm, and in the case of MWNTs typical diameters varies from 30–50 nm [4]. This unique structure has resulted in CNTs possessing higher mechanical strength.
Similar to graphite CNTs are composed entirely of sp2 bonds, which are stronger than the sp3 bonds found in diamond. This bonding structure is responsible for their unique strength.
CNTs naturally align themselves into ropes which are held together by Van der Waals forces. CNTs can merge together at high pressure, trading some sp2 bonds for sp3 bonds which are responsible for production of strong, unlimited-length wires in high pressure nanotubes [8].
2.2.2 CNTs as a means of hydrogen storage
CNTs appear to be one of the most promising candidates for the hydrogen absorption. One of the main advantages of CNTs is that it is one of the very few light materials which is solid at room temperature [9]. CNTs are well known as being good adsorbents for gases
[10]. The origin of this property is due to the existence of these materials in a very finely powdered form with highly porous structure and because of the existence of special interactions between carbon atoms and gas molecules [10-11]. Hydrogen exists in the form of H2 molecules in empty spaces inside CNTs [12].
12 Literature review
Table 2.1: Experimental findings of hydrogen storage capacities in CNTs
Hydrogen storage amount, Material References wt%
CNTs 0.1-67 [9]
CNTs >3 [13]
SWNTs 5-10 [12, 14-16]
SWNTs 4.2-65 [17]
MWNTs(K and Li-doped) 14-20 [4], [2], [18]
Several investigations have been carried out on hydrogen absorption into CNTs. The experiemental summary of the hydrogen absorption in CNTs is shown in Table 2.1. Several reseach groups have reported that CNTs exhibit a high hydrogen absorption capacity, >3wt
% at ambient temperatures. Most of this research indicates that SWNTs have hydrogen storage capacity of 5–10 wt% at pressures less than 1 bar near room temperature. Chen et al. has claimed a high H2 uptake of 20 and 14 % can be achieved respectively for Li-doped and K-doped MWNTs in milligram quantities under ambient pressure [18, 2].
Recently CNTs have been reported to be very attractive candidates for hydrogen absorption in electrochemistry research, Figure 2.2. The large surface area and low resistivity of CNTs makes them of great interest in electrochemical studies. The large surface area and tubular structure of CNTs suggest that capillarity effects are important. Thus, CNTs could be very useful for the absorption of hydrogen, a key aspect of the clean energy economy [19].
13 Literature review
Figure 2.2: The power density versus energy density plot distinguishes various electricity storage systems
[19]
The hydrogen absorption results are based on the nature of bonding between carbon and hydrogen. Hydrogen absorption in the CNTs originate either from the van der Waals interaction with binding energy ~0.1 eV, known as physisorption or from covalent C-H interaction with binding energy more than 2-3 eV, known as chemisorption [2, 14, 20].
Hydrogen absorption-desorption is directly related to the surface absorption process. The recent discovery of hydrogen storage in CNTs has stimulated interest in the high and reversible hydrogen storage capacity of CNTs [2, 4, 15]. Currently the applications of
CNTs in the electrodes of hydrogen storage battery, super capacitor and Li-ion battery is being carried out to improve the electrochemical hydrogen storage facility.
14 Literature review
2.2.3 Metal hydrides as a means for hydrogen storage
In 1970’s work on hydrogen storage alloys were initiated and developed to a great extent
[21, 22]. These have two basic properties: high and reversible hydrogen storage capacity per mole of compound and high energy stored per unit volume, making MH an attractive option.
Hydrogen storage alloy material can form hydride through interaction with hydrogen gas or by the electrochemical method by the following equations
l MHHM (2.1)
2 l OHMHeOHM (2.2)
where M stands for a hydrogen storage alloy and MH for hydride. Not all hydrogen storage alloys can be charged and discharged through equation (2). Hydrogen storage alloys play an important role not only as a catalyst for the charge-discharge of the hydrogen but also act as a hydrogen reservoir [21, 23]. As a result good hydrogen storage alloys must have the following requirements:
x High reversible hydrogen storage capacity, i. e.> 1 wt%
x Good electrochemical catalyst characteristics for hydrogen charge-discharge
characteristics
x Easy and fast activation
x Excellent corrosion resistance in the concentrated KOH electrolyte solution
x Have the capability of being cycled without alteration in pressure or temperature
during the life of the system 15 Literature review
x Very slow degradation of alloy with cycling
x Good thermal and electrical conductivity
The main idea for hydrogen absorbing alloy is the combination of strong hydride forming elements A with weak hydride forming elements B, where A is usually a rare earth or an alkaline earth metal and element B is often a transition metal. These are generally brittle highly ordered alloys, usually with narrow, integral stoichiometries (AaBb) that can react with hydrogen to form an intermediate strength hydride AaBbHx. The variation in composition of these alloys results in a wide variation in the properties of these hydrides.
There are certain classes of multi-component hydrogen absorbing alloys based on their microstructures that seem to work better than the others for reversible hydrogen storage.
Among them AB2 and AB5 have drawn much attention due to their higher absorption properties.
The general form of AB5 compound is Mm (Ni, Co, Mn, Al)5 or LaNi5, where Mm
(Mischmetal) is a mixture of rare earth elements; its composition corresponds to that of different natural ores. It contains 50-55% cerium, 18-28% lanthanum, 12-18% neodymium,
4-6% praseodymium and other rare earth elements in small quantities, as well as metallic impurities. In Ni-MH batteries, one often employs lanthanum-enriched Mischmetal, for example 50% lanthanum, 30% cerium, 14% neodymium and other rare earth metals [24].
The Ni-MH battery performance is measured by parameters such as capacity, cycle life and discharge capability. These parameters closely depend on the parameters of hydride
16 Literature review
forming alloy (AB2, AB5), which are used as the active material of the negative electrode.
The electrochemical characteristics can be changed by designing the composition of the hydrogen storage alloy to provide optimum performance.
AB5 alloys are capable of absorbing hydrogen 1000 times their own volume. They are designed in such a configuration that ‘A’ hydrides generate heat exothermically while ‘B’ hydrides generate heat endothermically in AB5 alloy. These reactions introduce binding energy that leads hydrogen to be released and absorbed [25]. The characteristics of AB2 and AB5 alloys are shown in Table 2.2.
Table 2.2: The advantages and disadvantages of AB2 and AB5 alloys.
Alloys Characteristics References
AB2 Advantages Higher hydrogen capacity [22, 26]
Rapid kinetics [27]
Disadvantages More sensitive to gaseous impurities [22]
Hard to active electrochemically [23]
Main elements are toxic [27]
Low cycle life [27]
High self-discharge rate [27]
Poor high rate discharge capability [28]
AB5 Advantages High hydrogen storage, ~250-300 mAh/g [29]
Easy activation [30]
Better discharge ability [31]
Light weight
Economic
17 Literature review
Disadvantages Slow discharge kinetics [32]
Low discharge ability at high power [33]
application
Suffer Severe corrosion during cycling [34, 35]
The ability to high rate discharge and low temperature for AB5 hydrogen storage alloy still need to be further improved to meet the requirements for extended application [26].
2.3 Ni-MH battery
Nowadays Ni-MH batteries based on MH negative electrodes is one of the major and important areas of electrochemical studies. These batteries are in high volume commercial production for portable batteries beginning in 1989 and achieving over 900 million annual worldwide cell productions in 1999. The rapid growth of Ni-MH batteries is due to both technical and environmental advantages, with energy and performance higher than that of
Ni-Cd batteries. The growth is also largely encouraged by the explosive growth of portable electronic devices. For the best overall performance Ni-MH batteries have become the dominant advanced battery technology for electric vehicle (EV) and hybrid electric vehicle
(HEV) applications such as Toyota, Mazda [7]. The energy density of the Ni-MH cells is ~
75% higher than that of Ni-Cd batteries [36].
In addition to the essential performance targets of energy, power, cycle life and operating temperature, the following features of Ni-MH have established the technology pre- eminence [36-39]:
18 Literature review
x Flexible cell sizes from 60 mAh–250 Ah
x Safe operation at high voltage (320+ volts)
x Excellent volumetric energy and power, flexible vehicle packaging, improved
energy density (up to 40 percent greater than nickel-cadmium cells)
x Safety in charge and discharge, including tolerance to abusive overcharge and over
discharge;
x Excellent thermal properties;
x Very low memory effect;
x Environmentally acceptable and recyclable materials.
Simplified incorporation into products currently using nickel cadmium cells because of the many design similarities between the two chemistries.
2.3.1 Comparison of rechargeable battery system
Typical characteristics of rechargeable batteries, including cell voltage, specific energy and cycle lifetime are listed in Table 2.3.
Table 2.3: Cell voltage, specific energy and cycle lifetime of rechargeable batteries
[40-41].
Characteristics Ni-Cd Ni-MH Lead acid Li-ion
Negative electrode Cd Hydrogen storage Pb LixC6
alloy
19 Literature review
Energy density, Wh/Kg 45-80 60-120 30-50 110-160
Cycle life 1500 300-500 200-300 500-1000
Fast charge time 1 h 2-4 h 8-16 h 2-4 h
Overcharge tolerance Moderate Low High Very low
Cell Voltage 1.25 1.25 2 3.6
Self-discharge/month 20% 30% 5% 10%
Typical cost (US$) $50 $60 $25 $100
Commercial use since 1950 1990 1970 1991
Even though Ni-MH batteries have higher specific energy than the other two alkaline electrolyte systems, they remain largely inferior to another rechargeable Li-ion battery. In very recent years, rechargeable Li-ion batteries have appeared on the market as the latest generation of portable, rechargeable energy sources; they have impressive characteristics and largely surpass Ni-MH batteries regarding energy delivered per unit weight or volume.
But Lithium batteries cannot be operated for safety reasons without electronic control of each individual cell. Thus, Li-ion batteries will only be used in applications where it is imperative to have maximum energy content.
For most uses and for better performance, the Ni-MH battery system will be preferred. Ni-
MH offers more energy per unit volume or weight than Ni-Cd or lead-acid batteries. They are free of toxic heavy metals, such as cadmium. 20 Literature review
2.3.2 Reaction mechanisms of Ni-MH battery
The nickel-metal hydride cell chemistry is a hybrid of the proven positive electrode chemistry with the energy storage of metal alloys developed for advanced hydrogen energy storage concepts. In negative electrode hydrogen is stored reversibly in the form of metal hydride and the positive electrode is a standard nickel oxide electrode, the electrode reactions are:
Charge
At the negative electrode, in the presence of the hydrogen storage alloy M and with an electrical potential applied, the water in the electrolyte is decomposed into hydrogen atoms, which are absorbed into the alloy, and hydroxyl ions as indicated below:
2 o OHMHeOHM E˚= -0.828V (2.3)
At the positive electrode, the charge reaction is based on the oxidation of nickel hydroxide just as it is in the nickel-cadmium cell.
()2 OHOHNi o 2 eOHNiOOH E˚= 0.490V (2.4)
Discharge
At the negative electrode, the hydrogen is desorbed and combines with a hydroxyl ion to form water while also contributing an electron to the circuit.
2 o eOHMOHMH (2.5)
At the positive electrode, nickel oxy-hydroxide is reduced to its lower valence state, nickel hydroxide.
2 ()2 OHOHNieOHNiOOH (2.6)
The reversible reaction of this electrode is
21 Literature review
arg edisch 2 l () 2 OHOHNieOHNiOOH (2.7) arg ech
During charging and overcharging oxygen is generated at the nickel electrolyte interfaces, it may evolve as gas, the oxygen can then be transported from the nickel to MH alloy electrode where the oxygen gas may dissolve in the electrolyte and be reduced at the MH electrolyte surface [42]. The overall schematic charge discharge phenomenon is shown in
Figure 2.3.
Figure 2.3: Schematic diagram of the electrochemical reaction process of a Ni-MH battery [38].
22 Literature review
MH battery operates in a strongly oxidizing medium composed of high concentration alkaline electrolyte. The electrochemical reaction processes at the metal hydride electrode take place at the interface between the MH alloy powder and electrolyte.
In an alkaline aqueous solution, the hydrogen atoms produced at the surface of the MH alloy powder are instantly adsorbed and then diffuse into the bulk of the MH alloy. The electrochemical reactions are therefore expressed as follows
2 ad l OHHeOH (2.8)
ad l HH ab (2.9 where Had and Hab denote the hydrogen atoms adsorbed on the surface of the MH alloy powder and absorbed in the bulk of MH alloy, respectively. Reaction (2.8) reflects the charge transfer process at the interface between the MH alloy powder and the electrolyte, and reaction (2.9) relates to the diffusion of hydrogen from the powder surfacing to the bulk of the MH alloy, which is mass transfer process [30, 43].
Hydrogen diffusivity in the alloy should be high to ensure the high discharge rate of the Ni-
MH battery, especially at high discharge rates. These parameters are related to the mass transfer and charge transfer processes, and the discharge capability of the battery depends critically on these processes, Figure. 2.3.
2.4 Electrochemical performance of Ni-MH battery
In several research electrochemical experiments of the Ni-MH battery focused mainly on activation characteristics, total charge-discharge cycles of the electrodes, high rate charge-
23 Literature review
discharge capabilities and self-discharge rate. The overall performance of the Ni-MH depends on the composition of the MH alloy, electrolyte concentration, binder and electrode fabrication techniques.
2.4.1 Discharge capacity of negative electrode
The high rate discharge capacity of the Ni-MH battery mainly reflects that energy storage into the negative electrode. It is a key parameter to determine whether the AB5 alloy or
AB5-CNTs can be used in the Ni-MH battery system. The electrochemical hydrogen storage capability of the LaNi5 electrodes doped with different amounts of CNTs (15%,
10%, 7% and 5%) under the similar charge-discharge conditions are shown in the Figure
2.4. A three-electrode system was introduced in this study: AB5-CNTs electrode was used as the negative electrode, NiOOH/Ni(OH)2 as the positive electrode and Hg/HgO as reference electrodes under the charge-discharge current density 200 and 100 mA/g. From
Figure 2.4, it can be seen, that the electrode with 10% CNTs has a discharge capacity of
407 mAh/g, which is the highest and most stable among all other electrodes.
24 Literature review
Figure 2.4: The discharge curves of LaNi5 electrodes doped with different ratios of CNTs of 15%, 10%, 7% and 5% in cycle number 40 [44].
Figure 2.5 shows the specific discharge capacity as a function of number of cycles using
MH alloy as active material in the negative electrode. The decay of these two samples is caused by the deterioration of the negative electrode and hence the MH alloy.
Figure 2.5: Variation of discharge capacity of the two samples with number of charge-discharge cycles at 25°
C [41]. 25 Literature review
It can be seen from Figure 2.5 that the specific discharge capacity shows a large variation for the first 30 cycles and this variation is caused by the deterioration of the negative electrode. It is believed that the grain boundaries in MH alloy have layers of segregated rare earth element La, which transforms into La(OH)3 and corrosion.
2.4.2 Activation profile and cycle life test
Although AB5 alloy are known to be rather easily activated, activation process took several repeated charge-discharge cycles, after which reversible and stable absorption-desorption characteristics are obtained. During the first five repeated charge-discharge cycles the particle size of the AB5 alloy powder decreases greatly and after that remains constant
(about 2µm) with increasing number of cycles. In these hydrogen absorption-desorption cycles particles are broke up with each cycles and thus the alloy powders easily oxidized in these first couples of cycles. After these first couples of cycles the particle size reaches a constant value which resists further oxidation and then remains almost constant with increasing cycle number [30].
Figure 2.6 shows the activation profiles of La0:85MgxNi4:5Co0:35Al0:15 electrodes. The activation of La0:85MgxNi4:5Co0:35Al0:15 alloy electrodes in the range of 0.05 26 Literature review Figure 2.6: Activation profile of La0.85MgxNi4.5Co0.35Al0.15 hydride electrodes [25]. This easy activation might be due to the addition of Ni, which is very useful to metal hydride decompositions. This result reveals that electrochemical hydrogen storage strongly depends on compositions of active material of the negative electrode. Figure 2.7: The hydrogen storage curves of CNTs-Ni electrode with different diameter [46]. 27 Literature review The cycle duration of Ni-MH battery is one of the most important performances that determine whether or not it can be used in a practical battery. During repeated charge- discharge cycles the specific discharge capacity gradually decreases due to oxidation, and degradation of the active material [45-46, 37]. The cycle life of the electrode usually measured by the total number of repeated charge-discharge cycles of the battery until the specific discharge capacity of the electrode decreases to a certain value. In order to compare the discharge performance of the CNTs-Ni electrodes with different tube diameters, their highest discharging curves are shown as Figure 2.7. This graph showed that the 10-30 nm CNTs-Ni electrode has the best charge-discharge characteristics. The author concluded that it is not necessary that the bigger or smaller diameters of MWNTs are better for electrochemical hydrogen storage property [46]. In CNTs hydrogen molecules can be stored both in the porous holes and between the layers of CNTs. The hydrogen molecules that get deposited between the layers of CNTs enter mainly by diffusing from the outside to inside. The layers of CNTs are in direct proportion to the tube diameter, and too many layers can keep hydrogen coming into the tube center from tube walls through diffusion, and therefore the deposition and release of hydrogen are reduced. On the other hand, when the diameter is too small (such as 10–20 nm), the hydrogen storage is limited by the few layers and hence resulted a decrease of charge-discharge capacity. So CNTs with smaller outside diameter have better charge-discharge performance, and higher diameter electrode has a lower discharge capacity and a lower discharge plateau. Among the CNTs investigated, H. Zhang et al. concluded that the 10–30 28 Literature review nm CNTs have the largest discharge capacity and highest voltage plateau with a good cyclic life. Figure 2.8 shows that the CNTs-LaNi5 electrodes with 20-40nm diameter CNTs heated at 800° C has the best charge-discharge characteristic [47]. This experiment also proved that a pure LaNi5 electrode only has a maximum discharge capacity of 265.6 mAh/g and the corresponding voltage plateau is 0.83 V, which is lower than that of any of the CNTs- LaNi5 electrodes. The author concluded that the hydrogen storage capacity has a close correlation with the structure of CNTs, and the hydrogen molecules can be stored both in the porous hole and between the layers of the CNTs. The hydrogen which is deposited between the layers of CNTs enter by diffusion process from outside to inside of the tube. CNTs heated to different temperatures in nitrogen can help hydrogen coming into the tube center by diffusion through the tube wall. Consequently the deposition and release of hydrogen are enhanced. So CNTs heated in nitrogen have improved charge-discharge cycles. This experiment showed that 20-40 nm MWNTs treated at 800°C in nitrogen has the best electrochemical hydrogen storage property. 29 Literature review Figure 2.8: Discharge curve of CNTs-LaNi5 electrode of 20-40 MWNTs treated at different temperatures in nitrogen [47] Figure 2.9 showed that the 20-40 nm CNTs-LaNi5 electrode has a best charge-discharge characteristic. 60-100 nm CNTs-LaNi5 electrode has the worst electrochemical hydrogen storage characteristic among all CNTs-LaNi5 electrodes. Figure 2.9: Discharge curve of CNTs-LaNi5 electrode with different diameter CNTs treated with 500° C in Vacuum [48]. 30 Literature review From the above results it can be observed that most of the researchers have concluded that the higher electrochemical performances depends on the diameter of the MWNTs. Too much or too small amount of CNTs has no effect on the improvement of overall performance of Ni-MH battery [46-48]. 2.4.3 Self-discharge One of the main disadvantages of Ni-MH battery with MH alloy electrode is the self- discharge. Self-discharge is the loss of all capacity of a fully charged cell while standing the open circuit. Self-discharge occurs by the slow decomposition of both positive and negative electrodes due to their intrinsic instability as well as reaction of hydrogen with nickel electrode. It is reported that the rate of self-discharge increases with increasing temperature, shown in Figure 2.10. Figure 2.10: Storage characteristics of Ni-MH battery at different temperature [24] 31 Literature review Possible mechanisms for self-discharge are [49-51]: x The reaction between residual hydrogen in the battery with the positive electrode x Slow decomposition of the positive and negative electrodes x Shuttle effect of nitrite from the separator or sintered positive electrode. During the storage period MH electrode at open circuit undergoes following anodic and cathodic reactions 5 o )( 3 eNiOHLaOHLaNi (2.10) 2 o HOHeOH 2 (2.11) When the MH alloy is partially oxidized to lanthanum oxide, the amount of active material is considerably reduced. As a result the rate of self-discharge rate increases. The self-discharge behavior of the LaNi5 system determines the capacity loss during storage in open-circuit conditions and can be divided into two parts namely, reversible and irreversible. The reversible capacity loss is due to the chemical desorption of hydrogen from the MH electrode and the irreversible capacity loss is caused by the degradation of the hydrogen storage alloy. 2.4.4 High rate charge-discharge capabilities Although Ni-MH batteries are commercially available, further research is required to improve the high rate charge-discharge capabilities (HRC and HRD) regarding their use in electrical and hybrid vehicles. Figure 2.11 shows a solar vehicle used for Ni-MH study, 32 Literature review Japan. HRD and HRC have a great effect on the closer contact between active material and current collector [52]. Figure 2.11: A solar vehicle used for Ni-MH study, Japan [53] Charge efficiency is the capacity obtained during discharge as a percentage of the charge capacity. To conducting charge efficiency test, two sample electrodes H149 (composition La0.7Ce0.3Ni3.55Al0.3Co0.75Mn0.4) and H153 (composition La0.6Ce0.3Zr0.1Ni3.55Al0.3Co0.75Mn0.4) were charged at C/4, C/2, C, 2C and 3C to full charge and then discharged at the C/2 rate. This experiment was performed after completely activated the electrodes. Figure 2.12 shows the charge efficiency of these two samples as a function of charge rate. The charge efficiency is higher than 80% at the 3 C rate for all electrodes. 33 Literature review Figure 2.12: Charge efficiency of the MH alloy sample [54] Figure 2.13 shows that the discharge capacity of the LaNi5 was comparable to that of La0:85MgxNi4:5Co0:35Al0:15 alloys. It is known that the high rate charge-discharge capabilities are determined by the charge transfer reaction at the alloy surface as well as the nature of the electrode additives and the amount of the active materials [54]. It is generally observed that high rate discharge ability can be improved by the changing the composition of alloys. 34 Literature review Figure 2.13: High rate discharge capability of the different electrodes [16]. 2.5 PCI measurements A powerful and popular way to measure the equilibrium hydrogen content of a hydrogen storage alloy under different hydrogen partial pressure is Pressure Composition Isotherm (PCI) method. As the pressure of the hydrogen slightly increases the absorbed hydrogen atom begin to occupy the interstitial sites in the metal lattice and forms a solid solutions. PCI curve is based on the Sievert’s law (equation 2.12) and van’s Hoff’s equation (equation 2.13). The amount of hydrogen absorbed or released, in equilibrium, in the alloy can be calculated by the pressure existing in a closed system with a constant volume 1 u u PKH 2 (2.12) Solid H2 ln '' // RSRTHP (2.13) H 2 35 Literature review Where Hsolid is the concentration of the hydrogen absorbed in alloy, PH2 is the equilibrium potential of hydrogen, R is the gas constant and K is the equilibrium constant, ǻH is the enthalpy and ǻS is the entropy of hydrogenation reaction [55]. 36 Literature review References [1] Seth Dunn, “Hydrogen futures: toward a sustainable energy system”, International Journal of Hydrogen Energy 27 (2002) 235-264. [2] Hui-Ming Cheng, Quan-Hong Yang, “Hydrogen storage in CNTs”, Carbon 39 (2001) 1447-1454. [3] Sumio Iijima, “CNTs: past, present and future”, Physica B 323 (2002) 1-5. [4] A. Zuttel, P. Sudan, “Hydrogen storage in carbon nanostructures”, International Journal of Hydrogen Energy 27 (2002) 203-212. [5] Arindam Sarkar, Rangan Banerjee, “A quantitative method for characterization of CNTs for hydrogen storage”, International Journal of Hydrogen Energy 29 (2004) 1487-1491. [6] www.pa.msu.edu/cmp/csc/nanotube.html, 19 January 2005. [7] J. Lv, J. P. Tu, “Effects of CNTs on the high rate discharge properties of nickel/metal hydride batteries”, Journal of Power Sources 132 (2004) 282-287. [8] http://en.wikipedia.org/wiki/carbon_nanotube, 19 January 2005. [9] Mustafa Musameh, Nathan S. Lawrence, “Electrochemical activation of CNTs”, Electrochemistry Communications 7 (2005) 14-18. [10] F. Lamari Darkinm, P. Malbrunot, “Review of hydrogen storage by absorption in CNTs”, International Journal of Hydrogen Energy 27 (2002) 193-202. [11] L. B. Wang, J. B. Wang, “An electrochemical investigation of Mg1-xAlxNi(0 hydrogen storage alloys”, Journal of Alloys and Compounds 385 (2004) 304-308. [12] Seung Mi Lee, Ki Soo Park, “Hydrogen adsorption and storage in CNTs”, Synthetic Metals 113 (2000) 209-216. 37 Literature review [13] H. Takagi, H. Hatori, “Adsorptive hydrogen storage in carbon and porous materials”, Materials Science and Engineering B108 (2004) 143-147. [14] P. Sudan, A. Zuttel, “Physisorption of hydrogen in single-walled CNTs”, Carbon 41 (2003) 2377-2383. [15] N. Rajalakshmi , K.S. Dhathathreyan, “Electrochemical investigation of single- walled carbon nanotubes for hydrogen storage”, Electrochimica Acta 45 (200) 4511-4515. [16] Andreas Züttel, Pascal Wenger, “Hydrogen density in nanostructured carbon, metals and complex materials”, Materials Science and Engineering B108 (2004) 9- 18. [17] Patrice Guay, Barry L. Standsfield, “On the control of carbon nanostructures for hydrogen storage applications”, Carbon 42 (2004) 2187-2193. [18] Brian Riestenberg, “Hydrogen Storage in CNTs For the purposes of pollution free vehicular fuel cells and secondary batteries”, 23 April 2002. [19] John Robertson, “Realistic applications of CNTs”, http://www.materialstoday.com/pdfs_7_10/Robertson.pdf, 21 February 2006. [20] S. Orimo, A. Zuttel, “Hydrogen interaction with carbon nanostructures: current situation and future prospects”, Journal of Alloys and Compounds 356-357 (2003) 716-719. [21] Kuochih Hong, “The development hydrogen storage electrode alloys for nickel hydride batteries”, Journal of Power Sources 96 (2001) 85-89. [22] P. Dantzer, “Properties of intermatallic compounds suitable for hydrogen storage applications” Materials Science and Engineering A329-331 (2002) 313-320. 38 Literature review [23] N. Cui, J. L. Luo, “Nickel-metal hydride (Ni-MH) battery using Mg2Ni-type hydrogen storage alloy”, Journal of Alloys and Compounds 302 (2000) 218-226. [24] Paul Ruetschi, Felix Meli, “Nickel-metal hydride batteries. The preferred batteries of the future?”, Journal of Power Sources 57 (1995) 85-91. [25] Ernest Worthman, “Batteries and battery technology for a wireless future”, http://rfdesign.com/images/archive/0900Worthman40.pdf, 21 February 2006. [26] Xinbo Zhang, Yujun Chai, “Crystal structure and electrochemical properties of rare earth non-stoichiometric AB5-type alloy as negative electrode material in Ni- MH battery”, Journal of Solid State Chemistry 177 (2004) 2373-2377. [27] H.Y. Park, I. Chang, “Electrode characteristics of the Cr and La doped AB2-type hydrogen storage alloys”, International Journal of Hydrogen Energy 26 (2001) 949- 955. [28] T. Mouri, H. Iba, “Hydrogen absorbing alloys with a large capacity for a new energy carrier”, Materials Science and Engineering A329-331 (2002) 346-350. [29] B. Knosp, L. Vallet, “Performance of an AB2 alloy in sealed Ni-MH batteries for electric vehicles: quantification of corrosion rate and consequences on the battery performance”, Journal of Alloys and Compounds 293-295 (1999) 770-774. [30] Mingming Geng, Jianwen Han, Feng Feng, “Charging/discharging stability of a metal hydride battery electrode”, Journal of the Electrochemical Society 146 (7) 2371-2375 (1999). [31] Han Shumin, Zhang Zhong, “The effect of vacuum evaporation plating on phase structure and electrochemical properties of AB5-5 mass% LaMg3 composite alloy”, Electrochimica Acta 50 (2005) 5491-5495. 39 Literature review [32] J. R. Ares, F. Cuevas, “Influence of thermal annealing on the hydrogenation properties of mechanically milled AB5 type alloys”, Materials Science and Engineering B 108 (2004) 76-80. [33] Chan-Yeol Seo, Seung-Jun Choi, “Effect of vanadium content on electrochemical properties of la-based AB5 type metal hydride electrodes”, International Journal of Hydrogen Energy 28 (2003) 967-975. [34] A. Merzouki, C. Cachet-Viver, “Microchemistry study of metal hydride battery materials cycling behaviour of LaNi3.55Mn0.4Al0.3Co0.75 compared with LaNi5 and its mono-substituted derivatives”, Journal of Power Sources 109 (2002) 281-286. [35] Y. Nakamuru, K. Oguro, “X-ray diffraction peak broadening and degradation in LaNi5 based alloys”, International Journal of Hydrogen Energy 25 (2000) 531-537. [36] A.M. Hermann, P.A Ramakrishnan, “Metal hydride batteries research using nanostructured additives”, International Journal of Hydrogen Energy 26 (2001) 1295-1299. [37] F. Haschka, W. Warthmann, “Alkaline batteries for hybrid and electric vehicles”, Journal of Power Sources 72 (1998) 32-36. [38] F. Feng, M. Geng, “Electrochemical behabiour of intermetallic based metal hydrides used in Ni/metal hydride batteries: a review”, International Journal of Hydrogen Energy 26 (2001) 725-734. [39] M. Ben Moussa, M. Abdellaoui, “Effect of substitution of Mm for La on the electrochemical properties of the LaNi3.55Mn0.4Al0.3Co0.75 compound”, Journal of Alloys and Compounds 399 (2005) 264-269. [40] http://www.buchmann.ca/chap2-page2.asp, 8 February 2006. 40 Literature review [41] Mingming Geng, Jianwen Han, “Electrochemical measurements of a metal hydride electrode for the Ni/MH battery”, International Journal of Hydrogen Energy 25 (2000) 203-210. [42] W. B. Gu, C. Y. Wang, “Modeling discharge and charge characteristics of Nickel- metal hydride batteries”, Electrochimica Acta 44 (1999) 4525-4541. [43] M. Geng, F. Feng, “Charge transfer and mass transfer reactions in the metal hydride electrode”, International Journal of Hydrogen Energy 26 (2000) 165-169. [44] Xiaojuan Fu, Haiyan Zhang, “The effect of CNTs on the electrochemical hydrogen storage performance of LaNi5 rare earth alloy”, Physica E 25 (2005) 414-420. [45] Laure Le Guenne, Patrick Bernard, “Life duration of Ni-MH cells for high power applications”, Journal of Power Sources 105 (2002) 134-138. [46] Haiyan Zhang, Xiaojuan Fu, “The effect of MWNTs with different diameters on the electrochemical hydrogen storage capability”, Physics Letters A 339 (2005) 370- 377. [47] Shuangping Yi, Haiyan Zhang, “The electrochemical properties of LaNi5 electrodes with multi-walled CNTs synthesized by chemical vapor deposition and treated at different temperatures in a nitrogen atmosphere”, Physica B 373 (2006) 131-135. [48] Shuangping Yi, Haiyan Zhang, “The study of electrochemical properties of CNTs- LaNi5 electrodes”, Materials Science and Engineering B (2006). [49] F. Feng, D. O. Northwood, “Self-discharge characteristics of a metal hydride electrode for Ni-MH rechargeable batteries”, International Journal of Hydrogen Energy 30 (2005) 1367-1370, 41 Literature review [50] Peter Kritzer, “Separators for Ni-MH and Ni-Cd batteries designed to reduce self- discharge rates”, Journal of Power Sources 137(2004) 317-321. [51] Chigyu Jeong, Wonsub Chung, “Effect of temperature on the laves phase alloy used in nickel/ metal-hydride batteries”, Journal of Power Sources 79 (1999) 19-24. [52] M.L. Soria, J. Chacon, “Metal hydride electrodes and Ni/MH batteries for automotive high power applications”, Journal of Power Sources 102 (2001) 97-104. [53] H. Hoshino, H. Uchida, “Preparation of a nickel-metal hydride (Ni-MH) rechargeable battery and its application to a solar vehicle”, International Journal of Hydrogen Energy 26 (2001) 873-877. [54] Wenlin Zhang, Arnaldo Visintin, “Investigation of changes in morphology and elemental distribution in metal hydride alloys after electrochemical cycling”, Journal of Power Sources 75 (1998) 84-89. [55] Zhuang Hui-zhen, “Study on the micro structures and hydrogenation properties of Ti-Zr based hydrogen storage alloy”, M.Sc thesis, Taiwan University, 1999. 42 Experimental details Chapter 3 Experimental details 3.1 Introduction This study had two main objectives. The first objective was to investigate the differences in the performance of the hydrogen storage electrodes fabricated using different compositions of the CNTs and to improve the characteristics of Ni-MH battery system. The properties of these Ni-MH batteries studied are the initial quick activation rate, high charge-discharge capabilities and cycle life. The second objective was to investigate the equilibrium hydrogen content of specimens at different hydrogen partial pressures using pressure composition isotherm (PCI) method. 3.2 Sample details The overall electrochemical performance of the Ni-MH battery depends on the composition of the active material alloy, electrolyte concentration, binder and electrode fabrication techniques. 3.2.1 Materials Commercial MmNi5 alloys were chosen as experimental active material in negative electrode for Ni-MH battery. Electrochemical characterization such as activation test, 43 Experimental details charge-discharge characteristics, cycle life test, high rate capability test, state of charge test, depth of discharge test and self-discharge test were carried out at room temperature. MmNi5 possesses good oxidation and corrosion resistance in alkaline medium as the Ni- MH battery in oxidizing medium contains highly alkaline electrolyte. Since many elements react to form oxides in electrolyte, it follows if these elements were used as electrodes, they will introduce oxidation and fail to store hydrogen. In this aggressively oxidizing environment, protection from oxidation and corrosion is very important. The duration of the battery strongly depends on the oxidation and corrosion resistance of the electrode materials [1-2]. The hydrogen storage alloy used in this study in the negative electrode had the composition of La0.58Ce0.25Pr0.06Nd0.11Co0.74Mn0.41Al0.18 [3]. Nickel (Ni) is used as the conductor for the MH electrode because of its microporous structure that helps increase the effective active area and conductivity. In this study, pure elemental Ni powder (200 mesh, 99.5% purity) was mixed with MmNi5 alloy by mechanical grinding. Ni also provides necessary contact between alloy powders and provides good electrical and thermal conduction [4]. CNTs (95% purity) obtained from China, were mixed with MmNi5 alloy because of their excellent hydrogen absorption capability. The cylindrical pore structure of CNTs indicates that they are likely to be ideal gas reserves. CNTs also have strong oxidation resistance [5]. Hydrogen can be absorbed by the CNTs in two ways, Physisorption and chemisorption. Physisorption of hydrogen dominates when CNTs trap hydrogen molecule inside the cylindrical structure of the nanotubes. Chemisorption occurs by hydrogen dissociation and reaction with carbon nanotubes [6]. 44 Experimental details Ball milling method is often used for producing nano-crystalline or amorphous alloys [7]. It is observed that after ball milling hydrogen storage materials showed a higher capacity [8]. Ball milling also increases specific surface area [9]. In order to investigate the hydrogen absorption-desoprtion behavior of CNTs, treatment has been carried out through ball milling for 1 hour. Ball milling was performed using a Rocklabs rotation mill for 1 hour in the school of Materials Science and Engineering, UNSW. The separator must be characterized by a low internal resistance to achieve high rate performances. As the separator, polypropylene non-woven materials grafted with acrylic monomers with hydrophilic properties was used in this study. The separator retains the electrolyte that acts as the channel for ion exchange between the positive and negative electrodes [10]. It acts as an electronic insulator to prevent short circuit between the electrodes. Small pores of the separator help to avoid short-circuit, high electrolyte absorption, good wettability and chemical stability in alkaline electrolyte [11]. During the entire experiments care has been taken to ensure that the separator was not dry out. A suspension of Polytetrafluorethylene (PTFE) was used as binder of the powders to avoid active material shedding after repeated charge-discharge cycles [12]. Both negative and positive electrodes were prepared with 3-dimentional Ni substrate in order to reduce electrode resistance [11]. Ni substrate is suitable for high power applications due to both high Ni content and small pore. These factors allow close contact between the current collector and the active materials during high rate charge-discharge rates [10]. 45 Experimental details 3.2.2 Experimental electrodes Different types of negative and commercial positive electrodes were used for electrochemical studies. Commercial positive electrode The commercial positive electrodes were obtained from a commercial supplier, in Taiwan. Prepared negative electrodes These electrodes were prepared using as received Ni powder, AB5 alloy, CNTs and PTFE dispersion. Prepared negative electrode with carbon nanotubes In this study MWNTs with diameter 20-40 nm were mixed with AB5 alloy to improve the battery performance. The CNTs were obtained from a commercial supplier, in China. Direct addition of CNTs into the negative hydrogen storage alloy is rather simple and quite economical. Electrodes were made using different amounts of CNTs to test the improved performance of Ni-MH battery system: Following percentages of CNTs were used in this investigation: 0.01 g CNTs (~ 2% of active materials) 0.03 g CNTs (~5% of active materials) 0.06 g CNTs (~10% of active materials) 0.09 g CNTs (~15% of active materials) Ball milled 20-40 nm CNTs (1 hour) 46 Experimental details 3. 3 Sample characterization techniques Scanning Electron Microscopy (SEM) Scanning electron microscopy (SEM) was used to characterize the microstructure of as received powders and alloys, determine the average particle size of the materials and the surface morphology of different electrodes. SEM was also used to observe the morphology of the negative electrodes after electrochemical experiments. In SEM a double-sided tape was placed and then sample was placed on top of the tape. Depending on the conductivity requirements, some specimen required carbon, chromium or gold coatings. Voltage ranging from 5-20 KV was used for resolution of the images at different magnification range depending on the samples. From recorded photographs, average particle sizes of alloys and powders were measured directly. SEM Hitachi S4500 and SEM Cambridge S900 scanning electron microscopes were used for SEM analysis at the Electron Microscope Unit at UNSW. Composition of the AB5 sample was examined using an electron diffraction spectroscopy (EDS) coupled with SEM machine, at the electron microscope unit, UNSW. The Electron probe micro analysis (EPMA) is the most powerful instrument for microanalysis of materials. Compositional information of a sample can be obtained by characteristic X-ray analysis. This instrument is capable of obtaining X-ray scanning picture that shows the elemental distribution in the area of interest. The EPMA analysis of the AB5 alloy was examined using the Cameca SX50 microprobe in the electron microscope unit, UNSW. 47 Experimental details X-ray Diffraction (XRD) The XRD analyses were used for phase identification, lattice parameter determination. A Siemens D5000 X-ray diffractometer was used for XRD analysis of the samples at the school of Materials Science and Engineering, UNSW. 30 KV acceleration voltage and 30 mA current was used for graphite monochromated CuKĮ radiation along with computer to count and display the diffraction pattern. The angular range scanned was 5˚ to 110˚ with a 0.02˚ scanning step size. Phase identification and lattice parameter calculations were accompanied through comparison of the results with data from database. Diffraction file cards were used to identify the diffraction peaks. Transmission Electron Microscopy (TEM) The most frequently used tool to characterize CNTs is the Transmission Electron Microscope. The structure of nanotubes was obtained by direct physical observation using TEM analysis. Generally TEM images can be used to express more detailed structure down to a few angstroms. A TEM Hitachi H7000 machine was used for physical observation of CNTs at the electron microscope unit, UNSW. The average particle size of CNTs was measured directly from the recorded photograph. Raman spectra Raman spectrum is a useful tool for understanding structure and composition of the materials. Raman spectra of the CNTs were obtained using a Renishaw 1000 Raman spectrometer focused through an Olympus optical microscope at the school of Materials Science and Engineering UNSW. The spot size at 50x magnifications is approximately 5 µm. The excitation light was obtained from green laser beams, 514 nm. 48 Experimental details Chemical analysis Chemical analysis of the material was done by the induced coupled plasma spectroscopy (ICP) from the school of Chemistry, UNSW. 3. 4 Preparation of electrodes Preparation of negative electrode The negative electrode was prepared using the pasting method. Table 3.1 shows the different materials used to prepare this experimental negative electrode. For pasting electrodes, alloy powder are mixed thoroughly with Ni powder and 0.96 g alcohol (50% water solution) is added for dispersion and 0.17g PTFE (3 wt% of the total weight of paste) was used as a binder. The mixing of the Ni powder with AB5 alloy increases the electrical conductivity and reaction kinetics. The paste was then placed on a foamed nickel substrate (diameter of paper core~152/300 nm) and mechanically pressed under 15 MPa for 30 minutes on both sides after folding the nickel foam, Figure 3.1. Table 3.1: Materials used in the electrochemical experiments. Materials Weight Ni powder 1.2 g MH alloy 0.6 g Alcohol-distilled water solution ratio 1:1 0.72 g Ni foam 30x30x1 mm3 Polytetrafluoroethylene 0.09 g CNTs 0.01-0.09 g 49 Experimental details The electrodes were dried in a vacuum oven at 117˚C for four hours before commencing electrochemical measurements. This was used as negative electrode in the experimental test cell, the photograph of the negative electrode is shown in Figure 3.2. Experimental flow chart for the preparation of the negative electrode is shown in Figure 3.3. Negative electrode Foam Alloy powder after pressing Nickel plate pasted on both sides Figure 3.1: Schematic diagram of experimental negative electrode. Figure 3.2: Photograph of the prepared negative electrode. 50 Experimental details MH powder Ni powder PTFE CNTs 0.6 gm 1.2 gm 5 wt% 0.01-0.09 gm Mixing of the materials by mechanical grinding Loading of paste on Ni- foam (155 MPa) Vacuum dried at 110°C for 4 hours Figure 3.3: Flow chart of the preparation of pasted type negative electrode. The negative electrode with the CNTs was made by the procedure discussed above by mixing CNTs homogeneously by mechanical grinding. In order to investigate the effect of CNTs, the proportion of CNTs was increased gradually. Improved positive electrode Sintered Ni(OH)2/NiOOH plates were used as positive electrodes. The active material Ni(OH)2 is semiconductor material, which is different from the conductive hydrogen storage alloy. The photograph of the commercial positive electrode is shown in Figure 3.4. 51 Experimental details Figure 3.4: Photograph of as received positive electrode. It is well known that during the charge-discharge process of the positive electrode, the electrode is expanding and shrinking, which reduces mechanical strength, result in higher electric resistance and capacity degradation [13]. To avoid the above shortcomings and to prevent the swelling of the positive electrode, the surface of the positive electrode were sandwiched between conductive Ni foam (thickness of about 1 mm), then pressed under 15 MPa for 5 minutes to form the improved positive electrode. Figure.3.5 shows the schematic diagrams of improved and original positive electrodes. Paste Nickel foam substrate Original electrode Improved electrode Conductive nickel foam layer Figure 3.5: The schematic diagram of the improved and original positive electrodes. 52 Experimental details 3.5 Current collector Different Ni: nickel mesh, nickel wire (diameter 0.5mm) and nickel foam were used as the current collector for the negative electrodes. In most cases, for positive electrode current collector was cut from the positive electrode itself, i.e., a small portion of the positive electrode was used as current collector. 3.6 Electrochemical test cell setup In this study, a Ni-MH test cell was designed with one negative and two positive electrodes. The objective of this set up was to test the electrochemical stability of the negative electrode. The dimensions of positive electrodes were 2mm larger than that of negative electrodes (30ɯ30ɯ2 mm3). Therefore the capacity of the positive electrodes was expected to be higher than that of the negative electrode. The cells were then assembled with one negative electrode sandwiched between two positive electrodes and separated by separator, as shown in Figure 3.6 and placed in an open cell. The approximate distance between negative and positive electrodes was about 5 mm. Figure 3.6: Schematic diagram of cell assembly. 53 Experimental details 3.7 Electrolyte soaking of electrodes The assembled electrodes were treated with electrolyte for 24 hours and stored in a desicator at room temperature. The composition of the electrolyte was as follows Composition: 6M KOH + 1 wt% LiOH This alkaline pre-treatment is introduced for not only dissociation of water but also for recombination of the water on the electrode surface. The possible explanations of these phenomena are that the initial hydrogen charge accelerated by the alkaline treatment activates the electrode surface and subsequent hydrogen discharge also becomes activated [14]. After pre-treatment of electrodes, the electrodes are now ready to perform the electrochemical test. 3.8 Apparatus for electrochemical measurements Electrochemical measurements on the electrochemical cell were carried out using REPOWER charging-discharging machine interfaced with a computer using CTO_EN.EXE cell test MFC Application software, Figure 3.7. Figure 3.7: Photograph of charge-discharge REPOWER machine. 54 Experimental details 3.9 Electrochemical testing parameters for negative electrodes The electrochemical measurements include maximum capacity of the cell, charge-discharge cycles and specific discharge capacity. Through these measurements, the electrochemical characteristics of the Ni-MH battery can be evaluated. The measurements of the electrochemical properties were carried out at room temperature in an open cell. The sample electrodes were cyclically charged and discharged for several cycles until the discharge capacity became reproducibly stable. Parameters used in this experiment (10% CNTs electrode) Theoretically estimated capacity used in these experiments= 250 mAh/g Active material of the 10% CNTs electrode= total weight of the AB5 alloy+ total weight of the CNTs = 0.6 + 0.06 g = 0.66 g Actual capacity of the 10% CNTs electrode= 250x0.66 mAh/g= 165 mAh/g Charge discharge rate, 0.1 C= 165x0.1= 17 mA 1 C rate= 165x1= 165 mA Specific discharge capacity of the active material= Discharge capacity obtained from charge discharge tests/ mass of the active material of the MH electrodes. 55 Experimental details 3.10 Experimental procedure In this research, in order to identify the crystal structure characteristics of the as received samples XRD analyses were performed. The surface analyses of the samples were observed by SEM equipped with EPMA. Gas phase hydrogen absorption-desorption measurements of the samples were carried out in a modified Sieverts type apparatus. To ensure good homogeneity, AB5 Alloy powders and CNTs were crush and ground manually in an agate mortar and pestle to prepare the negative electrode of the Ni-MH battery. A slurry containing ground metal hydride powder, CNTs and PTFE were pasted into foamed Nickel substrate. This Ni-MH electrode placed on to a current collector at room temperature and pressed and dried at 110 C for 4 hours to obtain experimental negative metal hydride electrode. As received positive electrode was folded by Ni foam and pressed to increase conductivity. The Ni-MH experimental test cell was then assembled with one prepared negative electrode sandwiched between two positive electrodes separated by a polypropylene separator in a flooded open cell. These experimental electrodes were immersed in 6M KOH solution for 1 day at room temperature. After soaking the electrochemical test cell are ready to perform several electrochemical tests, shown in Figure 3.8. For electrochemical measurements, the electrochemical tests were conducted with a charging-discharging machine to determine the charge-discharge characteristics curves and the maximum electrochemical capacity of all electrodes. The electrochemical test cells were activated through several (first couple of cycles) charge-discharge cycles to obtain 56 Experimental details reversible and stable charge-discharge characteristics. Afterwards, individual electrochemical tests were performed by individual test cells, shown in Figure 3.9. Physical Fabrication of Production of characterization of MH electrode Cell Ni-MH battery as-received activation samples PCI Electrochemical measurements Tests Figure 3.8: Flow chart of the Experimental tests of the negative electrodes. Activation test Maximum Cycle High rate High rate Self-discharge, capacity life test charge discharge state of charge, ability test capability depth of test discharge test Figure 3.9: Experimental procedure of the electrochemical measurements. Activation and cycle life characteristics To stabilize the charge-discharge capacity it takes about 10 cycles to fully activate the Ni- MH cell in these experiments. The electrodes were charged and discharged at 0.1 and 0.2 C rates respectively for 10 cycles to active the cells. 57 Experimental details Cycle life is the duration of the cell after which the battery can not be used any more because of the loss of its capacity. Different electrodes cyclic charge and discharge treatments were made at a short interval of 2 minutes. Table 3.2 shows the experimental procedure for activation and cycle life test. Table 3.2: Parameters and procedure for activation and cycle life test. Cut-off volt= 1V Electrode Charge- Charge Discharge with Cycles discharge current, current, %CNTs rate, C mA mA 1-5 0.1 7 7 0% and 2% >5 0.2 14 14 1-5 0.1 8 8 5% and 10% >5 0.2 16 16 15% and 1-5 0.1 9 9 20% >5 0.2 18 18 Discharge capacity The discharge capacity is a very important characteristic, which is calculated from discharge current and discharge time. The unit for discharge capacity is mAh. The negative electrode was charged initially under a 0.1 or 0.2 C rate. C rate is the charging or discharging current in mA flowing per unit time. After a short interval, the discharge process was conducted using the same rate. 58 Experimental details The discharge capacity for each cycle was calculated according to the following equation QIdischarg e t Where Q is the discharge capacity in mAh, I is the discharge current in mA and t is the discharge time in hours. The cut-off voltage was 1.00 V. Depending on the experiment several charge-discharge cycles were conducted for each electrode. By repeated charge- discharge cycles, the decay in the capacity can be found and the cycle life of the electrode can be obtained. Specific discharge capacity The specific discharge capacity of the electrode was calculated from capacity obtained from the experiments by the following equation Specific discharge capacity=discharge capacity/ weight of the active materials. The unit of specific discharge capacity is mAh/g. Self-discharge, state of charge, depth of discharge test SOC, DOD and SD test were performed by the following sequence with only one corresponding electrode: activation test, SOC, high rate discharge, DOD and SD test. SOC is the proportion of the capacity input to a cell which is subsequently available on discharge. An estimation of SOC assists optimum utilization of Ni-MH battery for a given application, e.g. an evaluation of its state of health. In this investigation, the cell was monitored at five different states of charge, namely 20, 40, 60, 80 and 100 percent state of charge. 59 Experimental details DOD is the ratio of the charge removed from a battery when its capacity is full. The typical end of discharge voltage is 1.0 V for Ni-MH battery. A battery often gets many short discharges with subsequent recharges. This behavior is similar to the charge efficiency. SOC and DOD experiments were conducted using different percentage of charge time and discharge time shown in Table 3.3 and 3.4 respectively. Table 3.3: Experimental procedure and parameters for SOC test. Cut-off volt= 1V Experiments Charge-discharge rate, C % of full charge time 100 80 State of charge test 0.2 60 40 20 Table 3.4: Experimental procedure and parameters for DOD test. Cut-off volt= 0 V Experiments Charge-discharge rate, C % of full discharge time 100 80 Depth of discharge test 0.2 60 40 20 60 Experimental details Self-discharge is the loss of capacity of a fully charged cell while standing at an open circuit. It is dominated by the slow decomposition of positive/negative electrode materials because of their intrinsic instability and also due to the reaction of hydrogen with the nickel substrate. The self-discharge was measured by first charging the electrodes at 0.1 C rate and keeping it at rest for 2 days and then discharging. After that discharge capacity is measured. Similarly self-discharge rate was checked after maintaining cells for 4, 6 and 8 days under open circuit at room temperature for 10% and 0% CNTs electrode during the available time frame. The procedure for self-discharge test is shown in Table 3.5. Table 3.5: Experimental procedure and parameters for self-discharge test. Cut-off volt= 1 V Experiments Charge-discharge rate, C Storage time, days 2 4 Self-discharge test 0.2 6 8 High rate capability test The experimental procedure and parameters for high rate discharge capability and high rate charge ability are shown in Table 3.6 and 3.7 respectively. 61 Experimental details Table 3.6: Experimental procedure and parameters for high rate discharge capability test. Discharge Cut-off Experiments No of cycles Charge rate, C rate, C voltage, V 1-3 0.1 0.1 Activation test 1 4 -10 0.2 0.2 11 High rate 2 0.8 >11 0.2 discharge test 3 0.6 4-10 0.5 Table 3.7: Experimental procedure and parameters for high rate charge ability test. Cut-off volt= 1 V Charge Discharge Experiments No. of cycles Charge rate, C time, rate, C min 1-3 0.1 0.1 600 Activation test 4-10 0.2 0.2 300 160 High rate charge test >11 230.2 0 3 20 62 Experimental details Pressure composition isotherm measurements In order to investigate the hydrogen absorption characteristics of hydrogen storage materials, PCI curves were obtained using the conventional Sieverts-type apparatus, Figure 3.10. Hydriding-dehydriding experiments for PCI were conducted at room temperature. The hydrogen absorption-desorption of a sample was calculated by the following relationships PV nRT Where P is the pressure in the reactor, V is the volume in m3 of the reactor, R is the universal gas constant, T is the absolute temperature of the reactor in Kelvin and n is the number of moles of hydrogen. Ar2 1 S1 4 Handle D S4 S3 C 6 HPu2 57810 S2 E A B a container Gas Chamber container Gas Figure 3.10: Sketch map of PCI system for hydrogen storage materials used in this experiment [15]. 63 Experimental details The sample MmNi5 and ball milled CNTs were used in this investigation. About 0.5g of the sample was weighed and placed in the reactor. High purity Argon gas along with vacuum pump used to purify the PCI system. Then high purity H2 gas was introduced into the reactor and kept a high pressure (~15 atm) in the reactor for 3 hours at 350° C to activate the sample and followed a desorption process at a high temperature (390° C) to get rid of the absorbed hydrogen from the samples. The PC isotherms were recorded after activation treatment this activation-desorption process was repeated three times. The absorption isotherm measurements were started when the sample was completely desorped. At last one hour was provided for equilibrium pressure to be obtained. To get an average absorption-desorption data several data were recorded within the range of 10-20 atm by setting the gas gauge. This procedure was repeated several times. 64 Experimental details References [1] Charles K. Witham, “The effect of alloy chemistry on the electrochemical and hydride properties on Ni-substituted LaNi5”, PhD thesis, California 2000, Institute of Technology, Pasadena, California. [2] A. Merzoukia, C. Cachet-Vivier, “Microelectrochemistry study of metal-hydride battery materials cycling behavior of LaNi3.55Mn0.4Al0.3Co0.75 compared with LaNi5 and its mono-substituted derivatives,” Journal of Power Sources 109 (2002) 281– 286. [3] Jeremy Kushner, “Study on gas/solid reactions at two different temperatures and on electrochemical reactions at four different discharge rates of a commercial AB5 hydrogen storage alloy”, Honors thesis, UNSW 2005. [4] M. M. Shaijumon, N. Bejoy, S. Ramapradhu “Catalytic growth of Carbon nanotubes over Ni/Cr hydrotalcite-type anionic clay and their hydrogen storage properties”, Applied Surface Science 242 (2005) 192-198. [5] M. M. Shaijumon, S. Ramaprabhu, “Studies of yield and nature of carbon nanostructures synthesized by pyrolysis of ferrocene and hydrogen absorption studies of carbon”, International Journal of Hydrogen Energy 30 (2005) 311-317. [6] Haiyan Zhang, Xiaojuan Fu, Jianfeng Yin, Chun Zhou, “The effect of MWNTs with different diameters on the electrochemical hydrogen storage capacity”, Physics Letters A 339 (2005) 370-377. [7] Myoung Youp Song, Dongsu Ahn, “Development of AB2 -type Zr–Ti–Mn–V–Ni–Fe hydride electrodes for Ni–MH secondary batteries”, Journal of Alloys and Compounds 298 (2000) 254–260. 65 Experimental details [8] Hao Niu, “Electrochemical performance of ball milled Mg2Ni electrodes”, M.Sc thesis, Windsor, Ontario, Canada, 1999. [9] Jianwen Han, “Electrochemical properties of metal hydride electrodes”, PhD thesis, Windsor, Ontario, Canada 2000. [10] M. L. Soria, J. Chacon, “Metal hydride electrodes and Ni/MH batteries for automotive high power applications”, Journal of Power Sources 102 (2001) 97-104. [11] M. Luisa Soria, Joaquin Chacon, “Nickel metal hydride batteries for high power applications”, Journal of Power Sources 96 (2001) 68-75. [12] Deyi Yan, Weiguo Cui, “Preparation and properties of no-binder electrode Ni-MH battery”, Journal of Alloys and Compounds 293-295 (1999) 780-783. [13] J.B. Wu, J. P. Tu, “High rate dischargeability enhancement of Ni/MH rechargeable batteries by addition of nanoscale CoO to positive electrodes”, Journal of Power Sources (2005). [14] H. Hoshino, H. Uchida, “Preparation of a nickel-metal hydride (Ni-MH) rechargeable battery and its application to a solar vehicle”, International Journal of Hydrogen Energy 26 (2001) 873-877. [15] Perfect communication with Dr. Zhongmin Wang, Visiting research fellow, UNSW, Australia. 66 Experimental results and discussion Chapter 4 Results and discussion 4.1 Introduction A number of techniques were used to characterize the samples. The surface morphology of the samples was determined using Scanning electron microscope (SEM), Electron diffraction spectroscopy (EDS), Transmission electron microscopy (TEM) and Electron probe micro analysis (EPMA). The crystal structure of the materials was obtained by XRD analysis. The chemical analysis of the MH alloy was carried out using Induced coupled plasma (ICP) analysis. The electrochemical experiments on MH negative electrodes were characterized by galvanostatic cycling test. In the following section, experimental results are presented along with discussion. 4.2 Surface morphology SEM and TEM images were used to determine the type of nanotubes and to verify their structure. 4.2.1 SEM, EDS, EPMA and chemical analysis Prepared electrodes The porous microstructures of the 0% and 5% CNTs electrodes were studied by SEM analysis, as shown in Figure 4.1. The surface morphology of 0% CNTs electrode is different from that in 5% CNTs electrode. The surface of 0% CNTs electrode looks 67 Experimental results and discussion much smoother than that of 5% CNTs electrode. The dispersion of CNTs within the active materials of the negative electrode was checked by SEM analysis. From Figure 4.2, it can be observed that some areas are covered with carbon nanotubes. Others are likely to be Ni-rich region, implying that carbon nanotubes were not dispersed uniformly in the powders. Further mechanical grounding was required for homogeneous distribution of the CNTs into the particles. Figure 4.1: SEM analysis of 0% CNTs electrode (upper picture) and 5% CNTs electrode (lower picture). 68 Experimental results and discussion Figure 4.2: SEM image of ground 10% CNTs, Ni and AB5 alloy. Metal hydride alloy powder The particle size and composition of the MH alloy powder in the negative electrode and compositions were examined using SEM and EDS analysis. Figure 4.3 shows the micrograph for AB5 alloys. Figure 4.3: The SEM image of as received AB5 alloy 69 Experimental results and discussion The multi component AB5 alloy exhibits a wide particle size distribution from ~2 to 12 µm. The particles are irregular in shape. The average diameter of the AB5 particle is about 900 nm. Figure 4.4 shows the EDS analysis spectrum of the AB5 alloy. Figure 4.4: EDS analysis of as received AB5 alloy. The result of EDS analysis, shown in Table 4.1 is slightly different from that of ICP (Table 4.2) and EPMA (Table 4.3) analysis and one can observe that the composition of MH alloys in table 4.1 does not match well with the nominal composition ( La0.58Ce0.25Pr0.06Nd0.11Co0.74Mn0.41Al0.18 ), because the EDS analysis only represents very local chemical composition. Table 4.1: Chemical composition of AB5 Powder from EDS analysis. Element Element (wt %) Atomic (%) Ne K 11.20 26.66 Al K 0.72 1.28 Cl K .30 0.41 Ca K 0.21 0.25 Sc K 0.05 0.06 70 Experimental results and discussion Cr K 0.75 0.69 Mn K 6.73 5.89 Co K 13.37 10.90 Ni K 72.30 59.15 Electron probe micro analysis The surface elemental distribution of AB5 powder was also studied by an electron probe micro analyzer. Table 4.2 represents the compositional result of AB5 alloy. EPMA analysis also represents very local chemical composition. Table 4.2: Composition of AB5 alloy obtained by EPMA analysis.* Voltage = 15kV, Current = 150nA Element Relative Conc. Magnesium Not detected Aluminium Low Titanium Not detected Vanadium Not detected Chromium Very faint trace Manganese Low-medium Iron Not detected Cobalt Low-medium Nickel Medium-high Barium Not detected Lanthanum Medium Cerium Low Neodymium Strong trace 71 Experimental results and discussion * Please see Appendices for spectrum obtained from EPMA analysis of AB5 alloy. Chemical analysis The chemical composition of the MH alloy powder was determined by induced coupled plasma spectroscopy. Table 4.3 shows the composition of the MH alloy powder. Table 4.3: Composition of MH alloy by ICP analysis (weight %). Ni La Co Mn Al Fe Mg V Others 59 10 11 0.9 2.6 0.034 0.025 0.020 15.4 ICP analysis showed that three constituent elements Ce, Pr and Nd of the alloy were not detected by this facility. From the results of EDS, EPMA and ICP analysis, it can be observed that the result of EPMA analysis shows more accurate result of nominal composition ( La0.58Ce0.25Pr0.06Nd0.11Co0.74Mn0.41Al0.18 ) of the as received MH alloy used in this investigation. Carbon nanotubes (20-40 nm) Figure 4.5 shows the structural analysis of CNTs. The CNTs are featured having an average diameter of 25 nm. Most of CNTs are aligned and packed together to form ropes of several parallel tubes as shown in Figure 4.5. 72 Experimental results and discussion Figure 4.5: SEM image of as received CNTs diameter ranging from 20-40 nm. Separator Figure 4.6 shows the surface morphology of the non-woven polypropylene separator. The surface of the separator consists of long and thick fibers of polymer. Figure 4.6: SEM image of as received polypropylene separator. 73 Experimental results and discussion 4.2.2 TEM analysis of CNTs Microstructure of the CNTs, diameter 20-40 nm was examined using TEM analysis. The TEM images of CNTs are shown in Figure 4.7. It is observed that CNTs have an inner cylindrical hollow cavity structure without any aperture in the tube wall. CNTs which have clearly hollow inside show open ends (providing inner surface) in Figure 4.7 (lower picture), indicates purity and increased hydrogen absorption capacity since hydrogen could enter nanotubes through the open ends. [1]. Open tubes enable gas absorption from both inside and outside the tubes whereas closed tubes do not. Regarding the opening of the tubes, it is essential to ensure that hydrogen molecules can enter the tube because those molecules can not penetrate through the carbon atom hexagonal structural network. Open end make the cavity accessible for the hydrogen gas [2]. At room temperature, opened CNTs in a hexagonal arrangement exhibit higher hydrogen absorption rates than any other carbon materials [3, 4]. 74 Experimental results and discussion Figure 4.7: TEM image of as received CNTs, diameter ranging from 20-40 nm. The tube is curved with a diameter of about 40-60 nm and a hollow core of about 20-40 nm and the length is ~ 900 nm. The tube thicknesses of various nanotubes are different. The inner hollow cavity can serve as a storage medium for hydrogen and is capable of drawing gas inside by capillarity [5]. 4.3 Structure analysis The crystalline structure of the AB5, CNTs and electrodes were examined by XRD techniques. 4.3.1 Crystal structure analysis Crystal structure of AB5, CNTs and electrodes 75 Experimental results and discussion AB5 have a hexagonal structure (Haucke compound). A schematic diagram of AB5 unit cell is shown in Figure 4.8. The alloy unit cell is shown is hexagonal with lattice parameters a=5.017 Å and c=3.987Å [6]. Figure 4.8: Schematic diagram of the crystal cell of AB5 alloy, 1=La, 2=2c Ni and 3=3g Ni [6]. Figure 4.9 shows the XRD patterns of the MmNi5. The sharp peak indicates a long range crystallographic order and good crystallinity [7]. 1 1 185 1 As received metal hydride alloy powder 1 165 0 1 145 1 2 1 0 125 0 1 2 105 1 2 Counts 0 1 2 85 0 2 2 3 0 0 65 0 2 1 45 25 25 35 45 55 65 75 85 95 Two theta Figure 4.9: XRD pattern of as-received AB5 alloy. 76 Experimental results and discussion The XRD of CNTs and ball milled CNTs are shown in Figure 4.10. XRD peaks are centered at 25 and 45 degrees indicate the presence of graphitic crystalline structure of MWNTs; it is quite likely that other peaks are corresponding to catalytic impurities. With increasing diameter CNTs, XRD peak at 45° broadens indicating a deterioration in crystalline long range order and crystallinity [8]. As received 20-40 nm and ball milled CNTs 800 1 600 0 0 0 400 2 1 20-40 nm 200 CNTs 0 Counts 800 600 20-40 nm ball 400 milled CNTs (1hour) 200 0 5 152535455565758595105115 Two theta As received 10-20 nm and 60-100 nm CNTs 800 600 400 200 10-20 nm 0 800Counts 600 400 200 60-100 nm 0 5 152535455565758595105115 Two theta Figure 4.10: XRD diffraction of as received CNTs, diameter ranging from 20-40 nm and ball milled CNTs (1hour) (upper graph) and 10-20 and 60-100 nm CNTs (lower graph). 77 Experimental results and discussion It is observed that the intensity of the peak centered at 25 degree increases with decreasing diameter of CNTs, from 10-20 to 60-100 nm. This phenomenon indicates that the crystalline quality of the graphite structure in the MWNTs becomes poor with an increase in diameter. Diffraction peaks become sharp after ball milling. This might be due to defects introduced into CNTs. Because of contamination of sample container from the ball milling machine, after ball milling the color of whole black CNTs became green [9]. The XRD pattern of the as received Ni powder is shown in Figure 4.11. Sharp and characteristic peaks for Ni are observed in this graph. As received nickel powder 1 3515 1 1 3015 2515 2015 2 1515 0 Counts 0 2 3 1015 2 1 0 1 515 15 25 35 45 55 65 75 85 95 105 Two theta Figure 4.11: XRD diffraction of as received Ni powder. Prepared negative electrodes with CNTs have almost similar structure, shown in Figure 4.12. The XRD analysis of 0% CNTs electrode clearly revealed the sharp peaks of Ni and characteristic peaks of the crystalline AB5 alloy. 78 Experimental results and discussion Prepared CNTs electrode Ni Ni CNTs 400 10% CNTs Ni Ni 200 AB5 electrode 0 400 5% CNTs electrode 200 Counts 0 400 0% CNTs electrode 200 0 525456585105 Two theta Figure 4.12: XRD diffraction of prepared CNTs electrodes. This result shows that the electrodes examined were a single phase corresponding to a hexagonal CaCu5 type structure. At 15-20 degree, the diffraction peaks become sharper for 10% and 5% CNTs electrodes respectively. From this analysis it is likely to be that the peak at 16 degree is for CNTs. 4.3.2 Structure analysis of CNTs MWNTs were also characterized by Raman Spectroscopy, which is a very powerful technique for structure analysis. Figure 4.13 shows the micro Raman spectra of as received CNTs. The spectrum shows two strong peaks at about 1348 and 1578 cm-1. The peak at 1578 cm-1 known as G band from the MWNTs structure and the peak at 1348 cm-1 known as D band which is mainly derived from carbon defects of MWNTs [1, 10-12]. A large ratio of these two peaks indicates that this kind of CNTs posses a 79 Experimental results and discussion better graphitic structure of MWNTs [8, 13]. The Raman spectrum of CNTs indicates that this kind of CNTs is not SWNTs but MWNTs. As received 20-40nm CNTs 205 185 165 145 125 105 85 65 Raman Intensity Raman 45 25 5 200 700 1200 1700 2200 Wave Number (/cm) Figure 4.13: Raman spectrum of as received CNTs diameter ranging from 20-40 nm. 4.4 Electrochemical measurements One each of the prepared electrode with 0%, 2%, 5%, 10%, 15% and 20% CNTs electrode were used for individual test. These experiments include cycle life test, high rate discharge capability, and high rate charge ability test. The state of charge, depth of discharge and self-discharge experiments were carried out using the following sequence: activation test, high rate discharge capability test, depth of discharge test and self-discharge test respectively by the available time frame. All electrochemical experiments were conducted at room temperature. 80 Experimental results and discussion 4.4.1 Activation performance and maximum capacity For good technological performance, Ni-MH battery requires several repeated charge- discharge cycles for fully activation to ensure that the cell is electrochemically stable. This activation phenomenon results in the production of finely divided powders (particles size usually at the order of few dozen of micros or less). [12-14]. Figure 4.14 shows the activation characteristics of Ni-MH batteries with different composition of CNTs. It seems that the activation of 2% and 5% CNTs electrode is easier than that of other electrodes and these electrodes have almost similar capacity for the first cycle, 147 mAh/g. On the other hand, 0% and 20% CNTs electrodes showed almost similar initial, ~ 68 mAh/g and maximum capacity, ~ 224 mAh/g. The maximum specific discharge capacity 243 mAh/g is observed for 5% and 15% CNTs electrodes. To achieve maximum capacity electrode took 7 to 10 cycles in this investigation. Activation for prepared electrodes 300 250 5% CNTs 15% CNTs 0% CNTs 2% CNTs 200 2% CNTs 150 5% CNTs 10% CNTs 10% CNTs 100 15% CNTs Specific Discharge Capacity, mAh/g 20% CNTs 20% CNTs 0% CNTs 50 0 5 10 15 20 25 Cycle Number Figure 4.14: Activation profile of different kind of CNTs electrode. 81 Experimental results and discussion During activation performance alkaline atoms (Li, K, Na) from the rare earth based hydrogen storage alloy penetrate into the surface oxide layers by heating the alloys in electrolyte [15]. The active material in the negative electrode breaks up into micro particles with increasing the number of cycles. The particle size of the AB5 alloy decreases greatly in first few cycles and after that it remains almost constant with increasing number of cycles. This phenomenon introduces oxidation for first few cycles and after that when the particle size reaches a constant value it prevents further oxidation [16]. The maximum capacity is the stable capacity after activation. In order to achieve maximum discharge capacity, a fresh electrode needs couple of charge-discharge cycles to remove oxide layer or absorbed gas into the negative electrode which reacts with the hydrogen [17]. Table 4.4 shows the maximum capacity of the electrodes. Table 4 .4: Activation and maximum capacity of the electrodes Specific discharge Maximum Capacity, Electrodes capacity on 1st cycle, Activation cycles mAh/g mAh/g 0% CNTs 67 10 229 2% CNTs 146 7 239 5% CNTs 148 8 243 10% CNTs 128 8 232 15% CNTs 139 8 243 20% CNTs 68 9 223 82 Experimental results and discussion It appears that the initial discharge capacity is higher as addition of adequate amount of CNTs. This might be explained that the decrease of the electrode oxidation due to a lower surface area in contact with the electrolyte. It is likely to happen that the electrode oxidation is accelerated with decreasing particle size [18]. Due to un-oxidized material properties of CNTs might increase the initial discharge capacity. As compared with other carbon materials, such as carbon black and graphite, the volume of CNTs is much smaller, which indicates the specific discharge capacity is much higher [19]. It is believed that in higher percentage of CNTs electrode introduce higher amount of CNTs and less packing density, as a result no close contact with the molecules. Consequently it introduces high resistance for diffusion of electrolyte into the negative electrode. On the contrary, 5% CNTs electrode introduce lower resistance because of close contact with the nano molecules. Thus lower percentage of CNTs might provide enough pores and space for diffusion of electrolyte though the negative electrode. From this study it may be concluded that too much CNTs have no significant influence on activation performance of Ni-MH battery. The results indicate that the significant improvements for electrodes with 5% CNTs may be due to the increases of effective surface area and a reduction of oxidation between electrode particles. Effect of different diameter and ball milled CNTs on activation and maximum capacity Figure 4.15 represents the effect of different diameter and ball milled 5% CNTs electrode using similar experimental procedure. The initial discharge capacity is less than 100 mAh/g for all electrodes except 20-40 nm electrode. 10-20 nm electrode shows low maximum capacity compared to other electrodes. 60-100 nm electrode shows the highest maximum stable specific discharge capacity among all of the electrodes though 83 Experimental results and discussion its initial discharge capacity is low, ~ 91 mAh/g, Table 4.5. Activation profiles of these electrodes are similar to different composition of CNTs electrode shown in Figure 4.14. Activation of different diameter of CNTs 300 Ball milled 60-100nm 250 20-40 nm 200 10-20 nm 150 60-100nm 100 Ball milled 10-20 nm 20-40 nm 50 Specific discharge capacity,mAh/g discharge Specific 0 024681012 Cycle number Figure 4.15: Effect of diameter and ball milled CNTs on activation profile. Although CNTs electrodes are made using similar experimental procedure the specific discharge capacity of CNTs with different diameter is different. It can be observed from Table 4.5 that 20-40 nm CNTs have performed well with overall excellent activation and maximum specific discharge capacity. Some previous research also mentioned that MWNTs with small diameter has good electrochemical performance [20]. Nanomaterials absorb large quantities of gas by virtue of their high surface area [21]. Table 4.5: Activation and maximum capacity of different diameter and ball milled electrodes. Specific discharge Maximum Capacity, 5% CNTs electrodes capacity on 1st cycle, Activation cycles mAh/g mAh/g 84 Experimental results and discussion 10-20 nm 61 219 20-40 nm 148 242 Ball milled 20-40 8-10 65 244 nm 60-100 nm 91 252 The gas absorption depends on tube diameter and length of the CNTs [3]. The following reasons might be responsible for absorption of gas in CNTs [3, 4]: x Highly specific areas would permit the gas absorption. x Optimized absorbent structure would enable high absorptive property by introducing both the appropriate tube diameter and the inner-tube spacing for different experimental conditions /the space between nanotubes in the bundles. The hydrogen storage depends on the structure of CNTs, and the hydrogen molecules can be stored both in the porous hole and between the layers of CNTs [22]. The hydrogen molecules that deposited between the layers of CNTs might enter by diffusing from outside to inside of the nanotube. The layers of CNTs are direct proportion to the diameter, and having too many layers can resist hydrogen from coming into the tube center from tube wall through diffusing. As a result the deposition and release of hydrogen likely to be less. So CNTs with smaller outside diameter have better activation performance, on the other hand, when the diameter is too small, the hydrogen storage is limited by the few layers and that might result decrease of capacity. 85 Experimental results and discussion 4.4.2 Discharge potential analysis Table 4.6 presents the discharge middle voltages of electrodes with different CNT contents. The discharge middle voltage is the average data that observed at the 10th charge–discharge cycle. The discharge middle voltage is an important parameter of the rechargeable batteries [23]. Higher discharge middle voltage is associated with better performance in discharging [17, 24]. Figure 4.16 represents the discharge curve for six electrodes at a charge-discharge rate of 0.1-0.2 C after being fully activated in 6 M KOH electrolyte for 24 hours. This result indicates that the discharge voltage does not remain constant but decreases with the extent of discharge in each corresponding electrodes. Higher discharge voltage indicates the electrode has lower resistance. The specific discharge capacities of the CNTs electrodes were calculated at a cut-off voltage of 1.0 V. Table 4.6: Discharge mid point voltage obtained from different electrodes. Specific Discharge time, Discharge mid point CNTs electrodes discharge hours voltage, V capacity, mAh/g 0% 4.9 229 1.274 2% 4.9 223 1.260 5% 4.8 242 1.269 10% 4.8 233 1.266 15% 4.5 241 1.266 20% 4.4 223 1.299 86 Experimental results and discussion Discharge potential at 10 cycle 1.5 10% CNTs 2% CNTs 1.45 20% CNTs 1.4 5% CNTs 15% CNTs 1.35 0% CNTs 1.3 1.25 20% CNTs Voltage, V 1.2 2% CNTs 1.15 15% CNTs 1.1 10% CNTs 5% CNTs 1.05 0% CNTs 1 0 30 60 90 120 150 180 210 240 Specific Discharge Capacity, mAh/g Figure 4.16: Discharge curves of MH electrodes with various CNTs composition. The discharge capacity of the electrode gets better with addition of 5 to 15% CNTs. Table 4.6 shows the discharge midpoint voltages of the CNTs electrodes. 5%, 10% and 15% CNTs electrodes have higher midpoint voltages, whereas 5% and 15% CNTs electrode has a steady discharge plateau voltage, which are superior to the other CNTs electrodes. This indicates that these electrodes have lower resistance and higher specific discharge capacity compared to other electrodes. 5% and 10% electrodes took 4.8 hours to fully discharge while 15% CNTs electrode took only 4.5 hours. On the contrary, though 20% CNTs electrode shows the highest midpoint discharge voltage among all of the electrodes, its capacity is low. Because the discharge time limit is low, 4.4 hours, compared to other electrodes and hence provides lower specific discharge capacity. Discharge duration depends mainly on the discharge voltage that is on the internal 87 Experimental results and discussion resistance of the electrode [25]. On the other hand, though 0% CNTs electrode has second highest midpoint voltage and the longest discharge time, 4.9 hour; its specific discharge capacity is low compared to those of other electrodes. The specific discharge capacity is related to the ratio of the two active materials in the electrode. The optimum value of the ratio is neither too much nor too small ratio of CNTs to AB5 to achieve higher discharge performance. It might be assumed that the increase in internal resistance of the CNTs batteries was lower than that that 0% CNTs batteries CNTs. The electrode with lower CNTs has a lower discharge capacity than that with adequate CNTs electrode. It can be concluded that the optimum percentage of CNTs in AB5 electrode is 5% in our experimental range. 4.4.3 Cycle life characteristics and cycle stability Figure 4.17 presents the variation of the discharge capacity of 0%, 2%, 5%, 10%, and 20% CNTs electrode as a function of cycle number. 0% CNTs Electrode The specific discharge capacity for 0% CNTs in the first cycle is about 143 mAh/g . The discharge capacity reaches a value of 242 mAh/g after 8 cycles and the discharge capacity remains constant over 34 cycles. Afterwards it starts to decrease steeply from cycle 38 to 63 cycles. The capacity decays from approximately 243 to 109 mAh/g and remains constant after that. The 0% CNTs electrode was charged and discharged over 221 cycles. 88 Experimental results and discussion 280 Cycle life test of prepared electrodes 230 5% CNTs 180 20% CNTs 10% CNTs 130 0% CNTs 80 pcfcDcag aaiy mAh/g Capacity, Dicharge Specific 2% CNTs 30 0 20 40 60 80 100 120 140 160 180 200 220 240 Cycle Number Figure 4.17: Variation of the electrochemical discharge capacity of different composition of CNTs electrode as a function of cycle number. 2% CNTs Electrode The specific discharge capacity for the first cycle of 2% CNTs is only about 130 mAh/g, but specific discharge capacity increases steadily to reach a value about 239 mAh/g after 7 cycles and it remains constant over 36 cycles. The specific discharge capacity decreases steeply from ~ 240 mAh/g in the 36 cycles to 78 mAh/g in the 62 cycles. After 62 cycles the specific discharge capacity decreases slowly. The total number of cycles is 113 for this electrode. The specific discharge behavior of 2% CNTs electrode is pretty much similar to specific discharge capacity of 0% CNTs electrode but with a low specific discharge capacity along with low cycle number. 5% CNTs Electrode The specific discharge capacity for 5% CNTs electrode is about 153 mAh/g for the first cycle which is the highest among all the electrodes. The specific discharge capacity 89 Experimental results and discussion reaches a value of 246 mAh/g after 8 cycles and afterwards it almost remains constant over 53 cycles. After that it decreases slowly to a value of 241 mAh/g from 53 cycles to a value of 224 mAh/g after 72 cycles. Due to software limitations about 20 cycles were lost after that. The specific discharge capacity starts decreasing steeply from 72 cycles. The total number of cycle is 130 in this electrode. 10% CNTs Electrode The specific discharge capacity of the 10% CNTs electrode is about 143 mAh/g for the first cycle and its value reaches about 243 mAh/g after 8 cycles, afterwards it remains constant over 53 cycles. The specific discharge capacity starts decreasing slowly from 53 cycles. Its capacity does not decrease steeply like above mentioned electrodes. Total number of charge-discharge cycle of this electrode is 223. The cycle stability of this electrode is close to the one with 0% CNTs electrode but with a higher specific discharge capacity. So far, this is the highest cycle stability performed by these two electrodes in this investigation. 20% CNTs Electrode The capacity of the first cycle of this electrode is ~ 66 mAh/g, which is the lowest among all of the electrodes. The capacity increased to a value of 241 mAh/g after 5 cycles. After that the capacity gradually decreased to a value of 225 mAh/g over 56 cycles. From 56 cycles to 73 cycles the capacity decreased steeply and from 93 cycles to 101 cycles the capacity decreased almost exponentially. The total number of cycles for this electrode is 101. 90 Experimental results and discussion From the above results 5% CNTs electrode displays initial higher specific discharge capacity, confirming the positive effect on the addition of 5% CNTs in the MH electrodes and 20% CNTs shows the lowest initial specific discharge capacity, Table 4.7. Table 4.7: Experimental findings obtained by cycle life test of the electrodes. Initial capacity, Maximum Activation Life stability, CNTs electrodes mAh/g capacity, mAh/g cycles cycles 0% 143 242 8 221 2% 130 239 7 113 5% 153 246 8 130 10% 143 243 8 223 20% 66 241 5 101 Due to volume expansion and shrinkage of MH electrode or experimental error (less timing during press), connection between Ni substrate and active materials might become poor, shows in Figure 4.18(a).This loose connection problem was fixed by increasing timing during press. This incident could lower the specific discharge capacity of MH electrode. It is also observed that SEM image of electrode after experiment, Figure 4.18(a), the electrode are covered with long fibers of separator and there might not enough CNTs in this region due to inhomogeneous distribution of the CNTs into the electrode. Figure 4.18(b) shows positive electrode after experiment. Positive electrode lost its active materials into the separator and the connection between current collector and positive electrode was loose. This deterioration of the positive electrode and loose connection might contribute the decease of specific discharge capacity. Afterwards care was taken to prevent loose connection between positive electrode and current collector 91 Experimental results and discussion by increasing timing (during press) or improved the current collector (used positive electrode itself for good conductivity). In other experiments, positive electrode was also improved by folding with Ni foam to prevent deterioration of electrodes. Crack Prepared 2% CNTs electrode Separator AB5 Loose connection SEM image of 2% CNTs electrode after 2% CNTs electrode after experiment 100 cycles Figure 4.18(a): Photograph of the 2% CNTs electrode after cycle life test. Loose connection Experimental electrodes in a test cell Deteriorated positive electrode Separator contaminated with Ni(OH)2 Figure 4.18(b): Photograph of the deterioration of positive electrode after cycle life test. 92 Experimental results and discussion The decay in specific discharge capacity might results from a deterioration of the negative electrode due to oxidation because of La(OH)3 or LaO3 formation during repeated charge-discharged cycles [26]. The XRD pattern of the 2% CNTs in Figure 4.19 after cycle life tests shows new peaks appear at 26-29 degrees. During charge- discharge negative alloy might expand and shrink due to charge and discharge or absorption and desorption of hydrogen. This introduces oxidation into the MH electrode. XRD of 2% CNTs electrode 600 500 400 300 ground MH alloy, Ni and Counts 2% CNTs 200 100 Ne w pe ak s After cycle life 0 5 152535455565758595105115 Two theta Figure 4.19: XRD pattern of the 2% CNTs, Ni and MH powder and their corresponding XRD analysis after cycle life test. Due to the volume change of the MH alloy during hydrogen absorption, the grains and particles size decrease during repeated charge-discharge cycles [27]. The introduction of oxidation of the negative electrode and loss of active materials with repeated charge- discharge cycles can result in capacity decay [27-29]. This results in a large increase in surface area and hence increased corrosion. Alloys with a small volume expansion tend 93 Experimental results and discussion to have longer cycles stability than those with high volume expansions. The La containing alloy has higher capacity with less stability [24]. The corrosion elements of the AB5 alloy might introduce oxidation in the electrolyte and deposited in the positive electrodes. After repeated charge-discharge cycle, the deposition amounts of the corrosion products of all elements of AB5 alloy in the positive electrodes increases [30, 21]. Another possible reason of the capacity decay is the decrease of the electrolyte during charge-discharge cycles. With respect to the specific discharge capacitates of these electrodes it might be concluded that the addition of CNTs improves the stability and lifetime of the charge- discharge cycles. CNTs seem to prevent further oxidation during repeated charge- discharge cycles and therefore, increased discharge capacity. It is likely to happen that the repeated charge-discharge process; the battery suffers lower mechanical strength, losing of the active materials from the substrate, high internal resistance and capacity decay. It might be assumed that CNTs electrode can introduce higher mechanical strength, lower internal resistance and hence increase the capacity [21, 31]. Hydrogen absorption mechanism is related to the surface of the samples [2]. It is assumed that the highly accessible electrode-electrolyte interface of the CNTs electrode introduced good electrical conductivity and better conductive network. Hydrogen might get packed within the tubes at a higher density. It can be concluded that addition of CNTs in small amount is helpful in improving charge-discharge cycle stability, but an extra addition of CNTs (e.g. 20%) deteriorates the stability. Because the high volume of CNTs will reduce the lower density which 94 Experimental results and discussion results in poor connection between the active materials. As a result, resistance becomes higher than that of lower CNTs additions. Due to an ideal pore structure CNTs are suitable for hydrogen storage and exhibits a good electrochemical hydrogen storage performance [32]. Researchers have shown that the decline in the cyclic life of an electrode has partially a relationship with the activated material shedding. But the experiment also indicated that CNTs in the electrodes have a great discrepancy in the electrochemical hydrogen storage capacity, especially the electrodes prepared with CNTs of different diameters have a great diversity in the electrochemical hydrogen storage capability. The influence on different diameter and ball milling (1hour) on the cyclic specific discharge capacities of the 5% CNTs electrodes is presented in Figure 4.20. It is observed that initial and maximum discharge capacity is higher for 10-20 nm CNTs electrode. Effect of diameter and ball milled CNTs in cycle life test 350 300 10-20 nm Ball milled 250 CNTs 20-40nm 60-100nm 200 40-60 nm 150 Ball Milled CNTs 10-20 nm 100 20-40nm 60-100nm Specific Discharge Capacity, mAh/g Capacity, Discharge Specific 50 40-60 nm 0 0 5 10 15 20 25 30 35 40 45 50 55 Cycle Number Figure 4.20: Effect of diameters and ball milled in the cycle life performance of 5% CNTs electrode. 95 Experimental results and discussion The influence in different diameter in CNTs and effect of ball milling CNTs (1 hour) in the cycling discharge capacities are shown in Figure 4.20. This graph presents the discharge capacity of 5% CNTs electrode with 10-20, 20-40, 40-60, 60-100 nm and ball milled CNTs electrode. The specific discharge capacity of 10-20 nm CNTs electrode reaches about 280 mAh/g after 5 cycles. From Figure 4.20, it is observed that specific discharge capacity decreases very slowly from 5 to 19 cycles, after 19 cycles specific discharge capacity starts to increase again with cycle number. The specific discharge capacity of this electrode starts to increase to a value of almost 300 mAh/g after 36 cycles, which is the theoretical capacity used in these experiments. The discharge capacity of the 20-40, 60-100 nm and ball milled CNTs are almost similar except for the first few cycles. The maximum specific discharge capacity almost similar for these three electrodes ~ 245 mAh/g even though the activation cycle is different, Table 4.8. After 7 cycles the discharge capacity remains almost constant for these three electrodes. One of the possible reasons of this might be the XRD diffraction of ball milled and 60-100 nm CNTs possess almost similar structure, Figure 4.9. The initial specific discharge capacity of the 40-60 nm CNTs is 35 mAh/g which is similar to 37 mAh/g of 60-100 nm CNTs electrodes. The total number of cycles obtained from this electrode is 22 during the available time frame. Table 4.8: Effect of diameter and ball milled in 5% CNTs electrodes. 5%CNTs Initial capacity, Maximum Activation Life stability, electrodes mAh/g capacity, mAh/g cycles cycles 10-20 nm 107 280 5 43 20-40 nm 153 246 8 50 96 Experimental results and discussion 40-60 nm 35 251 6 22 60-100 nm 37 245 4 47 Ball milled 69 242 2 50 It appears that the maximum capacity, 280 mAh/g is higher for 10-20 nm CNTs electrode as compared to other electrodes. The initial activation is poor for 40-60 nm and 60-100 nm electrodes. There is no significant difference between 20-40 nm and ball milled CNTs except ball milled CNTs shows better activation properties, it only takes 2 cycles to reach its maximum capacity where as other one takes 8 cycles. It can be observed from Figure 4.20 that the small diameter electrode shows a higher specific discharge capacity. It is believed that smaller diameter introduces higher capacity because of its higher surface area and high activity for hydrogen absorption and desorption. Hydrogen absorption-desorption depends on the structure of the nanotubes. Hydrogen molecules are stored into the aggregated pores and between the layers of CNTs, Figure 4.21 [33]. Figure 4.21: Illustrated diagram of the aggregated pores of the CNTs [33]. 97 Experimental results and discussion Hydrogen enters into the CNTs generally through diffusion [34]. Due to wetting properties of CNTs, their inner hollow cavity can serve as a storage medium for other materials. They are able to draw liquid or gas inside by capillarity [32]. It is assumed that the hydrogen condenses in the cavity of the nanotube or forms an absorbed layer of hydrogen at the surface of the tube [2]. The hydrogen storage density due to condensed hydrogen in the cavity of the tube depends on the tube diameter. These layers of CNTs are direct ratio to the diameter, and too many layers might introduce prevention of hydrogen diffusion [31]. Consequently, hydrogen absorption-desorption will be low. As a result, CNTs with a smaller diameter have better electrochemical performance. On the contrary, when the diameter is too small, the hydrogen storage is limited by the few layers and hence results lower capacity [32, 22, 35]. As a result, CNTs with a smaller diameter has better electrochemical performance. 4.4.4 SOC, DOD and Self-discharge tests The charging efficiency for different state of charge has been calculated using different percentage of charge at cut-off 1.0 V. The different state of charge has been measured for three electrodes. However, the values obtained from the graph, Figure 4.22, indicate there is no significant effect of the SOC condition during the discharge. The electrode with 10% CNTs has shown most steady state of different charge conditions among the three of the electrodes. Figure 4.23 shows the depth of discharge (DOD) of the two electrodes. The power available from a battery depends on the DOD [36]. The DOD test for different state of discharge has been calculated using different percentage of discharge at a cut-off 0.0 V. 98 Experimental results and discussion State of charge test 120 100 10% CNTs 80 % 0%CNTs 2% CNTs 60 Charge, 40 10% CNTs 0% CNTs 20 2% CNTs 0 0 50 100 150 200 250 300 Specific discharge Capacity, mAh/g Figure 4.22: Specific discharge capacity at different state of charge. This experiment is very similar to the SOC test. 10% CNTs electrode shows a higher specific discharge capacity 192 and 240 mAh/g at 80% and 100% discharge state respectively, as compared to those of 0% CNTs electrode. The DOD test shows the prepared 0% CNTs electrode is not stable at different rate of discharge. Further experiments need to be carried out to get an improved accuracy. Depth of discharge test 120 100 0% CNTs 80 10% CNTs 60 40 Discharge, % Discharge, 10% CNTs 20 0% CNTs 0 0 50 100 150 200 250 300 Specific Discharge Capacity, mAh/g Figure 4.23: Specific discharge capacity at different depth of discharge. 99 Experimental results and discussion After completing the DOD test self-discharge of the electrodes was measured by monitoring open circuit voltage decay of fully charged electrodes with open circuit storage time. Figure 4.24 shows the self-discharge rate of the electrodes fabricated with containing 0% and 10% CNTs. The electrodes were fully charged and stored in the open circuit condition for 2, 4, 6 and 8 days at room temperature. Figure 4.24 shows there is no significant difference in the capacity retained by the electrodes. The results obtained are satisfactory, as none of the electrodes experienced complete self-discharge during this storage period. 10% CNTs electrode losses 14% capacity during 2 days storage time, but for 4 to 8 days storage time its capacity lost is negligible, Table 4.9. It can be clearly observed that 10% electrode shows remarkably low self-discharge rate. On the other hand, the available capacity is 0% CNTs electrode on storage time 6 and 8 days are 69% and 58% respectively. To get more information further experiments need to be conducted for long storage times at high temperatures. Self-discharge trend 300 250 10% CNTs 200 150 100 0% CNTs 50 10% CNTs Specific discharge capcity, mAh/g 0% CNTs 0 0123456789 Open circuit storage time, Day Figure 4.24: Self-discharge behavior of different storage time of the electrodes. 100 Experimental results and discussion 1.52 Self-discharge test 1.5 1.48 1.46 1.44 10% CNTs 1.42 0% CNTs 1.4 10% CNTs Open circuit V Voltage, 1.38 1.36 0% CNTs 1.34 0 5 10 15 20 25 30 35 40 45 50 Open circuit storage time, Hour Figure 4.25: Voltage decay on open circuit at room temperature. Cell voltage of the battery decreases with increasing storage time, which is similar to charge retention to storage time. As can be seen from Figure 4.25, that open circuit voltage drops quickly during the first 24 hours and then decreases gradually with increasing storage time. 10% CNTs show higher and stable open circuit voltage after 24 hours storage time. This phenomenon suggests that the open circuit voltage might be an important indicator of decrease of capacity. Table 4.9: Available specific discharge capacity after self-discharge. Specific Spe. discharge Storage time, discharge Available Electrodes capacity after day capacity, capacity, % storage, mAh/g mAh/g 101 Experimental results and discussion 2 249 219 86 4 224 218 97 10% CNTs 6 225 220 97 8 225 220 97 2 133 109 82 4 135 102 76 0% CNTs 6 134 93 69 8 133 76 58 Self-discharge occurs due to the following reasons: when the hydrogen pressure is higher than the inner cell pressure, hydrogen will be easily removed from the MH electrode resulting in decreased in discharge capacity. On the other hand, when the positive electrode reacts with hydrogen to form nickel hydroxide, the cell inner pressure decreases, this little drop of the inner pressure also responsible for self-discharge. Another phenomenon of self-discharge is the deterioration of the negative electrode due to electrochemical oxidation of the alloys [17, 37], consequently hydrogen release from the MH alloy [38]. Self-discharge may be accelerated from the hydrolysis of separator [27]. Storing the electrode at open circuit above 1.0 V can lower the self-discharge rate [30]. This can prevent oxidation of hydrogen on the surface of the electrode. [39]. Self-discharge does not depend on the type of the active material alloy of the negative electrode [26]. From the above findings it can be shown that 10% CNTs electrode shows lower self-discharge rate, while in this case Ni-MH battery lose about 30% of their capacity by self-discharge during one month of storage period. [21]. 102 Experimental results and discussion 4.4.5 High rate discharge characteristics The power capability of the electrodes was tested at different SOC and DOD tests. The high rate discharge capability of the electrodes was measured after complete activation. The high rate discharge ability was carried out from the 0.2 to 10 C discharge rate of CNTs electrodes at room temperature. Figure 4.26 shows comparison of specific discharge capacity of the electrodes with different C rates. It can be seen that the specific discharge capacity decreases rapidly with increasing discharge rates. From the table 4.10 it can be seen that 2 to 15% electrodes show better high rate capability. Specific discharge capacity of 5% electrode reaches a maximum value of 248 mAh/g at 1 C rate and after that it remains almost constant over 2 C rate. Although 5% CNTs electrode showed highest discharge capacities from 1C rate to 6 C rate among all of the electrodes, its capacity deteriorated at higher discharge rate. 0% and 20% electrodes show similar low discharge properties at high rate discharge, 10 and 14 mAh/g at 10 C rate respectively. The 10% CNTs exhibit the highest specific discharge capacity of 30 mAh/g compare to other electrodes from 7 to 10 C rate. From 1 to 3 C rate 15% CNTs electrode represent higher specific discharge capacity compare to 10% CNTs electrode, but after 3 C rate its capacity starts to decrease rapidly. 103 Experimental results and discussion The specific discharge capacity of the 2 % CNTs shows higher capacity from 1 to 5 C rate, whose value is higher than 10% but lower than 5% CNTs electrode. After 5 C rate its capacity starts to decrease. From the Table 4.10 it can be seen that 2 to 15% electrodes show better high rate discharge capability. Table 4.10: High rate discharge performance of the electrodes at room temperature. Specific discharge capacity, mAh/g Electrodes 1C 3C 5C 7C 10C 0% CNTs 201 83 29 22 10 2% CNTs 217 173 89 49 19 5% CNTs 242 183 100 60 25 10% CNTs 217 142 91 64 30 15% CNTs 224 135 74 45 20 20% CNTs 158 80 39 26 14 High rate discharge of electrodes 300 15% CNTs 250 10% CNTs 5% CNTs 5% CNTs 20% CNTs 200 2% CNTs 0% CNTs 15% CNTs 150 2% CNTs 20% CNTs 100 0% CNTs 50 10% CNTs Specific discharge capacity, mAh/g 0 12345678910 C rate Figure 4.26: High rate discharge capabilities of different types of CNTs electrodes. 104 Experimental results and discussion From the above analysis it can be seen that the remarkable high rate dischargeability can be obtained by addition of 5 to 10% CNTs. As a result, with increasing appropriate amount of CNTs into the MH electrode, the high rate discharge capabilities can be enhanced. This phenomenon requires a certain amount to modify the electrodes performance at high discharge rates. Too much and less percentage CNTs contributed no effect in the improvement of high-rate capability of the electrodes. The rate capability depends strongly on: the nature of the electrode additives, and the amount of active materials [15]. The mesoporous character of carbon nanotubes plays a dominant role in their electrochemical properties [40]. Compared with traditional carbon materials, carbon nanotubes have a higher rate of electron transfer. It is assumed that discharging capacity of CNTs electrode is related to the ratio of the two active materials in the electrode. There exists an optimum value of the ratio: neither too big nor too small ratio of CNTs electrode can attain higher and more stable discharging performance. The electrode with excess CNTs has a lower discharging capacity than that with adequate CNTs. We conclude that the optimum percentage CNTs electrode is 5% in our experiment range. The CNTs electrode provides better conductivity during high rate charge-discharge cycles. Considering the excellent high rate discharge ability, lower percentage of CNTs electrodes are doubtlessly has the potential use in the battery systems for electric vehicles. The effect of different diameter and ball milled CNTs are presented in Figure 4.27. It can be seen clearly that 20-40 nm CNTs shows remarkable improvement on 105 Experimental results and discussion high rate discharge capabilities on continuous discharge up to the 10 C rate as compared to other electrodes, Table 4.11. 300 Effect of diameter and ball milled CNTs g 250 60-100 nm 200 Ball milled Ball Milled CNTs CNTs 20-40 nm 150 10-20 nm 40-60 nm 40-60 nm 100 10-20 nm 20-40 nm 50 Specific discharge capacity, mAh/ 60-100 nm 0 1357911 C rate Figure 4.27: Effect of diameter and ball milled CNTs on high rate discharge capabilities. Ball milled and 60-100 nm CNTs electrode show almost similar characteristics at high rate discharge. Comparing 10-20 and 40-60 nm CNTs electrodes it can be seen that from 1 to 5 C rate, the specific discharge capacity is higher for 10-20 nm CNTs electrode, but after that the capacity becomes almost same for these two electrodes. It is well known that mainly high rate discharge capability depends on ohmic resistance (such as connection, conductivity) and resistance from electrolyte (ion diffusion) and hydrogen diffusion. Better ion and hydrogen diffusion increases the capacity. 2 to 10% CNTs show better capacity because there is low resistance between the CNTs molecule and 20% CNTs introduces low packing density, consequently it exhibits higher resistance and hence lower capacity. Too much CNTs do not contribute enough space and pore for hydrogen absorption and desorption. 106 Experimental results and discussion Table 4.11: High rate discharge performance of 5% electrodes at room temperature. CNTs Specific discharge capacity, mAh/g electrodes 1C 3C 5C 7C 10C 10-20 nm 216 67 19 13 9 20-40 nm 242 183 100 60 25 Ball milled 233 127 38 16 11 40-60 nm 156 42 15 13 12 60-100 nm 243 118 41 16 It is believed that high rate discharge capability is introduced by the electrochemical kinetics on the surface and the diffusion of hydrogen in the lattice [41]. The surface of the lower percentage CNTs has high electro catalytic activity and also often lower resistance to the diffusion of hydrogen. The specific surface area of the hydrogen storage alloy may be increases with the addition of CNTs. All these effects lead to an improvement in the performance of the MH electrode. As a result, CNTs could work not only as hydrogen reservoirs but also as a catalyst for charge-discharge cycles. Research has shown that carbon nanotubes have a higher rate of electron transfer and the largest hydrogen storage capacity among the hydrogen storage materials at present so far [42]. In CNTs a possible method of hydrogen storage may be the reversible chemisorption of hydrogen into the CNTs [43]. The performance of the battery depends on the hydrogen storage capacity of the CNTs. The nanotubes have a significant effect in improving the battery performance at a high discharge rate [21]. 107 Experimental results and discussion 4.4.6 High rate charge characteristics The charge curve is presented in Figure 4.28 at different high state of charge current. The rapid charge capability of the electrodes has been determined in this experiment. High values of capacity of the electrode obtained in the high rate charge ability tests enable the definition of a quick charge procedure, at least for 20-30 minutes duration. Figure 4.28 and Figure 4.29 clearly show that best performance can be obtained when the electrodes are charged at the 0.2 C rate up to 1.5 C rate, in these range the capacity is almost 100%. From Table 4.12 and 4.13, it can be observed that 5% and 10% CNTs electrode posses highest specific capacity in terms of charge-discharge performance. Like other experiments described above, this experimental results also prove that small percentage CNTs is helpful for electrochemical characteristics. High rate charge of electrodes 270 5% CNTs 10% CNTs 220 0% CNTs 170 2% CNTs 20% CNTs 15% CNTs 2% CNTs 5% CNTs 120 10% CNTs 0% CNTs 20% CNTs Specific Charge Capacity, mAh/g 15% CNTs 70 0.2 0.7 1.2 1.7 2.2 2.7 3.2 C rate Figure 4.28: High rate charge capabilities of the CNTs electrodes. 108 Experimental results and discussion Specific discharge capacity of the electrodes 0% CNTs 350 5% CNTs 2% CNTs 300 10% CNTs 15% CNTs 20% CNTs 250 10% CNTs 200 5% CNTs 20% CNTs 150 0% CNTs 100 2% CNTs 15% CNTs Specific discharge capacity, mAh/g capacity, discharge Specific 50 0.2 0.7 1.2 1.7 2.2 2.7 3.2 Charge rate, C rate Figure 4.29: High rate discharge capabilities of the CNTs electrodes. Table 4.12: Specific charge capacity at different high rate charge current. Specific charge capacity, mAh/g Electrodes 0.5C 1C 3C 0% CNTs 250 250 104 2% CNTs 247 251 140 5% CNTs 253 249 251 10% CNTs 248 248 250 15% CNTs 247 249 82 20% CNTs 248 250 161 Table 4.13: Specific discharge capacity at different high rate charge current. Discharge rate=0.2 C 109 Experimental results and discussion Specific discharge capacity, mAh/g Electrodes 0.5 C 1 C 3 C 0% CNTs 246 235 101 2% CNTs 239 230 100 5% CNTs 239 231 197 10% CNTs 251 250 218 15% CNTs 259 252 88 20% CNTs 271 252 162 Figure 4.30 and Table 4.14 show the effect of diameter and ball milled CNTs on high rate charge characteristics. 20-40 nm and 40-60 nm show that specific charge capacity is almost 100% at 3 C rate, that means the electrodes can be charged in 20 minutes. On the contrary, 60-100 nm electrode shows worst high rate charge condition, which again proves that bigger diameter CNTs does not further improve the electrochemical performance tests. Effect of diameter and ball milled CNTs 300 250 20-40 nm Ball milled 40-60 nm CNTs 200 10-20 nm Ball Milled 150 10-20 nm 60-100nm 100 40-60 nm 20-40 nm 60-100nm 50 Specific ChargeSpecific mAh/g Capacity, 0 0.2 0.7 1.2 1.7 2.2 2.7 3.2 C rate Figure 4.30: Effect of diameter and ball milled CNTs on high rate charge capabilities. 110 Experimental results and discussion Table 4.14: Effect of diameter and ball milled CNTs on specific charge capacity at different high rate charge current. Specific charge capacity, mAh/g Electrodes 0.5C 1C 3C 10-20 nm 250 250 104 20-40 nm 253 249 251 Ball milled 247 251 140 40-60 nm 248 248 250 60-100 nm 247 249 82 4.5 PCI measurements Several researches show that native nanotubes are not open. The core of CNTs might become accessible to guest molecules after ball milling process. In general two principal surface sites are inherent in CNTs: (a) intercrystalline or aggregate pores corresponding to the external surface; (b) Nanoscales cavities in the central core of the nanotubes. This type of pore might have a huge adsorption capacity which is responsible for 78.5% of the total adsorbed amount [44]. Therefore an understanding of the intercrystalline pore structure in relation to the aggregation mechanisms is necessary for improvement of efficient hydrogen storage materials. Now a day, it has become an efficient way of shortening carbon nanotubes by applying ball milling procedure [45]. This process not only separates CNTs agglomerates, but also fragments the nanotubes [46]. During this process, the long nanotubes might be reduced and shortened with open tips [47]. It is believed that some CNTs are changed 111 Experimental results and discussion into amorphous carbon after milling treatment. The experiment also shows that the CNTs become shorter and shorter with milling time increasing and also thinner as a function of time due to expansion and peeling of graphene layers. According to researcher that the thinner tubes are more reactive than tubes with a larger diameter [46]. In this study, the hydrogen absorption-desorption behavior of MH alloy powder and ball milled CNTs were carried out at room temperature. Figure 4.31 shows the hydrogen absorption of the MH powder increases with increasing pressure and the desoprtion increases with decreasing pressure. In this sample the hydrogen is not released completely. A small amount of hydrogen remains into the alloy below 0.07 atm pressure. This alloy absorbed the maximum amount of hydrogen, 1.4wt%. This result is similar to some previous research [48]. MH alloy exhibit sloped plateaus between 0.22 to 0.95 bar which is suitable range for electrochemical applications [49]. The long plateau pressure at lower pressure is responsible for large amount of hydrogen absorption and desorption [48]. The desorption behavior of MH alloy is not that good, may be due to some experimental error desorption is taking long time. Absorption-desorption for as received MH alloy powder 10 9 8 7 6 5 4 Absorption curve 3 2 Pressure (kg/m2) 1 Desorption curve 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 wt % Figure 4.31: Hydrogen absorption-desorption of as received MH alloy. 112 Experimental results and discussion Figure 4.32 shows the absorption desorption of the ball milled CNTs, this experiment was conducted with Jerry Leuo, honors student, UNSW. The absorption behavior of both ball milled CNTs 20-40 is good but hydrogen does not desorption properly of this sample. Ball milled CNTs absorbs 0.5 wt% hydrogen at 25 atm pressure. This result is similar to some previous results [48]. No plateau observes in this result. This experiment should carry out at high pressure to observe the absorption-desorption behavior properly. Absorption-desorption curve of ball milled CNTs 30 25 20 Absorption curve 15 Desorption curve 10 Pressure (Kg/m2)Pressure 5 0 0 0.1 0.2 0.3 0.4 0.5 0.6 wt% Figure 4.32: Hydrogen absorption-desorption curve of ball milled CNTs. 4.6 Limitations of this study This section focuses on various design limitations as well as features of the present CNTs based negative electrode. Identifying these limitations is considered extremely crucial for future improvement of the proposed topology. 113 Experimental results and discussion The connection between active material electrode and current collector was not of a very good quality and needs improvements. x Shedding of the positive electrode was quite severe. x Loss of the electrolyte during open cell experiment made the electrode performance unstable, drying out of the electrolyte decrease in capacity. x Some times, important data could not be recorded by the software. It is believed 5% CNTs would exhibit the higher and stable cycle life if it was possible to record all of the corresponding data’s. x Dispersion of CNTs into the MH and Ni powder could be improved further. x Statistics needs to be improved further with additional data. 114 Experimental results and discussion References [1] M. M. Shaijumon, N. Bejoy, S. Ramapradhu, “Catalytic growth of Carbon nanotubes over Ni/Cr hydrotalcite-type anionic clay and their hydrogen storage properties”, Applied Surface Science 242 (2005) 192-198. [2] A. Zuttel, P. Sudan, “Hydrogen storage in carbon nanostructures”, International Journal of Hydrogen Energy 27 (2002) 203-212. [3] F. Lamari Drakim, “Review of hydrogen storage by adsorption in carbon nanotubes”, International Journal of Hydrogen Energy 27 (2002) 193-202. [4] Arindam Sarkar, “A quantitative method for characterization of carbon nanotubes for hydrogen storage”, International Journal of Hydrogen Energy 29 (2004) 1487- 1491. [5] M. M. Shaijumon, S. Ramaprabhu “Studies of yield and nature of carbon nanostructures synthesized by pyrolysis of ferrocene and hydrogen absorption studies of carbon”, International Journal of Hydrogen Energy 30 (2005) 311-317. [6] Charles K Witham, “The effect of alloy chemistry on the electrochemical and hydriding properties of Ni-substitute LaNi5”, California Institute of Technology, Pasadena, California 2000. [7] Xinbo Zhang, Yujun Chai, “ Crystal structure and electrochemical properties of rare earth non-stoichiometric AB5-type alloy as negative electrode material in Ni- MH batter”, Journal of Solid State Chemistry 177 (2004) 2373-2377. [8] Chien-Sheng Kuo, Allen Bai, “Diameter control of multiwalled carbon nanotubes using experimental strategies”, Carbon 43(2005) 2760-2768. [9] Perfect communication with Dr. Q. Song, visiting research fellow, UNSW. 115 Experimental results and discussion [10] X. P. Gao, X. Qin, F. Wu, “Synthesis of carbon nanotubes by catalytic decomposition of methane using LaNi5 hydrogen storage ally as a catalyst”, Chemical Physics Letters 327 (2000) 271-276. [11] Seung Mi Lee, Ki Soo Park, Young Chul Choi, “Hydrogen absorption and storage in carbon nanotubes”, Synthetic Metals 113 2000, 209-216. [12] M. M. Shaijumon, S. Ramapradhu, “ Synthesis of carbon nanotubes by pyrolysis of acetylene using alloy hydride materials as catalysts and their hydrogen adsorption studies”, Chemical Physics Letters 374 (2003) 513-520. [13] S. G. Wang, Qing Zhang, “Synthesis and characterization of MWNTs with narrow diameter over nickel catalyst by MPCVD”, Scripta Materialia 48 (2003) 409-412. [14] P. Dantzer,“Properties of intermatallic compounds suitable for hydrogen storage applications”, Materials Science and Engineering A329-331 (2002) 313-320. [15] H. Hoshino, H. Uchida, “Preparation of a nickel-metal hydride (Ni-MH) rechargeable battery and its application to a solar vehicle”, International Journal of Hydrogen Energy 26 (2001) 873-877. [16] Mingming Geng, Jianwen Han, Feng Feng, “Charging/discharging stability of a metal hydride battery electrode”, Journal of the Electrochemical Society 146 (7) 2371-2375 (1999). [17] Chigyu Jeong, Wonsub Chung, “Effect of temperature on the laves phase alloy used in nickel/ metal-hydride batteries”, Journal of Power Sources 79 (1999) 19-24. [18] C. Rongeat, L. Roue, “On the cycle life improvement of MgNi-based alloy for Ni- MH batteries”, Journal of alloys and Compounds (2005). [19] J. Lva, J.P. Tu, “Effects of carbon nanotubes on the high-rate discharge properties of nickel/metal hydride batteries”, Journal of Power Sources 132 (2004) 282–287. 116 Experimental results and discussion [20] Chenguo Hu, Yiyi Zhang, “Diameter-dependent voltammetric properties of carbon nanotubes”, Chemical Physics Letters 418 (2006) 524-529. [21] A.M. Hermann, P.A Ramakrishnan, “Metal hydride batteries research using nanostructured additives”, International Journal of Hydrogen Energy 26 (2001) 1295-1299. [22] Shuangping Yi, Haiyan Zhang, “The study of electrochemical properties of CNTs- LaNi5 electrodes”, Materials Science and Engineering B. [23] M. Ben Moussa, M. Abdellaoui, “Effect of substitution of Mm for La on the electrochemical properties of the LaNi3.55Mn0.4Al0.3Co0.75 compound”, Journal of Alloys and Compounds 399 (2005) 264-269. [24] M. Luisa Soria, Joaquin Chacon, “Nickel metal hydride batteries for high power applications”, Journal of Power Sources 96 (2001) 68-75. [25] Chunshen Wang, Mariza Marrero-Rivera, “The self-discharge mechanism of AB5- type hydride electrodes in Ni-MH batteries”, International Journal of Hydrogen Energy. [26] Paul Ruetschi, Felix Meli, “Nickel-metal hydride batteries. The preferred batteries of the future?”, Journal of Power Sources 57 (1995) 85-91. [27] Jeng-Kuei Chan, Dyi-Nan Simon Shong, “Effect of Ni content on the electrochemical characteristics of the LaNi5-based hydrogen storage alloys” Materials Chemistry and Physics 83 (2004) 361-366. [28] Laure Le Guenne, Patrick Bernard, “Life duration of Ni-MH cells for high power applications”, Journal of Power Sources 105 (2002) 134-138. 117 Experimental results and discussion [29] Xiaofeng Li, Baojia Xia, “Corrosion of AB5 alloy during the storage on the performance of nickel-metal-hydride battery”, Journal of Alloys and Compounds 391 (2005) 190-193. [30] N. Rajalakshmi, “Electrochemical investigation of single-walled carbon nanotubes for hydrogen storage”, Electrochimica Acta 45 (2000) 4511-4515.] [31] Haiyan Zhang, Xiaojuan Fu, “The effect of MWNTs with different diameters on the electrochemical hydrogen storage capability”, Physics Letters A 339 (2005) 370- 377. [32] H. Takagi, “Adsorptive hydrogen storage in carbon and porous materials”, Materials Science and Engineering B108 (2004) 143-147.] [33] Quan-Hong Yang, Peng-Xiang, “Adsorption and capillarity of nitrogen in aggregated multi-walled carbon nanotubes” Chemical Physics Letters 345 (2001) 18-24. [34] Shuangping Yi, Haiyan Zhang, “The electrochemical properties of LaNi5 electrodes with multi-walled carbon nanotubes synthesized by chemical vapor deposition and treated at different temperatures in a nitrogen atmosphere”, Physica B 373 (2006) 131-135. [35] F. Haschka, W. warthmann, “Alkaline batteries for hybrid and electric vehicles”, Journal of Power Sources 72 (1998) 32-36. [36] N. Cui, J.L. Luo, “An AC impedance of self-discharge mechanism of nickel-metal hydride (Ni-MH) battery using Mg2Ni-type hydrogen storage alloy anode”, Electrochimica Acta 45 (2000) 3973-3981. 118 Experimental results and discussion [37] F. Feng, D. O. Northwood, “Self-discharge characteristics of a metal hydride electrode for Ni-MH rechargeable batteries”, International Journal Hydrogen Energy 30 (2005) 1367-1370. [38] Jon-Ha Lee, Ki-Young Lee, “Self-discharge behavior of sealed Ni-MH batteries using MmNi3.3+xCo0.7Al1.0-x anodes”, Journal of Alloys and Compounds 232 (1996) 197-203. [39] Xiaojuan Fu, Haiyan Zhang, “The effect of carbon nanotubes on the electrochemical hydrogen storage performance of LaNi5 rare earth alloy”, Physica E 25 (2005) 414-420. [40] Weixiang Chen, Zhiyuan Tang, “Effects of surface treatment on performances of metal hydride electrodes and Ni/MH batteries”, Journal of Power Sources 74 (1998) 34-39. [41] V. Ya. Davydov, N. Sheppard, “Reversible hydrogen chemisorption at near- ambient temperatures and pressures on unsaturated and aromatic hydrocarbon complexes with finely divided metals”, International Journal Hydrogen Energy 29 (2004) 1157-1164. [42] M. Bououdina, D. L. Sun, “Improved kinetics by the multiphase alloys prepared from laves phases and LaNi5”, Journal of Alloys and Compounds 288 (1996) 229- 237. [43] M. Latroche, A. Percheron-Guegna, “Influence of cobalt content in MmNi4.3- xMn0.3Al0.4Cox alloy (x=0.36 and 0.69) on its electrochemical behaviour studied by neutron diffraction”, Journal of alloys and Compounds 293-295 (1999) 637-642. 119 Experimental results and discussion [44] K.V. Romanenko, A. Fonseca, S. Dumonteil, “Xe NMR study of Xe adsorption on multiwall carbon nanotubes”, Solid State Nuclear Magnetic Resonance Volume 28, Issues 2-4, September 2005, Pages 135-141. [45] Li Chen, Men-zhen Qu, “PC-mediated shortening of carbon nanotubes”, Materials Letter 58 (2004) 3737-3740. [46] Yao Wang , Jun Wu, Fei Wei, “A treatment method to give separated multi-walled carbon nanotubes with high purity, high crystallization and a large aspect ratio”, Carbon 41 (2003) 2939–2948. [47] Zhejuan Zhang, Z. Sun, “Improve the field emission uniformity of carbon nanotubes treated by ball-milling process”, Applied Surface Science (2006). [48] C. H. Peng, M. Zhu, “Microstructure and hydrogen storage properties of a multi- phase Ml0.7Mg0.3Ni3.2 hydrogen storage alloy”, Journal of Alloys and Compounds 375 (2004) 324-329. [49] Patrice Guay, Barry L. Standsfield, “On the control of carbon nanostructures for hydrogen storage applications”, Carbon 42 (2004) 2187-2193. 120 Conclusions Chapter 5 5.1 Conclusions A number of important performance parameters, such as activation behavior, discharge capacity, self-discharge, state of charge, depth of discharge, high rate charge-discharge capability are improved by the addition of CNTs. It is believed that the CNTs have good electrochemical and thermal conductivity, and they introduce large surface area into the electrodes [1]. All these characteristics lead to an improvement in the performance of the Ni-MH battery system. In this chapter a summary of the main results from this work is presented which is followed by future directions for research in this field. This thesis concludes with potential usage of carbon nanotube based electrodes in industrial/commercial applications. 5.2 A brief summary of main results In spite of the rather poor quality of the electrical contact between the active material and the current collector, a great deal of novel information has been obtained in this study. A maximum specific discharge capacity of 243 mAh/g was obtained in this investigation. In addition, it was also possible to achieve 220 charge-discharge cycles as compared to 50 cycles in similar experimental conditions [1-2]. Experimental results have shown that the best performance for hydrogen storage capacity was observed with 121 Conclusions 5% CNTs and 20-40 nm diameter CNTs in the negative electrodes. The improved performance of Ni-MH batteries with different concentrations and diameters of CNTs are briefly summarized below: Electrode Performance 5% CNTs Fast activation performances Fast charging (20 minutes). Maximum specific discharge capacity, 243 mAh/g Higher discharge midpoint voltage, 1.269 V Cycle life with higher specific discharge capacity Excellent high rate discharge characteristics up to 10 C rate 10% CNTs Low self-discharge rate Better high rate discharge characteristics up to 10 C rate Fast charging (20 minutes) 20-40 nm CNTs Fast activation performances Fast charging (20 minutes). Maximum specific discharge capacity, 243 mAh/g Excellent high rate discharge characteristics up to 10 C rate 10-20 nm CNTs stable cycle life with highest specific discharge capacity A few additional features observed in this study are described below along with interpretation wherever possible: 122 Conclusions Deterioration in specific discharge capacity was mainly caused by the poor electrical connection between the active material and the current collector in this study. Specific discharge capacity also decayed due to a decrease in the amount of active material in the negative electrode caused by corrosion of AB5 alloy. Activation characteristics improve with the addition of CNTs. This phenomenon indicates that CNTs provide large surface area and faster rate of electron transfer. In Ni-MH battery CNTs also absorb hydrogen by chemisorption process. In addition, CNTs prevent further oxidation during repeated charge-discharge cycles. High rate charge-discharge capabilities and kinetic factors such as ohmic resistance, resistance from electrolyte, and proton and hydrogen diffusion improve with the addition of CNTs. Improvement of high charge-discharge capabilities indicate that CNTs provide large effective surface area. It is assumed that during high rate charge-discharge cycles CNTs provide better conductive network between the active materials. An optimum percentage of CNTs plays an important role in improving the performance of a MH electrode. A high percentage of CNTs (15% to 20% CNTs of active material of Ni- MH battery) in this study led to poorer electrochemical performances. The addition of high percentage of CNTs results in additional resistance due to slower hydrogen diffusion, lower density and poor contact of active materials. On the contrary, low percentage of CNTs addition (e.g. 2%) to the electrode does not alter the charging-discharging behavior significantly. Therefore lower amounts of CNTs also lead to poor electrochemical 123 Conclusions characteristics. In this investigation, optimum electrochemical performances were observed with 5% CNTs electrode. Overall, the 20-40 nm diameter CNTs showed the best electrochemical performances. It is assumed that 20-40 nm CNTs introduced higher surface area as well as reaction activity as compared to larger diameter CNTs. Hydrogen enters into the CNTs mainly through diffusion and it is believed that the chemisorption of hydrogen on carbon nanotubes surface, allows hydrogen to have access from both ends of CNTs. It is likely that the hydrogen absorbed with formation of covalent bonds at the CNTs surfaces. The layers on the walls of CNTs, are in direct proportion to their diameter, and too many layers might inhibit the diffusion of hydrogen. Consequently, hydrogen absorption-desorption will be poor. As a result, CNTs with a smaller diameter have a better electrochemical performance. But when the diameter is too small, such as 10- 20 nm, hydrogen storage is limited to a few layers only resulting in capacity decay. Best hydrogen absorption capacity was observed in 20-40 nm CNTs. It can be concluded from the above discussion that the electrochemical characteristics of Ni-MH battery can be improved considerably by the addition of 5%, 20-40 nm CNTs in the negative electrode. Though carbon nanotubes are a promising material for use in hydrogen storage, the mechanisms and optimum conditions for hydrogen adsorption onto carbon nanotubes are not yet fully understood. At this point it is not economically feasible to produce electrochemical cells utilizing carbon nanotubes. These experimental results, however, 124 Conclusions indicate that it should be possible in a not too distant future. The main areas of research required to advance this material into widespread research are the tube size, diameter, impurities, temperature and pressure affects on hydrogen adsorption and desorption behavior. 5.3 Future Directions Based on the results of present research, the following areas are recommended for future work: x The behavior of each material especially corrosion or oxide formation during repeated charge-discharge cycles needs to be investigated. In order to obtain information on the basic reactions related to elemental corrosion, each reaction process corresponding to different experimental condition should be investigated separately. Electrochemical measurement using “cyclic voltammogram” is one for the powerful methods for analyzing this specific reaction. x To understand the basic mechanisms of electrochemical performances, such as charge and mass transfer process during high rate charge-discharge research needs to be carried out using “AC impedance” machine. x Investigation on the self-discharge rates at different temperatures should be carried out in order to find out the optimum percentage and diameter of CNTs. x Investigate the pressure composition isotherm of different diameter CNTs and different percentage of CNTs at different pressure and temperature. 125 Conclusions x Performance on the electrochemical experiments with purified MWNTs and SWNTs should be conducted to find out the optimum performances of the Ni-MH battery with CNTs. x Since CNTs are very thin tubes, the dispersion of CNTs is usually very difficult. Further research should emphasize on improving the dispersion of CNTs into the metal hydride alloy and Ni powder. The following experiments might be carried out to conduct homogeneous distribution of CNTs into the alloy powder: .1. High energy ball milling of CNTs for fewer minutes (5-10min) might be conducted to break down the CNTs clusters. Advantage of using these parameters might be helpful to nanotubes remaining intact. Afterwards CNTs electrode can be made mixing alloy powder with CNTs using mechanical grinding or sheer mixer or homogenizer. .2. The CNTs suspension could be prepared with ethanol and the dispersion of CNTs in the solution can be obtained by proper sonication for several hours to get small bundles of CNTs. After those CNTs can be mixed with ball milled or as received alloy powder to get low viscosity paste and can be used sheer mixer or homogenizer to mix properly. Research shows ball milling followed by ultrasonic dispersion can give many more agglomerates CNTs smaller than 10 µm [3]. 126 Conclusions 5.4 Potential applications in electric vehicles Pure electric vehicles are already using Ni-MH battery, e.g, Toyota, Honda, Mitsubishi, Mazda, Subaru, and Nissan. Hydride electrical vehicle also use Ni-MH battery for starting and at low speeds, and use gasoline for higher speeds [4-5]. Toyota has developed a hybrid electric vehicle with Ni-MH battery as well as Mazda. If above mentioned vehicles become more popular due to environmental concerns and along with the need of greater fuel efficiency, Ni-MH battery could become a major energy supplier for the transportation system. The development of the electric vehicles requires the development of a battery pack with high energy density, high specific power and long cycle life. 127 Conclusions References [1] Perfect communication with Dr. Q. Song, visiting research fellow, UNSW. [2] Perfect communication with Dr. Z. M. Wang, visiting research fellow, UNSW. [3] Yao Wang , Jun Wu, Fei Wei “A treatment method to give separated multi-walled carbon nanotubes with high purity, high crystallization and a large aspect ratio”, Carbon 41 (2003) 2939–2948. [4] Tomohiko Ikeyaa, Nobuyuki Sawadab, “Multi-step constant-current charging method for an electric vehicle nickel/metal hydride battery with high-energy efficiency and long cycle life”, Journal of Power Sources 105 (2002) 6-12. [5] F. Haschka, W. Warthmann, “Alkaline batteries for hybrid and electric vehicles”, Journal of Power Sources 72 (1998) 32-36. 128