Page 1

Development of Lithium Battery Materials

Using Phase Separated Electrolytes

相分離電解質を用いたリチウム二次電池材料の開発

February, 2008

by

Hiroki Nara

Graduate School of Science and Engineering, Waseda University

Applied chemistry major

i

Preface

Recently, the development of rechargeable batteries indicating higher

performance are strongly desired as the power supplies of mobile equipment

evolving remarkably, such as cellular phones, lap top PC and so on. The batteries

with high energy density should be achieved when lithium metal anode is applied to

batteries; because the redox potential of lithium is lowest, indicating highest

operation voltage, and electrochemical equivalent of lithium is large, indicating

high gravimetric capacity. The primary batteries with Li metal anode has been

put into practical use, and used widely. However, when the Li metal is applied to

anode for the secondary battery and charge-discharge cycles are repeated, the

dendritic lithium is deposited on the anode. The dendritic deposition of lithium

could cause the short-circuit in the battery, resulting in ignition and explosion, and

the capacity loss, so the lithium metal secondary batteries have not put into

practical use. Thus, new electrolytes and anodes have been investigated because

of the dendritic deposition control of Li.

In the field of electrolytes, polymer electrolytes and inorganic solid

electrolytes are actively investigated as substitutes of the organic electrolytes with

the danger of the ignition and the explosion by the short-circuit. Especially, the

polymer gel electrolytes with relatively higher ionic conductivity are most

practicable in these electrolytes at present. In addition, electrolytes are required

to be enough strength mechanically for physical suppression of Li dendritic

deposition and formability. However the ionic conductivity and the mechanical

strength of polymer electrolytes are incompatible characters; it is difficult to obtain

the polymer electrolyte with both high ionic conductivity and good mechanical

strength. As the preparation technique of electrolytes with the both

characteristics, the polymer blend method to obtain the polymer blend that has

mutual co-continuous micro phase separated structures is proposed, and a PEO/PS

polymer blend gel electrolyte that consists of Polyethylene oxide (PEO) and

Polystyrene (PS) is reported as an electrolyte that meets the requirement.

However the PEO/PS polymer blend gel electrolyte has an apprehension to lose

PEO, which is the ionic conductive phase, from PS matrix by the dissolution of PEO

with the addition of a large amount of plasticizer.

In the field of anodes, the carbon anode where the dendritic deposition was

controlled by storing the lithium as not the lithium metal but lithium ion has

appeared on the market widely as a product. However its capacity at operation

has almost reached the theoretical capacity of 372 mAh / g. Thus the

ii

developments of new anode materials for higher capacity batteries are desired.

Tin anode has attracted attention as an alloy anode system, because of its high

theoretical capacity of 994 mAh / g, about 2.7 times the carbon anode. The

expansion and constriction of tin anode at alloying and dealloying with lithium lead

to its disintegration: as the results, a remarkable decrease in discharge capacity

because of the electrical charge and discharge cycle occurs. The disintegration

with the charge and discharge cycle is a matter for its practical application.

In this thesis, phase separated electrolytes were adopted to the material

preparation technique, to improve the retention of ionic conductive phase on the

polymer gel electrolyte, composed of PEO and PS, and charge-discharge cycle

durability of a tin anode. On polymer gel electrolytes, to improve the retention of

ionic conductive phase, two approaches were attempted. One is the application of

a PEO-LiX complex to the ionic conductive phase in the polymer blend gel

electrolyte. The other is the adoption of a PEO-PS di-block copolymer, composed of

PEO chain and PS chain covalently bonded, instead of blended PEO and PS. On

tin anodes, to improve its charge-discharge cycle durability, a porous tin anode was

electrodeposited with phase separated electrolyte bath composed of a surfactant and

a Sn solution. These obtained materials were characterized and discussed.

iii

Outline of this thesis

Chapter 1 is a general introduction to the thesis. Progress of lithium

secondary batteries, especially electrolytes and anodes in development of secondary

batteries with high power and capacity and their issues were described to clarify the

standpoint of this thesis.

In chapter 2, PEO-LiBF4 complex was applied to the ionic conductive phase

in the polymer blend gel electrolyte by the addition of LiBF4 in hot-blending process.

The PEO-LiBF4 complex formation was confirmed with Raman spectra. The

mutual co-continuous micro phase separated structures was also indicated in the

case of the PEO-LiBF4/PS polymer blend system. The electrochemical properties

required as an electrolyte for lithium batteries were characterized.

In chapter 3, the PEO-PS di-block copolymer was employed as polymer

electrolyte material. The PEO-PS diblock copolymer gel electrolyte was

demonstrated to have a micro phase separated structure by its self-assembling; and

prevented PEO phase to flow out from PS matrix when soaked in plasticizer for

PEO. The PEO-PS di-block copolymer gel electrolyte was characterized. It was

revealed that an increase of the interfacial resistance between electrolyte and

lithium originated in liquefied PEO, and the fixed PEO on PS matrix was not affect

the increase of the interfacial resistance.

In chapter 4, the potential of mesoporous tin anode to improve its cycle

durability and rate property is discussed. The mesoporous structure was

introduced to tin anode by electrodepositing with phase separated electrolyte bath

composed of a surfactant and a Sn solution. The reason of improvement on the

cycle durability and rate property of tin anode was discussed.

Chapter 5 summarizes the results from chapter 2 to chapter 4, and finally

indicates the future perspectives of this research.

After the final chapter, a work that is not the subject of screening is added

as an appendix. The downsizing possibility of a hybrid power source composed of a

direct methanol fuel cell and an electric double layer capacitor is discussed by

numerical simulation.

iv

Contents

Preface......................................................................................................................i

Chapter 1:

General Introduction................................................................................................... 1

1.1 Background............................................................................................................ 3

1.2 Polymer electrolytes .............................................................................................. 6

1.2.1 Basic properties of polymer electrolytes.......................................................... 6

1.2.2 Polymer gel electrolytes................................................................................... 9

1.3 Anode materials................................................................................................... 16

1.3.1 Carbon anodes ............................................................................................... 16

1.3.2 Alloy anodes ................................................................................................... 19

1.4 Phase separated structure .................................................................................. 26

1.4.1 Polymer blend ................................................................................................ 26

1.4.2 Liquid crystals ............................................................................................... 30

1.5 Summary ............................................................................................................. 37

Reference ................................................................................................................... 38

v

Chapter 2:

PEO / PS polymer blend gel electrolyte .................................................................... 47

2.1 Introduction ......................................................................................................... 49

2.2 Experimental ....................................................................................................... 52

2.3 Results and Discussion........................................................................................ 54

2.3.1 Confirmation of phase separated structure .................................................. 54

2.3.2 Confirmation of PEO-LiBF4 complex formation............................................ 55

2.3.3 Electrochemical properties of IPN gel electrolyte......................................... 57

2.4 Summary ............................................................................................................. 68

Reference ................................................................................................................... 69

Chapter 3:

PEO-PS diblock polymer gel electrolyte................................................................. 71

3.1 Introduction ......................................................................................................... 73

3.2 Experimental ....................................................................................................... 74

3.3 Results and Discussion........................................................................................ 76

3.3.1 Confirmation of phase separated structure .................................................. 76

3.3.2 Electrochemical properties of IPN gel electrolyte......................................... 77

3.4 Summary ......................................................................................................... 91

Reference ................................................................................................................... 92

vi

Chapter 4:

Mesoporous Sn Anode electrodeposited with lyotropic liquid crystals .................... 93

4.1 Introduction ......................................................................................................... 95

4.2 Experimental ....................................................................................................... 96

4.3 Results and Discussion........................................................................................ 99

4.3.1 Confirmation of mesoporous structure.......................................................... 99

4.3.2 Electrochemical properties of mesoporous Sn anode .................................. 101

4.4 Summary ........................................................................................................... 113

Reference ................................................................................................................. 114

Chapter 5:

General Conclusion ................................................................................................. 117

Appendix ............................................................................................................. 123

Trial for Hybrid System of Fuel cell and Capacitor

List of Achievements ........................................................................................... 141

Acknowledgements.............................................................................................. 147

Chapter 1:

General Introduction

Chapter 1

3

1.1. Back Ground(1, 2)

A battery is composed of several electrochemical cells that are connected in

series and/or in parallel to provide the required voltage and capacity, respectively.

Each cell consists of a positive and a negative electrode (both sources of chemical

reactions) separated by an electrolyte solution containing dissociated salts, which

enable ion transfer between the two electrodes. Once these electrodes are

connected externally, the chemical reactions proceed in tandem at both electrodes,

thereby liberating electrons and enabling the current to be tapped by the user.

The amount of electrical energy, expressed either per unit of weight (Wh/kg) or per

unit of volume (Wh/l), that a battery is able to deliver is a function of the cell

potential (V) and capacity (Ah/kg), both of which are linked directly to the chemistry

of the system. The cell potential is determined by Gibbs free energy change, -ΔG.

G-

zEF

∆

=

where z is electron number for reaction, F is Faraday constant.

When the anode materials which have strong reducing power and the cathode

materials which have strong oxidizing power are combined, -∆G increases, and the

electromotive force increases. The reducing and oxidizing power are expressed by

the electrode potential, therefore the electromotive force is determined by the

potential difference between the anode and cathode.

The capacity of a battery is determined by electrochemical equivalent,

because the electrochemical reaction follows Faraday law. Therefore the high

electromotive force and capacity battery that is high energy density battery can be

achieved by applying the anode which has low redox potential and small

electrochemical equivalent and the cathode which has high redox potential and

small electrochemical equivalent.

As described above, the voltage and current of a battery is directly related

to the types of materials applied in the electrodes and electrolyte. The propensity

of an individual metal or metal compound to gain or lose electrons in relation to

another material is known as its electrode potential. Thus the strength of

oxidizing and reducing agents are indicated by their standard potential.

Compounds with a positive electrode potential are used for anodes and those with a

negative electrode potential for cathodes. The larger the difference between the

electrode potentials of the anode and cathode, the greater the electromotive force of

the battery and the greater the amount of energy that can be produced by the

battery.

Chapter 1

4

High specific energies are theoretically available with most metals in the

first, second third principal groups of the Periodic Table, but only metallic lithium(3)

has found wide application. Among the various existing batteries (Fig. 1.1.1),

lithium based batteries – because of their high energy density and design flexibility

– currently outperform other systems, accounting for 63 % of worldwide sales values

in portable batteries(4).

Figure 1.1.1 Comparison of the different battery technologies in terms of volumetric

and gravimetric energy density.(1)

Chapter 1

5

The energy properties of metallic lithium are the most advantageous because

its redox potential of -3.01 V(3) with respect to the standard hydrogen electrode is

the highest and its atomic mass is small. Of all the alkali metals, it is also the

least reactive toward organic solvents. Lithium can be electrodeposited from

solutions of its salts in organic solvents with a 100 % yield. However this does not

mean that such an electrode can be used in a storage battery offering a large

number of charge-discharge cycles, because the metal is often deposited as a loose,

inhomogeneous dendrite layer containing badly with the operating electrode (the

remaining mass of metal). Moreover, some of the deposited metal may simply fall

away from the electrode surface to the bottom of the cell. So the main problem

when looking for a reversible, multiple-use lithium cell is to find a suitable lithium

anode at which lithium can be deposited in a compact, electron-conducting form at

least several hundred times, i.e. an anode with service life, which means almost

100 % useful recovery. Thus, in order to achieve a reversible lithium battery, a

suitable electrolyte and/ or anode must be developed.

Chapter 1

6

1.2. Polymer electrolytes(5)

1.2.1. Basic properties of polymer electrolytes

Although, tremendous research activities by Sony leading to

commercialization of lithium batteries incorporating liquid electrolytes took place

over the last decade(6), many advantages of solid polymer electrolytes over their

liquid counterparts such as organic solutions and inorganic and molten salts can be

seen. The possibility of internal shorting, leaks, and producing combustible

reaction products at the electrode surfaces, existing in the liquid electrolytes, is

eliminated by the presence of a solid polymer electrolyte. Nevertheless, the

polymer electrolytes should exhibit ionic conductivities, at least, of the order of 10-3

to 10-2 S/cm at room temperature and play the role of a separator, played by the

liquids. The polymer should also allow good cycle lives, low temperature

performances, and good thermal and mechanical strengths in order to withstand

internal temperature and pressure buildup during the battery operation. The

polymers, in general, being light-weight and non-combustible materials can be

fabricated to requirements of size and shape, thus offering a wide range of designs.

Since stable thin films of the polymers can be easily made, high specific energy

(low-mass) and high specific power (less volume) batteries can be expected for use in

electric devices and EV. Internal voltage drops may be low, about 50 mV at 10

mA/cm2, when films as thin as 40 µm are made(7, 8).

The field of polymer electrolytes has gone through three stages: “dry solid

systems”, “polymer gels”, and “polymer composites”. The “dry systems” use the

polymer host as the solid solvent and do not include any organic liquids. The

“polymer gels” contain organic liquids as plasticizers which with a lithium salt

remain encapsulated in a polymer matrix, whereas the “polymer composites”

include high surface area inorganic solids in proportion with a “dry solid polymer”

or “polymer gel” system. In this thesis, more detailed aspects of the polymer gels

system with individual examples are discussed below. The advancement of the

field can be found in literature publications and reviews(7, 9-11).

Chapter 1

7

Ionic conductivity

One of the important properties of a polymer electrolyte leading to its

development activity is the ionic conductivity. Temperature dependence on

conductivity of amorphous polymer electrolytes generally follows the

Vogel-Tammann-Fulcher (VTF) equation

(

)0

2

1

exp TTB

AT

-

-

=

-

σ

where, T0 is the glass transition temperature of the polymer electrolyte measured

by DSC, T is the temperature of measurement, A is the pre-exponential factor, and

E is the activation energy which can be evaluated either from the configurational

entropy theory or the free-volume theory, and hence relates to the segmental motion

of polymer chains(12). The ionic conductivity is usually measured by AC impedance

techniques(13).

Another important property of the polymer electrolyte is the lithium ion

transference number (tLi+) which ideally for lithium battery applications should be

unity. A value of tLi+ lower than 1 would tend to develop concentration gradients at

electrode surfaces leading to limiting currents. Thus, both the parameters, ionic

conductivity and lithium ion transference number are important in order to choose

a polymer electrolyte for a practical lithium battery. The maximum power

obtainable in a lithium cell can be related to the conductivity of the electrolyte,

whereas, the maxi-mum limiting current that can be drawn from the cell and the

cycleability of the cell can be related to the tLi+. The lithium transference number

(and its associated diffusion coefficient) measurements are usually made using

techniques such as concentration cell method(14, 15), Tubandt method(16, 17),

LiNMR(18), and electrochemical(19, 20), and electrophoretic NMR techniques(21, 22).

The techniques have their advantages and disadvantages and differ by theoretical

models used for interpretation of data. Analysis of some of the problems and

limitations associated with some of the techniques have been described by Fritz and

Kuhn(23).

An understanding to the interactions of the various species in the polymer

electrolytes has made possible to choose appropriate polymer hosts, complexing

salts and salt concentrations so that the ionic conductivities are optimized.

Evidence of salt dissolution into a polymer host is mainly based on spectroscopic

methods such as IR and Raman(24, 25). The anion-cation and ion-polymer

associations in both crystalline and amorphous phases have been studied using both

Chapter 1

8

these which monitor changes in the vibrational modes of the polymer host and the

anions(26-29). For example, in the case of lithium triflate (LiCF3SO3)(28) and lithium

bis(trifluoromethane sulfonyl)imide [LiTFSI] complexes(27)

with oxyethylene

containing polymer chains, SO3 symmetric stretching modes, CF3 bending modes,

and Li-O stretching modes were associated to ion-ion interactions and information

on ion pairing and ion association was obtained(30).

X-ray and neutron diffraction methods have given de-tailed information on

the structural aspects of coordination around the ions in crystalline phases of

polymer electrolytes(31-33). The EXAFS has revealed the local environment of the

ion(34) in both crystalline and amorphous phases, including the existence of ion pairs

such as [MX]0, [M2X]+, [MX2]-, etc. and the determination of bond lengths of the

ion-polymer interactions(35). The EXAFS technique is more useful when theoretical

models depicting the most probable situations are at hand so that fitting of the

EXAFS spectrum of known structures similar to the chemical environment of the

polymer electrolyte is possible.

Stability of polymer electrolytes

The electrochemical stability of an electrolyte is one of the essential

parameters when rechargeable lithium batteries are concerned. The instability in

the electrolyte is known to bring out irreversible reactions and capacity fading in

the battery(36). The stability commonly expressed as “electrochemical stability

window” of the polymer electrolyte (units of volts) should be the same as that of the

electrode potential or higher so that overpotentials during re-charge are

compensated by the higher stability window. For example, for a 4-V lithium

battery a window of at least 4.5 V vs. Li / Li+ in the polymer electrolyte is required

for compatibility with a lithium metal anode and lithium-ion intercalating electrode

materials.

Thermal and mechanical stabilities of a polymer electrolyte during

charge-discharge cycles are vital for a safe and endurable battery. During charge

and discharge, heat is known to get generated in the battery which increases the

surface area of an existing passive layer at the electrode surface(37). The heat can

also melt and degrade the polymer electrolyte within the battery and cause internal

short circuits. Increase in heat due to environmental factors during storage

increases the self discharge reactions in the battery and shorten life.

Chapter 1

9

1.2.2. Polymer gel electrolytes

In 1973, the first measurements on conductivities of poly(ethylene oxide)

(PEO) complexes with alkali metal salts were made by Wright et al.(38, 39). It was

after Armand that the potential of these new materials were realized for future

battery applications(40). PEO with high molecular weight of about 5 x 106 and 80 %

crystallinity was usually employed as the polymer host to form complexes with

lithium salts. Apart from the ability of the sequential oxyethylene group,

-CH2-CH2-O- in PEO to complex with lithium salts, polymer hosts containing

sequential polar groups such as O-, -NH- and -C-N- in the polymer chain were also

found to dissolve lithium salts(41).

The lithium salt complexed-poly(ethylene oxide) (PEO)(42)

and

-poly(propylene oxide) (PPO)(43) are the most widely investigated “dry solid polymer”

electrolyte systems in all solid state lithium batteries. The main reason to choose

these two polymer hosts is because they form more stable complexes and possess

higher ionic conductivities than any other group of solvating polymers without the

addition of organic solvents. Complex formation in PEOn, -salt (n = number of

ether oxygens per mole of salt) is governed by competition between solvation energy

and lattice energy of the polymer and the inorganic salt(44). Low lattice energies of

both the polymer and the complexing salt have been found to increase stabilities in

the resultant polymer electrolyte(44).

On dry solid polymer systems, to enhance the ionic conductivity, the

improvement of host polymers was attempted by the reduction of its crystallinity.(7,

45-50)

The performances of electrochemical cells incorporating the “dry solid”

polymer electrolytes and lithium metal electrodes were not satisfactory, and cycle

lives were as low as 200 to 300 cycles. The poor performance of the cells was

mainly attributed to the poor conductivity of the electrolytes, along with the

reactivity of the anion of the electrolyte towards the lithium metal electrodes.

Therefore, the polymer gel systems, whose crystallinity is reduced by adding

plasticizer, have attracted attention.

Chapter 1

10

In 1975, Feuillade and Perche demonstrated the idea of plasticizing a

polymer with an aprotic solution containing an alkali metal salt in which the

organic solution of the alkali metal salt remained trapped within the matrix of the

polymer(51). Such mixings have resulted into formation of gels with ionic

conductivities close to those of the liquid electrolytes, arising from similar

conductivity mechanisms taking place in both the systems. However, one would

expect slightly lower conductivities in the polymeric gels, if more viscous solvents

are used, as compared to those used in the liquid electrolytes.

Since then, various polymeric hosts such as, poly-(vinylidene fluoride)

(PVdF)(52), poly(vinylidene carbonate) (PVdC), poly(acrylonitrile) (PAN)(53, 54),

poly(vinyl chloride) (PVC)(55), poly(vinyl sulfone) (PVS)(56), poly(p-phenylene

terepthalamide) (PPTA)(57), and poly(vinyl pyrrolidone) (PVP), have been found to

form electrolytes with conductivities ranging between 10-4 and 10-3 S/cm at 20 oC

(Table 1.2.1)(7). These systems are presently expressed in various terms such as

“plasticized polymer electrolytes”, “polymer hybrids”, “gelionics” and “gel

electrolytes”. The electrolytes are easily prepared by heating a mixture containing

the appropriate amounts of the polymer, solvents and a lithium salt to about 120 -

150 oC, a temperature above the glass transition temperature of the polymer, to

form viscous clear liquids. Films of the gels are usually made by solution casting

in hot state allowing the solution to cool under pressure of electrodes.

Less-evaporating solvents, such as ethylene carbonate (EC), propylene carbonate

(PC), dimethyl formamide (DMF), diethyl phthalate (DEP), di-ethyl carbonate

(DEC), methylethyl carbonate (MEC), dimethyl carbonate (DMC), γ-butyrolactone

(BL), glycol sulfite (GS), and alkyl phthalates have been commonly investigated as

“plasticizer” solvents for the gel electrolytes(7). The solvents have been used

separately or as mixtures.

The phenomenon of plasticization was found to increase the amorphous

phase in the polymer system with a single glass transition temperature as low as

-40 oC, the latter varying with the amounts of solvent and polymer containing in the

gels. It was generally observed that up to 80 % of solvent could be trapped into the

polymer matrix. The high permittivity solvents allowed a greater dissociation of

the lithium salt and increased the mobility of the cation. Much research work in

the area of gel electrolytes can be found in literature reviews(7, 44, 58).

Chapter 1

11

Table 1.2.1 Ionic conductivities of some gel polymer electrolytes and polymer

composite electrolytes.

Chapter 1

12

PEO-based gels

Plasticization of high molecular weight P(EO)n-LiX electrolytes with PC

and/or EC was found to form soft solids with poor mechanical stabilities, although

room temperature conductivities as high as the order of 10-3 S/cm were obtained(59,

60). The poor mechanical stability was accounted to be mainly due to the solubility

of the PEO in the solvents(60).

Cross-linking of polymers by methods such as, UV(61), thermal radiation(62),

photo-polymerization(63), and electron beam radiation polymerization(64) was found

to reduce the solubility of the polymers with the organic solvents and also helped to

trap the liquid electrolyte within the polymer matrix. Low molecular weight PEO

was cross-linked and plasticized with 50 % PC by Borghini et al.(65). This material

showed good mechanical proper-ties and conductivities of the order 10-4 S/cm at 20

oC were two orders of magnitude higher than the unplasticized amorphous

cross-linked PEO-LiClO4

complex. In general, the room temperature

conductivities of gels based on polymers and copolymers prepared by crosslinking

methods were found to be in the range of 10-5 - 10-4 S/cm(65).

Plasticizers such as dioctyl-, dibutyl-, and dimethyl-phthalate were recently

investigated for PEO-LiClO4

complexes(66). In comparison with the three

plasticizer-based systems, lowest crystallinity in the gel containing the dioctyl

phthalate (DOP) was observed showing the best room temperature conductivity of

9.67 x 10-5 S/cm for the composition (PEO)8-(LiClO4 : DOP, 99.9 : 0.1 wt.%).

However, the room temperature conductivities of the complexes containing the

dibutyl phthalate and the dimethyl phthalate were slightly lower than those

containing the DOP plasticizer.

Chapter 1

13

PAN- and PVDF-based gels

The PAN-based(67-69) and PVdF-based(70) gel electrolytes are the most widely

investigated polymer gel electrolyte systems and remarkable conductivities of the

order of 10-3 S/cm at 20 oC have been obtained. A fully amorphous gel of

PAN-LiClO4 (1 : 0.2) in EC showed room temperature conductivity of 1 x 10-3 S/cm

and an estimated activation energy of 86 kJ/mol(71). Because of the absence of any

oxygen atoms in the PAN polymer matrix, the PAN-based gels were found to have

lithium ion transference numbers greater than 0.5(72). With a greater dissociation

of salts such as, LiTFSI and LiTFSM in

Figure 1.2.1 Probable interaction of the lithium ion in gel electrolytes.

PAN-based gels, transference numbers as high as 0.7 could be obtained. The

delocalization of electron density around the large organic anions in the salts is

believed to promote greater dissociation and increase the transference number.

Strong polar groups in the polymer are undesirable because of the

complexation of lithium ion by both the polymer matrix and the solvent(73). A

probable interaction of the lithium ion with the polymer chain and the solvent is

shown in Fig. 1.2.1 for a PAN-based gel system(74). Thus, the polymer in the gel

electrolytes would mainly act as an encapsulating matrix with electrostatic

interactions with the solvated lithium salt so that the lithium ion mobility is least

hindered. Yang et al.(71) studied a PAN-LiClO4-DMF gel in order to understand the

interaction of the Li+ ions with the PAN chain. Using FTIR they showed the Li+

ions to form bonds with C=N groups of the PAN as well as C=O of the DMF which

would support the structure in Fig. 1.2.1. Using the same technique, in

PAN-LiClO4-DMF, Wang et al. reported interactions between Li+ ions and the

oxygen and/or nitrogen atoms of DMF along with interactions between the O atoms

of DMF and N atoms of the nitrile of PAN(75). Unfortunately, no interactions

between LiClO4 and PAN could be detected with the use of the FTIR technique.

Chapter 1

14

Jiang et al. recently studied PVdF-based gels consisting of EC, PC and a LiX

salt(76). As expected they found the presence of plasticizers EC and PC to

significantly disorder the crystalline structure and reduce crystallinity in the gels

than in the parent polymer host, PVdF. They also reported the mechanical

strength of the resulting gel to be depended on the PVdF content, whereas the

conductivity to be mainly influenced by the viscosity of the medium and the

concentration of the lithium salt. Room temperature conductivities as high as 2.2

x 10-3 S/cm for the LiN(CF3SO2)2 salt containing mixture were reported(76).

Increase in gel properties were observed when mixtures of polymer hosts or

copolymers were used in the gel preparation. Increase in gel properties were

observed for PVdF co-polymerized with hexafluoropropylene (HFP). The

PVdF-HFP co-polymer in the gel showed greater solubility for organic solvents, and

lower crystallinity and glass transition temperature than the PVdF polymer alone

in the gel(77).

Others

Wieczorek and Stevens studied blends polyether, PMMA, and LiCF3SO3

(78).

The electrolytes showed a maximum room temperature conductivity of 3 x 10-5 S/cm.

Morita et al. also prepared gels consisting of PMMA grafted with PEO and

containing Li salts, and room temperature conductivities of the order 10-3 were

reported(79). Addition of crown ethers such as 12-crown-4 and 15-crown-5 to the

PMMA-PEO-Li+ gel was found to increase the conductivity to some extent, whereas

the Li ion mobility was found to increase significantly with the addition of the

15-crown-5 ether(79). Lithium transference numbers greater than 0.5 were found in

the gel electrolytes based on poly(methyl methacrylate)(80)

and

poly(tetrahydrofuran)(81). The values were found to be dependent on the amounts

of organic solvent present in the gel.

Blends of polymers, PVC and PMMA were studied using PC as the

plasticizer and LiCF3SO3 as the salt(82). Due to insolubility of PVC in the solvent

PC, phase separation was observed. The inclusion of PVC into PMMA helped to

increase the mechanical stability of the gel, but unfortunately decreased the lithium

ion conductivity. The ions were found to preferentially move towards the

plasticizer-rich phase or the PMMA-rich phase.

Dimensionally stable gels consisting of PEG-PAN-PC-EC-LiClO4

were

prepared by Munichandraiah et al.(83). Compared with gels PEO-PC-LiClO4 and

Chapter 1

15

PAN-PC-EC-LiClO4, the PEG containing gels showed lower room temperature

conductivities, but higher mechanical stabilities.

PVC-based electrolytes consisting of LiClO4, LiTFSM, or LPF6 salt and

THF/PC mixture as plasticizer gave ionic conductivity of the order 10-4 S/cm at 20 oC

with respective lithium ion transference numbers of 0.26, 0.40, and 0.45(84). Polymer

gels consisting of PVC, LiTFSI, and solvents such as dibutyl phthalate (DBP) and

dioctyl adipate (DOP) gave room temperature conductivities of the order 10-4

S/cm(85).

Low molecular ethers such as poly(ethylene glycol) dimethyl ether

(PEGDME) was used as a solvent for salts such as LiN(CF3SO2)2, LiCF3SO3, and

LiPF6 to which poly(vinylidene fluoride)-hexafluoropropene copolymer was added as

the encapsulating polymer matrix(86). PEGDME was also added to a copolymer

formed from triethylene glycol dimethacrylate (TRGDMA) and acrylonitrile (AN)

and LiCF3SO3 salt. Room temperature conductivities in the range from 10-5 to 10-4

S/cm were reported(87). The conductivity was found to increase in line with the

molar ratios of both AN : TRGDMA and (EO) : LiCF3SO3

(87).

In the gel mixture consisting of poly(p-phenylene terephthalamide) (PPTA),

polyethylene glycol (PEG), polycarbonate in PC-EC, and a lithium salt,

conductivities as high as 2.2 x 10-3 S/cm at room temperature were observed at 0.8

M LiBF4 salt per mole of PPTA content(57). Above 1 M of the salt content, the

conductivity was found to fall rapidly suggesting the ion conductivity to be due to

LiBF4 interaction at the amide bond sites of the PPTA(57).

In the polymer gel electrolytes, the incorporation of liquid electrolytes into

homo- or co-polymer hosts has allowed room temperature conductivities as high as

10-3 S/cm. The mechanical stability of the gels is determined by the ratios of the

polymer and solvent (plasticizer) in the gel. Highly mechanically stable gels with

10 - 12 wt.% of plasticizer and 70-80 wt.% of polymer, with room temperature

conductivities between 10-4

and 10-3

S/cm have been generally observed.

Plasticizer solvents such as EC and PC have been much studied due to their high

polarity and low vapor pressure, which also allow greater plasticizing effect to the

polymer host. Most of the studied plasticizers have shown electrochemical

instabilities at lithium metal electrode surfaces, and thus a search for new

plasticizers is seen necessary. Due to greater dissociation of the lithium salt, much

higher lithium ion transference numbers than in the solvent-free electrolytes (about

0.6) have been observed.

Chapter 1

16

1.3.

Anode materials(88, 89)

1.3.1. Carbon anodes

Since the lithium ion secondary battery with carbon anode had

commercialized by Sony Corporation in 1991(6), the carbon anode has been

investigated. At present mostly carbons are used as the negative electrode of

commercial rechargeable lithium batteries: i) because they exhibit both higher

specific charges and more negative redox potentials than most metal oxides,

chalcogenides, and polymers; and ii) due to their dimensional stability, they show

better cycling performance than Li alloys. The insertion of lithium into carbon,

habitually named “intercalation”, proceeds according to Equation (1).

(1)

Due to electrochemical reduction (charge) of the carbon host, lithium ions

from the electrolyte penetrate into the carbon and form a lithium/carbon

intercalation compound, LixCn. The reaction is reversible.

The quality of sites capable of lithium accommodation strongly depends on

the crystallinity, the microstructure, and the micromorphology of the carbonaceous

material(90-107). Thus, the kind of carbon determines the current/potential

characteristics of the electrochemical intercalation reaction, and also the risk of

solvent co-intercalation.

Carbonaceous materials suitable for lithium intercalation are commercially

available in hundreds of types and qualities(90, 91, 108-111). Many exotic carbons have

been synthesized in laboratories by pyrolysis of various precursors, some of them

with a remarkably high specific charge. Fullerenes have also been tested(112, 113).

Carbons that are capable of reversible lithium intercalation can roughly be

classified as graphitic and non-graphitic (disordered). Graphitic carbons are

carbonaceous materials with a layered structure but typically with a number of

structural defects. From a crystallographic point of view the term “graphite” is

only applicable to carbons having a layered lattice structure with a perfect stacking

order of graphene layers, either the prevalent AB (hexagonal graphite, Fig. 1.3.1) or

the less common ABC (rhombohedral graphite). Due to the small transformation

energy of AB into ABC stacking (and vice versa), perfectly stacked graphite crystals

are not readily available. Therefore, the term “graphite” is often used regardless of

stacking order. The actual structure of carbonaceous materials typically deviates

LixCn

xLi+ + xe- + Cn

discharge

charge

Chapter 1

17

more or less from the ideal graphite structure. Materials consisting of aggregates of

graphite crystallites are named graphite as well. For instance, the terms natural

graphite, artificial or synthetic graphite, and pyrolytic graphite are commonly used,

although the materials are polycrystalline. The crystallites may vary considerably

in size. In some carbons, the aggregates are large and relatively free of defects, for

example, in highly oriented pyrolytic graphite (HOPG). In addition to graphitic

crystallites, other carbons also include crystallites containing carbon layers (or

packages of stacked carbon layers) having significant misfits and misorientation

angles of the stacked segments to each other (turbostratic orientation or

turbostratic disorder)(114). The latter disorder can be identified from an increased

average planar spacing compared to graphite(115).

Figure 1.3.1 Left: schematic drawing of the crystal structure of hexagonal graphite,

showing the AB layer stacking sequence and the unit cell. Right: view perpendicular

to the basal plane of hexagonal graphite. Prismatic surfaces can be subdivided into

arm-chair and zig-zag faces.

Non-graphitic carbonaceous materials consist of carbon atoms that are

mainly arranged in a planar hexagonal network but without far-reaching

crystallographic order in the c-direction. The structure of those carbons is

characterized by amorphous areas embedding and crosslinking more graphitic

ones(116) (Fig. 1.3.2). The number and the size of the areas vary, and depend on

both the precursor material and the manufacturing temperature. Non-graphitic

carbons are mostly prepared by pyrolysis of organic polymer or hydrocarbon

precursors at temperatures below ~1500 oC. Heat treatment of most non-graphitic

(disordered) carbons at temperatures from ~1500 to ~3000 oC allows one to

Chapter 1

18

distinguish between two different carbon types. Graphitizing carbons develop the

graphite structure continuously during the heating process, as crosslinking between

the carbon layers is weak and, therefore, the layers are mobile enough to form

graphite-like crystallites. Non-graphitizing carbons show no true development of

the graphite structure even at high temperatures (2500-3000 oC), since the carbon

layers are immobilized by crosslinking. Since non-graphitizing carbons are

mechanically harder than graphitizing ones, it is common to divide the

non-graphitic carbons into “soft” and “hard” carbons(116).

Figure 1.3.2 Schematic drawing of a non-graphitic (disordered) carbonaceous

material.

Chapter 1

19

1.3.2. Alloy anodes

Various insertion materials have been proposed for negative electrodes of

rechargeable lithium batteries, for example, transition-metal oxides and

chalcogenides, carbons, lithium alloys, and polymers. Table 1.3.1 shows that both

the specific charges and the charge densities of lithium insertion materials are

theoretically lower than that of metallic lithium. However, considering that the

cycling efficiency of metallic lithium is ≤ 99 %, one has to employ a large excess of

lithium(91, 117, 118) to reach sufficient cycle life. The practical charge density of a

secondary lithium electrode is therefore much lower than the theoretical one, so

that it is comparable with the charge densities of alternative lithium-containing

compounds. However, the potential of the electrode materials also has to be

considered because a higher potential versus Li/Li+ of the negative electrode means

a lower cell voltage. For instance, the potential of many Li alloys is ~0.3 to ~1.0 V vs.

Li/Li+ whereas it is only ~0.1 V vs. Li/Li+ for graphite (Fig. 1.3.3).

Table 1.3.1 Characteristics of representative negative electrode materials for

lithium batteries, calculated by using data from(117, 119-121). The values are for fully

lithiated host materials except for the values in parentheses, which are for

lithium-free host materials. Li4 denotes a four-fold lithium excess, which is

necessary to reach a sufficient cycle life.

[a] In some cases a considerably lower amount of the specific charge/charge density can be cycled

reversibly in practice.

Chapter 1

20

The replacement of metallic lithium by lithium alloys has been under investigation

since Dey(122) demonstrated the feasibility of electrochemical formation of lithium

alloys in liquid organic electrolytes in 1971. The reaction usually proceeds

reversibly according to the general scheme shown in Equation (2).

(2)

With only a few exceptions (such as hard metals, M = Ti, Ni, Mo, Nb), Li alloys are

formed at ambient temperature by polarizing the metal M, for example, Al, Si, Sn,

Pb, In, Bi, Sb, Ag, and some multinary alloys,(93, 120, 123-134) sufficiently negatively in a

Li+- containing electrolyte. In most cases even the binary systems Li-M are very

complex. Sequences of stoichiometric intermetallic compounds and phases LixM

with considerable phase range are usually formed during lithiation of the metal M,

characterized by several steps and/or slopes in the charge diagram (Fig. 1.3.3). The

formation of Li-M phases is in many cases reversible, so that subsequent steps and

slopes can also be observed during discharge.

Figure 1.3.3 Charging curves of some matrix metals (M), compared to highly

oriented turbostratic mesophase pitch carbon fibers (P 100, FMI Composites-Union

Carbide, characterized in LiClO4 / propylene carbonate). Modified and redrawn

from(135).

LixM

xLi+ + xe- + M

discharge

charge

Chapter 1

21

A Li+ ion transfer cell with the trademark Station, announced recently by

Fujifilm Celltech Co., Ltd.,(136) uses an “amorphous tin-based composite oxide

(abbreviated TCO or ATOC)” for the negative electrode. The TCO combines both: i) a

promising cycle life and ii) a high specific charge (>600 Ah/kg) and charge density

(>2200 Ah/L). (137)

The TCO is synthesized from SnO, B2O3, Sn2P2O7, Al2O3, and other

precursors. However, only the SnII compounds in the composite oxide are said to

form the electrochemically active centers for Li insertion. The oxides of B, P, or Al,

which are electrochemically inactive, have glass forming properties and form a

network stabilizing the dimensional integrity of the composite host during

synthesis.(137) The improvement of cycle life by using composite amorphous lithium

insertion materials, such as V2O5, together with a network former, such as P2O5, has

been reported in the literature.(138-144)

In order to explain the high specific charge a mechanism can be suggested

in which the tin oxide reacts to form Li2O and metallic Sn.(145, 146) This reaction is

associated with large charge losses due to the irreversible formation of Li2O. In a

second step the Sn then alloys with lithium reversibly. On the other hand, according

to Fujifilm Celltech(137) no Li2O was found after lithium insertion. However, the idea

that the high specific charge of the TCO is due to the alloying of metallic tin has led

to a renaissance of Li-alloy negative electrode research and development.(145)

The electrochemical reduction of a tin electrode in a Li+-containing

electrolyte leads to the subsequent formation of a number of intermetallic phases

LixSny at high temperature (415 oC)(147) as well as at room temperature(145, 148-150).

Diffusion data (Table 1.3.2) disclose that the Li+ cation mobility in lithium-rich

phases is quite acceptable both at 415 oC(148) and at room temperature(123, 132)

allowing reasonable charge/discharge current densities. However, whereas very

slow measurements at near equilibrium conditions allow to characterize several

LixSny phases (particular at 415 oC, cf. data in Table 1.3.2), at room temperature

and under practical charging conditions only the lithium-poor phases Li2Sn5 and

LiSn can be clearly distinguished in the constant current charge curve (Figure

1.3.4) as well as in X-ray diffraction patterns.(145, 151) It has been suggested that the

lithium-rich phases LixSny do not form long-range ordered structures, because the

atom mobility is too low at room temperature in this case.(145, 151)

Chapter 1

22

Table 1.3.2 Melting points, densities from X-ray data dx), specific charges, charge

densities as well as plateau potentials and maximum chemical diffusion coefficients

((Dchem-max) of several LixSny phases at 415 oC (LiCl-KCl eutectic melt as electrolyte)

and 25 oC (LiAsF6/PC or LiClO4/PC as electrolytes).

LixSny

x in

LixSn

Melting

point

(oC)

dx

(g/cm3)

Specific

charge

(Ah/kg)

Charge

density

(Ah/L)

Plateau potential

for formation

at 415 oC

(V vs. Li/Li+)

Dchem-max

at 415 oC

(cm2/s)

Plateau potential

for formation

at 25 oC

(V vs. Li/Li+)

Dchem-max at

25 oC

(cm2/s)

Li2Sn5

0.4

319

6.11

88.3

539.5

N/A

N/A

~0.760

N/A

LiSn

1

≥485

5.10

213.3

1087.8

0.569

4.10x10-6

0.660

8.0x10-8

‘LI3Sn2’

1.5

465

*

*

*

*

*

*

*

Li7Sn3

2.33

500

3.67

463.6

1701.4

0.456

4.07x10-5

0.530

N/A

Li5Sn2

2.5

≥698

3.54

492.5

1743.5

0.430

5.89x10-5

0.485

5.0x10-7

Li13Sn5

2.6

716

3.46

509.6

1763.2

0.390

7.59x10-4

0.485

N/A

Li7Sn2

3.5

783

2.96

656.0

1941.8

0.284

7.76x10-5

0.420

N/A

‘Li4Sn’

4

≥684

*

*

*

*

*

*

*

Li22Sn5

4.4

758

2.56

790.2

2022.9

0.170

1.91x10-5

0.380

5.9x10-7

*Existence doubtful

Figure 1.3.4 Characteristic charge curve of electroplated Sn in 1 M LiClO4/PC,

i =0.025 mA/cm2.

The lithium-rich phases exhibit high theoretical specific charges and charge

densities (Table 1.3.2). On the other hand, the tin pulverizes rapidly when it

experiences full lithiation due to the large density decrease (is volume increase).

Limiting the degree of lithium uptake to the stoichiometry of LiSn relieves the

Chapter 1

23

pulverization problem to some extent, because the differences in density between

metallic Sn (7.29 g/cm3) and the phases Li2Sn5 and LiSn are relatively small (Table

1.3.2). Moreover, these phases are structurally related(152), so that the consequences

of reconstitution reactions are small as well. However, further alloying with lithium

beyond the stoichiometry LiSn leads to a rapid density decrease (Table 1.3.2) and

causes major structural rearrangements, which considerably enlarge the

mechanical stresses on the host material. In order to improve the dimensional

stability and thus the rechargeability for higher lithium uptakes, the morphology

and chemistry of the host material has to be specifically designed.

Host metal properties, such as particle size, shape, texture, porosity, etc.

strongly affect the macroscopic dimensional stability during lithium alloying and

dealloying and thus the cycling behavior.(135, 153, 154) Though the volume expansions of

the metal hosts upon alloying with lithium are in the order of several 100 %, large

absolute volume changes can be avoided, when the size of the metallic host particles

is kept small (Fig. 1.3.5).

The practical feasibility of this concept was checked by employing tin

anodes of different particle sizes. These have been prepared by electroplating on

copper substrates from aqueous solutions containing Sn2+ cations. The morphology

of the metals can be drastically changed by a control of the plating conditions, for

example, by variation (i) of the chemical composition of the solutions (e.g. by

addition of leveling and complexing agents), (ii) of the solution temperature, (iii) of

the stirring conditions and (iv) of the plating current densities. The detailed

preparation procedures can be found elsewhere.(135, 154)

The thickness of the

deposits is in the range of a few micrometers, i.e. too thin for practical application in

current cylindrical and prismatic lithium ion cells. Nevertheless, since they are

binder-free they are good model substrates for understanding the relation between

metal anode properties and electrochemical performance.

Chapter 1

24

Figure 1.3.5 Model: 1st lithiation (is 1st alloying with lithium) of a loosely packed

small particle size metallic material. Even 100% volume expansion of the

individual particles does not crack them as their absolute changes in dimensions

are still small.

The cycling performance of lithium alloys can be significantly improved if

intermetallic and/or composite hosts are employed instead of pure metals. The

basic idea is that at a certain stage of charge/discharge, i.e. at a certain electrode

potential, one (or more) components/phases of the composite or intermetallic are

able to store lithium (`reactant'), i.e. expand/contract, whereas the other

components/phases are less active or even inactive(88), i.e. perform as a `matrix'

buffering the expansions of the reactant (Fig. 1.3.6). Additional preconditions for a

successful application of this concept are (i) that the reactant is finely dispersed in

the matrix, (ii) that the matrix allows the electrons and lithium ions to move

between the reactive domains or is a mixed (electron and lithium cation) conductor

and (iii) that the reactive domains are sufficiently small, so that the above discussed

benefits of `small particle size' are effective.

Figure 1.3.6 Model: strong expansions of the `reactant' domains due to lithiation

can be buffered by the inactive or less active `matrix' domains, thus keeping the

extent of crack formation in the overall multiphase material small.

Thus, the selection of an adequate matrix is the key to a successful Sn

compound anode. Elements that are inactive against Li are assumed to suppress

the volume change effectively without much irreversible capacity. The use of Sn

Chapter 1

25

compounds with elements such as Ni(155-157), Fe(158-162), Cu(163-166), Mn(160, 167), and

Co(160) has been investigated based on this assumption. Recently, an amorphous

ternary Sn-Co-C anode has been introduced to a practical use.(168)

Chapter 1

26

1.4. Phase separated structure

1.4.1. Polymer blend

A kind and quantity of the polymer materials were rapidly developed in

these dozens of years. We can not live without the polymer materials over a wide

area. However, the polymer materials are said to be reached the age of puberty, the

characteristic of polymer are demanded to meet a broad applications. However,

the performance as the single material has already come to the limit. The composite,

copolymerization, and polymer blend of polymers are needed as the metallic and

inorganic materials had done. The multicomponent polymer will satisfy enough

demanded performance by a suitable combination from many existing polymers.

Polymer blend composed of two immiscible polymers(169, 170)

A co-continuous morphology is a non-equilibrium morphology that is

generated during mixing of two polymers. As such, it is an unstable morphology,

and it starts changing through filament break-up and retraction as soon as the fluid

blend comes out of the mixer. However, the blend may remain co-continuous, if it

is frozen fast. Considerable attention has been given to the conditions that make

co-continuous morphologies possible in blends during the mixing. It has been

generally suggested that co-continuity occurs at the phase inversion point.

Existing empirical relations(171-173) and theories(174) give a volume fraction for phase

inversion as a function of the viscosity ratio, as shown in Fig. 1.4.1. However,

many experimental results(170) cannot be described with these relations (Fig. 1.4.1).

By basing the phase inversion point on the viscosity ratio of the components only,

these relations neglect the influence of the blending conditions and material

properties, e.g. the interfacial tension. Moreover, they do not take into account any

requirements as to the shape of the dispersed component necessary to obtain

co-continuity. Especially at low volume fractions, a co-continuous network can

only exist if the minor blend component consists of structures with an extended

shape(170). These structures can be formed and remain so only under appropriate

blending conditions. For this reason, it is to be expected that the existence of a

co-continuous morphology in the blend will be strongly dependent both on the

processing conditions, e.g. the stress levels, and the processing properties of the

blend components, e.g. the matrix viscosity, viscosity ratio and interfacial tension.

In the literature(170), an equation has been derived that describes the critical volume

Chapter 1

27

fraction of the minor phase for complete co-continuity as a function of the matrix

viscosity, interfacial tension, shear rate and phase dimensions, and it was shown

that a high viscosity of the matrix component was favorable for co-continuity over a

broad composition range in blends of commercial grades of polyethylene and

polystyrene.

Figure 1.4.1.Phase inversion as a function of the viscosity ratio, p = ηd/ηm, according

to several empirical relations summarized. * and ●: points found for PE/PS systems

in Ref. (170).

Chapter 1

28

Polymer blend composed of a di-block copolymer (175)

Block copolymers have recently received much attention not only thanks to the

scale of the microdomains (tens of nanometers), their various chemical and physical

properties but also due to the convenient size and shape tunability of microdomains

afforded by simply changing their molecular weights and compositions. Many

potential uses of block copolymers for different nanotechnologies have been proposed

based on principally their ability to form interesting patterns. However, the main

challenge of using block copolymers lies with control of microstructure. Achievement

of precise microdomain location, orientation, and elimination of various defects requires

introduction of external fields during the processing step. A variety of mechanical,

electrical, magnetic biases and surface interactions have been proposed to manipulate

and guide the microstructures of block copolymers.

Block copolymers consist of chemically distinct polymer chains covalently

linked to form a single molecule. Owing to their mutual repulsion, dissimilar blocks

tend to segregate into different domains, the spatial extent of the domains being limited

by the constraint imposed by the chemical connectivity of the blocks. Area

minimization at the interface (the IMDS) of two blocks takes place to lower the

interfacial energy. From an entropic standpoint, the molecules prefer random coil

shapes but the blocks are stretched away from the IMDS to avoid unfavorable contacts.

As a result, of these competing effects, self-organized periodic microstructures emerge

on the nanoscopic length scale. Various microdomain structures are achieved,

depending on relative volume ratio between blocks and chain architecture as well as the

persistence lengths of the respective blocks.

In the simplest case of non-crystalline flexible coil AB di-block copolymers, the

composition of the AB di-block (i.e. the volume fraction f of block A) controls the

geometry of the microdomain structure. As shown in Fig. 1.4.2, for nearly symmetric

di-blocks (f ~ 1/2) a lamellar phase occurs. For moderate compositional asymmetries, a

complex bi-continuous state, known as the double gyroid phase, has been observed in

which the minority blocks form domains consisting of two interweaving

threefoldcoordinated networks. At yet higher compositional asymmetry, the minority

component forms hexagonally packed cylinders and then spheres arranged on a

body-centered cubic lattice. Eventually, as f → 0 or 1, a homogeneous phase results.(176)

Chapter 1

29

Figure 1.4.2. Schematic phase diagram showing the various ‘classical’ block

copolymer morphologies adopted by non-crystalline linear diblock copolymer. The

blue component represents the minority phase and the matrix, majority phase

surrounds it.(177)

Chapter 1

30

1.4.2. Liquid crystals(178)

Polyoxyethylene surfactants are widely used as emulsifying agents and

detergents. Like ionic surfactants they form micelles above a critical concentration

in water (the c.m.c.) with liquid crystals frequently occurring at higher

concentrations. They are unusual in having a lower consolute temperature, termed

the cloud point. For many years this was attributed to the presence of giant

micelles. However, it is now thought to correspond to the phase separation of a

concentrated surfactant solution containing small micelles from a more dilute

solution, perhaps due to the presence of significant intermicellar

attractions.(179-181) Some surfactants even show a ‘double cloud point’, where a

1 % aqueous solution goes cloudy then clears and clouds a second time when the

temperature is increased.(182)

Over years a number of theories to describe micelle shapes have been

published.(183, 184)

These involve a balance of alkyl-chain/water repulsions and

repulsion between adjacent head groups within the micelle, together with surface

curvature and limitations due to alkyl-chain packing. Most lyotropic liquid crystals

consist of ordered micelles.(185)

It should be possible to extend the theories of

micelle shapes to account for mesophase formation by addition of appropriate terms

for intermicellar forces. Parsegian(186) has already successfully described the

hexagonal/lamellar transition of ionic surfactants using Poisson-Boltzmann theory

to account for electrostatic repulsions. Wennerstram and co-workers(187, 188)

have developed this approach further. For non-ionic surfactants there is no

quantitative theory to describe intermicellar repulsions. However, with

poly(oxyethylene) alkyl ethers [n-CnH2n+1(OCH2CH2)mOH,CnEOm] it is possible to

change systematically the surfactant chemical structure by alteration of m and n,

hence varying head-group interactions and micelle size. By determining

water/surfactant phase diagrams of a range of compounds, the validity of this

theoretical description and the importance of each contributing factor (alkyl-chain

conformations, curvature, head-group area etc.) for the structure and stability of

mesophases can be assessed.

The shape of micelles just above the c.m.c. is highly dependent upon

surfactant type and solution conditions (concentration, electrolyte level,

temperature). It appears that spherical micelles, rod micelles and oblate spheroid

(bilayer) micelles all occur, but under different circumstances. The structures of

Chapter 1

31

normal hexagonal (H1), lamellar (Lα) and reversed hexagonal (H2) lyotropic

mesophases are well known.(185, 189)

The lamellar phase consists of equidistant

parallel surfactant bilayers separated by water layers. In the hexagonal phases

very long normal or reversed rod micelles are packed in a hexagonal array. Two

different classes of cubic phases have been discovered,(185, 190) one occurs at

compositions between micellar solution (L1) and H1 phases, consisting of ordered

spherical micelles packed in body- or face-centred arrays (labelled I1), and the

second occurs between Lα and H1 (labelled V1) or H2 (labelled V2). While the

structures of these phases are not fully known, the best candidates(185, 190) are

regular ‘bicontinuous’ networks where the chain/water interface has both positive

and negative curvatures.

The forces responsible for micellar shape and mesophase structure can be

divided into intramicellar and intermicellar contributions. The former determine

the shape just above the c.m.c. while the latter take account of intermicelle

interactions at higher concentrations. The description given below is a summary

and development of several treatments.(183-185, 190, 191)

At the c.m.c. micelle shape is determined by the surface area per molecule

(a) at the alkyl-chain/water interface and the interface curvature. The possible

micelle shapes are spheres, rods (prolate spheroids) or discs (oblate spheroids,

bilayer micelles). For a spherical micelle, the volume of the surfactant alkyl chain

(v), micelle radius (r) and a are related by

3

ar

v =

........................................................................................................(1)

Since r cannot extend beyond the ‘all-trans’ length of alkyl chain (lt), while v is

invariant, then a cannot fall below a certain limit (as

c). From the known densities of

alkanes, assuming a micelle radius of 1.25 Ǻ per C-C bond this is estimated to be as

c

= 70 Ǻ 2. Similar arguments apply to rod micelles, with the limit (ar

c ) being ar

c =

47 Ǻ 2. Thus if a ≥ 70 Ǻ, the possible micelle shapes are spheres, rods or discs. With

a in the range 70 > a/ Ǻ 2 ≥ 47, only rod or disc micelles can form. With a < 47 Ǻ 2,

rod micelles are excluded and only bilayer micelles (or reversed phases) can form.

This limitation of micelle type is generally referred to as the ‘alkyl-chain packing

constraint’.

The detailed shapes adopted by rod and disc micelles are not known.

Prolate and oblate ellipsoids, or rods with hemi-spherical ends and bilayers with

hemi-cylindrical edges, are both popular pictorial representations. It has been

Chapter 1

32

argued that only ellipsoids with low eccentricity can occur because alkyl chains are

unable to fill up completely the highly curved regions at the edge of ellipsoids with

high eccentricity.(183)

This conclusion involves an assumption that the alkyl-chain

axis is, on average, normal to the interface. There seems to be no physical reason

why a few chains at the ellipsoid–micelle edge should not be tilted to allow packing

into the highly curved region, and thus the restraint is too severe.

Figure 1.4.3 Schematic illustration of forces at the micelle surface (see text). Plane x

represents alkyl-chain/water interface, inter-head-group repulsions act in plane y.

Given that the value of a is sufficiently large for any particular shape to

occur, the actual form adopted is determined by surface curvature. The forces

present at the chain/water interface (intramicellar forces) can be represented as

occurring in different planes, as shown schematically in Fig. 1.4.3. The

chain/water interface is at x. Interactions between hydrated head groups

(electrostatic, solvation, steric) usually result in a repulsive force in some plane y,

further into the water than x. The minimization of hydrocarbon-chain/water

interactions gives an attractive force in plane x. These are dominant contributions.

However, depending on the value of a, there will also be an apparent

repulsive force in some plane z within the alkyl-chain region. This arises from the

unfavorable entropy change caused by restrictions on the conformations of the

chains with small a values. While no calculations of the magnitude of this effect

have been published, it is obvious that it will increase with temperature and with

alkyl-chain length, since the number of available conformations is proportional to

3n-2 (n = chain carbon number). The effect will also depend on the shape of the

micelle. Without detailed model calculations of chain conformations it is not

possible to estimate this contribution to the stability of spheres, rods or bilayers at

large a value (a > as

c). However, any restriction on the chain taking the all-trans

conformation will increase the limiting values as

c, and ar

c. Moreover, with a > as

c,

the average chain length within a binary is < lt/3. This will involve numerous

y

x

z

Chapter 1

33

gauche conformations which have an unfavorable enthalpy, so disfavoring the

lamellar phase. For low values of a, spheres and rods cannot pack. On comparing

reversed phases and the lamellar phase as a approaches the limit of the all-trans

cross-sectional area within the lamellar phase (ab

c), the reversed phases will be

favored at sufficiently high temperatures because they have more conformational

states available to the chains. Thus in practice the alkyl-chain curvature effect is

likely to transform bilayers into reversed structures when a is too small for rods or

spheres to occur.

The contributions listed above can be represented by an equation of the

form

g

a

a

rC +

+

=

)(

0

γ

μ

..........................................................................................(2)

where μo is the energy per surfactant molecule, and C(r) is a curvature free-energy

expression, being a function of micelle radius; the alkyl-chain/water repulsion is

given by γa, where γ represents surface tension, and all other contributions are

included in g. One could employ expressions involving higher powers of a but

these do not qualitatively alter the description,(183) nor does inclusion of an

allowance for the area of the chain/water interface occupied by the head groups.(184)

To summarize, the balance of forces indicated in Fig. 1.4.3 determines the

aggregate shape within packing constraints. A large repulsion in plane y gives

water continuous phases while small repulsions give reversed micelles. With large

repulsions spherical micelles occur, where a (sphere) < a (bilayer) [provided that a

(sphere) ≥ as

c]. When a is just too small for spheres, one expects rod micelles

(prolate ellipsoids), and as a is reduced further, oblate spheroids (bilayers) with a <

ar

c. The particular values of a where shape changes occur will depend on the

detailed nature of the forces in planes x and y (and z). Reversed phases are

expected at still smaller values of a.

There are two major effects of intermicellar forces. First, if we consider

the micelles formed at the c.m.c. as hard-core particles, then as the volume fraction

of micelles is increased, we will observe order/disorder transitions. Spherical

micelles will close-pack into a regular cubic array. Long rod micelles will form a

hexagonal array, while a lamellar phase will form with bilayers. The limiting

volume fractions are 0.74 for face-centred cubic and 0.91 for a hexagonal phase.

Theoretically, a lamellar phase can occur without water being present, i.e. limiting

Chapter 1

34

volume fraction = 1. (In practice it is well known that anhydrous soaps form a

lamellar phase at high temperatures.(185, 189)

The actual volume fractions for

order/disorder transitions appear to be in the range 0.7-0.8 of the close-packed

volume values(192) (i.e. ca. 0.5 and ca. 0.7 for spheres and rods, respectively). For

lamellae, two large bilayers in any volume will be aligned to some extent because

they cannot pass through each other. Thus the lowest theoretical volume fraction

for a lamellar phase is just above the c.m.c. However, van der Waals attractions

between bilayers can be sufficiently large to overcome entropy and interbilayer

repulsions (193) causing phase separation of a lamellar phase where the water layers

do not swell indefinitely. (Similar considerations apply to disordered oblate micellar

solutions, hence the occurrence of the cloud point.(180, 185)) Thus instead of a dilute

solution of large bilayers, a lamellar-phase + water dispersion occurs just above the

c.m.c. Starting with various different micellar shapes, one would expect the

following phase sequences (the numbers refer to the volume fraction of micelles

required for the phase transition):

A hexagonal phase occurs at volume fractions above the close-packing limit of

spherical micelles because the rods can relieve some of the surface strain due to

curvature better than the lamellar phase. In the second and third cases a lamellar

phase occurs only when the close-packing volume of rods is exceeded. In practice,

the limit of mesophase formation is often determined by the stability of a crystalline

surfactant phase, particularly with ionic or zwitterionic compounds. This

description has nothing to say about the crystalline state.

Of course, the repulsions between micelles are not of the hard-sphere type,

but occur at a range of distances from the micelle surface. This soft-core repulsion

can extend to many micelle diameters if ionic surfactants with low concentrations of

added electrolyte are present. For zwitterionic surfactants (lecithins) the existence

of an apparent ‘hydration’ repulsion force has been demonstrated.(193)

The force

can be thought of as arising from interactions between hydrogen-bonded water

network structures on adjacent micelle surfaces. Theoretical considerations(194)

oblate spheroid → lamellar phase

(disc micelles)

prolate spheroid → hexagonal phase → lamellar phase

(rod micelles)

spherical phase → cubic phase → hexagonal phase → lamellar phase

0-1.0

0.7

0.91

0.5

0.74

0.91

Chapter 1

35

suggest that it could have an exponential fall-off with increasing distance from the

surface. For polyoxyethylene surfactants we expect that this force will be

accompanied by another repulsive force due to limitations on the conformation of

EO groups arising from steric hindrance when micelles are close together.

One possible consequence of the soft-core repulsions is the formation of

cubic structures other than the face-centred variety. Recent results suggest that

primitive or body-centred structures can occur in addition to face-centred cubic.

Thus the minimum volume fraction for cubic structures is further reduced. On

increasing volume fraction, one might expect to observe the sequence: primitive

cubic → body-centred cubic → face-centred cubic → other phases, with the volume

fraction for these transitions being in the ratio 0.52 : 0.68 : 0.74, respectively.

However, the actual structures observed could vary from this sequence according to

the form of the intermicellar repulsions.

The soft-core repulsion between micelles has a second effect: it causes the

value of a to decrease with increasing volume fraction (unless a is limited by

alkyl-chain packing). This leads to an increase in micelle size. More importantly,

a micelle shape transition of the type sphere → rod → bilayer can occur with a > ac,

where the unfavorable curvature energy is balanced by a contribution from micelle

interactions. This type of transition will be accelerated when a approaches the

packing limits of spheres or rods. If the repulsions between the new shapes are

sufficiently large, then an ordered phase can form immediately. Thus one could

observe the sequences:

What is observed in practice will depend on the relative magnitudes of the forces

and their detailed dependence on area (a) or distance (micelle separation, i.e.

concentration). An illustration of the possible phase sequences at different

curvatures and as a function of volume fraction is given in Fig. 1.4.4. Note that the

micelle shape transitions in disordered solutions occur over a range of volume

fractions (dotted lines) while transitions involving mesophases occur at constant

volume fraction. Also, it was emphasized that mesophases occur from interactions

between micelles. If interactions are absent then ordered phases cannot form.

spherical micelles → hexagonal phase → lamellar phase

(disordered)

rod micelles→ lamellar phase

(desordered)

Chapter 1

36

So far the effects of entropy were ignored. This is because there was no obvious

model to allow for the change caused by forming small micelles from large ones

which includes the effects of alkyl-chain conformations and changes in

water/head-group orientations. However, any effect is likely to scale as kT/N,

where N is the aggregation number of the small micelle. For a given area per

molecule, N will be a function of r2, where r is the micelle radius, which in turn is

proportional to the alkyl-chain length. Thus the entropy contribution will decrease

rapidly with increasing alkyl-chain length.

Figure 1.4.4 Schematic illustration of mesophase structures as a function of

surfactant volume fraction and increasing curvature. Dotted lines indicate micelle

shape transitions. Mesophase regions are cubic (I1), hexagonal (H1) and lamellar

(Lα).

Chapter 1

37

1.5. Summary

The energy properties of metallic lithium are the most advantageous

because its redox potential is the highest and its atomic mass is small. However

when the lithium metal anode is charge and discharged several times, it is often

deposited as a loose, inhomogeneous dendrite layer containing badly with the

operating electrode. Moreover, some of the deposited metal may simply fall away

from the electrode surface to the bottom of the cell. Thus, in order to solve this

problem for achievement of a reversible lithium battery, a suitable electrolyte and/

or anode must be developed.

In the approach from electrolytes, the possibility of internal shorting, leaks,

and producing combustible reaction products at the electrode surfaces, existing in

the liquid electrolytes, is eliminated by the presence of a solid polymer electrolyte.

The polymers, in general, being light-weight and non-combustible materials can be

fabricated to requirements of size and shape, thus offering a wide range of designs.

Nevertheless, the polymer electrolytes should exhibit ionic conductivities, at least,

of the order of 10-3 to 10-2 S/cm at room temperature and play the role of a separator,

played by the liquids. The polymer should also allow good cycle lives, low

temperature performances, and good thermal and mechanical strengths in order to

withstand internal temperature and pressure buildup during the battery operation.

These problems were solved at some level by the application of gel electrolyte

system. Nevertheless these techniques need expensive materials, complicated

process, and/or hazardous solvent. Further improvements are required for

industry.

In the approach from anodes, the lithium ion secondary batteries which

indicate high power and high capacity were commercialized due to the development

of carbon anodes. Nevertheless, higher capacity is required to anodes of lithium

ion secondary batteries, whereat alloy anodes, i.e. Sn, Si, have been investigated.

The problem of these anodes, that is the capacity degradation due to the volume

change during charge-discharge cycles, are being improved by alloying of inactive

metals with lithium. Now the interests on alloy anode are next stage, that is

improvement for high rate property.

Finally the phase separated structures described in this chapter have a

tremendous amount of potential for application of electrochemistry, because these

materials can be selected as electrolyte.

In this thesis, the improvements of the materials for lithium secondary

batteries by applying the phase separated structures were discussed.

Chapter 1

38

Reference

1.

J. M. Tarascon and M. Armand, Nature, 414, 359 (2001).

2.

Axeon Power Ltd, http://www.axeonpower.com/chemistries.htm, in (2006).

3.

R. Bauer, Chemie Ingenieur Technik 44, 147 (1972).

4.

H. Takeshita, Portable Li-ion, worldwide, in Proc. Conf. Power 2000, San Diego, CA

(2000).

5.

F. B. Dias, L. Plomp and J. B. J. Veldhuis, Journal of Power Sources, 88, 169 (2000).

6.

T. Nagaura and K. Tozawa, Prog. Batteries Solar cell, 9, 209 (1990).

7.

M. Alamgir and K. M. Abraham, in Lithium Batteries - New Materials,

Developments and Perspectives, G. Pistoia Editor, p. 93, Elsevier (1994).

8.

Z. Ogumi, Y. Uchimot, E. Endo, K. Yasuda and Z. Takehara, Polym. Adv. Technol., 4,

188 (1993).

9.

M. B. Armand, Annual Review of Materials Science, 16, 245 (1986).

10.

P. G. Bruce and C. A. Vincent, Journal of the Chemical Society-Faraday

Transactions, 89, 3187 (1993).

11.

B. Scrosati, Materials Science and Engineering B-Solid State Materials for

Advanced Technology, 12, 369 (1992).

12.

M. A. Ratner, in Polymer Electrolyte Reviews-1, J. R. MacCallum and C. A. vincent

Editors, Elsevier, London (1987).

13.

P. W. M. Jacobs, J. W. Lorimer, A. Russer and M. Wasiucionek, Journal of Power

Sources, 26, 483 (1989).

14.

P. M. Blonsky, D. F. Shriver, P. Austin and H. R. Allcock, Solid State Ionics, 18-19,

258 (1986).

15.

A. Bouridah, F. Dalard, D. Deroo and M. B. Armand, Journal of Applied

Electrochemistry, 17, 625 (1987).

16.

P. G. Bruce, M. T. Hardgrave and C. A. Vincent, Solid State Ionics, 53-56, 1087

(1992).

17.

M. Leveque, J. F. Lenest, A. Gandini and H. Cheradame, Makromolekulare

Chemie-Rapid Communications, 4, 497 (1983).

18.

P. G. Bruce and C. A. Vincent, Faraday Discussions, 88, 43 (1989).

19.

J. J. Xu and G. C. Farrington, in Lithium Polymer Batteries, J. Broadhead and B.

Scrosati Editors, p. 121, Pennington, NJ (1997).

20.

P. R. Sorensen and T. Jacobsen, Solid State Ionics, 9-10, 1147 (1983).

21.

H. Dai, S. Sanderson, J. Davey, F. Uribe and T. A. Zawodzinski Jr., in Lithium

Polymer Batteries, J. Broadhead and B. Scrosati Editors, p. 111, Pennington, NJ

Chapter 1

39

(1997).

22.

T. A. Zawodzinski Jr., H. Dai, S. Sanderson, J. Davey and F. Uribe, in Lithium

Polymer Batteries, J. Broadhead and B. Scrosati Editors, p. 103, Pennington, NJ

(1997).

23.

H. P. Fritz and A. Kuhn, Journal of Power Sources, 41, 253 (1993).

24.

R. Frech, J. Manning, D. Teeters and B. E. Black, Solid State Ionics, 28-30, 954

(1988).

25.

L. M. Torell and S. Schantz, in Polymer Electrolyte Review-2, J. R. MacCallum and

C. A. vincent Editors, p. 1, Elsevier, London (1989).

26.

B. L. Papke, M. A. Ratner and D. F. Shriver, Journal of Physics and Chemistry of

Solids, 42, 493 (1981).

27.

B. L. Papke, M. A. Ratner and D. F. Shriver, Journal of the Electrochemical Society,

129, 1434 (1982).

28.

G. Petersen, P. Jacobsson and L. M. Torell, Electrochimica Acta, 37, 1495 (1992).

29.

D. Teeters and R. Frech, Solid State Ionics, 18-19, 271 (1986).

30.

W. W. Huang and R. Frech, Polymer, 35, 235 (1994).

31.

Y. Chatani, Y. Yakura and T. Ishioka, Polymer, 31, 208 (1990).

32.

P. Lightfoot, M. A. Mehta and P. G. Bruce, Science, 262, 883 (1993).

33.

P. Lightfoot, J. L. Nowinski and P. G. Bruce, Journal of the American Chemical

Society, 116, 7469 (1994).

34.

D. C. Koningsberger, X-ray absorption, Wilwy, UK (1998).

35.

R. J. Latham, R. G. Linford and W. S. Schlindwein, Faraday Discussions, 103

(1989).

36.

H. Mao, J. N. Reamers, Q. Zhong and U. von Sacken, in Rechargeable Lithium and

Lithium-ion Batters, S. Megahed Editor, p. 245 (1995).

37.

Y. Matsuda, Journal of Power Sources, 43, 1 (1993).

38.

D. E. Fenton, J. M. Parker and P. V. Wright, Polymer, 14, 589 (1973).

39.

P. V. Wright, Br. Polym. J., 7, 319 (1975).

40.

M. B. Armand, S. M. Chabagno and M. Duclot, in Second International Meeting on

Solid Electrolyte, St. Andrews, Scotland (1978).

41.

F. M. Gray, in Solid Polymer Electrolytes, F. M. Gray Editor, p. 4, VCH, New York

(1991).

42.

M. B. Armand, in Polymer Electrolyte Reviews-1, J. R. MacCallum and C. A.

Vincent Editors, p. 1, Elsevier, London (1989).

43.

J. R. MacCallum and C. A. vincent, in Polymer Electrolyte reviews-1, J. R.

MacCallum and C. A. vincent Editors, p. 23, Elsevier, London (1987).

Chapter 1

40

44.

M. B. Armand, W. Gorecki and R. Andreani, in Second International Meeting on

Polymer Electrolytes, B. Scrosati Editor, p. 91, New York (1990).

45.

L. M. Ding, Polymer, 38, 4267 (1997).

46.

L. M. Ding, J. Shi and C. Z. Yang, Synthetic Metals, 87, 157 (1997).

47.

L. M. Ding, J. Shi, C. Z. Yang and S. J. Dong, Polymer Journal, 29, 410 (1997).

48.

F. M. Gray, in Solid Polymer Electrolytes, F. M. Gray Editor, p. 183, VCH, New York

(1991).

49.

A. Killis, J. F. Lenest, H. Cheradame and A. Gandini, Makromolekulare

Chemie-Macromolecular Chemistry and Physics, 183, 2835 (1982).

50.

D. W. Kim, J. K. Park, M. S. Gong and H. Y. Song, Polymer Engineering and

Science, 34, 1305 (1994).

51.

G. Feuillade and P. Perche, Journal of Applied Electrochemistry, 5, 63 (1975).

52.

T. Nagatomo, C. Ichikawa and O. Omoto, Journal of the Electrochemical Society,

134, 305 (1987).

53.

F. Croce, F. Gerace, G. Dautzemberg, S. Passerini, G. B. Appetecchi and B. Scrosati,

Electrochimica Acta, 39, 2187 (1994).

54.

D. Peramunage, D. M. Pasquariello and K. M. Abraham, Journal of the

Electrochemical Society, 142, 1789 (1995).

55.

M. Alamgir and K. M. Abraham, Journal of the Electrochemical Society, 140, L96

(1993).

56.

H. S. Choe, J. Giaccai, M. Alamgir and K. M. Abraham, Electrochimica Acta, 40,

2289 (1995).

57.

I. Kuribayashi, M. Yamashita, S. Muraoka and K. Nagasawa, Journal of Power

Sources, 63, 121 (1996).

58.

R. Koksbang, I. I. Olsen and D. Shackle, Solid State Ionics, 69, 320 (1994).

59.

J. M. North, European Patent No. EP0279554 A2, in (1988).

60.

S. Chintapalli and R. Frech, Solid State Ionics, 86-88, 341 (1996).

61.

H. W. Rhee, W. I. Jung, M. K. Song, S. Y. Oh and J. W. Choi, Molecular Crystals and

Liquid Crystals Science and Technology Section a-Molecular Crystals and Liquid

Crystals, 294, 225 (1997).

62.

D. G. H. Ballard, P. Cheshire, T. S. Mann and J. E. Przeworski, Macromolecules, 23,

1256 (1990).

63.

M. Morita, T. Fukumasa, M. Motoda, H. Tsutsumi and Y. Matsuda, Journal of the

Electrochemical Society, 137, 3401 (1990).

64.

D. W. Xia, D. Soltz and J. Smid, Solid State Ionics, 14, 221 (1984).

65.

M. C. Borghini, M. Mastragostino and A. Zanelli, Electrochimica Acta, 41, 2369

Chapter 1

41

(1996).

66.

M. S. Michael, M. M. E. Jacob, S. R. S. Prabaharan and S. Radhakrishna, Solid

State Ionics, 98, 167 (1997).

67.

G. B. Appetecchi, F. Croce and B. Scrosati, Electrochimica Acta, 40, 991 (1995).

68.

G. B. Appetecchi, G. Dautzenberg and B. Scrosati, Journal of the Electrochemical

Society, 143, 6 (1996).

69.

B. Huang, Z. Wang, L. Chen, R. Xue and F. Wang, Solid State Ionics, 91, 279 (1996).

70.

F. Croce, G. B. Appetecchi, S. Slane, M. Salomon, M. Tavarez, S. Arumugam, Y.

Wang and S. G. Greenbaum, Solid State Ionics, 86-8, 307 (1996).

71.

C. R. Yang, J. T. Perng, Y. Y. Wang and C. C. Wan, Journal of Power Sources, 62, 89

(1996).

72.

, in Lithium Polymer Batteries, J. Broadhead and B. Scrosati Editors, Pennington,

NJ (1997).

73.

L. A. Dominey, in Lithium Batteries - New Materials, Developments and

Perspectives, G. Pistoia Editor, p. 137, Elsevier (1994).

74.

M. Alamgir and K. M. Abraham, in Lithium Batteries - New Materials,

Development and Perspectives, G. Pistoia Editor, p. 114, Elsevier (1994).

75.

Z. Wang, B. Huang, H. Huang, L. Chen and R. Xue, Solid State Ionics, 86-88, 313

(1996).

76.

Z. Jiang, B. Carroll and K. M. Abraham, Electrochimica Acta, 42, 2667 (1997).

77.

A. S. Gozdz, J. M. Tarascon, C. N. Schmutz, P. C. Warren, O. S. Gebizlioglu and F.

Shokoohi, in ECS Fall Meeting, Miami, Florita (1994).

78.

W. Wieczorek and J. R. Stevens, Journal of Physical Chemistry B, 101, 1529 (1997).

79.

M. Morita, H. Tanaka, M. Ishikawa and Y. Matsuda, Solid State Ionics, 86-88, 401

(1996).

80.

G. B. Appetecchi, G. Dautzemberg and B. Scrosati, in Workshop on Advanced

Lithium Batteries,, p. 72, Osaka, Japan (1995).

81.

M. Alamgir, R. D. Moulton and K. M. Abraham, Electrochimica Acta, 36, 773

(1991).

82.

H. J. Rhoo, H. T. Kim, J. K. Park and T. S. Hwang, Electrochimica Acta, 42, 1571

(1997).

83.

N. Munichandraiah, G. Sivasankar, L. G. Scanlon and R. A. Marsh, Journal of

Applied Polymer Science, 65, 2191 (1997).

84.

Y. Matsuda, T. Teramine and K. Nishizawa, in The 1997 Joint International

Meeting, The Electrochemical Society and The International Society of

Electrochemistry, p. 186, Paris (1997).

Chapter 1

42

85.

A. M. Sukeshini, A. Nishimoto and M. Watanabe, Solid State Ionics, 86-8, 385

(1996).

86.

K. M. Abraham and Z. Jiang, Journal of the Electrochemical Society, 144, L136

(1997).

87.

H. Bischoff, D. Rahner, K. Wiesener, T. Steurich and R. Sandner, Journal of Power

Sources, 44, 473 (1993).

88.

M. Winter, J. O. Besenhard, M. E. Spahr and P. Novak, Advanced Materials, 10,

725 (1998).

89.

M. Winter and J. O. Besenhard, Electrochimica Acta, 45, 31 (1999).

90.

J. R. Dahn, A. K. Sleigh, H. Shi, J. N. Reimers, Q. Zhong and B. M. Way,

Electrochimica Acta, 38, 1179 (1993).

91.

J. R. Dahn, A. K. Sleigh, H. Shi, B. W. Way, W. J. Weydanz, J. N. Reimers, Q. Zhong

and U. von Sacken, in Lithium Batteries - New Materials, Developments and

Perspectives, G. Pistoia Editor, p. 1, Elsevier, Amsterdam (1994).

92.

M. Endo, J. I. Nakamura, A. Emori, Y. Sasabe and K. Takeuchi, Molecular Crystals

and Liquid Crystals Science and Technology Section a-Molecular Crystals and

Liquid Crystals, 244, A171 (1994).

93.

D. Fauteux and R. Koksbang, Journal of Applied Electrochemistry, 23, 1 (1993).

94.

G. B. Li, Z. G. Lu, B. Y. Huang, H. Huang, R. J. Xue and L. Q. Chen, Solid State

Ionics, 81, 15 (1995).

95.

G. B. Li, R. J. Xue, L. Q. Chen and Y. Z. Huang, Journal of Power Sources, 54, 271

(1995).

96.

A. Satoh, N. Takami and T. Ohsaki, Solid State Ionics, 80, 291 (1995).

97.

K. Sawai, Y. Iwakoshi and T. Ohzuku, Solid State Ionics, 69, 273 (1994).

98.

N. Takami, A. Sato and T. Ohsaki, in Lithium Batteries, S. Surampudi and V. R.

Koch Editors, p. 44, Pennington, NJ (1993).

99.

N. Takami, A. Satoh, M. Hara and I. Ohsaki, Journal of the Electrochemical

Society, 142, 371 (1995).

100.

O. Yamamoto, N. Imanishi, Y. Takeda and H. Kashiwagi, Journal of Power Sources,

54, 72 (1995).

101.

O. Yamamoto, Y. Takeda, N. Imanishi and R. Kanno, in New Sealed Rechargeble

Batteries and Supercapacitor, B. m. Barnett, E. Dowgiallo, G. Halpert, Y. Matsuda

and Z. Takehara Editors, p. 302, Pennington, NJ (1993).

102.

J. Yamaura, Y. Ozaki, A. Morita and A. Ohta, Journal of Power Sources, 43, 233

(1993).

103.

R. Yazami, in Lithium Batteries - New Materials, Developments and Perspectives,

Chapter 1

43

G. Pistoia Editor, p. 49, Elsevier, Amsterdam (1994).

104.

R. Yazami and D. Guerard, Journal of Power Sources, 43, 39 (1993).

105.

R. Yazami, K. Zaghib and M. Deschamps, Molecular Crystals and Liquid Crystals

Science and Technology Section a-Molecular Crystals and Liquid Crystals, 244,

A165 (1994).

106.

T. Zheng, Y. H. Liu, E. W. Fuller, S. Tseng, U. Vonsacken and J. R. Dahn, Journal of

the Electrochemical Society, 142, 2581 (1995).

107.

T. Zheng, J. S. Xue and J. R. Dahn, Chemistry of Materials, 8, 389 (1996).

108.

S. Miura, H. Nakamura and M. Yoshio, Prog. Batteries Battery Mater., 12, 115

(1993).

109.

H. Shi, J. Barker, M. Y. Saidi and R. Koksbang, Journal of the Electrochemical

Society, 143, 3466 (1996).

110.

H. Shi, J. Barker, M. Y. Saidi, R. Koksbang and L. Morris, Journal of Power Sources,

68, 291 (1997).

111.

T. D. Tran, J. H. Feikert, X. Song and K. Kinoshita, Journal of the Electrochemical

Society, 142, 3297 (1995).

112.

L. Firlej, A. Zahab, F. Brocard and P. Bernier, Synthetic Metals, 70, 1373 (1995).

113.

S. Lemont, J. Ghanbaja and D. Billaud, Molecular Crystals and Liquid Crystals

Science and Technology Section a-Molecular Crystals and Liquid Crystals, 244, 203

(1994).

114.

B. E. Warren, Physical Review, 59, 693 (1941).

115.

R. Franklin, Acta Crystallographica, 4, 253 (1951).

116.

R. E. Franklin, Proceedings of the Royal Society of London. Series A, Mathematical

and Physical Sciences (1934-1990), 209, 196 (1951).

117.

K. Brandt, Solid State Ionics, 69, 173 (1994).

118.

K. Brandt, Journal of Power Sources, 54, 151 (1995).

119.

D. Billaud, E. McRae and A. Herold, Materials Research Bulletin, 14, 857 (1979).

120.

R. Nesper, Progress in Solid State Chemistry, 20, 1 (1990).

121.

M. S. Whittingham, Progress in Solid State Chemistry, 12, 41 (1978).

122.

A. N. Dey, Journal of the Electrochemical Society, 118, 1547 (1971).

123.

A. Anani, S. Crouch-Baker and R. A. Huggins, Journal of the Electrochemical

Society, 134, 3098 (1987).

124.

J. O. Besenhard and H. P. Fritz, Electrochimica Acta, 20, 513 (1975).

125.

A. V. Churikov, E. S. Nimon and A. L. Lvov, Electrochimica Acta, 42, 179 (1997).

126.

I. Hauke, S. Machill, D. Rahner and K. Wiesener, Journal of Power Sources, 44, 421

(1993).

Chapter 1

44

127.

R. A. Huggins, Journal of Power Sources, 22, 341 (1988).

128.

R. A. Huggins, Journal of Power Sources, 26, 109 (1989).

129.

E. S. Nimon and A. V. Churikov, Electrochimica Acta, 41, 1455 (1996).

130.

S. V. Sazhin, A. V. Gorodyskii and M. Y. Khimchenko, Journal of Power Sources, 47,

57 (1994).

131.

E. M. Shembel, I. M. Maksyuta, L. I. Neduzhko, A. I. Belosokhov, A. F. Naumenko

and V. V. Rozhkov, Journal of Power Sources, 54, 416 (1995).

132.

J. Q. Wang, I. D. Raistrick and R. A. Huggins, Journal of the Electrochemical

Society, 133, 457 (1986).

133.

W. Weppner and R. A. Huggins, Journal of the Electrochemical Society, 124, 1569

(1977).

134.

M. Morita and Y. Matsuda, Journal of Power Sources, 26, 573 (1989).

135.

J. O. Besenhard, J. Yang and M. Winter, Journal of Power Sources, 68, 87 (1997).

136.

L. Fujifilm Celltech Co., in Nippon Denki Shinbun (1996).

137.

Y. Idota, T. Kubota, A. Matsufuji, Y. Maekawa and T. Miyasaka, Science, 276, 1395

(1997).

138.

T. Shodai, Y. Sakurai and S. Okada, in Rechrgeble Lithium and Lithium-Ion

Batteries, S. Megahed and B. M. Barnett Editors, p. 224, Pennington, NJ (1995).

139.

M. Arakawa, S. Tobishima, T. Hirai and J. Yamaki, Journal of the Electrochemical

Society, 133, 1527 (1986).

140.

T. Pagnier, M. Fouletier and J. L. Souquet, Solid State Ionics, 9-10, 649 (1983).

141.

Y. Sakurai, S. Okada, J. Yamaki and T. Okada, Journal of Power Sources, 20, 173

(1987).

142.

Y. Sakurai and J. Yamaki, Journal of the Electrochemical Society, 132, 512 (1985).

143.

Y. Sakurai and J. Yamaki, Journal of the Electrochemical Society, 135, 791 (1988).

144.

S. Tobishima, M. Arakawa, T. Hirai and J. Yamaki, Journal of Power Sources, 20,

293 (1987).

145.

I. A. Courtney and J. R. Dahn, Journal of the Electrochemical Society, 144, 2045

(1997).

146.

M. Szekely, E. Bernard, E. Caillot, M. Herlem, A. Etcheberry, C. Mathieu and B.

Fahys, Journal of Electroanalytical Chemistry, 391, 69 (1995).

147.

C. J. Wen and R. A. Huggins, Journal of the Electrochemical Society, 128, 1181

(1981).

148.

C. J. Wen and R. A. Huggins, Journal of Solid State Chemistry, 35, 376 (1980).

149.

J. Sangster and C. W. Bale, Journal of Phase Equilibria, 19, 70 (1998).

150.

I. A. Courtney, J. S. Tse, O. Mao, J. Hafner and J. R. Dahn, Physical Review B, 58,

Chapter 1

45

15583 (1998).

151.

J. R. Dahn, I. A. Courtney and O. Mao, Solid State Ionics, 111, 289 (1998).

152.

C. Julien, in Lithium Batteries - Development and Perspectives, G. Pistoia Editor,

p. 167, Elsevier, Amsterdam (1994).

153.

J. Yang, M. Wachtler, M. Winter and J. O. Besenhard, Electrochemical and Solid

State Letters, 2, 161 (1999).

154.

J. Yang, M. Winter and J. O. Besenhard, Solid State Ionics, 90, 281 (1996).

155.

J. H. Ahn, Y. J. Kim, G. X. Wang, M. Lindsay, H. K. Liu and S. X. Dou, Materials

Transactions, 43, 63 (2002).

156.

O. Crosnier, T. Brousse, X. Devaux, P. Fragnaud and D. M. Schleich, Journal of

Power Sources, 94, 169 (2001).

157.

G. M. Ehrlich, C. Durand, X. Chen, T. A. Hugener, F. Spiess and S. L. Suib, Journal

of the Electrochemical Society, 147, 886 (2000).

158.

D. Larcher, L. Y. Beaulieu, O. Mao, A. E. George and J. R. Dahn, Journal of the

Electrochemical Society, 147, 1703 (2000).

159.

O. Mao and J. R. Dahn, Journal of the Electrochemical Society, 146, 414 (1999).

160.

O. Mao and J. R. Dahn, Journal of the Electrochemical Society, 146, 423 (1999).

161.

O. Mao, R. A. Dunlap and J. R. Dahn, Journal of the Electrochemical Society, 146,

405 (1999).

162.

O. Mao, R. L. Turner, I. A. Courtney, B. D. Fredericksen, M. I. Buckett, L. J. Krause

and J. R. Dahn, Electrochemical and Solid State Letters, 2, 3 (1999).

163.

K. D. Kepler, J. T. Vaughey and M. M. Thackeray, Electrochemical and Solid State

Letters, 2, 307 (1999).

164.

D. Larcher, L. Y. Beaulieu, D. D. MacNeil and J. R. Dahn, Journal of the

Electrochemical Society, 147, 1658 (2000).

165.

N. Tamura, R. Ohshita, M. Fujimoto, S. Fujitani, M. Kamino and I. Yonezu,

Journal of Power Sources, 107, 48 (2002).

166.

G. X. Wang, L. Sun, D. H. Bradhurst, S. X. Dou and H. K. Liu, Journal of Alloys and

Compounds, 299, L12 (2000).

167.

L. Beaulieu, D. Larcher, R. A. Dunlap and J. R. Dahn, Journal of Alloys and

Compounds, 297, 122 (2000).

168.

H. Inoue, in International Meeting on Lithium Batteries, IMLB 2006, Biarritz,

France (2006).

169.

R. C. Willemse, A. P. de Boer, J. van Dam and A. D. Gotsis, Polymer, 40, 827 (1999).

170.

R. C. Willemse, A. P. de Boer, J. van Dam and A. D. Gotsis, Polymer, 39, 5879

(1998).

Chapter 1

46

171.

R. M. Ho, C. H. Wu and A. C. Su, Polymer Engineering and Science, 30, 511 (1990).

172.

G. M. Jordhamo, J. A. Manson and L. H. Sperling, Polymer Engineering and

Science, 26, 517 (1986).

173.

I. S. Miles and A. Zurek, Polymer Engineering and Science, 28, 796 (1988).

174.

V. I. Metelkin and V. S. Blekht, Colloid Journal of the Ussr, 46, 425 (1984).

175.

C. Park, J. Yoon and E. L. Thomas, Polymer, 44, 6725 (2003).

176.

M. W. Matsen and F. S. Bates, Macromolecules, 29, 7641 (1996).

177.

F. S. Bates and G. H. Fredrickson, Physics Today, 52, 32 (1999).

178.

D. J. Mitchell, G. J. T. Tiddy, L. Waring, T. Bostock and M. P. McDonald, Journal of

the Chemical Society, Faraday Transactions 1, 79, 975 (1983).

179.

M. Corti and V. Degiorgio, J. Phys. Chem, 85, 1442 (1981).

180.

E. J. Staples and G. J. T. Tiddy, J. Chem. Soc., Faraday Trans. 1, 74, 2530 (1978).

181.

C. Tanford, Y. Nozaki and M. F. Rhodes, J. Phys. Chem, 81, 1555 (1977).

182.

F. Harusawa, S. Nakamura and T. Mitsui, Colloid & Polymer Science, 252, 613

(1974).

183.

J. N. Israelachvili, D. J. Mitchell and B. W. Ninham, J. Chem. Soc., Faraday Trans.

2, 72, 1525 (1976).

184.

D. J. Mitchell and B. W. Ninham, j. Chem. Soc., Faraday Trans. 2, 77, 601 (1981).

185.

G. J. T. Tiddy, Phys. Rep., 57, 1 (1980).

186.

V. A. Parsegian, Trans. Faraday Soc., 62, 848 (1966).

187.

B. Jonsson and H. Wennerstrom, J. Colloid Interface Sci., 80, 482 (1981).

188.

B. Jonsson, H. Wennerstrom and B. Halle, J. Phys. Chem, 84, 2179 (1980).

189.

P. A. Winsor, Chem. Rev., 68, 1 (1968).

190.

K. Fontell, Mol. Cryst. Liq Cryst., 63, 59 (1981).

191.

C. Tanford, J. Phys. Chem, 76, 3020 (1972).

192.

J. A. Barker and D. Henderson, Rev. Mod. Phys., 48, 587 (1976).

193.

R. P. Rand, Annual Review of Biophysics and Bioengineering, 10, 277 (1981).

194.

S. Marcelja and N. Radic, Chem. Phys.Lett., 42, 129 (1976).

Chapter 2:

PEO/PS polymer blend gel electrolyte

Chapter 2

49

2.1. Introduction

As it has been introduced in chapter 1, polymer electrolytes for lithium

secondary batteries have been studied for years for safety issues in future lithium

batteries, and gel polymer electrolytes are one of the most promising candidates to

be considered in the coming first stage because of its higher ionic condudtivity.

However, gel polymer electrolytes have two major problems to be realized in

practical use that relates to the nature of the polymer-solvent interaction. For

example, polymers with weak solvent interaction such as PVdF

(polyvinylidenefluoride)(1-5) and PAN (polyacrylonitrile)(6-12) form unstable gels, and

are likely to loose the plasticizer from the gel matrix, even though they exhibit good

mechanical strength. On the other hand, polymers with strong solvent interaction

such as PEO (polyethylene oxide)(13-18) and PMMA (polymethyl-methacrylate)(6, 19-26)

form stable gels with the plasticizer and show higher ionic conductivity than that of

solid polymer electrolytes, while the gels in the system have poor mechanical

strength. It is indeed difficult to achieve both properties of high ionic conductivity

and mechanical strength.

However, recently, many excellent studies were undertaken to tackle these

problems, i.e., a co-polymerization of the host polymer, a three-dimensional

cross-link of the host polymer, or a polymerization of room temperature molten salt

(ionic liquid)(27-33). In these studies, S. Passerini et al. proposed a polymer blending

method to get the novel polymer matrix of gel electrolyte for lithium secondary

batteries(28, 29) with low-cost, green, and easy processes. This proposed method was

to use the simple hot blending method for combining common polymers. Two

immiscible polymers, PEO and polystyrene (PS), were blended to obtain the

electrolyte demonstrating both merits of ionic conduction and mechanical strength.

The PEO was plasticized to provide the ionic conductivity whereas the PS as the

framed network to yield the membrane mechanical strength. This blended

polymer system showed co-continuous morphology. This morphology was formed

with the IPN (interpenetrated polymer network) of the two polymers. Figure 2.1.1

illustrates a schematic model for a co-continuous polymer blend system, where

lithium ion transfers through the ion conductive phase. The phase of frame of the

film keeps the mechanical strength when the ion conductive phase is plasticized.

Figure 2.1.2 shows the photographs comparing the PEO film with and without PS,

i.e., PEO film (a) and IPN film (b) after the plasticizing process. As seen in Fig.

2.1.2, the IPN film kept its shape, however the PEO films with plasticizing process

could not keep its shape due to its soft and viscous texture. This plasticized IPN

Chapter 2

50

electrolyte was expected to posses a high ionic conductivity and a good mechanical

strength, while there are some concerns to loose the ion conductive phase by flowing

out of the soft plasticized PEO phase from PS frame with the large amount of

plasticizer. In the practical use, a higher conductivity is requested, and one of the

solutions would be to increase the amount of plasticizer for gel electrolytes.

The aim of this chapter is to improve the retention of ionic conductive phase on

the PEO/PS polymer blend gel electrolyte system by the application of PEO-LiBF4

complex to the polymer blend gel electrolyte. There are some reports on that a

complex is formed when PEO and LiX are mixed(34, 35). The lithium ion serves as

the bridging of PEO chains as shown in Fig. 2.1.3. The PEO-LiX complex should

keep high viscosity in spite of impregnation with the plasticizer. Therefore, the

formation of PEO-LiX complex is to give the potential to prevent that conductive

phase runs out from the frame phase in the IPN. Characteristics of the gel

polymer electrolyte made with IPN film of PEO-LiBF4 complex and PS, and the

possibility of the electrolyte formed with this IPN matrix were examined for the

future Li metal secondary batteries.

Figure 2.1.1 Schematic model for co-continuous polymer blend interpenetrated

polymer network system.

PS : The frame of the film

PEO-LiX : Ion conductive phase

PS : The frame of the film

PEO-LiX : Ion conductive phase

Enabling ionic conduction

Imparting mechanical strength

Enabling ionic conduction

Imparting mechanical strength

Chapter 2

51

Figure 2.1.2 Photographs of the films of PEO (a) and PEO-PS IPN (b) films which

are plasticized with LiBF4/EC-PC.

Figure 2.1.3. Schematic illustration of the formation of PEO-LiX complex.

(a)

(b)

PEO polymer chain

Li+

Li+

O

O

O

O

O

O

O

O

PEO polymer chain

Li+

Li+

Li+

Li+

O

O

O

O

O

O

O

O

Chapter 2

52

2.2. Experimental

Sample preparation and characterization

PEO (POLYOX WSR N-300, Union Carbide; Mw=400,000), PS (STYRON 656D,

Dow Chemical; Mw=30,000) and LiBF4 (Stella Chemifa, Japan) were dried under

dry airflow (dew point below -60 oC). PS was porphyrized before hot-blending.

The mixture of weighted PEO, PS and LiBF4 powders was hot-blended with blend

equipment (Small Amount Blend Machine, Imoto Machinery Co. LTD, Japan). The

blending conditions were as follows, PEO: PS = 48 : 52 by weight., O : Li = 6 : 1 by

mol, 180 oC, 30 min., 107 rpm. The blended samples were quenched in liquid N2.

Adequate amounts of the obtained blends were hot-pressed at 120 oC under 30 MPa

with sandwiching between aluminum foils to form thin films. The pressing time

was reduced to a minimum by setting the hold-under-pressure time to zero. The

thickness of the obtained films was c.a. 100 μm.

To investigate the morphology of the obtained film by hot-blending, the

PEO-LiBF4 phase in the IPN films was removed by dissolution of PEO-LiBF4 in

water. After the dissolution, the samples were rinsed with water and dried under

vacuum. The sample was sputtered by Pt-Pd for observation with Scanning

Electron Microscope (SEM). SEM images were taken with a Hitachi, S-2500CX.

The formation of PEO-LiBF4 complex was confirmed by means of Raman

spectroscopy. Raman spectra were recorded using JASCO spectrometer equipment

(RFT-800, FT/IR-800), with the condition as follows, excitation laser: He-Ne, 1064

nm, 10 mW, resolution: 4 cm-1, scanning rate: 0.5 mm/s.

Chapter 2

53

Electrochemical characterization

To make the obtained film into the gel electrolyte, a LiBF4/EC (ethylene

carbonate)-PC (propylene carbonate) (1:1 vol.) was used as a plasticizer.

The PEO-LiBF4 phase was plasticized with the solution containing the desired

concentration of Li+, which was quickly absorbed into the PEO-LiBF4 phase, by

wetting the surfaces of the IPN film. SUS / IPN-gel electrolyte / SUS cell was

assembled and the ionic conductivity of IPN gel electrolyte was evaluated by a.c.

impedance measurements (FRA 5080, NF, Japan, GPIB Potentiostat / Galvanostat

HA-501G, Hokuto Denko, Japan). The impedance spectrum was measured in the

constant potential mode by sweeping the frequencies from 20 kHz to 1 Hz range at

a.c. amplitude of 10 mV. And the temperature dependency of the conductivity of

the gel electrolyte was also examined in the temperature range of 25 to 70 oC in the

same way with Ni / IPN-gel electrolyte / Ni cell.

Li / IPN-gel electrolyte / Li cell was assembled, and the chemical stability at the

interface between the IPN gel electrolyte and lithium metal was evaluated by a.c.

impedance measurements with the conditions of a.c. amplitude voltage of 10 mV

and of the frequency range of 20 kHz - 100 mHz.

Li / IPN-gel electrolyte / Au cell was assembled to evaluate the oxidation stability

of the IPN gel electrolyte by linear sweep voltammetry (LSV) (Function Generator

HB-104, Hokuto Denko, Japan, Potentiostat / Galvanostat HA-501G, Hokuto Denko,

Japan) at the scan rate of 20 mV/s in the potential ranging from the open circuit

potential (OCP) to 6.0 V vs. Li/Li+.

The charge-discharge test was conducted to evaluate the performance of the IPN

gel electrolyte as an electrolyte for lithium secondary batteries. Li / IPN-gel

electrolyte / LiCoO2 cell was assembled with the LiCoO2 supplied from Japan

Chemical Industry. The charge-discharge test was performed (HJ1010mSM8,

Hokuto Denko, Japan) under constant current-constant voltage (CC-CV) mode at

C/5 rate of loading current and the voltage upper limit of 4.2 V for charging and CC

mode for discharging with C/5 rate of current with cut-off voltage of 3.0 V.

Chapter 2

54

2.3. Results and Discussion

2.3.1. Confirmation of phase separated structure

There are some reports that a blended polymer composed of PEO and PS

indicated the phase separated structure(28, 29). Therefore, to confirm the structure

of the polymer blend film applied PEO-LiBF4 complex, it was observed with SEM.

Because PEO and LiBF4 can be dissolved in H2O, the polymer blend film was dipped

in distilled water to confirm the matrix formations, where the PS phase remains

after PEO and LiBF4 are dissolved into H2O. The remained PS phase of the

polymer blend film was observed with SEM, shown in Fig. 2.3.1. The pore in the

image is the part occupied by PEO-LiBF4 phase. From the morphology of the PS

phase in the resulting film, the IPN structure was also demonstrated in the

PEO-LiBF4/PS polymer blend system.

Fig. 2.3.1 SEM image of PS phase of polymer blend film. The PEO-LiBF4 phase was

removed by rinsing the polymer blend with H2O

50 μm50

μm50

μm

Chapter 2

55

2.3.2. Confirmation of PEO-LiBF4 complex formation

In order to confirm the formation of PEO-LiBF4 complex expected during

hot-blending, Raman spectra of the IPN film were measured and are shown in Fig.

2.3.2. In Fig. 2.3.2 the spectra of four samples are indicated, i.e., hot-blended

PEO-LiBF4, cooled down PEO melted with the heat treatment of PEO powder,

LiBF4 powder, and mixture of PEO powder with LiBF4 powder. The spectra of the

heated and cooled PEO without LiBF4 and the mixture of PEO powder and LiBF4

powder show similar peaks except the peak at 800 cm-1 that appeared in the

spectrum of the mixture of PEO powder and LiBF4 powder. In contrast, both of the

spectra of LiBF4 powder and the mixture of PEO powder and LiBF4 powder has the

peak at 800 cm-1. From these results, it is concluded that the peak at 800 cm-1 is

due to LiBF4, and the operation of melting PEO powder without LiBF4 does not

alter the solid-state properties of PEO. On the other hand, the peak around 800

cm-1 in the spectrum of the hot-blended PEO-LiBF4 shifted to lower wave number,

777cm-1. Generally, the peak of BF4

- anion is observed at 777cm-1, whereas the

peak of LiBF4 powder was observed at 800 cm-1. This might be the effect of Li+

cation. It might be considered that the peak of BF4

- anion changed back into the

original peak by the formation the complex of PEO and Li+. This result suggests

that the complex of PEO and LiBF4 may be formed as reported in the literatures(36,

37). The formation of PEO-LiBF4 complex in the matrix is thus confirmed.

Furthermore, it is expected that the ion conducting phase would keep high viscosity

by the interaction between PEO and LiBF4, when large amount of plasticizer is

added.

Chapter 2

56

Fig. 2.3.2 Raman spectra of four samples of hot-blended PEO-LiBF4, PEO melted

once and cooled from PEO powder, LiBF4 powder, and mixture of PEO powder +

LiBF4 powder. The excitation laser: He-Ne, 1064 nm, 10 mW, resolution: 4 cm-1,

scanning rate: 0.5 mm/s.

100

400

700

1000

1300

1600

Wavenumber / cm

-1

In

ten

sity / a.u.

LiBF4 Powder

PEO Powder + LiBF4 Powder

PEO Chunk

PEO-LiBF4 Complex

600

700

800

900

1000

Wavenumber / cm

-1

Intensity / a.u.

LiBF4 Powder

PEO Powder + LiBF4 Powder

PEO Chunk

PEO-LiBF4 Complex

Chapter 2

57

2.3.3. Electrochemical properties of IPN gel electrolyte

Figure 2.3.3 shows the variation of the ionic conductivity of IPN gel electrolytes

as function of the concentration of the supporting electrolyte in the plasticizer. For

the sake of comparison, the ionic conductivity of gelled PEO-LiBF4/PS polymer

blend film and PEO/PS polymer blend film were also measured at room

temperature. Uptake of the plasticizer is calculated according to the following

formula (1).

100

][

][

][

][

[%]

4

×

+

+

=

gr

Plasticize

g

LiBF

g

PEO

gr

Plasticize

r

plasticize

the

of

Uptake

(1)

The ionic conductivity of PEO-LiBF4/PS is higher than that of PEO/PS in the range

of the investigated concentration. This is presumably due to the anion of LiBF4

added during the hot-blending process. The cation Li+ is used to form the complex,

while the anion BF4

- is comparatively free in the conductive phase in the IPN.

Besides, the ionic conductivity of PEO-LiBF4/PS has a peak at 0.5 M, and that of

PEO/PS has a peak at 0.7 M. This is presumably due to the high concentration of

the supporting electrolyte in the PEO-LiBF4/PS. LiBF4

added during the

hot-blending process would increase the concentration of LiBF4

in the

PEO-LiBF4/PS gel electrolyte. However, accurate concentrations of Li+ and BF4

-

ion were unidentified, because the percent of LiBF4 that was used for the formation

of PEO-LiBF4 complex and the degree of LiBF4 dissociation in the matrix are

unknown. Therefore, the increase of concentration of LiBF4 causes solvated ions,

ternary ions and other ion complexes to start forming at lower concentration of the

supporting electrolyte in the plasticizer. Owing to these multiple factors, the ionic

conductivity as a function of the supporting electrolyte in the plasticizer showed the

phenomena in Fig. 6. In the following investigation, 0.5 M LiBF4 / EC-PC would be

used to plasticize the PEO-LiBF4/PS films, and the 0.7 M LiBF4 / EC-PC films would

be used to plasticize PEO/PS.

Chapter 2

58

Fig. 2.3.3 Room temperature ionic conductivity of IPN gel electrolyte as a function

of the concentration of the supporting electrolyte in plasticizer. Uptake of the

plasticizer was 50 wt%. The impedance spectrum was measured in the constant

potential mode by sweeping the frequencies from 20 kHz to 1 Hz range at a.c.

amplitude of 10 mV.

0

0.2

0.4

0.6

0.8

1

0

0.4

0.8

1.2

1.6

Concentration of the supporting electrolyte / M

Ionic conductivity / m

S

cm

-

1

PEO-LiBF4 / PS

PEO / PS

Chapter 2

59

The temperature dependency of the conductivity of the of the PEO-LiBF4/PS

polymer blend gel electrolyte is shown in Fig. 2.3.4. The figure shows a straight

line, indicating that the conductivity of the PEO-LiBF4/PS polymer blend gel

electrolyte obeys Arrhenius law. This implies that the conductive environment of

Li in the polymer gel electrolyte is liquid like and remains unchanged in the

investigated temperature region. From Fig. 2.3.4, it is clear that the conductivity

increases evidently with increasing temperature and the conductivity at 25 oC was

calculated to be 1.14 mS / cm. The activation energy (Ec) of conductivity for this

film can be calculated from Arrhenius equation

│

⎠

⎞

│

⎝

⎛ -

=

RT

Ec

exp

0

σ

σ

where T is temperature on the Kelvin scale and σ0 is a proportional constant. Ec

value of this polymer film is 20.64 kJ/mol and this value is lower than the value of

PEO gel electrolyte reported in the literature of 30.8kJ/mol(38).

Chapter 2

60

Fig. 2.3.4 Arrhenius plot of ionic conductivity for gel electrolyte containing 0.5 M

LiBF4/EC-PC in the temperature range of 25 to 70 oC.

-3

-2.5

-2

2.80

2.90

3.00

3.10

3.20

3.30

3.40

1000/T / K

-1

L

og (cond./S

cm

-1

)

Chapter 2

61

Figure 2.3.5 shows the Cole-Cole plots of the Li / IPN-gel electrolyte / Li cell

investigated by a.c. impedance measurement. For the sake of comparison,

PEO-LiBF4/PS and PEO/PS gel electrolytes were applied. Assuming these results

due to the interfacial resistance derived from the charge transfer and the surface

film resistance on lithium electrode, the equivalent circuit was constructed as

shown in Fig. 2.3.6, with which the fitting were carries out on these results. The

fitted data of the Cole-Cole plots on PEO-LiBF4/PS and PEO/PS gel electrolytes are

shown in Table 2.3.1 and Table 2.3.2 respectively. From the fitted data of the

Cole-Cole plots, the semicircles observed in higher and lower frequency regions

should be ascribed due to the interfacial resistance and the surface film resistance

on lithium electrode respectively, because the value of the electric double layer

capacitance between gel electrolyte and lithium electrode was the order of μF/cm2.(39,

40)

Here the interfacial resistance was focused on because its increase was more

remarkable. The diameter of the semi-circle observed in higher frequency region is

assigned to be the sum of interfacial resistances of the two interfaces of the gel

polymer electrolyte and lithium metal in the sandwiched type cell. Figure 2.3.7 (a)

shows the change in the interfacial resistance of the gel polymer electrolyte and

lithium metal as a function of the time, and the normalized change by the

interfacial values of both resistances are shown in Fig.2.3.7 (b). The interfacial

resistance was calculated by the radius of fitting curve for Fig. 2.3.5 divided by 2

with the assumption of the interfaces of both side of gel electrolyte is almost the

same. The resistance of PEO-LiBF4/PS gel electrolyte shows lower than that of

PEO/PS gel electrolyte, while both of the normalized change of PEO-LiBF4/PS and

PEO/PS gel electrolyte are the same. This is presumably due to that a small

amount of plasticizer covered the Li surface in contact with the PEO-LiBF4/PS gel

electrolyte. Because PEO-LiBF4/PS gel electrolyte contains large amount of

plasticizer compared to the PEO/PS. The Li surface covered with plasticizer has less

chance to contact with PEO or PS polymer molecules and the solid electrolyte

interface (SEI) formed with the reaction between the plasticizer and Li might have

smaller interfacial resistance than that formed by the reaction between Li and PEO

or PS used this study. This suggests that application of PEO-LiBF4 complex to IPN

system improved the interfacial properties of the gel polymer electrolyte and

lithium metal.

Chapter 2

62

Fig. 2.3.5 Cole-Cole plot of Li / IPN-Gel / Li cell. The IPN film was plasticized with

plasticizer containing electrolyte. (a) PEO-LiBF4/PS gel electrolyte, (b) PEO/PS gel

electrolyte. The impedance spectrum was measured in the constant potential mode

by sweeping the frequencies from 20 kHz to 100 mHz range at a.c. amplitude of 10

mV.

0

1000

2000

3000

4000

0

1000

2000

3000

4000

Zreal / Ω cm

2

-Z

im

a

g

/ Ω

cm

2

1h

4h

36h

100h

385

915h

0

5000

10000

15000

20000

0

5000

10000

15000

20000

Zreal / Ω cm

2

-Z

im

a

g

/ Ω

cm

2

1h

4h

36h

100h

385h

915h

(b)

(a)

Chapter 2

63

Fig. 2.3.6 Assumed equivalent circuit for the fitting of impedance spectra. Rs:

electrolyte resistance, Ri: interfacial resistance, Rf: surface film resistance, Cdl:

electric double layer capacitance, Cf: capacitance due to surface film

Table 2.3.1 Parameters of determined from fitting the experimental impedance

spectra of PEO-LiBF4/PS gel electrolyte shown in Fig.2.3.5 (a) to the equivalent

circuit of Fig. 2.3.6.

Time [hour] Rs [Ωcm2]

R1 [Ωcm2]

C1 [μF/cm2]

R2 [Ωcm2]

C2 [μF/cm2]

0

49

125

1.8

36

8990

1

39

153

2.2

90

9000

4

34

230

2.2

126

9000

36

34

547

2.4

283

1000

100

32

727

2.4

398

1800

385

32

1625

2.8

647

2000

915

32

2415

2.8

802

2000

Table 2.3.2 Parameters of determined from fitting the experimental impedance

spectra of PEO/PS gel electrolyte shown in Fig. 2.3.5 (b) to the equivalent circuit of

Fig. 2.3.6.

Time [hour] Rs [Ωcm2]

R1 [Ωcm2] C1 [μF/cm2]

R2 [Ωcm2] C2 [μF/cm2]

0

500

931

1.6

770

30000

1

184

839

1.4

770

30200

4

186

1145

2.0

768

30600

36

163

1845

2.6

1100

30000

100

152

2729

2.6

1200

30000

385

150

7027

2.4

2003

30000

915

148

11700

3.0

3000

30000

R

s

R

i

C

dl

C

f

R

f

R

s

R

i

C

dl

C

f

R

f

Chapter 2

64

Fig. 2.3.7 Interfacial resistance change of Li / various electrolytes over time during

static storing. The interfacial resistances in this figure were calculated from

impedance data. (a) variation per hour, (b) change per hour. Uptake of the

plasticizer is 80 wt%.

0

1000

2000

3000

4000

5000

6000

1

10

100

1000

Time / hour

In

terfacial restan

ce / Ω

cm

2

PEO-LiBF4 / PS (0.5M LiBF4 / EC-PC)

PEO / PS (0.7M LiBF4 / EC-PC)

0

200

400

600

800

1000

1200

1400

1600

1800

1

10

100

1000

Time / hour

R

ate of ch

an

ge of R

i

/ %

PEO-LiBF4 / PS (0.5M LiBF4 / EC-PC)

PEO / PS (0.7M LiBF4 / EC-PC)

C

h

an

ge of R

i

/ %

0

1000

2000

3000

4000

5000

6000

1

10

100

1000

Time / hour

In

terfacial restan

ce / Ω

cm

2

PEO-LiBF4 / PS (0.5M LiBF4 / EC-PC)

PEO / PS (0.7M LiBF4 / EC-PC)

0

200

400

600

800

1000

1200

1400

1600

1800

1

10

100

1000

Time / hour

R

ate of ch

an

ge of R

i

/ %

PEO-LiBF4 / PS (0.5M LiBF4 / EC-PC)

PEO / PS (0.7M LiBF4 / EC-PC)

C

h

an

ge of R

i

/ %

(a)

(b)

Chapter 2

65

Figure 2.3.8 shows LSV curves of Li / IPN-gel electrolyte / Au cells. For the

sake of comparison, PEO-LiBF4/PS gel electrolyte (Fig.2.3.8 (a)) and PE

(polyethylene) separator with electrolytic solution (Fig.2.3.8 (b)) were also

investigated. The same solution was used as plasticizer of the polymer blend film

and electrolytic solution. From this result, both of the curves in PEO-LiBF4/PS gel

electrolyte and 0.7 M LiBF4 / EC-PC stand about 4.5 V vs. Li/Li+. The oxidation

current starting at 4.5 V vs. Li/Li+ is ascribed to the oxidation of 0.7 M LiBF4 /

EC-PC. This result suggests that PEO-LiBF4/PS can be applied to Li / LiCoO2 cell

as the electrolyte in the range between 0 and 4.2 V.

Chapter 2

66

Fig. 2.3.8 Linear sweep voltammogram of Li / electrolyte / Au cell with (a)

PEO-LiBF4 / PS gel electrolyte, (b) polyethylene separator with electrolyte solution

at the scan rate of 20 mV/s.

C

u

rren

t d

en

sity / m

A

cm

-2

Cell voltage / V vs. Li

0.1

0.2

0.3

0.4

0

1

2

3

4

5

6

0

C

u

rren

t d

en

sity / m

A

cm

-2

Cell voltage / V vs. Li

0.1

0.2

0.3

0.4

0

1

2

3

4

5

6

0

0.1

0.2

0.3

0.4

0

1

2

3

4

5

6

0

0.1

0.2

0.3

0.4

0

1

2

3

4

5

6

0

C

u

rren

t d

en

sity / m

A

cm

-2

Cell voltage / V vs. Li

0

1

2

3

4

5

6

0.1

0.2

0.3

0.4

0.5

C

u

rren

t d

en

sity / m

A

cm

-2

Cell voltage / V vs. Li

0

1

2

3

4

5

6

0.1

0.2

0.3

0.4

0.5

0

1

2

3

4

5

6

0

1

2

3

4

5

6

0.1

0.2

0.3

0.4

0.5

0.1

0.2

0.3

0.4

0.5

(b)

(a)

Chapter 2

67

Figure 2.3.9 shows the discharge capacity per gram of cathode. For the sake of

comparison, PEO-LiBF4/PS polymer blend film and PE separator with electrolytic

solution were used. The capacity of the cell using PEO-LiBF4/PS is a little lower

than that of the cell using PE separator with electrolytic solution. Besides, there

are some cycles with deathly low capacity. However, the degradation rate is

unimportant. The capacity per gram of cathode keeps 130 mAh/g over 30 cycles.

Fig. 2.3.9 Charge-discharge test of Li / (PEO-LiBF4/PS) gel electrolyte / LiCoO2 cell

and Li / (0.7 M LiBF4 / EC-PC) / LiCoO2 cell. The charge-discharge rate was C/5 in

the CC-CV charge mode and CC discharge mode, and the voltage range was 3.0 –

4.2 V.

0

20

40

60

80

100

120

140

160

180

0

5

10

15

20

25

30

35

Cycle number / -

D

isch

arge cap

acity / m

A

h

g

-1

Discharge Capacity (PEO-LiBF4 / PS)

Discharge Capacity (electrolytic solution)

Chapter 2

68

2.4.

Summary

In this chapter, the possibility of the PEO-LiBF4/PS polymer blend as electrolyte

for lithium secondary batteries. A PEO-LiBF4/PS polymer blend with three-

dimensionally IPN structure was obtained by hot-blending of the powders of PEO,

LiBF4, and PS. PEO-LiBF4 complex was confirmed to be formed in IPN prepared

by the polymer blend method.

The ionic conductivity of PEO-LiBF4/PS gel electrolyte was ~ 1 mS/cm as the

maximum value. The chemical stability of the interface between the IPN gel

electrolyte and lithium metal was successfully improved by applying PEO-LiBF4

complex to IPN. PEO-LiBF4/PS gel electrolyte demonstrated the stability against

oxidation below 4.5 V vs. Li/Li+. This result revealed that LiCoO2 cathode can be

adapted to the IPN gel electrolyte. With the charge-discharge rate of C/5, the

capacity per gram of cathode was found to be over 130 mAh/g over 30 cycles.

Consequently, these results should suggest that PEO-liBF4/PS polymer blend can be

applied to lithium secondary battery as the electrolyte.

Chapter 2

69

References

1.

F. Croce, G. B. Appetecchi, S. Slane, M. Salomon, M. Tavarez, S. Arumugam, Y.

Wang and S. G. Greenbaum, Solid State Ionics, 86-88, 307 (1996).

2.

C. Capiglia, Y. Saito, H. Kataoka, T. Kodama, E. Quartarone and P. Mustarelli,

Solid State Ionics, 131, 291 (2000).

3.

A. Manuel Stephan, S. Gopu Kumar, N. G. Renganathan and M. Anbu

Kulandainathan, European Polymer Journal, 41, 15 (2005).

4.

H.-S. Kim, P. Periasamy and S.-I. Moon, Journal of Power Sources, 141, 293 (2005).

5.

D. Saikia and A. Kumar, Electrochimica Acta, 49, 2581 (2004).

6.

G. B. Appetecchi, F. Croce and B. Scrosati, Electrochimica Acta, 40, 991 (1995).

7.

G. B. Appetecchi, G. Dautzenberg and B. Scrosati, Journal of the Electrochemical

Society, 143, 6 (1996).

8.

B. Huang, Z. Wang, L. Chen, R. Xue and F. Wang, Solid State Ionics, 91, 279 (1996).

9.

Z. Wang, B. Huang, H. Huang, L. Chen and R. Xue, Solid State Ionics, 86-88, 313

(1996).

10.

C. R. Yang, J. T. Perng, Y. Y. Wang and C. C. Wan, Journal of Power Sources, 62, 89

(1996).

11.

L. A. Dominey, in Lithium Batteries - New Materials, Developments and

Perspectives, G. Pistoia Editor, p. 137, Elsevier (1994).

12.

M. Alamgir and K. M. Abraham, in Lithium Batteries - New Materials,

Development and Perspectives, G. Pistoia Editor, p. 114, Elsevier (1994).

13.

D. G. H. Ballard, P. Cheshire, T. S. Mann and J. E. Przeworski, Macromolecules, 23,

1256 (1990).

14.

M. Morita, T. Fukumasa, M. Motoda, H. Tsutsumi and Y. Matsuda, Journal of the

Electrochemical Society, 137, 3401 (1990).

15.

H. W. Rhee, W. I. Jung, M. K. Song, S. Y. Oh and J. W. Choi, Molecular Crystals and

Liquid Crystals Science and Technology Section a-Molecular Crystals and Liquid

Crystals, 294, 225 (1997).

16.

M. C. Borghini, M. Mastragostino and A. Zanelli, Electrochimica Acta, 41, 2369

(1996).

17.

S. Chintapalli and R. Frech, Solid State Ionics, 86-88, 341 (1996).

18.

D. W. Xia, D. Soltz and J. Smid, Solid State Ionics, 14, 221 (1984).

19.

O. Bohnke, G. Frand, M. Rezrazi, C. Rousselot and C. Truche, Solid State Ionics, 66,

97 (1993).

20.

O. Bohnke, C. Rousselot, P. A. Gillet and C. Truche, Journal of the Electrochemical

Chapter 2

70

Society, 139, 1862 (1992).

21.

F. Croce, S. D. Brown, S. G. Greenbaum, S. M. Slane and M. Salomon, Chemistry of

Materials, 5, 1268 (1993).

22.

T. Iijima, Y. Toyoguchi and N. Eda, Denki Kagaku, 53, 619 (1985).

23.

P. E. Stallworth, S. G. Greenbaum, F. Croce, S. Slane and M. Salomon,

Electrochimica Acta, 40, 2137 (1995).

24.

J. Vondrak, M. Sedlarikova, J. Reiter and T. Hodal, Electrochimica Acta, 44, 3067

(1999).

25.

J. Vondrak, M. Sedlarikova, J. Velicka, B. Klapste, V. Novak and J. Reiter,

Electrochimica Acta, 46, 2047 (2001).

26.

O. Bohnke, G. Frand, M. Rezrazi, C. Rousselot and C. Truche, Solid State Ionics, 66,

105 (1993).

27.

H. Ohno, Electrochimica Acta, 46, 1407 (2001).

28.

S. Passerini, F. Alessandrini, T. Momma, H. Ohta, H. Ito and T. Osaka,

Electrochemical and Solid State Letters, 4, A124 (2001).

29.

S. Passerini, M. Lisi, T. Momma, H. Ito, T. Shimizu and T. Osaka, Journal of the

Electrochemical Society, 151, A578 (2004).

30.

M. Watanabe, Y. Suzuki and A. Nishimoto, Electrochimica Acta, 45, 1187 (2000).

31.

M. Watanabe, H. Tokuda and S. Muto, Electrochimica Acta, 46, 1487 (2001).

32.

A. Noda, T. Kaneko and M. Watanabe, POLYMER PREPRINTS, JAPAN, 49, 754

(2000).

33.

A. Noda, J. Nishimoto and M. Watanabe, POLYMER PREPRINTS, JAPAN, 46, 511

(1997).

34.

F. Croce, L. Persi, F. Ronci and B. Scrosati, Solid State Ionics, 135, 47 (2000).

35.

P. P. Prosini and S. Passerini, Solid State Ionics, 146, 65 (2002).

36.

J. Adebahr, P. Gavelin, P. Jannasch, D. Ostrovskii, B. Wesslen and P. Jacobsson,

Solid State Ionics, 135, 149 (2000).

37.

C. Capiglia, N. Imanishi, Y. Takeda, W. A. Henderson and S. Passerini, Journal of

the Electrochemical Society, 150, A525 (2003).

38.

T. C. Wen and W. C. Chen, Journal of Applied Polymer Science, 77, 680 (2000).

39.

T.-C. Wen and W.-C. Chen, Journal of Power Sources, 92, 139 (2001).

40.

W. Xu, K. S. Siow, Z. Gao and S. Y. Lee, Solid State Ionics, 112, 1 (1998).

Chapter 3:

PEO-PS diblock polymer gel electrolyte

Chapter 3

73

3.1. Introduction

In chapter 2, the improvement of the retention of ionic conductive phase on

the PEO/PS polymer blend gel electrolyte system(1-3) by the application of PEO-LiX

complex was discussed. And the potential of the application of PEO-LiX complex

to the PEO/PS polymer blend gel electrolyte system was demonstrated.(4)

In this chapter, to improve the retention of ionic conductive phase on the

polymer gel electrolyte, another approach which is different from the approach

discussed in chapter 2 is investigated. A di-block copolymer, composed of PEO

chain and PS chain covalently bonded, was applied to polymer gel electrolyte for the

restraint of the ionic conductive phase running out from the frame phase in the IPN,

and also to achieve both properties of high ionic conductivity at RT and excellent

mechanical strength. There are some reports on solid electrolyte of block

copolymer(5-8), however a study on gel electrolyte of block copolymer has not been

reported. It is expected that the gel electrolyte formed by self-assembling of the

di-block copolymer has micro phase separation structure(9), that is IPN structure.

The characteristic and possibility of the prepared PEO-PS di-block copolymer gel

electrolyte were examined for the future lithium secondary batteries.

Chapter 3

74

3.2. Experimental

Sample preparation and characterization

PEO-PS di-block copolymer (P1614-SEO, Sowa Science Co., Mn (PEO /PS) =

29,000 / 51,000) was dried under dry airflow (dew point below -60 oC). LiBF4 /

ethylene carbonate (EC) - propylene carbonate (PC) (1:1 vol.) and ethylhexylacetate

were used as plasticizer for PEO and PS respectively. The plasticizer for PEO was

prepared by dissolving LiBF4 (Stella Chemifa corporation, Japan) into EC-PC

(Kishida Chemical Co., Ltd.) under dry air condition. As the plasticizer for PS(10),

ethylhexylacetate (Kanto Chemical Co., Inc.) was used as received. The PEO-PS

di-block copolymer, LiBF4 / EC-PC, and ethylhexylacetate were dissolved into

toluene. The mixture solution was cast into a glass fiber separator in Teflon cup.

The solvent was allowed to evaporate at room temperature, covered with a Petri

dish to prevent rapid evaporation of toluene under dry air atmosphere overnight.

Then, the Petri dish was removed for further toluene evaporation. The thickness

of the IPN film was ~ 700 µm, due to the quantity of casted mixture solution and the

thickness of the glass fiber separator. The prepared gel electrolyte was dipped into

the plasticizer, whose Li+ concentration was the same as that used its preparation,

and wiped the residual plasticizer before electrochemical measurements. In this

study, a glass fiber separator was applied just as supporting film so as not to crack

easily when treated coarsely; the glass fiber separator should not affect any

electrochemical properties of polymer gel electrolyte. The polymer gel electrolyte

was enough strong mechanically to keep its shape without the glass fiber separator.

To confirm the phase separated structure of the obtained film, the surface of film

was observed with Atomic Force Microscope (AFM, Veeco, Multimode Dimension

3100). The film for AFM observation was prepared on a Cu substrate by

spin-coating, because the film prepared by casting in Teflon cup was too rough to

observe with AFM. The morphology difference of a prepared IPN film between the

upper side exposed to dry air and the bottom side contacted with Teflon sheet

during evaporating process were investigated with Scanning Electron Microscope

(SEM, Hitachi, S-2500CX).

Chapter 3

75

Electrochemical characterization

The ionic conductivity of the polymer gel electrolyte was measured by a.c.

impedance measurement (FRA 5080, NF, Japan, Automatic Polarization System,

HZ-3000, Hokuto Denko, Japan), using Ni / polymer gel electrolyte / Ni cell.

The impedance spectrum was measured in the constant voltage mode at open

circuit potential (OCP) by sweeping the frequencies range from 10 kHz to 100 mHz

with a.c. amplitude of 10 mV. And the temperature dependency of the conductivity

of the gel electrolyte was also examined in the temperature range of 25 to 100 oC in

the same way.

Li half cell with the polymer gel electrolyte was assembled to investigate the

effect of suppressing dendritic deposition, and the chemical stability of polymer gel

electrolyte against Li metal at its static storing. To investigate the effect of

suppressing dendritic deposition of Li, lithium deposition and dissolution were

cycled 21 times at current density of 2.5 mA / cm2 (1st cycle for pretreatment) and

0.5 mA / cm2 (2nd -21st cycle), Li deposition and dissolution time of 2 hours with

Battery Charge / Discharge System (HJ1010mSM8, Hokuto Denko, Japan), then,

the Li / polymer gel electrolyte interfaces were observed.

The chemical stability at the interface between the polymer gel electrolyte and

Li metal was evaluated in light of the interfacial resistance shift at static storing

with time measured by a.c. impedance measurement with the conditions of a.c.

amplitude voltage of 10 mV and of the frequency range of 10 kHz - 100 mHz.

Li / polymer gel electrolyte / Ni cell was assembled to evaluate the oxidation

stability of the polymer gel electrolyte by normal pulse voltammetry (NPV) with

Automatic Polarization System at the scan rate of 50 mV / s in the potential rang

from OCP to the positive potential with the pulse period and width are 0.5 and 0.1

sec respectively.

The charge-discharge test was conducted to evaluate the performance of the

polymer gel electrolyte as an electrolyte for lithium secondary batteries. Li /

polymer gel electrolyte / LiFePO4 cell was assembled. LiFePO4 cathode was

composed of LiFePO4 (battery grade, Aldrich), acetylene black (AB), and PVdF-HFP

gel. The charge-discharge test was performed with Battery Charge / Discharge

System under CC-CV (constant current-constant voltage) mode at C/5 rate of

loading current and the voltage upper limit of 3.6 V with 2 hours for voltage holding

for charging and CC mode for discharging with C/5 rate of current with cut-off

voltage of 2.0 V.

Chapter 3

76

3.3. Results and Discussion

3.3.1. Confirmation of phase separated structure

For confirmation of the phase separated structure of PEO-PS di-block copolymer

film, a spin-coated PEO-PS di-block copolymer film with LiBF4 / EC-PC and

ethylhexylacetate, which are plasticizer for PEO and PS respectively, on Cu

substrate was observed with AFM. Figure 3.3.1 shows AFM image of the film,

bright and dark region are PS and PEO phase respectively; it was confirmed from

expanding bright region of the film prepared by the addition of a PS polymer to the

casting solution. It suggested that double gyroid or double diamond structure, that

is interpenetrated polymer network (IPN) structure, was formed in the film

prepared from PEO-PS di-block copolymer, LiBF4 / EC-PC, and ethylhexylacetate.

Fig. 3.3.1 AFM image of spin-coated PEO-PS di-block copolymer film (PEO content

of 36 wt %), with LiBF4 / EC-PC (1:1 vol.) and ethylhexylacetate as plasticizer for

PEO and PS respectively, on Cu substrate.

250nm

0.0

nm

100.0

nm

0.0

nm

100.0

nm

250nm

0.0

nm

100.0

nm

0.0

nm

100.0

nm

Chapter 3

77

3.3.2. Electrochemical properties of IPN gel electrolyte

The ionic conductivity of IPN gel electrolyte at room temperature was evaluated.

Figure 3.3.2 shows the variation of the ionic conductivity of interpenetrated

polymer network gel electrolytes as function of the Li+ concentration in LiBF4 /

EC-PC used as the plasticizer for PEO. The ionic conductivity of IPN gel

electrolyte had a peak when 0.84 M LiBF4 / EC-PC was used as the plasticizer; the

maximum ionic conductivity was 2.65 mS/cm. The phenomenon of indicating

maximum as a function of the Li+ concentration of the plasticizer for PEO on the

IPN gel electrolyte is presumably similar to that of liquid electrolyte(11, 12). The

phenomenon in this PEO-PS interpenetrated polymer network gel electrolyte would

be explained by the deficiency of carrier ion, Li+, at lower Li+ concentration in the

plasticizer for PEO and due to the limited mobility of Li+ by forming ternary ions

and other ion complexes at higher Li+ concentration in that. Therefore the ionic

conductivity of IPN gel electrolyte had a peak at a Li+ concentration of the

plasticizer. In this study, the IPN gel electrolyte added 1.12 M LiBF4 / EC-PC was

used in the following electrochemical measurements.

Chapter 3

78

Fig. 3.3.2 Ionic conductivity of interpenetrated polymer network gel electrolytes as a

function of Li+ concentration in LiBF4 / EC-PC used as plasticizer at room

temperature. The impedance spectrum was measured in the constant voltage mode

at open circuit potential (OCP) by sweeping the frequencies range from 10 kHz to

100 mHz with a.c. amplitude of 10 mV.

0.1

1

10

0

1

2

3

Li

+

concentration of LiBF4/EC-PC as plasticizer / M

Ionic conductivity / m

S

cm

-1

Chapter 3

79

The temperature dependency of the conductivity of the of the PEO-PS di-block

copolymer gel electrolyte is shown in Fig. 3.3.3. The figure shows a straight line,

indicating that the conductivity of the PEO-PS di-block copolymer gel electrolyte

obeys Arrhenius law. This implies that the conductive environment of Li in the

polymer gel electrolyte is liquid like and remains unchanged in the investigated

temperature region. From Fig. 3.3.3, it is clear that the conductivity increases

evidently with increasing temperature and the conductivity at 25 oC was calculated

to be 0.54 mS / cm. The activation energy (Ec) of conductivity for this film can be

calculated from Arrhenius equation

│

⎠

⎞

│

⎝

⎛ -

=

RT

Ec

exp

0

σ

σ

where T is temperature on the Kelvin scale and σ0 is a proportional constant. Ec

value of this polymer film is 33.57 kJ/mol and this value is almost same as PEO gel

electrolyte reported in the literature(13).

Chapter 3

80

Fig. 3.3.3 Arrhenius plot of ionic conductivity for gel electrolyte containing 1.12 M

LiBF4/EC-PC in the temperature range of 25 to 100 oC.

-3.5

-3

-2.5

-2

-1.5

2.60

2.80

3.00

3.20

3.40

1000/T / K

-1

L

og (cond./S

cm

-1

)

Chapter 3

81

The chemical stability of electrolytes is important factor to assemble the cell

with Li anode. To investigate the chemical stability of interpenetrated polymer

network gel electrolyte against Li metal, Li / IPN gel electrolyte / Li cell was

assembled and a.c. impedance measurement was carried out. Figure 3.3.4 shows

the Cole-Cole plots of the Li / IPN gel electrolyte / Li cell measured with the

symmetric cell. Assuming these results due to the interfacial resistance derived

from the charge transfer and the surface film resistance on lithium electrode, the

equivalent circuit was constructed as described in chapter 2, with which the fitting

were carries out on these results. The fitted data of the Cole-Cole plots on PEO-PS

di-block copolymer gel electrolyte and 1.12 M LiBF4/EC-PC as a reference are shown

in Table 3.3.1 and Table 3.3.2 respectively. From the plots fitting and analysis of

the Cole-Cole plots, the diameter of the semi-circle observed in higher frequency

region is assigned to be the sum of interfacial resistances of the two interfaces of gel

polymer electrolyte and Li metal in the sandwiched type cell as described in chapter

2. The interfacial resistance was calculated from the radius of fitting curve for Fig.

3.3.4 divided by 2 with the assumption of the interfaces of both side of gel electrolyte

is the same in this symmetrical cell. Figure 3.3.5 shows the change in the

interfacial resistance between PEO-PS di-block copolymer gel electrolyte and Li

metal analyzed from the Cole-Cole plots in Fig. 3.3.4 as a function of the time. For

comparison, those using PEO/PS polymer blend film(3) and liquid electrolyte, LiBF4 /

EC-PC, in a separator were also analyzed. The interfacial resistance of PEO-PS

di-block copolymer gel electrolyte was lower than that of PEO/PS polymer blend gel

electrolyte; both of those of PEO-PS di-block copolymer gel electrolyte and liquid

electrolyte, LiBF4 / EC-PC, were almost the same. The difference between the

interfacial resistances of PEO/PS polymer blend gel electrolyte and liquid

electrolyte of LiBF4 / EC-PC is existence or nonexistence of PEO, and the difference

between those of PEO-PS di-block copolymer and PEO/PS polymer blend gel

electrolyte is whether PEO phase is fixed or not to PS matrix by covalent bonding.

Therefore the difference of chemical stability between PEO-PS di-block copolymer

and PEO/PS polymer blend gel electrolyte should be derived from the diffusivity of

PEO; in the case of PEO/PS polymer blend gel electrolyte, as the result of the

reaction of solved PEO and Li, the interfacial resistance should be increased. On

the other hand, in the case of PEO-PS di-block copolymer gel electrolyte, the

increase of the interfacial resistance should be able to be suppressed by fixing PEO

phase to PS matrix absolutely. Consequently, the PEO-PS di-block copolymer gel

electrolyte exhibited excellent chemical stability against Li.

Chapter 3

82

Fig. 3.3.4 Change in impedance spectra of a half cell during static storage. The half

cell of Li / interpenetrated polymer network gel electrolyte (with 1.12 M

LiBF4/EC-PC used as plasticizer). Inset is enlarged view of the high frequency

domain of the impedance spectra. The data were obtained under the following

conditions: frequency range of 10 k - 0.1 Hz, amplitude voltage of 10 mV, and room

temperature.

0

50

100

150

0

50

100

150

Zreal

/ Ωcm

2

-Z

im

ag

/ Ω

cm

2

1 hr

2 hr

4 hr

8 hr

12 hr

0

10

20

0

10

20

30

40

50

Chapter 3

83

Table 3.3.1 Parameters of determined from fitting the experimental impedance

spectra of PEO-PS di-block copolymer gel electrolyte.

Time [hour]

Rs [Ωcm2]

R1 [Ωcm2] C1 [μF/cm2] R2 [Ωcm2] C2 [μF/cm2]

0

30

45

2.4

201

1224

1

35

76

4.2

301

1286

2

40

113

5.0

346

1286

4

40

151

5.2

403

1428

6

40

179

5.4

422

1560

8

40

221

5.2

446

1628

12

40

257

5.4

525

1768

23

40

284

5.4

553

2068

59

40

505

5.2

822

2262

466

40

1264

5.0

1647

2294

Table 3.3.2 Parameters of determined from fitting the experimental impedance

spectra of 1.12 M LiBmeF4/EC-PC electrolyte as reference.

Time [hour]

Rs [Ωcm2]

R1 [Ωcm2] C1 [μF/cm2] R2 [Ωcm2] C2 [μF/cm2]

0

9

29

4.0

176

1768

1

11

58

6.4

295

1900

2

11

82

6.4

378

1894

4

11

128

6.8

481

2016

6

11

160

7.4

525

2252

8

11

193

7.4

566

2356

12

11

241

7.4

660

2356

23

11

332

7.0

662

2722

59

11

589

7.0

929

3276

466

12

1419

7.4

1915

3400

Chapter 3

84

Fig. 3.3.5 Interfacial resistance change of Li / various electrolytes over time during

static storing. The interfacial resistances in this figure were calculated from

impedance data.

0

500

1000

1500

2000

2500

3000

1

10

100

1000

Time / hour

Interfacial resistance / Ω

cm

2

diblock polymer gel

polymer blend gel

liquid electrolyte

Chapter 3

85

Electrolytes are also required to suppress the growth of dendritic deposition of Li

for lithium secondary batteries. To confirm the suppression effect of the PEO-PS

di-block copolymer gel electrolyte against dendritic deposition Li, Li / IPN gel

electrolyte half cell was assembled and dissolving-deposition reactions were cycled

with a constant current. Figure 3.3.6 shows the voltage profile during lithium

dissolving-deposition reactions cycling. At 1st cycle as pretreatment, the current

of 2.5 mA/cm2 was flown to remove impurity on Li electrode surface by anode

stripping. The larger overvoltage was observed at both of first dissolving and

deposition process. These should be derived from removing impurity on Li

surfaces. In 2nd-21st cycles, the current of 0.5 mA/cm2 was flown to form the

dendritic lithium. Stable overvoltage was observed during 2nd-21st cycles,

indicating that dissolving-deposition reactions occurred stably. After 21 cycles of

dissolving-deposition reactions, the Li / IPN gel electrolyte half cell was

disassembled to observe the surfaces of IPN gel electrolyte contacted with Li. The

dendritic deposition of Li was not observed on one surface as shown in Fig. 3.3.7 (a),

however that was observed on the other surface as shown in Fig. 3.3.7 (b). This

difference can be explained with the SEM images indicating the dense side of IPN

gel electrolyte prepared on Teflon sheet (Fig. 3.3.8 (a)) and the porous side of that

prepared facing air (Fig. 3.3.8 (b)). Though the dendritic deposition of Li grew into

the voids of PEO-PS copolymer at its porous side, the dendritic deposition of Li was

not able to grow by IPN gel electrolyte at its dense side. Consequently, it was

suggested that PEO-PS di-block copolymer gel electrolyte was able to suppress the

dendritic deposition of Li at charge-discharge cycles.

Chapter 3

86

Fig. 3.3.6 Voltage profile during lithium dissolving-deposition reactions cycling with

a half cell of Li / interpenetrated polymer network gel electrolyte (1.12 M

LiBF4/EC-PC was used as plasticizer). The conditions of the dissolving-deposition

reactions were as follows: constant current flow of 2.5 mA / cm2 at the 1st cycle and

0.5 mA / cm2 during the 2nd - 21st cycles.

-400

-200

0

200

400

0

20

40

60

80

Time / hour

O

vervoltage / m

V

Chapter 3

87

Fig. 3.3.7 Micrographs of interpenetrated polymer network gel electrolyte (1.12 M

LiBF4/EC-PC was used as plasticizer) after 21 cycles with a constant current flow of

2.5 mA / cm2 at the 1st cycle and 0.5mA / cm2 at the 2nd – 21st cycles. (a) Dense side

of the gel electrolyte prepared in a Teflon cup, (b) Porous side of the gel electrolyte

prepared facing air.

Fig. 3.3.8 SEM images of interpenetrated polymer network gel electrolyte (1.12 M

LiBF4/EC-PC was used as plasticizer) before dissolving-deposition cycles of Li. (a)

Dense side of gel electrolyte prepared in a Teflon cup, (b) Porous side of gel

electrolyte prepared facing air.

80μm

80μm

(a)

(b)

(a)

(b)

5 mm

5 mm

Chapter 3

88

The electrochemical stability of electrolytes against oxidation is important to

achieve high power batteries. To investigate the electrochemical stability against

oxidation, Li / IPN gel electrolyte / Ni cell was assembled and NPV was carried out,

shown in Fig. 3.3.9. In Fig. 3.3.9, the oxidation current started to flow from 3.0 V

vs. Li/Li+, and then its increasing rate changed around ca. 3.9 V vs. Li/Li+. From

this result, the IPN gel electrolyte should be indicated to stand about 3.9 V vs. Li/Li+

by extrapolating the NPV curve measured with IPN gel electrolyte. The oxidation

current observed 2.0-3.9 V vs. Li/Li+ was presumably due to a decomposition of

impurity. This should be able to explain from the charge-discharge efficiency of the

full cell using IPN gel electrolyte, shown in Fig. 3.3.10. Figure 3.3.10 indicates the

discharge capacity per gram of cathode and charge-discharge efficiency. The

LiFePO4 cathode was selected on the bases of the assumption that the IPN gel

electrolyte is stable under the potential of 3.9 V. The charge-discharge efficiency

except for 1st cycle was ca. 99 %, though that of 1st cycle was 88 %. This result

implies that impurity was almost decomposed at 1st cycle. Therefore it was

suggested that IPN gel electrolyte can be applied to Li / LiFePO4 cell as the

electrolyte in the range between 0 and 3.6 V.

Chapter 3

89

Fig. 3.3.9 Normal pulse voltammogram of Li / interpenetrated polymer network gel

electrolyte (1.12 M LiBF4 /EC-PC was used as plasticizer) / Ni cell. The electrode

potential was scanned from open circuit potential to positive potential at a scan rate

of 50 mV / s, and the pulse period and width were 0.5 and 0.1 sec, respectively.

0

0.5

1

1.5

3

3.5

4

4.5

5

5.5

Voltage / V vs. Li/Li

+

C

urrnt dencity / m

A

cm

-2

Chapter 3

90

The discharge capacity of the cell using IPN gel electrolyte kept 124 mAh/g with

~ 99 % of coulombic efficiency over 30 cycles. This discharge capacity was lower

than theoretical capacity of LiFePO4, due to IR drop derived from the electrolyte

resistance. However, its improvement should be achieved by thinning the

electrolyte film.

Fig. 3.3.10 Discharge capacity and coulomb cycle efficiency of Li / interpenetrated

polymer network gel electrolyte (1.12 M LiBF4/EC-PC was used as plasticizer) /

LiFePO4 cell. The charge-discharge rate was C/5 (195 μA / cm2) in the CC-CV charge

mode and CC discharge mode, and the voltage range was 2.0 - 3.6 V

.

80

90

100

110

120

130

140

0

10

20

30

Cycle number / -

D

ischarge capacity / m

A

h g

-

1

50

60

70

80

90

100

110

C

ycle efficiency / %

Chapter 3

91

Summary

In this chapter, the possible application of a PEO-PS di-block copolymer gel

electrolyte for lithium secondary batteries was investigated. Since the PEO-PS di-block

copolymer gel electrolyte was too stiff to handle, a plasticizer for PS, ethylhexylacetate,

was added to the electrolyte for the casting process. Indications were that the structure

of the electrolyte was a co-continuous interpenetrated polymer network, consisting of

PS domains with a columnar or spherical shape. The IPN gel electrolyte had a high

ionic conductivity (2.65 mS / cm at maximum value) and also good mechanical strength.

It was confirmed visually that the dendritic deposition at the interface between the Li

electrode and the gel electrolyte was suppressed as a result of the electrolyte’s good

mechanical strength. The electrolyte also showed excellent chemical stability when it

contacted the Li electrode, which probably is the result of the covalent bonding between

the PEO chain and the PS chain. Furthermore, the IPN gel electrolyte showed oxidation

stability below 3.9 V vs. Li / Li+. The electrolyte, moreover, was applied to a Li / LiFePO4

cell. With a charge-discharge rate of C/5, the capacity per gram of cathode was found to

be over 120 mAh / g over 30 cycles. This suggests that the PEO-PS di-block copolymer

gel electrolyte can be successfully applied as the electrolyte in lithium secondary

batteries.

Chapter 3

92

Reference

1.

F. Croce, L. Persi, F. Ronci and B. Scrosati, Solid State Ionics, 135, 47 (2000).

2.

S. Passerini, F. Alessandrini, T. Momma, H. Ohta, H. Ito and T. Osaka,

Electrochemical and Solid State Letters, 4, A124 (2001).

3.

S. Passerini, M. Lisi, T. Momma, H. Ito, T. Shimizu and T. Osaka, Journal of the

Electrochemical Society, 151, A578 (2004).

4.

T. Momma, H. Ito, H. Nara, H. Mukaibo, S. Passerini and T. Osaka,

Electrochemistry, 71, 1182 (2003).

5.

G. Liu, M. T. Reinhout and G. L. Baker, Solid State Ionics, 175, 721 (2004).

6.

T. Niitani, M. Shimada, K. Kawamura, K. Dokko, Y. H. Rho and K. Kanamura,

Electrochemical and Solid State Letters, 8, A385 (2005).

7.

T. Niitani, M. Shimada, K. Kawamura and K. Kanamura, Journal of Power Sources,

146, 386 (2005).

8.

D. R. Sadoway, B. Y. Huang, P. E. Trapa, P. P. Soo, P. Bannerjee and A. M. Mayes,

Journal of Power Sources, 97-8, 621 (2001).

9.

C. Park, J. Yoon and E. L. Thomas, Polymer, 44, 6725 (2003).

10.

T. A. Martin and D. M. Young, Polymer, 44, 4747 (2003).

11.

Y. Matsuda, M. Morita and T. Yamashita, Journal of the Electrochemical Society,

131, 2821 (1984).

12.

M. Ue and S. Mori, Journal of the Electrochemical Society, 142, 2577 (1995).

13.

T. C. Wen and W. C. Chen, Journal of Applied Polymer Science, 77, 680 (2000).

Chapter 4:

Mesoporous Sn Anode electrodeposited

with lyotropic liquid crystals

Chapter 4

95

4.1. Introduction

There is a strong incentive to develop and characterize non-carbonaceous

materials for use as anodes for lithium ion secondary batteries that deliver higher

capacities than carbon. A Sn anode has been reported to have higher theoretical

capacity (994 mAh/g) than that of carbon (372 mAh/g).(1-3)

However, the Sn anode

has a problem that its cycle life is unsatisfactory because of Sn disintegration due to

the volume change during the charge and discharge cycling of Li-ions. To solve

this problem, Sn-based alloys with inactive elements against lithium such as Ni(4-7),

Fe(8-12), Cu(13-16), Mn(9, 17), and Co(9) have been investigated. Recently, an amorphous

ternary Sn-Co-C anode has been introduced to a practical use.(18)

On the other hand, the study of the synthesis, properties, and possible uses of

mesoporous materials has been devoted, since the discovery of FSM-16(19) and

MCM-41(20), both silica and aluminumsilicate. Already in 1993 it was suggested, on

the basis of mechanistic ideas, that it should be possible to synthesize non-siliceous

materials.(21)

The first examples have been reported already in 1994.(22)

However, for

these materials it had not been possible to remove the template and thus no mesoporous

materials, but only mesostructured materials, could be obtained. The first mesoporous

nonsiliceous frameworks were reported in 1995-1996,(23, 24)

from which rapid

development started. Since the first report by Attard et al.(25), several mesoporous

metals including Sn(26-28) have been prepared by the reduction of the corresponding

metal ions in the presence of lyotropic liquid crystals (LLC)(29, 30) made of nonionic

surfactants(29, 31) which is described in Section 1.4.2. The mesoporous metals have been

mainly prepared by electrodeposition for the reduction of metal ions to form thin films

on conductive substrates. Thus, many mesoporous materials with specific structural

features (e.g., uniform mesopore size and high surface area) have extensively been

investigated. In particular, mesoporous metals with high electroconductivity are very

promising for various electrochemical applications.

In this chapter, the cycle and rate properties of mesoporous Sn anode

prepared from a LLC bath were investigated for lithium secondary batteries. The

mesoporous structure should suppress the influence of the volume change during

the charge and discharge cycling of Li-ions, leading to a longer life. Also, the

mesoporous structure with high surface area can provide a low diffusion resistance

of Li-ions into the electrode, which should contribute to an increase of the reaction

rate.

Chapter 4

96

4.2. Experiments

Sample preparation and characterization

A mesoporous Sn was prepared on a copper foil by electrodeposition with a

LLC bath. The LLC including Sn ions was prepared by mixing 0.300 g nonionic

surfactant, octaethyleneglycol monohexadecyl ether (C16EO8), and an aqueous Sn

solution (0.300 g) which was prepared by dissolving both 1.69 g SnCl2･2H2O and

9.15 g of H2SO4 aq. (18.3 mol/L) into water and fixing the volume to 50 mL. The LLC

takes a 2D-hexagonal symmetry, as is confirmed by low angle XRD analysis (Figure

4.2.1). The electrodeposition was carried out at room temperature and H2O

saturated atmosphere with a constant voltage of -100 mV till the current of 1.5

C/cm2 was passed. Sn plate was used as the counter electrode.

And a Sn anode, called a “dense Sn” in this thesis, was electrodeposited on a

copper foil without nonionic surfactant to evaluate its electrochemical properties as

control. The bath used for this preparation is shown in Table 4.2.1(32). The

electrodeposition was conducted under room temperature with current density of 5

mA/cm2 for 166 sec, constant agitation by magnetic stirrer. Pt wire was used as

the counter electrode and an Ag/AgCl electrode was used as reference electrode.

The thickness of the mesoporous and dense Sn anode were 1.2 and 1.0 μm,

respectively, which were confirmed from cross-sectional scanning electron

microscope (SEM, HITACHI S2500CX) image.

The crystalline structure analysis of both Sn anodes was conducted with a

X-ray diffractometer (XRD). The XRD apparatus (Rigaku, Rint-TTR3) with Cu Kα

radiation was equipped with the voltage and current of 50 kV and 300 mA,

respectively. The nano-structure and the electron diffraction (ED) pattern of both

samples were evaluated using a transmission electron microscope (TEM: JEOL,

JEM-2010), with the powdery samples which were scratched from the substrate and

dispersed in ethanol and mounted on a TEM microgrid.

Chapter 4

97

Figure 4.2.1 Small-angle X-ray diffraction patterns of lyotropic liquid crystal bath

containing a Sn solution.

Table 4.2.1 Bath composition of a “dense Sn” anode.(32)

Chemical

Concentration

SnSO4

40 g/L

H2SO4

60 g/L

ο-Cresolsulfonic acid (CH3C6H3(OH)SO3H

40 g/L

Polyethylene glycol (Polymerization degree: 20,000)

((CH2CH2O-)n)

100 ppm

1

2

3

4

5

6000

5000

4000

3000

2000

1000

0

2θ / degree

In

te

n

s

ity

/ a

.u

.

3.6 3.8 4.0 4.2 4.4

200

150

100

50

0

2θ / degree

In

te

n

s

ity

/ a

.u

.

Chapter 4

98

Electrochemical characterization

Cycle properties were evaluated using conventional glass cells with two

pieces of lithium foil as counter and reference electrodes, and 1 M LiClO4 / ethylene

carbonate (EC) + propylene carbonate (PC) (1:1 vol. %) (as purchased from Kishida

Chemical Co.,Ltd.) as the organic electrolyte. All the cells were sealed under Ar

atmosphere. For evaluation of the cycling, both charge and discharge of Li-ions were

carried out by constant current (CC) mode at current density of 0.994-16.9 A/g

(1-17C rate)in the potential range of 0.01 to 1 V vs. Li/Li+, with Battery Charge /

Discharge System (HJ1010mSM8, Hokuto Denko, Japan).

Cyclic voltammetry was carried out with Automatic Polarization System,

HZ-3000, Hokuto Denko, Japan), in the potential range of open-circuit potential

(OCP) → 0.01 to 1 V vs. Li/Li+, at scan rate of 0.2 mV/sec.

Chapter 4

99

4.3. Results and Discussion

4.3.1. Confirmation of mesoporous structure

Morphological and structural differences of Sn anodes electrodeposited with

or without nonionic surfactant were investigated with TEM. Figure 4.3.1 indicates

the bright-field TEM images and selected ED patterns in a 100 nm region of

as-prepared mesoporous Sn and dense Sn. Figure 4.3.1 (a) shows the TEM image

of the as-prepared sample with nonionic surfactant and its selected-electron

diffraction (ED) pattern in a 100 nm region. Though the nonionic surfactant

template was ordered with 2D-hexagonal symmetry of a d spacing of 5.8 and 2.8 nm,

confirmed by low angle XRD analysis (Fig. 4.2.1), a disordered mesoporous

structure was observable over the entire area (Fig. 4.3.1 (a)). This reason should

be contributed by the dendritic Sn deposition; this dendritic deposition of Sn should

dislodge the nonionic surfactant template. The bright and dark areas indicate

pores and deposited Sn, respectively. The ED patterns are assignable to (111)

diffraction of the α-Sn structure and (200), (101), (220), and (211) diffractions of the

β-Sn structure. The ring pattern indicates that the deposited Sn is composed of

very minute polycrystals. On the other hand, Figure 4.3.1 (b) shows the TEM

image of the as-prepared sample without nonionic surfactant and its selected-ED

pattern in a 100 nm region. A mesoporous structure was not observed on the

as-prepared sample without nonionic surfactant. The ED patterns are assignable to

(111), (220), and (311) diffraction of the α-Sn structure.

Thus, the introduction of mesoporous structure to Sn anode with nonionic

surfactant, and the crystallinity of both samples were confirmed.

Chapter 4

100

Figure 4.3.1 Bright-field TEM images and selected ED patterns in a 100 nm region.

(a) As-prepared mesoporous Sn. (b) As-prepared dense Sn. The sample were

scratched from the substrate. The powdery samples were dispersed in ethanol and

mounted on a TEM microgrid.

20 nm

(a)

β-Sn (220)

β-Sn (101)

α-Sn (111)

β-Sn (211)

β-Sn (200)

20 nm

(b)

α-Sn (311)

α-Sn (220)

α-Sn (111)

Chapter 4

101

4.3.2. Electrochemical properties of mesoporous Sn anode

To evaluate the influence of the introduction of mesoporous structure to Sn

anode, the charge-discharge tests were carried out. Figure 4.3.3 and Figure 4.3.4

show charge-discharge curves of the mesoporous Sn anode, and dense Sn anode is

also shown for comparison, at initial stage (1-ca.5th cycle) and second stage or later

(ca. 5-100th cycle), respectively.

In Fig. 4.3.3, the plateaus at 0.58, 0.72, and 0.78 V vs. Li/Li+ on discharge

curves; assigned to the reactions of Li7/3Sn → LiSn, LiSn → Li0.4Sn, and Li0.4Sn →

Sn, respectively(33), were observed. The potentials of these plateaus of mesoporous

Sn anode at discharging were ca. 20 mV lower than that of dense Sn. This

difference should be contributed by the reduction of iR drop due to the higher

surface area of mesoporous Sn anode compared with dense anode. This discussion is

also supported by the cyclic voltammogram of both anodes, shown in Fig. 4.3.5. The

current density of the mesoporous Sn anode was ca. 3 times as high as that of the

dense Sn anode; this indicates that the active site of the mesoporous Sn with Li was

larger than that of the dense Sn.

Figure 4.3.6 shows the discharge capacity of the mesoporous Sn anode and

dense Sn anode. In Fig. 4.3.6, the discharge capacities of both the mesoporous and

dense Sn anodes, increased at the initial stage (0-5th cycle). This reason is

described later. At the second stage (ca. 5-20th cycle), the discharge capacity of the

dense Sn anode was dramatically decreased due to electrical disconnection. The

disintegration of the dense anode from the Cu foil was visually observed. On the

other hand, in the case of the mesoporous Sn anode, such a phenomenon was not

confirmed. The mesoporous structure suppresses the influence of the volume

change during the cycling to some extent, which can prevent the disintegration of

the Sn anode from the substrate. This reason is also described later. At the

following stage (over 20th cycle), the discharge capacities of both Sn anodes were

maintained with small changes. After the 100th cycle, the discharge capacity of

the mesoporous anode was 425 mAh/g, while that of the dense anode was only 46

mAh/g. For comparison, when nanosized carbon particles were cycled at a

constant current of 100 mA/g, the discharge capacity after 100th cycle was 290

mAh/g.(34)

Consequently, the introduction of the mesoporous structure is effective

to improve the cycle durability.

Chapter 4

102

Figure 4.3.3 The charge-discharge curves of (a) mesoporous Sn, (b) dense Sn at

initial stage. The charge-discharge current density was 0.994 A/g (1 C rate) in the

potential range of 0.01 to 1 V vs. Li/Li+.

0.0

0.2

0.4

0.6

0.8

1.0

1.2

-1000 -800

-600

-400

-200

0

200

400

600

800

Capacity / mAh g

-1

P

otential / V

vs. L

i/L

i+

0

0.2

0.4

0.6

0.8

1

1.2

-1000 -800 -600 -400 -200

0

200

400

600

800

Capacity / mAh g-1

P

o

tential / V

vs. L

i/L

i+

(a)

(b)

1st charge

1st discharge

5th charge

3rd discharge

3rd charge

5th discharge

1st charge

1st discharge

6th charge

3rd discharge

3rd charge

6th discharge

Chapter 4

103

Figure 4.3.4 The charge-discharge curves of (a) mesoporous Sn, (b) dense Sn at

second stage or later. The charge-discharge current density was 0.994 A/g (1C

rate) in the potential range of 0.01 to 1 V vs. Li/Li+.

0.0

0.2

0.4

0.6

0.8

1.0

1.2

-800 -600 -400 -200

0

200

400

600

800

Capacity / mAh g

-1

P

otential / V

vs. L

i/L

i+

100th 30th 20th 15th 5th

0

0.2

0.4

0.6

0.8

1

1.2

-800 -600 -400 -200

0

200

400

600

800

Capacity / mAh g

-1

P

otential / V

vs. L

i/L

i+

(a)

(b)

5th 15th 20th 30th 100th

100th 40th 20th 10th 6th

6th 10th 20th 40th 100th

Chapter 4

104

Figure 4.3.5 The cyclic voltammogram of the mesoporous Sn (black line) and the

dense Sn anodes (gray line) at 1st cycle. The potential range was open-circuit

potential → 0.01 to 1 V vs. Li/Li+, at a scan rate of 0.2 mV/sec.

-1.0

-0.5

0.0

0.5

1.0

0

0.5

1

1.5

Potential / V vs. Li/Li+

C

urrent D

ensity / m

A

cm

-

2

Chapter 4

105

Figure 4.3.6. The discharge capacities of the mesoporous Sn (●) and the dense Sn

anodes (○). The charge-discharge current density was 994 mA/g (1 C rate) in the

potential range of 0.01 to 1 V vs. Li/Li+.

0

200

400

600

800

0

20

40

60

80

100

Cycle number / -

D

is

c

h

a

rg

e

c

a

p

a

c

ity

/ m

A

h

g

-1

Chapter 4

106

Figure 4.3.7 shows the discharge capacity at the 1st cycle with various C

rates. With increasing C rate, the discharge capacity decreased for both anodes.

At lower C rate, the difference between mesoporous and dense Sn anodes was

hardly observed. However at higher C rate, a remarkable difference was

observed; only the mesoporous Sn anode operated at as high as 17 C rate. And

the discharge capacity at the 1st cycle with 10 C rate was 352 mAh/g, which is

almost the theoretical capacity of carbon anodes. This result is ascribable to the

alleviation of voltage loss due to the moderation of actual current density and

should be ascribed to the shortening of the solid state diffusion distance of Li. The

alleviation of voltage loss was confirmed in Figure 4.3.8. The plateaus of the

reaction (LiSn → Li7/3Sn) on both mesoporous and dense Sn anode were observed

at ca. 0.40 V vs. Li/Li+ when they were charged with 1 C rate. However the plateau

on the dense Sn anode was drastically shifted to ca. 0.12 V vs. Li/Li+ when it was

charged with 7 C rate, though the plateau on the mesoporous Sn anode was shifted

to as small as 0.26 V vs. Li/Li+ when it was charged with 10 C rate. As the result,

rate property was improved by introducing mesoporous structure to the Sn anode.

Chapter 4

107

Figure 4.3.7. The discharge capacity at the first cycle of the mesoporous Sn anode

(♦) and the dense Sn anode electrodeposited with a conventional bath (◊). The

charge-discharge current density was varied from 0.994 to 1.69 A/g (1 – 17 C rate) in

the potential range of 0.01 to 1 V vs. Li/Li+.

C rate / -

D

is

c

h

a

rg

e

c

a

p

a

c

ity

/ m

A

h

g

-1

0

200

400

600

800

0

5

10

15

20

Chapter 4

108

Figure 4.3.8. The charge-discharge curves of (a) mesoporous Sn, (b) dense Sn at 1st

cycle. The charge-discharge current density was varied from 0.994 A/g (1 C rate) to

16.9 A/g (17 C rate) in the potential range of 0.01 to 1 V vs. Li/Li+.

0

0.2

0.4

0.6

0.8

1

1.2

-1000 -800 -600 -400 -200

0

200 400 600 800

Capacity / mAh g

-1

P

otential / V

vs. L

i/L

i+

1C

2C

3C

5C

10C

15C

17C

0

0.2

0.4

0.6

0.8

1

1.2

-1000 -800 -600 -400 -200

0

200 400 600 800

Capacity / mAh g

-1

P

otential / V

vs. L

i/L

i+

1C

2C

3C

7C

(a)

(b)

Chapter 4

109

4.3.3. Morphology and Structure change with cycle

Figure 4.3.7 shows the surface morphology of the mesoporous and the dense

Sn anode, observed with SEM. The ditch aligned in constant direction was

observed on the both the mesoporous and the dense Sn anode as deposited. These

ditches were derived from the surface of Cu substrate. After a few cycles, both the

mesoporous and the dense Sn anode formed cracks like island due to the volume

change with cycles. These clacks contributed to the increase the electrochemically

active surface area, so that the discharge capacities were increased at fist stage,

described in Fig. 4.3.6.(35)

Figure 4.3.8 shows the wide-area surface morphology of

the mesoporous and the dense Sn anode after 3 cycles. The clacks of the

mesoporous Sn anode after 3 cycles were smaller than that of the dense Sn anode.

This smaller clack formation should contribute to the suppression of the stress in

the electrode. As the result, the discharge capacity degradation, which was

observed at second stage indicated in Fig. 4.3.6, by the disintegration of the anode

from the Cu substrate should be controlled on the mesoporous Sn anode.

Chapter 4

110

Figure 4.3.7 SEM images of mesoporous Sn anode and dense Sn anode (as-deposited,

after 1st cycle, and after 3rd cycle at 1 C rate in the potential range of 0.01 to 1 V vs.

Li/Li+.).

Figure 4.3.8 Wide-area SEM images of (a) mesoporous Sn anode and (b) dense Sn

anode after 3rd cycle shown in Fig. 4.3.7.

As Deposited

After 3 cycling

Mesoporous

Sn anode

After 1 cycling

12μm

12μm

12μm

12μm

12μm

12μm

Dense

Sn anode

12μm

12μm

12μm

12μm

12μm

12μm

80 μm

80 μm

(a)

(b)

Chapter 4

111

Figure 4.3.9 shows the TEM image of the mesoporous and dense Sn anode

after the 100 cycles and its selected-ED pattern. From Fig. 4.3.9 (a), the formation

of the Sn grains of 10 nm in size was observable, indicated by the arrow (Fig. 4.3.9

(a)). The ring pattern including the intense spots (Fig. 4.3.9 (a)) shows the

presence of both very minute polycrystalline phase and crystal Sn grains. From

Fig. 4.3.9 (b), the formation of the Sn grains was not observed in this area, however

the formation of the Sn grains of 20-50 nm in size was observed from the dense Sn

after the 30 cycles, hence the Sn grains could be formed in the other area. There

was the report that an amorphous-like phase should be effective to improve the

cycle durability of Sn anode.(36)

Therefore, the very minute polycrystalline phase

should contribute to the improvement of the cycle durability of the mesoporous Sn

anode. When the dense Sn anode was used, larger Sn crystals grow easily, after

the charge-discharges of Li ions are repeated for several times.(35)

On the contrary,

the mesoporous Sn anode still retains its very minute polycrystalline phase even

after the 100th cycle. Therefore, the presence of the mesoporosity should prevent

the larger Sn crystal growth during cycling, leading to a longer life.

Chapter 4

112

Figure 4.3.9 Bright-field TEM images and selected ED patterns in a 100 nm region.

(a) the mesoporous Sn after 100 cycles. (b) the dense Sn after 100 cycles. The

samples were scratched from the substrate. The powdery samples were dispersed in

ethanol and mounted on a TEM microgrid.

20 nm

20 nm

(a)

(b)

Chapter 4

113

4.4. Summary

In summary, a mesoporous Sn anode was prepared from LLC bath by

electrodeposition. The cycle durability and charge-discharge property with high

current density were improved. The discharge capacity of the mesoporous anode

was 425 mAh/g after the 100th cycle, while that of the dense anode was only 46

mAh/g. The mesoporous structure suppressed the influence of the volume change

during the charge-discharge cycling of Li-ions, and provided the high cycle

durability to the Sn anode. The high cycle durability should be affected by a very

minute polycrystalline phase. Moreover, the improvement of the charge-discharge

property with high current density at the first cycle was demonstrated by the

introduction of the mesoporous structure. Only the mesoporous Sn anode operated

at as high as 17 C rate; the discharge capacity at the 1st cycle with 10 C rate was

352 mAh/g, which is almost the theoretical capacity of carbon anodes. This result

is ascribable to the alleviation of voltage loss due to the moderation of actual

current density and should be ascribed to the shortening of the solid state diffusion

distance of Li. Thus, it was demonstrated that the introduction of mesoporous

structure was effective to improve the cycle and rate properties of the Sn anodes.

Chapter 4

114

References

1.

I. A. Courtney and J. R. Dahn, Journal of the Electrochemical Society, 144, 2943

(1997).

2.

I. A. Courtney and J. R. Dahn, Journal of the Electrochemical Society, 144, 2045

(1997).

3.

Y. Idota, T. Kubota, A. Matsufuji, Y. Maekawa and T. Miyasaka, Science, 276, 1395

(1997).

4.

H. Mukaibo, T. Sumi, T. Yokoshima, T. Momma and T. Osaka, Electrochemical and

Solid State Letters, 6, A218 (2003).

5.

J. H. Ahn, Y. J. Kim, G. X. Wang, M. Lindsay, H. K. Liu and S. X. Dou, Materials

Transactions, 43, 63 (2002).

6.

G. M. Ehrlich, C. Durand, X. Chen, T. A. Hugener, F. Spiess and S. L. Suib, Journal

of the Electrochemical Society, 147, 886 (2000).

7.

O. Crosnier, T. Brousse, X. Devaux, P. Fragnaud and D. M. Schleich, Journal of

Power Sources, 94, 169 (2001).

8.

O. Mao and J. R. Dahn, Journal of the Electrochemical Society, 146, 414 (1999).

9.

O. Mao and J. R. Dahn, Journal of the Electrochemical Society, 146, 423 (1999).

10.

O. Mao, R. A. Dunlap and J. R. Dahn, Journal of the Electrochemical Society, 146,

405 (1999).

11.

O. Mao, R. L. Turner, I. A. Courtney, B. D. Fredericksen, M. I. Buckett, L. J. Krause

and J. R. Dahn, Electrochemical and Solid State Letters, 2, 3 (1999).

12.

D. Larcher, L. Y. Beaulieu, O. Mao, A. E. George and J. R. Dahn, Journal of the

Electrochemical Society, 147, 1703 (2000).

13.

K. D. Kepler, J. T. Vaughey and M. M. Thackeray, Electrochemical and Solid State

Letters, 2, 307 (1999).

14.

D. Larcher, L. Y. Beaulieu, D. D. MacNeil and J. R. Dahn, Journal of the

Electrochemical Society, 147, 1658 (2000).

15.

G. X. Wang, L. Sun, D. H. Bradhurst, S. X. Dou and H. K. Liu, Journal of Alloys and

Compounds, 299, L12 (2000).

16.

N. Tamura, R. Ohshita, M. Fujimoto, S. Fujitani, M. Kamino and I. Yonezu,

Journal of Power Sources, 107, 48 (2002).

17.

L. Beaulieu, D. Larcher, R. A. Dunlap and J. R. Dahn, Journal of Alloys and

Compounds, 297, 122 (2000).

18.

H. Inoue, in International Meeting on Lithium Batteries, Biarritz, France (2006).

Chapter 4

115

19.

T. Yanagisawa, T. Shimizu, K. Kuroda and C. Kato, Bulletin of the Chemical

Society of Japan, 63, 988 (1990).

20.

C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and J. S. Beck, Nature, 359,

710 (1992).

21.

A. Monnier, F. Schuth, Q. Huo, D. Kumar, D. Margolese, R. S. Maxwell, G. D.

Stucky, M. Krishnamurty, P. Petroff, A. Firouzi, M. Janicke and B. F. Chmelka,

Science, 261, 1299 (1993).

22.

Q. S. Huo, D. I. Margolese, U. Ciesla, P. Y. Feng, T. E. Gier, P. Sieger, R. Leon, P. M.

Petroff, F. Schuth and G. D. Stucky, Nature, 368, 317 (1994).

23.

D. M. Antonelli and J. Y. Ying, Angewandte Chemie-International Edition in

English, 34, 2014 (1995).

24.

U. Ciesla, S. Schacht, G. D. Stucky, K. K. Unger and F. Schuth, Angewandte

Chemie-International Edition in English, 35, 541 (1996).

25.

G. S. Attard, C. G. Goltner, J. M. Corker, S. Henke and R. H. Templer, Angewandte

Chemie-International Edition in English, 36, 1315 (1997).

26.

G. S. Attard, J. M. Elliott, P. N. Bartlett, A. Whitehead and J. R. Owen,

Macromolecular Symposia, 156, 179 (2000).

27.

A. H. Whitehead, J. M. Elliott and J. R. Owen, Journal of Power Sources, 82, 33

(1999).

28.

A. H. Whitehead, J. M. Elliott, J. R. Owen and G. S. Attard, Chemical

Communications, 331 (1999).

29.

S. S. Funari, M. C. Holmes and G. J. T. Tiddy, Journal of Physical Chemistry, 98,

3015 (1994).

30.

S. S. Funari, M. C. Holmes and G. J. T. Tiddy, Journal of Physical Chemistry, 96,

11029 (1992).

31.

Y. Yamauchi, T. Yokoshima, H. Mukaibo, M. Tezuka, T. Shigeno, T. Momma, T.

Osaka and K. Kuroda, Chemistry Letters, 33, 542 (2004).

32.

T. Homma, H. Sato, H. Kobayashi, T. Arakawa, H. Kudo, T. Osaka, S. Shoji, Y.

Ishisaki, T. Oshima, N. Iyomoto, R. Fujimoto and K. Mitsuda, Journal of

Electroanalytical Chemistry, 559, 143 (2003).

33.

M. Winter and J. O. Besenhard, Electrochimica Acta, 45, 31 (1999).

34.

Z. H. Yi, Y. G. Liang, X. F. Lei, C. W. Wang and J. T. Sun, Materials Letters, 61,

4199 (2007).

35.

H. Mukaibo, T. Momma and T. Osaka, Journal of Power Sources, 146, 457 (2005).

36.

M. Mohamedi, S. J. Lee, D. Takahashi, M. Nishizawa, T. Itoh and I. Uchida,

Electrochimica Acta, 46, 1161 (2001).

Chapter 5:

General Conclusion

Chapter 5

119

The aim of this dissertation is to investigate the potentials of the

introduction of phase separated electrolytes to preparation process of materials for

lithium secondary batteries and suggest the availability of phase separated

electrolytes for improvement of the materials.

In this chapter, the results obtained from the research described in Chapter

2, 3, and 4 were summarized and a possibility of phase separated electrolytes for a

future work.

I. The polymer blend gel electrolyte composed of PEO-LiBF4 complex and PS.

(Chapter 2)

The PEO-LiBF4/PS polymer blend gel electrolyte indicated the ionic conductivity

of 0.9 x 10-3 S/cm as the maximum value by optimizing the Li+ concentration in the

plasticizer, even though it possessed mechanically strong enough to handle, because

it had co-continuous three-dimensionally interpenetrated polymer network (IPN)

structure. In addition, the chemical stability of the interface between the IPN gel

electrolyte and lithium metal was improved compared with the PEO/PS polymer

blend gel electrolyte, because of the retentivity improvement of the ionic conductive

phase due to PEO-LiBF4 complex. The stability against the oxidation was enough

to be applied to LiCoO2

cathode which is 4 V class cathodes. With the

charge-discharge rate of C/5, the capacity per gram of cathode was found to be over

130 mAh/g over 30 cycles.

II. The di-block copolymer gel electrolyte composed of PEO and PS. (Chapter 3)

The ionic conductivity of the PEO-PS di-block copolymer gel electrolyte

demonstrated to be 2.65 mS/cm at maximum value, even though it was

mechanically strong enough to suppress the dendritic deposition of lithium at the

interface between Li electrode and the gel electrolyte. The excellent chemical

stability of the PEO-PS di-block copolymer gel electrolyte with Li electrode was

revealed. This result was led by fixation of the PEO with PS matrix by covalent

bond. The stability against the oxidation was found to be enough for application of

LiFePO4 cathode. With the charge-discharge rate of C/5, the capacity per gram of

cathode was found over 120 mAh/g over 30 cycles.

Chapter 5

120

III. Mesoporous Sn anode electrodeposited with a lyotropic liquid crystal bath.

(Chapter 4)

The significant improvements of Sn anode on the cycle durability and

charge-discharge property with high current density were discovered by

introduction of mesoporous structure. The discharge capacity of the mesoporous

Sn anode was 425 mAh/g after the 100th cycle, while the anode without mesoporous

structure was only 46 mAh/g. The mesoporous structure suppressed the influence

of the volume change during the charge and discharge cycling of Li-ions, and the

high cycle durability was realized. The high cycle durability should be affected by

a very minute polycrystalline phase. Moreover, the improvement of the

charge-discharge property with high current density at the first cycle was

demonstrated by the introduction of the mesoporous structure. Only the

mesoporous tin anode operated at as high as 17 C rate; the discharge capacity at the

1st cycle with 10 C rate was 352 mAh/g, which is almost the theoretical capacity of

carbon anodes. This result is ascribable to the alleviation of voltage loss due to the

moderation of actual current density and should be ascribed to the shortening of the

solid state diffusion distance of Li.

Chapter 5

121

These materials indicated potential performances as described above,

moreover the benefits are expected as follows.

The polymer blend gel electrolyte can be prepared from PEO and PS, which

have been widely used industrially, by simple hot-blending method without volatile

organic solvents, which are used for most preparation of polymer electrolytes.

The interfacial stability between the di-block copolymer gel electrolyte and

Li metal anode was established by its covalent bond, enable to suppress the

performance degradation at long storage. And the large film of block copolymer has

come to be prepared recently.

The mesoporous Sn anode which has high performances can be prepared by

electrodeposition, hence it can be prepared on large substrate. It is beneficial for

mass production.

The author believes that these results demonstrate the potential of phase

separated electrolytes, and has visions that anode and electrolyte would be

prepared all together three-dimensionally; the surfactant itself act as polymer

electrolyte by controlling the surfactant which is used as template at

electrodeposition of anode in future. If this process is brought to realization, some

benefit can be received; the degradation of Sn anode due to its volume change at

charge-discharge cycles can be suppressed by polymer matrix, and the organic

solvent for removal of surfactant as template becomes to unnecessary.

Appendix

“Numerical Simulation of DMFC-Capacitor Hybrid Power Supply

System for Small Electric Devices”

Appendix

125

1. Introduction

In recent years, mobile electronic devices including cellular phone, laptop computer,

and PDA, which are generally equipped with Li ion batteries, require large power output of the

electric power supply used to operate those devices. Besides the development of high energy

density, rechargeable Li ion batteries and small fuel cells are proposed as the electric power

supply for small electric devices of the future. Fuel cells attract many researchers because of

conceivable advantages, such as an enhanced energy density and a shorter period of time

required for recharging the energy consumed. Fuel cells, especially direct liquid-fuel fuel cells

such as direct methanol fuel cell (DMFC), are being developed to fulfill the need for high energy

density power supplies [1,2]. The direct liquid-fuel fuel cells have a high energy density, and

much efforts are being expended to improve their output power density to reduce the

production cost and the size of the fuel cells.

In designing a power supply for electric devices, both maximum and mean power

requirements should be taken into account together with the length of time during which a

large power consumption occurs. To satisfy the requirement of the electric devices, a power

supply consisting of fuel cells only should be designed and fabricated to provide maximum

power output. However, hybrid power supply systems, such as DMFC connected with another

power supply, capacitors [1, 4, 5] or secondary batteries [6-10], level the requirement of DMFC

output into the mean value in a short period of time by supplying the pulsed output from

another power supply. Especially capacitors are advantageous for use with a pulsed load,

because of its character, short charge-discharge time and extensive cycle durability.

The simulation of hybrid power supply systems is important, because, for constructing

such a system, it is useful to know the required power output of DMFC and capacitor for the

specific electric device without performing actual fabrication. The DMFC and capacitor can

then be designed for the required power output which has been calculated by the simulation.

In the present work, an internal resistance of DMFC was approximated by a simple

resistance based on a current-voltage profile of a DMFC. The simple numerical simulation of a

hybrid power supply system consisting of a DMFC and a capacitor connected in parallel was

attempted and the obtained results were discussed with the current profile of hybrid power

supply set up. In addition, the possibility for constructing a smaller hybrid power supply

system was confirmed for the system consisting of our DMFC and micro electrochemical

capacitor (MECC).

Appendix

126

2. Modeling

As electric devices used for the modeling study, two Japanese cellular phones (2G;

KDDI au, 2004 product, and 3G; NTT Docomo, 2006 product) were chosen to obtain power

consumption profiles. The power consumption profiles, when the constant voltage of 3.7 V was

applied to the cellular phone with removal of lithium battery pack, was determined by recorded

from the current flow to the cellular phone from the DC power supply (potentiostat /

galvanostat, HA-301, Hokuto Denko Corp.), every 20 and 50 msec with a voltage recorder

(Voltage Data Logger, VR-71, T&D corporation) as shown in Fig. 1 (a) and (b), respectively. The

power consumption profile shown in Fig. 1 (c) is described in simulation and discussion section.

The current profiles were assumed as merely a pulsed load, though a capacitor might have

been built in the 3G cellular phone.

Appendix

127

Fig. 1. Power consumption profiles of a cellular phone. (a) when calling with 2G, recorded every

20 msec, (b) when watching TV with 3G, recorded every 50 msec, and (c) when watching TV

and changing the channel with 3G, recorded every 20 msec. Note the power consumption

profile of (c) was derived from the sum of the current flow passed through the DC power supply

and capacitor connected in parallel.

0

0.2

0.4

0.6

0

20

40

60

80

100

Time / sec

I lo

ad

/ A

0

0.2

0.4

0.6

0

100

200

300

400

Time / sec

I lo

ad

/ A

0

0.2

0.4

0.6

0

10

20

30

40

Time / sec

I lo

ad

/ A

(a)

(b)

(c)

Appendix

128

A parallel connection of a DMFC and a capacitor was considered in formulating the

model of hybrid power supply in this work. The capacitor in this system is required to supply a

pulsed high power when demanded by the electric device. The DMFC supplies the power

leveled by the capacitor to the electric device and charges the capacitor.

In this study, the DMFC was fabricated in a same way presented by Shimizu et al. [3],

to obtain the current-voltage and current-power curves in its operation as shown in Fig. 2. The

operation range of the power supply for cellular phone was considered in the current range

lower than the value at which the maximum power was obtained. As many factors:

electrochemical reaction, methanol crossover, and water management, have to be considered

for a DMFC performance, the current-voltage data actually measured as shown in Fig. 2 should

be containing all factors affecting its performance; while the current-voltage profile shown in

Fig. 2 can be ascribed linearly aligning as long as in the current range of 0.025 to 0.7 A where

the DMFC will be operated. Accordingly, it would be possible to simulate with a simple

resistance which approximated an internal impedance of a DMFC. While the electric double

layer of the electrode in the DMFC is considered to be take into account, in this simulation the

electric double layer capacitance is omitted because of it’s negligibly small charge compared

with the generated current during the fuel cell operation; actually the simulation results

indicated in Fig. 4 showed good agreement with the current profile in the real world. Then,

with the assumptions listed below, an equivalent circuit of the hybrid type power supply

system shown in Fig. 3 was constructed.

Fig. 2. Polarization (■) and power density (□) curves. of DMFC (25 cm2 area).Catalysts:

unsupported Pt–Ru anode (6.4 mg cm−2 of Pt–Ru) and unsupported Pt cathode (3.9 mg cm−2 of

Pt), current collector SUS/Au, 2 M methanol solution, air passive, room temperature, and

Nafion 117.

0

0.1

0.2

0.3

0.4

0.5

0.6

0

0.2

0.4

0.6

0.8

1

IFC / A

V

F

C

/ V

0

0.1

0.2

0.3

P

o

w

er / W

Voltage

Power

Appendix

129

1.

The internal impedance of the fuel cell, which is a combination of ohmic resistance,

charge transfer reaction resistance, mass transfer resistance and double layer

capacitance, is represented as a simple ideal resistor in the operation range. DMFC

generates the power output with a voltage which is linearly related to the output

current with the ideal resistance described above. The virtual open circuit voltage is

the value of intersection found by extrapolating the current-voltage plots of the

operation current range to the voltage axis. The value of intersection is lower than the

real open circuit voltage of DMFC by the voltage drop caused by the charge-transfer

resistance.

2.

Capacitor is described as a serial connection of ideal capacitance and ideal resistance,

that is equivalent series resistance (ESR), which includes resistances of electrolyte,

electrodes, and lead wire in a capacitor.

The parameters in Fig. 3 are shown in Table 1. The resistance and the virtual open circuit

voltage of DMFC were calculated from the current-voltage profile of DMFC shown in Fig. 2;

this graph yields the resistance value of 0.27 Ω, and the virtual open circuit voltage of 0.485 V.

As the maximum output of the DMFC was about 0.2 W shown in Fig. 2, the DMFC stack of 10

DMFCs connected in series was designed in this study to modeling for operation of cellular

phone with its consumption of 2 W estimated from Fig. 1. RFC and VFC_ocv were then decided

from the resistance value and the virtual open circuit voltage for 10 DMFCs connected in series

for the operation of the cellular phone. The values of Rcap and C were measured by the a.c.

impedance method. For the equivalent circuit, the following numerical expressions were used

with the assumption that these values are constant during a length of time (the finite element

method). The parameters in below expressions are described in Table. 1.

Appendix

130

Fig. 3. Equivalent circuit of hybrid power supply system. The parameters are listed in Table. 1.

Table 1. Parameters in the equivalent circuit of hybrid power supply system.

VFC_ocv

Virtual open circuit voltage

V

Voltage of hybrid power supply system

Iload

Current flow through the electric devices

IFC

Current flow through DMFC

Icap

Current flow through capacitor

RFC

Inner resistance of DMFC

Rcap

Inner resistance of capacitor, ESR

Q

Quantity of charged electricity

C

Capacitance

V

Load

Iload

C

VFC_ocv

FC

Rcap

IFC

Icap

C

R

I

Appendix

131

0

)(

)(

)(

=

+

+

m

m

m

t

load

t

cap

t

FC

I

I

I

(1)

FC

t

FC

t

ocv

FC

R

I

V

V

m

m

×

=

-

)(

_

(2)

cap

t

cap

t

t

R

I

VCQ

m

m

m

×

=

-

)(

(3)

cap

I

dt

dQ

-

=

(4)

)(

_

m

m

m

t

load

cap

FC

cap

FC

ocv

FC

cap

FC

cap

t

cap

FC

FC

t

I

R

R

R

R

V

R

R

R

C

Q

R

R

R

V

×

+

×

+

×

+

+

×

+

=

(5)

（∵ (1), (2), (3)）

)

(

1

)

(

1

1

-

-

×

-

=

-

-

m

m

t

cap

t

t

tt

I

Q

Q

m

m

m

(6)

（∵(4)）

where tm indicates time.

With this model, the numerical simulation of current flow was performed with Microsoft Excel

by assigning the power consumption profiles as Iload (tm) shown in Fig. 1. As the results, V(tm),

IFC (tm), Icap (tm) and Q(tm) were calculated. The resolution time of the simulation was used the

same value at recording of the power consumption profiles. At the starting point the capacitor

was assumed to be made at the fully charged state by the fuel cell after a period of no

operation.

Appendix

132

3. Simulation and Discussion

To confirm the appropriateness of the proposed numerical simulation, the magnitudes

of currents actually measured were compared with currents calculated. For actual

measurement, a hybrid power supply was constructed with parallel connection of the DC power

supply with a series resistor, which represents the internal resistance of DMFC, and two

laminate type series capacitors, which are EDLCs, for achieving durability at cellular phone

operation voltages. The power consumption profile shown in Fig.1 (c) was determined from the

sum of the current flow passed through the DC power supply and electric double layer

capacitor (EDLC) connected in parallel, which are recorded every 20 msec with ampere meter

(zero shunt ammeter, HM-104, Hokuto Denko Corp.) and the voltage recorder; the parallel

connection of the DC power supply and EDLC were assumed to be hybrid power supply.

Though the power consumption profile shown in Fig.1 (c) was measured in another way, it can

be dealt as the power consumption profile shown in Fig. 1 (a), (b). The actual currents flowing

through the DC power supply and the capacitor to the cellular phone were shown by gray lines

in Fig. 4. The simulated currents based on the load current shown in Fig.1 (c) were shown by

black lines in Fig. 4. The simulated currents were calculated by using the values of Rcap, C, RFC,

and VFC_ocv. RFC is the value of the resistor connected with the DC power supply, and VFC_ocv is

the value applied with the DC power supply. The current profiles, i.e., values of measured IFC,

Icap and simulated IFC, Icap, are compared in Fig. 4. The small difference between measured and

simulated profiles may be attributable to the resistance of the circuit. This result demonstrates

that our numerical simulation method is applicable to the case using the hybrid power supply.

Appendix

133

Fig. 4. Profiles of current flowing through DMFC and capacitor. Simulated current profiles

(black line) and measured current profiles (gray line).

-0.2

0

0.2

0.4

0.6

0

20

40

60

80

100

Time / sec

C

urrent / A

Current flowing through capacitor

Current flowing through DMFC

Appendix

134

Figure 5 shows examples of simulated results for load current with two laminate type

capacitors in series, 1.92 F and 0.42 Ω, and with a virtual capacitor, with 1.92 F and 4.2 Ω, the latter

being 10 times as larger as the laminate type capacitor resistance, that is ESR. The load leveling rate

was calculated using the formula defined as follows:

100

[%] rate

leveling

Load

max

_

max

_

max

_

×

-

=

load

FC

load

I

I

I

(7)

where Iload_max, and IFC_max are maximum load current and current flow through the DMFC in the

range of measuring time, respectively. The load leveling rate were calculated with various load

currents and capacitances. The load leveling rate of (a), (b), (c), (d), (e) and (f) in Fig. 5 were 62.7,

27.3, 22.1, 38.9, 14.7, and 12.9 % respectively. These results indicate the importance of ESR,

implying that the higher ESR disables the capacitor in the hybrid power supply system for charging

and discharging. Therefore, the influence of capacitor with several values of capacitance and ESR

to the load leveling rate was simulated. Figure 6, (a), (b), (c) show the load leveling rate for the load

indicated in Fig. 1, (a), (b), (c), respectively. These results indicated that the load leveling rate is

converged with increasing ESR, even if the capacitor in the hybrid power supply system has a high

capacitance. Consequently, it was demonstrated that ESR is a very important factor to consider,

when a hybrid power supply system is constructed. In addition, this load leveling effect suggests the

possibility of downsizing the surface area of the DMFC electrode. In the case of (a) in Fig. 5, only

ca. 40 % of the DMFC electrode area is necessary for the cellular phone operation, compared with

the area required without the laminate type capacitors.

Appendix

135

Fig. 5. Profiles of current flowing through DMFC (black line) and capacitor (gray line) with

different capacitances; (a) - (c) 1.92F, 0.42 Ω, and (d) - (f) 1.92F, 4.2 Ω, (a) and (d): when making

a call with 2G; (b) and (e): when watching TV with 3G; (c) and (f) when watching TV and

change channels with 3G.

Fig. 6. Load leveling rate vs. ESR for capacitors with various capacitances, calculated on the

load current profiles; (a) when making a call with 2G, (b) when watching TV with 3G, (c) when

watching TV and change channels with 3G.

(a)

(d)

(b)

(e)

(c)

(f)

-0.2

0

0.2

0.4

0.6

C

u

rre

n

t / A

0

20

40

60

80

100

-0.2

0

0.2

0.4

0.6

0

10

20

30

40

Time / sec

C

u

rre

n

t / A

0

100

200

300

400

Time / sec

0

20

40

60

80

100

Time / sec

(a)

(b)

(c)

Appendix

136

The possibility of constructing a smaller hybrid power supply system was also examined

for our μDMFC [11-14] and micro electrochemical capacitor (MECC) [15, 16] for the 2G cellular

phone. The parameters of μDMFC were determined from the current-voltage curve [17]. The

resistance and the value of virtual open circuit voltage were 361 Ω and 0.333 V respectively. Values

of RFC and VFC_ocv were derived from the resistance and the virtual open circuit voltage of 15

μDMFCs, note that μDMFCs connected in series is called a “μDMFC-stack” in this paper,

assembled in series for the cellular phone application. To obtain a sufficiently large current, these

components of the stacks were assumed to be connected in parallel. The parameters of MECC used

were that announced by Takasu et al. [15], and MECCs were also assumed to be connected in

parallel. In this case, 1000 μDMFC-stacks and 3000 MECC units were connected in parallel. Fig. 7

shows voltage and current profiles for μDMFC and MECC. From Fig. 7 (a), (b), the minimum

voltage is 3.37 V, which is large enough to operate the cellular phone, in the range of measurement

time. The load leveling rate is 48.5 %, which indicates the MECCs functioned effectively. Here, the

volume of each cell was 1 cm

3

. Therefore the volume of this hybrid power supply system was

estimated to be 18000 cm

3

, which is smaller than 25000 cm

3

, the volume of the power supply

system without MECC for operating the cellular phone.

Appendix

137

Fig. 7. Voltage (a), (c) and current (b), (d) profiles of the hybrid power supply system consisting

of μDMFC and MECC. The profiles of (c), (d) refer to the case of the voltage raised with a

DC-DC converter (efficiency; 80 %). Profiles of current flowing through μDMFC and MECC

indicated by black line and gray line.

0

2

4

6

V

o

ltag

e / V

0

4

8

12

0

10

20

30

40

Time / sec

V

o

ltag

e / V

(a)

(b)

(c)

(d)

-0.2

0

0.2

0.4

0

10

20

30

40

Time / sec

C

u

rren

t / A

-0.2

0

0.2

0.4

C

u

rren

t / A

Appendix

138

In addition, a DC-DC converter that is charge pump assumed to be adapted to this system,

which enables to boost the voltage of the power supply, with reducing the current flowing, so that

the influence of the IR drop can be alleviated. The calculation with a DC-DC converter was simply

conducted; when it raises the voltage of the system, the current of the system is reduced

commensurately and is reduced in consideration of the loss by its efficiency. Figure 7 (c), (d) shows

the voltage and current profiles in case that a DC-DC converter (efficiency was assumed to be

80 %) which decuple the voltage of the system was adapted to this system which is composed of

4500 μDMFC-stacks, each consisting of 3 μDMFCs in series, and 1000 MECC units were

connected in parallel. The total volume of this hybrid power supply system was estimated to be

14500 cm

3

. Thus the DC-DC converter enables to decrease the volume by about 20 %. In spite of

the decrease in the volume, the load leveling rate is as high as 59.9 %. Consequently, it was

suggested that with the use of a DC-DC converter, further downsizing was possible. The μDMFC

and MECC used in this simulation were so of prototype that the electrode areas of μDMFC and

MECC were considerably smaller against the cell size. Therefore, those electrode areas would be

able to be expanded by a factor of 50 by the accumulation of the electrodes. When the electrodes of

μDMFC and MECC are accumulated, the volume of the hybrid power supply system would got

smaller by a factor of 50, i.e., to 290 cm

3

. Moreover, downsizing of μDMFC and MECC should be

achievable by increasing the effective electrode area, improving electrode performances, and

suppressing methanol crossover. Improvement of these factors may enable our hybrid power supply

system to become comparable to the contemporary standard system developed by the industry.

Appendix

139

4. Conclusion

Numerical simulation of a hybrid power supply system consisting of a DMFC and a capacitor

connected in parallel was simply investigated based on a known current-voltage data. The

possibility of designing a smaller hybrid power supply system was demonstrated for use with our

μDMFC and MECC. The simulated profiles of current flowing through the DMFC and the

capacitor demonstrated an excellent agreement with experimental profile. The ESR was revealed to

be a very important factor in the hybrid power supply system of DMFC and capacitor. The

numerical simulation helps to find improvements required of hybrid power supply system for

practical applications.

Appendix

140

Reference

1) T. Yoshitake, H. Kimura, S. Kuroshima, S. Watanabe, Y. Shimakawa, T. Manako, S.

Nakamura, and Y. Kubo, Electrochemistry, 70, 966 (2002).

2) T. Shimizu, M. Mohamedi, T. Momma, and T. Osaka, Electrochemistry, 72, 637 (2004).

3) T. Shimizu, T. Momma, M. Mohamedi, T. Osaka and S. Sarangapani, J. Power Sources, 137,

277 (2004).

4) L.P. Jarvis, T.B. Atwater, and P.J. Cygan, J. Power Sources, 79, 60 (1999).

5) K. Park, H. Ahn, and Y. Sung, J. Power Sources, 109, 500 (2002).

6) J. Han and E. Park, J. Power Sources, 112, 477 (2002).

7) Z. Jiang, L. Gao, M.J. Blackwelder, and R.A. Dougal, J. Power Sources, 130, 163 (2004).

8) L. Gao, Z. Jiang, and R.A. Dougal, J. Power Sources, 130, 202 (2004).

9) B. Lee, D. Jung, and Y. Ko, J. Power Sources, 131, 207 (2004).

10) M.J. Blackwelder and R.A. Dougal, J. Power Sources, 134, 139 (2004).

11) S. Motokawa, M. Mohamedi, T. Momma, S. Shoji, and T. Osaka, Electrochemistry

communications, 6, 562 (2004).

12) S. Motokawai, H. Obata, M. Mohamedi, T. Momma, S. Shoji, and T. Osaka,

Electrochemistry, 5, 352 (2005).

13) S. Motokawa, M. Mohamedi, T. Momma, S. Shoji, and T. Osaka, Electrochemistry, 5, 346

(2005).

14) M. Ishizuka, H. Houjou, S. Motokawa, J. Mizuno, T. Momma, T. Osaka, and S. Shoji,

JAPANESE JOURNAL OF APPLIED PHYSICS PART 1-REGULAR PAPERS BRIEF

COMMUNICATIONS & REVIEW PAPERS, 45, 7944 (2006).

15) M. Omoto, W. Sugimoto, K. Yokoshima, J. Park, T. Momma, T. Osaka, and Y. Takasu, 73th

Meeting of The Electrochemical Society of Japan, Abstract #1L28 (2006).

16) W. Sugimoto, K. Ohuchi, K. Yokoshima, J. Park, T. Momma, T. Osaka, and Y. Takasu, 208th

Electrochemical Society Meeting, Abstract #833 (2005).

17) S. Tominaka, H. Obata, T. Momma, J. Park, and T. Osaka, EMNT2006 (2006).

List of Achievement

141

List of Achievements

1. Original Articles

“Characteristics of Interpenetrated Polymer Network System made of

Polyethyleneoxide -LiBF4 Complex and Polystyrene as the Electrolyte for Lithium

Secondary Battery”

Toshiyuki Momma, Hiroaki Ito, Hiroki Nara, Hitomi Mukaibo, Stefano Passerini,

Tetsuya Osaka

Electrochemistry, 71, 1182 (2003).

“Numerical Simulation of DMFC-Capacitor Hybrid Power Supply System for Small

Electric Devices”

Toshiyuki Momma, Hiroki Nara, Tetsuya Osaka

Electrochemistry (Accepted)

“Feasibility of an Interpenetrated Polymer Network System Made of Diblock

Copolymer Composed of Polyethylene oxide and Polystyrene as the Gel Electrolyte

for Lithium Secondary Batteries”

Hiroki Nara, Toshiyuki Momma, Tetsuya Osaka

Electrochemistry (Accepted)

“Cycle and Rate Property of Mesoporous Tin Anode for Lithium Ion Secondary

Batteries”

Hiroki Nara, Yoshiki Fukuhara, Hitomi Mukaibo, Yusuke Yamauchi, Toshiyuki

Momma, Kazuyuki Kuroda, and Tetsuya Osaka.

Chemistry Letters, 37, 142 (2008).

“Mechanical Analysis and In-situ Structural and Morphological Evaluation of Ni-Sn

Alloy Anodes for Li Ion Batteries”

J. Chen, S. J. Bull, S. Roy, H. Mukaibo, H. Nara, T. Momma and T. Osaka, Y.

Shacham-Diamand

J. Phys. D: Appl. Phys., 41, (2008) 025302 (13pp).

List of Achievement

142

2. Oral Presentations

A．International Conferences (Presented by H. Nara, et al.)

“Cycle Property of Mesoporous Sn Anode for Lithium Ion Secondary Batteries”

212th Meeting of The Electrochemical Society, Washington, DC, USA, October

2007.

“Application of Diblockpolymer Gel Electrolyte Having Micro Phase Separation

Structure to Lithium Secondary Batteries”

The 56th Annual Meeting of International Society of Electrochemistry, Busan,

Korea, September 2005.

“Gel Electrolyte Having Micro Phase Separation Structure for Lithium Secondary

Batteries”

2nd International Conference on Polymer Batteries and Fuel Cells, Las Vegas,

USA, June 2005.

“Preparation of Gel Electrolyte Using Self-assembling Diblockpolymer for Lithium

Secondary Battery”

The Electrochemical Society (ECS) 2004 Joint International Meeting, Honolulu,

Hawaii. October 2004.

List of Achievement

143

B．Domestic Conferences (Presented by H. Nara, et al.)

“Effects of Dissolved O2 on Charge-Discharge Performance of Lithium Metal

Electrode in Organic Electrolytes with TFSI Anion Based Ionic Liquids as an

Additive”

The 48th Battery Symposium in Japan, Fukuoka, Japan, November 2007.

“Improvement of NiSn Alloy Powder for Lithium Ion Secondary Batteries by

Surface Oxide Control”

The 48th Battery Symposium in Japan, Fukuoka, Japan, November 2007.

“Cycle Performance of Sn Anode with Mesoporous Structure for Lithium-ion

Secondary Batteries”

The Electrochemical Society of Japan, Tokyo, Japan, September 2007.

“Improved Mechanical Property of Diblock-polymer Gel Electrolyte for Lithium

Secondary Battery and Battery Performance”

The chemical society of Japan, Osaka, Japan, March 2007.

“Voltage Loss Profile within the Electrolyte Membrane Caused by Proton Transfer

for Numerical Simulation of DMFC-Capacitor Hybrid Power Supply”

The 47th Battery Symposium in Japan, Tokyo, Japan, November 2006.

“Load Leveling Simulation of Direct Methanol Fuel Cell-Capacitor Hybrid System”

The Electrochemical Society of Japan, Tokyo, Japan, April 2006.

“Application of Diblockpolymer Gel Electrolyte to Lithium Secondary Batteries”

The Electrochemical Society of Japan, Kumamoto, Japan, April 2005.

“Preparation of Gel Electrolyte Using Self-assembling of Diblockpolymer for

Lithium Secondary Battery”

The chemical society of Japan, Hyogo, Japan, March 2004.

List of Achievement

144

“Characterization of the Polymer Blend with PEO-LiX Complex as the Electrolyte

for Lithium Secondary Battery”

The Electrochemical Society of Japan, Tokyo, Japan, April 2003.

“Polymer Blend Gel Electrolytes by Complex of PEO-LiX and PS”

The Electrochemical Society of Japan, Kanagawa, Japan, September 2002.

List of Achievement

145

3. Patent

Application No. 2007-293318

“DENKAIEKI, RICHIUMUNIJIDENCHI, OYOBI RICHIUMU NIJIDENCHI NO

SEIZOHOHO”

Tetsuya Osaka, Toshiyuki Momma, Chika Tatsumi, Hiroki Nara, Kunihiko Kojima,

Kentaro Tada, Satoshi Obara, Hitoshi Iwaki

Waseda University, Toyo Gosei Co., Ltd, November 2007.

Application No. 2007-119848

“HUKUGODENGEN”

Tetsuya Osaka, Toshiyuki Momma, Hiroki Nara, Masahiko Usuda, Masataka

Takeuchi

Waseda University, Showa Denko K.K, April 2007.

147

Acknowledgement

I would like to express my sincerely gratitude to Professor Dr. Tetsuya

Osaka for his supervision and continuous encouragement. I wish to express my

hearty gratitude to Professor Dr. Takayuki Homma for his advice and

encouragement. I would express my sincere appreciation to Professor Dr.

Kazuyuki Kuroda, chapter 4 in my thesis would not be achieved without his

cooperation. I would appreciate Professor Dr. Bruno Scrosati, University "La

Sapienza" Rome for valuable discussion and encouragement. I would extend my

sincere appreciation to Professor Dr. Yoshiyuki Sugawara, Professor Dr. Hiroyuki

Nishide, and Professor Dr. Yosi Shacham-Diamand for their valuable suggestions

and encouragement. And I would like to thank Associate Professor Dr. Toshiyuki

Momma for his valuable suggestions and discussion, my thesis would not be

reached completion without his advice. I would like to express my appreciation to

Professor Dr. Toru Asahi, Professor Dr. Kiyoshi Kanamura, Tokyo Metropolitan

University, Professor Dr. Masayoshi Watanabe, Yokohama National University,

Professor Dr. Hideki Masuda, Tokyo Metropolitan University, Professor Dr. N.

Furukawa, Dr. J.-E. Park and Professor Dr. Soo-Gil Park, Chungbuk National

University, for their valuable advice and discussions. It is a great pleasure to

express my deepest gratitude to Dr. Y. Okinaka for his advice, correction and

encouragement. I would tender my acknowledgments to Dr. Y. Yamauchi, Dr. H.

Mukaibo, Mr. Y. Fukuhara, Ms. A. Takai，and Mr. M. Komatsu, chapter 4 in my

thesis would not be accomplished without their cooperation. My appreciation goes

to all the people who have shared their time and talent with me in my everyday

battery life: Dr. S Motokawa, Mr. S. Kamei, Mr. H. Ito, Dr. T. Shimizu, Ms. J. Kim,

Mr. T. Sumi, Mr. N. Chihara, Ms. M. Hosoda, Mr. H. Kitoh, Mr. S. Tominaka, Mr. Y.

Wada, Ms. M. Iwasaki, Mr. M. Ueda, Mr. H. Obata, Ms. K. Saigusa, Mr. N.

Akiyama, Mr. R. Sebata, Mr. K. Takada, Mr. Y. Fukuhara, Ms. C. Tatsumi, Mr. S.

Ohta, Mr. K. Goto, Mr. S. Yamagami, Mr. H. Kawase, Mr. T. Kawano, and Mr. Y.

Nakamura. My colleagues, Mr. H. Iida, Ms. M. Matsunaga, Ms. K. Sakata and all

the other members of applied physical chemistry lab. So many of you have made me

feel at home and sometimes work hard. Finally I express greatest thanks to my

parents, Mr. Shinji Nara and Ms. Sayoko Nara, for deep affection and constant

financial support.

Hiroki Nara