1.1. GENERAL INTRODUCTION Conduction of electric c u m t is of two form a) e1ectmnic conduction and b) ionic conduction. The former case is due to the flow of electrons and later one is due to flow of ions i.e., electric charges accompanied by mass transport. Liquid electrolyte solutions exhibit ionic conduction, on applied field, by the dissociation of electrolytes. Similarly, some solid substances exhibit high ionic conductivity of the order of liquid electrolytes are called Solid Electrolytes (SEsj or Fast Ion Conductors (FICs) or Superionic Conductors (SICs) and these materials come under the field of Solid State Ionics [l - 71. Solid state ionics is an interdisciplinary area of science, which encloses the fundamental and application of physics, chemistry and material science involving all kinds of ionic transports in solid state. Wide range of SIC materials are known starting from crystalline to organic macmmolecules of different microstructures, includes single & polycrystaUine, composites, glass/amorphous and polymers of different forms like bulk, powder, monolith, thin films and etc. The SICs possess ion transport phenomena in solid state resulting in high value of conductivity (a) in the range of 1 - 10d - 10-9 Sun1 comparable to that of the order of Liquid Electrolytes (LEs) with neghgible electronic conductivity and low activation energy for ion migration \I - 3). The hi ion conduction is due to the migration of ions through inter-granular or inter-particle boundaries. The SIC materials are applied in elecb-ochemical systems to form solid state ionic devices such as solid state battaies, capacitors, gas sensors, fuel cell, e l m e m i c a l miniature cells etc. 11 - 19) and these are shown in the scheme 1.1 [3]. In focus of all these features, the chemists are enabled to synthesis many newer superionic conductme materials to sohe the major problem of energy conversion, storage and other applications [I - 251.
40
Embed
GENERALshodhganga.inflibnet.ac.in/bitstream/10603/966/7/07_chapter 1.pdf · pmicncr of dekts or imperfections to sustain.The ionic solids were classfied into two types besed on type
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1.1. GENERAL INTRODUCTION
Conduction of electric c u m t is of two form a) e1ectmnic conduction and
b) ionic conduction. The former case is due to the flow of electrons and later one
is due to flow of ions i.e., electric charges accompanied by mass transport. Liquid
electrolyte solutions exhibit ionic conduction, on applied field, by the dissociation
of electrolytes. Similarly, some solid substances exhibit high ionic conductivity of
the order of liquid electrolytes are called Solid Electrolytes (SEsj or Fast Ion
Conductors (FICs) or Superionic Conductors (SICs) and these materials come
under the field of Solid State Ionics [ l - 71.
Solid state ionics is an interdisciplinary area of science, which encloses the
fundamental and application of physics, chemistry and material science involving
all kinds of ionic transports in solid state. Wide range of SIC materials are known
starting from crystalline to organic macmmolecules of different microstructures,
includes single & polycrystaUine, composites, glass/amorphous and polymers of
different forms like bulk, powder, monolith, thin films and etc. The SICs possess
ion transport phenomena in solid state resulting in high value of conductivity (a)
in the range of 1 - 10d - 10-9 Sun1 comparable to that of the order of Liquid
Electrolytes (LEs) with neghgible electronic conductivity and low activation energy
for ion migration \I - 3). The h i ion conduction is due to the migration of ions
through inter-granular or inter-particle boundaries. The SIC materials are applied in elecb-ochemical systems to form solid state ionic devices such as solid
state battaies, capacitors, gas sensors, fuel cell, e l m e m i c a l miniature cells
etc. 11 - 19) and these are shown in the scheme 1.1 [3]. In focus of all these
features, the chemists are enabled to synthesis many newer superionic
conductme materials to sohe the major problem of energy conversion, storage
and other applications [ I - 251.
w n d a r y (-blegeable) battery electric
T cleclrolysis primary w full cell
c power
m t
coloration electric power electrochrornic display (EW photo cell
electric spa1 chemical sensors
Schcmc 1 1 ELbXTKOCt1EMICAL DEVICES
Ttus chapter p m d e s a brid description on the cladication of ion
conducting solids, the types of superionic conductors and the various models for
explaining thc ion transport mechanism m &ses of solid ionic conductors.
~ ~ y , a brid now on various applications of solid electrolytes and the aims of
the present invcsbgation an pnsented.
The phawmnon of ionic conduction is that if an dcctric fi& is applied
acms the adid dectdyte, the randomly oriented ions m m as a who& along the
direction of the appliai Ud. The ionic conduction talas place through the
pmicncr of dekts or imperfections to sustain. The ionic solids were classfied
into two types besed on type of defect or disorder as a) point defect and b) molten
sub-lattice. The point defect types of solids possess ionic mnduction through
l,'mnkel or Schottky defects and hence, the conductivity in- with
tc.mperaturr (1 , 251. In molten sub-lattice type d solids, all ions are available for
4 ontluction since the number of sites in the sublattice is more than the number
of ions and so ions can rnwe freely h r n one position to another with low
;~ct~vation energy psxsing high conductivity 111. The other method of
c~lassification is based on the mobile ions and defkrts concentration (n) into three
T y p e s : a) dilute point defects, n 5 loi8 crn considered to be normal ion
t-onduc-tors and b) txtended defects, n = loxJ cm3 & c) liquid-like molten sub-
latticr, n = 1 W cm 3 wns~derrd to be supenonn conductors
Normal ionic conductors (NICs) are malerials that exhibit low conductivity
of 10 lZ to 10 Ib Scm at ambient temperatuw e.g. NaCI, liCI. CaC12 etc. The ionic
cunduct~on mechanism m NlCs arise from the points defect or the imperfection in
the laturn s ~ t c cmted thermal& and the activation process involves both energy
due lo dckct formation (hd as weU as ion rmgration (hm) [I , 25, 261. The
conductivity expression is orpressed as follows
w h m the tx, prccxponenUal lactor, can be rpresented by
u, = e2vJ7.',4k 1 k 1.2
wherr e is the charge, vo jump or attempt frequency, f correlation factor, h jump
diapna, x fm3b of rnobik charge carriers and N charge canicr density.
1.2.3. Supsrionlc Conductors (SIC.)
Superionic conductors are the dass of ionic solids that exhibit very tugh ionic conductivity of the order 1.0 San-1 with n-ble electronic conductivity of
the order of 10.12 Scrn.1 at ambient as well--as hgh temperature [I]. The
conductivity of SIC mataials is based on the canier ion concentration of about
10" crn =, and hence. no thermally generated defccts i.e., hr = 0. Thus, its
conductivity is expressed as
Tablc 1.1. The d~tIerences between NICs and SICS
- - Ropem I I NICS I SICS I
I
Elecmn~c conduct~vlty , appreuable j n-ble
1
i , I just below the rneltmg / well below the melting 1 I High c o n d u m t y I I j pamt / ~ o m t I
Mobllc chargc carriers
and trmpaatun e&ct
low (1016 - 1018 an3] hgh ( l W cm9)
lughfy dependent / almost independent I I
k t i v a t h a w r ~ y
~=wb
I I ! high (hr and hm) 1 low (hm)
1-2 ev for NaCl 11, I1 / 0.1 ev for 111 i
From the early history of development of science, it is quite well known
that metals arc generally good conductors of electricity and most of the non-
metals are nonconductors or insulators. Later, it was rrahd that certain i
substances, such as common salt in aqueous solution, are good electrolytes of
electricity and other substances, like sugar in aqueous solution, are bad
rlrctrolytes. This led to the development of liquid electrolytes for ionic devices.
I fwever, it had various disadvantages Like
;I. limited Me time of the device because of the corrosion wactions occuning
txween the electrot).te solution and the electrodes
t). leakage of the Liquid electrolyte from the device due to corrosion
c. non-function of the devias below the freezing & above the b i h g points of the
electrolytes and
ti. dpcomprrition reaction occurring at high temperature of liquid electrolyte in
the device.
Iience, the unproved interest to search for newer materials that could
wemmc the above-mentioned disadvantages resulted in the dimvery of solid
electrolytes (St%). In 1833, Farday reported silver ion conductor in silver sulfide - 1301 and m 1899, Nemst found that the oxygen ion conductivity at lugh
temperatun in Yttria slatliked drcOnia ( 8 5 O . 0 Y203 - 1% ZrO?) 1311. In 1913,
Tubandt et al. have reported a h i electrical conductivity in qystalline a-AgI on
transformation of p to a phase at 422 K that has i n s p d large interest in the fieki
of solid-state electrochemistry, to synthesis a number of superionic conductors
126 - 291. During 1932, Tubandt reported a hlgh ionic conducting phase at high
ternperaturn in A&&. &Te, CuBr B Cul 1321. Later, in 1966 and 1%7, the
wpaiaslic muchng A&Sl and sodium p-alurninia wen reported as arhibiting
hi@ ion conductivity at ambient temperature 133, 341. Subsequently in 1967,
Bndly & Or&ne and Owens & Argue independently discovered the silver ion
CuCl
conduction in htAg.4~ (M - K*, Rb*, NH4') which was a break through in the
history of SICS (20a & b]. RbAg& synthesized based on the a - Agl model
consided as a good example of mattrial design, though the structures of these
two materials are not analogous. During the same period, Goodenough et al.
synthesized NASICON (Na,+xZrzPsxSi&,z], a - tailor made crystal in crystal
chemistry to understand the ionic conduction in thmdirnensional tunnel
suucture 1351. Since then, various kinds of newer solid state ion conducting
rlwterials have been synthesized and applied in the fabrication of suitable solid
state ionic devics.
1.4. CIASNFlCATIOll OF SUPERIONIC CONDUCTORS
Superionic Conducting materials are synthffized in large numbers and
c W 1 e d in diITercnt ways according to the properties possessed by the materials.
One of the ways is based on the microstructures and phases that can be broadly
dnided into four categories as follows
a) smp;k/po3rcrystalline, b] glasses/arnorphous, c) polymers and d) composites.
Superionic conducting crystalbe materials of ditferent cation (Ag, Cu*,
ti'. Na', H' etc] and anions (W . F-) conductors as charge canier ions have been
LiF - LhO - AI(POJ]~ were prepared and found to &bit conductivity in the order
of - l o 2 to 103 San at 573K 1102 - 1071. During the same mod, Takahashi
and Yarnamoto studied the glass forming ability in h 0 - LiCl - A l z a - systems and pursued for increase in Li' ion concentration by a c h k h g
conductivity of the order 103 Sun1 at 573K [1081. In 1982, large interrst on
lithiurn &loroborate glass oompounds has increased due its hlgh conductivity of
S m 1 at 573K Further, in the WOs - h 0 systun, the rapidly quenched
materials found to exhibit aonductivity of lo4 S m l at 4%K, W S - L b 0 , hO-
SiG sh6wcd 103 Sun at 673K 1109, I 101. Later, in order to enhance the
conductivity of Li' ion conducting glass, the mixed former effect was attempted
and found it suassful in obtaining the conductivity of 1W Sari.' at 473K [I l l ] .
In 1980, lithium ion inun-poratcd sulphide glasses w m synthesi?;ed and found to
be high ion conducting 11 121. In general, sulphide glasses exhibit more than
thnx ordm af conductivity compared m the c w m p o n d q oxide glasses. M e tt el. &orted that &S-m-U system to possss conductivity of 10S
S W ~ at 293K 11 131. The sulfide glasses such as, && - hs s h d the
CoaduaMV of pLnost 104 San.1 at 298K 11 141. Fg. 1.4 shows conductivity vs.
Ng. 1.4. lag a M W / T pbts of some of the tithiurn based SIC glassy systems
tanperRNrC pbts of some Uthium based SIC glassy systems and were also
summarized in table 1.6 198 - 1441.
Table 1.6 Exampks of lithium based superionic conducting glassy materials
Type of electrolyte
42.5 W - 57.5 &OJ 35LbCblOI&C12-3OLi$iQ- 12.5&0~12.5SiO1
63 ti& - 37 W3
3 1.8 Liz0 - 12.3 LiCl - 59.9 &03
29Li30-24LiCl-3AlZOj-44&03
22.1L1&12LiF- 15.8Li&0~-6Li~S04-28.5&r03
2 5 W - 1OLhPO~-65&03
2 0 M - 3 6 L i F - 4 4 & 0 3
4OLhO - 60 S i a
39Li30 - 13&03 - 48SiOa
7 1.5LiSds - 28.5Li&W4
W - L i S C k - S i O l
67LiKl~ - 3 3 M 0 4
65 Wb0~ " 35 SiOa
60 Lip03 - 40 LiF
30W-70LiPO:,
15W-7OLiF- 15AI(POs)s
33 LhS-66 RSs
37!&3-45Lil-l8RSs
50!&3-50W
50Li30-5ONthCb
soIM)-5o'hdh
Conductivity
!%-I)
6.1 x 1V3
9.7 x lo-=
6.3 x 10"
3 . 2 ~ 10"
1 . 5 ~ 1 0 ~
2.3 x
1.0 x lo-3
3 . 1 4 ~ lo4
1 x.106
4 . 3 ~ 1V9
1.9 x
1x105
3 . 5 ~ 1V3
1 x lo4
6 . 7 ~ 1V9
1x106
1.3 x 1W
1 . 1 ~ 1 0 4
1 x lW
4x10J
5 . 3 ~ 1W
6.4 x 1W
Temp
(K)
623
623
298
298
623
473
603
473
373
298
623
373
523
473
298
298
493
298
298
298
623
623
Ref.
98
98
115
115
108
106
116
106
117
118
119
112
120
121
122
122
123
124
125
112
110
126,127
The experimentally detamined conductivity is generally arm@& as a product of a oania concentration and mobility. In the crystalline solids, the
can* concentration may be defined as the density of defects with n femce to
the perfect latikc. Since each carrier is identically situated and has the same
mobility, the m t may be due to the ensemble of ions and it is a simple
multiple of thr single ion cumnt. In case of glasses, it is not simple to explain
because of the lack of r e f m c e of a ptdect lattice and the definition of defect
becomes rather arbitmy. Thus, various theories have been proposed to explain
the bhavior of the conductivity in glass samples 191.
Random sitc model was proposed based on the consideration that all ions
of a partjrular type arc potent$l carriers with a gaussian dstribution of activation
eneves 11451. The potcnual canier ion mobility varies with the distriiution of
activation entrgy and in turn varies with glass composition. In general, the
variation in canier concentration with glass composition is almost constant, and
hcncr the conductivity variation with glass composition is controlled only by a
change in ion mobility. This model was successful in explamng the composition
dependen& of conductivity in superionic conducting glassy materials.
Ravh.re and Souquet haw proposed a theory of the weak electrolyte model
to a@& the km transport in SICS glasss 1146, 1471. In this model, it is
oonskkrrd that e ~~ of the total ions are mbile that amtn'bute to the
amducthi, whik other ions are immobile and canicr ion mobility is Mependent
d cuq&km, in turn with smcture. The basis of the model is the
cornlation of the ionic conductivity (a] with the thermodynamic activity (a) and
their dependence is given by a = K lajll2 where K is a constant (148- 1491. The
ionic conductivity (a) and the activity (a) are correlated in silica based glasses
1146, 1471.
Minami et al. proposed diffusion path model on the bas15 that the
depmdmct of ion mobity, canier ion concentration on glass compositions and
the distribution of conducting species in diaerent anionic environment. In this
model, there exists two types of ion population in the glass matrix as, ion located
in the W e surmundjngs, wtuch are found to be mobile and hence contribute to
Fig 1.5. Pormtia] energy dmgram~ cxpbmhg the conduction
'p~#rssinthtSlCs
conduction, thwc associated with the oxy-anion that are immobile and paml
covalena exists between canier ion & non-bridging oxygen ions 1150j. The
repmentation of the potential energy by a schematic diegram is shown in 6g 1.5.
The wide ahabw potential is formed of alkali ion wiq halide ions & the n m ,
deep potcntial weil is of alkali ion interaction with oxide ions. Since glass is
considemi to be highly disordered structure, it also possesses random
distribution of activation energy among the potential well formed as shallow-
s h a h , shallowdecp and deep - deep. The increase in the alkali halide
conantration inotases the mobile ion concentration present in shallow wells,
and the shallow wells arc connected for long period with the formation of
favorable diffusion path for the ion transport and thereby increase conductivity
11511. Several experimental EXAFS results support the existence of mobile and
immobik silver ions in these types of glasses 1 152- 1561.
Anderson and Stuart proposed a model for the microscopic transport
mechanism of ions m glasss. Accordingly, ionic transport in glasses occurs by
means of difusive motion of ions betwen encrgchcally stable sites located in the
glass s t rum. The potential barrier W that the ion must surmount can be
apParacd aa the sum of two emxgia W= Eb + Es. Eb is an elcamstatic binding
magy, whb3 is coulomb turn associated with the removal of an ion h m its
mtavailing charged environment at one site to a position midway between two
neighboring Bftek i3 is strain energy encountered by the ion when it pses
thmugh the gateway lormed by the fully bonded bridging ion sites (157-1591. ?he
~ a n d S n L a r t m o d e l a I l o w s ~ t i o n o f t h e t w o c o n t n ' b u t i o n S t o t h e
activatkm in a number of silicate glasses. Letex waal sevsaltions have
been mede to the Andem and Stuart model and this plwides a usdul
~~ errptanatlan tw the ion transport phenommn in glasses.
The duster bypass model considers that glass can be mgarded as a
cmgclation (fmEen state] or ordered microdomains or clustns corn& by a
connective tissue. On cooling the glass below T, the &ual liquid that inihally
sumunds the dusters solidifits to give a highly disordered phase refemd to as
connective tissues 1160, 1611. The essence of cluster bypass model, as suggested
ty Ingram et al, is that the preferred pathways for ion migration lie outside the
cluster but lie within the connective tissue. This model, besides explauung the
ton transport in various glasses, prwides a simple explanation for the mixed
alkali effect in ion conducting glasses.
Elliott & Owens proposed diffusion controlled relaxation model for the
r n i m p i c tmnspo~ mechanism of ions in glasses. The DCR model assumes
the applicability of an Andason-Stuart model i.e., in addition to the ekctmstatic
binding en= and strain energy contributions to the activation energy for
diffU.sion motion of ions, a contri%ution to the p h k a t i o n (a.c.) conductivity is
attniuted to the double occupancy effect [162, 1631. In thii modei, it is
wndlerrd that there are energctxdy stable sites (NBOs) for the mobile ions to
rrsidc in the oxide gla~sy mcture . Ionic transport occurs by means of diffusive
motion between the equivalent sites resulting in the primmy rrlaxational event
with a characteristic miaoscopic relaxation time r. Howeva, when a cation hops
into one of the vidnal equivalent sites will result in the a'cation of double
occupancy Imorm m intasti- effm on instantaneous d a x a t h process. 'l'hia modd qWns the bquency dependence of conductivity that what the c i r t r u ~ ~ I . e i a x a t i o n i s t q ~ e c t h v n s p o n s i b l e f o r ( ~ ) ~ n l a x a t i o n . Tbc
ch- Brom nonaponcntial (KWW) behavior to DebycWP (sin@
behavior both for very small ion concentrations and at high temperatures, can be
readily explained by this model.
Bunde et al. proposed a dynamic structure model based on the m t i o n of
fluctuating pathways, within a dynamically determined structure, for ion
transport in glasses (164, 1651. The main features of this model are i) the ion
transport is a hopping process ii) the mobile ions are acfive in creating glass
structure and iQ the glass structure continues to change at a local level even far
below TI. Based on the assumption that the process of ion hopping with bxaJ
stnrctural rrlaxation OCCUIS together, the ion mwcs on from its own site to the
naghboring site than the m m r y of the cation mts and so a empty site
remains there for sometime, I. Such empty site constitutes the p r e f d
pathways for ion migration and the effect caused is known as site memory effect
The succfirr of this model includes the dixovery of a simple power law (a- cv)
opaative in sin& cation glasses and the quantitafjve eluadation of many
featurn of the mixed alkali effect. The main feature of this model is that the glass
structure must p i s t for time scales much longer than the inverse hopping
ram of individual cations.
Funkc, Ingram and Bunde have developed a unikd site relaxation model
baaed on the jump relaxation model and dynamic saucture model (1661. The
beaic idea ol the model is that the ion transport is stmngly influenced by their
mutually rrplbive , c o u h b intaactions. As a amscquena, their hopping