Transport characteristics of ZrO2 dispersed mixed

(BaCl2)1�x–(KCl)x solid electrolytes

Archana Guptaa, Anjan Silb,*

aSchool of Physics and Materials Science, Thapar Institute of Engineering and Technology,

Patiala 147004, Punjab, IndiabDepartment of Metallurgical and Materials Engineering, Indian Institute of Technology, Roorkee 247667, India

Received 7 August 2003; received in revised form 17 September 2004; accepted 22 September 2004

Abstract

Composite solid electrolytes in the system [(BaCl2)1�x:(KCl)x]1�y:(ZrO2)y were prepared following the

conventional ceramic powder processing route. In the mixed matrix system prepared by melt quench technique,

a nominal increase in conductivity (s) was found in (BaCl2)0.9:(KCl)0.1. On ZrO2 particle dispersion in this mixed

matrix, the maximum conductivity (�90 times that of base matrix value) was found to occur with 50 m/o of ZrO2.

Conductivity increases monotonically over the temperature range from 100 to 300 8C studied and attains the value

of 10 � 10�6 S cm�1 at 300 8C. The mobility (m) of the charge carriers at room temperature was found to be

18.5 � 10�2 cm2 V�1 s�1 and the increase in m with temperature was not very significant. The transference ionic

number determination showed that the electrical conductivity of the electrolyte is predominantly due to ions. This

study indicates that the conductivity is governed by mobile ion concentration.

# 2004 Elsevier Ltd. All rights reserved.

Keywords: D. Ionic conductivity; Solid electrolyte; C. Impedance spectroscopy; Transient ionic current; Ionic transference

number

1. Introduction

The conductivity of many ionic solids is increased by several orders of magnitude when the materials

are prepared as composite with a finely divided second phase. In most of the cases, the second phase is an

www.elsevier.com/locate/matresbu

Materials Research Bulletin 40 (2005) 67–77

* Corresponding author. Fax: +91 1332 285243/273560.

E-mail addresses: asil1fmt@iitr.ernet.in, anj_sil@yahoo.co.uk, anj_60@rediffmail.com (A. Sil).

0025-5408/$ – see front matter # 2004 Elsevier Ltd. All rights reserved.

doi:10.1016/j.materresbull.2004.09.011

insulating material such as Al2O3, SiO2 or Fe2O3 which are chemically inert and nearly insoluble in the

host matrix material such as LiI, AgI, CuCl, etc. The high ionic conductivity makes these materials as

good candidates for solid electrolytes used in devices such as batteries, sensors and fuel cells, etc., where

ion transport through the electrolyte can be current-limiting process [1–3].

Though the conductivity enhancement in the two phase system was known for about 80 years, the

systematic investigations started only after 1973, when Liang found that Li+ ion conductivity in LiI was

enhanced by Al2O3 dispersion without any reaction products or solid solution [4]. Since the time Liang

reported, this effect was observed in numerous systems [5–9]. Considerable efforts were made to develop

the understanding of the conductivity enhancement [10–12]. It is widely accepted that the effect is a

result of an enhanced conductivity path along the interface between the conducting matrix and the

dispersed particles due to the formation of space charge layers [12,13]. Besides the oxide dispersions, the

ionic conductivity (s) can also be altered by replacing the single halide matrix with the mixed matrix in

which the factors (i) the aliovalent ionic substitution in the regular lattice sites [14–16], (ii) distortion of

the regular lattice structure caused by substitution of ions with wrong size ion [17,18] effect the senhancement. The s enhancement as a result of the first factor is generally limited to 50–100 times,

however, this study is exceedingly important in elucidating the thermodynamics and kinetic properties of

point defects.

It can be evidenced from the literature that a considerable amount of work has been carried out on Ag+

ion conducting composite electrolytes [19–22], whereas not much attention has been paid to alkali halide

based composite electrolytes with the exception of Li+ conductors [23]. The primary objective for

choosing KCl as the host matrix is that it does not undergo phase transformation before melting. Thus, it

provides a scope for studying over a wide range of temperatures. The matrix (mixed halide) modification

was considered in view of the size difference as well as the valency difference of the cations K+ and Ba2+.

In the present study, the mixed matrix in the system (BaCl2)1�x:(KCl)x was undertaken to determine the

composition for which the conductivity is maximum. The composite electrolytes were made by

dispersing the mixed matrix of maximum conducting composition with ZrO2 in varying composition.

The crystallographic structures of KCl and BaCl2 are rock salt and fluorite, respectively. Therefore, the

predominating bulk defects are of Schottky and Frenkel types in KCl and BaCl2 crystals, respectively. In

BaCl2, Ba2+ ions form FCC structure and the tetrahedral voids are occupied by Cl� ions. The structure of

KCl is cubic with the anions arranged in cubic close packing with all the interstitial octahedral sites

occupied by the cations. On adding KCl into BaCl2, the following defect reaction is possible:

2KCl )BaCl2

K0Ba þ K

�

i þ 2Cl�Cl

On the other hand, if the BaCl2 addition is made into KCl the possible defect reactions satisfying the

preservation of the regular site ratios of the host crystal, would be as follows:

BaCl2 )2KCl

Ba�

K þ V0K þ 2Cl�Cl

and

BaCl2 )KCl

Ba�

K þ Cl0i þ Cl�Cl

In view of the fact that K+ ion is smaller (smaller than Ba2+ ion size) the replacement of Ba2+ site by K+

would be relatively easy provided other thermodynamics and kinetic properties of point defects are

satisfied.

A. Gupta, A. Sil / Materials Research Bulletin 40 (2005) 67–7768

Therefore, an appreciable effect on defect mobility as reflected by the above defect reaction will be the

cause of the conductivity enhancement in the BaCl2/KCl mixed matrix.

As the defects in KCl and BaCl2 are of different types (Schottky type in KCl and Frenkel type in

BaCl2), a difference in diffusivities (be they ionic or vacancy) will result at the interfaces BaCl2/KCl. For

BaCl2, KCl or the mixed matrix BaCl2–KCl in contact with ZrO2 which being oxide; is normally

nucleophilic, the characteristic defect inducing process responsible for the increased vacancy concen-

tration as compared to the bulk can be written as

VA þ KK ,K�

A þ V0K

VA þ BaBa ,Ba��

A þ V00Ba

and

VA þ K0Ba þ K

�

i ,K�

A þ K0Ba

respectively. VA denotes a free active surface site which is responsible for the K+/Ba2+ cation adsorption.

V0K and V00

Ba are the potassium and barium vacancies formed in the space charge region. While the

adsorption process from the halide components KCl and BaCl2 separately creates potassium and barium

vacancies, respectively in the space charge region; the adsorption process from the mixed matrix will

make the oxide/halide interface positive and as a result K0Ba may get bound near the interface in the space

charge region.

2. Experimental

Commercially available chemicals KCl (Purity > 99%, E. Merck), BaCl2 (s.d. fine) and ZrO2

(Purity > 99%, C.S. Zircon) of particle size of about 1 mm were used as the raw materials without

further purification. The homogeneous mixtures of (BaCl2)1�x:(KCl)x (x in m/o) were melted in the

temperature range of 700–1000 8C. The melt was then quenched to the room temperature. The materials

thus obtained were ground using mortar and pestle and then pressed at a pressure level of 200 KN to

prepare the samples in the form of pellets having diameter of 2 cm and thickness of 3 mm. The pellets

were then sintered in a muffle furnace in the temperature range of 500–700 8C for the soaking time period

of 5 h. For conductivity measurements, the flat surfaces of the samples were electroded with silver paste.

The bulk resistances (dc) of the samples were determined from their complex impedance plots of the data

obtained at the room temperature with the help of LCR meter (HP, Model 4274A) in the frequency range

100 Hz to 100 KHz. Hence the dc conductivity values were estimated from the low frequency side

interceptions of the semicircular arcs on the real axis. The dc conductivity values were plotted as a

function of the composition to determine the maximum conducting composition of the matrix. In this

matrix, the ZrO2 dispersion was made and the conductivity behaviour of the dispersed electrolytes was

studied as a function of ZrO2 content. The sample for which the maximum conductivity was obtained was

characterized by SEM (LEO 435VP) and XRD. The XRD patterns were taken using X-ray diffractometer

(Rigaku, Model D max III C) employing Cu Ka radiation.

The ionic mobilities (m) and the ionic transference number (ti) were determined for the maxi-

mum conducting composite electrolyte using transient ionic current (TIC) technique [24] and

Hebb-Wagner dc polarization method [25], respectively. The schematic diagram of the electrical circuit

A. Gupta, A. Sil / Materials Research Bulletin 40 (2005) 67–77 69

used for the TIC measurement is shown in Fig. 6. A fixed potential V of 1.77 V was applied across the

sample having thickness d of 3.14 mm through graphite electrodes. Here the graphite electrode is treated

as a blocking electrode. The sample was polarized for a sufficiently long time (�2 min) so that the

complete polarization takes place. The polarity of the electric field was then reversed. The time

dependent TIC response was monitored using an X–Y–t recorder. Accordingly, the response time (t)

at which the current shoots up was noted and hence the ionic mobility (m) was estimated using the

relation:

m ¼ d2

tVðcm2 V�1 s�1Þ

For the polarization method a fixed potential (V � 1.77 V), with the polarity shown in Fig. 7 was

applied across the sample pellet. The sample was electroded with reversible (silver) on one side and

irreversible (graphite) electrode on the other side. The current iT (in arbitrary units) through the sample

was monitored with time using X–Y–t recorder.

In order to investigate the temperature dependent transport characteristics of the material, the variation

of s andmwith temperature were studied. Every measurement point was obtained by allowing the sample

to attain the thermal equilibrium.

3. Results and discussion

3.1. XRD analyses of the samples

The XRD analyses of sintered KCl, BaCl2, mixed matrix (BaCl2)0.9:(KCl)0.1 and composite electro-

lyte [(BaCl2)0.9:(KCl)0.1]0.5:[ZrO2]0.5 are shown in Fig. 1. It is evident from Fig. 1c that for the

mixed matrix, the major peaks belong to either BaCl2 or KCl phase. There is no strong evidence of

compound formation between BaCl2 and KCl. When compared with the patterns of KCl and BaCl2 given

in Fig. 1a and b, respectively, it can be seen that most of the peaks of mixed matrix remain in their

position. However, a slight shift was observed in case of a few peaks. In the pattern given in Fig. 1d, the

presence of ZrO2 was found while BaCl2 and KCl peaks are found to remain in their position with

intensity reduction as compared to those seen in the mixed matrix pattern (Fig. 1c). However, in the

composite electrolyte, a few unidentified peaks were also observed, but could not be matched with data in

the JCPDS file.

3.2. Electrical conductivity of matrix

Fig. 2 shows the variation of conductivity (s) versus composition (x) of the mixed matrix

(BaCl2)1�x:(KCl)x (0 � x � 0.5) at room temperature. From Fig. 2, it can be seen that for

(BaCl2)0.9:(KCl)0.1 composition s becomes maximum, however, the increase is nominal. The following

gives an explanation of the composition dependent conductivity behaviour and the arise of maxima. On

substituting the host lattice ion by guest ion, the lattice distortion takes place, which is characterized by

the factor j1 � rh/rgj where rh and rg are the ionic radii of the host and guest ions, respectively [26]. In the

present case, the magnitude of this factor (�0.014) is negligibly small. Therefore, the effect of lattice

distortion towards the conductivity enhancement is insignificant. The increase in conductivity of the

A. Gupta, A. Sil / Materials Research Bulletin 40 (2005) 67–7770

A. Gupta, A. Sil / Materials Research Bulletin 40 (2005) 67–77 71

Fig. 1. XRD patterns of (a) KCl; (b) BaCl2; (c) (BaCl2)0.9:(KCl)0.1 and (d) [(BaCl2)0.9:(KCl)0.1]0.5:[ZrO2]0.5.

mixed matrix with x was observed to reach a maximum at about x = 0.1 and then it decreased (Fig. 2). The

enhancement in conductivity could be attributed to the net charge transfer across the interface of the

biphasic mixture.

3.3. Electrical conductivity of composite electrolyte

The conductivities (s) of the samples in the system [(BaCl2)0.9:(KCl)0.1]1�y:(ZrO2)y were determined

from their complex impedance plots. The s values were plotted as a function of composition y in Fig. 3.

From the plot, it can be seen that a significant conductivity enhancement (almost two orders of

magnitude) occurs at about 50 m/o ZrO2. This enhancement is attributed to the maximum percolation

pathways at 50 m/o ZrO2. The pathways are developed due to the formation of space charge layer at the

interface between matrix and dispersoid. The indication of such interface formation was found by

microstructural investigation of the sample.

3.4. Conductivity as a function of temperature

As the ionic conductivity is a very sensitive function of temperature, the temperature dependent

conductivity was also studied. The study was carried out for [(BaCl2)0.9:(KCl)0.1]0.5:[ZrO2]0.5 samples in

the temperature range 100–300 8C. The conductivities were plotted as a function of temperature in the

form of log s versus 1/T in Fig. 4. From the plot, it can be seen that the s increases from 10�8 S cm�1 (at

100 8C) to about 10�5 S cm�1 (at 270 8C) yielding the activation energy for ion transport Q = 0.34 eV.

When compared with other reported electrolyte systems, this Q value is found to be comparable with that

of PbX2–Al2O3 (where X = Cl, Br, I), KCl–ZrO2 and AgX based systems [21,27,28].

A. Gupta, A. Sil / Materials Research Bulletin 40 (2005) 67–7772

Fig. 2. The room temperature conductivity of mixed matrix [(BaCl2)1�x:(KCl)x] as a function of x (0.0 � x � 0.5).

A. Gupta, A. Sil / Materials Research Bulletin 40 (2005) 67–77 73

Fig. 3. Variation of the room temperature conductivity of composite electrolyte [(BaCl2)0.9:(KCl)0.1]1�y:[ZrO2]y as a function of

y (0.0 � y � 0.55).

Fig. 4. Variation of the conductivity with temperature in [(BaCl2)0.9:(KCl)0.1]0.5:[ZrO2]0.5 sample.

A. Gupta, A. Sil / Materials Research Bulletin 40 (2005) 67–7774

Fig. 5. SEM micrograph of [(BaCl2)0.9:(KCl)0.1]0.5:[ZrO2]0.5 sintered sample.

Fig. 6. Typical TIC plot of the composite electrolyte [(BaCl2)0.9:(KCl)0.1]0.5:[ZrO2]0.5 at 250 8C. Inset: schematic experimental

circuit for TIC measurement.

3.5. SEM micrograph

The microstructural investigation was carried out for the composite electrolyte [(BaCl2)0.9:

(KCl)0.1]0.5:[ZrO2]0.5 sample. The micrograph is shown in Fig. 5. From the micrograph, the evidence

of grain formation due to sintering is clear. The micrograph suggests that the system is a composite,

wherein ZrO2 particles are dispersed in the halide matrix grains.

As can be seen from the micrograph that the interconnecting interphase regions are quite significant.

Thus, the microstructural evidence of the sample supports well the development of higher percolation

pathways in it.

3.6. Ionic mobility determination

In order to estimate the conductivity contribution due to the ionic mobility factor as expressed

by s(T) = n(T)qm(T), where n and q are the charge carrier concentration and the charge of the mobile

ion, respectively, the ionic mobilities of the composite electrolyte [(BaCl2)0.9:(KCl)0.1]0.5:[ZrO2]0.5

were determined at different temperatures. Some of the mobility values, e.g. 0.185, 0.230, 0.328

and 0.383 cm2 V�1 s�1 determined at 25, 100, 250 and 300 8C, respectively are given here as

typical data. A transient ionic current profile at a typical temperature of 250 8C is presented in Fig.

6. From the m values, it can be seen that the mobility at 300 8C becomes just two times the room

temperature value. Therefore, the conductivity variation with temperature is a combined effect of the

mobility and the carrier concentration enhancements. The mobility contribution is limited as observed in

this system.

A. Gupta, A. Sil / Materials Research Bulletin 40 (2005) 67–77 75

Fig. 7. Time dependent current response of the composite electrolyte [(BaCl2)0.9:(KCl)0.1]0.5:[ZrO2]0.5. Inset: schematic

experimental circuit diagram for the transference ion number study.

3.7. Ionic transference number determination

The performance of solid electrolytes for their application in electrochemical devices (sensors),

energy storage devices (batteries), etc. depends upon the ionic contribution of the conductivity.

Therefore, the desirable feature for these electrolytes is the conductivity due to the ions only with a

negligible contribution from the electrons. To ascertain the extent of the ionic and electronic contribu-

tions to the total conductivity, a measurement of the ionic transference number (ti) was made. For systems

with sole charge carriers as ions, ti is equal to unity. A profile showing the time dependent ionic current at

room temperature for the composition [(BaCl2)0.9:(KCl)0.1]0.5:[ZrO2]0.5 is shown in Fig. 7. The current

value decreased to a negligibly small value. Therefore, it can be concluded that the contribution to

conductivity is predominantly due to the ions and from the plot the transference number has been

estimated as ti � 0.964 at room temperature.

4. Conclusions

In an attempt to prepare an oxide dispersed composite solid electrolyte on a mixed matrix base, the

mixed matrix (BaCl2–KCl) was prepared and the composition (BaCl2)0.9:(KCl)0.1 resulted maximum

conductivity. An oxide (ZrO2) dispersion of 50 m/o in this mixed matrix increased the conductivity (�2

orders of magnitude). About 40 times enhancement in s was observed at 300 8C. The proportionate

increase in m was not found under the same temperature rise, which leads one to conclude that the

enhancement is predominantly due to the charge carrier concentration.

References

[1] T. Minami, J. Non-Cryst. Solids 56 (1983) 15.

[2] C.A. Angell, Solid State Ionics 18/19 (1986) 72.

[3] G. Rog, A. Kielski, A. Kozlowska-Rog, M. Bucko, Ceram. Int. 24 (1998) 91.

[4] C.C. Liang, J. Electrochem. Soc. 120 (1973) 1289.

[5] T. Jow, J.B. Wagner Jr., J. Electrochem. Soc. 126 (1979) 1963.

[6] K. Shahi, J.B. Wagner Jr., J. Electrochem. Soc. 128 (1981) 6.

[7] S. Fujitsu, K. Koumoto, H. Yanagida, Solid State Ionics 18/19 (1986) 1146.

[8] M.C.R. Shastry, K.J. Rao, Solid State Ionics 51 (1992) 311.

[9] G. Rog, M.M. Bucko, A. Kielski, A. Kozlowska-Rog, Ceram. Int. 25 (1999) 623.

[10] A. Bunde, W. Dieterich, E. Roman, Phys. Rev. Lett. 55 (1985) 5.

[11] R. Blender, W. Dieterich, J. Phys. C 20 (1987) 6113.

[12] N.F. Uvarov, V.P. Isupov, V. Sharma, A.K. Shukla, Solid State Ionics 51 (1992) 41.

[13] U. Lauer, J. Maier, Solid State Ionics 53–56 (1992) 885.

[14] H.H. Hoffer, W. Eysel, U.V. Alpen, J. Solid State Chem. 36 (1981) 365.

[15] S. Hull, P. Berastegui, J. Phys.: Condens. Matter 11 (1999) 5257.

[16] S. Hull, D.A. Keen, P. Berastegui, Solid State Ionics 147 (2002) 97.

[17] P. Manoravi, K. Shahi, Solid State Ionics 58 (1992) 243.

[18] K. Singh, S.D. Wachasunder, S.S. Bhoga, Bull. Mater. Sci. 18 (1995) 147.

[19] R.K. Gupta, R.C. Aggarwal, Solid State Ionics 72 (1994) 314.

[20] J.S. Lee, S. Adams, J. Maier, J. Electrochem. Soc. 147 (2000) 2407.

[21] R.C. Agrawal, R.K. Gupta, Bull. Mater. Sci. 19 (1996) 573.

A. Gupta, A. Sil / Materials Research Bulletin 40 (2005) 67–7776

[22] J.S. Lee, S. Adams, J. Maier, J. Phys. Chem. Solids 61 (2000) 1607.

[23] D. Lubben, F.A. Modine, J. Appl. Phys. 80 (1996) 5150.

[24] S. Chandra, S.K. Tolpadi, S.A. Hashmi, Solid State Ionics 28/30 (1988) 651.

[25] S. Chandra, Superionic Solids—Principles and Applications, 1981, p. 141.

[26] K. Shahi, J.B. Wagner Jr., J. Phys. Chem. Solids 43 (1982) 713.

[27] A. Kumar, K. Shahi, J. Mater. Sci. 30 (1995) 4407.

[28] A. Mangla, A. Sil, Ind. J. Pure Appl. Phys. 39 (2001) 267.

A. Gupta, A. Sil / Materials Research Bulletin 40 (2005) 67–77 77