This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Electrical transport and crystallization in Cu+ ion substituted AgI–Ag2O–V2O5 glassysuperionic system

Neha Gupta a, Anshuman Dalvi a,⁎, A.M. Awasthi b, S. Bhardwaj b

a Physics Group, Birla Institute of Technology and Science Pilani (RJ), 333031, Indiab UGC-DAE-Consortium for Scientific Research, Indore (MP), 452017, India

a b s t r a c ta r t i c l e i n f o

Article history:Received 26 May 2009Received in revised form 18 September 2009Accepted 9 November 2009

Keywords:Superionic glassesCrystallizationSilver ion conductorsFast ion conductors

A new glassy solid electrolyte system CuxAg1−xI–Ag2O–V2O5 has been synthesized. The structural, thermal andelectrical properties of the samples have been investigated. The glassy nature of the samples is confirmed by X-Ray diffraction and Differential Scanning Calorimetry studies. The electrical conductivity of these samplesincreases with CuI content and approaches a maximum value of ∼10−2 Ω−1 cm−1 for x=0.35 at roomtemperature. Ionic mobility measurements suggest that enhancement in the conductivity with Cu+ ionsubstitution may be attributed to increase in the mobility of Ag+ ions. The electrical conductivity versustemperature cycles carried out at well-controlled heating rate above Tg and Tc reveal interesting thermalproperties. For lower CuI content samples conductivity exhibits anomalous rise above Tg and subsequent fall at Tc.It is also found that CuI addition into AgI–Ag2O–V2O5 matrix reduces the extent of crystallization.

© 2009 Elsevier B.V. All rights reserved.

1. Introduction

Fast ion conducting glasses have been subject of interest for the lasttwo decades because of their possible applications to electrochemicaldevices. As a result, Ag+, Na+ and Li+ ion based glasses have beendeveloped, characterized and widely used in ionic devices [1]. Inparticular, AgI–oxysalt glasses are known to show high ionic conduc-tivity up to ∼10−2 Ω−1 cm−1 at room temperature [2–6]. The effect ofthe presence of two mixed glass formers (mixed former effect) [7] andmixed alkalis (mixed alkali effect) [8,9] on electrical conductivity hasalso been studied for various glassy systems. Apart from mixed alkalisystems, the effect of the presence of two different cations (e.g. Ag+ andCu+) has also been investigated [10]. Interestingly, addition of CuIprovides AgI in the system by an exchange reaction according toPearson's theory [11]. Moreover, due to low cost, non hygroscopicnature and similar chemical properties to that of AgI, CuI has been usedby variousworkers as a possible dopant in silver oxysaltmatrix [12–15].Similarly, combination of different dopants like, PbI2, NaI and CdI2 in thesilver oxysalt matrix has also been developed [16–18].

Thus, there are studies on effect of CuI addition on AgI–oxysaltsystem with emphasis on development of electrolytes for button typecells [19], possible glass forming ability and fundamental structuralaspects [13]. No studies in our knowledge are available on effect of Cu+

ion substitution into AgI–oxysalt matrix on thermal properties, viz.,crystallization, structural relaxation at Tg and the process of phasetransformation. This work is an attempt to investigate important

thermal events in Cu+ ion substituted AgI–Ag2O–V2O5 glass usingelectrical conductivity versus temperature cycles, as a tool, preciselyabove Tg and Tc. Glass system having composition 50(CuxAg1−x)I–33.33Ag2O–16.67 V2O5 with x=0–0.4 has been chosen for the presentstudy. Thus, the preparation, structural, thermal and electrical proper-ties of the chosen system is discussed.

2. Experimental

The compositions 50(CuxAg1− x)I–33.33Ag2O–16.67 V2O5 forx=0.1, 0.15, 0.20, 0.25, 0.30 and 0.40 (corresponding to 5, 7.5, 10,12.5, 15 and 20 m/o of CuI in AgI–Ag2O–V2O5 system, respectively)were chosen and abbreviated as 10, 15, 20, 25, 30 and 40CI-SISOVO,respectively, for the present investigation. Using the high purity rawmaterials AgI, CuI, Ag2O and V2O5, the samples were prepared byconventional melt quenching. Powder X-ray diffraction of the sampleswas carried out with CuKα radiation (1.54 Å) using RIGAKU X-Raydiffractometer. The amount of the sample used for X-ray diffraction iscarefully kept same for all the compounds for a comparative study.The amorphous/glassy nature of the samples was also confirmed bydifferential scanning calorimetry (Shimadzu, DSC-60) measurements.The electrical conductivity measurements (25–175 °C) were carriedout at a controlled heating rate of 1 °C/min using computer controlledHIOKI LCR meter model 3532-50 and Libratherm programmable PIDtemperature controller. Ionic mobility has also been measured usingtransient ionic current (TIC) technique [20,21]. The glassy sample inthe shape of a cylindrical pellet is sandwitched between blockingplatinum electrodes and the current is recorded as a function of timeusing computer interfaced Rishcom 15S electronic digital multimeter.To confirm the pure ionic nature, Ag/I2 button-type electrochemical

Solid State Ionics 180 (2010) 1607–1612

⁎ Corresponding author. Tel.: +91 1596 515426; fax: +91 1596 244183.E-mail address: anshumandalvi@gmail.com (A. Dalvi).

0167-2738/$ – see front matter © 2009 Elsevier B.V. All rights reserved.doi:10.1016/j.ssi.2009.11.007

Contents lists available at ScienceDirect

Solid State Ionics

j ourna l homepage: www.e lsev ie r.com/ locate /ss i

Author's personal copy

cells were prepared using these glassy samples as electrolytes, withAg metal as anode and I2 (in tetra methyl ammonium iodide andgraphite) as cathode as described elsewhere [22].

3. Results and discussion

3.1. X-ray diffraction

The X-ray diffraction (XRD) patterns for melt-quenched as well asfor the 10 and 30CI-SISOVO samples annealed after melt quenchingare shown in Fig. 1. Fig. 1a and d shows the patterns of as preparedglassy samples. Absence of any significant peak apparently confirmsthe glassy/amorphous nature of these samples. Further, the glassysamples are annealed above crystallization temperatures to investi-gate the precipitated compounds. The XRD results on glassy samplesannealed at 140 °C (Fig. 1b and e) and 170 °C (Fig. 1c and f) are alsoshown. The annealing of 10CI-SISOVO sample at 140 °C results intomajor precipitation of two compounds, viz., Ag4V2O7 and Ag8I4V2O7,as evident in Fig. 1b. Moreover, AgI also precipitates in a very smallamount. Interestingly, when the 10CI-SISOVO sample is furtherannealed at 170 °C (Fig. 1c), the intensity of the peaks correspondingto Ag4V2O7 and AgI remain unchanged and that correspond toAg8I4V2O7 appears to grow slightly. These results suggest that majorprecipitation of Ag4V2O7 and AgI occurs at lower temperatures(∼140 °C). Nevertheless, crystallization of Ag8I4V2O7 also begins atlower temperatures but subsequently completes at relatively highertemperatures. The XRD patterns of annealed glassy samples having CuIin relatively large amount do reveal interesting results. When the 30CI-SISOVO is annealedat140 °C (Fig. 1e), again thereappearminorpeaks ofAgI and Ag4V2O7 in the XRD patterns along with those of the Ag8I4V2O7

major peaks. Apparently, the height of all the peaks for annealed 30CI-SISOVO sample is significantly smaller than those of the peakscorrespond to 10CI-SISOVO samples (Fig. 1b). When the 30CI-SISOVOis further annealed at 170 °C, there is a slight increase in the height ofpeaks corresponding to Ag8I4V2O7. Further, there is negligible change inthe height of peaks corresponding to Ag4V2O7 and AgI.

It may be suggested here that annealing of 10CI-SISOVO exhibitsmajor precipitation of Ag4V2O7 and AgI at low temperatures and thatof the Ag8I4V2O7 compound whose precipitation also begins at low

temperatures but is progressive and subsequently ends at somewhathigher temperature. In case of the 30CI-SISOVO sample, the XRDresults suggest that on annealing, the overall precipitation ofcompounds is relatively small. Height of peaks corresponding to AgIis also found to be comparable in 10 and 30CI-SISOVO samples.

Thus it may be concluded here that (i) in both the samplescrystallization occurswith precipitation ofmultiple phases, (ii) additionof CuI reduced the formation of the two compounds (viz., Ag4V2O7 andAg8I4V2O7) and (iii) silver iodide precipitates in both the compositions.X-ray diffraction results for the other compositions do confirm glassynature of the samples. It is observed that the AgI–Ag2O–V2O5 systemdissolves CuI at least up to 17.5 m/o (corresponding to x=0.35, viz,35CI-SISOVO).

3.2. Differential scanning calorimetry

The results of DSC are shown in Fig. 2 for 10, 15 and 30CI-SISOVOsamples for a typical heating rate of 10 °C/min. The second heatingcycles for the samples (i.e. those of preannealed) are also shown.

In the DSC scans for all the as prepared glassy samples (Fig. 2a, cand e), there appear an endothermic smooth dip in 60–75 °C rangethat correspond to respective glass transition temperature (Tg) ofthese samples. Apparently, all the samples are glassy in nature. TheTg shows a decrease with CuI content. Fig. 2a shows the pattern for10CI-SISOVO melt-quenched sample. For this, there appears a broadcrystallization peak with exothermic rise (Tc) at ∼113 °C, that maybe visualized as two merged up exothermic peaks having maximumat ∼128 (Tp1) and ∼148 °C (Tp2).

For the 15CI-SISOVO sample (Fig. 2c), similar merged up peaksappear to begin at ∼107 °C and have first crystallization peak Tp1 at∼135 °C and Tp2 at ∼154 °C.

The DSC scans for 30CI-SISOVO are shown in Fig. 2e and f. Here thepatterns are found to be significantly different from those of the 10 and15CI-SISOVO samples. In the first cycle, first major exothermic peakappears to begin at∼107 °C and ends at the∼120 °C followed by a smallexothermic peak at ∼140 °C. On further heating, a small endothermicdip appears at ∼150 °C that may correspond to characterization β→αphase transition of AgI.

Fig. 1. X-Ray Diffraction patterns for as prepared glasses (a and d), glassy samplesannealed at 140 °C (b and e), and at 170 °C (c and f). Symbols denote (○) Ag8I4V2O7 (●)AgI (▲) Ag4V2O7 peaks.

Fig. 2. DSC scans for different compositions of CI-SISOVO (Ist) glassy and (IInd)annealed-glassy samples.

1608 N. Gupta et al. / Solid State Ionics 180 (2010) 1607–1612

Author's personal copy

In the second heating cycle no significant thermal event is seen incase of 10CI-SISOVO, whereas, for 15 and 30CI-SISOVO a smallendothermic dip appears at ∼147 °C exhibiting the presence of AgIthat may have precipitated during the crystallization.

The DSC results suggest that there are at least three compounds thatprecipitate at Tp1 andTp2 for lowCuI content samples. The crystallizationof these compounds may be simultaneous, however their respectivemaximum rates of crystallization (appear at Tp1 and Tp2, respectively)are significantly different. For low CuI content samples, Tp1 and Tp2maycorrespond to precipitation of Ag4V2O7+AgI and Ag8I4V2O7, respec-tively. Further, it is important to note that the higher CuI contentsamples (e.g. 30CI-SISOVO) again exhibit major precipitation of at leasttwo compounds with maximum crystallization rate at Tp1 and Tp2,respectively. In this case at Tp1, there is major precipitation of AgI alongwith minor precipitation of Ag4V2O7. Further, at Tp2 precipitation ofAg8I4V2O7 is likely.

Apparently, the XRD and DSC results are in good agreement witheach other.

For the further understandingof thermal events in the samples,withinthe glass forming region, the parameters defining the thermal stability ofglassy system are obtained from DSC plots and given in Table 1. The totalenthalpy release (ΔH) in the samples during crystallization is calculatedfrom the total area under the exothermic peak. The value of ΔH found tobedecreasingwithCuI content clearly suggests that CuI substitution in theSISOVO matrix suppresses the crystallization. Similarly, Tc, Tg, Tc −Tg andTg/Tm also consistently decreases with CuI content. Apparently thethermal stability decreases with CuI content.

It is shown by various workers that annealing of 50AgI–33.33Ag2O–16.67V2O5 (50SISOVO) above Tc results into precipitationof Ag8I4V2O7 compound [23,24]. Our results suggest that theprecipitation of this compound reduces with CuI addition. It is likelythat the added CuI reacts with Ag2O introducing Cu2O in the glassmatrix. Thus the Ag2O deficiency in the matrix may reduce theAg8I4V2O7 and Ag4V2O7 formation in the glass.

The above discussion further suggests that for low CuI contentsamples there may be at least three types of Ag+ ions surrounded bydifferent environments. The first type of Ag+ ions may be surroundedby V–O units, whereas, the second type of Ag+ ions surrounded by V–O along with I− units. There is also a possibility of third type of Ag+

ions may be surrounded by I− ions alone. Such surrounding of Ag+

ions may be responsible for precipitation of three compounds, viz.,Ag4V2O7, AgI and Ag8I4V2O7 during crystallization.

3.3. Electrical conductivity

At the outset, the ionic nature of all the samples is confirmed bygalvanic cell method using the following relation

tþ =VObserved

VCalculatedð1Þ

Using Eq. (1) the ionic transport number is found to be near unity forall the samples. In view of structural and thermal studies, whichrevealed precipitation of various compounds during crystallization, theelectrical conductivity is also studied above the Tg and Tc, i.e. in themetastable regions, for various compositions. Theelectrical conductivity

(1 kHz) measured (for first and second heating cycle) at a typicalheating rate of 1 °C/min versus temperature is shown in Fig. 3.

Fig. 3a, c and e represents the first and Fig. 3b, d and f representsthe second heating cycles of 10, 15 and 30CI-SISOVO samples,respectively. The conductivity and temperature cycles for all thesamples are reversible at least up to 65–70 °C (T≤Tg).

For the 10CI-SISOVO sample (Fig. 3a) above the glass transitiontemperature (Tg∼65 °C) and prior to crystallization temperature(Tc∼110 °C), the conductivity increases at appreciably faster rate anddeviates from linearity. On further heating, the conductivity shows anotable drastic fall. It may be stressed here that fall in the conductivitycorresponds to beginning and saturation to that of completion thecrystallization process. On further heating above 130 °C, the conduc-tivity again increases linearly.

The σ–T characteristics of 15 and 30CI-SISOVO (Fig. 3c and e,respectively) are rather dissimilar to that of 10CI-SISOVO sample. Forthe 15CI-SISOVO sample conductivity exhibits a little deviation fromthe Arrhenius behavior at Tg and on further heating a slight fall nearTc. Subsequently, conductivity attains a plateau for a small temper-ature range and further increases linearly with temperature.

The σ–T behavior for 30CI-SISOVO (Fig. 3c) exhibit no significantdeviation from Arrhenius behavior at Tg, though a plateau like region isapparent in the vicinity of Tc of this sample that may be an indication ofsmaller and slower crystallization. Earlier structural studies on glassy[26] and mechanochemically synthesized [27] AgI–Ag2O–V2O5 systemhave confirmed that precipitation of Ag8I4V2O7 affects the conductivityvery significantly and is indeed responsible for its sharp fall at Tc. In thepresent investigation, as discussed earlier (Section 3.2), the precipitationof Ag8I4V2O7 decreases notably with increasing CuI content. Thus, thehigh CuI content samples may exhibit relatively smaller and slowerprecipitation of Ag8I4V2O7 during crystallization and hence the signif-icant conductivity change at Tc is not seen in the σ–T characteristics.

After the completion of first heating cycle, the electrical conduc-tivity of the 10, 15, and 30CI-SISOVO is found to have dropped, byfactors of 6, 5 and 4, respectively at room temperature. In the secondheating cycles, for all three samples, the conductivity increasesArrheniously. Further the 15 and 30CI-SISOVO samples exhibit asignificant deviation from linearity followed by a rise in conductivity

Table 1Characteristics temperatures and total enthalpy release during crystallization in the CI-SISOVO system.

x ΔH (J/g) Tc (°C) Tg (°C) Trg=Tg/Tm Tc−Tg (°C)

0.10 20±0.40 110 70 0.37 400.15 19±0.38 102 65 0.34 370.20 18±0.36 96 62 0.32 340.25 14±0.28 94 61 0.30 330.30 10±0.20 92 60 0.29 32

Fig. 3. Electrical conductivity as a function of temperature at a typical heating rate of1 °C/min. Symbols denote (●) first and (○) second heating cycles.

1609N. Gupta et al. / Solid State Ionics 180 (2010) 1607–1612

Author's personal copy

at ∼147 °C that may be due to β→α phase transition in theprecipitated AgI. Such rise is more apparent for 30CI-SISOVO andalmost negligible in case of 10CI-SISOVO sample.

The electrical conductivity–temperature cycles are further scruti-nized to unfold the information about thermal events in the system.Thus for Tg≤T≤Tc and T≥Tc, the conductivity as a function oftemperature has been replotted for 50SISOVO (that contains no CuI)and 10, 15CI-SISOVO samples in Fig. 4. As noticed, the fall in theconductivity above the crystallization temperature apparentlydepends on the CuI content in the samples. For 15CI-SISOVO, at Tcfall in the conductivity is drastically small, whereas, for x=0.3 only aplateau is seen in σ–T plot (Fig. 4) with no further fall in theconductivity near Tc. Thus, as the CuI content increases, fall in theconductivity at crystallization decreases. For low CuI content samples,on the completion of crystallization, conductivity attains a constantvalue of σXL and further found to be increasing linearly with thetemperature. To further understand the relation between electricalconductivity changes associated with the crystallization in these CuIcontained AgI–Ag2O–V2O5 samples, two parameters are defined. Todefine the first parameter, the conductivity versus temperature cyclein linear region (TbTg) is extrapolated to find the conductivity value atTg (i.e. σexp). For the compositions well within in the glass formingregion, the parameter Rs is defined as

Rs =σp

σexpð2Þ

Thus it is the ratio of value of conductivity at the peak (σp) to theextrapolated conductivity value at Tg, viz., σexp. This parameterdefines the extent of anomalous rise in conductivity above Tg. Higherthe Rs, more will be the deviation in conductivity from Arrheniousbehavior. This, in turn, may signify the structural relaxation and hencehigher free volume change at Tg. Anomalous rise in the conductivity atand above Tg is generally expressed by free volume approachproposed by Souquet [28]. According to this model below Tg is theglassy state where glass can be visualized as “frozen liquid” with afreezed viscosity. Between glass transition and crystallization tem-peratures (TcNTNTg) the glass is in the supercooled state that is arelatively less viscous state. Thus, at Tg, due to decrease in viscositythe free volume in the glass increases. The increase in the electricalconductivity at Tg may be attributed to this free volume change thatfacilitates the ion motion. The Rs is plotted as a function of CuI contentin Fig. 5a. Apparently Rs is found to be decreasing with CuI content. Itmay suggest that Cu+ substitution either (i) decreases the freevolume change associated with structural relaxation at Tg, or (ii) such

a free volume change at Tg becomes less significant due to overallincrease in room temperature conductivity of the samples.

Secondly, for such samples in which conductivity shows appre-ciable fall at Tc (σXLbσp), one can define the ratio of the conductivityat the peak (σp) to that of its value attained by the sample aftercrystallization (σXL) in the σ vs T plot as the second parameter Rc,

Rc =σp

σXL; Rc≥1 ð3Þ

Our definition of Rc excludes those glasses in which conductivity ofcrystalline samples is found to be higher than those of respectiveamorphous/glassy state [29]. Moreover, our samples in which crystal-lization rate is very slow (e.g. x≥0.2) are also excluded from discussion.

Since any sudden and significant change in the conductivityindicates a structural transformation in the system, the value of Rcwould infer the extent of crystallization during the phase transforma-tion. Higher value of Rc corresponds to massive fall in the conductivitythat in turn indicates larger magnitude of crystallization. Thus Rc givesvaluable information about the extent of crystallization. The parameterRc is plotted in Fig. 5b and interestingly also found to be decreasingwithCuI content. Such behaviour of the parameter suggests that CuI additionsuppresses the crystallization in the samples.

The conductivity at room temperature is plotted as a function of CuIcontent (Fig. 6a). It is apparent that conductivity increases with CuIcontent, reaches to a maximum of ∼10−2 Ω−1 cm−1 at x=0.35 (35CI-SISOVO) and further decreases for higher CuI content. The decrease inthe conductivity above x=0.35may be due to increased crystallinity inthe samples. The conductivity of the annealed (above Tc) glassy samplesalso exhibits an increase with Cu+ ion substitution in the matrix.

The activation energy of ionic conduction as a function of CuIcontent is found to be decreasing (Fig. 6b), which shows consistencywith conductivity–temperature cycles. The activation energy ofcrystallized samples is found to be lower than those of the virginsamples and follows same trend with CuI content. Similar results onCuI substituted systems are obtained in earlier investigations [15]. Theenhancement in the conductivity is further investigated.

Fig. 4. Electrical conductivity–temperature plot on an extended scale. Conductivity atpeak (σp), after crystallization (σXL) and extrapolated at peak temperature (σexp) areshown for 50SISOVO sample.

Fig. 5. The two parameters Rs and Rc (obtained from conductivity–temperature cycles)versus CuI content.

1610 N. Gupta et al. / Solid State Ionics 180 (2010) 1607–1612

Author's personal copy

The electrical conductivity is given by

σ = ∑iniqiμi ð4Þ

where, ni is the density and qi and µi the charge and mobility of the ithspecies of the charge carriers, respectively. As also discussed earlier(Section 3.3), it is proposed by various workers that substituted CuIeventually transforms into Cu2O due to following reaction [25]:

2CuI þ Ag2O→2AgI þ Cu2O ð5ÞThus, the total majority charge carriers i. e., Ag+ ions in the system

are constant andhence increase in the conductivity cannot be attributedto the increase in number of charge carriers (ni). Secondly, investigationby variousworkers [19,30] have shown that Cu+ ion in such systems donot contribute in the ionic conductivity and the total electricalconductivity is only due to those of Ag+ ions. Thus, the only possibilityof the conductivity increase should be solely due to increase inmobilityof ions due to the formation of favorable structure for smooth ionicmovement. Thus, to further understand the conductivity rise with CuIcontent, roleof substitutionof Cu2O inplace ofAg2O is also studied in theSISOVO system. These samples viz., 50AgI–33.33(Ag2O1−y–Cu2Oy)–16.67V2O5 (y=0.1–0.3) are found to be purely glassy in nature. Theelectrical conductivity as function of temperature is studied for thesesamples and shown for two of these in Fig. 7. The inset of Fig. 7 showsionic conductivity (300 K) with Cu2O content. As apparent, theconductivity increases till y=0.3 and subsequently falls at higher y.For y=0.1 conductivity increases linearly till ∼65 °C and on furtherheating conductivity shows anomalous deviation from Arrheniusbehavior. At ∼110 °C there is a drastic fall in the conductivity similarto CI-SISOVO system due to crystallization in the sample. For higherCu2O content samples, viz., y=0.2, it is evident that conductivity attainsa plateau at Tc. Since, Cu2O addition do not supply free Ag+ ion in thesystem, thus, a favorable structural formation in the samples due toCu2O addition is expected. The above results further strongly suggest

that Cu2O formation in CuI–AgI–Ag2O–V2O5 system, leads to theincrease in the conductivity.

To further understand, the increase in the conductivity with CuIcontent, the ionic mobility (µ) is measured as a function of CuI contentby the transient ionic current (TIC) technique [20,21]. Blockinggraphite electrodes are pasted on both sides of the glassy sample andcylindrical pellet is further sandwiched between platinum electrodes.The dc voltage of V≈0.4 V is applied for 1 h to completely polarize thesample. The polarity is then reversed and the transient current ismeasured as a function of time. The mobility µ (cm2 V−1 s−1) canthus be calculated by the relation

μ =d2

Vτð6Þ

Where d is the thickness of the samples and V is the appliedvoltage. The transient current increases rapidly, attains a maximumand subsequently falls (30CI-SISOVO Fig. 8a). Time (τ) in which

Fig. 6. Electrical conductivity at 300 K (a) and activation energy (T≤Tg) (b) as afunction of CuI content. Symbols denote: (●) room temperature conductivity before Istheating cycle; (○) room temperature conductivity after Ist heating cycle; (▲)Activation energy measured during Ist heating cycle; (Δ) for IInd heating cycle.

Fig. 7. Electrical conductivity versus temperature cycles for 50AgI–33.33[(Ag2O)1− y–

(Cu2O)y]–16.67V2O5 system. (⊕) y=0.1 and (●) y=0.2. Inset: Conductivity (300 K) asa function of Cu2O content.

Fig. 8. (a) Transient current versus time plot on an extended scale for 30CI-SISOVOsample (b) d2 versus Vτ plot for 10 (●) and 30 (○) CI-SISOVO samples.

1611N. Gupta et al. / Solid State Ionics 180 (2010) 1607–1612

Author's personal copy

current attains amaximum is obtained from this I−t plot (Fig. 8a). Forbetter accuracy, mobility is measured for different thicknesses. Fig. 8shows d2 versus Vτ plot for two samples 10 and 30CI-SISOVO.Mobility is measured from the slope. Apparently, mobility increaseswith CuI content in the sample. The value of mobility is found to be0.34±0.03 and 0.93±.09 cm2 V−1 s−1 for 10 and 30CI-SISOVOsamples, respectively.

Thus, it may be concluded here that the rise in conductivity withCuI content is essentially due to Cu2O substitution in the glass matrixthat in turn increases the ionic mobility.

4. Conclusions

Cu+ ion substitution significantly affects electrical as well asthermal properties of the AgI–Ag2O–V2O5 system. Samples tillx=0.35 are highly glassy in nature. The ionic conductivity is foundto be increasing with CuI content that may be attributed to increase inthe mobility due to Cu2O incorporation into the matrix, whereas,subsequent drop above x=0.35 may be due to crystallization in thesamples. The system is found to exhibit multiple crystallization andelectrical conductivity versus temperature cycles successfully revealthe crystallization behavior of these samples. The massiveness/extentof the crystallization decreases with CuI content in the glass matrix.

Acknowledgements

This work is supported by Department of Science and Technology(India), SERC fast track project, SR/FTP/PS-77/2005 and the support isgratefully acknowledged. Thanks are also due to Dr N. P. Lalla UGCDAE CSR, Indore (India) for X-ray diffraction measurements.

References

[1] H.L. Tuller, D.P. Button, D.R. Uhlmann, J. Non-Cryst. Solids 40 (1980) 93.[2] T. Saito, M. Tatsunisago, T. Minami, Solid State Ion. 61 (1993) 285.[3] T. Minami, Y. Takuma, M. Tanaka, J. Electrochem. Soc. 124 (1977) 1659.[4] R.V.G.K. Sharma, S. Radhakrishna, Solid State Ion. 40/41 (1990) 483.[5] A. Dalvi, K. Shahi, J. Phys. Chem. Sol. 64 (2003) 813.[6] M. Foltyn, M.Wasiucionek, J. Garbarczyk, J.L. Nowinski, Solid State Ion. 176 (2005)

2137.[7] A. Magistris, G. Chiodelli, M.J. Duclot, Solid State Ion. 9/10 (1983) 611.[8] D.E. Dey, J. Non-Cryst. Solids 21 (1976) 343.[9] J.N. Mundy, G.L. Jin, Solid State Ion. 25 (1987) 71.[10] M.K.P. Seydei, S.A. Suthanthiraraj, Bull. Electrochem. 12 (1996) 683.[11] R.G. Peason, J. Am. Chem. Soc. 85 (1963) 3533.[12] B. Rivolta, F. Bonino, B. Scrosati, Mater. Chem. Phys. 19 (1988) 557.[13] K. Hariharan, C. Sangamitra, A.M. Sukeshini, Solid State Ion. 53–56 (1992) 1179.[14] V.G. Chandrasekhar, S.A. Suthanthiraraj, J. Mater. Sci. 28 (1993) 4043.[15] S.A. Suthanthiraraj, S. Murugesan, P. Maruthamuthu, Mater. Res. Bull. 37 (2002)

2145.[16] K.P. Padmasree, D.K. Kanchan, A.R. Kulkarni, Solid State Ion. 177 (2006) 475.[17] R. Chitradevi, S. Selvasekarapandian, Ionics 6 (2006) 203.[18] S.S. Sekhon, J. Mater. Sci. 34 (1999) 2899.[19] S. Murugesan, B. Bergman, Electrochim. Acta 52 (2007) 8064.[20] S. Chandra, S.K. Tolpadi, S.A. Hashmi, J. Phys., Condens. Matter 1 (1989) 9101.[21] R.C. Agrawal, R. Kumar, J. Phys., D, Appl. Phys. 27 (1994) 2431.[22] A. Dalvi, K. Shahi, Solid State Ion. 148 (2002) 431.[23] S. Adams, K. Hariharan, J. Maier, Solid State Ion. 75 (1995) 193.[24] A. Dalvi, A.M. Awasthi, S. Bharadwaj, Solid State Ionics, in: B.V.R. Chowdari, et al.,

(Eds.), Proceedings of 11th Asian Conference on Solid State Ionics, Coimbatore,India, Narosa Publications, New Delhi, 2008, p. 555.

[25] S. Murugesan, A. Wijayasinghe, B. Bergman, Solid State Ion. 178 (2007) 779.[26] St. Adams, K. Hariharan, J. Maier, Solid State Ion. 86 (1996) 503.[27] A. Dalvi, K. Shahi, J. Non-Cryst. Solids 341 (2004) 124.[28] J.L. Souquet, Annu. Rev. Mater. Sci. 11 (1981) 211.[29] M. Tatsumisago, T. Saito, T. Minami, Chem. Lett. (2001) 790.[30] S. Murugesan, S.A. Suthanthiraraj, P. Maruthamuthu, Mater. Res. Bull. 42 (2007)

2017.

1612 N. Gupta et al. / Solid State Ionics 180 (2010) 1607–1612