THERMAL BEHAVIOUR OF THE LOCAL ENVIRONMENT AROUND IODINE IN FAST-ION-CONDUCTING AGI-DOPED GLASSES

(1)

HAL Id: hal-00513794

https://hal.archives-ouvertes.fr/hal-00513794

Submitted on 1 Sep 2010

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

THERMAL BEHAVIOUR OF THE LOCAL ENVIRONMENT AROUND IODINE IN

FAST-ION-CONDUCTING AGI-DOPED GLASSES

Andrea Sanson, Francesco Rocca, Paolo Fornasini, Giuseppe Dalba, Rolly Grisenti, Andrea Mandanici

To cite this version:

Andrea Sanson, Francesco Rocca, Paolo Fornasini, Giuseppe Dalba, Rolly Grisenti, et al.. THER- MAL BEHAVIOUR OF THE LOCAL ENVIRONMENT AROUND IODINE IN FAST-ION- CONDUCTING AGI-DOPED GLASSES. Philosophical Magazine, Taylor & Francis, 2007, 87 (3-5), pp.769-777. �10.1080/14786430601032394�. �hal-00513794�

(2)

For Peer Review Only

THERMAL BEHAVIOUR OF THE LOCAL ENVIRONMENT AROUND IODINE IN FAST-ION-CONDUCTING AGI-DOPED

GLASSES

Journal: Philosophical Magazine & Philosophical Magazine Letters Manuscript ID: TPHM-06-May-0140.R3

Journal Selection: Philosophical Magazine Date Submitted by the

Author: 15-Sep-2006

Complete List of Authors: Sanson, Andrea; Istituto di Fotonica e Nanotecnologie del Consiglio Nazionale delle Ricerche

Rocca, Francesco; Istituto di Fotonica e Nanotecnologie del Consiglio Nazionale delle Ricerche

Fornasini, Paolo; University of Trento, Physics Dalba, Giuseppe; University of Trento, Physics Grisenti, Rolly; University of Trento, Physics Mandanici, Andrea; University of Messina, Physics

Keywords: diffusion, EXAFS, glass, ionic conductivity, thermal properties Keywords (user supplied):

Note: The following files were submitted by the author for peer review, but cannot be converted to PDF. You must view these files (e.g. movies) online.

ManuscriptRevised2_Sanson.tex

http://mc.manuscriptcentral.com/pm-pml

Philosophical Magazine & Philosophical Magazine Letters

(3)

For Peer Review Only

Page 1 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(4)

For Peer Review Only

0 5 10 15

-0.05 0.00 0.05

k (Å

^-1

)

-0.05 0.00 0.05

k F k (Å

-1

)

-0.05 0.00 0.05

Page 2 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(5)

For Peer Review Only

0.00 0.04 0.08

298 K 149 K 39 K

borate

0 2 4 6

0.00 0.04 0.08

molybdate x=0.75 150 K

300 K 25 K

R (Å)

0.00 0.04 0.08

151 K

M o d u li F .T . (Å

-2

)

^{25 K}

Page 3 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(6)

For Peer Review Only

0 100 200 300

2.5 3.0 3.5 4.0

I- A g c o o rd in at io n n u m b er

Temperature (K)

0 100 200 300

Temperature (K)

Page 4 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(7)

For Peer Review Only

0 100 200 300

-0.03 -0.02 -0.01 0.00 0.01

I- A g d is ta n ce v ar ia ti o n ( Å )

Temperature (K)

0 100 200 300

Temperature (K)

Page 5 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(8)

For Peer Review Only

0.000 0.005 0.010 0.015 0.020

C

2

* ( Å

2

)

0 100 200 300

0.0000 0.0005 0.0010

C

3

* ( Å

3

)

Temperature (K)

0 100 200 300

Temperature (K)

Page 6 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(9)

For Peer Review Only

Philosophical Magazine,

Vol. 00, No. 00, DD Month 200x, 1–7

THERMAL BEHAVIOUR OF THE LOCAL ENVIRONMENT AROUND IODINE IN FAST-ION-CONDUCTING AGI-DOPED GLASSES

A. SANSON^†∗, F. ROCCA^†, P. FORNASINI^‡, G. DALBA^‡, R. GRISENTI^‡, A. MANDANICI^§

†Istituto di Fotonica e Nanotecnologie del Consiglio Nazionale delle Ricerche, Sezione CeFSA di Trento - Via Sommarive 14, I-38050 Povo (Trento), Italy

‡Dipartimento di Fisica dell’Universita`di Trento, Via Sommarive 14, I-38050 Povo (Trento), Italy

§Dipartimento di Fisica dell’Universita`di Messina, Salita Sperone 31, I-98100 Messina, Italy

(30 April 2006)

The local environment around iodine has been studied as a function of temperature by extended x-ray absorption fine structure (EXAFS) in fast-ion-conducting AgI doped glasses (borate and molybdate): silver borate glass (AgI)0.55(Ag2O:4B2O3)0.45 has been investigated from 39 to 298 K, silver molybdate glasses (AgI)x(Ag2MoO4)1−x, with x=0.67 and x=0.75, from 25 K to Tg (330÷350 K).

In spite of the very different character of their host matrix, these two families of glasses display a very similar local thermal behaviour around iodine: the mean distance and the coordination number of the short-range I-Ag distribution decrease when temperature increases.

These experimental results can be related to temperature dependence of the diffusion of the silver ions, giving new elements to better understand the mechanism of ionic conduction.

1 INTRODUCTION

Fast-ion-conducting (FIC) glasses have been attracting much attention for their potential applications to solid-state electrochemical devices, including batteries, sensors, and smart windows [1, 2]. Because of their technological importance, FIC glasses have been investigated extensively to establish the chemical and structural characteristics responsible for their high ionic conductivity. The acquisition of this knowledge is critical not only for explaining the mechanism of ion transport in glasses, but also for designing systems with improved performance appropriate for the current needs [1–7].

Among FIC glasses, AgI-doped glasses represent a wide family, with a great variety of host matrices, characterized by different former oxides: borates, phosphates, molybdates, tungstates, and so on [8–11].

Many approaches have been proposed in order to explain the transport properties of AgI-based glasses:

the weak electrolyte model [12], the random site model [13], the cluster and cluster-bypass models [14–16], which assume the formation of AgI micro-clusters within the glass; the dynamic structure and the diffusive pathway models [17–19] where, in contrast, the dopant salt is assumed to be homogeneously distributed in the glass matrix; the jump diffusion model [20], which attempts at giving a microscopic interpretation of the conductivity spectra. In spite of the great variety of developed models, the full relationship between structure and dynamics at short and medium range order and ionic transport are not yet well understood.

For this reason, new experimental investigations are necessary.

According to the Arrhenius law, temperature is the main physical property responsible for the variation of the mobility of the diffusing ions. Extended x-ray absorption fine structure (EXAFS) is a powerful technique for investigating the local environment around selected atomic species [21]. EXAFS measurements as function of temperature are a very effective tool for obtaining new insights into the local origin of the ionic conduction in AgI-rich glasses. The aim of this work is to investigate the temperature effect on the short-range order of AgI-doped FIC glasses and the corresponding relation with the transport properties.

∗Corresponding author. Email: sanson@science.unitn.it

Philosophical Magazine

ISSN 1478-6435 print/ISSN 1478-6443 online c° 200x Taylor & Francis http://www.tandf.co.uk/journals

DOI: 10.1080/1478643YYxxxxxxxx Page 7 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(10)

For Peer Review Only

2 A. Sanson et al.,

0 5 10 15

-0.05 0.00 0.05

k (Å^-1)

-0.05 0.00 0.05

kFk(Å-1 )

-0.05 0.00 0.05

0.00 0.04 0.08

298 K 149 K 39 K

borate

0 2 4 6

0.00 0.04 0.08

300 K 25 K

R (Å) 0.00

0.04 0.08

151 K

Moduli F.T. (Å-2 ) ^{25 K}

Figure 1. EXAFS signals (left panels) at low and room temperatures (solid and dashed lines, respectively) in silver borate glass (AgI)0.55(Ag2O:4·B2O3)0.45(top panel), in silver molybdate glasses (AgI)x(Ag2MoO4)1−xwith x=0.67 (middle panel) and x=0.75

(bottom panel). The moduli of Fourier transform of EXAFS signals are shown in the right panels at three selected temperatures.

2 EXPERIMENT AND DATA ANALYSIS

EXAFS measurements at the K edge of Iodine have been performed at the BM29 beamline of ESRF in Grenoble (France) on the silver borate glass (AgI)_0.55(Ag₂O:4B₂O₃)_0.45 in the temperature range 39÷298 K, and on silver molybdate glasses (AgI)_x(Ag₂MoO₄)_1−x with x=0.67 and x=0.75 between 25 K and Tg (330÷350 K), using a cold-hot finger sample-holder in vacuum. Also crystalline β-AgI has been measured at low temperatures in order to make a comparison with glasses. The glasses were prepared by the melting- quenching technique. Appropriate mixtures of AgI (Aldrich 99 %) with silver and boron oxides (for borate glass) and with Ag₂MoO₄ (for molybdates, prepared by precipitation from aqueous solution of AgNO₃ and Na₂MoO₄ [22]) were melted at 800 and 600 ^◦C, respectively, for 2 hours and then quenched down to room temperature in stainless steel molds. The presence of possible crystallites in the glasses is ruled out by x-ray diffraction measurements. The beamline was set in the flat crystal monochromator Si(311) configuration (with mirrors), obtaining high photon flux and stability. Good quality EXAFS data were obtained in transmission mode up to 15 ˚A⁻¹, also for the highest temperature. Generally, two spectra were collected for each temperature.

The analysis of data was performed according to well-established procedures, within the approach based on the cumulant expansion method [21, 23, 24]. The edges of all spectra have been aligned to within 0.1 eV or better, and the k · χ(k) EXAFS signals were Fourier transformed in the interval k =2.4÷14 ˚A⁻¹ (up to 15 ˚A⁻¹ for molybdate glasses) using a Gaussian window (Fig. 1). The first-shell signal has been isolated from noise by Fourier back-transform, and the EXAFS cumulants and coordination number have been obtained using the method based on phases difference and amplitudes ratio [24,25], taking the lowest

Page 8 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(11)

For Peer Review Only

Thermal behaviour of the local environment around Iodine in fast-ion-conducting AgI-doped glasses 3

temperature spectrum as reference for backscattering amplitudes, phase shifts, and inelastic terms. The low temperature spectra of glasses have been in turn calibrated against the low temperature spectra of β-AgI.

The EXAFS cumulants are directly related to position and shape of the distance distribution probed by EXAFS: the first cumulant C₁^∗ is the mean value, the second cumulant C₂^∗ is the variance, the higher order cumulants measure the deviations from the gaussian shape. In highly disordered systems, the cumulants given by the EXAFS analysis could not reproduce the exact values of the cumulants of the complete distance distribution, since the convergence of the cumulant series can be limited to a low-k interval [21,26].

In their pioneering study on the superionic phase transition of AgI, Boyce et al. [27], by a standard analysis of EXAFS, found a 25 % reduction of the Ag-I coordination number in the superionic α-phase with respect to the expected value N = 4. Such reduction was attributed to the the Ag ions in diffusion. In a subsequent work, by an excluded-volume model, they were able to reconstruct the whole Ag-I distance distribution containing the right number of four silver ions. This distribution is characterized by a tail extending to long distances, which represents the conduction pathway [28]. To hypothesize a model distribution is relatively easy for crystals like AgI, where in the α-phase the iodine ions still form a rigid sublattice, and the available sites for silver ions have well defined geometrical shapes. It is by far less simple in the case of liquids and non-crystalline solids, like FIC glasses. A solution to this problem has been suggested by Filipponi [29].

The basic idea is to separate the nearest-neighbours distribution into a short-range narrow asymmetric distribution and a long-range tail distribution; the long-range tail is obtained from other techniques, like diffraction experiments with reverse Monte Carlo (RMC) modeling. Only the parameters of the short-range distribution are used as fitting parameters in the EXAFS analysis. An application of this procedure has been recently made in both superionic and liquid phases of CuI [30].

The accuracy of this mixed approach depends on the reliability of the long-range tail distribution, which can be difficult to obtain in the case of multi-component glasses at various temperatures, like required by our present study on FIC glasses. At the present, to our knowledge, only a work based on diffraction + RMC modeling at room temperature is available, and not the partial distributions have been published, but only the interference functions [31]. For this reason, we follow here the first pioneering approach of Boyce: we use the cumulants approach extending the data analysis to the whole available k range (Boyce used a simple gaussian standard analysis, i.e. first and second cumulant), but the polynomial coefficients obtained from the ratio method cannot be safely considered as the cumulants of the ”whole” effective distribution of all nearest-neighbour silver ions. We make the assumption that the EXAFS polynomial coefficients only give a meaningful parametrization of the short-range narrow component of the whole I-Ag distribution, corresponding to the Ag ions more tightly bound to iodines. The remaining long-range tail, which contains Ag ions participating to conduction, escapes EXAFS detection, as assumed by Boyce.

The reasonability of this assumption is given by the following consideration: typical values of the jumping- length, which approximates the extent of the long-range tail distribution, is within 1÷3 ˚A in FIC glasses [36]. Then the Ag ions in the long-tail distribution contribute to EXAFS only by scattering the long- wavelength photoelectrons (low k values); the other photoelectrons of higher wavelengths, when scattered by different regions of the long-range tail, give rise to signals that interfere destructively. As final result the Ag ions in the long-range tail nearly completely escape detection by EXAFS. It will be shown below that the temperature dependence of the cumulants extracted from our analysis can reasonably be interpreted in terms of silver ion mobility and conduction properties of the glasses, proving the soundness of our assumption.

3 RESULTS

According to previous considerations, the cumulants obtained from the EXAFS analysis refer to a short- range distribution of I-Ag distances, that is assumed as due to silver ions bound to iodine, i.e. not in diffusion. The temperature dependence of the coordination number, say of the average number of Ag ions surrounding I ions in this short-range distribution, is reported in Fig. 2 for the two families of investigated glasses: when temperature increases, the coordination number decreases, and is anyway smaller than the value 4 of crystalline β-AgI. Another striking result is given by the temperature dependence of the I-Ag Page 9 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(12)

For Peer Review Only

0 100 200 300

2.5 3.0 3.5 4.0

I-Ag coordination number

Temperature (K)

0 100 200 300

Temperature (K)

Figure 2. Temperature dependence of the short-range I-Ag coordination number in silver borate glass (AgI)0.55(Ag2O:4·B2O3)0.45

(left panel) and in silver molybdate glasses (AgI)x(Ag2MoO4)1−x with x=0.67 and x=0.75 (open and full circles, respectively, in the right panel). The dashed lines are the corresponding linear fits.

0 100 200 300

-0.03 -0.02 -0.01 0.00 0.01

I-Ag distance variation (Å)

Temperature (K)

0 100 200 300

Temperature (K)

Figure 3. Temperature dependence of the short-range I-Ag distance in silver borate glass (AgI)0.55(Ag2O:4·B2O3)0.45(left panel) and in silver molybdate glasses (AgI)x(Ag2MoO4)1−x with x=0.67 and x=0.75 (open and full circles, respectively, in the right panel).

distance (first cumulant C₁^∗): in both families of glasses, borates and molybdates, the average value of the I-Ag short-range distance distribution decreases when temperature increases (Fig. 3), the contraction being about 0.02 ˚A within the explored temperature interval. On the contrary, in crystalline β-AgI, the mean I-Ag distance expands of about 0.02 ˚A within the same temperature interval [32]. In the glasses here investigated, the contraction of the I-Ag distance is clearly evident independently from the method of analysis. The EXAFS signals at room temperature reported in Fig. 1, in fact, show a lower frequency with respect to the EXAFS signals at lower temperature, indicating an expansion of the first-shell distance while temperature is decreasing. This gives evidence of the reasonableness of our cumulants analysis.

Also the temperature dependence of the higher order cumulants has been extracted from EXAFS: the second cumulants C₂^∗, that measure the variance of the short-range I-Ag distance distribution, are displayed in the top panels of Fig. 4; the third cumulants C₃^∗, that measure the asymmetry of the distribution, are displayed in the bottom panels of the same figure; the fourth cumulant C₄^∗, which measures the flatness of the distribution with respect to a gaussian behaviour, is about (1.0 ± 0.1)·10⁻⁴ A˚⁴ in both families of glasses, with a negligible dependence on the temperature. At low temperature, where diffusion effects are negligible and the short-range distribution is, to an excellent approximation, equal to the total distribution, the absolute values of the three cumulants C₂^∗, C₃^∗ and C₄^∗ are much larger than the corresponding values for β-AgI; this behaviour is a direct evidence of the presence of structural disorder around iodine in the glasses.

Page 10 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(13)

For Peer Review Only

0.000 0.005 0.010 0.015 0.020

C2* (Å2 )

0 100 200 300

0.0000 0.0005 0.0010

C3* (Å3 )

Temperature (K)

0 100 200 300

Temperature (K)

Figure 4. Temperature dependence of the second (top panels) and third (bottom panels) cumulants extracted by EXAFS: in silver borate glass (AgI)0.55(Ag2O:4·B2O3)0.45(left panels) and in silver molybdate glasses (AgI)x(Ag2MoO4)1−x with x=0.67 and x=0.75

(open and full circles, respectively, in the right panels). With the triangles are indicated the second and third cumulant of crystalline β-AgI, from ref. [32] (down triangles) and from unpublished measurements here analyzed with the same parameters of analysis used in

the glasses (up triangles).

4 DISCUSSION

A first striking result of this investigation is the great similarity of the thermal behaviors of the two families of investigated glasses, in spite of the structural difference of their host matrices. In silver borate glasses, the binary matrix Ag₂O-B₂O₃ forms a continuous random network, while AgI-doped molybdate glasses are depolymerised systems where the oxyanions, generally (MoO₄)²⁻ tetrahedra, do not form networks.

Notwithstanding this relevant difference, the local environment around iodine ions is very similar in the two families of AgI-doped glasses: the short-range I-Ag distance shrinks of about 0.02 ˚A between low and room temperature, the I-Ag coordination number reduces from about 3.7÷3.9 to 2.9÷3.1 (with a larger variation in molybdate glasses), and the higher order cumulants are very similar. These results support the idea that, at least from the local point of view of iodine, the mechanism of ionic conduction is the same in the two different families of glasses.

Let us now try to establish a connection between the EXAFS results and the transport properties of these investigated systems. The short-range I-Ag coordination number displayed in Fig. 2, which represents the mean number of Ag ions tightly bound to iodine or, equivalently, the mean number of Ag ions not undergoing diffusion, decreases when temperature increases. According to the Arrhenius law, when temperature increases the ionic conductivity increases, and the number of Ag ions undergoing diffusion increases.

A reasonable interpretation of EXAFS results is the following: when temperature increases, the long-tail of the whole I-Ag distance distribution is progressively populated by the growing number of diffusing Ag ions, while the contribution to the short-range distribution decreases and, accordingly, the coordination number measured by EXAFS is reduced.

A linear fit to the temperature dependence of coordination numbers in Fig. 2 gives their mean slope for each glass family: for the silver borate glass the slope ∆N/∆T is about -(2.2 ± 0.1)·10⁻³ K⁻¹, while for silver molybdate glasses the slope, averaged over the two compositions, is about -(3.2 ± 0.1)·10⁻³ K⁻¹. The absolute value of slope ∆N/∆T can be related to the activation energy for ionic conductivity, about 0.23 eV for molybdate glasses (0.24-0.26 eV for glass x=0.67 and 0.20-0.23 eV for glass x=0.75 [33, 34]), Page 11 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(14)

For Peer Review Only

and about 0.35 eV for the borate glass [35]. Temperatures being equal, the glasses with lower activation energy are characterized by a more favorable thermal activation of the mobile cations, corresponding, in our glasses, to a higher probability of jump of the Ag ions and to a greater de-population of the short-range distribution. In agreement, we have found a larger reduction of the EXAFS I-Ag coordination number in silver molybdate glasses, where the activation energy is lower, than in silver borate glasses; in particular, the ratio between the corresponding glass activation energies is about 1.5 and, approximately, also the ratio between the slopes ∆N/∆T of the two glasses has the same value. At the present, this empirical (and sensible) result cannot be generalized to the other families of FIC glasses, but anyway it cannot be excluded.

Also the decrease of the I-Ag average distance displayed in Fig. 3 can be explained in terms of the diffusion of Ag ions. While in crystalline β-AgI each I ion is surrounded by a tetrahedron of four Ag ions at the same crystallographic distance, in AgI-doped glasses at low temperature each iodine ion has a different environment and the four I-Ag distances are different. The low-temperature values of C₂^∗ and C₃^∗ displayed in Fig. 4, much higher than the corresponding values for β-AgI, are a clear indication of this structural disorder. It is reasonable to assume that, of the different I-Ag distances in the low-temperature glasses, the longer ones correspond to weaker I-Ag bonds. It is reasonable to assume that, when temperature increases, the Ag ions undergoing diffusion are the ones less tightly bound to iodine, and are thus found at a greater distance from iodine. This gives a direct explanation of the reduction of the short-range I-Ag distance displayed in Fig. 3: the Ag ions that participate to ionic diffusion mainly jump from the longer I-Ag distances and, as consequence, the mean distance of the short-range I-Ag distribution is reduced when temperature increases.

5 CONCLUSIONS

In this work the short-range structure order around iodine ions has been investigated by EXAFS in two families of AgI-doped FIC glasses, borates and molybdates, as a function of temperature. The results are interpreted as monitoring the progressive transition of Ag ions from the short-range distribution (say the distribution of Ag ions not undergoing diffusion) to the long-range conduction tail (containing the Ag ions participating to conduction), which escapes EXAFS detection. The thermal behaviour of the local environment around iodine is very similar in the two types of glasses, despite the diversity of their glass matrices. In both glasses, the coordination number and the mean value of the short-range I-Ag distance distribution are reduced when temperature increases. The ionic diffusion, thermally activated, mainly involves those Ag ions which are more distant from their iodine nearest-neighbours. These aspects should be taken into account which aim to develop suitable theoretical models for the mechanism of ionic conduction.

6 ACKNOWLEDGEMENTS

The authors are grateful to C. Armellini for samples preparation and to the staff of the BM29 beamline at ESRF for technical assistance. The financial support by ESRF (Projects HS-1666 and HS-2463) is acknowledged. A.S. aknowledges also the support of the Provincia Autonoma di Trento, through the LOTHEX project at IFN-CNR.

References

[1] Fast Ion Transport in Solids, edited by P. Vashista, J. N. Mundy, and G. K. Shenoy (Elsevier, New York, 1979).

[2] T. Minami, J. Non-Cryst. Solids 73 273 (1985).

[3] H. L. Tuller, D. P. Button, and D. R. Uhlmann, J. Non-Cryst. Solids 40 93 (1980).

[4] D. Ravaine, J. Non-Cryst. Solids 73 287 (1985).

[5] M. D. Ingram, Phys. Chem. Glasses 28 215 (1985).

[6] J. L. Souquet, Solid State Ionics 28-30 693 (1988).

[7] C. A. Angell, Annu. Rev. Phys. Chem. 43 693 (1992).

[8] J. Kawamura and M. Shimoji, J. Non-Cryst. Solids 88 281 (1986).

Page 12 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(15)

For Peer Review Only

[9] J.P. Malugani, A. Wasniewski, M. Doreau, G. Robert, and A. Al Rikabi, Mat. Res. Bull. 13 427 (1978).

[10] G. Chiodelli, G. Campari Vigano, G. Flor, A. Magistris, and M. Villa, Solid State Ionics 8 311 (1983).

[11] M. Hosono, J. Kawamura, H. Itoigawa, N. Kuwata, T, Kamiyama, and Y. Nakamura, J. Non-Cryst. Solids 244 81 (1999).

[12] D. Ravaine and J.L. Souquet, Phys. Chem. Glasses 19 115 (1978).

[13] A.M. Glass and K. Nassau, J. Appl. Phys. 51 3756 (1980) [14] J.P. Malugani and R. Mercier, Solid State Ionics 13 293 (1984).

[15] A. Fontana, F. Rocca, and M.P. Fontana, Phys. Rev. Lett. 58 503 (1987).

[16] M.D. Ingram, Philos. Mag. B 60 729 (1989).

[17] P. Maass, A. Bunde, and M.D. Ingram, Phys. Rev. Lett. 68 3064 (1992).

[18] G.N. Greaves, J. Non-Cryst. Solids 71 203 (1985).

[19] T. Minami, J. Non-Cryst. Solids 73 273 (1985).

[20] K. Funke, Prog. Solid State Chem. 22 111 (1993).

[21] E.D. Crozier, J.J. Rehr, and R. Ingalls, in X-ray Absorption: Principles, Applications, Techniques of EXAFS, SEXAFS and XANES, D.C. Koningsberger and R. Prins editors, John Wiley and Sons, New York (1988).

[22] R. Kohlmuller and J.P. Faurie, Bull. Soc. Chim. Fr. 11 4379 (1968).

[23] G. Dalba, P. Fornasini, and F. Rocca, Phys. Rev. B 47 8502 (1993).

[24] G. Bunker, Nucl. Instrum. Methods Phys. Res. 207 437 (1983).

[25] P. Fornasini, F. Monti, and A. Sanson, J. Synchrotron Rad. 8 1214 (2001).

[26] D.S. Yang, D.R. Fazzini, T.I. Morrison, L. Trger, and G. Bunker, J. Non-Cryst. Solids 210 275 (1997).

[27] J.B. Boyce, T.M. Hayes, W. Stutius, and J.C. Mikkelsen Jr, Phys. Rev. Lett. 38 1362 (1977).

[28] J.B. Boyce, T.M. Hayes, and J.C. Mikkelsen Jr, Phys. Rev. B 23 2876 (1981).

[29] A. Filipponi, J. Phys.: Condens. Matter 13 R23 (2001).

[30] A. Trapananti, A. Di Cicco, and M. Minicucci, Phys. Rev. B 66 14202 (2002).

[31] J. Swenson, R.L. McGreevy, L. B¨orjesson, J.D. Wiks, and W.S. Howells, J. Phys.: Condensed Matter 8, 3545 (1996).

[32] G. Dalba, P. Fornasini, R. Gotter, and F. Rocca, Phys. Rev. B 52 149 (1995).

[33] J. Kawamura and M. Shimoji, J. Non-Cryst. Solids 88 281 (1986).

[34] S. Bhattacharya and A. Ghosh, J. Phys.: Condens. Matter 17 5655 (2005).

[35] G. Chiodelli, G. Campari Vigan, G. Flor, and A. Magistris, Solid State Ionics 8 311 (1983).

[36] D.L. Sidebottom, Phys. Rev. Lett. 83 983 (1999).

Page 13 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60