Electrical behavior of some rare-earth-doped Nd0.33Ln0.34Sr0.33MnO3 manganites

Electrical behavior of some rare-earth-doped Nd0.33Ln0.34Sr0.33MnO3 manganites

ARTICLE IN PRESS Journal of Magnetism and Magnetic Materials 309 (2007) 237–243 www.elsevier.com/locate/jmmm Electrical behavior of some rare-earth-...

291KB Sizes 0 Downloads 16 Views

ARTICLE IN PRESS

Journal of Magnetism and Magnetic Materials 309 (2007) 237–243 www.elsevier.com/locate/jmmm

Electrical behavior of some rare-earth-doped Nd0.33Ln0.34Sr0.33MnO3 manganites K. Padmavathi, G. Venkataiah, P. Venugopal Reddy Department of Physics, Osmania University, Hyderabad 500 007, India Received 8 April 2006; received in revised form 20 May 2006 Available online 31 July 2006

Abstract A series of rare-earth-doped Nd–Sr–Mn–O manganites having the compositional formula Nd0.33Ln0.34Sr0.33MnO3, (where Ln ¼ Gd, Eu, Sm, Nd and Ce) were prepared by citrate-based sol–gel route with ethylene glycol as a gelating reagent. After characterizing the samples by X-ray diffraction and scanning electron microscopy, a systematic study of magnetic susceptibility, electrical resistivity and magnetoresistance has been carried out over a temperature range 80–300 K. The variation of both electrical and magnetic transition temperatures with average ionic radii of A-site cation is explained. The electrical resistivity data were analyzed using different theoretical models and it has been concluded that at low temperatures (ferromagnetic metallic region) the resistivity may originate from grain/domain boundary, electron–electron scattering and two-magnon scattering effects, while in the paramagnetic insulating regime, the variation of resistivity with temperature may be explained by adiabatic small polaron and variable-range hopping mechanisms. The values of activation energies are found to decrease, while the density of states at the Fermi-level, N(EF) are increasing with increasing /rAS. A suitable explanation for the observed behavior is given. r 2006 Elsevier B.V. All rights reserved. Keywords: Resistivity; Magnetoresistance; Sol-gel process; Small polaron; Magnon

1. Introduction The perovskite manganite material, AMnO3 (A ¼ trivalent rare-earth ion) is an antiferromagnetic insulator. When A-site ion is replaced by a divalent ion, A0 (A0 ¼ Ca, Sr, Ba, Pb etc.), the resulting compound, depending on the concentration (0.2oxo0.5), becomes a ferromagnetic material with metallic nature. In the presence of an external magnetic field, these materials exhibit a substantial decrease of electrical resistivity and the phenomena is known as the ‘Colossal Magneto Resistance’ (CMR) [1]. These materials have been extensively studied during the last 10 years not only to understand the basic physical phenomena but also due to their potential applications in electrical and electronic technologies. The cause for simultaneous occurrence of ferromagnetism and metallic behavior in the manganites at temperatures, TpTC, was explained by Zener’s double-exchange (DE) mechanism [2]. Corresponding author. Tel.: +91 40 27682287; fax: +91 40 27009002.

E-mail address: [email protected] (P. Venugopal Reddy). 0304-8853/$ - see front matter r 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jmmm.2006.07.006

Although, there is an agreement that DE mechanism is successful in explaining the electrical resistivity behavior in the ferromagnetic metallic (FMM) region, it has been emphasized that large changes in resistivity in the paramagnetic region could not be explained within its framework. Meanwhile, there is an evidence from both the experiments and theoretical calculations that the polaron formed due to strong Jahn–Teller (JT) interactions, plays a major role in explaining the underlying Physics in these materials [3] especially in the paramagnetic regime. Therefore, it is very much essential to understand the scenario of complex electrical conduction mechanism clearly in terms of polarons as function of temperature, magnetic field and also the average ionic radius of A-site, /rAS. Further, the foremost obstacle in the field of CMR manganites is their use in electronic devices. From the applications point of view, particle size of the material and microcrystalline structure, etc., are the key parameters. Unfortunately, these parameters are highly affected by the preparation routes along with heat treatments. In this context, it is noteworthy that, CMR materials prepared by

ARTICLE IN PRESS 238

K. Padmavathi et al. / Journal of Magnetism and Magnetic Materials 309 (2007) 237–243

the conventional ceramic methods are not suitable for advanced and technological applications, as very often materials produced by this method are having faulty homogeneity, secondary phases apart from large particle size. Alternatively, the materials prepared by sol–gel routes are of high quality, homogeneous and fine-particle materials [4]. It is important to note that A-site cations with different sizes can produce a JT distortion of MnO6 octahedron thereby influencing the transport properties. For example, while keeping the level of hole-doping constant, change in /rAS by A-site substitution with different rare earths, results in significant variation of Mn–O–Mn bond angles, which in turn causes change in hybridization between Mn3d and O-2p orbital states. Further, rare-earth substitution also microscopically relates to the transfer integral b describing electron hopping between Mn sites, which in turn influences the DE [5,6]. Therefore, the variation of Asite cation radius /rAS, irrespective of the individual rareearth element influences various magnetic properties including magnetoresistance (MR). In view of these facts, an attempt has been made to explain the electrical behavior of sol–gel prepared Nd-based manganites with compositional formula Nd0.33Ln0.34Sr0.33MnO3 (Ln ¼ Gd, Eu, Sm, Nd and Ce) and the results of such an investigation are presented here. 2. Experimental 2.1. Preparation Polycrystalline materials with the compositional formula, Nd0.33Ln0.34Sr0.33MnO3 (Ln ¼ Gd, Eu, Sm, Nd and Ce) were prepared by the sol–gel method. In this method, the stoichiometric amounts of metal nitrates were taken and citric acid was added to convert them into metal citrates. After getting a sol on slow evaporation, a gelating reagent—ethylene glycol was added and heated between 160 and 180 1C to get a gel. This solution on further heating yields a dry fluffy porous mass (precursor), which was

calcined at 700 1C followed by sintering at 1000 1C. More details are given in an earlier publication [7]. 2.2. Characterization All the materials were characterized by X-ray diffraction (XRD) studies using Philips (Expert) diffractometer with Cu Ka radiation. The morphology of the materials was studied by using a scanning electron microscope (SEM) (Philips-FEIXL30 ESEM). The magnetic transition temperatures (TC) were obtained by measuring AC susceptibility (w0 ) over a temperature range 80–300 K, while the metal–insulator transition temperature (TP) values were arrived at using the electrical resistivity measurements. Finally, the MR measurements were also performed using a JANI’S ‘supervaritemp’ cryostat in applied magnetic fields ranging from 1 to 7 T over the temperature range 80–300 K using four-point probe technique. 3. Results and discussion 3.1. General The XRD patterns of all the samples were recorded at room temperature and data were analyzed using the least squares refinement technique. It has been found that all the materials of present investigation, with the exception of cerium-doped one, are having single-phase and orthorhombic structure with Pbnm space group. However, in the case of Ce-doped material (NCSMO), the XRD pattern is found to have extra peaks. After close observation and analysis, it has been concluded that the extra peaks may be due to unreacted CeO2. Fig. 1 shows the XRD pattern of NCSMO and the extra peaks were indexed with star (*) mark. In fact, the observed behavior is in conformity with the conclusion arrived at by several investigators that it is very difficult to prepare a single-phase cerium-doped manganites even by the well-known preparation techniques [8]. Finally, the unit cell parameters of all the samples were obtained and are given in Table 1.

Fig. 1. XRD pattern of Nd0.33Ce0.34Sr0.33MnO3 (NCSMO) sample, the star mark (*) in the pattern indicate the presence of CeO2 impurity peak.

ARTICLE IN PRESS K. Padmavathi et al. / Journal of Magnetism and Magnetic Materials 309 (2007) 237–243

The SEM images of all the samples were recorded and the average grain sizes were found to be in 1–3 mm range except in the case of NCSMO. In the case of NCSMO, the grain sizes are not uniform and are found to be large when compared with those of other samples (Fig. 2) and this may be due to the presence of unreacted CeO2. 3.2. Magnetic and electrical behavior A systematic investigation of AC susceptibility as a function of temperature has been undertaken and the ferroto paramagnetic transition temperature (TC) values were determined from the inflection point of dw0 /dT versus T(K) curves (Table 1). The TC values of all the samples, except in the case of Ce-doped one, are found to increase with increasing ionic radius of the A-site cation. Further, electrical resistivity measurements were also carried out to obtain the metal–insulator transition temperatures (TP) over a temperature range 80–300 K and are given in Table 1. It can be seen from the table that there is difference between values of TC and TP and as a matter of fact, a similar difference between these two transition temperatures was reported earlier in the case of several sol–gelprepared samples [9,10]. The observed behavior may be attributed to the following reasons. One possibility is the surface-strain-induced grain boundary effects [11], while the second one could be due to the phase-separation phenomenon. According to the phase-separation phenomenon, the two phases viz., antiferromagnetic insulating (AFMI) and FMM phases coexists just below TC. Due to the presence of AFMI regions near the grain boundaries and as TC values are governed by only ferromagnetic core, TC values might have not been modified. However, AFMI regions influence the electrical resistivity resulting in shifting of the metal–insulator transition temperature (TP) to lower-temperature side [9]. It is clear from the table that the values of both TC and TP are found to increase with increasing /rAS (except in the case of NCSMO) and the observed behavior may be explained as follows. When the value of /rAS becomes small, the oxygen ions tend to move towards the center of MnO6 octahedra by reducing Mn–O bond distances and Mn–O–Mn bond angle by causing distortion in the lattice. This lattice distortion provides a local trap for the eg electron and causes possible phase or domain separation.

239

As a result, hopping amplitude of the carriers moving from Mn3+ to Mn4+ decreases, which in turn causes local lattice distortion of MnO6 octahedra [12]. Therefore, the values of TP and TC increase with increasing /rAS. In contrast to the general trend of increasing of TP with increasing ionic radius, the TP value of NCSMO sample is found to decrease to 180 K. It is clear from the XRD and SEM analyses that the unreacted CeO2 is present in between the regions of the grains. This may increase the residual resistivity of the material, which in turn decreases the density of metallic particles, thereby enhancing the electrical resistivity. However, as CeO2 does not affect the magnetic core present in the sample, TC might have not been affected significantly when compared with TP. Further, the electrical resistivity measurements were also carried out in magnetic fields of 1–7 T in the temperature range 70–300 K and a typical plot of NSSMO is shown in Fig. 3. It can be seen from the figure that the resistivity is

Fig. 2. Typical SEM images of (a) NCSMO and (b) NSSMO samples.

Table 1 Experimental data of some CMR materials Sample code

NGSMO NESMO NSSMO NNSMO NCSMO

Compositional formula

Nd0.33Gd0.34Sr0.33MnO3 Nd0.33Eu0.34Sr0.33MnO3 Nd0.33Sm0.34Sr0.33MnO3 Nd0.33Nd0.34Sr0.33MnO3 Nd0.33Ce0.34Sr0.33MnO3

Lattice parameters a (A˚)

b (A˚)

c (A˚)

5.4297 5.4400 5.4388 5.4622 5.7276

5.4272 5.4320 5.4375 5.4286 5.4574

7.6849 7.6882 7.6866 7.6856 7.7338

/rAS (A˚)

TC (K)

TP (K)

DT ¼ TC–TP

Max. MR (%)

1.192 1.197 1.201 1.212 1.223

160 185 217 265 252

145 160 190 210 180

15 25 27 55 72

90 89 71 61 63

ARTICLE IN PRESS K. Padmavathi et al. / Journal of Magnetism and Magnetic Materials 309 (2007) 237–243

240

found to decrease with increasing field and that TP values are shifting towards high-temperature side. The observed behavior may be due to the fact that the applied magnetic field induces delocalization of charge carriers, which in turn suppresses the resistivity causing local ordering of spins. Due to this ordering, the FMM state may suppress the paramagnetic insulating (PMI) regime thereby resulting in complete polarization of conduction electrons (e1g) inside the magnetic domains and are easily transferred between the pairs of Mn3+ and Mn4+ via oxygen. Due to these reasons, TP shifts towards high-temperature side with the application of magnetic field [13]. 3.3. Magnetoresistance With a view to understand the variation of magnetoresistance (MR) with /rAS, the percentage of MR is calculated for all the samples by the formula rð0Þ  rðHÞ MR% ¼  100, (1) rð0Þ where r(0) is the resistivity at zero magnetic field and r(H) is the resistivity in a magnetic field. The calculated MR values of all the samples are given in Table 1. The MR values are found to be high in the vicinity of TP. Further, it can also be seen from the table that MR values are decreasing with increasing ionic radius of the A-site cation. It is also interesting to note that the samples with low TC exhibit large MR while those with high TC exhibit low MR and the behavior is in conformity with the universal MR–TC relationship [14]. 3.4. Conduction mechanism Inspite of extensive experimental and theoretical work to understand the conduction mechanism of CMR materials

in general and rare-earth manganites in particular, the present scenario is more confusing than earlier. Therefore, an effort has been made to understand the conduction mechanism of these materials by analyzing the experimental data of both the ferro-as well as the paramagnetic regions using various theoretical models. 3.4.1. Low-temperature behavior (ToTP) The relative strengths of the different scattering mechanisms originating from different contributions to the resistivity have been investigated by fitting the resistivity data at ToTP to various empirical equations [13,15–17]. It has been concluded that the electrical resistivity data of the samples of the present investigation are found to fit well with the equation r ¼ r0 þ r2 T 2 þ r4:5 T 4:5 ,

(2)

where r0 arises due to the grain or domain boundary scattering [16,17]. As the polycrystalline materials contain grain boundaries, their significant contribution to the resistivity is proved [18] and hence the term r0 plays a major role in the conduction process. On the other hand, r2T2 term explains the contribution of electron–electron scattering process to the resistivity, while the term r4.5T4.5 may be attributed to two-magnon scattering process in the ferromagnetic region [19]. As a matter of fact, it was concluded that the two-magnon scattering process is more favorable in half-metallic band structured materials such as manganites [20]. A typical plot of resistivity versus temperature of NNSMO along with its best fit to the equation r ¼ r0 þ r2 T 2 þ r4:5 T 4:5 are shown in Fig. 4. The best-fit parameters r0, r2 and r4.5 for all the samples, both in presence and absence of magnetic fields, are given in Table 2 and it is clear that all the three parameters are found to decrease with increasing magnetic field and the observed behavior may be explained as follows. When

25 NSSMO

7.5 NNSMO

20

15

10

ρ(Ωcm)

ρ(Ωcm)

6.0 0T 1T 3T 4T 5T 6T 7T

4.5

0T 3T 7T

3.0

1.5

5

75 100

150

200 T(K)

250

300

Fig. 3. A plot of resistivity (r) versus temperature (T) of Nd0.33Sm0.34Sr0.33MnO3 (NSSMO) sample at different magnetic fields.

100

125

150 T(K)

175

200

225

Fig. 4. A typical plot resistivity (r) versus temperature (T) of NNSMO sample at different magnetic fields and the solid line represents best to the equation r ¼ r0 þ r2 T 2 þ r4:5 T 4:5 .

ARTICLE IN PRESS K. Padmavathi et al. / Journal of Magnetism and Magnetic Materials 309 (2007) 237–243

magnetic field increases, the domain becomes large reducing the value of r0, while the reduction in r2 and r4.5 could be attributed to the decrease of electron spin fluctuations in the presence of magnetic field. Intrinsically, the scattering effects are suppressed from various contributions because of the parallel configuration of the spins present in the domain [15] and as such all the contributing parameters viz., r0, r2 and r4.5 may decrease with the application of magnetic field [16]. Further, the values of r0 are also found to decrease continuously with increasing /rAS, except in the case of NCSMO, while there is no systematic variation of r2 and r4.5 values with /rAS. Therefore, it has been concluded that the values of r2 and r4.5 might not be sensitive to the varying values of /rAS. 3.4.2. High-temperature behavior (T4TP) The variation of electrical resistivity with temperature (above TP) may be explained on the basis of two different models. To explain the electrical conduction just above TP, i.e., TPoToyD/2, (yD is Debye’s temperature) variablerange hopping (VRH) model has been suggested, while small polaron hopping (SPH) model is considered at temperatures beyond yD/2. In the latter case, a polaron can be thought to be trapped inside a local energy well of height Ea and when the field is applied, one side of the well is lowered slightly with respect to the other. This makes the polaron to hop in that direction [21]. The high-temperature resistivity data (TPoToyD/2) were fitted to the equation [22] s ¼ s0 exp ðT 0 =TÞ1=4 ,

(3)

241

where T0 ¼ 16a3/kBN(EF) and N(EF) is the density of states at the Fermi level. Here T0 values have been obtained using the slope of ln(s) versus T1/4 plots. Finally, the values of N(EF) have been obtained using the values of T0 by taking a as a constant [13]. The computed values of T0 and N(EF) are given in Table 3 and are found to be in agreement with the reported ones [13,15,20]. The values of T0 are decreasing continuously with increasing /rAS, whereas the values of N(EF) are found to increase continuously. Generally, it is expected that higher T0 values indicate an increase in bending of Mn–O–Mn bond which in turn reflects the enhancement of carrier effective mass or the narrowing of the bandwidth, resulting in a drastic change in the resistivity and sharpening the resistivity peak in the vicinity of TP. Further, for a given material, as T0 values are smaller in the presence of field, the density of state values at the Fermi level are expected to increase. This could be due to the suppression of magnetic domain scattering with the application of field [15]. The conduction mechanism of manganites at high temperatures, T4yD/2 is governed by thermally activated small polarons and polaronic models. These polaronic models could be due to either adiabatic or non-adiabatic approximations [23] and are given by r ¼ ra T exp ðE P =kB TÞ ðadiabaticÞ,

(4)

r ¼ ra T 3=2 exp ðE P =kB TÞ

(5)

ðnonadiabaticÞ,

where EP is the activation energy and ra is the residual resistivity and is given by ra ¼ 2kB =3ne2 a2 v

(6)

Table 2 The best-fit parameters obtained from low-temperature (ToTP) resistivity data Sample code

NGSMO NESMO NSSMO NNSMO NCSMO

r2 (  104 Ocm K2)

r0 (Ocm)

r4.5 (  1010 Ocm K4.5)

B ¼ 0T

B ¼ 3T

B ¼ 7T

B ¼ 0T

B ¼ 3T

B ¼ 7T

B ¼ 0T

B ¼ 3T

B ¼ 7T

7.0724 5.8750 4.2978 1.5445 194.7795

2.7169 0.7394 2.1941 0.8881 78.008

0.4991 0.2164 1.6215 0.6501 48.6132

11.20 3.40 5.40 1.10 114.20

3.00 2.20 4.10 1.00 106.60

2.00 1.10 2.30 0.70 66.30

138.80 37.60 2.32 0.78 78.97

22.09 0.69 2.28 0.57 65.61

0.44 0.45 1.83 0.23 63.24

Table 3 The best-fit parameters obtained from high-temperature (TPoT) resistivity data at different fields Sample code

NGSMO NESMO NSSMO NNSMO NCSMO

T0 (  106 K)

N(EF) (  1021 eV1 cm3)

EP (meV)

B ¼ 0T

B ¼ 3T

B ¼ 7T

B ¼ 0T

B ¼ 3T

B ¼ 7T

B ¼ 0T

B ¼ 3T

B ¼ 7T

18.2009 4.2709 2.3843 1.2351 0.4243

2.4194 3.0048 1.1854 0.3694 0.1611

1.1442 1.3196 0.8761 0.1524 0.1108

0.1115 0.4752 0.8512 1.6433 4.7838

0.8389 0.6754 1.7122 5.4937 12.6003

1.7739 1.5381 2.3167 13.3167 18.3129

160.0347 148.1359 143.5416 113.6947 61.4812

150.3233 135.0041 119.8541 107.2288 57.9943

116.5498 132.1762 92.1123 86.7961 55.9961

ARTICLE IN PRESS K. Padmavathi et al. / Journal of Magnetism and Magnetic Materials 309 (2007) 237–243

here kB is Boltzmann’s constant, e is electron charge, n is number of density of charge carriers, a is site-to-site hopping distance and v is longitudinal optical phonon frequency. It has been explained by Jung [24] that higher (two to three orders higher than those of oxide semiconductors) density of state values at Fermi level, N(EF) could be due to their high value of conductivity and that the higher values of N(EF) are clear signatures of the applicability of adiabatic hopping mechanism. As N(EF) values of the samples of the present investigation are comparable with those reported by Jung, the electrical resistivity data, in the temperature region T4yD/2, were fitted to the adiabatic small polaron-hopping model represented by the equation, r ¼ raT exp (EP/kBT) and from the best fits, the activation energy values were calculated. Further, yD values were also estimated from the plots of ln (r/T) versus (1/T) (Fig. 5) by taking the deviation from linearity in the low-temperature region of this plot as equal to yD/2. The activation energy values, (EP) are also included in Table 3 and one can note that all these parameters are in agreement with the reported ones [12,13]. It is also clear from the table that the

0.004

0.006

T-1(K-1) 0.008 0.010

0.012

0.014

-2.0 -2.5 NSSMO

-3.0 -3.5

0T

-4.0 -4.5

160

140

EP (meV)

242

120

100

80

60 1.192

1.200

1.208

1.216

1.224

(Å) Fig. 6. Variation of activation energy, EP(meV) with /rAS(A˚) in zero magnetic field.

activation energies are found to decrease continuously with increasing value of /rAS. The variation of EP with /rAS is found to be almost linear and is shown in Fig. 6 and the observed behavior may be due to the fact that when /rAS decreases, conduction bandwidth narrows increasing effective band gap thereby enhancing the activation energies with decreasing /rAS. Similarly, the values of EP are decreasing with increasing applied magnetic field and the behavior may be attributed to the decrease in the values of charge localization under the influence of magnetic field [12]. 4. Conclusions

-2.5

As TC or TP values are found to increase with increasing /rAS, it has been concluded that manganites doped with the rare earths having higher ionic radii are more useful. As NGSMO and NESMO materials exhibit highest MR of E90%, one may conclude that these two materials may be exploited as sensors for some practical applications.

ln(ρ / T)

-3.0 -3.5 -4.0 3T -4.5

Acknowledgments The authors are grateful to DRDO, Government of India for sponsoring a research Project no. ERIP/ER/ 0103326/M/01. The authors also thank Dr. V. Prasad, Indian Institute of Science, Bangalore for providing facilities to undertake magnetoresistance measurements. One of the authors, Mrs. K. Padmavathi thanks UGC, Government of India for providing teaching fellowship.

-3.5

-4.0 7T -4.5

0.004

0.006

0.008

0.010

0.012

0.014

T-1(K-1) Fig. 5. Variation of ln(r/T) versus inverse temperature (1/T) of NSSMO sample at different magnetic fields. The solid line gives the best fit to the equation r ¼ raT exp (EP/kBT).

References [1] S. Jin, T.H. Tiefel, M. Mc Cormack, R.A. Fastnacht, R. Ramesh, L.H. Chen, Science 264 (1994) 413. [2] C. Zener, Phys. Rev. 82 (1951) 403.

ARTICLE IN PRESS K. Padmavathi et al. / Journal of Magnetism and Magnetic Materials 309 (2007) 237–243 [3] A.J. Millis, P.B. Littlewood, B.I. Shraiman, Phys. Rev. Lett. 74 (1995) 5144. [4] P.N. Lisboa-Filho, A.W. Mombru, H. Pardo, E.R. Leite, W.A. Ortiz, Solid State Commun. 130 (2004) 31. [5] H.Y. Hwang, S.-W. Cheong, P.R. Radaelli, M. Marezio, B. Batlogg, Phys. Rev. Lett. 75 (1995) 914. [6] L.M. Rodriguez-Martinez, J.P. Attfield, Phys. Rev. B 54 (1996) R15622. [7] G. Venkataiah, D.C. Krishna, M. Vithal, S.S. Rao, S.V. Bhat, V. Prasad, S.V. Subramanyam, P. Venugopal Reddy, Physica B 357 (2005) 370. [8] C. Mitra, P. Raychaudhuri, J. John, S.K. Dhar, A.K. Nigam, R. Pinto, J. Appl. Phys. 89 (2001) 524. [9] Y.H. Huang, C.H. Yan, Z.M. Wang, C.S. Liao, G.X. Xu, Solid State Commun. 118 (2001) 541. [10] A. Dutta, N. Gayathri, R. Ranganathan, Phys. Rev. B 68 (2003) 054432. [11] J. Rivas, L.E. Hueso, A. Fondado, F. Rivadulla, M.A. LopezQuintela, J. Magn. Magn. Mater. 221 (2000) 57. [12] S. Bhattacharya, R.K. Mukherjee, B.K. Chaudhuri, Appl. Phys. Lett. 82 (2003) 4101. [13] A. Banerjee, S. Pal, B.K. Chaudhuri, J. Chem. Phys. 115 (2001) 1550.

243

[14] A.P. Ramirez, J. Phys.: Condens. Matter. 9 (1997) 8171. [15] G. Venkataiah, P. Venugopal Reddy, J. Magn. Magn. Mater. 285 (2005) 343. [16] G.J. Snyder, R. Hiskers, S. DiCarolis, M.R. Beasley, T.H. Geballe, Phys. Rev. B 53 (1996) 14434. [17] J.M. De Teresa, M.R. Ibarra, J. Blasco, J. Garcı´ a, C. Marquina, P.A. Algarabel, Z. Arnold, K. Kamenev, C. Ritter, R. von Helmolt, Phys. Rev. B 54 (1996) 1187. [18] M. Dominguez, S.M. Bhagat, S.E. Lofland, J.S. Ramachandran, C.G. Xiong, H.L. Ju, R.L. Greene, T. Venkatesan, Eur. Phys. Lett. 32 (1995) 349. [19] A. Urushibara, Y. Moritomo, T. Arima, A. Asamitsu, G. Kido, Y. Tokura, Phys. Rev. B 51 (1995) 14103. [20] G. Venkataiah, P. Venugopal Reddy, Solid state Commun. 136 (2005) 114. [21] D.C. Worledge, G.J. Snyder, M.R. Beasley, T.H. Geballe, J. Appl. Phys. 80 (1996) 5158. [22] N.F. Mott, E.A. Davis, Electronic Process in Noncrystalline Materials, Clarendon, Oxford, 1971. [23] D. Emin, T. Holstein, Ann. Phys. 53 (1969) 439. [24] W.H. Jung, J. Mater. Sci. Lett. 17 (1998) 1317.