Magnetic inhomogeneity and Griffiths phase in Bi ...

Physica B 498 (2016) 82–91

Contents lists available at ScienceDirect

Physica B journal homepage: www.elsevier.com/locate/physb

Magnetic inhomogeneity and Griffiths phase in Bi substituted La0.65
xBixCa0.35MnO3 manganites S.O. Manjunatha a,b, Ashok Rao a,n, P. Poornesh a, W.J. Lin c, Y.-K. Kuo c,n a

Department of Physics, Manipal Institute of Technology, Manipal University, Manipal 576104, India Department of Physics, M.S. Ramaiah University of Applied Sciences, Bengaluru 560058, India c Department of Physics, National Dong-Hwa University, Hualien 974, Taiwan b

art ic l e i nf o

a b s t r a c t

Article history: Received 3 November 2015 Received in revised form 10 May 2016 Accepted 28 June 2016 Available online 29 June 2016

In the present communication, we report the effect of Bi substitution on structural, magneto-transport, magnetic and thermal properties of La0.65
xBixCa0.35MnO3 (0 rx r 0.2) compounds. Rietveld refined XRD patterns confirm that all the samples are single phase and crystallize in rhombohedral symmetry with R3C space group. It was observed from electrical and magnetic studies that with increasing Bi content both the metal-insulator transition temperature (TMI) and ferromagnetic–paramagnetic (FM–PM) transition temperature (TC) shift towards lower temperatures. Deviation of the temperature dependence of inverse susceptibility curves from the Curie–Weiss law suggests the existence of Griffiths-like phase. The electrical resistivity data were analyzed by utilizing various theoretical models. It is revealed that the electron–electron scattering is dominating in the metallic region, while the insulating region is well-described by the polaron hopping model. Analysis of thermoelectric power data further reveals that the small polaron hopping (SPH) mechanism is operative in the high-temperature insulating regime. The entropy change associated with the FM–PM transition is found to decrease with increasing x, which is presumably due to the increase in magnetic inhomogeneity with increasing Bi content. & 2016 Elsevier B.V. All rights reserved.

Keywords: Manganites Griffiths phase Thermoelectric power

1. Introduction Perovskite manganites with general formula R1
xAxMnO3 (R is trivalent rare earth and A is divalent alkaline earth) have been extensively investigated due to the unusual electrical and magnetic properties they exhibit [1–4]. In addition to their peculiar physical properties, the existence of inhomogeneity and preformation of ferromagnetic clusters in the paramagnetic region, known as Griffiths phase (GP), have attracted considerable interest [5–8]. The Griffiths phase in manganites arises at a characteristic temperature TG in which the Curie–Weiss law is inapplicable in the temperature range TC o To TG. The system exhibits neither pure paramagnetic phase nor the long-range FM order in this range of temperature. Usually A-site substitution is one the main factors for the existence of GP in manganites [5,9–12]. However, there are other factors, viz. Jahn–Teller distortion [9,13] and grain size effect [14] which could also lead to the appearance of Griffiths phase in manganites. Salmon et al. proposed that the colossal magnetoresistance property of manganites is a Griffiths singularity [5]. On n

Corresponding authors. E-mail addresses: [email protected] (A. Rao), [email protected] (Y.-K. Kuo). http://dx.doi.org/10.1016/j.physb.2016.06.031 0921-4526/& 2016 Elsevier B.V. All rights reserved.

the other hand, Pramanik et al. argued that the quenched disorder is prerequisite for the formation of GP [6]. Recently, Dayal et al. observed an evolution of Griffiths phase around TC while Ti was substituted at Mn site in the La0.4Bi0.6MnO3 system. The presence of GP is attributed to the increase in magnetic inhomogeneity with Ti substitution [7]. The size of the ferromagnetic clusters in the Griffiths phase region can be diverged by application of an external magnetic field which results in the sharp increase in the magnetization [15]. Jiang et al. [15] have reported an extreme sensitivity of GP to the applied magnetic field in single crystal La0.73Ba0.27MnO3. They have reported that the depression of χ
1 vs T curves below the GP is completely quenched on application of only 150 Oe of static magnetic field. Such sensitivity can be used for applications in magnetic sensors. Nevertheless, there are systems which exhibit colossal magnetoresistance (CMR) effect without the presence of GP [13,16]. Therefore, the physical mechanisms of the GP need to be further investigated. Recently, the effect of Bi þ 3 substitution on a variety of properties in La1
xCaxMnO3 (LaCaMnO) system has been studied [17– 21]. It is well-known that BiMnO3 exhibits multiferroic nature where phases like ferroelectric, ferromagnetic and ferroelastic are coexisting in this material. It is also important to mention that Bisubstituted compounds exhibit a high charge ordering temperature (TCO  475 K) [22]. Gencer et al. and Atalay et al. have

S.O. Manjunatha et al. / Physica B 498 (2016) 82–91

proposed that substituting Bi in LaCaMnO system introduces excellent magnetocaloric properties with a high relative cooling power [17,18]. However, there are relatively few reports in exploring the existence of Griffiths phase in the Bi-substituted LaCaMnO systems. In this regard, we have performed a throughout investigation on the structural, magnetic, magneto-transport and thermoelectric properties of Bi-substituted La0.65
xBixCa0.35MnO3 (0 rx r0.2) system. We have also made an effort to account for the enhancement of the Griffiths-like phase in the paramagnetic region with increase in Bi concentration.

83

and thermal conductivity measurements were carried out in a closed cycle refrigerator using a heat-pulse technique. Specific heat measurements were performed in temperature range 90– 300 K using a high resolution ac calorimeter. The details of the thermal measurements techniques have been described elsewhere [23].

3. Results and discussion 3.1. Structure and surface morphology

2. Experimental The compounds La0.65
xBixCa0.35MnO3 (0 rx r0.2) [LaBiCaMnO] were prepared using solid state reaction method. Stoichiometric proportions of high purity (99.9% pure, Sigma-Aldrich) La2O3, Bi2O3, Ca2CO3 and MnO2 precursors were mixed homogeneously in an agate mortar for about 6 h to obtain a homogenous mixture. The mixture was calcined at 900 °C with two intermediate grindings to exhaust carbonates present in the mixture. The calcined mixtures were taken in the form of rectangular shaped pellets and were sintered at 1000 °C for 24 h. Finally, the samples were furnace cooled to room temperature. X-ray diffraction (XRD) studies were done using Bruker D8 Advance X-ray diffractometer. The surface morphology of the samples was examined using scanning electron microscopy (SEM) using Oxford EVO MA18. The temperature-dependent electrical resistivity and magnetoresistance measurements were performed using the standard four probe technique with a superconducting magnetic system (Oxford Spectromag) in the temperature range 5–300 K. Magnetic measurements were carried out using quantum design magnetic property measurement system (MPMS). Thermoelectric power

We have performed the X-ray diffraction studies to confirm the purity of the samples and determine the lattice parameters. Recorded XRD patterns were further analyzed by employing the Rietveld refinement technique using Fullprof program. The Rietveld refined XRD pattern for all the samples is depicted in Fig. 1 and the structural parameters along with Mn–O–Mn bond angle and bond lengths for all the samples are summarized in Table 1. There is a slight decrease in lattice parameters a and c with increase in Bi-content. In addition to this the cell volume also decreases with Bi-content. This trend in cell parameters can be attributed to the presence of magnetic polarons which are produced due to competing behavior of larger ionic radii generated ferromagnetic-double exchange and charge ordered state. The results indicate that the Mn–O–Mn bond angle decreases and Mn–O bond length increase with substituting smaller Bi ion. This may be due to successive substitution of Bi at the slightly larger La, resulting in reduced values A-site atomic radius. The Rietveld refinement parameters such as Rp, Rwp and Rexp are also presented in Table 1 and the estimated values for presently investigated samples suggest that the fittings are good. It is found from the Rietveld analysis that all the samples in the present work are single phase and crystallize in rhombohedral structure with R-3C space group.

Fig. 1. Rietveld refinement for the La0.65
xBixCa0.35MnO3 samples.

84

S.O. Manjunatha et al. / Physica B 498 (2016) 82–91

Table 1 The parameters deduced from the structural studies of the La0.65
xBixCa0.35MnO3 (0 r xr 0.2) series. Composition “x”

0.00

System Space group Lattice parameters a (Å) c (Å) V (Å)3 Mn–O–Mn bond angle Mn–O bond length Rf Rp Rwp Rexp χ2

Rhombohedral R-3C 5.447 13.339 342.8 159.85 1.956 7.69 6.49 8.9 5.96 1.01

0.05

5.454 13.333 342.6 159.84 1.954 5.44 5.05 7.01 4.00 3.08

0.10

0.20

5.452 13.331 342.3 159.84 1.953 5.08 4.52 6.00 3.75 2.57

5.441 13.327 341.8 159.81 1.951 8.52 6.07 8.88 3.98 2.98

Fig. 2 depicts the microstructure of the samples in the present investigation. It is seen that the grains of all the samples (except the sample with x¼ 0.20) are distinct and the grain boundaries are well-defined. However, sample with x¼ 0.20 shows an agglomerated distribution of grains. It is also observed that the grains size decreases with increase in Bi concentration. This is attributed to the increase in strain in the sample, as the addition of Bi restricts the growth of grains. 3.2. Magnetic studies It is well-known that the manganites exhibit a magnetic transition from ferromagnetic to paramagnetic phase at the Curie temperature TC. Therefore, a systematic investigation of

temperature-dependent magnetization for La0.65
xBixCa0.35MnO3 (0 rx r0.2) was undertaken in the temperature range 5–300 K. We have carried out magnetization measurements in zero field cooled (ZFC, without any external magnetic field) and field cooled (FC, with application of a constant magnetic field of 250 Oe) conditions and the results are depicted in Fig. 3. It is observed from Fig. 3 that as the temperature decreases from room temperature, an abrupt jump in magnetization is observed at a temperature corresponding to the Curie temperature TC. We have determined TC from the inflection point of temperature-dependent dM/dT plots as depicted in the insets of Fig. 3 and the values of TC are summarized in Table 2. It is observed that TC decreases with increase in Bi content. This is presumably due to the relatively smaller ionic size of Bi (compared to that of La) tilts the MnO6 octahedra, resulting in a reduced overlap between the Mn-3d and O-2p orbitals [24]. As discussed in last section, we have seen a decrease in bond angle with increase in Bi-content, which localizes the electrons and also causes possible phase or domain separation. It also hinders the hopping probability of carriers from Mn3 þ to Mn4 þ as the bond angle deviates from 180° and causes local lattice distortions of the MnO6 octahedra and reduces the mobility of the carriers [25]. This may possibly be the reason for shifting of TC towards lower temperature side. It is also noticed from Fig. 3 that there exists a bifurcation in the magnetization data measured in ZFC and FC conditions below TC. Such an observation is ascribed to the appearance of the spin glass or cluster glass state in these samples, suggesting the presence of magnetic inhomogeneity in the long range ferromagnetic ordering [7]. The lower values of measured magnetization of ZFC as compared to FC data may be due the fact that the movements of magnetic domains along the direction of magnetic field are restricted as a result of the pinning of magnetic domain walls [26,27]. The bifurcation feature is found to be more

Fig. 2. SEM micrographs of the La0.65
xBixCa0.35MnO3 (0r x r0.2) series.

S.O. Manjunatha et al. / Physica B 498 (2016) 82–91

85

Fig. 3. Temperature-dependent magnetization of the La0.65
xBixCa0.35MnO3 (0 rx r 0.2) compounds at 250 Oe. Insets show the corresponding dM/dT vs. T plots.

Table 2 The parameters deduced from the magnetic studies of the La0.65
xBixCa0.35MnO3 (0 rx r 0.2) series. Composition 'x'

TC (K)

Θ (K)

TG (K)

GP (TC %)

exp μeff ( μB )

thoe μeff ( μB )

MS (emu/g) (5 K)

0.00 0.05 0.10 0.20

263.8 246.4 240.5 201

274.1 243.2 238.8 225.8

270 260 258 272

2.29 5.23 6.78 26.10

5.92 6.21 6.98 6.42

4.6 4.6 4.6 4.6

99.32 90.74 117.75 96.44

distinct in the high Bi content samples, indicating that Bi substitution enhances the magnetic anisotropy due to the bending of MnO6 octahedra [28] which is consistent with the trends of bond angle and bond lengths. Alternatively, the increase in bifurcation with increasing Bi content may also be attributed to the decrease in grain size with Bi substitution. In this case, the decrease in grain size enhances the grain boundary effect due to the enhanced surface to volume ratio. This in turn disturbs the long range ferromagnetic order and the uncompensated spins start to account for the large irreversibility [7]. We have also verified the applicability of Curie–Weiss law to the samples studied under present investigation. The experimental data of all the samples were fitted to the Curie–Weiss law: χ ¼C/(T
Θ), where χ, Θ, and C are magnetic susceptibility, Curie– Weiss temperature and Curie constant, respectively. Fig. 4 depicts the linear fit of inverse susceptibility versus temperature. We have evaluated the parameters C and Θ by linear fitting the inverse susceptibility versus temperature plots above TG and the obtained parameters C and Θ are presented in Table 2. The values of Θ are

positive for all the samples, indicating the predominant ferromagnetic interactions in these samples. In addition, the inverse susceptibility shows a deviation from the linearity and a downturn behavior is observed. The onset of deviation from the linearity happens at a characteristic temperature TG and the values of TG for the studied samples are tabulated in Table 2. The Curie–Weiss law is inapplicable in the temperature range between TC and TG (TC oTo TG) where the system contains ferromagnetic clusters in the paramagnetic phase. This behavior indicates the possible existence of randomly distributed ferromagnetic clusters within the paramagnetic region which is known as Griffiths phase [29,30]. It is also crucial to mention that the temperature region over which the Griffiths phase is active increases with increasing Bi content as one can clearly see in Fig. 4 that the temperature region is significantly enhanced for the x¼ 0.20 sample. We have also evaluated the temperature range of GP normalized with TC's using the equation GP(TC%)¼[(TG
TC)/TC]  100. It is found that the GP(TC%) also increases with increasing Bi concentration, suggesting the increase in magnetic inhomogeneity near TC. Therefore, we can

86

S.O. Manjunatha et al. / Physica B 498 (2016) 82–91

C=

NAμB2 μeff 2 3kB

(1)

where NA (mol ) is Avogadro number, μB (emu) is Bohr magnetheo is ton, and kB (erg/K) is Boltzmann constant. The theoretical μeff estimated using the following equation [32]:
1

theo μeff =

(

)

(

0.65μeff 2 Mn3 + = 4.9μB + 0.35μeff 2 Mn4 + = 3.87μB

)

(2)

exp theo The parameters μeff and μeff are summarized in Table 2. It is theo exp is much greater as compared to μeff . This found that the μeff

Fig. 4. Inverse magnetic susceptibility (H/M) versus temperature for the samples La0.65
xBixCa0.35MnO3 (0 rx r 0.2). The solid lines represent the fits to the Curie– Weiss law.

conclude that the substitution of Bi further enhances the appearance of Griffiths phase in the LaCaMnO system. exp ) We can determine the effective paramagnetic moment ( μeff

discrepancy validates the formation of ferromagnetic spin clusters within the paramagnetic state [33]. Fig. 5 depicts the isothermal field-dependent magnetization measurements which were carried out in the range of 77 T at 5 and 200 K. All samples except x¼ 0.20 exhibit the characteristic behavior of long range ferromagnetic ordering as the magnetization rises piercingly at low magnetic fields and saturates rapidly at higher fields. On the other hand, the sample with x¼0.20 exhibits entirely different behavior. The M–H curve at 5 K shows irreversible magnetization effect whereas at 200 K the magnetization does not saturate even up to 7 T. This behavior in M–H curve is attributed to the existence of antiferromagnetic component coupling with the ferromagnetic state in the La0.45Bi0.2Ca0.35MnO3 sample [34].

from the Curie constant using the following equation [31]: 3.3. Resistivity measurement The

temperature-dependent

Fig. 5. Field-dependent magnetization for the La0.65
xBixCa0.35MnO3 (0 r x r0.2) series.

of

electrical

resistivity

for

S.O. Manjunatha et al. / Physica B 498 (2016) 82–91

87

Table 3 Coefficients of fits deduced from the analysis of electrical resistivity (at 0 T, 2 T, 8 T) using various theoretical models for the La0.65
xBixCa0.35MnO3 (0 r xr 0.2) series. Sample Metallic region

Fig. 6. Temperature-dependent electrical resistivity of the La0.65
xBixCa0.35MnO3 (0 r xr 0.2) compounds.

x ¼0.00 0T 2T 8T x ¼0.05 0T 2T 8T x ¼0.10 0T 2T 8T x ¼0.20 0T 2T 8T

SPH model (Ω-mm/K4.5) Eρ (meV)

ρ0 (Ωmm)

ρ2 (Ω-mm/K2) ρ4.5

0.58 0.39 0.29

5.64  10
5 5.18  10
5 3.03  10
5

1.07  10
11
2.01  10
12
1.61  10
12

0.22 0.15 0.10

1.56  10
5 1.35  10
5 8.13  10
6

0.35 0.24 0.16 1.11 0.29 0.18

Percolation model TC-mod (K)

U0/kB (K)

143.7 115.5 55.4

263 287.5 301.7

2.89  104 1.62  103 1.33  103

2.91  10
12 9.86  10
13 1.13  10
13

137.24 112.49 28.90

248.5 264.7 220.6

1.85  104 1.79  104 1.36  103

2.62  10
5 2.29  10
5 1.40  10
5

1.02  10
12
2.37  10
12
3.21  10
12

122.92 101.06 26.33

239.4 254.9 212.9

1.82  104 1.48  104 4.87  103

2.29  10
5 1.46  10
5 1.11  10
5

5.18  10
11 2.14  10
11 3.44  10
12

165.10 158.05 123.06

200.1 209.5 234.8

3.34  103 3.22  103 2.58  103

insulating region (T 4TMI), and percolation region (T  TMI). The nature of conduction mechanism in metallic region is analyzed using the following equation [27]

ρFM ( T ) = ρ0 + ρ2 T2 + ρ4.5 T 4.5

Fig. 7. Temperature-dependent electrical resistivity (symbol) and theoretical fit (solid line) using percolation model at different magnetic fields (0 T, 2 T, and 8 T) for a typical sample with x ¼0.05. Insets (a) and (b) show the fit in metallic region and insulating region respectively at different magnetic fields.

LaBiCaMnO manganites is illustrated in Fig. 6. It is seen that all samples exhibit a clear transition from paramagnetic insulating state to ferromagnetic metallic state at temperature TMI with lowering temperature from room temperature. The La0.65Ca0.35MnO3 sample (x¼ 0) exhibits TMI at around 265 K and it shifts towards lower temperatures with increasing Bi content (TMI 190 K for the sample with x ¼0.20). The decrease in TMI with increasing Bi content is presumably due to reduction in Mn3 þ –O– Mn4 þ bond angle which in turn reduces the double exchange interaction in the La0.65
xBixCa0.35MnO3 system. We have also studied the electrical resistivity in the presence of applied magnetic field of 2 T and 8 T to evaluate the CMR effect in these compounds. Fig. 7 displays the MR plot for a typical sample with x ¼0.05. It is seen that the applied magnetic field greatly suppresses the resistivity and shifts the TMI towards higher temperatures, thereby resulting in an increase in the metallic phase fraction. This is due to the alignment of localized t2g spins in the direction of applied magnetic field which increases the charge exchange integral between Mn3 þ and Mn4 þ ions via the intermediate oxygen ion. We now analyze the temperature-dependent electrical resistivity by considering three regions viz., metallic region (T oTMI),

(3)

where ρ0 is the residual resistivity due to grain boundary scattering, ρ2 arises from electron–electron scattering (e
–e
), and ρ4.5 is a resistivity factor due to assorted effects such as electronmagnon and electron-phonon scattering processes. Theoretical fitting to the experimental data for a typical sample with x ¼0.05 is shown in the inset (a) of Fig. 7 and the fitting parameters ρ0, ρ2, and ρ4.5 are listed in Table 3. A satisfactory agreement between the fit and measured data is clearly seen. It is noted from the fit that the contribution from the second term (ρ2T2) is considerably greater than the third term (ρ4.5T4.5), suggesting that the electron– electron scattering play a vital role in electrical conduction process in the metallic region. It is also important to note that the fitting parameters decrease with the application of magnetic field, in agreement with the MR studies. We have utilized the polaron hopping model to explain the temperature variation of electrical resistivity in the insulating region. The equation is given by

⎛ E ⎞ ρPM ( T ) = ρ0 T exp⎜ A ⎟ ⎝ kBT ⎠

(4)

where EA is the activation energy for hopping conduction and ρ0 is a proportionality constant. The activation energy is evaluated from a linear fit of ln(ρ/T) versus 1/T plot for a typical sample with x¼ 0.05 as depicted in the inset (b) of Fig. 7. The obtained EA values for all samples at various applied fields are given in Table 3. The activation energy is found to decrease with increase in Bi content (except for the x¼ 0.20 sample), most likely due to the fact that the Bi substitution hinders the double exchange mechanism and hence lowers the possibility of conduction electrons to hop to the neighboring sites [35]. It is well-known that pure ferromagnetic metallic (FMM) phase is realized well below TMI in manganites while they are in paramagnetic insulating (PMI) phase well above TMI. However, it is important to note that there exists a competition between the FMM and PMI phases near TMI and therefore none of the above models can explain the electrical resistivity in the entire

88

S.O. Manjunatha et al. / Physica B 498 (2016) 82–91

temperature range. A phenomenological model known as percolation model based on the phase segregation mechanism has been utilized to analyze the resistivity data in the temperature range near TMI. The model can be quantitatively described using the following equation,

ρ(T ) = ρFM (T )f + ρPM (T )( 1 − f )

(5)

where f is the ferromagnetic phase volume fraction. It is wellknown that volume fraction of FM and PM phases obey Boltzmann distribution,

f=

contribution arises due to the polarization of spins on application of external magnetic field. Extrinsic contribution on the other hand arises due to spin polarized tunneling between the grains across the grain boundaries [40]. The intrinsic MR in general has the maximum value at TC while the extrinsic MR increases with decrease in temperature. We can thus conclude that both the contributions are present in the total MR(%) of samples under present investigation. 3.4. Thermoelectric power (TEP)

1 1 + exp

( ) U kBT

⎛ where, U ≈ − U0⎜ 1 − ⎝

(6) T

⎞ ⎟ is the energy difference between FM

TCmod ⎠

and PM states and TCmod is the temperature in the neighborhood where resistivity has a maximum value. Fig. 7 depicts a good fit between experimental data (symbols) and the theoretical fit (solid line) at various magnetic fields for a typical sample with x ¼0.05. The fitting parameters TC-Mod and U0 are tabulated in Table 3. It is observed that the value of TC-Mod is close to TMI for all the samples which is expected theoretically. We have estimated MR(%) using the equation; MR(%) ¼ [(ρ0
ρH)/ρH]  100, where ρ0 and ρH are resistivity of the samples without and with the application of magnetic field, respectively. Fig. 8 depicts the variation of MR(%) versus temperature when the samples are subjected to external magnetic fields of 2 T and 8 T. It is observed that the MR(%) increases with increasing Bi content. Similar results have been reported in the literature [36]. The observed MR% for the samples in the present investigation are 57, 65, 68, 88% for x¼ 0, 0.05, 0.1, 0.2 respectively. Jiang et al. [37] have reported MRE70% in La0.594Bi0.066Ca0.33MnO3 sample. Srinivasan et al. [38] have reported MRE48% in La0.62Bi0.05Ca0.33MnO3 sample at 4 T. Xia et al. [19,39], in their various reports on Bi doped LCMO compounds have reported MR% of E50%, 100%, 100% for La0.67Ca0.33MnO3, La0.49Ca0.18Ca0.33MnO3 and La0.536Bi0.134Ca0.33 MnO3 at 3 T, respectively. The samples in the present investigation are showing less MR% as compared with the reports in the literature. This may be due to the variation in synthesis technique and related parameters like temperature and grinding time adopted to prepare the samples. It is also observed that MR(%) peaks near TC and then show a slow rise in MR(%) with lowering temperature. Such a behavior can be understood by various types of contribution to magneto-resistance in manganites, viz., intrinsic and extrinsic contributions to the total MR(%). Intrinsic

We have performed thermoelectric power measurement to further explore the nature of polarons responsible for the electrical conduction in paramagnetic insulating phase. Fig. 9 depicts the temperature-dependent TEP of La0.65
xBixCa0.35MnO3 (0 rxr 0.2) compounds in the temperature range 5–300 K. It is seen from the Fig. 9 that all the samples have a negative TEP in the entire temperature range, indicating that the electrons are the dominant charge carriers in their thermoelectric transport. It is also observed that all the samples shows a marked slope change in TEP in the vicinity of magnetic and electrical transitions. The slope change in TEP shifts towards lower temperature with increasing Bi content which is consistent with the magnetic and electrical resistivity measurements. In order to understand the various scattering mechanisms, the TEP data in the metallic region is analyzed using the following equation [27]

S( T ) = S0 + S3/2T 3/2 + S4T 4

(7)

where S0 is a temperature-independent constant arising from impurity, S3/2T3/2 corresponds to electron-magnon scattering, and S4T4 corresponds to spin wave fluctuations in the FM phase. The fitted curve for a typical sample with x ¼0.05 is presented in the inset (a) of Fig. 9 and a good fit between measured and theoretical data is clearly seen. The fitting parameters S0, S3/2, and S4 for the La0.65
xBixCa0.35MnO3 series are tabulated in Table 4. The evaluated values of S3/2T3/2 are found to be larger than S4T4, suggesting that the contribution from electron-magnon scattering dominates the thermoelectric transport in the metallic region. It is well-established that thermoelectric power in manganite systems is governed by thermally activated carriers at high temperatures, therefore the TEP in PMI region is analyzed within the framework of Mott's polaron hopping model [27] given by the equation,

Fig. 8. Temperature-dependent MR(%) of the La0.65
xBixCa0.35MnO3 (0 rx r 0.2) series under applied field of (a) 2 T and (b) 8 T.

S.O. Manjunatha et al. / Physica B 498 (2016) 82–91

Fig. 9. Temperature-dependent thermoelectric power of La0.65
xBixCa0.35MnO3 (0 r xr 0.2) compounds. The insets represent the fitted thermoelectric power (solid line) as a function of temperature of La0.60Bi0.05Ca0.35MnO3 compound (x¼ 0.05) according to (a) Eq. (5) in the metallic state and (b) Eq. (6) in the insulating state.

Table 4 Fitting parameters of thermoelectric power in metallic region (S0, S3/2, and S4) and insulating region (ES and α) for La0.65
xBixCa0.35MnO3 (0 r x r0.2) compounds. The values of magnetic entropy (ΔS) are also given. x

S0 (μV/K) S3/2 (μV/K5/2)

0.00
15.94 0.05
5.585 0.10
6.342 0.20 –

S=

1.113 0.547 0.586 –

S4 (μV/K5)

ES (meV) α
9


3.340  10
2.290  10
9
2.453  10
9 –

⎤ kB ⎡ ES + α⎥ ⎢ e ⎣ kBT ⎦

7.816 3.457 3.989 5.574


0.00358
0.00221
0.00257
0.00288

ΔS (R) 0.25 0.23 0.17 0.05

(8)

where ES is the activation energy for thermoelectric power and α is a constant related to the kinetic energy of the polarons. We evaluated the ES and α by linear fitting the S versus 1/T curve in the PMI region and the obtained values are tabulated in Table 4. Inset (b) of Fig. 9 depicts the linear fit for S versus 1/T curve for a typical sample with x ¼0.05 and it is obvious that polaron hopping model satisfactorily describes the measured data in the insulating region. It is noted that the deduced α values are found to be less than one, implying that small polaron hopping governs the transport of the carriers in these Bi-substituted LaCaMnO manganites. Moreover, the magnitude of ES is much smaller than that of EA, which is also the characteristic of small polaron hopping conduction. We thus confirm here that the small polaron hopping conduction is operative in paramagnetic insulating region for the La0.65
xBixCa0.35 MnO3 system. 3.5. Thermal conductivity (κ) The temperature dependence of thermal conductivity (κ) for La0.65-xBixCa0.35MnO3 (0 rx r0.2) compounds measured in the temperature range 10–310 K is shown in Fig. 10. The magnitude of room-temperature κ is found to be rather low and lies between 18 (x ¼0.20) and 30 mW/cm K (x ¼0). For a crystalline solid, such a low value of thermal conductivity could be originated from random distortions of the lattice which lead to high degree of disorder. In manganites, it suggests the presence of Jahn–Teller

89

Fig. 10. Temperature-dependent thermal conductivity of La0.65
xBixCa0.35MnO3 (0 rx r 0.2) compounds.

distortions of the Mn þ 3O6 octahedra [41,42]. Furthermore, the overall magnitude of κ decreases with increasing Bi content, which is attributed to the increased phonon scattering due to the increase in the formation of John–Teller polarons that act as the scattering centers. All the samples exhibit a step-like feature (indicated by arrows in Fig. 10) at temperature Tκ near to their corresponding TC. This is attributed to the fluctuations of the magnetic order parameter in the vicinity of TC which tend to scatter the phonons [43]. At higher temperatures (above Tκ), thermal conductivity for all measured samples increases with increasing temperature. Usually, high-temperature thermal conductivity of crystalline insulators is mostly a decreasing function of temperature. The increase in κ(T) above Tκ cannot be attributed to high temperature electron or phonon processes and may be attributed to the local an harmonic lattice distortions associated with the thermally activated polarons [44,45]. On the other hand, κ(T) is observed to increase as the temperature decreases below Tκ. This is connected to the suppression of phonon-phonon Umklapp scattering due to the destabilization of John–Teller polarons, as the materials enter the ferromagnetic metallic state [44,46]. With further lowering temperature, a broad peak appears around 40 K for these LaBiCaMnO samples. This is a typical feature for the reduction of thermal scattering in crystalline solids at low temperatures where the mean free path of phonons becomes approximately equal to the crystal site distance, ascribed to the Umklapp process. At even lower temperatures the rapid decrease in κ(T) below the broad peak indicates the crossover from Umklapp to defect limited scattering [46]. A clear trend found in κ is that the height of the low-temperature peak gradually decreases with increasing x, indicating a strong enhancement in the phonon scattering due to Bi substitution for La. In general, the total thermal conductivity of magnetic materials is a sum of contributions from various components viz., electronic (κe), phonon (κph) and magnetic (κm). However, it has been established that, in manganites, the electronic and magnetic contributions are negligible compared to the phonons [41]. The electronic thermal conductivity κe can be evaluated using the Wiedemann–Franz law κeρ/T¼ L. Here ρ is the dc-electric resistivity and the Lorentz number L ¼2.45  10
8 W Ω K
2. We thus obtained the room-temperature electronic thermal conductivity κe for La0.60Bi0.05Ca0.35MnO3 sample (x ¼0.05, ρ300  0.6 Ω mm) as 0.12 mW/cm K. The magnetic thermal conductivity κm can be

90

S.O. Manjunatha et al. / Physica B 498 (2016) 82–91

ferromagnetic–paramagnetic phase transition (R is the ideal gas constant). This discrepancy is ascribed to the inhomogeneity of the sample, or partially canted spins in the ferromagnetic state. An obvious trend of the estimated ΔS of the La0.65
xBixCa0.35MnO3 series is that it decreases monotonically with increasing Bi content. Such a result suggests an increase in magnetic inhomogeneity due to the weakening of the double-exchange mechanism with Bi substitution, or due to the decrease in grain size which affects the long range ferromagnetic ordering as we discussed in the magnetic studies section. It is also noticed that the sample with x ¼0.20 exhibits a rather broad slope change instead of a peak in CP near TC and consequently a very small entropy change (see Table 4) associated with the transition, which may be due to the lack of long range magnetic ordering in the sample [47]. This is in good agreement with the ZFC magnetization data where the magnetic moment decreases considerably below TC.

Fig. 11. Temperature-dependent specific heat of La0.65
xBixCa0.35MnO3 (0 r xr 0.2) compounds. Inset (a) shows CP data (symbols) of the x¼ 0.05 sample with estimated lattice background (solid line) and inset (b) shows ΔCP/T vs. T plot.

estimated through the relation κm ¼(1/3)Cm(vm) τm (Cm is the magnetic specific heat, v m and τ m are the velocity and life-time of magnons respectively) and it is usually orders of magnitude smaller than κph. Thus we can conclude that the measured κ of the La0.65
xBixCa0.35MnO3 series is mainly due to lattice phonons with a negligibly small contribution from electrons and magnons. 2

3.6. Specific heat (CP) The temperature dependence of specific heat (CP) measured in the temperature range 90–300 K are presented in Fig. 11. It is seen from Fig. 11 that the La0.65Ca0.35MnO3 (x¼0.00) sample exhibits a sharp peak in CP at T ¼263.0 K during the corresponding magnetic phase transition, which is very close to its Curie temperature TC ¼ 263.8 K as determined from the magnetization measurement. With increasing Bi substitution, the peak shifts towards lower temperature and the peak height is progressively suppressed. Such an observation is consistent with the report in the literature [28]. In order to evaluate the excess specific heat (ΔCP) associated with the transition, we have estimated the lattice background from 90 K to 300 K for each sample. In this temperature range, we assume that the lattice specific heat has the form given from the Einstein model,

⎤ ⎡ xi2e xi ⎥ CEinstein = 3nR∑ ai⎢ , 2⎥ ⎢ xi i ⎣ ( e − 1) ⎦

(9)

here xi ¼hνE/kBT, n is atoms per molecule (in present case n¼ 5), and R is the ideal gas constant. Three optical phonons (i¼1, 2, 3) having Einstein frequencies νE and relative occupations ai with ∑i ai = 1 were used. The best fit of the data for a typical sample with x¼ 0.05 is shown as a solid line in inset (a) Fig. 11, and the estimated Einstein temperatures ΘE ¼hνE/kB are 200 K, 500 K, and 700 K, respectively. It is worth mentioning that since the lattice specific heat is quite similar for all samples in the LaBiCaMnO series, and so are their Einstein temperatures (not shown here). The magnetic contribution for the specific heat can be then obtained by subtracting the lattice part from the experimental data. The change in entropy (ΔS) is evaluated by integrating the area under ΔCP/T versus T curves as depicted in the inset (b) of Fig. 11 and the obtained values of ΔS are given in Table 4. The entropy change (ΔS¼ 0.25R) for the un-substituted sample La0.65Ca0.35 MnO3 is about 40% that of the theoretical value Rln2 for a

4. Conclusions Systematic investigation on structural, magneto-transport, magnetic and thermal properties of La0.65
xBixCa0.35MnO3 (0 rx r0.2) compounds have been carried out. Structural studies confirmed that all the samples are single phase and crystallize in rhombohedral symmetry with R-3C space group. Electrical and magnetic studies consistently showed that the TMI and TC shift towards lower temperature side with increasing x. In addition, MZFC is found to be lower as compared to MFC well below TC which is ascribed to the appearance of the spin glass or cluster glass state in the samples. This bifurcation also suggests the presence of magnetic inhomogeneity in the long range ferromagnetic ordering. A more pronounced bifurcation feature is seen with increasing Bi substitution, indicating an enhancement in magnetic anisotropy in the LaCaMnO system. This magnetic anisotropy in turn leads to the appearance of Griffiths-like phase in the Bi-substituted LaCaMnO samples. The electrical transport data were analyzed by utilizing various theoretical models. From the analyses, it is revealed that electron–electron scattering dominates the conduction mechanism in the metallic region whereas the insulating region the electrical transport is well-explained using polaron hopping model. The percolation model is also used to describe the resistivity data in the entire temperature range. Analysis of thermoelectric power data reveals that the small polaron hopping mechanism is operative in the high temperature insulating regime. The magnitude of room-temperature thermal conductivity for these manganites lies in the range of 18–30 mW/cm K. Such a low value of κ suggests the presence of Jahn–Teller effect in the samples studied. The change in entropy associated with the FM–PM transition is evaluated from the specific heat measurements. The change in entropy is found to decrease with increasing Bi content as a result of the enhancement of magnetic inhomogeneity, in good agreement with the finding from magnetic measurements.

Acknowledgments One of the authors (S.O.M) acknowledges Manipal University for providing financial support to pursue Ph.D. The authors are thankful to Dr. Rajeev Rawat and Mr. Sachin Kumar, UGC-DAE, CSR Indore for resistivity and MR measurement. The thermal and magnetic measurements are supported by the Ministry of Science and Technology of Taiwan under Grant no. MOST 103-2112-M259-008-MY3 (Y.K.K.).

S.O. Manjunatha et al. / Physica B 498 (2016) 82–91

References [1] S. Jin, T.H. Tiefel, M. McCormack, R.A. Fastnacht, R. Ramesh, L.H. Chien, Science 264 (1994) 413. [2] C.N.R. Rao, B. Raveau (Eds.), Colossal Magneto-Resistance, Charge Ordering and Related Properties of Manganese Oxides, World Scientific, Singapore, 1998. [3] P.K. Siwach, H.K. Singh, O.N. Srivastava, J. Phys.: Condens. Matter 20 (2008) 273201. [4] V. Dayal, V.P. Kumar, J. Magn. Magn. Mater. 361 (2014) 212–218. [5] M.B. Salmon, P. Lin, Phys. Rev. Lett. 88 (2002) 197203. [6] A.K. Pramanik, A. Banerjee, Phys. Rev. B 81 (2010) 024431. [7] V. Dayal, V.P. Kumar, R.L. Hadimani, D.C. Jiles, J. Appl. Phys. 115 (2014) 17E111. [8] H.G. Zhang, L.S. Chen, Y.T. Li, H. Liu, Y.Y. Chen, K. Chen, X.G. Dong, H. Yang, T. Gao, Q. Li, J. Supercond. Nov. Magn. 24 (2011) 1665. [9] J. Deisenhofer, D. Braak, Krug von Nidda, J. Hemberger, R.M. Eremina, V. A. Ivanshin, A.M. Balbashov, G. Jug, A. Loidl, T. Kimura, Y. Tokura, Phys. Rev. Lett. 95 (2005) 257202. [10] W.J. Jiang, X.Z. Zhou, Phys. Rev. B 76 (2007) 092404. [11] W.J. Jiang, X.Z. Zhou, G. Williams, Phys. Rev. B 77 (2008) 064424. [12] C.L. Lu, K.F. Wang, S. Dong, J.G. Wan, J. -M. Liu, Z.F. Ren, J. Appl. Lett. 103 (2008) 07F714. [13] W. Jiang, Xue Zhi Zhou, Gwyn Williams, Y. Mukovskii, K. Glazyrin, Phys. Rev. Lett. 99 (2007) 177203. [14] W.J. Lu, X. Luo, C.Y. Hao, W.H. Song, Y.P. Sun, J. Appl. Phys. 104 (2008) 113908. [15] Wanjun Jiang, X.Z. Zhou, Gwyn Williams, Phys. Rev. B 76 (2007) 092404. [16] J.A. Souza, J.J. Neumeier, Y.-K. Yu, Phys. Rev. B 78 (2008) 014436. [17] H. Gencer, S. Atalay, H.I. Adiguzel, V.S. Kolat, Phys. B 357 (2005) 326. [18] S. Atalay, V.S. Kolat, H. Gencer, H.I. Adiguzel, J. Magn. Magn. Mater. 305 (2006) 452. [19] Z.C. Xia, L.X. Xiao, C.H. Fang, G. Liu, B. Dong, D.W. Liu, L. Chen, L. Liu, S. Liu, D. Doyananda, C.Q. Tang, S.L. Yuan, J. Magn. Magn. Mater. 297 (2006) 1–6. [20] Renwen Li, Zhe Qu, Jun Fang, Xiaoman Yu, Lei Zhang, Yuheng Zang, Solid State Commun. 150 (2010) 389. [21] Renwen Li, Zhe Qu, Jun Fang, Phys. B 406 (2011) 1312. [22] S. Zemni, M. Baazaoui, Ja Dhahri, H. Vincent, M. Oumezzine, Mater. Lett. 63 (2009) 489. [23] Y.-K. Kuo, C.S. Lue, F.H. Hsu, H.H. Li, H.D. Yang, Phys. Rev. B 64 (2001) 125124. [24] Y. Tokura (Ed.), Colossal Magneto-Resistance Oxides, Gordon and Breach, London, 2000. [25] G. Venkataiah, V. Prasad, P. Venugopal Reddy, Phys. Status Solidi A 203 (2006) 2478. [26] D.C. Krishna, P. Venugopal Reddy, J. Alloy. Compd. 479 (2009) 661.

91

[27] S.O. Manjunatha, Ashok Rao, Subhashini, G.S. Okram, J. Alloy. Compd. 640 (2015) 154. [28] D.D. Mamatha, Neeraj Kumar, V.P.S. Awana, Bhasker Gahtori, J. Benedict Christopher, S.O. Manjunath, K.Z. Syu, Y.-K. Kuo, Ashok Rao, J. Alloy. Compd. 588 (2014) 406. [29] D. Fatnassi, J.L. Rehspringer, E.K. Hlil, D. Niznansky, M. Ellouze, F. Elhalouani, J. Supercond. Nov. Magn. 28 (2015) 2401. [30] W. Tang, W.J. Lu, X. Luo, B.S. Wang, X.B. Zhu, W.H. Song, Z.R. Yang, Y.P. Sun, Phys. B 405 (2010) 2733. [31] N. Mtiraoui, A. Dhahri, M. Oumezine, H. Dhahri, E. Dhahri, J. Magn. Magn. Mater. 323 (2011) 2831. [32] P.T. Phong, S.J. Jang, B.T. Huy, Y.-I. Lee, I.-J. Lee, J. Mater. Sci.: Mater. Electron. 24 (2013) 2292. [33] G. Narsinga Rao, J.W. Chen, S. Neeleshwar, Y.Y. Chen, M.K. Wu, J. Phys. D: Appl. Phys. 42 (2009) 095003. [34] L. Righi, J. Gutierrez, J.M. Barandiaran, J. Phys.: Condens. Matter 11 (1999) 2831. [35] S.O. Manjunatha, Ashok Rao, V.P.S. Awana, G.S. Okram, J. Magn. Magn. Mater. 394 (2015) 130. [36] L. Righi, J. Gutierrez, J.M. Barandiaran, J. Phys.: Condens. Matter 11 (1999) 2831–2840. [37] X.L. Jiang, J. Gao, S.K. Ren, Q.Y. Xu, H. Sang, F.M. Zhang, Y.W. Du, J. Alloy. Compd. 384 (2004) 261. [38] G. Srinivasan, R.M. Savage, V. Suresh Babu, M.S. Seehra, J. Magn. Magn. Mater. 168 (1997) 1. [39] (a) Z.C. Xia, G. Liu, B. Dong, L. Chen, D.W. Liu, C.H. Fang, D. Doyananda, L. Liu, S. Liu, C.Q. Tang, S.L. Yuan, J. Magn. Magn. Mater. 292 (2005) 260; (b) Z.C. Xia, Y.Y. Wu, G.F. Zhu, A.Z. Hu, X.N. Feng, Y. Wang, C.S. Xiong, S.L. Yuan, C.Q. Tang, J. Magn. Magn. Mater. 320 (2008) 368. [40] G. Venkataiah, V. Prasad, P. Venugopal Reddy, J. Alloy. Compd. 429 (2007) 1. [41] S.K. Agarwal, Neeraj Kumar, Neeraj Panwar, Bhasker Gahtori, Ashok Rao, P. C. Chang, Y.K. Kuo, Solid State Commun. 150 (2010) 684. [42] R. Berman, Thermal Conduction in Solids, Oxford University Press, Oxford, 1976. [43] I.K. Kamilov, A.G. Gamzatov, A.M. Aliev, A.B. Batdalov, B. Abdulvagidov Sh, O. V. Melnikov, O.Yu. Gorbenko, A.R. Kaul, Low Temp. Phys. 33 (2007) 829. [44] Dinesh Varshney, Irfan Mansuri, M.W. Shaikh, Y.K. Kuo, Mater. Res. Bull. 48 (2013) 4606. [45] M. Ikebe, H. Fujishiro, Y. Konno, J. Phys. Soc. Jpn. 67 (1998) 1083. [46] G.L. Bhalla Vikram Sen, W.K. Neeraj Panwar, N. Syu, Y.K. Kaurav, Kuo, Ashok Rao, S.K. Agarwal, Physica B 405 (2010) 1. [47] I. Dhiman, A. Das, A.K. Nigam, R.K. Kremer, J. Magn. Magn. Mater. 334 (2013) 21.