Influence of BiFeTaO3 addition on the electrical properties of Na0.4725K0.4725Li0.055NbO3 ceramics system using impedance spectroscopy

Journal of Alloys and Compounds 637 (2015) 203–212

Contents lists available at ScienceDirect

Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom

Influence of BiFeTaO3 addition on the electrical properties of Na0.4725K0.4725Li0.055NbO3 ceramics system using impedance spectroscopy Poonam Kumari a, Radheshyam Rai a,⇑, A.L. Kholkin b a b

School of Physics, Shoolini University, Solan, HP, India Department of Glass and Ceramics, Aveiro University, Aveiro, Portugal

a r t i c l e

i n f o

Article history: Received 27 September 2014 Received in revised form 18 February 2015 Accepted 20 February 2015 Available online 26 February 2015 Keywords: Ceramics X-ray diffraction Dielectric properties Impedance spectroscopy

a b s t r a c t Polycrystalline samples of (1  x)(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007 were prepared by using a solid state reaction technique. The XRD patterns of the samples at room temperature shows perovskite phase with monoclinic structure. The dielectric constant for x = 0.007 is maximum. Detailed studies of dielectric and impedance properties of the materials in a wide range of frequency (100 Hz–1 MHz) and temperatures (RT–500 °C) shows that dielectric properties were strongly temperature and frequency dependent. Dielectric and electrical properties of samples, indicate that the Curie temperature shifted to higher temperature side with the increase in frequency. The AC conductivity also increases with increase in frequency. The low value of activation energy obtained for the ceramic samples could be attributed to the influence of electronic contribution to the conductivity. The plots of Z00 and M00 vs frequency at various temperatures shows peaks in the higher temperature range (>320 °C). The compounds show dielectric relaxation, which is found to be of non-Debye type and the relaxation frequency shifted to higher side with increase in temperature. The Nyquist plot and conductivity studies showed the NTCR character of samples. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction In the sensor based modern society, functional ceramics plays a vital role. These functional ceramics are used in piezoelectric fuel injection, piezoelectric motors, piezoelectric printing machines, piezoelectrically controlled thread guides, micro positioning systems, and piezoelectric sensors. Most successful piezoelectrics, from the application point of view, belongs to the solid solution of lead based system [1]. The major drawback of lead based ceramics is the high contents of lead (60 wt%), which is toxic in nature and harmful for humans and environment. With the current most restrict environment directives; there is a strong need to the search for lead free alternatives to replace lead based ceramics. K0.5N0.5NbO3 (KNN) is currently being considered one of the most promising lead free composition to replace PZT [2]. K0.5Na0.5NbO3 experience two phase transition from room temperature to Curie temperature, that is, orthorhombic–tetragonal phase transition at about 200 °C and tetragonal–cubic transition at 420 °C. When the temperature reaches 400 °C, the ferroelectric hysteresis loop of ⇑ Corresponding author. E-mail address: [email protected] (R. Rai). http://dx.doi.org/10.1016/j.jallcom.2015.02.149 0925-8388/Ó 2015 Elsevier B.V. All rights reserved.

KNN still can be detected, indicating KNN represents good ferroelectric properties at high temperature [3]. Shirane et al. made the first systematic structural and dielectric investigation on KNN ceramics with the help of X-ray and dielectric measurements in 1954 [4]. Piezoelectricity of KNN was first studied by Egerton and Dillon [5], and they found that the radial coupling coefficient of KNN is excellent. KNN is a solid solution of ferroelectric KNbO3 and antiferroelectric NaNbO3 [4,6]. KNN has poor densification due to the inherent volatility of both K and Na [7] at the conventional sintering and therefore shows the lower piezoelectric characteristics (d33 = 80 pC/N and kp = 36%). To improve the densification and piezoelectric properties of KNN ceramics, a number of additions were added into KNN to form new KNN-based ceramics[8], such as KNN–LiNbO3 [9], KNN–Bi2O3 and Ta doped KNN ceramics [10,11], leading to dense microstructures without requiring sophisticated processes such as hot pressing. Due to dopants, its sinterability and piezoelectric, dielectric properties are improved while the phase transition temperatures are lowered in most cases [12,13]. Guo et al. reported that the doping of Li1+ on (K0.50Na0.50)1xLixNbO3 increased the piezoelectric properties [14] and the orthorhombic–tetragonal phase transition temperature decreases monotonically [15]. Saito et al. [12] reported that sinterability and

204

P. Kumari et al. / Journal of Alloys and Compounds 637 (2015) 203–212

dielectric properties of KNN ceramics have been improved by adding LiTaO3 [16]. Zhou et al. reported that BiFeO3 (BFO) also enhanced sintering densification behavior of KNN and change the room-temperature phase structure from a dominant orthorhombic state to the coexistence of orthorhombic and tetragonal phases, resulting in peak piezoelectric properties of d33 = 340 pC/N and kp = 47% [17]. BFO is representative of materials exhibiting a multiferroism that simultaneously possesses electric and magnetic orderings at room temperature [18,19] and widely studied in both ceramic and film forms [20]. Pure BiFeO3 (BFO) is a promising candidate as a multiferroism material because it is antiferromagnetic with a relatively high Neel temperature (TN = 370–380 °C) and ferroelectric with a high Curie temperature (Tc = 810–850 °C) [21,22]. The substitution of Ta was used to enhance the electrical resistivity of BiFeO3. In addition, the activation energy increased from 0.58 to 1.1 eV with Ta substitution in BiFeO3 ceramics. A close examination of the complex impedance plots at similar temperatures indicates that both grain and grain-boundary resistivities increased with Ta substitution, but the grain-boundary enhancement was much more significant [23]. The values for the conductivity and activation energy in Ta-substituted BiFeO3 were close to those in the BiFeO3– BaTiO3 solid-solution system or Nb-doped BiFeO3 [20]. The application of BiFeO3 is due its thermal metastability and electrical conduction due to non-stoichiometry [24]. In recent years, the research is going on BFO which shows the co-existence of both ferroelectric and ferromagnetic ordering in the same phase [25]. Particularly, the (BFT) BiFe0.5Ta0.5O3 ceramics has generated significant interest because it exhibit high piezoelectric coefficient, high dielectric constant, low dielectric loss and Curie temperature in comparison to the conventionally used Na0.4725K0.4725Li0.055NbO3 (NKLN). Among Moreover, it has been demonstrated that multilayer ceramic actuators (MLCAs) can be successfully fabricated using Li and Ta modified KNN ceramics and AgPd inner electrodes with a high electric field-induced strain of 292 pm/V [26]. In polycrystalline ceramics, defects and inhomogeneities such as grain boundaries (intrinsic defects) and point defects are inevitable. Major defects associated with KNN, which can influence its conductivity are oxygen vacancies, alkaline vacancies and space charges. These defects can interact with domain walls, which may strongly affect the electrical properties of these ceramics [27,28]. Indeed, understanding the nature of these defects and charge transport is fundamental to control the electrical performance of KNN ceramics. In case a material has more than one contribution to the impedance, which is often the case with polycrystalline ceramics where grain, grain boundaries and electrode–ceramic interface have distinct contributions, then one can witness more than one semi-circle, often overlapping each other. One of the ways to model such a behavior may be using three series–parallel RC elements circuit. As we know, complex impedance spectroscopy (CIS) is a nondestructive method to study microstructure and electrical properties of solids [29]. It enables us to evaluate the relaxation frequency (xmax) like, bulk effects, grain boundaries and electrode interface effects in the frequency domain of the material. The relaxation frequency of the material, at a given temperature, is only an intrinsic property of the material independent of geo-metrical factors of the sample. Consequently, an analysis of the electrical properties (conductivity, dielectric constant/loss, etc.) carried out using relaxation frequency (xmax) gives unambiguous results when compared with those obtained at arbitrarily selected fixed frequencies. In this way, the impedance measurements enable us to eliminate the error, if any, due to stray frequency effects. The impedance measurements on a material give us data having both resistive (real part) and reactive (imaginary part) components. It is conventionally displayed in a complex plane plot (Nyquist diagram) [30].

In this work, we studied the effects of BiFe0.5Ta0.5O3 content on the phase structure, microstructure and electrical properties of (Na0.4725K0.4725Li0.055NbO3) system synthesized by solid-state reaction method. A detailed and comparative investigation of the conduction behavior, impedance, electric modulus and transport properties of samples were studied with the help of Impedance Spectroscopy method. 2. Experimental procedures Polycrystalline samples of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) were synthesized from high purity oxides Na2CO3 (99.99 wt%), K2CO3 (99.99 wt%), Li2CO3 (99.99 wt%) Nb2O5 (99.99 wt%), Ta2O5 (99.99 wt%), Fe2O3 (99.99%), Bi2O3 (99.9 wt%), by using a solid state reaction technique. The constituent compounds in suitable stoichiometric were thoroughly mixed in a ball milling unit for 48 h. Then powder was dried at 100 °C and calcined at 900 °C for 4 h in alumina crucibles. The calcined fine powder was cold pressed into cylindrical pellets of 10 mm in diameter and 1–2 mm in thickness using a hydraulic press with a pressure of 50 MPa. These pellets were sintered at a temperature i.e. 1000 °C for 3 h. The formation and quality of compounds were verified with X-ray diffraction (XRD) technique. The XRD patterns of the compounds were recorded at room temperature using X-ray powder diffractometer (Rigaku Minifiex, Japan) with Cu Ka radiation (k = 1.5405 Å) in a wide range of Bragg angles 2h (20° 6 2h 6 60°) at a scanning rate of 2 deg min1. The flat polished surfaces of the sintered samples were electroded with air drying silver paste. Dielectric and impedance were determined by use of PSM1734 Impedance Analyzer at frequencies 1 kHz–1 MHz; samples were heated from room temperature to 500 °C.

3. Results and discussion 3.1. Structural studies Fig. 1 shows the X-ray diffraction pattern of crystalline powder 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics at room temperature. All patterns match with JCPDS#770038 data samples which shows that the samples have a mixed phase with a monoclinic structure at room temperature. The diffraction peaks of the compound were indexed indifferent crystal systems and unit cell configurations using a standard computer program package ‘‘POWD’’ as shown in Table 1. The main peak of the samples are located at approximately 2h = 31.86°, having h k l value h2 2 0i. Because of the kinetics of formation, the other impurity phases are obtained during synthesis. In all the samples, the impurity phase was observed and it was shown as ⁄ in Fig. 1 and it may be attributed to KTa (K6Ta10O13) (JCPDS#010701088). From XRD pattern we found that the intensity increases with doping of (BFT) BiFe0.5Ta0.5O3, which indicates

Fig. 1. XRD patterns of the 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics.

P. Kumari et al. / Journal of Alloys and Compounds 637 (2015) 203–212 Table 1 Lattice parameters and observed inter-planar spacing d for 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007.). Lattice parameters a = 7.975 b = 7.862 c = 7.956

dobs.

hkl

3.987 3.987 2.808 2.799 1.993 1.989 1.778 1.778 1.624 1.620

200 002 202 220 400 004 402 420 224 422

that the better crystalline phase. The diffraction peaks shift slightly to a higher angle as doping concentration of BFT increases in NKLN ceramics.

x = 0.005

205

The microstructure of ceramics was investigated by scanning electron microscope (SEM). A SEM image of samples was taken from freshly fractured surface as shown in Fig. 2. The fine-grained material with an average grain size of 0.5–2 lm was obtained. 3.2. Dielectric studies Fig. 3 shows the variation of dielectric constant (e) as a function of temperature of BFT doped NKLN at frequencies 1 kHz and 10 kHz. Characterization of dielectric constant as a function of temperature is the most common tool to observe phase transition in ferroelectric system. As in normal ferroelectrics, the dielectric constant of KNN ceramics increases gradually with increasing temperature up to the transition temperature and thereafter it decreases with increasing temperature [31]. It was found that the value of the dielectric constant increases and ferroelectric–paraelectric phase transition temperature shifts toward the higher temperature side as we increase BFT substitution from x = 0.005 to

x = 0.007

Fig. 2. SEM micrographs of surface of the 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics.

Fig. 3. Variation of dielectric part (e0 ) as function of temperature of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics at 1 kHz and 10 kHz frequencies.

206

P. Kumari et al. / Journal of Alloys and Compounds 637 (2015) 203–212 250

x =0.005

3000

o 25 C o 100 C o 200 C o 300 C o 400 C o 500 C

2500 2000 1500 1000

x =0.005

o 25 C o 100 C o 200 C o 300 C o 400 C o 500 C

200 150 100 50

500 0 x =0.007

8000

o 25 C o 100 C o 200 C o 300 C o 400 C o 500 C

6000 4000 2000

x =0.007

1000

o 25 C o 100 C o 200 C o 300 C o 400 C o 500 C

800 600 400 200 0

0 0.1

1

10

100

log f (kHz)

0.1

1000

1

10

100

1000

log f (kHz)

Fig. 4. Variation of real (e0 ) and imaginary part of dielectric constant (e00 ) as a function of frequency of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics at different temperatures.

Table 2 Some physical parameters for different compositions of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics. Compositions

Freq. (kHz)

Tmax (°C)

emax

eRT

Activation energy (eV)

x = 0.005

1 10

400 403

866.918 584.560

538.630 381.879

0.85 0.73

x = 0.007

1 10

402 405

1443.594 889.233

500.576 359.279

0.92 0.87

0.007. The ferroelectric–paraelectric phase transition temperature was observed around 400 °C for x = 0.005, but in case of x = 0.007 two peaks were observed around at temperatures 210 °C and 405 °C. Two peaks in the dielectric constant can be correlated to the two phase transitions from ferroelectric to ferroelectrics and

-4

-5 -8

fitted line

-7

-5

1kHz

10kHz

-9

-7.5

x = 0.005

x = 0.005 1kHz

-6

ferroelectrics to paraelectric at temperatures 210 °C and 405 °C respectively. For BFT doped NKLN (x = 0.007), the peaks in the dielectric constant can be ferroelectric–paraelectric phase transition (Tc) was measured as 405 °C from cubic to monoclinic upon cooling. In both cases, at higher temperatures, a strong increase of the dielectric constant was observed which could be related to thermally-induced enhancement of the hopping conduction and/ or dipole orientation thereby enhancing the dielectric constant. Real and imaginary part of dielectric constant decreases with the increasing frequency in both cases as shown in Fig. 4. The decrease in dielectric constant with increased frequency could be explained on the basis of total polarization arising from dipoles and trapped charge carriers [32]. All the curves show an intense increase of the loss magnitude at increased temperatures as compared to their room temperature values. At high frequencies, the loss is higher than one occurring at low frequencies. This kind of

x =0.007 1kHz 10kHz fitted line

x = 0.007 1kHz

-8.5

-10

10kHz

= 1 .1 5

= 1.38

-6 -9.0

= 1 .2 7 -11

-8

-8.0

-7

10kHz

= 1.10

-9.5

-10.0 -12 1.5

2.0

2.5

3.0

3.5

4.0

-8

ln (T -T m a x ) (K )

-10.5 2.0

2.5

-9

3.0

3.5

4.0

ln(T-Tmax ) (K)

-9 -10

-10

-11

-11

-12

-12 2

3

ln(T-Tmax ) (K)

4

5

2

3

4

ln(T-Tmax ) (K)

Fig. 5. Variation of ln (1/e  1/emax) vs ln (T  Tmax) of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics at 1 kHz and 10 kHz frequencies.

207

P. Kumari et al. / Journal of Alloys and Compounds 637 (2015) 203–212 -6

-4

-7

-6.6

x = 0.005

-6.8

1kHz

-7.0

10KHz

-7.2

Fitted line

-7.4

-8

-5.2

x = 0.007 1kHz

-5.4

10KHz -5.6

-6

E a = 0.73

Fitted line

E a = 0.87 eV

-5.8

-7.6 -7.8

-6.0

-8.0

-9

-8.2

-8

E a= 0.85 eV

-6.2

-8.4

E a = 0.92 eV

-6.4

-8.6 0.00130

-10

0.00135

1 /T (K -1 )

0.00140

0.00145

0.00132

0.00134

0.00136

0.00138

1 /T (K -1 )

-10

0.00140

0.00142

-11

-12

-12

x = 0.007

x = 0.005

-13

1kHz 10KHz

-14

1kHz

-14

10KHz

0.0015

0.0020

0.0025

0.0030

0.0035

0.0015

0.0020

0.0025

0.0030

0.0035

1/T(K -1)

1/T(K -1)

Fig. 6. Variation of ac conductivity ln rac as a function of inverse of absolute temperature 1/T of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics at 1 kHz and 10 kHz frequencies.

600

1500

30

x=0.007

x = 0.005

100

o

o

400 C 425 oC

80

o

425 C

450 oC

20

o

475 C

60

400 C

25

o

450 C

500 oC

o

475 C

15

40

o

500 C

400

1000

20

10

0 0.1

1

10

100

5

1000

log f (kHz) 0

x = 0.005 o

325 C

500

o

o

425 C

0.1

1

450 C

375 oC

475 C

400 oC

500 C

100

x=0.007

200

o

325 C o 350 C o 375 C o 400 C

o

350 C

10

log f (kH z)

o o

1000

o

425 C o 450 C o 475 C o 500 C

0

0 0.1

1

10

100

0.1

1000

1

log f (kHz)

10

100

1000

log f (kHz)

Fig. 7. Variation of real part modulus (Z0 ) with frequency at different temperatures of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics.

dependence of the e00 with frequency is associated with the dielectric loss due to the conduction mechanism occurring at high temperatures for ceramics. As the frequency of the applied field increases, the dipoles will not be able to orient themselves in the direction of the applied field, and hence the value of the dielectric constant decreases at high frequency [33]. BFT doped NKLN sample with x = 0.007 shows the maximum value of dielectric constant for 1 kHz frequency as shown in Table 2. To observe the effect of temperature on diffuseness, its variation as a function of temperature is plotted at 10 kHz frequency for BFT doped NKLN ceramics in Fig. 5. The degree of diffusivity in phase transition of BFT doped NKLN ceramics was calculated using Eq. (1).

ln



1

e



1

emax



¼ c lnðT  T max Þ þ c

ð1Þ

where e is the dielectric constant at temperature T, c is constant and emax is dielectric constant value at Tmax [34]. The parameter c gives information on the character of the phase transition and only depending on the composition of the specimens. The values c = 1 indicating the normal ferroelectric and c = 2 indicating ideal relaxor ferroelectric respectively [35]. It is clear that diffusivity increases with increasing temperature. The value of c found between (1.1 6 c 6 1.38) in the both cases. Since the c value confirm the normal behavior of dielectrics/ferroelectrics for BFT doped NKLN compounds.

208

P. Kumari et al. / Journal of Alloys and Compounds 637 (2015) 203–212 -700

-10

-40

x=0.007

o

o

400 C

400 C

o

-30

-600

-8

-200

425 C

o

425 C o

o

450 C

450 C

o

-6

o

475 C

475 C

-20

o

500 C

o

500 C

-4

-500 -10

-400

-2

-150

0

0 0.1

1

10

log f (kHz)

100

x = 0.005 o o 325 C 425 C

-300

o

450 C

o

475 C

o

500 C

350 C

-200

375 C -100

400 C

0.1

1000

-100

100

-50

1000

x=0.007

o

425 oC

o

350 C

450 oC

375 oC

475 oC

400 oC

500 oC

325 C

o

0

10

lo g f (k H z )

o o

1

0 0.1

1

10

100

1000

log f (kHz)

0.1

1

10

100

1000

log f (kHz)

Fig. 8. Variation of imaginary part modulus (Z00 ) with frequency at different temperatures of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics.

Fig. 6 shows the variation of log rac (s/m) vs 1/T (K1) of BFT doped NKLN ceramics at 10 kHz frequencies in the temperature range from RT to 500 °C. The AC electrical conductivity was calculated from the impedance data collected with LCR meter using the formula r = xeeo tan d, where eo is the vacuum dielectric constant, x is the angular frequency and kB is the Boltzmann constant. The value of activation energy was calculated in the paraelectric region from the plot of ln(rac) vs 1/T using the conductivity relation r = ro exp(Ea/kBT)[36]. As the temperature increases the AC conductivity also increases for BFT doped NKLN ceramics. At high temperature, the donor cations have a major part to play in the conduction process and activated with small energy which is called activation energy. The activation energy for all composition at different frequency was found to be very low in the paraelectric region. This may be due to ionic solids having a limited number of mobile ions being trapped in relatively stable potential wells during their motion through the solid. Due to a rise in temperature the donor cations are taking a major part in the conduction process. The donors have created a level (i.e. band-donor level), which is much nearer to the conduction band. Therefore, only a small amount of energy is required to activate the donors. In addition to this, a slight change in stoichiometry in multi-metal complex oxides causes the creation of large number of donors or acceptors, which creates donor or acceptors like states in the vicinity of conduction or valance bands. These donors or acceptors may also be activated with small energy [37].

3.3. Electrical studies The electrical properties of BFT doped NKLN materials were investigated by a complex impedance spectroscopy (CIS) technique. It is important to transform the dielectric and electrical data in different formalism and analyze them to get true picture of the material. The use of function Z⁄ and Y⁄ is particularly appropriate for the resistive and/or conductive analysis where the long-range conduction dominates, whereas the e⁄ and M⁄ functions are suitable when localized relaxation dominates. So the plotting of ac data in terms of impedance, electric modulus, and dielectric per-

mittivity simultaneously gives a complete assignment of all the physical processes taking place in the material. The variation of real part of impedance (Z0 ) with frequency at various temperatures (from 325 to 500 °C) for 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (x = 0.005 and 0.007) ceramics is shown in Fig. 7. The pattern shows a sigmoid variation as a function of frequency in the low frequency region followed by a saturation region in the high frequency region. This suggests the presence of mixed nature of polarization behavior in the material, such as electronic, dipolar and orientation polarization. A decreasing trend of Z0 with rise in temperature suggests the presence of negative temperature coefficient of resistance (NTCR) in the material in the low frequency region but tends to merge in the high frequency region at almost all temperatures. A possibility of increase in AC conductivity (rac) in the sample was observed due to the decrease in the magnitude of Z0 with rise in temperature. This increase in AC conduction with temperature may be due to the contribution of defects like oxygen vacancies. Generally, the contribution due to oxygen vacancies is more prominent in perovskite structures at higher temperatures. These results indicate increase in AC conductivity with rise in temperature in the high frequency region (possibly) due to the release of space charge and lowering in the barrier properties of the material. Fig. 8 presents the variation of imaginary part of impedance (Z00 ) as a function of frequency at different set of temperatures (from 325 to 500 °C) for BFT doped NKLN. With the increase of frequency, imaginary part of impedance (Z00 ) increases initially, attain a peak (Z00 max) and then decreases with frequency at all measured temperatures. At higher frequency side all the curves are merged which might be due to the reduction in space charge polarization at higher frequency. The peak shifts toward higher frequency side with increasing temperature showing that the resistance of the bulk material is decreasing. Also, the magnitude of Z00 decreases with increasing temperature. This would imply that dielectric relaxation is temperature dependent, and there is apparently not a single relaxation time. Fig. 9 shows the temperature-dependent spectra (Nyquist plot) of BFT doped NKLN materials. By impedance spectrum we got the semicircular arcs at the set of higher temperatures. The nature of

209

P. Kumari et al. / Journal of Alloys and Compounds 637 (2015) 203–212 -1400

x=0.005 o

350 C

450 C o

375 C

o

500 C

400 C

o

-300

o

o

0

20

40

60

80

100

o

475 C

o

475 C

-600

o

475 C o

500 C

-10

-5

450 C

0

450 C

o

450 C -15

425 C

-20

425 C

-800

o

425 C

-20

o

400 C

o o

o

400 C

o

-400

475 C -40

x=0.007 -25

o

o

o

375 C

325 C

425 C

-60

-30

o

o

-80

o

350 C -1000

x=0.007

-500 o

400 C

325 C

-1200

-100

0 0

5

10

15

20

25

30

o

500 C

-200

o

500 C -400

-100 -200

0

0 0

200

400

600

800

1000

1200

0

1400

100

200

300

400

500

Fig. 9. Variation of real (Z0 ) and imaginary part (Z00 ) of impedance with different temperatures of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics.

0.005

x = 0.005

0.0035

x=0.007 o

o

325 C

325 C

0.004

o

350 C

0.0030

o

350 C

o

375 C 0.0025

o

375 C

o

400 C

0.003

o

425 C o

450 C o

475 C

0.0015

o

400 C o

M'

M'

0.0020

425 C o

0.002

450 C

o

500 C

o

475 C

0.0010

o

0.001

500 C

0.0005

0.000

0.0000

0.1

1

10

log f (kHz)

100

1000

0.1

1

10

log f (kHz)

100

1000

Fig. 10. Variation of real part modulus (M0 ) with frequency at different temperatures of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics.

variation of the arcs with temperature and frequency provides various clues of the materials. However, as the temperature increased the slope decreased and found to curve toward the major (Z0 ) axis forming clear semicircular arcs. The radius of curvature was found to decrease with increasing temperature, which indicates the increase in conductivity of the sample with temperature. Generally, existence of a single semicircular arc represents only one contribution e.g. grain interior (bulk) property of the material. However, in the present case, the spectrum comprises of suppressed semicircular arcs indicating two different contributions from the grain interior (bulk) and grain boundary. The observed semicircular arcs have their centers lying off the real (Z’) axis which is an indication of non-Debye type relaxation with a distribution of relaxation times instead of a single relaxation process. In general, the relaxation time for grain boundary region is much larger than that for the grains and, therefore, its response relaxes at lower frequencies. Thus, the low frequency arc in the Nyquist plot corresponds to the

grain boundary effects and the smaller high frequency arc to the grain/bulk effect of the material. In the other words the semicircular arc with the real axis (Z0 ) gives us an estimate of the bulk resistance (Rb) of the material. It has been observed that the bulk resistance of the material decreases with increase in temperature showing a typical semiconducting property, i.e. negative temperature coefficient of resistance (NTCR) behavior. Electrical response of the materials can also be analyzed through complex electric modulus formalism, which provides an alternative approach based on polarization analysis. Complex electric formalism gives the inhomogeneous nature of the polycrystalline ceramic, which can be probed into bulk and grain boundary effects. Fig. 10 shows the variation of real part of electric modulus (M0 ) with frequency at higher temperatures between 325 °C to 500 °C for BFT doped NKLN ceramics. It is characterized by a low value of M0 in the low frequency region followed by a continuous disper-

210

P. Kumari et al. / Journal of Alloys and Compounds 637 (2015) 203–212

0.0015

x = 0.005

o

400 C

o

425 C

o

475 C

0.0020

o

325 C

x=0.007

o

450 C o

400 oC

o

425 oC

o

450 oC

350 C

o

375 C

o

325 C

o

350 C

500 C

0.0012

0.0015

375 C

475 oC 500 oC

M"

M"

0.0009 0.0010

0.0006 0.0005

0.0003

0.0000

0.0000

0.1

1

10

100

1000

0.1

1

10

100

1000

log f (kHz)

log f (kHz)

Fig. 11. Variation of imaginary part modulus (M00 ) with frequency at different temperatures of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics.

o o

450 C

0.0020

M"

o

0.0015

500 C

M"

o

375 C

o

0.003

o

o

425 C o

0.001

M'

0.002

0.003

450 C

0.002

o

400 C 425 C

o

0.0005

0.003

350 C

375 C

x=0.007 400 o C 425 o C 450 o C 475 o C 500 o C

0.004

o

325 C o

o

400 C

0.0010

0.0000 0.000

0.004

350 C

o

475 C

0.002

o

325 C

425 C

0.0025

x=0.007

x = 0.005

0.001

o

M"

0.003

o

400 C

M"

x = 0.005 0.0030

450 C 0.002

o

o

475 C

0.000 0.000

0.001

0.002

0.003

0.004

o

475 C

500 C

o

500 C 0.001

0.000 0.000

0.001

0.000 0.000 0.001

0.002

0.003

0.001

0.002

0.003

0.004

M'

M' Fig. 12. Variation of real (M0 ) and imaginary part (M00 ) of modulus with different temperatures of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007). Ceramics.

sion with increase in frequency. It was found that M0 values saturated to a maximum (M1 – the asymptotic value of M0 at higher frequencies) in the high frequency region for all temperatures. The asymmetric plot of M0 is because of the stretched exponential character of relaxation time of the material. Monotonous dispersion on increasing frequency at lower temperatures may be caused by short range mobility of charge carriers. Such results may possibly be related to a lack of restoring force governing the mobility of the charge carriers under the action of an induced electric field. The value of M0 decreases with rise in temperature in the observed frequency range. Fig. 11 shows the variation of imaginary part of electric modulus (M00 ) with frequency at higher temperatures for BFT doped NKLN ceramics. By these graphs we found that the position of the peak M00 max shifted to higher frequencies as the temperature was increased. Physically, the peak in the imaginary part of the electric modulus defines the regions where the carrier can move at long distances. At frequency above peak maximum, the carriers are confined to potential wells, being mobile on short distances. The peaks are asymmetric and broader than the ideal Debye curve.

Also, a peak in the M00 imaginary part indicates a dielectric relaxation process in the solid, and the frequency to the maximum indicates the mean relaxation time of this process. As can be seen, the imaginary part of the electric modulus exhibits a very well defined peak. The frequency range where the peaks occur is indicative of transition from long range to short range mobility [38,39]. The complex electric modulus spectrum M0 vs M00 is shown in Fig. 12 for all the samples at different temperatures. Two arcs are clearly observed over the entire measured frequency range in both the samples. In fact, the first arc is the contribution of grain boundaries whereas the second arc is the contribution of grains. The patterns are characterized by the presence of little asymmetric and depressed semicircular arcs whose center does not lie on M0 axis. The behavior of electric modulus spectrum is suggestive of the temperature dependent hopping type of mechanism for electric conduction (charge transport) in the system and non-Debye type dielectric relaxation. In a relaxation system, one can determine the probable relaxation time (s) from the position of the loss peak in the Z00 as well as M00 vs log f plots according to the relation: s = 1/ x = 1/2pf (f is the relaxation frequency). We found that the double

211

P. Kumari et al. / Journal of Alloys and Compounds 637 (2015) 203–212

x = 0.005

1.2

1.0

x=0.007 o 325 C

o

325 C

1.0

o

350 C

o

375 C

o

375 C o

400 C 0.6

o

425 C o

450 C

0.4

o

Z" / Z "max

0.8

Z" / Z "max

350 C

0.8

o

400 C

0.6

o

425 C o

450 C

0.4

o

475 C

o

475 C o

500 C

0.2

0.0

o

500 C

0.2

0.0 0.01

0.1

1

10

100

1000

1E-3

0.01

0.1

log f/f max

1

10

100

1000

log f/f max

Fig. 13. Modulus scaling behavior of (Z00 /Z00 max) vs log (f/fmax) of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics.

-30

-100

x = 0.005 -80

fitted line

-5

Fitted Line

-25

-4

450 oC

-3

475 oC

o

400 C o

450 C

500 oC

-20

-2

o

-60

475 C o

500 C

x = 0.007 Fitted line

-15

-1 0

1

o

2

3

4

5

400 C o 450 C

-40 -10

o

475 C o

500 C

-20

-5

0

0 0

20

40

60

80

100

0

10

15

20

25

30

Fig. 14. Fitting of Cole–Cole plot of impedance with 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) (where x = 0.005 and 0.007) ceramics for high temperature, (at 400 °C, 450 °C, 475 °C and 500 °C).

semicircular arcs in the complex modulus plots with a small semicircle at high frequency and a large semi-circular arc in the low frequency region at the temperatures 300–375 °C as shown in Fig 12. The modulus spectrum shows a marked change in the shape with rise in temperature suggesting a probable change in the capacitance value as a function of temperature. Fig. 13 shows the normalized plot of Z00 /Z00 max at different temperatures for BFT doped NKLN ceramics. The normalized plot overlaps on a single master curve at different temperatures (i.e. same shape and pattern in the peak position with slight variation in full width at half maximum FWHM with rise in temperature). Thus the dielectric processes occurring in the material can be investigated via master impedance plot [40]. The value of FWHM evaluated pffiffi from the normalized spectrum is greater than log 2þp3ffiffi, and this 2 3

indicates about non-Debye type behavior which is well supported by complex impedance plot. The peaks Z00 max shifts and broadening was observed at higher frequencies with increasing temperature. Liu et al. [41] indicated that this broadening of the response might be due to diffusion of motion of the ions leading to a spatial distribution of electrical response times. The appearance of temperature dependent peaks (Z00 max) at a characteristic frequency (xmax = 2pfmax) indicates the presence of relaxation process, which is temperature dependent [42]. These relaxation processes may be due

to the presence of immobile species at low temperature and defects which became mobile at high temperature. Fig. 14 shows the fitting of Nyquist plot for BFT doped NKLN ceramics and it observed from the figure that with the increase in temperature the slope of the lines decrease and the lines bend toward real (Z0 ) axis and at higher temperatures (400 °C, 450 °C, 475 °C and 500 °C); a semicircle could be traced, indicating the increase in conductivity of the sample. It can also be observed that the peak maxima of the plots decrease and the frequency for the maximum shifts to higher values with the increase in temperature. It can be noticed that the complex impedance plots are not represented by full semicircle, rather the semicircular arc is depressed and the center of the arc lies below the real (Z0 ) axis suggesting the relaxation to be of polydispersive non-Debye type in samples. This may be due to the presence of distributed elements in the material electrode system [29,43]. An equivalent circuit is being used to provide a complete picture of the system and establish the structural property relationship of the materials. Comparison of complex impedance plots (symbols) with fitted data (lines) using commercially available software ZSimpwin Version. To model the non-Debye response, constant phase element (CPE) is used in addition to resistors and capacitors. The fitted values of for the sample at various temperatures are enlisted in Table 3. Here it

212

P. Kumari et al. / Journal of Alloys and Compounds 637 (2015) 203–212

Table 3 Resistance and capacitance of bulk and grain boundaries of 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) ceramics after fitting. Compositions

T (°C)

Rg (ohm cm2)

Rgb (ohm cm2)

CPE (Q)

x = 0.005

400 450 475 500

10.95 670.7 6691 89,070

98,510 37,420 22,440 14,340

3.33E9 1.58E9 9.92E10 1.67E9

x = 0.007

400 450 475 500

26.49 94.65 79.8 82.35

28,670 5222 3647 3131

6.64E9 3.318E9 1.918E7 2.172E9

has also been clearly observed from the Nyquist plots that the influence of grain size on the inter grain resistivity increases with decreasing grain size. 4. Conclusions (BFT doped NKLN) 1  x(Na0.4725K0.4725Li0.055NbO3)  x(BiFe0.5Ta0.5O3) solid solution ceramics were prepared using solid state reaction method. BFT doped NKLN ceramics were investigated for its dielectric properties by impedance spectroscopy (IS) in the temperature range of RT–500 °C and in the frequency range of 100 Hz– 1 MHz. The samples have curie temperatures 400–405 °C and exhibiting a monoclinic crystal system. Real and imaginary parts of complex impedance and modulus properties of the materials were investigated by using complex impedance spectroscopy (CIS) technique. At a higher temperatures, the observed double semicircles in electric modulus plots confirms about the formation of samples in single phase and indicates the presence of both bulk and grain boundary contributions. Impedance analysis indicates the presence of mostly bulk (grain) resistive contributions in the materials at higher temperatures whereas complex modulus plots shows the presence of grains as well as grain boundary contributions in the materials. It is due to the fact that impedance plot highlights the phenomenon with largest resistance whereas electric modulus plot highlights the phenomenon with smallest capacitance. Due to the large difference between resistive values of grains and grain boundaries, it is not possible to get two semicircles on the same impedance plot. Both impedance and modulus analysis support the typical behavior of negative temperature coefficient of resistance (NTCR) of the materials. They also confirm the presence of non-Debye type of relaxation phenomenon in the materials. References [1] S.-E. Park, T.R. Shrout, J. Appl. Phys. 82 (1997) 1804–1811. [2] E. Ringgaard, T. Wurlitzer, J. Eur. Ceram. Soc. 25 (2005) 2701–2706. [3] S. Li, Y. Yue, X. Ning, M. Guo, M. Zhang, J. Alloys Comp. 586 (2014) 248–256.

[4] G. Shirane, R. Newnham, R. Pepinsky, Phys. Rev. 96 (1954) 581. [5] L. Egerton, D.M. Dillon, J. Am. Ceram. Soc. 42 (1959) 438–442. [6] P.W. Barnes, M.W. Lufaso, P.M. Woodward, Acta Crystallogr., Sect. B: Struct. Sci. 62 (2006) 384–396. [7] R. Jaeger, L. Egerton, J. Am. Ceram. Soc. 45 (1962) 209–213. [8] R.A. Bucur, I. Badea, A.I. Bucur, S. Novaconi, J. Alloys Comp. 630 (2015) 43–47. [9] M.-S. Chae, J.-H. Koh, S.-K. Lee, J. Alloys Comp. 587 (2014) 729–732. [10] W. Jo, R. Dittmer, M. Acosta, J. Zang, C. Groh, E. Sapper, K. Wang, J. Rödel, J. Electroceram. 29 (2012) 71–93. [11] M.S. Kim, S.J. Jeong, J.S. Song, J. Am. Ceram. Soc. 90 (2007) 3338–3340. [12] Y. Saito, H. Takao, T. Tani, T. Nonoyama, K. Takatori, T. Homma, T. Nagaya, M. Nakamura, Nature 432 (2004) 84–87. [13] B. Noheda, D. Cox, G. Shirane, J. Gao, Z.-G. Ye, Phys. Rev. B 66 (2002) 054104. [14] Y. Guo, K.-I. Kakimoto, H. Ohsato, Appl. Phys. Lett. 85 (2004) 4121–4123. [15] X. Niu, J. Zhang, L. Wu, P. Zheng, M. Zhao, C. Wang, Solid State Commun. 146 (2008) 395–398. [16] Y. Guo, K.-i. Kakimoto, H. Ohsato, Mater. Lett. 59 (2005) 241–244. [17] J.-J. Zhou, J.-F. Li, L.-Q. Cheng, K. Wang, X.-W. Zhang, Q.-M. Wang, J. Eur. Ceram. Soc. 32 (2012) 3575–3582. [18] Y.E. Roginskaya, Y.Y. Tomashpol’Skii, Y.N. Venevtsev, V. Petrov, G. Zhdanov, Sov. J. Exp. Theor. Phys. 23 (1966) 47. [19] F. Kubel, H. Schmid, Acta Crystallogr., Sect. B: Struct. Sci. 46 (1990) 698– 702. [20] Y.-K. Jun, W.-T. Moon, C.-M. Chang, H.-S. Kim, H.S. Ryu, J.W. Kim, K.H. Kim, S.H. Hong, Solid State Commun. 135 (2005) 133–137. [21] M. Popa, D. Crespo, J.M. Calderon-Moreno, S. Preda, V. Fruth, J. Am. Ceram. Soc. 90 (2007) 2723–2727. [22] C. He, Z.-J. Ma, B.-Z. Sun, R.-J. Sa, K. Wu, J. Alloys Comp. 623 (2015) 393– 400. [23] M. Mahesh Kumar, A. Srinivas, S. Suryanarayanan, T. Bhimasankaram, Phys. Status Solidi (a) 165 (1998) 317–326. [24] W. Eerenstein, F. Morrison, J. Dho, M. Blamire, J. Scott, N. Mathur, Science 307 (2005). 1203-1203. [25] J. Wang, J. Neaton, H. Zheng, V. Nagarajan, S. Ogale, B. Liu, D. Viehland, V. Vaithyanathan, D. Schlom, U. Waghmare, Science 299 (2003) 1719–1722. [26] M.-S. Kim, S. Jeon, D.-S. Lee, S.-J. Jeong, J.-S. Song, J. Electroceram. 23 (2009) 372–375. [27] K.G. Webber, R. Zuo, C.S. Lynch, Acta Mater. 56 (2008) 1219–1227. [28] D. Kobor, B. Guiffard, L. Lebrun, A. Hajjaji, D. Guyomar, J. Phys. D: Appl. Phys. 40 (2007) 2920. [29] J.R. Mcdonald, A Wiley-Interscience Publication, John Wiley & Sons, New York, 1987. [30] S.K. Barik, R. Choudhary, A. Singh, Adv. Mater. Lett. 2 (2011) 419–424. [31] P. Jakes, E. Erdem, R.-A. Eichel, L. Jin, D. Damjanovic, Position of defects with respect to domain walls in Fe 3+-doped Pb [Zr 0.52 Ti 0.48] O 3 piezoelectric ceramics, Appl. Phys. Lett. 98 (2011) (072907-072907-072903). [32] D. Varshney, A. Kumar, K. Verma, J. Alloys Comp. 509 (2011) 8421–8426. [33] P. Kumar, M. Kar, J. Alloys Comp. 584 (2014) 566–572. [34] Z.-G. Ye, A.A. Bokov, J. Adv. Diel. 02 (2012) 1241010. [35] L. Khemakhem, I. Kriaa, M. Derbel, N. Abdelmoula, Int. J. (2009). [36] P. Uniyal, K.L. Yadav, J. Phys.: Condens. Matter 21 (2009) 012205. [37] L. Shu, X. Wei, L. Jin, Y. Li, H. Wang, X. Yao, Appl. Phys. Lett. 102 (2013) 152904. [38] R. Rai, I. Coondoo, R. Rani, I. Bdikin, S. Sharma, A.L. Kholkin, Curr. Appl. Phys. 13 (2013) 430–440. [39] A. Shukla, R. Choudhary, A. Thakur, Therm. J. Phys. Chem. Solids 70 (2009) 1401–1407. [40] A. Shukla, R. Choudhary, A. Thakur, J. Mater. Sci. Mater. Electron. 20 (2009) 745–755. [41] J. Liu, C.-G. Duan, W.-G. Yin, W.-N. Mei, R.W. Smith, J.R. Hardy, Phys. Rev. B 70 (2004) 144106. [42] K. Prasad, K. Kumari, K. Chandra, K. Yadav, S. Sen, Mater. Sci. – Poland 27 (2009) 373–384. [43] D.K. Sharma, R. Kumar, R. Rai, S. Sharma, A.L. Kholkin, J. Adv. Diel. 2 (2012) 1250002.