EDP Sciences logo
Subscriber Authentication Point
Search
Menu
All issues Volume 80 / No 1 (October 2017) Eur. Phys. J. Appl. Phys., 80 1 (2017) 10401 Full HTML
Issue
Eur. Phys. J. Appl. Phys.
Volume 80, Number 1, October 2017
Article Number 10401
Number of page(s) 5
Section Nanomaterials and Nanotechnologies
DOI https://doi.org/10.1051/epjap/2017170054
Published online 02 October 2017

© EDP Sciences, 2017

1 Introduction

Increasing attention to ferroelectric nanostructures is caused by the broad application prospects of their use in non-volatile electronic memories, sensors, pyroelectric and electrooptic devices [1]. Functional properties of these materials are based on an existence of the stable ferroelectric phase. Barium titanate (BaTiO3) is one of the most widely known materials used for practical applications.

However, ferroelectricity does not appear in ultrafine particles due to the so-called “size effect”. According to references [2,3], the critical size of BaTiO3 nanoparticles varies from 9 nm to more than 100 nm, depending on the method of their synthesis. The difference in the values of the critical size are caused by the presence of different defects and different their concentration. These defects are reagents and resultants of reactions, which predominantly occur due to the nonequilibrium conditions of nanoparticles preparation [3–5].

Thermal treatment in air atmosphere lead to the lattice defects disappearance that results in an emergence of ferroelectric properties in nanoparticles [6].

Thus, the purpose of this paper is to study structure and dielectric properties of BaTiO3 nanoparticles evolution during their heat treatment.

2 Methods

Samples in the form of disc were prepared by compacting of barium titanate powder (Sigma–Aldrich, USA). Initial particles of cubic perovskite modification of BaTiO3 had a spherical shape with 100 nm average diameter (Fig. 1a). Obtained samples were annealed in the electrical resistive furnace at air atmosphere under the following scheme: 700 °C (1 h) → 1000 °C (5 h) → 1200 °C (1 h).

X-ray diffraction (XRD) patterns of the materials under study (CuKα1 radiation) were obtained in the range 2θ of 10°–70° with the step scan 0.02°, using CBG EMMA diffractometer. The microstructure of the samples was studied by FESEM (Quanta 650). FTIR spectra (Nicolet iS-50) were obtained in transmission mode. A KBr pellet was prepared using a manual press (CrushIR, PikeTechnologies Inc.) for the background spectrum. 1 wt.% BaTiO3 nanoparticles (before and after thermal treatment) were homogeneously mixed with KBr for spectroscopic measurement.

For the studies of dielectric properties, the samples in the form of disc (diameter and thickness were 10 mm and 1 mm, respectively) were used. The electrodes were made by silver paste coating. The sample was placed before experiments in a thermostat where the temperature was varied from 20 to 300 °C and measured with an error less than ±1 °C. Measurements of dielectric permittivity (ε) and dielectric losses (tgδ) were performed at slow heating/cooling (1.5–2 °C/min) using LCR-meter E7-20.

thumbnail Fig. 1

SEM image of nanoparticles: (a) in the initial state and (b) after the thermal treatment at 1000 °C.

3 Results and discussion

Diffraction patterns of samples obtained by the heat treatment of initial nanoparticles at the temperatures 700 °C, 1000 °C and 1200 °C are presented in Figure 2. The splitting of 002/200 reflex observed for the samples after their annealing at temperatures of 1000 °C, indicates formation of the tetragonal crystal phase.

Parameters of the tetragonal cell are c ≈ 4.0301 Å and 4.0335 Å, and a ≈ 4.0335 Å and 3.996 Å for the samples annealed previously at 1000 °C and 1200 °C, respectively. Annealing leads to the appearance and increase of the tetragonal distortion: c/a ≈ 1.0085 and 1.0097 for the samples annealed at 1000 °C and 1200 °C, respectively. Structural variations are accompanied by some reduction of the unit cell volume V. It was found that V ≈ 64.45, 64.35 and 64.35 Å3 for the initial sample and after its annealing at 1000 °C and 1200 °C, respectively.

At the same time, the crystalline structure of the sample after the thermal treatment at 700 °C remains the cubic one.

An average particle size has been determined by the broadening of the reflexes of diffraction patterns accordingly to the Debye–Scherer equation: (1) where k is the constant equal to 0.9, λ = 0.1540598 nm for CuKα1 line, B is the reflex broadening on its half height and θ is the Bragg angle. The resulting estimates of particle sizes are about 100 nm for all the samples. Thus, the heat treatment of BaTiO3 nanoparticles under experimental conditions used in this work does not lead to any visible increasing of their sizes.

However, the SEM image showed a change in shape and size of particles after the thermal anneling. Some particles increased in their size up to ≈400 nm (Fig. 1b). Taking into account the results of X-ray analysis, we suppose that the coarse granules are agglomerates of small crystallites.

Let us discuss the peculiarities of FTIR spectra of both initial and annealed samples (Fig. 3). One can see the three broad absorption bunds are located within the range 400–3700 cm−1. The absorption band 520 cm−1 emerges due to the stretching vibrations of the Ti─O bond. The line in the vicinity of 1435 cm−1 can be caused by presence of carbonates [7]. The broad absorption line near 3400 cm−1 is associated with the stretching O─H vibrations of hydroxyl groups located both at the surface and at the inside of crystallites (the so-called lattice hydroxyl groups) [8]. The peak in the vicinity of 2931 cm−1 in the FTIR spectrum observed for initial nanoparticles unambiguously shows the presence of lattice hydroxyl groups [8]. These groups cause the increase of unit cell volume and appearance of barium vacancies (2[OHO•] = VBa″) [9].

Annealing of nanoparticles leads to the significant decrease of the intensity of the peak 1435 cm−1 and the disappearance of absorption lines near 3400 and 2931 cm−1 due to the OH-groups, that indicates the removal of these defects.

Temperature dependences of ε for the samples under study are shown in Figure 4. Curves 1a and 1b correspond to the BaTiO3 sample previously annealed at temperature 700 °C during 1 h.

It is seen that the dependence of ε(T) obtained at heating passes through a maximum at the temperature near 52 °C. The maximum of dielectric permittivity observed at cooling is significantly lower and shifted to high temperatures (≈62 °C).

The shapes of these maxima are quite different from the shape of he ε(T) dependence observed for the bulk BaTiO3 at the ferroelectric phase transition [10–12]. Therefore, the obtained dielectric anomalies are not related to any ferroelectric phase transition in the material. Probably, they are caused by the release of traps charge carriers at heating. Really, the maximum of ε appearing during heating disappears practically at cooling.

The heat treatment at temperatures 1000 and 1200 °C leads to the substantial increase of the dielectric permittivity and appearance of the characteristic peak of ε at Curie temperature TC ≈ 120 °C, as shown in Figure 4. The Curie temperature depends on the regime of measurement: heating or cooling, that is characteristically for first order ferroelectric phase transition. Corresponding values of TC will denote as TCh and TCc. The temperatures are TCh ≈ 138 and 133 °C, and TCc ≈ 119 and 121 °C for the samples annealing previously at 1000 and 1200 °C, respectively.

The observed thermal hysteresis of the Curie temperature indicates the first order phase transition. The width of hysteresis is ΔTC = TCh−TCc ≈ 19 and 12 °C for the samples previously annealed at 1000 and 1200 °C, respectively.

It should be noted that the obtained values of ΔTC exceed the width of thermal hysteresis for the bulk crystals BaTiO3, where ΔTC ≈ 7 °C [10]. We suppose that this anomalously broad thermal hysteresis of the Curie temperature emerges due to the interaction of interphase boundaries and “random field” type lattice defects, similarly to the case of ferroelectrics with incommensurate phase [13]. These defects pins boundaries of polar phase and stabilizes it above phase equilibrium temperature. Therefore, TCh is decreased after thermal treatment, which not influences practically on TCh. Thus the thermal treatment leading to decreasing of the lattice defect concentration results in a decrease of ΔTC.

The study of the influence of external d.c. electric field (E=) on (TCh) in the BaTiO3 sample annealed at 1000 °C speaks in favor of the supposition mentioned above. The obtained dependence of  δTCh = TCh (E=) − TCh (E= = 0) is showed in Figure 5.

This dependence differs quite drastically from the similar dependence observed for the BaTiO3 ceramic, where TC increases linearly with the bias electric field E= [11]: (2) where A ≈ 1.4 °C/kV [11].

However, the experimental TCh(E=) curve is typical for ferroelectric materials with a high concentration of defects of type “random field” [14–16]. Note that the polar microregions near TC are under influence of random fields induced by lattice defects in such ferroelectrics. The Curie temperature does not shift until the bias field E= will exceed the intensity of the internal random fields. When the field E= exceeds this threshold, the rapid increase of TCh occurs.

Above the Curie temperature, ε(T) dependence follows the Curie–Weiss law [11,12]: (3) where ε1 is the temperature-independent component of the dielectric permittivity, CCW is the Curie–Weiss constant and θ is the Curie–Weiss temperature.

An applicability of the expression (3) to describe the experimental results for the samples undergoing ferroelectric phase transition are illustrated in Figure 6. The best approximation of our experimental data by the expression (3) was obtained at following parameters: CCW ≈ 21 370 K and 77 950 K; θ ≈ 98 and 90 K and ε1 ≈ 85 and 10, respectively, for the samples annealed at 1000 and 1200 °C. However, it should be noted that the values of the Curie–Weiss constant for the samples studied, are significantly lower than in the bulk BaTiO3, where CCW ≈ 142 000 K [12].

thumbnail Fig. 2

Diffraction patterns for BaTiO3 samples after thermal treatment at 700 °C (a), 1000 °C (b) and 1200 °C (c).
thumbnail Fig. 3

FTIR spectra of initial nanoparticles (dotted line) and nanoparticles after their thermal treatment at the 1000 °C (solid line).
thumbnail Fig. 4

Temperature dependences of ε observed at 10 kHz during heating (1a–3a) and cooling (1b–3b) for samples annealed previously at 700 (1a and 1b), 1000 (2a and 2b) and 1200 °C (3a and 3b).
thumbnail Fig. 5

Dependence δTCh(E=) for BaTiO3 sample previously annealed during 5 h at the temperature of 1000 °C.
thumbnail Fig. 6

Temperature dependences of (ε−ε1)−1, obtained at heating for nanostructured BaTiO3, annealed previously during 5 h at the temperatures of 1000 °C (a) and 1200 °C (b).

4 Conclusions

Analysis of the experimental results allows the following conclusions:

  • thermal annealing of BaTiO3 nanoparticles with initially cubic perovskite crystalline lattice at 1000 °C during 5 h leads to the formation of ferroelectric tetragonal phase. It was found that the increase of tetragonal distortions increase simultaneously with the annealing temperature (c/a ≈ 1.0085 and 1.00976 for the samples annealed at 1000 and 1200 °C respectively);

  • heat treatment of BaTiO3 nanoparticles under experimental conditions does not lead to any visible increasing of their sizes. However, this annealing leads to the formation of small crystallites agglomerates;

  • appearance of ferroelectricity in BaTiO3 nanoparticles is not produced by the increase of their sizes, but caused by the lattice defects concentration decrease owing to thermal annealing;

  • above the Curie temperature ε(T), the dependence satisfies the Curie–Weiss law. Increasing of the thermal treatment temperature leads to the increase of the Curie–Weiss constant;

  • temperature hysteresis of dielectric permittivity observed at cyclic temperature variation in the vicinity of Curie temperature indicates the first order phase transition in the annealed BaTiO3 nanoparticles. The width of thermal hysteresis decreases after high-temperature annealing of the sample. It shows that the anomalously broad hysteresis of the dielectric permittivity emerges due to an interaction of interphase boundaries and “random field” type lattice defects.

Acknowledgments

The authors are grateful to Prof. A.P. Kuzmenko and P.A. Abakumov (Southwest State University, Kursk, Russia) for completed studies of the FTIR spectra.

Nikita A. Emelianov is supported by the Ministry of Education and Science of the Russian Federation within research project #3.9499.2017 included into the basic part of research funding assigned to Kursk State University.

References

  1. C.H. Ahn, K.M. Rabe, J.M. Triscone, Physics of Ferroelectrics: A Modern Perspective (Springer-Verlag, Berlin, 2007) [Google Scholar]
  2. V. Fridkin, S. Ducharme, Ferroelectricity at the Nanoscale: Basic and Applications (Springer-Verlag, Berlin, Heidelberg, 2014) [CrossRef] [Google Scholar]
  3. M.D. Glinchuk, A.V. Ragulya, Nanoferroics (Naukova Dumka, Kiev, 2010) [Google Scholar]
  4. C. Fang, D.-X. Zhou, S.P. Gong, Physica B 406, 1317 (2011) [CrossRef] [Google Scholar]
  5. Ch.-F. Chen, G. King, R.M. Dickerson, P.A. Papin, Sh. Gupta, W.R. Kellogg, G. Wu, Nano Energy 13, 423 (2015) [CrossRef] [Google Scholar]
  6. X. Wei, Y. Li, J. Ceram. Process. Res. 6, 250 (2005) [Google Scholar]
  7. S. Chang, W. Liao, C. Ciou, J. Lee, C. Li, J. Colloid Interface Sci. 329, 300 (2009) [CrossRef] [PubMed] [Google Scholar]
  8. R. Kota, B.I. Lee, J. Mater. Sci.: Mater. Electron. 18, 1221 (2007) [CrossRef] [Google Scholar]
  9. J. Adam, G. Klein, T. Lehnert, J. Am. Ceram. Soc. 96, 2987 (2013) [CrossRef] [Google Scholar]
  10. B. Jaffe, W.R. Cook, H. Yaffe, Piezoelectric Ceramics (Academic Press, London, New York, 1971) [Google Scholar]
  11. F. Iona, G. Shirane, Ferroelectric Crystals (Pergamon Press, Oxford-London-New York-Paris, 1962) [Google Scholar]
  12. L.E. Lines, A.M. Glass, Principles and Application of Ferroelectrics and Related Materials (Clarendon, Oxford, 1977) [Google Scholar]
  13. K. Hamano, T. Ikeda, T. Fujimoto, J. Phys. Soc. Jpn. 49, 2278 (1980) [CrossRef] [Google Scholar]
  14. L.N. Korotkov, Phys. Status Solidi (b) 222, R1 (2000) [CrossRef] [Google Scholar]
  15. S.N. Dorogovtsev, Sov. Phys.: Solid State Phys. 24, 948 (1982) [Google Scholar]
  16. M.D. Glinchuk, V.A. Stephanovich, J. Phys.: Condens. Matter 9, 11081 (1998) [CrossRef] [Google Scholar]

Cite this article as: Leonid N. Korotkov, Wisam M. Mandalavi, Nikita A. Emelianov, Jose Angel Roldan Lopez, Influence of the thermal treatment on structure and dielectric properties of nanostructured BaTiO3, Eur. Phys. J. Appl. Phys. 80, 10401 (2017)

All Figures

thumbnail Fig. 1

SEM image of nanoparticles: (a) in the initial state and (b) after the thermal treatment at 1000 °C.
In the text
thumbnail Fig. 2

Diffraction patterns for BaTiO3 samples after thermal treatment at 700 °C (a), 1000 °C (b) and 1200 °C (c).
In the text
thumbnail Fig. 3

FTIR spectra of initial nanoparticles (dotted line) and nanoparticles after their thermal treatment at the 1000 °C (solid line).
In the text
thumbnail Fig. 4

Temperature dependences of ε observed at 10 kHz during heating (1a–3a) and cooling (1b–3b) for samples annealed previously at 700 (1a and 1b), 1000 (2a and 2b) and 1200 °C (3a and 3b).
In the text
thumbnail Fig. 5

Dependence δTCh(E=) for BaTiO3 sample previously annealed during 5 h at the temperature of 1000 °C.
In the text
thumbnail Fig. 6

Temperature dependences of (ε−ε1)−1, obtained at heating for nanostructured BaTiO3, annealed previously during 5 h at the temperatures of 1000 °C (a) and 1200 °C (b).
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.