APPLIED PHYSICS LETTERS
VOLUME 76, NUMBER 14
3 APRIL 2000
Dielectric loss of SrTiO3 single crystals under direct current bias Chen Ang,a) A. S. Bhalla, Ruyan Guo, and L. E. Cross Materials Research Laboratory, The Pennsylvania State University, University Park, Pennsylvania 16802
共Received 16 November 1999; accepted for publication 6 February 2000兲 The dielectric behavior of SrTiO3 single crystals under high dc electric field 共up to 50 kV/cm兲 is reported in this letter. The rounded dielectric constant peaks are induced by the application of dc bias, and the corresponding dielectric losses are observed. The results show that dielectric loss under dc bias consists of several components coming from ‘‘defects mode’’ and ‘‘induced mode.’’ The field dependence of these modes is studied and their physical nature is discussed. © 2000 American Institute of Physics. 关S0003-6951共00兲04314-X兴
In order to meet the needs of continuously increasing development in communication, field and frequency agile materials for microwave electronics 共FAME兲 are desirable.1,2 Currently the most studied system for FAME is the perovskite SrTiO3. 3–6 SrTiO3 shows a high nonlinear electric-field effect at low temperatures and a reasonably low dielectric loss.3,4 However, it is also found that when SrTiO3 is made as a thin film, the dielectric loss is increased by more than an order of magnitude.5–7 Understanding of the higher losses in thin film SrTiO3 is not very convincing yet. In fact, knowledge concerning the dielectric loss under high dc electric field for single crystal is almost lacking. To achieve dielectric tuning in thin films very high fields are often applied. This is possible because of the enhanced breakdown strength of the films and the possibility to achieve high electric field at low terminal voltages in these very thin samples. The understanding of the physical nature of dielectric loss in SrTiO3, especially under high dc electric field, is highly desirable with the current importance of the FAME materials. Pure SrTiO3 is an intrinsic quantum paraelectric.8,9 It is known that permittivity peaks can be induced in SrTiO3 by the application of external electrical field.10 The effect of dc electric field on dielectric behavior of SrTiO3 single crystals has been extensively studied by several groups.10–12 However, the dielectric loss or a good understanding of it under dc electric field has not been reported. In this letter, we measured the effect of dc electric field up to 50 kV/cm on dielectric properties of SrTiO3 single crystals, in an effort to obtain more experimental data to enhance our understanding of the electric-field dependence of the dielectric loss and the physical mechanisms responsible. The single crystal samples with 共100兲 polished surfaces were obtained from commercial source. Complex dielectric permittivity was measured using an HP 4284A LCR meter with an ac field of 1 V/mm parallel to the 关100兴 direction. Temperature dependence of the dielectric properties was measured in a cryostat system in the temperature range 12– 300 K, while the specimen was being cooled or heated up at a cooling/heating rate of 1 K/min and readings were taken at
every 1 or 2 K interval. The dc voltage is applied to the samples and a blocking circuit is adopted to separate the high dc voltage from the LCR meter. Temperature T dependence of the dielectric constant and dielectric loss (tan ␦) as a function of frequency for the SrTiO3 sample are shown in Fig. 1. From 300 to 12, K, increases continuously with decreasing temperature attaining the value ⫽10 300 at 12 K. However, for tan ␦, in the temperature range 50–100 K, there are a set of peaks with frequency dispersive behavior. With further decreasing temperature, as T⬍50 K, the tan ␦ increases sharply corresponding to the increases in . Temperature dependence of the dielectric constant and loss under different dc fields 共0–50 kV/cm兲, is shown in Figs. 2 and 3. The results show that a dielectric peak is induced by applying dc electrical field; this is similar to those widely reported results in the earlier literature.10–12 For the dielectric loss, the curve shows more complicated behavior. In the temperature range of 12–120 K and the field range of 2–20 kV/cm, as shown in Fig. 2, there are several sets of peaks under dc electric-field: 共1兲 mode I around 75 K and 共2兲 mode II around 50 K. The T m of both modes I and II are field independent. Below 50 K, it seems several peaks overlap. As suggested in the earlier
a兲
Author to whom correspondence should be addressed; on leave from Department of Physics, Zheijiang University, Hangzhou, 310027, P.R. China; electronic mail:
[email protected]
0003-6951/2000/76(14)/1929/3/$17.00
FIG. 1. Temperature dependence of and tan ␦ as function of frequency for as-grown SrTiO3 single crystal. 1929
© 2000 American Institute of Physics
1930
Appl. Phys. Lett., Vol. 76, No. 14, 3 April 2000
Ang et al.
FIG. 4. Phase diagram of the dielectric relaxation modes for different electric fields 0–50 kV/cm.
FIG. 2. Temperature dependence of the 共upper curve兲 and tan ␦ 共lower curve兲 for the SrTiO3 single crystal for the different electric fields from 0 to 20 kV/cm at 10 kHz. Solid curves: peak A, dashed curves: mode III.
literature,10–12 the peak induced by the electric-field in the dielectric constant indicates an occurrence of a ‘‘ferroelectric phase transition,’’ i.e., the peak could be recognized as a ‘‘ferroelectric peak.’’ Generally, for a ferroelectric peak, the corresponding loss peak should occur at the same temperature. Therefore, it is reasonable to assign a loss peak that occurs at the temperature of the ferroelectric peak in the
FIG. 3. Temperature dependence of the 共upper curve兲 and tan ␦ 共lower curve兲 for the SrTiO3 single crystal for the different electric fields from 25 to 50 kV/cm at 3 kHz.
dielectric constant, as peak A 共shown as solid curves in Fig. 2兲. In addition to peak A, another peak is also noticeable, whose T m is found to be field independent 共denoted as mode III, shown as a dashed line兲. With further increasing field, from 25 to 50 kV/cm 共see Fig. 3兲, it is observed that modes I, II, and III almost disappear, and only one peak remains, whose temperature corresponds exactly to that of the dielectric constant maximum; hence, this peak is peak A. Figure 4 shows a summary of the dielectric relaxation modes for the different electric fields 0–50 kV/cm in the form of a phase diagram. The T m for three modes, I, II, and III does not change with electric field. However, modes I, II, and III vanish at high electric fields. The T m of peak A shifts to higher temperatures with increasing electric fields, which is the only one that stays with the increase in electric field E⭓25 kV/cm. The relaxation modes I 共⬃75 K兲 and III 共⬃28 K兲 observed have also been observed in single crystals by Mizaras and Loidl13 and Viana et al.;14 in polycrystalline samples by Chen et al.,15,16 and in thin films by Li et al.,17 Mizaras and Loidl attributed mode III to the dynamic response of the domain walls that occur at the cubic-to-tetragonal phase transition.9,18 Viana et al. attributed mode III to a possible coherent state.14,19 Mode II has not been observed or discussed by the previous workers. However, it is recognized that the possibility of the existence of the quantum coherent state around 37 K is still an open question. Mu¨ller et al.19 suggested that it is a static response rather than a dynamic response, but mode III obviously exhibits a dynamic behavior. Mode III thus cannot be explained as a quantum coherent state. In addition, the important and common characteristic of modes I, II, and III is their field-independent T m . Similar dielectric modes were also reported in pure and Bi doped SrTiO3 with T m independent of the concentration of the doping impurity.15,16 This indicates that besides the contribution of the defects 共or impurities兲, an intrinsic mechanism should be involved. It is well known that SrTiO3 is a typical soft mode quantum paraelectric. In nominally pure SrTiO3, low levels of unavoidable defects and impurities 共for example, oxygen vacancies兲 may be present. Therefore, it is highly possible that such defects could interact with the soft modes at low temperatures, and contribute to dielectric relaxation behavior. In SrTiO3 doped with a small amount of Bi 共500 ppm兲,15,16 obvious enhancement of the similar relaxation process supports this conjecture. Indeed, in the soft mode quantum paraelectric KTaO3, the dielectric relaxation at low tempera-
Ang et al.
Appl. Phys. Lett., Vol. 76, No. 14, 3 April 2000
1931
that the dielectric losses consist of several components coming from the defects modes and the induced mode at low electric field; at higher electric field 共⭓25 kV/cm兲, the defects modes disappear, and only the induced mode remains. One of the authors 共C.A.兲 would like to thank Dr. Zhi Yu for the stimulating discussion. This work is supported by the Defense Advance Research Projects Agency under Grant No. DABT63-98-1-0002.
O. G. Vendik, Ferroelectrics 12, 85 共1976兲. O. G. Vendik, I. G. Mironenko, and L. T. Ter-Martirosyan, Microwaves RF 33, 67 共1994兲; O. G. Vendik, E. Kollberg, S. S. Gevorgian, A. B. Kozyrev, and O. I. Soldatenkov, Electron. Lett. 31, 654b 共1995兲. 3 A. B. Kozyrev, T. B. Samoilova, A. A. Golovkov, E. K. Hollmann, D. A. Kalinikos, V. E. Loginov, A. M. Prudan, O. I. Soldatenkov, D. Galt, C. H. Mueller, T. V. Rivkin, and G. A. Koepf, J. Appl. Phys. 84, 3326 共1998兲. 4 F. W. Van Keuls, R. R. Romanofsky, D. Y. Bohman, M. D. Winters, F. A. Miranda, C. H. Mueller, R. E. Treece, T. V. Rivkin, and D. Galt, Appl. Phys. Lett. 71, 3075 共1997兲. 5 R. E. Treece, J. B. Thompson, C. H. Mueller, T. Rivkin, and M. W. Cromar, IEEE Trans. Appl. Supercond. 7, 2363 共1997兲. 6 D. Galt, J. Price, J. A. Beall, and R. H. Ono, Appl. Phys. Lett. 63, 3078 共1993兲; D. Galt, J. Price, J. A. Beall, and T. E. Harvey, IEEE Trans. Appl. Supercond. 5, 2575 共1995兲. 7 D. Fuchs, C. W. Schneider, R. Schneider, and H. Rietschel, J. Appl. Phys. 85, 7362 共1999兲. 8 K. A. Mu¨ller and H. Burkhard, Phys. Rev. B 19, 3593 共1979兲. 9 M. E. Lines and A. M. Glass, Principle and Application of Ferroelectrics and Related Materials 共Oxford University Press, Oxford, 1977兲. 10 C. Frenzel and E. Hegenbarth, Phys. Status Solidi 23, 517 共1974兲. 11 H. Unoki and T. Sakudo, J. Phys. Soc. Jpn. 23, 546 共1967兲. 12 P. A. Fleury and J. M. Worlock, Phys. Rev. 174, 613 共1968兲, J. Hemberger, P. Lunkhemer, R. Viana, R. Bohmer, and A. Loidl, Phys. Rev. B 52, 13159 共1995兲. 13 R. Mizaras and A. Loidl, Phys. Rev. B 56, 10726 共1997兲. 14 R. Viana, P. Lunkenheimer, J. Hemberger, R. Bo¨hmer, and A. Loidl, Phys. Rev. B 50, 601 共1994兲. 15 C. Ang, J. F. Scott, Z. Yu, H. Ledbetter, and J. L. Baptista, Phys. Rev. B 59, 6661 共1999兲. 16 C. Ang, Z. Yu, J. Hemberger, P. Lunkhemer, and A. Loidl, Phys. Rev. B 59, 6665 共1999兲. 17 H.-C. Li, W. Si, A. D. West, and X. X. Xi, Appl. Phys. Lett. 73, 464 共1998兲. 18 M. Liu, T. R. Finlayson, and T. F. Smith, Phys. Rev. B 55, 3480 共1997兲. 19 K. A. Mu¨ller, W. Berlinger, and E. Tosatti, Z. Phys. B: Condens. Matter 84, 277 共1991兲. 20 B. Salce, J. L. Gravi, and L. A. Boatner, J. Phys.: Condens. Matter 6, 4077 共1994兲. 21 V. Trepakov, F. Smutny, V. Vikhnin, V. Bursian, L. Sochava, L. Jastrabik, and P. Syrnikov, J. Phys.: Condens. Matter 7, 3765 共1995兲. 1 2
FIG. 5. Dielectric loss tan ␦ of modes I, II, and III, and peak A vs electric field from 0 to 20 kV/cm at 10 kHz.
ture 共⬃40 K兲 was found to be related to the defects/impurities.20,21 In the present work, we suggest that the modes I, II and III observed, similar to the ‘‘defects mode’’ in Bi doped SrTiO3, could be attributed to the reorientation of the dipoles 共formed by the impurities and defects兲 that interact with the soft modes. The dielectric loss amplitudes of modes I, II, and III, and peak A verses electric field at 10 kHz are shown in Fig. 5. It can be seen that for all the anomalies, the loss amplitudes are varied nonmonotonously with electric field, i.e., with increasing electric field, the dielectric losses initially increase, then with further increasing electric field, the losses decrease. For modes I, II, and III, as the electric field reaches 20 kV/ cm, the typical loss levels are 0.0002–0.0004. For peak A, at ⬃5 kV/cm, the loss reaches a maximum, then decreases with increasing fields. As electric field E⭓25 kV/cm, modes I, II, and III almost disappear, and only peak A remains. Similar behavior is also observed for the ‘‘defect mode’’ in Bi doped SrTiO3. 15,16 The physical mechanism needs further study. In conclusion, the effect of dc electrical field on dielectric properties of SrTiO3 has been studied. A rounded dielectric constant peak is induced by the application of dc bias. The dielectric losses are more complicated than the dielectric constant. There are four anomalous peaks: three peaks are identified as associating with ‘‘defects modes,’’ whose T m is independent of electric field; the fourth is an electric-field dependent mode or ‘‘induced mode,’’ which corresponds to the induced peak in the dielectric constant. The results show
