X-ray diffuse scattering associated with ferroelectric microregions in KTa1-xNbxO3


H. Abe1, 2, K. Harada3, R. J. Matsuo3, H. Uwe3 and K. Ohshima3

1Department of Materials Science and Engineering, National Defense Academy, 1-10-20 Hashirimizu, Yokosuka 239-8686, Japan
2Department of Physics, University of Houston, Houston, 77204-5506 Texas, USA
3Institute of Materials Science, University of Tsukuba, 1-1-1 Tennoudai, Tsukuba 305-8573, Japan

J. Phys.: Condens. Matter 13, 3257 (2001).


  X-ray diffuse scattering for single crystals of KTa1-xNbxO3 (x=0, 0.007 and 0.011) has been measured. For the sample of x= 0.011, peculiar X-ray diffuse scattering is observed above the phonon-frozen-in temperature (30K). By an analysis of the diffuse scattering intensities, the average radius of the ferroelectric microregions is found to grow up to around 4 nm with decreasing temperature from 100K down to 30 K. Anisotropic distributions of the diffuse scattering indicate off-centered shift of Nb and Ta ions to the<111>-direction. Moreover, lattice constants for x=0.011 below 30 K exhibit negative expansion which is found neither in pure KTaO3 nor in KTa1-xNbxO3 (x=0.007) at low-temperatures. These results are successfully explained by introducing the effective classical potential (ECP) method for the anharmonic oscillators. The quantum contribution to mean square displacement of Nb and Ta ions from an average position is also calculated by the ECP method.

 
Figure 1. Lattice constants of KTa1-xNbxO3 as a function of Nb composition at room temperature. 
Figure 2. (a) Lattice constants of KTa1-xNbxO3 as a function of temperature. Open circles, open squares and solid circles are corresponding to x=0.000, x=0.007 and x=0.011, respectively. Each solid line is calculated by using the effective classical potential (ECP) method (see text), where only a solid curve in pure KTaO3 is calculated using parameters in a previous paper [1]. (b) The temperature dependence of the isotropic temperature factor, B(T). Open circles, open squares and solid circles are corresponding to x=0.000, x=0.007 and x=0.011, respectively. B(T) is determined by both of temperature dependence of integrated intensities and lattice constants. Solid curves are calculated by using ECP method.
Figure 3. Temperature dependence of full width at half maximum (FWHM) of 002 Bragg reflection along the radial direction in KTa1-xNbxO3 (x=0.011) (open circles). FWHM also increases below 30 K. To emphasize the anomaly below 30 K, lattice constants (closed circles) are also seen.
Figure 4. Temperature dependence of diffuse scattering at  =0.629 along the G002 + q [ƒÄ  0 0] direction in KTa1-xNbxO3 (x=0.011), where the unit of q is reciprocal lattice one. A solid line shows the summation of the calculated thermal diffuse scattering and Huang diffuse scattering. While, a broken line shows the calculated thermal diffuse scattering. 
Figure 5. Temperature dependence of the calculated radius of FMR cluster, Rcl. Rcl values along various direction are analyzed by the self-consistent Monte Carlo method using the observed diffuse scattering.
Figure 6. Temperature dependence of (a) the observed diffuse scattering, (b) the calculated one, (c) the calculated Huang diffuse scattering (HDS) and (d) the calculated thermal diffuse scattering (TDS) along the G002 + q [ƒÄ  0 0] in KTa1-xNbxO3 (x=0.011).
Figure 7. Temperature dependence of the observed diffuse scattering at 60 K along transverse- and longitudinal-directions around 002, 202 and 103 Bragg reflections.
Figure 8. The calculated distributions of Huang diffuse scattering around 202 Bragg reflection using the shift of off-centered Nb, whose direction is (a) [100]-direction, (b) [110]-direction and (c) [111]-direction. (d) For comparison, the calculated distribution of thermal diffuse scattering around 202 Bragg reflection is also shown.
Figure 9. q dependence of Huang diffuse scattering along the G002 + q [0 0  ] at 60 K. An crossover point, qc, between q-2 and q-4, can provide the average size of FMR. It is estimated to be around 10 nm.


References

[1] Vogt H 1995 Phys. Rev. B51 8046
[2] Uwe H 1985 Jpn. J. Appl. Phys. Suppl. 24-2 513
[3] Uwe H, Lyons K B, Carter H L  and Fleury P A 1986 Phys. Rev. B33 6436
[4] Klink J J, Rod S and Chatelain A 1986 Phys. Rev. B33 2084
[5] Stachiotti M G and Migoni R L  1990 J. Phys.: Condens. Matter 2 4341
[6] Rytz D, Hochli U T and Bilz H 1980 Phys. Rev. B22 359
[7] Rytz D and Scheel H J 1982 J. Cryst. Growth 59 468
[8] Gehring P M, Chow H, Shapiro S M, Hriljac J A, Chen D H, Toulouse J, Rytz D and Boatner L A 1992 Phys. Rev. B46 5116
[9] Kleenman W, Schafer F J and Rytz D 1985 Phys. Rev. Lett. 54 2038
[10] Ang C, Yu Z, Vilarinho P M and Baptista J L 1998 Phys. Rev. B57 7403
 [11] Chou H, Shapiro S M, Lyons K B, Kejems J and Rytz D 1990 Phys. Rev. B41 7231
[12] Gehring P M, Chow H, Shapiro S M, Hriljac J A, Chen D H, Toulouse J, Rytz D and L. Boatner A 1993 Ferroelectrics 150,  47
[13] Axe J D, Harada J and Shirane G 1970 Phys. Rev. B1 1227
[14] Comes R and Shirane G 1972 Phys. Rev. B5 1886
[15] Andrews S R 1985 J. Phys. C18 1357
[16] Warren B E 1990 X-RAY DIFFRACTION (New York: Dover)
[17] Wooster W A 1997 Diffuse X-ray Reflection from Crystals (New York: Dover)
[18] Dederichs P H 1973 J. Phys. F3 471
[19] Barrett H H 1968 Phys. Lett. 26A 217
[20] Perry C H, Currant R, Buhay H, Migoni R M, Stirling W G and Axe J D 1989 Phys. Rev. B39 8666
[21] Landolt Boernstein 16, p. 408 (Springer, 1981)
[22] Eglitis R I, Kotomin E A, Borstel G and Dorfman S 1998 J. Phys.: Condens. Matter. 10 6271

Back to Publications List


ab@nda.ac.jp
Department of Materials Science and Engineering
National Defense Academy

Last Modified: April 1, 2009