Trivalent Dopant Size Influences Electrostrictive Strain in Ceria Solid Solutions

The technologically important frequency range for the application of electrostrictors and piezoelectrics is tens of Hz to tens of kHz. Sm3+- and Gd3+-doped ceria ceramics, excellent intermediate-temperature ion conductors, have been shown to exhibit very large electrostriction below 1 Hz. Why this is so is still not understood. While optimal design of ceria-based devices requires an in-depth understanding of their mechanical and electromechanical properties, systematic investigation of the influence of dopant size on frequency response is lacking. In this report, the mechanical and electromechanical properties of dense ceria ceramics doped with trivalent lanthanides (RE0.1Ce0.9O1.95, RE = Lu, Yb, Er, Gd, Sm, and Nd) were investigated. Young's, shear, and bulk moduli were obtained from ultrasound pulse echo measurements. Nanoindentation measurements revealed room-temperature creep in all samples as well as the dependence of Young's modulus on the unloading rate. Both are evidence for viscoelastic behavior, in this case anelasticity. For all samples, within the frequency range f = 0.15-150 Hz and electric field E <= 0.7 MV/m, the longitudinal electrostriction strain coefficient (vertical bar M-33 vertical bar) was 10(2) to 10(4)-fold larger than expected for classical (Newnham) electrostrictors. However, electrostrictive strain in Er-, Gd-, Sm-, and Nd-doped ceramics exhibited marked frequency relaxation, with the Debye-type characteristic relaxation time tau <= 1 s, while for the smallest dopants-Lu and Yb-little change in electrostrictive strain was detected over the complete frequency range studied. We find that only the small, less-studied dopants continue to produce useable electrostrictive strain at the higher frequencies. We suggest that this striking difference in frequency response may be explained by postulating that introduction of a dopant induces two types of polarizable elastic dipoles and that the dopant size determines which of the two will be dominant.