Quantum critical points in ferroelectric relaxors : stuffed tungsten bronze K3Li2Ta5O15 and lead pyrochlore (Pb2Nb2O7)

Abstract

We have synthesized ceramic specimens of the tetragonal tungsten bronze K3Li2Ta5O15 (KLT) and characterized its phase transition via X-ray diffraction, dielectric permittivity, resonant ultrasonic spectroscopy, and heat capacity measurements. The space group of KLT is reported as both P4/mbm and Cmmm with the orthorhombic distortion occurring when there are higher partial pressures of volatile K and Li used inside the closed crucibles for the solid state synthesis. The data show strong relaxor behavior, with the temperature at which the two dielectric relative permittivity peaks decreasing, with 104 ≥ Tm1 ≥ 69 K and 69 ≥ Tm2 ≥ 46 K as probe frequency f is reduced from 1 MHz to 316 Hz. F tests show that the data satisfies a Vogel-Fulcher model better than Arrhenius with an extrapolated freezing temperature for ε’ and ε” of Tf1 = +15.8 and –11.8 K and Tf2 = –5.0 and –15.0 K for f -> 0 (tending to dc). This difference between Tf from real and imaginary values, albeit counterintuitive, is mandatory, according to the theory of Tagantsev. Therefore, by tuning frequency, the transition could be shifted to absolute zero, suggesting KLT has a relaxor-type quantum critical point. In addition, we have reanalyzed the conflicting literature for Pb2Nb2O7 pyrochlore which suggests that this also has a relaxor-type quantum critical point since the freezing temperature from the Vogel-Fulcher fitting is below absolute zero. Since the transition temperature evidenced in the dielectric data at approximately 100 kHz shifts below 0 K for very low frequencies, this transition would not be seen with heat capacity data collected in the zero-frequency (dc) limit. Both of these materials show promise for possible new relaxor-type quantum critical points with non-perovskite based structures.