Resonant piezoelectric spectroscopy shows polar resonances in paraelectric SrTiO 3 at temperatures below 80 K. These resonances become strong at T < 40 K. The resonances are induced by weak electric fields and lead to standing mechanical waves in the sample. This piezoelectric response does not exist in paraelastic SrTiO 3 nor at temperatures just below the ferroelastic phase transition. The interpretation of the resonances is related to ferroelastic twin walls which become polar at low temperatures in close analogy with the known behavior of CaTiO 3 . SrTiO 3 is different from CaTiO 3 , however, because the wall polarity is thermally induced; i.e., there exists a small temperature range well below the ferroelastic transition point at 105 K where polarity appears on cooling. As the walls are atomistically thin, this transition has the hallmarks of a two-dimensional phase transition restrained to the twin boundaries rather than a classic bulk phase transition.

A novel experimental technique, resonant piezoelectric spectroscopy (RPS), has been applied to investigate polar precursor effects in highly (65%) B-site ordered PbSc 0.5 Ta 0.5 O 3 (PST), which undergoes a ferroelectric phase transition near 300 K. The cubic-rhombohedral transition is weakly first order, with a coexistence interval of ∼4 K, and is accompanied by a significant elastic anomaly over a wide temperature interval. Precursor polarity in the cubic phase was detected as elastic vibrations generated by local piezoelectric excitations in the frequency range 250-710 kHz. The RPS resonance frequencies follow exactly the frequencies of elastic resonances generated by conventional resonant ultrasound spectroscopy (RUS) but RPS signals disappear on heating beyond an onset temperature, T onset , of 425 K. Differences between the RPS and RUS responses can be understood if the PST structure in the precursor regime between T onset and the transition point, T trans = 300 K, has locally polar symmetry even while it remains macroscopically cubic. It is proposed that this precursor behavior could involve the development of a tweed microstructure arising by coupling between strain and multiple order parameters, which can be understood from the perspective of Landau theory. As a function of temperature the transition is driven by the polar displacement P and the order parameter for cation ordering on the crystallographic B site Q od . Results in the literature show that, as a function of pressure, there is a separate instability driven by octahedral tilting for which the assigned order parameter is Q. The two mainly displacive order parameters, P and Q, are unfavorably coupled via a biquadratic term Q 2 P 2 , and complex tweedlike fluctuations in the precursor regime would be expected to combine aspects of all the order parameters. This would be different from the development of polar nanoregions, which are more usually evoked to explain relaxor ferroelectric behavior, such as occurs in PST with a lower degree of B-site order.

An experimental method, Resonant Piezoelectric Spectroscopy (RPS), is introduced for the detection of polar precursor effects in ferroelectric and multiferroic materials. RPS is based on the excitation of elastic waves through the piezoelectric effect in a sample. As the intensity of these waves is significantly amplified through mechanical resonances, RPS is very sensitive to the development of polar nanostructures. Using RPS, we identify polar nanostructures in BaTiO 3 as a precursor in the cubic phase. Results are compatible with polar tweed structures which persist up to 613 K. This temperature is much higher than previously reported. V

Ultrasonic velocity measurements on the magnetoelectric multiferroic compound CuFeO2 reveal that the antiferromagnetic transition observed at TN1 = 14 K might be induced by an R3m C2/m pseudoproper ferroelastic transition [1]. In that case, the group theory states that the order parameter associated with the structural transition must belong to a two dimensional irreducible representation Eg (x 2 − y 2 , xy). Since this type of transition can be driven by a Raman Eg mode, we performed Raman scattering measurements on CuFeO2 between 5 K and 290 K. Considering that the isostructural multiferroic compound CuCrO2 might show similar structural deformations at the antiferromagnetic transition TN1 = 24.3 K, Raman measurements have also been performed for comparison. At ambient temperature, the Raman modes in CuFeO2 are observed at ωE g = 352 cm −1 and ωA g = 692 cm −1 , while these modes are detected at ωE g = 457 cm −1 and ωA g = 709 cm −1 in CuCrO2. The analysis of the temperature dependence of modes shows that the frequency of all modes increases down to 5 K. This typical behavior can be attributed to anharmonic phonon-phonon interactions. These results clearly indicate that none of the Raman active modes observed in CuFeO2 and CuCrO2 drive the pseudoproper ferroelastic transition observed at the Néel temperature TN1. Finally, a broad band at about 550 cm −1 observed in the magnetoelectric phase of CuCrO2 below TN2 could be attributed to a magnon mode.

International audienceFerroic domain walls could play an important role in microelectronics, given their nanometric size and often distinct functional properties. Until now, devices and device concepts were mostly based on mobile domain walls in ferromagnetic and ferroelectric materials. A less explored path is to make use of polar domain walls in nonpolar ferroelastic materials. Indeed, while the polar character of ferroelastic domain walls has been demonstrated, polarization control has been elusive. Here, we report evidence for the electrostatic signature of the domain-wall polarization in nonpolar calcium titanate (CaTiO 3). Macroscopic mechanical resonances excited by an ac electric field are observed as a signature of a piezoelectric response caused by polar walls. On the microscopic scale, the polarization in domain walls modifies the local surface potential of the sample. Through imaging of surface potential variations, we show that the potential at the domain wall can be controlled by electron injection. This could enable devices based on nondestructive information readout of surface potential

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.Microstructural patterns of twin boundaries and tweed in ferroelastic materials display typical aspects of glasses. The patterns are complex, their dynamics follows Vogel-Fulcher statistics and their field cooling-non-field cooling hysteresis is similar to those described in this issue as 'strain glasses'. The difference is that domain glasses do not need extrinsic defects to form. In the paraelastic phase, an intrinsic tweed pattern dominates the high temperature precursor regime. Experimentally, massive elastic precursor softening is related to polar standing waves, which are attributed to the glassy relaxation of the tweed pattern. In the ferroelastic phase we find a complex twin pattern when the sample is strained with a constant strain rate. The dynamics of the pattern formation is a-thermal at low temperatures and follows Vogel-Fulcher statistics at moderately high temperatures. It is argued that domain boundary patterns can hence evolve glasslike states while the underlying matrix remains fully crystalline without any defect induced disorder.1 Introduction Non-ergodic, glass like structures are commonly observed whenever the degree of disorder in a structural phase transition is large. Non-ergodicity is the defining quantity in relaxor materials [1-3] but it is not known how the breaking of ergodicity occurs in the limiting case of weakly disordered systems. Decreasing the strength of the random field in a ferroic phase transition will make the transition increasingly more ergodic and one may be tempted to assume that the fully ordered system undergoes a 'classic' phase transition without any ergodicity breaking. It is not clear whether the transition between relaxor-type, non-ergodic behaviour and an ergodic, ferroic transition mechanism is stepwise or continuous. Ren and collaborators [4] have argued that an abrupt transition exists between the long range ordered state and the non-ergodic glass state. Nevertheless, even for weak disorder, the ferroic state contains significant fluctuations of the order parameter and relaxor type behaviour cannot be excluded. Lloveras et al. [5] have shown that spatially heterogeneous states that occur in ferroelastic transitions depend crucially on the elastic anisotropy with tweed type microstructures for anisotropic interactions and mottled structures with almost spherical nano domains for isotropic interactions (e.g. in NiTi). They argue that such microstructures lead to structural disorder that gives rise to a distribution of energy barriers that, when overcoming a well-defined threshold, screen the long-range interactions

No abstract

Highly cation-ordered, ferroelectric PbSc 0.5 Ta 0.5 O 3 (PST) crystals have been studied by acoustic emission over a wide temperature range. The high degree of order leads to a non-dispersive dielectric anomaly at T trans = 300 K of a weakly first order phase transition. Acoustic emission (AE) was found at three characteristic temperatures, 330 K, 409 K and ∼600 K, but none between these temperatures. These temperatures are close to those known from cation-disordered relaxor PST, containing polar nano regions (PNRs). The microstructure in our ferroelectric PST contains structural tweed rather than PNRs. The coincidence of the AE temperatures points towards a close structural relationship between PNRs and tweed. Furthermore, when electric fields are applied, we observe shifts of these temperatures which are similar to 'critical end point' behavior. The similarity of AE signals in relaxors and tweed ferroelectrics proves that AE detects signals in a wider parameter space than previously expected.Random fields and local phase transitions are commonly evoked when symmetry changes in heterogeneous systems. While some models predict a well-defined transition behavior, such as the spherical random-bondrandom-field (SRBRF), 1,2 where the system transforms at a well-defined temperature from a paraelectric phase to a spherical glass phase without long range order. Other approaches predict smeared cross-overs such as Vugmeister and Rabitz, 3 and the traditional composition fluctuations model, 4 and the super-paraelectric model by Cross. 5 In none of these cases would a classic criticality analysis in terms of Wilson exponents of single thermodynamic order parameters be applicable. Experimentally, smeared crossovers are indeed seen in relaxor ferroelectrics (RFEs). These are materials with wide frequency dispersions of the dielectric response in the transition region. In addition, the transition between a paraelectric and a ferroelectric phase is not only smeared but also split into a number of other singularities which are commonly ascribed to the existence and temperature evolution of polar nanosized regions (PNRs), 6,7 The structural properties of PNRs are still not fully understood with few studies focused on atomic-level correlation effects between PNRs. 8Smooth crossover near the Burns temperature T d at high temperature range is seen in RFEs by the temperature evolution of the refractive index, 9-11 dielectrics and Brillouin light scattering, 12-19 polarized Raman spectra, 20,21 thermal expansion, 22-24 and acoustic emission (AE). 25-29 Similar behavior is found near the intermediate temperature T * . 12-21,30-35 On further cooling, a smeared frequency dependent temperature maximum of the dielectric response occurs near T m and, slightly below T m the freezing occurs at the freezing temperature T f , similar to polar-glass phases. [36][37][38][39] In contrast to these observations, one finds that AE studies [12][13][14][15][16][17][18][19]22,23,[25][26][27][28][29][30] show extremely sharp singularities at T d and T...