There is a general perception that a large piezoelectric response in ferroelectric solid solutions requires a morphotropic/polymorphic phase boundary (MPB/PPB), i.e., a composition driven interferroelectric instability. This correlation has received theoretical support from models which emphasize field driven polarization rotation and/or interferroelectric transformations. Here, we show that the ferroelectric system $\left(1-x\right)\mathrm{PbTi}{\mathrm{O}}_3-\left(x\right)\mathrm{Bi}\left({\mathrm{Zr}}_{1/2}{\mathrm{Ni}}_{1/2}\right){\mathrm{O}}_3$ (PT-BNZ), which shows $d_{33}$ ($\sim 400\mathrm{p}\mathrm{C}/\mathrm{N}$) comparable to the conventional MPB/PPB systems, does not belong to this category. In the unpoled state the compositions of PT-BNZ showing large $d_{33}$ exhibit a coexistence of tetragonal and cubiclike (CL) phases on the global length scale. A careful examination of the domain strucures and global structures (both in the unpoled and poled states) revealed that the CL phase has no symptom of average rhombohedral distortion even on the local scale. The CL phase is rather a manifestation of tetragonal regions of short coherence length. Poling increases the coherence length irreversibly which manifests as poling induced $\mathrm{CL}\to P4mm$ transformation on the global scale. PT-BNZ is therefore qualitatively different from the conventional MPB piezoelectrics. In the absence of the composition and temperature driven interferroelectric instability in this system, polarization rotation and interferroelectric transformation are no longer plausible mechanisms to explain the large electromechanical response. The large piezoelectricity is rather associated with the increased structural-polar heterogeneity due to domain miniaturization without the system undergoing a symmetry change. Our study proves that attainment of large piezoelectricity does not necessarily require interferroelectric instability (and hence morphotropic/polymorphic phase boundary) as a criterion.

• Received 8 December 2017
• Revised 9 May 2018

DOI:https://doi.org/10.1103/PhysRevB.97.224109