Novel Hexagonal Ferroelectric LuGaO₃

The presence of simultaneous ferroelectric and magnetic order in hexagonal LuFeO₃ has made it a promising candidate for next-generation microelectronics. At the same time, the metastable nature of this phase makes it an interesting case study for the synthesis of novel epitaxial materials. Its utility, however, is limited by its low polarization (&lt;10 μC/cm²) and often poor film quality. Recent first principles predictions have identified a new hexagonal ferroelectric, LuGaO₃, which is isostructural to LuFeO₃ but is predicted to be thermodynamically stable and exhibit a higher polarization than LuFeO₃. While nonmagnetic, the enhanced polarization in LuGaO₃ could make it a more suitable ferroelectric, and alloying it with LuFeO₃ may serve to enhance the magnetic character while retaining more robust ferroelectricity. Overall, understanding the synthesis of newly predicted materials provides an important opportunity to verify and correct computational models, making them more reliable for future materials predictions.<br/><br/>Here, we demonstrate the synthesis of epitaxial, hexagonal LuGaO₃ via pulsed-laser deposition, which has not yet been reported. Extensive X-ray diffraction-based structural studies verify the hexagonal nature of this phase, while piezoresponse force microscopy studies show indications of stable, ferroelectric switching. While first principles calculations predict this phase to be thermodynamically stable (energy above hull = 0), the bulk material separates into Lu₂O₃ and Lu₃Ga₅O₁₂ phases. Film synthesis is even more complex and highly sensitive to growth conditions and epitaxial constraints. While (111)-oriented yttria-stabilized zirconia is the preferred substrate for hexagonal ferroelectrics like LuGaO₃, it is also closely lattice matched with Lu₂O₃, which forms epitaxially with surprisingly high quality under a wide range of conditions, with excess gallium being segregated to large precipitates on the surface of the film. Energy dispersive X-ray spectroscopy shows a large gallium deficiency (as much as 30% off from the target stoichiometry) in these phase-separated films, likely reducing the stability of the perovskite phase. Depositing the films at room temperature and subsequently annealing them circumvents the gallium loss enough to stabilize the hexagonal phase in extremely thin (~5 nm) films, while depositing in abnormally high oxygen partial pressures (1 Torr or more) has the same effect for thicker films, enabling more extensive structural and dielectric characterization which confirms the presence of a hexagonal, ferroelectric phase. We further examine the structure and dielectric properties as a function of synthesis conditions as the material transforms from predominately nonpolar Lu₂O₃ to ferroelectric hexagonal LuGaO₃. We then compare this to LuFeO₃, which does not exhibit the same phase separation and largely maintains the hexagonal phase regardless of the synthesis conditions. Finally, we examine the influence of alloying and multilayering of both materials to understand their stability and produce a more robust multiferroic. Interestingly, providing a thin (as little as 1-2 nm) buffer layer of LuFeO₃ is enough to stabilize hexagonal LuGaO₃ where it would otherwise phase separate on a bare substrate, further demonstrating the importance of epitaxy in phase selection. Extending this to a multilayer or solid solution has a similar effect. While such mixtures maintain robust ferroelectricity, a small degree of phase separation still persists and will likely require further tuning of the gallium chemistry to fully correct. Nevertheless, this work represents an important step toward finding a more robust hexagonal perovskite multiferroic and offers valuable insight on metastable materials synthesis.