Manipulating Ferroelectric α-In₂Se₃/GaN Dipole Interactions by Polarization Engineering

By utilizing the dipole interactions, polar heterostructures are developing toward two-dimensional/three-dimensional (2D/3D) hybrid dimensions for unique phenomena such as polarization-induced electric field, charge separation, band-structure modification, etc. In this study, we have constructed 2D/3D α-In2Se3/GaN polar heterostructures by stacking 2D α-In2Se3 and 3D wurtzite GaN semiconductors with distinct polarity configurations. Upon GaN contacting with α-In2Se3, the structural stability demonstrates a direct dependence on the polarity assembly, as evidenced by the change of binding energy from −1.39/–1.83 eV in Ga-polarity to −2.99/–1.95 eV in N-polarity. In particular, the polarity inversion of GaN not only dictates the direction of charge transfer but also increases the electrostatic potential differences from 2.92/1.84 eV in N-polarity to 3.65/5.10 eV in Ga-polarity. Moreover, the interfacial polarization and electronic band profiles of α-In2Se3/GaN have been further adjusted by applying the external electric field. It is found that the interfacial dipole interaction plays a crucial role in the external electric field modulation, with a direct correlation to the charge distribution at α-In2Se3/GaN polar heterointerfaces. By sweeping the external electric field from −0.4 to 0.4 V/Å, the α-In2Se3(↑)/GaN(↑) and α-In2Se3(↓)/GaN(↑) polar heterostructures maintain the type-III electronic band profiles, attributed to the electrostatic shielding effect of high-density charge carriers. However, the band alignments of α-In2Se3(↑)/GaN(↓) and α-In2Se3(↓)/GaN(↓) polar heterostructures have been extensively transformed from type-III, to type-II, and then to type-I. Our findings demonstrate the effective capability of polarity inversion and external electric field on manipulating dipole interactions of 2D/3D polar heterostructures, paving the way for the innovative design and optimization of next-generation optoelectronics, information memory, and neural computing applications.