Chemically produced two-dimensional conducting interfaces for memory and neuromorphic applications

Two-dimensional electron gases (2DEGs) at epitaxial oxide interfaces, such as LaAlO3/SrTiO3 heterostructures, have long served as model platforms for exploring emergent interfacial phenomena, including superconductivity, magnetism, and tunable metal–insulator transitions. While such systems were initially investigated primarily within the context of condensed matter physics, their relevance is being redefined by recent advances in non-epitaxial oxide 2DEGs, which offer new opportunities for practical applications. In contrast to polarity-driven epitaxial interfaces, non-epitaxial oxide 2DEGs are often governed by chemically driven carrier generation mechanisms, most notably oxygen vacancy (VO) formation induced by interfacial redox reactions. These mechanisms enable systematic tuning of interfacial electronic properties using amorphous or polycrystalline oxides and industry-compatible processes. This review provides a unified perspective linking the fundamental formation mechanisms of oxide 2DEG to their emerging applications. We first survey the principal physical origins of oxide 2DEGs, followed by an in-depth discussion of VO generation pathways in non-epitaxial oxide systems, including ionic bombardment, metal-induced redox reactions, and surface chemical effects. Building on this mechanistic foundation, we connect interfacial chemistry to application-level performance metrics, with particular emphasis on memory and neuromorphic technologies motivated by rapidly expanding computational demands in the AI era. In memory architectures, oxide 2DEGs are evaluated as enablers of both selector devices and storage elements, while in neuromorphic hardware, sensor platforms exploiting 2DEG transport are highlighted for their capability to emulate biological sensory functions and dynamics. Key challenges for practical deployment of oxide 2DEGs, including compatibility with three-dimensional integration, thermal stability during processing, and limitations in material selection, are critically examined. Potential strategies for overcoming these constraints, such as interfacial chemistry control, defect engineering, and process–structure–property correlations, are discussed. By bridging the fundamental physics and chemistry of oxide 2DEGs with emerging device-level applications, this review underscores their potential as a cornerstone material platform for next-generation, energy-efficient computing architectures.