Two-Dimensional Flash Memory: Materials, Interfaces, and Storage-Computing Convergence

Flash memory remains the dominant nonvolatile storage technology, yet its continued scaling is increasingly constrained by the coupled requirements of fast programming, low operating energy, and long-term reliability. Recent advances in 2D materials and van der Waals heterostructures provide an opportunity not merely to miniaturize flash cells, but to reconfigure the charge-programming physics itself. In this Review, we introduce interface-controlled charge programming as a unifying framework for understanding and designing 2D flash memories. We discuss how atomically thin channels, clean van der Waals interfaces, engineered tunneling barriers, controllable trap states, and contact modulation regulate charge injection, transport, trapping, and confinement. This framework connects device-level metrics, including nanosecond-to-subnanosecond programming, low-energy operation, long retention, and high endurance, with emerging system-level functions such as multilevel storage, in-memory computing, sensing-memory-computing integration, and CMOS-compatible heterogeneous integration. We further identify critical challenges in wafer-scale synthesis, device variability, reliability benchmarking, and array-level implementation. By shifting flash memory design from bulk-field-driven tunneling to interface-programmed charge control, 2D materials may enable a new class of nonvolatile devices for storage-computing convergence beyond conventional silicon flash.