STTR Phase I: Ultrascalable Non-volatile Memory with Multifunctionality by MOCVD Direct Growth Two-Dimensional Materials

The broader impact/commercial potential of this Small Business Technology Transfer (STTR) project is to advance the development of high-performance, energy-efficient artificial intelligence (AI) hardware. The proposed innovation enables the scalable integration of adaptive, non-volatile memory elements based on atomically thin molybdenum disulfide (MoS2) into complex silicon trench structures. This approach promises to overcome density and bandwidth limitations in current memory technologies, providing a platform for specialized AI accelerators with tightly coupled computation and storage. Successful commercialization could significantly enhance the capabilities of machine learning systems across various domains, offering societal benefits in fields such as healthcare, transportation, and scientific research. The project fosters collaboration among academic institutions, government agencies, and industry partners, strengthening the U.S. position in the strategically important AI hardware sector. This STTR Phase I project proposes to develop a novel manufacturing process for integrating two-dimensional (2D) MoS2 material into trenches composed of CMOS-compatible materials to make high-density memristors in a high-aspect-ratio microstructure. The goal is to demonstrate the feasibility of directly growing conformal MoS2 monolayers on complex 3D topographies using a low-temperature metalorganic chemical vapor deposition (MOCVD) technique. The research objectives include optimizing the growth parameters to achieve reliable resistive switching performance and assessing the scalability of the integration scheme. The anticipated technical results comprise a proof-of-concept demonstration of multifunctional MoS2 memristor arrays with improved storage density as well as fabrication process uniformity compared to planar designs. This project aims to establish the groundwork for further development of this technology toward commercially viable AI hardware solutions for less energy-hungry, multifunctional, and highly efficient computing. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.