Toward Tunable Conductance States in Layered Materials for Neuromorphic Computing using Metal Intercalation

As computing needs and accompanying energy costs continue to rise exponentially, neuromorphic (brain-inspired) computing has the potential to provide crucial improvements to computing speed and efficiency over traditional CMOS technology. Metal-intercalated, layered van der Waals materials present a promising strategy for tuning conductivity and insulator-metal transitions in devices for neuromorphic computing. Ion-insertion materials have been well studied for energy storage applications; however, the field focuses on high, reversible loading of ion materials. Here, we investigate the intercalation of metal heteroatoms into layered chalcogenide and oxide host materials, with the goal of understanding the associated modulation of the host electronic structure. We measure electrical transport of bulk materials and provide local insight into the structural and chemical effects of intercalation using scanning transmission electron microscopy (STEM) techniques. We further discuss how these effects differ based on host and intercalant species and intercalant concentration. These findings help provide insight into the dependencies of the host/guest interactions to understand the modulation of conductance and insulator-metal transitions for neuromorphic devices.