Dec 3, 2024
3:30pm - 3:45pm
Sheraton, Second Floor, Republic A
Nitin Kumar1,Karsten Beckmann2,3,Nathaniel Cady3,Sambandamurthy Ganapathy1
University at Buffalo, The State University of New York1,NYcreats2,University at Albany, State University of New York3
Nitin Kumar1,Karsten Beckmann2,3,Nathaniel Cady3,Sambandamurthy Ganapathy1
University at Buffalo, The State University of New York1,NYcreats2,University at Albany, State University of New York3
NbO<sub>2</sub> exhibits an electric field driven insulator-to-metal transition at room temperature, making it a promising candidate for scalable elements in next-generation computing applications such as artificial neurons, oscillators, selector devices, and memory elements. These applications leverage the resistive switching properties of NbO<sub>2</sub> that can be altered based on device size. The effect of scaling on resistive switching parameters for device sizes ranging from 30 nm × 30 nm × 25 nm to 170 nm × 170 nm × 25 nm is investigated. Larger devices are found to exhibit only one or potentially no regions of negative differential resistance (NDR). As the device size is reduced, the on-off ratio increases, and the smaller devices manifest dual NDR regions. Ultra-low frequency noise measurements are carried out to understand the role of electrical domains in various applications of NbO<sub>2</sub> and noise levels were orders of magnitude higher in smaller devices. Oscillators (with tunable frequencies up to 5 MHz) based on resistive switching are built, and the dependence of resistive switching parameters on oscillator frequency across different sizes, applied voltages, and external parameters such as resistance and capacitance are studied. The transport measurements are supported by NSF award 1726303.