Impact of humidity and temperature on nanowire networks dynamics

The use of neural networks in the hardware processing for Artificial Intelligence (AI) has gained significant traction, particularly for solving complex, non-linear, and chaotic problems. A primary objective of this research is to explore the optimal environment for neuromorphic chips, enabling advancements such as increased neural connections, self-healing capabilities, enhanced memory capacity, and energy efficiency. This study focuses on the impact of two critical environmental factors—humidity and temperature cycles—on the dynamic behaviour of Ag-PVP nanowire networks. Through conductance measurements, this research seeks to demonstrate both potentiation and relaxation correlations with external conditions in these networks. The experiments were conducted in a controlled laboratory environment at the Polytechnic University of Turin, utilizing the Keithley 4200-SCS Semiconductor Characterization System for data collection and the Precision Humidity Control System (HCS-2M) for environmental regulation. The findings support the theoretical background and align with previous research, offering generalized insights into the behaviour of these “alive” materials and their potential applications in neuromorphic systems.