Investigation of Metal Ions Drift in Memristive Devices using Perturbed Angular Correlation (PAC) method

The distinctive electrical characteristics of memristive devices render them an optimal choice for enhancing the computational efficiency of computer systems and facilitating their integration into neuromorphic applications. Particularly noteworthy is their potential application in the field of Artificial Intelligence. This study aims to elucidate the atomic-level structure of memristive devices employing Perturbed Angular Correlation (PAC) spectroscopy. Specifically, our investigation seeks to scrutinize the impact of external voltage on the local electrical field gradients within memristive devices and gain insights into ion channel formation mechanisms. Our findings underscore the potential of PAC as a promising technique for probing the local structure of memristive devices. However, the examined samples were found unsuitable for PAC measurements, primarily due to the diminutive size of the active zone, which hindered the attainment of statistically significant results. Consequently, we propose an alternative device configuration for future investigations. Furthermore, our study underscores the critical importance of controlled aging and storage protocols for memristive units, attributable to the formation of silver sulfide on the contact pads of the samples. These insights are pivotal for the development and longevity of memristive device technologies.