This is the authors' abstract. We don't add key points for this paper.
Understanding the resistive switching (RS) mechanisms in different types of microelectronic devices is a critical step in the development of next-generation energy-efficient memory and computing units. Over the past decade, numerous studies on various metal-insulator-metal (MIM) structures have revealed the filament-type RS mechanism, namely the formation and rupture of conductive channels that lead to the low resistance state (LRS) and high resistance state (HRS) of the device. Yet, the filament-type RS devices usually suffer from undesirable temporal and spatial variations due to the random formation/rupture of the conductive filaments. In contrast, another interface-type RS devices whose resistive state is dominated by the semiconductor (or insulator) and metal interfaces, exhibit lower threshold switching voltage and more reliable switching repeatability, thus showing great promise in data storage memory and neuromorphic computing [1, 2]. Several different models have been proposed to explain the interface-type switching mechanisms, such as charge carriers (i.e. hole or electron) trapping and/or detrapping, electrochemical migration of oxygen vacancies, and even Mott transition induced by carriers doped at the interface [3-5]. However, due to the limited materials examples and challenges for implementing in situ characterization experiments, the RS mechanisms of interface-type RS devices are much less studied, which undoubtedly hinders their potential applications. In this study, we utilized in situ scanning transmission electron microscopy (STEM) and electron energy loss spectroscopy (EELS) techniques to investigate the RS mechanism of a nitride (TiN) thin film based memristor device. The TEM lamella device was prepared by the focused ion beam (FIB) technique and then transferred onto the MEMS-based E-chip [6, 7]. In situ S/TEM and EELS biasing experiments were conducted to characterize the microstructural evolution and chemical oxidation states variation inside the device during the RS process. Our in situ EELS Ti L2/L3 ratio maps clarify the Ti valence states variation inside the device as a result of oxygen vacancies migration under negative and positive electric fields. The RS mechanism is further discussed based on the interplay of ionic charges and electronic conduction, namely the electric field-driven n-p/p-n junction formation and Schottky barriers modulations at the upper and lower metal/semiconductor interfaces, which lead to the bipolar diode-like I-V characteristic of the device [8, 9]. (a) Schematic of the MEMS-based E-chip and in situ biasing experiment. (b) SEM images showing the TEM lamella mounted on the E-chip. (c) Schematic of the positive and negative bias directions applied to the vertical Pt/TiOx/TiN/Si device. (d) I-V and (e) log-scale I-V characteristic curves of the device obtained during experiment. (a) Cross-sectional ADF image, (b) Ti L2/L3 ratio maps derived from the in situ Ti-L2,3 EELS spectra imaging (SI) at unbiased, negative and positive biased conditions.
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: