As nanoscale electronic devices are being packed into dense three-dimensional arrays, the effects of the thermal environment of the system during device operation become critical, but are not clearly understood. Predicting the temperature evolution using a robust model will provide critical design guidelines for complex memory and computing systems. Here, we used in-operando thermal and x-ray mapping with sub-micrometer spatial and sub-microsecond temporal resolutions on functioning tantalum oxide memristive switches and observed hot spots corresponding to oxygen concentration gradients, indicating the presence of localized conductive filaments. We constructed a hybrid electro-thermal model comprising 3D heat transfer and 0D resistive switching models to predict electrical characteristics and the temperature rise and calibrated it against the measurements. We also demonstrated thermal crosstalk in an array of memristors to illustrate localized heating. Such a model will guide system design by considering thermal performance, which is critical to most future electronic chips.
CITATION STYLE
Shen, W., Kumar, S., & Kumar, S. (2021). Experimentally calibrated electro-thermal modeling of temperature dynamics in memristors. Applied Physics Letters, 118(10). https://doi.org/10.1063/5.0039797
Mendeley helps you to discover research relevant for your work.