Defects Induced Charge Trapping/Detrapping and Hysteresis Phenomenon in MoS2 Field-Effect Transistors: Mechanism Revealed by Anharmonic Marcus Charge Transfer Theory

One critical problem inhibiting the application of MoS2 field-effect transistors (FETs) is the hysteresis in their transfer characteristics, which is typically associated with charge trapping (CT) and charge detrapping (CDT) induced by atomic defects at the MoS2-dielectric interface. Here, we propose a novel atomistic framework to simulate electronic processes across the MoS2-SiO2 interface, demonstrating the distinct CT/CDT behavior of different types of atomic defects and further revealing the defect type(s) that most likely cause hysteresis. An anharmonic approximation of the classical Marcus theory is developed and combined with state-of-the-art density functional theory to calculate the gate bias dependent CT/CDT rates. All the key electr onic quantities are calculated with Heyd-Scuseria-Ernzerhof hybrid functionals. The results show that single Si-dangling bond defects are active electron trapping centers. Single O-dangling bond defects are active hole trapping centers, which are more likely to be responsible for the hysteresis phenomenon due to their significant CT rate and apparent threshold voltage shift. In contrast, double Si-dangling bond defects are not active trap centers. These findings provide fundamental physical insights for understanding the hysteresis behavior of MoS2 FETs and provide vital support for understanding and solving the reliability of nanoscale devices.

Automated summary

The study proposed a novel atomistic framework to simulate electronic processes across the MoS2-SiO2 interface, revealing the distinct CT/CDT behavior of different types of atomic defects and identifying the defect type(s) most likely responsible for hysteresis. Results showed that single Si-dangling bond defects are active electron trapping centers, while single O-dangling bond defects are active hole trapping centers, more likely to cause hysteresis.