Publication
With the rapid advancement of quantum computing, there is a growing demand for reliable quantum devices operating at cryogenic temperatures to achieve fault-tolerant systems. Cryo-CMOS has recently gained much attention as the key solution to manipulating large-scale qubits, but challenges remain in designing and optimizing Cryo-CMOS systems. Therefore, this thesis focuses on the cryogenic characterization and modeling of MOSFETs for quantum computing applications.
Some physical phenomena that occur at cryogenic temperatures are not well understood, and needless to say, they are not formulated in compact models for circuit simulations. The shortfall of documenting particular cryogenic behaviors leads to inaccurate predictions of device performance at cryogenic temperatures using industrial standard PDKs developed for use near room temperature. Hence, the primary objective of this research is to develop analytical models that can accurately capture the unique behaviors of MOSFETs under cryogenic conditions. To achieve this goal, the thesis proposes new perspectives based on current studies in device physics at cryogenic temperatures. Both theoretical analysis and experimental validations were employed to achieve this goal.
The physics-based models of subthreshold behavior, threshold voltage, radio-frequency performance, and dynamic self-heating are proposed, which improve the existing models. Moreover, these models provide valuable insights into the device behavior, paving the way for designing and optimizing MOSFET technologies at cryogenic temperatures. Nevertheless, reliable cryogenic compact models still have a long way to go. It requires extensive collaboration with foundries. Therefore, in addition to physics-based models, this thesis validates the semi-empirical model down to 4 K, which can be used in the early design phase for cryogenic applications.
In conclusion, this thesis contributes to advancing the cryogenic modeling of MOSFETs, offering a promising avenue for developing cryogenic electronics. The findings of this study have significant implications for optimizing cryogenic devices and circuits.