Efficient and accurate modeling of double-gate MOSFETs: a computationally feasible approach for potential, subthreshold, and drain current characteristics

Abstract The continuous advancement of semiconductor technology has been driven by the scaling of metal-oxide-semiconductor field-effect transistors (MOSFETs) to smaller dimensions, but this scaling is limited in the nanoscale regime due to short-channel effects (SCE). To address this limitation, the double-gate (DG) MOSFET has been proposed. For the effective utilization of the DG MOSFET in future semiconductor applications, accurate and robust models depicting its characteristics are essential. These models include potential distribution, subthreshold current, and drain current, where accuracy, complexity, and computational efficiency are crucial. This paper presents a modeling approach that reduces complexity in potential distribution, speeds up subthreshold current computation, and enhances drain current accuracy. The potential distribution is modeled using the superposition method, achieving a computational complexity of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mi class="MJX-tex-calligraphic" mathvariant="script">O</mml:mi> <mml:mo stretchy="false">(</mml:mo> <mml:mi>N</mml:mi> <mml:mo stretchy="false">)</mml:mo> </mml:math> , which is more efficient than the widely used Green’s function method. The subthreshold current is modeled numerically, achieving a faster computation time of 0.1202 s compared to existing methods. A curve-fitting approach is applied to model the drain current, improving accuracy in both the transfer and output characteristics across different temperatures. The model is further integrated with artificial neural networks, achieving high predictive performance. The proposed models are validated using an industry-standard device simulator and benchmarked against existing methods.