DESIGN AND MODELING OF TUNNEL FIELD-EFFECT TRANSISTOR AND THEIR CIRCUIT APPLICATIONS

Abstract

The thermionic injection of electrons over the barrier in the metal-oxide-semiconductor field-effect transistors (MOSFETs) imposes a fundamental limitation on the steep transition in the subthreshold region. Therefore, a new device is required to provide steep slope and high drive current to improve the energy efficiency of the circuits. Different novel devices like tunnel field-effect transistors (TFETs), negative capacitance field-effect transistors (NCFETs), impact ionization FETs, and nano-electromechanical FETs have demonstrated potential for steep-slope devices. Among these, significant progress on the TFET technology has been achieved in the last decade due to their CMOS compatible fabrication process steps and promise to provide MOSFET like characteristics. Recently, the tunnel FETs for internet-of-things applications have been successfully demonstrated experimentally. TFETs utilize band-to-band tunneling mechanism for injection of charge carriers from source to the channel region. Over the last decade, the journey of Tunneling FETs (TFETs) for improving the drive capability and the subthreshold slope (SS) has led to different new device architectures. The device structure, source/channel/drain materials and doping have been optimized to improve the performance. Point (lateral) tunneling and line (vertical) tunneling are the two main tunneling mechanisms. In point tunneling, the tunneling takes place in parallel to the gate. On the other hand, in line tunneling, the tunneling direction is perpendicular to the gate. In the latter case, the area of the cross-section for tunneling increases that improves the drive current and SS. Application of novel materials like III-V/2D materials improves the performance of the device due to the direct tunneling process. Despite of the advances in device architectures and material systems, the performance of the TFETs are still not acceptable for the application at future technology node. Therefore, line tunneling mechanism and novel materials are extensively being explored to achieve an acceptable performance. The main concern in 2D material based TFETs is the doping of the source and drain region. The conventional methods of doping are not feasible in 2D materials. For example, the ion implantation method of doping is not possible for 2D materials since the placement of dopant atoms in a single thin layer is a very difficult task. The chemical doping by using the diffusion process is also not a good option because the diffusion of dopant atoms can lead to an undesired chemical reaction in 2D materials. In addition, it creates the dopant states in the bandgap of the semiconducting materials that degrade the IOFF and SS of the device. A thin body of 2D semiconductors creates an ideal situation for electrostatic of the device. Different tunnel FET structures using electrostatic doping have been reported. These devices use multiple supply voltages for source and drain doping that is not an energy-efficient technique. The charge plasma doping technique has the potential for electrostatic doping of 2D materials without using extra supply voltages. In charge plasma doping, the source and drain regions are induced in the intrinsic semiconductor body by employing work function engineering. However, to achieve high and abrupt doping in the source region of the TFET structure, the charge plasma TFET structure needs to be explored by utilizing the different combinations of source and drain metal gates. The impact of strain on the bandgap of the transition metal dichalcogenides (TMDs) has been investigated in using density functional theory (DFT) based calculations. However, the impact of strain on the effective mass of TMD material and on the electrical parameters of TMD material based TFET has not been studied so far. Therefore, there is a strong need to investigate the impact strain on the performance of TMD material based TFETs and their potential applications in flexible circuits and strain sensors.