In this paper, we systematically evaluate dc/ac performances of sub-7-nm node fin field-effect transistors (FinFETs) and nanosheet FETs (NSFETs) using fully calibrated 3-D TCAD. The stress effects of all the devices were carefully considered in terms of carrier mobility and velocity averaged within the active regions. For detailed AC analysis, the parasitic capacitances were extracted and decomposed into several components using TCAD RF simulation platform. FinFETs improved the gate electrostatics by decreasing fin widths to 5 nm, but the fin heights were unable to improve RC delay due to the trade-off between on-state currents and gate capacitances. The NSFETs have better on-state currents than do the FinFETs because of larger effective widths (Weff) under the same device area. Particularly p-type NSFETs have larger compressive stress within the active regions affected by metal gate encircling all around the channels, thus improving carrier mobility and velocity much. On the other hand, the NSFETs have larger gate capacitances because larger Weff increase the gate-to-source/drain overlap and outerfringing capacitances. In spite of that, sub-7-nm node NSFETs attain better RC delay than sub-7-nm node as well as 10-nm node FinFETs for standard and high performance applications, showing better chance for scaling down to sub-7-nm node and beyond.

In this paper, multi-threshold voltage ( V _th ) scheme of 7-nm node nanosheet FETs (NSFETs) with narrow NS spacing were successfully achieved by metal-gate work function (WF) and channel doping ( N _ch ) using fully calibrated 3-D TCAD simulations. The limited NS spacing, which allows TiN capping layer only, makes different WF between the edge and the middle part of NS circumference. Unfortunately, this causes non-linear Vth shifts and dc performance degradation as a function of WF due to one-side turn-on phenomena between the edge and the middle part. Furthermore, the fixed WF of TiN capping layer limits Vth shifts toward ultra-low-power applications. To enable multi-Vth of NSFETs, several possible solutions are addressed: changing the Nch and the WF of TiN capping layer. The higher N _ch enables lower off-state current while 50-nm-wide three-stacked NS decreases dc performance variations effectively. Changing the WF of TiN capping layer can extend V _th margins, but degrade DC performance as a trade-off. Nonetheless, 7-nm node NSFETs adopting these techniques have multi- V _th options to satisfy wide ranges from ultra-low-power to high-performance applications.

Analog/RF performances of 5-nm node bulk fin-shaped field-effect transistors (FinFETs) and nanosheet FETs (NSFETs) were investigated and compared thoroughly using fully-calibrated TCAD. NSFETs have greater current drivability and gate-to-channel controllability than FinFETs under the same footprint, thus achieving larger intrinsic gain. But the cutoff frequencies ( $F_{t}$ ) of FinFETs and NSFETs are comparable due to larger gate capacitances ( $C_{gg}$ ) of NSFETs compensating DC performance improvements. Gate resistances ( $R_{g}$ ) of NSFETs are larger because of their metal gate (MG) configuration surrounding the channels, longer MG height by the top-most NS spacing region, and the bottom transistor, thus degrading maximum oscillation frequency ( $F_{max}$ ). Device design guidelines of FinFETs and NSFETs are also studied for better intrinsic gain, $F_{t}$ , and $F_{max}$ . Intrinsic gain is improved by better electrostatics, whereas $F_{t}$ increases by greater current drivability over $C_{gg}$ . For larger $F_{max}$ , careful device design is required to compensate between $R_{g}$ , $C_{gg}$ , output resistance, and $F_{t}$ . Overall, NSFETs outperform FinFETs in terms of intrinsic gain, $F_{t}$ , and $F_{max}$ , thus NSFETs are promising for analog/RF applications.

Process (systematic) variations of sub-5-nm node fin field-effect transistors (FinFETs) and nanosheet field-effect transistors (NSFETs) were investigated thoroughly using fully calibrated TCAD. All the process parameters consisting of front-end- as well as middle-of-line structure were independently randomized within feasible process conditions. A novel process scheme called source/drain patterning (SDP), having a superior performance by decreasing outer fringing capacitance through the downsized source/drain (S/D) epi, has also been analyzed for the variability study. SDP FinFETs have smaller variations of threshold voltages (Vth), off-state currents (Ioff), and effective currents (Ieff) than conventional ones because the subfin leakage is effectively controlled by bottom oxide (BO) beneath the source/drain instead of high punchthrough-stopper doping. The Spearman's correlation results between process parameters, Ioff, and Ieff, showed that the process parameters affecting the short-channel effects vary Ioff and Ieff greatly. Especially, the most critical one was the fin width (Wfin) for FinFETs. SDP NSFETs have the smallest variations of Vth, Ioff, and Ieff. The BO blocks the bottom leakage completely, and the variations of nanosheet (NS) thickness (TNS) are much smaller than those of Wfin due to the different process flows: epitaxial growth for TNS versus patterning for Wfin. Therefore, NSFETs are promising to reduce the variations of Vth, Ioff, and Ieff in the sub-5-nm node.

Excess source and drain (S/D) recess depth (TSD) variations were analyzed comprehensively as the most critical factor to DC/AC performances of sub 5-nm node Si-Nanosheet (NS) FETs for system-on-chip (SoC) applications. The variations of off-, on-state currents (Ioff, Ion) in three-stacked NS channels and parasitic bottom transistor (trpbt), gate capacitance (Cgg), intrinsic switching delay time (td), and static power dissipation (Pstatic) are investigated quantitatively according to the TSD variations. The more S/D dopants diffuse into the trpbt with the deeper TSD, so the Ioff and Ion increase by raised current flowing through the trpbt. Especially, the Ioff of PFETs remarkably increases above the certain TSD (TSD,critical) compared to NFETs. Furthermore, the Ion contribution of each channels having the TSD,critical is the largest at the top NS channel and the trpbt has the ignorable Ion contribution. Among the NS channels, top (bottom) NS channel has the largest (smallest) Ion contribution due to its larger (smaller) carrier density and velocity for both types of NSFETs. The Cgg also increases with the deeper TSD by increasing parasitic capacitance, but fortunately, the td decreases simultaneously due to the larger increase rate of the Ion than that of the Cgg for all SoC applications. However, the Pstatic enormously increases with deeper TSD, and low power applications are more sensitive to the TSD variations than other applications. Comprehensive analysis of the inevitable trpbt effects on DC/AC performance is the most critical indicator whether Si-NSFETs could be adopted to the sub 5-nm node CMOS technology.

Structural advancements of 5-nm node bulk fin-shaped field-effect transistors (FinFETs) without punch-through-stopper (PTS) were introduced using fully calibrated TCAD for the first time. It is challenging to scale down conventional bulk FinFETs into 5-nm technology node due to the sub-fin leakage increase. Meanwhile, bottom oxide deposition after anisotropic etching for source/drain (S/D) epi formation prevents the sub-fin leakage effectively even without the PTS doping, thus achieving better gate-to-channel controllability. Bottom oxide FinFETs also have smaller gate capacitances than do conventional FinFETs because the parasitic capacitances decrease by smaller S/D epi separated from the bottom Si layer, which reduces junction and outer-fringing capacitances. But smaller S/D epi decreases the stresses along the channel direction, and the effective widths decrease by the bottom oxide layer blocking the current paths at the bottom side of fin channels. Furthermore, increase of the interconnect resistance and capacitance parasitics down to 5-nm node diminishes the improvements of total delays as the interconnect wire length increases greatly. In spite of these drawbacks, 5-nm node bottom oxide FinFETs achieve smaller total delays than do the 7-nm node conventional FinFETs, especially for low-power applications, thus promising for the scalability of bulk FinFETs along with simple and reliable process by avoiding PTS step.