This work highlights the work-function-fluctuation (WKF) of the strained Si 0.6 Ge 0.4 p-n-p-n tunneling field-effect transistor (TFET) using nanosheet geometry at sub-3 nm technology nodes. The reported work exploits 3000 samples to signify the key results of several dc factors such as ON- and OFF-state currents ( ${I}_{ \mathrm{\scriptscriptstyle ON}}$ and ${I}_{ \mathrm{\scriptscriptstyle OFF}}$ ), threshold voltage ( ${V}_{\text {th}}$ ), and the subthreshold swing (SS). Fewer WKFs are identified for higher metal grain number (MGN) due to the shaped low grain size. Collective average energy reduction of 10–15 meV as low with high MGN and 40–60 meV as high with low MGN for a group of metal-grains closer to the tunneling junction is observed. This huge impact of energy reduction proportionally affects electron transport due to reduction in tunneling length. High variability in ${V}_{\text {th}}$ ( $\geq {15}$ %) is identified at progressively diminished MGN due to parabolic behavior of variation. Furthermore, marginal variation in SS, ${I}_{ \mathrm{\scriptscriptstyle ON}}$ , and ${I}_{ \mathrm{\scriptscriptstyle OFF}}$ are observed because of the WKF dependability over the subthreshold region of operation.
With the rapid growth of the semiconductor manufacturing industry, it has been evident that device simulation has been considered a sluggish process. Therefore, due to downscaling of semiconductor devices, it is significantly expensive to obtain the inevitable device simulation data because it requires complex analysis of various factors. To develop a competent technique to analyze the performance of the line tunnel field-effect transistors (TFETs), the 3-D stochastic device simulation is integrated with a machine learning (ML) algorithm, named random forest regressor (RFR). Despite producing tremendous researches by the RFR model in the field of computer vision, the adoption of these ML algorithms in the field of the semiconductor industry has a lot of margin for progress. The ML-based RFR model is exploited to predict the effect of variability sources of line TFET under different biasing conditions. Results are promising and reducing the computational cost of device simulation by 99%. The prediction of effect of source variation is less than 1% as compared to the device simulation of line TFET. The application of the RFR on the line TFET device exhibits the power and flexibility of this approach because its evaluation with different bias conditions shows outstanding results.
1 January 2020 in IEEE Open Journal of Nanotechnology
This work comprises of design and simulation of multi-channel line tunnel field-effect transistors (mCLTFETs) by scaling inter-gate separation (IGS) and overlapped source (L OV ). The scope of the work is to explore the performance boost and optimization of the studied devices by considering geometrical structures, low-bandgap materials, IGS and L OV of the mCLTFETs. The structure is designed without diminishing the subthreshold swing (SS) and the leakage currents through a spacer technology and strained Si 0.6 Ge 0.4 . The optimal values of IGS and L OV for the multi-channel concept are estimated subject to several physical constraints of the proposed device. An IGS ≈ 10 nm and a L OV ≈ L G /2 are reported as suitable choice for sub-8-nm technological nodes, where SS = 18 mV/dec and I on /I off = 10 9 are achieved.
Tunnel field-effect transistors (TFETs) are the decent performance estimators in the prospective of short-channel effects. In such structures, a small inter-gate separation (IGS) is a key factor that appraises for high packed-density with more number of channels (N) to deliver superior performance. Hence, the investigation is majorly focused on scaling IGS and its fringing-field impact on device behavior for the first time. The outcomes reveal that the high fringing-field initiates for IGS <; 10 nm and influences the tunneling probability and scattering strongly at 1-nm IGS, which affect the DC and RF characteristics; hence, optimized values of IGS are investigated and determined as IGS > 10 nm. The results state that the optimized IGS can provide source to deliver high ratio of on- and off-current (I on /I off ). Even though, a small IGS is beneficial for reduction in the total capacitance, the RF performance improvement depends on a large IGS. The investigation is further extended and quantified for the finest IGS in multi-channel TFETs when N varies from 1 to 10. These analyses are assessed for the emerging technological nodes.
31 January 2020 in Japanese Journal of Applied Physics
Using a dual spacer consisting of 50% SiO 2 and 50% HfO 2 , the ratio of the on-state current/the off-state current in the order of 10 6 is achieved for the explored devices. Based on the experimentally validated simulation, the result indicates that the variation of gate-capacitance is significant owing to the sizeable parasitic capacitance resulting from the spacer. The role of the spacer acts as the parallel plate capacitor, the amount of gate capacitance will be increased largely with the HfO 2 due to its high parasitic capacitance. For the devices without any spacers, induced by a donor-type single charge trap (SCT), the statistically calculated highest amplitude of random telegraph noise (RTN) is 2.32%. It occurs when the SCT locates in the middle of the channel due to the high occurrence of capture and emission for SCT under low gate voltage. Notably, for devices with considerable spacers, the RTN can be significantly suppressed (≤1%).
This work illustrates the most effective way of utilizing the ferroelectricity for tunneling field-effect transistors (TFETs). The ferroelectric (Hf 0.5 Zr 0.5 O 2 ) in shunt with gate-dielectric is utilized as an optimized metal–ferroelectric–semiconductor (OMFS) option to improve the internal voltage ( Vint ) for ample utilization of polarization and electric fields of Hf 0.5 Zr 0.5 O 2 across the tunneling region. The modeling of Vint signifies 0.15–1.2 nm reduction in tunneling length ( λ ) than the nominal metal–ferroelectric–insulator–semiconductor (MFIS) option. Furthermore, the TFET geometry with the scaled-epitaxy region as vertical TFET (VTFET), strained Si 0.6 Ge 0.4 as source, and gate-all-around nanowire options are used as an added advantage for further enhancement of TFET’s performance. As a result, the proposed design (OMFS-VTFET) achieves superior DC and RF performances than the MFIS option of TFET. The figure of merits in terms of DC characteristics in the proposed and optimized structure are of improved on-current (=0.23 mA μ m −1 ), high on-to-off current ratio (=10 11 ), steep subthreshold swing (=33.36 mV dec −1 ), and superior unity gain cut-off frequency (≥300 GHz). The design is revealed as energy-efficient with significant reduction of energy-efficiency in both logic and memory applications.