Deep Learning Algorithms for the Work Function Fluctuation of Random Nanosized Metal Grains on Gate-All-Around Silicon Nanowire MOSFETs

Akbar, Chandni; Li, Yiming; Sung, Wen-Li

doi:10.1109/access.2021.3079981

Cited by 20 publications

(8 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These ML frameworks are entirely based on device data and do not include device physics. In addition to this, the ML techniques have been further used to predict and model the characteristics variations that occurred in different semiconductor devices due to WKF [49], random dopant distribution (RDD) [50], line-edge-roughness (LER) [51], [52], process variation effect (PVE) [53], etc. In some work, ML is also applied for the estimation of threshold voltage variability by random telegraph noise fluctuation [54], and by device parameter variability [55], etc.…”

Section: A Preliminaries and Related Workmentioning

confidence: 99%

A Machine Learning Approach to Modeling Intrinsic Parameter Fluctuation of Gate-All-Around Si Nanosheet MOSFETs

Butola

Kola

2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

The sensitivity of semiconductor devices to any microscopic perturbation is increasing with the continuous shrinking of device technology. Even the small fluctuations have become more acute for highly scaled nano-devices. Therefore, these fluctuations need to be addressed extensively in order to continue further device scaling. In this paper, we mainly focus on three intrinsic parameter fluctuation sources, work function fluctuation (WKF), random dopant fluctuation (RDF), and interface trap fluctuation (ITF) for gate-all-around (GAA) silicon nanosheet (NS) MOSFETs. Generally, the effect of these fluctuations is analyzed using a time-consuming device simulation process. A machine learning (ML) based powerful and efficient artificial neural network (ANN) model is used to accelerate this process. Firstly, the effects of fluctuation sources are analyzed individually by using the ANN model and results have been presented that show the WKF variations dominate the threshold voltage (VTH), off-state current (IOFF), and on-state current (ION) variations among other fluctuation sources. Next, we examine the combined effect of all three fluctuation sources. It is crucial because considering only one fluctuation can result in unexpected variations due to other fluctuations present in the device. Therefore, the ANN model is used to estimate the combined effects as well. The results show that the proposed model predicts the outputs with an R 2 -score of 99% and an error rate of less than 1%. In addition, the ML is also utilized to calculate the permutation importance of input variables as a measure to investigate the contribution of each fluctuation source.

show abstract

Section: A Preliminaries and Related Workmentioning

confidence: 99%

A Machine Learning Approach to Modeling Intrinsic Parameter Fluctuation of Gate-All-Around Si Nanosheet MOSFETs

Butola

Kola

2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…Thus, they proposed an artificialneural-network-based machine learning (ML) approach with more accurate results for process variations [25]. Recently, various deep learning (DL) techniques have been studied for the WKF on the GAA Si NW devices with high efficiency and accuracy [26]. The advantage of these ML/DL models is data-oriented, not model-oriented.…”

Section: Table I the Process Parametersmentioning

confidence: 99%

A Nanosized-Metal-Grain Pattern-Dependent Threshold Voltage Model for the Work Function Fluctuation of GAA Si NW MOSFETs

Sung

Li²

2021

IEEE Access

Self Cite

View full text Add to dashboard Cite

To estimate characteristic fluctuation of emerging devices, three-dimensional device simulation has been performed intensively for various random cases; however, it strongly relies on huge computational resource. In this paper, we report a nanosized-metal-grain pattern-dependent model by using the method of multi-variable non-linear regression (MVNLR) for the threshold voltage fluctuation (Vth) of gate-all-around silicon nanowire metal-oxide-semiconductor field-effect transistors. The model is developed by perturbing the local metal grains of specific patterns and summing up each metal grain of various grain patterns. The method of MVNLR is applied to build equations for nominal devices, and devices with high work-function (WK) and low WK with respective to process parameters (radius of channel, gate length, channel doping, and oxide thickness), which are generated by the design of experiments. The proposed model can estimate the magnitude of Vth accurately (error rate < 1%) for all parameters, where an error-correction process is further adopted by using a simplified one-perceptron method.

show abstract

“…Akbar et al used ML methods to assist device simulation of work function fluctuations for 3D multi-channel gates around silicon nanosheet MOSFETs [14]. DL algorithms are also used to evaluate the work function fluctuations of GAAFETs [15,16].…”

Section: Introductionmentioning

confidence: 99%

Prediction of electrical properties of GAAFET based on integrated learning model

Zhang,

Chen,

Wang

et al. 2024

Nanotechnology

View full text Add to dashboard Cite

As device feature sizes continue to decrease and fin field effect transistors (FinFETs) reach their physical limits, gate all around field effect transistors (GAAFETs) have emerged with larger gate control areas and stackable characteristics for better suppression of second-order effects such as short-channel effects due to their gate encircling characteristics. Traditional methods for studying the electrical characteristics of devices are mostly based on the technology computer-aided design (TCAD). Still, it is not conducive to developing new devices due to its time-consuming and inefficient drawbacks. Deep learning (DL) and machine learning (ML) have been well-used in recent years in many fields. In this paper, we propose an integrated learning model that integrates the advantages of DL and ML to solve many problems in traditional methods. This integrated learning model predicts the direct current characteristics, capacitance characteristics, and electrical parameters of GAAFET better than those predicted by DL or ML methods alone, with a linear regression factor (R2) greater than 0.99 and very small root mean square error (RMSE). The proposed integrated learning model achieves fast and accurate prediction of GAAFET electrical characteristics, which provides a new idea for device and circuit simulation and characteristics prediction in microelectronics.

show abstract

Deep Learning Algorithms for the Work Function Fluctuation of Random Nanosized Metal Grains on Gate-All-Around Silicon Nanowire MOSFETs

Cited by 20 publications

References 68 publications

A Machine Learning Approach to Modeling Intrinsic Parameter Fluctuation of Gate-All-Around Si Nanosheet MOSFETs

A Machine Learning Approach to Modeling Intrinsic Parameter Fluctuation of Gate-All-Around Si Nanosheet MOSFETs

A Nanosized-Metal-Grain Pattern-Dependent Threshold Voltage Model for the Work Function Fluctuation of GAA Si NW MOSFETs

Prediction of electrical properties of GAAFET based on integrated learning model

Contact Info

Product

Resources

About