Chuyang Hong

A Surface-Field-Based Model for Nanowire MOSFETs With Spatial Variations of Doping Profiles

We report a novel method to solve the nonlinear 1-D Poissons equation for the gate-all-around (GAA) nanowire MOSFETs with nonuniform doping profiles. An algebraic relation between the electric field and potential is identified to develop a surface-field-based compact model. Drain current is derived from the oxide-interface boundary condition to avoid the complicated surface-potential approach. As verified by the TCAD simulations, this analytic model is capable of continuously covering all operating regions of GAA nanowire MOSFETs with slowly varying doping profiles.

An Analytic Surface-Field-Based Quasi-Atomistic Model for Nanowire MOSFETs With Random Dopant Fluctuations

For the first time, an analytic surface-field-based model for nanowire MOSFETs with random dopant fluctuations (RDF) is reported. In this model, the depletion charge due to the discrete dopant distribution is described by the Dirac δ functions, while the mobile charge keeps its continuous form. By introducing two new variables, the discrete 1-D Poissons equation is transformed into a simple algebraic equation to correlate the surface potential with the field (due to the inversion charge). Without solving the potential distribution, the drain current can be calculated from the Pao-Sah integral using the oxide-interface boundary condition. This model is shown to be more accurate in predicting the RDF effects than the continuous TCAD simulations for all the operating regions. We also discuss the RDF-incorporated short-channel effects by solving the discrete 2-D Poissons equation in the subthreshold regime.

A Unified Continuous and Discrete Model for Double-Gate MOSFETs With Spatially Varying or Pulsed Doping Profiles

This paper presents a unified continuous and discrete model covering all device operating regions of double-gate MOSFETs for the first time. With a specific variable transformation method, the 1-D Poisson’s equation in the Cartesian coordinate for double-gate MOSFETs is transformed into the corresponding form in the cylindrical coordinate. Such a transformed cylindrical Poisson’s equation results in a simple algebraic equation, which correlates the (inversion-charge induced) surface potential to the field and allows the long-channel drain-current formula to be derived from the Pao–Sah integral. This model can be readily applied to predict the effects of both continuous and discrete doping variations. The short-channel-effect model is also developed by solving the 2-D Poisson’s equation using the eigenfunction-expansion method. The accuracy of both long-channel and short-channel models is confirmed by the numerical calculations and TCAD simulations.

A General and Transformable Model Platform for Emerging Multi-Gate MOSFETs

The complete general solution of nonlinear 1-D undoped Poisson’s equation, in both Cartesian and cylindrical coordinates, is derived by employing a special variable transformation method. A general model platform for various types of emerging multi-gate MOSFETs is further constructed and verified with TCAD simulations. It is shown that this model platform is suitable for analyzing a series of emerging devices, such as double-surrounding-gate, inner-surrounding-gate, and outer-surrounding-gate nanoshell MOSFETs, all of which require different boundary conditions from the conventional gate-all-around nanowire device.

A Nonlinear Surface-Field Compact Model for Junctionless Nanowire MOSFETs

A nonlinear surface-field-based model for heavily doped JL nanowire MOSFETs is developed. By introducing two specific transformation variables, the surface potential to the field (due to the mobile charge) are correlated by a simple algebraic relation. Without solving the electrostatic potential, a drain current model is derived by the Pao-Sah integral. A second-order correction is carried out to improve the model accuracy.

A Continuous Compact Model Incorporating Higher-Order Correction for Junctionless Nanowire Transistors With Arbitrary Doping Profiles

A continuous surface-field-based compact model for heavily doped junctionless nanowire transistors is developed. By constructing specific transformation variables, an algebraic relation between the surface field and potential is identified to facilitate a computationally efficient model for calculating the (long-channel) drain current from the oxide-interface boundary condition with no need of the surface-potential approach. The model accuracy at high doping levels is further improved by incorporating a second-order correction. A short-channel-effect model is also developed using the eigenfunction-expansion method to solve the 2-D Poissons equation. A good agreement between the model prediction and TCAD simulations is observed. Finally, high model accuracy is verified by the experimental data.

Author’s Reply to “Comments on ‘A General and Transformable Model Platform for Emerging Multi-Gate MOSFETs’”

The authors would like to apologize for and correct a few errors about the references in our recently published paper ( [1] of this reply), and make some comments: 1) References [26] (by T. Alfrey et al. ) and [32] in our paper should be [2] (by R. M. Fuoss et al. ) and [3] as listed in this reply, respectively. 2) We would also like to recognize the work of Paolucci et al. ( [4] of this reply), in particular, their introduction of two transformation variables (Eq. (4)) to solve the nonlinear cylindrical 1-D Poisson’s Equation. We were not aware of their work at the time of our paper publication. Actually, the involved transformation variables/method were first reported by Fuoss et al. [2] , hereinafter referred to as Fuoss’ transformation variables/method . On the other hand, it should be pointed out that before the paper by Paolucci et al. was submitted for consideration of publication, we had been aware of the work of Fuoss et al. The related early work of our corresponding author (Chen) dates back to 2001 (e.g., [37] in our paper). All the research reports of our students, including the cited thesis ([38] in our paper) of Jun Zhou (one of our authors), have been well documented in the database and library of our university. Jun Zhou’s first report on Fuoss’ transformation variables was submitted in November of 2014 (the evidence material has been submitted to the editor for a review). In his report, Jun Zhou used Fuoss’ transformation variables to prove that the cylindrical nonlinear Poisson’s equation can be transformed to the Cartesian form. His thesis proposal [5] was submitted in November of 2015 for an approval from his advisor and the thesis committee. The thesis was completed and officially signed (and documented in our university library) in May of 2016.

A surface-field-based compact model for double-gate MOSFETs with continuous and discrete doping profiles

A unified continuous and discrete compact model covering all device operating regions for double-gate MOSFETs is developed. By applying special variable transformation method, a simple algebraic equation is derived to correlate the inversion-charge induced surface potential to field. Such an algebraic relation allows the drain-current formula to be derived without solving the surface-potential distribution.

A surface-field-based modeling platform for heavily doped junctionless nanowire MOSFETs

A continuous surface-field based modeling platform for heavily doped junctionless nanowire transistors is presented. By constructing specific transformation variables, a straightforward relation between the surface potential and field is identified to facilitate an efficient model for calculating the (long-channel) drain current from the oxide-interface boundary condition with no need of a surface-potential approach. Such a surface-field-based modeling platform may enable a paradigm shift in compact modeling strategy, which is also applicable to device doping optimization.

Impacts of process variability of alternating-material self-aligned multiple patterning on SRAM circuit performance

In this paper, we propose a novel modular patterning technology to reduce the edge-placement errors (EPE) significantly by combining alternating-material self-aligned multiple patterning (altSAMP) and selective etching processes. It is assumed that gates and fins are fabricated by the same type of altSAMP process as mixing two different processing techniques will drive up the manufacturing costs. Process variability induced circuit performance degradation is shown to be a serious issue as FinFET devices are scaled down to sub-10nm. We analyze the dependence of FinFET-based SRAM circuit performance on supply voltage, fin-width and gate-length variations. Improved device control with narrower fins helps to increase the static noise margin (SNM) in all SRAM cell designs. Higher supply voltage is also beneficial to the SNM performance. Our simulation results show that 6-T SRAM circuit design does not meet the six-sigma yield requirement when the half pitch is scaled down to sub-7 nm. To reduce the SRAM circuit variability, we study an 8-T SRAM cell and show that it significantly improves the SRAM performance.