3D coupled electro-thermal simulations for SOI FinFET with statistical variations including the fin shape dependence of the thermal conductivity

Abstract

Thermal and electrical transport in FinFET with statistical variations is investigated by 3D coupled electro-thermal simulation, using statistical-variabilityaware device simulator GARAND with built-in thermal simulation module. The module employs a new formula for the calculation of the thermal conductivity in the fin region with fin shape dependence. An SOI FinFET structure with combined statistical sources of GER, FER and MGG is studied, with a focus on the lattice temperature profile and the statistical distribution of on current.

Impact of Self-Heating on the Statistical Variability in Bulk and SOI FinFETs

“…[12], we have developed a thermal simulation module to investigate the impact of self-heating on FinFET DC operation and on the corresponding statistical variability. This module is based on the solution of the coupled Heat Flow, Poisson, and Current Continuity Equations [8].…”

“…We employ a new approximate formula for the calculation of the thermal conductivity in the fin region, which extends the previous 1D formula [15] to 2D [8]. Considering a fin of height h and width w as shown in Fig.…”

“…This is complemented by lower statistical variability due to a tolerance of lower channel doping compared to bulk transistors. Self-heating has been highlighted as one of the areas of concern for FinFETs because of the 3D geometry, the impact of the corresponding 3D heat flow through the fin to the substrate and the impact of the fin geometry on the local thermal conductivity [2][3][4][5][6][7][8]. These perceived self-heating problems could be exacerbated in This Manuscript received December 10, 2014; revised April 21, 2015.…”

“…Therefore analyzing and modeling the self-heating in FinFETs and the influence on device performance has become one of the topics attracting a lot of recent interest [2][3][4][5][6][7][8].…”

“…In this paper we use the enhanced self-heating simulation capabilities, recently introduced in the GSS 'atomistic' device simulator GARAND [8,12], complemented by the unique capabilities for physical simulation of different sources of statistical variability [13][14], to study, for the first time, the impact of self heating on FinFET variability. The test-bed transistors in this study are bulk and SOI FinFETs designed to meet the specifications for the 14/16nm CMOS technology generation [11].…”