Electrical and Computer Engineering ETDs

Publication Date

7-12-2014

Abstract

Carbon nanotube field-effect transistor (CNFET) is one of the promising candidates as extensions to silicon CMOS devices. The CNFET, which is a 1-D structure with a near-ballistic transport capability, can potentially offer excellent device characteristics and order-of-magnitude better energy-delay product over standard CMOS devices. Significant challenges in CNT synthesis prevent CNFETs today from achieving such ideal benefits. CNT density variation and metallic CNTs are the dominant type of CNT variations/imperfections that cause performance variation, large static power consumption, and yield degradation. We present an imperfection-aware design technique for CNFET digital VLSI circuits by: 1) Analytical models that are developed to analyze and quantify the effects of CNT density variation on device characteristics, gate and system levels delays. The analytical models, which were validated by comparison to real experimental/simulation data, enables us to examine the space of CNFET combinational, sequential and memory cells circuits to minimize delay variations. Using these model, we drive CNFET processing and circuit design guidelines to manage/overcome CNT density variation. 2) Analytical models that are developed to analyze the effects of metallic CNTs on device characteristics, gate and system levels delay and power consumption. Using our presented analytical models, which are again validated by comparison with simulation data, it is shown that the static power dissipation is a more critical issue than the delay and the dynamic power of CNFET circuits in the presence of m-CNTs. 3) CNT density variation and metallic CNTs can result in functional failure of CNFET circuits. The complete and compact model for CNFET probability of failure that consider CNT density variation and m-CNTs is presented. This analytical model is applied to analyze the logical functional failures. The presented model is extended to predict opportunities and limitations of CNFET technology at todays Gigascale integration and beyond.'

Keywords

CNFET, CNT Density Variation, Metallic CNTs, Imperfection, Variation, Failure Analysis, Digital VLSI

Document Type

Dissertation

Language

English

Degree Name

Computer Engineering

Level of Degree

Doctoral

Department Name

Electrical and Computer Engineering

First Advisor

ZARKESH-HA, PAYMAN

First Committee Member (Chair)

Hossein-Zadeh, Mani

Second Committee Member

Leseman, Zayd

Third Committee Member

Plusquellic, James

Fourth Committee Member

ZARKESH-HA, Payman

Available for download on Wednesday, May 16, 2018

Share

COinS