University of Illinois Chicago
Browse

Mitigation of Threshold Voltage Variability by Exploiting Interaction of Oxygen Vacancies and MGG

Download (2.45 MB)
thesis
posted on 2020-05-01, 00:00 authored by Madhu Padmanabha Sumangala
The scaling of transistor beyond 65nm was backed by replacing SiO2 by a high-k dielectric material such as HfO2. The use of HfO2 on the silicon bulk introduced positively charged thermodynamic point defect oxygen vacancy. The surplus electrons transferred to the gate of the transistor by oxygen vacancy, creates a dipole and hence affects the local electrostatics. As formation of oxygen vacancy is random and statistical in nature, this will create its own variability. The deposition of metal gate of the transistor in the form of grains of various size along different orientation is known to posses varying workfunctions. This alters the localized flatband voltage causing threshold voltage variability. The higher value of workfunction reduces the surface potential, whereas lower value of the workfunction increases the surface potential. Interestingly, the positively charged oxygen vacancies create voltage spikes. Hence, by carefully placing oxygen vacancies via altering processing conditions, the effect of metal grain granularity can be invalidated. The law of mass action based oxygen vacancy generation model was used to simulate the theory in Synopsys's Sentaurus TCAD Tool. With the proposed method, the threshold voltage variability induced by metal grain granularity was reduced by 50% for 9nm average edge length.

History

Advisor

Trivedi, Amit Ranjan

Chair

Trivedi, Amit Ranjan

Department

Electrical and Computer Engineering

Degree Grantor

University of Illinois at Chicago

Degree Level

  • Masters

Degree name

MS, Master of Science

Committee Member

Strocio, Michael Chen, Pai-Yen

Submitted date

May 2020

Thesis type

application/pdf

Language

  • en

Usage metrics

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC