• DocumentCode
    2922313
  • Title

    Study of Transistor and Product NBTI Lifetime Distributions

  • Author

    Qin, Jin ; Yan, Baoguang ; Shoshany, Yossi ; Roy, Druker ; Rahamim, Hezi ; Marom, Haim ; Bernstein, Joseph B.

  • Author_Institution
    Reliability Eng., Univ. of Maryland, College Park, MD
  • fYear
    2008
  • fDate
    12-16 Oct. 2008
  • Firstpage
    64
  • Lastpage
    67
  • Abstract
    NBTI has been extensively studied to understand physics of degradation in recent years. However, little has been done to find out the lifetime distributions of NBTI at both transistor and product level, which are important in reliability prediction and improvement. In this paper, Monte-Carlo simulation is carried out to study the NBTI lifetime distribution at transistor level. Lognormal distribution is found to have the best fit. Product level NBTI lifetime distribution is studied through rare event simulation. Result shows that Weibull distribution has a better fit than lognormal distribution at product level. Acceleration test result of 90 nm SRAM cache NBTI degradation is compared with the simulation results and a good agreement is observed.
  • Keywords
    MOSFET; Monte Carlo methods; SRAM chips; Weibull distribution; log normal distribution; semiconductor device reliability; semiconductor device testing; thermal stability; Monte-Carlo simulation; NBTI lifetime distribution; SRAM cache; Weibull distribution; acceleration test; deep submicron CMOS process; lognormal distribution; negative bias temperature instability; rare event simulation; transistor level; Degradation; Discrete event simulation; Equations; Integrated circuit reliability; Life estimation; Niobium compounds; Predictive models; Stress; Titanium compounds; Weibull distribution;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Integrated Reliability Workshop Final Report, 2008. IRW 2008. IEEE International
  • Conference_Location
    S. Lake Tahoe, CA
  • ISSN
    1930-8841
  • Print_ISBN
    978-1-4244-2194-7
  • Electronic_ISBN
    1930-8841
  • Type

    conf

  • DOI
    10.1109/IRWS.2008.4796088
  • Filename
    4796088