Abstract

Discrete impurity effects in terms of their statistical variations in number and position in the inversion and depletion region of a MOSFET, as the gate length is aggressively scaled, have recently been researched as a major cause of reliability degradation observed in intra-die and die-to-die threshold voltage variation on the same chip resulting in significant variation in saturation drive (on) current and transconductance degradation—two key metrics for benchmark performance of digital and analog integrated circuits. In the following chapter, the authors have highlighted the random dopant fluctuation (RDF) based Ensemble Monte Carlo(EMC)device simulation study conducted by the Computational Electronics (COMPUTEL) research group of Arizona State University. In addition to RDF, random number and position of interface traps lying close to Si:SiO2 interface engender additional concerns leading to enhanced experimentally observed fluctuations in drain current and threshold voltage. In this context, the authors of this chapter present novel EMC based simulation studies on trap induced random telegraph noise (RTN) responsible for statistical fluctuation pattern observed in threshold voltage, its standard deviation and drive current in saturation for 45 nm gate length MOSFET device. From the observed simulation results and their analysis, it can be cogently projected that with continued scaling in gate length and width, RTN effect will eventually supersedeas a major reliability bottleneck over the typical RDF phenomenon. The fluctuation patterns observed by EMC simulation outcomes for both drain current and threshold voltage have been analyzed and explained from analytical device physics perspectives.

Original languageEnglish (US)
Title of host publicationCircuit Design for Reliability
PublisherSpringer New York
Pages47-67
Number of pages21
ISBN (Print)9781461440789, 9781461440772
DOIs
StatePublished - Jan 1 2015

ASJC Scopus subject areas

  • General Engineering

Fingerprint

Dive into the research topics of 'Atomistic simulations on reliability'. Together they form a unique fingerprint.

Cite this