Failure Distributions of Successive Dielectric Breakdown Events

Abstract

Experimental time-dependent dielectric breakdown (TDDB) distributions of standard CMOS hardware are used to demonstrate the problem of detecting a range of successive breakdown events, e.g., from the 5th to the 20th. Specifically, the range and the statistical distribution of successive breakdown events significantly change the shape of the cumulative failure distribution, which is crucial for the low percentile extrapolation. The observed behavior affects the correlation of integrated circuit failure distributions to dielectric breakdown as it is expected that circuit malfunction is caused by a range of successive breakdown events.Index Terms-CMOS integrated circuits, dielectric breakdown, oxide reliability, ultrathin gate oxides.

The “buffering” role of high-к in post breakdown degradation immunity of advanced dual layer dielectric gate stacks

“…Considering the Weibull parameters for the time-to-first-IL SBD after area scaling from single transistor to circuit level taking 10 9 transistors (β = 0.85, η = 9.49 × 10 5 sec at V op = 1V) and also the scale and shape parameters for leakage current distribution after IL SBD (Fig. 8) → (β I = 0.869, η I = 28.1 nA at V op = 1V, assumed Weibull here), we make use of multiple BD statistics [44] and Monte Carlo simulations [45] to evaluate the time needed for reaching the standard circuit failure criteria of I gl ~ 10μA. The results of the statistical simulation of percolated current increase with time are plotted in Fig.…”

Statistical characterization and reliability modeling of novel high-K gate dielectric stacks

“…The distributions are non-Weibull and the Weibull slope increases for higher number of BD events, as justified previously in Ref. [234].…”

“…4.27 for a particular stack of t HK : t IL = 32Ǻ : 16Ǻ. While the statistics of multiple BD is known to be non-Weibull (even for SiO 2 ) [234], our simulation shows that the 1% time to failure for 10 SBD events is only an order of magnitude higher than that taken for the 1 st SBD event. For increasing number of BD events, the distributions get closer to each other and therefore, even for 1000 BD events (considering nano-ampere level leakage from every IL SBD event), the lifetime enhancement relative to the 1 st BD event will not be more than a factor of 1000.…”

“…This suggests that low percentile lifetime enhancement becomes less prominent for a large number of multiple BD events. At high percentiles, it is well known that the improvement in lifetime is less evident for all multiple BD events [234]. For K > 50 BD events, the maximum lifetime enhancement possible is of the order of 10 9 seconds, corresponding to F = 1 ppm.…”

On dielectric breakdown statistics

“…The Weibull statistic (Weibull 1939(Weibull , 1951 has frequently been adopted to investigate the breakdown behaviour of different systems (Ribes et al 2005, Pompl and Kerber 2005, Dissado 1990, Rowland et al 1986, Hill and Dissado 1983, Coppard et al 1990, Dissado 2002, Cran 1976, Filliben 1975, Looney and Gulledge Jr 1985. In these investigations an 3 Author to whom any correspondence should be addressed.…”