Unveiling the electromigration physics of ULSI interconnects through statistics

Electromigration is an important failure phenomenon in ULSI. There are several underlying physical mechanisms in an electromigration process depending on the stress conditions. In this work, we use various statistical techniques to investigate the number of such mechanisms for a given set of electromigration data. The statistical distribution of each mechanism can be found and the failure units that belong to the different statistical distributions can be identified. Through this methodology, the existence of the incubation phenomena for aluminum electromigration is shown, and the time for the incubation is found to be different for different underlying mechanisms.

[1]  Wei Li,et al.  Development of highly accelerated electromigration test , 2006, Microelectron. Reliab..

[2]  Cher Ming Tan,et al.  Reliability Improvement in Al Metallization: A Combination of Statistical Prediction and Failure Analytical Methodology , 2004, Microelectron. Reliab..

[3]  H. Toyoda,et al.  Electromigration induced aluminum atom migration retarding by grain boundary structure stabilization and copper doping , 1999 .

[4]  Joel Keizer,et al.  Statistical Thermodynamics of Nonequilibrium Processes , 1987 .

[5]  T. L. Hill,et al.  An Introduction to Statistical Thermodynamics , 1960 .

[6]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[7]  H. Akaike.  A new look at the statistical model identification , 1974 .

[8]  A. Krishnamoorthy,et al.  Current crowding effect on copper dual damascene via bottom failure for ULSI applications , 2005, IEEE Transactions on Device and Materials Reliability.

[9]  Emmanuel Yashchin,et al.  Threshold electromigration failure time and its statistics for Cu interconnects , 2006 .

[10]  Dimitri Kececioglu,et al.  Maximum likelihood estimates, from censored data, for mixed-Weibull distributions , 1992 .

[11]  Benoit B. Mandelbrot,et al.  An outline of a purely phenomenological theory of statistical thermodynamics-I: Canonical ensembles , 1956, IRE Trans. Inf. Theory.

[12]  T Nogami,et al.  The electromigration lifetime determined from the minimum time to failure in an acceleration test , 1995 .