Impact of crystallization statistics on data retention for phase change memories

Author

Redaelli, A. ; Ielmini, D. ; Lacaita, A.L. ; Pellizzer, F. ; Pirovano, A. ; Bez, R.

Author_Institution

Dipt. di Elettronica e Informazione, Politecnico di Milano

fYear

2005

fDate

5-5 Dec. 2005

Firstpage

742

Lastpage

745

Abstract

The stochastic nature of percolation is shown as a possible issue for retention in PCM devices, due to the occurrence of unlikely crystallization events that early decrease the device resistance. Failure time dispersions at high temperatures are measured and analyzed through a crystallization Monte Carlo model. A physical insight into nucleation and growth mechanisms is thus provided and a maximum working temperature of 105degC is extracted to guarantee, on large statistics, the 10 years data retention requirement for non volatile applications

Keywords

Monte Carlo methods; crystallisation; percolation; phase change materials; random-access storage; 10 years; 105 C; Monte Carlo models; crystallization events; crystallization statistics; data retention; failure time dispersions; phase change memories; Crystallization; Dispersion; Electrical resistance measurement; Failure analysis; Phase change materials; Phase change memory; Statistics; Stochastic processes; Temperature measurement; Time measurement;

fLanguage

English

Publisher

ieee

Conference_Titel

Electron Devices Meeting, 2005. IEDM Technical Digest. IEEE International

Conference_Location

Washington, DC

Print_ISBN

0-7803-9268-X

Type

conf

DOI

10.1109/IEDM.2005.1609460

Filename

1609460