Here is the abstract you requested from the hitec_2014 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|A 256 kb (32kx8) EEPROM for >200 C Applications|
|Keywords: EEPROM, retention, temperature|
|Dennis Adams1, Randall Lewis1, Jason O’Brien1, Bill Hand1, Greg Marsh1, Ian Manwaring3, Donald Pierce2, Cory Sherman1, Sze Wong1 December 19, 2013 A memory retention study was performed on the W28C256 32kx8 EEPROM device to assess its suitability for use in high temperature applications above 200 °C. This study indicates this device has memory retention well in excess of 10 years at 225 °C. During 2014, characterization and extended life testing is planned to further assess device reliability for high temperature applications. The Northrop Grumman W28C256 32kx8 EEPROM is a radiation hardened device that has been successfully used in space applications since 1995. Nonvolatility is achieved using SONOS (Silicon-Oxide-Nitride-Oxide-Silicon) charge storage technology. This design utilizes a four transistor memory cell (compared to a two transistor memory cell used in the W28C0108 128kx8 EEPROM). The objective of this study was to evaluate whether the four transistor memory cell and increased programming voltage (10 V vs 7.5 V for the 128kx8 device) would lead to improved high temperature memory retention. Cumulative time to retention failure distributions were generated by programming devices with a checkerboard pattern, followed by unbiased exposure to high temperatures, with interval retention testing at room temperature. Sample sizes varied from 14 to 33 parts. The failure criterion was defined as a single bit failure in a device memory. The activation energy was then derived based on traditional Arrhenius relationship for median time to failure (MTTF); MTTF = A*exp(Ea/kT) where Ea is activation energy, k= Boltzman’s constant and T= temperature The MTTFs measured were 1.6 days, 6 days and >42 days at 375, 350 and 300 °C, respectively (no failures were seen out to 42 days at 300 °C and testing was suspended). From these data, the thermal activation energy for memory retention was calculated to be 2.27 eV. Estimated retention life time at +225 C was 410 years. From this work, a wafer level screen was established to remove devices from the population that do not meet established retention requirements for >200 °C applications Fixturing is currently being built to allow electrical characterization of this device up to 300 °C. Accelerated life testing at 250 °C will be performed in 2014. A study is also planned to assess memory retention when devices are programmed at elevated temperatures for in situ reprogrammability. 1 – Northrop Grumman Corporation (Baltimore, MD) 2 – Sandia Technologies Incorporated (Albuquerque, NM) 3 – Rochester Institute of Technology (Rochester, NY)|
|Dennis A. Adams, Senior Consulting Engineer
Northrop Grumman Corporation