Please use this identifier to cite or link to this item:
http://localhost/handle/Hannan/587445
Title: | A Quantitative Analysis of DICE SRAM SEU Caused by Heavy Ion Elastic Scattering |
Authors: | Ming Zhu;Hengjing Zhu;Wei Zhang;Qingkui Yu;Min Tang |
subject: | DICE|Monte Carlo simulation|TCAD simulation|NRED|single event upset |
Year: | 2016 |
Publisher: | IEEE |
Abstract: | A 4 Mbit DICE latch SRAM exhibits single event upset(SEU) cross-section of 1014 -1012 cm2 /bit in vertically incident heavy-ion beam tests. TCAD simulations show that the DICE latch works, and that the SEU is not caused by charge sharing. Numerical simulations on particle elastic collision gives quantitative evidence that it is caused by elastic scattering between the incident ion and target nucleus. Three scenarios of elastic scattering SEU are calculated separately in order to get a deeper understanding. The upset cross-section's dependencies on ion energy, incident angle and the distance between DICE nodes are analyzed. A designing suggestion that paired nodes of DICE latch should be at least 6 μm away is firstly presented. |
URI: | http://localhost/handle/Hannan/167301 http://localhost/handle/Hannan/587445 |
ISSN: | 0018-9499 1558-1578 |
volume: | 63 |
issue: | 4 |
Appears in Collections: | 2016 |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
7473863.pdf | 1.4 MB | Adobe PDF | ![]() Preview File |
Title: | A Quantitative Analysis of DICE SRAM SEU Caused by Heavy Ion Elastic Scattering |
Authors: | Ming Zhu;Hengjing Zhu;Wei Zhang;Qingkui Yu;Min Tang |
subject: | DICE|Monte Carlo simulation|TCAD simulation|NRED|single event upset |
Year: | 2016 |
Publisher: | IEEE |
Abstract: | A 4 Mbit DICE latch SRAM exhibits single event upset(SEU) cross-section of 1014 -1012 cm2 /bit in vertically incident heavy-ion beam tests. TCAD simulations show that the DICE latch works, and that the SEU is not caused by charge sharing. Numerical simulations on particle elastic collision gives quantitative evidence that it is caused by elastic scattering between the incident ion and target nucleus. Three scenarios of elastic scattering SEU are calculated separately in order to get a deeper understanding. The upset cross-section's dependencies on ion energy, incident angle and the distance between DICE nodes are analyzed. A designing suggestion that paired nodes of DICE latch should be at least 6 μm away is firstly presented. |
URI: | http://localhost/handle/Hannan/167301 http://localhost/handle/Hannan/587445 |
ISSN: | 0018-9499 1558-1578 |
volume: | 63 |
issue: | 4 |
Appears in Collections: | 2016 |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
7473863.pdf | 1.4 MB | Adobe PDF | ![]() Preview File |
Title: | A Quantitative Analysis of DICE SRAM SEU Caused by Heavy Ion Elastic Scattering |
Authors: | Ming Zhu;Hengjing Zhu;Wei Zhang;Qingkui Yu;Min Tang |
subject: | DICE|Monte Carlo simulation|TCAD simulation|NRED|single event upset |
Year: | 2016 |
Publisher: | IEEE |
Abstract: | A 4 Mbit DICE latch SRAM exhibits single event upset(SEU) cross-section of 1014 -1012 cm2 /bit in vertically incident heavy-ion beam tests. TCAD simulations show that the DICE latch works, and that the SEU is not caused by charge sharing. Numerical simulations on particle elastic collision gives quantitative evidence that it is caused by elastic scattering between the incident ion and target nucleus. Three scenarios of elastic scattering SEU are calculated separately in order to get a deeper understanding. The upset cross-section's dependencies on ion energy, incident angle and the distance between DICE nodes are analyzed. A designing suggestion that paired nodes of DICE latch should be at least 6 μm away is firstly presented. |
URI: | http://localhost/handle/Hannan/167301 http://localhost/handle/Hannan/587445 |
ISSN: | 0018-9499 1558-1578 |
volume: | 63 |
issue: | 4 |
Appears in Collections: | 2016 |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
7473863.pdf | 1.4 MB | Adobe PDF | ![]() Preview File |