Predicting the vulnerability of memories to muon-induced SEUs with low-energy proton tests informed by Monte Carlo simulations

Abstract

A method for predicting device vulnerability to muon induced single event upsets (SEUs) using proton tests and Monte Carlo simulations was developed and validated using a 28 nm commercial static random access memory (SRAM). This method replaces testing with muon beams while still incorporating the device's response to stopping charged particles. Monte Carlo simulations are used to relate energy deposition by stopping protons to that of muons. This mapping can be employed to determine the proton test energies required to simulate exposure to a muon environment. This method is superior to simulations since the device's empirical response to proton exposure is used to determine upper and lower particle energy bounds for muon susceptibility. This window of vulnerability can then be used to predict a conservative estimate of the soft error rate for muon exposures.