T. Nuns

Low energy proton induced SEE in memories

This paper presents proton sensitivity curves obtained on memories from various generations. Highlight is set on component response at very low energy (down to 5 MeV) and implications on the calculated SEE rates are discussed.

Memories Response to MBU and Semi-Empirical Approach for SEE Rate Calculation

This paper deals with SEE rate prediction and proposes a semi-empirical approach that makes no use of the RPP concept. The investigation is based on a reduced set of ground data and SEU/MBU results are compared with in-flight data obtained on memories on-board ICARE (SAC-C orbit). Two-dimensional mixed-mode simulations complete this study and provide first insight to understand the observed behavior

In flight observation of proton induced destructive single event phenomena

MEX Experience Module is a part of CARMEN2 instrument launched in June 22 2008 aboard JASON2 satellite. This scientific instrument is dedicated to the study of the effects of space radiation environment on various electronic devices. Among all the phenomena studied in this experiment, this paper focuses on the data collected on destructive SEEs: Latch-up on commercial SRAMs and Burnout on power MOSFETs.

In Flight SEU/MCU Sensitivity of Commercial Nanometric SRAMs: Operational Estimations

A method and the corresponding platform devoted to operational SEE-rate prediction are presented and illustrated by experimental results. Predicted error-rates are in well agreement with results issued from the activation of an SRAM platform, in 90 nm technology node, on board stratospheric balloons flights. Direct ionization of protons is investigated for a 65 SRAM memory virtually boarded on the balloon flight.

“Effective NIEL” in Silicon: Calculation Using Molecular Dynamics Simulation Results

The non-ionizing energy loss (NIEL), often used to scale the damage coefficients of irradiated electronic components, sometimes fails to make good degradation predictions. The classical non-ionizing energy loss calculation, performed under binary cascade approximation, lacks accuracy at low incident energy for some light particles such as electrons. The amount of displacement damage predicted by molecular dynamics simulations where many-body interactions are naturally included can be significantly different from the classical binary collision calculation. The creation of damage predicted by molecular dynamics technique, within the low energy deposition regime, can be quite different from what can be classically simulated. For instance, earlier studies demonstrated that, under suitable conditions, low energy transfer below the atomic displacement threshold can increase damage production. In the case of silicon material, molecular dynamics simulation results have been incorporated within the definition of a new energy partition function. In comparison with the classical Kinchin Pease approach, our improved model more accurately estimates the number of displacements generated by a primary knock-on atom. This new energy partition function is used to calculate corrected non- ionizing energy loss.

Annealing of Proton-Induced Random Telegraph Signal in CCDs

A silicon CCD imager has been irradiated with 10 MeV protons and measurements focused on random telegraph signal (RTS). A variance detection method is applied for quantifying the number of RTS after irradiation and through isochronal annealing. The observed behavior is analyzed and corresponds to the annealing of phosphorus-vacancies.

Measurements of Random Telegraph Signal in CCDs Irradiated With Protons and Neutrons

CCD imagers have been irradiated with 10 to 100 MeV protons, 45 MeV neutrons and measurements focussed on random telegraph signal (RTS) characterization. The objective is to propose a method for RTS detection and to analyse pixels behaviour with temperature, particle species and energy.

Light Particle-Induced Single Event Degradation in SDRAMs

SDRAMs have been exposed to protons, neutrons, alpha particles, heavy ions and Co60 irradiations. Numerous cells exhibit large data retention time degradation that are attributed to single interactions. The physical process leading to the degradation is discussed. The annealing behavior and leakage current order of magnitude are compatible with the properties, stability and emission rate of clusters of defects in the silicon

Using Subthreshold Heavy Ion Upset Cross Section to Calculate Proton Sensitivity

This paper reports on SEU proton calculations using the very low LET heavy ion upset cross section measurements. This approach allows for improving the prediction of the proton sensivity of devices based on heavy ion data.

Simulation of heavy-ion-induced failure modes in nMOS cells of ICs

Simulation results show leakage current induced by the impact of a single heavy ion in an nMOSFETs birds beak accountable for experimentally observed stuck bits. Parametric extrapolation shows dramatic sensitivity increase for recent local oxidation of silicon technologies.