Search

Single event latchup (SEL) in a 65 nm CMOS SRAM technology due to heavy ions is observed and device sensitivity is shown to be a strong function of lateral beam orientation, angle of incidence, and temperature. Experimental results show the importance of testing at multiple lateral beam orientations to properly characterize device sensitivity. Index Terms--65 nm, azimuthal angle, grazing angle, lateral angle, radiation effects, single event latchup.

SEUs due to combinational logic in 90 nm CMOS is analyzed at various speeds using a new design approach called the Combinational Circuit for Radiation Effects Self-Test (C-CREST). C-CREST allows the cross-section of combinational logic to be increased while minimizing propagation delay. The design was fabricated in IBM's 9SF CMOS process and underwent broadbeam testing that distinguished combinational logic errors from latch errors. Results confirm that the design is effective in testing combinational logic for SE vulnerabilities with minimum speed penalty. Index Terms--Combinational logic, DICE latch, single event, single event transient, single event upset, window of vulnerability.

The minimum laser energy required to cause sustained harmonic oscillations in a 201-stage ring oscillator varies little with increasing power supply voltage when operating in the subthreshold region. These small changes in threshold laser energy in the subthreshold region suggest that the cross-section curves as a function of power supply voltage remain relatively constant. Simulations show that the minimum pulsewidth required to generate higher-order oscillations decreases as operating voltage increases. Single-event transients are wider when circuits are operating in the subthreshold region than when operating at the nominal power supply voltages. Narrower single-event transients result for strikes on PMOS transistors as compared to those for strikes on NMOS devices in the subthreshold region; the opposite is observed when circuits are operating at the nominal power supply voltages. Index Terms--CMOS circuits, digital circuits, radiation effects, radiation effects in ICs, single-event effects, single event transients.

A well-collapse source-injection mode for SRAM SEU is demonstrated through TCAD modeling. The recovery of the SRAM's state is shown to be based upon the resistive path from the p+-sources in the SRAM to the well. Multiple cell upset patterns for direct charge collection and the well-collapse source-injection mechanisms are predicted and compared to SRAM test data. Index Terms--Heavy ion testing, multiple cell upset, single event modeling, SRAM.

Displacement-damage induced degradation in AlGaN/AlN/GaN HEMTs with polarization charge induced 2DEGs is examined using simulations and experiments. Carrier removal in the unintentionally doped AlGaN layer changes the space charge in the structure and this changes the band bending. The band bending decreases the 2DEG density, which in turn reduces the drain current in the device. The effect of the defect energy levels on the 2DEG density is also studied. The interplay between carrier removal, mobility degradation, and the charged defects is analyzed and quantified. Index Terms--Deep level traps, displacement damage, GaN, high electron mobility transistor (HEMT), polarization charge, proton radiation.

Simulations are used to characterize the single event transient current and voltage waveforms in deep submicron CMOS integrated circuits. Results indicate that the mechanism controlling the height and duration of the observed current plateau is the redistribution of the electrostatic potential in the substrate following a particle strike. Quantitative circuit and technology factors influencing the mechanism include restoring current, device sizing, and well and substrate doping. Index Terms--Field funneling, potential modulation, pulse shape, pulse width, single event, TCAD.

Experimental evidence and Monte-Carlo simulations for several technologies show that accurate SEE response predictions depend on a detailed description of the variability of radiation events (e.g., nuclear reactions), as opposed to the classical single-valued LET parameter. Rate predictions conducted with this simulation framework exhibit excellent agreement with the average observed SEU rate on NASA's MESSENGER mission to Mercury, while a prediction from the traditional IRPP method, which does not include the contribution from ion-ion reactions, falls well below the observed rate. While rate predictions depend on availability of technology information, the approach described here is sufficiently flexible that reasonably accurate results describing the response to irradiation can be obtained even in the absence of detailed information about the device geometry and fabrication process. Index Terms--Nuclear reactions, single event effects (SEE), single event effect rate.

The effects of heavy ion energy and nuclear interactions on the single-event upset (SEU) and single-event latchup (SEL) response of commercial and radiation-hardened CMOS ICs are explored. Above the threshold LET for direct ionization-induced upsets, little difference is observed in single-event upset and latchup cross sections measured using low versus high energy heavy ions. However, significant differences between low- and high-energy heavy ion test results are observed below the threshold LET for single-node direct ionization-induced upsets. The data suggest that secondary particles produced by nuclear interactions play a role in determining the SEU and SEL hardness of integrated circuits, especially at low LET. The role of nuclear interactions and implications for radiation hardness assurance and rate prediction are discussed. Index Terms--Indirect ionization, integrated circuit reliability, nuclear reactions, radiation effects, radiation hardness assurance, single-event effects, single-event latchup, single-event upset, soft errors.

A radiation-hardened-by-design phase-locked loop (PLL)--designed and fabricated in 130 nm CMOS--is shown to mitigate single-event transients (SETs). Two-photon-absorption (TPA) laser tests were used to characterize the error signatures of the PLL and to perform single-event upset (SEU) mapping of the PLL sub-components. Results show that a custom, voltage-based charge pump reduces the error response of the PLL over conventional designs by more than two orders of magnitude as measured by the number of erroneous PLL clock pulses following a single-event. Additionally, SEU mapping indicates a 99% reduction in the vulnerable area of the radiation-hardened-by-design (RHBD) charge pump over a conventional design. Furthermore, the TPA experiments highlight the importance of the voltage-controlled oscillator in the overall SET response of the PLL implementing the RHBD charge pump. Index Terms--Charge pumps, mixed analog-digital integrated circuits, phase-locked loops (PLLs), radiation effects, radiation hardening, radiation-hardened-by-design (RHBD), RF operation, single-event effects (SEEs), single-event transients (SETs).

This paper presents a combined Monte Carlo and analytic approach to the calculation of the pixel-to-pixel distribution of proton-induced damage in a HgCdTe sensor array and compares the results to measured dark current distributions after damage by 63 MeV protons. The moments of the Coulombic, nuclear elastic and nuclear inelastic damage distributions were extracted from Monte Carlo simulations and combined to form a damage distribution using the analytic techniques first described by Marshall et al. The calculations show that the high energy recoils from the nuclear inelastic reactions (calculated using the Monte Carlo code MCNPX) produce a pronounced skewing of the damage energy distribution. While the nuclear elastic component (also calculated using the MCNPX) contributes only a small fraction of the total nonionizing damage energy, its inclusion in the shape of the damage across the array is significant. The Coulombic contribution was calculated using the Monte Carlo radiative energy desposition (MRED), a Geant4 application. The comparison with the dark current distribution strongly suggests that mechanisms which are not linearly correlated with nonionizing damage produced according to collision kinematics are responsible for the observed dark current increases. This has important implications for the process of predicting the on-orbit dark current response of the HgCdTe sensor array. Index Terms--CCD, dark current distribution, GEANT4, HgCdTe, proton effects.

The interaction between a heavy ion and the overlayer materials in an integrated circuit may result in a nuclear reaction. This reaction leads to a charge generation profile that is substantially altered from the profile generated during a direct ionization event. In this work, nuclear reactions are integrated into the modeling of the SEU response of an SRAM cell using GEANT4-based simulations. The simulated transient response is compared to the response obtained using a typical heavy ion model that includes only direct ionization. Index Terms--Heavy, ion, Monte Carlo, nuclear event, SRAM, TCAD.

A Hubble space telescope wide field camera 3 e2v CCD was irradiated while operating at -84[degrees]C and the dark current studied as a function of temperature while the charge coupled device was warmed to a sequence of temperatures up to a maximum of +30[degrees]C. The device was then cooled back down to--84[degrees] and remeasured. Hot pixel populations were tracked during the warm up and cool down. Hot pixel annealing began below -40[degrees]C and the anneal process was largely completed by the time the detector reached + 20[degrees]C. There was no apparent sharp annealing temperature. Although a large fraction of the hot pixels fell below the threshold to be counted as a hot pixel, they nevertheless sustained a higher leakage rate than the remaining population. The mechanism for hot pixel annealing is not presently understood. Room temperature irradiations do not adequately characterize the hot pixel distributions for cooled applications. Index Terms--CCDs, displacement damage, hot pixels, radiation effects.

We investigate the effects of metallization layers on the radiation hardness of an epitaxial CMOS memory technology using Monte Carlo simulations. A geometrically and compositionally realistic three-layer metallization scheme is employed in detailed radiation transport simulations that include contributions from discrete [delta]-rays and nuclear reactions. The presence of high-Z plugs used to connect different metallization layers can have a significant effect on the single-event sensitivity depending on the location of the high-Z material relative to the sensitive region of the underlying device. Index Terms--Energy deposition, Geant4, ion track, metal-oxide semiconductor (MOS), metallization layers, Monte Carlo simulation, radiation effects, single-event effects, tungsten.

The energy-deposition spectra measured by the CRRES (Combined Release and Radiation Effects Satellite) pulse height analyzer (PHA) experiment are reexamined to determine the relative contributions from the different mechanisms by which large localized energy-deposition events can be generated in silicon detector volumes flown in space. Comparisons with calculations generated by the CUPID simulation model modified to include pion-production and elastic-scattering events are used to generate a 'best fit' trapped proton spectrum that is compared to the predictions of the NASA models for the CRRES orbit. Index Terms--Dosimetry, protons, radiation effects, radiation monitoring.

The complex single-event upset (SEU) signatures produced by high-speed analog-to-digital converter devices are analyzed to establish magnitude, duration, and pattern characteristics. Histograms and other descriptive summaries of event characteristics are identified and their use in performing system effects analysis is described. Index Terms--Analog-to-digital conversion, ion radiation effects, proton radiation effects.

We present a study of proton transient effects in metal-semiconductor-metal (MSM) photodetectors, which demonstrates their inherent advantage for minimizing Single Event Effects (SEEs) in proton environments. Upset mechanisms are characterized for 830 nm GaAs and 1300 nm InGaAs detectors. Only protons incident at grazing angles are likely to cause a bit errors by direct ionization. The MSM technology appears to be a more robust to single bit errors than thicker 1300 nm p-i-n diode structures which we have previously shown to be susceptible to errors from direct ionization events at all angles, and also at relatively high optical powers [1]. For a given receiver, the relative contributions of direct ionization and nuclear reaction upset mechanisms at a specific data rate and optical power are determined by the geometry of the charge collection volume of the detector. We show that state-of-the-art p-i-n detectors can also display a reduced sensitivity to direct ionization by incident protons except at grazing angles.

We present ground test and space flight data describing a single event anomaly that affects multiple bytes in a stacked DRAM module. A 12 Gbit solid state recorder containing 1,440 DRAM die experiences the anomalous events at a rate requiring testing of a large sample set of these modules.

Radiation-induced permanent degradation and single event transient effects for optocouplers are discussed in this paper. These two effects are independent to the first order and will be addressed separately. Displacement damage-induced degradation of optocoupler current transfer ratio is reviewed. New data are presented that show the importance of application specific testing and that generalized quantification of optocoupler CTR degradation can lead to incorrect predictions of actual circuit performance in a radiation environment. Data are given for various circuit loading and drive current parameters. Previous work that introduces the idea that two mechanisms exist for inducing transients on the optocoupler output is discussed. New data are presented that extends the evidence of this dual mechanism hypothesis. In this work measurements show that single event transient cross sections and transient propagation varies with circuit filtering. Finally, we discuss utilization of the optocouplers in the space environment. New data are applied to two examples: one on permanent degradation and the other on single event transient rates in high bandwidth applications.

Individual ionizing heavy ion events are shown to cause two or more adjacent memory cells to change logic states in a high density CMOS SRAM. A majority of the upsets produced by normally incident heavy ions are due to single-particle events that causes a single cell to upset. However, for grazing angles a majority of the upsets produced by heavy-ion irradiation are due to single-particle events that cause two or more cells to change logic states. Experimental evidence of a single proton-induced spallation reaction that causes two adjacent memory cells to change logic states is presented. Results from a dual volume Monte-Carlo simulation code for proton-induced single-event multiple upsets are within a factor of three of experimental data for protons at normal incidence and 70 degrees.

Multiple Bit Upsets (MBU) have been studied in a 256k CMOS static RAM irradiated at normal incidence and grazing angle. In normal incidence the sensitive areas have been identified with pulsed laser irradiation. The laser power thresholds have been determined for single to quadruple upsets in adjacent cells. Both experimental data and 3D simulations emphasize the role of delayed charge collection, by diffusion, and charge sharing between sensitive areas. Upset tracks have been recorded at grazing angle and used to determine the charge collection depth. These data revealed the existence of an LET threshold for MBU at grazing angle. As the ion LET increases different types of tracks are observed and correlated to the topological pattern in adjacent memory cells. This phenomenon is due to an unexpected charge collection mechanism, which couples adjacent sensitive areas and results in charge transfer between memory cells. The comparison with previous data on the same device indicates a strong influence of both ion energy and angle of incidence on the cross section, emphasizing the intrinsic limitation of standard characterizations with low energy ions. These results indicate that the basic assumption of a rectangular parallelepipedic volume does not take into account coupling phenomena, such as occurs in MBUs, and is no longer valid at grazing angle.

We present data which show that Single Event Upset (SEU) cross section varies linearly with frequency for most devices tested. We show that the SEU cross section can increase dramatically away from a linear relationship when the test setup is not optimized, or when testing near the maximum operating frequency. We also observe non-linear behavior in some complex circuit topologies. Knowledge of the relationship between SEU cross section and frequency is important for estimates of on-orbit SEU rates.

A new simulation code, the Clemson Omnidirectional Spallation Model for Interaction in Circuits (COSMIC), is described and its predictions agree with SEU data from four devices flown as part of the microelectronics package experiment on the CRRES satellite. The code uses CUPID for determining the energy depositions in the sensitive volumes; it allows proton exposures with arbitrary angles of incidence including random omnidirectional exposure; and the user specifies the thickness of shielding on six sides of the sensitive volume. COSMIC is used as part of an algorithm developed to predict the rate proton induced single event upsets occur in the space radiation environment given by AP-8. In testing the algorithm, the position coordinates are taken from the satellite's ephemeris data, but calculations based on position coordinates from orbital codes were also in agreement with the measured values.

An instrument is described which is designed to characterize the complex radiation environments inside spacecraft and airplanes in terms of the risk of SEEs in the present and planned microelectronic systems and in terms of the risk to flight crews and passengers.

Simulations show that ignoring the angular dependence of proton SEU cross sections produces errors in predictions of SEU rates in space. Moreover, they suggest that devices with thin sensitive volumes may upset to protons at grazing incidence despite high threshold LET values (>80 MeV [cm.sup.2]) at normal incidence. Incorporating angular effects in space predictions requires accurate knowledge of the dimensions of the sensitive volume associated with the SEU-sensitive junction, especially the thickness. A method is proposed for using proton SEU measurements at different angles and energies combined with simulations to determine the thickness of the sensitive volume and to test the reliability of the predictions.

Charge collection exhibited by p-n junctions, which have at least one small dimension, deviates from the geometric assumptions commonly used in SEU testing. The amount of charge collected did not increase with the secant of the angle of incidence. The number of events under the peak in the charge collection spectrum did not decrease as the cosine of the angle of incidence. Both the position of the peak and the number of events under the peak measured at a given angle of incidence depended upon which symmetry axis of the device was chosen to be the axis of rotation.

The analysis of the electrical stress-induced recoverable degradation in InAs/AlSb high-electron mobility transistors (HEMTs), which is attributed to metastable defects in the AlSb layer generated by the injected holes, is presented. The analysis of candidate defects has shown that substantial and interstitial oxygen are responsible for the degradation and reducing the oxygen contamination during device fabrication is vital to improve the reliability of InAs/AlSb HEMTs.

Petersen has introduced a one parameter characterization of a device by the Figure Of Merit (FOM). It was claimed that this parameter was sufficient to estimate the SEU rate in almost all orbits. The present paper presents an analytic study of the FOM concept and compares the FOM model with other empirical models. It is found that indeed the FOM parameter gives, in most cases, a good agreement with the rates found using the full SEU cross section plots of the devices. The agreement is poorer in cases where a high portion of the proton flux comes from low energy protons and for very SEU-hard devices. This is demonstrated for certain devices (FPGAs) where the FOM predicted by proton may be smaller by an order of magnitude than the FOM from heavy ions.