SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Sensitivity of a variety of candidate spacecraft electronics to proton and heavy ion induced single event effects is presented. Devices tested include digital, linear, and hybrid devices.
The LTC6103HMS8#PBF (henceforth abbreviated as LTC6103) current sense amplifier from Linear Technology was tested for both destructive and non-destructive single-event effects (SEE) using the heavy-ion cyclotron accelerator beam at Lawrence Berkeley National Laboratory (LBNL) Berkeley Accelerator Effects (BASE) facility. During testing, the input voltages and output currents were monitored to detect single event latch-up (SEL) and single-event transients (SETs).
The XQR18V04 was evaluated for single event upset rates using proton and heavy ions. The PROM was demonstrated to be immune to latch-up, as well as to static upset in the flash memory cells, to an LET > 125 MeV/mg/cmz (effective). The PROM was also tested in a dynamic mode, which revealed three distinct error modes: Read Bit Errors, Address Errors, and a Single Event Functional Interrupt (SEW which affected the data output drivers. Saturation cross-sections, and onset thresholds, for these errorinodes were measured at the heavy ion facility at Texas A&M University, and the proton facility at UC Davis. Additional testing was performed at UC Davis and the Cobalt 60 source at McClellan Air Force Base to examine the effect to TID life as a function of power biasing. The PROM demonstrated a 100% improvement in total TID life with an 84% percent decrease in device usage.
The AMD Radeon Embedded e9170 Graphics Processing Unit (GPU), notably the e9173 Peripheral Component Interconnect Express (PCIE) variant, is of interest to Artemis generation programs with requirements for graphics rendering, compute, artificial intelligence (Ai) with a constraints-requiring piece-part procurement and power consumption of less than 50W. In addition to collecting heavy ion data on this device, a secondary purpose of this test campaign was to validate video capture hardware and software workflows used with GPU, microprocessor and system-on-chip device testing. Five (5) test patterns from the NEPP Processor Enclave (NPE) test suite were used with the e9173. The test patterns covered the operating system’s (OS) idle contribution towards the cross section, matrix math using tensorflow-rocm, two artificial intelligence models developed at NASA GSFC, and an industry standard GPU benchmarking application called Mesa GLXGears.
This paper reports heavy ion, proton, and laser induced single event effects results for a variety of microelectronic devices targeted for possible use in NASA spacecrafts. The compendium covers devices tested within the years of 2010 through 2012.
This study is to characterize the destructive single-event latchup (SEL) susceptibility of the LTC1604AIG#90117 (LTC1604) analog/digital converter (ADC). The device was monitored for SEL current events on the power supplies during exposure to a heavy ion beam at Lawrence Berkeley National Laboratory (LBNL) 88-inch Cyclotron. Testing was performed on September 11, 2021.
In this study the 14 nm Intel Broadwell 5th generation core series 5005U-i3 and 5200U-i5 was mounted on Dell Inspiron laptops, MSI Cubi and Gigabyte Brix barebones and tested with Windows 8 and CentOS7 at idle. Heavy-ion-induced hard- and catastrophic failures do not appear to be related to the Intel 14nm Tri-Gate FinFET process. They originate from a small (9 m 140 m) area on the 32nm planar PCH die (not the CPU) as initially speculated. The hard failures seem to be due to a SEE but the exact physical mechanism has yet to be identified. Some possibilities include latch-ups, charge ion trapping or implantation, ion channels, or a combination of those (in biased conditions). The mechanism of the catastrophic failures seems related to the presence of electric power (1.05V core voltage). The 1064 nm laser mimics ionization radiation and induces soft- and hard failures as a direct result of electron-hole pair production, not heat. The 14nm FinFET processes continue to look promising for space radiation environments.
The key objectives of this study are to investigate, both computationally and experimentally, which forms, compositions, and layerings of hydrogen, boron, and nitrogen containing materials will offer the greatest shielding in the most structurally robust combination against galactic cosmic radiation (GCR), secondary neutrons, and solar energetic particles (SEP). The objectives and expected significance of this research are to develop a space radiation shielding materials system that has high efficacy for shielding radiation and that also has high strength for load bearing primary structures. Such a materials system does not yet exist. The boron nitride nanotube (BNNT) can theoretically be processed into structural BNNT and used for load bearing structures. Furthermore, the BNNT can be incorporated into high hydrogen polymers and the combination used as matrix reinforcement for structural composites. BNNT's molecular structure is attractive for hydrogen storage and hydrogenation. There are two methods or techniques for introducing hydrogen into BNNT: (1) hydrogen storage in BNNT, and (2) hydrogenation of BNNT (hydrogenated BNNT). In the hydrogen storage method, nanotubes are favored to store hydrogen over particles and sheets because they have much larger surface areas and higher hydrogen binding energy. The carbon nanotube (CNT) and BNNT have been studied as potentially outstanding hydrogen storage materials since 1997. Our study of hydrogen storage in BNNT - as a function of temperature, pressure, and hydrogen gas concentration - will be performed with a hydrogen storage chamber equipped with a hydrogen generator. The second method of introducing hydrogen into BNNT is hydrogenation of BNNT, where hydrogen is covalently bonded onto boron, nitrogen, or both. Hydrogenation of BN and BNNT has been studied theoretically. Hyper-hydrogenated BNNT has been theoretically predicted with hydrogen coverage up to 100% of the individual atoms. This is a higher hydrogen content than possible with hydrogen storage; however, a systematic experimental hydrogenation study has not been reported. A combination of the two approaches may be explored to provide yet higher hydrogen content. The hydrogen containing BNNT produced in our study will be characterized for hydrogen content and thermal stability in simulated space service environments. These new materials systems will be tested for their radiation shielding effectiveness against high energy protons and high energy heavy ions at the HIMAC facility in Japan, or a comparable facility. These high energy particles simulate exposure to SEP and GCR environments. They will also be tested in the LaRC Neutron Exposure Laboratory for their neutron shielding effectiveness, an attribute that determines their capability to shield against the secondary neutrons found inside structures and on lunar and planetary surfaces. The potential significance is to produce a radiation protection enabling technology for future exploration missions. Crew on deep space human exploration missions greater than approximately 90 days cannot remain below current crew Permissible Exposure Limits without shielding and/or biological countermeasures. The intent of this research is to bring the Agency closer to extending space missions beyond the 90-day limit, with 1 year as a long-term goal. We are advocating a systems solution with a structural materials component. Our intent is to develop the best materials system for that materials component. In this Phase I study, we have shown, computationally, that hydrogen containing BNNT is effective for shielding against GCR, SEP, and neutrons over a wide range of energies. This is why we are focusing on hydrogen containing BNNT as an innovative advanced concept. In our future work, we plan to demonstrate, experimentally, that hydrogen, boron, and nitrogen based materials can provide mechanically strong, thermally stable, structural materials with effective radiation shielding against GCR, SEP, and neutrons.
We will present the first ever single-event effects testing results on a 22 nm fully-depleted silicon-on-insulator test chip. The 128 MB SRAMs were irradiated with heavy ions and the results are compared to previous technology generations.
The principal research tasks of this investigation are: (1) specification of the energetic (keV to MeV) ion environments upstream of the four Galilean satellites and (2) data analysis and numerical modeling of observed ion interactions with the satellites. Differential flux spectra are being compiled for the most abundant ions (protons, oxygen, and sulfur) from measurements at 20 keV to 100 MeV total energy by the Energetic Particle Detector (EPD) experiment and at higher ion energies by the Heavy Ion Counter (HIC) experiment. Runge-Kutta and other numerical techniques are used to propagate test particles sampled from the measured upstream spectra to the satellite surface or spacecraft through the local magnetic and corotational electric field environment of each satellite. Modeling of spatial variations in directional flux anisotropies measured during each close flyby provides limits on atomic charge states for heavy (O, S) magnetospheric ions and on internal or induced magnetic fields of the satellites. Validation of models for magnetic and electric field configurations then allows computation of rates for ion implantation, sputtering, and energy deposition into the satellite surfaces for further modeling of observable chemical changes induced by irradiation. Our ongoing work on production of oxidants and other secondary species by ice irradiation on Europa's surface has significant applications, already acknowledged in current literature, to astrobiological evolution. Finally, the work will improve understanding of energetic ion sources and sinks at the satellite orbits for improved modeling of magnetospheric transport processes. The scope of the research effort mainly includes data from the primary Galileo mission (1995-1997) but may also include some later data where directly relevant (e.g., comparison of J0 and I27 data for Io) to the primary mission objectives. Funding for this contract also includes partial support for our related education and public outreach activity, 'Interaction of Magnetospheric Particles Applied to Classroom Teaching (IMPACT).'
Ionizing radiation need not trigger unintended self-testing. Protocols installed in certain microcircuits permitting functional self-testing do not consider possibility of inadvertent triggering by protons or energetic heavy ions in space or satellite orbit. Recommended devices be rendered immune to so-called single event effects of energetic ions in cosmic rays, flares, or trapped-charge belts surrounding Earth and some other planets. Proposed pair of stimuli be applied to device in specified manner to start self-test. Protocol can require two or more stimuli at single node, or pair of stimuli at two nodes, or some more complicated combination.
Spaceflight system electronic devices must survive a wide range of radiation environments with various particle types including energetic protons, electrons, gamma rays, x-rays, and heavy ions. High-energy charged particles such as heavy ions can pass straight through a semiconductor material and interact with a charge-sensitive region, generating a significant amount of charge (electron-hole pairs) along their tracks. These excess charges can damage the device, and the response can range from temporary perturbations to permanent changes in the state or performance. These phenomena are called single event effects (SEE). Before application in flight systems, electronic parts need to be qualified and tested for performance and radiation sensitivity. Typically, their susceptibility to SEE is tested by exposure to an ion beam from a particle accelerator. At such facilities, the device under test (DUT) is irradiated with large beams so there is no fine resolution to investigate particular regions of sensitivity on the parts. While it is the most reliable approach for radiation qualification, these evaluations are time consuming and costly. There is always a need for new cost-efficient strategies to complement accelerator testing: pulsed lasers provide such a solution. Pulsed laser light can be utilized to simulate heavy ion effects with the advantage of being able to localize the sensitive region of an integrated circuit. Generally, a focused laser beam of approximately picosecond pulse duration is used to generate carrier density in the semiconductor device. During irradiation, the laser pulse is absorbed by the electronic medium with a wavelength selected accordingly by the user, and the laser energy can ionize and simulate SEE as would occur in space. With a tightly focused near infrared (NIR) laser beam, the beam waist of about a micrometer can be achieved, and additional scanning techniques are able to yield submicron resolution. This feature allows mapping of all of the sensitive regions of the studied device with fine resolution, unlike heavy ion experiments. The problematic regions can be precisely identified, and it provides a considerable amount of information about the circuit. In addition, the system allows flexibility for testing the device in different configurations in situ.
The purpose of these tests was to characterize the susceptibility of the LSK389 and LSK489 Dual N-Channel JFET from Linear Systems to Single-Event Latchup due to heavy ions and to Total Ionizing Dose (TID) degradation at -65 °C and intermediate dose rate. The test was carried out to assess suitability of the amplifiers for the Ocean Color instrument (OCI) on the Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) spacecraft.
Traditionally, accelerator SEE testing is accomplished by removing the tops of packages so that the IC chips are accessible to heavy ions. However, ICs in some advanced packages cannot be de-lidded so a different approach is used that involves grinding and/or chemically etching away part of the package and the chip from the back side. The parts are then tested from the back side with ions having sufficient range to reach the sensitive volume. More recently, the entire silicon substrate in an SOI/SRAM was removed, making it possible to use low-energy ions with shorter ranges. Where removal of part of the package is not possible, facilities at Michigan State, NASA Space Radiation Laboratory, GANIL (France) and GSI (Germany) offer high-energy heavy ions with long ranges so that the ions can reach the devices' sensitive volumes without much change in the LET. Unfortunately, a run will typically involve only one ion species having a single energy and LET due to the long time it takes to tune a new energy. The Variable Depth Bragg Peak (VDBP) method is similar to the above method in that it involves the use of high-energy heavy ions that are able to pass through the packaging material and reach the device, obviating the need to remove the package. However, the method provides a broad range of LETs from a single ion by inserting degraders in the beam that modify the ion energy and, therefore, the LET. The crux of the method involves establishing a fiduciary point for degrader thickness, i.e., where the Bragg peak is located precisely at the sensitive volume in the device, for which the measured SEU cross-section and the ion LET are both also maxima and can be calculated using a Monte-Carlo program, TRIM. Once the fiduciary point has been established, calibrated high density polyethylene (HDPE) degraders are inserted into or removed from the beam to vary the ion LET at the device in a known manner. After each change of degrader thickness, the SEU cross-section is measured and the corresponding LET calculated from the change in degrader thickness. That information is used to generate a plot of cross-section as a function of ion LET. The advantages of this approach are that the part does not have to be de-lidded and a broad range of LETs is available from a single heavy ion without having to go to non-normal angles of incidence to change the "effective" LET. As we will show, it is possible to obtain an entire curve of cross-section versus LET using just two or three ions. Fig. 1 shows curves of cross-section vs LET for a Freescale 4 Mbit SOI/SRAM measured at the 88" Cyclotron at Berkeley and at NSRL. The open symbols are the data obtained from Berkeley for top-side and back-side irradiation. The solid data points are for the data obtained at NSRL using a device for which the package was intact. The data are for Iron and Gold and cover a range of LETs from 4 MeV.cm2/mg to 84 MeV.cm2/mg. The agreement between the data obtained from Berkeley and from NSRL is excellent, demonstrating that the VDBP method is capable of providing accurate values of cross-section versus LET, at least for the 4 Mbit SRAM. Details of the technique will be included in the final presentation.
Pulse Width Modulator (PWM) Controllers are the heart of switching power supply systems in development today. The PWMs considered here have the same integration advantages as many other controllers but it also includes the interface drivers for the follow-on power Field Effect Transistors (FET). Previous work on these types of devices looked into the required test methodologies [ 11 and the impact of radiation on the soft start and shutdown circuits of typically incorporated in the technology [2]. Taking advantage of this previous work this study was undertaken to determine the single event destructive and transient susceptibility of the Linfinity SG1525A Pulse Width Modulator Controller. The device was monitored for transient interruptions in the output signals and for destructive events induced by exposing it to a heavy ion beam at the Texas A&M University Cyclotron Single Event Effects Test Facility. After exposing these devices to the beam, a new upset mode has been identified that can lead to catastrophic power supply system failure if this event would occur while drive power FETs off the two device outputs. The devices and the test methods used will be described first. This will be followed by a brief description of the data collected to date (not all data can be presented with the length constraints of the summary) and a summary of the key results.
Rolling contact fatigue (RCF) test specimens of heat treated 440C bearing steel were chromium ion plated in thicknesses from 0.1 to 8.0 micron and tested in RCF tester using 700 ksi maximum Hertzian stress. Heavy coatings, greater than about 5 micron in thickness, peeled off or spalled readily, whereas thin coatings, less than 3 micron thick, were tenacious and did not come off. Furthermore, significant improvement in RCF life was obtained with thin chromium ion plated test specimens. The average increase in B10 life was 75% compared with unplated 440C. These preliminary results indicate that ion plating is a promising way to improve bearing life.
Arrays of fibers made of scintillating material were used as position-sensitive detectors or hodoscopes for beam-finding at ion accelerators. Experiments were made with alpha's from an Am241 source incident upon one end of the fiber, the other end being viewed with a photomultiplier tube. The scintillation light was not detected in any of the fibers tested beyond about 5 cm. The effective useful lengths for detection of relativistic heavy ions were given.
Heavy ion single-event measurements on a variety of high density commercial NAND flash memories are reported. Three single event effect (SEE) phenomena are examined: single effect upsets (SEUs), single event functional interrupts (SEFIs), and catastrophic loss of ability to erase and program the device. Also, radiation results of multilevel flash technology are compared with results from single-level flash technology. These commercial high density memories appear to be much less susceptible to SEE and have better TID response compared to older generations of flash memories. The single-level devices are less sensitive to SEUs than multi-level devices. The charge pump in this study survived up to 600 krads. A new high current phenomenon in high density NAND flash memories is also discussed.