Trends in radiation-hardened electronics testing shape future designsStory
June 13, 2013
Recent trends in electronics testing for space missions are focused on rigorous testing with attention to the complexities of space weather and the impact on the component and system design.
Historical testing, often based on military standards, has been increasingly replaced with a “test as you fly” rigor. Historical radiation testing has its basis in military standards where high dose rates and neutron fluence from a weapons environment were the prime concerns. These historical methods do not test for many of the important, emergent space requirements. Thus, the test standards for components and systems, although still used today, are being replaced or supplemented in many cases. The recent emphasis on the effects of Enhanced Low Dose Radiation Sensitivity (ELDRS), Single Event Effects (SEEs), and proton belts, which contribute to ionization and displacement damage simultaneously, are examples of such changes.
Mission-specific requirements are becoming increasingly common. The environments for Earth-orbiting satellites versus interplanetary missions to the gas giants are very different, with the latter introducing requirements for operation in extreme cold and extremely high ionizing dosage.
Consequently, this shift in understanding and approach to radiation hardness assurance has changed the way the space electronics industry tests, and therefore designs, components and systems. The following examines space radiation effects and how historical test methods fit into emerging trends.
Space radiation effects: A closer look
The space environment for electronics is complex, with a broad range of particle types and energies, as well as electromagnetic radiation. The effects of these many sources of radiation on electronic devices depend on the component type and technology. They are often dependent on the rate and energy of the radiation source. The space environment is dominated by proton and electron belts with particle energies ranging from 1 MeV to 100 MeV for electrons and up to 400 MeV for protons arranged in belts as shown in Figure 1. The total ionizing dose rate in Earth orbit does not exceed 10 mrad(Si) per second. Single events with high-energy particles are occurring with a high frequency for low energies and low frequency for higher energies.
Figure 1: Top image: proton belt, scale in earth radii. Bottom image: electron belts, scale in earth radii (courtesy of NASA).
(Click graphic to zoom)
The resultant radiation effects on components are often broadly categorized as ionization, displacement damage, and single event effects:
Ionization radiation effects are induced by electromagnetic radiation greater than electromagnet radiation ultraviolet and beyond: by alpha particles (helium nuclei) or by beta particles (electrons). The effect of ionization on semiconductors is the addition or removal of electrons from atoms because of the columbic interaction with the radiation source. Ionizing radiation is measured by the Total Ionizing Dosage (TID) in Rad or Gray units. TID damage is generally reversible with time and temperature as recombination occurs with drift and diffusion. As will be discussed later and has only recently been understood, the rate at which ionization occurs can affect the results for some component types.
Displacement damage is caused by energetic protons and neutrons impacting the crystal lattice of the semiconductor material. Vacancies are created in the lattice of silicon resulting in changes in gain and leakage. The resultant radiation damage is permanent, affecting components like bipolar devices and opto-isolators in particular. The effect on the semiconductor can be very different for the two particles and depends on the particle energy. Protons, because of their charge, will cause ionization in addition to displacement damage. Neutrons will only contribute to displacement damage. In the space environment, protons are the prime source of radiation effects. Neutrons can be generated as a result of protons interacting with materials in the spacecraft.
Single event effects
Single event effects result from highly energetic particles, either protons or cosmic rays. Cosmic rays are most frequently protons, less frequently alpha particles and heavy nuclear ions. These highly energetic ions leave tracks of electron-hole pairs in the semiconductor material or in the dielectric. The unit used to measure the particle energy and impact on the material is Linear Energy Transfer (LET). If the charge created reaches a critical level, it can affect the semiconductor device with soft errors or state changes in memory or computing devices. In analog devices transients can be induced. Depending on the depth of penetration in multilayer devices, it can result in a latchup and high current draw. The ionic track in dielectrics can result in a conductive channel and current flow in MOS devices and can be destructive.
Historical test methods versus emerging trends
As mentioned, historical test methods for the effect of radiation on electronic devices have often failed to simulate the space environment for Earth-orbiting missions.
TID testing to ensure component survival has historically been performed at very high dose rates, typically greater than 50 rad(Si) per second. At these rates, the TID that would require 10 years in orbit can be accomplished in less than an hour. The dose rate in orbit will not exceed 10 mrad(Si) per second and will be orders of magnitude lower. The high dose rate test is often followed by an anneal period of a day to a week in an effort to account for the low dose rate in orbit. The test method most often used is detailed in military standard MIL-STD-883, Test Method 1019. However, studies show that there can be serious radiation effects on bipolar devices that cannot be predicted based on high dose rate testing. With the trend toward reduced supply voltages for digital and analog components and the reduction in device geometries, these effects have become more pronounced.
ELDRS refers to the low dose rate effect. Requirements for ELDRS tests have been added to Test Method 1019 in the most recent revisions in an effort to address this requirement. Unfortunately, testing at low dose rates to achieve the mission TID objectives can take six months to a year or longer, creating a real problem for component design and mission planning. Efforts have been made to develop methods to accelerate the effect by using elevated temperatures and slightly higher dose rates. Nearly all customers and agencies now require that the potential for ELDRS be addressed.
Test methods like 1019 include requirements for displacement damage tests using neutrons. Neutron testing is important in a weapons environment and these test methods were developed as military specifications. Increasingly, customers are requesting proton displacement testing at multiple particle energies. The proton sources complicate radiation testing because inevitably, there is total dose accumulation along with the displacement damage and effects cannot be easily separated. In addition, there can be an interaction with the metal packaging or shielding, which produces a greater TID, particularly with lower energy protons. Current standards often require testing for TID and displacement damage as separate tests and do not combine the effects on the same electronic component. Recent product testing performed by Crane Aerospace & Electronics used a combination of proton energies on packaged electronics to achieve a more realistic simulation of an orbital environment.
Single event testing has historically been performed at low particle energies often with low mass particles, in part resulting from facility availability and cost. The space environment includes a broad range of energies and particle mass. Higher mass and high energies result in the deposition of energy levels deeper in the device. In space, the particle impact can occur at any angle with respect to the device features. Current standards such as ESCC25100 and JESD57 do not adequately address or require multi-angle testing or testing with a variety of particle species and energies. The increasing complexity and density of modern digital devices have increased the sensitivity of these components to single event effects. The lack of effective standards has resulted in the development of many internal test methods. Crane Aerospace & Electronics, in union with NASA, performs SEE testing using a range of particle species and with LET ranges that ensure that data sheet specifications are met. This kind of testing is costly and can only be performed at a select group of facilities; however, it does more realistically reflect the conditions of a space environment.
Industry efforts unfold to keep pace with trends
In summary, there is a greater understanding of the limitations of traditional test methods for electronics in simulating the complex nature of the space environment. A notably significant trend in radiation hardness design and assurance is a greater sophistication in the approach to more accurately simulate the space environment. This trend is resulting in a considerable effect on component design and development as well as system-level tests.
Customers increasingly require testing tailored to specific mission requirements and are less reliant on traditional test methods. The “test as you fly” philosophy is being driven by the radiation hardness assurance groups at major agencies like NASA and space hardware suppliers. Significant experimentation is underway to effectively accelerate ELDRS tests. New methods are in development using laser simulation to induce single event effects. More sophisticated proton fluence tests utilizing multiple particle energies and updates to software modeling for the effect have been published. Also, new versions of standards, for instance, MIL-STD-883, TM 1019, MIL PRF 38534, Appendix G, and ESCC 25100 are in the works, and new standards that address single event tests and proton tests will emerge from JEDEC and NASA to better address limitations.
Jay Kuehny is a Principal Engineer with Crane Aerospace & Electronics at the Redmond, WA location, where he also serves in a Radiation Hardness Assurance role. Jay has more than 20 years of experience in the design and development of Interpoint power converter products for space applications, which have been used on missions including Mars Science Lab Curiosity, Hubble, Mars Rover, and Cassini in addition to most military and space satellites. Contact him at [email protected].
Crane Aerospace & Electronics 425-895-4051 www.craneae.com