Radiation environment

'Radiation' in the spacecraft environment context generally refers to subatomic particles in space. Of course, the Sun and other astrophysical sources yield electromagnetic radiation (hard UV, X-rays and gamma rays) that are somewhat damaging to materials and living things, but these effects are generally small. In this chapter we discuss briefly the sources of energetic particles and their effects on spacecraft systems (Trainor, 1994); effects on living things are discussed in Section 14.3

Note that because the missions of entry probes and landers tend to be short, and the radiation environment at or near a planetary surface is more benign than in orbit, the radiation hazard is generally not as significant a concern as it is for orbiters. Landers on airless bodies (the Moon, Mercury, and especially Europa) may be exceptions, due to secondary radiation from the surface. However, all landers will need a radiation tolerance in that they spend time, perhaps many years, in the space environment.

There are four principal sources of radiation that must be considered. First is any radiation source carried by the spacecraft, such as a radioisotope thermoelectric generator (RTG), radioisotope heaters or sources associated with instruments such as X-ray fluorescence spectrometers. A characteristic of RTGs is their neutron flux.

A second source is galactic cosmic rays (GCRs). These are high-energy particles, usually nuclei of high atomic number ('heavy-Z' or 'high-Z' particles) from astrophysical sources. These are damaging, both directly, and indirectly, in that they may produce a shower of secondary particles and quanta by a number of methods. Heavy particles (protons, nuclei), shatter nuclei into lighter particles that in turn generate cascades of short-lived radiations by collisions and pair production from Bremsstrahlung X-rays. Energetic electrons generate X-rays when striking shielding, via the Bremsstrahlung process, and generate no particulate radiations.

Particles from the Sun form another population. These are usually less energetic, of lower atomic mass, but far higher in number. The flux of solar particles can be strongly enhanced during high solar activity (flares, coronal mass ejections etc.) and deleterious effects on Mars-orbiting spacecraft have been noted.

Usually the strongest sources of concern are the particles trapped in a planet's magnetic field. This is particularly the case for the planet Jupiter (and a Jupiter flyby may be the dominant radiation dose for an outer-solar-system mission beyond Jupiter itself). In general the field concentrates the particles in toroidal 'radiation belts', and thus the orbital design of a mission around a magnetic planet must be done carefully to minimize the dose. Around Jupiter, the moons Io and Europa are immersed in these belts, and thus radiation hardness is essential: on Europa's surface it may be advantageous to bury a lander to gain some shielding effect from the ground.

On Earth-orbiting satellites, trapped particles in the Earth's magnetic field are responsible for the bulk of radiation problems. They tend to occur predominantly in the auroral ovals (i.e. latitude belts approx 60°-70° from the equator, where magnetic field lines funnel in towards the Earth's poles) and in the South Atlantic Anomaly (SAA). This region, roughly over Brazil, is one where the Earth's net field is rather weak, and trapped particles penetrate to lower altitudes leading to increased interaction with low-orbiting satellites.

Modelling the radiation effects on components is a challenging task. Not only are various components susceptible in varying degrees to the different sources, but the effects will depend in a complex manner on the mass distribution and thus the shielding effects around the relevant component. Optimum shielding materials depend on the expected radiation source: for example, tantalum is particularly effective at shielding against stray neutrons from RTGs. And more shielding (usually expressed as an equivalent thickness of aluminium) is not necessarily better, in that GCRs often produce even more damaging Bremsstrahlung upon striking the shield. While shielding may reduce the total dose in a radiation belt, during a long cruise in deep space a modest amount of shielding may in fact increase the radiation damage. Various simulation codes are available to model these effects.

The radiation hazard is in general worse in orbit than on a planetary surface, where the atmosphere can shield a large fraction of the incoming particles. Titan and Venus are particularly benign in this regard; Mars less so. Asteroids and comets may endure a comparable radiation flux to that received en route in heliocentric orbit.

Radiation doses are usually expressed in units of rads: (this unit prevails in parallel with the corresponding SI unit, the gray: 1 rad = 0.01 gray; 1 gray corresponds to 1 joule absorbed per kg. For comparison, a prompt dose of a few hundred rad is typically fatal to humans). The Galileo spacecraft was designed to endure a dose of around 150 kilorads. Around Europa, in, but not in the worst part of, Jupiter's radiation belts, a spacecraft would endure 4 megarads in one month. Note that the energy of an individual particle is usually expressed in electron volts (1 eV ~ 1.6 X 10~19J).

Radiation damage usually manifests itself in effects on semiconductor devices, although very high doses can render optical components opaque or degrade the strength of materials. The main 'total dose effect' in electronic components is a steady increase in the gate voltage or leakage current. Ultimately, these parameters may exceed the levels at which the circuit will function as intended. A similar effect is seen in certain detectors like CCDs, whose dark current may increase ('hot pixels'). Some total dose effects can be at least partly cured by 'annealing', running at a high temperature for a short time.

In addition to steady total dose effects, there are 'sudden death' radiation damage mechanisms. One of these is the 'single event upset' (SEU), wherein the passage of a particle through a digital component alters the state of that component. A bit, most typically in a computer memory, is 'flipped', from '0' to '1' or vice versa. Where that bit is simply some data, such as a single pixel in an image, such a change is not usually catastrophic. However, if the bit is in a computer instruction, the effects may be profound and impossible to predict.

A principal protection against SEUs is to have rad-hard memory and processors. These critical functions are made less vulnerable to SEUs by, for example, the use of alternative substrates (e.g. silicon-on-sapphire) and by the use of larger gates. The energy required to flip a bit will depend on the operating voltage and the capacitance of the memory cell-more modern, high-density memories use lower voltages and smaller cells and can thus, in fact, be more vulnerable. A second approach is to use coded memory, whereby two distinct words differ by more than one bit-change. Thus, an inconsistent single bit can indicate that a memory cell has been flipped, and the incorrect bit identified and corrected by a software process or dedicated circuitry.

A final damage mechanism is not reversible, but is preventable. This is 'latch-up'. In this mechanism, the passage of a charged particle through a semiconductor creates a parasitic transistor. A large current can flow if the device is powered up, and the heating produced by the current will destroy the device. Latch-up protection consists of fast current-sensing logic that determines whether a latch-up has occurred, and if so, shuts the circuit down before heat has built up to damaging levels.

0 0

Post a comment