# How does radiation degrade mechanical parts and electronic devices?

I'm running out of places to look (lots of Googling, SE, [articles and books are too specific and never give a good overview]), and yet I am still unsure about how exactly radiation can degrade mechanical parts and electronic devices. Could someone summarise the mechanisms involved, for either of those or both?

I'm working as a mechanical+electrical designer in the Space industry, that is why I am particularly interested in that. This is purely out of curiosity though.

Here is a start, please correct and improve:

We call radiation high energy photons (X-rays & gamma rays) or high energy particles (i.e. travelling in a straight line through space):

1. Alpha particles, which are Helium atoms
2. Beta particles, which are electrons
3. Neutrons
4. Ions (I'm surprised they don't capture/release electrons rapidly)

Galactic cosmic rays can apparently be any atom, though usually protons (after all, that's what space is mostly made of).

Effects:

• High energy photons can give sufficient energy to electrons to make them leave their orbits, ionising the atom. I assume the cristal lattices of metals try to rearrange for example, making it swell and embrittle, but I don't know what the resulting ions become (I've always thought ions were only temporary, except dissolved in e.g. water)
• High energy particles certainly have the same effect by ejecting atoms on impact.
• It seems like transmutation, where high energy particles break nucleii so that the atom is transformed into an another element, is very rare.
• I assume ionisation is the main failure mechanism in electronic devices, mainly in transistors because it can create currents where there is supposed to be none and perhaps trigger a runaway (affecting doping of the junctions, thanks dmckee).

Protection (bonus):

• High energy rays can be effectively absorbed by dense materials with high atomic numbers such as lead because their many electrons increase the probability of absorbtion of the photons
• Alpha particles don't go very far; even though they carry lots of energy they are big and any material can stop them. beta particles are lighter, faster and penetrate further but still are stopped relatively easily because they are charged. Neutrons are somtething else, stopped by nucleii and therefore smaller lighter elements with small atom diameters can be packed in the same volume to stop them more effectively.

Related:

How does radiation shielding using absorbing materials work?

How would steel degrade in space

Edit: Wikipedia has a nice page covering effects on electronics here (keywords are key...): http://en.wikipedia.org/wiki/Radiation_hardening

• I have always assumed that silicon and germanium electronics are sensitive to local aberrations in the lattice structure. That is, after all, how semiconductor doping works. – dmckee --- ex-moderator kitten May 13 '15 at 0:03
• The topic, even on the most naive level, fills entire engineering textbooks, so I am surprised that an aerospace engineer doesn't have access to the literature. If you need it for your work, your workplace should have some of the basic textbooks and publications. A short search for "radiation hardness" brings up tons of material. Did you even try to do serious research? – CuriousOne May 13 '15 at 0:23
• I don't need it (we are limited in the number of components and materials we can use), this is purely out of curiosity. The litterature is not doing a good job at summarising, I would like to get an overview to be able to understand the details that the books and articles get into. – Mister Mystère May 13 '15 at 7:43
• Thanks, "Hardening" is a keyword that gave me a very nice result on Wikipedia for the electronics aspect, see updated post. – Mister Mystère May 13 '15 at 9:38
• Alpha's, beta's and neutrons are created by interactions of the primary cosmic rays, because they decay they are not in the primordial flux. In your list of particles you should add the muons which are the result of decays from cosmic interactions and actually are the main component reaching sea at a rate of one per square cm per minute.hyperphysics.phy-astr.gsu.edu/hbase/particles/muonatm.html – anna v May 20 '15 at 8:33

Ionizing radiation loses energy in matter by creating electron-ion pairs.

Suppose you have an 1 MeV charged particle stopping in a silicon crystal. The first ionization energy for free silicon atoms is about 8 eV. The ionization energy will be a little different for silicon atoms on the lattice, but not grossly so: your 1 MeV charged particle is going to turn into $10^4$-$10^5$ electron-ion pairs. This is a femtocoulomb or so of charge, which doesn't sound like much. But if the charge is deposited in a nanosecond the instantaneous current might be a microampere. Currents that large and that fast are actually relatively easy to detect --- that's how silicon radiation detectors work, after all. In an electronic circuit that is operating with a gigahertz clock with low-current field effect or CMOS transistors, a microampere integrated over a nanosecond at the right gate input might be enough charge to flip a bit from false to true, and poof! you've changed your running program code to something different that what the author wrote. The problem obviously gets worse with more frequent interactions with radiation.

A second mechanism is dislocation. If your incoming radiation strikes a nucleus rather than an electron, it can remove that nucleus from its "proper" location in the crystal lattice. Suppose your crystal lattice was a perfect crystal of silicon, the starting point for all sorts of semiconductor devices: it's not perfect any more!

The first of these is purely a concern for electronics; the second will affect the functioning of electronics first, but will affect structural integrity eventually. Another interaction to consider is ionization-induced chemical reactions. Suppose I have a material which is chemically inert in its neutral state, but reactive when ionized. Exposure to radiation will allow the "forbidden" reactions involving the ionized material to proceed. I had some coaxial cable in an experiment once which saw an unexpectedly large dose; the insulating layer between the central and outer conductors turned from a nice sturdy white plastic into sort of a gray powder, and stopped insulating.

Ionization-induced chemistry is probably the main reason by radiation exposure is bad for biological organisms. Biology involves an enormous number of carefully regulated chemical reactions; replace some neutral particle with an ion and your reaction will go quite differently.

According to the Review of Particle Physics (Section 33.7.4 of the 2014 edition) there are two main causes of radiation damage for electronic devices:

1. Bulk damage due to displacement of atoms from their lattice sites. This leads to increased leakage current, carrier trapping, and build-up of space charge that changes the required operating voltage. Displacement damage depends on the non-ionizing energy loss and the energy imparted to the recoil atoms, which can initiate a chain of subsequent displacements, i.e., damage clusters. Hence, it is critical to consider both particle type and energy.
2. Surface damage due to charge build-up in surface layers, which leads to increased surface leakage currents. In strip detectors the inter-strip isolation is affected. The effects of charge build-up are strongly dependent on the device structure and on fabrication details. Since the damage is proportional to the absorbed energy (when ionization dominates), the dose can be specified in rad (or Gray) independent of particle type.

The review is freely accessible and, although brief and technical, you can find more details there. Also, the review suggests to look at the Handbook of Radiation Effects for a detailed explanation on the matter. From what I can see from the ToC, there is also a chapter named The interaction of radiation with shielding materials that might answer your curiosity about mechanical damage.

Although there are different types of "radiation," their common effect is to transfer some/most of their energy to the material they "hit," resulting in the breaking of the atomic bonds and or structures of the material. When "enough" bonds and/or structures are broken, the material will fail. Since the electrical characteristics of electronic components are more "sensitive," they will fail electrically much sooner than mechanically.