Some Notes on Radiation

To talk about the health effects of “radiation” is much like talking about the health effects of “viruses”. If I tell you I’ve been exposed to a virus, you really can’t say anything much about my likely health outcome. Maybe I’ll die–if the virus is rabies, or H1N1. Or maybe I’ll get the sniffles, or nothing at all.

It all depends on the kind of virus I’ve been exposed to, and my susceptibility to it.

Radiation has hardly ever killed anyone, so we have very little data on its effects. Radiation deaths at Hiroshima and Nagasaki are still heavily studied. Although a tiny minority of the people who were killed in the bombings of those cities died from radiation effects, they are still the largest cohort of radiation victims in history. Modelling the radiation dose across the landscape, and identifying exactly where each individual was at the time of the bombing, was at least up until the mid-1990′s a significant ongoing project.

So we don’t know as much as we would like, empirically, and our theoretical knowledge is complicated. Radiation comes in multiple forms and has somewhat different effects depending on energy. The biological nature of radiation damage is understood in a broad sense, but the details are complex and sometimes fuzzy, possibly even quantum.

The basic physics of the ways in which different types of radiation interact with matter is pretty well understood, but again, there is a diversity of kinds and the interactions depend on energy and are frequently not simple. For example, radioactive nuclei that emit beta particles generally also emit gamma-rays, and when gamma rays interact with matter they create beta particles. Neutrons are another story entirely, as are cosmic ray muons, which are an important component of background radiation, particularly for air travellers and people who live at high altitudes.

Rather than mutter on at length about generalities, I’m going to focus on one specific instance here: potassium-40 versus iodine-131.

Potassium-40 is the most significant radioactive element in the human body. The average adult has over 100 g of potassium in their body, and 0.0118% of that is radioactive potassium-40.

It might be worth mentioning where naturally occurring radioactives come from. Unsurprisingly it is the same place virtually everything heavier than oxygen comes from: supernova, the explosive death of large stars. Our solar system and others like it are formed by condensation out of the cloud of gas and dust left over from one or more exploding stars. During a supernova explosion every isotope imaginable gets created, including radioactive ones.

Most radioactive isotopes are short-lived. Iodine-131 has a half-life of a little over a week, so any that was produced in a supernova four or five billion years ago is long gone. Potassium-40, in contrast, has a half-life of a little over a billion years, so it’s only been about four half-lives since the supernova whose remnants we are made out of exploded. Ergo, 0.5*0.5*0.5*0.5 = 6.25% of the original potassium-40 is still around, even though it continues to slowly decay.

There are three different units of radiation exposure, which adds to the already considerable confusion around the it. One unit is the bequerel (Bq) which is simply the rate of decay, the number of decays per second. Another is the gray (Gr) which is the total energy deposited in a given mass of tissue (J/kg). Finally, there is the sievert (Sv) which is the value in Gr multiplied by a “quality factor”, Q, which attempts to account for the amount of biological damage done by that form of radiation, and which I’ll talk about more in a later post. For beta and gamma particles Q is considered to be 1, so the values in Gr and the value in Sv are numerically the same, although conceptually distinct.

When comparing the effects of potassium-40 and iodine-131, there are a number of important factors to take into account:

1) decay rate
2) decay mode
3) decay energy
4) distribution in the body

The current health flap in Japan is based on a standard for infants that is stated in Bq: 100 decays per second of iodine-131 per kilogram (litre, more-or-less) of drinking water is the maximum allowed.

By contrast, naturally occurring potassium-40 has a decay rate of about 400 Bq in the average infant (figuring 4400 Bq for the average adult male and cutting the mass down by a bit more than factor of ten for an infant.)

So the rate is about four times higher for potassium-40 than the legal limit for iodine-131. What goes on?

To compare apples to sort-of-apples (we are never going to get to quite an apples-to-apples comparison, unfortunately) we have to look at decay schemes and decay energies. Iodine-131 decays via beta-decay–the conversion of a neutron to a proton via the emission of a electron and a neutrino. The decay energy is split between the electron (also called a beta-particle) and the neutrino, which is an almost massless particle that has negligible interaction with matter. Because the decay energy is split, and because of quantum, the beta-particle can be emitted with any energy between zero and the full energy of the decay.

In this case the maximum energy is just under 0.4 MeV and the average energy is just under 0.2 MeV. “MeV” stands for “million electron-volts”: an electron-volt is the energy an electron gets with accelerated by a potential of one volt. It isn’t very much: a 10 watt lightbulb uses about 1,600,000,000,000,000,000,000 electron-volts every second.

Given how tiny the energy is, it may be a little surprising that radiation like this can do any damage at all, and this is a cautionary tale as to why we need to think carefully about this stuff. The history of science tells us that our intuitions about what is “absurd” are if anything slightly worse than random chance at picking out what is likely to be true or false. After all, what seems more intuitively absurd: that the obviously stationary Earth is hurtling around the Sun, or that it is the stationary centre of all things? Or how about disease being caused by tiny animals rather than the merciless wrath and hatred of your favourite all-merciful and loving god? I know which one seemed intuitively obvious to my ancestors.

But the energy deposited by a beta particle is all deposited in a very narrow band around its track, breaking bonds between atoms in a way that the same amount of energy deposited as heat cannot.

Gamma rays, in contrast, pass through matter without interaction until they scatter off an electron, losing energy to it. The scattered electron is then indistinguishable from a beta-particle, moving quickly and losing energy as it passes through matter. This is why gammas and betas have the same Q: their terminal energy deposition process is identical.

Although iodine-131 is a beta-emitter, it also emits gamma rays, generally with an energy of 0.4 MeV. It works like this: Iodine-131 emits a beta-particle to become xenon-131, but the xenon-131 that is created is typically created in an excited nuclear state, which decays by emitting a gamma ray.

The process of electromagnetic de-excitation of nuclear states is very similar to the somewhat more familiar process of fluorescence in atomic physics, but a million times more energetic. When you snap a peppermint candy and it emits a flash of light it does so because you are breaking atomic bonds and leaving the resulting atoms in excited atomic states, which decay to their ground state by emitting visible photons. Nuclear states do exactly the same thing, but the photons–light quanta–have vastly more energy.

Thus, most nuclei that decay by beta emission are also gamma emitters, and both iodine-131 and potassium-40 fall into this category. Potassium-40 actually has more exotic decay modes: it can decay by either ordinary electron (beta) emission or positron (anti-electron) emission. In the latter case the positron will lose most of its energy and then annihilate when it runs into an atomic electron, creating two 0.5 MeV gamma rays in the process, which will eventually scatter and lose energy to the resulting beta particles.

This kind of process is sometimes called an “electron-gamma shower”, and one of the most important radiation transport codes–developed primarily at Canada’s National Research Council–is called EGS for this reason.

So for iodine-131 there is a beta particle with a peak energy of 0.4 MeV and an average energy of 0.2 MeV, and a gamma ray with an energy of 0.4 MeV.

For potassium-40 there is a beta particle with an average energy of 0.5 MeV and a peak energy of over 1 MeV, and a gamma ray of almost 1.5 MeV.

The modes of the two radiation sources are the same, but their energies are quite different. Potassium-40 deposits something like twice as much energy in the body as iodine-131 for every decay. Thus, for equal bodily loads in bequerel, potassium-40 will produce about twice the dose in grays or sieverts.

Finally, there is the critical question of distribution in the body. I have been critical of the American TSA for using bulk dosimetry to evaluate a technology that deposits almost all its energy in the skin. In the case of iodine versus potassium, iodine is of particular concern because it tends to concentrate in the thyroid, which is a radio-sensitive organ.

If we suppose that 100% of an iodine dose ends up in the thyroid, and that the thyroid has a mass of 0.01 kg in an infant with a mass of 6 kg (600x mass ratio) then a dose of iodine-131 would be 300 times worse than an equal dose of potassium-40, even given the factor of two difference in decay energies.

But… iodine-131 has a half-life of 8.3 days. That means after three months a spike of iodine-131 has dropped to less than 1/1000 of what it was originally when ingested, while an equal dose of potassium-40 would still be going at exactly the same rate, and the three months after that, and the year after that, and the decade after that…

Very high doses of iodine-131 are clearly bad things. They are correlated with a statistically significant increase in thyroid cancer, which if left untreated can be fatal, and typically kills about 5% of the people who get it even with treatment. It is a bad disease.

But… so is iodine poisoning. And so is dehydration.

This post was prompted by reports that an advisory has been issued in Tokyo to not give infants formula mixed with tap water due to iodine-131 levels of “twice the legal limit” (a slogan I’m sure Greenpeace will be screaming in its delightfully context-free way for decades to come.)

But the legal limit is 100 Bq/kg of water. Infants–who are better off with breast milk in any case–rarely drink a full kilogram of water per day, and radioactive iodine is no longer being released into the environment by any of the Fukushima reactors or spent fuel pools, as near as anyone can tell.

I hope after reading all this it is clear that the “health risk” associated with that level of iodine-131 is difficult to quantify on a theoretical basis. It is also, more significantly, impossible to measure. There is simply no evidence, nor any way of producing evidence, of the health risks because they are so small.

The health risk to infants from dehydration, on the other hand, is well known and all too easily quantified.