Over the weekend, I wrote two diaries on the isotopes of concern to scientists and the public in the event of a possible release of radioactive material from a nuclear power plant (part 1, on iodine-131; part 2, on cesium-137 and strontium-90).
Now, I want to turn to those various units of radiation you've been hearing about: rads, rems, and so on.
As I mentioned in part 1, nuclear fission occurs when the nucleus of an atom is bombarded with other particles; in a nuclear reactor, those particles are neutrons.
When the fission occurs, at least some of the resulting smaller nuclei are themselves radioactive, and capable of decaying spontaneously; as those two earlier diaries noted, uranium, with a mass of 238, produces atoms such as iodine-131 and strontium-90. While there are several ways this can happen, the key thing to note is that radioactive decay does not produce these large particles. Instead, it generally produces one of three other types of radiation:
1. Alpha particles, which are bits of the nucleus containing two protons and two neutrons (in other words, a helium nucleus).
2. Beta particles, which are either negatively-charged electrons or their positively-charged antiparticles, positrons.
3. Gamma radiation, which are photons that carry energy. Of the three forms of radiation, gamma rays are the ones that carry the most energy.
The most basic way of measuring radiation is to simply measure the number of particles released; in other words, to count the number of atoms that disintegrate. There are two units used for this: the becquerel, which is very simply one disintegrating atom per second; and the curie (named after Marie and Pierre Curie), which is about 3.7 x 10^10 (that is, 37 billion) disintegrations per second. [If you're wondering where that number comes from, it's approximately the number of disintegrations per second in one gram of radium-226, an isotope discovered by the Curies.]
Of course, from a practical standpoint of measuring exposure, this isn't particularly useful; as you may have heard already, not all radiation is created the same. What's more important is the total amount of energy carried by that radiation, and how concentrated it is. [Here's an analogy: saying that an acre of land contains a "ton of trash" could mean that you have 2,000 pounds of various debris spread about, or one junked car right in the middle of an otherwise clean field.]
Without going into the technical details, we can determine the basic methods by which radioactive isotopes decay, and the exact amounts of energy that are associated with those methods of decay. [If people really want to know the gory details, I can present it separately.] We can then figure out the amount of energy that human tissue would be exposed to, and based on that calculate the radiation absorbed dose, or rad level. [For the record, there's another unit, the gray that corresponds to 100 rads, or 1 joule per kilogram of tissue.]
One rad corresponds to an absorbed dose of 0.01 joules per kilogram of tissue. To give you an idea of how "small" this energy is, a Big Mac contains about 2 million joules of energy. The reason why these relatively small amounts of energy are so dangerous is that it's composed of a small number of relatively high-energy particles, rather than a large number of low-energy particles. Those high-energy particles basically act like little bullets, damaging or killing cells (which leads to radiation sickness) or, even worse in the long run, creating mutations in DNA (which leads to increased risk of cancer).
But, again, not all radiation affects the body equally.
Gamma radiation and beta particles are, per unit of energy, actually less dangerous than alpha particles, because they are so small they might actually pass through cells without affecting them. This is why you need more shielding for some forms of radiation than others: for X-rays, you need lead aprons, while alpha particles can be stopped by a single sheet of paper. On the other hand, the fact that just about anything can stop an alpha particle pretty much means that the large alpha particles are going to be stopped by, and transfer their energy, to whatever tissue is nearby them. Fortunately, all three of the isotopes of concern—iodine-131, cesium-137, and strontium-90—all decay by beta and gamma radiation, so we can ignore this particular complication for a moment. [Also, it should be noted that alpha particles inflict far more damage when they enter the body than when they are outside; outside, they are basically all stopped by the skin, which is much less sensitive to radiation than most internal tissues.]
If you multiply the number of rads from these isotopes by a weighting factor based on the relative vulnerabilities of various types of tissues (for example, your eyes are more easily damaged by radiation than your skin), you get the roentgen equivalent in man, or rem level. This is the number you most often hear quoted on the TV (although if you read a scientific paper, you may see the sievert, which is equivalent to 100 rems).
Now here's the important thing in determining the hazard from radioactivity. What matters is both the number of rems and the rapidity of the dose. Most Americans are exposed to something on the order of 300-600 millirems of radiation each year (a millirem is 1/1,000th of a rem).
Some of the power plants in Japan have been cited as causing exposure levels of 400 millirems per hour. In other words: one hour near those reactors subjects tissue to as much radiation as a full year of everyday life. Please note that I am not saying that the people living in the towns near the reactor are being exposed to that much radiation; the dosage drops proportionally with distance (the farther away, the lower the dose, in much the same way that the light from a flashlight becomes dimmer and more diffuse the further it travels).
A sudden dose of less than 25 rems—that would be about two days' exposure to the levels cited near the plant—generally does not lead to any sort of acute radiation poisoning. It does, however, pose a potential cancer risk later in life.
Lesions (e.g., bleeding, sores on the skin) become visible with a sudden dose on the order of 50-100 rems—four to eight days' exposure at 400 millirems/hour. The LD50—the dose at which about 50 percent of people exposed will die within ~30 days—is on the order of 200 rems, which would be more than a week's exposure at those massively high levels. [On Sunday night, radiation levels were about 100 times lower, which would require far too long an exposure to cause acute radiation poisoning.]
Let me close by pointing out that things right now are scary. Nobody can be happy about the fact that radioactive isotopes are being vented into the atmosphere to prevent the possibility of a massive explosion. But, after reading this diary, I hope you can understand why that is vastly, vastly preferable to a sudden explosion. [If you'd like more technical details, such as the formula for calculating rads, please ask; I'll be glad to answer questions as best I can.]
Updated by Samer at Tue Mar 15, 2011 at 08:24 PM EDT
As cwillis notes below, that number of "400 millirems per hour" was actually "400 miliisieverts per hour"—40,000 millirem, which is obviously 100 times higher. That said, later that day, the reported number had fallen to 0.6 millisieverts, or 60 millirems.