Technically Speaking: Capture
versus Beta Decay

The question of whether neutron capture is "fast" or "slow" must be asked relative to the timescale set by typical beta decays in and near the stability valley. Generally, the beta decay half-life gets shorter as one moves further out of the stability valley.

For red giants, the neutron capture rate on a given nucleus may be about once per year. Thus, neutron capture will, on average, proceed until an isotope is made that has a half-life for beta-decay that is considerably less than a year. In the iron neutron capture sequence discussed in the section on the s-process, iron-59 has a beta decay half-life of about a month, so it is unlikely that neutron capture can proceed beyond iron-59 in red giants before a beta decay converts iron to cobalt. On the other hand, in a supernova the capture rate may be many neutrons per second and only nuclei very near the neutron drip line have beta-decay half lives short enough to compete with this.

The r-Process

The s-process can build up heavier nuclei slowly. However, it is restricted to the stability valley because, if enough neutrons are captured to move out of it, one or more beta decays will occur to return to the stability valley. Thus, the s-process cannot account for the observed existence of nuclei out of the stability valley. Also, because there is a gap in the stability valley beyond 209-Bi (the mass-209 isotope of the element bismuth) the s-process cannot cross the gap to build the heaviest nuclei like uranium, thorium, or plutonium (these are termed the transuranium elements).

Rapid Neutron Capture
But suppose the density of free neutrons were so high that instead of capturing one or two neutrons before a beta decay there was a very high probablility for capturing 5 or 10 or even 20 neutrons before a beta decay could take place, as illustrated in the following figure.

Then the rapid capture of neutrons would take us far to the right in the Segrè chart and out of the stability valley. This could do two things: First, it could produce a whole series of "neutron-rich" isotopes (isotopes having more neutrons than those in the stability valley) when the population beta decays back toward the stability valley after the flux of neutrons decreases. Second, if the capture is rapid enough the gap noted above can be bridged, and beta decay back to the stability valley can produce the stable isotopes of the transuranium elements that cannot be produced in the s-process.

The r-Process
This is exactly the idea of the rapid neutron capture or r-process. The difference between the s-process and the r-process is that in the r-process the flux of neutrons is large and the capture rate is very high; thus it is a rapid capture. The s-process is thought to take place primarily in red giant stars, but the large flux of neutrons required for the r-process cannot be produced in a normal star. The most likely site for the r-process is in a Type II supernova (see the right panel).

The r-Process Path
The path followed in the r-process during a supernova is illustrated by the red area in the following diagram.

Notice that the r-process path runs very close to the neutron drip line. That is, it runs through extremely neutron-rich isotopes. When the neutrons moderate in intensity as the supernova explosion begins to wane, the resulting population then beta decays back toward the stability valley. This animation illustrates the r-process path in the chart of the nuclides.