Hi, I just read about an interesting characteristic of the nucleon-nucleon interaction (e.g., the scattering of a proton with a neutron or a proton with a proton or a neutron with a neutron). I wonder whether this characteristic is behind what we've been referring to here as "neutron tunneling" or "neutron stripping" in earlier posts to this list.
This characteristic is called the "one pion exchange potential," and it can be seen in low-energy neutron-proton scattering experiments (< ~ 20 MeV). In these experiments a neutron becomes a proton and a proton becomes a neutron through the exchange of a charged pi meson, or pion. In the scattering experiments the pion is not a real pion but a virtual one, limited by the constraints of the uncertainty principle. For this reason it has a very short range, on the order of femtometers, which is the range at which nuclear reactions take place. To see how virtual pion exchange works in an experimental context, consider a neutron that is incident upon a proton. The outgoing neutron will scatter in any number of angles, with differing probabilities. A likely scenario is that the neutron will scatter at a small angle -- some angle close to 0 degrees, for example. In these cases the neutron is deflected a little bit from its initial trajectory. In other cases it will scatter into a larger angle, perhaps up to 90 degrees or more. These outcomes are seen less and less at larger angles. Where things get weird is that beyond a certain angle the probability starts to increase again and reaches a maximum comparable to the small scattering angles. To the observer this looks as though what is happening is that the neutron is incident upon the proton, and the neutron transfers all of its momentum to the proton, and the proton then leaves the scattering along a trajectory similar to the one in which the neutron entered it. These interactions occur with a probability that is comparable to the small-angle scatterings. For a number of reasons that go beyond my knowledge of the system, physicists prefer a different explanation for what is happening in this second case. Instead of an account in which the momentum is transferred from the incident neutron to the outgoing proton, this second case is explained by the neutron becoming a proton, and the proton becoming a neutron, and the neutron-now-proton continuing on its way at a small deflection angle. In this account there is a virtual pion that is exchanged between the neutron and the proton, which causes the proton to become a neutron and vice versa. In one direction the exchange involves a positively charged pion, and in another direction it involves a negatively charged pion. Because these pions have ~ 134 MeV mass, their range is quite short, and beyond this range, the effect becomes negligible. What I wonder is what the distribution of the range of this interaction looks like going out one or two standard deviations, sort of like the high-energy tail in the Boltzmann distribution. Perhaps over the duration of a low-energy np scattering, the virtual pion can only make it out to several femtometers before the uncertainty principle makes it improbable. But if there's a long-range tail to this kind of interaction, which occurs with less probability, longer range interactions might be able to occur at significant probabilities when considered over longer periods of time. If so, we would not have a ~ 1GeV rest-mass neutron traveling from a 7Li to a nickel nucleus, but instead a ~ 134 MeV virtual pion being exchanged between a neutron in the 7Li and a proton in the nickel nucleus, causing them to change into one another. Eric
