Message: Re: Charge, momentum and mass are not enough to compute the force | Not Logged In (login) |
Hi Peter, Thank you for your answer. No, I haven't seen before how the spin is propagated in GEANT. But I'm afraid that wouldn't help in my case because the ground-state antihydrogen atoms that I'm dealing with have four different hyperfine substates, and these four states have different magnetic moments. Moreover, the magnitudes of the magnetic moments of two of them depend on the magnitude of the external magnetic field. (The two substates in one substate pair always have the same magnitude but opposite sign). The force is magnetic moment magnitude times field gradient. (They are in a sextupole field.) So in order to calculate the force, I have to know in which substate the atom is. One possibility would be to encode this information in e.g. y[42] and y[137], but I don't know if and how it is possible. (The manual doesn't say anything about this.) Or another possibility is that instead of just one antihydrogen particle, I define two: one with a field-independent magnetic moment and another one with field-dependent magnetic moment, and use the spin as the magnetic moment, as you suggested. In that case the magnetic moment magnitude would be magnetic moment vector * field direction (scalar product). However, sometimes the atom jumps from one substate to another, so I might have to convert one antihydrogen particle type to the other one (by killing it and creating the new one), and I don't like this idea. Anyway, I think to let the equation of motion calculator know about the particle type (the way I wrote it in my previous post) is the clearest and easiest way to eliminate this and similar future problems as well. If you wish, I can send you the files that I changed to achieve this. (There are not so many of them, and the changes are simple.) However, they are still for version 6.1. Best regards, Bertalan |
Inline Depth: | Outline Depth: | Add message: |
to: |