Date: Sun, 18 May 2003 10:38:29 -0400 (EDT) From: Wally Melnitchouk To: Sergey Kulagin Subject: Re: spectral function Hi Sergey, Thanks for checking the numbers. My fortran program gives for the normalizations: > Number of nucleons: > R0P = 1.99838 and R0N = 0.998754 1.99229 0.988317 pmax=6.5/fm (1,000 pts) Emax=150 MeV (10,000 pts) 1.98944 0.982703 pmax=6.5/fm (1,000 pts) Emax=150 MeV (1,000 pts) 1.97177 0.969760 pmax=6.5/fm (1,000 pts) Emax= 68 MeV (10,000 pts) > Polarizations: > R1P = -0.044047 and R1N = 0.884194 -0.0490068 0.873550 -0.0518748 0.867901 -0.0454000 0.873542 > Tensor polarization: > R2P = -0.233399 and R2N = 0.0351692 -0.235589 0.0370413 -0.235572 0.0370434 -0.254418 0.0256887 One can see there is some sensitivity to the cut-off, and to the number of integration points. I expect this would decrease if we had a spectral function on a finer grid. > The normalization of the spin independent part looks OK. The > polarizations are somewhat different from those I was able to extract > from Friar et al. paper (-0.056 for the proton and 0.86 for neutron). > However this could be caused by different wave functions used by SS > and Friar et al. The proton tensor polarization looks surprisingly > large (this is what you have complained about?). It would be nice to Yes, this number looked anomalously large. It leads to a rather large proton contribution to the g1 and g2 He structure functions. The polarizations R1 one could live with I guess as they're reasonably close to typical values, but the R2 values I don't know what to compare with (other than what you suggest below). > have an independent check of this. In principle we could try to > evaluate the expectation value of the tensor operator \hat T with the > 3He ground state wave function and express it in terms of the orbital > probabilities, unless we do not find it in literature. Send me your > numbers for comparison. Cheers Wally