Hello, I've simulated a propagation of low energy (KE~1 MeV) KLong0 beam through 1 meter of a concrete slab. I compared Geant4 versions 10.4 and 10.5. The attenuation factor of KLong0 beam in versions 10.4 and 10.5 are off by two orders of magnitude. KLong0's are attenuated by a factor of ~1.2x in the version 10.4, which seems unphysical. Is there a bug in the Kaon0 cross-sections in the version 10.4? If so, can you please provide a reference to the bug description and it's possible resolution in version 10.5. Thank you, Yuri.
I used QGSP_BERT in this study
Hello, Likely you face the same problem as reported in #2220. Please, try out recommended PhysicsList FTFP_BERT. VI
Please try the latest patch, Geant4 10.6p01; the problem you see with QGSP_BERT is likely fixed there. Let us know how it goes.
Dear Dennis and Vladimir, We've recently simulated a large MC sample with Geant4 10.5.p01. I observed a large difference between 10.5 (significant attenuation of KL in concrete) and 10.4 (almost no attenuation of KL in concrete). Do you know if the bug is present in 10.5 as well? Do we have to use 10.6 to not being affected by that bug? Can you please describe what the bug was? Best, Yuri.
Hello Yuri, Looking into development history it is possible to restore situation as following. For 10.4 kaon cross sections for high energy have been fixed and verified. The fix was back-ported to 10.3. No attention was to low-energy case. In 10.5 a problem #2220 was introduced to the GSHEISHA cross sections. We start to use the Glauber cross section by default. In 10.6p01 we fix #2220. In 10.6 kaon cross sections were revised and elementary cross section parameterisation was improved. See http://vnivanch.web.cern.ch/vnivanch/hadronic/HadrHN/20190819/A_kaon+_proton_VG.png and http://vnivanch.web.cern.ch/vnivanch/hadronic/HadrHN/20190819/A_kaon-_proton_VG.png . From elementary cross sections we obtain kaon-nuclear cross section using Glauber approach. Some description is available (J. Allison et al., Nucl. Instrum. Meth. A 835, 186-225, 2016). KL cross section is defined as a half sum of K+ and K- according to isotopic invariance. We do not have data for KL, which may be parameterised. Isotopic invariance works fine if the electromagnetic interaction may be neglected, which is true for high energy. For your use-case there is a problem. We cannot guarantee good accuracy for KL absorption at 1 MeV because K+ x-section has a Coulomb barrier, which is absent for neutral kaons. Also K- cross section in the current parameterisation is increased for low energies faster than expect from quantum mechanics 1/v dependence. We may improve our KL cross sections at low-energy but for that we need to have, at least, few KL data points for whatever materials. Do you have such information? best regards, Vladimir
Hello, cross sections of kaons were updated at low energy in Geant4 10.7beta. This report may be closed. Vladimir
