Mean ionisation potential is different in G4NistMaterialBuilder::NistSimpleMaterials and G4DensityEffectData::Initialize functions. This may cause confusion when creating materials. For example, for carbon made via G4NistManager::FindOrBuildMaterial("G4_C"): I=81, C=3.0679807, x0=-0.018709584, x1=2.5023904, for carbon made "manually" via G4Material: I=81, C=2.9925, x0=0.0351, x1=2.486, and in G4DensityEffectData::Initialize (Sternheimer (1984) tables): I=78.0, C=2.9925, x0=-0.0351, x1=2.486.
Hello, parameterisation of Sternheimer was done in 1984. NIST mean ionisation potentials was extracted in 2005 and updated later after ICRU73 and recently with ICRU90 reports. So, Nist material data are supported more by experiment. At the same time, we cannot change original parameterisation of the density effect made in 184, so we keep the table from this publication unchanged. During initialisation we scale the density effect shape to have relativistic plato according to the NIST ionisation potential. VI
