Semi-automated spectral quantification of the concentration of ingredients in gun propellants

Using IR Spectroscopy to determine the concentration profile of deterrents of gun propellants is a well-known and established method, as presented for example by Vogelsanger et. al or Kaiser et. al. (1-5). This measurement methodology among others was used in the EDA project "Prediction Models for the Implementation of Munition Health Management" (PREMIUM) to further investigate the effects of ageing on the ballistic stability and performance parameters of ammunition. In this context a semi-automated procedure for evaluating the IR spectroscopy images was introduced, which includes the baseline correction "adaptive smoothness parameter penalized least squares" introduced by Zhang et. al. and machine-learning-supported band detection (6). This method was tested on 80 measurements in total from the gun propellants K5810 which is used in DM41 and K15230 which is currently used for the 4.6 x 30 mm for example for the MP7. The results are precise spatially resolved concentration profiles of relevant propellant components such as nitrocellulose (NC) or dibutyl phthalate (DBP) with little effort, which allows the further analysis of the diffusion effect of the latter.