The proposed project will investigate the physical mechanisms governing the industrial production of tungsten powder, and will aim to accurately capture these physics in a numerical modeling approach. Of pivotal interest is the prediction for the powder’s grain size distribution under realistic industrial conditions because this distribution directly influences key material properties (e.g. toughness, strength) of tungsten carbide based products.A numerical modeling methodology is particularly valuable in the case of tungsten production because the physical environment is very hostile, leaving little scope for experimental investigations. Capturing the governing physical-chemical processes would enable the exploration of different operation conditions and the control of the optimal powder size distributions.This novel modeling approach will be incorporated into the eXtended Discrete Element Method (XDEM) that extends the classical Discrete Element Method by a coupling via mass, heat and momentum transfer between a particulate phase and a liquid or gas represented by state-of-the-art Computational Fluid Dynamics (CFD). Hence, CFD serves to predict gaseous products and volatiles flow as well as their development in the void space between particles. The thermochemical and morphological conversion of tungsten oxide particles is described by novel statistical approaches based on XDEM, whereby the tungsten formation process will be represented for industrial reactors in reasonable computational cost and time. As revealed by previous studies performed by the University of Luxembourg, in collaboration with CERATIZIT Luxembourg S.à r.l., grain size predictions will provide to the Powder Metallurgy (PM) industry with a valuable tool for gaining insight to control the quality of tungsten made products. Thus, this research significantly strengthens the Luxembourg PM sector and the European research environment. Furthermore, these efforts also contribute to the growth and wealth of the Greater Region and its university and align very well with the FNR Foresight priority of “Sustainable Resource Management”.