Numerical simulations of plasmas are an indispensable tool of investigation in plasma physics and plasma engineering. Building a plasma device is expensive, and iterations for its optimization are in many cases prohibitive or limited by economic resources. Empirical methods of optimization often lead to an incomplete understanding, forcing a trial and error process way too expensive for a sustainable progress. On the other side, computer simulations are much cheaper, and can provide a much bigger amount of information on a given plasma process. Computer aided design of plasma devices can allow huge reductions of development costs in both research and industry, giving the possibility to “know before building” what the performance of a plasma device will be. However, computer models are often restricted by a big number of assumptions, restricting the analysis to a limited scale of space and time. The problem is that each plasma process is intrinsically a multiscale process of huge complexity, involving many interconnected physical and chemical processes acting at different spatial and temporal scales.
In this seminar I will review my research activity of multiscale plasma modeling, encompassing fluid plasma models, particle-in-cells and microscopic kinetic models. After a presentation of the multiscale nature of matter in the state of plasma, I will show models for each scale, examples of numerical solutions and applications tightly connected with past and ongoing experimental projects, comprising: helicon plasmas, plasma for space propulsion, plasma sources of exotic isotopes beams and plasma-material interaction.