Fundamental interactions in plasmas are very well defined and researched, but the sheer number of possible interactions between particles and materials makes plasma research very complicated. Processes in plasma occur over many orders of magnitude in time and spatial scale, requiring careful selection of physical and mathematical models for describing included processes. Geometry simplification and reduction of spatial complexity are made to include more detailed and complicated reaction kinetics (reaction sets). Global models (volume-averaged or 0D model), well known for their fast evaluation times, are often employed where spatial dependence of plasma parameters is not significant and more complicated chemistry is required.The electron energy distribution function (EEDF) profoundly affects reaction kinetics as many reactions depend on the electron energy. The Maxwellian EEDF is often assumed, leading to overestimated electron impact reaction rates and time evolution of species densities. A self-consistent EEDF evaluation is incorporated in the global model framework to capture the temporal evolution of the EEDF and further increase the fidelity of the results. The EEDF evaluation frequency is user-definable and affects both simulation accuracy and run times. The Kinetic Global Model framework (KGMf), coupled with self-consistently evaluated EEDFs, has been used to study complex reaction chains in pulsed low-temperature plasmas across a range of pressures with good comparison to theoretical, computational, and experimental results from the literature.Many input parameters in plasma simulations, i.e., impact cross-sections and the reaction rate coefficients, are derived from experiments and approximate theories, and they contain uncertainties. Additionally, experimentally measured values from different experiments can deviate significantly. Sensitivity analysis can quantify the effect of input parameter uncertainties on plasma behavior for a given system. The reaction rate coefficients were selected as input parameters for global sensitivity analysis (GSA), and a high-dimensional input parameter space was reduced with Saltelli's sampling. The effects of reaction rate coefficient uncertainties on species densities in a nanosecond pulse discharge inan H2-O2-Ar gas mixture were explored and quantified using sensitivity indices.Large reaction sets are prohibitively expensive in spatially dependent simulations, limiting detailed chemistry in fluid and PIC simulations. Reactions and species have different contributions to the selected plasma quantities of interest, e.g., species density or temperature. Defining the importance of reactions and species for a given system can reduce there action set with known effects on the selected quantity of interest. The reduced reaction set has lower computational complexity and is suitable for global sensitivity analysis or spatially dependent simulations. The directed relation graph (DRG) method is presented and deployed in a nanosecond pulse discharge in an H$_2$-O$_2$-Ar gas mixture, reducing the reaction set while preserving the H and H$. {+}$ densities. The reduction steps were compared in terms of the size of the reduced reaction set and the simulation run time reduction.
Read
