# Low-pressure hydrogen plasmas explored using a global model

 Authors Cameron M Samuell, Cormac S Corr Journal/Conference Name Plasma Sources Science and Technology Paper Category Condensed Matter Physics Paper Abstract Low-pressure hydrogen plasmas have found applications in a variety of technology areas including fusion, neutral beam injection and material processing applications. To better understand these discharges, a global model is developed to predict the behaviour of electrons, ground-state atomic and molecular hydrogen, three positive ion species (H+, $\text{H}_{2}^{+}$ , and $\text{H}_{3}^{+}$ ), a single negative ion species (H−), and fourteen vibrationally excited states of molecular hydrogen (${{\text{H}}_{2}}left(upsilon =1\right.$ –14)). The model is validated by comparison with experimental results from a planar inductively coupled GEC reference cell and subsequently applied to the MAGPIE linear helicon reactor. The MAGPIE reactor is investigated for a range of pressures from 1 to 100 mTorr and powers up to 5 kW. With increasing power between 50 W and 5 kW at 10 mTorr the density of all charged species increases as well as the dissociative fraction while the electron temperature remains almost constant at around 3 eV. For gas pressures from 1–100 mTorr at an input power of 1 kW, the electron density remains almost constant, the electron temperature and dissociative fraction decreases, while $\text{H}_{3}^{+}$ density increases in density and also dominates amongst ion species. Across these power and pressure scans, electronegativity remains approximately constant at around 2.5%. The power and pressure determines the dominant ion species in the plasma with $\text{H}_{3}^{+}$ observed to dominate at high pressures and low powers whereas H+ tends to be dominant at low pressures and high powers. A sensitivity analysis is used to demonstrate how experimental parameters (power, pressure, reactor wall material, geometry etc) influence individual species' density as well as the electron temperature. Physical reactor changes including the length, radius and wall recombination coefficient are found to have the largest influence on outputs obtained from the model. Date of publication 2015 Code Programming Language MATLAB Comment