A two dimensional axisymmetric quasi-neutral fluid model of an emissive hollow cathode that includes neutral xenon, single charge ions and electrons has been developed. The gas discharge is coupled with a thermal model of the cathode into a self-consistent generic model applicable to any hollow cathode design. An exhaustive description of the model assumptions and governing equations is given. Boundary conditions for both the gas discharge and thermal model are clearly specified as well. A new emissive sheath model that is valid for any emissive material and in both space charge and thermionic emission limited regimes is introduced. Then, setting the emitter temperature to an experimentally measured profile, we compare simulation results of the plasma model to measurements available in the literature for NASA NSTAR barium oxide cathode. Qualitative discrepancies between simulation results and measurements are noted in the cathode plume regarding the simulated plasma potential. Motivated by experimental evidence supporting the occurrence of ion acoustic instabilities in the cathode plume, an enhanced model of electron transport in the plume is presented and its consequences analyzed. Using the obtained plasma model, simulated quantities in the plume are qualitatively comparable with measurements. Inside the cathode, the simulated plasma density agrees well with measurements and is within the ±50 % experimental uncertainty associated with these measurements. A comparison of simulation results of the full coupled cathode model for the NASA NSTAR cathode with experimental measurements is presented in a companion paper, as well as a physical analysis of the cathode behavior and a parametric study of the influence of the operating point and key design choices.

A numerical emissive hollow cathode model which couples plasma and thermal aspects of the NASA NSTAR cathode has been presented in a companion paper and simulation results obtained using the plasma model were compared to experimental data. We now compare simulation results with measurements using the full coupled model. Inside the cathode, the simulated plasma density profile agrees with the experimental data up to the ±50 % experimental uncertainty while the simulated emitter temperature differs from measurements by at most 5. We then proceed to an analysis of the cathode discharge both inside the cathode where electron emission is dominant and outside in the near plume where electron transport instabilities are important. As observed previously in the literature, the total emitted electron current is much larger (34) than the set discharge current collected at the anode (13) while ionization plays a negligible role. Extracted electrons are emitted from a region much shorter than the full emitter (0.9 vs. 2.5). The influence of an applied axial magnetic field in the plume is also assessed and we observe that it leads to a 10-fold increase of the plasma density 1 cm downstream of the orifice entrance while the simulated discharge potential at the anode is increased from 10 up to 35.5. Lastly, we perform a parametric study on both the operating point (discharge current, mass flow rate) and design (inner radius) of the cathode. The simulated useful operating envelope is shown to be limited at low discharge current mostly because of the probable ion sputtering of the emitter and at high discharge current because of emitter evaporation, plasma oscillations and sputtering of the keeper electrode. The behavior of the cathode is also analyzed w.r.t. its internal radius and simulation results show that the useful emitter length scales linearly with the cathode radius.