GPT Simulation Checklist

Checklist for GPT simulations

Done

• Initial electron bunch simulation - centering electron bunch through 1.5m.
  • See Simulation Results
• Ion production rate benchmarked against analytical theory and against IBSimu.
  • Benchmarking against analytical theory
    • eN and seed tests - Ensuring the ion production rate is not affected by the number of macro particles (eN) in the simulation.
  • Mean Free Path Comparison
    • GPT Simulation: Simulation Video
    • GPT Simulation Results: media:IBSimu_Comparison_Tests_Data.xlsx
    • IBSimu Simulation: media:P_zx_se10.png
    • GPT/IBSimu Results Comparison: media:IBSimu_Comparison_Tests_Data-RicardoComparison.xlsx
• Secondary electron energy distribution (from SEDCS) implemented. Initial tests result in SE energy histograms with the correct shape.
  • media:SEDCS_Benchmarking_IPR_Info_Histograms.xlsx
• Gas species H2, He, CO, and CH4 implemented. H2 and He have the correct SEDCS, CO and CH4 are assumed to have the same SEDCS as H2.
• GPT writeremove custom element has been implemented and tested to produce correct output.
  • Sample GPT Simulation: media:TestMaxwellian_200us_noRand_Efield.pptx
  • Writeremove output (Side view): media:RMaxPic_SideView.png
  • Writeremove output (Front view): media:RMaxPic_FrontView.png
• Created simulation of "DC" beam pulses that include space charge, the T-cathode electric field, and secondary electron production for H2 gas. "DC" implies the simulation consists of a very long electron (macro-particle) bunch with a user-defined total charge that is uniformly distributed to all electron macro-particles.
  • See sample GPT simulation in last bullet - This simulation shows two "DC" pulses 5ns in length separated by 10ns. Each pulse has total charge 1nC distributed uniformly to 10^3 particles. The simulation is split into 3 parts with different timesteps: the first 18ns of the simulation uses a 50ps timestep. Then the timestep increases to 1ns until the simulation reaches 500ns. Then the timestep increases to 1us until the simulation reaches 200us.
• Creating color coding custom element for color coding simulation particles by type.
  • media:TestMaxwellian_200us_noRand_Efield_colored.pptx -- same as sample simulation above, but color-coded. Blue=primary electron, Green=secondary electron, Red=ion.

In Progress

• Maxwellian distribution for ion energies has been implemented and benchmarked externally. Need to show that it also works in the GPT ionization routine.
  • Initial tests using external C++ code and comparison with Cristhian's code: File:Maxwellian_Distribution.xlsx
• The spacecharge3D routine is being benchmarked against analytical theory. An initially stationary ion a certain distance away from a DC electron beam will oscillate about the electron beam due to the electric field of the electron beam. At the same time, the ion will circle the wire due to the magnetic field generated by the electron current.
• The secondary electron and ion energy routines are being benchmarked against theory. Although both have been correctly implemented and histograms of secondary electron and ion energies appear to have the correct shape, an analytical calculation to determine whether these histograms are indeed correct and consistent with the probability distribution equations needs to be done.
• GPT simulations of DC and pulsed electron beams are being created to show whether ions can remain trapped within the electron beam potential. The total simulation time will need to be rather long to show ions moving within the beam potential.
• Optimization of ionization code to improve simulation run time.

To Be Done

• Create a new custom element that will allow GPT to import vacuum data to use as the local gas density for ionization calculations instead of assuming a constant gas density throughout the simulation.
• Derive an approximation for SEDCS for CO and CH4. Kim (et. al.) provide a way to approximate the SEDCS using the so-called Binary Encounter Bethe Model, as the differential oscillator strengths necessary for calculating the SEDCS for CO and CH4 are not available.
• Create a new custom element to simulate secondary electron yield from interior surfaces.

Go to Ionization Effects Meetings

Return to Ionization Effects