I am currently using Garfield++ to simulate the drift behavior of electrons and ions in gas detectors, and I have a technical question regarding the calculation of drift velocity.
As I understand from the Garfield++ documentation, the drift velocity of electrons/ions is directly derived from input electric and magnetic field parameters (based on pre-calculated transport parameter tables) rather than being computed by integrating acceleration (i.e., accounting for particle inertia a=qE/m).
For the drift trajectory of ions (which have much larger mass than electrons and thus more significant inertial effects), will this simplification (ignoring inertia and directly assigning velocity) introduce non-negligible errors in the simulation results (e.g., trajectory deviation)?
Hi,
for electrons it depends on the method used for simulating the drift lines. AvalancheMicroscopic uses the second-order equation of motion a =q E /m for calculating the trajectory of the electron between free-flight steps, and the microscopic electron-atom/molecule cross-sections for calculating the inverse mean free path. AvalancheMC and DriftLineRKF use indeed the average macroscopic drift velocity (normally pre-computed using Magboltz together with other swarm parameters like diffusion coefficients, Townsend coefficient, etc.).
For ions, only methods based on the macroscopic drift velocity (AvalancheMC, DriftLineRKF) are available but this is usually a good enough approximation (better than for electrons).