Collision History of a Single Particle

Hi Marton,

Is it possible to read out information from the history of a single particle? In my case, I would like to adjust the design of a curved tube so that a typical particle has a particular number of wall collisions before exiting the tube.

If I could make a histogram of the number of collisions per particle (between the gas input at the upstream end and a facet with sticking factor 1 at the downstream end), I could then alter the tube geometry until I matched the intended number of collisions. Of course, there are probably more elegant solutions, and I look forward to hearing your ideas.

Many thanks,

Alec

Hello Alec,

What you can do - since you're interested in a typical particle - is to set only one desorbing surface, then you'll have the following info (for example gained from the formula editor):

  • Number of hits / number of desorption: average number of hits your particle does (not a histogram, but still a good indicator)
  • MPP formula: mean pumping path, i.e. average length travelled by a single particle until pumped (in cm)
  • MFP formula: mean free path, average path between two collisions (in cm)

You're not the first one to request the "number of wall bounces" histogram, though, so I'll think of adding it. Of course, with every histogram comes some memory and performance drawback, and also the fact that the binning (max. distance and number of bins) have to be set up before the simulation.

If you're really pro, then you can in fact create a histogram of the travelled distance. In the moments editor, disable the Maxwell-Boltzmann distribution, then all particles will have a constant speed (which can be queried at different points, for example by an incident velocity profile). Then using the Pressure evolution plotter, you can create a plot similar to this at the exit location, from which it is possible to deduce by a division the flight time, thus the flight distance of particles.

Details of the procedure are in the time-dependent mode description.