Deuterium Diffusion Query

[Tom Banks <tbanks@socrates.berkeley.edu>]

I have a question for our in-house deuterium diffusion experts, but first
I need to describe my current situation:

As I mentioned at the close of Tuesday's teleconference, I am working on a
fast Monte Carlo program to simulate diffusion effects and explore the
possibility that they might be responsible for the gondola effect.  At
present I have implemented (thermal) mu-p diffusion only, which is a
relatively simple process to model: you sample from the normalized
diffusion function

    f(r) = sqrt(2/pi) * r^2 * exp(-r^2/2) ,

and then scale the sampled value according to the time of decay

    r -> r * sqrt(2Dt) ,

where D is the diffusion constant.


I am now at the point where I would like to incorporate deuterium
diffusion into my program as well.  I know that mu-p -> mu-d transfer and
subseqent diffusion is a complex dynamic process, but is there any way I
could crudely model the situation by sampling from some generic "mu-d
diffusion" function and scaling the result in time, analogous to the mu-p
diffusion algorithm described in the preceding paragraph?  According to
Brendan's plots in ELOG #19,

    http://kaon.physics.berkeley.edu:8080/analysis-run8/19

the mu-d diffusion splits into 2 peaks at late times, so I know that this
wouldn't be completely accurate.  But I'd like to know if you think I
could model mu-d diffusion by sampling from some standard mu-d function
(the sampling frequency would probably be determined by the deuterium
concentration) and then scaling the result according to an appopriate
effective diffusion constant.

Thanks,
Tom