ePIC pfRICH mailing list

["Kiselev, Alexander" <ayk AT bnl.gov>]

  Hello colleagues,

  Concerning the definitions on slide 6 of Andy's talk yesterday

https://indico.bnl.gov/event/20307/contributions/79775/attachments/49183/83806/pfRICH-Prototype-081423%201.pdf

, see Ra and Rz here:

https://upmold.com/surface-finish-ra-rz/

while the difference between Ra and Rq is "Average roughness (Ra) is the average of the individual heights (asperities) and depths from the arithmetic mean elevation of the profile. Root mean square roughness (Rq) is the square root of the sum of the squares of the individual heights and depths from the mean line."

  The problem with all this is that it is all about surface roughness in the transverse direction (dimension of [distance]), while we would naively expect a dimensionless quantity aka [angle]. 

  However, as it seems to be indicated in this old paper https://apps.dtic.mil/sti/pdfs/AD0262572.pdf, once surface roughness as quantified by Ra/Rq/Rz becomes much smaller than the wave length, the slopes do not matter, the reflection is specular rather than diffuse (?), with a net effect of an overall (small) reflectivity loss, going further down with roughness. That's why CMA who can achieve an exceptional polishing quality can claim >95% reflectivity overall. For us however there is no difference between say 92% and 95%, especially if this difference comes at a cost of a few hundred $k.

  I'd appreciate if somebody finds time to cross-check this assessment. Volunteers?

  All in all, if we are not getting down to dozens of nm roughness with our in-house production, and end up around ~100nm or so, we probably just need to measure the actual reflectivity, and the actual angular spread, plug the numbers into GEANT as measured (this is very straightforward) and run a simulation.

  Regards,

    Alexander.