What chemical is best for etching MgF2?

Factual Questions
1

I am trying to turn a single crystal VUV MgF2 window into a weak diffuser, kind of like shower glass. I have done things like this before with fused silica or other optical glasses like BK7, but never for a single crystal like MgF2. I thought at first about using an acid, but after talking to a friend that is more knowledgable of chemistry than I, they said that an acid would probably oxidize the MgF2 and ruin the optical properties of the window (I am using this in transmission down to Lyman Alpha, 121 nm). They suggested that just using water would probably work, but after trying it I find that it is not working. I think they were thinking of LiF which is much more hydroscopic and is also used in VUV systems.

Anyway, any thoughts?

2

I was thinking acetic acid (vinegar) would be the best place to start. Will it oxidize the surface and ruin the transmission properties? These windows are expensive and I only really have 1 to play with…

3

How about using something like this, with as fine a grit as you like? You can get grit down to the micron range in size.

4

I actually already did this with one window and it was effective. The problem is that it is too diffuse (it spreads the angle of my beam too much). The way to control how much a diffuser spreads the light is by controlling the feature size (actually the spatial frequency). The sand blast produces very small features (high spatial frequency) on the the surface, no matter how fine or coarse the grain size is and this produces a lot of spreading. I want to only spread the beam by about 5 degrees, 1 degree or less would be even better. There are commercial products that do this (beam shaping diffusers), but none that work in the UV thus my attempt to make one.

5

I think chemical treatment may soften the features on the MgF2 window you have already roughened. Based on the limited solubility, I would run it under DI water briefly and see how that effects the diffusion. If it starts making it better, go back and do it some more. I would actually start very briefly, since the first rinse will likely wash off loose particulate that maybe responsible for a lot of diffusion.

I think you are unlikely to get the effect you want from a chemical treatment alone without a lot of trial and error, and it sounds like you can’t afford that.