Re: Machining diamond

[Richard Jones <richard.t.jones@uconn.edu>]

Jim Stewart wrote:
> Dear Tagger
>
> While at BNL I spoke with Triveni Rao about thinning diamond. She is the
> head of the laser source group and is working on a high current electron
> source. For this she needs 1cm diameter 30um thick diamond wafers.
> This means they are also looking for ways to thin diamond wafers. They
> tried reactive ion etching but this took a very long time and was
> expensive. They are now looking into ablating the diamond with an excimer
> laser. They have ordered a new laser and will start testing soon. She
> would be interested in collaboration on developing techniques to hold the
> samples.
>
> Does anyone know what surface flatness and roughness we would require? 
There are no specifications on surface quality, per se.  As far as the 
surfaces are concerned, the only requirements are to have them 
reasonably normal to the 001 direction (within a few mr) and keep the 
thickness reasonably uniform (within +/- 5 microns).
> An interesting possibility, in the event that they succeed in thinning the
> diamond, would be to leave the outside of the sample thick and then thin
> the middle. This would leave a strong structural support around the edge
> of the sample and hopefully keep the diamond surface flat.
>   
Yes, we have discussed that possibility with our colleagues at CHESS.  
So far, the only thinning option we have found so far -- grinding with a 
diamond-impregnated plate -- requires that the entire wafer be thinned 
at the same time.  Even then, the uniformity in thickness is not much 
better than our spec of  +/- 25%.  This is a very good idea, and might 
be the only way to maintain sufficient planarity at 20 microns.  It 
requires two things: very large diamonds, and a way to thin them 
differentially.  I say very large because of the size of the beam spot 
at the radiator: we need the thin part to cover at least 3 sigma around 
the center of the beam, or else a sizable fraction of our photon beam 
will come from the tails of the beam interacting in the thick part of 
the diamond.  This might severely restrict the usable area of the 
diamond that is available for spot moves, effectively limiting the 
radiation lifetime of the radiator.  We need to talk about this more.  
Did you ever run into someone named Ben-Zvi?
> How thick are the wires used to hold the diamonds?
>   
The diamonds that we have from Element 6 come mounted with 25 micron 
tungsten wires.

Richard Jones