Message

Silicon is quite strong. However, it is also brittle. For detectors I have been able to bend 200 micron silicon to less than a 200mm radius. That should be plenty of strain. Virginia Semiconductor is great. Look at their double-side polished stock and you can typically find one or two batches with less than 200 micron thickness. Wafers thinner than 100 microns begin to get quite fragile.

Chemical Engineering Division

Argonne National Laboratory

I have still not decided on the final substrate material, but I have a couple of samples coming from Dupont of polyimide (same composition as Kapton) material called vespel. It is essentially inert as one would expect (like Kapton), but due to differences in manufacture I understand it is hydroscopic. From this I assume that the reaction for making Kapton is diffusion limited and Vespel is essentially a "polycrystalline" -- in the sense there are grains between which water can absorb -- form of Kapton. The energy in question is 11.6 keV (the Ge edge). Si is an interesting idea but the loss for a 200 micron thick substrate is essentially 1/e. It is certainly worth considering for higher energies though! The 1/e value for polyimide is about 3500 microns in contrast, while the 1/e value for MgO is about 300. As in my experiment I want to create biaxial stress in a thin film on the substrate, I worry that the stress/strain curves for MgO are too stiff. On the other hand, thin Si is a real possibility (darn, Si technology is everywhere!). I like the Si idea and might try that in parallel. Has anyone tried using thinned Si wafers (Virginia Technology ? sells mechanically thinned wafers I think -- I saw them at a MRS booth a long time ago). How fragile are the wafers?

On 2004/04/28, at 0:32, Jeff Terry wrote:

Both items are good to know. I didn't realize that the laminated kapton structures still had good heat resistance.

On Apr 27, 2004, at 10:22 AM, Matt Newville wrote:

I think you can get polyimide thicker than 175 microns. It may not

go as Kapton, but maybe as Cirlex or Torlon or something else. I

think goodfellow.com carries these in millimeter thick sheets and

rods, and that they're still radiation and heat resistant.

Goodfellow tends to be pricey, but has excellent information on

thermal and mechanical properties.

Using MgO, sapphire, or even diamond might be reasonable too.

PS: I have a working build of PGPLOT with all of Aquaterm, X11,

Postscript, and Png devices on Mac OS X. I'm still tweaking

the makefile so that it links directly to the png objects to

avoid possible conflicts with dynamic png and zlib libraries,

but I should have a working ifeffit binary using this in a

matter of days, and then be ready to tweak horae's Makefile.PL

so that horae_update works.

_______________________________________________

Ifeffit mailing list

Ifeffit@millenia.cars.aps.anl.gov

http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

_______________________________________________

Ifeffit mailing list

Ifeffit@millenia.cars.aps.anl.gov

http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

National Institute for Advanced Industrial Science & Technology

Center for Applied Near-Field Optics Research (CANFOR)

AIST Central 4, Higashi 1-1-1

Tsukuba, Ibaraki JAPAN 305-8568

tel. +81-298-61-5636

fax. +81-298-61-2939

email: paul-fons@aist.go.jp

The lines below are in a Japanese font

〒３０５－８５６８

茨城県つくば市東１－１－1

つくば中央第４

近接場光応用工学センター

ポール・フォンス主任研究官