Silicon is unfortunately even MORE difficult. It is very transmissive, so you have three sources of exiant radiation instead of just two with the aluminium. I have some amazing images of silicone, but I'm not going to post them.

 You can work with both of these materials, but they are VERY difficult and the imformation you get may be more limited than you expect. You will definitely have more success with aluminium than silicon.

Why can you not change the emissivity?

---

Bob Berry
BINDT, IRT Level 3 EL, Level 3 CV (ISO 18436)
Thermal Vision
8 Old Fair Green
Dunboyne
Co Meath
Ireland
bob@thermalvision.ie
www.thermalvision.ie

Each piece of silicon has about 300 holes, each 30 microns in diameter.  These holes cannot get plugged, and the plate is narrow enough that I don't think painting a spot on it would give enough area to measure.  That is why I didn't want to change the emissivity of the block of aluminum.  I wanted to see what can be done with a a low e and highly reflective surface.

Ralph Felice has a graph showing the emissivity of Silicon versus wavelength and temperature www.pyrometry.com/semiconductors.php.   His company thrives on ACCURATE noncontact measurement of stuff that is very hard to measure.

 

InGaAs short wave cameras are commonly used to find defects in silicon wafers.  InGaAS typically sees from 0.9 to 1.7 microns.  Typically they are used to see through silicon and only defects show up.  Perhaps that type of camera could be filtered to show a thermal image.  If thermal imaging is something you have to do give me a call and we can see if an InGaAS camera can be pushed to do what you want.

 

--
John C. Lafeber, P.E.
NCII - North Coast Industrial Imaging
4165 Stein Road
Ann Arbor, MI 48105

734-904-0870
LeanIR@gmail.com

Isaac -

I don't know if you are dealing with an issue of confidentiality, but it strikes me that you are spoon feeding us information and getting mixed guidance as a result.  If you can give us a fuller picture of what you need to do, on what, and with what, I think you will get better guidance.  Your recent post about the silicon with all the holes and your use of the aluminum as a substitute for testing was, I think, critical. 

 The fact that they are both shiny or polished, does not mean that aluminum is a good substitute for silicon in your test development.  As John Lefeber's referenced graph shows, the emissivity of silicon is no worse than about 0.2 in the waveband of your camera.  This is many times higher than polished aluminum.  Polished aluminum is therefore a much more difficult material to successfully image than the silicon.  In addition, you note that the the silicon has many small holes in it.  These will, if they are deep enough and frequent enough, give you a much higher emissivity than the unaltered silicon or the graph referenced by John. 

If possible, I suggest you get a sample of the silicon that can be "wasted", that is, that you will be allowed to touch and handle.  Do your testing with it. 

Not all shiny surfaces are low emissivity.  Try looking at the data side of a CD. Under the transparent coating it is an aluminum disk, however the data writing process generates many small holes that end up giving it a usable emissivity. 

 

Jack

 

---

Jack M. Kleinfeld, P.E.
Kleinfeld Technical Services, Inc.
Bronx, NY
718-884-6644
JKEngineer@KleinfeldTechnical.com
come see what we can do for you: http://www.KleinfeldTechnical.com

Isaac,

Pure silicon is very transmissive, it is used for low cost IR lenses, measuring it directly will be almost impossible. On the other hand doped silicon is not always transparent. It sounds like you are dealing with some kind of through silicon via technology, the wafers I have come across using this technology are all doped silicon, and might actually be quite easy to measure. If you want to test this, take your IR camera and try looking through one of your silicon wafers at a halogen lamp, that will let you know if there is much transmission.

Assuming it is transmissive and you can't measure the silicon directly, there are other options; you must be mounting these dies somehow, that could be your key. If they are in a package with die attach, the die attach compound should have high emissivity, and ought to be in thermal equlibrium with the silicon. You'll be looking through the silicon to the attach compound, but it won't matter. Or, you might be clipping these in some kind of mount, if so, the die will be pretty much at the temperature of the mount, measure the mount. If you want to get an accurate profile of temperature across the die, then you are pretty much going to have to use an IR microscope lens, so you'll probably have to pattern some oxide or nitride targets on the wafer during production, and measure off those. Microscope lenses usually suffer badly from the Narcissus effect, where you see the detector reflected in the image, that makes reflective surfaces *very* difficult to deal with.
 

Luke

Luke,

 

I agree doped silicon is workable, it is possible that etched and maybe even masked silicon is workable too, but pure silicon is as you said VERY transmissive.

---

Bob Berry
BINDT, IRT Level 3 EL, Level 3 CV (ISO 18436)
Thermal Vision
8 Old Fair Green
Dunboyne
Co Meath
Ireland
bob@thermalvision.ie
www.thermalvision.ie