Can a CMOS sensor be used as a Geiger nounter

[swt94025]

Can a CMOS sensor be used as a Geiger counter

To help the folks in Japan protect themselves from contaminated food and to get some assurance of the safety of their homes, we are putting together a team to evaluate the feasibility of writing an iPhone/Android application that would use the CMOS sensor as an ionizing radiation detector.

There are several issues that have to be overcome: battery life, the lack of a shutter, attenuation by the camera housing, but by far the biggest issue is the sensitivity due to the thin layer of active silicon in a CMOS camera sensor compared with that in a commerial radiation sensor.

Is there anyone in this group that has access to the simulation tools for energy deposition from ionizing radiation and the expertise to use them?

If the numbers work out even remotely (overnight integration times maybe?) we are going to write an app and put it up in the App store to help the Japanese get some visibility.

 

[Bob S]

Solid state radiation detectors (e.g., PIN diodes) can be used in both ion chamber (continuous current) and pulse (Geiger counter) mode. See for example

http://www.carroll-ramsey.com/detect.htm

I disagree with their ion chamber calculation; the current depends on the sensitive volume, not area.

I calculate 1 rad/hr = 0.01 Sievert/hr = 1.77 nanoamps per cubic cm.

You need about 100:1 signal to noise ratio at 1 rad/hr.

What is the size of the sensitive volume in your CMOS device?

Bob S

 

[swt94025]

Bob,

That is a very useful figure. Is it for Gamma or Beta radiation? I would imagine that for sensors thick enough to reliably generate a signal that they would be dependent on area not volume, so should I assume this figure is for gamma rays?

The sensor has an area of about 0.4 cm^2 and (unfortunately) an epi layer thickness that is probably around 5 microns. I don't have an accurate figure for the latter, but it won't be an order of magnitude off from that. I'm not 100% sure that the epilayer is the operative thickness, as I'm not sure if there might be some sensitivity from deeper parts of the sensor, but if it works the same as vible light photons, then the end of the epilayer is the end of the sensitive region.

So the total volume is 2e-4 cm^3. At this rate we have 0.35 picoamps per rad per hour. That is a reasonable starting point.

Not knowing the secondary electron energy, I'm concerned that the range of those electrons will require a (downward) adjusted value for such thin layer, as there might be fewer secondaries coming from the space above the sensor. Ironically, the previous iPhone might have better sensitivity, not being a back-thinned detector.

Solid state radiation detectors (e.g., PIN diodes) can be used in both ion chamber (continuous current) and pulse (Geiger counter) mode. See for example

http://www.carroll-ramsey.com/detect.htm

I disagree with their ion chamber calculation; the current depends on the sensitive volume, not area.

I calculate 1 rad/hr = 0.01 Sievert/hr = 1.77 nanoamps per cubic cm.

You need about 100:1 signal to noise ratio at 1 rad/hr.

What is the size of the sensitive volume in your CMOS device?

Bob S

 

[Bob S]

I believe using current integration is out of the question, because you will need a leakage current in the range of fempto-amps, (which is out of the question, but please check), so let's explore pulse mode.

For gamma detection, say a Cesium-137 gamma, it will convert by producing a Compton electron, which will have on average an energy of about 300 KeV and a range of about 0.1 grams/cm² or about 0.04 cm (400 microns). It will produce about 100,000 (electron) ion pairs, or about 250 ion pairs per micron. So for a 5 micron epilayer, there will be nominally 1200 ion pairs, or 2 x 10^-16 Coulombs. If you can get the capacitance of te epilayer down to say 10 picoFarads, this would produce a 20 microvolt pulse. This is very small; could you reliably detect it? Please check the CMOS junction capacitance for me.

Doesn't the epilayer thickness of a reverse-biased junction vary as the square root of applied voltage? How much voltage can you apply, and what maximum thickness can you get?

Once you work these questions out, here is what we will need to get the quantitative counting rate per millirad per hour. What kind of sensitivity do you want (need) (millirads per hour)?

Photon yield per millirad (specific gamma constant)
http://researchcompliance.uc.edu/radsafety/isotope/isds-cs137.html

and

Photon attenuation (conversion) cross sections in silicon
http://physics.nist.gov/PhysRefData/XrayMassCoef/ElemTab/z14.html

Bob S

 

[swl]

Not an expert, but I know Geiger counters are now much mor affordable in Japan. New Chinese junk and old Russian relics are available for under $300 on auction now. I bought an old (NOS) Russian unit and it seems to work well enough for checking around the house, schools and parks.
I've got an old IPhone 3G, so let me know if you want me to try taking pictures of radioactive stuff in the dark.

BTW: Is there a cheap scintillating material that could be placed over the camera lens to improve the sensitivity?

 

[jim hardy]

Not an expert, ...

...BTW: Is there a cheap scintillating material that could be placed over the camera lens to improve the sensitivity?

when i was a grade-schooler in early 50's my chemistry set included a "spinthariscope", a gizmo that you looked into and could see flashes of light from radioactive decay.
Apparetnly they use zinc sulfide to make the radiation events emit visible light ...
http://en.wikipedia.org/wiki/Spinthariscope


a quick google finds lots of info, including this youtube on how to make one.
http://www.unitednuclear.com/index.php?main_page=index&cPath=2_12

worth an experiment?

 

[swl]

Looks like it has been done with Geiger Camera...

http://www.youtube.com/watch?v=k0B2wfS1GH0&feature=related

I hope your's works better.

 

[Bob S]

In my previous post #4, please replace "electron-ion" with "electron hole".

Bob S