0
\$\begingroup\$

In the video Why all solar panels are secretly LEDs (and all LEDs are secretly solar panels) a silicon photovoltaic cell from a calculator is purposefully damaged and then a 36 V bias is applied. When imaged with a normal silicon CCD (or CMOS) camera with its infrared filter removed, infrared light can be seen emitted from the photovoltaic cell and the damage visualized.

The pattern is well delineated and instantaneous as the voltage is turned on and off, and the device is hand-held so it's not thermal infrared; cutoff for a silicon CCD is 1.1 eV or about 1100 nm.

This sounds familiar to me. I remember hearing about the use of infrared imaging to localize defects or failures in ICs a long time ago, and I think it was explained to me that the radiation was direct emission from the silicon and not just blackbody thermal emission.

Question: Am I remembering correctly? Is direct infrared photoemission from silicon used in IC manufacturing diagnostics? If so, then since silicon is not a direct band gap material, how does it produce at least small amount of light?

Note: I was reminded of this when reading the recently-posted Physics SE question Do silicon solar cells act like an LED when you flip the voltage, or are we just seeing black body radiation?

infrared light emission from photovoltaic from Why all solar panels are secretly LEDs (and all LEDs are secretly solar panels)

\$\endgroup\$
0
\$\begingroup\$

I've used those emissions in diagnosing high-voltage breakdown. Gotta turn off the lights, or use lid that closes.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ Does this test have a name? Can you add some supporting link? Thanks! \$\endgroup\$ – uhoh Jan 29 at 3:25
  • \$\begingroup\$ I also recall this technique being used to diagnose faults in CMOS integrated circuits...possibly junction breakdown? I don't know of a common name for the technique but the legends are true. And nobody said it was easy. \$\endgroup\$ – Elliot Alderson Jan 29 at 12:50
0
\$\begingroup\$

Maybe you're asking about Photoemission electron microscopy but this isn't generally used for manufacturing. It is used for (failure) analysis.

How a semiconductor junction can produce photons is explained in the Wikipedia page.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ It's hard to successfully manufacture ICs without failure analysis on-site; FA is an integral part of manufacturing. Photoemission electron microscopy uses photons to stimulate the emission of electrons, which is unrelated to my question about light emission. Also, I'd like to know which Wikipedia page explains how an indirect band gap material like silicon makes enough photons to be seen during failure analysis. (your link explains how electrons are imaged and detected, not photons) Thanks! \$\endgroup\$ – uhoh Jan 29 at 9:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.