4
$\begingroup$

In science communicator Steve Mould's Youtube video Why all solar panels are secretly LEDs, he supplies a voltage across a solar cell. Using an IR camera, we see IR emission coming from the solar cell, and he claims that is LED behavior.

Is it not more likely that the IR we're seeing is blackbody radiation? My understanding of LEDs is that indirect semiconductors (i.e. silicon) make for terrible light-emitters, and the energy will be emitted non-radiatively by a phonon.

$\endgroup$
6
  • 1
    $\begingroup$ You indeed get some IR, which can be used as a measure of how good the panel is. But, yes, the efficiency is horrible. $\endgroup$
    – Jon Custer
    Commented Jan 29, 2020 at 1:26
  • $\begingroup$ Different but related from 2016: Do III-V based photovoltaics “glow” (photo-luminesce) when illuminated but not loaded? While normal silicon diodes make very little if any light when forward biased, many/most photovoltaics in space are made from III-V semiconductors with direct band gaps, and often are multi-junction for higher efficiency. $\endgroup$
    – uhoh
    Commented Jan 29, 2020 at 1:31
  • 1
    $\begingroup$ I've just asked Is infrared photoemission from silicon used in IC manufacturing diagnostics? and linked back to your question. $\endgroup$
    – uhoh
    Commented Jan 29, 2020 at 1:58
  • 1
    $\begingroup$ As terrible light emitters as they are, they are also terrible light absorbers. Still, silicon solar panels work. Why wouldn't they be able to emit back? $\endgroup$
    – Ruslan
    Commented Jan 29, 2020 at 10:43
  • 1
    $\begingroup$ @uhoh your point in your post about it being instantaneous makes it seem clear that it isn't black body radiation, so thanks for pointing that out. $\endgroup$
    – Roy
    Commented Jan 30, 2020 at 16:51

1 Answer 1

2
$\begingroup$

Indeed, silicon does not act as an infrared LED. That is because the band gap is indirect, the maximum of the valence band and the minimum of the conduction band are not at the same point in $k$-space. Momentum must be conserved in recombination, and this is impossible because the infrared photon has negligible momentum.

This is why one uses GaAs for IR LEDs.

But he is using a lot of current, so maybe enough to get some near-IR photons anyway.

$\endgroup$
2
  • $\begingroup$ Does that mean solar panels also can't be made out of silicon? $\endgroup$ Commented Oct 29, 2020 at 17:50
  • $\begingroup$ @user253751 No, of course not. $\endgroup$
    – user137289
    Commented Oct 29, 2020 at 19:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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