Accidental LED fabrication

[Grenadier] had a piece of silicon carbide sitting around that he planned to use when making a primitive diode called a Cat’sWhisker Diode. While probing he noticed that one of the crystals threw off a bit of light. He popped it off and used JB Weld to attach it to a brass plate. The peculiar thing is that it generates light when power is run through it both forward and reverse biased. So what’s going on here? According to an informed discussion on the phenomenon there’s actually a pair of diodes in series but with their polarity reversed.


  1. grenadier says:


  2. bluewraith says:

    The most exciting phrase to hear in science, the one that heralds new discoveries, is not ‘Eureka!’ (I’ve found it!), but ‘That’s funny…’ -Isaac Asimov.

  3. Migsantiago says:

    He should be careful… what if the light also includes UV or infrared lights?

    Anyways, good accidental discovery.

  4. osgeld says:

    similar to how LED’s were first noticed back in the day

  5. fartface says:

    Someone repeats what happened at HP labs 100 years ago.

  6. DarkFader says:

    not in parallel?

  7. TheTechninja says:

    I think you mean that the pair of diodes is in Parallel with each other. Unless they are Zener-Diodes, having two in series with opposite polarities would just mean that current would never make it through.
    I am imagining something like this:

  8. This phenomenon was reported May 5, 1861, in letters to Michael Faraday:

    Also, many blue and UV LEDs are made from Silicon Carbide.

  9. Novajeeper says:

    The original red ring of death.

  10. featheredfrog says:

    seems to me two opposite-polarity diodes of any type in series wouldn’t pass current in either direction, so, really? “Series” and not “Parellel”?

    props to DarkFader

  11. Eli-0 says:

    @DarkFader – The description is correct. The site mentions one of the diodes has it’s reverse breakdown voltage exceeded (allowing it to conduct in the reverse direction).

  12. Dr. Edison says:

    I noticed a similar thing 40 years ago when I sawed the top off of a TO-5 transistor and reverse biased the base– the die lit up white! I thought it might have been the metallization going incandescent, but nothing melted– so perhaps it was luminescence. There is still much we don’t know about physics… stay curious, my friends!

  13. alankilian says:

    One of the first comments on the “Informed Discussion” link says:

    “What you have is two schottky diodes (metal semiconductor junction) in antiseries. Each metal-semiconductor contact forms a diode, which can point in either direction depending on whether the work function of your contact is greater or smaller than that of the SiC. In either case, when sufficient bias is applied, one will enter reverse breakdown, and the other will then be forward biased and conducting current, and then able to emit light.”

    That explains how they can be in series and still have one forward biased with enough current to light up.

  14. chango says:

    Needs Jeri’s perforated aluminum electrode

  15. Alan says:

    I thought parallel too, but the informed debate suggests they’re in series and breaking down in reverse bias. So yea, zener like diodes, or just plain big and crude enough to take it.

  16. grenadier says:

    @Rob Wentworth

    That is awesome, thanks for the link.

  17. Alex Rossie says:


  18. threepointone says:

    Are you kidding me? This is blackbody radiation. At 24W you have enough probably based on the thermal characteristics of the material to light up the contacts. The Schottky diodes are indeed relevant, and they affect what area gets heated most, but seriously: look at the color. THIS IS BLACKBODY! hackaday, can you use a bit of discretion when calling everything informed?

    To be fair, even researchers make this mistake at first–you get really excited that something’s emitting light, and then you look at the spectrum. . .and realize it’s just blackbody.

  19. grenadier says:

    Yeah, this is not black body. The Light is emitted instantly; as soon as the power is applied. No waiting period for it to get hot.

    Not only that, I used some canned air to chill the LED and it still emitted light, at a different color too.

  20. grenadier says:

    A video as proof. Not sure how to embed so here are a couple codes. Don’t know if any of them will work.

  21. grenadier says:

    Lol, Looks like the codes worked too well.

  22. ysa says:

    “Not only that, I used some canned air to chill the LED and it still emitted light, at a different color too.”

    I’m not an expert, but doesn’t the fact that the light has a different color at a different temperature indicate that it is indeed black body radiation?

  23. Ugly American says:

    @Rob Wentworth – surely you realize that is a fiction site?

    For reality of LEDs, try:

  24. grenadier says:

    No, chilling an LED changes the color.

  25. Mikey says:

    @Edison – just about every active component in electronics either gives off heat or light in some form. So it’s no surprise that some random component in a TO-220 package gave off light. Parts inside your computer’s CPU probably give off light as well (though there are enough that give off heat that it needs active cooling obviously.)

  26. George Johnson says:

    This is the kind of “playing around” that makes life long electronics “guru’s.”

    This is great, I love this kind of stuff.

    A lot of people don’t know they can open the metal can of a common transistor and see it throw off light too in many cases.

    I was amazed when my boss (at a electronics R&D place) didn’t know that regular CCD’s were sensitive to IR light (and that a filter is needed to block it).

    Cool stuff!!

  27. strider_mt2k says:

    The jig he made for it also looks kinda old school sciency too.

    Really cool stuff.

  28. i did this too a while back, got yellow AND green light from the same crystal using a piece of conducting thread as the contact.
    the wavelength emitted depends on the number of hydrogen atoms within the SiC lattice IIRC, as this alters the bandgap.
    Interestingly other people have observed negative resistance in SiC (this was in an IEE article) and the procedure to make them requires only a piece of raw silicon, crystal of SiC and a way to heat it up so the silicon diffuses into the SiC and forms a molecule thick barrier which is the tunnel junction.

  29. Philippe says:


    A century ago? Really?

    HP was founded in the 1939 and LEDs didn’t turn up at HP labs until 1966.

  30. Michael says:

    Another possibility is that it’s actually incandescing, does it say that it isn’t heating? A semiconductor in contact with a metal with produce heat on one end and will cool on the other. This is unlikely but a possibility. The diode scenarios posited seem unfinished at best to be sure though.

  31. ChipsAhoy says:

    A pair of diodes in series but with their polarities reversed? Sounds like your run of the mill transistor to me. As Dr. Edison said, he noticed the same thing with his TO5 transistors, though he reverse biased the base. Amplifier operation of the transistors would always have the base-collector junction reverse-biased, while the base-emitter junction forward biased.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

Join 96,353 other followers