5
\$\begingroup\$

I’m a complete beginner working on a simple experiment to measure the I–V characteristics of LEDs. For most visible LEDs, my results look as expected: the current increases exponentially once the forward voltage exceeds the built-in potential, and then the curve stays smooth.

However, with an infrared LED, I noticed something strange. After the threshold voltage (around 1 V), the current rises exponentially at first, but then the slope suddenly changes and the curve becomes almost linear and much flatter than the exponential trend before it.

This is the setup:

  • Infrared LED polarized directly
  • Series resistor: 500 Ω
  • Current measured with an ammeter, voltage measured across the LED with a voltmeter
  • Current limited to about 20 mA
  • Same setup used for other LEDs (red, green, blue) gives perfect curves

What could cause this sudden change of slope after the built-in voltage in an infrared LED?

P.S. The LED wasn't warm at the touch.

These are the graphs

\$\endgroup\$
6
  • 2
    \$\begingroup\$ If you happen to know it, can you please provide us the part number for the LED? Does the LED feel warm to the touch? The forward voltage of LEDs do change with temperature. \$\endgroup\$ Commented Nov 5 at 17:02
  • \$\begingroup\$ I can't right now, I was in the uni lab, but I will. The LED was't warm at the touch. @C.Dunn \$\endgroup\$ Commented Nov 5 at 17:28
  • 1
    \$\begingroup\$ @PietroSchiavone LED type diodes have non-ideality all over the place. So I can't suggest some bright-line reason between a red LED and a near-IR LED, for example. But there is a process that takes place, very near the point where the applied electron volts (applied volts times electron charge) equals the band-gap energy, that causes the non-ideality factor to momentarily decline before starting to increase again. To test this idea, you'd need to use 100 us pulses to avoid heating effects but now increase the current still further. Also, the bulk impedance varies but dominates later. \$\endgroup\$ Commented Nov 6 at 4:53
  • 2
    \$\begingroup\$ @PietroSchiavone As depletion region narrows, diffusion injection current in the quasi-neutral regions takes over from depletion region recombination. Look up dominant transport. Or recombination/diffusion crossover. But every time I've really tried to understand LEDs -- as opposed to nice and simple diodes -- I've found that there was way more to learn about than I wanted to absorb. And it would not surprise me if new behaviors emerge as the eV level declines towards atomic/molecular bonds and away from valence transition energies. Maybe IR LEDs do exhibit a more pronounced behavior. +1. \$\endgroup\$ Commented Nov 6 at 5:02
  • \$\begingroup\$ What do you mean by "curve becomes linear and flatter"? From what I can see in the second plot, it's still an exponential increase, and an even faster one at that.. \$\endgroup\$ Commented Nov 6 at 5:39

1 Answer 1

Reset to default
2
\$\begingroup\$

What could cause this sudden change of slope after the built-in voltage in an infrared LED?

Resistance in the quasi-neutral (i.e. "bulk") regions of the semiconductor. All diodes have this effect to one degree or another, but often it is only noticeable in high current pulses because continuous operation at these levels are likely to be destructive to the device.

I can't explain why you are able to observe this effect in IR LEDs but not visible. But there obviously need to be differences in chemistry between different "colors" of LED.

\$\endgroup\$

Your Answer

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

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.