An apparatus was constructed in the lab to measure charge carrier lifetime.

  • An LED illuminates an n-typed germanium semi-conductor with a square function with a constant frequency (the period of the illumination is such that the charged carrier density reaches saturation).
  • A constant current flows through the semiconductor so that the voltage on it is measured time dependent using a scope.

voltage on semiconductor vs. time

Using the described apparatus, an exponential decay of the voltage is measured each time the LED is turned off, and its life time is measured.

When the current that flows through the semiconductor is increased, the lifetime is changed (the negative current part can be ignored for this question)

life time vs. current

My question is how can one explain this phenomenon? I might be able to describe the decrease in life time, by knowing that the temperature increases with current, hence electron density increases which lowers the lifetime (more electrons for the holes to perform recombination).

But the part i really can't understand is the increase in lifetime at the low current values. Any insight will be valuable, thank you.

*EDIT - it may be important to mention that the experiment was reproduced on 3 different samples.

  • $\begingroup$ I don't have a real answer, but consider 1. other sources of illumination on the sample. 2. Is the sample small enough that some carriers are swept to the contacts before they can recombine? 3. Any other recombination mechanisms (surface states, Auger, ...) that could be confusing your curve fitting? $\endgroup$ – The Photon Jul 17 at 16:15
  • $\begingroup$ Also, is the voltage signal really pure AC as shown in your 1st chart? that seems very strange for the experiment you described. $\endgroup$ – The Photon Jul 17 at 16:15
  • $\begingroup$ 1. There IS a light source, but it operates at a period of 2e-2 sec which is much larger than the lifetime. 2. I did check it. The sample's length is about 1cm, and the drift velocity makes the holes move about 1mm for the LED period. 3. There might be other recombinations affecting this, but won't they just make up an effective lifetime so that it will still be approximately an exponent? $\endgroup$ – dor00012 Jul 17 at 16:56
  • $\begingroup$ In my experience you can get quite large errors in the time constant when trying to fit an exponential decay from imperfect time domain data. I don't know if that's what's going on here, but it's something to watch out for. $\endgroup$ – The Photon Jul 17 at 17:05

Your Answer

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

Browse other questions tagged or ask your own question.