How to determine Threshold Voltage for an LED

[Plexonite]

Attached is a Current Density vs. Voltage plot which is generated by one of my OLED Labview measurement systems.  I am trying to programmatically determine the threshold voltage.  When the data is viewed in log scale it is possible to observe a clear charge injection at (0.8, 0.000203).  The problem is that I can observe this value visually but I am not able to solve for it programmatically.  I have tried taking the derivative to observe slope changes but that only helps with determining the turn-on voltage.  Any ideas how I could solve for this value?

Thanks

E

0.000203

[dthor]

Perhaps I'm not understanding your data correctly... But I don't see any voltages below 4V on your plots. Where exactly are you seeing 0.8V, 0.000203mA/cm^2 ? Am I looking at the wrong plots?

Perhaps you could just provide a screenshot of the graph and where you see Vth.

[Plexonite]

Sorry wrong File

[dthor · Accepted solution]

You're on the right track, you just have to do deeper. Try taking the 2nd derivative and finding the points where it is greatest. This will show you where the Log graph is concave up and where your Vth is. I've included a plot of Log(data), D(Log(data), and D(Log(Data),2). I also zoomed in on the area of interest, so pay attention to the scales.

You might have to make some assumptions though. For example, with exactly what I posted above, you get Vth = -0.4 or so. If you assume Vth > 0, then you get Vth = 0.8.

Also, you will probably want to do some more fancy stuff such as: interpolate the data to get more points, smooth the interpolation, hack off the ends of the derivative data (they go to 0 depending on your derivative method).

[Plexonite]

Genius! Ahh I was so close.  Thanks you very much. 

You should theoretically be able to assume that Vth > 0 due to the fact that charge should not be flowing through the device in reverse bias.