For this week's HW perhaps you could write a two page report outlining what you have learned in this class, (including what you expect to learn about lasers and photo-detectors next week). You can also post outlines of this here or comments, questions etc. related to this matter.
Also, suggested problems for the final are encouraged.
Subscribe to:
Post Comments (Atom)
Sorry if this is obvious, but I am trying to figure out why drift current doesn't compensate the diffusion current under forward bias.
ReplyDeletePage 173 discusses the importance of a supply of minority carriers - the drift current sensitive to this quantity.
Wasn't our whole argument that there was an excess of minority carriers? Why do these preferentially recombine outside of the depletion region as opposed to getting swept back across the junction?
I think the minority carriers participate in the drift current and do not depend on forward bias of reverse bias. I think the drift current depends on the amount of minority carriers, so I am guessing the more minority carriers you have do to thermal excitation, the greater the drift current. Since the diffusion current does depend on bias, a forward bias can cause the diffusion current to overwhelm the drift current unless you obtain more minority carriers to counteract it. I am not sure if I am right but this is what I think. Hope this helps
ReplyDeleteEthan -
ReplyDeleteI'm still kind of confused on this too, but here's my take.
Under forward bias, more recombinations occur just outside of the depletion region because there are more minority carriers there than there "should be" - that is, there are more minority carriers there than there would be without a bias. $\tau$, the average lifetime of a carrier, is shortened in this situation, which means more recombinations are happening in that region. Also, as far as I can tell, this excess is caused in the first place by the diffusion current blowing up under a forward bias - I'm not sure if that's right though.
Anyway, minority carriers are not swept back across the junction because there is actually a higher concentration of minority carriers in the depletion region than there is outside of it - in other words, there are more holes per cm^3 in the depletion region of the n-type material than there are in the bulk of the n-type material. So, the diffusion current will push them away from the junction and not back towards it, so as long as it is greater than the drift current (like under forward bias).
Rodney -
Each type of carrier has both a drift and diffusion current associated with it; neither overall current is only determined by the minority carrier concentration. Also, thermal generation is assumed constant, and shouldn't be affected by an applied bias (in our model, anyway.)
The diffusion current increases exponentially with bias, so it quickly overwhelms the drift current opposing it, like you said. Interestingly, though, it seems the minority diffusion is what really makes a p-n junction work - so having more minority carriers wouldn't stop the current. In fact, it's the excess of minority carriers immediately outside of the junction that seems to drive the whole thing by encouraging more recombinations and allowing more current to flow into the device.
Anyway, that's my understanding of it - if any of that is not correct, please let me know. I'm probably going to be perpetually unsure on this.
Hm, maybe it does have something to do with tau... and the [shorter?] lifetime associated with penetrating into the neutral regions. But wouldn't being swept by the field be a pretty quick process?
ReplyDeleteI guess it still doesn't make sense to me why we say that there are a wealth of minority carriers to participate in diffusion, but not drift. Both drift and diffusion depend on the amount of carriers, right? Apparently, only the thermally generated carriers are available for drift...?
Other books have very complicated looking diagrams revealing which current dominates in which region. I was hoping we had a way of deconstructing this that I didn't catch.
Looking back now, I realize there are so many things that I could know so much better and there are almost too many questions to ask. This might be a wonderful HW to assign one week earlier. This would give a chance for people to really discover what they don’t understand about these topics and allow discussion to try and solidify(haha get it) some things.
ReplyDeleteI agree, tthompso. I have some questions even going back to "q" at the beginning; maybe I should just retake the class!
ReplyDeleteDave:
ReplyDeleteQuestions about q are probably best left for solid state theory, 155--I highly encourage you to take that course next year if you can. It is very interesting and helps explain much of what we did in this course.