Howdy, y’all! I bet you thought you’d seen the last of me! To be perfectly honest, I thought so, too—I’ve been trying to draw illustrations of octahedral and tetrahedral interstitial sites for the past fifteen minutes, with basically no success. Because of that, I’ve returned to the part of inorganic that I’ve alluded to but glossed over for the past week or so: semiconductors!
Boring, you say? I thought so too, but then I really thought about it, and, well, I like my computer (and my phone and my calculator and…)
In my post on metals, I talked about why metals conduct electricity. It has to do with that lovely little thing called band theory, which my professor referred to as “solid state physics.” Altogether now! Uuuuuughh…
Now, if we undergraduate (or high school or…) students had our way, we would call it quits there and say, “Okay! Metals are hard! Everything else is nonmetal!” However, because the world is big and beautiful, it doesn’t quite work that way. You know from looking at the periodic table that there are a handful of these lovely, cringe-worthy things called “metalloids,” and, as you expect, they can’t make up their minds. Although indecisiveness is generally called something like “irresponsibility” or “you just want go wherever you’ll get food” in my house, when it comes to elements, these actually serve a pretty good purpose.
Take that, leadership workshops!
Yup, the metalloids are a very important class of materials called “intrinsic semiconductors.” Now, what does that mean? I’m so very glad you asked, because I have to tell you regardless. (I have a test tomorrow, you know.)
Well, semiconductors are solids with an energy gap between their filled band and their empty band. This is pretty much what you’d expect from something straddling the “metal” fence: metals have a single partially-filled band, nonmetals have separate orbitals, and metalloids have separate bands.
Now, the key here is that, while semiconductors do have separate bands, they aren’t that far apart in energy. Ideally (at 0 Kelvin), all of their electrons are in their lower-energy band (the valence band) and their higher-energy band (the conduction band) is completely empty. However, as you start adding energy to the semiconductors with heat, electrons gain enough energy that they can hop the gap.
This sounds really useful at first pass, but if you think about it, it’s tragically inconvenient; yes, semiconductors are very useful for use in electronics and such, but imagine if the conductivity of the semiconductors in your computer varied wildly with temperature. It’d make things a little hard to manage, really. For this reason, the only truly useful purpose that intrinsic semiconductors serve is to measure temperature—they’re put in temperature-monitoring devices called thermistors, in fact.
Add some impurities in, though, and it’s a whole different ball game. This is called “doping,” and it alters the electronics of the semiconductors just enough to make them actually useful and reliable at the temperatures they’ll be used in. This produces “defect” semiconductors of two types: p-type and n-type.
P-type semiconductors result from doping an intrinsic semiconductor with a relatively electron-deficient atom. For example, add boron to silicon, and you’ve got yourself a p-type defect semiconductor. These work by introducing atoms with empty energy levels that lie just above the valence (full) band in the semiconductor. Electrons can easily hop from the valence band to these isolated energy levels, generating positive (thus the p) holes in the valence band that will conduct charge.
N-type semiconductors, in contrast, are produced when a pure semiconductor is doped with an electron-rich atom, like putting phosphorus in silicon. This introduces filled energy levels just below the conduction band of the semiconductor; electrons easily hop from the dopant to the conduction band of the semiconductor, generating negative (n) charges that conduct charge.
It’s very important to note that p-type semiconductors are higher in energy than their respective n-type semiconductors. This is because the electron-deficient dopants experience a smaller effective nuclear charge than the semiconductor, whereas the electron-rich dopants experience a higher effective nuclear charge than the semiconductor.
(I can’t explain this better, unfortunately. The subject has been overrun by physicists online. If you find a good resource, let me know! Please! Pls. Pls.)
Okay, so that’s great, but what about it? Well, turns out, defect semiconductors have some awesome applications. If I have time, I might get into it. First, though, I have to go back and review crystal closest packing, which is sure to be oodles of joy.
Questions? Comments? A good (chemical??) explanation of band-bending across a p/n junction? Put ‘er down there, please!