Doping in semiconductors

dE_logics

Addition of pentavalent impurities to Si will result in each dopant forming bonds with 4 si atoms and, so 1 atom of the dopant will be left out.

On adding a trivalent impurity, the dopant will form a bonds with 3 Si atoms...so how does an electron disappear to make a hole?

Why does the dopant form only a single bond with silicon?...why not triple? Chemical properties I presume?

bpsbps

Assuming that we're only talking about "shallow dopants"...

You're on the right path...

With a pentavalent impurity where does the extra electron go? It is loosely bound to the impurity atom and is delocalized in the lattice in the vicinity of the impurity. The charge can be easily ionized and can conduct -- hence it is an n-type donor.

With a trivalent impurity where does the missing electron go? (Does that make sense?) Well, a negative charge is taken from the lattice and localized in the bond but that means that there is now an absence of charge, or a "hole" that is delocalized in the lattice in the vicinity of the impurity. This positive charge can be ionized and conduct -- hence it is a p-type donor.

Thinking in terms of electrons and holes is the simple way to approach the problem. You can also think purely in terms of electrons, but it is more complicated.

dE_logics

That dopant which lost the election should have a positive charge on it. But this charge is not strong enough to hold the electron tightly...so for a specific voltage, it leaves the nucleus.

Am I right?

You mean, since the position of the electron has now been changed, cause of the nucleus around, a positive charge will be imposed in the same place.

If that's true, then I suppose, there should be a negative charge formed somewhere within the covalent bond, and so there should be a shift in the positioning of other electron also (I presume that's how the holes move).

Also why does this effect not work with a donar atom...why does 4 holes not get formed when a donar atom is imposed in a semiconductor?...in that case, in this case of acceptor atom...why does it not happen that 3 holes are formed cause there will be 3 delocalization of elections?

I think the answer lies with the symmetry of the final arrangement formed with the donor/acceptor...in cause of 3 they can't cancel out.

PhaseShifter

And where would the electrons go when they left 4 holes behind?

Are you thinking of the dopant atom as an interstitial or substitutional defect?

dE_logics

The electrons are within the crystal lattice; as said by bpsbps, since holes are formed by delocalization of electrons...I think whenever a covalent bond gets formed, there is a delocalization...since there is a delocalization, there should be a hole formed by virtue of each covalent bond.

I don't know what's a substitutional defect, but it is a defect in the lattice I suppose.

PhaseShifter

If that were the case, even an intrinsic semiconductor would form four holes for every atom.

It helps to think of the crystal as a macromolecule.

The valence band as the set of all bonding molecular orbitals for the molecule, the conduction band is the set of all antibonding orbitals.

dE_logics

No!...in an intrinsic conductor the electrons are not delocalized!

Antibonding?

PhaseShifter

If the dopant atom contains only one extra (or missing) electron when compared to the atom it replaces, how would you then get multiple charge carriers?

dE_logics

Are you tying to clear out your concepts or are those questions for me?

Cause I'm trying to clear out my concepts...so I might convey wrong information...however I do know (for sure shot) the answer to your last question.

PhaseShifter

Questions for you.

dE_logics

Don't we get just 1 extra charge carrier?

However following -

There should be a total of 5 charge carriers formed...which I know does not happen.

PhaseShifter

Where are you getting that second quote? As far as I can tell, you're just making it up.

You only form holes by delocalizing electrons when thermal or optical excitation takes place, and then you form an electron-hole pair. (This is what happens in intrinsic semiconductors and separates them from insulators.)

Dopants are a different story altogether.

dE_logics

I learned that from someone here -


And this is exactly what I wanna know.

PhaseShifter

You'll notice he isn't referring to a valence electron in that paragraph, but a "missing electron".

(Although I admit his statement that "a negative charge is taken from the lattice and localized in the bond" is a bit misleading, since the network of bonds is the same thing as the lattice, if it's a thing you can remove electrons from. The valence electrons used to form the bonds aren't as localized as a Lewis diagram would have you believe, thus the term "valence band".)

dE_logics

aaa...I don't know, but I think only valance electrons can go missing...

PhaseShifter

But he's not referring to an actual electron...he's referring to the "missing" 4th valence electron from a trivalent dopant atom.

dE_logics

Ok...now I see.

The p type dopant, will just act like another silicon atom...the only difference is that it will be missing an electron.

So considering the normal structure of a lattice, there will be a missing electron around the dopant.

But again, how did a positive charge emerge in a lattice where the the number of electrons are enough to satisfy the positive charge given by the nucleus?