What is the significance of the contact potential in metal-metal interfaces?

[VortexLattice]

Hi all,

I'm trying to find resources for how to calculate and find what I should expect as the result of the interface between two metals with different work functions.

Anderson's[/PLAIN] Rule tells us how to do this for two semiconductors and the Schottky-Mott Rule tells us how to do it for a metal-semiconductor interface (and those articles both mention that in reality those rules rarely actually predict something close to reality).

All I can find for a metal-metal interface (assuming there are no oxide layers between them or anything) is this article, but it's not clear to me if that's a very oversimplified model or not.

Does anyone have any suggestions, or know of any programs that could calculate this?

Thank you!

 

[geologic]

For starters, can you limit these theories to the metal-metal case? Maybe make the gap go to zero?

Definitely you expect impedance from the differing crystal structures and gap between the materials (although it may be microscopic).

 

[VortexLattice]

Well, maybe that would work, but metals are fundamentally different from semiconductors... For example, if the top of the valence band is now 5eV above the bottom of the conduction band (so they overlap), would you say that the band gap is now 0, or maybe -5eV? I kind of suspect you can't naively just apply a simplified version of these.

But I think what I'm looking for is this. However, I'm still unsure about a bunch of stuff about it. So the basic idea is, when you connect two different metals with a conducting wire, a tiny bit of charge flows from one to another until the Fermi levels (not equal to the chemical potentials!) are equal.

So, is it true that you have this strange situation where there is a measurable voltage across the two metals, but charge isn't constantly flowing (obviously) due to that?

So, what happens if you just apply an exterior voltage to this setup? Because there is a voltage by default across the metals, would you see no current until you got to some threshold voltage?

 

[ZapperZ]

In the Kelvin probe method, the physics of metal-metal interface is central to the ability to measure the work function/surface potential. So maybe you might want to try looking into that technique, because it tends to cover the physics of metal-metal interface.

The brief summary of it here is that, similar to the PN junction case, the Fermi levels of the two metals will align themselves, creating a potential gradient across the metal-metal junction.

You may try looking at this paper for a start.

http://kummelgroup.ucsd.edu/pubs/paper/110.pdf

Zz.

 

[VortexLattice]

In the Kelvin probe method, the physics of metal-metal interface is central to the ability to measure the work function/surface potential. So maybe you might want to try looking into that technique, because it tends to cover the physics of metal-metal interface.

The brief summary of it here is that, similar to the PN junction case, the Fermi levels of the two metals will align themselves, creating a potential gradient across the metal-metal junction.

You may try looking at this paper for a start.

http://kummelgroup.ucsd.edu/pubs/paper/110.pdf

Zz.

Hi, thank you for the reply, I read the relevant part of that. But there's still something I don't really understand, what is the significance of that contact potential, even if you can measure it? Does it affect the two metals if they are part of a circuit?