Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

The Case of Electronic structure of Liquid Metals

  1. Oct 5, 2014 #1
    I have been contemplating for a few days now about how one could think about the electronic structure of Liquid metals. The cases of an isolated atom or that of solid metal crystal are well known and easy to understand. Though, a google search reveals (to me) that nothing much has come out for liquid metals.
    I have had a semester of Solid state physics and have just started Grad school. Pardon my naivetè if any, in the following arguments or if they are over simplified.

    Some curious points to me :

    01. Solid metals are generally conductive : this is usually explained by invoking the idea that the conduction and valence band nearly overlap and hence even a small bit of energy is sufficient to get the electrons to flow.
    Liquid mercury is definitely highly electrically conducting. How can one expect it to even have bands, given that (atleast in the simplistic arguments ) band formation is attributed to a periodic potential - something not present in a liquid.

    02. Semiconductors like Silicon become highly conducting and acquire a metallic luster in liquid state: Strangely could the same be related to the present Question.

    03. Experimental verification : What kind of Experimentation could be undertaken to determine the same. What could be tried ?

    04. Modeling : How could one EVEN begin to model the situation ??

    Is there a strange idea lurking out there, or is it just a figment of my imagination to fail to see the obvious ?
  2. jcsd
  3. Oct 6, 2014 #2


    User Avatar
    Science Advisor

    A metal forming from the overlap of valence and conduction band is not the generic case. Rather the conduction band is simply not completely filled. In deed, band structure is rather secondary to explain metallic behaviour. Rather you can start from a free electron picture. Maybe you have already seen the Lindhard dielectric function of a free electron gas in your class. From it you learn that the electron gas tends to shield electric disturbances on an atomic length scale. Especially the electrons near the Fermi energy don't see the full potential of the ionic cores but only a shielded version of it, so that the interaction with the cores is effectively rather weak. So the interaction of the electrons with the ionic cores in a liquid metal can be handled by perturbation theory and there are powerful diagrammatic techniques to do so.
    Nevertheless there are interesting phenomena, like Anderson localization.

    I think the following thesis is quite interesting, also for localizing the original literature:
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook