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

How do free electrons impact reflectivity

  1. Mar 22, 2010 #1
    How do free electrons (metals) impact reflectivity. Why are they more reflective than another surface without free electrons (valence).

    Does this mean that a medium with no valence electrons will be 100% non reflective.

  2. jcsd
  3. Mar 28, 2010 #2
  4. Mar 29, 2010 #3


    User Avatar
    Science Advisor

    Well, free electrons will shield any electric field (faradays cage) so that the electric field has to vanish at the surface of a conductor. This is only possible for a 50:50 superposition of forward and backward travelling waves which corresponds to (almost) perfect reflectivity.

    In general, any deviation of the refractive index from its vacuum value 1 will also lead to reflectivity, so free electrons are not necessary for reflection.

    Furthermore, the reflectivity depends on the angle of incidence. So any material becomes a nearly 100% reflector for almost 90 degree angle of incidence.
  5. Mar 29, 2010 #4
    Well.. I was looking for reflectivity for radiation, not the static case.

    I think it has something to do with "free" electron can match the incomming radiation pretty much 100% up to the plasma frequency which is derivable via drude model. Above this, a metals reflectivity of radiation starts to fall off

    For the case with No free electrons, I'm assuming that the atomic bonds (through dampening) prevent frequency matching , so reflectivity and absorption only occur for resonant frequencies.
  6. Mar 29, 2010 #5


    User Avatar
    Science Advisor

    I did not refer to the static case either. The absolute value of the dielectric constant of a metal is quite large for frequencies well below the plasma frequencies (it falls approximately off from its static value with an inverse power of frequency) so that the electric fields inside the metal (and at the surface) is generally small, which is sufficient to explain the high reflectivity.

    While it is true that absorption occurs only for resonant frequencies, this is not true for reflectivity. Reflectivity depends on both the real and the imaginary part of the index of refraction, while absorption depends only on the imaginary part. However, although the real and imaginary part of the index of refraction are related by a Kramers Kronig relation, the latter is non-local in frequency space so that the index of refraction can deviate considerably from its vacuum value for frequencies far off resonance.
  7. Jun 9, 2010 #6
    I think in case of metals where we have free electrons available for conduction, when EM radiations fall on the surface of metals, electrons very close to Fermi surface vibrate thus and in this process they emit radiations. This why metals have high absorption and reflectivity and low transmission. That is to say at particular frequency, which might be their optical band gap, transmission falls steeply before and after that.
  8. Jun 27, 2010 #7
    Materials having free electron are good reflectors because the free electron can oscillate under the influence of the incident light's electric field. The oscillating electrons then re-emit those radiations with the same frequency. This is the reason for their high reflectivity.
    The re-emitted light is scattered in 2Pi direction. If the surface is smooth, these reflected wavefronts interfere constructively along the direction of reflection (according to laws of reflection). If the surface is rough, the addition is not coherent and the reflection is diffuse.
    For material with bandgap greater than the energy of the incident photon, light gets transmitted and the material will appear to be transparent is there are no absorbing centers (defects etc.).
    Metals do not have optical band gaps.
  9. Jun 27, 2010 #8
    Metals do not have optical band gaps.[/QUOTE]

    i think metals do have optical gaps which should not be confused with semiconducting gap however they might be very narrow. e.g. in rare earth metals 4f core electrons do not take part in conduction or bonding but they are active to make optical transistions.
  10. Jun 27, 2010 #9
    I am not sure if we can call them band gaps. Those are essentially atomic levels.
  11. Jun 27, 2010 #10
    absolutely, atomic/molecular orbitals are fundamentally same thing, probably that is why distinction is made by writing "optical" or "semiconducting" .
  12. Jun 27, 2010 #11


    User Avatar
    Science Advisor

    Yes, e.g. the red colour of copper is due to interband transitions from the d-band to the conduction band. This is a true absorption phenomenon, in contrast to reflectivity in case of an " ideal " metal. An ideal metal is non-absorptive as it reflects all the intensity. In the case of copper or gold this has a strange consequence, i.e. the reflectivity of copper or gold is lower in regions where the metal is absorbing and transmittion increases there. This can be seen in some light bulbs which are covered with a thin sheet of copper or gold. The light which passes trough the metal is of greenish colour. That´s where absorption is highest.
  13. Jun 27, 2010 #12
    Ya correct. However, the definition of "ideal metal" can be different in different contexts. Like Cu can be an ideal metal while discussing conductivity.
    Agreed with the discussions above.
    But I think there is no difference between the "semiconducting" and "optical" band gap. I am saying this because again I do not think that the atomic levels and bands are the same thing. The rules and absorption processes are different for the case of continuum absorption and discrete absorption. The absorption spectra are completely different in both cases. Lots of concepts that hold for the case of atomic levels lose their meaning for bands.
  14. Jun 27, 2010 #13
    I would like to know what concepts you are refering to. i have encountered such materials which do posses band gaps (optical) but their temperature vs resistivity plots clearly show that these are not semiconducting gaps. however one can speculate that there are defects or heavy doping. none-the-less i guess i dont have any confusion about this but happy to discuss it further.
  15. Jun 27, 2010 #14
    I was referring to the properties like the angular momentum quantum numbers. Even the idea that a particular state is due to certain atom/ion may not hold good which is actually the whole idea behind calling them "bands". The spread in energy (causing bandwidth) will be another obvious answer.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook