Meaning of soulution of Central Equation: Nearly free electron model

  1. Considering the Nearly Free Electron model of solids, where we assume the valence electrons of some one dimensional(!) solid to move in a weak, periodic (with respect to the solids lattice constant) potensial.

    We may derive (which I assume you are familiare with, and will not do here) the central equation as an algebraic reformulation of the time independent Schrödinger eq. corresponding to the model/potential at hand,

    [itex]
    (\lambda_k - \epsilon)C_k + \Sigma_G U_G C_{k-G} = 0
    [/itex]



    where [itex] \lambda_k = (\hbar^2 k^2) / (2m_e) [/itex], [itex] G [/itex] is the set of possible reciprocal lattice vectors and [itex] C_k [/itex] is det fourier coefficients corresponding to the solution of the Schrödinger eq.:

    [itex]
    \psi_k = \Sigma_k C_k e^{ikx}
    [/itex].


    My problem is that I do not understand what exactly we do find if we solve the central equation.

    Say for instance I solve it and find the energy [itex] \epsilon_\pm = \lambda_k \pm U_0 [/itex] for some [itex]k[/itex]. Then I am told the energy gap, [itex] \epsilon_{gap} = \epsilon_+ - \epsilon_- [/itex], between two energy bands for this [itex]k[/itex] at hand. Please (dis)confirm!?

    ... and then WHICH two bands are this gap between? (If that makes sense). And is it possible to find values for [itex] C_k [/itex], how? .. Assuming we know the periodicity of the potential and [itex] k [/itex].
     
  2. jcsd
  3. yes we can;
    in this case C is equal to: +_sgn(U)C
    u can find the exact equation in,Solid State Physics By Ashcroft&Mermin.chapter9,equation (9.29)-
     
Know someone interested in this topic? Share a link to this question via email, Google+, Twitter, or Facebook