What is the coinciding point limit of these two parametrized terms?

  • Thread starter Thread starter Willi Tschau
  • Start date Start date
  • Tags Tags
    Integral Sum
Willi Tschau
Messages
1
Reaction score
0

Homework Statement


The story is that I would like to evaluate the coinciding point limit (when ## (x^0, x^1)→(y^0,y^1)##) of these two terms:

\begin{eqnarray*}
&&\frac{1}{2L}e^{\frac{i}{2}eE\left((x ^1)^2-(y^1)^2\right)}\left( im\left( x^0-y^0+ x^1-y^1\right) \right) \sum_{n=0}^{\infty}e^{-ik_n( x^0-y^0+x^1-y^1)}\\
&&\times \frac{1}{L}\int_{-L/2}^{L/2}dx\left( -1\right)^{n}\cos \left( eE\left(\frac{L^2}{4} -x^2\right) +\left( 2n+1\right) \pi \frac{x}{L}\right) \\
&&+\frac{1}{2L}e^{\frac{i}{2}eE\left((x ^1)^2-(y^1)^2\right)}m \sum_{n=0}^{\infty} e^{-ik_n( x^0-y^0+x^1-y^1)}\\
&&\times \frac{1}{L}\int_{-L/2}^{L/2}dx\left(-1 \right)^{n}\left(2x-L \right)\sin\left( eE\left(\frac{L^2}{4}-x^2 \right)+\left(2n+1\right)\pi \frac{x}{L} \right)
\end{eqnarray*}

m, e, E, L are respectively the mass, charge, electric field strength, length of the box, and are constant. The energy levels ##k_n##'s are given by ##k_n=(n+1/2)\frac{\pi}{L}##.

Homework Equations

& 3. The Attempt at a Solution [/B]

1. There will be no singularities since another calculation of mine has shown that.

2. When I put ## \frac{1}{L}\int_{-L/2}^{L/2}dx\left( -1\right)^{n}\cos \left( eE\left(\frac{L^2}{4} -x^2\right) +\left( 2n+1\right) \pi \frac{x}{L}\right) ## in Mathematica, I got something that decays like ##\frac{1}{n}##, when ##n## is large.

So I expect the sum will give me ##\log (1-exp...)## (appropriate dissipative term ##\pm i \epsilon## may be needed); of course there will be some residue.

3. My burning question is: is there an analytic way to expand the integral which depends on ##n## as ##\frac{C}{n}+D+...##, for some constant ##C, D##, which are to be found?

Thanks!
 
Physics news on Phys.org
The limit for the first term should be obvious from inspection.

The limit for the second term has a divergent series ##\sum_{n=0}^{\infty} 1##.

The integrals look like monsters to do analytically (although it appears possible), why not just observe the behaviour of the other terms and use some simple limit laws?
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top