0
$\begingroup$

Does the gauge field, $\left(\frac{y}{x^2+y^2},-\frac{x}{x^2+y^2}\right)=\nabla\arctan(x/y)\equiv\nabla f$ has effect on Schrödinger equation

$$\left[\left(\hat{p_x}-\frac{y}{x^2+y^2}\right)^2+\left(\hat{p_y}+\frac{x}{x^2+y^2}\right)^2\right]\psi=E\psi\,?$$

Since we can perform a gauge transformation $A'\rightarrow A+\nabla f,\psi'\rightarrow e^{if}\psi$ to gauge out the field and lead to the equation below:

$$[\hat{p_x}^2+\hat{p_y}^2]\psi'=E\psi'\,?$$

$\endgroup$
3
$\begingroup$

I am assuming that all charges of interest in the problem are $1$.

You can gauge out the field provided that you can find some $f$ such that $\nabla f=A$. Now, in your case, there is no such $f$ because the $f=\rm{arctan}(x/y)$ is not well defined when $x=y=0$. This is why the gauge field is not trivial.

More precisely, $f$ exists everywhere except at the origin. At the origin $B=-\nabla \times A=2\pi \delta(\bf{r}-0)$ (this you can get by applying Stokes theorem). Thus the gauge field configuration corresponds to an infinitesimally thin flux tube through the center carrying $2\pi$ flux.

However, if you want those solutions of the Schrodinger equation that vanish at the origin (Have not checked this, I think self-adjointness of the domain+operator would require that this is true), then you can restrict the Schrodinger equation to multiply connected domain $\mathbb{R}^2-\{0\}$, in which $f$ exists. Here the gauge field can be removed after a transformation.

You might find the following paper interesting (Byers and Yang PRL 1961): http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.7.46

and this wiki entry: https://en.wikipedia.org/wiki/Byers-Yang_theorem

Also pay attention to the periodicity of $f$ in the azimuthal direction. The flux at the center is an integer multiple of $2\pi$ and therefore $f$ is single valued. If the flux were not single valued, you would have ended up with the same Schrodinger equation, but the solutions $\psi$ will have to satisfy twisted boundary conditions due to the non-periodicity of $f$ in the azimuthal direction.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.