So the differential equation looks as follows:

$$i \hbar \frac{\partial \psi}{\partial t} = - \frac{\hbar^2}{2m} \Delta \psi$$

where $\hbar, m > 0$, $\psi(t,x) \in \mathbb{C}$, $t > 0$, $x \in \mathbb{R}^3$ and

$$\psi(0,x) = \exp(-|x|^2).$$

I think this can be solved elegantly using a spatial Fourier transform. I know that:

$$\hat{\frac{\partial \psi}{\partial t}} = \frac{\partial \hat{\psi}}{\partial t},$$

but how do I calculate

$$\hat{\Delta \psi} \; \; $$

and then use it to solve the PDE?


You shouldn't do a Fourier in the time coordinate just in the position coordinates (i.e. x y z). And you know by simple calculations that $\hat{\Delta \psi}$ is just $-k^2\psi(t, \vec{k})$ where $\vec{k}$ is (x, y, z) after the Fourier transform. And the by substituting this to the equation you get a simple ode for $\hat{\psi}(t, \vec{k})$ because $\hat{\frac{\partial \psi}{\partial t}} = \frac{\partial \hat{\psi}}{\partial t},$ when you transform in the position coordinates as you said.

| cite | improve this answer | |
  • $\begingroup$ Thank you. Could you clarify on how to calculate $\hat{\Delta \psi}$? $\endgroup$ – maths_student Jan 26 at 11:16
  • $\begingroup$ Just plug in the fourier transform df/dx and get -ik$\hat{f}$ and the other coordinates are the same. To get the laplcian just apply the derivative twice to get $(-ik)^2 = k^2$. $\endgroup$ – Joe Jan 26 at 16:14
  • $\begingroup$ In addition after you plug in df/dx do integration by parts to get what I said $\endgroup$ – Joe Jan 26 at 16:47

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.