I was curious what would happen if we had a free particle hamiltonian and saw what would happen if your initial wavefunction was a delta function and time evolve it using Schrodinger's Equation (say in one dimension). Here were my thoughts:

$$-\frac{\hbar^2}{2m}\partial^2_x\psi = i\hbar \partial_t \psi$$

Taking the Fourier transform, which I call $F(\cdot)$, and $F(\psi) = \tilde{\psi}$:

$$\partial_t \tilde{\psi} = \frac{\hbar i}{2m} (2\pi i k)^2 \tilde{\psi}$$

$$\implies \tilde{\psi} = \tilde{\psi}(t=0)\exp(\frac{-2\pi^2\hbar k^2 it}{m})$$

If we our initial wavefunction is a delta function (say centered at $x=0)$:

$$\tilde{\psi}(t=0) = F(\delta(x)) = \int \delta(x) \exp(-2\pi i xk) dx= 1$$

$$\implies \tilde{\psi}(k, t) = \exp(\frac{-2\pi^2\hbar k^2 it}{m})$$

Thus, in position space:

$$\psi(x, t) = F(\tilde{\psi}) = \int \exp(\frac{-2\pi^2\hbar k^2 it}{m}) \exp(2\pi ixk)dk$$

I believe this is correct, and (if it is) my main question is how do you interpret it?

  • 2
    $\begingroup$ Did you try to evaluate the last Gaussian integral? $\endgroup$
    – Qmechanic
    Aug 27, 2020 at 14:27
  • 1
    $\begingroup$ Your maximally localized function in position space is a function that is maximally delocalized in momentum space, which is why you get a constant function of k when you transform it to k space, which is nothing else but momentum space except some constants. The time evolution shows that each momentum component of your function has a phase factor that oscillates with the corresponding energy $E(k)=(\hbar k)^2/2m$. It is important to remember that these states are idealized and cannot be realized as physical state. They are useful mathematical tools but physically impossible. $\endgroup$
    – Hans Wurst
    Aug 27, 2020 at 14:43
  • $\begingroup$ en.wikipedia.org/wiki/… $\endgroup$ Aug 26, 2021 at 3:03

1 Answer 1


You've found the integral representation of Green's function for the free particle Schrodinger equation. Delta-function is not a square-integrable function, but in quantum mechanics we often consider $\delta(x)$ as the eigenfunction of the operator $\hat{x}$ in coordinate representation: $$ \hat{x}|x'\rangle = x'|x'\rangle, \quad \langle x| x'\rangle = \delta(x-x'). $$ From your question formulation follows equality $$ \psi(x,t) = \langle x | e^{-\frac{i}\hbar \hat{H} t} | x' = 0 \rangle \qquad (*) $$ Free particle problem possess translational invariance, so $$ \psi(x,t) = \langle x+x_0 | e^{-\frac{i}\hbar \hat{H} t} | x_0 \rangle $$ for any $x_0$. As $\psi(x,t)$ is the matrix element of the evolution operator, it is straightforward to express the solution to the non-stationary Schrodinger equation in the following form: $$ \phi(x,t) = \int \psi(x-x',t) \phi(x',0) dx' $$


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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