Problem: In the context of quantum harmonic oscillator the eigenfunctions are given by: $ u_n(x) = (N_n/\sqrt{b}) H_n(x/b) \exp\left[-x^2/(2b^2)\right] $, where $N_n$ is the normalization factor: $ N_n = \left(\frac{1}{\sqrt{\pi} 2^n n!}\right)^{1/2} $ and $b = \left(\frac{\hbar^2}{m k}\right)^{1/4}$, where $m$ is the mass and $k$ is the spring constant.

My book asks me to prove that the following equation holds, where $\varphi_n(k)$ are the eigenfunctions in the momentum space:

$ \varphi_n(k) = \sqrt{b} N_n H_n(b k) \exp(-b^2 k^2 / 2), k=p/\hbar $

Attempted solution:

norm[n_] := (1/(Sqrt[π] 2^n n!))^(1/2)

u[n_, x_] := 1/Sqrt[b] norm[n] HermiteH[n, x/b] Exp[-(x^2/(2 b^2))]

φ1[n_, k_] :=
  {n ∈ Integers, n > 0, x ∈ Reals, k ∈ Reals, b > 0},
  1/Sqrt[2 π] Integrate[u[n,x]Exp[-I k x], {x, -∞, +∞}]]

φ2[n_, k_] :=
  {n ∈ Integers, n > 0, b > 0},
  Sqrt[b] norm[n] HermiteH[n, b k] Exp[-((b^2 k^2)/2)]]

    {j, Simplify[φ1[j, k] == φ2[j, k]]},
    {j, 1, 8}] // TableForm

And here is the output:

enter image description here

My question is: Why does the relation hold only for $4k, k \in Z$ ? If I print $φ1[n,k]$ and $φ2[n,k]$ I see that in $φ1$ there are $(-i)^j$ terms, which is why the equation holds only for multiples of 4.

Does anyone have any idea at all on where the error is ? (In my book? In how I do the Fourier transform? Something else I'm missing?)

  • $\begingroup$ Hi Zet, I think this question looks more likely belong to the Physics site. $\endgroup$
    – Silvia
    Dec 10, 2013 at 4:05

1 Answer 1


Your code is ok. What is missing is the phase factor in the definition of momentum-space wavefunction $\varphi_2(n,k)$ that you use. The proper definition is:

$\varphi_2(n,k) = (-i)^n \sqrt{b} N_n H_n(b k) \exp(-b^2 k^2 / 2) ,$

as can be verified for example in these lecture notes.

Also, I've found that a slight change in your testing function is helpful in making the comparisons:

Table[{j, Expand[φ1[j, k]] == Expand[φ2[j, k]]}, {j, 1, 8}] // TableForm

Note also that for a quantum harmonic oscillator, the possible values of the quantum number $n$ are $n = 0,1,2,3,\ldots$, so you probably want also to check the equality for $n=0$.


Your Answer

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

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