$$ \psi_E(x)=\begin{cases} e^{ikx}+Ae^{-ikx} \quad \quad x \le0 \\ Ce^{ikx} \quad \quad x\ge a \\ \end{cases} $$ where $k^2=\frac{2mE}{\hbar^2}$.

Now what I read in my notes is

"since the eigenfunctions of SE equation must not be equal to zero in a point with their first derivatives, then $C \neq 0$".

How can I prove this statement? Is due to the fact that the SE is $\psi''(x)=-\frac{2mE}{\hbar^2}\psi(x)$ so that the eigenfunction must have concavity facing upward when it's negative and concavity facing downward when it's positive?


1 Answer 1


The eigenfunction of the 1D Schrödinger equation satisfy a second order linear differential equation. If there is a point $x$ where $\psi(x) = \psi'(x) = 0$, then $\psi$ is zero everywhere, which is not allowed.

  • $\begingroup$ I can't understand. Why would be zero everywhere? I just have a Cauchy's problem that "select" one of the solutions of the TISE. $\endgroup$
    – Salmone
    May 31, 2022 at 11:27
  • $\begingroup$ The constant zero function is a solution of the TISE and has the same Cauchy data $\endgroup$ May 31, 2022 at 11:54
  • $\begingroup$ Ok thank you. Are we assuming that there is a unique solution for Cauchy problem? Otherwise, why the fact that the constant zero is a solution of the TISE with same Cauchy data would be a problem? I mean: couldn't there be other functions that solve the TISE and Cauchy problem other than zero? $\endgroup$
    – Salmone
    May 31, 2022 at 15:30
  • 1
    $\begingroup$ In your particular case, the solutions are of the form $\psi(x) = A e^{ikx} + Be^{-ikx}$. It is not hard to see that the Cauchy problem has a unique solution. In general, you may need to assume some regularity for the potential $V(x)$. $\endgroup$ May 31, 2022 at 16:16

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.