I am in the 3-dimensional radial Schrödinger equation, in the spherical coordinates, where we try to find the separable solutions

$$\psi(r) = R_l(r) Y_l^m(\theta, \varphi) \equiv \frac{u_l(r)}{r}Y_l^m(\theta, \varphi).$$ ($\theta$ is the colatitude and $\varphi$ the azimuthal angle.)

I've been trying for an hour to get from this line to the next one:

$$\left[-\frac{\hbar}{2m} \left( \frac{1}{r} \frac{d^2}{dr^2} r - \frac{l(l+1)}{r^2} \right) + V(r)\right] R_l(r) Y_l^m(\theta, \varphi) = ER_l(r) Y_l^m(\theta, \varphi)$$

$$\Longleftrightarrow \left[-\frac{\hbar}{2m} \left( \frac{d^2}{dr^2} - \frac{l(l+1)}{r^2} \right) + V(r)\right] u_l(r) = Eu_l(r)$$ How did the $\frac{1}{r}$ and $r$ dissapear with $u_l(r)$?

  • $\begingroup$ Hint: Check the definition in the first equation. $\endgroup$
    – Qmechanic
    Commented Apr 23, 2022 at 7:56
  • $\begingroup$ Are you sure that the line you are starting from is right? According to Griffiths 2nd edition Eq. (4.35) the second derivative term is different. It should be $$ \frac{1}{r^{2}}\frac{d}{dr}\left( r^{2}\frac{dR}{dr}\right) $$ $\endgroup$
    – erikasan
    Commented Apr 23, 2022 at 9:16
  • $\begingroup$ I just checked in my textbook, the equation is the one I wrote. Maybe they are equivalent. $\endgroup$ Commented Apr 23, 2022 at 10:42

2 Answers 2


While it seems like you've resolved your particular issue, this might be a good opportunity to think about how operations work more generally, as you will come to see a lot of operator algebra in quantum mechanics.

Taking your first equation and substituting $R(r) \rightarrow u(r)/r$ and dividing by $Y(\theta,\varphi)$, we get \begin{equation} \left[ -\frac{\hbar}{2m} \left( \frac{1}{r} \frac{d^2}{dr^2} r - \frac{l(l+1)}{r^2} \right) + V(r) \right]\frac{u(r)}{r} = E \frac{u(r)}{r}. \end{equation} Someone who is used to regular (scalar) algebra might be tempted to simply cancel the $1/r$ factors multiplying $u(r)$ on both sides. This would be fine if not for a certain term on the left-hand side; \begin{equation} \tag{1} \label{term} -\frac{\hbar}{2m} \left( \frac{1}{r} \frac{d^2}{dr^2} r \right)\frac{u(r)}{r}. \end{equation} There are two important reasons why we can't just multiply this term by $r$ to get rid of the $r$ dividing $u(r)$.

  1. Operators are applied from the left, by convention.
  2. The differential operator $d/dr$ does not commute with multiplication by a factor which involves $r$;

\begin{equation} \frac{d}{dr}f(r) \neq f(r)\frac{d}{dr}. \end{equation} With this in mind, if we multiply (\ref{term}) by $r$ we should do so from the left (you can think of it as applying a multiplication operator), which cancels the $1/r$ factor there (scalar multiplications do, of course, commute). The remaining part then tells us to first multiply $u(r)/r$ by $r$ and then differentiate the result twice with respect to $r$, which is why we get \begin{equation} -\frac{\hbar}{2m}\frac{d^2u}{dr^2}. \end{equation}


A friend just explained me that actually, turning $R_l(r)$ into $u_l(r)/r$ made a $\frac{1}{r}$ appear everywhere, except in the $\frac{d^2}{dr^2}$ term where the $r R_l(r)$ turns into $u_l(r)$. As there was already a $\frac{1}{r}$ in front of $\frac{d^2}{dr^2}$, all the $\frac{1}{r}$ and $Y_l^m(\theta, \varphi)$ just dissapear.


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.