Homework Help: Time-independent Schrodinger equation in term of the TDSE

1. Jan 6, 2015

rwooduk

1. The problem statement, all variables and given/known data
Write down the general solution of the time-dependant schrodinger equation in terms of the solutions of the time-independant Schrodinger equation.

2. Relevant equations
TDSE
TISE

3. The attempt at a solution
I'm really not sure how to interpret this question, I could write the general solution to the time dependant schrodinger equation:

$$\Psi (r,t) = \sum_{n} a_{n}(t) \Psi _{n} exp (-\frac{iE_{n}t}{\hbar})$$

insert this into the time dependant schrodinger equation and show that it is satisfied, i.e. equal at both sides

OR

Does it want me to write ψ(r,t) = R(r)T(t) insert this into the time dependant schrodinger equation and use separation of variables to show that there are two solutions, one with time and one with position dependance.

Thanks for any help interpreting what the question is asking me to do.

Last edited: Jan 6, 2015
2. Jan 6, 2015

BvU

Your problem statement has independent twice.
I think what you did is fine if you reconsider the $a_n(t)$ :
you want to give expressions for them
if they are functions of t the whole idea of separation of variables is up the creek ​

3. Jan 6, 2015

rwooduk

Yes apologies it should have read:

Write down the general solution of the time-dependant schrodinger equation in terms of the solutions of the time-independant Schrodinger equation.

So if i insert

$$\Psi (r,t) = \sum_{n} a_{n}(t) \Psi _{n} exp (-\frac{iE_{n}t}{\hbar})$$

into the time dependant schrodinger equation and show that it is satisfied, do you think that's what the question is asking me to do?

also how would i expand an(t)?

4. Jan 6, 2015

BvU

Still won't work because you want $i\hbar\;{\partial \Psi\over \partial t}=E\;\Psi$ and $da\over d t$ spoils that...

 sorry, that's for the $\Psi_n$, not for $\Psi (r,t)$. Nevertheless, I think you dont want the $a_n$ time dependent. No time to dig up a reasonable argument now. Have to think that over.

Last edited: Jan 6, 2015
5. Jan 6, 2015

rwooduk

That's fine, I appreciate the help anyway. I will also think on it some more and post if I get anywhere. I look forward to any further ideas you or anyone else may have on this one.

6. Jan 6, 2015

BvU

The TISE comes about by separation of variables: split $\Psi ({\bf r},t) = f(t) \Psi ({\bf r})$. For f(t) the separated eqn becomes $\displaystyle {i\hbar\over f}\; {df\over dt} = E$.
Linearity of the SE allows coefficients $a_n$, but there is no room in f(t) for functions $a_n(t)$.

7. Jan 6, 2015

rwooduk

ahh, ok so it's what I was thinking in the second part of the original post. That's great, thanks very much for the help!

8. Jan 6, 2015

BvU

Depending on where you are in the curriculum, there's more that can be loaded onto this exercise:
showing that any $\Psi({\bf r},0)$ can be written as such a summation (i.e. that the $\Psi_n$ form a basis),
the expressions for $a_n$, etc

9. Jan 7, 2015

rwooduk

We have done both, which is why I found the question so disambiguous, if both methods show the same thing I'll go with the separation of variables method.

thanks again