1
$\begingroup$

Given is the potential :$V(x) = \frac{-\hbar^2}{m}D\delta(x+a) - \frac{-\hbar^2}{m} D\delta(x-a)$ with $a >0$ and $ D > 0$. A stream of particles from the positive $x$-axis are falling towards the potential and I want to calculate the probability of transmission. I would need help with the boundary conditions. My Ansatz is: $\psi_1(x) = Ae^{ikx}+Be^{-ikx}$ for the incoming wave, $\psi_2(x) =Ce^{ikx}+De^{-ikx}$ for the wave between the delta distributions and $\psi_3(x) = Ee^{ikx}+Fe^{-ikx}$ for the wave function $\psi > a$ with $F = 0$. These are my conditions until now: $\psi_1(-a) = \psi_2(-a)$ $\psi_2(a) = \psi_3(a)$ What are the other conditions? I was thinking of calculating the transfer matrix first and afterwards the transmission.

$\endgroup$

1 Answer 1

1
$\begingroup$

Ok your ansatz about the form of the $\psi$. Now, the other conditions you need come from the discontinuity of the derivative through the delta potential. If you take the time-independent Schrödinger equation and integrate it around a delta (see, for example, here for the steps), then it can be shown that in the case of a potential like $V(x)=\gamma\delta(x-x_0)$ the derivative must possess a discontinuity given by $$ \left.\frac{d\psi}{dx}\right|_{x\to x_0^+} - \left.\frac{d\psi}{dx}\right|_{x\to x_0^-} = \frac{2m\gamma}{\hbar^2}\psi(x_0) $$ in your case $|\gamma|=D\hbar^2/m$ and hence for the two $\delta$s you have $$ \left.\frac{d\psi}{dx}\right|_{x\to -a^+} - \left.\frac{d\psi}{dx}\right|_{x\to -a^-} = -2D\psi(-a) $$ and $$ \left.\frac{d\psi}{dx}\right|_{x\to +a^+} - \left.\frac{d\psi}{dx}\right|_{x\to +a^-} = +2D\psi(+a) $$

Since you have 6 constants $A,B,\dots,F$ to determine, these 2 conditions plus the 2 you wrote plus the $F=0$ will leave us with just one free parameter that should be absorbed in the computation of the reflection and transmission probabilities.

Edit: in the last line I originally wrongly wrote $\psi(-a)$ instead of $\psi(+a)$, now it is corrected.

$\endgroup$
4
  • $\begingroup$ Why do you have $+2D(-a)$ when $/gamma$ is $\frac{-D\hbar^2}{m}$? When I try to solve this, I get $A = -Be^{2ika}$ can you confirm this? I get $B=D$ and $A=C$ by Equating the coefficients and I'm not quite shure if this is allowed. I'm also not shure which $\psi$ you mean after the equal sign when you say $ \frac{d\psi}{dx}_{x\rightarrow a^+}-\frac{d\psi}{dx}_{x\rightarrow a^-} = -2D\psi(-a)$ In fact I thought I can choose which Ansatz I like for that $\psi$. $\endgroup$
    – gamma
    Commented Feb 27, 2018 at 14:45
  • $\begingroup$ I made a typo sorry: in the last line it is a $+$ and not a $-$ sign so that it is $+2D\psi(+a)$. Probably the results you found are due to this. I will correct the answer. $\endgroup$
    – ndrearu
    Commented Feb 27, 2018 at 14:48
  • $\begingroup$ But still I don't understand why it's $+2D\psi(a)$ instead of $-2D\psi(a)$? $\endgroup$
    – gamma
    Commented Feb 27, 2018 at 15:14
  • $\begingroup$ In the potential you wrote there is a minus between the deltas, so in turn the minus in front of $\delta(x-a)$ gives a $+$ in the discontinuity. $\endgroup$
    – ndrearu
    Commented Feb 27, 2018 at 15:19

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.