Draft: Master
1 unresolved thread
1 unresolved thread
Course material
Edited by Michael Wimmer
Merge request reports
Activity
assigned to @michaelwimmer
General organization remark: I see that the
docs
folder is also in the repository, but that is the one generated byexecute.py
as far as I understand. Basically, all material (.md
files, figures, etc.) should go into thesrc
folder.Could you please reorganize this?
Additionally, it's fine to have
svg
s as figures, not necessary to convert topng
.
WKB
QM in a smooth potential
-
In the classical section, we should have sketches for the cases E>V(x) and E<V(x) e.g. like this:
and
Initially I was thinking whether we should have an animation here, but I think it's overkill.
WKB wave function
Edited by Michael Wimmer-
In the classical section, we should have sketches for the cases E>V(x) and E<V(x) e.g. like this:
Connection formulas
- Is it possible to put the final connection formulas into a table? I was wondering if that improves the formatting.
Bound states
-
The picture of the initial situation would fit better as a summary at the end of the section (The reader will not know what is)
-
I'm not entirely sure what is a good initial plot. I think what is important to bring across is that we are considering a situation with $V(x)>E$ for large enough x such as
-
I would suggest to have for each of the subsections (each of the different cases) to have one individual, but more detailed sketch. I.e. indicating in the sketch what andare, and also indicating which of them depends onexplicitly (i.e.).
Edited by Michael Wimmeradded 1 commit
- 4e6e1be4 - correct animations landau zener and add tables wkb connection
Tunneling
I think we should reorganize this section, and essentially revert the order.
- Change the title into "Transport"
-
Start with a sketch explaining that we now deal with a problem opposite to the bound state problem, namely E>V(x) for |x| -> \infty. And that then you can define a transport problem similar to what they did in QM with tunneling:
- Then continue with the case where E>V(x) everywhere, and hence T=1. Instead of the wave packet propagation we could instead simply have a plot of the WKB stationary wavefunction (specifically, the real part of psi, not the absolute value, in order to see oscillations). There you also see how the wave length and amplitude change. Ideal would be to have a slider where one can change e.g. the potential height (maybe even from negative [dip] to positive [peak, but always staying in E>V(X)) and see how the wave function changes - like was done in "Validity of the connection formulas"
-
Only then I would go to tunneling. Now you show there 2D kwant simulations I believe. I actually meant to simply do a 1D kwant simulation, and plot a 1D wave function (e.g. the real part of psi should look like that
). Maybe also there we could have a slider changing the energy/potential height?
Please register or sign in to reply