| Author | Message | ||
     anghel (aanghel) Junior Member Username: aanghel Post Number: 3 Registered: 06-2005 |
Does anybody knows how to implement absorbing boundary conditions for the time dependent 2D Schroedinger equation | ||
| anghel (aanghel) Member Username: aanghel Post Number: 4 Registered: 06-2005 |
I came back to my question with some clarifying coments since I did not yet get an answer. I want to solve the time dependent Schrodinger equation in 2D. the initial value of the wave function is a gaussian centered somwhere in the problem domain. This gausian wave packet moves an when aproches the boundary it start reflect and interfere if I set normal BC on this boundary. This is not what I want. I want the wave packet to move unperturbed out of the doman, i.e. through the boundary as if it were not there. In the literature this is called absorbing boundary conditions. | ||
| Robert G. Nelson (rgnelson) Moderator Username: rgnelson Post Number: 692 Registered: 06-2003 |
Well, what does the literature tell you about the way to pose an absorbing boundary condition? FlexPDE is just a finite element model builder for partial differential equations. If you know how to do it in a text book, then you know how to do it in FlexPDE. | ||
| anghel (aanghel) Member Username: aanghel Post Number: 5 Registered: 06-2005 |
The literature does not tell me nothing about. Thatīs why I am asking the community. All you see are nice results with the mention that in order to save memory, absorbing boundary conditions have been imposed on the boundaries. I simply hopped you or someone else has a trick or knows some reference. | ||
| Robert G. Nelson (rgnelson) Moderator Username: rgnelson Post Number: 693 Registered: 06-2003 |
Prof. Backstrom discusses FlexPDE for the Schroedinger equation in his book "Waves by Finite Element Analysis", available from the link on www.pdesolutions.com/bookstore.html. He gets the same kind of relation as the Sommerfeld condition for electromagnetic waves, namely natural(Psi_real) = -k*Psi_imag natural(Psi_imag) = k*Psi_real |