Dear Lorenzo, See the solution posted on the link below 👇 https://www.mail-archive.com/[email protected]/msg00278.html Hope this helps
On Tue, Nov 19, 2024, 17:02 Lorenzo CAVICCHI <[email protected]> wrote: > Dear all, > > I am trying to implement a transport calculation for the Haldane model on > armchair and zig-zag graphene nanoribbons (with open boundary conditions > along the y-direction and periodic boundary conditions along the > x-direction, i.e., the transport direction). In particular, my system > consists of a finite-sized scatterer attached to two semi-infinite leads. > The system is "homogeneous," meaning that all three components (left lead, > scatterer, and right lead) share the same Haldane Hamiltonian ( > https://doi.org/10.1103/PhysRevLett.61.2015). > I am writing to seek assistance because I am encountering two major > problems when extracting the eigenstates of the scatterer at a fixed energy: > > 1)When the system is constructed with zig-zag edges, I find a limit on the > width of the sample in the y-direction, beyond which the wave functions > cannot be calculated using the kwant.wave_function() method. Specifically, > if the ribbon exceeds this width, for certain energy values, calling > kwant.wave_function(sys, energy) produces the following error: > "RuntimeError: Numbers of left- and right-propagating modes differ, > possibly due to a numerical instability." > This appears to result from an accumulation of lead bands, which makes the > identification of propagating modes inaccurate. > 2)When the system is constructed with armchair edges, the > kwant.wave_function(system, energy) method again fails to properly return > the eigenstates of the scatterer. However, in this case, the error occurs > for all ribbon widths (in the y-direction) and seems unrelated to the > energy value. The error is: > "LinAlgError: Reordering failed; problem is very ill-conditioned." > > I have not been able to resolve these issues and would greatly appreciate > any insights or suggestions you might have for addressing them. > Thank you for your time. > > Best regards, > Lorenzo Cavicchi >
