Support Options

Submit a Support Ticket

  • Discoverability Visible
  • Join Policy Invite Only
  • Created 29 Jun 2010

K point treatment for bulk-oriented ballistic transport.

  1. Ganesh Krishna Hegde

    The scenario I have in mind is this – say I have a certain structure that is periodic along all three crystal_directions. Out of the three crystal_directions, I choose one to be my direction of transport making the other two transverse directions. In the reciprocal lattice, I will have forward moving modes travelling along direction of transport and transverse modes. In such a scenario, how do I specify the k-point grid in NEMO5 in such a way that I have ‘n’ band structures corresponding to n transverse modes? Let’s take a concrete example. Cu nanowire. transport to be calculated along z. I orient 001 along z and 100 and 010 along x and y respectively. I have a cross section that is 2 atoms thick along z and 10 atoms thick along x and y respectively. In reciprocal space I have a (n X m X p) grid, but from a transport perspective, I have p forward moving per transverse mode. i.e. I have nXm independent band structures plotted versus kz. So, how do I specify the k-point grid to handle such a situation in Nemo5?

    Report abuse

  2. Michael Povolotskyi

    Let us clarify the question. First of all, if you have a periodic bulk crystal you will have a 3D k-space.

    I guess what you need is to have a unit cell with open boundary conditions along z and periodic along x and y. In this case you will have a 2D k-space (reciprocal to the (x,y) plane) and propagating modes along z axis.

    Currently the open boundary conditions are not part of the Schroedinger class, they are part of the Propagation class family.

    Report abuse

  3. Ganesh Krishna Hegde

    Michael, I understand your point. But my intention is to have 3D k-space. But I would like to format my 3D k space a little differently as I would like to divide into ‘forward moving’ and ‘transverse’ modes as though I expect a certain direction to be direction of ‘transport’. So then I would have a band structure (E vs kz, say) for each transverse mode (each distinct grid point of the (kx, ky) grid). Is there any way to format my output to look like this in the Schrodinger class?

    Report abuse

  4. Ganesh Krishna Hegde

    I think I have this figured out. I can go from say (0,0,0) to (0,0,0.5) to (0,0.1,0) and finally (0,0.1,0.5) with just 2 steps for the segment (0,0,0.5) to (0,0.1,0). This essentially would imply that we have two band structures being appended one below the other. I guess I hadn’t really understood how the k-segmentation works. Thanks for your comments anyways.

    Report abuse, a resource for nanoscience and nanotechnology, is supported by the National Science Foundation and other funding agencies. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.