Hello, I am trying to use MPB to plot the bands for a strongly dispersive material. I have chosen to use the find-k function as I can change the epsilon based on the input frequency to the find-k function. I first iterate over the "num-bands" values, then over some range of frequencies I define, and finally over some list of k-points that are interpolated between the corners of the irreducible Brillouin zone. In my case I am using a triangular lattice so the corners are defined as follows:
(vector3 0 0 0) ; gamma (vector3 0 0.5 0) ; M (vector3 (/ -3) (/ 3) 0) ; K A 4 point interpolation between the M and K point would result in the following intermediate k-points: #(0 0.5 0) #(-0.0666666666666667 0.466666666666667 0.0) #(-0.133333333333333 0.433333333333333 0.0) #(-0.2 0.4 0.0) #(-0.266666666666667 0.366666666666667 0.0) #(-0.333333333333333 0.333333333333333 0) I can then input these vectors as kdir to the find-k function. I multiply the output (kdir1,kdir2,kdir3) by kmag and have some k-point at a given frequency for a given band. For example, for band 1, kdir (-0.0666666666666667 0.466666666666667 0.0), the outputs are: band, frequency, k-point: 1 0.25 #(-0.0649337682727207 0.454536377909045 0.0) 1 0.2 #(-0.0468389697323159 0.327872788126211 0.0) 1 0.15 #(-0.0341039817856597 0.238727872499618 0.0) 1 0.1 #(-0.0223618836388612 0.156533185472028 0.0) 1 0.05 #(-0.011082887945324 0.0775802156172679 0.0) I think that in order to plot the outputs similarly to what can be done for the normal run functions, the outputs would have to match the input k-point exactly. So I would want to use this output to find some frequency that has the k-point (-0.0666666666666667 0.466666666666667 0.0) via interpolation or some other means. The problem is that there are no output k-points that bound the input, i.e. one that has a larger and smaller magnitude, so I'm not sure how to approximate this. I'm not entirely sure my train of thought is correct on this one. Any help would be appreciated! Regards, Erika
_______________________________________________ mpb-discuss mailing list mpb-discuss@ab-initio.mit.edu http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/mpb-discuss
