Dear Jinxi, the definition of irmin and irmax depends on your choice of the volume where the local DOS has to be integrated. Reasonable numbers go from 1 (first point in the FFT grid) to N (last point in the FFT grid) where N is output by pw.x (see "FFT grid:"). Points in real space going from irmin to irmax are included. This is done for the three dimensions independently. So suppose you want to take a volume which goes from the origin of the unit cell to half of it in the three dimensions (12.5% of total volume), and your FFT grid contains 64x64x50 points, then you would set: irmin(1)=1 irmax(1)=32 irmin(2)=1 irmax(2)=32 irmin(3)=1 irmax(3)=25
HTH Guido PS: Unless for specialized applications, you might perhaps be interested in the "projected" (on the atomic wavefunctions) rather than "local" DOS, both are computed by projwfc.x. Il 12/06/2011 10:17 AM, jinxi ha scritto: > Dear all: > I try to get local DOS using projwfc.x, but i can't understand the > description in "INPUT_projwfc": how to set parameters irmin and irmax? > I've read the input files in example16, but i still don't know how to > locate the local DOS by irmin and irmax. Are there general methods to > calculate the number and position of grid points? > If there are some general references (books, articles, presentations) > for the basic knowledge, please recommend to me. > Many thanks! > best wishes > jinxi > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Guido Fratesi Dipartimento di Scienza dei Materiali Universita` degli Studi di Milano-Bicocca