Yes, they are not implemented which really means that with -so the forces are incorrect. You have to run a position optimization without -so and as the final step when these are zero include -so. You have to hope that -so does not change the positions much. While this is not ideal, until/if someone works out the appropriate terms for -so it is all one can do in Wien2k automatically. If you are really concerned at the end you can change slightly the positions at the surface, reconverge and look at the change in energies -- i.e. a numerical differentiation of the energy to get the forces. To date nobody has automated this in Wien2k.
On Tue, Aug 7, 2012 at 9:12 AM, Kateryna Foyevtsova <foyevtsova at th.physik.uni-frankfurt.de> wrote: > Hi, > > according to the wien2k web page > > http://www.wien2k.at/reg_user/limitations/ > > calculation of forces with spin-orbit is not yet implemented. > > > > > On 07/08/12 16:05, Alexey Korshunov wrote: >> Prof. Blaha and wien users. >> >> I am running wien version 11.1 on a intel core i7 machine with >> Mandriva 2010 operating system, fortran compiler ifort 11.1 and math >> libraries mkl 10. >> The purpose of my calculations is to get forces to perfom surface >> structure relaxation. >> I am calculating tungsten slab with the following initial structure >> >> W_surface >> CXY LATTICE,NONEQUIV.ATOMS: >> 765_Cmmm >> MODE OF CALC=RELA >> unit=bohr >> 5.981401 8.458979 84.589790 90.000000 90.000000 >> 90.000000 >> ATOM -1: X=0.00000000 Y=0.00000000 Z=0.00000000 >> MULT= 1 ISPLIT= 8 >> W 1 NPT= 781 R0=0.00000500 RMT= 2.2000 Z: >> 74.0 >> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000 >> 0.0000000 1.0000000 0.0000000 >> 0.0000000 0.0000000 1.0000000 >> ATOM -2: X=0.50000000 Y=0.00000000 Z=0.05000000 >> MULT= 2 ISPLIT= 8 >> -2: X=0.50000000 Y=0.00000000 Z=0.95000000 >> W 2 NPT= 781 R0=0.00000500 RMT= 2.2000 Z: >> 74.0 >> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000 >> 0.0000000 1.0000000 0.0000000 >> 0.0000000 0.0000000 1.0000000 >> ATOM -3: X=0.00000000 Y=0.00000000 Z=0.10000000 >> MULT= 2 ISPLIT= 8 >> -3: X=0.00000000 Y=0.00000000 Z=0.90000000 >> W 3 NPT= 781 R0=0.00000500 RMT= 2.2000 Z: >> 74.0 >> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000 >> 0.0000000 1.0000000 0.0000000 >> 0.0000000 0.0000000 1.0000000 >> ATOM -4: X=0.50000000 Y=0.00000000 Z=0.15000000 >> MULT= 2 ISPLIT= 8 >> -4: X=0.50000000 Y=0.00000000 Z=0.85000000 >> W 4 NPT= 781 R0=0.00000500 RMT= 2.2000 Z: >> 74.0 >> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000 >> 0.0000000 1.0000000 0.0000000 >> 0.0000000 0.0000000 1.0000000 >> ATOM -5: X=0.00000000 Y=0.00000000 Z=0.20000000 >> MULT= 2 ISPLIT= 8 >> -5: X=0.00000000 Y=0.00000000 Z=0.80000000 >> W 5 NPT= 781 R0=0.00000500 RMT= 2.2000 Z: >> 74.0 >> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000 >> 0.0000000 1.0000000 0.0000000 >> 0.0000000 0.0000000 1.0000000 >> ATOM -6: X=0.50000000 Y=0.00000000 Z=0.25000000 >> MULT= 2 ISPLIT= 8 >> -6: X=0.50000000 Y=0.00000000 Z=0.75000000 >> W 6 NPT= 781 R0=0.00000500 RMT= 2.2000 Z: >> 74.0 >> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000 >> 0.0000000 1.0000000 0.0000000 >> 0.0000000 0.0000000 1.0000000 >> ATOM -7: X=0.00000000 Y=0.00000000 Z=0.30000000 >> MULT= 2 ISPLIT= 8 >> -7: X=0.00000000 Y=0.00000000 Z=0.70000000 >> W 7 NPT= 781 R0=0.00000500 RMT= 2.2000 Z: >> 74.0 >> LOCAL ROT MATRIX: 1.0000000 0.0000000 0.0000000 >> 0.0000000 1.0000000 0.0000000 >> 0.0000000 0.0000000 1.0000000 >> 8 NUMBER OF SYMMETRY OPERATIONS >> >> using GGA PBE XC potential, spin-orbit interaction without spin >> polarization, RKmax=7, k-mesh=2000. >> The commandline is run_lapw -p -so -cc 0.00001 -ec 0.0001 -fc 1.0 -i >> 120 -r 20. >> There is the force result of last two iterations: >> >> :ITE069: 69. ITERATION TOTAL FORCE WITH RESPECT TO THE GLOBAL >> COORDINATE SYSTEM: >> :FGL001: 1.ATOM 0.000000000 0.000000000 >> 0.000000000 partial forces >> :FGL002: 2.ATOM 0.000000000 0.000000000 >> 3.035000000 partial forces >> :FGL003: 3.ATOM 0.000000000 0.000000000 >> 2.956000000 partial forces >> :FGL004: 4.ATOM 0.000000000 0.000000000 >> 16.247000000 partial forces >> :FGL005: 5.ATOM 0.000000000 0.000000000 >> 79.306000000 partial forces >> :FGL006: 6.ATOM 0.000000000 0.000000000 >> 5.081000000 partial forces >> :FGL007: 7.ATOM 0.000000000 0.000000000 >> 828.277000000 partial forces >> >> :ITE070: 70. ITERATION TOTAL FORCE WITH RESPECT TO THE GLOBAL >> COORDINATE SYSTEM: >> :FGL001: 1.ATOM 0.000000000 0.000000000 >> 0.000000000 total forces >> :FGL002: 2.ATOM 0.000000000 0.000000000 >> 289.947000000 total forces >> :FGL003: 3.ATOM 0.000000000 0.000000000 >> 123.015000000 total forces >> :FGL004: 4.ATOM 0.000000000 0.000000000 >> 72.397000000 total forces >> :FGL005: 5.ATOM 0.000000000 0.000000000 >> 33.811000000 total forces >> :FGL006: 6.ATOM 0.000000000 0.000000000 >> 25.803000000 total forces >> :FGL007: 7.ATOM 0.000000000 0.000000000 >> -56.857000000 total forces >> >> The difference is because of appearance of huge valence forces in last >> iteration: >> >> VALENCE-FORCE IN mRy/a.u. = |F| Fx Fy Fz >> :FVA001: 1.ATOM 0.000 0.000 0.000 >> 0.000 >> VALENCE-FORCE IN mRy/a.u. = |F| Fx Fy Fz >> :FVA002: 2.ATOM 287.029 0.000 0.000 >> 287.029 >> VALENCE-FORCE IN mRy/a.u. = |F| Fx Fy Fz >> :FVA003: 3.ATOM 119.172 0.000 0.000 >> 119.172 >> VALENCE-FORCE IN mRy/a.u. = |F| Fx Fy Fz >> :FVA004: 4.ATOM 55.430 0.000 0.000 >> 55.430 >> VALENCE-FORCE IN mRy/a.u. = |F| Fx Fy Fz >> :FVA005: 5.ATOM 51.852 0.000 0.000 >> -51.852 >> VALENCE-FORCE IN mRy/a.u. = |F| Fx Fy Fz >> :FVA006: 6.ATOM 29.373 0.000 0.000 >> 29.373 >> VALENCE-FORCE IN mRy/a.u. = |F| Fx Fy Fz >> :FVA007: 7.ATOM 684.540 0.000 0.000 >> -684.540 >> >> The total forces in the 70-th iteration look unphysical. A very high >> force has appeared on the next to the internal layer of the slab >> (it is the 6-th layer from the surface). >> This leads to an icrease of interlayer distance between internal >> monolayers. On the other hand, the total forces in the 69-th iteration are >> rather realistic. However, they do not contain the valence-force >> corrections. >> This effect appears only in the spin-orbit mode. >> >> The questions are: >> 1. Is it possible to use the forces without valence-force corrections, >> e.g. pure HELLMAN-FEYMAN forces, for the relaxation? >> >> 2. Can the spin-orbit mode lead to the wrong forces? >> >> Sincerely, Korshunov Alexey, Sarov State Physical Tecnical Institute, >> Russia. >> al.korshunof at gmail.com <mailto:al.korshunof at gmail.com> >> >> >> >> _______________________________________________ >> Wien mailing list >> Wien at zeus.theochem.tuwien.ac.at >> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > > _______________________________________________ > Wien mailing list > Wien at zeus.theochem.tuwien.ac.at > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu 1-847-491-3996 "Research is to see what everybody else has seen, and to think what nobody else has thought" Albert Szent-Gyorgi