Dear Julia, As all DFT code Wien2k can only work on a clearly defined atomic structure in the 3D space. As a consequence, each crystallographic sites are fully occupied. When a partial occupancy is observed experimentally, one solution consists to create a supercell (which will be lower in symmetry) that will allow the distribution of Cu1 and Sn1 on this site. By such an approach you will create an atomic order and for this reason it is usually necessary to consider different models in order to check that the order does not perturb the property you are looking for.
Another approach "the virtual crystal approximation" consists to mix two atoms in one in a given proportion. Then you don't need to create a supercell and you have an effective treatment of the partial occupancy. However, I don't think that such an approach will be valid in the present case, because of the strong difference between Cu and Sn atoms. Regards Xavier Julia Messel a ?crit : > Dear wien users, > > is it possible to consider for s.o.f.'s < 1 and mixed occupancy in struct > file? I create a struct-file out of a cif-file, which contains the position > x=0.32445 y=0.25 z=0.49864 shared by 0.728 Cu1 and 0.272 Sn1, but wien2k does > not accept this. > > Thanking you in anticipation, > > Julia > > > SOF > Cu1 Cu Uani 0.32445(10) 0.25 0.49864(12) 0.0156(4) 4 0.728(12) > Sn1 Sn Uani 0.3245 0.25 0.4986 0.0156(4) 4 0.272(12) > Sn2 Sn Uani 0.67730(12) 0.21643(12) 0.40562(14) 0.0370(6) 8 0.5 > P1 P Uani 0.1799(3) 0.25 0.3511(3) 0.0199(10) 4 1 > P2 P Uani 0.4699(3) 0.25 0.3449(3) 0.0168(10) 4 1 > P3 P Uani 0.07948(16) 0.0796(2) 0.4183(2) 0.0148(6) 8 1 > P4 P Uani 0.43279(16) 0.0835(2) 0.2108(2) 0.0142(7) 8 1 > P5 P Uani 0.31577(17) 0.0781(2) 0.8318(2) 0.0160(6) 8 1 > P6 P Uani 0.35655(16) 0.0532(2) 0.6295(2) 0.0202(7) 8 1 > P7 P Uani 0.03056(16) 0.0780(2) 0.8522(2) 0.0144(6) 8 1 > P8 P Uani 0.4126(2) 0.25 0.9038(3) 0.0145(10) 4 1 > P9 P Uani 0.0583(2) 0.25 0.7201(3) 0.0150(10) 4 1 > _______________________________________________ > Wien mailing list > Wien at zeus.theochem.tuwien.ac.at > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > >
