Dear all, perhaps some more comments on this, because the comments by Markus are true, but might be misunderstood
Markus Valkeap�� schrieb: > At 14:24 2004-02-21 +0000, you wrote: > > > Someone could please guide me through the definitions of magnetic moment > > > components Mx, My, Mz in Fullprof. How is Magnetic moment M given in the > > > out file related to these components. > > > >here I cannot answer because I do not use Fullprof. However, Mx,My,Mz are > >the projections of the magnetic moment vector on the crystal axes. For > >rectangular systems, it is easy to calculate the overall M from these > >(Pytagoras), for hexagonal, monoclinic and triclinic you have correctly > >account for the angles between the Mx,My,Mz. > > > > > > That's the way Fullprof seems to do it too. However, isn't is so that, at > least for powders, we loose one dimension. That is, for instance in > hexagonal and tetragonanl system one gets Mx and Mz only (or My and Mz, > difference can't be seen). Thus Pythagoras is valid in hexagonal case also. > Below a line from Fullprof out-file: > > ---clip--- > Atom Mx dMx sMx My dMy sMy Mz dMz sMz > M sM MPhas dMPhas sMPhas > Mn1 2.382 0.000 0.066 2.382 0.000 0.066 0.422 0.000 0.354 > 2.4190 0.0238 0.0000 0.0000 0.0000 > ---clip--- The "loss of one dimension" refers to cases where the "configurational symmetry" remains cubic (where all directions are lost) tetragonal or hexagonal.(G. Shirane, Acta Crystallogr. 12 (1959) 282.) > > As I understand it "M" is the magnetic moment: sqr(2.382^2 + 0.422^2) =3D > 2.4190. "My" is not taken into account, program assumes restrictions due to > the crystal system? > > Best Regards to all, > -Markus > This is here the case, because My is zero or can be set zero and Mx is perpendicular to Mz, because here, apparently, the 'configurational symmetry' is hexagonal (although the magnetic space group symmetry must be lower than hexagonal, because (at least for a collinear case) a magnetic moment along a particular direction perpendicular to c breaks the hexagonal symmetry.) I have to admit that I do not know how the magnetic symmetry is dealt with in Fullprof, but, it is true that if the configurational symmetry is hexagonal, it makes no sense to refine Mx and My simultaneously, and thus the problem to perform a vector addition of non-perpendicular Mx and My components of the magnetic moment is unlikely to occur. But in general, there can be structures with hexagonal nuclear structure and a configurational symmetry lower than hexagonal. In that case the magnetic moment direction perpendicular to the c axis does not get lost. Therefore, one must be careful to what types of symmetry the terms "cubic", "hexagonal" and "tetragonal" refer to (nuclear structure, configurational symmetry, or full magnetic space group symmetry (Schubnikov group)). Best regards Andreas -- Dr. Andreas Leineweber Max-Planck-Institut fuer Metallforschung Heisenbergstrasse 3 70569 Stuttgart Germany Tel. +49 711 689 3365 Fax. +49 711 689 3312 e-mail: [EMAIL PROTECTED] home page of department: http://www.mf.mpg.de/de/abteilungen/mittemeijer/english/index_english.htm
