RE: mismatch between ICSD cif file and GSAS

[Von Dreele, Robert B.]

Title: Message

Don, As Allen Larson noted GSAS only uses the 2nd setting for space groups that have two (e.g. F d 3 m) so that he inversion center is placed at the origin. The eta alumina structure you have is for the 1st setting. To convert from 1st to 2nd subtract 1/8,1/8,1/8 from all xyz. That will fix the structure factors & the site multiplicity problems. Bob Von Dreele     R.B. Von Dreele IPNS Division Argonne National Laboratory Argonne, IL 60439-4814   -----Original Message----- From: Donald Wang [mailto:[EMAIL PROTECTED] Sent: Sunday, May 15, 2005 9:31 PM To: rietveld_l@ill.fr Subject: mismatch between ICSD cif file and GSAS Dear all,   I have a puzzle here.  I have been learning GSAS for a while but still at a relatively low level.  I am trying to calculate % of glass phase in a plasma sprayed alumina coating (eta alumina, ISCD #39104, space group Fd-3M, 227).  The structure is copy-pasted as below:     Atom    #          OX        SITE      x           y             z           SOF            H        ITF(B)     O         1          -2         32 e     0.37        0.37        0.37        1.               0              3.  Al        1           +3        8 a       0.25        0.25        0.25        0.503         0              3.  Al        2           +3        48 f      0.25        0             0            0.083         0              3.  Al        3           +3       16 d      0.625      0.625       0.625      0.564         0              3.  Al        4           +3       16 c      0.125     0.125        0.125       0.27          0              3.     However, when this CIF file is imported into GSAS (using EXPGUI), EXPGUI shows the following corresponding multiplicity 32 32 32 8 8   and GSAS-simulated XRD pattern is much different from the one generated from ICSD.  Please help on this mismatch between ICSD and GSAS-EXPGUI.      Many thanks in advance Don   Donald Wang University of Connecticut