Dear Andrew, You only need to constrain RM11 and RM33 to be the same for both phases; GSAS takes care of RM22 by symmetry. Bob Von Dreele
-----Original Message----- From: Andrew Hector [mailto:[EMAIL PROTECTED] Sent: Thu 3/25/2004 9:44 AM To: [EMAIL PROTECTED] I am trying to refine a magnetic structure in Cmma based on a tetragonal nuclear cell. To do this I need to constrain together the a and b parameters. I have been constraining the RM11 and RM22 terms to move together, this keeps a and b the same but introduces a high level of instability to the refinement, the fit slowly deteriorates in every cycle whenever lattice parameters are refined. The following message appears in the lst file: 2 Columns of the 4 Column matrix are 0.0 Variable NoVarabl was not refined Variable NoVarabl was not refined I'd appreciate anyone's thoughts on how to get around this. Andrew Hector. __________________________________ Dr Andrew L Hector Royal Society University Research Fellow School of Chemistry University of Southampton, SO17 1BJ, UK
