From: "Dr J.Ian Langford" (langfoji@novell1.bham.ac.uk) Organization: The University of Birmingham To: L.M.D.Cranswick@dl.ac.uk (L. Cranswick) Date: Thu, 20 May 1999 17:02:49 BST Subject: Re: LAPOD LAPOD: a note on the correction of powder data for instrumental peak shifts In my view, it is better to obtain instrumental peak shifts by using data from an internal or external standard, rather than to include relevant parameters as refinable quantities. At Birmingham our correction for instrumental shifts is a two stage process. We use 5-10 micron silver (high purity, Koch-Light) as an external standard for reflections at medium and high angles (NIST SRM675 at low angles) and assume that shifts are due only to zero error and specimen surface displacement (SSD). Transparency is a problem with Si, unless a smear sample is used. We then obtain a least-squares fit of the difference between the observed Ag line positions and those calculated, at the temperature of the instrument, versus cos theta. This will be linear, unless some other misalignment is appreciable, and it gives the zero error and SSD. Microsoft EXCEL is used for this purpose. We then use a standard program for producing hkl, d and 2theta from cell parameters for the sample of interest, for a given system and S.G. and with allowance for instrumental shifts as determined from the Ag data. This provides the hkl and corrected 2theta input for LAPOD. If the sample is not highly absorbing, the effective linear absorption coefficient, taking into account packing density, if the sample is a powder, can also be input. However, at present the correction on;y applies to the case os a 'thick' sample [mu*t*cosec(theta) >> 1, where t is the sample thickness]. mu can be measured directly by transmission or calculated, if the packing density is ascertained. The best account of sample preparation for obtaining precise cell parameters, and for XRD in general, is probably Chapter 9 in Jenkins & Snyder (1996) 'Introduction to X-ray Powder Diffractometry' (New York: Wiley) - and indeed of most other aspects of experimental powder XRD. Ian Langford