Dear Martin,
The main problems of the e.s.u. expressions used in Rietveld programs are the
neglect of correlations between residuals and the loss of interrelations
between particular residuals and particular parameters in terms of their mutual
influence.
Some references on the topic:
J. Appl. C
Dear Apu,
If you use the Rietveld-based QPA you may assess the detection limit for a
phase from the estimated standard uncertainty of its weight (or volume)
fraction. In this case one should also take into account the fact that the
Rietveld standard uncertainties of QPA are underestimated about
