13 - PRACTICAL ISSUES IN ANALYTICAL CHEMISTRY

Summary

Producing a good chemical analysis, particularly of difficult materials such as those often encountered in archaeology, is as much an art as a science. It requires experience, ingenuity, perseverance, and, often, luck. There are, however, a number of steps which can be taken to minimize the dependence on the element of luck. These include well-thought-out and appropriate sampling procedures, high quality standards matched to the material being analyzed, a rigorous approach to calibration, and, above all, a laboratory dedicated to good quality assurance procedures. Here we outline some of the procedures used for the preparation of solutions for inorganic trace element and organic residue analysis, and give a series of basic “recipes”. A more comprehensive treatment can be found in Mester and Sturgeon (2003). We then discuss in detail the types of analytical standards used to calibrate analytical instruments (synthetic, primary, secondary, and internal). The meaning of the terms accuracy and precision is explained, together with some mathematical procedures for calculating the errors associated with the analysis, particularly those arising from linear regression, which is almost invariably used to quantify analytical data. The classic work on quantifying errors of measurement is Topping (1972), and Barford (1985) and Miller and Miller (1993) present much of the basic mathematics necessary to deal with the statistical side of analytical chemistry. The last section discusses appropriate overall quality assurance procedures – essential to ensure (and demonstrate) that the analytical data are fit for purpose.