Fine chemicals and analytical reagents

In a number of cases, simple dissolution of a solid sample in an appropriate solvent is possible and some laboratory reagents may even be analysed without further treatment. Prior to flame analysis, the best solvent is dilute hydrochloric acid, provided of course that the major matrix elements are not silver, lead or another element which forms a sparingly soluble chloride. If additional oxidising ability is required, concentrated nitric acid may be added to the solvent. This acid is the preferred solvent when the analysis is to be completed by electrothermal atomisation. If the material contains large amounts of silica it may be necessary to add hydrofluoric acid after preliminary digestion with hydrochloric acid (see Chapter 4g). Care should of 

Acid dissolution is a particularly favourable approach for carbonates and sulphides, where the matrix anion will be removed during the evolution of carbon dioxide or hydrogen sulphide, and for salts of organic acids, where the anion seldom causes interference problems. Conversely, sulphates can cause problems during flame atomisation and chlorides during furnace atomisation ways of dealing with such problems are discussed below. 

The standards are best prepared with a matching sodium chloride content (15 gl 1) to minimise errors from differing viscosity and ionisation suppression. 

Blank determinations following any of the above dissolution procedures are important and may prove significant at the trace level. While it is always advisable to prepare the standards in the same concentration of acid and major matrix elements as the sample, to overcome variations in uptake rate and atomisation, it is also preferable to take a blank through the whole dissolution procedure. Losses of trace metals will be minimised if dilute solutions are kept at a pH 2 and analysis is not delayed. 

Chloride media present a particular problem in electrothermal atomisation. Manning and Slavin [3] have described the determination of lead in magnesium and sodium chlorides at the 0.1/ugg-1 level, and their paper contains many points of relevance to other analyses. Pyrolytic carbon-coated 

---