Recent achievements in the field of instrumental analytical chemistry are associated with the development of devices that register a variety of analytical signals and use a variety of information processing methods. Hence, the use of advanced mathematical techniques for the treatment of a large amount of data obtained is required. Multivariate approaches are of the greatest interest for the Earth and Life sciences since most studies are based on measuring the chemical composition of complex natural organomineral substances and describing their compositions, on which their properties directly depend. Natural objects are multidimensional in their origin, so several analysis methods and one-dimensional processing techniques are traditionally used for their study and analytical control in order to achieve the required accuracy of the results. The special mathematical techniques for processing the analytical signals in emission spectra objectively reduces costs while improving accuracy. The authors in the current paper attempt to assess the need for the use of multivariate calibrations in atomic emission spectrometry with arc discharge while simultaneously analyzing different types of objects. Different variants of calibrations are considered for two techniques of direct atomic emission determination of F as well as Li, P, B, Mn, Ni, Co, V, Cr, W, Mo, Sn, Ga, Pb, Cu, Zn, Ag, Sb, As, Tl, Ge, Bi and Cd (22 elements) in powders of rocks, loose and bottom sediments, soils, ashes, gold-silver ores and products of their processing.
Read full abstract