Multivariate statistical characterization of the tolerance of argon inductively coupled plasmas to organic solvents

The tolerance of Argon-inductively coupled plasmas to the introduction of organic solvents, defined by Matsunaga's parameter and by “limiting aspiration rate”, has been correlated with their nature by statistical methods. The main physical variables of solvents obtained from the technical literature were used as independent variables to perform principal component analysis (PCA) and factorial discriminant analysis (FDA). Thus the variables which had the greatest influence on heat exchange nebulization and transport processes were determined. Organic solvents were classified into different groups characterized by their tolerance in plasma operation. In the PCA study, the second principal component was the most significative to differentiate between the tolerance of solvents and it showed a high correlation with surface tension, viscosity and heat of vaporization. Scores of the organic solvents belonging to the same class can be adjusted to straight lines and their positions in the plane (expressed as ordinates in the origin) as well as the slope values are characteristics of organic solvent groups. The results of FDA confirms those obtained with PCA. Multivariable regression was applied to obtain predictive equations of the limiting aspiration rate for organic solvents, in terms of their main physical variables.